Neuropsychological functioning and brain energetics of drug resistant mesial temporal lobe epilepsy patients

Epilepsy Research 138 (2017) 26–31

Contents lists available at ScienceDirect

Epilepsy Research journal homepage: www.elsevier.com/locate/epilepsyres

Neuropsychological functioning and brain energetics of drug resistant mesial temporal lobe epilepsy patients

MARK

Camila Moreira Osórioa, Alexandra Latinia,b, Rodrigo Bainy Leala,c, Maria Emília Rodrigues de Oliveira Thaisa, Helena Dresch Vascoutoa, Aline Pertile Remorb, Mark William Lopesb, Marcelo Neves Linharesa,d,e, Juliana Bena,c, Roberta de Paula Martinsa,b, Rui Daniel Predigerf, Alexandre Ademar Hoellera, Hans Joachim Markowitschg, Peter Wolfa,i,j, ⁎ Kátia Lina,h,i, Roger Walza,h,i, a

Centro de Neurociências Aplicadas, Hospital Universitário (HU), Universidade Federal de Santa Catarina (UFSC), Florianópolis, SC, Brazil Laboratório de Bioenergética e Estresse Oxidativo, LABOX, Depar tamento de Bioquímica, UFSC, Florianópolis, SC, Brazil Laboratório de Transdução de Sinal no Sistema Nervoso Central, Departamento de Bioquímica, UFSC, Florianópolis, SC, Brazil d Divisão de Neurocirurgia, Departamento de Cirurgia, HU, UFSC, Florianópolis, SC, Brazil e Serviço de Neurocirurgia, Hospital governador Celso Ramos (HGCR), Florianópolis, SC, Brazil f Departamento de Farmacologia, UFSC, Florianópolis, SC, Brazil g Physiological Psychology, University of Bielefeld, Bielefeld, Germany h Centro de Epilepsia do Estado de Santa Catarina, CEPESC, HU, UFSC, Florianópolis, SC, Brazil i Serviço de Neurologia, Departamento de Clínica Médica, HU, UFSC, Florianópolis, SC, Brazil j Danish Epilepsy Centre, Dianalund, Denmark b c

A R T I C L E I N F O

A B S T R A C T

Keywords: Brain metabolism Cognition Temporal lobe epilepsy Hippocampus

Interictal hypometabolism is commonly measured by 18-fluoro-deoxyglucose Positron Emission Tomography (FDG-PET) in the temporal lobe of patients with mesial temporal lobe epilepsy (MTLE-HS). Left temporal lobe interictal FDG-PET hypometabolism has been associated with verbal memory impairment, while right temporal lobe FDG-PET hypometabolism is associated with nonverbal memory impairment. The biochemical mechanisms involved in these findings remain unknown. In comparison to healthy controls (n = 21), surgically treated patients with MTLE-HS (n = 32, left side = 17) had significant lower scores in the Rey Auditory Verbal Learning Test (RAVLT retention and delayed), Logical Memory II (LMII), Boston Naming test (BNT), Letter Fluency and Category Fluency. We investigated whether enzymatic activities of the mitochondrial enzymes Complex I (C I), Complex II (C II), Complex IV (C IV) and Succinate Dehydrogenase (SDH) from the resected samples of the middle temporal neocortex (mTCx), amygdala (AMY) and hippocampus (HIP) were associated with performance in the RAVLT, LMII, BNT and fluency tests of our patients. After controlling for the side of hippocampus sclerosis, years of education, disease duration, antiepileptic treatment and seizure outcome after surgery, no independent associations were observed between the cognitive test scores and the analyzed mitochondrial enzymatic activities (p > 0.37). Results indicate that memory and language impairment observed in MTLE-HS patients are not strongly associated with the levels of mitochondrial CI, CII, SDH and C IV enzymatic activities in the temporal lobe structures ipsilateral to the HS lesion.

1. Introduction Mesial temporal lobe epilepsy with hippocampal sclerosis (MTLEHS) is the most frequent drug-resistant epilepsy referred for surgery (Fiest et al., 2014; Pauli et al., 2017a, 2017b, 2012; Wiebe et al., 2001). Epilepsy may affect different domains of cognitive functioning

depending on the relationship between epileptogenic and symptomatogenic zone (Rosenow and Lüders, 2001). MTLE-HS is characterized by temporal lobe dysfunction, affecting particularly memory and language (Pauli et al., 2017a; Knopman et al., 2015), but impairments may extend beyond those functions thought to be mediated by the temporal lobe. These include attention, working memory, speed of processing,

⁎ Corresponding author at: Departamento de Clínica Médica, Hospital Universitário, 3 andar, Universidade Federal de Santa Catarina (UFSC), Trindade, Florianópolis, SC CEP 88.040970, Brazil. E-mail address: [email protected] (R. Walz).

http://dx.doi.org/10.1016/j.eplepsyres.2017.10.009 Received 10 March 2017; Received in revised form 20 September 2017; Accepted 9 October 2017 Available online 12 October 2017 0920-1211/ © 2017 Elsevier B.V. All rights reserved.

