Cyclic nucleotides in the (rd) retinal degenerate chicken retina

MISSING IMAGE

Material Information

Title:
Cyclic nucleotides in the (rd) retinal degenerate chicken retina
Physical Description:
viii, 160 leaves : ill. ; 29 cm.
Language:
English
Creator:
Lee, Nancy Ruth, 1961-
Publication Date:

Subjects

Subjects / Keywords:
Research   ( mesh )
Retina -- physiology   ( mesh )
Retina -- chemistry   ( mesh )
Retinal Degeneration -- genetics   ( mesh )
Cyclic GMP   ( mesh )
Cyclic AMP   ( mesh )
Chickens   ( mesh )
Phosphoric Diester Hydrolases   ( mesh )
Department of Neuroscience thesis Ph.D   ( mesh )
Dissertations, Academic -- College of Medicine -- Department of Neuroscience -- UF   ( mesh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1991.
Bibliography:
Includes bibliographical references (leaves 145-159).
Statement of Responsibility:
by Nancy Ruth Lee.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001664817
oclc - 25668342
notis - AHX6591
System ID:
AA00009065:00001


This item is only available as the following downloads:


Full Text













CYCLIC NUCLEOTIDES IN THE (rd) RETINAL DEGENERATE
CHICKEN RETINA



















By

NANCY RUTH LEE


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1991


































This thesis is dedicated to the memory of Robert J.

Ulshafer, Ph.D., my mentor, friend and inspiration for this

work.















ACKNOWLEDGEMENTS


I thank my advisor Dr. Robert J. Ulshafer for sharing his

knowledge of the retina and expertise in electron microscopy,

for his patience as a mentor and for his kindness, generosity

and sense of humor as a friend. His enthusiasm for his

research sparked my interest in the retina and was the

inspiration for my disseration project.

I am grateful to Dr. William G. Luttge for his support,

encouragement and advice as both coadvisor and chairman, which

have facilitated completion of this dissertation.

I thank committee member Dr. Robert J. Cohen for his

technical advice and knowledge concerning various biochemical

aspects of the project and for his support and interest

throughout the course of the dissertation studies.

For her enthusiasm, shared interest in the rd chick model

and assistance with dissections, I thank committee member Dr.

Susan Semple-Rowland.

I am grateful to committee member Dr. William W. Dawson

for his technical suggestions and the use of equipment and to

previous committee member Dr. Adrian J. Dunn for his technical

recommendations.

I thank Dr. Melvin L. Rubin, Chairman of the Department

of Ophthalmology, for the use of the department's facilities

iii










and equipment throughout the course of my dissertation.

I thank Connie Daughtry and Evelyn Clausnitzer for their

assistance in the preparation of this manuscript.

A special thank you to my husband, Roger Lee, for his

continuous patience, love and support, which have made

possible the realization of both a career and family life.

And a final thanks to Mom and Dad, who made this possible

and who have always been there.

The present studies were supported by Office of Naval

Research Grant No. N00014-88-J-1137, N.I.H. Grant No. EY04590,

and by March of Dimes Basic Research Grant No. 1-1039.
















TABLE OF CONTENTS


page

ACKNOWLEDGEMENTS ....................................... iii

ABSTRACT ............... ................... ................ vii

CHAPTERS

1 LITERATURE REVIEW................................. 1

Transduction in Rods.............................. 3
Transduction in Cones............................. 8
Chicken Retina Development and
Cyclic Nucleotides............................... 15
The Retinal Degenerate (rd) Chicken.............. 16
Abnormal cGMP Metabolism and
Retinal Degeneration............................. 23

2 LIGHT DECREASES cGMP LEVELS IN THE
CONE-DOMINANT CHICKEN RETINA....................... 26

Introduction...................................... 26
Methods............................................ 28
Effect of Freezing on Retinal cGMP and cAMP.... 29
Effect of Light Adaptation on cGMP and cAMP.... 30
Effect of Brief Light Pulse on cGMP and cAMP... 31
Results.......... ............................... 32
Effect of Freezing on Retinal cGMP and cAMP.... 32
Effect of Light Adaptation on cGMP and cAMP.... 34
Effect of Brief Light Pulse on cGMP and cAMP... 36
Effect of Ischemia on Dark-Adapted Levels
of Cyclic Nucleotides.......................... 36
Discussion ........................................ 37

3 LIGHT MODULATION AND DEVELOPMENTAL TIMECOURSE
OF CYCLIC NUCLEOTIDE LEVELS IN THE
rd CHICKEN RETINA ................................ 41

Introduction...................................... 41
Methods............................. ........... 44
Effect of Light Adaptation...................... 44
Development of Cyclic Nucleotide Levels........ 45
Extraretinal Tissues............................. 46










Results .......................................... 47
Effect of Light Adaptation..................... 51
Extraretinal Tissue Levels of cGMP and cAMP.... 53
Developmental Timecourse........................ 53
Central vs. Peripheral Retinal Cyclic
Nucleotide Levels............................ 58
Discussion......................................... 61

4 DISTRIBUTION OF cGMP AND cAMP IN THE
CONE-DOMINANT RETINAS OF SIGHTED AND
BLIND (rd) CHICKENS................................ 68

Introduction...................................... 68
Methods.......................... ................. 70
Results......................... .................. 72
Distribution of cGMP............................ 72
Distribution of cAMP............................ 77
Discussion....................................... 78
cGMP............................................. 79
cAMP............................................. 83

5 HISTOCHEMICAL LOCALIZATION AND KINETIC
PROPERTIES OF cGMP AND cAMP PHOSPHODIESTERASES
IN THE rd CHICKEN RETINA........................... 88

Introduction...................................... 88
Methods.......................................... 89
Histochemical Techniques....................... 90
Kinetic Analyses................................. 92
Results ......................... ................. 94
Histochemical Localization of cGMP PDE......... 94
Histochemical Localization of cAMP PDE......... 101
Kinetic Properties of cGMP PDE and cAMP PDE.... 105
Discussion.............................. ....... 109
Phosphodiesterase in Carrier Chick Retinas..... 109
Phosphodiesterase in rd Chick Retinas.......... 111

6 KINETIC ANALYSIS OF GUANYLATE CYCLASE ACTIVITY
IN THE rd CHICKEN RETINA ......................... 114

Introduction ..................................... 114
Methods......................... .................. 115
Guanylate Cyclase Assay......................... 115
Results.......................................... 120
Discussion...................................... 126
Speculation..................................... 134

7 OVERALL DISCUSSION................................. 138

REFERENCE LIST ......................................... 145

BIOGRAPHICAL SKETCH .................................... 160
















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

CYCLIC NUCLEOTIDES IN THE (rd) RETINAL DEGENERATE
CHICKEN RETINA


By

Nancy Ruth Lee

August 1991


Chairperson: William G. Luttge, Ph.D
Major Department: Neuroscience

This study investigated the light modulation, retinal

distribution and developmental timecourse of cGMP and cAMP

levels as well as the kinetic properties and histochemical

localizations of cyclic nucleotide metabolizing enzymes in the

cone-dominant chicken retina. Comparisons were made between

sighted chicks and congenitally-blind rd (retinal degenerate)

chickens, in which visual transduction is severely compromised

prior to the onset of photoreceptor pathology. In sighted

chicks, levels of cGMP and cGMP phosphodiesterase (PDE)

activity were concentrated within the photoreceptor cells and

were highest in the outer segment (OS) region. Levels of cAMP

and cAMP PDE activity were more evenly distributed between

outer and inner retina and were lowest in the OS region. Light

adaptation significantly reduced levels of both cGMP and cAMP


vii










in the proximal photoreceptor layers. In the rd retina,

defective transduction was found to be accompanied by a severe

and photoreceptor-specific deficiency in cGMP levels. rd

retinas failed to accumulate normal levels of cGMP during

photoreceptor outer segment development. On the day of hatch,

photoreceptor levels of cGMP were reduced 75% within the OS

region and were unresponsive to light adaptation. Levels of

cAMP in rd retinas were not severely affected in association

with blindness. Kinetic analysis of cGMP PDE activity in rd

retinas revealed a Vmax comparable to sighted chicks and a 28%

reduced apparent Km for cGMP, indicating that rd

photoreceptors may have an enhanced capacity for cGMP

hydrolysis. Kinetic analysis of guanylate cyclase activity in

rd retinas revealed a 78% increased Km for GTP and significant

inhibition of enzyme activity at GTP levels above

approximately 250 uM. The kinetic abnormalities apparent in

guanylate cyclase are sufficient to largely explain the 75%

deficiency in rd outer segment levels of cGMP. Our results

suggest that inadequate synthesis of photoreceptor cGMP levels

may underly defective transduction in the rd chick retina and

provide in vivo evidence supportive of a cGMP-mediated

transduction mechanism in cone photoreceptors.


viii















CHAPTER 1
LITERATURE REVIEW



Visual signalling begins with absorption of light by

photosensory receptors. Most vertebrate retinas possess two

types of photoreceptors, rod cells and cone cells. Rods are

extremely sensitive to light and mediate vision in dim light.

These cells contain a single visual pigment, rhodopsin, which

absorbs light over a broad spectral range, but maximally at

500 nm. Rod vision is therefore achromatic. Cones generally

require more light to operate than rods and mediate daylight

vision. They possess one of several visual pigments with

narrow yet overlapping absorption spectra. Cones are

responsible for color vision.

Rods and cones are highly specialized to capture light

and convert it into a neural signal which may ultimately be

transmitted to the brain. Phototransduction takes place in the

photoreceptor outer segment, an elongated specialization at

the cell's apical end. The outer segment is comprised of a

dense stack of 500-2000 flattened membranous disks in which

light-absorbing visual pigment molecules are embedded.

Surrounding the disks is a plasma membrane which contains

light-sensitive ion channels. In rods, the disk membranes are

physically separated from the outer plasma membrane with the

1










2

exception of a few basal disks. In cones, the disk membranes

are continuous with the plasma membrane throughout the length

of the outer segment. At the base of the outer segment is a

thin cilium which connects the outer segment to the inner

segment. Outer segment membranes are continuously renewed,

forming by evagination of the ciliary plasma membrane.

The inner segment contains the metabolic machinery of the

photoreceptor cell. Numerous mitochondria, densely packed in

the apical "ellipsoid" region of the inner segment, generate

the energy required for inner segment needs as well as for

transduction and membrane renewal processes in the outer

segment. A somatic region and connecting fiber joins the inner

segment to the cell's synaptic terminal, which possesses

specialized ribbon synapses for chemical communication with

retinal interneurons.

Unlike most other sensory receptors, vertebrate

photoreceptors hyperpolarize in response to light stimulation.

In the dark, the cation channels in the outer segment plasma

membrane are primarily in an open state. A dark current of Na*

ions flows into the outer segment, depolarizing the cell.

Light stimulation leads to the closure of the outer segment

cation channels which reduces the circulating current. The

cell hyperpolarizes and neurotransmitter release from the

synaptic terminal is subsequently decreased in graded response

to light.











Transduction in Rods


The molecular mechanism linking photon absorption to

closure of outer segment cation channels is tentatively

understood in rods. Extensive biochemical and

electrophysiological evidence indicates that 3'5' guanosine

monophosphate, cyclic (cGMP) plays a central role. First, rod

outer segments contain a uniquely high concentration of CGMP

in the dark (approximately 70 uM) (Goridis et al., 1974) and

possess high levels of the enzymes which synthesize (guanylate

cyclase) and hydrolyze (phosphodiesterase) it (Berger et al.,

1980). Second, light activates rod outer segment

phosphodiesterase (PDE) (Bitensky et al., 1981) and the

subsequent hydrolysis of cGMP is both fast enough and of

sufficient magnitude to account for the photoelectric response

of rods observed in vitro (reviewed in Pugh and Lamb, 1990).

Third, injection of cGMP into isolated rod outer segments

mimics the electrophysiological dark response of rods and

increases the latency of the light response (Miller and Nicol,

1979).

Recent electrophysiological evidence indicates that cGMP

directly binds to a component of the outer segment cation

channel, holding it open in a reversible and concentration-

dependent manner (Fesenko et al., 1985; Nakatani and Yau,

1985). When bound, cGMP holds the channels open, permitting

the dark current to flow into the outer segment and depolarize









4

the cell. The binding is cooperative, requiring 2-4 cGMP

molecules to keep one channel open.

The sequence of molecular events thought to link photon

absorption to cGMP-mediated closure of the outer segment ion

channels is depicted in Fig. 1.1 and is described briefly

below (reviewed in Stryer, 1986; Lamb, 1986; Pugh and Cobbs,

1986). In the dark, the high concentration of cGMP in the rod

outer segment maintains the opened state of the cGMP-regulated

ion channels and the cell is depolarized. Light catalyzes a

cascade of reactions which rapidly reduces the rod outer

segment concentration of cytosolic cGMP. Three disc membrane

proteins, rhodopsin, phosphodiesterase (PDE) and transducin

are major components of the cGMP cascade. First, absorption of

a photon activates rhodopsin via isomerization of the 11-cis

retinal chromophore. Photoexcited rhodopsin (*Rh) then

activates transducin, a G-protein composed of three subunits,

by catalyzing the exchange of GTP for GDP on the G-alpha

subunit. Multiple transducin molecules are triggered by each

photoexcited rhodopsin molecule, amplifying the signal

approximately 500 fold. Once charged with GTP, the G-alpha

subunit dissociates from the G-beta:gamma subunit complex and

activates PDE by removing the inhibitory gamma subunits of the

PDE protein. Each activated PDE molecule is capable of

hydrolyzing 1000 molecules of cGMP, providing a second

amplification step in rod transduction. The hydrolysis of cGMP










0 ) + rcM 44 + -4 ) n
0 r-. +. 0 -4 0U + U i. ,.
4J r > M-4 4 1 4 0 > B )
4) Ap 45 r. v0 q r
_u u o o r .A w

-4 1 -H 144 >4 4"-O4- -

4 00 .0 V + 0- i
c 4 J N 0 00 go u o c U O "

o 0




Ur t -" -- 0 0

to ..o1, ,.,j ,r f WA m 0i:

. A 8 .q V 3&
4 -14 0 b I -.. 4 -4 4J H mrl

A 4-- r 0 4 C O -4 4 0M' 1-
$4 j U U 4 04t r.ON1 m pi-
- (a- to .4 to 0 9 ; ,, 04 ,4j -


o.4 044 W
O4 0 X:04 >O 0 W 0V 4J


4 0o r. E0 J 'o 0. > 0 V






O 04 0 m4 0
m- iC -A 400 5 t


. 0, 0 A 0 0 >-4 0 O0





,0 014 ,,. O A v
0 v00 0) 4 U) (0

* ONm 0 0 0 W V


S0 OV44 0 W" 0,.,


to 4 0 0 -,4 W *0 a
-* 0 : to o 0 C" V -
W 4r-4 0 0A M -H a c OS04
































+










7
causes fewer channels to open, thus reducing the circulating

current and hyperpolarizing the cell.

