Download ARTICLE Visual pigment gene structure and expression in human

 1997 Oxford University Press
Human Molecular Genetics, 1997, Vol. 6, No. 7
981–990
ARTICLE
Visual pigment gene structure and expression in
human retinae
Tomohiko Yamaguchi, Arno G. Motulsky and Samir S. Deeb*
Departments of Medicine and Genetics, University of Washington, Seattle, WA 98195, USA
Received February 14, 1997; Revised and Accepted April 10, 1997
We determined the genotypes of the X-chromosome-linked red/green color vision genes by a novel
PCR/SSCP-based method and assessed expression by mRNA analysis in retinae of 51 unselected post mortem
eye specimens from Caucasian males of unknown color vision status. All individuals had a single red (long-wave)
pigment gene and one or more (an average of two) green (middle-wave) pigment genes. Four males had
5′green–red3′ hybrid genes in addition to normal red and green pigment genes. These findings are consistent with
earlier studies on human visual pigment gene structure using Southern blotting and with a recent study using
pulsed-field electrophoresis. We interpret claims of much larger numbers of red, green and green–red hybrid
genes to be technical artifacts. The ratio of expressed red to green pigment retinal mRNA varied widely (1–10 with
a mode of 4) and was not correlated with that of red to green pigment genes. In one individual with a green–red
hybrid gene in addition to normal red and green pigment genes, the normal red pigment gene and the hybrid gene
were both expressed, but the normal green gene was not. This person presumably had deuteranomalous color
vision. Two with green–red hybrid genes expressed the normal red and green pigment genes, but not the hybrid
genes. These two individuals presumably had normal color vision. We interpret the failure to express their
green–red hybrid genes to be caused by their location at a more distal position in the visual pigment gene array.
INTRODUCTION
Human color vision is initiated by the absorption of light by three
classes of cone photoreceptors, the short-wave or blue-,
middle-wave or green- and long-wave or red-sensitive cones that
have overlapping spectral sensitivity curves with wavelengths of
maximum absorption at ∼420, 530 and 560 nm, respectively (1–5).
The red and green pigment genes are arranged in a head-to-tail
tandem array on the long arm of the X-chromosome (Xq28) (6–8).
In individuals with normal color vision, the red pigment gene is
located 5′ of one or more green pigment genes. The most common
number of green pigment genes in an array was shown to be two
(with a range of 1–5) in the Caucasian population (9,10), but the
frequency distribution varies among males of different ethnic
origin (11,12). Each repeat unit in the array is comprised of a
pigment gene (15 kb) of six exons and of 24 kb of intergenic
sequences. The red and green pigment gene repeats are 98%
identical at the DNA sequence level (including introns and
3′-flanking sequences), and the different green pigment genes are
99.9% identical (6).
Individuals with normal color vision can match the color of a test
light of any wavelength composition by mixing the right
proportions of three primaries (red, green and blue), and their color
vision is therefore referred to as trichromatic. Individuals who lack
functional red or green cones are known as protanopes or
deuteranopes, respectively. They are able to match the color of a
test light by mixing two primaries only and therefore have
dichromatic color vision. Dichromacy occurs in ∼1–2% of
Caucasian males.
A milder form of defective color vision is anomalous
trichromacy which occurs in ∼5% of Caucasian males. Although
anomalous trichromats are believed to have three classes of
photoreceptors, one of these classes has anomalous spectral
sensitivity. Protanomalous males have normal blue and green
cones but anomalous red cones with a spectral sensitivity that is
significantly shifted towards that of the green cone.
Deuteranomalous individuals have normal blue and red cones
together with anomalous green cones (13) with a spectral
sensitivity that is significantly shifted towards that of the red cone.
The various common (8% of Caucasian males) red–green color
vision defects are associated with complete loss of green
photopigments or with the formation of red–green or green–red
hybrid pigments with shifted absorption maxima. Unequal
homologous recombination explains both the loss of pigment
genes and the formation of hybrid pigment genes (9).
5′Red–green3′ and 5′green–red3′ hybrid genes were first
discovered among individuals with protan and deutan color
vision defects, respectively (9,12). 5′Red–green3′ hybrid pigments are green-like and the 5′green–red3′ hybrid pigments are
*To whom correspondence should be addressed. Tel: +1 206 543 1706; Fax: +1 206 543 0754; Email: [email protected]
982
Human Molecular Genetics, 1997, Vol. 6, No. 7
red-like in their wavelength of maximum absorption (3–5). We
also observed that a small fraction (4–8% of unselected
Caucasian males) with normal color vision had, in addition to
normal red and green pigment genes, 5′green–red3′ hybrid genes
(10,12). To explain the lack of correspondence between genotype
(green–red hybrid genes) and color vision phenotype, we
hypothesized that not all opsin genes in an array are expressed in
a sufficient number of retinal cones to influence color vision. We
suggested that in deuteranomalous individuals the red and
green–red hybrid but not the normal green pigment genes were
expressed. In contrast, individuals with normal color vision who
carried normal red and green pigment genes as well as green–red
hybrid genes did not express the hybrid genes (12,14).
More recently, Neitz and Nietz claimed a much larger number
of green pigment genes in many individuals (15) and reported the
frequent finding of more than one red pigment gene per array
(15). However, in a later publication (16), they referred to the
additional red pigment genes as green–red hybrid genes and
suggested that such hybrid genes occurred in almost 50% of the
population.
To re-examine these problems of structure and function, we
devised a novel method to assess the number of red and green
pigment genes, as well as of various visual pigment hybrid genes.
We then studied expression of these visual pigment genes in post
mortem retinae by mRNA analysis. This approach allowed us to
determine directly the ratio of expressed red to green pigment
genes and whether green–red hybrid genes were expressed or not.
Our findings are consistent with the interpretation that only
expressed green–red hybrid genes cause deuteranomaly.
RESULTS
Structure of the red/green pigment gene arrays
A novel method for determining genomic structure. We developed
a rapid and reliable method for the determination of the structure
of the X-chromosome-linked visual pigment gene arrays. This
method was applied to arrays of 51 unselected Caucasian male post
mortem retinae to study the relationship of gene structure to the
patterns of retinal expression of these genes by quantitative mRNA
analysis. The strategy used in determining genomic structure is
depicted in Figure 1A. Competitive PCR, using primers that
completely match the sequence of both red and green pigment gene
sequences, was used to amplify segments from the proximal
promoter. The ratio of green to red promoter fragments amplified
from genomic DNA of males was used to calculate the total
number of genes in an array. Competitive PCR of red and green
exon 4 sequences was then used to detect hybrid genes in which the
fusion occurred 5′ of exon 4. Such amplified sequences were
separated electrophoretically (based on a few differences in
nucleotide sequence) on a single-strand conformation polymorphism (SSCP) gel and quantified by phosphorimage analysis.
