15681637-s1568163713x00042-s1568163713000676-main

Ageing Research Reviews 12 (2013) 941 966

Contents lists available at ScienceDirect

Ageing Research Reviews

j ourna l h om epage: www.elsev ier .com/ locate /ar r

Review

Pathogenesis of acute stroke and the role of inammasomes

David YaPrasad Ca Department ob School of Biomc Department o

a r t i c l

Article history:Received 21 AReceived in re12 SeptemberAccepted 19 S

Keywords:InammasomeIschemic strokCaspase-1CytokinesTreatment ways and current evidence for inammasome activation following cerebral ischemia, and the potential

for future treatments for stroke that may involve targeting inammasome formation or its products inthe ischemic brain.

2013 Elsevier B.V. All rights reserved.

Contents

1. Introd2. Stroke

2.1. 3. Mech

3.1. 3.2. 3.3.

4. DAMP4.1. 4.2.

4.3. 4.4.

CorresponE-mail add

1568-1637/$ http://dx.doi.ouction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 942. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 942Ischemic stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 942

anisms of cell death in ischemic stroke: ischemic cascade . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943Neuronal depolarization and excitotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943Reactive oxygen species and oxidative stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 943Inammation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944s and inammasomes: an overview in stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 944Molecular structure of NLRP1 and NLRP3 receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 946Potential stimulus of NLRP1 and NLRP3 receptor activation in stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9464.2.1. ATP-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9474.2.2. Acidosis-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9474.2.3. Cathepsin-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9474.2.4. K+-mediated NLRP activation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9494.2.5. ROS-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9494.2.6. Ca2+-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9504.2.7. Edema-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9504.2.8. PKR-mediated NLRP activation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 950Mechanism of NLR activation: NLRP1 and NLRP3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 951Formation of the NLRP1 and NLRP3 inammasome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 951

ding author at: Department of Physiology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore.resses: [email protected], [email protected] (T.V. Arumugam).

see front matter 2013 Elsevier B.V. All rights reserved.rg/10.1016/j.arr.2013.09.004ng-Wei Fanna,b, Seung-Yoon Leea, Silvia Manzaneroa,hunduria, Christopher G. Sobeyc, Thiruma V. Arumugama,b,

f Physiology, Yong Loo Lin School of Medicine, National University of Singapore, Singaporeedical Sciences, The University of Queensland, St Lucia, Queensland, Australia

f Pharmacology, Monash University, Clayton, Victoria, Australia

e i n f o

pril 2013vised form

2013eptember 2013

e

a b s t r a c t

Inammation is an innate immune response to infection or tissue damage that is designed to limit harmto the host, but contributes signicantly to ischemic brain injury following stroke. The inammatoryresponse is initiated by the detection of acute damage via extracellular and intracellular pattern recogni-tion receptors, which respond to conserved microbial structures, termed pathogen-associated molecularpatterns or host-derived danger signals termed damage-associated molecular patterns. Multi-proteincomplexes known as inammasomes (e.g. containing NLRP1, NLRP2, NLRP3, NLRP6, NLRP7, NLRP12,NLRC4, AIM2 and/or Pyrin), then process these signals to trigger an effector response. Briey, signalingthrough NLRP1 and NLRP3 inammasomes produces cleaved caspase-1, which cleaves both pro-IL-1and pro-IL-18 into their biologically active mature pro-inammatory cytokines that are released intothe extracellular environment. This review will describe the molecular structure, cellular signaling path-

942 D.Y.-W. Fann et al. / Ageing Research Reviews 12 (2013) 941 966

4.5. NLRP1 and NLRP3 inammasome-mediated cell death in stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9524.5.1. IL-1 and glutamate excitotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9524.5.2. IL-1 and oxidative stress . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9534.5.3. IL-1, IL-18 and IL-12. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 953

. . . . . 5. Evide . . . . . .6. Curre . . . . . .7. Futur . . . . . .

7.1. . . . . . 7.2. . . . . . .7.3. chan7.4. . . . . . 7.5. 8R. .

8. Summ . . . . . .Ackno . . . . . Refer . . . . . .

1. Introdu

Stroke iresulting inmajor caus2010). An estroke in thtions, it is eby 2030 (Soccurs wheor thromboby bleedingThe pathopand extenshomeostasitotoxicity, rarachidoniction of neurof the bloodet al., 2011)gen activatoocclusion isstroke therits narrow terebral hemrescue dyinframe (NINalternative tection. Dedeath and respectivelytrials due toal., 2000; CFosphenytoVan der Wo

Recent inammatotem that mcerebral iscplasma meglial cells ctor kappa Bpathways (endogenounecrotic ceduction of cell death. Tcomplexes

013; In pan)-likLRP3mune andurinressvolvts th

ke

ke isd oogicaeprintain

storlly, s. Ischtrokeccluts fotureic stl isch

Globr a m

comrdialselyuate

botic4.5.4. Pyroptosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .nce of inammasome activity in cerebral ischemia . . . . . . . . . . . . . . . . . . . . . . . . .nt treatments in stroke . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . e treatments in stroke: targeting inammasome signaling . . . . . . . . . . . . . . . . .Targeting signaling pathways: NF-B and MAPK pathway . . . . . . . . . . . . . . . .Targeting inammasome components: NLRPs, ASC and caspase-1 . . . . . . .Targeting receptors and ion channels: P2X7 receptor, Pannexin 1 and K+

Targeting secondary messengers: ROS, PKR and -arrestin-2. . . . . . . . . . . . .Targeting cytokines and cytokine receptors: IL-1, IL-18, IL-1R1 and IL-1ary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .wledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ction

s the second leading cause of mortality worldwide approximately 6 million deaths every year and is ae of long-term disability (World Health Organization,stimated 16 million people suffered from a rst-evere year 2005, and in the absence of any clinical interven-stimated that 23 million rst-ever strokes will occur

trong et al., 2007; Mukherjee and Patil, 2012). Stroken blood ow to the brain is interrupted by an embolictic occlusion of a cerebral artery (ischemic stroke) or

from a ruptured blood vessel (hemorrhagic stroke).hysiological processes following stroke are complexive, and include bioenergetic failure, loss of cell ions, acidosis, increased intracellular calcium levels, exci-eactive oxygen species-mediated toxicity, generation of

acid products, cytokine-mediated cytotoxicity, activa-onal and glial cells, complement activation, disruptionbrain barrier and inltration of leukocytes (Woodruff. Currently, intravenous recombinant tissue plasmino-r (r-tPA) to induce thrombolysis following a thrombotic

the only pharmacological agent approved for acuteapy. However, a major limitation of r-tPA therapy isherapeutic window of 3 h. An increased risk of intrac-orrhage, neuronal excitotoxicity and an inability to

g neurons preclude the use of r-tPA beyond this timeDS, 1995; Smith et al., 2008; Taschner et al., 2011). Anapproach for treating acute ischemic stroke is neuropro-spite neuroprotective agents decreasing neuronal cellinfarct size in cell culture and animal stroke models,, all such agents tested in patients have failed in clinical

deleterious side effects and/or low efcacy (Ahmed ethan et al., 1998; Cheng et al., 2004; Davis et al., 2000;in Internet Stroke Centre, 2007; Furuya et al., 2001;

et al., 22000).domaiand Nand imdamaginjury for exptheir inof agen

2. Stro

Stroin blooneuroldiate dto maienergyClinicarhagicof all sbotic oaccounthe rupIschemor foca2007).brain owhichmyocaConveris attenthromrp et al., 2002).ndings have provided insight into a newly describedry mechanism fundamental to the innate immune sys-ay contribute to neuronal and glial cell death duringhemia. There is emerging evidence to suggest that

mbrane pattern recognition receptors on neurons andan play an important role in activating nuclear fac-

(NFB) and mitogen activated protein kinase (MAPK)Tang et al., 2007, 2013). This occurs in response tos danger signals initiated by substances released fromlls in the ischemic core, leading to an increased pro-pro-inammatory cytokines and to neuronal and glialhese effects are mediated by intracellular multi-proteintermed inammasomes (Abulaa et al., 2009; Deroide

cerebral artcategories,

2.1. Ischem

Ischemiwithin the by an ischemerfusion (Kuand penumthe cerebrations of nehippocamptate granul. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 953 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 954 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 955nels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 956. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 957

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 958. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 958

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 958

Kono and Rock, 2008; Legos et al., 2001; Tamatani et al.,rticular, the NOD (nucleotide-binding oligomerizatione receptor (NLR) Pyrin domain containing 1 (NLRP1)

inammasomes, expressed abundantly in the braine cells, may play important roles in detecting cellulard mediating inammatory responses to aseptic tissueg ischemic stroke. This review will describe evidence

ion of NLRP1 and NLRP3 inammasomes in the brain,ement in ischemic stroke, and the therapeutic potentialat modify inammasome signaling.

an acute condition characterized by a sudden decreasew to brain tissue resulting in impairment or loss ofl function. The condition typically involves an imme-vation of both glucose and oxygen, which are needed

the metabolic demands of the brain as it holds noes that can be drawn upon (Ahmad and Graham, 2010).troke can be classied as either ischemic or hemor-emic stroke commonly accounts for approximately 80%

cases, and can be instigated by an embolic or throm-sion of a cerebral artery, whereas hemorrhagic stroker approximately 1520% of all cases and is initiated by

of a cerebral blood vessel (Gilgun-Sherki et al., 2002).roke can be further divided into two categories globalemia (Bacigaluppi et al., 2010; Durukan and Tatlisumak,al ischemic stroke occurs when blood ow to the entireajority part of the brain is stopped or severely reduced,monly occurs during a cardiac arrest associated with

infarction (Bottiger et al., 1999; Yonekura et al., 2004)., focal ischemic stroke occurs when cerebral blood owd in a specic brain region, prompted by an embolic or

occlusion (either transiently or permanently) in a major

ery (Hata et al., 2000; McAuley, 1995). Amongst the twofocal ischemic stroke is by far the most prevalent.

ic stroke

c stroke is characterized by the formation of two regionsischemic territory, a central ischemic core surroundedic penumbra (or peri-infarct zone) due to focal hypop-mar et al., 2010; Lo, 2008). The size of the ischemic corebra will usually depend on the severity and duration ofl artery occlusion and vulnerability of certain popula-urons to ischemia (e.g. CA1 pyramidal neurons in theus are more susceptible to ischemic damage than den-e neurons) (Brouns and De Deyn, 2009; Mattson et al.,

D.Y.-W. Fann et al. / Ageing Research Reviews 12 (2013) 941 966 943

2001). The formation of the ischemic core and penumbra duringischemic stroke is limited by the level of cerebral blood ow thatcontinues to perfuse the affected tissue (Mehta et al., 2007).

