<%BANNER%>

Corneal Vascularization in the Florida Manatee (Trichechus manatus latirostris)


PAGE 1

CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE ( Trichechus manatus latirostris ) By JENNIFER YOUNG HARPER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Jennifer Young Harper

PAGE 3

ACKNOWLEDGMENTS I would like to thank my mentor, Dr. Don Samuelson. He has been a wonderful source of knowledge, inspiration, and motivation. Without his help and guidance, I could not have accomplished any of this work. I would also like to thank Dr. Roger Reep for all of his help and support along the way. He too has acted as a rock and support system. My additional committee members (Dr. Peter McGuire, Dr. Dennis Brooks, and Dr. Gordon Bauer) have been tremendous support and I thank them for all they have offered. Their guidance has been appreciated beyond belief. Laboratory technologists Pat Lewis and Maggie Stoll were extremely helpful and supportive during much of my work. I would have not been able to accomplish the first procedure without their help. I thank these fine ladies from the bottom of my heart. My parents, Jim and Marion Young, have meant more to me than I can ever describe or explain. I appreciate all their love and support. Finally, I thank my husband Ridge Harper. He has been the strongest support system I could ever ask for and makes me happier than I ever knew I could be. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT.........................................................................................................................x CHAPTER 1 LITERATURE REVIEW.............................................................................................1 The Florida Manatee.....................................................................................................1 The Tactile Sensory System of Marine Mammals.......................................................2 The Auditory System of Marine Mammals..................................................................5 The Olfaction and Gustation of Marine Mammals.......................................................7 The Vision of Marine Mammals...................................................................................8 The Cornea..................................................................................................................13 Angiogenesis...............................................................................................................17 Objectives...................................................................................................................22 2 CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE AND THREE-DIMENSIONAL RECONSTRUCTION.....................................................23 Introduction.................................................................................................................23 Materials and Methods...............................................................................................32 Specimen Collection............................................................................................32 Results.........................................................................................................................36 Anatomical Description.......................................................................................36 Morphometry.......................................................................................................39 Statistics...............................................................................................................40 Discussion...................................................................................................................42 3 VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR-1 AND -2 IN THE FLORIDA MANATEE CORNEA....................................................................96 Introduction.................................................................................................................96 Materials and Methods.............................................................................................106 iv

PAGE 5

Specimen Collection..........................................................................................106 Preparation of Eyes............................................................................................106 Histology...........................................................................................................107 Immunohistochemistry......................................................................................107 Results.......................................................................................................................108 Vascular endothelial growth factor receptor-1 (VEGFR-1)..............................108 Vascular endothelial growth factor receptor-2 (VEGFR-2)..............................109 4 ADDITIONAL STUDIES........................................................................................127 Corneal Vascularization in the Antillean Manatee...................................................127 Ophthalmic Examinations in Two Captive Florida Manatees..................................128 5 GENERAL CONCLUSIONS...................................................................................139 LIST OF REFERENCES.................................................................................................142 BIOGRAPHICAL SKETCH ...........................................................................................15 4 v

PAGE 6

LIST OF TABLES Table page 2-1 Information on manatee tissue used and volume of vasculature at 25 X.................41 2-2 Volume of vasculature at 200 X...............................................................................42 vi

PAGE 7

LIST OF FIGURES Figure page 2-1 Florida manatee eye with ocular tissue and fat removed revealing the eye and optic nerve..............................................................................................................55 2-2 Histological pictures of MNW-9111.....................................................................56 2-3 These pictures were taken of MSW-9114 of the same blood vessel at varying magnifications to show the penetration of this vessel into the anterior epithelium..............................................................................................................57 2-4 UCF-9114 shows the smaller vessels that were outlined within the stroma for three-dimensional reconstruction (200X)..............................................................58 2-5 Pictures from MNW-9016.....................................................................................59 2-6 These pictures were taken from the unknown specimen. (A) Shows the edge of the limbal vascular arrangement at the peripheral cornea (200X), and (B) larger vessels that go right up to the anterior epithelium at 200X...................................60 2-7 Anatomical orientation of manatee eyes as they are described in three-dimensional reconstruction....................................................................................61 2-8 MNW-9016............................................................................................................62 2-9 MNW-9022(R).......................................................................................................63 2-10 MNW-9111............................................................................................................64 2-11 MNW-9116............................................................................................................65 2-12 MSE-9101..............................................................................................................66 2-13 MSE-9105..............................................................................................................67 2-14 MSE-9107..............................................................................................................68 2-15 MSE-9115..............................................................................................................69 2-16 MSE-9131..............................................................................................................70 vii

PAGE 8

2-17 MSW-9114.............................................................................................................71 2-18 MSW-9131(R).......................................................................................................72 2-19 MSW-9137(L)........................................................................................................73 2-20 MSW-9137(R).......................................................................................................74 2-21 MSW-9141.............................................................................................................75 2-22 MSW-9143.............................................................................................................76 2-23 TM-8619(L)...........................................................................................................77 2-24 TM-8619(R)...........................................................................................................78 2-25 TM-8630(L)...........................................................................................................79 2-26 TM-8630(R)...........................................................................................................80 2-27 UCF-9114..............................................................................................................81 2-28 UCF-9137..............................................................................................................82 2-29 UCF-9138..............................................................................................................83 2-30 UCF-9141..............................................................................................................84 2-31 Unknown................................................................................................................85 2-32 Fetus(L)..................................................................................................................86 2-33 Fetus(R)..................................................................................................................87 2-34 Dorsal quadrant of MNW-9016 at 200X...............................................................88 2-35 Medial quadrant of MNW-9016 at 200X...............................................................89 2-36 Dorsal quadrant of MSW-9114 at 200X................................................................90 2-37 Ventral quadrant of MSW-9114 at 200X...............................................................91 2-38 Dorsal quadrant of TM-8619(L) at 200X..............................................................92 2-39 Medial quadrant of TM-8619(L) at 200X..............................................................93 2-40 Ventral quadrant of TM-8619(L) at 200X.............................................................94 2-41 Lateral quadrant of TM-8619(L) at 200X..............................................................95 viii

PAGE 9

3-1 Florida manatee eye.............................................................................................118 3-2 Pictures taken from TM-0310 showing a positive reaction of VEGFR-1...........119 3-3 VEGFR-1 positive reactions in TM-0310 showing the prevalence of the positive reaction at 200X with 1:50.....................................................................120 3-4 VEGFR-1 positive reactions in TM-0310 (A) at 400X with 1:100.....................121 3-5 VEGFR-1 reactions in TM-0310 of control samples...........................................122 3-6 Pictures taken from TM-0311 showing a positive reaction of VEFGR-1...........123 3-7 VEGFR-1 positive reactions in TM-0311............................................................124 3-8 VEGFR-1 reactions in TM-0311 of control samples...........................................125 3-9 Pictures taken of TM-0310 showing no positive reactions of VEGFR-2 that appears to be different from the control samples.................................................126 4-1 The cornea from an Antillean manatee eye.........................................................130 4-2 This corneal section was cut at 10 microns and stained using Massons Trichrome stain....................................................................................................131 4-3 These pictures show the unusually nonorganized, wavy stroma observed in the Antillean manatee cornea...............................................................................132 4-4 Extensive vasculature was observed in the (A) stroma of the Antillean manatee cornea at 100X and (B) as the stroma approaches the anterior epithelium at 100X.....................................................................................................................133 4-5 These pictures illustrate the vessels in the anterior epithelium............................134 4-6 Vasculature can be observed (A) beneath the anterior epithelium at 200X and (B) penetrating the anterior epithelium at 400X..................................................135 4-7 The unusual material thought to be mucus surrounding the anterior epithelium............................................................................................................136 4-8 Unusual projections were observed in the posterior stroma and adjacent endothelium..........................................................................................................137 4-9 A PAS stain of the Antillean manatee cornea showing no reaction to a fungus in the anterior epithelium at (A) 100X and (B) 200X..........................................138 ix

PAGE 10

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE (Trichechus manatus latirostris) By Jennifer Young Harper May 2004 Chair: Don A. Samuelson Cochair: Roger L. Reep Major Department: Veterinary Medical Sciences The cornea of the Florida manatee is anatomically unique because blood vessels are found throughout the cornea. In all other species of animals, this is considered a pathological condition that impedes vision and is usually caused by injury or trauma of some sort. The purpose of this study was to describe more clearly corneal vascularization by (1) examining the architecture through three-dimensional reconstruction in order to find possible patterns in size, distribution, and location of blood vessels that may be related to age, gender, location found, and environment, and (2) examining vascular endothelial growth factor receptors-1 and -2 in the cornea. Twenty-six eyes from 22 individuals were prepared for histological examination and subsequent three-dimensional reconstruction. Every specimen examined possessed blood vessels in the cornea, mostly comprising 0.3% of total surface density (volume) of the cornea. However, two individuals had a greater density of vascularization (1.2% and 0.7%). The size and x

PAGE 11

distribution of vessels appeared to be greatest toward the mid-periphery of the cornea. No differences were found among individuals based on age, gender, and site of recovery. Environmental influences were not a significant factor either, which was not originally anticipated. The presence of vessels at the level of the anterior epithelium was surprising, and it appeared that the vascularization was directed more anteriorly than originally thought. Blood vessels were also found in fetal eyes. In all the eyes examined, no signs of injury or trauma could be observed. Vascular endothelial growth factor receptor-1 (VEGFR-1) was found in the corneas of 2 individuals examined through immunohistochemistry. However, vascular endothelial growth factor receptor-2 (VEGFR-2) was not, indicating that a noninflammatory and nonpathological process of angiogenesis might be taking place in the Florida manatee cornea. The presence of blood vessels appears to impair vision minimally based on the blood vessels low density, size, and location. The penetration of vessels into the anterior epithelium and development of vessels within the fetus points to an evolutionary adaptation, possibly due to the manatees unique ability to move freely in salt, brackish, and fresh water. xi

PAGE 12

CHAPTER 1 LITERATURE REVIEW The Florida Manatee The Florida manatee, Trichechus manatus latirostris, is perhaps one of the most interesting marine mammals. It is a subspecies of the West Indian manatee and is in the order Sirenia, which includes two additional species, the Amazonian manatee and the West African manatee. The order Sirenia is made up of two families, Trichechidae (manatees) and Dugongidae (dugongs). Manatees are the only marine mammal herbivores (Beck and Barros 1991; Reynolds and Odell 1991) and the Florida manatee feeds on over 60 different species of aquatic vegetation (Provancha and Hall 1991). Trichechids are also unique among marine mammals in that they are able to live in different types of aquatic habitats including fresh water, salt water and brackish water (Hartman 1979). During warmer months of the year, manatees can be found in subtropical and tropical coastal waters of the Southeastern portion of the United States. However, during winter months, manatees are limited to the warmer waters of Florida because they have relatively low metabolic rates that prevent them from producing heat efficiently in cooler water (Gallivan et al. 1983; Irvine 1983). Therefore, Florida manatees are particularly vulnerable to water temperatures that fall below 20C (Bossart et al. 2002; Bossart 2001). Florida manatees are listed as endangered species in the United States (United States Marine Mammal Commission 2001) and their survival is jeopardized by many environmental and human factors (United States Fish and Wildlife Service 2001). 1

PAGE 13

2 Environmental influences include infection caused by epizootic outbreaks (Bossart et al. 1998) and exchange of morbillivirus between individuals (Duignan et al. 1995). However, human influences are more often the cause of death or injury to manatees, including the ingestion of debris such as monofilament fishing line (Beck and Barros 1991), oil spills and habitat alteration (Lefebvre and OShea 1996). Perhaps the most direct human impact on manatees survival is collision with watercraft. Mortality rates from 1974 to 1996 indicate that 23.4% of manatee deaths were caused by watercraft (Reynolds 1999). Due to the endangered species status of the manatee, the United States Fish and Wildlife Service (2001) designed a recovery plan for the Florida manatee. This recovery plan outlines actions that need to be taken in order to reduce threats to Florida manatees. Included are four primary objectives: (1) minimize causes of manatee disturbance, harassment, injury and mortality; (2) determine and monitor the status of manatee populations; (3) protect, identify, evaluate, and monitor manatee habitats; and (4) facilitate manatee recovery through public awareness and education. In efforts to reduce the threats to manatees using the recovery plan, the scientific community has taken quite an interest in learning as much as possible about these unique gentle giants. Establishing a better understanding of the five sensory systems, touch (tactile), hearing (auditory), smell (olfaction), taste (gustation), and vision, is a step in the right direction in order to discover methods to conserve the species, learn about their biology, and how they interact in their environment. The Tactile Sensory System of Marine Mammals Approximately 120 million years ago mammals began to develop the one of their sensory systems, that of touch through the use of whiskers or vibrissae. These tactile

PAGE 14

3 structures have evolved to increase the search area for an animal (Brecht et al. 1997). Vibrissae can be found on many areas of animals, but most commonly the feet or pads and face. The vibrissal follicle-sinus complexes of the mystacial vibrissae, located on the face, are very well innervated (Ebara et al. 2002) and are used during active touch to discriminate objects (Bachteler and Dehnhardt 1999). It has been shown that rats use their mystacial vibrissae exclusively to discriminate textured surfaces. Additionally, walruses and California sea lions use their mystacial vibrissae to distinguish different shapes (Bachteler and Dehnhardt 1999). Mammals living in aquatic environments often need to interpret information about their surrounding conditions when visibility is very poor if not impossible. Therefore, several species of mammals living in these conditions have developed the sensory abilities to compensate for visual losses (Dehnhardt et al. 2001). Most marine mammals have some mechanism for tactile sensation in which they interpret information about their environment, and for these aquatic animals, vibrissae are one of the ways to do so. In marine mammals, vibrissae are most commonly associated with pinnipeds, being what appear to be the whiskers of the animal. The vibrissae of pinnipeds are innervated peripherally and are found in large sections of the animals bodies to maximize the amount of information gained about the surrounding environment. Vibrissae are not limited to pinnipeds, however, as all marine mammals have some form of vibrissae varying in tactile sensitivity (Wartzok and Ketten 1999). The Florida manatee has thick, dense lips with attached rigid vibrissae that create a unique anatomy of the face (Marshall et al. 1998a). These vibrissae flare out and are used to grab vegetation and to manipulate objects in their environment (Reep et al. 1998). Additional vibrissae are found in the oral

PAGE 15

4 cavity, being smaller and less noticeable, and still concerned with feeding behaviors (Marshall et al. 1998b). Florida manatees also have facial hairs, as well as stiff hairs found on the oral disk (Reep et al. 1998). The perioral bristles found on the manatees muscular snouts, which are used in a prehensile manner to forage on aquatic vegetation, are distinctive to Sirenians (Marshall et al. 2000). Reep et al. (2001) examined the follicles of the face and found that they showed characteristics of vibrissae that included a well-defined blood sinus, a connective tissue capsule, and dense innervation. This finding suggests that the anatomy of these follicles is comparable to that of many other mammals that possess vibrissae used for tactile exploration, such as the closest living land relative of the manatee, the elephant. The tip of the trunk of elephants is referred to as the finger and is extremely mobile in that it can be used for a wide range of behaviors such as grasping food, tactile sensitivity, and chemosensory recognition. The skin of the trunk tip has a high density of nerve endings, several convoluted branched corpuscles, and vibrissae that are densely innervated. The unique innervation found in the trunk tip of Asian elephants is similar to that of the mystacial vibrissae found in rodents and the lip tissue found in some monkey species (Rasmussen and Munger 1996). A study conducted by Bachteler and Dehnhardt (1999) investigating the tactile hairs on the face of the Antillean manatee found that it had a tactile sensitivity that compared closely to the trunk tip of Asian elephants, but was less sensitive than that of the vibrissae of pinnipeds. More recent studies conducted by Reep et al. (2002) suggest that in addition to sensory vibrissae located on the facial region, Florida manatees also possess sensory hairs located on the body that may be analogous to the lateral line found on fish. The follicles

PAGE 16

5 of these sensory hairs were examined histologically and also showed characteristics of follicle-sinus complexes. It was found that the hairs varied in follicle size and there appeared to be no major differences in follicle geometry. The follicles found on the body differed from those on the face in that body follicles lacked a ring sinus, were less innervated, and the capsules were thinner in body follicles than those on the face. Reep and his colleagues suggested that these hairs are actually vibrissae used to detect water displacements connected to environment stimuli, such as other manatees, currents and water flows. The hairs found on the body of the manatee might be similar to the whiskers of harbor seals (Phoca vitulina), which also act as the fish lateral line by detecting water movements and other information about the surrounding environment (Dehnhardt et al. 2001). If these body hairs of the manatee are indeed tactile hairs, then this information combined with the studies of the facial vibrissae, indicates that manatees most developed sensory system would be the tactile sensory system (Reep et al. 2002). The Auditory System of Marine Mammals Hearing in mammals is a remarkable process that involves several very small bones in the ear. The malleus, incus, and stapes in the middle ear transmit sound to the inner ear from the tympanic membrane. In land mammals, vibrations occur in the tympanic membrane where sound is received and then the sound is passed on to the malleus. The malleus and incus then rotate together producing a back and forth movement at the vestibular window where sound pressure is amplified (Kinkel et al. 2001; Mallo et al. 2000). This process that takes place in the middle ear of land mammals allows them to efficiently receive airborne sounds. However, for marine mammals, this process is not efficient because once sound has passed through water into the air-filled middle ear, it has been obstructed. Most marine mammals have developed unique middle ears to

PAGE 17

6 compensate for this loss and are well adapted to hearing sounds in their underwater world (Ketten 1991). The ossicles in the middle ears of cetaceans have evolved to transmit vibrations to the oval window where the velocity is amplified and not the pressure, as in land mammals. The ossicles of cetaceans are also inflated and much denser than that of any other mammals (Ridgway et al. 2001; Nummela et al. 1999). The Florida manatee also adapted to hearing underwater. The ears of the manatee are located just behind the eyes and have very small openings. Like pinnipeds, the ear bones of manatees are in direct contact with the cranium (Wartzok and Ketten 1999). Because of this anatomical positioning, it has been suggested that sound passes through the lower jaw and is carried to the ear bones (Reynolds and Odell 1991). Experiments conducted by Gerstein et al. (1999) tested the auditory abilities of two captive manatees to measure the underwater hearing sensitivity with regard to frequency and hearing threshold. When measuring hearing sensitivity, an audiogram is often performed. An audiogram is also referred to as a hearing curve and graphs the range of frequencies that can be heard. Additionally, an audiogram measures an individuals sensitivity within its hearing range. Most mammals have a U-shaped plot on a hearing curve because the intensity of a signal at its lowest detection threshold is measured. The highest threshold areas indicate the lowest sensitivity and are found on either side of the frequency range (Gerstein 2002). The experimenters discovered that manatees displayed an inverted U-shaped hearing range from 500 Hz to 38 kHz. This information suggested that manatees hear low frequency sounds below 16 kHz (i.e., the frequency at which they can discern the lowest amplitude or quietest sound) as well as other marine mammals (Gerstein et al. 1999). Audiograms conducted on other marine mammals have produced U-shaped

PAGE 18

7 hearing ranges also. The arctic white whale, Delphinapterus leucas, has a hearing range of 40 Hz to 150 kHz (Ridgway et al. 2001); the pacific walrus, Odobenus rosmarus divergens, has a range of 1 to 12 kHz (Kastelein et al. 2002a); the harbor porpoise, Phocoena phocoena, has a range between 250 Hz and 180 kHz (Kastelein et al. 2002b); and the striped dolphin has a range of 29 to 123 kHz (Kastelein et al. 2003). Compared with other marine mammals, manatees appear to have a similar hearing sensitivity with pinnipeds and some cetaceans of a range between 10 and 26 kHz. This range of frequency seems to allow these marine mammals to hear best in low noise conditions underwater (Gerstein et al. 1999; Ketten 1991). It also appears that manatees and killer whales (Orcinus orca) share a peak hearing sensitivity around 16 kHz (Gerstein et al. 1999). To put this in terms relative to humans, normal hearing range in adults is between 0.04 and 16 kHz (Wartzok and Ketten 1999). The Olfaction and Gustation of Marine Mammals In order to detect materials in the water and in the air, smell and taste are usually closely associated senses. Neither smell nor taste has been examined thoroughly in marine mammals. Smell sensations decreased in marine mammals as they adapted to their aquatic environment, and while taste sensations have been observed in dolphins, there is a lack of experimentation to support this with other marine mammals (Wartzok and Ketten 1999). Anatomical studies have shown that taste and smell of terrestrial mammals are far more developed than that of certain marine mammals including mysticetes, odontocetes, and sirenians. Manatees have a rudimentary olfactory system and both cetaceans and manatees are deficient of a vomeronasal organ, which is an essential factor of the olfactory system (Wartzok and Ketten 1999; Mackay-Sim et al.

PAGE 19

8 1985). Studies done by Hartman (1979) indicated that manatees might use their sense of smell more through their mouth than their nose. Manatees have taste buds at the back of their tongues (Levin and Pfeiffer 2002; Reynolds and Odell 1991), located on the dorsal and posterior lateral swellings (Levin and Pfeiffer 2002; Wartzok and Ketten 1999). The taste responses in manatees appear to be more developed than those in cetaceans. Manatees may be using taste and smell to avoid eating certain plants, as well as for recognition of other manatees and determining if a female is in estrous (Reynolds and Odell 1991). Overall, the taste and smell capabilities of manatees are not well known and require further investigation. The Vision of Marine Mammals The amount of light present in any environment affects the way a terrestrial or marine mammal will view its world. However, for marine mammals, the amount of light in their aquatic environment can be very different from that of the environment of land mammals. Light passing through the water is absorbed, refracted, and scattered, and can then be affected by its wavelength, chlorophyll concentration, and organic matter concentration in the water. All of these factors can potentially influence visual detection and visual acuity (Wartzok and Ketten 1999). Visual detection for marine mammals refers to adaptations that facilitate its becoming aware of predators or prey in the aquatic environment. When an animal identifies a predator or prey, it must take some type of action. If a marine mammal is to find prey at deep depths in the water, then it needs to develop visual adaptations such as corresponding receptor pigment sensitivity to low light environments, having an increased number of photoreceptors, and being able to increase the light intensity captured through the tapetum (Wartzok and Ketten 1999).

PAGE 20

9 Most marine mammals photoreceptors will have the highest sensitivity in the light range of water where they usually feed. For example, the pigments of rod receptors of pinnipeds and cetaceans have their highest sensitivity near the blue end of the spectrum because this is the same as the wavelengths of light piercing the open ocean (Wartzok and Ketten 1999). The number of photons captured dictates what visual signals will be detected (Saari 1992). Marine mammals are able to detect objects in their environment in a special way, especially in the open ocean. Their well-developed tapetum reflects photons and this reflection process gives the visual pigment further opportunity to secure a photon, increasing the chance for detection of objects in the environment (Wartzok and Ketten 1999). Not surprisingly, marine mammals have more highly developed tapeta than any other mammal (Supin et al. 2001; Walls 1942). Visual acuity measures an animals ability to resolve features in its environment and is measured in degrees or minutes of arc (Wartzok and Ketten 1999; Westheimer 1992). The refraction of light rays depends on differences in refractive indices and the angle of light. The lens of terrestrial mammals depends on this refractive index and the radius of curvature, which can generate a lens with a power of 25 to 40 diopters (a diopter is the measure within the lens of refractive power). As marine mammals adapted to their aquatic environment, the strength of their principal refractive structure, the lens, was markedly reduced. This loss of refractive strength occurred because as light passed through water into the corneas, there was a very small change in refractive index due to the refractive index of water and the interior of the eye being so similar. To compensate for this, marine mammals developed spherical lenses, which are more similar to fish than to terrestrial mammals (Wartzok and Ketten 1999).

