- Email: [email protected]
Otoprotective effect of recombinant erythropoietin in a model of newborn hypoxic-ischemic encephalopathy
International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
Contents lists available at SciVerse ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Otoprotective effect of recombinant erythropoietin in a model of newborn hypoxic-ischemic encephalopathy Yu¨ksel Olgun a,*, Gu¨nay Kırkım a, Efsun Kolatan b, Mu¨ge Kıray c, Alper Bag˘rıyanık d, Bu¨lent S¸erbetc¸iog˘lu a, Osman Yılmaz b, Necati Go¨kmen e, Hu¨lya Ellidokuz f, Abdullah Kumral g, Semih Su¨tay a a
Dokuz Eylu¨l University, School of Medicine, Department of Otorhinolarngology, Turkey Dokuz Eylu¨l University, School of Medicine, Department of Laboratory of Animal Science, Turkey Dokuz Eylu¨l University, School of Medicine, Department of Physiology, Turkey d Dokuz Eylul University, School of Medicine, Department of Histology, Turkey e Dokuz Eylu¨l University, School of Medicine, Department of Anesthesiology and Reanimation, Turkey f Dokuz Eylu¨l University, School of Medicine, Department of Biostatistics, Turkey g Dokuz Eylu¨l University, School of Medicine, Department of Pediatrics, Turkey b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 November 2012 Received in revised form 19 January 2013 Accepted 26 January 2013 Available online 20 February 2013
Objective: The aim of this study is to test the hypotheses that central auditory pathology as well as inner ear pathology is contributing mechanisms to observed hypoxic-ischemic encephalopathy (HIE) induced hearing loss and that recombinant erythropoietin (rhEPO) will reduce this cellular pathology and attenuate hearing loss. Methods: Twenty-eight 7-day Wistar albino rat pups were divided into four groups: Control group (n = 8) was given only intraperitoneal saline solution. Sham group (n = 5) had only a midline neck incisions without carotid ligation under general anesthesia and administration of intraperitoneal saline solution. HIE group (n = 8) and rhEPO treated group (n = 7) were subjected to left common carotid artery ligation followed by 2.5 h hypoxia exposure to a mixture of 8% oxygen and 92% pure nitrogen. HIE group was injected with intraperitoneal saline solution, while the rhEPO treated group received rhEPO 100 U/ kg within the same volume as the saline-alone solution. At the end of the seventh week of age hearing (ABRs) was evaluated in response to clicks, 6 kHz and 8 kHz tone burst stimuli. Animals were sacrificed and both temporal lobes, cochleas and brainstems of the animals were collected. Tissue samples were evaluated with light microscopy, immunohistochemical studies, including TUNEL and caspase-3 stainings, and electron microscopy. Results: Hearing thresholds were elevated in HIE animals. In rhEPO treated animals, ABR values were similar to controls. HIE caused apoptotic changes in brainstem structures as shown by light microscopy and immunohistochemical methods. Apoptotic changes also were found within the organ of Corti, spiral ganglion cells and neurons of temporal lobe by electron microscopic investigation. In rhEPO animals many of these apoptotic changes were observed, but reduced compared to untreated animals. Conclusions: Mechanisms underlying HIE-induced hearing loss are based on apoptosis in inner ear; however central auditory pathway pathology occurs as well, likely contributing to changes in auditory processing and perception of complex signals not reflected by the ABR threshold shifts. For both clinical and basic significance ‘rhRPO’ is found to reduce those effects. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Hypoxic-ischemic encephalopathy Hearing loss Erythropoietin
1. Introduction Central nervous system damage due to hypoxic-ischemic encephalopathy (HIE) is one of the leading causes of mortality
* Corresponding author at: Dokuz Eylu¨l University, School of Medicine, Department of Otorhinolaryngology, Mithatpasa Caddesi, No. 1606, I˙nciraltı, I˙zmir, Turkey. Tel.: +90 5067021434; fax: +90 2322789495. E-mail addresses: [email protected], [email protected] (Y. Olgun). 0165-5876/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijporl.2013.01.029
and morbidity in newborns. Nearly 80–85% of babies recover but varying degrees of neurological sequela are observed among those babies [1–3]. HIE is also known to deteriorate hearing in affected individuals [4]. Studies in children recovered from HIE showed that ABR thresholds are elevated [4–6]. The mechanism of hearing loss is not well studied. In a limited number of postmortem studies, some pathological changes were found in cochlea, brainstem and auditory cortex; however all of these changes were observed in severe cases of HIE resulting in death [1,6,7]. Thus the results are not easily generalizable to cases that recovered with hearing
740
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
impairment. In organotopic cochlea cell cultures HI conditions led to hair cell damage [8,9] but there are no in vivo studies investigating the effects of HIE on the hearing of neonates. Thus the aim of this study is to investigate the effect of HI on hearing and auditory pathways. In addition, recently recombinant erythropoietin (rhEPO); an endogenous glycoprotein was found to reduce brain damage in newborn rats [10–13]; and receptors of EPO were shown in the organ of Corti of guinea pigs, and rats [14,15]. Recombinant EPO was found to prevent apoptosis in organotopic cochlea cultures of newborn rats in HI conditions [9]. Moreover rhEPO was found to attenuate acoustic trauma [16–18]. Thus a second aim of this study is to assess the efficacy of rhEPO to prevent HIE-induced hearing loss and associated pathology. The purpose of this study is to test the hypotheses that central auditory pathology as well as inner ear pathology are correlated with HIE-induced hearing loss and that rhEPO will reduce this cellular pathology and attenuate hearing loss. 2. Methods This study was performed in accordance with guidelines established by the Experimental Animal Laboratory and approved by the Animal Care and Use Ethical Committee of the University (Protocol Number: 27/2010). Twenty-eight Wistar albino rat pups of either sex weighting 9–11 g were used in experiments. Seven-day old pups were chosen because the neurodevelopmental stage of these animals corresponds to that of newborn infants [19]. Animals were housed in controlled environmental conditions (20 2 8C, 55% relative humidity and 12 h light/dark cycle) and had free access to pulverized standard rat pellet food and tap water. Animals in all groups were similar in weight, and weight gain was similar through the study period. Animals with signs of external ear or middle ear disorders were eliminated from the study. 2.1. Experimental design Rat pups were divided into 4 groups. In group 1 (control group n = 8), only 0.2 ml saline was given intraperitoneally. In group 2 (sham group n = 5); a midline vertical neck incision was performed and sutured, without carotid artery ligation, under halothane anesthesia; then 0.2 ml saline was given intraperitoneally. In group 3 (HIE group n = 8) HIE was induced, including midline incision and artery ligation, under halothane anesthesia; and 0.2 ml saline was administered intraperitoneally. In group 4 (HIE + rhEPO group n = 7) HIE was induced under halothane ¨ /kg rhEPO in 0.2 ml saline was given anesthesia and 1000 U intraperitoneally. 2.2. Induction of HIE A modification of Levine preparation was used as model of perinatal hypoxic ischemic brain injury [20]. In groups 3 and 4 the left common carotid arteries were ligated permanently. Under anesthesia a midline 0.5–1 cm vertical neck incision was performed and the left common carotid artery was identified and ligated with 6/0 silk under the microscope. Total time of surgery never exceeded 6 min. Animals were excluded from the study if there was bleeding during ligation or respiratory arrest. Following a 2 h period of recovery and breast feeding, animals were exposed to a 2.5 h period of hypoxia (92% N2, 8% O2) in a hypoxia chamber, an airtight container partially submerged in a 37 8C water bath to maintain a constant thermal environment. If respiration of any animal arrested before completing 2.5 h of hypoxia it was excluded from study.
