Abstract

   Orexin-A is a neuropeptide with potent neuroprotective activity towards
   cerebral ischemia-reperfusion (I/R) injury, but few studies have
   attempted to elucidate the mechanism. Herein, we performed global gene
   expression profiling of the hippocampus following reperfusion with
   Orexin-A using RNA sequencing (RNA-seq). RNA-seq identified 649
   differentially expressed genes (DEGs) in the Orexin-A group compared
   with saline controls (I/R group), of which 149 were up-regulated and
   500 were down-regulated. DEGs were confirmed using qRT-PCR, their
   molecular functions, biological processes and molecular components were
   explored using Gene Ontology (GO) analysis and 206 KEGG pathways were
   associated with Orexin-A treatment. MAPK, chemokine and calcium
   signalling pathways were mainly responsible for the neuroprotective
   effects of Orexin-A. Hspb1, Igf2 and Ptk2b were selected for functional
   interaction analysis by GeneMANIA. The results suggest that Orexin-A
   modifies gene expression in the hippocampus, leading to neuroprotection
   from I/R injury. The study provides a basis for future elucidation of
   the molecular mechanisms underlying Orexin-A.

   Keywords: Orexin-A, ischemia-reperfusion injury, RNA sequencing,
   differential gene expression, gene network

INTRODUCTION

   Orexin-A is a novel neuropeptide involved in the regulation of feeding
   behaviour [[34]1, [35]2], energy metabolism [[36]3], hormone secretion
   [[37]4], the sleep-wake cycle [[38]5] and anaesthesia [[39]6]. Orexin-A
   plays a pivotal role in neurotrauma [[40]7], Parkinson's disease
   [[41]8], narcolepsy [[42]9], vascular diseases [[43]10] and brain
   tumours [[44]11, [45]12]. In addition, Orexin-A protects against
   ischemia-reperfusion (I/R) injury by reducing gastric lesions through
   increasing HO-2 expression and decreasing HO-1 expression [[46]13].
   After intestinal I/R injury, hypothalamus Orexin-A levels gradually
   decrease from 30 to 150 min of reperfusion, and then recover slowly
   [[47]14]. Orexin-A levels change during hepatic reperfusion, indicating
   involvement in the regulation of liver injury induced by this process
   [[48]15].

   Similarly, Orexin-A has a prominent neuroprotective effect on cerebral
   I/R injury. Kitamura et al. found that the number of neurons expressing
   Orexin-A was significantly lower in the non-ischemic side than the
   ischemic side [[49]16]. Additionally, they found Orexin-A pre-treatment
   significantly reduced the brain infarction volume induced by
   reperfusion injury, ameliorated neurologic deficit scores and reduced
   the infarction volume after rat cerebral I/R injury. The
   neuroprotective effect of Orexin-A was shown to be mediated by an
   increase in hypoxia-inducible factor-1 (HIF-1α) activity [[50]17], and
   our lab demonstrated that Orexin-A protects SHY-5Y cells against
   oxidative stress by activating the PI3K/AKT signalling pathway.

   Few studies have investigated the neuroprotective effects of Orexin-A
   in I/R injury. Therefore, in the present study we performed global gene
   expression profiling in the hippocampus after reperfusion with Orexin-A
   by RNA sequencing (RNA-seq). We identified differentially expressed
   genes (DEGs) and signalling pathways putatively associated with
   Orexin-A reperfusion. The findings provide a basis for further
   elucidation of the molecular mechanisms underlying the neuroprotective
   effects of Orexin-A.

RESULTS

Orexin-A decreases the infarct volume after cerebral I/R injury

   Infarction volumes in model rats were measured using
   triphenyltetrazolium chloride (TTC) staining. As shown in Figure [51]1,
   normal tissues were stained red, while infarction areas were stained
   white. No infarction area was identified in the sham group, while the
   I/R group displayed a significantly higher rate of infarction (30.620%
   ± 2.665%) than the sham group. In addition, the Orexin-A group
   displayed a rate of infarction that was 12.497% ± 1.912% lower than
   that displayed by the I/R group (^**p < 0.01), indicating that
   pre-treatment with Orexin-A significantly decreased the infarction
   area, and Orexin-A therefore exhibited a clear neuroprotective effect
   on I/R injury.

Figure 1. Orexin-A decreases the cerebral infarction volume following middle
cerebral artery occlusion (MCAO) in rats.

   [52]Figure 1
   [53]Open in a new tab

   (A) Representative images from sham and MCAO groups treated with or
   without Orexin-A after a 24 h reperfusion using TTC staining. Brain
   sections from the sham group are red, and infarction volumes are not
   visible. Infarction volumes in the I/R group are white and obviously
   increased in size compared with the sham group. Reperfusion of Orexin-A
   significantly decreases the infarction volume compared with the I/R
   group. (B) Percentage of cerebral infarction volumes to total brain
   volumes. Columns represent infarction volumes as a percentage of total
   volumes. Data are expressed as mean ± SD (n = 6). ^**p < 0.01 vs. the
   I/R group.