Hybrid genes with fusion in intron 4 were detected by
amplification with gene-specific primers (Table 1 and Fig. 1A)
followed by electrophoresis on agarose gels as previously
described (12). Autoradiographs of SSCP gels showing red and
green DNA fragments amplified from the promoter and exon 4 are
shown in Figure 1B and C, respectively.
Table 1. Sequence and location of oligodeoxynucleotide primers
I.D. no.
Position and Sequence (5′→3′)
Location
80
(–190) *CCAGCAAATCCCTCTGAGCCG
Promoter
81
(41) GGCTATGGAAAGCCCTGTCCC
Promoter
55
(286) *AGAAGCTGCGCCACCCGCT
Exon 2
24B
(449) ACACAGGGAGACGGTGTAGC
Exon 2
30
(621) *TACTGGCCCCACGGCCTGAAG
Exon 4
2B
(785) CGCTCGGATGGCCAGCCACAC
Exon 4
7G
(730) ACCCCACTCAGCATCATCGT
Exon 4
7R
(730) ATCCCACTCGCTATCATCAT
Exon 4
3C
(786) GTGGCAAAGCAGCAGAAAGAG
Exon 5
78
(922) *TTGGCAGCAGCAAAGCATGCG
Exon 5
8G
(878) GAAGCAGAATGCCAGGACC
Exon 5
8R
(878) GACGCAGTACGCAAAGAT
Exon 5
The asterisk (*) indicates the primer in each pair that was 5′ end-labeled prior
to use in PCR amplification.
Number of red and green pigment genes per array. Earlier studies
indicated the presence of a single red pigment gene and one or
more green pigment genes per X chromosome (6–9). The
position of the red pigment gene at the 5′ edge of the array,
abutting single copy DNA sequence, explains lack of variation in
its copy number. Therefore, in a single X chromosome, only one
copy of the red pigment gene promoter should exist, while one or
more copies of the green pigment gene promoter (depending upon
the number of green pigment genes) may be found. Band density
ratios corresponding to the green and red pigment gene promoters
plus one equals the total number of genes in the array, i.e. a ratio
of 2:1 would indicate a total number of three genes.
In order to compare the results obtained by the PCR/SSCP
method with those determined by Southern blot analysis earlier
(9), we selected 20 individuals for whom we had determined the
ratio of green to red pigment genes by Southern blot analysis
(10,11) and subjected their DNA to PCR/SSCP analysis of the
promoter region. The plot (Fig. 1D) of the results obtained by
these two methods shows excellent correspondence for green to
red pigment gene ratios up to 4.
The ratios of green to red promoter segments of X chromosomes from 51 unselected Caucasian male eyes are given in Table
2 (those who did not carry hybrid genes) and Table 3 (those who
carried hybrid genes). The structures of gene arrays predicted
from these ratios are shown in Figure 2. Fifteen individuals
(29.4%) had one red pigment and one green pigment gene, 26
(51.0%) had one red and two green pigment genes and five (9.8%)
had one red and three or more green pigment genes. The
frequency distribution of the total number of genes per array was
similar to that reported previously for unselected Caucasian
males using Southern blot analysis (6,10).
Frequency of individuals with hybrid pigment genes. Hybrid
genes in which the fusion occurred in intron 4 were detected by
PCR amplification using green- or red-specific primers in exons
4 and 5 (Fig. 1A). Hybrids in which the fusion occurred 5′ of exon
4 were detected by competitive PCR amplification of red and
green exon 4 followed by SSCP analysis and band intensity
comparisons.
983
Human
Genetics,
1997,
6, No.
NucleicMolecular
Acids Research,
1994,
Vol. Vol.
22, No.
1 7
Table 2. Ratios of green to red sequences in human genomic DNA and in
retinal mRNA for individuals with no hybrid pigment genes
I.D. number
G/R promoter
G/R exon 4
genomic DNA
R/G exon 4
mRNA
2507
2534
2759
2760
2791
2792
2817
2832
2836
2871
2970
2996
3011
3022
3024
3041
3042
3043
3049
3050
3051
3056
3062
3063
3064
3068
3069
3076
3077
3078
3079
3098
3099
3100
3103
3106
3120
3401
3402
3403
3406
3411
3417
3443
3445
3447
3448
3457
3463
1.92 (0.08) ∼2
1.13 (0.02) ∼1
1.97 (0.12) ∼2
1.95 (0.16) ∼2
1.70 (0.13) ∼2
1.99 (0.01) ∼2
0.93 (0.02) ∼1
0.95 (0.07) ∼1
1.82 (0.03) ∼2
1.16 (0.06) ∼1
1.11 (0.08) ∼1
1.81 (0.02) ∼2
1.53 (0.11) ∼1
0.93 (0.08) ∼1
0.89 (0.01) ∼1
1.82 (0.00) ∼2
4.48 (0.21) ∼4
1.94 (0.03) ∼2
2.14 (0.02) ∼2
1.88 (0.11) ∼2
4.10 (0.03) ∼4
2.07 (0.14) ∼2
2.15 (0.05) ∼2
0.98 (0.04) ∼1
1.95 (0.21) ∼2
1.94 (0.07) ∼2
3.89 (0.01) ∼4
2.77 (0.25) ∼3
1.93 (0.11) ∼2
0.95 (0.03) ∼1
0.97 (0.02) ∼1
2.05 (0.33) ∼2
1.22 (0.27) ∼1
1.05 (0.08) ∼1
1.07 (0.06) ∼1
1.86 (0.02) ∼2
1.89 (0.01) ∼2
1.18 (0.06) ∼1
1.91 (0.04) ∼2
1.92 (0.06) ∼2
1.95 (0.03) ∼2
2.01 (0.04) ∼2
1.96 (0.01) ∼2
1.85 (0.13) ∼2
1.66 (0.09) ∼2
1.01 (0.04) ∼1
1.06 (0.04) ∼1
2.27 (0.07) ∼2
3.18 (0.33) ∼3
2.28 (0.60) ∼2
1.30 (0.38) ∼1
1.77 (0.42) ∼2
1.96 (0.55) ∼2
1.93 (0.12) ∼2
1.87 (0.07) ∼2
1.05 (0.08) ∼1
1.53 (0.40) ∼1
1.93 (0.07) ∼2
1.06 (0.05) ∼1
1.02 (0.02) ∼1
2.02 (0.13) ∼2
1.12 (0.06) ∼1
1.10 (0.03) ∼1
1.12 (0.03) ∼1
1.98 (0.06) ∼2
2.61 (0.87) ∼3
1.94 (0.42) ∼2
1.88 (0.18) ∼2
2.18 (0.37) ∼2
4.15 (1.96) ∼4
1.86 (0.10) ∼2
1.89 (0.08) ∼2
0.98 (0.03) ∼1
1.89 (0.12) ∼2
1.87 (0.07) ∼2