2.3. Intraperitoneal EPO application Recombinant EPO was given in 0.2 ml saline solution. After retrieval from hypoxia chambers group 4 received an intraperitoneal injection of rhEPO at a dose of 1000 U/kg whereas HIE, sham and control groups received intraperitoneal 0.2 ml saline solution only. After those procedures all pups were returned to their dams until the end of their seventh week of age. 2.4. Auditory assessment Auditory brainstem response (ABR) testing was performed at the end of the seventh week. Only animals demonstrating a normal otoscopic examination were kept on study. They were anesthetized with ketamine hydrochloride (60 mg/kg) and xylazine hydrochloride (5 mg/kg). ABRs were recorded using ICS Medical Charter equipment via the insert earphones, in a silent room. The insert earphones were placed directly into rat’s external auditory canals. A neonatal probe tip was used to seal the external auditory canal. Subdermal needle electrodes were placed over the vertex (active), the right and left retro auricular regions (reference) and on the dorsum (ground). The stimuli used were alternating clicks (pulse duration 0.1 ms) at a rate of 21.1/s, and tone bursts of 6 kHz and 8 kHz (1 ms plateau, 2 ms rise/fall times) at a rate of 31.1/s. EEG activity was preamplified (100,000) and band-pass analogfiltered from 100 to 3000 Hz for click stimuli and 50–1500 Hz for tone bursts. The number of averaged responses was 1024 and each averaged response was replicated. Each series of stimuli began from supra-threshold level with subsequent measures using lower intensity levels. Stimulus level was decreased progressively and intensities that appeared to be near threshold were replicated. Threshold was defined as the lowest intensity to elicit repeatable components of ABR in at least two trials. 2.5. Sacrifice and tissue sampling All rats were sacrificed under ether anesthesia after auditory assessments. Tissues, including temporal lobes of the cerebrum, cochlear nuclei of the brainstems and both temporal bones, were sampled. 2.6. Histopathological procedures The brain, brainstem and inner ear tissue samples were prepared for light and electron microscopic assessment. For light microscopic assessment brainstem samples were fixed by immersion in 10% formalin in phosphate buffer (pH 7.4) overnight, then dehydrated through a series of increasing concentrations of ethanol (60%, 70%, 80%, 90%, absolute ethanol) and xylene, and then embedded in paraffin blocks. Paraffin blocks were placed in Leica RM2255 rotary microtome (Germany) and sections of 5 mm thickness were obtained. In brainstem samples, the cochlear nucleus was identified according to rat brain atlas [21]. Sections were deparaffinized, hydrated and stained with cresyl violet and TUNEL (terminal deoxynucleotidyl transferase-mediated dUTPbiotin nick end labeling), and caspase-3 immunohistochemistry was performed. The images were analyzed using a computerassisted image analyzer system consisting of a microscope (Olympus BX-51, Japan) equipped with a high-resolution video camera (Olympus DP71, Japan). For ultrastructural investigations of organ of Corti, temporal bones were carefully harvested, soft tissues were removed and lateral walls of the cochlea were opened. The cochleas were then placed in 2.5% glutaraldehyde for 24 h for fixation. Temporal bones decalcified in EDTA, sucrose, 2.5% glutaraldehyde with phosphate saline, pH 7.0 for 2 months, to verify decalcification of the cochlea.
[(Fig._1)TD$IG]
[(Fig._2)TD$IG]
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
741
Fig. 1. The left ear hearing levels. The left ear mean and SEM ABR values in dB SPL for each of the experimental groups (1–4) to click and tone burst (6 kHz and 8 kHz) stimuli. Hearing loss was greater in the left, carotid-ligated side than the right.
Fig. 2. The right ear mean and SEM ABR values in dB for each of the experimental groups (1–4) to click and tone burst (6 kHz and 8 kHz) stimuli.
Temporal lobe tissue samples were fixed in gluteraldehyde with phosphate saline. All the tissues were postfixed with osmium tetroxide (OsO4), dehydrated in a graded series of alcohols, and then embedded in Araldite1 CY212. The thin (60–90 nm) sections were obtained with an ultra-microtome (Leica) and stained with uranyl acetate and lead citrate, examined by transmission electron microscopy (Carl Zeiss Libra 120 EFTEM), and digitally photographed. Terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL) method: detection of DNA fragmentation in situ was visualized with the use of the ApopTag Plus Peroxidase In Situ Apoptosis Detection Kit (Chemicon International, USA). Deparaffinized tissue sections were incubated with proteinase K (20 mg/ ml). Tissue sections were subjected to 3% H2O2 for endogenous peroxidase inhibition and were incubated with 1 equilibration buffer at room temperature for 30 min. The digoxigenin-labeled dNTP tail was incubated with Tdt (terminal deoxynucleotidyl transferase) for 1 h at 37 8C, and sections were washed in stop/ wash buffer for 10 min at room temperature. Tissue sections were incubated with anti-digoxigenin-peroxidase antibody at room temperature for 30 min and were stained with diaminobenzidine (DAB) as a peroxidase substrate. Staining was evaluated using a light microscope after counterstaining with haemotoxylin. Five [(Fig._3)TD$IG] fields were randomly chosen for each slide and the TUNEL-positive
cells per field were determined. The apoptotic index (percentage of apoptotic nuclei) was calculated as apoptotic nuclei/total nuclei counted 100%. All counting procedures were performed blindly. Caspase-3 immunohistochemistry: For visualization of the caspase-3 expression, caspase-3 immunohistochemistry was performed using an active anti-caspase-3 antibody. The sections were incubated overnight with active anti-caspase-3 antibody (1:100; Millipore AB3623) and then for another 30 min with the biotinylated mouse secondary antibody. The bound secondary antibody was then amplified with Histostain1 plus bulk kit (Invitrogen, 85–9043). The antibody–biotin–avidin–peroxidase complexes were visualized using DAB. The sections were finally mounted onto lysine-coated slides. The percentage of caspase 3positive cells was determined by counting the positive cells on 5 random fields in each group. 2.7. Statistical analysis All statistical analyses were done using SPSS 15.0 software program. All results are expressed means SEM. The mean values of groups were compared by Kruskal–Wallis variance analysis with post hoc Bonferroni test. For the comparison of right and left ear ABR values Wilcoxon test was used (P < 0.05 was considered statistically significant).
Fig. 3. Representative sections from the cochlear nucleus of control (A), sham (B), HIE-alone (C), and HIE-rhEPO treated (D) groups, showing cresyl violet, TUNEL and caspase 3 staining. In figure IC and ID arrows show pyknotic cells, in figure IIC and IID arrows show TUNEL+ cells, in figure IIIC and IIID arrows show caspase 3-positive cells.
742
[(Fig._4)TD$IG]
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
3. Results 3.1. Auditory findings Hearing loss was demonstrated bilaterally in HIE group. Within each of the four groups the difference between the left and the right ear click, 6 kHz, 8 kHz ABR values was not statistically significant (p > 0.05). For the left ears, click, 6 kHz, 8 kHz ABR values of HIE group were significantly higher than values of other groups (p < 0.05). For the right ears, ABR values in HIE group were higher than sham, control and the rhEPO treated groups; however this difference was significant only for click ABR values (p < 0.05). The mean left and right ABR threshold values are given in Figs. 1 and 2. 3.2. Histological examination 3.2.1. Cresyl violet staining Control and sham groups with cresyl violet stained sections taken through the cochlear nucleus showed normal morphology (Fig. 3, IA). Across the groups and animals within each group samples were from the same area. In the HIE group, most of the neurons were shrunken and darkly staining-pyknotic (Fig. 3, IB). However in rhEPO treated group 4 majority of the neurons were normal and there were very few pyknotic neurons (Fig. 3, IC). 3.2.2. TUNEL assay The present study shows that HIE induced apoptotic cell death in the cochlear nucleus. Control and sham group rats showed fewer
[(Fig._5)TD$IG]
Fig. 4. Mean and SEM of apoptotic cell ratios (number of stained nuclei/total nuclei counted 100%) with TUNEL and caspase-3 stainings for each of the experimental groups 1–4.
