TOXICOLOGICAL SCIENCES 42, 2 8 - 3 5 (1998)
ARTICLE NO. TX972413
Trichloroethylene Ototoxicity: Evidence for a Cochlear Origin1
Laurence D. Fechter,*-2 Ye Liu,*-2 David W. Herr.f and Kevin M. Croftonf3
'Toxicology Program, The University of Oklahoma College of Pharmacy, Oklahoma City, Oklahoma; and \Neurotoxicology Division, National Health and
Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711
Received July 14, 1997; accepted November 30, 1997
Trichloroethylene (TCE) is one of several volatile organic
solvents that have been identified as auditory toxicants (ototoxicants) in laboratory animals (Rebert et al, 1991; Jaspers et
al., 1993; Crofton and Zhao, 1993, 1997; Crofton etal., 1994).
Trichloroethylene (TCE) is known to produce an unusual pat- TCE, styrene, toluene, and p-xylene all produce an atypical
tern of hearing impairment in laboratory animals marked by a pattern of auditory dysfunction marked by a preferential impreferential loss of threshold sensitivity at midfrequencies. The pairment of auditory function at middle frequencies of the
purpose of this research was to determine whether the TCE- audiogram (Pryor et al, 1987; Mattsson et al., 1990; Yano et
induced auditory deficit results from cochlear dysfunction. Adult al., 1992; Crofton and Zhao, 1993; Jaspers et al., 1993; Crofton
Long Evans hooded rats were exposed via inhalation to either 0 et al., 1994; Johnson and Canlon, 1994a,b; Campo et al.,
(clean air) or 4000 ppm TCE (6 h/day for 5 days). Auditory 1997). The pattern of ototoxicity produced by these volatile
thresholds for 1-40 kHz tones were determined 3 weeks after
organic solvents can be contrasted against the preferential
exposure using reflex modification audiometry (RMA; n = 12/
group). Cochlear electropotentials were measured during subse- disruption of high-frequency auditory function seen with the
quent testing (n = 3-10/group) 5 to 7 weeks after exposure, well-studied ototoxic aminoglycoside antibiotics and antineincluding thresholds for cochlear action potentials (CAP) and the oplastics (cf. Brown and Feldman, 1978; Rybak, 1986, 1992;
1-H.V cochlear microphonic for 2-40 kHz tones, and the N, am- Harpur, 1987) as well as some chemical contaminants such as
plitude intensity function (40-90 dB SPL). Cochlear histopathol- trimethyltin (Crofton et al., 1990; Fechter and Carlisle, 1990;
ogy was assessed in midmodiolar preparations of a separate set of Clerici et al., 1991) and triethyltin (Clerici et al, 1991).
animals, exposed as before (n = 4/group). RMA testing confirmed
The cochlea has been shown to be a target for some ototoxic
a TCE-induced loss in midfrequency threshold sensitivity (8 and solvents. Toluene exposure resulted in a loss of middle and
16 kHz). CAP thresholds were elevated at midfrequencies (8 and lower turn outer hair cells in rats (Pryor et al, 1984; Sullivan
16 kHz) among TCE-treated subjects, along with a suppression of
et al, 1988; Johnson and Canlon, 1994b). Consistent with the
the N, amplitude from 50 to 90 dB SPL. The cochlear microouter
hair cell damage, toluene exposure also caused a reducphonic, a nonpropagated ac potential generated largely by the
tion
in
midfrequency distortion product otoacoustic emissions
outer hair cells, was not affected by the TCE treatment. Cochlear
(Johnson
and Canlon, 1994a) and an increase in behavioral
histopathology revealed a loss of spiral ganglion cells that was
significant in the middle turn, but not in the basal turn. There was thresholds for midfrequency tones (Crofton et al., 1994). Styan inconsistent loss of hair cells among treated subjects. The data rene exposure has been shown to disrupt the initial negative
suggest strongly that the behaviorally determined loss in auditory peak of the auditory-evoked brainstem potential and to damage
function can be accounted for by a cochlear impairment and that outer hair cells in treated subjects (Yano et al, 1992). With
the spiral ganglion cell may be a prominent target of TCE. c IWS respect to TCE, published reports do not clearly address the
involvement of the cochlea in the auditory impairment. BehavSociety of Toxicology.
Key Words: trichloroethylene; ototoxicity; cochlea; spiral gan- ioral studies (Jaspers et al, 1993; Crofton and Zhao, 1993,
1997; Crofton et al., 1994) demonstrate a clear pattern of
glion cell.
midfrequency hearing impairment, but are not able to determine the site within the auditory system where this impairment
occurs. Electrophysiological studies have consisted of far-field
1
This article has been reviewed by the National Health and Environmental
auditory brain stem response measurements that show a drop in
Effects Research Laboratory, U.S. Environmental Protection Agency, and
amplitude of all components of the click and tone auditory
approved for publication. Mention of trade names or commercial products does
not constitute endorsement or recommendation for use.
evoked potential following TCE exposure (Rebert et al, 1991,
2
Supported in part by PHS Grant ES02852 and by NSF EPSCoR Cooper1993). The initial peak of the auditory brain stem evoked
ative Agreement OSR-9550478.
potential is thought to have a cochlear origin, while more
3
To whom correspondence should be addressed at Neurotoxicology Divicentral structures are thought to generate later potentials (cf.
sion, MD-74B, NHEERL, USEPA, Research Triangle Park, NC 27711.
Chen and Chen, 1991; M0ller et al, 1988; Popper and Fay,
E-mail: [email protected].
Trichloroethylene Ototoxicity: Evidence for a Cochlear Origin.
Fechter, L. D., Liu, Y., Herr, D. W.( and Crofton, K. M. (1998).
Toxicol. Sri. 42, 28-35.
1096-6080/98 $25.00
Copyright O 1998 by the Society of Toxicology
All rights of reproduction in any form reserved.
