Role of Mitochondrial Uncoupling Protein 4 in Rat Inner Ear

Molecular and Cellular Neuroscience 47 (2011) 244–253

Contents lists available at ScienceDirect

Molecular and Cellular Neuroscience

journal homepage: www.elsevier.com/locate/ymcne

Role of mitochondrial uncoupling protein 4 in rat inner ear

Alina Smorodchenko a,c, Anne Rupprecht a,c, Julia Fuchs b, Johann Gross b, Elena E. Pohl a,c,⁎ a Department of Physiology and Biophysics, University of Veterinary Medicine, Vienna, Austria b Department of Othorhinolaryngology, Charité—Universitaetsmedizin Berlin, Germany c Institute of Cell Biology and Neurobiology, Charité—Universitaetsmedizin Berlin, Germany article info abstract

Article history: The uncoupling protein 4 (UCP4) belongs to the mitochondrial anion transporter family. Protein tissue Received 3 September 2010 distribution and functions are still a matter of debate. Using an antibody we have previously shown that UCP4 Revised 3 March 2011 appears in neurons and to a lesser extent in astrocytes of murine neuronal tissue as early as days 12–14 of Accepted 3 March 2011 embryonic development (Smorodchenko et al., 2009). Here we demonstrated for the first time that Available online 11 March 2011 neurosensory cells such as hair cells of the inner ear and mechanosensitive Merkel cells in skin also express a significant amount of UCP4. We tested the hypothesis about whether UCP4 contributes to the regulation of Keywords: Mitochondrial membrane protein oxidative stress using the model of oxygen deprivation. For this we compared the protein expression level in Voltage dependent anion channel freshly isolated explants of organ of Corti, modiolus and stria vascularis from neonatal rats with explants Spiral ganglion neurons cultured under hypoxia. Western blot analysis revealed that the UCP4 level was not increased under hypoxic Cochlea development conditions, when compared to the mitochondrial outer membrane protein VDAC or to the anti-oxidative Hypoxia enzyme SOD2. We moreover demonstrated that UCP4 expression is differently regulated during postnatal Oxidative stress stages and is region-specific. We hypothesized that UCP4 may play an important role in functional maturation Reactive oxygen species of the rat inner ear. Hair cells © 2011 Published by Elsevier Inc. Merkel cell Sensory cell

Introduction of fetal and adult animals with the maximal content in the cortex. Based on the finding that UCP4 expression coincided with the UCP4 is a mitochondrial inner membrane protein, which belongs beginning of neurogenesis, we recently hypothesized that the to the anion carrier family. The physiological role of UCP4 is still not protein may be involved in neuronal cell differentiation and/or clear, but several putative functions such as a regulation of free apoptosis (Smorodchenko et al., 2009). Zhang et al. (2006) earlier radicals production and calcium homeostasis (Chan et al., 2006; Wu reported that the overexpression of UCP4 in preadipocyte inhibits et al., 2009), involvement in apoptosis in the mammalian brain (Mao their differentiation into adipocytes, but stimulates preadipocyte et al., 1999; Hanak and Jezek, 2001), modulation of neuronal activity proliferation and protects preadipocytes from apoptosis induced by (Mattson and Liu, 2003) and participation in thermogenesis (Liu et al., serum deprivation. 2006) are in discussion. UCP4 mRNA in the rat inner ear was first described by Recently, we designed a pure affinity anti-UCP4 antibody and Kitahara et al. (Kitahara et al., 2004). In subsequent reports showed that the protein is mainly expressed in neuronal mito- Kitahara et al. (Kitahara et al., 2005, 2007) claimed that different chondria and to a lesser extent in astrocytes (Smorodchenko et al., UCP subfamily members including UCP4 contribute to the 2009). The protein expression level varied in different brain regions neuroprotection against the pharmacologically (kanamycin intox- ication) and mechanically (labyrinthectomy) induced oxidative damage and have a signaling role for the neuromodulation in Abbreviations: AB, antibody; CK18, cytokeratin 18; COX IV, mitochondrial Complex vestibular nerve. ′ IV; DAPI, (4 ,6-diamidino-2-phenylindole, dihydrochloride salt); Dcx, doublecortin; To investigate the UCP4 expression in different structures of the GFAP, glial fibrillar acidic protein; IHC, inner hair cells; MOD, modiolus; NF200, neurofilaments 200; OC, organ of Corti; OHC, outer hair cells; PB, phosphate buffer; PC, inner ear at the protein level and to examine its role in ROS pillar cells; PhC, phalangeal cells; PFA, paraformaldehyde; ROS, reactive oxygen species; regulation we now employ a well-established model of hypoxia SD, standard deviation; SEM, standard errors of mean; SG, spiral ganglion; SL, spiral (Gross et al., 2007). We compared the UCP4 expression in early limbus; SOD2, superoxide dismutase 2; SV, stria vascularis; VDAC, voltage dependent (postnatal day 2, P2), late postnatal (P5, P8, prior to the onset of anion channel; UCP, uncoupling protein; WB, Western blot. hearing) and adult (P28) rat cochlea to evaluate further hypotheses ⁎ Corresponding author at: Department of Physiology and Biophysics, University of Veterinary Medicine, 1210 Vienna, Austria. concerning the possible UCP4 participation in postnatal develop- E-mail address: [email protected] (E.E. Pohl). ment of the inner ear.

