Modulation of androgen and estrogen receptor expression by ...

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

FULL-LENGTH ORIGINAL RESEARCH

Modulation of androgen and estrogen receptor expression by antiepileptic drugs and steroids in hippocampus of patients with temporal lobe epilepsy *yNina Killer, *yMonika Hock, *Marcel Gehlhaus, *Philipp Capetian, *Rolf Knoth, zGeorgios Pantazis, *Benedikt Volk, and *Ralf P. Meyer *Pathologisches Institut, Abteilung Neuropathologie, Neurozentrum, Freiburg, Germany; yDepartment Mechanical & Process Engineering, Hochschule Furtwangen University, Villingen-Schwenningen, Germany; and zInstitut fu¨r Hirnforschung, Universita¨tsklinikum Tu¨bingen, Tu¨bingen, Germany

SUMMARY Purpose: Many of the antiepileptic drugs (AED) used in therapy of temporal lobe epilepsy (TLE) are known as cytochrome P450 (CYP, P450) inducers. These AEDs are thought to modulate androgen and estrogen pathways in hippocampus, and therefore cause mental and reproductive disorders found in TLE patients. In the present study, we analyzed expression of androgen receptor (AR), estrogen receptor a (ERa), and CYP3A in the hippocampus of TLE patients and in murine hippocampal cell line HN25.1. Methods: Patients and cell lines had been treated with P450-inducing or noninducing AEDs, or with prednisolone, applied to prevent oedema formation prior to neurosurgical resection of the epileptic hippocampus. Human patient samples were analyzed by immunohistochemical approach, the HN25.1 cell line by quantitative RT-PCR, CAT reporter gene assay, and immunoblot.

Epilepsy is a paroxysmal disorder of the brain, which affects about 1% of the population (Forsgren et al., 2005). Temporal lobe epilepsy, the most common form of the chronic intractable epilepsies, is correlated with 70% with the morphological sign of ammon’s horn sclerosis Accepted April 1, 2009; Early View publication June 1, 2009. Address correspondence to Ralf P. Meyer, Ph.D., Pathologisches Institut, Abt. Neuropathologie, Neurozentrum, Universitt Freiburg, Breisacherstraße 64, D-79106 Freiburg, Germany. E-mail: ralf.meyer@ uniklinik-freiburg.de Nina Killer and Monika Hock contributed equally to this article. Wiley Periodicals, Inc. ª 2009 International League Against Epilepsy

Results: In both, humans and cell lines, the expression of testosterone metabolising CYP3A4 (human) or CYP3A11 (mouse) and AR was up-regulated when P450-inducing AEDs and/or prednisolone had been applied. AR responsive CAT reporter gene assay indicated an increase of ARsignalling after treatment of the HN25.1 cells with the P450-inducers phenytoin and carbamazepine. ERa expression was increased only by the P450inducing AEDs, but not by prednisolone, which indicates that pathways different from CYP3A4/11 led to ERa enhancement. Discussion: We conclude that P450-inducing AEDs influence AR expression and signalling in hippocampus most likely via CYP3A4/11-induction. The HN25.1 cell line holds promise to investigate the correlation between drug application and AR regulation, and to specifically address issues that are relevant to human TLE patients. KEY WORDS: Androgen receptor, Antiepileptic drug, Estrogen receptor a, Hippocampus, Cytochrome P450, Temporal lobe epilepsy.

(AHS). AHS is characterized by dispersion and damage of granule cells of the dentate gyrus, neuronal loss mainly in the CA1, CA3, and CA4 regions of the hippocampal formation, gliosis and microglial activation (Freiman et al., 2002). Initial choice of medication are the antiepileptic drugs (AEDs) carbamazepine, oxcarbazepine, lamotrigine, in special cases phenytoin and phenobarbital, which usually act on ion-channels and the GABAergic-system (Rogawski & Loscher, 2004; Berkovic, 2005). Carbamazepine, phenytoin, phenobarbital and oxcarbazepine have the potency to induce cytochrome P450 (P450), mainly CYP3A and CYP2C isoforms, while lamotrigine

1875

1876 N. Killer et al. is considered as noninducer (Flockhart, 2008). Isoforms of the P450 superfamily play an important role in the oxidative metabolism of endogenous and foreign compounds like steroid hormones and pharmaceuticals (Nelson, 2008). After P450 induction, this metabolism can be dramatically altered, leading to adverse drug reactions (ADR) (Wilkinson, 2005). Besides the liver and other peripheral organs, these effects occur also in the brain (Meyer et al., 2007). Consistent with this, the enhanced incidence of reproductive disorders and sexual dysfunction in epileptic patients has been suggested to arise from therapies with P450-inducing antiepileptic drugs, as decreased serum levels of bioactive testosterone and increased 17b-estradiol were found in these patients in several clinical studies (Herzog et al., 1991, 2004; Isojarvi et al., 2005). Members of the CYP3A subfamily have been reported as the major P450-isoforms affected in hippocampal pyramidal neurons after AED treatment. Long-term administration of phenytoin leads to induction of CYP3A11 and to CYP3A-dependent depletion of testosterone, increased 17b-estradiol levels, together with an upregulation of androgen receptor (AR) and estrogen-forming CYP19 (aromatase), predominantly in CA1 pyramidal neurons (Meyer et al., 2006). Cell culture studies substantiated that AR up-regulation and its signal transduction due to AED application are mediated by CYP3A2 (rat) or CYP3A11 (mouse) (Gehlhaus et al., 2007). Additionally, in clinical case reports and several mouse models 17b-estradiol responsive estrogen receptor a (ERa) is described as the main estrogen receptor involved in mood and cognitive disorders (Ohlsson et al., 2000; Bao et al., 2008). These data favor the hypothesis, that P450-inducing AEDs influence the patients’ endocrine system directly inside the hippocampus by P450-dependent alteration of steroid hormone level and consecutive modulation of the associated steroid receptors AR and ERa, which may lead to altered steroid hormone signalling. Therefore, the initial aim of the present study was to analyze the expression of AR, ERa and CYP3A4 in hippocampal neurons of epileptic patients in order to evaluate differences in endocrine configuration due to patient’s treatment with the respective antiepileptic drug. Patients with neurosurgical resection of the hippocampus had been treated additionally with the glucocorticoid prednisolone as antiphlogistic and antioedematous agent. The second aim of the present study was to establish an in vitro model which closely matches the human and animal situation, in order to get more reliable information on the functional coherence between drug administration, P450 induction, and steroid receptor regulation. The hippocampal cell line HN25.1, derived from mouse immortalized hippocampal neurons, is reported to be drug Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

