An Atlas of the Quantitative Protein Expression of Anti-Epileptic-Drug Transporters, Metabolizing Enzymes and Tight Junctions at the Blood–Brain Barrier in Epileptic Patients

The purpose of the present study was to quantitatively elucidate the levels of protein expression of anti-epileptic-drug (AED) transporters, metabolizing enzymes and tight junction molecules at the blood–brain barrier (BBB) in the focal site of epilepsy patients using accurate SWATH (sequential window acquisition of all theoretical fragment ion spectra) proteomics. Brain capillaries were isolated from focal sites in six epilepsy patients and five normal brains; tryptic digests were produced and subjected to SWATH analysis. MDR1 and BCRP were significantly downregulated in the epilepsy group compared to the normal group. Out of 16 AED-metabolizing enzymes detected, the protein expression levels of GSTP1, GSTO1, CYP2E1, ALDH1A1, ALDH6A1, ALDH7A1, ALDH9A1 and ADH5 were significantly 2.13-, 6.23-, 2.16-, 2.80-, 1.73-, 1.67-, 2.47- and 2.23-fold greater in the brain capillaries of epileptic patients than those of normal brains, respectively. The protein expression levels of Claudin-5, ZO-1, Catenin alpha-1, beta-1 and delta-1 were significantly lower, 1.97-, 2.51-, 2.44-, 1.90- and 1.63-fold, in the brain capillaries of epileptic patients compared to those of normal brains, respectively. Consistent with these observations, leakage of blood proteins was also observed. These results provide for a better understanding of the therapeutic effect of AEDs and molecular mechanisms of AED resistance in epileptic patients.


Introduction
Epilepsy is one of the most common central nervous system (CNS) disorders, with an incidence of approximately 1% of the population. Post-traumatic epilepsy as a comorbidity of traumatic brain injuries is also becoming a universal challenge for brain health due to the increasing incidence of brain trauma [1]. About one third of patients eventually develop drug resistance to some anti-epileptic drugs (AEDs) and fail to respond to treatment with medication [2], although a variety of combination pharmacotherapy of multiple drugs was investigated [3]. The blood-brain barrier (BBB) regulates the penetration of AEDs into the CNS by forming tight junctions and by expressing drug transporters and drugmetabolizing enzymes. Quantitative analysis of the expression and function of transporters, enzymes and tight junction proteins at the BBB in epilepsy patients is important in terms All epilepsy patients had severe seizure disorder refractory to anti epileptic medications since all patients underwent surgery to try to cure the seizures. Brain capillaries were isolated from brain focal site in epilepsy patients or normal cortex/white matter in non-epileptic donors, and used for the SWATH analysis.

Isolation of Brain Capillaries from Human Brains
The brain capillaries were isolated as previously described [21,22]. The representative microscopic images of the isolated brain capillaries were shown in Figure S1, indicating that the purity of the isolated brain capillaries was not significantly different among epileptic focal site, control cortex and control white matter groups.

Sample Preparation for SWATH-Based Quantitative Proteomics
Protein digestion was performed as described previously [23]. The tryptic digests were cleaned up with a self-packed SDB-XD 200 µL tip (3M, Maplewood, MN, USA) as previously described [24].

LC-MS/MS Measurement for SWATH-Based Quantitative Proteomics
The cleaned peptide samples were injected into an NanoLC 425 system (Eksigent Technologies, Dublin, CA, USA) coupled with an electrospray-ionization Triple TOF 5600 mass spectrometer (SCIEX, Framingham, MA, USA), which was set up for a single direct injection and analyzed by SWATH-MS acquisition, as previously described [20].

Data Analysis for SWATH-Based Quantitative Proteomics
Spectral alignment and data extraction from the SWATH chromatogram were performed with the SWATH Processing Micro App (SCIEX, Toronto, Canada) in Peakview (SCIEX) using in-house spectral libraries as previously described [20] According to the procedure previously described [19], unreliable peaks and peptides were removed based on the criteria of data selection and amino acid sequence-based peptide selection, and the peak areas at the peptide level were calculated as an average of those in the transition level after normalizing the differences in signal intensity between the different transitions. The details have been reported in our previous study [19]. The peak areas of individual proteins were calculated as an average of those at the peptide level, and were compared between epileptic patients and normal groups.

