- Email: [email protected]
Combined effects of noise and styrene exposure on hearing function in the rat
Hearing Research 139 (2000) 86^96 www.elsevier.com/locate/heares
Combined e¡ects of noise and styrene exposure on hearing function in the rat R. Lataye, P. Campo *, G. Loquet Institut National de Recherche et de Se¨curite¨, Laboratoire de Neurotoxicologie et Immunotoxicologie, Avenue de Bourgogne, P.O. Box 27, 54501 Vandoeuvre, France Received 23 January 1999; received in revised form 30 August 1999; accepted 9 September 1999
Abstract Combined exposure to both noise and aromatic solvents such as styrene is common in many industries. In order to study the combined effects of simultaneous exposure to both noise and styrene on hearing, male adult Long-Evans rats were exposed either to 750 ppm styrene alone, to a 97 dB SPL octave band of noise centered at 8 kHz, or to a combination of noise and styrene. The exposure duration was 6 h/day, 5 days/week, for 4 consecutive weeks. Auditory function was tested over a frequency range from 2 to 32 kHz by recording near field potentials from the inferior colliculus, whereas histopathological analyses of the cochleae were performed with conventional morphometric approaches. Whereas both noise and styrene each caused permanent threshold shifts, the mechanisms of cochlear damage were different. Noise-induced hearing loss was mainly related to injuries of the stereocilia, whereas styrene-induced hearing loss was related to outer hair cell losses. Following the combined exposure, the threshold elevations as well as the cell losses exceeded the summed loss caused by noise and by styrene alone in the range of 8^16 kHz. Therefore, these results suggest that the two ototoxicants can cause a permanent synergistic loss of auditory sensitivity. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Styrene; Noise; Synergism; Rat
1. Introduction Styrene is an organic solvent widely used in the manufacturing of reinforced plastics, resins, synthetic rubbers and insulating materials. The highest occupational exposures to styrene occur during the production of glass-reinforced polyester products, especially large items such as boats that involve manual lay-up and spray-up operations. During this operation, the peak styrene exposure concentrations may be higher than 100 ppm in the industry. For work environments, the standards are 50 ppm averaged over an 8 h working day, with a peak level of 200 ppm. It is common that workers are exposed to styrene fumes in an environment where noise pollution is also common (Miller et al., 1994; Morata et al., 1994). Hu-
* Corresponding author. Fax: +33 3-83-50-21-85; E-mail: [email protected]
man studies have shown that chronic exposure to noise (Borg et al., 1995) or styrene alone (Mo«ller et al., 1990; Calebrese et al., 1996) can cause auditory de¢cits in workers. Similarly, several animal experiments have shown that noise exposure can induce cochlea damage such as stereocilia injuries or hair cell losses (Liberman and Dodds, 1987 ; Liberman, 1987; Gao et al., 1992), and that styrene can also disrupt auditory function. Yano et al. (1992) showed styrene-induced hearing losses using electrophysiological techniques, and Crofton et al. (1994) used behavioral techniques to perform their experiments with solvents. Both Yano et al. (1992) and Crofton et al. (1994) showed a mid-frequency hearing de¢cit induced by styrene in rats. Although both animal and human studies have examined the e¡ects of noise or styrene, few studies have focused on the e¡ects of a combined exposure to solvent and noise on hearing (see Johnson and Nyle¨n, 1995 ; Cary et al., 1997 for reviews). Brie£y, ototraumatic interaction continues to be a subject of controv-
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 9 9 ) 0 0 1 7 4 - 4
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
ersy in humans. Several recent studies have suggested a synergistic interaction between noise and solvents (Barrega®rd and Axelsson, 1984; Morata et al., 1994), whereas Jacobsen et al. (1993) and Sass-Kortsak et al. (1995) reported that the e¡ects of noise dominated. In the rat, Johnson et al. (1988) showed that sequential exposures to toluene followed by noise produced hearing losses which were greater than the summated loss caused by toluene and noise alone. Such ¢ndings were con¢rmed later by Lataye and Campo (1997) who exposed rats to toluene and noise simultaneously. To the best of our knowledge, only Fechter (1993) reported the e¡ects of an acute exposure of inhaled styrene (500 ppm for 7 h) combined with a 95 dB(A) simultaneous noise exposure in the guinea pig. He did not ¢nd an acute cochlear e¡ect due to styrene alone, nor an enhancement of the noise-induced hearing loss by the solvent. However, according to the author, the protocol was inappropriate for detecting such e¡ects. Because of the controversy in the literature regarding noise-styrene interactions, and the prevalence of both types of exposure, the present investigation was designed to study the e¡ects of simultaneous exposure to noise and/or to inhaled styrene on hearing in adult Long-Evans rats. To achieve this purpose, auditory function was tested by recording the near ¢eld auditory evoked potentials from the inferior colliculus, whereas the histopathological analyses of the cochleae were performed using conventional techniques.
87
by Campo et al. (1997) and will only be brie£y summarized here. After deep anesthesia was induced with a mixture of ketamine (50 mg/kg) and xylazine (3 mg/kg), rats were placed on a stereotaxic table in order to implant a chronic recording tungsten electrode in the right inferior colliculus (AP = 39.3 mm; L = 1.5 mm ; D = 4.2 mm). A second electrode was implanted in the rostral cranium just below the dura mater to serve as the ground electrode. These two electrodes were then fastened to a transistor socket and ¢xed with dental cement to the skull. The animals were allowed to recover for 4 weeks before any testing. One month after surgery, audiometric testing was performed in an audiometric room (Amplisilence0 G11) on awake rats placed in a restraining device. Typical recordings of inferior colliculus potential (ICP) responses from normal rats were depicted in a previous paper (Campo et al., 1997). The acoustic stimuli were generated by a Medelec ISD 45 and transduced by a speaker (JBL, 2405) positioned 15 cm from the pinna of the left ear. The acoustic stimuli were ¢ltered clicks (2 cycles for the rise/fall ramp, 4 cycles for the plateau) at frequencies of 2^32 kHz, presented at a rate of 20/s. After ampli¢cation (104 ) and ¢ltering between 32 and 8000 Hz, the ICPs were fed into a signal averager (n = 256) in order to determine the auditory thresholds (using an analysis window of 30 ms). A response amplitude of 20 WV (trough-to-peak) was considered the threshold value in our experimental conditions. 2.3. Noise exposure
2. Materials and methods 2.1. Animals Male Long-Evans rats (450^500 g) were purchased from Janvier Laboratories in France. The animals were housed for 1 month before the start of the experiments in individual cages (350U180U184 mm) with steam-cleaned pinewood bedding. Food (UAR Cie. France, ref.: A04 10) and tap water was available ad libitum except during the exposure period. Light (£uorescent lighting) was on from 7.00 h to 19.00 h. The temperature in animal quarters was 22 þ 1³C and the relative humidity ranged from 50 to 55%. While conducting the research described in this article, the investigators adhered to the Guide for Care and Use of Laboratory Animals, as promulgated by the French Conseil d'Etat through the De¨cret No. 87-848 published in the French Journal O¤ciel on 20 October 1987. 2.2. Audiometry The animal preparation has been previously detailed
The animals were exposed to noise inside the inhalation chambers. They were housed alone in individual cages with a speaker above the cages. The rats were exposed to a 97 dB ( þ 1 dB) octave band noise centered at 8 kHz for 6 h/day, 5 days/week, for 4 consecutive weeks. The exposure spectrum was chosen in order to cause hearing losses where auditory sensitivity is the highest. The noise level inside the chambers did not exceed 66 dB SPL for animals not exposed to noise. 2.4. Styrene exposure Animals were exposed to 750 ppm styrene (SigmaAldrich, 99%) with the same schedule as for the noise, namely 6 h/day, 5 days/week, for 4 consecutive weeks in inhalation chambers. The chambers contained up to eight animals where each animal was housed in an individual cage. Designed to sustain dynamic and adjustable air£ow (10^20 m3 /h), the chambers (200 l) were maintained at a negative pressure of no more than 3 mm H2 O. Input air was ¢ltered and conditioned to a temperature of 22^24³C and a relative humidity of 50^ 55%. Styrene was vaporized by bubbling an additional
