Article history:Accepted 30 November 2010Available online 4 December 2010
IL-10 reduces pro-inflammatory responses after ischemic stroke primarily by acting on gliaand endothelium, but relatively little is known about the direct effects of IL-10 on corticalneurons, which are often damaged in stroke. We found by PCR and immunohistochemistrythat cortical neurons express IL-10 receptor. Treatment of primary cortical neurons inculture with IL-10 increased neuronal survival after exposure to oxygen-glucose deprivation(OGD) or glutamate toxicity. IL-10 also induced phosphorylation of AKT in cortical neurons.Pretreatment with the specific PI-3K inhibitor, wortmannin, attenuated IL-10 mediatedneuroprotection against OGD and glutamate. In addition, IL-10 induced STAT-3phosphorylation. Pre-treatment with a functional blocking antibody to the IL-10 receptorreduced both Stat-3 and AKT phosphorylation and blocked IL-10 mediated protection ofcortical neurons. These data suggest that IL-10 provides neuroprotection by acting via IL-10receptor and PI3K/AKT and STAT-3 signal transduction pathways.
IL-10 is a cytokine that suppresses various aspects of theimmune response including the synthesis of TNFα, IL-1β, andIL-8. The administration of exogenous IL-10 protects tissueexposed to a variety of ischemia-reperfusion regimens andreduces lesion size in focal ischemic stroke (Engles et al., 1997;Spera et al., 1998). Knockout mice for IL-10 have larger infarctsafter stroke comparedwith wild-type controls (Grilli et al., 2000)and transgenic mice overexpressing IL-10 have attenuatedinfarct maturation in a permanent model of focal cerebralischemia (de Bilbao et al., 2009). Collectively, these studiessuggest that IL-10 is neuroprotective through indirect effects onpro-inflammatory pathways. However, other studies alsosuggest that IL-10 may directly protect neurons. For example,it has recently been shown that IL-10 protects spinal neurons
logy, University of Texas M
u (S.I. Savitz).
er B.V. All rights reserved
from glutamate excitotoxicity (Zhou et al., 2009). We therefore,conducted the present study to assess whether IL-10 directlyprotects culturedprimary cortical neurons fromoxygen-glucosedeprivation and glutamate toxicity, two injury models that arerelevant to ischemic stroke.
2. Results
2.1. IL-10 receptor is expressed by neurons
Using specific primers for IL-10R and RT-PCR, we demonstratethat IL-10 receptor (IL-10R)mRNA ispresent in corticalprimary ratcultured neurons, as presented in Fig. 1A. Using scrambledprimers for IL-10Rprimers,weno longer detected anmRNAbandfor IL-10R. Under themicroscope, we identified neurons by their
edical School at Houston, Houston, TX 77030, USA. Fax: +1 713 500
Fig. 1 – (A) RT-PCR for IL-10 receptor (IL-10R) and β-actin mRNA expression in primary rat cultured neurons. Lanes 1: 1 kb DNAladder, 2: upper band: IL-10R amplificationproduct and lower band:β-actin amplification product, 3: the absence of IL-10Rproductusing scrambled primers and lower band:β-actin amplification product. (B) Microscopic picture of immunofluorescence of IL-10R(left panel); FITC-conjugated anti-IL-10 receptor antibody (1:500), NeuN (middle picture), PE-conjugated anti-NeuN antibody(Dilution: 1:100), and co-localization of IL-10R and MAP-2 (right panel) in primary rat neurons in culture.
190 B R A I N R E S E A R C H 1 3 7 3 ( 2 0 1 1 ) 1 8 9 – 1 9 4
characteristic morphology and by NeuN immunostaining whilethe presence of IL-10R1 on neurons was determined by co-localization of immunoreactivity for IL-10R1 and NeuN (Fig. 1B).
