Evaluation of Sensory Evoked Potentials in Long Evans Rats Gestationally Exposed to Mercury (Hg0) Vapor

David W. Herr*,1, Sushmita M. Chanda{dagger},{ddagger}, Jaimie E. Graff*, Stanley S. Barone, Jr*, Robert P. Beliles§ and Daniel L. Morgan{ddagger}

* Neurotoxicology Division, MD B105-05, NHEERL, ORD, U.S. E.P.A., Research Triangle Park, North Carolina 27711; {dagger} Roche Palo Alto LLC, Palo Alto, California 94304-1397; {ddagger} Respiratory Toxicology, NIEHS, Research Triangle Park, North Carolina 27709; and § NCEA, ORD, U.S. E.P.A., Washington, District of Columbia 20460

Received July 2, 2004; accepted July 31, 2004


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Mercury is known to alter neuronal function and has been shown to cross the placental barrier. These experiments were undertaken to examine if gestational exposure to mercury vapor (Hg0) would result in alterations in sensory neuronal function in adult offspring. Dams were exposed to 0 or 4 mg/m3 Hg0 for 2 h/day from gestational days 6–15. This exposure paradigm has been shown to approximate a maximal tolerated dose of Hg0 for the dams. Between postnatal days 140–168, male and female offspring (one of each gender/dam) were examined using a battery of sensory evoked potentials. Peripheral nerve action potentials, nerve conduction velocity, somatosensory evoked responses (cortical and cerebellar), brainstem auditory evoked responses, pattern evoked potentials, and flash evoked potentials were quantified. Gestational exposure to 4 mg/m3 Hg0 did not significantly alter any of the evoked responses, although there was a suggestion of a decrease in compound nerve action potential (CNAP) amplitudes in male animals for the 3 mA stimulus condition. However, this possible change in CNAP amplitudes was not replicated in a second experiment. All evoked potentials exhibited predictable changes as the stimulus was modified. This shows conclusively that the evoked responses were under stimulus control, and that the study had sufficient statistical power to detect changes of these magnitudes. These results indicate that gestational exposure to 4 mg/m3 Hg0 did not result in changes in responses evoked from peripheral nerves, or the somatosensory, auditory, or visual modalities.

Key Words: mercury vapor; evoked potentials; sensory toxicity.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Elemental (metallic) mercury (Hg0) is a highly toxic metal that can cause serious adverse health effects. Because Hg0 has a high vapor pressure, the primary route of human exposure is by inhalation of the Hg0 vapors. Occupational exposure to Hg0 vapor is associated with the chloralkalai industry, combustion of fossil fuels, mercury mining, the refining of gold, and the dental industry. Exposure of the general population to Hg0 vapor occurs primarily through inhalation of Hg0 vapor released from dental amalgams (International Programme on Chemical Safety, 1991Go).

The toxicokinetics of inhaled Hg0 have been well studied. Approximately 80% of inhaled Hg0 vapor is absorbed into the blood as it passes through the pulmonary circulation (Hursh et al., 1976Go). A portion of the dissolved Hg0 can be trapped in erythrocytes. However, Hg0 remains dissolved in the blood long enough for it to be distributed throughout the body. Because of its lipid solubility, Hg0 readily diffuses through cell membranes where it is oxidized by cytosolic catalase-hydrogen peroxide to mercuric mercury (Hg2+), the reactive species for most mercury compounds (Clarkson, 1997Go; International Programme on Chemical Safety, 1991Go). Mercuric Hg rapidly combines with intracellular ligands such as sulfhydryls, potentially disrupting enzymes and proteins essential to normal organ function.

Much less is known about the toxicokinetics of Hg0 vapor in the pregnant female as it relates to fetal development and reproductive outcome. During pregnancy, a number of physiological changes occur that can alter the tissue distribution and therefore the toxicity of xenobiotics (Corley et al., 2003Go). Because it is lipid soluble, Hg0 penetrates the placental barrier (Clarkson et al., 1972Go; Kosta et al., 1975Go; Lutz et al., 1996Go; Warfvinge et al., 1994Go; Yoshida et al., 1990Go) and is taken up by fetal tissues (Morgan et al., 2002Go). As in maternal tissues, oxidation of Hg0 in the fetal tissues converts it to Hg2+, which is less likely to exit the fetus by crossing the placental barrier. As such, Hg2+ from Hg0 vapor can accumulate at higher levels in the fetus than in the mother (Clarkson, 1992Go).

The ability of Hg to accumulate in the fetus has led to concerns about potential developmental toxicity resulting from in utero Hg0 exposure. All forms of Hg administered to animals have been shown to result in developmental problems such as spontaneous abortion, stillbirths, and congenital malformations (for reviews: Barlow and Sullivan, 1982Go; Schuurs, 1999Go). A limited number of poorly documented studies (Baranski and Szymczyk, 1973Go; Steffek et al., 1987Go) indicate that exposure to Hg0 vapor during pregnancy can cause maternal toxicity and adverse effects to the fetus and offspring. In animal studies, Baranski and Szymczyk (1973)Go reported that rat pups exposed prenatally to Hg0 vapor (approximately 2.5 mg/m3, 6 h/day for 21 days before fertilization and over gestational days 7–20), died within 6 days after birth. Other investigators reported that exposure to 1.5 mg/m3 Hg0 vapor for 1 or 3 h/day over gestation days (GD) 6–11 or 13–18 altered levels of nerve growth factor and its receptors on postnatal days (PND) 21 and 60, suggesting that gestational exposure to Hg0 could alter trophic factor's regulation of brain development (Söderström et al., 1995Go). Gestational exposure to Hg0 vapor has also been reported to produce behavioral changes in offspring. Exposure to 1.8 mg/m3 Hg0 vapor over GD 11–14 and GD 17–20 (1 or 3 h/day) produced hypoactivity at three months, decreased performance in a radial arm maze at four months, reduced habituation to activity chambers at seven months, followed by hyperactivity at 14 months of age (Danielsson et al., 1993Go). Another study exposed pregnant dams to 1.8 mg/m3 Hg0 for 1.5 h/day over GD 14–19. These investigators reported that Hg0 exposure resulted in hyperactivity at four months, decreased performance in a swim maze at 4.5 months, and hyperactivity coupled with decreased performance in a radial arm maze at five months of age (Fredriksson et al., 1996Go). These few studies indicate that gestational exposure to Hg0 may alter brain development and can result in long lasting changes in behavior of the offspring.

Furthermore, exposure to Hg0 vapor has been shown to alter electrophysiological endpoints in adult humans. Studies in workers exposed to Hg0 vapor have shown decreased nerve conduction velocities (Albers et al., 1982Go; Andersen et al., 1993Go; Ellingsen et al., 1993Go; Levine et al., 1982Go; Singer et al., 1987Go), altered sensory nerve amplitudes (Albers et al., 1982Go; Ellingsen et al., 1993Go), and alterations in visual evoked responses (Andersen et al., 1993Go; Ellingsen et al., 1993Go). Several of these studies were conducted years after exposure to Hg0 vapor had ended, indicating long-lasting changes in neural function (Albers et al., 1988Go; Andersen et al., 1993Go; Ellingsen et al., 1993Go). Because of the possibility of irreversible changes in neural function following gestational exposure to Hg0, this study used a battery of electrophysiological measures (similar to those shown to be altered in human studies) to examine peripheral nerve function, somatosensory system responses, brainstem auditory system potentials, and visual system evoked potentials in adult rats that had been exposed in utero to Hg0 vapor.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals, exposure, and surgery. Timed pregnant Long Evans rats (225–250 g; Charles River Laboratories, Portage, MI) were received on GD 2 at NIEHS. Gestation day 0 was defined as the day of confirmed mating (presence of vaginal plug). Dams were housed individually in plastic cages with wood chip bedding, and food (NIH-07) and water (deionized, filtered tap water) were provided ad libitum except during Hg0 exposure. Feed contained less than 0.025 ppm total Hg. The animal rooms were maintained on a 12 h light/dark cycle (lights on 0600 h). This study was conducted under federal guidelines for the use and care of laboratory animals and was approved by the NIEHS and U.S.E.P.A Animal Care and Use Committees, which require compliance with National Institute of Health guidelines. Animals were housed in a humidity- and temperature-controlled, high efficiency particulate air (HEPA)-filtered, mass air displacement room in facilities accredited by the American Association for Accreditation of Laboratory Animal Care. Sentinel animals were housed in the animal facility as part of an ongoing surveillance program for parasitic, bacterial, and viral infections and were pathogen-free throughout the study.

