Drosophila in Vision Research
The Friedenwald Lecture
William L. Pak
I n this article, I first discuss the initial motive behind
importantly for our purpose, because of its well-develdeveloping a genetic approach to study the photooped compound eyes. I began in the fall of 1966 by
transduction process using electroretinogram (ERG)writing to several people who I thought would be fadefective mutants of Drosophila. Then I describe how miliar with Drosophila to see if they had ever come
this approach is contributing to the understanding of
across blind mutants. The few replies I received, and
phototransduction events in Drosophila and how the even more tellingly the lack of them, convinced me
results obtained in Drosophila may also be applicable that any mutants that I might need would have to be
to our understanding of photoreceptor processes in
generated and isolated de novo. Thus we began a longmammalian systems. I then describe how the mechaterm mutagenesis program in which male flies were
nisms of synaptic transmission may also be amenable
mutagenized with the chemical mutagen, ethyl methane sulfonate, and the offspring were screened for
to study using ERG-defective mutants of Drosophila.
defects shown by ERG.4
Since I began working on the early receptor potential in the 1960s, the major focus of research in
This approach was greeted with much skepticism.
my laboratory has been elucidating phototransduction
One of the problems was that the initial assumption
and closely related photoreceptor events. Because the
that I made—that a series of proteins is involved in
early receptor potential is generated with a virtually
phototransduction—was not widely accepted at the
undetectable latency, it was thought that its generatime. Many believed that phototransduction was much
tion is closely linked to the phototransduction process,
too fast a process to involve many proteins. Another
about which little was known at the time.1"3 It soon
criticism was that any molecule important in the phobecame clear, however, that electrophysiologic analytotransduction process is likely to be important also
sis of the early receptor potential, although interestin vital processes of the body. Consequently, any mutaing, was not likely to lead to molecular understanding
tion in the gene encoding such a protein is likely to
of phototransduction events. Therefore, I took a combe lethal, precluding isolation of mutants in the gene.
pletely different approach to the study of photorecepStill another criticism was that photoreceptors of any
tor function. The approach that I eventually chose was
phototransduction-null mutants would probably dea genetic one. I assumed that the phototransduction
generate from disuse, making analysis very difficult,
process would involve a series of proteins. If this were
even if it were possible to isolate such mutants. Finally,
the case, the proteins would have to be encoded by
it was contended that even if everything worked out,
genes, and if a suitable organism were chosen, mutants
any results one might obtain would be so specific to
could be isolated in the genes. I reasoned that mutants
Drosophila that they would not be applicable to any
defective in the proteins involved in phototransducother system. Events of the past 30 years have shown
tion would have defective responses of the photorethat most of these criticisms were unfounded or inceptor and that they could be identified by ERG revalid. To address the last point specifically, it might
cording. The approach I was proposing was mutant
be noted that evolution implies conservation as well
screening. At the time, however, mutant screening was
as change. If evolution has occurred at all, any basic
largely restricted to prokaryotic systems, and this work
molecular processes should be conserved across speturned out to be one of the first attempts at genetic
cies. They need not be identical but, nevertheless, sufdissection of a neural function.
ficiently conserved to allow meaningful deduction of
mechanisms
across species.
I chose the fruit fly Drosophila melanogaster for this
work because of its well-studied genetics and, just as
From the Department of Biological Sciences, Purdue University, West Lafayette,
Indiana.
Reprint requests: William L. Pak, Department of Biological Sciences, Purdue
University, Lilly Hall of Life Sciences, Stair. Street, West Lafayette, IN 47907.
2340
ERG-DEFECTIVE MUTANTS
It turned out that ERG-defective mutants could be
isolated at the rate of about 2 to 3 mutants per 1000
Investigative Ophthalmology & Visual Science, November HW.1"), Vol. 3f>, No. 12
Copyright © Association for Research in Vision and Ophthalmology
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2341
Drosophila in Vision Research
chromosomes screened in ethyl methane sulfonate
rnutagenesis. During a period of many years, we isolated 200 to 300 mutants that fell into more than 60
complementation groups by screening for the ERG
phenotype. In the meantime, Benzer's group performed a mutagenesis program to screen for "behavioral mutants."' Some of the X-chromosome mutants
they isolated by phototaxis were allelic to some of the
ones we had isolated.(> In addition, Heisenberg7
screened for optomotor-defective mutants, and Koenig and Merriam8 for ERG-defective autosomal mutants. Some of the mutants these investigators isolated
were also allelic to some of the ones we obtained.
Figure 1 illustrates the ERG phenotypes of three
classes of the mutants isolated in our screen. Because
a defect in the ERG does not necessarily mean that
phototransduction events are affected, an analysis of
the ERG, supplemented by whatever other histologic,
optical, and spectrophotometric information was
available, was used to classify the mutants and to iden-
wild type —
nonA
ninaA
norpA
| io mv
10 sec
Or
Or
Or
stimulus
FIGURE l. Representative mutant electroretinogram (ERG)
phenotypes. ERGs elicited from three representative mutants are compared with that of wild type (top trace). The
nonA ERG (second trace) lacks the on- and off-transients,
the ninaA ERG (third trace) lacks the prolonged depolarizing afterpotential, and the norpA ERG (fourth trace) shows
no response to light stimuli. The stimulus (bottom trace)
consists of a series of five -4-sec light pulses given in the
sequence: orange, blue, blue, orange, orange. Or = orange;
B = blue.
TABLE
l. Classification of Mutants
Phenotypes
Rhodopsin deficient
Small or no receptor potentials
Retinal degeneration
Altered time course of
receptor potential decay
Defective responses of secondorder neurons
Complementation
Groups (Genes)
10
15
12
14
31
tify those that were likely to be defective in phototransduction. Most, although not all, of the mutants could
be tentatively classified into several overlapping classes
in this manner (Table 1). In Table 1, mutants of the
first class are those that are deficient in rhodopsin
content. These mutants were identified from a severe
reduction in, or the absence of, the prolonged depolarizing afterpotential (PDA) in the ERG." The PDA
is generated when a strong, colored stimulus converts
a substantial amount (>20%) of rhodopsin to metarhodopsin. If the rhodopsin content is low, the
amount of metarhodopsin that can be photoconverted is also low, no matter how strong the stimulus,
and the PDA is not generated. The ninoE gene, which
encodes the major class of opsin, which I will discuss
later, was identified from a complementation group
in this class of mutants.10 The second class of mutants
listed in Table 1 are those with small or no receptor
potential—a phenotype readily identifiable in the
ERG. Many of the genes identified by this class of
mutants are likely to be important in phototransduction. The norpA gene, encoding phospholipase C
(PLC) involved in phototransduction, which I will also
discuss, was identified from one of the complementation groups comprising this class of mutants. Many
mutants of the second class and some mutants of the
first class display photoreceptor degeneration. These
mutants and other mutants displaying a similar phenotype have been grouped together to form the third
class. The last class, those with "defective responses
of the second-order neurons," is of interest because
mutations of this class have little or no effect on the
responses of photoreceptors. Instead, they severely reduce or eliminate the on- and off-transients of the
ERG originating from the laminar neurons, on which
Rl-6 photoreceptor endings synapse. Thus it was hypothesized that, in many of these mutants, synaptic
transmission between Rl-6 photoreceptors and their
target neurons in the lamina is blocked.
Drosophila was one of the first metazoans for which
modern molecular techniques became available, and
it is still one of few metazoans to which a wide variety
of modern experimental techniques can be readily
applied." As a consequence, our original objective,
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
2342
TABLE 2.
Gene
Genes Identified by ERG-Defective Mutants Cloned to Date (1/95)
Protein Product
ninaE
Rl-6 opsin
norpA
Phospholipase C
ninaC
PK-myosin I
ninaA
Cyclophilin
trp
Light-activated Ca++
channel (?)
RNA/DNA-binding
protein
Phosphatidylinositol
transfer protein
Protein kinase C
Histidine decarboxylase
Diacylglycerol kinase
nonA
rdgB
inaC
hdc
rdgA
InaD
?
Gene Cloned by
Mutant Reported by
9
Stephenson et al (1983)
Scavardaetal (1983)10
Paketal (1970)"'
Hotta and Benzer (1970)6
Pak (1979)1(i
Stephenson et al (1983)''
Pak (1979)l(i
Larrivee et al (1981)l8
Cosens and Manning (1969)21
Paketal (1970)M
Hotta and Benzer (1970)G
Hotta and Benzer (1970)°
Pak (1979)lfi
Burgetal (1993)27
Hotta and Benzer (1970)°
Pak (1979)16
molecular elucidation of phototransduction and related events, began to be realized. Now many laboratories are studying ERG-defective mutants of Drosophila. Approximately one dozen genes corresponding to
the mutants identified in our original screens have
now been cloned (Table 2). Some of these mutants
were also isolated by others, as indicated in Table 2.
One of the genes listed, trp, is the only one in that
table that was first identified from a spontaneously
occurring mutant line displaying defective light-dependent behavior.12 Several new alleles of this mutation, however, did appear in our screen. In these cloning efforts, the availability of mutants allowed the identification, localization, and cloning of the genes
carrying the mutations, and mutants also provided
clues to the potential functions of the proteins encoded by the genes.
In the rest of this article, I discuss four genes,
ninaE, norpA, ninaA, and hdc, that have been cloned
and analyzed. The protein products of the first two
genes are direcdy involved in phototransduction. The
protein product of the third gene, ninaA, although
not directly involved in phototransduction, is essential
for transport/maturation of the major subclass of rhodopsin. The protein product of the fourth gene, hdc,
catalyzes the synthesis of transmitter released by the
photoreceptors.
THE Rl-6 OPSIN GENE, ninaE
The ninaE gene was thefirstgene to yield to the molecular genetic approach described above. It was identified from the "rhodopsin deficient" class of mutants
l2
OTousaetal (1985)
IS
Zuker etal (1985)
Bloomquist et al (1988)l5
Montell and Rubin (1988)17
Shiehetal (1989)l9 20
Schneuwly et al (1989)
Montell and Rubin (1989)22
Wong etal (1989)23
Jones and Rubin (1990)24
Vihtelicetal (1991)25
Smith etal (1991)26
Burg et al (1993)27
Masai et al (1993)28
Shieh and Niemeyer (1995)29
listed in Table 1. These mutants were named nina
(neither inactivation nor afterpotential) based on
their ERG phenotype.9'13 The nina mutants available
at the time fell into five complementation groups, ninaA, B,. . . E (Fig. 2). Because in all of these mutants
the amount of rhodopsin in the majority class of photoreceptors, Rl-6, is drastically reduced, it seemed possible that one of the five complementation groups
might correspond to the structural gene for opsin in
Rl-6 photoreceptors. Two tests were used to identify
which, if any, of the five complementation groups
might correspond to the gene encoding Rl-6 opsin:
cell-class specificity of mutations and gene-dosage effect.
