Cell Tissue Res (2008) 332:403–414 DOI 10.1007/s00441-008-0575-y

REGULAR ARTICLE

Biochemical analysis of selenoprotein expression in brain cell lines and in distinct brain regions Barbara Hoppe & Anja U. Bräuer & Markus Kühbacher & Nicolai E. Savaskan & Dietrich Behne & Antonios Kyriakopoulos

Received: 19 June 2007 / Accepted: 9 January 2008 / Published online: 4 March 2008 # Springer-Verlag 2008

Abstract Selenium is present in various biologically important selenoproteins. The preferential incorporation of selenium into the brain indicates its significance for this organ, but so far knowledge concerning the cerebral selenoproteome is scarce. We therefore investigated the expression of selenoproteins in various regions of the rat brain, various subcellular fractions and several brain cell lines by 75Se-labelling, gel electrophoretic separation and autoradiography, with the 75Se tracer as the selenoprotein marker. Quantitative evaluation of the labelled proteins in selenium-deficient rats revealed information regarding preferentially supplied selenoproteins and their distribution; 21 selenoproteins could be distinguished, among them a novel or modified 15-kDa selenoprotein enriched in the cerebellum cytosol. The selenoproteins differed in the degree of their expression among the brain regions and within a region among the subcellular fractions. Some cellBarbara Hoppe and Anja U. Bräuer contributed equally to this work. This study was supported by grants from the Deutsche Forschungsgemeinschaft (SPP 1087 and SA 1041/2–3). B. Hoppe : M. Kühbacher : D. Behne : A. Kyriakopoulos (*) Department of Molecular Trace Element Research in the Life Sciences, Hahn-Meitner-Institut, Glienicker Straße 100, 14109 Berlin, Germany e-mail: [email protected] A. U. Bräuer Institute of Cell Biology & Neurobiology, Center for Anatomy, Charité-Universitätsmedizin Berlin, Schumannstraße 20/21, 10098 Berlin, Germany N. E. Savaskan Department of Neuromorphology, Brain Research Institute, Swiss Federal Institute of Technology (ETH) and University of Zürich, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland

type-specific selenium-containing proteins were found in the cell lines. Differences in the distribution patterns between mono-cultured and co-cultured endothelial cells and astrocytes showed that mediators produced by other cells could affect the selenoprotein expression of a specific cell-type. This effect might play a role in the uptake and distribution of selenium in the brain but could also be of significance in the selenium metabolism of other tissues. Keywords Selenium . Selenoproteins . Brain . Brain cell lines . Rat (Wistar)

Introduction Selenium is an essential micronutrient in vertebrates and is needed during brain development and metabolism, as recently reviewed (Röthlein 1999; Chen and Berry 2003; Brauer and Savaskan 2004; Schweizer et al. 2004). Measurements of selenium levels exhibit a considerable variation among different tissues (Behne and Wolters 1983). Regulation mechanisms have been found to exist that, during insufficient selenium intake, ensure preferential supply of the central nervous system (CNS) with this element. Consequently, its content in the brain is maintained for prolonged periods of selenium deficiency to a much greater degree than that in the other body compartments, a finding that suggests the especially important functions of selenium in this organ (Behne et al. 1988; Savaskan et al. 2007). Selenium is present in numerous selenoproteins including several enzymes with physiological key roles, such as the glutathione peroxidases (GPx)1, the deiodinases and 1 Abbreviations: (GPx), glutathione peroxidases; (D), deiodinases; (TR), thioredoxin reductases; (SECIS), selenocysteine-insertionsequence; (CNS), central nervous system; (Sec), selenocysteine; (UGTR), UDP-glucose:glycoprotein glucosyltransferase.

404

the thioredoxin reductases (TR). Here, it is contained in the form of the rare amino acid selenocysteine, which is also known as the 21st amino acid (Stadtman 1996), encoded by a UGA codon in combination with a specific stem loop (selenocysteine-insertion-sequence) downstream of the codon (Berry et al. 1991; Walczak et al. 1998). Most of the members of the selenoprotein family have been detected by using bioinformatical approaches (Kryukov et al. 2003). The functions of many of them are still unknown and indications have been obtained of the existence of some further forms of selenoproteins that have not yet been identified (Behne and Kyriakopoulos 2001). So far, selenium has been established to play a critical role in various health conditions of the CNS, such as seizure (Weber et al. 1991; Ramaekers et al. 1994; Savaskan et al. 2003), cancer (Ganther 2001; Brenneisen et al. 2005; Squires and Berry 2006), brain development (Mitchell et al. 1998; Kohrle 2000; Loflin et al. 2006) and mood disorders (Hawkes and Hornbostel 1996; Benton 2002). However, information concerning the role of the selenoproteins in these processes remains scarce. The glutathione system seems to have important functions in controlling cellular redox states and as a primary defence against H2O2 and peroxides in brain cells (Dringen et al. 2005; Aguirre et al. 2006). In accordance with this finding, an increased susceptibility to oxidative stress in the cortical neurons (de Haan et al. 1998) and a higher vulnerability to mitochondrial toxins have been observed (Klivenyi et al. 2000) in mice after the homozygous disruption of the cytosolic GPx (GPx1). Elevated expression of GPx has been detected in glial cells around damaged sites in infarct patients (Takizawa et al. 1994) and in patients who died following Parkinson’s disease (Damier et al. 1993). An increase in the TR level has also been reported in patients with Alzheimer’s disease (Lovell et al. 2000). The significance of GPx4 and selenoprotein P (SelP) for the brain has been shown in cell culture studies of neurons (Yan and Barrett 1998; Savaskan et al. 2007) and astrocytes (Steinbrenner et al. 2006) and in experiments on mice in which neurological cell death in GPx4 null mutants and neurological dysfunctions in mice with disrupted SelP genes have been found (Schomburg et al. 2003; Hill et al. 2004; Borchert et al. 2006). In order to be able to elucidate the role of selenium in the brain in more detail, knowledge of the various selenoproteins present in the brain is required. Of special interest in this respect is the investigation of differences in their distribution among the different brain regions, cell types and subcellular compartments in order to obtain new information regarding their possible sites of action and functions. So far, no extensive studies have been carried in this field. This may partly be attributable to the lack, in