PAGE 21

10 Within the last century, manatee eye anatomy and vision have not been studied extensively. Early studies conducted by Walls (1942) suggested that manatees have limited visual capabilities and acuity. Other research conducted (West et al. 1991; Piggins et al. 1983) supported Walls theory that manatee eyes were adapted for low light conditions and contained very few ganglion cells, as well as no mechanism for accommodation, thus limiting vision. Behavioral research conducted by Hartman (1979) suggested that vision is the sensory mechanism which manatees rely on the most for receiving information about their environment and that manatees were most likely hypermetropic (vision for distant objects is better than vision for near objects). However, the research of Piggins et al. (1983) suggested that manatees were emmetropic (an eye with normal vision that sharply focuses vision at all distances). Research conducted by Cohen et al. (1982) further supports the idea that manatees vision is not as poor as previously thought. Cohen and his colleagues found two types of photoreceptors, rod-like and cone-like, in the eyes of the West Indian manatee. The photoreceptors were found in the retina at low rod:cone ratios suggesting that the occurrence of a large number of cone cells in the rod-dominated retina would indicate relatively good visual acuity. They also suggested that by having both rods and cones in the retina, manatees were likely to have vision functioning more on a diurnal level than nocturnal. Mass et al. (1997) studied the topographic order of ganglion cells in the retina of the Florida manatee. They found that the ganglion-cell distribution did not appear to be even and was diverse across the retina with higher ganglion cell densities near the center of the retina. Based on these findings using ganglion cell densities, the investigators were able to calculate retinal resolution. Retinal resolution can be used as an accurate measure

PAGE 22

11 for visual acuity because the retinal resolution in the normal eye has evolved to have a similar resolution as the eye optics (Supin et al. 2001). Visual acuity can be measured in minutes or degrees of visual arc, which is also called cycles per degree. Mass and his team (1997) found that the Florida manatee had limited visual resolution; 20 of visual arc, suggesting manatees are myopic (nearsightedness or when blurred vision results as objects are moved further away). These anatomical findings are consistent with behavioral studies conducted by Bauer et al. (1999) and Colbert et al. (1999). Two manatees were trained with two-choice simultaneous discrimination tasks. The manatees were trained to discriminate vertical black and white bands equated for brightness. During each training trial, a standard and a target with wider vertical stripes were presented. The manatees were then reinforced with food for choosing the target with wider stripes. Visual acuity of the manatees was measured in degrees of visual arc depending on the size of the vertical bands and the distance the eyes were from the target. Based on these visual discrimination tasks that the two manatees were trained to do, the research team found that one manatee had a visual arc of 23 and the other manatee had a much lower level of visual acuity at 1 degree of visual arc. Compared with other mammals, both terrestrial and marine, there is a wide range of visual acuity results. It has been found that mixed breed kittens have a visual acuity measure of 3.86 cycles per degrees (Whittle et al. 1987); horses have a measure of 23.3 cycles per degree (Timney and Keil 1992); the bottlenose dolphin, Tursiops truncates, measures at 9 in water (Mass and Supin 1995); the harbor porpoise measures between 2.2 and 2.6 cycles per degree; the gray whale, Eschrichtius gibbosus, has a visual acuity of 2.4 to 2.9 cycles per degree;

PAGE 23

12 and the Amazon river dolphin, Inia geoffrensis, has a very poor visual acuity of 0.6 to 0.75 cycles per degree (Supin et al. 2001). Recent studies conducted by Griebel and Schmid (1996) found that manatees do have limited color vision. Manatees were trained in captivity to discriminate a gray target during a two-fold simultaneous choice test. Three different shades of blue, green, and red were tested against varying gray colors. A color would first be presented with a gray that was almost black. As trials continued, the brighter grays were presented until a gray almost white in color was reached. If the manatee discriminated the color from gray, then it was assumed that color was perceived. Discrimination between some shades of gray and a color was not possible if they had a similar brightness. The manatees were able to discriminate blue and green from several gray hues, but could not distinguish red and blue-green from gray hues. Griebel and Schmid suggested that the color vision of manatees was similar to that of fur seals and California sea lions in which similar methodologies were used to test color vision. The method used to test color vision of manatees suggests they possess dichromatic color vision. In addition, manatees were also able to distinguish a color independently of the shape of the stimulus and other colors. Additionally, the spotted seal seems to possess some capacity for color discrimination based on behavioral studies (Peichl et al. 2001). However, another marine mammal, the bottlenose dolphin, does not seem to have the ability to discriminate colors based on behavioral tests (Madsen and Herman 1980). An additional study conducted by Griebel and Schmid (1997) investigated brightness discrimination in the manatee. Two manatees were trained to discriminate gray targets that varied in brightness in a two-fold simultaneous choice test. Thirty

PAGE 24

13 different shades of gray were used and the manatee made a decision between different targets by touching its choice with its snout. The Weber fraction, which is a formula used to calculate the difference in relative reflection, was used to determine how well the manatees discriminated differing shades of brightness. It was found that manatees have brightness discrimination ability three times less than that of humans. This additional study further suggested that manatees do possess dichromatic color vision and that the brightness discrimination ability is similar to that of the fur seal. A histological examination conducted by Samuelson et al. (1994) examining the entire eye of four Florida manatees found some very interesting and unique results. They found that the corneas were small and round with anterior epithelium of 14-18 cell layers. A diffused capillary bed was found under the anterior epithelium in each eye. A large majority of the stroma in the cornea appeared to be vascularized. This was an amazing find because in no other species of animal have blood vessels been consistently found in the cornea with no signs of injury or infection (Klintworth 1991). An additional discovery of interest made by Samuelson et al. (1994) was that manatees lack a nontapetal choroid. Based on their histological and anatomical findings, they suggested that manatees visual systems function diurnally and are unique from other marine mammals because they have no tapetum. The Cornea Several interesting discoveries were made by Samuelson et al. (1994) upon histologically examining the eye of the Florida manatee, especially the cornea. In order to understand the significance of finding blood vessels in several corneas without injury, it is important to understand the morphology and anatomy of the normal cornea in most mammalian species. The cornea is the anterior-most, transparent avascular layer of the

PAGE 25

14 eye. Its primary functions are to provide support for the contents within the globe, refract light, and transmit light. Corneal shape is elliptical, and in most species, the horizontal diameter is larger than the vertical diameter. Cats and dogs have a diameter that is very similar, but ungulates have a much wider horizontal diameter, which aids in their horizontal field of view. The thickness of the cornea varies between species. Usually the thickest area of the cornea is along the peripheral edges and the thinnest area of the cornea is in the center (Samuelson 1999). The cornea is a transparent tissue because it lacks blood vessels, is supplied with sensory nerves, has a nonkeratinized surface epithelium, and lacks pigmentation (Muller et al. 2003). The cornea is composed of four (and sometimes five) layers including the anterior epithelium, Bowmans layer, stroma, Descemets membrane, and the endothelium (Samuelson 1999). The anterior epithelium covers the corneal anterior surface. It is nonkeratinized and made up of stratified squamous cells. In carnivores, the epithelium is 25 to 40 m thick, while in ungulates the epithelium can be two to four times thicker. The epithelium in cats, dogs, and birds consists of a single layer of columnar basal cells, which lie on a thin basement membrane and give rise to several layers of wing-shaped cells and two to three layers of nonkeratinized squamous cells. Larger animals have more layers of wing-shaped and squamous cells (Agrawal and Tsai 2003). A basement membrane lies under the epithelium. The basement membrane is usually 30 to 55 nm thick and is anchored to the stroma by hemidesmosomes. The hemidesmosomes vary among species; they are linear in mammals and amphibians, rosettes in birds and reptiles, and without arrangement or absent in fish. Regenerative capabilities of epithelial cells are good (Buck 1983).

PAGE 26

15 The stroma, which comprises approximately 90% of the corneal thickness, is made up of sheets of lamellae tissue that are transparent. Cells called keratocytes can be found among the lamellae sheets. These keratocytes aid in the maintenance and formation of the lamellae, and can form into fibroblasts if injury within the cornea takes place. The lamellae run parallel with the diameter of the cornea. Maintenance of corneal clarity is the primary function of the stroma. The stromas organization allows 99% of light to pass through the cornea without scattering and the primary support structure of the stroma is made up of collagen fibrils, proteoglycans, and glycoproteins, which constitute 15 to 25% of the stroma (Joyce 2003; Hogan et al. 1971). In humans, nonhuman primates, and avian species, an additional corneal layer exists. This layer is called Bowmans layer and is found in the anterior-most region of the stroma. Collagen fibrils in Bowmans layer are found randomly and are small in diameter compared to the stroma. The Bowmans layer is 10 to 15 m thick and is acellular. The collagen found in Bowmans layer is produced by the anterior epithelium (Hayashi et al. 2002). Posteriorly, there is another membrane, Descemets membrane, which is an acellular membrane that forms a protective border within the cornea. Descemets membrane is produced by the posterior endothelium and has been found to be elastic with fine collagen fibrils. The membrane has a thin anterior zone next to the stroma and two broad zones located in a posterior region next to the endothelium (Hayashi et al. 2002). The innermost cornea is lined with one layer of flattened cells called the corneal endothelium. Mitosis takes place in immature animals within the endothelium usually, but in older individuals the ability to divide is lost. In adult eyes, the surface of the corneal endothelium has small microvillae and pores spotted on it with

PAGE 27

16 hexagonally shaped cells. With age, there is a loss in hexagonal shape due to a decrease in the density of the epithelium (Laing et al. 1976). Even though the normal corneas of terrestrial and aquatic mammals are similar, there are some differences. Studies conducted by Dawson et al. (1987) with the corneas of cetacean species found few differences from the corneas of terrestrial mammals. However, a more current study conducted by Miles (2002, unpublished) found several differences between the corneas of marine and terrestrial mammals. Miles found through histological examination and three-dimensional reconstruction that cetacean corneas were thinner centrally than that of terrestrial mammals. Miles also found that cetacean corneas lacked a Descemets membrane. It was postulated that this membrane was absent to allow the cornea to change its curvature to enhance refractive capabilities. Additionally, Miles found that some species of cetaceans lacked a Bowmans membrane (Kogia breviceps and Kogia simus) while other species had a thick, prominent Bowmans membrane (Tursiops truncates and Globicephela macrorhyncus). Collins and Collins (2000) found that the mean density of the anterior epithelium was thicker in aquatic mammals than that of terrestrial ones. These variations are most likely due to differences in environments. An additional study found that sea otters (Enhydra lutris) have an extensively developed anterior epithelium. It is believed that the epithelium has developed this way as an adaptation to the salinity of the aquatic environment in which it lives (Murphy et al. 1990). An investigation into the corneal anatomy of the Florida manatee conducted by Samuelson et al. (1997) found blood vessels consistently present in eight eyes that were examined histologically. The number, size, and depth of the vessels varied between the

PAGE 28

17 eyes. In all the specimens examined, no signs of infection or injury were present. This was a very unusual finding because normal mammalian corneas are avascular (Samuelson 1999; Klintworth 1991). The presence of blood vessels in the cornea is usually considered a pathological condition associated with infection or injury (Klintworth 1991). Angiogenesis Angiogenesis is the biological process in which blood vessels form from existing vasculature (Nicosia and Villaschi 1999; Battegay 1995, Folkman 1995; Schultz and Grant 1991). Angiogenesis occurs in almost every tissue and is a necessary process that is critical in normal conditions such as development, reproduction, healing of wounds, bone repair, ischemic heart disease, ischemic peripheral vascular disease, tumor growth and metastasis, and diabetic retinopathy (Battegay 1995; Adamis et al. 1994). Angiogenesis can also occur under pathological conditions such as cancer, rheumatoid arthritis, trauma, or infection (Schultz and Grant 1991). Early vessel formation occurs through a process called vasculogenesis, in which endothelial cells separate and grow in an avascular tissue, and then unite to form a primitive tubular network. This early network of vessels includes the aorta and major veins. Interconnected branching patterns of vessels form through angiogenic remodeling that are characteristic of mature vasculature. At this stage, endothelial cells join together securely with supporting cells to form mature vessel walls (Yancopoulos et al. 2000). New blood vessels form after the first capillary tubes have developed through angiogenesis (Nicosia and Villaschi 1999). These vessels sprout into previously avascular tissue (Yancopoulos et al. 2000) as a result of endothelial cell migration and proliferation, and then further develop (Nicosia and Villaschi 1999). This process is

PAGE 29

18 referred to as angiogenic sprouting and is responsible for vascularizing some tissues throughout normal development (i.e., neural tube and retina) (Yancopoulos et al. 2000). Mitosis of endothelial cells increases the sprout and a small lumen develops due to the curve within each cell. Eventually a loop is formed and blood flow occurs when two hollow sprouts join at the ends. The process of angiogenesis is complete when pericytes travel within the newly formed loop structure (Schultz and Grant 1991). Many factors, including cytokines, control the establishment and alteration of blood vessels. Perhaps the most important determinant in vessel formation is vascular endothelial growth factor, VEGF. It is distinguished for its capacity to cause vascular leakage and permeability, in addition to its capacity to promote vascular endothelial cell production (Yancopoulos et al. 2000). VEGF is a cytokine that is an endothelial cell mitogen and heparin-binding, dimeric protein (Miller et al. 1994; Klagsbrun and DAmore 1991). This cytokine is necessary in initiating the formation of immature vessels during vasculogenesis or angiogenic sprouting that occurs in development and in adults (Yancopoulos et al. 2000). VEGF is able to stimulate endothelial migration, proliferation, and proteolytic activity and has a signal peptide that is secreted through numerous pathways. VEGF is made by many different cells, including endothelial cells, smooth muscle cells, and fibroblasts (Nicosia and Villaschi 1999). Mesenchymal and stromal cells also secrete VEGF (Beck and DAmore 1997). The VEGF gene is composed of eight exons that are separated by seven introns. Alternative splicing of the VEGF gene has produced four isoforms, including VEGF121, VEGF189, VEGF206, and VEGF165 (Nicosia and Villaschi 1999). Four VEGF-related genes have been discovered (VEGF-B, VEGF-C, VEGF-D, and placenta growth factor),

PAGE 30

19 and they produce glycosylated dimers that are secreted after cleavage of their signal peptide. Two classes of high-affinity binding sites on endothelial cells have been found for VEGF with the molecular mass of these binding sites of 180-200 kDa (Ortega et al. 1999). The tyrosine kinases primarily expressed by the endothelium, Flt-1 (VEGFR-1) and Flt-1/KDR (VEGFR-2), are receptors for VEGF. The secretion of VEGF is noticeably stimulated by hypoxic conditions (Nicosia and Villaschi 1999). Both VEGFR-1 and VEGFR-2 bind VEGF with a high affinity and have seven immunoglobulin-like domains, one transmembrane region, and a tyrosine kinase sequence that is interrupted by a kinase insert domain. There are two promoters of VEGFR-1 and VEGFR-2 that contain a 5 flanking sequence that is needed for endothelial specific expression. An additional receptor for VEGF is neuropilin-1. It modulates the interaction between VEGF and VEGFR-2 (Ortega et al. 1999). The expression of VEGFR-1 in naive cells mediates cell migration. Both VEGF receptors are tyrosine phosphorylated. However, the phosphorylation of VEGFR-1 is not required for function in vivo. Adaptators of the Src family, including Fyn and Yes, are phosphorylated in response to VEGF/VEGFR-1 but not to VEGF/VEGFR-2 interactions. In contrast, phosphorylated VEGFR-2 links with Shc, Grb2 and Nck, but also with phosphatases of SHP-1 and SHP-2, and initiates the phosphorylation of the MAPK cascade through the stimulation of Raf. VEGF has also been found to induce the tyrosine phosphorylation and recruitment of focal adhesion kinase (Ortega et al. 1999). Angiogenesis can occur in endothelial cells during events such as tumor progression, diabetic retinopathy, and rheumatoid arthritis. Local angiogenesis may be due to the release of soluble mediators by the involved tissues. At this point there is a

PAGE 31

20 switch in the phenotype of the quiescent endothelial cell to an activated phenotype. The endothelial cells are then able to respond to mitogenic signals (Ortega et al. 1999). There is a larger increase in the upregulation of genes modulated by VEGFR-2 than with VEGFR-1. These induced genes include growth factors, protease inhibitors, receptors, transcription factors, and actin-binding proteins (Yang et al. 2002). At this point, mitogenic growth factors are released and this changes the activated phenotype to an angiogenic phenotype. Interactions between angiogenic growth factors and their receptors provide the signals for cell migration, proliferation, and differentiation to form new vessels (Ortega et al. 1999). The normal cornea is usually an avascular structure. If blood vessels are present within the cornea, this is referred to as corneal angiogenesis or vascularization and is considered a pathological condition that can affect visual acuity. The causes of corneal vascularization are varied and in humans can include infections (such as trachoma and herpes), immunologic processes, alkali burns, toxic and nutritional deficiencies, and the wearing of contact lenses. In experimental settings, corneal vascularization can be induced through chemical, microbiological, and physical injuries (such as burns or implantation of pellets containing a wide range of chemicals), nutritional deficiencies, and immunological reactions (Klintworth 1991). Rabbits have often been used to study experimentally induced corneal vascularization. After an injury of some type has occurred in the cornea, cell migration and mitosis are involved in healing of the epithelial corneal surface. If abrasion occurs, cell migration takes place almost immediately. In order to return the epithelium to its usual thickness, cell division follows. Cell migration can heal small injuries within 24

PAGE 32

21 hours. If a larger portion of the anterior epithelium is injured, the limbus is the main source for production of epithelial stem cells. Endothelial cells are not able to respond as quickly to cell loss. In the absence of new cell formation, the presence of too few endothelial cells may result in edema, which can result in corneal disease (Samuelson 1999). The stromal edema that occurs with corneal vascularization aids in the growth of blood vessels in the cornea and this is an inflammatory process. There are many inflammatory mediators that are involved in this process including histamine, serotonin, ADP, prostaglandins, many cytokines, free radicals, and immune aggregates. However, several factors can help reverse the process including epinephrine, cortisone, and some drugs such as antihistamines and glucocorticoids (Klintworth 1991). When an injury of the cornea occurs, stroma edema takes place and neutrophils move to the source of the injury to release enzymes and chemotaxic agents. Epithelial wound healing begins within one hour of the injury and wing layer cells flatten and slide next to the injury site in order to cover the area within 24 to 96 hours. The epithelium returns to its full thickness by the action of basal stem cells of the limbus. Proliferation of the stromal fibroblast takes place after 24 hours (Ahmadi and Jakobiec 2002). Leukocytes move in the extravascular space within 24 hours and move through the corneal stroma to the site of injury. The first new vessels appear after 27 hours. Advancing new vessels remain permeable to fluid, plasma proteins, and smaller proteins. Forming corneal blood vessels do not have firm intercellular complexes; whereas, well-formed corneal vessels in advanced stages of wound healing do not leak. Once new vessels have developed, the basal lamina breaks down allowing the endothelial cells to

PAGE 33

22 move into the extracellular matrix for further vessel formation. Progression of these endothelial cells at the end of the developing new blood vessels extends towards the damaged area (Klintworth 1991). The movement of endothelial cells, which elongate and form solid sprouts developing lumen, follows this. Two sprouts form a loop and blood flows, and the process is repeated from the top of the loops (Pepose and Ubels 1992). New blood vessels can emerge within 33 hours after injury. A network of vessels is formed that becomes linked and these vessels then begin to circulate (Klintworth 1991). Objectives This study will examine several aspects of corneal vascularization in the Florida manatee, Trichechus manatus latirostris. The purpose of the first part of this study is to describe more clearly corneal vascularization by examining the architecture through three-dimensional reconstruction in order to find possible patterns in size, distribution, and location of blood vessels that may be related to age, gender, and environment. Secondly, immunohistochemistry will be performed to determine if there is a presence or absence of vascular endothelial growth factor (VEGF) receptors-1 and -2 in the corneas of two manatees varying in age. This will be done in order to try and further explain what is occurring in the corneas at a biochemical level.

PAGE 34

CHAPTER 2 CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE AND THREE-DIMENSIONAL RECONSTRUCTION Introduction The eye is an amazing organ that allows organisms the ability to interpret light-based information about their environment. Normal ocular functions are, in part, supported by temporary and permanent vascular beds found throughout the developing and mature eye. During embryonic development, temporary vascular beds are associated with the lens. Posteriorly, the lens is supplied with blood from the hyaloid artery system (Drohan et al. 2002), which consists of an extensive arterial, arteriolar and capillary network throughout the vitreous body (Los et al. 2000). The hyaloid artery moves toward the lens from the optic nerve head and provides a network of vessels that covers the posterior side of the lens (tunica vasculosa lentis). The lens is also supplied anteriorly with blood from vascular branches that form the pupillary membrane and are fed by long ciliary arteries (Drohan et al. 2002) and are drained by vessels in the area of the future ciliary body (Los et al. 2000). The vessels in the tunica vasculosa lentis connect with the pupillary membrane (Strek et al. 1993). Later in gestation the hyaloid artery involutes, resulting in a transparent remnant called the canal of Cloquet that leads to the lens and becomes avascular (Drohan et al. 2002). The regression of the hyaloid artery begins when cellular components of the blood vessel walls break down (Los et al. 2000). In addition to temporary vascular beds, permanent vascular beds can also be found in the eye. The ciliary body is the site for one of the key vascular structures occurring in 23

PAGE 35

24 the eye by the major arterial circle (MAC). In most terrestrial mammals, the ciliary body is supplied by the MAC, which originates from combinations of the anterior ciliary artery and the long posterior ciliary arteries (Morrison et al. 1987). The most extensive vascular bed within the ciliary body occurs within the ciliary processes and is fed by arterioles derived from the MAC. In rats and guinea pigs the blood vessels within the ciliary processes are concentrically parallel and travel posteriorly emptying into the choroidal veins, forming a vascular pattern that is similar to that of primates (Morrison and Van Buskirk 1984). The ciliary process in cats and dogs is supplied by one arteriole that also travels posteriorly and sends out capillary branches that drain outwardly into venous sinuses. Sheep, goats, pigs, and cows have ciliary processes that receive blood from many arterioles and veins that empty into the choroidal circulation (Morrison et al. 1987). In a study conducted by Miles (2002) examining the angioarchitecture of the ciliary body of different whale species, similar vascular patterns were observed in the ciliary body of all species. It was found that a single arteriole located in the apical area of the ciliary body supplied the capillary bed in the bottlenose dolphin. In three other species examined, Globicephela, K. simus, and K. breviceps, several arterioles supplied the capillaries of each ciliary process. Other vessels in the ciliary body of the whale species examined were hypothesized to aid in venous outflow, including many small vessels located at the base of the processes, which collateralized with vessels responsible for collecting aqueous humor. In addition to the smaller veins, larger veins were also found that aided in the movement of blood of the iridal and ciliary processes to the choroidal venous outflow system.

PAGE 36

25 The ciliary body vasculature has also been investigated in the Florida manatee by Natiello (1999). It was found that the ciliary process in the manatee contains many capillaries, partly similar to that observed in cetaceans and pinnipeds. Abundant blood vessels were observed on the outside folds that lead into the choroidal vasculature. The major arterial circle (MAC), found at the base of the iris, fed the ciliary bodys vasculature. The anterior ciliary artery and two long posterior ciliary arteries formed the MAC. Venous outflow of the ciliary body consisted of a unique bifurcated system, suggesting an interesting vascular dynamic associated with aqueous humor production. The choriocapillaris, located in the posterior uvea, provides another example of a permanent vascular bed in the eye, especially for animals with a tapetum lucidum. Blood vessels penetrate the tapetum at right angles and are parallel to the incident light in order to supply the choriocapillaris. The blood vessels go straight through the tapetum so as not to interfere with light reflection. In carnivores such as the dog, cat, ferret, and grey seal, the tapetum under the choriocapillaris is cellular (tapetum cellulosum) and indents into the RPE so that the tapetum has a flat, reflective surface (Braekevelt 1986). With some species of animals that have a tapetum fibrosum, such as the sheep, the choriocapillaris is not indented into the RPE (Braekevelt 1986b). The vasculature of the tapetum in most domestic species consists of medium-sized vessels in the dorsal section of the choroid. Variations in the reflective color of the tapetum can result from choroidal blood vessels. Many small vessels pass through the tapetal layer and form the choriocapillaris, which is a single layered capillary bed. The lumen of the choriocapillaris is wide, which allows red blood cells to go through easily (Samuelson 1999). In the posterior pole of the choriocapillaris, alternating feeding

PAGE 37

26 arterioles and draining venules are found (Alm 1992). The area including the RPE, choriocapillaris and Bruchs membrane is responsible for a number of functions including moving metabolites to photoreceptors, stabilization of the structural design of outer segments of the photoreceptors, storing vitamin A for visual pigments, and phagocytosis and lysosomal break down of photoreceptor outer segments (Braekevelt 1983). Bhutto and Amemiya (2001) conducted a study to investigate the architecture of the rat choroid by using scanning electron microscopy of vascular corrosion casts. In this study, it was found that the choriocapillaris viewed from the retinal side formed a network of capillaries that ranged in diameter and appeared to resemble a dense honeycomb pattern of vessels. Additionally, an irregular pattern of vessels was also observed. The two different vascular patterns in the choriocapillaris were evenly distributed throughout the choroid except in the peripheral area. When viewed from the posterior side, the choriocapillaris vascular pattern resembled a palm-like pattern that ended at the ora serrata. Of particular interest is that vision was not impeded even with a dense network of blood vessels forming the choriocapillaris. Permanent vascular beds can also be found in the mammalian retina. In order for normal function to occur in a retina over 160 m in thickness, ocular blood flow must be strictly maintained. The central retinal vein controls blood flow from the retina and is formed from retinal veins meeting on the optic nerve head. The central retinal vein has two stems at the disc that connect while they are still in the optic nerve (Ruskell 1998). Retinal perfusion for most mammals is carried out by the choroidal vessels, which are found between the outer retina and sclera (Steinle et al. 2000). The vascular needs of the

PAGE 38

27 outer retina are also met by choroidal circulation. Species that have a highly vascularized inner retina usually have a wide range of vessel patterns and these vascular patterns are determined by density of neural cells in the ganglion cell layer forming an area centralis (Dunlop et al. 1997). Capillaries within the retina have been found to be very thin in diameter and contain fewer corpuscular elements than the choriocapillaris (Ninomiya and Kuno 2001). Major blood vessels in the retina avoid areas of high neural cell density. By doing this, visual acuity is not negatively affected and many species of animals display a similar retinal vascular pattern (Dunlop et al. 1997). In some snake and lizard species, there is a unique ocular structure called the spectacle or brille, which is a fused eyelid that forms a clear covering over the cornea. In some species of snakes the spectacle is about the same size as the cornea, and in other species the spectacle is much larger than the cornea and can even cover a portion of the head (Sivak 1977). Walls (1942) originally believed that a sebaceous material entered between the spectacle and cornea to act as a lubricant. Walls also thought that if a spectacle had a flattened curvature, then the refractive involvement of the cornea would be decreased. Sivak (1977) found that the spectacle essentially replaced the cornea as a refractive component and that the spectacle plays the major refractive role. Mead (1976) injected microsilicone into the vessels of the spectacle and found that the spectacle was indeed vascularized, having vessels that transversed the stromal layer of the spectacle. Visualization, using a slit lamp at a power of 32 X, was possible and showed small vessels with transparency in the walls of the vessels. The vessels were only visible because of the reflection of red blood cells that were circulating through them. It was also found that there was a vertical arrangement of vessels in the spectacle

PAGE 39

28 of members of some snake families. The most interesting finding of this study was that the spectacle maintained a vascular pattern and still retained a high level of transparency. The normal healthy cornea, unlike other regions such as the retina and choriocapillaris, is an example of an ocular structure that is normally avascular. However, if blood vessels are present within the cornea, this is referred to as corneal angiogenesis or vascularization and is considered a pathological condition that affects visual acuity. The causes of corneal vascularization are varied and in humans can include infections (such as trachoma and herpes), immunologic processes, alkali burns, toxic and nutritional deficiencies, and the wearing of contact lenses. Except for contact lenses, identical causes have been observed in domestic species. In experimental settings, corneal vascularization can be induced similarly through chemical, microbiological, and physical injuries, nutritional deficiencies, and immunological reactions (Klintworth 1991). Rabbits have often been used to study experimentally induced corneal vascularization. After an injury of some type has occurred in the cornea, cell migration and mitosis are involved in healing of the epithelial corneal surface. If abrasion occurs, cell migration takes place almost immediately. In order to return the epithelium to its usual thickness, cell division follows. Cell migration can heal small injuries within 24 hours. If a larger portion of the anterior epithelium is injured, the limbus is the main source for production of epithelial stem cells. Endothelial cells are not able to respond as quickly to cell loss. In the absence of new cell formation, existing endothelial cells compensate for the loss. Too few endothelial cells are unable to provide adequate deturgescence, resulting in corneal edema (Terry and Ousley 2003; Vemuganti et al.