The Drosophila compound eye contains three
classes of photoreceptors, each containing a different
visual pigment: the numerically preponderant class,
Rl-6, and two minor classes, R7 and R8. If opsins in
different classes of photoreceptors are encoded by different genes, one would expect that mutations in the
gene encoding Rl-6 opsin would only affect Rl-6 photoreceptors, and not R7 or R8. To test cell-class specificity, rhodopsin molecules were visualized in freezefracture electron microscopy,14 and the PDA was elicited from the ultraviolet-sensitive R7 photoreceptors
using ultraviolet stimuli.9 These tests established unequivocally that only the mutations in the ninaA and
E complementation groups are specific for Rl-6 photoreceptors. The gene-dosage effect was tested by examining the dependence of the amount of Rl-6 rhodopsin expressed on the number of copies of the wildtype allele of ninaAXb or E.10 Results showed that the
amount of Rl-6 rhodopsin covaries with the dosage
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2343
Drosophila in Vision Research
Rl-6 RHODOPSIN
CONTENT (%)
100
wild type
ninaC?238
32 ±6
13±4
18 + 11
3±1
ninaEP332
FIGURE 2. Electroretinograms (ERGs) and rhodopsin content of nina mutants. ERGs recorded fromfivenina mutants, each from a different nina complementation group, ninaA,
B, C, D, or E, are compared with that of wild type. To the right of each ERG trace is an
approximate rhodopsin content in Rl-6 photoreceptors given as a percentage of the wildtype level. The stimulus sequence consists of three 4-second blue light pulses (solid bars
just below the top ERG trace) followed by three 4-second orange light pulses (open bars)
given at 30-second intervals. A 5-mV calibration pulse precedes each response. (Modified
from Figure 1 and Table 2 of Stephenson et al9 and reproduced with permission from
Cambridge University Press.)
of ninaE+ but not with ninaA+, suggesting that ninaE
is probably the structural gene for Rl-6 opsin. The
ninaE gene was subsequently localized to 92B6-7 of
the third chromosome by deficiency mapping.16
The ninaE gene was cloned17'18 by low-stringency,
interspecies hybridization to bovine rod opsin cDNA,
which had been cloned by then.32 Sequence analysis
of the gene revealed that it encodes a protein of 373
amino acids containing seven hydrophobic domains
that could be interpreted as transmembrane domains,
as had been shown for bovine and human opsins.32'33
Although the overall sequence identity with mammalian opsins is only about 36%, the protein contained
several local stretches of strikingly high sequence identity. The protein also contained all other known molecular hallmarks of opsin, including the presumed
site of chromophore attachment, a lysine residue near
the middle of the seventh transmembrane domain;
potential phosphorylation sites near the C-terminus,
consisting of several serine and threonine residues;
and a potential N-glycosylation site near the N-terminus. The NinaE protein was diefirstinvertebrate opsin
to be elucidated molecularly and thus provided evi-
dence for the essential structural identity between vertebrate and invertebrate opsins.
The first indication that mutation in a rhodopsin
gene might affect the photoreceptor structure appeared in the study of the Drosophila mutant oraJKS4
(outer rhabdomere absent) .6 In this mutant, the rhabdomeres of Rl-6 photoreceptors are well in the process of degeneration already at edosion and disappear
rapidly thereafter (Fig. 3).31-3437 Even in the early
1980s, ora/K84 was recognized as an allele of ninaE.10
However, there was also evidence that ora/KS4 acted as
an allele of another group of mutants. O'Tousa et al31
showed definitively that orafKS4 is a double mutant with
lesions in ninaE and another closely linked gene, and
that the lesion in ninaE is solely responsible for its
morphologic defect. The mutation in die ninaE gene
was isolated and called "nma£ w ." 31 Subsequently it
was shown diat many other mutations in the ninaE
gene, which result in the alteration of the protein
product, can cause degeneration of photoreceptors
with allele-dependent variations in the severity and
time course of degeneration.37
Meanwhile, it was discovered that some patients
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2344
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
FIGURE 3. Photoreceptor degeneration in ninaE mutant.
Transverse sections of ninaE1'3*2 mutant ommatidia
(C and D) are compared with
those of wild type (A and B)
at 0 day and 3 weeks after
eclosion. A and C are from 0day-oldflies,and B and D are
from 3-week-oldflies.All sections were taken from the
level of Rl-6 photoreceptor
cell nuclei. Rhl, Rh2, and so
on: Rhabdomeres of photoreceptors Rl, R2, and so
forth. Scale bar = 1.0 fim.
(Modified from Figures 2
and 6 of Leonard et al,37 and
reproduced with permission
from John Wiley and Sons,
Inc.)
with retinitis pigmentosa have mutations in the rhodopsin gene,38"40 paralleling the discovery in Drosophila that mutations in the rhodopsin gene cause photoreceptor degeneration. Many mutations in the rhodopsin gene associated with dominant retinitis
pigmentosa have now been reported (reviewed in Berson41). Approximately 25% to 30% of dominant retinitis pigmentosa is associated with mutations in the rhodopsin gene.41 In contrast, only one case of recessive
retinitis pigmentosa associated with a defect in the
rhodopsin gene has been identified.4^ The finding,
however, is significant in that die expected effect of
this mutation on the opsin protein is almost identical
to that of ninaEom of Drosophila.^ In both cases, the
mutation results in a stop codon that would prematurely truncate the opsin protein in the third cytoplasmic loop eliminating the sixth and seventh transmembrane domains (Fig. 4), and, in both cases, individuals homozygous for the mutation have degeneration of photoreceptors, suggesting an unexpected
degree of conservation in the mechanism of certain
forms of retinal degeneration.
However, ninaEmutations initially identified were
all recessive, probably because recessive mutations are
generally more common and, in Drosophila but not in
humans, any recessive mutation can be made homozygous to reveal its phenotype. Recendy, dominant ninaE mutations that cause photoreceptor degeneration
have been reported."14'45 Some of the missense mutations identified would result in the substitutions of
exactly the same conserved amino acid residues as
those reported for human dominant retinitis pigmentosa (Fig. 4)>444S again suggesting remarkable
conservation of the mechanisms of degeneration in
the two species.
PHOSPHOLIPASE C GENE, norpA
Isolation of mutants in the norpA gene was first reported independently by Pak et al14 and by Hotta and
Benzer.6 These were among the first ERG-defective
mutants to be generated by chemical mutagenesis,
and the mutants ultimately provided the most compelling evidence available that PLC is involved in invertebrate photo transduction.
The effects of norpA mutations on die ERG are
very striking (Fig. 5). Strong mutant alleles of norpA
essentially abolish the photoreceptor response and
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Drosophila in Vision Research
2345
Cytoplasmic
350
l
ARRP:
E249
:E
249-stop|
nmoE onl :Q251-stop
Dros: PI 841
ADRP: P171L
Dros: E194K
ADRP: E181K
Dros: G195S
ADRP: G182S
FIGURE 4. Rhodopsin gene mutations resulting in retinal degeneration in both human and
Drosophila. The protein alteration expected from a human rhodopsin gene mutation identified in a family with recessive autosomal retinitis pigmentosa, E249 — Stop,'12 is shown at
the corresponding Drosophila amino acid position. The Drosophila ninaE""' mutation would
result in Q251 — Stop.43 Both mutations thus would truncate the opsin protein at the third
cytoplasmic loop, and both are associated with recessive photoreceptor degeneration. The
remaining five human mutations are associated with autosomal dominant retinitis pigmentosa (reviewed in Berson"), and the resulting amino acid changes are shown at the
conserved Drosophila amino acid positions. Drosophila mutations at the identical codons also
cause dominant retinal degeneration.4'1'15 ARRP, autosomal recessive retinitis pigmentosa;
ADRP, autosomal dominant retinitis pigmentosa; Dros, Drosophila.
render the fly blind (reviewed in Paklb), suggesting
that the mutations affect an important component of
the phototransduction machinery. It was soon shown
that the norpA gene does not encode rhodopsin and is
not likely to encode a component of the light-activated
channel.4h However, the identity of the norpA-enco&ed.
protein remained elusive for some time. Yoshioka et
al47 showed that PLC activity, which is normally highly
concentrated in the eye of Drosophila, is greatly reduced in norpA mutants, providing for the first time
a potential link between norpA mutations and PLC
activity in the eye. Selinger and Minke48 extended the
findings of Yoshioka et al47 and showed that, in a temperature-sensitive norpA mutant, the PLC activity is
normal at permissive temperatures but blocked at restrictive temperatures, thus implying that the norpA
gene might encode PLC.
In the meantime, evidence began appearing to
indicate that invertebrate photoreceptors might employ PLC in phototransduction.48"'2 Drawing on the
available evidence and information from other PLCmediated signaling systems,53 Fein49 first proposed
that invertebrate phototransduction involves an inositol phospholipid-based cascade of events. In this hy-
pothesis, photoexcitation of rhodopsin activates an
inositol phospholipid-specific PLC through a G protein, and the activated PLC, in turn, catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate
(PIP2) to diacylglycerol and inositol 1,4,5-trisphosphate (IP3), with IP3 acting as the second messenger
for both excitation and adaptation. The hypothesis
was supported by several lines of electrophysiologic,
biochemical, and pharmacologic evidence. For example, pressure injection of IPS into the ventral photoreceptor of Limulus was found to mimic the effects of
light,54'55 and light-dependent activation of PLC has
been demonstrated in several invertebrate eye preparations.51 f)25b A major deterrent to wide acceptance
of the hypothesis, however, was that the calcium buffer ethyleneglycol-bis-(/?-aminoethylether)-N,N,N',N'tetraacetic acid did not have the same effect on IP ;r
mediated excitation as that mediated by light.57'58
The norpA gene was cytogenetically mapped to the
4B6-4C1 region of the X chromosome by deficiency
mapping5' and cloned by a combination of chromosomal walking and transposon tagging.15 Conceptual
translation of the open reading frame yielded a protein of 1095 amino acids. It was possible to compare
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Investigative Ophthalmology 8c Visual Science, November 1995, Vol. 36, No. 12
2346
Wild Type
•V-
10 mV T
2m
10 sec
norpAP24
norpApn
49-
norpAP16
norpA™
the offspring, definitively establishing that a lesion in
the cloned gene is solely responsible for the norpA
phenotype.6869 Similarly, the initial identification of
the NorpA protein as PLC was based purely on sequence homology with mammalian PLCs. Since then,
Toyoshima et al70 biochemically purified the major
PLC in the eye of Drosophila and showed that sequences of several peptide fragments generated from
it completely matched those found in the deduced
NorpA sequence.15 Together these results leave little
doubt that the gene cloned by Bloomquist et al15 is
the gene identified by norpA mutants and that the
gene encodes a major functional PLC in the eye of
Drosophila. Because strong mutations in this gene abolish the response of the photoreceptor, the conclusion
seems inescapable that the no^A-encoded PLC is a
critical component of the phototransduction machinery in Drosophila.
