Figure Legends
Figure 1: Histology and immunohistochemistry reveal features of hippocampal sclerosis. Photomicrographs showing histology of the MTLE-HS hippocampal samples and non-seizure control tissue sections. Hematoxylin & Eosin stained MTLE-HS hippocampal specimens show (a) loss of neurons, (b) highlighted by NeuN, (c) GFAP shows reactive astrogliosis (arrow heads). Autopsy control specimens show (d) no loss of neurons in H & E stained sections, (e) supported by NeuN IHC, (f) no reactive gliosis as revealed in GFAP IHC.
Figure 2: Quantitative estimation of tryptophan-kynurenine pathway metabolites. (a) concentration of tryptophan (autopsy control, n=6, MTLE-HS, n=20; 1.068 ± 0.28 ng/mg of wet tissue in autopsy control vs 0.086 ± 0.013 ng/mg of wet tissue in MTLE-HS), (b) concentration of kynurenine (autopsy control, n=10, MTLE-HS, n=43; 2.895 ± 1.05 ng/mg of wet tissue in autopsy control vs 1.74 ± 0.57 ng/mg of wet tissue in MTLE-HS), (c) tryptophan kynurenine ratio (autopsy control, n=6, MTLE-HS, n=20; 1.58 ± 0.80 in autopsy control vs 0.05 ± 0.009 in MTLE-HS), (d) concentration of kynurenic acid (autopsy control, n=10, MTLE-HS, n=43; 0.184 ± 0.046 ng/mg of wet tissue in autopsy control vs 0.0289 ± 0.0039 ng/mg of wet tissue in MTLE-HS), (e) concentration of quinolinic acid (autopsy control, n=10, MTLE-HS, n=14; 11.67 ± 2.98 ng/mg of wet tissue in autopsy control vs 39.22 ± 9.23 ng/mg of wet tissue in MTLE-HS), (f) quinolinic acid kynurenic acid ratio (autopsy control, n=10, MTLE-HS, n=14; 90.34 ± 28.97 in autopsy control vs 242.0 ± 20.25 in MTLE-HS), (g) concentration of glutamate (autopsy control, n=10, MTLE-HS, n=12; 12.08 ± 3.01 ng/µg protein in autopsy control vs 10.57 ± 4.7 ng/µg protein in MTLE-HS), (h) concentration of GABA (control, n=10, MTLE-HS, n=12; 1.18 ± 0.32 ng/µg protein in autopsy control vs 1.15 ± 0.16 ng/µg protein in MTLE-HS), (i) glutamate GABA ratio (autopsy control, n=10, MTLE-HS, n=12; 15.34 ± 4.98 in autopsy control vs 8.46 ± 2.02 in MTLE-HS) in autopsy control samples and hippocampal samples obtained patients with MTLE-HS. (j) Concentration ofde novo synthesis of kynurenic acid (non-seizure control, n=18, MTLE-HS, n=26; 5.28 ± 1.13 ng/mg of wet tissue in non-seizure control vs 0.10 ± 0.03 ng/mg of wet tissue in MTLE-HS) and (k) concentration ofde novo synthesis of quinolinic acid (non-seizure control, n=12, MTLE- MTLE-HS, n=12; 8.66 ± 2.28 ng/mg of wet tissue in non-seizure control vs 22.33 ± 7.47 ng/mg of wet tissue in MTLE-HS) in non-seizure control samples and hippocampal samples obtained patients with MTLE-HS. Data are presented as a mean ± SEM. * denotes p<0.05, ** p<0.01, *** p<0.001. Mann Whitney Test.
