**3.2 Activation of the inflammasome**

Activation of inflammasome requires the interaction between its receptors and the specific ligands grouped in the name of pathogen-associated molecular patterns (PAMPs) and danger-associated molecular patterns (DAMPs) [11, 12]. A large number of inflammasome ligands have been identified; the major ones are presented in **Table 1**. Receptors implicated in the inflammasome structure are located on the intracellular site of cell membrane. This localization means that the inflammasome's receptors are activated by ligands present on the inner aspect of the cell [20]. In other words, an inflammasome is activated in a cell only if the considered cell is infected, mutated, or damaged. The most studied inflammasome platform is the NLRP3 or cryopirine; its activation can be mediated by a double stimulus. The first is the stimulation of a toll-like receptor (TLR) which leads to the activation of the transcription pathway of pro-IL-1β that raises the transcription of the genes of NLRP3 and its deubiquitination [22]. The second stimulus is


#### **Table 1.**

*Major Inflammasome activators (modified from [20]).*

done directly on the NLRP3 through its receptors by a DAMP expressed by the cell secondary to the first stimulus and linked to a cell membrane damage, trouble of cell ionic or metabolic homeostasis, etc. Another activation mechanism of NLRP3 is described in Alzheimer's disease and implicates the β-amyloid protein [23]. Betaamyloid proteins activate the inflammasome pathway in the microglial cells and thus provoke the liberation of IL-1β and its pyropoptosis which lead to neural cell death. Inflammasomes are also activated by reactive oxygen species resulting from mitochondrial malfunctioning or destruction [20].

Regardless of the inflammasome receptor activating stimulus, it causes a conformational modification of the receptor with liberation of the NBD domain. This liberation of the NBD domain permits the oligomerization of the inflammasome receptor into a hexamer or heptamer and recruitment of an adaptor protein by homotypic PYD-PYD interaction in the case of NLRP3. The recruited adaptor protein also recruits the procaspase-1 by homotypic CARD-CARD interaction. The obtained conformational two-by-two rapprochement of procaspase-1 leads to their autoproteolytic cleavage and their autoactivation [20]. On active form, caspase-1 is a tetramer formed by two pairs P10 and P20 subunits. Active caspase-1 produces activation of IL-1β and IL-18 and the outbreak of pyropoptosis by induction of cell membrane pore formation, which leads to water influx into the cell, swelling, and then osmotic lysis. Interleukin-1β and IL-18 amplify inflammation reaction and activities of all types of lymphocyte. Pyropoptosis, defined as inflammatory programmed cell death, has been found in macrophages, dendritic cells, and neurons [24]. So, in the CNS, inflammation through inflammasome and caspase-1 activation leads to pyroptosis of neurons and microglial cells that play the role of macrophages. This cellular death occurs indirectly in the case of microglial cell death or directly in the case of neuronal death resulting in significant neurotoxicity observed in many diseases.

#### **4. Inflammation in neurotoxicity and neuroprotection**

#### **4.1 Inflammation in neurotoxicity**

At the level of the central nervous system (CNS) as we have shown previously, the inflammasome effects are much more detrimental than beneficial for its homeostasis. This detrimental effect has been observed in many neurological disorders where inflammasomes seem to provoke neurotoxicity, both directly or indirectly [2]. Among these disorders we have Alzheimer's disease, bacterial meningitis, mouse's equivalent multiple sclerosis, depression, etc. [2, 23, 25]. This evidence, built from clinical and experimental researches, is more often based on the observation of a rise in the expression of inflammasome NLRP3 in the CNS or in the peripheral blood or on the discovery of an anti-inflammasome activity of the drugs used in the treatment of these disorders. **Table 2** summarizes for each neurological disorder the role played by inflammasome and inflammation in its pathogenesis. Another fact is that a unique neuron culture treatment with IL-1β does not produce deleterious effect; however, when the administration is prolonged for several days, it leads to neurotoxicity [34]. The negative impact of pro-inflammatory cytokines on the CNS is also seen on glial cells. Indeed, glial cells are the targets of pro-inflammatory cytokines and are activated by an inflammatory stimulus (PAMPs or DAMPs). This glial cell activation leads to the production of cytokines responsible of a local inflammatory response. Astrocytes activated by inflammation produce neurotrophins and growth factors like nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and glial cell