W
Or
Or
FIGURE 5. Comparison of norpA electroretinograms (ERGs)
with that of wild type. ERGs elicited from three norpA mutants are compared with that recorded from wild type. The
severity of mutant ERG phenotype is allele dependent. norpA
responses are often greatly prolonged and far outlast the
stimuli. The prolonged time course of norpA responses is
illustrated at slower sweep speeds on the right-hand side of
the traces, with breaks in the traces indicated by double S's.
The stimulus consists of a white light pulse followed by two
orange pulses, all lasting 4 seconds. W, white; Or, orange.
the deduced NorpA protein sequence with mammalian inositol phospholipid-specific PLC sequences because several mammalian protein sequences had just
become available from cDNA cloning at about the
same time.60"63 These mammalian PLCs had two restricted regions of high-sequence identity, called domains X and Y, and could be classified into three
classes, a, /3, and y on the basis of primary structure
and immunologic properties.64 As Figure 6 shows, the
NorpA protein closely resembles the ft class PLC in
structure and has high-sequence identity with mammalian PLCs within domains X and Y (Fig. 6).
The original identification that the cloned gene,
indeed, is the norpA gene heavily relied on a transposon-induced norpA mutant allele.1566 The mutant allele, isolated in transposon mutagenesis, was shown to
harbor the transposon sequence in the coding region
identified in the cloned cDNA,'5 and remobilization
of the transposon was found to revert the mutant phenotype to wild type.66 Since that time, even more rigorous proof has been obtained. Introduction of the
cloned cDNA, driven by a suitable promoter, such as
the ninaE promoter, into the germline of a norpA null
mutant by P-element-mediated transformation67 was
shown to completely rescue the mutant phenotype of
Immunolocalization of the NorpA protein completely supports this view.66 Immunofluorescence
staining of adult head sections with a polyclonal antiserum generated against the NorpA protein showed that
the staining is confined mainly to the rhabdomeres of
the compound eyes and ocelli. Moreover, subcellular
localization of the protein by immunogold electron
microscopic analysis showed that the labeling is found
Extended Dom X
Extended Dom V
NorpA
60%
61%
Bov PLCI
SH2
49%
SH2
SH3
39%
47%
Bov PLC Y,
Bov PLC 5,
FIGURE 6. Comparison of the NorpA protein with three bovine inositol phospholipid-specific phosopholipase C (PLC):
PLC/?i, 71, and <5|.hO~fa3 The NorpA protein most closely resembles the 0 class PLC in overall structure. Sequence identity within domains X and Y as well as extended domains X
and Y is also the highest between the NorpA protein and
the P class bovine PLC. Domains X and Y are two small,
most highly conserved regions among inositol phospholipidspecific PLCs,64 and extended domains X and Y are more
extended regions of homology.b5 The larger numbers over
each bovine PLC represent sequence identities between the
NorpA protein and the particular PLC within extended domains X and Y, and the smaller numbers are corresponding
identities within domains X and Y. PLCys have src homologous sequences not found in other classes of PLC in the
region between domains X and Y
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2347
Drosophila in Vision Research
in the rhabdomeres of both the compound eye and
ocelli in close association with the rhabdomeric membrane but not within the membrane. Because the
rhabdomere corresponds to the outer segment of the
vertebrate photoreceptor and is a specialized structure
for light reception and phototransduction, the localization of the NorpA protein just described within this
structure is exactly what one would expect if the
NorpA protein played a central role in phototransduction.
Extended DomX
Extended DomY
Pomx|
[pom Y
Bov Ret p| „ i. NH
(PLC W '
76%
65%
Dro NorpA
52%
61%
62%
63%
Bov PLCP,
53%
MAMMALIAN HOMOLOGS OF THE
NORPA PROTEIN
In contrast to invertebrate phototransduction, phototransduction in vertebrate photoreceptors is based on
G-protein-dependent activation of cyclic guanosine
monophosphate phosphodiesterase (see reviews in
references 71 to 73, and no role has been determined
for PLC in vertebrate phototransduction. Thus there
seemed to be no reason to suppose that a NorpA-like
protein would be present in vertebrate photoreceptors. Yet there have been persistent reports of lightactivated PLC activity in vertebrate photoreceptors
(see reviews in Jelsema and Alexrod74 and Anderson
and Brown75).
To determine whether NorpA-like, retina-preferentially expressed PLCs might be present in mammals,
probes derived from norpA cDNA were used to screen
bovine retinal cDNA libraries. Analysis of positively
hybridizing clones isolated in the screen led to the
identification of two major, alternatively processed isozymes of bovine PLC with novel properties.76 The
structure of the two isozymes, bovine retinal PLC/5,
and /?„, is shown schematically at the top of Figure 7.
The shorter isozyme, /3U, which is about 900 amino
acids long, shares its entire protein sequence with the
larger form. Northern blot analysis revealed that the
cognate mRNAs are expressed predominantly in the
retina with a minor expression in the cerebellum. No
expression was detected in any other tissue tested.
Figure 7 shows the sequence comparison between the
new bovine retinal PLCs and representative other
known PLCs. It may be seen that sequence identity
within domains X and Y or within extended domains
X and Y, which are more extended regions of homology among known PLCs,fi5 is more than 10 percentage
points higher with the NorpA protein than any other
PLCs. Together these results suggest that these PLCs
are bovine homologs of the NorpA protein.
Distribution of the new PLCs within the retina
was examined by immunocytochemical analysis using
polyclonal antisera generated against poorly conserved regions of the proteins.79 Results showed an
intense band of staining in the photoreceptor layer
just distal to the outer limiting membrane, staining of
60%
Dro plc-21
52%
Hum PLCp,
34%
37%
Bov PLC y.
38%
Bov PLC 5,
FIGURE 7. Sequence comparison of bovine retinal PLCs with
seven other inositol phospholipid-specific PLCs. The bovine
retinal PLCs are shown at the top. Two major isozymes,
types I and II, generated by alternative RNA splicing, are
indicated. In addition, each isozyme can have two alternative
forms, a or b, depending on whether a 12-amino acid region
is present between domains X and Y. The bovine retinal
PLCs display more than 10 percentage points higher sequence identity with the NorpA protein than with any other
PLC within domains X and Y or extended domains X and
Y. The larger numbers over each PLC represent sequence
identities between the NorpA protein and the particular
PLC in the extended domains X and Y, and the smaller
numbers are corresponding identities in domains X and Y.
Bovine Ret/?,,,,, bovine retinal PLC type I and II; Drosophila
NorpA, Drosophila NorpA protein; bovine PLC/?i, bovine
brain PLC/?6062; Drosophila?LC-21, Drosophila PLC expressed
in the central nervous system65; human PLC/?2> human
PLC/?2 from HL-60 cell line77; bovine PLCy,, bovine brain
PLCy,61-63; bovine PLC<5,, and bovine brain PLG5,.02 (Modified from Figure 2 in Ferreira and Pak,78 and reproduced
with permission from Plenum Press.)
some cells of the inner nuclear layer and ganglion
cells, but a conspicuous lack of staining in the rod
outer segment and outer nuclear layers (Fig. 8A).
Careful examination of staining in the photoreceptor
layer, using sections that had been mechanically disturbed to disperse the cells, showed that both the inner and outer segments of cones, but no rods, are
stained (Fig. 8B).
Protein that probably corresponds to one of the
PLC isozymes identified in the present work has been
biochemically purified from the bovine retina80 and
cerebellum.81 Sequences of several peptides generated
from it match those contained within the protein sequence deduced by Ferreira et al,7b indicating the
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2MB
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
B
ROSROS
COS
COS
ONL-
GCL-
FIGURE 8. Immunocytochemical localization of bovine retinal
phosophipase C (PLC) in bovine retina. (A) Bovine retinal
section immunostained with a polyclonal anti.sen.im generated
against the bovine retinal PLCs and viewed in Nomarski optics.
ROS, rod outer segments; ONL, outer nuclear layer; OPL, outer
plexiform layer; INL, inner nuclear layer; IPL, inner plexiform
layer; GCL, ganglion cell layer. Curved arrows point to stained
cells in the outer margin of the inner nuclear layer. Scale bar,
70 (xm. (B) Photoreceptor layer of a mechanically disturbed
retinal section stained with the antisenim and viewed in Nomarski optics. ROS, rod outer segments; COS, cone outer segments;
CIS, cone inner segments. Scale bar, 30 fj,m. (Reproduced from
Ferreira and Pak,™ with permission from the Journal of Biological
Chemistry.)
identity of these proteins. These authors have named
invertebrate phototransduction involves a PLC-medithe new protein "PLC-/34."
ated cascade. Moreover, the combined work of many
Several conclusions may be drawn from these reinvestigators now working in this field is beginning to
sults. First, PLCs highly homologous to the NorpA
provide an outline of the entire process, although the
protein are, indeed, present in the mammalian photopicture is still incomplete, particularly in steps after
receptors, even though one might have thought that
the cleavage of PIP2 by PLC. I will not try to present
the NorpA protein is a specialized protein present
a complete account of the current status of Drosophila
only in invertebrate photoreceptors. Second, the fact
phototransduction in this article. More complete acthat the new PLCs are present in cone outer segments
counts may be found, for example, in recent reviews
raises the possibility that they may somehow be inby Selinger et alH4 and by Ranganathan et al.Hft
volved in cone phototransduction. Although verteFigure 9 presents a working model of the Drosophbrate phototransduction has been studied extensively,
ila phototransduction cascade. In this model, lightmost of the information has come from rods. The
excited rhodopsin activates a G protein, presumably
present results suggest that much more needs to be
by catalyzing the exchange of guanosine triphosphate
learned about signal transduction in cones. Finally,
for guanosine diphosphate on its a subunit. The actistrong mutations in the norpA gene of Drosophila, in vated G , in turn stimulates PLC to catalyze the hydrolyt
addition to blocking phototransduction, cause lightsis of PlPa, thus generating two potential second mes2
dependent degeneration of photoreceptors." '^ The
sengers, diacylglycerol and inositol 1,4,5-trisphospresence of NorpA-homologous PLCs only in cones
phate, IP3. Inositol 1,4,5-trisphosphate releases Ca2+
suggests that mutations in the gene encoding these
from the subrhabdomeric cisternae (SRC) lying at the
PLCs may cause specific degeneration of cones in
base of the microvilli comprising the rhabdomere.
mammals.
