- Email: [email protected]
Intrathecal lentivirus-mediated RNA interference targeting nerve growth factor attenuates myocardial ischaemia–reperfusion injury in rat
British Journal of Anaesthesia, xxx (xxx): xxx (xxxx) doi: 10.1016/j.bja.2019.06.024 Advance Access Publication Date: xxx Laboratory Investigation
LABORATORY INVESTIGATION
Intrathecal lentivirus-mediated RNA interference targeting nerve growth factor attenuates myocardial ischaemiaereperfusion injury in rat Mengyun Dou1,2,y, Zhenxiao Ma1,2,y, Xueying Cheng1,2, Guichang Zou3, Yan Xu3, Cheng Huang1,2, Wei Xiong3, Shufang He1,2,* and Ye Zhang1,2,* 1
Department of Anaesthesiology and Perioperative Medicine, The Second Hospital of Anhui Medical University, Hefei,
China, 2Key Laboratory of Anaesthesiology and Perioperative Medicine of Anhui Higher Education Institutes, Anhui Medical University, Hefei, China and 3Key Laboratory of Brain Function and Disease, School of Life Sciences, University of Science and Technology of China, Chinese Academy of Sciences, Hefei, China *Corresponding authors. E-mails: [email protected], [email protected] y
These authors contributed equally to this work.
Abstract Background: Nerve growth factor (NGF) has been implicated in hyperalgesia by sensitising nociceptors. A role for NGF in modulating myocardial injury through ischaemic nociceptive signalling is plausible. We examined whether inhibition of spinal NGF attenuates myocardial ischaemiaereperfusion injury and explored the underlying mechanisms. Methods: In adult rats, lentivirus-mediated short-hairpin RNA targeted at reducing NGF gene expression (NGF-shRNA) or a transient receptor potential vanilloid 1 (TRPV1) antagonist (capsazepine) was injected intrathecally before myocardial ischaemiaereperfusion. Infarct size (expressed as the ratio of area at risk) and risk of arrhythmias were quantified. Whole-cell clamp patch electrophysiology was used to record capsaicin currents in primary dorsal root ganglion neurones. The co-expression of substance P (SP) and calcitonin gene-related peptide (CGRP), plus activation of TRPV1, protein kinase B (Akt) and extracellular signal-regulated kinase (ERK) were also quantified. Results: NGF levels increased by 2.95 (0.34)-fold in dorsal root ganglion and 2.12 (0.27)-fold in spinal cord after myocardial ischaemiaereperfusion injury. Intrathecal injection of NGF-shRNA reduced infarct area at risk from 0.58 (0.02) to 0.37 (0.02) (P<0.01) and reduced arrhythmia score from 3.67 (0.33) to 1.67 (0.33) (P<0.01). Intrathecal capsazepine was similarly cardioprotective. NGF-shRNA suppressed expression of SP/CGRP and activation of Akt/ERK and TRPV1 in spinal cord. NGF increased capsaicin current amplitude from 144 (42) to 840 (132) pA (P<0.05), which was blocked by the TRPV1 antagonist 50 -iodoresiniferatoxin. Exogenous NGF enhanced capsaicin-induced Akt/ERK and TRPV1 activation in PC12 neuroendocrine tumour cells in culture. Conclusions: Spinal NGF contributes to myocardial ischaemiaereperfusion injury by mediating nociceptive signal transmission. Keywords: lentivirus; myocardial ischaemiaereperfusion; nerve growth factor; nociceptive signal; transient receptor potential vanilloid 1
Editorial decision: 10 June 2019; Accepted: 10 June 2019 © 2019 Published by Elsevier Ltd on behalf of British Journal of Anaesthesia. For Permissions, please email: [email protected]
1
2
-
Dou et al.
Editor’s key points Perioperative myocardial ischaemiaereperfusion injury is associated with excess morbidity and mortality, yet many conventional treatments are ineffective. Nociceptive signalling from ischaemic cardiac tissue may aggravate both infarct size and cardiac arrythmias. Nerve growth factor modulates nociceptive signalling. In this preclinical study, myocardial infarction was reduced by genetic inhibition of nerve growth factor expression using intrathecally delivered small interfering RNA in adult rats.
Myocardial ischaemia and reperfusion injury are independently associated with increased risk of perioperative mortality after cardiac1 and noncardiac surgery.2 Chemical mediators produced during ischaemia and reperfusion, including protons, adenosine, and bradykinin, activate cardiac afferent nerves and transmit ischaemic nociceptive signals to the thoracic spinal cord through dorsal root ganglia (DRG).3,4 This results in cardiac pain and excitatory cardiovascular reflex responses.4 Modulation of cardiac nociceptive signalling may therefore serve as an useful perioperative strategy against myocardial ischaemia/reperfusion (I/R) injury. Nerve growth factor (NGF) is a neurotrophic factor that rapidly sensitises nociceptive sensory neurones to painful stimuli.5 When NGF binds to its high-affinity receptor (tropomyosin receptor kinase A [TrkA]), NGF sensitises transient receptor potential vanilloid 1 (TRPV1), a non-selective cation channel mainly expressed on nociceptive neurones. This leads to enhanced neuronal discharge and the release of neuropeptides.6 NGF induces the sensitisation of TRPV1 through various different signalling pathways,6e8 although the precise mechanisms remain unclear. NGF levels are up-regulated in the myocardium after myocardial infarction.9,10 Local anaesthetic blockade of the spinal nerves abolishes the up-regulation of NGF in ischaemic myocardium, suggesting that NGF may modulate myocardial I/R injury.9 Local increases of NGF in ischaemic myocardium may reduce reperfusion injury and improve post-ischaemic cardiac recovery.11,12 However, the neuronal effect of NGF in myocardial I/R injury has not been elucidated. Lentiviral-mediated gene transfer into neuronal cells has been shown to be effective and safe, without significant immune responses or unwanted side-effects.13,14 We constructed a lentiviral-mediated short hairpin RNA (shRNA) to specifically inhibit spinal NGF gene expression in the thoracic spinal cord via intrathecal administration. Our data suggest that neuronal NGF contributes to myocardial I/R injury through ischaemic nociceptive signal transmission.
Methods Animals Male adult SpragueeDawley (SD) rats, weighing 260 (20) g, or 1week-old SD rats (for neonatal DRG neurone isolation) were used in this study. All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publication No. 85-23, revised 1996), and ARRIVE
(Animal Research: Reporting of In Vivo Experiments) guidelines. The study was approved by the Institutional Animal Care and Use Committee of Anhui Medical University. For animals undergoing surgery, the adequacy of anaesthesia was confirmed by the loss of righting reflex and lack of response to tail and hind limb toe pinch. Buprenorphine (0.05 mg kg1) was administered s.c. every 12 h for the first 3 postoperative days to ensure postoperative analgesia. Animals were euthanised by overdose (100 mg kg1) of pentobarbital sodium injection (adult rats) or inhalation of carbon dioxide (neonatal rats). A total of 60 adult rats and 20 neonatal rats were used in this study. Three rats were excluded because of infection (one rat) and motor dysfunction (two rats) after intrathecal injection. The other five animals were discarded because of death ahead of schedule during ischaemia (two rats, both in the I/R group) and reperfusion (three rats, one in the I/R group and two in the NC-shRNA group).
NGF shRNA We designed lentiviral vectors encoding a shRNA specifically targeting the NGF gene, which was delivered intrathecally at the level of the upper thoracic spinal cord segments (T2eT6). The cell bodies of the spinal cardiac afferent fibres originating from the heart are mainly located in the DRGs of the T2eT6 spinal segments and terminate in the laminae within the same segments of spinal cord.4 We examined the inhibitory effects of the NGF-shRNA in vitro and in vivo, and confirmed NGF at both mRNA and protein levels.
Intrathecal injections of lentiviral-shRNA and TRPV1 antagonist Intrathecal catheterisation at the T2eT6 level of the thoracic spinal cord was performed as described previously.15,16 After recovery for 3 days, the animals received a single injection with NGF-shRNA, NC-shRNA or normal saline in the same volume (10 ml) via the implanted catheter, followed by additional 10 ml saline infusion. The dead space of the catheter lumen was approximately 10 ml. The final virus titres of NGFshRNA and NC-shRNA were 4108 and 1109 TU ml1, respectively, as determined by GFP fluorescence in H293T cells (Supplementary Fig. S1). We established that 7 days was sufficient for shRNA interference before myocardial ischaemia. In another set of experiments, the intrathecally catheterised animals were administrated 10 ml capsazepine (1 mg ml1) and additional 10 ml saline infusion via the catheter. After 10 min, the animals were subjected to myocardial I/R procedure.
Myocardial I/R injury Rats were anaesthetised with 3% pentobarbital sodium i.p. (60 mg kg1), intubated, and ventilated with 100% oxygen. After thoracotomy, ischaemia was achieved by ligating the left anterior descending coronary artery (LAD) using a 6-0 silk suture for 30 min, followed by 120 min reperfusion by releasing the ligature.17 During experiments, ECG, HR, MAP, and rate pressure product (RPP) were continuously recorded. At the end of experiments, the animals were sacrificed via removal of heart under anaesthesia. Heart slices were prepared and stained by triphenyltetrazolium chloride (TTC) for infarct size determination. Tissue samples from thoracic spinal cord and DRG were collected for immunoblot, quantitative reverse
Spinal NGF gene silence reduces cardiac I/R injury
transcriptionepolymerase chain reaction, and immunofluorescence assays.