Results spiral ganglion (SG, blue) and axons (black). WB analysis (Fig. 1B) shows that UCP4 is expressed as a band of approximately 36 kDa with Pattern of UCP4 distribution in the cochlea under physiological the highest level in OC, less in MOD. No UCP4 was detected in SV of P3 conditions rat cochlea. To obtain the pattern of protein distribution in different inner ear structures, we performed the immunohistochemistry of First, we have examined the cellular distribution of UCP4 in Wistar cryosections from P3 and P28 rats' cochleae. Fig. 1C (left panel) and rat cochlea using highly speciﬁc anti-UCP4 antibody (Smorodchenko Fig. S1 show the UCP4 immunoreactivity in OC, SG, SL and inner sulcus et al., 2009). The protein content in structures of newborn (postnatal cells of P3 animals. UCP4 is localized in mitochondria as conﬁrmed by day 3, P3) and adult (postnatal day 28, P28) rats inner ear was a co-localization with mitochondrial porin VDAC (Fig. S1B). No UCP4 analyzed by WB and immunohistochemistry. was detected in SV (data not shown) and in SL of P28 rats Fig. 1C, right The schematic drawing (Fig. 1A) shows the cochlea emphasizing panel). Protein expression in cochleae of adult rats was mostly the structures on which we focused: spiral limbus (SL, gray), organ of restricted to spiral ganglion neurons and structures of OC (Fig. 1C, Corti (OC, pink), stria vascularis (SV, yellow), modiolus (MOD, green), right panel, Fig. 2A). The high-resolution image of newborn cochlea

Fig. 1. UCP4 distribution in cochlea under physiological conditions. A. Schematic drawing shows the cochlea. SL-spiral limbus (gray), axons (black), MOD-modiolus (green), OC-organ of Corti (pink), SV-stria vascularis (yellow), SG-spiral ganglion (blue). B. Representative Western blot of lysates from freshly isolated organ of Corti (OC) with spiral limbus (SL), modiolus (MOD) and stria vascularis (SV) using anti-UCP4. Anti-actin and anti-VDAC antibodies were used as control for the sample loading and as mitochondria marker respectively. OC, MOD and SV were isolated from P3 rats. For each lane 20 μg protein were loaded. C. Confocal laser scanning microscopy image of OC, SL and SG of P3 rats (left panel) in comparison with P28 rats (right panel). Double-labeling with anti-UCP4 antibody (visualized by Alexa-488, green) and DAPI was performed on cryosections (20 μmthick).In P3 rat cochlea (left panel) UCP4 is homogenously distributed in different structures: SG neurons, SL, supporting cells and hair cells of OC (white arrows). In P28 rat cochlea (right panel) UCP4 is restricted to SG and OC cells. The complete images (left and right panels) are a result of a montage of two microscopical ﬁelds, taken under the same conditions. Scale bar, 40 μm. 246 A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253 demonstrated, that under physiological conditions, UCP4 immunore- UCP4 detection in hair cells activity is detected in mitochondria of outer (OHCs) and inner hair cells (IHCs) and, surprisingly, in mitochondria of supporting cells Data from literature (Mao et al., 1999) and from our previous study (inner phalangeal, Deiters' and pillar cells) (Fig. 2B). (Smorodchenko et al., 2009) showed that UCP4 mRNA and protein

Fig. 2. UCP4 expression in supporting, outer and inner hair cells. A. Confocal laser scanning microscopy images of OC from freshly isolated rat cochlea (P28 rats). UCP4 in cytoplasm of OHCs and IHCs is stronger expressed than in inner phalangeal cell, PhC, (visualized by Alexa-488, green) and outer phalangeal cells (Deiters' cells, DC, arrowhead). Staining was performed on cryosections from freshly isolated cochlea (20 μm thick) and counterstained with DAPI (blue). PC, pilar cells. Scale bar, 50 μm. B. Merged confocal laser scanning microscopy image of OC (newborn, P3) shows the triple-labeling with anti-UCP4 (visualized by Alexa-488, green), anti-NF200 (visualized by Alexa-568, red) antibodies and DAPI (blue). Anti-UCP4 immunoreactivity is seen on the apical pole of all neurosensory hair cells and in the cytoplasm of supporting epithelial cells. Anti-NF200 staining shows contact of SG neuron processes with supporting cells. Arrows indicate basilar membranes. Staining was performed on cryosections (20 μmthick) from freshly isolated cochlea. Scale bar, 20 μm. C. Confocal laser scanning microscopy image of the neurosensory hair cells under normoxia conditions. Triple-labeling was performed on whole cochlea (middle turn) with anti-UCP4 (visualized by Alexa-488, green) antibodies, phalloidin (red) and DAPI (blue). UCP4 immunostaining is exclusively present in apical poles of hair cells cytoplasm (under the stereocilia) and Deiters' cells. The protein was not detected in stereocilia of hair cells (arrows). Scale bar, 20 μm. A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253 247 were exclusively detected in neuronal cells and to a lesser extent in present not only in SG neurons (NF-200 positive cell bodies) but also astrocytes. The presence of UCP4 in hair cells, shown in Fig. 2A, is of in other cells of the surrounding connective tissue such as fibrocytes high interest and was therefore examined in detail. Double immuno- and satellite cells. To a lesser extent, UCP4-immunoreactivity was also labeling of P3 rat cochleae with anti-UCP4 antibodies and phalloidin observed in neuronal processes (arrows). In MOD of P28 rats, UCP4 (a marker for actin filaments) revealed the high density of UCP4 expression was strongly restricted to SG neuron bodies (Fig. 3B, upper immunostaining in the apical pole of hair cell cytoplasm, directly panel) and glial cells (Fig. 3B, lower panel, arrow heads). Only few under the stereocilia and was not detected in stereocilia themselves mitochondria of neuronal processes contained UCP4 (Fig. 3B, upper (Fig. 2C, arrows). Confocal images of hair cells stained with the panel, arrows). selective marker for OHCs prestin, anti-UCP4 AB and DAPI at two different z-levels (Fig. S2A) confirmed the localization of UCP4 in UCP4 co-localization with glial cells OHCs. Apico-basal pattern of UCP4 distribution reflected the mitochondria distribution in the cytoplasm of hair cells (Saito, 1983), GFAP is known as a classical marker for astrocytes, non-myelinating visualized by immunostaining with the mitochondrial marker COX IV Schwann cells and satellite cells in a late postnatal period. However, it is (Fig. S2B). also used as a marker for immature neuronal cells in embryonic development and the early postnatal stage (Rio et al., 2002). UCP4 expression in spiral ganglion neurons In newborn rats (P3) the double-immunopositivity (anti-UCP4 and anti-GFAP) was detected in structures of OC (supporting cells), According to Kitahara et al.(2004) UCP4 mRNA is present in spiral SL (fibroblasts and interdental cells) and MOD (Schwann and ganglia (SG) of adult mice ear and varies between different mouse satellite cells) (Fig. S3B). Interestingly, in adult animals (P28) the strains. We double-immunostained cochlea cryosections with an anti- co-localization was only restricted to the glial cells in MOD and UCP4 antibody and a neuronal marker anti-NF200 to examine UCP4 supporting cells of OC (data not shown). WB confirmed the presence in rat SG. Fig. 3A (arrows) shows that in P3 animals UCP4 is immunohistochemistry findings (Fig. 5C).

