WIUHIWIIUHHIHIUHWWWHlllWllNllHWl

   

This is to certify that the

dissertation entitled

A QUANTITATIVE ASSESSMENT OF TPP-INDUCED DELAYED NEUROPATHY
IN THE RETINA AND LATERAL GENICULATE NUCLEUS OF THE EUROPEAN
FERRET (MUSTELA PUTORIUS FURO)

presented by

Lee Lipsitz

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Ana t omy

 

 

Myth

@jor professor

Date / [Tlgfl/O 0

MS U is an Afflrmatiw Action/Equal Opportunity Institution 0-12771

 

 

LIBRARY

Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/Dataouepes-nts

 

 

A QUANTITATIVE ASSESSMENT OF TPP-INDUCED DELAYED
NEUROPATHY IN THE RETINA AND LATERAL GENICULATE NUCLEUS
OF THE EUROPEAN FERRET (MUSTELA PUTORIUS FURO)

By

Lee Lipsitz

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
Department of Anatomy

2000

ABSTRACT
A QUANTITATIVE ASSESSMENT OF TPP-INDUCED DELAYED
NEUROPATHY IN THE RETINA AND LATERAL GENICULATE NUCLEUS
OF THE EUROPEAN FERRET (MUSTELA PUTORIUS FURO)
By

Lee Lipsitz

Triphenyl phosphite (TPP), an organophosphorous delayed neurotoxicant, has
been examined extensively in our lab to assess its degenerative effects on the central
visual pathway Of the European ferret. Tanaka et al. (1994, 1999) reported an age-
related pattern of fiber and cell body degeneration in TPP-treated animals,
progressing from retinal axons and lateral geniculate nucleus (LGN) neurons to
visual cortex. These studies, however, did not address whether TPP exposure also
results in retinal ganglion cell (RGC) degeneration, nor did they examine
quantitatively the degenerative effects in the LGN. The purpose of this study was tO
quantify the effects of TPP on RGCs and LGN neurons in the ferret, to describe
functional deficits associated with TPP exposure using electroretinography (ERG),
and to assess changes in whole-brain neurotoxic esterase (NTE); compounds that
induce organophosphorous delayed neuropathy typically inhibit NTE. We
administered Single subcutaneous injections Of TPP (1184 mg/kg) to 13 adult ferrets.
Nine were examined 5 days after treatment and 4 were examined 7 days after
treatment. Five additional ferrets were injected with TPP and NTE analysis was
performed 48 hours after treatment (Johnson, 1977). For histological analysis, the

fixed retinae were whole-mounted and the brains were sectioned parasagittally (50

am) using a Vibratome. All tissue was processed using standard histological and
analysis techniques. ERGS were recorded with a DTL PlusTM disposable electrode
(Retina Technologies Inc.. Hixson, TN) over a range Of 4.5 log units below
maximum intensity up to maximum intensity (2.5 cd-s/mz). RGC counts from
matched nasal regions Showed Significantly fewer (21%) neurons in the TPP-treated
ferrets (5d = 282 i 52 SD; 7d = 284 i 12 SD) compared with normal (359 i 42
SD). No Significant difference in cell number was found in temporal retina, even
though this region contained, on average, 13% fewer ganglion cells in the TPP-
treated ferrets (5d = 334 i 44; 7d = 357 i 39 vs normal = 394 i 72). Mean soma
size comparisons and cell size distributions for RGCS in both nasal and temporal
retina showed no Significant differences for any Of the animals studied. LGN
neurons were significantly smaller (28%) than normal in the TPP-treated ferrets (5d
= 155 um2 i 23; 7d = 152 um2 i 28 vs normal = 214 um2 i 9). The cell Size
distributions for LGN neurons were shifted toward smaller cell Sizes in both the 5-
and 7-day TPP-treated groups compared tO normal. There were no functional
deficits detected with the ERG as a result of TPP exposure. TPP produced a 75%
reduction in NTE activity compared with normal. The results of this study provide
for the first time quantitative evidence that exposure to TPP results in severe

degenerative changes in both the retina and LGN of the European ferret.

This dissertation is dedicated to the memory Of Dr. Duke Tanaka, Jr.

ACKNOWLEDGMENTS

The completion of this dissertation would not have been possible without the
support Of many individuals. I would like to begin by thanking Dr. Steve J. Bursian,
who has been with me since the beginning of my graduate career. Without his
support and willingness to help me in any way possible I would not have completed
this degree. Thank you to Dr. Arthur J. Weber, who also played a critical role in
my Obtaining this degree. His unwavering support and assistance on every aspect of
this project was instrumental in the completion of this degree. I am grateful that he
stepped into the role as my primary advisor after the loss of Dr. Tanaka, Jr. Thank
you to the remainder Of my graduate committee—Drs. William Falls, Sharleen
Sakai, and David Ramsey—for your support and guidance throughout my program.
Thank you Dr. Richard Aulerich for your continuing support throughout my
graduate career. I am grateful to the Department of Anatomy faculty and staff who
saw me through some very difficult times and encouraged me to not give up.
Thanks are due to the Graduate School, College of Veterinary Medicine, and the
Michigan Agricultural Experiment Station for their financial support to see me
through the last years of this degree. And Of course thank you tO my family and

friends, without whom none Of this matters.

TABLE OF CONTENTS

LIST OF TABLES ....................................... vii
LIST OF FIGURES ....................................... viii
INTRODUCTION ........................................ 1
MATERIALS AND METHODS .............................. 5
Subjects and Procedures .................................. 5
Histology ............................................ 5
Tissue Processing ................................... 5
Retinal Ganglion Cell Analysis .......................... 6
Lateral Geniculate Neuron Analysis ....................... 7

NTE Assay ........................................... 8
Electroretinography ..................................... 9
Corneal Recording Electrodes ........................... 10
Statistical Analysis ..................................... 12
RESULTS ............................................. l3
Retinal Ganglion Cells ................................... 13
Qualitative Observations ............................... 13
Quantitative Analyses ................................ l3
Lateral Geniculate Nucleus ................................ 15
Qualitative Observations ............................... 15
Quantitative Analyses ................................ 15

NTE Assay ........................................... 16
ERG Recordings ....................................... 16
DISCUSSION ........................................... 18
APPENDIX A: Tables 1—4 .................................. 29
APPENDIX B: Figures 1—9 ................................. 34
BIBLIOGRAPHY ........................................ 49

vi

Table 1

Table 2

Table 3

Table 4

LIST OF TABLES

The Effect of Triphenyl Phosphite (TPP) on Retinal Ganglion
Cell (RGC) Measurements in the European Ferret. .........

The Effect of Triphenyl Phosphite (TPP) on Lateral Geniculate
Nucleus (LGN) Soma Sizes in the European Ferret. ........

The Effect of Triphenyl Phosphite (TPP) on Whole-Brain
Neuropathy Target Esterase (NTE) Activity in the European
Ferret. ........................................

Mean ERG Latency and Amplitude Measurement for the STR,
a-wave, and b-wave, at Each Stimulus Intensity ...........

vii

29

3O

31

32

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

LIST OF FIGURES

Schematic drawing Of the matched nasal and temporal
regions Of the retina. The boxed regions within each retina
represent the sample areas while the arrows represent their
approximate projections to the LGN. The filled circle
represents the Optic disc in each retina. In the LGN laminae
C, A, and A1 are labeled as well as the perigeniculate
nucleus (P) ...................................

Photomicrographs of cresyl-violet stained sections from
matched nasal regions of the retina in a normal (A) and a
TPP-treated (B) ferret. Note the decrease in retinal ganglion
cell density in the TPP-treated retina compared to the normal.
Scale bar = 50 um. .............................

Histograms comparing the mean number of retinal ganglion
cells in matched nasal and temporal regions Of the retina in
normal and TPP-treated ferrets 5 and 7 days after treatment
(* p < 0.05 Kruskal-Wallis w/Mann Whitney). Nasally there
were 21% fewer cells in the TPP-treated ferrets than in the
normal ferrets, while temporally there were 13% fewer
ganglion cells .................................

Histograms comparing the mean soma diameters Of retinal
ganglion cells in matched nasal and temporal regions Of
normal and TPP-treated ferrets 5 and 7 days after treatment.
Mean soma size was similar for all groups (Kruskal-Wallis
waann Whitney) ..............................