On GD 4, the animals were weighed and randomly assigned to one of two treatment groups. The animals were exposed by nose-only to avoid contamination of the fur and subsequent oral and dermal exposure to Hg0 vapor. Control animals were exposed to conditioned air in a stainless steel, 52-port nose-only exposure system (Lab Products, Rockville, MD). A second nose-only exposure system was used for Hg0 vapor exposures. Dams were placed in cylindrical holding tubes during nose-only exposure to either conditioned air or 4 mg/m3 Hg0 vapor for 2 h/day for 10 consecutive days from GD 6 through GD 15. Exposures were limited to 2 h a day to minimize stress caused by restraint during nose-only exposure. This restraint procedure resulted in no significant decreases in body weight gain of the dams or offspring weight. Additionally, this exposure concentration was previously found to result in a maximum tolerated dose of Hg0 (Morgan et al., 2002Go).

Elemental Hg vapor was generated by passing HEPA filtered, charcoal scrubbed, and temperature and humidity controlled (20–24°C; 40–60% humidity) air through a flask containing 10–20 g of Hg0 (CAS# 7439-97-6; Aldrich Chemical Co., Milwaukee, WI; 99.99 + % pure). The flask was placed in a water bath which was maintained at approximately 2°C above ambient temperature, and the airstream containing Hg0 vapor was diluted and delivered to the exposure system using mass flow controllers at about 12 l/min. Exposure concentrations were measured from the nose-only system once every 15–30 min and air samples from the room, scrubber, and the exposure system enclosure were analyzed once every hour. Analysis was accomplished using a Jerome mercury analyzer specific for Hg0 vapor (Model 431-X; Arizona Instruments, Phoenix, AR).

On GD 18, the dams were transferred to the animal facilities at the U.S. E.P.A.. The number of litters in each exposure group was as follows: conditioned air (n = 12), and 4 mg/m3 Hg0 (n = 10). The animals were then housed singly in animal colonies with a 12 h light dark cycle (lights on 0600 h; 22 ± 2°C; 40 ± 20% relative humidity) with food (#5001, Purina Lab Chow, St. Louis, MO) and tap water provided ad libitum. Postnatal day 0 was defined as the day of birth (between 0700 and 1600 h), and PND 1 was defined as the day following birth. On PND 4, the litters were culled to 10 pups (1:1 male:female ratio). The offspring not used in these experiments were used in studies to follow the elimination of Hg from various tissues and to examine possible changes in postnatal brain neurochemistry (manuscript in preparation).

On PND 140–168, one male and one female offspring per dam were surgically implanted with electrodes using procedures that have been described previously (Herr et al., 1992Go, 1994Go; Herr and Boyes, 1997Go). Rats were anesthetized with sodium pentobarbital (50 mg/kg ip) and implanted, using blunt ear bars, with stainless-steel epidural screw electrodes (00–90 x 1/16 in) pre-soldered to nichrome wire which was crimped to gold pins (Dyer et al., 1987bGo). The surface area of the electrode in contact with dura was approximately 0.8 mm2. Electrode locations were as follows: (1) the active electrode for flash evoked potentials (FEP) and pattern evoked potentials (PEP) was 1 mm anterior to lambda and 4 mm to the left of midline, (2) the active electrode for cortical somatosensory evoked potentials (SEPcortex) was 2 mm posterior to bregma and 2 mm to the left of midline, (3) the active electrode for brainstem auditory evoked responses (BAER) and cerebellar somatosensory evoked potentials (SEPcerebellum) was 3 mm posterior to lambda on the midline, (4) a ground electrode was 2 mm anterior to bregma and 2 mm to the left of midline, (5) a reference electrode for FEP and PEP was 2 mm anterior to bregma and 2 mm to the right of midline, and (6) a reference electrode for BAER, SEPcortex, and SEPcerebellum was 7 mm anterior to bregma and 2 mm to the right of midline. The incision was painted with 10% Povidone Iodine creme (E. Fougera & Co., Melville, NY) and closed with wound clips. The rats were allowed approximately a week to recover prior to testing.

Physiological testing. Subjects were allowed to acclimate to the laboratory for at least 15 min prior to initiating testing. The unanesthetized subjects were restrained in a plastic cone (decapicone; Braintree Scientific, Inc., Braintree, MA), their head and pinnae exposed, placed in a custom designed testing apparatus contained inside a sound attenuated Faraday box, and their tails inserted into a teflon stimulation and recording tray, as previously described (Hamm et al., 2000Go; Herr et al., 1996aGo, 1998Go). Stainless steel syringe needles (25-gauge) were used for stimulation of the ventral caudal tail nerves and recording the compound nerve action potential (CNAP). The electrodes were placed at the following distances (in cm) posterior to the hairline of the tail: 12 (second anode), 11 (second cathode), 10 (temperature probe), 9 (first anode), 8 (first cathode), 6 (ground), 4 (active), and 1 (reference). The temperature probe was a 26-gauge needle thermistor (MT-26/2, Physitemp Instruments, Inc., Clifton, NJ), with shielded and grounded wiring connected to a thermometer (Thermalert Model TH-8, Physitemp Instruments, Inc.) located outside the chamber (Faraday cage). Tail temperature was not regulated, but the temperature of the tail in close proximity to the ventral nerves was recorded with each waveform. Sixty seconds were allowed to elapse prior to stimulation to reduce the number of artifacts created by movements of the test subjects. General sensory stimuli and recording conditions are described in Table 1. All EEG signals were amplified 10,000X and averaged using the evoked potential system previously described (Hamm et al., 2000Go). Briefly, the evoked responses were differentially amplified and bandpass filtered (Model 12A5 Neurodata Acquisition System, Grass Instruments, Astro-Med, Inc., West Warwick, RI). The signals were digitized with 12 bit resolution, using a VAX 4000-100 and ADQ32 analog-to-digital conversion boards (Digital Equipment Corporation, Woburn, MA). The amplitude and latency response factors for both the amplifiers and computer were calibrated using sine waves of 178 µV RMS at 3 Hz, 100 Hz, and 3 kHz. The data were analyzed using custom written software (Hamm et al., 2000Go). All evoked potentials were recorded during a single test session in the following order: CNAP, SEPcortex, SEPcerebellum, nerve conduction velocity (NCV), BAER, PEP, and FEP. The total test session was approximately 45 min. Multiple stimulus intensities were used to produce intensity-response functions for the evoked potentials. This procedure was used to demonstrate that the evoked responses were under stimulus control, and that recording conditions were adequate to quantify changes in the evoked potentials of the magnitude produced by the different stimulus intensities. This procedure is recommended in EPA test guidelines (United States Environmental Protection Agency, 1998Go). The specifics regarding the individual stimuli are detailed as follows.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Stimulus and Recording Parameters

 
CNAP, NCV, SEPcortex, and SEPcerebellum. The stimulus for the ventral caudal tail nerve was a constant current biphasic square wave (anodal followed by cathodal, 50 µs each polarity, 100 µs total duration). Current intensities were 1, 2, and 3 mA (each polarity). Stimuli were generated as previously described (Hamm et al., 2000Go; Herr et al., 1998Go) and delivered by a stimulus isolation unit (Model BSI-1; BAK Electronics, Inc., Germantown, MD) which had been custom modified to reduce output electrical noise levels. Prior to initiating the investigation, the electrical stimulus was calibrated using a 20.5 k{Omega} resistor to simulate the rat's tail, the delivered voltage was quantified, and the necessary changes in the system's output voltage were stored to be used as a calibration factor by the stimulus delivery system (Hamm et al., 2000Go). Presentation of the different stimulus intensities was randomized between animals. Stimuli were alternately applied to the first (9 and 8 cm from tail hairline) and second (12 and 11 cm from tail hairline) pairs of stimulating electrodes. A CNAP, a SEPcortex, and a SEPcerebellum were simultaneously recorded following stimulation from the first pair of electrodes. After stimulation using the second pair of electrodes, only a CNAP was recorded for calculation of NCV. Tail temperature was recorded following each stimulus intensity. Data for one control animal was lost due to operator error, thus only 11 animals have data for the CNAP, NCV, SEPcortex, and SEPcerebellum endpoints.

BAERs. The stimuli used were rarefaction clicks and simulated filtered clicks centered at 4 and 16 kHz (all at 50, 65, and 80 dBpeak SPL; re: 20 µPa), and 64 kHz (65, 70, and 80 dBpeak SPL). Calibration of auditory stimuli was performed at ear level in the test chamber using a Brüel & Kjær Measuring Amplifier (Model 2636) with a 0.635 cm microphone (Model 4135; Brüel & Kjær, Marlborough, MA) and a 200 Hz high-pass filter (Model 3343; Krohn-Hite, Avon, MA). Stimuli were generated as previously described (Hamm et al., 2000Go; Herr et al., 1998Go). The signal duration varied among the stimuli: 50.0 µs (click), 2.54 ms (4 kHz), 634 µs (16 kHz), and 159 µs (64 kHz). The speaker was placed approximately 10.7 cm in front of and 25.5 cm above the animal's auditory canals. This resulted in a distance of about 27.7 cm between the speaker and the auditory canals, at an angle of approximately 67°. The acoustic travel time was approximately 825 µs. The stimuli were presented in a random order between animals.

