Small RNA Sequencing Reveals Differentially Expressed piRNAs Related to Hypoxia in Tibetan Chickens

Zengrong ZHANG, Xiaosong JIANG, Mohan QIU, Chunlin YU, Huarui DU, Qingyun LI, Han PENG, Xia XIONG, Xiaoyan SONG, Chenming HU, Jialei CHEN, Chaowu YANG

Abstract The relevance of genetic mechanism to the phenotype of hypoxic adaptation remains elusive. Tibetan chickens typically used to investigate the mechanism for the adaptation of hypoxia and the recognition of hypoxia-related piRNA remains an open issue. The purpose of this study was to illustrate whether the piRNAs were related to hypoxic adaptation. First of all， the differentially expressed piRNAs （DEpiRNAs） were identified through RNA sequencing between the Tibetan chickens and Daheng broilers. Subsequently， the target genes of DEpiRNAs were predicted and annotated by software. The network was constructed by Cytoscape. In our study， a total of 277 DEpiRNAs （33 down-regulated， 244 up-regulated） were identified in the Tibetan chickens compared with the Daheng broilers. All of the 277 DEpiRNAs predicted 36 658 targeted genes. Gene Ontology （GO） analysis showed that the target genes were significantly enriched in the biological process correlated with proliferation and apoptosis of cells， including cell cycle， mitochondrial outer membrane permeabilization， and positive regulation of stress-activated mitogen-activated protein kinase cascade. Kyoto Encyclopedia of Genes and Genomes （KEGG） analysis implicated that the DEpiRNAs were mainly involved in immune and metabolism， including natural killer cell-mediated cytotoxicity， toll-like receptor signaling pathway and fatty acid metabolism. Furthermore， a predicted network with four piRNAs acted on 11 pathways via interacting with 22 target genes， in which piR-gga-1368839 regulated metabolic pathways by acting on DHCR24. In conclusion， we determined the DEpiRNAs in the Tibetan chickens and found that these piRNAs were associated with metabolism， which may be favorable for researching the biological adaptation to hypoxic stress.

Key words piRNAs; Tibetan chicken; Hypoxic adaptation; RNA sequencing

Received： February 9， 2022  Accepted： April 8， 2022

Supported by Science and Technology Support Planning Project of Sichuan Province （2021YFYZ0031; 2020YFN0146; SASA2022CZYX002）; National Modern Agricultural Technology System Construction of China （CARS-41-G04）.

Zengrong ZHANG （1980-）， male， P. R. China， associate research fellow， devoted to research about poultry breeding and production.

*Corresponding author： E-mail： zhangzengrong2004@163.com; cwyang@foxmail.com.

Tibetan chicken is a native Chinese species， which distributes in the semi-agricultural and semi-pastoral areas with an altitude of 2.2-4.1 km. Through millennia of adaptation， Tibetan chickens have adapted to the condition of altitude hypoxia. Furthermore， the hatching ability of Tibetan chickens was higher than that of other chicken breeds[1]. It implied the special mechanism in Tibetan chickens for maintaining oxygen balance. Therefore， to study the mechanisms of hypoxic adaptation， the Tibetan chickens were utilized to discover the unique genes related to hypoxia.

Oxygen plays an important role in sustaining life activities and metabolism of the body. Hypoxia results in the abnormal changes in the metabolism， function， and morphological structure of the tissue. The hypoxic condition leads to a serial of poor effects， such as cardiovascular disease， rheumatoid arthritis， ischemic injury， and the metastasis and invasion of the tumor[2-6]. Hypoxia exaggerated the risk of the cardiovascular demise of the fetus. It also resulted in reducing bone density， increasing bone resorption and bone brittleness. In solid tumors， hypoxia promoted cancer invasion and metastasis by increasing the expression of hypoxia-inducible factor-1α （HIF-1A）. Moreover， hypoxia frequently activates HIF-1A to promote the progression and metastasis of the tumor[7]. HIF is a core regulator in inducing hypoxic genes and repairing of the oxygenated environment of cell[8]. HIF signaling and its cascading pathways played a key role in hypoxic adaptation[9]. Exploring the molecular mechanisms related to hypoxia is of great importance for understanding the hypoxia-related disease.

