Analysis of deep sequencing Exosome-microRNA expression profile from Chicken Type Ⅱ Pneumocytes derived reveals potential role of gga-miRNA-451 in inflammation

Background: Exosomes are nanosized extracellular vesicles secreted by multiple cells in the body, including those located in the respiratory tract and lungs. They are emerging as important inflammatory mediators and can release their contents, especially microRNAs (miRNAs), to both neighboring and distal cells. Mycoplasma gallisepticum (MG) can target host cell and cause chronic respiratory disease (CRD) in chickens. Although exosomal miRNAs have been demonstrated to produce an important effect on microbial pathogenesis and inflammatory response as crucial regulatory noncoding RNAs, the mechanism by which exosomal miRNAs regulate MG-induced inflammation remains to be elucidated. Methods: the expression of exosome-microRNA derived from MG-infected chicken type Ⅱ pneumocytes (CP-Ⅱ) was screened, and the target genes and function of differentially expressed miRNAs (DEGs) were predicted. To verify the inflammatory functions of exosomal gga-miR-451 via targeting YWHAZ, Western blot, ELISA, and RT-qPCR were used in this study. Results: A total of 722 miRNAs were identified from the two exosomal small RNA (sRNA) libraries, and 279 novel miRNAs were discovered; 30 miRNAs (9 up-regulated and 21 down-regulated) were significantly changed (P0.05). Function annotation analysis of DEGs showed that the target miRNAs were significantly enriched in treatment group, such as cell cycle, Toll-like receptor signaling pathway and MAPK signaling pathway, etc. The results have also confirmed that gga-miR-451-absent exosomes derived from MG-infected CP-Ⅱ cells increased inflammatory cytokine production in DF-1 (chicken embryo fibroblast) cells, and Wild Type CP-Ⅱ cells-derived-exosomes displayed protective effects.

Agricultural University (Wuhan, China) (25,26). The MG-HS strain was cultured, and the concentration was determined as previously described. The viable number of MG-HS in suspension was measured using a color-changing unit (CCU) assay (27).

Infection Experiments
When CP-Ⅱ cells in the experimental group reached 80-90% confluence, twelve 150 mm plates of the cells were incubated in medium without antibiotics. Six plates of the cells were infected with 2 ml/plate of MG-HS at the mid-exponential phase (1×10 12 CCU/ml), whilst the other six plates of the cells uninfected with MG-HS used as a control. At 12 h post-infection, the medium of the cells in both groups was changed to exosome-depleted media prepared by ultracentrifugation of FBS for 70min at 100,000×g (28). The culture supernatant and the cells were collected at 12-60 hpi for further experiments.

Exosome isolation and characterization
The CP-Ⅱ cells culture supernatants in both groups were collected and immediately filtered through a 0.22-mm filter (Millipore, USA) and centrifuged within 30 min at 2000×g for 10 min at 4 °C. Cell culture exosomes were isolated and characterized as described previously (29)(30)(31). Briefly, the supernatant was collected and then centrifuged at 10,000g for 40 min at 4 °C. The resulting supernatant was transferred to a new ultracentrifuge tube and centrifuged at 100,000×g for 2 h at 4°C (Beckman Optima XE-90, SW32 Ti rotor). The supernatant was aspirated and the pellet was suspended in pure PBS (HyClone, USA) and centrifuged at 100,000×g for another 2 h at 4 °C. The purified exosomes were resuspended in 50μL PBS and used for experimental procedures or stored at -80 °C. For nanoparticle tracking analysis (NTA), exosomes were diluted with PBS over a range of concentrations to obtain between 10 and 100 particles per image before analysis. The ZetaView Nanoparticle Tracking Analyzer (Particle Metrix, Germany) was used to automatically measure the average diameter and concentration. For Transmission electron microscopy, a 10-μL aliquot of the suspended exosomes was applied to a carboncoated copper grid. Then the sample was negatively stained with 2% uranyl acetate after drying. Micrographs were obtained under a HITACHI H-7650 transmission electron microscope (HITACHI, Japan).

Exosome labeling
Exosomes secreted by CP-Ⅱ cells were labeled using a PKH67 green fluorescent labeling kit (Sigma-Aldrich, MINI67, MO, USA) to examine the uptake of exosomes by DF-1 cells in vitro. Labeled exosomes were incubated with DF-1 cells at 39 °C for 6 h and then fixed. Fluorescent images were taken with a Confocal Laser Scanning Microscope (ZEISS LSM 800 META, Carl Zeiss Imaging, Germany).

Exosome sRNA sequencing and data processing
Exosomes were isolated from samples and characterized as described above. Total RNA from exosomes in both groups (n=3) was used for sRNA sequencing. Library preparation and sRNA sequencing was performed by Ribobio (Guangzhou, China). In brief, total RNA samples were fractionated and only small RNAs ranging from 18 to libraries were sequenced using the Illumina HiSeq TM 2500 platform. Clean reads were collected from raw reads by removing the adapter dimers, glow quality, and contaminated reads. Then, clean reads of sRNA were mapped to a reference sequence by BWA 0.7.12 (32). Mapped sRNA tags were used to look for known sRNA, and miRbase version 21.0 (www.mirbase.org), Rfam12.1 (rfam.xfam.org), and Pirnabank (pirnabank.ibab.ac.in) were used for reference.