Arrhythmia score Arrythmia burden after myocardial I/R injury was qualitatively recorded according to the following criteria18: 0, premature ventricular complexes (PVC) <50 times; 1, PVC 50e499 times; 2, PVC >500; 3, ventricular tachycardia/ventricular fibrillation (VT/VF), total duration <1 min; 4, total VT/VF continued 1e2 min; 5, total VT/VF >2 min.
-
3
Cell models: DRG and PC12 cells One-week-old SD rats were sacrificed via inhalation of carbon dioxide followed by decapitation. DRG cells were isolated from T2eT6 segments and maintained in neurobasal medium with 50 ng ml1 NGF for 24 h, followed by 24 h in medium without NGF.19,20 After 48 h of culture, the primary DRG cells were preincubated with 100 ng mL1 NGF for 1 h and then exposed to100 nmol L1 capsaicin, in the presence or absence of TRPV1 specific antagonist 50 -iodoresiniferatoxin (I-RTX; 100 nmol L1), followed by whole-cell clamp patch recording or
Fig 1. Up-regulation of NGF in thoracic DRG and spinal cord after myocardial ischaemiaereperfusion. (a) Representative photographs of heart sections from Sham and ischaemiaereperfusion group. Myocardial infarct size is expressed as a ratio of IS/AAR. (b) Arrhythmia score was determined based on the frequency and duration of arrhythmias during experiment (where 0 was the lowest and 5 was the highest arrhythmia incidence). All above values are presented as mean (SEM); n¼6, **P<0.01 by unpaired t-test. Representative images of NGF protein immunoreactive staining in (c) T2eT6 DRG and (d) spinal dorsal horns. Bar¼50 mm. The relative levels of NGF protein in (e) T2eT6 DRG and (f) spinal cord were normalised by b-actin. The values are expressed as mean (SEM); n¼6, **P<0.01 by unpaired t-test. NGF, nerve growth factor; DRG, dorsal root ganglia; SEM, standard error of the mean; IS/AAR, infarct size/area at risk; I/R, ischaemiaereperfusion.
4
-
Dou et al.
Fig 2. Intrathecal injection of lentiviral NGF-shRNA reduces infarct size and arrhythmia. (a) The GV248 vector carried one pair of oligonucleotides to generate shRNA targeting NGF gene. (b) T2eT6 spinal cord and DRG tissue sections were prepared 3 days after intrathecal injection of lentivirus. Both light field (A, C) and green fluorescent protein (GFP) fluorescence (B, D) were examined under a fluorescent microscope to confirm the in vivo infection efficiency. Bar¼50 mM. The levels of NGF mRNA (c) and protein (d) in thoracic spinal cord were detected 7 days after intrathecal injection of lentivirus. The values are expressed as mean (SEM); n¼3, **P<0.01 as compared with CON group by one-way ANOVA with post hoc Tukey test. (e) Experimental design for intrathecal injection of lentivirus-mediated NGF-shRNA. The effects on myocardial ischaemiaereperfusion infarct size (f), and arrhythmia scores (g) were examined in each group. Values are expressed as mean (SEM); n¼6, ##p<0.01 as compared with ischaemiaereperfusion group by one-way ANOVA with post hoc Tukey test. Data in ischaemiaereperfusion group used here are same as those in Figure 1. NGF, nerve growth factor; shRNA, short hairpin RNA; DRG, dorsal root ganglia; SEM, standard error of the mean; ANOVA, analysis of variance; NC, negative control; CON, control; IS/AAR, infarct size/area at risk; I/R, ischaemiaereperfusion; SEM, standard error of the mean.
Spinal NGF gene silence reduces cardiac I/R injury
immunofluorescence staining. In addition, PC12 neuroendocrine tumour cells were incubated with 100 ng ml1 NGF for 1 h and then exposed to 100 nmol L1 capsaicin for 30 min for western immunoblot analysis. The doses and time of above drugs were determined based on preliminary experiments. Expanded methods are presented in Supplementary Materials.
-
5
Statistical analysis Data are expressed as mean (standard error of the mean, SEM). Quantification of infarct size/area at risk, arrhythmia score, densitometric analysis of protein bands, and patch clamp recordings were undertaken by research personnel
Fig 3. Immunofluorescence staining of substance P (SP) and calcitonin gene-related peptide (CGRP) in DRG neurones and spinal cord. (a) The localisation of SP (green) and CGRP (red) in T2eT6 DRG tissue sections were evaluated by double immunofluorescence. White arrows indicate merge staining (yellow). Bar¼50 mm. (b) The immunostaining of SP or CGRP in T2eT6 spinal cord tissue sections. Bar¼100 mm. ANOVA, analysis of variance; DRG, dorsal root ganglia; NGF, nerve growth factor; I/R, ischaemiaereperfusion; shRNA, short hairpin RNA.
6
-
Dou et al.
masked to treatment allocations. Results were analysed using GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA, USA). Unpaired t-test was used for comparisons between two experimental groups. Comparisons among multiple groups were conducted via one-way analysis of variance (ANOVA) followed by post hoc Tukey’s test. Repeatedmeasures ANOVA followed by a Bonferroni’s test were performed to analyse the haemodynamic parameters at different time points. Statistical significance was set at P<0.05. The sample size in each group was calculated using € t Kiel, GerG*Power Version 3.1.7 (Franz Faul, Universita many),21 based on our previous work.17
Results NGF protein expression in thoracic DRG and spinal cord after myocardial I/R Myocardial I/R injury caused a high ratio of infarct size to area at risk (0.58 [0.02]; Fig. 1a). Arrhythmias occurred commonly during I/R injury (Supplementary Fig. S2), resulting in high arrhythmia scores (Fig. 1b). The immunoreactive staining of NGF protein in DRG (Fig. 1c) and spinal cord (Fig. 1d) were increased after myocardial I/R. Immunoblots showed that NGF protein level was increased by 2.95 (0.34)-fold in DRG (Figs 1e) and 2.12 (0.27)-fold in spinal cord (Fig. 1f), respectively.
Fig 4. NGF enhanced TRPV1 currents and SP/CGRP expression in thoracic DRG neurones. (a) Representative trace records of inward currents elicited by capsaicin (100 nmol L1) in primary DRG neurones. (b) The data points of capsaicin-elicited inward current amplitude (n¼9 from three rats in each group). The values are expressed as mean (SEM); *P<0.05, **P<0.01 by one-way ANOVA with post hoc Tukey test. (c) Representative images of SP (green) and CGRP (red) double immunofluorescence in DRG neurones. The furthest right lane shows the enlarged merge images. Bar¼50 mm. NGF, nerve growth factor; TRPV1, transient receptor potential vanilloid 1; DRG, dorsal root ganglia; SP, substance P; CGRP, calcitonin gene-related peptide; ANOVA, analysis of variance; SEM, standard error of the mean; CON, control; I-RTX, 50 iodoresiniferatoxin; CON, control.
Spinal NGF gene silence reduces cardiac I/R injury
-
7
Fig 5. NGF-dependent activation of Akt and ERK signalling in spinal cord. (a) NGF-shRNA suppressed the protein expression of NGF and its receptor TrkA in spinal cord. (b) The relative expression of NGF or TrkA was normalised to b-actin and the value in Sham group was assigned as 1. (c) The phosphorylated (p) and total Akt protein levels in spinal cord upon NGF-shRNA treatment. (d) The relative expression of pAkt and Akt was normalised to b-actin and the value in Sham group was assigned as 1. (e) The phosphorylated (p) and total ERK protein levels in spinal cord upon NGF-shRNA treatment. (f) The relative expression of pERK was normalised to total ERK and b-actin and the value in Sham group was assigned as 1. The values are expressed as mean (SEM); n¼3, **P<0.01, compared with Sham; #P<0.05, ##P<0.01, compared with ischaemiaereperfusion by one-way ANOVA with post hoc Tukey test. (g) The phosphorylation of Akt and ERK upon exogenous NGF treatment in PC12 cells. (h) The relative expression of and pAkt or pERK was normalised to total Akt or ERK, and b-actin. The value in CON group was assigned as 1. The values are expressed as mean (SEM); n¼3, *P<0.05, compared with CON. NGF, nerve growth factor; Akt, protein kinase B; NGF, nerve growth factor; shRNA, short hairpin RNA; ERK, extracellular signal-regulated kinase; ANOVA, analysis of variance; SEM, standard error of the mean; I/R, ischaemiaereperfusion; CON, control; CAP, capsaicin; NC, negative control.
Inhibition of spinal NGF expression limited myocardial I/R injury We generated three lentiviral vectors that carried shRNA targeting the NGF gene (Supplementary Table S1) and identified an effective shRNA (NGF-shRNA3; Supplementary
Fig. S3). In vivo, the lentiviral vectors transfected thoracic spinal cord and DRG cells, as revealed by GFP fluorescence (Fig. 2b). Both NGF mRNA and protein levels in thoracic spinal cord were reduced by the lentiviral-mediated NGF-shRNA (Fig. 2ced).