With the extensive application of bioinformatics， exploring and excavating the differentially expressed （DE） non-coding RNAs （ncRNAs） in Tibetan chicken and illustrating the molecular mechanism related to hypoxic became possible. ncRNA includes miRNAs， small interfering RNAs and PIWI-interacting small RNA （piRNA）. piRNA is originally discovered in the small profile of Drosophila. And subsequently， it is discovered in the somatic cells of diverse organisms. Most of the study about piRNA has focused on the reproductive system， hepatopathy and cancer[10-12]. A previous study showed piRNAs were abnormally elevated in the hypoxic pulmonary hypertension model， compared with the control group[13]. Wang et al.[14-15] found that the expression of piRNA-63076 increased under hypoxic conditions， and the up-regulation was connected with cell proliferation. piRNA-823 aggravated the glucose consumption of carcinoma cells by inhibiting the ubiquitination of HIF-1A[16]. The knowledge about piRNA in hypoxia is a limitation. In the hypoxic animal model， the discovery of piRNAs may be a promising strategy to investigate the hypoxic adaptation[13]. Given these， we believed that piRNA maybe play a key role in hypoxia-related molecular processes. However， the expression of piRNA in Tibetan chickens （hypoxic adapted species） is still unclear. It is necessary to explore the DEpiRNAs in Tibetan chickens.

Our previous study proved that DElncRNAs could regulate target gene expression and contribute to hypoxic adaption in Tibetan chickens[17-18]. Simultaneously， our previous study showed that DEmiRNAs were correlated with hypoxic adaptation， especially gga-miR-34c-5p. In this study， we selected Tibetan chicken as the hypoxic adaptive samples and Daheng broilers as the control samples. The DEpiRNAs were screened by small RNA sequencing. The Gene Ontology （GO） and Kyoto Encyclopedia of Genes and Genomes （KEGG） analysis were used to display the enrichment of functional and metabolic pathways of DEpiRNA. We provided further insights into the DEpiRNAs within Tibetan chickens and Daheng broilers by analyzing the expression of piRNAs.

Materials and Methods

Sample collection

The Tibetan chickens and Daheng broilers （22-55 weeks old） were purchased from Mao County Jiuding Original Ecological Livestock and Poultry Breeding Co. Ltd （Aba Autonomous Prefecture， Sichuan Province， China）. The chickens were given food and water randomly. All of our actions about animals were approved and authorized by the animal care and ethical committee of Sichuan Animal Science Academy. The eggs of Tibetan chickens and Daheng broilers within 7 d were collected and incubated for 16 d. Eggs from Tibetan chickens （n=120） were incubated in a hypoxic incubator with 37.5 ℃ and relative humidity of 65%， whereas the eggs of Daheng broilers （n=100） were in a normoxic incubator. Then， the embryonic heart tissue was obtained from chicken and frozen in the liquid nitrogen for further study.

RNA isolation， small RNA libraries construction， and sequencing

The total RNA was obtained from the embryonic heart tissues of chicken by the RNeasy Mini Kit （Qiagen， Hilden， Germany）. The quality of RNA was detected by the 1% agarose gel electrophoresis. The concentration and integrity of total RNA were measured by the NanoDrop ND-1000 spectrophotometer （Thermo Fisher Scientific， Waltham， USA） and Agilent 2100 Bioanalyzer （Agilent Technologies， Santa Clara， USA）. Eight independent small RNA libraries of Tibetan chickens （TC， n=4） and low altitude chickens （LC， n=4） were constructed by the manufacturers protocol. Following passing the quality control tests， the Multiplex Small RNA Library Prep Set for Illumina （NEB， MA， USA） was leveraged to establish the library. Briefly， the rRNA was removed and the remaining RNA was cut into small pieces. RNase H-reverse transcriptase （NEB， MA， USA） was used to synthesize the first-strand cDNA and the end repair was conducted. Then， the fragments were sorted and PCR was performed to amplify the cDNA. An Illumina Hiseq2500 platform （Illumina， San Diego， California） was used for the sequencing of library preparations.

