Neuroimmune interaction in seizures and epilepsy: focusing on monocyte infiltration

Abstract

Epilepsy is a major neurological condition that affects millions of people globally. While a number of interventions have been developed to mitigate this condition, a significant number of patients are refractory to these treatments. Consequently, other avenues of research are needed. One such avenue is modulation of the immune system response to this condition, which has mostly focused on microglia, the resident immune cells of the central nervous system (CNS). However, other immune cells can impact neurological conditions, principally bloodborne monocytes that can infiltrate into brain parenchyma after seizures. As such, this review will first discuss how monocytes can be recruited to the CNS and how they can be distinguished from there immunological cousins, microglia. Then we will explore what is known about the role monocytes have within seizure pathogenesis and epilepsy. Considering how little is known about monocyte function in seizure and epilepsy related pathologies, further studies are warranted that investigate infiltrated bloodborne monocytes as a potential therapeutic target for epilepsy treatment.

1. Introduction:

Epilepsy is a condition affecting upwards of 65 million people worldwide, and is the fourth most common neurological disorder in the United States [1, 2]. A major hallmark of epilepsy is the development of spontaneous seizures due to dysregulation of neuronal activity, synchronization, and excitability [1]. Although a number of medications have been developed to treat epilepsy, up to a third of patients are refractory to treatment [3]. Consequently, exploration of new avenues of epileptic intervention is needed. Epilepsy research has predominantly had a “neurocentric” focus, while the contributions of immunological systems have been largely understudied.

Within the central nervous system (CNS), microglia play the role of the resident immune cell type. Indeed, microglia contribute to the development of a variety of pathologies (e.g., pain, stroke, Alzheimer’s disease), and that modulation of their activity can impact the progression of these conditions [4–8]. In regards to epilepsy, microglia become readily activated taking on a larger, less ramified, amoeboid morphology. The pathological consequences of this activation include promotion of inflammation (i.e., expression of pro-inflammatory cytokines), regulation of neuronal activity, phagocytic clearance of neurons, and promotion of chronic-seizures [1, 9–11]. Moreover, it has been suggested that non-inflammatory alterations to microglia are sufficient to induce epilepsy [12]. However, microglia are not the only immune cell type that can influence epilepsy. Monocytes, an immunological relative to microglia, have also been suggested to affect a number of neurological disorders, including pain, stroke, amyotrophic lateral sclerosis (ALS), and epilepsy [13–17]. Monocytes are bloodborne myeloid-progenitor-derived cells that can respond rapidly to infected and damaged tissues. Monocyte infiltration also contributes significantly to the apparent microgliosis following status epilepticus (SE) [18]. However, the contributions of infiltrating monocytes and subsequent macrophages to neurological disorders, including epilepsy, have often been poorly defined. As such, this review seeks to explore the current literature regarding the role of monocytes within epileptic pathogenesis and promote novel directions for research into epileptic intervention that targets infiltrating monocytes.

2. Distinguishing Infiltrating Monocytes from Resident Microglia:

One key consideration when discussion monocyte activity within the CNS is how to distinguish them from resident microglia under pathological conditions. It is currently hypothesized that during early development, primitive macrophages contained within the yolk sac infiltrate into the nascent CNS prior to blood-blain-barrier (BBB) closure and give rise to microglia [19, 20]. Microglia are then thought to maintain their numbers via self-replication, albeit slowly [21, 22]. However, it has been postulated monocytes can also differentiate into “microglia-like” cells, which would make it difficult to distinguish these infiltrated “microglia-like” cells from true resident microglia [19, 23] (Figure 1). Nevertheless, this remains a controversial hypothesis, as many argue that monocytes can still be distinguished for microglia and are not a source of microglia in the adult CNS [18, 22, 24–26]. Further confounding this controversy, recent microglial ablation studies have argued both for and against monocytes as a source of microglial repopulation [27–29]. Thus, to fully characterize the role of monocytes within CNS pathologies, great care is needed when distinguishing between resident microglia and infiltrated monocytes. The following are some of the ways in which these types of investigations have been conducted.