Deactivation of the cGMP cascade occurs through a variety

of mechanisms, only a few of which are currently understood.

Photoexcited rhodopsin is deactivated by multiple ATP-

dependent phosphorylations via the enzyme rhodopsin kinase

(Miller and Dratz, 1984; Miller et al., 1986) as well as by

the inhibitory "capping" action of a 48 kD protein (arrestin)

(Miller et al., 1986). GTP-ase activity inherent in transducin

spontaneously hydrolyzes GTP to GDP, thereby deactivating PDE

and restoring transducin to its dark form (Stryer, 1983,

1986). PDE may be deactivated more quickly by the ATP-

dependent binding of arrestin at transducin binding sites on

PDE (Zuckerman and Cheasty, 1986). Dark levels of cGMP are

regenerated by the enzyme guanylate cyclase (GC), the activity

of which is stimulated by the reduced Ca++ concentration in

the rod outer segment that occurs following closure of the

light-sensitive channels (Pepe et al., 1986). Recent studies

indicate that stimulation of guanylate cyclase activity by low

levels of Ca"+ is mediated by a 26 kD Ca+-binding protein,

recovering (Dizhoor et al., 1991). Calcium may also modulate

the cGMP cascade by inhibiting PDE activity (Kawamura and

Bownds, 1981; Lolley and Racz, 1982) and by binding to the

extracellular surface of the light-sensitive channels (Stern

et al., 1986). Biochemical investigations of cyclic

nucleotides in rod-dominant retinas support a cGMP-mediated










8

transduction mechanism in rod photoreceptors; levels of cGMP

in rod-dominant retinas, both in vivo (Orr et al., 1976; De

Vries et al., 1978; de Azeredo et al., 1981), in vitro

(Kilbride and Ebrey, 1979), as well as in isolated rod outer

segments (Woodruff and Bownds, 1979), have been shown to be

reduced 5- to 10-fold following light exposure.


Transduction in Cones


The mechanism of transduction in cone photoreceptors is

not well understood. Emerging evidence suggests that cones

possess a cGMP-mediated transduction mechanism similar to that

in rods. First, cone outer segment membranes have been shown

(Haynes and Yau, 1985) to contain cGMP-sensitive conductances

with characteristics strikingly similar to those recorded from

rod outer segment membranes. Like the rod outer segment

conductance, the cone conductance appears to be directly gated

by cGMP with cGMP binding in a positively cooperative manner.

Neither cAMP nor Ca+* has significant effect on the cone outer

segment conductance.

Second, infusion of cGMP into isolated larvae salamander

cones produces a large persistent increase in the light-

sensitive current (Cobb et al., 1985). Like the rod response,

the cone current increases 10-fold in magnitude and duration

after cGMP infusion and demonstrates a 5- to 10-fold increase

in light-sensitivity. Under the same conditions cAMP has no

effect.











Finally, cone outer segments possess the major protein

components of the cGMP cascade. Immunologically-related but

distinct forms of both cGMP PDE (Hurwitz et al., 1985) and

alpha-transducin (Lerea et al., 1986) are expressed in rod and

cone outer segments. Also, the photopigments of rods and

cones, although structurally different, possess significant

homology in the domains thought to interact with transducin

(Nathans et al., 1986). It thus appears that, although the

components of the visual cascade are somewhat different in

rods and cones, the two vertebrate photoreceptor types may

operate in essentially the same way (Pugh and Cobbs, 1986).

In light of the evidence supporting a cGMP-mediated

transduction mechanism in cones, it is puzzling that

biochemical assays have failed to detect light-induced

decreases in cGMP levels in the cone-dominant retina (De Vries

et al., 1979; Farber et al., 1981, 1983). Perhaps even more

perplexing is that some investigators have observed large,

light-sensitive levels of cAMP in the cone dominant retinas of

the ground squirrel and western fence lizard, suggesting the

involvement of cAMP in cone transduction (Farber et al., 1981,

1982, 1983). Other investigators (Berger et al., 1980; De

Vries et al., 1979), however, have reported that

concentrations of cGMP and cAMP in the ground squirrel retina

are approximately equal and that neither cAMP nor cGMP levels

in this retina change in response to light exposure.












The reports also disagree as to the retinal distributions

of cAMP and cGMP in the cone-dominant retina. Evidence

obtained from freshly-dissected ground squirrel retinas in

which the photoreceptors were selectively destroyed suggests

that cone photoreceptors possess large levels of cAMP (Farber

et al., 1981, 1983). Analyses of individual cell layers of

freeze-dried ground squirrel retinas (De Vries et al., 1979),

however, suggest a retinal distribution of cyclic nucleotides

similar to that of rod-dominant species: cGMP levels (as well

as GC and PDE activities) highly-concentrated in the

photoreceptor layers, especially the outer segment region, and

cAMP levels more diffusely distributed with minimal amounts in

the photoreceptor outer segments.

These few and controversial reports concerning the

concentration and light modulation of cyclic nucleotides in

the cone-dominant retina may be confounded by the different

sensitivities of retinal cGMP and cAMP to various experimental

factors such as ischemia and freezing. Moreover, a lack of

consideration for certain morphological and physiological

properties of cone-dominant retinas may have also contributed

to the discrepant results. Factors known and proposed to

influence cyclic nucleotides in the cone dominant retina are

discussed individually below.

Ischemia. While the effect of ischemia on cyclic

nucleotide levels in the cone-dominant retina has not been

reported, studies performed on rod-dominant retinas (Orr et











al., 1976; Mitzel et al., 1978) and the brain (Steiner et al.,

1972; Lenox et al., 1982; Jones and Stavinoha, 1979; Lust and

Passonneau, 1979; Nemoto, 1982) reveal that even brief periods

(seconds) of ischemia significantly elevate cAMP levels and

depress cGMP levels. As in the brain, the effects of ischemia

in the retina are localized (Orr et al., 1976). cGMP levels

decrease selectively in photoreceptors while cAMP levels

increase primarily in the inner retinal layers. Moreover,

dark-adapted retinas are much more sensitive to ischemia than

light-adapted retinas (Orr et al., 1976).

In the published reports of cyclic nucleotides in cone-

dominant retinas (De Vries et al., 1979, 1982b; Farber et al.,

1981, 1982, 1983), the minimum time from decapitation to

halting of cyclic nucleotide metabolism was on the order of 1

minute, a delay sufficient to significantly alter both cAMP

and cGMP levels in the brain (Lust and Passonneau, 1979; Lenox

et al., 1982). It is possible that cyclic nucleotide levels in

cones respond more quickly to or are affected more by ischemia

than those in rods and that ischemia-induced decreases in cGMP

have prevented the detection of light-induced decreases in

cGMP in the cone-dominant retina. Ischemia-induced increases

in cAMP may also explain the observed light-induced decreases

in cAMP in certain cone-dominant retinas.

It is also possible that the observed light-induced

decreases in cAMP are associated with synaptic transmission of

neurons in the inner layers of the retina. The retinas of rod-











dominant (rat, mouse, cat, calf) and cone-dominant (chick,

ground squirrel) species possess dopamine-sensitive adenylate

cyclase (AC) activity in the inner plexiform layer and

photoreceptor inner segment layer (De Vries et al., 1982a, de

Mello, 1978; Lolley et al., 1974). In the rod-dominant retina

(Orr et al., 1976) light exposure causes significant decreases

in cAMP levels which are restricted to the layers possessing

AC activity.

Freezing. Inactivation of retinal metabolism by freezing

has been reported to depress cAMP levels in the cone-dominant

retina (Farber et al., 1981). In the ground squirrel, cAMP

levels in retinas inactivated by freezing were significantly

reduced compared to cAMP levels in unfrozen retinas

inactivated by acid extraction. A further reduction in cAMP

was observed upon exposure of dark-adapted frozen retinas to

30 seconds of laboratory illumination, indicating that

freezing does not completely inactivate light-sensitive cAMP

metabolism. Subsequent freeze-drying reduced cAMP levels even

further. These effects were selective for cAMP levels (cGMP

levels were not effected by freezing or freeze-drying),

suggesting that the reason other studies failed to detect

light-induced decreases in cAMP in the ground squirrel retina

was due to freezing and freeze-drying of the retinas. The

freezing effect appears to be unique to cone-dominant retinas

(or perhaps only to the ground squirrel retina) since large

decreases in cAMP levels are observed in the inner retinal










13
layers of freeze-dried rod-dominant retinas following exposure

to light (Orr et al., 1976). Since the techniques of freezing

and freeze-drying are widely-used for rapid fixation and

microdissection of retina, it is important to determine if the

decreases in cAMP observed in ground squirrel occur in other

cone-dominant animals. It is possible that the reported

results reflect different exposure times of fresh and frozen

retinas to ischemia, loss of outer segments, or a ground

squirrel-specific phenomenon.

Morphological and physiological considerations. The

different morphological and physiological properties of rods

and cones may also affect the assessment of cyclic nucleotide

concentration and light regulation in the cone-dominant

retina. For example, in the retinas of many cone-dominant

species, including the ground squirrel (De Vries et al.,

1979), chicken, and chameleon, the photoreceptor outer

segments tightly interdigitate between the villous processes

of the overlaying pigment epithelium. Rapid dissection of

these cone-dominant retinas results in the substantial loss of

cone outer segments and thus their cyclic nucleotides. Loss of

outer segments may explain the failure of studies using

freshly-dissected retinas to detect light-induced decreases in

cGMP.

Rods and cones also respond differently to light and dark

adaptation. Each receptor type produces a sustained

hyperpolarization in the presence of background illumination.












In cones, however, the hyperpolarization rapidly decays from

its peak value and is sustained at a less polarized level

(Normann and Werblin, 1974). Thus, it is possible that cone-

dominant retinas which have been light-adapted or dark-adapted

for a long period may not exhibit significant changes in cGMP

content. A more effective stimulus to examine light-induced

changes in cone cyclic nucleotides levels is, perhaps, a brief

flash of light following dark adaptation.

Since electrophysiological studies have shown that the

light-sensitive conductance of cone outer segments is directly

regulated by the concentration of cGMP (Haynes and Yau, 1985;

Cobbs et al., 1985), it seems reasonable that light-induced

decreases in cGMP may be biochemically detected in the cone-

dominant retina if the light stimulus is appropriate and if

factors such as ischemia and freezing are controlled for. The

first aim of the proposed research is to determine the

concentration and distribution of cGMP and cAMP in the cone-

dominant chicken retina and to determine whether these

nucleotide pools are sensitive to light. Retinas will be

rapidly fixed to minimize possible ischemia-induced changes in

cyclic nucleotide levels. The results of these studies will

provide information concerning cyclic nucleotides in a

relatively nonischemic cone-dominant retina of a species that

has not previously been explored in detail.











Chicken Retina Development and Cyclic Nucleotides


Cone photoreceptors predominate in the chicken retina,

outnumbering rods 6 to 1 in the central retina and 3 to 1 in

the peripheral retina (Morris and Shorey, 1967). Cones are

easily distinguished from rods by the presence of oil droplets

in the apical ends of cone inner segments. Rod outer segments

are also larger in diameter (about 2.0 um) than those of cones

(1.5 um) and this difference is greater in peripheral retina,

where the rod outer segments are longer and wider (Young,

1978).

Development and differentiation of the chick retina is

similar to that of other vertebrate retinas. The retina arises

from the neural ectoderm of the ventrolateral medullary plate.

Ganglion cells start differentiating 2-4 days post-

fertilization (pf), followed by the photoreceptor cells at

approximately day 7 pf (Meller and Tetzlaff, 1976; Coloumbre,

1955). Inner segments begin forming on day 10 pf (Olson, 1979)

and outer segments are first visible between day 15 and 16 pf

(Meller and Tetzlaff, 1976). An adult-like

electroretinographic (ERG) response may be elicited from the

developing retina on day 17 pf (Rager, 1979). By embryonic day

19-20, the chick retina is fully differentiated and

functionally complete (Olson, 1979; Coloumbre, 1955).

Information concerning cyclic nucleotide metabolism in

the chick retina is primarily restricted to developmental

studies performed on light-adapted animals (de Mello, 1978;











Chader et al., 1974b). The retinal concentration of cAMP in

day 7 pf embryos is 7 pmole/mg protein. Between days 16 and 18

pf, cAMP levels rise sharply to 21 pmole/mg protein. The

concentration of cGMP between days 8 and 15 pf is much lower

(0.5 pmole/mg protein), but shows an abrupt increase of up to

14 pmole/mg protein at hatching. The sharp rise in cAMP and

cGMP levels after day 15 pf coincides with a marked increase

in cGMP PDE activity (Chader et al., 1974b) and with the first

appearance of outer segments in the photoreceptor cells. This

is also the time of synapse formation between amacrine cells

and ganglion cells (Coloumbre, 1955), both of which possess

dopamine-dependent adenylate cyclase activity (de Mello, 1978;

Schwarz and Coyle, 1976). A dopamine-dependent increase in

cAMP can be elicited as early as day 7 pf and is maintained

throughout retinal development.


The Retinal Degenerate (rd) Chicken


Rods and cones are highly specialized to convert light

into a neural response, and each type of photoreceptor cell

possesses a unique set of protein and enzyme systems to carry

out this function. A genetic defect in any component of the

transduction mechanism (its activation, inactivation or

modulation) may therefore specifically affect photoreceptor

cell function without affecting other cell types in the

retina.












Our laboratory is currently investigating a strain of

chicken which appears to possess a genetic defect in

transduction. The defective gene(s) results in blindness in

the chicken at the time of hatch followed by complete

degeneration of photoreceptor cells by six months posthatch.

This retinal degenerate (rd) strain of Rhode Island Red

chicken, reported in 1980 by Cheng et al., is believed to have

developed in response to the chemical mutagen, ethyl methane

sulfonate. The visual defect is inherited as a simple

recessive; matings between rd and genetically-normal chickens

produce heterozygotes (carriers) which are sighted and retain

normal retinal morphology throughout life.

On the day of hatch, rd chicks are behaviorally and

electrophysiologically blind (Cheng et al., 1980; Wolf, 1982;

Ulshafer et al., 1984). The chicks fail to peck at small dark

targets and to avoid visually-threatening stimuli as do

normally-sighted hatchlings. The rd chick retina also fails to

exhibit normal electrical responses to light at this time.

Under both scotopic and photopic conditions, all components of

the ERG are unmeasurable, indicating that blindness occurs at

the levels of the photoreceptor cells and that transduction is

defective in both rod and cone photoreceptor cell types

(Ulshafer et al., 1984).