Identification of DEpiRNA

The raw reads were processed by Fast-QC （http：∥www.bioinformatics.babraham.ac.uk/projects/fastqc/）. The BWA algorithm was used to map the chicken genome and miRBase Version 21 （http：∥www.mirbase.org/） were leveraged to annotate the piRNA. The DE-Seq 2.0 algorithm was utilized to analyze the DEpiRNAs between Tibetan chickens and Daheng broilers with |log2 fold change|>1 and P<0.05 as the inclusion criteria.

Target prediction and annotation

RNAHybrid （http：∥bibiserv.techfak.uni-bielefeld.de/rnahybrid/） and miRanda （http：∥www.microrna.org/microrna/home.do） were used to predict the targeted genes of piRNAs， and miRanda predicted the target genes of DEpiRNAs with Score>=150 and Energy<-20 as the inclusion criteria. Whereas， inclusion criteria of target genes were Energy<-25 in RNAhybrid. Ultimately， the target genes were determined by the intersection of the two databases. GO （http：∥www.geneontology.org） enrichment analysis was performed to annotate the target genes from the respect of cellular component， molecular function and biological process. KEGG （http：∥www.genome.jp/kegg） database was used to analyze the pathway of target genes.

Construction of the piRNA-mRNA-pathway network

To display the relevance of piRNA expression to the regulation of mRNA， a network of piRNA-mRNA-pathway was established. HIF is widely recognized as a major regulator of hypoxic response. In this study， we screened HIF-1A related piRNAs by Target prediction. Moreover， the corresponding target mRNAs of those piRNAs and the enriched pathways of these mRNAs were shown in the network diagram. Cytoscape software 3.6.1 （https：//cytoscape.org/） was utilized to draw the piRNA-target genes-pathway interaction network.

Results and Analysis

Overall of piRNAs profile in Tibetan chickens and Daheng broilers

The sequencing data were shown in Table 1 to display the reliability of data. The clean reads acquired form the Tibetan chickens ranged from 6 173 789 to 8 237 479， and the clean reads of Daheng broilers ranged from 6 224 495 to 7 338 193. Simultaneously， the mapping rates of eight samples were all over 70%. These suggested that the clean reads were well mapped to the reference genome （Table 1）. And， the quality of the data was good enough for subsequent analysis.

Identification of DEpiRNAs in Tibetan chickens

To identify the potential DEpiRNAs related to hypoxia in Tibetan chickens， we performed DESeq2.0. A total of 277 DEpiRNAs were screened between Tibetan chickens and Daheng broilers， as shown in the Volcano plot （Fig. 1A）. Among them， 33 piRNAs were down-regulated， whereas 244 piRNAs were up-regulated in the Tibetan chickens compared to Daheng broilers. A heat map was utilized to analyze the expressed pattern of DEpiRNAs and the results showed that DEpiRNAs were significantly separated in the two groups （Fig. 1B）.

