28
Lars Chittka and Adriana Briscoe
chimeric opsin genes in the Rl-6 cells restored normal spectral sensitivity function to the mutant flies. Using electrophysiology and microspectrophotometry,
Britt et al. (1993) measured changes in, respectively, the spectral sensitivities of
the R 1-6 photoreceptor cells and the absorption spectra of the chimeric visual
pigments. This approach allowed them to monitor the effects of exchanging
different Rh I and Rh2 transmembrane domains on the spectral tuning of the
opsins.
Britt et al. (1993) found that no single transmembrane domain was responsible
for the 60 nm difference between Rh 1 ( 480 nm) and Rh2 ( 420 nm). Exchange of a
single Rh 1 transmembrane domain (TM) for the corresponding Rh2 domain
resulted in a 4-10 nm shift towards shorter wavelengths for most domains except
TM 4, which resulted in an 11 nm shift to longer wavelengths. Replacement of
almost all TM domains (TM2-7) was required to convert a Rh 1 transgene into a
Rh2-like opsin ( 436 nm). Unlike the three to seven amino acid residues responsible for tuning the primate red and green cone pigments (Neitz et al. 1991),
which differ by 15-30 nm, the Drosophila Rhl and Rh2 TM domains do not
interact in an additive fashion. For example, replacement of only Rh1 TM6 (with
a Rh2 TM6) results in a 12 nm shift to shorter wavelengths, and replacement of
Rh 1 TM7 alone results in a 4 nm in the same direction. However, replacement of
both Rh 1 TM6 and 7 simultaneously results in a 20 nm shift to longer wavelengths. The authors propose two mechanisms to account for spectral tuning, one
of wTiich is involved in large-scale or coarse spectral tuning, and the other in finescale tuning, as exemplified by the human red and green cone pigments. While
fine tuning in the vertebrate pigments is apparently nearly additive in effect (Hunt
et al. 1996), coarse tuning in the case of the Drosophila opsins occurs in a combinatorial manner, involves many more TM domains (Britt et al. 1993), and occurs
over a larger evolutionary time scale. Finally, not all mutations in visual pigment
genes are neutral or cause functional changes; some are downright deleterious.
For example, three different missense mutations in the gene that codes for the
human blue sensitive pigment all cause complete loss of the human short-wave
receptor function, or even cell death (Deeb and Motulsky 1996).
Besides spectral fme tuning of existing photopigments, color vision systems can
also evolve by changing the number of color receptor types. Although clearly the
number of such types is conservative in many taxa, there is also some variation.
For example, more than 40 species of Hymenoptera have 3 color receptor types
most sensitive in the UV, blue, and green, but there are at least three species which
have red receptors in addition (Peitsch et al. 1992). Such increases in color
receptor types occur through gene duplication and subsequent spectral tuning
(Briscoe 1999; Briscoe 1998a). How common are opsin gene duplications? The
crab Hemigrapsus sanguineus (Sakamoto et al. 1996) and the horseshoe crab
Limulus polyphemus (Smith et al. 1993) have unique opsin gene duplications,
found in no other species so far. The problem is that we are not sure if these
duplications are species specific, or whether they are more basal in larger
taxonomic groups, such as genera or even orders. The butterflies Papilio glaucus
and Papilio xuthus, for example, share two gene opsin gene duplications (Briscoe
1998a; Kitamoto et al. 1998); as do Drosophila melanogaster and Drosophila
Lars Chittka and Adriana Briscoe
chimeric opsin genes in the Rl-6 cells restored normal spectral sensitivity function to the mutant flies. Using electrophysiology and microspectrophotometry,
Britt et al. (1993) measured changes in, respectively, the spectral sensitivities of
the R 1-6 photoreceptor cells and the absorption spectra of the chimeric visual
pigments. This approach allowed them to monitor the effects of exchanging
different Rh I and Rh2 transmembrane domains on the spectral tuning of the
opsins.
Britt et al. (1993) found that no single transmembrane domain was responsible
for the 60 nm difference between Rh 1 ( 480 nm) and Rh2 ( 420 nm). Exchange of a
single Rh 1 transmembrane domain (TM) for the corresponding Rh2 domain
resulted in a 4-10 nm shift towards shorter wavelengths for most domains except
TM 4, which resulted in an 11 nm shift to longer wavelengths. Replacement of
almost all TM domains (TM2-7) was required to convert a Rh 1 transgene into a
Rh2-like opsin ( 436 nm). Unlike the three to seven amino acid residues responsible for tuning the primate red and green cone pigments (Neitz et al. 1991),
which differ by 15-30 nm, the Drosophila Rhl and Rh2 TM domains do not
interact in an additive fashion. For example, replacement of only Rh1 TM6 (with
a Rh2 TM6) results in a 12 nm shift to shorter wavelengths, and replacement of
Rh 1 TM7 alone results in a 4 nm in the same direction. However, replacement of
both Rh 1 TM6 and 7 simultaneously results in a 20 nm shift to longer wavelengths. The authors propose two mechanisms to account for spectral tuning, one
of wTiich is involved in large-scale or coarse spectral tuning, and the other in finescale tuning, as exemplified by the human red and green cone pigments. While
fine tuning in the vertebrate pigments is apparently nearly additive in effect (Hunt
et al. 1996), coarse tuning in the case of the Drosophila opsins occurs in a combinatorial manner, involves many more TM domains (Britt et al. 1993), and occurs
over a larger evolutionary time scale. Finally, not all mutations in visual pigment
genes are neutral or cause functional changes; some are downright deleterious.
For example, three different missense mutations in the gene that codes for the
human blue sensitive pigment all cause complete loss of the human short-wave
receptor function, or even cell death (Deeb and Motulsky 1996).
Besides spectral fme tuning of existing photopigments, color vision systems can
also evolve by changing the number of color receptor types. Although clearly the
number of such types is conservative in many taxa, there is also some variation.
For example, more than 40 species of Hymenoptera have 3 color receptor types
most sensitive in the UV, blue, and green, but there are at least three species which
have red receptors in addition (Peitsch et al. 1992). Such increases in color
receptor types occur through gene duplication and subsequent spectral tuning
(Briscoe 1999; Briscoe 1998a). How common are opsin gene duplications? The
crab Hemigrapsus sanguineus (Sakamoto et al. 1996) and the horseshoe crab
Limulus polyphemus (Smith et al. 1993) have unique opsin gene duplications,
found in no other species so far. The problem is that we are not sure if these
duplications are species specific, or whether they are more basal in larger
taxonomic groups, such as genera or even orders. The butterflies Papilio glaucus
and Papilio xuthus, for example, share two gene opsin gene duplications (Briscoe
1998a; Kitamoto et al. 1998); as do Drosophila melanogaster and Drosophila