Metabolically, the 1 day rd chick retina is also abnormal

(Fig. 1.2). Dark-adapted rd chick retinas utilize

significantly less glucose than those of dark-adapted sighted























Figure 1.2. Effect of light adaptation on 3H 2-deoxy-D-[l-
3H]glucose (3H-2DG) uptake in retinas of 1-day posthatch
carrier and rd chicks. Values represent the mean SEM of 5-6
retinas per group. Chicks were either dark-adapted (total
darkness) or light-adapted (approximately 1600 lux) for 2
hours, and were then injected intraperitoneally with 2
uCi/gram body weight 3H-2DG (specific activity 25 Ci/mMole,
Amersham). Thirty minutes post-injection, chicks were
sacrificed by decapitation and enucleated under the
appropriate lighting condition. Retinas were isolated,
solubilized in lN NaOH and aliquots measured for radioactivity
in a scintillation counter (LKB Instruments) and for protein
content (method of Lowry et al., 1951). Glucose utilization is
expressed as total dpm per ug retinal protein. Thin layer
chromatography (data not shown) verified for all groups that
approximately 80% of retinal 3H-2DG had been phosphorylated
into 3H-2-deoxyglucose-6-phosphate, indicative of cellular
glucose utilization.




























M Dark
W Light


Figure 1.2


I *


30--

0

c

a20Q
20
'-D-


10 1


KL _


Carrier









20

chicks, with usage levels comparable to those found in sighted

retinas in the light (Ruth et al., 1985). The rd retina

behaves metabolically as if it is always in the light,

suggesting that the photoreceptor cation current is

permanently shut off in this retina, possibly due to

insufficient energy to drive the Na'-K ATPase pump in the

inner segment.

These behavioral, electrophysiological and metabolic

defects are present in the rd chick retina before any

morphological abnormality is apparent. At the time of

hatching, the retina of the rd chick appears morphologically

normal. Light and electron microscopy studies (Ulshafer et

al., 1984; Ulshafer and Allen, 1985a, 1985b) show that all

retinal cell types and layers are present in their normal

positions, numbers and densities. The pigment epithelium,

which is the blood-retinal barrier, and the choroid are

similar in appearance to those of sighted hatchmates.

Initial signs of retinal degeneration appear 7-10 days

posthatching (ph). At this time, degenerative changes are

primarily restricted to the photoreceptor cells; rod and cone

outer segments become swollen and disorganized and associated

inner segments become distended at their tips. As the

photoreceptors degenerate, the number of outer segments

decreases, large spaces develop between photoreceptor inner

segments and pycnotic nuclei appear in the nuclear layer

(Ulshafer and Allen, 1985a).









21

The pathological changes in the photoreceptor layers

proceed swiftly from central to peripheral regions of the

retina. By 8 weeks ph, very few outer segments remain and the

inner segments are in various stages of degeneration (Ulshafer

et al., 1984). Both rods and cones are susceptible to

degeneration with the possible exception of the double cone

cell complex which appears somewhat more resistant to cell

disruption. By 6 months ph, virtually all visual cells are

gone, the inner retinal layers appear thinned and pathological

changes are present in the pigment epithelium (Ulshafer and

Allen, 1985b).

The rd chick retina does not appear to possess a primary

defect in pigment epithelium phagocytosis. Phagosomes and

large outer segment inclusions are frequently observed in the

cytoplasm of the rd pigment epithelium cells (Ulshafer and

Allen, 1985b) and cultured rd chick retinal pigment epithelium

cells phagocytize latex beads as well as dissociated outer

segments from sighted and rd chick retinas (Ulshafer et al.,

1985). It is also unlikely that blindness and retinal

degeneration in the rd chick result from a defect in the

vitamin A cycle since the quantities of rhodopsin, 11-cis

retinyl esters, 11-cis protein retinol and interphotoreceptor

retinoid-binding protein (IRBP) are comparable in 3 day ph rd

and normal-sighted chicks (Bridges et al., 1987).

Recent studies indicate that the cone photopigment may be

abnormal in the rd chick. Early receptor potentials (ERP),










22

non-ionic electrical responses thought to be produced by

conformational changes in the cone pigment molecule during

light activation (Pak, 1968), are either absent or severely

reduced in 1-3 day ph rd chicks (Dawson et al., 1990). Also,

the labelling of rd cones by antibodies that recognize a cone

outer segment-specific protein (thought to be a component of

the cone visual pigment) is significantly reduced both in the

number of outer segments labelled and in the density of label

(Ulshafer et al., 1990a). These observations indicate that

cone photopigment in the rd chick may be absent, structurally

abnormal or positioned incorrectly within the outer segment

membrane. An inability of photopigments to trigger visual

transduction may explain lack of vision in rd chick cone

photoreceptors. Visual pigment in rd rod photoreceptors,

however, appears to be normal. Rhodopsin-specific antibodies

label comparable numbers of rods at comparable densities in rd

and carrier retinas (Ulshafer et al., 1990a) and the amount of

spectrophotometrically-detectable rhodopsin in homogenates of

rd retinas is comparable to that in carrier retinas at 1-3

days ph (Bridges et al., 1987). Defective rhodopsin therefore

seems unlikely to explain lack of vision in rd rod

photoreceptors.

Since transduction appears to be mediated by cGMP

hydrolysis in both rods and cones, it seems reasonable that a

defect in the photopigment or in any component of the visual

cascade is likely to manifest itself as abnormal cGMP levels,









23

especially in response to light stimulation. The second aim of

the proposed research is to determine if blindness and

subsequent photoreceptor degeneration in the rd chick retina

are associated with abnormal cyclic nucleotide levels and to

examine the enzymatic basis of the abnormality. It is proposed

that alterations in cyclic nucleotide metabolism accompany

blindness in the rd chick retina and that this defective

metabolic state perhaps leads to subsequent photoreceptor cell

degeneration. Since blindness is apparent a week or more

before the onset of pathology in the rd chick retina, it is

possible to study the transduction defect in this mutant

independent of confounding morphological degeneration.


Abnormal cGMP Metabolism and Retinal Deqeneration


Disorders of cGMP metabolism are currently implicated as

the possible cause of photoreceptor degeneration and

subsequent blindness in three other mutants possessing

hereditary retinal degeneration; the rd mouse (Lolley and

Farber, 1976), Irish setter (Aguirre et al., 1982) and collie

(Woodford et al., 1982). In these mutants, abnormally-high

levels of cGMP are apparent in the retina during inner segment

and outer segment development. The cGMP accumulation is

associated with deficient PDE activity (Schmidt and Lolley,

1973; Farber and Lolley, 1976) and occurs prior to or during

the early stages of photoreceptor degeneration. The deficiency

in PDE activity appears to be expressed differently in each










24
animal (Lolley and Lee, 1984; Lee et al., 1985). Recent

molecular analysis of cDNA clones indicates that the genetic

defect in the rd mouse resides within the beta subunit of cGMP

PDE (Bowes et al., 1990).

PDE activity is also depressed in the rds mouse (Cohen,

1983), a mutant in which photoreceptor outer segments fail to

develop and retinal degeneration proceeds slowly. In this

retina, however, cGMP levels are abnormally low (Sanyal et

al., 1984) despite the presence of normal-appearing connecting

cilia, a site associated with guanylate cyclase activity in

the rodent retina (Fleischman et al., 1980).

It is thought that the abnormally-low PDE activity or the

associated elevations in cGMP levels may be responsible in

some way for photoreceptor degeneration in these mutants.

Incubation of genetically-normal Xenopus Laevis eye rudiments

(Lolley et al., 1977) and human retinas (Ulshafer et al.,

1980) with various PDE inhibitors or high concentrations of

dibutyryl cGMP produces photoreceptor-specific degeneration

similar to that seen in vivo, with minimal effect on inner

retinal layers. The PDE inhibitors and cyclic nucleotide

analogues (dibutyryl cGMP as well as dibutyryl cAMP) inhibit

retinal protein synthesis in vitro, prior to changes in

retinal morphology (Ulshafer and Hollyfield, 1982). Rod

receptors are most affected.

Cone-dominant retinas also appear vulnerable to abnormal

cyclic nucleotide metabolism. Incubations of all-cone lizard









25

eyecups with the PDE inhibitor isobutylmethylxanthine (IMBX)

or high concentrations of cAMP or cGMP analogues produces

selective degeneration of cone cells and inhibits retinal

protein synthesis, with cAMP apparently more toxic to cones

than cGMP (Williams et al., 1985, 1987). It is thus possible

that abnormal PDE activity and/or cyclic nucleotide levels may

induce photoreceptor-specific degeneration in some inherited

retinal disorders by inhibition of protein synthesis.

Unlike most retinal degenerate species currently under

study, the rd chick possesses a cone-dominant retina. Thus,

examination of cyclic nucleotide levels and associated enzyme

activities in the rd chick retina will provide information

concerning potential abnormalities in cyclic nucleotide

metabolism associated with cone-dominant retinal degeneration.
















CHAPTER 2
LIGHT DECREASES cGMP LEVELS IN THE
CONE-DOMINANT CHICKEN RETINA


Introduction


Extensive research has established that visual

transduction in vertebrate rod photoreceptors is mediated by

light-stimulated decreases in outer segment levels of cGMP. In

darkness, high levels of cGMP in the rod outer segment

directly maintain the open state of cation channels in the

outer segment plasma membrane (Fesenko et al., 1985; Nakatani

and Yau, 1985). This allows a depolarizing current to flow.

Light stimulation initiates the sequential activation of

rhodopsin, transducin, and cGMP phosphodiesterase (reviewed in

Stryer, 1986), which rapidly reduces outer segment levels of

cGMP. This results in membrane channel closure and subsequent

photoreceptor hyperpolarization. Experimental evidence in

support of this mechanism is the demonstration of significant

light-induced decreases in cGMP levels in rod-dominant retinas

in vivo (Orr et al., 1976) and in vitro (Kilbride and Ebrey,

1979), as well as in isolated rod outer segments (Woodruff and

Bownds, 1979).

Visual transduction in cone photoreceptors is also

believed to be mediated by light-induced decreases in levels









27

of cGMP. Like rods, cones have been shown to possess a light-

sensitive conductance directly gated by cGMP concentration

(Haynes and Yau, 1985; Cobbs et al., 1985) and to have all the

major protein components thought to mediate the cGMP cascade

(Hurwitz et al., 1985; Lerea et al., 1986; Nathans et al.,

1986). Light-induced decreases in cGMP levels, however, have

not yet been detected in a cone-dominant retina (De Vries et

al., 1979; Farber et al., 1981, 1982, 1983). Rather, light-

stimulated decreases in cAMP levels have been observed in

cone-dominant retinas under certain conditions [i.e., in

freshly-dissected retinas (Farber et al., 1981, 1982, 1983)

but not in frozen or freeze-dried retinas (De Vries et al.,

1979; Farber et al., 1981)]. Based on their observations,

Farber et al. (1980) suggested that cAMP rather than cGMP may

mediate visual excitation in cones.

These few and controversial reports concerning cyclic

nucleotides in cone-dominant retinas have been limited largely

to one species (the ground squirrel) and may potentially be

confounded by factors known and proposed to influence cyclic

nucleotide levels in the retina such as ischemia (Orr et al.,

1976; De Vries et al., 1982b) and freezing (Farber et al.,

1981).

In the present study, we examined retinal cyclic

nucleotide levels in a different cone-dominant species, the

chicken, which possesses a cone:rod ratio of 6 to 1 in the

central retina (Morris and Shorey, 1967). cGMP and cAMP levels










28
were measured in dark-adapted central retinal punches and were

compared to those exposed to either long-term light adaptation

or to a 5 sec pulse of light. In addition, we examined the

effects of freezing and ischemia (2-3 min) on cyclic

nucleotide levels in this retina.


Methods


Animals. White Leghorn chicks, 1-2 days posthatch, were

obtained from the IFAS poultry science unit at the University

of Florida. Rhode Island Red chicks, 1-2 days posthatch, were

reared from eggs produced by controlled matings between

homozygous rd/rd hens and homozygous, genetically-normally

roosters maintained at the University of Florida Animal Care

Facility. Eggs were incubated in an automatic rotating forced-

draft incubator until embryonic day 19 (E19) and then

transferred to a brooder for hatching. These "carrier" chicks,

which are heterozygous for the rd mutation described by

Ulshafer et al. (1984), retain vision and normal retinal

morphology throughout life (Ulshafer et al., 1984) and their

eyes elicit early retinal potentials similar to those of

genetically-normal chick eyes (Dawson et al., 1990). Vision

was verified behaviorally in carrier hatchlings by the pecking

of visual targets.











Effect of Freezing on Retinal cAMP and cGMP


Rhode Island Red carrier chicks (rd/+) were light-adapted

under cool white fluorescent light (illumination level

approximately 1500 lux) or dark-adapted in total darkness for

2-3 hr, after which time they were sacrificed by decapitation

and enucleated in either room light or dim red light,

respectively. Following the removal of the anterior segment

and vitreous, the vascular pecten area was excised with a

razor blade and the entire retina-pigment epithelium-choroid

(RPEC) complex was isolated using forceps. The pigment

epithelium-choroid unit was included with the retina to insure

the acquisition of photoreceptor outer segments, which, in the

chicken and other cone-dominant retinas (De Vries et al.,

1979), are deeply embedded in the overlying pigment

epithelium. The RPEC complex from one eye was frozen in liquid

nitrogen, while, at the same time, the contralateral RPEC

complex was sonicated (20-30 sec at 70% output power,

Ultrasonic Disruptor, Heat Systems Inc.) in 400 ul ice cold

acid (1.0 N HC1 or 10% TCA). Frozen complexes were

subsequently sonicated under the appropriate lighting

condition. The time from decapitation to freezing or

sonication (designated as time of ischemia) was measured with

a stopwatch. The ischemic times for frozen and nonfrozen RPEC

complexes within a pair were identical and, for all pairs,

averaged 130 + 18 sec (n=9) in the dark and 104 + 12 sec (n=9)

in the light.












Effect of Light Adaptation on cGMP and cAMP Levels


White Leghorn chicks were either light-adapted under

bright cool white fluorescent light (illumination

approximately 3408 lux) or dark-adapted in total darkness for

2 hr, after which time, under the same lighting conditions,

they were sacrificed by decapitation and the heads immediately

frozen in liquid nitrogen. Heads were transferred to a -35C

cryostat in room light, bissected and mounted cornea-side up

onto a dissection chuck without thawing. Using a prechilled

scalpel, the eyelid and anterior segment were removed and the

vitreous was scraped from the eyecup. A cold trephine was then

used to remove a 4 mm punch of RPEC from the posterior eye

pole (central retina). Punches were stored in liquid nitrogen

until sonication (30 sec, 70% output power) in 300 ul cold 10%

TCA.

The rate of freezing of central retina upon immersion of

decapitated heads into liquid nitrogen was measured in

additional chicks using a thermistor probe-ohmeter assembly.