inducible and shows CYP3A-mediated enhanced testosterone metabolism after phenytoin treatment (Thuerl et al., 1997; Hagemeyer et al., 2000). We demonstrate here, that the expression of CYP3A4 and AR in hippocampal pyramidal neurons is closely correlated and concordantly up-regulated in P450 inducerand prednisolone-treated patients. ERa expression appeared to be regulated differently from AR in that it is not correlated with CYP3A4. Investigations of the HN25.1 cell line closely matched these human data. We conclude, that treatment of AHS patients with P450 inducing AEDs or glucocorticoids leads to alterations in androgen and estrogen signalling pathways in the principal neurons of the hippocampus. This highlights that AR expression and signalling are obviously influenced by a P450-dependent pathway. The hippocampal cell line HN25.1 appears as a promising tool to investigate functional correlation between drug application and steroid receptor regulation in the central nervous system.

Methods Subjects We analyzed resection material from patients with temporal lobe epilepsy associated with AHS, from patients with cortical brain tumors, and from a control group of deceased individuals without known (brain) pathology (Table 1 and supplementary Table S1) in retrospective fashion. In all samples investigated in the present study, written informed consent had been obtained from the patient or its relatives in charge for the use of resection material for research. This was approved by the institutional Ethics Committee (EK Freiburg 169/05); all samples were irreversibly anonymized before analysis. Hippocampal surgery was performed in the Department of Neurosurgery of the University Hospital Freiburg, paraffin sections were cut in the Department of Neuropathology. The case history and the following clinical details were recorded whenever available: (1) the dosage and the duration of antiepileptic drug therapy, (2) the dosage and the duration of pre-and postsurgical prednisolone treatment, (3) the occurrence of ADRs, and (4) the use of invasive presurgical epilepsy monitoring, as the resulting gliosis may influence the results. Only patients with robust case history in terms of medication were included in this study. This precluded available patient material with only limited or fragmentary medication data, which inhere the danger of result falsification. Brain tissue All brain samples used in this study had been fixed in buffered formalin solution (4%, v/v). Neurosurgical resected samples were directly fixed for 1 day before paraffin embedding. Autopsy material with post-mortem delay 0.05, the results were considered as not significant (n.s.).

Results Effects of AED and steroid hormone treatment in hippocampus of human subjects Sex differences in the patient groups We used mixed male/female patient population of comparable age in all patient groups (supplementary Table S1). In order to assess possible male/female differences with respect to drug or hormone treatment, we first analysed AR, ERa, and CYP3A4 protein expression using the four scale rating score described in Methods (Fig. 2). After calculation of the median values (Table 2), we investigated the sex dependent target protein expression by evaluation of the variation from the median (Fig. 3). We could not detect any prevalence for reduced or enhanced AR, ERa, or CYP3A4 expression with respect to sex in each of the patient groups. Only a small portion of patients demonstrated variational enhanced or reduced protein expression (Fig. 3). As a conclusion, we could not detect noticeable differences of hippocampal AR, ERa and CYP3A4 expression between both sexes with regard to drug and steroid hormone treatment. Expression of AR and ERa in pyramidal neurons of human hippocampus Androgen receptor and ERa revealed different expression pattern in hippocampal pyramidal neurons of the subjects investigated in the present study. Fig. 4 demonstrates AR and ERa expression in the CA2 pyramidal neurons of the hippocampus proper from representative individuals of each patient group. This region was chosen for analysis, because the pyramidal neurons of CA2 persist in all patients of this study up to Wyler grade IV, and they were observed as the main site of steroid hormone receptor

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ! Figure 4. Expression of AR, ERa, and CYP3A4 in human hippocampal pyramidal neurons of CA 2 region, depending on drug and steroid hormone treatment of the subjects. High-power confocal images of double-labelled sections illustrating colocalization of AR (green; A, E, I, M, Q, U) and ERa (red; B, F, J, N, R, V). Colocalization of the antigens appears yellow in the overlay mode (C, G, K, O, S, W). Immunohistochemical detection of CYP3A4 (brown) and counterstain of cell nuclei with Mayer’s haematoxylin (blue) (D, H, L, P, T, X). (A–D) patient group I; (E–H) patient group II; (I–L) patient group III; pyramidal neurons from grey matter (L III) (glioblastoma multiforme, no epilepsy); (M–P) patient group IV; (Q–T) patient group V; (U–X) patient group VI. Scale bars: 10 lm for the fluorescence doublelabelled sections of AR and ERa, and 25 lm for the immunohistochemical detection of CYP3A4. Epilepsia ILAE