5.15 (0.38) ∼5
2.79 (0.20) ∼3
1.91 (0.45) ∼2
1.01 (0.12) ∼1
1.02 (0.08) ∼1
1.97 (0.14) ∼2
1.07 (0.08) ∼1
1.09 (0.09) ∼1
1.04 (0.04) ∼1
1.96 (0.07) ∼2
2.00 (0.15) ∼2
1.15 (0.07) ∼1
1.98 (0.16) ∼2
1.99 (0.29) ∼2
1.94 (0.18) ∼2
1.82 (0.04) ∼2
1.92 (0.14) ∼2
2.05 (0.04) ∼2
2.31 (0.45) ∼2
1.16 (0.06) ∼1
1.09 (0.04) ∼1
2.47 (0.22) ∼2
3.16 (0.25) ∼3
4.63 (1.52) ∼5
3.25 (0.04) ∼3
3.47 (0.99) ∼3
3.52 (0.55) ∼4
3.88 (0.70) ∼4
7.42 (2.08) ∼7
4.14 (0.22) ∼4
3.77 (0.98) ∼4
1.68 (1.13) ∼2
10.41 (1.70) ∼10
4.08 (1.18) ∼4
5.22 (0.60) ∼5
6.85 (2.56) ∼7
3.83 (0.34) ∼4
3.96 (0.52) ∼4
4.68 (1.65) ∼5
1.23 (0.19) ∼1
4.25 (0.92) ∼4
2.73 (1.04) ∼3
2.63 (0.37) ∼3
3.80 (0.56) ∼4
4.96 (0.70) ∼5
3.58 (1.42) ∼4
4.26 (0.21) ∼4
2.72 (1.01) ∼3
4.38 (1.02) ∼4
3.12 (0.45) ∼3
2.88 (0.61) ∼3
3.32 (1.04) ∼3
3.44 (0.09) ∼3
4.21 (0.61) ∼4
5.16 (0.48) ∼5
3.44 (0.37) ∼3
8.26 (0.95) ∼8
3.52 (0.65) ∼4
5.46 (1.07) ∼5
5.91 (0.77) ∼6
4.86 (0.84) ∼5
4.26 (0.68) ∼4
7.28 (1.59) ∼7
4.01 (1.31) ∼4
3.12 (0.90) ∼3
2.47 (0.52) ∼2
2.01 (0.72) ∼2
4.15 (1.03) ∼4
6.58 (3.15) ∼7
12.73 (8.36) ∼13
2.3 (0.14) ∼2
3.2 (0.37) ∼3
Numbers represent means (standard error) of three independent determinations
with the values rounded to the nearest whole number. Green to red (G/R) ratios
for the promoter and exon 4 were determined using total genomic DNA as template. Red to green (R/G) mRNA ratios for exon 4 were determined using reverse-transcribed total retinal RNA as template.
983
Hybrid genes in which the points of fusion occurred 5′ of exon
4 were detected by determining the ratio of green to red exon 4
genomic sequences and comparing it with that for the promoter
segments. If the gene array is comprised of normal red and green
pigment genes and no hybrid genes, then the green to red ratio of
the amplified promoter segments would be the same as that of
fragments amplified from exon 4. Inequality in these ratios would
indicate the presence of green/red hybrid genes in the array. For
example, a green:red ratio of 2:1 for the promoter and a ratio of 1:2
for exon 4 would indicate the presence of a 5′ green–red 3′ hybrid
gene (fusion 5′ of exon 4), in addition to one normal red and one
normal green pigment genes (subject A in Fig. 2 and Table 3).
Five of 51 individuals were found to have hybrid genes. Three
had 5′green–red3′ hybrid genes: one with a G2-3/R4 (see Materials
and Methods for explanation of nomenclature of hybrid genes) and
two with a G4/R5 hybrid gene in addition to one normal red and
one normal green pigment genes (individuals labeled A, B and C,
respectively, in Fig. 2 and Table 3). The ratios of green/red genomic
segments corresponding to the promoter and exons 2, 4 and 5 are
given in Table 3A. Two others (labeled D and E in Fig. 2 and Table
3) had, in addition to normal green pigment genes, single red
pigment genes in which exon 4 only was replaced by exon 4 of a
green pigment gene (R3/G4/R5 type of hybrid). One of these (E)
had in addition a green–red hybrid gene with a fusion in exon 4
(G4/R5). Unlike the previously employed method of Southern
blotting, the PCR and SSCP methods used in this study allowed
unequivocal identification of R3/G4/R5 hybrid genes. Thus, a total
of four individuals (8%) had simple 5′green–red3′ hybrid genes.
The color vision phenotype of the donors of these four post mortem
specimens is unknown.
Expression of red and green visual pigment genes in
whole retinae
Expression of normal pigment genes. Total cellular RNA was
prepared from whole retinae of 51 male eye donors and analyzed
for the ratio of red to green exon 4 mRNA sequences by reverse
transcription, followed by competitive PCR amplification and
electrophoretic separation of red- from green-specific exon 4
DNA strands (SSCP) as described in Materials and Methods. A
standardization curve relating expected and observed ratios of red
to green exon 4 sequences was constructed by including purified
DNA templates that contained artificial mixtures of known ratios
of red to green exon 4 sequences in the PCR mixture. An
autoradiograph of the SSCP gel and a plot of the standard curve
are shown in Figure 3A and B. This assay gave a linear response
over a wide range of red to green ratios (0.2–10). Exon 4 was
chosen for this assay since it contains the least number of
nucleotide differences (three of 165) between red and green
sequences so that differences in amplification rate based solely on
sequence differences would be minimized. Furthermore, the
amplified DNA strands derived from red and green exon 4
sequences can be separated easily by conformational differences
on SSCP. Analysis of exon 5 by this method is complicated by the
presence of a common polymorphism in codon 283 (CCA/CCC)
of the green pigment gene. The red:green pigment mRNA ratios
(not corrected using a standard curve) determined using exon 5
on eight eye specimens were on average 14% higher than those
determined using exon 4 (data not shown).
