The hydroxycarboxylic acid receptor HCA2 is required for the protective effect of ketogenic diet in epilepsy

Abstract

One third of epilepsy patients are resistant to treatment with current anti‐seizure medications. The ketogenic diet is used to treat some forms of refractory epilepsy, but the mechanism of its action has not yet been elucidated. In this study, we aimed to investigate whether the hydroxycarboxylic acid receptor 2 (HCA2), a known immunomodulatory receptor, plays a role in mediating the protective effect of this diet. We demonstrate for the first time that selective agonists at this receptor can directly reduce seizures in animal models. Agonists also reduce network activity in rodent and human brain slices. Ketogenic diet is known to increase circulating levels of endogenous HCA2 agonists, and we show that the effect of ketogenic diet in reducing seizures in the 6 Hz seizure model is negated in HCA2‐deficient mice. Our data support the potential of HCA2 as a target for the treatment of epilepsy and potentially for neurodegenerative diseases.

Abbreviations

• aCSF
• artificial cerebrospinal fluid
• ACTH
• adrenocorticotropic hormone
• AD
• Alzheimer's disease
• ASM
• anti-seizure medication
• b.i.d.
• [twice per day]
• CC₅₀
• current causing half-maximal effect
• EAE
• experimental autoimmune encephalomyelitis
• FFA
• free fatty acid
• GLUT-1
• glucose transporter type-1
• HCA
• hydroxycarboxylic acid
• KO
• knock-out
• LPS
• lipopolysaccharide
• MMF
• mono-methylfumarate
• MS
• multiple sclerosis
• PD
• Parkinson's disease
• PTZ
• pentylenetetrazol
• SLE
• seizure-like event
• SRS
• spontaneous recurrent seizures
• WT
• wild-type
• β-HB
• β-hydroxybutyrate

INTRODUCTION

The hydroxycarboxylic acid receptors (HCA₁, HCA₂ and HCA₃) are a family of G-protein-coupled receptors that are critical for sensing endogenous intermediates of metabolism. All three receptors are expressed on adipocytes and mediate anti-lipolytic effects in response to starvation. Hydroxycarboxylic acid receptor 2 (HCA₂; previously referred to as HM74a or GPR109a) was originally identified as the receptor for niacin.¹ In addition to its expression on lipid cells, HCA₂ is also present on immune cells including macrophages, monocytes, microglia, neutrophils and dermal dendritic cells.² Under basal conditions expression of HCA₂ is low, but in peritoneal macrophages and RAW 264.7 cells, HCA₂ expression is increased 20- to 80-fold by LPS.^2,3 Numerous recent studies have demonstrated that HCA₂ mediates immunomodulatory effects in various immune cell types, including macrophages.⁴ Nicotinic acid (niacin), monomethyl fumarate (MMF) and β-hydroxybutyrate (3-hydroxybutyrate; β-HB) are agonists of HCA₂ and demonstrate immunomodulatory effects in several chronic inflammatory disease models^5,6 and at least some of their effects have been shown to be elicited via peripherally expressed HCA₂ on macrophages.⁶ The anti-inflammatory action of β-HB in rat primary cultured microglia has been shown to be mediated through HCA₂.⁷ Nicotinic acid, ketogenic diet (high-fat, low-carbohydrate diet) and MMF reduce clinical score in the experimental autoimmune encephalomyelitis (EAE) model of multiple sclerosis (MS) and the effect was shown to be both macrophage and HCA₂-dependent for MMF using HCA₂-deficient mice.⁵ There is a growing literature supporting a role for the innate immune system in the initiation and progression of epilepsy.^8–12 Inflammation increases neuronal excitability and there are an increasing number of examples of immunomodulatory drugs demonstrating efficacy in chronic models of seizures.¹³ Similarly, the role of autoimmunity and neuroinflammation in development of clinical seizures and epilepsy is being increasingly recognized.^14,15 For instance, patients with autoimmune and inflammatory encephalitides develop epilepsy; and some epilepsy syndromes (e.g. West syndrome, Rasmussen's encephalitis) respond to anti-inflammatory (e.g. steroids, ACTH), immunosuppressive or immunomodulatory treatments.^16–18 Tecfidera^R (the oral prodrug dimethyl fumarate, which is rapidly de-esterified to MMF on absorption) is used to treat relapsing remitting multiple sclerosis, although activation of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) and inhibition of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) by MMF may contribute to its efficacy, in addition to agonism of HCA₂.^19,20 Ketogenic diet has itself begun to be investigated in treatment of multiple sclerosis (reviewed in²¹).

β-HB, a ketone body, has been proposed by some reports to mediate the action of the ketogenic diet in treatment of refractory epilepsy, though this is disputed^22–24 and other mechanisms have been implicated.^25–32 β-HB is the main endogenous ligand of human HCA₂ with an EC₅₀ of around 0.7–0.8 mM in mammalian assays, a concentration that this ketone body reaches only after fasting or following at least 2–3 weeks of ketogenic diet (that elevated plasma β-HB level to 2-4 mM³³). Both fasting and ketogenic diet have been shown to mediate anti-seizure effects. To our knowledge, there are no data demonstrating that a selective HCA₂ agonist is efficacious in a pre-clinical model of epilepsy. We hypothesised that the efficacy in epilepsy of ketogenic diet is via agonism of HCA₂ by β-HB, and that synthetic HCA₂ agonists (e.g. GSK256073) may be able to recapitulate effects of the ketogenic diet in a model of epilepsy.