Statistical Analysis
All statistical analyses were performed under the null hypothesis, assuming that the means for the compared groups were equal. Comparison between two groups was performed by an unpaired two-tailed Student's t-test (equal variance) or Welch's test (unequal variance) according to the result of the F test, and followed by a Benjamini-Hochberg (BH) correction. If the p-value was less than 0.05, the difference was considered to be statistically significant and the null hypothesis was rejected. No formal power calculation was performed to estimate the required sample size. No randomization or blinding was performed in this study.

Protein Expression Levels of AED Transporters in Brain Capillaries
Three transporters (MDR1, BCRP and MCT8) and LAT1 are involved in the efflux and influx of AEDs at the BBB, respectively (Table 2) [7,25,26]. The brain capillaries were isolated from the focal sites of epileptic brains or normal brains, and subjected to the SWATH analysis. A total of 2964 proteins were quantified and compared between the epilepsy and normal brain groups (Table S1). The protein expression levels of MDR1 and BCRP were significantly smaller in the epilepsy group than in the normal group. MCT8 and LAT1 were not significantly different between the two groups ( Figure 1).   [12] The AED transporters and metabolizing enzymes detected by SWATH analysis are listed in this table. The substrates (AEDs) of the corresponding transporters and metabolizing enzymes are listed on the basis of literature search.

Figure 2.
Protein expression level of AED-metabolizing enzymes in epileptic brain capillaries compared to controls. Brain capillaries were isolated from focal sites in six epilepsy patients and five normal brains; tryptic digests were produced and subjected to SWATH analysis. The band inside the box represents the median, and the bottom and top of the box indicate the first and third quartiles, respectively. The whiskers indicate the minimum and maximum values for the protein levels. X plots show the average in each group. The data are normalized by the median of the protein expression level in controls. * BH-adjusted p value < 0.05, ** BH-adjusted p value < 0.01 and *** BH-adjusted p value < 0.001, significantly different between the epilepsy and control groups. N.S., not significantly different. GSTP1, Glutathione S-transferase pi; GSTO1, Glutathione S-transferase omega-1; GSTM3, Glutathione S-transferase Mu 3; GSTK1, Glutathione S-transferase kappa 1; MGST2, Microsomal glutathione S-transferase 2; CYP2E1, Cytochrome P450 2E1; ALDH1A1, Aldehyde dehydrogenase family 1 member A1; ALDH2, Aldehyde dehydrogenase 2; ALDH3A2, Aldehyde dehydrogenase family 3 member A2; ALDH6A1, Aldehyde dehydrogenase family 6 member A1; ALDH7A1, Aldehyde dehydrogenase family 7 member A1; ALDH9A1, Aldehyde dehydrogenase family 9 member A1; ADH5, Alcohol dehydrogenase 5; EPHX1, Epoxide hydrolase 1; CBR1, Carbonyl reductase 1; CBR3, Carbonyl reductase 3.

Protein Expression Levels of Tight Junction Molecules in Brain Capillaries
Decreased expression levels of tight junction molecules weaken the integrity of tight junction at the BBB, which would be predicted to increase the permeability of AEDs into brain. As shown in Figure 3, the protein expression levels of Claudin-5, ZO-1, Catenin alpha-1, beta-1 and delta-1 were significantly 1.97-, 2.51-, 2.44-, 1.90-and 1.63-fold smaller in the brain capillaries of epileptic patients than those of normal brains, respectively Protein expression level of AED-metabolizing enzymes in epileptic brain capillaries compared to controls. Brain capillaries were isolated from focal sites in six epilepsy patients and five normal brains; tryptic digests were produced and subjected to SWATH analysis. The band inside the box represents the median, and the bottom and top of the box indicate the first and third quartiles, respectively. The whiskers indicate the minimum and maximum values for the protein levels. X plots show the average in each group. The data are normalized by the median of the protein expression level in controls. * BH-adjusted p value < 0.05, ** BH-adjusted p value < 0.01 and *** BH-adjusted p value < 0.001, significantly different between the epilepsy and control groups. N.S., not significantly different. GSTP1, Glutathione S-transferase pi; GSTO1, Glutathione S-transferase omega-1; GSTM3, Glutathione S-transferase Mu 3; GSTK1, Glutathione S-transferase kappa 1; MGST2, Microsomal glutathione S-transferase 2; CYP2E1, Cytochrome P450 2E1; ALDH1A1, Aldehyde dehydrogenase family 1 member A1; ALDH2, Aldehyde dehydrogenase 2; ALDH3A2, Aldehyde dehydrogenase family 3 member A2; ALDH6A1, Aldehyde dehydrogenase family 6 member A1; ALDH7A1, Aldehyde dehydrogenase family 7 member A1; ALDH9A1, Aldehyde dehydrogenase family 9 member A1; ADH5, Alcohol dehydrogenase 5; EPHX1, Epoxide hydrolase 1; CBR1, Carbonyl reductase 1; CBR3, Carbonyl reductase 3.