Cell Tissue Res (2008) 332:403–414

most of the cases, of the immuno-assays to that allow specific determination of the expressed proteins. We have therefore examined the patterns of selenoproteins in various brain regions of the rat, in subcellular fractions of these samples and in various brain cell lines in radiotracer experiments with 75Se as a selenoprotein marker. Cell cultures have frequently been used to investigate selenoproteins in body compartments. However, whether interactions among the different cell types influence selenoprotein expression has not yet been established; hence, the selenoprotein pattern in a mono-culture of a specific cell type may differ from that found in the presence of the other cells contained in the tissue in question. Experiments have therefore been included (1) to obtain information on this novel aspect and (2) to determine to what extent this effect may play a role in selenium metabolism and whether it may have to be considered in the application of cell studies in selenoprotein research.

Materials and methods Animals and materials Animals Wistar rats were fed a low selenium diet with a selenium content of 8–15 μg/kg (ICN Biochemicals, Cleveland, Ohio, USA) for several generations. The composition of the diet and the animal care has been described in detail elsewhere (Behne et al. 1990). The experiments were approved by the Senatsverwaltung für Gesundheit und Soziales, Berlin, Germany (research project no. G 0266/00) as the institution responsible for animal research. Cell cultures The immortalized brain cell types used in the investigation included the neuronal cell line HT22 (mouse), the microglial cell line BV-2 (mouse), the astrocytoma cell line U373 (human), the oligodendrocyte cell line OLN-93 (rat) and the cerebral endothelial cell line rBCEC4 (rat). Production of

75

Se

Metallic 74Se, enriched from the natural abundance of 0.9% to more than 99%, was transformed into ammonium selenite and irradiated in the Berlin nuclear reactor BER II for several months at a neutron flux density of 1.3× 1014 cm−2 s−1. In this way, 75Se with an extremely high

Cell Tissue Res (2008) 332:403–414

specific activity was produced. It was dissolved in physiological saline. Methods Labelling and preparation of rat brain samples Three male selenium-deficient Wistar rats (240–270 g) were labelled in vivo by intraperitoneal injection of 75Se (18 MBq/μg selenium per animal). The rats were sacrificed after 6 days by anaesthesia with isoflurane (CuraMED Pharma, Karlsruhe, Germany) and cardiopuncture. Brains were divided into brain stem/medulla oblongata, cerebellum, hippocampus and cerebellar cortex, the frontal and temporal poles of which were separated by using a binocular set up. The regions were identified according to a rat brain atlas (Paxinos et al. 1980). The tissues were homogenized by means of a polytron (Kinematica, Heidelberg, Germany) in a threefold volume of a TRIS/HCl buffer (20 mM, pH 7.4). The cytosols were separated by centrifugation at 120,000g. Nuclear fractions were obtained by centrifugation of the resuspended pellets in a sucrose buffer solution (0.25 mM sucrose, 5 mM MgCl2, 20 mM TRIS/HCl) at 1,000g. Differential centrifugation of the supernatants at 10,000g and 120, 000g then resulted in the isolation of the mitochondrial and microsomal fractions, respectively. Labelling and preparation of brain cell line cultures All cell culture media contained 10% (v/v) fetal calf serum (FCS; Biochrom, Berlin, Germany). As most of their selenium stemmed from the FCS, all media had the same selenium concentration of 13.6 nM, as determined by atomic absorption spectrometry. The cerebral endothelial cells rBCEC4, kindly provided by I. E. Blasig (FMP Berlin, Germany), were cultured in a medium consisting of high glucose DMEM with 1.2 mM L-glutamine, 100 U penicillin/ml, 100 μg streptomycin/ml (all purchased from Invitrogen, Eggenstein, Germany), 100 μg heparin/ml, 110 μg sodium pyruvate/ml, 10 μg ECGF/ml (all purchased from Sigma, Deisenhofen, Germany), 2.5 μg amphoterizin B/ml (Biochrom) and 10% (v/v) FCS (charge A 661; Biochrom). The culture medium used for the other cell cultures was high glucose DMEM with 2 mM L-glutamine, 100 U penicillin/ml, 100 μg streptomycin/ml and 10% FCS. The cells were cultured at 37°C and 5% CO2. They were plated at a density of 1.4–2×104 cells/cm2 and labelled by adding 80 nM sodium selenite with 0.8 MBq 75Seselenite corresponding to 27 kBq 75Se/ml. After 72 h, the cells were homogenized in lysis buffer (20 mM TRIS/HCl, pH 7.4) by sonication. Three quarters of the lysate volumes