28
TRICHLOROETHYLENE OTOTOXICITY
1993; Starr and Zaaroor, 1990). These data are consistent with,
but do not firmly establish, a cochlear locus of dysfunction.
There are no TCE data available that allow a differentiation of
potential sites of damage within the cochlea.
Based on the known midfrequency hearing loss induced by
TCE exposure (Rebert et al, 1991; Jaspers et al, 1993; Crofton and Zhao, 1993, 1997; Crofton et al, 1994) and morphological evidence of outer hair cell loss with other solvents
(Sullivan et al, 1988; Yano et al, 1992; Johnson and Canlon,
1994b), we hypothesized that TCE exposure would disrupt the
function and/or integrity of cochlear outer hair cells. To test
this hypothesis, we evaluated cochlear function in rats using
behavioral, electrophysiological, and morphological methods,
after inhalation exposure to TCE. Behavioral thresholds were
measured with reflex modification audiometry (RMA) and
used to confirm previous findings (Crofton and Zhao, 1993,
1997; Crofton et al, 1994). Electrophysiological measures
were selected to provide diagnostic information on different
cochlear cell types. The l-/xV cochlear microphonic potential
(CM) was used to characterize the function of outer hair cells.
The CM is known to represent predominantly the response of
the outer hair cells (Dallos et al, 1972; Dallos and Cheatham,
1976; Sellick and Russell, 1980). Cochlear action potential
(CAP) thresholds were used to characterize the output of the
cochlea via the auditory nerve. The amplitude of the CAP
reflects neural activity in the auditory nerve, which consists
primarily of Type I afferents that propagate activity generated
by inner hair cells at Type I spiral ganglion cells (cf. M0ller,
1983). Wang et al (1997) demonstrated a loss in the maximal
CAP output from the cochlea of chinchillas in which inner hair
cell loss was produced by treatment with carboplatin. To the
extent that the CM amplitude remains unaffected while the
CAP is disrupted, it is possible to attribute impaired function to
the inner hair cells and/or Type 1 spiral ganglion cells. We also
measured the growth of the N, potential as a function of
increasing stimulus intensity at suprathreshold levels (N, input/
output). Hall (1990) has demonstrated that the P, wave of the
brainstem auditory-evoked potential, equivalent to the N, recorded from the round window, is highly correlated with the
number of spiral ganglion cells that remain after toxic injury.
An increase in the slope of the N, input/output function, in
conjunction with an increase in the threshold, is diagnostic of
loudness recruitment (cf. Eggermont, 1983). Histopathological
data were also obtained to confirm and further refine the
conclusions reached from the electrophysiological studies.
METHODS
Subjects. Male Long Evans hooded rats (Charles River, Inc., Raleigh, NC)
were obtained at approximately 60 days of age and were housed two per cage
in standard plastic hanging cages (24 X 20 X 45 cm) in an AAALACapproved facility (National Health Effects and EnvironmentaJ Research Laboratory, USEPA, Research Triangle Park, NC). All animals were given a
10-day acclimation period and were maintained on a 12:12-h photoperiod, L:D
(0600:1800). Food (Purina Rat Chow) and tap water were available freely.
29
Temperature was maintained at 21 ± 2°C and relative humidity was maintained at 50 ± 10*.
A total of 48 animals (maximal use of exposure chambers) were exposed to
either air (n = 24) or 4000 ppm TCE (n = 24) as described below. Behavioral
thresholds were determined in 12 animals exposed to TCE and 12 were
exposed to air. A random selection of 40 animals (n = 20/group) were coded
as to group condition and shipped via air to the animal facilities at the
Department of Environmental Health Science at the Johns Hopkins School of
Hygiene. Animals remaining in the EPA colony were used in separate experiments not described here. Coding of shipped animals ensured that physiological testing and anatomical analyses of subjects were blind with respect to
treatment. Subjects were housed singly in the AAALAC animal facility until
tested under environmental conditions similar to those at EPA. Animals were
given a 4-day (minimum) acclimation period. Cochlear potentials were measured 5 to 6 weeks postexposure in a total of 20 subjects (n = 10/group) that
were randomly selected from those sent to Johns Hopkins. The cochleas of
eight rats sent to Johns Hopkins, but not tested electrophysiologically, were
maintained for 11 weeks following TCE (n = 4) or air exposure (n = 4) and
then utilized for histopathology. This late time point was chosen to ensure that
there was no ongoing pathological processes. The remaining 12 unused animals were euthanized at the end of the study. All investigations were performed in accordance with the principles of the Declaration of Helsinki and in
accordance with animal use protocols approved by the appropriate institutional
animal care and use committees.
Chemical exposure. Animals were exposed to 1,1,2-trichloroethylene vapor for 6 h/day for 5 days at 0 (air only) or 4000 ppm as previously described
(Crofton and Zhao, 1993). Briefly, separate Hinners type exposure chambers
(Hinners el al., 1968) were used that had a volume of 422 L. Inlet air was first
carbon and HEPA filtered and then conditioned to 22 ± 2°C, 50 ± 1 0 *
relative humidity. Chamber air flow rates were set to 100 L/min or 14.2
changes/h. The vapor generation system was of the ' 'J-tube'' design (Miller et
al, 1980), with zero-grade nitrogen used for the generator sweep gas. Reagentgrade 1,1,2-trichloroethylene (99 + %, Aldrich) was metered into the generator using a rotary piston pump. Chamber concentrations were monitored using
long infrared spectrophotometry (Foxboro, Miran 1A, Foxboro, MA). All
exposure concentrations were within ± 1.3% of expected values.