Epilepsy Research 138 (2017) 26–31

C.M. Osório et al.

or less) and psychiatric diagnoses were excluded. This less rigid criterion for IQ is due to the high prevalence of low IQ in our patients. WADA or fMRI were not performed to asses the hemispheric dominance for language. Clinical variables analyzed were age, years of education, gender, hand dominance, side of the hippocampus sclerosis (HS), duration of epilepsy (in years), age of epilepsy onset (recurrent seizure) and monthly frequency of seizures impairing awareness. Patients who used only one AED were rated as monotherapies. Patients using two or more AEDs, associated or not with benzodiazepines, were classified as being under polytherapy. The benzodiazepines were clobazam or clonazepam. The AEDs were carbamazepine, phenobarbital, diphenylehydatoin, valproic acid, lamotrigine or topiramate. Control subjects matched for gender, hand dominance, age and education level and recruited during the same period of patients were companion persons of patients from other outpatient clinics and had no previous history of neurological or psychiatric disorders.

visuospatial and executive deficits. (Bell et al., 2013; Oyegbile et al., 2004; Sherman et al., 2011) Several studies identified hypometabolism by 18-fluoro-deoxyglucose positron emission tomography (FDG-PET) in the typically epileptogenic areas of MTLE-HS (Akman et al., 2010; Chassoux et al., 2016; Knopman et al., 2015) The degree of hypometabolism in these regions was shown to correlate to some degree with neuropsychological test results (Nickel et al., 2003). Dominant temporal lobe hypometabolism has been associated with verbal memory impairment, while nondominant temporal lobe hypometabolism was associated with nonverbal memory performance (Knopman et al., 2015). Mitochondria are critical modulators of cell function and are recognized as proximal metabolic sensors and effectors (Babcock and Wikström, 1992). The mitochondrial respiratory chain (RC) consists of five enzyme complexes that are distributed in a special way in the inner mitochondrial membrane. The electrons coming from the Krebs cycle and other reactions catalyzed by dehydrogenases are transferred to the RC with molecular oxygen as the final acceptor. Along with this process, there is translocation of protons across the inner mitochondrial membrane and ATP synthesis. (Babcock and Wikström, 1992). The RC occurs due to the presence of four enzymatic complexes. The electrons are transported through these four complexes and the last one reduces O2 to H2O. Respiration begins with the oxidation of fuels in metabolic pathways that transfer electrons to NAD+ and FAD. These coenzymes can come from various metabolic processes including the tricarboxylic acid cycle (TCA). Energy from the reoxidation of NADH and FAD(2H) by O2 is converted to the high energy phosphate bonds of ATP via oxidative phosphorylation (Milane et al., 2015). Epilepsy surgery offers an opportunity for association studies between brain mitochondrial metabolism and clinical variables of patients (Osório et al., 2017; Marcelo Fernando Ronsoni et al., 2016). We identified the neuropsychological tests significantly impaired in our MTLE-HS patients in comparison to healthy controls matched for age and sociocultural characteristics. Thereafter we investigated if the neuropsychological test results showing impairments in patients correlated with the mitochondrial respiratory chain complex enzyme activities analyzed in their brain samples from middle temporal neocortex (mTCx), amygdala (AMY) and hippocampus (HIP), resected during epilepsy surgery. We hypothesized the mitochondrial enzyme activities would be significantly associated with the cognitive performance of our patients. The results serve to understand the biochemical and metabolic mechanisms involved in patients with cognitive impairment after MTLE-HS.

2.2. Anesthesia protocol, surgery and brain tissue sampling The anesthesia protocol, surgical procedures and brain tissue sampling were performed by the same team of neurosurgery as described previously (Lopes et al., 2016; Ronsoni et al., 2016). Surgeries followed the standard procedure for MTLE-HS surgery were a maximum of 4 cm of the anterior lateral temporal lobe including the middle and inferior temporal gyrus. Was ressected (Pauli et al., 2017a, 2017b; Wiebe et al., 2001). The mesial resection included the amygdala and up to 3 cm of the anterior hippocampus. For further biochemical analysis, samples were frozen in liquid nitrogen immediately after collection and then transferred to a −80 °C freezer. 2.3. Tissue homogenization procedure Brain samples were processed 24 (SD ± 9) months after the surgery. After being defrosted, samples were weighed and mechanically homogenized with a ground glass Potter-Elvehjem homogenizer in four volumes (v/v) of 50 mM Tris, pH 7.0, 1.0 mM EDTA, 100 mM NaF, 0.1 mM PMSF, 2.0 mM Na3VO4, 1% Triton X-100, 10% glycerol, and protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA). Homogenates were placed on ice for 10 min and centrifuged at 3000xg at 4 °C for 10 min. Aliquots of the supernatants were used to measure the mitochondrial enzyme activities after thawing the samples three times. Sample preparation was carried out in an Eppendorf 5415 R centrifuge (Eppendorf, Hamburg, Germany).

