Effect of dietary zinc levels on photic retinopathy in the pig


Material Information

Effect of dietary zinc levels on photic retinopathy in the pig
Physical Description:
vii, 260 leaves : ill. ; 29 cm.
Smith, Patricia J., 1963-
Publication Date:


Subjects / Keywords:
Swine   ( mesh )
Zinc -- adverse effects   ( mesh )
Retina -- physiopathology   ( mesh )
Retina -- chemistry   ( mesh )
Light -- adverse effects   ( mesh )
Food, Fortified   ( mesh )
Retinal Diseases -- chemically induced   ( mesh )
Retinal Diseases -- veterinary   ( mesh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1995.
Includes bibliographical references (leaves 242-259).
Statement of Responsibility:
by Patricia J. Smith.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 50514609
System ID:

Full Text








I am most grateful for the support and guidance of my committee chairman,

Dr. Don A. Samuelson, who has helped me throughout my career as a veterinary

ophthalmologist With out his support and friendship many of my accomplishments

would not have been possible. I would also like to thank Dr. Dennis Brooks, my

friend and colleague, who has always supported my endeavors with all his resources

and undying friendship. Dr. David Whitley has worked with me for several years and

I am grateful to him for his help and support. Special thanks go to Dr. Robert

Cousins who was willing to explore working with a veterinary ophthalmologist and a

graduate student who also had clinical duties. I am very grateful to him for his

support and guidance as well as the opportunity to work in his laboratory and utilize

its facilities. Other thanks go to my friends and family who have supported all my

academic endeavors, and especially to Ms. Patricia Lewis who has always been there

for technical support and as a friend.


ACKNOWLEDGMENTS .........................................ii

ABSTRACT ................. .. ............... ............ vi


1. INTRODUCTION ............. .............. .............

2. LITERATURE REVIEW ...................................4

Photic injury ........... ....... ......... .... ... .. ....... .4

Nechnis s .. o ...... ... ... ... ... ... .. 6
Protective Adaptations in the Eye ........................... 12

Zincand the Eye ......................... .... ... ........ 17
Metallthionein .......................... ..... ........ 26
Light, Zinc and Age-related Macular Degeneration ................ 29

3. MATERIALS AND METHODS .............................. 31

M materials ..............................................31

Animals .................... ........................31
Phtic Injury Unit and Lighting Conditios .................... 31
Dies ......................................... ... ...33

Methods ............................... ..... ........ ..36

Photic Injury ................. ....................... .......36

Pilot study ............................... .... .. ....36
Part 1: Intermittent light stress .......................... 36
Part 2: Continuous light stress .......................... 36


Clinical Exa nations .................................. 38
Tissue Collection and Processing ........................... 39
Morphological Examination and Energy Dispersive
X-ray Microanalysis ................................. 42
Zinc and Copper Detemination by Atomic Absorption
Spectrophotometry................................... 44
Immunohistochemistry ..................................46
MetalothioneiM Assay .................................. 49
Statisical Methods ..................................... 51

4. RESULTS ................................................ 55

Animals ............................................... 55
Funds Photography ...................................... 55
Electrophysiologic Studies .................................. 58
Blood and Tissue Analysis .................................. 58
Normal Porcine Retina and Control (unexposed) Animals ............ 67

Control lighting (14 days on diet) .......................... 75
Control Lighting (33 days on diet) .......................... 78

Photic Injury ...................................

Pilot Study: 36 and 72 hours of Intermnent Light and
Heat Stress ...............................
Part 1: Intermittent Light Stress (36 hours : 72 hours Post-
Exposure ........................... ....
Part 1: Intermittent Light Stress (36 hours): 3 weeks Post-
Exposure ................................
Part 1: Intermittent Light Stress (72 hours): 72 hours Post-
Exposure ...............................
Part 1: Intermittent Light Stress (72 hours): 3 weeks Post-
Exposure ................................
Part 2: Cotinuous Light Stress (72 horns): 72 hours Post-
Exposure ................................
Part 2 Continnous Light Stress (72 homrs): 3 weeks Post-
Exposure ................................

TI-nunnhistoclue istry ............................
Metallothionein Assay ............................
Energy Dispersive X-ray Microanalysis of RPE and Choroidal
Melanin Granules .............................

......... 81

......... 81



........ 103

........ 107

........ 170

........ 176
........ 181


Control Animals ......... ..... ........... .... .... .. 191
Continuous Light Stress (72 hours): 72 hours Post-
Exposre .......1.... .... ................... .. 195
Three Weeks on Experimental Diets Light Stress and
Contrls ................................. .......199

5. DISCUSSION ........... ......................... ....202

Porcine Animal Model ........ .......................... .. 202
ZincStatus ...... .... ....... ...... ....... ...... ......205
Photic Injury ..................... .... .... .. ..... ......209

Metallothiodne ....................................... .. ......228
Elemental Analysis of Melanin Ganules .................... 232
Summary ............................................. 238

REFERENCES .................. ..................... .......242

BIOGRAPHICAL SKETCH ..................................... 260

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Patricia J. Smith

May 1995

airman: Dr. Don A. Samuelson
Major Department Veterinary Medicine

The pig was used as a model to study the effects of light and dietary zinc on the

rtina. Animals were fed one of three experimental diets with low, normal and high zinc

levels for one week prior to light exposure. Animals were then exposed to various

schedules of intermittent or continuous high intensity light (630 foot candles) in a photic

injury unit and then sacrificed 72 hours or 3 weeks post-exposure. The zinc status of

animals was evaluated by several methods to allow observation of any influence of

dietary zinc an light damage. Metafloonin in the neural retina was quantitated by the

cadmium-hemoglobin binding assay.

Two types of light damage were induced in the porcine retina. The first type was

chracterized by multifocal outer nuclear layer pyknosis and outward movement through

the external limiting membrane. The second type of damage was characterized by

diffuse retinal edema, INL and ONL necrosis or sweling, and RPE damage. This more

severe type of damage was rarely observed in the inennittent light group, was common

in the continuous light group, and was noted in heat stressed animals exposed to

intermittent light.

The light induced retinal damage was mild and asymmetric in many animals and

at three weeks post-exposure recovery had occuned which suggests the light dose was

near the threshold for photic injury. The experimental diets did not significantly alter the

zinc status of animals so no conclusions can be drawn about the effect of zinc status on

photic injury in the porcine retina. There was, however, a trend towards greater outer

nuclear layer cell death in animals with lower kidney metallothionein, a good indicator

of zinc status. Metallthionein levels were quantitated in the neural retina but were not

significantly affected by dietary zinc level or by light-stress

A low level of melanogenesis was observed in the RPE of unexposed eyes of

young pigs. Low and high dietary zinc levels appeared to increase melanogenesis in the

porcine RPE. Light stress resulted in an increased frequency of melanogenesis in the

porcine RPE.


Excessive exposure to high intensity light results in pronounced damage to the retina.

The cumulative effect of light exposure over many years is one factor thought to contribute

to the development of age-related macular degeneration, the major cause of blindness in

humans over fifty. The concentration of zinc in ocular tissues is higher than in any other

tissue in the body, suggesting that zinc has important functions in the normal eye. A recent

study investigating the effect of oral zinc administration on the visual acuity of human

subjects with age-related macular degeneration revealed that individuals receiving zinc

supplements demonstrated signifianly less visual loss thannon-treated individuals' Using

the pig eye as a model for the human eye, the proposed study will investigate the effect of

zinc status on experimental induced photic retinopathy.

The primary goals of this study are 1) to examine clinically and morphologically the

damaging effects of excessive light exposure on the retina and choroid of the pig; 2) to

examine by energy dispersive x-ray micronalysis any change in the elemental content of

melanin granules in the retinal pigment epithelium and charoid that may occur during

excessive light exposure; 3) to document the presence and distribution of the imptant zinc

binding protein metallothionein in ocular tissues utilizing imminohisochemical staining

and a cadmium-hemoglobin binding assay and 4) to examine and compare the influence

of zinc supplementation and deficiency on photic injury, melanin granule elemental content

and metallothionein levels.

Evidence exists which indicates that zinc can provide the retina cytoprotection during

illumination. Exposure to excessive light has been shown to damage the outer layers of the

retina and retinal pigment epithelium via disruption of their cell membranes.4 The

production of free radicals, H20O, OH and 02-, during excessive light exposure are key

elements in the pathogenesis of photic retinopathy." Zinc is found in high concentrations

in the eye, especially the retina and choroid. Certain zinc-associated proteins including

superoxide dismutase (SOD) and metallothionein are believed to play important roles as

antioxidants in protecting tissues against free radical damage.4 Superoxide dismutase is

abundant inthe retina and RPE.' A metallothionein-like protein was isolated in the bovine

retina." Subsequently, metallothionein has been found in human RPE cells and in the RPE

and inner retinal layers ofthe rat"I" In studies on the bovine retina, zinc had a non-unifonr

distribution being most concentrated in the outer segments which are very susceptible to

lipid peroxidation during light and/or oxidative stress." The protective role of zinc against

photic injury is substantiated by another report that demonstrated zinc and tamine had a

protective effect on peroxidation-induced damage of photoreceptor outer segments. In

these studies protection was believed to result from a membrane stabilizing effect

Zinc may provide indirect protection against photic injury due to its essential role in

maintaining the health of the RPE and retina. For example, retinol dehydrogenase is a zinc

metalloenzyme that catalyzes the conversion of retinol to retinal which is used to recycle the

photopigment rhodopsin.

The importance of zinc for normal retinal health is exemplified by the significant

visual impairment and pathologic changes that can occur in ocular tissues of zinc-deficient

animals and humans. Studies in boars fed diets deficient in zinc led to the development of

a retinal degeneration characterized by dilation and disruption of the outer segments

(especially of cone cells in the central retina) at the face of the retinal pigment epithelium."

Since zinc is such a key element in the health of the retina, and zinc deficiency has

been shown to induce retinal degeneration, it was hypothesized that zinc deficiency might

increase the susceptibility of the retina to light-induced injury. Conversely, supplementation

of zinc, an antioxidant nutrient, might protect against the oxidative damage that occurs in

photic injury. Specifically, the hypotheses are as follows: 1) Dietary zinc supplementation

has no protective effect on the degree of retinal damage produced by excessive light

exposure, and the alternative, zinc supplementation is protective against photic injury to the

retina, and 2) a zinc-deficient state does not increase the susceptibility of the retina tophotic

injury, and the alternative, a zinc-deficient state increases the susceptibility of the retina to

photic injury. This investigation will examine the effect of altered zinc status on light

induced retinal damage in order to provide a better undstandig of how zinc and light may

contribute to the retinal pathology seen in age-related macular degeneration.

The pig was chosen as ananimal model due to important anatomical similarities with

the human eye, including the presence of numerous cones that are concentrated centrally,

the presence of red, blue and green sensitive cones, a similar retinal vascular pattern, and the

absence ofa tapetum lucidum that is present in most nonprimate mammalian species.2


Photic Iniry

Retinal damage due to excessive or prolonged light exposure is well documented and

has been produced in the monkey, rat, mouse, rabbit, pigeon, dog, piglet, and ma~n.- The

degree and type of light damage to the retina depends on the type of stimulus used to

produce the damage. Light damage may be assessed by light and electron microscopy, and

by the electroretinogram. The most definitive and detailed morphologic picture of light

damage is presented by studies with light and electron microscopy?.31'

Several factors influence the type ofretinal damage induced by lit. Variables that

have beenshowntoinflence light damage includebodytemperatur, nutritional status, age,

species, strain or breed, light intensity, duration of exposure, schedule of exposure

(continuous vs. intermittent), light source (candeent vs fluorescent wavelength, and

lighthistory (cyclic, dark-adapted, tc.).e Becausemanyvariables influence thedegreeand

type of damage, valid cmpariso between studies is difficult. The greatest amount of

coistency is seen in the order and type of structural changes that occur in the retina

Most photic injury studies have found that the first alterations occur in the outer

segments(OS) of the photoreceptor cells. In ats, mice, monkeys and in piglets, rod OS

appear swollen, lame structure is disrupted, and disc membra thin and form tubules

and vesicle9s."'*13 Similar disruption of cone outer segments was observed in pigeons


exposed to fluorescent light Pyknosis and swelling of mitochndria is then observed in

the inner segment. Rod OS may become large, round or pear-shaped bodies. Microvilli of

the RPE increase in length and number, and increased phagocytosis of the disrupted OS is

evident Outer segments lose connections with inner segments and begin to disappear. Inner

segments shrink and pyknosis begins to occur in the outer nuclear layer. Ultimately, all

damaged photorecept s disappear. Kuwabara and Gom commented that with extremely

intense light or elevation of body temperature changes may occur in the cell body prior to

outer segment disruption. Although inner retinal layers usually remain normal in

appearance, severe photic damagemay result in degeneration of inner retinal layers. Miller

cells proliferate to maintain retinal architect. If photoreceptors completely disappear

Miller cell processes and RPE microvili interdigitate in a permanent adhesion.0

In Kuwabam and Gom's studies in rats, bids and primates, the changes described

above occurred over a one month period following exposure to fluorescent light. Similar

changes have also been observed in primates m, pigeons piglets", and rabbits.