RESULTS

Sub-chronic dosing of selective HCA₂ agonists demonstrates anti-convulsant effects in mouse 6Hz model

We chose the 6 Hz and PTZ animal models of epilepsy based on published data showing ketogenic diet to reduce seizure activity in these systems.^34,35 We assessed structurally distinct HCA₂ agonists, to gain evidence that anti-convulsant effects were mediated by direct activation of HCA₂. Compounds used were the selective HCA₂ agonists GSK256073 and 780A which are in the same structural class (xanthines), and 4a which is structurally unrelated to the xanthines. Later studies included MMF, which is less selective for HCA₂ but has been studied extensively and is in clinical use. Compound 245A is a xanthine differing from GSK256073 by a single methylene group and was included to support human dose prediction (structures are presented in Figure S1). Intravenous infusion of PTZ (20 mg/mL at 1 mL/min) in the rat induces myoclonic and forelimb tonic seizures. Table S1A demonstrates that GSK256073 administered orally (1, 3, 10, 30, 60 and 100 mg/kg) 30 mins prior to PTZ did not significantly change the dose of PTZ required to induce myoclonic or forelimb tonic seizures compared to the vehicle control group. Table S1B shows plasma concentrations of GSK256073 and PTZ, and concentrations of GSK256073 in cerebrospinal fluid, from this study. The positive control, diazepam (10 mg/kg) significantly increased the dose of PTZ required to induce both myoclonic (47.0% vs. vehicle) and forelimb tonic (114.4% vs. vehicle) seizures as expected. Since the efficacy of the ketogenic diet is observed only after several weeks administration, we investigated whether GSK256073 could reduce seizures after sub-chronic (repeat) dosing in the rat PTZ model. Table S2 demonstrates that GSK256073 at 60 mg/kg administered orally (b.i.d. dosing for days 1–7 followed by a single dose on the morning of day 8), significantly increased the dose of PTZ (113.8% vs. vehicle) required to induce forelimb tonic seizures. None of the doses of GSK256073 administered (10, 30 and 60 mg/kg) significantly changed the dose of PTZ required to induce myoclonic seizures compared to the vehicle control group. Diazepam administered orally at 10 mg/kg, significantly increased the dose of PTZ required to induce both myoclonic (154.3% vs. vehicle) and forelimb tonic (183.6% vs. vehicle) seizures as expected.

Table S3A demonstrates that compound 4a at both 30 mg/kg and 60 mg/kg and compound 780A at 60 mg/kg administered orally, after repeat dosing (b.i.d. dosing for days 1–7 followed by a single dose on the morning of day 8) significantly increased the dose of PTZ (4a (30 mg/kg): 92.36% vs. vehicle; 4a (60 mg/kg): 81.64% vs. vehicle; 780A (60 mg/kg): 76.49% vs. vehicle) required to induce forelimb tonic seizures. Neither compound significantly changed the dose of PTZ required to induce myoclonic seizures compared to the vehicle control group. Diazepam administered orally at 10 mg/kg, significantly increased the dose of PTZ required to induce both myoclonic (88.94% vs. vehicle) and forelimb tonic (152.54% vs. vehicle) seizures as expected. Table S3B shows blood and brain concentrations of compounds 4a and 780A in this study. Overall, we have shown in the rat PTZ model that three structurally distinct, orally available and selective HCA₂ agonists exhibit an anti-convulsant action, after sub-chronic dosing.

Sub-chronic dosing of selective HCA₂ agonists demonstrates anti-convulsant effects

Of all the commonly used models of epilepsy, data in the 6 Hz mouse model has provided the most robust anti-seizure effects of the ketogenic diet.^36,37 To further validate whether HCA₂ agonism may mediate the effect of this diet, we investigated the effects of HCA₂ agonists in this model. In the 6 Hz test, a 32-mA electrical stimulus resulted in the expression of at least one of the following behaviours: immobility/stun, lateral head movements, straub tail and forelimb clonus in vehicle treated mice. Typically, immobility/stun, lateral head movement and straub tail were the most common signs and were frequently co-expressed in affected mice. Pre-treatment with the positive control, levetiracetam (100 mg/kg) significantly reduced seizures measured either as incidence or as a total score based on the individual seizure related behaviours expressed (p < 0.05 vs. acute vehicle pre-treatment, Kruskall-Wallis test). Pre-treatment with both 4a (30 mg/kg) and 780A (60 mg/kg) also reduced the incidence of seizure induced by a 6 Hz stimulus, although the magnitude of effect was not as marked as that of levetiracetam and not all mice were affected by treatment. The only significant difference noted between 4a or 780A treatment groups and vehicle control was in mice treated for 7 days with 780A on average seizure score (p < 0.05 vs. 7-day vehicle pre-treatment, Kruskall–Wallis test). However, there were trends following both acute and 7-day treatment with 4a (30 mg/kg) and acute treatment with 780A (60 mg/kg, Figure 1 a,b). Blood and brain levels of test articles were analysed to confirm levels sufficient for target engagement at the HCA₂ receptor (Table S5). Pre-treatment with GSK256073 also appeared to reduce the incidence of seizure induced by the 6 Hz stimulus, although again the magnitude of effect was not as pronounced as levetiracetam (Figure 1c,d). For example, while levetiracetam completely inhibited seizures in 8/8 (100%) mice, the most pronounced effect of GSK256073 was in the 60 mg/kg sub-chronic group, where seizures were prevented in 5/8 (62.5%) of mice. Both groups treated with sub-chronic 30 and 60 mg/kg exhibited seizure scores significantly lower than vehicle controls (*p < 0.05; **p < 0.01 vs. 7-day vehicle pre-treatment, Kruskall–Wallis test). No significant effect of acute treatment with GSK256073 (3–100 mg/kg) was noted on seizure score compared to acute vehicle pre-treatment, although a modest dose-related trend to reduce both seizure incidence and score was noted. Blood and brain concentrations of GSK256073 in this study are presented in Table S6. The demonstration that selective and structurally distinct HCA₂ agonists exhibit anti-convulsant actions in the 6 Hz model suggests that these effects are dependent on the HCA₂ receptor.

