Computational Pharmacogenetics of P-Glycoprotein Mediated Antiepileptic Drug Resistance

1. INTRODUCTION

PGP (HGNC nomenclature: ABCB1) is a key transmembrane protein conserved from bacteria to man, and its function is to protect the organism from toxic xenobiotics. PGP has turned out to be a critical player in multiple drug resistance phenomena. Here, we are interested in its role in antiepileptic drug resistance. Epilepsy is a chronic neurological condition affecting more than 50 million people worldwide and 1-2% of the population [1]. The recurring limitation in the treatment protocol of epilepsy is the failure of drug-response in more than one-third of cases. This is the case with the > 30 FDA-approved drugs for epilepsy. PGP is an ATP-coupled efflux pump documented as an epileptogenic mediator [2]. It is known to be highly expressed in the blood-brain barrier, which is pharmacologically crucial for the bioavailability of drugs acting on the central nervous system [3]. The experimental evidence for the role of PGP polymorphisms in antiepileptic drug resistance has not been conclusive so far [4-10]. There is evidence for its cognate role in antidepressant therapy [11]. There are at least two mechanisms by which PGP could mediate refractory epilepsy. First, elevated levels of PGP protein might be linked with the low intracellular drug concentration in cortical cells observed in epilepsy treatment. Elevated abundance could result from multiple mechanisms, for example, promoter sequence mutations, decrease in methylation-mediated epigenetic silencing, change in miRNA-mediated transcript abundance, and decrease the rate of protein degradation. PGP is well-known for its broad substrate specificity, and greater PGP abundance would result in phenotypic drug resistance due to effective drug extrusion. Alternatively, a gain-of-function mutation or changes in protein folding might enhance its functional activity, resulting in the same phenotype, i.e., hyperactive PGP leading to pharmacoresistant epilepsy (Fig. 1)[12, 13].

PGP consists of two homologous halves, each consisting of a transmembrane (TM) domain with six alpha helices and a Nucleotide-Binding Domain (NBD) [14]. A large, hydrophobic and polyspecific drug-binding pocket resides in the inverted V-shaped internal cavity formed by the transmembrane domains [15]. It is clear that he key to epilepsy treatment would involve control of the epileptogenic mediator proteins, including hPGP. Structural studies of hPGP might enhance our current understanding of the role of PGP in drug resistance mechanisms, and provide any evidence of the relationship between specific MDR1 haplotypes and altered drug pharmacokinetics. Given that little information is available on the pharmacology of missense polymorphisms of MDR1, analysis of the role of gain-of-function muta tions in PGP would be valuable. Earlier bioinformatic studies of susceptible genetic Loci have identified and analyzed key polymorphisms specific to pathological phenotypes, for e.g., polymorphisms associated with methylenetetrahydrofolate reductase (MTHFR) gene [16-21]. Genetic polymorphisms are known to influence sus ceptibility to epilepsy (for e.g., BDNF [22, 23]). Here, we have attempted an in silico study of PGP polymorphisms and their structure-activity relationships with respect to drug-substrates to obtain a comprehensive picture of PGP role in the emergence of antiepileptic drug insensitivity. PGP-mediated processes are also the major con tributors to the emergence of drug resistance in cancer therapy and other conditions. Our results could extend to examining the role of P-glycoprotein in generalized drug-resistance in multiple conditions [24].

2. MATERIALS AND METHODS

3. RESULTS

Nearly 500 hPGP SNPs were retrieved, however most of these were unannotated, and we obtained a set of 20 hPGP SNPs for further study, none of whose functional effects were known in the literature (Table 1). The results of our assessment of functional impact by various approaches are summarised in Table 2. Most of the SNPs were determined to be neutral, not disease-causing or deleterious. Five SNPs were predicted to be functionally important by at least one of the tools, as shown in Table 2.

Table 3 provides the representative structures of P-glycoprotein in the PDB. Of these homologous hits, the mouse structures cover the full length of the hPGP. Some mouse structures co-crystallised with a ligand might not be representative of the native PGP conformation. When 4Q9H was superimposed with the 3G5U structure, it was observed that the register of the C-terminal half of the ’inverted-V’ of 4Q9H was displaced relative to that of 3G5U (Fig. 2), which rendered 4Q9H unsuitable for modeling the full hPGP structure. The alignment between the hPGP and 3G5U is very good, showing > 87.5 % sequence identity and good sequence coverage (Fig. 3). 3G5U was used as the template for homology modeling. The target structures of the hPGP wildtype and the five variants were independently modeled and energy- minimised five times each, and the best model was used for further studies. All the modelled structures are available in the Supporting Information and their mutual rms deviations are shown in Table 4.

Phenobarbital was first used as an antiseizure drug in 1912, followed by phenytoin. Today, more than 30 drugs are FDA-approved in the treatment of epilepsy, yet all of them face pharmacoresistance and more than one-third of epilepsy cases remain untreatable. In addition to phenobarbital and phenytoin, two other common antiepileptic medications, namely valproate and carbamazepine, were included in the set of ligands studied (Table 5).