2. Material and methods 2.4. Mitochondrial respiratory chain complex enzyme activities 2.1. Patients Mitochondrial enzyme activities were measured spectrophotometrically with a Varian Cary 50 spectrophotometer with temperature control (Varian Inc., Palo Alto, CA, USA). Complex I (CI) activities were measured by the rate of NADH-dependent ferricyanide reduction at 420 nm. The activities of succinate-2,6- dichloroindophenol (DCIP)-oxidoreductase (Complex II [C II]), succinate phenazyne oxidoreductase (succinate dehydrogenase [SDH]; Complex II) and cytochrome c oxidase (Complex IV [C IV]) were assayed as previously described elsewhere (Ronsoni et al., 2016). The activities were calculated as nanomol min−1 mg −1 protein. In order to control variations in the mitochondrial mass, the mitochondrial content of Mitofusin1 (Mfn1) was determined by western blot.

Thirty two consecutive adult patients with drug-resistant MTLE-HS were included in the research protocol (265-FR304969) previously described by our group (Lopes et al., 2016; Ronsoni et al., 2016). All patients failed to respond to adequate treatment with at least two antiepileptic drugs in monotherapy and had seizures impairing awareness at least once a month. They were treated surgically between May 2009 and December 2012 at the Centro de Epilepsia de Santa Catarina (CEPESC). All patients had complete medical history, seizure semiology, neurological examination, neuropsychological and psychiatric evaluation, interictal and ictal video-EEG analysis and MRI (1.5 T) findings consistent with unilateral MTLE-HS. We excluded patients with any diagnosis of an epilepsy syndrome other than unilateral MTLE-HS, focal motor or sensory abnormalities on physical examination, and generalized or extra-temporal interictal EEG spikes. (Araújo et al., 2006; de Lemos Zingano et al., 2015; Guarnieri et al., 2009; Pauli et al., 2017a, 2017b, 2012; Velasco et al., 2011). Patients with mental retardation determined by neuropsychological tests (Intelligence Quotient, IQ of 60

2.5. Western blot analysis of Mfn1 Because variations in the histopathologic distribution of neuronal lesions and gliosis in the brain tissue of MTLE-HS patients can influence the mitochondrial content, the mitochondrial mass variation between samples was controlled through the determination of Mfn1 content by 27

Epilepsy Research 138 (2017) 26–31

C.M. Osório et al.

identify cognitive tests that were significantly impaired in our MTLE-HS patients. We also compared the neuropsychological performance among controls, patients with right or left side HS by one-way ANOVA followed by Bonferroni test. For these analysis a “p” level < 0.05 was considered statistically significant. Pearson’s correlation was used to investigate the correlation between the mitochondrial enzyme activities in their brain samples from mTCx, AMY and HIP and the cognitive tests scores that were impaired in the patiens. The analysis was done separately for patients with right or left side HS. Mitochondrial enzymes showing a correlation with cognitive tests with a “p” level < 0.20 were further included in the analysis of covariance (ANCOVA). This level of significance for the variable inclusion in the ANCOVA was adopted empirically. Analysis of Covariance (ANCOVA) was used to determine the effect of the enzymatic activity (independent variable) previously selected (see Pearson’s correlation above) on the cognitive score (dependent variables), adding seizure outcome after surgery, side of HS, seizure outcome after surgery and antiepileptic drug treatment (mono- or polytherapy) as co-factors and disease duration (in years), education level (in years) as covariates. Seizure outcomes were categorized according to Engel classification I, II, III or IV. The initial model shows the effect of all the included independent variables together, and the final model shows only the independent variables that remained significantly associated with the cognitive tests for “p” levels < 0.05. The analysis was performed with the software SPSS 17.0 (Chicago, IL).