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

1883 Androgen Receptor in Epileptic Hippocampus

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

1884 N. Killer et al. immunoreactivity within the hippocampus (Fig. 1D). Extensive AR expression was found predominantly in the nuclei of pyramidal neurons of those patients, who had been treated with P450-inducing drugs (group I, V, Fig. 4A, Q), and of those with presurgical administration of the glucocorticoid prednisolone (group I, II, III, Fig. 4A, E, I). We could not detect a noticeable difference in AR expression between group I (P450-inducer) and II (noninducer) in the nuclei of CA2 neurons, as deduced from semiquantitative double-blind evaluation (Fig. 4A, E; Table 2). However, we have to keep in mind that prednisolone had been applied to all these patients. Within the patients which had neither AED nor glucocorticoid treatment (groups IV and VI), the main AR signal was detectable in the cytoplasm, while the nuclei showed mild staining at the utmost, with some slight nucleolar fluorescence in group VI (Fig. 4M, U; Table 2). Therefore, manifestation of AHS itself appeared not to have major influence on AR expression in hippocampal CA2 neurons (group IV). However, we still have to put a question mark behind the old debate, if epilepsy itself is associated with endocrine disorders, as we could not obtain samples from untreated epileptic patients with definite seizure frequency, but without hippocampal damage. Expression of ERa was clearly less pronounced as compared to AR and localized predominantly in the cytoplasm (Fig. 4, ‘‘merged’’ column, C, G, K, O, S, W). Furthermore, substantial ERa expression was detected only in hippocampal pyramidal neurons of those patients, who had been treated with an AED (groups I, II, V) (Fig. 4B, F, R; Table 2). Patients who had received prednisolone only (group III), and those of the control group (group VI) did not exert any ERa immunosignal (Fig. 4J, V) (Table 2). Noteworthy, slight nuclear localization of ERa could be found in isolated patients treated with CYP inducers (groups I and V), but its expression was too low to contribute substantially within the respected patient group (Table 2). Expression of CYP3A4 in pyramidal neurons of human hippocampus We next investigated the occurrence of CYP3A4 in hippocampal pyramidal neurons within all patient groups. With great agreement the expression pattern of CYP3A4 matched that of AR (Fig. 4, Table 2). Strong CYP3A4 expression was observed in hippocampal pyramidal neurons of patients representing groups I, II, and V (Fig. 4D, H, T). 8/8 patients of group I, 7/8 patients of group II, and 3/5 patients of group V demonstrated extensive expression of CYP3A4 (Table 2). CYP3A4 also exerted moderate expression in pyramidal neurons of the cortical layer (L III) of group III, whose patients had received glucocorticoid treatment only (Fig. 4L). In patients with non-AED-treated epilepsy (group IV), CYP3A4 expression appeared to be negligible (Fig. 4P; Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

Table 2). Those patients representing the untreated control group VI exerted only mild constitutive expression of CYP3A4 in pyramidal neurons (Fig. 4X, Table 2). Analysis of immortalized pyramidal neurons from mouse hippocampus (HN25.1 cell line) We next performed in vitro investigations on murine hippocampal cell line HN25.1 to analyze the coherence between CYP3A11, AR, and ERa regulation. We initially checked the conditions to resemble the findings from the human study. In case of phenytoin and carbamazepine a concentration of 10 lM revealed as the best match. In case of lamotrigine, we could not detect any difference in cell line response in the concentration range from10 to100 lM (supplementary Fig. S1). Transcriptional regulation of CYP3A11, AR, and ERa after drug- and steroid hormone treatment The treatment of the cells with P450-inducing drugs phenytoin and carbamazepine led to an enhancement of cyp3a11 mRNA expression by the threefold (phenytoin), or the 1.6-fold (carbamazepine), as evaluated by quantitative RT-PCR referring to vehicle control (Taqman) (Fig. 5, cyp3a11). Lamotrigine, applied in 100 lM, did not alter cyp3a11 mRNA expression significantly. Interestingly, application of the glucocorticoid prednisolone led to a 2.3-fold increase of cyp3a11 mRNA. The combination of both, prednisolone and phenytoin, exerted a 1.8 fold increase of cyp3a11 mRNA, which was within the same range as prednisolone or the AEDs alone (Fig. 5, cyp3a11). The expression pattern of AR revealed to be in accordance to that of cyp3a11 after AED or prednisolone treatment. Both P450-inducing AEDs, phenytoin and carbamazepine, led to a 3.8- and 1.9-fold enhancement, respectively, of AR mRNA. Interestingly, the glucocorticoid prednisolone and its combination with phenytoin exerted a 7.6- and 6.2-fold increase in AR mRNA (Fig. 5, AR). Lamotrigine, however, demonstrated transcriptional activation of AR expression by 2.4-fold, although it did not regulate cyp3a11 mRNA in the hippocampal cell line (Fig. 5, AR, cyp3a11). With the exception of lamotrigine, it appeared that AR was regulated similar to cyp3a11. Also in these experiments, the expression pattern of ERa was different as compared to AR after treatment of the cells with either AED or prednisolone (Fig. 5, ERa). Though phenytoin and carbamazepine caused a significant upregulation of ERa mRNA by the 1.5- or 2.1-fold, respectively, lamotrigine, prednisolone, and the combination of prednisolone and phenytoin did not cause any significant alteration of ERa mRNA expression (Fig. 5, ERa). This is in good agreement to the human data, where ERa also appeared to be regulated by the P450-inducers, but not by prednisolone (Fig. 4, Table 2).