Reflex modification audiometry thresholds. Auditory function was monitored using a modification of the procedures of Young and Fechter (1983) and
Ison (1984). This procedure estimates RMA thresholds to auditory tones (see
Crofton, 1992; Crofton et ai, 1990, for details). Testing was conducted in eight
sound-attenuated chambers, each containing a wire mesh plastic-framed test
cage (7.6 X 7.6 X 23 cm), mounted on a load cell/force transducer assembly
designed to measure vertical force (Ruppert et al, 1984).
Each rat was placed in a test cage and, following a 10-min adaptation period
at an ambient noise level of 28 dB (A-weighted), received a total of 240 trials.
There were 10 trials at each of 23 different sound pressure levels (SPL) of the
prepulse stimulus (S,, -6 to 90 dB SPL in 3- or 6-dB increments, 20-ms
duration, 2.5-ms rise/fall). The eliciting stimulus (SJ was a 120-dB, 40-ms,
burst of white noise (generated by a 32 stage linear feedback shift register,
rise/fall time = 2.5 ms). All sound measurements were decibels of SPL (linear,
with a 10 Hz high pass cutoff; re: 20 /xPa), unless otherwise stated. Each rat
also received 10 blank control trials, during which only S2 was presented. The
order of presentation of the trials was computer-generated in a semirandom,
but balanced fashion, so that within each of ten 24-trial blocks, each S,/S 2
condition was presented once. The intertrial interval was 15 s, and the interstimulus interval (measured from onset of S, to onset of Sj) was 90 ms.
Animals were tested for no more than two frequencies on any one day to
ensure no interaction between testing and auditory function. The following
frequencies were tested in random order 1, 4, 8, 16, 24, 32, and 40 kHz. RMA
thresholds were determined using a modified segmented-line model and defined as the joint point in the segmented line model, which corresponds to the
S, intensity above which the response was inhibited (see Crofton, 1990, 1992;
Crofton et al., 1990, for details). Note that data from all trials were analyzed.
Acoustical measurements were made with a Brflel and Kjar (B&K) 2636 (or
2610) measuring amplifier equipped with a B&K 4135 0.25-in. microphone or
30
FECHTER ET AL.
aB&K2209 impulse level meter with a B&K 4165 0.5-in. microphone. Output
was calibrated at I kHz with a B&K 4230 sound level calibrator. Ambient
noise in the EPA animal colony averaged 78 dB with most power at the very
low frequency end (<200 Hz), and a 4O-dB drop from 200 Hz to 10 kHz.
Behavioral test chamber noise averaged 69.1 dB (28.4 dB A-weighting). The
noise in the solvent exposure chamber averaged 76.2 dB: 55 dB peak power at
0.5—1 kHz with an approximate 15 dB drop per octave (see Crofton and Zhao,
1993, for specifics).
Cochlear electrodes. Animals were anesthetized with urtthane (1.5 g/kg)
administered intrapentoneally (ip) for general anesthesia. The disappearance of
the blink reflex and the pedal withdrawal reflex to digit compression were the
indicators for adequacy of anesthetic level. As necessary, this dosage was
supplemented with additional urethane (0.15 g/kg) injections. Lidocaine (2%)
was injected at the incision for local analgesia. The surgery was conducted in
a double-walled audiometric booth (IAC Model 122A, Bronx, NY). The rats
were positioned dorsally on a thermostatically controlled dc heating table
(Narco Biosystems) in order to maintain the body temperature at 39 ± 1 "C, the
temperature normally seen in anesthesized rats. The subject was held rigidly by
a head holder and tracheotomized to ensure unobstructed breathing, and the
right pinna was removed so that the tympanic membrane could be visualized.
The bulla was exposed ventrally by removing the overlying skin and muscles
and the cochlea was revealed by gentle shaving of the bulla using a surgical
blade to produce a hole through which the electrodes could be placed. A low
voltage lamp was directed at the cochlea to prevent cooling A Teflon-coated
silver wire (0.11 mm) recording electrode was positioned on the surface of the
round window membrane and was referenced to a Ag/AgCl electrode placed in
neck musculature. A custom made Plexiglas ear speculum was carefully
positioned into the external auditory meatus such that the tympanic membrane
was visible through the speculum.
CAP detection threshold and CM measurement.
Measurement of CAP
thresholds and the 1-^.V RMS CM isopotential curve were made from the
round window for tone stimuli of 2, 4, 8, 16, 32, and 40 kHz. The tones were
generated by an oscillator (Stanford Research Systems Model SR 530, Sunnyvale, CA), shaped by a tone switch (Model BSiT), attenuated by a programmable attenuator (Model PATT, Willsonics, Los Angeles, CA), and amplified
by a high impedance amplifier (Interstellar Research, Ann Arbor, MI) in order
to drive a 0.5-in. condenser microphone (ACO Pacific, Belmont, CA) placed
within the plastic speculum fitted inside the rat's external auditory meatus. The
CAP was generated using pure tone pips of 10 ms duration with 1 ms
onset-offset ramps which were presented at a rate of 9/s. The electrophysiological response was bandpass filtered between 300 and 1000 Hz and amplified
1000X (Grass, P15 ac preamplifier, Quincy, MA). CAP thresholds were
defined as the stimulus intensity that produced a just detectable N, wave
observed on a Nicolet Model 2090 digital oscilloscope (Madison, WI). The
approximate minimal detectable potential was 1 fiV baseline to peak.
The cochlear microphonic was elicited by continuous pure tone stimuli
generated as described above. Tone intensity was increased automatically until
a 1 piV RMS signal was achieved. The cochlear microphonic was high pass
filtered above 300 Hz, amplified 1000X, as above, and filtered using a lock-in
amplifier (Stanford Research Systems Model SR 530, Sunnyvale, CA).
Calibration of tone intensities was conducted on a subject-by-subject basis.
Subsequent to the last threshold determination, the sound intensity of each tone
was measured for each subject by using the lock-in amplifier to measure the
voltage output of a 0.5-in. B&K 4134 microphone coupled to a probe tube
inserted into the speculum to within approximately 1 mm of the tympanic
membrane.