**327**

*Neuroprotection: The Way of Anti-Inflammatory Agents DOI: http://dx.doi.org/10.5772/intechopen.90509*

> **Inflammasome actors**

NLRP3, ASC, IL-1β

NLRP1, NLRP3, CARD8, ASC, IL-1β, IL-18

*Implication of inflammasome in neurological pathologies (modified from [2]).*

**Neurological pathologies**

Alzheimer's disease

Multiple sclerosis

Amyotrophic lateral sclerosis

Parkinson's disease

Pneumococci meningitis

**Table 2.**

line-derived neurotrophic factor (GDNF) [34]. These trophic factors have a neuroprotective effect. In contrast, microglial cell activation leads to the release of neurotoxic factors such as pro-inflammatory cytokines, chemokines, free radicals, nitric oxide, and metalloproteases [34]. For example, in the case of stroke, vascular interruption provokes an ischemia with neural lysis. This neural lysis is associated with a massive release of intracellular contents into the extracellular compartment, among which is glutamate. At this stage two neurotoxicity pathways are triggered: the excitotoxicity pathway by massive glutamate release and the inflammation pathway by activation of microglial cells. Microglial cells are activated by the ischemic danger signal or through N-Methyl-D-aspartate (NMDA) receptors on their surface membrane that are sensible to glutamate [34]. This microglial cell activation leads to the production and release of pro-inflammatory cytokines and other molecules as specified previously. The consequences are neurotoxicity and in stroke an increase

NLRP3, IL-1β Activation of inflammasome NLRP3 by

and neurons death.

**Experimental justifications References**

[23, 25]

[26–28]

[29]

[30, 31]

[32, 33]

β-amyloid protein and production IL-1β by microglial cell leading to neuro-inflammation

Presence of SEP-like lesions in Muckle-Wells syndrome. Rise of gene's expression and concentrations of caspase-1, IL-18 in peripheral mononuclear cells. Gene's polymorphisms of

of superoxide dismutase in microglial mouse cell

α-synuclein protein. Neuro-degenerescence is

Gene's polymorphisms of NLRP1 and CARD8 and spinal fluid concentration of IL-1β and IL-18 are associated to clinical prognostic of meningitis. Low severity of meningitis in mouse's models deficient to NLRP3and ASC or

caspase-1 is associated with SEP.

provokes neuro-inflammation.

after inhibition of IL-1 or IL-18.

accelerated by excess IL-1.

Caspase-1, IL-1β Activation of caspase-1 and IL-1β by a mutant

NLRP3, IL-1 Activation of inflammasome NLRP3 by

of the core ischemia at the expense of ischemic penumbra.

**4.2 Metabolic syndrome as a cause of inflammation in neurotoxicity**

Neurotoxicity as we aforementioned results from multiple biochemical processes

including inflammation. Whether it is initiated and amplified at the level of the CNS or at the periphery, inflammation remains harmful to the CNS. As a matter of fact, when it comes to inflammation, there is a communication between the periphery and the CNS [34]. Before addressing, at the end of this section, this connection between CNS and periphery, we would first of all want to present the metabolic syndrome as a cause of peripheral inflammation that could have an impact on the CNS. The metabolic syndrome is in fact a metabolic disorder characterized by a group of conditions that increase the risk of developing cardiovascular diseases and type 2 diabetes mellitus. Two mechanisms are suggested in an attempt to explain


*Neuroprotection: The Way of Anti-Inflammatory Agents DOI: http://dx.doi.org/10.5772/intechopen.90509*

**Table 2.**

*Neuroprotection - New Approaches and Prospects*

mitochondrial malfunctioning or destruction [20].