Under physiological conditions, cerebral tissue requires con-tinuous bloadequate suto maintaintriphosphatphorylationcerebral bloing ischemi1981; Bisdawill then uing in an inelectrically if cerebral ban ischemicnormal heamay generasue that is impaired (Aity of glucosblood vesseof cell deathlevels of perwill effectivsize (Baronischemic peical outcomthe most cli

3. Mechancascade

The ischular and cenecrosis or aevents bitive stress aischemia is both glucosgetic failureHertz and Daddition, reysis, whichlactate withdosis) and acid-sensinDing et al., et al., 2011;results in dying neuronsA major coinsult withmonly elicimembranes(Kaplan, 20will inducecells (Higuc2013; Mong

3.1. Neuron

Anoxic dgated calciu

an inux of Ca2+ ions, inducing uncontrolled release into thesynaptic cleft of glutamate, the major excitatory neurotransmitter(Arundine and Tymianski, 2003; Zhang et al., 2006). Energy failurewill also impair the re-uptake of glutamate by glutamate trans-

s (EAnaptiu, 20ate ay-5-thyl-s, drted ski, et a

Ca2+

of of Nthroain en, 20e, theally itly, t

of aecepe decger as of thal m005;

activ

en cia, Camaes (e

of NADnd ntic c+ ioonucpain)beensis, wnclus, in

oma-anoding cs (A

al., 2s-Lewwn m, relel wh1998

procontial ceed inshimeuronismod ow of at least 50 mL/100 g/min to sustain anpply of both glucose and oxygen, which are utilized

neurological function through energy (i.e. adenosinee; ATP) production by glycolysis and oxidative phos-

(Bisdas et al., 2004; Mehta et al., 2007). Conversely, ifod ow is reduced to less then 10 mL/100 g/min dur-c stroke, an ischemic core will develop (Astrup et al.,s et al., 2004; Mehta et al., 2007). This ischemic corendergo rapid, irreversible, necrotic cell death, result-farcted region of cerebral tissue that is metabolically,and functionally inactive (Mehta et al., 2007). However,lood ow remains between 10 and 50 mL/100 g/min,

penumbra may form between the ischemic core andlthy tissue (Astrup et al., 1981; Hossmann, 1994). Thiste a heterogeneous, meta-stable region of cerebral tis-metabolically active but electrically and functionallystrup et al., 1981; Moskowitz et al., 2010). The availabil-e and oxygen in the ischemic penumbra from collateralls will usually lead to a slower energy-dependent mode, known as apoptosis (Broughton et al., 2009). If normalfusion are not restored in sufcient time, the penumbraely merge with the ischemic core and increase infarct, 1999; Weinstein et al., 2004). Since salvage of thenumbra may be associated with improved neurolog-e and recovery, this region is currently considered to benically relevant target for acute stroke therapy.

isms of cell death in ischemic stroke: ischemic

emic cascade is a complex series of interlinked molec-llular mechanisms involved in ischemic cell death bypoptosis. In general, it is characterized by the following

oenergetic failure, acidotoxicity, excitotoxicity, oxida-nd inammation. The primary insult caused by cerebralhypoperfusion, and accordingly, insufcient delivery ofe and oxygen to the brain, which will induce bioener-

by stopping or slowing ATP production (Hertz, 2008;ienel, 2002; Hertz et al., 2007; Rossi et al., 2007). In

duced oxygen availability will initiate anaerobic glycol- leads to an increased production and accumulation ofin the ischemic tissue, decreasing intracellular pH (aci-causing acidotoxicity mediated by calcium-permeableg ion channels (ASICs) in the brain (Brouns et al., 2008;2000; Katsura et al., 1994; Park et al., 1999; Sherwood

Xiang et al., 2004; Xiong et al., 2004). The loss of ATPsfunction of all ATP-dependent ion pumps, thus render-

and glial cells highly susceptible to cerebral ischemia.nsequence of ATP loss occurs within minutes of the

the inhibition of Na+/K+-ATPase pumps, which com-ts rapid deterioration of ionic gradients across plasma, resulting in an inux of Na+ and efux of K+ ions02; Lipton, 1999; Mongin, 2007). This ionic imbalance

widespread anoxic depolarization in neurons and glialhi et al., 2002; Jarvis et al., 2001; Leichsenring et al.,in, 2007; White et al., 2012).

al depolarization and excitotoxicity

epolarization in neurons causes opening of voltage-m channels at the pre-synaptic terminal and allows

porterpre-syMassieglutamhydrox(N-meneuronnels gaTymianZhang furthercationinux water and brMongiin timespecicurreneffectsmate rand thexchana serienon-leet al., 2

3.2. Re

Whischemtiple denzymmationchain, nase aapoptolar Ca2

as end(m-calIt has apoptoteins, iproteinlymphand ryincludproteinLiu et Robertunknosomescytosoet al., Such aThe unand gldetect(Yama

In nmechaAT2; excitatory amino acid transporter 2) located onc neurons and surrounding astrocytes (Camacho and06; Rossi et al., 2000). The resultant accumulation oft synapses will then overstimulate AMPA (-amino-3-methyl-4-isoxazole propionic acid), kainate and NMDAd-aspartic acid)-type glutamate receptors on otheriving a further inux of Na+ and Ca2+ ions through chan-by these receptors (Arias et al., 1999; Arundine and2003; Li et al., 2007; Seo et al., 2001; Suzuki et al., 2012;l., 2006). Depolarization of additional neurons causesion inux and glutamate release, leading to local ampli-the initial ischemic insult. In addition, the increaseda+ ions into neurons causes an osmotic movement ofugh aquaporins into the cell, leading to cell swellingdema (Ayata and Ropper, 2002; Breder et al., 2000;07; Simard et al., 2007). If energy supply is not restoredse changes will result in rapid necrotic cellular lysis,

n the ischemic core (Sattler and Tymianski, 2000). Con-he increased Ca2+ ion inux mediated by the combinedctivation of voltage-gated Ca2+ channels, ASICs, gluta-tors and reverse operation of the Na+/Ca2+ exchanger,reased Ca2+ ion efux due to inhibition of the Na+/Ca2+

nd plasma membrane Ca2+-ATPase pump, will initiatenuclear and cytoplasmic events that lead to lethal oretabolic derangements known as excitotoxicity (Bano

Jeffs et al., 2007; Li et al., 2007; Schwab et al., 2002).

e oxygen species and oxidative stress

alcium homeostasis is disrupted during cerebrala2+ ions can become a powerful activator of mul-ging mechanisms, including activation of catabolicspecially endonuclease and calpain) and increased for-reactive oxygen species (from the electron transportPH oxidases, phospholipase A2, xanthine dehydroge-itric oxide synthase), ultimately leading to necrotic orell death. The increased concentration of intracellu-ns can activate nuclear and cytosolic proteases suchlease and calpains, i.e. calpain I (-calpain) and II, respectively (Lee et al., 2005; Neumar et al., 2001).

shown that endonuclease can cleave DNA to causehile activated calpain can hydrolyse cytoskeletal pro-

ding spectrin, fodrin, actin and tubulin; anti-apoptoticcluding Bcl-2 (B-cell lymphoma 2) and Bcl-xL (B-cellextra large); membrane proteins, including glutamateine receptors; and regulatory and signaling proteins,almodulin-binding protein, protein kinase C and G-ki et al., 2002; Buddle et al., 2003; Ling et al., 2002;004b; Nakagawa and Yuan, 2000; Neumar et al., 2001;is et al., 1994; Xu et al., 2009). In addition, through anechanism(s), calpain can induce the rupture of lyso-asing cathepsins (i.e. cathepsin B, D and L) into theich can hydrolyse similar calpain targets (Yamashima; Yamashima, 2004; Yamashima and Oikawa, 2009).cess is known as the calpain-cathepsin hypothesis.rolled proteolysis of these cellular proteins in neuronslls is an important component of neurodegeneration

necrosis that is observed primarily in the ischemic corea, 2004; Yamashima and Oikawa, 2009).ns and glial cells during cerebral ischemia, the primary

of Ca2+ ion uptake into the mitochondrial matrix is


through the calcium uniporter (Kirichok et al., 2004; Triantalouet al., 2013). Consequently, abnormal accumulation of Ca2+ ionswithin the mitochondrial matrix will decrease the mitochondrialtransmembrane potential and induce the formation of calciumprecipitateswithin the tron transpwith oxyge2004; Niemcytosolic CaNADPH (niproducing Oet al., 2011the transloplasma mewhich is utprostaglandproductionIn addition,dase by Ca2

of O2 (Abr2009). Incrlead to thegen peroxid(ONOO) bactivated n2001; Heeb

The incrinduces oxiimpact multeins and lcan stimula(NF-B) dirvating mitop38 MAPK aglial damaget al., 2001Schwaningestress can d(i.e. the majfacilitate thteins (such cytosol, leaby Ca2+ ionsis throughHayashi et severe oxidmoderate omarily in th

3.3. Inamm

Inammof ischemicsword, initto tissue rcharacterizcytokines, 1 (IL-1),astrocytes, and glial ce2001; Vila ethe expresssion molecuselectins (e1) on endo

al., 2004a; Ehrensperger et al., 2005; Huang et al., 2000; Yilmazand Granger, 2008; Zhang et al., 1998). These adhesion moleculesare crucial for the inltration of immune cells, particularly neu-trophils and monocytes/macrophages, into the ischemic territory

repee knt al., P-1/Cased

guirijev

studd nohancobseatherafterpite ompytes pro-i

NADse (iMP-

ellulademAma7a).

nvolv seveylato

of thhich

2004006;