PEPs. The stimuli were patterns of alternating light and dark vertical bars with sinusoidal spatial illuminance profiles. The bars comprised 0.080, 0.16, or 0.32 cpd of visual angle, and had 15, 30, or 60% contrast at each spatial frequency. Calibration was performed using a photometer (Model DR-2550 with Model DR-2550-2B multiprobe, Model TC284 photometric filter, Model 700-3B fiber optic probe, and/or Model DR-2550-18 cosine detector, Gamma Scientific, Inc., San Diego, CA) as previously described (Hamm et al., 2000Go). Percent contrast is calculated as (((Lmax Lmin)/(Lmax + Lmin)) * 100), where Lmax and Lmin are the maximum and minimum illuminance values of the light and dark bars, respectively. The stimuli were sinusoidally modulated in an appearance/disappearance manner with a mean illuminance of 50 lux. The monitor was centered 17.5 cm directly in front of the animal's head. The stimuli were presented in a random order between subjects. Each subject's averaged waveform was processed with a fast Fourier transform (Bergland and Dolan, 1979Go) using a Kaiser-Bessel window (Gade and Herlufsen, 1988Go; Harris, 1978Go; Kaiser, 1969Go) to obtain a spectrum.

FEPs. The stimulus was a 10 µs flash generated by a photic stimulator (Model PS22, Grass Instrument Division, Astro-Med, Inc., West Warwick, RI) enclosed in a foam-lined and electrically shielded wooden box mounted approximately 37 cm above the animal's head. Three flash intensities were used: 16, 56, and 191 lux-s (strobe settings 1, 4, and 16). Stimulus intensities were presented in a counterbalanced order between animals and chambers. Ambient illumination of the test chamber at about 19 lux was produced by an overhead DC light bulb, resulting in relative flash intensities (RFIs) of 49, 55, and 60 dB (Herr et al., 1991Go). Acoustic white noise (80 dB SPL, flat response from 0 to 15 kHz with an approximate 30 dB linear decrease from 15–20 kHz) was generated (Model 1405, Brüel & Kjær), amplified (Model SM-100, Opamp Labs, Inc., Los Angeles, CA), and delivered through a 10 cm cone speaker placed approximately 57 cm above the rat's ear level. The white noise was present during FEP testing to mask any auditory potentials produced by the strobe discharge (Herr et al., 1996bGo; Shaw, 1992Go). Flash intensity and ambient illumination were quantified using a photometer (Model DR-2550 with Model DR-2550-18 Cosine Probe, Gamma Scientific). Calibration of the masking auditory white noise was performed at the animal's ear level in the test chamber using a 1.27 cm microphone (Model 4166, Brüel & Kjær) connected to a measuring amplifier (Model 2636, Brüel & Kjær).

Colonic temperature. Due to the known influence of temperature on evoked potentials (de Jesus et al., 1973Go; Hetzler et al., 1988Go; Hetzler and Dyer, 1984Go; Janssen et al., 1991Go; Miyoshi and Goto, 1973Go; Petajan, 1968Go), colonic temperature was quantified immediately following the animal's removal from the test chamber. A temperature probe (Model RET-1; Physitemp Instruments, Inc., Clifton, NJ), connected to a thermometer (Model BAT-10, Physitemp Instruments, Inc.), was inserted approximately 8 cm rectally and deep colonic temperature was recorded. These data were used to ascertain if changes in body temperature could be related to alterations in evoked potentials.

Replicate experiment. In order to confirm or refute the putative changes in CNAPs in male rats that were observed in the main experiment, a second cohort of animals was generated. Based on the magnitude of the changes in peak N1 and P2 amplitudes observed in male rats of the main study, sample size calculations were performed to provide an {alpha} = 0.025 (p-value observed in main experiment) and a ß = 0.80. These calculations indicated that an n-size of 18 animals should provide adequate power to detect the previously observed decreased in peak N1 amplitude. The second cohort consisted of 20 litters for each of the air and 4 mg/m3 Hg0 vapor groups, which were generated in the same manner as the rats described previously. The animals were implanted with electrodes at approximately PND 133, allowed to recover for one week, and had CNAPs, NCV, and colonic temperature recorded using the procedures described for the main experiment. Other electrophysiological endpoints were not recorded due to the lack of possible changes produced by Hg0 vapor exposure in the earlier study.

Statistical analysis. Peak amplitudes and latencies were measured from each animal's average waveform. Peak amplitudes (in µV) were measured from baseline (defined as the average voltage over the pre-stimulus period). Peak latencies (in ms) were calculated from stimulus onset. Peaks were identified by their polarity and latency according to the average waveform from each treatment group. For CNAPs, the duration and area of the negative peak (N1) were also quantified. The duration was calculated as the time interval between the ascending and descending sides of peak N1 and the intersection with a horizontal line extending from the baseline. The area of peak N1 was calculated as the negative area that occurred over the duration of peak N1. Nerve conduction velocity (in m/s) was calculated as the latency between the first positive peak of the two CNAPs divided by the distance between the two cathodes (3 cm) (Gagnaire et al., 1986Go). If there was no clear initial positive peak in the CNAP, the latency was calculated at a point immediately before the initial negative deflection in the CNAP. The NCV was calculated for the 3 mA stimulus only. For the PEP spectra, the 1F and 2F (first and second harmonic of the stimulation frequency) were quantified. Data were analyzed using a repeated measures analysis of variance (ANOVA; PROC GLM) (SAS Institute Inc., 1989Go, 1997Go) using a Greenhouse-Geisser correction factor ({varepsilon}) (Geisser and Greenhouse, 1958Go; Greenhouse and Geisser, 1959Go; Keselman and Rogan, 1980Go) for degrees of freedom for within-subject and/or litter effects. The dose of Hg0 was a between-litter factor, and stimulation condition and gender were within-litter factors. A significant main effect of treatment, or significant interaction(s) of gender and/or stimulus condition with treatment were followed by step-down ANOVAs, that examined treatment effects at each stimulus condition and/or gender. The critical {alpha} level for the ANOVAs was determined for each evoked response using a Bonferroni correction. Peak amplitudes and latencies had an overall {alpha} = 0.025 (0.05/2), which was further adjusted based on the number of peak amplitudes and latencies which were analyzed (e.g., critical {alpha} for peak amplitudes or latencies = 0.025/number of peaks). Any step-down ANOVAs also used Bonferroni adjusted critical {alpha} levels. This procedure was used to minimize the number of Type I statistical errors and maintains the family-wise {alpha} ≤ 0.05 (Abt, 1981Go; Muller et al., 1983Go). Due to the potentially decreased statistical power produced by these adjustments (Muller et al., 1983Go), if an overall effect was significant but subsequent step-down ANOVAs failed to reach the corrected significance level (but had an p ≤ 0.05), the actual probability values are reported. This allows the readers to apply their own judgement as to the biological significance of the results. Group mean comparisons were performed using a Tukey-Kramer multiple comparison test ({alpha} = 0.05) (Kramer, 1956Go). Data are reported as mean ± SE. Group averaged waveforms were calculated from individual animal data, and are presented for illustrative purposes.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
General Health
Body weight. The effects of exposure to 4 mg/m3 Hg0 on the body weight gain of the dams and the neonatal weights on PND1 have been previously reported (Morgan et al., 2002Go), and are not reproduced in this article. The 4 mg/m3 Hg0 treatment results in the exposed dams gaining approximately 7% less weight than controls (Morgan et al., 2002Go). In the offspring, there were no treatment-related differences in body weight either at the time of surgery or at testing for male or female animals (p-values ≥0.1099). As expected, male rats weighed more than female subjects (F[1,20] = 572.04, p ≤ 0.0001). There was a slight indication that female animals recovered faster from surgery than males (time by gender interaction: F[1,20] = 4.74, p = 0.0417). At surgery, male rats weighed 560.6 ± 18.8 and 566.2 ± 12.6 g, while female rats weighed 290.4 ± 7.3 and 282.2 ± 6.6 g, for control and Hg0-exposed groups, respectively. At testing, male rats weighed 548.1 ± 16.2 and 558.5 ± 14.4 g, while female rats weighed 293.3 ± 6.9 and 276.2 ± 7.5 g, for control and Hg0-exposed groups, respectively.

Electrophysiological Measures
CNAP, NCV, SEPcortex, and SEPcerebellum. There were no significant (critical {alpha} = 0.0083; 0.025/3 peaks) treatment-related changes in CNAP peak amplitudes (Fig. 1). However, there was a nonsignificant (p = 0.0472) suggestion of a treatment by gender by stimulus intensity interaction for peak P2 amplitude. Further analysis suggested that any treatment-related changes in CNAP waveforms occurred only in male rats using a 3 mA stimulus for the amplitudes of peaks N1 (F[1,19] = 5.89, p = 0.0253) and P2 (F[1,19] = 9.57, p = 0.0060) (Table 2). There were no indications of treatment-related changes in the area of peak N1 (p-values ≥ 0.05).