Infiltration of Blood Proteins
If the tight junction is weakened, blood proteins can infiltrate to the lateral spaces of endothelial cells or the brain-side surface of endothelial cells, resulting in increased amounts of blood proteins being in the isolated brain capillary fraction. As shown in Figure 4, the levels of the hemoglobin subunit delta, alpha, beta, fibrinogen beta chain, gamma chain, serum albumin, fibrinogen alpha chain and complement C3 in the brain capillary fractions were significantly greater (26.  (Figure 4). This suggests that the blood proteins infiltrate into the lateral spaces of endothelial cells or the brain-side surface of endothelial cells in epilepsy patients.  Figure 3). This suggests that the protein expression of major tight junction proteins are downregulated at the BBB in the focal sites of epileptic patients.

Infiltration of Blood Proteins
If the tight junction is weakened, blood proteins can infiltrate to the lateral spaces of endothelial cells or the brain-side surface of endothelial cells, resulting in increased amounts of blood proteins being in the isolated brain capillary fraction. As shown in Figure 4, the levels of the hemoglobin subunit delta, alpha, beta, fibrinogen beta chain, gamma chain, serum albumin, fibrinogen alpha chain and complement C3 in the brain capillary fractions were significantly greater (26.   Infiltration of blood proteins in epileptic brain capillaries compared to controls. Brain capillaries were isolated from focal sites in six epilepsy patients and five normal brains; tryptic digests were produced and subjected to SWATH analysis. Blood proteins are defined as the molecules whose tissue specificity in the Uniprot database is classified as "blood" or "plasma". Among such molecules, the proteins showing the average ratio (Epilepsy/Control) > 5 and BH-adjusted p value < 0.05 are shown in this table. The band inside the box represents the median, and the bottom and top of the boxes indicate the first and third quartiles, respectively. The whiskers indicate the minimum and maximum values of the protein levels. X plots show the average in each group. The data are normalized by median of the infiltration in control. * BH-adjusted p value < 0.05 and *** BH-adjusted p value < 0.001, significantly different between epilepsy and control groups.

Discussion
The protein expression of AED-related proteins in cerebral vessels of epilepsy patients has been quantified by immunohistochemical analysis [6,37]. The detection of proteins in human tissue sections such as formalin-fixed paraffin-embedded tissue section by immunohistochemistry is semi-quantitative. The accurate and robust quantification of protein expression is a challenging task [38,39]. While antibody-based analysis with high quantitative accuracy has recently emerged, the quantitative analysis of cerebral blood vessels in epilepsy described above has been performed only by conventional immuno- Infiltration of blood proteins in epileptic brain capillaries compared to controls. Brain capillaries were isolated from focal sites in six epilepsy patients and five normal brains; tryptic digests were produced and subjected to SWATH analysis. Blood proteins are defined as the molecules whose tissue specificity in the Uniprot database is classified as "blood" or "plasma". Among such molecules, the proteins showing the average ratio (Epilepsy/Control) > 5 and BH-adjusted p value < 0.05 are shown in this table. The band inside the box represents the median, and the bottom and top of the boxes indicate the first and third quartiles, respectively. The whiskers indicate the minimum and maximum values of the protein levels. X plots show the average in each group. The data are normalized by median of the infiltration in control. * BH-adjusted p value < 0.05 and *** BH-adjusted p value < 0.001, significantly different between epilepsy and control groups.

