Ketone Body Rescued Seizure Behavior Of LRP1 De ciency By Modulating Glutamate Transport

Jin-Ming Zhang Second A liated Hospital of Guangzhou Medical University Ming-Jie Chen The Third Medicine School of Guangzhou Medical University Jiong-Hui He The Third Medicine School of Guangzhou Medical University Ya-Ping Li Second A liated Hospital of Guangzhou Medical University Zhi-Cai Li The First Clinical Medicine School of Guangzhou Medical University Zi-Jing Ye Second A liated Hospital of Guangzhou Medical University Yong-Hui Bao Guangzhou Medical University Bing-Jun Huang Guangzhou Medical University Wen-Jie Zhang Guangzhou Medical University Ping Kwan University of Sydney Yu-Ling Mao The Third A liated Hospital of Guangzhou Medical University Jing-Da Qiao (  joaquinqjd@163.com ) Second A liated Hospital of Guangzhou Medical University https://orcid.org/0000-0002-4693-8390


Introduction
Epilepsy is a common neurological disorder affecting all age groups and is related to seizures, which are abnormal discharge of brain neurons. It is one of the transient dysfunctions of the brain that can occur in sudden and can be recurring [1]. Recent studies have shown that abnormal brain energy metabolism is closely related to the onset of epilepsy [2]. Myoclonic epilepsy with ragged-red bers (MERRF), affecting complex I of the electron transport chain, is one of the examples concerning mitochondrial DNA (mtDNA) mutations that result in epileptic phenotypes [3]. Glucose is the predominant energy source in the brain [4]. The imbalance of glucose can lead to seizures [2,[5][6][7]. However, the mechanisms underlying glucose regulation and altered neuronal excitability remain poorly understood [8]. LRP1, the low-density lipoprotein receptor 1, has more than 40 different ligands and participates in many physiological processes. In the central nervous system (CNS), LRP1 is ubiquitously expressed and serves as a critical transport receptor as well as a modulator of several distinct signaling pathways in neurons [9], astrocytes [10], and microglia. It has been demonstrated that LRP1 is involved in nerve excitability. In neurons, the deletion of LRP1 caused hyperactivity, dystonia [9]. In astrocytes, early astrocyte dysfunction caused by LRP1 deletion is an important factor leading to abnormal excitability and morphologic changes of epileptic seizures during brain development in mice [11]. In embryonic cortical radial glia stem cells located in the telencephalon, the deletion of LRP1 caused a severe epileptic phenotype [9]. However, the mechanism of LRP1 induced epilepsy remains unknown.

Page 4/16
The ketogenic diet (KD) is a high-fat, calorie-restricted diet used to treat childhood epilepsies that do not respond to available drugs. Despite its clinical use for nearly 100 years, how the KD controls seizures remains unknown [12]. KD was used to regulate energy metabolism in order to manage epilepsy [13], beta-hydroxybutyrate is the major bioactive metabolite [14,15]. In this study, we are aimed to explore the role of LRP1 in Epilepsy and whether KD can treat epilepsy caused by LRP1 knockdown by using betahydroxybutyrate. Our results highlighted that the knockdown of LRP1 can cause epilepsy and abnormal brain structure, and we also found that LRP1 induced epilepsy is associated with glutamate imbalance and KD can rescue LRP1 induced epilepsy.

Data sources
To explore novel epilepsy-related genes by bio-information program., we downloaded the gene expression pro les acquired by high-throughput sequencing of GSE139914 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE139914) and GSE134697 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE134697) from the GEO database (https://www.ncbi.nlm.nih.gov/geo/) after searching for keywords related to epilepsy. Then the gene expression matrix data of the epilepsy group and the control group were imported into GraphPad Prism (version 7.0.0) for t-tests. P < 0.05 was used to indicate a signi cant difference.

Fly stocks
Drosophila melanogaster (Fruit y) is a classical animal model for epilepsy studies [16]. It had been used to investigate the pathogenic genes for epilepsy and other neuro-development disorders [17,18]. Recently, we also used Drosophila as a model to discover the novel epilepsy gene-UNC13B [19]. Thus, we used Drosophila as a model here to study the role of LRP1 in epilepsy.