HEARES 3344 7-12-99
88
R. Lataye et al. / Hearing Research 139 (2000) 86^96
air£ow through a £ask containing the test compound. The solvent concentration in the chambers was continuously monitored using a gas chromatograph. 2.5. Experimental design The rats (n = 64) were divided into four groups of 16 animals. Group S was exposed to styrene alone (747 þ 22 ppm), group N was exposed only to noise (97 þ 1 dB) and group S+N was exposed both to noise (97 þ 1 dB) and styrene (753 þ 27 ppm). One unexposed group (group C) was used as control. Audiometric thresholds were determined once prior to exposure (T1), the day following the end of exposure (T2) and after a 6 week recovery period (T3). The compound threshold shift (CTS = T23T1) and the permanent threshold shift (PTS = T33T1) were calculated for each animal. 2.6. Histology The rats were killed right after the last audiometry (T3), i.e. 6 weeks post exposure. The animals (7^8 months old) were ¢rst injected with a lethal dose of ketamine (75 mg/kg), and then ¢xed by intracardiac perfusion with 400 ml of a mixture of 4% paraformaldehyde and 3% glutaraldehyde in a trihydrate solution of sodium cacodylate (0.1 M, pH 7.4). The temporal bones were then removed, the tympanic bullae opened and the cochleae ¢xed again by perilymphatic perfusion with 3% glutaraldehyde solution. Following primary 24 h ¢xation, the cochleae were post¢xed with OsO4 1% in 0.1 M cacodylate bu¡er (pH 7.4) for 1 h and ¢nally washed in a trihydrate solution of sodium cacodylate. After dehydration in graded concentrations of ethanol (EtOH) from 30 to 70%, the cochleae were drilled to obtain a thin layer of cochlear bone. 2.6.1. Surface preparation The left cochleae (n = 8 per group) were dissected in 70% EtOH and the organ of Corti (including hook) was mounted in glycerin on microscope slides. The number of outer (OHC) and inner hair cells (IHC) was determined based on the presence of either the stereocilia, the cuticular plate or the cell nucleus. A cytocochleogram showing the percentage of hair cell loss as a function of distance from the base of the cochlea was plotted for each animal. The frequency-place map established by Mu«ller (1991) was used to superimpose the frequency coordinates on the length coordinates of the organ of Corti. 2.6.2. Scanning electron microscopy The right cochleae obtained from each group were further dissected in order to remove the bony capsule,
the spiral ligament, the stria vascularis, and Reissner's and the tectorial membranes. The exposed organ of Corti (OC) was dehydrated in ascending concentrations of EtOH up to 100%. Next, the samples were immersed in hexamethyldisilazane (Merck, 804324) and placed in a vacuum drying chamber overnight. The dried specimens were mounted on brass stubs using Eukitt0 with conductive silver paint and ¢nally sputter-coated with gold. The sputter coater (Valzers sputtering device) used with a current of 15 mA for 3 min allowed us to cover the specimen with a 25 nm deep gold coat. The tissues were viewed on a Jeol 840A scanning electron microscope. 2.7. Statistical analysis The interaction between noise and styrene was tested by running a linear model. The ¢xed e¡ects studied in the present investigation are those caused by either the styrene or the noise exposure or their interaction nested within frequency [styrene(freq.)+noise(freq.)+i[styrene, noise](freq.). In the model equation i is de¢ned as the interaction coe¤cient. A mixed procedure from SAS (Institute Inc., SAS/SAT0 software : Changes and Enhancements, Release 6.07. SAS Institute Inc., Cary, NC, 1992) was used. This procedure takes into account correlations between frequencies and variance heterogeneity between groups as well (covariance structure). The recovery, or in other terms the (CTS3PTS), was statistically tested using Student's t-test on paired samples at particular audiometric frequencies. Con¢dence levels for the tests and for the con¢dence intervals were 95%. P values are reported in Section 3. 3. Results 3.1. Hearing loss 3.1.1. Noise-induced hearing loss Fig. 1A shows signi¢cant CTS and PTS [F(12,656) = 22.31, P = 0.0001] in the noise-exposed group compared to the unexposed controls. Noise-induced hearing loss (NIHL) was located in a broad frequency range (8^20 kHz) with the peak of the loss occurring near 12 kHz. The maximum hearing loss was therefore positioned approximately one-half octave (11.3 kHz) above the center frequency of the exposure (8 kHz). The CTS and PTS at 10 kHz were 30.6 and 15.3 dB, respectively, at 12 kHz they were 30.8 and 18.3 dB, and at 16 kHz CTS and PTS were 25.4 and 16.1 dB. Signi¢cant recovery from CTS was observed from 4 (P = 0.002) to 24 kHz (P = 0.038). For instance, the values at 8, 10, 12 and 16 kHz were respectively 12.4, 15.3, 12.5 and 9.3 dB, respectively.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
89
Fig. 1. A: CTS and PTS (n = 16) versus frequencies obtained from noise-exposed animals. Noise exposure: octave band noise centered at 8 kHz at 97 dB SPL, 6 h/day, 5 days/week, 4 weeks. B: Obtained from styrene-treated animals. Styrene exposure: 750 ppm, 6 h/day, 5 days/week, 4 weeks. C: Obtained from animals exposed to both noise and styrene. The bars represent the 95% con¢dence intervals.
3.1.2. Styrene-induced hearing loss Fig. 1B shows signi¢cant CTS and PTS [F(12,656) = 4.34, P = 0.0001] in the styrene-exposed group compared to the unexposed controls. Styrene-induced hearing loss (SIHL) was located in the region of 16^20 kHz, i.e. in a narrower frequency range than that obtained with noise alone. The values for CTS and PTS at 16 kHz were 9.7 and 7.1 dB respectively, whereas at
20 kHz the values were 13.5 and 9.2 dB. The CTS were signi¢cantly di¡erent from the PTS, or in other terms the recovery from CTS was signi¢cant from 2 (P = 0.006) to 20 kHz (P = 0.002). The recovery at 24 and 32 kHz was not signi¢cant. 3.1.3. Styrene- and noise-induced hearing loss In Fig. 1C, signi¢cant CTS and PTS [F(12,656) =
HEARES 3344 7-12-99
90
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 1 (continued).
3.66, P = 0.0001] were obtained in the styrene- andnoise group compared with the unexposed controls. Styrene- and noise-induced hearing loss (SNIHL) was located in a frequency range (8^20 kHz) similar to that obtained with noise alone. As with the noise alone, the peak of the loss was positioned at 12 kHz. However, the PTS at 10, 12 and 16 kHz were 21.9, 29.0 and 26.2 dB, respectively, signi¢cantly larger than the values of 15.3, 18.3, and 16.1 dB observed for the noise alone group. In addition to exceeding the PTS in the region of maximal hearing loss, the spectrum of hearing loss was broader than that observed with noise alone, as it extended to 8 and 20 kHz in the combined condition. Signi¢cant recovery from CTS was observed over the
frequency range from 2 (P = 0.034) to 24 kHz (P = 0.014). The amount of recovery varied from 2.3 dB at 2 kHz to 14.8 dB at 8 kHz. 3.1.4. Interaction PTS values obtained in the three experimental conditions are summarized in Fig. 2. In this ¢gure, the arithmetic sum (in dB) was plotted to compare the predicted with the experimental losses. The combined exposures induced PTS values signi¢cantly larger [F(12,656) = 3.66, P = 0.0001] than the arithmetic sum (in dB) of NIHL+SIHL over the frequency range from 6 to 12 kHz. The signi¢cance and the statistical values for all audiometric frequency are detailed in Ta-
TBL1 Interaction statistics of the e¡ects of noise and styrene as a function of frequency Frequency (kHz)
Interaction coe¤cient (i)
S.D.
t
P
Lower limit
Upper limit
2 3 4 5 6 8 10 12 16 20 24 32
1.82 3.62 2.01 1.34 6.76 9.58 6.14 10.08 2.78 0.72 32.62 34.13
2.17 2.09 2.14 2.10 2.68 2.61 3.64 3.60 3.83 4.04 1.88 1.55
0.84 1.72 0.94 0.64 2.52 3.68 1.69 2.80 0.72 0.18 31.39 32.66
0.40 0.08 0.35 0.52 0.01 0.00 0.09 0.00 0.47 0.86 0.16 0.01
32.45 30.5 32.19 32.78 1.49 4.46 31.01 3.01 34.74 37.21 36.32 37.18
6.08 7.74 6.21 5.47 12.02 14.70 13.29 17.16 10.30 8.66 1.07 31.08
The interaction coe¤cient is de¢ned as i = (PTScombined 3PTScontrol )3[(PTSnoise 3PTScontrol )+(PTSctyrene 3PTScontrol )]; t = i/S.D.; P = P (dtd 6 dTQ d), TQ is the Student variable with Q degrees of freedom; limits are i þ TQ;0:025 US.D.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
91
Fig. 2. PTS versus frequencies obtained after 750 ppm styrene exposure, 97 dB octave band noise centered at 8 kHz and simultaneous noise and styrene exposures for 6 h/day, 5 days/week, 4 weeks. The arithmetic sum of the e¡ects of noise and styrene was also plotted. The bars represent the 95% con¢dence intervals.