Ps

ammdam

Howsue ssioncitsse isnd ind toted m

term andllow6; K

200Meyl

of ans ageolecu, NLRis sig2004ter et al., ent echadurin (i.e. calcium phosphate and calcium hydroxyapatite)inner mitochondrial membrane, perturbing the elec-ort chain and causing electron leakage that can reactn to produce superoxide (O2) (Green and Kroemer,inen, 2003; Triantalou et al., 2013). The increase in2+ can activate protein kinase C, which in turn activatescotinamide adenine dinucleotide phosphate) oxidase,2

(Brennan et al., 2009; Kahles et al., 2010; Yoshioka). Accumulation of Ca2+ within neurons can inducecation of cytosolic phospholipase A2 (PLA2) into thembrane, catalyzing the formation of arachidonic acid,ilized by cyclooxygenase and lipoxygenase to produceins and leukotrienes, respectively, with a concomitant

of O2 (Kishimoto et al., 2010; Tomimoto et al., 2002). conversion of xanthine dehydrogenase to xanthine oxi-+-activated proteases can result in an increased outputamov et al., 2007; Al-Gonaiah et al., 2009; Ono et al.,eased production of O2 from numerous sources can

formation of additional free radicals, such as hydro-e (H2O2), hydroxyl radical (OH), and peroxynitrite

y reacting with nitric oxide (NO) produced from Ca2+-euronal and endothelial NO synthase (n/eNOS) (Chan,a and El-Hanafy, 2012; Nanetti et al., 2007).eased production of reactive oxygen species (ROS)dative stress, a major cause of tissue damage that cantiple cellular components, including nucleic acids, pro-ipids (Allen and Bayraktutan, 2009). In addition, ROSte transcription factors such as nuclear factor kappa Bectly, and activator protein-1 (AP-1) indirectly by acti-gen activated protein kinases (MAPKs) (in particularnd c-Jun-N-terminal kinase; JNK) to cause neuronal ande by modulating caspase-mediated apoptosis (Barone; Chen et al., 2011; Kratsovnik et al., 2005; Ridder andr, 2009; Suzuki et al., 1997). Furthermore, oxidativeamage organelles such as the endoplasmic reticulumor site of calcium storage) and mitochondria, which cane release of additional Ca2+ ions and pro-apoptotic pro-as cytochrome c and apoptosis inducing factor) into theding to local amplication of the initial ischemic insults, and both endoplasmic reticulum stress and apopto-

the intrinsic and extrinsic pathway (Cao et al., 2004;al., 2005; Malhotra and Kaufman, 2007). In general,ative stress can cause cell death through necrosis, whilexidative stress can elicit apoptosis that is observed pri-e ischemic penumbra (Chen et al., 2011).

ation

ation plays a signicant role in the overall pathogenesis stroke. The inammatory response is a double-edgedially contributing to ischemic brain injury and thenegeneration (Chamorro and Hallenbeck, 2006). It ised by the production and release of pro-inammatorysuch as tumor necrosis factor- (TNF), interleukin-

IL-6 and IL-18, by activated cells including neurons,microglia and endothelial cells, resulting in neuronalll death during cerebral ischemia (Allan and Rothwell,t al., 2000). Pro-inammatory cytokines can also induceion of adhesion molecules, such as intercellular adhe-le-1 (ICAM-1), vascular adhesion molecules (VCAMs),

.g. P-selectin, E-selectin) and integrins (e.g. Mac-1, LFA-thelial cells, leukocytes and platelets (Arumugam et

duringdamagTang e1 (MCis reletant in(Dimitrecentshoweand enThese lium raction

Desing incleukoctional IL-18),synthalarly Mextracbrain edeath (al., 200to be ition ofanaphmationcells, wet al., et al., 2

4. DAM

Intissue 2008).bral tisprogreical deresponlular aresponassociasignalsPAMPscells foal., 2002002a,1997; tiationnoxiouand mNLRP7cess thet al., KempsMiao e

Rectory mdeath rfusion, which paradoxically, often leads to secondaryown as ischemic reperfusion injury (Buck et al., 2008;2006). In addition, monocyte chemoattractant proteinCL2), the major chemokine in mammalian systems,

by activated neurons and glial cells and is impor-ding leukocyte migration toward the damaged tissueic et al., 2006; Stamatovic et al., 2003). However, ay suggested that neurovascular leukocyte accumulation

spatial correlation with increased vessel permeabilityed expression of endothelial cell adhesion molecules.rvations may indicate that the neurovascular endothe-

than the brain parenchyma is the site of leukocyte stroke (Enzmann et al., 2013).the mechanisms of ischemic reperfusion injury remain-letely understood, it has been shown that inltratingcan release a variety of cytotoxic agents, including addi-nammatory cytokines (i.e. TNF, IL-1, IL-6, IL-12 andPH oxidase-derived ROS, NO from inducible nitric oxideNOS), and matrix metalloproteinases (MMPs, particu-2 and MMP-9). These MMPs can cause damage to ther matrix and blood brain barrier (BBB), exacerbatinga, hemorrhage, and ultimately, neuronal and glial cellntea et al., 2009; Asashi et al., 2001; Kriz, 2006; Yang et

In addition, the complement cascade has been showned in ischemic reperfusion injury through the produc-ral inammatory mediators, including C1, C3a and C5axins, that are involved in leukocyte recruitment and for-e membrane attack complex (MAC) in neurons and glial

causes cell lysis and further tissue damage (Arumugamb; Barnum et al., 2002; Gesuete et al., 2009; Leinhase

Van Beek et al., 2000) (Fig. 1).

and inammasomes: an overview in stroke

ation is an innate immune response to infection andage designed to limit harm to the host (Medzhitov,ever, as mentioned, the inammatory response in cere-damaged following ischemic stroke contributes to the

of ischemic brain injury and exacerbation of neurolog- (Chamorro and Hallenbeck, 2006). The inammatory

initiated by the detection of acute damage via extracel-tracellular pattern recognition receptors (PRRs), which

conserved microbial structures, termed pathogen-olecular patterns (PAMPs) and/or host-derived dangered damage-associated molecular patterns (DAMPs).

DAMPs may be released from stressed or damageding either microbial or non-microbial insults (Akira etono and Rock, 2008; Maslanik et al., 2013; Matzinger,2b, 2012; Medzhitov, 2008; Medzhitov and Janeway,an et al., 2006; Rock and Kono, 2008). Hence, the ini-

inammatory response requires sensors to detect anynt or irregularity within the cellular microenvironment,lar platforms such as the NLRP1, NLRP2, NLRP3, NLRP6,P12, NLRC4, AIM2 and Pyrin inammasomes, that pro-nal to trigger an appropriate effector response (Agostini; Chae et al., 2011; Fernandes-Alnemri et al., 2009;t al., 2011; Khare et al., 2012; Martinon et al., 2002;2010; Minkiewicz et al., 2013; Vladimer et al., 2012).ndings have provided insight into new inamma-nisms that may contribute to neuronal and glial cellg cerebral ischemia. There is emerging evidence to


Fig. 1. A schem glialbioenergetic f bral idelivery of bot or sloaddition, reduintracellular pdependent ionpre-synaptic tre-uptake of gdriving a furthwater into theConcurrently, derangementsof ROS (oxidatendoplasmic rthe initial ischbrain cells to endothelial ceIn addition, chintracellular; N

suggest thacan play anpathways. Tnecrotic celof pro-inamediated bimately 700Deroide et et al., 2001;Tamatani et

At presesome in nedetecting ceto aseptic 2009; Derothe NLRP1 ponents: thdomain)-likASC (apoptpase recrui(Lamkan ethe NLRP1 ipro-caspasein humans)atic diagram of the major cell injury mechanisms involved in causing neuronal andailure, acidotoxicity, excitotoxicity, oxidative stress and inammation. During cereh glucose and oxygen to the brain, which will induce bioenergetic failure by stopping

ced oxygen availability will initiate anaerobic glycolysis, which leads to increased productH (acidosis) causing acidotoxicity and necrotic cell death in the brain. During cerebral i

pumps (e.g. Na+/K+-ATPase pumps) to fail causing widespread anoxic depolarization erminals and allows an inux of Ca2+ ions, inducing uncontrolled release of glutamate ilutamate by glutamate transporters. The resultant build-up of glutamate at synapses wer inux of Na+ and Ca2+ ions through channels gated by these receptors. The increased

cell, leading to cell swelling and brain edema. If energy supply is not restored in time,the increased concentration of Ca2+ ions within neurons can initiate a series of nuclear

known as excitotoxicity by activating catabolic enzymes, NO and ROS generating enzymeive stress) that degrade key cellular components inducing apoptotic or necrotic cell deeticulum and mitochondria, which can facilitate the release of additional Ca2+ ions and premic insult by Ca2+ ions, and both endoplasmic reticulum stress and apoptosis through tincrease the production and release of proinammatory cytokines, which can cause cells and leukocytes to facilitate leukocyte inltration into the ischemic territory during remokines can be released by activated brain cells and contribute to guiding leukocyte mO, nitric oxide; ROS, reactive oxygen species; Pro, pro-inammatory).

t plasma membrane PRRs on neurons and glial cells important role in activating NF-B and MAPK signalinghis is in response to endogenous DAMPs released byls in the ischemic core, leading to increased productionmmatory cytokines, and neuronal and glial cell deathy large intracellular multi-protein complexes (approx-

kDa) termed inammasomes (Abulaa et al., 2009;al., 2013; Iyer et al., 2009; Kono and Rock, 2008; Legos

