ZPT-related distal axonopathy: Behavioral and electrophysiologic correlates in rats

Neurotoxicologyand Teratology, Vol. 12, pp. 153-159. ©Pergamon Press plc, 1990. Printed in the U.S.A.

0892-0362/90 $3.00 + .00

ZPT-Related Distal Axonopathy: Behavioral and Electrophysiologic Correlates in Rats J. F. R O S S A N D G. T. L A W H O R N Human and Environmental Safety Division, Miami Valley Laboratories, P.O. Box 398707 The Procter & Gamble Company, Cincinnati, O H 45239-8707 R e c e i v e d 5 July 1989

ROSS, J. F. AND G. T. LAWHORN. ZPT-related distal axonopathy: Behavioral and electrophysiologic correlates in rats. NEUROTOXICOL TERATOL 12(2) 153-159, 1990.--These experiments examined the relationship between behavioral and electrophysiologic signs of neuromuscular dysfunction in rats with zinc pyridinethione (ZPT)-induced neurotoxicity. ZPT added to the diet of adult rats at 50 ppm produced a moderate (approximately 200 g) reduction in forelimb and hindlimb grip strength which occurred during the second week of dosing. Other behavioral tests of peripheral nervous system toxicity were affected inconsistently. Electrophysiologic changes included a marked (maximum 95%) reduction of indirect muscle-evoked potential (M-wave) and a decrement (maximum 30%) during repetitive M-wave elicitation. Electrophysiologic changes were greater in hindlimb than in forelimb. Needle electromyography revealed denervation potentials in ZPT-treated rats which appeared after M-wave changes and recovered faster than did M-wave amplitude. Caudal nerve conduction velocity was unaffected, indicating that peripheral neurotoxic effects were confined to the neuromuscular junction. When ZPT exposure was discontinued, grip strength recovered in about 1 week. In contrast, electrophysiologic measures required 42 days to recover completely. These results indicate that deficits in neuromuscular junction physiology are a sensitive index of ZPT-related distal axonopathy. Rat

Zinc pyridinethione

Distal axonopathy

Neurophysiology

THE addition of zinc pyridinethione (ZPT) or certain other pyrithiones to the diet of rats at levels -->50 ppm produces signs of neurotoxicity characterized behaviorally by progressive hindlimb weakness and morphologically by a distal axonopathy highlighted by the intraneuronal accumulation of tubulovesicular profiles. Behavioral and morphologic signs of toxicity begin during the second week of dosing. If dosing is continued past this time, weakness becomes more generalized, and pathology spreads centfipitally from its origin in the neuromuscular junction. On the basis of subjective evaluations it has been suggested that, upon termination of exposure the weakness reverses rapidly--within several days to a few weeks. The time required for reversal depends on the initial severity of the weakness (14, 15, 17-19). The electrophysiologic correlates of the muscle weakness are not clear. It has been reported that rats eating a diet with 250 ppm ZPT for 10 days required a greater than normal stimulation current on the sciatic nerve to evoke indirect muscle potentials (M-waves) comparable to those of controls, and that the resulting contractile force was less than that of controls (18). However, in those studies there is insufficient detail to determine: 1) at what stage of dosing/recovery the animals were tested, 2) the degree of weakness at the time of testing and 3) the actual amplitude of the M-wave responses. Moreover, there are additional diagnostic tests (e.g., repetitive M-wave elicitation) which are commonly used clinically and experimentally as electrophysiologic correlates of muscle weakness.

Neuromuscular junction

Behavior

The validation of an electrophysiologic correlate of pyrithionerelated neuromuscular dysfunction would provide an attractive alternative to current safety testing procedures for this category of chemicals. Significant ZPT-related lesions are seen by electron microscopy in neuromuscular junction well before axonal swellings can be demonstrated by light microscopy in the associated nerve trunks (14). Thus, current NOEL testing based on the most sensitive morphologic method requires a necropsy and is both time consuming and expensive. The minimally invasive procedures used in the present experiments allow each animal to be used repeatedly, thus reducing animal use for onset/recovery studies. Furthermore, the data are available on the day of testing, so that safety decisions and follow-up experiments can be organized quickly. Thus, the following series of experiments had two objectives: 1) to demonstrate the electrophysiologic correlates of ZPT-induced neuromuscular dysfunction and 2) to determine if electrophysiologic tests could provide a sensitive and efficient test method for safety testing of ZPT. METHOD