Seizure behavior test
Stress sensitivity or bang sensitivity (BS) experiments to assess stress-induced seizures were conducted as originally described by Ganetzky and Wu [21] with some modi cations that have been described in detail previously [22]. The bang-sensitive (BS) test was conducted on ies 3-5 days after eclosion and seizure-like behavior was assessed [16,23]. Flies were anesthetized with CO 2 and were transferred to new clean food vials one day before testing. About two to ve ies were collected in empty plastic y vials (3-6 ies/vial). Flies were allowed to recover for at least 30 min and mechanically stimulated with a vortex mixer (VWR, Radnor, PA, USA) at maximum speed for 10 second. The time for each y to recover, which de ned as standing, was scored, and a mean value from the scoring was taken for each vial. One vial was calculated as one N number in each group.

Three types of Diet
Standard medium: 7.12g Cornmeal, 1.68g of yeast powder, 0.56g AGAR, 8.5g sucrose, and 100ml D-D water, was blended and heated to the appropriate temperature (about 100 ℃). After that, 0.47ml propionic acid was added.
Ketone Body medium: 3mM 0.04g beta-HB was added to the standard medium.

Statistical analysis
All quantitative data are presented as mean±S.D. The Student's t-test was used to compare 2 independent or paired samples. One-way ANOVA was used to compare multiple samples, and Tukey's posthoc test was used to evaluate differences between two groups. Statistical analyses were performed with GraphPad Prism 7.00 and SPSS 20. The cutoff value for statistical signi cance is P<0.05.

LRP1 Gene was Low Expression in Epilepsy Patients
GSE139914 contained 6 epilepsy samples and 41 controls from the human Brodmann Area 38. Samples of GSE134697 were from Neocortex and consisted of 17 epilepsy samples and 2 controls. With P <0.05 as the threshold, we found that LRP1 expression was statistically signi cantly higher in the Epilepsy group than that in the control group. (Figure 1).

Global Knockdown of LRP1 Induced Seizure Behavior
To further con rm the role of LRP1 in seizure, we examined BS seizure-like behavior in tub-Gal4>LRP1-RNAi LRP1 ies. The tub-Gal4>LRP1-RNAi ies exhibited the typical seizure-like behavior [24]. About 30.72% of tub-Gal4>LRP1-RNAi ies showed obvious seizure-like behavior, which was higher than the rate in LRP1-RNAi control ies (

Ketone body can Reduce Seizure-Like Behavior in LRP1 Defect Flies
To further explore the effects of the ketogenic diet and high glucose diet on the incidence of epilepsy in tub-Gal4>LRP1-RNAi LRP1 knockdown ies and WT ies (Canton-S), we performed a bang-sensitive (BS) test to different diet application groups. About 30.72% of tub-Gal4>LRP1-RNAi ies feeding Standard food showed obvious seizure-like behavior, which was higher than the rate in tub-Gal4>LRP1-RNAi ies feeding Ketone body (beta-hydroxybutyrate, BHB) food (30.72%±13.28% [n=10] vs 2.08%±5.89% [n=8]; P<0.001) (Figure. 2C). The Canton-s ies feeding high sucrose food also had a higher rate of seizures than those feeding Standard food (

Glial LRP1 Defect is Su cient to Induce Seizure Behavior and sensitive to Ketone body
It is well known that glial LRP1 performs in regulating energy homeostasis, so we speculate that the expression of LRP1 in glia is closely related to epilepsy. Therefore, we established repo-Gal4>LRP1-RNAi LRP1 ies and elav-Gal4>LRP1-RNAi ies. In the same way, we analyzed the BS percentage of the three models. We found that about 17.94% of repo-Gal4>LRP1-RNAi ies showed obvious seizure-like behavior, which was higher than the rate in elav-Gal4>LRP1-RNAi LRP1 knockdown ies These results indicate that the incidence of epileptic behavior in Drosophila by repo-Gal4>LRP1-RNAi LRP1 knockdown can be further increased, which further suggests that not only is epilepsy caused by astrocytes closely related to LRP1 but also there are other ways causing epilepsy induced by LRP1 knockdown, suggesting a new research direction for us. To our surprise, contrasting with tub-Gal4>LRP1-RNAi ies feeding high sucrose diet previously mentioned, the elav-Gal4>LRP1-RNAi ies feeding high sucrose food had a higher rate of seizures than that feeding standard food (30.75%±15.93% [n=5] vs 4.60%±8.46% [n=21]; ****P<0.001), which is worth further discussion. (Figure. 4B). The same as tub-Gal4>LRP1-RNAi ies feeding KD, the repo-Gal4>LRP1-RNAi ies feeding standard food also had a higher rate of seizures than that feeding BHB ( knockdown ies could partially rescue (29.69% ± 9.35%, n=9 v.s. 9.06% ± 5.44%, n=7, ***p=0.0001) ( Figure. 5B). It suggested that downregulated Eaat1 was the mechanism of LRP1 defect causing seizure and the ketogenic diet could partially rescue it by up-regulating EAAT1.