The posterior eye pole reached -10C within 14.0 + 2.9 sec

(n=7) after immersion into liquid nitrogen, cooling at an

average rate of 3.6 + 0.7 OC/sec. Since the time from

decapitation to immersion of heads into liquid nitrogen was 1

sec or less, central retinal metabolism was effectively

inactivated in less than 15 sec after sacrifice. This short










31

ischemic interval enabled us to measure relatively nonischemic

levels of cAMP and cGMP in central RPEC complexes.


Effect of Brief Light Pulse on cGMP and cAMP Levels


Chicks were dark-adapted in total darkness for 2 hr,

then, under infared illumination (FJW Find-R-Scope),

sacrificed by decapitation and enucleated. The anterior

segment and vitreous were removed from each eye. One eyecup

was then placed under a fiber optic light (MKII, Nikon Inc.)

and exposed to 5 sec high intensity illumination

(approximately 13566 lux) while the other eyecup was

maintained in total darkness. Both eyecups were then

immediately and simultaneously frozen in liquid nitrogen while

being maintained under their respective lighting conditions.

The time between decapitation and immersion of eyecups into

liquid nitrogen was identical for eyes within each pair and

averaged 144 + 30 sec (n=7) for all pairs. Trephine punches of

central RPEC (4 mm) were subsequently isolated at -350C as

described above and sonicated (approx. 30 sec, 70% output

power) in 300 ul ice cold 10% TCA.

Cyclic nucleotide assays. Homogenates in 1 N HCl were

heated to 100C for 3 min to insure protein denaturation,

centrifuged (13,600 xG) for 10 min at 4C and the supernatants

neutralized with 1.0 N NaOH. Homogenates in 10% TCA were

centrifuged and the supernatants extracted with (4 x 5 vol)

water-saturated ether. All supernatants were then diluted with










32

0.05 N Na acetate buffer (pH 6.2) and assayed for cGMP and

cAMP content by radioimmunoassay (Steiner et al., 1972) using

commercial kits (Biomedical Technologies, Inc.). Duplicate or

triplicate determinations were performed on 1-2 dilutions of

acetylated samples. Proteins were measured in the solubilized

pellets using the method of Lowry et al. (1951) with BSA

standard.

Assay specificity was established by incubating some

samples and standards with 50 ug/ml commercial cyclic

nucleotide phosphodiesterase, PDE, (Sigma Chemicals, St.

Louis). Assessed levels of both cGMP and cAMP were below

detection after 1 hr of PDE hydrolysis (see Chapter 3, Fig.

3.1). Additional samples were purified prior to assay on

Biorad AG1-X8 format columns (200-400 mesh), using 1.0 N

formic acid to elute cAMP and 4.0 N formic acid to elute cGMP

(Farber and Lolley, 1982). Both cAMP and cGMP levels were very

similar in chromatographically-purified and nonpurified

retinal extracts, ruling out sample interference (see Chapter

3, Fig. 3.2). Statistical significance between experimental

groups was determined using the Student's t-test with values

of p<.05 considered significant.


Results


Effect of Freezing on Retinal cGMP and cAMP


Freezing was found to have no effect on RPEC complex

levels of cGMP or cAMP in the chicken (Table 2-1). Under both
















Table 2-1. Effect of Freezing on Cyclic Nucleotide Levels
in the Retina-PE-Choroid Complex of the Chicken


cGMP


NONFROZEN


FROZEN


Differ.


p
value


DARK 10.20 + 1.32+ 10.16 + 2.11 NSD p=.47

LIGHT 7.37 + 1.20 6.93 + 1.23 NSD p=.46

Difference 28% 32%

p value *p<.001 *p=.001



cAMP


NONFROZEN FROZEN Differ. p
value


DARK 10.77 + 2.96 12.76 + 2.98 NSD p=.17

LIGHT 11.33 + 1.99 11.86 + 2.98 NSD p=.64

Difference NSD NSD

p value p=.64 p=.54

+ For each group, values are expressed as pmole cyclic
nucleotide per mg RPEC protein and represent the mean + SD of
9 eyes from 1-day posthatch Rhode Island Red chicks
heterozygous for the rd mutation. Dark values significantly
higher than light values, using the student's t test.










34

light-adapted and dark-adapted conditions, levels of each

cyclic nucleotide were comparable in frozen and nonfrozen RPEC

complexes. In contrast to the report of freeze-induced

decreases in cAMP in ground squirrel (Farber et al., 1981), we

found that freezing tended to slightly increase retinal levels

of cAMP in dark-adapted chickens. This difference, however,

was not statistically significant.

Comparison of cyclic nucleotide levels in dark-adapted

and light-adapted RPEC complexes demonstrates that light

adaptation significantly reduced levels of cGMP. In both

nonfrozen and frozen preparations, levels of cGMP were

approximately 30% lower in light-adapted RPEC complexes

compared to dark-adapted complexes. Levels of cAMP, on the

other hand, showed no significant light-dark difference in

either nonfrozen or frozen RPEC complexes.


Effect of Light Adaptation on cGMP and cAMP Levels


Central retinal RPEC complexes of White Leghorn chicks

isolated under relatively nonischemic conditions possessed

approximately equal levels of cGMP and cAMP in the dark (12.77

+ 1.95 pmole/mg protein and 13.64 + 1.18 pmole/mg protein,

respectively) (Fig. 2.1). Light adaptation significantly

decreased central RPEC levels of cGMP by 22% (p<.001). Light

adaptation had no effect on central RPEC levels of cAMP.











SLDARK
EI LIGHT


15+


10+


I


cGMP


cAMP


Figure 2.1. Effect of light adaptation on cyclic nucleotide levels in the central
retina-PE-choroid (RPEC) complex of 1-day posthatch White Leghorn chicks.
Values represent the mean + SD of 7-10 eyes per group. For all eyes, the
ischemic duration was approximately 15 sec. *p<.O01


DARK
-t T-- LIGHT


10+


cGMP


cAMP


Figure 2.2. Effect of 5 sec 'iqht stimulation on cyclic nucleotide levels in the
central retina-PE-choroid (RPEC) complex of 1-day posthatch White Leghorn
chicks. Values represent the mean + SD of 6-7 eyes per group. The ischemia
duration for all eyes averaged 144 + 30 sec. *p<.O01


O0


0 i












Effect of Brief Light Pulse on cGMP and cAMP Levels


Figure 2.2 shows that a 5 sec exposure to an illumination

level of approximately 13566 lux reduced dark-adapted levels

of cGMP by 36% in central RPEC complexes of White Leghorn

chicks. Levels of cAMP, however, were not altered by this

brief intense illumination.


Effect of Ischemia on Dark-Adapted Levels
of Cyclic Nucleotides


The ischemic duration of RPEC complexes employed in the

brief light study shown in Fig. 2.2 (144 + 30 sec) was

significantly longer than that of RPEC complexes employed in

the light adaptation study shown in Fig. 2.1 (< 15 sec).

Comparison between cyclic nucleotide levels in dark-adapted

RPEC complexes of Fig. 2.1 and Fig. 2.2 thus provided

information concerning the effect of a 2-3 min ischemic

interval on retinal levels of cGMP and cAMP. Levels of cGMP in

dark-adapted RPEC complexes exposed to approximately 144 sec

of ischemia (12.63 + 0.56 pmoles/mg RPEC protein) were found

to be very similar to the levels measured in dark-adapted RPEC

complexes exposed to approximately 15 sec of ischemia (12.77

+ 1.95 pmoles/mg RPEC protein). However, levels of cAMP in

dark-adapted RPEC complexes exposed to 144 sec of ischemia

(17.73 2.24 pmoles/mg RPEC protein) were found to be

significantly higher (p=0.001) than those measured in dark-










37

adapted RPEC complexes exposed to 15 sec of ischemia (13.64 +

1.18 pmoles/mg RPEC protein).


Discussion


The results of this study demonstrate for the first time

significant light-induced decreases in cGMP levels in a cone-

dominant retina. In the central retinal complex of the

chicken, where approximately 86% of the photoreceptors are

cones (Morris and Shorey, 1967), both light adaptation in vivo

and brief light stimulation in vitro significantly reduced

dark-adapted levels of cGMP. Under these same conditions,

light had no effect on retinal levels of cAMP.

In our studies, we included the PE-choroid complex with

the retina to insure the procurement of photoreceptor outer

segments. Existing evidence argues that retina-PE-choroid

levels of cGMP reflect photoreceptor levels of cGMP in the

chicken. First, levels of cGMP in chicken retina dramatically

and monotonically increase as photoreceptor outer segments

develop (de Mello, 1978; Lee et al., 1990; Chapter 3). Second,

analysis of the distribution of cyclic nucleotide levels in

the central retina-PE-choroid complex of the chick (Chapter 4)

showed that approximately 80% of the total retinal cGMP

content is located within the photoreceptor cells. It is

therefore likely that the light-associated decreases in cGMP

levels in the chicken RPEC complex are associated with the

photoreceptors. Unpublished studies in our laboratory (Chapter










38

3) have also shown that levels of cGMP in central retinal RPEC

complexes of the chick (where cones outnumber rods 6 to 1) are

similar to those measured in peripheral retinal RPEC complexes

(where the cone:rod ratio is 3:1). These findings indicate

that light-sensitive levels of cGMP are present in both cone

and rod photoreceptor cell types. Our findings add to the

growing evidence that cGMP mediates visual transduction in

cone as well as rod photoreceptor cells.

Unlike the reports of freeze-induced decreases in cAMP

levels in cone-dominant ground squirrel retinas (Farber et

al., 1981), our results show that freezing has no effect on

either cAMP or cGMP levels in the cone-dominant chicken

retina. Ischemia, however, does appear to increase levels of

cAMP in the central retina of the chick. Significant ischemia-

induced increases in cAMP levels have been demonstrated in

rod-dominant retinas (Orr et al., 1976; Mitzel et al., 1978)

as well as in whole ground squirrel eyes (De Vries et al.,

1982b). It is therefore important to minimize ischemia for the

accurate measurement of physiologically-relevant levels of

cyclic nucleotides in both rod and cone-dominant retinas.

In our light adaptation study, central retinal metabolism

was inactivated approximately 15 sec after sacrifice, and was

not interposed with anesthesia or tissue trauma, both of which

have been shown to alter cyclic nucleotide levels in central

nervous system tissue (Lenox et al., 1982; Cheng, 1982). Under

these conditions of minimal ischemia, dark-adapted central










39

RPEC complexes of the chick had approximately 13 pmoles/mg

protein each of cGMP and cAMP, giving a cGMP:cAMP ratio of

about 1. These values are similar to those reported by De

Vries et al., (1979) for the ground squirrel retina, but are

not in agreement with the high levels of cAMP reported by

Farber et al., (1981) for ground squirrel and western fence

lizard retinas. Levels of cGMP in rod-dominant retinas are

approximately 4-fold higher than in cone-dominant retinas and

are reduced by light adaptation by approximately 50% (De Vries

et al., 1978; Farber et al., 1981; Mitzel et al., 1978). This

is compared to the approximate 30% decrease in cGMP we

observed in the chick retina. The differences in cyclic

nucleotides in rod-enriched and cone-enriched retinas may be

related to the morphological, physiological and functional

differences that exist between rod and cone cell types. Cone

photoreceptors, which subserve daylight vision, require

approximately 400 times more light than rods to elicit a half-

maximal photoresponse (Schnapf and Baylor, 1987), and the

photoresponse to a given light stimulus is approximately 100

times smaller and is roughly 4 times faster in cones than in

rods (Schnapf and McBurney, 1980). Also, unlike rods, cones

begin to repolarize with sustained illumination (Normann and

Werblin, 1974). Thus, it is not unreasonable that steady-state

levels of cGMP, as well as the magnitude of the light-induced

decreases in cGMP, appear to be lower in cones than in rods.










40

In the chick central retina, a 5 sec pulse of light

produced a significantly greater decrease in the level of cGMP

compared to 2 hours of light adaptation (36% and 22%,

respectively). It is therefore possible that light-associated

decreases in cGMP levels may be better quantified in other

cone-dominant retinas if a brief light stimulus is employed.
















CHAPTER 3
LIGHT MODULATION AND DEVELOPMENTAL TIMECOURSE OF
CYCLIC NUCLEOTIDE LEVELS IN THE rd CHICKEN RETINA


Introduction


Cyclic nucleotides play a key role in the function and

integrity of vertebrate photoreceptor cells. Light-stimulated

decreases in outer segment levels of cGMP have been shown to

mediate visual transduction within rod photoreceptor cells

(reviewed in Stryer, 1986). While much less is understood

about transduction in cone photoreceptor cells, growing

evidence indicates that cones possess a cGMP-mediated

transduction mechanism similar to that found in rods (Haynes

and Yau, 1985; Cobbs et al., 1985; Hurwitz et al., 1985; Lerea

et al., 1986). In support of this view, we recently reported

significant light-induced decreases in cGMP levels in the

cone-dominant retina of the chicken (Chapter 2).

Excessive levels of cyclic nucleotide levels, however,

appear to be toxic to photoreceptor cells. Incubation of

retinas with high concentrations of cAMP or cGMP analogues

results in the selective degeneration of photoreceptors in

both rod-dominant (Lolley et al., 1977; Ulshafer et al., 1980)

and cone-dominant species (Williams et al., 1987). Retinal

levels of cGMP have also been shown to be elevated 2- to 10-










42

fold prior to photoreceptor degeneration in three strains of

rod-dominant animals possessing early onset inherited retinal

degeneration; the rd mouse (Lolley and Farber, 1976), Irish

Setter dog (Aguirre et al., 1982) and Collie dog (Woodford et

al., 1982). In these animals, which model retinitis pigmentosa

in man, high levels of photoreceptor cGMP are believed to

trigger the degeneration process by disrupting cellular

activities that are regulated by cGMP such as entry of Na* and

Ca'+ ions into the cell and the phosphorylation of proteins

which may potentially be involved in the modulation of

cellular metabolism, ion fluxes and/or neurotransmitter

release (Lolley et al., 1987). Studies conducted in vitro

suggest that high levels of cGMP may also disrupt

photoreceptor morphology by the inhibition of protein

synthesis within the cell (Ulshafer and Hollyfield, 1982). In

cone-dominant retinas, levels of cGMP and cAMP have been shown

to be elevated prior to cone cell damage induced by

iodoacetate injection (Farber et al., 1983). Cyclic nucleotide

levels have not yet been examined, however, in a mutant cone-

dominant animal possessing inherited retinal

degeneration.

The retinal degenerative (rd) chicken is a cone-

dominant animal possessing an autosomal recessive mutation

which results in blindness at hatch, followed by the

progressive degeneration of photoreceptor cells (Ulshafer et

al., 1984). On the day of hatch, scotopic and photopic










43

electroretinographic (ERG) responses are virtually absent in

the rd chick retina, indicating that transduction is defective

in both rod and cone photoreceptor cell types (Ulshafer et

al., 1984; Wolf, 1982). Paradoxically, however, the rd chick

retina appears morphologically normal at this time, complete

with fully-differentiated photoreceptors and normal outer

segment ultrastructure (Ulshafer and Allen, 1985a).