Isolation and characterization of the ninaE and
norpA genes, encoding two main components of the
pathway to this point, opsin and PLC, have already
PHOTOTRANSDUCTION CASCADE
been described. Two genes encoding eye-specific G
Largely because of the critical piece of evidence proprotein subunits have been isolated from Drosophila,
vided by norpA mutants, it is now widely accepted that one (dgq) encoding the a subunit8'1 and the other
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2349
Drosophila in Vision Research
^
SRC
L
ePKC
*
—
^L
....T)?
^
< - *
—
Microvillui — '
>- Ca**-perm. chonnel
Ifrp?)
h- cation chann
Bx fwc#ll vl w
rtoimo Membrane — " ^ ^
FIGURE 9. Working model of phototransduction cascade in
Ihosofihila. See text for details. Rh, rhodopsin; G, G protein;
PLC, inositol phospholipid-specific phospholipase C; PIP.,,
phosphatidylinositol 4,5-bisphosphate; ePKC, eye-specific
protein kinase C; PI-TP, phosphatidylinositol transfer protein; PI, phosphatidylinositol; IP,, inositol 1,4,5-trisphosphate; IPiR, inositol 1,4,5-trisphosphate receptor; SRC,
subrhabdomeric cisternae; \w, photon.
(d(w) encoding the (3 subunit.H7 The a subunit is most
homologous to the vertebrate G(|(, proteins involved
in activation of PLC, consistent with its proposed role
as activator of the NorpA protein. Phenotypes of mutants isolated in the dgq gene have also been reported
to be consistent with this role of the protein product.Hr''KH
From electrophysiologic studies of the ERG-defective mutant trjinm (Table 2), Hardie and Minke"" obtained evidence that there are two classes of lightactivated channels in Drosophila photoreceptors: one
permeable primarily to Ca"'', and the other permeable
to cations nonselectively. The Ca"''-permeable channel is hypothesized to be encoded by the trp gene.""
Thus IPi binds to the IP:, receptor on the SRC membrane and releases Ca"'+ from the intracellular stores,
SRC. This event leads to the opening of both classes
of channels. Minke and Selinger" suggested that the
IP:i receptor on the SRC membrane and the ir^cncoded channel on the plasma membrane directly interact to open the latter in a manner similar to the
interaction between the ryanodine receptor and the
dihydropyridine receptor in skeletal muscles.""' Alternatively, depletion of intracellular Ca"' stores could
trigger the Ca"' -permeable channel on the plasma
membrane through an unknown second messenger
(see review by Clapham"). Thus the entry of Ca"^
through the trpenco&cd channel replenishes the Ca" +
stores in SRC and is responsible for the steady-state
phase of the receptor potential. The non-<r/>-encoded
cation channel is responsible for the transient phase
of the receptor potential. Its molecular identity and
the pathway leading to it have not yet been elucidated.
The other product of PIPL> hydrolysis, diacylglycerol, appears to take part in adaptation through the
activation of photoreceptor-specific protein kinase C
(ePKC). The gene for this substance was cloned from
a pool of eye-preferentially expressed genomic clones
obtained by a subtractive hybridization protocol'" and
was shown to correspond to inaC mutants"' (Table 2)
by Smith et al."1' Hardie et al"' showed that adaptation
is severely impaired in an inaC null mutant, suggesting
that the m«Gencoded photoreceptor-specific protein
kinase C is an important component of adaptation.
Another protein shown in Figure 9 is that encoded by
the rdg gene11 (Table 2). Vitehlic et al!Hi isolated the
rdgB gene and showed that it encodes a Ca"*-binding
integral membrane protein with significant sequence
identity with a rat brain phosphatidyl inositol transfer
protein in its amino terminal domain. Although the
protein is expressed in photoreceptors as well as the
antennae and the brain, in photoreceptors it is localized in the region of the SRC. Thus the protein may
function in transporting phospholipids and proteins
used in phototransduction from the SRC to the rhabdomeres to replenish their rhabdomeric stores.
Much is now known about inactivation of metarhodpsin and regeneration of rhodopsin in Drosophila,'n'm a process not illustrated in Figure 9. As in the
case of vertebrates,'''1'"0 metarhodopsin inactivation
involves phosphorylation and arrestin binding. Regeneration of rhodopsin requires its dephosphorylation
by a protein phosphatase encoded by the rdgC
gene."7"" Two arrestin genes have been isolated from
Drosophila: arrestin 1 (phosrestin l)'""'" w and arrestin
2 (phosrestin 2)."""' r ' Subsequently, mutations have
been isolated in both these genes.'" The rdgC gene
was cloned using mutants isolated in a screen for those
showing retinal degeneration. ""•'"•' Byk et al"7 showed
that metarhodopsin undergoes rapid phosphorylation
upon photoconversion from rhodopsin, decreasing its
ability to activate the G protein, and binding of arrestin results in further quenching of metarhodopsin
activity. Moreover, arrestin protects metarhodopsin
from dephosphorylation by the rdgC'rencoded protein
phosphatase. Once metarhodopsin is photoconverted
to rhodopsin, however, arrestin is released and rhodopsin is regenerated through dephosphorylation by
the RdgC protein. By generating mutants in the arrestin genes, Dolph et al"* demonstrated the critical
role of arrestin in metarhodopsin inactivation in vivo.
Another important insight to emerge from these studies is the origin of the PDA as persistent activation
resulting from an inadequate supply of arrestin to inactivate metarhodopsin completely when the latter is
produced in large amounts.'17""
In addition, several other Drosophila genes encoding proteins more indirectly involved in phototransduction than the proteins just described have also
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2350
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
been identified. For example, I discuss in the next
section the ninaA gene encoding a photoreceptor-specific cyclophilin important in opsin transport and folding. The ninaC gene9'16 (Table 2) encodes two closely
related unconventional myosins each consisting of two
linked domains: one homologous to protein kinases
and the other to the myosin heavy-chain head.17 Noting that normal localization of calmodulin in photoreceptors depends on these proteins, Porter et al107 have
suggested that they may function to control the subcellular distribution of calmodulin, which is normally
highly concentrated in the rhabdomeres and is probably important in phototransduction and other photoreceptor processes. The rdgA gene6 (Table 2) encodes
an eye-specific diacylglycerol kinase,28 which catalyzes
the conversion of diacylglycerol to phosphatidic acid
in inositol phospholipid metabolism.
I hope that this brief and incomplete overview
makes it clear that much progress has been made in
the understanding of invertebrate phototransduction
in the past 10 years largely through the use of Drosophila and Drosophila mutants. It is also important to note
that almost every gene direcdy or indirectly involved
in phototransduction, when defective, causes retinal
degeneration. Thus it has been reported that mutations in the ninaE, norpA, trp, ninaC, rdgA, rdgB, rdgC,
and arr2 genes all cause degeneration of photoreceptors (reviews by Pak83 and by Ranganathan85). In almost all cases, the degeneration is either completely
or partially light dependent. The observation suggests
that much of light-dependent retinal degeneration
may be associated with defects in phototransduction.
PHOTORECEPTOR-SPECIFIC
CYCLOPHILIN GENE, ninaA
As with the ninaE gene, ninaA was identified as one
of the complementation groups in the "rhodopsindeficient" class of mutants (Table 1; Fig. 2). The study
of the ninaA gene first provided evidence that rhodopsin requires another protein, a cyclophilin, for its normal expression. As previously discussed, mutations in
this gene along with those in ninaE were found to
affect the rhodopsin level only in Rl-6 photoreceptors.9'18'35 The ninaE gene encodes Rl-6 opsin, accounting for die specificity of ninaE mutations for Rl6 rhodopsin. The molecular basis of specificity of ninaA mutations for Rl-6 rhodopsin is less clear.
The ninaA gene was localized to the 21D1-2 to
21E1-2 region of die second chromosome by deficiency mapping1820 and cloned both by chromosomal
walking20 and by using eye-preferentially expressed
clones mapping to the region, obtained by a subtractive hybridization protocol.19 The gene was found
to encode a 237-amino acid protein with strong sequence homology to cyclophilin.19'20 Unlike previously
identified mammalian cyclophilins, however, the NinaA protein is an integral membrane protein108 that
uses a putative transmembrane domain at its C-terminus20 to anchor to the membrane. The gene is expressed specifically in all classes of adult photoreceptors, but the protein product is required by only a
subset of rhodopsin present in Rl-6 photoreceptors
and, possibly, ocelli, hence the observed specificity of
ninaA mutations for Rl-6 rhodopsin.108
Cyclophilins are members of a highly conserved
family of proteins that are thought to be the primary
cellular targets of the potent immunosuppressive
drug, cyclosporine A, used to suppress graft rejection
and to treat immune disorders (reviewed in Heitman
et al109 and in Sigal et al110). Because the enzyme,
peptidyl prolyl cis-trans isomerase, known to accelerate
protein folding in vitro, was found to be a cyclophilin," 1112 it was hypodiesized that the NinaA protein
acts on rhodopsin as a folding catalyst.19'20108 The
findings that ninaA mutations inhibit opsin transport
from the endoplasmic reticulum and that the NinaA
protein is present throughout the secretory pathway,
including the endoplasmic reticulum, were viewed as
consistent with this hypothesis."3 More recently,
Baker et al1 H showed that NinaA forms a specific stable
complex with Rl-6 rhodopsin in vivo and suggested
that it functions as a chaperon or as both a chaperon
and foldase for Rl-6 rhodopsin.
