Journal of Pathology J Pathol 2015; 237: 249–262 Published online 22 June 2015 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/path.4565
ORIGINAL PAPER
Tollip is a critical mediator of cerebral ischaemia–reperfusion injury Mingchang Li,1# Bin Feng,2# Lang Wang,3# Sen Guo,3,4 Peng Zhang,3,4 Jun Gong,3,4,5 Yan Zhang,3,4 Ankang Zheng3,4 and Hongliang Li3,4* 1 2 3 4 5
Department of Neurosurgery, Renmin Hospital of Wuhan University, People’s Republic of China School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan, People’s Republic of China Department of Cardiology, Renmin Hospital of Wuhan University, People’s Republic of China Cardiovascular Research Institute, Wuhan University, People’s Republic of China College of Life Sciences, Wuhan University, People’s Republic of China
*Correspondence to: H Li, Department of Cardiology, Renmin Hospital of Wuhan University, Cardiovascular Research Institute, Wuhan University, Jie Fang Road 238, Wuhan 430060, People’s Republic of China. E-mail: [email protected] #
These authors contributed equally to this study.
Abstract Toll-like receptor (TLR) signalling plays an important role in regulating cerebral ischaemia–reperfusion (I/R) injury. Toll-interacting protein (Tollip) is an endogenous negative modulator of TLR signalling that is involved in several inflammatory diseases. Our previous study showed that Tollip inhibits overload-induced cardiac remodelling. However, the role of Tollip in neurological disease remains unknown. In the present study, we proposed that Tollip might contribute to the progression of stroke and confirmed this hypothesis. We found that Tollip expression was significantly increased in I/R-challenged brain tissue of humans, mice and rats in vivo and in primary neurons subjected to oxygen and glucose deprivation in vitro, indicating the involvement of Tollip in I/R injury. Next, using genetic approaches, we revealed that Tollip deficiency protects mice against I/R injury by attenuating neuronal apoptosis and inflammation, as demonstrated by the decreased expression of pro-apoptotic and pro-inflammatory genes and the increased expression of anti-apoptotic genes. By contrast, neuron-specific Tollip over-expression exerted the opposite effect. Mechanistically, the detrimental effects of Tollip on neuronal apoptosis and inflammation following I/R injury were largely mediated by the suppression of Akt signalling. Additionally, to further support our findings, a Tollip knockout rat strain was generated via CRISPR-Cas9-mediated gene inactivation. The Tollip-deficient rats were also protected from I/R injury, based on dramatic decreases in neuronal apoptosis and ischaemic inflammation through Akt activation. Taken together, our findings demonstrate that Tollip acts as a novel modulator of I/R injury by promoting neuronal apoptosis and ischaemic inflammation, which are largely mediated by suppression of Akt signalling. Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
Keywords: Tollip; stroke; apoptosis; Akt; knockout rat
Received 22 January 2015; Revised 11 May 2015; Accepted 20 May 2015
No conflicts of interest were declared.
Introduction Stroke is the second leading cause of morbidity and mortality in industrialized countries. Ischaemic stroke, which is typically caused by middle cerebral artery occlusion (MCAO), accounts for the majority of strokes [1,2]. In the past few decades, although abundant basic research and more than 100 clinical trials in stroke patients have been conducted to explore effective treatments for ischaemic stroke, only thrombolysis using recombinant tissue plasminogen activator (rt-PA) has been approved by the US Food and Drug Administration (FDA) [3]; however, the number of patients who received this thrombolytic therapy is limited to about 10%, due to the 4.5 h time window for rt-PA treatment Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
[4]. Therefore, there is a need for improved and effective therapies for acute stroke. Toll-interacting protein (Tollip), initially identified as an intermediate in interleukin (IL)-1 signalling, is ubiquitously expressed in various tissues, including brain [5]. As an endogenous negative modulator of TLR signalling during inflammatory responses in various immune cells, Tollip prevents TLR2and TLR4-mediated cell signalling via direct binding to TLRs, or via its blocking capacity on IL-1 receptor-associated receptor kinases (IRAKs) [6]. Although TLR-mediated signalling is involved in several neurological diseases, eg Alzheimer’s disease, neonatal brain injury, multiple sclerosis, traumatic brain injury and cerebral ischaemia–reperfusion (I/R) injury J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
250
[7–10], the function of Tollip in neurological disease is largely unknown. In the present study, we show that ischaemic injury is limited in Tollip-deficient mice but exaggerated in Tollip-over-expressing mice. Activation of the Akt signalling pathway mainly accounts for the protective effect observed in Tollip-deficient mice. Based on these data, we propose that Tollip is a novel modulator of cerebral I/R injury, and that inhibition of Tollip in the brain may represent an effective therapeutic strategy for the treatment of cerebral I/R injury.
Materials and methods A detailed description of the materials and methods used is included in Supplementary materials and methods (see supplementary material).
Animals All protocols in the present study were approved by the Animal Care and Use Committee of the Renmin Hospital of Wuhan University, China. All surgeries and subsequent analyses were performed in a blinded manner. Neuron-specific Tollip transgenic mice (Tollip-TG) were generated by micro-injection. Tollip-knockout (KO) mice were purchased from the European Mouse Mutant Archive (EMMA; B6.Cg-Tolliptm1Kbns/Cnrm)]. TLR2-KO (cat. no. 005846) and TLR4-KO mice (cat. no. 007227) were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). Sprague–Dawley rats were purchased from the Vital River Company (Beijing, China; Strain code 101). All the animals were exposed to a 12 h light/dark cycle with controlled temperature and humidity; food and water were given ad libitum.
Generation of Tollip knockout rat Tollip-knockout SD rats were created by CRISPRCas9-mediated gene inactivation, as described previously [11].
Transient middle cerebral artery occlusion (tMCAO) model The mouse and rat tMCAO model was performed as described previously [12,13].
M Li et al
neurological deficits of the rats were assessed using a five-point scale at 24 and 72 h after I/R, as described previously [16].
Adenoviral vectors and neonatal rat neuron cultures Adenoviral vectors containing the rat Tollip gene were used to over-express Tollip. RNA interference techniques were used to silence Tollip expression. Primary cultures of neonatal rat neurons were prepared as described previously [12]. The cultured neurons were exposed to transient oxygen and glucose deprivation (OGD) for 60 min to model I/R in vitro.
LDH assay and cell viability LDH assay and cell viability were examined as previously described [17]. For cell viability, a non-radioactive cell-counting kit-8 (CCK-8) assay (CK04, Dojindo, Kumamoto, Japan) was used, in accordance with the manufacturer’s protocol. LDH release was determined via a colorimetric LDH cytotoxicity assay (G1782, Promega, Madison, WI, USA). Three independent experiments were performed.
Immunofluorescence, TUNEL and Fluoro-Jade B staining The staining protocols used were described previously [12,15]. We prepared 5 μm-thick sections in mouse and 10 μm-thick sections in rat. For TUNEL staining, the sections were stained using an ApopTag® Plus In Situ Apoptosis Fluorescein Detection Kit (S7111, Millipore, Temecula, CA, USA), according to the manufacturer’s protocol. For Fluoro-Jade B staining, the sections were stained with the fluorescent dye Fluoro-Jade B (AG310, Millipore), according to the manufacturer’s protocol. Image analysis was performed using Image Pro Plus 6.0 software. All the counts were determined by a single observer who was blinded to the treatment protocols.
Quantitative real-time PCR (qRT–PCR) Tissue preparation for qRT–PCR analysis was performed as described in our previous study [12,15]. The sequence-specific primers for TNFα, MCP-1, IL-1β, IL-2, ICAM-1, VCAM-1, COX2, F4/80 and GAPDH have been described in our previous studies [15,18,19].
Human brain tissue samples
Western blot analysis
Human samples were obtained from patients with fatal intracerebral haemorrhage (ICH), as described in our previous study [14].
Protein samples from brain tissue or cultured cells were extracted and homogenized in lysis buffer. Western blot analyses were performed using 50 μg of the extracted protein samples. The supernatants were then separated by 10% SDS–PAGE and transferred to an Immobilon-FL membrane. After blocking with Tris-buffered saline (TBS) containing 5% non-fat milk, the membranes were probed with primary antibodies overnight at 4 ∘ C.
Neurological deficit scores and infarct volume The neurological deficits of the mice were assessed using a none-point scale at 24 or 72 h after ischaemia onset, as described in our previous studies [12,15]. The Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
251
Figure 1. Tollip expression is increased in neurons after I/R injury. (A, B) (Left) Western blots showing the expression of Tollip in intracerebral haemorrhage (ICH) and normal human brain tissues (A) and in the brains of WT mice in response to tMCAO injury (B); (right) quantitative analysis of the blot results; n = 4 for humans, n = 3 for mice. (C) Immunofluorescence staining of Tollip (red) and NeuN (green), showing the localization of Tollip in the cortex, striatum and hippocampus of the mouse brain: (right) quantitative analysis of Tollip-expressing neurons in the cortex and striatum; scale bar = 50 μm; *p = 0.0029 versus the contralateral tissue; n = 4. (D) (Left) Western blot showing the expression of Tollip in primary neurons exposed to OGD for 3, 6, 12 or 24 h; (right) quantitative analysis of the western blot results. Data are presented as mean ± SE
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
252
M Li et al
Figure 2. Tollip exaggerates cerebral I/R injury. (A) Brains of Tollip-KO and WT mice were stained with TTC at the indicated time points after I/R injury. (B, C) Quantification of infarct volumes (B) and neurological deficit scores (C) at the indicated time points after I/R injury; n = 5 or 6/group; *p = 0.0023 (B, 24 h), 0.0065 (B, 72 h), 0.0285(B, 7d), 0.0493 (C, 24 h), 0.0203(C, 72 h), 0.0179(C, 7d) versus WT mice. (D) Brains of Tollip-TG3, Tollip-TG4 and NTG mice were stained with TTC at 24 or 72 h after I/R injury. (E, F) Quantification of infarct volumes (E) and neurological deficit scores (F) of Tollip-TG3, Tollip-TG4 and NTG mice at the indicated time points after I/R injury; n = 6/group; *p = 0.0415 [(E) TG3, 24 h], 0.0318 [(E) TG4, 24 h], 0.0388 [(E) TG3, 72 h], 0.0257 [(E) TG4, 72 h], 0.0137 [(F) TG3, 24 h], 0.0255 [(F) TG4, 24 h], 0.0104 [(F) TG3, 72 h] and 0.0443 [(F) TG4, 72 h] versus NTG mice. Scale bar = 10 mm. Data are presented as mean ± SE
Statistical analysis All data are expressed as mean ± standard error (SE). One-way ANOVA followed by post hoc Tukey test was used to determine the differences between the groups. Unpaired Student’s t-test was used to perform comparisons between two groups; p < 0.05 was accepted as significant.
Results Tollip expression is up-regulated following cerebral I/R injury To investigate the potential role of Tollip in stroke, we first examined Tollip expression in brain tissue from ICH patients who presented with secondary cerebral ischaemia. Western blots showed that Tollip expression was significantly up-regulated in the brain samples of ICH patients (Figure 1A). Additionally, a similar time-dependent up-regulation of Tollip expression was observed in the brains of mice subjected to tMCAO and reperfusion (Figure 1B). At 2 h following the onset of ischaemia, the level of Tollip protein was increased by Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
2.49-fold and was further increased by 9.48-fold at 24 h compared with sham injury (Figure 1B). Furthermore, immunofluorescence staining for Tollip and NeuN, a marker of neurons, showed that Tollip expression was significantly increased in neurons of the cortex and the striatum, but not the hippocampus, in the mouse brain after ischaemic injury (Figure 1C). More importantly, consistent with the in vivo results obtained from human and mouse tissues, the up-regulated trends of Tollip were observed in primary neurons isolated from WT mice and exposed to OGD (Figure 1D; see also supplementary material, Figure S1). Taken together, these findings suggest that Tollip may be involved in the progression of ischaemic stroke.