N, input-output functions.
Once the CAP threshold and the 1-ftV CM
response were measured, the amplitude of the N, was determined for 16 kHz
tones with a duration of 10 ms, a 1-ms rise/fall time, and stimulus rate set at
9.7/s. Tone intensity was increased in 10-dB steps between 40-90 dB SPL.
The CAP amplitude was tested in stepwise fashion starting at the lowest
stimulus intensity (40 dB SPL) and progressing to die highest test intensity (90
dB SPL) in order to minimize the confounding effect of auditory fatigue
produced by high sound intensities on subsequent trials. Responses at each
stimulus intensity were averaged over 256 stimulus presentations by the
computer. N, input/output responses were saved on a computer using a custom
software program. The N, amplitude voltage was measured from the baseline
to the peak of the N,.
Cochlear histopathology. Eleven weeks following the termination of
chemical or air exposure subjects were euthanized and cochlea tissues were
harvested for histological examination of the basal and middle turns. The
apical turn which encodes the lowest sound frequencies was not examined. The
methods used have been described previously by Fechter el al. (1992). Subjects were anaesthetized with urethane (1.5 g/kg), transcardially perfused with
phosphate buffered saline and then with 3% glutaraldehyde and 2% paraformaldehyde in phosphate buffered saline. The auditory bullae were removed
from the subjects and opened to expose the cochlea, and the round and oval
windows of the cochlea were opened. A hole was made in the apex to enhance
perfusion and fixation. The tissue was gently agitated overnight in fresh
fixative. The tissue was washed in 0.1 M sodium cocodylate buffer and then
decalcified in 10% EDTA at 4°C, with agitation, for approximately 3 days. The
processed tissue was washed thoroughly with fresh cocodylate buffer and then
postfixed with 2% osmium tetroxide for 60 min. The tissue was again washed
in 0.1 M sodium cocodylate buffer and allowed to remain in buffer overnight.
The tissue was dehydrated using a series of ethanol washes and embedded in
JB4 resin and allowed to cure. The tissue from four subjects per group was cut
in the modiolar plane using a Zeiss HM 350 automated microtome at a
thickness of 3 fim and stained with toluidine blue and sections were coded as
to exposure group. Tissue sections were matched for depth of section in the
cochlea using landmarks, including appearance of all scale in section, size of
Rosenthal's canal, and dimensions of the organ of Corti. Qualitative assessment of the tissue was followed by quantification of total spiral ganglion cell
numbers, the diameter of spiral ganglion cells, and the area of Rosenthal's
canal. Area measurements were made with a computer-based morphometric
program. Cell density was determined for each section by dividing the number
of cells by the area of Rosenthal's canal. This was completed for four subjects
per exposure group. While slides were coded to obscure exposure group, there
were obvious differences between groups.
Statistical analysis. Analysis of variance (ANOVA) procedures were used
for all main and simple effects tests (SAS, 1989). Univariate two-way ANOVAs
with Geiser-Greenhouse corrections used to control the a for multiple withinsubject comparisons. In the case of more than one independent variable,
significant interactions were followed by simple-effects ANOVA tests for each
independent variable. When necessary, mean contrast comparisons between
treatment groups at specific test frequencies were made using the TukeyKramer studentized range test (SAS, 1989). An a level of 0.05 was used for all
step-down (simple-effects) and mean contrast tests. Higher frequency data for
the 1 -/iV CM were lost for some of the animals due to technical problems (for
2, 4, and 8 kHz, n = 10 and 10; for 16 kHz, n = 8 and 9; for 32 kHz, n = 4
and 5; for 40 kHz, n = 2 and 1; for AIR and TCE groups, respectively).
Therefore, data for the 40-kHz CM data was not analyzed. The RMA threshold
segmented-hne analysis (see Crofton et al, 1990; Crofton, 1992) was performed using PROC NONLIN with the DUD algorithm (SAS, 1989). RMA
amplitude data were weighted inversely proportional to the variance of the
mean at each intensity.
RESULTS
RMA thresholds. Trichloroethylene exposure resulted in a
mid-frequency hearing deficit (Fig. 1). Thresholds for the 8and 16-kHz tones were increased by approximately 25 dB in
the TCE-exposed animals compared to controls. These observed increases in the midfrequency tones were supported by
a significant frequency-by-treatment interaction (F(7,147) =
5.59, p < 0.0001) and significant effects of treatment at 8 kHz
(F(l,22) = 27.42, p < 0.0001) and 16 kHz (F(l,21) =
41.45, p < 0.0001). There were no significant effects of treatment for any of the other frequencies (all p's > 0.05).
31
TRICHLOROETHYLENE OTOTOXICITY
RMA Thresholds
16
N 1 Amplitude Input/Output -16 kHz
3240
Frequency (kHz)
FIG. 1. Trichloroethylene exposure resulted in a midfrequency hearing
loss as evidence by reflex modification audiometry data. Animals tested 2
weeks after exposure to air only (AIR) or 4000 ppm TCE (TCE) 6 h/day for
5 consecutive days. Data are presented as group means (±SE) ('Significantly
different from controls, p < 0.05; n — 12/group).
CAP detection thresholds. Assessment of the threshold for
eliciting a CAP showed a selective impairment in the TCEexposed group of approximately 20 dB which was restricted to
8 and 16 kHz (Fig. 2). Thresholds for the TCE-exposed group
were similar to air-exposed controls at both lower and higher
test frequencies. These conclusions were confirmed by a significant frequency-by-treatment interaction (F(l,18) = 14.69,
p < 0.0012) and significant effects of treatment only at 8 kHz
(F(l,18) = 14.69, p< 0.0012) 16 kHz (F(l,18) = 24.68, p<
CAP Thresholds
8
16
50
60
70
80
90
Stimulus Intensity (dB SPL)
32 40
Frequency (kHz)
FIG. 2. Trichloroethylene exposure resulted in a midfrequency hearing
loss as evidenced by elevation in sound level necessary to elicit a CAP.