**4. Inflammation in neurotoxicity and neuroprotection**

At the level of the central nervous system (CNS) as we have shown previously, the inflammasome effects are much more detrimental than beneficial for its homeostasis. This detrimental effect has been observed in many neurological disorders where inflammasomes seem to provoke neurotoxicity, both directly or indirectly [2]. Among these disorders we have Alzheimer's disease, bacterial meningitis, mouse's equivalent multiple sclerosis, depression, etc. [2, 23, 25]. This evidence, built from clinical and experimental researches, is more often based on the observation of a rise in the expression of inflammasome NLRP3 in the CNS or in the peripheral blood or on the discovery of an anti-inflammasome activity of the drugs used in the treatment of these disorders. **Table 2** summarizes for each neurological disorder the role played by inflammasome and inflammation in its pathogenesis. Another fact is that a unique neuron culture treatment with IL-1β does not produce deleterious effect; however, when the administration is prolonged for several days, it leads to neurotoxicity [34]. The negative impact of pro-inflammatory cytokines on the CNS is also seen on glial cells. Indeed, glial cells are the targets of pro-inflam-

matory cytokines and are activated by an inflammatory stimulus (PAMPs or DAMPs). This glial cell activation leads to the production of cytokines responsible of a local inflammatory response. Astrocytes activated by inflammation produce neurotrophins and growth factors like nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), and glial cell

observed in many diseases.

**4.1 Inflammation in neurotoxicity**

done directly on the NLRP3 through its receptors by a DAMP expressed by the cell secondary to the first stimulus and linked to a cell membrane damage, trouble of cell ionic or metabolic homeostasis, etc. Another activation mechanism of NLRP3 is described in Alzheimer's disease and implicates the β-amyloid protein [23]. Betaamyloid proteins activate the inflammasome pathway in the microglial cells and thus provoke the liberation of IL-1β and its pyropoptosis which lead to neural cell death. Inflammasomes are also activated by reactive oxygen species resulting from

Regardless of the inflammasome receptor activating stimulus, it causes a conformational modification of the receptor with liberation of the NBD domain. This liberation of the NBD domain permits the oligomerization of the inflammasome receptor into a hexamer or heptamer and recruitment of an adaptor protein by homotypic PYD-PYD interaction in the case of NLRP3. The recruited adaptor protein also recruits the procaspase-1 by homotypic CARD-CARD interaction. The obtained conformational two-by-two rapprochement of procaspase-1 leads to their autoproteolytic cleavage and their autoactivation [20]. On active form, caspase-1 is a tetramer formed by two pairs P10 and P20 subunits. Active caspase-1 produces activation of IL-1β and IL-18 and the outbreak of pyropoptosis by induction of cell membrane pore formation, which leads to water influx into the cell, swelling, and then osmotic lysis. Interleukin-1β and IL-18 amplify inflammation reaction and activities of all types of lymphocyte. Pyropoptosis, defined as inflammatory programmed cell death, has been found in macrophages, dendritic cells, and neurons [24]. So, in the CNS, inflammation through inflammasome and caspase-1 activation leads to pyroptosis of neurons and microglial cells that play the role of macrophages. This cellular death occurs indirectly in the case of microglial cell death or directly in the case of neuronal death resulting in significant neurotoxicity

**326**

*Implication of inflammasome in neurological pathologies (modified from [2]).*

line-derived neurotrophic factor (GDNF) [34]. These trophic factors have a neuroprotective effect. In contrast, microglial cell activation leads to the release of neurotoxic factors such as pro-inflammatory cytokines, chemokines, free radicals, nitric oxide, and metalloproteases [34]. For example, in the case of stroke, vascular interruption provokes an ischemia with neural lysis. This neural lysis is associated with a massive release of intracellular contents into the extracellular compartment, among which is glutamate. At this stage two neurotoxicity pathways are triggered: the excitotoxicity pathway by massive glutamate release and the inflammation pathway by activation of microglial cells. Microglial cells are activated by the ischemic danger signal or through N-Methyl-D-aspartate (NMDA) receptors on their surface membrane that are sensible to glutamate [34]. This microglial cell activation leads to the production and release of pro-inflammatory cytokines and other molecules as specified previously. The consequences are neurotoxicity and in stroke an increase of the core ischemia at the expense of ischemic penumbra.

