<<

FULL PAPER Anatomy

Relationship between Distribution of Tapetum Fibrosum and Retinal Pigment Epithelium in the

Aya SHINOZAKI1,2), Yoshinao HOSAKA2), Tomohiro IMAGAWA2) and Masato UEHARA2)*

1)The United Graduate School of Veterinary Science, Yamaguchi University, 1677–1 Yoshida, Yamaguchi 753–8515 and 2)Department of Veterinary Medicine, Tottori University, 4–101 Koyama Minami, Tottori 680–8553, Japan

(Received 14 September 2009/Accepted 10 October 2009/Published online in J-STAGE 27 November 2009)

ABSTRACT. The tapetum lucidum is a -reflective device improving visual sensitivity in mesopic vision. There have been few studies on tapetal distribution and its relationship with degree of pigmentation in the retinal pigment epithelium (RPE). In the present study, sheep’s were used for macroscopical observation of the tapetum, and then histological sections of the posterior eyecups were made to analyze the distribution of tapetal thickness and degree of pigmentation in the RPE. Macroscopically, with available light, the tapetum was located in the dorsal eye to the and showed an L-shape with the horizontally elongated nasal part and the dorsally expanded temporal part. In with a , the tapetal area expanded and showed a more triangular shape. The tapetum his- tologically consisted of layers of dense fibers and was thicker in the temporal part than in the nasal part. The maximum tapetal thickness was approximately 70 m. The histological tapetal area was similar to the tapetal shape with a flash light. The pigmentation of the RPE was divided into three types, nonpigmented, transitional, and pigmented areas. The nonpigmented area was similar to the tapetal shape with available light. It is suggested that approximately 55% of the histological tapetal area is covered with the nonpig- mented area and is functional under a natural light condition. The functional tapetal area was similar to the L-shaped high-density area of retinal ganglion cells. KEY WORDS: area centralis, pigmentation, , tapetum lucidum. J. Vet. Med. Sci. 72(2): 211–215, 2010

The tapetum lucidum is a light-reflective device found in ever, there have been few studies on the distribution of many and invertebrates. The tapetum is consid- tapetal thickness, degree of pigmentation in the RPE, and ered to improve mesopic vision under a low light condition the relationship with the . Furthermore, the shape of by reflecting light that was not absorbed and passed through the sheep’s tapetum is somewhat different in some studies the photoreceptors and providing the photoreceptors with a and controversial [1, 6, 7, 13]. Thus, the objectives of the second chance for absorbing the reflected light [13, 14]. present study were to investigate the histological distribu- Most of the mammalian tapetum is the choroidal tapetum tion of tapetum tissues and the degrees of pigmentation in located outside the retina and between the choriocapillary the RPE on the tapetum throughout the sheep’s eye fundus layer and the proper substance of the with many and to compare with distribution of retinal neurons. From melanocytes. The choroidal tapetum is divided into two the results, we are able to explain the difference in the types, the tapetum cellulosum and the tapetum fibrosum. sheep’s tapetal shape. To our knowledge, this is the first The tapetum cellulosum consists of layers of rectangular- report showing the topographic map of the tapetal thickness shaped tapetal cells containing numerous rodlets. The tape- and the degrees of pigmentation in the RPE throughout the tum fibrosum in sheep consists of closely and regularly eye fundus. arranged layers of collagen fibers, and the tapetum produces a structural metallic in the sheep’s eye fundus [13, 14, MATERIALS AND METHODS 16]. The retinal pigment epithelium (RPE) is the outermost Right eyes from three adult Suffolk sheep (Ovis aries) layer of the retina, and the epithelial cells generally have (females, 4 years old, 74–83 kg) were used for the present numerous melanin granules. However, the RPE overlying study. The sheep were obtained after other experiments the tapetum has no melanin granules in order to allow light unrelated to those described here and killed by exsan- to arrive at the tapetum and to be reflected back to photore- guination under deep anesthesia by xylazine. All experi- ceptors [3, 5, 13, 14, 16]. ments were conducted in compliance with the guidelines of There have been many morphological studies on the tape- the Animal Care and Use Committee of Tottori University tum cellulosum using electron microscopy and some studies and the AVMA Guidelines on Euthanasia. on the tapetum fibrosum and the RPE [1–3, 5, 17]. The Each enucleated right eye was perfused with 2.5% glut- tapetum is considered to have some effects on vision, how- araldehyde-2% paraformaldehyde/0.1 M phosphate buffer (pH 7.4) into the eyeball in order to avoid separation of the *CORRESPONDENCE TO: UEHARA, M., Department of Veterinary Medicine, Tottori University, 4–101 Koyama Minami, Tottori retina and choroid from the . After immersion in the 680–8553, Japan. same fixative for a few days, the eye was opened by an e-mail: [email protected] encircling cut, and the posterior eyecup was again fixed in 212 A. SHINOZAKI, Y. HOSAKA, T. IMAGAWA AND M. UEHARA

