Deletion of Mitochondrial Uncoupling Protein-2 Increases Ischemic Brain Damage after Transient Focal Ischemia by Altering Gene Expression Patterns and Enhancing Inflammatory Cytokines

Abstract

Mitochondrial hyperpolarization inhibits the electron transport chain and increases incomplete reduction of oxygen, enabling production of reactive oxygen species (ROS). The consequence is mitochondrial damage that eventually causes cell death. Uncoupling proteins (UCPs) are inner mitochondrial membrane proteins that dissipate the mitochondrial proton gradient by transporting H⁺ across the inner membrane, thereby stabilizing the inner mitochondrial membrane potential and reducing the formation of ROS. The role of UCP2 in neuroprotection is still in debate. This study seeks to clarify the role of UCP2 in transient focal ischemia (tFI) and to further understand the mechanisms of ischemic brain damage. Both wild-type and UCP2-knockout mice were subjected to tFI. Knocking out UCP2 significantly increased the infarct volume to 61% per hemisphere as compared with 18% in wild-type animals. Knocking out UCP2 suppressed antioxidant, cell-cycle, and DNA repair genes, including Sod1 and Sod2, Gstm1, and cyclins. Furthermore, knocking out UCP2 significantly upregulated the protein levels of the inflammatory cytokines, including CTACK, CXCL16, Eotaxin-2, fractalkine, and BLC. It is concluded that knocking out the UCP2 gene exacerbates neuronal death after cerebral ischemia with reperfusion and this detrimental effect is mediated by alteration of antioxidant genes and upregulation of inflammatory mediators.

Keywords

chemokines MCAO mitochondria neuronal death stroke UCP

Introduction

Oxidative stress and inflammation are crucial factors in mediating neuronal death after cerebral ischemia–reperfusion injury (Chan, 1996). It is well established that reactive oxygen species (ROS) production is increased after cerebral ischemia and reperfusion; such increases initiate the expression of inflammatory cytokines (Minami et al, 2006), which in turn stimulate the innate immune system to generate more ROS, creating a positive feedback mechanism. Inflammation signaling occurs directly in the brain and in the peripheral immune system after stroke (Offner et al, 2006). It has been reported that the levels of inflammatory cytokines and chemokines are increased after focal ischemia (Minami et al, 2006). Chemokines are cytokines that have the ability to induce chemotaxis on neighboring cells, particularly those involved in inflammatory actions. While some cytokines may offer protection, many cytokines and most chemokines have been shown to participate in the neuronal damage processes because of the neuroprotective effect observed after inhibition of inflammatory cytokines and chemokines achieved by neutralizing antibodies, gene knockouts, or pharmacological inhibition (Beech et al, 2001; Minami et al, 2006).

Published data have shown that occurrence of hyperpolarization of mitochondrial membrane potential precedes ROS production and cell death in cultured neurons exposed to oxygen and/or glucose deprivation (Iijima et al, 2003; Ouyang et al, 2006). Inhibition of hyperpolarization protected neuronal cells from oxidative stress-induced cell death (Choi et al, 2009). It is likely that mitochondrial hyperpolarization slows down the speed of electron transport, thereby increasing the chance for incomplete oxygen reduction when there is a persistent flow of electrons from NADH and FADH₂, a condition that occurs after recirculation or re-oxygenation after stroke in vivo or hypoxia in vitro. Uncoupling proteins (UCPs) are located in the inner mitochondrial membrane and function to transport protons into the mitochondrial matrix. UCPs were first described for their role in generating heat without shivering in brown adipose tissue. Subsequent studies showed that reduction of the proton motor force across the mitochondrial inner membrane by UCP2 decreased the formation of ROS (Mehta and Li, 2009). A small reduction in the mitochondrial membrane potential induced by mild uncoupling has a significant effect in ameliorating ROS production (Teshima et al, 2003). UCP2 is ubiquitously expressed in all tissues, with more levels in the brain, liver, and spleen at levels 1000 times less than UCP1 in brown adipose tissue. The proposed functions of UCP2 include preventing the formation of ROS and atherosclerosis, participation in inflammation, regulation of body weight, adaptive thermogenesis, and aging (Jezek, 2002).