Tollip exacerbates I/R-induced cerebral injury Based on the observation that Tollip expression was significantly increased in both in vivo and in vitro stroke models, we next investigated the function of Tollip in cerebral I/R injury, using Tollip-KO mice. Western blots showed that the Tollip protein was undetectable in the brains of Tollip-KO mice compared with their WT littermates (see supplementary material, Figure S2A). Subsequently, the Tollip-KO and WT mice were subjected to J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
253
Figure 3. Legend on the next page Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
254
tMCAO and reperfusion, in parallel, and the outcome was evaluated. TTC staining showed that the infarct volume was decreased by 63.01% at 24 h, by 70.12% at 72 h and by 57.26% at 7 days, respectively, following ischaemia onset in Tollip-KO mice compared with WT mice (Figure 2A, B). In addition, the neurological deficit scores were reduced in Tollip-KO mice at 24 and 72 h after ischaemia onset (Figure 2C). To further confirm the function of Tollip in promoting the progression of stroke, four independent neuron-specific Tollip-TG mouse lines were established (see supplementary material, Figure S2B) ,with dramatically increased expression of Tollip compared with their non-TG (NTG) littermates. Considering the highest expression in transgenic lines 3 (TG3) and 4 (TG4) (see supplementary material, Figure S2C), mice in these two lines were used to evaluate the effect of Tollip over-expression on stroke. TTC staining and the neurological deficit scores revealed an expanded infarct volume and enhanced neurological deficits in both TG3 and TG4 mice compared with NTG mice after cerebral I/R surgery (Figure 2D–F). Collectively, these data indicate that Tollip exerts an exacerbating effect on I/R-induced brain damage.
Tollip promotes ischaemic injury-induced neuronal apoptosis Neuronal apoptosis is an important pathological process during cerebral I/R injury that determines the outcome of ischaemia [20]. We therefore evaluated whether Tollip exacerbated ischaemic injury by regulating neuronal apoptosis. Fluoro-Jade B and TUNEL staining revealed that neuronal degradation and apoptosis were markedly inhibited in the ischaemic brains of Tollip-KO mice at 24 h after cerebral injury, whereas these processes were up-regulated in Tollip-TG mice (Figure 3A, B). Furthermore, western blots showed that the expression of the pro-apoptotic proteins Bax, Bid, Bak and cleaved caspase-3 was dramatically down-regulated, and that the expression of the anti-apoptotic factor Bcl2 was up-regulated, in the brains of the Tollip-KO mice compared with WT mice, whereas the contrary changes were observed in Tollip-TG mice (Figure 3C). Immunofluorescence staining showed that the expression of Bcl2 was significantly increased in the neurons of Tollip-KO mice but was decreased in those of Tollip-TG mice
M Li et al
(see supplementary material, Figure S3A). More importantly, in accordance with the in vivo results, in vitro experiments showed that Tollip deficiency led to an increase in viable neurons and a decrease in lactate dehydrogenase (LDH) release in cultured neurons exposed to OGD. In contrast, Tollip over-expression exerted the opposite effects (Figure 3D, E). In addition, similar trends of the protein levels of apoptosis-related factors were observed in primary neurons infected with AdshTollip or AdTollip in response to OGD treatment as those observed in the in vivo experiments (see supplementary material, Figure S3B). These data demonstrate that Tollip promotes neuronal apoptosis after cerebral I/R injury.
Tollip enhances post-ischaemic inflammation in the ischaemic brain The inflammatory response after I/R is another important pathological characteristic of ischaemic stroke [21]. The acute expression of pro-inflammatory cytokines, including TNFα, MCP-1, IL-1β, IL-2, VCAM-1, ICAM-1, COX-2 and F4/80, is involved in the response to cerebral injury following stroke [21]. Tollip deficiency and over-expression resulted in decreased and increased expression of these pro-inflammatory cytokines, respectively, as indicated by their mRNA levels at 24 h after ischaemia onset (Figure 4A). Given that NF-κB signalling plays a crucial role in mediating the expression of inflammation-associated genes, we next evaluated the activity of NF-κB signalling. Tollip deficiency inhibited NF-κB activity, as revealed by the decreased protein expression of phosphorylated p65, IκBα and IKKβ in the ischaemic brain, whereas the over-expression of Tollip up-regulated NF-κB activity (Figure 4B). In addition, similar changes in NF-κB signalling were observed in primary neuron cultures infected with AdshTollip or AdTollip that were exposed to OGD (Figure 4C). These findings indicated that Tollip enhances post-ischaemic inflammation.
Tollip exacerbates cerebral ischaemic injury via the inhibition of Akt signalling The potent regulatory capacity of Tollip on ischaemic stroke prompted us to explore the molecular
Figure 3. Tollip deficiency attenuates neuronal apoptosis after I/R injury. (A) Fluoro-Jade B and TUNEL (green) staining, showing apoptotic cells in the cortex of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; scale bar = 50 μm; n = 4/group. (B) Quantitative analysis of Fluoro-Jade B- and TUNEL-positive cells in the cortex of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; *p < 0.0001, = 0.0015 (TG, TUNEL) versus WT or NTG mice. (C) (Top) Western blots, showing the expression of Bcl2, Bax, Bid, Bak and cleaved caspase-3 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham surgery or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, Bcl2, Bid, Bak, cleaved-casp3, Caspase3), = 0.0168 (WT, Bax), < 0.0001 (NTG, Bcl2, Bax, cleaved-casp3, Caspase3), = 0.0022 (NTG, Bid), 0.0035 (NTG, Bak) versus sham WT or NTG mice; # p = 0.0357 (KO, Bcl2), < 0.0001 (KO, Bax, Bak), = 0.0005 (KO, Bid), 0.0002 (KO, cleaved-casp3), 0.0064 (KO, Caspase3), 0.0027 (TG, Bcl2), 0.0057 (TG, Bax), 0.0455 (TG, Bid), 0.0115 (TG, Bak), 0.0338 (TG, cleaved-casp3), 0.0034 (TG, Caspase3) versus I/R-injured WT or NTG mice. (D, E) Cell viability (D) and LDH release (E) analyses indicating the survival status of primary neurons isolated from Tollip-KO, WT, Tollip-TG and NTG mice that were exposed to OGD for 12 or 24 h; *p = 0.0129 [(D) KO, 12 h], 0.0045 [(D) KO, 24 h], 0.0116 [(D) TG, 12 h], 0.0063 [(D) TG, 24 h], 0.0179 [(E) KO, 12 h], 0.0156 [(E) KO, 24 h], 0.0166 [(E) TG, 12 h], 0.0174 [(E) TG, 24 h] versus neurons from WT or NTG mice; n = 9. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
255
Figure 4. Legend on the next page
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
256
mechanisms by which Tollip regulates neuronal apoptosis and inflammation upon ischaemic injury. As the MAPK signalling pathway plays an important role in the regulation of stroke [22], we first evaluated whether Tollip regulates stroke via the MAPK signalling pathway. As shown in Figure 5A, the protein levels of phosphorylated MEK1/2, ERK1/2, JNK1/2 and p38 were up-regulated in ischaemic brain compared with sham-injured brain. However, unexpectedly, no significant change in these protein levels was observed between Tollip-KO and WT mice, or between Tollip-TG and NTG mice (see supplementary material, Figure S4A), indicating that other, more important, molecular events are involved in, and might even be responsible for, the regulatory function of Tollip in cerebral ischaemic injury. Cumulative evidence has shown that Akt signalling is involved in the progress of stroke, based on its modulation of apoptotic and inflammatory responses [23,24]. In this study, we found that the phosphorylation of Akt and its downstream signalling molecules GSK3β and mTOR was up-regulated in the brains of the Tollip-KO mice compared with WT mice, whereas the activation of Akt signalling was dramatically blunted in Tollip-TG mice compared with NTG mice challenged by cerebral I/R injury (Figure 5A). Consistent with this, the blockage of Akt signalling by Tollip was further validated using cultured primary neurons infected with AdshTollip or AdTollip and exposed to OGD treatment (see supplementary material, Figure S4B). Furthermore, immunofluorescence staining showed that a consistent change in the level of phosphorylated Akt was primarily localized to neurons in the ischaemic brain (see supplementary material, Figure S4C). CREB is a downstream nuclear transcription factor of Akt, as Akt activates CREB by phosphorylating CREB at Ser133, resulting in the up-regulation of pro-survival CREB target genes, such as BDNF and Bcl-2 [25]. In the present study, phosphorylated CREB and BDNF were up-regulated in the brains of Tollip-KO mice and were down-regulated in those of Tollip-TG mice (Figure 5B). These data implied that Akt–CREB–BDNF signalling is the likely molecular basis of Tollip’s function in the pathological process of cerebral I/R injury. To confirm
M Li et al
this hypothesis, we modified Akt activity by infecting primary neurons isolated from Tollip-KO or Tollip-TG mice with Ad-dominant negative Akt (Addn-Akt) or Ad-constitutively active Akt (Adca-Akt), respectively. Under basal conditions, no significant difference in neuronal survival was observed between the treatments. However, when challenged with OGD, Addn-Akt infection, which inhibited Akt activity, led to fewer viable neurons and increased LDH release in Tollip-KO neurons compared with WT neurons (Figure 5C, D). By contrast, Tollip-TG neurons infected with Adca-Akt displayed significantly increased viability and decreased LDH release (Figure 5C, D). Taken together, these results demonstrate that Tollip exacerbates cerebral ischaemic injury via the inhibition of Akt signalling.