8
-
Dou et al.
Fig 6. Contribution of NGF-dependent TRPV1 activation in myocardial ischaemiaereperfusion injury. (a) The phosphorylated (p) and total TRPV1 levels in spinal cord after myocardial ischaemiaereperfusion injury. (b) The relative expression of pTRPV1 and TRPV1 was normalised to b-actin and the value in Sham group was assigned as 1. The values are expressed as mean (SEM); n¼3, **P<0.01, compared with Sham; ##P<0.01, compared with ischaemiaereperfusion by one-way ANOVA with post hoc Tukey test. (c) The expression of pTRPV1 and TRPV1 upon exogenous NGF treatment with or without TRPV1 antagonist I-RTX in PC12 cells. (d) The relative expression of pTRPV1 and TRPV1 was normalised to b-actin. The value in CON group was assigned as 1. The values are expressed as mean (SEM); n¼3, *P<0.05, compared with CON; &P<0.05, compared with NGF by one-way ANOVA with post hoc Tukey test. (e) Myocardial infarct size and (f) arrhythmia scores were reduced by intrathecal administration of TRPV1 antagonist capsazepine. The values are expressed as mean (SEM); n¼6, ##P<0.01, #P<0.05, as compared with ischaemiaereperfusion group by one-way ANOVA with post hoc Tukey test. (g) Suggested mechanism through which lentivirus-mediated NGF-shRNA exerts cardioprotective effects against myocardial ischaemic injury. NGF, nerve growth factor; shRNA, short hairpin RNA; TRPV1, transient receptor potential vanilloid 1; ANOVA, analysis of variance; SEM, standard error of the mean; I/R, ischaemiaereperfusion; TrkA, tropomyosin receptor kinase A; I-RTX, 50 -iodoresiniferatoxin; CON, control; IS/AAR, infarct size/ area at risk.
Spinal NGF gene silence reduces cardiac I/R injury
Rats were randomised to undergo myocardial I/R injury having received either NGF-shRNA or control (NC-shRNA) lentiviral GV248 vectors intrathecally (Fig. 2e). Infarct size was reduced from 0.58 (0.02) to 0.37 (0.02) by NGF-shRNA (Fig. 2f). There were no differences in the ratio of area at risk/left ventricle and right ventricle between groups (Supplementary Fig. S4a). The frequency of myocardial I/R injury-induced arrhythmias was also reduced by intrathecal injection of NGF-shRNA but not NC-shRNA (Fig. 2g). Haemodynamic parameters were similar between shRNA-treated groups (Supplementary Table S2).
Inhibition of spinal NGF reduced the expression of substance P and calcitonin gene-related peptide Double immunofluorescence showed moderate staining of substance P (SP) and calcitonin gene-related peptide (CGRP) in DRG slices from the sham-operated group. After myocardial I/ R injury, both SP and CGRP immunoreactivity were increased and co-localised in DRG neurones. The immunoreactive staining of both SP and CGRP was reduced by NGF-shRNA but not the NC shRNA (Fig. 3a). Similar staining was observed in the spinal cord. Immunofluorescent staining of both SP and CGRP was enhanced after myocardial I/R injury, particularly in the superficial laminal neurones of spinal dorsal horn. This expression was inhibited by NGF-shRNA (Fig. 3b).
NGF enhanced TRPV1 currents and SP/CGRP expression in thoracic DRG neurones Rat thoracic DRG neurones displayed moderate inward currents after application of 100 nmol L1 capsaicin. When preincubated with NGF, the DRG neurones were more sensitive to capsaicin, with an increase in current amplitude from 144 (42) to 840 (132) (P<0.05) However, these NGF-sensitised DRG neurone currents were completely abolished by the addition of I-RTX, the specific TRPV1 antagonist (Fig. 4a and b). SP and CGRP were mostly co-localised in DRG neurones, mainly within cell bodies after capsaicin treatment alone (Fig. 4c). Both neuropeptides were increased after NGF treatment. When TRPV1 channel was blocked by I-RTX, immunostaining of both SP and CGRP was lower (Fig. 4c).
NGF-dependent activation of protein kinase B and extracellular signal-regulated kinase signalling in spinal cord NGF-shRNA reduced not only I/R injury-induced NGF upregulation, but also the expression of TrkA receptor (Fig. 5a and b). To explore the signalling pathway downstream of NGF/TrkA, we further examined the influence of NGF-shRNA on the phosphorylation levels of protein kinase B (Akt) and extracellular signal-regulated kinase (ERK1/2), which represents the activation of these kinases. Myocardial I/R injury increased both the total protein level and the phosphorylation level of Akt, which was reduced by NGF-shRNA (Fig. 5c and d). Phosphorylation of ERK kinase was induced by I/R injury but reduced after NGF gene knockdown. The total protein levels of ERK were similar among groups (Fig. 5e and f). Exogenous NGF treatment enhanced capsaicin-mediated phosphorylation of both Akt and ERK in neuronal PC12 cells (Fig. 5g and h).
-
9
NGF-dependent TRPV1 activation contributed to myocardial I/R injury The activation of Akt/ERK signalling induced by NGF may result in sensitisation of TRPV1 receptor.7 We found that both total and phosphorylated TRPV1 levels were increased in the spinal cord after myocardial I/R injury. Knockdown of NGF gene expression in spinal cord attenuated the increase of total and phosphorylated TRPV1 proteins after myocardial I/R injury (Fig. 6a and b). Moreover, capsaicin-induced up-regulation of TRPV1 in PC12 cells was further enhanced when pretreated with NGF, which was abolished by the specific TRPV1 antagonist I-RTX (Fig. 6c and d).
TRPV1 antagonist reduced infarct size We finally examined whether spinal TRPV1 activation contributes to myocardial I/R injury by intrathecally administering capsazepine, a TRPV1 antagonist. Infarct size was reduced from 0.51 (0.02) to 0.33 (0.02) after TRPV1 inhibition with intrathecal capsazepine (Fig. 6e; Supplementary Fig. S4b). Capsazepine also reduced the arrhythmia score from 4.17 (0.31) to 2.50 (0.43) (Fig. 6f). There were no haemodynamic differences between groups (Supplementary Table S3).
Discussion NGF-induced sensitisation of nociceptive sensory neurones has been implicated in inflammatory and neuropathic hyperalgesia. However, little is known about how NGF participates in sensory neurones that may become sensitised during myocardial ischaemic injury. The present study revealed an up-regulation of NGF protein levels in the upper thoracic DRG and spinal cord in response to myocardial I/R injury. By silencing NGF gene expression using lentivirus-mediated shRNA, we demonstrated, for the first time, that inhibition of spinal NGF expression limited cardiac reperfusion injury. Mechanistically, the cardioprotection afforded by reducing spinal NGF protein levels appears to require TRPV1-dependent nociceptive transmission involving spinal Akt and ERK signalling. NGF promotes the release of SP and CGRP from both peripheral and central endings of sensory neurones.22 These two neuropeptides are involved in the transmission of noxious stimuli, conveying nociceptive signalling from primary afferent fibres to the superficial spinal dorsal horn. We found that SP and CGRP were highly expressed and co-localised in DRG neurones after myocardial I/R injury. SP and CGRP immunoreactive staining was reduced after NGF silencing, indicating that the expression of SP and CGRP is at least partly dependent on spinal NGF. CGRP and SP released from TRPV1expressing cardiac sensory nerves into myocardium may reduce cardiac ischaemic injury, contributing to preconditioning induced cardioprotection.23 In DRG and spinal cord, however, we speculate that the increase of SP and CGRP induced by NGF might be responsible for high neuronal activity after ischaemic stimuli. In support of this hypothesis, we used DRG cells isolated from T2eT6 segments to confirm NGF increased TRPV1-dependent inward currents and the expression of SP and CGRP. Although it is known that NGF increases DRG neuronal sensitivity to capsaicin and neuropeptides, in hyperalgesia or neuropathic pain,6,24 our data also show that NGF increased neuropeptide expression and neuronal activity after cardiac ischaemic injury.
10
-
Dou et al.