1885 Androgen Receptor in Epileptic Hippocampus

Figure 5. Alteration of mRNA expression of cyp3a11, AR, and ERa in immortalized hippocampal neurons (HN25.1), dependent from drug and steroid hormone treatment (48 h incubation time). Data from quantitative real-time RT-PCR (Taqman) were presented as alteration of mRNA expression of each treatment referring to vehicle control (mean ± SEM, n = 6 (vehicle control and 10 lm phenytoin), n = 3 (all others), measured in triplicate). Statistical significance is demonstrated by asterisks (***p £ 0.001; **p £ 0.01; *p £ 0.05; n.s.: alteration not significant). (DPH: phenytoin, CBZ: carbamazepine, Ltg: lamotrigine, Pred: prednisolone). Epilepsia ILAE

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

1886 N. Killer et al.

Figure 6. Expression of AR and ERa protein in subfractions of immortalized hippocampal neuron cell line (HN25.1) after drug and steroid hormone treatment. Phenytoin was added in concentration of 10 lm, prednisolone of 5 lm (48 h incubation time). (A) Representative immunoblot with polyclonal anti-AR antibodies (110 kD) and monoclonal anti-ERa antibody. Fifty micrograms protein was loaded per lane in case of AR detection, 150 lg in that of ERa. (ho: 9,000 g supernatant, cleared lysate; cyt: cytosol; mc: microsomes, nc: nuclear fraction). (B) Visualization of antibody specificity for AR and ERa. AR demonstrated the known bands for AR-A and AR-B at 87 and 110 kD, resp., ERa at 67 kD. Immunosignals were visualized using enhanced chemoluminescence. Epilepsia ILAE

Protein expression of AR and ERa in HN25.1 cell line after drug- and steroid hormone treatment Our previous experiments support differences in regulation of AR- and ERa expression after AED- and prednisolone treatment. To gather further evidence for this concept, we performed immunoblots using AR- and ERa specific antibodies. The size and intensity of the immunoreactive proteins (110 kDa for AR and 67 kDa for ERa, Fig. 6) confirmed our hitherto existing data. After treatment with phenytoin, prednisolone, or a combination of both, AR was excessively upregulated (Fig. 6). In these samples, the AR-immunosignal was present in the purified nuclear fraction derived from the homogenized cells. The major portion of immunoreactivity was found in the cytosol (Fig. 6, AR). Regulation of ERa revealed to be less prominent than that of AR. ERa was found exclusively localized in cytosol. Phenytoin treatment caused a slight upregulation of ERa, while prednisolone, either alone or after combination with phenytoin, apparently did not alter or even reduced the expression of ERa (Fig. 6, ERa). We could not find any nuclear localization of ERa by immunoblot analysis. Activation of AR signalling after drug treatment In order to prove, if the observed upregulation of AR mRNA and protein after drug treatment comes along with activation of AR mediated signal transduction, we performed a reporter gene assay using an androgen receptor sensitive ARE-tk-CAT construct (see ‘‘Methods’’ section for details on the construct). As presented in Fig. 7, both P450-inducing AEDs used in the present study, phenytoin and carbamazepine, caused a highly significant upregulaEpilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

Figure 7. AR-sensitive reporter gene assay in immortalized hippocampal neuron cell line (HN25.1). After pre-treatment with 10 lm phenytoin or carbamazepine for 48 h, cells were transfected with ARE-tk-CAT reporter construct. AR-dependent CAT expression was analyzed after additional 48 h. Results were presented as alteration of CAT expression of each treatment referring to untreated controls. Statistical significance is demonstrated by asterisks (***p £ 0.001). Epilepsia ILAE

tion of the AR sensitive CAT expression about the threefold (3.3-fold in case of phenytoin, 3.1-fold in that of carbamazepine), referring to untreated controls.

Discussion In this report, we describe that expression of AR and CYP3A4 is closely correlated and concordantly

1887 Androgen Receptor in Epileptic Hippocampus up-regulated in pyramidal neurons of the human hippocampus, when the patients had taken P450-inducing AEDs and/or glucocorticoids like the antiphlogistic and antioedematous prednisolone. The expression of the other steroid receptor investigated, ERa, appeared to be regulated differently from AR in that it appears not correlated with the action of CYP3A4, as prednisolone in fact induced CYP3A4 and AR, but not ERa. These findings were supported substantially by investigations of a murine hippocampal in vitro model, the HN25.1 cell line. Our observations point to alterations in the endocrine state in hippocampus of epileptic patients with reflection to their antiepileptic therapy. Neuroactive steroids are main regulators of brain functions and affect a wide range of neurophysiologic parameters like mood, behaviour, sexuality, memory, cognition, etc. (McEwen, 1994). Especially the hippocampus is considered as a prominent area of neurosteroid biosynthesis and action of steroid hormones derived from the gonads or adrenals, like testosterone, estradiol and glucocorticoids (Leranth et al., 2003; Yau et al., 2003). Their corresponding receptors AR, ERa+b, and glucocorticoid receptor (GR) show high concentrations and considerable overlap of expression in the hippocampus (Kawata, 1995; McEwen et al., 2001). Several pathological conditions have perturbing influence on the steroid hormone balance in the hippocampus. Reduced bioavailable testosterone and elevated estradiol levels in blood serum were found in clinical studies investigating patients with temporal lobe epilepsy. Steroid hormone mediated behaviour is altered in a significant fraction of these patients. This includes affective, cognitive, and reproductive functions in both sexes, like e.g., bad or depressive mood state, detrimental effects on cognition, impotence, polycystic ovaries, and menstrual problems (Herzog & Fowler, 2005; Frye, 2006). Interestingly, such disorders are much more pronounced in patients who had been treated with P450-inducing AEDs compared to such, who had been treated with noninducers (Stores, 1975; Isojarvi et al., 2005). The question arises, whether these complications are derived originally from dysfunctions of the damaged hippocampus or secondary from drug mediated alterations directly within the hippocampus. In a previous study using a mouse in vivo model, we observed after treatment with phenytoin a collective increase in AR, CYP3A11 and CYP19, consecutively enhanced testosterone metabolism leading to its decreased level in hippocampus, and additional enhancement of 17b-estradiol (Meyer et al., 2006). Consistently, it has been demonstrated, that androgen levels of epileptic patients are altered mainly in the hippocampal tissue rather than in blood serum (Frye, 2006). These findings assign for an apparently P450-dependent upregulation of steroid metabolism and signalling pathways directly within the hippocampus.