Discussion
The protein expression of AED-related proteins in cerebral vessels of epilepsy patients has been quantified by immunohistochemical analysis [6,37]. The detection of proteins in human tissue sections such as formalin-fixed paraffin-embedded tissue section by immunohistochemistry is semi-quantitative. The accurate and robust quantification of protein expression is a challenging task [38,39]. While antibody-based analysis with high quantitative accuracy has recently emerged, the quantitative analysis of cerebral blood vessels in epilepsy described above has been performed only by conventional immunohistochemical analysis, and an accurate quantitative analysis of the BBB in epilepsy patients has not yet been reported. In contrast, it was reported that disease-associated changes and differences in protein expression levels, as quantified by quantitative proteomics, accurately reflects differences in the in vivo transport activity of the transporter [40][41][42]. Therefore, in the present study, we analyzed AED transporters, enzymes and tight junction molecules at the BBB using the SWATH method, which has excellent quantitative accuracy, and then quantitatively clarified changes in their expression at the BBB in epilepsy patients for the first time. Unlike antibody-based analyses, the level of expression of the MDR1 protein was found to be decreased in epilepsy subjects (Figure 1). BCRP levels were also decreased. Other AED transporters (MCT8 and LAT1) were unchanged (Figure 1). The expression of various AED metabolizing enzymes in the human BBB is reported for the first time, and the protein expression levels of GSTP1, GSTO1, CYP2E1, ALDH1A1, ALDH6A1, ALDH7A1, ALDH9A1 and ADH5 were found to be significantly increased at the BBB in the epileptic focus ( Figure 2). These metabolic enzymes may be involved in the tolerance of AEDs. Regarding tight junction molecules, the protein expression levels were decreased at the BBB of epileptic patients (Figure 3), as has also been reported in animal models of epilepsy [13,14]. In line with this, the leakage of blood proteins was also shown (Figure 4). MDR1 at the BBB has been reported to be involved in the resistance to AEDs [4]. In contrast, the MDR1 inhibitor verapamil, when administered in combination with AEDs, failed to improve the frequency of epileptic seizures [5]. In the present study, the level of expression of MDR1 at the BBB was lower in epileptic patients compared to the BBB of normal subjects (Figure 1), suggesting that MDR1 has little influence on tolerance to AEDs. Furthermore, because of weakened tight junctions (Figures 3 and 4), the MDR1-mediated resistance to AEDs may be less pronounced. MDR1 at the BBB has been thought to be involved in drug resistance, as previous studies in rodents have shown that MDR1 expression limits the brain transfer of antiepileptic drugs. However, in humans, there is little data to suggest that MDR1 is involved in antiepileptic drug resistance. Immunostaining experiments have shown that the signal intensity of MDR1 is increased, but the specificity and quantitative performance of the antibody is limited. In this study, we applied the SWATH method, which has excellent specificity and quantitative accuracy, and for the first time we were able to accurately quantify the expression levels of MDR1 and BCRP. Therefore, unlike rodents, MDR1 and BCRP do not seem to have a significant effect on antiepileptic drug resistance in humans. Tight junctions were weakened in the focal area of human epilepsy, in agreement with reports in rodents [13,14]. Weakened tight junctions allow blood components to infiltrate the brain, and these have been reported to increase the frequency of epileptic seizures [13]. Thus, the brain transfer of antiepileptic drugs may be also increased, but the degree to which they induce convulsive seizures could be also increased, which may apparently indicate a weaker effect of antiepileptic drugs.
Angiogenesis has been reported to occur in the epileptic brain [43]. No significant increase in the levels of expression of the MDR1 protein was detected in the present SWATH analysis (Figure 1), which indicates that the amount (moles) of MDR1 protein expressed per µg of blood vessel protein was not increased. If angiogenesis increases the amount of blood vessels in the brain per unit volume, then the total levels of expression of efflux transporters (MDR1, BCRP and MCT8) in the unit volume would also increase. It is therefore possible that the distribution of various AEDs (Table 2) in the brain could be limited, which would then lead to the pharmaco-resistance for AEDs. As shown in Figure 2, the expression of the ALDH family members was increased. ALDH has been reported to promote angiogenesis [44]. Therefore, angiogenesis may be promoted by an increased expression of the ALDH family of proteins at the epileptic focus. The number of donors was limited in this study (Table 1), and quantitative analyses using further specimens and investigations of vessel density will be required to understand the overall involvement of efflux transporters in the tolerance to AEDs.
As shown in Table 2, a variety of metabolic enzymes are involved in the metabolism of AEDs. The findings reported in this study show that the expressions of GSTP1, GSTO1, CYP2E1, ALDH1A1, ALDH6A1, ALDH7A1, ALDH9A1 and ADH5 are increased in the BBB of epilepsy patients (Figure 2). Because CYP2E1 and GSTs metabolize a range of AEDs, it is possible that their increased expression could affect AEDs resistance. At the normal BBB, GSTP1 is expressed six-fold more abundantly than MDR1, and GSTO1 is expressed in absolute amounts comparable to MDR1, suggesting that GSTP1 and GSTO1 could sufficiently contribute to the substrate metabolism at the human BBB [9]. At the BBB in the epileptic focus, GSTP1 and GSTO1 were upregulated by 2.13-and 6.23-fold, respectively ( Figure 2). GSTs are involved in the metabolism of major AEDs such as phenytoin, valproate and carbamazepine (Table 2). Furosemide, an inhibitor of GSTs, has been reported to potentiate the anticonvulsant action of valproate [45]. These findings suggest that GSTP1 and GSTO1 could be involved in the drug resistance of AEDs. In future studies, in vivo analyses will be needed to demonstrate whether the enzymes that are up-regulated at the epileptic BBB ( Figure 2) are actually involved in the drug resistance of AEDs.
A limitation of the present study is the small number of donors. Therefore, it cannot be said that all types of epilepsy patients are covered. In terms of age, patients with epilepsy tend to be younger than controls: Control 1 and patients with epilepsy 3, 4 and 6 overlap in age; for the molecules shown in Figures 1-4, there is no significant difference in expression between control 1 and older control donors (donors 2, 3 and 4) (Table S1). Therefore, the quantitative differences between epilepsy patients and controls shown in this study may be due to the pathogenesis of epilepsy. Only epileptic patient 5 was female, and the expression of GSTP1 was lower than in male epileptic patients (Table S1). GSTP1 is expressed in the liver at lower levels in females than in males [46]. Therefore, in BBB, the amount of GSTP1 may be lower in females. Further studies are needed to increase the number of samples and to equalize the age range and gender between epilepsy and control groups.
We have previously reported that MDR1 protein expression at the BBB is increased in mouse models of epilepsy [41], but in contrast, it is decreased in patients with epilepsy ( Figure 1). Table 1 lists the medications that epilepsy patients have been taking. These have not been reported to reduce MDR1 protein expression. While the different age range of the epilepsy and control groups may have an effect, the expression of MDR1 was lower in the epilepsy group when compared between control 1 and epilepsy patients of the same age (donors 3, 4 and 6). In young adulthood, the expression of MDR1 in the BBB is highly variable in mice, but not in primates [47,48]. While the detailed mechanisms are unknown, it may be likely that differences in the regulation of MDR1 expression between mice and primates result in the contrasting changes in expression levels at the BBB in mice and humans with epilepsy.
To investigate whether there are regional differences in the expression levels of the BBB, we isolated blood vessels from white matter and cortex for control 4 and performed SWATH analysis. The results showed that there was no significant difference in the expression of molecules in Figures 1-4 between white matter and cortical BBB (Table S1).