Fig. 3. Expression of UCP4 in spiral ganglion neurons in P3 and P28 rats. A. 3D-reconstructed confocal laser scanning microscopy image of SG neurons from freshly isolated cochlea (modiolus, P3 rats) demonstrates expression of UCP4 in cell bodies of SG and adjacent glial cells. Double-labeling with anti-UCP4 (visualized by Alexa-488, green) and anti-NF200 (visualized by Alexa-568, red) antibodies was performed on whole cochlea. Scale bar, 10 μm. B. Confocal laser scanning microscopy image of modiolus from P28 rat. Double-labeling with anti-UCP4 antibody (visualized by Alexa-488, green) and anti-NF-200 (visualized by Alexa-568, red) antibodies was performed in cross cryostat sections (20 μm thick). Upper panel: expression of UCP4 in cell bodies of SG, single UCP4-positive mitochondria in neuronal processes (arrows). Scale bar, 20 μm. Lower panel: single UCP4-positive glial cells (arrowheads). Scale bar, 30 μm. 248 A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253

UCP4 co-localization with epithelial cells in MOD and had statistically higher levels in adult animals in comparison to P2 rats (3-fold increased). In contrast, the relative To confirm the origin of UCP4 positive cells, we performed double- level of UCP4 expression in OC reached the adult level at day P8 and its immunolabeling of whole-mount prepared cochlea (Fig. 4A) from P3 lowest value at early postnatal developmental stage (P2, Fig. 5B). The and P28 cochlea using anti-UCP4 and anti-CK18 antibodies. Cytoker- UCP4 protein amount was normalized to the amount of the atin filament proteins are present in cells of the epithelial origin mitochondrial marker VDAC, to relate the UCP4 concentration to the (Kuijpers et al., 1991). Yellow color in the merged image (Fig. 4A) mitochondrial number. The parallel increase of VDAC/actin ratio shows that all epithelial cells of SL were positive, both for UCP4 and implies that the mitochondrial amount in cochlea is increasing with CK18 in P3 rat cochlea. In contrast, no UCP4 immunoreactivity was age (Fig. 5B). detected in SL's epithelial cells from adult animals (P28, Fig. 1C), WB analysis of OC lysates revealed that GFAP expression is higher whereas high UCP4 levels were observed in hair and supporting cells in P2 rats (Fig. 5C) than in P5 and P8 animals. Only a negligible amount (Fig. 2A). of protein was detected at the P28 stage. Such immunoreactivity in adult rats can be explained by the presence of supporting cells, with a weak GFAP positivity (Rio et al., 2002). In contrast, the GFAP UCP4 co-localization with Merkel cells immunoreactivity in MOD increased with an increase in the rats' To evaluate the presence of UCP4 in other neurosensory receptor age, reaching the highest level in adult animals. The latter was fi cells, the footpads of adult rats (P28) were stained with the specific con rmed by immunohistochemistry analysis (data not shown). High marker of Merkel cells CK18 (Kim and Holbrook, 1995; Lucarz and amounts of another marker for immature neuronal cells, doublecortin Brand, 2007). Fig. 4B shows the significant amount of protein in (Dcx, von Bohlen und Halbach, 2007; Francis et al., 1999), was found in Merkel cells localized in the basal layer of the epidermis. MOD 204 and OC of P2 rats (Fig. 5D). Its expression is negligible at P8 stage and completely disappeared in adult rats.