Animals Male rats of the Fischer 344 strain, weighing approximately 250 g (90 days of age) at the beginning of the study, were obtained from Charles River, Kingston, NY. Animals were housed indi-

153

154

ROSS AND LAWHORN

Control

vidually in stainless steel cages, identified with individual ear tags, and maintained on a 12-hour light-dark cycle (0700-1900). Powdered lab chow (Purina, St. Louis, MO; abbreviated PLC) and water were available ad lib.

ZPT-Treated

Behavioral Tests Food intake was measured by weighing feed cups every 3--4 days. Trays were checked daily for gross amounts of spillage, which was a rare occurrence, and added to feed cup weight. All other behavioral tests were conducted " b l i n d . " Rats were tested in random order, determined each day, just prior to testing. Grip strength was measured 3 days/week (Monday/Wednesday/Friday). Other behavioral tests were conducted twice weekly (Tuesday/ Thursday). Forelimb/hindlimb grip strength (FL/HL-GS). Grip strength was measured by standard apparatus/procedures (8). The gauges were calibrated weekly by holding them vertically and placing weights of known value (100, 500 and 1,000 g) on the triangular metal-grasping bar. Each rat was tested three times by obtaining one forelimb and one hindlimb measurement from all rats during three cycles through all subjects. For each rat the median of the individual measurements was designated the daily score. Extensor postural thrust. A modified version of the " s p l a y " test (5,21) was used. Rats were suspended vertically in air (approximately 50 cm above a table top) with the hand around the rats' thorax from behind. Body weight support was removed suddenly, although the hand was kept loosely around the thorax. The normal response is bilateral extension of hindlimbs. The rat (and the tester's hand) continues to fall until it reaches a table top, where the hindlimbs support body weight firmly and steadily. The response was quantitated by pressing the rats' feet into modeling clay which covered the table top. We measured the distance between the footpad proximal to the 3rd digit on each foot. The median of three successive tests was used as the daily score for each rat. Tail-lift hindlimb extensor response. This test was developed to provide an early, quantitative measure of "scissoring" or hindlimb adduction which has been reported after methylmercury exposure (7). The test was performed by lifting the rats' hindquarters by the base of the tail to an approximate 45-degree angle with respect to the table surface. The normal response was bilateral hindiimb extension. Interhock distance was measured with a ruler to the nearest 5 nun. The median of three successive tests was used as the daily score for each rat.

Electrophysiologic Tests Anesthesia was induced with a 5% isoflurane/95% oxygen mixture and maintained at approximately 2.5% isoflurane/97.5% oxygen. Gas was delivered though a nosecone and evacuated to a class A hood. The rats were prepared for testing by shaving the hair from the thigh to the hock with small animal clippers. A small area of skin overlying the thoracic vertebrae was also shaved. Core temperature and subcutaneous temperature over the tail, forelimb and/or thigh were monitored with a rectal probe [Yellow Springs Instrument Co. (YSI) Model 402] or subcutaneous thermal probe (YSI Model 41TA) respectively, connected to a temperature monitor (YSI Model 41). The animals were tested while lying on a circulating heat pad (approximately 37°C). Animals were tested in lateral recumbency with the appropriate (superior) limb held in full extension with a restraining device. The tail was housed in a tail tray which lay on a circulating heat pad. Electrophysiologic tests were conducted with a digital electromyography system (Nicolet Viking, Madison, WI) and consisted

~"

Repetitive M-Wave

'°°°'

ZPT-Treated Flbrlllation Potentials

, 1

11 2o~, J

I

I '_..I

FIG. 1. Representative examples of evoked (A--C) and spontaneous (D-E) waveforms. Small arrows indicate the location of stimulation artifact. (A) M-wave recorded from extensors of hock; note differences in amplitude gain needed to record M-waves for ZPT-treated rats and control rats. (B) Repetitive M-wave; note change in time base from (A). (C) Mixed nerve response recorded from caudal tail nerve; top response evoked from distal site, bottom response evoked from proximal site. (D) Fibrillation potentials (fast waves <100 microvolts) recorded from extensors of hock in pyrithione-treated rats. (E) Positive sharp waves [large waves with initial fast positive (downward) component] recorded from extensors of hock in pyrithione-treated rats.