984
Human Molecular Genetics, 1997, Vol. 6, No. 7
Figure 1. Determination of the number and composition of the red–green opsin genes in arrays of unselected Caucasian male eye donors. (A) Strategy for determining
the structure of the array. DNA fragments derived from the red and green pigment genes were competitively amplified by PCR using genomic DNA as template and
primers (arrows) that completely matched both red and green gene sequences. The sequence of the primers (the identification numbers of which are given below the
arrows) is given in Table 1. One of the primers was radiolabeled at its 5′ end. The PCR-amplified segments were denatured and the single strands of the red and green
gene segments, which differ in sequence at a few positions, were resolved by electrophoresis on a non-denaturing polyacrylamide gel (SSCP). (B) Autoradiograph
of an SSCP gel showing the ratios of green to red (G/R) promoter DNA fragments (G, green strand; R, red strand) generated using total genomic DNA as template.
The results of two independent PCR amplification reactions are shown for eye donor (I.D. numbers for each pair, from left to right, are 3103, 3120, 3401, 3402, 3403
and 3406). The level of radioactivity in the DNA strands derived from the red and green pigment genes was determined by phosphorimage analysis. The ratios were
rounded to the nearest whole number. (C) Autoradiograph of an SSCP gel showing the ratios of green to red (G/R) exon 4 DNA fragments generated using total genomic
DNA as template. Ratios (above lanes) were rounded to the nearest whole numbers. The DNA samples shown in this autoradiograph belong to the following subjects
(see Table 2): from left to right 2534, 2759, 2760, 2817, 2832, 2970, 3063, 3078, 3099, 3401 and 3463). (D) Correspondence between green to red gene ratios obtained
by Southern blotting and PCR/SSCP analyses of the promoter region.
An autoradiograph of an SSCP gel showing single strands of
exon 4 of the red and green pigment genes in 14 males is shown
in Figure 4A. The red:green exon 4 mRNA ratios in retinae of the
46 (out of 51) male eye donors without hybrid genes are given in
Table 2 and the frequency distribution is shown in Figure 4B.
There was wide variation in this ratio. The red pigment gene,
which is located closest to the locus control region (LCR) of the
pigment genes (see Fig. 6), was expressed at higher levels (on
average 4-fold) than the green pigment gene in 45 of 46 subjects.
Furthermore, this ratio was only moderately and inversely
correlated (r = 0.351) to the ratio of green to red pigment genes
in the arrays of these individuals (Fig. 4C). There was no
statistically significant difference (t-test) between the means of
red:green mRNA ratios of the groups who had one or two green
pigment genes.
Expression of hybrid pigment genes. Among three eye donors (of
a total of 51) who carried 5′green–red3′ hybrid genes in addition
to normal green and red pigment genes (subjects A, B and C in
Fig. 2 and Table 3A), we investigated whether the normal green
or the 5′green–red3′ hybrid genes were expressed. Failure of
expression of the hybrid gene may provide an explanation for the
association of normal color vision with the presence of 5′green–
red3′ hybrid genes.
985
Human
Genetics,
1997,
6, No.
NucleicMolecular
Acids Research,
1994,
Vol. Vol.
22, No.
1 7
985
Table 3. Ratios of green to red sequences in human genomic DNA and in retinal mRNA for subjects A–E (see Fig. 2) who carried hybrid genes
Subject I.D. number
A. Green/red ratios in genomic DNA
Promoter
Exon 2
Exon 4
Exon 5
A
3263
2.04 (0.04) ∼2
1.99 (0.22) ∼2
0.54 (0.02) ∼0.5
0.49 (0.06)∼0.5
B
3023
1.92 (0.12) ∼2
1.79 (0.24) ∼2
1.96 (0.08) ∼2
0.44 (0.05)∼0.5
C
3061
1.85 (0.16) ∼2
1.84 (0.15) ∼2
1.97 (0.15) ∼2
0.46 (0.03)∼0.5
D
3119
1.52 (0.63) ∼2
1.22 (0.02) ∼1
green only
0.63 (0.03) ∼1
E
2858
2.72 (0.01) ∼3
2.86 (0.12) ∼3
green only
1.82 (0.19) ∼2
B. Red/green mRNA ratios in whole retinae
Subject
I.D. number
Exon 2
Exon 4
Exon 5
A
3263
9.35 (1.47) ∼9
red only
red only
B
3023
4.38 (0.46) ∼4
5.08 (1.10) ∼5
4.15 (0.46) ∼4
C
3061
4.98 (1.22) ∼5
4.00 (1.02) ∼4
4.37 (0.46) ∼4
D
3119
5.18 (0.32) ∼5
green only
7.77 (0.95) ∼8
E
2858
3.57 (0.29) ∼4
green only
4.12 (0.54) ∼4
Numbers represent means (standard error) of three independent determinations and the values rounded to the nearest whole number.
Exons 2, 4 and 5 of the red and green pigment genes and of the
corresponding retinal RNA from subjects A, B and C were
subjected to competitive PCR/SSCP analysis as described in
Materials and Methods. The results (Fig. 5A) showed that the
retinae of all three subjects contained mRNA sequences corresponding to exons 2, 4 and 5 of the red pigment gene in ratios
shown in Table 3B. Donor A expressed red and green exon 2 but
only red exons 4 and 5, indicating that he expressed the normal
red and green–red hybrid genes but not the normal green pigment
gene of his array. Donors B and C expressed both red and green
exons 2, 4 and 5, indicating that they expressed both the normal
red and green pigment genes of their arrays (B and C gave the
same results and only that of B is shown in Fig. 5A). However,
this experiment could not rule out expression of their green–red
hybrid in addition to the normal red and green pigment genes.