ble 1. Below and above these frequencies, the hearing loss induced by the combined exposure was quite similar to the summated losses obtained by each toxic agent alone. 3.2. Morphological results 3.2.1. Light microscopy The average cytocochleogram of eight animals belonging to group C is £at for all the rows of hair cells (OHC1^3) and is therefore not depicted. Fig. 3A shows the averaged (n = 8) cytocochleogram of group N. For OHC1, approximately 12% were missing at 17 kHz frequency place. The two other rows had only slight damage: 3% for OHC2 and 5% for OHC3. Fig. 3B illustrates the cytocochleogram (n = 8) of group S. The largest losses are located at OHC3, 86% at 20 kHz and 70% at 4 kHz. OHC2 is less damaged (25% at 20 kHz and 36% at 4 kHz) than OHC3, but more than OHC1 (15% at 20 kHz and 17% at 4 kHz). Frequencies above 30 kHz seem to be well preserved for each row of OHCs. Similarly, IHC were well preserved. Fig. 3C depicts the cytocochleogram of group S+N. As for group S, the order of the trauma can be characterized as follows : missing OHC3 s OHC2 s OHC1 in terms of missing cells. However, OHC2^3 show larger losses in the vicinity of 20 kHz (OHC3: 94%; OHC2 : 42%) and 4 kHz (OHC3: 86% ; OHC3: 62%) compared to those obtained with group S.
3.2.2. Scanning electron micrography All the scanning electron micrographs (SEM) are from the second turn and correspond to the 16 kHz region. Fig. 4 (#3491 ; U2000) is a SEM of the OC from a noise-exposed rat. OHC1 is more severely damaged than OHC2 and OHC3. Three phalangeal scars (butter£y-like shape) and fusion of stereocilia can be observed at the level of OHC1, whereas splayed stereocilia were observed on IHC. Splayed stereocilia could be observed at the most damage frequencies, from 8 to 20 kHz. Fig. 5 (#3425 ; U2000) is a SEM of the OC from a styrene-treated rat. Only ¢ve hair cells remain in row 3, whereas OHC1^2 are well preserved. Scars have replaced missing hair cells in OHC3. Note the concavity of the scar formation in the reticular lamina at the level of OHC3. The inner hair cells do not seem to be injured. Fig. 6 (#3499, U2000) is a SEM of the OC from a rat exposed to both noise and styrene. The loss of outer hair cells is large even in the second row. Similar to that observed following exposure to noise alone, scar formation (butter£y-like shape) replaced missing OHC1. The third row shows the most damage with only one hair cell remaining. Two concavities can be observed similar to those observed in animals exposed only to styrene (Fig. 5). Splayed stereocilia were observed on the IHC. As for the noise exposure, splayed stereocilia could be observed at the most damaged frequencies, from 8 to 20 kHz.
HEARES 3344 7-12-99
92
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 3. A: Average cytocochleogram (n = 8) obtained from rats exposed to a 97 dB octave band noise centered at 8 kHz for 6 h/day, 5 days/ week, 4 weeks. B: Obtained from rats exposed to 750 ppm styrene for 6 h/day, 5 days/week, 4 weeks. C: Obtained from rats exposed to both noise and styrene. Abscissa, upper trace: length (mm) of the entire spiral course of the organ of Corti from the bottom to the hook; lower trace: frequency map according to Mu«ller (1991). Ordinate: hair cell loss in percent. IHC: inner hair cells; OHC1: ¢rst row of outer hair cells; OHC2: second row; OHC3: third row.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
93
Fig. 4. Scanning electron micrograph (#3491, U2000) of the 16 kHz region of the OC from a noise-exposed rat. Three phalangeal scars (stars) and fusion of stereocilia (black arrows) can be observed at the level of OHC1. Splayed stereocilia (arrowheads) are noted among the stereociliary bundles of IHC.
4. Discussion Following simultaneous exposure to noise and styrene, the auditory threshold shifts were increased compared to auditory de¢cits caused either by noise or by styrene alone (Fig. 1C). Indeed, both CTS and PTS were signi¢cantly greater (38.4 and 29 dB, respectively) than those obtained following noise exposure alone at the peak of hearing loss (12 kHz). Similarly, the cochleae of the combination group showed a large increase in the number of missing cells compared to either the noise alone or the styrene alone group (Fig. 3B,C). The pattern of damage (order and location of the trauma) in the combined exposures appears more like that observed with styrene than with noise. Thus, both the electrophysiological and morphological results support an interaction between noise and styrene on hearing loss and cochlear damage in the rat cochlea : frequencies and the order of the trauma look like those obtained with styrene. It is useful to clarify the terminol-
ogy related to the interaction concept, especially synergism and additivity. As reported by Nyle¨n (1994) and Fechter (1995), additivity is obtained when the toxic e¡ect of the combined exposures can be predicted by the arithmetic sum of the e¡ects observed by each individual exposure. As for the concept of synergism, the toxic e¡ect of the combined exposures must be greater than that expected by the sum of the e¡ects observed by each individual exposure. The authors will refer to this terminology in the present study. Fig. 2 illustrated the overall interactions between the two agents on hearing. There was a signi¢cant synergy between the e¡ects of both agents within the range 6^12 kHz (Table 1), as the PTS values induced by the combined exposure exceeded the summed losses caused by styrene alone and noise alone. A simple additivity was noted at and above 16 kHz. As with the noise exposure, the styrene exposure caused a long-lasting threshold shift in auditory sensitivity in the rat. This ¢nding con¢rmed the results of
HEARES 3344 7-12-99
94
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 5. Scanning electron micrograph (#3425; U2000) of the OC from a styrene-treated rat. Note that concavities in the reticular membrane replaced missing OHC3 (arrowheads).
previous studies with this solvent (Crofton et al., 1994). After a 6 week post-exposure period, slight (4.3 dB at 20 kHz) but signi¢cant recovery was observed over a large frequency range from 2 to 20 kHz. This result has never been reported previously. The maximum rate of metabolism (Vmax ) might explain, at least partly, the recovery measured in the present study. Indeed, according to Filser et al. (1993), Vmax is close to 700 ppm in rats. Thus, one can consider that the rat is capable of rapidly metabolizing accumulated styrene in tissues after 4 weeks of 750 ppm exposure, decreasing the risk of injury to the hair cells. PTS also con¢rmed that styrene caused a selective mid-frequency (16^20 kHz) hearing de¢cit. This mid-frequency susceptibility to styrene was consistent with data obtained from other ototoxic solvents such as toluene (Campo et al., 1997) or trichloroethylene (Crofton et al., 1994). As for the histological data, two major permanent morphological changes were observed in the noise-exposed cochleae: hair cell losses and stereocilia changes.
The missing hair cells reported in an averaged cytocochleogram (Fig. 3A) showed that OHC1 were more susceptible to acoustic injury than the two other rows of OHCs. However, the use of surface preparations underestimated the damage of the organ of Corti, since no missing hair cells were counted at 10 kHz for instance, whereas a 16 dB PTS was measured by electrophysiology. Using SEM, the detailed observations of the organ of Corti showed numerous clumped and fused stereocilia on OHC1 and splayed stereocilia on IHCs (Fig. 4) at the most damaged frequencies (8^20 kHz). Thus, in our experimental conditions, the acoustic trauma was characterized less by the lack of cells than by injured stereocilia. These aspects of stereocilia injuries may result from excessive motion of the basilar membrane which caused collisions among stereocilia. Thus, the order of hair cell damage (OHC1 s IHC s OHC2 and OHC3) caused by the 97 dB noise exposure was in agreement with the progression established in the guinea pig earlier by Robertson and Johnstone
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
95
Fig. 6. Scanning electron micrograph (#3499; U2000) of the OC from a rat exposed to both noise and styrene. Phalangeal scars (stars) and concavities (large arrowhead) replaced respectively missing OHC1 and OHC3. Splayed stereocilia can be observed in the IHC row (small arrowheads).