PAGE 40

29 2002). The cornea is a relatively dehydrated tissue being made up of 75 to 85% water. Deturgescence is the state of dehydration in the cornea and is carried out by the endothelium and epithelium (Samuelson 1999). The stromal edema that occurs with corneal vascularization aids in the growth of blood vessels in the cornea and this is an inflammatory process. There are many inflammatory mediators that are involved in this process including histamine, serotonin, ADP, prostaglandins, many cytokines, free radicals, and immune aggregates (Klintworth 1991). Angiogenesis is the biological process in which new blood vessels form from existing vasculature (Nicosia and Villaschi 1999; Battegay 1995; Folkman 1995; Schultz and Grant 1991). Angiogenesis occurs in almost every tissue and is a necessary process that is critical in normal conditions such as development, reproduction, healing of wounds, bone repair, ischemic heart disease, ischemic peripheral vascular disease, and diabetic retinopathy (Battegay 1995; Adamis et al. 1994). Pathological conditions in which angiogenesis can occur include cancer, rheumatoid arthritis, trauma, tumor growth and metastazation, or infection (Schultz and Grant 1991). Early vessel formation occurs through a process called vasculogenesis, in which endothelial cells separate and grow in an avascular tissue, and then unite to form a primitive tubular network. The aorta and major veins are examples of this early network of vessels. Interconnected branching patterns of vessels that are characteristic of mature vasculature form through angiogenic remodeling. At this stage, endothelial cells join together securely with supporting cells to form mature vessel walls (Yancopoulos et al. 2000). New blood vessels form after the first capillary tubes have developed through angiogenesis (Nicosia and Villaschi 1999). These vessels sprout into previously

PAGE 41

30 avascular tissue as a result of endothelial cell migration and proliferation, and then further develop (Yancopoulos et al. 2000; Nicosia and Villaschi 1999). This process is referred to as angiogenic sprouting and is responsible for vascularizing some tissues throughout normal development (i.e., neural tube and retina) (Yancopoulos et al. 2000). Mitosis of endothelial cells increases the sprout and a small lumen develops due to the curve within each cell. Eventually a loop is formed and blood flow occurs when two hollow sprouts join at the ends (Schultz and Grant 1991). When an injury of the cornea occurs it is accompanied with stromal edema and eventually corneal angiogenesis as neutrophils move to the source of the injury to release enzymes and chemotaxis agents. Epithelial wound healing begins within one hour of the injury and wing layer cells flatten and slide next to the injury site in order to cover the area within 24 to 96 hours. The epithelium returns to its full thickness by the action of basal stem cells of the limbus. Proliferation of the stromal fibroblast takes place after 24 hours (Ahmadi and Jakobiec 2002). Leukocytes move in the extravascular space within 24 hours as well as traveling through the corneal stroma to the site of injury. The first new vessels appear after 27 hours. Advancing new vessels remain permeable to fluid, plasma proteins, and smaller proteins. Forming corneal blood vessels do not have firm intercellular complexes, whereas well-formed corneal vessels in advanced stages of wound healing do not leak. Once new vessels have developed, the basal lamina breaks down at specific sites, allowing the endothelial cells to move into the extracellular matrix for further vessel formation. Progression of these endothelial cells at the end of the developing new blood vessels extends toward the damaged area (Klintworth 1991). The movement of endothelial cells, which elongate and form solid sprouts developing lumen,

PAGE 42

31 follows this. Two sprouts form a loop and blood flows, and the process is repeated from the top of the loop (Pepose and Ubels 1992). New blood vessels can emerge within 33 hours after injury. A network of vessels is formed that become linked and these vessels then begin to circulate (Klintworth 1991). An investigation into the corneal anatomy of the Florida manatee (Trichechus manatus latirostris) conducted by Samuelson et al. (1997) consistently found the presence of blood vessels within the corneas of eight eyes that were examined histologically. The number, size, and depth of the vessels varied among eyes. In all the specimens examined, no signs of infection or injury were present. The interference of corneal vascularization on the manatees ability to see, i.e., visual hindrance, had not been assessed. Within the last century, manatee eye anatomy and vision has not been studied extensively. Early studies conducted by Walls (1942) suggested that manatees have limited visual capabilities and acuity. Other research conducted supported Walls theory that manatee eyes were adapted for low light conditions and contained very few ganglion cells, as well as no mechanism for accommodation, thus limiting vision (West et al. 1991; Piggins et al. 1983). However, behavioral research conducted by Hartman (1979) suggested that vision is the sensory mechanism in which manatees rely on the most for receiving information about their environment. More recent research carried out by Mass et al. (1997) found that the ganglion-cell distribution in the retina did not appear to be even and was diverse across the retina with higher ganglion cell densities near the center of the retina suggesting a limited visual resolution. These anatomical findings were

PAGE 43

32 consistent with behavioral studies conducted by Bauer et al. (1999) and Colbert et al. (1999) where they also found that manatees had a limited visual ability. We propose as our central hypothesis that corneal vascularization in the Florida manatee affects the vision of these animals. The amount of vascularity from collected specimens will be determined. The pattern(s) and amount of corneal vascularization from previously collected specimens will be determined by light microscopic three-dimensional reconstructions of the vascular elements found in the corneas from 22 deceased manatees. Correlations of the amount of vascularity with factors including age, gender, time of year, and location where the animals were found will be made. This study will lay the foundation for determining the extent to which corneal vascularization exists in manatees and for understanding the potential impact of this unique condition on their ability to survive. Materials and Methods Specimen Collection Florida manatee eyes were collected from Dr. Gordon Bauer at New College in Sarasota, Florida. Dr. Bauer received the eyes from Florida Marine Research Institute (FMRI) in St. Petersburg, Florida. Twenty-six eyes from 22 individuals were used and the eyes were taken from necropsied animals over a period of 5 years. Each eye was labeled with an identification number given to it by FMRI. Information including gender, size, county where the animal was found, and time when the animal was found was recorded. Manatee eyes were placed in fresh 10% neutral buffered formalin solution (Luna 1968).

PAGE 44

33 Preparation of Eyes All extraneous tissue surrounding the globe was removed using a scalpel (see figure 2-1). Individual corneas were measured in millimeters from dorsal to ventral and medial to lateral sections. An additional measurement was taken from the cornea to the optic nerve. After measurements were taken and recorded, the corneas were removed from the globe by cutting 1 mm from the limbus into the sclera with a sharp razor. A notch was cut into the limbus at 12:00 in the dorsal section of each cornea in order to know orientation during histological examination. The remainder of the eye was stored and the cornea was placed in a tissue cassette, labeled and then returned to the 10% neutral buffered formalin solution until processing. Histology In preparation for embedding, the corneas were dehydrated by placing them in the following ethyl alcohols for an hour each: 70%, 80%, two changes of 95%, followed by three changes of 100%. All dehydration steps were prepared by using ethyl alcohol. The tissue was then cleared with two changes of xylene. The tissues were infiltrated with two changes of Fisher Tissue Prep at 57 Celsius. The tissues were then embedded using Fisher Tissue Prep T565 in metal molds and allowed to harden. Using a Reichert-Jung 2030 microtome, the embedded corneas were cut in serial dorsal/ventral sections at a thickness of 10 microns. The tissue ribbons were floated out on a 60 Celsius water bath and then placed on 3 x 1 x 1 Superfrost microscope slides (Fisher Scientific, Pittsburgh, Pennsylvania) with three sections per slide. Each slide was labeled with the identification number of the animal, order of section, and orientation of tissue. Approximately 35 slides and 105 sections per eye where cut. After cutting was

PAGE 45

34 completed, the tissues were placed in a slide rack and into an oven at 60 Celsius overnight to dry. After the slides had dried completely, they were deparaffinized and then stained using a Massons Trichrome Stain (Humason 1972). When staining was completed, the slides were coversliped using synthetic xylene based mounting media, Eu-kitt. Three-Dimensional Reconstruction Three-dimensional reconstructions of corneal vasculature where carried out using the Light Microscopy facility in the McKnight Brain Institute at the University of Florida. Slides that had been histologically processed were placed on a Ludl-motorized scanning stage of an AxioPlan 2 Zeiss Microscope. A Sony model DXC 970MD camera was used to take individual pictures of each serial section in sequential order. Imaging Research Incorporated (MCID) created the computer software used to produce the images. Initially each section was digitized at a magnification of 25 X (2.5-Zeiss-Plan/NEO Floar). The sections then were aligned based on landmarks (e.g., blood vessels or the outline of the cornea) that were common with the previous section. The number of images taken for each cornea varied from 16 to 48 sections (each section being 10 microns in thickness). Once all images per cornea were taken, the blood vessels were outlined in red and the corneal tissue was outlined in yellow for each section using a mouse driven color-coded screen cursor. The MCID program has an alignment feature that allows the user to align the sections optically. Once the images of each cornea were stacked and aligned, the MCID program then created a three-dimensional cube. The cube could be rotated 360 degrees, as well as display the outlined structures (blood vessels and corneal outline) within the stacked images that comprised each cornea. Other factors

PAGE 46

35 including transparency, lighting, and viewing orientation could be manipulated allowing for several images to be taken of each cornea. Due to these functions on the MCID program, detailed maps of the spatial arrangement of blood vessels in the cornea could be made. Three maps were made for each cornea that included a face-on view, a dorsal or ventral view, and a lateral or medial view. For each map that was constructed, an image was made that included both outlined features and an additional map was also made that included only the blood vessels present in the cornea. The MCID program was also able to calculate the volume and area of outlined features in each cornea, which allowed us to analyze and measure the vascularization as a percent of the total volume of tissue examined. After examination of the maps constructed at 25 X, three additional corneal maps were constructed at 200 X (20X-Zeiss-Plan/APO Chromat). Two of the corneas were highly vascularized and one had a low percentage of vascularization. Each cornea was divided into dorsal, ventral, medial and lateral quadrants. The same process was carried out within each quadrant as previously described, but at 200 X instead of 25 X. However, instead of outlining all the corneal tissue, only the stroma was outlined in order to visualize vasculature that was passing from the stroma to the anterior epithelium. This process of increasing the magnification and separating the stroma and anterior epithelium was done in order to examine the vascular patterns of smaller, feeder vessels passing between the two cornea layers, as well as determine the vascularization by these blood vessels as a percent of the total volume. Additionally, the diameter of blood vessels at 25 X and 200 X was calculated in microns using a Nikon Labophot-2 microscope.

PAGE 47

36 Statistics One-way ANOVAs were carried out using the SAS computer program based on blood vessel as a percent of the total volume in each cornea calculated for 20 individuals. Four different parameters were analyzed including gender (male versus female), age class (calf, juvenile, or adult), coast found (east versus west), and season found (Spring, Summer, Fall, or Winter). Results Anatomical Description The stroma of the cornea stained a light blue to medium blue color. Within the stroma, blood vessels can be seen that are either red in color due to red blood cells or lack color and can be seen because of their hollowed shape (Figures 2-2 through 2-6). The border surrounding the blue stroma is stained a deep red to purple and is the anterior epithelium. In Figures 2-2 and 2-3 vessel penetration between the two layers can be easily observed. As Figures 2-2 through 2-6 illustrate, the vessels found in the cornea vary in shape, size, and location. Diameters of these vessels were measured at 25 X in microns using several specimens including MNW-9016, MNW-9114, MNW-9022, and MSE-9105. The average vessel diameter for MNW-9016 was 60 m, 75 m for MSW-9114, 82.1 m for MNW-9022, and 121.4 m for MSE-9105. The overall average for vessel diameter at 25 X was 84.6 m. Orientation (left eye, right eye, dorsal, ventral, lateral, or medial) of the manatee cornea as used for three-dimensional reconstruction can be viewed in Figure 2-7 and the actual reconstructions of manatee corneal tissue at 25 X are represented in Figures 2-8 through 2-33. In Figures 2-8 through 2-33 A, a face-on view of both the corneal tissue and vasculature can be seen. The tissue present represents approximately one-fourth of

PAGE 48

37 the entire cornea. If the corneal tissue is removed, then only the vessels of the cornea can be viewed (Figures 2-8 through 2-33 B). This allows for better visualization of the vasculature present without it being obscured by corneal tissue. Figures 2-8 through 2-33 C represent a dorsal or ventral view of each tissue with the vessels present. The view is dorsal or ventral depending on where most vasculature was observed through three-dimensional reconstructions. In Figures 2-8 through 2-33 D, only the vasculature within the cornea can be seen, which allows observations to be made of vascular patterns as they pass through the tissue. In Figures 2-8 through 2-33 E and F, the same applies as with Figures 2-8 through 2-33 C and D. However, the views represent a medial or lateral (horizontal) perspective of the tissue and vasculature (Figures 2-8 through 2-33 E) or just the vasculature within the corneal tissue (Figures 2-8 through 2-33 F). If left or right orientation of an individual eye was not determined, then the medial or lateral perspective was labeled horizontal instead. The three-dimensional reconstructions of the corneal vascularization have the appearance of spaghetti-like entangled networks. Tissues with less vascularization make up the majority of the samples (Figures 2-10, 2-11, 2-14, 2-16, 2-18 through 2-30, and 2-33) and appear to have randomly scattered vessels across the area of the tissue. No apparent patterns can be seen with the vasculature and the vessels were found throughout all parts of the cornea. Corneal tissues that are more highly vascularized (Figures 2-8, 2-9, 2-12, 2-13, 2-15, 2-17, 2-31, and 2-32) have greater amounts of blood vessels that can be seen in the reconstructions and cover more area of tissue. These tissue specimens resemble the less vascularized tissue specimens in that they appear random in nature and no apparent pattern can be observed.

PAGE 49

38 It should be noted that the most highly vascularized cornea (Figure 2-17) had an interesting vessel formation. This cornea appeared to contain a jumble of vessels that had no discernable pattern. Vessels were seen criss-crossing throughout the entire area represented by the three-dimensional reconstruction. Additional corneas of interest were those of the fetus (Figures 2-32 and 2-33). A network of vessels was observed in both corneas of the fetus, possibly indicating a developmental process. The vascularization in the fetal corneas resembled that of the other corneas in that the vessels appeared random in nature with no apparent pattern. Additionally, three corneas were examined at a higher magnification of 200 X. The tissue chosen for these examinations included MNW-9016, MSW-9114, and TM-8619 (left) because MNW-9016 and MSW-9114 were the two most highly vascularized corneas and TM-8619 (left) was one of the least vascularized corneas. The area of interest at 200 X was along the junction between the stroma and anterior epithelium of the cornea. On the histologically processed slides, the stroma and anterior epithelium were very similar in appearance, however; red blood cells were observed in the anterior epithelium and not in the stroma. Additionally, the anterior epithelium stained a dark red color and the stroma usually stained a light purple or blue color. The area between the stroma and anterior epithelium was of interest because vessels not only came close to the anterior epithelium via the stroma, but also penetrated into it without any apparent connective tissue elements (see Figures 2-2 through 2-6). Three-dimensional reconstructions were also carried out with these corneas at 200 X in order to examine the smaller, feeder vessels present in the stroma and anterior epithelium. Four quadrants (dorsal, medial, ventral, and lateral) where mapped out for

PAGE 50

39 each cornea and vasculature within each quadrant was examined. A larger number of vessels were observed in the tissues examined at 200 X (Figures 2-34 through 2-41). The figures were arranged in a similar pattern as with those at 25 X. However, instead of being six images representing each cornea from various angles, there are eight images of MNW-9016 (Figures 2-34 and 2-35), eight images of MSW-9114 (Figures 2-36 and 2-37) and sixteen images of TM-8619 (Figures 2-38 through 2-41) because only one quadrant was reconstructed for each cornea at 25 X and two to four quadrants were reconstructed at 200 X. The vasculature at 200 X is similar to that at 25 X in that the vessels do not form patterns and seem randomly distributed throughout the corneal area. However, the most obvious difference between 200 X reconstructions versus 25 X reconstructions is that the vascular patterns at 200 X occupy more of the volume and are far more noticeable. Additionally, vessels could be observed penetrating the anterior epithelium from the stroma with the higher magnification reconstructions. This level of detailed vascularization could not be detected at 25 X. The diameters of blood vessels at 200 X in MNW-9016, MSW-9114, and TM-8619 (L) were also measured. The average vessel diameter for MNW-9016 was 29.3 m, 27.1 m for MSW-9114, and 26.8 m for TM-8619 (L). The overall average for vessel diameters at 200 X was 27.3 m. Morphometry Using the MCID computer program implemented on a M5 system (Imaging Research, Inc.) at the Light Microscopy Facility in the McKnight Brain Institute, vascularization as a percent of the total volume was calculated for each manatee cornea examined. The area of the corneal tissue covered in the three-dimensional reconstructed images at 25 X represented approximately one-fourth of the total corneal area present. Photographing and imaging the entire corneal tissue area at 25 X was not possible due to

PAGE 51

40 screen size limitations. The reconstructed image that was performed in each specimen represented the most vascularized area in the cornea. However, 25 X was selected because this magnification allowed the maximum area to be represented for each specimen while still being able to observe vasculature. The vasculature volumes calculated for each manatee cornea ranged from 0.1% to 1.2 %, with average blood vessel volume coverage of 0.3%. The following values for each manatee cornea, according to its identification number (see Table 2-1), represent the volume of vascularization based on the total area of the cornea examined at 25 X. Additional volumes were calculated for three corneas that were further examined at 200 X (MNW-9016, MSW-9114, and TM-8619 left) in order to better view the vasculature of smaller, feeder vessels between the stroma and anterior epithelium. The following values for each manatee cornea, according to its identification number (see Table 2-2), represent the volume of vascularization based on the total area of the cornea examined in four separate quadrants at 200 X (dorsal, ventral, medial, and lateral or side quadrants if the eye was not labeled left or right). Statistics The statistics were calculated using the SAS computer program by means of a one-way ANOVA. The effect of gender (male versus female), age class (calf, juvenile, or adult), coast found (east versus west), and season found (Spring, Summer, Fall, or Winter) based on blood vessel as a percent of the total volume (which was calculated using the MCIP computer program) of tissue was examined in 20 individuals. The unknown and fetus samples were not used in calculating statistics because no additional information was available for these samples. A one-way ANOVA was used because little variation was seen among the separate factors analyzed. In the case of three individuals

PAGE 52

41 where both eyes were examined, the average between the two eyes was taken and used as the volume value for comparing within the statistics (MSW-9137=0.3%, TM-8619=0.2%, and TM-8630=0.2%). There was no significant gender effect found (F1,18=1.34, P=0.2622); no significant age class effect found (F2,17=0.82, P=0.4588); no significant coast effect found (F1,18=0.07, P=0.7996); and no significant season effect found (F3,16=0.30, P=0.8259). Table 2-1. Information on manatee tissue used and volume of vasculature at 25 X Manatee I.D. number % volume Date found Season found Length (cm) Class Sex County found MNW-9016 0.7 9-28-90 Fall 320 Adult M Sarasota MNW-9022 0.6 11-16-90 Fall 147 Calf F Manatee MNW-9111 0.1 4-8-91 Spring 264 Adult M Dixie MNW-9116 0.1 6-6-91 Summer 210 Calf M Collier MSE-9101 0.5 1-13-91 Winter 335 Adult M Broward MSE-9105 0.6 2-24-91 Winter 323 Adult M Palm Beach MSE-9107 0.3 4-12-91 Spring 232 Calf M Broward MSE-9115 0.5 8-9-91 Summer 292 Adult M Martin MNE-9131 0.1 10-31-91 Fall 348 Adult M Duval MSW-9114 1.2 5-22-91 Spring 340 Adult F Lee MSW-9131 0.1 9-14-91 Fall 292 Adult M Charlotte MSW-9137 0.3 10-31-91 Fall 289 Adult M Okeech-obee MSW-9141 0.3 11-28-91 Fall 252 Juvenile M Lee MSW-9143 0.1 4-17-86 Spring 260 Juvenile M Lee TM-8619 0.2 6-22-86 Summer 219 Calf F Collier TM-8630 0.2 4-6-91 Spring 246 Juvenile F Volusia UCF-9114 0.2 10-26-91 Fall 293 Adult F Brevard UCF-9137 0.4 10-30-91 Fall 220 Calf M Brevard UCF-9138 0.4 12-27-91 Winter 296 Adult F Brevard UCF-9141 0.3 1-13-91 Winter 306 Adult M Brevard Unknown 0.7 NA NA NA NA NA NA Fetus 0.5 NA NA NA NA NA NA

PAGE 53

42 Table 2-2. Volume of vasculature at 200 X Manatee I.D. number Dorsal quadrant Medial quadrant Ventral quadrant Lateral quadrant Side quadrant MNW-9016 5.36% 5% 0.44% NA -MSW-9114 1.28% -1.31% -2.8% TM-8619 4.93% 0.13% 0.15% 4.59% -Discussion The three-dimensional reconstructions of the Florida manatee corneas produced interesting results. In the 26 corneas examined from 22 individuals, vascular beds were present in every cornea. This finding varies little from previous studies conducted by Samuelson et al. (1997 and 1994). The vascularization occurred primarily in the stroma with some extension of vessels into the anterior epithelium. There were no large vessels found in the cornea and vascularization was not heavily extensive throughout. The vascularization that was present was found throughout, but the area where vessels were occurring most commonly varied among individuals and even between eyes from the same animal. The general pattern of vascularization throughout the cornea was highly irregular with criss-crossing vessels that extended through several layers in the stroma and anterior epithelium. The reconstructions of vessels found at 25 X (see Figures 2-8 through 2-33) best resembles that of a spaghetti-like network with no specific design (Figures 2-2 through 2-6). No major differences in density or size were found in the general pattern of vascularization reflecting a morphological pattern with few outliers. The average amount of vascularization present in the corneas examined in this study was 0.3%. Only two animals examined contained appreciably greater amounts of vascularization in their corneas with 1.2% (MSW-9114, Figure 2-17) and 0.7% (MNW-9016, Figure 2-8). This overall pattern in MSW-9114, with it being the highest vascularized cornea examined,

PAGE 54

43 did not differ greatly from the other corneas studied. It only contained more vessels. The only common factor between the two animals with the highest amount of vascularization found was that they were adults. The animals were not found in the same year or season, were different sexes, and were found in different counties. However, the counties where the animals were found were in close proximity to each other; Lee and Sarasota counties are only separated by Charlotte County on the southwest coast of Florida. Interestingly, the third highest amount of vascularization recorded was a calf at 0.6% (MNW-9022), suggesting that age of the animals does not play a critical role in determining the amount of vascularization present in their corneas. However, this calf was found in Manatee County, which is located directly north of Sarasota County. While, this finding may indicate that some environmental influence found in the water in that general area may be playing a role in the higher amounts of vascularization found in these corneas, another individual (MSE-9105) with 0.6% of its cornea vascularized was found in Palm Beach County, which is directly across the state on the southeast coast of Florida. Based on its location, the influence of the environment may not play an important role in determining vascularizing factors after all. Overall, Table 2-1 shows little variation in the percent of vascularization found in the corneas of the manatees examined. Nonetheless, several differences are seen in such factors as date found, age of the animal, size of the animal, and location where the animal was found. Statistically, these differences were not found to be significant. An unexpected aspect of this study was the unique opportunity to study the cornea of a manatee fetus. Reconstructions of the left and right corneas of this individual showed that both contained vascularization, but with varying amounts. This interesting

PAGE 55

44 developmental case allowed us to examine endogenous versus exogenous factors that may affect vascularization. Interestingly, the irregular vascular patterns observed in the other manatees were also observed in the fetus. This observation suggests that the vascularization is stimulated in the animal during development and not through exogenous factors such as the environment, especially when this animal was never exposed to an aquatic environment outside that of its mothers uterus. The findings in the fetus suggest that corneal vascularization in the Florida manatee may be an evolutionary adaptation process. Other areas that could be investigated to further explore this possibility include examining other vascular patterns throughout the body that may be unusual and/or unique to the manatee. It is also possible that some change in organ development occurs that alters vascular patterns. For example, a recently discovered low-density receptor-related lipoprotein called Lrp5 has been found to provide dual function in very different organ systems when there is a defect that occurs in it. The Lrp5 protein functions as a Wnt coreceptor. Wnt proteins help control many developmental processes such as mesoderm induction, cell fate determination, limb patterning, joint formation, and organogenesis. Lrp5 binds to Wnt proteins in osteoblasts and Lrp5 is needed for normal deterioration of embryonic vasculature in the eye. Lrp5 and Wnt proteins work in conjunction to regulate osteoblast proliferation and function, and eye vascularization during late development and after birth. Recent studies have discovered that Lrp5 becomes inactivated in human patients with osteoporosis-pseudoglioma syndrome and is mutated in patients that have high bone mass syndrome. Mice were used to further investigate phenotypes that occurred when the Lrp5 protein was deficient. It was found that they displayed two phenotypes; a low bone mass density caused by a

PAGE 56

45 reduction in bone formation and they retained their hyaloid vasculature in the eye throughout life (Kato et al. 2002). Currently it is not known if manatees have an altered Lrp5 protein. However, further investigations into correlations between the Lrp5 protein and other vascular components may prove to be very insightful. Manatees do have a high bone density so it may be possible that the corneal vascularization occurring is a by product of their high bone density. However, it should be noted that no hyaloid vasculature has been observed in the adult manatee eye. This may rule out the possibility of an Lrp5 deficiency, but there are other aspects of the manatee genome that have not been explored which may be causing the vascularization observed in the cornea. Three corneas were reconstructed at a higher magnification of 200X (Figures 2-34 through 2-41), as opposed to 25X, to demonstrate angiogenic patterns of small blood vessels and their association and penetration between the stroma and anterior epithelium. This was a very interesting find because this penetration of vessels between layers does not usually occur without the apparent presence of connective tissue (see Figures 2-2 through 2-6). By having been divided into quadrants based on its anatomical position, essentially the entire cornea was evaluated. Vascularization was not uniform throughout the quadrants of each individual examined (see Table 2-2). Vascularization was present regionally and varied in location between the three corneas reconstructed. The area covered by vessels present at this magnification was greater than that of vessel coverage at the lower magnification. This was most likely due to the ability to better visualize and trace smaller vessels at the higher magnification. Even though the vessels did cover more area in the cornea at this magnification, they were smaller in size and helped demonstrate the amount of vascularization. These vessels may be too small to interfere with vision.

PAGE 57

46 As with the larger vessels observed at 25X, the smaller vessels at 200X did not have a regular pattern. In the future, it may be of interest to reconstruct all corneas in this study at 200X in order to determine if any repeating patterns do arise. However, the vascular pattern at this magnification currently appears to be random in nature. The most interesting aspect of this part of the study was the presence of the vessel penetration between the stroma and anterior epithelium. To the best of our knowledge, this occurrence has not been documented previously. Our original hypothesis that blood vessels would interfere with vision was not supported because the vessels observed did not appear to be large enough in either size or amount to interfere with vision. Vascular beds throughout the eye occur under normal conditions without interfering with light transmission. One such example is the choriocapillaris. The choriocapillaris vascular structure is different from that observed in the manatee corneal vascular structure in that the choriocapillaris is only a single layer of broad to narrow small vessels (Bhutto and Amemiya 2001; Buggage et al. 1996), whereas the corneal vasculature observed penetrated several layers and continued into adjacent anterior epithelial cells as seen through tangential sections. However, a corrosion cast study conducted on Wistar Kyoto rats investigating the choriocapillaris did find some interesting vascular patterns that are comparable to those found in the manatee cornea. The vascular pattern of the choriocapillaris in these rats viewed from the retinal side near the posterior pole was a nonhomogeneous complex of capillaries that differed in diameter size and was composed of a thick honeycomb pattern of vessels. However, near the peripheral areas of the choriocapillaris, a more regular pattern was formed that was elongated and palm-like in structure. This peripheral palm-like vascular pattern found in

PAGE 58

47 the choriocapillaris terminated at the ora ciliaris retinae (Bhutto and Amemiya 2001). Even though no regular vascular patterns were found in the peripheral sections of the manatee cornea, the irregular patterns found in the peripapillary area of the choriocapillaris appear to be similar to that found in the manatee corneal vascular structure. Irregular vascular patterns in the choriocapillaris have also been found in the rhesus monkey. It was found in these mammals that the posterior pole of the choriocapillaris also had a nonhomogeneous structure consisting of distinct lobular segments (Flower 1993). The choriocapillaris and tapetum lucidum work in conjunction to enhance vision for many species of animals. The vasculature of the tapetum in most domestic species consists of medium-sized vessels in the dorsal section of the choroid. Many vessels pass through the tapetal layer and form the choriocapillaris. The tapetum inserts among branching vessels found in the choroid and the choriocapillaris (Samuelson 1999). The vessels that pass through the tapetum and supply the choriocapillaris in species such as the cat, dog, ferret, and grey seal, do so at right angles to the tapetum. Following this pattern, the blood vessels travel parallel to the incident light, which interferes with the reflective function of the tapetum as little as possible (Braekevelt 1986). This does not seem to be occurring in the vessels found in the manatee cornea. Instead, the vessels are found in any orientation and direction. It is important to note that even though the vascular patterns of the manatee cornea and choriocapillaris of some areas of various species are similar in that they both have irregular patterns; they most likely share more differences than similarities in vascular structure. The choriocapillaris is only a single layer of vessels, whereas the corneal

PAGE 59

48 vascular extends through several layers and even penetrates into adjacent cells. Still, it is important to note that the choriocapillaris does have a relatively high surface area of vessel coverage, which does not interfere with light transmission, thus affecting vision (Buggage et al. 1996). If the vascular patterns in the choriocapillaris are disrupted for any reason, it can result in insufficient removal of waste that is generated by the RPE cells causing an accumulation of additional wastes at Bruchs membrane (Cao et al. 1998). This information regarding the choriocapillaris suggests that not only are the vascular patterns in and around the choriocapillaris important for not interfering with vision in animals with tapeta, but also needed for normal retinal functions to occur. Perhaps the vasculature in the corneas of the manatees observed is more comparable to the vasculature of the retinal vessels. Vascular patterns in the retina can be found in the center and peripheral areas. In the outer retina, vascular demands are met by choroidal circulation. Species of animals that have vascularized inner retinas have a wide range of blood vessel patterns, and these patterns are most often determined by the density distributions of neural cells in the ganglion cell layer. In areas of high cell density, large capillary beds are found. For example, in the domestic cat (Felis domesticus), the squirrel monkey (Saimiri sciureus), and human a high concentration of capillaries are found in the area centralis and fovea where high cell densities are found. Large blood vessels that are found in the retina differ from retinal capillaries in that they will not be found in areas of high cell density. This feature allows visual acuity to be maintained without compromise by allowing light to be transmitted without interference (Dunlop et al. 1997).