TUNEL-positive cells (Fig. 3, IIA) than both HIE and HIE-rhEPO groups. There were more TUNEL positive cells in HIE group, as seen in Fig. 3, IIB. Quantification and statistical analysis of the TUNEL staining showed that the number of TUNEL positive cells increased significantly in HIE group compared with control and sham groups. (p < 0.05) (Fig. 4). Moreover, while there was an elevation in TUNEL positive cells in the HIE-rhEPO treated group compared to sham and control subjects, the number of TUNEL positive cells were significantly decreased in rhEPO treated group (p < 0.05) compared to the HIE group alone (Fig. 4).
Fig. 5. Organ of Corti. In A (control), B (sham), D (HIE-rhEPO) normal hair cell morphology (HC), normal nucleus borders (N), ciliae (black arrows), regular cell membrane (blue arrows) and mitochondria (red arrows) were seen. In C (HIE-alone) loss of ciliae (green arrow) and electron dense cytoplasm (*) were observed. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
3.2.3. Caspase-3 immunohistochemistry Apoptosis was further confirmed by caspase-3 immunohistochemistry. Immunohistochemical evaluation based on the intensity of caspase-3 immunoreactivity in the rat brainstem is shown in Fig. 3, III A, B and C. The number of caspase -3-positive cells in HIE group was significantly increased compared to control and sham groups (p < 0.05) (Fig. 4). The number of TUNEL positive cells was significantly decreased in rhEPO treated group (p < 0.05) compared to the HIE-only group (Fig. 4). 3.2.4. Ultrastructural findings Ultrastructural findings of organ of Corti, spiral ganglion and temporal lobe were similar for both the left and the right sides in each group. Ultrastructural findings of organ of corti: In control and sham groups, the ultrastructure of hair cells and supporting cells was normal. The boundaries of the cell membrane and nucleus were regular. Mitochondria showed normal properties (Fig. 5A and B). In HIE group, hair cells were shrunken and became electron-dense where ultrastructural findings became difficult to distinguish. Cilia loss on apical surface of the cells was seen clearly. In these cells [(Fig._6)TD$IG]intracellular degenerative areas were also observed. Apical
743
intercellular junctions were damaged (Fig. 5C). In rhEPO treated group the cellular organization of hair cells and supporting cells was normal and similar to that of the control and sham groups (Fig. 5D). Ultrastructural findings of spiral ganglion: In control and sham groups, the ultrastructure of ganglion cells and satellite cells was normal. The boundaries and shapes of the cells were regular (Fig. 6A and B). In HIE group (Fig. 6C), changes in cell shapes and irregularity at the cell membranes were seen. There were cytoplasmic degenerative areas and loss of satellite cells. In rhEPO treated group ganglion cell structures were protected and all findings were similar to control and sham groups (Fig. 6D) Ultrastructural examination of temporal lobe: In control and sham group, neurons with a regular cell membrane, normal ultrastructure and euchromatic nuclei were seen. Neuropil and perivascular area appeared normal (Fig. 7A and B). In the HIE group, the neurons were damaged, the nuclei of the cells were heterochromatic and nuclei borders were irregular. Neuropil has degenerative areas and perivascular edema was prominent (Fig. 7C). In rhEPO treated group, neurons and perivascular structures were well protected and appeared similar to tissues of the control and sham groups (Fig. 7D).
Fig. 6. The ultrastructural morphology of Spiral ganglion cells (GC). A and B: The ganglion cells in the control and sham groups showed normal morphology. C: The ganglion cells in the HIE group generally showed aponecrotic and apoptotic character. Because of shrinkage of the perikarya cell membranes have deep interdigitations (red arrow), and there was chromatin clusters in nuclei (N). D: The ganglion cells in the EPO group have euchromatic nuclei (N), the cell membranes showed minimal irregularities (red arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
[(Fig._7)TD$IG]
744
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
Fig. 7. Temporal lobe. In A (control), B (sham), D (rhEPO), the cells have euchromatic nucleus (N) and nuclei borders (black arrows) were regular and nucleolus was prominent. Mitochondria (red arrows) and perivascular area (*) was normal. In C (HIE) the neurons were damaged, diminished in size, became electron-dense and displayed wrinkled nucleus. The nuclei of the cells were heterochromatic and nuclei borders were irregular. Perivascular edema was prominent. BV, blood vessel. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
4. Discussion Prelingual hearing loss if untreated inevitably leads to deterioration of speech and language development. It also causes low educational performance and loss of future job opportunities. Perinatal asphyxia is known to cause various neurological development delays; and hearing loss has been reported in 17.1% of infants following perinatal asphyxia [4]. Hearing loss can also exist in 6.3–12.2% of babies recovered from HI insult without any other neurological sequalae [4,22]. On the other hand in postnatal period exposure to HI does not cause a permanent hearing loss which implies a sensitive period in the development of human hearing organ. Recent ultrastructural and micromolecular studies indicate that sensorineural hearing loss is a result of hypo-oxygenization at the cellular level, with appropriate biochemical sequela. Noise exposure, vascular occlusion, aging and ototoxic substances can trigger overexpression of reactive oxygen species (ROS), alteration of calcium metabolism, up regulation of cell death gene pathways and release of apoptotic proteins which lead to apoptotic cell death.
In perinatal HIE the exact site(s) and mechanisms of hearing loss have not been fully understood. In postmortem studies cochleosaccular atrophy, loss of outer hair cells (OHC) and edema in the stria vascularis are prominent findings. It was found that caspase-3 staining was positive in both inner and outer hair cells, spiral ganglion cells and stria vascularis. Those apoptotic changes preceded changes in brain tissue [1]. However nearly all postmortem studies were conducted on newborns who died after relatively severe HI insults and little evidence is available on the pathological changes in babies recovered from HIE. Effects of HI were also investigated in organotopic cochlea cultures of newborn rats. In one study [8] hair cell loss was not found in hypoxic-only conditions, but if hypoxia is induced together with ischemia, the loss of hair cells became evident. Although those in vitro studies point out the cochlea of newborn rat is susceptible to HI there is no in vivo study to our knowledge which investigated the effects of HI on the peripheral and central auditory system in neonates. Mechanisms of neuronal injury in HIE have been investigated in different models. One of the widely accepted models of HIE is Levine’s preparation described by Rice et al. [20]. In this model,
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
