Differential Activation of Neuroinflammatory Pathways in Children with Seizures: A Cross-Sectional Study

Background: Experimental and clinical findings suggest a crucial role of inflammation in epileptogenesis. We aimed to analyze levels of inflammatory cytokines in plasma and saliva from children with acute seizures and healthy controls and measure their associations with HHV6 and EBV infection. Methods: We analyzed plasma from 36 children within 24 hours of acute seizures (cases) and 43 healthy controls and saliva from 44 cases and 44 controls with a multiplex immunoassay. Saliva from all controls and 65 cases and blood from 26 controls and 35 cases were also analyzed by ddPCR for viral DNA. Statistical analysis included Wilcoxon Rank Sum test, Fisher’s exact test, ANOVA and Spearman correlation. Results: Compared to controls, children with breakthrough seizures (n=18) had higher levels of CCL11 (p<0.001), CCL26 (p<0.001), IL-8 (p=0.03), CCL4 (p=0.02) in plasma. Children with new onset seizures (n=13) showed higher levels of CCL11 (p=0.05) and IL-6 (p=0.01). Patients with febrile seizures (n=5) had higher levels of IFNg (p<0.001), IL-6 (p<0.001), IL-10 (p<0.001), CXCL10 (p=0.001). CCL11 was higher with 3 or more seizures (p=0.01), seizures longer than 10 minutes (p=0.001) and when EEG showed focal slowing (p=0.02). In saliva, febrile seizures had higher levels of IL-1β (n=7, p=0.04) and new onset seizures had higher IL-6 (n=15, p=0.02). Plasma and saliva cytokine levels did not show correlation. Frequency of HHV-6 and EBV detection was similar across seizure types and not different than controls. We found no correlation between viral load and cytokine levels Conclusions: We showed differential activation of neuroinflammatory pathways in plasma from different seizure etiologies compared to controls, unrelated to HHV-6 infection.


Background
Approximately 15 million people worldwide are affected by pharmaco-resistant epilepsy and experience seizures despite complex therapeutic regimens, often burdened by significant side-effects. Current antiepileptic drugs target seizures symptomatically but not underlying pathophysiological mechanisms. 1 Experimental and clinical findings suggest a crucial role of inflammation in epileptogenesis. 2 New therapeutic strategies are necessary to improve seizure control and quality of life for people with epilepsy. The first step in developing novel therapies is improving understanding of pathophysiological mechanisms of epileptogenesis.
An emerging hypothesis is that various brain insults, including viral infections, can contribute to epileptogenesis by inducing a cascade of chronic central nervous system (CNS) inflammatory processes and increased blood-brain barrier permeability, leading to enhanced neuronal excitability. 3,4 Previous studies highlighted the role of glial cells (astrocytes and microglia) and neurons in production of inflammatory cytokines. 5 Preclinical studies showed that interleukin (IL)1β overproduction could enhance susceptibility to adult seizures with hippocampal neuronal injury, and might contribute to development of temporal lobe epilepsy and hippocampal sclerosis. 6 Elevated levels of IL1β, IL6, and IL8 have been linked to neuronal hyperexcitability 2 and were demonstrated in serum and cerebrospinal fluid (CSF) of patients with epilepsy. Etiology-specific meta-analyses revealed elevated IL6 levels in temporal lobe epilepsy patients. 7 IL6, produced by resident cells of the CNS, contributes to development of seizures following viral infections. 8 In vitro stimulation studies showed increased production of IL1β, IL6 and IL10 by peripheral blood mononuclear cell (PBMCs) in epilepsy and febrile seizure patients. 9 In resected epileptogenic tissue, hippocampal C-C Motif Chemokine Ligand (CCL)11 levels are higher than those in entorhinal and temporal cortices. 10 There is growing evidence that neurotropic viral infections, such as those from herpesviruses, play an important role in the pathogenesis of seizures. Among herpesviruses, human herpesvirus (HHV)-6 occupies a prominent place, as many studies have linked it to seizures and epilepsy. 11 Primary infection with HHV-6B occurs in almost 90% of children by age two years. 12 HHV-6 DNA is present in children with febrile status epilepticus 13 and can be detected in resected epileptogenic tissue from temporal lobe epilepsy patients. [14][15][16] Pathological analyses revealed that HHV-6 can infect astrocytes and oligodendrocytes leading to upregulation of several proinflammatory cytokines. Host gene expression analysis of brain samples of patients with mesial temporal lobe epilepsy (MTLE) revealed expression of monocyte chemoattractant protein (MCP)1 and glial fibrillary acidic protein (GFAP) in the HHV-6-positive versus HHV-6-negative amygdala tissues, with positive correlation with viral load, suggesting an inflammatory process triggered by the virus. 17 HHV-6 has been reported to induce IL6 both in vitro 18,19 and in vivo 20 and to up-regulate the production of CCL2 and IL8. 21,22 IL6-mediated cellular and humoral immune responses play a crucial role in determining the outcome of viral infection. 23 In a preliminary feasibility study, 24 we showed higher levels of IL8 and IL1β in saliva from 32 children with seizures compared to 30 age-matched controls with a febrile illness and no seizures.
With the present study, we sought to expand the initial data and aimed to investigate the inflammatory response in blood and saliva for different types of seizures in children and elucidate a potential role of HHV-6 infection in triggering activation of a different neuroinflammatory pathway. This is meant as a first step towards the identification of novel therapeutic targets. Medical records were reviewed for laboratory, imaging and electroencephalogram (EEG) results, if available. Study data were collected and managed using a password-protected database (REDCap electronic data capture tools).