Figure 1: Comparison of monocyte and macrophage characteristics.

One of the key difficulties in delineating the function of monocytes within any neurological condition is separating them from resident microglia. Differentiated macrophages, which arise from infiltrated monocytes, display similar morphologies and phenotypes as activated microglia. However, monocyte and microglia specific proteins may still be able to separate these populations. Finally, the controversial hypothesis that infiltrating monocytes can also differentiate into resident microglia within the adult CNS should be considered. Abbreviations: CCL2, chemokine (C-C motif) ligand 2; CCR2, C-C chemokine receptor type 2; CNS, central nervous system; CX3CR1, C-X3-C motif chemokine receptor 1; IBA1, ionized calcium binding adaptor molecule 1; TMEM119, transmembrane protein 119.

Some experiments have utilized bone-marrow chimeras to generate and trace labeled monocytes infiltrating into the CNS [23, 30, 31]. However, it is now accepted that utilization of lethal irradiation, which is used in the generation of these chimeric animals, is damaging to the BBB [32–34]. Therefore, the degree to which monocytes infiltrate into the CNS and the level of inflammation described in these models should be considered. However, other methods are being developed that can be used to distinguish infiltrating monocytes from resident microglia, without damaging radiation.

Flow cytometry remains a relatively popular method to investigate monocyte populations. Typical monocytes in humans can be identified by their expression of, CD14, CD16, CD64, and C-C chemokine receptor type 2 (CCR2; CD192) [35, 36]. For mice, the expression of Ly6C, CD43, CD11b, and CCR2 is used to help identify monocyte populations [37–39]. However, microglia have also been shown to express CD11b, CD14, CD16, and CD64 [40, 41]. Consequently, in disease contexts, like epilepsy, that requires separation of these two populations, more specific markers are needed. Traditionally, when microglia and monocytes need to be separated, relative expression of CD45 has often been employed [39]. However, with the recent advances in genetic profiling, markers specific to microglia have been identified (e.g., P2Y12, Transmembrane protein 119 (TMEM119), Hexosaminidase B (HexB), Olfactomedin-like protein 3 (Olfml3)) [42]. These markers are now being incorporated into fluorescence-activated cell sorting (FACS) activities [39]. Conversely, CCR2 has been shown to lacking on microglia and could also be used to separate these two cell types [43] (Figure 1).

While separation of microglia and monocytes by relative protein expression is possible with flow cytometry, direct visualization methods require more specific means. Under normal brain conditions, the small number of monocytes residing in blood vessels typically take on a small round morphology with little to no processes, while microglia are highly ramified with numerous long processes. However, under pathological conditions these differences in morphology fade away as infiltrated monocytes differentiate into macrophages, and microglia soma size increases and processes shorten [44] (Figure 1). Nevertheless, other means of investigation can be used to distinguish monocytes/macrophages from microglia, such as serial block-face scanning electron microscopy (SBF-SEM) [45]. It has also recently shown that electrophysiology approaches can be used to separate microglia and monocyte. Specifically, following SE induction activated microglia were observed to have both inward and outward potassium currents, while infiltrated monocytes displayed principally inwards currents [18].

However, genetic modulation may offer a more substantive means to tag and investigate monocytes and microglia independently. For example, CX3CR1^CreER/+:R26^tdTomato/+ (Ai14) mice have previously been used to separate microglia from monocytes [13, 46, 47]. By taking advantage of the fact that monocytes have a rapid turn-over rate (>90% after 3 weeks) when compared to microglia (<10% after 1 year), microglia could then be specifically labeled with tdTomato [13, 22, 47]. Then staining for a common marker, like ionized calcium binding adaptor molecule 1 (IBA1) could distinguish monocytes from microglia. Recently, more straightforward labeling methods have been developed. The use of CCR2-RFP mice has become a popular means to identify monocytes in vivo in a variety of contexts [43, 48–51]. This genetic labeling method has also been used to investigate monocytes within epilepsy [14, 15, 18]. For instance, when coupled to 2-photon imaging, CCR2-RFP labeling allowed for the identification of newly infiltrated monocytes following SE that displayed typical round morphologies [18]. It should be noted however, that CCR2 is suggested to be expressed by a number of cell types in the CNS (e.g., astrocytes, microglia) [52–54]. However, CCR2-RFP labeling seems to be specific to monocytes when investigating CNS pathologies [55].