7-day old rat pups are exposed to hypoxia after unilateral ligation of common carotid artery. Developmental timing of the insult represents the newborn period of the human [23,24]. In this model neuroprotective effects of erythropoietin have been studied [13,19]. We have adopted this HIE model for the investigation of HI on hearing. In the first part of the current study, ABRs revealed bilateral hearing loss in all HIE-alone treated animals. Hearing loss was more prominent on the ligated site. Across animals, within the HIEonly treated group, there were some differences in the extent of hearing loss and associated pathology. This finding is consistent with clinical reports in which various degrees of hearing loss are seen after HIE. Cochleas, brainstems and temporal lobes of the pups were histopathologically evaluated. Apoptotic changes in brainstem at the level of cochlear nuclei were shown by cresyl violet staining, TUNEL assay and caspase-3 stainings in HIE group. Our observation indicated that the brain stem tissue damage mainly reflected apoptotic cell death. Apoptotic changes within hair cells, spiral ganglion cells and temporal lobes were also detected by using ultrastructural assessments. A prominent finding was cellular degeneration with the loss of ciliae of hair cells in the cochlea. Ultrastructural damage in spiral ganglion cells and temporal lobes was also clearly detected in the HIE-alone group. Those findings are in accordance with previous postmortem studies as well as in vitro experiments [1,6,7]. Studies on prevention of apoptosis in the central nervous system induced by HIE increased after 2000. Specifically, detection of EPO receptors first in guinea pig [14] and later in rat brains [15] led to use of rhEPO in experimental studies. As EPO is an endogenous protein and has clinical application for a number of different indications, including use in newborns, it has gained attention. It has now been demonstrated that hypoxic conditions activated EPO receptors and increased endogenous EPO production in organotypic cochlea cell cultures [25]. Those findings indicate that rhEPO given exogenously can easily bind and activate EPO receptors. Recombinant EPO seems to be advantageous in protective strategies. The potential efficacy of EPO as an otoprotective agent is supported by the observation that EPO prevents gentamycin ototoxicity in rat cochlea cell lines [16]. rhEPO also prevented apoptosis and necrosis due to hypoxia in newborn rat cochlea cell culture [9]. For those reasons in the second part of the study, rhEPO was chosen for investigation of possible otoprotective properties. ABR results demonstrated that hearing loss was not induced in HIE animals with rhEPO treatment. Those electrophysiological findings were also supported by histopathological investigations. TUNEL assay and caspase-3 staining procedures showed that apoptosis was prevented by the application of rhEPO. Similar results were found on the electronmicroscopic studies of the rhEPO treated group; mitochondria and ciliae of the hair cells were in normal configuration, similar to the control and sham groups. Spiral ganglion cells were also intact, as well as temporal lobe structures. These findings are consistent with results in in vitro studies. 5. Conclusion Our findings confirm the hypotheses that HIE-induced hearing loss in the newborn is based on upregulation of cell death pathways leading to apoptosis and that these changes are observed in both the peripheral and central auditory system from cochlear nucleus to the temporal cortex and that rhEPO can significantly attenuate these changes and protect hearing. In a newborn rat HIE induced hearing loss bilaterally. HIE causes
745
apoptosis in hair cells and spiral ganglion cells, which may well explain the induced changes in hearing sensitivity shown in the ABR results. However as well apoptotic cell death was clearly evident in the brainstem cochlear nucleus and temporal lobe tissues in the newborn. These changes indicate that changes in central processing of auditory signals and the perception of complex signals such as speech and signals in noise may be compromised by HIE. Our finding indicates that rhEPO is effective in the protection of hearing as shown by ABR records. However of equal clinical importance rhEPO also reduced central auditory pathology. Thus, apoptotic changes due to HIE within the cochlea, brainstem and temporal lobe are all significantly attenuated by rhEPO. Our results suggest that rhEPO may have a protective effect against HIE induced hearing loss, both in peripheral sensitivity as well as complex sound processing and perception in the central nervous system. These studies provide the scientific rational to support the needed clinical trials to translate these observations to eventual prevention of hearing loss in human infants. References [1] J. Schumutzhard, R. Glueckert, C. Sergi, I. Schwentner, I. Abraham, A. SchrottIscher, Does perinatal asphyxia induce apoptosis in the inner ear? Hear. Res. 250 (2009) 1–9. [2] R.C. Vannucci, S.J. Vannucci, A model of perinatal hypoxic-ischemic brain damage, Ann NY Cad. Sci. 835 (1997) 243–249. [3] M. Sano, K. Kaga, E. Kitazumi, K. Kodama, Sensorineural hearing loss in patients with cerebral palsy after asphyxia and hyperbilirubinemia, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 1211–1217. [4] Z.D. Jiang, Long-term of perinatal and postnatal asphyxia on developing human auditory brainstem responses: peripheral hearing loss, Int. J. Pediatr. Otorhinolaryngol. 33 (1995) 225–238. [5] Z.D. Jiang, D.M. Brosi, A. Wilkinson, Depressed brainstem auditory electrophysiology in preterm infants after perinatal hypoxia-ischemia, J. Neurol. Sci. 281 (2009) 28–33. [6] Y. Orita, I. Sando, M. Miura, S.I. Haginomori, B.E. Hirsch, Cochleosaccular pathology after perinatal and postnatal asphyxia: histopathologic findings, Otol. Neurotol. 23 (2002) 34–38. [7] S. Koyama, K. Kaga, H. Sakata, Y. Ino, K. Kodere, Pathological findings in the temporal bone of newborn infants with neonatal asphyxia, Acta Otolarngol. 125 (2005) 1028–1032. [8] B. Mazurek, E. Winter, J. Fuchs, H. Haupt, J. Gross, Susceptibility of the hair cells of the newborn rat cochlea to hypoxia and ischemia, Hear. Res. 182 (2003) 2–8. [9] N. Andreeva, A. Nyamaa, H. Haupt, J. Gross, B. Mazurek, Recombinant human erythropoietin prevents ischemia-induced apoptosis and necrosis in explant cultures of the rat organ of Corti, Neurosci. Lett. 396 (2006) 86–90. [10] C.T. Noguchi, P. Asavaritikrai, R. Teng, Y. Jia, Role of erythropoietin in the brain, Crit. Rev. Oncog. 64 (2007) 159–171. [11] R.J. Mc Pherson, S.E. Juul, Recent trends in erythropoietin-mediated neuroprotection, Int. J. Dev. Neurosci. 26 (2008) 103–111. [12] M. Yamada, C. Burke, P. Colditz, D.W. Johnson, G.C. Gobe, Erythropoietin protects against apoptosis and increases expression of non-neuronal cell markers in the hypoxia-injured developing brain, J. Pathol. 224 (2011) 101–109. [13] A. Kumral, H. Baskın, D.C. Yesilirmak, B.U. Ergur, S. Aykan, S. Genc, Erythropoietin attenuates lipopolysaccharide-induced white matter injury in the neonatal rat brain, Neonatology 92 (2007) 269–278. [14] P. Caye´-Thomasen, N. Wagner, B.L. Frederiksen, K. Asal, J. Thomsen, Erythropoietin and erythropoietin receptor expression in the guinea pig inner ear, Hear. Res. 203 (2005) 21–27. [15] A.M. Naldi, M. Gassman, D. Bodmer, Erythropoietin but not VEGF has a protective effect on auditory hair cells in the inner ear, Cell. Mol. Life Sci. 66 (2009) 3593–3599. [16] A. Monge, I. Nagy, S. Bonaby, S. Schmid, M. Gassman, D. Bodmer, The effect of erythropoietin on gentamicin-induced auditory hair cell loss, Laryngoscope 116 (2006) 312–316. [17] B.L. Frederiksen, P. Caye´-Thomasen, S.P. Lund, N. Wagner, K. Asal, N.V. Olsen, et al., Does erythropoietin augment noise induced hearing loss? Hear. Res. 223 (2007) 129–137. [18] K. Tabuchi, K. Oikawa, I. Uemaetomari, S. Tsuji, T. Wada, A. Hara, A glucocorticoids and dehydroeiandrosterone sulfate ameloriate ischemia-induced injury of the cochlea, Hear. Res. 180 (2003) 51–56. [19] A. Kumral, N. Uysal, K. Tugyan, A. Sonmez, O. Yilmaz, N. Gokmen, Erythropoietin improves long-term spatial memory deficits and brain injury following neonatal hypoxia-ischemia in rats, Behav. Brain Res. 153 (2004) 77–86. [20] J.E. Rice, R.C. Vannucci, T.B. Bierly, The influence of immaturity on hypoxicischemic brain damage in the rat, Ann. Neurol. 9 (1981) 131–141. [21] G. Paxionos, C. Watson, Atlas: The Rat Brain in Sterotaxic Coordinates, fourth ed., Academic Press, 1998.
746
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
[22] Z.D. Jiang, X.Y. Liu, B.P. Shi, L. Lin, C.F. Bu, A.R. Wilkinson, Brainstem auditory outcomes and correlation with neurodevelopment after perinatal asphyxia, Pediatr. Neurol. 39 (2008) 189–195. [23] H.S. Romijn, M.A. Hofman, A. Gramsbergen, At what age is the developing cerebral cortex of the rat comparable to that of the full-term human baby? Early Hum. Dev. 26 (1991) 61–67.
[24] J. Tomfighi, D. Mauger, R.C. Vannucci, S.J. Vannucci, Influence of age on the cerebral lesions in an immature rat model of cerebral hypoxic-ischemia: a light microscopic study, Dev. Brain Res. 100 (1997) 149–160. [25] J. Gross, R. Moller, N. Amarjagal, A. Machulik, Fuchs J., U. Ungethu¨m, Expression of erythropoietin and angiogenic growth factors following inner ear injury of newborn rats, Prague Med. Rep. 110 (2009) 310–331.