western blot. Mfn1 is an integral mitochondrial outer membrane protein that together with Mfn2 (Mitofusin 2), participates in the mitochondrial dynamics, mediating part of the mitochondrial fusion process (Chen et al., 2003). Mfn1 was determined in 24 hippocampi samples that left over from the 32 used in the initiall analysis. Samples were homogenized by the same researcher in the same day, placed in liquid nitrogen, and stored at−80 °C until analysis. Samples were homogenized in 2.5 volumes of 50 mM Tris, pH 7.0, containing 1 mM EDTA, 100 mM NaF, 0.1 mM PMSF, 2 mM Na3VO4, 1% Triton X- 100, 10% glicerol and protease inhibitor cocktail (Roche, Mannheim, Germany). Then, the samples were centrifuged at 10,000×g for 10 min at 4 °C, boiled at 100 °C for 5 min. The protein content was quantified by Lowry’s method (Lowry et al., 1951). Samples were diluted in 25% 100 mM Tris Buffer (pH 6.8, with 40% glycerol and bromophenol blue) and 8% β-mercaptoethanol. An aliquot of 50 μg of total protein was size-separated by electrophoresis in 12% SDS–polyacrylamide gel, under reducing conditions, and transferred to a nitrocellulose membrane. After washing and blocking, the membranes were incubated overnight with anti-Mfn1 (1:2000; Cell Signaling; Danvers, USA) primary antibody. Afterwards, membranes were exposed to the anti-rabbit (1:2000; Cell Signaling; Danvers, USA) secondary antibody. The immune complexes were visualized using the ECL chemiluminescence detection system (GE Healthcare, São Paulo, SP, Brazil). Membranes were stripped and the content of anti-Actin (1:1000; Sigma, Saint Louis, USA) was evaluated to verify loading evenness. In order to prevent intra-day biases of western blot quantification, an internal control (IC) was applied in all electrophoresis. Three pooled hippocampi homogenized in the same way and time as all the studied samples were used as the reference sample (and considered 100%) in all the western blotting analysis as previously described (Lopes et al., 2016). The immunocontent was determined as a ratio of the Optic Density (OD) of the protein band by the OD of the β-actin band. The bands were quantified using the Scion Image software (Frederick, MD, USA). All the persons engaged in the biochemical analyses were blind with respect to all the clinical data.

3. Results The clinical and demographic variables of patients and controls are presented in Table 1. There were no statistical differences between patients and controls according to gender, age, education level and hand dominance. The mean post-operative follow-up was 69 months (SD = 10; range 50–85) and the seizure outcome results after surgery according to the Engel Class were: - Engel 1A (completely seizure free since surgery) = 15 (46.9%); - Engel 1B (Auras only) = 2 (6.3%); - Engel 1C (some seizures impairing awareness after surgery, but none for at least 2 years) = 2 (6.3%);

2.6. Neuropsychological assessments Neuropsychological assessments were done by a neuropsychologist blind for all clinical, neurosurgical and laboratory variables except for the type of epilepsy (MTLE-HS). The average time for neuropsychological assessment was 2 months before surgery (range from 1 to 3 months) and the raw scores of the tests were analyzed. The cognitive assessments included: Letters and Category Fluency, Rey Auditory Verbal Learning Test (RAVLT-Total, RAVLT-Retention, RAVLT-Delayed and RAVLT-Recognized), Wechsler Memory Scale III (VMS-III), subtests Logical Memory First Recall (LM 1st), Logical Memory I (LM I), Logical Memory II (LM II), Wechsler Adult Intelligence Scale III (WAIS-III), subtests Vocabulary, Similarities, Information, Block Design, Digit Span, Matrix Reasoning and Picture Completion, Rey-Osterrieth Complex Figure (ROCF-Copy, ROCFImmediate, ROCF-Delayed) and Boston Naming Test (Suppl. Table 1).

Table 1 Clinical and demographic variables of MTLE-HS patients and controls. Variables

2.7. Ethical aspects The local Research Ethics Committee approved the study (365FR304969) and informed consent was collected from all patients and controls. All procedures performed in the study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. 2.8. Statistical analysis

“p”

Groups All Cases n = 32

Control n = 21

level

Gender Female Male

19 (59.4) 13 (40.6)

12 (57.1) 9 (42.9)

0.87

Hand Dominance Right Non-right

28 (87.5) 4 (12.5)

19 (90.5) 2 (9.5)

0.74

Side of the HSa Right Left

17 (53.1) 15 (46.9)

– –

–

AEDsb treatment Monotherapy Polytherapy

10 (31.3) 22 (68.8)

– –

–

Age (years) Education (years) Disease duration (years) Monthly frequency of seizures

Mean (SEM) 36,70 (2.16) 6.75 (0.54) 24.84 (2.03) 8.48 (1.51)

40.33 (2.93) 7.86 (0.89) – –

0.31 0.24 – –

a

HS, Hippocampus sclerosis. AEDs, Antiepileptic drugs were: carbamazepine (n = 24), phenobarbital (n = 10), diphenylehydantoin (n = 3), Oxcarbazepine (n = 3), Lamotrigine (n = 4), Topiramate (n = 2), Valproic acid (n = 3). Adjuvant benzodiazepines were used in 16 patients. b

The comparisons of neuropsychological test scores between controls (n = 21) and patients (n = 32) were analyzed by Student t-test to 28

Epilepsy Research 138 (2017) 26–31

C.M. Osório et al.