View larger version (28K):
[in this window]
[in a new window]
 
FIG. 1. Average CNAP and waveforms showing NCV between the two stimulation sites (n = 10–11 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following tail stimulation with a 1, 2, or 3 mA stimulus (3 mA only for NCV calculations). CNAP waveforms are plotted with negativity upward. The shaded regions represent the 95% amplitude confidence intervals for the waveforms. For ease of presentation in this and all other figures, waveforms whose confidence limits overlap with controls are displayed without confidence regions in all panels. There was a suggestion that treatment with Hg0 vapor decreased the amplitude of CNAP peaks N1 and P2 in male animals when a 3 mA stimulus was used, and increased the latencies of peaks P1, N1, and P2 when stimulated with an intensity of 2 or 3 mA. NCV was not affected by Hg0 treatment. Female rats had greater amplitude CNAPs than males. Increasing stimulus intensities resulted in larger CNAP peak amplitudes and decreased peak latencies.

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 CNAP Peak Amplitudes and Latencies

 
There were no significant (critical {alpha} = 0.0083) treatment-related changes in the latencies of CNAP peaks (Fig. 1). However, there was a nonsignificant (p-values ≥ 0.0267) suggestion of a treatment main effect (averaged over genders and stimulus intensities) to increase the latencies of peaks P1, N1, and P2. Further step-down analysis indicated that these effects were only present in male animals for the 2 and 3 mA stimulus conditions (p-values ≥ 0.0139). There were no significant effects of Hg0 vapor exposure indicated for the peak N1 duration for either male or female rats under any of the stimulus conditions (p-values ≥ 0.05).

There were significant gender differences and effects of stimulus intensity on CNAP peaks (Fig. 1). Female animals had greater peak P2 amplitude than male rats (gender effect: F[1,19] = 23.98, p ≤ 0.0001) and the latency for peaks P1 and N1 was less in females than in males (gender effect: F-values[1,19] ≥ 16.32, p-values ≤ 0.0007). Larger stimulus intensities produced greater amplitudes for peaks P1 and P2 (stimulus intensity effect: F-values[2,38] ≥ 6.51, p-values ≤ 0.0055, {varepsilon}-values ≥ 0.8805), and resulted in a decrease in peak P1 latency (stimulus intensity effect: F[2,38] = 11.06, p = 0.0002, {varepsilon} = 0.9895), when averaging male and female data. Additionally, female rats had a greater rate of increase for peak N1 amplitude and peak N1 area than male rats as the stimulus intensity increased (gender by stimulus intensity interaction: F-values[2,38] ≥ 12.08, p-values ≤ 0.0003, {varepsilon}-values ≥ 0.8024).

Gestational exposure to Hg0 vapor had no effect on NCV in the ventral caudal tail nerves (Fig. 1) of male or female rats (p-values > 0.05). This conclusion was sustained when tail temperature was used as a covariate in the analysis (p-values > 0.05). The NCV of male rats was 29.7 ± 0.8 and 28.3 ± 1.8 m/s for control and treated subjects, respectively. The NCV of female rats was 30.1 ± 1.1 and 30.4 ± 1.1 m/s for control and treated subjects, respectively. The tail nerve temperature averaged 22.7 ± 0.2 and 22.8 ± 0.3°C for male and female rats, respectively.

There were no significant effects (all p-values > 0.05) of gestational exposure to Hg0 vapor on the amplitudes or latencies of peaks in the SEPcortex recordings (Fig. 2). However, the waveforms were altered by the different stimulus intensities. Increasing the stimulus intensity resulted in increased amplitudes for peaks N27 and N53, and increased the latency of peak P74 (stimulus intensity effect: F-values[2,38] ≥ 9.61, p-values ≤ 0.0009, {varepsilon}-values ≥ 0.8216). There were few gender-related differences between the SEPcortex waveforms, but females had a shorter P14 latency than observed in males (gender effect: F[1,19] = 15.18, p = 0.0010).



View larger version (30K):
[in this window]
[in a new window]
 
FIG. 2. Average SEPcortex waveforms (n = 10–11 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following tail stimulation with a 1, 2, or 3 mA stimulus. Waveforms are plotted with negativity upward. The shaded regions represent the 95% amplitude confidence intervals for the control waveforms. Treatment with Hg0 did not alter any SEPcortex peak amplitude or latency. Increasing the stimulus intensity increased the peak amplitudes, and decreased peak latency. Additionally, peak P14 latency was less in females than in males.

 
Gestational exposure to Hg0 vapor did not significantly (critical {alpha} = 0.0063; 0.025/4 peak amplitudes or latencies) alter peak amplitudes or latencies of SEPcerebellum waveforms (Fig. 3). Stimulus-related changes in the waveforms were indicated by larger peak P55 amplitudes with greater stimulus intensities (stimulus intensity effect: F[2,38] = 49.72, p ≤ 0.0001, {varepsilon} = 0.8274). Additionally, gender-related differences were indicated by reduced peak N10 latencies in females compared to males (gender effect: F[1,19] = 15.39, p = 0.0009).



View larger version (28K):
[in this window]
[in a new window]
 
FIG. 3. Average SEPcerebellum waveforms (n = 10–11 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following tail stimulation with a 1, 2, or 3 mA stimulus. Waveforms are plotted with negativity upward. The shaded regions represent the 95% amplitude confidence intervals for the control waveforms. Treatment with Hg0 did not alter any SEPcerebellum peak amplitude or latency. Increasing the stimulus intensity resulted in larger peak P55 amplitudes. Additionally, peak N10 latency was less in females than in males.

 
BAERs. Treatment with Hg0 vapor did not produce significant (critical {alpha} = 0.0036; 0.025/7 peak amplitudes or latencies) changes to peak amplitudes or latencies in BAER waveforms following auditory stimulation with rarefaction clicks or 4, 16, or 64 kHz tone pips (Fig. 4). There were only minor gender-related differences observed in BAER waveforms. When averaged over stimulus intensities, females had a larger peak P4 amplitude than males when a rarefaction click (gender effect: F-value[1,20] = 12.24, p-value = 0.0023) or 64 kHz tone pip (gender effect: F-value[1,20] = 15.33, p-value = 0.0009) were used as the auditory stimuli.



View larger version (51K):
[in this window]
[in a new window]
 
FIG. 4. Average BAER waveforms (n = 10–12 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following stimulation with a rarefaction click, or 4, 16 (50, 65, 80 dB SPL), or 64 kHz (65, 70, 80 dB SPL) tone pips. Waveforms are plotted with positivity upward. The shaded regions represent the 95% amplitude confidence intervals for the control waveforms. Treatment with Hg0 did not alter any BAER peak amplitude or latency. Increasing the stimulus intensities resulted in larger BAER waveforms, and decreases in BAER peak latencies.

 
The auditory responses were under stimulus control, as evidenced by the intensity-related changes in the BAER waveforms. As louder rarefaction click stimuli were presented, the amplitude of peaks P1, P2, P3, and P4 increased, while the amplitude of peak N3 decreased (more positive value) (stimulus intensity effect: F-values[2,40] ≥ 202.63, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.9275). Also, as greater stimulus intensities were used, the latencies of peaks P1, N1, P2, N2, P3, N3, P4 decreased (stimulus intensity effect: F-values[2,40] ≥ 134.55, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.8381). When the animals were presented with a 4 kHz tone pip, the amplitudes of peaks P1, P2, P3, and P4 increased, and the amplitudes of peaks N2 and N3 decreased (greater positive value), with louder stimuli (stimulus intensity effect: F-values[2,40] ≥ 27.80, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.7071). As the 4 kHz tone pip intensity increased, the latencies of peaks P1, N1, P2, N2, P3, N3, and P4 decreased (stimulus intensity effect: F-values[2,40] ≥ 212.20, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.7560). When a 16 kHz tone pip was used as a stimulus, the amplitudes of peaks P1, P2, P3, and P4 increased, while the amplitude of peak N3 decreased (greater positive value), with louder stimuli (stimulus intensity effect: F-values[2,40] ≥ 62.73, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.8173). Additionally, the latencies of peaks P1, N1, P2, N2, P3, N3, and P4 decreased (stimulus intensity effect: F-values[2,40] ≥ 55.55, p-values ≤ 0.0001, {varepsilon}-values ≥ 0.6343) as the stimulus intensity increased. Finally, when a 64 kHz tone pip was used as the auditory stimulus, the amplitudes of peaks P1, P3, and P4 increased, and the amplitudes of peaks N2 and N3 (greater positive value) decreased, with louder stimuli (stimulus intensity effect: F-values[2,40] ≥ 9.81, p-values < 0.0005, {varepsilon}-values ≥ 0.9361). Also, the latencies of peaks P1, N1, N2, N3, and P4 decreased (stimulus intensity effect: F-values[2,40] ≥ 11.64, p-values < 0.0007, {varepsilon}-values ≥ 0.6203) with increasing stimulus intensity for the 64 kHz tone pip.