[IMAGE OMITTED. SEE PDF]

We proceeded to confirm whether the effect of HCA₂ agonists in the 6 Hz model was indeed target dependent by investigating whether the protective effect of GSK256073 in reducing 6 Hz seizures was maintained in HCA₂-deficient mice (KO, Study 2). Both WT and KO groups following vehicle pre-treatment exhibited similar CC₅₀ values (WT: 21.1 (18.8–23.8) mA; KO: 23.4 (18.8 to 29.0) mA, Figure S2). Within each stimulus intensity direct comparisons between treatment groups within each genotype were evaluated. Using seizure incidence (±) as the dependent measure, treatment comparisons were made using Chi-squared test within each stimulus intensity. Using seizure score (0–4) as the dependent measure, treatment comparisons were made using Kruskall-Wallis test within each stimulus intensity. In WT mice, treatment with GSK256073 at 60 mg/kg b.i.d. for 7 days produced a modest protective effect against 6 Hz psychomotor seizures. This was evidenced by an increase in the CC₅₀ measure (WT (Vehicle): 21.1 (18.8–23.8) mA; WT (GSK256073): 27.9 (21.9–35.6) mA, Figure S2). Furthermore, analysis of seizure incidence at 22 mA revealed a significant difference between WT mice treated with GSK256073 compared to vehicle (p < 0.01; Chi-squared test, Figure 2A). Analysis of seizure score also showed a trend for GSK256073 to protect against 6 Hz seizures at the 22 mA and 32 mA stimulus intensity (p < 0.1; Mann–Whitney U-test, see Figure 2C). In contrast, GSK256073 showed no protective effect in KO mice (Figure 2B,D); CC₅₀ measures were equivalent (KO (Vehicle): 23.4 (18.85 to 29.05) mA; KO (GSK256073): 23.8 (16.53 to 34.27) mA, Figure S2). Blood and brain concentrations of GSK256073 in all animals was not significantly different across treatment groups (Table S7). The anti-convulsant action of HCA₂ agonist GSK256073 in the 6 Hz model is therefore dependent on the presence of the HCA₂ receptor.

[IMAGE OMITTED. SEE PDF]

HCA₂ mediates the protective effect of the ketogenic diet in the 6 Hz model

In the 6 Hz test, both WT and KO groups following vehicle pre-treatment had similar and overlapping CC₅₀ values based on seizure score (WT: 19.3 (16.5–22.6) mA; KO: 24.1 (5.3–109) mA, Figure 3 and Figure S3). Pre-study baseline levels of glucose and β-HB were not measured but a comparison of WT and KO mice fed normal diet (day 4) identified no significant genotype differences on blood glucose (WT: 8.1 + 0.2 mM; KO: 7.7 + 0.1 mM, p = 0.07, T-test) and β-HB levels (WT: 0.41 + 0.02 mM; KO: 0.40 + 0.02 mM, NS, T-test). In WT mice, ketogenic diet as sole food source for 14 days produced a modest protective effect against 6 Hz psychomotor seizures (Study 3). This was evidenced by an increase in the CC₅₀ measure (WT (normal diet): 19.3 (16.5–22.6) mA; WT (ketogenic diet): 34.0 (9.9–116) mA, Figure 3 and Figure S3). Also, analysis of seizure incidence at 22 mA revealed a significant difference between dietary groups (p < 0.01; Chi-squares test), and at 18 mA and 32 mA the difference was of borderline significance (p = 0.07; Chi-Squares test). Analysis of seizure score also showed a significant difference between dietary groups at the 22 mA and 32 mA stimulus intensity (p < 0.05; Mann–Whitney U-test, Figure 3). In WT mice, treatment with ketogenic diet increased blood β-HB levels 3-fold (normal: 0.4–0.5 mM; ketogenic diet: 1.4–1.5 mM; p < 0.01), and reduced blood glucose levels (normal: 8–9 mM; ketogenic diet: 6–7 mM; p < 0.01). These changes were maintained throughout the 14 days of diet (not shown). In KO mice, treatment with a ketogenic diet as sole food source did not produce a reliable protective effect against 6 Hz seizures. For example, the CC₅₀ measure was very similar between the treatment groups (KO (normal diet): 24.1 (5.3–109) mA; KO (ketogenic diet): 20.0 (17.8–22.5) mA, Figure 3 and Figure S3). Analysis of seizure score failed to identify significant difference between the dietary groups at the 18 mA, 22 mA and 32 mA stimulus intensities (p > 0.05; Mann–Whitney U-test). Furthermore, seizure incidence was unaffected by diet at the 18 mA, 22 mA and 32 mA stimulus (p > 0.05, Chi-squared test). In summary, in contrast to WT mice, no protective effect of the ketogenic diet against 6 Hz seizures was observed in the KO mouse line. In KO mice, treatment with ketogenic diet increased blood β-HB levels approximately 3-fold (regular: 0.4–0.5 mM; ketogenic diet: 1.4–1.5 mM; p < 0.01; Mann–Whitney U-test), and reduced blood glucose levels (regular: 8–9 mM; ketogenic diet: 6–7 mM; p < 0.01; Mann–Whitney U-test). These changes were maintained throughout the 14 days of diet. Thus, the effects of ketogenic diet on blood β-HB and glucose levels were similar between the WT and KO groups.

[IMAGE OMITTED. SEE PDF]

The exact molecular action of the ketogenic diet in epilepsy has not been reported. Our data are the first to show a direct link between the ketogenic diet and HCA₂ agonism. Similarly, there are no reports demonstrating the anti-convulsant effects of selective HCA₂ agonists in any model of epilepsy.