PAGE 60

49 Studies using scanning electron microscopy (SEM) investigating vascular patterns in the retina of rats have shown small stem arterioles that are thin in diameter and are nearly straight. These arterioles branch and meet with capillaries that go on to form massive plexuses and sparse vascular networks in the center and periphery of the retina. The capillaries found in the retina are thin in diameter, measuring 3 to 4 m (Ikebe et al. 2001). Additionally, it has been found in rats using vascular corrison casts that side branching of arterioles is the main vascular pattern observed in the arterial tree of the retina. This vascular pattern most likely allows the best flow of materials to the most superficial layers of the retina (Pannarale et al. 1996). The corneal vessels appear to be most similar to retinal vessels in that they both form multiple layers. The vasculature of the retina allows light to pass through with minimal interference, especially towards the center of the retina where large vessels are rarely found under normal conditions. Larger vessels are more commonly found in the peripheral areas of both the retina and cornea where there is minimal impedance of light transmission. The importance of retinal blood vessel structure is especially noticeable when there is some type of abnormality. Normal choroidal vasculature is necessary for normal retinal function and if the vasculature is compromised for any reason, the result can be dysfunction or death of photoreceptors. This can be a sign of a larger problem such as diabetic retinopathy (Cao et al. 1998). Other physiologic changes that may occur in retinal vasculature patterns under abnormal conditions can result in diseases such as glaucoma and age-related macular degeneration. Disturbances most commonly noted with these disorders include a change in ocular pressure and metabolic changes in autoregulation of oxygenated blood delivery to the retina (Yazdanfar et al. 2003).

PAGE 61

50 An additional vascular pattern occurring under normal conditions in reptilian eyes is in the spectacle. The spectacle is a fused eyelid that forms a clear covering over the cornea and is the same size or larger than the cornea. Microsilicone injected into the spectacle found that the structure was vascularized and that the vessels were in a vertical arrangement in some snake species (Mead 1976). The most interesting finding of this study was that the spectacle maintained a vascular pattern and still retained a high level of transparency. These vessels seem to be similar to those found within the manatee cornea in that they are both small and similar in size throughout the area. However, not much work has been conducted on the spectacle of reptilian species, so it is hard to make any definite comparisons between the manatee cornea and reptilian spectacle until more research has been carried out. In the normal, healthy cornea, light passes through unimpeded. The stroma, which makes up the majority of the cornea, has an arrangement of fibrils in such a fashion that does not interfere with light transmission (Freund et al. 1995). However, conditions such as disease, injury, trauma, or clinically induced malformations can change the balance within the cornea leading to inflammation or eventually, more serious problems. Such side effects occur in many species of animals under pathological conditions (Klintworth 1991). In human ophthalmology there is a need for immediate and quick diagnosis and medical aid in response to injury or disease causing corneal vascularization. At the General Eye Service of the Massachusetts Eye and Ear Infirmary, 35 of 845 patients (4.14%) who came in for an eye examination had some form of corneal vascularization. This 4.14% of patients represents 1.4 million Americans with overwhelming infections

PAGE 62

51 that can cause blindness. The primary causes of corneal vascularization in humans result from chlamydial infections (affecting 400 million people worldwide and blinding 6 million), onchocerciasis infections (affecting 50 million people and blinding 1 million), herpes simplex eye infections (affecting 500,000 in the United States), and the wearing of contact lenses (affecting 125,000 to 470,000 people in the United States). This information suggests that corneal vascularization is a major contributor to eye disease in humans (Lee et al. 1998). An additional condition that can occur in humans and cause vascularization in the cornea is homocystinuria. Homocystinuria is an autosomal-recessive disorder of amino acid metabolism that affects several organ systems including the skeletal system, central nervous system, vascular systems, and the eyes. Even though corneal opacities with homocystinuria are not common, 8 out of 90 eyes examined with homocystinuria contained corneal opacities. Eyes that were found to have corneal opacities caused by homocystinuria were found with deep, central scarring and opacification associated with vascularization caused by long-term corneal edema and dislocation of the lens (Rao et al. 2002). Such disorders can most likely be ruled out in the manatee because pathology reports of the necropsied animals examined did not indicate major problems occurring in organ systems caused by genetic problems nor were dislocated lens and edema observed in the manatee cornea during gross and histological examination. Therefore, it is most likely safe to postulate that this genetic disorder or others like it are not responsible for the vascularized corneas found in the Florida manatees examined, especially because research investigating genetic disorders has not found much if any evidence to indicate such problems exist. Currently the only known genetic disorder to occur in manatees is a

PAGE 63

52 defect called congenital ectrodactyly, which is a malformation that is characterized by one or more phalanges missing from one or more digits (Watson and Bonde 1986). However, more research is needed in this area of study to draw any conclusive results suggesting that corneal vascularization in the Florida manatee may be a possible side effect from some other physiological factor. More conclusive results can be drawn from other research carried out on various animal species investigating corneal vascularization and its impact on the surrounding tissue, which indicates that vascularization observed in the manatee cornea is not caused by a pathological condition such as injury, disease or trauma. When injury occurs to the cornea, often the epithelium will contain goblet cells and the stroma becomes vascularized leading to a deficiency in vision. Corneal function will then be interfered with resulting in recurring corneal erosions, decrease in vision caused by uneven corneal surfaces, weakened tensile strength, and ineffectual barrier function (Lee et al. 1998). While not all of these factors could be examined at the histological level with the manatee cornea, no corneal erosions were found. Mice with induced corneal vascularization caused by cauterization were found to have several symptoms not found in the manatee such as edema and opacity occurring in all specimens examined, along with a loss of epithelial and endothelial cells and a large amount of infiltrating neutrophils and macrophages into the stroma (Sonoda et al. 1998). Rats that were used to study inflammation associated with induced corneal vascularization through krypton laser burns demonstrated different histological observations than those found in the manatee. After the laser burns were performed on the rats, epithelial wounding of the limbus was found along with an increase in the

PAGE 64

53 number of cells around the limbal vessels. Additionally, infrequent inflammatory cells were observed and vascularization eventually advanced towards the center of the cornea leading to the vascularization of the entire cornea including the deep stroma (Kvanta et al. 2000). Induced corneal vascularization in rabbits through alkali injury also resulted in trauma not observed in manatee corneas. Damage to all cellular components including the epithelium, keratocytes, and endothelium was observed light microscopically. Edema in the stroma caused it to increase by 35% and on the edge of the stroma, damaged cells and nuclear debris were seen. Towards the peripheral portion of the stroma, fibroblast and inflammatory cells were found. Additionally, the structure of the stroma was disrupted so that lamellae separated and collagen fibrils lost normal parallel arrangement leading to large spaces between the fibrils or fibrils that were almost touching (Huang et al. 2001). Dogs can be afflicted by corneal vascularization through such disorders as spontaneous chronic corneal epithelial defects (SCCED). This defect causes erosions in the cornea that are characterized by sheets of loosely arranged epithelium that can take up to six months to heal. Similar erosions can also be found in humans that take long periods of time to repair and are often associated with afflictions such as traumatic abrasions, epithelial membrane dystrophy, anterior stromal dystrophies, and neurotrophic keratitis. In dogs with SCCED, the basement membrane of the anterior epithelium will detach from the stroma indicating that the extracellular matrix components in these animals may have been previously compromised (Bentley et al. 2001). In the manatee corneas studied, no erosions or detached basement membranes were observed, once again

PAGE 65

54 suggesting that the vascularization found is not an abnormal condition for the Florida manatee. Collectively, signs of inflammation and concurrent side effects found in humans and other mammals have not been observed in the vascularized corneas of the Florida manatee, indicating that the vessels present are not occurring under abnormal conditions. In summary, every Florida manatee cornea examined (n=26) contained vascularization. The vessels observed varied in location, amount, and even between corneas of the same individual. However, no side effects such as edema or erosions were observed in the surrounding tissue indicating that this may not be an abnormal condition in the corneas of Florida manatees. Because vascularization was observed in the fetus, corneal vascularization in the manatee may be a developmental or evolutionary process possibly occurring as a side effect from some other unknown physiological factor. The significance of these findings represent the first study known to us in which corneal vascularization has been found that did not occur because of some type of pathological condition. Additionally, the size of the vessels found in the corneas indicates that they are too small to interfere with light transmission and are most likely not affecting manatees ability to survive in their different aquatic environments.

PAGE 66

55 =100 mm Figure 2-1: Florida manatee eye with ocular tissue and fat removed revealing the eye and optic nerve. The eye is being compared to a dime for size comparison.

PAGE 67

56 A B C Figure 2-2: These histological pictures of MNW-9111 show (A) a long blood vessel in the stroma (blue) that approaches the anterior epithelium (purple) at 100X, (B) blood vessel penetration between the stroma and anterior epithelium at 200X, and (C) close detail of the vessels present in the stroma at 400X.

PAGE 68

57 C B A Figure 2-3: These pictures were taken of MSW-9114 of the same blood vessel at varying magnifications to show the penetration of this vessel into the anterior epithelium (purple): (A) 200X, (B) 400X, and (C) 1000X.

PAGE 69

58 B A Figure 2-4: (A) The arrangement of smaller vessels near the anterior epithelium (purple) from the larger trunk vessel in the stroma (blue) at 200X, specimen MSW-9114. (B) This picture of UCF-9114 shows the smaller vessels (see arrow) that were outlined within the stroma for three-dimensional reconstruction (200X).

PAGE 70

59 B A Figure 2-5: Pictures from MNW-9016 showing (A) a blood vessel passing through the stroma (blue) at 200X, and (B) red blood cells present in the vessel as it penetrates the anterior epithelium (purple) at 1000X.

PAGE 71

60 B A Figure 2-6: These pictures were taken from the unknown specimen. (A) Shows the edge of the limbal vascular arrangement at the peripheral cornea (200X), and (B) larger vessels that go right up to the anterior epithelium at 200X.

PAGE 72

61 D F E C B A Figure 2-7: Anatomical orientation of manatee eyes as they are described in three-dimensional reconstruction: (A) left eye orientation (face-on view), (B) right eye orientation (face-on view), (C) dorsal eye orientation, (D) ventral eye orientation, (E) lateral eye orientation (or horizontal view if left or right is unknown), and (F) medial eye orientation (or horizontal view if left or right is unknown).

PAGE 73

62 Figure 2-8: MNW-9016, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 74

63 E D F C B A Figure 2-9: MNW-9022(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position).

PAGE 75

64 Figure 2-10: MNW-9111, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) medial view of vessels and corneal tissue, and (F) medial view of only the vessels at 25X (arrow indicates anterior position). E D F C A B

PAGE 76

65 Figure 2-11: MNW-9116, (A) face-on on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) horizontal view of vessels and corneal tissue, and (D) horizontal view of only the vessels at 25X. D C B A

PAGE 77

66 Figure 2-12: MSE-9101, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C A B

PAGE 78

67 Figure 2-13: MSE-9105, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) medial view of vessels and corneal tissue, and (F) medial view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 79

68 Figure 2-14: MSE-9107, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 80

69 Figure 2-15: MSE-9115, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 81

70 Figure 2-16: MSE-9131, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 82

71 Figure 2-17: MSW-9114, (A) face-on on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 83

72 Figure 2-18: MSW-9131(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 84

73 Figure 2-19: MSW-9137(L), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 85

74 Figure 2-20: MSW-9137(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 86

75 Figure 2-21: MSW-9141, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 87

76 E D F C B A Figure 2-22: MSW-9143, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position).

PAGE 88

77 Figure 2-23: TM-8619(L), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 89

78 E D F C B A Figure 2-24: TM-8619(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position).

PAGE 90

79 Figure 2-25: TM-8630(L), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) medial view of vessels and corneal tissue, and (F) medial view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 91

80 Figure 2-26: TM-8630(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) medial view of vessels and corneal tissue, and (F) medial view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 92

81 Figure 2-27: UCF-9114, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 93

82 Figure 2-28: UCF-9137, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 94

83 Figure 2-29: UCF-9138, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 95

84 E D F C B A Figure 2-30: UCF-9141, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position).

PAGE 96

85 Figure 2-31: Unknown, (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) horizontal view of vessels and corneal tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 97

86 Figure 2-32: Fetus(L), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 98

87 Figure 2-33: Fetus(R), (A) face-on view of vessels and corneal tissue, (B) face-on view of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F) lateral view of only the vessels at 25X (arrow indicates anterior position). E D F C B A

PAGE 99

88 D C B A Figure 2-34: Dorsal quadrant of MNW-9016 at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from horizontal view (arrow indicates anterior position).

PAGE 100

89 D C B A Figure 2-35: Medial quadrant of MNW-9016 at 200X, (A) vasculature and corneal tissue outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from horizontal view (arrow indicates anterior position).

PAGE 101

90 D C B A Figure 2-36: Dorsal quadrant of MSW-9114 at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from horizontal view (arrow indicates anterior position).

PAGE 102

91 D C B A Figure 2-37: Ventral quadrant of MSW-9114 at 200X, (A) vasculature and corneal tissue outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from horizontal view (arrow indicates anterior position).

PAGE 103

92 D C B A Figure 2-38: Dorsal quadrant of TM-8619(L) at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from medial view (arrow indicates anterior position).

PAGE 104

93 D C B A Figure 2-39: Medial quadrant of TM-8619(L) at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from medial view (arrow indicates anterior position).

PAGE 105

94 D C B A Figure 2-40: Ventral quadrant of TM-8619(L) at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from medial view (arrow indicates anterior position).

PAGE 106

95 D C B A Figure 2-41: Lateral quadrant of TM-8619(L) at 200X, (A) vasculature and corneal outline from face-on view, (B) vasculature only from face-on view, (C) vasculature only from dorsal view, and (D) vasculature only from medial view (arrow indicates anterior position).

PAGE 107

CHAPTER 3 VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR-1 AND -2 IN THE FLORIDA MANATEE CORNEA Introduction Cytokines are very small, nonstructural soluble proteins or glycoproteins made by leukocytes and other cell types. Cytokines act as communicators between cells, and most cytokines help regulate host responses to pathogens, disease, infection, or inflammation (Fitzgerald et al. 2001; Dinarello 2000). Cytokines have often been compared to hormones. However, this is not accurate because highly specialized tissues express hormones and almost every tissue in the body makes cytokines. Hormones are also the primary product produced by the cell, while cytokines account for only a small portion of cell output (Dinarello 2000). Cytokines bind to specific receptors located on the surface of target cells and they display four key features: pleiotropy, redundancy, potency, and act as a network or cascade. Because the receptors of cytokines are expressed on a variety of cell types and their activated signaling pathways increase gene expression, they are pleiotropic. Cytokines are redundant because there are many similarities in the amino acid sequences of receptors. Cytokine receptors have a high affinity to ligands, which makes them potent, act in synchronization, and counter-regulate inhibitory cytokines (Fitzgerald at al. 2001). Environmental influences have the greatest effect on the function of cytokines, being most commonly demonstrated with inflammation. A variety of cytokines and their 96

PAGE 108

97 inhibitors can be detected at the area of inflammation. For example, anti-inflammatory agent anti-TNF, the anticytokine that increases production of IL-1 receptor antagonist (IL-1Ra), and the inhibitory cytokine IL-10 can be found at the same time and result in an outcome that will be determined by this balancing of cytokines. There are cytokines that act as inducers including IL-1, TNF, and IFN. Each of these cytokines can induce cytokine gene expression and are called proinflammatory cytokines. Inhibitors of cytokine gene expression also exist, such as the glucocorticoid hormones and synthetic steroids, which are used as immunosuppressants and anti-inflammatory drugs. Cytokines are also known to play important roles in blood vessel development (Fitzgerald et al. 2001). Angiogenesis is the biological process that produces new blood vessels from existing vasculature (Nicosia and Villaschi 1999; Battegay 1995; Folkman 1995; Schultz and Grant 1991). Angiogenesis occurs in almost every tissue and is a necessary process that is critical in normal conditions such as development, reproduction, healing of wounds, and bone repair (Battegay 1995; Adamis et al. 1994). Angiogenesis can also occur under pathological conditions such as cancer, tumor growth and metastasis, rheumatoid arthritis, trauma, age-related macular degeneration, ischemic heart disease, ischemic peripheral vascular disease, diabetic retinopathy or infection (Rosen 2002; Schultz and Grant 1991). The complex process of angiogenesis is highly regulated by pro-angiogenic and anti-angiogenic growth factors. Several growth factors are very specific for the endothelium (such as vascular endothelial growth factor, VEGF) and others carry out several different functions (such as matrix metalloproteinases, MMPs) (Rosen 2002).

PAGE 109

98 Endothelial cells separate and grow in an avascular tissue, and then unite to form a primitive tubular network during early blood vessel formation. This early network of vessels includes the aorta and major veins. Interconnected branching patterns of vessels form through angiogenic remodeling that are characteristic of mature vasculature. At this stage, endothelial cells join together securely with supporting cells to form mature vessel walls (Yancopoulos et al. 2000). The process of angiogenesis is a series of involved steps. Initially, diseased or injured cells release pro-angiogenic growth factors into surrounding tissue, such as vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), or platelet-derived growth factor (PDGF). Angiogenic growth factors are then released into the surrounding environment and bind to and activate existing endothelial cells of nearby blood vessels. The newly activated endothelial cells signal the nucleus, which initiates the production of enzymes such as MMPs. The extracellular matrix surrounding present blood vessel is broken down by the enzymes, making it possible for endothelial cells to enter the matrix and divide. These endothelial cells then travel towards the growth factor stimulus, which is mediated by integrins (adhesion molecules) at the cell surface. Endothelial cells create new blood vessels by forming hollow tubes. Blood begins to circulate when individual vessels connect and form loops or networks, which can now carry blood to the tissue that originally released the pro-angiogenic growth factors (Rosen 2002). These vessels sprout into previously avascular tissue (Yancopoulos et al. 2000) as a result of endothelial cell migration and proliferation and further develop (Nicosia and Villaschi 1999). This process is referred to as angiogenic sprouting and is responsible for

PAGE 110

99 vascularizing some tissues throughout normal development (i.e., neural tube and retina), as well as in pathological conditions (i.e., tumors) (Yancopoulos et al. 2000). Perhaps the most important determinant in blood vessel formation during angiogenesis is vascular endothelial growth factor, VEGF. It is distinguished for its capacity to cause vascular leakage and permeability, in addition to its capacity to promote vascular endothelial cell production (Yancopoulos et al. 2000). VEGF is a cytokine that is an endothelial cell mitogen and a heparin-binding, dimeric protein (Miller et al. 1994; Klagsbrun and DAmore 1991) and is necessary for the initiation of immature vessel formation during vasculogenesis or angiogenesis (Yancopoulos et al. 2000). VEGF is able to stimulate endothelial migration, proliferation, and proteolytic activity and has a peptide that is signaled through numerous pathways (Dunk and Ahmed 2001). VEGF is made by a variety of cells, including endothelial cells, smooth muscle cells, and fibroblasts (Nicosia and Villaschi 1999). The VEGF gene is composed of eight exons that are separated by seven introns. Alternative splicing of the VEGF gene produces four isoforms, including VEGF121, VEGF189, VEGF206, and VEGF165 (Nicosia and Villaschi 1999). Five VEGF related genes have been discovered (VEGF-B, VEGF-C, VEGF-D, VEGF-E, and placenta growth factor) each producing glycosylated dimers that are secreted after cleavage of their signal peptide (Rosen 2002). VEGF binds to two high affinity tyrosine kinase receptors, VEGF receptor 1 (VEGFR-1/Flt) and VEGF receptor 2 (VEGFR-2/KDR) (Beck and DAmore 2001; Dunk and Ahmed 2001). Both VEGFR-1 and VEGFR-2 bind VEGF with a high affinity and have seven immunoglobulin-like domains, one transmembrane region, and a tyrosine kinase sequence that is interrupted by a kinase

PAGE 111

100 insert domain. Expression of VEGFR-1 and 2 is restricted to the vascular endothelium. It is currently believed that VEGFR-1 and 2 mediate all VEGF effects on vascular endothelium. When VEGF binds to VEGFR-1 and 2, pro-angiogenic signals are transmitted by the receptors to proteins located downstream and a signal cascade is initiated (Rosen 2002). An additional receptor for VEGF is neuropilin-1. It modulates the interaction between VEGF and VEGFR-2 (Ortega et al. 1999). The expression of VEGFR-1 in naive cells mediates cell migration. While both VEGF receptors are tyrosine phosphorylated, the phosphorylation of VEGFR-1 is not required for function in vivo. The Src family of non-receptor tyrosine kinases comprises several highly homologous kinases, including Fyn and Yes (Bevilagua et al. 2003), and are phosphorylated in response to VEGF/VEGFR-1 but not to VEGF/VEGFR-2 interactions (Ortega et al. 1999). In contrast, phosphorylated VEGFR-2 links with Shc (adaptors in tyrosine phosphorylation signal transduction pathways) (Lee et al. 2004), Grb2 (domain-containing adapter proteins) (Schulze and Mann 2003), and Nck (a protein required for translation and mediating signaling from the plasma membrane) (Kebache et al. 2003), but also with phosphatases of SHP-1 (a protein-tyrosine phosphatase that is a negative regulator of multiple signal transduction pathways) (Frank et al. 2003) and SHP-2 (a protein-tyrosine phosphatase that plays an important role in intracellular signaling brought forth by growth factors, hormones, and cytokines) (Tartaglia et al. 2004), and initiates the phosphorylation of signaling molecules such as mitogen-activated protein kinases (MAPKs), which transmit signals from the membrane to the nucleus in response to growth factors and cellular stress (McMullen et al. 2003) through the stimulation of Raf, a protein that responds to growth factor signals (Weber et al. 2003). VEGF has also

PAGE 112

101 been found to induce the tyrosine phosphorylation and recruitment of focal adhesion kinase (Ortega et al. 1999). Additionally, it has been shown that VEGFR-1 mediates monocyte migration leading to the induction of nitric oxide (NO) production (Dunk and Ahmed 2001). Angiogenesis can occur in endothelial cells during such events as tumor progression, diabetic retinopathy, and rheumatoid arthritis. Local angiogenesis may be due to the release of soluble mediators by the involved tissues. At this point there is a switch in the phenotype of the quiescent endothelial cell to an activated phenotype. The endothelial cells are then able to respond to mitogenic signals (Ortega et al. 1999). There is a larger increase in the upregulation of genes modulated by VEGFR-2, than with VEGFR-1. These induced genes include growth factors, protease inhibitors, receptors, transcription factors, and actin-binding proteins (Yang et al. 2002). At this point, mitogenic growth factors are released and these change the activated phenotype to an angiogenic phenotype. Interactions between angiogenic growth factors and their receptors provide the signals for cell migration, proliferation, and differentiation to form new vessels (Ortega et al. 1999). Corneal angiogenesis is often used as a model to investigate other types of vascularization. The causes of corneal vascularization are varied and in humans can include infections (such as trachoma and herpes), immunologic processes, alkali burns, toxic and nutritional deficiencies, and the wearing of contact lenses. In experimental settings, corneal vascularization can be induced through chemical, microbiological, and physical injuries, nutritional deficiencies, and immunological reactions (Klintworth 1991). Rabbits have most often been used to study experimentally induced corneal

PAGE 113

102 vascularization. After an injury of some type has occurred in the cornea, cell migration and mitosis are involved in healing of the epithelial corneal surface. If abrasion occurs, cell migration takes place almost immediately. In order to return the epithelium to its normal visual thickness, cell division follows. Cell migration can heal small injuries within 24 hours. If a larger portion of the anterior epithelium is injured, the limbus is the main source for production of epithelial stem cells. Endothelial cells are not able to respond as quickly to cell loss. Endothelial cells can produce edema, which results in corneal disease (Samuelson 1999). The stromal edema that occurs with corneal vascularization aids in the growth of blood vessels in the cornea and is usually an inflammatory process. There is an increase in vascular permeability due to the inflammatory response as a result of endothelial cell contractions between endothelial cell junctions. There are many inflammatory mediators that are involved in this process including histamine, serotonin, ADP, prostaglandins, many cytokines, free radicals, and immune aggregates. However, several factors can help reverse the process including epinephrine, cortisone, and some drugs such as antihistamines and glucocorticoids (Klintworth 1991). Injuries to the cornea result in stromal edema with neutrophils moving to the source of the injury to release enzymes and chemotaxic agents. Anteriorly, epithelial wound healing begins within one hour of the injury and wing layer cells flatten and slide next to the injury site in order to cover the area within 24 to 96 hours. The epithelium returns to its full thickness by the action of basal stem cells of the limbus. Production of the stromal fibroblast takes place after 24 hours (Ahmadi and Jakobiec 2002).

