© 2002 Oxford University Press
Human Molecular Genetics, 2002, Vol. 11, No. 1
23–32
ARTICLE
The molecular basis of dichromatic color vision in
males with multiple red and green visual pigment genes
Wolfgang M. Jagla1, Herbert Jägle1, Takaaki Hayashi2, Lindsay T. Sharpe1,3,* and
Samir S. Deeb2
1Department
of Neuro-Ophthalmology, University Eye Hospital, Roentgenweg 11, D-72076 Tübingen, Germany,
of Medicine and Genome Sciences, University of Washington, Seattle, WA, USA and
3Department of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, UK
2Departments
Received September 3, 2001; Revised and Accepted November 2, 2001
We investigated the genotypic variation in 50 red–green color vision deficient males (27 deuteranopes and
23 protanopes) of middle European ancestry who possess multiple genes in the X-linked photopigment gene
array. We have previously shown that only the first two genes of the array are expressed and contribute to the
color vision phenotype. Therefore, the hypothesis is that the first two genes possessed by multigene-dichromats
encode pigments of identical or nearly identical spectral sensitivity: one gene normal (R or G) and the other a
hybrid (G/R or R/G). The spectral sensitivities of the encoded pigments were inferred from published in vitro
and in vivo data. The color vision phenotype was assessed by standard anomaloscopy. Most genotypes (92%)
included hybrid genes whose sequence and position and whose encoded pigment correlated exactly with the
phenotype. However, one and possibly two of the protanopes had gene arrays consistent with protanomaly
rather than protanopia, since two spectrally different pigments may be encoded by their arrays. Two of the
deuteranopes had only R- and G-photopigment genes, without any detectable G/R-hybrid genes or any as-of-yet
identified point mutation or coding/promoter sequence deletions. Further, an unexpectedly high number of
multigene-deuteranopes (11%) had the C203R mutation in their most upstream G-pigment gene, suggesting a
founder effect of middle European origin for this mutation. About half of the protanopes possessed an
upstream R/G-hybrid gene with different exon 2 coding sequences than their downstream G-pigment gene(s),
which is inconsistent with published data implying that a single amino acid substitution in exon 2 can confer
red–green color discrimination capacity on multigene-protans by altering the optical density of the cones.
INTRODUCTION
Human color vision is trichromatic due to the existence of three
different photopigments with distinct, spectral sensitivities
expressed in three separate classes of photoreceptor cell: the
short-wavelength sensitive [blue (B-)], middle-wavelength
sensitive [green (G-)] and long-wavelength-sensitive [red (R-)]
cones. The three photopigments share the same small conjugated
chromophore (11-cis-retinal), but differ in their covalently
linked heptahelical transmembrane protein or opsin. The genes
encoding the G- and R-opsins are arranged in a head-to-tail
tandem array on Xq28 (1–3) which has been directly visualized
by in situ hybridization (4). A single R-photopigment gene is
located at the 5′ position of the array followed by one or more
G-photopigment genes. A locus control region (LCR), located
upstream of the R-photopigment gene, has been shown to be
essential for cone photoreceptor-specific expression of all
genes in the array (5,6).
The R- and G-photopigment genes, each possessing a
translated region of six exons, are highly homologous in their
coding sequences. They differ in only 15 amino acids (1); seven of
which are believed to be responsible for the ∼32 nm difference in
the peaks of the absorbance spectrum of the R- (∼528 nm) and
G- (∼560 nm) cone photopigments (Table 1) (7–9).
Owing to the high degree of homology of the R- and G-photopigment genes and their juxtaposition in the head-to-tail
tandem array (1,10) the genes are predisposed to unequal
crossing over, which results either in a change in gene number
or in the formation of red/green (R/G) or green/red (G/R)
hybrid genes, leading in some cases to color vision deficiencies
(1,11). Those deficiencies affecting the function of the R-cones
are termed protan defects, with protanopia denoting the loss of
R-cone function (a form of dichromacy as opposed to
*To whom correspondence should be addressed at: Department of Psychology, University of Newcastle, Newcastle-upon-Tyne NE1 7RU, UK.
Tel: +44 191 2227514; Fax: +44 191 2225622; Email: [email protected]
Correspondence may also be addressed to S. S. Deeb at: Departments of Medicine and Genome Sciences, Box 357360 (Room 206), University of Washington,
Seattle, WA, USA. Tel: +1 206 543 1706; Fax: +1 206 543 0754; Email: [email protected]
24
Human Molecular Genetics, 2002, Vol. 11, No. 1
Table 1. Predicted peak absorbances of the R, G, R/G-hybrid and G/R-hybrid photopigments
Genotype
In vitro
In vitro
In vivo
In vivo
Recombinant pigments,
Merbs and Nathans (7)
Recombinant pigments,
Asenjo et al. (8)
Psychophysics,
Sharpe et al. (9)
Electroretinography,
Neitz et al. (34)
G(A180)/R1G2(A180)
529.7 ± 2.0
532 ± 1.0
527.8 ± 1.1
530
G(S180)
–
534 ± 1.0
–
–
R2G3(A180)
529.5 ± 2.6
532 ± 1.0
528.5 ± 0.7
530
R3G4(S180)
533.3 ± 1.0
534 ± 1.0
531.5 ± 0.8
–
R4G5(A180)
531.6 ± 1.8
–
535.4
–
R4G5(S180)
536.0 ± 1.4
538 ± 1.0
534.2
537
G2R3(A180)
549.6 ± 0.9
–
–
–
G2R3(S180)
553.0 ± 1.4
559 ± 1.0
–
–
G3R4(A180)
548.8 ± 1.3
555 ± 1.0
–
–
G4R5(A180)
544.8 ± 1.8
551 ± 1.0
–
–
R(A180)
552.4 ± 1.1
556 ± 1.0
557.9 ± 0.4
558
R(S180)
556.7 ± 2.1
563 ± 1.0
560.3 ± 0.3
563
trichromacy) and protanomaly (or protanomalous trichromacy)
denoting the alteration of R-cone function (associated with the
presence of R/G-hybrid genes which encode photopigments
with anomalous spectral sensitivity). Similarly, those deficiencies
affecting the function of the G-cones are termed deutan defects,
with deuteranopia denoting the loss of G-cone function and
deuteranomaly (or deuteranomalous trichromacy) denoting the
alteration of G-cone function (associated with the presence of
G/R-hybrid genes that encode photopigments with anomalous
spectral sensitivity) (reviewed in 12,13).
