Do blue-eyed white cats have normal or abnormal retinofugal pathways? R. W. Guillery, T. L. Hickey, and P. D. Spear

Do blue-eyed white cats have normal or abnormal retinofugal pathways? R. W. Guillery, T. L. Hickey, and P. D. Spear Three white cats that had blue eyes and no tapetum were studied by behavioral, electrophysiological, and anatomical methods in order to determine whether they showed evidence of abnormal retinofugal pathways comparable to those found in Siamese cats and in other mammalian forms having melanin deficits. The three cats were normal in evert/ respect. However, several other white cats, obtained subsequently, do show an abnormality of the retinogeniculate pathway identical to the abnormality of Siamese cats. Cats of the second type are thought to be homozygous for the Siamese gene and also to express the White gene. Because the characteristic Siamese pigmented "points" fail to develop in the presence of the white gene, cats of the second type are not distinguishable from other white cats on the basis of eye color or coat color. In terms of their central visual pathways and of their breeding patterns, however, they are recognizably Siamese. It is not known how common such "crypto-siamese" cats are in the white cat population, but the possibility of their occurrence suggests that, in general, white cats should not be used for studies of the central visual pathways. Key words: Siamese cats, pathway abnormalities, albinism, white cats, inherited visual abnormalities A misrouting of some optic nerve fibers occurs in the optic chiasm of individuals of many different mammalian species. An abnormally high proportion of decussating retinofugal fibers has been found in individuals having a significant reduction in the melanin of the retina, and this association has been found consistently, no matter what genetic mechanism is responsible for the melanin From the Department of Pharmacological and Physiological Sciences, University of Chicago, Chicago, 111. (R. W. G.); School of Optometry, the Medical Center, University of Alabama in Birmingham (T. L. H.); and Department of Psychology, the University of Wisconsin, Madison (P. D. S.) Supported by National Institutes of Health grants R01 EY02374, EY01338, EY01916, EY02545, P30EY 03039, and K04EY00089. Submitted for publication Aug. 27, 1980. Reprint requests: R. W. Guillery, Department of Pharmacological and Physiological Science, University of Chicago, 947 East 58th Street, Chicago, 111. 60637. deficit. 5 ' 13 * 17 The chiasmatic abnormality has been studied in particular detail in Siamese cats, which are homozygous for an allele of the albino series and which consequently have an abnormality of melanin formation. 18 In Siamese cats the abnormality has been related to an abnormal pattern of geniculate lamination and to a virtual absence of binocularly innervated neurons in the striate cortex. Within the visual cortex of Siamese cats one finds either an abnormal pattern of the visual field representation or a significant suppression of inputs from the ipsilateral eye. In association with the second type of change, it has been possible to demonstrate that many Siamese cats appear to be blind for the nasal hemifield. 3 ' 6-7 ' 8-10 - 19 ' 20 Blue-eyed white cats show little melanin in their coats or in their eyes, and some years ago, after occasional verbal reports about white cats that had possibly demonstrated one or another abnormality of the visual 0146-0404/81/070027+07$00.70/0 1981 Assoc. for Res. in Vis. and Ophthal., Inc. 27

28 Guillery et al. Invest. Ophthalmol. Vis. Sci. July 1981 pathways, we undertook a careful study of three white cats from this point of view. Our results showed that the white cats were normal in every respect. A similar result has been noted by Levick et al. 13a More recently we have obtained some white cats that are clearly abnormal. In order to clarify the extent to which one can expect to find abnormal visual pathways in white cats, we here report our observations on some of these white cats. Methods and results Initial observations of three white cats Appearance and behavior. These three cats were obtained from a breeder of white cats. Each cat had blue eyes and a pure white coat, corresponding in appearance to the description given by Searle 18 for the dominant white gene (W). One of the cats was an adult female and the other two were her young adult male offspring. One of the mother's parents was a blue-eyed white cat, and the other parent was a white cat with one yellow eye and one blue eye. The father of the two males was also a blue-eyed white cat. He had one blueeyed white parent, and one yellow-eyed white parent. The pedigree of all three experimental cats shows only white cats for two preceding generations and shows no Siamese cats for four preceding generations. Attempts to breed from these three animals were entirely unsuccessful. Ophthalmoscopic study of the three experimental cats showed a pink fundus with no evidence for pigment in the central parts of the retina and with no tapetal reflection. Below the level of the optic disc, where a normal cat shows clear evidence of pigmentation, these white cats showed no discernible pigment ophthalmoscopically. The visual fields of the three cats were tested behaviorally by methods described previously. 