Tollip deficiency in rats protects against I/R-induced cerebral injury Given the profound effects of both the absence and over-expression of Tollip on cerebral ischaemia– reperfusion in mice, we were interested in whether Tollip exerts the same function in rats. We first measured the expression of Tollip in the rat brain following tMCAO injury and found that, consistent with the mouse brain, Tollip expression was significantly increased in a time-dependent manner, with a 2.86-fold change observed at 6 h after ischaemic injury and a 7.70-fold change observed at 24 h (Figure 6A). This increase was primarily limited to the neurons in the cortex of the rat brain (see supplementary material, Figure S5A). These results indicated that Tollip is involved in the progression of ischaemic injury in rats. Next, to confirm its function in rats, we generated a Tollip-KO rat strain using CRISPR-Cas9 technology. One single-guide RNA (sgRNA) that targets downstream of the 3′ end of exon 1 of the Tollip gene in rat was designed and constructed (Figure 6B). After micro-injecting 28 embryos with Cas9 mRNA and sgRNA, seven pups were generated. PCR products from the genomic DNA of these founders were subjected to the T7 endonuclease 1 (T7E1) cleavage assay. Four founders were detected with cleavage products
Figure 4. Tollip inhibits inflammation in response to I/R injury. (A) Real-time PCR, showing expression of the pro-inflammatory cytokines TNFα, MCP-1, IL-1β, IL-2, VCAM-1, ICAM-1, COX-2 and F4/80 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; *p = 0.0014 (KO, TNFα), 0.0038 (KO, MCP-1), 0.0096 (KO, IL-1β), 0.0361 (KO, IL-2), 0.0070 (KO, VCAM-1), 0.0080 (KO, ICAM-1), 0.0105 (KO, COX-2), 0.0010 (KO, F4/80), 0.0070 (TG, TNFα), 0.0026 (TG, MCP-1), 0.0009 (TG, IL-1β), 0.0055 (TG, IL-2), 0.0226 (TG, VCAM-1), 0.0082 (TG, ICAM-1), 0.0017 (TG, COX-2), 0.0206 (TG, F4/80) versus WT or NTG mice; n = 4. (B) (Top) Western blots, showing the expression of p-IKKβ, p-IκBα and p-p65 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p = 0.0001 (WT, p-IKKβ), 0.0009 (WT, p-IκBα), 0.0002 (WT, IκBα), < 0.0001 (WT, p-p65), < 0.0001 (NTG, p-IKKβ, p-p65), = 0.0001 (NTG, p-IκBα), 0.0001 (NTG, IκBα) versus sham WT or NTG mice; # p = 0.0107 (KO, p-IKKβ), 0.0233 (KO, p-IκBα), 0.0003 (KO, IκBα), 0.0126 (KO, p-p65), 0.0046 (TG, p-IKKβ), 0.0088 (TG, p-IκBα), 0.0003 (TG, IκBα), < 0.0001 (TG, p-p65) versus I/R-injured WT or NTG mice. (C) (Top) Western blots showing the expression of p-IKKβ, p-IκBα and p-p65 in primary neurons infected with AdshTollip, AdshRNA, AdTollip or AdGFP and then either exposed to OGD or not for 6 h; (bottom) quantitative analysis of the western blots; *p < 0.0001 (AdshRNA, p-IKKβ, p-IκBα, IκBα), = 0.0001 (AdshRNA, p-p65), < 0.0001 (AdGFP, p-IKKβ, p-IκBα), 0.0015 (AdGFP, IκBα), = 0.0001 (AdGFP, p-p65) versus the control-AdshRNA or AdGFP; # p < 0.0001 (AdshTollip, p-IKKβ, IκBα), = 0.0002 (AdshTollip, p-IκBα), = 0.0001 (AdshTollip, p-p65), < 0.0001 (AdTollip, p-IKKβ, IκBα), 0.0036 (AdTollip, p-IκBα), 0.0405 (AdTollip, p-p65) versus OGD-AdshRNA or AdGFP. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
257
Figure 5. Legend on the next page Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
258
(Figure 6C), indicating there are mutations in DNA templates of these rats. We then cloned the PCR products of these founders and randomly sequenced several clones from each rat. Seven distinct mutations were identified from four individual founders (Figure 6D). Founder no. 29–7 contained an allele with out-of-frame 16 bp addition, and 1 bp mutation was mated to wild-type Ntac:SD strain to obtain the F1 generation. Heterozygous F1 offspring were interbred to establish the Tollip−/− rat strain (Figure 6E). The Tollip-KO rats grew normally and were active and fertile. Western blots showed that Tollip protein was absent in the brain of Tollip-KO rats (Figure 6F). Subsequently, the Tollip-KO and Tollip+/+ rats underwent tMCAO treatment, with the result that at 24 and 72 h after ischaemia onset, the infarct volumes and neurological deficit scores were dramatically lower in Tollip-KO than in Tollip+/+ rats (Figure 6G–I). In addition, neuronal degeneration and apoptosis were inhibited in Tollip-KO rats (see supplementary material, Figure S5B). The expression of Bcl-2 was increased in Tollip-KO rats, whereas the protein levels of Bax, Bid, Bak and cleaved caspase-3 were consistently decreased (see supplementary material, Figure S5C). We further measured the activity of the NF-κB and Akt signalling pathways. Western blot analysis showed that NF-κB activity was inhibited (see supplementary material, Figure S5D) and that Akt signalling was promoted in Tollip-KO rats compared with Tollip+/+ controls (see supplementary material, Figure S5E, F), in accordance with our observed changes in mice. Collectively, our findings demonstrated that the deletion of Tollip in rats exerts potent protection against I/R-induced cerebral injury.
Discussion Several pathological processes are involved in I/R-induced cerebral injury, such as neuronal degeneration, cellular apoptosis, inflammatory responses and oxidative stress [20,26]. The dysfunction of signal transduction and gene regulation after I/R is the basis for these pathological processes. Thus, identifying novel modulators of the relevant cell signalling pathways and
M Li et al
elucidating their underlying molecular mechanisms are highly warranted. However, it was unclear whether Tollip plays a role in the progression of ischaemic stroke. In the present study, we identified Tollip as a novel regulator of stroke through its inhibition of I/R-induced inflammation and neuronal apoptosis, thereby reducing the infarct area and the extent of neurological deficit; conversely, the over-expression of Tollip led to more serious ischaemic injury in mice. Notably, this reduction in ischaemic injury was even more remarkable in rats lacking Tollip than in WT rats. These data provide the first direct evidence that Tollip mediates cerebral injury after I/R. Originally identified as an adaptor protein in IL-1 receptor signalling, Tollip is extensively expressed in various adult human tissues, including brain tissue [27]; however, the role of Tollip in brain damage induced by I/R insult remains largely unknown. In the present study, we discovered that the expression of Tollip is significantly up-regulated in the brain tissue of ICH patients who experienced secondary cerebral ischaemia. A murine tMCAO model and a neuronal OGD exposure model were used to confirm this change in Tollip expression. Both western blot and immunofluorescence staining analyses showed that Tollip expression was markedly increased after I/R or OGD exposure. This dramatic change in Tollip expression indicated that Tollip might play an important role in the pathological progression of cerebral I/R injury. Neuronal apoptosis, as a determinant of stroke outcome, occurs within several min of ischaemia onset and continues for several days [26]; thus, inhibition of apoptosis of neurons provides a promising strategy for the prevention of ischaemic stroke [28,29]. In the present study, Tollip was found to facilitate neuronal apoptosis, not only in the ischaemic brain but also in OGD-exposed neurons. Corresponding changes of the expression of apoptosis-related molecules, eg Bax, Bad, Bid and cleaved caspase-3, were observed to be induced by artificial up- or down-regulation of Tollip in vivo and in vitro upon I/R injury. Apart from neuronal apoptosis, inflammation is another important pathological process involved in ischaemic stroke, especially during the acute phase [30]. Previous studies demonstrated that Tollip inhibited LPS-induced NF-κB activity
Figure 5. Tollip regulates neuronal apoptosis by suppressing Akt signalling. (A) (Top) Western blots, showing the expression of p-Akt, p-GSK3β and p-mTOR in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, p-Akt, p-GSK3β, p-mTOR), = 0.0001 (NTG, p-Akt), < 0.0001 (NTG, p-GSK3β, p-mTOR) versus the sham WT or NTG mice; # p = 0.0004 (KO, p-Akt), 0.0003 (KO, p-GSK3β), < 0.0001 (KO, p-mTOR), = 0.0078 (TG, p-Akt), 0.0004 (TG, p-GSK3β), < 0.0001 (TG, p-mTOR) versus I/R-injured WT or NTG mice. (B) (Top) Western blots showing the expression of p-CREB and BDNF in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, p-CREB, BDNF), < 0.0001 (NTG, p-CREB), 0.0002 (NTG, BDNF) versus the sham WT or NTG mice; # p < 0.0001 (KO, p-CREB, BDNF), 0.0011 (TG, p-CREB), 0.0002 (TG, BDNF) versus I/R-injured WT or NTG mice. (C, D) Cell viability (C) and LDH release (D) analyses, showing the apoptosis status of primary neurons isolated from Tollip-KO or WT mice infected with AdGFP or Addn-Akt, and that of primary neurons from Tollip-TG or NTG mice infected with AdGFP or Adca-Akt and then either exposed to OGD or not for 24 h; *p = 0.0001 [(C) Addn-AKT], < 0.0001 [(C) Adca-AKT], 0.0155 [(C) Addn-AKT], < 0.0001 [(D) Adca-AKT] versus the AdGFP-infected neurons from WT or NTG mice after OGD treatment for 24 h; # p < 0.0001 [(D) Addn-AKT], < 0.0001 [(D) Adca-AKT], < 0.0001 [(D) Addn-AKT, < 0.0001 [(D) Adca-AKT] versus AdGFP-infected neurons from Tollip-KO or Tollip-TG after OGD treatment for 24 h; n = 9. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
259
Figure 6. Tollip knockout rats generated by Cas9 exhibit protection against I/R injury. (A) (Left) Western blot, showing the expression of Tollip in the rat brain at 6, 12 and 24 h following tMCAO injury; (right) quantitative analysis of the western blots. (B) sgRNA was designed to target downstream of the 3′ end of exon 1 of the rat Tollip gene; the red sequence shows the sgRNA target; the green sequence indicates the protospacer adjacent motif (PAM). (C) Photograph of an agarose gel from the T7 endonuclease 1 digestion assay, showing uncut products (WT) and cut products (mutant). (D) Sequencing results, showing the DNA sequence of WT and four individual founders; additions are shown in bold and the red sequence indicates mutation. (E) Agarose gel electrophoresis, showing the genotype of WT (+/+ ), heterozygous (+/− ) and homozygous (−/− ) rats; the 196 bp band represents the WT allele and the 180 bp band represents the mutant allele. (F) Western blot, showing Tollip protein expression in the brains of the Tollip-KO and WT rats. (G) TTC staining of the brains of Tollip-KO and WT rats at the indicated time points after I/R injury. (H, I) Quantification of infarct volumes (H) and neurological deficit scores (I) at the indicated time points after I/R injury; n = 5 or 6/group; Scale bar = 10 mm. *p = 0.0001 [(H) 24 h], 0.0442 [(H) 72 h], 0.0018 [(I) 24 h], 0.0203 [(I) 72 h] versus WT rats. Data are presented as mean ± SE
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
260
and pro-inflammatory cytokine expression [31,32]. Interestingly, inhibition of the inflammatory response was found with Tollip deficiency after I/R surgery or OGD stimulation, whereas over-expression of Tollip in the brain promoted the inflammatory response. This opposite function of Tollip in regulating cell apoptosis and inflammation depends largely on the cell types, exogenous stimuli and corresponding molecular events. Akt, a key regulator of apoptosis, is normally phosphorylated in the non-ischaemic brain and can be further enhanced after reperfusion in ischaemia models [33–35]. It has been reported that the exogenous application of growth factors protects against stroke by activating the Akt pathway [36–38], and the consequent blockage of Bax, which increases membrane permeability by interacting with pore proteins on the mitochondrial membrane [39]. The profound role of Akt during ischaemic stroke has also been firmly validated in our previous studies [15,19,40]. Apart from its influence on apoptosis, activated Akt showed a capacity to phosphorylate GSK3β, thereby inactivating GSK3β and resulting in the down-regulation of its downstream target NF-κB, potently suppressing the expression of inflammatory genes [41]. Other studies suggested that the downstream factors Akt, mTOR and FOXO1 are also involved in its ability to suppress inflammatory gene expression [42,43]. In the present study, we found that the phosphorylation of Akt and its direct downstream molecules GSK3β, mTOR, CREB and BDNF was up-regulated in the brains of Tollip-KO mice but down-regulated in those of Tollip-TG mice, indicating that Tollip inhibits the Akt signalling pathway in neurons following I/R injury, which is consistent with our previous study showing that Tollip suppressed Akt signalling in the heart [44]. More importantly, suppression of Akt activity by Addn-Akt in the neurons of Tollip-KO mice significantly promoted apoptotic responses, and increased Akt activity by Adca-Akt blocked the promotional effect of Tollip over-expression on neuronal apoptosis. These findings demonstrated that Tollip inhibits neuronal apoptosis largely through suppression of Akt activity. An interesting and seemingly paradoxical observation in our present study is that, as a negative regulator of TLR2/4 in the immune network [6,45], Tollip did not exhibit an opposite function in ischaemic stroke to that of TLR2/4. Considering that both TLRs and Tollip are involved in the process of stroke and can regulate Akt activation [46], a relationship between TLR2/4 and Tollip during ischaemic stroke merits clarification. Surprisingly, although TLR2 and TLR4 respectively enhanced or blunted the phosphorylation of Akt, deficiency of neither TLR2 nor TLR4 affected the level of Tollip expression or Tollip-regulated activation of Akt in response to I/R insult in vivo or OGD challenge in vitro. In turn, no significant influence of Tollip on TLR2/4 was found. More importantly, we identified that Tollip interacts directly with Akt in neurons after OGD treatment (see supplementary material, Figure S6D). Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
M Li et al
Collectively, Tollip might act as a negative regulator of TLR signalling during the immune response in immune cells, whereas during immune-unrelated cellular behaviour in non-immune cells, eg apoptosis of neurons, Tollip might function as an apoptosis-related factor in a TLR-independent manner. In conclusion, the present study provides the first evidence that Tollip mediates cerebral I/R injury by regulating neuronal apoptosis and inflammatory responses. Akt signalling accounts for the biological function of Tollip in this cerebral I/R model. We propose that the targeting of Tollip may result in novel and promising strategies for the mitigation of I/R injury in ischaemic stroke patients.