The abovedescriptions represent the general sequence of changes that occurinphotic

injury, bt as mentioned previously the changes seen can vary depending on the

experimental conditions Depending on the study, varying degrees of RPE damage have

also been observed. For example, in some studies, incandescent light, very intense or long

light exposure, and elevated body temperature resulted in more marked pigment epithelial

damage.m 44' It has been proposed that RPE damage is a result of thermal injury due to

absorption of light and subsequent heat generation by melanin granules of the RPE.

investigators also disagree on when RPE damage occus. For example, in the rat photic

injury model, Hansson believes damage occurs prior to receptor destruction, Noell believes

it occurs simultaneously and Dowling believes it occurs subsequent to receptor


Most studies have shown that with mild or moderate injury, retinal changes are

reversible. Recovery begins in the RPE and proceeds very slowly with the formation of new

lamellar membranes in the outer segments. Kuwabara believes the preservation of the

pigment epithelium and the photoreceptor cell body are necessary for recovery. 34w


Light maydamagethe retina in three ways: thrmal acoustic and photoche 2ical.-

In theral damage significant temperature elevation is produced in the retina to cause

coagulation of proteins and membranes leading to cellular damage. Acoustic damage is the

result ofultrashot exposures to electric fields, acoustic signals or shock waves. Exposure

results in production of sonic transients that mechanically disrupt tissues. In photochemical

damage molecular sensitizer is activated by incident photons which result in the production

of free radicals. Free radicals are short-lived atoms or molecules that have unpaired

electrons in their outer orbital rim. Because of this unstable structure they are reactive with

other molecules and toxic to tissues." Photochemical injury is usually produced with light

intensiies several orders of magnitude below that needed to produce a thermal bum and

requires longer exposure times to produce damage, i.e, seconds, minutes or days vs. pico-

or nano- seconds for the other mechanisms. It is important to remember that there is no

sharp line of demarcation between these three types of retinal injury. Other proposed

mecanisms of photic damage to the retina include metabolic disorders from extended

overbleaching and production of toxic products other than free radicals in the outer layers

of the retina.

Photooxidation and lipid peroxidation. The mammalian retna is unique in that it is

the only tissue where light is focused on a group of highly oxygenated cells. In the eye,

blood flow per gram of tissue is higher than any other body tissue; for example, it is four

times greater in the choroid than in the renal cortex.C The mammalian retina again is

unique in its high content of polyunsaturated fatty acids (PUFAs). Polyunsaturated fatty

acids are highly susceptible to lipid peroxidation induced by free radicals (esp. oxyradicals)

and their susceptibility is directly proportional to the number of double bonds, i.e., the

degree ofnsaturation. The content and distribution of fatty acids in the retinas of humans,

pigs, dogs, sheep, rabbits and cows are nearly identical." Approximately 45% of the fatty

acids in retinal membranes contain polyunsaturatcs and about thirty percent of the PUA's

in the retina is 22:6 doosahexaenoic acid (6 double bonds).Y A reinvestigation of the

PUFA content in frog, bovine and rat eyes found 46-50% 22:6 dcosahexaenoic acid in rod

outer segment membranes. The combination of high oxygen tension and light exposure

to the rod OS renders them particularly susceptible to light-induced oxidative damage

(photoxidation). Potoxidation and lipid peroxida are both believed to be significant

contributors to photic retinopathy.4 '

In photooxidative damage a molecular sensitizerS) absorbs a photon of light and is

converted to an activated single state(S). The 'S molecule is very unstable and short-lived

and may dissipate its energy by florescence, interaction with the solvent or by radiatiauless

transition to the triplet sttae(S). "S The 3S molecule mediates two types of

photosensitized reactions. In type I the 3S reacts with a substrate molecule, which may be

another sensitizer molecule. In type II reactions the sS molecular reacts with oxygen to

generate single oxygen or the superoxide ion radical (O4-). Singlet 06 reacts with water

or other oxygen molecules and through univalent elect transfer can generate superoxide

ion, hydroxyl radicals(OH) and hydrogen peroxide. In addition, hydrogen peroxide can

react withthe superoxide radical or ferrous iron (Feton's reaction) to produce the extremely

reactive hydroxyl radical. These oxyradicals can directly damage tissues and especially can

react with biomembranes to induce lipid peroxidation

Lipid peroxidation is induced when a free radical (R), ie., the hydroxyl radical,

extracts a hydrogen and electron from a PUFA (LF) to form a lipid free radicaLSs This

represents the initiation of the lipid peroxidation chain reaction:

RH + LH,- RH, + L

Propagation occurs when the lipid radical reacts with oxygen to produce lipid peroxide

radicals (LHO0) and these peroxyradicals can react with other PUFA's to generate lipid

hydropenoxides (ULOOH) and more lipid free radicals. Ie net effect is to oxidize lipids

to hydroperoxides:

LH" + 0 LHOO


Since more lipid radicals are generated that may react with other PUFA's, the cycle

continues and is known as autoxidation." Lipid hydroperoxides are capable of causing

lipid peroxidations in vivo and in vitro by a chain reaction which may occur by

fragmentation to an alknxy radical and hydroxyl radical each of which continue chain

reactions with additional saturated lipids. The reactions above are only a few of many

potential reactions that have the potential for enormous amplification of the initiating event

which can be caused by any radical and could account for extensive lipid peroxidation

initiated by very small amounts of oxygen derived radicals.

The lipid peroxide radical can decompose to form the highly reactive molecule

malonaldehyde. Malnaldehyde is stable enough to diffuse to distant sites within the cell

such as the nucleus." Malnaldehyde readily condenses via Schiff base formation with

nitrogenous materials including amino acids peptides, phospholipids and nucleic acids As

a result, imprtantfunctionsofproteins structurall and enzymatic) and DNA (replicati and

transcription) are inhibited leading to cell damage and death.

Several investigators have provided in vivo and in vitr evidence that substantiates

the occurrence ofpoto dati and lipid peroxidation in retinas exposed to light. '4

Ham has shown a sharp drop in the radiant exposure threshold with increased partial

pressure of oxygen in material blood in macaque monkeys exposed to blue light.4 In that

experiment petreatmet with methylprednisone (amemnmne stabilizer) rbeta carotene (an

antioxidant andpotent scavenger ofsinglet oxygen) protected againsttheoxygenpotentiated

damage. In another study, dark adapted frog retinas were shown to accumulate

hydroperoxides when exposed to light. Thme action spectrum of this process paralleled the

absorption spectrum ofhodopsinhaving suggested this photopigmt may participateinthe

induction of lipid peroxidation. Delmelle exposed retinal enriched liposomes to light and

observed malonadehyde production along with increased hydroperoxides, increased

membrane fluidity and liposome lysis.n Porcine retinas inadiated with ultraviolet light had

a significant increase in malonaldehyde, a product of lipid peroxidation of membranes"

Increase in hydroperkides has also been shown in rabbit retinas exposed to damaging

light.' Vitamin E deficient Wiser rats exposed to high intensity light had three times the

accunmlatio of hydroperoxides compared to controls indicating this antioxidant nutrient

was protecting against lipid peoxidation mediated damage.3

In rats exposed to varied lighting conditions photoreceptor degeneration was

accompanied by a selective loss of 22:6 docosahexaenoic acid and a simultaneous increase

in hydroperaides."- Since 22-6 is the major PUFA of rod outer segment membranes, its

selective loss indicates lipid peroxidation of outer segment membranes may be a significant

factor in phoic retinopathy.

An important unanswered question is which molecules) in the retina represent

sensitizes Absorption of photons by sensizers generates the first excited molecule that

canprtiiptein potoxidationandinitiatelipidperoxidation. Potentialsensitizers include

exogeous substances (drugs, foods or dyes) and endogenous molecules. Rhodopsin,cone

photopigm s r(.e. iodopsin), flavin, melankn, retinal, retinal and the intermediates that

occur in the light induced isomerization of rhodopsin (i.e. metahodopsi I and I) a

potential sensitizers. Rhodopsin and cone photopigments are probable mediators of

photodyramic reactions (directly orindirectly via their int ediates) asthe action spectrum

of light damage in the rat parallels the absorption spectrum of hdopsin and in experiments

utilizing narrow band filters selective cone damage has been produced. S"A

The above theories and experiments indicate that the retina is susceptible to

photooxidative and lipid peroxidation mediated damage. Anaxidation in the RPE cell and

phagocytosis of shed disc outer segments results in accrlinn fthe end prodctsoflipid

perodation, i.e., cross linked proteins, DNA and lipids in the RPE. Te RPE enzymes,

presented with an altered substrate, cannot always completely digest these mantrals and as

a consequence residual bodies and lipofuscin accumulate in the RPE cells." Lipid

peroxidation probably occurs throughout an individuals lifespan. The age-related

accmulation of lipofuscin in cells such as the RPE is ubiquitous across species and is

thought to be important in the pathogenesis of ARMD.

Other mechanism. Several investigators have shown tha the threshold for light

damage decreases exponentially as a function of decreasing wavelength (L.e. increasing

energy) indicating other mechanisms besides photopigment meniated reactkis contribute

to photic retinopathy. Y41^

Inats exposed to continuous light,changes includeacivationaflysosoaal enzymes

and elevation of cathepsin D (total and free) in the neuPa tinas, RPE and caroid when

compared to comtr animals Lysosomal enzymes may contribute to photic tinopathy

if excessive activation occurs under excessive light stress.

A recent study has shownthat inactivation of cyclic GMP(cGMP)phospb iesterase

(PDE) accmaieslight damage.' Cyclic GMP is a central molecule pin ho ansduction;

the convesia of light energy to a nerve impulse. After light strikes a photarcptor outer

segment the phopigment 0.e. rhodopsin) is isoenrine which induces a series ofreactions

that activate cGMP PDE. A subsequent decrease in cGMP results in closure of sodium

channels and hyperpolarization of the photoreceptr azd initiates an action potential. In the

Irish Sette and CH mouse an increase in cGMP smcdary to a decrease in the PDE activity

is thought to contribute to retinal degeneration. 6 The inactivation noted in light stressed

retinas would result in an accumulation of cGMP tha could contribute to the light induced

retinal damage.

Other studies have shown inactivation of cytocbnme oxidase occurs during exposure

to ultraviolet light. 'Tis mitochondrial enzyme is a component of the electron transport

chain that is responsible for energy production for the phoreeptor. Photoreceptors have

high energy demand to fuel the continual synthesis ofdozns of outer segment discs that are

dsed daily. Inactivation of cytochrome oxidase durin excessive light exposure could

contribute to retinal damage.

In conclusion, multiple factors and mecWarnism contribute to light mediated retinal

damage. High oxygen tension, focusing of light on the retina, high PUA content of outer

segments and high metabolic rate of the photoreceptcr make the retina very susceptible to

damage mediatd by photooxidaton and lipid pea oidatin. I addition, inactivation of

inmartant photareceptor enzymes may contribute to Egh induced retinal damage.

Protective Adatations in the Ey

It is evident the eye is susceptible to damage frn excessive light exposure mediated

by photoxidative and lipid peroxidative damage. Because the eye has evolved under these

streses a very effective set of physiologic and molecular adaptations have developed that

pioaect the eye from light induced damage.

As mentioned previously the photoreceptos are extremely metabolically active. The

ellipsoid region of the photoreceptor (PR) is packed with mitochondria to provide energy

(ATP) for the cel. Te primary activity fueled by the mitochandria is the constant renewal

ofoutersegments. Both cones and rods continually synthesieouter segment discs, the discs

move distally toward the RPE, and are eventually shed and phagocytized by the RPE.-77

Each etinal rod produces and sheds about 80-90 discs per day and the entire compliment of

discs ae replaced every 9 to 13 days." The outer segments are very susceptible to

photoaxidative damage because of their high concentration of PUFA's The constant

shedding of older, more err-ridden parts of the PR outer segment allow fa partial erasure

of light induced molecular ens. Biological renewal, however, is not perfect, and the

altered outer segments (Le. with lipid peroxides) represent potential indigestible substrates

for the RPE phagolysosomal system. Incomplete digestion of the shed outer segments

results in accumulation of lipofuscin in the RPE and is pat of normal senescence.4'".

Excessive accumulation of lipofuscin can lead to pathology by further inhibition of RPE

Melanin granles in the iris, RPE and choraid represent another protective

merhmism. One function of melanin is thought to be absorption of photons not absorbed

by the outer segments and conversion of this scattered light o heat. Recent investigations

suggest melanin is a supprssor of photosensitized molecules including singlet oxygen, as

wel as a free radical scavenger.' In contrast, there is evidence that at high rates of

exposure to short wavelength light melanin may become cytotoxic by production of free

radicals.'" This free radical generation is reversible and decay of the radicals occurs

in the dark" Melanin granules are in the pical region and microvilli of the RPE cells and

the close proximity of these organelles to the rods and cones suggests they play an important

role inthe visual process. Some studies have shown pigmented animals are less susceptible

to lightdamage The mechanism of this prection is relatedto irispigmentationdecreasing

incident light on the retina. When retinal exposure intensity is identical, no difference in

susceptibility of pigmented vs. albino rats was observed.3

In primates xanthophyll and other macular pigments are thought to be protective by

absorbing the shorter and more damaging wavelengths of light Some macular pigments

may actually act as free radical scavengers.

Many studies have shown that elevated body temperature lowers the threshold for

retinal damage by light. "0 In addition to thermal mediated damage, chemical reactions

that ae facilitated by increased temperature may contribute to increased damage. As

discussed above, the melanin granules in the RPB and choroid absorb excess photons and

convert the energy to heat. The choroidal circulation is wel deigned to dissipate the heat

generated by light absorption The rich vascular plexus, the chriocapllaris, is separated

from the RPE only by Bruch's membrane. Also, the larger chridal vessels are arranged

to resulting a untcune flow pattern that aids in heat dissipating4 Recent studies have

shown that cboroidal blood flow increases with increased ilumiation." The investigators

believe the choroidal circulation, via regulation of blood flow, maintains normal body

temperature in the RPE and outer retinal layers during light exposure and prevents excessive

increases in temperature that pedispose the retina to photic injury.

Several other inacellular mechanisms that protect against free radical and

photochemical damage are present throughout the eye. Antioxidant nutrients such as alpha

tocopherol (vitamin E) and ascorbic acid (vitamin C) are in high anetratios in the retina

and other ocular structures

Many studies have shown vitamin C protects against light induced retinal

damage." In rats, guinea pigs and monkeys, intense light exposure leads to a loss of

reduced ascorbate indicating it is oxidized during light exposure." Admiistration of

ascorbate before (but not after) light exposure led to a preservatiaof docsahexaenic acid

in rats having suggested this vitamin had inhibited oxidation of thismajor PUFA inrod outer

segment membranes

Vitamin B protects biological membranes from autoidation and is present in high

concentrations in the ltina and RplE.49 In experiments with light-induced retinal

damage, vitamin B spplementation decreased retinal damage and decreased accumulation

of hydroperoxides (end products of lipid peroxidation of memnanes) when compared to


The RPE phagocytzes rod outer segments and with its active lysosomal system

digeststhese membnanousPganelles. Incomplete digestion duetoabnomal substrates C(e.,

lipid peroxides) or decreased enzyme function results in fcmnation of residual bodies or

lipofuscinY A um latin of lipofuscin occurs normally with increasing age.