Table 2 Neuropsychological evaluation of patients and controls. Cognitive tests, mean (SE)

Controls n = 21

Patients n = 32

Table 3 Association between mitochondrial respiratory chain enzyme activities and the impaired cognitive tests scores of patients controlling for education level, side of hippocampus sclerosis and seizure outcome after surgery.

“p” Levela

RAVLT Total Retention Delayed Recognized

39.9 (2.2) 8.7 (0.6) 8.9 (0.7) 10.9 (0.8)

35.2 (1.6) 5.9 (0.5) 4.6 (0.6) 9.9 (1.3)

0.08 0.001 < 0.001 0.58

Logical Memory LM 1st recall LM I LM II

19.3 (1.5) 31.5 (1.9) 20.2 (1.3)

16.5 (1.5) 26.4 (2.2) 13.3 (1.7)

0.21 0.11 0.005

ROCF Copy Immediate Delayed

25.9 (1.6) 13.1 (1.3) 14.1 (1.3)

24.7 (1.2) 10.7 (1.1) 11.6 (1.0)

0.52 0.09 0.12

Digit Span Boston Naming Test Letters Fluency Category Fluency Vocabulary Information Similarities Block Design Picture Completion Matrix Reasoning

12.9 (0.9) 47.1 (2.0) 25.8 (1.8) 14.5 (0.8) 22.9 (2.5) 8.3 (1.0) 13.7 (1.9) 21.8 (2.6) 11.9 (1.4) 9.7 (1.6)

11.2 (0.5) 35.0 (1.7) 18.2 (1.6) 11.0 (0.7) 22.1 (1.9) 6.7 (0.7) 14.2 (1.0) 19.8 (1.5) 14.5 (0.9) 7.8 (0.8)

0.08 < 0.001 0.003 0.002 0.86 0.19 0.81 0.50 0.12 0.24

Cognitive tests and predictors

F

“p” levela

RAVLT retention Initial Model Education Age Disease duration Monthly frequency of seizures Side of HS AEDs treatment Seizure outcome Complex II activity in the Hippocampus

1.17 1.81 0.08 6.00 0.88 1.05 0.49 0.73

0.30 0.20 0.78 0.03 0.36 0.38 0.63 0.41

5.37 3.93 5.04

0.03 0.05 0.01

2.52 1.35 0.04 3.08 0.53 1.35 0.62 0.12

0.14 0.27 0.84 0.10 0.48 0.29 0.56 0.74

5.74 3.93

0.02 0.05

1.08 0.74 1.12 0.10 1.63 0.18 0.83 0.01

0.33 0.42 0.32 0.76 0.23 0.84 0.47 0.93

16.84 4.67

< 0.0001 0.04

0.30 0.54 1.03 0.32 1.94 0.09 2.62 0.005 0.01

0.60 0.48 0.34 0.58 0.21 0.91 0.13 0.94 0.91

4.14 15.15 5.04

0.05 0.01 0.01

3.11 0.17 0.38 0.25 0.18 1.13 1.43 0.98

0.12 0.69 0.55 0.63 0.68 0.37 0.29 0.35

Final Model - Adjusted R2 = 0.23 Education Monthly frequency of seizures Seizure outcome after surgery RAVLT delayed Initial Model Education Age Disease duration Monthly frequency of seizures Side of HS AEDs treatment Seizure outcome Complex II activity in the Hippocampus Final Model - Adjusted R2 = 0.24 Education Monthly frequency of seizures

- Engel 1D (seizures impairing awareness only with AED discontinuation) = 4 (12.5%); - Engel II (up to 2 seizures impairing awareness per year) = 4 (12.5%); - Engel III (more than 2 seizures per year but significant improvement) = 5 (15.6%); - Engel IV (no significant improvement or worse) = none.