405

were taken for the separation of the cytosolic fractions by centrifugation at 120,000g. Co-culturing experiment For the isolation of primary astrocytes, the cerebrum of Wistar rats was taken on days 2 and 3 post natum and transferred into cooled PBS. After removal of the pia mater and arachnoid meninges, the tissue was homogenized by means of a fire-polished Pasteur pipette and the homogenate was centrifuged at 800g. The resuspended cells were plated on poly-L-lysine (10 μg/ml)-coated dishes (Nunc, Wiesbaden, Germany) and incubated at 37°C and 5% CO2 by using high glucose DMEM with 2 mM L-glutamine, 100 U penicillin/ml, 100 μg streptomycin/ml and 10% FCS as a culture medium. On days 7 and 10 after isolation, the dishes were shaken for 10 min and rinsed with warm PBS to remove microglial cells; the astrocytes were then replated in the culture medium at low density. Immunocytochemical staining showed that, of the cultured cells treated in this way, more than 90% reacted with anti-glial fibillary acidic protein, a marker for astrocytes and less than 5% with antiisolectin B4, a marker for microglia. After 12 days, the astrocytes were transferred onto the bottom surface of PET membranes of cell culture inserts (Falcon, Heidelberg, Germany) and cultured in the astrocyte medium. After 3 days, the cerebral endothelial cells (rBCEC4) were seeded on the top surface of the membrane and cultured in their medium. Both cell types were labelled by adding 800 kBq in 35 nM sodium selenite to the co-culture and harvested after 3 days. Monocultures of primary astrocytes and endothelial cells (rBCEC4) were labelled in the same way. The treatment of the harvested cells corresponded to that of the other cell lines described above. Gel electrophoretic separation Normalization of the protein content was performed in all samples was first performed by after determining the protein content by means of the Bradford method. Next, 50 μg of the protein mixture of rat brain samples or cell cultures was were separated either by using 15% onedimensional SDS-polyacrylamide gel electrophoresis (SDSPAGE) in 15% gels (Laemmli 1970) or by two-dimensional isoelectric focusing (IEF)/SDS-PAGE (Eckerskorn et al. 1988) with using the Multiphor II system (Amersham Biotech, Munich). The quality of the protein separation was established in each gel by the staining of the separated protein bands by means of Coomassie Blue or silver staining. For the determination of the molecular weight of the proteins separated by SDS-PAGE, a molecular weight marker kit (Sigma-Aldrich, Deisenhofen, Germany) was applied. In both methods, the gels were stained, dried

406

Cell Tissue Res (2008) 332:403–414

between sheets of cellophane in a gel-drying apparatus (Biometra, Göttingen. Germany) and prepared for documentation to be used for autoradiography. Determination of

75

Se

The tracer distribution in the separation gels was determined autoradiographically by means of photostimulable phosphorus plates (BAS 1000, Fuji Film, Tokyo, Japan), which in connection with an imaging analyzer (FLA-3000, Fuji Film, Tokyo, Japan) allowed the analysis of the 75Se activity in the separated labelled proteins. The quantitative evaluation was performed by means of the software AIDA 2.43 (Raytest, Straubenheim, Germany).

Results Expression of selenoproteins in brain cell cultures In order to obtain information concerning the selenium compounds in different types of brain cells, the selenoproteins present in a neuronal cell line (HT-22), a microglial cell line (BV-2), an oligodendrocyte cell line (OLN-93), an astrocytoma cell line (U-373) and a cerebral endothelial cell line (rBCEC4) were metabolically labelled with 75Se, separated by gel electrophoresis and analysed by autoradiography. The autoradiograms of the proteins in lysates and cytosols of these cell lines after SDS-PAGE are shown in Fig. 1. The molecular weights of the labelled bands and their relative tracer content, estimated by the degree of blackening in the autoradiograms, are listed in Table 1. Twenty selenium-containing bands could be distinguished after separation by SDS-PAGE. Of those six were Fig. 1 Selenoproteins in the lysate and cytosol of several brain cell lines determined autoradiographically after 75Selabelling and protein separation (50 μg protein per lane) by SDS-polyacrylamide gel electrophoresis (SDS-PAGE)

present in all cell types investigated. Distinct differences in the selenoprotein patterns between the cell lines found for the more strongly labelled bands include a 25-kDa band and a 13-kDa band present in HT-22, OLN-93 and rBCEC4 but not in the other two cell lines, a U-373 specific 21-kDa band and the lack of the 23-kDa-band in HT-22. Weakly labelled bands at 44 kDa, 12 kDa and 11 kDa were only visible in BV-2 and at 42–40 kDa only in HT-22. Further weakly labelled bands at 62–60 kDa, 34–33 kDa and 17 kDa were observed in some but not all cell lines. However, in the cases where the analytical signal was near the limit of detection, final conclusions on cell-specific expressions of the selenoproteins could not be reached. The labelled band at 18 kDa was detected in all cell lysates but not in the cytosols, a finding that indicated that this selenoprotein was membrane-bound. In addition to the investigations by SDS-PAGE, twodimensional IEF/SDS-PAGE was applied in the investigation of the cytosolic selenoproteins in the various cell lines. The autoradiograms of the separated proteins in the cytosols of HT-22 and rBCEC4 are shown as examples in Fig. 2. In all cell lines, two labelled spots with the molecular weight of 15 kDa and isoelectric point (IP)values of 4.6–4.8 and 7.8–8.2 were detected (Fig. 2a). In the rBCEC4 cells, two further 15-kDa spots appeared at IP 5.0–5.2 and 5.8–6.2 (Fig. 2b). Effects of co-culturing on selenoprotein patterns in brain cells New information on interactions between different types of brain cells with regard to selenoprotein expression was obtained by comparing the patterns of 75Se-labelled proteins of primary astrocytes and the endothelial cell line

Cell Tissue Res (2008) 332:403–414

407

Table 1 Distribution of 75Se among labelled selenoprotein bands in several brain cell lines as estimated from the autoradiograms in Fig. 1 (l lysate, c cytosol, empty boxes no detected labelling, − slight labelling, (+) weak labelling, + labelling, ++ strong labelling) 75

Se-labeled bands

74 kDa 62–60 kDa 57–53 kDa 48–46 kDa 44 kDa 42–40 kDa 34–33 kDa 25 kDa 24 kDa 23 kDa 21 kDa 20 kDa 18 kDa 17 kDa 16 kDa 15 kDa 13 kDa 12 kDa 11 kDa 10 kDa

HT-22

BV-2

l

c

l

+

(+)

(+)

++ + (+) (+) ++ +

++ + (+) ++ +

++ ++ (+)

+

++ ++

+ (+)

+

+

+

++ + (+)

(+) ++

OLN-93 c

++ + (+)

++ (+) -

++

(+) ++ (+) ++

++ (+) (+)

+ (+)

+

(+)

+ + (+)

+ + (+)