However, besides the clinical and animal studies mentioned so far, the lack of systematic investigations in human brain restrains the understanding of how AEDs influence brain’s steroid regulation and signalling. The present study was set up to elucidate the coherence between AED therapy and alterations in steroid receptor configuration and signaling in the hippocampus of epileptic patients with AHS. Therefore, a carefully defined set of AHS patients, who had been treated with AEDs and/or the glucocorticoid prednisolone, was analyzed concerning the expression of the steroid receptors AR and ERa in pyramidal neurons of the hippocampus. Prednisolone is a corticosteroid drug with predominantly glucocorticoid activity, used pre- and postsurgically as antiphlogistic and antioedemic therapy. However, prednisolone and glucocorticoids in general are supposed to interact with P450 isoforms, namely CYP2C19 and CYP3A4 (Flockhart, 2008). In order to evaluate cross-reactions due to prednisolone therapy, we included a thorough analysis of its effects in the present study. The use of well defined postmortem tissue in the groups IV, V, and VI allowed us to set up indispensable controls with respect to epilepsy, AED-treatment, and AHS manifestation. These advantages overcome potential caveats emerging when using post-mortem tissue. We found AR immunoreactivity highly increased in the nuclei of pyramidal neurons of AED- and/or prednisolone-treated patients. This was supported by in vitro data using the hippocampal cell line HN25.1, which consists of immortalized hippocampal pyramidal neurons from mice (Lee et al., 1990). Also here, the P450-inducing AEDs phenytoin and carbamazepine, and the glucocorticoid prednisolone caused up-regulation of the androgen signalling cascade, proceeding over AR mRNA and protein expression to enhanced androgen responsive reporter gene expression. However, it is known, that metabolic trapping or protein binding of these drugs can alter their biological activity. Such effects may vary dramatically when changing models from animals to cell lines or human samples (Meyer et al., 2001a, 2001b). Nevertheless the drug concentrations used in the present in vitro assay led to similar results as compared to the outcomes from the human study. This enables us to use this cell line as a model for AED – steroid hormone crosstalk. We speculate that CYP3A is decisive for AR regulation. CYP3A4 and its mouse analogue CYP3A11 were found to be strongly induced in all those patients and cell lines, where AR was highly expressed, induced, or nuclear accumulated. This occurred in all cases, where patients and the cell line had been treated with the P450-inducers oxcarbazepine (human), with carbamazepine and phenytoin, or with prednisolone (human and mouse). These data are in good agreement with previous studies in mice (Meyer et al., 2006), and with available clinical studies analyzing circulating hormone levels (Isojarvi et al., Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

1888 N. Killer et al. 2005). In addition to these data we recently demonstrated, that gene silencing of CYP3A11 or its chemical inhibition by ketoconazole led to decreased AR expression and signal transduction in neuronal and hepatic cell lines (Gehlhaus et al., 2007). These observations clearly indicate, that AEDs affect neuronal androgen signalling via a P450dependent pathway. The AEDs themselves inhere the potency of CYP3A4/11 induction within the hippocampus. The induced CYP3A in turn metabolizes its substrate testosterone to the mostly inactive 2b-, and 6b-hydroxytestosterones (Meyer et al., 2006). The androgen-sensitive cells in hippocampus might get a signal to compensate for reduced testosterone levels by receptor up-regulation to maintain overall hormonal influence on cellular procedures. This ligand metabolism-based hypothesis is corroborated by several in vitro and in vivo studies that have revealed similar effects about androgen-dependent AR regulation (Quarmby et al., 1990; Krongrad et al., 1991). Estrogen receptor a has recently been described as the main estrogen receptor involved in mood and cognitive disorders (Ohlsson et al., 2000; Bao et al., 2008). This discriminates ERa from the other estrogen receptor, ERb, which appears to function in cortical development of the brain, although recent findings on brain ERb has made think it controls functions previously attributed to ERa (Wang et al., 2003). In our hands, ERa seems to be regulated differently from AR in that it is most likely not affected by the action of CYP3A4/CYP3A11. We could find slight nuclear translocation only in isolated patients of the groups I and V, all of them having been treated with P450-inducers only. In all other groups, ERa was localized in the cytoplasm, indicating its inactivated state. In the in vitro assay, ERa could be found exclusively in the cytosol. This is consistent with other studies demonstrating ERa as mainly extranuclear in hippocampal pyramidal neurons (McEwen et al., 2001; Hart et al., 2007). There are several studies demonstrating that testosterone in brain can be converted to 17b-estradiol by CYP19, and to the estrogen receptor ligands 5a-androstan-3a,17b-diol and its 3b-isomer, both synthesized by corresponding 3a- or 3b-hydroxysteroid dehydrogenases from intermediate 5a-dihydrotestosterone (Balthazart & Ball, 2006; MacLusky et al., 2006; Scharfman & MacLusky, 2006). These reactions are independent from metabolizing P450s like CYP3A4/11. Because it is known from animal studies, that steroidogenic CYP19 is induced by AEDs like phenytoin (Meyer et al., 2006), we speculate, that P450inducing drugs may influence ERa via such estrogenic pathways, independent from CYP3A4. This may explain, why prednisolone could not activate ERa, as this glucocorticoid is not metabolized by CYP19 or 5a-reductase. In the present study, prednisolone revealed analogous effects on CYP3A4/11 and AR as did the P450-inducing AEDs, but obviously different ones with respect to ERa. This is in accordance with other reports demonstrating, that Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