(1980). Therefore, the electrophysiological and histological data clearly demonstrate that our experimental conditions produced a typical acoustic injury based on a mechanical process. From a morphological point of view, the lesions of the organ of Corti from styrene-exposed animals were very di¡erent from those in noise-exposed animals. For instance, severe and progressive OHC losses (OHC3 s OHC2 s OHC1) were detected (Fig. 3B), whereas no obvious stereocilia damage was noted (Fig. 5). In contrast, OHC losses were not only located in the mid-frequency area (as reported by Yano et al., 1992) but also in the mid- to low-frequency area (4 kHz). In fact, this discrepancy between morphological and electrophysiological ¢ndings had already been discussed regarding toluene (Campo et al., 1997). Brie£y, that might be due to the enhancement of the low-frequency amplitude responses by the inferior colliculus following trauma (Salvi et al., 1990), and/or to the speci¢c sensitivity of the apical turn of the rat cochlea (Prosen and Moody, 1991). Although the exact mechanism of sty-
rene toxicity is not understood, it is likely that styrene impairs preferentially the basal pole of OHC and/or the supporting cells by tissue contamination. Indeed, a possible route to reach the OHC is the lipid-rich content of the membranes of the di¡erent cells of the organ of Corti. The styrene could di¡use through the outer sulcus and reach the lipid-rich Hensen's cells which are connected with the Deiters' cells (Campo et al., 1999). Based on these results, the e¡ect of styrene is therefore di¡erent from that of noise and is mainly caused by a chemical process. Based on the ¢ndings reported above, it is realistic to think that the damage caused by combined exposure is generated by the same mechanisms (chemical or mechanical) as that described previously for each individual agent. In other terms, the coexistence of both mechanisms does not seem to generate a new ototraumatic mechanism. As a result, the synergy could be the result of a potentiation of one agent by the other. The potentiation could be the result of two phenomena: ¢rst, a weakening of the OHC membranes induced by the sol-
HEARES 3344 7-12-99
96
R. Lataye et al. / Hearing Research 139 (2000) 86^96
vent which might increase the micro-lesions due to noise exposure (Mulroy et al., 1998) ; second, a cell poisoning of the organ of Corti due to the solvent, which would reduce the availability of necessary energy during noise exposure. Given the risks encountered by subjects exposed to solvents and noise, it is reasonable to wonder if the solvent as well as the noise threshold limit values are really adequate in multiple occupational exposures. 5. Conclusion Rats exposed simultaneously to styrene and noise su¡er more hearing loss and greater cochlear damage than those caused by each agent alone. This synergistic e¡ect is primarily limited to the spectrum of the noise. The damage to the ear caused by noise and styrene occurs due to two di¡erent mechanisms (chemical versus mechanical). Acknowledgements The authors wish to thank Dr. F. Boettcher (Medical University of South Carolina, Charleston, SC, USA) and Dr. T. Morata (NIOSH, USA) for their comments on an earlier draft of the manuscript. They also wish to thank C. Barthe¨le¨my, M. Roure, P. Bonnet and J.C. Vigneron (INRS, FRANCE) for their technical assistance. M. Grzebyk (INRS) is acknowledged for the statistical analyses. References Barrega®rd, L., Axelsson, A., 1984. Is there an ototraumatic interaction between noise and solvents? Scand. Audiol. 13, 151^155. Borg, E., Canlon, B., Engstro«m, B., 1995. Noise-induced hearing loss. Scand. Audiol. 24 (Suppl. 40). Calebrese, G., Martini, A., Sessa, G., Cellini, M., Bartolucci, G.B., Marcuzzo, G., De Rosa, E., 1996. Otoneurological study in workers exposed to styrene in the ¢berglass industry. Int. Arch. Occup. Environ. Health 68, 219^223. Campo, P., Lataye, R., Cossec, B., Placidi, V., 1997. Toluene-induced hearing loss: a mid-frequency location of the cochlear lesions. Neurotoxicol. Teratol. 19, 129^140. Campo, P., Lataye, R., Loquet, G.M., 1999. Toluene and styreneinduced hearing losses: a comparative study. In: Prasher, D., Canlon, B. (Eds.), Cochlear Pharmacology and Noise Trauma. NRN, London, pp. 113^124. Cary, R., Clarke, S., Delic, J., 1997. E¡ects of combined exposure to noise and toxic substances ^ critical review of the literature. Ann. Occup. Hyg. 41, 455^465. Crofton, K.M., Lassiter, T., Rebert, C., 1994. Solvent induced ototoxicity in rats: An atypical selective mid-frequency hearing de¢cit. Hear. Res. 80, 25^30.
Fechter, L.D., 1993. E¡ects of acute styrene and simultaneous noise exposure on auditory function in the guinea pig. Neurotoxicol. Teratol. 15, 151^155. Fechter, L.D., 1995. Combined e¡ects of noise and chemicals. Occup. Med. State Art Rev. 10, 609^621. Filser, J.G., Schwegler, U., Csana¨dy, G.A., Greim, H., Kreuzer, P.E., Kessler, W., 1993. Species-speci¢c pharmacokinetics of styrene in rat and mouse. Arch. Toxicol. 67, 517^530. Gao, W.Y., Ding, D.I., Zheng, X.Y., Run, F.M., Liu, Y.J., 1992. A comparison of changes in the stereocilia between temporary and permanent hearing losses in acoustic trauma. Hear. Res. 62, 27^41. Jacobsen, P., Hein, H.O., Suadicani, P., Parving, A., Gyntelberg, F., 1993. Mixed solvent exposure and hearing impairment: an epidemiological study of 3284 men. The Copenhagen male study. Occup. Med. 43, 180^184. Johnson, A.C., Juntunen, L., Nyle¨n, P., Borg, E., Ho«glund, G., 1988. E¡ect of interaction between noise and toluene on auditory function in the rat. Acta Otolaryngol. (Stockh.) 105, 56^63. Johnson, A.C., Nyle¨n, P., 1995. E¡ects of industrial solvents on hearing. Occup. Med. State Art Rev. 10, 623^640. Lataye, R., Campo, P., 1997. Combined e¡ects of a simultaneous exposure to noise and toluene on hearing function. Neurotoxicol. Teratol. 19, 373^382. Liberman, M.C., 1987. Chronic ultrastructural changes in acoustic trauma: serial section reconstruction of stereocilia and cuticular plates. Hear. Res. 26, 65^88. Liberman, M.C., Dodds, L.W., 1987. Acute ultrastructural changes in acoustic trauma: Serial section reconstruction of stereocilia and cuticular plates. Hear. Res. 26, 45^64. Miller, R.R., Newhook, R., Poole, A., 1994. Styrene production, use, and human exposure. Crit. Rev. Toxicol. 24 (Suppl. 1), S1^S10. Mo«ller, C., Odkvist, L., Larsby, B., Tham, R., Ledin, T., Bergholtz, L., 1990. Otoneurological ¢ndings in workers exposed to styrene. Scand. J. Work Environ. Health 16, 189^194. Morata, T.C., Dunn, D.E., Sieber, W.K., 1994. Occupational exposure to noise and ototoxic organic solvents. Arch. Environ. Health 49, 359^365. Mu«ller, M., 1991. Frequency representation in the rat cochlea. Hear. Res. 51, 247^254. Mulroy, M., Henry, W.R., McNeil, P.L., 1998. Noise-induced transient microlesions in the cell membranes of auditory hair cells. Hear. Res. 115, 93^100. Nyle¨n, P., 1994. Organic solvent toxicity in the rat; with emphasis on combined exposure interactions in the nervous system. Arbete Ha«lsa 3, 1^50. Prosen, C.A., Moody, D.B., 1991. Low-frequency detection and discrimination following apical hair cell destruction. Hear. Res. 57, 142^152. Robertson, D., Johnstone, B.M., 1980. Acoustic trauma in the guinea pig cochlea: early changes in ultrastructure and neural threshold. Hear. Res. 3, 167^179. Salvi, R., Saunders, S., Gratton, M., Arehole, S., Powers, N., 1990. Enhanced evoked response amplitude in the inferior colliculus of the chinchilla following acoustic trauma. Hear. Res. 50, 245^ 258. Sass-Kortsak, A.M., Corey, P.N., Robertson, J.M., 1995. An investigation of the association between exposure to styrene and hearing loss. Ann. Epidemiol. 5, 15^24. Yano, B.L., Dittenber, D.A., Albee, R.R., Mattsson, J.L., 1992. Abnormal auditory brain stem responses and cochlear pathology in rats induced by an exaggerated styrene exposure regimen. Toxicol. Pathol. 20, 1^6.
