Evolution of Colour 2

Evolution and Genetics of Colour Vision

Basic Mammalian Condition

The basic mammalian condition is dichromatic, possessing a shortwave cone and a single spectral class of cone in the middle to long wave region.
This allows sensitivity to color differences along the blue to yellow axis but not the green to red axis.
As wavelengths increase toward the red end, the excitation of the shortwave cone diminishes, leaving only the signal from the middle wave cone.
Deep reds may appear as dark grey to black due to weak signals at longer wavelengths.

Gene Duplication and Spectral Sensitivity

Approximately 30 to 40 million years ago, a gene duplication of the middle wave opsin occurred.
Opsins are visual pigments that determine spectral sensitivity in cones.
Following gene duplication and minor amino acid sequence divergence, spectral sensitivity changed, resulting in two classes: middle wave and long wave sensitive cones.
A handful of amino acid substitutions could produce a three-nanometer difference in peak sensitivity between the medium and the long wave cones.
This mutation may have provided a selective advantage to tree-dwelling, fruit-feeding primates by enhancing their ability to distinguish yellow, orange, and reddish colors against green foliage.

Transduction of Light into Electrical Signals

Transduction of light into electrical signals relies on visual pigments comprising an opsin bound to a chromophore.
The chromophore is 11-cis-retinal which absorbs photons.
Opsins determine the likelihood of absorbing photons of different wavelengths, thus controlling spectral sensitivity.
Amino acid substitutions at key positions alter the wavelength of peak spectral sensitivity.
The difference in spectral sensitivity between M and L cones (approximately 530 and 560 nanometers) arises from a few amino acid differences in their respective opsin sequences.

Evolutionary History of Opsin Genes

Sequence analysis of opsin amino acid sequences across species helps reconstruct evolutionary history.
Vertebrates possess four classes of opsins at the base of their evolution.
Gene duplications have led to the emergence of long wave sensitive, shortwave sensitive one, shortwave sensitive two, and rod opsin families.
All classes of vertebrates have examples of opsins from these five families, indicating their presence at the base of vertebrate evolution around 540 million years ago.
Vertebrates typically have a duplex retina with a rod opsin and up to four spectral classes of cones.
Some species may have secondarily lost an opsin gene, as seen in mammals.

Mammalian Evolution and Dichromacy

Mammals evolved as small, nocturnal creatures, making four cone classes redundant for night vision.
They retained two cone classes: one with an opsin from the shortwave sensitive family and one with a sensitivity in the middle to long wave end of the spectrum.
Each species has its own versions of opsins from some or all of the four cone families and a version of one opsin from the rod family.
Spectral sensitivity can be altered by a few key amino acid substitutions.
Individual opsins from the long wave sensitive family can have spectral sensitivities ranging from approximately 500 to 570 nanometers.
Human long and middle wave cone opsins are derived from this family, with spectral sensitivities around these ranges.
The short wave sensitive family has versions tuned to around 400 to almost 450 nanometers in different species, with the human version around 420 nanometers.

Cone Spectral Sensitivities in Vertebrates

Parrots have retained members of all four classes of opsins, with two curves (red and yellowish) for each opsin family.
Humans have three cone classes: middle wave cone, long wave cone, and a short wave cone (S cone).
Humans lack opsin representatives from some families but have two from the long wave sensitive family (M and L cones).
Duplication of a long wave sensitive gene from the long wave opsin family, followed by divergence, created the M and L cones, enabling color vision on the red to green axis.
This occurred relatively recently, perhaps 30 million years ago.

Evolution of Mammalian Opsins

Modern mammals (therians) have retained a cone from the short wave sensitive family one and a cone from the long wave sensitive family, giving them the dichromatic condition.
All mammals have retained the rod opsin. More primitive mammals like monotremes and marsupials have taken different routes to reach the dichromatic condition.
About 30 million years ago, along the primate lineage in Africa, a gene duplication and sequence divergence occurred.
This resulted in two versions of the single gene from the long wave sensitive family, which are the M cones and L cones found in Old World primates.