Alternatively, microglia specific markers can be harnessed to separate monocytes from microglia. For instance, the microglial marker TMEM119 has recently been used to generate microglia specific transgenic reporter mice [56–58]. Similarly, P2Y12 has also been shown to distinguish resident microglia from infiltrated monocytes [25, 59]. As such, this marker is also being targeted for the development of genetic tools that can be utilize for the labeling and separation of microglia from monocytes [60]. However, it should be noted that under certain pathological conditions, microglia specific markers (i.e., P2Y12) can be down-regulated following activation [61, 62]. These markers, including TMEM119 and P2Y12, are also rapidly lost by microglia in culture [63]. While in vitro expression levels may not be of issue, for in vivo studies the observation that expression of these markers is not entirely consistent should be considered when designing experiments.

3. Epilepsy, Blood-Brain-Barrier Disruption, and Monocyte Infiltration:

While many studies do not attempt to separate microglia from infiltrating monocytes and subsequent macrophages, recent studies has shown that monocyte infiltration and microglial proliferation both contribute to observed hippocampal microgliosis following kainic acid (KA)-induced seizures [18]. Consequently, since it is possible that both microglia and monocytes contribute to seizure and epilepsy related pathologies, it is important to first understand how monocytes gain entry into the CNS. A key contributor to the immune privilege of the CNS is the blood-brain-barrier (BBB) [64]. The BBB is a selectively permeable boundary composed of brain microvascular endothelial cells (BMECs) supported by pericytes and the end-feet of astrocytes [65–67]. Generally, the BBB will only allow the passage of water, glucose, and other essential nutrients while excluding most other serum factors [68]. Moreover, under normal physiological conditions the BBB limits the passage of peripheral immune cells into the CNS [69]. However, under pathological condition the BBB can become compromised allowing previously excluded serum components and immune cells into the CNS inducing numerous adverse effects (Figure 2).

Figure 2: Blood-brain-barrier under physiological and pathological conditions.

The BBB is an integral part of healthy brain function. Under typical conditions the BBB is composed of tightly associated BMECs, supported by basement membrane, pericytes, and astrocyte endfeet. It will strictly limit passage into the greater brain to only select factors. However, under pathological conditions this tight barrier can become permeable allowing for previously excluded factors into the brain. Moreover, damage to the BBB can facilitate release of chemoattractants such as CCL2 into the bloodstream leading to the recruitment of monocytes to the CNS. Abbreviations: BBB, blood-brain-barrier; BMEC, brain microvascular endothelial cell; CCL2, chemokine (C-C motif) ligand 2; CNS, central nervous system.

The most notable example of BBB disruption is in stroke, particularly ischemic stroke. Specifically, endothelial tight junction degradation, breakdown of the underlying basement membrane, and modulation of cellular transport mechanism have been observed within the first few hours following ischemic stroke [70, 71]. In regards to epilepsy, there is significant evidence indicating that during the acute phase there is significant opening of the BBB similar to that observed during stroke [72]. This has been seen in humans and both infection-related (e.g., viral encephalitis) and “sterile” (e.g., pilocarpine, KA) animal models of epilepsy [15, 73, 74]. Moreover, studies have also shown that BBB disruption can directly induce seizure activity and exacerbate epileptogenesis [75–78].