If a Drosophila rhodopsin requires a cyclophilin,
an important question is whether mammalian rhodopsins also similarly require cyclophilins. However, more
than a century of work on mammalian rhodopsins
provided no clue that they might require such a protein. Stamnes et al108 reported on a bovine protein that
cross-reacts widi NinaA antibodies, but the protein is
not retina-specifically expressed and was never characterized. Accordingly, NinaA cDNA was used as probe
to isolate bovine cDNA clones that are expressed specifically in the retina." 5 Analysis of these clones identified two NinaA-related proteins that are generated by
alternative RNA splicing. The C-terminal region
shared by these two isozymes contains a cyclophilin/
peptidyl prolyl cis-trans isomerase domain that is
highly homologous to other known cyclophilins. Otherwise, they are unlike any other cyclophilin characterized to date. To begin with, these bovine proteins are
large for cyclophilins (~250 and >1000 amino acids
compared with <200 amino acids for most other
cyclophilins), characterized by the presence of extended domains N-terminal to the cyclophilin domain. Moreover, Northern blot analysis showed diat
the cognate messages are expressed specifically in the
retina. Within the retina, however, both the RNA and
protein products are found in photoreceptors as well
as other retinal neurons of the inner nuclear and ganglion cell layers. Among photoreceptors, however,
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Drosophila in Vision Research
the proteins are found primarily in cones. The bovine cyclophilin constructs expressed in vitro displayed both peptidyl prolyl as-trans isomerase and
cyclosporine A-binding activities which, however, were
strongly reduced compared with other known mammalian cyclophilins.
Thus mammalian retinas do contain retina-specifically expressed cyclophilins. In fact, these bovine
cyclophilins are the only retina-specifically expressed
cyclophilins, other than the NinaA protein, identified
to date. Although their overall structure is very different from that of the NinaA protein, their specific expression in the retina indicates their involvement in
retina-specific functions. In addition, their cone-preferential localization among photoreceptors suggests
that one of these functions might be specific interactions with cone opsins as chaperons or foldase during
cone opsin maturation, as has been proposed for the
NinaA protein with a subset of Drosophila opsin.
SYNAPTIC TRANSMISSION
Although the original intended use of the ERG-defective mutants was to elucidate phototransduction and
related photoreceptor events, many of the isolated
mutants were defective in processes other than phototransduction (Table 1). Thus the use of ERG-defective
mutants need not be restricted to phototransduction.
The synaptic process, in particular, appears amenable
to an approach similar to that applied to phototransduction using another class of ERG-defective mutants.
The insect ERG is dominated by a corneal-negative component corresponding to the light-induced
depolarization of photoreceptors, but it also contains
components that arise postsynaptic to photoreceptors.
The on- and off-transients of the ERG (Fig. 10), in
particular, have long been thought to originate from
the first optic ganglion, the lamina (reviewed by Goldsmith and Bernard" 7 and by Pak118). Coombe"9 provided the most direct evidence for the cellular origin
of these components using the Drosophila mutant Vam,
in which the large monopolar neurons, LI and L2, in
the lamina degenerate. He showed that, in this mutant, the on- and off-transients disappear concomitantly with degeneration of LI and L2, implicating
these cells as the site of origin of these ERG components. LI and L2 are the major synaptic targets of the
Rl-6 photoreceptors.120121
The last mutant class, labeled "defective responses of the second-order neurons" in Table 1, consists of those mutants in which the on- and off-transients are reduced or missing but the receptor component of the ERG is normal or nearly normal (Fig. 10).
The existing mutants of this class (called "transientdefective" mutants) fall into about 30 complementation groups (Table 1). Because saturation mutagenesis
2351
Wild type
Transient-defective
mutant
Intracellular
recording
Receptor
A\ f
component-' « '
FIGURE 10. Schematic representation of synaptic connections
in the lamina and comparison of responses elicited from a
transient-defective mutant and wild type. Shown on the lefthand side of the figure is a schematic representation of
synaptic connections between Rl-6 photoreceptors and LI
and L2 neurons in the lamina. The top two traces on the
right compare the receptor potentials recorded intracellularly from the Rl-6 photoreceptors of wild-type and transient-defective mutant flies. The bottom two traces are a
similar comparison of electroretinograms (ERGs). Rl-6, Rl6 photoreceptors; LI and L2, large monopolar neurons 1
and 2; AM, amacrine cell. (The schematic drawing on the
right was modified from Figure 39 of Nassel,"1' and reproduced with permission of the author and Pergamon Press.)
has never been carried out, it is almost certain that
there are more mutants of this class still to be isolated.
In view of the site of origin of the ERG on- and offtransients, previously discussed, a reasonable hypothesis is that in these mutants, synaptic transmission between Rl-6 photoreceptors and their target neurons
in the lamina is blocked or defective (Fig. 10). The
reason for the many complementation groups presumably reflects the fact that synaptic transmission is
a complex process requiring many proteins. Some of
these mutations may affect synaptic transmission indirecdy, and the corresponding proteins are not integral
components of the machinery of synaptic transmission. Even if a sizable fraction of the 30 complementation groups consists of such mutations, the number
remaining still would seem large enough to begin a
serious attempt at a molecular genetic analysis of synaptic transmission.
Since the mid-1980s, evidence began accumulating to indicate that histamine is probably the major
transmitter used by invertebrate photoreceptors. Immunocytochemical, biochemical, and physiologic
lines of evidence have shown that, in several invertebrate species, high levels of histamine are present in
the photoreceptor and synaptic layers, that it is synthesized there, and that exogenous application of histamine to postsynaptic neurons mimics the effects of
light.122"129 As a first step toward use of this class of
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Investigative Ophthalmology 8c Visual Science, November 1995, Vol. 36, No. 12
2352
2.5
tt> -«•
-i-i
•
2.0
1.5
E g
B 1 1.0
«/> ^
re •— 0.5
-
m-
r
T
LnfTlr
-
00
wild
type
J]
•
eya chrom. 1
mutants
chrom. 2
mutants
chrom. 3
mutants
FIGURE 11. Screening transient-defective mutants for defective histamine synthesis in the head. Using isolated heads,
29 transient-defective mutants, each from a different complementation group, were screened for a defect in their
ability to synthesize and accumulate histamine from exogenously supplied radiolabeled histidine. The amounts of histamine detected in the assay are shown as a percentage of
exogenously supplied histidine. The mutant identified as
defective in histamine synthesis is shown as afilledbar. Wildtype flies and eya mutants were used as controls, eya mutants
lack the compound eyes but retain the ocelli. Histamine
synthesis in the ocelli accounts for the relatively high levels
of histamine synthesis still remaining in eya. (Unpublished
results obtained by M. Burg and V. Sarthy.)
mutants, we decided to look for those that are defective in the synthesis of histamine by screening the
existing mutants. If our collection of transient-defective mutants represented a sufficiently wide array of
proteins involved in synaptic transmission, we reasoned, there was a chance that the synthetic enzyme
for histamine might be represented in it.
Histamine is synthesized from histidine catalyzed
by the pyridoxal phosphate-dependent enzyme, histidine decarboxylase. Thirty transient-defective mutants, each from a different complementation group,
were tested for histamine synthesis by examining the
fraction of exogenously applied histidine converted to
histamine in isolated head preparations.130 The screen
identified a single complementation group consisting
of four members with very low levels of histamine synthesis (Fig. 11). These mutants, when heterozygous,
displayed HDC activity levels intermediate between
those of wild type and homozygotes, suggesting that
the HDC activity level is gene dosage dependent. The
gene-dosage dependence, in turn, implied that the
gene corresponding to this complementation group
probably is the structural gene for HDC.
The putative Drosophila HDC gene was cloned130
by low-stringency, interspecies hybridization using rat
HDC cDNA131 as probe. It encodes an 847-amino-acid
protein, the N-terminal 476-amino-acid region of
which displays 62% and 60% identity with the corresponding regions of human132 and rat HDC,131 respec-
tively. The Drosophila protein is larger than the mammalian HDCs (847 compared with 662 and 655 amino
acids), and its C-terminal 370 amino acid region has
no significant homology to any known protein. The
gene is expressed most abundantly in the retina, but
it is also expressed in discrete regions of the brain and
thoracic ganglia. The proof of identity between the
gene that was cloned and the one that was identified
by the mutants was obtained by P-element-mediated
germline rescue of the mutants with cloned genomic
DNA (Fig. 12) or cloned cDNA driven by a heat-shock
promoter.
Thus mutations in the gene encoding HDC block
histamine synthesis and synaptic transmission between
photoreceptors and laminar neurons, providing
strong genetic support for the hypothesis that Drosophila photoreceptors use histamine as their major transmitter. Even more importantly for our purpose, the
work provided direct evidence that at least one of the
complementation groups identified by ERG transientdefective mutants does, indeed, correspond to a gene
encoding a protein involved in synaptic transmission.
Moreover, because the gene was identified simply by
screening the existing transient-defective mutants, the
B
10mV
500 msec
stimulus
FIGURE 12. P-element-mediated transformation rescue of a
histidine decarboxylase mutant. Electroretinograms (ERGs)
recorded from a wild-type fly (A), an hdc mutant (B), and
a sibling transgenic fly carrying a genomic fragment containing the hdc gene in an hdc mutant background (C).
(Unpublished results by M. Burg.)
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
Drosophila in Vision Research
~30 complementation groups, identified by this class
of mutants, may well correspond to a sufficiently representative array of proteins involved in synaptic transmission for fruitful use of these mutants in a study
of the synaptic process. In recent years, remarkable
progress has been made in identifying proteins involved in the molecular machinery of secretion, including transmitter release (reviewed in references
133 to 135). However, identification of these proteins,
as impressive as it is, still represents only one aspect
of the study of synaptic transmission. Much work still
needs to be done. Drosophila mutants, such as the transient-defective mutants, could be used to make a valuable contribution to our understanding of this important process of the nervous system.