NGF induces acute sensitisation of TRPV1 receptor on nociceptive sensory neurones via PI3K/Akt and ERK signalling in inflammatory or injury induced hyperalgesia.6,7,25 In line with previous studies, we found that capsaicin-induced TRPV1 expression and activation were enhanced by NGF treatment, whereas NGF-shRNA suppressed the activation of downstream Akt and ERK signal kinases, and the expression and phosphorylation of TRPV1 receptor. Furthermore, intrathecal administration of TRPV1 antagonist exhibited similar protective effects against myocardial injury, indicating that NGFshRNA may reduce myocardial injury through inhibiting TRPV1 activation. These results suggest the involvement of PI3K/Akt and ERK signalling in NGF-induced TRPV1 activation during ischaemia and reperfusion, which may contribute to myocardial injury. It is known that cardiac ischaemia and reperfusion evoke reflex cardiovascular excitation through sympathetic nerves in the spinal cord, resulting in the imbalance of myocardial oxygen supplyedemand, tachycardia, and arrhythmias.26,27 Our recent study revealed an increase in neuronal excitability in the spinal cord because of I/R injury using spinal cord functional MRI (fMRI) and patch clamp recording techniques.28 We further demonstrated that ischaemic preconditioning exerted cardioprotection through inhibiting I/R injury-induced neuronal excitability and neuropeptide expression in the spinal cord.28 However, the activation of sympathetic nerves innervating heart and blood vessels increases heart rate, contractility, and vasoconstriction, which may be deleterious. In previous studies, we have established the method of intrathecal injection and delivered drugs such as opioids to limit cardiac injury, demonstrating an important role of spinal neural signalling in cardioprotection.16,29,30 Our work confirms that now intrathecal application of shRNA is a useful approach for elucidating gene function in pain and nociceptive signalling.31e33 It may be possible to administer small molecules targeting NGF or TRPV1 locally in the spinal cord before surgery to prevent perioperative myocardial injury. This study has several limitations. We detected NGF expression only in DRG and spinal cord but not in the myocardium, although up-regulation of NGF protein levels in post-ischaemic myocardium has been already reported.11,12 Despite showing that NGF enhanced capsaicin-induced currents in DRG cells, the inhibitory effects of NGF-shRNA on spinal cord neuronal activity would be compelling. Although we revealed the involvement of spinal PI3K/Akt and ERK signalling in NGF-induced TRPV1 activation, the use of their specific inhibitors would provide more direct evidence. In conclusion, our study demonstrates that lentivirusmediated NGF gene silencing in the spinal cord reduced the extent of myocardial I/R injury. Although the underlying mechanisms require further elucidation, our findings suggest that spinal NGF is a molecular target for limiting perioperative myocardial I/R injury.
Authors’ contributions Conception of the project: YZ, SFH, WX Design of the study: SFH, YZ, MYD, ZXM Performance of the experiments: MYD, ZXM, XYC, CH Initial data collection and analysis: MYD, ZXM, GCZ, YX Final data analysis and writing of the manuscript: MYD, ZXM, SFH Critical revision of the manuscript: YZ, SFH, XW
Acknowledgements The authors thank Li Zhang, from the Laboratory for Integrative Neuroscience, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health (Bethesda, MD, USA), for reviewing and commenting on the manuscript.
Declaration of interest The authors declare that they have no conflicts of interest.
Funding National Natural Science Foundation of China (No. 81471145 to YZ).
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.bja.2019.06.024.
References 1. Hausenloy DJ, Boston-Griffiths E, Yellon DM. Cardioprotection during cardiac surgery. Cardiovasc Res 2012; 94: 253e65 2. van Waes JA, Nathoe HM, de Graaff JC, et al. Myocardial injury after noncardiac surgery and its association with short-term mortality. Circulation 2013; 127: 2264e71 3. Pan HL, Chen SR. Sensing tissue ischemia: another new function for capsaicin receptors? Circulation 2004; 110: 1826e31 4. Foreman RD. Mechanisms of cardiac pain. Annu Rev Physiol 1999; 61: 143e67 5. Mills CD, Nguyen T, Tanga FY, et al. Characterization of nerve growth factor-induced mechanical and thermal hypersensitivity in rats. Eur J Pain 2012; 17: 469e79 6. Zhu W, Oxford GS. Phosphoinositide-3-kinase and mitogen activated protein kinase signaling pathways mediate acute NGF sensitization of TRPV1. Mol Cell Neurosci 2007; 34: 689e700 7. Zhuang ZY, Xu H, Clapham DE, Ji RR. Phosphatidylinositol 3-kinase activates ERK in primary sensory neurons and mediates inflammatory heat hyperalgesia through TRPV1 sensitization. J Neurosci 2004; 24: 8300e9 8. Bonnington JK, McNaughton PA. Signalling pathways involved in the sensitisation of mouse nociceptive neurones by nerve growth factor. J Physiol 2003; 551: 433e46 9. Yue W, Guo Z. Blockade of spinal nerves inhibits expression of neural growth factor in the myocardium at an early stage of acute myocardial infarction in rats. Br J Anaesth 2012; 109: 345e51 10. Hiltunen JO, Laurikainen A, Vakeva A, Meri S, Saarma M. Nerve growth factor and brain-derived neurotrophic factor mRNAs are regulated in distinct cell populations of rat heart after ischaemia and reperfusion. J Pathol 2001; 194: 247e53 11. Meloni M, Caporali A, Graiani G, et al. Nerve growth factor promotes cardiac repair following myocardial infarction. Circ Res 2010; 106: 1275e84 12. Strande JL, Routhu KV, Lecht S, Lazarovici P. Nerve growth factor reduces myocardial ischemia/reperfusion injury in rat hearts. J Basic Clin Physiol Pharmacol 2013; 24: 81e4
Spinal NGF gene silence reduces cardiac I/R injury
13. Wong LF, Goodhead L, Prat C, Mitrophanous KA, Kingsman SM, Mazarakis ND. Lentivirus-mediated gene transfer to the central nervous system: therapeutic and research applications. Hum Gene Ther 2006; 17: 1e9 14. Palfi S, Gurruchaga JM, Ralph GS, et al. Long-term safety and tolerability of ProSavin, a lentiviral vector-based gene therapy for Parkinson’s disease: a dose escalation, openlabel, phase 1/2 trial. Lancet 2014; 383: 1138e46 15. Yaksh TL, Rudy TA. Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976; 17: 1031e6 16. Jiang L, Hu J, He S, Zhang L, Zhang Y. Spinal neuronal NOS signaling contributes to morphine cardioprotection in ischemia reperfusion injury in rats. J Pharmacol Exp Ther 2016; 358: 450e6 17. Zhang Y, Irwin MG, Wong TM. Remifentanil preconditioning protects against ischemic injury in the intact rat heart. Anesthesiology 2004; 101: 918e23 18. Surinkaew S, Kumphune S, Chattipakorn S, Chattipakorn N. Inhibition of p38 MAPK during ischemia, but not reperfusion, effectively attenuates fatal arrhythmia in ischemia/reperfusion heart. J Cardiovasc Pharmacol 2013; 61: 133e41 19. Repic T, Madirazza K, Bektur E, Sapunar D. Characterization of dorsal root ganglion neurons cultured on silicon micro-pillar substrates. Sci Rep 2016; 6: 39560 20. Zhang X, Huang J, McNaughton PA. NGF rapidly increases membrane expression of TRPV1 heat-gated ion channels. Embo J 2005; 24: 4211e23 21. Faul F, Erdfelder E, Lang AG, Buchner AG. *Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007; 39: 175e91 22. Verge VM, Richardson PM, Wiesenfeld-Hallin Z, Hokfelt T. Differential influence of nerve growth factor on neuropeptide expression in vivo: a novel role in peptide suppression in adult sensory neurons. J Neurosci 1995; 15: 2081e96 23. Shen S, Al-Thumairy HW, Hashmi F, Qiao LY. Regulation of transient receptor potential cation channel subfamily V1 protein synthesis by the phosphoinositide 3-kinase/ Akt pathway in colonic hypersensitivity. Exp Neurol 2017; 295: 104e15
-
11
24. Armour JA. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 1999; 41: 41e54 25. Howard-Quijano K, Takamiya T, Dale EA, et al. Spinal cord stimulation reduces ventricular arrhythmias during acute ischemia by attenuation of regional myocardial excitability. Am J Physiol Heart Circ Physiol 2017; 313: H421e31 26. Huang C, Wang J, Wang N, et al. Effect of myocardial ischemic preconditioning on ischemiaereperfusion stimulation-induced activation in rat thoracic spinal cord with functional MRI. Int J Cardiol 2019; 285: 59e64 27. Li R, Wong GT, Wong TM, Zhang Y, Xia Z, Irwin MG. Intrathecal morphine preconditioning induces cardioprotection via activation of delta, kappa, and mu opioid receptors in rats. Anesth Analg 2009; 108: 23e9 28. Lu Y, Hu J, Zhang Y, Dong C. Spinal neuronal NOS activation mediates intrathecal fentanyl preconditioning induced remote cardioprotection in rats. Int Immunopharmacol 2014; 19: 127e31 29. Meng FF, Xu Y, Dan QQ, et al. Intrathecal injection of lentivirus-mediated glial cell line-derived neurotrophic factor RNA interference relieves bone cancer-induced pain in rats. Cancer Sci 2015; 106: 430e7 30. Zou W, Song Z, Guo Q, Liu C, Zhang Z, Zhang Y. Intrathecal lentiviral-mediated RNA interference targeting PKCgamma attenuates chronic constriction injuryinduced neuropathic pain in rats. Hum Gene Ther 2011; 22: 465e75 31. Hou L, Zhang Y, Yang Y, Xiang K, Tan Q, Guo Q. Intrathecal siRNA against GPNMB attenuates nociception in a rat model of neuropathic pain. J Mol Neurosci 2015; 55: 533e40 32. Park KA, Fehrenbacher JC, Thompson EL, Duarte DB, Hingtgen CM, Vasko MR. Signaling pathways that mediate nerve growth factor-induced increase in expression and release of calcitonin gene-related peptide from sensory neurons. Neuroscience 2010; 171: 910e23 33. Zhong B, Wang DH. TRPV1 gene knockout impairs preconditioning protection against myocardial injury in isolated perfused hearts in mice. Am J Physiol Heart Circ Physiol 2007; 293: H1791e8 Handling editor: G.L. Ackland
LABORATORY INVESTIGATION
Intrathecal lentivirus-mediated RNA interference targeting nerve growth factor attenuates myocardial ischaemiaereperfusion injury in rat Mengyun Dou1,2,y, Zhenxiao Ma1,2,y, Xueying Cheng1,2, Guichang Zou3, Yan Xu3, Cheng Huang1,2, Wei Xiong3, Shufang He1,2,* and Ye Zhang1,2,* 1
Department of Anaesthesiology and Perioperative Medicine, The Second Hospital of Anhui Medical University, Hefei,
China, 2Key Laboratory of Anaesthesiology and Perioperative Medicine of Anhui Higher Education Institutes, Anhui Medical University, Hefei, China and 3Key Laboratory of Brain Function and Disease, School of Life Sciences, University of Science and Technology of China, Chinese Academy of Sciences, Hefei, China *Corresponding authors. E-mails: [email protected], [email protected] y
These authors contributed equally to this work.