prednisolone acts as inducer of CYP3A4, and that serum testosterone levels are significantly reduced after glucocorticoid treatment (Contreras et al., 1996; Pascussi et al., 2000). On the other hand, in peripheral tissues prednisolone can act as antiandrogen (Zouboulis & PiqueroMartin, 2003). These reports point to the well known effect of tissue specific differences in steroid hormone action (MacLean et al., 1997; Turgeon et al., 2004). The potency of prednisolone to induce CYP3A4/11 can most likely explain the finding that in patients group II, treated with the P450 noninducer lamotrigine, CYP3A4 and AR were upregulated, whereas in the cell line assay only AR was enhanced. These patients underwent a neurosurgical resection of the hippocampus, and received prednisolone in addition to lamotrigine. Because section material from autopsy of patients, who received monotherapy with non-P450 inducers like lamotrigine without neurosurgery, is not available so far, we investigated the effects of lamotrigine in the in vitro assay. In these experiments, lamotrigine was applied without additional glucocorticoid, and revealed to enhance AR mRNA independent from murine CYP3A11. The endocrine effects of the new AEDs like lamotrigine have not been widely studied so far. There are recent reports using rat models, which demonstrate either possible toxic effects on sexual organs and the liver, when lamotrigine is applied at high dose (30 mg/kg) (Daoud et al., 2004), or no significant changes in gonadal morphology and no disturbances in reproductive hormones like FSH and LH at lamotrigine serum levels of about 45 lM (Roste et al., 2003). Others speculate that lamotrigine improves mood and cognitive tasks in humans (Brown et al., 2003). It turned out, that lamotrigine has the potential to slightly induce some endogenous enzymes, e.g., in thyroid hormone pathway (Strolin et al., 2005). Our observation, that lamotrigine enhances AR, most likely independent of CYP3A11, may contribute to further characterization of its pharmacological effects. We conclude that AHS patients, who are treated by P450 inducing AEDs and/or by glucocorticoids like prednisolone, exhibit alterations in androgen and estrogen signalling pathways in the principal neurons of the hippocampus. The influence of the damaged hippocampus itself on endocrine function appears much less, as evidenced from the findings with patient group IV. In case of AR, modulation of its expression and signalling is most likely caused by drug mediated induction and consecutive enhanced testosterone metabolic activity of CYP3A4. In case of ERa regulation, drug mediated effects on CYP19 and androgen metabolism independent of CYP3A4 appear causative. In vitro tools like the HN25.1 cell line assay appear as good and valuable models for investigation and prediction of neuroactive drug action.

1889 Androgen Receptor in Epileptic Hippocampus

Acknowledgments The present study was supported by grants of the German Research Foundation, DFG (DFG-Me1544/4), and the Forschungskommission of the University Hospital Freiburg (MEY 322/04, MEY 375/05 and MEY 465/06). The authors want to thank the following persons for their assistance during the genesis of the present study: Margarethe Ditter for her technical assistance, and Dr. Marie Follo from the Core Facility of the Department of Internal Medicine I, Universittsklinik Freiburg, for her assistance in performing the QPCR analysis. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Disclosure: All the authors state that they have no conflicts of interest.