Enrollment
Written informed consent was obtained from a parent or legal guardian and written assent from the child, when indicated. Children's National Medical Center Institutional Review Board approved the study.
Sample collection, processing and analysis: For each participant, simultaneous saliva and blood samples were obtained. All samples were collected, handled and processed following standard biosafety procedures. Saliva was collected utilizing a validated pediatric swab (SalivaBio Oral Swab, Salimetrics). Whole blood was collected in EDTA tubes via a venous puncture. Samples were then centrifuged at 2,300 g for 10 minutes immediately after collection and saliva, plasma, and whole blood were aliquoted. Samples were then then frozen at -80 C immediately after processing and shipped on ice to the Viral Immunology Section of the National Institutes of Health for analysis. After thawing the samples, DNA was extracted utilizing a commercially available kit (DNeasy Blood & Tissue Kit, Qiagen) following the manufacturer's protocol for plasma and a previously validated protocol for saliva. 24 HHV-6 and Epstein-Barr Virus (EBV) viral DNA in saliva and whole blood was quantified using digital droplet PCR (ddPCR). Primers from the highly conserved region u57 (HHV-6) and bamHI (EBV) were selected. Different probes were used to distinguish between HHV-6A and HHV-6B. Ribonuclease P Subunit P30 (RPP30) was used as a cellular housekeeping gene.

Statistical analysis and study outcomes
The primary outcome was cytokine levels in cases vs. controls. Secondary outcomes included frequency of detection of HHV-6 and EBV viral DNA in cases vs. controls and viral loads in cases vs. controls.

Statistical analysis was conducted utilizing R version 3.5.3 and included Pearson
Chi-squared test, Fisher's exact test for relative frequencies for HHV-6 detection, Wilcoxon rank-sum test for cytokine analysis, one-way analysis of variance (ANOVA) on ranks and Spearman's correlation for correlations between cytokine levels and clinical variables and HHV-6 viral load and clinical variables.
A p value < 0.05 was considered significant.

Cytokine analysis
Clinical characteristics and cytokine levels are summarized in Table 1 (plasma) and  In saliva, we observed higher levels of IL1b in febrile seizures (n=7, p=0.04) and IL-6 in new onset seizures (n=15, p=0.02) (Supplemental Figure 3). Cytokine levels in plasma and saliva were not associated with the height of fever. We did not observe a correlation between plasma and saliva cytokine levels (data not shown).