Abstract Background: Nerve growth factor (NGF) has been implicated in hyperalgesia by sensitising nociceptors. A role for NGF in modulating myocardial injury through ischaemic nociceptive signalling is plausible. We examined whether inhibition of spinal NGF attenuates myocardial ischaemiaereperfusion injury and explored the underlying mechanisms. Methods: In adult rats, lentivirus-mediated short-hairpin RNA targeted at reducing NGF gene expression (NGF-shRNA) or a transient receptor potential vanilloid 1 (TRPV1) antagonist (capsazepine) was injected intrathecally before myocardial ischaemiaereperfusion. Infarct size (expressed as the ratio of area at risk) and risk of arrhythmias were quantified. Whole-cell clamp patch electrophysiology was used to record capsaicin currents in primary dorsal root ganglion neurones. The co-expression of substance P (SP) and calcitonin gene-related peptide (CGRP), plus activation of TRPV1, protein kinase B (Akt) and extracellular signal-regulated kinase (ERK) were also quantified. Results: NGF levels increased by 2.95 (0.34)-fold in dorsal root ganglion and 2.12 (0.27)-fold in spinal cord after myocardial ischaemiaereperfusion injury. Intrathecal injection of NGF-shRNA reduced infarct area at risk from 0.58 (0.02) to 0.37 (0.02) (P<0.01) and reduced arrhythmia score from 3.67 (0.33) to 1.67 (0.33) (P<0.01). Intrathecal capsazepine was similarly cardioprotective. NGF-shRNA suppressed expression of SP/CGRP and activation of Akt/ERK and TRPV1 in spinal cord. NGF increased capsaicin current amplitude from 144 (42) to 840 (132) pA (P<0.05), which was blocked by the TRPV1 antagonist 50 -iodoresiniferatoxin. Exogenous NGF enhanced capsaicin-induced Akt/ERK and TRPV1 activation in PC12 neuroendocrine tumour cells in culture. Conclusions: Spinal NGF contributes to myocardial ischaemiaereperfusion injury by mediating nociceptive signal transmission. Keywords: lentivirus; myocardial ischaemiaereperfusion; nerve growth factor; nociceptive signal; transient receptor potential vanilloid 1
Editorial decision: 10 June 2019; Accepted: 10 June 2019 © 2019 Published by Elsevier Ltd on behalf of British Journal of Anaesthesia. For Permissions, please email: [email protected]
1
2
-
Dou et al.
Editor’s key points Perioperative myocardial ischaemiaereperfusion injury is associated with excess morbidity and mortality, yet many conventional treatments are ineffective. Nociceptive signalling from ischaemic cardiac tissue may aggravate both infarct size and cardiac arrythmias. Nerve growth factor modulates nociceptive signalling. In this preclinical study, myocardial infarction was reduced by genetic inhibition of nerve growth factor expression using intrathecally delivered small interfering RNA in adult rats.
Myocardial ischaemia and reperfusion injury are independently associated with increased risk of perioperative mortality after cardiac1 and noncardiac surgery.2 Chemical mediators produced during ischaemia and reperfusion, including protons, adenosine, and bradykinin, activate cardiac afferent nerves and transmit ischaemic nociceptive signals to the thoracic spinal cord through dorsal root ganglia (DRG).3,4 This results in cardiac pain and excitatory cardiovascular reflex responses.4 Modulation of cardiac nociceptive signalling may therefore serve as an useful perioperative strategy against myocardial ischaemia/reperfusion (I/R) injury. Nerve growth factor (NGF) is a neurotrophic factor that rapidly sensitises nociceptive sensory neurones to painful stimuli.5 When NGF binds to its high-affinity receptor (tropomyosin receptor kinase A [TrkA]), NGF sensitises transient receptor potential vanilloid 1 (TRPV1), a non-selective cation channel mainly expressed on nociceptive neurones. This leads to enhanced neuronal discharge and the release of neuropeptides.6 NGF induces the sensitisation of TRPV1 through various different signalling pathways,6e8 although the precise mechanisms remain unclear. NGF levels are up-regulated in the myocardium after myocardial infarction.9,10 Local anaesthetic blockade of the spinal nerves abolishes the up-regulation of NGF in ischaemic myocardium, suggesting that NGF may modulate myocardial I/R injury.9 Local increases of NGF in ischaemic myocardium may reduce reperfusion injury and improve post-ischaemic cardiac recovery.11,12 However, the neuronal effect of NGF in myocardial I/R injury has not been elucidated. Lentiviral-mediated gene transfer into neuronal cells has been shown to be effective and safe, without significant immune responses or unwanted side-effects.13,14 We constructed a lentiviral-mediated short hairpin RNA (shRNA) to specifically inhibit spinal NGF gene expression in the thoracic spinal cord via intrathecal administration. Our data suggest that neuronal NGF contributes to myocardial I/R injury through ischaemic nociceptive signal transmission.
Methods Animals Male adult SpragueeDawley (SD) rats, weighing 260 (20) g, or 1week-old SD rats (for neonatal DRG neurone isolation) were used in this study. All animal experiments were performed in accordance with the Guide for the Care and Use of Laboratory Animals (NIH publication No. 85-23, revised 1996), and ARRIVE
(Animal Research: Reporting of In Vivo Experiments) guidelines. The study was approved by the Institutional Animal Care and Use Committee of Anhui Medical University. For animals undergoing surgery, the adequacy of anaesthesia was confirmed by the loss of righting reflex and lack of response to tail and hind limb toe pinch. Buprenorphine (0.05 mg kg1) was administered s.c. every 12 h for the first 3 postoperative days to ensure postoperative analgesia. Animals were euthanised by overdose (100 mg kg1) of pentobarbital sodium injection (adult rats) or inhalation of carbon dioxide (neonatal rats). A total of 60 adult rats and 20 neonatal rats were used in this study. Three rats were excluded because of infection (one rat) and motor dysfunction (two rats) after intrathecal injection. The other five animals were discarded because of death ahead of schedule during ischaemia (two rats, both in the I/R group) and reperfusion (three rats, one in the I/R group and two in the NC-shRNA group).
NGF shRNA We designed lentiviral vectors encoding a shRNA specifically targeting the NGF gene, which was delivered intrathecally at the level of the upper thoracic spinal cord segments (T2eT6). The cell bodies of the spinal cardiac afferent fibres originating from the heart are mainly located in the DRGs of the T2eT6 spinal segments and terminate in the laminae within the same segments of spinal cord.4 We examined the inhibitory effects of the NGF-shRNA in vitro and in vivo, and confirmed NGF at both mRNA and protein levels.
Intrathecal injections of lentiviral-shRNA and TRPV1 antagonist Intrathecal catheterisation at the T2eT6 level of the thoracic spinal cord was performed as described previously.15,16 After recovery for 3 days, the animals received a single injection with NGF-shRNA, NC-shRNA or normal saline in the same volume (10 ml) via the implanted catheter, followed by additional 10 ml saline infusion. The dead space of the catheter lumen was approximately 10 ml. The final virus titres of NGFshRNA and NC-shRNA were 4108 and 1109 TU ml1, respectively, as determined by GFP fluorescence in H293T cells (Supplementary Fig. S1). We established that 7 days was sufficient for shRNA interference before myocardial ischaemia. In another set of experiments, the intrathecally catheterised animals were administrated 10 ml capsazepine (1 mg ml1) and additional 10 ml saline infusion via the catheter. After 10 min, the animals were subjected to myocardial I/R procedure.
Myocardial I/R injury Rats were anaesthetised with 3% pentobarbital sodium i.p. (60 mg kg1), intubated, and ventilated with 100% oxygen. After thoracotomy, ischaemia was achieved by ligating the left anterior descending coronary artery (LAD) using a 6-0 silk suture for 30 min, followed by 120 min reperfusion by releasing the ligature.17 During experiments, ECG, HR, MAP, and rate pressure product (RPP) were continuously recorded. At the end of experiments, the animals were sacrificed via removal of heart under anaesthesia. Heart slices were prepared and stained by triphenyltetrazolium chloride (TTC) for infarct size determination. Tissue samples from thoracic spinal cord and DRG were collected for immunoblot, quantitative reverse
Spinal NGF gene silence reduces cardiac I/R injury
transcriptionepolymerase chain reaction, and immunofluorescence assays.
Arrhythmia score Arrythmia burden after myocardial I/R injury was qualitatively recorded according to the following criteria18: 0, premature ventricular complexes (PVC) <50 times; 1, PVC 50e499 times; 2, PVC >500; 3, ventricular tachycardia/ventricular fibrillation (VT/VF), total duration <1 min; 4, total VT/VF continued 1e2 min; 5, total VT/VF >2 min.
-
3
Cell models: DRG and PC12 cells One-week-old SD rats were sacrificed via inhalation of carbon dioxide followed by decapitation. DRG cells were isolated from T2eT6 segments and maintained in neurobasal medium with 50 ng ml1 NGF for 24 h, followed by 24 h in medium without NGF.19,20 After 48 h of culture, the primary DRG cells were preincubated with 100 ng mL1 NGF for 1 h and then exposed to100 nmol L1 capsaicin, in the presence or absence of TRPV1 specific antagonist 50 -iodoresiniferatoxin (I-RTX; 100 nmol L1), followed by whole-cell clamp patch recording or
Fig 1. Up-regulation of NGF in thoracic DRG and spinal cord after myocardial ischaemiaereperfusion. (a) Representative photographs of heart sections from Sham and ischaemiaereperfusion group. Myocardial infarct size is expressed as a ratio of IS/AAR. (b) Arrhythmia score was determined based on the frequency and duration of arrhythmias during experiment (where 0 was the lowest and 5 was the highest arrhythmia incidence). All above values are presented as mean (SEM); n¼6, **P<0.01 by unpaired t-test. Representative images of NGF protein immunoreactive staining in (c) T2eT6 DRG and (d) spinal dorsal horns. Bar¼50 mm. The relative levels of NGF protein in (e) T2eT6 DRG and (f) spinal cord were normalised by b-actin. The values are expressed as mean (SEM); n¼6, **P<0.01 by unpaired t-test. NGF, nerve growth factor; DRG, dorsal root ganglia; SEM, standard error of the mean; IS/AAR, infarct size/area at risk; I/R, ischaemiaereperfusion.
4
-
Dou et al.
Fig 2. Intrathecal injection of lentiviral NGF-shRNA reduces infarct size and arrhythmia. (a) The GV248 vector carried one pair of oligonucleotides to generate shRNA targeting NGF gene. (b) T2eT6 spinal cord and DRG tissue sections were prepared 3 days after intrathecal injection of lentivirus. Both light field (A, C) and green fluorescent protein (GFP) fluorescence (B, D) were examined under a fluorescent microscope to confirm the in vivo infection efficiency. Bar¼50 mM. The levels of NGF mRNA (c) and protein (d) in thoracic spinal cord were detected 7 days after intrathecal injection of lentivirus. The values are expressed as mean (SEM); n¼3, **P<0.01 as compared with CON group by one-way ANOVA with post hoc Tukey test. (e) Experimental design for intrathecal injection of lentivirus-mediated NGF-shRNA. The effects on myocardial ischaemiaereperfusion infarct size (f), and arrhythmia scores (g) were examined in each group. Values are expressed as mean (SEM); n¼6, ##p<0.01 as compared with ischaemiaereperfusion group by one-way ANOVA with post hoc Tukey test. Data in ischaemiaereperfusion group used here are same as those in Figure 1. NGF, nerve growth factor; shRNA, short hairpin RNA; DRG, dorsal root ganglia; SEM, standard error of the mean; ANOVA, analysis of variance; NC, negative control; CON, control; IS/AAR, infarct size/area at risk; I/R, ischaemiaereperfusion; SEM, standard error of the mean.
Spinal NGF gene silence reduces cardiac I/R injury
immunofluorescence staining. In addition, PC12 neuroendocrine tumour cells were incubated with 100 ng ml1 NGF for 1 h and then exposed to 100 nmol L1 capsaicin for 30 min for western immunoblot analysis. The doses and time of above drugs were determined based on preliminary experiments. Expanded methods are presented in Supplementary Materials.
-
5
Statistical analysis Data are expressed as mean (standard error of the mean, SEM). Quantification of infarct size/area at risk, arrhythmia score, densitometric analysis of protein bands, and patch clamp recordings were undertaken by research personnel
Fig 3. Immunofluorescence staining of substance P (SP) and calcitonin gene-related peptide (CGRP) in DRG neurones and spinal cord. (a) The localisation of SP (green) and CGRP (red) in T2eT6 DRG tissue sections were evaluated by double immunofluorescence. White arrows indicate merge staining (yellow). Bar¼50 mm. (b) The immunostaining of SP or CGRP in T2eT6 spinal cord tissue sections. Bar¼100 mm. ANOVA, analysis of variance; DRG, dorsal root ganglia; NGF, nerve growth factor; I/R, ischaemiaereperfusion; shRNA, short hairpin RNA.
6
-
Dou et al.
masked to treatment allocations. Results were analysed using GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA, USA). Unpaired t-test was used for comparisons between two experimental groups. Comparisons among multiple groups were conducted via one-way analysis of variance (ANOVA) followed by post hoc Tukey’s test. Repeatedmeasures ANOVA followed by a Bonferroni’s test were performed to analyse the haemodynamic parameters at different time points. Statistical significance was set at P<0.05. The sample size in each group was calculated using € t Kiel, GerG*Power Version 3.1.7 (Franz Faul, Universita many),21 based on our previous work.17
Results NGF protein expression in thoracic DRG and spinal cord after myocardial I/R Myocardial I/R injury caused a high ratio of infarct size to area at risk (0.58 [0.02]; Fig. 1a). Arrhythmias occurred commonly during I/R injury (Supplementary Fig. S2), resulting in high arrhythmia scores (Fig. 1b). The immunoreactive staining of NGF protein in DRG (Fig. 1c) and spinal cord (Fig. 1d) were increased after myocardial I/R. Immunoblots showed that NGF protein level was increased by 2.95 (0.34)-fold in DRG (Figs 1e) and 2.12 (0.27)-fold in spinal cord (Fig. 1f), respectively.
Fig 4. NGF enhanced TRPV1 currents and SP/CGRP expression in thoracic DRG neurones. (a) Representative trace records of inward currents elicited by capsaicin (100 nmol L1) in primary DRG neurones. (b) The data points of capsaicin-elicited inward current amplitude (n¼9 from three rats in each group). The values are expressed as mean (SEM); *P<0.05, **P<0.01 by one-way ANOVA with post hoc Tukey test. (c) Representative images of SP (green) and CGRP (red) double immunofluorescence in DRG neurones. The furthest right lane shows the enlarged merge images. Bar¼50 mm. NGF, nerve growth factor; TRPV1, transient receptor potential vanilloid 1; DRG, dorsal root ganglia; SP, substance P; CGRP, calcitonin gene-related peptide; ANOVA, analysis of variance; SEM, standard error of the mean; CON, control; I-RTX, 50 iodoresiniferatoxin; CON, control.
Spinal NGF gene silence reduces cardiac I/R injury
-
7
Fig 5. NGF-dependent activation of Akt and ERK signalling in spinal cord. (a) NGF-shRNA suppressed the protein expression of NGF and its receptor TrkA in spinal cord. (b) The relative expression of NGF or TrkA was normalised to b-actin and the value in Sham group was assigned as 1. (c) The phosphorylated (p) and total Akt protein levels in spinal cord upon NGF-shRNA treatment. (d) The relative expression of pAkt and Akt was normalised to b-actin and the value in Sham group was assigned as 1. (e) The phosphorylated (p) and total ERK protein levels in spinal cord upon NGF-shRNA treatment. (f) The relative expression of pERK was normalised to total ERK and b-actin and the value in Sham group was assigned as 1. The values are expressed as mean (SEM); n¼3, **P<0.01, compared with Sham; #P<0.05, ##P<0.01, compared with ischaemiaereperfusion by one-way ANOVA with post hoc Tukey test. (g) The phosphorylation of Akt and ERK upon exogenous NGF treatment in PC12 cells. (h) The relative expression of and pAkt or pERK was normalised to total Akt or ERK, and b-actin. The value in CON group was assigned as 1. The values are expressed as mean (SEM); n¼3, *P<0.05, compared with CON. NGF, nerve growth factor; Akt, protein kinase B; NGF, nerve growth factor; shRNA, short hairpin RNA; ERK, extracellular signal-regulated kinase; ANOVA, analysis of variance; SEM, standard error of the mean; I/R, ischaemiaereperfusion; CON, control; CAP, capsaicin; NC, negative control.
Inhibition of spinal NGF expression limited myocardial I/R injury We generated three lentiviral vectors that carried shRNA targeting the NGF gene (Supplementary Table S1) and identified an effective shRNA (NGF-shRNA3; Supplementary
Fig. S3). In vivo, the lentiviral vectors transfected thoracic spinal cord and DRG cells, as revealed by GFP fluorescence (Fig. 2b). Both NGF mRNA and protein levels in thoracic spinal cord were reduced by the lentiviral-mediated NGF-shRNA (Fig. 2ced).
8
-
Dou et al.
Fig 6. Contribution of NGF-dependent TRPV1 activation in myocardial ischaemiaereperfusion injury. (a) The phosphorylated (p) and total TRPV1 levels in spinal cord after myocardial ischaemiaereperfusion injury. (b) The relative expression of pTRPV1 and TRPV1 was normalised to b-actin and the value in Sham group was assigned as 1. The values are expressed as mean (SEM); n¼3, **P<0.01, compared with Sham; ##P<0.01, compared with ischaemiaereperfusion by one-way ANOVA with post hoc Tukey test. (c) The expression of pTRPV1 and TRPV1 upon exogenous NGF treatment with or without TRPV1 antagonist I-RTX in PC12 cells. (d) The relative expression of pTRPV1 and TRPV1 was normalised to b-actin. The value in CON group was assigned as 1. The values are expressed as mean (SEM); n¼3, *P<0.05, compared with CON; &P<0.05, compared with NGF by one-way ANOVA with post hoc Tukey test. (e) Myocardial infarct size and (f) arrhythmia scores were reduced by intrathecal administration of TRPV1 antagonist capsazepine. The values are expressed as mean (SEM); n¼6, ##P<0.01, #P<0.05, as compared with ischaemiaereperfusion group by one-way ANOVA with post hoc Tukey test. (g) Suggested mechanism through which lentivirus-mediated NGF-shRNA exerts cardioprotective effects against myocardial ischaemic injury. NGF, nerve growth factor; shRNA, short hairpin RNA; TRPV1, transient receptor potential vanilloid 1; ANOVA, analysis of variance; SEM, standard error of the mean; I/R, ischaemiaereperfusion; TrkA, tropomyosin receptor kinase A; I-RTX, 50 -iodoresiniferatoxin; CON, control; IS/AAR, infarct size/ area at risk.
Spinal NGF gene silence reduces cardiac I/R injury
Rats were randomised to undergo myocardial I/R injury having received either NGF-shRNA or control (NC-shRNA) lentiviral GV248 vectors intrathecally (Fig. 2e). Infarct size was reduced from 0.58 (0.02) to 0.37 (0.02) by NGF-shRNA (Fig. 2f). There were no differences in the ratio of area at risk/left ventricle and right ventricle between groups (Supplementary Fig. S4a). The frequency of myocardial I/R injury-induced arrhythmias was also reduced by intrathecal injection of NGF-shRNA but not NC-shRNA (Fig. 2g). Haemodynamic parameters were similar between shRNA-treated groups (Supplementary Table S2).
Inhibition of spinal NGF reduced the expression of substance P and calcitonin gene-related peptide Double immunofluorescence showed moderate staining of substance P (SP) and calcitonin gene-related peptide (CGRP) in DRG slices from the sham-operated group. After myocardial I/ R injury, both SP and CGRP immunoreactivity were increased and co-localised in DRG neurones. The immunoreactive staining of both SP and CGRP was reduced by NGF-shRNA but not the NC shRNA (Fig. 3a). Similar staining was observed in the spinal cord. Immunofluorescent staining of both SP and CGRP was enhanced after myocardial I/R injury, particularly in the superficial laminal neurones of spinal dorsal horn. This expression was inhibited by NGF-shRNA (Fig. 3b).
NGF enhanced TRPV1 currents and SP/CGRP expression in thoracic DRG neurones Rat thoracic DRG neurones displayed moderate inward currents after application of 100 nmol L1 capsaicin. When preincubated with NGF, the DRG neurones were more sensitive to capsaicin, with an increase in current amplitude from 144 (42) to 840 (132) (P<0.05) However, these NGF-sensitised DRG neurone currents were completely abolished by the addition of I-RTX, the specific TRPV1 antagonist (Fig. 4a and b). SP and CGRP were mostly co-localised in DRG neurones, mainly within cell bodies after capsaicin treatment alone (Fig. 4c). Both neuropeptides were increased after NGF treatment. When TRPV1 channel was blocked by I-RTX, immunostaining of both SP and CGRP was lower (Fig. 4c).
NGF-dependent activation of protein kinase B and extracellular signal-regulated kinase signalling in spinal cord NGF-shRNA reduced not only I/R injury-induced NGF upregulation, but also the expression of TrkA receptor (Fig. 5a and b). To explore the signalling pathway downstream of NGF/TrkA, we further examined the influence of NGF-shRNA on the phosphorylation levels of protein kinase B (Akt) and extracellular signal-regulated kinase (ERK1/2), which represents the activation of these kinases. Myocardial I/R injury increased both the total protein level and the phosphorylation level of Akt, which was reduced by NGF-shRNA (Fig. 5c and d). Phosphorylation of ERK kinase was induced by I/R injury but reduced after NGF gene knockdown. The total protein levels of ERK were similar among groups (Fig. 5e and f). Exogenous NGF treatment enhanced capsaicin-mediated phosphorylation of both Akt and ERK in neuronal PC12 cells (Fig. 5g and h).
-
9
NGF-dependent TRPV1 activation contributed to myocardial I/R injury The activation of Akt/ERK signalling induced by NGF may result in sensitisation of TRPV1 receptor.7 We found that both total and phosphorylated TRPV1 levels were increased in the spinal cord after myocardial I/R injury. Knockdown of NGF gene expression in spinal cord attenuated the increase of total and phosphorylated TRPV1 proteins after myocardial I/R injury (Fig. 6a and b). Moreover, capsaicin-induced up-regulation of TRPV1 in PC12 cells was further enhanced when pretreated with NGF, which was abolished by the specific TRPV1 antagonist I-RTX (Fig. 6c and d).
TRPV1 antagonist reduced infarct size We finally examined whether spinal TRPV1 activation contributes to myocardial I/R injury by intrathecally administering capsazepine, a TRPV1 antagonist. Infarct size was reduced from 0.51 (0.02) to 0.33 (0.02) after TRPV1 inhibition with intrathecal capsazepine (Fig. 6e; Supplementary Fig. S4b). Capsazepine also reduced the arrhythmia score from 4.17 (0.31) to 2.50 (0.43) (Fig. 6f). There were no haemodynamic differences between groups (Supplementary Table S3).
Discussion NGF-induced sensitisation of nociceptive sensory neurones has been implicated in inflammatory and neuropathic hyperalgesia. However, little is known about how NGF participates in sensory neurones that may become sensitised during myocardial ischaemic injury. The present study revealed an up-regulation of NGF protein levels in the upper thoracic DRG and spinal cord in response to myocardial I/R injury. By silencing NGF gene expression using lentivirus-mediated shRNA, we demonstrated, for the first time, that inhibition of spinal NGF expression limited cardiac reperfusion injury. Mechanistically, the cardioprotection afforded by reducing spinal NGF protein levels appears to require TRPV1-dependent nociceptive transmission involving spinal Akt and ERK signalling. NGF promotes the release of SP and CGRP from both peripheral and central endings of sensory neurones.22 These two neuropeptides are involved in the transmission of noxious stimuli, conveying nociceptive signalling from primary afferent fibres to the superficial spinal dorsal horn. We found that SP and CGRP were highly expressed and co-localised in DRG neurones after myocardial I/R injury. SP and CGRP immunoreactive staining was reduced after NGF silencing, indicating that the expression of SP and CGRP is at least partly dependent on spinal NGF. CGRP and SP released from TRPV1expressing cardiac sensory nerves into myocardium may reduce cardiac ischaemic injury, contributing to preconditioning induced cardioprotection.23 In DRG and spinal cord, however, we speculate that the increase of SP and CGRP induced by NGF might be responsible for high neuronal activity after ischaemic stimuli. In support of this hypothesis, we used DRG cells isolated from T2eT6 segments to confirm NGF increased TRPV1-dependent inward currents and the expression of SP and CGRP. Although it is known that NGF increases DRG neuronal sensitivity to capsaicin and neuropeptides, in hyperalgesia or neuropathic pain,6,24 our data also show that NGF increased neuropeptide expression and neuronal activity after cardiac ischaemic injury.
10
-
Dou et al.
NGF induces acute sensitisation of TRPV1 receptor on nociceptive sensory neurones via PI3K/Akt and ERK signalling in inflammatory or injury induced hyperalgesia.6,7,25 In line with previous studies, we found that capsaicin-induced TRPV1 expression and activation were enhanced by NGF treatment, whereas NGF-shRNA suppressed the activation of downstream Akt and ERK signal kinases, and the expression and phosphorylation of TRPV1 receptor. Furthermore, intrathecal administration of TRPV1 antagonist exhibited similar protective effects against myocardial injury, indicating that NGFshRNA may reduce myocardial injury through inhibiting TRPV1 activation. These results suggest the involvement of PI3K/Akt and ERK signalling in NGF-induced TRPV1 activation during ischaemia and reperfusion, which may contribute to myocardial injury. It is known that cardiac ischaemia and reperfusion evoke reflex cardiovascular excitation through sympathetic nerves in the spinal cord, resulting in the imbalance of myocardial oxygen supplyedemand, tachycardia, and arrhythmias.26,27 Our recent study revealed an increase in neuronal excitability in the spinal cord because of I/R injury using spinal cord functional MRI (fMRI) and patch clamp recording techniques.28 We further demonstrated that ischaemic preconditioning exerted cardioprotection through inhibiting I/R injury-induced neuronal excitability and neuropeptide expression in the spinal cord.28 However, the activation of sympathetic nerves innervating heart and blood vessels increases heart rate, contractility, and vasoconstriction, which may be deleterious. In previous studies, we have established the method of intrathecal injection and delivered drugs such as opioids to limit cardiac injury, demonstrating an important role of spinal neural signalling in cardioprotection.16,29,30 Our work confirms that now intrathecal application of shRNA is a useful approach for elucidating gene function in pain and nociceptive signalling.31e33 It may be possible to administer small molecules targeting NGF or TRPV1 locally in the spinal cord before surgery to prevent perioperative myocardial injury. This study has several limitations. We detected NGF expression only in DRG and spinal cord but not in the myocardium, although up-regulation of NGF protein levels in post-ischaemic myocardium has been already reported.11,12 Despite showing that NGF enhanced capsaicin-induced currents in DRG cells, the inhibitory effects of NGF-shRNA on spinal cord neuronal activity would be compelling. Although we revealed the involvement of spinal PI3K/Akt and ERK signalling in NGF-induced TRPV1 activation, the use of their specific inhibitors would provide more direct evidence. In conclusion, our study demonstrates that lentivirusmediated NGF gene silencing in the spinal cord reduced the extent of myocardial I/R injury. Although the underlying mechanisms require further elucidation, our findings suggest that spinal NGF is a molecular target for limiting perioperative myocardial I/R injury.