References Balthazart J, Ball GF. (2006) Is brain estradiol a hormone or a neurotransmitter? Trends Neurosci 29:241–249. Bao AM, Meynen G, Swaab DF. (2008) The stress system in depression and neurodegeneration: focus on the human hypothalamus. Brain Res Rev 57:531–553. Berkovic SF. (2005) Treatment with anti-epileptic drugs. Aust Fam Physician 34:1017–1020. Blackbeard J, O’Dea KP, Wallace VC, Segerdahl A, Pheby T, Takata M, Field MJ, Rice AS. (2007) Quantification of the rat spinal microglial response to peripheral nerve injury as revealed by immunohistochemical image analysis and flow cytometry. J Neurosci Methods 164:207–217. Bradford MM. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of proteindye binding. Anal Biochem 72:248–254. Brown ES, Frol A, Bobadilla L, Nejtek VA, Perantie DC, Dhillon H. (2003) Effect of lamotrigine on mood and cognition in patients receiving chronic exogenous corticosteroids. Psychosomatics 44:204–208. Cao S, Hudnall SD, Kohen F, Lu LJ. (2000) Measurement of estrogen receptors in intact cells by flow cytometry. Cytometry 41:109–114. Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld MG, Sawyers CL. (2004) Molecular determinants of resistance to antiandrogen therapy. Nat Med 10:33–39. Colburn RW, DeLeo JA, Rickman AJ, Yeager MP, Kwon P, Hickey WF. (1997) Dissociation of microglial activation and neuropathic pain behaviors following peripheral nerve injury in the rat. J Neuroimmunol 79:163–175. Contreras LN, Masini AM, Danna MM, Kral M, Bruno OD, Rossi MA, Andrada JA. (1996) Glucocorticoids: their role on gonadal function and LH secretion. Minerva Endocrinol 21:43–46. Daoud AS, Bataineh H, Otoom S, Abdul-Zahra E. (2004) The effect of Vigabatrin, Lamotrigine and Gabapentin on the fertility, weights, sex hormones and biochemical profiles of male rats. Neuro Endocrinol Lett 25:178–183. Flockhart DA. (2008) P450-Drug interaction table. Available at: http:// medicine.iupui.edu/flockhart/ [Accessed September 30, 2008] Forsgren L, Beghi E, Oun A, Sillanpaa M. (2005) The epidemiology of epilepsy in Europe – a systematic review. Eur J Neurol 12:245– 253. Freiman TM, Gimbel K, Honegger J, Volk B, Zentner J, Frotscher M, Deller T. (2002) Anterograde tracing of human hippocampus in vitro-a neuroanatomical tract tracing technique for the analysis of local fiber tracts in human brain. J Neurosci Methods 120:95– 103. Frye CA. (2006) Role of androgens in epilepsy. Expert Rev Neurother 6:1061–1075. Gehlhaus M, Schmitt N, Volk B, Meyer RP. (2007) Antiepileptic drugs affect neuronal androgen signaling via a cytochrome P450-dependent pathway. J Pharm Exp Ther 322:550–559. Hagemeyer CE, Rosenbrock H, Singec I, Knoth R, Volk B. (2000) Different testosterone metabolism by immortalized embryonic and

postnatal hippocampal neurons from C57BL/6 mice: a crucial role for androstenedione. J Neurosci Res 60:106–115. Hagemeyer CE, Rosenbrock H, Ditter M, Knoth R, Volk B. (2003) Predominantly neuronal expression of cytochrome P450 isoforms CYP3A11 and CYP3A13 in mouse brain. Neuroscience 117:521– 529. Hart SA, Snyder MA, Smejkalova T, Woolley CS. (2007) Estrogen mobilizes a subset of estrogen receptor-alpha-immunoreactive vesicles in inhibitory presynaptic boutons in hippocampal CA1. J Neurosci 27:2102–2111. Herzog AG, Levesque LA, Drislane FW, Ronthal M, Schomer DL. (1991) Phenytoin-induced elevation of serum estradiol and reproductive dysfunction in men with epilepsy. Epilepsia 32:550–553. Herzog AG, Drislane FW, Schomer DL, Pennell PB, Bromfield EB, Kelly KM, Farina EL, Frye CA. (2004) Differential effects of antiepileptic drugs on sexual function and reproductive hormones in men with epilepsy: interim analysis of a comparison between lamotrigine and enzyme-inducing antiepileptic drugs. Epilepsia 45:764–768. Herzog AG, Fowler KM. (2005) Sexual hormones and epilepsy: threat and opportunities. Curr Opin Neurol 18:167–172. Isojarvi JI, Tauboll E, Herzog AG. (2005) Effect of antiepileptic drugs on reproductive endocrine function in individuals with epilepsy. CNS Drugs 19:207–223. Joe I, Ramirez VD. (2001) Binding of estrogen and progesterone-BSA conjugates to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and the effects of the free steroids on GAPDH enzyme activity: physiological implications. Steroids 66:529–538. Kawata M. (1995) Roles of steroid hormones and their receptors in structural organization in the nervous system. Neurosci Res 24:1–46. Krongrad A, Wilson CM, Wilson JD, Allman DR, McPhaul MJ. (1991) Androgen increases androgen receptor protein while decreasing receptor mRNA in LNCaP cells. Mol Cell Endocrinol 76:79–88. Lee HJ, Hammond DN, Large TH, Roback JD, Sim JA, Brown DA, Otten UH, Wainer BH. (1990) Neuronal properties and trophic activities of immortalized hippocampal cells from embryonic and young adult mice. J Neurosci 10:1779–1787. Leranth C, Petnehazy O, MacLusky NJ. (2003) Gonadal hormones affect spine synaptic density in the CA1 hippocampal subfield of male rats. J Neurosci 23:1588–1592. Lutz LB, Jamnongjit M, Yang WH, Jahani D, Gill A, Hammes SR. (2003) Selective modulation of genomic and nongenomic androgen responses by androgen receptor ligands. Mol Endocrinol 17:1106– 1116. MacLean HE, Warne GL, Zajac JD. (1997) Localization of functional domains in the androgen receptor. J Steroid Biochem Mol Biol 62:233–242. MacLusky NJ, Hajszan T, Prange-Kiel J, Leranth C. (2006) Androgen modulation of hippocampal synaptic plasticity. Neuroscience 138:957–965. McEwen BS. (1994) How do sex and stress hormones affect nerve cells? Ann N Y Acad Sci 743:1–16. McEwen B, Akama K, Alves S, Brake WG, Bulloch K, Lee S, Li C, Yuen G, Milner TA. (2001) Tracking the estrogen receptor in neurons: implications for estrogen-induced synapse formation. Proc Natl Acad Sci USA 98:7093–7100. Meyer RP, Hagemeyer CE, Knoth R, Kurz G, Volk B. (2001a) Oxidative hydrolysis of scoparone by cytochrome P450 Cyp2c29 reveals a novel metabolite. Biochem Biophys Res Commun 285:32–39. Meyer RP, Knoth R, Schiltz E, Volk B. (2001b) Possible function of astrocyte cytochrome P450 in control of xenobiotic phenytoin in the brain: in vitro studies on murine astrocyte primary cultures. Exp Neurol 167:376–384. Meyer RP, Podvinec M, Meyer UA. (2002) Cytochrome P450 CYP1A1 accumulates in the cytosol of kidney and brain and is activated by heme. Mol Pharmacol 62:1061–1067. Meyer RP, Hagemeyer CE, Knoth R, Kaufmann MR, Volk B. (2006) Anti-epileptic drug phenytoin enhances androgen metabolism and androgen receptor expression in murine hippocampus. J Neurochem 96:460–472. Meyer RP, Gehlhaus M, Knoth R, Volk B. (2007) Expression and function of cytochrome p450 in brain drug metabolism. Curr Drug Metab 8:297–306.