Photoreceptor degeneration begins 7-10 days after hatching and

proceeds swiftly from the central to peripheral retina. At 8

weeks posthatch (ph) very few outer segments remain in the rd

chick retina and by 6 months ph virtually all photoreceptors

are gone.

The unique defect in the rd chicken has enabled us to

examine cyclic nucleotide levels in a nontransducing yet

morphologically-intact cone-dominant retina as well as in a

degenerating cone-dominant system. This study reports the

effect of light adaptation on retinal levels of cGMP and cAMP

in rd chicks at 1-2 days ph, a time well prior to

photoreceptor degeneration in this retina. Retinal levels of

cGMP and cAMP were also measured in rd chickens at various

ages before and after hatching in order to investigate the

relationship between retinal cyclic nucleotide levels and

photoreceptor outer segment development, maturation and

degeneration in the rd chicken.











Methods


Animals. A breeding colony of Rhode Island Red chickens

homozygous for the rd mutation is maintained at the University

of Florida and is cared for according to NIH guidelines. Blind

rd/rd chicks were produced from controlled matings between

homozygous rd hens and rd roosters, while sighted heterozygous

chicks (rdd/+, termed carriers), were generated from matings

between rd/rd hens and normal-sighted (+/+) roosters. Eggs

were incubated in a forced-draft incubator until time of use

or until embryonic day 19 (E19), when they were placed in a

brooder for hatching. Posthatch chicks were maintained under

a 12h:12h light:dark cycle until used. Vision (or lack of

vision) was verified by the pecking (or lack of pecking) at

visual targets. Carrier chicks, which retain vision and normal

retinal morphology throughout life (Ulshafer et al., 1984),

served as sighted controls for all experiments.


Effect of Light Adaptation


At 1-2 days ph, chicks were light-adapted under

laboratory illumination (approximately 1600 lux) or were kept

in total darkness for 2-3 hr, after which time they were

sacrificed in the same lighting by decapitation. The heads

were placed immediately into liquid nitrogen. In some chicks,

the external eyelids were surgically removed prior to light or

dark adaptation, in an attempt to accelerate freezing of the

retina, since skin has been shown to be the major factor which










45

slows freezing of the brain (Swaab, 1971). Prior to surgery,

chicks were temporarily anesthetized with CO2 vapors and,

following surgery, local anesthetic (xylocaine) was applied to

the affected area and chicks were allowed to regain

consciousness before light or dark adaptation. Since this

procedure did not significantly alter the results of our

experiments we did not continue this practice and combined the

results of animals with eyelids removed and those with eyelids

intact in our analysis. Frozen heads were dissected in a -35C

cryostat (IEC CTF Microtome-Cryostat, International Equipment

Co.) to expose the retinal surface as previously described

(Chapter 2). Retina-pigment epithelium-choroid (RPEC)

complexes were either removed from the whole retinal area

(central plus peripheral) using a scalpel or were removed from

central retina only, using a 4 mm trephine punch. RPEC

complexes were stored in liquid nitrogen until sonicated in

300 ul ice cold 1.0 N HC1 for approx. 30 sec at 70% output

power using an Ultrasonic Disruptor (Heat Systems Inc.). The

PE-C unit was included with the retina to insure the

acquisition of photoreceptor outer segments.


Development of Cyclic Nucleotide Levels


For embryonic timepoints, eggs of the appropriate age

were removed from a transparent incubator during the light

cycle. The embryos were isolated quickly, sacrificed by

decapitation and the heads placed immediately into liquid










46

nitrogen. Embryonic age was verified by measuring toe and beak

lengths as described by Hamilton (1952). For posthatch

timepoints, chicks aged 1-56 days ph were light-adapted in

room light (approximately 1600 lux) for 2 hrs and subsequently

sacrificed by decapitation and the heads promptly immersed

into liquid nitrogen.

Heads were dissected in a -350C cryostat to expose the

retinal surface as described in Chapter 2. A punch of RPEC

complex was removed from the central eyecup using a cold

trephine appropriately-sized for the eye. Punches were 4 mm

diameter in chicks ages embryonic day 12 (E12) through day 7

ph, and were 5 mm, 7 mm and 8 mm diameter in chicks ages 16,

35, and 56 days ph, respectively. In some 1 day ph and 35 day

ph chicks, RPEC punches were also removed from peripheral

retina. Punches were kept in liquid nitrogen until sonication

(30-45 sec, 70% output power--Ultrasonic Disruptor, Heat

Systems Inc) in 300-400 ul ice cold acid (1.0 N HCl).


Extraretinal Tissues


Tissue from cerebellum, optic tectum, heart and liver was

surgically removed from decapitated 1-2 day ph light-adapted

chicks and sonicated (30-60 sec, 70% output power) in 600-800

ul ice cold 1.0 N HC1.

Cyclic nucleotide assay. After sonication, tissue

homogenates were heated in boiling water for 3 min to insure

protein precipitation and centrifuged for 10 min at 40C.










47
Supernatants were adjusted to pH 6-7 with 1.0 N NaOH, diluted

to an appropriate volume with 0.05 M Na acetate buffer (pH

6.2) and, after acetylation, assayed for cGMP and cAMP content

by radioimmunoassay (Steiner et al., 1972) using commercial

kits (Biomedical Technologies, Inc.). Two to four

determinations were performed on dilutions within the assay

range. Proteins were measured in the solubilized pellets using

the method of Lowry et al. (1951) with BSA standard.

Statistical Tests. Statistical significance between data

groups was determined using the Student's t-test for unrelated

samples. Values of p<.05 were considered significant.


Results


Validity of the cyclic nucleotide assay procedure was

tested by reacting some samples with 50 or 100 ug/ml

commercial cyclic nucleotide phosphodiesterase, PDE, (Sigma).

Following 1 hr PDE hydrolysis, levels of both cGMP and cAMP

were below assay detection (Fig. 3.1), verifying the

specificity of the RIA procedure for measurement of cGMP and

cAMP. Additionally, some samples were purified prior to

radioiummunoassay on BioRad AG1-X8 format columns (200-400

mesh), using 1.0 N formic acid to elute cAMP and 4.0 N formic

acid to elute cGMP (Farber and Lolley, 1982) (Fig. 3.2).

Levels of cAMP and cGMP were similar in purified and

nonpurified retinal extracts of both sighted and blind chicks,















100-


0 30 60 120
minutes of PDE hydrolysis


100-


60
minutes of PDE hydrolysis


120


Figure 5.1. Percentage of original content (approx. 30nM) of cGMP (top)
and cAMP (bottom) remaining in RPEC supernatants after incubation
with 100ug/ml PDE in 0.1M Tris HCI, pH 7.5, with 2mM MgSO4 at 300C.


0--0 standard
A carrier
O rd












\


0 -O standard
A carrier
0 rd











0 -0----_
A


I





































- cGMP
----- cAMP
cGMP + cAMP


I ~ ~I*~-I*-


1 2 3 4 5 6 7 8 910 1 2 3 45 6 7 8 910 1 2 3 4 5 6 7 8 9 10


H20 2N Formate 4N Formate
Eluted Fraction (ml)

Figure 3.2. Elution profile of 3HcGMP and 3HcAMP from AG1-X8 format columns.
After subtraction of 3H20, 96% of 3HcGMP and 98% of 3HcAMP were recovered.


9000

8000

7000

6000


E 5000

4000

3000

2000

1000

0


I I
































4) V-1 r-
0 H < H

4)
0 N ,
a)
S- ----
-4
0

Eu *
O V N 0 -


0 rm c 0 c-
) a7 o 0 H
SH H


.gc



i H1
.
Cu v H H4 H H






0 v H H H H







U) 4J
R &


0 0
HW


>4J
00

V E
CUr



o o

>)4

'$4 I



0 00
S00
t4 X
0) ) .-4
0H4-li


0
.-4






>1














0




V

r)
0










4)
t0






e
rO1










51

ruling out possible sample interference in the nonpurified

extracts (Table 3-1).


Effect of Light Adaptation


Figure 3.3 shows the effect of light adaptation on cGMP

levels in whole RPEC complexes of 1 day ph carrier and rd

chicks. Carrier chick RPEC complexes contained an average of

10.5 pmole cGMP/mg RPEC protein in the dark. Light adaptation

significantly reduced carrier RPEC levels of cGMP by 19%

(p=0.001), a decrease comparable to that observed in

genetically-normal chicks (Chapter 2). In the nontransducing

yet morphologically-normal RPEC complexes of rd chicks, cGMP

levels were severely depressed. Both light-adapted and dark-

adapted levels of cGMP were about 87% lower in rd RPEC

complexes compared to carrier RPEC complexes. In addition,

unlike sighted chick RPEC complexes, blind rd RPEC complexes

demonstrated no light-dark difference in cGMP content.

Our finding of significant light-induced decreases in

cGMP levels in carrier chick whole RPECs, where cones

outnumber rods approximately 3 to 1, prompted us to

investigate the effect of light adaptation on cGMP levels in

central RPECs, where the cone:rod ratio increases to 6 to 1.

As shown in Fig. 3.4, both dark-adapted and light-adapted

levels of cGMP, as well as the percent decrease in cGMP levels

in response to light adaptation, were similar in whole and

central RPEC complexes of carrier chicks.


















I DARK
=-- LIGHT


0'


Carrier


Figure 3.3. Effect of light adaptation on cGMP levels in whole RPEC complexes
of carrier and rd chicks. Values represent the mean + SD of 7-11 eyes per
group. *p=.001


Figure 3.4. Effect of light adaptation on cGMP levels in RPEC complexes
isolated from the whole retina and the central retina of carrier chicks.
Values represent the mean SD of 6-8 eyes per group. *p(.001.


15





0
10-
a-

E


S5-
E
0-
EL


0
C
r10


a-1

a-
0
E5

E
a-


p-- I-










53
Levels of cAMP in whole RPEC complexes averaged about 8

pmole/mg protein in dark-adapted carrier chicks (Fig. 3.5).

Light adaptation had no effect on carrier levels of cAMP. In

rd whole RPEC complexes, levels of cAMP were elevated 50% in

the light and 41% in the dark compared to carrier cAMP levels.

As in carriers, light adaptation had no effect on cAMP levels

in rd RPEC complexes. The ratio of cGMP:cAMP in rd RPEC

complexes was found to be 0.13, 10-fold lower than that in

carrier RPECs.


Extraretinal Tissue Levels of cGMP and cAMP


Carrier and rd chicks had essentially the same levels of

cGMP in each extraretinal tissue examined (Table 3-2). Levels

of cAMP were also quite similar in all tissues except the

optic tectum, where rd levels of cAMP were elevated 22%.


Developmental Time Course


In the developing chicken retina, photoreceptor inner

segments begin forming on embryonic day 10 (E10) (Olson, 1979)

and outer segments are first visible between days E15-E16

(Meller and Tetzlaff, 1976). Measurable ERGs are present on

E17 (Rager, 1979) and, by E19-E20, the chick retina is fully

differentiated and functionally complete (Olson, 1979;

Coloumbre, 1955).

Figure 3.6 shows that on E12, prior to the formation of

outer segments, levels of cGMP were comparably low in central









































S- DARK
T C LIGHT


Carrier


Figure 3.5. Effect of light adaptation on cAMP levels in whole RPEC complexes
of carrier and rd chicks. Values represent the mean + SD of 7-10 eyes per
group.


15





0
Of
"10-
bJ
0-

E
0C

o
-6

E
0-


0'


I I
















Table 3-2. Cyclic Nucleotide Levels in Extra-retinal
Tissues of Carrier and rd Chicks


cGMP
(pmoles/mg protein)


cAMP
(pmoles/mg protein)


rd/+ 0.26 + 0.07a 30.73 + 6.29
Cerebellum
(5) rd/rd 0.36 + 0.12 32.96 + 5.58
Optic rd/+ 0.67 + 0.05 32.48 + 2.24
Tectum
(5) rd/rd 0.74 + 0.10 39.60 + 4.83 *
rd/+ 0.26 + 0.07 4.71 + 0.27
Heart (5)
rd/rd 0.32 + 0.10 5.29 + 0.55
rd/+ 0.09 + 0.01 5.04 + 0.84
Liver (7)
___i__ver (rd/rd 0.09 + 0.01 5.59 + 1.07

a Each value represents the mean + SD of the number of
samples designated in parentheses.
* rd significantly differs from carrier, p=.017


















0-0 Cr
*--* rd




T T
T /0 0
0



TO
0
v
0


/ 0- -


20-







0-


LJ
15-
0-
a-





0 5-




0-


AGE (days)
Figure 3.6. cGMP in central RPEC complexes of light-adapted carrier and
rd chicks as a function of age. Day 1 is day of hatch. Values represent
the mean SD of 6-11 eyes for each data point. Absent error bars
fall within the size of the points.


E12 E18 1 7 16 35
AGE (days)
Figure 3.7. cAMP in central RPEC complexes of light-adapted carrier and
rd chicks as a function of age. Day 1 is day of hatch. Values represent
the mean +.SD of 6-11 eyes for each data point. Absent error bars
fall within the size of the points.


56


4-


E12 E18 1 7 16 35


20-


.-

215
a-
0







E 5



0


0-0 Cr
I*-- rd


0




0 0 0
T0
0 T T

-


I ....


I


. I I I I . .


I


I I I I I I I I.










57

RPEC complexes of carrier and rd chicks (approximately 2 pmole

cGMP/mg protein). Between E12 and day of hatch (day 1 ph),

levels of cGMP in carrier central RPECs increased 4-fold, in

temporal association with outer segment development. cGMP

levels continued to rise during the first week posthatch, as

outer segments elongated, then stabilized at 12 pmole/mg

protein on day 7 ph, by which time outer segments had reached

adult length. Levels of cGMP in rd central RPEC complexes

failed to increase during outer segment development. cGMP in

rd complexes remained at pre-outer segment levels (2 pmole/mg

protein) throughout the course of outer segment development

(E12-1 ph), maturation (1-7 ph) and photoreceptor degeneration

(7-56 ph), 2- to 6-fold lower than the cGMP levels found in

carrier central RPEC complexes.