However, while under normal circumstances the BBB regulates monocyte entry into then CNS, BBB disruption does not necessitate immediate infiltration of monocytes into the CNS. For example, following SE induction via intraperitoneal (i.p.) KA monocyte infiltration was not observed till 24 hours later [14, 15]. Our own preliminary investigations also observed that monocyte infiltration is delayed into the hippocampus following seizure induced BBB disruption, even though BBB permeability is rapidly disrupted (<6 hours). This can also be seen in other types of injury, such as traumatic brain injury (TBI), where neutrophils are observed to rapidly infiltrate into areas where the BBB is disrupted, while monocytes/macrophages only begin to appear 24 hours later [79]. Furthermore, MRI studies of experimental allergic encephalomyelitis (EAE) demonstrated that BBB disruption and monocyte infiltration are two distinct events occurring days apart [80]. While BBB disruption does have notable pathological consequences (e.g., leakage of plasma proteins, ion imbalance, neuronal dysfunction), it alone may not result in significant monocyte infiltration [79–81]. However, BBB disruption could facilitate the release of chemoattractants into circulation. Thus, it is important to discuss the possible factors that may facilitate monocyte attraction into seizure-injured brain regions.

4. Importance of the CCL2-CCR2 Axis to Monocyte Recruitment:

It is becoming increasingly apparent that the chemokine (C-C motif) ligand 2 (CCL2)-CCR2 axis is of particular importance to the activities of monocytes within the CNS and elsewhere. Expression of CCL2 seems to be related to a number of neuropathologies including cerebral ischemia, Alzheimer’s disease, and traumatic brain injury [82–84]. Furthermore, CCL2 is highly up-regulated in patients suffering from human immunodeficiency virus (HIV) associated neurocognitive disorders, which facilitates the attraction of monocytes into the brain [85, 86]. In regards to seizures and epilepsy, several studies have shown that CCL2 is extremely up-regulated, as much as 200 fold, in the brain following SE and in epileptic patients [14, 15, 87–89]. It has also been shown that cerebral CCL2 expression is elevated in patient’s suffering from intractable epilepsy [90–92]. Furthermore, systematic administration of the CCL2 transcription inhibitor, bindarit, or CCR2 antagonist, RS102895, suppressed the seizure exacerbating effects of lipopolysaccharide (LPS) [87]. A number of sources of CNS CCL2 have been described in the literature, including astrocytes, microglia, and endothelial cells. However, considering that CCL2 released from damaged neurons has been shown to be principal to the development and progression of neuropathic pain and encephalopathy, it is likely also the driver of the apparent increase following seizure. Indeed, it was found that CCL2 expression was mostly upregulated in microglia and neurons following SE [14] (Figure 3A).

Figure 3: CCL2 drives monocytes infiltration and activation after seizures.

(A) It has been shown that following seizures, neurons and microglia will secrete high levels of CCL2. This chemokine may then pass into circulation, attracting monocytes into the CNS via CCR2 signaling. (B) It has also been demonstrated that CCL2 may induce monocyte-derived macrophages to secrete the inflammatory cytokine IL-1β via STAT3 signaling mechanisms. IL-1β could then contribute to neuroinflammation and seizure related pathology. Abbreviations: CCL2, chemokine (C-C motif) ligand 2; CCR2, C-C chemokine receptor type 2; CNS, central nervous system; IL-1β, Interleukin 1 beta; STAT3, signal transducer and activator of transcription 3.

As a chemoattractant, CCL2 expression significantly influences monocyte recruitment via CCR2 signaling [93]. So much so that an alternative name for CCL2 is monocyte chemoattractant protein 1 (MCP-1). Additionally, CCR2⁺ monocytes are typically considered to be inflammatory [94]. Studies using CCR2 KO mice have shown stunted infiltration of monocytes following ischemia, traumatic brain injury, and experimental autoimmune encephalomyelitis [74, 95–97]. It has also been shown that neuronal expression of CCL2 following Theiler’s Murine Encephalomyelitis Virus (TMEV) infections will direct monocytes to infiltrate into the CNS [98]. Further demonstrating the importance of CCL2-CCR2, when CCR2 was knocked-out there was a significant reduction in monocytes infiltration into the hippocampus following kainic-acid-induced SE [14, 15]. Interestingly, one report showed that in La Crosse Virus encephalitis entry of Ly6C^hi monocytes into the CNS was CCR2 independent, while still being necessary for their movement within the CNS [99].