UCP4 expression during inner ear development UCP4 expression after induction of normoxia and hypoxia The maturation of the inner ear is accompanied by functional and morphological changes as well as by expression changes of several The susceptibility of SG neurons to the oxidative stress caused by proteins. To quantitatively examine the UCP4 expression during hypoxia and ischemia is the main reason for hearing disorders (Orita postnatal inner ear development, we isolated cochlea from newborn et al., 2002; Mazurek et al., 2003). Because the members of UCP (P2), young (P5, P8) and adult (P28) rats. Western blot results subfamily are alleged to regulate reactive oxygen species (ROS) and to (Fig. 5A) and their quantiﬁcations (Fig. 5B) showed that relative UCP4 be regulated by ROS (Echtay et al., 2002) we were interested in testing expression (normalized to actin) gradually increased from P2 to P28 the UCP4 expression in hypoxia.

Fig. 4. A. UCP4 expression in epithelia of spiral limbus (SL). Confocal laser scanning microscopy image of freshly isolated rat cochlea (P3). Double-labeling with anti-UCP4 (visualized by Alexa-488, green) and anti-CK18 (visualized by Alexa-568, red) antibodies. Staining was performed on cross cryostat 20 μm thick sections. Scale bar, 50 μm. B. Confocal laser scanning microscopy image of the rat footpad (P28). Double-labeling with anti-UCP4 (visualized by Alexa-488, green) and anti-CK18 (visualized by Alexa-568, red) antibodies. Yellow color in the merged image indicates co-localization of UCP4 with CK18-positive Merkel cell. Staining was performed on cryosections (20 μmthick).Scale bar, 10 μm. A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253 249

Fig. 5. UCP4 expression in cochlea during development. A–D. Representative Western blots from freshly isolated OC and MOD from P2, P5, P8 and P28 day old rats. (A) Immunostaining with anti-UCP4 (UCP4), anti-VDAC (VDAC) and anti-actin antibodies; (B) Quantitative analysis of Western blots showing the relative distribution of UCP4 (normalized to β-actin and VDAC) and VDAC normalized to β-actin in freshly isolated OC and MOD. The deviation of each ratio from the mean ratio is given as a percentage, whereby the mean ratio of all values in the experiment was set to 100%. Quantitative results are based on 6 Western blots from 3 independent experiments and expressed as means+SEM. *Pb0.005; **Pb0.05. (C) Immunostaining with anti-GFAP and (D) with anti-doublecortin (Dcx). Lysates were extracted from 6 OC and 6 MOD per group (at least three independent experiments). Twenty microgram proteins were loaded per lane.

We used the model of mild hypoxia (5 h) which is characterized by although the changes in the immunoreactivity pattern of another a low cell death rate and was validated by biochemical (release of mitochondrial marker COX IV mirrored morphological alterations of lactate dehydrogenase) and morphological (cytotoxicity test using mitochondria (disintegration) under hypoxic conditions (Fig. 6D). propidium iodide) criteria in our previous work (Gross et al., 2007). Using the same model, we have shown previously that hypoxic We demonstrated that organotypic cultures of OC, MOD and SV conditions led to the down-regulation of several proapoptotic genes maintain viability under normoxic conditions. Hypoxia exposure and to the prominent up-regulation of the mitochondrial anti- induced moderate cell damage (see Fig. 1 and Figs. S1–S4 in Gross et oxidative enzyme SOD2 in MOD (Gross et al., 2007). In the present al. (2007)). Using this setup, we thus compared the UCP4 expression study we observed UCP4 downregulation instead of UCP4 over- in OC and MOD of P3 rats at different culture conditions: in freshly expression (similar to SOD2) under hypoxic conditions. prepared samples, during normoxia and hypoxia (see “Experimental methods”). UPC4 expression was investigated relative to the expression of two other proteins: β-actin as a cell marker and VDAC Discussion as a mitochondria marker. Additionally we measured VDAC expression relative to β-actin expression as a criterion for the possible In this study we evaluated the expression of UCP4 in the rat change in mitochondria number during hypoxia (Fig. 6B). cochlea at physiological conditions during development, and tested WB analysis of the protein expression in OC and MOD revealed that the hypothesis of the possible UCP4 up-regulation under oxidative the initial content of UCP4 in freshly isolated OC was higher than in MOD stress conditions. (Fig. 6A). No significant changes in UCP4 expression were observed in For the first time, we now show that not only neurons and OC at normoxic conditions in comparison to freshly isolated tissue. In astrocytes (Smorodchenko et al., 2009), but also neurosensory contrast, hypoxia for the duration of 5 h led to the decrease of UCP level receptor cells, hair cells, express a significant amount of UCP4. in both regions (Fig. 6A–B and Fig. S4). Because normoxia and hypoxia Support for the idea that the presence of UCP4 may be a common also led to the decrease of VDAC in MOD (Fig. 6B VDAC/actin), we feature of neurosensory cells comes from our present data showing believe that the decrease in UCP4 expression (Fig. 6B UCP4/actin) may the presence of UCP4 in Merkel cells (Fig. 4B). The latter are neuro- be partly due to mitochondria damage. Immunohistochemistry con- sensory receptors in the skin of vertebrates that provide touch in- firmed the highest UCP4 level in freshly prepared OC (Fig. 6C). Fig. 6D formation to the brain and originate from the neural crest (Szeder shows similar UCP4 expression at normoxia and hypoxia conditions, et al., 2003), similar to peripheral neurons and glia cells. 250 A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253