CP-Ⅱ cells-derived exosomes were isolated and identified in morphology and
phenotype.
The CP-Ⅱ cells shaped monolayers when adherent cultured for 18h, were small and cycloid, grouped together to form islands (Fig. 1A). The morphology of CP-Ⅱ cells was observed by optical microscope and transmission electron microscopy (TEM).
The cells have one or more osmophilic lamellar bodies and the surface microvilli ( Fig. 1B). The exosomes were isolated from the media and collected from MGinfected (MG) and non-infected (NC) CP-Ⅱ cells by high-speed centrifugation and were characterized by nanoparticle tracking analysis (NTA) and transmission electron microscope (TEM). These results showed an average particle diameter of 30-150 nm (Fig. 1C), double layer membrane structure of round-shaped vesicles, and diameters about 100nm (Fig. 1D). The exosomal surface markers (cluster of differentiation 9 (CD9) and CD63) could be detected using western blotting (Fig.  1E). There was difference in the MG-infected and non-infected CP-Ⅱ cells, the group of MG-Infection is more consistent with previous reports on exosomes (35). These results demonstrate that CP-Ⅱ cells-derived particles collected in this experiment were identified as exosomes.

Sequence analysis
To identify exosomal miRNAs and their expression levels from MG-infected and noninfected CP-Ⅱ cells, two small RNA libraries were constructed for deep sequencing.
The total reads reached more than thirteen million, and approximately 90.16% clean reads remained in Table S3. These clean reads were matched to authoritative miRNA/rRNA/tRNA/snRNA/snoRNA databases, such as miRbase version 21, Rfam12.1, and Pirnabank. The length distribution range of these libraries is 19-24 nt, and 22 nt is the main length in the miRNAs sequence distribution.

Differentially expressed miRNAs
A total of 722 mature chicken miRNAs were identified from the two exosomal sRNA libraries, and 279 novel miRNAs were discovered using miRDeep2 software ( Table   S4). As shown in Fig. 2A, the differential enriched miRNA profiles of all samples of MG-infected (MG) and non-infected (NC) were evaluated by DESeq. The transcriptional level of co-expressing miRNAs was compared between MG-infected and non-infected libraries (Fig. 2B). There were 469 of 722 miRNAs (65%) were equally-expressed in the two libraries, while 253 miRNAs (35%) were differential expressed with a fold-change >2 (|log 2 (Fold Change) |≥1). Among these differential expressed miRNAs, only 30 miRNAs (9 up-regulated and 21 down-regulated) were significantly changed (P-value 0.05) and 17 miRNAs (3 up-regulated and 14 downregulated) were significantly changed (P-value 0.01) in the MG-infected vs. noninfected groups (Fig. 2C). The details of the known chicken miRNAs and their log2 (fold-change) and P-value were shown in Table 1. and then applied to evaluation.

Functional annotation and enrichment analysis of predicted target genes of the differentially expressed miRNAs
To further understand of the biological function of miRNAs in CP-Ⅱ cells-derived exosomes, a target prediction of miRNAs with log 2 (fold-change) ≥1 and P-value 0.05 was performed using miRanda, Pita, and RNAhybrid. Gene Ontology (GO) functional annotation of target genes was conducted with assignments to biological process (BP), cellular component (CC) and molecular function (MF) categories. The results showed that some items of BP were involved in regulating cellular process, singleorganism process, and metabolic process, some items of CC were involved in regulating cell part, intracellular, and intracellular membrane-bounded organelle, and some of items of MF were involved in regulating protein binding, organic cyclic compound binding, and catalytic activity ( Fig. 3; corrected p-value< 0.05). Analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway for these target genes was performed (p<0.05). These target genes were mainly enriched in the following pathways: apoptosis, cell cycle, focal adhesion, tight junction, influenza A, Toll-like receptor signaling pathway, Wnt signaling pathway, Cytokine-cytokine receptor interaction and MAPK signaling pathway (Fig. 4). These pathways reflect the physiological processes of MG-infection and the potential regulatory mechanisms underlying them.

Exosomal miRNAs interaction network for elements involved in cell cycle, apoptosis, and Toll-like receptor signals
Our previous studies have shown that cell cycle, apoptosis, and Toll-like receptor signaling pathways are activated and play an important role in the inflammatory response to MG infection (16,36). We found that 20 predicted targets of DEGs are involved in regulating apoptosis, 16 predicted targets of DEGs are involved in cell cycle, and 16 predicted targets of DEGs are involved in Toll-like receptor signaling pathway (Fig. 5). Interestingly, many miRNAs (21 of 30) negatively regulate these three pathways by targeting the positive genes. Moreover, some other pathways are also modulated by DEGs, such as mitogen-activated protein kinase (MAPK) and Cytokine-cytokine receptor interaction signals.