In addition to providing a chemoattractant signal for monocytes, CCL2 may also support monocyte recruitment by increasing the permeability of the BBB. Several studies have shown that direct injection of CCL2 into the brain will result in BBB disruption and that this effect is plasmin-dependent [100–104]. Studies investigation the plasmin-dependence of CCL2-mediated BBB disruption also observed increased monocyte infiltration that correlated with increased plasmin activity [105, 106]. However, CCL2-mediate BBB disruption also seems to be CCR2-dependent. For instance, CCL2 biding to CCR2 receptors located on BMECs will disrupt the integrity of BBB tight junctions [107]. It has also demonstrated that CCR2 deficiency will reduced BBB disruption due to both seizures and stroke [15, 108]. There is also indication that monocytes will participate in CCL2 mediated BBB disruption, likely because of CCR2 [103]. Consequently, it is possible that following an initial neuropathological event, like a seizure, BBB disruption will allow for release of upregulated CCL2, thereby attracting monocytes, which in turn can further exacerbate BBB permeability.

However, it is unclear if sustained CCL2-CCR2 signaling is necessary for monocyte activities once they are recruited to the CNS. Following SE some infiltrated CCR2⁺ monocytes within the hippocampus were observed to also express C-X3-C motif chemokine receptor 1 (CX3CR1), and display morphological alterations (i.e., development of processes) [18]. It has also been shown that CCR2⁺ monocytes are only a portion of the total circulating monocyte population [109–111]. Moreover, inflammation related CCR2⁺ monocytes typically express low levels of CX3CR1, while non-inflammatory monocytes express low CCR2 but high CX3CR1 [111]. This dual CCR2⁺:CX3CR1⁺ population may represent an intermediate cell type that is transitioning away from a pro-inflammatory phenotype. This is also suggested to occur with infiltrated monocytes responding to ischemic stroke [112]. Nevertheless, even if monocytes can transition after arriving into the CNS, the initially responding CCR2⁺ monocytes would still be considered inflammatory. Thus, it is important to consider the role monocytes may have in promoting inflammation within epilepsy.

5. CNS Inflammation and Epilepsy:

CNS inflammation is now recognized as a key mediator of SE and epilepsy pathogenesis [1, 113–115]. It has also been suggested that neuroinflammatory pathways and factors are important targets for therapeutic intervention [116]. For instance, reducing oxidative stress and reactive oxygen species levels can improve epilepsy pathology [117, 118]. The effects of inflammation can be further magnified if there is also a pre-conditioning source (e.g., LPS stimulation) [119, 120]. Extrinsic inflammatory conditions such as obesity have also been linked to more sever seizure effects [121].

Additionally, a number of immunological cell types have been related to seizure and epilepsy induced inflammation. Several clinical reports have shown increased neutrophil levels within epileptic patients, which is a noted indicator of inflammation [122, 123]. Moreover, pro-inflammatory CD4⁺ and CD8⁺ T cells have been shown to significantly infiltrate into both the neocortex and hippocampus of mice shortly after seizure (<24h) and within the hippocampus of epilepsy patients [124, 125]. Production of pro-inflammatory IL-17 by these cells may also contribute to refractory pediatric epilepsy [126]. Interestingly though, it has also been observed that within recombination activating gene 1 (RAG1) KO mice, which lack T and B cells, there was significantly more neurodegeneration and an earlier onset of spontaneous recurrent seizures following KA treatment [127]. Thus, there are clear implications for innate immune response in seizure and epilepsy pathology.

