Allostatic Load of Immune Cells in the Spleen and Brain of the Kcna1-null Mouse Model of Temporal Lobe Epilepsy

Abstract

Human temporal lobe epilepsy (TLE), particularly drug-resistant TLE, is associated with chronic peripheral immune activation. Here, we determined whether a similar association was detectable in a genetic mouse model of TLE with spontaneous recurrent seizures (SRS), Kcna1-null mice. Flow cytometry was used with fluorescence-activated cell sorting to determine the presence of lymphocytes, macrophages and granulocytes in isolated brain and spleen of wildtype (WT) and Kcna1-null mice. Splenic analysis revealed uniformly elevated Mac-1⁺MHC-II⁺ macrophages across all epileptic mice, whereas CD8⁺ cytotoxic T-cells increased proportionally with severe seizure frequency and burden, resulting in reduced CD4⁺/CD8⁺ ratios—an immune risk phenotype. Brain tissue showed increased infiltration of both CD4⁺ and CD8⁺ T-cells. Importantly, Kv1.1 was not expressed on T-cells of WT mice. Using an allostatic interpretative framework, our results indicate that the immune system anticipates ongoing challenges (SRS) and maintains elevated readiness with an allostatic shift towards chronic adaptation (macrophages), but in doing so potentially damaging dynamic allostatic loads (CD8⁺ T-cells) are accumulated. The Kcna1-null model provides a valuable tool for investigating epilepsy immunopathogenesis without chemoconvulsant confounds. These findings support the notion that TLE has a significant immunological component which may participate in pathology and be a target for intervention.

1. Introduction:

People with temporal lobe epilepsy and/or drug resistant epilepsy may experience chronic activation of the peripheral adaptive immune response associated with altered leukocyte profiles, the presence of peripheral proinflammatory markers and brain infiltration of T-cells.^1,2 This involves, but is not limited to, increases in T-cells and may be related to seizure frequency. Preclinical animal studies using chemoconvulsants to induce acute seizures also report this phenomenon and suggest a role for T-cells in seizure susceptibility. Unfortunately, mechanistic and therapeutic investigations are hampered and interpretations limited because chemoconvulsants can directly activate the adaptive immune response, including lymphocytes.^2–6 To circumvent this limitation and provide means to gain better understanding of the immunopathogenesis of epilepsy, here we determined whether an adaptive immune response is detectable in a model of spontaneous recurrent seizures (SRS) and temporal lone epilepsy using Kcna1-null mice. Kcna1-null mice lack the alpha subunit of the Kv1.1 delayed rectifier potassium channels, develop SRS in the fourth postnatal week which progressively worsen, and exhibit drug-resistant epilepsy.^7,8 In addition, to assist interpreting potential variability between subjects we employed the conceptual framework of allostasis which states that physiologic processes may alter homeostatic set points to maintain stability in response to challenges. The difference between old and new homeostatic set points is referred to as allostatic load, which if prolonged can induce pathology.⁶ In the context of our study, SRS are the challenges and an immune response is the allostatic load.

2. Material and Methods:

Animal husbandry, behavioral seizure monitoring and fluorescence-activated cell sorting procedures are provided in the supplemental methods section.

Data were analyzed with Prism10.3.0 (Graphpad Software, Inc., La Jolla, CA) and reported as the mean ± standard error. Statistical significance was determined with an unpaired t-test, Fisher’s exact test for proportions or linear regression with Pearson’s correlation coefficient where appropriate. Pearson’s correlation coefficient parametric were met (independence, linearity, no outliers, and no heteroscedasticity).