HCA₂ agonists reduce seizure-like events in rodent and human brain slices

To further understand whether selective HCA₂ agonists act in a direct manner in reducing seizures in the epilepsy animal models, we next investigated the effects of selective HCA₂ agonists for their ability to modify induced seizure-like events (SLEs; the in vitro correlate of seizures) in acute brain slices in vitro from both humans and rodents. For the human tissue studies, 245A was used in place of GSK256073, to support dose prediction. Based on previous pharmacological data from a range of assays, an optimal concentration of compound was used to determine efficacy for reducing the frequency and magnitude of induced epileptiform neuronal activity initiated by a change of external ion concentrations. Aberrant ion homeostasis has been implicated in epilepsy³⁸ and SLEs can be facilitated in vitro by elevating extracellular potassium (K⁺)₀^39–41 and/or lowering magnesium (Mg²⁺)₀.³⁹ In this study, SLEs were induced by replacing normal aCSF with “modified” aCSF (mACSF) containing reduced external (Mg²⁺)₀ and raised (K⁺)₀ (0.25 μM Mg²⁺/8 mM K⁺).⁴²

The statistical measures used in this study identified a significant reduction in the power spectra of epileptic activity between 245A and control condition and between MMF and control condition in rodent tissue (n = 9; Figure 4 and Tables S4a and S4b). For compounds MMF and 245A, statistical analysis showed that in general, washout of compound did not reverse the effect upon epileptic activity (Figure 4C). It is not clear if this is due to the mode of action of compounds or due to intrinsic factors of the activity within the slices. We also observed effects of a range of HCA₂ agonists (of different chemotypes) on reducing the power of epileptic activity in human tissue (summarised in Table S4c). Clearly, translation into human slices is advantageous in terms of proving target engagement in human material is associated with a functional effect, but additional repeats are required before any statistical analyses can be provided.

[IMAGE OMITTED. SEE PDF]

Discussion

The HCA₂ receptor has roles in immunometabolism, acting as a sensor for its primary ligand, the ketone body β-HB. Under certain conditions, β-HB activates HCA₂ and downstream signalling to control adipocyte function and elicit anti-inflammatory effects.⁶ β-HB reaches concentrations sufficient to activate HCA₂ during fasting or following administration of a ketogenic diet,³³ both of which are known to have anti-seizure effects in intractable epilepsy.⁴³ Many studies have been conducted to investigate the mechanism of action of the ketogenic diet in pre-clinical models of epilepsy, but none have so far implicated HCA₂. The consensus remains that the mechanism of action of the ketogenic diet is not fully understood, with multiple targets being implicated.^29,30 Niacin, β-HB and MMF are reported to mediate neuroprotective effects via HCA₂ agonism in models of neurodegeneration,⁶ but none have been tested in models of epilepsy. We therefore sought to investigate whether selective HCA₂ agonists would exhibit anti-convulsant effects. We chose induced models of seizure based on the published correlation of β-HB levels with anti-seizure effect after administration of ketogenic diet.³⁴ We show efficacy of structurally distinct HCA₂ agonists in both the PTZ and 6 Hz models. Rat (PTZ studies) was chosen to support dose prediction, since safety studies are more commonly performed in this species. Mouse was preferred for the 6 Hz model, allowing inclusion of HCA₂-deficient mice. Translation of antiseizure efficacy to other models or to patients is difficult to infer from our 6 Hz data alone, because the activity profile of HCA₂ agonists (efficacy at an intermediate stimulus intensity of 22 mA) is common to other CNS-acting compounds that lack clinical antiseizure efficacy. However, antiseizure effects in human brain slices suggest the efficacy of HCA₂ agonists could translate to human disease and merits further investigation. Future work will be required to extend these findings to models of epilepsy with spontaneous recurrent seizures (SRS), where ketogenic diet is known to also be efficacious.⁴⁴ The magnitude of effects we observe is moderate, consistent with the involvement of multiple beneficial mechanisms associated with ketogenic diet. However, our data showing reduction of seizures with ketogenic diet in wild-type but not HCA₂ knock-out mice in the 6 Hz model implicate HCA₂ agonism as a component of ketogenic diet efficacy. In addition, emerging evidence of sustainability of the effect of ketogenic diet following withdrawal of the diet in patients, suggests an effect on disease progression.^45,46 As such, there is potential for an HCA₂ agonist to exert a similar disease modifying effect, not observed with any standard anti-seizure medication (ASM), in addition to the anti-seizure effects we have reported.

The cellular site-of-action for HCA₂ agonists in reducing seizures is not yet defined. If the clinical effect of ketogenic diet is mediated through activation of the HCA₂ receptor, this might occur via modulation of immune cells to a more protective phenotype. In mice, HCA₂ appears to activate a Ly6C^lo population of resident macrophages, which are protective (‘M2’) in models of neurodegeneration.⁶ HCA₂ receptor mRNA is expressed at low levels in unstimulated conditions in rat primary cultured microglia but is induced in response to LPS, suggesting involvement of HCA₂ during the early stages of microglial activation. The anti-inflammatory action of β-HB (1.5 mM) in rat primary cultured microglia has been shown to be mediated through the HCA₂ receptor and downstream NF-kB activation.^7,47 Microglia are highly relevant to the pathophysiology of epilepsy,⁴⁸ though the neuroprotective role of HCA₂ may be mediated by macrophages in some models of neurodegenerative disease. After induction of ischemic stroke, HCA₂ is expressed on monocytes/macrophages that infiltrate the brain at the site of injury. When fed a ketogenic diet or treated with β-HB or niacin, cell death and infarct size are reduced, and cognition improved. These neuroprotective effects are lost in HCA₂ deficient mice, and studies in chimeric mice demonstrate that the neuroprotective effect of these ligands is mediated by bone marrow-derived monocytes/macrophages.⁶

From our data, HCA₂ agonists may have therapeutic utility in lowering the propensity to seizures, potentially by promoting a switch to a more regenerative microglial phenotype, differentiating from ASMs acting through neuronal mechanisms. HCA₂ agonists could impact disease progression and might offer an alternative to ketogenic diet. Efficacy of the ketogenic diet appears comparable to that of modern ASMs. Ketogenic diet is considered an effective treatment for epilepsy patients unresponsive or poorly responsive to ASMs, regardless of age^49–51 or seizure type, with positive effects on seizure frequency, cognition and behaviour.⁵² Strong evidence supports use of the ketogenic diet in children including as an early or a first line treatment option in several paediatric epilepsy syndromes including Dravet syndrome and GLUT-1 deficiency.^36,53–55 Unfortunately, compliance to the ketogenic diet is challenging especially among adults, so an HCA₂ agonist may be a useful alternative therapeutic approach.⁵⁶