Acknowledgements This study was supported by the National Natural Science Foundation of China (Grant Nos 81171112, 81371272 and 81170086), the National Science and Technology Support Project (Grant Nos 2011BAI15B02, 2012BAI39B05, 2013YQ030923-05, 2014BAI02B01 and 2015BAI08B01), the Key Project of the National Natural Science Foundation (Grant No. 81330005), the National Basic Research Programme of China (Grant No. 2011CB503902) and the National Science Fund for Distinguished Young Scholars (Grant No. 81425005).
Author contributions In this study, MCL and LW designed and performed the experiments, analysed the data and wrote the manuscript; BF, SG and PZ analysed the data and wrote the manuscript; JG, YZ and AKZ performed the experiments; and HL designed the experiments and wrote the manuscript.
Abbreviations BDNF, brain-derived neurotrophic factor; CREB, cAMP-response element binding protein; COX2, cytochrome c oxidase subunit II; GSK3β, glycogen synthase kinase 3β; ICAM-1, intercellular adhesion molecule 1; ICH, fatal intracerebral hemorrhage; IL, interleukin; I/R, ischaemia–reperfusion; IRAK, interleukin-1 receptor-associated kinase; LDH, lactate dehydrogenase; LPS, lipopolysaccharide; MAPK, mitogen-activated protein kinase; MCAO, middle cerebral artery occlusion; MCP-1, chemokine (C–C motif) ligand; mTOR, mechanistic target of rapamycin; MyD88, myeloid differentiation primary response protein 88; OGD, oxygen and glucose deprivation; rt-PA, recombinant tissue plasminogen activator; TLR, Toll-like receptor; TNFα, tumour necrosis factor-α; Tollip, toll-interacting protein; TRAF1, TNF J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
receptor-associated factor 1; TRAF5, TNF receptorassociated factor 5; TUNEL, terminal dexynucleotidyl transferase (TdT)-mediated dUTP nick end labelling; VCAM-1, vascular cell adhesion molecule 1.
References 1. Bronner LL, Kanter DS, Manson JE. Primary prevention of stroke. N Engl J Med 1995; 333: 1392–1400. 2. Tu JV. Reducing the global burden of stroke: INTERSTROKE. Lancet 2010; 376: 74–75. 3. Lo EH, Dalkara T, Moskowitz MA. Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci 2003; 4: 399–415. 4. Hacke W, Kaste M, Fieschi C, et al. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. The European Cooperative Acute Stroke Study (ECASS). J Am Med Assoc 1995; 274: 1017–1025. 5. Burns K, Clatworthy J, Martin L, et al. Tollip, a new component of the IL-1RI pathway, links IRAK to the IL-1 receptor. Nat Cell Biol 2000; 2: 346–351. 6. Zhang G, Ghosh S. Negative regulation of toll-like receptor-mediated signaling by Tollip. J Biol Chem 2002; 277: 7059–7065. 7. Gambuzza ME, Sofo V, Salmeri FM, et al. Toll-like receptors in Alzheimer’s disease: a therapeutic perspective. CNS Neurol Disord Drug Targets 2014; 13: 1542–1558. 8. Miranda-Hernandez S, Baxter AG. Role of toll-like receptors in multiple sclerosis. Am J Clin Exp Immunol 2013; 2: 75–93. 9. Stridh L, Mottahedin A, Johansson ME, et al. Toll-like receptor-3 activation increases the vulnerability of the neonatal brain to hypoxia-ischemia. J Neurosci 2013; 33: 12041–12051. 10. Wang JH, Ahn IS, Fischer TD, et al. Autophagy suppresses age-dependent ischemia and reperfusion injury in livers of mice. Gastroenterology 2011; 141: 2188–2199, e2186. 11. Hu X, Chang N, Wang X, et al. Heritable gene-targeting with gRNA/Cas9 in rats. Cell Res 2013; 23: 1322–1325. 12. Xiang M, Wang L, Guo S, et al. Interferon regulatory factor 8 protects against cerebral ischaemic–reperfusion injury. J Neurochem 2014; 129: 988–1001. 13. Wu Y, Wang YP, Guo P, et al. A lipoxin A4 analog ameliorates blood–brain barrier dysfunction and reduces MMP-9 expression in a rat model of focal cerebral ischemia–reperfusion injury. J Mol Neurosci 2012; 46: 483–491. 14. Chen HZ, Guo S, Li ZZ, et al. A critical role for interferon regulatory factor 9 in cerebral ischemic stroke. J Neurosci 2014; 34: 11897–11912. 15. Wang L, Lu Y, Zhang X, et al. Mindin is a critical mediator of ischemic brain injury in an experimental stroke model. Exp Neurol 2013; 247: 506–516. 16. Henninger N, Bouley J, Bratane BT, et al. Laser Doppler flowmetry predicts occlusion but not tPA-mediated reperfusion success after rat embolic stroke. Exp Neurol 2009; 215: 290–297. 17. Guo S, Li ZZ, Jiang DS, et al. IRF4 is a novel mediator for neuronal survival in ischaemic stroke. Cell Death Differ 2014; 21: 888–903. 18. Xue Y, Lim S, Yang Y, et al. PDGF-BB modulates hematopoiesis and tumor angiogenesis by inducing erythropoietin production in stromal cells. Nat Med 2012; 18: 100–110. 19. Wang L, Lu Y, Guan H, et al. Tumor necrosis factor receptor-associated factor 5 is an essential mediator of ischemic brain infarction. J Neurochem 2013; 126: 400–414. 20. Broughton BR, Reutens DC, Sobey CG. Apoptotic mechanisms after cerebral ischemia. Stroke 2009; 40: e331–339. 21. Lindsberg PJ, Grau AJ. Inflammation and infections as risk factors for ischemic stroke. Stroke 2003; 34: 2518–2532. Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
261
22. Bogoyevitch MA, Boehm I, Oakley A, et al. Targeting the JNK–MAPK cascade for inhibition: basic science and therapeutic potential. Biochim Biophys Acta 2004; 1697: 89–101. 23. Chan PH. Mitochondria and neuronal death/survival signaling pathways in cerebral ischemia. Neurochem Res 2004; 29: 1943–1949. 24. Zhao H, Shimohata T, Wang JQ, et al. Akt contributes to neuroprotection by hypothermia against cerebral ischemia in rats. J Neurosci 2005; 25: 9794–9806. 25. Zhang L, Zhao H, Zhang X, et al. Nobiletin protects against cerebral ischemia via activating the p-Akt, p-CREB, BDNF and Bcl-2 pathway and ameliorating BBB permeability in rat. Brain Res Bull 2013; 96: 45–53. 26. Brouns R, De Deyn PP. The complexity of neurobiological processes in acute ischemic stroke. Clin Neurol Neurosurg 2009; 111: 483–495. 27. Nishimura M, Naito S. Tissue-specific mRNA expression profiles of human toll-like receptors and related genes. Biol Pharm Bull 2005; 28: 886–892. 28. Rabuffetti M, Sciorati C, Tarozzo G, et al. Inhibition of caspase-1like activity by Ac–Tyr–Val–Ala–Asp–chloromethyl ketone induces long-lasting neuroprotection in cerebral ischemia through apoptosis reduction and decrease of proinflammatory cytokines. J Neurosci 2000; 20: 4398–4404. 29. Yepes M, Sandkvist M, Wong MK, et al. Neuroserpin reduces cerebral infarct volume and protects neurons from ischemia-induced apoptosis. Blood 2000; 96: 569–576. 30. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of inflammatory cells. J Leukoc Biol 2010; 87: 779–789. 31. Byun EB, Yang MS, Choi HG, et al. Quercetin negatively regulates TLR4 signaling induced by lipopolysaccharide through Tollip expression. Biochem Biophys Res Commun 2013; 431: 698–705. 32. Yu JL, Dou DQ, Chen XH, et al. Immunoenhancing activity of protopanaxatriol-type ginsenoside-F3 in murine spleen cells. Acta Pharmacol Sinica 2004; 25: 1671–1676. 33. Li F, Omori N, Jin G, et al. Cooperative expression of survival p-ERK and p-Akt signals in rat brain neurons after transient MCAO. Brain Res 2003; 962: 21–26. 34. Osuka K, Watanabe Y, Usuda N, et al. Modification of endothelial NO synthase through protein phosphorylation after forebrain cerebral ischemia/reperfusion. Stroke 2004; 35: 2582–2586. 35. Jin KL, Mao XO, Nagayama T, et al. Induction of vascular endothelial growth factor receptors and phosphatidylinositol 3′ -kinase/Akt signaling by global cerebral ischemia in the rat. Neuroscience 2000; 100: 713–717. 36. Jin KL, Mao XO, Greenberg DA. Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc Natl Acad Sci USA 2000; 97: 10242–10247. 37. Hui L, Pei DS, Zhang QG, et al. The neuroprotection of insulin on ischemic brain injury in rat hippocampus through negative regulation of JNK signaling pathway by PI3K/Akt activation. Brain Res 2005; 1052: 1–9. 38. Brywe KG, Mallard C, Gustavsson M, et al. IGF-I neuroprotection in the immature brain after hypoxia–ischemia, involvement of Akt and GSK3β? Eur J Neurosci 2005; 21: 1489–1502. 39. Tsuruta F, Masuyama N, Gotoh Y. The phosphatidylinositol 3-kinase (PI3K)–Akt pathway suppresses Bax translocation to mitochondria. J Biol Chem 2002; 277: 14040–14047. 40. Lu YY, Li ZZ, Jiang DS, et al. TRAF1 is a critical regulator of cerebral ischaemia–reperfusion injury and neuronal death. Nat Commun 2013; 4: 2852. 41. Martin M, Rehani K, Jope RS, et al. Toll-like receptor-mediated cytokine production is differentially regulated by glycogen synthase kinase 3. Nat Immunol 2005; 6: 777–784. 42. Brown J, Wang H, Suttles J, et al. Mammalian target of rapamycin complex 2 (mTORC2) negatively regulates Toll-like receptor J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
262
4-mediated inflammatory response via FoxO1. J Biol Chem 2011; 286: 44295–44305. 43. Brown J, Wang H, Hajishengallis GN, et al. TLR-signaling networks: an integration of adaptor molecules, kinases, and cross-talk. J Dent Res 2011; 90: 417–427. 44. Liu Y, Jiang XL, Liu Y, et al. Toll-interacting protein (Tollip) negatively regulates pressure overload-induced ventricular hypertrophy in mice. Cardiovasc Res 2014; 101: 87–96.
M Li et al
45. Lee H, Lee S, Cho IH, et al. Toll-like receptors: sensor molecules for detecting damage to the nervous system. Curr Protein Pept Sci 2013; 14: 33–42. 46. Fallah MP, Chelvarajan RL, Garvy BA, et al. Role of phosphoinositide 3-kinase-Akt signaling pathway in the age-related cytokine dysregulation in splenic macrophages stimulated via TLR-2 or TLR-4 receptors. Mech Ageing Dev 2011; 132: 274–286.
SUPPLEMENTARY MATERIAL ON THE INTERNET The following supplementary material may be found in the online version of this article: Supplementary materials and methods Figure S1. Tollip expression is increased in neurons after OGD treatment Figure S2. The generation and identification of Tollip-KO and neuron-specific Tollip-TG mice Figure S3. Tollip deficiency attenuates neuronal apoptosis after I/R injury Figure S4. Tollip regulates neuronal apoptosis by suppressing Akt signalling Figure S5. Deletion of Tollip in rats protects against I/R injury Figure S6. Tollip regulates the progression of ischaemic stroke in a TLR2/4-independent manner
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
ORIGINAL PAPER
Tollip is a critical mediator of cerebral ischaemia–reperfusion injury Mingchang Li,1# Bin Feng,2# Lang Wang,3# Sen Guo,3,4 Peng Zhang,3,4 Jun Gong,3,4,5 Yan Zhang,3,4 Ankang Zheng3,4 and Hongliang Li3,4* 1 2 3 4 5
Department of Neurosurgery, Renmin Hospital of Wuhan University, People’s Republic of China School of Electronic Information and Communications, Huazhong University of Science and Technology, Wuhan, People’s Republic of China Department of Cardiology, Renmin Hospital of Wuhan University, People’s Republic of China Cardiovascular Research Institute, Wuhan University, People’s Republic of China College of Life Sciences, Wuhan University, People’s Republic of China
*Correspondence to: H Li, Department of Cardiology, Renmin Hospital of Wuhan University, Cardiovascular Research Institute, Wuhan University, Jie Fang Road 238, Wuhan 430060, People’s Republic of China. E-mail: [email protected] #
These authors contributed equally to this study.