3. Results:

3.1. Allostatic profile of peripheral immune cells from the spleen.

Prior to tissue isolation for flow cytometry, mice were continuously video-recorded for 48-hours and behavioral seizures assessed. The average frequency of SRS, as well as the seizure burden index (SBI), per subject varied (ranging from 1-26 seizures per day) similar to our previous reports^7,8 (Figure 1A, Supplemental Figure 3A–D). To determine differences in the profile of the peripheral immune response, splenic leukocytes were analyzed. Of the total number of leukocytes, the proportion of lymphocytes and granulocytes/monocytes did not differ between genotypes (Figure 1B). Assessment of the granulocyte population indicated that the number of Gr-1⁺ neutrophils did not differ (Figure 1C). Mac-1 can be expressed on a variety of cells, including granulocytes, T-cells, B-cells, NK-cells, dendritic cells, and monocytes.¹⁰ Co-expression of Mac-1⁺ cells with MHC-II⁺ is commonly used to label macrophages.¹⁰ Mac-1⁺MHC-II⁺ cells were elevated in Kcna1-null mice (Figure 1C). This increase occurred in all Kcna1-null mice, irrespective of SRS (p>0.05, Pearson’s correlation, data not shown). Assessment of the lymphocyte population indicated that the numbers of B220⁺ B-cells and CD4⁺ T-cells were similar between genotypes, while the number of splenic CD8⁺ cytotoxic T-cells was significantly increased in Kcna1-null mice (Figure 1D). This increase was positively correlated with seizure frequency and SBI when considering all seizure types or only the more severe Type 3-5 seizures (Figure 1E, Supplemental Figure 3A,B,D), but the correlation was lost when Type 1 brief myoclonic seizures were considered in isolation (Supplemental Figure 3C). The greater amount of CD8⁺ T-cells resulted in lower CD4⁺/CD8⁺ ratios which indicate an immune risk phenotype and were negatively associated with only Type 3-5 seizure frequency and SBI (Figure 1E, Supplemental Figure 3D). Collectively, these data suggest increased allostatic load of select factors of innate and adaptive immune cells in the spleen of mice experiencing frequent, severe SRS.

Figure 1. Increased macrophages and CD8⁺ T-cells in the spleen of Kcna1-null mice.

(A) Daily frequency of generalized convulsive seizures (Type 3-5) for individual Kcna1-null mice. (B) Percentage population of leukocytes in the spleen of WT and Kcna1-null mice with contribution of granulocytes and lymphocytes identified. (C) Percent granulocyte population in WT and Kcna1-null mice in the spleen: (left) Gr1⁺ neutrophils and (right) Mac-1⁺ MHC-II⁺ macrophages. (D) Percent lymphocyte population in WT and Kcna1-null mice in the spleen: (left) B220⁺ B-cells, (middle) CD4⁺ T-cells, and (right) CD8⁺ T-cells. (E) Correlation analysis of seizures and lymphocytes: frequency of Type 3-5 generalized convulsive seizures compared to (left) splenic CD8+ T-cells and (right) splenic CD4⁺/CD8⁺ ratio in Kcna1-null mice. Unpaired Student’s t test and Pearson’s correlation were used to determine significance. Data are mean±SEM; n=8-10; *p<0.05. **p<0.01.

3.2. Allostatic profile of immune cells in brain tissue.

The overall proportion of lymphocytes and granulocytes/monocytes in brain tissue did not differ between genotype (Figure 2A). Specifically, Gr-1⁺ neutrophils, Mac-1⁺MHC-II⁺ macrophages and B220⁺ B-cells did not differ (Figure 2B,C). In contrast, CD4⁺ T-cells and CD8⁺ cytotoxic T-cells were significantly increased (Figure 2C). The simultaneous increase in both populations, resulted in no change in the CD4⁺/CD8⁺ ratio. Interestingly, the degree of infiltration of CD4⁺ T cells was negatively correlated with seizures, whereas the increased CD8⁺ T cells were not associated with seizure frequency (Supplemental Figure 4). Collectively, these data suggest that the presence of severe seizures, even as low as once per day, is associated with increased allostatic load of adaptive immunity in the brain.

Figure 2. Increased CD4⁺ and CD8⁺ T-cells in the brain of Kcna1-null mice.

(A) Percentage population of leukocytes in the brain of WT and Kcna1-null mice with contribution of granulocytes and lymphocytes identified. (B) Percent granulocyte population in WT and Kcna1-null mice in the brain: (left) Gr1⁺ neutrophils and (right) Mac-1⁺ MHC-II⁺ macrophages. (C) Percent lymphocyte population in WT and Kcna1-null mice in the brain: (left) B220⁺ B-cells, (middle left) CD4⁺ T-cells, (middle right) CD8⁺ T-cells, and (right) CD4⁺/CD8⁺ brain ratio. (D) Kv1.1 expression on (left) CD4⁺ and (right) CD8⁺ T-cells in the brain and spleen of WT mice. Student’s t test was used to determine significance. Data are mean±SEM; n=8-10 (A-C) and n=3 (D); *p<0.05. **p<0.01.

To address potential direct effects of Kv1.1 loss from T cells, we determined whether Kv1.1 is expressed on CD4⁺ or CD8⁺ T cells. Previous studies have reported conflicting results regarding the presence of Kv1.1 in T-cells.^11–13 Here, co-expression of Kv1.1 was not found with either CD4⁺ or CD8⁺ when gating for CD3⁺ T-lymphocytes specifically (Figure 2D), or when gating for total lymphocytes isolated from WT mice (data not shown).

