Background

Histomoniasis, also known as infectious cecum hepatitis or “blackhead disease”, is a disease of gallinaceous birds (turkeys, chickens, quails, and peacocks) caused by the H. meleagridis protozoan parasite [1]. H. meleagridis has a complex life history and transmission routes [2]. In the natural environment, it usually parasitizes in Heterakis gallinarum (H. gallinarum) eggs or earthworm and survive for a long time [3, 4]. H. meleagridis mainly parasitize the cecum and liver of the host [1], causing cecum mucosal lesion, intestinal wall hypertrophy, caseous cecum core, and yellowish-green round inflammatory necrotic focus in the liver, which seriously affects the metabolism and absorption of nutrients, in severe cases, excessive inflammatory response and immune dysfunction caused infected avian death [5]. The intestinal mucosal is the host’s first barrier against H. meleagridis infection. IgA, an important part of mucosal immunity against pathogen invasion, has been shown that it will be continuously elevated in response to H. meleagridis’ invasion of cecum mucosa [6]. Presently, Histomoniasis have become a worldwide disease. In Europe and America where farming a large number of turkeys, with high morbidity and fatality [7]. Although chickens are less vulnerable to the disease than turkeys (100%), the epidemic of the disease in Chinese chicken flocks is also very serious, and the mortality rates are 20 to 30% [8]. Since most chemical drugs that can effectively control and prevent the disease were banned owing to their potential carcinogenicity [9], which has contributed to the incidence of the disease is increasing year by year and causing severe economic losses.

MicroRNA (miRNA) is a class of short non-coding RNA molecules expressed by animals, plants, viruses, and some single-celled organisms, with a length of approximately 22 nucleotides [10]. MiRNAs play important role in the regulation of cellular signal networks in both normal and diseased conditions. In eukaryotes, 2/3 of the coding genes are regulated by miRNAs [11], which participate in the regulation of many physiological processes [12], such as development, cell division, proliferation, and metabolism, and play an essential role in the inflammatory response [13], immune-related pathway [14], tumorigenesis [15] and so on. A growing number of studies have shown that host miRNAs modulate target gene expression at the post-transcriptional level by inhibiting translation and promoting degradation of target genes, and play an important role in against parasite infection [16, 17].

Since the first report of Histomoniasis, increased attention has been paid to histopathological features [18], etiology [19], virulence [20], and gene expression analysis [21]. However, the molecular mechanism of the interaction between H. meleagridis and chicken is not clear, especially the study of miRNA expression in chicken cecum after H. meleagridis infection. This study investigated the expression profile of chicken cecum miRNA at 10 and 15 days post-infection. To our knowledge, this is the first report on the expression of miRNA in chicken cecum during H. meleagridis infection. This article will further our understanding of the interactions between the host and H. meleagridis. This will aid the development of novel therapies against H. meleagridis in the future.

Methods

Animals and experimental infection

An F strain of H. meleagridis, obtained from a home-bred chicken in Jiangsu Province, China, was cryopreserved in liquid nitrogen in our laboratory. 40 SPF White Leghorn layers (15-day old) were used in this study. 30 chickens were divided into the infection groups and 10 chickens were divided into the control group. Chickens in the infection group were inoculated with 2 × 105 H. meleagridis through the cloaca, and the control group chickens were not treated.

Sample collection and preparation

At 10 and 15 days post-infection (DPI), half of the chickens in the two groups were killed, respectively. The chickens cecum samples aseptically collected, were thoroughly rinsed in PBS, and immediately frozen in liquid nitrogen. All the cecum samples were stored at − 80 °C until RNA extraction [17].

Total RNA was prepared from individual cecum samples using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instruction. The purity and integrity of RNA samples were assessed using the RNA Nano 6000 Assay Kit of the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA) and spectrophotometer (IMPLEN, CA, USA), respectively [17].