++ ++

rBCEC4

c

l

c

++

(+) (+) +++

(+) +++

-

-

-

++

rBCEC4 grown either as mono-cultures or as a co-culture. The autoradiograms of the cytosolic proteins after separation by SDS-PAGE (Fig. 3) show distinct changes in both cell types in the co-culture with regard to the selenoprotein bands at around 25 kDa, with a shift in the tracer distribution to a protein with a slightly higher molecular weight. Another change attributable to co-culturing was a more intensively labelled selenoprotein in the 15-kDa range in the endothelial cells. The same effect was observed in a mono-culture of rBCEC4 grown in a culture medium that consisted to (40%) of the conditioned medium of the Fig. 2 Selenoproteins in the cytosol of the neuronal cell line HT-22 (a) and the cerebral endothelial cell line rBCEC4 (b) determined autoradiographically after 75Se-labelling and twodimensional protein separation. Two 15-kDa selenoprotein spots (black arrows, red circles) were found in all brain cell lines investigated, two additional 15-kDa selenoproteins (grey arrows, blue circles) were only detected in rBCEC4

l

U-373

(+) ++ (+)

+

+++ ++

+ (+)

++

+

++ ++ ++ ++ + (+) (+) +

++ ++ ++ ++

+

+

(+) (+) +

l

c

++ (+) ++ (+) -

(+) (+) ++ (+) -

(+) ++ (+) ++

++ +

++ ++

+

+ ++ ++

+ + (+)

(+)

(+)

primary astrocytes. We therefore concluded that the increased expression of this selenoprotein was caused directly by a mediator released from the astrocytes, rather than by other interactions between the two cell types. Expression of selenoproteins in various rat brain regions Further information on the selenoproteome in the CNS was obtained by investigating the selenoproteins in several brain regions of 75Se-labelled rats. The autoradiograms of the labelled proteins after SDS-PAGE in the homogenates and

408 Fig. 3 Selenoprotein expression in mono-cultures of primary astrocytes and the cerebral endothelial cell line rBCEC4 and the co-culture of the two cell types as determined autoradiographically after 75Se-labelling and separation of the proteins (50 μg per lane) by SDS-PAGE

cytosolic, mitochondrial, microsomal and nuclear fractions of the frontal pole of the cortex, temporal pole of the cortex, cerebellum, brain stem/medulla oblongata and hippocampus are shown in Fig. 4a-e. The tracer distribution in the homogenates was similar, with distinct bands at 74, 57, 50, 25, 23, 20, 18, 15 and 12–10 kDa being found in all regions. As in the cell lines, the 18-kDa selenoprotein was not present in the cytosols. In selenium-deficient animals, most of the metabolized selenium is not excreted but is reutilized and incorporated once again into selenoproteins (Behne and Hofer-Bosse 1984). 75Se administered to deficient animals is therefore distributed relatively quickly among the selenoproteins in the same way as native selenium, irrespective of differences in their turnover rate. Thus, quantitative information on the distribution of the selenoproteins in the subcellular fractions can be obtained by determining the 75Se activity in the bands as a percentage of the total protein-bound 75Se activity in the lane in question (see Fig. 4f-j). Among the most strongly labelled proteins (more than 15% of the total 75Se activity) found in all brain regions were the 57-kDa band in the cytosols (25%–35%), the 20kDa band in the mitochondria (30%–35%), microsomes (20%–25%) and nuclei (30%–50%), the 18-kDa band in the mitochondria (25%–30%), microsomes (15%–25%) and nuclei (15%–25%) and the 15-kDa band in the cytosols (25%–40%) and microsomes (15%–25%). The most prominent region-specific differences were the intense labelling

Cell Tissue Res (2008) 332:403–414

of the 15-kDa band in the cytosol of the cerebellum, with about 40% of the total protein-bound 75Se activity compared with about 25% in the other cytosols (red rectangle in Fig. 4b), the 20-kDa band in the nuclei of the brain stem and hippocampus (50% and 40% compared with about 30% in the other three nuclear fractions) and the 23kDa band in the cytosol of the brain stem (20% compared with values between 10% and 15% in the other cytosols). The distribution of the selenoproteins in the homogenates and cytosols of the frontal and temporal poles of the cortex, hippocampus, cerebellum and brain stem/medulla oblongata were also analysed after two-dimensional separation. An example of an autoradiogram of the temporal pole of the cortex homogenate, on which the selenoprotein spots are indicated by circles, is presented in Fig. 5a. Overall, 21 spots originating from selenium-containing proteins or protein subunits could be distinguished in the various homogenates. In the autoradiograms of the cytosols, the 18 kDa selenoprotein was missing; this protein has previously been shown to be membrane-bound (Kyriakopoulos et al. 2002). The same was true with the 15-kDa spot with an IP-value of 5.8–6.2, but an additional 15-kDa spot with an IP-value of 5.0–5.2 was visible. This spot was much more strongly labelled in the cytosol of the cerebellum than in the other cytosols investigated (Fig. 5d). The selenoproteins found after two-dimensional separation in the homogenates and cytosols of the different brain regions, as characterized by the ranges of their molecular weights and IP-values, are compiled in Table 2. Comparison of selenoprotein spots with known selenoproteins For allocation to the known selenoproteins and those identified in silico, we compared the apparent molecular masses of the selenoprotein spots found in SDS-PAGE with the molecular masses as determined or calculated in other studies. These data are listed in Table 3.