By analyzing the protein expression during development, we astrocytes previously described in CNS (Smorodchenko et al., 2009; Liu revealed that prior to the onset of hearing (P8–P10), UCP4 is broadly et al., 2006). Results showing the stable expression of UCP4 through the distributed in cochlea of the newborn animals. Taking into account the postnatal development in MOD, where spiral ganglion neurons, nerve embryogenesis of the inner ear, we considered that UCP4 positive cells fibers, and glial cells are localized, are in line with the findings of in rat cochlea in addition to spiral ganglion neurons can be neuroe- Kitahara et al.(2004), who detected UCP4 mRNA in SG of adult rodents. pithelial cells (Fritzsch et al., 2006). UCP4 immunoreactivity was not Supporting cells of OC are a potential source for the differentiation seen in these cells after their differentiation to SL epithelial cells (Fig. 1C, of new hair cells in the postnatal mammalian ear (Raphael et al., right panel). Double immunostaining with the marker for immature 2007). Moreover, it is assumed that both hair and supporting cells neuronal cells (GFAP) and anti-UCP4 antibody clearly shows the co- have a common epithelial precursor during regeneration (Jones and localization of both proteins in all investigated structures of rat cochlea Corwin, 1996; Fekete et al., 1998). During further differentiation, hair at P3 (Fig. S3B). GFAP immunoreactivity almost disappeared in OC of the cells lose their cytokeratin expression (Raphael et al., 1987). In adult adult inner ear (Fig. 5C), but was increased in MOD, where glial cells human cochlea, the cytokeratin marker CK18 is present in supporting, were also accumulated. This finding is in line with data reported by Rio epithelial cells, Reissner's membrane, but not in hair cells and et al.(2002) and confirms the presence of UCP4 in GFAP positive fibroblasts (Anniko et al., 1989; Bauwens et al., 1991). These facts

Fig. 6. UCP4 expression in organ of Corti after normoxia and hypoxia. A. Representative Western blots from freshly isolated OC and MOD, after induction of 24 h normoxia and 5 h hypoxia (P3 rats) immunostained with anti-UCP4 (UCP4), anti-VDAC (VDAC) and anti-actin antibodies. Twenty microgram proteins were loaded per lane. B. Quantiﬁcation of Western blots showing the relative distribution of UCP4 (normalized to β-actin and VDAC) and VDAC (normalized to β-actin) in freshly isolated OC and MOD, after induction of 24 h normoxia and 5 h hypoxia. Quantitative results are based on 4 Western blots from 4 independent experiments and expressed as means+SEM for n=6 cultured specimens. *Pb0.005. C. Confocal laser scanning microscopy image of hair cells of OC from freshly isolated rat cochlea and after induction of 24 h normoxia and 5 h hypoxia. Labeling with anti- UCP4 (visualized by Alexa-488, green) antibodies was performed on whole medial turn of cochlea. Images were taken from apical pole of hair cells. Scale bar, 20 μm. D. Confocal laser scanning microscopy image of epithelial cells in SL from P3 rat cochlea after induction of 24 h normoxia and 5 h hypoxia. Triple-labeling with anti-UCP4 (visualized by Alexa-488, green) and COX IV (visualized by Alexa-568, red) antibodies (marker of inner mitochondrial membrane) was performed on whole cochlea. The staining patterns for UCP4 and COX IV are overlapped (yellow) in cell bodies. Hypoxia leads to the mitochondria morphology alteration, visualized by COX IV staining. Culture specimens were counterstained with DAPI (blue). Scale bar, 10 μm. A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253 251

Fig. 6 (continued).