The role of UCP2 in brain ischemic stroke is still a matter of debate. Upregulation of UCP2 and UCP5 genes has been reported to reduce neuronal damage in cerebral stroke and brain trauma injury models (Bechmann et al, 2002; Kim et al, 2007; Mattiasson et al, 2003; Nakase et al, 2007). Contradictorily, de Bilbao et al (2004) observed that deletion of UCP2 gene in mice conferred resistance to ischemic brain damage in a permanent middle cerebral arterial occlusion (MCAO) model. Della-Morte et al (2009) later suggested the protective role of resveratrol was due to inhibition of the expression of UCP2. Furthermore, it is not clear whether UCP2 exerts any influence on the expressions of oxidation-related genes and inflammatory cytokines after transient cerebral ischemia. The objectives of this study were to clarify the effects of UCP2 on the outcome of brain ischemic stroke and its influences on gene expression patterns and inflammatory cytokines. We induced 60-minute transient focal ischemia (tFI) in both wild-type and UCP2 homozygous knockout (UCP^−/−) mice and collected brain tissues at 5 and 24 hours after reperfusion for determination of infarct volume, gene expression pattern, and inflammatory cytokine enzyme activities. Our results showed that deletion of the UCP2 gene significantly increased the infarct volume, suppressed neuroprotective genes, and increased the levels of inflammatory proteins.

Materials and methods

Animals

Eighty-three animals (41 wild-type and 42 UCP2^−/− mice) were used in this study. The experimental group and animal numbers are listed in Table 1. All procedures were performed in strict compliance with the National Institutes of Health guidelines for animal research and were approved by the University of Hawaii's Institutional Animal Use and Care Committee. UCP2^−/− mice were generated by targeting the UCP2 exons 3 and 4, and replacing it with a neomycin cassette (Arsenijevic et al, 2000). Mice were backcrossed onto a C57/Bl6 background. Mice were fasted overnight with free access to water. Anesthesia was induced with 3% and maintained at 1 to 1.5% isoflurane in 30% oxygen and 70% nitrous oxide. Body temperatures were maintained between 36.5°C and 37.5°C with a combination of a heating blanket and a lamp. Animals with fasting blood glucose level between 4 and 6mmol/L were used for the experiment. Pre-ischemic blood pressure was 100 to 120 mmHg, with no statistical difference between the two types of animals.

Table 1

Number of animals in each group

	TTC	India ink	PCR array	Protein array	IHC
Wild-type
Control	—	4	4	4	4
5-h Reperfusion	—	—	—	—	4
24-h Reperfusion	8	—	6	3	4
UCP^−/−
Control	—	4	4	4	4
5-h Reperfusion	—	—	—	—	4
24-h Reperfusion	8	—	6	4	4

IHC, immunohistochemistry; PCR, polymerase chain reaction; TTC, 2,3,5-triphenyltetrazolium chloride; UCP, uncoupling protein.

Ischemic Model

Transient MCAO was induced as described before (Li et al, 1998). Briefly, a nylon monofilament (Doccol, Redlands, CA, USA) coated with silicon, size 6 to 0, was inserted into the common carotid artery to the internal carotid artery to block the MCA. After occlusion, the mice were examined and only animals with neurologic signs of diminished resistance to lateral push, walking to the left after being pulled backwards by the tail, or with spontaneous contralateral circling, were included in the study. After 1 hour of ischemia, mice were re-anesthetized and the filament was removed to restore blood flow.

Measuring Brain Infarction

Eight mice in each group were used to determine infarct volume by 2,3,5-triphenyltetrazolium chloride (TTC) staining. After 1 hour of MCAO and 24 hours of reperfusion, mice were deeply anesthetized with 5% isoflurane and transcardially perfused with ice-cold saline. Brains were removed and sectioned coronally at a thickness of 1 mm using a brain matrix (Harvard Apparatus, Holliston, MA, USA) and incubated in 2% TTC for 15 minutes at room temperature. Brain slices were then fixed in 4% paraformaldehyde, scanned (Hewlett Packard, Palo Alto, CA, USA) into a computer, and quantified using the NIH imaging software (rsb.info.nih.gov/nih-image).

Anatomy of the MCA and Circle of Willis

Naïve mice (n=4 in each type of mouse) were deeply anesthetized and transcardially perfused with 2% India ink in 20% gelatin in saline. Mice were decapitated after 30 seconds, and brains were removed and fixed with 4% paraformaldehyde. Brain images were captured using a Leica dissecting scope (Leica Microsystems, Wetzlar, Germany).

PCR Array

At 24 hours after reperfusion, mice were decapitated after being deeply anesthetized with 5% isoflurane. Brains were extracted within 30 seconds and frozen in liquid nitrogen and stored at −80°C for later dissection. A peripheral area of the ipsilateral cortex (equivalent to the penumbra area in this model) was dissected in a −20°C glove box. RNA was isolated using Mini RNeasy Columns (Qiagen, Rockville, MD, USA) and stored at −80°C. cDNA was synthesized using Superscript III and oligo-dT (Invitrogen, Carlsbad, CA, USA). Ninety-six-well PCR plates pre-spotted with oligonucleotides for cellular stress genes (SuperArray, Frederick, MD, USA) were used with the SYBR Green RT-PCR master mix (Invitrogen) and run on the ABI 9600 (Applied Biosystems, Foster City, CA, USA). Crossing threshold (CT) values were manually set at 2.0 and were normalized to the housekeeping genes β-actin and HPRT1.