2.2. IL-10 activates the Jak/Stat3 and PI3K/AKT pathwayin cortical neurons
To further study whether IL-10 has direct effects on culturedcortical neurons, we studied its signaling responses. Byperforming cell-basedmicro-titer phospho-ELISA on neuronalextracts, we established that recombinant IL-10 in a concen-tration-dependent fashion increased phosphorylation of Stat3at Tyr705 and AKT at S473 (Figs. 2A and B respectively). Theseresponses demonstrate that in cortical neurons, IL-10 activatesboth the Jak/STAT-3 and PI3K/Akt pathways. To furtherauthenticate these observations, we assessed STAT-3 and PI3Kphosphorylation in response to IL-10 treatment in the presenceor absence of IL-10 receptor blocking antibody. Indeed, IL-10receptor blocking antibody reduced PI3-mediated AKTphosphorylation and STAT-3 phosphorylation in response toIL-10 (Figs. 3A and B). Finally, to substantiate a role for PI3K inAkt phosphorylation, we used the specific PI3K inhibitor,wortmannin, and found that it inhibited IL-10-mediated Aktphosphorylation in primary neuronal cultures (Fig. 3C).
2.3. IL-10 reduces OGD and glutamate-induced injury toprimary cortical neurons
To demonstrate the direct neuroprotective effect of IL-10, wesubjected cortical neurons in culture to OGD (90 min) orglutamate (100 μM) in the presence or absence of IL-10. In aconcentration-dependent manner, IL-10 attenuated neuronal
death inducedbyOGDorglutamate (Figs. 4AandB). Inagreementwith studies showing a link between Akt phosphorylation andneuronal survival, wortmannin under conditions that reduceAkt-phosphorylation, reduced the neuroprotective effect of IL-10(Fig. 4C). In addition, pre-treatment with IL-10 receptor blockingantibody completely abolished the protective effect of IL-10(Fig. 4D). However, pre-treatment with heat denatured IL-10receptor blocking antibody had no effect on the protectionafforded by IL-10.
3. Discussion
In the present study, we show that IL-10 directly protectscultured rat cortical neurons exposed to conditions modelingischemic stroke. Our studies indicate that cortical neuronsexpress IL-10 receptor and IL-10 activates canonical pro-survival pathways including Akt and STAT-3. In response toglutamate excitotoxicity or OGD, IL-10 directly protectscultured neurons in a dose-dependent manner.
IL-10 is a pleiotropic cytokine mainly produced by microgliaand has a range of known effects on the inflammatory cascadeafter various brain injuries. Several studies, for example, haveshown that IL-10may play an important neuroprotective role inin vitro and in vivo models of ischemic stroke. Mixed corticalglial–neuronal co-cultures derived from IL-10 KOmice aremoresensitive to NMDA, glutamate, or OGD-mediated injury (Grilliet al., 2000) and IL-10 KO mice have increased infarct size afterfocal cerebral ischemia comparedwithwild-type controls (Grilliet al., 2000). Also, a direct injection of human IL-10, whenadministered intracerebroventricularly or systemically, reducesinfarct volume in a severe focal ischemia model using
Fig. 3 – (A) Inhibitionof IL-10-mediatedAkt phosphorylation inprimary rat neuronal cultures by IL-10 receptor blockingantibody (IL-10RB: Dilution: 1:500) determined at 30min afteradding IL-10, IL-10RB, or both. N=8 per group. (B) Inhibition ofIL-10-mediated STAT-3 phosphorylation in primary ratneuronal cultures by IL-10 receptor blocking antibody (IL-10RB:Dilution: 1:500) within 30min of incubation. N=8 per group.(C) Inhibition of IL-10-mediated Akt phosphorylation inprimary rat neuronal cultures by PI3 kinase inhibitor,wortmannin (30 nM) within 30min of incubation. The controlgroup received IL-10 vehicle. IL-10 is 20 ng/ml. *p<0.05compared with control. N=8 per group.
Fig. 2 – (A) Concentration-dependent increase in Aktphosphorylation inprimary ratneuronal cultureswithin30minof incubation. N=8 per group. (B) Concentration-dependentincrease in STAT-3 phosphorylation in primary rat neuronalcultures within 30min of incubation. The control groupreceived IL-10 vehicle. N=8 per group. *p < 0.05 comparedwith control; **p < 0.05 comparedwith the groupof neurons thatwere treated with 10 ng/ml of IL-10. ***p < 0.05 comparedwith the group of neurons that were treated with 30 ng/ml ofIL-10.