can explain the presence of UCP4 immunoreactivity in both cell types oxidative stress no UCP4 up-regulation occurred in the ear; (iv) UCP4 (hair and supporting cells) of newborn and adult rat cochlea. In may rather play a distinct role in functional maturation of rat inner cochlea of P28 rats the co-localization of UCP4 with CK18 expression ear, supporting our hypothesis about UCP4 participation in brain was only observed in supporting cells of OC. No UCP4 immunoreac- neuronal development (Smorodchenko et al., 2009). tivity was detected in epithelial cells and fibroblasts of SL at P28 (Fig. 1C). The involvement of UCP4 in oxidative stress regulation was studied Experimental methods in the model of oxygen deprivation (Gross et al., 2007). Hypoxia is known to be accompanied by reactive oxygen (ROS) and nitrogen Explant cultures of organ Corti and modiolus, hypoxia induction species (RNS) formation and is characterized by a set of metabolic changes named as the “oxidative stress response” (Blokhina and Explant cultures of organ of Corti (OC), modiolus (MOD) and stria Fagerstedt, 2010). UCP4 was proposed to regulate free radicals vascularis (SV) were prepared as previously described (Gross et al., production, based mainly on its putative ability to transport protons, 2010; Gross et al., 2007). In brief, the membranous cochleae were similar to UCP1 or UCP2 (Klingenberg et al., 2001; Beck et al., 2006; isolated from 2-, 5-, 8- and 28-day-old Wistar rats (animal facility of Urbankova et al., 2003; Rupprecht et al., 2010; Jezek et al., 1989). the Free University Berlin, Germany) and dissected into OC, MOD and However, no direct evidence for such UCP4 transport behavior was SV (Sobkowicz et al., 1993) in buffered saline glucose solution (BSG; obtained so far. Ectopic expression of UCP4 in mammalian cells led to 116 mM NaCl, 27.2 mM Na2HPO4, 6.1 mM KH2PO4, 11.4 mM glucose). the reduction of mitochondrial membrane potential, ΔΨm, which is The MOD fragments contained the main part of spiral ganglion (SG) consistent with an uncoupling function (Mao et al., 1999). On the neurons; the OC were dissected together with spiral limbus (SL) and other side, phylogenetic analysis shows that UCP4 with approxi- are indicated as OC in the text below. For tissue culture, cochlea mately 30% homology to UCP1 is more closely related to oxogluta- fragments were plated in 4-well tissue culture dishes in 500 μl rate and dicarboxylate carriers than to uncoupling proteins (Borecky Dulbecco's modified Eagle medium/F12 nutrient (1:1) mixtures et al., 2001). Using the model of oxygen deprivation, we now (DMEM/F12, Gibco, Karlsruhe, Germany) supplemented with 10% demonstrated that hypoxia did not lead to the increase of the fetal bovine serum (FBS, Biochrom AG, Berlin, Germany), 50 mM protein expression level, neither in MOD nor in OC (Fig. 6B). These glucose, insulin–transferrin–Na–Selenit-Mix 2 μl/ml (Roche Diagnos- results are in disagreement with conclusions of Wu et al. which tics GmbH, Mannheim, Germany), penicillin 100 U/ml (Grünenthal demonstrated the up-regulation of UCP4 protein levels by superox- GmbH, Aachen, Germany) and maintained at 37 °C and 5% CO2 in a ide anion (Wu et al., 2009). The exposure of neuroblastoma cells to humidified tissue culture incubator. After 3 h the specimens were 1 mM neurotoxin MPP+ (methyl-4-phenylpyridinium ion) for 72 h divided into experimental groups: freshly isolated, normoxia and showed only modestly increased steady-state UCP4 mRNA levels hypoxia. Freshly isolated specimens were fixed in 4% PFA. Specimens (Ho et al., 2005). of “normoxia” group were first incubated for up to 24 h at the same It should be mentioned that the absence of UCP4 up-regulation in conditions and then fixed in 4% paraformaldehyde (PFA) in 0.1 M our experiments does not necessarily mean the absence of the UCP4 phosphate buffer (PB) at pH 7.4. The exposure to hypoxia was protective role under hypoxic conditions. It was shown for other UCP performed for 5 h at 37 °C in a gas mixture of 5% CO2, 95% N2 as proteins (UCP1-UCP3) that mitochondria-derived free radicals and described previously (Gross et al., 2007). Thereafter, cultures were lipid peroxidation products such as 4-hydroxy-2-nonenal can regu- kept under normoxic conditions for another 16 h. Freshly isolated late their activity (Echtay et al., 2003). cochlea fragments and cochlea specimens after 24 h normoxia and 5 h In summary, (i) UCP4 expression in cochlea is region-specific and hypoxia were frozen in liquid nitrogen and stored for further Western is developmentally regulated; (ii) for the first time, UCP4 expression blot analyses at −80 °C. Tissues were collected and fixed in 4% PFA for in neurosensory cells (hair cells, Merkel cells) was shown; (iii) under immunohistochemistry. 252 A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253

Studies were performed in accordance with the German Preven- Acknowledgments tion of Cruelty to Animals Act and approved by Berlin Senate Ofﬁce for Health. This work was supported by Deutsche Forschungsgemeinschaft (Po-524/3, Po-524/5). Immunohistochemistry