HEARES 3344 7-12-99
Combined e¡ects of noise and styrene exposure on hearing function in the rat R. Lataye, P. Campo *, G. Loquet Institut National de Recherche et de Se¨curite¨, Laboratoire de Neurotoxicologie et Immunotoxicologie, Avenue de Bourgogne, P.O. Box 27, 54501 Vandoeuvre, France Received 23 January 1999; received in revised form 30 August 1999; accepted 9 September 1999
Abstract Combined exposure to both noise and aromatic solvents such as styrene is common in many industries. In order to study the combined effects of simultaneous exposure to both noise and styrene on hearing, male adult Long-Evans rats were exposed either to 750 ppm styrene alone, to a 97 dB SPL octave band of noise centered at 8 kHz, or to a combination of noise and styrene. The exposure duration was 6 h/day, 5 days/week, for 4 consecutive weeks. Auditory function was tested over a frequency range from 2 to 32 kHz by recording near field potentials from the inferior colliculus, whereas histopathological analyses of the cochleae were performed with conventional morphometric approaches. Whereas both noise and styrene each caused permanent threshold shifts, the mechanisms of cochlear damage were different. Noise-induced hearing loss was mainly related to injuries of the stereocilia, whereas styrene-induced hearing loss was related to outer hair cell losses. Following the combined exposure, the threshold elevations as well as the cell losses exceeded the summed loss caused by noise and by styrene alone in the range of 8^16 kHz. Therefore, these results suggest that the two ototoxicants can cause a permanent synergistic loss of auditory sensitivity. ß 2000 Elsevier Science B.V. All rights reserved. Key words: Styrene; Noise; Synergism; Rat
1. Introduction Styrene is an organic solvent widely used in the manufacturing of reinforced plastics, resins, synthetic rubbers and insulating materials. The highest occupational exposures to styrene occur during the production of glass-reinforced polyester products, especially large items such as boats that involve manual lay-up and spray-up operations. During this operation, the peak styrene exposure concentrations may be higher than 100 ppm in the industry. For work environments, the standards are 50 ppm averaged over an 8 h working day, with a peak level of 200 ppm. It is common that workers are exposed to styrene fumes in an environment where noise pollution is also common (Miller et al., 1994; Morata et al., 1994). Hu-
* Corresponding author. Fax: +33 3-83-50-21-85; E-mail: [email protected]
man studies have shown that chronic exposure to noise (Borg et al., 1995) or styrene alone (Mo«ller et al., 1990; Calebrese et al., 1996) can cause auditory de¢cits in workers. Similarly, several animal experiments have shown that noise exposure can induce cochlea damage such as stereocilia injuries or hair cell losses (Liberman and Dodds, 1987 ; Liberman, 1987; Gao et al., 1992), and that styrene can also disrupt auditory function. Yano et al. (1992) showed styrene-induced hearing losses using electrophysiological techniques, and Crofton et al. (1994) used behavioral techniques to perform their experiments with solvents. Both Yano et al. (1992) and Crofton et al. (1994) showed a mid-frequency hearing de¢cit induced by styrene in rats. Although both animal and human studies have examined the e¡ects of noise or styrene, few studies have focused on the e¡ects of a combined exposure to solvent and noise on hearing (see Johnson and Nyle¨n, 1995 ; Cary et al., 1997 for reviews). Brie£y, ototraumatic interaction continues to be a subject of controv-
0378-5955 / 00 / $ ^ see front matter ß 2000 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 5 9 5 5 ( 9 9 ) 0 0 1 7 4 - 4
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
ersy in humans. Several recent studies have suggested a synergistic interaction between noise and solvents (Barrega®rd and Axelsson, 1984; Morata et al., 1994), whereas Jacobsen et al. (1993) and Sass-Kortsak et al. (1995) reported that the e¡ects of noise dominated. In the rat, Johnson et al. (1988) showed that sequential exposures to toluene followed by noise produced hearing losses which were greater than the summated loss caused by toluene and noise alone. Such ¢ndings were con¢rmed later by Lataye and Campo (1997) who exposed rats to toluene and noise simultaneously. To the best of our knowledge, only Fechter (1993) reported the e¡ects of an acute exposure of inhaled styrene (500 ppm for 7 h) combined with a 95 dB(A) simultaneous noise exposure in the guinea pig. He did not ¢nd an acute cochlear e¡ect due to styrene alone, nor an enhancement of the noise-induced hearing loss by the solvent. However, according to the author, the protocol was inappropriate for detecting such e¡ects. Because of the controversy in the literature regarding noise-styrene interactions, and the prevalence of both types of exposure, the present investigation was designed to study the e¡ects of simultaneous exposure to noise and/or to inhaled styrene on hearing in adult Long-Evans rats. To achieve this purpose, auditory function was tested by recording the near ¢eld auditory evoked potentials from the inferior colliculus, whereas the histopathological analyses of the cochleae were performed using conventional techniques.
87
by Campo et al. (1997) and will only be brie£y summarized here. After deep anesthesia was induced with a mixture of ketamine (50 mg/kg) and xylazine (3 mg/kg), rats were placed on a stereotaxic table in order to implant a chronic recording tungsten electrode in the right inferior colliculus (AP = 39.3 mm; L = 1.5 mm ; D = 4.2 mm). A second electrode was implanted in the rostral cranium just below the dura mater to serve as the ground electrode. These two electrodes were then fastened to a transistor socket and ¢xed with dental cement to the skull. The animals were allowed to recover for 4 weeks before any testing. One month after surgery, audiometric testing was performed in an audiometric room (Amplisilence0 G11) on awake rats placed in a restraining device. Typical recordings of inferior colliculus potential (ICP) responses from normal rats were depicted in a previous paper (Campo et al., 1997). The acoustic stimuli were generated by a Medelec ISD 45 and transduced by a speaker (JBL, 2405) positioned 15 cm from the pinna of the left ear. The acoustic stimuli were ¢ltered clicks (2 cycles for the rise/fall ramp, 4 cycles for the plateau) at frequencies of 2^32 kHz, presented at a rate of 20/s. After ampli¢cation (104 ) and ¢ltering between 32 and 8000 Hz, the ICPs were fed into a signal averager (n = 256) in order to determine the auditory thresholds (using an analysis window of 30 ms). A response amplitude of 20 WV (trough-to-peak) was considered the threshold value in our experimental conditions. 2.3. Noise exposure
2. Materials and methods 2.1. Animals Male Long-Evans rats (450^500 g) were purchased from Janvier Laboratories in France. The animals were housed for 1 month before the start of the experiments in individual cages (350U180U184 mm) with steam-cleaned pinewood bedding. Food (UAR Cie. France, ref.: A04 10) and tap water was available ad libitum except during the exposure period. Light (£uorescent lighting) was on from 7.00 h to 19.00 h. The temperature in animal quarters was 22 þ 1³C and the relative humidity ranged from 50 to 55%. While conducting the research described in this article, the investigators adhered to the Guide for Care and Use of Laboratory Animals, as promulgated by the French Conseil d'Etat through the De¨cret No. 87-848 published in the French Journal O¤ciel on 20 October 1987. 2.2. Audiometry The animal preparation has been previously detailed
The animals were exposed to noise inside the inhalation chambers. They were housed alone in individual cages with a speaker above the cages. The rats were exposed to a 97 dB ( þ 1 dB) octave band noise centered at 8 kHz for 6 h/day, 5 days/week, for 4 consecutive weeks. The exposure spectrum was chosen in order to cause hearing losses where auditory sensitivity is the highest. The noise level inside the chambers did not exceed 66 dB SPL for animals not exposed to noise. 2.4. Styrene exposure Animals were exposed to 750 ppm styrene (SigmaAldrich, 99%) with the same schedule as for the noise, namely 6 h/day, 5 days/week, for 4 consecutive weeks in inhalation chambers. The chambers contained up to eight animals where each animal was housed in an individual cage. Designed to sustain dynamic and adjustable air£ow (10^20 m3 /h), the chambers (200 l) were maintained at a negative pressure of no more than 3 mm H2 O. Input air was ¢ltered and conditioned to a temperature of 22^24³C and a relative humidity of 50^ 55%. Styrene was vaporized by bubbling an additional