Spotted Cytokine Array

A separate set of animals was used for spotted cytokine array study. At 24 hours after reperfusion, mice were deeply decapitated under deep anesthesia with 5% isoflurane. Brains were excised, frozen in liquid nitrogen, and stored at −80°C for later dissection. A peripheral area of the ipsilateral cortex was dissected in a −20°C glove box. The brains were homogenized using a tissue homogenizer (Cole-Palmer, Vernon Hills, IL, USA) at 14,000 r.p.m. in Sigma Protein Isolation Buffer containing 1 mmol/L EDTA, 5 mmol/L DTT (Sigma-Aldrich, St Louis, MO, USA), and protease inhibitors (Thermo Scientific, Rockford, IL, USA). The homogenates were centrifuged at 750 g for 15 minutes to separate the nuclear fraction from the cytosolic and mitochondria fraction. The supernatant containing the cytosolic and mitochondrial proteins was used for cytokine quantification. Protein was quantified using the A280 protein quantification program on the NanoDrop 2000 (Thermo Scientific). Protein was incubated on a cytokine array membrane (Ray Biotech, Norcross, CA, USA); signal was detected by chemiluminesence (Thermo Scientific) onto a film (Kodak, Rochester, NY, USA), scanned (Hewitt Packard), and quantitated using Quantity One (Bio-Rad, Hercules, CA, USA).

Immunohistochemistry

At 5 and 24 hours after reperfusion, mice were transcardially perfused under anesthesia with ice-cold phosphate-buffered saline for 30 seconds and then with 4% paraformaldehyde. Brains were removed and fixed overnight in 4% paraformaldehyde at 4°C. The brains were sectioned coronally at 30-μmol/L thickness in ice-cold phosphate-buffered saline using a vibrating microtome (Leica Microsystems). The sections were placed in an anti-freeze solution and stored at −20°C for later use. The sections were washed, the nonspecific binding sites on sections were blocked with 3% bovine serum albumin, and the sections were incubated overnight with a primary antibody against fractalkine (mAb IgG; 1 μg/mL; Invitrogen) and nuclear factor-κB (NF-κB) (1:100; Santa Cruz Biotechnology, Santa Cruz, CA, USA). The sections were incubated with an AlexaFluor-594-conjugated secondary antibody (1:200; Invitrogen), mounted with Vectashield mounting medium containing DAPI (Vector Laboratories, Burlingame, CA, USA), and scanned under a Nikon Eclipse Ti-u laser-scanning confocal microscope (Nikon, Tokyo, Japan) at × 400 magnification. Three fields per section were captured and analyzed.

Statistical Analyses

All data are presented as means ± s.d. Student's t-test was used to analyze the difference between infarct volumes that were calculated by multiplying the infarct areas measured on six TTC-stained brain sections by the brain section's thickness. Student's t-test was also used to analyze the data collected from PCR array and spotted cytokine array from the two types of animals mentioned above. P-value <0.05 was considered as significant.

Results

Infarct Volume

A 60-minute MCAO induced mild brain damage located predominantly in the caudoputamen at 24 hours of recovery in wild-type mice. Mice lacking UCP2 had a significant increase in infarction after tFI (Figure 1). The brain damage extended to almost the entire cortex in the ipsilateral hemisphere of the UCP^−/− mice. As a result, the percent infarct volume increased significantly in knockout mice as compared with that in wild-type mice (61.6 ± 32.8 versus 17.8 ± 4.6, P<0.0001). Neurologic behavior was only examined after MCAO to validate the success of occlusion. No difference was observed between the wild-type and knockout mice.

Figure 1

Infarction after 1-hour middle cerebral arterial occlusion (MCAO) and 24-hour reperfusion was determined by staining 1-mm sections with triphenyltetrazolium chloride (TTC) (A). UCP2^−/− mice had a significant increase in infarction area after 1-hour MCAO and 24-hour reperfusion (B).

Cerebral Vasculature

To evaluate whether globally knocking out UCP2 caused a phenotypic change in the cerebral vasculature, we transcardially injected carbon black ink and imaged the cerebral blood vessels (Figure 2). Both wild-type and UCP2^−/− mice showed intact and correct alignment of the Circle of Willis, anterior cerebral arteries, MCAs, and posterior arteries with no remarkable difference.

Figure 2

Mice were perfused with carbon black to determine whether there were vascular abnormalities in the UCP2^−/−mice. The Circle of Willis, anterior cerebral arteries, middle cerebral arteries (MCAs), and posterior arteries all appear normal as compared with those in wild-type controls.