Clinically, the protan and deutan phenotypes are characterized
by the types of Rayleigh (red–green) color matches (14) made
on a small viewing field (<2° in diameter) anomaloscope. The
observer is required to match a spectral yellow (∼589 nm)
primary light to a juxtaposed mixture of spectral red (∼679 nm)
and green (∼544 nm) primary lights. Individuals with normal
color vision choose a unique match between the red–green
mixture ratio and the yellow intensity. Dichromats, on the
other hand, because they possess only one functioning class of
cone, R- or G- or a single hybrid pigment, are able to fully
match the spectral yellow primary to any mixture of the spectral
red and green primaries by merely adjusting the intensity of the
yellow, regardless of the red-to-green ratio. Thus, instead of a
unique match, they will have a fully extended matching range
that encompasses both the red and green primaries. Anomalous
trichromats have either an anomalous red (protanomalous) or
green (deuteranomalous) hybrid pigment. Their match ranges
are intermediate between those of normals and dichromats and
therefore, signify less severe forms of color vision deficiency
(12,15).
The various red–green color deficiency phenotypes are
closely associated with the type of crossing-over event.
However, the association is complex because the same phenotype may be associated with different genotypes. Unequal
recombination, in which the crossing-over point occurs in the
region between genes (intergenic crossing-over), can reduce
the gene array to a single R-cone pigment gene, resulting in
single-gene deuteranopia (the loss of function of the G-cones).
On the other hand, unequal recombination, in which the
crossing-over point typically occurs within the genes (intragenic
crossing-over), can have several outcomes. It can reduce the
array to a single R/G-hybrid gene that encodes a photopigment
with a spectral sensitivity either identical to or resembling that of the
normal G-cone pigment (Table 1). Alternatively, the R/G-hybrid
gene may coexist with one or more downstream normal G-cone
pigment genes (9,16). Generally, protanopia (the loss of R-cone
function) may be implied if the fusion in the hybrid gene
occurs before exon 3 because the gene will essentially encode
a G-cone pigment. However, this implication has yet to be
exhaustively tested and even very small differences in the λmax of
the two expressed photopigments can confer color discrimination
in some individuals in the Rayleigh task (W.M.Jagla,
T.Breitsprecher, C.Wolf, M.Pirzer, H.Jaegle, I.Kucsera,
G.Kovacs, B.Wissinger, S.S.Deeb and L.T.Sharpe, unpublished
data). Otherwise, protanomaly is predicted—although this again
needs to be thoroughly investigated and confirmed—because the
gene will encode an anomalous pigment that differs in spectral
sensitivity from the normal G-cone pigment by 2–8 nm,
depending on which amino acid residues have been substituted
(Table 1 has a summary of the published in vivo and in vitro
λmax estimates of normal and hybrid pigments). Finally,
unequal intragenic recombination can replace the normal G-cone
pigment gene with a G/R-hybrid gene (1,12,16). In such cases, the
array consists of a normal R-cone pigment gene followed by
a G/R-hybrid pigment gene (and possible additional G-cone
pigment genes). Deuteranopia may only be implied if the
fusion occurs before exon 3 because the gene will encode a
pigment with a spectral sensitivity similar to that of the normal
R-cone pigment. However, this has yet to be conclusively
demonstrated and, once again, exceptions have been observed
in performance on the classical Rayleigh task (W.M.Jagla,
T.Breitsprecher, C.Wolf, M.Pirzer, H.Jaegle, I.Kucsera, G.Kovacs,
B.Wissinger, S.S.Deeb and L.T.Sharpe, unpublished data).
Otherwise, deuteranomaly is predicted—although once again this
needs to be investigated and confirmed—because the gene will
encode an anomalous pigment that differs in spectral sensitivity
from the normal R-cone pigment by 4–12 nm, depending on
which amino acid residues have been substituted.
In the present study, we determine the variability in genotype
among dichromats with multiple photopigment genes in their
Human Molecular Genetics, 2002, Vol. 11, No. 1
array and correlate this to the color vision phenotype. First, we
determined the variability in the size of the gene array and of
the position of hybrid genes in relation to expression. This
issue has been looked at closely before. The existence of a
hybrid gene in the photopigment gene array does not have to be
associated with any color vision deficiency at all. What seems
to be decisive is the position of the hybrid gene in the array.
There is good evidence to suggest that only the first two genes
of the array are expressed in the retina and hybrid genes in a
more distal position do not disrupt or influence the normal
color vision (11,17,18). In fact, 4–8% of Caucasian males with
normal color vision have hybrid genes that are either not or
only insufficiently expressed (10,16).
Secondly, we investigated whether multi-gene dichromacy
is always caused by the presence of R/G- or G/R-hybrid genes
in the array or whether other causes can be found, such as point
mutations and major sequence deletions. So far, a C203R
missense mutation has been found in all of the G-cone pigment
genes in one individual with deuteranopia or extreme deuteranomaly (11) and in the cone pigment genes of multiple bluecone monochromats, who lack the function of both the R- and
G-cones (5,19). And a single case has been reported in which a
rearrangement, possibly a deletion, between exons 1 and 4 of
the R-cone opsin gene engendered a protan defect with a
progressive macular degeneration (19). For the general
understanding of photopigment gene expression, it would be
worthwhile determining the frequency of such occurrences.
Thirdly, we wanted to establish which R/G- and G/R-hybrid
genes in combination with normal genes engender dichromacy
and which engender anomalous trichromacy. What is the
minimum difference between the spectral sensitivities of the
normal R- (or normal G-) pigment and the anomalous pigment
encoded by a G/R-hybrid gene (or by a R/G-hybrid gene) that
confers some red–green color discrimination and allows an
individual to be an anomalous trichromat rather than a
dichromat? We already know that the primary determinants of
the spectral shift are located in exon 5. The R/G sequence
differences in exon 5—principally the residues at 277 and
285—result in spectral shifts of 15–25 nm, the exact value
depending on sequences in exons 2–4. The in vivo measured
data (9) further suggest that substitutions at the sites confined
to exons 2–4 produce much smaller spectral shifts. Exon 2
contributes at most 0.0–2.0 nm; exon 3, 1.0–4.0 nm; and exon
4, 2.4–4.0 nm. These results are in approximate agreement
with the in vitro results (7,8) and with inferences based on
comparison of primate visual pigment gene sequences and
cone spectral sensitivity curves (20–22). However, there is still
considerable uncertainty about the actual amount of spectral
shifting associated with sequence variations in each exon and
about how the shifts implied by different sequence variations
accumulate and interact. Moreover, there is little evidence about
how spectral differences between two photopigments actually
translate to color discrimination performance and about how
other retinal and cortical factors (such as individual variability
in the gain of post-receptoral color-opponent channels) act upon
the photopigment differences and influence discrimination.
Finally, in relation to the above issue, we wanted to take a
closer look at the role of exon 2 in conferring color discrimination
capacity and in influencing the optical density of photopigments. Codon differences between the R- and G-cone pigment
genes are confined to three sites: codons 65, 111 and 116 (1).