Li et al., 2009; Martinon et al., 2002; Savage et al., 2012; al., 2000).nt, it is thought that the NLRP1 and NLRP3 inamma-urons and glial cells may play an important role inllular damage and mediating inammatory responses

tissue injury during ischemic stroke (Abulaa et al.,ide et al., 2013; Savage et al., 2012). In humans,inammasome is composed of four cytoplasmic com-e NLRP1 (NOD (nucleotide-binding oligomerizatione receptor (NLR) Pyrin domain containing 1) receptor;osis-associated speck-like protein containing a cas-tment domain); pro-caspase-1 and pro-caspase-4 or 5t al., 2002; Martinon et al., 2002). However, in mice,nammasome is composed of the NLRP1 receptor, ASC,-1, pro-caspase-11 (homologs to pro-caspase-4 or -5

and XIAP (X-linked inhibitor of apoptosis) (De Rivero

Vaccari et a2009). The Ncomponentdomain)-likASC and pr2004; Schro

ActivatioNLRP3 recethe NLRP1 verts pro-cMartinon ewill cleaveinammatoare then relet al., 2011and a particet al., 2009;2006; LamkZhang et aland pro-IL-tinct regula2010; Mart(Fig. 2). Themembrane cell death in cerebral ischemia. These cell injury mechanisms includeschemia there is decreased blood ow, and accordingly, insufcientwing ATP production via glycolysis and oxidative phosphorylation. In

ion and accumulation of lactate within the ischemic tissue decreasingschemia there is a decreased production of ATP, which causes ATP-in neurons. This causes voltage-gated Ca2+ channels to open at thento the synaptic cleft. In addition, the energy failure will impair theill then overstimulate glutamate receptors on neighboring neurons,

inux of Na+ ions into neurons will cause an osmotic movement of these changes will result in rapid necrotic cellular lysis of neurons.

and cytoplasmic events that lead to lethal or non-lethal metabolics, and causing mitochondrial failure, which increases the productionath depending on severity. ROS can damage organelles such as theo-apoptotic proteins into the cytosol, leading to local amplication ofhe intrinsic and extrinsic pathway. Finally, ROS can activate residentll damage and induce the expression of cell adhesion molecules oneperfusion releasing additional proinammatory cytokines and ROS.igration toward the ischemic tissue (ATP, adenosine triphosphate; I,

l., 2008, 2009; Mawhinney et al., 2011; Silverman et al.,LRP3 inammasome is composed of three cytoplasmic

s: the NLRP3 (NOD (nucleotide-binding oligomerizatione receptor (NLR) Pyrin domain containing 3) receptor;o-caspase-1 in both mice and humans (Agostini et al.,der and Tschopp, 2010).n and subsequent oligomerization of the NLRP1 andptors independently will lead to the formation ofand NLRP3 inammasomes, respectively, which con-aspase-1 into cleaved caspase-1 (Agostini et al., 2004;t al., 2002). Following activation, cleaved caspase-1

pro-IL-1 and pro-IL-18 into biologically active pro-ry cytokines mature IL-1 and mature IL-18, whicheased into the extracellular environment (Bauernfeinda). In addition, cleaved capase-1 may induce apoptosisular type of cell death known as pyroptosis (Bergsbaken

Erener et al., 2012; Fink et al., 2008; Fink and Cookson,an, 2011; Sagulenko et al., 2013; Walsh et al., 2011;., 2003). The production and maturation of pro-IL-118 is a tightly regulated process, and involves two dis-tory signals (Bauernfeind et al., 2011a; Khare et al.,inon et al., 2009; Medzhitov, 2008; Yu and Finlay, 2008)

rst signal (Priming) involves the activation of plasmaPRRs (e.g. toll-like receptors, TLRs; TLR2 and TLR4),


Fig. 2. Mechan codedand MAPK-de and the activation in theformation is re pro-Ireleased from 18 carespective rec by thethrough apop e receinterleukin-1 , mito(nucleotide-bi nd 3);recruitment d rsor).

receptor foreceptor 1 box 1; and (Eigenbrod This up-regand both psignaling p2013; FredeKang et al.,2004a, 201Savage et asecond signof the NLRPirregularitiestress, resumasome, wcaspase-1 (Maslanik et

4.1. Molecu

In humatural domai(NAIP, CIITAFIIND (funcment domaSchultz et ais absent (2009). Thedomains: ana C-termina

ns arisms involved in the production and maturation of pro-IL-1 and IL-18. DAMPs enpendent transcription of NLRP1 and NLRP3 inammasome proteins, and pro-IL-1

and oligomerization of the NLRP1 and NLRP3 receptors independently, resulting sponsible for activating pro-caspase-1 into cleaved caspase-1, which then cleaves

the cell. These mature proinammatory cytokines mature IL-1 and mature IL-eptors on the same cell or neighboring neurons, astrocytes or microglia indicated tosis and pyroptosis (DAMPs, damage-associated molecular patterns; TLR, toll-likreceptor 1; IL-18R, interleukin-18 receptor; NF-B, nuclear factor kappa-B; MAPKnding oligomerization domain)-like receptor (NLR) Pyrin domain containing 1 aomain; XIAP, X-linked inhibitor of apoptosis; Cl, Cleaved; IL, interleukin; Pre, Precu

r advanced glycation end products (RAGE), and IL-1 domai

(IL-1R1), by DAMPs (e.g. HMGB1, High mobility groupIL-1) released from necrotic cells in the ischemic coreet al., 2008; Tang et al., 2007, 2013; Zheng et al., 2013).ulates gene transcription of inammasome proteins,ro-IL-1 and pro-IL-18 mediated by NF-B and MAPKathways (Bauernfeind et al., 2011b, 2009; Budai et al.,rick Lo et al., 2008; He et al., 2012; Juliana et al., 2010;

2000; Legos et al., 2001; Liao et al., 2012; Liu et al.,3; Mariathasan and Monack, 2007; Qiao et al., 2012;l., 2012; Tamatani et al., 2000; Zhao et al., 2013). Theal involves activation and consequent oligomerization1 and NLRP3 receptors individually by either DAMPs, ors within the cellular microenvironment from cellularlting in the formation of the NLRP1 and NLRP3 inam-hich then activates pro-caspase-1 to produce cleavedFaustin et al., 2007; Li et al., 2009; Martinon et al., 2002;

al., 2013; Savage et al., 2012).

lar structure of NLRP1 and NLRP3 receptors

ns the NLRP1 receptor is characterized by ve struc-ns: an N-terminal PYD (pyrin) domain, a central NACHT, HET-E and TP1) domain, LRRs (leucine rich repeats),tion to nd) and C-terminal CARD (caspase recruit-in) domain (Letunic et al., 2009; Martinon et al., 2009;l., 1998). However, in mice the N-terminal PYD domainFaustin et al., 2007; Hsu et al., 2008; Jha and Ting,

NLRP3 receptor is characterized by three structural N-terminal PYD domain, a central NACHT domain andl LRR domain (Fig. 3). The functions of the following

downstreamprotein ASCNACHT domtor, once acinammaso2007; Faustal., 2012; Mimplicated deletion of tor by remoMogridge, is autolyticnecessary fDOsualdo downstreamprotein pro

4.2. Potentistroke

The precreceptor acthe extensithe NLRP1 and bindingthe NLRP1 athe intracel2011; Kerssa fully detor activati by PRRs such as TLRs, RAGE and IL-1R1 allow the activation of NF-Bpro-IL-18 known as Signal 1 (Priming). The second signal involves

formation of the NLRP1 and NLRP3 inammasome. InammasomeL-1 and pro-IL-18 into mature proinammatory cytokines that aren then initiate autocrine and paracrine effects by binding onto their

dashed line. In addition, cleaved caspase-1 can initiate cell deathptor; RAGE, receptor for advanced glycation end products; IL-1R1,gen activated protein kinase; AP-1, activator protein-1; NLRP, (NOD

ASC, apoptosis-associated speck-like protein containing a caspase

e as follows: the N-terminal PYD domain facilitates

homotypic PYDPYD interactions with the adaptor

(Masumoto et al., 1999; Srinivasula et al., 2002). Theain is responsible for both the NLRP1 and NLRP3 recep-tivated, to oligomerize and form the central core of theme, which is an ATP-dependent process (Duncan et al.,in et al., 2007; Koonin and Aravind, 2000; Levinsohn etartinon et al., 2002). The LRR domain is considered to bein ligand sensing and autoregulation due to reports thatthe LRR domain results in a constitutively active recep-ving a possible autoinhibitory role of the LRR (Liao and2009; Truhlar and Komives, 2008). The FIIND domainally cleaved upon NLRP1 receptor activation, which isor NLRP1 inammasome activity (Finger et al., 2012;et al., 2011). The C-terminal CARD domain facilitates

homotypic CARDCARD interactions with the effector-caspase-1 (Srinivasula et al., 2002).

al stimulus of NLRP1 and NLRP3 receptor activation in

ise molecular and cellular stimuli for NLRP1 and NLRP3tivation during cerebral ischemia are unknown. Despiteve list of stimuli described to be capable of activatingand NLRP3 receptor, there is no evidence of direct lig-

(Petrilli et al., 2007a). Hence, it is now proposed thatnd NLRP3 receptor is a sensor for abnormal changes inlular environment in times of cellular stress (Davis et al.,e et al., 2011; Schroder and Tschopp, 2010). Althoughned mechanism leading to NLRP1 and NLRP3 recep-on has not been elucidated during cerebral ischemia,


Fig. 3. Structu roteinstructural dom e, the structural dom rizedcaspases from ding o1 and 3); ASC, a YD, pyrich repeats; F

numerous including eintracellulamitochondrcell swellinal., 2012; LArlehamn eNakahira etRossol et al(Fig. 4).