PAGE 114

103 Leukocytes move in the extravascular space within 24 hours and move through the corneal stroma to the site of injury. The first new vessels appear after 27 hours. Advancing new vessels remain permeable to fluid, plasma proteins, and smaller proteins. Forming corneal blood vessels do not have firm intercellular complexes, whereas well formed corneal vessels in advanced stages of wound healing do not leak. Once new vessels have developed, the basal lamina breaks down, allowing the endothelial cells to move into the extracellular matrix. Progression of these endothelial cells at the end of the developing new blood vessels extends toward the damaged area (Klintworth 1991). The movement of endothelial cells, which elongate and form solid sprouts developing lumen, follows this. Pairs of sprouts form loops and blood flows, and the process is repeated from the top of each loop (Pepose and Ubels 1992). New blood vessels can emerge within 33 hours after injury. A network of vessels is formed that becomes linked and these vessels then begin to pass materials between the new connections (Klintworth 1991). Experimental induction of corneal neovascularization has been used for years to describe vascularization induced by tumors, inflammatory and growth factors, and corneal injuries (Ueda et al. 1997). There are two commonly used bioassays to measure angiogenesis, the developing chick chorioallantoic membrane (CAM) and implantation in the rabbit cornea. Angiogenic materials are modified during normal development of vascular patterns in the CAM assay and with cornea implantation, the growth of blood vessels is induced from existing normal vessels (Gaudric et al. 1992). Many animal models demonstrate that cytokines and angiogenic factors are involved in induced ocular neovascularization. The most commonly used animals are

PAGE 115

104 non-human primates and small mammals such as rabbits and rats. VEGF has been shown to increase vascular permeability and angiogenesis in the iris of cynomolgus monkeys (Maccaca fascicularis). Induced ocular angiogenesis through laser photocoagulation of retinal veins in cynomolgus monkeys showed a temporal association with changing aqueous VEGF levels and iris neovascularization. No VEGF was detected prior to photocoagulation, and afterwards levels of VEGF increased dramatically. These findings showed a relationship between VEGF (a secreted angiogenic factor) and increased development of ocular neovascularization (Miller et al. 1994). Not only has VEGF been found to increase with iris vascularization, but also with retinal neovascularization. Retinal ischemia produced by retinal vein occlusion in monkeys produces a large increase of VEGF in the vitreous and anterior chamber of the eye (Folkman 1995). Additional studies of the angiogenic cytokine factor VEGF, involved in induced corneal vascularization, have shown interesting results. VEGF has been demonstrated to advance a variety of steps in angiogenesis including proliferation, migration, proteolytic activity, and capillary tube development of endothelial cells in human corneas. In normal corneas, it was found that corneal endothelial cells and vascular endothelial cells of limbal vessels made VEGF. In inflamed corneas, VEGF expression was much higher in corneal endothelial cells, keratocytes, and vascular endothelial cells of new vessels. These findings indicate that VEGF is expressed at higher levels in inflamed and vascularized human corneas than in normal corneas and suggest that VEGF is involved in the pathogenesis of corneal vascularization (Philipp et al. 2000). Differences seen in VEGF levels before and after corneal injury have also been described in male Sprague-Dawley rats. Induced corneal vascularization occurred in these rats by implantation of

PAGE 116

105 anti-VEGF antibody or non-immune goat IgG pellets. Levels of VEGF in the rat corneas increased 10-fold after being wounded. This information indicates that VEGF is needed for wound and inflammation-related corneal vascularization (Amano et al. 1998). Corneal vascularization induced in rats by cautery of silver nitrate shows inflammation and vascularization that can be easily studied in order to investigate inflammatory gene expression and the role of leukocytes in inflammatory angiogenesis. Silver nitrate cautery of the rat cornea resulted in blood vessel growth from the limbus to the lesion. Using the corneal micropocket assay, it was found that basic fibroblast growth factor (bFGF) and VEGF did induce angiogenesis from the limbus to the implant. Basic FGF was released after injury to promote healing and VEGF production was increased subsequently. It was also found that leukocytes associated with inflammation were the primary source of VEGF in the rat cornea after cautery. During inflammation, cytokines helped regulate the expression of VEGF (Edelman et al. 1999). Other cytokines besides VEGF have been shown to affect corneal angiogenesis because a balance is needed between angiogenic factors and inhibitors (Yoshida et al. 1998). CAM assays were used to describe the effects of developing vascular patterns of rabbits induced with corneal vascularization by implanted pellets containing bFGF. Blood vessel growth was observed over progressive days after the implantation of the pellet. However, inflammation in the cornea did not stimulate angiogenesis (as it might with VEGF), but vascularization was induced by bFGF (DAmato 1994). Interleukin (IL)-8 appears to induce angiogenic factors in induced ocular vascularization. IL-8 is the major regulator of a variety of inflammatory and proliferative response genes and is

PAGE 117

106 found at higher levels in human patients with vascularization than without (Yoshida et al. 1998). An interesting case of corneal vascularization was originally discovered by Samuelson et al. (1997 and 1994) in which blood vessels were consistently present in the anterior portion of the corneas in the Florida manatee. However, no signs of infection, injury or other pathological conditions could be detected in the corneas. The reason for the presence and function of the blood vessels is currently unknown. The objective of this study is to investigate the presence or absence of vascular endothelial growth factor receptors-1 and -2 in the corneas of the Florida manatee, Trichechus manatus latirostris, through the use of immunohistochemistry. This will determine what role, if any, VEGFR-1 and VEGFR-2 have in the angiogenic process occurring in the blood vessels found in the cornea of these animals. Materials and Methods Specimen Collection Florida manatee eyes were collected from Sea World of Florida in Orlando, Florida within 24 hours after death. Each eye was labeled with an identification number given to it by Sea World. Information including gender, size, and time when the animal died was recorded. Manatee eyes from specimen TM-0310 (juvenile animal) were collected on April 16, 2003 and eyes from specimen TM-0311 (fetal animal) were collected on April 23, 2003. Manatee eyes were placed in fresh 10% neutral buffered formalin solution (Luna 1968). Preparation of Eyes All extraneous tissue surrounding the globe was removed using a scalpel. The corneas were then removed from the globe by cutting 1 mm from the limbus into the

PAGE 118

107 sclera with a sharp razor. A notch was cut into the limbus at 12:00 in the dorsal section of each cornea in order to know orientation during histological examination. Photographs were then taken of the corneas (see Figure 3-1). The remainder of the eye was stored and the cornea was placed in a tissue cassette, labeled and then returned to the 10% neutral buffered formalin solution for a total of 24 hours. The tissue was changed to phosphate buffered saline solution (PBS) for 12 hours and then processed. Histology In preparation for embedding, the corneas were dehydrated by placing them in ascending concentrations of ethyl alcohol for an hour each: 70%, 80%, two changes of 95%, followed by three changes of 100%. The tissues were then cleared with two changes of xylene and infiltrated with two changes of Fisher Tissue Prep at 57 Celsius. The tissues were then embedded using Fisher Tissue Prep T565 in metal molds and allowed to harden. Using a Reichert-Jung 2030 microtome, the embedded corneas were cut in serial dorsal/ventral sections at a thickness of 5 microns. The tissue ribbons were floated out on a 60 Celsius water bath and placed on Fisherbrand Superfrost microscope slides (3 x 1 x 1) with one section per slide. Each slide was labeled with the identification number of the animal, order of section, and orientation of tissue. Twenty sections per eye where cut. After cutting was completed, the slides were placed in a slide rack and into an oven at 60 Celsius overnight to dry. Immunohistochemistry Sections were deparaffinized with three changes of xylene for five minutes each, two changes of 100% ethanol alcohol, 95% ethanol alcohol, and distilled water for two minutes each. The slides were rehydrated with Tris buffered saline (TBS) solution

PAGE 119

108 (Humason 1972). The sections were then quenched in 3% hydrogen peroxide for 20 minutes. TBS was used to wash the slides three times for two minutes each. Blocking was performed with 5% normal goat serum (53 L of normal goat serum and 1000 L of Dako antibody diluent, no. S0809) (DakoCytomation, Carpinteria, California) for 20 minutes. The slides were then washed three times in TBS for two minutes each. A primary antibody (Flt-1, polyclonal rabbit, catalog number SC316 or KDR, monoclonal mouse, no. SC251, both from Santa Cruz Biotechnology, Santa Cruz, California) was combined with the Dako antibody diluent at different concentrations of 1:50, 1:100, and 1:200. The Flt-1, polyclonal rabbit primary antibody was used for detection of VEGFR-1 and the KDR, monoclonal mouse primary antibody was used for VEGFR-2 detection. The sections were incubated in a humidified chamber overnight at 10 Celsius. The staining procedure was completed by using a detection kit from Dako (no. K0673). After the sections were stained and rinsed twice with distilled water, twice with 95% ethanol alcohol, and three times with xylene, they were coverslipped using Richard-Allan Scientific mounting medium (Richard-Allan Scientific, Kalamazoo, Michigan). Results Vascular endothelial growth factor receptor-1 (VEGFR-1) Both the juvenile (TM-0310) (see Figures 3-2, 3-3, and 3-4) and fetal (TM-0311) (see Figures 3-6 and 3-7) animals had a positive reaction for VEGFR-1 in their corneal tissue. The control samples (or negative control) had no positive reaction occur in them (sees Figures 3-5 and 3-8). TM-0310 reacted positively in both corneas (left and right) at 1:50 and 1:100 concentrations of polyclonal rabbit primary antibody in Dako antibody diluent. TM-0311 also showed positive reactions in both corneas at concentrations of 1:50 and 1:100, however; the positive reaction was more evident in the 1:100

PAGE 120

109 concentration (Figure 3-6). Even though positive reactions for VEGFR-1 could be seen in both individuals, the reaction was more evident in the juvenile manatee (TM-0310). Preparation and fixation of corneal tissue with both animals was carried out in the same way, so the difference seen in the degree of positive reaction between the animals was most likely due to individual variation. An adult manatee cornea was not examined in this study due to the difficulty in obtaining samples within 24 hours post mortem. However, it would be beneficial to look at an adult cornea in the future. With both the juvenile and fetal samples, the positive reaction was most apparent in the stromal vessels and anterior epithelial vessels (Figures 3-3 and 3-6). Interestingly, no positive reaction was observed in the limbal vessels of these animals (Figure 3-7). Only a very slight positive reaction could be seen in the larger limbal vessels of the juvenile animal, but it was minimal. It is possible that this may be an age characteristic. Again, it would be beneficial to have an adult cornea for examination to investigate VEGFR-1 expression as a function of age. Interestingly, in the juvenile sample (TM-0310), some larger vessels had no positive reaction, while other larger vessels did (Figure 3-3). It is possible that the larger vessels that did not exhibit a positive reaction were more established vessels and smaller vessels were forming from these that were positive for VEGFR-1. Once again, it would be beneficial to examine an adult cornea in order to understand what is occurring with these larger vessels and if they are positive for VEGFR-1 or not. Vascular endothelial growth factor receptor-2 (VEGFR-2) With both the juvenile and fetal samples, no positive staining of VEGFR-2 was observed in any of the immunohistological procedures carried out. The concentrations at 1:50, 1:100, and 1:200 of monoclonal mouse primary antibody to Dako antibody diluent

PAGE 121

110 produced no reaction at all and the results were no different from that of the control samples (negative control). The reaction was absent with respect to vessels in the cornea, limbus, or conjunctiva. The experiment investigating VEGFR-2 was then carried out using a positive control to check for possible defects in the monoclonal mouse primary antibody or other substances used. Corneal tissue from a pitbull (canine) with glaucoma and vascularization in the stroma and corneal tissue from a horse with keratitis and vascularization throughout the entire cornea were used as positive controls. When the immunohistochemical procedures were completed, these specimens did have a positive reaction in their stromal vessels. This finding indicated that the procedure and substances were not defective in some way. The experiment was then carried out a second time on both of the manatee samples and again showed no positive reaction for VEGFR-2 (Figure 3-9). Discussion Of the two samples examined (TM-0310 and TM-0311), both had positive reactions for VEGFR-1 in the corneal tissue at 1:50 and 1:100 concentrations of polyclonal rabbit primary antibody in Dako antibody diluent. The juvenile sample (TM-0310) contained more positive reactions than did the fetal sample (TM-0311). This may be due to individual or age differences. To determine which, if not both are the case, additional immunohistochemical procedures need to be carried out on more samples from varying age groups. The difficulty with this goal, however, is obtaining samples from manatees in different age groups within 24 hours after their death. In order for the immunohistochemistry to be valid and work properly, the tissue samples must be prepared and fixed immediately after death. It may be possible that an age pattern in

PAGE 122

111 manatees could be discovered based on positive reactions of VEGFR-1 in corneal vessels. If more positive reactions of VEGFR-1 are found in the cornea as the animal ages, then some scale may be developed based on those results. Combined with the size of their Descemets membrane and anterior lens capsule, it may be possible to determine approximate age groups of manatees. Descemets membrane, the basement membrane of the corneal endothelium, increases in thickness consistently over time with other mammals and the anterior capsule of the lens, another basement membrane, also increases in thickness with age. Even though these two ocular structures were not examined in this study, it may be a possible path for investigation in the future for age determination in the manatee. With VEGFR-1, while positive reactions were found in the corneal tissue of both samples including the stromal and anterior epithelial vessels and in the conjunctiva, no comparable reactions of VEGFR-1 could be found in the limbal vessels. Even though we had the advantage of examining the corneal, limbal, and associated conjunctival vessels, which collectively comprise the entire vascular bed for the anterior-most portion of the eye of the manatee and may be working in conjunction with each other to carry out undetermined vascular functions within this region of the eye, we did not focus on the conjunctiva. Since the conjunctiva and cornea are in such close proximity and can carry out associated functions, it may be important to look at the conjunctiva more closely in the future to determine whether these vessels play a similar role as the corneal vessels, and what role that might be. By examining this entire vascular bed making up the corneal, limbal, and conjunctival vessels, it may be inferred that this section of the anterior globe is working together to facilitate some unknown component.

PAGE 123

112 Neither the juvenile nor fetal sample examined demonstrated any positive reaction with VEGFR-2 at any concentration of monoclonal mouse primary antibody. While the immunohistochemical procedures carried out examining VEGFR-2 appeared to be no different from the negative control samples, we are confident that the procedures were carried out correctly because positive controls that were used did result in positive reactions of VEGFR-2 in corneal vessels. These positive controls both suffered from pathological problems resulting in vascularization in their corneas. Still, there is a possibility that the antibody for VEGFR-2 was not sensitive to the manatee. Once again, additional tissue would be very interesting to test in the future, especially samples that vary in age. If the VEGFR-2 immunohistochemical findings for the manatee are correct, the presence or lack of these VEGF receptors is a significant finding. Since the two receptor tyrosine kinases of vascular endothelial growth factor (VEGF) are VEGFR-1 and VEGFR-2 and they start the signaling pathways during vasculogenesis and pathological angiogenesis (Meyer and Rahimi 2003), the action of VEGF is mainly restricted to endothelial cells to induce such responses as cell proliferation, migration, survival, differentiation, and permeability to macromolecules (Kojima-Yuasa et al. 2003; Meyer and Rahimi 2003). VEGF is able to stimulate endothelial cells by binding to and activating VEGFR-1/Flt-1 and VEGFR-2/FLK-1 (Meyer and Rahimi 2003). When VEGF binds to its receptors, dimerization and activation of the inherent kinase results. The VEGF receptors will then autophosphorylate and ensuing signal transduction then occurs (Kojima-Yuasa et al. 2003). Therefore, the effects of VEGF are highly mediated

PAGE 124

113 by VEGFR-1 and VEGFR-2 (Suzuma et al. 1998) and it may be possible that all the effects of VEGF are mediated by these receptors (Rosen 2002). Even though all the specific roles of VEGFR-1 and 2 have not been defined in the current literature and are still being investigated, it has been suggested that VEGFR-1 has several roles including signal transduction, activation of unique signaling proteins found in endothelial cells (Meyer and Rahimi 2003), and cell migration (Ortega et al. 1999) and can be expressed in both endothelial cells and pericytes (Suzuma et al. 1998). Additionally, VEGFR-1 is thought to be responsible for chemotaxis (McLeod et al. 2002), induction of nitric oxide (NO) production in trophoblasts, and stimulation of matrix metalloproteinase expression in vascular smooth muscle cells (Dunk and Ahmed 2001). VEGFR-2 is thought to be responsible for migration of endothelial cells (Meyer and Rahimi 2003; Ortega et al. 1999) and VEGFR-2 is expressed in microvascular endothelial cells (Suzuma et al. 1998). It also is thought that VEGFR-2 is the primary signal transduction pathway controlling angiogenesis (Rosen 2002) and may be responsible for the initiation of mitogenesis (McLeod et al. 2002). Both receptors are very different in their ability to go through autophosphorylation and kinase activation in that VEGFR-2 has the ability to undergo ligand-dependent tyrosine phosphorylation and kinase activation and VEGFR-1 cannot (Meyer and Rahimi 2003). Based on this information and VEGFR-2 not being found in the manatee corneas examined, we postulate that the process of angiogenesis occurring in this tissue is a non-pathological and non-inflammatory response to currently unknown factors, and most likely due to developmental or evolutionary influences.

PAGE 125

114 Interestingly, it has been shown that VEGF is expressed in normal endothelial and epithelial cells in the cornea, throughout the epithelium of the cornea, and by vascular endothelial cells in limbal vessels that have no pathological or experimentally induced conditions occurring in them, suggesting that VEGF may have a maintenance role in the cornea (Philipp et al. 2002; Kvanta et al. 2000). This additional evidence further suggests that the receptors for VEGF found in the manatee corneas examined are occurring there normally and not under conditions brought on by such factors as trauma, injury, or infection. The presence of the blood vessels in the corneas of all manatees examined thus far still remains enigmatic. The normal cornea in most mammalian species is a transparent avascular layer of the eye that is able to refract and transmit light, as well as provide support for the contents within the globe. The cornea is able to transmit light clearly in part because it lacks blood vessels, has a keratinized surface epithelium, and is pigmented (Muller et al. 2003). The avascularity found in the cornea is maintained through such components as the major keratan sulfate proteoglycans (including lumican, keratocan, and mimecan) of the stroma and interfibrillar proteins (including collagens type VI and XII) (Kurpakus et al. 1999). The arrangement of collagen and proteoglycans in the stroma allows light to pass without restriction through the extracellular matrix aiding in corneal transparency (Twining et al. 1999). It has also been thought that additional proteoglycans in the aqueous humor may help maintain corneal clarity. The aqueous humor acts as a circulatory system in the eye by clearing it of harmful agents and waste products from the surrounding environment. The replacement of aqueous humor connected with heparan sulfate bind to and prevents growth factors from potentially

PAGE 126

115 causing vascularity within the cornea (Fannon et al. 2003). Other factors such as transketolase and aldehyde dehydrogenase 3 located in corneal cells are also important in maintaining corneal clarity because these cells are able to respond quickly to injuries or infections (Twining et al. 1999). It is possible that there is an abnormality occurring with the proteoglycans in the manatee cornea or even within the corneal cells themselves. This could help explain why these vessels are found. There are other factors to consider that may be causing this rare phenomenon. There is the possibility that an undiscovered genetic adaptation may be occurring in the manatee. For example, a recently discovered low-density receptor-related lipoprotein called Lrp5 has been found to provide dual function in very different organ systems when there is a defect. The Lrp5 protein functions as a Wnt coreceptor for Wnt proteins, which help control many developmental processes such as mesoderm induction, cell fate determination, limb patterning, joint formation, and organogenesis. Specifically, Lrp5 binds to Wnt proteins in osteoblasts and Lrp5 is needed for normal deterioration of embryonic vasculature in the eye. Lrp5 and Wnt proteins work in conjunction to regulate osteoblast proliferation and function, and eye vascularization during late ocular development. Recent studies have discovered that Lrp5 becomes inactivated in human patients with osteoporosis-pseudoglioma syndrome and is mutated in patients that have high bone mass syndrome. Mice were used to further investigate phenotypes that occurred when the Lrp5 protein was deficient. It was found that they displayed two phenotypes; a low bone mass density caused by a reduction in bone formation and the retention of their hyaloid vasculature in the eye throughout life (Kato et al. 2002). Currently, it is not known if manatees have a deficient Lrp5 protein or some other

PAGE 127

116 comparable genetic mutation. However, further investigations into correlations between the Lrp5 protein and other vascular components may prove to be very insightful. It should be noted that no hyaloid vasculature has been observed in the adult manatee eye. This may rule out the possibility of an Lrp5 deficiency, but there are other aspects of the manatee genome that have not been explored which may be causing the vascularization observed in the cornea. Corneal vascularization in the Florida manatee may be caused by some other factor such as ocular hypoxia. Angiogenesis often takes place when there is cell replication, inflammation and tissue repair, and tissue hypoxia. When such an event occurs and vascularization follows, large amounts of cells in the same region compete for oxygen and existing nutrients. Metabolites accumulate, which can change the cellular makeup and even cause some cells to release factors (such as cytokines) that stimulate vascular growth and development. In some scenerios in which tissue hypoxia occurs which results in angiogenesis, low oxygen tension is thought to be a major contributing factor in vascularization. Additionally, the corneal epithelium relies on atmospheric oxygen for anaerobic metabolism of carbohydrates to provide energy needs. If some barrier or other obstacle blocks this oxygen supply from occurring, corneal vascularization often results (Klintworth 1991). Ocular hypoxia is also known to alter gene expression. For example, a recently discovered protein called inhibitory PAS (Per/Arnt/Sim-IPAS) is linked with low levels of VEGF expression in the cornea during hypoxic conditions. When an IPAS antisense oligonucleotide was used in a mouse cornea to produce angiogenesis under normal oxygen conditions, stimulation of hypoxia-dependent VEGF gene expression took place.

PAGE 128

117 This finding showed that IPAS negatively regulates angiogenesis and produces an avascular phenotype (Makino et al. 2001). It could be possible that some type of hypoxic conditions may be taking place in the corneas of manatees. Perhaps some of the fresh water areas where the manatees are found may be too slow moving and even depleted of oxygen, therefore preventing proper oxygen exchange from occurring between the water and the anterior portion of the cornea. If there is a lack of oxygen, this could lead to hypoxic conditions where blood vessels may develop. This may also explain why some vessels have been observed developing in the conjunctiva of the manatee eye. In the future, oxygen exchange rates or possible oxygen depletion could be tested by investigating oxygen levels found in some of the waters throughout the state of Florida (salt, fresh, and brackish waters alike). However, this is an unlikely explanation because vascularization was observed in all manatee specimens examined and not just a few from a certain area where low oxygen levels are suspected. A more likely explanation for the vascularization observed may be caused by the unusually thick tear film found in the manatee and other marine mammals such as the bottlenose dolphin (Dawson, personal communication). Perhaps the vascular bed found in the manatee cornea is an adaptation for this aquatic mammal that allows it to have better control in dealing with changes in osmotic conditions as it travels from fresh to salt to brackish waters and back again. Whatever the cause for this vascularization found in the cornea of manatees, it definitely is a unique and interesting characteristic that requires further investigation to better understand the mechanisms causing such an occurrence, be it environmental, genetic, or evolutionary. Learning more about this condition in a non-inflammatory and non-pathological setting

PAGE 129

118 could possibly lead to a better understanding of how to treat and provide increased care for other species that suffer from corneal vascularization associated with trauma, injury, or infection. C B A Figure 3-1: Florida manatee eye (A) showing the left cornea only and all other ocular tissue removed, (B) showing the right cornea only with all other ocular tissue removed, and (C) ocular tissue and fat removed revealing the eye and optic nerve compared to a dime for size comparison.

PAGE 130

119 A B Figure 3-2: Pictures taken from TM-0310 showing a positive reaction of VEGFR-1 at (A) 20X with a 1:50 concentration of polyclonal rabbit primary antibody in Dako antibody diluent nicely indicating where the vessels are present and (B) 100X at 1:50 showing location of basal cells which are positive for VEGFR-1.

PAGE 131

120 C B D A Figure 3-3: VEGFR-1 positive reactions in TM-0310 (A) showing the prevalence of the positive reaction at 200X with 1:50, (B) a larger vessel at 400X with 1:50 most likely more established with no positive reaction in it (as seen in smaller vessels) but can see positive reaction in anterior epithelium along basal layer, (C) a larger vessel at 200X with 1:100 that did have a positive reaction unlike in B, and (D) a branching vessel at 400X with 1:100 showing a positive reaction.

PAGE 132

121 C B A Figure 3-4: VEGFR-1 positive reactions in TM-0310 (A) at 400X with 1:100 showing a prevalent positive reaction around the large vessel indicating that growth may still be occurring, (B) 400X with 1:100 showing good localization of positive reaction, and (C) 200X with 1:100 of a good overview with more vessels forming around already existing vessels, possibly indicating the process of angiogenesis taking place.

PAGE 133

122 B A Figure 3-5: VEGFR-1 reactions in TM-0310 of control samples (A) at 20X and (B) 400X showing no positive reaction in either sample.

PAGE 134

123 D C B A Figure 3-6: Pictures taken from TM-0311 showing a positive reaction of VEFGR-1 at (A) 20X with 1:100 concentration, indicating a good overview but not as strong of a positive reaction on periphery as with TM-0310, (B) 100X with 1:100, indicating a positive reaction but still not as strong as with TM-0310, (C) 200X with 1:100, indicating a positive reaction in larger vessels, and (D) 200X with 1:100, also indicating a positive reaction.

PAGE 135

124 C B D A Figure 3-7: VEGFR-1 positive reactions in TM-0311 (A) at 400X with 1:100, showing a positive reaction that is similar to those seen in TM-0310, (B) 200X with 1:100, showing no reaction in limbal vessels, indicating that the vessels are most likely already developed and are not forming any new vessels, (C) 400X with 1:50, showing several vessels but not as strong of a positive reaction a seen in TM-0310, indicating individual differences in vessel development, and (D) 400X with 1:50, with no reaction at all in the limbus (colored cells are pigment cells).

PAGE 136

125 B A Figure 3-8: VEGFR-1 reactions in TM-0311 of control samples (A) at 200X and (B) at 400X, showing no positive reactions, but showing the pigmentary patterns of melanocytes in the limbus.

PAGE 137

126 C B A Figure 3-9: Pictures taken of TM-0310 showing no positive reactions of VEGFR-2 that appears to be different from the control samples at (A) 20X with a 1:100 concentration of monoclonal mouse antibody to Dako antibody diluent, (B) 100X with 1:100, and (C) 200X with 1:100.