Global analysis of RNA-seq data

   Clean reads from RNA-seq were filtered, and 14,199,598, 13,499,205 and
   13,116,887 total reads were acquired in the sham, I/R and Orexin-A
   groups, respectively. Analysis of global gene expression is shown in
   Figure [54]2. About 89% of reads were successfully mapped, 11% were
   unmapped, 71% were perfectly matched with reference sequences and
   17-18% were mismatched (Figure [55]2A). Figure [56]2B shows the number
   of common and unique genes in the sham, I/R and Orexin-A groups. A
   total of 12,457 genes were found to be commonly expressed in all three
   groups, and 221 genes were specific to the sham sample, 294 were unique
   to the I/R group and 206 were expressed only in the Orexin-A group.

Figure 2. Analysis of global gene expression using RNA-seq data.

   [57]Figure 2
   [58]Open in a new tab

   (A) Summary of RNA-seq reads mapped. The y-axis represents the
   percentage of mapped reads or genes. (B) Venn diagrams showing the
   number of commonly and uniquely expressed genes among the three
   treatment groups.

Screening of DEGs

   Genes were considered differentially expressed when FDR ≤0.001 and
   |log2 ratio| >1. As shown in Figure [59]3A, 156 genes were up-regulated
   and 26 genes were down-regulated in the sham group compared with the
   I/R group. After reperfusion with Orexin-A, 649 genes were
   differentially expressed compared with the I/R group, of which 149 were
   up-regulated and 500 were down-regulated. In addition, there were 186
   up-regulated genes and 353 down-regulated genes in the Orexin-A group
   compared with the sham group. Figure [60]3B shows the number of DEGs
   and the fold change in expression between the Orexin-A and I/R groups.
   The absolute fold change (log2 ratio) ranged from 1 to 17.13. Figure
   [61]3C shows the expression levels of DEGs in the I/R and Orexin-A
   groups. Nine genes (Glycam1, Hfe2, Hmgn5, Lect1, LOC100911576, Mybl2,
   Scgb1c1, Tmem72 and Tmem27) not expressed in the I/R group were
   significantly up-regulated after reperfusion with Orexin-A. Eight genes
   (Hmgn5b, LOC100362172, Upk1b, Tmprss5, Msln, Pcdha1, Pcdh11x and Hdx)
   expressed in the I/R group were not expressed at all after reperfusion
   with Orexin-A. Additionally, 632 genes were detected in both groups. A
   selection of DEGs between the Orexin-A and I/R groups is listed in
   Table [62]1. These DEGs are likely to be involved in the
   neuroprotective effects of Orexin-A, and were further investigated.

Figure 3. Diagram of differentially expressed genes.

   [63]Figure 3
   [64]Open in a new tab

   (A) Number of up- and down-regulated genes in each group. (B) Scatter
   plot displaying differentially expressed genes between the I/R and
   Orexin-A groups. The x- and y-axes show the fold change (log2 ratio) in
   gene expression. Yellow points represent up-regulated genes with a fold
   change less than 1 and p < 0.05; blue points denote down-regulated
   genes with a fold change greater than 1 and p < 0.05. Brown points
   represent genes with no significant differences. (C) Venn diagrams
   showing the number of commonly and uniquely differentially expressed
   genes between the I/R and Orexin-A groups. Nine genes were expressed
   only in the I/R group, and eight genes were expressed only after
   reperfusion with Orexin-A.

Table 1. List of 20 differentially expressed genes in Orexin-A and I/R
groups.

   Up-regulated genes
   Gene symbol Description Log2 ratio p-value
   Scgb1c1 secretoglobin, family 1C, member 1 13.8946 1.99E-08
   Hmgn5 high mobility group nucleosome binding domain 5 13.5781 7.09E-29
   Tmem72 transmembrane protein 72 13.3148 2.53E-41
   Tmem27 transmembrane protein 27 11.8455 9.26E-06
   Lect1 leukocyte cell derived chemotaxin 1 11.7313 2.37E-06
   Hfe2 hemochromatosis type 2 10.7228 7.17E-05
   Mybl2 myeloblastosis oncogene-like 2 9.8765 7.17E-05
   Ttr transthyretin 6.7934 1.20E-06
   Psca prostate stem cell antigen 5.1408 4.18E-10
   Cdh3 cadherin 3 4.6621 1.64E-19
   Fibcd1 fibrinogen C domain containing 1 1.4395 2.14E-73
   Down-regulated genes
   Gene symbol Description Log2 ratio p-value
   LOC100362172 LRRGT00112-like -12.6795 1.27E-05
   Upk1b uroplakin 1B -10.9729 5.18E-05
   Tmprss5 transmembrane protease, serine 5 -10.7566 1.27E-05
   Pcdha1 protocadherin alpha 1 -10.3106 1.11E-08
   Pcdh11x protocadherin 11 X-linked -9.7813 2.56E-05
   Hdx highly divergent homeobox -9.7813 5.18E-05
   Shox2 short stature homeobox 2 -6.2204 1.82E-21
   Bcl3 B-cell CLL/lymphoma 3 -4.9307 4.51E-16
   Aspg asparaginase -4.8379 4.25E-08
   Socs3 suppressor of cytokine signalling 3 -4.0020 1.20E-11
   Hspb1 heat shock protein B1 -2.6018 1.48E-58
   Slc24a2 solute carrier family 24, member 2 -1.20343 1.64E-05
   [65]Open in a new tab