While infiltrating monocytes have also been suggested to promote inflammation after SE, relatively little is known about the mechanisms by which monocytes can promote inflammation under epileptic conditions [15]. Besides CCL2, several other pro-inflammatory cytokines have been linked to epileptogenesis such as interleukin 6 (Il-6), IL-1β, and tumor necrosis factor alpha (TNF-α) [128–130]. Monocytes have been shown to express IL-6, while IL-1β and TNF-α will affect monocyte recruitment and function [14, 131–133]. Moreover, SE-induced expression of pro-inflammatory IL-1β was almost completely abolished within CCR2 KO mice [14]. It was then determined that IL-1β was being produced by both microglia and monocyte-derived macrophages in a signal transducer and activator of transcription 3 (STAT3) dependent manner (Figure 3B). With CCL2-CCR2 being possibly linked to neuroinflammation after infiltration, it further highlights the question as to the role of CCR2 dependent mechanisms beyond recruitment.

6. Effect of Monocytes upon Status Epilepticus and Epilepsy:

While there is evidence that the immune system has a role in aggregate epileptogenesis, there are several causes of epilepsy that are directly related to immunological effectors. Notably, epilepsy can arises due to viral encephalitis (VE), a viral infection that produces inflammation within the brain parenchyma. VE has been shown to induce seizures during the acute infection period, and later unprovoked epileptic events following infection resolution. In regards to monocytes there is clear evidence demonstrating a significant CCL2-dependent infiltration of monocyte into the CNS following viral infections, specifically into the hippocampus [98]. Studies utilizing CCR2 KO mice showed fewer activated myeloid cells, ostensibly monocytes, correlated with reduced hippocampal damage, but not seizure presentation [46]. Additionally, immunodepletion and adaptive bone marrow transfer experiments demonstrated a potential role for inflammatory monocytes in hippocampal damage [134, 135].

However, other studies involving depletion of macrophages, which are derived from monocytes, reduced the number of acute seizures and epileptic events due to VE, but not hippocampal damage [136–138]. It has also been suggested that this effect on seizures is dependent upon macrophage IL-6 expression [136, 139]. Furthermore, direct injection of monocytes into the brain seemed to exacerbate the seizure burden of TMEV infected mice [138, 140]. Consequently, it would seem the role monocytes play is still inconclusive. It is possible that alternative monocyte-derived populations, like differentiated macrophages, may affect seizure severity and hippocampal damage in different ways. Another possible explanation may be that microglia contribute to hippocampal damage, and that this activity is somehow affected by CCR2 KO, while infiltrating monocytes can affect seizure burden in a CCR2 independent fashion. Regardless, more research is needed to conclusively determine the role of monocytes in VE-induced seizures.

Viral infection is not the only means by which epilepsy and seizures can occur, consequently investigations into the role of monocytes in “sterile” models of SE and epilepsy are also important. Most, non-infection related models of SE and epilepsy utilize either kainic acid (KA) or pilocarpine to induce seizures and have often been correlated with temporal lobe epilepsy (TLB) [141–143]. As mentioned previously, CCR2⁺ monocyte infiltration has been suggested to promote inflammation following KA-induced SE, which in turn was correlated with neuronal damage but not seizure severity [14, 15]. Further, studies utilizing CCR2 KO mice revealed similar results indicating that monocyte infiltration exacerbates the pathological effects of KA-induced SE [14, 15]. However, it has also been shown that i.v. transplantation of bone marrow mononuclear cells (BMMCs) into lithium–pilocarpine-treated rats reduced seizure frequency and improved spatial memory deficits [144]. It should be noted though that the identity of these BMMCs was not determined, so this effect may not be due specifically to monocytes. Moreover, if these cells did contain monocytes, they may present alternative anti-inflammatory phenotypes not typically associated with monocytes that naturally respond to CNS pathologies (i.e., inflammatory CCR2⁺ monocytes).