RT-qPCR verification
To validate the reliability of the sequencing results, a total of 8 differentially expressed exosomal miRNAs were selected for the Poly (A) Plus real-time PCR. RT-qPCR results showed that the expression of gga-let-7d, gga-miR-451, gga-miR-133-3p and gga-miR-223 were significantly downregulated in MG-infected group compared to control groups. gga-miR-193a-3p and gga-miR-33-5p were up-regulated in MG-infected group as compared with the control groups. In contrast, gga-miR-460b-5p and gga-miR-202-5p had no significant difference (Fig. 6). These results indicated that exosomes derived from MG-infected CP-Ⅱ cells in early infection may play a significant role in the immune inflammatory process. Consistent with this hypothesis, gga-miR-451 is reportedly involved in the immune inflammatory process (16), suggesting that gga-miR-451 may play a crucial role in immune regulation during early infection.

gga-miR-451-desent exosomes derived from MG-infected CP-Ⅱ cells increases inflammatory cytokine production in DF-1 cells
Previous data showed that gga-miR-451 expression is significantly elevated during MG infection in DF-1 cells. To verify this result in chicken lung epithelial cells, CP-Ⅱ cells were infected with MG-HS. At 48 h post-infection, RT-qPCR result showed that gga-miR-451 level was significantly increased in the CP-Ⅱ cells (Fig. 7A).
Interestingly, gga-miR-451 level is significantly decreased in the exosomes derived from MG-infected CP-Ⅱ cells (Fig. 6). showed that gga-miR-451 is significantly decreased in Exo-MG as compared with control group (Fig. 7C). Recently, we reported that gga-miR-451 decreased the inflammatory cytokine production, including TNF-α and IL-1β (16). We hypothesize mainly from a class of membrane-associated or organelle proteins ("membrane", "intracellular membrane-bounded organelle", "intracellular part", "cytoplasm") with small molecule binding and enzyme catalysis functions ("organic cyclic compound binding", "nucleic acid binding", "protein binding", "catalytic activity"). They are involved in the regulation of cellular process, metabolism, and protein modification ("cellular process", "metabolic process regulation of response to stimulus", "biological regulation"). The KEGG pathways is widely used to predict the gene function and genome information, which is helpful for us to study the metabolism function, genetic information processing, cellular processes and diseases (48). In Our results revealed that the concentration of exosomes derived from MG-infected CP-Ⅱ cells was increased as non-infected cells (Fig. 1E, Table S3),, which was similar to the results of previous reports in other diseases (35,50). there are only 5 differentially expressed miRNAs (gga-miR-365-3p, gga-miR-449b-5p, gga-miR-133c-3p, gga-miR-451, gga-miR-223) which are consistent between the results of this study and our previous results of miRNAs sequencing of MGinfected chicken lungs at 3 and 10 days post-infection (9). This study demonstrated that gga-miR-451 level was significantly increased in MG-infected CP-Ⅱ cells, which was shared by the results of our previous report in DF-1 cells (16). While the exosomal gga-miR-451 was significantly decreased in the exosomes derived from MG-infected CP-Ⅱ cells compared to control group ( Fig. 6A and 5). In non-small cell lung cancer, plasma exosomal miR-451a might serve as a reliable biomarker for prediction of recurrence and prognosis (57). It is reasonable to believe that gga-miR-451-desent exosomes derived from MG-infected CP-Ⅱ cells increases inflammatory cytokine production, including TNF-α and IL-1β.
Growing evidences have investigated YWHAZ protein (14-3-3ζ) in multiple systems and found that is a target gene and is negatively regulated by miR-451 in mammals and chickens (16,58). We also found that gga-miR-451-desent exosomes can regulate the expression of YWHAZ protein in DF-1 cells (Fig. 7). These results suggest that the exosomes derived from CP-Ⅱ cells may deliver less gga-miR-451 to DF-1 cells to induce inflammatory cytokines by targeting YWHAZ and increasing its expression. In addition to gga-miR-451, other exosomal gga-miRNAs may also regulate the immunity processes in MG infection. For example, exosomal miR-223 targets the transcription factor STAT5 and is implicated in inflammatory reactions in multiple sclerosis (59). The serum exosome derived microRNA-193 was reported to be a potential Alzheimer's Disease biomarker (60). However, the molecular mechanism of these exosomal miRNAs that shuttle and regulate chronic respiratory inflammation needs to be further studied.

Conclusions
The present study provides evidence to prove that the mode of intercellular communication between chicken cells is mediated by exosomes. About 30 exosome-microRNAs derived from CP-Ⅱ were significantly changed following MG infection. The    KEGG enrichment analysis of differentially expressed miRNAs in chicken type II pneumocytes Figure 5 Network of miRNA-target pathways involved in cell cycle, apoptosis and Toll-like receptor sig Figure 5 Network of miRNA-target pathways involved in cell cycle, apoptosis and Toll-like receptor sig Figure 6 The validation of the selected exosome-microRNA expression profile approach by using RT-qP Figure 6 The validation of the selected exosome-microRNA expression profile approach by using RT-qP