191B R A I N R E S E A R C H 1 3 7 3 ( 2 0 1 1 ) 1 8 9 – 1 9 4
spontaneously hypertensive rats subjected to permanent MCAocclusion (Spera et al., 1998).
The majority of the existing studies on IL-10 suggest thatthis cytokine is neuroprotective through its inhibitory effectson pro-inflammatory cells. However, in addition, neuronsmayalso be directly responsive to IL-10. The expression of IL-10receptors has been found on embryonic spinal neurons, whichare directly protected by IL-10 against glutamate toxicity(Zhou et al., 2009). Other studies indicate that IL-10 protectscerebellar granule cells from NMDA toxicity (Grilli et al., 2000)and IL-10 protects retinal ganglion cells in the setting of serumdeprivation (Boyd et al., 2003). In the present study, we nowadd evidence for the direct effects of IL-10 on cortical neurons.We found mRNA expression of IL-10R in cultured corticalneurons and, via co-localization of IL-10R and NeuN immuno-reactivity,wedemonstrated that IL-10 receptorsare localizedonthe neuronal membrane. Furthermore, we used cell-based
phospho-ELISA to establish the mechanisms by which theIL-10 receptor signals in neurons. In agreement with others(Zhouetal., 2009),we find that IL-10 receptor activatescanonicalpro-survival pathways such as AKT and STAT-3. Finally, ourdata indicate that IL-10 dose-dependently protects neuronsfrom OGD or glutamate neurotoxicity. This protection isinhibited by the IL-10 receptor blocking antibody, pointingtoward specific binding of IL-10 with its receptor on neurons.
Fig. 4 – IL-10 reduces cell death of primary cultured neurons exposed to OGD (A) or glutamate (100 μM) (B). IL-10was added rightbefore the onset of injury. MTT survival assay was completed at 24 h after treatment. The 0 ng/ml group received vehicle forIL-10. *p<0.05 compared with 0 ng/ml. **p<0.05 compared with 1 ng/ml. N=8 per group. (C) Protection of cultured neuronsagainst OGD by IL-10 (30 ng/ml) is partially abolished if neurons are pre-treated for 30 min with the PI-3 kinase inhibitor,wortmannin. Media represent neurons exposed to OGD and IL-10 vehicle. *p<0.05 compared with the media group. N=8 pergroup. (D) Protection of cultured neurons from OGD by IL-10 (30 ng/ml) is specifically mediated by its receptor. *p<0.05compared with neurons not exposed to OGD. **p<0.05 compared with neurons exposed to OGD+IL-10. BA=blocking antibody.HD=heat denatured. N=8 per group.
192 B R A I N R E S E A R C H 1 3 7 3 ( 2 0 1 1 ) 1 8 9 – 1 9 4
Protection was also inhibited with the PI3K inhibitor,wortmannin, implicating PI3K-mediated AKT phosphorylationas an important pro-survival pathwayof IL-10 in cortical neurons.Whether inhibition of STAT-3 would similarly attenuate IL-10mediatedneuroprotection is an importantquestion to investigatein the future. In addition, it is also important to determine infuture studies whether neurons may also produce IL-10 whichwould point toward a potential endogenous neuroprotectiveautocrine loop under conditions of ischemia.
In conclusion, the data from this study provides evidencethat IL-10 directly protects cortical neurons through receptor-mediated mechanisms including the activation of the AKT andSTAT-3 pathways. IL-10 not only behaves as an anti-inflamma-tory cytokine but also directly protects neurons in modelssimulating the pathophysiology of stroke. IL-10 is a particularlyattractive, therapeutic candidate given its pleiotropic effects onpro-inflammatory and neuroprotective pathways. These in vitrostudiessupport further investigating IL-10upregulation inanimalmodels as a potential treatment approach for ischemic stroke.