PEPs. Gestational exposure to 4 mg/m3 Hg0 vapor did not significantly (critical {alpha} = 0.0250; 0.05/2 peak amplitudes) alter the 1F or 2F amplitudes resulting from steady state pattern visual stimulation (Fig. 5). The biological responses were under stimulus control, as both the 1F and 2F peak amplitudes increased with increasing stimulus contrast (contrast effect: F-values[2,40] ≥ 5.95, p-values ≤ 0.0105, {varepsilon}-values ≥ 0.7773). Further evidence for stimulus control of the responses is indicated by the observation that the 2F amplitude was less at 0.32 cpd than at the other spatial frequencies (spatial frequency effect: F[2,40] = 15.70, p ≤ 0.0001, {varepsilon} = 0.8569). There were no gender-related differences in the pattern-evoked visual responses (all p-values ≥ 0.05).



View larger version (29K):
[in this window]
[in a new window]
 
FIG. 5. Average PEP amplitude spectra (n = 10–12 rats/spectra) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following stimulation with sinusoidally modulated light and dark vertical bar stimuli with spatial frequencies of 0.08, 0.16, or 0.32 cpd using 60, 30, or 15% contrast. The shaded regions represent the 95% amplitude confidence intervals for the control spectra. Treatment with Hg0 did not alter the 1F or 2F amplitudes. The biological responses were under stimulus control as the amplitudes were altered by the different stimulus contrast levels and/or the spatial frequencies.

 
FEPs. Gestational exposure to 4 mg/m3 Hg0 vapor did not produce significant (critical {alpha} = 0.0031; 0.025/8 peak amplitudes or latencies) changes to peak amplitudes or latencies in FEP waveforms following visual stimulation with a flash stimulus (Fig. 6). However, the biological responses were under stimulus control as the amplitudes and latencies of the FEP peaks differed over the flash intensities. The amplitudes of peaks P28, N36, P59, P72, and N166 increased, while the amplitudes of peaks N65 and N84 decreased (due to increased net positivity in this region of the waveform) as the flash intensity increased (stimulus intensity effect: F-values[2,40] ≥ 8.33, p-values ≤ .0030, {varepsilon}-values ≥ 0.6895). The latencies of peaks P28, N36, and N166 decreased, while the latency of peak P72 tended to increase, with greater flash intensities (stimulus intensity effect: F-values[2,40] ≥ 6.89, p-values ≤ 0.0031, {varepsilon}-values ≥ 0.7248). There were no significant gender-related differences in FEP peak amplitudes or latencies.



View larger version (32K):
[in this window]
[in a new window]
 
FIG. 6. Average FEP waveforms (n = 10–12 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following stimulation with a 16, 56, and 191 lux-s light flash. Waveforms are plotted with negativity upward. The shaded regions represent the 95% amplitude confidence intervals for the control waveforms. Treatment with Hg0 did not result in significant changes in FEP peak amplitudes or latencies. The visual responses were under stimulus control, as greater flash intensities generally resulted in larger amplitudes and reduced peak latencies.

 
Colonic temperature. There were no treatment-related differences in colonic temperature at the conclusion of physiological testing (all p-values > 0.05). Control animals colonic temperatures were 38.1 ± 0.2 and 37.5 ± 0.3 while the Hg0-exposed animals colonic temperatures were 37.7 ± 0.5 and 37.9 ± 0.3°C for males and females, respectively.

Replicate Experiment
CNAP and NCV. In the replicate experiment, gestational exposure to 4 mg/m3 Hg0 did not alter (p-values > 0.05) the amplitude of any of the CNAP peaks nor the area of peak N1 (Fig. 7, Table 2). Female animals had significantly (critical {alpha} = 0.0083; 0.025/3 peak amplitudes or latencies) greater peak amplitudes than males for CNAP peaks N1 and P2 (gender effect: F-values[1,38] ≥ 187.47, p-values ≤ 0.0001).



View larger version (23K):
[in this window]
[in a new window]
 
FIG. 7. Average CNAP and waveforms showing NCV between the two stimulation sites (n = 20 rats/waveform) for animals gestationally exposed to conditioned air or 4 mg/m3 Hg0 vapor following tail stimulation with a 3 mA stimulus from the replicate experiment. Waveforms are plotted with negativity upward. The shaded regions represent the 95% amplitude confidence intervals for the waveforms. Treatment with Hg0 did not result in significant changes in CNAP peak amplitudes or latencies, peak N1 area or duration, or NCV. Female rats had greater amplitude CNAPs than males.

 
Similar to the results for CNAP amplitudes, treatment with 4 mg/m3 Hg0 did not significantly (critical {alpha} = 0.0083) change the latencies for any of the CNAP peaks (Fig. 7). The duration of peak N1 was not changed by gestational exposure to Hg0 vapor (p-values > 0.05). A slight difference between the genders was indicated as females had a shorter peak P1 latency than males (gender effect: F[1,38] = 9.18, p = 0.0044).

Comparison of the CNAP peaks for the control animals (Table 2) from the two experiments indicated significant (critical {alpha} = 0.0083) differences for the latencies of peaks N1 (study effect: F[1,29] = 18.32, p = 0.0002) and P2 (study effect: F[1,29] = 15.38, p = 0.0005) only in female animals.

In concordance with the results of the main experiment, gestational exposure to Hg0 vapor had no effect on the NCV in the ventral caudal tail nerve (Fig. 7) of male or female offspring (p-values > 0.05). Again, this conclusion was sustained when tail temperature was used as a covariate in the analysis (p-values > 0.05). The NCV in males was 27.9 ± 0.6 and 27.9 ± 0.4 m/s for control and treated subjects, respectively. The NCV in female rats was 29.0 ± 0.6 and 29.7 ± 0.6 m/s for control and treated subjects, respectively. The tail nerve temperature averaged 20.9 ± 0.1 and 20.8 ± 0.1°C for male and female rats, respectively.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Gestational exposure to 4 mg/m3 Hg0 vapor for 2 h/day over GD 6–15 did not alter compound nerve action potentials, nerve conduction velocity, cortical or cerebellar somatosensory potentials, brainstem auditory evoked responses, or visual evoked potentials in male or female offspring when tested as adults. Exposure to this treatment regime has been shown to represent a maximally tolerated dose for pregnant dams, resulting in a slight reduction in weight gain (Morgan et al., 2002Go). In contrast to maternal effects, there were no treatment-related differences in body weight of the offspring in this study and all animals appeared healthy. The animals examined in the main study were litter mates of a subset of the dams used in the study by Morgan and coworkers (Morgan et al., 2002Go), so the exposure paradigm was identical and slight effects produced by Hg0 on the dams were present. On PND 1, fetal brains contained approximately 20 ng/g total Hg, as previously reported (Morgan et al., 2002Go). Thus, the lack of effects on the electrophysiological measures should not be interpreted as a lack of exposure of the fetus to Hg0.

Exposure to Hg0 vapor has been shown to produce changes in electrophysiological measures in humans. Occupational and epidemiological studies often involve years of exposure to Hg0 vapor. For example, studies of chloralkali workers utilized people previously exposed to a mean concentration of 59 µg Hg/m3 (range: 10–162 µg Hg/m3) for an average duration of 7.9 years (Andersen et al., 1993Go; Ellingsen et al., 1993Go). Other investigations have involved workers with ongoing Hg0 exposure, and have estimated exposure levels based on urinary or blood Hg levels (Albers et al., 1982Go; Levine et al., 1982Go; Singer et al., 1987Go), making direct comparisons to the exposures in the current rodent study difficult. The human studies have frequently reported decreases in nerve conduction velocity (Albers et al., 1982Go; Andersen et al., 1993Go; Ellingsen et al., 1993Go; Levine et al., 1982Go; Singer et al., 1987Go), and some investigators have reported decreases and/or increases in the amplitude of nerve action potentials (Albers et al., 1982Go; Ellingsen et al., 1993Go). Increases in the latencies of visual evoked potentials have also been reported following occupational exposure to Hg0 vapor (Andersen et al., 1993Go; Ellingsen et al., 1993Go; Shikata et al., 1998Go). Decreased latencies for cortical somatosensory evoked potentials, but no changes in brainstem auditory evoked responses, were also reported (Ellingsen et al., 1993Go). In contrast, other researchers have reported increased conduction times for brainstem auditory and somatosensory evoked potentials in workers exposed to Hg0 vapor (Chang et al., 1995Go). Unfortunately, comparable electrophysiological assessments of neural function have not been performed in animal models of adult or developmental exposure to Hg0 vapor.