qRT-PCR validation

   qRT-PCR was performed to validate selected DEGs identified by RNA-seq.
   The relative expression levels of eight DEGs were determined using a
   LightCycler 480 (LC480). The sequences of primers were listed in Table
   [66]2. As shown in Figure [67]4, the relative expression levels
   determined by qRT-PCR were in good agreement with the RNA-seq data.
   Fives genes (Enpp2, Slc24a2, Pk2b, Cacna1e and Hspb1) were up-regulated
   in the I/R group compared with the sham group (^**p < 0.01, ^***p <
   0.001), and all were down-regulated following pre-treatment with
   Orexin-A compared with the I/R group (^$p < 0.05, ^$$p < 0.01). Three
   genes (Fibcd1, Wfs1 and Ttr) were down-regulated at 24 h after
   reperfusion compared with the sham group (^**p < 0.01), whereas
   Orexin-A reperfusion clearly increased their expression compared with
   the I/R group (^$p < 0.05, ^$$p < 0.01). The results confirmed the
   reliability of the RNA-seq data and further indicated the involvement
   of the identified DEGs in the neuroprotective effects of Orexin-A.

Table 2. Sequences of primers used for amplification.

   Primer name Forward sequence (5’-3’) Reverse sequence (5’-3’)
   β-actin
   Enpp2 tggaatcctgtggcatccatgaa
   cccttcagtccgagtttgacc taaaacgcagctcagtaacagtccg
   gccgtccatacaggagatgt
   Slc24a2
   Ptk2b ccgaggaagatgatgaccag
   gccgagttcaagcagatcag tccagagagggaacacgatg
   aggtcttgacggccacatta
   Cacna1e
   Hspb1 gcccttaaagctcgtgtcag
   acgaagaaaggcaggatgaa gcattggaagacagtcagca
   gacagggaagaggacaccag
   Fibcd1 ggtgtgggcttgttctctgt gttctctgaatggtcgctgtc
   Wfs1 cagccgagaagggacagata ggcaagtcgcaggtagtgtt
   Ttr ctgctggaaagcctggaa cactgctctgctcctcttca
   [68]Open in a new tab

Figure 4. Validation of a selection of differentially expressed genes using
qRT-PCR.

   [69]Figure 4
   [70]Open in a new tab

   (A-B) Fives genes were up-regulated and three were down-regulated in
   the I/R group compared with the sham group. Treatment with Orexin-A
   reversed these expression changes. ^**p < 0.01, ^***p < 0.001 vs. sham
   group; ^$p < 0.05, ^$$p < 0.01 vs. I/Rgroup.

GO annotation of DEGs

   GO annotation was used to further investigate the functions of the
   identified DEGs. As shown in Figure [71]5, DEGs were annotated into
   three categories: molecular function, biological process and cellular
   component. The molecular function category was divided into eight main
   subcategories, and the top four were binding (GO:0005488), catalytic
   activity (GO:0003824), receptor activity (GO:0004872) and transporter
   activity (GO:0005215; Figure [72]5A). The top hit in the biological
   process category was cellular process (GO:0009987), metabolic process
   (GO:0044238), biological regulation (GO:0065007) and response to
   stimulus (GO:0050896) were also well represented (Figure [73]5B). Cell
   part (GO:0044464), membrane (GO:0016020), organelle (GO:0043226) and
   macromolecular complex (GO:0032991) were the main cellular component
   subcategories (Figure [74]5C). A selection of DEGs classified into
   different subcategories is listed in Table [75]3.

Figure 5. PANTHER classification analysis of differentially expressed genes
between I/R and Orexin-A groups.

   [76]Figure 5
   [77]Open in a new tab

   (A) Molecular function categories. (B) Biological process categories.
   (C) Cellular component categories.

Table 3. GO classification of differentially expressed genes.