PAGE 138

CHAPTER 4 ADDITIONAL STUDIES Corneal Vascularization in the Antillean Manatee Since corneal vascularization was found in every specimen of the Florida manatees examined, investigation additional non-Florida manatees could determine if similar or even the same unusual occurrence was taking place. Robert K. Bonde of the United States Geological Service/Sirenia Project located in Gainesville, Florida was able to collect an eye from an Antillean manatee that was found stranded in Puerto Rico on July 21, 2003. The eye was immediately placed in 10% neutral buffered formalin after necropsy and upon receiving the manatee eye on August 25, 2003, it was examined and photographed (see Figure 4-1). The Antillean manatee eye was very similar in size and shape to that of the Florida manatee. However, this cornea was much more opaque and cloudy in appearance than any observed with the Florida manatee specimens. After photographs of the cornea were taken, it was then processed histologically using the same methods as with the Florida manatee corneas (see Chapter 2). The histological findings of this cornea were very interesting. It too was vascularized, but much more extensively than that observed in the Florida manatee corneas (see Figures 4-2 through 4-6). The vessels appeared to be arranged randomly as with the Florida manatee, but showed a much greater density, having been observed throughout the entire corneal tissue. Once again, vessels were observed extending from the stroma penetrating the anterior epithelium (see Figures 4-5 and 4-6). Furthermore, the stromal collagen was arranged in a wavy pattern that was not observed in the Florida 127

PAGE 139

128 manatee (see Figure 4-3). Unusual projections were also seen in the posterior stroma and endothelium that were not seen in the Florida manatee (see Figure 4-8). It is currently not known what function these projections have and whether these projections are found in all Antillean manatees. An additional feature that was observed in the Antillean manatee cornea and not in the Florida manatee cornea was a light yellow layer on the outside of the tissue surrounding the anterior epithelium that looked similar to adipose tissue (see Figure 4-7). Additional slides were stained with PAS to detect the presence of fungi that might have caused such an occurrence. The stain indicated that the sections were PAS negative for fungi (see Figure 4-9). Currently, it is believed that this layer is mucus and may be more predominant in the Antillean manatee than the Florida manatee because of the primarily salt water habitat in which the Antillean manatee lives. At this point it is not possible to speculate whether morphological features seen in this individual occur in all Antillean manatee corneas. It would be very interesting and informative to examine additional corneas from Antillean manatees and perform three-dimensional reconstructions, which were not carried out on this specimen. It would be interesting to determine if any patterns do exist and how much actual blood vessel coverage there is. The presence of corneal vascularization in an Antillean manatee helps further supports the premise that this is a normal process taking place in these animals especially in the absence of any signs of inflammation, infection, or injury. Ophthalmic Examinations in Two Captive Florida Manatees Two captive Florida manatees, Hugh and Buffett, at Mote Marine Laboratory in Sarasota, Florida had been trained by Dr. Gordon Bauer of New College and Debi Colbert and Joe Gaspard of Mote Marine Laboratory to receive ophthalmic examinations.

PAGE 140

129 The animals were trained to place their heads on a platform located directly next to the concrete edge of their enclosure approximately six inches above the waters surface. They were also trained to keep their eyes open for up to a duration of thirty seconds while remaining stationary. Dr. Chris Murphy from University of Wisconsins College of Veterinary Medicine performed streak retinoscopy, infrared photorefraction, and keratometry on the manatees eyes while they had their heads stationed on the platform. Additionally, refraction was carried out under water through the use of streak retinoscopy. It was found that both manatees had poor reflexes that were most likely caused by their vascularized corneas and ophthalmic lesions. Small central cataracts were observed in both eyes of Buffett, whereas Hugh seemed to have a rather clear visual axis in his left eye. It was noted that Hughs right eye did have an axial opacity. Underwater examinations to determine reflexes could not be obtained because of obstructions within the manatees enclosure (i.e., plexiglass). Even though the ophthalmic examinations did not take place under ideal conditions, the best images were obtained with Buffett, both above and below the water. It appeared that underwater and above water Buffett was emmetropic and sometimes slightly hyperopic by approximately one diopter. These findings suggest that the manatees examined are able to change their focus by a very small amount in both the air and water rapidly.

PAGE 141

130 Figure 4-1: The cornea from an Antillean manatee eye. Notice the cornea is cloudy.

PAGE 142

131 Figure 4-2: This corneal section was cut at 10 microns and stained using Massons Trichrome stain. The red tissue is the anterior epithelium and the blue to purple is stroma. This corneal tissue from the Antillean manatee was taken at 25X.

PAGE 143

132 Figure 4-3: These pictures show the unusually nonorganized, wavy stroma observed in the Antillean manatee cornea at (A) 25X, (B) 100X, and (C) 200X. C B A

PAGE 144

133 B A Figure 4-4: Extensive vasculature was observed in the (A) stroma of the Antillean manatee cornea at 100X and (B) as the stroma approaches the anterior epithelium at 100X.

PAGE 145

134 Figure 4-5: These pictures illustrate the vessels in the anterior epithelium at (A) 200X, (B) 400X, and (C) capillaries in the anterior epithelium at 400X. C B A

PAGE 146

135 Figure 4-6: Vasculature can be observed (A) beneath the anterior epithelium at 200X and (B) penetrating the anterior epithelium at 400X. B A

PAGE 147

136 A Figure 4-7: The unusual material thought to be mucus surrounding the anterior epithelium at (A) 200X and (B) 400X. B

PAGE 148

137 A Figure 4-8: Unusual projections were observed in the posterior stroma and adjacent endothelium at (A) 25X and (B) 100X. B

PAGE 149

138 A Figure 4-9: A PAS stain of the Antillean manatee cornea showing no reaction to a fungus in the anterior epithelium at (A) 100X and (B) 200X. B

PAGE 150

139 CHAPTER 5 GENERAL CONCLUSIONS All of the manatee corneas examined in th is study (26 Florida manatee corneas and one Antillean manatee cornea) were vascul arized. The vessels observed varied in location, amount, and even between corneas of the same individual. However, no pathological signs such as edema or erosi ons were observed in the surrounding tissue, indicating that this may not be an abnormal condition in the co rneas of Florida manatees. Similar characteristics were observed among the corneas, including irregular vessel patterns, vessel penetration between the stro ma and anterior epithelium, and no signs of injury or trauma in the tissue. The significan ce of these findings is th at they represent the first study known to us in which corneal vasc ularization has been f ound that did not occur because of an underlying abnorma l or pathological condition. The blood vessel coverage expressed as a percentage in the Florida manatees ranged from 0.1% to 1.2%. This level of va scularization most likely does not interfere with visual functions at a point that would cau se significant visual loss. Additionally, the size of the vessels found in the corneas indicates that they are too sm all to interfere with light transmission and are most likely not aff ecting manatees ability to survive in their different aquatic environments. Varying factors including gender, age, season, and location along the coast where the animal was found were exam ined statistically based on the percentage of blood vessel coverage in the corneas and were not found to be significant, indicating that the vascularization occurs most likely at similar rates and frequencies among the manatees sampled.

PAGE 151

140 An unexpected aspect to this study was th e unique opportunity to study the cornea of a manatee fetus. Surprising ly, the irregular corneal vascul ar patterns observed in the other manatees were also observed in the fetus. This finding suggests that the vascularization is stimulated in the animal during development and not through exogenous factors such as the environment, especially when this animal was never exposed to an aquatic environment outside that of its mothers uterus. Because vascularization was observed in the fetal speci men, this lead us to believe that corneal vascularization in the manatee may be a deve lopmental and evolutionary process. In the immunohistochemical procedures that were car ried out on two manatee samples investigating the presence or absence of vascular endothelial growth receptors (VEGFR-1 and VEGFR-2), VEGFR-1 was f ound in both samples of corneal tissue, whereas VEGFR-2 was not. This information re-enforces the postula tion that the process of angiogenesis occurring in this tissue is a non-pathol ogical and non-inflammatory response to currently unknown f actors, being most likely developmental or evolutionary influences. Interestingly, VEGF is expresse d in normal endothelial and epithelial cells in the cornea, throughout the epithelium of the co rnea, and by vascular endothelial cells in limbal vessels that have no pathological or experimentally induced conditions occurring in them suggesting that VEGF may have a maintenance role in the cornea. This additional evidence further suggests that th e receptors for VEGF found in the manatee corneas examined are occurring there norma lly and not under conditions brought on by such factors as trauma, injury, or infection. Perhaps the vascular bed found in the ma natee cornea is an adaptation for this aquatic mammal that allows it to compensate fo r control changes in os motic conditions as

PAGE 152

141 it travels from fresh to salt to brackish waters and back again. Whatever the cause for this vascularization found in the cornea of manatees, it definitely is a unique and interesting characteristic th at requires further investig ation to better understand the mechanisms causing such an occurrence, be it environmental, genetic, or evolutionary. Learning more about this condition in a noninflammatory and non-pathological setting could possibly lead to a better understanding of how to treat and pr ovide increased care for other species that suffer from corneal vasc ularization associated with trauma, injury, or infection.

PAGE 153

142 LIST OF REFERENCES Adamis, A.P., J.W. Miller, M.T. Bernal, D. J. DAmico, J. Folkman, T. Yeo, and K. Yeo. Increased vascular endothelial growth factor levels in the vitreous of eyeswith proliferative diabetic retinopathy. Am erican Journal of Ophthalmology, 118: 445450. Ahmadi, A.J. and F. A. Jakobiec. 2002. Corneal wound healing: Cytokines and extracellular matrix protei ns. International Ophthalm ology Clinics, OcularTrauma, 42 (3); 13-22. Agrawal, V.B. and R.J. Tsai. 2003. Corneal epithelial wound heali ng. Indian JournalOf Ophthalmology, 51 (1): 5-15. Alm, A. 1992. Ocular Circ ulation. Pages 198-227. In W. Hart Jr., editor. Adlers Physiology of the Eye, Ninth Edition. Mosby-Year Books, Inc., St. Louis, Missouri. Amano, S., R. Rohan, M. Kuroki, M. Tolentino, and A.P. Adamis. 1998. Requirementfor vascular endothelial grow th factor in woundand inflammationrelated corneal ne ovascularization. I nvestigative Ophthalmology and Visual Research,39 (1): 18-22. Bachteler, D. and G. Dehnhardt. 1999. Active touch performance in the Antillean manatee: Evidence for a functional differe ntiation of facial tactile hairs. Zoology,102: 61-69. Battegay, E.J. 1995. Angiogenesis: mechanis tic insights, neovascular diseases, and therapeutic prospects. Journal of Molecula r Medicine, 73: 333-346. Bauer, G.B., D.E. Colbert, and W. Fellner 1999. Underwater visual acuity of the Florida manatee, Trichechus manatus latirostris Paper presented at the 13th Biennial Conference on the Biology of Marine Mammals. Maui, Hawaii, November 28December 3. Beck, C. and and N. Barros. 1991. The imp act of debris on the Florida manatee. MarinePollution Bull etin, 22: 508-510. Beck, L. and P.A. DAmore. 1997. Vasc ular development: cellular and molecular regulation. FASEB Journal, 11: 365-373.

PAGE 154

143 Bentley, E., G.A. Abrams, D. Covitz, C.S. Cook, C.A. Fischer, D. Hacker, C.M. Stuhr,T.W. Reid, and C.J. Murphy. 2001. Morphology and immunohistochemistry of spontaneous chronic corneal epithelial defects (SCCED) in dogs. Investigative Ophthalmology and Visual Science, 42 (10): 2262-2269. Bevilaqua, L.R., J.I. Rossato, J.H. Medina, I. Izquierdo, and M. Cammarota. 2003. Src kinase activity is required for a voidance memory and formation and recall.Behavioral Pharmo cology, 14 (8): 649-652. Bhutto, I.A. and T. Amemiya. 2001. Microvascular archit ecture of the rat choroid:corrosion cast study. The Anatomical Record, 264: 63-71. Bossart, G. 2001. Manatees. Pages 939-960. In L. Dierauf and F. Gulland, editors.Marine Mammal Medicine. CRC Press, Boca Raton, Florida. Bossart, G., D. Baden, R. Ewing, B. Roberts, and S. Wright. 1998. Brevotoxicosis in manatees ( Trichechus manatus latirostris ) from the 1996 epizootic: Gross,histologic, and immunohistochemi cal features. Toxicologic Pathology, 26:276-282. Bossart, G.D., R.A. Meisner, S.A. Ro mmel, S. Ghim, and A.B. Jenson. 2002. Pathological features of the Florida ma natee cold stress syndrome. Aquatic Mammals, 29 (1): 9-17. Braekevelt, C.R. 1983. Fine structure of the choriocapillaris, Bruchs membrane, andretinal epithelium in the sheep. Anatomy and Embryology, 166 (3): 415-425. Braekevelt, C.R. 1986. Fine structure of the tapetum cellulosum of the grey seal ( Halichoerus grypus ). Acta Anatomica, 127: 81-87. Braekevelt, C.R. 1986b. Re tinal epithelial fine stru cture in the grey seal ( Halichoerus grypus ). Acta Anatomica, 127: 255-261. Brecht, M., B. Preilowski, and M.M. Merzen ich. 1997. Functional architecture of the mystacial vibrissae. Behaviour al Brain Research, 84: 81-97. Buck, R.C. 1983. Ultrastructu ral characteristics associat ed with the anchoring of corneal epithelium in several classes of vertebrates. Journal of Anatomy, 137: 743756. Buggage, R.R., E. Torczynski, and H. E. Grossniklaus. 1996. Choroid and suprachoroid.Pages 1-33. In W. Tasmna and E.A. Jaeger, editors. Duanes Flundations of Clinical Ophthalmology, Vo lume 1. Lippincott-Raven Publishers, Philadelphia. Cao, J., S. McLeod, C.A. Merges, and G.A. Lutty. 1998. Choriocapillaris degenerationand related pathologic cha nges in human diabetic eyes. Arch Ophthalmology, 116: 589-597.

PAGE 155

144 Cohen, J.L., G.S. Tucker, and D.K. Odell. 1982. The photoreceptors of the West Indianmanatee. Journal of Morphology, 173: 197-202. Colbert, D.E. and G.B. Bauer. 1999. Basic husbandry training of two Florida manatees, Trichechus manatus Soundings, 24 (3): 18-21. Collin, S.P. and H.B. Collin. 2000. A comp arative SEM study of the vertebrate cornealepithelium. Cornea, 19 (2): 218-230. DAmato, R.J., M.S. Loughnan, E. Flynn, and J. Folkman. 1994. Thalidomide is an inhibitor of angiogenesis. Proceedings of the National Academy of Sciences of the United States of America, 91 (9): 4082-4085. Dawson, W.W., J.P. Schroeder, and S.N. Sh arpe. 1987. Corneal surface properties of two marine mammal species. Marine Mammal Science, 3 (2): 186-197. Dehnhardt, G., B. Mauck, W. Hanke, and H. Bleckmann. 2001. Hydrodynamic trailfollowing in Harbor seals ( Phoca vitulina ). Science, 293 (5527): 102-104. Dinarello, C.A. 2000. Impact of basic research on tomorrows medicine. Chest, 118 (2):503-508. Drohan, L., C.E. Colby, M.E. Brindle, S. Sanislo, and R.L. Ariagno. 2002. Candida (Amphotericin-Sensitive) lens abscess associated with decreasing arterial blood flow in a very low birth weight preterm infant. Pe diatrics, 110 (5): 1-4. Duignan, P., C. House, M. Walsh, T. Campbe ll, G. Bossart, N. Duffy, P. Fernandes, B.Rima, S. Wright, and J. Geraci. 1995. Conservation and Management of MarineMammals. Smithsonian Institution Press, Washington. Dunk, C. and A. Ahmed. 2001. Vascular endothelial growth factor receptor-2mediatedmitogenesis is nega tively regulated by vascular endothelial growth factor receptor-1 in tumor epithelia l cells. American Journal of Pathology, 158 (1): 265273. Dunlop, S.A., S.R. Moore, and L.D. Beazley. 1997. Changing patterns of vasculature in the developing amphibian retina. The J ournal of Experimental Biology, 200: 24792492. Ebara, S., K. Kumamoto, T. Matsuura, J.E. Mazurkiewicz, and F.L. Rice. 2002. Similarities and differences in the innervat ion of mystacial vibr issal follicle-sinus complexes in the rat and cat: A confocal microscopic study. The Journal ofComparative Neurology, 499: 103-119. Edelman. J.L., M.R. Castro, and Y. Wen. 1999. Correlation of VEGF expression byleukocytes with the growth and re gression of blood vessels in the rat cornea.Investigative Ophthalmology and Visual Research, 40 (6): 1112-1123.

PAGE 156

145 Fannon, M., K. Forsten-Williams, C.J. Dow d, D.A. Freedman, J. Folkman, and M.A. Nugent. 2003. Binding inhibition of angiogenic factors by heparan sulfateproteoglycans in aque ous humor: potential mechanism for maintenance of an avascular cornea. FASEB Journal, 17: 902-904. Fitzgerald, K.A., L.A.J. ONeill, A.J.H. Gearing, and R.E. Callard. 2001. The CytokineFacts Book, Second Edition. Acad emic Press, San Diego. 515 pages. Flower, R.W. 1993. Extraction of c horiocapillaris hemodynamic data from ICGfluorescence angiograms. Investigative Ophthalmology and Visual Science, 34 (9): 2720-2729. Folkman, J. 1995. Angiogenesis in cancer, va scular, rheumatoid and other disease. Nature Medicine, 1 (1): 27-31. Frank, C., C. Burkhardt, D. Imhof, J. Ri ngel, O. Zschornig, K. Wieligmann, M. Zacharias, and F.D. Bohmer. 2003. Eff ective dephoshorylation of Src substrates by SHP-1. Journal of Biological Chemistry. Freund, D.E., R.L. McCally, R.A. Farrell, S.M. Cristol, N.L. LHernault, and H.F.Edelhauser. 1995. Ultrastructure in an terior and posterior stroma of perfused human and rabbit corneas. Relation to tr ansparency. Investigative Ophthalmology and Visual Science, 36 (8): 1508-1523. Gallivan, G., R. Best, and J. Kanwisher. 1983. Temperature regulation in the Amazonian manatee, Trichechus inunguis Physiological Zoology, 56: 255-262. Gaudric, A., T. Nguyen, M. Moenner, A. Glacet-Bernard, and D. Barritault. 1992. Quantification of angiogenesis due to basic fibroblast growth fa ctor in a modified rabbit corneal model. Ophthalmic Research, 24: 181-188. Gerstein, E.R. 2002. Manatees, bioacoustics and boats. American Scientist, 90: 154163. Gerstein, E.R., L. Gerstein, S.E. Forsyt he, and J.E. Blue. 1999. The underwater audiogram of the West Indian manatee ( Trichechus manatus ). Journal of Acoustic Society of America, 105 (6): 3575-3583. Griebel, U. and A. Schmid. 1996. Color vision in the Manatee ( Trichechus manatus ).Vision Research, 36 (17): 2747-2757. Griebel, U. and A. Schmid. 1997. Bright ness discrimination ability in the West Indianmanatee ( Trichechus manatus ). The Journal of Experimental Biology, 200: 1587-1592. Hartman, D. 1979. Ecology and behavior of the manatee ( Trichechus manatus ) in Florida. Special publication number 5, American Society of Mammalogists.

PAGE 157

146 Hayashi, S., T. Osawa, and K. Tohyama. 2002. Comparative obser vations on corneas, with special reference to Bowmans layer and Descemets membrane in mammals and amphibians. Journal of Morphology, 254 (3): 247-258. Hogan, M.J., J.A. Alvarado, and J.E. Weddell. 1983. Histology of the human eye. WBSaunders, Philadelphia. Huang, Y., K.M. Meek, M. Ho, and C.A. Paterson. 2001. Analysis of birefringenceduring wound healing and re modeling following alkali burns in rabbit cornea.Experimental Eye Research, 73: 521-532. Humason, G.L. 1972. Animal Tissue Techniques, Third Edition. W.H. Freemanand Company, San Francisco. 276 pages. Ikebe, T., T. Shimada, K. Ina, H. Kitamura, and K. Nakatsuka. 2001. The threedimensional architecture of retinal blood ve ssels in KK mice, with special reference to the smooth muscle cells and pericytes. Journal of Electron Microscopy, 50 (2): 125-132. Irvine, A. 1983. Manatee metabolism and its influence on distribution in Florida. Biological Conservation, 25: 315-334. Joyce, N.C. 2003. Proliferative capacity of th e corneal endothelium. Progress in Retinal and Eye Research, 22: 359-389. Kastelein, R.A., P. Bunskoek, M. Hagedoor n, W.W. Au, and D. de Haan. 2002b. Audiogram of a ha rbor porpoise ( Phocoena phocoena ) measured with narrow-band frequency-modulated signals. Journal of Acoustic Soci ety of America, 112 (1): 334-344. Kastelein, R.A., M. Hagedoorn, W.W. Au, a nd D. de Haan. 2003. Audiogram of a striped dolphin ( Stenella coeruleoalba ). Journal of Acoustic Society of America, 113 (2): 1130-1137. Kastelein, R.A., P. Mosterd, B. van Sa nten, M. Hagedoorn, and D. de Haan. 2002a.Underwater audiogram of a Pacific walrus ( Odobenus rosmarus divergens ) measured with narrow-band frequency-m odulated signals. Journal of Acoustic Society of America, 112 (1): 2173-2182. Kato, M., M.S. Patel, R. Le vasseur, I. Lobov, B.H.J. Chang, D.A. Glass II, C. Hartmann, L.Li, T. Hwang, C.F. Brayton, R.A. La ng, G. Karenty, and L. Chan. 2002.Cbfa1independent decrease in osteoblas t proliferation, osteopenia, and persistentembryonic eye vascularizati on in mice deficient in Lrp5, a Wnt coreceptor. Journal of Cell Biology, 157 (2): 303-314. Kebache, S., E. Cardin, D.T. Nguyen, E. Chevet, and L. Larose. 2003. Nck-1 antagonizes endoplasmic reticulum stress-indu ced inhibition of tr anslation. Journal of Biological Chemistry.

PAGE 158

147 Ketten, D.R. 1991. The marine mammal ear: Specializations for aquatic audition and echolocation. Pages 717-754. In D. Webster, R. Fay, and A. Popper, editors. The Evolutionary Biology of Hearing. Springer-Verlag, New York. Kinkel, M.D., J.G.M. Thewissen, and H.A. Oelschlager. 2001. Rotation of middle earossicles during cetacean developmen t. Journal of Morphology, 249: 126-131. Klagsbrun, M. and P.A. DAmore. 1991. Re gulators of angiogenesis. Annual Review of Physiology, 53: 217-239. Klintworth, G.K. 1991. Corneal Angiogenesi s: A Comprehensive Clinical Review. Pringer-Verlag Press, New York. 135 pages. Kojima-Yuasa, A., J.J. Jua, D.O. Kennedy, and I. Matsui-Yuasa. 2003. Green tea extractinhibits angiogenesis of human umbilical vein endothelial cells through reductionof expression of VEGF recep tors. Life Sciences, 73: 1299-1313. Kurpakus, W.M., K.A. Kernacki, and L.D. Hazlett. 1999. Corneal cell proteins and ocular surface pathology. Biotechnol ogy and Histochemistry, 74 (3): 146-159. Kvanta, A., S. Sarman, P. Fagerholm, S. Seregard, and B. Steen. 2000. Expression of matrix metalloproteinases-2 (MMP-2) and vascular endothelial growth factor(VEGF) in inflammation-associ ated corneal neovascularization. ExperimentalEye Research, 70: 419-428. Laing, R.A., M.M. Sanstrom, A.R. Berrospi, and H.M. Leibowitz. 1976. Changes in the corneal endothelium as a func tion of age. Experimental Eye Research, 22 (6): 587594. Lee, M.S., T. Igawa, S.J. Chen, D. Van Bemmel, J.S. Lin, F.F. Lin, S.L. Johansson,J.K. Christman, and M.K. Lin. 2004. p66S hc protein is upregulated by steroidhormones in hormone-sensitive can cer cells and in primary prostate carcinomas. International Journa l of Cancer, 108 (5): 672-678. Lee, P., C.C. Wang, and A.P. Adamis. 1998. Ocular neovascularization: Anepidemiologic review. Survey of Ophthalmology, 43 (3): 245-269. Lefebvre, L. and T. OShea. 1996. Florida manatees. National Bi ological Service: OurLiving Resources-Coastal and Marine Ecosystems, 267-269. Levin, M.J. and C.J. Pfeiffer. 2002. Gross and microscopic observations on the lingual Structure of the Florida manatee, Trichechus manatus latirostris Anatomy, Histology, and Embryology, 31 (5): 278-285. Los, L.I., M.J.A. Van Luyn, P.S. Eggli, F. Dijk, and P. Nieuwenhuis. 2000. Vascularremnants in the rabbit vitr eous body II. Enzyme digestion and immunohisto-chemical studies. Experi mental Eye Research, 71: 153-165.

PAGE 159

148 Luna, L.G. 1968. Manual of Histologic Stai ning Methods of the Armed Forces Institute of Pathology, Third Edition. McGraw-Hill Book Company, New York. 258 pages. Mackay-Sim, A., D. Duvall, and B.M. Gr aves. 1985. The West Indian manatee ( Trichechus manatus ) lacks a vomeronasal organ. Brain, Behavior, and Evolution, 27 (2-4): 186-194. Madsen, C.J. and L.M. Herman. 1980. Soci al and ecological corr elates of cetacean Vision and visual appearance. Pages 101-147. In L.M. Herman, editor.Cetacean Behavior: Mechanisms and Functions. John Wiley and Sons, New York. Makino, Y., R. Cao, K. Svensson, G. Bertil sson, M. Asman, H. Tanaka, Y. Cao, A. Berkenstam, and L. Poellinger. 2001. Inhi bitory PAS domain pr otein is a negative regulator of hypoxia-inducible gene expression. Nature, 414: 550-554. Mallo, M., H. Schrewe, J.F. Martin, E.N. Olson, and S. Ohnemus. 2000. Assembling afunctional tympanic membrane: Signals from the external acoustic meatus coordinate development of the malleal manubrium. Development, 127: 4127-4136. Marshall, C., L. Clark, and R. Reep. 1998a. The muscular hydrostat of the Floridamanatee ( Trichechus manatus latirostris ): A functional morphological modelof perioral bristle use. Ma rine Mammal Science, 14 (2): 290-303. Marshall, C., G. Huth, V. Edmonds, D. Hali n, and R. Reep. 1998b. Prehensile use of perioral bristles during feeding and associ ated behaviors of the Florida manatee ( Trichechus manatus latirostris ). Marine Mammal Science, 14 (2): 274-289. Marshall, C., P. Kubilis, G. Huth, V. Ed monds, D. Halin, and R. Reep. 2000. Foodhandling ability and feeding-cycle length of manatees feeding on several speciesof aquatic plants. Journal of Mammalogy, 81 (3): 649-658. Mass, A.M., D.K. Odell, D.R. Ketten, and A.Y. Supin. 1997. Ganglion layer topographyand retinal resolution of the Caribbean manatee, Trichechus manatus. DokladyBiological Sciences, 355: 392-394. Mass, A.M. and A.Y. Supin. 1995. Ganglion ce ll topography of the re tina in the bottlenosed dolphin, Tursiops truncates Brain, Behavior, a nd Evolution, 45 (5): 257265. McMullen, M., R. Keller, M. Sussman, and K. Pumiglia. 2003. Vascular endothelialGrowth factor-mediated activ ation of p38 is dependent upon Src and RAFTK/Pyk2. Oncogene. McLeod, D.S., M. Taomoto, J. Cao, Z. Zhu, L. Witte, and G.A. Lutty. 2002. Localization of VEGF receptor-2 (KDR/Flk -1) and effects of blocking it in oxygeninduced retinopathy. Inves tigative Ophthalmology and Vi sual Science,43 (2): 474482.