Figure 3: Spontaneous excitatory postsynaptic currents were enhanced in MTLE-HS hippocampal samples but spontaneous inhibitory postsynaptic currents were unaffected. (a) Sample recordings of spontaneous EPSCs recorded from pyramidal neurons in the cortical sample obtained from non-seizure controls. Inset shows a single EPSC event at an expanded time scale. The second trace shows absence of any spontaneous EPSCs following perfusion of the slice with glutamate receptor antagonists APV (50 µM) and CNQX (10 µM) for 10-min proving that these events are glutamate receptor. The bottom trace shows sample recordings of spontaneous EPSCs recorded from pyramidal neurons in the hippocampal sample obtained from patients with MTLE-HS. Plots represent data from eight neurons from eight patients for non-seizure control, nine neurons from the hippocampal samples of nine MTLE-HS patients. (b) In the hippocampal samples, cumulative distribution of inter-event interval displaced toward lower intervals and (c) that of peak amplitude displaced towards longer amplitude. (d) Sample recordings of spontaneous IPSCs recorded from pyramidal neurons in the cortical sample obtained from non-seizure controls and the hippocampal samples obtained from patients with MTLE-HS. The bottom trace shows absence of any spontaneous IPSCs following perfusion of the slice with GABAAreceptor antagonist bicuculline (10 µM) for 10-min proving that these events are GABAA receptor mediated. Plots represent data from five neurons from five patients for non-seizure control, five neurons from the hippocampal samples of five MTLE-HS patients. (e) Cumulative distribution of inter-event interval and (f) peak amplitude of the IPSCs also did not alter. (g) Frequency and amplitude ratio of EPSC/IPSC both were significantly higher in the hippocampal samples. Data represented as mean ± SEM. *p<0.01, **p<0.01, ***p<0.001. Two tailed unpaired T-test/ Mann Whitney Test.
Figure 4: Spontaneous excitatory postsynaptic currents were not suppressed after incubation with 200 µM kynurenine in the hippocampal samples obtained from patients with MTLE-HS. (a) Sample recordings of spontaneous EPSCs recorded from pyramidal neurons in cortical sample obtained from non-seizure controls before and after incubation with200 µM kynurenine for 85 mins. Plots represent data from six neurons from six patients for non-seizure control. (b) Cumulative distribution of inter-event interval displaced towards longer intervals while that of (c) peak amplitude displaced towards shorter amplitude. (d) Sample recordings of spontaneous EPSCs recorded from pyramidal neurons in hippocampus sample obtained from MTLE-HS patients without and with 200 µM kynurenine. Plots represent data from sixteen neurons from sixteen patients for MTLE-HS. (e) Cumulative distribution of inter-event interval and (f) peak amplitude did not displace. (g) Percentage reduction of frequency was significantly reduced in the hippocampal samples after incubation with 200 µM kynurenine. Data represented as mean ± SEM. ***p<0.001. Two tailed unpaired T-test/ Mann Whitney Test.
Figure 5: Spontaneous excitatory postsynaptic currents were suppressed after 30 mins of perfusion with 10 µM kynurenic acid in the hippocampal samples obtained from patients with MTLE-HS. (a) Sample recordings of spontaneous EPSCs recorded from pyramidal neurons in cortical sample obtained from non-seizure controls without and with 10 µM kynurenic acid. Plots represent data from six neurons from six patients for non-seizure control. (b) Cumulative distribution of inter-event interval displaced towards longer intervals while that of (c) peak amplitude displaced towards shorter amplitude. (d) Sample recordings of spontaneous EPSCs recorded from pyramidal neurons in the hippocampus sample obtained from patients with MTLE-HS without and with 10 µM kynurenic acid. Plots represent data from ten neurons from ten patients for MTLE-HS. (e) Cumulative distribution of inter-event interval displaced towards longer intervals while that of (f) peak amplitude displaced towards shorter amplitude. (g) Percentage reduction of frequency was significantly increased in the hippocampal samples after perfusion with 10 µM kynurenic acid. (h) Perfusion with 50 µM APV significantly reduced frequency of spontaneous excitatory postsynaptic currents in both non-seizure controls and hippocampal samples (n=6; 0.98 ± 0.018 Hz without APV vs 0.41 ± 0.25 Hz with APV). (i) Percentage reduction of frequency was significantly increased in the hippocampal samples after perfusion with 50 µM APV respectively. (j) Within the hippocampal samples, percentage reduction of frequency was significantly increased after kynurenic acid treatment in comparison to kynurenine incubation and after APV treatment in comparison to kynurenic acid. Data represented as mean ± SEM. ***p<0.001, #p<0.05. Two tailed unpaired T-test/ Mann Whitney Test/one way ANOVA with Dunn’s posthoc test.