We therefore examined directly whether the green–red hybrid
genes of subjects B and C, together with their normal red and
green pigment genes, are expressed in their retinae. Primers 7G
[specific for green exon 4 sequence (12)] and 78 (matches both
red and green exon 5 sequences) (Table 1, Fig. 1A) were used to
amplify competitively a segment extending from exon 4 to exon
5 sequences that are transcribed from either the normal green or
the green–red hybrid pigment genes. SSCP analysis of the
RT-PCR products obtained from retinal mRNA of donors A and
B, as well as from donors who had normal red and green but no
hybrid genes, is shown in Figure 5B. RNA from donor E served
as a positive control since he expressed green exon 4–red exon 5
hybrid RNA from his proximal red pigment gene with the green
exon 4. No red–green hybrid mRNA sequences were detected
using competitive PCR amplification in retinal RNA of donors B
and C.
The above results suggest that the red and hybrid but not the
green pigment gene were expressed in subject A, whereas the red
and the normal green, but not the hybrid, genes were expressed
in subjects B and C. Therefore, we infer that donor A had
deuteranomalous color vision whereas donors B and C had
normal color vision.
The lack of expression of the green pigment gene of donor A
and of the green–red hybrid genes of donors B and C in their
retinae is unlikely to be due to the presence of inactivating
mutations, since the sequence of the proximal 230 bp of the
promoter as well as of all coding sequences were found to be
normal, as determined by SSCP analysis (data not shown).
Subject E also had a G4/R5 hybrid gene in addition to a
R3/G4/R5 hybrid and two normal green pigment genes, but he
was not informative regarding hybrid gene expression since
mRNA corresponding to exons 4 and 5 of his two hybrid genes
would be indistinguishable. This subject expressed a G4/R5
hybrid mRNA (but not a red exon 4 mRNA) as well as a normal
green pigment mRNA (Fig. 5B, Table 3B). Subject D, who had
a R3/G4/R5 in addition to a normal green pigment gene, also
expressed a G4/R5 hybrid mRNA (from his hybrid red pigment
gene) as well as a normal green pigment mRNA (Table 3B).
DISCUSSION
Structure of the red/green visual pigment gene arrays
We have developed a rapid and reliable SSCP-based method for
determining the structure of the X-chromosome-linked red and
green color vision genes. Using this method, we determined the
numbers and structure of red and green pigment genes in 51
unselected Caucasian male eye donors. The results show that the
great majority of males have arrays with one red pigment gene and
an average of two green pigment genes and rarely more than three
green pigment genes. The number and gross structure of gene
arrays was very similar to those previously reported (9,10) for
unselected Caucasian males (combined total of 152) as determined
by Southern blot analysis. The frequency distribution of the total
number of genes in an array determined by the PCR/SSCP method
was also similar to that observed in 67 unselected males using the
direct method of pulsed field gel electrophoresis (17). All these
results are unlike those reported by Neitz and Neitz who claimed
a much larger number of pigment genes per array (15,16). These
authors determined the structure of red/green gene arrays of 27
Caucasian color-normal males using PCR to amplify corresponding segments of the red and green pigment genes, followed by
digestion with restriction enzymes that distinguish red from green
986
Human Molecular Genetics, 1997, Vol. 6, No. 7
Figure 2. Diagram illustrating the types and frequencies of gene arrays observed among 51 unselected male eye specimens. Black and white arrows and arrow segments
represent red (R) and green (G) pigment gene sequences, respectively. Ellipses represent promoter (P) regions. The dotted line in the third array from the top indicates
the presence of three or more green pigment genes. G3/R4 hybrid indicates that exons 1–3 are derived from green and exons 4–6 are derived from red pigment gene
sequences. G4/R5 are hybrids comprised of green exons 1–4 and red exons 5 and 6. R3/G4/R5 are hybrids in which exon 4 of the red pigment gene has been replaced
by a green exon 4.
sequences. They reported that more than half of these 27 males
have arrays that contained a total number of four or more genes. It
is unlikely that this discrepancy is due to the small size of the
sample they studied, since individuals with six or more genes are
extremely rare in the general Caucasian population. Artifacts
related to their PCR method used to estimate the ratio of green to
red gene promoter sequences may explain both the unusually large
numbers of genes per array and the high proportion of arrays with
hybrid genes. It appears that the promoter ratio for most of their
individuals who carried more than three genes per array was
overestimated by a factor of two. This misinterpretation is
suggested by the high proportion (12/15) of such individuals with
an even number of genes.
We observed that four of 51 subjects had 5′green–red hybrid
genes in addition to normal red and green pigment genes. This type
of hybrid gene is typically found in deuteranomalous individuals
(9,12,18), but occasionally among individuals with normal color
vision (12). Again, these results are at odds with those of Neitz and
Neitz (15) who reported that more than half of the 27 color-normal
males carried two or more red pigment genes. However, these
additional red pigment genes were referred to as 5′green–red3′
hybrid genes in a subsequent report on the gene arrays of the same
individuals (16). The hybrid genes in their study occurred only in
males with four or more genes per array. This interpretation could
have also resulted from overestimation of the green to red promoter
ratio relative to that of exon 5 in those subjects who are presumed
to carry more than three genes per array.
Expression of the red and green visual pigment genes in
retinae
The ratio of red to green mRNA transcripts observed in extracts
of whole retinae varied widely among males, with a modal value
of four. This ratio may reflect a higher number of red cones in the
normal retina. However, variation in the level of expression of
these genes among individual cones may also contribute to the
observed mRNA ratios. A number of investigators, using
psychophysical techniques, have observed wide individual
variations in the red to green cone ratio with a modal value of ∼2
(19–22). A narrower range of ratios (1.46–2.36) was observed by
Cicerone and Nerger (23). All these studies used small and
centrally located test fields. The observed ratios were therefore
representative of the foveal region. Microspectrophotometric
measurements on seven eyes of individuals with normal color
vision gave an average red to green cone ratio in the foveal region
of 1.39 (1). One explanation for the discrepancy between our
results on cone ratios derived from mRNA analysis and the foveal
ratio of ∼2 may be that this ratio is higher in the peripheral retina
than in the fovea. Since only ∼10% of cones are located in the
fovea, the ratios we obtained from whole retinal mRNA represent
largely extrafoveal cones. There is evidence to suggest that the
ratio of red to green cones, as determined by retinal mRNA
analysis, is higher (4:1 vs 2:1) in the peripheral than in the central
retina (24). On the other hand, estimates of red to green cone
ratios (average of 2 with a range of 0.67–9.0) obtained on 16
subjects with normal color vision by flicker photometric electroretinography (ERG) were similar using either small or very
large test fields (25) and suggest that the red to green ratio does
not change significantly throughout the retina. More studies will
be necessary to resolve the differences between results of mRNA
analysis and ERG.