Small RNA library preparation and sequencing

Twelve libraries were constructed from the cecum of 10 DPI (CE), and control group (CC), 15 DPI (CEH), and control group (CCH), with three in each group. The small RNA libraries were prepared from a total of 2 μg total RNA isolated from each sample using NEBNext® Multiplex Small RNA Library Prep Set for Illumina® (NEB, USA) according to the manufacturer’s instruction and sequenced at the Novogene Bioinformatics Institute (Beijing, China) on an Illumina Novaseq 6000 platform (Illumina, San Diego, CA, USA) following the vendor’s instructions.

Basic data processing

Raw data (raw reads) of Fastq format were firstly processed through custom Perl and Python scripts. In this step, clean data (clean reads) were obtained by removing reads containing ploy-N, ploy A or T or G or C, or with 5′ adapter contaminants, or without 3′ adapter or the insert tag. Moreover, the 3′ adapter sequences were trimmed. At the same time, Q20, Q30, and GC content of the raw data were calculated. Clean reads with 18 to 35 nt length range were chosen for downstream analyses.

The processed small RNA reads were used in Bowtie [22] for read mapping to reference sequence. This allowed for 1 mismatch base.

Processed reads of length at 18 to 35 nt were then mapped to their reference genome and analyzed using the bowtie package (no mismatch). To identify conserved miRNAs, the predicted miRNA hairpins were compared against miRNA precursor sequences from miRBase22.0 (http://www.mirbase.org/) using mirDeep2 [23]. Srna-tools-cli (http://srna-tools.cm.uea.ac.uk/) were used to obtain the potential miRNA and draw the secondary structures. MirDeep2’s quantifier. pl were used to obtain the miRNA counts, and custom scripts were used to obtain base bias on the first position of identified miRNA with 18 to 35 nt length and each position of all identified miRNA respectively.

The available software miREvo [24] and MirDeep2 [23] were integrated to predict novel miRNA through exploring the secondary structure, the Dicer cleavage site and the minimum free energy of the small RNA reads unannotated in the former steps.

All sequence data were submitted to the NCBI Gene Expression Omnibus (GEO) public database (http://www.ncbi.nlm.nih.gov/geo/) with the GEO accession number GSE193859.

Analysis of differentially expressed miRNAs

The expression levels of known and new miRNAs in each sample were counted, and transcripts per million clean tags (TPM) [25] was used to normalize the expression levels. Differential expression analysis of two groups was performed using the DESeq R package (1.24.0). The P-value was adjusted using the Benjamini & Hochberg method [26]. Corrected P-value < 0.05 was set as the threshold for screening differentially expressed genes.

Predicted target genes of miRNAs and bioinformatics analysis

miRanda [27] and RNAhybrid [28] were used to predict the target gene of miRNA. GOseq based Wallenius non-central hypergeometric distribution [29] which could adjust for gene length bias, was implemented for GO enrichment analysis. GO terms with a P-value < 0.05 were regarded as significantly enriched terms. Additionally, KEGG [30] pathway with a P-value < 0.05 were considered significantly enriched pathways. KOBAS [31] software was used to test the statistical enrichment of the target gene candidates in KEGG pathways.

miRNA-gene network

To dissect the role of DEMs in inflammatory and immune, Cytoscape3.9.0 software was used to construct DEMs and immune and inflammatory-related genes regulatory networks at 10 and 15 DPI.

Quantitative real-time qPCR validation

Nine DEMs, including 3 miRNAs in 10 DPI, 2 miRNAs in 15 DPI, and 4 miRNAs shared at 10 and 15 DPI, were selected and measured using SYBR green-based RT-qPCR to verify the sequencing results. miRNA sequences in which uracil was replaced by thymine were used as the forward primers for the real-time PCR described in Table 1. The miRNA primers were synthesized by BGI Co. Ltd. (Shenzhen, China). The total RNA was extracted from cecum samples, and reverse transcribed into cDNA using a Mir-x™ miRNA First-Strand Synthesis and SYBR® RT-qPCR Kit (TaKaRa, Dalian, China) following the manufacturer’s instruction. RT-qPCR cycling conditions were as follows: 95 °C for 5 mins; followed by 45 cycles of 95 °C for 10 s, 60 °C for 10 s, and 72 °C for 15 s; and melting curve analysis from 60 °C to 97 °C. All reactions were carried out with three repeats. U6 snRNA were used as internal reference gene for quantifying miRNA expression analysis. The expression of each miRNA relative to U6 was calculated using the 2 − ΔΔCT method [17].