Excessive accumaation of lpofascin in the RPE may actually inhibit RPE function and

result in retinal degeneration. In rats and monkeys fed vitamin E deficient diets, massive

amounts of lipofuscin had accumulated and severe retinal degeneration had occurred, both

having been attributed to lipid peroxidationa.'"

Taurine, an amino acid, is another potential antioxidant nutrient in the eye. Taurine

deficiency causes retinal degeneration in cats." The mechanism by which taurine maintains

photoreceptr structure and function is thought to be a membrane stabilizer. Evidence for

this was found in studies that showed taurine prevented peroxidation-induced damage in

isolated frog rod outer segments via membrane stabilizatioaL" In studies with rats, the

retinas of taurine-depleted animals were not more susceptible to damaging light intensities

than controls." Unlike primates, rats or pigs, taurine is an essential amino acid in cats and

the dietary-induced retinopathy seen in cats is probably unique to this species.

Zinc is another nutrient that indirectly and directly provides protection against light

induced oxidative damage to the retina as metalloenzymes (Le., superoxide dismtase and

caalase), metalloprteins .e., metallothinein) and/or by stabilizing membranes.

Specific antioxidantenzymes, superoxidedismutase (SOD),catalase andglutathione

peroxidase, catalyze the the enzymatic breakdown and detoxification of superoxide and

hydrogen pergide. SOD catalyzes the dismutation of te superoxide radical to hydrogen

peroxide and oxygen. SOD is 200 times more c ncentrated in bovine and frog rod outer

segments than the rest of the retina.'" Thus, this potent anti idant enzyme is present in

the portion of the retina most susceptible to oxidative damage.

Catalas converts hydrogen peroxide to water and is found in microperxisomes in

the RP" il Microperoxisomes are intracellular organeles found in cells involved with

lipid metabolism and are thought to provide protection from hydrogen peroxide damage."

Microperoxisomes actually decrease in number in the RPE at the comesponding burst of

outer segment disc shedding that occurs with the onset of light." The investigators propose

micoperoxisomes fuse with RPE phagosomes and facilitate digestio. Microperoxismes

are probably an important component of the steady state equilibrium of disc shedding and

RPE phagocytosis/digestion because they prevent accumulation of undigestible and

damaging lipoperoxides (which result in lipofuscin formation) in the retina."

Glutathione peroxidase is another antioxidant enzyme present in the retina.' In the

presence of reduced ghOathione it can detoxify lipohydroperoxides, hydrogen peroxides,

nucleic acid peroxides and other radicals. It also can recycle other anioxidans, i.e., by

reducing oxidized ascorbic acid produced during high light expose.

Zinc and the Ef

In 1869 zinc was found to be essential for growth in fungi, in 1926 for higher

plantstm and in 1934 essential for rats.'1 In 1939 carbonic anhydrase was proven to be a

zinc metalloenzyme.1T In 1955 parakeratosis in swine was found to be secondary to zinc

deficiency.1m Several other studies have shown the importance of zinc for the growth and

maintenance of animals."'" In 1963,Prasad and others documented the importanceofzinc

for human health? Clinical manifestations of moderate to severe zinc deficiency include

anorexia, growth failure, alopecia, thickening and hyperkeratinizat of the epidmis

gonadal hypoplasia, diarrhea, mental disturbance, altered cell mediated immunity leading

to recurrent infections, delayed wound healing, abnormal dark adaptation, hypoguesia, and

if untreated, death."n" These studies and many others continue to document the

importance of zinc for normal health in animals and humans.

The importance of zinc to the eye was suspected, when it was determined the content

of zinc in the eye is higher than that of most mammalian tissues. The highest concentration

is in the chlaid and retina followed by the ciliary body, iris, optic nerve, sclera, cornea and

lens."'4 In addition, investigators have shown a rapid uptake of zinc by melanin

containing ocular tissues in several species.1"'~' The copper content of the eye is also

proportionately high compared to other tissues."'l' In 1939, Keilin and Mann showed zinc

was an essential component of carbonic anhydrase.'" This discovery provided unequivocal

evidence f zinc's role in ocular metabolism and paved the way for further investigation into

the biological role of zinc in the eye.

Thmbas diverse functions in metalloenzymes, immunecompetence, visual acuity,

wound heatlinr stabilization of membranes, gene expression, DNA structure and in protein,

carbohydrate and lipid metabolism. ,1~"-,"17-U

TZi can have a structural, catalytic or regulatory role in metalloenzymes. At least

51 zincetalloenzymes have been identified and examples in all six enzyme classes can be

found.' I addition to carbonic anhydrase several other zinc metallonzymes function in

ocular tissues: retinal dehydrogenase, superoxide dismntase, leucine aminopeptidase,

collagease and alpha mannosidase have been identified in ocular tissues.1"'J1i"2

Carbonic anhydrase is important for aqueous humor production and is also present in the

RPE". Retinal dehydrogenase converts retinol (vitamin A) to retinal which is a component

of the visual pigment hodopsin. Superoxide dismutase is an important antioxidant enzyme

found in high concentrations in the retina. Leucin aminopeptidase is utilized for the

degradation of lens proteins. Collagenase is important in coeal wound healing. Alpha

nannosidase is alysosomal enzyme in the RPE participating in the digestion of shed rod and

cone outer segments.a Frxn these examples it is obvious that alterations in zinc status

could significantly influence metabolism in the eye and vision.

In all tissues, zinc in metalloenzymes that participate in nucleic acid and protein

synthesis, is important in cell differentiation and proliferation. For example, in human RPE

cell cultures, decreased zinc resulted in decreased cell proliferation (63%) and protein

synthesis (50%)." In low zinc medium other parameters were also reduced in the RPE cells:

zinc (40%), catalase (68%), alaline phosphatase (60%), alpha manosidase (68%), and

metallothionein (82%). After zinc repletion catalase and alpha mannosidase activities rose

by 86% and 106%, respectively, suggesting inactive enzymes were present but were not

functioning due to lack of zinc." Another study on human donor eyes determined the

amount of zinc in the peripheral and macula retina. Cytosolic zinc was higher in the

peripheral retina vs the macula.m A positive correlation between macular and peripheal

zinc, and between age and disease was found. A negative correlation between macala zinc

and age, and macular zinc and disease was observed. Fr example, total zinc was 38% less

in eyes of people over 70 compared to those under 70 years of age. It is possible that age

related decline in enzymes such as catalase, important proteins like metallothionein and

other human RPE functions can be influenced by the amount of bioavailable zinc at the

cellular level" This reduction in the function of the RPE with age is an important

component in the pathogenesis of ARMD.

Zinc may also provide protection against light-induced free radical damage via its

role in promoting membrane stability. It is increasingly evident that zinc ions are an integral

part of membranes."na A substantial amount of zinc is located in the plasma membrane

bound to lipopoteins and other membane-associated proteins, Le., enzymes and structural

proteins.m Elevated extracellular zinc leads to increased membrane zinc with a stabilization

effect, and, conversely, decreased extracellular zinc leads to loss of plasma membrane-

associatedzincand destabilizio1 Chavpil and co-wokers believe membrane pathology

is a primary component of zinc deficiency pathology." One of the mechanisms by which

zinc promotes membrane stability is via prevention of lipid peroxidation

Several experiments have shown zinc deficiency increases lipid peroxidation in

membranes and that zinc protects against peroxidative damage induced by heavy metals,

carbon tetrachloride and high oxygen tension. 1190Slm In zinc deficient rats, liver

microsomes had increased lipid peroxidation vitro and in vo. Liver mitochondria

and microsomes from rats fed high levels of dietary zinc showed resistance to in vitro

induced lipid peroxidation.1 Dietary induced zinc deficiency in rats increases endogenous

free radical production in lung icosome1 Zinc aspartate protected mice against lethal

effects of innizing radiation.m The radioprotection was attributed to formation of zinc

complexes with low or high molecular weight thiols (sulfur ontining agents) within cels.

Zincis thought to prevent free radical formation and oxidation of protein sulfhydryl grouOps

by completing with sulfur molecules and protecting against radical induction.'","2

The effects of altered zinc metabolism on the eye and vision are rapidly becoming

an area of interest for many researchers. Specifically, an association between zinc


deficiency and the development of age-related macular degeneration has been proposed.'

It is currenly believed suboptimal zinc nutrition poses a problem for a substantial section

of the United States population." Children, the aged, and pregnant individuals are

especially at risk.'Oua3""'m Several studies as well as clinical diseases hae shown zinc

deficiency results in significant structual and physiological alterations within the eye.

In rtslow maternal plasma and embryonic zinc concentration during oganogenesis

caused abenant morphogenesis of the optic vesicle and incomplete close of the fetal

fissure leading to anophthalmia or colobomatous microphthalmia.'" Mirphthalmia has

also been seen in chicks from zinc deficient hens."7 Because zinc is impitant in protein

metabolism and has a beneficial effect on wound healing in vascular tissues,'"-"Nm4 it is

thought to promote coaneal healing. Collagenases which contribute to devastating melting

cmIeal ulcke contain zinc. Treatment of melting ulcers with zinc chelatcra (e. EDTA)

pro tes healing. These contrasts reflect the complexity of zinc metablism within the

eye. Zinc deficiency induces cataracts in fish3" and some believe may play a role in the

etiology of cataracs in humans; however, evidence to substantiate this is lacking." "4n

Several human diseases, including acrodermatitis eteropthica, alcoholism,

mnalawb ption sicle cel anemia and chronic renal, pancreatic or debilitating diseases, are

associated with zinc deficiency. A common manifestation of zinc deficiency is abnormal

dark daptation (poor night vision)."U11U.-3 The rod photrecptoms with the

photopigmntrodopsin areresponsible for night vision. Two mechanisms contribute to the

alhnrmal dark adaptation observed in individuals that are deficient in zinc.

Retinal alcohol dehydrogenase converts retino (vitamin A) to retinal which

condenses with psin to form rhodopsin. One mechanism by which zinc deficiency results

in abnormal dak adaptation or night vision is felt to be from decreased activity of retinal

alcohol dehydrgenase. In zinc deficient rats retinal and liver alcohol dehydrogenase

activity were siificantly lowered and conversion of retnol to retinal reduced. The retina

was more sensitive to the lack of dietary zinc than the liver.' Dietary orpathological zinc

deficiency causa functional vitamin A deficiency at the level of the retina because retinol

is not converted to retinal. Vitamin A deficiency is a well documented cause of night

blindness. In alholic cirrhotics, zinc and vitamin A deficiency are common. Morrison

et aL found that in some patients that were initially treated with vitamin A, only after

addition of zincdid dark adaptation return to normal They suggest the improvement seen

in dark adaptatir may be due to enhanced activity of previously depressed retinal alcohol


ZiTh deicy may also influence vitamin A metabolismat the level of the liver by

decreasing synesis and/or release of retinol-binding protein (RBP) from the liver."Y'-m-

Vitamin A isbomd to RBP and pralbumin for transport to the retina and other tissues.

Retinal bindingptein levels in liver and plasma were found to be 50% and 25% of onmal

in zinc deficit~ rats.m In zinc deficiency, albmnin levels are also low, which may

contribute to inequxate transfer of vitamin A to the retina." 1

Arod atitis nteopathica is a rare hereditary abnormality in zinc absorption

characterized by dnatiis, chronic diarrhea, central nervous system abnormalities and

impaired ummo function with frequent infections. Several ocular manifestations have

been reported. Ocular signs including conjunctivitis, canthal dermatiti linear subepithelial

coneal opacities, cataacts, optic atrophy, ciliary body atrophy, retinal degeneration and

RPE depigmentation have been reported.'" Gaze aversion and photophobia have also been

reportedand are suggestive ofimpaired cone vision." Similarly,inalcoholic cihotics with

zinc deficiency, a few reports of abnormal color vision and macular vision have been made,

indicating cane vision may be affected by zinc deficiency."

Silverstone and coworkers believe alterations in zinc and copper metabolism are

involved in pigmentary retinopathies including retinitis pigmentosa and ARMD.1' In a

group of retinitis pigntosa patients they documented low plasma zinc while in ARMD

patients they found elevated plasma zinc and copper." The elevation of plasma zinc and

copper was postulated tobe due to a constant release of these elements from the pigment in

degenerating RPE and chroid occning in ARMD.

In spite of Silversone's findings, zinc deficiency has been proposed to be one of the

etiologic facts inARMD. u4'" InastudybyNewsoimeandcowods ofagroupof 151

subjects, the zinc treated group had significantly less visual loss than the placebo group.'

Using the technique of elemental analysis, Ulshafer found that zinc levels in degenerating

RPE cells adjacent to drusen deposits (both of which occur in ARMD) were decreased.7

I should be mentionedhat Silverstone's finding of increased plasma zinc does not rule out

a subclinical zinc deficiency in patients as assessment of zinc status is difficult, and plasma

values are a poor indicator of status.

Another possible way zinc deficiency may contibute to ARMD is via alpha

mannosidase. a recent study an age-dependent decrease in alpha mannosidase activity

was found in RPE cell cultures and addition of zinc to the media increased activity 3-fold.

The same investigators found a decrease in alpha mannoidase activity in the serum of

ARMD patients. Addition of zinc augmented serum enzyme activity 2-fold.T Since alpha

mannosidase is required for the degradation of phagocytized outer segment membranes in

the phagolysosomal system of the RPE, decrease in its activity could contribute to the

accumulation of undigested outer segments as drusen and lipofusin, both of which are

associated with ARMD. Zinc status of donor patients was not assessed in this study.