Figure 6: Altered enzyme expression levels of tryptophan-kynurenine pathway contributes to reduced KYNA levels in the hippocampal samples obtained from patients with MTLE-HS. (a) KAT II mRNA expression did not alter significantly (autopsy control, n=6, MTLE-HS, n=12; ΔCT 1.49 ± 0.25 in autopsy control vs ΔCT 1.43 ± 0.44 in MTLE-HS) but (b) protein level expression was significantly decreased in the hippocampal samples (autopsy control, n=8, MTLE-HS, n=10; normalised expression 0.90 ± 0.03 in autopsy control vs normalised expression 0.76 ± 0.062 in MTLE-HS). (c) IDO mRNA expression (autopsy control, n=6, MTLE-HS, n=10; ΔCT 11.12 ± 1.0 in autopsy control vs ΔCT 1.67 ± 0.89 in MTLE-HS) as well as (d) protein expression (autopsy control, n=8, MTLE-HS, n=10; 246.4 ± 11.36 ng/ml in autopsy control vs 294.7 ± 2.77 ng/ml in MTLE-HS) both significantly increased in the hippocampal samples. (e) KMO mRNA (autopsy control, n=6, MTLE-HS, n=12; ΔCT 7.11 ± 0.81 in autopsy control vs ΔCT 5.95 ± 0.49 in MTLE-HS) as well as (f) protein level expression (autopsy control, n=8, MTLE-HS, n=10; normalised expression 0.79 ± 0.073 in autopsy control vs normalised expression 0.74 ± 0.049 in MTLE-HS) did not alter. (g) PLP concentration was significantly reduced in the hippocampal samples (autopsy control, n=6, MTLE-HS, n=27; 5.03 ± 1.01 ng/mg of wet tissue in autopsy control vs 2.04 ± 0.75 ng/mg of wet tissue in MTLE-HS). (h) mRNA expression of PNPO was significantly reduced in the hippocampal samples (autopsy control, n=6, MTLE-HS, n=9; ΔCT 3.41 ± 0.43 in autopsy control vs ΔCT 5.8 ± 0.29 in MTLE-HS). (i) Protein expression of PNPO was also significantly reduced in the hippocampal samples (autopsy control, n=8, MTLE-HS, n=10; normalised expression 0.58 ± 0.042 in autopsy control vs normalised expression 0.39 ± 0.045 in MTLE-HS). The Western blot images of KAT II, KMO, PNPO, and GAPDH represent single band of KAT II at the predicted size of 50 KDa, KMO at 45 KDa, PNPO at 30 KDa and GAPDH at 37 KDa. Data represented as mean ± SEM. * p<0.05, ** p<0.01, *** p<0.001. Mann Whitney Test.
Figure 7: The concentration of metabolites in the hippocampal samples was correlated with seizure duration of patients with MTLE-HS.(a) Scatterplot shows inverse correlation (sample size 40; Spearman’s non-parametric correlation, r= -0.4712, p=0.0021; line of best fit Y= -0.001385*X+0.06080, slope= -0.001385 ± 0.0004206, R2=0.2220, F=10.84, p=0.0021) between concentration of KYNA in the hippocampal samples obtained from patients with MTLE-HS and seizure duration (years). (b) Scatterplot shows inverse correlation (sample size 26; Spearman’s non-parametric correlation, r= -0.6173, p=0.0008; line of best fit Y= -0.009452*X+0.2413, slope= -0.009452 ± 0.003495, R2=0.2336, F=7.313, p=0.0124) betweende novo synthesis of KYNA in the hippocampal samples obtained from patients with MTLE-HS and seizure duration (years). (c) Scatterplot shows correlation (sample size 14; Spearman’s non-parametric correlation, r= 0.5423, p=0.0476; line of best fit Y= 3.683*X-10.84, slope= 3.683 ± 1.135, R2=0.4675, F=10.53, p=0.0070) between concentration of QUIN in the hippocampal samples obtained from patients with MTLE-HS and seizure duration (years). (d) Scatterplot shows no significant correlation (sample size 12; Spearman’s non-parametric correlation, r= 0.5750, p=0.0542) between de novosynthesis of QUIN in the hippocampal samples obtained from patients with MTLE-HS and seizure duration (years). (e) Scatterplot shows inverse correlation (sample size 29; Spearman’s non-parametric correlation, r= -0.4647, p=0.0111; line of best fit Y= -0.04474*X+1.880, slope= -0.04474 ± 0.01921, R2=0.1673, F=5.425, p=0.0276) between concentration of PLP in the hippocampal samples obtained from patients with MTLE-HS and seizure duration (years).