Contents lists available at SciVerse ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Otoprotective effect of recombinant erythropoietin in a model of newborn hypoxic-ischemic encephalopathy Yu¨ksel Olgun a,*, Gu¨nay Kırkım a, Efsun Kolatan b, Mu¨ge Kıray c, Alper Bag˘rıyanık d, Bu¨lent S¸erbetc¸iog˘lu a, Osman Yılmaz b, Necati Go¨kmen e, Hu¨lya Ellidokuz f, Abdullah Kumral g, Semih Su¨tay a a
Dokuz Eylu¨l University, School of Medicine, Department of Otorhinolarngology, Turkey Dokuz Eylu¨l University, School of Medicine, Department of Laboratory of Animal Science, Turkey Dokuz Eylu¨l University, School of Medicine, Department of Physiology, Turkey d Dokuz Eylul University, School of Medicine, Department of Histology, Turkey e Dokuz Eylu¨l University, School of Medicine, Department of Anesthesiology and Reanimation, Turkey f Dokuz Eylu¨l University, School of Medicine, Department of Biostatistics, Turkey g Dokuz Eylu¨l University, School of Medicine, Department of Pediatrics, Turkey b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 November 2012 Received in revised form 19 January 2013 Accepted 26 January 2013 Available online 20 February 2013
Objective: The aim of this study is to test the hypotheses that central auditory pathology as well as inner ear pathology is contributing mechanisms to observed hypoxic-ischemic encephalopathy (HIE) induced hearing loss and that recombinant erythropoietin (rhEPO) will reduce this cellular pathology and attenuate hearing loss. Methods: Twenty-eight 7-day Wistar albino rat pups were divided into four groups: Control group (n = 8) was given only intraperitoneal saline solution. Sham group (n = 5) had only a midline neck incisions without carotid ligation under general anesthesia and administration of intraperitoneal saline solution. HIE group (n = 8) and rhEPO treated group (n = 7) were subjected to left common carotid artery ligation followed by 2.5 h hypoxia exposure to a mixture of 8% oxygen and 92% pure nitrogen. HIE group was injected with intraperitoneal saline solution, while the rhEPO treated group received rhEPO 100 U/ kg within the same volume as the saline-alone solution. At the end of the seventh week of age hearing (ABRs) was evaluated in response to clicks, 6 kHz and 8 kHz tone burst stimuli. Animals were sacrificed and both temporal lobes, cochleas and brainstems of the animals were collected. Tissue samples were evaluated with light microscopy, immunohistochemical studies, including TUNEL and caspase-3 stainings, and electron microscopy. Results: Hearing thresholds were elevated in HIE animals. In rhEPO treated animals, ABR values were similar to controls. HIE caused apoptotic changes in brainstem structures as shown by light microscopy and immunohistochemical methods. Apoptotic changes also were found within the organ of Corti, spiral ganglion cells and neurons of temporal lobe by electron microscopic investigation. In rhEPO animals many of these apoptotic changes were observed, but reduced compared to untreated animals. Conclusions: Mechanisms underlying HIE-induced hearing loss are based on apoptosis in inner ear; however central auditory pathway pathology occurs as well, likely contributing to changes in auditory processing and perception of complex signals not reflected by the ABR threshold shifts. For both clinical and basic significance ‘rhRPO’ is found to reduce those effects. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Hypoxic-ischemic encephalopathy Hearing loss Erythropoietin
1. Introduction Central nervous system damage due to hypoxic-ischemic encephalopathy (HIE) is one of the leading causes of mortality
* Corresponding author at: Dokuz Eylu¨l University, School of Medicine, Department of Otorhinolaryngology, Mithatpasa Caddesi, No. 1606, I˙nciraltı, I˙zmir, Turkey. Tel.: +90 5067021434; fax: +90 2322789495. E-mail addresses: [email protected], [email protected] (Y. Olgun). 0165-5876/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijporl.2013.01.029
and morbidity in newborns. Nearly 80–85% of babies recover but varying degrees of neurological sequela are observed among those babies [1–3]. HIE is also known to deteriorate hearing in affected individuals [4]. Studies in children recovered from HIE showed that ABR thresholds are elevated [4–6]. The mechanism of hearing loss is not well studied. In a limited number of postmortem studies, some pathological changes were found in cochlea, brainstem and auditory cortex; however all of these changes were observed in severe cases of HIE resulting in death [1,6,7]. Thus the results are not easily generalizable to cases that recovered with hearing
740
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
impairment. In organotopic cochlea cell cultures HI conditions led to hair cell damage [8,9] but there are no in vivo studies investigating the effects of HIE on the hearing of neonates. Thus the aim of this study is to investigate the effect of HI on hearing and auditory pathways. In addition, recently recombinant erythropoietin (rhEPO); an endogenous glycoprotein was found to reduce brain damage in newborn rats [10–13]; and receptors of EPO were shown in the organ of Corti of guinea pigs, and rats [14,15]. Recombinant EPO was found to prevent apoptosis in organotopic cochlea cultures of newborn rats in HI conditions [9]. Moreover rhEPO was found to attenuate acoustic trauma [16–18]. Thus a second aim of this study is to assess the efficacy of rhEPO to prevent HIE-induced hearing loss and associated pathology. The purpose of this study is to test the hypotheses that central auditory pathology as well as inner ear pathology are correlated with HIE-induced hearing loss and that rhEPO will reduce this cellular pathology and attenuate hearing loss. 2. Methods This study was performed in accordance with guidelines established by the Experimental Animal Laboratory and approved by the Animal Care and Use Ethical Committee of the University (Protocol Number: 27/2010). Twenty-eight Wistar albino rat pups of either sex weighting 9–11 g were used in experiments. Seven-day old pups were chosen because the neurodevelopmental stage of these animals corresponds to that of newborn infants [19]. Animals were housed in controlled environmental conditions (20 2 8C, 55% relative humidity and 12 h light/dark cycle) and had free access to pulverized standard rat pellet food and tap water. Animals in all groups were similar in weight, and weight gain was similar through the study period. Animals with signs of external ear or middle ear disorders were eliminated from the study. 2.1. Experimental design Rat pups were divided into 4 groups. In group 1 (control group n = 8), only 0.2 ml saline was given intraperitoneally. In group 2 (sham group n = 5); a midline vertical neck incision was performed and sutured, without carotid artery ligation, under halothane anesthesia; then 0.2 ml saline was given intraperitoneally. In group 3 (HIE group n = 8) HIE was induced, including midline incision and artery ligation, under halothane anesthesia; and 0.2 ml saline was administered intraperitoneally. In group 4 (HIE + rhEPO group n = 7) HIE was induced under halothane ¨ /kg rhEPO in 0.2 ml saline was given anesthesia and 1000 U intraperitoneally. 2.2. Induction of HIE A modification of Levine preparation was used as model of perinatal hypoxic ischemic brain injury [20]. In groups 3 and 4 the left common carotid arteries were ligated permanently. Under anesthesia a midline 0.5–1 cm vertical neck incision was performed and the left common carotid artery was identified and ligated with 6/0 silk under the microscope. Total time of surgery never exceeded 6 min. Animals were excluded from the study if there was bleeding during ligation or respiratory arrest. Following a 2 h period of recovery and breast feeding, animals were exposed to a 2.5 h period of hypoxia (92% N2, 8% O2) in a hypoxia chamber, an airtight container partially submerged in a 37 8C water bath to maintain a constant thermal environment. If respiration of any animal arrested before completing 2.5 h of hypoxia it was excluded from study.