Smoothed histograms showing the Size distributions of retinal
ganglion cells from matched nasal (A) and temporal (B)
regions of the retinae in normal and TPP-treated ferrets 5 and
7 days after treatment. The cell size distributions were
similar for all groups (Kolmogorov-Smirnov test) ........

viii

34

36

37

38

40

Figure 6

Figure 7

Figure 8

Figure 9

Photomicrographs Of cresyl-violet stained parasagittal sections
Of the ferret LGN. (A) Low power (4x) view illustrating the
laminar divisions C, A1, A, and the perigeniculate nucleus
(P). (B) Low power view of a 5-day post-treatment LGN.
Note the pale appearance of the TPP-treated LGN compared
to normal. Scale bar = 50 pm. (C) High power (40x) view
of neurons in lamina A1 of a normal LGN. (D) High power
view of neurons in lamina A1 of a 5-day post-treatment
LGN. Note the irregular Shaped, fragmented, pale staining,
and pyknotic nuclei (arrowheads) in the TPP-treated in
comparison to normal. Scale bar = 500 pm ............ 42

Histogram comparing the mean cross-sectional area in LGN

neurons from matched regions in lamina A and A1 in normal

and TPP-treated ferrets 5 and 7 days after treatment. The

mean cross-sectional areas of LGN neurons in TPP-treated

ferrets were 30% smaller than those of normal ferrets (* p <

0.05 Kruskal-Wallis w/Mann Whitney) ............... 43

Smoothed histograms comparing the size distributions of

neurons from matched regions in lamina A and A1 of the

LGN in normal and TPP-treated ferrets 5 and 7 days after

treatment. There was a significant shift toward smaller soma

Sizes in the TPP-treated ferrets compared to normal (* p <

0.05, Kolmogorov-Smirnov test) .................... 44

Corneal ERG responses. (A) Responses from a normal ferret

using a DTL fiber electrode at each stimulus intensity on day

l and day 5. The STR, b-wave, and a-wave are labelled with

arrows. The arrowheads point out the artifact. (B)

Responses from a day 5 post-treatment ferret using a DTL

fiber electrode at each stimulus intensity on day 1 before

treatment and day 5 after treatment. The STR, b-wave, and

a-wave are labelled with arrows. The arrowheads point out

the artifact ................................... 46

ix

INTRODUCTION

Organophosphorous compounds (OPS) are carbon-containing chemicals
derived from phosphorous acids. Organophosphates are used extensively as
insecticides. They are also used as intermediates in the manufacturing Of industrial
compounds, pharmaceuticals, fire retardants, antifungal agents, and herbicides, and
they have been developed as chemical warfare agents (Padilla et al., 1978; Stephans.
1991; Marrs, 1993; US EPA, 1985; Windebank, 1987; Ecobichon, 1994; Ehrich.
1996). Along with these many uses, OPS appear to have several mechanisms by
which they can cause toxicity. Organophosphates are anticholinesterase agents and
their acute toxic effects are a direct result of interference with cholinergic
transmission. Certain OPS also have been implicated in producing
organophosphorous-induced delayed neurotoxicity (OPIDN). Although the
mechanism has not been elucidated, it appears unrelated to its anticholinesterase
activity (Johnson, 1977; Veronesi, 1987; Abou-Donia, 1992; Ecobichon, 1994;
Koelle, 1994; Sultatos, 1994; Kang et al., 1995; Richardson, 1995). It is apparent
that the broad applications Of use and the many as yet unexplained mechanisms Of
toxicity pose a significant threat to the health Of humans and animals.

Clinically, OPIDN is characterized by a delayed onset of ataxia, and
subsequent paralysis (Abou-Donia, 1981; Baron, 1981). Exposure to the OP
triphenyl phosphite (TPP), an OPIDN compound, results in widespread axonal
degeneration in the mammalian and avian central nervous system (Tanaka et al.,
1990, 1992, 1994; Varghese et al., 1995a; Lehning et al., 1996). Of particular
interest are the degenerative effects TPP has on the visual system Of the European

1

ferret (Tanaka et al., 1990, 1994). The degeneration pattern found in these animals
appears to be age-related. Ferret kits 4 weeks Of age or younger do not Show any
evidence Of degeneration in the visual pathway as demonstrated by silver-
impregnation, even though degeneration is evident in other brain regions. At 5
weeks of age the kits begin to Show retinogeniculate axon degeneration, and by 7
weeks of age there is evidence Of lateral geniculate nucleus (LGN) cell body
degeneration. Ferret kits at 10 weeks of age have degeneration patterns consistent
with the adult animal. This includes retinogeniculate axon degeneration, LGN cell
body degeneration, and geniculocortical axon degeneration. This sequence Of events
led Tanaka et a1. (1994, 1999) to hypothesize a pattern Of anterograde transneuronal
degeneration beginning with the retinal ganglion cells, the first-order neurons in the
visual pathway. These studies, however, did not examine directly whether TPP
produced damage within the retina. Examining the effects of TPP on the retina was
one focus of this study.

OPIDN compounds inhibit certain esterases, including neuropathy target
esterase (NTE) (Johnson, 1982, 1990). NTE is an enzyme found primarily in nerve
tissue, and at present its function is unknown; It is thought that to initiate OPIDN,
NTE must go through a two-step process of phosphorylation by the OPIDN
compound, followed by cleavage of an alkyl group from the phosphorylated enzyme,
a process called aging (Johnson, 1982; Lotti et al., 1984; Berends, 1987; Abou-
Donia and Lapadula, 1990; Kamata et al., 1999). It is presumed that NTE must
reach a 70% inhibition threshold before clinical Signs and neuropathy are Observed.
At present, the validity Of measuring NTE inhibition to identify OPIDN compounds
is controversial. Nonetheless, inhibition Of NTE in the domestic female chicken

2

remains the standard test used by the Enviromnental Protection Agency (EPA) to
identify OPIDN compounds (Varghese et al., 1995; Jamal, 1997; Kamata et al.,
1999).

Risk assessment is based not only on the mechanism of a toxic compound.
but also on the significance of its effects. Determining whether a toxicant alters
visual function is an important component of risk assessment. Although it may take
some time to elucidate the mechanism Of OP neurotoxicity, it is important to
establish functional assays that can be performed on humans and animals exposed to
these compounds. This information will enhance the possibility Of detecting adverse
effects at the earliest possible stage Of exposure. With respect to the visual system.
electroretinograms (ERGS) are used routinely to assess the functional integrity of the
retina. The ERG consists Of an initial negative deflecting a-wave, followed by
positive deflecting b- and c-waves. Recently, a negative component Of the ERG, the
scotopic threshold response (STR), was described (Sieving, 1986). The STR
originates from the inner retina, and may provide functional information regarding
either the direct or indirect responses of retinal ganglion cells (Sieving et al., 1986;
Sieving and Wakabayashi, 1991; Frishman et al., 1996).

Based on the above information, the 3 main Objectives Of this study were to:
1) determine whether TPP causes degeneration Of ganglion cells in the retina, 2)
quantify the extent Of postsynaptic damage also present in the LGN, and 3) establish
whether the ERG might serve as a reliable means to document TPP-induced
functional deficits. Since the EPA continues to use NTE inhibition as a standard
indicator Of the ability of organophosphorous compounds to produce OPIDN, and
because the effect of TPP on ferret whole-brain NTE activity has not been

3

examined, a final Objective of this study was to determine if TPP causes a reduction

in whole-brain NTE activity.

MATERIALS AND METHODS

Subjects and Procedures

Thirty European ferrets (Mustela putorius furo), of both sexes, ranging in age
from 1 to 3 years Old, were used in this study. Ferrets were housed individually in
wire cages at the Michigan State University Experimental Fur Farm. Each cage was
bedded with excelsior (wood wool), and food and water were available ad libitum.
Temperature and humidity were ambient. Experimental ferrets received a single
subcutaneous injection of TPP (1184 mg/kg, 97% pure, Aldrich Chemical CO.,
Milwaukee, WI). Normal ferrets were not injected. The dose of TPP used has been
Shown previously to cause clinical deficits and neuronal degeneration characteristic
of OPIDN in the ferret (Tanaka et al., 1994, 1999). Animals were Observed daily
for acute systemic effects, such as difficulty in breathing or vomiting, and Signs
indicative of OPIDN, such as paresis or paralysis. Any animal displaying signs
Significant enough to compromise its ability to eat or drink was euthanized with an

overdose Of pentobarbital sodium (> 50 mg/kg).

Histology
Tissue Processing

Seven normal ferrets and 13 TPP-treated ferrets were used for histological
analysis of the retina and LGN. Of the TPP-treated ferrets, 9 were examined 5 days
after treatment and 4 were examined 7 days after treatment. All animals were first
anesthetized with pentobarbital sodium (50 mg/kg) and perfused transcardially with
1.5% paraformaldehyde and 2% glutaraldehyde in a 0.1M phosphate buffer solution

5

(pH 7.5). The brain and eyes were removed immediately, and the brains post-fixed
in the same solution. The anterior segment Of each eye was removed and the
posterior eyecups then were placed in 0.01 M phosphate buffered saline (pH 7.5).
Following 3 hours Of post-fixation, the retinae were removed from the eyecup,
whole mounted onto glass slides, dehydrated in graded alcohols, defatted, and
stained with cresyl violet. The brains remained in fixative for 24 hours. The
forebrain, containing the LGN, was sectioned (50 um) parasagittally using a
Vibratome. The sections were mounted onto glass slides, dehydrated in graded

alcohols, defatted, and stained with cresyl violet.

Retinal Ganglion Cell Analysis

Retinal ganglion cell measurements were Obtained from 6 normal and 11
TPP-treated ferrets. All comparisons were made in matched regions Of the nasal and
temporal retina. The representation of the visual field on the ferret retina has been
defined as 12.4°/mm (labs and Stryker, 1985). Using this as a reference, it was
possible to transpose the visual field representation in the LGN back to the correct
physical location on the retina. The regions of retina examined were located
approximately 1.61 mm superior to the optic disc and 2.02 mm away from the optic
disc in either the nasal or temporal direction. These retinal regions were selected for
analysis because the axons Of ganglion cells located here project to that part of the
LGN shown previously to undergo TPP-induced degeneration (Tanaka et al., 1994,
1999). In addition, because these regions are located outside the area centralis and
visual streak, ganglion cell density and Size are relatively constant. Figure 1 shows
the locations of the retinal sample regions and their projection patterns to the LGN.