4.2.1. ATP-mThe NLR

by energy blast cells sMogridge, decrease thof ATP to Aactivate theprotein kinan unknowNLRP1 receATP depletiNLRP1 recethat mutaticaused consinhibit the Nwhich is relevels of cyactivity seereceptor (Li

4.2.2. AcidoThe NLR

extracellulain human m

ia, e of chemamolic arougon ofviro

Roin4). Hral domains in the NLRP1 and NLRP3 receptors and associated inammasome pains in humans a PYD, NACHT, LRRs, FIIND and CARD domain. However, in micains in humans and mice a PYD, NACHT and LRRs. The adaptor ASC, is characte

human and mouse are characterized by a CARD domain (NLRP, (NOD (nucleotide-binpoptosis-associated speck-like protein containing a caspase recruitment domain; PIIND, function to nd; CARD, caspase recruitment domain).

contributing cellular events are considered plausible,nergy depletion, acidosis, cathepsin release, decreasedr K+ concentration, increased ROS production, oxidizedial DNA, increased intracellular Ca2+ concentration,g, and protein kinase R (PKR) activation (Compan etee et al., 2012; Liao and Mogridge, 2012; Lindestamt al., 2010; Lu et al., 2012; Munoz-Planillo et al., 2013;

al., 2011; Petrilli et al., 2007b; Rajamaki et al., 2013;., 2012; Shimada et al., 2012; Zhou et al., 2010a, 2011)

ischemreleasepost-isdent inmetabtion thsecretilular en2013; al., 200ediated NLRP activationP1 receptor was recently demonstrated to be activateddepletion under in vitro conditions in human bro-ubjected to oxygen and glucose deprivation (Liao and2012). It was shown that ischemic conditions coulde levels of cytosolic ATP, thereby lowering the ratioMP (adenosine monophosphate) in the cytoplasm, to

main cellular energy sensor, AMPK (AMP-activatedase), and promote NLRP1 receptor activation throughn mechanism. Importantly, although AMPK promotesptor activation, activation of AMPK in the absence ofon in the cytoplasm was not sufcient to activate theptor (Liao and Mogridge, 2012). Lastly, it was shownon of the ATPase binding motif in the NLRP1 receptortitutive activation, suggesting that ATP might normallyLRP1 receptor instead of being required for assembly,

dundant under ischemic conditions due to decreasedtosolic ATP. This is in direct contrast to the abolition ofn when the same mutation is introduced into the NLRP3ao and Mogridge, 2012).

sis-mediated NLRP activationP3 receptor was recently shown to be activated byr and intracellular acidosis under in vitro conditionsacrophages (Rajamaki et al., 2013). During cerebral

tions is usuinitiates anmulation oet al., 2000;Xiang et al.vate the NLASIC1a, on et al., 2010et al., 2004been suggemore detaiment appedecreased ia mechanis2013).

4.2.3. CatheThe NLR

sins releaspermeabilizlate crystalet al., 20092010). Durisive releaseat the site Ca2+ ions fs in humans and mice. The NLRP1 receptor is characterized by vePYD domain is absent. The NLRP3 receptor is characterized by three

by two structural domains a PYD and CARD domain. Inammatoryligomerization domain)-like receptor (NLR) Pyrin domain containingrin domain; NACHT, NAIP, CIITA, HET-E and TP1 domain; LRR, leucine

xtracellular acidosis may be caused by either passiveH+ ions from necrotic cells in the ischemic core oric inammation. Inltration and activation of resi-matory cells in damaged tissue leads to an increase inctivity due to increased energy and oxygen consump-h anaerobic glycolysis, resulting in the production and

lactic acid, decreasing extracellular pH in the extracel-nment (acidosis) (Krawczyk et al., 2010; Rajamaki et al.,iotis et al., 2009; Tannahill and ONeill, 2011; Xiong etowever, intracellular acidosis under ischemic condi-ally caused by a reduction in oxygen availability, whichaerobic glycolysis resulting in the production and accu-f lactic acid within the cell (Brouns et al., 2008; Ding

Katsura et al., 1994; Park et al., 1999; Rossi et al., 2007;, 2004). An acidic extracellular environment may acti-RP3 receptor by H+ ions binding to ASICs, in particularneurons and glial cells and resulting in Ca2+ inux (Li; Pignataro et al., 2007; Sherwood et al., 2011; Xiong). Increased cytosolic Ca2+ concentration has recentlysted to activate the NLRP3 receptor, and is discussed inl below. Nevertheless, an acidic intracellular environ-ars to activate the NLRP3 receptor in synergy with antracellular K+ concentration seen in ischemia throughm that remains to be fully determined (Rajamaki et al.,

psin-mediated NLRP activationP1 and NLRP3 receptor may be activated by cathep-ed into the cytoplasm due to lysosomal membraneation, destabilization and rupture induced by particu-s (Hornung et al., 2008; Hoegen et al., 2011; Newman; Savage et al., 2012; Shi et al., 2013; Terada et al.,ng cerebral ischemia, this could be caused by the pas-

of cholesterol crystals from atherosclerotic plaquesof occlusion, or the release of soluble uric acid androm necrotic cells in the ischemic core undergoing


Fig. 4. Potential stimulus involved in NLRP1 and NLRP3 receptor activation in cerebral ischemia. The precise molecular and cellular stimuli of NLRP1 and NLRP3 receptoractivation during cerebral ischemia are unknown. However, several cellular events are considered plausible during cerebral ischemia including energy depletion, acidosis,cathepsin release, decreased intracellular K+ concentration, increased ROS production, oxidized mitochondrial DNA, increased intracellular Ca2+ concentration, cell swellingand PKR activation. During cerebral ischemia there is decreased levels of cytosolic ATP, thereby lowering the ratio of ATP/AMP in the cytoplasm, which activates AMPK topromote NLRP1 receptor activation. Extracellular and intracellular acidosis may activate the NLRP3 receptor. Extracellular acidosis caused by passive release of H+ ions fromnecrotic cells in the ischemic core or secretion from metabolically active leukocytes may activate the NLRP3 receptor by H+ ions binding onto ASIC1a on neurons and glialcells resulting in the inux of Ca2+ ions into the cytoplasm. Consequently, increasing the concentration of Ca2+ ions in the intracellular environment, which has recentlybeen suggested to activate the NLRP3 receptor. However, intracellular acidosis caused by a reduction in oxygen availability under ischemic conditions, initiates anaerobicglycolysis in the mitochondria resulting in the production and accumulation of lactic acid within the cell leading to a decrease in intracellular pH (acidosis) that appears toactivate the NLRP3 receptor in synergy with a decreased intracellular K+ concentration. The NLRP1 and NLRP3 receptor can be activated by cathepsins caused by lysosomalmembrane permeabilization, destabilization, and rupture releasing cathepsins into the cytoplasm induced by particulate crystals. The NLRP1 and NLRP3 receptor can beactivated by a decrease in K+ levels in the cytoplasm caused by dysfunction of the Na+/K+-ATPase pump due to a decreased production of ATP resulting in both an increasedinux and efux of Na+ and K+ ions, respectively. The increased inux of Na+ ions will promote an osmotic movement of water through aquaporins into the cell dilutingthe concentration of K+ ions in the cytoplasm; together with an increased efux of K+ ions into the extracellular environment by dysfunctional Na+/K+-ATPase pumps willboth decrease the concentration of K+ ions inside the cell. Alternatively, the passive release of ATP from necrotic cells in the ischemic core may bind onto plasma membraneP2X4 receptors on neurons, astrocytes or microglia, which can cause P2X4 receptors to open allowing an efux of K+ ions along its concentration gradient out of the celldecreasing the concentration of K+ ions in the cytoplasm. In addition, necrotic cells in the ischemic core will passively release K+ ions into the extracellular environment.Collectively these mechanisms increase the amount of K+ ions in the extracellular environment and activate Pannexin 1 channels on the plasma membrane. Opening ofPannexin 1 channels will lead to the release of more ATP, which can further activate more P2X4 and now P2X7 receptors on the same cell causing additional K+ efuxcreating a positive feedback loop. The NLRP3 receptor can be activated by localized increases in ROS levels in the cytoplasm. This may occur through perturbation of theelectron transport chain in the mitochondria or impaired mitophagy during cerebral ischemia causing TXNIP to bind with the NLRP3 receptor leading to its activation. TheNLRP3 receptor can be activated by oxidized mitochondrial DNA released by the mitochondria due to an increase in K+ efux and ROS. The NLRP3 receptor can be activatedby an increased intracellular Ca2+ concentration. The passive release of Ca2+ ions from necrotic cells in the ischemic core can bind to and activate CaSRs and GPR6CAreceptors on neighboring cells. Consequently, activation of CaSRs and GPR6CA receptors can interact with Gq and activate PLC, which cleaves PIP2 into DAG and InsP3.InsP3 will bind onto InsP3-R on the endoplasmic reticulum to stimulate the release of Ca2+ ions into the cytoplasm. A reduced concentration of cAMP in the cytoplasm couldpromote NLRP3 receptor activation caused by passive release of Ca2+ ions from necrotic cells in the ischemic core binding to and activating CaSRs but interacting with Gi toinhibit adenylate cyclase, which converts ATP to cAMP. Therefore, inhibition of adenylate cyclase will decrease the formation and concentration of cAMP in the cytoplasm,which is suggested to inhibit the NLRP3 receptor. The NLRP3 receptor is activated by cell swelling caused by an increased inux of Na+ ions into neurons, which causesan osmotic movement of water through aquaporins into the cell. The NLRP1 and NLRP3 receptor can be activated by PKR, which is activated by cellular stress, including adecreased intracellular K+ concentration, increased intracellular ROS production and increased intracellular Ca2+ concentration, which all occur during cerebral ischemia(ATP, adenosine triphosphate; AMP, adenosine monophosphate; cAMP, cyclic adenosine monophosphate; AMPK, adenosine monophosphate-activated protein kinase; CaSR,calcium-sensing receptor; AC, adenylate cyclase; ASIC, acid sensing ion channel; PIP2, phosphatidylinositol-4,5-bisphosphate; PLC, phospholipase C; DAG, diacylglycerol;InsP3, inositol triphosphate; InsP3R, inositol triphosphate receptor; ROS, reactive oxygen species; Ox, oxidized; TXNIP, thioredoxin-interacting protein; PKR, protein kinaseR; NLRP, (NOD (nucleotide-binding oligomerization domain)-like receptor (NLR) Pyrin domain containing 1 and 3); ASC, apoptosis-associated speck-like protein containinga caspase recruitment domain; Cl, cleaved; Ex, extracellular; IL, interleukin; In, intracellular; Pre, precursor).


crystallization to produce monosodium urate (MSU) and calciumphosphate (i.e. calcium pyrophosphate dihydrate and octacalciumphosphate) crystals, respectively, in the extracellular environment.These particulate crystals may then be taken up by resident cellssuch as astrcytosis, phaCD36) to b2010; DenoGasse et alLatz, 2013;2006; Nara2010; Rockuptake of ceacidic lysosmeabilizatimechanismsin B and Lstimulate thinhibitory dciated withNLRP1 and activated stFukuda et aet al., 2001;liposomes aNLRP3 receliposomes iwhich substatin 2 (i.e. and glial ce

4.2.4. K+-mThe NLR

in K+ levelsimmune ceal., 2013; Peresult fromNa+/K+-ATP2002; Liptoinux of Nthrough aquconcentratitogether wi1999; Mongnecrotic celP2X4 recepreceptor opet al., 2009;et al., 2010)sively releathese mechlar environmembrane nels will leaP2X7 recepttional K+ efFranchi et aFeuvre et a2006). The ltors being mP2X7 recept2007). In adstimuli, incCa2+ and inmate that o2006; Thom

concentration of K+ ions in the cytoplasm will create an environ-ment that is favorable for activating the NLRP1 and NLRP3 receptor(Kahlenberg and Dubyak, 2004; Lindestam Arlehamn et al., 2010;Munoz-Planillo et al., 2013; Petrilli et al., 2007b; Yu, 2003).