Conclusions
SWATH-based quantitative proteomics with excellent quantitative accuracy was applied to the analysis of components of the BBB in the focal site of epilepsy patients, and finally gave a quantitative atlas illustrated in Figure 5. The results show that the protein expression levels of MDR1 and BCRP are downregulated, and those of MCT8 and LAT1 are comparable to a normal BBB. The findings also revealed that the protein expression levels of eight AED-metabolizing enzymes (GSTP1, GSTO1, CYP2E1, ALDH1A1, ALDH6A1, ALDH7A1, ALDH9A1 and ADH5) are significantly upregulated at the BBB in the focal site of epilepsy patients, whereas the protein expression levels of tight junction molecules are significantly downregulated and this is accompanied by an increased infiltration of blood proteins. The present study has only revealed protein expression levels and it is not yet known whether these changes in expression levels affect the brain pharmacokinetics and pharmacological effects of antiepileptic drugs. Further experimental justification is required.
Pharmaceutics 2021, 13, x FOR PEER REVIEW 11 of 14 Figure 5. Summarized illustration of molecular mechanisms at the BBB in focal site of epileptic patient. These molecular mechanisms were suggested on the basis of the present SWATH results, but the functional demonstration will be needed in future.

Supplementary Materials:
The following are available online at www.mdpi.com/xxx/s1. Figure S1, Microscopic photographs of the isolate brain capillaries; Table S1, Raw proteomic data for the levels of protein expression for all quantified proteins in the brain capillaries of the epilepsy and control groups.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the Figure 5. Summarized illustration of molecular mechanisms at the BBB in focal site of epileptic patient. These molecular mechanisms were suggested on the basis of the present SWATH results, but the functional demonstration will be needed in future.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/pharmaceutics13122122/s1. Figure S1, Microscopic photographs of the isolate brain capillaries; Table S1, Raw proteomic data for the levels of protein expression for all quantified proteins in the brain capillaries of the epilepsy and control groups.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: Not applicable.