Viral droplet digital PCR
Frequency of HHV-6 and EBV detection was similar across seizure types and not different from controls (Supplemental Table 1). We found no correlation between viral load and cytokine levels.

Discussion
In our cross-sectional study we analyzed blood and saliva samples from children with different types of acute seizures to investigate the levels of inflammatory cytokines and the presence of HHV-6 and EBV viral DNA. We showed differential activation of inflammatory pathways in plasma from different seizure etiologies vs. it is reasonable to conclude that CCL11 may have a proepileptogenic effect. This hypothesis has been corroborated by studies that have shown that in resected epileptogenic tissue, hippocampal CCL11 levels are higher than those in the entorhinal and temporal cortices. 38 In our study, we observed higher levels of CCL11, and of the other Eotaxin family member CCL26, in children with chronic epilepsy and breakthrough seizures and to a lesser extent in new onset of seizures.
This finding may suggest that Eotaxins could be an early biomarker of epileptogenesis in children with seizures and warrant further studies. Interestingly, we also observed that CCL11 levels positively correlated with clinical variables of severity such as seizure duration and number of seizures and with focal slowing on EEG, which is a common finding in the context of focal epilepsy. Other EEG features, such as generalized slowing and epileptiform discharges had no such correlation, possibly indicating that CCL11 may be a biomarker of localized brain dysfunction rather than diffuse or excitatory processes.
We also observed elevation in other cytokines such as IL8 in blood from children with chronic epilepsy. Several studies have shown that this cytokine is increased after seizures, including focal, generalized tonic-clonic, myoclonic, atypical absence, and typical absence seizures in serum and CSF of patients with epilepsy. 39-41 In a previous study 24 found higher levels of this cytokine in saliva from children with epilepsy but this time we could not confirm this finding in the same biological compartment.
The only cytokine in our current study that was consistently elevated in both plasma and saliva in children with new onset of seizures was IL6. In previous studies, this mediator is increased within 24 hours after generalized tonic-clonic seizures and febrile seizures but is not changed after seizures in patients with chronic focal epilepsy. 42 At 6 hours after focal unaware or secondary generalized tonic-clonic seizures in patients with MTLE or extratemporal epilepsies, only the MTLE group showed a significant rise in plasma levels of IL6. 43 Saliva may represent a less invasive and less expensive method for quantification of this biomarker and further studies are needed to validate this finding. While prior evidence suggests that salivary components may originate from the salivary glands or may be derived from the blood by passive diffusion or active transport, 44 studies reveal mixed results when comparing blood and salivary cytokines both in physiologic and pathologic conditions. Some reveal no cross-talk between the two compartments, 45  The main strength of our study is that we examined a population of young children with seizures and analyzed samples from different biological compartments for presence of viral DNA from common viruses and at the same time we studied the levels of a pool of cytokines that are associated with neuroinflammation. We also included several clinical variables in our analysis and compared the results with simultaneous age-balanced healthy controls. The main limitation of our study is the sample size, which will need to be expanded in further studies to validate our preliminary findings. In addition, we were not able to match all blood and saliva samples for PCR and cytokine analysis, we did not perform PCR for other viruses than HHV-6 and EBV and we did not have serological data. In this study, Type-I error and false positive results were not of primary concern as the focus was on hypothesis generating and exploration. All cytokines that indicate some utility across key outcomes of interest may be further investigated in future trials in a more tightly controlled manner. Also, CSF could not be obtained from study participants and therefore our observations may represent indirect measures of inflammatory activation in the periphery.

Conclusions
In our study we showed differential activation of neuroinflammatory pathways in

Consent for publication
Not applicable.
Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests:
The authors declare that they have no competing interests. JC contributed to study design, data analysis and interpretation, and drafting and revising of the manuscript for intellectual content.
WT contributed to study design, data analysis and interpretation, and drafting and revising of the manuscript for intellectual content.
WG contributed to study design, data analysis and interpretation, and drafting and revising of the manuscript for intellectual content.
SJ contributed to study design, data analysis and interpretation, and drafting and revising of the manuscript for intellectual content.