6 ' 21 After 25 trials at each of 12 visual field positions (15 intervals between positions), we obtained quantitative results that were indistinguishable from normal, and careful observation of the animals during the testing procedures at no point gave any indication of abnormalities comparable to those that have been described in Siamese cats by Elekessy et al. 3 or by Guillery and Casagrande. 6 Electrophysiological study of the visual cortex. Since only some Siamese cats show abnormalities of visually guided behavior, 6 ' 23 our next step was to use two of the animals for electrophysiological studies of single cells in the visual cortex by means of methods that have been described in detail by Kratz and Spear. 12 Three microelectrode penetrations were made in one of the cats (W-l) between stereotaxic AP coordinates +2 mm and 5.8 mm, and a single penetration was made at 7.5 mm in the other. All the penetrations entered striate cortex within 0.5 to 1.0 mm of the border between areas 17 and 18 (determined subsequently from Nissl-stained sections) and continued for 5 to 6 mm down the medial wall of the hemisphere. These experiments showed no abnormality and were clearly distinct from the results obtained in Siamese cats. The posterior penetrations first recorded cells having receptive fields within a degree of the area centralis 15 and then recorded cells with receptive fields progressing to 10 in the contralateral hemifield, roughly along the zero horizontal meridian. In the most anterior penetration, cells near the 17/18 border had receptivefieldson the vertical meridian, 7 into the lower hemifield. As the electrode was advanced, receptive field positions progressed 20 into the contralateral hemifield. Thus the cortical map was normal and not like the map seen in Boston Siamese cats. 8 ' 9 The relative input from the two eyes showed an essentially normal ocular dominance pattern. Comparison of the results with the results obtained from comparable cortical regions (i.e., regions with similar receptive field locations) of normal cats 1 ' 2l 1K 26 demonstrates that the ocular dominance distributions obtained from these two white cats were within the normal range. The majority of cells received some input from the ipsilateral eye (42 out of 59), and a high proportion of the cells were binocularly driven; neither finding is expected for Siamese cats for receptive field positions within about 20 of the vertical meridian, and we conclude that these white cats showed neither the Boston nor the Midwestern pattern of geniculo-cortical projections. 9 Retinogeniculate pathways. Each of the three white cats was given an injection of 500 /^Ci of 3 H-proline into one eye 1 day before death or preceding the electrophysiological experiments. Ketamine anesthesia was used. Subsequently the brains werefixedby an arterial perfusion with 10% formol saline under pentobarbital (Nembutal) anesthesia, and blocks containing the lateral geniculate nuclei were cut as frontal series and prepared for autoradiography. 16 The autoradiographs showed a normal retinogeniculate projection in all three cats. Lamina Al received an entirely uncrossed input and showed no interruptions. It was, as far as could be determined by inspection, normal in size. The reduced ipsilateral input and the disrupted lamina Al that characterize Siamese

Volume 21 Number 1, Part 1 Abnormal visual pathways in cats 29 cats 4 7> ' 10 ' 20 were not seen. The distribution of retinal afferents to the C laminae also appeared to be normal. Pigment distribution in the eyes. One eye from each cat was embedded in paraffin after the cornea and lens had been removed, and sagittal sections were cut at 14 /xm and stained with either cresyl violet or thionin. These sections confirmed the observation of Thibos et al. 25 that the tapetum is absent and that there is no melanin in the choroid of such white cats' eyes. However, in contrast to the observations of Thibos et al., 25 these three cats also showed significantly less pigment than normal in the retinal pigment epithelium. Comparison with normal cats and with Siamese cats showed that the pigment-free area, which in a normal cat corresponds roughly to the tapetal area, was somewhat greater in Siamese cats and in these white cats than in normal cats. In addition, the density of the pigment was roughly comparable in the white and the Siamese cats and is somewhat sparser than in the normal cats. It should be noted that the three white cats were pedigree cats selected and bred for their white coat color and for their delicate blue eyes. Thus they may represent a relatively extreme form of pigment reduction, possibly due to the action of several unidentified genes in addition to the dominant White gene. Other blue-eyed white cats we have studied have shown rather more pigment in the retinal pigment epithelium. We have found in general that the distribution of pigment in the retinal pigment epithelium varies considerably not only from one part of the retina to another but also between individuals with the same coat color genes. The density of the pigment increases gradually from the edge of the tapetum towards the periphery, but the size of the pigment-free "tapetal" area varies as does the intensity of