It is also possible that differences in the pathological mechanisms governing the various models of epileptogenesis could result in the observed discrepancies in monocyte-related activity. When blood monocytes from patients suffering from Neurocysticercosis (NCC), a disease related to Taenia solium (T. solium) infection and a leading cause of epilepsy in the developing world, were compared to those derived from patients with idiopathic epilepsy via microarray analysis, it was revealed that NCC-related monocytes had higher association with inflammatory/immune pathways than idiopathic epilepsy patients [145]. It has also been suggested that NCC will induce inflammatory monocyte-astrocyte crosstalk [146]. Consequently, when determining the role monocytes have within epilepsy it is important to not only consider their relation to CNS factors, but also the root cause of the condition.

Finally, seizure and epileptic conditions have been liked to a number of co-morbidities including anxiety, depression, and memory deficits [147, 148]. Infiltration of monocytes into the CNS has also been shown to affect these behavioral and memory effects in other conditions such as prolonged stress and Alzheimer’s disease [149–151]. Furthermore, stress-induced anxiety may be related to recruitment of IL-1β producing monocytes [152]. In regards to epilepsy, when CCR2 KO mice were evaluated for anxiety and memory following KA-induced SE, it was found that CCR2 KO mice had better performance in these evaluations than did wild-type mice. These improvements were also correlated with reduced monocyte infiltration and microglial activation [14]. Consequently, while more work needs to be done to evaluate the contribution of monocytes to epilepsy-related co-morbidities, there are indications that infiltrating monocytes can modulate a wide range of effects (Table 1).

Table 1:

Selected references and key findings on monocytes in seizures and epilepsy.

Model	Administration	Species	Key Findings	Reference
Clinical	N/a	Humans	Monocytes derived from patients with neurocysticercosis-associated epilepsy displayed significantly different gene expression profiles than monocytes derived from idiopathic epilepsy patients.	[145]
Kainic Acid	Intracerebroventricle (ICV)	Mice	CCL2-CCR2 mediates seizure-induced monocyte infiltration. STAT3 facilitate CCL2 induced production of IL-1β by monocyte-derived macrophages and microglia. CCR2 KO reduced seizure-induced neurodegeneration and behavioral impairments.	[14]
Kainic Acid	ICV	Mice	Both monocyte infiltration and microglia proliferation contribute to seizure-induced microgliosis.	[18]
Kainic Acid	ICV	Rats	2-Deoxyglucose treatment altered the expression pattern of monocyte-derived macrophages, reducing seizure-induced hippocampal damage.	[155]
Kainic Acid, Pilocarpine	Intraperitoneal (IP)	Mice	CCR2 KO reduced seizure-induced monocyte infiltration and IL-1β expression into the hippocampus. Reduced monocyte infiltration correlated with reductions in seizure-induced BBB disruption and neurodegeneration.	[15]
Pilocarpine	IP	Rats	Injection of BMMCs reduced seizure burden and improved memory deficits.	[144]
Pilocarpine	IP	Rats	Roscovitine treatment reduced seizure-induced monocyte infiltration.	[154]
Viral Encephalitis	Theiler’s murine encephalomyelitis virus (TMEV) Daniels (DA) strain Parietal cortex injection	Mice	Microglia post-infection had higher expression of TNF-α then infiltrating macrophages, while infiltrating macrophages had higher expression IL-6 than microglia. Treatment with either minocycline or wogonin reduced the number of infiltrating macrophages and the number of seizures post-infection.	[136]
Viral Encephalitis	TMEV DA strain Intracerebral (IC)	Mice	Systemic clodronate liposome treatment, which ablates circulating monocytes and macrophages, reduced seizure burden post-infection, but not hippocampal damage.	[137]
Viral Encephalitis	TMEV DA strain Intracerebral (IC)	Mice	RAG1 KO did not affect seizure burden post-infection. Depletion of macrophages improved seizure burden post-infection.	[138]
Viral Encephalitis	TMEV DA strain Intracerebral (IC)	Mice	IL-6 produced by microglia and infiltrating macrophages may have a role in VE-induced seizures. Minocycline treatment reduced seizure burden post-infection.	[139]
Viral Encephalitis	TMEV DA strain Intracerebral (IC)	Mice	Both CCR2 and CX3CR1 affect myeloid cell accumulation due to infection. KO of either CCR2 or CX3CR1 reduced hippocampal damage, but not seizure development.	[46]