Retinal Circuitry and Color Vision

The evolution of a new spectral class of opsin poses the question of whether new circuitry is needed to process signals from the new cone class.
This assumes only one opsin out of the M and L is expressed in any given cone, which is true.
Before duplication, primates had a single L cone instead of separate orange and green blobs organized in a centre-surround fashion.
The ancient subsystem, present in all mammals and primates, including New World primates, is served by the small bistratified ganglion cells.
In primates, prior to the split into M and L cones, there would have been a single, let's say, L cone here instead of just orange here, instead of green and orange blobs organized in a centre-surround fashion. That is via the midget bipolar cells. The centre is fed from a single cone in the centre of the receptive field to the ganglion cell, whereas the surround is fed via inhibition from a number of cones. Very different bipolar cells in addition to the immediate ganglion cell, hence excitation from the centre and inhibition from the surround. This is what gives you your chromatic contrast, which is particularly sensitive to borders.
After duplication and sequence divergence, two cone classes (orange and green) exist.
In the fovea, the centre of the receptive field receives input from either an M cone or an L cone because it comes from a single cone. It has to be one of these. So the centre here is going to have excitation in this case from an L cone. Now the inhibition is still coming from the surround, but the inhibition now could be coming from a random mixture of these two classes of cones.
Inhibition from the surround could come from a random mix of these two cone classes, resulting in spectral opponent signals.
The pre-existing centre-surround circuitry in the primate visual system was ready for wavelength opposition prior to the duplication and divergence splitting into M and L cones.
There may be no need to evolve additional circuitry.
As receptive fields get bigger as you move further away from the centre, then both the centre and the surround will sample multiple cones. So the possibility that opponents might get broken down does exist, although there is still a likelihood that there is going to be a spectral difference between the centre and the surround.
It remains possible that there are developmental mechanisms that cause the neurons taking inputs from the surround to be selective in the cones that they select to sample, such that if the central cone was L, maybe they selectively sample the M cones.
Evidence suggests that random sampling may be sufficient, but specific sampling might also occur to maintain wavelength opposition.

Genetic Mechanisms of Opsins

Mammalian opsins from the long wave sensitive family are found on the X chromosome.
In humans and other New World primates, two opsins are close together on the X chromosome: the long wave sensitive and, slightly downstream, the middle wave sensitive.
Upstream, a locus control region randomly selects either the M or the L cone opsin to be expressed in any given cone.
The original duplication likely involved a segment of chromosome containing the opsins in tandem, followed by sequence divergence, resulting in our current long wave and middle wave opsins.
A shift in spectral sensitivity of about 30 nanometers between M and L cones is thought to be accounted for by just a few amino acid substitutions in a few key positions in the opsin sequence.
Two amino acid substitutions in exon five, such as a shift at amino acid position 277 from tyrosine to phenylalanine, account for 21 nanometers of the spectral difference, with other shifts having smaller contributions.
Human L and M opsins broadly have peak spectral sensitivities of around about 513 and about 560
There can be allelic variations of a few nanometers in peak spectral sensitivity due to inter-individual differences.
An M cone might have 533 or 530, and an L cone 559, 562 or whatever, according to different allelic variants.

Human Color Deficiencies

Human dichromats inherit just either an M or an L cone opsin gene from that X chromosome
The sensitivities that you could get could be S combined with M (a form of dichromacy called protanopia) or S combined with L (dichromacy)
Anomalous trichromats inherit two genes which are allelic versions of the same coding class represented here by M and M dash.
They inherit cone opsins of slightly different spectral sensitivity, spectral sensitivity that might vary by three, four, even five nanometers.
Discrimination along the sort of yellowish-green to yellow to red end of the spectrum would be somewhat compromised, although somewhat better than in the case of a dichromat.
The same principle applies. It could be anomalous trichromats come in the two forms with an S cone combined with either two allelic versions of the M cone or of the L cone.

Polymorphic Colour Vision in New World Primates

Old World primates, including humans, are trichromats, with typical spectral sensitivities of S, M, and L.
New World primates are dichromatic.
In many species of New World primates, the long wave sensitive opsin comes in multiple alleles (e.g., three alleles).
The spectral separation in the peak sensitivities can be significant

Sex-linked color vision variance

Males inherit one copy of the X chromosome and are dichromatic, having an S opsin and one from the long wave sensitive class.
Females inherit two copies of the X chromosome and can be homozygous (same allele on both X chromosomes) or heterozygous (different alleles on different X chromosomes).
X inactivation means that only one X chromosome is read in a given cone.
Heterozygous females will randomly express one or the other alleles and will express any two of these possible different alleles. So they could be genuinely trichromatic, expressing this allele similar to our M opsin or similar to our L opsin.
Homozygous females are dichromatic like the males.
Heterozygous females genuinely have trichromatic colour vision, which their male and homozygous female conspecifics lack.
Trichromatic females may have more success foraging for ripe fruits with reddish colors.
Evidence supports the theory that divergence followed by duplication, giving rise to different cone classes, can be immediately exploited by existing retinal circuitry.