Discussion The finding that the brain is preferentially supplied with selenium has suggested important functions of the selenoproteins in this organ (Behne et al. 1988). Information

Fig. 4 Selenoprotein expression in subcellular fractions of several rat„ brain regions. a-e Autoradiograms of selenoproteins after labelling rats in vivo with 75Se; proteins (50 μg per lane) were separated by SDSPAGE (red rectangle intense labelling of the 15-kDa band in the cytosol of the cerebellum). f-j Quantitative determination of the 75Se activity in the protein bands expressed as a percentage of the total protein-bound 75Se activity in each lane

Cell Tissue Res (2008) 332:403–414

409

410

Cell Tissue Res (2008) 332:403–414

Fig. 5 Selenoprotein expression in rat brain fractions determined autoradiographically by labelling rats in vivo with 75Se and two-dimensional protein separation. a Example of the selenoprotein pattern in the temporal pole of cortex protein homogenate, with the protein spots (black circles) identified in this brain area and listed in Table 2 (red circles selenoproteins absent from this protein sample, but detectable in other brain homogenates, as also listed in Table 2). b-d Differences in expression of selenoproteins, with an 18-kDa selenoprotein and a 15-kDa selenoprotein (IP 5.8–6.2) found in the temporal pole of the cortex homogenate but not in the cytosol (black arrows, red circles) and a 15-kDa selenoprotein (IP 5.0– 5.2) most strongly evident in the cerebellum cytosol (grey arrow, blue circle)

concerning the presence of several of these compounds in the brain has been derived from mRNA analyses for proteins, such as glutathione peroxidase (Dreher et al. 1997; Mitchell et al. 1998), TR (Whanger 2001; Jeong et al. 2004), selenoprotein W (Sun et al. 2001; Jeong et al. 2004; Loflin et al. 2006), type 2 deiodinase (Salvatore et al. 1996), selenoprotein T (Ikematsu et al. 2007), selenoprotein P (Saijoh et al. 1995; Scharpf et al. 2007), methionine sulphoxide reductase B (Moskovitz and Stadtman 2003) and the 15-kDa selenoprotein (Kumaraswamy et al. 2000). However, knowledge with regard to the actual expression in the CNS has so far been restricted to a few selenoproteins that have been analysed by specific antibodies; these include gluthathione peroxidase (Damier et al. 1993; Takizawa et al. 1994), selenoprotein W (Gu et al. 2000) and selenoprotein P (Burk et al. 1997; Scharpf et al. 2007). More extensive information on the expressed selenoproteins in the brain has been obtained in the present study by the labelling of rats with 75Se. In order to increase the

retention of the tracer, strongly selenium-depleted animals have been used and the power of detection has further been improved by the administration of 75Se with an extremely high specific activity. The results show that a larger number of selenoproteins is expressed in the CNS. So far, they have been characterized by their molecular masses and their IPvalues. Because of to the similarities in their molecular masses, some of them have been allocated to known selenoproteins. Of special interest is the detection of several selenoproteins with molecular weights in the 14–15 kDa range. Two 15-kDa selenoproteins have previously been described: (1) the 15-kDa selenoprotein found in the rat has with an IP-value of 4.5–4.7 and a native molecular weight of 250 kDa (Kalcklosch et al. 1995) and that in humans (Sep15) has with an IP-value of 4.7 and a native molecular weight of 200–250 kDa (Gladyshev et al. 1998) and (2) the other 15-kDa selenoprotein has with an IP-value of 7.0 and a native molecular mass of 30 kDa found in the rat where it is enriched in the brain (Röthlein 1999). Other selenopro-

Cell Tissue Res (2008) 332:403–414

411

Table 2 Selenoproteins found in rat brain by means of 75Se-labelling and two-dimensional protein separation (IP isoelectric point) Spot number

Approximate molecular weight (kDa)

IP-range

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

74 56–58 54–57 53–56 53–56 48 46–47 45–46 44–45 32 27–28 25 23 20 18 14–15 14–15 14–15 14–15 14–15 10

4.8–5.0 5.8–6.0 5.5–5.6 5.3–5.5 5.3–5.5 6.0–6.4 5.3–5.5 5.1–5.2 4.9–5.0 5.0–5.2 5.0–5.2 6.2–6.8 6.2–6.8 7.4–7.8 7.2–8.0 5.8–6.2 5.0–5.2 6.8–7.2 6.8–7.2 4.6–4.8 8.5–8.6

tein spots in this molecular weight range have not previously been detected. Sep15 has been purified in a complex with UDP-glucose:glycoprotein glucosyltransferase, which is known to be involved in the quality control of protein folding. The function of Sep15 in this complex Table 3 Comparison of selenoprotein spots found in rat brain with known selenoproteins by means of similarities in their molecular masses (Sel selenoprotein, D deiodinase, TR thioredoxin reductase, SPS2 selenophosphate synthetase 2, GPx glutathione peroxidases, SePP selenoprotein P)

might be the recognition and/or control of protein modification (Korotkov et al. 2001; Labunskyy et al. 2005). Nothing is known as yet about the characteristics of the newly found 15-kDa selenoproteins. However, as the two 15-kDa selenoproteins with IP-values of 5.0–5.2 and 5.8– 6.2 have only been detected in the endothelial cells out of the five cell lines investigated, these compounds might be expressed in a cell-type-specific manner. The selenoproteins listed in Table 2 are present in all brain regions but differ with regard to the degree of their expression among the regions and within a region among the subcellular fractions. It has to be borne in mind here that, in selenium-deficient animals, the tracer distribution does not reflect the selenoprotein concentrations in the normal state as, because of the hierarchy in the selenium supply, the selenoproteins low in the ranking order such as GPx1 are expressed to a lesser extent during periods of insufficient selenium intake (Kyriakopoulos and Behne 2002). Information can therefore be obtained in this way with regard to the preferentially supplied selenoproteins and their specific sites. These include GPx4 (20 kDa), which has been found to be highly expressed in the microsomal, mitochondrial and nuclear fractions, the last-mentioned showing elevated levels in the hippocampus and brain stem samples (Fig. 4). It is worthy of note in this respect, that the disruption of the GPx4 gene results in early embryonic lethality (Imai et al. 2003; Yant et al. 2003). Other highly expressed selenoproteins are those in the 57-kDa band in the cytosols (probably members of the TR family), the 18-kDa

Spot

Predicted selenoprotein

Molecular mass (Reference)