25
In vitro (7,8) and in vivo (9) data suggest that substitutions at
these sites contribute very little to spectral tuning (but this
should be examined more closely). However, recently it has been
speculated that differences in the amino acids encoded by red
and green exon 2 (in particular codon 116) are involved in
determining the stability and therefore the optical density of the
L- and M-cone photopigments, which can also contribute to
red–green color discrimination capability (23). This is because
increasing the photopigment optical density broadens the relative sensitivity of the pigment away from the absorption peak.
In particular, if two genes that differ only in their exon 2
sequences are expressed, then the resulting differences in the
optical density of the pigments may engender some differences
in spectral sensitivity that may be responsible for some
residual red–green color discrimination. If so, this would imply
that multi-gene protanopic subjects would generally not be
expected to have different exon 2 variants, whereas some
protanomalous subjects, whose opsin genes are otherwise
identical in their sequences, would.
RESULTS
The ratios of G/R sequences (promoters and exons 2–5) of the
photopigment gene arrays of 23 protanopes and 27 deuteranopes were determined by single-strand conformation polymorphism (SSCP) analysis as described in Materials and
Methods. The mean total number of genes within the arrays
differs between the investigated gene arrays of the protanopic
(2.83 genes) and deuteranopic (4.13 genes) subjects. This
finding is consistent with the prediction from the mechanism
of recombination that generates R/G- and G/R-hybrid pigment
genes (12,13).
Gene arrays of protans
The mean ratios of green-to-red sequences are given in
Table 2. The estimated gene numbers ranged from two to four:
six subjects (26.1%) had two-gene arrays; 15 (65.2%) had
three-gene arrays; and two (8.7%) had four-gene arrays. In all
23 protanopic subjects, one R/G-hybrid gene and one or more
(up to three) normal G-genes were found (Fig. 1).
The first gene of the array of all protanopic subjects was
analyzed by long-range (LR)-PCR using a red-specific primer
in the promoter and a green-specific primer in exon 5 (24). All
subjects were found to have a R/G-hybrid gene in the proximal
position of the array: 10 (43.5%) had a R1G2(A180) hybrid
gene; 10 (43.5%) a R2G3(A180) hybrid gene; one (4%) a
R3G4(S180) hybrid gene; and two (9%) a R4G5(A180) hybrid
gene.
The 3′-terminal gene in the array was characterized by LR-PCR
and amplification of its exon 5, in the single subject (2324)
who was inferred to have a second (G/R) hybrid gene in
addition to his proximal (R/G) hybrid gene (18). It established
that the G/R-hybrid gene was in the terminal position (third
gene) of his array.
The presence of Ala or Ser at position 180 is given separately
for the proximal (first) gene and the distal (downstream) genes
of each array in Table 2. Of the 23 proximal, R/G-hybrid
genes, 21 had Ala and two had Ser at position 180. Of the
90 distal green genes, 83 had Ala and seven had Ser. The single
G/R-hybrid gene (subject 2324) had Ser at position 180.
26
Human Molecular Genetics, 2002, Vol. 11, No. 1
Table 2. Determination of total gene number, detection of G- and R-sequences in human genomic DNA for protanopic subjects with a R/G-hybrid gene in the
proximal position and determination of a serine/alanine polymorphism in exon 3 of proximal (R/G-hybrid) and distal (G and/or G/R-hybrid) photopigment genes
Subject ID
number
Promotor (G/R-ratio)
Estimated total
gene number
Exon 2
Exon 3
proximal gene
Exon 3
distal genes
Exon 4
Exon 5
2190
3.20 (∼3)1
4
G
A180
A180
G
G
2191
1.90 ± 0.57 (∼2)2
3
G
A180
A180
G
G
2192
1.55 ± 0.07 (∼2)2
3
G
A180
A180
G
G
2193
2.17 ± 0.06 (∼2)
3
R+G
A180
A180
G
G
2194
1.10 ± 0.26 (∼1)
2
(1.12 ± 0.08)
S180
A180
G
G
2195
2.00 ± 0.42 (∼2)2
3
R
A180
A180
G
R
2199
2.08 ± 0.41 (∼2)
3
R+G
A180
A180
G
G
2200
1.85 ± 0.07 (∼2)2
3
G
A180
A180
G
G
2203
2.07 ± 0.32 (∼2)
3
R+G
A180
A180
(0.55 ± 0.05)
G
2204
2.03 ± 0.60 (∼2)
3
G
A180
A180
G
G
2224
1.10 ± 0.10 (∼1)
2
G
A180
A180
G
G
2225
1.10 ± 0.00 (∼1)
2
R+G
A180
A180
G
G
2233
2.00 ± 0.42 (∼2)2
3
G
A180
A180
G
G
2236
2.00 ± 0.42 (∼2)2
3
G
A180
A180
G
G
2286
1.13 ± 0.06 (∼1)
2
R+G
A180
A180
G
G
2288
1.98 ± 0.34 (∼2)
3
R+G
A180
A180
G
G
2293
1.13 ± 0.06 (∼1)
2
R+G
A180
A180
G
G
2303
1.97 ± 0.25 (∼2)
3
G
A180
A180
G
G
2312
3.37 ± 0.69 (∼3)
4
(3.08 ± 0.52)
S180
A180
(2.8 ± 0.28)
G
2314
1.46 ± 0.30 (∼1)
2
R+G
A180
A180
G
G
2323
2.18 ± 0.11 (∼2)
3
(2.27 ± 0.34)
A180
A180
G
G
2324
2.20 ± 0.26 (∼2)
3
R+G
A180
A180 (G) + S180 (G/R)
(1.89 ± 0.31)
(2.17 ± 0.74)
2326
1.67 ± 0.21 (∼2)
3
G
A180
A180
G
G
Numbers represent means ± SE of at least three independent determinations (exceptions are marked with the number of determinations as superscript). In the promotor
column, the values in parentheses are rounded to the nearest whole number. In the exon 2, 4 and 5 columns, the values in parentheses represent means ± SE for G/R ratios.
The putative R/G-gene arrays of all protanopic subjects are
shown schematically in Figure 1. The molecular analyses
allowed inference of the spectral sensitivities of the opsin
encoded by each pigment gene. The two most frequent hybrid
genes in the arrays were R1G2 (43.5%) and R2G3 (43.5%)
with Ala at position 180. According to the spectral properties
of in vitro expressed and in vivo photopigments (Table 1),
these hybrid genes are predicted to encode photopigments
having identical [R1G2(A180)] or almost identical [R2G3(A180)]
absorbance spectrum peaks (λmax) to that of the normal pigment
[the estimated difference in λmax between the R2G3(A180) and
G(A180) pigment is 0.0–0.7 nm]. Although one of the individuals
with a R2G3(A180) hybrid gene (2324) also has a G2R3(S180)
hybrid gene, it occupies the most distal position (third) in the
array and is therefore assumed to have no significant influence on
phenotype (18).