Functional and pathway analysis of the predicted targeted genes of DEpiRNAs

RNAHybrid and miRanda were used to screen the target genes of DEpiRNAs. A total of 36 658 target genes were obtained from the intersection of the two databases （Fig. 2A）. To explore the potential of DEpiRNAs involved in the hypoxic adaptability of Tibetan chickens， we used the GO and KEGG enrichment analysis for annotating the function of DEpiRNAs. As illustrated in Fig. 3A， GO term analysis revealed that target genes were enriched in the biological process related to hypoxic adaptation， such as cell cycle， mitochondrial outer membrane permeabilization， and positive regulation of stress-activated mitogen-activated protein kinase cascade （MAPK） cascade. As illustrated in previous studies， hypoxia-induced the proliferation of pulmonary arterial smooth muscle cells （PASMCs） were inhibited by lincRNA-COX2 via affecting the G2/M phase of the cell cycle[19]. Hypoxia-ischemia （HI） aggravated mitochondrial outer membrane permeabilization[20]. Hypoxia upregulated EGFR1 and drug resistance in cancer cells through the MAPK pathway[21]. KEGG analysis demonstrated that DElncRNAs were mainly enriched in pathways associated with hypoxic adaptation， including natural killer cell-mediated cytotoxicity， toll-like receptor signaling pathway， fatty acid metabolism， and amino sugar and nucleotide sugar metabolism. Larimichthys crocea regulated the hypoxia-induced immune response through participating in B cell receptor signaling pathway， natural killer cell-mediated cytotoxicity and NF-κB signaling pathway[22]. Inhibiting the TLR4/NF-κB/STAT3 pathway relieved neonatal hypoxic-ischemic brain damage[23]. HIF-1A increased fatty acid synthesis and inhibited fatty acid oxidation in colon cancer[24]. These results gave further insight into the view that Tibetan chickens maybe regulate their metabolism to adapt to the hypoxic environment （Fig. 3B）.

The network of DEpiRNA-mRNA-pathway

To further visualize the relevance of DEpiRNAs to their target mRNAs， we established the DEpiRNA-mRNA-pathway network. Because of the crucial effect of HIF-1A in hypoxia， we identified four piRNAs targeted HIF-1A （detailed information was shown in Fig. 2B）. As shown in Fig. 4， piR-gga-80861， piR-gga-1368839， piR-gga-127600 and piR-gga-1440613 regulated the HIF-1A through the mTOR signaling pathway in Tibetan chickens. Previous studies demonstrated that HIF-1A play a critical role in regulating hypoxic adaptation in humans[25]. DHCR24 and TOLLIP affect the death of cardiomyocytes elicited by hypoxia[26-27]. piR-gga-80861 targeted the TOLLIP by the toll-like receptor pathway. DHCR24 interacted with piR-gga-80861 and piR-gga-1368839 via the metabolic pathways or steroid biosynthesis pathway. Therefore， we speculated that these four DEpiRNAs maybe participate in the hypoxic adaptation of Tibetan chickens by regulating these pathways via target genes.

Conclusions and Discussion

Oxygen participates in maintaining life activities and metabolism， which is necessary for survival. Hypoxia is unfavorable for the progression of disease. Tibetan chickens live in high-altitude environments （hypoxic） due to stable genetic information of hypoxic adaptation. Nevertheless， the underlying molecular mechanisms of hypoxia adaptation remain ambiguous. Genomics technologies have been extensively used to illustrate the relationship between adaptive phenotype and genetic information in plants， animals and humans[28-30]. In our study， we performed RNA-seq to identify the DEpiRNAs between Tibetan chickens and Daheng broilers. Among them， 33 down-regulated and 244 up-regulated DEpiRNAs were identified in the Tibetan chickens， compared with the Daheng broilers. And these piRNAs were significantly concentrated in the biological process， including cell cycle， mitochondrial outer membrane permeabilization， and positive regulation of stress-activated mitogen-activated protein kinase cascade.

piRNAs mediated the degradation of mRNA and maintaining the stability of the genome[31]. piRNA is differently expressed in many diseases and exert different biological functions， such as fertility， neurodegenerative disease， cardiovascular diseases， and cancer progression. There were 32% of the preferentially female piRNAs and 93% of the male-identified piRNAs[32]. A previous study showed approximately 100 DEpiRNAs in Alzheimers disease patients， and the up-regulation of piRNAs may result in the cell apoptosis and neurodegeneration[33]. hsa-piR-020009 and hsa-piR-006426 were increased in heart failure patients， whereas most DEpiRNAs were decreased[34]. The down-regulation of piR-001773 and piR-017184 significantly inhibited the progress of prostatic cancer[35]. In our study， we discovered DEpiRNAs in the tissues of Tibetan chickens and Daheng broilers. We suspected that these piRNAs were related with hypoxic adaptation.