4. Experimental procedures
4.1. Cell cultures
Neuronal cultures were prepared from 17 day embryos ofSprague–Dawley rats as we previously described (Zhao et al.,2006a), using Neurobasal medium (GibcoBRL, Gaithersburg, MD)with B27. The purity of the neuronal cultures was confirmed byMAP-2 staining showing the cultures were >95% neurons.
4.2. Oxygen glucose deprivation
Neurons at 13 days in vitro (DIV) were exposed to combinedoxygen and glucose deprivation (OGD). In brief, cultures werereplaced with serum- and glucose-free medium then placed in aBillups-Rothenberg modular incubator chamber (Del Mar, SanDiego, CA, USA), which was flushed for 5min with 95% nitrogenand5%CO2andthensealed (hypoxia).Thechamberwasplaced in
193B R A I N R E S E A R C H 1 3 7 3 ( 2 0 1 1 ) 1 8 9 – 1 9 4
a water-jacketed incubator (Forma Scientific, USA) at 37 °C for90min. Control glucose-containing normoxic cultures wereincubated for the same periods of time at 37 °C in humidified95% air and 5% CO2. IL-10 was added into neuronal cultures rightbefore exposure toOGD. In separate IL-10R blocking experiments,we pre-treated neuronal cultures with the anti-IL-10R1 antibody(R&D Systems) for 30min at 37 °C prior to OGD exposure in thepresence and absence of IL-10.
4.3. Glutamate excitotoxicity
Neurons were incubated for 15 min with or without IL-10 inHEPES-saline buffer and then exposed for 30 min to 100 μMglutamate in the same buffer.
4.4. MTT cell survival assay
Theviability of the cells inneuronalmonocultureswasassessedby their ability to uptake thiazolyl blue tetrazolium bromide(MTT) according to established protocols (Sharma et al., 2010).
4.5. RT-PCR of IL-10R
Total RNAwasextractedwith theTriazol (Invitrogen, La Jolla, CA)reagent according to the manufacturer's instructions. Reversetranscription polymerase chain reaction was carried out aspreviously described (Zhao et al., 2006b). The following cDNAprimers (Invitrogen) were used: IL-10 Receptor 1 Forward:CATTCCTCGTCTCGATCTCCAG; Reverse: CCAGATTAGTGC-CAAGGCTATC. β-Actin Forward: CAGTTCGCCATGGATGACGA-TATC; Reverse: CACGCTCGGTCAGGATCTTCATG. Scrambled IL-10 Receptor 1 primer pairs were also used as negative controls.Forward: CAGTCATCTTCACGCTCACCG; Reverse: CCGCAG-TACTGACATGGATAC. RT-PCR was performed using Fast-StartDNA Master plus reaction mix according to the manufacturer'sprotocol. The optical density of each band was quantified usingNIH Image-J software.
4.6. Akt (Ser 473) and STAT-3 (Tyr 705) phosphorylation
The relative amount of phospho-Akt and phospho-STAT-3protein in primary rat neuronal cultures was determinedusing ELISA kits for phospho-Akt and phospho-STAT-3(RayBiotech, Inc, Norcross, GA, USA) as described by Hsieh etal. (2009). Briefly, a 96-well microplate was coated withphospho-Akt or phospho-STAT-3 capture antibody and incu-bated overnight at room temperature before use. Culturedneurons were solubilized in lysis buffer according to themanufacturer's instructions and the cell lysates were addedinto wells. The phospho-Akt or phospho-STAT-3 detection Abwas then added to each well after washing, followed by 1 hincubation. The relative amount of phospho-Akt and phospho-STAT-3 proteinwas thendetectedusing a standard streptavidin-HRP format and the absorbance was read at 450 nm followed by570 nm subtraction. The value of phospho-Akt and phospho-STAT-3 was normalized to the protein content of each group.This approach to measure phosphorylated proteins has beenvalidated by Western blot (Chakrabarti et al., 2005; Chen et al.,2009; Lucas et al., 2010).