Subject 2203 has a red exon 4 in the first R4G5(Ala180)
hybrid gene and also in one of his two G-pigment genes. The
order of the two G-pigment genes is unknown, therefore, two
arrangements are possible. However, we predict that the
second gene in the array is the one with a red exon 4, which
would result in the first two genes of the array encoding
pigments with identical or near identical absorption maxima,
because they differ only in their exon 2 coding sequences,
consistent with protanopia (the estimated difference in λmax is
0.0–0.7 nm; Table 1).
Interestingly, there are two apparent exceptions to the rule
that multigene protanopes have identical or nearly identical
pigments encoded by their first two genes: subject 2194, who
has a R3G4(S180) hybrid gene, and subject 2312, who has a
R4G5(S180) hybrid gene, paired with a G(A180) gene. These
combinations are predicted to encode pigments with absorption
maxima that differ by 2–3.7 and 6 nm, respectively (Table 1).
Gene arrays of deuteranopes
The gene arrays of multi-gene deuteranopes were characterized
by the same methods as used to characterize the gene arrays of
the multi-gene protanopes. The ratios of their green-to-red
sequences for the promoter and exon 2, 4 and 5 fragments, as
well as their estimated total gene number, are given in Table 3.
As for the protanopic subjects, the results of SSCP analysis
concerning total gene number confirmed those of a previous
study in which Southern blot hybridization after gel electrophoresis was performed (9). However, the size of the arrays in
those deuteranopic subjects having more than four photopigment
genes can only be roughly inferred since the G/R ratios cannot be
determined exactly by SSCP analysis. Interestingly, the incidence
of (estimated) large gene arrays is higher among multi-gene
Human Molecular Genetics, 2002, Vol. 11, No. 1
27
Figure 1. Multi-gene protanopes (N = 23). Photopigment gene arrays inferred from the detection of hybrid genes and the estimation of the total gene number.
Rectangles represent the six exons of the prototype R- (closed) and G- (open) pigment genes. The number of genes of the same kind is indicated as subscript. In
the distal genes, the exon 3 polymorphism A180S is shown closed for S180 and open for A180. Subject (2203) has a R4G5(A180) hybrid gene and two variants
in exon 4 of his distal G-genes, which would predict different protan phenotypes depending on the gene order. The shown order in square brackets is consistent
with the diagnosed protanopic phenotype.
deuteranopes than among protanopes: 16 (59.26%) of the
deuteranopes had arrays with four or more genes, with the
largest array estimated to have eight genes, as compared with
two (8.7%) of the protanopes, with the largest array estimated
to have four genes.
The first gene of the array of all deuteranopic subjects was
determined by LR-PCR using red-specific primers in the
promoter and exon 5. All subjects were found to have a normal
R-pigment gene, with either the Ala or Ser polymorphic
variant, in the first position. However, in one subject (2311), a
green exon 4 was found in the R-pigment gene.
Of the 27 deuteranopic subjects, 23 had arrays comprised of
one normal R-pigment gene (occupying the first position), at
least one G/R-hybrid gene, and one or more (up to six) normal
G-pigment genes (Fig. 2). Five different types of G/R-hybrid
gene were identified: G1R2(A180) in one individual,
G2R3(S180) in six individuals, G3R4(A180) in 15 individuals
and G4R5(S180) in one individual. Among these 23 arrays,
65% contained a G3R4(A180) hybrid gene paired with a
R(A180)-pigment gene in the first position. This pair of genes
encodes identical or nearly identical pigments (Table 1), which
is consistent with deuteranopia.
The nature of the 3′-terminal gene in arrays was determined
in all subjects by LR-PCR. No G/R-hybrid genes were detected
in the terminal position of any of the arrays. In seven subjects
(2230, 2300, 2175, 2301, 2309, 2289, 2311) who have only
two or three genes in the array, we could directly confirm by
LR-PCR of the terminal gene that the G/R-hybrid gene occupies
the second position in the array. And, in the rest of the cases,
this is a justified assumption to make because only a single
hybrid gene is present in the array.
In two of the 27 subjects (2187, 2308), deuteranopia was not
caused by the presence of a G/R-hybrid gene (Fig. 2B), but
rather by the occurrence of a C203R mutation in exon 4 of their
most proximal (upstream) G-pigment gene (Table 3).
The C203R mutation was also present in one of the G-pigment
genes of a third subject (2183) who also possesses a
G2R3(S180) hybrid gene in his array (Fig. 2C). LR-PCR
established that a normal G(A180) gene occupied the most
distal position in his four pigment gene array. Thus, it is not
possible by present methods to resolve if his deuteranopia is
caused by the C203R mutation in the G-pigment gene or by the
G/R-hybrid gene; since we cannot resolve which of these two
genes occupies the most proximal (second) position in his array
(see the two alternative gene arrangements in Fig. 2C). But both
arrangements are compatible with a deuteranopic phenotype.
Two other subjects (2170, 2246) had neither a hybrid gene
nor a gene with a C203R mutation in their pigment gene arrays
(Fig. 2D). Therefore, to determine if some other factor was
causing their deuteranopia, we sequenced all exons, intron–exon
junctions and the proximal promoter regions of their single Rand G-pigment genes. No as-of-yet known mutation was
28
Human Molecular Genetics, 2002, Vol. 11, No. 1
Table 3. Ratios of G- and R-sequences in human genomic DNA for deuteranopic subjects with a normal R-gene in the proximal position and determination of a
serine/alanine polymorphism in exon 3 of proximal (red) and distal (G and/or G/R-hybrid) photopigment genes
Subject ID
number
Promotor
(G/R ratio)
Estimated total Exon 2
gene number
(G/R ratio)
Exon 3
proximal gene
Exon 3 distal genes
Exon 4