Discussion
Epilepsy is a chronic brain disease that affects approximately 65 million people worldwide [25]. Therefore, it is important to understand the underlying mechanisms for epilepsy, which can improve our management to this disease. In our present study, the bioinformation results indicated that LRP1 would be a potential epilepsy gene. Animal experiments con rmed that LRP1 defects could induce seizures and abnormal brain structure. Additionally, we found that KD is a speci c treatment with excellent outcomes, through the modulation of glutamate transporters, for epilepsy induced by LRP1 de ciency.
LRP1 is plays a role in neural excitability. Here we used bioinformatic and y models to further con rm its important role in Epilepsy. Our ies' data indicated that LRP1 defect could induce seizure behavior and abnormal brain structure, which is consistent with the previous mouse study [26].
It has been reported that LRP1 is closely related to glucose homeostasis. Neuronal LRP1 de ciency impairs insulin signal transduction and downregulates GLUT3 expression in neurons, leading to reduced glucose uptake. [27]. In astrocytes, LRP1 mediates the molecular interaction between insulin-like growth factor 1 receptor (IGF1-R) and GLUT1 [28]. Glucose imbalance leads to seizures, and LRP1 plays an important role in maintaining glucose homeostasis. Thus, we use high sucrose and KD to feed the ies. Our data demonstrated that high sucrose diet can induce epilepsy in wild type ies, consistent with previous studies that high glucose concentrations provoked seizure in the adult rat model [8]. However, to our surprise, LRP1 de ciency in global cells did not show a higher seizure rate in high sucrose feeding. This suggested the presence of another mechanism that LRP1 de ciency could results in further glucose accumulation, which meant that great seizure susceptibility could not be further aggravated by high sucrose. Further study required to demonstrate this hypothesis.
In this study, we found that LRP1 de ciency can rescue epilepsy in a ketone body-speci c manner. The ketone body triggers a systemic shift from glucose metabolism toward the metabolism of fatty acids yielding ketone bodies, such as acetoacetate and BHB as substrates for energy [29]. Ketones, unlike glucose, are not likely deliver an immediate and large amount of energy necessary to initiate or sustain seizure activity [30]. But in our case of LRP1 de ciency, the BHB should not only affected by the glucose metabolism although high sucrose feeding did not signi cantly affect LRP1 knockdown ies ( gure 2C). Instead, BHB can rescue seizure in LRP1 de ciency by modulating synaptic glutamate cycling. Our behavior data showed that overexpression of glutamate transport gene EAAT1 could reduce the seizure rate of LRP1 de ciency ies, and BHB could not decrease seizure rate in EAAT1 knockdown ies. This set of data suggested that glutamate transporter would be one of the targets of KD for epilepsy. Additionally, in a mouse study, LRP1 de ciency result in NMDA receptor decrease [26]. These evidences suggested that KD may rescue epilepsy in LRP1 de ciency by modulating synaptic glutamate cycling.
In summary, we found that LRP1 can be a potential novel epilepsy gene. Screening for LRP1 mutations can identify individuals who require clinical attention for possibly frequent daily seizures. And we rstly found that BHB showed a rescue effect in the LRP1 de ciency model by modulating glutamate transport, which may help to understand the KD mechanism and establish precise treatment toward epilepsy case by case.

Statements & Declarations
Jin-Ming Zhang and Jing-Da Qiao designed this study, Ya-Ping Li performed the bioinformatic experiment, all the authors performed the animal experiments and wrote the manuscript.

Data availability
Raw data were generated at Institute of Neuroscience, The Second A liated Hospital of Guangzhou Medical University. Derived data supporting the ndings of this study are available from the corresponding author on request.

Ethics approval
No ethical approval is required in this study.

Consent to participate
Not applicable Consent to publish Not applicable    Glial knockdown of LRP1 su ciently induced Seizure behavior. (A) Seizures occurred at a higher rate in tub-Gal4>LRP1-RNAi ies than in repo-Gal4>LRP1-RNAi ies. The repo-Gal4>LRP1-RNAi ies had a higher rate of seizures than of the elva-Gal4>LRP1-RNAi group. (B) The pan-neural LRP1 knockdown (elav-Gal4>LRP1-RNAi) ies feeding high sucrose food had a higher rate of seizures than that feeding standard food. Ketone body diet completely inhibited the seizure of those ies. (C) Seizures occurred at a higher rate in glial knockdown (repo-Gal4>LRP1-RNAi) ies feeding standard food than that feeding for ketone body diet.