2.3. Intraperitoneal EPO application Recombinant EPO was given in 0.2 ml saline solution. After retrieval from hypoxia chambers group 4 received an intraperitoneal injection of rhEPO at a dose of 1000 U/kg whereas HIE, sham and control groups received intraperitoneal 0.2 ml saline solution only. After those procedures all pups were returned to their dams until the end of their seventh week of age. 2.4. Auditory assessment Auditory brainstem response (ABR) testing was performed at the end of the seventh week. Only animals demonstrating a normal otoscopic examination were kept on study. They were anesthetized with ketamine hydrochloride (60 mg/kg) and xylazine hydrochloride (5 mg/kg). ABRs were recorded using ICS Medical Charter equipment via the insert earphones, in a silent room. The insert earphones were placed directly into rat’s external auditory canals. A neonatal probe tip was used to seal the external auditory canal. Subdermal needle electrodes were placed over the vertex (active), the right and left retro auricular regions (reference) and on the dorsum (ground). The stimuli used were alternating clicks (pulse duration 0.1 ms) at a rate of 21.1/s, and tone bursts of 6 kHz and 8 kHz (1 ms plateau, 2 ms rise/fall times) at a rate of 31.1/s. EEG activity was preamplified (100,000) and band-pass analogfiltered from 100 to 3000 Hz for click stimuli and 50–1500 Hz for tone bursts. The number of averaged responses was 1024 and each averaged response was replicated. Each series of stimuli began from supra-threshold level with subsequent measures using lower intensity levels. Stimulus level was decreased progressively and intensities that appeared to be near threshold were replicated. Threshold was defined as the lowest intensity to elicit repeatable components of ABR in at least two trials. 2.5. Sacrifice and tissue sampling All rats were sacrificed under ether anesthesia after auditory assessments. Tissues, including temporal lobes of the cerebrum, cochlear nuclei of the brainstems and both temporal bones, were sampled. 2.6. Histopathological procedures The brain, brainstem and inner ear tissue samples were prepared for light and electron microscopic assessment. For light microscopic assessment brainstem samples were fixed by immersion in 10% formalin in phosphate buffer (pH 7.4) overnight, then dehydrated through a series of increasing concentrations of ethanol (60%, 70%, 80%, 90%, absolute ethanol) and xylene, and then embedded in paraffin blocks. Paraffin blocks were placed in Leica RM2255 rotary microtome (Germany) and sections of 5 mm thickness were obtained. In brainstem samples, the cochlear nucleus was identified according to rat brain atlas [21]. Sections were deparaffinized, hydrated and stained with cresyl violet and TUNEL (terminal deoxynucleotidyl transferase-mediated dUTPbiotin nick end labeling), and caspase-3 immunohistochemistry was performed. The images were analyzed using a computerassisted image analyzer system consisting of a microscope (Olympus BX-51, Japan) equipped with a high-resolution video camera (Olympus DP71, Japan). For ultrastructural investigations of organ of Corti, temporal bones were carefully harvested, soft tissues were removed and lateral walls of the cochlea were opened. The cochleas were then placed in 2.5% glutaraldehyde for 24 h for fixation. Temporal bones decalcified in EDTA, sucrose, 2.5% glutaraldehyde with phosphate saline, pH 7.0 for 2 months, to verify decalcification of the cochlea.
[(Fig._1)TD$IG]
[(Fig._2)TD$IG]
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
741
Fig. 1. The left ear hearing levels. The left ear mean and SEM ABR values in dB SPL for each of the experimental groups (1–4) to click and tone burst (6 kHz and 8 kHz) stimuli. Hearing loss was greater in the left, carotid-ligated side than the right.
Fig. 2. The right ear mean and SEM ABR values in dB for each of the experimental groups (1–4) to click and tone burst (6 kHz and 8 kHz) stimuli.
Temporal lobe tissue samples were fixed in gluteraldehyde with phosphate saline. All the tissues were postfixed with osmium tetroxide (OsO4), dehydrated in a graded series of alcohols, and then embedded in Araldite1 CY212. The thin (60–90 nm) sections were obtained with an ultra-microtome (Leica) and stained with uranyl acetate and lead citrate, examined by transmission electron microscopy (Carl Zeiss Libra 120 EFTEM), and digitally photographed. Terminal deoxynucleotidyl transferase mediated dUTP nick end labeling (TUNEL) method: detection of DNA fragmentation in situ was visualized with the use of the ApopTag Plus Peroxidase In Situ Apoptosis Detection Kit (Chemicon International, USA). Deparaffinized tissue sections were incubated with proteinase K (20 mg/ ml). Tissue sections were subjected to 3% H2O2 for endogenous peroxidase inhibition and were incubated with 1 equilibration buffer at room temperature for 30 min. The digoxigenin-labeled dNTP tail was incubated with Tdt (terminal deoxynucleotidyl transferase) for 1 h at 37 8C, and sections were washed in stop/ wash buffer for 10 min at room temperature. Tissue sections were incubated with anti-digoxigenin-peroxidase antibody at room temperature for 30 min and were stained with diaminobenzidine (DAB) as a peroxidase substrate. Staining was evaluated using a light microscope after counterstaining with haemotoxylin. Five [(Fig._3)TD$IG] fields were randomly chosen for each slide and the TUNEL-positive
cells per field were determined. The apoptotic index (percentage of apoptotic nuclei) was calculated as apoptotic nuclei/total nuclei counted 100%. All counting procedures were performed blindly. Caspase-3 immunohistochemistry: For visualization of the caspase-3 expression, caspase-3 immunohistochemistry was performed using an active anti-caspase-3 antibody. The sections were incubated overnight with active anti-caspase-3 antibody (1:100; Millipore AB3623) and then for another 30 min with the biotinylated mouse secondary antibody. The bound secondary antibody was then amplified with Histostain1 plus bulk kit (Invitrogen, 85–9043). The antibody–biotin–avidin–peroxidase complexes were visualized using DAB. The sections were finally mounted onto lysine-coated slides. The percentage of caspase 3positive cells was determined by counting the positive cells on 5 random fields in each group. 2.7. Statistical analysis All statistical analyses were done using SPSS 15.0 software program. All results are expressed means SEM. The mean values of groups were compared by Kruskal–Wallis variance analysis with post hoc Bonferroni test. For the comparison of right and left ear ABR values Wilcoxon test was used (P < 0.05 was considered statistically significant).
Fig. 3. Representative sections from the cochlear nucleus of control (A), sham (B), HIE-alone (C), and HIE-rhEPO treated (D) groups, showing cresyl violet, TUNEL and caspase 3 staining. In figure IC and ID arrows show pyknotic cells, in figure IIC and IID arrows show TUNEL+ cells, in figure IIIC and IIID arrows show caspase 3-positive cells.
742
[(Fig._4)TD$IG]
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
3. Results 3.1. Auditory findings Hearing loss was demonstrated bilaterally in HIE group. Within each of the four groups the difference between the left and the right ear click, 6 kHz, 8 kHz ABR values was not statistically significant (p > 0.05). For the left ears, click, 6 kHz, 8 kHz ABR values of HIE group were significantly higher than values of other groups (p < 0.05). For the right ears, ABR values in HIE group were higher than sham, control and the rhEPO treated groups; however this difference was significant only for click ABR values (p < 0.05). The mean left and right ABR threshold values are given in Figs. 1 and 2. 3.2. Histological examination 3.2.1. Cresyl violet staining Control and sham groups with cresyl violet stained sections taken through the cochlear nucleus showed normal morphology (Fig. 3, IA). Across the groups and animals within each group samples were from the same area. In the HIE group, most of the neurons were shrunken and darkly staining-pyknotic (Fig. 3, IB). However in rhEPO treated group 4 majority of the neurons were normal and there were very few pyknotic neurons (Fig. 3, IC). 3.2.2. TUNEL assay The present study shows that HIE induced apoptotic cell death in the cochlear nucleus. Control and sham group rats showed fewer
[(Fig._5)TD$IG]
Fig. 4. Mean and SEM of apoptotic cell ratios (number of stained nuclei/total nuclei counted 100%) with TUNEL and caspase-3 stainings for each of the experimental groups 1–4.