Abstract Toll-like receptor (TLR) signalling plays an important role in regulating cerebral ischaemia–reperfusion (I/R) injury. Toll-interacting protein (Tollip) is an endogenous negative modulator of TLR signalling that is involved in several inflammatory diseases. Our previous study showed that Tollip inhibits overload-induced cardiac remodelling. However, the role of Tollip in neurological disease remains unknown. In the present study, we proposed that Tollip might contribute to the progression of stroke and confirmed this hypothesis. We found that Tollip expression was significantly increased in I/R-challenged brain tissue of humans, mice and rats in vivo and in primary neurons subjected to oxygen and glucose deprivation in vitro, indicating the involvement of Tollip in I/R injury. Next, using genetic approaches, we revealed that Tollip deficiency protects mice against I/R injury by attenuating neuronal apoptosis and inflammation, as demonstrated by the decreased expression of pro-apoptotic and pro-inflammatory genes and the increased expression of anti-apoptotic genes. By contrast, neuron-specific Tollip over-expression exerted the opposite effect. Mechanistically, the detrimental effects of Tollip on neuronal apoptosis and inflammation following I/R injury were largely mediated by the suppression of Akt signalling. Additionally, to further support our findings, a Tollip knockout rat strain was generated via CRISPR-Cas9-mediated gene inactivation. The Tollip-deficient rats were also protected from I/R injury, based on dramatic decreases in neuronal apoptosis and ischaemic inflammation through Akt activation. Taken together, our findings demonstrate that Tollip acts as a novel modulator of I/R injury by promoting neuronal apoptosis and ischaemic inflammation, which are largely mediated by suppression of Akt signalling. Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
Keywords: Tollip; stroke; apoptosis; Akt; knockout rat
Received 22 January 2015; Revised 11 May 2015; Accepted 20 May 2015
No conflicts of interest were declared.
Introduction Stroke is the second leading cause of morbidity and mortality in industrialized countries. Ischaemic stroke, which is typically caused by middle cerebral artery occlusion (MCAO), accounts for the majority of strokes [1,2]. In the past few decades, although abundant basic research and more than 100 clinical trials in stroke patients have been conducted to explore effective treatments for ischaemic stroke, only thrombolysis using recombinant tissue plasminogen activator (rt-PA) has been approved by the US Food and Drug Administration (FDA) [3]; however, the number of patients who received this thrombolytic therapy is limited to about 10%, due to the 4.5 h time window for rt-PA treatment Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
[4]. Therefore, there is a need for improved and effective therapies for acute stroke. Toll-interacting protein (Tollip), initially identified as an intermediate in interleukin (IL)-1 signalling, is ubiquitously expressed in various tissues, including brain [5]. As an endogenous negative modulator of TLR signalling during inflammatory responses in various immune cells, Tollip prevents TLR2and TLR4-mediated cell signalling via direct binding to TLRs, or via its blocking capacity on IL-1 receptor-associated receptor kinases (IRAKs) [6]. Although TLR-mediated signalling is involved in several neurological diseases, eg Alzheimer’s disease, neonatal brain injury, multiple sclerosis, traumatic brain injury and cerebral ischaemia–reperfusion (I/R) injury J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
250
[7–10], the function of Tollip in neurological disease is largely unknown. In the present study, we show that ischaemic injury is limited in Tollip-deficient mice but exaggerated in Tollip-over-expressing mice. Activation of the Akt signalling pathway mainly accounts for the protective effect observed in Tollip-deficient mice. Based on these data, we propose that Tollip is a novel modulator of cerebral I/R injury, and that inhibition of Tollip in the brain may represent an effective therapeutic strategy for the treatment of cerebral I/R injury.
Materials and methods A detailed description of the materials and methods used is included in Supplementary materials and methods (see supplementary material).
Animals All protocols in the present study were approved by the Animal Care and Use Committee of the Renmin Hospital of Wuhan University, China. All surgeries and subsequent analyses were performed in a blinded manner. Neuron-specific Tollip transgenic mice (Tollip-TG) were generated by micro-injection. Tollip-knockout (KO) mice were purchased from the European Mouse Mutant Archive (EMMA; B6.Cg-Tolliptm1Kbns/Cnrm)]. TLR2-KO (cat. no. 005846) and TLR4-KO mice (cat. no. 007227) were obtained from the Jackson Laboratory (Bar Harbor, ME, USA). Sprague–Dawley rats were purchased from the Vital River Company (Beijing, China; Strain code 101). All the animals were exposed to a 12 h light/dark cycle with controlled temperature and humidity; food and water were given ad libitum.
Generation of Tollip knockout rat Tollip-knockout SD rats were created by CRISPRCas9-mediated gene inactivation, as described previously [11].
Transient middle cerebral artery occlusion (tMCAO) model The mouse and rat tMCAO model was performed as described previously [12,13].
M Li et al
neurological deficits of the rats were assessed using a five-point scale at 24 and 72 h after I/R, as described previously [16].
Adenoviral vectors and neonatal rat neuron cultures Adenoviral vectors containing the rat Tollip gene were used to over-express Tollip. RNA interference techniques were used to silence Tollip expression. Primary cultures of neonatal rat neurons were prepared as described previously [12]. The cultured neurons were exposed to transient oxygen and glucose deprivation (OGD) for 60 min to model I/R in vitro.
LDH assay and cell viability LDH assay and cell viability were examined as previously described [17]. For cell viability, a non-radioactive cell-counting kit-8 (CCK-8) assay (CK04, Dojindo, Kumamoto, Japan) was used, in accordance with the manufacturer’s protocol. LDH release was determined via a colorimetric LDH cytotoxicity assay (G1782, Promega, Madison, WI, USA). Three independent experiments were performed.
Immunofluorescence, TUNEL and Fluoro-Jade B staining The staining protocols used were described previously [12,15]. We prepared 5 μm-thick sections in mouse and 10 μm-thick sections in rat. For TUNEL staining, the sections were stained using an ApopTag® Plus In Situ Apoptosis Fluorescein Detection Kit (S7111, Millipore, Temecula, CA, USA), according to the manufacturer’s protocol. For Fluoro-Jade B staining, the sections were stained with the fluorescent dye Fluoro-Jade B (AG310, Millipore), according to the manufacturer’s protocol. Image analysis was performed using Image Pro Plus 6.0 software. All the counts were determined by a single observer who was blinded to the treatment protocols.
Quantitative real-time PCR (qRT–PCR) Tissue preparation for qRT–PCR analysis was performed as described in our previous study [12,15]. The sequence-specific primers for TNFα, MCP-1, IL-1β, IL-2, ICAM-1, VCAM-1, COX2, F4/80 and GAPDH have been described in our previous studies [15,18,19].
Human brain tissue samples
Western blot analysis
Human samples were obtained from patients with fatal intracerebral haemorrhage (ICH), as described in our previous study [14].
Protein samples from brain tissue or cultured cells were extracted and homogenized in lysis buffer. Western blot analyses were performed using 50 μg of the extracted protein samples. The supernatants were then separated by 10% SDS–PAGE and transferred to an Immobilon-FL membrane. After blocking with Tris-buffered saline (TBS) containing 5% non-fat milk, the membranes were probed with primary antibodies overnight at 4 ∘ C.
Neurological deficit scores and infarct volume The neurological deficits of the mice were assessed using a none-point scale at 24 or 72 h after ischaemia onset, as described in our previous studies [12,15]. The Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
251
Figure 1. Tollip expression is increased in neurons after I/R injury. (A, B) (Left) Western blots showing the expression of Tollip in intracerebral haemorrhage (ICH) and normal human brain tissues (A) and in the brains of WT mice in response to tMCAO injury (B); (right) quantitative analysis of the blot results; n = 4 for humans, n = 3 for mice. (C) Immunofluorescence staining of Tollip (red) and NeuN (green), showing the localization of Tollip in the cortex, striatum and hippocampus of the mouse brain: (right) quantitative analysis of Tollip-expressing neurons in the cortex and striatum; scale bar = 50 μm; *p = 0.0029 versus the contralateral tissue; n = 4. (D) (Left) Western blot showing the expression of Tollip in primary neurons exposed to OGD for 3, 6, 12 or 24 h; (right) quantitative analysis of the western blot results. Data are presented as mean ± SE
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
252
M Li et al
Figure 2. Tollip exaggerates cerebral I/R injury. (A) Brains of Tollip-KO and WT mice were stained with TTC at the indicated time points after I/R injury. (B, C) Quantification of infarct volumes (B) and neurological deficit scores (C) at the indicated time points after I/R injury; n = 5 or 6/group; *p = 0.0023 (B, 24 h), 0.0065 (B, 72 h), 0.0285(B, 7d), 0.0493 (C, 24 h), 0.0203(C, 72 h), 0.0179(C, 7d) versus WT mice. (D) Brains of Tollip-TG3, Tollip-TG4 and NTG mice were stained with TTC at 24 or 72 h after I/R injury. (E, F) Quantification of infarct volumes (E) and neurological deficit scores (F) of Tollip-TG3, Tollip-TG4 and NTG mice at the indicated time points after I/R injury; n = 6/group; *p = 0.0415 [(E) TG3, 24 h], 0.0318 [(E) TG4, 24 h], 0.0388 [(E) TG3, 72 h], 0.0257 [(E) TG4, 72 h], 0.0137 [(F) TG3, 24 h], 0.0255 [(F) TG4, 24 h], 0.0104 [(F) TG3, 72 h] and 0.0443 [(F) TG4, 72 h] versus NTG mice. Scale bar = 10 mm. Data are presented as mean ± SE
Statistical analysis All data are expressed as mean ± standard error (SE). One-way ANOVA followed by post hoc Tukey test was used to determine the differences between the groups. Unpaired Student’s t-test was used to perform comparisons between two groups; p < 0.05 was accepted as significant.
Results Tollip expression is up-regulated following cerebral I/R injury To investigate the potential role of Tollip in stroke, we first examined Tollip expression in brain tissue from ICH patients who presented with secondary cerebral ischaemia. Western blots showed that Tollip expression was significantly up-regulated in the brain samples of ICH patients (Figure 1A). Additionally, a similar time-dependent up-regulation of Tollip expression was observed in the brains of mice subjected to tMCAO and reperfusion (Figure 1B). At 2 h following the onset of ischaemia, the level of Tollip protein was increased by Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
2.49-fold and was further increased by 9.48-fold at 24 h compared with sham injury (Figure 1B). Furthermore, immunofluorescence staining for Tollip and NeuN, a marker of neurons, showed that Tollip expression was significantly increased in neurons of the cortex and the striatum, but not the hippocampus, in the mouse brain after ischaemic injury (Figure 1C). More importantly, consistent with the in vivo results obtained from human and mouse tissues, the up-regulated trends of Tollip were observed in primary neurons isolated from WT mice and exposed to OGD (Figure 1D; see also supplementary material, Figure S1). Taken together, these findings suggest that Tollip may be involved in the progression of ischaemic stroke.