   Biological process
   GO Term Symbol
   GO:0009987: cellular process Arhgap5, Cobl, Ube2l6, Aqp4, Nrxn1,
   Inpp5f, Fmr1, Casq2, Capza1, Fras1, Upp1, Clock, Enpp2, Dis3, Aspg;
   GO:0044238: metabolic process Klk8, Cyp2d4, Eif1a, Cfap52, Plcxd3,
   Lnpep, Eri1, Prkab2, Sult1d1, Gnpda2, Hmgn5, Cdkl1, Met, Pcmtd1, Dbp;
   GO:0065007: biological regulation Oas1i, Gabrb2, Slc17a6, Cd38, Gdf10,
   Zic1, Taok1, Sostdc1, Dnm1l, Jak2, Bdnf, Otx2, Timp1, Kcne2, Cd44;
   GO:0050896: response to stimulus Klk8, Stc1, Grik2, Gabra1, Ago4,
   Shox2, Ucp2, Cdc73, Stat3, Necab2, Eltd1, Prkacb, Crem, Shc3, Lcn2;
   GO:0032502: developmental process Gabrb3, Fnbp1l, Ntrk1, Megf10, Gfap,
   Kdr, Rfx3, Usp13, Igfbp2, G2e3, Ptgs2, Lect1, Ntng1, Hook1, Bmpr2
   Cellular component
   GO Term Symbol
   GO:0005623: cell Ubn2, Bdnf, Jak2, Prkg2, Gad2, Unc5d, Msx1, Wfs1,
   Prlr, Rb1cc1, Adra2a, Pdp1, PVR, Car6, Scn3a;
   GO:0044464: cell part lkbh7, Fhod3, Zmat4, Tubb6, Smarcad1, Parm1,
   Negr1, Ptprt, Camk2d, Doc2b, Vgf, Nexn, Bcl3, Gabra3, Pak3;
   GO:0043226: organelle Rbbp8, Hspa4l, Map3k2, Aqp1, Trib1, Psme4, Guf1,
   Scn1a, Pds5a, Adamts1, Kitlg, Dnase2, Tmbim1, Akap12, Stc1;
   GO:0044425: membrane part Synpr, Aqp4, Nrxn1, Ap5m1, Shc4, Fras1,
   Stxbp6, Clic6, Slc6a20, Il13ra1, Fcgr3a, Itm2a, Pqlc1, Bst2, Met
   Molecular function
   GO Term Symbol
   GO:0005488: binding Wipf1, Necab1, Grin3a, Mmp9, Kcnj3, Spta1, Wnt9b,
   Zfp483, Usp32, Dgkb, myd1, Clk1, Atf3, Fibcd1, Atf2;
   GO:0003824: catalytic activity Ace, Mtif2, Trib1, Guf1, Ube3a, Dnase2,
   Far1, Cdk18, Clybl, Ddx3, March7, Ascc3, Spint2, Ntrk1, Papd4;
   GO:0060089: molecular transducer activity Epha3, S1pr3, Hif1a, Gpr63,
   Cd3e, Clock, Map3k2, Acvr2a, Bmpr2, Opcml, Il13ra1, Bag4, Mier1,
   Plcxd3, Bst2;
   GO:0005215: transporter activity Kpna3, Slc24a5, Aqp4, Grin2b, Gabrb2,
   Gria4, Folr1, Grin3a, Kcnj3, Vldlr, Aqp1, Clic6, Cacnb4, Trpc6, Hcn4
   [78]Open in a new tab

   We chose two subclasses for further exploration using AmiGO 2 (Figure
   [79]6). DEGs related to programmed cell death could be involved in
   autophagic cell death, cornification, pyroptosis, mitochondria
   programmed cell death and death receptor activity. DEGs associated with
   defence responses may play an important role in the cellular defence
   response, the innate immune response, the inflammatory response,
   clearance of foreign intracellular DNA and the defence response to
   tumour cells. Subcategorising DEGs could help us to better understand
   their potential roles in Orexin-A reperfusion.

Figure 6. Further analysis of two selected biological process subcategories.

   [80]Figure 6
   [81]Open in a new tab

   (A) Thirteen subcategories associated with programmed cell death. (B)
   Fifteen subcategories related to defence responses.

Pathway analysis of DEGs

   KEGG pathway analysis was performed to determine the biological
   pathways associated with DEGs. Among the 206 identified pathways in the
   Orexin-A group compared with the I/R group, the top 15 were Metabolic
   pathways (ko01100; 34 DEGs), Endocytosis (ko04144; 29 DEGs), HTLV-I
   infection (ko05166; 28 DEGs), Pathways in cancer (ko05200; 25 DEGs),
   Cell adhesion molecules (ko04514; 23 DEGs), Mitogen-activated protein
   kinase (MAPK) signalling pathway (ko04010; 21 DEGs), Phagosome
   (ko04145; 21 DEGs), Tight junction (ko04530; 19 DEGs), Focal adhesion
   (ko04510; 18 DEGs), Wnt signalling pathway (ko04310; 16 DEGs), Calcium
   signalling pathway (ko04020; 16 DEGs), Dopaminergic synapse (ko04728;
   16 DEGs), Chemokine signalling pathway (ko04062; 15 DEGs), GnRH
   signalling pathway (ko04912; 13 DEGs) and Neurotrophin signalling
   pathway (10 DEGs; Figure [82]7 and Table [83]4). These signalling
   pathways likely play important roles in a various manner in the
   neuroprotective effects of Orexin-A.

Figure 7. Top 15 enriched pathways between I/R and Orexin-A groups.

   [84]Figure 7
   [85]Open in a new tab

Table 4. KEGG pathway classification of differentially expressed genes.