Histological and ulrastructural studies of the effects of zinc deficiency on the retina

are limited. The eyes of a male child with acrodennatitis eMeropathica were obtained for

histopathological exam.a Extensive degeneration of the RPE through t the posterior pole

was observed as well as atrophy of the inner and outer retinal layers. Interestingly, loss of

rod and cone outer segments was most severe anterior to the equate. Extensive loss of the

ganglion cell and nerve fiber layers was present. Complete atrophy of the optic nerve was

also observed. The optic nerve atrophy was attributed to the severe nervous system

involvement, several historical episodes of cardiopulmonary anest and previous treatment

with diodoquin, all of which could contribute to severe optic nerve atrophy. The above

history coupled withrepeated infections in the child may also have contributed to the retinal

degeneration so it is difficult to attribute the retinal changes to zinc deficiency alone.

Leure-duPreet al have shown several alterations in the retina of rats on zinc

deficient diets or treated with zinc chelatorsms2'" They observed accumulation of

osmiophilic inclusic bodies within the RPE as well as vesiculation and degmmeratin of the

photoreceptor outer segments. Te photoreceptor degeneration was usually in the areas

where many inclusions had accumulated in RPE cells. The inclusions did not appear to be

the products of phagocytosis, nor did they resemble lipofuscin granules or

niroperoisoes. The inclusions were lipid in nate but otherwise the chemical

composition has not been investigated. Inclusions were usually associated with

mitochondria and smooth endoplasmic reticulum. Modified light regimes did not affect or

initiate fomnation of the inclusions.

Samuelson and coworkers have investigated how zinc deficiency affects the retina

inswine."-"' Because extensive literature is available on zinc deficiency in pigs and the

zinc deficiency syndrome can be consistently reproduce, this species represents a good

model for studying the effects of zinc deficiency in the eye. "'Sm In boars maintained on

low zinc diets for 4 months, numerous cone nuclei and their inner segments were displaced

towards the RPE. Occasional cones were observed in various stages of degenation. The

number of cone inner segments appeared distinctly fewer in those animals that had been an

the low zinc diet fri 4 12 months. In boas on the diet for 8 months disrienation of the

outer segments were noted. Boars fed a control diet and peniclmine had similar retinal

changes. Current studies of sows manaintai n chronic low and moderately low zinc diets

have shown similar disruption of outer nuclear layers and the development of large

melanomes in the choroid. Elemental analysis of RPE and melanin granules in these

animals showed interesting alterations in zinc, copper and calcium including a decrease in

zinc in RPE melanin granules vs controlsf.4'


Metallothioncin (MT) is a low molecular weight cytosolic protein that selectively

binds metal ions such as copper and zinc. This protein also has affinity for potentially toxic

elements such as cadmium and mercury.Y'" Some of the metalothionein's biological

functions include protection against heavy metal toxicosis, and regulation of copperand zinc

metabolism.''" Metallahiaoein can donate its zinc ions to activate zinc metalloenzymes.

It is involved in cell polifration and differentiatim most likely as a provider of zinc ions,

since some enzymes responsible for synthesis of nucleic acids and proteins require zinc.

Metallothionein may also play a role in maintenance of protective barriers since it has been

found in the placenta, RPE (the blood- retinal barrier) and blood-brain baniers.Y'

Metallothionein can provide cytoprotection in several ways. It has been shown to

be an efficient scavenger of free hydroxyl ions in vitro.'1 It may protect against potential

damage caused by heavy metals, as wel as other compounds involved in inflammation or

physiologic stress. For example, deamethasone, cAMP and intedeulkn-I induce synthesis

ofMT andmany of these necales are importantprticipants inthe inflammatory responses

or stress.*'S"'*61 Induced MTmay bind toxic metals or donate zinc ions for stabilization of

membranes or important metalloenzymes like the antioxidant enzyme SOD. Ultraviolet

light has also been reported to induce synthesis of MT.' The increase in MT protein results

from increased transcription of MT genes.

A metallothionelike protein has recently been identified in the bovine retina.1

Subsequently, MT has been found in human and rat RPE, and in the inner retinal layers of

the rat" MT was also localind by iummnohistochemistry in the comeal endothelium and

epithelium, and in the lens epithelium." In adult human RPE cell cultures mean MT levels

were 18 +/- 2.2 pg MT/mg protein" in one study and 8 pg MT/ mg protein in another."

Studies on metallothimein inhuman RPE cell culture have revealed that MT (similar tohe

distribution of zinc) is present in higher levels in the macula (20 ig MT/ mg protein) than

in the peripheral retina (11 (g MT / mg protein). They also found that MT levels decrease

with age, with a much greater decrease occurring in the macula vs. the periphery (Le. from

28 pg MT mg protein in the late 20's to 4 ig MT / mg protein in the late 8W(s. The

significant decrease in MT levels in the RPE that occurs with age may play a role in the

pathogenesis of age-elated macular degeneration and warrants further study.

Metallothionein synthesis is induced in RPE cultures by incubation with Znu,

cadmium, dexamethasone, phagocytosis of latex beads or rod outer segments 1,25

dihydroxy vitamin D, and paraquat (induces oxidative stress).*" The induction of MT

during oidative stress (incubation with phorbol myristate acetate, lipopolyscchmar

ferrous sulfate, hydrogen peroxide, paraquat, and phagocytosis of rod outer segments) has

been correlated with the activation of two nuclear regulatory proteins, NF-kB and Ap-l,

again indicating the increase in MT is due to increased gene transcripion" Free radical

scavengers lke DMSOblocked the induction ofMT that occurred withadditin hydrogen

peroide, again suggesting the oxidative stress caused the increase in MT synthesis.

Another experiment showed induction of MT after addition of ferric iron to hmnan

RPE cell cultures." The 2-fold increase in MT was blocked by the free radical scavenger,

N-acetyl cysteine (AC) or cycloheximide. Ferric iron activated the DNA binding of NF-

kB. Again NAC inhibited this activation. Phagocytosis of bovine rod outer segments

induced MT and activated NF-kB, which were both blocked by the iron chelaor

desfeaoxamine during phagocytosis. The authors postulate that free iron may play a role

in generatim of an RPE stress response that can occur as a result of phagocytosis of shed

photoreceptor outer segments." This and other studies substantiate other data that suggests

MT can be an acute phase reactant protein, which presumably plays a role in protecting the


Excessivelightexposure, whichinducesoxidativestrssmightalsoinduce MT. The

proposed study in which MT will be identified in the porcin retina utilizing the cadmium-

hemoglobin binding assay and immunohistochemical staining will attempt to document

whether changes in levels or distribution of MT result from photic injury andlor

manipulation of dietary zinc levels. Immunohistochemical staining has been utlized to

localize MT in the kidney and liver of rats, and in the brain of humIns.." Most of these

studies utilized innmmnawddase staining to localize MT.

Colloidal gold immn ytochemistry has been found to yield superior results for

localization at the electronmicroscopic level.'"m With the silver intensifiction technique

applied to colloidal gold stained tissues, specimens can also be examined at the light

microscopic level m imnn~ ytochemistry of the retina at light microscopic level

allows localization to the rtinal layers but is not as useful for lolization at the cellular

level. Use of the colloidal gold technique will allow more detailed localization of MT so

will be utilized in this study. Specimens will be examined at the light and electron

microscopic level to try to determine where MT is distributed in retina and specicay in

which cells. This will add to our knowledge of the biological function of zinc and

metallothionein in ocular tissues.

Light Zinc. and Age-related Macular Degeneation

Age-related macular degeneration (ARMD) in humans is the major cause of

blindness in people over fifty.l 6 This disease involves degeneration of the macula, a cone

dense area of the human retina. Major features of ARMD include its central location, its

association with age, disruption of retinal pigment epithelium (RPE) pigmentation,

formation of drusen, destruction of Bruch's membrane, and bsal laminar deposits of

amorphous or granular material between the RPE cells and their basal lamina. All the

major signs of ARMD increase in prominence with advancing age, but only in some

individuals do they progress to the stage of pathology. The RPE plays a prominent role in

this macular disease. Throughout life photoreceptor discs are shed in association with the

visual process. The RPE phagocytizes and degrades these membranes in conjunction with

lysosomes. It is proposed that abnormal molecules gradually accumulate within the RPE due

to imperfections in the cell's digestive mechanisms.5 The residues of incomplete molecular

degradation accumulate in the cell and increasingly interfere with normal metabolism which

presumably leads to abnormal excretion of debris and drusen under the RPE cell. In support

of this theory, a recent study has shown a significant age-dependent decrease in activity of

alpha mannosidase in cultured RPE cells.' Furthermore, the researchers demonstrated a

2-fold increase in the enzyme's activity in aged RPE cell cultures with addition of zinc.

Since alpha mannosidase is a lysosomal enzyme that normally functions in RPE lysosomal

digestion of photoreceptor disc segments, a decrease in this enzyme's activity could lead to

accumulation of undigested rod outer segments and drusen, both of which are associated

with ARMD. These theories suggest that ARMD is an advanced stage of a deterioative

process that takes place in all eyes, and the progression of any of several factors may lead

to pathology.

The etiology of ARMD is multifactoria Some of the proposed factors that

contribute to the development of pathology include genetics, light iris pigmentation,

prolonged light exposure, and most recently, nutritional deficiencies."tA'f"m A study on

the effect of oral zinc administration and the visual acuity outcome in 151 subjects with

drusen or macular degeneration showed that the treated group had significantly less visual

loss than the placebo group after a follow-up of 12-24 months.' Using the pig as a model,

this study will investigate the effects of zinc deficiency and supplementation photc inju

utilizing clinical, hiopathologica immunistochemical and ultstructural studies. The

changes in elemental content of different components of the retina and choroid will also be

observed utilizing energy dispersive X-ray microanalysis. The primary goals win be to

determine if zinc supplementation is protective against photic injury and if zinc deficient

states increase the susceptibility of the retina and RPB to photic injury.




Thirty-one six- to eight-week old lightly pigmented pigs were obtained from the

University of Florida Swine Unit Animals were obtained a minimum of 4 days prior to

starting the experimental diet to allow acclimation to their new environment and recovery

from the stress of transport. Animals from Part 2 of the study were treated with oral

trimethoprim-sulfa suspension 30 mg/ kg P.O. q 24 h for the first 7 days of housing. When

not in the eerimental photic injury unit, animals were housed in the University of Florida

College of Veterinary Medicine Metabolic building in air conditioned pes with free access

to water. The front of the pens had galvanized chain link fencing with a door.

Photic Inuy Unit and LiUhtin Conditions

The experimental design of the lighting conditions and phatic injury unit was in part

from the advice from Dr. William Dawson, Department of Opthahmology, University of

Florida College of Medicine, who has graciously agreed to advise us in this matter, and from

a variety of past studies involved in photic injury. The photic injury unit was crcula

with white formica on the internal walls. It was necessary to elevate the unit 4 inches off the

floor to facilitate cleaning. A reflective white wall board was placed on the floor to help

maximize light intensity. The unit was elevated five inches off the floor by legs to facilitate

cleaning and air circulation. Incandescent tungsten-halogen light sources, 4 outdoor flood

lights (Philips Lighting Co., Somereset, NJ), were suspended over the unit These lights

generate broad spectrum visible light possessing wavelengths from 400-900 nm. The

relative power (radiant energy in pwl 10 mnm lumens) along the wavelength spectrum

increases steadily being less than 100 for shorter wavelengths, e., less than 500am,

modete (100-200) for intermediate wavelengths. e500-700nm, and greatest (200-300)

fir nger wavelengths of 700-900 nm. The incandesct bulbs also have a significant

amontof radiant energy inthe infrared (non-visible) spectum with wavelengths up to 1500

nm wavelength.' The photic injury unit could house only two animals at one time under the

guideline of the University of lorida animal welfare committee necessitating staggering

of times of exposure to the light in the unit. Therefore at the time of exposure pigs were 8 -

11 weeks old.

During the initial pilot study the temperature in the unit averaged 90-93F and

animals exhibited a tendency towards hyperthennia with rectal temperatures of 105-106F

compared to the normal range of 01.6-103.6'F.r Rectal temperatures the animals prior

to placement in the unit wee around 103P. Several studies have demonstrated hat

elevated body temperature (hyperthema) accentuates photic injury." To minimize the

number of pemental variales that might influence photic injury in subsequent trials, a

100% acrylic infrared (heat) absorbing shield (K-S-H Inc., Olive Branch, MI) and fans were

utilized to maintain the temperature in the unit at 80T. This resulted in lower rectal

temperatures of 102-103'F.

l ncadescent Lamps, General Electric Co. 1977

The plastic shield, floor and walls of the unit were cleaned once or twice daily to

remove accumulated dust in order to maintain maximum illumination within the unit. Light

intensity was measured with a J16 digital photometer (Tektronix, Beaverton, OR). The

photometer was placed in the center of the circular unit with the sensor directed at the wall

of the unit A measurement was obtained and then the photometer rotated 90" for the next

measurement This was repeated until twenty readings were obtained. Mean illuminance

of experimental lighting (with the plastic shield in place) was 630.7 ft-candles +/- 81.6.

Control lighting had a mean luminance of 194.7 ft-candles +-- 77.2 (n 20). Control

fluorescent lighting illuminated cages housing unexposed animals a generated a gradient of

illumination from front (261 ft-candles) to back (115 ft-candles) of the cages. For

comparison normal indoor iluminatio office lighting, ranges from 50-100ft.candles.