In comparison to cGMP, substantial levels of cAMP were

present in central RPEC complexes of carrier and rd chicks

prior to outer segment development on E12 (Fig. 3.7). Levels

of cAMP in both carrier and rd RPEC complexes oscillated with

age and the fluctuations showed no consistent relationship

with the development of photoreceptor outer segments. Carrier

and rd central RPEC complexes demonstrated a similar

qualitative pattern. Between E12 and E18, at the start of

outer segment formation, levels of cAMP increased 40% in both

carrier and rd RPEC complexes. With continued outer segment

development (E18-1 ph) and elongation (1-7 ph), however,

levels of cAMP declined in carrier and rd RPEC complexes and,










58

thereafter, increased gradually with age. At E12 and E18,

levels of cAMP in rd central RPEC complexes were 18% higher

than those in carrier RPECs. On the day of hatch, however,

central RPEC levels of cAMP were comparable in rd and carrier

chicks. By the onset of photoreceptor degeneration on day 7

ph, rd central RPEC levels of cAMP were depressed 32% relative

to those in the carrier. Thereafter, with the progressive loss

of photoreceptors, rd levels of cAMP remained approximately

30% lower than carrier levels of cAMP, but, as in sighted

birds, gradually increased with age.


Central vs. Peripheral Retinal Cyclic Nucleotide Levels


Table 3-3 compares cyclic nucleotide levels in 4 mm

punches of RPEC complexes isolated from central and peripheral

retinal areas of light-adapted carrier and rd chicks. cGMP

levels in central retinal RPEC complexes were comparable to

those in peripheral retinal RPEC complexes in both 1 day ph

carrier and rd chicks, as well as in 35-day-old rd chicks that

had incurred substantially more photoreceptor degeneration in

the central retina than in the peripheral retina. Central RPEC

levels of cAMP, on the other hand, were significantly higher

than peripheral RPEC levels of cAMP in all groups examined. At

hatch, central cAMP levels were elevated relative to

peripheral cAMP levels 39% and 20% in carrier and rd chicks,

respectively. At 35 days ph, central levels of cAMP remained

significantly higher than peripheral levels in rd chicks,
















Table 3-3. Cyclic Nucleotide Levels in RPEC Complexes
Isolated From Central and Peripheral Retinal
Areas Light-adapted Carrier and rd Chickens

cGMP (pmole/mg RPEC protein)
Central Peripheral %Difference

Id ph Cr (6) 8.46 + 0.60a 8.92 + 1.58 NSD

id ph rd (8) 2.41 + 0.36 1.95 + 0.21 NSD

35d ph rd (5) 1.79 + 0.14 1.60 + 0.26 NSD



cAMP (pmole/mg RPEC protein)
Central Peripheral %Difference

ld ph Cr (6) 11.76 + 0.28 8.44 + 1.25 39% p<.001

id ph rd (8) 12.60 + 0.89 10.49 + 1.24 20% p=.002

35d ph rd (5) 9.32 + 0.86 6.72 + 0.41 39% p<.001


a Values represent the mean + SD of the number of RPEC
complexes designated in parentheses.










60

despite a substantial loss of photoreceptors in the central

retina. Comparison of carrier and rd chicks reveals that, at

hatch, rd levels of cAMP were significantly elevated only in

the peripheral retina. Central retinal levels of cAMP were

similar in blind and sighted chicks at hatch. Increased levels

of cAMP in peripheral, but not central, RPEC complexes of rd

chicks is consistent with our observation of elevated cAMP

levels in rd whole RPEC complexes (Fig. 3.5), which consists

predominantly of peripheral retina.

Based on our measurement of cyclic nucleotide levels in

4 mm punches of RPEC, we calculated the approximate molar

concentrations of cyclic nucleotides in the chicken RPEC

complex as follows; the width of the RPEC in the central area

of 1-day-old chicken retina is approximately 230 um, as

measured with an ocular micrometer in JB4-embedded, light

microscope sections (Chapter 4). The volume occupied by a 4 mm

punch of central retinal chick RPEC is therefore approximately

2.9 ul. Protein levels in RPEC punches averaged 1.08 mg/ml

0.08 (n=12) and were similar in carrier and rd chicks. Each 4

mm RPEC punch thus contained approximately 0.65 mg protein.

Using these values, the calculated concentrations of cGMP in

central RPEC complexes of lighted-adapted carrier and rd

chicks were 2 uM and 0.5 uM, respectively. The concentration

of cAMP was estimated to be approximately 3 uM in both carrier

and rd chick central RPEC complexes.












Discussion

The results presented in this study show that defective

transduction in the cone-dominant retina of the rd chicken is

associated with a severe reduction in retinal levels of cGMP

and that the existing levels of cGMP in the rd retina are not

reduced in response to light adaptation. Combined with our

findings in sighted chick RPECs that cGMP levels are

significantly decreased following light adaptation and that

most of this cGMP develops in temporal association with the

photoreceptor outer segments, our results strongly support

mounting evidence that cGMP is involved in the mediation of

visual transduction in cones.

In recent years, cone photoreceptors have been shown to

possess all of the established major proteins necessary for

cGMP-mediated transduction, including visual pigments (Nathans

et al., 1986), transducin (Lerea et al., 1986) and cGMP

phosphodiesterase (PDE) (Hurwitz et al., 1985). Additionally,

cGMP has been shown to directly gate ion channels in cone

outer segment membranes in a concentration-dependent manner

(Haynes and Yau, 1985) and to modulate light-sensitive current

in isolated cones (Cobbs et al., 1985). We recently reported

significant light-induced decreases in cGMP levels in the

cone-dominant retina of genetically-normal chicks, providing

the first in vivo evidence that cGMP mediates cone

transduction (Chapter 2).










62

In the present study, RPEC levels of cGMP in sighted

chicks heterozygous for the rd mutation were found to

significantly decrease in response to light adaptation. In

addition, 80% of the total cGMP content of the central retina

was found to develop concomitantly with the photoreceptor

outer segments. Similar developmental increases in outer

segment-associated cGMP levels have been observed by others in

genetically-normal chick retina (de Mello, 1978). Since

approximately 87% of the photoreceptors in the central chick

retina are cones (Morris and Shorey, 1967), our results

suggest 1) that cones contain a major portion of the cGMP in

the chick retina and 2) that the light-stimulated decreases in

cGMP levels in RPEC complexes of sighted chicks are associated

with the photoreceptor cells. We additionally found in carrier

chicks that dark-adapted levels of cGMP, as well as the

percentage decrease in cGMP by light adaptation, were

comparable in RPEC complexes isolated from central retina

(where rods constitute approximately 13% of the

photoreceptors) and from whole retina, which consists

predominantly of peripheral retina (where the percentage of

rods increases to 25%). These results imply that the light-

sensitive pool of cGMP in the chick retina is associated with

cone as well as rod photoreceptor cell types.

In the nontransducing rd chick retina, central RPEC

levels of cGMP were normal prior to outer segment development

but failed to increase as outer segments formed. This suggests










63

that cGMP metabolism in the rd chick retina is disrupted

within the photoreceptor cells. On the day of hatch, rd RPEC

levels of cGMP were depressed 87% and were unaltered by light

exposure, despite the presence of well-developed and

morphologically-normal outer segments. Thus, defective

transduction in the rd chick retina is closely associated with

the failure to accumulate normal levels of retinal cGMP during

outer segment development. These results imply that rd chick

retinas may not develop sufficient photoreceptor levels of

cGMP to support visual transduction. Abnormally-low levels of

cGMP could account for the absence of electrical activity in

rd chick photoreceptor cells by causing the permanent closure

of outer segment membrane channels, leaving the photoreceptors

in a constant state of hyperpolarization.

The results of the present study are not supportive of a

critical role for cAMP in cone transduction. First, the

presence of substantial levels of cAMP in central RPEC

complexes prior to outer segment formation as well as the

oscillation of cAMP levels during outer segment development

make it unlikely that the photoreceptors are the primary

source of cAMP in the chicken retina. Our findings are in

agreement with reports showing that cAMP is evenly distributed

throughout the cone-dominant retina of the ground squirrel (De

Vries et al., 1979) and the rod-dominant retina of the rabbit

(Orr et al., 1976). Second, RPEC levels of cAMP were also not

significantly affected by light adaptation in sighted carrier










64

chicks. However, since photoreceptors do not appear to be a

major contributor of cAMP in chick retina, it is possible that

light-induced changes within these cells are being masked by

cAMP levels in other retinal layers. Examination of cAMP

levels in the individual cell layers of the chicken retina are

needed to investigate this possibility. In the cone-dominant

ground squirrel, significant light-induced decreases in cAMP

levels have been demonstrated in freshly-dissected whole

retinas (Farber et al., 1981) but have not been detected in

the retinal layers of freeze-dried preparations (De Vries et

al., 1979).

Levels of cAMP in 1 day ph rd chicks were elevated

relative to carrier chick cAMP levels in peripheral retina and

in whole retina (which consists predominantly of peripheral

retina), but were normal in central retina. Levels of cAMP in

rd central RPECs were also fairly comparable to those in

carriers RPECs during outer segment development. These results

indicate that cAMP levels are not significantly altered in

association with blindness in the rd chick retina. cAMP levels

were decreased in temporal association with morphological

degeneration in the rd chick retina. On day 7 ph, levels of

cAMP were reduced 30% in rd central RPECs, coincident with the

onset of degenerative changes in rd photoreceptors. Between

day 7-56 ph, rd RPEC levels of cAMP remained reduced by 30%

and increased with age in a pattern qualitatively similar to

that observed in carriers, even after substantial










65

photoreceptor cell loss. These observations support our notion

that the photoreceptors contribute only a portion of the cAMP

content in the chick RPEC complex. In addition, in 35 day ph

rd chicks, a time at which more photoreceptors have

degenerated in the central retina compared to peripheral

retina, central RPEC levels of cAMP were significantly higher

than peripheral RPEC levels of cAMP. This suggests that cone

cells are not the primary source of cAMP making central

retinal levels of cAMP higher than peripheral retinal levels

of cAMP in the chicken. This is in contrast to the conclusion

of a study involving postmortem human eyes (Farber et al.,

1985) in which the authors attributed elevated levels of cAMP

in the central retina (foveal area) to the high density of

cone photoreceptors in this region. In the chicken retina,

central RPEC levels of cAMP may be higher than peripheral RPEC

levels of cAMP because of the increased thickness of the cAMP-

rich inner plexiform layer and choroid layer in this region

(Coloumbre, 1955; Chapter 4).

In contrast to cAMP, there was no apparent correlation

between morphological degeneration in the rd chick retina and

changes in central RPEC levels of cGMP. Indeed, a 4-fold

reduction in retinal cGMP levels does not appear to adversely

affect the morphological development of rd chick photoreceptor

cells. It is possible, however, that the abnormally-low levels

of cGMP apparent in the rd retina at E18 may lead (directly or

indirectly) to subsequent photoreceptor cell death at 7-10










66

days posthatch. Effects of reduced photoreceptor levels of

cGMP that could potentially disrupt cellular integrity include

1) reduced intracellular levels of Ca++ resulting from reduced

Ca++ influx through the cGMP-gated conductance, 2) reduced

activity of cGMP PDE and, as a consequence, reduced proton

production, 3) disrupted protein phosphorylation secondary to

the altered activity of cyclic nucleotide dependent protein

kinase, and 4) sustained cellular hyperpolarization.

Degeneration may result either from the general disruption of

cell homeostasis and/or from the disruption of a process or

protein that is specifically involved in photoreceptor cell

maintenance.

Abnormally-low levels of photoreceptor cGMP have also

been shown to precede photoreceptor degeneration in the rds

mouse mutant, in which photoreceptors degenerate during the

early stages of outer segment development (Cohen, 1983). The

apparent defect in cGMP metabolism in the rds mouse, however,

is different from the that in the rd chicken in that, in the

rds mouse, retinal levels of cGMP are reduced in response to

light, despite being abnormally low (Cohen, 1983). The light-

insensitivity of cGMP levels in rd chick RPECs is consistent

with defective cGMP metabolism being temporally-associated

more with a functional defect within the photoreceptors rather

than with their degeneration.

The apparent defect in cGMP metabolism in the rd chick

appears to be restricted to the retina since levels of cGMP in










67

all extraretinal tissues examined, including the optic tectum,

were normal. Abnormal cGMP levels in the rd chick retina may

be the result of a mutation affecting either the primary

structure or the amounts of an enzyme that plays a key role in

cGMP metabolism such as cGMP PDE or guanylate cyclase. We have

examined this possibility and will report on it in subsequent

chapters. Alternately, disrupted cGMP metabolism may be due to

a defect elsewhere in the transduction cycle or in a factor

critical to the maintainance of transduction such as post-

translational modification, ion regulation or energy

metabolism.
















CHAPTER 4
DISTRIBUTION OF cGMP AND cAMP IN THE CONE-DOMINANT
RETINAS OF SIGHTED AND BLIND (rd) CHICKENS


Introduction


In rod-dominant retinas from several species, cGMP levels

are concentrated 10-100 fold within the photoreceptors while

levels of cAMP are more uniformly distributed throughout the

retina (Orr et al., 1976; Steiner et al., 1972; De Vries et

al., 1978; Berger et al., 1980; Ferrendelli and Cohen, 1976).

Light adaptation significantly decreases cGMP levels

throughout the rod receptor cell (Orr et al., 1976; Fletcher

and Chader, 1976; de Azeredo et al., 1981). In the rod outer

segment, the light-stimulated hydrolysis of cGMP has been

shown to mediate visual transduction by bringing about the

closure of cGMP-gated cation channels in the outer segment

plasma membrane (reviewed by Stryer, 1986). Light adaptation

also decreases cAMP levels in rod outer nuclear and outer

plexiform layers (Orr et al., 1976; De Vries et al., 1978), an

effect believed to be associated with synaptic transmission

(Blazynski et al., 1990) and light-entrained events (Burnside

et al., 1982) in the photoreceptor.

Studies concerning the distribution of cyclic

nucleotides in the cone-dominant retina have been limited to










69

the ground squirrel but reveal a similar pattern: cGMP is

concentrated 10- to 40-fold within the cones, while cAMP is

dispersed throughout the retina (De Vries et al., 1979).

Unlike rod-dominant retinas, however, light adaptation was

reported to have no effect on levels of cGMP (or cAMP) in any

layer of the ground squirrel retina (De Vries et al., 1979).

This report is puzzling in view of growing evidence that

light-stimulated decreases in outer segment levels of cGMP

mediate visual transduction in both rod and cone photoreceptor

cell types (Haynes and Yau, 1985; Cobbs et al., 1985; Hurwitz

et al., 1985; Lerea et al., 1986; Nathans et al., 1986).

We recently reported significant decreases in cGMP levels

in the cone-dominant chicken retina following light

stimulation (Chapters 2 & 3). Since approximately 80% of the

total cGMP content of these retinas develops concomitantly

with the photoreceptor outer segments (Chapter 3), this points

to the existence of a light-sensitive pool of cGMP within the

cone photoreceptor cells. Perhaps more importantly have been

our observations that, in the nontransducing, yet normal-

appearing retinas of congenitally-blind rd chicks, cGMP levels

do not increase during outer segment development and, on the

day of hatch, are severely depressed compared to normal and

are unaffected by light adaptation (Chapter 3). These findings

indicate the presence of abnormally-low and light-insensitive

levels of cGMP within the nontransducing cones of rd chicks.










