- Top of page
- MATERIALS AND METHODS
Summary: Purpose: The multidrug resistance (mdr) gene family encodes the drug transport macromolecule P-glycoprotein (P-gp), which contributes to the functionality of the blood–brain barrier. Recent evidence suggests that P-gp–mediated drug extrusion may play a facilitatory role in refractory epilepsy. We investigated the regional expression of mdr genes in genetically epilepsy-prone rat (GEPR) brain after a single audiogenic seizure.
Methods: Three groups of adult male GEPRs (n = 5/group) were exposed to a seizure-inducing audiogenic stimulus and killed at 4 h, 24 h, and 7 days thereafter. A further group (n = 5) served as a stimulus-naïve control. Expression of mdr1a and mdr1b in distinct anatomic brain regions (cortex, midbrain, pons/medulla, hippocampus) was determined by quantitative reverse transcriptase–polymerase chain reaction (RT-PCR) in the presence of competitive internal standards.
Results: When compared with control, mdr1a expression in cortex and midbrain was significantly (p < 0.05) increased at 24 h after a single audiogenic seizure. Cortical mdr1a expression remained elevated at 7 days after stimulus. In contrast, mdr1a expression in pons/medulla and hippocampus was unchanged. The mdr1b isoform was quantifiable in hippocampus alone and not influenced by seizure activity.
Conclusions: These findings suggest that acute seizures in the GEPR can induce the expression of mdr genes. The pattern of increased expression appears to follow the anatomic pathway of audiogenic seizures in these animals, with initiation in the midbrain and propagation to the cortex. Further studies are required to investigate the effects of recurrent seizure activity and to characterise mdr expression in other experimental seizure models.
P-glycoprotein (P-gp) is an efflux transporter encoded by the multidrug resistance (MDR1) gene in humans and mdr1a and mdr1b isoforms in rodents (1). It is expressed predominantly in organs with excretory functions (e.g., liver, kidney, gastrointestinal tract) and at blood–tissue barriers (e.g., testis, placenta) (2). P-gp is expressed to a high level in the cerebrovascular endothelium, where it contributes to the functionality of the blood–brain barrier (BBB) (3,4). P-gp–mediated efflux is believed to act as a physiological defense mechanism, extruding xenobiotics from mammalian cells and affording protection to sensitive organs (5).
Accumulating evidence suggests that P-gp may play a facilitatory role in refractory epilepsy (6). Elevated MDR1 expression has been reported in the region of surgically excised temporal lobe epileptic foci (7), and subsequent studies have revealed increased immunohistochemical staining of P-gp in temporal lobe tissues of patients with intractable mesial temporal sclerosis (8) and in tuber cells associated with tuberous sclerosis and uncontrolled seizures (9). Positive perivascular P-gp labeling also has been identified in the region of “pre-epileptic” cortical dysplastic tissue obtained at postmortem (10), and increased P-gp expression observed in endothelial cells isolated from the resected temporal lobes of epilepsy patients (11).
This evidence contributes to an emerging consensus that overexpression of P-gp in the cerebrovascular endothelium, in the region of the epileptic focus, may limit the access of antiepileptic drugs (AEDs) to their site of action and thus play a facilitatory role in refractory epilepsy. This hypothesis is based on the prerequisite that multiple AEDs are substrates for P-gp–mediated efflux. Experimental findings are now beginning to support this premise (4,7,12–14). There remains, however, little or no evidence to suggest how the fundamental overexpression of P-gp in pharmacoresistant epilepsy might originate.
In the field of oncology, overexpression of P-gp is widely recognised to underlie the intrinsic and acquired resistance of several tumour types to chemotherapeutic agents (15), and genetic studies suggest that a number of different stimuli can induce the expression of P-gp (16). It is our hypothesis that, in refractory epilepsy, seizure activity may be responsible for the overexpression of P-gp. We have, therefore, investigated the effects of acute audiogenic seizures in the genetically epilepsy-prone rat (GEPR) on the expression of mdr1a and mdr1b genes in several anatomically distinct brain regions.
- Top of page
- MATERIALS AND METHODS
Despite the introduction of nine new AEDs in the last decade, >30% of epilepsy patients continue to experience seizures with otherwise optimal drug treatment (19). The cause of this poor response remains unknown. In refractory epilepsy, the seizure disorder is resistant to multiple agents with a variety of mechanisms of action, administered either singly or in combination (20). This observation suggests the operation of a general mechanism influencing responsiveness to AEDs. Recent evidence points to the possible involvement of P-gp in the causation of drug-resistant epilepsy (6–11). P-gp is an efflux transporter, encoded by the mdr genes, which is widely acknowledged to contribute to the phenomenon of drug resistance in cancer chemotherapy. We speculate that P-gp may play a similar role in refractory epilepsy and that seizure activity may increase P-gp expression in the BBB. Accordingly, we investigated the regional expression of mdr genes in GEPR brain tissue after acute audiogenic seizures.
Studies in the field of oncology suggest that a number of chemical, physiologic, and pathologic stimuli influence the expression of P-gp (16,21). It is widely accepted that substrate compounds, most notably antineoplastic agents, can induce the expression of mdr genes (22,23). However, intrinsic MDR1 expression also has been observed in untreated tumours derived from tissues that do not ordinarily express the gene (24). In experimental studies, P-gp immunoreactivity in rat brain capillary endothelial cells is immediately decreased after a period of focal ischaemia (25), and prolonged culture of these cells can alter the expression of individual mdr isoforms (26). P-gp expression is also influenced by such diverse stimuli as hepatocarcinogens (27,28), heat shock (29), partial hepatectomy (30), sodium butyrate (31), protein kinase C agonists (32), transient oxidative stress (33), and even P-gp inhibitors such as verapamil (34). Whether seizure activity can be added to this extensive list remains to be determined.