Animals tested 4 to 5 weeks after exposure to air only (AIR) or 4000 ppm
trichloroethylene (TCE) 6 h/day for 5 consecutive days. Data are presented as
group means (±SE) ('Significantly different from controls, p < 0.05; n =
10/group).
FIG. 3. Growth of the CAP amplitude in response to increasing sound
levels at 16 kHz for subjects exposed to air only (AIR) or 4000 ppm trichloroethylene (TCE) 6 h/day for 5 consecutive days. TCE subjects showed a
depressed CAP amplitude at sound levels from 50 to 90 dB. Data are presented
as group means (±SE) ('Significantly different from controls, p < 0.05; n =
10/group).
0.0001). There were no significant effects of treatment for any
of the other frequencies (all p's > 0.05).
N, Input-output functions. The growth in the N, amplitude
for a 16-kHz tone was attenuated in TCE-treated subjects for all
sound intensities between 50 and 90 dB (Fig. 3). The CAP N,
amplitude increased as a function of stimulus intensity in both
groups, but TCE exposure resulted in smaller N, amplitudes
relative to controls. Statistical analyses revealed a significant
treatment-by-intensity interaction (F(5,9O = 4.11, p < 0.0429)
and significant effects of treatment at the 50- 90-dB intensities (all
F's(l>18) > 6.65, p < 0.0189). There was no significant effect of
treatment at 40 dB (F(l,18) = 3.70, p < 0.0703).
1 ILV Cochlear microphonic. Figure 4 illustrates the effect
of TCE exposure on the 1-/AV RMS CM. In contrast to the
disruption of the CAP elicited by TCE for midrange frequencies, there was no difference between control subjects and
TCE-treated subjects in the sound intensity needed to elicit a
1-JAV cochlear microphonic response from the round window.
This is supported by a nonsignificant treatment-by-frequency
interaction (F(4,28) = 0.93. p < 0.4633) and a nonsignificant
main effect of treatment (F(l,4) = 0.12, p < 0.7411). There
was a significant effect of frequency (F(4,28) = 19.47, p <
0.0001) indicating an increase in the stimulus intensity required to produce a l-/xV response at the higher frequencies
(Fig. 4).
Cochlear histopathology. The most remarkable histopathological finding in cochleas from TCE-exposed rats was a
marked decrease in the density of spiral ganglion cells among
treated subjects (Figs. 5A and 5B). This loss was apparent in all
TCE-treated sections. In contrast, only rarely were hair cells
32
FECHTER ET AL.
1 uV Cochlear Microphonic
16
32
Frequency (kHz)
FIG. 4. Sound levels necessary to elicit a 1-jiV
phonic potential in subjects exposed to air only (AIR)
ethylene (TCE) 6 h/day for 5 consecutive days. Data
means (±SE). There were no significant differences
4—10/group).
RMS cochlear microor 4000 ppm trichloroare presented as group
between groups (n =
observed to be missing. In many sections the loss of spiral
ganglion cells was obvious even though inner and outer hair
cells observed in the same section remained intact (Fig. 5B).
However, in other instances, the loss of outer hair cells was
noted along with the decline in spiral ganglion cell number.
Because a surface preparation of the organ of Corti was not
conducted, it is not possible to reconstruct the loss of hair cells
along the length of the cochlear partition in treated and control
subjects. Thus, it was not possible to quantify the hair cell loss.
The subjective impression, however, was that spiral ganglion
cell loss was consistent in all sections studied while hair cell
loss was an infrequent event.
Quantitative analysis of spiral ganglion cells revealed a
significant effect of TCE on cell density (Fig. 6). TCE exposure
resulted in decreased numbers of spiral ganglion cells in the
lower-middle (47%) and upper-middle turns (43%), relative to
controls. There was no difference in cell density in the basal
turn. These conclusions were supported by a significant turnby-treatment interaction (F(2,12) = 7.37, p < 0.0114) and a
significant main effect of treatment for the lower-middle
(F(l,6) = 12.04, p < 0.0133) and upper-middle turns (F(l,6)
= 15.72, p < 0.0074). Spiral ganglion cell diameters did not
differ between groups or among the turns of the cochlea. The
mean spiral ganglion cell diameter was 12.7 ± 0.3 /xm in
control subjects and 12.9 ± 0.6 /xm in TCE-treated subjects.
DISCUSSION
The data presented here clearly demonstrate that inhalation
exposure to TCE results in both functional and structural damage
to the cochlea in the rat The finding of a midfrequency selective
impairment for the RMA thresholds is consistent with previous
reports for TCE and other volatile organic solvents (Jaspers et al.,
1993; Crofton and Zhao, 1993; Crofton et al, 1994). The present
data extend the previous findings by characterizing the auditory
effects of TCE at the level of the cochlea. TCE exposure elevated
CAP thresholds for midfrequency tones and reduced N, amplitude at 16 kHz even at suprathreshold stimulus intensities. In
contrast, TCE exposure failed to alter the CM. These data, suggestive of an effect on inner hair cells and/or spiral ganglion cells,
were supported by histopathological observations of a loss of
spiral ganglion cells in the middle cochlear turns. Taken together,
these data clearly demonstrate the ototoxic effects of TCE and
suggest that the cochlear site of action is limited primarily to the
spiral ganglion cells.
The midfrequency hearing deficit reported here for the RMA
behavioral thresholds is consistent with numerous previous reports for a number of volatile organic solvents. Midfrequency
hearing deficits have been reported now for styrene, xylene,
toluene, and TCE (cf. Yano et al, 1992; Johnson and Canlon,
1994a,b; Crofton et al., 1994). The present data also support the
reliability of this effect within this laboratory.