Authors’ contributions Conception of the project: YZ, SFH, WX Design of the study: SFH, YZ, MYD, ZXM Performance of the experiments: MYD, ZXM, XYC, CH Initial data collection and analysis: MYD, ZXM, GCZ, YX Final data analysis and writing of the manuscript: MYD, ZXM, SFH Critical revision of the manuscript: YZ, SFH, XW
Acknowledgements The authors thank Li Zhang, from the Laboratory for Integrative Neuroscience, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health (Bethesda, MD, USA), for reviewing and commenting on the manuscript.
Declaration of interest The authors declare that they have no conflicts of interest.
Funding National Natural Science Foundation of China (No. 81471145 to YZ).
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.bja.2019.06.024.
References 1. Hausenloy DJ, Boston-Griffiths E, Yellon DM. Cardioprotection during cardiac surgery. Cardiovasc Res 2012; 94: 253e65 2. van Waes JA, Nathoe HM, de Graaff JC, et al. Myocardial injury after noncardiac surgery and its association with short-term mortality. Circulation 2013; 127: 2264e71 3. Pan HL, Chen SR. Sensing tissue ischemia: another new function for capsaicin receptors? Circulation 2004; 110: 1826e31 4. Foreman RD. Mechanisms of cardiac pain. Annu Rev Physiol 1999; 61: 143e67 5. Mills CD, Nguyen T, Tanga FY, et al. Characterization of nerve growth factor-induced mechanical and thermal hypersensitivity in rats. Eur J Pain 2012; 17: 469e79 6. Zhu W, Oxford GS. Phosphoinositide-3-kinase and mitogen activated protein kinase signaling pathways mediate acute NGF sensitization of TRPV1. Mol Cell Neurosci 2007; 34: 689e700 7. Zhuang ZY, Xu H, Clapham DE, Ji RR. Phosphatidylinositol 3-kinase activates ERK in primary sensory neurons and mediates inflammatory heat hyperalgesia through TRPV1 sensitization. J Neurosci 2004; 24: 8300e9 8. Bonnington JK, McNaughton PA. Signalling pathways involved in the sensitisation of mouse nociceptive neurones by nerve growth factor. J Physiol 2003; 551: 433e46 9. Yue W, Guo Z. Blockade of spinal nerves inhibits expression of neural growth factor in the myocardium at an early stage of acute myocardial infarction in rats. Br J Anaesth 2012; 109: 345e51 10. Hiltunen JO, Laurikainen A, Vakeva A, Meri S, Saarma M. Nerve growth factor and brain-derived neurotrophic factor mRNAs are regulated in distinct cell populations of rat heart after ischaemia and reperfusion. J Pathol 2001; 194: 247e53 11. Meloni M, Caporali A, Graiani G, et al. Nerve growth factor promotes cardiac repair following myocardial infarction. Circ Res 2010; 106: 1275e84 12. Strande JL, Routhu KV, Lecht S, Lazarovici P. Nerve growth factor reduces myocardial ischemia/reperfusion injury in rat hearts. J Basic Clin Physiol Pharmacol 2013; 24: 81e4
Spinal NGF gene silence reduces cardiac I/R injury
13. Wong LF, Goodhead L, Prat C, Mitrophanous KA, Kingsman SM, Mazarakis ND. Lentivirus-mediated gene transfer to the central nervous system: therapeutic and research applications. Hum Gene Ther 2006; 17: 1e9 14. Palfi S, Gurruchaga JM, Ralph GS, et al. Long-term safety and tolerability of ProSavin, a lentiviral vector-based gene therapy for Parkinson’s disease: a dose escalation, openlabel, phase 1/2 trial. Lancet 2014; 383: 1138e46 15. Yaksh TL, Rudy TA. Chronic catheterization of the spinal subarachnoid space. Physiol Behav 1976; 17: 1031e6 16. Jiang L, Hu J, He S, Zhang L, Zhang Y. Spinal neuronal NOS signaling contributes to morphine cardioprotection in ischemia reperfusion injury in rats. J Pharmacol Exp Ther 2016; 358: 450e6 17. Zhang Y, Irwin MG, Wong TM. Remifentanil preconditioning protects against ischemic injury in the intact rat heart. Anesthesiology 2004; 101: 918e23 18. Surinkaew S, Kumphune S, Chattipakorn S, Chattipakorn N. Inhibition of p38 MAPK during ischemia, but not reperfusion, effectively attenuates fatal arrhythmia in ischemia/reperfusion heart. J Cardiovasc Pharmacol 2013; 61: 133e41 19. Repic T, Madirazza K, Bektur E, Sapunar D. Characterization of dorsal root ganglion neurons cultured on silicon micro-pillar substrates. Sci Rep 2016; 6: 39560 20. Zhang X, Huang J, McNaughton PA. NGF rapidly increases membrane expression of TRPV1 heat-gated ion channels. Embo J 2005; 24: 4211e23 21. Faul F, Erdfelder E, Lang AG, Buchner AG. *Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007; 39: 175e91 22. Verge VM, Richardson PM, Wiesenfeld-Hallin Z, Hokfelt T. Differential influence of nerve growth factor on neuropeptide expression in vivo: a novel role in peptide suppression in adult sensory neurons. J Neurosci 1995; 15: 2081e96 23. Shen S, Al-Thumairy HW, Hashmi F, Qiao LY. Regulation of transient receptor potential cation channel subfamily V1 protein synthesis by the phosphoinositide 3-kinase/ Akt pathway in colonic hypersensitivity. Exp Neurol 2017; 295: 104e15
-
11
24. Armour JA. Myocardial ischaemia and the cardiac nervous system. Cardiovasc Res 1999; 41: 41e54 25. Howard-Quijano K, Takamiya T, Dale EA, et al. Spinal cord stimulation reduces ventricular arrhythmias during acute ischemia by attenuation of regional myocardial excitability. Am J Physiol Heart Circ Physiol 2017; 313: H421e31 26. Huang C, Wang J, Wang N, et al. Effect of myocardial ischemic preconditioning on ischemiaereperfusion stimulation-induced activation in rat thoracic spinal cord with functional MRI. Int J Cardiol 2019; 285: 59e64 27. Li R, Wong GT, Wong TM, Zhang Y, Xia Z, Irwin MG. Intrathecal morphine preconditioning induces cardioprotection via activation of delta, kappa, and mu opioid receptors in rats. Anesth Analg 2009; 108: 23e9 28. Lu Y, Hu J, Zhang Y, Dong C. Spinal neuronal NOS activation mediates intrathecal fentanyl preconditioning induced remote cardioprotection in rats. Int Immunopharmacol 2014; 19: 127e31 29. Meng FF, Xu Y, Dan QQ, et al. Intrathecal injection of lentivirus-mediated glial cell line-derived neurotrophic factor RNA interference relieves bone cancer-induced pain in rats. Cancer Sci 2015; 106: 430e7 30. Zou W, Song Z, Guo Q, Liu C, Zhang Z, Zhang Y. Intrathecal lentiviral-mediated RNA interference targeting PKCgamma attenuates chronic constriction injuryinduced neuropathic pain in rats. Hum Gene Ther 2011; 22: 465e75 31. Hou L, Zhang Y, Yang Y, Xiang K, Tan Q, Guo Q. Intrathecal siRNA against GPNMB attenuates nociception in a rat model of neuropathic pain. J Mol Neurosci 2015; 55: 533e40 32. Park KA, Fehrenbacher JC, Thompson EL, Duarte DB, Hingtgen CM, Vasko MR. Signaling pathways that mediate nerve growth factor-induced increase in expression and release of calcitonin gene-related peptide from sensory neurons. Neuroscience 2010; 171: 910e23 33. Zhong B, Wang DH. TRPV1 gene knockout impairs preconditioning protection against myocardial injury in isolated perfused hearts in mice. Am J Physiol Heart Circ Physiol 2007; 293: H1791e8 Handling editor: G.L. Ackland