HIF pathway drove the low-oxygen tolerance by the coordinate shift in metabolism and the activation of the transcriptional program[36]. The activity of HIF-1 is determined by the oxygen regulatory subunit HIF-1A， which plays a critical role in regulating the gene expression induced by hypoxia. In this study， we identified four DEpiRNAs （piR-gga-80861， piR-gga-1368839， piR-gga-127600， and piR-gga-1440613） by targeting HIF-1A. And then， we found that the four DEpiRNAs regulated the mTOR signaling pathway， metabolic pathways， insulin signaling pathway， and toll-like receptor signaling pathway via acting on target genes. In traumatic brain injury， hypoxia increased the remyelination of bone marrow stromal cells through the mTOR/HIF-1A/VEGF pathway; mTOR inhibitor rapamycin hinder the effect of hypoxia. Hypoxia-induced cognitive impairment was reported to be link with the dysfunction of the mTOR signaling pathway[37]. Accordingly， we considered that the mTOR signaling pathway participate in the hypoxic environment. Similarly， hypoxia interfered with fatty acid metabolism and insulin signaling. Under the hypoxic condition， enhancing fatty acid metabolism increased the endurance performance of mice. The synthesis of unsaturated fatty acids was increased in rat cardiomyocytes. On the other hand， a previous study revealed that hypoxia is sensed by fat tissue via the HIF-1A prolyl hydroxylase. Meanwhile， fat inhibited insulin secretion of the brain by releasing HIF-1A-dependent humoral factors， thereby restricting systemic growth. Toll-like receptor signaling pathway is correlated with inflammation regulation and immune function. Hypoxia and HIF-1A-induced inflammation via provoking toll-like receptor signaling in rheumatoid arthritis[38]. Another study revealed that ablating toll-like receptor 4 improved renal injury， inflammation， and fibrosis after hypoxic injury[39]. Given this， we thought that DEpiRNAs regulated the hypoxic adaptation of Tibetan chickens by the above signaling pathways.

In summary， we analyzed the piRNAs profiles of Tibetan chicken. Our study implicated DEpiRNAs may be one of the genetic mechanisms contributing to hypoxic adaption in Tibetan chickens. Further studies are needed to verify the specific piRNA and clarify the mechanism of action.

References

[1] LIU C， ZHANG L， SONG M， et al. Highly efficient dissociation of oxygen from hemoglobin in Tibetan chicken embryos compared with lowland chicken embryos incubated in hypoxia[J]. Poult Sci， 2009， 88（12）： 2689-2694.

[2] ALLISON BJ， BRAIN KL， NIU Y， et al. Altered cardiovascular defense to hypotensive stress in the chronically hypoxic fetus[J]. Hypertension， 2020： A12015384.

[3] HU F， MU R， ZHU J， et al. Hypoxia and hypoxia-inducible factor-1alpha provoke toll-like receptor signalling-induced inflammation in rheumatoid arthritis[J]. Annals of the Rheumatic Diseases， 2014， 73（5）： 928-936.

[4] SONG J， CHEN W， ZHU G， et al. Immunogenomic profiling and classification of prostate cancer based on HIF-1 signaling pathway[J]. Frontiers in Oncology， 2020（10）： 1374.

[5] HAMAD HA， ENEZEI HH， ANMAR A， et al. Identification of potential chemical substrates as fuel for hypoxic tumors that may be linked to invadopodium formation in hypoxia-induced MDA-MB-231 breast-cancer cell line[J]. Molecules， 2020， 25（17）： 3876.

[6] COLLINO F， LOPES JA， STEPHANY C， et al. Adipose-derived mesenchymal stromal cells under hypoxia： Changes in extracellular vesicles secretion and improvement of renal recovery after ischemic injury[J]. Cellular Physiology and Biochemistry， 2019， 52（6）： 1463-1483.