Alteration of Gene Expression

To determine the mRNA expression profile in the ischemic penumbra area of wild-type and UCP2-knockout mice after tFI, a Mouse Stress Toxicity PCR array (Super Array) was used to measure the transcript levels of 84 genes by quantitative PCR (ABI 7300). Housekeeping genes HPRT1 and β-actin were used to normalize the results. The results showed that ischemia and reperfusion injury induced 34.8-, 63.1-, and 14.7-fold increase in the level of the cell-cycle regulation genes cyclin-C, cyclin-D1, and cyclin-G1, respectively. It is worth mentioning that cyclin-D1 and cyclin-G1 participate in neuroregeneration and DNA repair. The mRNA levels of murine double minute-2 (Mdm2), which mediates p53 degradation, Cu/Zn-superoxide dismutase (Sod1), and Mn-superoxide dismutase (Sod2), were also elevated and the antioxidant gene glutathione-S-transferase murine type-1 (Gstm1) was suppressed. Deletion of UCP2 gene led to pronounced suppression of cyclins, Gstm1, Mdm2, Sod1, and Sod2 (Table 2). Please visit http://www.nature.com/jcbfm for Supplementary PCR array data accompanies this publication.

Table 2

Real-time PCR array data

Gene	Full name	Function	(A) UCP2^−/−/WT	(B) WT/sham
Ccnc	Cyclin-C	Cell-cycle regulation	−88.1	34.8
Ccnd1	Cyclin-D1	Neuronal regeneration	−9.8	63.1
Ccng1	Cyclin-G1	DNA repair	−238.6	14.7
Gstm1	Glutathione-S-transferase murine type-1	Antioxidant	−151.8	−4.5
Mdm2	Murine double minute-2	Mediate p53 degradation	−25.7	2.25
Sod1	Cu/Zn superoxide dismutase-1	Antioxidant enzyme	−14.7	4.1
Sod2	Mn superoxide dismutase-2	Antioxidant enzyme	−7.3	5.3

PCR, polymerase chain reaction; UCP, uncoupling protein; WT, wild type.

Increased Inflammatory Cytokines

A protein-spotted array was used to determine the change in 24 cytokine protein levels in the ischemic penumbra of the wild-type and UCP2-knockout mice after tFI. There were no detectable levels of the 24 detected cytokines in both UCP^−/− and wild-type mice without ischemic injury. After ischemia and reperfusion, 10 cytokines showed more than twofold increase after MCAO in wild-type mice in the cortical penumbra area (Figure 3). These include B-lymphocyte chemoattractant (BLC) (also known as B-cell-attracting chemokines-1 or CXCL13), cytokine-responsive gene-2 (CRG-2 or CXCL10, in human IP-10), cutaneous T-cell-attracting chemokine (CTACK/CCL27), chemokine (C–X–C motif) ligand-16 (CXCL16), Eotaxin-2 (CCL24), fractalkine (CXCL1), insulin-like growth factor-binding protein-6 (IGFBP-6), macrophage inflammatory protein-1 (MCP-1, CCL2), and macrophage inflammatory protein-5 (MCP-5, CCL12). Ischemia in UCP2^−/− mice led to more pronounced elevation in the levels of eight cytokines (BLC, CRG-2, CTACK, CXCL16, Eotaxin-2, fractalkine, and IGFBP-6) and less pronounced increase in MCP-1 and MCP-5 as compared with that in wild-type ischemic mice.

Figure 3

Protein detection of inflammatory and apoptotic proteins was measured using protein-spotted arrays. Selected cytokines from a protein-spotted array showed significant difference in wild-type versus UCP2^−/− after middle cerebral arterial occlusion (MCAO).

Immunohistochemistry of Fractalkine and NF-κB

For fractalkine/CXCL1 (Figure 4A), virtually no immunoreactivity was observed in the control brain sections of the wild-type and knockout mice. There was no noticeable change observed at 5 hours of recirculation in both groups (data not shown). Fractalkine immunoreactivity enhanced markedly at 24 hours of recovery in the UCP2-knockout animals, whereas only a mild enhancement was observed in the wild-type animals (Figure 4A). Predictably, the fractalkine immunostaining localized to the cytoplasm.

Figure 4

Immunohistochemistry of the ipsilateral cortex. Fractalkine staining of the sham and ischemic penumbra of wild-type and UCP2^−/− mice (A). Nuclear factor (NF)-κB staining of the sham and ischemic penumbra of wild-type and UCP2^−/− mice (B).

For NF-κB (Figure 4B), a faint NF-κB background immunostaining was observed in the cytosol of both wild-type and UCP2^−/− control animals. At 5 hours of recovery, immunoreactivity increased slightly but the staining remained in the cytosol, with no marked difference between that in the wild-type and knockout animals (data not shown). At 24 hours of recirculation, NF-κB immunoreactivity increased in both types of animals and, more importantly, NF-κB translocation to the nucleolus occurred in the knockout animals. Thus, while the majority of the cells that stained positively for NF-κB had cytosolic localization in the wild-type brain sections, the majority of the NF-κB-positive neurons had nuclear localization (Figure 4B).