Table 1 The sequences of miRNAs used for RT-qPCR validation
Full size table

Results

Sequencing of small non-coding RNAs in cecum of H. meleagridis-induced chickens

Twelve cecum libraries were constructed from CC, CE, CCH, and CEH groups, with 3 in each group. High-throughput sequencing generated 11,310,686, 11,759,594, 11,103,844, and 12,587,674 average reads in the CE, CC, CEH, and CCH libraries, respectively. After removal of low-quality and adaptor contamination reads, the average clean reads obtained at each group were 11,038,098, 11,527,878, 10,876,563, and 12,315,338 respectively (Table 2).

Table 2 The list of data filtering (%)
Full size table

Most of the sRNA length in the 12 libraries were 21-24 nt. Among these, the sRNA tags of 83.28–96.43% were mapped to the chicken genome. The repetitive sequence, exon sequence, intron sequence, tRNA, rRNA, snRNA, and snoRNA were successfully annotated. 57.12–78.38% of the reads in each library were identified as known miRNAs, 0.01–0.04% were predicted to be new miRNAs (Fig. 1). A total of 797 known and 91 novel mature miRNAs, corresponding to 667 and 96 precursors, respectively, were identified with a BLAST search against the miRBase or by recognition of standard stem-loop structures (Table 3).

Fig. 1
figure 1

The annotation statistics of the unique reads from samples in each group. ‘other’ represents the unannotated small RNA. (CC1, CC2, CC3 = 10 DPI control groups; CE1, CE2, CE3 = 10 DPI infected groups; CCH1, CCH2, CCH3 = 15 DPI control groups; CEH1, CEH2, CEH3 = 15 DPI infected groups)

Full size image
Table 3 The known and novel miRNAs mapped in chicken genome
Full size table

Differentially expressed miRNAs in cecum of infected and control chickens at different time points

Pearson correlation coefficients were used to estimate expression levels to examine the gene expression patterns of miRNAs in different samples. The correlation coefficients ranged from 0.885 for CEH2 versus CC1 to 0.992 for CE1 versus CE2 (Fig. 2). A total of 161 unique chicken-encoded miRNAs were significantly differentially expressed between the infected and control samples at 10 and 15 DPI, including one (gga-novel-123) novel miRNA from 15 DPI. 94 and 127 miRNAs were identified as DEMs at 10 and 15 DPI, respectively (Fig. 3; Additional files 1, 2). A total of 101 DEMs were identified to be sample-specific, including 34 from 10 DPI and 67 from 15 DPI. 60 DEMs were shared at 10 and 15 DPI, of which 24 DEMs were up-regulated and 35 DEMs were down-regulated, interestingly, one (gga-miR-2954) of the DEMs was up-regulated at 10 DPI and down-regulated at 15 DPI (Fig. 4).

Fig. 2
figure 2

The correlation analysis between samples in different groups. Pearson correlation coefficients were calculated to estimate the association of expression levels between samples

Full size image
Fig. 3
figure 3

The volcanoplot and heatmap of the differentially expressed miRNAs. a, b The volcanopiot of the differentially expressed miRNAs at 10 and 15 DPI, respectively. c The heatmap of the differentially expressed miRNAs at 10 and 15 DPI. The blue indicated no significant difference, while the red and green indicated miRNA with significant difference (a, b). The red indicated higher miRNA expression level and the blue showed lower miRNA expression level (c)

Full size image
Fig. 4
figure 4

The differential expression analysis of chicken cecal miRNAs between infected and control groups at different periods. a The Venn diagrams of the DEMs between the infected and control groups. b The number of DEMs between the infected and control groups. (CE vs CC = numbers of 10 DPI DEMs; CEH vs CCH = numbers of 15 DPI DEMs)

Full size image

Functional enrichment analysis of target genes of differentially expressed miRNAs

RNAhybrid and Miranda software were used to predict the candidate target genes of each differentially expressed miRNA. A total of 2170 target genes for the 94 DEMs at 10 DPI and 2445 target genes for the 127 DEMs were predicted at 15 DPI. GO functional and KEGG pathway enrichment analyses were performed to better illuminate the functions of the DEMs.