GSK256073 was investigated in phase II clinical trials for dyslipidaemia and diabetes and shown to have favourable safety and tolerability profiles. Target engagement was demonstrated, as evidenced by the reduction of plasma free fatty acids (FFA), but no sustained modulation of relevant clinical end points was demonstrated (HDL/LDL-cholesterol⁵⁷ or HbA1c⁵⁸ for dyslipidaemia or diabetes, respectively). Studies using HCA₂-deficient mice later confirmed that the effect of niacin on FFA is mediated via HCA₂ whereas its effects on other plasma lipids (e.g. triglycerides, HDL/LDL-cholesterol) are independent of HCA₂, challenging the long-standing FFA hypothesis previously used to explain the lipid benefits of niacin.⁵⁹ Interestingly, niacin shows protection in models of atherosclerosis via both HCA₂-dependent and HCA₂-independent effects; the HCA₂-dependent effects are mediated by macrophages,^59,60 warranting further investigation of HCA₂ in more chronic indications targeting immune cells.⁴ As noted, MMF, a more potent HCA₂ agonist than niacin, was also shown to mediate its protective effect in the EAE model of multiple sclerosis via HCA₂, although the specific cell type of its action was not confirmed.⁵

Based on the evidence of neuroprotective and anti-neuroinflammatory effects, GSK256073 would be predicted to “reset” homeostasis of the innate immune system and have potential utility in a range of neurodegenerative diseases including AD, PD, stroke and MS.⁶¹ This idea is further supported by the mounting evidence for potential efficacy of a ketogenic diet in neurodegenerative diseases, including AD.^62–65 Efficacy in disease might be further enhanced by modulators binding at an HCA2 allosteric site, to potentiate agonists acting at the orthosteric binding site of niacin, MMF, acipimox, GSK256073 and (presumptively) β-HB.^66,67

In summary, our data demonstrate that structurally distinct HCA₂ agonists are efficacious in acute seizure models and the effect is further shown to be HCA₂-dependent using HCA₂-deficient mice. Encouragingly, we observed effects of HCA₂ agonists in reducing network and seizure like activity in both cultured rodent and human brain slices. The effect of the ketogenic diet on reducing seizures in wild-type but not HCA₂-deficient mice in the 6 Hz model provides the first data to directly link HCA₂ agonism to the mechanism of action of this clinically precedented diet used to treat refractory epilepsy patients. HCA₂ agonists could provide an advantage over ketogenic diet in terms of increased tolerability, improving compliance. GSK256073, which is thought to act via immune cells, may differentiate from ASMs that act through primarily neuronal mechanisms, for example via ion channel modulation. The combined impact of activation of the HCA₂ receptor on macrophage polarisation, reducing seizure activity and potentially affecting disease progression supports a rationale for the development of HCA₂ agonists for treatment of epilepsy.

METHODS

All animal studies were ethically reviewed by the institution according to local regulations and conducted in accordance with the Animals (Scientific Procedures) Act 1986 or Canadian Council on Animal Care (CCAC) and the GSK Policy on the Care, Welfare and Treatment of Animals. Detailed methods for rat and human brain slice electrophysiology are presented in Data S1.

HCA2 Agonists

GSK256073 (8-chloro-3-pentyl-3,7-dihydro-1H-purine-2,6-dione) was used as the Tris salt throughout. Compound 245A (example 10 in patent WO2005077950: Medicaments with HM74A receptor activity; Pinto, I.L., Rahman, S.S. and Nicholson, N.H., SmithKline Beecham corporation) is structurally similar to GSK256073, differing by a methyl group. Compound 4a is 3-methyl-5-carboxyl-isoxazole.⁶⁸ Compound 780A is example 32 from patent WO2007017261 (Xanthine derivatives as selective HM74A agonists; Hatley, R.J.D., Mason, A.M. & Pinto, I.L., SmithKline Beecham corporation). Monomethylfumarate (MMF) was obtained from Sigma Aldrich (UK). Structures are shown in Figure S1.

Rat PTZ model

Naïve rats (male, CD strain, weight 185–240 g, 6–7 weeks of age from Charles River UK) were acclimatised to the procedure room in their home cages, with food and water available ad libitum. All rats were tail marked and weighed. Rats were dosed twice daily (am and pm) p.o. with either test compound or vehicle (according to group) on days 1–7. On day 8, rats were dosed am, p.o. with either test compound or vehicle 30 min prior to PTZ administration (according to group). Rats were lightly restrained and injected intravenously using a butterfly cannula (size 27G ¾ in) secured to the tail by tape. The needle was attached to a 5 mL syringe prefilled with heparinised PTZ solution (20 mg/mL 0.9% Saline), which was held in the adjustable motor driven perfusion pump. PTZ was infused at a constant rate of 1.0 mL/min to induce seizures. During the infusion, the rats were observed for the onset of myoclonic and forelimb tonic seizures. The latencies (in s) from start of infusion to the appearance of myoclonic and forelimb tonic seizures were recorded. Infusions were stopped at the appearance of forelimb tonic seizures in each animal, up to a cut-off of 120 s. For animals reaching the cut-off, the dose of PTZ in mg/kg infused over the 120 s was calculated as the threshold dose.

The threshold dose in mg/kg for the appearance of clonic and tonic seizure myoclonic and forelimb tonic seizures was calculated using the following formula: Time to seizure (s) x concentration of PTZ (mg/mL) x flow rate (mL/min) x 1000/60 x body weight of animal (g). Animals were culled immediately by concussion of the brain by striking the cranium, followed by decapitation for the collection of blood. Blood was collected in EDTA (ethylenediaminetetraacetic acid), centrifuged and plasma removed. Mean threshold dose ± SEM for each behaviours/seizure were recorded.