TUNEL-positive cells (Fig. 3, IIA) than both HIE and HIE-rhEPO groups. There were more TUNEL positive cells in HIE group, as seen in Fig. 3, IIB. Quantification and statistical analysis of the TUNEL staining showed that the number of TUNEL positive cells increased significantly in HIE group compared with control and sham groups. (p < 0.05) (Fig. 4). Moreover, while there was an elevation in TUNEL positive cells in the HIE-rhEPO treated group compared to sham and control subjects, the number of TUNEL positive cells were significantly decreased in rhEPO treated group (p < 0.05) compared to the HIE group alone (Fig. 4).
Fig. 5. Organ of Corti. In A (control), B (sham), D (HIE-rhEPO) normal hair cell morphology (HC), normal nucleus borders (N), ciliae (black arrows), regular cell membrane (blue arrows) and mitochondria (red arrows) were seen. In C (HIE-alone) loss of ciliae (green arrow) and electron dense cytoplasm (*) were observed. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
3.2.3. Caspase-3 immunohistochemistry Apoptosis was further confirmed by caspase-3 immunohistochemistry. Immunohistochemical evaluation based on the intensity of caspase-3 immunoreactivity in the rat brainstem is shown in Fig. 3, III A, B and C. The number of caspase -3-positive cells in HIE group was significantly increased compared to control and sham groups (p < 0.05) (Fig. 4). The number of TUNEL positive cells was significantly decreased in rhEPO treated group (p < 0.05) compared to the HIE-only group (Fig. 4). 3.2.4. Ultrastructural findings Ultrastructural findings of organ of Corti, spiral ganglion and temporal lobe were similar for both the left and the right sides in each group. Ultrastructural findings of organ of corti: In control and sham groups, the ultrastructure of hair cells and supporting cells was normal. The boundaries of the cell membrane and nucleus were regular. Mitochondria showed normal properties (Fig. 5A and B). In HIE group, hair cells were shrunken and became electron-dense where ultrastructural findings became difficult to distinguish. Cilia loss on apical surface of the cells was seen clearly. In these cells [(Fig._6)TD$IG]intracellular degenerative areas were also observed. Apical
743
intercellular junctions were damaged (Fig. 5C). In rhEPO treated group the cellular organization of hair cells and supporting cells was normal and similar to that of the control and sham groups (Fig. 5D). Ultrastructural findings of spiral ganglion: In control and sham groups, the ultrastructure of ganglion cells and satellite cells was normal. The boundaries and shapes of the cells were regular (Fig. 6A and B). In HIE group (Fig. 6C), changes in cell shapes and irregularity at the cell membranes were seen. There were cytoplasmic degenerative areas and loss of satellite cells. In rhEPO treated group ganglion cell structures were protected and all findings were similar to control and sham groups (Fig. 6D) Ultrastructural examination of temporal lobe: In control and sham group, neurons with a regular cell membrane, normal ultrastructure and euchromatic nuclei were seen. Neuropil and perivascular area appeared normal (Fig. 7A and B). In the HIE group, the neurons were damaged, the nuclei of the cells were heterochromatic and nuclei borders were irregular. Neuropil has degenerative areas and perivascular edema was prominent (Fig. 7C). In rhEPO treated group, neurons and perivascular structures were well protected and appeared similar to tissues of the control and sham groups (Fig. 7D).
Fig. 6. The ultrastructural morphology of Spiral ganglion cells (GC). A and B: The ganglion cells in the control and sham groups showed normal morphology. C: The ganglion cells in the HIE group generally showed aponecrotic and apoptotic character. Because of shrinkage of the perikarya cell membranes have deep interdigitations (red arrow), and there was chromatin clusters in nuclei (N). D: The ganglion cells in the EPO group have euchromatic nuclei (N), the cell membranes showed minimal irregularities (red arrow). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
[(Fig._7)TD$IG]
744
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
Fig. 7. Temporal lobe. In A (control), B (sham), D (rhEPO), the cells have euchromatic nucleus (N) and nuclei borders (black arrows) were regular and nucleolus was prominent. Mitochondria (red arrows) and perivascular area (*) was normal. In C (HIE) the neurons were damaged, diminished in size, became electron-dense and displayed wrinkled nucleus. The nuclei of the cells were heterochromatic and nuclei borders were irregular. Perivascular edema was prominent. BV, blood vessel. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
4. Discussion Prelingual hearing loss if untreated inevitably leads to deterioration of speech and language development. It also causes low educational performance and loss of future job opportunities. Perinatal asphyxia is known to cause various neurological development delays; and hearing loss has been reported in 17.1% of infants following perinatal asphyxia [4]. Hearing loss can also exist in 6.3–12.2% of babies recovered from HI insult without any other neurological sequalae [4,22]. On the other hand in postnatal period exposure to HI does not cause a permanent hearing loss which implies a sensitive period in the development of human hearing organ. Recent ultrastructural and micromolecular studies indicate that sensorineural hearing loss is a result of hypo-oxygenization at the cellular level, with appropriate biochemical sequela. Noise exposure, vascular occlusion, aging and ototoxic substances can trigger overexpression of reactive oxygen species (ROS), alteration of calcium metabolism, up regulation of cell death gene pathways and release of apoptotic proteins which lead to apoptotic cell death.
In perinatal HIE the exact site(s) and mechanisms of hearing loss have not been fully understood. In postmortem studies cochleosaccular atrophy, loss of outer hair cells (OHC) and edema in the stria vascularis are prominent findings. It was found that caspase-3 staining was positive in both inner and outer hair cells, spiral ganglion cells and stria vascularis. Those apoptotic changes preceded changes in brain tissue [1]. However nearly all postmortem studies were conducted on newborns who died after relatively severe HI insults and little evidence is available on the pathological changes in babies recovered from HIE. Effects of HI were also investigated in organotopic cochlea cultures of newborn rats. In one study [8] hair cell loss was not found in hypoxic-only conditions, but if hypoxia is induced together with ischemia, the loss of hair cells became evident. Although those in vitro studies point out the cochlea of newborn rat is susceptible to HI there is no in vivo study to our knowledge which investigated the effects of HI on the peripheral and central auditory system in neonates. Mechanisms of neuronal injury in HIE have been investigated in different models. One of the widely accepted models of HIE is Levine’s preparation described by Rice et al. [20]. In this model,
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