Experimental data does show that exposure to Hg0 vapor produces neuroanatomical changes in adult animals, and alters cognitive and motor function after gestational treatments. Adult rats exposed to about 480 µg/m3 Hg0 for 5 h/day, 4–5 days/week, for 8 weeks had decreased cross-sectional myelin area of neurons in the dorsal root ganglion cells, and a tendency for a reduction in axonal area (Schiønning et al., 1998Go). This same treatment regime produced loss of cerebellar Purkinje and granule cells, and a reduction in the granule cell layer volume (Sørensen et al., 2000Go). Prenatal exposure to Hg0 vapor has been reported to produce behavioral changes in rodent models (Danielsson et al., 1993Go; Fredriksson et al., 1996Go). These changes consisted of alterations in motor activity (most frequently hyperactivity) and decreased performance in radial arm and swimming mazes. Exposure of pregnant squirrel monkeys to 500–1000 µg Hg/m3 increased the variability in the performance of offspring in concurrent random-interval schedules of reinforcement. One of the treated monkeys underwent electroretinographic examination. No changes in the rod or cone response of this monkey were observed (Newland et al., 1996Go). Although these studies suggest that exposure to Hg0 vapor may produce neuronal damage and alter cognitive performance, the lack of assessment of sensory neuronal function prevents comparison to the data presented in this article.

This article is the first known report of electrophysiological assessment of sensory function in adult rats after in utero exposure to Hg0 vapor. Unlike reports of occupational exposure to Hg0 in humans, no treatment-related changes in visual, somatosensory, or auditory evoked potentials, or nerve conduction velocity were observed. The first experiment suggested a decrease in the amplitude of tail compound nerve action potentials in male rats, but only at the largest stimulus intensity (3 mA). In contrast to this gender-related difference, female mice have been shown to have a greater accumulation of Hg than male mice in motor neurons after ip injection with 0.5 mg/kg HgCl2 (Pamphlett et al., 1997Go). Additionally, changes in Hg0-related CNAPs were not replicated in the second experiment. The lack of replication of Hg0-related changes was not due to differences in the control CNAPs between the two studies, as the only significant differences between control responses were in peak N1 and P2 latencies in female animals. These endpoints were not suggested to be altered by Hg0 exposure in the first study. Therefore, the decreased CNAP amplitudes suggested in the first study may represent a Type I error.

Our data suggest that the neurophysiological function of sensory systems in rodents is not a primary toxicological target for gestational exposure to levels up to 4 mg/m3 Hg0 vapor. Other investigators have used whole body exposure (vs. nose-only exposure) of dams to Hg0 vapor and reported altered performance in radial arm and swim mazes in the offspring (Danielsson et al., 1993Go; Fredriksson et al., 1996Go). Our data may be useful in interpreting these behavioral changes, as it suggests that the animals were not likely to have large sensory deficits that decreased their ability to utilize extra-maze cues (D'Hooge and De Deyn, 2001Go; Dudchenko et al., 1997Go) during testing. It is interesting to note that brain levels of Hg on PND 1 from litter mates of the animals tested in the main experiment (plus some additional neonates) contained approximately 20 ng/g total Hg (Morgan et al., 2002Go). Litter mates of animals reported to show deficits in performance in maze tasks had brain levels of Hg in the range of 5–12 ng/g on PND 2–3 (Danielsson et al., 1993Go; Fredriksson et al., 1996Go). This again argues that the behavioral changes were not related to large changes in sensory neural function.

A concern with negative results relates to the ability of the dependant measures to adequately detect biologically relevant differences. These procedures are able to detect significant differences in evoked responses produced by manipulation of the stimulus parameters (Herr et al., 1991Go, 1995Go, 1998Go, 2001Go) or due to gender differences (Herr et al., 1998Go, 2001Go). Predictable stimulus-related changes in peak amplitudes and latencies of evoked responses were observed for CNAPs, SEPcortex, SEPcerebellum, BAERs, PEPs, and FEPs. Gender-related differences were seen for the amplitudes and latencies of CNAPs, the latencies of SEPcortex and SEPcerebellum peaks, and the amplitude of certain BAER peaks. These results replicate our previous findings of reduced latencies for early peaks in SEPcortex and SEPcerebellum potentials, and CNAP peak latencies, in females compared to males (Herr et al., 2001Go). The greater CNAP peak amplitudes in females compared to males that we observed also replicates previous work in our laboratory (Herr et al., 1998Go, 2001Go). One possible explanation for the reduced SEP peak latencies in females may relate to their smaller body size compared to males. A smaller distance between the stimulating (tail location) and recording (head location) electrodes would result in a gender-related reduction in peak latency. However, this possibility does not apply to the CNAP peak latencies, as the inter-electrode distances were controlled. Additionally, we have previously reported that differences in tail diameter did not correlate with gender-related differences in CNAP peak amplitudes or latencies (Herr et al., 1998Go). Therefore, although the biological basis of the gender-related differences in evoked responses has not been conclusively identified these differences are highly replicable in our laboratory. Together, these results demonstrate that the responses were under stimulus control and that sufficient statistical power was present to detect differences of the magnitude produced by varying the stimulus intensity or between genders.

The neurophysiological endpoints that were quantified involved many portions of the neuroaxis. The CNAPs and NCV examined the peripheral tail nerves directly. Somatosensory evoked potentials were recorded from the cortex and cerebellum. Changes in these responses could reflect dysfunction in the peripheral nerves, spinal cord, cerebellum, thalamus, or cortical regions (for reviews see: Herr and Boyes, 1995Go; Mattsson et al., 1992Go; Mattsson and Albee, 1988Go; Rebert, 1983Go). The brainstem auditory evoked responses indicate neurotransmission from the auditory nerve through the lateral lemniscus (Hall, 1992Go; Mattsson et al., 1992Go; Rebert, 1983Go; Shaw, 1988Go). The visual evoked potentials (flash and pattern) are generated in the visual cortex (Barth et al., 1995Go; Brankack et al., 1990Go; Dyer et al., 1987aGo,bGo; Schroeder et al., 1991Go). Changes in these visual evoked responses could reflect alterations in the retina, lateral geniculate nucleus, sub-cortical regions, or cortical processing. Although we have shown changes in FEPs after retinal alterations produced by 3,3'-iminodipropionitrile (IDPN) (Barone et al., 1995Go; Herr et al., 1995Go), it is possible that small changes in retinal function may not be detected by FEPs due to cortical magnification of the visual response. However, it is unlikely that gestational exposure to Hg0 vapor produced large changes in sensory neuronal function that were not detected with the battery of tests used in these studies. One possibility (that we have not examined) is that gestational exposure to Hg0 vapor could produce alterations in sensory function in neonates or juvenile rats that recovers by adulthood. These experiments were designed to test for irreversible changes in neurophysiological function by testing the animals as adults. As such, reversible changes that may have occurred at an earlier life stage would have not been detected.

We have used the same electrophysiological procedures to detect changes in visual, auditory, and peripheral nerve function in rodent models. Previous studies have shown alterations in FEPs after ip injection of 115 mg/kg dichloromethane, 86 mg/kg 1,3-dichloropropane, 105 mg/kg 1,2-dichlorobenzene (Herr and Boyes, 1997Go), 400 mg/kg/day IDPN for three days (Herr et al., 1995Go), or 200 mg/kg carbon disulfide (CS2) (Herr et al., 1992Go). We have reported that gestational exposure to 1, 4, or 8 mg/kg/day Aroclor 1254 altered BAERs when the animals were tested as adults (Herr et al., 1996aGo). We have also shown changes in peripheral nerve function using electrical stimulation of caudal tail nerves (as in this study for SEPcortex, SEPcerebellum, CNAP and NCV) after inhalation exposure to 500–800 ppm CS2 for 13 weeks (Herr et al., 1998Go). However, even though these techniques have been shown to have sufficient sensitivity to detect alterations in evoked responses after exposure to xenobiotics, subtle changes in the function of the peripheral nerve, somatosensory, auditory, or visual systems cannot be completely excluded.

In summary, these experiments indicate that gestational exposure (GD 6–15) to 4 mg/m3 Hg0 vapor, a dose previously shown to produce decreased rate of maternal weight gain and exposure of the fetal brain to Hg (Morgan et al., 2002Go), failed to significantly alter compound nerve action potentials, nerve conduction velocity, somatosensory, brainstem auditory, or visual evoked potentials when the animals were tested as adults.


    ACKNOWLEDGMENTS
 
The authors wish to thank Dr. Timothy Shafer and Dr. Robert Sills for helpful discussions and critique of the manuscript. We also wish to thank Mr. C. Hamm for the design and implementation of the software used for electrophysiological recordings.


    NOTES
 
The information in this document has been funded wholly (or in part) by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.

Portions of this manuscript were presented as a poster at the 38th annual meeting of the Society of Toxicology (1999, Toxicol. Sci. 48(Suppl. 1), 242).