Our results which indicate that not all genes of the array are
expressed in the retina are consistent with our earlier findings in
which differential expression of green pigment genes in the retina
was observed. In these studies, we took advantage of a common
synonymous polymorphism (C/A) in exon 5 of the red pigment
987
Human
Genetics,
1997,
6, No.
NucleicMolecular
Acids Research,
1994,
Vol. Vol.
22, No.
1 7
987
located downstream of the two proximal pigment genes are less
likely. We therefore suggest that only the two most proximal
pigment genes of an array are expressed in sufficient cones to
influence color vision as measured anomaloscopically. Based on
this model, if a male subject has a normal red, normal green and
a green–red hybrid gene, deuteranomaly would result only if the
green–red hybrid gene occupies the proximal position and is
expressed preferentially over the normal green opsin gene.
R3/G4/R5 hybrid genes would be expected to encode a
red-sensitive pigment with an absorption maxima shifted by 2–3
nm towards the green (3–5). Two of the amino acid residues
(positions 230 and 233) that contribute to spectral tuning of the
red and green visual pigments are located in exon 4. The normal
red pigment gene has Ile and Ala while the green pigment gene
has Thr and Ser at these two position, respectively. We previously
observed two males who had R3/G4/R5 hybrid genes in addition
to normal green pigment genes (28). Although these two subjects
had normal color vision as determined by anomaloscopy, they
required more red light in the mixture of red and green lights to
match the standard yellow light than did individuals with normal
red pigment genes (28).
MATERIALS AND METHODS
Subjects
Human retinal tissue and blood samples were collected from 51
unselected post mortem male Caucasian donors from between 4
and 8 h after death. These specimens were obtained through the
Lions Eye Bank of the University of Washington, Seattle. DNA
was prepared from peripheral blood leukocytes and total cellular
RNA was prepared from whole retinae as previously described
(14). The anterior part (including the lens) was removed from the
enucleated eyes and the retinae were peeled off and either
immediately used for extracting RNA or stored at –70C.
Figure 3. Standardization curve relating red:green ratios of artificial mRNA
transcript mixtures of known composition (input) with the observed red:green
ratios of these mixtures. Templates that contained mixtures of known ratios of
red to green exon 4 mRNA sequences were used as templates in quantitative
PCR/SSCP analysis. (A) Autoradiograph of the SSCP gel. (B) A plot of
expected vs observed red to green mRNA ratios.
gene to address this question (14). Analysis in 22 male eye donors
(14) who had two or more green pigment genes in their genomic
DNA showed clearly that, whenever the two alleles of exon 5
were present in genomic DNA, only one was represented in
retinal mRNA in sufficient quantities to be detected by competitive PCR amplification.
The model illustrated in Figure 6 was proposed (14) to explain
such selective expression in the retina. An LCR, located between
3.1 and 3.7 kb 5′ of the red pigment gene transcription initiation
site, was shown to be required for cone photoreceptor-specific
expression of both red and green pigment genes (26,27). The
LCR was postulated to regulate expression of the opsin genes of
the array in a distance-dependent manner (14). Thus, if the LCR
forms a stable transcriptionally active complex with the red opsin
gene promoter, a red-sensitive cone will result. If the LCR forms
a stable complex with the proximal green pigment gene promoter
instead, a green-sensitive cone is formed. Transcriptionally active
complexes between the LCR and promoters of pigment genes
Rapid method for determination of number and ratios
of red and green pigment genes
DNA fragments derived from the red and green pigment gene
promoters (from –190 to +41 with respect to the transcription start
site) were amplified competitively by PCR using genomic DNA
as template and primers (#80 and 81) that completely matched
both red and green gene sequences (Table 1 and Fig. 2). One of
the primers (#80) was radiolabeled with 32P at its 5′ end, which
resulted in tagging of the sense strands of the red and green
amplified products. The labeling reaction for primers contained,
in a total volume of 15 µl, 15 pmol of deoxyoligonucleotide, 90
µCi of [γ-32P]ATP (New England Nuclear; sp. act. 3000
Ci/mmol), 15 U of bacteriophage T4 polynucleotide kinase and
kinase buffer (Bethesda Research Laboratories). The reaction
mixture was incubated at 37C for 45 min and then at 70C for
15 min. The labeled oligonucleotide was purified on a G-25 spin
column (Pharmacia) according to the manufacturer’s protocol.
Typically, the oligonucleotide primer was labeled to a specific
activity of 1×106 c.p.m./pmol, diluted with cold oligonucleotide
to a specific activity of 5×104 c.p.m./pmol and the concentration
adjusted to 5 µM for use in PCR amplification.
The amplification reaction contained, in a total volume of 10 µl,
100 ng of genomic DNA, 0.5 U of Taq polymerase, amplification
buffer (Perkin-Elmer), 200 µM (each) dNTPs, 0.5 µM of each
988
Human Molecular Genetics, 1997, Vol. 6, No. 7
Figure 4. Frequency distribution of the ratio of red to green pigment mRNAs in whole retinae of male eye specimens and its relationship to the number of green pigment
genes per array. Total RNA was extracted from retinae of human male donors and analyzed for relative abundance of red to green exon 4-encoded sequences. (A)
Autoradiograph of an SSCP gel showing intensities of bands derived from red (R) and green (G) exon 4 mRNA sequences. Lanes marked M contain markers for the
red and green pigment bands. Phosphorimage analysis was used to determine band intensity. Ratios of red to green band intensities were rounded to the nearest whole
numbers. (B) Frequency distribution of the ratio of red to green exon 4 mRNA sequences derived from exon 4. Individuals with hybrid pigment genes (Fig. 2) have
been excluded from this histogram and their ratios are presented separately in Figure 5. (C) A plot of the number of green pigment genes per X-chromosome as a
function of the red:green ratio in exon 4 mRNA. Linear, exponential and logarithmic correlation coefficients were 0.35, 0.39 and 0.40, respectively.
primer (one of which is labeled) and 10% glycerol. An initial
denaturation at 96C for 5 min was followed by 26 cycles of
amplification at 94C for 30 s and extension at 64C for 1 min
and a final 4 min of extension.