#### **4.2 Metabolic syndrome as a cause of inflammation in neurotoxicity**

Neurotoxicity as we aforementioned results from multiple biochemical processes including inflammation. Whether it is initiated and amplified at the level of the CNS or at the periphery, inflammation remains harmful to the CNS. As a matter of fact, when it comes to inflammation, there is a communication between the periphery and the CNS [34]. Before addressing, at the end of this section, this connection between CNS and periphery, we would first of all want to present the metabolic syndrome as a cause of peripheral inflammation that could have an impact on the CNS. The metabolic syndrome is in fact a metabolic disorder characterized by a group of conditions that increase the risk of developing cardiovascular diseases and type 2 diabetes mellitus. Two mechanisms are suggested in an attempt to explain

the genesis of inflammation in metabolic syndrome. The first is a dysfunction of the organelles of adipocytes, observed in obesity; the second is adipose tissue hypoxia also observed in obesity [35]. The first mechanism suggests that hypertrophic adipose tissue found in obesity undergoes excessive lipolysis resulting in hyperlipidemia and an increase in circulating fatty acid levels. This increase in circulating levels of fatty acids, coupled with an abundance of carbohydrates, results in an increase in the oxidative activity of mitochondria that produce excess energy. As time goes by, this state results in a dysfunction of the mitochondria freeing a large quantity of electrons responsible for an increased production of reactive oxygenated compounds. This oxidative stress can subsequently activate the innate immune system and thus cause inflammation. Furthermore, the excess of nutrients overruns the endoplasmic reticulum, resulting in a faulty plication of proteins which activates the response to faulty plication of proteins. This response stimulates the activation of three membranous proteins: PKR-like eukaryotic initiation factor 2-alpha kinase (PERK), inositol requiring enzyme-1 (IRE-1), and activating transcription factor-6 (AFT-6). PERK, IRE-1, and AFT-6 significantly enhance inflammation by activating the signaling pathway NF-kB [35].

Concerning the second mechanism, it is suggested that a localized hypoxia could initiate a dysregulation of adipokines in obesity. As a matter of fact, adipose tissue is mainly made up of adipocytes, but also preadipocytes, resident macrophages, fibroblasts, and endothelial cells. With the increase in adipose tissue observed in obesity, there is a need for a significant angiogenesis. The hypoxic signal present during this expansion results in the activation of transcription factors like the hypoxiainducible factors which are required in the activation of genes associated with angiogenesis, glucose metabolism, stress, and inflammation. Moreover, in vitro data reveal that human preadipocytes, when exposed to hypoxia, increase their expression of leptin and reduce their expression of the peroxisome proliferator-activated receptor gamma (PPARγ). Yet, agonists of the PPARγ stimulate insulinosensitivity and reduce inflammation. Furthermore, exposed to hypoxia, resident macrophages produce pro-inflammatory cytokines [35]. In type 2 diabetes, coupled with the mechanisms mentioned above, chronic hyperglycemia maintains a vicious circle. In fact, chronic hyperglycemia is responsible for an increase in glycation end products (AGEs) whose receptors belong to the family of PRRs. So, glycated plasma proteins, glycated lipids, or nucleic acids bind to AGE receptors present at the surface of macrophages and provoke a pro-inflammatory and pro-oxidative response [35].

Therefore, the metabolic syndrome induces a state of peripheral inflammation that becomes chronic because it is maintained by its causative process. This peripheral inflammation can directly affect the CNS through produced and circulating inflammatory mediators. These mediators penetrate the CNS via areas without a blood–brain barrier like the periventricular choroid plexuses following which they cause the aforementioned neurotoxic effects [34]. Furthermore, the blood–brain barrier is capable of transmitting an inflammatory message from the vascular endothelium to the CNS via active mechanisms involving cyclooxygenases [34]. Through these mechanisms, an inflammation at the periphery, if it lasts long enough, can extend to the CNS and result in neurotoxicity and subsequent neurologic disorders.