Discussion

Our data showed that knocking out UCP2 significantly increased the infarct volume after ischemia–reperfusion injury. This damage was not due to vasculature abnormalities, as no major differences in the major cerebral vasculature were found between wild-type and UCP2-knockout mice. We acknowledge that the carbon black perfusion could only provide a gross morphologic assessment of the vasculature. Laser Doppler detection would provide actual tissue perfusion during MCAO. Brain damage may progress for several days before reaching its maturity if the duration of ischemia is brief. In our study, however, nearly 100% of the MCA territory had developed into infarction in the UCP2^−/− animals after 24 hours of reperfusion, suggesting that brain damage had been matured at 24 hours of recovery in UCP2^−/− animals.

Our data complements the work of Wieloch's group, which showed that enhanced expression of UCP2 attenuates damage from cerebral ischemia–reperfusion injury (Deierborg et al, 2008; Mattiasson et al, 2003). However, our findings differ from those of the previous study published by de Bilbao et al (2004) showing that infarct volume was significantly reduced after MCAO in UCP^−/− mice. There are two major differences between the above work and our study. First, the surgical technique in the previous study occluded the distal MCA. As a result, it only induced infarction in the cortical area. In our study, the MCA was occluded from the proximal part, yielding an infarct in both the striatum and the overlaying cortex. Second and more importantly, a permanent MCAO model was used in the previous study, whereas a transient MCAO with blood recirculation was used in our study. By contrast with transient MCAO, ROS production after permanent MCAO does not seem to be a major contributor in ischemic damage progressions. It has been reported that reperfusion produced a burst in ROS formation after transient MCAO, whereas increase in ROS production was much less prominent after permanent MCAO (Peters et al, 1998). Studies on Sod1^−/+ animals lend support to the concept that ROS production is significantly different between transient and permanent MCAO. Thus, although overexpression of Sod1 failed to protect neurons after a permanent MCAO (Chan et al, 1993), it showed neuroprotection after transient MCAO (Saito et al, 2003). Therefore, the negative effect of UCP2 observed by de Bilbao et al is probably ascribed to the permanent MCAO model being used.

The mechanisms responsible for downregulation of cell-cycle and antioxidant genes, and upregulation of inflammatory cytokines, in UCP^−/− mice are not clear. We hypothesize that deletion of the UCP2 gene causes increased ROS production, which directly activates NF-κB, a master transcription factor, to control a variety of inducible genes that regulate cell cycle, anti-oxidation, and inflammatory processes (Lentsch and Ward, 1999; Ridder and Schwaninger, 2009). Our results support this hypothesis. Thus, NF-κB immunoreactivity increased and there was an evident nuclear translocation of NF-κB in UCP2^−/− mice after tFI. Previous studies have shown that knockout of UCP2 persistently increases NF-κB activation in the spleen and increases translocation to the nucleus after endotoxin stimulation as compared with wild-type controls (Bai et al, 2005). Similarly, levels of inflammatory cytokines such as nitric oxide synthase, interferon-γ, tumor necrosis factor-α, inteleukin-1β, and inteleukin-6 were all increased in the UCP2-knockout mice (Bai et al, 2005). It is likely that NF-κB suppresses the cell-cycle and antioxidant genes, and activates the inflammatory mediators responsible for the increased ischemic damage in the UCP2-knockout mice.

Our study showed that knocking out UCP2 suppressed antioxidant genes after ischemia and reperfusion injury. As we have stated in the introduction, upregulation of UCPs is capable of reducing mitochondrial ROS formation by preventing hyperpolarization or stabilizing the mitochondrial membrane potential. Conversely, deletion of UCP2 is expected to increase ROS production. This has been proven in reports that have shown that endogenous superoxide and hydrogen peroxide increased in UCP2-knockout mice at baseline levels and after endotoxin challenge (Arsenijevic et al, 2000; Bai et al, 2005; Pi et al, 2009). Our study showed that ischemia induced the expression of antioxidant Sod1 and Sod2 genes and cell-cycle regulation genes in wild-type mice. These changes may reflect a stress response and imply enhanced ROS production after ischemia and reperfusion. Interestingly, ischemia and reperfusion in UCP2-knockout mice caused marked suppression of cell-cycle genes, antioxidant Gstm1, Sod1 and Sod2 genes, and Mdm2 gene that mediates the degradation of p53. Suppression of cell-cycle genes results in inhibition of cell differentiation and repair. For example, the level of cyclin-G1, which is activated by the PI3K/Akt pathway, increased and translocated to the nucleolus after various brain injuries, including knife cut, cold injury, kinate injection, in vitro NMDA exposure, hypoxia, and transient and permanent focal cerebral ischemia (Maeda et al, 2005; van Lookeren Campagne and Gill, 1998a). Cyclin-G1 has been proposed as a DNA repair gene because it is increased in post-hypoxic neuronal stem cells that undergo proliferation and DNA repair (Chen et al, 2009; van Lookeren Campagne and Gill, 1998b). Therefore, suppression of cyclin-G1 will likely lead to suppression of DNA repair after ischemia.