6

Each retina was analyzed using a microscope-based digitizing system
(Minnesota Datametrics, St. Paul, MN). Using a 40x Objective, 4 adjacent retinal
areas were captured using a high resolution CCD camera (Hamamatsu model C5985)
and image analysis software (Image Pro Plus, Media Cybernetics, Inc.). Cells
identified within each sample area were used as references to avoid duplicate cell
counts. The stage was moved systematically until images from 4 sequential, but
separate, retinal areas were captured (total sample area = 0.164 mmz).

Measurements of the cross-sectional areas of ganglion cells displaying clearly
visible nucleoli were Obtained using the image analysis software. The somal
diameters Of these neurons then were calculated from the area measurements. Since
retinal ganglion cells have highly circular somata, the areal measurements provide a

relatively true estimate of cell diameter.

Lateral Gem'culate Neuron Analysis

The LGNS of 6 normal and 13 TPP-treated ferrets were examined. Tanaka et
al. (1994) reported heavy neuronal degeneration over the full rostral-caudal extent of
the ventral region Of laminae A and A1. In order to correlate the region Of the LGN
examined with the retinal areas studied, the known projection pattern Of the ferret‘s
visual field within the LGN was used (labs and Stryker, 1985; Figure 1). The
parasagittal sections (50 um) were examined light microscopically to describe the
degenerative changes present. Cell measurements were made systematically in each
ferret beginning at the rostral border Of lamina A and moving sequentially to the
caudal edge of lamina A1, where it borders with lamina C. If 100 cells were not
counted in an initial pass, a second adjacent pass was made across the ventral region

7

of laminae A and A1. For all LGNS, the cells were drawn at 1000x using a camera
lucida. Cross-sectional areas of the drawn cells were determined using a digitizing

tablet (Sigma Scan, Jandel Scientific).

NTE Assay

Forty-eight hours after administration of TPP, 5 normal and 5 treated ferrets
were sedated with ketamine HCl (11 mg/kg), and then killed by cervical
decapitation. The brains were removed quickly, weighed, and frozen on dry ice.
Assessment of NTE activity was performed using the method Of Johnson (1977).
Briefly, whole brains were homogenized in 50 mM Tris buffer (pH 8.0). An aliquot
of the homogenate containing 6 mg of brain tissue was incubated with 40 pM
paraoxon, or paraoxon plus 50 NM mipafox, in a final volume of 2 mls. Two
milliliters of a phenyl valerate solution (1 vol Of 15 mg phenyl valerate/ml redistilled
dimethylformamide in 30 vol Of 0.03% Triton-X 100), was added to each tube.
Tubes were incubated at 37°C for 20 minutes. The reaction was stopped by adding
2 mls Of 1% sodium dodecyl sulfate solution containing 0.025% 4-aminoantipyrene.
One ml Of 0.4% potassium ferricyanide then was added for color development.
After 30 minutes, absorbance for each tube was read at a wavelength of 510 nm
using a spectrophotometer. Enzyme activity was expressed as nmoles phenyl
valerate hydrolyzed/min/gram brain. The percent inhibition of NTE was calculated
by dividing the difference between the mean enzyme activity measured in normal
and TPP-treated animals by that Obtained from the brains of normal animals, and

multiplying by 100.

Electroretinography

Usable ERG recordings were Obtained from 9 of 30 animals after numerous
modifications to the system. Baseline ERGS were recorded with varying light
intensities immediately before dosing with TPP. After recording, 5 Of the 9 ferrets
were injected with TPP. The ERG light intensity series then was repeated on all 9
ferrets at 5 days after treatment. Prior to testing, all ferrets were dark-adapted for 1
hour. Under red light illumination, ferrets were sedated with xylazine (2.2 mg/kg),
followed by ketamine HCl (11 mg/kg) given intramuscularly, 20 minutes later.
Sedation Of all ferrets began 20 minutes prior to each ferret’s ERG recording
session. This was done to reduce any potential variability in ERG recordings arising
from different levels of sedation between animals. The pupil of the right eye was
dilated with 1% tropicamide HCl (Mydriacyl®, Alcon Laboratories, Ft. Worth, TX)
and the cornea anesthetized with 0.5% proparacaine HCL (Alcaine®, Alcon). The
eyelids were held Open with sutures, and a recording electrode was placed on the
cornea and held in place with a small amount of methylcellulose. Platinum
subderrnal electrodes (Grass Instrument CO., Cambridge. MA) were used for the
differential and ground electrodes (see below for electrode placement). A Ganzfeld
stimulator equipped with a xenon bulb (2.5 cd-s/mz) was used to elicit all ERG
responses. The light intensity was varied in 0.5 log unit steps using Kodak Wratten
neutral density filters over a range Of 4.5 log units below maximum intensity up to
maximum intensity (2.5 cd-s/mz). A bandpass of 03-70 Hz was used to filter the
ERG responses. At light intensities ranging from —4.5 log units to —1.0 log unit
below maximum intensity, 4 flashes, set 5 s apart, were averaged. The final 2 light
intensity settings had an interstimulus time Of 30 S between the 4 flashes used for

9

averaging. Signals were amplified and digitized using the Electrophysiologic
Personal Interfaced Computer-2000 (EPIC-2000, LKC Technologies, Inc,
Gaithersburg, MD) software package. The ERG recordings then were displayed on
the EPIC-2000 computer monitor, and using the EPIC 2000 software package,
latency and amplitude measurements of the STR, a-wave, and b-wave were made.
All latency measurements were made from the onset of the flash stimulus to the start
Of the corresponding response. STR amplitude was measured from baseline to the
lowest point of the STR response, the a-wave amplitude was measured from baseline
to the peak of the a-wave response, and b-wave amplitude measurements were made

from the peak of the a-wave response to the peak Of the b-wave response.

Corneal Recording Electrodes

An attempt to record stable ERGS was made using 3 different corneal
recording electrodes. Initially, a homemade monopolar contact lens containing a
platinum wire loop was placed on the eye with a small amount Of methylcellulose
for adhesion. With the monopolar system, a differential electrode was placed 1 cm
from the lateral canthus of the recorded eye, and a ground electrode was placed over
the occipital protuberance of the Skull. Recordings Obtained with this system were
not consistent within the same animal, and it often was difficult to record any
response from the retina. In addition, the stimulus produced a significant electrical
artifact, usually a large negative potential, that occurred prior to the start of the a-
wave. The second corneal electrode system involved the use Of a Burian-Allen
bipolar corneal electrode (Hansen Ophthalmic Development Laboratory, Iowa City,
Iowa) designed originally for use in the rat. With this electrode, only a ground

10

electrode was necessary, and it was placed over the hindquarters of the animal in
order to minimize any potential electrical interference. Unlike other Burian-Allen
corneal electrodes, those designed for rats are not attached to a lid speculum. This
is because the rat eye, unlike the ferret eye, protrudes forward enough that keeping
the lids Open during an ERG recording is not an issue. The Burian-Allen electrodes
also are designed differently for each species based on the average curvature of the
cornea. This ensures proper contact with the cornea. Proper fit of the Burian-Allen
electrode was difficult to achieve on the ferret eye. A lid Speculum attachment most
likely would have improved stability and enhanced the contact Of the Burian-Allen
electrode on the cornea. The ERG responses generated with this electrode also were
not consistent within the same animal, and Often no response from the retina could
be recorded. In addition, the large negative potential that was Observed with the first
electrode system remained. The third ERG electrode tried was the DTL PlusTM
disposable electrode (Retina Technologies Inc., Hixson, TN). The DTL electrode is
a silver/nylon fiber that is placed across the cornea with an artificial tear solution
(IsoptO-Tears®, Alcon). The fiber can be cut appropriately to fit across any size
cornea. Since this is a monopolar system, differential and ground electrodes again
were used, and placed in the nasal dermis and subdermally over the hindquarters,
respectively. The ERG responses obtained using this system were greatly improved
over those recorded with the other two systems. ERGS were recorded for all
animals, and the response was more consistent within the same animal. However,
the negative artifact, although diminished, still was present. Attempts to eliminate
this artifact were made by replacing the xenon bulb and by conducting the ERGS in
a different recording facility. The artifact never was eliminated fully, and it

11

interfered with interpretation Of the a-wave and the STR. The STR was one of the

primary ERG components this study intended to examine.

Statistical Analysis

Comparisons of retinal ganglion cell number and mean cell diameter were
made using the Kruskal-Wallis and Mann-Whitney tests (SPSS, Chicago, IL).
Differences in mean soma size in the LGNS Of normal and TPP-treated ferrets were
compared using one-way analysis Of variance (ANOVA; SPSS). The Kolmogorov-
Smimov test for two independent samples was used for comparisons of retinal
ganglion cell and LGN cell size distributions (SPSS). Differences in mean latency
and amplitude measurements at each light intensity setting for the STR, a- and b-
waves were compared using a multivariate ANOVA (SAS, SAS Institute, Cary, NC).