#### **4.3 Inflammation and neuroprotection**

Actually, even if some anti-inflammatory strategies have proven their efficacy in animal models, none have demonstrated efficacy in humans in the prevention or treatment of neurological diseases associated with neurotoxicity. However, with conclusive experimental results on the use of anti-inflammatory drugs in neuroprotection, this therapeutic approach presents encouraging prospects for clinical

**329**

*Neuroprotection: The Way of Anti-Inflammatory Agents DOI: http://dx.doi.org/10.5772/intechopen.90509*

expression in CNS cells.

of inflammation [38].

research. In doing so, after bringing out the negative impact of inflammation on the central nervous system (CNS), it seems appropriate to present some strategies explored or still to be explored in an attempt to inhibit neuro-inflammation and prevent or treat neurotoxicity associated with many neurological disorders. Glucocorticoid and general anesthesia products have stimulated a strong interest in neuroprotection in the cases of stroke on experimental animal models; this has not been demonstrated yet in humans [34]. Indeed, glucocorticoids have been found to be ineffective in stroke, head trauma, and meningeal hemorrhage [34]. And classic hypnotic agents like thiopental, midazolam, or propofol have peripheral immune-modulatory effects and are capable of inhibiting inflammatory response. They inhibit chemotaxis, adherence of neutrophils, phagocytosis, and liberation of free radicals and pro-inflammatory cytokines like IL-1β and TNFα in experimental mouse model; however, these activities have not been demonstrated in humans yet [34]. In general, having in mind previously described inflammation and inflammatory neurotoxicity mechanisms, we can conclude that neuroprotection strategies based on modulation of inflammation have to maintain the beneficial roles of immunological defense and healing of inflammation while neutralizing its neurotoxic consequences. Thus, three anti-inflammatory strategies for neuroprotection axis can be developed: the modulation of the communication between peripheral inflammation and CNS, the modulation of interaction between pro-inflammatory cytokines and their intracerebral targets, and the modulation of inflammasome

In relation to the first axis, namely, the modulation of the communication between peripheral inflammation and CNS and the COX inhibitors (nimesulide and indomethacin) has shown a neuroprotective activity in baby mice with brain lesions. This neuroprotective activity is made possible by inhibition of the communication through the blood–brain barrier between activated peripheral inflammatory cells and the CNS [34]. With the same idea, the COX inhibitors have been presented as potentially beneficial in the treatment of major depression and other psychiatric disorders. Indeed, celecoxib has presented a beneficial effect in the treatment of major depression and schizophrenia especially in early stages [36]. Acetyl salicylic acid in particular seems to have both a preventive and therapeutic effect on schizophrenia [36]. Communication between peripheral inflammation and the CNS does not occur solely via the blood–brain barrier as it can also be done through the parasympathetic and sympathetic systems. Indeed, immune cells present at their surfaces nicotinic receptors for acetylcholine and β-adrenergic receptors for catecholamine [34]. These receptors link immune cells to parasympathetic and sympathetic systems respectively. Thus, a pharmacologic vagal or noradrenergic stimulation could represent a potential target for neuroprotection. For this purpose, vagal stimulation potentially passing through the modulation of lipocalin prostaglandin D2 synthase (L-PGDS) has shown in rat models with ischemic stroke a neuroprotective effect against ischemia reperfusion [37]. Also, a noradrenergic stimulation has shown, in Parkinson's disease, a neuroprotective effect by inhibition

Concerning the modulation of interaction between pro-inflammatory cytokines and their intracerebral target strategy, specific receptor antagonist of IL-1 appears to be the most conclusive therapeutic approach. This antagonist is produced endogenously following brain injury, and its administration by systemic or intracerebral route leads to a reduction in the size of lesions in mouse models [34]. Furthermore, Veltkamp et al*.* report the use, via general route of anakinra, of an antagonist of IL-1 receptors in a clinical trial on a patient having stroke [39]. This clinical trial has shown a great reduction of national institute of health stroke scale (NIHSS), and it also shows more patient with modified Rankin score (mRS) of 0–1 in 3 months [39].

#### *Neuroprotection: The Way of Anti-Inflammatory Agents DOI: http://dx.doi.org/10.5772/intechopen.90509*

*Neuroprotection - New Approaches and Prospects*

activating the signaling pathway NF-kB [35].