7-day old rat pups are exposed to hypoxia after unilateral ligation of common carotid artery. Developmental timing of the insult represents the newborn period of the human [23,24]. In this model neuroprotective effects of erythropoietin have been studied [13,19]. We have adopted this HIE model for the investigation of HI on hearing. In the first part of the current study, ABRs revealed bilateral hearing loss in all HIE-alone treated animals. Hearing loss was more prominent on the ligated site. Across animals, within the HIEonly treated group, there were some differences in the extent of hearing loss and associated pathology. This finding is consistent with clinical reports in which various degrees of hearing loss are seen after HIE. Cochleas, brainstems and temporal lobes of the pups were histopathologically evaluated. Apoptotic changes in brainstem at the level of cochlear nuclei were shown by cresyl violet staining, TUNEL assay and caspase-3 stainings in HIE group. Our observation indicated that the brain stem tissue damage mainly reflected apoptotic cell death. Apoptotic changes within hair cells, spiral ganglion cells and temporal lobes were also detected by using ultrastructural assessments. A prominent finding was cellular degeneration with the loss of ciliae of hair cells in the cochlea. Ultrastructural damage in spiral ganglion cells and temporal lobes was also clearly detected in the HIE-alone group. Those findings are in accordance with previous postmortem studies as well as in vitro experiments [1,6,7]. Studies on prevention of apoptosis in the central nervous system induced by HIE increased after 2000. Specifically, detection of EPO receptors first in guinea pig [14] and later in rat brains [15] led to use of rhEPO in experimental studies. As EPO is an endogenous protein and has clinical application for a number of different indications, including use in newborns, it has gained attention. It has now been demonstrated that hypoxic conditions activated EPO receptors and increased endogenous EPO production in organotypic cochlea cell cultures [25]. Those findings indicate that rhEPO given exogenously can easily bind and activate EPO receptors. Recombinant EPO seems to be advantageous in protective strategies. The potential efficacy of EPO as an otoprotective agent is supported by the observation that EPO prevents gentamycin ototoxicity in rat cochlea cell lines [16]. rhEPO also prevented apoptosis and necrosis due to hypoxia in newborn rat cochlea cell culture [9]. For those reasons in the second part of the study, rhEPO was chosen for investigation of possible otoprotective properties. ABR results demonstrated that hearing loss was not induced in HIE animals with rhEPO treatment. Those electrophysiological findings were also supported by histopathological investigations. TUNEL assay and caspase-3 staining procedures showed that apoptosis was prevented by the application of rhEPO. Similar results were found on the electronmicroscopic studies of the rhEPO treated group; mitochondria and ciliae of the hair cells were in normal configuration, similar to the control and sham groups. Spiral ganglion cells were also intact, as well as temporal lobe structures. These findings are consistent with results in in vitro studies. 5. Conclusion Our findings confirm the hypotheses that HIE-induced hearing loss in the newborn is based on upregulation of cell death pathways leading to apoptosis and that these changes are observed in both the peripheral and central auditory system from cochlear nucleus to the temporal cortex and that rhEPO can significantly attenuate these changes and protect hearing. In a newborn rat HIE induced hearing loss bilaterally. HIE causes
745
apoptosis in hair cells and spiral ganglion cells, which may well explain the induced changes in hearing sensitivity shown in the ABR results. However as well apoptotic cell death was clearly evident in the brainstem cochlear nucleus and temporal lobe tissues in the newborn. These changes indicate that changes in central processing of auditory signals and the perception of complex signals such as speech and signals in noise may be compromised by HIE. Our finding indicates that rhEPO is effective in the protection of hearing as shown by ABR records. However of equal clinical importance rhEPO also reduced central auditory pathology. Thus, apoptotic changes due to HIE within the cochlea, brainstem and temporal lobe are all significantly attenuated by rhEPO. Our results suggest that rhEPO may have a protective effect against HIE induced hearing loss, both in peripheral sensitivity as well as complex sound processing and perception in the central nervous system. These studies provide the scientific rational to support the needed clinical trials to translate these observations to eventual prevention of hearing loss in human infants. References [1] J. Schumutzhard, R. Glueckert, C. Sergi, I. Schwentner, I. Abraham, A. SchrottIscher, Does perinatal asphyxia induce apoptosis in the inner ear? Hear. Res. 250 (2009) 1–9. [2] R.C. Vannucci, S.J. Vannucci, A model of perinatal hypoxic-ischemic brain damage, Ann NY Cad. Sci. 835 (1997) 243–249. [3] M. Sano, K. Kaga, E. Kitazumi, K. Kodama, Sensorineural hearing loss in patients with cerebral palsy after asphyxia and hyperbilirubinemia, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 1211–1217. [4] Z.D. Jiang, Long-term of perinatal and postnatal asphyxia on developing human auditory brainstem responses: peripheral hearing loss, Int. J. Pediatr. Otorhinolaryngol. 33 (1995) 225–238. [5] Z.D. Jiang, D.M. Brosi, A. Wilkinson, Depressed brainstem auditory electrophysiology in preterm infants after perinatal hypoxia-ischemia, J. Neurol. Sci. 281 (2009) 28–33. [6] Y. Orita, I. Sando, M. Miura, S.I. Haginomori, B.E. Hirsch, Cochleosaccular pathology after perinatal and postnatal asphyxia: histopathologic findings, Otol. Neurotol. 23 (2002) 34–38. [7] S. Koyama, K. Kaga, H. Sakata, Y. Ino, K. Kodere, Pathological findings in the temporal bone of newborn infants with neonatal asphyxia, Acta Otolarngol. 125 (2005) 1028–1032. [8] B. Mazurek, E. Winter, J. Fuchs, H. Haupt, J. Gross, Susceptibility of the hair cells of the newborn rat cochlea to hypoxia and ischemia, Hear. Res. 182 (2003) 2–8. [9] N. Andreeva, A. Nyamaa, H. Haupt, J. Gross, B. Mazurek, Recombinant human erythropoietin prevents ischemia-induced apoptosis and necrosis in explant cultures of the rat organ of Corti, Neurosci. Lett. 396 (2006) 86–90. [10] C.T. Noguchi, P. Asavaritikrai, R. Teng, Y. Jia, Role of erythropoietin in the brain, Crit. Rev. Oncog. 64 (2007) 159–171. [11] R.J. Mc Pherson, S.E. Juul, Recent trends in erythropoietin-mediated neuroprotection, Int. J. Dev. Neurosci. 26 (2008) 103–111. [12] M. Yamada, C. Burke, P. Colditz, D.W. Johnson, G.C. Gobe, Erythropoietin protects against apoptosis and increases expression of non-neuronal cell markers in the hypoxia-injured developing brain, J. Pathol. 224 (2011) 101–109. [13] A. Kumral, H. Baskın, D.C. Yesilirmak, B.U. Ergur, S. Aykan, S. Genc, Erythropoietin attenuates lipopolysaccharide-induced white matter injury in the neonatal rat brain, Neonatology 92 (2007) 269–278. [14] P. Caye´-Thomasen, N. Wagner, B.L. Frederiksen, K. Asal, J. Thomsen, Erythropoietin and erythropoietin receptor expression in the guinea pig inner ear, Hear. Res. 203 (2005) 21–27. [15] A.M. Naldi, M. Gassman, D. Bodmer, Erythropoietin but not VEGF has a protective effect on auditory hair cells in the inner ear, Cell. Mol. Life Sci. 66 (2009) 3593–3599. [16] A. Monge, I. Nagy, S. Bonaby, S. Schmid, M. Gassman, D. Bodmer, The effect of erythropoietin on gentamicin-induced auditory hair cell loss, Laryngoscope 116 (2006) 312–316. [17] B.L. Frederiksen, P. Caye´-Thomasen, S.P. Lund, N. Wagner, K. Asal, N.V. Olsen, et al., Does erythropoietin augment noise induced hearing loss? Hear. Res. 223 (2007) 129–137. [18] K. Tabuchi, K. Oikawa, I. Uemaetomari, S. Tsuji, T. Wada, A. Hara, A glucocorticoids and dehydroeiandrosterone sulfate ameloriate ischemia-induced injury of the cochlea, Hear. Res. 180 (2003) 51–56. [19] A. Kumral, N. Uysal, K. Tugyan, A. Sonmez, O. Yilmaz, N. Gokmen, Erythropoietin improves long-term spatial memory deficits and brain injury following neonatal hypoxia-ischemia in rats, Behav. Brain Res. 153 (2004) 77–86. [20] J.E. Rice, R.C. Vannucci, T.B. Bierly, The influence of immaturity on hypoxicischemic brain damage in the rat, Ann. Neurol. 9 (1981) 131–141. [21] G. Paxionos, C. Watson, Atlas: The Rat Brain in Sterotaxic Coordinates, fourth ed., Academic Press, 1998.
746
Y. Olgun et al. / International Journal of Pediatric Otorhinolaryngology 77 (2013) 739–746
[22] Z.D. Jiang, X.Y. Liu, B.P. Shi, L. Lin, C.F. Bu, A.R. Wilkinson, Brainstem auditory outcomes and correlation with neurodevelopment after perinatal asphyxia, Pediatr. Neurol. 39 (2008) 189–195. [23] H.S. Romijn, M.A. Hofman, A. Gramsbergen, At what age is the developing cerebral cortex of the rat comparable to that of the full-term human baby? Early Hum. Dev. 26 (1991) 61–67.
[24] J. Tomfighi, D. Mauger, R.C. Vannucci, S.J. Vannucci, Influence of age on the cerebral lesions in an immature rat model of cerebral hypoxic-ischemia: a light microscopic study, Dev. Brain Res. 100 (1997) 149–160. [25] J. Gross, R. Moller, N. Amarjagal, A. Machulik, Fuchs J., U. Ungethu¨m, Expression of erythropoietin and angiogenic growth factors following inner ear injury of newborn rats, Prague Med. Rep. 110 (2009) 310–331.