ROS- NLRvels iZhoueissncur tochoolip

all ofov e

and KKishi

al., 2 acti, andgers; Cast al.009)ent erce

2011niqupiratctionivatiia (N

receticulutowa, wh-tub

the the e

mitial arane.ondr

receZhouocho

to dse insis, w

reag totS-geagy)agicB (Le dued aondred ac2011tophion, or. Hventsion (ting ocytes, microglia and inltrating leukocytes via endo-gocytosis or membrane-bound scavenger receptors (i.e.e degraded by lysosomes in the cell (Duewell et al.,ble et al., 2011; Ea et al., 2011; Freigang et al., 2011;., 2009; Ghaemi-Oskouie and Shi, 2011; Grebe and

Hoffman et al., 2010; Jin et al., 2011; Martinon et al.,yan et al., 2011; Pazar et al., 2011; Rajamaki et al.,

et al., 2013; Sheedy et al., 2013). Consequently, thertain particulate crystals by endosomes that fuse withomes downstream induces lysosomal membrane per-on, destabilization and rupture through an unknown. This releases proteases such as cathepsins (e.g. cathep-) into the cytoplasm, which are proposed to eithere receptor itself, its receptor activators, or cleave eitheromains within the receptor or inhibitory proteins asso-

the receptor. Such a sequence is thought to release theNLRP3 receptor from an inactive conformation to anate during cerebral ischemia (Benchoua et al., 2004;l., 2004; Kilinc et al., 2010; Qin et al., 2008; Seyfried

Wen et al., 2008). Despite a recent study demonstratings a new type of particulate matter that can activate theptor, an alternative mechanism was observed wherebynduced the production of ROS from the mitochondria,equently activated transient receptor potential melas-TRPM2) channels to induce calcium inux in neuronslls, to activate the NLRP3 receptor (Zhong et al., 2013).

ediated NLRP activationP1 and NLRP3 receptor can be activated by a decrease

(


by oxidoreductase thioredoxin. Following an increase in cytoplas-mic ROS, this complex dissociates and allows TXNIP to bind withthe NLRP3 receptor (mainly in the LRR), leading to NLRP3 receptoractivation (Lane et al., 2013; Zhou et al., 2010a).

Recent sand generadrial DNA, oxidized mthe NLRP3 The study stion, demondamage dudrial DNA intransition (membrane chondrial Dmembrane major sourcConsequentdrial DNA cwith the msignaling (Sdence for aROS may ac

4.2.6. Ca2+-The NLR

an increasein vivo conet al., 2012bral ischemdecreased cfrom intracage and forendoplasm2005; JeffsSchwab et studies havtion of Ca2+

through botand GPR6Cof intracelling cerebraof Ca2+ froactivating Cand glial ce2012; TzimG-protein cproteins in et al., 2009Conigrave, mediated awith Gq acleaves phoglycerol (DAHofer and B2012). The the phosphPKC has anyto be deterplasm can reticulum tet al., 2012Rossol et al

Lastly, itcytoplasm

Kim et al., 2007; Lee et al., 2012; Peters-Golden, 2009; Trophy,1998). During cerebral ischemia, this may be caused by passiverelease of Ca2+ from necrotic cells in the ischemic core, which thenbinds to and activates CaSRs on neighboring neurons and glial cells

et aly, Cai a, whion oncen

the t prtrastP onich ised cor ac

Edem NLRellin2012ia, a+/K+-on ofeasedn, 20ent ells, pper2007nts pion or wn ma

rece dep012)fciethrouivati

Morial (T

receally T

recermor, whted Tquiret apph a p

and l sweia.

PKR- NLRed blysacal., 2

kin idenrferitioned btudies elegantly connected both an increase in K+ efuxtion of ROS with the production of oxidized mitochon-and demonstrated that once released into the cytosol,itochondrial DNA acts as a danger signal and activatesreceptor (Mathew et al., 2012; Shimada et al., 2012).howed that K+ efux-induced mitochondrial dysfunc-strated by a decreased MMP (a marker of mitochondrialring apoptosis), which released oxidized mitochon-to the cytosol through the mitochondrial permeabilityMPT) pore that forms across the inner mitochondrialduring ischemic conditions. This occurs because mito-NA that is normally attached to the inner mitochondrialwill be prone to oxidation due to its close proximity to ae of ROS during cerebral ischemia (Shimada et al., 2012).ly, this report demonstrated that oxidized mitochon-an bind to and activate the NLRP3 receptor, consistentitochondria playing a key role in NLRP3 inammasomehimada et al., 2012). Overall, the study has provided evi-

potentially unied mechanism by which K+ efux andtivate the NLRP3 receptor during cerebral ischemia.

mediated NLRP activationP3 receptor was recently shown to be activated byd intracellular Ca2+ concentration under in vitro andditions (Lee et al., 2012; Murakami et al., 2012; Rossol; Triantalou et al., 2013). As mentioned, during cere-ia this may occur by an increased calcium inux, aalcium efux and/or an increased release of calciumellular stores (mediated primarily by oxidative dam-mation of MAC on Ca2+ storing organelles such as theic reticulum) in neurons and glial cells (Bano et al.,

et al., 2007; Li et al., 2007; Murakami et al., 2012;al., 2002; Triantalou et al., 2013). In addition, recente shown that an increased extracellular concentra-ions can indirectly mediate NLRP3 receptor activationh plasma membrane calcium-sensing receptors (CaSRs)A receptors, together with a decreased concentrationular cAMP (Lee et al., 2012; Rossol et al., 2012). Dur-l ischemia, this may be achieved by passive releasem necrotic cells in the ischemic core binding to andaSRs and GPR6CA receptors on neighboring neuronslls (Korff et al., 2006; Lee et al., 2012; Rossol et al.,as et al., 2004). CaSRs and GPR6CA receptors are bothoupled receptors that can interact with Gq and Githe plasma membrane (Christiansen et al., 2007; Faure; Hofer and Brown, 2003; Pi et al., 2005; Khan and2010; Riccardi and Kemp, 2012). Consequently, Ca2+-ctivation of CaSRs and GPR6CA receptors can interactnd activate membrane-bound phospholipase C, whichsphatidylinositol-4,5-bisphosphate (PIP2) into diacyl-G) and inositol triphosphate (InsP3) (Lee et al., 2012;rown, 2003; Khan and Conigrave, 2010; Rossol et al.,main effect of DAG is to activate PKC, which catalyzesorylation of a variety of intracellular proteins. Whether

effect on the activation of the NLRP3 receptor remainsmined. Moreover, InsP3 that is released into the cyto-bind to InsP3 receptors (InsP3-R) on the endoplasmico stimulate the release of Ca2+ into the cytoplasm (Lee; Hofer and Brown, 2003; Khan and Conigrave, 2010;., 2012).

was shown that a reduced concentration of cAMP in thecould promote NLRP3 receptor activation (Bos, 2003;

(Korff quentlwith Gcyclaseinhibitand coinhibitwithouIn conof cAMby whdecrearecept

4.2.7. The

cell swet al., ischemthe Narioratiin incrMongimovemcell swand Roet al., lar eveactivatreceptstate iNLRP3changeet al., 2was suefux tor act2012).potentNLRP3especiNLRP3Furthevationpromowas reThus, ithrougvationfor celischem

4.2.8. The

activatlipopo(Lu et proteinmarilyby inteIn addactivat., 2006; Lee et al., 2012; Tzimas et al., 2004). Conse-2+-mediated activation of CaSRs can similarly interactnd inhibit the membrane-bound enzyme adenylateich converts ATP to cAMP (Lee et al., 2012). Therefore,f adenylate cyclase will tend to decrease the formationtration of cAMP in the cytoplasm, which is thought toNLRP3 receptor by interfering with the NACHT domaineventing ATP from binding onto the NLRP3 receptor., Rossol and colleagues (2012) detected no inuence

NLRP3 receptor activation. Hence, the mechanism(s)ncreased concentration of Ca2+ in conjunction with aoncentration of cAMP in the cytoplasm promotes NLRP3tivation in cerebral ischemia remains to be claried.

a-mediated NLRP activationP3 receptor was recently shown to be activated byg under in vitro conditions in immune cells (Compan; Schorn et al., 2011). This may occur during cerebrals a major consequence of ATP loss is the inhibition ofATPase pumps, which will commonly elicit rapid dete-

ionic gradients across the plasma membrane resulting Na+ inux and K+ efux (Kaplan, 2002; Lipton, 1999;

07). The inux of Na+ into neurons will result in osmoticof water through aquaporins into the cell so that thecausing brain edema in the ischemic penumbra (Ayata, 2002; Breder et al., 2000; Schorn et al., 2011; Simard). A recent study provided insight into the molecu-otentially driving volume-dependent NLRP3 receptor