**4.3 Inflammation and neuroprotection**

the genesis of inflammation in metabolic syndrome. The first is a dysfunction of the organelles of adipocytes, observed in obesity; the second is adipose tissue hypoxia also observed in obesity [35]. The first mechanism suggests that hypertrophic adipose tissue found in obesity undergoes excessive lipolysis resulting in hyperlipidemia and an increase in circulating fatty acid levels. This increase in circulating levels of fatty acids, coupled with an abundance of carbohydrates, results in an increase in the oxidative activity of mitochondria that produce excess energy. As time goes by, this state results in a dysfunction of the mitochondria freeing a large quantity of electrons responsible for an increased production of reactive oxygenated compounds. This oxidative stress can subsequently activate the innate immune system and thus cause inflammation. Furthermore, the excess of nutrients overruns the endoplasmic reticulum, resulting in a faulty plication of proteins which activates the response to faulty plication of proteins. This response stimulates the activation of three membranous proteins: PKR-like eukaryotic initiation factor 2-alpha kinase (PERK), inositol requiring enzyme-1 (IRE-1), and activating transcription factor-6 (AFT-6). PERK, IRE-1, and AFT-6 significantly enhance inflammation by

Concerning the second mechanism, it is suggested that a localized hypoxia could initiate a dysregulation of adipokines in obesity. As a matter of fact, adipose tissue is mainly made up of adipocytes, but also preadipocytes, resident macrophages, fibroblasts, and endothelial cells. With the increase in adipose tissue observed in obesity, there is a need for a significant angiogenesis. The hypoxic signal present during this expansion results in the activation of transcription factors like the hypoxiainducible factors which are required in the activation of genes associated with angiogenesis, glucose metabolism, stress, and inflammation. Moreover, in vitro data reveal that human preadipocytes, when exposed to hypoxia, increase their expression of leptin and reduce their expression of the peroxisome proliferator-activated receptor gamma (PPARγ). Yet, agonists of the PPARγ stimulate insulinosensitivity and reduce inflammation. Furthermore, exposed to hypoxia, resident macrophages produce pro-inflammatory cytokines [35]. In type 2 diabetes, coupled with the mechanisms mentioned above, chronic hyperglycemia maintains a vicious circle. In fact, chronic hyperglycemia is responsible for an increase in glycation end products (AGEs) whose receptors belong to the family of PRRs. So, glycated plasma proteins, glycated lipids, or nucleic acids bind to AGE receptors present at the surface of macrophages and provoke a pro-inflammatory and pro-oxidative response [35]. Therefore, the metabolic syndrome induces a state of peripheral inflammation that becomes chronic because it is maintained by its causative process. This peripheral inflammation can directly affect the CNS through produced and circulating inflammatory mediators. These mediators penetrate the CNS via areas without a blood–brain barrier like the periventricular choroid plexuses following which they cause the aforementioned neurotoxic effects [34]. Furthermore, the blood–brain barrier is capable of transmitting an inflammatory message from the vascular endothelium to the CNS via active mechanisms involving cyclooxygenases [34]. Through these mechanisms, an inflammation at the periphery, if it lasts long enough, can extend to the CNS and result in neurotoxicity and subsequent neurologic disorders.

Actually, even if some anti-inflammatory strategies have proven their efficacy in animal models, none have demonstrated efficacy in humans in the prevention or treatment of neurological diseases associated with neurotoxicity. However, with conclusive experimental results on the use of anti-inflammatory drugs in neuroprotection, this therapeutic approach presents encouraging prospects for clinical