Fig. 1. Photomicrographs of the tapetum and retinal pigment epithelium (RPE) in HE-stained vertical sections of a sheep eye. The tape- tum shows even and eosinophilic stainability. The RPE is variable in pigmentation. A: Thick tapetal area covered with nonpigmented RPE. B: Tapetal area covered with the transitional area of the RPE. C: Thin tapetal area covered with pigmented RPE. D: Nontapetal area covered with heavily pigmented RPE. The arrows with both heads show tapetal thickness. RPE=retinal pigment epithelium, CCL=choriocapillary layer, TL=tapetum lucidum, PSC=proper substance of choroid, BV=blood vessel, Sc=sclera. Scale bar =25 m. the same fixative. After fixation, the tapetal shape was RESULTS observed macroscopically and photographed with available light and with a flash. The posterior eyecup was divided Macroscopical observation of the tapetum (Fig. 2): Mac- into some parts in order to be flattened. Celloidin as embed- roscopically, with available light, the tapetum was located in ding medium was used to avoid separation of the retina and the dorsal eye fundus to the optic disc and showed an L- choroid, to accurately measure the tapetal thickness, and to shape consisting of the horizontally elongated nasal part and confirm the location of a section. Two serial vertical sec- the dorsally expanded temporal part (Fig. 2A). In photo- tions were made at 40–50 m in thickness every 1 mm or graphs with a flash, the tapetal area expanded, paticulally in every 200 m in a part including the area centralis of the ret- the dorso-central to dorso-nasal edge, and showed a more ina. The sections were stained with hematoxylin-eosin (HE) triangular shape (Fig. 2B). or trichrome stain with hematoxylin, acid fuchsin, and Histology of the tapetum and the RPE (Figs. 1 and 3): aniline . The specimens were pasted with a film includ- The tapetum was present between the choriocapillary layer ing lines at intervals of 1 mm. and the pigmented proper substance of the choroid. The Tapetal thicknesses were measured at 40 objective with tapetum consisted of many layers of dense collagen fibers HE specimens every 1 mm using an eyepiece with a scale including a few fibroblasts and a few small blood vessels attachment. The outer border of the tapetum was unclear penetrating the tapetal layers. The tapetum showed even because melanocytes were scattered in the outer border. and eosinophilic stainability in HE specimens and was blue Thus, the tapetal thicknesses were defined as the distance in color by aniline blue in trichrome staining. The outer from the inner point contacting the choriocapillary layer to tapetal edge was irregular due to irregular appearance of the outer point contacting melanocytes appearing in the melanocytes in the tapetal layers. tapetal layers (Fig. 1). The RPE was a single layer overlying the choriocapillary Degree of pigmentation in the RPE was observed at 20 layer. The RPE contained melanin granules of various objective between 1-mm-interval lines on the same HE amounts and sizes. In the thick tapetal area, few melanin specimens as those used for measurement of tapetal thick- granules were found in the RPE (Fig. 1A). In the thinner ness. RPE sampling areas were divided into three types by tapetal area, melanin granules increased gradually with degree of pigmentation. Firstly, the area in which nonpig- mented cells made up more than half of the RPE was regarded as the nonpigmented area with high light transmis- sion properties (Fig. 1A). In this area, there was little pig- mentation, if any. Secondly, the area in which pigmented cells made up more than half of the RPE was regarded as the transitional area with low light transmission properties (Fig. 1B). Thirdly, the area composed of only pigmented cells was regarded as the pigmented area with light impassable properties (Figs. 1C and 1D). From these data, topographic maps of tapetal thickness and degree of pigmentation in the RPE were made. The values were expressed as means  SD. Fig. 2. Photographs of the sheep’s right eye fundus with avail- able light (A) and with a flash (B). The tapetal area shows an L- shape in A and a triangular shape in B. Right side is nasal. Scale bar =1 cm. TAPETUM FIBROSUM OF THE SHEEP 213