In all cases, the level of Significance was p < 0.05.

12

RESULTS

Retinal Ganglion Cells
Qualitative Observations

Photomicrographs of matched regions Of the nasal retina of a normal and 5-
day post-TPP-treated ferret are shown in Figure 2. Since there were no differences
between the 5-day and 7-day post-TPP ferrets, only an example from a 5-day animal
is Shown for comparison to normal. Light microscopically, there is no evidence Of
any cellular degeneration, as indicated by pyknotic nuclei, in any Of the retinal
ganglion cells in the TPP-treated animal. The figure also illustrates that a Similar
range of cell sizes (large, medium, and small) is present in both the normal and
TPP-treated retinae. However, the retina of the TPP-treated ferret shows a clear

decrease in ganglion cell density.

Quantitative Analyses

Table 1 presents the number and mean soma diameters of retinal ganglion
cells in the nasal and temporal retinae Of normal and TPP—treated ferrets. The
number Of retinal ganglion cells and the mean soma diameters are greater in the
temporal retinae compared to the nasal retinae in normal and TPP-treated ferrets. In
the nasal retina there were Significantly fewer (21%) neurons in the TPP-treated
ferrets (5d = 282 j: 52 SD; 7d = 284 i 12 SD) than in the retina Of normal ferrets
.(359 i 42 SD). Cell counts from the temporal retinae of TPP-treated ferrets (5d =
334 :t 44 SD; 7d = 358 i 39 SD) were not significantly different from normal (394
i 72 SD), even though this region contained, on average, 13% fewer ganglion cells

13

in the TPP-treated ferrets. There were no differences in mean somal diameters
between any of the normal and TPP-treated ferrets in either the nasal or temporal
regions. The histograms in Figure 3 illustrate the decrease in cell number in the
TPP-treated ferrets compared to normal. The histograms in Figure 4 illustrate that
the mean soma sizes Of ganglion cells in both the nasal and temporal regions of the
retina were similar for the normal and TPP-treated eyes. The smoothed histograms
in Figure 5 compare the Size distributions of ganglion cells located in the nasal and
temporal retinae of normal and TPP-treated animals. There were no significant
differences in the cell Size distributions for any Of the groups. Nasally the range and
mean soma Sizes are Similar (normal = 6026.4 um, 12.3 um i 0.8 SD vs. 5d = 6.5-
270 um, 13.1 um i 0.7 SD; 7d = 7.1-22.4 urn, 12.3 um i 0.7 SD). The range of
soma Sizes, and the mean soma sizes, are similar temporally as well (normal = 6.6-
26.7 um, 13.2 um i 0.6 SD vs. 5d = 6.5-25.4 um, 13.7 um i 0.8 SD; 7d = 7022.2
um, 13.0 pm i 0.8 SD). Previously, Vitek et al. (1985) reported retinal ganglion
cell soma diameters in the ferret retina; 3-4% Of the retinal ganglion cells had
somata > 20 am in diameter, (classified as alpha cells), while the remaining 96-97%
of ganglion cells had soma diameters measuring < 20 am, (all other cell classes). In
the present study, 6% (x = 22) Of the ganglion cells in the nasal region sampled had
soma diameters > 20 pm, and 94% (x = 337) had soma diameters < 20 pm. The
percentage of retinal ganglion cells with soma diameters > 20 pm decreased to 2%
(x = 6) at 5-days after treatment and 0.3% (x = l) at 7-days after treatment.
Similarly, the percent of retinal ganglion cells in the nasal region sampled with soma
diameters < 20 um increased to 98% (x = 276) and 99.7% (x = 283) for the 5-day
and 7-day post-treatment ferrets, respectively. In the temporal retina the percent of

14

cells with soma diameters > 20 um were 3%, 4%, and 1% for normal, 5-day, and 7-
day treatment ferrets, respectively. The percent of cells with soma diameters < 20
nm were 97%, 96%, and 99% for normal, S-day, and 7-day treatment ferrets.

respectively.

Lateral Geniculate Nucleus
Qualitative Observations

There was histological evidence of degenerative changes in the LGNS Of
TPP-treated ferrets. Qualitatively there were nO apparent differences between the
degenerative changes in the 5-day and 7-day post-treatment LGNS. The
photomicrographs in Figures 6a (normal) and 6b (7-day post-treatment) Show the
laminar divisions in parasagittal sections Of the LGN. The LGN in the TPP-treated
ferret stains less intensely due tO the loss Of Nissl substance, compared to the normal
LGN. The individual cells in the rostral-caudal extent Of the ventral region of
laminae A and Al were well defined in the normal LGN (Figure 6c). In
comparison, the photomicrograph in Figure 6d illustrates that the neurons in the
LGNS Of the TPP-treated ferrets are irregular in Shape, fragmented, pale staining,

and pyknotic.

Quantitative Analyses

The cross-sectional areas of all LGN neurons measured in the normal and
TPP-treated ferrets are presented in Table 2, and these data are presented graphically
in the histograms Of Figure 7. The data show that the mean cross-sectional areas Of
LGN neurons in both 5- and 7-day TPP-treated ferrets (5d = 155 um2 :I: 23 SD; 7d =

15

152 um2 i 28 SD) are significantly smaller (30%) than those Of neurons in normal
animals (214 um2 i 10 SD). The smoothed histograms in Figure 8 compare the Size
distributions of LGN neurons measured in the normal and TPP-treated animals.
There is a significant Shift toward smaller cell sizes for both the 5-day and 7-day
post-treatment ferrets compared with normal. The histograms suggest a decrease in
the percent of large cells and an increase in the percent Of small cells in the LGNS

of the 5-day and 7-day post-treatment ferrets.

NTE Assay

TPP had a Significant effect on whole-brain NTE activity. NTE activity in
TPP-treated ferrets was reduced 75% when compared to normal ferrets (Table 3).
Mean NTE activity in the normal ferrets measured 1519 i 161 nmoles
phenylvalerate hydrolysed/min/gram brain and in the TPP-treated ferrets the mean

activity measured 371 i 124 nmoles phenylvalerate hydrolysed/min/gram brain.

ERG Recordings

To aid in the comparison Of ERG components, ERG recordings from all 9
ferrets on day 1 were grouped as day l-normals, recordings from the 4 untreated
ferrets on day 5 were grouped as day S-normals, and recordings from the 5 TPP-
treated ferrets were grouped as day 5-TPP animals. Figures 9a and 9b Show the
ERG recordings from the days 1- and 5-normal and day l-normal and day 5-TPP-
treated ferrets, respectively. The STR, a-, and b-waves, and the recurring electrical
artifact are labeled. The electrical artifact is first evident at the lowest stimulus
intensity (-4.5 log cd-S/mz) and is visible in day 1-normal, day 5-normal, and day

16

5-TPP recordings. At each Stimulus intensity, the amplitude and shape of the artifact
is different in the ERG recordings shown in Figures 9a and 9b. The artifact was
present only when the corneal electrode was on the ferret’s eye. Despite the
unpredictable nature Of the artifact, it was small enough that latency and amplitude
measurements Of the STR and the a- and b-waves still could be made (Table 4).
The STR first was detected at the lowest stimulus intensity (-4.5 log cd-S/mz), and
was measurable until approximately the onset of the b-wave (-3.0 cd-S/mz). The
mean latency of the STR ranged from 45-90 msec for day l-normals, 60-105 msec
for day S-normals, and 45-90 msec for day 5-TPP ferrets over the intensity series.
The maximum amplitude Of the STR was 133 NV, 117 UV, and 233 ”V for the day
l-normal, day S-normal, and day 5-TPP ferrets, respectively. The amplitude of the
STR in the day 5-TPP recordings was significantly different compared to the day 5-
normal recordings but not from the day-1 normal recordings. The a—wave was first
detected at —2.5 log cd-s/mz. The mean latency range Of the a-wave was 0-45 msec
for all groups, and the maximum amplitude was 433 ”V, 383 UV, and 483 ”V for
the day l-normal, day 5-normal, and day 5-TPP ferrets, respectively. The b-wave
was first detected at —3.0 log cd-S/mz. The mean latency of the b-wave ranged from
30-120 msec for day l-normals, 30-105 msec for day 5-normals, and 30-105 msec
for day 5-TPP ferrets. The maximum amplitude of the b-wave was 433 NV, 383
UV, and 383 NV for the day l-normal, day S-normal, and day 5-TPP ferrets,
respectively. The amplitude Of the b-wave in the day 5-normal and day 5-TPP
recordings were significantly different compared to the day l-normal recordings (p <

0.05).

17

DISCUSSION

In this study we used standard histological techniques to quantify the
degenerative effects that TPP has on retinal ganglion cells and their target neurons in
the LGN of the European ferret. Electroretinography was employed to describe
functional deficits resulting from TPP exposure. The effect of TPP on whole-brain
NTE activity also was assessed. The results Of this study show that exposure to TPP
causes a reduction in the total number Of retinal ganglion cells. In addition, TPP
exposure results in a decrease in mean LGN soma size in lamina A and A1, as well
as a Shift in the distribution of LGN neurons toward smaller cell Sizes. Functional
deficits were not detected with the ERG, as a result of TPP exposure. Finally, TPP
exposure produced a 75% reduction in NTE activity compared to control.

This study shows for the first time that TPP, an OPIDN compound, causes
severe degenerative changes in the mammalian retina. The reduction in the number
Of ganglion cells. in the retina (Table 1; Figure 3) suggests that TPP exposure results
in retinal ganglion cell death. This is consistent with the ability Of TPP to induce
neuronal cell death in other regions of the visual system, including the LGN and
visual cortex (Tanaka, 1994, 1999). Hamm et al. (1998) recently reported retinal
ganglion cell degeneration in fish embryos exposed to the organophosphate pesticide
diazinon, and to diisopropylphosphorofluoridate (DFP), an OPIDN compound. The
degenerative effects were more severe in embryos exposed to the OPIDN compound
DFP than those exposed to diazinon. TO the author’s knowledge there are no other

reports of retinal ganglion cell degeneration from exposure to an OPIDN compound.

18

In the nasal retina the reduction in retinal ganglion cell number (21%) was
Significant. However, in the temporal retina the reduction in retinal ganglion cell
number (13%) was not significantly different in TPP-treated ferrets when compared
to normals. Although it is difficult to explain why there appears to be a differential
effect in the temporal vs nasal retina, it may relate to differences in the proposed
mechanism of TPP-induced neuronal death, functional differences Of ganglion cells.
and the overall size Of the cells themselves. It has been suggested that TPP induces
cell death by mitochondrial toxicity. In vitro studies with cultured chromaffin cells
have shown that TPP causes mitochondrial swelling and inhibition of ATP synthesis
(Knoth-Anderson et al., 1993). In vivo studies have shown that TPP decreases
mitochondrial metabolic enzyme activity in Skeletal muscle (Konno et al., 1989).
Mitochondria are the principle organelles responsible for ATP production, through
cellular respiration. The number Of mitochondria in cells is highly variable and
dependent on the metabolic needs Of the individual cell (Wheater and Burkitt, 1987).
Regional differences in metabolic activity in the central nervous system have been
documented by examining mitochondrial number and by measuring cytochrome
oxidase (CO) activity, a mitochondrial enzyme important in the production of ATP
(Carroll and Wong-Riley, 1984; Kageyama and Wong-Riley, 1982, 1984).
Kageyama and Wong-Riley (1984) examined CO activity in the ferret retina and
found that ganglion cells of the same size formed two general populations, those
whose dendrites and somata stained heavily for CO and those that stained lightly.
This suggested that cells of the same size can have different metabolic requirements.
They reported that the two populations Of cells, those with high metabolic activity
and those with low metabolic activity, appear to represent OFF-center and ON-center

19

ganglion cells, respectively. It is known that each region Of the mammalian retina
has functionally distinct (ON vs OFF) subsets Of ganglion cells (Dowling, 1987).
Ganglion cells can be considered either OF F-center or ON-center cells based on their
response to changes in illumination (Dowling, 1987). Ganglion cells respond to
light from a relatively restricted region of Space, referred to as the receptive field.
An increase in illumination in the receptive field center of an ON-center ganglion
cell will cause this cell to increase its firing rate, whereas an OFF—center ganglion
cell responds to a decrease in illumination. The increase in metabolic activity in the
OFF-center cells in comparison to the ON-center cells reported by Kageyama and
Wong-Riley (1984) suggests that these two cell populations might contain different
numbers Of mitochondria. Although we did not identify cells based on function, if
TPP induces cell death by mitochondrial toxicity and ganglion cells in temporal
retina contain, on average, more mitochondria than those in nasal retina, then it is
possible that the cells with fewer mitochondria are more susceptible to the same
concentration of TPP. Additionally, if there is a threshold at which TPP causes cell
death, because of their smaller Size, nasal cells may reach this critical threshold
sooner than the larger temporal cells. In the present study, we found that, in
general, cells are larger in temporal retina compared to nasal retina (Table 1).
Fitzgibbon et al. (1996) and Vitek et al. (1985) also reported that in the ferret the
cell size distribution Shifts toward larger sizes in temporal retina. A differential
effect on retinal ganglion cells based on the function and size Of the cell might
explain why a systemically applied neurotoxicant might not affect all neurons

equally.

20

It has been suggested that retinal ganglion cells die by apoptosis as well as
by necrosis (JOO et al., 1999; Bien et al., 1999). Cell death by apoptosis or necrosis
has been distinguished based on distinctive histopathologic morphologies. Apoptotic
cells tend to shrink while their organelles maintain their integrity until eventually the
nucleus fragments. Apoptosis is associated with a minimal inflammatory response.
Cells are genetically programmed for apoptotic cell death during normal
development and it has been suggested that physiological, pathological, or
pharmacological agents can activate these genetic pathways after maturation
(Thompson, 1998). Cells that die as a result of necrosis appear swollen. Organelles
begin to degenerate and eventually the plasma membrane ruptures. Necrotic cell
death causes an inflammatory response (Thompson, 1998; Albensi, 1999). In our
examination Of retinal ganglion cells exposed to TPP we did not see cellular changes
characteristic Of necrosis, such as swollen somata, pyknOSiS, or inflammation. Since
it was apparent that TPP exposure resulted in a decrease in the number of retinal
ganglion cells, we conclude that this loss was primarily by apoptosis.

In agreement with Tanaka et a1. (1994, 1999) we found severe neuronal
changes in the LGNS Of ferrets exposed to TPP. We extended their findings by
quantifying the degree Of degeneration by measuring LGN soma area. There was no
difference in LGN soma Size between the 5-day and 7-day post-treatment ferrets. but
neurons from both groups Of animals were significantly smaller than normal.
Because cells in lamina A1 and the ventral region Of lamina A are uniform in Size
(Linden et al., 1981), the inclusion of cells located near laminar borders Should not

have affected our results.

21

Based on the results of their previous work Tanaka et al. (1994, 1999)
hypothesized that the age-related sequence of TPP-induced degeneration seen in the
ferret central visual pathway was a result Of anterograde transneuronal degeneration.
By definition, transneuronal degeneration occurs because the insult has effects across
a synapse (Cowan 1970). In order for transneuronal degeneration to be responsible
for the TPP-induced degeneration of the ferret visual system, the first-order neuron
in the central visual pathway, the retinal ganglion cell, must be affected by TPP
exposure, and the synaptic connections between the retina, LGN, and visual cortex
must be mature. In the present study, we demonstrated that TPP affects the retinal
ganglion cells in adult ferrets and we can assume that the synaptic connections along
the central visual pathway in an adult are mature. These findings provide support
for Tanaka et al.’s hypothesis Of anterograde transneuronal degeneration. It is
possible that the age-related pattern seen by Tanaka et al. (1994, 1999) was due to
the immaturity Of the synaptic connections along the central visual pathway.

It is known that the ferret visual system undergoes a Significant period of
postnatal development during the first 4-5 weeks of age (Greiner and Weidman,
1981; Linden et al., 1981; Peduzzi, 1988; Jackson et al., 1989; Hutchins and
Casagrande, 1990; VOigt et al., 1993; Wong et al., 1993). Synaptic connections
within the retina are not fully developed until at least 30 days Of age, which suggests
that the retinal circuitry is not functionally mature until then. Retinal afferents,
initially widespread at birth, segregate into ipsilateral and contralateral projection
patterns within their respective LGN territories by 15-21 days of age. Lamination of

the LGN begins at 7-14 days after birth, achieving adult-like architecture by 35-40

22

days, after the retinal circuitry has matured. In visual cortex, formation and
development Of synapses in all 6 cortical layers occur by 4 weeks Of age. If the
retinal circuitry, retinal afferents, and LGN lamination patterns are not mature until
35-40 days of age, it is possible that retinogeniculate connections are not mature.
Without a mature synapse transneuronal degeneration would not be present at these
earlier ages. It would be necessary to examine the retinal ganglion cells Of ferrets
prior to the onset of retinogeniculate axon degeneration and to examine the
retinogeniculate synapse maturity to further investigate the age-related sequence of
degeneration reported by Tanaka et al. (1994, 1999).

It is important to note that regardless Of the anatomical changes we found in
the retina as a result of TPP exposure, we found no functional deficits that could be
detected using the ERG. The origins Of the a- and b-waves and the STR differ and
have been well documented. Briefly, the a-wave originates from the activity of the
photoreceptors (Brown and Wiesel, 1961; Dowling, 1987) while the b-wave is a
direct result of ON-bipolar cell activity (Tian and Slaughter, 1995). Of particular
interest in the present study was the STR response, since our anatomical analysis
focussed on ganglion cells in the retina.

The STR was first described by Seiving et al. (1986), and it is thought to
originate from either direct or indirect amacrine cell (amacrine cells synapse on
retinal ganglion cells), or retinal ganglion cell activity. The STR can be eliminated
when animals are given an intravitreal injection of aspartate (Wakabayashi et al.,
1988) since aspartate blocks the synaptic transmission between the photoreceptors
and the horizontal cells (Dowling and Ripps, 1972). Furthermore, Bush et al. (1995)
found that the STR remained in rats even after significant losses in photoreceptors.

23

Frishman et al. (1996) recently reported that the STR has a fast component, arising
from amacrine cells, and a Slow component, arising from retinal ganglion cells.
Although many questions remain regarding the origin of the STR, and what the STR
means with respect to retinal function, the present study suggests that retinal
ganglion cell degeneration, as a result of TPP exposure, does not affect the STR
response significantly. Either the STR is not retinal ganglion cell dominated or
ganglion cell numbers can be reduced significantly without affecting the STR.

The results of the ERG recordings do not indicate a treatment effect. Even
though the waveform variability and artifact were present, and the artifact was
unpredictable, it appears from Figure 9a and 9b that the tracings are affected
relatively equally. Figure 9a shows the waveform variability in ERG response that
can be seen between the day 1-norrnal recordings and the day 5-normal recordings.
It is apparent that the amplitudes are smaller in the day 5-normal recordings
compared to the day l-normal recordings, although the overall shapes of the
recordings are similar. The variability in ERG responses was a technical problem
that never was resolved and probably is partially related to the corneal electrode that
was used. The DTL electrode is a thread composed Of many Silver-coated nylon
fibers that are placed over the cornea. It is possible that some Of the fibers did not
lay flat against the cornea or were damaged when the thread was adapted to the
smaller eye of the ferret. Other investigators have reported similar problems with
the variability of ERG recordings in the same and in different animals using the
DTL fiber electrodes (Frishman et al., 1996; Paul Sieving personal comm.)

A second technical problem encountered with the ERG recordings was the
presence Of an electrical artifact. The artifact was unpredictable. it did not

24

consistently increase or decrease as the stimulus intensity increased, and it reversed
polarity at random stimulus intensities. The origin Of the artifact likely arose from
the corneal electrode for two reasons. First, the artifact was considerably diminished
in amplitude as each new electrode system was used, and second, the artifact was
only present when the corneal electrode was on the eye. Even though the presence
and shape of the artifact was random, it appears that it has a very Short latency and
disappears quickly. Typically the a-wave of the ERG is a very fast response
occurring within 10 msec (Wakabayashi et al., 1988). Similarly the STR also is a
rapid response occurring within 75 msec (Wakabayashi et al., 1988; Sieving and
Wakabayashi, 1991; Bush et al., 1995; Frishman et al., 1996). The b-wave has a
longer latency than the a-wave or STR, occurring within 100 msec (Sieving and
Wakabayashi, 1991). Although we did not measure the latency of the artifact it can
be seen in Figures 9a and 9b that it also is relatively short, and thus could
potentially affect the a-wave but probably not the b-wave and the STR. Regardless.
Of the technical difficulties, there still did not appear tO be any evidence of
functional deficits as a result of TPP exposure.

Previous studies have reported changes in a-wave and b-wave latencies and
amplitudes as a result Of exposure to organophosphorous compounds. Imai et al.
( 1974, 1983) exposed rats to fenthion, a non-OPIDN compound, and found that at
low doses the latency Of the a—wave was reduced and the amplitudes Of the a- and b-
wave were increased. At higher doses the reverse was true, the latency increased
and the amplitudes decreased. Yoshikawa et al. (1990) exposed rats to 5 different

non-OPIDN compounds. They found increases and decreases in latency and

25

amplitude measurements of the a- and b-wave that were dependent on which
compound the animals were exposed to. Kreft et al. (1985) exposed trout to an
OPIDN compound and found an increase in a-wave amplitude and a decrease in b-
wave amplitude. Because the results of these studies are equivocal they have
generated some controversy over the role of OPS in ocular toxicology. This in part
has led to the re-evaluation of these compounds by the EPA (Boyes et al., 1994).

Finally, this study measures for the first time the TPP-induced inhibition Of
NTE in the ferret. Our results Show inhibition in excess of the threshold currently
used by the EPA, as an indicator Of OPIDN. Varghese et al. (1995b) have reported
NTE inhibition below the 70% threshold in Japanese quail exposed to TPP, yet these
quail had clinical signs and neuropathology consistent with OPIDN. Whether the
differences are due to species variations or not, the EPA continues to use inhibition
Of NTE in hens as a biomarker. Clearly this may no longer be the most accurate
assay.

In summary, the results of this study provide for the first time quantitative
evidence that exposure to TPP results in severe degenerative changes in both the
retina and the LGN of the European ferret, as well as a significant inhibition of NTE
activity. Furthermore, by documenting degeneration in the retinal ganglion cells we
have provided further evidence that the TPP-induced degeneration previously
described by Tanaka et al. (1994, 1999) in the LGN and visual cortex may be a
result Of anterograde transneuronal degeneration originating at the level of the retinal
ganglion cells. TPP exposure does not result in functional deficits that can be

detected using the ERG, even though there is retinal ganglion cell degeneration.

26

APPENDICES

27

APPENDIX A

Tables 1-4

28

Table 1. The Effect of Triphenyl Phosphite (TPP) on Retinal Ganglion Cell (RGC)
Measurements in the European Ferret.

 

Nasal retina Temporal retina

 

 

 

 

 

 

 

Treatment RGC RGC
(days post- # Of diametersa # of diametersa

Animal ID treatment) RGC (um) RGC (um)
NF55 Normal --- --- 279 14.01 i 2.27
NF13 Normal --- --- 377 13.68 i 2.75
NF52X Normal 414 11.09 i 2.32 497 12.18 i- 2.09
NFSOX Normal 370 12.67 i 2.63 434 13.28 i 2.57
NF28 Normal 335 12.61 i 2.64 376 13.19 i 2.63
NF38 Normal 317 12.85 i 2.59 403 13.18 i 2.91
Mean :2 SD 359 i 12.31 i 0.82 394 13.25 i 0.62
NF36 TPP(5) 244 13.23 i 2.41 --- ---
NF41 TPP(5) 243 13.50 i 2.26 333 13.48 i 2.34
NF10 TPP(5) 287 12.84 i 2.62 348 13.49 i: 2.70
NF25 TPP(5) 309 13.61 i 3.24 298 14.37 i 3.22
NF5 TPP(5) 252 12.63 i 2.09 317 13.78 i 2.35
NF52 TPP(5) 209 13.90 i 2.40 295 14.62 i 2.75
NF54 TPP(5) 331 11.83 i 1.90 415 12.27 i- 2.37
Mean i SD 282 13.08 i 0.70 334 13.67 i 0.83
NF33 TPP(7) 278 13.04 i 1.73 328 14.00 i 2.03
NF56 TPP(7) 302 12.70 i 2.06 320 13.18 i 1.92
NF9 TPP(7) 284 11.76 i 1.88 393 12.56 i 2.54
NF7 TPP(7) 274 11.61 i 2.01 390 12.13 i 2.09
Mean i SD 285 12.28 i 0.70 358 12.97 i 0.81

 

alData presented as mean iSD.

bSignificantly different from normal (p<0.05).

29

Table 2. The Effect of Triphenyl Phosphite (TPP) on Lateral Geniculate Neuron (LGN)
Soma Sizes in the European Ferret.

 

 

 

 

 

 

 

Treatment Mean LGN soma sizea

Animal ID (Days post-treatment) (umz)

NFSS Normal 229 i 112
NF12 Normal 201 i- 93
NF52X Normal 214 i 118
NF50X Normal 217 i 96
NF28 Normal 219 i 118
NF38 Normal 206 i 102
Mean i SD 214 i 10
NF24 TPP(5) 178 i 99
NF32 TPP(5) 182 i 82
NF36 TPP(5) 176 i 97
NF41 TPP(5) 162 i- 67
NF10 TPP(5) 121 i 87
NF25 TPP(5) 171 i 116
NF5 TPP(5) 143 i 78
NF52 TPP(5) 132 i 67
NF54 TPP(5) 131 i 57
Mean i SD 155 i 23b
NF33 TPP(7) 184 i 76
NF56 TPP(7) 168 i 82
NF9 TPP(7) 124 i 42
NF7 TPP(7) 134 i 53
Mean _+_. SD 152 3: 28b

 

8Data presented as mean 2% SD.
bSignificantly different from normal (p < 0.05).

30

Table 3. The Effect Of Triphenyl Phosphite (TPP) on Whole-Brain Neuropathy Target
Esterase (NTE) Activity in the European Ferret.

 

 

Treatment Whole-brain NTE activitya % Inhibition
Normal 1519 i161b --
TPP 371 i 124b 75

 

aActivity expressed as nmoles phenylvalerate hydrolysed/min/gram brain.
bData presented as mean iSD.

31

Table 4. Mean ERG Latency and Amplitude Measurement for the STR, a-wave, and

b-wave, at Each Stimulus Intensity.

 

 

 

 

 

 

 

Day 1 - normal Day 5 - normal Day 5-TPP

Stimulus
intensity
(log Latencya Amplitudea Latency Amplitude Latency Amplitude
cd/s/m2 (msec) (UV) (msec) (NV) (msec) (11V)
-4.5 75:34 71 :37 83:11 83:24 84:8.2 113:74
-4.0 69:31 86:49 83 :11 117:10 75 :10 150:59
-3.5 60:29 81 :44 90 :21 117 :47 60 :20 133 : 54
a-wave
-2.5 26 :14 89 : 52 23 :15 46 : 34 25 :11 97 :14
-2.0 23 :15 98 : 33 26 :14 50 :14 25 :11 93 : 25
-1.5 15:17 132:35 19:8 83:54 10:7 123 :42
-1.0 11:8 211:37 8:9 121:21 5:8 213:100
-0.5 8:9 315 :50 16:1 213 :93 7:13 317: 128
0 8:9 343 :60 10:9 272 :139 0 377 :98
b-wavec
-3.5 75 : 44 112 : 68 49 : 57 108 : 96 72 : 44 53 : 49
-3.0 60 : 21 219 : 90 83 :19 117 : 41 81 :13 123 : 101
-2.5 41 :14 324 : 86 34 :12 217 : 45 63 :16 227 :133
-2.0 38: 15 324:76 41 :8 258 :69 54:8 233 : 115
-1.5 49 : 28 259 : 77 38 : 9 229 : 103 48 : 7 183 : 76
-1.0 38 : 9 294 : 74 41: 8 233 : 62 42 : 7 220 : 49
-0.5 34 : 8 302 : 86 41 : 8 267 : 43 45 : 11 230 : 85

0 34:8 376:59 40:9 294:77 51:8 273:80

 

aData presented as mean : SD.
”Day S-TPP amplitude measurements are significantly different from Day l-normal

(p < 0.05).

cDay 5-normal and Day S-TPP amplitude measurements are significantly different
from Day l-normal ( p <0.05).

32

APPENDIX B

Figures 1-9

Temporal . Nasal . . - Temporal
O O
I . ,

 

f

R
Caudal V Rostral V Caudal
c ‘ l A i<P II) >

V

 

 

Figure 1. Schematic drawing of the matched nasal and temporal regions of the retina
The boxed regions within each retina represent the sample areas while the arrows
represent their approximate projections to the LGN. The filled circle represents the optic

disc in each retina Laminae C, A, and A1 are labeled as well as the perigeniculate
nucleus (P).

34

Figure 2. Photomicrographs of cresyl-violet stained sections from matched nasal
regions Of the retina in a normal (A) and a TPP-treated (B) ferret. Note the decrease
in retinal ganglion cell density in the TPP-treated retina compared to the normal.
Scale bar = 50 um.

35

 

36

500 -

400 -

300 -

200 -

Number of Cells

100 a

 

 

Normal 5-day 7-day

 

Days Post TPP Treatment

 

- Nasal
- Temporal

 

 

 

Figure 3. Histograms comparing the mean number of retinal ganglion cells in matched
nasal and temporal regions of the retina in normal and TPP-treated ferrets 5 and 7 days
after treatment (* p < 0.05 Kruskal-Wallis w/Mann Whitney). Nasally there were 21%
fewer cells in the TPP-treated ferrets than in the normal ferrets, while temporally there
were 13% fewer ganglion cells.

16*

Mean Cell Diameter (um)

 

 

Normal 5-Day 7-Day

 

Days Post TPP Treatment

 

- Nasal
- Temporal

 

 

 

Figure 4. Histograms comparing the mean soma diameters Of retinal ganglion cells in
matched nasal and temporal regions of normal and TPP-treated ferrets 5 and 7 days after
treatment. Mean soma size was similar for all groups (Kruskal-Wallis w/Mann Whitney).

38

Figure 5. Smoothed histograms showing the size distributions of retinal ganglion
cells from matched nasal and temporal regions of the retinae in normal and TPP-
treated ferrets 5 and 7 days after treatment. The cell Size distributions were similar
for all groups (Kolmogorov-Smimov test).

39

Percent of Cells

Percent of Cells

Nasal

4o-

30-

 

20~

 

 

 

10-

 

 

 

 

Cell Diameter (um)

Temporal

40-

30‘

 

20‘

 

 

 

10~

 

 

 

 

Cell Diameter (um)

40

Figure 6. Photomicrographs Of cresyl-violet stained parasagittal sections of the ferret
LGN. (A) Low power (4x) view illustrating the laminar divisions C, A1, A, and the
perigeniculate nucleus (P). (B) Low power view of a 5-day post-treatment LGN.
Note the pale appearance Of the TPP-treated LGN compared tO normal. Scale bar =
50 um. (C) High power (40x) view of neurons in lamina A1 Of a normal LGN. (D)
High power view Of neurons in lamina A1 of a 5-day post-treatment LGN. Note the
irregular shaped, fragmented, pale staining, and pyknotic nuclei (arrows) in the TPP-
treated in comparison to normal. Scale bar = 500 um.

41

 

 

 

42

250 -

200 -

150 -

100 -

Mean Soma Area (umz)

50-

 

 

Figure 7. Histogram comparing the mean cross-sectional area in LGN neurons from
matched regions in lamina A and A1 in normal and TPP-treated ferrets 5 and 7 days after
treatment. The mean cross-sectional areas of LGN neurons in TPP-treated ferrets were
30% smaller than those of normal ferrets (* p < 0.05 Kruskal-Wallis w/Mann Whitney).

43

18" *
16-
14-

12-

 

10*

 

 

 

Percent of Cells
m

 

 

 

5 -
4 ..
2 .,
o —1
0 100 200 300 400 500 600 700 800

Cell Area (umz)

Figure 8. Smoothed histograms comparing the size distributions of neurons from matched
regions in lamina A and A1 Of the LGN in normal and TPP-treated ferrets 5 and 7 days
after treatment. There was a Significant shift toward smaller soma sizes in the TPP-
treated ferrets compared to normal (* p < 0.05, Kolmogorov-Smimov test).

44

Figure 9. Corneal ERG responses. (A) Responses from a normal ferret using a
DTL fiber electrode at each stimulus intensity on day 1 and day 5. The STR, b-
wave, and a-wave are labelled with arrows. The arrowheads point out the artifact.
(B) Responses from a 5-day post-treatment ferret using a DTL fiber electrode at
each stimulus intensity on day 1 before treatment and day 5 after treatment. The
STR, b-wave, and a-wave are labelled with arrows. The arrowheads point out the
artifact.

45

(:d"S/rn2 Day 1 Day 5
-4.5

—4.0
-3.5
-3.0

-2.5

-2.0
-l.5

-1.0

-0.5

 

46

y
a
D
b V
2m 0. 5.
vs 4 2..
d

BIBLIOGRAPHY

48

BIBLIOGRAPHY

Abou-Donia MB. Organophosphorus ester-induced delayed neurotoxicity. Annu Rev
Pharmacol Toxicol 1981;21:511-548.

Abou-Donia MB, Lapadula DM. Mechanisms Of organophosphorus ester-induced
delayed neurotoxicity: type I and type II. Annu Rev Pharmacol Toxicol 1990;30:405-
440.

Abou-Donia MB. Triphenyl phosphite: A type II organophosphorous compound-
induced delayed neurotoxic agent. In: J. E. Chambers and P. E. Levi, eds.
Organophophates: Chemistry, Fate and Effects. San Diego: Academic Press,
1992;327-351.

Albensi BC. Apoptosis and biochemical cascades. Drug News Perspect 1999;12:1-5.

Baron RL. Delayed neurotoxicity and other consequences Of organophosphorous
esters. Annu Rev Entomol 1981;26:29-48.

Berends F. Mechanisms of ageing of organophosphate-inhibited esterases. In: F. De
Matteis and E. A. Lock, eds. Selectivity and Molecular Mechanisms of Toxicity. New
York: Macmillan Publishing Company, 1987;125-152.

Bien A, Seidenbecher CI, Bockers TM, Sabel BA, Kreutz MR. Apoptotic versus
necrotic characteristics Of retinal ganglion cell death after partial Optic nerve injury.
JNeurotrauma 1999;16:153-163.

Boyes WK, Tandon P, Barone JS, Padilla S. Effects Of organophosphates on the
visual system of rats. J Appl Toxicol 1994;14:135-143.

Brown KT, Wiesel TN. Localization of origins of electroretinogram components by
intraretinal recording in the intact cat eye. J Physiol 1961;158:257-280.

Bush RA, Hawks KW, Sieving PA. Preservation Of inner retinal responses in the
aged royal college Of surgeons rat. Evidence against glutamate excitotoxicity in
photoreceptor degeneration. Invest Ophthalmol Vis Sci 1995;36:2054-2062.

Carroll EW, Wong-Riley MTT. Quantitative light and electron microscopic analysis
of cytochrome oxidase-rich zones in the striate cortex Of the squirrel monkey
(Saimiri sciureus). J Comp Neurol 1984;222:1-17.

Cowan WM. Anterograde and retrograde transneuronal degeneration. In: W. J. H.

Nauta and S. O. E. Ebbesson, eds. Contemporary Research Methods in
Neuroanatomy. New York: Springer-Verlag, 1970;217-251.

49

Dowling JE, Ripps H. Adaptation in Skate photoreceptors. J Gen Physiol
1972;60:698-719.

Dowling JE. The Retina.“ An Approachable Part of the Brain. Cambridge: Harvard
University Press, 1987.

Ecobichon DJ. Organophosphorus ester insecticides In: D. J. Ecobichon and R. M.
Joy, eds. Pesticides and Neurological Diseases. 2nd ed. Boston: CRC Press,
1994;171-249.

Ehrich M. Neurotoxic esterase inhibition: Predictor of potential for
organophosphorus-induced delayed neuropathy. In: J. N. Blancato, R. N. Brown, C.
C. Dary and M. A. Saleh, eds. Biomarkers for Agrochemicals and Toxic Substances.
Washington, DC: American Chemical Society, 1996;79-93.

F itzgibbon T, Wingate RJ, Thompson ID. Soma and axon diameter distributions and
central projections Of ferret retinal ganglion cells. Vis Neurosci 1996;13:773-786.

Frishman LJ, Shen FF, Du L, Robson JG, Harwerth RS, Smith EL, Carter-Dawson
L, Crawford ML. The scotopic electroretinogram of macaque after retinal ganglion
cell loss from experimental glaucoma. Invest Ophthalmol Vis Sci 1996;37:125-141.

Greiner JV, Weidman TA. Histogenesis of the ferret retina. Exp Eye Res
1981;33:315-332.

Hutchins JB, Casagrande VA. Development Of the lateral geniculate nucleus:
Interactions between retinal afferent, cytoarchitectonic, and glial process lamination
in ferrets and tree shrews. J Comp Neurol 1990;298:113-128.

Imai H. Toxicity Of organophosphorous pesticde (Fenthion) on the retina:
Electroretinographic and biochemical study. Acta Soc Ophthalmol Jpn 1974;78:163-
172.

Imai H, Miyata M, Uga S, Ishikawa S. Retinal degeneration in rats exposed to an
organophosphate pesticide (Fenthion). Environ Res 1983;30:453-465.

Jackson CA, Peduzzi JD, Hickey TL. Visual cortex development in the ferret. 1.
Genesis and migration of visual cortical neurons. J Neurosci 1989;9:1242-1253.

Jamal GA. Neurological syndromes Of organophosphorus compounds. Adverse Drug
React T oxicol Rev 1997;16:133-170.

Johnson MK. Improved assay of neurotoxic esterase for screening organophosphates
for delayed neurotoxic potential. Arch T oxicol 1977;37:113-115.

50

Johnson MK. The target for initiation Of delayed neurotoxicity by organophosphorus
esters: Biochemical studies and toxicological applications. In: E. Hodgson, J. R.
Bend and R. M. Philpot, eds. Reviews in Biochemical Toxicology. New York:
Elsevier, 1982;141-212.

Johnson MK. Organophosphates and delayed neuropathy- is NTE alive and well?
T oxicol Appl Pharmacol 1990;102:385-399.

JOO CK, ChOi JS, KO HW, Park KY, Sohn S, Chun MTT. Oh YJ, Gwag BJ.
Necrosis and apoptosis after retinal ischemia: Involvement of NMDA-mediated
excitotoxicity and p53. Invest Ophthalmol Vis Sci 1999;40:713-720.

Kageyama GH, Wong-Riley MTT. Light and EM. localization of cytochrome
oxidase staining in the lateral geniculate nucleus. Soc Neurosci Abstr 1982;8z208.

Kageyama GH, Wong-Riley MTT. The histochemical localization Of cytochrome

oxidase in the retina and lateral geniculate nucleus Of the ferret, cat, and monkey,
with particular reference to retinal mosaics and ON/OFF-center visual channels. J
Neurosci l984;4:2445-2459.

Kamata R, Suzuki T, Saito S, Kofujita H, Ota M, Kobayashi H. Lack of correlation
of organophosphorus-induced delayed neurOpathy with neuropathy target esterase in
hens and japanese quails. J Health Sci 1999;45:209-216.

Knoth-Anderson J, Abou-Donia MB. Differential effects of triphenyl phosphite and
diisopropyl phosphorofluoridate on catecholamine secretion from bovine
adrenomedullary chromaffin cells. J T oxicol Environ Health 1993;38:103-114.

Koelle GB. Pharmacology of organophosphates. J Appl Toxicol 1994;14:105-109.
KonnO N, Katoh K, Yamauchi T, Fukushima M. Delayed neurotoxicity of triphenyl
phosphite in hens: Pharmacokinetic and biochemical studies. T oxicol Appl Pharm

1989;100:440-450.

Kreft WD, Hoffert JR, Fromm PO. Action Of organophosphates on the
electroretinogram Of rainbow trout. Exp Biol 1985;44:19-27.

Lehning EJ, Tanaka D, Jr., Bursian SJ. Triphenyl phosphite and diisopropyl
phosphorofluoridate produce separate and distinct axonal degeneration patterns in the

central nervous system Of the rat. F undam Appl Toxicol 1996;29:110-118.

Linden DC, Guillery RW, Cucchiaro J. The dorsal lateral geniculate nucleus Of the
normal ferret and its postnatal development. J Comp Neurol 1981;203:189-211.

Lotti M, Becker CE, Aminoff MJ. Organophosphate polyneuropathy: Pathogenesis
and prevention. Neurology 1984;34:658-662.

51

Marrs TC. Organophosphate poisoning. Pharmac Ther 1993;58:51-66.

Peduzzi J. Relationship of geniculocortical afferents to developing neurons in the
ferret visual cortex. Soc Neurosci Abstr 1988;14:743.

Richardson RJ. Assessment of the neurotoxic potential of chlorpyrifos relative to
other organophosphorus compounds: A critical review of the literature. J T oxicol
Environ Health 1995;44:135—165.

Sieving PA, Frishman LJ, Steinberg RH. Scotopic threshold response of proximal
retina in cat. J Neurophysiol 1986;56:1049-1061.

Sieving PA, Wakabayashi K. Comparison of rod threshold ERG from monkey, cat
and human. Clin Vis Sci 1991;6zl71-179.

Stephans T. Organophosphate pesticides: Neurotoxins threaten farm workers. The
Journal of NIH Research 1991;3221-24.

Sultatos LG. Mammalian toxicology of organophosphorus pesticides. J Toxicol
Environ Health 1994;43:271-289.

Tanaka D, Jr., Bursian SJ, Lehning EJ, Aulerich RJ. Exposure to triphenyl phosphite
results in widespread degeneration in the mammalian central nervous system. Brain
Res 1990;531:294-298.

Tanaka D, Jr., Bursian SJ, Aulerich RJ. Age-related effects of triphenyl phosphite-
induced delayed neuropathy on central visual pathways in the European ferret
(Mustela putorius furo). Fundam Appl Toxicol 1994;22:577-587.

Tanaka D, Jr., Lin M, Powell D, Morgan M, Lipsitz L, Aulerich RJ, Bursian SJ.
Effects of dark-rearing on triphenyl phosphate-induced neuropathy in the visual

system of the developing European ferret (Mustela putorius furo). J T oxicol Environ
Health 1999;58:215-231.

Thompson EB. Special topic: Apoptosis. Annu Rev Physiol 1998;60:525-532.

Tian N, Slaughter MM. Correlation of dynamic responses in the ON bipolar neuron
and the b-wave of the electroretinogram. Vision Res 1995;35:1359-1364.

U. S. EPA. Triphenyl Phosphite. CAS No. 101-02-0, Washington, D. C. 1985
Varghese RG, Bursian SJ, Tobias C, Tanaka D, Jr. Organophosphorus-induced
delayed neurotoxicity: a comparative study of the effects of tri-ortho-tolyl phosphate

and triphenyl phosphite on the central nervous system of the Japanese quail.
Neurotoxicology 1995a;1 6:45-54.

52

Varghese RG, Bursian SJ, Tobias C, Tanaka D, Jr. Triphenyl phosphite-induced
neuropathy in the avian forebrain: A silver impregnation study of the visual and
auditory systems of the Japanese quail. Neurataxicalagy 1995b;16:lOS-113.

Veronesi B, Dvergsten C. Triphenyl phosphite neuropathy differs from
organophosphorus-induced delayed neuropathy in rats. Neurapathal App] Neurabial
1987;13:193-208.

Vitek DJ, Schall JD, Leventhal AG. MorphologY, central projections, and dendritic
field orientation of retinal ganglion cells in the ferret. J Comp Neural 1985;241:1-11.

Voigt T, De Lima AD, Beckmann M. Synaptophysin immunocytochemistry reveals
inside-out pattern of early synaptogenesis in ferret cerebral cortex. J Comp Neural
1993;330:48-64.

Wakabayashi K, Gieser J, Sieving PA. Aspartate separation of the scotopic response
(STR) from the photoreceptor a-wave of the cat and monkey ERG. Invest
Ophthalmol Vis Sci 1988;29:1615-1622.

Wheater PR, Burkitt HG. Cell structure and function. Functional Histology. A Text
and Colour Atlas. 2nd ed. New York: Churchill Livingstone, 1987;2-28.

Windebank AJ. Peripheral neuropathy due to chemical and industrial exposure. In:
W. B. Matthews, ed. Handbook of Clinical Neurology: Neurapathies: Elsevier
Science Publishers, 1987;263-292.

Wong ROL, Meister M, Shatz CJ. Transient period of correlated bursting activity
during development of the mammalian retina. Neuron 1993;11:923-938.

Yoshikawa H, Yoshida M, Hara 1. Electroretinographic changes induced by
organophosphorous pesticdes in rats. J T axicol Sci 1990;15:87-95.

Zahs KR, Stryker MP. The projection of the visual field onto the lateral geniculate
nucleus of the ferret. J Comp Neural 1985;241:210-224.

53