More importantly, the present results demonstrate that TCE
ototoxicity has a cochlear site of action. TCE exposure produced
a 20- to 25-dB elevation in the CAP thresholds for 8- and 16-kHz
tones, but failed to alter thresholds at higher or lower frequencies.
This midfrequency effect is highly consistent with the effects on
RMA thresholds as well as previous reports for TCE and other
ototoxic solvents (cf. Crofton et al., 1994). CAP N, amplitude is
known to reflect the magnitude of neural activity in the auditory
nerve and gives evidence for the integrity of the overall peripheral
auditory system (cf. Davis et al., 1958; M0ller, 1983). Therefore,
the increase in CAP thresholds suggests a site of action at, or more
peripheral to, the auditory nerve.
TCE failed to produce any change in the l-/xV CM at any
frequency tested. The CM is known to represent predominantly
the response of the outer hair cells (Dallos et al., 1972; Dallos
and Cheatham, 1976; Sellick and Russell, 1980). This suggests
that TCE exposure had minimal or no impact on outer hair
cells. However, an alternative explanation for a lack of effect
of TCE on the CM is the overrepresentation of basal turn hair
cells in the CM recorded from the round window, which may
lead to an underestimation of the effects of TCE on middle or
upper turn hair cells (cf. Dallos, 1973; Sohmer et al, 1980).
The lack of any effect at the middle frequencies, coupled with
the correlative evidence of loss of ganglion cells and little or no
damage to hair cells (see discussion below) make this explanation unconvincing. Furthermore, the lack of effect of TCE on
the CM is different from other ototoxicants that damage outer
hair cells and also disrupt the CM (e.g., Fechter et al., 1992;
Davis et al, 1958). Together with the CAP data, these data
suggest that the effect of TCE may be restricted to inner hair
cells and/or spiral ganglion cells.
TCE exposure caused a reduction in the N, input-output
amplitude at all suprathreshold stimulus intensities for the
16-kHz stimulus. This type of response, a parallel shift to the
right in the input-output function, is unlike most classic oto-
TRICHLOROETHYLENE OTOTOXICrTY
100pm ,
FIG. 5. Light micrographs depicting loss of spiral ganglion cells (SPG) from the middle turn of the cochlea in a control subject (A) compared to a rat exposed
to TCE (B). Scale bar, 100 /xm. Note the normal presence of inner hair cell (IHC) and outer hair cells (OHC) in both sections.
toxicants that typically cause loudness recruitment. A comparable lack of loudness recruitment in the intensity function for
brain stem-auditory evoked potential amplitudes has been reported following TCE exposure (Rebert et aL, 1991). An
increase in the slope of the N, input-output function, in
conjunction with an increase in the threshold, is characteristic
of loudness recruitment with a loss in threshold sensitivity (cf.
Eggermont, 1983; Gulick et aL, 1989). Age-related hearing
dysfunction (presbycusis) and aminoglycoside antibiotic ototoxicity represent two common instances of cochlear impairment where loudness recruitment occurs (Aran and Darrouzet,
1975; Dallos and Wang, 1974; Henry and Lepkowski, 1978;
Hunter and Willow, 1987). By contrast, when inner hair cells
are selectively impaired a decrement in maximal CAP output is
observed rather man loudness recruitment (Wang et aL, 1997).
Loudness recruitment may result from a loss of a population of
34
FECHTER ET AL.
Spiral Ganglion Cell Density
0.005
Basal
Lower Middle
Upper Middle
Basilar Location
FIG. 6. TCE decreases the density of spiral ganglion cells in the lowermiddle and upper-middle turns of the cochlea. Data are presented as group
means (±SE) ('Significantly different from controls, p < 0.05; n = 4/group).
cells that detect weak stimuli and are important for sound
detection but are not important for encoding loudness (stimulus
intensity above threshold). Therefore, once sound levels exceed a critical value, the unaffected cell population generates a
rapid increase in cochlear signal amplitude yielding normal
cochlear output, i.e., increased slope of the input-output function above the threshold elevated by damage (Pickles, 1988).
Indeed, Davis et al. (1958) suggested that these two cell
populations correspond to the inner and outer hair cells with
their associated spiral ganglion cells. The outer hair cells play
a critical role in sound detection at low stimulus intensities,
whereas the more numerous inner hair cells and Type 1 spiral
ganglion cells play a more significant role in determining the
maximal output from the cochlea and thereby the response to
suprathreshold stimuli. The present data showing a lack of
loudness recruitment are consistent with impairment of the
inner hair cell and or Type 1 spiral ganglion cells.
The histopathology of cochleas from TCE-exposed rats
plainly demonstrates a loss of spiral ganglion cells. In addition,
the loss of spiral ganglion cells appears to be restricted to the
lower-middle and upper-middle turns of the cochlea. Surprisingly, this damage to the spiral ganglion cells occurred with
only sporadic and infrequent loss of hair cells as seen in the
midmodiolar sections. Admittedly, definitive evidence for or
against an effect of TCE on hair cells can only be found in
serial reconstructions and/or surface preparations. More importantly, the midcochlear location of damage to the spiral ganglion cells is consistent and supportive of the RMA and CAP
threshold deficits seen at 8 and 16 kHz. Studies relating cochlear function with the labeling of spiral ganglion cells in the
rat indicate that stimuli of 7.7 and 16 kHz are associated with
maximal labeling of spiral ganglion cells 65 and 50% of the
distance along the cochlea from the base (Miiller, 1991). This
too is consistent with the functional data reported here as
higher tone frequencies are known to be encoded at more basal
regions along the cochlea and spiral ganglion cells in this
region appear normal among TCE-exposed subjects.