(G/R ratio)
Exon 5
(G/R ratio)
2169
3.30 ± 1.23 (∼3)
4
0.80
A180
A180
1.20 ± 0.14
1.05 ± 0.21
2170
1.20 ± 0.24 (∼1)
2
1.16 ± 0.06
A180
S180 (G/R) + A180 (G)
1.08 ± 0.05
0.85 ± 0.07
2172
3.00 ± 0.61 (∼3)
4
2.54 ± 0.07
A180
S180 (G) + A180 (G/R)
0.90 ± 0.28
1.00 ± 0.00
2174
3.92 ± 0.83 (∼4)
5
A180
S180 + A180
0.30 ± 0.17
0.22 ± 0.04
2175
2.08 ± 0.28 (∼2)
3
A180
A180
0.55 ± 0.07
0.50 ± 0.00
2178
3.70 ± 1.56 (∼4)
4
A180
A180
1.19 ± 0.41
1.10 ± 0.14
2179
2.95 ± 0.49 (∼3) 2
4
A180
A180
0.99 ± 0.02
0.85 ± 0.07
2180
3.27 ± 0.71 (∼3)
4
A180
A180
1.23 ± 0.33
0.85 ± 0.07
2183
2.83 ± 0.65 (∼3)
4
S180
S180 (G/R) + A180 (G)
0.95 ± 0.21a
0.90 ± 0.00
2186
2.65 ± 0.20 (∼3)
4
A180
A180
1.09 ± 0.59
1.30 ± 0.14
2187
2.10 ± 0.11 (∼2)
3
A180
A180
1.74 ±
0.37a
1.85 ± 0.09
2223
5.98 ± 0.53 (∼6)
7
A180
S180 (G) + A180 (G/R)
2.72 ± 0.34
2.65 ± 0.24
2230
2.37 ± 0.45 (∼2)
3
S180
S180
0.62 ± 0.10
0.54 ± 0.06
2240
5.05 ± 3.04 (∼5)
6
S180
A180
1.80 ± 0.01
1.55 ± 0.35
2242
3.20 ± 1.04 (∼3)
4
A180
A180
1.19 ± 0.16
0.90 ± 0.00
2246
1.30 ± 0.14 (∼1)
2
A180
S180
1.09 ± 0.02
1.00 ± 0.23
2289
2.37 ± 0.12 (∼2)
3
A180
A180
1.09 ± 0.07
0.88 ± 0.07
2300
2.13 ± 0.32 (∼2)
3
S180
S180
2.05 ± 0.35
0.53 ± 0.12
2301
2.03 ± 0.25 (∼2)
3
A180
A180
0.51 ± 0.16
0.62 ± 0.07
2302
2.98 ± 0.98 (∼3)
4
A180
A180
0.85 ± 0.07
0.95 ± 0.05
2308
2.20 ± 0.20 (∼2)
3
A180
A180
2.05 ±
0.41a
1.95 ± 0.13
2309
1.97 ± 0.46 (∼2)
3
G only
A180
A180
0.60 ± 0.14
0.56 ± 0.04
2311
2.07 ± 0.21 (∼2)
3
G only
S180
S180 (G/R) + A180 (G)
G only
0.43 ± 0.05
2316
3.08 ± 1.14 (∼3)
4
A180
A180
1.11 ± 0.13
1.00 ± 0.17
2320
7.15 ± 3.80 (∼7)
8
2.98 ± 0.28
A180
S180 (G) + A180 (G/R)
2.67 ± 0.06
2.23 ± 0.24
2321
2.97 ± 1.04 (∼3)
4
2.70 ± 0.47
S180
S180 (G/R) + A180 (G)
3.00 ± 0.25
2.25 ± 0.52
2330
2.76 ± 0.03 (∼3)
4
4.88 ± 1.02
S180
S180 (G/R) + A180 (G)
0.98 ± 0.04
1.19 ± 0.08
1.92 ± 0.11
2.28 ± 0.23
2.65 ± 0.26
2.78 ± 0.32
1.13 ± 0.06
Numbers represent means ± SE of at least three independent determinations (exceptions are marked with the number of determinations as superscript). In the
promotor column, the values in parentheses are rounded to the nearest whole number. Column 6 (Exon 3, distal genes): (G), located in G-gene; (G/R), located in
G/R-hybrid gene.
aC203R mutation in exon 4 in at least one distal gene.
identified. Thus, the cause of deuteranopia in these subjects
remains to be determined.
DISCUSSION
There are four points to consider about our data: (i) what is the
role of normal and hybrid genes in multigene dichromacy;
(ii) which R/G- and G/R-hybrid genes in combination with
normal genes engender dichromacy; (iii) how frequently do
factors other than hybrid genes (e.g. point mutations) cause
multigene dichromacy; and (iv) what is the role of exon 2 in
conferring color discrimination capacity and in influencing the
optical density of photopigments.
The role of hybrid genes in multigene dichromacy
Hybrid genes play the most important role in determining
multiple-gene dichromacy. In all 23 protanopic subjects, an
R/G-hybrid gene was found as the first gene in the array; and
in 23 of the 27 deuteranopes, a G/R-hybrid gene was found and
either directly identified (nine subjects) or assumed (14 subjects)
to be the second gene in the array. In the majority of protanopes
(87%), the hybrid gene was the result of recombinations in
either intron 1 (43%) or intron 2 (43%); in the majority of
deuteranopes (87%), it was the result of recombinations in
either intron 2 (22%) or intron 3 (65%).
All deuteranopes who had hybrid genes also had normal
G-pigment genes. The expression of these normal G-pigment
genes would have resulted in normal color vision or a reduced
severity of the defect had they been sufficiently expressed.
Subjects who have highly biased G/R-cone ratios (G/R: 0.09
and 0.03) (25) have been shown to have normal color vision.
So even 3% expression of a photopigment gene may suffice to
provide full trichromatic vision. However, as has been demonstrated before (17,18,26), only the first two genes of the array
are expressed and contribute to color vision. Here, we found
Human Molecular Genetics, 2002, Vol. 11, No. 1
29
Figure 2. Multi-gene deuteranopes (N = 27). (A) Photopigment gene arrays of 22 deuteranopic subjects with hybrid genes inferred from the detection of hybrid
genes and the estimation of the total gene number. See Figure 1 for more details. In the distal genes, the exon 3 polymorphism A180S is shown closed for S180,
open for A180 and as a gray gradient when both A180 and S180 are present. (B) Photopigment gene arrays of two deuteranopic subjects with a C203R point
mutation inferred from the C203R screening results and the estimation of the total gene number. (C) Deuteranopic subject with a G2R3(S180) hybrid gene and a
C203R mutation in one of his two G(A190) genes. Neither of these genes occupies the most distal position. Thus, two different gene orders are possible to explain
the deuteranopic phenotype. The order in which the gene with the C203R mutation occupies the upstream position is shown in square brackets. (D) Photopigment
gene arrays of two deuteranopic subjects without hybrid genes, as inferred from the SSCP-based G/R-ratio analysis and LR-PCR-based hybrid gene screening.
that in all seven deuteranopes who carried a total of three genes
in the array, the G/R-hybrid gene occupied the second position
in the array. In all others who had more than three genes, a
normal G-pigment gene occupied the terminal position in the
array. These results are in agreement with earlier results on
deuteranomals and normals (17,18,26). An unlikely alternative
to the influence of gene position on expression (i.e. that only
the two most upstream genes in the array are expressed), which
has been suggested by Neitz and Neitz (27), is that all the extra
G-pigment genes in deuteranopes have inactivating mutations.
However, our sequence analysis failed to detect any such mutations.
Which R/G- and G/R-hybrid genes in combination with
normal genes engender dichromacy?