1 2–5 6–9

SelO TR1 and TR3 homodimer subunits and/or SePP Sel I and/or SPS2

10

D1 or D2

11–13

GPx-1–3 and/or SelV

14

GPx-4 and/or SelS and/or SelT

15

SelM and/or Sel 18

16–20

SelH and/or Sel15 and/or SelR

21

SelR and/or SelW and/or SelK

SelO≅73 kDa (Kryukov et al. 2003) TR1 and TR3≅55 (Sun et al. 2001) SePP≅57 kDa (Scharpf et al. 2007) Sel I≅45 kDa (Kryukov et al. 2003) SPS2≅48 kDa (Kim et al. 1997) D1≅29 kDa (Curcio-Morelli et al. 2003) D2≅31 kDa (Curcio-Morelli et al. 2003) GPx1–3≅22–23 kDa (Ballihaut et al. 2007) GPx1≅25 kDa (Kryukov et al. 1999) SelV≅25 kDa (Kryukov et al. 2003) GPx4≅20–22 kDa (Ballihaut et al. 2007) SelS≅21 kDa (Kryukov et al. 2003) SelT≅19 kDa (Kryukov et al. 1999) SelM≅17 kD (Korotkov et al. 2002) Sel18≅18 kDa (Kyriakopoulos et al. 2002) SelH≅13 kDa (Kryukov et al. 2003) Sel15≅15 kDa (Behne and Kyriakopoulos 2001; Novoselov et al. 2006) SelR≅16 kDa (Lescure et al. 1999) SelR≅12 kDa (Kryukov et al. 2002) SelW≅10 kDa (Kryukov et al. 2003) SelK≅9 kDa (Kryukov et al. 2003)

412

selenoprotein (probably selenoprotein T) in the non-cytosolic fractions and 15-kDa selenoproteins in the cytosolic and microsomal fractions. The enrichment of the 15-kDa selenoprotein with an IP-value of 5.0–5.2 in the cytosol of the cerebellum is of special interest as it suggests a specific function of this selenoprotein in this brain compartment. The autoradiograms of the brain regions and several brain cell lines after protein separation by SDS-PAGE have shown similar distribution patterns of the radiotracer for most of the more strongly labelled selenoproteins. Exceptions are a 21-kDa band found only in the astrocytoma cell line and a 13-kDa band in the oligodendrocyte, endothelial and neuronal cell lines. Several of the more weakly labelled bands in the cell lines have likewise not been observed in the brain samples. The question of whether these compounds are selenoproteins present in certain cell types in the in vivo state but undetectable in brain samples, because of the decrease in the selenoprotein:total protein ratio by proteins from other tissue components, or whether they are only produced in vitro in the cell lines still has to be clarified. An interesting phenomenon is the difference in the distribution pattern of selenoproteins found between monocultured and co-cultured endothelial cells and primary astrocytes (Fig. 3). These two cell types are constituents of the blood-brain-barrier and thus these interactions might represent specific processes related to the passage of selenium through this barrier and its incorporation into the CNS. However, the selenoprotein expression of a certain cell type may also be affected by mediators produced by other cells. Overall, this study has shown that a multitude of selenoproteins are expressed in the brain and provides information regarding their distribution among the different regions and subcellular fractions. Further experiments are planned for a more detailed characterization in order to be able to distinguish between known selenoproteins and novel modifications. Functional tests also have to be carried out to study alterations in expression during co-cultivation and to investigate the significance of this effect in selenium metabolism. Acknowledgements We thank G. Niggemann and J. Franke, Nuklearmedizin, Charité-Universitätsmedizin Berlin, Campus Benjamin Franklin, for their help with the animal experiments. We are also indebted to I. E. Blasig, Forschungsinstitut für Molekulare Pharmakologie, Berlin for kindly providing the cerebral endothelial cell line and to D. Richter and E. Kwidzinski, Institut für Anatomie, Charité-Universitätsmedizin Berlin for their help in the preparation of primary astrocytes.

References Aguirre P, Mena N, Tapia V, Rojas A, Arredondo M, Nunez MT (2006) Antioxidant responses of cortex neurons to iron loading. Biol Res 39:103–104

Cell Tissue Res (2008) 332:403–414 Ballihaut G, Mounicou S, Lobinski R (2007) Multitechnique massspectrometric approach for the detection of bovine glutathione peroxidase selenoprotein: focus on the selenopeptide. Anal Bioanal Chem 388:585–591 Behne D, Wolters W (1983) Distribution of selenium and glutathione peroxidase in the rat. J Nutr 113:456–461 Behne D, Hofer-Bosse T (1984) Effects of a low selenium status on the distribution and retention of selenium in the rat. J Nutr 114:1289–1296 Behne D, Kyriakopoulos A (2001) Mammalian selenium-containing proteins. Annu Rev Nutr 21:453–473 Behne D, Hilmert H, Scheid S, Gessner H, Elger W (1988) Evidence for specific selenium target tissues and new biologically important selenoproteins. Biochim Biophys Acta 966: 12–21 Behne D, Scheid S, Hilmert H, Gessner H, Gawlik D, Kyriakopoulos A (1990) Combination of neutron activation analysis, tracer techniques, and biochemical methods in the investigation of selenium metabolism. Biol Trace Elem Res 26–27:439–447 Benton D (2002) Selenium intake, mood and other aspects of psychological functioning. Nutr Neurosci 5:363–374 Berry MJ, Banu L, Chen YY, Mandel SJ, Kieffer JD, Harney JW, Larsen PR (1991) Recognition of UGA as a selenocysteine codon in type I deiodinase requires sequences in the 3' untranslated region. Nature 353:273–276 Borchert A, Wang CC, Ufer C, Schiebel H, Savaskan NE, Kuhn H (2006) The role of phospholipid hydroperoxide glutathione peroxidase isoforms in murine embryogenesis. J Biol Chem 281:19655–19664 Brauer AU, Savaskan NE (2004) Molecular actions of selenium in the brain: neuroprotective mechanisms of an essential trace element. Rev Neurosci 15:19–32 Brenneisen P, Steinbrenner H, Sies H (2005) Selenium, oxidative stress, and health aspects. Mol Aspects Med 26:256–267 Burk RF, Hill KE, Boeglin ME, Ebner FF, Chittum HS (1997) Selenoprotein P associates with endothelial cells in rat tissues. Histochem Cell Biol 108:11–15 Chen J, Berry MJ (2003) Selenium and selenoproteins in the brain and brain diseases. J Neurochem 86:1–12 Curcio-Morelli C, Gereben B, Zavacki AM, Kim BW, Huang S, Harney JW, Larsen PR, Bianco AC (2003) In vivo dimerization of types 1, 2, and 3 iodothyronine selenodeiodinases. Endocrinology 144:937–946 Damier P, Hirsch EC, Zhang P, Agid Y, Javoy-Agid F (1993) Glutathione peroxidase, glial cells and Parkinson’s disease. Neuroscience 52:1–6 Dreher I, Schmutzler C, Jakob F, Kohrle J (1997) Expression of selenoproteins in various rat and human tissues and cell lines. J Trace Elem Med Biol 11:83–91 Dringen R, Pawlowski PG, Hirrlinger J (2005) Peroxide detoxification by brain cells. J Neurosci Res 79:157–165 Eckerskorn C, Mewes W, Goretzki H, Lottspeich F (1988) A new siliconized-glass fiber as support for protein-chemical analysis of electroblotted proteins. Eur J Biochem 176:509–519 Ganther HE (2001) Selenium metabolism and mechanisms of cancer prevention. Adv Exp Med Biol 492:119–130 Gladyshev VN, Jeang KT, Wootton JC, Hatfield DL (1998) A new human selenium-containing protein. Purification, characterization, and cDNA sequence. J Biol Chem 273:8910–8915 Gu QP, Sun Y, Ream LW, Whanger PD (2000) Selenoprotein W accumulates primarily in primate skeletal muscle, heart, brain and tongue. Mol Cell Biochem 204:49–56 Haan JB de, Bladier C, Griffiths P, Kelner M, O'Shea RD, Cheung NS, Bronson RT, Silvestro MJ, Wild S, Zheng SS, Beart PM, Hertzog PJ, Kola I (1998) Mice with a homozygous null mutation for the most abundant glutathione peroxidase, Gpx1, show increased