SUMMARY
A genetic approach involving the use of ERG-defective
Drosophila mutants, described in this article, was developed with the explicit goal of elucidating phototransduction and related photoreceptor events at the molecular level. Advances in the past 30 years have made
this goal a reality. Although the phototransduction
process in Drosophila is not yet fully understood, enormous strides are being made by many investigators
now working in this field. In addition to elucidating
phototransduction events in invertebrate photoreceptors, works of these investigators are providing insights
into the functional machinery of signaling systems in
general. For example, mobilization of Ca2+ and replenishment of intracellular Ca2+ stores after their
depletion is a current topic of intense interest (see,
for example, references 93, 136, and 137). Study of
trp mutants is providing insights into this process.84'91
Similarly, the mechanisms of inactivation of G-protein
coupled receptors appear to be highly conserved in
diverse transduction systems, and inactivation of Drosophila metarhodopsin is proving a valuable model system for this study.8598
The identification of the NinaA protein19'20 and
its role as chaperon and/or foldase for opsin would
not have been possible without the genetic approach
described in this article. In fact, study of the NinaA
gene is providing fresh insights not only into the rhodopsin maturation process but also into the role of
cyclophilins in general.113'114 Moreover, recent results
suggest that related retina-specific proteins are also
present in mammals.115 As valuable as the NinaA protein has been, it may be only the first of a number of
such novel proteins or novel isozymes of known proteins required in the retina to be discovered through
this approach.
For two of the Drosophila proteins discussed in this
article, NorpA and NinaA, related mammalian proteins expressed in photoreceptors have been identi-
2353
fied, even though it seemed unlikely that such proteins would have any role in mammalian photoreceptors.70'"5 A recent report suggests that the Trp protein
has human homologs expressed most heavily in the
brain (C. Montell, cited in reference 93). It may well
be that almost any protein identified in the Drosophila
retina has its counterpart(s) in mammals. Discovery
of these mammalian homologs of Drosophila proteins
will probably spur new lines of investigation in mammalian photoreceptor function. For example, because
no role had been assigned to PLC in cone phototransduction, the discovery that NorpA-homologous PLCs
are present in cone outer segments79 raises questions
about what role(s) these PLCs might play in signal
transduction in cone outer segments.
Most Drosophila genes encoding proteins directly
or indirectly involved in phototransduction, when mutated, cause photoreceptor degeneration. These include ninaE, norpA, trp, ninaC, rdgA, rdgB, rdgC, and
arr. Because it appears likely that the protein products
of these genes have counterparts in mammalian photoreceptors and because many of the basic mechanisms of photoreceptor degeneration are probably
conserved across species (see discussion on rhodopsin
gene mutations), I suggest that any gene that, when
defective, causes photoreceptor degeneration in Drosophila should be seriously considered a potential
"identifier" of human candidate genes for retinitis
pigmentosa. It should be remembered that it was first
discovered in Drosophila that mutations in the rhodopsin gene cause photoreceptor degeneration.10
Finally, the use of ERG-defective mutants need
not be limited to the study of photoreceptor function.
Synaptic transmission, in particular, appears well
suited for investigation using ERG-defective mutants.
Synaptic transmission is a basic and universal property
of the nervous system. It might be thought, at first
sight, that any mutation that blocks this process would
be lethal to the organism. However, many proteins
important in the eye are encoded by genes expressed
specifically or preferentially in the eye. Consequently,
mutations in these genes need not be lethal to the
organism. It was this property of the eye that enabled
ready isolation of mutants defective in photoreceptor
function. Similarly, many candidate mutants defective
in synaptic transmission have been identified among
the ERG-defective mutants isolated because the synaptic process in question occurs entirely within the eye.
Thus it should be possible to explore synaptic transmission as a next extension of the genetic approach
based on the use of ERG-defective mutants.
Acknowledgments
I gratefully acknowledge the contributions of past and present collaborators, students, associates, and laboratory personnel. I thank Martin Burg and Vijay Sarthy for allowing
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2354
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
me to present previously unpublished figures. I thank Lydia
Randall, Chenjian Li, and Mark O'Neil for help in preparing
figures, and Ann Pellegrino for help in preparing the manuscript.
19.
References
1. Brown KT, Watanabe K, Murakami M. Early and late
receptor potentials of monkey cones and rods. Cold
Spring Harb Symp Quant Biol. 1965; 30:457-482.
2. Cone RA. Early receptor potential of the vertebrate
eye. Cold Spring Harb Symp Quant Biol. 1965; 30:483492.
3. Pak WL. Some properties of the early electrical response in the vertebrate retina. Cold Spring Harb Symp
Quant Biol. 1965; 30:493-499.
4. Pak WL, Grossfield J, White NV. Nonphototactic mutants in a study of vision of Drosophila. Nature.
1969;222:351-354.
5. Benzer S. From gene to behavior. JAMA. 1971;
218:1015-1022.
6. Hotta Y, Benzer S. Genetic dissection of the Drosophila
nervous system by means of mosaics. Proc Natl Acad
Sri USA. 1970;67:1156-1163.
7. Heisenberg M. Isolation of mutants lacking the optomotor response. Drosophila Information Service. 1971;
46:68.
8. KoenigJ, Merriam JR. Autosomal ERG mutants. Drosophila Information Service. 1977; 52:50-51.
9. Stephenson RS, O'Tousa J, Scavarda NJ, Randall LL,
Pak WL. Drosophila mutants with reduced rhodopsin
content. In: Cosens DJ, Vince-Price D, eds. The Biology
of Photoreception. Cambridge: Cambridge University
Press; 1983:477-501.
10. Scavarda NJ, O'Tousa J, Pak WL. Drosophila locus with
gene-dosage effects on rhodopsin. Proc Natl Acad Sri
USA. 1983; 80:4441-4445.
11. Rubin GM. Drosophila melanogaster as an experimental
organism. Srience. 1988;240:1453-1459.
12. O'Tousa JE, Baehr W, Martin RL, Hirsh J, Pak WL,
Applebury ML. The Drosophila ninaEgene encodes an
opsin. Cell. 1985; 40:839-850.
13. Zuker CS, Cowman AF, Rubin GM. Isolation and structure of a rhodopsin gene from D. melanogaster. Cell.
1985;40:851-858.
14. Pak WL, Grossfield J, Arnold KS. Mutants of the visual
pathway of Drosophila melanogaster. Nature. 1970; 227:
518-520.
15. Bloomquist BT, Shortridge RD, Schneuwly S, et al.
Isolation of a putative phospholipase C gene of Drosophila, norpA, and its role in phototransduction. Cell.
1988;54:723-733.
16. Pak WL. Study of photoreceptor function using Drosophila mutants. In: Breakefield X. Neurogenetics: Genetic Approaches to the Nervous System. New York: Elsevier-North Holland; 1979:67-99.
17. Montell C, Rubin GM. The Drosophila ninaC locus encodes two photoreceptor cell specific proteins with
domains homologous to protein kinases and the myosin heavy domain head. Cell. 1988;52:757-772.
18. Larrivee DC, Conrad SK, Stephenson RS, Pak WL.
Mutation that selectively affects rhodopsin concentra-
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
tion in the peripheral photoreceptors of Drosoj)hila
melanogaster. J Gen Physiol. 1981; 78:521 -545.
Shieh B-H, Stamnes MA, Seavello S, Harris GL, Zuker
CS. The ninaA gene required for visual transduction
in Drosophila encodes a homologue of cyclosporin Abinding protein. Nature. 1989;338:67-70.
Schneuwly S, Shortridge, RD, Larrivee DC, Ono T,
Ozaki M, Pak WL. Drosoj)hila ninaA gene encodes an
eye-specific cyclophilin (cyclosporine A binding protein). Proc Natl Acad Sri USA. 1989; 86:5390-5394.
Cosens DJ, Manning A. Abnormal retinogram from a
Drosophila mutant. Nature. 1969;224:285-287.
Montel C, Rubin GM. Molecular characterization of
the Drosophila trp locus: A putative integral membrane
protein required for phototransduction. Neuron.
1989;2:1313-1323.
Wong F, Schaefer EL, Roop BC, LaMendolaJN, Johnston-Seaton D, Shao D. Proper function of the Drosophila trp gene product during pupal development is
important for normal visual transduction in the adult.
Neuron. 1989; 3:81-94.
Jones KR, Rubin GM. Molecular analysis of no-on-transient A, a gene required for normal vision in Drosophila.
Cell. 1990; 4:711-723.
Vihtelic TS, Hyde DR, O'Tousa JF. Isolation and characterization of the Drosophila retinal degeneration B
(rdgB)gene. Genetics. 1991; 127:761-768.
Smith DP, Ranganathan R, Handy RW, Marx J, Tsuchida T, Zuker C. Photoreceptor deactivation and retinal
degeneration mediated by a photoreceptor-specific
protein kinase C. Srience. 1991;254:1178-1181.
Burg MG, Sarthy PV, Koliantz G, Pak WL. Genetic
identification and cDNA cloning of a Drosophila histidine decarboxylase gene and its requirement for photoreceptor
transmitter
synthesis.
EMBO
J.
1993;12:911-919.
Masai I, Okazaki A, Hosoya T, Hotta Y. Drosophila retinal degeneration A gene encodes an eye-specific diacylglycerol kinase with cysteine-rich zinc-finger motifs
and ankyrin repeats. Proc Natl Acad Sri USA.
1993:90:11157-11161.
Shieh B-H, Niemeyer B. A novel protein encoded by
the inaD gene regulates recovery of visual transduction in Drosophila. Neuron. 1995; 14201-210.
Pak WL, Conrad SK, Kremer NE, Larrivee DC, Schinz
RH, Wong F. Photoreceptor function. In: Siddiqi O, Babu
P, Hall LM, Hall JC, eds. Development and. Neurobiobgy of
Drosophila. New York: Plenum Press; 1980:331 -346.
O'Tousa JE, Leonard DS, Pak WL. Morphological defects
in ord™ photoreceptors caused by mutation in Rl-6 opsin gene of Drosophila. J Neurogenet. 1989; 6:41-52.
Nathans J, Hogness DS. Isolation, sequence analysis
and intron-exon arrangement of the gene encoding
bovine rhodopsin. Cell. 1983;34:807-814.
Nathans J, Hogness DS. Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc
Natl Acad Sri USA. 1984;81:4851-4855.
Harris WA, Stark WS, Walker JA. Genetic dissection
of the photoreceptor system in the compound eye of
Drosophila melanogaster. J Physiol (Lond). 1976; 256:415439.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2355
Drosophila in Vision Research
35. Schinz RH, Lo M-VC, Larrivee DC, Pak WL. Freezefracture study of the Drosophila photoreceptor membrane: Mutations affecting membrane particle density. JCell Biol. 1982:93:961-969.