Docking between each hPGP protein (wildtype + 5 variants) and ligand was carried out. Ten docking runs were performed per receptor-ligand pair. Each run provides one low-energy docked conformation of the respective receptor-ligand pair. The site corresponding to the least-energy binding mode was taken as the binding site of the ligand with the receptor. To ascertain convergence to the lowest-energy binding mode, the ten runs of each receptor-ligand binding conformations were clustered at 2.0Å r.m.s. The lowest-energy binding modes showed good convergence. Of the 24 receptor-ligand pairs, 21 had energy-histograms showing the least binding energy (±0.2 kcal/mol) as the most probable conformation and the least binding energies of the rest were within 1 kcal/mol of the binding energies of the most prob- able conformation. This provided confidence that the docking procedure resulted in convergence to the optimum receptor-ligand conformation. The structures of the receptor-drug complexes, as well as the best poses (defined as within 4.5Å of the ligand), are available in the Supporting Information. A comparison of the best poses between the wildtype and one of the variants is illustrated in Fig. (4). The groups of contacting residues specific to each docked complex are shown in Table 6. These residues were defined at a contact distance of < 4.5Å from the drug in the bound conformation. The summary of the docking analysis is presented in Table 7.

4. DISCUSSION

The evolutionary conservation of the location of the prioritised SNPs was computed using the multiple alignments constructed from PGP homologous sequences non-redundant at 40% sequence identity. The quantile scores of each position with respect to hPGP sequence are given in the Supporting information. All five polymorphism locations ranked at least in the 50^th quantile, and the E566K in particular ranked 97^th. The polymorphisms were found in evolutionarily conserved positions. Experimental evidence is available on some of the polymorphisms [39-41].

The hPGP residues binding the ligand in each hPGP-drug pair represent the drug-specific binding pocket (Table 6). These residues contributed to stabilizing the docked complex by forming hydrogen bonds and Van der Waals interactions with the substrates. It is observed that the binding site of wildtype hPGP is identical for phenytoin and carbamazepine. This is a 14-residue binding pocket in the internal cavity lined by four hydrophobic residues (Ile144, Val179, Leu890, Leu924), three charged residues (Arg148, Asp886, Lys934) and four polar residues (Ser180, Asn183, Tyr928, Ser931). In contrast, the variant proteins bound the drugs in alternative variable regions, notably a binding pocket involving Gln132, Val133, Trp136, Ile190, Glu875, Leu879, Phe938, Thr941, and Phe942 that bound all drugs except valproate. It was remarkable that for a given variant hPGP, the binding pocket interactions differed for each drug. The R593C hPGP variant bound phenytoin very close to the mutation site, suggesting evidence for a local conformational change in binding the drug.

Further clarity on these observations could be obtained on an examination of the estimated binding free energies of the wildtype and variant hPGPs with the different drugs of interest and the corresponding predicted differential ligand affinity given by ∆∆G_mut = ∆G_bind,mut − ∆G_bind,wt (Table 7). It was observed that all but three of the differential ligand affinities were negative. This implied that the variant hPGP bound each drug with a stronger affinity than the wildtype hPGP. The maximum range of differential response was observed with variant E566K (−1.25 kcal/mol < ∆∆G_mut < 0.80 kcal/mol).

Two features indicated the neutrality of wildtype hPGP with respect to binding anti-epileptic drugs. First, the binding pocket appeared constant for both phenytoin and carbamazepine. Second, the binding energy with the drug was higher relative to the variants and hence less tight. On the other hand, there were two features that indicated that variant hPGPs would assist in the development of drug resistance.

First, variant hPGPs bound each drug in a different location in the internal cavity. Variability in location affords a better search of the optimal binding modes of the drug. Second, consistently lower binding energies were observed, implying stable drug-PGP complexes for possible energetic extrusion of the drug. The in silico analysis showed that polymorphisms could have played a role in relocating the optimal drug-binding cavity for a higher affinity, relative to the wild-type hPGP.

An elevated affinity between a variant hPGP and the drug could suggest a potential differential adverse response to therapy. From Table 7, it is observed that this is the case for 17 out of the 20 drug-protein combinations studied. Experimental studies are necessary to validate these results and determine whether the mag nitude of any differential adverse responses could translate to the threshold for the development of pharmacoresistance. By contributing to drug resistance, PGP polymorphisms could conceivably play an adverse role in the treatment of multiple epilepsy subtypes. One example is electrical status epilepticus during slow sleep (’ESES’), a condition which could primarily affect children [42] and should best be treated early. The response to use of valproate and diazepam treatment regimens for ESES is quite heterogeneous [43-45], indicating inter-individual variability and a possible role for polymorphisms of pharmacogenomic importance, including those of PGP analyzed in our study.

CONCLUSION

Though hPGP is well-documented as a modifier of drug bioavailability in many conditions, its role in antiepileptic drug resistance has been controversial. At least two alternative mechanisms could explain the hPGP-mediated epileptogenic phenotype. Our work suggests that polymorphisms are a viable mechanism of PGP action that could lead to drug resistance acquisition independent of other mechanisms. It is interesting that all the polymorphisms appeared to result in gain-of-function. Coupled with the observation that somatic mutations could have a similar effect to identical inherited polymorphisms, this would suggest that PGP is a potential oncogene in the context of cancer drug resistance.

Developing a drug resistance strategy to combat drug resistance is a top priority. Our work has highlighted that MDR1 polymorphisms could potentially lower the threshold for development of pharmacoresistance. This gain-of-function process in hPGP offers a novel candidate target in the fight against antiepileptic drug re sistance. Experimental validation of our work is necessary to apply our findings towards achieving pharmacosensitive response in epilepsy treatment. Our methodology is extendable to studies investigating the effect of genetic polymorphisms on phenotypes in other diseases and conditions.

REFERENCES