PAGE 160

149 Mead, A.W. 1976. Vascularity in the reptil ian spectacle. Inves tigative Ophthalmology, 15 (7): 587-591. Meyer, R.D. and N. Rahimi. 2003. Compara tive structure-function analysis of VEGFR1 and VEGFR-2. Annals of the New York Academy of Sciences,995: 200-207. Miles, S.A. 2002. Angioarchitecture of th e Ciliary Body in Four Cetacean Species. Masters Thesis, University of Fl orida. Gainesville, Florida. Miller, J.W., A.P. Adamis, D.T. Shima, P.A. DAmore, R.S. Moulton, M.S. OReilly, J.Folkman, H.F. Dvorak, L.F. Brown, B. Berse, T. Yeo, and K. Yeo. 1994. Vascular endothelial growth factor/vascular permeability factor is temporally and spatially correlated with ocular angiogenes is in a primate model. AmericanJournal of Pathology, 145 (3): 574-584. Morrison, J.C., M.P. DeFrank, and E.M. Van Buskirk. 1987. Comparative microvascular anatomy of mammalian cilia ry processes. Investigative Ophthalmologyand Visual Science, 28 (8): 1325-1340. Morrison, J.C. and E.M. Van Buskirk. 1984. Ciliary process micr ovasculature of the primate eye. American Journa l of Ophthalmology, 97 (3): 372-383. Muller, L.J., C.F. Marfurt, F. Kruse, and T. M.T. Tervo. 2003. Corneal nerves: Structure, contents, and function. Experi mental Eye Research, 76: 521-542. Murphy, C.J., R.W. Bellhorn, T. Williams, M.S. Burns, F. Schaeffel, and H.C. Howland. Refractive state, ocular anatomy, and accommodative range of the sea otter ( Enhydra lutris ). Vision Research, 30 (1): 23-32. Natiello, M. 1999. The ciliar y body and its vasculature in the West Indian manatee asdetermined by histological and threedimensional reconstruction. Masters thesis.University of Florida. Gainesville, Florida. Nicosia, R.F. and S. Villaschi. 1999. Au toregulation of angiogenesis by cells of the vessel wall. Pages 1-43. In Kwang W. Jeon, editor. Intern ational Review of Cytology, A Survey of Cell Biology. Academic Press, London. Ninomiya, H. and H. Kuno. 2001. Microva sculature of the rat eye: scanning electronmicroscopy of vascular corrosion casts. Veterinary Ophthalmology, 4: 5559. Nummela, S., T. Wagar, S. Hemila, and T. Reuter. 1999. Scaling of the cetacean middleear. Hearing Research, 133: 71-81. Ortega, N., H. Hutchings, and J. Plou et. 1999. Signal relays in the VEGF system.Frontiers in Bioscience, 4: 141-152.

PAGE 161

150 Pannarale, L., P. Onori, M. Ripani, and E. Gaudio. 1996. Precapillary patterns and perivascular cells in the retinal microvasc ulature. A scanning electron microscope study. Journal of Anatomy, 188 (3): 693-703. Peichl, L., G. Behrmann, and R.H.H. Kroger. 2001. For whales and seals the ocean is not blue: A visual pigment loss in marine mammals. European JournalNeuroscience, 13: 1520-1528. Pepose, J. and J. Ubels. 1992. The cornea. Chapter 2. Pages 29-70. In W. Hart Jr.,editor. Adlers Physiology of the Ey e, Ninth Edition. Mosby-Year Books, Inc., St. Louis, Missouri. Philipp, W., L. Speicher, and C. Humpel. 2000. Expression of va scular endothelial growth factor and its receptors in inflamed and vascularized human corneas.Investigative Ophthalmology and Visual Research, 41 (9): 2514-2522. Piggins, D., W.R.A. Muntz, and R.C. Best. 1983. Physical and mo rphological aspectsof the eye of the manatee Trichechus inunguis NATTERER 1883: (Sirenia:mammalia). Mar. Behav. Physiology, 9: 111-130. Provancha, J. and C. Hall. 1991. Observations of associations be tween seagrass bedsand manatees in east central Florida. Florida Scientist, 54: 87-98. Rasmussen, L.E.L. and B.L. Munger. 1996. The sensorineural specializations of the trunk tip (finger) of the Asian elephant, Elephas maximus The AnatomicalRecord, 246: 127-134. Rao, S.K., S. Krishnakumar, R.R. Sudhir, K.N. Sulochana, and P. Padmanabhan. 2002.Bilateral corneal fibrosis in homocystinuria. Cornea, 21 (7): 730-732. Reep, R., C. Marshall, and M. Stoll. 2002. Tactile hairs on the postcranial body inFlorida manatees: A mammalian lateral line? Brain, Behavi or, and Evolution, 59: 141-154. Reep, R., C. Marshall, M. Sto ll, and D. Whitaker. 1998. Di stribution and innervation of facial bristles and hairs in the Florida manatee ( Trichechus manatus latirostris ).Marine Mammal Science, 14 (2): 257-273. Reep, R., M. Stoll, C. Marshall, B. Homer, and D. Samuelson. 2001. Microanatomy of facial vibrissae in the Florida manatee: The basis for specialized sensory function and oripulation. Brain, Behavi or, and Evolution, 58: 1-14. Reynolds, J. 1999. Efforts to conserve the manatees. Pages 267-295. In J. Twiss and R.Reeves, editors. Conservation and Management of Marine Mammals. Smithsonian Institution Press, Washington. Reynolds, J. and D. Odell. 1991. Manat ees and Dugongs. Facts on File, Inc., New York.

PAGE 162

151 Ridgway, S.H., D.A. Carder, T. kamolnic k, R.R. Smith, C.E. Schlundt, and W.R. Elsberry. 2001. Hearing and whistling in the deep sea: Depth influences whistleSpectra but does not attenua te hearing by white whale ( Delphinapterus leucas ) (Odontoceti, Cetacea). The Journal of Experi mental Biology, 204: 38293841. Rosen, L.S. 2002. Clinical experience with angiogenesis signaling inhibitors: Focus on vascular endothelial growth factor (VEGF) blockers. Cancer Control, 9(2): 36-44. Ruskell, G. 1998. The retinopial vein: A vein passing directly from the retina to thepia mater at the optic nerve head. Britis h Journal of Ophthalmology, 82: 495-497. Saari, J.C. 1992. The biochemistry of sensory transduction in vertebrate photoreceptors.Pages 460-484. In W.M. Hart Jr., editor. Adlers Physiology of the Eye, NinthEdition. Mosby-Year Book, Inc., St. Louis, Missouri. Samuelson, D. 1999. Ophthalmic anatomy, ch apter 1. Pages 31-150. In Kirk N. Gelatt, editor. Veterinary Ophthalmology. Lippinc ott Williams and Wilkin s, Philadelphia. Samuelson, D.A., P.A. Lewis, and M. Pinkwa sser. 1997. Corneal vascularization in the West Indian manatee ( Trichechus manatus ). Investigative Ophthalmology and Visual Science (ARVO Supplement), 38: S517. Samuelson, D.A., R.L. Reep, P.A. Lewis, and W.M. Chisholm. 1994. The ocular anatomy of the West Indian manatee. First International Manatee and DugongConference, 1: 33-34. Gainesville, Florida. SAS/STAT. Guide for Personal Computers, Version 6th edition, SAS Institute, Inc.,Cary. Schultz, G.S. and M.B. Grant. 1991. Neova scular growth factors. Eye, 5: 170-180. Schulze, W.X. and M. Mann. 2003. A novel proteomic screen for peptideproteininteractions. Journa l of Biological Chemistry. Sivak, J.G. 1977. The role of the spectacle in the visual optics of the snake eye. VisionResearch, 17: 293-298. Sonoda, K., T. Sakamoto, H. Yoshikawa, S. Ashizuka, Y. Ohshima, K. Kishihara, K.Nomoto, Y. Ishibashi, and H. Inom ata. 1998. Inhibition of corneal inflammationby the topical use of Ras farn esyltransferase inhi bitors: Selective inhibitionof macrophage localization. Investigative Ophthalmology and Visual Science,39 (12): 2245-2251. Steinle, J.J., D. Krizsan-Agbas, and P.G. Smith. 2000. Regional regulation of choroidal blood flow by autonomic innervation in the rat. American Journal of Physiological, Regulatory, and Integrat ive Comparative Physiology, 279: 202-209.

PAGE 163

152 Strek, W., P. Strek, M. Nowogrodzka-Zagor ska, J.A. Litwin, K. Pitynski, and A.J.Miodonski. 1993. Hyaloid vessels of the human fetal eye. A scanning electronmicroscopic study of corrosion casts. Archives of Ophthalmology, 111 (11): 1573-1577. Supin, A.Y., V.V. Popov, and A.M. Mass. 2001. The Sensory Physiology of AquaticMammals. Kluwer Academic Publishers, Boston. Suzuma, K., H. Takagi, A. Otani, I. Suzuma, and Y. Honda. 1998. Increased expressionof KDR/Flk-1 (VEGFR-2) in murine model of ischemia-induced retinalNeovascularization. Mi crovascular Research, 56: 183-191. Tartaglia, M., C.M. Niemeyer, K.M. Shannon, and M.L. Loh. 2004. SHP-2 and myeloidmalignancies. Current Opinion in Hematology, 11 (1): 44-50. Terry, M.A. and P.J. Ousley. 2003. Re placing the endotheliu m without corneal surfaceincisions or sutures: The first Unite d States clinical series using the deep lamellarendothelial keratoplasty pro cedure. Ophthalmology, 110 (4): 755-764. Timney, B. and K. Keil. 1992. Visual acu ity in the horse. Vision Reseach, 32 (12): 2289-2293. Twining, S.S., P.M. Wilson, and C. Ngamkitidechakul. 1999. Extrahepatic synthesis of Plasminogen in the human cornea is up-regulated by interleukins-1 and 1 .Biochemical Jour nal, 339: 705-712. Ueda, T., T. Ueda, S. Fukuda, R. Browne, E. Jenis, R. Spengler, R. Chou, P. Buch, A.Aljada, P. Dandona, R. Sasisekharan, C.K. Dorey, and D. Armstrong. 1997.Lipid hydroperoxide-induced tu mor necrosis factor (TNF), vascular endothelial growth factor a nd neovascularization in the ra bbit cornea: effect of TNF inhibition. Angiogenesis, 1 (2): 174-184. United States Fish and Wildlife Service. 2001. Florida Manatee Recovery Plan. 3rdRevision, pages 1-109. United States Fi sh and Wildlife Service, Atlanta, Georgia. United States Marine Mamma l Commission. 2001. Annual Report to Congress, pages United States Marine Mammal Commission, Bethesda, Maryland. Vemuganti, G.K., M.S. Sridhar, D.P. Edward, and S. Singh. 2002. Subepithelial amyloid deposits in congenital hereditary endothelial dystrophy: A histo-pathologic study of five cases. Cornea, 21 (5): 524-529. Walls, G.L. 1942. The Vertebrate Eye and Its Adaptive Radiation. Hafner Press, New York, N.Y.

PAGE 164

153 Wartzok, D. and D. Ketten. 1999. Marine mammal sensory systems. Pages 117-175.In J. Reynolds and S. Rommel, editors. Bi ology of Marine Mammals. Smithsonian Instiution Press, New York. Watson, A.G. and R.K. Bonde. 1986. Congenital malformations of the flipper in three West Indian manatees, Trichechus manatus and a proposed mechanism for development of ectrodactyly and cleft hand in mammals. Clinical Orthopaedics,202: 294-301. Weber, A., U.R. Hengge, D. Urbanik, A. Markwart, A. Mirmohammadsaegh, W.B. Reicehl, C. Wittekind, P. Wiedemann, and A. Tannapfel. 2003. Absence ofmutations of the BRAF gene and constitutive activations of extracellularregulated kinase in malignant melanomas of the uvea. Laboratory Investigation,12: 1771-1776. West, J.A., J.G. Sivak, C.J. Murphy, and K.M. Kovaks. 1991. A comparative study ofthe anatomy of the iris and ciliary body in aquatic mammals. Canadian Journal of Zoology, 69: 2594-2607. Westheimer, G. 1992. Visual acuity. Pages 531-547. In W.M. Hart, Jr., editor. Adlers Physiology of the Eye, NinthEdition. Mo sby-Year Book, Inc., St. Louis, Missouri. Whittle, E.J., C. Bryans, and B. Timney. 1987. Visual acuity in esotropic cats followingOcclusion of the non-deviating ey e. Behavioural Brain Research, 24: 101-109. Yancopoulos, G.D., S. Davis, N.W. Gale, J.S. Rudge, S. J. Wiegand, and J. Holash. Vascular-specific growth factors and bl ood vessel formation. Nature, 407: 242248. Yang, S., K. Toy, G. Ingle, C. Zlot, P.M. Williams, G. Fuh, B. Li, A. Vos, and M.E. Gerritsen. 2002. Vascular endothelial growth factor-induced genes in human umbilical vein endothelial cells, relative roles of KDR and Flt-1 receptors.Arteriosclerosis, Thrombosis, and Vascular Biology, 22: 1797-1803. Yazdanfar, S., A.M. Rollins, and J.A. Izatt. 2003. In vivo imaging of human retinal flow dynamics by color Doppler optical cohe rence tomography. Arch Ophthalmology, 121: 235-239.

PAGE 165

154 BIOGRAPHICAL SKETCH Jenny Harper was born in Georgia on Febr uary 3, 1978, and lived in Bainbridge, Georgia, until she was 19. Jenny received an Associate of Arts degree from Bainbridge College in May 1997 and a bachelors degree in biology from the University of Georgia in May 1999. Jenny then did research at Ge orgia Southern Univer sity for her thesis involving the captive behavior of Florida ma natees. She graduated with her masters degree from Georgia Southern Univers ity in December 2001. Jenny arrived in Gainesville to start her PhD at the Univers ity of Florida under Dr. Don Samuelson in August 2001. After completion of her PhD, Je nny hopes to work in th e field of wildlife biology and also plans on enjoying time with her and her husbands new baby.


Permanent Link: http://ufdc.ufl.edu/UFE0003600/00001

Material Information

Title: Corneal Vascularization in the Florida Manatee (Trichechus manatus latirostris)
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0003600:00001

Permanent Link: http://ufdc.ufl.edu/UFE0003600/00001

Material Information

Title: Corneal Vascularization in the Florida Manatee (Trichechus manatus latirostris)
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0003600:00001


This item has the following downloads:


Full Text












CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE
(Trichechus manatus latirostris)


















By

JENNIFER YOUNG HARPER


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2004

































Copyright 2004

by

Jennifer Young Harper















ACKNOWLEDGMENTS

I would like to thank my mentor, Dr. Don Samuelson. He has been a wonderful

source of knowledge, inspiration, and motivation. Without his help and guidance, I could

not have accomplished any of this work. I would also like to thank Dr. Roger Reep for

all of his help and support along the way. He too has acted as a rock and support system.

My additional committee members (Dr. Peter McGuire, Dr. Dennis Brooks, and

Dr. Gordon Bauer) have been tremendous support and I thank them for all they have

offered. Their guidance has been appreciated beyond belief.

Laboratory technologists Pat Lewis and Maggie Stoll were extremely helpful and

supportive during much of my work. I would have not been able to accomplish the first

procedure without their help. I thank these fine ladies from the bottom of my heart.

My parents, Jim and Marion Young, have meant more to me than I can ever

describe or explain. I appreciate all their love and support.

Finally, I thank my husband Ridge Harper. He has been the strongest support

system I could ever ask for and makes me happier than I ever knew I could be.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ......... .................................................................................... iii

LIST OF TABLES .................................................... ... .................. vi

LIST OF FIGURE S ......... ....................... ............. ........... vii

A B STR A C T ................................................. ..................................... .. x

CHAPTER

1 L ITER A TU R E R E V IEW .............................................................. .. ....... ..............

The Florida M anatee ................. .......... ..... ............................. .... ...............
The Tactile Sensory System of Marine Mammals ........................................... 2
The Auditory System of M arine M ammals....... ....... ........ ...................................5
The Olfaction and Gustation of M arine M ammals.....................................................7
The Vision of M arine M ammals........................... .................... ... .......... ..... 8
T h e C orn ea .......................................................................................... 13
A ngiog en esis ................................................................................................ 17
O b j e c tiv e s ........................................................................................................2 2

2 CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE AND
THREE-DIMENSIONAL RECONSTRUCTION............................................... 23

In tro du ctio n ...................................... ................................................ 2 3
M materials and M methods ....................................................................... ..................32
Specim en C collection .......... ...................................................... .. ........... ..... 32
R e su lts .................. ........................................................................3 6
A natom ical D description ............................................................ .....................36
M orphom etry .................................................................................. 39
Statistics ...................................... ............................... .......... ...... 4 0
D iscu ssio n ...................................... ................................................. 4 2

3 VASCULAR ENDOTHELIAL GROWTH FACTOR RECEPTOR-1 AND -2 IN
THE FLORIDA MANATEE CORNEA....................................... ............... 96

Introduction .............. ....... .............. ...... .... .......... ............ 96
M materials and M methods ........................................... ....................................... 106










Specim en C collection ................................................. ............................. 106
P reparation of E yes......... ........................................................ ............ 106
H isto lo g y ...............................................................10 7
Im m unohistochem istry .................................................................. ..................107
R e su lts ............................................................................... ....... ......... 1 0 8
Vascular endothelial growth factor receptor-1 (VEGFR-1)............................. 108
Vascular endothelial growth factor receptor-2 (VEGFR-2)..............................109

4 ADDITIONAL STUDIES ......................................................... ...............127

Corneal Vascularization in the Antillean Manatee..............................127
Ophthalmic Examinations in Two Captive Florida Manatees..............................128

5 GENERAL CONCLUSIONS...................................................... ............... 139

L IST O F R E FE R E N C E S ......... ................................................................ ..................142

BIOGRAPH ICAL SKETCH .............. ......................... ................... ............... 154




































v
















LIST OF TABLES

Table pge

2-1 Information on manatee tissue used and volume of vasculature at 25 X.................41

2-2 Volum e of vasculature at 200 X ..................................................... .................42
















LIST OF FIGURES

Figure page

2-1 Florida manatee eye with ocular tissue and fat removed revealing the eye and
o p tic n erv e ........................................................................ 5 5

2-2 Histological pictures of M NW -9111 ......................................................... 56

2-3 These pictures were taken ofMSW-9114 of the same blood vessel at varying
magnifications to show the penetration of this vessel into the anterior
epithelium ........................................... ........................... 57

2-4 UCF-9114 shows the smaller vessels that were outlined within the stroma for
three-dimensional reconstruction (200X). .................................. .................58

2-5 Pictures from M N W -9016 ........................................................................ ... ... 59

2-6 These pictures were taken from the unknown specimen. (A) Shows the edge of
the limbal vascular arrangement at the peripheral cornea (200X), and (B) larger
vessels that go right up to the anterior epithelium at 200X. ................................60

2-7 Anatomical orientation of manatee eyes as they are described in three-
dim ensional reconstruction ........................................................ ............. 61

2-8 M N W -9016 ............................................................................................. ........62

2-9 M N W -9022(R) ............................................................. ...... .... .. 63

2-10 M N W -9111 ................................................................................................. ......64

2-11 M N W -9116 ................................................................ .......65

2-12 M SE -9101 .................................................................................................. ......66

2-13 M SE -9 105 .............................................................................................. ........67

2-14 M SE-9107 .............................................................................................. ....... 68

2-15 M SE -9115 .................................................................................................. ......69

2-16 M SE -9131 ............................................................................................ .......70









2 -17 M S W -9 1 14 ..................................................................................... 7 1

2-18 M SW -913 1(R). .................................. .. .. .. ..... .. ............72

2-19 M SW -9137(L).................. .......................................... ....... ............. 73

2-20 M SW -9137(R). .................................... .. .. ....... .. ............74

2-21 M SW -9141 ................................................................ ......75

2-22 M SW -9143 ............................................................... ......76

2-23 TM -8619(L ) ................................................................................................. ...... 77

2-24 TM -8619(R ) .................................................................................................. ......78

2 -2 5 T M -8 6 3 0 (L ) .................................................................. .................................7 9

2-26 T M -8630(R ) .................................................................80

2-27 U C F-9114. .................................................................. 81

2-28 U C F -9 137 ................................................................... 82

2-29 U C F -9 138 ................................................................... 83

2-30 U CF-9141. ..................................................................84

2-3 1 U know n ..................................................................................................... 85

2 -3 2 F etu s(L ) ......................................................................................................8 6

2-33 F etu s(R ) ...................................................................................................... ......87

2-34 Dorsal quadrant of MNW-9016 at 200X ....................... .......................88

2-35 Medial quadrant of MNW-9016 at 200X .......................................................89

2-36 Dorsal quadrant of MSW-9114 at 200X .............................................................90

2-37 Ventral quadrant of MSW-9114 at 200X .......................................................91

2-38 Dorsal quadrant of TM-8619(L) at 200X .......... ..................... 92

2-39 Medial quadrant of TM-8619(L) at 200X........ .......................93

2-40 Ventral quadrant of TM-8619(L) at 200X ........ .......................94

2-41 Lateral quadrant of TM-8619(L) at 200X ............................................. ....... 95









3-1 Florida m anatee eye ......................................................................... 118

3-2 Pictures taken from TM-0310 showing a positive reaction of VEGFR-1 .........19

3-3 VEGFR-1 positive reactions in TM-0310 showing the prevalence of the
positive reaction at 200X with 1:50. ........................................ ............... 120

3-4 VEGFR-1 positive reactions in TM-0310 (A) at 400X with 1:100....................121

3-5 VEGFR-1 reactions in TM-0310 of control samples............. ................122

3-6 Pictures taken from TM-0311 showing a positive reaction of VEFGR-1 ...........123

3-7 VEGFR-1 positive reactions in TM-0311................................ .....124

3-8 VEGFR-1 reactions in TM-0311 of control samples................................ 125

3-9 Pictures taken of TM-0310 showing no positive reactions of VEGFR-2 that
appears to be different from the control samples........................................126

4-1 The cornea from an Antillean manatee eye.. ................................................. 130

4-2 This corneal section was cut at 10 microns and stained using Masson's
Trichrom e stain. ................... .. .............. .. ... .. .................... 131

4-3 These pictures show the unusually nonorganized, wavy stroma observed in
the A ntillean m anatee cornea .................................... ............................. ........ 132

4-4 Extensive vasculature was observed in the (A) stroma of the Antillean manatee
cornea at 100X and (B) as the stroma approaches the anterior epithelium at
1 0 0 X ...................................... ................................................... 1 3 3

4-5 These pictures illustrate the vessels in the anterior epithelium............................134

4-6 Vasculature can be observed (A) beneath the anterior epithelium at 200X and
(B) penetrating the anterior epithelium at 400X. ............................................135

4-7 The unusual material thought to be mucus surrounding the anterior
ep ith eliu m ...................................................... ................ 13 6

4-8 Unusual projections were observed in the posterior stroma and adjacent
en d oth eliu m ............................ ..................................................... ............... 13 7

4-9 A PAS stain of the Antillean manatee cornea showing no reaction to a fungus
in the anterior epithelium at (A) 100X and (B) 200X............... .... ..............138















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE
(Trichechus manatus latirostris)

By

Jennifer Young Harper

May 2004

Chair: Don A. Samuelson
Cochair: Roger L. Reep
Major Department: Veterinary Medical Sciences

The cornea of the Florida manatee is anatomically unique because blood vessels are

found throughout the cornea. In all other species of animals, this is considered a

pathological condition that impedes vision and is usually caused by injury or trauma of

some sort. The purpose of this study was to describe more clearly corneal vascularization

by (1) examining the architecture through three-dimensional reconstruction in order to

find possible patterns in size, distribution, and location of blood vessels that may be

related to age, gender, location found, and environment, and (2) examining vascular

endothelial growth factor receptors-1 and -2 in the cornea. Twenty-six eyes from 22

individuals were prepared for histological examination and subsequent three-dimensional

reconstruction. Every specimen examined possessed blood vessels in the cornea, mostly

comprising 0.3% of total surface density (volume) of the cornea. However, two

individuals had a greater density of vascularization (1.2% and 0.7%). The size and









distribution of vessels appeared to be greatest toward the mid-periphery of the cornea.

No differences were found among individuals based on age, gender, and site of recovery.

Environmental influences were not a significant factor either, which was not originally

anticipated. The presence of vessels at the level of the anterior epithelium was surprising,

and it appeared that the vascularization was directed more anteriorly than originally

thought. Blood vessels were also found in fetal eyes. In all the eyes examined, no signs

of injury or trauma could be observed. Vascular endothelial growth factor receptor-1

(VEGFR-1) was found in the corneas of 2 individuals examined through

immunohistochemistry. However, vascular endothelial growth factor receptor-2

(VEGFR-2) was not, indicating that a noninflammatory and nonpathological process of

angiogenesis might be taking place in the Florida manatee cornea. The presence of blood

vessels appears to impair vision minimally based on the blood vessels' low density, size,

and location. The penetration of vessels into the anterior epithelium and development of

vessels within the fetus points to an evolutionary adaptation, possibly due to the

manatee's unique ability to move freely in salt, brackish, and fresh water.














CHAPTER 1
LITERATURE REVIEW

The Florida Manatee

The Florida manatee, Trichechus manatus latirostris, is perhaps one of the most

interesting marine mammals. It is a subspecies of the West Indian manatee and is in the

order Sirenia, which includes two additional species, the Amazonian manatee and the

West African manatee. The order Sirenia is made up of two families, Trichechidae

(manatees) and Dugongidae (dugongs). Manatees are the only marine mammal

herbivores (Beck and Barros 1991; Reynolds and Odell 1991) and the Florida manatee

feeds on over 60 different species of aquatic vegetation (Provancha and Hall 1991).

Trichechids are also unique among marine mammals in that they are able to live in

different types of aquatic habitats including fresh water, salt water and brackish water

(Hartman 1979). During warmer months of the year, manatees can be found in

subtropical and tropical coastal waters of the Southeastern portion of the United States.

However, during winter months, manatees are limited to the warmer waters of Florida

because they have relatively low metabolic rates that prevent them from producing heat

efficiently in cooler water (Gallivan et al. 1983; Irvine 1983). Therefore, Florida

manatees are particularly vulnerable to water temperatures that fall below 20C (Bossart

et al. 2002; Bossart 2001).

Florida manatees are listed as endangered species in the United States (United

States Marine Mammal Commission 2001) and their survival is jeopardized by many

environmental and human factors (United States Fish and Wildlife Service 2001).






2


Environmental influences include infection caused by epizootic outbreaks (Bossart et al.

1998) and exchange of morbillivirus between individuals (Duignan et al. 1995).

However, human influences are more often the cause of death or injury to manatees,

including the ingestion of debris such as monofilament fishing line (Beck and Barros

1991), oil spills and habitat alteration (Lefebvre and O'Shea 1996). Perhaps the most

direct human impact on manatees' survival is collision with watercraft. Mortality rates

from 1974 to 1996 indicate that 23.4% of manatee deaths were caused by watercraft

(Reynolds 1999).

Due to the endangered species status of the manatee, the United States Fish and

Wildlife Service (2001) designed a recovery plan for the Florida manatee. This recovery

plan outlines actions that need to be taken in order to reduce threats to Florida manatees.

Included are four primary objectives: (1) minimize causes of manatee disturbance,

harassment, injury and mortality; (2) determine and monitor the status of manatee

populations; (3) protect, identify, evaluate, and monitor manatee habitats; and (4)

facilitate manatee recovery through public awareness and education. In efforts to reduce

the threats to manatees using the recovery plan, the scientific community has taken quite

an interest in learning as much as possible about these unique gentle giants. Establishing

a better understanding of the five sensory systems, touch (tactile), hearing (auditory),

smell (olfaction), taste gustationn), and vision, is a step in the right direction in order to

discover methods to conserve the species, learn about their biology, and how they interact

in their environment.