References After fixation in 4% PFA, whole cochlea and cochlea segments (apical, medial and basal) from P3 and P28 rats were incubated in Anniko, M., Thornell, L.E., Ramaekers, F.C., Stigbrand, T., 1989. Cytokeratin diversity in phosphate-buffered sucrose solution (1.4 M) overnight at 4 °C for epithelia of the human inner ear. Acta Otolaryngol 108, 385–396. cryoprotection. The cochleae were then embedded in Tissue freezing Bauwens, L.J., DeGroot, J.C., Ramaekers, F.C., Veldman, J.E., Huizing, E.H., 1991. Cytokeratin expression in the epithelia of the adult human cochlea. Eur Arch medium (Leica Microsystems GmbH, Germany). For investigation of Otorhinolaryngol 248, 293–297. immunofluorescence the tissues were cryosectioned (20 μm thick) at Beck, V., Jaburek, M., Breen, E.P., Porter, R.K., Jezek, P., Pohl, E.E., 2006. A new automated −26 °C (Jung Frigocut Cryostat Microtome, Leica, Germany), mounted technique for the reconstitution of hydrophobic proteins into planar bilayer ™ fi membranes. Studies of human recombinant uncoupling protein 1. Biochim Biophys on Superfrost microscope slides (Erie Scienti c, Portsmouth NH) Acta 1757, 474–479. and stored at −20 °C until staining. Blokhina, O., Fagerstedt, K.V., 2010. Oxidative metabolism, ROS and NO under oxygen Cochlea cross sections and whole samples of OC were permeabi- deprivation. Plant Physiol Biochem 48, 359–373. fi Borecky, J., Maia, I.G., Arruda, P., 2001. Mitochondrial uncoupling proteins in mammals lized, then blocked for non-speci c binding as previously described and plants. Biosci Rep 21, 201–212. (Smorodchenko et al., 2009) and double-labeled with appropriate Chan, S.L., Liu, D., Kyriazis, G.A., Bagsiyao, P., Ouyang, X., Mattson, M.P., 2006. primary antibodies (AB) (see Table S1, Supplementary materials) at Mitochondrial uncoupling protein-4 regulates calcium homeostasis and sensitivity to store depletion-induced apoptosis in neural cells. J Biol Chem 281, 37391–37403. 4 °C overnight. Negative control sections were incubated in a similar Echtay, K.S., Esteves, T.C., Pakay, J.L., Jekabsons, M.B., Lambert, A.J., Portero-Otin, M., Pamplona, manner, except that the primary AB was replaced by normal goat R., Vidal-Puig, A.J., Wang, S., Roebuck, S.J., Brand, M.D., 2003. A signalling role for 4- serum (Vector Laboratories, UK). For visualization of primary ABs, hydroxy-2-nonenal in regulation of mitochondrial uncoupling. EMBO J 22, 4103–4110. Echtay, K.S., Roussel, D., St Pierre, J., Jekabsons, M.B., Cadenas, S., Stuart, J.A., Harper, J.A., secondary ABs (Alexa-488, Alexa-568, Alexa-633-conjugated; see Roebuck,S.J.,Morrison,A.,Pickering,S.,Clapham,J.C.,Brand,M.D.,2002. Table S1, Supplementary materials) in blocking solution were applied Superoxide activates mitochondrial uncoupling proteins. Nature 415, 96–99. for 90 minutes at RT. The sections and whole samples were covered Fekete, D.M., Muthukumar, S., Karagogeos, D., 1998. Hair cells and supporting cells – with DAPI mounting medium for nuclei counterstaining. share a common progenitor in the avian inner ear. J Neurosci 18, 7811 7821. Francis, F., Koulakoff, A., Boucher, D., Chafey, P., Schaar, B., Vinet, M.C., Friocourt, G., McDonnell, N., Reiner, O., Kahn, A., McConnell, S.K., Berwald-Netter, Y., Denoulet, P., Confocal microscopy Chelly, J., 1999. Doublecortin is a developmentally regulated, microtubule-associated protein expressed in migrating and differentiating neurons. Neuron 23, 247–256. Fritzsch, B., Beisel, K.W., Hansen, L.A., 2006. The molecular basis of neurosensory cell Confocal images were acquired in sequential scanning mode with formation in ear development: a blueprint for hair cell and sensory neuron an upright laser microscope (Leica DM 2500) equipped with a 63× regeneration? Bioessays 28, 1181–1193. oil-immersion objective. For the excitation of Alexa-488, Alexa-568, Gross, J., Machulik, A., Amarjargal, N., Moller, R., Ungethum, U., Kuban, R.J., Fuchs, F.U., Andreeva, N., Fuchs, J., Henke, W., Pohl, E.E., Szczepek, A.J., Haupt, H., Mazurek, B., Alexa-633 and DAPI an argon-ion laser (488 nm), helium-neon laser 2007. Expression of apoptosis-related genes in the organ of Corti, modiolus and (543 and 633 nm) and blue diode laser (405 nm) were used stria vascularis of newborn rats. Brain Res 1162, 56–68. respectively. During the processing stage individual image channels Gross, J., Stute, K., Moller, R., Fuchs, J., Amarjargal, N., Pohl, E.E., Angerstein, M., Smorodchenko, A., Mazurek, B., 2010. Expression of prestin and Gata-3,-2,-1 mRNA were pseudocolored with RGB colors corresponding (if not otherwise in the rat organ of Corti during the postnatal period and in culture. Hear Res 261, 9–21. indicated) to the fluorophore emission spectrum. Image J software Hanak, P., Jezek, P., 2001. Mitochondrial uncoupling proteins and phylogenesis–UCP4 as (http://rsb.info.nih.gov/ij/) was used for the image merging, back- the ancestral uncoupling protein. FEBS Lett 495, 137–141. Ho, P.W., Chan, D.Y., Kwok, K.H., Chu, A.C., Ho, J.W., Kung, M.H., Ramsden, D.B., Ho, S.L., ground correction and adjustment of brightness and contrast. 3D 2005. Methyl-4-phenylpyridinium ion modulates expression of mitochondrial reconstruction was performed with Volocity 4 software (Improvision uncoupling proteins 2, 4, and 5 in catecholaminergic (SK-N-SH) cells. J Neurosci Res Ltd., Coventry, England). 81, 261–268. Jezek, P., Krasinskaya, I.P., Smirnova, I., Drahota, Z., 1989. Carnitine cycle in brown adipose tissue mitochondria as a tool for studying the regulatory role of fatty acids Western blot analysis (WB) in the uncoupling protein function. FEBS Lett 243, 37–40. Jones, J.E., Corwin, J.T., 1996. Regeneration of sensory cells after laser ablation in the lateral line system: hair cell lineage and macrophage behavior revealed by time- Protein samples for Western blots (WB) were obtained from OC, – – lapse video microscopy. J Neurosci 16, 649 662. MOD and SV. Tissues were homogenized in RIPA buffer (Tris HCl, Kim, D.K., Holbrook, K.A., 1995. The appearance, density, and distribution of Merkel 50 mM; NaCl, 250 mM; SDS 0.1%, Aprotinin, 1 mg/l; NP40, 1%; Na- cells in human embryonic and fetal skin: their relation to sweat gland and hair deoxycholate, 0.5%). SDS-PAGE and WB were performed as described follicle development. J Invest Dermatol 104, 411–416. Kitahara, T., Horii, A., Kizawa, K., Maekawa, C., Kubo, T., 2007. Changes in mitochondrial previously (Smorodchenko et al., 2009). Membranes were incubated uncoupling protein expression in the rat vestibular nerve after labyrinthectomy. with an affinity purified AB against mUCP4 (1:1000). ABs coupled to Neurosci Res 59, 237–242. horseradish peroxidase (anti-rabbit IgG, GE Healthcare, UK) were Kitahara, T., Li, H.S., Balaban, C.D., 2004. Localization of the mitochondrial uncoupling protein family in the rat inner ear. Hear Res 196, 39–48. used as secondary antibodies. The chemiluminescence reaction of the Kitahara, T., Li-Korotky, H.S., Balaban, C.D., 2005. Regulation of mitochondrial ECL-plus system (Amersham, Freiburg, Germany) was used for uncoupling proteins in mouse inner ear ganglion cells in response to systemic visualization of AB-binding. The membranes were then stripped kanamycin challenge. Neuroscience 135, 639–653. β Klingenberg, M., Winkler, E., Echtay, K., 2001. Uncoupling protein, H+ transport and and sequentially incubated with ABs against VDAC and -actin regulation. Biochem Soc Trans 29, 806–811. (see Table S1) as a control for uniform loading. Immunoblots were Kuijpers, W., Peters, T.A., Tonnaer, E.L., Ramaekers, F.C., 1991. Expression of cytokeratin quantified by densitometry using Image J. polypeptides during development of the rat inner ear. Histochemistry 96, 511–521. Liu, D., Chan, S.L., Souza-Pinto, N.C., Slevin, J.R., Wersto, R.P., Zhan, M., Mustafa, K., de Cabo, R., Mattson, M.P., 2006. Mitochondrial UCP4 mediates an adaptive shift in Statistic analysis energy metabolism and increases the resistance of neurons to metabolic and oxidative stress. Neuromolecular Med 8, 389–414. Data were presented as mean±standard deviation (SD) from at Lucarz, A., Brand, G., 2007. Current considerations about Merkel cells. Eur J Cell Biol 86, 243–251. least three independent experiments with six specimens per each Mao, W., Yu, X.X., Zhong, A., Li, W., Brush, J., Sherwood, S.W., Adams, S.H., Pan, G., 1999. condition. Data were considered statistically significant at pb0.005 by UCP4, a novel brain-specific mitochondrial protein that reduces membrane using the Student's t test. potential in mammalian cells. FEBS Lett 443, 326–330. Mattson, M.P., Liu, D., 2003. Mitochondrial potassium channels and uncoupling proteins Supplementary materials related to this article can be found online in synaptic plasticity and neuronal cell death. Biochem Biophys Res Commun 304, at doi: 10.1016/j.mcn.2011.03.002. 539–549. A. Smorodchenko et al. / Molecular and Cellular Neuroscience 47 (2011) 244–253 253