4. Discussion

We provide evidence that the Kcna1-null mouse model of TLE experiences a chronic adaptive immune response establishing it as a valuable tool for studying epilepsy immunopathogenesis. Previous preclinical studies report that pilocarpine-induced and kainate-induced prolonged status epilepticus (SE) increases CD4⁺ and CD8⁺ T-cells in the spleen and brain;^4,14,15 however, T-lymphocytes express receptors for these drugs (i.e. muscarinic acetylcholine receptors and glutamate receptors), thus confounding interpretations that emphasize the role of seizures.^4,5 Here, we demonstrate that mouse T-cells do not express the Kv1.1alpha subunit, thus utilization of epileptic Kcna1-null mice removes this limitation of potential confounding external variables. Our findings support an adaptive immune response in chronic epilepsy with a strong association with SRS frequency and severity.

Kcna1-null mice exhibit increased splenic granulocytes and lymphocytes. Specifically, Mac-1⁺MHCII⁺ macrophages increased uniformly across all Kcna1-null mice regardless of SRS. MHC-II expression on splenic macrophages (innate immunity) contributes to the induction of the adaptive immune response, i.e. naïve CD8⁺ or CD4⁺ T-cells (adaptive immunity) are activated only after antigen presentation via MHC-I or II molecules.¹⁶ This suggests a baseline shift in innate immune activation associated with the epileptic state rather than a direct response to seizure. In contrast, the increase of CD8⁺ cytotoxic T-cell positively correlated with frequency and burden of severe convulsive seizures, but not brief myoclonic seizures. The severe seizures likely cause greater neuronal damage, blood-brain barrier disruption, and release of brain-derived antigens into the periphery, thereby providing stronger signals for adaptive immune activation.¹⁷ Alternatively, or in concert, stimulation of the sympathetic system by severe SRS could enhance expansion of CD8⁺ T-cells in a graded manner.¹⁸ Studies are needed to explore these possibilities.

Elevated CD8⁺ T-cells resulted in reduced CD4⁺/CD8⁺ ratios in Kcna1-null mice which is particularly noteworthy as this represents an immune risk phenotype associated with chronic inflammation, accelerated immune aging, and increased morbidity in various clinical contexts.¹⁹ The negative correlation between CD4+/CD8+ ratios and severe SRS further reinforces that severe seizures drive a shift toward a pro-inflammatory, cytotoxic immune profile. This phenotype may contribute to disease progression, as chronic CD8+ T-cell activation can perpetuate inflammation and potentially lower seizure thresholds, creating a vicious cycle of seizures and immune activation.²⁰

In the Kcna1-null brain, infiltration of CD4⁺ T-cells and CD8⁺ cytotoxic T-cells increased, further supporting that peripheral adaptive immune activation is accompanied by central immune involvement. This finding parallels clinical studies where genetic and acquired epilepsies experienced infiltration of CD4⁺ and CD8⁺ T-lymphocytes in the brain parenchyma.^2,21 The CD4⁺/CD8⁺ did not change due to the simultaneous increase in both T-cell subsets suggesting balanced recruitment into the CNS. The increase in CD4⁺ T-cells in the brain but not in the periphery may reflect increased blood-brain barrier permeability in Kcna1-null mice as we have previously reported.²² Alternatively, increases could be due to other sources like the thymus, blood, or lymph nodes. The unexpected and apparent paradoxical negative correlation between CD4⁺ T-cell infiltration and SRS could reflect early recruitment of anti-inflammatory regulatory T-cells (Treg) that is overwhelmed or exhausted with increased SRS. Tregs have been demonstrated to protect against seizures.² Further phenotypic characterization of CD4⁺ subsets is necessary to distinguish between these possibilities. In contrast, the lack of correlation between CD8+ T-cell brain infiltration and SRS, despite the strong peripheral correlation, suggests that once CD8+ T-cells enter the brain, they establish residence and persist locally, contributing to chronic neuroinflammation regardless of current seizure burden.