pigmentation. A comparison between a normal cat and a Siamese cat is relatively straightforward at or close to the tapetal edge, since this can be used as a landmark. A comparison with an a-tapetal retina from a white cat is more difficult because there is no relevant landmark. In terms of pigment in the retinal pigment epithelium alone, it has not been possible always to distinguish clearly between Siamese cats and white cats or between normal cats and white cats. Observations demonstrating the occurrence of abnormal retinogeniculate pathways in some white cats. At this point of our studies we had convinced ourselves that the visual pathways of white cats are normal. We were next interested in studying the development of pigment in the white cats and obtained a white torn for breeding purposes.* This cat had a good record of successful matings, but initially he was unsuccessful with our white queens. In order to use the torn and to demonstrate his potency we mated him with some of our established Siamese queens. These matings, involving three Siamese cats, have so far produced five white and eight Siamese offspring. The latter, which have developed the characteristic Siamese "points", demonstrate that the white torn carries at least one Siamese gene, and raise the possibility that some or all of the white offspring might be homozygous for the Siamese gene. That is, the offspring can be regarded as having one dominant White gene, which prevents the expression of the characteristic Siamese pigmented points but which should not be expected to prevent the development of the equally characteristic but less obvious abnormality of the retinogeniculate pathways. Two of the white offspring have been used for studying the retino-geniculate pathways autoradiographically. t These two cats showed no evidence for a tapetal reflection and showed only a limited amount of pigment in the peripheral parts of the retinal pigment epithelium. When the cats were 3V2 months old, an injection of 500 fxci of 3 H-proline was made into one eye and the autoradiographic method 16 showed the retinogeniculate projection, which was abnormal in both cats, as illustrated in Fig. 1. The basic pattern of the retinogeniculate projection was that described for Siamese cats. 4 9 ' Much of lamina Al received a crossed input, and this segment of the lamina showed fusions with lamina A. A very small medial and a larger lateral segment of lamina Al received a normal uncrossed innervation. In addition there was a significant crossed input to lamina Cl. In comparison to the situation seen in most Siamese cats, the medial and lateral segments of lamina Al receiving an uncrossed input were relatively small. Thus it is possible that the abnormality in these white cats is slightly more severe than in Siamese cats and that one has to consider some *This was a cat that had been raised by Dr. W. L. Salinger with a binocular lid suture. He used it for breeding but stopped using it when he found electrophysiological evidence of an abnormal visualfieldmap in the lateral geniculate nucleus of a sibling. Before sending the cat to Chicago, Dr. Salinger checked one eye (under anesthesia) and noted a blue iris and pink fundus. twe thank Dr. Torrealba for making these injections and for providing us with Figs. 1 and 2.

30 Invest, Ophthalmot. Vis. Sci. July 1981 Guillery et al. A M I N A A Fig. 1. Pairs of brightfield {1 and 2) and darkfield {3 and 4) photomicrographs of two horizontal sections through the dorsal lateral geniculate nucleus of the white kitten described in the text. Note that the ipsilateral section (1 and 3) shows a small uncrossed input going to a small lateral segment of lamina Al (LAI), to two tiny medial patches of Al (MAI), and to a small marginal part of the medial interlaminar nucleus (MIN). The contralateral section (2 and 4) shows labeled axons that distribute to all other parts of the geniculate complex, including the geniculate wing (W), possible additive effect of the two genes. However, because the size of the abnormality was not obviously outside the range of variation seen in Siamese cats in general, one would need a rather large sample to determine whether the abnormality in these white cats is larger than the abnormality in pigmented Siamese cats. Two of the three other white cats produced by the above matings were used in infancy for different experiments, so that we did not study their re- tinofugal pathways. However, Nissl-stained sections from one of these kitten brains show an abnormally interrupted lamina Al that fuses extensively with lamina A in the medial parts of the nucleus. That is, this animal can reasonably be regarded as having the Siamese abnormality of the retinogeniculate pathway. No sections were available for study from the second kitten. The third white kitten was used for an electrophysiological study when it was approximately