Logical Memory II Initial Model Education Age Disease duration Monthly frequency of seizures Side of HS AEDs treatment Seizure outcome SDH activity in the amygdala

There was no significant difference (p ≥ 0.14) in the enzymatic activities of Complex I, II, IV and SDH in the neocortex, amygdala and hippocampus according to the brain side of hippocampal sclerosis (Suppl. Fig. 1). The western blot results of Mfn1 content in the analyzed samples of hippocampus are showed in the Suppl. Fig. 2. There was no association between mitochondrial enzymatic activities and the Mfn1 content in the hippocampus, suggesting that the mitochondrial activity was not related to variation on the mitochondrial mass in the tissue analyzed (Suppl. Table 2). Table 2 provides data on the cognitive performance of patients and controls. Patients had significantly lower performance in RAVLT-Retention (p = 0.001) and RAVLT-Delayed (p < 0.001), LM II (p = 0.005), Boston Naming Test (p < 0.001), Letters (p = 0.003) and Category Fluency (p = 0.002) tests. The Suppl. Figs. 3A and B show the comparisons between controls, patients with right and left HS. The Suppl. Tables 2 and 3 show the Pearson correlation between the RAVLT-Retention, RAVLT-Delayed, LM II, Boston Naming Test, Letters and Category Fluency tests scores and the enzymatic activities of the mitochondrial complexes in the analyzed brain structures from right (Supplem. Tables 2) and left (Supplem. Tables 3) temporal lobe, respectively. Table 3 shows the effect of enzymatic activity on the respective cognitive test score, adding seizure outcome after surgery, side of HS, seizure outcome after surgery and antiepileptic drug treatment (monoor polytherapy) as co-factors and disease duration (years) and education level (year) as covariates. Seizure outcome were grouped as Engel I (n = 23), Engel II (n = 4) or Engel III (n = 5). The initial model included all the predictive variables investigated. The final model shows only the variables that remained associated with the cognitive test score

Final Model - Adjusted R2 = 0.40 Education Side of HS Boston Naming test Initial Model Education Age Disease duration Monthly frequency of seizures Side of HS AEDs treatment Seizure outcome SDH in the Temporal Neocortex Complex I activity in the Hippocampusb Final Model - Adjusted R2 = 0.50 Monthly frequency of seizures Seizure outcome Education Category Fluency Initial Model Education Age Disease duration Monthly frequency of seizures Side of HS AEDs treatment Seizure outcome Complex I activity in the Temporal Neocortex Final Model - Adjusted R2 = 0.34 Education

29

3.81 0.05 (continued on next page)

Epilepsy Research 138 (2017) 26–31

C.M. Osório et al.

studies and the reader should be aware of the possibility of type II error results in our analyses. Although MTLE-HS is a relatively uniform epileptic syndrome, there are considerable variations in the histopathologic distribution of neuronal lesions, gliosis and the connectivity of mTCx, HIP and AMY that may result in mitochondrial content variation across the analyzed samples. The imbalances in cellular/structural distribution (and consequently mitochondrial content) were partially controlled by corrections for the protein amount of each analyzed sample and Mfn1 content in the hippocampus. However, compensatory mechanisms may differ according to the sub-regions or cellularity distribution that may not be differentiated when the whole sample is homogenized for western blot analysis. We would also like to emphasize as positive aspects of our study i) the data collection was done in a prospective manner and by using a previously well-established protocol; ii) biochemical analyses were done blind for the neuropsychological evaluation and clinical parameters; iii) extensive controls were applied with respect to surgical and anesthetic parameters during brain sample collection (Lopes et al., 2016; Ronsoni et al., 2016). Our findings indicate that memory and language impairment observed in MTLE-HS patients is not strongly associated with the levels of mitochondrial C I, C II, SDH and C IV enzymatic activities in the temporal lobe structures ipsilateral to the HS lesion.

Table 3 (continued) Cognitive tests and predictors Seizure outcome

F

“p” level

7.74

< 0.002

a

a

Level of significance determined by ANCOVA analysis. Because the activity of Complex II and II in the Hippocampus were correlated (data no shown), to avoid the collinearity only Complex II was included in the ANCOVA to determine variables affecting the Boston Naming Test score. b

