- Email: [email protected]
Effect of round window membrane application of nitric oxide on hearing and nitric oxide concentration in perilymph
International Journal of Pediatric Otorhinolaryngology (2003) 67, 585 /590
www.elsevier.com/locate/ijporl
Effect of round window membrane application of nitric oxide on hearing and nitric oxide concentration in perilymph Jonathan B. Hanson, Paul T. Russell, Andy T.A. Chung, Claire S. Kaura, Samantha H. Kaura, Earnest O. John, Timothy T.K. Jung* Department of Surgery, Division of Otolaryngology /Head and Neck Surgery, Loma Linda University School of Medicine, Jerry L. Pettis Memorial Veterans Medical Center, Loma Linda, CA 92354, USA Received 29 July 2002; received in revised form 2 January 2003; accepted 5 January 2003
KEYWORDS Nitric oxide; S -Nitroso-N -acetylpenicillamine; Round window membrane; Hearing-loss
Summary Nitric oxide (NO), a free radical, has been found to be important in the development of middle ear effusions. However, the effect of NO in the middle ear effusion on cochlear function and on perilymph concentrations of NO has not been reported. We placed S -nitroso-N -acetylpenicillamine (SNAP), a NO donor compound, on the round window membrane (RWM) of adult chinchillas. Auditory brainstem response (ABR) thresholds were measured before and after the placement of SNAP on the RWM and hourly for 8 h after SNAP placement. Samples of perilymph were collected 2 h after application of SNAP and were assayed for total nitrate and nitrite, the end products of NO. Experimental ears demonstrated significant ABR threshold elevations after 5 h and elevated nitrate/nitrite in the perilymph. These findings suggest that NO present in the middle ear passes through the RWM into the inner ear and can cause significant hearing loss. – 2003 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction Sensorineural hearing loss (SNHL) is associated with many causes such as increasing age [1], otitis media with effusion [2], chronic otitis media [3] and bacterial meningitis [4]. The pathophysiologic mechanism of SNHL in otitis media is not completely understood. It has been suggested that SNHL in otitis media is caused by toxic substances, such as nitric oxide present in the middle ear effusion. The exact role of nitric oxide (NO), a free radical, in the *Corresponding author. Tel.: /1-909-352-7920; fax: /1-909352-7908. E-mail address: [email protected] (T.T.K. Jung).
pathogenesis of middle ear inflammation and in cochlear injury is still unknown. Evidence suggests that as an inflammatory mediator, NO increases middle ear vascular permeability [5] and mucin secretion [6] in endotoxininduced otitis media with effusion. It is strongly suspected that NO plays an important role in the ototoxicity of pneumolysin, an exotoxin from Streptococcus pneumoniae [7]. Amaee et al. demonstrated a protective effect of NG-methyl-Larginine, an inhibitor of NO release, on cochlear injury [8]. Our previous work has shown that NO causes irreversible morphologic changes in isolated cochlear outer hair cells using an in vitro model (unpublished data).
0165-5876/03/$ - see front matter – 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0165-5876(03)00035-1
586
The permeability of the RWM and the passage of compounds from the middle ear into the perilymph have been well documented [9 /11]. Inflammatory mediators have been implicated as a cause of SNHL by crossing the RWM and injuring the sensory cells of the cochlea [12]. If NO in middle ear effusion crosses through the RWM to the inner ear space, it may be possible that the NO radical may induce SNHL. Although NO appears to be important in the formation of middle ear effusions and can cause cochlear injury, the effect on the inner ear of NO passing from the middle ear space has yet to be documented. The purpose of this study was to determine the change in cochlear function and the concentration of NO in the perilymph when the NO donor compound S -nitroso-N-acetyl, L-penicillamine (SNAP) is placed on the RWM.
2. Materials and methods Randomly selected healthy male and female adult chinchillas were used for the study. All experiments were performed under anesthesia using intramuscular ketamine (20 mg/kg) and xylazine (0.5 mg/kg). Baseline ABRs were determined for all test animals in both ears. The right RWM was then exposed through a posterior inferior approach to the bulla, leaving the tympanic membrane and ossicles undisturbed. A small piece (2 / 2 mm2) of pressed Gelfoam (Pharmacia and Upjohn, Inc., Bridgewater) was saturated with experimental or control solution and applied on the RWM. SNAP was purchased from Alexis, San Diego. This experiment was performed in two parts. In the first part, we studied the effect of NO on the cochlear function of chinchillas. The second part consisted of perilymph assays that we performed to demonstrate the diffusion of NO through the RWM into the inner ear. Eleven healthy adult chinchillas, weighing between 400 and 600 g, were randomly assigned to the first part of the experiment, while seven of the animals were assigned to the second part of the experiment. The cochlear function part consisted of two groups. One group of animals (Group 1, n /7) had one of their ears operated on and received 4.89/1.5 ml of freshly prepared SNAP (0.5 mg/250 ml) on Gelfoam placed on the right RWM of the operated ear. The contralateral left ear was not operated on and did not receive SNAP on the RWM. The contralateral ear served as an intra-animal control (n /7). ABR thresholds on Group 1 were noted on both ears before the operation at t /0 and after the operation at t /1, 2, 3, 4, 5, 6, 7, 8
J.B. Hanson et al.
Fig. 1 Mean ABR threshold loss in SNAP-OP versus Inactivated SNAP-OP groups from 1 to 8 h after application of inactive or fresh SNAP solution on the RWM. There was significant (P B/0.05) elevation of ABR threshold after 5 h in the experimental group compared to controls. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, L-penicillamine; RWM, round window membrane.
h. Hearing loss was calculated and compared for the SNAP-operated, SNAP contralateral ears (Fig. 2, Table 1, column 3). Threshold losses for these two ears (SNAP operated and SNAP contralateral) were compared using a two-tailed Student’s t -test. The second group (Group 2) consisted of a group of animals (n /4) in which an operated ear received 4.89/1.5 ml of inactivated SNAP (0.5 mg/ 250 ml) on Gelfoam placed on the right RWM to note the toxic effects of SNAP metabolites. The inactivated SNAP was prepared by heating (458 for 2 h) freshly prepared SNAP (0.5 mg/250 ml) and leaving the SNAP for 3 /4 days at room temperature (t1/2 /10.3 h) in light for complete evaporation of NO. The contralateral ear (n /4) in this group was not operated on and did not receive inactivated SNAP on the RWM. In the inactivated SNAP
Fig. 2 Mean ABR threshold loss in SNAP-OP versus SNAP Non-OP groups 1 /8 h after application of fresh SNAP solution on the RWM. There was significant (P B/0.05) elevation of ABR threshold after 3 h in the experimental group compared to contra-lateral ear. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, L-penicillamine; RWM, round window membrane.
P values obtained by t -test, OP, operated ears; Non-OP, Contra-lateral ears. * P 5/0.05.
0.50 0.40 0.97 0..92 0.64 0.76 0.51 0.45 0.21 0.09 0.03* 0.002* 0.007* 0.008* 0.001* 0.002* 0.37 0.73 0.77 0.35 0.13 0.02* 0.02* 0.009* 1 2 3 4 5 6 7 8
SNAP-OP versus Inactivated SNAP-OP versus Contralateral SNAP-OP (Fig. 1) SNAP Non-OP (Fig. 2)
Experimental and control groups Time (h)
Statistical analysis of experimental and control groups Table 1
group, the operated ears and the contralateral ears served as controls. ABR thresholds on Group 2 were noted on both ears before the operation at t /0 and after the operation at t /1, 2, 3, 4, 5, 6, 7, 8 h. The hearing loss was calculated and compared for both ear types in Group 2 (inactivated SNAPoperated, inactivated SNAP contralateral) (Fig. 4; Table 1, column 5). Threshold losses were compared using a two-tailed Student’s t -test. We also compared SNAP-OP versus Inactivated SNAP-OP (Fig. 1; Table 1, column 2) and Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP (Fig. 3; Table 1, column 4). Auditory brainstem response potentials were recorded between subdermal needle electrodes at the vertex and the ipsilateral mastoid portion of the bulla. Responses were recorded using clinical averager (Nicolet CA 1000), physical amplifier (Nicolet HGA-200A) and click stimulus generator (NIC-1001). The click response was averaged selecting the settings of averager at: 150 /1500 Hz band width filter, 10 ms time, 1125 repetition, 33.3/s rate and 100 ms duration. Bandwidth selection was based on filters set on the CA 1000 (low 150 Hz, high 1500 Hz). The filter range was chosen because of myogenic artifact from animals precluded using higher frequencies. The probe used was the Nicolet Tubal Insert Earphone (TIP-10, 10 ohms) with a frequency response of 0.1 /10 kHz. The physical amplifier was set at a gain of 104 with an electrode impedance of B/5. The hearing threshold was determined by the lowest auditory stimulus that produced detectable and reproducible ABR waveforms. For the perilymph assay part of the experiment, seven animals were used. All chinchillas used in
Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP (Fig. 3)
Fig. 3 Mean ABR threshold loss in Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP groups 1 /8 h after initial ABR threshold measurement. There was no significant difference in hearing loss between these two groups throughout the 8 h duration of the experiment. At t /8, P /0.05. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, Lpencillamine.
0.05* 0.33 0.33 0.38 0.33 0.64 0.74 0.89
587
Inactivated-SNAP OP versus Contralateral Inactivated SNAP Non-OP (Fig. 4)
Effect of round window membrane application of nitric oxide
588
J.B. Hanson et al.
Welfare Act (7 USC et seq.). The animal use protocol was approved by the Institutional Animal Care and Use Committee of J.L. Pettis Memorial Veterans Hospital.
3. Results
Fig. 4 Mean ABR threshold loss in Inactivated SNAP-OP versus Contralateral Inactivated SNAP Non-OP 1 /8 h after initial application of Inactivated SNAP solution on the right RWM. There was a significant difference between these two groups for hearing loss at t /1 (P /0.04), but there was no significant difference between these groups at all other hours. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, Lpencillamine; RWM, round window membrane.
this study were screened at 0 dBN HL. Animals not showing replicable responses at 0 dBN HL or better were not included. The 0 dBN HL setting corresponds to a peak equivalent of 30 dBN SPL. SNAP was dissolved in normal saline solution at a concentration of 10 mM. SNAP (4.89/1.5 ml) on Gelfoam was applied on the RWM to one ear (operated) on all animals and after 2 h perilymph was collected from both ears and was analyzed separately (Table 1). After the placement of freshly prepared SNAP, animals were euthanized by decapitation while under deep anesthesia using ketamine (80 mg/kg) and xylazine (2 mg/kg) 2 h following application. The temporal bones and cochleae were quickly harvested from each animal and perilymph was collected from each side (10 /15 ml) using a blunt 28 gauge needle and a 1 ml syringe. The samples of perilymph were assayed by colorimetric assay (Griess method) for total nitrite/nitrate using a kit purchased from Cayman Chemical Co., Ann Arbor. Each sample was diluted to 80 ml with assay buffer. In addition, 10 ml of enzyme cofactor and 10 ml of nitrate reductase were added. The sample was incubated at room temperature for 3 h to convert nitrate to nitrite and then color reagents were added. Absorbance at 540 nm was measured using a plate reader and absorbance data was converted to concentration values for each specimen. The samples for NO assays were blinded. Statistical significance was noted using a two-tailed Student’s t-test. This study was performed in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, the NIH Guide for the Care and Use of Laboratory Animals and the Animal
3.1. Part 1: cochlear function group The results of this study are summarized in Table 1. When comparing SNAP-OP versus Inactivated SNAP-OP, there were significant moderate to severe ABR threshold losses after 6 h following application of SNAP on the RWMs (Fig. 1). The average hearing loss observed at t /8 h for the SNAP-OP group was 529/7 dB and for the Inactivated SNAP-OP group, it was 169/8 dB (Fig. 1). The hearing losses for the SNAP-OP group were significantly different from the Inactivated SNAP-OP group at t /6, 7, 8 h (Table 1, column 2). The SNAP-OP group was also compared with the Contralateral SNAP Non-OP group. It was found that these two groups had significantly different average hearing losses at t /3, 4, 5, 6, 7, 8 h (Table 1, column 3). At t /8, the average hearing loss was 529/7 dB for the SNAP-OP group and 219/4 dB for the Contralateral SNAP Non-OP group (Fig. 2). When the Contralateral SNAP Non-OP group was compared with the Contralateral Inactivated SNAP Non-OP group, no significant differences in hearing loss were observed for all 8 h (Table 1, column 4; Fig. 3). Similarly, when the Inactivated SNAP-OP group was compared with the contralateral Inactivated SNAP Non-OP group, no significant differences in hearing loss were observed for all 8 h (Table 1, column 5; Fig. 4). Table 2 Concentrations of NO-metabolites in the perilymph Experiment No.
Contra lateral
Operated ear
1 2 3 4 5 6 7
154 220 27 219 210 19 55
687 551 690 131 229 200 289
Mean
1299/35
3979/90
NO metabolites levels (NO2 /NO3 ) in mM, (P / 0.02).
Effect of round window membrane application of nitric oxide
3.2. Part 2: perilymph assay group The results of part 2 are summarized in Table 2. In operated ears with fresh SNAP placed on the RWM, the total nitrate/nitrite concentration in perilymph ranged from 131 to 690 mM (mean9/S.E. 3979/90, n /7). The range for the non-operated contralateral ears was 19 /220 mM (mean9/S.E. 1299/35). The levels of NO metabolites (NO2 / NO3 ) in the perilymph of operated ears were three times greater than in the non-operated ears, which is statistically significant (P /0.02).
4. Discussion Nitric oxide is a biologically active molecule with many functions. It is produced in trace quantities by neurons, endothelial cells, platelets and neutrophils in response to homeostatic stimuli. Other cells, such as macrophages, fibroblasts and hepatocytes, produce NO in micromolar concentrations in response to inflammatory or mitogenic stimuli. These higher levels of NO lead to the formation of peroxynitrite, destruction of iron-sulphur clusters, thiol nitrosation and nitration of protein trypsin residues. The production of NO in different biological systems varies over several orders of magnitude. In addition to the presence of NO in the middle ear as an inflammatory mediator [6], as a free radical, it may be directly responsible for significant cellular injury. Halliwell describes the cytotoxicity of reactive oxygen species such as NO, hydrogen peroxide, the superoxide anion, the hydroxyl radical and peroxynitrite [13]. Each of these free radicals or radical producing compounds can cause lipid peroxydation, DNA strand breakage, carbohydrate and protein damage and can alter membrane-bound enzymes and receptors [14]. This damage changes cell membrane fluidity and inhibits the cell’s ability to maintain ionic gradients [13,15]. The NO donor compound, SNAP, is a stable analog of endogenous S -nitroso compounds and a good source of NO [16,17]. Amaee et al. [8] found that SNAP solution exposed to light for 3 /4 days loses its ability to injure the cochlea when directly infused into the scala tympani of guinea pigs. By using light-exposed SNAP as the control solution and comparing it to freshly prepared SNAP, the threshold losses induced in experimental ears can be attributed directly to the presence of the NO radical itself in the inner ear. Indeed, significant threshold elevations were found after 5 h following the application of fresh SNAP to the RWM when compared to the control (Fig. 1; Table 1,
589
column 2). The hearing loss for the SNAP-OP ears ranged from mild to severe (9 /52 dB). Mild hearing loss possibly caused by fatigue and the effect of anesthesia was observed in Contralateral Non-OP ears with SNAP (3 /21 dB), Inactivated SNAP-OP ears (16 /20 dB) and Inactivated SNAP-Non-OP ears (2 /15 dB). There may have been some cross-over of NO from SNAP-OP ear to Contralateral Non-OP ear. We strongly suspect that the NO radical is passing through the RWM into the perilymph and damaging the cochlea. Molecular size should not prevent passage of NO through the RWM as much larger molecules, such as albumin, have been found to cross it. It is known that in vivo nitric oxide has a high diffusion coefficient (3.3 /10 5 cm2/s) and because of the permeability and thinness of the RWM (11 mm in rodents), it can reach the perilymph within fraction of seconds [19,20]. The final products of NO release in vivo are nitrite and nitrate ions (NO2 and NO3 ). The relative amounts of NO2 and NO3 formed are variable, thus the best quantitative index of exposure to the NO radical is the sum of these two ion groups [18]. The levels of NO metabolites (NO2 /NO3 ) in the perilymph of operated (3979/90 mM) ears were three times greater and statistically different (P /0.02) compared to non-operated ears (1299/35 mM), which validates that NO radical passed through the RWM into the perilymph. The unilateral auditory threshold loss found in our study suggests a local cytotoxic process affected by the NO radical itself. This is a control based study and we believe that random errors causing lack of reproducibility in a control group are the same as in the experimental group. Low precisions and wide ranges for NO metabolites for the contralateral (19 /220 mM) and the operated ear (131 /687 mM) were attributed to random errors caused by inconsistencies associated with SNAP application on the RWM, individual differences between animals and assay technique. Some possible sources of error associated with SNAP applications are: surface area of RWM touching the Gelfoam, amount of SNAP solution on Gelfoam (4.89/1.5 ml), Gelfoam soaked with SNAP solution contamination with middle ear secretion or blood. Application of SNAP on RWM is a tricky procedure because the RWM is slightly exposed and most it is hidden between the bones and sources of error associated with RWM applications are difficult to detect and correct. Our findings support that NO, a free radical detected in middle ear effusions, when applied to the RWM, passes through it into the perilymph and induces significant changes in cochlear function and levels of NO-metabolites. These findings sug-
590
J.B. Hanson et al.
gest involvement of the NO radical in the pathogenesis of SNHL associated with otitis media.
References [1] R.W. Babin, L.A. Harker, The vestibular system in the elderly, Otolaryngol. Clin. North Am. 15 (1982) 387 /393. [2] A. Aviel, E. Ostfeld, Acquired irreversible sensorineural hearing loss associated with effusion, Am. J. Otolaryngol. 3 (1982) 217 /222. [3] M.M. Paparella, D.R. Brady, R. Hoel, Sensori-neural hearing loss in chronic otitis media and mastoiditis, Trans. Am. Acad. Ophthalmol. Otolaryngol. 74 (1970) 108 /115. [4] P.R. Dodge, H. Davis, R.D. Feign, Prospective evaluation of hearing impairment as a sequela of acute bacterial meningitis, New Engl. J. Med. 311 (1984) 869 /874. [5] S.S. Ball, J. Prazma, C.G. Dais, et al., Nitric oxide: a mediator of endotoxin-induced middle ear effusions, Laryngoscope 106 (1996) 1021 /1027. [6] A.S. Rose, M.J. Prazma, S.H. Randell, et al., Nitric oxide mediates secretion in endotoxin-induced otitis media with effusion, Otolaryngol. Head Neck Surg. 116 (1997) 308 / 316. [7] F.R. Amaee, S.D. Comis, M.P. Osborne, NG-Methyl-L-arginine protects the guinea pig cochlea from the cytotoxic effects of pneumolysin, Acta Otolaryngol. (Stockh.) 115 (1995) 386 /391. [8] F.R. Amaee, S.D. Comis, M.P. Osborne, et al., Possible involvement of nitric oxide in the sensorineural hearing loss of bacterial meningitis, Acta Otolaryngol. (Stockh.) 117 (1997) 329 /336. [9] M. Suzuki, H. Kawauchi, T. Fujiyoshi, et al., Round window membrane and perilymph in experimental otitis media with effusion, Ann. Otol. Rhinol. Laryngol. 98 (1989) 980 /987.
.
[10] M.V. Goycoolea, M.M. Paparella, B. Goldberg, et al., Permeability of the round window membrane in otitis media, Arch. Otolaryngol. 106 (1980) 430 /433. [11] S.H. Lee, H.W. Woo, T.T.K. Jung, et al., Permeability of arachidonic acid metabolites through the round window membrane in chinchillas, Acta Otolaryngol. Suppl. (Stockh.) 493 (1992) 165 /169. [12] T. Morizono, T. Tono, Middle ear inflammatory mediators and cochlear function, Otolaryngol. Clin. North Am. 24 (1991) 835 /843. [13] B. Halliwell, J.M.C. Gutterridge, Free Radicals in Biology and Medicine, Clarendon press, Oxford, 1989. [14] J.F. Ghersi-Egea, M.H. Livertoux, A. Minn, et al., Enzyme mediated superoxide radical formation initiated by exogenous molecules in rat brain preparations, Toxicol. Appl. Pharmacol. 110 (1991) 107 /117. [15] B. Halliwell, Reactive oxygen species and the central nervous system, J. Neurochem. 59 (1992) 1609 /1623. [16] J.E. Shaffer, B.J. Han, W.H. Chern, et al., Lack of tolerance to a 24-hour infusion of S -nitroso N -acetylpenicillamine (SNAP) in conscious rabbits, J. Pharmacol. Exp. Ther. 260 (1992) 286 /293. [17] J.A. Bauer, H.L. Fung, Differential hemodynamic effects and tolerance properties of nitroglycerin and an S -nitrosothiol in experimental heart failure, J. Pharmacol. Exp. Ther. 256 (1991) 249 /254. [18] R.W. Nims, J.F. Darbyshire, J.E. Saavedra, et al., Colorimetric methods for the determination of nitric oxide concentration in neutral aqueous solutions, Methods 7 (1995) 48 /54. [19] J. Garthwaite, C.L. Boulton, Nitric oxide signaling in the central nervous system, Annu. Rev. Physiol. 57 (1995) 683 / 706. [20] L. Nordang, M. Anniko, Hearing loss in relation to round window membrane morphology in experimental chronic otitis media, ORL J. Otorhinolaryngol. Relat. Spec. 63 (6) (2001) 333 /340.
www.elsevier.com/locate/ijporl
Effect of round window membrane application of nitric oxide on hearing and nitric oxide concentration in perilymph Jonathan B. Hanson, Paul T. Russell, Andy T.A. Chung, Claire S. Kaura, Samantha H. Kaura, Earnest O. John, Timothy T.K. Jung* Department of Surgery, Division of Otolaryngology /Head and Neck Surgery, Loma Linda University School of Medicine, Jerry L. Pettis Memorial Veterans Medical Center, Loma Linda, CA 92354, USA Received 29 July 2002; received in revised form 2 January 2003; accepted 5 January 2003
KEYWORDS Nitric oxide; S -Nitroso-N -acetylpenicillamine; Round window membrane; Hearing-loss
Summary Nitric oxide (NO), a free radical, has been found to be important in the development of middle ear effusions. However, the effect of NO in the middle ear effusion on cochlear function and on perilymph concentrations of NO has not been reported. We placed S -nitroso-N -acetylpenicillamine (SNAP), a NO donor compound, on the round window membrane (RWM) of adult chinchillas. Auditory brainstem response (ABR) thresholds were measured before and after the placement of SNAP on the RWM and hourly for 8 h after SNAP placement. Samples of perilymph were collected 2 h after application of SNAP and were assayed for total nitrate and nitrite, the end products of NO. Experimental ears demonstrated significant ABR threshold elevations after 5 h and elevated nitrate/nitrite in the perilymph. These findings suggest that NO present in the middle ear passes through the RWM into the inner ear and can cause significant hearing loss. – 2003 Elsevier Science Ireland Ltd. All rights reserved.
1. Introduction Sensorineural hearing loss (SNHL) is associated with many causes such as increasing age [1], otitis media with effusion [2], chronic otitis media [3] and bacterial meningitis [4]. The pathophysiologic mechanism of SNHL in otitis media is not completely understood. It has been suggested that SNHL in otitis media is caused by toxic substances, such as nitric oxide present in the middle ear effusion. The exact role of nitric oxide (NO), a free radical, in the *Corresponding author. Tel.: /1-909-352-7920; fax: /1-909352-7908. E-mail address: [email protected] (T.T.K. Jung).
pathogenesis of middle ear inflammation and in cochlear injury is still unknown. Evidence suggests that as an inflammatory mediator, NO increases middle ear vascular permeability [5] and mucin secretion [6] in endotoxininduced otitis media with effusion. It is strongly suspected that NO plays an important role in the ototoxicity of pneumolysin, an exotoxin from Streptococcus pneumoniae [7]. Amaee et al. demonstrated a protective effect of NG-methyl-Larginine, an inhibitor of NO release, on cochlear injury [8]. Our previous work has shown that NO causes irreversible morphologic changes in isolated cochlear outer hair cells using an in vitro model (unpublished data).
0165-5876/03/$ - see front matter – 2003 Elsevier Science Ireland Ltd. All rights reserved. doi:10.1016/S0165-5876(03)00035-1
586
The permeability of the RWM and the passage of compounds from the middle ear into the perilymph have been well documented [9 /11]. Inflammatory mediators have been implicated as a cause of SNHL by crossing the RWM and injuring the sensory cells of the cochlea [12]. If NO in middle ear effusion crosses through the RWM to the inner ear space, it may be possible that the NO radical may induce SNHL. Although NO appears to be important in the formation of middle ear effusions and can cause cochlear injury, the effect on the inner ear of NO passing from the middle ear space has yet to be documented. The purpose of this study was to determine the change in cochlear function and the concentration of NO in the perilymph when the NO donor compound S -nitroso-N-acetyl, L-penicillamine (SNAP) is placed on the RWM.
2. Materials and methods Randomly selected healthy male and female adult chinchillas were used for the study. All experiments were performed under anesthesia using intramuscular ketamine (20 mg/kg) and xylazine (0.5 mg/kg). Baseline ABRs were determined for all test animals in both ears. The right RWM was then exposed through a posterior inferior approach to the bulla, leaving the tympanic membrane and ossicles undisturbed. A small piece (2 / 2 mm2) of pressed Gelfoam (Pharmacia and Upjohn, Inc., Bridgewater) was saturated with experimental or control solution and applied on the RWM. SNAP was purchased from Alexis, San Diego. This experiment was performed in two parts. In the first part, we studied the effect of NO on the cochlear function of chinchillas. The second part consisted of perilymph assays that we performed to demonstrate the diffusion of NO through the RWM into the inner ear. Eleven healthy adult chinchillas, weighing between 400 and 600 g, were randomly assigned to the first part of the experiment, while seven of the animals were assigned to the second part of the experiment. The cochlear function part consisted of two groups. One group of animals (Group 1, n /7) had one of their ears operated on and received 4.89/1.5 ml of freshly prepared SNAP (0.5 mg/250 ml) on Gelfoam placed on the right RWM of the operated ear. The contralateral left ear was not operated on and did not receive SNAP on the RWM. The contralateral ear served as an intra-animal control (n /7). ABR thresholds on Group 1 were noted on both ears before the operation at t /0 and after the operation at t /1, 2, 3, 4, 5, 6, 7, 8
J.B. Hanson et al.
Fig. 1 Mean ABR threshold loss in SNAP-OP versus Inactivated SNAP-OP groups from 1 to 8 h after application of inactive or fresh SNAP solution on the RWM. There was significant (P B/0.05) elevation of ABR threshold after 5 h in the experimental group compared to controls. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, L-penicillamine; RWM, round window membrane.
h. Hearing loss was calculated and compared for the SNAP-operated, SNAP contralateral ears (Fig. 2, Table 1, column 3). Threshold losses for these two ears (SNAP operated and SNAP contralateral) were compared using a two-tailed Student’s t -test. The second group (Group 2) consisted of a group of animals (n /4) in which an operated ear received 4.89/1.5 ml of inactivated SNAP (0.5 mg/ 250 ml) on Gelfoam placed on the right RWM to note the toxic effects of SNAP metabolites. The inactivated SNAP was prepared by heating (458 for 2 h) freshly prepared SNAP (0.5 mg/250 ml) and leaving the SNAP for 3 /4 days at room temperature (t1/2 /10.3 h) in light for complete evaporation of NO. The contralateral ear (n /4) in this group was not operated on and did not receive inactivated SNAP on the RWM. In the inactivated SNAP
Fig. 2 Mean ABR threshold loss in SNAP-OP versus SNAP Non-OP groups 1 /8 h after application of fresh SNAP solution on the RWM. There was significant (P B/0.05) elevation of ABR threshold after 3 h in the experimental group compared to contra-lateral ear. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, L-penicillamine; RWM, round window membrane.
P values obtained by t -test, OP, operated ears; Non-OP, Contra-lateral ears. * P 5/0.05.
0.50 0.40 0.97 0..92 0.64 0.76 0.51 0.45 0.21 0.09 0.03* 0.002* 0.007* 0.008* 0.001* 0.002* 0.37 0.73 0.77 0.35 0.13 0.02* 0.02* 0.009* 1 2 3 4 5 6 7 8
SNAP-OP versus Inactivated SNAP-OP versus Contralateral SNAP-OP (Fig. 1) SNAP Non-OP (Fig. 2)
Experimental and control groups Time (h)
Statistical analysis of experimental and control groups Table 1
group, the operated ears and the contralateral ears served as controls. ABR thresholds on Group 2 were noted on both ears before the operation at t /0 and after the operation at t /1, 2, 3, 4, 5, 6, 7, 8 h. The hearing loss was calculated and compared for both ear types in Group 2 (inactivated SNAPoperated, inactivated SNAP contralateral) (Fig. 4; Table 1, column 5). Threshold losses were compared using a two-tailed Student’s t -test. We also compared SNAP-OP versus Inactivated SNAP-OP (Fig. 1; Table 1, column 2) and Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP (Fig. 3; Table 1, column 4). Auditory brainstem response potentials were recorded between subdermal needle electrodes at the vertex and the ipsilateral mastoid portion of the bulla. Responses were recorded using clinical averager (Nicolet CA 1000), physical amplifier (Nicolet HGA-200A) and click stimulus generator (NIC-1001). The click response was averaged selecting the settings of averager at: 150 /1500 Hz band width filter, 10 ms time, 1125 repetition, 33.3/s rate and 100 ms duration. Bandwidth selection was based on filters set on the CA 1000 (low 150 Hz, high 1500 Hz). The filter range was chosen because of myogenic artifact from animals precluded using higher frequencies. The probe used was the Nicolet Tubal Insert Earphone (TIP-10, 10 ohms) with a frequency response of 0.1 /10 kHz. The physical amplifier was set at a gain of 104 with an electrode impedance of B/5. The hearing threshold was determined by the lowest auditory stimulus that produced detectable and reproducible ABR waveforms. For the perilymph assay part of the experiment, seven animals were used. All chinchillas used in
Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP (Fig. 3)
Fig. 3 Mean ABR threshold loss in Contralateral SNAP Non-OP versus Contralateral Inactivated SNAP Non-OP groups 1 /8 h after initial ABR threshold measurement. There was no significant difference in hearing loss between these two groups throughout the 8 h duration of the experiment. At t /8, P /0.05. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, Lpencillamine.
0.05* 0.33 0.33 0.38 0.33 0.64 0.74 0.89
587
Inactivated-SNAP OP versus Contralateral Inactivated SNAP Non-OP (Fig. 4)
Effect of round window membrane application of nitric oxide
588
J.B. Hanson et al.
Welfare Act (7 USC et seq.). The animal use protocol was approved by the Institutional Animal Care and Use Committee of J.L. Pettis Memorial Veterans Hospital.
3. Results
Fig. 4 Mean ABR threshold loss in Inactivated SNAP-OP versus Contralateral Inactivated SNAP Non-OP 1 /8 h after initial application of Inactivated SNAP solution on the right RWM. There was a significant difference between these two groups for hearing loss at t /1 (P /0.04), but there was no significant difference between these groups at all other hours. ABR, auditory brainstem-evoked response; SNAP, S -nitroso-N -acetyl, Lpencillamine; RWM, round window membrane.
this study were screened at 0 dBN HL. Animals not showing replicable responses at 0 dBN HL or better were not included. The 0 dBN HL setting corresponds to a peak equivalent of 30 dBN SPL. SNAP was dissolved in normal saline solution at a concentration of 10 mM. SNAP (4.89/1.5 ml) on Gelfoam was applied on the RWM to one ear (operated) on all animals and after 2 h perilymph was collected from both ears and was analyzed separately (Table 1). After the placement of freshly prepared SNAP, animals were euthanized by decapitation while under deep anesthesia using ketamine (80 mg/kg) and xylazine (2 mg/kg) 2 h following application. The temporal bones and cochleae were quickly harvested from each animal and perilymph was collected from each side (10 /15 ml) using a blunt 28 gauge needle and a 1 ml syringe. The samples of perilymph were assayed by colorimetric assay (Griess method) for total nitrite/nitrate using a kit purchased from Cayman Chemical Co., Ann Arbor. Each sample was diluted to 80 ml with assay buffer. In addition, 10 ml of enzyme cofactor and 10 ml of nitrate reductase were added. The sample was incubated at room temperature for 3 h to convert nitrate to nitrite and then color reagents were added. Absorbance at 540 nm was measured using a plate reader and absorbance data was converted to concentration values for each specimen. The samples for NO assays were blinded. Statistical significance was noted using a two-tailed Student’s t-test. This study was performed in accordance with the Public Health Service Policy on Humane Care and Use of Laboratory Animals, the NIH Guide for the Care and Use of Laboratory Animals and the Animal
3.1. Part 1: cochlear function group The results of this study are summarized in Table 1. When comparing SNAP-OP versus Inactivated SNAP-OP, there were significant moderate to severe ABR threshold losses after 6 h following application of SNAP on the RWMs (Fig. 1). The average hearing loss observed at t /8 h for the SNAP-OP group was 529/7 dB and for the Inactivated SNAP-OP group, it was 169/8 dB (Fig. 1). The hearing losses for the SNAP-OP group were significantly different from the Inactivated SNAP-OP group at t /6, 7, 8 h (Table 1, column 2). The SNAP-OP group was also compared with the Contralateral SNAP Non-OP group. It was found that these two groups had significantly different average hearing losses at t /3, 4, 5, 6, 7, 8 h (Table 1, column 3). At t /8, the average hearing loss was 529/7 dB for the SNAP-OP group and 219/4 dB for the Contralateral SNAP Non-OP group (Fig. 2). When the Contralateral SNAP Non-OP group was compared with the Contralateral Inactivated SNAP Non-OP group, no significant differences in hearing loss were observed for all 8 h (Table 1, column 4; Fig. 3). Similarly, when the Inactivated SNAP-OP group was compared with the contralateral Inactivated SNAP Non-OP group, no significant differences in hearing loss were observed for all 8 h (Table 1, column 5; Fig. 4). Table 2 Concentrations of NO-metabolites in the perilymph Experiment No.
Contra lateral
Operated ear
1 2 3 4 5 6 7
154 220 27 219 210 19 55
687 551 690 131 229 200 289
Mean
1299/35
3979/90
NO metabolites levels (NO2 /NO3 ) in mM, (P / 0.02).
Effect of round window membrane application of nitric oxide
3.2. Part 2: perilymph assay group The results of part 2 are summarized in Table 2. In operated ears with fresh SNAP placed on the RWM, the total nitrate/nitrite concentration in perilymph ranged from 131 to 690 mM (mean9/S.E. 3979/90, n /7). The range for the non-operated contralateral ears was 19 /220 mM (mean9/S.E. 1299/35). The levels of NO metabolites (NO2 / NO3 ) in the perilymph of operated ears were three times greater than in the non-operated ears, which is statistically significant (P /0.02).
4. Discussion Nitric oxide is a biologically active molecule with many functions. It is produced in trace quantities by neurons, endothelial cells, platelets and neutrophils in response to homeostatic stimuli. Other cells, such as macrophages, fibroblasts and hepatocytes, produce NO in micromolar concentrations in response to inflammatory or mitogenic stimuli. These higher levels of NO lead to the formation of peroxynitrite, destruction of iron-sulphur clusters, thiol nitrosation and nitration of protein trypsin residues. The production of NO in different biological systems varies over several orders of magnitude. In addition to the presence of NO in the middle ear as an inflammatory mediator [6], as a free radical, it may be directly responsible for significant cellular injury. Halliwell describes the cytotoxicity of reactive oxygen species such as NO, hydrogen peroxide, the superoxide anion, the hydroxyl radical and peroxynitrite [13]. Each of these free radicals or radical producing compounds can cause lipid peroxydation, DNA strand breakage, carbohydrate and protein damage and can alter membrane-bound enzymes and receptors [14]. This damage changes cell membrane fluidity and inhibits the cell’s ability to maintain ionic gradients [13,15]. The NO donor compound, SNAP, is a stable analog of endogenous S -nitroso compounds and a good source of NO [16,17]. Amaee et al. [8] found that SNAP solution exposed to light for 3 /4 days loses its ability to injure the cochlea when directly infused into the scala tympani of guinea pigs. By using light-exposed SNAP as the control solution and comparing it to freshly prepared SNAP, the threshold losses induced in experimental ears can be attributed directly to the presence of the NO radical itself in the inner ear. Indeed, significant threshold elevations were found after 5 h following the application of fresh SNAP to the RWM when compared to the control (Fig. 1; Table 1,
589
column 2). The hearing loss for the SNAP-OP ears ranged from mild to severe (9 /52 dB). Mild hearing loss possibly caused by fatigue and the effect of anesthesia was observed in Contralateral Non-OP ears with SNAP (3 /21 dB), Inactivated SNAP-OP ears (16 /20 dB) and Inactivated SNAP-Non-OP ears (2 /15 dB). There may have been some cross-over of NO from SNAP-OP ear to Contralateral Non-OP ear. We strongly suspect that the NO radical is passing through the RWM into the perilymph and damaging the cochlea. Molecular size should not prevent passage of NO through the RWM as much larger molecules, such as albumin, have been found to cross it. It is known that in vivo nitric oxide has a high diffusion coefficient (3.3 /10 5 cm2/s) and because of the permeability and thinness of the RWM (11 mm in rodents), it can reach the perilymph within fraction of seconds [19,20]. The final products of NO release in vivo are nitrite and nitrate ions (NO2 and NO3 ). The relative amounts of NO2 and NO3 formed are variable, thus the best quantitative index of exposure to the NO radical is the sum of these two ion groups [18]. The levels of NO metabolites (NO2 /NO3 ) in the perilymph of operated (3979/90 mM) ears were three times greater and statistically different (P /0.02) compared to non-operated ears (1299/35 mM), which validates that NO radical passed through the RWM into the perilymph. The unilateral auditory threshold loss found in our study suggests a local cytotoxic process affected by the NO radical itself. This is a control based study and we believe that random errors causing lack of reproducibility in a control group are the same as in the experimental group. Low precisions and wide ranges for NO metabolites for the contralateral (19 /220 mM) and the operated ear (131 /687 mM) were attributed to random errors caused by inconsistencies associated with SNAP application on the RWM, individual differences between animals and assay technique. Some possible sources of error associated with SNAP applications are: surface area of RWM touching the Gelfoam, amount of SNAP solution on Gelfoam (4.89/1.5 ml), Gelfoam soaked with SNAP solution contamination with middle ear secretion or blood. Application of SNAP on RWM is a tricky procedure because the RWM is slightly exposed and most it is hidden between the bones and sources of error associated with RWM applications are difficult to detect and correct. Our findings support that NO, a free radical detected in middle ear effusions, when applied to the RWM, passes through it into the perilymph and induces significant changes in cochlear function and levels of NO-metabolites. These findings sug-
590
J.B. Hanson et al.
gest involvement of the NO radical in the pathogenesis of SNHL associated with otitis media.
References [1] R.W. Babin, L.A. Harker, The vestibular system in the elderly, Otolaryngol. Clin. North Am. 15 (1982) 387 /393. [2] A. Aviel, E. Ostfeld, Acquired irreversible sensorineural hearing loss associated with effusion, Am. J. Otolaryngol. 3 (1982) 217 /222. [3] M.M. Paparella, D.R. Brady, R. Hoel, Sensori-neural hearing loss in chronic otitis media and mastoiditis, Trans. Am. Acad. Ophthalmol. Otolaryngol. 74 (1970) 108 /115. [4] P.R. Dodge, H. Davis, R.D. Feign, Prospective evaluation of hearing impairment as a sequela of acute bacterial meningitis, New Engl. J. Med. 311 (1984) 869 /874. [5] S.S. Ball, J. Prazma, C.G. Dais, et al., Nitric oxide: a mediator of endotoxin-induced middle ear effusions, Laryngoscope 106 (1996) 1021 /1027. [6] A.S. Rose, M.J. Prazma, S.H. Randell, et al., Nitric oxide mediates secretion in endotoxin-induced otitis media with effusion, Otolaryngol. Head Neck Surg. 116 (1997) 308 / 316. [7] F.R. Amaee, S.D. Comis, M.P. Osborne, NG-Methyl-L-arginine protects the guinea pig cochlea from the cytotoxic effects of pneumolysin, Acta Otolaryngol. (Stockh.) 115 (1995) 386 /391. [8] F.R. Amaee, S.D. Comis, M.P. Osborne, et al., Possible involvement of nitric oxide in the sensorineural hearing loss of bacterial meningitis, Acta Otolaryngol. (Stockh.) 117 (1997) 329 /336. [9] M. Suzuki, H. Kawauchi, T. Fujiyoshi, et al., Round window membrane and perilymph in experimental otitis media with effusion, Ann. Otol. Rhinol. Laryngol. 98 (1989) 980 /987.
.
[10] M.V. Goycoolea, M.M. Paparella, B. Goldberg, et al., Permeability of the round window membrane in otitis media, Arch. Otolaryngol. 106 (1980) 430 /433. [11] S.H. Lee, H.W. Woo, T.T.K. Jung, et al., Permeability of arachidonic acid metabolites through the round window membrane in chinchillas, Acta Otolaryngol. Suppl. (Stockh.) 493 (1992) 165 /169. [12] T. Morizono, T. Tono, Middle ear inflammatory mediators and cochlear function, Otolaryngol. Clin. North Am. 24 (1991) 835 /843. [13] B. Halliwell, J.M.C. Gutterridge, Free Radicals in Biology and Medicine, Clarendon press, Oxford, 1989. [14] J.F. Ghersi-Egea, M.H. Livertoux, A. Minn, et al., Enzyme mediated superoxide radical formation initiated by exogenous molecules in rat brain preparations, Toxicol. Appl. Pharmacol. 110 (1991) 107 /117. [15] B. Halliwell, Reactive oxygen species and the central nervous system, J. Neurochem. 59 (1992) 1609 /1623. [16] J.E. Shaffer, B.J. Han, W.H. Chern, et al., Lack of tolerance to a 24-hour infusion of S -nitroso N -acetylpenicillamine (SNAP) in conscious rabbits, J. Pharmacol. Exp. Ther. 260 (1992) 286 /293. [17] J.A. Bauer, H.L. Fung, Differential hemodynamic effects and tolerance properties of nitroglycerin and an S -nitrosothiol in experimental heart failure, J. Pharmacol. Exp. Ther. 256 (1991) 249 /254. [18] R.W. Nims, J.F. Darbyshire, J.E. Saavedra, et al., Colorimetric methods for the determination of nitric oxide concentration in neutral aqueous solutions, Methods 7 (1995) 48 /54. [19] J. Garthwaite, C.L. Boulton, Nitric oxide signaling in the central nervous system, Annu. Rev. Physiol. 57 (1995) 683 / 706. [20] L. Nordang, M. Anniko, Hearing loss in relation to round window membrane morphology in experimental chronic otitis media, ORL J. Otorhinolaryngol. Relat. Spec. 63 (6) (2001) 333 /340.