Mazurek, B., Winter, E., Fuchs, J., Haupt, H., Gross, J., 2003. Susceptibility of the hair cells Smorodchenko, A., Rupprecht, A., Sarilova, I., Ninnemann, O., Brauer, A.U., Franke, K., of the newborn rat cochlea to hypoxia and ischemia. Hear Res 182, 2–8. Schumacher, S., Techritz, S., Nitsch, R., Schuelke, M., Pohl, E.E., 2009. Comparative Orita, Y., Sando, I., Miura, M., Haginomori, S., Hirsch, B.E., 2002. Cochleosaccular analysis of uncoupling protein 4 distribution in various tissues under physiological pathology after perinatal and postnatal asphyxia: histopathologic findings. Otol conditions and during development. Biochim Biophys Acta 1788, 2309–2319. Neurotol 23, 34–38. Sobkowicz, H.M., Loftus, J.M., Slapnick, S.M., 1993. Tissue culture of the organ of Corti. Raphael, Y., Kim, Y.H., Osumi, Y., Izumikawa, M., 2007. Non-sensory cells in the Acta Otolaryngol Suppl 502, 3–36. deafened organ of Corti: approaches for repair. Int J Dev Biol 51, 649–654. Szeder, V., Grim, M., Halata, Z., Sieber-Blum, M., 2003. Neural crest origin of mammalian Raphael, Y., Marshak, G., Barash, A., Geiger, B., 1987. Modulation of intermediate- Merkel cells. Dev Biol 253, 258–263. filament expression in developing cochlear epithelium. Differentiation 35, 151–162. Urbankova, E., Voltchenko, A., Pohl, P., Jezek, P., Pohl, E.E., 2003. Transport kinetics of Rio, C., Dikkes, P., Liberman, M.C., Corfas, G., 2002. Glial fibrillary acidic protein uncoupling proteins: analysis of UCP1 reconstituted in planar lipid bilayers. J Biol expression and promoter activity in the inner ear of developing and adult mice. Chem 278, 32497–32500. J Comp Neurol 442, 156–162. von Bohlen und Halbach, H.O., 2007. Immunohistological markers for staging Rupprecht, A., Sokolenko, E.A., Beck, V., Ninnemann, O., Jaburek, M., Trimbuch, T., neurogenesis in adult hippocampus. Cell Tissue Res 329, 409–420. Klishin, S.S., Jezek, P., Skulachev, V.P., Pohl, E.E., 2010. Role of the transmembrane Wu, Z., Zhang, J., Zhao, B., 2009. Superoxide anion regulates the mitochondrial free potential in the membrane proton leak. Biophys J 98, 1503–1511. Ca2+ through uncoupling proteins. Antioxid Redox Signal 11, 1805–1818. Saito, K., 1983. Fine structure of the sensory epithelium of guinea-pig organ of Corti: Zhang, M., Wang, B., Ni, Y.H., Liu, F., Fei, L., Pan, X.Q., Guo, M., Chen, R.H., Guo, X.R., 2006. subsurface cisternae and lamellar bodies in the outer hair cells. Cell Tissue Res 229, Overexpression of uncoupling protein 4 promotes proliferation and inhibits 467–481. apoptosis and differentiation of preadipocytes. Life Sci 79, 1428–1435.