(p < 0.05). No significant effect of mitochondrial enzymes activity was observed on the cognitive tests analyzed. 4. Discussion This is the first report on data adressing associations between cognitive functioning and mitochondrial energetics in human samples from neocortical and limbic structures of MTLE-HS patients. As previously demonstrated in patients from Brazil (Castro et al., 2013) and other countries (Castro et al., 2013; Knopman et al., 2015), drug resistant MTLE-HS patients showed lower performances in language and memory (verbal and non-verbal) tests than controls. However, contrary to our initial hypothesis, the impaired cognitive tests were not associated with Complex I, II, IV and SDH enzymatic activities, analyzed in the temporal lobe structures of our patients. In fact, cognitive impairment in temporal lobe epilepsy probably results from brain network disruptions rather than from localized damage. (Bell et al., 2013). The network interactions may be particularly complex as demonstratas showed in Brazilian patients with unilateral HS in which the non-lateralizing profiles of selective verbal and nonverbal memory deficits related to left and right HS (Castro et al., 2013). The exact mechanisms involved are not clear, but we speculate that there are metabolic implications for those findings, including mitochondrial enzyme activities modulation in widespread brain regions other than the temporal lobe. Our results are complementary to previous studies about FDG-PET temporal metabolism and memory performance in MTLE-HS. (Knopman et al., 2015). Considering the high glucose affinity by the hexokinase and its maximal activity under physiologic glucose plasmatic levels conditions (Wilson, 1997), an enhancement in the Complex I, II, IV and SDH enzymatic activities has no implications on the FDG uptake. However, an impairment in the analyzed mitochondrial respiration enzymes activity could result in Glucose-6-phosphate accumulation, inhibiting the hexoquinase, and resulting in less FDG phosphorylation (Wilso, 1985). Taken together, our results suggest that the FDG-PET hypometabolism associated with memory impairment in MTLE-HS patients is not strongly related to variations in the Complex I, II, IV and SDH enzymatic activities in the temporal lobe structures ipsilateral to the HS. Although the observed enzymatic activities determined “in vitro” do not represent the absolute values expected “in vivo”, we believe the proportion of variation observed “in vitro” estimates the same “in vivo” variation among the patients and is therefore acceptable for our purpose of aur association study design. Considering the kinetic modeling of brain FDG uptake (Morris et al., 2004), changes in the blood-brainbarrier, regional distribution of neuronal and glial cellularity and network may affect the temporal lobe physiology an the FDG-PET imaging in MTLE-HS patients (Knopman et al., 2015). Interestingly, using the same sample of patients of the present study we demonstrated that Complex I, II, IV and SDH enzymatic activities did not differ with respect to the presence or absence of Axis I psychiatric diagnosis (Osório et al., 2017) In addition, previous authors did not find any association between FDG-PET metabolism in temporal lobe structures of MTLE-HS and presence of a diagnosis of depression (Osório et al., 2017; Salzberg et al., 2006). However, there are some limitations to be considered as well: The relatively small sample size is a limitation inherent to enzymology

Acknowledgments This work was supported by PRONEX Program (Programa de Núcleos de Excelência - NENASC Project) of FAPESC-CNPq-MS, Santa Catarina Brazil (Grant 56802/2010). Prof. Dr. Hans J. Markowitsch (Process 406929/2013-0) and Prof. Dr. Peter Wolf (Process 88881.030478/2013-01) are Special Visitor Professors (PVE) supported by The Science Without Borders Program Project from MEC/MCTI/ CAPES/CNPq/FAPs. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.eplepsyres.2017.10.009. References Akman, C.I., Ichise, M., Olsavsky, A., et al., 2010. Epilepsy duration impacts on brain glucose metabolism in temporal lobe epilepsy: results of voxel-based mapping. Epilepsy Behav. 17, 373–380. Araújo, D., Santos, A.C., Velasco, T.R., et al., 2006. Volumetric evidence of bilateral damage in unilateral mesial temporal lobe epilepsy. Epilepsia 47, 1354–1359. Babcock, G.T., Wikström, M., 1992. Oxygen activation and the conservation of energy in cell respiration. Nature 356, 301–309. Bell, B., Lin, J.J., Seidenberg, M., et al., 2013. The neurobiology of cognitive disorders in temporal lobe epilepsy. Nat. Rev. Neurol. 7, 154–164. Castro, L.H., Silva, L.C.A.M., Adda, C.C., et al., 2013. Low prevalence but high specificity of material-specific memory impairment in epilepsy associated with hippocampal sclerosis. Epilepsia 54, 1735–1742. Chassoux, F., Artiges, E., Semah, F., et al., 2016. Determinants of brain metabolism changes in mesial temporal lobe epilepsy. Epilepsia 57, 907–919. Chen, H., Detmer, S.A., Ewald, A.J., et al., 2003. Mitofusins Mfn1 and Mfn2 coordinately regulate mitochondrial fusion and are essential for embryonic development. J. Cell Biol. 160, 189–200. de Lemos Zingano, B., Guarnieri, R., Diaz, A.P., et al., 2015. Validation of diagnostic tests for depressive disorder in drug-resistant mesial temporal lobe epilepsy. Epilepsy Behav. 50, 61–66. Fiest, K.M., Sajobi, T.T., Wiebe, S., 2014. Epilepsy surgery and meaningful improvements in quality of life: results from a randomized controlled trial. Epilepsia 55, 886–892. Guarnieri, R., Walz, R., Hallak, J.E.C., et al., 2009. Do psychiatric comorbidities predict postoperative seizure outcome in temporal lobe epilepsy surgery? Epilepsy Behav. 14, 529–534. Knopman, A.A., Wong, C.H., Stevenson, R.J., et al., 2015. The relationship between neuropsychological functioning and FDG-PET hypometabolism in intractable mesial temporal lobe epilepsy. Epilepsy Behav. 44, 136–142. Lopes, M.W., Leal, R.B., Guarnieri, R., et al., 2016. A single high dose of dexamethasone affects the phosphorylation state of glutamate AMPA receptors in the human limbic system. Transl. Psychiatry 6, e986. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275.