70

In this study, we measured the effect of light

adaptation on cyclic nucleotide levels in the various retinal

cell layers of sighted and rd chicks. The retinal

distributions of cGMP and cAMP were determined in central

chick retina, where approximately 87% of photoreceptors are

cones (Morris and Shorey, 1967).


Methods


Animals. Controls were sighted carrier chicks,

heterozygous for the rd mutation. Carrier chicks retain vision

and normal retinal morphology throughout life (Ulshafer et

al., 1984), and at the age employed in this study (1-2 days

posthatch), possess ERG signals which are comparable to those

in homozygous (wild type) normal chicks (Dawson et al., 1990).

At 1-2 days posthatch, levels of cGMP in dark-adapted central

retina-pigment epithelium-choroid (RPEC) complexes of carrier

chicks (11.08 + 0.72 pmole/mg protein, n=6) are not

significantly different (p=.07) from those in normal chicks

(12.77 + 1.95 pmole/mg protein, n=7), while levels of cAMP in

carrier RPEC complexes (9.96 0.87 pmole/mg protein, n=7) are

27% lower than those in normal RPECs (13.64 + 1.18 pmole/mg

protein, n=7).

Carrier chicks and blind homozygous rd chicks were

produced from a breeding colony at the University of Florida

as described in Chapter 3. At age 1-2 days posthatch (ph),

chicks were either dark-adapted in total darkness or light-









71

adapted under 27000 lux cool white fluorescent illumination

for 1-2 h and were then sacrificed by decapitation under the

same light conditions. Heads were quickly frozen in liquid

nitrogen. The time from decapitation to freezing of the

central retina to -10C was about 15 sec, a minimal ischemic

interval (Chapter 2). Heads were bissected longitudinally in

a -25C cryostat and mounted eyelid-side down onto a

dissection chuck without thawing. The posterior eyepole was

cleared of extraneous tissue with a prechilled scalpel and 7

um sections of central retina cut at an angle tangential to

the posterior eyepole in order to maximize the widths of the

retinal cell layers. Sections were transferred to a drying

assembly (Ace Glass) and dried overnight at -40C under

vacuum. Dried sections were stored under vacuum at -700C until

further dissection, which was carried out in a draft-free

glove box maintained at 40-55% relative humidity. Individual

retinal cell layers were hand-dissected under 40X

magnification using hair loops and razor blade shards

following the method of Lowry and Passoneau (1972).

Undissected, stained sections aided identification of retinal

layers. Like dissected layers of 10-30 sections (1 eye) were

combined, extracted in 200 ul ice cold 10% TCA, frozen/thawed

4 times and centrifuged (13,600 xG) at 4C for 10 min.

Supernatants were extracted with 4 x 5 volumes of water-

saturated ether, diluted with 0.05 M Na acetate buffer (pH

6.2) and assayed in duplicate for cGMP and cAMP using an










72

acetylated radioimmunoassay as previously described (Chapter

3). Proteins were assayed using the Lowry method (Lowry et

al., 1951), modified for 1-10 ug amounts of protein.


Results


Figure 4.1 shows an unstained freeze-dried retinal

section prior to dissection. The orange-colored layer consists

of oil droplets, which are present in the apical ends of cone

inner segments in the avian retina. Photoreceptor outer

segments are located directly above the oil droplets and

project deeply into the overlying pigment epithelium (PE).

During dissection, the PE layer tended to separate from the

retina, leaving outer segments both embedded within the PE as

well as extending from the oil droplet layer. Thus, in order

to better estimate cone outer segment levels of cyclic

nucleotides, we procured both PE and oil droplet (OD) layers,

with their associated outer segments. The dissected layers of

a representative freeze-dried retinal section are shown in

Fig. 4.2.


Distribution of cGMP


The distribution of cGMP in the central retina of carrier

chicks and rd chicks is shown in Fig. 4.3. The retinal cell

layers are scaled on the X-axis according to their actual

widths in the central retina, as measured with an ocular

micrometer in JB4-embedded, light microscope sections. In

























Figure 4.1. Freeze-dried section of retina-PE-choroid from the
central chick retina as viewed under a dissection microscope
at 40X magnification. The section is unstained and unmounted.
The retinal cell layers can be distinguished by their various
colors and shadings. Note the separation between the PE
(black) and OD (orange) layers at the left.


Figure 4.2. Section as in Fig.4.1 following dissection into
retinal layers. The layers are, from right to left, choroid,
pigment epithelium plus outer segments, oil droplets plus
outer segments, inner segment-outer nuclear layer-outer
plexiform layer, inner nuclear layer, inner plexiform layer.
Ganglion cell-nerve fiber layer not shown. The distance
between bars at the top is 1 mm.












Figure 4.3. Distribution of cGMP in the central retina of
dark-adapted and light-adapted sighted (carrier) and rd
chicks. Freeze-dried sections were hand-dissected into eight
layers (CHOROID; PE-OS, pigment epithelium plus outer segment;
OS-OD, outer segment plus oil droplets; IS-ONL, inner segment
plus outer nuclear layer; OPL, outer plexiform layer; INL,
inner nuclear layer; IPL inner plexiform layer; GC-NFL,
ganglion cell plus nerve fiber layer). In dark-adapted
retinas, the IS-ONL-OPL was dissected as a unit; values were
thus placed at an intermediate position between IS-ONL and OPL
layers. Retinal layers are scaled to their actual widths in
central retina, as measured with an ocular micrometer in cross
sections (n=8) of JB4-embedded 1-day sighted chick eyes. The
points represent the mean + SD of 3-4 eyes (outer retinal
layers of all groups except rd dark) or the range of 2 eyes
(remaining points), each measured in duplicate. Absent error
bars fall within the size of the points.


Figure 4.4. Distribution of cAMP in central retina of dark-
adapted and light-adapted sighted (carrier) and rd chicks.
Abbreviations, standard deviations etc. are as in Fig. 4.3.











0-0 carrier dark
0- 0 carrier light
25 A- A rd dark
SA-- A rd light
220 I

E
_15
0
S15-- I


610 /1-
E O I





0 I I I .I I I I.-
CHOROID PE-OS IS-ONL OPL INL IPL GC-NFL
OS-OD

Figure 4.3



30
-- 0 carrier dark
O-O carrier light
25 A- A rd dark
0 A- A rd light

* 20--


E 15--
-5 I \2

o _
0,

10-





0 I II I I 1
CHOROID PE-OS IS-ONL OPL INL IPL GC-NFL
OS-OD


Figure 4.4










76

carrier retinas, 80% of the total measured cGMP was located

within the photoreceptor cells. Photoreceptor levels of cGMP

were 2- to 8-fold higher than those in non-photoreceptor

regions in both darkness and light. In the dark, carrier

photoreceptor levels of cGMP were high throughout the cell.

Both outer segment (OS-OD layer) and proximal (IS-ONL-OPL

layer) regions of the cell contained approximately 25 pmole

cGMP/mg protein. cGMP levels in the outer segment-containing

PE-OS layer averaged about 11 pmole/mg protein in darkness.

(The actual molar concentration of cGMP within the outer

segments is likely to be much higher than indicated by either

measurement, especially by PE-OS levels, since the protein

content of the outer segments contributes only a portion of

the total protein in these layers.) Light adaptation

significantly reduced cGMP levels in the proximal

photoreceptor cell layers (IS-INL and OPL) by approximately

65% but had no significant affect on cGMP levels in the outer

segment region (PE-OS and OS-OD). No light-dark differences in

cGMP content were observed in the non-photoreceptor cell

layers.

In rd chick retina, levels of cGMP in all retinal layers

containing the photoreceptor cells were significantly lower

than those observed in carrier retinas. In particular, outer

segment-containing layers in the rd retina had cGMP levels 4-

fold lower than in carrier retinas in both the light and dark.

Proximal photoreceptor layers (IS-INL, OPL) in the rd retina










77
had cGMP levels that were depressed 90% in the dark and,

unlike sighted retinas, were unaffected by light adaptation.

cGMP levels in the non-photoreceptor cell layers were

comparable in rd and carrier chicks in both the light and

dark.


Distribution of cAMP


Compared to cGMP, levels of cAMP were distributed more

evenly throughout the chicken retina (Fig. 4.4). In dark-

adapted carrier chicks, cAMP levels peaked in the proximal

(IS-INL-OPL) photoreceptor layers (23 pmole/mg protein) and in

the IPL (20 pmole/mg protein), and were lowest in the outer

segment-containing PE-OS layer (7 pmole/mg protein).

Photoreceptors contained approximately 46% of the total cAMP

content. Light adaptation decreased cAMP levels 40-50% in both

the proximal photoreceptor (IS-ONL-OPL) layers and the INL.

In the rd chick retina, light-adapted levels of cAMP were

similar to those in sighted carriers in all layers except the

GC-NFL, where levels were reduced by approximately 30%. Dark-

adapted cAMP levels in rd retinas were significantly less than

those observed in carrier retinas in the IS-ONL-OPL (72%), INL

(53%) and in the outer segment-containing OS-OD layer (30%).

In the outer segment-containing PE-OS layer, however, cAMP

levels were relatively high and were comparable between rd and

carrier chicks. Since the OS-OD layer is partially comprised

of inner segments, in which rd levels of cAMP were severely









78

reduced, it is likely that low cAMP levels in the OS-OD layer

are predominantly associated with the inner segment and that

outer segment levels of cAMP are not significantly disrupted

in the rd retina. Light adaptation significantly altered

proximal photoreceptor (IS-ONL-OPL) levels of cAMP in the

nontransducing rd retina. However, in contrast to sighted

chicks, rd levels of cAMP were increased rather than decreased

following light exposure.


Discussion


The present study demonstrates that the general

distribution of cyclic nucleotides in the cone-dominant

chicken retina is similar to that found in rod-dominant

retinas (Orr et al., 1976; Berger et al., 1980; De Vries et

al., 1978) and in the cone dominant ground squirrel retina (De

Vries et al., 1979). In all these retinas, cGMP is

concentrated almost exclusively in the photoreceptor cells

with high levels present throughout the cell, while cAMP is

more uniformly dispersed throughout the retina. The

distribution of cGMP in the chick retina at hatch is

consistent with our earlier observations that increases in

cGMP in developing chick retina are temporally associated with

photoreceptor outer segment development (Chapter 3). The

similar distribution of cGMP and cAMP in rod- and cone-

dominant retinas suggests cyclic nucleotides function

similarly in both photoreceptor cell types.












cGMP


Our results reveal significant light-induced decreases in

cGMP levels within cone photoreceptors. The 65% reduction in

cGMP levels in the proximal (IS-ONL and OPL) layers of carrier

chick cones is sufficient to account for the 19% reduction in

cGMP levels previously observed in RPEC complexes (Chapter 3).

Comparable light-induced decreases in cGMP levels have been

demonstrated in the IS, ONL and OPL layers of rod

photoreceptors (Orr et al., 1976; de Azeredo et al., 1981) but

have not previously been observed in cones (De Vries et al.,

1979; Farber et al., 1981, 1982, 1983). In rabbit (Orr et al.,

1976) and frog rods (de Azeredo et al., 1981) the light

stimulated decreases in cGMP levels in IS and ONL layers

actually exceeded those within the outer segment.

The presence of a substantial and light-sensitive pool of

cGMP in the proximal layers of both rod and cone

photoreceptors suggests that appropriate cGMP levels in all

portions of the photoreceptor may be important to visual cell

function. The fact that cGMP levels are severely reduced

throughout the nontransducing rd cone photoreceptor cells

supports this idea. cGMP-gated ion channels, identical to

those found in the rod outer segment plasma membrane, have

been found in the inner segment membranes of rods (Matthews

and Watanabe, 1987, 1988; Watanabe and Matthews, 1988, 1990).

Thus, light-stimulated decreases in inner segment cGMP levels

could potentially alter local current flow into the inner










80
segment, modifying the electrical potential of the cell and

associated neurotransmitter release. In addition, Matthews et

al. (1985) and Cobbs and Pugh (1986) have shown that rod cGMP

can migrate from the inner segment into the outer segment in

seconds to tens of seconds. Inner segment levels of cGMP may

therefore be directly coupled to cGMP levels within the outer

segment. While cGMP migration occurs too slowly to influence

transduction activation, it may be involved in the recovery of

outer segment levels of cGMP as well as in processes

accompanying light and dark adaptation.

Similar to reports on the ground squirrel (De Vries et

al., 1979), we found light adaptation had no effect on cGMP

levels in the outer segment region of chicken cones. Whereas

these data may not seem in concordance with cGMP-mediated

transduction in cones, our finding of severely depressed

levels of cGMP in the nontransducing outer segments of rd

cones does indicate that adequate outer segment levels of cGNP

are necessary for normal cone function. Recent studies showing

that cones, like rods, possess 1) cGMP-gated ion channels in

their outer segment membranes (Haynes and Yau, 1985), 2) a

light sensitive current which is modulated by cGMP

concentration (Cobbs et al., 1985), and 3) all components

necessary for cGMP-meditated transduction (Hurwitz et al.,

1985; Lerea et al., 1986; Nathans et al., 1986), support the

hypothesis that cGMP is the second messenger in cones. It

therefore seems likely that light-stimulated decreases in










81

outer segment levels of cGMP mediate cone transduction but

that the physiological properties of cone cells make these

changes difficult to detect. The photoresponse in cones has

been shown to be 2-4 times faster and 50-100 times smaller

than that in rods (Schnapf and Baylor, 1987; Bader et al.,

1978; Schnapfand McBurney, 1980). The corresponding changes in

cone outer segment levels of cGMP may therefore occur too

rapidly or be of insufficient magnitude to be detected by our

method. The use of 1-2 hr light adaptation in the present and

in previous studies (De Vries et al., 1979) may have

additionally compromised our ability to detect light-

stimulated changes in outer segment cGMP levels since cones,

unlike rods, begin to repolarize during sustained illumination

(Normann and Werblin, 1974). We have previously demonstrated

in sighted chick RPEC's, significantly greater decreases in

cGMP levels following a 5 sec, intense light pulse (36%)

compared to 1-2 hr light adaptation (22%) (Chapter 2), a

result consistent with Normann and Werblin's observations. The

possibility that brief, intense light stimulation may achieve

measurable decreases in cone outer segment levels of cGMP

warrants investigation.