Outdoor illumination on a clear sunny day is 10,000 f-candles, and in shade or on an

overcast day may be around 1,000 ft-candles or less.m


Diet analysis was performed by the University of Florida Animal Nutrition

laboratory under Dr. Joel Brndmuhl's direction. Prir to the acquisition a animals were

fed a con-soybean meal based grower diet (the control diet in part 1) at the swine unit One

week prior to the time they were placed into the photic injury unit, the anima were placed

on their designated perimental diet The experimental diets were fed until the day before

animal sacrifice (animals were fsted for 18-24 hours before sacrifice). The pigs were fed

either a control diet (ZnN, 100 ppm Zn), zinc deficient diet (Zn-, 11 ppm Zn) cr control diet

that was supplemented with zinc sulfate powder to contain approximately 300 ppm Zn

(Zn+). The control diet with 100 ppm zinc met all the nutritional requirements for pigs of

this age.' In the interittent light stress group and their controls, the zinc deficient diet was

a casein-sucrose based feed supplemented with mineral/ vitamin mix lacking zinc to end up

with the low zinc level of l ppmn. The other diets were corn-soybean based (Table 1). In

the studiesutilizing continuouslighting,the experimental diets were all casin-sose based

and had similar zinc levels; Zn- 11 ppm, ZnN 93 ppm and Zn+ 240 ppm (Table 2).

Unfortunately the results of diet analysis were not available until after the experiment and

it is apparent that the supplanted diet in part 2 of the experiment did not contain 300 ppm

but only had 236 ppm of zinc. The average incoming weight of the animals was 20 kg.

Animals were pair fed 1.5 kg feed per day which was within the range of the average

expected feed intake for pigs f this age and size.m Animals kept alive fo three weeks after

light exposure had their feed take level increased as body weight increased to2 kg of feed

per day. Animals were fed from plastic or stainless steel containers to avoid extra zinc

ingesion. Animals being fed the zinc deficient diet were provided with distilled deionized

water free choice. As mentioned previously, when not in the photic injury unit animals

were in cement pens with galvanized fencing at the fron. Galvanized metal contains zinc

so pens that housed animals a zinc deficient diets had peg board stopped to the chain link

to prevent the animals from chewing n the metal and obtaining exogenous zinc

Table 1: Diet analysis from Part 1 of the experiment The control and zinc supplemented
diets were crn and soybean meal based. The zinc deficit diet was a casein and sucrose
based die. (NDF non digestible fiber; ppm parts per million mgkg; *determined on
a dry matter basis, all others are on an as-fed basis)

Zn Deficient Control (Zn normal) Zinc supplemented
Casein-sucrose Com-soybean Corn-soybean
Dry Matter 95.34 89.74 89.24
Crude Protein 17.90 18.06 1802
Crude Fat % 4.99 2.00 1.99
Ash % 3.27 5.20 5.26
NDF % 3.06 11.40 2.51
Zinc ppm 11.17 100.39 296.86
Copper ppm* 22.46 16.68 12.22
Calcium ppm 7608 8359 5741

Table 2: Diet analysis for Part 2 of experiment All diets were sucrose casein based.

Zn Deficient Control (Zn nonal) Zinc supplemented
Dry Matter S 91.12 90.82 91.06
Crude Protein % 18.46 18.44 17.85
Crude Fat % 5.78 6.98 5.85
Ash % 4.72 4.67 4.79
NDF % 4.02 3.9 4.27
Zinc ppm 10.85 92.03 236.87
Copper ppm* 18.42 20.91 17.97
Calcium ppm* 7542 7573 7682

Photic Inur

Pilot study. Two young pigs were obtained from the University of Florida swine unit.

They were fed the control diet and were placed in the unit for either 36 or 72 hours of

intermittent light. These animals were sacrificed 72 hours after exposure and tissues were

processed for light and electron microscopy (see details in next section). Rectal

temperatures were taken every morning. Results ofmorphologic analysis of these tissues

were used to guide subsequent experiments.

Part 1: Intermittent light stress. Eighteen animals were utilized for experiments with

inermittent light stress. From each diet group (Zn-, Zn+ and ZnN) 2 pigs were exposed to

one of three lighting conditions: control lighting, high intensity-short-exposure (18 hr light

6 hr dark for two consecutive days) and high intensity-chronic exposure (18 hr light and 6

hr dark for four consecutive days). Therefore animals received intermittnt high intensity

light for 36 or 72 hours as indicated in Table 3. Prior to and after photic injury the animals

were exposed to control light intensity for 18 hours each day until sacrifice which was

performed 72 hons and 3 weeks post-exposure.

Part 2: Continuous light stress. An additionalgroupofthirteenanimals was obtained

after completion of the first study. Four animals were in each diet group (Zn-, Zn+ and

ZnN). Experimetal diets were started one week prior tolight exposure Due to constraints

in expense of diet and board, only one animal, fed the control diet, was not exposed to light

stress. All three diets were casein-sucrose based. Animals on the zinc deficient diet were

kept in stainless steel cages and provided with &deknized water delivered in plastic tubing

Table 3. Summary of experimental design for Part 1 utilizing intenmittent ligt exposure.
Six animals were in each diet group. Two animals of each group were exposed to the three
lighting regimes. One animal in each diet group was sacrificed 72 hours or three weeks after
light exposure. LI: control lighting, L2: 36h intermittent light stress, L3: 72h intermittent
light stress.

DIET TOTAL ANIMALS Kill 72 hours Kill 3 weeks
Control diet (lOOppm Za) 6 (2-LI; 2-L2; 2-L3) 3 3
Zn deficient (1 ppm Zn) 6 (2-L1; 2-L2; 2-L3) 3 3
Zn supplemented (300ppm Zn) 6 (2-L1; 2-L2; 2-L3) 3 3

from plastic reservoirs. Animals in each group were exposed to 72 hours of continuous

light (630.7 ft. candles) in the photic injury unit Three of the four animals in each group

were sacrificed by lethal injection 72 hours post light exposure and the fourth three weeks

post-exposure. The experimental design is summarized in Table 4. One animal, on the

control diet, was not exposed to light and was utilized as the control for this part of the

experiment along with the control animals from part 1.

Table 4: Summary of experimental design for Part 2 utilizin 72 hours of continuous light
exposure. Four animal were in each diet group and all received light Another animal not
exposed to light was fed the control diet for 10 days and served as the control for this

DIET TOTAL ANIMALS Kill 72 hours Kill 3 weeks
Control diet (l00ppmZn) 4 (L4) 3 1
Zn deficient (1lppm Zn) 4(L4) 3 1
Zn supplemented (300ppm Z) 4 (L4) 3 1

Clinical Examinatios

Initial physical examination (icl ting weig), ni harm examination with funds

photography, and blood sample collection was permnred during the 4-7 day acclimation

period prior to starting the experimental diets. T" procedures wee repeated prior to

sacrifice. For these procedures animals wee aneiheized with xylazine (Img/kg) and

ketaminehydrochloride (10-20mg/kg) injected inu nlryin the neck Euthanasia was

performed using sodium pntobarbital injected tdng= t an ear vein.

To determine if retinal damage induced i this experiment could be detected by

electrodiagnostic testing with a Nicolet Compact Aditory unit (Nicolet Instrument Corp.,

Madison, WI), animals exposed to 72 hours of aimiren high intensity light (L3) and

maintained for 3 weeks postexposure had electrarcdogtmns (ERG)performed prior tonight

exposure and prior to their sacrifice. For ERG emams, animals were induced with the

xylazine-ketamine combination and maint;iainie g Teralanestesiawithalothane gas

delivered via amask. The pupils were dilated with : 'rmpicamide Ten flashes were signal

averaged to generate the ERG wave-frm. ight-aiapoed (photopic) ERGs were obtained

and then the animal dark-adapted for 5 minuat Low and high intensity dark-adapted

(scotopic) ERGs were then obtained.

Blood was collected for dterinati onfcmpe blood count, chemistry analysis,

and serum zinc and copper prior to and at the cod f the experiment. Frzen serum was

stored so determination Za and Cu levels by arnic absorption spectrnmetry could be

performed on all samples simultaneously.

Tissue Collection and Processing

After euthanasia both globes were enucleate and transferred to the lab for

immediate fixation. The anterior portion of the globe w zDanoved at the a serrata. Four




Figure 1: The pie portrays the posterior eyecup of a pacine eye after removal of the
anterior segment Four wedges were removed from .ea anima's eye and preserved by
different fixation. The center of the pie represents the pti~ nerve. The area surrounding
the junction of wedge I and 2 in the dorsolateral quad represents the area centralis.

wedges of freshly dissected tissue from the temporal pcar of the retina were made

(Figure 1). In part one ofthe experiment, the most suped wedge (wedge 1) was placed in

2.5% glutaraldehdee (360mOsm, pH 7.4) and subsequc~ty processes for morphological

examination. Wedge 2 was placed in 10% buffered neatl formalin (BN); wedge 3 in

2.5% glutaaldehyde to be processed for immunndbiisto nrical tainina. and wedge 4 was

slam frozen in liquid nitrogen. In the slam freezing procedure, the neural retina was gently

removed from the wedge and placed ona molybdenum (Mb) strip, grasped with pre-cooled

copper block forceps and immersed in the liquid nitrogen. The rest of the wedge (with the

RPE and choroid positioned to contact the copper block), was placed on a Mb strip and slam

frozen as described above.

For eyes obtained from part 2 of the experiment wedge 1 was placed in 10% BNF,

wedge 2 in combined 0.2% glutaraldehyde and 2.5% parafrmaldehyde to be processed for

immnnohistochemical studies; wedge 3 was placed in 2.5% glutaraldehyde and processed

for morphological studies; and wedge 4 was rapidly frozen by cold metal block freezing

technique (slam freezing) with liquid nitrogen for x-ray microanalysis.

In mrphological studies wedge 1 was used in pat 1 (intennittent light stress and

non-exposed controls) and wedge 3 in part two (continuous light stress) to enable detailed

mrphological examination of two different regions of the temporal retina utilizing both

light and tran mission electron microscopy. Wedge 1 includes the superotemporal retina

including part of the area centralis and wedge 3 the inferotmporal retina.

After removal of the four wedges, the remaining neural retina was removed and

frozen by immersion in liquid nitrogen. The underlying retinal pigment epithelum (RP

and choroid were removed fiom the scera and also fozen by immersion in liquid nirogen

These tissues were stored for deteamintion ofmathi i by the edmiu moglobin

binding assay.

Liver and kidney cortex were harvested from each animal Tissues were fixed in the

same fixatives described above and froen by immersiinliquid nitrogen. Metallothicein

is abundant in the liver and kidney cortex so sections of liver and kidney were used as

positive controls for immunhistochemical studies. Frozen liver and kidney tissue were

analyzed for copper and zinc content by atomic absorption spectrometry and for

metaothionein content by the cadmium-hemoglobin binding assay.

Gluaraldehyde fixed tissues were left in fixative for at least 24 hours, washed in

phosphate buffered saline, post-fixed in osmium tetroxide and dehydrated ina graded series

of alcools, alcohol propylene oxide mixtures and finally propylene oxide: plastic mixtures

until embedding in 100% epon-araldite Tissues for immunohistochemicaltechniques were

processed without ossmicatin through graded series of alcohol, alcohol-plastic and finally

embedded in 100% LR-whiteplastic. For morphological and immiunhistocherical studies,

semithin (1 micron) and ultrathin (80-0lnm) sections were made. One micron sections

were stained with toluidine blue for light microscopic exam. Ultrathin sections were placed

on copper grids for ultrastctural studies and post-stained in uranyl acetate and lead

citrate.m Tissues fixed in buffered neutral formalin were routinely processed and

embedded in paraffm. Five micron sections were cut and stained with hematoxylin and

eosin for histological examintion

For energy dispersive x-ray microanalysis (EDX), following freezing, specimen

were stored in liquid nitrogen until they were lyophilized by a fireze drying apparatus. The

slam freezing technique is believed to be superior to other methods to allow x-ray analysis

of biological tissues.m The specimens were then embedded in a low viscosity plastic

resin (eponaraldite) and 200-400nm semithin sections placed on fonnvar coated nylon grids

(no post-aining).

Morphological Examination and Energy Dispersive X-ray Microanalysis

For the pilot study and part 1, a block of tissue from the center of wedge 1

(superotemporal), encompassing the area centralis, was mounted. For part 2 of the study,

the similar central area of wedge 3 (inferionasal) was mounted and the adjacent mid-

peripheral area was also mounted for sectioning with subsequent examination by light and

electron microscopy. In part 2, selected eyes had central sections from the unosmicated

wedge 2 mounted and examined as this wedge encompasses the area centralis. Electron

microscopy was performed utilizing a Hitachi H-7000 transmission electron microscope.

Quantitative analysis of photic injury was attempted on tissues from animals in part

2 of the experiment. Light microscopic slides were placed on a microscope linked to an

IBAS image analysis unit. The average of 25 measurements of retinal, ONL and RPE

thickness were made on each eye. At 400x, an area was outlined and the total number of

nuclei within the area was counted by placing a mark within each identifiable nucleus. This

allowed determination of ONL density. After identifying the total number of nuclei in the

outlined area, cone nuclei were identified to allow determination of the rod:cone ratio. This

was performed by a computer program which subtracted the cones from the total number of

nuclei in the area to determine the number of rods and then dividing the number of rods by

the number of cones. Lastly, in the same outlined area, the number of degenerating/ dying

nuclei were identified to allow calculation of the % dying cells. A nucleus was considered

degenerating/dying if any of the following was present; pyknosis, significant alteration of

chromatin pattern from normal, and displacement of the nuclei through the external limiting

membrane (ELM). All of these counts were repeated five times on multiple sections from


each eye. On the several sections from each eye, to facilitate statistical analysis, damage

was also subjectively graded and the grades of damage were assigned numerical values as

follows, no damage (0), mild damage (1), moderate damage (2), and severe damage (3).