4.7. Immunohistochemistry
Cultured neurons were fixed with 4% PFA, permeabilized, andincubated with primary antibodies directed against IL-10R(Santa Cruz, USA, 1:300) and Neu-N (Millipore, 1:500). Afterwashing, the cultureswere incubatedat roomtemperaturewithphycoerytherin (PE) or FITC-conjugated secondary antibodies(Santa Cruz, USA). Subsequently, cells were visualized using aninverted Olympus fluorescence microscope.
4.8. Data analysis
Data are presented as mean±SEM and analyzed by one wayANOVA for multiple groups with post-hoc pair-wise compar-ison tests. p<0.05 was considered statistically significant.
4.9. Blinding
All experiments were performed and analyzed in a blindedmanner.
Acknowledgments
This work was funded by the Howard Hughes MedicalInstitute, the American Heart Association, and the NotsewOrm Sands Foundation.
R E F E R E N C E S
Boyd, Z.S., Kriatchko, A., Yang, J., Agarwal, N., Wax, M.B., Patil,R.V., 2003. Interleukin-10 receptor signaling throughSTAT-3 regulates the apoptosis of retinal ganglion cells inresponse to stress. Investig. Ophthalmol. Vis. Sci. 44,5206–5211.
Chen, Q.W., Edvinsson, L., Xu, C.B., 2009. Role of ERK/MAPK inendothelin receptor signaling in human aortic smooth musclecells. BMC Cell Biol. 10, 52.
de Bilbao, F., Arsenijevic, D., Moll, T., Garcia-Gabay, I., Vallet, P.,Langhans,W., Giannakopoulos, P., 2009. In vivo over-expressionof interleukin-10 increases resistance to focal brain ischemia inmice. J. Neurochem. 110, 12–22.
Grilli, M., Barbieri, I., Basudev, H., Brusa, R., Casati, C., Lozza, G.,Ongini, E., 2000. Interleukin-10 modulates neuronal thresholdof vulnerability to ischaemic damage. Eur. J. Neurosci. 12,2265–2272.
Hsieh, C.H., Nickel, E.A., Chen, J., Schwacha, M.G., Choudhry, M.A.,Bland, K.I., Chaudry, I.H., 2009.Mechanismof the salutary effectsof estrogen on Kupffer cell phagocytic capacity followingtrauma-hemorrhage: pivotal role of Akt activation. J. Immunol.182, 4406–4414.
Lucas, A., Kim, Y., Rivera-Pabon, O., Chae, S., Kim, D.H., Kim, B.,2010. Targeting the PI3K/Akt cell survival pathway to induce
194 B R A I N R E S E A R C H 1 3 7 3 ( 2 0 1 1 ) 1 8 9 – 1 9 4
cell death of HIV-1 infected macrophages withalkylphospholipid compounds. PLoS One 5 88(13), 2869–2876.
Sharma, S., Yang, B., Strong, R., Xi, X., Brenneman, M., Grotta, J.C.,Aronowski, J., Savitz, S.I., 2010. Bone marrow mononuclearcells protect neurons and modulate microglia in cell culturemodels of ischemic stroke. J. Neurosci. Res. 251(3), 189–192.
Spera, P.A., Ellison, J.A., Feuerstein, G.Z., Barone, F.C., 1998. IL-10reduces rat brain injury following focal stroke. Neurosci. Lett.251, 189–192.
(PPARgamma) activation protects neurons from NMDAexcitotoxicity. Brain Res. 1073–1074, 460–469.
Zhao, X., Zhang, Y., Strong, R., Grotta, J.C., Aronowski, J., 2006b.15d-Prostaglandin J2 activates peroxisomeproliferator-activatedreceptor-gamma, promotes expression of catalase, and reducesinflammation, behavioral dysfunction, and neuronal loss afterintracerebral hemorrhage in rats. J. Cereb. Blood FlowMetab. 26,811–820.
Zhou, Z., Peng, X., Insolera, R., Fink, D.J., Mata, M., 2009.Interleukin-10 provides direct trophic support to neurons. J.Neurochem. 110, 1617–1627.