Distributions of tapetal thickness and nonpigmented area of the RPE (Fig. 4): The tapetal area histologically including tapetal layers (histological tapetal area) (Fig. 4A) was simi- lar to the macroscopic tapetal shape with a flash light (Fig. 2B) and showed a triangular shape. The tapetal thickness increased inward, and the increase was steeper in the nasal, ventral and temporal edges than in the dorso-central to dorso-nasal edges. The tapetum was thicker in the temporal part than in the nasal part. The maximum thickness was 70  6.1 m. The mean tapetal thickness was 34  3.7 m. The percentages of each thickness area showed a single-peaked pattern and dominant thickness of 31–50 m (Fig. 5). The transitional and pigmented areas of the RPE partially covered the peripheral part, especially the dorso-central and dorso-nasal parts of the histological tapetal area. Thus, the nonpigmented area of the RPE (Fig. 4B) was similar to the macroscopic tapetal shape with available light (Fig. 2A) and showed an L-shape. The percentage of the nonpigmented and transitional areas to the histological tapetal area was 77  2.8%, and that of the nonpigmented area was 55  3.4%. Tapetum and RPE in the area centralis (Figs. 3 and 4): The area centralis, the area with maximum density of retinal ganglion cells in the temporal retina to the optic disc, was found as a stratified structure of concentrated cells in the (Fig. 3A and 3B). The tapetal thickness in the area centralis was near or below the mean (Figs. 3 and 4A). The area centralis fell within the nonpigmented area of the RPE including some epithelial cells with a few large melanin granules (Figs. 3C and 4B).

DISCUSSION

The sheep’s tapetum fibrosum consisted of layers of dense collagen fibers including a few fibroblasts and a few small blood vessels penetrating the tapetal layers (Figs. 1 and 3) and was morphologically similar to previous reports [1, 2, 13, 16]. The sheep’s RPE was also morphologically similar to previous reports [3, 13]. In the present study, we defined the thickness of the tape- Fig. 3. Photomicrographs of the sheep’s retina, choroid and sclera tum as the distance from the inner point contacting the cho- in vertical sections of the area centralis by HE (A, C) or riocapillary layer to the outer point contacting melanocytes trichrome stain (B). The area centralis shows a stratified arrange- ment of cells in the ganglion cell layer and overlaps with the appearing in the tapetal layers. Thus, the thickness mea- tapetum (A, B). Some large melanin granules are found in the sured in this study means not thickness of entire tapetal lay- RPE of the area centralis (C). Re=retina, Ch=choroid, Sc=sclera, ers but functional thickness. In the topographic map of GCL=ganglion cell layer, INL=, ONL=outer tapetal thickness (Fig. 4A), the tapetum was thicker in the nuclear layer, RPE=retinal pigment epithelium, CCL=choriocap- temporal part, and the maximum thickness was 70  6.1 m. illary layer, TL=tapetum lucidum, PSC=proper substance of The mean tapetal thickness was 34 ± 3.7 m. The area choroid. Scale bars =50 m in A and B; 25 m in C. showing thickness of 31–50 m was dominant (Fig. 5). There are a few reports on the thickness of the tapetum decrease in tapetal thickness (Figs. 1B and 1C). The RPE fibrosum [1, 2], and it has been shown that the tapetal thick- without the tapetum was heavily pigmented and was brown ness is about 80 m [1] or 100–200 m [2] in sheep. These or black (Fig. 1D). Most of the melanin granules were values were obtained by electron microscopy on a small part small. By way of exception, epithelial cells containing a of the tapetum and probably mean the depth of entire tapetal few large melanin granules were often found between the layers. nonpigmented area and the transitional area, particulary in The sheep’s tapetum was macroscopically L-shape with the temporal periphery including the area centralis (Fig. available light (Fig. 2A). The same observation has been 3C). reported previously [1, 13]. With strong light such as a flash 214 A. SHINOZAKI, Y. HOSAKA, T. IMAGAWA AND M. UEHARA