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

1890 N. Killer et al. Nelson DR. (2008) P450 nomenclature and overview. Available at: http://drnelson.utmem.edu/CytochromeP450.html [Accessed September 30, 2008]. Niki T, Takahashi-Niki K, Taira T, Iguchi-Ariga SM, Ariga H. (2003) DJBP: a novel DJ-1-binding protein, negatively regulates the androgen receptor by recruiting histone deacetylase complex, and DJ-1 antagonizes this inhibition by abrogation of this complex. Mol Cancer Res 1:247–261. Ohlsson C, Hellberg N, Parini P, Vidal O, Bohlooly Y, Rudling M, Lindberg MK, Warner M, Angelin B, Gustafsson JA. (2000) Obesity and disturbed lipoprotein profile in estrogen receptoralpha-deficient male mice. Biochem Biophys Res Commun 278:640– 645. Pascussi JM, Gerbal-Chaloin S, Fabre JM, Maurel P, Vilarem MJ. (2000) Dexamethasone enhances constitutive androstane receptor expression in human hepatocytes: consequences on cytochrome P450 gene regulation. Mol Pharmacol 58:1441–1450. Quarmby VE, Yarbrough WG, Lubahn DB, French FS, Wilson EM. (1990) Autologous down-regulation of androgen receptor messenger ribonucleic acid. Mol Endocrinol 4:22–28. Reid JM, Kuffel MJ, Ruben SL, Morales JJ, Rinehart KL, Squillace DP, Ames MM. (2002) Rat and human liver cytochrome P-450 isoform metabolism of ecteinascidin 743 does not predict gender-dependent toxicity in humans. Clin Cancer Res 8:2952–2962. Rogawski MA, Loscher W. (2004) The neurobiology of antiepileptic drugs. Nat Rev Neurosci 5:553–564. Roste LS, Tauboll E, Isojarvi JI, Berner A, Berg KA, Pakarinen AJ, Huhtaniemi IT, Knip M, Gjerstad L. (2003) Gonadal morphology and sex hormones in male and female Wistar rats after long-term lamotrigine treatment. Seizure 12:621–627. Sawa A, Khan AA, Hester LD, Snyder SH. (1997) Glyceraldehyde3-phosphate dehydrogenase: nuclear translocation participates in neuronal and nonneuronal cell death. Proc Natl Acad Sci USA 94:11669–11674. Scharfman HE, MacLusky NJ. (2006) The influence of gonadal hormones on neuronal excitability, seizures, and epilepsy in the female. Epilepsia 47:1423–1440. Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM. (1990) Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497– 504. Stores G. (1975) Behavioural effects of anti-epileptic drugs. Dev Med Child Neurol 17:647–658.

Epilepsia, 50(8):1875–1890, 2009 doi: 10.1111/j.1528-1167.2009.02161.x

Strolin BM, Ruty B, Baltes E. (2005) Induction of endogenous pathways by antiepileptics and clinical implications. Fundam Clin Pharmacol 19:511–529. Tauer U, Knoth R, Volk B. (1998) Phenytoin alters Purkinje cell axon morphology and targeting in vitro. Acta Neuropathol 95:583–591. Thuerl C, Otten U, Knoth R, Meyer RP, Volk B. (1997) Possible role of cytochrome P450 in inactivation of testosterone in immortalized hippocampal neurons. Brain Res 762:47–55. Turgeon JL, McDonnell DP, Martin KA, Wise PM. (2004) Hormone therapy: physiological complexity belies therapeutic simplicity. Science 304:1269–1273. Wang L, Andersson S, Warner M, Gustafsson JA. (2003) Estrogen receptor (ER) beta knockout mice reveal a role for ERbeta in migration of cortical neurons in the developing brain. Proc Natl Acad Sci USA 100:703–708. Wilkinson GR. (2005) Drug metabolism and variability among patients in drug response. N Engl J Med 352:2211–2221. Yau JL, Rasmuson S, Andrew R, Graham M, Noble J, Olsson T, Fuchs E, Lathe R, Seckl JR. (2003) Dehydroepiandrosterone 7-hydroxylase CYP7B: predominant expression in primate hippocampus and reduced expression in Alzheimer’s disease. Neuroscience 121:307– 314. Zouboulis CC, Piquero-Martin J. (2003) Update and future of systemic acne treatment. Dermatology 206:37–53.

Supporting Information Additional Supporting Information may be found in the online version of this article: Figure S1. Validation data for the experiments using the hippocampal cell line HN25.1. Table S1. Clinical data of human subjects used in present study. Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.