The GO teams include biological process (BP), cellular component (CC), and molecular function (MF). A total of 881 and 1027 significantly enrichen GO terms (P < 0.05) were identified from 10 and 15 DPI, respectively. Some of these GO terms were shared at 10 and 15 DPI, for example, single-organism process, single-organism carbohydrate metabolic process in BP, cytoplasm, intracellular part, an intracellular membrane-bounded organelle in CC, protein binding, anion binding, kinase binding in MF (Fig. 5). Moreover, GO terms related to inflammation (e.g., cell proliferation, inflammatory response, interleukin-6 biosynthetic process, chemokine production) and immune function (e.g., death, response to stimulus, mast cell activation involved in immune response, T cell proliferation, B cell selection) were found at 10 and 15 DPI.

Fig. 5
figure 5

The GO enrichment analysis for target genes of DEMs. a The significant enriched GO terms of BP, CC, and MF for target genes of DEMs at CE vs CC. b The significant enriched GO terms of BP, CC, and MF for target genes of DEMs at CEH vs CCH. (BP = biological process; CC = cellular component; MF = cellular component)

Full size image

A total of 148 and 149 KEGG pathways were obtained at 10 and 15 DPI, respectively. The top 20 pathways of KEGG pathway analysis of differentially expressed miRNA target genes are shown in Fig. 6 (Additional file 4). At 10 DPI, only 4 pathways were significantly enriched (P < 0.05). Of these, the Hedgehog signaling pathway was the most significantly enriched, which played an important role in the repair of injury and cell proliferation. At 15 DPI, only 3 pathways were significantly enriched (P < 0.05). Among them, Endocytosis and Phagosome were associated with defense responses against pathogenic microorganism.

Fig. 6
figure 6

The KEGG pathway analysis for target genes of DEMs. The top 20 KEGG pathways of DEMs at 10DPI (a) and 15DPI (b). (The point size representss the number of genes enriched in the pathway. The x-axis shows the rich factor; the y-axis shows the pathway names)

Full size image

miRNA-gene network analysis

At 10 DPI (Fig. 7a), 24 different genes were possibly regulated by 20 DEMs. Among these genes, 6 target genes were regulated by up-regulated miRNA while 18 target genes were regulated by down-regulated miRNA. Among these DEMs, gga-miR-214 regulated the greatest number of target genes, with 5 targets. The target gene regulated by the greatest number of DEMs was TRAF2, with 3 miRNAs. At 15 DPI (Fig. 7b), 29 different genes were possibly regulated by 22 DEMs. Among these genes, 13 target genes were regulated by up-regulated miRNA while 16 target genes were regulated by down-regulated miRNA. Of these DEMs, gga-miR-12,265-5p regulated the greatest number of target genes, with 4 targets and followed by gga-miR-2954, regulated 3 targets. The target gene regulated by the greatest number of DEMs was IL5RA, with 4 miRNAs.

Fig. 7
figure 7

The network analysis of the interaction between the DEMs and their potential target genes. Red indicated up-regulated expressed miRNAs, blue indicated down-expressed miRNAs, and green indicated inflammatory and immunity-related target genes. a, b The interaction between the target genes and the DEMs in chicken cecum at 10 and 15 DPI

Full size image

Quantitative real-time qPCR validation

The expression patterns of 9 miRNAs measured with RT-qPCR consistent with the high-throughput sequencing results (Fig. 8). The results verified the accuracy and reliability of the high-throughput sequencing results.