Fig. 4. Topographic maps of tapetal thickness (A) and degree of pigmentation in the RPE (B) of a sheep right eye. dots show the location of the area centralis. Right side is nasal. Scale bar =1 cm.

tological tapetal area. This is a reason why the tapetal area changes macroscopically in shape depending on brightness. Pigmentation in the RPE controls the functional tapetal area. In the sheep’s retina, the area with a high density of reti- nal ganglion cells shows an L-shape consisting of horizontal and vertical visual streaks, and the maximum density in the area centralis is located in the temporal end of the horizontal visual streak [Shinozaki et al., unpublished]. The functional tapetal area was similar to the L-shaped distribution of reti- nal ganglion cells. An overlap of the tapetal area and the specialized distribution of retinal ganglion cells has been described in some species. The tapetal area seems to over- lap with the horizontal visual streak in sheep [1], , bovines, and [9], with the area centralis in [15], Fig. 5. Percentages of each tapetal thickness area. Values are the and with the L-shaped high-density area in [11]. The means of three sheep  SD. shape of the sheep’s functional tapetal area indicates that sheep achieve better mesopic vision in horizontal and antero-inferior visual fields. Sheep are basically diurnal but light, the tapetal area was expanded, particularly in the active from dawn to evening and graze even at nighttime dorso-central to dorso-nasal part. Thus, the tapetum showed [4]. The better mesopic vision in the horizontal visual field a more triangular shape (Fig. 2B). The same findings in would be effective for detecting . The thicker ophthalmoscopic studies have been reported [6, 7]. The tapetal area located in the temporal part would have higher shape with available light was similar to that of the nonpig- reflective power. Thus, the antero-inferior visual field cor- mented area of the RPE, and the shape with a flash light was responding to the thicker tapetal area would be more impor- similar to that of the histological tapetal area. Thus, under tant for sheep in a low light environment. The better natural light, only the tapetal area consisting of relatively mesopic vision in the antero-inferior field would be effec- thick tapetal layers and covered with the nonpigmented area tive for scanning the ground. of the RPE would effectively reflect light and be functional. There is concern that the tapetum and its overlap with the The functional tapetal area was about half (55%) of the his- retinal ganglion cell distribution may cause not only an TAPETUM FIBROSUM OF THE SHEEP 215 increase in visual sensitivity but also weakening of visual 2. Bortolami, R., Callegari, E. and Lucchi, M. L. 1979. Ultra- resolution [8, 9, 13], because the reflection by the tapetum structural observations on the tapetum lucidum fibrosum. Anat. may be too glaring in a bright environment and the tapetum Histol. Embryol. 8: 72–78. may produce scattered light. In the sheep’s retina, there are 3. Braekevelt, C. R. 1983. Fine structure of the choriocapillaris, three types of photoreceptors, i.e., rod, middle to long wave- Bruch’s membrane and retinal epithelium in the sheep. Anat. Embryol. 