7. Interplay between Monocytes and Microglia within Seizure and Epilepsy Pathology:

Finally, it is important to note the possibility of combinatory and compensatory effects of monocytes and microglia within the context of seizure and epilepsy pathologies. As previously noted, monocyte recruitment and microglial proliferation both contribute to the apparent microgliosis following seizure [18]. This may result in a possibly compensatory effect when either microglia or monocytes are targeted by an intervention. Moreover, it has been suggested that microglia can act as a source of CCL2, which may facilitate monocyte recruitment [153]. This could mean that activation of microglia could result in further infiltration of monocytes and microgliosis. Yet, the degree to which either cell type contributes to seizure and epilepsy pathology remains largely unknown. Our own work indicates however, that both monocyte-derived macrophages and microglia may contribute to pathological IL-1β production after seizure [14]. Consequently, future explorations into the role of monocytes and microglia should consider how each cell type contributes to the observations. With the novel molecular tools currently being developed these investigations should become easier to perform.

8. Conclusion and Future Directions:

In summary, lines of evidence have shown that monocytes play a substantive role in pathogenesis of seizures and epilepsy. Now that new tools are being developed that can separate monocyte populations from resident microglia, there is significant opportunity in trying to investigate and target these cells for therapeutic purposes. For instance several of the previously cited studies have shown modulation of the CCL2-CCR2 axis will reduce monocyte infiltration and improve seizure-related outcomes. Additionally, Roscovitine, a cyclin-dependent kinase 5 (CDK5) inhibitor, was shown to mitigate monocyte infiltration and suppress SE-induced neuroinflammation [154]. Another possible avenue is directing infiltrated monocytes towards anti-inflammatory and restorative activities. It has been recently reported that 2-Deoxyglucose treatment will shift monocytes towards a restorative macrophage phenotype following KA induced seizures, which resulted in reduced hippocampal neuron loss [155]. However, several questions still remain. For example, the exact role of monocytes in epilepsy is still relatively unclear as there are conflicting reports depending on the animal model of epilepsy. It also remains unclear whether monocytes that do not express high levels of CCR2 have a role in epileptogenesis. Nevertheless, infiltrating monocytes offer novel possible research directions for the treatment of seizures and epilepsy.

Abbreviations:

ALS

amyotrophic lateral sclerosis

BBB

blood-brain-barrier

BMEC

brain microvascular endothelial cell

BMMC

bone marrow mononuclear cells

CCL2

chemokine (C-C motif) ligand 2

CCR2

C-C chemokine receptor type 2

CCR2-RFP

CCR2 knock-in red fluorescent protein

CNS

central nervous system

CX3CR1

C-X3-C motif chemokine receptor 1

FACS

fluorescence-activated cell sorting

HexB

hexosaminidase B

HIV

human immunodeficiency virus

IBA1

ionized calcium binding adaptor molecule 1

IL-1β

Interleukin 1 beta

IL-6

Interleukin 6

KA

kainic acid

LPS

lipopolysaccharide

NCC

neurocysticercosis

Olfml3

olfactomedin-like protein 3

RAG1

recombination activating gene 1

SBF-SEM

serial block-face scanning electron microscopy

SE

status epilepticus

STAT3

signal transducer and activator of transcription 3

TBI

traumatic brain injury

TMEM119

transmembrane protein 119

TMEV

Theiler’s murine encephalomyelitis virus

TNF-α

tumor necrosis factor alpha

VE

viral encephalitis