Cell Tissue Res (2008) 332:403–414 susceptibility to the oxidative stress-inducing agents paraquat and hydrogen peroxide. J Biol Chem 273:22528–22536 Hawkes WC, Hornbostel L (1996) Effects of dietary selenium on mood in healthy men living in a metabolic research unit. Biol Psychiatry 39:121–128 Hill KE, Zhou J, McMahan WJ, Motley AK, Burk RF (2004) Neurological dysfunction occurs in mice with targeted deletion of the selenoprotein P gene. J Nutr 134:157–161 Ikematsu K, Tsuda R, Tsuruya S, Nakasono I (2007) Identification of novel genes expressed in hypoxic brain condition by fluorescence differential display. Forensic Sci Int 169:168–172 Imai H, Hirao F, Sakamoto T, Sekine K, Mizukura Y, Saito M, Kitamoto T, Hayasaka M, Hanaoka K, Nakagawa Y (2003) Early embryonic lethality caused by targeted disruption of the mouse PHGPx gene. Biochem Biophys Res Commun 305:278–286 Jeong DW, Kim EH, Kim TS, Chung YW, Kim H, Kim IY (2004) Different distributions of selenoprotein W and thioredoxin during postnatal brain development and embryogenesis. Mol Cells 17:156–159 Kalcklosch M, Kyriakopoulos A, Hammel C, Behne D (1995) A new selenoprotein found in the glandular epithelial cells of the rat prostate. Biochem Biophys Res Commun 217:162–170 Kim IY, Guimaraes MJ, Zlotnik A, Bazan JF, Stadtman TC (1997) Fetal mouse selenophosphate synthetase 2 (SPS2): characterization of the cysteine mutant form overproduced in a baculovirusinsect cell system. Proc Natl Acad Sci USA 94:418–421 Klivenyi P, Andreassen OA, Ferrante RJ, Dedeoglu A, Mueller G, Lancelot E, Bogdanov M, Andersen JK, Jiang D, Beal MF (2000) Mice deficient in cellular glutathione peroxidase show increased vulnerability to malonate, 3-nitropropionic acid, and 1methyl-4-phenyl-1,2,5,6-tetrahydropyridine. J Neurosci 20:1–7 Kohrle J (2000) Thyroid hormone metabolism and action in the brain and pituitary. Acta Med Austriaca 27:1–7 Korotkov KV, Kumaraswamy E, Zhou Y, Hatfield DL, Gladyshev VN (2001) Association between the 15-kDa selenoprotein and UDPglucose:glycoprotein glucosyltransferase in the endoplasmic reticulum of mammalian cells. J Biol Chem 276:15330–15336 Korotkov KV, Novoselov SV, Hatfield DL, Gladyshev VN (2002) Mammalian selenoprotein in which selenocysteine (Sec) incorporation is supported by a new form of Sec insertion sequence element. Mol Cell Biol 22:1402–1411 Kryukov GV, Kryukov VM, Gladyshev VN (1999) New mammalian selenocysteine-containing proteins identified with an algorithm that searches for selenocysteine insertion sequence elements. J Biol Chem 274:33888–33897 Kryukov GV, Kumar RA, Koc A, Sun Z, Gladyshev VN (2002) Selenoprotein R is a zinc-containing stereo-specific methionine sulfoxide reductase. Proc Natl Acad Sci USA 99:4245–4250 Kryukov GV, Castellano S, Novoselov SV, Lobanov AV, Zehtab O, Guigo R, Gladyshev VN (2003) Characterization of mammalian selenoproteomes. Science 300:1439–1443 Kumaraswamy E, Malykh A, Korotkov KV, Kozyavkin S, Hu Y, Kwon SY, Moustafa ME, Carlson BA, Berry MJ, Lee BJ, Hatfield DL, Diamond AM, Gladyshev VN (2000) Structureexpression relationships of the 15-kDa selenoprotein gene. Possible role of the protein in cancer etiology. J Biol Chem 275:35540–35547 Kyriakopoulos A, Behne D (2002) Selenium-containing proteins in mammals and other forms of life. Rev Physiol Biochem Pharmacol 145:1–46 Kyriakopoulos A, Bertelsmann H, Graebert A, Hoppe B, Kuhbacher M, Behne D (2002) Distribution of an 18 kDa-selenoprotein in several tissues of the rat. J Trace Elem Med Biol 16:57–62 Labunskyy VM, Ferguson AD, Fomenko DE, Chelliah Y, Hatfield DL, Gladyshev VN (2005) A novel cysteine-rich domain of