Fig. 8
figure 8

The expression level of differentially expressed miRNAs validated by RT-qPCR. MicroRNA expression was quantified relative to the expression level of U6 using the comparative cycle threshold (△CT) method. (* P<0.05, ** P<0.01, *** P<0.005)

Full size image

Discussion

In this experiment, 94 and 127 miRNAs were identified as DEMs in chicken cecum samples at 10 and 15 DPI, respectively. It is obvious that more DEMs were identified in cecum samples at 15 DPI, and simultaneously, more severe pathological damage was also observed at this time point in this study. These not only represent the differences in the induced responses of the host between the two time points, but also imply that H. meleagridis infection may induce more biological processes involved in the host pathological formation at 15 DPI compared to 10 DPI. Among all of the identified DEMs, 60 DEMs were shared at 10 and 15 DPI. Interestingly, only one (gga-miR-2954) miRNA was up-regulated at 10 DPI and down-regulated at 15 DPI. Recent studies have revealed that miR-2954 plays an extensive regulatory role in normal development and disease [32,33,34,35]. And the other 59 DEMs showed the similar expression patterns at the two time points, indicating that these miRNAs were involved in the regulation of persistent infection of H. meleagridis. In addition, the numerous DEM homologs, including miR-29b-3p, miR-449a, let-7b, miR-146a-5p, miR-204, miR-128-3p, and miR-31-5p, miR-133a-3p, obtained from the chicken cecum in this study, has been shown to be associated with the intestinal mucosal integrity [36], intestinal inflammatory response [37,38,39,40,41,42], suggesting that these DEMs or DEM homologs may play the important roles in the cecum response against H. meleagridis. For instance, miR-133a-3p showed a downregulated expression in the cecum tissue of chickens following H. meleagridis infection at both time points, which was consistent with that in intestine of chicken coccidiosis [43], but showed an opposite expression pattern in cecum of chicken Salmonellosis [44]. A previous study found that miR-133a-3p inhibited the proliferation and promoted the apoptosis of intestinal epithelial cells by limiting the expression of TAGLN2 [36]. This indicates that miR-133a-3p might have a central role in chickens resistance to pathogenic, and could be responsible for the persistent inflammatory response and intestinal mucosal integrity in host cecum throughout the infection process.

At present, the research on host immune and inflammatory response caused by H. meleagridis infection was mainly focused on T cells, especially the immune and inflammatory response mediated by Th1 cells [45,46,47]. In this study, 27 GO terms related to T cell differentiation and function were screened to analyze the functions of the genes enriched into these terms (Additional file 3). The main findings of present work are summarized in Table 4. A large number of candidate target genes were involved in the promotion of Th1 and Th17 responses at both time points. In addition, we found that some candidate target genes inhibited Th2 responses at 10 DPI, while at 15 DPI, others candidate target genes promoted Th2 response. For example, RARA, having been shown to be involved in the differentiation of Th1 or Th2 cells [52, 58], was found to be potentially targeted by gga-miR-2954 that showed an up-regulated expression at 10 DPI and a down-regulated expression at 15 DPI. Moreover, PIPK2 [53], MALT1 [54], SOCS5 [50], STAT3 [51], IL12B [56], and IRF4 [57] has been shown to be associated with Th response. Among these, PIPK2 and MALT1 has been shown to be associated with promotion of Th1 and Th17 differentiation, respectively [53, 54]. STAT3 was capable of inhibiting Th2-mediated immune responses [51]. IRF4 has been shown to be related to promoting the differentiation of Th2 cell [57]. In this study, the potentially interactive combinations gga-miR-7460-3p/PIPK2 and gga-miR-145-5p/MALT1 at both time points, gga-miR-214/STAT3 at 10 DPI, and gga-miR-12,265-5p/IRF4 at 15 DPI may play the key roles in regulation of the Th cells responses in chicken cecum tissue infected with H. meleagridis. This showed that the host initiated different signaling pathways by the miRNAs and their targets to regulate Th cell differentiation to cope with the development of infection.