166: 415–425. length sensitive (M/L) cone, and short wavelength sensitive 4. Dwyer, C. M. 2008. Environment and the sheep. pp. 41–79. In: (S) cone. Rods contribute to the visual sensitivety in The Welfare of Sheep (Dwyer, C. M. ed.), Springer, Dordrecht. mesopic vision, M/L cones contribute to the visual resolu- 5. Feeney-Burns, L. and Mixon, R. N. 1979. Development of tion and color vision in photopic vision, and S cones con- amelanotic retinal pigment epithelium in eyes with a tapetum tribute to the color vision in photopic vision. Because the lucidum: melanosome autophagy and termination of melano- fresh tapetum is mainly bluish in color [1], the reflected genesis. Dev. Biol. 72: 73–88. light would be effectively absorbed by rods with a photopig- 6. Galán, A., Martín-Suárez, E. M., Granados, M. M., Gallardo, J. ment spectral peak of about 500 nm [10] and would not be M. and Molleda, J. M. 2006. Comparative fluorescein angiog- effectively absorbed by M/L cones with a photopigment raphy of the normal sheep and ocular fundi. Vet. Ophthal- spectral peak of 552 nm [12]. Furthermore, S cones have a mol. 9: 7–15. 7. Galán, A., Martín-Suárez, E. M. and Molleda, J. M. 2006. Oph- photopigment spectral peak of 446 nm [12] and are different thalmoscopic characteristics in sheep and goats: comparative in distribution from the tapetum [Shinozaki et al., unpub- study. J. Vet. Med. A Physiol. Pathol. Clin. Med. 53: 205–208. lished]. Thus, the tapetum may mainly contribute to visual 8. Hall, C. 2007. The impact of on equine learn- sensitivity and hardly interfere with visual resolution. ing. Behav. Processes 76: 29–33. In the present study, epithelial cells of the RPE containing 9. Hebel, R. 1976. Distribution of retinal ganglion cells in five a few large melanin granules were often found between the mammalian species (pig, sheep, ox, , ). Anat. nonpigmented area and the transitional area including the Embryol. 150: 45–51. area centralis (Fig. 3C). Similar large granules have been 10. Höglund, G., Nilsson, S. E. and Schwemer, J. 1982. Visual pig- reported in sheep [3] and in bovines as macromelanosomes ment and visual receptor cells in fetal and adult sheep. Invest. [5]. According to Feeney-Burns and Mixon [5], the epithe- Ophthalmol. Vis. Sci. 23: 409–418. 11. Hughes, A. and Whitteridge, D. 1973. The receptive fields and lial cells of the RPE in the future tapetal area are melanized topographical organization of goat retinal ganglion cells. along with those in the nontapetal area during early embryo- Vision Res. 13: 1101–1114. genesis, but melanogenesis in the future tapetal area gradu- 12. Jacobs, G. H., Deegan 2nd, J. F. and Neitz, J. 1998. Photopig- ally subsides. The macromelanosomes are a residual body ment basis for dichromatic color vision in cows, goats, and resulting from the prenatal melanolytic activity of the lysos- sheep. Vis. Neurosci. 15: 581–584. omal system in the RPE of the tapetal area, and they are 13. Ollivier, F. J., Samuelson, D. A., Brooks, D. E., Lewis, P. A., found even in bovines of 8–9 years in age. The presence of Kallberg, M. E. and Komáromy, A. M. 2004. Comparative macromelanosomes may produce some sort of effects on morphology of the tapetum lucidum (among selected species). light reflection by the tapetum. The relationships among the Vet. Ophthalmol. 7: 11–22. tapetum, RPE and retina in development must be deter- 14. Schwab, I. R., Yuen, C. K., Buyukmihci, N. C., Blankenship, T. N. and Fitzgerald, P. G. 2002. of the tapetum. mined in order to elucidate the significance of macromel- Trans. Am. Ophthalmol. Soc. 100: 187–200. anosomes in the area centralis and the difference between 15. Steinberg, R. H., Reid, M. and Lacy, P. L. 1973. The distribu- the histological tapetal area and the functional tapetal area. tion of rods and cones in the retina of the (Felis domesti- cus). J. Comp. Neurol. 148: 229–248. REFERENCES 16. Walls, G. L. 1967. The Eye and Its Adaptive Radia- tion, Hafner, New York. 1. Bellairs, R., Harkness, M. L. and Harkness, R. D. 1975. The 17. Young, N. M., Hope, G. M., Dawson, W. W. and Jenkins, R. L. structure of the tapetum of the eye of the sheep. Cell Tissue 1988. The tapetum fibrosum in the eyes of two small . Res. 157: 73–91. Mar. Mamm. Sci. 4: 281–290.