The histopathological effects reported here for TCE exposure in the rat include the extensive loss of ganglion cells with
only a sparse loss of hair cells. These results are not consistent
with previous reports for toluene and styrene, two other volatile organic solvents that also produce midfrequency hearing
loss in rodent species (Pryor et al, 1987; Mattsson et al., 1990;
Yano et al, 1992; Crofton et al, 1994; Johnson and Canlon,
1994a,b; Campo et al, 1997). Numerous authors have documented the loss of outer hair cells following exposure to
toluene (Pryor et al, 1984; Sullivan et al, 1988; Johnson and
Canlon, 1994b; Campo et al, 1997) and styrene (Yano et al,
1992). Moreover, Campo et al (1997) recently provided descriptive evidence that ganglion cells are spared after toluene
exposure. The reason for the fact that all three solvents would
produce a midfrequency hearing loss, yet produce two different
patterns of morphological damage remains to be determined.
In summary, the present data demonstrate a cochlear site of
action for the ototoxicity of TCE. Inhalation exposure to TCE
results in both functional and structural damage to the cochlea
in the rat. TCE exposure elevated CAP thresholds for midfrequency tones, reduced N, amplitude even at suprathreshold
stimulus intensities, and resulted in a loss of spiral ganglion
cells in the middle cochlear turns. In contrast, TCE exposure
failed to alter the CM. Taken together, these data suggest a
localized cochlear site of action of TCE in the spiral ganglion
cells and/or inner hair cells.
ACKNOWLEDGMENTS
The authors thank Paul Medensky for technical support with the exposure
facilities, John Peoples for technical and photographic assistance with the
histopathology, and Paula Meder for help in manuscript preparation.
REFERENCES
Aran, J. M., and Darrouzet, J. (1975). Observation of click-evoked compound
Vin nerve responses before, during and over seven months after Kanamycin
treatment in the guinea pig. Acta. Ololaryngol 79, 24-32.
Brown, R. D., and Feldman, A. M. (1978). Pharmacology of hearing and
ototoxicity. Annu. Rev. Pharmacol. Toxicol. 18, 233—252.
Campo, P., Lataye, R., Cossec, B., and Placidi, V. (1997). Toluene-induced
hearing loss: A mid-frequency location of the cochlear lesions. Neuroioxi-
col. Teraiol. 19, 129-140.
Clerici, W. J., Ross, B., Jr., and Fechter, L. D. (1991). Acute ototoxicity of
trialkyltins in the guinea pig. Toxicol Appl. Pharmacol. 109, 547-556.
Chen, T. J., and Chen, S. S. (1991). Generator study of brainstem auditory
evoked potentials by a radiofrequency lesion method in rats. Exp. Brain Res.
85, 537-542.
Crofton, K. M. (1990). Reflex modification and the detection of toxicantinduced auditory dysfunction. Neuroloxicol. Teratol. 12, 461-468.
Crofton, K. M., Dean, K. F., Menache, M. G., and Janssen, R. (1990).
Trimethyltin effects on auditory function and cochlear morphology. Toxicol.
Appl. Pharmacol. 10S, 123-132.
Crofton, K. M. (1992). Reflex modification and the assessment of sensory
TRICHLOROETHYLENE OTOTOXICITY
dysfunction. In Neurotoxicology (H. A. Tilson and C. L. Mitchell, Eds.), pp.
181-211. Raven Press, New York.
Crofton, K. M., Lassiter, T. L., and Rebert, C. S. (1994). Solvent-induced
ototoxicity in rats: An atypical selective mid-frequency hearing deficit.
Hear. Res. 80, 25-30.
Crofton, K. M., and Zhao, X. (1993). Mid-frequency hearing loss in rats
following inhalation exposure to trichloroethylene: Evidence from reflex
modification audiometry. Neurotoxicol. Teratol. 15, 413-433.
Crofton, K. M., and Zhao, X. (1997). The ototoxicity of trichloroethylene:
Extrapolation and relevance of high-concentration, short-duration animal
exposure data. Fundam. Appl. Toxicol. 38, 101-106.
Dallos, P., BiUone, M. C , Durrant, J. D., Wand, C. Y., and Raynor, S. (1972).
Cochlear inner and outer hair cells: Functional differences. Science 177,
356-358.
Dallos, P. (1973). The Auditory Periphery, p. 276. Academic Press, New York.
Dallos, P., and Wang, C. Y. (1974). Bioelectric correlates of kanamycin
intoxication. Audiology 13, 277-289.
Dallos, P., and Cheatham, M. A. (1976). Production of cochlear potentials by
inner and outer hair cells. J. Acoust. Soc. Am. 60, 510-512.
Davis, H., Deatherage, B. H., Rosenblut, B., Fernandez, C , Kimura, R., and
Smith, C. A. (1958). Modification of cochlear potentials produced by
streptomycin poisoning and by extensive venous obstruction. Laryngoscope
68, 596-627.
Eggermont, J. J. (1983). Audiologic disorders. In Bases of Auditory Brain-Stem
Evoked Responses (E. J. Moore, Ed), pp. 287-316. Grune & Stratton, New
York.
Fechter, L. D., and Carlisle, L. (1990). Auditory dysfunction and cochlear
vascular injury following trimethyltin exposure in the guinea pig. Toxicol.
Appl. Pharmacol. 105, 133-143.
Fechter, L. D., Clerici, W. J., Yao, L., and Hoeffding, V. (1992). Rapid
disruption of cochlear function and structure by trimethyltin in the guinea
pig. Hear. Res. 58, 166-174.
Gulick, W. L., Gescheider, G A., and Frisina, R. D. (1989). Hearing: Physiological Acoustics, Neural Coding, and Psychoacoustics, pp. 265-268.
Oxford Univ. Press, New York.