1 To whom correspondence should be addressed at 109 T.W. Alexander Drive, MD B105-05, NHEERL/NTD/NPTB, U.S. Environmental Protection Agency, Research Triangle Park, NC 27711. Fax: (919) 541-4849. E-mail: Herr.david{at}epamail.epa.gov.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Abt, K. (1981). Problems of repeated significance testing. Controlled Clin. Trials 1, 377–381.[CrossRef][ISI][Medline]

Albers, J. W., Cavender, G. D., Levine, S. P., and Langolf, G. D. (1982). Asymptomatic sensorimotor polyneuropathy in workers exposed to elemental mercury. Neurology 32, 1168–1174.[Abstract]

Albers, J. W., Kallenbach, L. R., Fine, L. J., Langolf, G. D., Wolfe, R. A., Donofrio, P. D., Alessi, A. G., Stolp-Smith, K. A., Bromberg, M. A., and Mercury Workers Study Group (1988). Neurological abnormalities associated with remote occupational elemental mercury exposure. Ann. Neurol. 24(5), 651–659.[ISI][Medline]

Andersen, A., Ellingsen, D. G., Morland, T., and Kjuus, H. (1993). A neurological and neurophysiological study of chloralkali workers previously exposed to mercury vapour. Acta Neurol. Scand. 88(6), 427–433.[ISI][Medline]

Baranski, B., and Szymczyk, I. (1973). Effects of mercury vapor on reproductive functions of female white rats. Med. Pr. 24(3), 249–261.

Barlow, S. M., and Sullivan, F. M. (1982). Mercury and its compounds (inorganic). In Reproductive Hazards of Industrial Chemicals: An Evaluation of Animal and Human Data (S. M. Barlow and F. M. Sullivan, Eds.), pp. 368–403. Academic Press, New York.

Barone, S., Jr., Herr, D. W., and Crofton, K. M. (1995). Effects of 3,3'-iminodipropionitrile on the peripheral structures of the rat visual system. Neurotoxicology 16(3), 451–468.[ISI][Medline]

Barth, D. S., Goldberg, N., Brett, B., and Di, S. (1995). The spatiotemporal organization of auditory, visual, and auditory-visual evoked potentials in rat cortex. Brain Res. 678(1–2), 177–190.[CrossRef][ISI][Medline]

Bergland, G. D., and Dolan, M. T. (1979). Fast Fourier Transform Algorithms. Programs for Digital Signal Processing, p. 1.2–1–1.2–18. IEEE Press, New York.

Brankack, J., Schober, W., and Klingberg, F. (1990). Different laminar distribution of flash evoked potentials in cortical areas 17 and 18b of freely moving rats. J. Hirnforsch. 31(4), 525–533.[ISI][Medline]

Chang, Y. C., Yeh, C. Y., and Wang, J. D. (1995). Subclinical neurotoxicity of mercury vapor revealed by a multimodality evoked potential study of chloralkali workers. Am. J. Ind. Med. 27(2), 271–279.[ISI][Medline]

Clarkson, T. W. (1992). Symposium review: Toxicity assessment of mercury vapor from dental amalgams. Fundam. Appl. Toxicol. 19, 319–329.[CrossRef][ISI][Medline]

Clarkson, T. W. (1997). The toxicology of mercury. Crit. Rev. Clin. Lab. Sci. 34, 369–403.[ISI][Medline]

Clarkson, T. W., Magos, L., and Greenwood, M. R. (1972). The transport of elemental mercury into fetal tissues. Biol. Neonate 21(239), 244.

Corley, R. A., Mast, T. J., Carney, E. W., Rogers, J. M., and Daston, G. P. (2003). Evaluation of physiologically based models of pregnancy and lactation for their application in children's health risk assessments. Crit. Rev. Toxicol. 33(2), 137–211.[ISI][Medline]

Danielsson, B. R. G., Fredriksson, A., Dahlgren, L., Teiling Gårdlung, A., Olsson, L., Dencker, L., and Archer, T. (1993). Behavioral effects of prenatal metallic mercury inhalation exposure in rats. Neurotoxicol. Teratol. 15(6), 391–396.[CrossRef][ISI][Medline]

D'Hooge, R., and De Deyn, P. P. (2001). Applications of the Morris water maze in the study of learning and memory. Brain Res. Rev. 36, 60–90.[ISI][Medline]

de Jesus, P. V., Jr., Hausmanowa-Petrusewicz, I., and Barchi, R. L. (1973). The effect of cold on nerve conduction of human slow and fast nerve fibers. Neurology 23, 1182–1187.[ISI][Medline]

Dudchenko, P. A., Goodridge, J. P., Seiterle, D. A., and Taube, J. S. (1997). Effects of repeated disorientation on the acquisition of spatial tasks in rats: Dissociation between appetitive radial arm maze and aversive water maze. J. Exp. Psychol. 23(2), 194–210.[ISI]

Dyer, R. S., Clark, C. C., and Boyes, W. K. (1987a). Surface distribution of flash-evoked and pattern reversal-evoked potentials in hooded rats. Brain Res. Bull. 18(2), 227–234.[CrossRef][ISI][Medline]

Dyer, R. S., Jensen, K. F., and Boyes, W. K. (1987b). Focal lesions of visual cortex – Effects on visual evoked potentials in rats. Exp. Neurol. 95, 100–115.[CrossRef][ISI][Medline]

Ellingsen, D. G., Mørland, T., Andersen, A., and Kjuus, H. (1993). Relation between exposure related indices and neurological and neurophysiological effects in workers previously exposed to mercury vapor. Br. J. Indust. Med. 50(8), 736–744.[ISI][Medline]

Fredriksson, A., Dencker, L., Archer, T., and Danielsson, B. R. G. (1996). Prenatal coexposure to metallic mercury vapor and methylmercury produce interactive behavioural changes in adult rats. Neurotoxicol. Teratol. 18(2), 129–134.[CrossRef][ISI][Medline]

Gade, S., and Herlufsen, H. (1988). Windows to FFT analysis. Sound and Vibration 22, 14–22.

Gagnaire, F., Nicot, T., Zissu, D., Bonnet, P., and de Ceaurriz, J. (1986). Assessment of tail nerve function in rats chronically exposed to vinyltoluene. Toxicol. Lett. 30, 27–34.[CrossRef][ISI][Medline]

Geisser, S., and Greenhouse, S. W. (1958). An extension of Box's results on the use of the F distribution in multivariate analysis. Annal. Mathemat. Stat. 29, 885–891.[ISI]

Greenhouse, S. W., and Geisser, S. (1959). On methods in the analysis of profile data. Psychometrika 24(2), 95–112.[ISI]

Hall, J. W., III (1992). Handbook of Auditory Evoked Responses, pp. 1–871. Allyn and Bacon, Boston.

Hamm, C. W., Ali, J. S., and Herr, D. W. (2000). A system for simultaneous multiple subject, multiple stimulus modality, and multiple channel collection and analysis of sensory evoked potentials. J. Neurosci. Meth. 102(2), 95–108.[CrossRef][ISI][Medline]

Harris, F. J. (1978). On the use of windows for harmonic analysis with the discrete fourier transform. Proc. IEEE 66(1), 51–83.[ISI]

Herr, D. W., and Boyes, W. K. (1997). A comparison of the acute neuroactive effects of dichloromethane, 1,3-dichloropropane, and 1,2-dichlorobenzene on rat flash evoked potentials (FEPs). Fundam. Appl. Toxicol. 35(1), 31–48.[CrossRef][ISI][Medline]

Herr, D. W., and Boyes, W. K. (1995). Electrophysiological analysis of complex brain systems: Sensory-evoked potentials and their generators. In Neurotoxicology. Approaches and Methods (L. W. Chang and W. Slikker, Jr., Eds.), pp. 205–221. Academic Press, New York.

Herr, D. W., Boyes, W. K., and Dyer, R. S. (1991). Rat flash-evoked potential peak N160 amplitude: Modulation by relative flash intensity. Physiol. Behav. 49(2), 355–365.[CrossRef][ISI][Medline]

Herr, D. W., Boyes, W. K., and Dyer, R. S. (1992). Alterations in rat flash and pattern reversal evoked potentials after acute or repeated administration of carbon disulfide (CS2). Fundam. Appl. Toxicol. 18(3), 328–342.[ISI][Medline]

Herr, D. W., Goldey, E. S., and Crofton, K. M. (1996a). Developmental exposure to Aroclor 1254 produces low-frequency alterations in adult rat brainstem auditory evoked responses. Fundam. Appl. Toxicol. 33(1), 120–128.[CrossRef][ISI][Medline]

Herr, D. W., Graff, J. E., Derr-Yellin, E. C., Crofton, K. M., and Kodavanti, P. R. S. (2001). Flash-, somatosensory-, and peripheral nerve-evoked potentials in rats perinatally exposed to Aroclor 1254. Neurotoxicol. Teratol. 23(6), 591–601.[CrossRef][ISI][Medline]

Herr, D. W., King, D., Barone, S., Jr., and Crofton, K. M. (1995). Alterations in flash evoked potentials (FEPs) in rats produced by 3,3'-iminodipropionitrile (IDPN). Neurotoxicol. Teratol. 17(6), 645–656.[CrossRef][ISI][Medline]

Herr, D. W., King, D., Griffin, V. T., Watkinson, W. P., Boyes, W. K., Ali, J. S., and Dyer, R. S. (1994). Within-session changes in peak N160 amplitude of flash evoked potentials in rats. Physiol. Behav. 55(1), 83–99.[CrossRef][ISI][Medline]

Herr, D. W., Vo, K. T., King, D., and Boyes, W. K. (1996b). Possible confounding effects of strobe "clicks" on flash evoked potentials in rats. Physiol. Behav. 59(2), 325–340.[CrossRef][ISI][Medline]

Herr, D. W., Vo, K. T., Morgan, D. L., and Sills, R. C. (1998). Carbon disulfide neurotoxicity in rats: VI. Electrophysiological examination of caudal tail nerve compound action potentials and nerve conduction velocity. Neurotoxicology 19(1), 129–146.[ISI][Medline]

Hetzler, B. E., Boyes, W. K., Creason, J. P., and Dyer, R. S. (1988). Temperature-dependent changes in visual evoked potentials of rats. Electroencephalogr. Clin. Neurophysiol. 70, 137–154.[CrossRef][ISI][Medline]

Hetzler, B. E., and Dyer, R. S. (1984). Contribution of hypothermia to effects of chloral hydrate on flash evoked potentials of hooded rats. Pharmacol. Biochem. Behav. 21(4), 599–607.[ISI][Medline]

Hursh, J. B., Clarkson, T. W., Cherian, M. G., Vostal, J. V., and Mallie, R. V. (1976). Clearance of mercury (197Hg,203Hg) vapour inhaled by human subjects. Arch. Environ. Health 31, 302–309.[ISI][Medline]

International Programme on Chemical Safety (1991). Inorganic Mercury, pp. 1–168. World Health Organization, Geneva.

Janssen, R., Hetzler, B. E., Creason, J. P., and Dyer, R. S. (1991). Differential impact of hypothermia and pentobarbital on brain-stem auditory evoked responses. Electroencephalogr. Clin. Neurophysiol. 80, 412–421.[CrossRef][ISI][Medline]

Kaiser, J. F. (1969). Digital Filters. In System Analysis by Digital Computer (F. F. Kuo, Ed.), pp. 218–285. Wiley, New York.

Keselman, H. J., and Rogan, J. C. (1980). Repeated measures F tests and psychophysiological research: Controlling the number of false positives. Psychophysiology 17(5), 499–503.[ISI][Medline]

Kosta, L., Byrne, A. R., and Zelenko, V. (1975). Correlation between selenium and mercury in man following exposure to inorganic mercury. Nature 254, 238–239.[ISI][Medline]

Kramer, C. Y. (1956). Extension of multiple range tests to group means with unequal numbers of replications. Biometrics 12, 307–310.[ISI]

Levine, S. P., Cavender, G. D., Langolf, G. D., and Albers, J. W. (1982). Elemental mercury exposure: peripheral neurotoxicity. Br. J. Indust. Med. 39(2), 136–139.[ISI][Medline]

Lutz, E., Lind, B., Herin, P., Karkau, I., Bui, T. I., and Vahter, M. (1996). Concentration of mercury, cadmium and lead in brain and kidney of second trimester fetuses and infants. J. Trace Elem. Exp. Med. 10, 61–67.[CrossRef][ISI]

Mattsson, J. L., and Albee, R. R. (1988). Sensory evoked potentials in neurotoxicology. Neurotoxicol. Teratol. 10(5), 435–443.[CrossRef][ISI][Medline]

Mattsson, J. L., Boyes, W. K., and Ross, J. F. (1992). Incorporating evoked potentials into neurotoxicity test schemes. In Neurotoxicology (H. Tilson and C. Mitchell, Eds.), pp. 125–145. Raven Press, New York.

Miyoshi, T., and Goto, I. (1973). Serial in vivo determinations of nerve conduction velocity in rat tails. Physiological and pathological changes. Electroencephalogr. Clin. Neurophysiol. 35, 125–131.[CrossRef][ISI][Medline]

Morgan, D. L., Chanda, S. M., Price, H. C., Fernando, R., Liu, J., Brambila, E., O'Conner, R. W., Beliles, R. P., and Barone, S., Jr. (2002). Disposition of inhaled mercury vapor in pregnant rats: Maternal toxicity and effects on developmental outcome. Toxicol. Sci. 66(2), 261–273.[Abstract/Free Full Text]

Muller, K. E., Otto, D. A., and Benignus, V. A. (1983). Design and analysis issues and strategies in psychophysiological research. Psychophysiology 20(2), 212–218.[ISI][Medline]

Newland, M. C., Warfvinge, K., and Berlin, M. (1996). Behavioral consequences of in utero exposure to mercury vapor: Alterations in lever-press durations and learning in squirrel monkeys. Toxicol. Appl. Pharmacol. 139(2), 374–386.[CrossRef][ISI][Medline]

Pamphlett, R., Ewan, K. B. R., McQuilty, R., and Waley, P. (1997). Gender differences in the uptake of inorganic mercury by motor neurons. Neurotoxicol. Teratol. 19(4), 287–293.[CrossRef][ISI][Medline]

Petajan, J. H. (1968). Changes in rat ventral caudal nerve conduction velocity during cold exposure. Am. J. Physiol. 214(1), 130–132.[Free Full Text]

Rebert, C. S. (1983). Multisensory evoked potentials in experimental and applied neurotoxicology. Neurobehav. Toxicol. Teratol. 5, 659–671.[ISI][Medline]

SAS Institute Inc. (1989). SAS/STAT User's Guide, Vol. 2, Version 6, Fourth Edition, pp. 1–846. SAS Institute Inc., Cary, NC.

SAS Institute Inc. (1997). SAS/STAT Software: Changes and Enhancements through Release 6.12, pp. 1–1167. SAS Institute Inc, Cary, NC.

Schiønning, J. D., Larsen, J. O., and Eide, R. (1998). A stereological study of dorsal root ganglion cells and nerve root fibers from rats exposed to mercury vapor. Acta Neuropathol. 96(2), 185–190.[CrossRef][ISI][Medline]

Schroeder, C. E., Tenke, C. E., Givre, S. J., Arezzo, J. C., and Vaughan, H. G. Jr. (1991). Striate cortical contribution to the surface-recorded pattern-reversal VEP in the alert monkey. Vision Res. 31(7/8), 1143–1157.[CrossRef][ISI][Medline]

Schuurs, A. H. B. (1999). Reproductive toxicity of occupational mercury. A review of the literature. J. Dentistry 27, 249–256.[CrossRef][ISI][Medline]

Shaw, N. A. (1988). The auditory evoked potential in the rat – A review. Prog. Neurobiol. 31, 19–45.[CrossRef][ISI][Medline]

Shaw, N. A. (1992). Auditory potentials elicited by the Grass photic stimulator in the rat. Physiol. Behav. 52(2), 401–403.[CrossRef][ISI][Medline]

Shikata, E., Mochizuki, Y., Oishi, M., and Takasu, T. (1998). A case of chronic inorganic mercury poisoning with progressive intentional tremor and remarkably prolonged latency of P300. Rinsho Shinkeigaku 38(12), 1064–1066.[Medline]

Singer, R., Valciukas, J. A., and Rosenman, K. D. (1987). Peripheral neurotoxicity in workers exposed to inorganic mercury compounds. Arch. Environ. Health 42(4), 181–184.[ISI][Medline]

Söderström, S., Fredriksson, A., Dencker, L., and Ebendal, T. (1995). The effect of mercury vapor on cholinergic neurons in the fetal brain: Studies on the expression of nerve growth factor and its low- and high-affinity receptors. Dev. Brain Res. 85, 96–108.[ISI][Medline]

Sørensen, F. W., Larsen, J. O., and Eide, R. (2000). Neuron loss in cerebellar cortex of rats exposed to mercury vapor: a stereological study. Acta Neuropathol. 100(1), 95–100.[CrossRef][ISI][Medline]

Steffek, A. J., Clayton, R., Siew, C., and Verrusio, A. C. (1987). Effects of elemental mercury vapor exposure on pregnant Sprague-Dawley rats. J. Dent. Res. 66, 239 (Abstract).

United States Environmental Protection Agency (1998). Health Effects Test Guidelines. OPPTS 870.6855. Neurophysiology: Sensory Evoked Potentials. EPA 712-C-98-242.

Warfvinge, K., Hua, J., and Lögdberg, B. (1994). Mercury distribution in cortical areas and fiber systems of the neonatal and maternal adult cerebrum after exposure of pregnant squirrel monkeys to mercury vapor. Environ. Res. 67, 196–208.[CrossRef][ISI][Medline]

Yoshida, M., Satoh, H., Kojima, S., and Yamamura, Y. (1990). Retention and distribution of mercury in organs of neonatal guinea pigs after in utero exposure to mercury vapor. J. Trace Elements Exp. Med. 3, 219–226.





This Article
Abstract
FREE Full Text (PDF)
All Versions of this Article:
82/1/193    most recent
kfh246v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Herr, D. W.
Articles by Morgan, D. L.
PubMed
PubMed Citation
Articles by Herr, D. W.
Articles by Morgan, D. L.