Mouse 6 Hz model

Hcar2 constitutive knock out mice (HCA2-KO) and age-matched wild-type were procured and supplied from Taconic Biosciences (Europe). All 6 Hz studies were performed in Canada. 6 Hz testing consisted of the delivery of an electrical stimulus (0.2 ms pulse width, 3 s duration at various stimulus intensities) via corneal electrodes moistened with saline according to standard operating procedures (ECT unit 57 800; Ugo Basile). The effect of treatment on subsequent seizure was noted. Protection was defined as complete absence of the following four behaviours: stun/immobility, forelimb clonus, straub tail, and/or lateral head movement, within 20 s of stimulus delivery. The experiment was terminated immediately once endpoint was met or if no seizure sign was observed within 20 s of stimulus delivery. To assess treatment effects on seizures, CC₅₀ values (current necessary to elicit a psychomotor seizure in 50% of sample population) were determined. Procedures were designed to avoid or minimize discomfort, distress, and pain to the animals in accordance with the principles of the Animal Research Act of Ontario and the guidelines of the Canadian Council on Animal Care (CCAC). Study 1 evaluated the effects of various HCA2 agonists in WT, and studies 2 and 3 evaluated the effects of HCA2 agonists and ketogenic diet respectively, comparing HCA2 WT and KO animals. Detailed study descriptions are provided in Data S1.

Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in , the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY,⁶⁹ and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20.⁷⁰

AUTHOR CONTRIBUTIONS

Conceived the experiments and analyses: Jill Richardson, Guy Higgins, Neil Upton, Peter Massey, Mark Cunningham, Steve Wilson, Joerg Holenz, Arseniy Lavrov, Hong Lin, and Yasuji Matsuoka. Performed the experiments: Peter Massey, Mark Cunningham, Colleen Taylor. Analyzed and interpreted data: Jill Richardson, Guy Higgins, Neil Upton, Peter Massey, and Andrew Brown. Wrote the paper: Jill Richardson, Arseniy Lavrov and Andrew Brown.

ACKNOWLEDGMENTS

The authors thank Zhengyi Wang, Taylor Guo, Daniel Sévin, Nao Iwamoto and Xueying Sun (all GSK) for their significant contributions to the HCA2 project, and Metul Patel (GSK) for assistance in compiling data.

FUNDING INFORMATION

Studies were funded by GlaxoSmithKline.

CONFLICT OF INTEREST STATEMENT

The authors declare the following interests: Jill Richardson, Steve Wilson, Joerg Holenz, Arseniy Lavrov, Hong Lin, Yasuji Matsuoka and Andrew Brown were employees of GlaxoSmithKline at the time these studies were conducted. Guy Higgins, Neil Upton, Peter Massey, Mark Cunningham, and Colleen Taylor conducted studies under contract to GlaxoSmithKline.

DATA AVAILABILITY STATEMENT

Data available on request due to privacy/ethical restrictions.

ETHICS STATEMENT

All animal studies were ethically reviewed by the institution according to local regulations and conducted in accordance with the Animals (Scientific Procedures) Act 1986 or Canadian Council on Animal Care (CCAC) and the GSK Policy on the Care, Welfare and Treatment of Animals. Human tissue was collected with the approval of, and according to the terms specified by, both the South Tees Local Research Ethics Committee protocol (06/Q1003/51) and Newcastle Upon Tyne Hospitals NHS Trust (approval—CM/PB/3707).

References

Wise A, Foord SM, Fraser NJ, et al. Molecular identification of high and low affinity receptors for nicotinic acid. J Biol Chem. 2003;278:9869‐9874.

Feingold KR, Moser A, Shigenaga JK, Grunfeld C. Inflammation stimulates niacin receptor (GPR109A/HCA2) expression in adipose tissue and macrophages. J Lipid Res. 2014;55:2501‐2508.

Zandi‐Nejad K, Takakura A, Jurewicz M, et al. The role of HCA2 (GPR109A) in regulating macrophage function. FASEB J. 2013;27:4366‐4374.

Graff EC, Fang H, Wanders D, Judd RL. Anti‐inflammatory effects of the hydroxycarboxylic acid receptor 2. Metabolism. 2016;65:102‐113.

Chen H, Assmann JC, Krenz A, et al. Hydroxycarboxylic acid receptor 2 mediates dimethyl fumarate's protective effect in EAE. J Clin Invest. 2014;124:2188‐2192.

Rahman M, Muhammad S, Khan MA, et al. The beta‐hydroxybutyrate receptor HCA2 activates a neuroprotective subset of macrophages. Nat Commun. 2014;5:3944.

Fu SP, Wang JF, Xue WJ, et al. Anti‐inflammatory effects of BHBA in both in vivo and in vitro Parkinson's disease models are mediated by GPR109A‐dependent mechanisms. J Neuroinflammation. 2015;12:9.

Auvin S, Cilio MR, Vezzani A. Current understanding and neurobiology of epileptic encephalopathies. Neurobiol Dis. 2016;92:72‐89.

Bertoglio D, Verhaeghe J, Santermans E, et al. Non‐invasive PET imaging of brain inflammation at disease onset predicts spontaneous recurrent seizures and reflects comorbidities. Brain Behav Immun. 2017;61:69‐79.

Bogdanovic RM, Syvanen S, Michler C, et al. (R)‐[11C]PK11195 brain uptake as a biomarker of inflammation and antiepileptic drug resistance: evaluation in a rat epilepsy model. Neuropharmacology. 2014;85:104‐112.

Devinsky O, Vezzani A, Najjar S, De Lanerolle NC, Rogawski MA. Glia and epilepsy: excitability and inflammation. Trends Neurosci. 2013;36:174‐184.

Vezzani A, French J, Bartfai T, Baram TZ. The role of inflammation in epilepsy. Nat Rev Neurol. 2011;7:31‐40.

Lee DH, Stangel M, Gold R, Linker RA. The fumaric acid ester BG‐12: a new option in MS therapy. Expert Rev Neurother. 2013;13:951‐958.