Hall, R. D. (1990). Estimation of surviving spiral ganglion cells in the deaf rat
using electrically evoked auditory brainstem response. Hear. Res. 45, 123-236.
Harpur, E. S. (1987). The pathophysiology of hearing. Br. Med. Bull. 43,
871-886.
Henry, K. R., and Lepkowski, C. M. (1978). Evoked potential correlates of
genetic progressive hearing loss. Ada Otolaryngol. 86, 366-374.
Hinners, R. G., Burkhart, J. K., and Punte, C. L. (1968). Animal inhalation
exposure chambers. Arch. Environ. Health 16, 194-206.
Hunter, K. P., and Willow, J. F. (1987). Aging and the auditory brainstem response
in mice with severe or minimal presbycusis. Hear. Res. 30, 207-218.
Ison, J. R. (1984). Reflex modification as an objective test for sensory processing
following toxicant exposure. Neurobehav. Toxicol Teratol 6, 437-445.
Jaspers, R. M., Muijser, H., Lammers, J. H., and Kulig, B. M. (1993). Midfrequency hearing loss and reduction of acoustic startle responding in rats
following trichloroethylene exposure. NeurotoxicoL TeratoL 15. 407-412.
Johnson, A. C , and Canlon, B. (1994a). Toluene exposure affects the functional activity of the outer hair cells. Hear. Res. 72, 189-196.
Johnson, A. C , and Canlon, B. (1994b). Progressive hair cell loss induced by
toluene exposure. Hear. Res. 75, 201-208.
Mattsson, J. L., Gorzinski, S. J., Albee, R. R., and Zimmer, M. A. (1990).
Evoked potential changes from 13 weeks of simulated toluene abuse. Pharmacol. Biochem. Behav. 36, 683-689.
35
Miller, R. R., Letts, R. L., Potts, W. J., and McKenna, M. J. (1980). Improved
methodology for generating controlled test atmospheres. Am. Indust. Hygiene Assoc. J. 41, 844-846.
Moller, A. R. (1983) Auditory Physiology, pp. 94-98. Academic Press, San
Diego.
M0ller, A. R., Jannetta, P. J., and Sekhar, L. N. (1988). Contributions from the
auditory nerve to the brain-stem auditory evoked potentials (BAEPs): Results of intracranial recording in man. Electroencephlogr. Clin. Neurophysiol. 71, 198-211.
MUller, M (1991). Frequency representation in the rat cochlea. Hear. Res. 51,
247-254.
Pickles, J. O (1988). An Introduction to the Physiology of Hearing. Academic
Press, San Diego.
Popper, A. N., and Fay, R. R. (1993). Sound detection and processing by fish:
Critical review and major research questions. Brain, Behav. Evol. 41,
14-38.
Pryor, G. T., Rebert, C. S., Dickinson, J., and Feeney, E. M. (1984). Factors
affecting toluene-induced ototoxicity in rats. Neurobehav. Toxicol. Teratol.
6, 223-228.
Pryor, G. T., Rebert, C. S., and Howd, R. A. (1987). Hearing loss in rats caused
by inhalation of mixed xylenes and styrene J. Appl. Toxicol. 7, 44—61.
Rebert, C S., Day, V. L., Matteucci, M. J., and Pryor, G. T. (1991). Sensoryevoked potentials in rats chronically exposed to trichloroethylene: Predominant auditory dysfunction. Neurotoxicol. Teratol. 13, 83-90.
Rebert, C. S., Boyes, W. K., Pryor, G. T., Svensgaard, D. J., Kassay, K. M.,
Gordon, G. R., and Shinsky, N. (1993). Combined effects of solvents on the
rat's auditory system: Styrene and trichloroethylene. Int. J. Psychophysiol.
14, 49-59.
Ruppert, P. H., Dean, K. F., and Reiter, L. W. (1984). Trimethyltin disrupts
acoustic startle responding in adult rats. Toxicol. Lett. 22, 33—38.
Rybak, L. P. (1986). Drug ototoxicity. Annu. Rev. Pharmacol. Toxicol. 26,
79-99.
Rybak, L. P. (1992). Hearing: The effects of chemicals. Otolaryngol. Head
Neck Surg. 106, 677-686.
SAS (1989). SAS/STAT Users Guide, Version 6, fourth ed., Vol. 2, pp.
891-996. SAS Institute, Cary, NC.
Sellick, P. M., and Russell, I. J. (1980). The responses of inner hair cells to
basilar membrane velocity during low frequency auditory stimulation in the
guinea pig cochlea. Hear. Res. 2, 439-445.
Starr, A., and Zaaroor, M. (1990). Eight nerve contributions to cat auditory
brainstem responses (ABR). Hear. Res. 48, 151-160.
Sohmer, H., Kinarti, R., and Gafni, M. (1980). The source along the basilar
membrane of the cochlear microphonic potential recorded by surface electrodes in man. Electroencephalogr. Clin. Neurophysiol. 49, 506-514.
Sullivan, M. J., Rarey, K. E., and Conolly, R. B. (1988). Ototoxicity of toluene
in rats. Neurotoxicol. Teratol. 10, 525-530.
Wang, J., Powers, N. L., Hofstetter, P., Trautwein, P., Ding, D., and Salvi, R.
(1997). Effects of selective inner hair cell loss on auditory nerve fiber
threshold, tuning and spontaneous and driven after discharge rate. Hear.
Res. 107, 67-82.
Yano, B. L., Dittenber, D. A., Albee, R. R., and Mattsson, J. L. (1992). Abnormal
auditory brainstem responses and cochlear pathology in rats induced by an
exaggerated styrene exposure regimen. Toxicol. PathoL 20, 1-6.
Young, J. S., and Fechter, L. D. (1983). Reflex inhibition procedures for
animal audiometry: A technique for assessing ototoxicity. J. Acoust. Soc.
Am. 73, 1686-1693.