Tollip exacerbates I/R-induced cerebral injury Based on the observation that Tollip expression was significantly increased in both in vivo and in vitro stroke models, we next investigated the function of Tollip in cerebral I/R injury, using Tollip-KO mice. Western blots showed that the Tollip protein was undetectable in the brains of Tollip-KO mice compared with their WT littermates (see supplementary material, Figure S2A). Subsequently, the Tollip-KO and WT mice were subjected to J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
253
Figure 3. Legend on the next page Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
254
tMCAO and reperfusion, in parallel, and the outcome was evaluated. TTC staining showed that the infarct volume was decreased by 63.01% at 24 h, by 70.12% at 72 h and by 57.26% at 7 days, respectively, following ischaemia onset in Tollip-KO mice compared with WT mice (Figure 2A, B). In addition, the neurological deficit scores were reduced in Tollip-KO mice at 24 and 72 h after ischaemia onset (Figure 2C). To further confirm the function of Tollip in promoting the progression of stroke, four independent neuron-specific Tollip-TG mouse lines were established (see supplementary material, Figure S2B) ,with dramatically increased expression of Tollip compared with their non-TG (NTG) littermates. Considering the highest expression in transgenic lines 3 (TG3) and 4 (TG4) (see supplementary material, Figure S2C), mice in these two lines were used to evaluate the effect of Tollip over-expression on stroke. TTC staining and the neurological deficit scores revealed an expanded infarct volume and enhanced neurological deficits in both TG3 and TG4 mice compared with NTG mice after cerebral I/R surgery (Figure 2D–F). Collectively, these data indicate that Tollip exerts an exacerbating effect on I/R-induced brain damage.
Tollip promotes ischaemic injury-induced neuronal apoptosis Neuronal apoptosis is an important pathological process during cerebral I/R injury that determines the outcome of ischaemia [20]. We therefore evaluated whether Tollip exacerbated ischaemic injury by regulating neuronal apoptosis. Fluoro-Jade B and TUNEL staining revealed that neuronal degradation and apoptosis were markedly inhibited in the ischaemic brains of Tollip-KO mice at 24 h after cerebral injury, whereas these processes were up-regulated in Tollip-TG mice (Figure 3A, B). Furthermore, western blots showed that the expression of the pro-apoptotic proteins Bax, Bid, Bak and cleaved caspase-3 was dramatically down-regulated, and that the expression of the anti-apoptotic factor Bcl2 was up-regulated, in the brains of the Tollip-KO mice compared with WT mice, whereas the contrary changes were observed in Tollip-TG mice (Figure 3C). Immunofluorescence staining showed that the expression of Bcl2 was significantly increased in the neurons of Tollip-KO mice but was decreased in those of Tollip-TG mice
M Li et al
(see supplementary material, Figure S3A). More importantly, in accordance with the in vivo results, in vitro experiments showed that Tollip deficiency led to an increase in viable neurons and a decrease in lactate dehydrogenase (LDH) release in cultured neurons exposed to OGD. In contrast, Tollip over-expression exerted the opposite effects (Figure 3D, E). In addition, similar trends of the protein levels of apoptosis-related factors were observed in primary neurons infected with AdshTollip or AdTollip in response to OGD treatment as those observed in the in vivo experiments (see supplementary material, Figure S3B). These data demonstrate that Tollip promotes neuronal apoptosis after cerebral I/R injury.
Tollip enhances post-ischaemic inflammation in the ischaemic brain The inflammatory response after I/R is another important pathological characteristic of ischaemic stroke [21]. The acute expression of pro-inflammatory cytokines, including TNFα, MCP-1, IL-1β, IL-2, VCAM-1, ICAM-1, COX-2 and F4/80, is involved in the response to cerebral injury following stroke [21]. Tollip deficiency and over-expression resulted in decreased and increased expression of these pro-inflammatory cytokines, respectively, as indicated by their mRNA levels at 24 h after ischaemia onset (Figure 4A). Given that NF-κB signalling plays a crucial role in mediating the expression of inflammation-associated genes, we next evaluated the activity of NF-κB signalling. Tollip deficiency inhibited NF-κB activity, as revealed by the decreased protein expression of phosphorylated p65, IκBα and IKKβ in the ischaemic brain, whereas the over-expression of Tollip up-regulated NF-κB activity (Figure 4B). In addition, similar changes in NF-κB signalling were observed in primary neuron cultures infected with AdshTollip or AdTollip that were exposed to OGD (Figure 4C). These findings indicated that Tollip enhances post-ischaemic inflammation.
Tollip exacerbates cerebral ischaemic injury via the inhibition of Akt signalling The potent regulatory capacity of Tollip on ischaemic stroke prompted us to explore the molecular
Figure 3. Tollip deficiency attenuates neuronal apoptosis after I/R injury. (A) Fluoro-Jade B and TUNEL (green) staining, showing apoptotic cells in the cortex of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; scale bar = 50 μm; n = 4/group. (B) Quantitative analysis of Fluoro-Jade B- and TUNEL-positive cells in the cortex of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; *p < 0.0001, = 0.0015 (TG, TUNEL) versus WT or NTG mice. (C) (Top) Western blots, showing the expression of Bcl2, Bax, Bid, Bak and cleaved caspase-3 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham surgery or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, Bcl2, Bid, Bak, cleaved-casp3, Caspase3), = 0.0168 (WT, Bax), < 0.0001 (NTG, Bcl2, Bax, cleaved-casp3, Caspase3), = 0.0022 (NTG, Bid), 0.0035 (NTG, Bak) versus sham WT or NTG mice; # p = 0.0357 (KO, Bcl2), < 0.0001 (KO, Bax, Bak), = 0.0005 (KO, Bid), 0.0002 (KO, cleaved-casp3), 0.0064 (KO, Caspase3), 0.0027 (TG, Bcl2), 0.0057 (TG, Bax), 0.0455 (TG, Bid), 0.0115 (TG, Bak), 0.0338 (TG, cleaved-casp3), 0.0034 (TG, Caspase3) versus I/R-injured WT or NTG mice. (D, E) Cell viability (D) and LDH release (E) analyses indicating the survival status of primary neurons isolated from Tollip-KO, WT, Tollip-TG and NTG mice that were exposed to OGD for 12 or 24 h; *p = 0.0129 [(D) KO, 12 h], 0.0045 [(D) KO, 24 h], 0.0116 [(D) TG, 12 h], 0.0063 [(D) TG, 24 h], 0.0179 [(E) KO, 12 h], 0.0156 [(E) KO, 24 h], 0.0166 [(E) TG, 12 h], 0.0174 [(E) TG, 24 h] versus neurons from WT or NTG mice; n = 9. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
255
Figure 4. Legend on the next page
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
256
mechanisms by which Tollip regulates neuronal apoptosis and inflammation upon ischaemic injury. As the MAPK signalling pathway plays an important role in the regulation of stroke [22], we first evaluated whether Tollip regulates stroke via the MAPK signalling pathway. As shown in Figure 5A, the protein levels of phosphorylated MEK1/2, ERK1/2, JNK1/2 and p38 were up-regulated in ischaemic brain compared with sham-injured brain. However, unexpectedly, no significant change in these protein levels was observed between Tollip-KO and WT mice, or between Tollip-TG and NTG mice (see supplementary material, Figure S4A), indicating that other, more important, molecular events are involved in, and might even be responsible for, the regulatory function of Tollip in cerebral ischaemic injury. Cumulative evidence has shown that Akt signalling is involved in the progress of stroke, based on its modulation of apoptotic and inflammatory responses [23,24]. In this study, we found that the phosphorylation of Akt and its downstream signalling molecules GSK3β and mTOR was up-regulated in the brains of the Tollip-KO mice compared with WT mice, whereas the activation of Akt signalling was dramatically blunted in Tollip-TG mice compared with NTG mice challenged by cerebral I/R injury (Figure 5A). Consistent with this, the blockage of Akt signalling by Tollip was further validated using cultured primary neurons infected with AdshTollip or AdTollip and exposed to OGD treatment (see supplementary material, Figure S4B). Furthermore, immunofluorescence staining showed that a consistent change in the level of phosphorylated Akt was primarily localized to neurons in the ischaemic brain (see supplementary material, Figure S4C). CREB is a downstream nuclear transcription factor of Akt, as Akt activates CREB by phosphorylating CREB at Ser133, resulting in the up-regulation of pro-survival CREB target genes, such as BDNF and Bcl-2 [25]. In the present study, phosphorylated CREB and BDNF were up-regulated in the brains of Tollip-KO mice and were down-regulated in those of Tollip-TG mice (Figure 5B). These data implied that Akt–CREB–BDNF signalling is the likely molecular basis of Tollip’s function in the pathological process of cerebral I/R injury. To confirm
M Li et al
this hypothesis, we modified Akt activity by infecting primary neurons isolated from Tollip-KO or Tollip-TG mice with Ad-dominant negative Akt (Addn-Akt) or Ad-constitutively active Akt (Adca-Akt), respectively. Under basal conditions, no significant difference in neuronal survival was observed between the treatments. However, when challenged with OGD, Addn-Akt infection, which inhibited Akt activity, led to fewer viable neurons and increased LDH release in Tollip-KO neurons compared with WT neurons (Figure 5C, D). By contrast, Tollip-TG neurons infected with Adca-Akt displayed significantly increased viability and decreased LDH release (Figure 5C, D). Taken together, these results demonstrate that Tollip exacerbates cerebral ischaemic injury via the inhibition of Akt signalling.
Tollip deficiency in rats protects against I/R-induced cerebral injury Given the profound effects of both the absence and over-expression of Tollip on cerebral ischaemia– reperfusion in mice, we were interested in whether Tollip exerts the same function in rats. We first measured the expression of Tollip in the rat brain following tMCAO injury and found that, consistent with the mouse brain, Tollip expression was significantly increased in a time-dependent manner, with a 2.86-fold change observed at 6 h after ischaemic injury and a 7.70-fold change observed at 24 h (Figure 6A). This increase was primarily limited to the neurons in the cortex of the rat brain (see supplementary material, Figure S5A). These results indicated that Tollip is involved in the progression of ischaemic injury in rats. Next, to confirm its function in rats, we generated a Tollip-KO rat strain using CRISPR-Cas9 technology. One single-guide RNA (sgRNA) that targets downstream of the 3′ end of exon 1 of the Tollip gene in rat was designed and constructed (Figure 6B). After micro-injecting 28 embryos with Cas9 mRNA and sgRNA, seven pups were generated. PCR products from the genomic DNA of these founders were subjected to the T7 endonuclease 1 (T7E1) cleavage assay. Four founders were detected with cleavage products
Figure 4. Tollip inhibits inflammation in response to I/R injury. (A) Real-time PCR, showing expression of the pro-inflammatory cytokines TNFα, MCP-1, IL-1β, IL-2, VCAM-1, ICAM-1, COX-2 and F4/80 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 24 h after I/R injury; *p = 0.0014 (KO, TNFα), 0.0038 (KO, MCP-1), 0.0096 (KO, IL-1β), 0.0361 (KO, IL-2), 0.0070 (KO, VCAM-1), 0.0080 (KO, ICAM-1), 0.0105 (KO, COX-2), 0.0010 (KO, F4/80), 0.0070 (TG, TNFα), 0.0026 (TG, MCP-1), 0.0009 (TG, IL-1β), 0.0055 (TG, IL-2), 0.0226 (TG, VCAM-1), 0.0082 (TG, ICAM-1), 0.0017 (TG, COX-2), 0.0206 (TG, F4/80) versus WT or NTG mice; n = 4. (B) (Top) Western blots, showing the expression of p-IKKβ, p-IκBα and p-p65 in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p = 0.0001 (WT, p-IKKβ), 0.0009 (WT, p-IκBα), 0.0002 (WT, IκBα), < 0.0001 (WT, p-p65), < 0.0001 (NTG, p-IKKβ, p-p65), = 0.0001 (NTG, p-IκBα), 0.0001 (NTG, IκBα) versus sham WT or NTG mice; # p = 0.0107 (KO, p-IKKβ), 0.0233 (KO, p-IκBα), 0.0003 (KO, IκBα), 0.0126 (KO, p-p65), 0.0046 (TG, p-IKKβ), 0.0088 (TG, p-IκBα), 0.0003 (TG, IκBα), < 0.0001 (TG, p-p65) versus I/R-injured WT or NTG mice. (C) (Top) Western blots showing the expression of p-IKKβ, p-IκBα and p-p65 in primary neurons infected with AdshTollip, AdshRNA, AdTollip or AdGFP and then either exposed to OGD or not for 6 h; (bottom) quantitative analysis of the western blots; *p < 0.0001 (AdshRNA, p-IKKβ, p-IκBα, IκBα), = 0.0001 (AdshRNA, p-p65), < 0.0001 (AdGFP, p-IKKβ, p-IκBα), 0.0015 (AdGFP, IκBα), = 0.0001 (AdGFP, p-p65) versus the control-AdshRNA or AdGFP; # p < 0.0001 (AdshTollip, p-IKKβ, IκBα), = 0.0002 (AdshTollip, p-IκBα), = 0.0001 (AdshTollip, p-p65), < 0.0001 (AdTollip, p-IKKβ, IκBα), 0.0036 (AdTollip, p-IκBα), 0.0405 (AdTollip, p-p65) versus OGD-AdshRNA or AdGFP. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
257
Figure 5. Legend on the next page Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
258
(Figure 6C), indicating there are mutations in DNA templates of these rats. We then cloned the PCR products of these founders and randomly sequenced several clones from each rat. Seven distinct mutations were identified from four individual founders (Figure 6D). Founder no. 29–7 contained an allele with out-of-frame 16 bp addition, and 1 bp mutation was mated to wild-type Ntac:SD strain to obtain the F1 generation. Heterozygous F1 offspring were interbred to establish the Tollip−/− rat strain (Figure 6E). The Tollip-KO rats grew normally and were active and fertile. Western blots showed that Tollip protein was absent in the brain of Tollip-KO rats (Figure 6F). Subsequently, the Tollip-KO and Tollip+/+ rats underwent tMCAO treatment, with the result that at 24 and 72 h after ischaemia onset, the infarct volumes and neurological deficit scores were dramatically lower in Tollip-KO than in Tollip+/+ rats (Figure 6G–I). In addition, neuronal degeneration and apoptosis were inhibited in Tollip-KO rats (see supplementary material, Figure S5B). The expression of Bcl-2 was increased in Tollip-KO rats, whereas the protein levels of Bax, Bid, Bak and cleaved caspase-3 were consistently decreased (see supplementary material, Figure S5C). We further measured the activity of the NF-κB and Akt signalling pathways. Western blot analysis showed that NF-κB activity was inhibited (see supplementary material, Figure S5D) and that Akt signalling was promoted in Tollip-KO rats compared with Tollip+/+ controls (see supplementary material, Figure S5E, F), in accordance with our observed changes in mice. Collectively, our findings demonstrated that the deletion of Tollip in rats exerts potent protection against I/R-induced cerebral injury.