Volume 21 Number 1, Part 1 Abnormal visual pathways in cats 31 3 months old. It showed reduced retinal pigmentation and a tapetal reflection. Two electrode penetrations were made in one hemisphere at the stereotaxic AP plane 6.0. The first penetration entered the cortex close to the 17/18 border, and the first recorded neurons had receptive fields about 2 ipsilateral to the zero vertical meridian close to the zero horizontal meridian. As the electrode was advanced through area 17 down the medial wall of the hemisphere, receptive field positions shifted by about 10 along the horizontal meridian and into the contralateral field. The second penetration, which was medial to the first, gave receptive fields in comparable positions. A total of 41 cells were studied in this hemisphere. Of these, five were nonresponsive and the other 36 were all driven through the contralateral eye only. No influence from the ipsilateral eye could be detected. In order to confirm that this eye was not blind, recordings were obtained from the other hemisphere. A total of nine cells were studied, and in this hemisphere, too, only the contralateral eye had access to the cortical neurons. Each of the nine cells was driven through the contralateral eye, and again the ipsilateral eye had no influence on cortical activity. On the basis of the receptive field positions and the entirely monocular input to the cortical cells, we concluded that this cat showed the Siamese abnormality (of the Midwestern pattern 9 ). Nissl-stained sections through the lateral geniculate nucleus have demonstrated the characteristically Siamese abnormal laminar pattern. Because four of the five white kittens produced by the white torn showed some evidence of the Siamese abnormality, the white torn is probably homozygous for the Siamese gene. This conclusion is reinforced by the fact that not one of the 13 kittens was normally pigmented even though eight were not white. Had the male been heterozygous for the Siamese gene, one would have expected about half of the nonwhite kittens to have been normally pigmented and half to have been Siamese. We therefore checked with Dr. Salinger regarding the white male's parentage. It appears that the mother was a white cat and the father was a grey and white cat. Neither parent was identified as Siamese, but one of the white tom's littermates that had been studied electrophysiologically (see footnote, p. 29) had shown some evidence of a pathway abnormality. Dr. Salinger has recently sent us this brain, and frontal sections stained with cresyl violet have clearly demonstrated the abnormal geniculate lamination expected for Siamese cats. A second series of matings has been undertaken between a Siamese male and a white female. One of the white female offspring of this mating was back-crossed with the Siamese male and produced a litter of three white, one Siamese, and one grey kitten. One of the white animals is still alive and two were used for different experiments. Nisslstained sections are available from one of these animals and these sections show the abnormal geniculate lamination characteristic of Siamese cats. Discussion The main conclusion one can draw from these observations is that even though, as Thibos et al. 25 have already pointed out, the White gene is not itself associated with an abnormality of the central visual pathways, blue-eyed white cats can have abnormal central visual pathways. The frequency with which one can expect to encounter such abnormal blue-eyed white, "crypto-siamese" cats is not known. Our experience suggests that they may not be uncommon; the Siamese gene may at some point have been used to enhance the blue eyes that are such an attractive feature of these white cats, 18 and further, since there is a certain cachet to owning a Siamese rather than an ordinary cat, the Siamese gene may be fairly common in the general cat population. A second point arising from our study concerns the relationship between the amount of melanin in the retinal pigment epithelium and the occurrence of the pathway abnormality. It has been shown, for many different genotypes in several different species, that there is an association between these' two features. Animals with abnormally low amounts of pigment in the pigment epithelium generally show the pathway abnormality. 13 ' 14> 17> 27 In contrast to this, in our comparisons of normal, Siamese, and blue-eyed white cats we have not been able to demonstrate clearly that the cats with abnormal pathways always have less melanin in the pigment epithelium than the normal cats. This could prove to be an important clue for separating the mechanisms that produce retinal melanin from those that produce the pathway abnormality. However, the abnormal pathways are established prenatally in

32 Guillery et al. Invest. Ophthalmol. Vis. Sci. July 1981 cats (Guillery, unpublished observations), and in Siamese cats there is a considerable and continuous postnatal increase of pigment. Preliminary observations suggest that this newborn Siamese cats have very little retinal melanin and that this increases during the first weeks of life whereas white cats that are not crypto-siamese are born with significantly more retinal melanin. We have no evidence about newborn crypto-siamese cats but would expect their eyes to be like those of Siamese cats. The details of early pigment development in white, Siamese, and normal cats merit further study, but the evidence available to date does not provide an argument against a close association between the mechanisms producing retinal melanin on the one hand and the abnormal chiasmatic pathway on the other. In the above discussion we have been concerned with