[7] SANHUEZA C， BENNETT JC， MANUEL V， et al. Caveolin-1-mediated tumor suppression is linked to reduced HIF1alpha s-nitrosylation and transcriptional activity in hypoxia[J]. Cancers （Basel）， 2020， 12 （9）： 2349.

[8] LEE KE， SIMON MC. SnapShot： Hypoxia-inducible factors[J]. CELL， 2015， 163（5）： 1288.

[9] SIMONSON TS， YANG Y， Chad DH， et al. Genetic evidence for high-altitude adaptation in Tibet[J]. Science， 2010， 329（5987）： 72-75.

[10] GUO B， LI D， DU L， et al. piRNAs： Biogenesis and their potential roles in cancer[J]. Cancer Metastasis Rev， 2020， 39（2）： 567-575.

[11] WU PH， FU Y， KATHARINE C， et al. The evolutionarily conserved piRNA-producing locus pi6 is required for male mouse fertility[J]. Nature Genetics， 2020， 52（7）： 728-739.

[12] XU J， YANG X， ZHOU Q， et al. Biological significance of piRNA in liver cancer： A review[J]. BIOMARKERS， 2020， 25（6）： 436-440.

[13] MA C， ZHANG L， WANG X， et al. piRNA-63076 contributes to pulmonary arterial smooth muscle cell proliferation through acyl-CoA dehydrogenase[J]. Journal of Cellular and Molecular Medicine， 2020， 24（9）： 5260-5273.

[14] WANG J， ZHANG P， LU Y， et al. piRBase： A comprehensive database of piRNA sequences[J]. Nucleic Acids Research， 2019， 47（D1）： D175-D180.

[15] ZHANG P， SI X， LIU P， et al. piRBase： A web resource assisting piRNA functional study[J]. Database （Oxford）， 2014（2014）： u110.

[16] FENG J， YANG M， WEI Q， et al. Novel evidence for oncogenic piRNA-823 as a promising prognostic biomarker and a potential therapeutic target in colorectal cancer[J]. Journal of Cellular and Molecular Medicine， 2020， 24（16）： 9028-9040.

[17] ZHANG Z， QIU M， DU H， et al. Identification of long noncoding RNAs involved in adaptability to chronic hypoxic by whole transcriptome sequencing[J]. 3 Biotech， 2020， 10（6）： 269.

[18] ZHANG Z， QIU M， DU H， et al. Small RNA sequencing reveals miRNAs important for hypoxic adaptation in the Tibetan chicken[J]. Br Poult Sci， 2020： 1-8.

[19] CHENG G， HE L， ZHANG Y， et al. LincRNA-Cox2 promotes pulmonary arterial hypertension by regulating the let-7a-mediated STAT3 signaling pathway[J]. Molecular and Cellular Biochemistry， 2020， 475（1-2）： 239-247.

[20] BABURAMANI AA， SOBOTKA KS， REGINA V， et al. Effect of Trp53 gene deficiency on brain injury after neonatal hypoxia-ischemia[J]. Oncotarget， 2017， 8（7）： 12081-12092.

[21] LU Y， LIU Y， SEBASTIAN O， et al. Hypoxia induces resistance to EGFR inhibitors in lung cancer cells via upregulation of FGFR1 and the MAPK pathway[J]. Cancer Research， 2020， 80（21）： 4655-4667.

[22] MU Y， LI W， WU B， et al. Transcriptome analysis reveals new insights into immune response to hypoxia challenge of large yellow croaker （Larimichthys crocea）[J]. Fish Shellfish Immunol， 2020（98）： 738-747.

[23] ZHOU Y， WANG S， ZHAO J， et al. Asiaticoside attenuates neonatal hypoxic-ischemic brain damage through inhibiting TLR4/NF-kappaB/STAT3 pathway[J]. Ann Transl Med， 2020， 8（10）： 641.