30

Epilepsy Research 138 (2017) 26–31

C.M. Osório et al.

improvement in quality of life after temporal lobe epilepsy surgery. A prospective study. Epilepsia 58, 755–763. Ronsoni, M.F., Remor, A.P., Lopes, M.W., et al., 2016. Mitochondrial respiration chain enzymatic activities in the human brain: methodological implications for tissue sampling and storage. Neurochem. Res. 41, 880–891. Rosenow, F., Lüders, H., 2001. Presurgical evaluation of epilepsy. Brain 124, 1683–1700. Salzberg, M., Taher, T., Davie, M., et al., 2006. Depression in temporal lobe epilepsy surgery patients: an FDG-PET study. Epilepsia 47, 2125–2130. Sherman, E.M.S., Wiebe, S., Fay-Mcclymont, T.B., et al., 2011. Neuropsychological outcomes after epilepsy surgery: systematic review and pooled estimates. Epilepsia 52, 857–869. Velasco, T.R., Wichert-Ana, L., Mathern, G.W., et al., 2011. Utility of ictal single photon emission computed tomography in mesial temporal lobe epilepsy with hippocampal atrophy: a randomized trial. Neurosurgery 68, 431–436. Wiebe, S., Blume, W.T., Girvin, J.P., Eliasziw, M., Effectiveness and Efficiency of Surgery for Temporal Lobe Epilepsy Study Group, 2001. A randomized, controlled trial of surgery for temporal-lobe epilepsy. N. Engl. J. Med. 345, 311–318. Wilso, J.E., 1985. Regulation of mammalian hexokinase activity. In: Beitner, R. (Ed.), Regulation of Carbohydrate Metabolism. CRC Press, Boca Raton, pp. 45–85. Wilson, J.E., 1997. An introduction to the isoenzymes of mammalian hexokinase types IIII. Biochem. Soc. Trans. 25, 103–107.

Milane, L., Trivedi, M., Singh, A., Talekar, M., Amiji, M., 2015. Mitochondrial biology, targets, and drug delivery. J. Control. Release 207, 40–58. Morris, E.D., Endres, C.J., Schmidt, K.C., Christian, B.T., Muzic Jr., R.F., Fisher, R., 2004. Kinetic modeling in positron emission tomography. In: Wernick, M.N., Aarsvold, J. (Eds.), The Fundamentals of PET and SPECT. Elsevier Inc., pp. 499–540. Nickel, J., Jokeit, H., Wunderlich, G., Ebner, A., Witte, O.W., Seitz, R.J., 2003. Genderspecific differences of hypometabolism in mTLE: implication for cognitive impairments. Epilepsia 44, 1551–1561. Osório, C.M., Lin, K., Guarnieri, R., et al., 2017. Mitochondrial respiratory chain complex enzyme activities of limbic structures and psychiatric diagnosis in temporal lobe epilepsy patients: preliminary results. CNS Neurosc. Ther. 23, 700–702. Oyegbile, T.O., Dow, C., Jones, J., et al., 2004. The nature and course of neuropsychological morbidity in chronic temporal lobe epilepsy. Neurology 62, 1736–1742. Pauli, C., de Oliveira Thais, M.E.R., Claudino, L.S., et al., 2012. Predictors of quality of life in patients with refractory mesial temporal lobe epilepsy. Epilepsy Behav. 25, 208–213. Pauli, C., de Oliveira Thais, M.E.R., Guarnieri, R., Schwarzbold, M.L., Diaz, A.P., Ben, J., Linhares, M.N., Markowitsch, H.J., Wolf, P., Wiebe, S., Lin, K., Walz, R., 2017a. Decline in word-finding: the objective cognitive finding most relevant to patients after mesial temporal lobe epilepsy surgery. Epilepsy Behav. 58, 218–224. Pauli, C., Schwarzbold, M.L., Diaz, A.P., et al., 2017b. Predictors of meaningful

31