36. Stark WS, Sapp R. Ultrastructure of the retina of Drosophila melanogaster: The mutant ora (outer rhabdomeres absent) and its inhibition of degeneration in rdgB (retinal
degeneration-B). J Neurogenet. 1987; 4:227-240.
37. Leonard DS, Bowman VD, Ready, DF, Pak WL. Degeneration of photoreceptors in rhodopsin mutants of
Drosophila. J Neurobiol. 1992; 23:605-626.
38. Dryja TP, McGee TL, Reichal E, et al. A point mutation of the rhodopsin gene in one form of retinitis
pigmentosa. Nature. 1990;343:364-366.
39. Dryja TP, Hahn LB, Cowley, GS, McGee TL, Berson
EL. Mutation spectrum of the rhodopsin gene among
patients with autosomal dominant retinitis pigmentosa. ProcNatlAcad Sci USA. 1991; 88:9370-9374.
40. Sung G-H, Davenport CM, Hennessey JC, et al. Rhodopsin mutations in autosomal dominant retinitis pigmentosa. ProcNatlAcad Sci USA. 1991;88:6481-6485.
41. Berson E. Retinitis pigmentosa: The Friedenwald Lecture. Invest Ophthalmol Vis Sci. 1993;34:1659-1676.
42. Rosenfeld PJ, Cowley GS, McGee TL, Sandberg MA,
Berson EL, Dryja TP. A null mutation in the rhodopsin
gene causes rod photoreceptor dysfunction and autosomal recessive retinitis pigmentosa. Nat Genet.
1992;l:209-213.
43. Washburn T, O'Tousa JE. Molecular defects in Drosophila rhodopsin mutants. / Biol Chetn. 1989; 264:
1564-1566.
44. Kurada P, O'Tousa JE. Retinal degeneration caused by
dominant rhodopsin mutations in Drosophila. Neuron.
1995; 14:571-579.
45. Colley NJ, Cassill JA, Baker EK, Zuker CS. Defective
intracellular transport is the molecular basis of rhodopsin-dependent dominant retinal degeneration.
Proc Natl Acad Sci USA. 1995;92:3070-3074.
46. Pak WL, Ostroy SE, Deland MC, Wu C-F. Photoreceptor mutant of Drosophila: Is protein involved in intermediate steps of phototransduction? Science. 1976;
194:956-959.
47. Yoshioka T, Inoue H, Hotta Y. Absence of phosphatidylinositol phosphodiesterase in the head of a Drosophila visual mutant, norpA (no receptor potential A ) . /
Biochem. 1985;97:1251-1254.
48. Selinger Z, Minke B. Inositol lipid cascade of vision
studied in mutant flies. In: Watson JD, Wigler M, Fera-
52.
53.
54.
55.
56.
57.
58.
59.
hydrolysis in fly photoreceptors. Proc Natl Acad Sci
USA. 1987; 84:6939-6943.
Baer KM, Saibil HR. Light- and GTP-activated hydrolysis of phosphatidylinositol bisphosphate in squid photoreceptor membranes./Biol Chem. 1988;263:17-2O.
Berridge MJ, Irvine RF. Inositol trisphosphate, a novel
second messenger in cellular signal transduction. Nature. 1984;312:315-321.
Fein A, Payne R, Corson DW, Berridge MJ, Irvine RF.
Photoreceptor excitation and adaptation by inositol
1,4,5-trisphosphate. Nature. 1984;311:157-160.
Brown JE, Rubin LJ, Ghalayini AJ, et al. wyo-inositol polyphosphate may be a messenger for visual excitation in
Limulus photoreceptors. Nature. 1984; 311:160-163.
Szutz EZ, Wood SF, Reid MS, Fein A. Light stimulates
the rapid formation of inositol trisphosphate in squid
retinas. Biochem J. 1986; 240:929-932.
Rubin LJ, Brown JE. Intracellular injection of calcium
buffers blocks IP3-induced but not light-induced electrical responses of Limulus ventral photoreceptors. Biophysj. 1985; 47:38a.
Payne R, Corson DW, Fein A. Pressure injection of
calcium both excites and adapts Limulus ventral photoreceptors. / Gen Physiol. 1986; 88:107-126.
Banga SS, Bloomquist BT, Brodberg RK, et al. Cytogenetic characterization of the 4BC region on the X
chromosome of Drosophila melanogaster: Localization
of the mei-9, norpA and omb genes. Chromosoma.
1986; 93:341-346.
60. Katan M, Kriz RW, Totty N, et al. Determination of
the primary structure of PLC-154 demonstrates diversity of phosphoinositide-specific phospholipase C activities. Cell. 1988; 54:171-177.
61. Stahl ML, Ferenz CR, Kelleher KL, Kriz RW, Knopf
JL. Sequence similarity of phospholipase C with the
non-catalytic region of src. Nature. 1988; 332:269-272.
62. Suh PG, Ryu SH, Moon KH, Suh HW, Rhee SG. Cloning and sequence of multiple forms of phospholipase
C. Cell. 1988:54:161-169.
63. Suh PG, Ryu SH, Moon KH, Suh HW, Rhee SG. Inositol phospholipid-specific phospholipase C: Complete
cDNA and protein sequences and sequence homology
to tyrosine kinase-related oncogene products. Proc
Natl Acad Sci USA. 1988; 85:5419-5423.
64. Rhee SG, Suh P-G, Ryu S-H, Lee SY. Studies of inositol
phospholipid-specific phospholipase C. Science. 1989;
244:546-550.
65. Shortridge RD, Yoon Y, Lending CR, Bloomquist BT,
miscoj, eds. Cold Spring Harbor Symposium on QuantitaPerdew MH, Pak WL. A Drosophila phospholipase C
tive Biology. Vol53: The Molecular Biology of Signal Transgene that is expressed in the central nervous system.
duction. Cold Spring Harbor, NY: Cold Spring Harbor
J Biol Chem. 1991;266:12474-12480.
Laboratory; 1988:333-341.
66. Schneuwly S, Burg MG, Lending C, Perdew MH, Pak
49. Fein A. Excitation and adaptation of Limulus photoreWL. Properties of photoreceptor-specific phospholiceptors by light and inositol 1,4,5-trisphosphate.
pase C encoded by the norpA gene of Drosophila melanoTrends Neurosci. 1986;9:110-114.
gaster. J Biol Chem. 1991;266:24314-24319.
50. Payne R. Phototransduction by microvillar photore67. Rubin GM, Spradling AC. Genetic transformation of
ceptors of invertebrates: Mediation of a visual cascade
by inositol trisphosphate. Photobiochemistry and PhotoDrosophila with transposable element vectors. Science.
biophysics. 1986; 13:373-397.
1982:218:348-353.
51. Devary O, Heichal O, Blumenfield A, et al. Coupling
68. McKay RR, Chen D-M, Miller K, Kim S, Stark WS,
of photoexcited rhodopsin to inositol phospholipid
Shortridge RD. Phospholipase C rescues visual defect
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2356
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
Investigative Ophthalmology & Visual Science, November 1995, Vol. 36, No. 12
in norpA mutant of Drosophila melanogaster. JBiol Chem.
1995;270:1327l-13276.
Pearn M, Randall LL, Shortridge RD, Pak WL. Molec85.
ular and electrophysiological characterization of Drosof)hila norpA mutants. J Biol Chem. 1995; in press.
Toyoshima S, Matsumoto N, Wang P, et al. Purifica86.
tion and partial amino acid sequences of phosphoinositide-specific phospholipase C of Drosophila eye. J Biol
Chem. 1990;265:14842-14848.
87.
Pugh NEJr, Lamb TD. Cyclic GMP and calcium: The
internal messengers of excitation and adaptation in
vertebrate photoreceptors. Vision Res. 1990; 30:192388.
1948.
Stryer L. Visual excitation and recovery. J Biol Chem.
1991;266:107ll-107l4.
89.
Kaupp UB, Koch K-W. Role of cGMP and Ca2+ in
vertebrate photoreceptor excitation and adaptation.
90.
Annu Rev Physiol. 1992;54:153-175.
Jelsema CL, Axelrod A. Light activation of phospholipases in rod outer segments. In: Hudspeth A, Macleish
P, Margolis F, eds. Sensory Transduction (Discussions in 91.
Neuroscience). Vol. 4. Geneva: Foundations for the
Study of the Nervous System. 1987; 79-84.
Anderson RE, Brown JE. Phosphoinositides in the retina. In: Progress in Retinal Research. Oxford: Pergamon;
92.
1988;8:211-228.
Ferreira PA, Shortridge RD, Pak WL. Distinctive subtypes of bovine phospholipase C that have preferential
expression in the retina and high homology to the
norpA gene product of Drosophila. Proc Natl Acad Sci
93.
USA. 1993;90:6042-6046.
Park D, Jhon D-Y, Kriz R, Knopf J, Rhee SG. Cloning,
sequencing, expression, and G(|-independent activa94.
tion of phospholipase G # . J Biol Chem. 1992;
267:16048-16055.
Ferreira PA, Pak WL. Characterization of vertebrate
homologs of Drosophila photoreceptor proteins. In:
95.
Anderson RE, LaVail MM, HollyfieldJG, eds. Degenerative Diseases of the Retina. New York: Plenum Press;
1995:263-274.
96.
Ferreira PA, Pak WL. Bovine phospholipase C highly
homologous to the NorpA protein of Drosophila is expressed specifically in cones. J Biol Chem. 1994; 269:
3129-3131.
97.
Lee G-W, Park DJ, Lee K-H, Kim CG, Rhee SG. Purification, molecular cloning, and sequencing of phospholipase Op.JBiol Chem. 1993; 268:21318-21327.
98.
Min DS, Kim DM, Lee YH, Seo J, Suh P-G, Ryu SH.
Purification of a novel phospholipase C isozyme from
bovine cerebellum. / Biol Chem. 1993;268:1220712212.
99.
Meyertholen EP, Stein PJ, Williams MA, Ostroy SE.
Studies of the Drosophila norpA phototransduction mutant. II. Photoreceptor degeneration and rhodopsin
maintenance. / Comp Physiol. 1987; 161:793.
100.
Pak WL. Retinal degeneration mutants of Drosophila.
In: Wright A, Jay B, eds. Molecular Genetics of Inherited
Eye Disorders. Chur, Switzerland: Harwood Academic
Publishers; 1994:29-52.
101.
Selinger Z, Doza YN, Minke B. Mechanisms and genet-
ics of photoreceptors desensitization in Drosophila
flies. Biochim Biophys Acta. 1993; 1179:283-299.