Britton J. Autoimmune epilepsy. Handb Clin Neurol. 2016;133:219‐245.

Tan TH, Perucca P, O'Brien TJ, Kwan P, Monif M. Inflammation, ictogenesis, and epileptogenesis: an exploration through human disease. Epilepsia. 2021;62:303‐324.

Spatola M, Dalmau J. Seizures and risk of epilepsy in autoimmune and other inflammatory encephalitis. Curr Opin Neurol. 2017;30:345‐353.

D'Alonzo R, Rigante D, Mencaroni E, Esposito S. West syndrome: a review and guide for Paediatricians. Clin Drug Investig. 2018;38:113‐124.

Varadkar S, Bien CG, Kruse CA, et al. Rasmussen's encephalitis: clinical features, pathobiology, and treatment advances. Lancet Neurol. 2014;13:195‐205.

Linker RA, Lee DH, Ryan S, et al. Fumaric acid esters exert neuroprotective effects in neuroinflammation via activation of the Nrf2 antioxidant pathway. Brain. 2011;134:678‐692.

Kornberg MD, Bhargava P, Kim PM, et al. Dimethyl fumarate targets GAPDH and aerobic glycolysis to modulate immunity. Science. 2018;360:449‐453.

Brockhoff JD, Bereswill S, Heimesaat MM. The impact of ketogenic diet on the onset and progression of multiple sclerosis. Eur J Microbiol Immunol (Bp). 2023;13:29‐36.

van Delft R, Lambrechts D, Verschuure P, Hulsman J, Majoie M. Blood beta‐hydroxybutyrate correlates better with seizure reduction due to ketogenic diet than do ketones in the urine. Seizure. 2010;19:36‐39.

Schoeler NE, Cross JH, Sander JW, Sisodiya SM. Can we predict a favourable response to ketogenic diet therapies for drug‐resistant epilepsy? Epilepsy Res. 2013;106:1‐16.

Buchhalter JR, D'Alfonso S, Connolly M, et al. The relationship between d‐beta‐hydroxybutyrate blood concentrations and seizure control in children treated with the ketogenic diet for medically intractable epilepsy. Epilepsia Open. 2017;2:317‐321.

Bough KJ, Rho JM. Anticonvulsant mechanisms of the ketogenic diet. Epilepsia. 2007;48:43‐58.

Danial NN, Hartman AL, Stafstrom CE, Thio LL. How does the ketogenic diet work? Four potential mechanisms. J Child Neurol. 2013;28:1027‐1033.

Masino SA, Li T, Theofilas P, et al. A ketogenic diet suppresses seizures in mice through adenosine a(1) receptors. J Clin Invest. 2011;121:2679‐2683.

Rho JM. How does the ketogenic diet induce anti‐seizure effects? Neurosci Lett. 2017;637:4‐10.

Simeone TA, Matthews SA, Samson KK, Simeone KA. Regulation of brain PPARgamma2 contributes to ketogenic diet anti‐seizure efficacy. Exp Neurol. 2017;287:54‐64.

Achanta LB, Rae CD. Beta‐hydroxybutyrate in the brain: one molecule, multiple mechanisms. Neurochem Res. 2017;42:35‐49.

Chang P, Augustin K, Boddum K, et al. Seizure control by decanoic acid through direct AMPA receptor inhibition. Brain. 2016;139:431‐443.

Augustin K, Williams S, Cunningham M, et al. Perampanel and decanoic acid show synergistic action against AMPA receptors and seizures. Epilepsia. 2018;59:e172‐e178.

Pan JW, Rothman TL, Behar KL, Stein DT, Hetherington HP. Human brain beta‐hydroxybutyrate and lactate increase in fasting‐induced ketosis. J Cereb Blood Flow Metab. 2000;20:1502‐1507.

Bough KJ, Eagles DA. A ketogenic diet increases the resistance to pentylenetetrazole‐induced seizures in the rat. Epilepsia. 1999;40:138‐143.

Bough KJ, Matthews PJ, Eagles DA. A ketogenic diet has different effects upon seizures induced by maximal electroshock and by pentylenetetrazole infusion. Epilepsy Res. 2000;38:105‐114.

Hartman AL. Does the effectiveness of the ketogenic diet in different epilepsies yield insights into its mechanisms? Epilepsia. 2008;49(Suppl 8):53‐56.

Dolce A, Santos P, Chen W, Hoke A, Hartman AL. Different ketogenesis strategies lead to disparate seizure outcomes. Epilepsy Res. 2018;143:90‐97.

Somjen GG. Ion regulation in the brain: implications for pathophysiology. Neuroscientist. 2002;8:254‐267.

Mody I, Lambert JD, Heinemann U. Low extracellular magnesium induces epileptiform activity and spreading depression in rat hippocampal slices. J Neurophysiol. 1987;57:869‐888.

Leschinger A, Stabel J, Igelmund P, Heinemann U. Pharmacological and electrographic properties of epileptiform activity induced by elevated K⁺ and lowered Ca2⁺ and Mg2+ concentration in rat hippocampal slices. Exp Brain Res. 1993;96:230‐240.

Frohlich F, Bazhenov M, Iragui‐Madoz V, Sejnowski TJ. Potassium dynamics in the epileptic cortex: new insights on an old topic. Neuroscientist. 2008;14:422‐433.

Huberfeld G, Menendez de la Prida L, Pallud J, et al. Glutamatergic pre‐ictal discharges emerge at the transition to seizure in human epilepsy. Nat Neurosci. 2011;14:627‐634.

Yuen AW, Sander JW. Rationale for using intermittent calorie restriction as a dietary treatment for drug resistant epilepsy. Epilepsy Behav. 2014;33:110‐114.

Dustin SM, Stafstrom CE. Ketogenic diet, but not polyunsaturated fatty acid diet, reduces spontaneous seizures in juvenile rats with kainic acid‐induced epilepsy. J Epilepsy Res. 2016;6:1‐7.