Our hypothesis is that dichromats should express either R- or
G-pigment genes in their retinae, but not both. Therefore, the
expectation is that the first two genes in the array encode
30
Human Molecular Genetics, 2002, Vol. 11, No. 1
identical cone photopigments or ones that differ in their λmax by
very small amounts (say >1 nm). This prediction is borne out in
all deuteranopes who carry G/R-hybrid genes (23/27) and in the
majority of protanopes (21/23), all of whom carry hybrid genes
(see the similarity between the λmax of their hybrid pigments
and those of the normal pigments in Table 1). However, among
protanopes, there were two potential exceptions. In subject
2194, the first two genes are estimated to encode photopigments
that are inferred to differ in λmax by 2.0–3.7 nm (Table 1). In the
other (2312), the estimated difference in λmax between the
photopigments is 1.7–3.9 nm or 1.9–7.6 nm, depending upon
whether the Ser180 or Ala180 carrying G-pigment gene occupies
the second position (Table 1). These two exceptions notwithstanding, we conclude that a λmax separation of >1 nm may be
required to confer color discrimination capacity. The two
exceptions may be explained by inaccuracies in the in vivo and
in vitro estimates of λmax (see the spread among estimates in
Table 1) or, perhaps, other factors than differences in the spectral
tuning of the hybrid and normal photopigments influence color
discriminability.
Notably, in addition to being associated with subtle differences among males with normal color vision (28), the common
Ser/Ala 180 polymorphism in exon 3 of the R-pigment gene
contributes to the phenotype of dichromacy by decreasing the
difference in λmax between the first two genes of the array. That
is, we would have found a lower frequency of dichromacy in
our population if the Ser/Ala polymorphism was not present
and all the R-pigment genes had Ser at that position.
The role of point mutations and other factors in multi-gene
dichromacy
To determine other causes of dichromacy than hybrid gene
expression was another main topic of our study. One already
identified cause of deutan defects [one such case was reported by
Winderickx et al. (11)] and a very important cause of X-linked
blue-cone monochromacy (29) is the C203R point mutation.
The replacement of Cys by Arg disrupts a highly conserved
disulfide bond, which impairs folding of the opsin leading to
the formation of an unstable pigment (30). We found three
subjects with this C203R mutation. In two of these individuals
(2187, 2308), no G/R-hybrid gene was found in the array and
their deuteranopia is straightforwardly explained by the
presence of a C203R mutation in a normal G(A180) gene,
which occupies the second position in the array (Fig. 2B). In
the third individual (2183), who was determined to have four
genes in total, one G/R-hybrid gene exists in addition to the
G-pigment gene with the C203R mutation (Fig. 2C). Due to
the inability to amplify DNA segments of ~39 kB, it is not
presently possible to directly determine whether the G/R-hybrid
gene occupies the second position in arrays consisting of a total
of more than three genes, unless the hybrid is located at the
3′end of the array. However, it is more likely that the C203Rcarrying gene occupies the second position rather than the
hybrid G/R-hybrid gene, because the λmax difference between
the normal R-pigment gene and the G/R-hybrid is ∼12 nm (see
the pigment absorbance peaks in Table 1). Such a large
spectral separation would be expected to provide a deuteranomalous phenotype rather than a deuteranopic one.
The unexpectedly high number of subjects (11.1% among
multi-gene deuteranopes and 6% among both deuteranopes
and protanopes) with a C203R point mutation is remarkable. It
suggests a founder effect of middle European origin for this
mutation. A C203R screening of 100 German females with
normal color vision, which found five incidences (in 2.5% of
all X chromosomes), supports this speculation (W.M.Jagla,
T.Breitsprecher, C.Wolf, M.Pirzer, H.Jaegle, I.Kucsera, G.Kovacs,
B.Wissinger, S.S.Deeb and L.T.Sharpe, unpublished data). In
contrast, in previous studies of a Caucasian population in the
United States, only one out of 84 color deficient males (1.2%)
carried the C203R mutation (16,31).
In two deuteranopic subjects (2170, 2246), the photopigment
gene array consists of one normal R- and one normal G-photopigment gene (Fig. 2C). No C203R mutation or any other
mutation was detected in all exons, exon–intron junctions and
proximal promoters of all the R- and G-pigment genes. Thus,
the gene arrangement would be predicted to give rise to normal
trichromacy. The reason for their dichromacy remains speculative.
It might be caused by an as-of-yet undetected mutation in an
unknown regulatory domain, or it could be due to a deficiency
involving a post-receptoral step. We propose that somatic
mosaicism may explain this apparent paradox. Unequal mitotic
recombination between the R- and G-pigment genes may have
occurred during development. This would have generated
some tissues with a normal gene and others with R/G- and
G/R-hybrid genes. Analysis of DNA from other tissues can
provide a test of this hypothesis. That is, the white blood cells,
from which we isolated DNA for analysis, may have derived
from cells that had not undergone mitotic unequal recombination between the R- and G-pigment genes carried by sister
chromatids, whereas the photoreceptor cells may have derived
from descendants of cells in which such a sister chromatid
exchange (to form hybrid genes) occurred during development.
The role of exon 2 in conferring color discrimination capacity
and in influencing the optical density of photopigments
Almost half (43.5%) of the multigene protanopes (10)
possessed an upstream R2/G3-hybrid gene with different
exon 2 coding sequences than their downstream G-pigment
gene(s). Thus our results are inconsistent with recently
published data (23) implying that a single amino acid substitution
in exon 2 can confer red–green color discrimination capacity
on individuals with multigene protan defects by altering the
optical density of the cones. Neitz et al. (23) found that in nine
protanopes, the first two genes in the arrays encoded pigments
predicted to be identical in λmax and had no sequence difference
in exon 2. However, they found that in five protanomalous
subjects the only difference between the first two genes of the
array was in exon 2, particularly at position 116 in three of
these subjects. Based on these observations they concluded
that differences in the exon 2 sequences at position 116 do not
contribute to spectral separation differences, but rather to the
stability of the pigments. This difference in stability was
hypothesized to confer some color discrimination. However,
our results do not support their interpretation. Rather, we
suggest that differences in cone photopigment optical density
may arise randomly, irrespective of differences in exon 2
sequences and may be responsible for residual color discrimination capacity (32,33). In support of our hypothesis, it has
been previously observed that a fraction of males (3/13) who
Human Molecular Genetics, 2002, Vol. 11, No. 1
have a single R-pigment gene in their array display anomalous
trichromatic color vision (16); for another exception, see
Nathans et al. (1).
CONCLUSIONS
We genotyped in detail 23 protanopic and 27 deuteranopic
males, with multiple genes in their X-linked photopigment
gene array, to determine the molecular basis for their phenotype,
which was diagnosed by the standard Nagel Type I anomaloscope.
All protanopes had R/G-hybrid genes, and the majority (85%)
of deuteranopes had G/R-hybrid genes. Thus, the major cause
of their dichromacy (92%) is the presence of hybrid genes.
Although we find a wide diversity among them in the nature of
their hybrid genes and total gene number, the variability is
consistent with only the first two genes in the array being
expressed and with the hybrid gene, in all but two protanopes,
encoding an anomalous pigment that differs in its λmax from
that of the paired normal pigment by <1 nm. Among the
deuteranopes, the normal G-pigment genes, even when present
in as many as six copies per array, do not contribute to any
color discrimination capacity. Of the 15% of deuteranopes who
did not carry a G/R-hybrid gene, two had the C203R mutation.