of: 1) needle electromyography, 2) determination of current intensities for maximum-evoked motor responses, 3) repetitiveevoked motor responses, and 4) mixed nerve conduction velocity. Needle electromyography was performed on extensor muscles of the tarsus (primarily gastrocnemius and soleus) with a concentric needle electrode. A cadmium/steel alligator clip (Mueller No. 30) was attached to the skin overlying the thoracic vertebrae as a ground electrode, and electrical contact was facilitated when necessary by the application of electrode paste (Nicolet No. 016-701200). The active electrode was a 20 mm long, 0.4 mm diameter concentric needle electrode (Nicolet No. 019-721300) inserted into the group of extensor muscles of the hock in the approximate middle of the muscle belly. The presence of insertional activity (normal response of the muscle to insertion and movement of the recording electrode) and spontaneous activity indicative of denervated muscles or myopathy (e.g., positive waves and fibrillation potentials; Fig. 1D and E) were recorded. Evoked muscle responses were recorded from flexor muscles of

FUNCTIONAL CORRELATES OF ZPT TOXICITY

the carpus (primarily flexor carpi ulnaris and digital flexors) and/or extensors of tarsus during stimulation of the brachial plexus or sciatic nerve at the hip, respectively. For recording of forelimbevoked potentials, a cathodal stimulating electrode (TEKA MF12; 12 mm long) was inserted under the axilla into the area of the brachial plexus. The anode was a cadmium/steel alligator clip (Mueller No. 30) attached to the skin between the shoulder blades. Recording electrodes were three cadmium/steel alligator clips (Mueller No. 30) attached to the skin overlying the caudal aspect of the humerus (ground), proximal lateral footpad (reference) and the approximate midbelly of the flexor muscles of the carpus, 20 mm proximal from the reference electrode (pickup). For recording of hindlimb-evoked muscle responses, the greater trochanter was palpated and the cathodal-stimulating electrode was inserted just caudal to this landmark and directed toward the sciatic nerve to a depth of approximately 5 mm. The anode was attached to the skin overlying the hip approximately 5 mm caudal to the cathode. Recording electrodes were attached to the skin overlying the thigh (ground), calcaneal tuber (reference) and approximate midbelly of the extensor muscles of the tarsus, 20 or 25 mm proximal from the calcaneal tuber (pickup), depending on age/size of rat. The resulting potentials were (Fig. 1A) diphasic with an initial large negative (upward) component and peak-to-peak amplitude of about 80-100 mV, which is consistent with published data (6), For repetitive stimulation tests, stimulation intensity was supramaximal (defined as current 25% greater than that which evoked a maximal-evoked potential). The responses to 10 consecutive stimuli at a rate of 5 Hz were recorded (Fig. 1B). The decrement was defined as the mean % amplitude change of the 4th and 5th responses compared with the first response in the series. In control rats, a small facilitation (negative decrement) of approximately 2.5-5% was observed (Figs. 1B, 3 and 5). Mixed nerve-evoked potentials were recorded from the longitudinal tail nerve at the tall base during stimulation at a distal tall and midtail sites. Ten percent platinum/iridium subdermal/EEG electrodes (Nicolet No. 019-737900; 12 mm long) were used at all sites and were positioned just lateral to the longitudinal nerve/ blood vessels. The recording site at the tail base consisted of reference (proximal), pickup (middle) and ground (distal) electrodes separated by 5 ram. The stimulating electrode pairs consisted of proximal (cathode) and distal electrodes (anode) separated by 5 ram. Stimulation intensity was 4.0 mA. The cathodes of the distal and midtail sites were separated from the pickup electrode at the tallbase by 100 mm and 50 ram, respectively. The potentials from 50 stimuli (7/sec) at each site were recorded, averaged and displayed as a single waveform with an initial negative (upward) deflection and a peak-to-peak amplitude of approximately 50-120 microvolts (proximal stimulation site; Fig. 1C, lower trace) or 5-20 microvolts (from distal stimulation site; Fig. 1C, upper trace). The amplitude and shape of the evoked response is consistent with published data (13). Latency of the evoked response was calculated as the time from stimulation to the takeoff of the first negative deflection. Caudal tail-mixed nerve conduction velocity was calculated by dividing the latency from the distal stimulating site by 100 mm. Proximal tail-mixed nerve conduction velocity was calculated by subtracting the proximal latency from the distal latency and dividing by 50 mm. The mixed nerve conduction velocity was the average of caudal and proximal values. Predose conduction velocity of approximately 35 m/sec for 90-day-old Fischer 344 rats was consistent with previously published data (16). Chemical Zinc pyridinethione was obtained from Olin (batch No. 8704-

155

DP5). Experimental diets were prepared by fast adding the appropriate amount of test material to approximately one tenth of the projected total weight of prepared diet. This premix was combined by hand. The remaining rat chow was mixed in a Hobart model A-200 mixer at a speed of approximately 45 rpm. The premix was slowly added over the surface of the feed and mixing continued for approximately 1 hour. Experimental Design General. Rats were divided (stratified by body weight) into groups (n = 6, unless specified). During one week of preexposure testing, all rats had free access to powdered laboratory chow (PLC). During the exposure phase of the experiment, CONTROL rats had access to PLC. Experimental rats ate chow containing ZPT 7 days/week. Exposure to test compounds was continued until at least one group exhibited a hindlimb grip strength deficit of >200 g relative to the CONTROL group. Experiment 1 The purpose of Experiment 1 was to determine the time course of onset and recovery for behavioral and electrophysiologic indicators of ZPT-related neuromuscular toxicity. The experimental group ate PLC containing 50 ppm ZPT for 14 days. On the following day, all rats had access to PLC for the remainder of the experiment. Electrophysiologic testing was conducted after 4, 9 and 14 days of exposure (days 5, 10 and 15) and after 7, 14, 28 and 42 days of recovery. Conduction velocity was not determined on the 42nd day of recovery. Experiment 2 The purpose of Experiment 2 was to determine if any of the behavioral or electrophysiologic changes seen in Experiment 1 were secondary to loss of body weight. Experimental rats ate chow containing 50 ppm ZPT for eight days. A third group (FOOD DEP) was restricted to 10 g feed/day. Electrophysiologic testing was conducted on days 9 and 10, beginning immediately after grip strength testing on day 9. Experiment 3 On the basis of gross behavioral observations, it has been suggested that ZPT produces primarily hindlimb weakness (14,18). The purpose of Experiment 3 was to compare the degree of electrophysiologic changes in forelimb function with those of the hindlimb. Experimental rats (n = 8) ate PLC containing 50 ppm ZPT for 11 days. Four control rats ate PLC. The repetitive M-wave test was conducted on forelimbs and hindlimbs after 6 and 10 days of dosing and after 18 days of recovery. Statistics All preexposure values for each subject were averaged. Group changes from this average baseline were compared for each dose day. If variances were equal by the F-test, comparisons were made by Student's t-test; otherwise, the groups were compared by a rank-sum test. Adjustment for multiple tests was made by the sequential Bonferroni-type procedure outlined by Tnkey et al. (22). In brief, p values for a set of H-correlated observations were ranked from smallest to largest (i.e., from P1 to PH) and corrected sequentially by the formula: Ptxl = 1 - ( 1 - Px)(H-x+1)1/2

156

ROSS AND LAWHORN

350 .--.

320.

r:

29o. I

f, ~

15.

L

RESULTS

! I

Experiment 1

O CONTROL • ~

,

A

,

ZPT i i

The food intake of rats fed 50 ppm ZPT was transiently depressed (Fig. 2), but a significant difference was noted only for days 3-4 of exposure. Nevertheless, the body weight of the ZPT group was significantly less than that of the CONTROL group beginning after 12 days of exposure and recovering only after 28 days on PLC. Both the magnitude and the duration of the body weight loss indicate that the weight loss cannot be accounted for by reduced food intake alone. The first somatomotor change seen in rats fed ZPT was a deficit in hindlimb grips strength (Fig. 2) which appeared after 8 days of testing and which recovered after 8 days on PLC. Forelimb grip strength deficits appeared slightly later, beginning after 12 days of exposure and recovered after only 4 days on PLC. Small but statistically significant reductions in extensor postural thrust distance (Fig. 2) were seen after 1 day of exposure and during the first week of recovery. No changes in the tail-lift extensor response were observed. Large reductions in hindlimb M-wave amplitude and a decrement during repetitive stimulation (Fig. 3) were observed beginning after 9 days of exposure to 50 ppm ZPT and recovering after 42 days on PLC. Denervation potentials (positive sharp waves and fibrillation potentials) appeared later and recovered faster than changes in average M-wave amplitude, beginning after 15 days of dosing, and recovering after 4 weeks on PLC (Fig. 3; control rats never exhibited denervation potentials, and so the percentage of occurrence is given only for ZPT-treated rats). Conduction velocity data were consistent with previously published data. Average caudal nerve conduction velocity was approximately 35 m/sec for 90-day-old rats of the Fischer 344 strain and rose gradually as the rats aged (16). In agreement with previous reports, conduction velocity was not affected at any timepoint by ZPT treatment (4,18).

i

! !

I, o Id~

== g,

o

!

~r,~

.

,

20.

!

1

230.

E

,

rT

z

.t ' ~ ' / 0

I

o l

1

~o

,?

"•m

I000

v

1

.o

._E_ o

200

500 400-

300"~

200 -

"I-

I DO-

!Lr.i

@

1 *O

I

L

! 11

,

i

,

i

|

.

.

.

where x = the rank of the p value in the original set of observations. If the corrected value Ptx] rejected the null hypothesis (i.e., was still significant), then Px was discarded and the procedure repeated with P(x+l) until a nonsignificant Ptxl was reached. A two-sided, adjusted p value of -<0.05 was considered statistically significant.

.

.

.

.

.

.

I

Experiment 2

BO-

P

On the 9th day of the experiment the mean body weights of groups CONTROL, FOOD DEP and ZPT were 2 9 6 _ 1 1 , 2 5 6 - 5 and 288 --- 10, respectively. The mean maximum reduction of body weight of group FOOD DEP relative to CONTROL was comparable to the maximum reduction of group ZPT in Experiment 1. Neither group FOOD DEP nor ZPT differed from CONTROL on any behavioral test because of the short duration of ZPT exposure. Group ZPT exhibited a reduction in hindlimb M-wave amplitude (M-wave a m p l i t u d e = 2 6 . 6 - 1 2 . 9 mV) relative to CONTROL (amplitude = 84.8 ± 9.8 mV) similar to that seen for a comparable duration of exposure in Experiment 1, while the FOOD DEP group (amplitude=86.0±5.8 mV) did not differ from CONTROL.

20--'

'l

. . . . . . . . . .

6c&rr~ot. • Zl:rl

'

80

X bJ

01

Experiment 3 The ZPT rats showed deficits in forelimb and hindlimb grip •

-5 PRE

, -

.

0

.

.

5

.

-

10

DAYS OF" DOSING

15

.

O

.

.

.

10

.

.

.

.

.

20 DAYS O F RECOVERY

,

30

4.0

FIG. 2. The effect of ZPT on behavioral measures in Experiment 1. CONTROL rats ate powdered lab chow (PLC) while ZPT rats ate 50 ppm ZPT in PLC for 14 days, then all rats ate PLC during recovery. Vertical bars are -2_SD. *Significantly different from CONTROL, p<0.05. **Significantly different from CONTROL, p<0.01.

FUNCTIONAL CORRELATES OF ZPT TOXICITY

157

120. ~

I I

T

o

-

1600

616

1o

I I I

t L 12oo

v t 4Ht ];

41,41"

I

aoo.

4H~

-d ~

3O

E

2o

~" ~o o

-~o2

J~

41"41"

I I

I

~*

I

•

i

I

•

o

Q

I

J~ I

I;*

o 19

o

I

!

! I !

~* I • I

I

01

,I

E

-

•

_|

.

.

.

.

.

.

.

!. . . .

I I !

8

'o

'

• ZPT

•

I I 41"J .Ire

c~

o

~

5o-

0

. I , - , -4 o 4 8 PRE I~YS OF I~SING

-

,

,

0

4.

,

.

,

8 12 I~Y5 OF" I~COVI[RY

-

,

16

.

,

20

O.

:>"

45

i

i

0 CONTROL

I I I

I I I

• ZPT

I "~O

FIG. 4. Forelimb(upper) and hindlimb(lower) grip strengthin Experiment 3. C O N T R O L rats ate powdered lab chow (PLC), while ZPT rats ate 50

°

ppm ZPT in PLC for 11 days, then all rats ate PLC during recovery. Verticalbars are ___SD. *SignificantlydifferentfromCONTROL,p<0.05. **Significantlydifferentfrom CONTROL, p<0.01.

35

I I

30. --5 PRE

I I S

10 15 DAYS OF OOSING

7

14

28 42 DAYS OF RECOVERY

FIG. 3. The effect of ZPT on electrophysiologicmeasures in Experiment 1. C O N T R O L rats ate powdered lab chow (PLC), while ZPT rats ate 50 ppm ZPT in PLC for 14 days, then all rats ate PLC during recovery. Verticalbars are __.SD. *SignificantlydifferentfromCONTROL,p<0.05. **Significantlydifferentfrom CONTROL, p<0.0l.

strength beginning simultaneously after 9 days of dosing (Fig. 4). Although Fig. 4 suggests that the hindlimb grip strength deficits recovered more slowly than forelimb deficits, the dilution of statistical effects to accommodate repeated measures analysis resulted in no statistical difference between the measures during recovery. The average hind- and forelimb-evoked potentials in control rats were approximately 80-100 mV, comparable to previously published data (6). The ZPT rats exhibited a significant decrease in hindlimb M-wave amplitude (Fig. 5) relative to forelimb amplitude during all three tests, even when hindlimb weakness was not evidenced by changes in grip strength. A relative increase in the hindlimb M-wave decrement was seen only during the second test time, when weakness was evident. DISCUSSION

The application of a series of quantitative behavioral tests confirmed previous observations that the major clinical correlate of ZPT-induced neurotoxicity is weakness. That is, grip strength was reduced, while few if any changes were seen in the extensor

postural thrust or tail-lift extensor response. However, the grip strength test provided a slightly different picture of mild to moderate ZPT-induced weakness than that obtained from gross observations. While descriptions based on observational methods have stressed the apparent hindlimb weakness (10, 17, 18), the grip strength tests gave little (Experiment 1) or no (Experiment 3) indication of differential effects on fore- and hindlimb strength. Although in theory hindlimb deficits might be expected to appear before forelimb changes during exposure to a distal axonopathic agent (19), the temporal appearance of grip strength deficits does not always confn'rn this expectation (12). The electrophysiologic data provide a picture of ZPT-induced neuropathy more consistent with selective vulnerability of hindlimb nerves. The hindlimb M-wave amplitude of ZPT-treated rats was reduced relative to forelimb amplitude prior to, during and after recovery from weakness. A similar vulnerability of hindlimb relative to forelimb M-wave has been reported for p-bromophenylacetylurea, which produces a morphologically similar peripheral neuropathy (6). The behavioral and electrophysiologic characterization of ZPTinduced neurotoxicity also differed during the recovery from the toxicity. While the grip strength data confirmed previous observations (14,17) that mild to moderate weakness reversed quickly upon removal of ZPT from the diet, electrophysiologic testing gave evidence for a slower time course to complete recovery. The greater sensitivity of M-wave amplitude relative to behavioral tests of weakness is not entirely unexpected. In dogs and cats, pharmacologic studies with competitive neuromuscular blocking drugs indicate a "margin of safety" for neuromuscular transmission, such that >75% of receptor sites must be occupied before a reduction in twitch tension is produced (11,23). Comparison of

158

ROSS AND LAWHORN

120'~

30 r--1 CONTROL

v

80.4

.~

40.

ZPT

v 20

~o

(3.

E

< 0) > o

o

0-

0

-I0

r'~ o

I

I

DAY 7: NO WEAKNESS 120-

80.

[

30

j

.r.

40,

20 10

0

O.

-10

DAY 11: WEAK 120

30

80

20

40

10

L,_j "," T

0

•-10

RECOVERY: NO WEAKNESS

FIG. 5. M-wave amplitude (left half) and M-wave decrement (right half) from control and ZPT rats after 7 (upper) and 11 (middle) days of dosing and 17 days (lower) of recovery. Control rats ate powdered lab chow (PLC), while ZPT rats ate 50 ppm ZPT in PLC for 11 days. All rats ate PLC during recovery. Vertical bars are +-SD. *Hindlimb minus forelimb significantly different from CONTROL, p<0.05. **Hindlimb minus forelimb significantly different from CONTROL, p<0.01.

group M-wave deficits with group grip strength values during onset and recovery from weakness (Experiments 1 and 3) indicated that behavioral deficits were observed only when the group M-wave amplitude was reduced by >80%. Denervation potentials (positive waves and fibrillation potentials) had a later onset and more rapid recovery than M-wave amplitude. These results are consistent with the hypothesis that such potentials appear only after days to weeks of neuromuscular junction denervation and disappear early in the process of reinnervation (2). The morphologic or physiologic correlate of the early electrophysiologic changes was not identified in the present experiments. It has been suggested that the early appearance of 1) tubulovesicular profiles in the motor nerve terminal and 2) Schwann cell processes in the synaptic cleft could render the nerve terminal nonfunctional prior to frank degeneration of the neuron (13). However, the possibility that pharmacologic neuromuscular blockade (1,24) could contribute to the early electrophysiologic changes cannot be overlooked. Food restriction which produced loss of body weight at a rate and amount comparable to 50 ppm ZPT in the diet failed to produce either behavioral or electrophysiologic changes indicative of neurotoxicity. These data indicate that the body weight loss commonly associated with pyrithione dosing to rats (10, 14, 18) contributes little, if any, to neurologic signs. Furthermore, it suggests that both the behavioral and electrophysiologic tests used in the present experiments can "read through" a moderate loss of body weight. Taken together, the data suggest that electrophysiology is considerably more sensitive than behavioral tests in detecting the neuromuscular deficits produced by pyrithione. Although neuropathology was not conducted in the present experiments, comparison with published data suggests that the electrophysiologic tests also compare favorably with electron microscopic histopathology. Sahenk and Mendell (14) reported that early morphologic changes could be demonstrated in rats eating 166 ppm ZPT for 7-10 days. We saw electrophysiologic changes as early as 6 days on a diet containing 50 ppm ZPT. Future work will help determine whether the relationship among test methods is true for other neurotoxins and for other types of lesions.

REFERENCES 1. Adams, M. B.; Wedig, J. H.; Jordan, R. L.; Smith, L. W.; Henderson, R.; Bozelleca, J. R. Urinary excretion and metabolism of salts of 2-pyridine-thiol-l-oxide following intravenous administration to female Yorkshire pigs. Toxicol. Appl. Phamaacol. 36:523-531; 1976. 2. Bowen, J. M. Peripheral nerve electrodiagnostics, electromyography, and nerve conduction velocity. In: Hoerlein, B. F., ed. Canine neurology. Philadelphia, PA: W. B. Saunders; 1978:268. 3. Daube, J. R. Nerve conduction studies. In: Aminoff, M. J., ed. Electrodiagnosis in clinical neurology. New York: Churchill Livingstone; 1986:270. 4. DeJesus, C. V. P.; Towfighi, J.; Snyder, D. R. Sural nerve conduction study in the rat: A new technique for studying experimental neuropathies. Muscle Nerve 1:162-167; 1978. 5. Edwards, P. M.; Parker, V. H. A simple, sensitive, and objective method for early assessment of acrylamide neuropathy in rats. Toxicol. Appl. Pharmacol. 40:589-591; 1977. 6. Jakobsen, J.; Lambert, E. H.; Carlson, G.; Brimijoin, S. Clinical and electrophysiological characteristics of the experimental neuropathy caused by p-bromophenylacetylurea. Exp. Neurol. 75:158-172; 1982. 7. Magos,L.; Peristianis, G. C.; Snowden, R. T. Postexposure preventive treamtent of methylmercury intoxication in rats with dimercaptosuccinic acid. Toxicol. Appl. Pharmacol. 45:463--475; 1978.

8. Meyer, O. A.; Tilson, H. A.; Byrd, W. C.; Riley, M. T. A method for the routine assessment of fore- and hindlimb grip strength of rats and mice. Neurobehav. Toxicol. 1:233-236; 1979. 9. Miyoshi, T.; Goto, I. Serial in vivo determinations of nerve conduction velocity in rat tails. Physiological and pathological changes. Electroencephalogr. Clin. Neurophysiol. 35:125-131; 1973. 10. Nolen, G. A.; Dierckman, T. A. Reproduction and teratology studies of ZPT administered orally or topically to rats and rabbits. Food Cosmet. Toxicol. 17:639-649; 1979. 11. Paton, W. D.; Waud, D. R. The margin of safety of neuromuscular transmission. J. Physiol. 191:59-90; 1967. 12. Pryor, G. T.; Bingham, L. R.; Dickinson, J.; Rebert, C. S.; Howd, R. A. Importance of schedule of exposure to hexane in causing neurotoxicity. Neurobehav. Toxicol. Teratol. 4:71-78; 1982. 13. Rebert, C. S.; Becker, E. Effects of inhaled carbon disulfide on sensory-evoked potentials of Long-Evans rats. Neurobehav. Toxicol. Teratol. 8:533-541; 1986. 14. Sahenk, Z.; Mendell, J. R. Ultrastructural study of zinc pyridinethione-induced peripheral neuropathy. J. Neuropathol. Exp. Neurol. 38:532-550; 1979. 15. Sahenk, Z.; Mendell, J. R. Zinc pyridinethione. In: Spencer, P. S.; Schaumburg, H. H., eds. Experimental and clinical neurotoxicology, chapter 40. Baltimore, MD: Williams and Wilkins; 1981:578-592.

FUNCTIONAL CORRELATES OF ZPT TOXICITY

16. Schmelzer, J. D.; Low, P.A. Electrophysiological studies on the effect of age on caudal nerve of the rat. Exp. Neurol. 96:612-620; 1987. 17. Snyder, D. R.; DeJesus, C. V. P.; Towfighi, J.; Jacoby, R. O; Wedig, J. H. Neurological, microscopic, and enzyme-histochemical assessment of zinc pyrithione toxicity. Food Cosmet. Toxicol. 17:651-660; 1979. 18. Snyder, D. R.; Gralla, E. J.; Coleman, G. L.; Wedig, J. H. Preliminary neurological evaluation of generalized weakness in zinc pyrithione-treated rats. Food Cosmet. Toxicol. 15:43-47; 1977. 19. Snyder, F. H.; Buehler, E. V.; Winek, C. L. Safety evaluation of ZPT in a shampoo formulation. Toxicol. Appl. Pharmacol. 7:425-437; 1965.

159

20. Spencer, P. S.; Schaumburg, H. H. Distal axonopathy: One common type of neurotoxic lesion. Environ. Health Perspect. 26:97-105; 1978. 21. Teal, J. J.; Evans, H. L. Behavioral effects of acrylamide in the mouse. Toxicol. Appl. Pharmacol. 63:470-480; 1982, 22. Tukey, J. W.; Ciminera, J. L.; Heyse, J. F. Testing the certainty of a response to increasing doses of a drug. Biometrics 41:295-301; 1985. 23. Wand, D. R.; Waud, B. E. In vitro measurement of margin of safety of neuromuscular transmission. Am. J. Physiol. 229:1632-1634; 1975. 24. Winek, C. L.; Buehler, E. V. Intravenous toxicity of zinc pyridinethione and several zinc salts. Toxicol. Appl. Pharmacol. 9:269-273; 1966.