Unlike homeostasis which employs changes to return systems to baseline, the concept of allostasis describes how physiological systems achieve stability by establishing new set points in response to chronic, repeated challenges.⁹ Applying the conceptual framework of allostasis to our findings reveals novel insights. The peripheral uniform elevation of Mac1⁺MHCII⁺ macrophages represent a stable allostatic shift or recalibrated baseline in which the immune system has established a new ‘set point’ for macrophages. This suggests a chronic adaptive state rather than acute reactivity. In contrast, splenic CD8⁺ T-cells show a dynamic allostatic load and scale with the challenge (SRS) severity. Furthermore, the loss of correlation with brief myoclonic seizures indicates the allostatic response has a minimal threshold. The resulting decrease in CD4⁺/CD8⁺ ratio indicates potential allostatic overload and the negative association with SRS suggests that the recalibration is adaptive initially but may become pathologic. Likewise, interpretations may be made for brain tissue, but the divergent pattern of brain infiltration suggests tissue-specific allostatic strategies. Overall, through the lens of allostasis our results indicate that the immune system anticipates ongoing seizure activity and maintains elevated readiness with chronic adaptation, but in doing so also accumulates potentially damaging allostatic loads. This framework suggests the immune system alterations are not simply ‘inflammation’ but rather an organized recalibration in an attempt to maintain stability under chronic neurological stress.

In conclusion, epileptic Kcna1-null mice exhibited an adaptive immune response with similarities to human temporal lobe epilepsy validating the use of this mouse model to probe mechanisms and potential therapies. These findings support the notion that epilepsy, particularly drug-resistant temporal lobe epilepsy, has a significant immunological component which may participate in pathology and be a target for intervention.

Supplementary Material

Suppl Figure 1

Supplementary Figure 1. Gating Strategy for leukocytes. (A) Flow cytometry dot-plots demonstrate the region of total neutrophils/macrophages and the data analyzed in (right, top) Mac-1⁺ or Gr-1⁺ or (right, bottom) Mac-1⁺MHCII⁺. (B) Flow cytometry dot-plots demonstrate the region of total lymphocytes and the data analyzed in (right, top) B220⁺ or CD4⁺, (right middle) B220⁺ or CD8⁺, and (right, bottom) CD8⁺ or CD4⁺.

Suppl Figure 2

Supplementary Figure 2. Gating Strategy for Kv1.1. (A) Flow cytometry dot-plots demonstrate the region of (in order from left to right): total lymphocytes, CD3⁺ T lymphocytes, CD4⁺ and CD8⁺ T lymphocytes and the data analyzed in (right, top) CD4⁺ and Kv1.1⁺ or (right, bottom) CD8⁺ and Kv1.1⁺.

Suppl Figure 3

Supplementary Figure 3. Correlation of splenic CD8⁺ T cells with seizure frequency and seizure burden. (A) Frequency of all types of seizures per day (left) were not correlated with splenic CD8⁺ T cells (middle) or the resulting CD4⁺/CD8⁺ ratio (right). (B) Seizure burden index (SBI) of all types of seizures per day (left) were correlated with splenic CD8⁺ T cells (middle) but not the CD4⁺/CD8⁺ ratio (right). (C) Frequency of Type 1 myoclonic seizures per day (left) were not correlated with splenic CD8⁺ T cells (middle) or resulting CD4⁺/CD8⁺ ratio (right). (D) SBI of severe Type 3-5 convulsive seizures per day (left) were correlated with splenic CD8⁺ T cells (middle) and the CD4⁺/CD8⁺ ratio (right).

Suppl Figure 4

Supplementary Figure 3. Correlation of brain CD4⁺ and CD8⁺ T cells with seizure frequency and seizure burden. (A) Frequency of all types of seizures per day were not correlated with brain CD4⁺ T cells. (B) Seizure burden index (SBI) of all types of seizures per day were not correlated with brain CD4⁺ T cells. (C) Frequency of Type 1 myoclonic seizures per day were not correlated with brain CD4⁺ T cells. (D) Frequency of severe Type 3-5 convulsive seizures per day were negatively correlated with brain CD4⁺ T cells. (E) SBI of severe Type 3-5 convulsive seizures per day were negatively correlated with brain CD4⁺ T cells. (F) Frequency of severe Type 3-5 convulsive seizures per day were not correlated with brain CD8⁺ T cells.

Highlights.

• Splenic macrophages increase uniformly in the Kcna1-null mouse model of TLE

• Spontaneous recurrent seizures are positively correlated with increased splenic CD8⁺ T-cells but negatively correlated with reduced CD4⁺/CD8⁺ ratios indicative of immune risk.

• CD4⁺ and CD8⁺ T-cells are increased in the brain of Kcna1-null mice.

• An allostatic interpretative framework suggests Kcna1-null mice experience an allostatic shift resulting in chronic adaptation of the immune system coupled with dynamic allostatic loads associated with seizures that may further contribute to the pathology of epilepsy.

Data Availability

The data that support the findings of this study are available from the corresponding authors upon reasonable request.