Unexpectedly, two others had neither hybrid genes nor point
mutations. The reason for their dichromacy remains inexplicable
in terms of the present state-of-the-art L-/M-photopigment
gene analysis.
MATERIALS AND METHODS
31
gene promoter regions were amplified by competitive PCR
using [γ-32P]ATP labeled primers and then subsequently
analyzed by SSCP (17). The ratio of green and red photopigment genes and the total gene number was then evaluated
by phosphoimaging. Because the R-photopigment promoter
exists as a single copy, the total number of genes in the array
could be estimated by adding one to the rounded promoter G/R
ratio. Inequality in the ratios for any individual indicated the
presence of hybrid genes. For example, if the G/R ratio for the
promoter and exon 4 fragments is 2 but that of exon 5 is 0.5, the
array was inferred to be composed of a normal R, a normal G
and a G/R-hybrid gene. Details of the experimental protocols
are given by Deeb et al. (24).
Detection of R/G-hybrid genes. LR-PCR using red-specific
primers in the promoter and green-specific primers in exon 5
was used to determine if the first gene in the array was red or a
R/G-hybrid (24).
Detection of normal G- and G/R-hybrid genes. LR-PCR using
green-specific primers in the promoter and red- or greenspecific primers in exon 5 was used to determine if the distal
genes in the array were green or a G/R hybrid or both (24).
Analysis of the sequence and position of G/R-hybrid genes.
LR-PCR combined with restriction fragment analysis was used
to identify the gene that occupies the 3′-terminal gene of the
array (24). The ratios of red–green promoter and exons 2, 4 and
5 were determined by SSCP analysis. The differences between
the ratios for the promoter and the exons establishes the point
of fusion in a hybrid gene.
Subjects
X chromosome-linked (red–green) color-blind young males
were recruited in Freiburg im Breisgau (Germany), Tübingen
(Germany) and Vienna (Austria) via word of mouth and by
advertising in local newspapers and cinemas. Their color
vision deficiencies were established by screening with the
Ishihara pseudoisochromatic plates (edition 5) and by the
Rayleigh (red–green) color-matching equation measured on a
Nagel Type I anomaloscope.
Genotyping
Detecting multiple-gene arrays. The initial genotyping was
performed by Southern blot hybridization after gel electrophoresis (1,28), and has been fully described in a previous
study (9). That study’s objective was to establish the correlation
between deduced protein sequences and psychophysically
measured spectral sensitivities in single-gene dichromats. After
their genotype was established, the multiple-gene dichromats
were not further investigated. For this study, only the dichromats
with a multiple gene array were investigated. The gene arrays of
multi-gene were characterized in detail by competitive/quantitative
PCR amplification, followed by SSCP, restriction enzyme
digestion, LR-PCR amplification and in some cases confirmed
by sequencing of PCR products. The protocols for analysis of
arrays have been described in detail by Deeb et al. (24).
Determination of number and ratios of red and green pigment
genes. To determine the total number and ratios of photopigment
genes in the X-linked gene array, the red and green photopigment
Determination of the genotype at position 180. The genotype
with respect to the Ser180Ala polymorphism in each of the
pigment genes of the array was first determined by amplification,
using total genomic DNA, of exon 3 followed by digestion
with Fnu4H as previously described by Deeb et al. (24). If both
the Ser and Ala alleles were detected, the genotype of the first
gene of the array was determined using the LR-PCR product of
the first gene (diluted 1:1000) as template for a second round
of amplification of the exon 3 fragment, followed by digestion
with Fnu4H. The same two-round amplification strategy was
performed on LR-PCR products of the downstream genes with
green-specific primers (normal G- or G/R-hybrid genes) (for
more details see 24).
Analysis of the sequence in R- and G-genes. The sequence of
the green pigment genes was determined by SSCP analysis of
PCR products. This covered the promotor and the coding
sequences. The first and all green pigment genes were amplified
separately and subjected to SSCP analysis, as well as
sequencing of the exon 2 of some samples to conform the
SSCP results. The LR-PCR products using green-specific
primers were used to amplify exon 4 of all G-genes. This was
followed by digestion with BstU1 to detect the C203R mutation.
Determination of the 3′-terminal gene of the array. The identity of the 3′-terminal gene in the array (normal or G/R-hybrid
gene) was determined by LR-PCR amplification followed by
restriction enzyme analysis of exon 5 as described previously
by Hayashi et al. (18).
32
Human Molecular Genetics, 2002, Vol. 11, No. 1
Derived spectral sensitivities for photopigments
The predicted peak absorbances of the R, G, R/G-hybrid and
G/R-hybrid photopigment spectra (Table 1) were inferred from
published data on photopigments expressed in vitro (7,8) and
in vivo (9).
ACKNOWLEDGEMENTS
This work was supported by a National Institutes of Health
grant no. EY08395 to S.S.D., a travel grant from the Smith
Kline Beecham-Stiftung to W.M.J. and by a grant from the
Deutsche Forschungsgemeinschaft (Bonn-Bad Godesberg)
SFB 430 (Tp A6) and a Hermann-und-Lilly-Schilling-Professur
(Stifterverband, Essen) awarded to L.T.S.
REFERENCES
1. Nathans,J., Piantanida,T.P., Eddy,R.L., Shows,T.B. and Hogness,D.S.
(1986) Molecular genetics of inherited variation in human color vision.
Science, 232, 203–210.
2. Vollrath,D., Nathans,J. and Davis,R.W. (1988) Tandem array of human
visual pigment genes at Xq28. Science, 240, 193–202.
3. Feil,R., Aubourg,P., Helig,R. and Mandel,J.L. (1990) A 195-kb cosmid
walk encompassing the human Xq28 color vision pigment genes.
Genomics, 6, 367–373.
4. Wolf,S., Sharpe,L.T., Schmidt,H.-J.A., Knau,H., Weitz,S., Kioschis,P.,
Poustka,A., Zrenner,E., Lichter,P. and Wissinger, B. (1999) Direct visual
resolution of gene copy number in the human photopigment gene array.
Invest. Ophthalmol. Vis. Sci., 40, 1585–1589.
5. Nathans,J., Davenport,C.M., Maumenee,I.H., Lewis,R.A., Hejtmancik,J.F.,
Litt,M., Lovrien,E., Weleber,R., Bachynski,B., Zwas,F., Klingaman,R.
and Fishman,G. (1989) Molecular genetics of human blue cone
monochromacy. Science, 245, 831–838.
6. Wang,Y., Macke,J.P., Merbs,S.L., Zack,D.J., Klaunberg,B., Bennett,J.,
Gearhart,J. and Nathans,J. (1992) A locus control region adjacent to the
human red and green visual pigment genes. Neuron, 9, 429–440.