It is increasingly recognised that experimental seizures can influence the expression of many genes, the most widely characterised of which are the immediate early genes, such as c-fos, which are upregulated within 1 h of seizure activity in a number of models (35–39). Kindled seizures, and those induced by kainic acid, have been reported to increase the mRNA levels of several neuropeptides (40–44), whereas audiogenic seizures are associated with increased expression of cholecystokinin (45) and the ubiquitous c-fos(46,47) and a reduced expression of tyrosine hydroxylase (48). However, this is almost certainly the tip of the iceberg as far as seizure-related gene expression is concerned. It is anticipated that a multitude of genes will ultimately prove susceptible to seizure-induced modulation, with model-specific patterns of expression in terms of both temporal relations and neuroanatomic substrates.
At this time, specific investigations of seizure-induced changes in mdr gene expression are limited. Enhanced mdr1 expression has been observed in hippocampus ≤24 h after seizures induced by systemic kainate (49) and increased immunohistochemical labelling of P-gp has been demonstrated in the cell bodies and processes of reactive astrocytes ≤10 weeks after intracerebroventricular kainate (50). In the current study, audiogenic seizures increased the expression of mdr1a in midbrain and cerebral cortex of the GEPR for ≤24 h and 7 days, respectively. Seizures were without effect on the expression of mdr1a in the pons/medulla and hippocampus. These findings are consistent with the recognised anatomic pathways of audiogenic seizures in the GEPR, with initiation in the inferior colliculus of the midbrain, subsequent propagation to the cerebral cortex and no known involvement of the hippocampus (51–53). The mdr1b isoform was quantifiable in hippocampus alone and was not influenced by seizure activity. The selective localisation of mdr1b to the hippocampus has been confirmed in a further study investigating the regional expression of mdr genes in naïve Sprague–Dawley rat brain (54). This differential expression may explain why mdr1b can be detected in whole rat brain homogenates but not in microvessels isolated from rat cerebral cortex (55). It also may account for the unexpected similarities in whole brain concentrations of digoxin, a strong P-gp substrate, between mdr1b knockout and wild-type mice (56), assuming that the differential pattern of expression is extended to this species.
Acute seizure activity is commonly acknowledged to elicit a nonspecific, transient increase in the permeability of the BBB during the ictal period (57). However, little is known about the long-term consequences of seizures on the functionality of the BBB. Downregulation of glucose transporter activity (58) and thickening of the cerebral capillary basement membrane (59,60) have been observed in therapeutically excised human epileptic tissue, suggesting long-term reinforcement of the BBB in response to repeated seizures. Recent evidence that P-gp expression is enhanced in resected epileptic tissue (6–11) would support this premise. However, studies using surgically excised human brain tissue are often beset with complications, not least of which is the identification of appropriate controls. These problems can be circumvented by the use of animal models to determine the precise effect of seizure activity on BBB functionality and/or integrity.
The GEPR was specifically selected for this investigation because, unlike other common experimental seizure models, it does not require physical restraint, implantation of electrodes or the administration of chemoconvulsant compounds, all of which could potentially influence gene expression. A further advantage lies in our understanding of the neuronal networks responsible for seizure generation in the GEPR (51–53), which facilitated direct comparison of mdr expression between implicated and excluded anatomic regions. Gene expression was determined by RT-PCR, which, compared with other techniques such as Northern blot, slot–blot, in situ hybridisation, or RNase protection assay, has the advantage of being highly sensitive and requiring considerably less tissue. The use of an internal standard in competition with the target sequence also allowed absolute quantification of mdr expression, whereas only a relative comparison can be made with other techniques.
When considering the findings of this study, it is important to appreciate that induction of mdr genes does not necessarily translate to increased P-gp expression. Although P-gp regulation is believed to occur primarily at the transcriptional level, with either increased transcription of mdr mRNA or enhanced stability of existing transcripts (2,28,61), there is some evidence that the P-gp protein also is subject to posttranslational control by glycosylation and/or phosphorylation (16). Furthermore, current evidence from studies in surgically excised human brain tissue suggests that mdr expression is elevated in the region of the epileptic focus alone. Normal brain tissue, both proximal and distal to the insult, is believed to be unaffected. In contrast, our experimental findings suggest that mdr expression is increased along the seizure axis. Further studies are clearly required, combining the technologies of quantitative RT-PCR with a concomitant protein-determination method to afford a clearer understanding of the relation between mdr expression and P-gp levels, and with long-term models in which the effect of repeated seizure activity on mdr expression can be more closely correlated with the human condition.
In conclusion, the results of this study suggest that short-term audiogenic stimulation can induce the expression of the mdr1a gene and, by extrapolation, that of P-gp, in the specific brain regions associated with reflex seizure activity in the GEPR. With AEDs now being identified as potential substrates of P-gp and accumulating evidence that P-gp expression is enhanced in the region of intractable human epileptic foci, these findings may have important implications for our understanding of refractory epilepsy and its pharmacologic treatment.