413 Sep15 mediates the interaction with UDP-glucose:glycoprotein glucosyltransferase. J Biol Chem 280:37839–37845 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685 Lescure A, Gautheret D, Carbon P, Krol A (1999) Novel selenoproteins identified in silico and in vivo by using a conserved RNA structural motif. J Biol Chem 274:38147–38154 Loflin J, Lopez N, Whanger PD, Kioussi C (2006) Selenoprotein W during development and oxidative stress. J Inorg Biochem 100:1679–1684 Lovell MA, Xie C, Gabbita SP, Markesbery WR (2000) Decreased thioredoxin and increased thioredoxin reductase levels in Alzheimer’s disease brain. Free Radic Biol Med 28:418–427 Mitchell JH, Nicol F, Beckett GJ, Arthur JR (1998) Selenoprotein expression and brain development in preweanling selenium- and iodine-deficient rats. J Mol Endocrinol 20:203–210 Moskovitz J, Stadtman ER (2003) Selenium-deficient diet enhances protein oxidation and affects methionine sulfoxide reductase (MsrB) protein level in certain mouse tissues. Proc Natl Acad Sci USA 100:7486–7490 Novoselov SV, Hua D, Lobanov AV, Gladyshev VN (2006) Identification and characterization of Fep15, a new selenocysteine-containing member of the Sep15 protein family. Biochem J 394:575–579 Paxinos G, Watson CR, Emson PC (1980) AChE-stained horizontal sections of the rat brain in stereotaxic coordinates. J Neurosci Methods 3:129–149 Ramaekers VT, Calomme M, Vanden Berghe D, Makropoulos W (1994) Selenium deficiency triggering intractable seizures. Neuropediatrics 25:217–223 Röthlein D (1999) Charakterisierung von 15 kD-Selenoproteinen in Geweben der Ratte. PhD-thesis, Freie Universität Berlin, Berlin Saijoh K, Saito N, Lee MJ, Fujii M, Kobayashi T, Sumino K (1995) Molecular cloning of cDNA encoding a bovine selenoprotein Plike protein containing 12 selenocysteines and a (His-Pro) rich domain insertion, and its regional expression. Brain Res Mol Brain Res 30:301–311 Salvatore D, Bartha T, Harney JW, Larsen PR (1996) Molecular biological and biochemical characterization of the human type 2 selenodeiodinase. Endocrinology 137:3308–3315 Savaskan NE, Brauer AU, Kuhbacher M, Eyupoglu IY, Kyriakopoulos A, Ninnemann O, Behne D, Nitsch R (2003) Selenium deficiency increases susceptibility to glutamate-induced excitotoxicity. FASEB J 17:112–114 Savaskan NE, Ufer C, Kuhn H, Borchert A (2007) Molecular biology of glutathione peroxidase 4: from genomic structure to developmental expression and neural function. Biol Chem 388:1007–1017 Scharpf M, Schweizer U, Arzberger T, Roggendorf W, Schomburg L, Kohrle J (2007) Neuronal and ependymal expression of selenoprotein P in the human brain. J Neural Transm 114:877–884 Schomburg L, Schweizer U, Holtmann B, Flohe L, Sendtner M, Kohrle J (2003) Gene disruption discloses role of selenoprotein P in selenium delivery to target tissues. Biochem J 370:397–402 Schweizer U, Brauer AU, Kohrle J, Nitsch R, Savaskan NE (2004) Selenium and brain function: a poorly recognized liaison. Brain Res Brain Res Rev 45:164–178 Squires J, Berry MJ (2006) Selenium, selenoproteins, and cancer. Hawaii Med J 65:239–240 Stadtman TC (1996) Selenocysteine. Annu Rev Biochem 65:83–100 Steinbrenner H, Alili L, Bilgic E, Sies H, Brenneisen P (2006) Involvement of selenoprotein P in protection of human astrocytes from oxidative damage. Free Radic Biol Med 40:1513–1523 Sun QA, Zappacosta F, Factor VM, Wirth PJ, Hatfield DL, Gladyshev VN (2001) Heterogeneity within animal thioredoxin reductases. Evidence for alternative first exon splicing. J Biol Chem 276: 3106–3114

414 Takizawa S, Matsushima K, Shinohara Y, Ogawa S, Komatsu N, Utsunomiya H, Watanabe K (1994) Immunohistochemical localization of glutathione peroxidase in infarcted human brain. J Neurol Sci 122:66–73 Walczak R, Carbon P, Krol A (1998) An essential non-Watson-Crick base pair motif in 3'UTR to mediate selenoprotein translation. RNA 4:74–84 Weber GF, Maertens P, Meng XZ, Pippenger CE (1991) Glutathione peroxidase deficiency and childhood seizures. Lancet 337:1443–1444

Cell Tissue Res (2008) 332:403–414 Whanger PD (2001) Selenium and the brain: a review. Nutr Neurosci 4:81–97 Yan J, Barrett JN (1998) Purification from bovine serum of a survivalpromoting factor for cultured central neurons and its identification as selenoprotein-P. J Neurosci 18:8682–8691 Yant LJ, Ran Q, Rao L, Van Remmen H, Shibatani T, Belter JG, Motta L, Richardson A, Prolla TA (2003) The selenoprotein GPX4 is essential for mouse development and protects from radiation and oxidative damage insults. Free Radic Biol Med 34:496–502