(Compan et al., 2012). It was shown that the NLRP3as oligomerized into inactive complexes in a restingcrophages. However, in a hypotonic environment theptor was activated and underwent a conformationalendent on decreased intracellular K+ levels (Compan. Hence, inhibition of K+ efux during hypotonic shocknt to block NLRP3 receptor activation. In addition, Cl

gh swell-sensing Cl channels reduced NLRP3 recep-on, although K+ efux was unaffected (Compan et al.,eover, in a hypotonic environment, transient receptorRP) channels have been implicated in cell swelling andptor activation as they respond to membrane stretch,RPV2 in macrophages, which was demonstrated whenptor activation was inhibited during TRPV2 blockade.e, a hypotonic environment caused TRP channel acti-ich induced changes to intracellular Ca2+ levels andGF-activated kinase 1 (TAK1) phosphorylation, whichd for NLRP3 receptor activation (Compan et al., 2012).ears that cell swelling may activate the NLRP3 receptorathway that involves K+ and Cl efux, TRP channel acti-TAK1 phosphorylation, suggesting a complicated rolelling in activating the NLRP3 receptor during cerebral

mediated NLRP activationP1 and NLRP3 receptors were recently shown to bey PKR in the cytoplasm under in vitro conditions incharide (LPS) primed immune cells during apoptosis012). PKR is a ubiquitously expressed serine/threoninease activated by double-stranded RNA that was pri-tied as an innate immune anti-viral protein inducedon (IFN) (Garcia et al., 2006; Nakamura et al., 2010)., PKR is involved in inammation and appears to bey cellular stress, including a decreased intracellular K+


concentration, increased intracellular ROS production, increasedintracellular Ca2+ concentration and pro-inammatory cytokines(TNF and IFN), all of which occur during cerebral ischemia (Luet al., 2012; Nakamura et al., 2010). However, the ability of PKR toact as a danto be determwill undergorder to phNLRP3 receet al., 2006evidence fothat is intendings wecaspase-1 aPKR by shoand IL-1 cdendritic cetion, the stuthe NLRP1, ated by autoprotein faile

A recentneeded forNLRP1 andtoxin infec2013). Thiscaspase-1 adependent LPS-primedinteractionstion and prpathway (Hpresent in while PKR is present) model. Thein two disttosis (Hett ecytosolic reNOD2, NLRconcluded NLRP1, NLRtion. This pknown to acable to activinto their mstrating thamacrophagto clarify th

4.3. Mechan

There arthe literaturThe principtation of thbased uponclosed inactdomain is fofrom bindinpromote anYuan et al., vidually, orthe regulattory LRR dothe formati

upon activation (Faustin et al., 2007; Kadota et al., 2009). The sec-ond hypothetical mechanism is based on the assumption that theNLR is present in a off state bound to a host guard complex, whichprotects the NLR from proteasomal degradation and keeps the NLR

nactiFontaer, dDAMR, proon aa et axperion c

rmat

resee asinalt wiadaphe seNLRP

inteb-do

Fau Effe

to b receDCAb; FaFig. 5omaimasomengest

sugg twoermin of p

et thatd, de

depion (

inaRP3

inteb; Mows caspaue

2002NLRPn wil

formtly, xim

morer toSalvehoulspaseial foes (am eger-sensing molecule to detect these stimuli remainsined. Nevertheless, upon activation by a stimulus, PKR

o dimerization and auto-phosphorylation reactions inosphorylate the target protein in this case NLRP1 orptors to induce activation (Dey et al., 2005; Garcia; Lu et al., 2012). The 2012 study by Lu et al. providesr a broader role for PKR as a danger-sensing moleculegral to inammasome assembly and activation. Majorre that overexpression of PKR substantially enhancedctivation and IL-1 cleavage, whereas knockdown ofrt hairpin RNA (shRNA) inhibited caspase-1 activationleavage in different cell types including macrophages,lls and embryonic kidney cells (Lu et al., 2012). In addi-dy demonstrated that PKR physically interacted withNLRP3, NLRC4 and AIM2 receptors, which was medi--phosphorylation of PKR, while a kinase-defective PKRd to bind to or activate the NLRP3 receptor.

study demonstrated that PKR kinase activity is not ASC oligomerization and caspase-1 activation in the

NLRP3 inammasome in non-primed anthrax lethalted macrophages undergoing pyroptosis (Hett et al.,

demonstrates that PKR has an uncharacterized role inctivation and pyroptosis that is distinct from its kinase-role in inammasome formation during apoptosis in

cells. This might possibly occur through PKR protein with the IK complex, which causes IB phosphoryla-oteosomal degradation activating the NF-B signalingett et al., 2013). In other words, PKR kinase activity isa primed apoptotic and pyroptotic cell death model,kinase activity is absent (i.e. PKR protein interactionin a non-primed apoptotic and pyroptotic cell deathrefore, this study establishes a different role for PKRinct cell death pathways during apoptosis and pyrop-t al., 2013). Moreover, PKR failed to interact with otherceptors or inammasome family members, includingP12 and NLRX1 (Lu et al., 2012). Hence, the authorsthat PKR selectively and directly interacted with theP3, NLRC4 and AIM2 receptors to induce their activa-roposal was recently challenged, however, as stimulustivate the NLRP3, NLRC4 and AIM2 receptors were alsoate pro-caspase-1, and process pro-IL-1 and pro-IL-18ature forms in PKR decient macrophages, demon-

t PKR is not required for inammasome activation ines (He et al., 2013). Hence, additional studies are needede precise role of PKR in inammasome activation.

ism of NLR activation: NLRP1 and NLRP3

e two proposed models of NLR activation suggested ine (Kadota et al., 2009; Mayor et al., 2007; Shirasu, 2009).al difference between the two models is the implemen-e activation signal. The rst hypothetical mechanism is

the assumption that the NLR is present in the cell in aive form (i.e. an off state), whereby the regulatory LRRlded onto the NACHT domain and thus preventing ATPg and initiating a structural rearrangement that would

open active state (Jha and Ting, 2009; Riedl et al., 2005;2010). However, direct binding of a PAMP/DAMP indi-

their associated complex with adaptor molecules, toory LRR domain on the NLR, would cause the regula-main to be released from the NACHT domain, leading toon of an open active NLR that is able to oliogomerize

in an i2001; HowevPAMP/the NLize upFontanmore eactivat

4.4. Fo

At pmasomC-Terminteracof the ever, tof the proteintwo su2011a;2002).of ASCNLRP1or CAR2011a,2002) (PYD dinamrequiretor sugmodelleavingthe C-tdomai(FingerrmedcleavestrictlyactivatNLRP3the NLprotein2011a,this allin pro-tions (Bet al., of the domaito thesequenvia protwo ortogeth2003;

It spro-caessentmasomRathinve conformation (Boyer et al., 2011; Dangl and Jones,na et al., 2011; Kadota et al., 2009; Mayor et al., 2007).irect or indirect activation of the guard complex by aP would lead to complete or partial dissociation fromducing an open and active NLR that is able to oligomer-ctivation (Boyer et al., 2011; Dangl and Jones, 2001;l., 2011; Kadota et al., 2009; Mayor et al., 2007). Clearly,imental evidence is needed before either model of NLRan be conrmed.

ion of the NLRP1 and NLRP3 inammasome

nt, there are three proposed models of NLRP1 inam-sembly in humans. The rst model suggests that the

CARD domain of the NLRP1 receptor is able to directlyth the CARD domain in pro-caspase-1 in the absencetor protein ASC (Martinon et al., 2002) (Fig. 5A). How-cond model suggests that the N-terminal PYD domain1 receptor facilitates downstream homotypic protein-ractions with the adaptor protein ASC, as it containsmains: a PYD and CARD domain (Bauernfeind et al.,stin et al., 2007; Martinon et al., 2009; Srinivasula et al.,ctively, this would allow the PYD and CARD domainind with the PYD domain and CARD domain in the

ptor and pro-caspase-1, respectively, through PYDPYDRD homotypic protein interactions (Bauernfeind et al.,

ustin et al., 2007; Martinon et al., 2009; Srinivasula et al.,B). However, it was recently proven that the N-terminaln of the NLRP1 receptor is not required for NLRP1me activity but the dependence upon ASC and thet of the C-terminal CARD domain of the NLRP1 recep-ed an alternative model (Finger et al., 2012). The thirdests that ASC dimers form via PYDPYD association

free CARD domains at either end, which can bind withnal CARD domain on the NLRP1 receptor and the CARDro-caspase-1 in order to form the NLRP1 inammasomeal., 2012) (Fig. 5C). In addition, it was recently con-

the FIIND domain on the NLRP1 receptor is autolyticallymonstrating that the NLRP1 inammasome activity isendent upon this cleavage following NLRP1 receptorFinger et al., 2012; DOsualdo et al., 2011). Regardingmmasome assembly the N-terminal PYD domain ofreceptor facilitates downstream homotypic PYDPYDractions with the PYD domain of ASC (Bauernfeind et al.,artinon et al., 2009; Srinivasula et al., 2002). Effectively,the CARD domain of ASC to bind with the CARD domainase-1 through homotypic CARDCARD protein interac-rnfeind et al., 2011a,b; Martinon et al., 2009; Srinivasula) (Fig. 6). Activation and subsequent oligomerization1 and NLRP3 receptors individually via their NACHTl subsequently recruit ASC and pro-caspase-1, leadingation of the NLRP1 and NLRP3 inammasome. Con-

this will activate pro-caspase-1 into cleaved caspase-1ity-induced auto-activation, which is a process wheree pro-caspase-1 proteins are brought sufciently close

induce their autocatalytic activation (Boatright et al.,sen and Dixit, 1999).d be recognized that the interaction between human-4/5 or murine pro-caspase-11 to pro-caspase-1 isr caspase-1 activation in both NLRP1 and NLRP3 inam-Kang et al., 2000, 2002, 2003; Martinon et al., 2002;t al., 2012; Salskov-Iversen et al., 2011; Sollberger et al.,


Fig. 5. Three pThe rst modeis able to direCARDCARD hASC. HoweverNLRP1 receptothe adaptor prThis allows thCARD domain PYDPYD or CAterminal PYD dactivity but thdomain of thesuggests that domains at eiNLRP1 receptoNLRP1 inammlike receptor (Nprotein contaiCIITA, HET-E aCARD, caspase

2012; Wancaspase-4/5remains to binding anmasome refull-length masome byXIAP becomtively due an attenuatKeane et a2011; Vince

LRP3of theractioof ASpic CArizatioFig. 6. Ndomain tein intedomain homotyoligomeroposed models for NLRP1 inammasome assembly in humans. (A)l suggests that the C-Terminal CARD domain of the NLRP1 receptorctly interact with the CARD domain in precursor caspase-1 throughomotypic protein interactions in the absence of the adaptor protein, the second model (B) suggests that the N-terminal PYD domain of ther is able to facilitate downstream homotypic protein interactions withotein ASC as it contains two sub-domains: a PYD and CARD domain.e PYD and CARD domain of ASC to bind with the PYD domain andin the NLRP1 receptor and precursor caspase-1, respectively, throughRDCARD homotypic interactions. Recently, it was proven that the N-omain of the NLRP1 receptor is not required for NLRP1 inammasomee dependence upon ASC and the requirement of the C-terminal CARD

NLRP1 receptor suggests an alternative model. (C) The third modelASC dimers form via PYDPYD association leaving two free CARDther end, which can bind with the C-terminal CARD domain on ther and the CARD domain of precursor caspase-1 in order to form theasome (NLRP1, (NOD (nucleotide-binding oligomerization domain)-LR) Pyrin domain containing 1); ASC, apoptosis-associated speck-like

ning a caspase recruitment domain; PYD, pyrin domain; NACHT, NAIP,nd TP1 domain; LRR, leucine rich repeats; FIIND, function to nd;

recruitment domain; Pre, Precursor).

g et al., 1998). However, the timing when human pro- or murine pro-caspase-11 binds to the inammasomebe claried. In addition, the precise location, time ofd role of XIAP in murine NLRP1 and NLRP3 inam-mains to be established. However, it is suggested thatXIAP may serve to inhibit the NLRP1 and NLRP3 inam-

inhibiting pro-caspase-1 activation, although oncees cleaved it is unable to inhibit pro-caspase-1 effec-

to the production of an XIAP fragment (BIR1-2) withed capacity to inhibit pro-caspase-1 (Katz et al., 2001;l., 2001; Lotocki and Keane, 2002; Mawhinney et al.,

et al., 2012). Therefore, stroke-induced XIAP cleavage

apoptosis-assoPYD, pyrin dorepeats; CARD

may reduceunrestraineVaccari et mals are nein stroke-in

4.5. NLRP1 stroke

The NLRglial cell deby increasicytokines Icleaved casmost studie1R1) on netissue durineffects thaton the timi(Bernardinoet al., 2007tissue injur1999; Shaftin IL-1 prresponse thcommon cethe mechanple, evidencexcitotoxicthat glutamlinked in cstroke.

4.5.1. IL-1There is

IL-1 actioin an experan NMDA atein expresparietal corNMDA-industudy revealocalized towidespreadrole for IL-an experim inammasome assembly in humans and mice. The N-terminal PYD NLRP3 receptor facilitates downstream homotypic PYDPYD pro-ns with the PYD domain of ASC. This effectively allows the CARDC to bind with the CARD domain in precursor caspase-1 throughRDCARD protein interactions (NLRP3, (NOD (nucleotide-bindingn domain)-like receptor (NLR) Pyrin domain containing 3); ASC,ciated speck-like protein containing a caspase recruitment domain;

main; NACHT, NAIP, CIITA, HET-E and TP1 domain; LRR, leucine rich, caspase recruitment domain; Pre, Precursor).

the threshold for activation of pro-caspase-1, allowingd maturation of both pro-IL-1 and IL-18 (De Riveroal., 2008). Additional studies with XIAP-decient ani-eded to determine the consequences of XIAP cleavageduced inammasome signaling.

and NLRP3 inammasome-mediated cell death in

P1 and NLRP3 inammasome can mediate neuronal andath in ischemic stroke via a number of mechanisms

ng the production and secretion of pro-inammatoryL-1 and IL-18, and through pleiotropic effects ofpase-1 in mediating apoptosis and pyroptosis. Whiles suggest that IL-1 binding to the IL-1 receptor 1 (IL-urons and glial cells is harmful to the injured cerebralg ischemic stroke, some studies report neuroprotective

seem to be dependent on the concentration of IL-1, andng of the response relative to the ischemic stroke insult

et al., 2005; Jones et al., 2005; Lu et al., 2005; Shaftela). Importantly, IL-1 alone, in the absence of cerebraly, is not neurotoxic (Lawrence et al., 1998; Rothwell,el et al., 2007b). It is thus proposed that the increaseoduction after ischemic stroke is part of a protectiveat goes wrong. A number of neurological disorders sharell injury mechanisms and could provide indications toisms underlying the harmful effects of IL-1. For exam-e has emerged on the relationship between glutamate

ity and oxidative stress with IL-1. Hence, it is proposedate excitotoxicity and oxidative stress with IL-1 are

ausing neuronal and glial cell death during ischemic

and glutamate excitotoxicity evidence to suggest that glutamate excitotoxicity andns are not mutually exclusive. This was demonstratedimental study that intracerebroventricular injection ofgonist (i.e. cis-2,4-methanoglutamate) increased pro-sion of IL-1 in neurons, astrocytes and microglia in thetex and striatum of rats following 30 min to 7 days ofced excitotoxicity (Pearson et al., 1999). In addition, theled an early temporal expression of IL-1 in microglia

the site of cerebral tissue damage and a delayed, expression of IL-1 in astrocytes suggesting a diverse1 following NMDA-induced excitotoxicity. Similarly,ental study demonstrated that pre-treatment with


MK-801, a non-competitive NMDA receptor antagonist decreasedgene expression of TNF- and IL-1 in the parietal cortex follow-ing 4, 16 and 24 h of ischemia in a photothrombosis model of focalischemic stroke (Jander et al., 2000). This study was the rst tosuggest thatory gene edeath induclink as to hicity. This cstudies. A stion of a reccell death NMDA-indufor ischemiIn addition,intracerebrto increaseincreased exa link betweby which tStroemer anstrated in awas able toand phosphto increase toxic cell det al., 2003)administrattransportermate and iin a dose-dan experim1 activateto increase levels inducco-culturesit appears tglutamate e

4.5.2. IL-1There is

are not mustrated thathippocampacetyl-cysteaddition, antion of NMuncoupled the producttical neuroal., 1995). intraperiton(IRFI 042) dof glutathioin the cortein mice (Mathese studieand that oxing a possib

4.5.3. IL-1Although

receptor (ILfunction is2006; Felde18 promote

play a critical role in the host defence against infection by induc-ing the production of interferon- (IFN-) from activated TH1,natural killer (NK) and B cells (Nakahira et al., 2002; Yoshimotoet al., 1998). However, the overproduction of IL-12 from inltrat-

cropal tistionphags TN, lea0; N

In ad to snt polls, ao et n, IL

FasLs a pologic.h IL-utocceptnd tcell ng pSedim

traning: p

chemhemles (sion

2005perg

and

Pyropoptoformrammclusi

At pmacr

al., 2o py. Pyre andr en

me-1 thnd Cellulaic moellind Coroptoosome the(Bre, cellent

emicitionave t NMDA-induced excitotoxicity can activate inamma-xpression independently from neuronal and glial celled by cerebral ischemia and may provide a mechanisticow IL-1 mediates cell injury by regulating excitox-oncept was supported by a number of experimentaltudy demonstrated that intracerebroventricular injec-ombinant IL-1 receptor antagonist decreased neuronaland infarct size following 24 h of focal ischemia andced excitotoxicity suggesting that IL-1 is a mediator

c and excitotoxic damage (Relton and Rothwell, 1992). a number of experimental studies demonstrated thatoventricular injection of IL-1 into the cortex failed

infarct volume in either the striatum or cortex, butcitotoxic damage in the striatum and cortex suggestingen excitotoxicity and IL-1, although the mechanism(s)hey converge may be diverse (Lawrence et al., 1998;d Rothwell, 1998; Allan et al., 2000). This was demon-n experimental study where administration of IL-1

increase activation of Src tyrosine kinase, which boundorylated the NMDA receptor subunits (i.e. NR2A/B)Ca2+ inux through NMDA receptors inducing excito-eath in primary hippocampal neurons in rats (Viviani. Moreover, an experimental study demonstrated thation of IL-1 decreased gene expression of glutamate

subtype-1, which decreased the re-uptake of gluta-ncreased excitotoxicity in primary human astroyctesependent manner after 24 h (Hu et al., 2000). Finally,ental study demonstrated that administration of IL-d the cystine/glutamate antiporter (i.e. System x(c))intracellular cystine levels and extracellular glutamateing glutamate excitoxicity in mixed neuron-astrocyte

under ischemic conditions (Fogal et al., 2007). Hence,hat IL-1 stimulates a variety of pathways to inducexcitotoxicity.

and oxidative stress evidence to suggest that oxidative stress and IL-1tually exclusive. A number of studies have demon-

ROS can induce the expression of pro-IL-1 in mixedal cultures and attenuated by antioxidants such as N-ine (Brabers and Nottet, 2006; Min et al., 2003). In

experimental study demonstrated that administra-DA increased intracellular Ca2+ concentration, whichthe mitochondrial electron transport chain increasingion of ROS and induces oxidative stress in mouse cor-ns under glutamate excitotoxic conditions (Dugan etSimilarly, an experimental study demonstrated thateal administration of a lipid peroxidation inhibitor

ecreased malondialdehyde (MDA) levels, prevented lossne-reduced (GSH) levels and gene expression of IL-1x and hippocampus of kainic-acid induced brain injuryrini et al., 2004; Reynolds and Hastings, 1995). Hence,s propose that excitotoxicity can cause oxidative stress,idative stress can induce pro-IL-1 expression suggest-le mechanistic link.

, IL-18 and IL-12 IL-18 is structurally homologous to IL-1, and its-18R) belongs to the same IL-1R/TLR superfamily, its

quite different from IL-1 (Boraschi and Dinarello,rhoff-Mueser et al., 2005). In synergy with IL-12, IL-s T helper 1 (TH1)-mediated immune responses, which

ing macerebrproducmacrosuch aand NOal., 2001999).(CD8+)a poteglial cesis (Daadditioity andIL-18 iphysiostroke

Bothave ative recells, atarget signali2003; ondaryincludIL-18);CX3C-cmolecureperfuet al., EhrensYilmaz

4.5.4. Pyr

other a progand ex2009).cytes, Fink etundergminedrupturcellulaplasmacapaseFink apate cosmoting swFink aning pychromproducptosis Hencedependin isch

Addto clehages and IL-18 from neuronal and glial cells in ischemicsue induces a pro-inammatory state via an increased

of IFN-. Consequently, this stimulates parenchymales to produce additional pro-inammatory cytokinesF and IL-6, and neurotoxic mediators such as ROSding to severe cerebral tissue damage (Monteforte etakanishi et al.,

15681637-s1568163713x00042-s1568163713000676-main

Documents