Discussion Several pathological processes are involved in I/R-induced cerebral injury, such as neuronal degeneration, cellular apoptosis, inflammatory responses and oxidative stress [20,26]. The dysfunction of signal transduction and gene regulation after I/R is the basis for these pathological processes. Thus, identifying novel modulators of the relevant cell signalling pathways and
M Li et al
elucidating their underlying molecular mechanisms are highly warranted. However, it was unclear whether Tollip plays a role in the progression of ischaemic stroke. In the present study, we identified Tollip as a novel regulator of stroke through its inhibition of I/R-induced inflammation and neuronal apoptosis, thereby reducing the infarct area and the extent of neurological deficit; conversely, the over-expression of Tollip led to more serious ischaemic injury in mice. Notably, this reduction in ischaemic injury was even more remarkable in rats lacking Tollip than in WT rats. These data provide the first direct evidence that Tollip mediates cerebral injury after I/R. Originally identified as an adaptor protein in IL-1 receptor signalling, Tollip is extensively expressed in various adult human tissues, including brain tissue [27]; however, the role of Tollip in brain damage induced by I/R insult remains largely unknown. In the present study, we discovered that the expression of Tollip is significantly up-regulated in the brain tissue of ICH patients who experienced secondary cerebral ischaemia. A murine tMCAO model and a neuronal OGD exposure model were used to confirm this change in Tollip expression. Both western blot and immunofluorescence staining analyses showed that Tollip expression was markedly increased after I/R or OGD exposure. This dramatic change in Tollip expression indicated that Tollip might play an important role in the pathological progression of cerebral I/R injury. Neuronal apoptosis, as a determinant of stroke outcome, occurs within several min of ischaemia onset and continues for several days [26]; thus, inhibition of apoptosis of neurons provides a promising strategy for the prevention of ischaemic stroke [28,29]. In the present study, Tollip was found to facilitate neuronal apoptosis, not only in the ischaemic brain but also in OGD-exposed neurons. Corresponding changes of the expression of apoptosis-related molecules, eg Bax, Bad, Bid and cleaved caspase-3, were observed to be induced by artificial up- or down-regulation of Tollip in vivo and in vitro upon I/R injury. Apart from neuronal apoptosis, inflammation is another important pathological process involved in ischaemic stroke, especially during the acute phase [30]. Previous studies demonstrated that Tollip inhibited LPS-induced NF-κB activity
Figure 5. Tollip regulates neuronal apoptosis by suppressing Akt signalling. (A) (Top) Western blots, showing the expression of p-Akt, p-GSK3β and p-mTOR in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, p-Akt, p-GSK3β, p-mTOR), = 0.0001 (NTG, p-Akt), < 0.0001 (NTG, p-GSK3β, p-mTOR) versus the sham WT or NTG mice; # p = 0.0004 (KO, p-Akt), 0.0003 (KO, p-GSK3β), < 0.0001 (KO, p-mTOR), = 0.0078 (TG, p-Akt), 0.0004 (TG, p-GSK3β), < 0.0001 (TG, p-mTOR) versus I/R-injured WT or NTG mice. (B) (Top) Western blots showing the expression of p-CREB and BDNF in the brains of Tollip-KO, WT, Tollip-TG and NTG mice at 6 h after sham or I/R injury; (bottom) quantitative analysis of the western blots; *p < 0.0001 (WT, p-CREB, BDNF), < 0.0001 (NTG, p-CREB), 0.0002 (NTG, BDNF) versus the sham WT or NTG mice; # p < 0.0001 (KO, p-CREB, BDNF), 0.0011 (TG, p-CREB), 0.0002 (TG, BDNF) versus I/R-injured WT or NTG mice. (C, D) Cell viability (C) and LDH release (D) analyses, showing the apoptosis status of primary neurons isolated from Tollip-KO or WT mice infected with AdGFP or Addn-Akt, and that of primary neurons from Tollip-TG or NTG mice infected with AdGFP or Adca-Akt and then either exposed to OGD or not for 24 h; *p = 0.0001 [(C) Addn-AKT], < 0.0001 [(C) Adca-AKT], 0.0155 [(C) Addn-AKT], < 0.0001 [(D) Adca-AKT] versus the AdGFP-infected neurons from WT or NTG mice after OGD treatment for 24 h; # p < 0.0001 [(D) Addn-AKT], < 0.0001 [(D) Adca-AKT], < 0.0001 [(D) Addn-AKT, < 0.0001 [(D) Adca-AKT] versus AdGFP-infected neurons from Tollip-KO or Tollip-TG after OGD treatment for 24 h; n = 9. Data are presented as mean ± SE Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
259
Figure 6. Tollip knockout rats generated by Cas9 exhibit protection against I/R injury. (A) (Left) Western blot, showing the expression of Tollip in the rat brain at 6, 12 and 24 h following tMCAO injury; (right) quantitative analysis of the western blots. (B) sgRNA was designed to target downstream of the 3′ end of exon 1 of the rat Tollip gene; the red sequence shows the sgRNA target; the green sequence indicates the protospacer adjacent motif (PAM). (C) Photograph of an agarose gel from the T7 endonuclease 1 digestion assay, showing uncut products (WT) and cut products (mutant). (D) Sequencing results, showing the DNA sequence of WT and four individual founders; additions are shown in bold and the red sequence indicates mutation. (E) Agarose gel electrophoresis, showing the genotype of WT (+/+ ), heterozygous (+/− ) and homozygous (−/− ) rats; the 196 bp band represents the WT allele and the 180 bp band represents the mutant allele. (F) Western blot, showing Tollip protein expression in the brains of the Tollip-KO and WT rats. (G) TTC staining of the brains of Tollip-KO and WT rats at the indicated time points after I/R injury. (H, I) Quantification of infarct volumes (H) and neurological deficit scores (I) at the indicated time points after I/R injury; n = 5 or 6/group; Scale bar = 10 mm. *p = 0.0001 [(H) 24 h], 0.0442 [(H) 72 h], 0.0018 [(I) 24 h], 0.0203 [(I) 72 h] versus WT rats. Data are presented as mean ± SE
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
260
and pro-inflammatory cytokine expression [31,32]. Interestingly, inhibition of the inflammatory response was found with Tollip deficiency after I/R surgery or OGD stimulation, whereas over-expression of Tollip in the brain promoted the inflammatory response. This opposite function of Tollip in regulating cell apoptosis and inflammation depends largely on the cell types, exogenous stimuli and corresponding molecular events. Akt, a key regulator of apoptosis, is normally phosphorylated in the non-ischaemic brain and can be further enhanced after reperfusion in ischaemia models [33–35]. It has been reported that the exogenous application of growth factors protects against stroke by activating the Akt pathway [36–38], and the consequent blockage of Bax, which increases membrane permeability by interacting with pore proteins on the mitochondrial membrane [39]. The profound role of Akt during ischaemic stroke has also been firmly validated in our previous studies [15,19,40]. Apart from its influence on apoptosis, activated Akt showed a capacity to phosphorylate GSK3β, thereby inactivating GSK3β and resulting in the down-regulation of its downstream target NF-κB, potently suppressing the expression of inflammatory genes [41]. Other studies suggested that the downstream factors Akt, mTOR and FOXO1 are also involved in its ability to suppress inflammatory gene expression [42,43]. In the present study, we found that the phosphorylation of Akt and its direct downstream molecules GSK3β, mTOR, CREB and BDNF was up-regulated in the brains of Tollip-KO mice but down-regulated in those of Tollip-TG mice, indicating that Tollip inhibits the Akt signalling pathway in neurons following I/R injury, which is consistent with our previous study showing that Tollip suppressed Akt signalling in the heart [44]. More importantly, suppression of Akt activity by Addn-Akt in the neurons of Tollip-KO mice significantly promoted apoptotic responses, and increased Akt activity by Adca-Akt blocked the promotional effect of Tollip over-expression on neuronal apoptosis. These findings demonstrated that Tollip inhibits neuronal apoptosis largely through suppression of Akt activity. An interesting and seemingly paradoxical observation in our present study is that, as a negative regulator of TLR2/4 in the immune network [6,45], Tollip did not exhibit an opposite function in ischaemic stroke to that of TLR2/4. Considering that both TLRs and Tollip are involved in the process of stroke and can regulate Akt activation [46], a relationship between TLR2/4 and Tollip during ischaemic stroke merits clarification. Surprisingly, although TLR2 and TLR4 respectively enhanced or blunted the phosphorylation of Akt, deficiency of neither TLR2 nor TLR4 affected the level of Tollip expression or Tollip-regulated activation of Akt in response to I/R insult in vivo or OGD challenge in vitro. In turn, no significant influence of Tollip on TLR2/4 was found. More importantly, we identified that Tollip interacts directly with Akt in neurons after OGD treatment (see supplementary material, Figure S6D). Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
M Li et al
Collectively, Tollip might act as a negative regulator of TLR signalling during the immune response in immune cells, whereas during immune-unrelated cellular behaviour in non-immune cells, eg apoptosis of neurons, Tollip might function as an apoptosis-related factor in a TLR-independent manner. In conclusion, the present study provides the first evidence that Tollip mediates cerebral I/R injury by regulating neuronal apoptosis and inflammatory responses. Akt signalling accounts for the biological function of Tollip in this cerebral I/R model. We propose that the targeting of Tollip may result in novel and promising strategies for the mitigation of I/R injury in ischaemic stroke patients.