   Pathway Gene symbol
   ko01100: Metabolic pathways Lect1; Ttr; Kl; Pla2g5; Nt5dc2; Hyal1;
   Fam213b; Phgdh; Hmgcs2; Csad; Smpd2; Lcn2; Plcb4; Inpp5f; C1galt1;
   Ndufa10l1; Upp1; Inpp4b; Ndst4; Ireb2;Gnpda2; Idi1; Pik3c2a; B3galt1;
   Rev3l; Pgap1; Ptgs2; Gad2; Csgalnact1; Dgkb; Nmnat2.
   ko04144: Endocytosis Folr1; Ntrk1; RT1-CE10; RT1-CE7; Rab11fip1;
   RT1-A3; RT1-A1; RT1-A2; Met LOC360231; RT1-T24-4; RT1-CE16; Rt1.aa;
   Hspa1a; RT1-T24-3; Hook1; Ofd1; Cav2; Ret; Txlng; Tmf1; Dnm1l; Filip1;
   Golga1; Wwp1.
   ko05166: HTLV-I infection Mybl2; Cd3e; RT1-CE10; Msx1; RT1-CE7; RT1-A3;
   RT1-A1; RT1-A2; LOC360231; RT1-T24-4; RT1-CE16; Rt1.aa; Wnt3; Fos;
   Atf3; RT1-T24-3;Wnt9b; Adcy8; Zfp36; Creb1; Arhgap29; Crem; Atf2; Fzd3;
   Mapk8; Myc.
   ko05200: Pathways in cancer Ntrk1; Gstm4; Mmp2; Wnt3; Fos; Tcf7l2;
   Mmp9; Kitlg; Wnt9b; Brca2; Vegfa; Ccne2; Arhgap29; Cpne9; Met; Ret;
   Hsp90aa1; Lef1; Rassf5; Fzd3; Hif1a; Stat3; Mapk8; Myc; Ptgs2.
   ko04514: Cell adhesion molecules Glycam1; Cldn2; Cdh3; RT1-CE10;
   Selplg; RT1-CE7; Cldn1; RT1-A3; F11r; RT1-A1; RT1-A2; LOC360231;
   RT1-T24-4; RT1-CE16; Rt1.aa; PVR; RT1-T24-3; Ncam2; Negr1; Cpne9;
   Nlgn1; Cadm1; Nrxn1.
   ko04010: MAPK signalling pathway Pla2g5; Ntrk1; Hspa1a; Fos; Hspb1;
   Map3k2; Dusp1; Cacnb4; Cacna2d1; Gadd45a; Arhgap29; Cpne9; Gadd45g;
   Taok1; Atf2; Ppm1l; Rasa1; Mapk8; Bdnf; Myc; Prkacb.
   ko04145: Phagosome RT1-CE10; RT1-CE7; RT1-A3; RT1-A1; Fcgr3a; RT1-A2;
   LOC360231; RT1-T24-4; RT1-CE16; Rt1.aa; Tubb6; RT1-T24-3; Hook1;
   Dync2h1; Ofd1; Tmf1; Filip1; Golga1; Eea1; Crispld2; Nos1.
   ko04530: Tight junction Cldn2; Cldn1; F11r; Prkcd; Synpo2; Spta1;
   Cpne9; Phf20l1; Bicd1; Txlng; Tmf1; Smtn; Filip1; Rb1cc1; Golga1;
   Cttnbp2nl; Ccdc39; Cep83; Ppfibp1.
   ko04510: Focal adhesion Nexn; Shc4; Vegfa; Akap12; Pak3; Rock1;
   Arhgap29; Cpne9; Met; Cav2; Smtn; Shc3; Pdlim4; Mapk8; Ppfibp1;
   Arhgap5; Kdr; Arid4a.
   ko04310: Wnt signalling pathway Sfrp1; Wnt3; Plcb4; Tcf7l2; Wnt9b;
   Grem2; Camk2d; Rock1; Cpne9; Lef1; Fzd3; Fhod3; Mapk8; Ppfibp1; Myc;
   Prkacb.
   ko04020: Calcium signalling pathway Calml4; Htr5b; Nexn; Plcb4; Adra1b;
   Adcy8; Camk2d; Akap12; Tacr1; Atp2b4; Cpne9; Grm5; Cd38; Prkacb; Nos1;
   Arid4a.
   Dopaminergic synapse Calml4; Fos; Plcb4; Creb1; Scn1a; Camk2d;
   Arhgap29; Cpne9; Grin2b; Kif5b; Atf2; Kcnj3; Gria4; Mapk8; Clock;
   Prkacb.
   ko04062: Chemokine signalling pathway Cxcl13; Ccl2; Prkcd; Plcb4;
   Adcy8; Shc4; Rock1; Arhgap29; Cpne9; Shc3; Stat3; Jak2; Pdlim4;
   Ppfibp1; Prkacb.
   ko04912: GnRH signalling pathway Pla2g5; Calml4; Mmp2; Mmp14; Prkcd;
   Plcb4; Map3k2; Adcy8; Camk2d; Hbegf; Cpne9; Mapk8; Prkacb.
   ko04728: Neurotrophin signalling pathway Ntrk1; Tp73; Calml4; Prkcd;
   Shc4; Camk2d; Arhgap29; Shc3; Mapk8; Bdnf.
   [86]Open in a new tab

Gene interaction network analysis of DEGs

   We selected three of the more interesting DEGs in the Orexin-A group
   compared with the I/R group for gene interaction analysis by GeneMANIA.
   The functional interaction network of Hspb1 (Figure [87]8A) includes
   seven Hspb1-interacting genes that were found to be differentially
   expressed by RNA-seq, of which five (Hspa1b, Odf1, Cryaa, Hspb2 and
   Flinc) were down-regulated and two (Hspb3 and Csrp3) were up-regulated
   (Table [88]5). Meanwhile, eight down-regulated and two up-regulated
   genes (Igf1, Igfbp1, Igfbp4, Gpc3, Padi3, Cyp4a8, Rln3, Insl3, Igfbp2
   and Ins1; Table [89]5) appeared to functionally interact with Igf2
   (Figure [90]8B). Similarly, six genes interacted with Ptk2b, of which
   Slc2a1, Jak2, Mcam and Epha7 were down-regulated and Slcola2 and Matk
   were up-regulated after reperfusion Orexin-A (Figure [91]8C and Table
   [92]5).