The PCR-amplified segments were denatured and the single
strands of the red and green gene promoter segments, which differ
in sequence at a few positions, were separated from each other by
electrophoresis (5–6 h at a gel temperature of 35C) on a 6%
non-denaturing polyacrylamide gel (SSCP, 29). The radioactivity
in the DNA strands derived from the red and green gene
promoters was determined by PhosphorImage analysis.
Detection of green/red hybrid genes
Detection of green/red hybrid genes with fusion points in intron
4 was performed using primers in exon 4 and 5 that are specific
for either red or green pigment gene sequences (#7G, 7R, and 8G
and 8R in Table 1 and Fig. 2) according to the procedure described
in detail elsewhere (12).
Detection of hybrid genes with fusion points 5′ of exon 4 was
done by competitively amplifying exon 4 of red and green
pigment genes followed by SSCP analysis. The amplification
primers for exon 4 (#30 and 2B) are shown in Table 1 and Figure
1A. The amplification reaction mixtures and conditions were the
same as for amplification of promoter sequences except that no
glycerol was included. The conditions for SSCP analysis were the
same as for the promoter segment except that the gel temperature
for exon 4 was 28C. The red:green ratios for the promoter and
exon 4 segments were compared. Inequality in these ratios
indicated the presence of hybrid genes with fusion points 5′ of
exon 4.
Nomenclature of hybrid genes. The hybrid pigment genes are
referred to here by the abbreviations originally used by Merbs and
Nathans (3) to indicate the origin of various exons. For example,
a G4/R5 gene is a hybrid in which exons 1–4 are derived from the
green pigment gene and exons 5 and 6 are derived from the red
pigment gene (point of fusion in intron 4). A G2-3/R4/G5 is a
hybrid in which only exon 4 of a red pigment gene is substituted
with that of a green pigment gene. The designation G2-3 is used
since exon 3 of the red pigment gene is indistinguishable from that
of a green pigment gene due to the existence of several shared
polymorphisms (30).
Determination of the ratio of red to green mRNA
transcripts in retinae
Total cellular RNA was prepared from whole retinae of 51 male
eye donors, reverse transcribed into cDNA and used as template
in PCR amplification and SSCP analysis essentially as described
(30). Exons 2, 4 and 5 of the red and corresponding green
sequences were amplified competitively in 25 cycles using the
primers shown in Figure 1A and Table 1, with one of each set
primers radiolabeled at its 5′ end with 32P as described above. The
amplification and SSCP conditions for exons 2 and 5 were the
same as described above for the promoter, except for the omission
of glycerol. As in genomic DNA analysis, one member of the pair
of primers was end labeled in order to radiolabel only one of the
989
Human
Genetics,
1997,
6, No.
NucleicMolecular
Acids Research,
1994,
Vol. Vol.
22, No.
1 7
989
Figure 5. Ratio of mRNA transcribed from green and green–red hybrid genes in retinae. (A) Expression of hybrid vs normal green pigment genes. Total retinal RNA
from post mortem eye donors A, B and C (see Fig. 2), whose arrays were comprised of a normal red, a normal green and a green–red hybrid gene, was used as template
in competitive RT-PCR/SSCP analysis to determine the ratio of mRNA derived from red and green exons 2, 4 and 5. Donors B and C gave similar patterns, and only
the pattern of B is shown. A donor (I.D #3448; Table 1) who had one normal red and one normal green and no hybrid genes (No Hyb.) in his array was included as
a control in this analysis. Results obtained using genomic DNA as template are also shown. D and R indicate DNA or RNA used as template, respectively. Two markers
for green exon 5 were included: one representing CCA and the other CCC of the polymorphic codon 283 of the green pigment gene. Subject A expressed red and
green exon 2 but only red exons 4 and 5, indicating lack of expression of the normal green pigment gene of his array. Donors B and C expressed both red and green
exons 2, 4 and 5. This result does not rule out expression of the hybrid gene as well by these two donors since exon 5 of the hybrid and normal pigment genes are
indistinguishable in this experiment. (B) Lack of expression of exon 4–5 hybrid mRNA in retinae of donors B and C. The letters above the lanes indicate the subject
analyzed. The gene arrays of subjects B, C and E are described in Figure 2 and are indicated in Table 2. N (I.D. #3448, Table 2) is a normal control. Primers 7G (matches
only green exon 4 sequence) and 78 (matches both red and green exon 5 sequence) were used to competitively amplify green exons 4–5 and green exon 4–red exon
5 hybrid segments using reverse-transcribed retinal RNA as template. RNA transcribed from the normal green but not from the hybrid pigment gene was observed.
As expected, subject E expressed the green exon 4–red exon 5 hybrid in his red pigment gene. The upper band in lanes B, C and N represent single strands of green
fragments with a less probable conformation than the main population of single strands.
Figure 6. Model illustrating the role of position of the green–red hybrid gene
in the color vision phenotype. Red-sensitive cones are postulated to result from
the exclusive expression of the red pigment gene (R, solid arrows) as a result
of stable coupling (mediated by DNA-binding proteins) of the LCR to the red
pigment gene promoter. Green-sensitive cones are formed if the LCR
preferentially and permanently couples to the proximal green pigment gene (G,
open arrows) promoter and turns on its expression. Distal green pigment gene
promoters (including those that belong to green–red hybrid genes) are activated
with very low probability (dotted arrows). Normal color vision results if the
green–red hybrid gene (H) occupies the distal position in the array, and
deuteranomal color results if the hybrid gene occupies the proximal position.
(Diagram adapted from 14).
DNA strands of the red and green exons and reduce the
complexity of autoradiographs. The choice of 25 cycles of
amplification was based on assays of DNA mixtures that
contained known ratios of red to green transcripts derived from
exon 4. Exon 4 fragments were amplified from genomic DNA of
individuals who have only red or green exon 4 genomic DNA
sequences, purified by electrophoresis on agarose gels and their
concentrations determined. Aliquots were then mixed with a
100-fold excess of yeast DNA and used as templates in
RT-PCR/SSCP analysis.
Detection of mRNA transcripts containing green–red hybrid
sequences with fusion points in intron 4 (G4/R5 hybrids) and of
transcripts derived from normal green pigment genes was done by
competitive amplification (after reverse transcription) using
primers #7G (specific for green exon 4) and #78 (identical in
sequence to both red and green exon 5) (Table 1 and Fig. 5B).