**328**

research. In doing so, after bringing out the negative impact of inflammation on the central nervous system (CNS), it seems appropriate to present some strategies explored or still to be explored in an attempt to inhibit neuro-inflammation and prevent or treat neurotoxicity associated with many neurological disorders. Glucocorticoid and general anesthesia products have stimulated a strong interest in neuroprotection in the cases of stroke on experimental animal models; this has not been demonstrated yet in humans [34]. Indeed, glucocorticoids have been found to be ineffective in stroke, head trauma, and meningeal hemorrhage [34]. And classic hypnotic agents like thiopental, midazolam, or propofol have peripheral immune-modulatory effects and are capable of inhibiting inflammatory response. They inhibit chemotaxis, adherence of neutrophils, phagocytosis, and liberation of free radicals and pro-inflammatory cytokines like IL-1β and TNFα in experimental mouse model; however, these activities have not been demonstrated in humans yet [34]. In general, having in mind previously described inflammation and inflammatory neurotoxicity mechanisms, we can conclude that neuroprotection strategies based on modulation of inflammation have to maintain the beneficial roles of immunological defense and healing of inflammation while neutralizing its neurotoxic consequences. Thus, three anti-inflammatory strategies for neuroprotection axis can be developed: the modulation of the communication between peripheral inflammation and CNS, the modulation of interaction between pro-inflammatory cytokines and their intracerebral targets, and the modulation of inflammasome expression in CNS cells.

In relation to the first axis, namely, the modulation of the communication between peripheral inflammation and CNS and the COX inhibitors (nimesulide and indomethacin) has shown a neuroprotective activity in baby mice with brain lesions. This neuroprotective activity is made possible by inhibition of the communication through the blood–brain barrier between activated peripheral inflammatory cells and the CNS [34]. With the same idea, the COX inhibitors have been presented as potentially beneficial in the treatment of major depression and other psychiatric disorders. Indeed, celecoxib has presented a beneficial effect in the treatment of major depression and schizophrenia especially in early stages [36]. Acetyl salicylic acid in particular seems to have both a preventive and therapeutic effect on schizophrenia [36]. Communication between peripheral inflammation and the CNS does not occur solely via the blood–brain barrier as it can also be done through the parasympathetic and sympathetic systems. Indeed, immune cells present at their surfaces nicotinic receptors for acetylcholine and β-adrenergic receptors for catecholamine [34]. These receptors link immune cells to parasympathetic and sympathetic systems respectively. Thus, a pharmacologic vagal or noradrenergic stimulation could represent a potential target for neuroprotection. For this purpose, vagal stimulation potentially passing through the modulation of lipocalin prostaglandin D2 synthase (L-PGDS) has shown in rat models with ischemic stroke a neuroprotective effect against ischemia reperfusion [37]. Also, a noradrenergic stimulation has shown, in Parkinson's disease, a neuroprotective effect by inhibition of inflammation [38].

Concerning the modulation of interaction between pro-inflammatory cytokines and their intracerebral target strategy, specific receptor antagonist of IL-1 appears to be the most conclusive therapeutic approach. This antagonist is produced endogenously following brain injury, and its administration by systemic or intracerebral route leads to a reduction in the size of lesions in mouse models [34]. Furthermore, Veltkamp et al*.* report the use, via general route of anakinra, of an antagonist of IL-1 receptors in a clinical trial on a patient having stroke [39]. This clinical trial has shown a great reduction of national institute of health stroke scale (NIHSS), and it also shows more patient with modified Rankin score (mRS) of 0–1 in 3 months [39]. Also based on this axis, sitagliptin, a molecule used in the treatment of type 2 diabetes since the discovery of incretin effect, has shown a great anti-inflammatory capacity. This anti-inflammatory activity of sitagliptin is linked to the inhibition of synthesis of pro-inflammatory cytokine and a raise in anti-inflammatory cytokine synthesis [40]. This property has been exploited in the treatment of Alzheimer's disease in mouse models, and the results were conclusive [40]. In humans, the administration of sitagliptin was associated with an amelioration of the minimental state examination (MMSE) score used to evaluate dementia [40]. All these axes remain focused on more or less advanced stages of inflammation. For this reason, they carry the risk of possibly altering the beneficial effects of inflammation. Thus, to reduce this intrinsic risk, it seems necessary to develop more specific methods to modulate the inflammation. One method could be the inhibition of inflammasomes. However, because of the lack or incomplete knowledge on inflammasome structure and activation, this approach remains difficult. Nevertheless, the inhibition of NLRP3, the most studied inflammasome, has been subjected to several studies in psychiatric disorders [41]. A specific inhibitor of NLRP3 has been developed which lays the foundation for further exploration of this axis [42].