Figure 8. GeneMANIA functional interaction network analysis of candidate rat
genes related to the neuroprotective effects of Orexin-A.

   [93]Figure 8
   [94]Open in a new tab

   (A) Functional interaction network of Hspb1. (B) Functional interaction
   network of Igf2. (C) Functional interaction network of Ptk2b. Red
   symbols represent genes down-regulated in the Orexin-A group compared
   with the I/R group. Green symbols represent up-regulated genes. Black
   symbols indicate genes with no significant differences in expression
   between groups.

Table 5. Genes interacting with Hspb1, Igf2 and Ptk2b.

   Genes interacting with Hspb1
   Gene symbol Log2 ratio Gene symbol Log2 ratio Gene symbol Log2 ratio
   Hspb2 -9.95 Hspa1b -1.51 Hspb3 8.45
   Cryaa -8.87 Flnc -0.60 Csrp3 8.28
   Odf1 -8.08
   Genes interacting with Igf2
   Gene symbol Log2 ratio Gene symbol Log2 ratio Gene symbol Log2 ratio
   Rln3 -10.19 Cyp4a8 -1.57 Igfbp2 1.75
   Insl3 -8.81 Igf1 -1.16 Ins1 0.97
   Igfbp1 -7.41 Igfbp4 -1.05
   Padi3 -6.32 Gpc3 -0.75
   Genes interacting with Ptk2b
   Gene symbol Log2 ratio Gene symbol Log2 ratio Gene symbol Log2 ratio
   Jak2 -1.13 Slc2a1 -0.56 Slcola2 0.75
   Epha7 -0.69 Mcam -0.53 Matk 0.71
   [95]Open in a new tab

   In addition, we analysed the signalling pathways involving Hspb1, Igf2
   and Ptk2b that may be involved in Orexin-A reperfusion. RNA-seq data
   showed that Hspb1 was mainly involved in the MAPK and VEGF signalling
   pathways, whereas Igf2 was most closely associated with the P53
   signalling pathway. Ptk2b could potentially affect four signalling
   pathways: the chemokine, nicotine addiction, calcium and Jak-STAT
   signalling pathways. Elucidation of these DEGs and their associated
   signalling pathways could prove critical for understanding the
   neuroprotective mechanisms of Orexin-A.

DISCUSSION

   RNA-seq technology has emerged as a powerful method for screening
   transcripts and has been applied to a broad range of fields [[96]18,
   [97]19]. Bermudez et al. used RNA-seq to screen DEGs and corresponding
   pathways involved in glucocorticoid responsiveness in a bovine
   trabecular meshwork (BTM) cell strain. Sequencing results showed that
   93 and 606 genes were differentially expressed in responder and
   non-responder BTM cells, respectively [[98]20]. Wang et al. analysed
   stage-specific genes associated with gastric cancer using RNA-seq and
   identified 2,224 genes specifically expressed in stages I and II, and
   539 genes specific to stage IV [[99]21]. Meanwhile, Holmes et al.
   assessed the impact of three different intensities of transcranial
   direct current stimulation (tDCS) on gene expression in rat cerebral
   cortex by RNA-seq. Compared with the sham group, ~1,000 DEGs were
   identified at each treatment intensity [[100]22]. In the current study,
   Orexin-A alleviated the cerebral infarction volume induced by I/R
   injury, consistent with the neuroprotective effects of Orexin-A on I/R
   injury. To more comprehensively reveal which genes and pathways are
   involved in the neuroprotective effects of Orexin-A, RNA-seq was
   carried out on sham, I/R and Orexin-A rat groups. To our knowledge,
   this is the first study to use RNA-seq to perform global gene
   expression profiling after reperfusion with Orexin-A in the
   hippocampus. Compared with the I/R group, ≥2-fold up-regulation of 149
   genes and ≥2-fold down-regulation of 500 genes was observed following
   reperfusion with Orexin-A, identifying 649 genes that are potentially
   involved in the neuroprotective effects of Orexin-A.

   Go analysis based on the RNA-seq data identified DEGs involved in
   adhesion, metabolism, localisation, immune reaction, programmed cell
   death and defence responses (Figures [101]5 and [102]6; Table [103]3).
   We screened several immune response-related genes including IL11
   (-2.359), Cd55/44 (-1.681/-3.4947), Ccl2 (-4.059), Ifit1 (-4.2288) and
   Cadm1 (-1.079) differentially expressed after reperfusion with
   Orexin-A. A number of genes related to cell death, including Bcl3
   (-4.9307), caspase 3/4 (-0.826/-1.838), Rassf5 (-1.2599), Myc
   (-1.0524), Casp8ap2 (-1.307) and Bag4 (-1.0895), were down-regulated in
   the Orexin-A group. In addition, reperfusion induced the up-regulation
   of Hif1a, Stat3, Hspb2, Hspa1a, Hspa1b and Hspb1, which play an
   important role in stimulus or stress, but Orexin-A lowered their
   expression levels (log2 ratio) by -1.1357-, -1.135-, -9.951-, -3.474-,
   -1.51- and -2.602-fold, respectively. These differences in expression
   after reperfusion with Orexin-A may contribute to its neuroprotective
   effects.

   Orexin-A has been shown to activate multiple protein kinases such as
   PKA, PKC, MAPK/Erk and PDK1 in various cellular contexts [[104]23].
   Orexin-A activates the mTORC1 pathway via extracellular calcium influx
   and the lysosome pathway in recombinant cell lines, leading to the
   central regulation of cell growth and metabolism [[105]24]. Orexin-A
   increases glucose transporter 4 expression and lipid accumulation in
   3T3-L1 adipocytes via ERK1/2, JNK and p38 MAPK signalling [[106]25,
   [107]26]. Orexin-A treatment promoted the uptake of Glu via the
   OX1R/PKCα/ERK1/2 pathway and protected co-cultured cells against
   anoxia/hypoglycaemic injury in astrocytes [[108]27]. Orexin-A
   stimulates ERK[1/2] phosphorylation and facilitates cell migration via
   the PLC-PKCα signalling pathway in cultured rat astrocytes [[109]28].
   In the current study, KEGG pathway analysis identified 206 pathways
   that were closely associated with Orexin-A reperfusion, of which
   metabolic pathways accounted for 34 DEGs, and MAPK signalling accounted
   for 21 DEGs. It was not surprising to find that metabolic pathways and
   MAPK signalling pathways may play important roles in Orexin-A
   reperfusion.

   Besides these two pathways, several others were also detected,
   including Wnt, calcium, chemokine, gonadotropin-releasing hormone
   (GnRH) and neurotrophin signalling pathways. Orexins play important
   roles in nociceptive modulation via activating intracellular calcium
   signalling [[110]29]. Orexin-A and Orexin-B significantly increase the
   expression of neurotrophin-3 (NT-3) at the mRNA level when primary
   cortical neurons are cultured with 0.01, 0.1 or 1 mM Orexins [[111]30],
   and Orexin-A directly stimulates GnRH neurons to secrete GnRH,
   implicating Orexin-A in GnRH signalling [[112]31]. Taken together,
   these results implied that Orexin-A might regulate the above signalling
   pathways, and thereby modulate cerebral I/R injury.

   Among the identified DEGs, three genes of particular interest were
   selected for analysis using GeneMANIA. Heat shock protein beta-1
   (Hspb1), also known as Hsp27, is a member of the small HSP family and a
   key mediator in the cellular response to environmental and
   physiological stresses [[113]32–[114]34]. Upon expose to oxidative
   stress, HSP27 functions as an antioxidant to decrease the levels of
   reactive oxygen species (ROS) [[115]35]. In benign prostatic
   hyperplasia (BPH), the expression of HSP27 is gradually increased and
   accompanied by inflammation [[116]36]. In retinal capillary endothelial
   cells, HSP27 functions as an anti-apoptotic protein to preserve the
   endothelial barrier [[117]37]. Others have similarly noted that HSP27
   inhibits cell death via a c-jun-mediated cascade in cerebral ischemia
   [[118]38]. HSP27 transgenic mice display amelioration of the infarction
   volume, behavioural deficits and blood brain barrier (BBB) damage in a
   stroke model [[119]39]. Insulin-like growth factor 2 (Igf2) is
   indispensable in a number of cells because it promotes cellular
   proliferation and differentiation [[120]40]. In medulloblastoma, Igf2
   promotes the proliferation of precursor cells and the PZp53 cell line
   [[121]41]. IGF2 also mediates the proliferation of hippocampal neural
   stem cells via IGF1R and AKT-dependent signalling [[122]42]. IGF2 is
   highly expressed in microglia and exhibits a strong protective effect
   against cytokine-mediated neuronal death in vitro and in vivo
   [[123]43]. Protein tyrosine kinase type 2 beta (Ptk2b) was found to
   play a vital role in the response to diverse cellular stimuli and
   diseases. Ptk2b specifically induces cell death and reduces clonogenic
   growth in myeloma cell lines and multiple myeloma stem cells.
   Furthermore, Ptk2b inhibition attenuates multiple myeloma progression
   [[124]44]. Ptk2b can be up-regulated by lovastatin, resulting in an
   anti-apoptotic effect on ox-LDL-induced cell injury in endothelial
   cells. The protective effect of Ptk2b is associated with the AKT
   signalling pathway [[125]45]. At 24-72 h after focal cerebral ischemia,
   phospho-Ptk2b was rapidly induced in microglia surrounding the necrotic
   infarction area, and phospho-Ptk2b was confirmed to colocalise with
   phospho-p38 and act as an upstream mediator of the p38 signalling
   pathway after cerebral ischemia [[126]46]. Consistent with these
   previous findings, these three genes likely play a vital role in
   Orexin-A-mediated neuroprotection, and their functions and molecular
   mechanisms will be investigated in future studies.

   It has found that orexin-A has multiple effects on different tissues.
   By comparing the mechanism of orexin-A in different tissues, it can be
   found that there is commonality and tissue specificity. It will provide
   a comprehensive understanding of Orexin-A effects on different organs
   and reveals whether its neuroprotective effects are tissue specific at
   molecular levels.

   In conclusion, our global expression profiling identified genes,
   pathways, functions and interaction networks potentially involved in
   the neuroprotective effects of Orexin-A reperfusion. However, the
   results obtained by RNA-seq and bioinformatic analysis were general and
   indicate the involvement of numerous complex processes, and further
   experiments are needed to validate the identified DEGs and signalling
   pathways to clarify the molecular mechanisms involved in the process of
   Orexin-A reperfusion.

MATERIALS AND METHODS

Animals

   Male Wistar rats, aged 6-8 weeks, were obtained from Pengyue
   Experimental Animal Breeding Institute of Jinan, China (License No.
   SCXK (Lu) 2014-0007). All rats were maintained in a room at 25 ± 3°C
   and 50-60% humidity, with a 12 h light-dark cycle and free access to
   food and water. All experiments were approved by the Ethics Committee
   of Jining Medical University and performed in accordance with the
   National Health Guide for the Care and Use of Laboratory Animals.

Animal grouping

   A total of 18 rats were randomly divided into three groups: (1)
   sham-treated group (sham), from which the right carotid artery was
   isolated without further processing; (2) I/R group, in which rats were
   subjected to a 2 h right middle cerebral artery occlusion (MCAO)
   followed by a 24 h reperfusion with saline (10 μl, 0.9% NaCl) after
   ischemia at the onset of reperfusion; (3) Orexin-A reperfusion group
   (Orexin-A), in which rats were injected with Orexin-A (50 ng/kg;
   Phoenix Pharmaceuticals, Inc., Burlingame, CA, USA) after reperfusion.

MCAO model

   Rat MCAO model experiments were performed using the bolt line method.
   Firstly, rats were anesthetised by intraperitoneal injection of 10%
   chloral hydrate (0.3 mL/100 g) and swabbed with iodophor. The common
   carotid artery (CCA) was isolated, and a midline incision was made to
   expose the internal carotid artery (ICA), external carotid artery (ECA)
   and right CCA. A nylon suture (0.26 mm diameter) was inserted into the
   right CCA lumen and gently injected into the ICA at a depth of 18 mm,
   and the neck incision was sutured. Finally, nylon sutures were gently
   removed at 2 h after ischemia, and reperfusion was performed for 24 h.

TTC staining

   The cerebral infarction volume was measured using TTC (Sigma, St Louis,
   MO, USA) staining. At 24 h after reperfusion, brains were removed,
   frozen at -20°C for 30 min, cut into 2 mm thick coronal slices and
   stained with 0.2% TTC solution in the dark for 30 min at 37°C. After
   fixing with 4% formalin, slices were photographed using a IX71
   microscope (Olympus, Japan). Infarction zones were analysed with
   Image-Pro Plus v 6.0 software (Media Cybernetics, Bethesda, MD, USA).
   The infarction vs. normal volume percentage was calculated.

Construction of cDNA library and RNA-seq

   Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA,
   USA) according to the manufacturer's instructions. RNA was quantified
   and the integrity was determined using a BioSpec-nano spectrophotometer
   (Shimadzu, Japan). Total RNA was treated with DNase I to remove DNA
   contamination, and RNA was used to generate a cDNA library. The mRNAs
   were multiplexed and tagged with standard Illumina tags, and fragments
   were amplified by PCR. Library quality and quantity were determined
   using an Agilent 2100 Bioanalyzer and an ABI StepOnePlus Real-Time PCR
   System. Library products were sequenced using an Illumina HiSeq TM 2000
   platform by Shenzhen Huada Genetics Corporation (Shenzhen, China).

Bioinformatics analysis

   Raw read data from RNA-seq were subjected to quality control (QC) that
   involved discarding low-quality reads, and the resulting clean reads
   were aligned with reference sequences using BWA software [[127]47].
   Expression levels of genes were quantified using RSEM [[128]46].
   Screening of DEGs was based on three methods: Poisson distribution, the
   NOISeq package [[129]48] and the EBSeq package [[130]49]. Gene Ontology
   (GO; [131]http://www.geneontology.org/), PANTHER classification
   analysis ([132]http://www.pantherdb.org/), KEGG Pathway enrichment
   analysis ([133]http://www.genome.jp/kegg/) and protein-protein
   interaction network analysis (GeneMANIA; [134]http://genemania.org/)
   were subsequently performed.

qRT-PCR

   The same RNA samples were reserve-transcribed into cDNA using a cDNA
   synthesis kit (Tiangen Biotech, Beijing, China) according to the
   manufacturer's protocol. qRT-PCR was performed in 96-well plates using
   a Roche LC480. Reactions (20 μL) were mixed with 10 μL of 2×SYBR
   SuperReal PreMix (Tiangen Biotech, Beijing, China), 2 μL of cDNA, 0.6
   μL of forward primer, 0.6 μL of reverse primer and 6.8 μL of ddH[2]O.
   The amplification procedure was as follows: 95°C for 10 min, followed
   by 40 cycles of 95°C for 15 s and 60°C for 1 min. PCR primers are
   listed in Table [135]1. β-actin was amplified as an endogenous control.
   The relative mRNA level of each gene was calculated using the 2 ^−ΔΔCT
   method. At least three replicates were performed for each cDNA sample,
   and reactions were prepared in triplicate.

Statistical analysis

   All data are expressed as mean ± standard error of the mean (SEM) and
   were analysed using GraphPad Prism one-way analysis of variance
   (ANOVA). A p-value less than 0.05 was considered statistically
   significant.

ACKNOWLEDGMENTS AND FUNDING