Suppression of antioxidant genes reduces ROS degradation and increases ROS accumulation. Gstm1 belongs to the Mu family of human glutathione-S-transferase. In addition to its location in the cytosol and membrane, Gstm1 is also located in the mitochondria and is a defense against oxidative stress (Raza et al, 2002). Sod1 and Sod2 are well known for their roles in protecting cells against oxidative stress by catalyzing the dismutation of superoxide into oxygen and hydrogen peroxide. Suppression of these antioxidant genes will impede the cell defense system and result in accumulation of ROS and, eventually, cell death. Mdm2 mediates the degradation of p53. Inhibition of Mdm2 increases the level of p53 and subsequently activates the p53-upregulated modulator of apoptosis (PUMA) and triggers apoptotic cell death (Jeffers et al, 2003).

ROS production has been shown to instigate downstream inflammatory responses in the brain after stroke (Lakhan et al, 2009). To study whether suppression of antioxidant genes in UCP2-knockout mice is linked to an increase in inflammatory cytokines, we performed the spotted cytokine array. The results showed that 60 minutes of ischemia induced a marked increase in BLC/CXCL13, CRG-2/CXCL10, CTACK/CCL27, CXCL16, Eotaxin-2/CCL24, fractalkine/CXCL1, and IGFBP-6 in UCP2-knockout as compared with wild-type mice. Among these cytokines, BLC/CXCL13 and CRG-2/CXCL10 have been linked to several inflammatory disorders in the central nervous system (Festa et al, 2009; Lane et al, 1998). CTACK/CCL27 recruits lymphocytes. Our results suggest a possible existence of a connection between CTACK and neuroinflammation. Eotaxins are potent eosinophil chemotractants. Our results, which showed increased levels of BLC/CXCL13, CRG-2/CXCL10, CTACK/CCL27, and Eotaxin-2/CCL24, imply that these cytokines may have a role in mediating the neuroinflammation set off by ischemia-reperfusion injury and that UCP2 may directly or indirectly regulate the inflammatory responses.

Both CXCL16 and fractalkine/CXCL1 are activated by the proteases and oxidative stress associated with cerebral ischemia. The level of fractalkine/CXCL1 is transiently increased after ischemia and reperfusion (Tarozzo et al, 2002). Knocking out fractalkine receptor (Denes et al, 2008) reduced the infarction by 56% as compared with that in wild-type controls after tFI. The increase in CXCL16 and fractalkine/CXCL1 in the UCP2-knockout mice after tFI would likely enhance the ischemic brain damage.

IGFBP-6 inhibits cell proliferation and increases apoptosis (Grellier et al, 1998; Seurin et al, 2002). Transgenic IGFBP-6 mice showed dysregulation of energy homeostasis, retarded growth, and downregulation of UCP1 in brown adipose tissue (Bienvenu et al, 2004). Therefore, the increase in IGFBP-6 in the UCP2-knockout mice after ischemia may be associated with increased damage observed in these mice.

Both MCP-1 and MCP-5 recruit leukocytes and mediate inflammatory responses. Knocking out MCP-1 decreased and overexpression of MCP-1 increased ischemic brain damage after permanent ischemia (Chen et al, 2003; Hughes et al, 2002). In this study we observed decreased levels of MCP-1 and MCP-5 in UCP2-deficient mice. This implies either that the increased damage in UCP2-knockout mice is not mediated by MCPs or that the decreased levels of the MCPs actually reduce the influx of neuroblasts to repair the damaged brain tissue.

In conclusion, our results showed that knocking out UCP2 increased brain damage induced by transient MCAO. Knocking out UCP2 suppressed genes related to DNA repair, antioxidation, and neuroprotection, and increased the protein levels of various inflammatory mediators. Our findings support the notion that UCP2 is an innate regulator of inflammation. With concluding verification that UCP2 is neuroprotective, future efforts to pharmaceutically target UCP2 will be the next logical step in this research. Ideal drug candidates will be multifunctional and will interrupt more than one specific step of the ischemic injury cascade. Compounds found to enhance the expression of UCP2 will likely suppress ROS generation and the inflammation that follows cerebral ischemia–reperfusion injury.

Footnotes

Acknowledgements

We thank Dr Mariana Gerschenson for helping with paper preparation.

The authors declare no conflict of interest.