In the rd chick retina, photoreceptors are

morphologically normal and fully differentiated at hatch

(Ulshafer et al., 1984; Ulshafer and Allen, 1985a) yet lack

any electophysiological evidence of visual transduction

(Dawson et al., 1990; Ulshafer et al., 1984). We have










82

previously reported that cGMP levels in the rd retina fail to

increase during outer segment development and, on the day of

hatch, are reduced 85% and are insensitive to light adaptation

(Chapter 3). The present study shows that this deficiency in

cGMP occurs specifically within the photoreceptor cells of the

rd chick retina, since cGMP levels in the non-photoreceptor

layers of the retina and in other tissues (Chapter 3) are

comparable in sighted and rd chicks. Deficient cGMP levels may

underlie the electrophysiological and behavioral blindness

observed in this mutant by causing the permanent closure of

cGMP-gated ion channels in the outer (and inner) segment

membranes. If the cGMP-gated channels are closed in rd chick

photoreceptors, the photoreceptors would be expected to be

chronically hyperpolarized. This idea is supported by our

observations that 1) glucose utilization in dark-adapted rd

retinas is low, resembling that found in normal retinas

exposed to light (Chapter 1; Ruth et al., 1985), 2) pigment

granules in the PE of dark-adapted rd chicks are chronically

dispersed, a condition found only in light-adapted normal

retinas (Chapter 7), and 3) levels of aspartate and glutamate,

excitatory amino acids believed to be neurotransmitters in

vertebrate photoreceptor cells, are elevated specifically in

the photoreceptor layers of rd retinas (Ulshafer et al.,

1990b), an accumulation consistent with reduced release of

neurotransmitter from hyperpolarized photoreceptor cells.











cAMP


In the present study, cAMP levels in the outer segment

region (PE-OS) of sighted chick cones were found to be

insensitive to light and lower than cAMP levels elsewhere in

the photoreceptor. Since levels of cAMP in rod outer segments

are similarly low and unresponsive to light adaptation (Orr et

al., 1976, De Vries et al., 1978), this suggests that outer

segment levels of cAMP may serve similar roles in rod and

cone photoreceptor cell types and thus argues against a

critical role for cAMP in the mediation of cone transduction.

Supportive of this suggestion is our finding in rd chicks that

cone outer segment levels of cAMP were not significantly

disrupted in association with defective transduction.

cAMP levels in the IS-ONL-OPL region of carrier chick

retinas were significantly reduced upon light adaptation.

Similar light-stimulated decreases in cAMP levels have been

shown to occur in the ONL and OPL layers of rod-dominant

species (Orr et al., 1976; De Vries et al., 1978), but have

not been detected in any layer of freeze-dried cone-dominant

ground squirrel retinas (De Vries et al., 1979). However,

since significant light-induced decreases in cAMP levels have

been observed in freshly-dissected whole ground squirrel

retinas (Farber et al., 1981, 1983), it is possible that a

light sensitive pool of cAMP exists within the proximal cone

cell layers of this cone-dominant retina. Several

neurotransmitters have been identified in the retina which










84
modulate levels of cAMP by interaction with receptors coupled

via GTP-binding proteins to adenylate cyclase. In chicken

retinal homogenates, adenosine (Paes de Carvalho and de Mello,

1980) and norepinephrine (Schwarz and Coyle, 1976) have been

shown to stimulate adenylate cyclase activity while dopamine

has been shown to stimulate (de Mello, 1978) as well as to

inhibit (Nowak et al., 1990) adenylate cyclase activity via

interaction with dopaminergic Dl and D2 receptor types,

respectively (Nowak et al., 1990). Since the OPL contains, in

addition to photoreceptor presynaptic terminals, the

postsynaptic terminals of horizontal cells and bipolar cells,

light-induced decreases in cAMP levels in the chick OPL may

occur in association with synaptic transmission. Consistent

with this hypothesis, Blazynski et al. (1990) have recently

shown that depolarizing agents increase OPL and INL levels of

cAMP in light-adapted rod-dominant retinas, yet have no effect

on levels in dark-adapted retinas. It has not yet been

established, however, which cell type in the OPL accumulates

cAMP.

cAMP have been implicated in the mediation of various

light-entrained events within photoreceptor cells, including

rod outer segment disk shedding (Besharse, 1982), melatonin

synthesis (Iuvone, 1990; luvone et al., 1990) and, in non-

mammalian vertebrates, movement of both rod and cone

photoreceptor cells (Dearry et al., 1990). In the chicken

retina, cAMP analogues have been shown to increase melatonin










85

levels and serotonin N-acetyltransferase activity, both of

which are elevated in the dark and inhibited by light

(reviewed in luvone, 1986). Our finding of high dark levels of

cAMP in IS-ONL-OPL layer of carrier retinas is consistent with

these findings and supports developmental studies indicating

that retinal melatonin is synthesized primarily within the

photoreceptors (Iuvone, 1990).

Interestingly, in rd chick retina, proximal photoreceptor

levels of cAMP were significantly altered by light adaptation

despite the absence of visual transduction. This suggests that

some information concerning environmental light levels may be

received by retinal photoreceptors through a pathway other

than the cGMP cascade. Light-induced changes have also been

observed in retinal levels of melatonin in rd chickens (4-10

weeks ph), even after substantial photoreceptor cell loss

(Pang et al., 1989). cAMP levels in the proximal photoreceptor

layers of the rd chick retina were increased in response to

light adaptation, a response opposite to that observed in

carrier retinas. One possible explanation for the opposite

light response of rd photoreceptor cAMP is that levels of cAMP

in nontransducing photoreceptors may by reflecting light

information received by the pineal gland that is normally

masked in transducing photoreceptors by changes in cAMP

associated with neurotransmitter release. Chick photoreceptor

levels of cAMP have the potential to receive light-dark

information from the pineal gland via the following









86
hypothetical sequence; 1) Upon light exposure, the amount of

melatonin secreted from the pineal gland is reduced (Binkley

et al., 1980). 2) This results in less melatonin bound to

melatonin binding sites in the tectum and other brain regions

(Dubocovich et al., 1989), altering their synaptic output.

3) The tectal output is received by the isthmooptic nucleus

(Crossland and Hughes, 1978), which in turn sends fibers to

amacrine cells in the retina (Crossland and Hughes, 1978).

4) Through direct centrifugal input or through amacrine cell-

amacrine cell interaction, the amount of dopamine released

from dopaminergic amacrine cells (Floren, 1979) is altered.

5) Dopamine, acting as a paracrine neuromodulator in the

retina (discussed by Dearry and Burnside, 1989), diffuses from

its site of release in the inner retina to the outer retina.

6) There, through interaction with nonsynaptic D1 or D2

receptors (Nowak et al., 1990) on the photoreceptor cells and

other neurons, reduced or elevated levels of dopamine,

respectively, result in decreased levels of cAMP in the IS-

ONL-OPL and INL layers of the rd retina. Pineal gland levels

of melatonin have been shown to be comparable in rd and

carrier chick at 4-10 weeks posthatch (Pang et al., 1989).

Thus, pineal-retinal interaction may be intact in the rd

retina at hatch. The light modulation of rd photoreceptor cAMP

levels may be associated with light-entrained events which

appear to be intact at this age, such as outer segment disc

shedding (Ulshafer and Allen, 1985a). The rd chick mutant is










87

a potential model for study of the extent to which visual

transduction is involved in the mediation of light-entrained

events within the eye.

In conclusion, 1) light adaptation significantly

decreases levels of both cGMP and cAMP in sighted chick cone

photoreceptor cells, 2) defective transduction in the rd chick

retina is accompanied by severely reduced and light-

insensitive levels of cGMP (but not cAMP) specifically within

the photoreceptor cells and, 3) despite the absence of visual

transduction, light adaptation significantly increased levels

of cAMP in rd chick photoreceptors, a light response opposite

to that observed in sighted chick photoreceptors. Our results

indicate that cGMP levels throughout the photoreceptor are

important to its function and support the notion that some

light-regulated events in the retina are signalled through

pathways other than those supporting visual transduction.















CHAPTER 5
HISTOCHEMICAL LOCALIZATION AND KINETIC PROPERTIES OF cGMP
AND cAMP PHOSPHODIESTERASES IN THE rd CHICKEN RETINA


Introduction


A congenital mutation in the rd chicken results in

blindness prior to onset of photoreceptor pathology in this

cone-dominant retina (Ulshafer et al., 1984). On the day of

hatch, rd chick photoreceptors are fully-differentiated and

possess normal ultrastructure (Ulshafer et al., 1985a). At

this time, however, both scotopic and photopic ERG responses

are virtually absent (Ulshafer et al., 1984), indicating that

visual transduction is defective in both rod and cone

photoreceptor cell types.

Visual transduction in rods and cones is believed to be

mediated by the light-stimulated hydrolysis of cGMP within the

photoreceptor outer segment (reviewed by Stryer, 1986; Pugh

and Cobbs, 1986). Thus, a defect in visual transduction might

be expected to result in abnormal photoreceptor levels of

cGMP, especially in response to light. Consistent with this

hypotheses, recent studies in our laboratory have shown that

defective transduction in rd chick photoreceptors is

accompanied by a photoreceptor-specific deficiency in cGMP

levels. On the day of hatch, rd chick photoreceptor levels of










89
cGMP are 80-90% lower than those in sighted heterozygous

chicks. In addition, unlike cGMP levels in sighted chick

photoreceptors, rd photoreceptor levels of cGMP fail to

decrease in response to light adaptation (Chapter 4).

Photoreceptor levels of cAMP, on the other hand, are

comparable in the central retinas of light-adapted sighted

heterozygous and rd chicks.

The presence of abnormally-low cGMP levels in rd chick

photoreceptors indicates either that insufficient levels of

cGMP are being synthesized and/or that excessive levels of

cGMP are being hydrolyzed. The present study examines cyclic

nucleotide hydrolysis in the rd chick retina. The

histochemical localization (at LM and EM levels) and kinetic

characteristics of cGMP phosphodiesterase (PDE) and cAMP PDE

enzymes were compared in the retinas of sighted heterozygote

chicks and rd chicks at 1-2 days posthatch, a time at which

photoreceptor morphology is normal in the rd retina.


Methods


Animals. Sighted carrier chicks, heterozygous for the rd

mutation and homozygous rd chicks were hatched from eggs

produced by a breeder colony at the University of Florida as

previously described (Chapter 2). For all experiments, chicks

were used at 1-2 days posthatch, a time prior to the onset of

photoreceptor degeneration in the rd retina (Ulshafer et al.,










90
1984). Vision in carrier chicks was verified by the pecking of

visual targets.


Histochemical Techniques


cGMP PDE and cAMP PDE activities were localized using the

cerium-lead double-capture agent method of Poeggel et al.

(1988). This method uses an exogenous 5'nucleotidase to cleave

the 5'GMP (5'AMP) reaction product of PDE into nucleoside and

phosphate. The liberated phosphate ions are precipitated first

by cerium and then by lead ions. The resulting metal-phosphate

complex can be visualized at the light microscope (LM) level

after staining or seen directly at the electron microscope

(EM) level as an electron-dense precipitate.

Chicks were sacrificed in room light (approximately 1600

lux) by decapitation, enucleated and the anterior segments

removed from the eyes with a razor blade. For LM

histochemistry, eyecups were cut into 5 mm2 pieces and fixed

in 2% paraformaldahyde, 0.2% glutaraldahyde containing 0.1 M

Na cacodylate buffer (pH 7.3) and 0.25 M sucrose at 4C for

30-45 min. Tissue was washed 4 x 10 min in the same buffer,

infiltrated with cacodylate buffers containing increase

percentages (14% then 20%) of sucrose, then rapidly frozen in

liquid nitrogen. Frozen sections (15 um) were cut in a -20C

cryostat (IEC-CTF Cryostat-Microtome), mounted on gelatin-

coated slides and either used immediately or stored at -70C

overnight. For EM localization, eyecups were bissected with









91

vitreous intact and fixed in the above media for 30 min at

4C. The vitreous was then removed from each eye and the

eyecups were fixed for an additional 30 min, washed 5 x 10 min

in buffer, and embedded in 2% agarose. Thick (40-50 um)

sections were cut on a vibratome, mounted on gelatin-coated

slides and used immediately or stored at 4C until use the

following day.

All sections were preincubated for 15 min at 24C at a

final pH of 7.5 in 80 mM tris maleate buffer, 0.25 M sucrose

and 1 mg/ml exogenous 5'nucleotidase (115 units/mg solid, from

Crotalus adamanteus Venom, Sigma). The sections were then

incubated for 10-30 min at 37C in the following media: 80 mM

tris maleate buffer, 3 mM cGMP (cAMP), 3 mM CeCl3, 2 mM MgCl2,

0.5 mg/ml snake venom, 0.25 M sucrose, final pH 7.5. After

incubation, sections were rinsed 3 x 5 min in 80 mM tris

maleate buffer (pH 7.4) containing 250 mM sucrose, placed in

5 mM lead citrate (made in the same buffer) for 5 min, then

rinsed in buffer 3 x 5 min. Light microscopic sections were

treated 1 min with 2% ammonium sulphide for visualization,

rinsed 3 x 5 min in dH20, and coverslipped with glycerine.

Sections for EM localization were postfixed in 2.5%

glutaraldahyde for 30 min, then in 1% Os04 for 10 min,

dehydrated in hexylene glycol and embedded in epon-araldite

resin using slide molds. The plastic-embedded sections were

removed from slides by rapid chilling in liquid nitrogen. They

were then mounted on plastic bullets and ultrathin sections










92

were cut using an ultramicrotome (LKB Ultratome III). Sections

were mounted on formvar-coated copper grids, carbon coated and

examined on a scanning transmission elecron microscope (STEM)

(Hitachi H-7000, Hitachi Ltd, Tokyo, Japan). Some sections

were analyzed in an energy-dispersive x-ray microanalysis

(EDX) unit (Kevex Delta V) interfaced with the STEM to

determine the ionic composition of the histochemical reaction

product.

To confirm cytochemical specificity, some sections were

incubated in the absence of substrate (cGMP or cAMP) or in the

presence of specific inhibitors of cyclic nucleotide PDE

activity, 2 mM 3-isobutyl-l-methylxanthine (IBMX) or 4 mM

theophylline, (Butcher and Sutherland, 1962; Chader et al.,

1974a). Endogenous 5'nucleotidase activity was investigated by

incubating some sections in media lacking snake venom.


Kinetic Analyses


Hydrolysis rates of cGMP PDE and cAMP PDE were measured

using the discontinuous radiochemical assay described by

Thompson and Appleman (1971). Chicks were sacrificed in room

light by decapitation, enucleated and the anterior segment,

vitreous, and pecten were removed from each eye. Retina-

pigment epithelium-choroid (RPEC) complexes were then isolated

and sonicated in 600 ul ice cold 40 mM Tris HC1 buffer

containing 5 mM MgCl2, 1 mM DTT, 0.1 M EDTA, and 0.1 M EGTA,

final pH 7.4. Homogenates were stored at -800C until assayed.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ERCUAW1OU_5240C8 INGEST_TIME 2012-02-20T22:34:33Z PACKAGE AA00009065_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES