Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Cone Visual Pigments in Monkeys and Humans JAM ES K. BOWMAKER Color vision in humans is based on three spectrally different classes of photoreceptors, that is, three classes of cones containing different visual pigments. The absorbance spectra of these different pigments were first es- tablished by direct microspectrophotometric measurements from individual cones by Marks et al. in 1964. Their data, along with the data of Brown and Wald (1964), showed the expected clear evidence for three types of cones with maximum absorbance, Ama=' at about 560, 530, and 440 to 460 nm. Since then various attempts have been made to establish with more precision the Oman and the shape of the absorbance spectra of the three pigments in humans (Dartnall et al., 1983) and to compare them with those found in Old World monkeys (Bowmaker et al., 1983; MacNichol et al., 1983~. The visual pigments of New World monkeys have also been inves- tigated and correlated with the marked variations in color vision found in these species (e.g., Mollon et al., 1984b). This paper reviews the present position, as determined from microspec- trophotometry, and addresses the following questions: How many visual pigments are available to primates? Are there spectrally preferred posi- tions for the pigments? If so, can they be related to anomalous pigments in humans? In addition, some controversy has arisen as to the spectral location of short-wave receptors in primates (Mansfield et al., 1984), and I shall present further evidence from both Old and New World monkeys suggesting differences between species. Microspectrophotometric recording from primate photoreceptors is difficult: the cone outer segments are small, with diameters of only 1 to 2 ~m, and they break down structurally within a relatively short time. Thus, the transverse absorbance spectra recorded from them have maximum densities of only about 0.015 to 0.020 with low signal-to-noise ratios. (For 19
20 JAMES K BOWMAKER a review see MacNichol et al., 1983~. In analyzing the data we set rigid criteria by which to select records from individual cones and then estimate the Oman by fitting a template curse to the absorbance spectrum (see Mollon et al., 1984, for details). Ideally, the template curve should be expressed in such a way that it remains constant in shape when displaced throughout the spectrum, and my co-workers and I have previously found that the Dartnall nomogram (based on frog rhodopsin with Max at 502 nm) expressed on a scale of >~/4 iS a reasonable approximation (Barlow, 1982; Mollon et al., 1984b). More recently, Mansfield (1985) and MacNichol (1986) have demonstrated that a transformation to F/Fmax (frequency/frequency maximum) produces a better fit to experimental data, and we have adopted this template in the analysis of all the data presented here. The difference in the Max determined from the two different templates is of course zero near 500 nm (both based on frog rhodopsin), but it increases to about 2 nm at 565 nm, with the F/Fmax value being the shorter, since at longer wavelengths the transformation, when expressed on a wavelength basis, yields a slightly broader curve than the >~/4 transformation. LONG- TO MIDDLE-WAVE PIGMENTS Old World Monkeys and Humans The cone pigments in macaques are well established. It is clear that the two pigments in the red-green spectral region have Max at about 565 and 535 nm (Bowmaker et al., 1983; MacNichol et al., 1983~. We have now had the opportunity to look at three other groups of Old World primates: the baboon, Papio papio (Bowmaker et al., 1983), and two types of guenon, the talapo~n, Cercopithecus (Miopithecus) talapoin, and the moustached guenon, Cercopithecus cephus. These species extend the range of platyrrhines studied. In addition, C. cephus is of particular interest because of its facial coloration. The moustached guenon, or blue-faced monkey, is characterized by a strikingly blue face surrounded by yellow hair, and we hoped, perhaps naively, that its retina would have a somewhat higher concentration of short-wave cones. We were fortunate in obtaining data from six such cones, about 10 percent of the total, more than double the numbers we have recorded from other species (Table 1~. It is clear from Figure 1 that the distributions of cones in the four species of Old World primates are similar and that the mean Oman for the long-wave cones and that for the middle-wave cones are almost identical, with overall means of 565.2 and 535.1 nm (Table 1~. The one rhesus monkey listed in the present data yielded relatively few long-wave cones with a mean Oman at about 562 nm, but our earlier data (Bowmaker et al., 1983) suggest a Oman closer to 565 nm.
CONE VISUAL PIGMENTS IN MONKEYS AND HUMANS 12 4 12 4 12 .~' 1 2 z 4 20 12 4 l ~ ILL PA TA MG l RH Total 530 550 570 Wavelength nm 21 FIGURE 1 Distribution of values of peak sensitivity of indiviual cones (maximally sensitive in the red-green spectral range) from four species of Old World primates: PA, Papio papio (baboon) (four animals); TA, Cercopithecus talapoin (talapoin); MG, Cercopithecus cephus (moustached guenon); and RH, Macaca mulatta (rhesus). The bottom panel is a summary histogram of the four species.
22 TABLE 1 Old World Primate Visual Pigments ~ n nm) JAMES K BOWMAKER Short-wave Middle-wave Long-wave Macaca mulatta 432.8 + 4.7 534.6 + 3.1 561.5 + 2.1 (Rhesus) (1) (1) (14) (7) Cercopithecus 430.5 + 2.2 533.0 + 1.1 566.5 + 1.4 cephus (1) (6) (23) (18) Cercopithecus 430.1 + 2.5 534.3 + 1.3 566.1 + 2.3 talapoin (1) (2) (23) (12) Papio papio 428.5 + 3.3 536.4 + 1.2 564.5 + 1.2 (baboon) (4) (4) (40) (30) Total Old 430.9 + 0.9 535.1 + 0.5 565.2 + 0.9 World (7) (13) (100) (67) Human 419.0 + 2.1 531.4 + 0.7 563.7 + 0.6 (5) (4) (71) (102) NOTE: Numbers in parentheses are either numbers of animals or numbers of cells analyzed. Max values are + S.D. Are the cone pigments in humans similar to those in Old World pri- mates? We have recently made measurements from another five human eyes enucleated because of melanoma; the distributions of the long- and middle-wave cones are shown in Figure 2. These distributions are remark- ably similar to those of the Old World primates, with mean Oman of 563.7 and 531.4 nm Table 1~: a similarity reflected in the spectral sensitivity mea- surements obtained from suction electrode techniques both in macaques (Baylor et al., 1987) and in humans (Schnapf et al., 1987~. The bimodality in the populations of both long- and middle-wave cones suggested in our earlier data from human eyes (Dartnall et al., 1983) was not apparent in the present study. A difference that is apparent between Old World monkeys and humans, however, is the ratio of long- to middle-wave cones. In each of the four species of Old World primates, there are fewer long-wave cones with an overall ratio of 0.77 (101:131), whereas in the human data the situation is reversed, with a ratio of long to middle of 1.67 (164:98~. These ratios are based on the total number of cells identified, not the numbers of records that passed our criteria, as listed in Able 1. Similar ratios were found in our earlier studies: 0.73 for macaques (Bowmaker et al., 1983) and 1.3 for humans (Dartnall et al., 1983~. [The ratios given by Boynton (1988, Table 1) are incorrect: he reversed our numbers of long- and middle-wave cones.] In their suction electrode measurements of macaque foveal cones, Baylor et al. (1987) found a ratio closer to 1.0 with 192 long-wave cones and 211 middle-wave cones. It is not clear whether the difference we found
CONE VISUAL PIGMENTS IN MONKEYS AND HUMANS 12 4 4 4 28 20 12 H15 i H13 H10 4 4 i L H14 H12 t ~ Total 530 550 570 Wavelength rim 23 FIGURE 2 Distribution of values of peak sensitivity of individual cones (maximally sensitive in the red-green spectral range) from the eyes of five persons. The bottom panel is a summary histogram of the five individuals.
24 JAMES K BOWM'9KER between Old World monkeys and humans is real or simply the result of sampling biases. New World Monkeys The cone pigments present in New World monkeys are markedly different from those of Old World monkeys, for New World monkeys show a striking polymorphism of visual pigments maximally sensitive in the red-green spectral region. This first became apparent in studies of color vision in the squirrel monkey, Saimi?~ sciureus (Jacobs et al., 1981~. It is now clear that this species has three pigments available in the middle to long wavelengths with Oman (determined from microspectrophotometry) at about 536, 550, and 563 nm (Mollon et al., 1984b). The pigments with Oman at about 536 and 563 nm are very similar in spectral location to those of Old World monkeys, but the P550 appears to be unique to New World species. The cone pigment complement in squirrel monkeys varies between individuals such that all males possess only one pigment in the red-green spectral region. This, combined with a short-wave pigment (see below), allows the animal only dichromatic color vision (Mollon et al., 1984b). However, the single pigment in the middle- to long-wave region may be any one of the three available to the species. In the females the position is more complex in that some animals are dichromats, resembling the males, whereas others enjoy trichromacy and possess, in addition to the short- wave pigment, a combination of any two of the three available longer-wave pigments (13owmaker et al., 1985, 1987~. The polymorphism of pigments and the resulting variation in color vision between individuals can be explained by a genetic model (Mollon et al., 1984b; Jacobs and Neitz, 1985) that is distinctly different from that proposed for the inheritance of color vision in humans and, by inference, in all Old World primates. The model postulates that there is only one genetic locus for a pigment in the red-green spectral range; · there are three alleles that can occur at the locus, the three al- leles corresponding to the three slightly different opsins of the photopigments; · the locus is on the X-chromosome; and · in those females that are heterozygous at the locus, only one of the two alleles is expressed, owing to Lyonization or X-chromosome inactivation, so that only one pigment is manufactured in any one ce11.
CONE VISUAL PIGMENTS IN MONKEYS AND HUMANS 25 The polymorphism expressed in squirrel monkeys is not unique to the cebid monkeys. It occurs in a similar form in the other major group of New World monkeys- the callitrichids (Travis et al., 1985; Jacobs et al., 1987~. In the common marmoset, Callithr~c jacchus jacchus, three pigments are available in the red-green spectral region, but these are different from those in Saimiri, with Oman at about 543, 557, and 564 nm Travis et al., 1988~. The P564 is again similar to the long-wave pigment in Old World monkeys, but the two pigments at shorter wavelengths are not common to Old World species, although there are reports of cones from macaques containing a pigment with Oman; at about 543 nm (Bowmaker et al., 1978; MacNichol et al., 1983~. As with Saimir`, it appears that all males are dichromats, pairing one of the three pigments with a short-wave pigment common to all individuals, whereas females may be either dichromatic or trichromatic. At present we have good evidence for only one form of trichromacy, the P564 combined with the P543 Travis et al., 1988) but with indications (unpublished) of the other two predicted variants. In its simplest form the genetic model suggests that two-thirds of females are trichromatic. However, within our limited sample of marmosets, there appear to be fewer trichromats than the model predicts. . Spectral Location of Pigments Within the primates both from the Old World and the New World, there would appear to be distinct spectral locations for cone photopigments in the red-green spectral range. Thus, Old World monkeys (including humans) have pigments at 535 and 565 nm, while New World monkeys may have one of these but may also have intermediate pigments with Oman at about 543, 550, and 557 nm. This is illustrated in Figure 3, which shows the pigments found in each species, and there are indications of specific spectral locations each separated by about 6 to 7 nm. The same spacing can be applied within mammals generally (see Jacobs, 1987) and is reminiscent of the idea of "clustering" of visual pigments first proposed by Dartnall and Lythgoe (1%5) for extracts of rod visual pigments (see also Knowles and Dartnall, 1977, pp. 195-199~. Presumably there are constraints on the structure of opsins such that variations in their amino acid sequences do not result in visual pigments with a continuum of spectral sensitivities but in pigments that lie at discrete spectral locations. Can such a hypothesis reveal anything about the spectral location of anomalous pigments in colour-deficient human observers? The work of Nathans et al. (1986) has shown that the opsin sequences of the long- and middle-wave cone pigments in humans are more than 90 percent homologous and that in anomalous trichromats the hybrid genes probably
26 0 ._ ~ 4 Q o 2 JAMES K BOWMAKER 430 530 550 570 Waveleng th nm FIGURE 3 Distnbution of values of the mean peak sensitivities of the cone types identified in humans, four species of Old World Inmates, and two species of New World primates. Note the dominant peak wavelengths at about 565, 535, and 430 nm and the suggestion of clustering with a wavelength interval of about 5 to 8 nm. encode for the pigments that lie between the normal long- and middle- wave pigments. The clustering of primate pigments implies that there are potentially three spectral locations for these pigments543, 550, and 557 nm. It is interesting that in analyses of psychophysical measurements on anomalous observers, Pokorny et al. (1973) estimated that the normal and anomalous pigments are separated by only 6 to 7 nm. Thus, protanomalous observers may have a P535-P542 combination, whereas deuteranomalous observers may have a P564 combined with a P557. Unfortunately, no direct measurements of anomalous pigments by either microspectrophotometry or suction electrode techniques have been made. The only such measurements from the eye of a color-deficient observer are those from a deuteranope, where only short- and long-wave cones were identified (Mollon et al., 1984a). SHORT-WAVE CONES Old World Monkeys and Humans Attempts to establish by microspectrophotometry the spectral location and shape of the absorbance spectra of short-wave cone pigments have been made more difficult by the relative rarity of short-wave cones, normally less than 10 percent of the cone population, and by the problems of short- wave scatter and the probable contamination of spectra by photoproducts (for a discussion, see Mansfield et al., 1984~. The best estimate for the Oman of short-wave cones in macaques would appear to be about 430 nm
CONE VISUAL PIGME~S IN MON=YS ED HUMS 27 (MacNichol et al., 1983; Mansfield et al., 1984~; our more recent data from four species of plattyrhines support this value (Table 1~. In the moustached guenon we obtained data from six short-wave cones; the mean absorbance spectrum is shown in Figure 4a. These prebleach spectra may be contaminated by short-wave light scatter and by photostable pigments. Each cone was exposed to white light to bleach the photopigment, and the spectrum was measured again, from which a difference spectrum representing the bleached photopigment could be derived (Figure 4a). The difference spectrum should eliminate problems caused by light scatter and photostable pigments, but it may be distorted by photoproducts. The AmaX of the prebleach and that of the difference spectra were then determined by fitting the template curve to the right-hand limb. As can be seen, both sets of data are best fitted by a template with Oman; at about 431 nm. Similar fits were obtained for the rhesus, talapoin, and baboons, and the overall Oman was again 431 nm. This value is identical to that reported by Baylor et al. (1987) from suction electrode measurements of spectral sensitivity in short-wave cones from macaques. In humans the situation appears to be different. In Figure 4b the mean spectra from four human short-wave cones are shown, with a template curve with )\maX 420 nm superimposed. Although the difference spectrum Is narrower than the template, it is clear that the spectra lie at wavelengths distinctly shorter than those of Old World monkeys. The Oman of 420 nm agrees with our earlier published data for human short-wave cones (Dartnall et al., 1983~. New World Monkeys In the rebid, Saimir' sciureus, we have data from 16 short-wave cones. The spectra, both prebleach and difference, are best fitted by a template curve with Oman at about 430 rim (for spectra, see Mollon et al., 1984b; Bowmaker et al., 1985~. The photopigment is clearly spectrally very similar to those in Old World monkeys. However, the short-wave cones in the callitrichid, Callithr~x jacchus, appear to be different. The mean spectra from 14 cones are shown in Figure 4c and are best fitted by a template with Ama2; at about 424 nm (Travis et al., 1988~. Spectral Location of Pigments As with the photopigments maximally sensitive in the red-green spectral region, the short-wave pigments of primates also appear to cluster at distinct spectral locations those of Old World monkeys and squirrel monkeys cluster at about 431 nm, that of the marmoset at about 424 nm, and that of humans at about 420 nm (Figure 3~. The mean spectra for the three pigments are shown together in Figure 4d, where the spectral displacements
28 JAMES K BOWMAKER 100 O L O I ~ I I r I I I r I ~ I r I I I r I I I lil I ~ I ~ ~ ! I , \ _~ 4) c o 53 o 100 _ ,f\~; ~ ,. '. ~ ~ a~ ~ . .~ - I ~ ~ ~ I ~ 1 l 1 i ~, ,,r I I' 'Tl 111 1 I r 1-11 _f _ _ !, 1 . , 1,, I 1 1 t' 1 1 ,, 1 . , 1 , 1 ~,, - | ! I I ~ | I ~ I I i I ~ ~ I I I I ~ ~ I I I I I I I ~ I I i ~ I ~ ~ I ~ I ~ I I ~ ~ I ~ I ' I ~ ~ I I I I ~ ~ ~ I I I 450 550 650 450 550 650 r I I T-r I I I 1 1 1 i I I 1 ~ 1 1 1 1 1 I I I I I I r r ~ I I r I I r I ~ I I ! I r I j I I I ! ! I 1nn ~ c ~ d - - ~ 1 [, 1 1 1 1,,,, ! i,, .~ ~, ~ ~ ' 450 550 650 450 550 650 Wavelength nm Wavelength nm FIGURE 4 Absorbance and difference spectra from short-wave cones of primates. The top panel of each pair plots the mean absorbance spectra before and after bleaching, and the bottom panel plots the difference spectrum derived from the bleaching. The prebleach spectra and difference spectra have been normalized to 100 percent. In the upper left are the mean spectra from SL~ cones from Cercopithecus cephus (moustached guenon), with a 431-nm template superimposed. In the upper right are the mean spectra from four cones from human eyes, with a 420-nm template. In the lower left are the mean spectra from 14 cones from Callahr~r jacchus (common marmoset), with a 424-nm template. In the lower right is a comparison of the mean spectra from Old World primates (longest-wave), Callithnx (middle-wave), and humans (shortest-wave).
CONE VISUAL PIGMENTS IN MONKEYS AND HUMANS 29 are clear. Again, this is reminiscent of the Am spacing of the longer- wave pigments (Figure 3) and may imply that there are constraints on the structure of visual pigments that determine their spectral sensitivities. Clearly, to elucidate such problems, more information is needed on the amino acid sequences of a number of primate opsins as well as a further understanding of the tertiary arrangement of opsin and retinal within the disc membranes of the cone outer segments. ACKNOWLEDGMENT lithe microspectrophotomet~y summarized here was conducted with support by the Medical Research Council of the United Kingdom grant G8512759N. REFERENCES Barlow, H.B. 1982 What causes trichromacy? A theoretical analysis using combfiltered spectra. Vision Research 22:635 643. Baylor, D.A., B.J. Nunn, and J.L. Schnapf 1987 Spectral sensitivity of cones of the monkey Macaca fascicularis. Joumal of Physiology 390:145-160. Bowmaker, J.K., H. J.N Dartnall, J.N. Lythgoe, and J.D. Mollon 1978 The visual pigments of rods and cones in the rhesus monkey, Macaca mulatta. Joumal of Physiology 274:329-348. Bowmaker, J.K., J.D. Mollon, and G.H. Jacobs 1983 Microspectrophotometric results for Old and New World primates. In Colour Vision: Physiology and Psychophysics, J.D. Mollon and Lye Sharpe, eds. London: Academic Press. Bowmaker, J.K., G. H. Jacobs, D .J. Spiegelhalter, and J. D. Mollon 1985 Two types of trichromatic squirrel monkey share a pigment in the red-green spectral range. Vision Research 25:1937-1946. Bowmaker, J.K., G.H. Jacobs, and J.D. Mollon 1987 Polymorphism of photopigments in the squirrel monkey: a sloth phenotype. Proceedings of the Royal Society of London B 231:38~390. Boynton, R.M. 1988 Colour vision. Annual Review of Psychology 39:69-100. Brown, P.K., and G. Wald 1964 Visual pigments of single rods and cones of the human retina. Science 144:4~51. Dartnall, H.J.N, and J.N. Lythgoe 1965 The spectral clustering of visual pigments. Vision Research 5:81-100. Dartnall, HJ.N, J.K Bowmaker, and J.D. Mollon 1983 Human visual pigments: microspectrophotometric results from the eyes of seven persons. Proceedings of the Royal Society of London B 220:115-130. Jacobs, G.H. 1987 Cone pigments and color polymorphism: a comparative perspective. In Frontiers of Usual Science: Proceedings of the 1985 Symposium. Committee
30 JAMES K BOWMAI~R on Vision, National Research Council. Washington, D.C.: National Academy Press. Jacobs, G.H., and J. Neitz 1985 Color vision in squirrel monkeys: sex-related differences suggest the mode of inheritance. Vision Research 25:141-143. Jacobs, G.H., J.K Bowmaker, and J.D. Mollon 1981 Behavioural and microspectrophotometric measurements of colour vision in monkeys. Nature 292 541-543. Jacobs, G.H., J. Neitz, and M. Crognale 1987 Color vision polymorphism and its photopigment basis in a callitrichid monkey (~Sant~inus fi`scicollis). Vision Research 27:2089~2100. Knowles, A., and H.J.A. Dartnall 1977 The photobiology of vision. In The Eye, vol. 2B, H. Davson, ed. New York: Academic Press. MacNichol, E.F. 1986 A unifying presentation of photopigment spectra. I^lsion Research 26:1543- 1556. MacNichol, E.F., J.S. Levine, R.J.W. Mansfield, L^E. Lipetz, and B.A. Collins 1983 Microspectrophotometry of visual pigments in primate photoreceptors. In Colour Vision: Physiology and Psychophysics, J.D. Mollon and L^T. Sharpe, eds. London: Academic Press. Mansfield, R.J.W. 1985 Primate photopigments and cone mechanisms. In The Usual System, A. Fein and J.S. Levine, eds. New York: Alan R. Liss. Mansfield, R.J.W., J.S. Levine, L.E. Lipetz, B.A. Collins, G. Raymond, and E.F. MacNichol 1984 Blue-sensitive cones in the primate retina: microspectrophotomet~y of the visual pigment. Experunental Brain Research 56:389-394. Marks, W.B., W.H. Dobelle, and E.F. MacNichol 1964 Visual pigments of single primate cones. Science 143:1181-1183. Mollon, J.D., J.K. Bowmaker, H.J.A. Dartnall, and A.C. Bird 1984a Microspectrophotometric and psychophysical results for the same deutera- nopic observer. In Colour vision deficiencies VII. G. Vemest, ed. The Hague: W. Junk. Mollon, J.D., J.K. Bowmaker, and G.H. Jacobs 1984b Variations in colour vision in a New World primate can be explained by polymorphism of retinal photopigments. Proceedings of the Royal Society of London B 322:373-399. Nathans, J., D. Thomas, and D.S. Hogness 1986 Molecular genetics of human color vision: the genes encoding blue, green and red pigments. Science 232:193 202. Pokorny, J., V.C. Smith, and I. Katz 1973 Derivation of the photopigment absorption spectra in anomalous trichromats. Joumal of the Optical Society of America 63:23~237. Schnapf, J.L^, T.~! Kraft, and D.N Baylor 1987 Spectral sensitivity of human cone photoreceptors. Nature 325:439~41. Travis, D.S., J.K. Bowmaker, and J.D. Mollon 1985 Polymorphism of retinal photopigments in the common marmoset (~Callithri~c jachus). Perception 14:A16. 1988 Polymorphism of visual pigments in a callitrichid monkey. Son Research 28:481090.