[24] NI T， HE Z， DAI Y， et al. Oroxylin A suppresses the development and growth of colorectal cancer through reprogram of HIF1alpha-modulated fatty acid metabolism[J]. Cell Death & Disease， 2017， 8（6）： e2865.

[25] MURRAY AJ， MONTGOMERY HE， MATTIN F， et al. Metabolic adjustment to high-altitude hypoxia： From genetic signals to physiological implications[J]. Biochem Soc Trans， 2018， 46（3）： 599-607.

[26] WAN N， LIU X， ZHANG X， et al. Toll-interacting protein contributes to mortality following myocardial infarction through promoting inflammation and apoptosis[J]. Br J Pharmacol， 2015， 172（13）： 3383-3396.

[27] HAN F， CHEN Q， SU J， et al. MicroRNA-124 regulates cardiomyocyte apoptosis and myocardial infarction through targeting Dhcr24[J]. Journal of Molecular and Cellular Cardiology， 2019（132）： 178-188.

[28] KAKIOKA R， KOKITA T， HIROKI K， et al. A RAD-based linkage map and comparative genomics in the gudgeons （genus Gnathopogon， Cyprinidae）[J]. Bmc Genomics， 2013（14）： 32.

[29] PENG J， LI Q， XU L， et al. Chromosome-level analysis of the Crassostrea hongkongensis genome reveals extensive duplication of immune-related genes in bivalves[J]. Molecular Ecology Resources， 2020， 20（4）： 980-994.

[30] ZHANG J， MA J， ZHOU X， et al. Comprehensive analysis of mRNA and lncRNA transcriptomes reveals the differentially hypoxic response of preadipocytes during adipogenesis[J]. Frontiers in Genetics， 2020（11）： 845.

[31] WATANABE T， CHENG EC， ZHONG M， et al. Retrotransposons and pseudogenes regulate mRNAs and lncRNAs via the piRNA pathway in the germline[J]. Genome Research， 2015， 25（3）： 368-380.

[32] BILLI AC， FREEBERG MA， Amanda MD， et al. A conserved upstream motif orchestrates autonomous， germline-enriched expression of Caenorhabditis elegans piRNAs[J]. PLoS Genetics， 2013， 9（3）： e1003392.

[33] QIU W， GUO X，YANG Q， et al. Transcriptome-wide piRNA profiling in human brains of Alzheimers disease[J]. Neurobiology of Aging， 2017（57）： 170-177.

[34] YANG J， XUE FT， LIU W， et al. Exosomal piRNA sequencing reveals differences between heart failure and healthy patients[J]. Eur Rev Med Pharmacol Sci， 2018， 22（22）： 7952-7961.

[35] ZHANG L， MENG X， LI D， et al. piR-001773 and piR-017184 promote prostate cancer progression by interacting with PCDH9[J]. Cellular Signalling， 2020： 109780.

[36] YUAN X， LUO Q， SHEN L， et al. Hypoxic preconditioning enhances the differentiation of bone marrow stromal cells into mature oligodendrocytes via the mTOR/HIF-1alpha/VEGF pathway in traumatic brain injury[J]. Brain and Behavior， 2020， 10（7）： e1675.

[37] ABDO QE， ZULKIPLI NN， RAHIMAH Z， et al. The role of mTOR signalling pathway in hypoxia-induced cognitive impairment[J]. International journal of neuroscience， 2020： 1-7.

[38] HU F， MU R， ZHU J， et al. Hypoxia and hypoxia-inducible factor-1alpha provoke toll-like receptor signalling-induced inflammation in rheumatoid arthritis[J]. Annals of the Rheumatic Diseases， 2014， 73（5）： 928-936.

[39] ZHANG Y， SU X， ZOU F， et al. Toll-like receptor-4 deficiency alleviates chronic intermittent hypoxia-induced renal injury， inflammation， and fibrosis[J]. Sleep and Breathing， 2019， 23（2）： 503-513.