Dressler A, Trimmel‐Schwahofer P, Reithofer E, et al. The ketogenic diet in infants—advantages of early use. Epilepsy Res. 2015;116:53‐58.

Dressler A, Trimmel‐Schwahofer P, Reithofer E, et al. Efficacy and tolerability of the ketogenic diet in Dravet syndrome–comparison with various standard antiepileptic drug regimen. Epilepsy Res. 2015;109:81‐89.

Fu SP, Li SN, Wang JF, et al. BHBA suppresses LPS‐induced inflammation in BV‐2 cells by inhibiting NF‐kappaB activation. Mediat Inflamm. 2014;2014: [eLocator: 983401].

Morris G, Puri BK, Maes M, Olive L, Berk M, Carvalho AF. The role of microglia in neuroprogressive disorders: mechanisms and possible neurotherapeutic effects of induced ketosis. Prog Neuro‐Psychopharmacol Biol Psychiatry. 2020;99: [eLocator: 109858].

Schoeler NE, Cross JH. Ketogenic dietary therapies in adults with epilepsy: a practical guide. Pract Neurol. 2016;16:208‐214.

Liu H, Yang Y, Wang Y, et al. Ketogenic diet for treatment of intractable epilepsy in adults: a meta‐analysis of observational studies. Epilepsia Open. 2018;3:9‐17.

Feng L, Wang J, Li X, Hu Y, Hong S, Jiang L. Prospective control study of efficacy and influencing factors of a ketogenic diet on refractory epilepsy in children. Transl Pediatr. 2022;11:138‐148.

Ruan Y, Chen L, She D, Chung Y, Ge L, Han L. Ketogenic diet for epilepsy: an overview of systematic review and meta‐analysis. Eur J Clin Nutr. 2022;76:1234‐1244.

Kossoff EH, Zupec‐Kania BA, Auvin S, et al. Optimal clinical management of children receiving dietary therapies for epilepsy: updated recommendations of the International Ketogenic Diet Study Group. Epilepsia Open. 2018;3:175‐192.

Martin‐McGill KJ, Bresnahan R, Levy RG, Cooper PN. Ketogenic diets for drug‐resistant epilepsy. Cochrane Database Syst Rev. 2020;6: [eLocator: CD001903].

Pizzo F, Collotta AD, Di Nora A, Costanza G, Ruggieri M, Falsaperla R. Ketogenic diet in pediatric seizures: a randomized controlled trial review and meta‐analysis. Expert Rev Neurother. 2022;22:169‐177.

Zimmern V, Minassian B, Korff C. A review of targeted therapies for monogenic epilepsy syndromes. Front Neurol. 2022;13: [eLocator: 829116].

Olson EJ, Mahar KM, Haws TF, et al. A randomized, placebo‐controlled trial to assess the effects of 8 weeks of administration of GSK256073, a selective GPR109A agonist, on high‐density lipoprotein cholesterol in subjects with dyslipidemia. Clin Pharmacol Drug Dev. 2019;8:871‐883.

Dobbins R, Byerly R, Gaddy R, et al. GSK256073 acutely regulates NEFA levels via HCA2 agonism but does not achieve durable glycaemic control in type 2 diabetes. A randomised trial. Eur J Pharmacol. 2015;755:95‐101.

Lauring B, Taggart AK, Tata JR, et al. Niacin lipid efficacy is independent of both the niacin receptor GPR109A and free fatty acid suppression. Sci Transl Med. 2012;4: [eLocator: 148ra115].

Lukasova M, Malaval C, Gille A, Kero J, Offermanns S. Nicotinic acid inhibits progression of atherosclerosis in mice through its receptor GPR109A expressed by immune cells. J Clin Invest. 2011;121:1163‐1173.

Moutinho M, Puntambekar SS, Tsai AP, et al. The niacin receptor HCAR2 modulates microglial response and limits disease progression in a mouse model of Alzheimer's disease. Sci Transl Med. 2022;14: [eLocator: eabl7634].

Cunnane SC, Courchesne‐Loyer A, St‐Pierre V, et al. Can ketones compensate for deteriorating brain glucose uptake during aging? Implications for the risk and treatment of Alzheimer's disease. Ann N Y Acad Sci. 2016;1367:12‐20.

Camberos‐Luna L, Massieu L. Therapeutic strategies for ketosis induction and their potential efficacy for the treatment of acute brain injury and neurodegenerative diseases. Neurochem Int. 2020;133: [eLocator: 104614].

Koh S, Dupuis N, Auvin S. Ketogenic diet and Neuroinflammation. Epilepsy Res. 2020;167: [eLocator: 106454].

Paoli A, Bianco A, Damiani E, Bosco G. Ketogenic diet in neuromuscular and neurodegenerative diseases. Biomed Res Int. 2014;2014: [eLocator: 474296].

Zhao C, Wang H, Liu Y, et al. Biased allosteric activation of ketone body receptor HCAR2 suppresses inflammation. Mol Cell. 2023;83(17):3171‐3187.

Park JH, Kawakami K, Ishimoto N, et al. Structural basis for ligand recognition and signaling of hydroxy‐carboxylic acid receptor 2. Nat Commun. 2023;14:7150.

Richman JG, Kanemitsu‐Parks M, Gaidarov I, et al. Nicotinic acid receptor agonists differentially activate downstream effectors. J Biol Chem. 2007;282:18028‐18036.

Harding SD, Sharman JL, Faccenda E, et al. The IUPHAR/BPS Guide to pharmacology in 2019: updates and expansion to encompass the new guide to immunopharmacology. Nucleic Acids Res. 2018;46:D1091‐D1106. doi:10.1093/nar/gkx1121

Alexander SPH, Christopoulos A, Davenport AP, et al. The concise guide to PHARMACOLOGY 2023/24: G protein‐coupled receptors. Br J Pharmacol. 2023;180:S23‐S144. doi:10.1111/bph.16177