7. Merbs,S.L. and Nathans,J. (1992) Absorption spectra of the hybrid
pigments responsible for anomalous color vision. Science, 258, 464–466.
8. Asenjo,A.B., Rim,J. and Oprian,D.D. (1994) Molecular determinants of
human red/green color discrimination. Neuron, 12, 1131–1138.
9. Sharpe,L.T., Stockman,A., Jägle,H., Knau,H., Klausen,G., Reitner,A. and
Nathans,J. (1998) Red, green, and red–green hybrid pigments in the
human retina: correlations between deduced protein sequences and
psychophysically measured spectral sensitivities. J. Neurosci., 18,
10053–10069.
10. Drummond-Borg,M., Deeb,S.S. and Motulsky,A.G. (1989) Molecular
patterns of X chromosome-linked color vision genes among 134 men of
European ancestry. Proc. Natl Acad. Sci. USA, 86, 983–987.
11. Winderickx,J., Sanocki,E., Lindsey,D.T., Teller,D.Y., Motulsky,A.G. and
Deeb,S.S. (1992) Defective colour vision associated with a missense
mutation in the human green visual pigment gene. Nat. Genet., 1, 251–256.
12. Sharpe,L.T., Stockman,A., Jägle,H. and Nathans,J. (1999) Opsin genes,
cone photopigments, color vision and colorblindness. Chapter 1.
In Gegenfurtner,K. and Sharpe,T.T. (eds), Color Vision: From Genes to
Perception. Cambridge University Press, Cambridge, pp. 3–52.
13. Motulsky,A.G. and Deeb,S.S. (2001) Color vision and its genetic defects.
In Scriver,C.R., Beaudet,A.L., Sly,W.S. and Valle,D. (eds),
The Molecular and Metabolic Bases of Inherited Disease, 8th edn.
McGraw-Hill, New York, Vol. 4, pp. 5955–5976.
14. Strutt,R.J. (1881) Experiments on colour. Nature, 25, 64–66.
15. Pokorny,J., Smith,V.C. and Verriest,G. (1979) Congenital color defects.
In Pokorny,J., Smith,V.C., Verriest,G. and Pinckers,A.J.L.G. (eds),
Congenital and Acquired Color Vision Defects. Grune and Stratton,
New York, pp. 183–241.
16. Deeb,S.S., Lindsey,D.T., Hibiya,Y., Sanocki,E., Winderickx,J.,
Teller,D.Y. and Motulsky,A.G. (1992) Genotype–phenotype relationships
in human red/green color-vision defects: molecular and psychophysical
studies. Am. J. Hum. Genet., 51, 687–700.
17. Yamaguchi,T., Motulsky,A.G. and Deeb,S.S. (1997) Visual pigment gene
structure and expression in human retinae. Hum. Mol. Genet., 6, 981–990.
18. Hayashi,T., Yamaguchi,T., Kitahara,K., Sharpe,L.T., Jägle,H.,
Motulsky,A.G. and Deeb,S.S. (2001) The importance of gene order in
expression of the red and green visual pigment genes and in color vision.
Color Res. Appl., 26, 79–83.
19. Reichel,E., Bruce,A.M., Sandberg,M.A. and Berson,E.L. (1989)
An electroretinographic and molecular genetic study of X-linked cone
degeneration. Am. J. Ophthalmol., 108, 540–547.
20. Neitz,M., Neitz,J. and Jacobs,G.H. (1991) Spectral tuning of pigments
underlying red–green color vision. Science, 252, 971–974.
21. Ibbotson,R.E., Hunt,D.M., Bowmaker,J.K. and Mollon,J.D. (1992)
Sequence divergence and copy number of the middle- and long-wave
photopigment genes in Old World monkeys. Proc. R. Soc. London B, 247,
145–154.
22. Williams,A.J., Hunt,D.M., Bowmaker,J.K. and Mollon,J.D. (1992)
The polymorphic photopigments of the marmoset: spectral tuning and
genetic basis. EMBO J., 11, 2039–2045.
23. Neitz,J., Neitz,M., He,J.C. and Shevell,S.K. (1999) Trichromatic color
vision with only two spectrally distinct photopigments. Nat. Neurosci., 2,
884–888.
24. Deeb,S., Hayashi,T., Winderickx,J. and Yamaguchi,T. (2000) Molecular
analysis of human red/green visual pigment gene locus: relationship to
color vision. Methods Enzymol., 316, 651–670.
25. Miyahara,E., Pokorny,J., Smith,V.C., Baron,R. and Baron,E. (1998)
Color vision in two observers with highly biased LWS/MWS cone ratios.
Vision Res., 38, 601–612.
26. Winderickx,J., Battisti,L., Hibiya,Y., Motulsky,A.G. and Deeb,S.S.
(1993) Haplotype diversity in the human red and green opsin genes:
evidence for frequent sequence exchange in exon 3. Hum. Mol. Genet., 2,
1413–1421.
27. Neitz,M. and Neitz,J. (1998). Molecular genetics and the biological basis
of color vision. In Backhaus, W.G.K., Kliegl, R. and Werner, J.S. (eds),
Color Vision: Perspectives from Different Disciplines. Walter de Gruyter,
Berlin, pp. 101–119.
28. Winderickx,J., Lindsey,D.T., Sanocki,E., Teller,D.Y., Motulsky,A.G. and
Deeb,S.S. (1992) Polymorphism in red photopigment underlies variation
in colour matching. Nature, 356, 431–433.
29. Nathans,J., Maumenee,I.H., Zrenner,E., Sadowski,B., Sharpe,L.T.,
Lewis,R.A., Hansen,E., Rosenberg,T., Schwartz,M., Heckenlively,J.R.
et al. (1993) Genetic heterogeneity among blue-cone monochromats.
Am. J. Hum. Genet., 53, 987–1000.
30. Kazmi,M.A., Sakmar,T.P. and Ostrer,H. (1997) Mutation of a conserved
cysteine in the X-linked cone opsins causes color vision deficiencies by
disrupting protein folding and stability. Invest. Ophthalmol. Vis. Sci., 38,
1074–1081.
31. Crognale,M.A., Teller,D.Y., Motulsky,A.G. and Deeb,S.S. (1998)
Severity of color vision defects: electroretinographic (ERG), molecular
and behavioural studes. Vision Res., 38, 3377–3385.
32. Sanocki,E., Teller,D.Y. and Deeb,S.S. (1997) Rayleigh match ranges of
red/green color-deficient observers: psychophysical and molecular
studies. Vision Res., 37, 1897–1907.
33. He,J.C. and Shevell,S.K. (1995) Variation in color matching and
discrimination among deuteranomalous trichromats: theoretical
implications of small differences in photopigments. Vision Res., 35,
2579–2588.
34. Neitz,M., Neitz,J. and Jacobs,G.H. (1995) Genetic basis of photopigment
variations in human dichromats. Vision Res., 35, 2095–2103.