For x-ray microanalysis the electron microscope was interfaced to a Kevex Super

8000 EDX unit. X-ray microanalysis is based on the principle that when atoms within an

ultrathin specimen are struck by the electron microscope beam (Le., electrons), x-ray waves

of various energies (KeV) are emitted. Each element yields a group of x-rays that are

characteristic of the atom (influenced by the element's atomic weight and valence for

example) and thus the x-rays may be used to identify the elements present in the areas being

scanned.' The EDX unit has a detector which collects the x-rays and displays them as a

spectra with the number of counts per minute (relates to the amount of element in the area

scanned) vs. the energy of the x-ay wave in KeV. Multiple x-ray emission referred to as

lines are generated as the electron beam stories an atom and induces electontransitions with

its electron orbits. For example, in the case of zinc the most intense lines, therefore most

easily detected by the EDX, are the K-lines: K, and K4 with energies of 8.638 and 8.615

KeV, respectively. The next most intense lines are the Kl with an x-ray energy of 9.572

KeV and the LI line with an energy of 1.011 KeV. Several other less intense x-rays are also

generated. The less intense x-ray waves are not usually detected unless very high

quantities of the element are present in the area being scanned.

Energydispersivex-raymicoanalysis was performed n tissues from animals inpat

2 of the experiment, e., 72 hours of continuous light exposure. Elemental x-ray spectra

were made by scanning a melanin granule with a preset time of 100 seconds A minimum

of five melanin granules were scanned from both retinal pigment epithelium and choroida

melanocytes from each eye. A background spectra of plastic without tissue was obtained

from each specimen in which melanin granules were scanned to enable subtraction of any

background elements.

Measurements were made on calcium, iron, copper and zinc peaks within each

spectra. Several background measureents were made immediately adjacent to both sides

of a peak and were averaged to determine the mean background height The height of the

peak was then measured and the peak/background (PK/BG) ratio determined (Figure 2). If

any calcium, iron, copper, or zinc was present in the background spectra, the PK/BG ratio

was determined similarly and then subtracted from the melanin granule spectra PK/BG

ratios. Within an animal the PK/BG ratios of a minimum of five spectra were averaged for

the RPE and the choroidal melanin. For the animls sacrificed 3 days post-exposure (3 in

each group, therefore 6 eyes) the PK/BG values were averaged and statistical analysis


Zinc and Copper Detem inatinm by Atomic Absorption SDectrometxy

These analyses were performed on a Perkin Elmer 360 Atomic Absorption

Spectrophometer in Dr. Robert J. Cousins' lab Department of Food Science and Human

Nutrition, University of Fkrida. To determine zinc and copper concentration in tissues a

nitric acid digest procedure followed by atomic absorption spectrmetry (AAS) wasutilized.

AAS techniques are preferred because of their specificity, sensitivity, precision, simplicity

and low cost per analysis.


A standard curve was constructed using standards with 0.1, 05 and 1.0 ppm of the

element of interest (copper or zinc The absorbance of each standard is measured 3 or 4

times before and after measuring the unknown samples. Using the average absorbance of

each standard and the known concentration a linear rgression was performed to construct

a standard curve. The absorbance fthed sample was then measured and the zinc or copper

concentration calculated by interplation from the standard curve. Fbr zinc deeminations

the wavelength was set at 213.8am and slit width 0.7nm. For copper determinations the

wavelength was set at 324.8nm and slit width 0.7nm.

Serum samples. Whole blood collected by venipuncture in plastic syringes was

centrifuged, the serum transferred intopolyethylene storage vials and stored ina-30'Ffieez-

er until analysis. One milliliter of serum was diluted with 4ml of distilled water and mixed

in a vortex. Distilled water was used as a blank. After calibration of the instrument and

measurement of standards, the absxrbance of each sample was measured

Tissue sales. One tenth of a gram of tissue was placed in a 2.0 volumetric glass

tube (wet weight). One millliter of75% nitric acid and 1 ml of 25% sufuric acid is added

to the tube and the tissues incubated in a hot bead bath ovemight After digestion of tissues

0.5m of the solution is diluted 1:4 in distilled water in a separate glass tube and analyzed



For immunocytochemical localization a gold labeled antibody technique was

utilized.'1n Silver imensiication was used on some sections for examination at the light

microscopic level. '" Tissues in Part were fixed in 2.5% glutaraldehyde and in Pat 2

in combined 0.2% glutaraldehyde-2.5% paraformaldehyde which is believed to provide

better fixation for immunohistochemical staining.'t After fixation, tissues were dehyd-

rated in a graded series of alcohols, and embedded in LR white resin (Electrm Micoscopy

Sciences, Fort Washington, PA). One micron and 80 nm sections were cut and shined

followed by light and election microscopic examination respectively.

For these studies, Dr. Mark Richards (USDA Agricultural Research Service, Belts-

ville, MD) generously provided approximately 10 micrograms of lyophilized prcine

metallothione Iland IIthat were purified by high performance liquid nch~aogp The

metallothionein fractions were divided into three portions for antibody production inNew

Zealand White rabbits. The rabbits were bled seven days prior to their first immuizaion.

Anmigen was mixed with 0.Sml complete Freunds adjuvant and emulsified. jectianmoute

was bcutaneous,intrademal and intamuscular. Thirty days after the initial immniation

an antigen boast was given. The atigen was mixed with 0.5ml of incomplete Remds

adjuvant and emulsified andinjectedby subcutaneous, intradermal and inrascularoe.

Ten days after the second injection a test sample was drawn.

The abbit serum provided was tested utiliing an EISA technique with a got andi-

rabbit IgG alaline phosphatase copnjgated antibody and also tested on sections of liver

using a gold conjugated goat anti-rabbit IgG antibody. Dilutions used wee 1200, 1:400,

1:600,1:800 and l:1000todeennineif antibody against theporcine MThadbeenproduced

and which dilutim was optimal fcr the immunohistochemical studies. A thirdantgenboost

was given 14 days following the second boost. Two weeks later, animals were bled out and

the seramn used in immunhistochenral studies.

For colloidal gold staining the Auroprope LM and IntnSE M kit (Jansen Biotedh

Olen, Belgium) kit was utilized. The manufacturers enmendats were followed with

some modifications of the technique. For studies at the light microscopic level one micrn

sections were placed on clean slides. The staining procedure was as follows. First a 20 pl

drop of 2% normal goat serum, a blocking agent, was placed on the sections for 1 hour at

room temperature. The blocking agent is then removed by suction or blotting. The second

step involves application of a 20pl drop of the primary antibody (the rabbit anti-pig MT

antibody) at various dilutions. The ideal dilution was determined by perfomning several

trials. The sections are incubated with the primary antibody overnight (usually around 16

hours) in a moist chamber at 4'C. The sections are then rinsed several times in phosphate

buffered saline (PBS). The secondary antibody was a goat anti-rabbit IgG gold labeled

antibody with 20 nm colloidal gold particles. The sections are incubated with a 20 pl drop

ofa 1:20 dilution of thesecondary anybody for 8-12 hours at 4C in a moist chamber. The

ideal length of time of the incubation was determined by performing several trials. The

sections were rinsed in phosphate buffered saline, a final wash in distilled water and then

silver intensification performed. Silver intensificatio was done as recommended by the


or immunogod staining ofutrathin sections to beevaluatedby electnmicros

silver intensification is not needed and the method was modified for use with ultrathin

sections on grids. The same antibody dilutions and incubation times were utilized. Grids

are floated on drops of antibody, PBS or distilled water instead of drops being applied to

sections on slides.

Metallothionein Assay

In this study, metallothionein (MT) levels were determined in the pig utilizing the

cadmium (Cd) hemoglobin (Hb) binding assay.' The assay was performed on aeural retina,

liver and kidney samples to add determine zinc status and any possible effects ofdiet and/or

light The method consistedof: 1) weighthe sample, dilute (usually l:5)withTRIS bufferand

homogenize (at this point, two lOpl aliquots of homogenate are removed for protein

determination by the Lowry method); 2) centrifuge to obtain a supernatant fraction; 3) heat-

denature the supernatant fraction, if high levels ofMT are anticipated which would be out of

the accurate range of the assay, dilutionsofthe supernatant with TRIS buffer canbe pefomned;

4) incubation of 200 pl the supernatant fraction for 10 minutes with a 200 pl ofasolution that

contains 2.0 pg/ml ofcadmium total and 0.5 pCi of 'Cd; 5) addition of 100 pl 2% bovine

hemoglobin (Hb) solution; 6) heat precipitation ofHb; 7) centrifuge; 8) repeat steps 5,6 and

7; and 9) analysis of 100pl aliquotofthe supernatantfaction for Cdremaining iaefraction

utilizing a gamma counter. In this assay mCd remains in the superatantfiactioaealyinthose

samples containing MT. With this assay the amount ofMT is calculated as follows

CPM (s) CPM(bkg)
X (16.52)X initial DF X supernatant DF= pg MT/I g sample

The value of 16.52 is a constant determined by expressing the concentration of

mallohionein the solution and was oained by dividing the grams ofcadmiumineachsample


(for 200il sample and 2 pg/ml Cd 0.4pg) by the atomic weight of Cd (0.1124 pg/mol)

and the MT sample size (0.2 ml) to yield units of nanomolesdofC/mlofsample~ This value

was multiplied by the molecular weight of MT (6500 Dalons) and then divided by 7, the

number of Cd atoms that is bound by each molecule of MT. It is stated in the original paper

that the accuracy of the assay is best if the sample counts me between 10 and 50% of the

total counts. Therefore additional dilution or less dilution of the tissue in step 1 or the

supernatant in step 2 was performed as needed if the samplcounts were out ofthis 10-50%


To avoid variability produced by different wet weights oftissues the final MT value

from the assay in ipgfmg tissue is divided by the protein ca nation (mg protein/mg

tissue) determined by the lowry method. This results in a value of pig of MTI mg of


Eyes from namal pigs at a University of Florida Meats Laboratory were utilized to

test the assay on ocular tissues. Due to the small amount ofocular tissue (0.06-0.15 gams)

modifications of the assay were required. The sensitivity of the assay was increased by

increasing the amount ofradioactive cadmium in the solution. Several runs were performed

utilizing two test solutions: 1)0.2 pg Cd total with 0.5 ipQ Cd"'ml and 2) 1.0 g Cd total

with lC Cdi mla. The chroid-RPE tissue was very scant, i..,usually 0.05 grams, and

also it was difficult to homogenize due to its high collagen content The colagenous tissue

tended to become caught inthe homogenizer blades resultingina loss of some ofthe already

very small quantity of tissue. An ultrasonic homogenizer did not successfully homogenize

the collagenous chroidal tissue without generating unacceptably high temperatues that


would denature protein. All these factors made it impossible to accurately determine MT

levels in the choroidal tissue. Several test runs of the assay did not result in accurate

measurement of MT in the choroid-RPE tissues. In all trials the sample counts were much

less (ie., 4-6% of total counts) than the ecomended minimum of 10% of the total counts

which probably results in an underesimtion of the amount of MT level" Therefoe, this

assay was not performed on the clhroid-RPE of experimental animals.

For neural retinal tissue, the solution with 1.0 pg Cd total with IpCi Cd' ml gave

consistent results in several test runs, so this solution was utilized on tissues from

experimental animals In addition to utilizing this solution in step 3, in step 1 all tissue was

weighed and 0.35 ml of RIS added (from the weight and the volume of TRIS the initial

dilution factor was calculated). Because a different cadmium solution was ntlized, the

constant, which is based on the amount of total cadminm in solution, was recalculated and

for the pg Cd/ ml solution was 9.20 Because the amount of neural retina was small the

tissue from both eyes of each animal was combined so only a single eye MT value was

available from each animal

Statistical Methods

Statistical analysis was not attempted n data from pat 1 of the experiment because

only one animal was present at each daa point. In prt 2 of the experiment, 3 animal

comprised each diet group exposed to lght and sacrificed 72 hours post-exposure, which

enabled some statistical analysis. However, only one control (unexposed animal) per diet

group limited the power of statistical tests.

Most variables of interest were observed in each of 2 individual eyes from 13 pairs

of eyes belonging to 13 pigs. These pigs belonged to I of 3 possible zinc diet groups (4 in

the Zn-, 5 in ZnN and 4 in Zn+ group). Within each of these diet groups, three pigs were

exposed to 72 hours of continuous excessive photic conditions intended to induce retinal

light damage, while the remaining pig in each group (the ZnN group had two non-exposed

controls) remained unexposed to excessive light and were considered diet groupspecific

controls in the design. Single kidney and eye MT (combined tissues from both eyes) were

detennined from 9 additional pigs exposed to intermittent excessive photic cnditions (2 fed

the Zn- diet, 4 fed the ZnN diet (pilot study animals included] and 2 fed the Zn+ diet). This

allowed comparison of kidney and eye MT values in non-exposed, intemittent and

continuous exposed groups irrespective of diet, i.e., to see the influence of light on MT

levels. For kidney and MT levels this resulted in an overall 2 factor independent groups

design, with the zinc diet group factor having three levels and the lght exposure factor also

having three levels

For response variables in which pairs of observations were available from 13 pigs,

Pearson'scoa elation coecient and Spearnan's rank coaelaticcoefficientwe computed

for OD vs. OS observations within subject to determine the degree of indepdence of

consistency between eyes within animal A paired T-test and the no-parametricaly

equivalent Wilcoxon signed rank test was also perfiamed on the mean pairwise difference

between eyes to detemnnie if OD differed significantly from OS in any response variable.

Paired observations within each pig were also averaged for all response variables except the

subjective retinal damage scales (0-3, nomal to severe) to obtainper-animal values for each


pig. The maximum score within a pair of eyes was used as the per-animal values for each

pig. All analyses were carried out separately on both individual eye datasets and per-animal

data sets.

For all continuous variables other than kidney and eye MT, a completely crossed 2-

factor independent groups analysis of variance (ANOVA) with 2 levels of light exposure

(none or continuous) and 3 levels of zinc diet groups (Zn-, ZnN and Zn+) was used to

determine statistically significant differences among various group means. These same

variables were also analyzed in an analysis of covariance (ANCOVA) model in which

kidney MT levels were considered as an indicator of dietary zinc status which could be

substituted for the zinc diet group factor and used to adjust the light exposure group means

before statistical comparison.

The outcome variable "percent cells dead" ranged from 0-100% and was therefore

transformed prior to analysis using the arcsinesquare root transfoaation in order to

stabilize variance as a function of mean levels. Kidney and eye MT levels were also

analyzed in a similar manner as described above except that an additional level for the light

exposre factor categorizing MT levels in anils exposed to intermittent light was included

inthe2 factor ANOVAs. One way ANOVA andthe non-pranetricay equivalentKrskal-

Wallis (KW) test were used to compare exposure group means within diet group and to

compIre diet group means within exposure group. The subjective retinal damage scale

observations were only compared using the KW test. One way ANOVA and the KW test

were then used to compare damage scale group means to see if continuous response

variables levels were associated with the degree of subjectively assessed retinal damage.


Finally, all response variables in both the individual eye and individual animal

datasets were correlated with one another using the Pearson and Spearman rank correlation

coefficients. Pearson correlation coefficients were determined for elemental specta (Ca,

Cu, Fe, and Zn), percent dying cells, kidney MT, eye MT, ONL density and subjective

damage. Only significant correlations will be presented. When the r is greater than 30%

it suggests the correlation may be real Le. the % of variability of ne variable is explained

by the other variable



For Part I of the experiment, the facility utilized to house the anima being fed the

zinc deficient diets was difficult to adapt to insure no exposure to exogenous zinc. The

cages consisted of three sides of cement block and the front a section of galvanized chain

link fence with a door. This section of galvanized metal was considered to be a potential

source of zinc so a piece of peg board was secured to the fence to prevent the pigs from

chewing an the metal Deicnized water was provided to all animal on the zinc deficient

diets. The pigs chewed away sections of the pegboard and/or pulled the peg board off to

expose the metal on several occasions. Thus they cold have obtained exogenous zinc. To

remedy this problem, in part two of the experimeo, stainless steel cages were obtained to

hose the pigs fed the zinc deficient diets.

Pundus Photopahy

The normal funds of the porcine eye is hown in figures 3 and 4. The retina ofthe

pig was holangiotic and ataptal. The optic nerve wa well myelimate and horintally oval

with retinal arteries and veins on its surface (Figure 3). There was three or four large veins

and 5-6 smaller arterioles. Dorsotemporally the vessels arced around a grossly avascular

area centalis (Figure 4). None of the experimeaal animal fundic examinations revealed

Figure 3: Normal porcine funds. The area of the optic nerve is shown. The holangiotic
vascular pattern is evident by the numerous vessels on the surface of the optic nerve. The
porcine retina however, does not possess a central retinal aretery as is present in humans.

Figure 4: Normal porcine funds with area centralis shown. Note vessels arcing around this
dorsotemporal area to create a relatively avascular area.


clinically apparent retinal damage at 72 hours or 3 weeks post-expure respective of diet

or light treatment

Electrophsiologic Studies

Electroretinograms were performed on selected animals from part I of the study

prior to light exposure and then three weeks after exposure prior to euthanasia The typical

electrretinogram obtained from all pigs is shown (Figure 5). The amplitude of the b-wave

actually increased during the three weeks between the initial pre-exposure evaluation and

the repeat test 3 weeks after 72 hours of light exposure. No difference in b-wave amplitude

or latency were noted in pre and post-exposr electrenograms of the animals fed the

Zn-, ZnN o Zn+ diets for three week periods after exposure to 72 hours ofintermittent light.

Because no changes were found in the intermittet light group, ERG testing was not

perfonned on the continuous light group.

Blood and Tissue Analyses

Complete blood count and chemistry analysis performed on all animals at the

beginning of the experiment revealed no abnomaliti Values were within normal limits

of young pigs for the clinical pathology laboratory at the University of Florida College of

Veterinary Medicine. In the control group fed the diet for 11-13 days there was one animal

fed theZn-diet and one fed theZn+ diet. Two animals were fed the ZN diet sotheaverage

of the two values, e. serum, kidney MT, were utilized in statistical analysis.

Serum samples drawn at the beginning and nd of the experiment (at sacrifice) were

analyzed for copper, zinc and calcium. Results are shown in Tables 5-7. Statistical analysis

revealed no significant change in serum levels of copper or zinc as a result of feeding the




z C


' S

o -5



1. -s

- >.
E =L






1 I



I g;

-o O Co 40 Co o
Sg o o on o
Sd 4 o ;
. d nddddoc

ao a io a e
Cso a q % *n V 4
o @o c 1 a r l
I~a O\ --r
$ : i i e 2 i C s i ee

oo 200 0

S c d c 2 o a
I -. 0s a 0s 0

% 9 -

O O a I O^
E 00 0

G o o o o a*Q

% _n
OC 0c- 01- U 0< we n i '
0 00 d 00 0000 0 0
S (EM O M Ot'0 -o l0 0 <

i2 '0C e4'0- ~
Ch oq t- W
^ 1 ri ?q" cea c 'n M tn op
=L v C4 % C-;ri oi

=L C -i ivi vi wi ; v 'lri v~i C*% cico C4


9 'V
-- -- -m -m -- -

d o r-

1 1;100000
tj to co cns^

N l IV on Cl Cli 'A 0 C -
4 v eq a V C 4 rrb

0 II O C t c
es. d o do o o

5^p ^. o t^^ oonr^^Q f S

cS~ -2 r> n '<- 5 r ooth
a2 dldld do' od
w ^

'PC NY '^ ^ Scn' O N n O -

lY vi I E= -wo r^ ^- oo
r c c t N N ^ -


. I




1. 1
I: j



0 00 C0 O m %D o0 0 CI 0 0 0m 0n t-
S O; od a o o0 o oo ao odZ oi CZ

0L t ro ; .- Cr ei n co vN -;
tS o tS -o 4 o %o

0 e on -o c- -'o
G Q q P

Go o o CI- -

o 0 S

6 5 t ffri F 1 o cs i cs g s d
sn en i "

aG w a %

m -^b a c o an -
8--- --cs-m-c--m

n- m m n-

- .6 9 W 9

_ O r- f e O O O
0v 0 0 ^ 0 0

at F= b S 5

0 N00o 0

0 0 0 0 0

d 0 00 d d
n i4& oo
e en e

c3 (n I e V I en

_ 1 1 1

different diets for 11-13 days or for the animals fed the diets for 3 weeks. The small number

of control animals made it difficult to compare the control vs. the exposed groups with

parametric tests such as the analysis of variance (ANOVA), so the non-parametric Kruskal-

Wallis (KW) test was also performed on the data.

At time of sacrifice, the mean serum zinc concencations for the control animals fed

the Zn-, ZnN or Zn+ diet for 11-13 days were respectively 031, 0.82 and 0.52 ppm.

Statistical analysis by the ANOVA and KW tests revealed no significant difference between

the three group means, p 0.34 and p 0.26. In the continuously exposed group the serum

zinc levels for the Zn-, ZnN and Zn+ fed animals were 0.98, 1.04, and 0.86. Again no

significant difference between the three groups was present (ANOVA p 0.60 and KW

p 051).

Copper and zinc levels in the liver and kidney were determined by atomic absorption

pectrophtmetry and are also shown in Tables 5-7. There was no significant difference

in the levd of these elements between the different diet groups. The liver zinc levels for

control animals fed the Zn-,ZnN or Zn+ diets for 11-13 days wererespectively, 44.88 (n-l),

51.71 (ii-2) and 44.77 (n-1) ppm. The ANOVA and KW tests revealed no significant

diffeence (p 0.70 and p 0.26). The kidney zinc levels in control animals fed the low,

normal and high levels of dietary zinc for 11-13 days, 28.4 (n-), 33.4 (n-2) and 35.2 (n-1)

ppm were not significantly different (ANOVA p 0.71 and KW p 0.41).

Kidney levels of Zn and MT are also shown in Table 5-7. Kidney MT levels, unli

liver MT, are a good indicator of zinc status that is not as affected by variables such as stress

or disease.= For control animals fed the Zn-, ZnN and Zn+ diets, the kidney MT levels

were 0.18, 0.26 and 0.26 pg MT/ mg protein. The ANOVA and KW tests revealed no

significant difference between the groups (p 0.11 and p 035) In the continuously

exposed group the kidney MT levels in the Zn-, ZnN and Zn+ fed animals were 0.09, 035

and 0.26 pg MT/mg protein. The ANOVA and KW showed that the three means were not

significantly different p 0.08 and p 0.09. The differences however approached

significance suggesting that in the cotinuously light exposed group, the animals fed the zin

deficient diet had lower kidney MT levels when compared to those fed the nominal and high

zinc diets. It is not suprising that the animals fed the high zinc diet did not have any

significant alteration in status because the experimental diet did not have the desired level

of 300 ppm zinc blt had only 240 ppm zinc.

Feeding the different levels of dietary zinc for three weeks appeared to influence the

liver and kidney zinc and MT levels to a greater extent than feeding for 11-13 days. In

several cases if livr and kidney Zn and MT values of the animals fed diets for 11-13 days

vs. 29-31 days are compared it is appears tat there is some influence of diet, e., values

from animals fed the Zn- diets are low, ZN intermediate and Zn+ higher (Table 8). In

animals fed the diets for 29-31 days statistical analysis could not be performed because only

one animal was in each diet or light group.

If the data in Tables 5-7 is closely examined certain animal may be identified that

had significant changes in the Zn and Cu in seum, liver, or kidney, or that had changes in

metallothionin levels that suggest some alteration in zinc status. In the control group it

appeared that #125 fed the Zn- diet for 11 days had a decrease in serum Zn, however, when

the liver and kidney Zn levels are examined no significant decrease in Zn content was

apparent The kidney MT level however was relatively low compared to the rest of the

population at 0.18 pg MT mg protein. Similar tissue levels of serum and organ Zn and

MT are present in #111, the animal fed the Zn- diet for 29 days The values suggest both

of these animals had a mild negative Zn balance. Animal#109, fed the Zn+diet for 29 days,

had positive changes in serum Zn, liver and kidney Zn and in MT (0.61 pig MTImg PR)

suggesting that feeding the Zn+ diet (with 300 ppm Zn) for29 days induced a supplemented


Examination of the intermittent light diet groups revealed a similar trend in

increasing kidney MT levels with increasing amount of dietay zinc. The mean kidney MT

levels for the animals fedthe low, normal and high zinc diets (n-2 per diet group) were 0.24,

0.40, and 035 pg MT/mg protein. Certain animals were identified that appeared to have

altered status. Animals #120 and 121 had low kidney MT values. In animal #120, which

also had decreased serum, liver and kidney Zn compared to the rest of the population, the

low dietary zinc level probably induced a mild deficiency afer 32 days of feeding. The zinc

deficiecy however appeared marginal as no overt clinical symptoms such as paraeratosis

were observed.

Inthe cotinuousy exposedgrop thetrend ofireasing kidney Mwith increasing

dietary zinc levelagain was apparent. Very notable waste low kidney MTlevelinanimals

#179 and 364 both which were fed the Zn- diet for 13 days. Seum, liver, and kidney zinc

leaves, however, did not substantiate significant low zinc status. Animal #182 had low liver

and kidney zinc and somewhat low kidney MT comparedto the control groups' values which

again may indicate a mild negative zinc balance. In contrast animal #366 seemed to have


a positive alteration in Zn status after being fed the Zn+ diet for 31 days since serum, liver,

kidney zinc and MT levels were greater than that of the rest of the population.

Nonnal Porcine Retina (unexosed)and Control Animals

The normal central porcine retina in the inferatempceal quadrant had a mean retinal

thickness of 272.7 pm. The outer nuclear layer (ONL) averaged 39.8 pm in thickness,

which was an average of 6-7 nuclei in height. The ONL was usually a little thicker than the

inner nuclear layer (INL). Cones were numerous and appeared as a layer of euchromatic

nuclei just internal to the external limiting membrane (Figure 6). The staining pattern of

rods and cones differed; the rod nuclei were more heterochromatic than cone nuclei (Figure

7). Internal to the ONL the cone pedicles were very promint The ellipsoid regions

(containing mitochondria) of porcine cone inner segments (IS) were large, plump, and stain

densely with toluidine blue (Figure 6) UbrastructuraUy rod and cone IS were packed with

mitochondria and the rod IS was longer and narrower than that of the cone.(Fgure 7). The

mitochondria in the IS of both types of phtoreceptcrs had numerous cristae indicative of

a high metabolic rate.

Sections were made from the peripapillary to the peripheral retina in the

inferotemparal wedge of eye tissue from a control animal on a nomal Zn diet. The nerve

fiberlayer progressively thins frm the optic nerve towards theperiphery. Larger aterioles

and veins were found in this layer (Figure 6). Small vessels and capillaries were found in

the ganglion cell layer (GCL), the middle of the inner plexifim layer (IPL) and within the

INL. The ganglion cell layer contained small, medium and large cells. The density of

ganglin cells decreased towards the periphery. The INL nucear thickness also decreased

0o s

. oo0 2

g r. x

0, -

U 0

an U


< I ad

20 A0
t -0 0

u Bo : *s t

0 2 2
t o
2^i -'

u~ 2 $ a
0 o s 3
^c -j '
-,: -
^1 g-g

c u2 4i




Figure 7: Outer retinal layers of normal porcine retina.
a) Electron micrograph of the ONL layer of a control animal fed the ZnN diet. The
rod nuclei (r) have more heterochromatin than the cones (c). Note lack of
intercellular space. (8000x) b) Rod (r) and cone (c) inner segment ellipsoid regions
are packed with mitochondria. The cone IS are short and plump compared to the
long narrow rod IS. Note that the mitochondrial cristae in most instances are easily
differentiated and very numerous, indicative of the high metabolism of the
photoreceptors. Some mitochondria are disrupted (arrowheads) but the infrequency
of this observation indicates that it is an artifact. (8000x)


4 t,


from the optic nerve to the periphery. ONL thickness decreased as the distance from the

optic nerve increased. The ONL, in number of nuclear layers, varied as follows:

peippillary, 6-7; central near optic nerve (paracental), 5-6; central, 4-5; and peripheral,

4-5. The ONL was notably thicker than the INL in the peripapillary retina. In the central

section near the optic nerve, the INL thickness approximated that of the ONL Inthe central

section nearer the periphery and the peripheral retina, the ONL was thicker than the INL.

he number of cones decreased from central to periphery. ON this inferotemporal wedge,

the rod to cone ratio in the ONL varied as follows: peripapillary, 8-9:1; paracentra6-7:l;

central 5-6:1; and peripheral, 8:1.

A similar series of central to peripheral sections were taken from a control animal's

superotemporal retina (Figure 8). The central area of this wedge represented the area

cenralis. The NFL and vessels were similarly distributed as the infertempral wedge. The

G( in peripapllary and central regions in this wedge was thicker than the infertemporal

wedge. The number of ganglio cells again decreased in the peripheral sections.

The area centralis (AC) can be identified by several histologic features. As the area

centralis is appached, the inner nuclear layerbecomes equal to or greater inthickness than

the ONL The GCL is 2-3 layers of densely packed cells. The rod to cone ratio in the AC

decreases to 4.4:1 indicating a relative increase in cone density. The average rod to cone

ratio is 6.7:1 in the inferotemporal quantrant and 5.5:1 in the superotemporal quadrant

susantiating the presence of a cone rich area in the superotempral quadrant.

In the srperotempral wedge, similar to the inferotemporal wedge, the ONL

thickness progressively decreased towards the periphery being 6-7 nuclei thick near the



t| 2

i n

o C6

4 -




4-4 )
0 "


(A cd 0i ~
U U8 M -b


~~~~~ 0 4)*~-

al 00-* C
Ipiu 4p
!z .s 5; Ecg g | a s

~ZU Ul ~ ~

c-o 0
o flSk U u iu-fr
i( z 1 2I< y
^ ^M ti^ a S^

g '- a s9 S3 aP'
D. i-i a a ^.o(c- o
o "o'G c3O. ^5&
f i- rS~ U r; < ._ '1,CCo

SE a

aD o
0 0

cn, 00
. -"' ea ^ -a .o o


-0 0
t) a I 'd w i

*u o 0 0 U -7
0 U OL s-


M- '

au R-5
|i ii!Cli^]

sa a gat g~ Pig Sl^P m

E4 C3 ^3 US J-i a)S < C B
5) r"% ^1 c ^
^S'S3 a u S'" -

s|!og |ii8

-n ^31 3 g f2^
bS(dS S







.J'. .b

E ~~~f~4klt-F


optic nerve, 5 in the central area near the optic nerve, 4 in the central area nearer the

periphery and 4 in the periphery. The rod to cone ratio in the area centralis usually was

between 4-5:1. From the optic nerve to the periphery the ratio changes from 8-9:1 near the

optic nerve, 55:1 in theparacentral sections, 4.4:1 in the central section (with AC, and 5.4:1

in the periphery.

The average retinal pigment epithelial height in theinferoten ral quadrant was 8.9

un. Uklastrcturay te RPB cytosol contained organeles typically found in the RPE of

other species including: melanin granules, residual bodies, phagocytized OS material,

lysosomes, golgi apparatus, and smooth and rough endoplasmic reticulum (Figre 9). The

basal cell membranehas some infolding but this was not aprominent feature as seen in other

species. The nucleus islarge, oval, and echromatic. The pig has a well developed Bruch's

membrane with five layers in areas adjacent to the choriocapillaris lament nmembanes

of RP, iner collagn zone elastic fiber layerouter cllagenos zone and the sent

membrane of endothelihn of the cboriocapillaris.

Control ightlnn (14 days on diet)

Normal zinc dit. This animal received the com-soybean based diet for 14 days.

Retinas of both eyes of this animal were normal Rarely, a cell was found degenerating and

moving through the external miting membrane (ELM). Sme ONL nuclei had an altered

chrmatin appearance with a rounding up of the nucleus and more echromatin. The left

eye of this animal had amoderate number of immature melanosmes in the RP indicative

of melanogeneis.

Figure 9: Normal porcine RPE-OS interface.
a) The OS are well embedded in the microvilli of the apical RPE. The cytosol of a
normal cell is dense due to the large amount of smooth endoplasmic reticulum that
fills all extra space. (8000x) b) Higher magnification of RPE cytosol demonstrates
the typical apical junctional complex (jc) that represents the outer blood retinal
barrier. Other organelles are evident: melanin granules (m), lysosome (1), and
mitochondria (arrow). (20,000x) c) Normal Bruch's membrane adjacent to a capillary
of the choriocapillaris. The membrane has five well developed layers as in humans:
1) basement membrane of the RPE, 2) inner collagenous zone, 3) elastic fiber layer,
4) outer collagenous zone, and 5) basement membrane of capillary endothelium.
(30,000x, TEM)


Mp a




:, 2 "

Zinc deficient di This animal received a zinc deficient casein-sucrose based diet

for 14 days. Mild disorganization of the outer segments was apparent at the ligbt

microscopic level and confirmed by transmission electron microscopy. Displacement of IS

into outer segments was seen. On ltrastructural examination, IS were swollen and

abnonmally shaped with sme degenerative mitrcondria (Figure lOa). Most nuclei in the

ONL were normal except for an occasional degenerating nucleus observed in the ELM.

Melnogenesis was observed in the left eye of this animal and incompletely melanized

granules were very numerous in several RPE cells. The other eye of this animal did not

exhibit melanogenesis.

Zinc supplemented dit. This animal received a zinc supplemented corn-soybean diet

for 14 days. The retina was normal except for an occasional degenerating cell located in the

ELM. One IS wasobserved between the RPE-OSinterfaceandmild OS disorganizationwas

observed in some areas. Evidence of melanogeneis was found in the left eye of this animal

In this eye immature mlanin granules were infrequently observed.

Control Ligtin. (33 days a diet)

Nomal zinc diet. This animal received a crn-soybean based diet for 33 days. The

retinas of this animal were nomna Immature meanr omes with visible meanfilaents

indicative of meanogenesis were observed in both eyes of this animal. In one eye the

immature granules were infrequently observed.

inT deficient die This animal received a zinc deficient casein-sucrose based diet

for 33 days. The retinas were normal with mininWa evidence ofdegeneration. There was the

occasional focus with disrupted OS and a few pyinotic nuclei in the ONL (Figure 10b). The

0y C

o "



0 0

o -a

d ed

-3 0.) -
U.. i

tod 4O ) 3
O C)


( U -

IaS 113



inner retina was normal The outer segment- RPE interface appeared loose as the OS were

not well embedded in the RPE microvilli Some disorganization of OS and some

degenerative IS were be observed. No melanogenesis was observed in either eye of this


Zinc supplemented diet This animal received a zinc supplemented corn-soybean

based diet for 33 days. The retinas of this animal were normal Immaturemelanosomes

were observed infrequently in one eye of this animal suggesting some melanogenesis was


Photic Inurv

Animals were observed twice daily while in the photic injury unit and did not exhibit

signs of stress. Behavior observed included laying down and sleeping, or walking around

the cage playing with the food dishes or water supply. Because the unit was elevated off the

floor to allow for cleaning, the pigs were able to and were often observed laying near the

edge of the unit and partially shading one eye by positioning their eyes under or near the

edge of the bottom of the unit.

Pilot Study: 36 and 72 hours of Intermittent LiUht and Heat Stress

The intermittent lighting schedule was 18 hours light and 6 hours dark with light

onset at 8 a.m. As mentioned previously, during the plot study the temperature in the unit

averaged 90-93F and animals exhibited a tendency towards hypothermia with rectal

temperatures of 105-106F compared to the normal range of 101.6-103.6'F.' 'Several

studies have demonstrated that elevated bodytemperature (hyperthennia) accentuatesphotic

injury '. To minimize the number of experimental variables that might influence photic

injury, in further trials, a plastic shield was used to absorb and reflect heat, nd along with

fans maintained the temperature in the unit at 80*F.

The retinas of animals exposed to 36 and 72 hours intennittent light stress and

sacrificed 72 bours after exposure were severely damaged. The type of damage was similar

in both animals but subjectively it was more severe in the animal exposed to 72 hours of

intemittent light.

Damage was diffuse and present in all retinal layers (Figure 11). In the animal

exposed to 72 h of intennittent light, some ONL nuclei were pyknotic but most appeared

swollen. All nuclei in the ONL exhibited chromatolysis with loss of distinct chromatin

(Figure 12). In both animals, damage to the inner retina was less pronounced and limited

toa generalied loss ofstaining which was interpreted as an edematous response. SomeINL

cells were pykotic; again this occurred moe frequently in the animal exposed to 72 hours

of intemttnt light (Figures 11 and 12).

Inner segment mitochondria were disrupted and swollen. The outer segments

appeared faiy normal with only sme exhibiting disorgani In the animal exposed

to 36 hours of inmittnt light the RPB appeared nrnal other than a few swollen cell

In the animal exposed to 72 hours of inemittent light, RPE cells were tall and had

vacoolatedcytoplasm suggestive ofintracelular edema (Figure 12). Many tohndriain

the RPE cytosol appeared swollen.

Part 1: Intmnittent Lh Stress (36 ours): 72 hours Post-Expoure (12 days n diet)

Three of six eyes examined had significant injury. The other 3 eyes had no damage

or only mild and often focal degeneration that was not significantly different from controls

Figure 11: Severely damaged retina of light-exposed (72 hours intermittent) and heat-
stressed animal.
a) Diffuse retinal edema is present. Several nuclei in the INL are pyknotic
(arrowheads). OS are disrupted, the RPE layer is tall due to edema and the RPE-OS
interface is compromised. (800x) b) Most nuclei in the ONL are swollen or
pyknotic and the ellipsoid region of the IS are edematous (arrows). Nearly all the IS
of the rods exhibit swollen mitochondria with this change occurring less frequently
in the cones. In many RPE cells, the cytosol appears vacuolated (arrowheads).

~s41r c<

E* ,, *


-1 -

o o o; n
o o


C 0


o o d E

0 o +
- W*
a -oo (u U

"00 o


* 5 U
0 0 2S
^*a fi aUU
^, a ^ .

- U 1 v -p-- -

. : Y


Two of the three damaged eyes were from the animal fed the zinc normal diet. One eye of

the animal fed the zin supplemented diet showed significant damage.

Normal zinc Moderately severe damage occuned in both eyes of this animal

The retina was diffusely edematous and artifctually detached presumably due to poor

attachment between the RPE and OS. If the attachment between the RPE and OS is firm,

artifactual detachment usually results in ton OS and remnants of OS are left an the surface

ofthe RPE. Threfate, when these changes are not observed it usually indicates poor RPPE-

OS attachment. Multiple nuclei in the ONL were pyknotic and located in the ELM (Figure

13). Some INLnudi were pyknotic. In some areas, inner and outer segment degeneration

was present. Rarely, the occasional RPE cell was degenerating with vacuolation of the

cytosol or pyknosis of the nucleus.

Zinc deficient dict The retina was normal except for one focus of moderate

degeneration at oe end of the section. In this degenerative area, changes observed were

similarto that foundinthe animal on the zincnormal diet Muiple nuclei inthe ONL were

pyknotic and located in the ELM. Some OS lameflae were disorganized and the

mitochondria in IS were unrecognizable due to swelling and disruption of normal

architecture Three other regions of this eye were examined and appeared nmomal so this

focus of degeneration was unlikely to have been induced by light. The otl eye of this

animal was nornaL

Zinc supplemented diet The left eye of this animal exhibited very mild damage

manifested as aheraion of the chromatin pattern of several nclei in the ONL (Figure 14a).

Affected rod ncle were mor euchromatic and rounded compared to normal nuclei in

o U

s 2
o .
- -

- ~oo
,U o

I mit

S E0

o 0

' O0

0 0
o o2
I 10

ca 2

c4L a -
a iflU
S 0U
U, u u u

- b U





5- *
o S
iC a


a 0

a B

I .2

S 0

. Io

0 0

5 "


. 8

- &

00 o


o o

o U

g 0)

O 0

0 2



ofr if



oanrol eyi. No nuclear displacement was noted and the rest of the retinal layes were

moonaL The right eye exhibited mild damage (Figure 14b). Numerous cells were

undergoing pyknosis and moving through the ELM. Several RPB cells appeared tall and


Pat 1: Intermittent Light Stress (36 hours 3 weeks Past-Exposme (30 days on die

Zinc nor diet. Both eyes of this animal had vry mild signs of degeneration not

significantly different from that observed in some control animals retinas. The occasional

rod nucleus was euchromatic and rounded. Some vaCudated IS were observed in the ONL

(Figure 15a). Subjectively, in some areas there appeared to be a decrease in ONL density.

These findings suggest that prior light-iduced damage had been partially repaired.

Zn defcient die. Both retinas appeared In one eye normal appearing IS

wee observed on the surface of the RB suggest pr displacement of uneahy IS

which then recovered.

Zrin supplemented diet The retinas of both eyes of this animal appeared nomnaL

Again increased intercellular space was apparent in the ONL suggesting los of nucli

Also, of the six o yesal examined had moderate to see damage. The nimal on hes

inc decent diet had mil damage ine eye andthe other was nonnal. The four damaged

eye were from the animals fed the zinc normal and :sc supplemented diets.

Figure 15: Comparison of retinas of animals fed the ZnN and Zn+ diets. Both animals were
exposed to 36 hours of intermittent light and sacrificed 3 weeks post-exposure.
a) Animal on ZnN diet. The retina is artifactually detached but the RPE-OS may
have been week as no OS remained adhered to the RPE. There appears to be some
loss of nuclei in the ONL layer as intercellular space is increased and cone nuclei are
not always on the border of the ELM (arrow). Some IS are vacuolated and within the
ONL (arrowheads). The IS and OS region is disorganized but rarely swollen or
degenerative suggesting recovery from prior damage may have occurred. The RPE
appears normal. (800x) b) Animal fed the Zn+ diet for 3 weeks. Retinal attachment
is strong. Several displaced cells and IS (arrowheads) are often apparent in the OS
layer on the surface of the RPE. The inner retina is normal. (800x)