Primer #78 was end labeled with 32P and the amplification
conditions were the same as for exon 4. The amplified fragments
(derived both from green–red hybrid and normal green pigment
genes) were subjected to SSCP analysis for 5 h at a gel
temperature of 33C. The radioactivity in bands corresponding to
hybrid and normal DNA segments was quantitated by phosphor
image analysis.
990
Human Molecular Genetics, 1997, Vol. 6, No. 7
Sequence analysis
Screening by SSCP for sequence variants in the coding regions
of the red and green pigment genes as well as direct sequencing
of PCR products have been described in detail elsewhere (30).
The proximal promoter sequences were amplified using primers
1 and 2 (Table 1) followed by SSCP analysis under the conditions
used above to determine the ratio of red to green promoter
sequences.
ACKNOWLEDGEMENTS
We thank the Lions Eye bank of the University of Washington for
supplying the specimens and Lori Iwasaki for the preparation of
DNA. Part of this work was presented at the XIII International
Conference on Color Vision Deficiencies in Pau, France, and
appeared in the proceedings of this conference: C.R. Cavonius,
Editor. Kluwer Academic Publishers, Doerdrecht, 1997, pp.
21–31. This work was supported by NIH Grant EY08395.
REFERENCES
1. Dartnall, H.J.A., Bowmaker, J.K. and Mollon, J.D. (1993) Human visual
pigments: microspectrophotometric results from the eyes of seven persons.
Proc. R. Soc .Lond, Ser B, 220, 115–130.
2. Smith, V.C. and Pokorny, J. (1975) Spectral sensitivity of the cone
photopigments between 400 and 500 nm. Vision Res., 15, 161–171.
3. Merbs, S.L. and Nathans, J. (1992) Absorption spectra of the hybrid pigments
responsible for anomalous color vision. Science, 258, 464–466.
4. Merbs, S.L. and Nathans, J. (1992) Absorption spectra of human cone
pigments. Nature, 356, 433–435.
5. Asenjo, A.B., Rim, J. and Oprian, D.D. (1994) Molecular determinants of
human red/green color discrimination. Neuron, 12, 1131–1138.
6. Nathans, J., Thomas, D. and Hogness, D.S. (1986) Molecular genetics of
human color vision: the genes encoding blue, green and red pigments.
Science, 232, 193–202.
7. Vollrath, D., Nathans, J. and Davis R.W. (1988) Tandem array of human
visual pigment genes at Xq28. Science, 240, 1669–1672.
8. Feil, R., Aubourg, P., Heilig, R. and Mandel, J.L. (1990) A 195-Kb cosmid
walk encompassing the human Xq28 color vision pigment genes. Genomics,
6, 367–373.
9. 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.
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. Jorgensen, A.L., Deeb, S. and Motulsky, A.G. (1990) Molecular genetics of X
chromosome-linked color vision among populations of African and Japanese
ancestry: high frequency of a shortened red pigment gene among AfroAmericans. Proc. Natl Acad. Sci. USA, 87, 6512–6516.
12. 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.
13. 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, p. 183.
14. Winderickx, J., Battisti, L., Motulsky, A.G. and Deeb, S.S. (1992) Selective
expression of the human X-linked green pigment genes. Proc. Natl Acad. Sci.
USA, 89, 9710–9714.
15. Neitz, M. and Neitz, J. (1995) Numbers and ratios of visual pigment genes for
normal red–green color vision. Science, 267, 1013–1016.
16. Neitz, M., Neitz, J. and Grishok, A. (1995) Polymorphism in the number of
genes encoding long-wavelength-sensitive cone pigments among males with
normal color vision. Vision Res., 35, 2395–2407
17. Macke, J.P. and Nathans, J. (1997) Individual variation in the size of the
human red and green visual pigment gene array. Invest. Ophthalmol. Vis. Sci.,
38, in press.
18. Neitz, M., Neitz, J. and Kainz, P.M. (1996) Visual pigment gene structure and
the severity of color vision defects. Science, 274, 801–804.
19. de Vries, H.L. (1947) The heredity of the relative numbers of red and green
receptors in the human eye. Genetica, 24, 199–212.
20. Rushton, W.A.H. and Baker, H.D. (1964) Red/green sensitivity in normal
vision. Vision Res., 4, 75–85.
21. Vimal, R.L.P., Pokorny, J., Smith, V.C. and Shevell, S.K. (1989) Foveal cone
thresholds. Vision Res., 29, 61–78.
22. Wesner, M.F., Pokorny,J., Shevell, S.K. and Smith, V.C. (1991) Foveal cone
detection statistics in color normals and dichromats. Vision Res., 31,
1021–1037.
23. Cicerone, CM and Nerger, J.L. (1989) The relative numbers of long-wavelength-sensitive and middle-wavelength-sensitive cones in the human fovea
centralis. Vision Res., 19, 115–128.
24. Neitz, M., Hagstrom, S.A., Kainz, P.M. and Neitz, J. (1996) L and M cone
gene expression in the human retina: relationship with gene order and retinal
eccentricity. Invest. Ophthalmol. Vis. Sci., 37, S448.
25. Jacobs, J. and Neitz, J. (1993) Electrophysiological estimates of individual
variation in the L/M cone ratio. In Drum, B. (ed.), Color Vision Deficiencies
XI. Proceedings of the International Symposium on Color Vision Deficiencies, Sydney, Kluwer, pp. 107–112.
26. Nathans, J., Maumenee, I.H., Zrenner, E., Sadowski, B., Sharpe, L.T., Lewis,
R.A. et al. (1993) Genetic heterogeneity among blue-cone monochromats.
Am.J. Hum. Genet., 53, 987–1000.
27. Wang, Y., Macke, J.P., Merbs, S.L., Zack, D., Klaunberg, B., Bennet, 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.
28. Winderickx, J. Lindsey, D.T., Sanocki, E., Teller, D.Y., Motulsky, A.G. and
Deeb, S.S. (1992) A Ser/Ala polymorphism in the red photopigment
underlies variation in colour matching among coulor-normal individuals.
Nature, 356, 431–433.
29. Orita, M., Suzuki, Y., Sekiyu, T. and Hayashi, K. (1989) Rapid and sensitive
detection of point mutations and DNA polymorphisms using the polymerase
chain reaction. Genomics, 5, 874–879.
30. 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.