Ranganathan R, Malicki DM, Zuker CS. Signal transduction in Drosophila photoreceptors. Annu Rev Neurosd. 1995;18:283-317.
Lee Y-], Dobbs MB, Verardi ML, Hyde DR. dgq: a Drosophila gene encoding a visual system-specific Ga molecule. Neuron. 1990; 5:889-898.
Yarfitz S, Niemi GA, McConnell JL, Fitch CL, Hurley JB.
A Gp protein in the Drosophila compound eye is different
from that in the brain. Neuron. 1991; 7:429-438.
Lee YJ, Shah S, Suzuki E, Zars T, O'Day P, Hyde DR. The
Drosophila dgq gene incodes a Ga protein that mediates
phototransduction. Neuron. 1994; 13:1143-1157.
Minke B, Wu C-F, Pak WL. Induction of photoreceptor voltage noise in the dark in Drosophila mutant.
Nature. 1975; 258:84-87.
Hardie RC, Minke B. The top gene is essential for a
light activated Ca2+ channel in Drosophila photoreceptors. Neuron. 1992;8:613-651.
Minke B, Selinger Z. Inositol lipid pathway in fly photoreceptors: Excitation, calcium mobilization and retinal degeneration. In: Osborne NN, Chader GJ, eds.
Progress in Retinal Research. Oxford: Pergamon;
1992:99-124.
Block BA, Imagawa T, Campbell KP, Franzini-Armstrong L. Structural evidence for direct interaction
between the molecular components of the transverse
tubule/sarcoplasmic reticulum junction in skeletal
muscle. / Cell Biol. 1988; 107:2587-2600.
Clapham DE. Replenishing the stores. Nature.
1995;375:634-635.
Schaeffer E, Smith D, Mardon G, Quinn W, Zuker C.
Isolation and characterization of two new Drosophila
protein kinase C genes, including one specifically expressed in photoreceptor cells. Cell. 1989;57:403-412.
Hardie RC, Peretz A, Suss-Toby E, et al. Protein kinase
C is required for light adaptation in Drosophila photoreceptors. Nature. 1993; 363:634-637.
Vihtelic TS, Goebl M, Milligan S, O'Tousa JE, Hyde
DR. Localization of Drosophila retinal degeneration B,
a membrane-associated phosphatidylinositol transfer
protein. / Cell Biol. 1993; 122:1013-1022.
Byk T, Bar-Yaacov M, Doza YN, Minke B, Selinger
Z. Regulatory arrestin cycle secures the fidelity and
maintenance of the fly photoreceptor cell. Proc Natl
Acad Sci USA. 1993;90:1907-1911.
Dolph PJ, Ranganathan R, Colley NJ, Hardy RW, Socolich M, Zuker CS. Arrestin function in inactivation
of G protein-coupled receptor rhodopsin in vivo. Science. 1993;260:1910-1916.
Liebman PA, Pugh EN. ATP mediates rapid reversal
of cyclic GMP phosphodiesterase activation in visual
receptor membranes. Nature. 1980;287:734-736.
Wilden U, Hall S, Kuhn H. Phosphodiesterase activation by photoexcited rhodopsin is quenched when
rhodopsin is phosphorylated and binds 48kDa protein. Proc Natl Acad Sci USA. 1986;83:1174-1178.
Steele FR, Washbum T, Rieger R, O'Tousa JE. Drosophila retinal degeneration C (rdgC) encodes a novel serine/
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017
2357
Drosophila in Vision Research
102.
103.
104.
105.
106.
107.
108.
109.
110.
111.
112.
113.
114.
115.
116.
117.
118.
119.
threonine protein phosphatase. Cell. 1992; 69:669676.
120.
Hyde DR, Mecklenburg KL, Pollock JA, Vihtelic TS,
Benzer S. Twenty Drosophila visual system cDNA
clones: One is a homolog of human arrestin. ProcNatl
AcadSci USA. 1990;87:1008-1012.
121.
Smith DP, Shieh B-H, Zuker CS. Isolation and structure of an arrestin gene from Drosophila. ProcNatl Acad
Sri USA. 1990;87:1003-1007.
122.
LeVine H, Smith DP, Whitney M, et al. Isolation of a
visual system specific arrestin: An in vivo substrate for
123.
light-dependent phosphorylation. Mech Dev. 1990;
33:19-26.
Yamada T, Takeuchi Y, Komori N, et al. A 49-kilodalton phosphoprotein in the Drosophila photoreceptor 124.
is an arrestin homolog. Srience. 1990; 248:483-486.
Steele F, O'TousaJE. Rhodopsin activation causes retinal degeneration in Drosophila rdgC mutant. Neuron.
1990; 4:883-890.
Porter JA, Yu M, Doberstein SK, Pollard TD, Montell
125.
C. Dependence of calmodulin localization in the retina on the NINAC unconventional myosin. Srience.
1993;262:1038-1042.
Stamnes MA, Shieh B-H, Chuman L, Harris GL, Zuker
126.
CS. The cyclophilin homolog ninaA is a tissue-specific
integral membrane protein required for the proper
synthesis of a subset of Drosophila rhodopsins. Cell. 127.
1991;65:219-227.
Heitman J, Mowa RN, Hall MN. Proline isomerase at the
crossroads of protein folding, signal transduction, and
immunosuppression. New Biology. 1992;4:448-460.
128.
Sigal NH, Dumont FJ. Cyclosporin A, FK-506, and rapamycin: Pharmacologic probes of lymphocyte signal
129.
transduction. Annu Rev Immunol. 1992; 10:519-560.
Fischer G, Wittmann-Liebold B, Lang K, Kiefhaber T,
Schmid FX. Cyclophilin and peptidyl-prolyl ris-trans
isomerase are probably identical proteins. Nature.
130.
1989;337:476-478.
Takahashi N, Hayano T, Suzuki M. Peptidyl-prolyl ristrans isomerase is the cyclosporin A-binding protein
131.
cyclophilin. Nature. 1989; 337:473-475.
Colley NJ, Baker EK, Stamnes MA, Zuker CS. The
cyclophilin homolog ninaA is required in the secretory pathway. Cell. 1991;67:255-263.
Baker EK, Colley NJ, Zuker CS. The cyclophilin homo132.
log NinaA functions as a chaperone, forming a stable
complex in vivo with its protein target rhodopsin.
EMBOJ. 1994; 13:4886-4895.
Ferreira PA, Horn JT, Pak WL. Retina-specifically ex133.
pressed novel subtypes of bovine cyclophilin. J Biol
Chem. 1995; in press.
134.
Nassel DR. Neurotransmitters and neuromodulators
in the insect visual system. Prog Neurobiol. 1991; 37:
135.
179-254.
Goldsmith TH, Bernard GD. The visual system of insects. In: Rockstein M, ed. The Physiology oflnsecta. Vol.
136.
2. 2nd ed. New York: Academic Press; 1974:166-272.
Pak WL. Mutations affecting the vision of Drosophila
melanogaster. In: King RC, ed. Handbook of Genetics. Vol.137.
3. New York: Plenum; 1975:703-733.
Coombe PE. The large monopolar cells LI and L2
are responsible for ERG transients in Drosophila. J
CompPhysiol [A]. 1986; 159:655-665.
Boschek CB. On the fine structure of the peripheral
retina and lamina ganglionaris of thefly,Musca domestica. ZZellforsch. 1971; 118:369-409.
Nicol D, Meinertzhagen IA. An analysis of the number
and composition of the synaptic populations formed by
photoreceptors of thefiy.JCompNeumL 1982; 207:29-44.
Elias MS, Evans PD. Histamine in the insect nervous
system: Distribution, synthesis and metabolism. J Neurochem. 1983; 41:562-568.
Hardie RC. Is histamine a neurotransmitter in insect
photoreceptors? / Comp Physiol [A]. 1987; 161:201213.
Nassel DR, Holmqvist MH, Hardie RC, Hakanson R,
Sundler F. Histamine-like immunoreactivity in photoreceptors of the compound eyes and ocelli of the flies
Calliphora erythrocephala and Musca domestica. Cell Tissue
Res. 1988; 253:639-646.
Pirvola U, Tuomisto L, Yamatodani A, Panula P. Distribution of histamine in the cockroach brain and visual
system: An immunocytochemical and biochemical
study. J Comp Neurol. 1988;276:514-526.
Calloway JC, Stuart AE. Biochemical and physiological
evidence that histamine is the transmitter of barnacle
photoreceptors. Vis Neurosri. 1989; 3:311-325.
Schlemermeyer E, Schutte M, Ammermuller J. Immunohistochemical and electrophysiological evidence
that locust ocellar photoreceptors contain and release
histamine. Neurosri Lett. 1989;99:73-78.
Battelle BA, Caiman BG, Andrews AW, et al. Histamine: A putative afferent neurotransmitter in Limulus
eyes. J Comp Neurol. 1991; 305:527-542.
Sarthy PV. Histamine: A neurotransmitter candidate
for Drosophila photoreceptors. / Neurochem. 1991;
57:1757-1768.
Burg MG, Sarthy PV, Koliantz G, Pak WL. Genetic
and molecular identification of a Drosophila histidine
decarboxylase gene required in photoreceptor transmitter synthesis. EMBOJ. 1993; 12:911-919.
Joseph DR, Sullivan PM, Wang Y-M, et al. Characterization and expression of the complementary DNA encoding rat histidine decarboxylase. Proc Natl Acad Sri
USA. 1990; 87:733-737.
Yamauchi K, Sata R, Tanno Y, et al. Nucleotide sequence of the cDNA encoding L-histidine decarboxylase derived from human basophilic leukemia cell line,
KU-812-F. NuclAridsRes. 1990; 18:5891.
Bennett MK, Scheller RH. A molecular description
of synaptic vesicle membrane trafficking. Annu Rev
Biochem. 1994; 63:63-100.
Ferronovick S, Jahn R. Vesicle fusion from yeast to
man. Nature. 1994; 370:6486.
Sollner T, Rotfiman JE. Neurotransmission harnessing
fusion machinery at the synapse. Trends Neurosci.
1994; 17:8.
Berridge MJ. Calcium oscillations. / Biol Chem.
1990; 265:9583-9586.
Irvine RF. "Quantal" Ca2+ release and control of Ca2+
entry by inositol phosphates: A possible mechanism.
FEBSLett. 1990;263:5-9.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933184/ on 07/31/2017