HEARES 3344 7-12-99
88
R. Lataye et al. / Hearing Research 139 (2000) 86^96
air£ow through a £ask containing the test compound. The solvent concentration in the chambers was continuously monitored using a gas chromatograph. 2.5. Experimental design The rats (n = 64) were divided into four groups of 16 animals. Group S was exposed to styrene alone (747 þ 22 ppm), group N was exposed only to noise (97 þ 1 dB) and group S+N was exposed both to noise (97 þ 1 dB) and styrene (753 þ 27 ppm). One unexposed group (group C) was used as control. Audiometric thresholds were determined once prior to exposure (T1), the day following the end of exposure (T2) and after a 6 week recovery period (T3). The compound threshold shift (CTS = T23T1) and the permanent threshold shift (PTS = T33T1) were calculated for each animal. 2.6. Histology The rats were killed right after the last audiometry (T3), i.e. 6 weeks post exposure. The animals (7^8 months old) were ¢rst injected with a lethal dose of ketamine (75 mg/kg), and then ¢xed by intracardiac perfusion with 400 ml of a mixture of 4% paraformaldehyde and 3% glutaraldehyde in a trihydrate solution of sodium cacodylate (0.1 M, pH 7.4). The temporal bones were then removed, the tympanic bullae opened and the cochleae ¢xed again by perilymphatic perfusion with 3% glutaraldehyde solution. Following primary 24 h ¢xation, the cochleae were post¢xed with OsO4 1% in 0.1 M cacodylate bu¡er (pH 7.4) for 1 h and ¢nally washed in a trihydrate solution of sodium cacodylate. After dehydration in graded concentrations of ethanol (EtOH) from 30 to 70%, the cochleae were drilled to obtain a thin layer of cochlear bone. 2.6.1. Surface preparation The left cochleae (n = 8 per group) were dissected in 70% EtOH and the organ of Corti (including hook) was mounted in glycerin on microscope slides. The number of outer (OHC) and inner hair cells (IHC) was determined based on the presence of either the stereocilia, the cuticular plate or the cell nucleus. A cytocochleogram showing the percentage of hair cell loss as a function of distance from the base of the cochlea was plotted for each animal. The frequency-place map established by Mu«ller (1991) was used to superimpose the frequency coordinates on the length coordinates of the organ of Corti. 2.6.2. Scanning electron microscopy The right cochleae obtained from each group were further dissected in order to remove the bony capsule,
the spiral ligament, the stria vascularis, and Reissner's and the tectorial membranes. The exposed organ of Corti (OC) was dehydrated in ascending concentrations of EtOH up to 100%. Next, the samples were immersed in hexamethyldisilazane (Merck, 804324) and placed in a vacuum drying chamber overnight. The dried specimens were mounted on brass stubs using Eukitt0 with conductive silver paint and ¢nally sputter-coated with gold. The sputter coater (Valzers sputtering device) used with a current of 15 mA for 3 min allowed us to cover the specimen with a 25 nm deep gold coat. The tissues were viewed on a Jeol 840A scanning electron microscope. 2.7. Statistical analysis The interaction between noise and styrene was tested by running a linear model. The ¢xed e¡ects studied in the present investigation are those caused by either the styrene or the noise exposure or their interaction nested within frequency [styrene(freq.)+noise(freq.)+i[styrene, noise](freq.). In the model equation i is de¢ned as the interaction coe¤cient. A mixed procedure from SAS (Institute Inc., SAS/SAT0 software : Changes and Enhancements, Release 6.07. SAS Institute Inc., Cary, NC, 1992) was used. This procedure takes into account correlations between frequencies and variance heterogeneity between groups as well (covariance structure). The recovery, or in other terms the (CTS3PTS), was statistically tested using Student's t-test on paired samples at particular audiometric frequencies. Con¢dence levels for the tests and for the con¢dence intervals were 95%. P values are reported in Section 3. 3. Results 3.1. Hearing loss 3.1.1. Noise-induced hearing loss Fig. 1A shows signi¢cant CTS and PTS [F(12,656) = 22.31, P = 0.0001] in the noise-exposed group compared to the unexposed controls. Noise-induced hearing loss (NIHL) was located in a broad frequency range (8^20 kHz) with the peak of the loss occurring near 12 kHz. The maximum hearing loss was therefore positioned approximately one-half octave (11.3 kHz) above the center frequency of the exposure (8 kHz). The CTS and PTS at 10 kHz were 30.6 and 15.3 dB, respectively, at 12 kHz they were 30.8 and 18.3 dB, and at 16 kHz CTS and PTS were 25.4 and 16.1 dB. Signi¢cant recovery from CTS was observed from 4 (P = 0.002) to 24 kHz (P = 0.038). For instance, the values at 8, 10, 12 and 16 kHz were respectively 12.4, 15.3, 12.5 and 9.3 dB, respectively.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
89
Fig. 1. A: CTS and PTS (n = 16) versus frequencies obtained from noise-exposed animals. Noise exposure: octave band noise centered at 8 kHz at 97 dB SPL, 6 h/day, 5 days/week, 4 weeks. B: Obtained from styrene-treated animals. Styrene exposure: 750 ppm, 6 h/day, 5 days/week, 4 weeks. C: Obtained from animals exposed to both noise and styrene. The bars represent the 95% con¢dence intervals.
3.1.2. Styrene-induced hearing loss Fig. 1B shows signi¢cant CTS and PTS [F(12,656) = 4.34, P = 0.0001] in the styrene-exposed group compared to the unexposed controls. Styrene-induced hearing loss (SIHL) was located in the region of 16^20 kHz, i.e. in a narrower frequency range than that obtained with noise alone. The values for CTS and PTS at 16 kHz were 9.7 and 7.1 dB respectively, whereas at
20 kHz the values were 13.5 and 9.2 dB. The CTS were signi¢cantly di¡erent from the PTS, or in other terms the recovery from CTS was signi¢cant from 2 (P = 0.006) to 20 kHz (P = 0.002). The recovery at 24 and 32 kHz was not signi¢cant. 3.1.3. Styrene- and noise-induced hearing loss In Fig. 1C, signi¢cant CTS and PTS [F(12,656) =
HEARES 3344 7-12-99
90
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 1 (continued).
3.66, P = 0.0001] were obtained in the styrene- andnoise group compared with the unexposed controls. Styrene- and noise-induced hearing loss (SNIHL) was located in a frequency range (8^20 kHz) similar to that obtained with noise alone. As with the noise alone, the peak of the loss was positioned at 12 kHz. However, the PTS at 10, 12 and 16 kHz were 21.9, 29.0 and 26.2 dB, respectively, signi¢cantly larger than the values of 15.3, 18.3, and 16.1 dB observed for the noise alone group. In addition to exceeding the PTS in the region of maximal hearing loss, the spectrum of hearing loss was broader than that observed with noise alone, as it extended to 8 and 20 kHz in the combined condition. Signi¢cant recovery from CTS was observed over the
frequency range from 2 (P = 0.034) to 24 kHz (P = 0.014). The amount of recovery varied from 2.3 dB at 2 kHz to 14.8 dB at 8 kHz. 3.1.4. Interaction PTS values obtained in the three experimental conditions are summarized in Fig. 2. In this ¢gure, the arithmetic sum (in dB) was plotted to compare the predicted with the experimental losses. The combined exposures induced PTS values signi¢cantly larger [F(12,656) = 3.66, P = 0.0001] than the arithmetic sum (in dB) of NIHL+SIHL over the frequency range from 6 to 12 kHz. The signi¢cance and the statistical values for all audiometric frequency are detailed in Ta-
TBL1 Interaction statistics of the e¡ects of noise and styrene as a function of frequency Frequency (kHz)
Interaction coe¤cient (i)
S.D.
t
P
Lower limit
Upper limit
2 3 4 5 6 8 10 12 16 20 24 32
1.82 3.62 2.01 1.34 6.76 9.58 6.14 10.08 2.78 0.72 32.62 34.13
2.17 2.09 2.14 2.10 2.68 2.61 3.64 3.60 3.83 4.04 1.88 1.55
0.84 1.72 0.94 0.64 2.52 3.68 1.69 2.80 0.72 0.18 31.39 32.66
0.40 0.08 0.35 0.52 0.01 0.00 0.09 0.00 0.47 0.86 0.16 0.01
32.45 30.5 32.19 32.78 1.49 4.46 31.01 3.01 34.74 37.21 36.32 37.18
6.08 7.74 6.21 5.47 12.02 14.70 13.29 17.16 10.30 8.66 1.07 31.08
The interaction coe¤cient is de¢ned as i = (PTScombined 3PTScontrol )3[(PTSnoise 3PTScontrol )+(PTSctyrene 3PTScontrol )]; t = i/S.D.; P = P (dtd 6 dTQ d), TQ is the Student variable with Q degrees of freedom; limits are i þ TQ;0:025 US.D.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
91
Fig. 2. PTS versus frequencies obtained after 750 ppm styrene exposure, 97 dB octave band noise centered at 8 kHz and simultaneous noise and styrene exposures for 6 h/day, 5 days/week, 4 weeks. The arithmetic sum of the e¡ects of noise and styrene was also plotted. The bars represent the 95% con¢dence intervals.
ble 1. Below and above these frequencies, the hearing loss induced by the combined exposure was quite similar to the summated losses obtained by each toxic agent alone. 3.2. Morphological results 3.2.1. Light microscopy The average cytocochleogram of eight animals belonging to group C is £at for all the rows of hair cells (OHC1^3) and is therefore not depicted. Fig. 3A shows the averaged (n = 8) cytocochleogram of group N. For OHC1, approximately 12% were missing at 17 kHz frequency place. The two other rows had only slight damage: 3% for OHC2 and 5% for OHC3. Fig. 3B illustrates the cytocochleogram (n = 8) of group S. The largest losses are located at OHC3, 86% at 20 kHz and 70% at 4 kHz. OHC2 is less damaged (25% at 20 kHz and 36% at 4 kHz) than OHC3, but more than OHC1 (15% at 20 kHz and 17% at 4 kHz). Frequencies above 30 kHz seem to be well preserved for each row of OHCs. Similarly, IHC were well preserved. Fig. 3C depicts the cytocochleogram of group S+N. As for group S, the order of the trauma can be characterized as follows : missing OHC3 s OHC2 s OHC1 in terms of missing cells. However, OHC2^3 show larger losses in the vicinity of 20 kHz (OHC3: 94%; OHC2 : 42%) and 4 kHz (OHC3: 86% ; OHC3: 62%) compared to those obtained with group S.
3.2.2. Scanning electron micrography All the scanning electron micrographs (SEM) are from the second turn and correspond to the 16 kHz region. Fig. 4 (#3491 ; U2000) is a SEM of the OC from a noise-exposed rat. OHC1 is more severely damaged than OHC2 and OHC3. Three phalangeal scars (butter£y-like shape) and fusion of stereocilia can be observed at the level of OHC1, whereas splayed stereocilia were observed on IHC. Splayed stereocilia could be observed at the most damage frequencies, from 8 to 20 kHz. Fig. 5 (#3425 ; U2000) is a SEM of the OC from a styrene-treated rat. Only ¢ve hair cells remain in row 3, whereas OHC1^2 are well preserved. Scars have replaced missing hair cells in OHC3. Note the concavity of the scar formation in the reticular lamina at the level of OHC3. The inner hair cells do not seem to be injured. Fig. 6 (#3499, U2000) is a SEM of the OC from a rat exposed to both noise and styrene. The loss of outer hair cells is large even in the second row. Similar to that observed following exposure to noise alone, scar formation (butter£y-like shape) replaced missing OHC1. The third row shows the most damage with only one hair cell remaining. Two concavities can be observed similar to those observed in animals exposed only to styrene (Fig. 5). Splayed stereocilia were observed on the IHC. As for the noise exposure, splayed stereocilia could be observed at the most damaged frequencies, from 8 to 20 kHz.
HEARES 3344 7-12-99
92
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 3. A: Average cytocochleogram (n = 8) obtained from rats exposed to a 97 dB octave band noise centered at 8 kHz for 6 h/day, 5 days/ week, 4 weeks. B: Obtained from rats exposed to 750 ppm styrene for 6 h/day, 5 days/week, 4 weeks. C: Obtained from rats exposed to both noise and styrene. Abscissa, upper trace: length (mm) of the entire spiral course of the organ of Corti from the bottom to the hook; lower trace: frequency map according to Mu«ller (1991). Ordinate: hair cell loss in percent. IHC: inner hair cells; OHC1: ¢rst row of outer hair cells; OHC2: second row; OHC3: third row.
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
93
Fig. 4. Scanning electron micrograph (#3491, U2000) of the 16 kHz region of the OC from a noise-exposed rat. Three phalangeal scars (stars) and fusion of stereocilia (black arrows) can be observed at the level of OHC1. Splayed stereocilia (arrowheads) are noted among the stereociliary bundles of IHC.
4. Discussion Following simultaneous exposure to noise and styrene, the auditory threshold shifts were increased compared to auditory de¢cits caused either by noise or by styrene alone (Fig. 1C). Indeed, both CTS and PTS were signi¢cantly greater (38.4 and 29 dB, respectively) than those obtained following noise exposure alone at the peak of hearing loss (12 kHz). Similarly, the cochleae of the combination group showed a large increase in the number of missing cells compared to either the noise alone or the styrene alone group (Fig. 3B,C). The pattern of damage (order and location of the trauma) in the combined exposures appears more like that observed with styrene than with noise. Thus, both the electrophysiological and morphological results support an interaction between noise and styrene on hearing loss and cochlear damage in the rat cochlea : frequencies and the order of the trauma look like those obtained with styrene. It is useful to clarify the terminol-
ogy related to the interaction concept, especially synergism and additivity. As reported by Nyle¨n (1994) and Fechter (1995), additivity is obtained when the toxic e¡ect of the combined exposures can be predicted by the arithmetic sum of the e¡ects observed by each individual exposure. As for the concept of synergism, the toxic e¡ect of the combined exposures must be greater than that expected by the sum of the e¡ects observed by each individual exposure. The authors will refer to this terminology in the present study. Fig. 2 illustrated the overall interactions between the two agents on hearing. There was a signi¢cant synergy between the e¡ects of both agents within the range 6^12 kHz (Table 1), as the PTS values induced by the combined exposure exceeded the summed losses caused by styrene alone and noise alone. A simple additivity was noted at and above 16 kHz. As with the noise exposure, the styrene exposure caused a long-lasting threshold shift in auditory sensitivity in the rat. This ¢nding con¢rmed the results of
HEARES 3344 7-12-99
94
R. Lataye et al. / Hearing Research 139 (2000) 86^96
Fig. 5. Scanning electron micrograph (#3425; U2000) of the OC from a styrene-treated rat. Note that concavities in the reticular membrane replaced missing OHC3 (arrowheads).
previous studies with this solvent (Crofton et al., 1994). After a 6 week post-exposure period, slight (4.3 dB at 20 kHz) but signi¢cant recovery was observed over a large frequency range from 2 to 20 kHz. This result has never been reported previously. The maximum rate of metabolism (Vmax ) might explain, at least partly, the recovery measured in the present study. Indeed, according to Filser et al. (1993), Vmax is close to 700 ppm in rats. Thus, one can consider that the rat is capable of rapidly metabolizing accumulated styrene in tissues after 4 weeks of 750 ppm exposure, decreasing the risk of injury to the hair cells. PTS also con¢rmed that styrene caused a selective mid-frequency (16^20 kHz) hearing de¢cit. This mid-frequency susceptibility to styrene was consistent with data obtained from other ototoxic solvents such as toluene (Campo et al., 1997) or trichloroethylene (Crofton et al., 1994). As for the histological data, two major permanent morphological changes were observed in the noise-exposed cochleae: hair cell losses and stereocilia changes.
The missing hair cells reported in an averaged cytocochleogram (Fig. 3A) showed that OHC1 were more susceptible to acoustic injury than the two other rows of OHCs. However, the use of surface preparations underestimated the damage of the organ of Corti, since no missing hair cells were counted at 10 kHz for instance, whereas a 16 dB PTS was measured by electrophysiology. Using SEM, the detailed observations of the organ of Corti showed numerous clumped and fused stereocilia on OHC1 and splayed stereocilia on IHCs (Fig. 4) at the most damaged frequencies (8^20 kHz). Thus, in our experimental conditions, the acoustic trauma was characterized less by the lack of cells than by injured stereocilia. These aspects of stereocilia injuries may result from excessive motion of the basilar membrane which caused collisions among stereocilia. Thus, the order of hair cell damage (OHC1 s IHC s OHC2 and OHC3) caused by the 97 dB noise exposure was in agreement with the progression established in the guinea pig earlier by Robertson and Johnstone
HEARES 3344 7-12-99
R. Lataye et al. / Hearing Research 139 (2000) 86^96
95
Fig. 6. Scanning electron micrograph (#3499; U2000) of the OC from a rat exposed to both noise and styrene. Phalangeal scars (stars) and concavities (large arrowhead) replaced respectively missing OHC1 and OHC3. Splayed stereocilia can be observed in the IHC row (small arrowheads).