Table 4 Target genes and miRNA affection on Th cells at 10 and 15 DPI in GO terms
Full size table

A total of 24 and 29 cytokines related to inflammation and immunity were selected at 10 and 15 DPI, respectively. The relationship between these target genes and Th cell and the results of these analyses are shown in Table 5. For example, IL11RA, having been shown to be involved in the differentiation of Th2 cells [59], was found to be potentially targeted by gga-miR-2954 that showed an up-regulated expression at 10 DPI and a down-regulated expression at 15 DPI. IL5RA [63], IFNGR2 [60], CD5 [61], CD2AP [68], CD276 [48], CCL19 [69], and TRAF1 [80] are involved in the regulation of Th cell responses. Among them, IL5RA has been shown to be associated with promote Th1 cell differentiation [63]. CD5 has been found to be involved in inhibiting Th2 cell differentiation [61]. TRAF1 has been shown to be associated with promote Th2 cell differentiation [80]. In this study, the potentially interactive combinations of gga-miR-16-5p/IL-5RA at both time points, gga-miR-129-5p/CD5 at 10 DPI, gga-miR-222b-5p/TRAF1 at 15 DPI may supports and extends the findings of GO enrichment analysis. These results may explain why more severe cecum lesions and inflammatory responses were observed from 10 to 15 days after infection, and why cecum lesions and inflammatory responses gradually abated from 15 days later.

Table 5 Cytokines and miRNA affection on Th cells at 10 and 15 DPI
Full size table

In the KEGG pathway enrichment analysis of this study, in different periods after H. meleagridis infection, the host responds to the development of the disease by regulating different signal pathways (Additional file 4). The Hedgehog signaling pathway (Additional files 5, 7) at 10 DPI has been proved to play an important role in the development and function of the intestinal mucosa, gastrointestinal inflammation, and immune regulation [81, 82]. Activating the Hedgehog signal pathway inhibits the development of colitis by up-regulating the expression of anti-inflammatory cytokine IL-10 [83] while inhibiting the Hedgehog signal pathway lead to inflammatory bowel disease [84]. Gli1 and Gli2 has been shown to involved in activate the Hedgehog signal pathway to inhibit intestinal inflammation and balance inflammatory cytokines [85, 86]. In DEMs involved in the Hedgehog signaling pathway, gga-miR-6606-5p/Gli1 and gga-miR-7460-3p/Gli2 may have an important role in activating Hedgehog signaling pathway. The Phagosome (Additional files 6, 8) at 15 DPI has been shown to play an important role in the removal of pathogenic microorganisms [87, 88]. ITGB2 and SCARB1 capable of regulating the phagocytosis-promoting receptors on the surface of phagocytes [89]. In addition, SCARB1 has been shown to increase the number of anti-inflammatory macrophages and the expression of anti-inflammatory genes and is beneficial to tissue repair and regeneration [90]. RILP, M6PR, and CTSS has been demonstrated to participate in the fusion of phagosomes and late endosomes, the transport of cathepsin precursors, and the formation of cathepsin in lysosomes, respectively [91,92,93]. In DEMs involved in the Phagosome pathway, gga-miR-146c-5p/ ITGB2, gga-miR-140-3p/ SCARB1, gga-miR-145-5p/ RILP, gga-miR-148b-5p/ M6PR, and gga-miR-146a-3p/CTSS potentially involved with improved phagocytosis efficiency of Phagosome. This shows that the responses of host to H. meleagridis infection were different at 10 and 15 DPI, it mainly through the regulation of inflammatory responses at 10 DPI, while at 15 DPI, it could be mainly to remove H. meleagridis.

Conclusions

At 10 and 15 DPI, a total of 161 DEMs were found, many of which are known to regulate host immune and inflammatory responses. Compared with 10 DPI, more DEMs were found at 15 DPI. KEGG enrichment analysis showed that the responses of the body to H. meleagridis infection were different at 10 and 15 DPI. GO enrichment analysis and miRNA-gene network analysis revealed that the immune response of the host caused by H. meleagridis infection is not limited to Th1.