Notes

Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website ()

References

Arsenijevic

Onuma

Pecqueur

Raimbault

Manning

Miroux

Couplan

Alves-Guerra

Goubern

Surwit

Bouillaud

Richard

Collins

Ricquier

(2000) Disruption of the uncoupling protein-2 gene in mice reveals a role in immunity and reactive oxygen species production. Nat Genet 26:435–9

Bai

Onuma

Bai

Medvedev

Misukonis

Weinberg

Cao

Robidoux

Floering

Daniel

Collins

(2005) Persistent nuclear factor-kappa B activation in Ucp2–/– mice leads to enhanced nitric oxide and inflammatory cytokine production. J Biol Chem 280:19062–9

Bechmann

Diano

Warden

Bartfai

Nitsch

Horvath

(2002) Brain mitochondrial uncoupling protein 2 (UCP2): a protective stress signal in neuronal injury. Biochem Pharmacol 64:363–7

Beech

Reckless

Mosedale

Grainger

Williams

Menon

(2001) Neuroprotection in ischemia–reperfusion injury: an antiinflammatory approach using a novel broad-spectrum chemokine inhibitor. J Cereb Blood Flow Metab 21:683–9

Bienvenu

Seurin

Grellier

Froment

Baudrimont

Monget

Le Bouc

Babajko

(2004) Insulin-like growth factor binding protein-6 transgenic mice: postnatal growth, brain development, and reproduction abnormalities. Endocrinology 145:2412–20

Chan

(1996) Role of oxidants in ischemic brain damage. Stroke 27:1124–9

Chan

Kamii

Yang

Gafni

Epstein

Carlson

Reola

(1993) Brain infarction is not reduced in SOD-1 transgenic mice after a permanent focal cerebral ischemia. Neuroreport 5:293–6

Chen

Tian

Yao

Zhang

Liu

(2009) Hypoxia stimulates proliferation of rat neural stem cells with influence on the expression of cyclin D1 and c-Jun N-terminal protein kinase signaling pathway in vitro. Neuroscience 165:705–14

Chen

Hallenbeck

Ruetzler

Bol

Thomas

Berman

Vogel

(2003) Overexpression of monocyte chemoattractant protein 1 in the brain exacerbates ischemic brain injury and is associated with recruitment of inflammatory cells. J Cereb Blood Flow Metab 23:748–55

10.

Choi

Kim

Choi

(2009) Oxidative stress-induced necrotic cell death via mitochondira-dependent burst of reactive oxygen species. Curr Neurovasc Res 6:213–22

11.

de Bilbao

Arsenijevic

Vallet

Hjelle

Ottersen

Bouras

Raffin

Abou

Langhans

Collins

Plamondon

Alves-Guerra

Haguenauer

Garcia

Richard

Ricquier

Giannakopoulos

(2004) Resistance to cerebral ischemic injury in UCP2 knockout mice: evidence for a role of UCP2 as a regulator of mitochondrial glutathione levels. J Neurochem 89:1283–92

12.

Deierborg

Wieloch

Diano

Warden

Horvath

Mattiasson

(2008) Overexpression of UCP2 protects thalamic neurons following global ischemia in the mouse. J Cereb Blood Flow Metab 28:1186–95

13.

Della-Morte

Dave

DeFazio

Bao

Raval

Perez-Pinzon

(2009) Resveratrol pretreatment protects rat brain from cerebral ischemic damage via a sirtuin 1-uncoupling protein 2 pathway. Neuroscience 159:993–1002

14.

Denes

Ferenczi

Halasz

Kornyei

Kovacs

(2008) Role of CX3CR1 (fractalkine receptor) in brain damage and inflammation induced by focal cerebral ischemia in mouse. J Cereb Blood Flow Metab 28:1707–21

15.

Festa

Hankiewicz

Kim

Skurnick

Wolansky

Cook

Cadavid

(2009) Serum levels of CXCL13 are elevated in active multiple sclerosis. Mult Scler 15:1271–9

16.

Grellier

De Galle

Babajko

(1998) Expression of insulin-like growth factor-binding protein 6 complementary DNA alters neuroblastoma cell growth. Cancer Res 58:1670–6

17.

Hughes

Allegrini

Rudin

Perry

Mir

Wiessner

(2002) Monocyte chemoattractant protein-1 deficiency is protective in a murine stroke model. J Cereb Blood Flow Metab 22:308–17

18.

Iijima

Mishima

Akagawa

Iwao

(2003) Mitochondrial hyperpolarization after transient oxygen–glucose deprivation and subsequent apoptosis in cultured rat hippocampal neurons. Brain Res 993:140–5

19.

Jeffers

Parganas

Lee

Yang

Wang

Brennan

MacLean

Han

Chittenden

Ihle

McKinnon

Cleveland

Zambetti

(2003) Puma is an essential mediator of p53-dependent and -independent apoptotic pathways. Cancer Cell 4:321–8

20.

Jezek

(2002) Possible physiological roles of mitochondrial uncoupling proteins—UCPn. Int J Biochem Cell Biol 34:1190–206

21.