Acknowledgements This study was supported by the National Natural Science Foundation of China (Grant Nos 81171112, 81371272 and 81170086), the National Science and Technology Support Project (Grant Nos 2011BAI15B02, 2012BAI39B05, 2013YQ030923-05, 2014BAI02B01 and 2015BAI08B01), the Key Project of the National Natural Science Foundation (Grant No. 81330005), the National Basic Research Programme of China (Grant No. 2011CB503902) and the National Science Fund for Distinguished Young Scholars (Grant No. 81425005).
Author contributions In this study, MCL and LW designed and performed the experiments, analysed the data and wrote the manuscript; BF, SG and PZ analysed the data and wrote the manuscript; JG, YZ and AKZ performed the experiments; and HL designed the experiments and wrote the manuscript.
Abbreviations BDNF, brain-derived neurotrophic factor; CREB, cAMP-response element binding protein; COX2, cytochrome c oxidase subunit II; GSK3β, glycogen synthase kinase 3β; ICAM-1, intercellular adhesion molecule 1; ICH, fatal intracerebral hemorrhage; IL, interleukin; I/R, ischaemia–reperfusion; IRAK, interleukin-1 receptor-associated kinase; LDH, lactate dehydrogenase; LPS, lipopolysaccharide; MAPK, mitogen-activated protein kinase; MCAO, middle cerebral artery occlusion; MCP-1, chemokine (C–C motif) ligand; mTOR, mechanistic target of rapamycin; MyD88, myeloid differentiation primary response protein 88; OGD, oxygen and glucose deprivation; rt-PA, recombinant tissue plasminogen activator; TLR, Toll-like receptor; TNFα, tumour necrosis factor-α; Tollip, toll-interacting protein; TRAF1, TNF J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
Tollip is a mediator of stroke
receptor-associated factor 1; TRAF5, TNF receptorassociated factor 5; TUNEL, terminal dexynucleotidyl transferase (TdT)-mediated dUTP nick end labelling; VCAM-1, vascular cell adhesion molecule 1.
References 1. Bronner LL, Kanter DS, Manson JE. Primary prevention of stroke. N Engl J Med 1995; 333: 1392–1400. 2. Tu JV. Reducing the global burden of stroke: INTERSTROKE. Lancet 2010; 376: 74–75. 3. Lo EH, Dalkara T, Moskowitz MA. Mechanisms, challenges and opportunities in stroke. Nat Rev Neurosci 2003; 4: 399–415. 4. Hacke W, Kaste M, Fieschi C, et al. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke. The European Cooperative Acute Stroke Study (ECASS). J Am Med Assoc 1995; 274: 1017–1025. 5. Burns K, Clatworthy J, Martin L, et al. Tollip, a new component of the IL-1RI pathway, links IRAK to the IL-1 receptor. Nat Cell Biol 2000; 2: 346–351. 6. Zhang G, Ghosh S. Negative regulation of toll-like receptor-mediated signaling by Tollip. J Biol Chem 2002; 277: 7059–7065. 7. Gambuzza ME, Sofo V, Salmeri FM, et al. Toll-like receptors in Alzheimer’s disease: a therapeutic perspective. CNS Neurol Disord Drug Targets 2014; 13: 1542–1558. 8. Miranda-Hernandez S, Baxter AG. Role of toll-like receptors in multiple sclerosis. Am J Clin Exp Immunol 2013; 2: 75–93. 9. Stridh L, Mottahedin A, Johansson ME, et al. Toll-like receptor-3 activation increases the vulnerability of the neonatal brain to hypoxia-ischemia. J Neurosci 2013; 33: 12041–12051. 10. Wang JH, Ahn IS, Fischer TD, et al. Autophagy suppresses age-dependent ischemia and reperfusion injury in livers of mice. Gastroenterology 2011; 141: 2188–2199, e2186. 11. Hu X, Chang N, Wang X, et al. Heritable gene-targeting with gRNA/Cas9 in rats. Cell Res 2013; 23: 1322–1325. 12. Xiang M, Wang L, Guo S, et al. Interferon regulatory factor 8 protects against cerebral ischaemic–reperfusion injury. J Neurochem 2014; 129: 988–1001. 13. Wu Y, Wang YP, Guo P, et al. A lipoxin A4 analog ameliorates blood–brain barrier dysfunction and reduces MMP-9 expression in a rat model of focal cerebral ischemia–reperfusion injury. J Mol Neurosci 2012; 46: 483–491. 14. Chen HZ, Guo S, Li ZZ, et al. A critical role for interferon regulatory factor 9 in cerebral ischemic stroke. J Neurosci 2014; 34: 11897–11912. 15. Wang L, Lu Y, Zhang X, et al. Mindin is a critical mediator of ischemic brain injury in an experimental stroke model. Exp Neurol 2013; 247: 506–516. 16. Henninger N, Bouley J, Bratane BT, et al. Laser Doppler flowmetry predicts occlusion but not tPA-mediated reperfusion success after rat embolic stroke. Exp Neurol 2009; 215: 290–297. 17. Guo S, Li ZZ, Jiang DS, et al. IRF4 is a novel mediator for neuronal survival in ischaemic stroke. Cell Death Differ 2014; 21: 888–903. 18. Xue Y, Lim S, Yang Y, et al. PDGF-BB modulates hematopoiesis and tumor angiogenesis by inducing erythropoietin production in stromal cells. Nat Med 2012; 18: 100–110. 19. Wang L, Lu Y, Guan H, et al. Tumor necrosis factor receptor-associated factor 5 is an essential mediator of ischemic brain infarction. J Neurochem 2013; 126: 400–414. 20. Broughton BR, Reutens DC, Sobey CG. Apoptotic mechanisms after cerebral ischemia. Stroke 2009; 40: e331–339. 21. Lindsberg PJ, Grau AJ. Inflammation and infections as risk factors for ischemic stroke. Stroke 2003; 34: 2518–2532. Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
261
22. Bogoyevitch MA, Boehm I, Oakley A, et al. Targeting the JNK–MAPK cascade for inhibition: basic science and therapeutic potential. Biochim Biophys Acta 2004; 1697: 89–101. 23. Chan PH. Mitochondria and neuronal death/survival signaling pathways in cerebral ischemia. Neurochem Res 2004; 29: 1943–1949. 24. Zhao H, Shimohata T, Wang JQ, et al. Akt contributes to neuroprotection by hypothermia against cerebral ischemia in rats. J Neurosci 2005; 25: 9794–9806. 25. Zhang L, Zhao H, Zhang X, et al. Nobiletin protects against cerebral ischemia via activating the p-Akt, p-CREB, BDNF and Bcl-2 pathway and ameliorating BBB permeability in rat. Brain Res Bull 2013; 96: 45–53. 26. Brouns R, De Deyn PP. The complexity of neurobiological processes in acute ischemic stroke. Clin Neurol Neurosurg 2009; 111: 483–495. 27. Nishimura M, Naito S. Tissue-specific mRNA expression profiles of human toll-like receptors and related genes. Biol Pharm Bull 2005; 28: 886–892. 28. Rabuffetti M, Sciorati C, Tarozzo G, et al. Inhibition of caspase-1like activity by Ac–Tyr–Val–Ala–Asp–chloromethyl ketone induces long-lasting neuroprotection in cerebral ischemia through apoptosis reduction and decrease of proinflammatory cytokines. J Neurosci 2000; 20: 4398–4404. 29. Yepes M, Sandkvist M, Wong MK, et al. Neuroserpin reduces cerebral infarct volume and protects neurons from ischemia-induced apoptosis. Blood 2000; 96: 569–576. 30. Jin R, Yang G, Li G. Inflammatory mechanisms in ischemic stroke: role of inflammatory cells. J Leukoc Biol 2010; 87: 779–789. 31. Byun EB, Yang MS, Choi HG, et al. Quercetin negatively regulates TLR4 signaling induced by lipopolysaccharide through Tollip expression. Biochem Biophys Res Commun 2013; 431: 698–705. 32. Yu JL, Dou DQ, Chen XH, et al. Immunoenhancing activity of protopanaxatriol-type ginsenoside-F3 in murine spleen cells. Acta Pharmacol Sinica 2004; 25: 1671–1676. 33. Li F, Omori N, Jin G, et al. Cooperative expression of survival p-ERK and p-Akt signals in rat brain neurons after transient MCAO. Brain Res 2003; 962: 21–26. 34. Osuka K, Watanabe Y, Usuda N, et al. Modification of endothelial NO synthase through protein phosphorylation after forebrain cerebral ischemia/reperfusion. Stroke 2004; 35: 2582–2586. 35. Jin KL, Mao XO, Nagayama T, et al. Induction of vascular endothelial growth factor receptors and phosphatidylinositol 3′ -kinase/Akt signaling by global cerebral ischemia in the rat. Neuroscience 2000; 100: 713–717. 36. Jin KL, Mao XO, Greenberg DA. Vascular endothelial growth factor: direct neuroprotective effect in in vitro ischemia. Proc Natl Acad Sci USA 2000; 97: 10242–10247. 37. Hui L, Pei DS, Zhang QG, et al. The neuroprotection of insulin on ischemic brain injury in rat hippocampus through negative regulation of JNK signaling pathway by PI3K/Akt activation. Brain Res 2005; 1052: 1–9. 38. Brywe KG, Mallard C, Gustavsson M, et al. IGF-I neuroprotection in the immature brain after hypoxia–ischemia, involvement of Akt and GSK3β? Eur J Neurosci 2005; 21: 1489–1502. 39. Tsuruta F, Masuyama N, Gotoh Y. The phosphatidylinositol 3-kinase (PI3K)–Akt pathway suppresses Bax translocation to mitochondria. J Biol Chem 2002; 277: 14040–14047. 40. Lu YY, Li ZZ, Jiang DS, et al. TRAF1 is a critical regulator of cerebral ischaemia–reperfusion injury and neuronal death. Nat Commun 2013; 4: 2852. 41. Martin M, Rehani K, Jope RS, et al. Toll-like receptor-mediated cytokine production is differentially regulated by glycogen synthase kinase 3. Nat Immunol 2005; 6: 777–784. 42. Brown J, Wang H, Suttles J, et al. Mammalian target of rapamycin complex 2 (mTORC2) negatively regulates Toll-like receptor J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
262
4-mediated inflammatory response via FoxO1. J Biol Chem 2011; 286: 44295–44305. 43. Brown J, Wang H, Hajishengallis GN, et al. TLR-signaling networks: an integration of adaptor molecules, kinases, and cross-talk. J Dent Res 2011; 90: 417–427. 44. Liu Y, Jiang XL, Liu Y, et al. Toll-interacting protein (Tollip) negatively regulates pressure overload-induced ventricular hypertrophy in mice. Cardiovasc Res 2014; 101: 87–96.
M Li et al
45. Lee H, Lee S, Cho IH, et al. Toll-like receptors: sensor molecules for detecting damage to the nervous system. Curr Protein Pept Sci 2013; 14: 33–42. 46. Fallah MP, Chelvarajan RL, Garvy BA, et al. Role of phosphoinositide 3-kinase-Akt signaling pathway in the age-related cytokine dysregulation in splenic macrophages stimulated via TLR-2 or TLR-4 receptors. Mech Ageing Dev 2011; 132: 274–286.
SUPPLEMENTARY MATERIAL ON THE INTERNET The following supplementary material may be found in the online version of this article: Supplementary materials and methods Figure S1. Tollip expression is increased in neurons after OGD treatment Figure S2. The generation and identification of Tollip-KO and neuron-specific Tollip-TG mice Figure S3. Tollip deficiency attenuates neuronal apoptosis after I/R injury Figure S4. Tollip regulates neuronal apoptosis by suppressing Akt signalling Figure S5. Deletion of Tollip in rats protects against I/R injury Figure S6. Tollip regulates the progression of ischaemic stroke in a TLR2/4-independent manner
Copyright © 2015 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd. www.pathsoc.org.uk
J Pathol 2015; 237: 249–262 www.thejournalofpathology.com