(1980). Therefore, the electrophysiological and histological data clearly demonstrate that our experimental conditions produced a typical acoustic injury based on a mechanical process. From a morphological point of view, the lesions of the organ of Corti from styrene-exposed animals were very di¡erent from those in noise-exposed animals. For instance, severe and progressive OHC losses (OHC3 s OHC2 s OHC1) were detected (Fig. 3B), whereas no obvious stereocilia damage was noted (Fig. 5). In contrast, OHC losses were not only located in the mid-frequency area (as reported by Yano et al., 1992) but also in the mid- to low-frequency area (4 kHz). In fact, this discrepancy between morphological and electrophysiological ¢ndings had already been discussed regarding toluene (Campo et al., 1997). Brie£y, that might be due to the enhancement of the low-frequency amplitude responses by the inferior colliculus following trauma (Salvi et al., 1990), and/or to the speci¢c sensitivity of the apical turn of the rat cochlea (Prosen and Moody, 1991). Although the exact mechanism of sty-
rene toxicity is not understood, it is likely that styrene impairs preferentially the basal pole of OHC and/or the supporting cells by tissue contamination. Indeed, a possible route to reach the OHC is the lipid-rich content of the membranes of the di¡erent cells of the organ of Corti. The styrene could di¡use through the outer sulcus and reach the lipid-rich Hensen's cells which are connected with the Deiters' cells (Campo et al., 1999). Based on these results, the e¡ect of styrene is therefore di¡erent from that of noise and is mainly caused by a chemical process. Based on the ¢ndings reported above, it is realistic to think that the damage caused by combined exposure is generated by the same mechanisms (chemical or mechanical) as that described previously for each individual agent. In other terms, the coexistence of both mechanisms does not seem to generate a new ototraumatic mechanism. As a result, the synergy could be the result of a potentiation of one agent by the other. The potentiation could be the result of two phenomena: ¢rst, a weakening of the OHC membranes induced by the sol-
HEARES 3344 7-12-99
96
R. Lataye et al. / Hearing Research 139 (2000) 86^96
vent which might increase the micro-lesions due to noise exposure (Mulroy et al., 1998) ; second, a cell poisoning of the organ of Corti due to the solvent, which would reduce the availability of necessary energy during noise exposure. Given the risks encountered by subjects exposed to solvents and noise, it is reasonable to wonder if the solvent as well as the noise threshold limit values are really adequate in multiple occupational exposures. 5. Conclusion Rats exposed simultaneously to styrene and noise su¡er more hearing loss and greater cochlear damage than those caused by each agent alone. This synergistic e¡ect is primarily limited to the spectrum of the noise. The damage to the ear caused by noise and styrene occurs due to two di¡erent mechanisms (chemical versus mechanical). Acknowledgements The authors wish to thank Dr. F. Boettcher (Medical University of South Carolina, Charleston, SC, USA) and Dr. T. Morata (NIOSH, USA) for their comments on an earlier draft of the manuscript. They also wish to thank C. Barthe¨le¨my, M. Roure, P. Bonnet and J.C. Vigneron (INRS, FRANCE) for their technical assistance. M. Grzebyk (INRS) is acknowledged for the statistical analyses. References Barrega®rd, L., Axelsson, A., 1984. Is there an ototraumatic interaction between noise and solvents? Scand. Audiol. 13, 151^155. Borg, E., Canlon, B., Engstro«m, B., 1995. Noise-induced hearing loss. Scand. Audiol. 24 (Suppl. 40). Calebrese, G., Martini, A., Sessa, G., Cellini, M., Bartolucci, G.B., Marcuzzo, G., De Rosa, E., 1996. Otoneurological study in workers exposed to styrene in the ¢berglass industry. Int. Arch. Occup. Environ. Health 68, 219^223. Campo, P., Lataye, R., Cossec, B., Placidi, V., 1997. Toluene-induced hearing loss: a mid-frequency location of the cochlear lesions. Neurotoxicol. Teratol. 19, 129^140. Campo, P., Lataye, R., Loquet, G.M., 1999. Toluene and styreneinduced hearing losses: a comparative study. In: Prasher, D., Canlon, B. (Eds.), Cochlear Pharmacology and Noise Trauma. NRN, London, pp. 113^124. Cary, R., Clarke, S., Delic, J., 1997. E¡ects of combined exposure to noise and toxic substances ^ critical review of the literature. Ann. Occup. Hyg. 41, 455^465. Crofton, K.M., Lassiter, T., Rebert, C., 1994. Solvent induced ototoxicity in rats: An atypical selective mid-frequency hearing de¢cit. Hear. Res. 80, 25^30.
Fechter, L.D., 1993. E¡ects of acute styrene and simultaneous noise exposure on auditory function in the guinea pig. Neurotoxicol. Teratol. 15, 151^155. Fechter, L.D., 1995. Combined e¡ects of noise and chemicals. Occup. Med. State Art Rev. 10, 609^621. Filser, J.G., Schwegler, U., Csana¨dy, G.A., Greim, H., Kreuzer, P.E., Kessler, W., 1993. Species-speci¢c pharmacokinetics of styrene in rat and mouse. Arch. Toxicol. 67, 517^530. Gao, W.Y., Ding, D.I., Zheng, X.Y., Run, F.M., Liu, Y.J., 1992. A comparison of changes in the stereocilia between temporary and permanent hearing losses in acoustic trauma. Hear. Res. 62, 27^41. Jacobsen, P., Hein, H.O., Suadicani, P., Parving, A., Gyntelberg, F., 1993. Mixed solvent exposure and hearing impairment: an epidemiological study of 3284 men. The Copenhagen male study. Occup. Med. 43, 180^184. Johnson, A.C., Juntunen, L., Nyle¨n, P., Borg, E., Ho«glund, G., 1988. E¡ect of interaction between noise and toluene on auditory function in the rat. Acta Otolaryngol. (Stockh.) 105, 56^63. Johnson, A.C., Nyle¨n, P., 1995. E¡ects of industrial solvents on hearing. Occup. Med. State Art Rev. 10, 623^640. Lataye, R., Campo, P., 1997. Combined e¡ects of a simultaneous exposure to noise and toluene on hearing function. Neurotoxicol. Teratol. 19, 373^382. Liberman, M.C., 1987. Chronic ultrastructural changes in acoustic trauma: serial section reconstruction of stereocilia and cuticular plates. Hear. Res. 26, 65^88. Liberman, M.C., Dodds, L.W., 1987. Acute ultrastructural changes in acoustic trauma: Serial section reconstruction of stereocilia and cuticular plates. Hear. Res. 26, 45^64. Miller, R.R., Newhook, R., Poole, A., 1994. Styrene production, use, and human exposure. Crit. Rev. Toxicol. 24 (Suppl. 1), S1^S10. Mo«ller, C., Odkvist, L., Larsby, B., Tham, R., Ledin, T., Bergholtz, L., 1990. Otoneurological ¢ndings in workers exposed to styrene. Scand. J. Work Environ. Health 16, 189^194. Morata, T.C., Dunn, D.E., Sieber, W.K., 1994. Occupational exposure to noise and ototoxic organic solvents. Arch. Environ. Health 49, 359^365. Mu«ller, M., 1991. Frequency representation in the rat cochlea. Hear. Res. 51, 247^254. Mulroy, M., Henry, W.R., McNeil, P.L., 1998. Noise-induced transient microlesions in the cell membranes of auditory hair cells. Hear. Res. 115, 93^100. Nyle¨n, P., 1994. Organic solvent toxicity in the rat; with emphasis on combined exposure interactions in the nervous system. Arbete Ha«lsa 3, 1^50. Prosen, C.A., Moody, D.B., 1991. Low-frequency detection and discrimination following apical hair cell destruction. Hear. Res. 57, 142^152. Robertson, D., Johnstone, B.M., 1980. Acoustic trauma in the guinea pig cochlea: early changes in ultrastructure and neural threshold. Hear. Res. 3, 167^179. Salvi, R., Saunders, S., Gratton, M., Arehole, S., Powers, N., 1990. Enhanced evoked response amplitude in the inferior colliculus of the chinchilla following acoustic trauma. Hear. Res. 50, 245^ 258. Sass-Kortsak, A.M., Corey, P.N., Robertson, J.M., 1995. An investigation of the association between exposure to styrene and hearing loss. Ann. Epidemiol. 5, 15^24. Yano, B.L., Dittenber, D.A., Albee, R.R., Mattsson, J.L., 1992. Abnormal auditory brain stem responses and cochlear pathology in rats induced by an exaggerated styrene exposure regimen. Toxicol. Pathol. 20, 1^6.
HEARES 3344 7-12-99