Why Sensory Ecology Needs to Become More Evolutionary
27
in combination with electrophysiological studies. References for measured spectra: Apis
mellifera UV and blue, Drosophila melanogaster Rh1-Rh4 (see Briscoe 1999). References
for in situ hybridization: D. melanogaster Rh5 (Papatsenko eta!. 1997); Rh6 (Huber eta!.
1997); Hemigrapsus sanguineus (Sakamoto et a!. 1996), Limulus polyphemus 1 and 2
(Smith et a!. 1993); Papilio xuthus PxRh1-3 (Kitamoto et a!. 1998). Reference for
immunohistochemistry: D. melanogaster Rh5 (Chou eta!. 1996)
This, in turn, is dependent on the ratio of amino acid substitutions that are
adaptively neutral to those that cause functional changes. If only a few amino acid
substitutions are responsible for spectral tuning, then the overwhelming majority
of amino acid changes used to reconstruct the phylogeny are going to reflect the
actual history of the gene family.
Although we do not yet know how many amino acid residues are involved in
spectral tuning of the insect visual pigments, we infer from research in vertebrates
(Hunt et al. 1996; Neitz et al. 1991) that this number is comparatively small
relative to the total number of amino acids of opsins. Therefore the pigments
which fall into distinct clades are similar because they share a common ancestry,
and not because natural selection has erased their history. This view is
corroborated by the fact that opsins in the distinct clades have intron splice sites
that are shared (Briscoe 1999).
In order to predict how easy it is to change spectral sensitivity as a response to
specific selective pressures, it is necessary to see how many amino acid replacements are required to cause such changes. The majority of studies on spectral
tuning have focused on the vertebrate visual pigments. These studies, which use
mutagenesis to tease apart the amino acids relevant to the spectral properties of
visual pigments, have made significant progress in furthering our understanding
of the relationship between vertebrate opsin structure and function.
Almost all of the variation in cone pigment absorption spectrum has been
accounted for: a mere one to nine specific amino acid substitutions are responsible
for the 10-100 nm differences between the vertebrate cone pigments (Asenjo et al.
1994; Sun et al. 1997; Lin et al. 1998). This suggests that adaptive spectral tuning
will be achieved relatively easy. On the other hand, previous studies have only
looked at the ease of tuning between pigments that already exist in nature. These
studies do not necessarily imply that spectral tuning in all spectral domains will
happen as readily. For example, the fact that only three amino acid replacements
are necessary to tum a human green receptor pigment O·max=530 nm) into a red
receptor pigment (Amax=560 nm) does not mean that generating a hypothetical
pigment with Amax=590 nm will require equally few changes.
In contrast to the large body of work that has led to the development of our
current model of spectral tuning in the vertebrate pigments, only one study to date
examines spectral tuning of insect opsins. Using a Drosophila transgenic expression system, Britt et al. (1993) studied the spectral properties of 13 chimeric
opsins created by exchanging one or more transmembrane domains of the
Drosophila Rh1 and Rh2 opsin genes. These chimeric opsin-encoding genes were
introduced into a mutant Drosophila strain using P element germline transformation, and expressed in the Rl-6 photoreceptor cells. The expression of the
27
in combination with electrophysiological studies. References for measured spectra: Apis
mellifera UV and blue, Drosophila melanogaster Rh1-Rh4 (see Briscoe 1999). References
for in situ hybridization: D. melanogaster Rh5 (Papatsenko eta!. 1997); Rh6 (Huber eta!.
1997); Hemigrapsus sanguineus (Sakamoto et a!. 1996), Limulus polyphemus 1 and 2
(Smith et a!. 1993); Papilio xuthus PxRh1-3 (Kitamoto et a!. 1998). Reference for
immunohistochemistry: D. melanogaster Rh5 (Chou eta!. 1996)
This, in turn, is dependent on the ratio of amino acid substitutions that are
adaptively neutral to those that cause functional changes. If only a few amino acid
substitutions are responsible for spectral tuning, then the overwhelming majority
of amino acid changes used to reconstruct the phylogeny are going to reflect the
actual history of the gene family.
Although we do not yet know how many amino acid residues are involved in
spectral tuning of the insect visual pigments, we infer from research in vertebrates
(Hunt et al. 1996; Neitz et al. 1991) that this number is comparatively small
relative to the total number of amino acids of opsins. Therefore the pigments
which fall into distinct clades are similar because they share a common ancestry,
and not because natural selection has erased their history. This view is
corroborated by the fact that opsins in the distinct clades have intron splice sites
that are shared (Briscoe 1999).
In order to predict how easy it is to change spectral sensitivity as a response to
specific selective pressures, it is necessary to see how many amino acid replacements are required to cause such changes. The majority of studies on spectral
tuning have focused on the vertebrate visual pigments. These studies, which use
mutagenesis to tease apart the amino acids relevant to the spectral properties of
visual pigments, have made significant progress in furthering our understanding
of the relationship between vertebrate opsin structure and function.
Almost all of the variation in cone pigment absorption spectrum has been
accounted for: a mere one to nine specific amino acid substitutions are responsible
for the 10-100 nm differences between the vertebrate cone pigments (Asenjo et al.
1994; Sun et al. 1997; Lin et al. 1998). This suggests that adaptive spectral tuning
will be achieved relatively easy. On the other hand, previous studies have only
looked at the ease of tuning between pigments that already exist in nature. These
studies do not necessarily imply that spectral tuning in all spectral domains will
happen as readily. For example, the fact that only three amino acid replacements
are necessary to tum a human green receptor pigment O·max=530 nm) into a red
receptor pigment (Amax=560 nm) does not mean that generating a hypothetical
pigment with Amax=590 nm will require equally few changes.
In contrast to the large body of work that has led to the development of our
current model of spectral tuning in the vertebrate pigments, only one study to date
examines spectral tuning of insect opsins. Using a Drosophila transgenic expression system, Britt et al. (1993) studied the spectral properties of 13 chimeric
opsins created by exchanging one or more transmembrane domains of the
Drosophila Rh1 and Rh2 opsin genes. These chimeric opsin-encoding genes were
introduced into a mutant Drosophila strain using P element germline transformation, and expressed in the Rl-6 photoreceptor cells. The expression of the