Kim

Park

Koo

Huh

Lee

(2007) The modulating effects of the overexpression of uncoupling protein 2 on the formation of reactive oxygen species in vascular cells. Diabetes Res Clin Pract 77(Suppl 1):S46–8

22.

Lakhan

Kirchgessner

Hofer

(2009) Inflammatory mechanisms in ischemic stroke: therapeutic approaches. J Transl Med 7:97

23.

Lane

Asensio

Paoletti

Campbell

Buchmeier

(1998) Dynamic regulation of alpha- and beta-chemokine expression in the central nervous system during mouse hepatitis virus-induced demyelinating disease. J Immunol 160:970–8

24.

Lentsch

Ward

(1999) Activation and regulation of NFkappaB during acute inflammation. Clin Chem Lab Med 37:205–8

25.

Gisselsson

Keuker

Vogel

Smith

Kuschinsky

Siesjo

(1998) Hyperglycemia-exaggerated ischemic brain damage following 30 mins of middle cerebral artery occlusion is not due to capillary obstruction. Brain Res 804:36–44

26.

Maeda

Ampo

Kiryu-Seo

Konishi

Ohba

Kadono

Kiyama

(2005) The p53-independent nuclear translocation of cyclin G1 in degenerating neurons by ischemic and traumatic insults. Exp Neurol 193:350–60

27.

Mattiasson

Shamloo

Gido

Mathi

Tomasevic

Warden

Castilho

Melcher

Gonzalez-Zulueta

Nikolich

Wieloch

(2003) Uncoupling protein-2 prevents neuronal death and diminishes brain dysfunction after stroke and brain trauma. Nat Med 9:1062–8

28.

Mehta

(2009) Neuroprotective role of mitochondrial uncoupling protein 2 in cerebral stroke. J Cereb Blood Flow Metab 29:1069–78

29.

Minami

Katayama

Satoh

(2006) Brain cytokines and chemokines: roles in ischemic injury and pain. J Pharmacol Sci 100:461–70

30.

Nakase

Yoshida

Nagata

(2007) Amplified expression of uncoupling proteins in human brain ischemic lesions. Neuropathology 27:442–7

31.

Offner

Subramanian

Parker

Afentoulis

Vandenbark

Hurn

(2006) Experimental stroke induces massive, rapid activation of the peripheral immune system. J Cereb Blood Flow Metab 26:654–65

32.

Ouyang

Sun

Giffard

(2006) Overexpression of inducible heat shock protein 70 and its mutants in astrocytes is associated with maintenance of mitochondrial physiology during glucose deprivation stress. Cell Stress Chaperones 11:180–6

33.

Peters

Back

Lindauer

Busch

Megow

Dreier

Dirnagl

(1998) Increased formation of reactive oxygen species after permanent and reversible middle cerebral artery occlusion in the rat. J Cereb Blood Flow Metab 18:196–205

34.

Bai

Daniel

Liu

Lyght

Edelstein

Brownlee

Corkey

Collins

(2009) Persistent oxidative stress due to absence of uncoupling protein 2 associated with impaired pancreatic beta-cell function. Endocrinology 150:3040–8

35.

Raza

Robin

Fang

Avadhani

(2002) Multiple isoforms of mitochondrial glutathione S-transferases and their differential induction under oxidative stress. Biochem J 366:45–55

36.

Ridder

Schwaninger

(2009) NF-kappaB signaling in cerebral ischemia. Neuroscience 158:995–1006

37.

Saito

Hayashi

Okuno

Ferrand-Drake

Chan

(2003) Overexpression of copper/zinc superoxide dismutase in transgenic mice protects against neuronal cell death after transient focal ischemia by blocking activation of the Bad cell death signaling pathway. J Neurosci 23:1710–8

38.

Seurin

Lassarre

Bienvenu

Babajko

(2002) Insulin-like growth factor binding protein-6 inhibits neuroblastoma cell proliferation and tumour development. Eur J Cancer 38:2058–65

39.

Tarozzo

Campanella

Ghiani

Bulfone

Beltramo

(2002) Expression of fractalkine and its receptor, CX3CR1, in response to ischaemia—reperfusion brain injury in the rat. Eur J Neurosci 15:1663–8

40.

Teshima

Akao

Jones

Marban

(2003) Uncoupling protein-2 overexpression inhibits mitochondrial death pathway in cardiomyocytes. Circ Res 93:192–200

41.

van Lookeren Campagne

Gill

(1998a) Increased expression of cyclin G1 and p21WAF1/CIP1 in neurons following transient forebrain ischemia: comparison with early DNA damage. J Neurosci Res 53:279–96

42.

van Lookeren Campagne

Gill

(1998b) Cell cycle-related gene expression in the adult rat brain: selective induction of cyclin G1 and p21WAF1/CIP1 in neurons following focal cerebral ischemia. Neuroscience 84:1097–112