the temporal relationships of pigment production and pathway formation. The spatial relationships also merit consideration, as has been pointed out by Thibos et al. 25 There is no correspondence in any species between the retinal distribution of melanin and the retinal distribution of abnormally connected ganglion cells. It would be a mistake to look for an interaction between pigment epithelial cells and retinal ganglion cells occurring evenly over the surface of the retina. The absence of pigment from the tapetal zone, stressed by Thibos et al., 25 reinforces the conclusion to be drawn from a study of the limited retinal distribution of the abnormally connected ganglion cells. Interactions that are not place-specific for each part of the retina must be sought. A small tongue of pigment epithelium that extends into the distal part of the eye stalk close to the developing retinofugal axons may be more relevant to the production of the pathway abnormality than is the melanin in the main part of the retina. This possibility has been discussed previously by one of us 24 and has recently been raised again by Silver and Sapiro, 22 who suggest that the pigment may act as a mechanical barrier that deviates outgrowing axons. We thank Jenny Hunter-Dobson for help with the histological procedures, Steven Price for help with the photography, and Debbie Teets for the typing of the manuscript. REFERENCES 1. Albus K: Predominance of monocularly driven cells in the projection area of the central visual field in cat's striate cortex. Brain Res 89:341, 1975. 2. Berman N, Murphy EH, and Salinger WL: Monocular paralysis in the adult cat does not change cortical ocular dominance. Brain Res 164:290, 1978. 3. Elekessy El, Campion JE, and Henry GH: Differences between visual fields of Siamese and common cats. Vision Res 13:2533, 1973. 4. Guillery RW: An abnormal retino-geniculate projection in Siamese cats. Brain Res 14:739, 1969. 5. Guillery RW: Visual pathways in albinos. Sci Am 230:44, 1974. 6. Guillery RW and Casagrande VA: Studies of the modifiability of the visual pathways in Midwestern Siamese cats. J Comp Neurol 174:15, 1977. 7. Guillery RW and Kaas JH: A study of normal and congenitally abnormal retino-geniculate projections in cats. J Comp Neurol 143:73, 1971. 8. Hubel DH and Wiesel TN: Aberrant visual projections in the Siamese cat. J Physiol 218:33, 1971. 9. Kaas JH and Guillery RW: The transfer of abnormal visual field representations from the dorsal lateral geniculate nucleus to the visual cortex in Siamese cats. Brain Res 59:61, 1973. 10. Kalil RE, Jhaveri SR, and Richards W: Anomalous retinal pathways in the Siamese cat: an inadequate substrate for normal binocular vision. Science 174: 302, 1971. 11. Kalil RE, Spear PD, and Langsetmo A: Response properties of striate cortex neurons in- cats raised with divergent or convergent strabismus. INVEST OPHTHALMOL VIS SCI 17(ARVO Suppl.):269, 1978. 12. Kratz KE and Spear PD: Effects of visual deprivation and alterations in binocular competition on responses of striate cortex neurons in the cat. J Comp Neurol 170:141, 1976. 13. LaVail JH, Nixon RA, and Sidman RL: Genetic control of retinal ganglion cell projections. J Comp Neurol 182:399, 1978. 13a. Levick WR, Thibos LN, and Morstyn R: Retinal ganglion cells and optic decussation of white cats. Vision Res 20:1001, 1980. 14. Lund RD, Lund JS, and Wise RP: The organization of the retinal projection to the dorsal lateral geniculate nucleus in pigmented and albino rats. J Comp Neurol 158:383, 1974. 15. Nikara T, Bishop PO, and Pettigrew JD: Analysis of retinal correspondence by studying receptive fields of binocular single units of cat striate cortex. Exp Brain Res 6:353, 1968. 16. Rogers AW: Techniques of Autoradiography. Amsterdam, 1973, Elsevier-NDU.

Volume 21 Number 1, Part 1 Abnormal visual pathways in cats 33 17. Sanderson KJ, Guillery RW, and Shackelford RM: Congenitally abnormal visual pathways in mink (Mustela vison) with reduced retinal pigment. J Comp Neurol 154:225, 1974. 18. Searle AG: Comparative Genetics of Coat Colour in Mammals. London, 1968, Logos Press. 19. Shatz C: Abnormal interhemispheric connections in the visual system of Boston Siamese cats: a physiological study. J Comp Neurol 171:205, 1977. 20. Shatz C: A comparison of visual pathways in Boston and Midwestern Siamese cats. J Comp Neurol 171:229, 1977. 21. Sherman SM: Visual field defects in monocularly and binocularly deprived cats. Brain Res 49:25, 1973. 22. Silver J and Sapiro J: The role of pigmented epithelia during morphogenesis of the optic nerve. INVEST OPHTHALMOL VIS SCI 19(ARVO Suppl.):3, 1980. 23. Simoni A and Sprague JM: Perimetric analysis of binocular and monocular visual fields in Siamese cats. Brain Res 111:189, 1976. 24. Stryker MP: Abnormal neural development. In Function and Formation of Neural Systems, Stent GS, editor. Berlin, 1977, Dahlem Konferenzen. 25. Thibos LN, Levick WR, and Morstyn R: Ocular pigmentation in white and Siamese cats. INVEST OPHTHALMOL VIS SCI 19:475, 1980. 26. Wilson JR and Sherman SM: Receptive-field characteristics of neurons in cat striate cortex: changes with visual field eccentricity. J Neurophysiol 49:512, 1976. 27. Wise RP and Lund RD: Retina and central projections of heterochromic rats. Exp Neurol 51:68, 1976.