The Tactile Sensory System of Marine Mammals

Approximately 120 million years ago mammals began to develop the one of their

sensory systems, that of touch through the use of whiskers or vibrissae. These tactile









structures have evolved to increase the search area for an animal (Brecht et al. 1997).

Vibrissae can be found on many areas of animals, but most commonly the feet or pads

and face. The vibrissal follicle-sinus complexes of the mystacial vibrissae, located on the

face, are very well innervated (Ebara et al. 2002) and are used during active touch to

discriminate objects (Bachteler and Dehnhardt 1999). It has been shown that rats use

their mystacial vibrissae exclusively to discriminate textured surfaces. Additionally,

walruses and California sea lions use their mystacial vibrissae to distinguish different

shapes (Bachteler and Dehnhardt 1999).

Mammals living in aquatic environments often need to interpret information about

their surrounding conditions when visibility is very poor if not impossible. Therefore,

several species of mammals living in these conditions have developed the sensory

abilities to compensate for visual losses (Dehnhardt et al. 2001). Most marine mammals

have some mechanism for tactile sensation in which they interpret information about their

environment, and for these aquatic animals, vibrissae are one of the ways to do so. In

marine mammals, vibrissae are most commonly associated with pinnipeds, being what

appear to be the whiskers of the animal. The vibrissae of pinnipeds are innervated

peripherally and are found in large sections of the animals' bodies to maximize the

amount of information gained about the surrounding environment. Vibrissae are not

limited to pinnipeds, however, as all marine mammals have some form of vibrissae

varying in tactile sensitivity (Wartzok and Ketten 1999). The Florida manatee has thick,

dense lips with attached rigid vibrissae that create a unique anatomy of the face (Marshall

et al. 1998a). These vibrissae flare out and are used to grab vegetation and to manipulate

objects in their environment (Reep et al. 1998). Additional vibrissae are found in the oral









cavity, being smaller and less noticeable, and still concerned with feeding behaviors

(Marshall et al. 1998b). Florida manatees also have facial hairs, as well as stiff hairs

found on the oral disk (Reep et al. 1998). The perioral bristles found on the manatees'

muscular snouts, which are used in a prehensile manner to forage on aquatic vegetation,

are distinctive to Sirenians (Marshall et al. 2000).

Reep et al. (2001) examined the follicles of the face and found that they showed

characteristics of vibrissae that included a well-defined blood sinus, a connective tissue

capsule, and dense innervation. This finding suggests that the anatomy of these follicles

is comparable to that of many other mammals that possess vibrissae used for tactile

exploration, such as the closest living land relative of the manatee, the elephant. The tip

of the trunk of elephants is referred to as the finger and is extremely mobile in that it can

be used for a wide range of behaviors such as grasping food, tactile sensitivity, and

chemosensory recognition. The skin of the trunk tip has a high density of nerve endings,

several convoluted branched corpuscles, and vibrissae that are densely innervated. The

unique innervation found in the trunk tip of Asian elephants is similar to that of the

mystacial vibrissae found in rodents and the lip tissue found in some monkey species

(Rasmussen and Munger 1996). A study conducted by Bachteler and Dehnhardt (1999)

investigating the tactile hairs on the face of the Antillean manatee found that it had a

tactile sensitivity that compared closely to the trunk tip of Asian elephants, but was less

sensitive than that of the vibrissae of pinnipeds.

More recent studies conducted by Reep et al. (2002) suggest that in addition to

sensory vibrissae located on the facial region, Florida manatees also possess sensory hairs

located on the body that may be analogous to the lateral line found on fish. The follicles









of these sensory hairs were examined histologically and also showed characteristics of

follicle-sinus complexes. It was found that the hairs varied in follicle size and there

appeared to be no major differences in follicle geometry. The follicles found on the body

differed from those on the face in that body follicles lacked a ring sinus, were less

innervated, and the capsules were thinner in body follicles than those on the face. Reep

and his colleagues suggested that these hairs are actually vibrissae used to detect water

displacements connected to environment stimuli, such as other manatees, currents and

water flows. The hairs found on the body of the manatee might be similar to the whiskers

of harbor seals (Phoca vitulina), which also act as the fish lateral line by detecting water

movements and other information about the surrounding environment (Dehnhardt et al.

2001). If these body hairs of the manatee are indeed tactile hairs, then this information

combined with the studies of the facial vibrissae, indicates that manatees' most developed

sensory system would be the tactile sensory system (Reep et al. 2002).

The Auditory System of Marine Mammals

Hearing in mammals is a remarkable process that involves several very small bones

in the ear. The malleus, incus, and stapes in the middle ear transmit sound to the inner

ear from the tympanic membrane. In land mammals, vibrations occur in the tympanic

membrane where sound is received and then the sound is passed on to the malleus. The

malleus and incus then rotate together producing a back and forth movement at the

vestibular window where sound pressure is amplified (Kinkel et al. 2001; Mallo et al.

2000). This process that takes place in the middle ear of land mammals allows them to

efficiently receive airborne sounds. However, for marine mammals, this process is not

efficient because once sound has passed through water into the air-filled middle ear, it

has been obstructed. Most marine mammals have developed unique middle ears to









compensate for this loss and are well adapted to hearing sounds in their underwater world

(Ketten 1991). The ossicles in the middle ears of cetaceans have evolved to transmit

vibrations to the oval window where the velocity is amplified and not the pressure, as in

land mammals. The ossicles of cetaceans are also inflated and much denser than that of

any other mammals (Ridgway et al. 2001; Nummela et al. 1999).

The Florida manatee also adapted to hearing underwater. The ears of the manatee

are located just behind the eyes and have very small openings. Like pinnipeds, the ear

bones of manatees are in direct contact with the cranium (Wartzok and Ketten 1999).

Because of this anatomical positioning, it has been suggested that sound passes through

the lower jaw and is carried to the ear bones (Reynolds and Odell 1991). Experiments

conducted by Gerstein et al. (1999) tested the auditory abilities of two captive manatees

to measure the underwater hearing sensitivity with regard to frequency and hearing

threshold. When measuring hearing sensitivity, an audiogram is often performed. An

audiogram is also referred to as a hearing curve and graphs the range of frequencies that

can be heard. Additionally, an audiogram measures an individual's sensitivity within its

hearing range. Most mammals have a U-shaped plot on a hearing curve because the

intensity of a signal at its lowest detection threshold is measured. The highest threshold

areas indicate the lowest sensitivity and are found on either side of the frequency range

(Gerstein 2002). The experimenters discovered that manatees displayed an inverted U-

shaped hearing range from 500 Hz to 38 kHz. This information suggested that manatees

hear low frequency sounds below 16 kHz (i.e., the frequency at which they can discern

the lowest amplitude or quietest sound) as well as other marine mammals (Gerstein et al.

1999). Audiograms conducted on other marine mammals have produced U-shaped









hearing ranges also. The arctic white whale, Delphinapterus leucas, has a hearing range

of 40 Hz to 150 kHz (Ridgway et al. 2001); the pacific walrus, Odobenus rosmarus

divergens, has a range of 1 to 12 kHz (Kastelein et al. 2002a); the harbor porpoise,

Phocoenaphocoena, has a range between 250 Hz and 180 kHz (Kastelein et al. 2002b);

and the striped dolphin has a range of 29 to 123 kHz (Kastelein et al. 2003).

Compared with other marine mammals, manatees appear to have a similar hearing

sensitivity with pinnipeds and some cetaceans of a range between 10 and 26 kHz. This

range of frequency seems to allow these marine mammals to hear best in low noise

conditions underwater (Gerstein et al. 1999; Ketten 1991). It also appears that manatees

and killer whales (Orcinus orca) share a peak hearing sensitivity around 16 kHz

(Gerstein et al. 1999). To put this in terms relative to humans, normal hearing range in

adults is between 0.04 and 16 kHz (Wartzok and Ketten 1999).

The Olfaction and Gustation of Marine Mammals

In order to detect materials in the water and in the air, smell and taste are usually

closely associated senses. Neither smell nor taste has been examined thoroughly in

marine mammals. Smell sensations decreased in marine mammals as they adapted to

their aquatic environment, and while taste sensations have been observed in dolphins,

there is a lack of experimentation to support this with other marine mammals (Wartzok

and Ketten 1999). Anatomical studies have shown that taste and smell of terrestrial

mammals are far more developed than that of certain marine mammals including

mysticetes, odontocetes, and sirenians. Manatees have a rudimentary olfactory system

and both cetaceans and manatees are deficient of a vomeronasal organ, which is an

essential factor of the olfactory system (Wartzok and Ketten 1999; Mackay-Sim et al.









1985). Studies done by Hartman (1979) indicated that manatees might use their sense of

smell more through their mouth than their nose.

Manatees have taste buds at the back of their tongues (Levin and Pfeiffer 2002;

Reynolds and Odell 1991), located on the dorsal and posterior lateral swellings (Levin

and Pfeiffer 2002; Wartzok and Ketten 1999). The taste responses in manatees appear to

be more developed than those in cetaceans. Manatees may be using taste and smell to

avoid eating certain plants, as well as for recognition of other manatees and determining

if a female is in estrous (Reynolds and Odell 1991). Overall, the taste and smell

capabilities of manatees are not well known and require further investigation.

The Vision of Marine Mammals

The amount of light present in any environment affects the way a terrestrial or

marine mammal will view its world. However, for marine mammals, the amount of light

in their aquatic environment can be very different from that of the environment of land

mammals. Light passing through the water is absorbed, refracted, and scattered, and can

then be affected by its wavelength, chlorophyll concentration, and organic matter

concentration in the water. All of these factors can potentially influence visual detection

and visual acuity (Wartzok and Ketten 1999). Visual detection for marine mammals

refers to adaptations that facilitate its becoming aware of predators or prey in the aquatic

environment. When an animal identifies a predator or prey, it must take some type of

action. If a marine mammal is to find prey at deep depths in the water, then it needs to

develop visual adaptations such as corresponding receptor pigment sensitivity to low

light environments, having an increased number of photoreceptors, and being able to

increase the light intensity captured through the tapetum (Wartzok and Ketten 1999).









Most marine mammals' photoreceptors will have the highest sensitivity in the light

range of water where they usually feed. For example, the pigments of rod receptors of

pinnipeds and cetaceans have their highest sensitivity near the blue end of the spectrum

because this is the same as the wavelengths of light piercing the open ocean (Wartzok

and Ketten 1999). The number of photons captured dictates what visual signals will be

detected (Saari 1992). Marine mammals are able to detect objects in their environment in

a special way, especially in the open ocean. Their well-developed tapetum reflects

photons and this reflection process gives the visual pigment further opportunity to secure

a photon, increasing the chance for detection of objects in the environment (Wartzok and

Ketten 1999). Not surprisingly, marine mammals have more highly developed tapeta

than any other mammal (Supin et al. 2001; Walls 1942).

Visual acuity measures an animal's ability to resolve features in its environment

and is measured in degrees or minutes of arc (Wartzok and Ketten 1999; Westheimer

1992). The refraction of light rays depends on differences in refractive indices and the

angle of light. The lens of terrestrial mammals depends on this refractive index and the

radius of curvature, which can generate a lens with a power of 25 to 40 diopters (a diopter

is the measure within the lens of refractive power). As marine mammals adapted to their

aquatic environment, the strength of their principal refractive structure, the lens, was

markedly reduced. This loss of refractive strength occurred because as light passed

through water into the corneas, there was a very small change in refractive index due to

the refractive index of water and the interior of the eye being so similar. To compensate

for this, marine mammals developed spherical lenses, which are more similar to fish than

to terrestrial mammals (Wartzok and Ketten 1999).









Within the last century, manatee eye anatomy and vision have not been studied

extensively. Early studies conducted by Walls (1942) suggested that manatees have

limited visual capabilities and acuity. Other research conducted (West et al. 1991;

Piggins et al. 1983) supported Walls' theory that manatee eyes were adapted for low light

conditions and contained very few ganglion cells, as well as no mechanism for

accommodation, thus limiting vision. Behavioral research conducted by Hartman (1979)

suggested that vision is the sensory mechanism which manatees rely on the most for

receiving information about their environment and that manatees were most likely

hypermetropic (vision for distant objects is better than vision for near objects). However,

the research of Piggins et al. (1983) suggested that manatees were emmetropic (an eye

with normal vision that sharply focuses vision at all distances). Research conducted by

Cohen et al. (1982) further supports the idea that manatees' vision is not as poor as

previously thought. Cohen and his colleagues found two types of photoreceptors, rod-

like and cone-like, in the eyes of the West Indian manatee. The photoreceptors were

found in the retina at low rod:cone ratios suggesting that the occurrence of a large

number of cone cells in the rod-dominated retina would indicate relatively good visual

acuity. They also suggested that by having both rods and cones in the retina, manatees

were likely to have vision functioning more on a diurnal level than nocturnal.

Mass et al. (1997) studied the topographic order of ganglion cells in the retina of

the Florida manatee. They found that the ganglion-cell distribution did not appear to be

even and was diverse across the retina with higher ganglion cell densities near the center

of the retina. Based on these findings using ganglion cell densities, the investigators were

able to calculate retinal resolution. Retinal resolution can be used as an accurate measure









for visual acuity because the retinal resolution in the normal eye has evolved to have a

similar resolution as the eye optics (Supin et al. 2001). Visual acuity can be measured in

minutes or degrees of visual arc, which is also called cycles per degree. Mass and his

team (1997) found that the Florida manatee had limited visual resolution; 20' of visual

arc, suggesting manatees are myopic (nearsightedness or when blurred vision results as

objects are moved further away). These anatomical findings are consistent with

behavioral studies conducted by Bauer et al. (1999) and Colbert et al. (1999). Two

manatees were trained with two-choice simultaneous discrimination tasks. The manatees

were trained to discriminate vertical black and white bands equated for brightness.

During each training trial, a standard and a target with wider vertical stripes were

presented. The manatees were then reinforced with food for choosing the target with

wider stripes. Visual acuity of the manatees was measured in degrees of visual arc

depending on the size of the vertical bands and the distance the eyes were from the target.

Based on these visual discrimination tasks that the two manatees were trained to do, the

research team found that one manatee had a visual arc of 23' and the other manatee had a

much lower level of visual acuity at 1 degree of visual arc. Compared with other

mammals, both terrestrial and marine, there is a wide range of visual acuity results. It has

been found that mixed breed kittens have a visual acuity measure of 3.86 cycles per

degrees (Whittle et al. 1987); horses have a measure of 23.3 cycles per degree (Timney

and Keil 1992); the bottlenose dolphin, Tursiops truncates, measures at 9' in water (Mass

and Supin 1995); the harbor porpoise measures between 2.2 and 2.6 cycles per degree;

the gray whale, Eschrichtius gibbosus, has a visual acuity of 2.4 to 2.9 cycles per degree;









and the Amazon river dolphin, Inia geoffrensis, has a very poor visual acuity of 0.6 to

0.75 cycles per degree (Supin et al. 2001).

Recent studies conducted by Griebel and Schmid (1996) found that manatees do

have limited color vision. Manatees were trained in captivity to discriminate a gray

target during a two-fold simultaneous choice test. Three different shades of blue, green,

and red were tested against varying gray colors. A color would first be presented with a

gray that was almost black. As trials continued, the brighter grays were presented until a

gray almost white in color was reached. If the manatee discriminated the color from

gray, then it was assumed that color was perceived. Discrimination between some shades

of gray and a color was not possible if they had a similar brightness. The manatees were

able to discriminate blue and green from several gray hues, but could not distinguish red

and blue-green from gray hues. Griebel and Schmid suggested that the color vision of

manatees was similar to that of fur seals and California sea lions in which similar

methodologies were used to test color vision. The method used to test color vision of

manatees suggests they possess dichromatic color vision. In addition, manatees were also

able to distinguish a color independently of the shape of the stimulus and other colors.

Additionally, the spotted seal seems to possess some capacity for color discrimination

based on behavioral studies (Peichl et al. 2001). However, another marine mammal, the

bottlenose dolphin, does not seem to have the ability to discriminate colors based on

behavioral tests (Madsen and Herman 1980).

An additional study conducted by Griebel and Schmid (1997) investigated

brightness discrimination in the manatee. Two manatees were trained to discriminate

gray targets that varied in brightness in a two-fold simultaneous choice test. Thirty









different shades of gray were used and the manatee made a decision between different

targets by touching its choice with its snout. The Weber fraction, which is a formula used

to calculate the difference in relative reflection, was used to determine how well the

manatees discriminated differing shades of brightness. It was found that manatees have

brightness discrimination ability three times less than that of humans. This additional

study further suggested that manatees do possess dichromatic color vision and that the

brightness discrimination ability is similar to that of the fur seal.

A histological examination conducted by Samuelson et al. (1994) examining the

entire eye of four Florida manatees found some very interesting and unique results. They

found that the corneas were small and round with anterior epithelium of 14-18 cell layers.

A diffused capillary bed was found under the anterior epithelium in each eye. A large

majority of the stroma in the cornea appeared to be vascularized. This was an amazing

find because in no other species of animal have blood vessels been consistently found in

the cornea with no signs of injury or infection (Klintworth 1991). An additional

discovery of interest made by Samuelson et al. (1994) was that manatees lack a

nontapetal choroid. Based on their histological and anatomical findings, they suggested

that manatees' visual systems function diumally and are unique from other marine

mammals because they have no tapetum.

The Cornea

Several interesting discoveries were made by Samuelson et al. (1994) upon

histologically examining the eye of the Florida manatee, especially the cornea. In order

to understand the significance of finding blood vessels in several corneas without injury,

it is important to understand the morphology and anatomy of the normal cornea in most

mammalian species. The cornea is the anterior-most, transparent avascular layer of the









eye. Its primary functions are to provide support for the contents within the globe, refract

light, and transmit light. Corneal shape is elliptical, and in most species, the horizontal

diameter is larger than the vertical diameter. Cats and dogs have a diameter that is very

similar, but ungulates have a much wider horizontal diameter, which aids in their

horizontal field of view. The thickness of the cornea varies between species. Usually the

thickest area of the cornea is along the peripheral edges and the thinnest area of the

cornea is in the center (Samuelson 1999). The cornea is a transparent tissue because it

lacks blood vessels, is supplied with sensory nerves, has a nonkeratinized surface

epithelium, and lacks pigmentation (Muller et al. 2003).

The cornea is composed of four (and sometimes five) layers including the anterior

epithelium, Bowman's layer, stroma, Descemet's membrane, and the endothelium

(Samuelson 1999). The anterior epithelium covers the corneal anterior surface. It is

nonkeratinized and made up of stratified squamous cells. In carnivores, the epithelium is

25 to 40 [tm thick, while in ungulates the epithelium can be two to four times thicker.

The epithelium in cats, dogs, and birds consists of a single layer of columnar basal cells,

which lie on a thin basement membrane and give rise to several layers of wing-shaped

cells and two to three layers of nonkeratinized squamous cells. Larger animals have

more layers of wing-shaped and squamous cells (Agrawal and Tsai 2003). A basement

membrane lies under the epithelium. The basement membrane is usually 30 to 55 nm

thick and is anchored to the stroma by hemidesmosomes. The hemidesmosomes vary

among species; they are linear in mammals and amphibians, rosettes in birds and reptiles,

and without arrangement or absent in fish. Regenerative capabilities of epithelial cells

are good (Buck 1983).









The stroma, which comprises approximately 90% of the corneal thickness, is made

up of sheets of lamellae tissue that are transparent. Cells called keratocytes can be found

among the lamellae sheets. These keratocytes aid in the maintenance and formation of

the lamellae, and can form into fibroblasts if injury within the cornea takes place. The

lamellae run parallel with the diameter of the cornea. Maintenance of corneal clarity is

the primary function of the stroma. The stroma's organization allows 99% of light to

pass through the cornea without scattering and the primary support structure of the

stroma is made up of collagen fibrils, proteoglycans, and glycoproteins, which constitute

15 to 25% of the stroma (Joyce 2003; Hogan et al. 1971).

In humans, nonhuman primates, and avian species, an additional corneal layer

exists. This layer is called Bowman's layer and is found in the anterior-most region of

the stroma. Collagen fibrils in Bowman's layer are found randomly and are small in

diameter compared to the stroma. The Bowman's layer is 10 to 15 .im thick and is

acellular. The collagen found in Bowman's layer is produced by the anterior epithelium

(Hayashi et al. 2002). Posteriorly, there is another membrane, Descemet's membrane,

which is an acellular membrane that forms a protective border within the cornea.

Descemet's membrane is produced by the posterior endothelium and has been found to

be elastic with fine collagen fibrils. The membrane has a thin anterior zone next to the

stroma and two broad zones located in a posterior region next to the endothelium

(Hayashi et al. 2002). The innermost cornea is lined with one layer of flattened cells

called the corneal endothelium. Mitosis takes place in immature animals within the

endothelium usually, but in older individuals the ability to divide is lost. In adult eyes,

the surface of the corneal endothelium has small microvillae and pores spotted on it with









hexagonally shaped cells. With age, there is a loss in hexagonal shape due to a decrease

in the density of the epithelium (Laing et al. 1976).

Even though the normal corneas of terrestrial and aquatic mammals are similar,

there are some differences. Studies conducted by Dawson et al. (1987) with the corneas

of cetacean species found few differences from the corneas of terrestrial mammals.

However, a more current study conducted by Miles (2002, unpublished) found several

differences between the corneas of marine and terrestrial mammals. Miles found through

histological examination and three-dimensional reconstruction that cetacean corneas were

thinner centrally than that of terrestrial mammals. Miles also found that cetacean corneas

lacked a Descemet's membrane. It was postulated that this membrane was absent to

allow the cornea to change its curvature to enhance refractive capabilities. Additionally,

Miles found that some species of cetaceans lacked a Bowman's membrane (Kogia

breviceps and Kogia simus) while other species had a thick, prominent Bowman's

membrane (Tursiops truncates and Globicephela macrorhyncus). Collins and Collins

(2000) found that the mean density of the anterior epithelium was thicker in aquatic

mammals than that of terrestrial ones. These variations are most likely due to differences

in environments. An additional study found that sea otters (Enhydra lutris) have an

extensively developed anterior epithelium. It is believed that the epithelium has

developed this way as an adaptation to the salinity of the aquatic environment in which it

lives (Murphy et al. 1990).

An investigation into the corneal anatomy of the Florida manatee conducted by

Samuelson et al. (1997) found blood vessels consistently present in eight eyes that were

examined histologically. The number, size, and depth of the vessels varied between the









eyes. In all the specimens examined, no signs of infection or injury were present. This

was a very unusual finding because normal mammalian corneas are avascular

(Samuelson 1999; Klintworth 1991). The presence of blood vessels in the cornea is

usually considered a pathological condition associated with infection or injury

(Klintworth 1991).

Angiogenesis

Angiogenesis is the biological process in which blood vessels form from existing

vasculature (Nicosia and Villaschi 1999; Battegay 1995, Folkman 1995; Schultz and

Grant 1991). Angiogenesis occurs in almost every tissue and is a necessary process that

is critical in normal conditions such as development, reproduction, healing of wounds,

bone repair, ischemic heart disease, ischemic peripheral vascular disease, tumor growth

and metastasis, and diabetic retinopathy (Battegay 1995; Adamis et al. 1994).

Angiogenesis can also occur under pathological conditions such as cancer, rheumatoid

arthritis, trauma, or infection (Schultz and Grant 1991).

Early vessel formation occurs through a process called vasculogenesis, in which

endothelial cells separate and grow in an avascular tissue, and then unite to form a

primitive tubular network. This early network of vessels includes the aorta and major

veins. Interconnected branching patterns of vessels form through angiogenic remodeling

that are characteristic of mature vasculature. At this stage, endothelial cells join together

securely with supporting cells to form mature vessel walls (Yancopoulos et al. 2000).

New blood vessels form after the first capillary tubes have developed through

angiogenesis (Nicosia and Villaschi 1999). These vessels sprout into previously

avascular tissue (Yancopoulos et al. 2000) as a result of endothelial cell migration and

proliferation, and then further develop (Nicosia and Villaschi 1999). This process is









referred to as angiogenic sprouting and is responsible for vascularizing some tissues

throughout normal development (i.e., neural tube and retina) (Yancopoulos et al. 2000).

Mitosis of endothelial cells increases the sprout and a small lumen develops due to the

curve within each cell. Eventually a loop is formed and blood flow occurs when two

hollow sprouts join at the ends. The process of angiogenesis is complete when pericytes

travel within the newly formed loop structure (Schultz and Grant 1991).

Many factors, including cytokines, control the establishment and alteration of blood

vessels. Perhaps the most important determinant in vessel formation is vascular

endothelial growth factor, VEGF. It is distinguished for its capacity to cause vascular

leakage and permeability, in addition to its capacity to promote vascular endothelial cell

production (Yancopoulos et al. 2000). VEGF is a cytokine that is an endothelial cell

mitogen and heparin-binding, dimeric protein (Miller et al. 1994; Klagsbrun and

D'Amore 1991). This cytokine is necessary in initiating the formation of immature

vessels during vasculogenesis or angiogenic sprouting that occurs in development and in

adults (Yancopoulos et al. 2000). VEGF is able to stimulate endothelial migration,

proliferation, and proteolytic activity and has a signal peptide that is secreted through

numerous pathways. VEGF is made by many different cells, including endothelial cells,

smooth muscle cells, and fibroblasts (Nicosia and Villaschi 1999). Mesenchymal and

stromal cells also secrete VEGF (Beck and D'Amore 1997).

The VEGF gene is composed of eight exons that are separated by seven introns.

Alternative splicing of the VEGF gene has produced four isoforms, including VEGF121,

VEGF189, VEGF206, and VEGF165 (Nicosia and Villaschi 1999). Four VEGF-related

genes have been discovered (VEGF-B, VEGF-C, VEGF-D, and placenta growth factor),









and they produce glycosylated dimers that are secreted after cleavage of their signal

peptide. Two classes of high-affinity binding sites on endothelial cells have been found

for VEGF with the molecular mass of these binding sites of 180-200 kDa (Ortega et al.

1999). The tyrosine kinases primarily expressed by the endothelium, Flt-1 (VEGFR-1)

and Flt-1/KDR (VEGFR-2), are receptors for VEGF. The secretion of VEGF is

noticeably stimulated by hypoxic conditions (Nicosia and Villaschi 1999).

Both VEGFR-1 and VEGFR-2 bind VEGF with a high affinity and have seven

immunoglobulin-like domains, one transmembrane region, and a tyrosine kinase

sequence that is interrupted by a kinase insert domain. There are two promoters of

VEGFR-1 and VEGFR-2 that contain a 5' flanking sequence that is needed for

endothelial specific expression. An additional receptor for VEGF is neuropilin-1. It

modulates the interaction between VEGF and VEGFR-2 (Ortega et al. 1999).

The expression of VEGFR-1 in naive cells mediates cell migration. Both VEGF

receptors are tyrosine phosphorylated. However, the phosphorylation of VEGFR-1 is not

required for function in vivo. Adaptators of the Src family, including Fyn and Yes, are

phosphorylated in response to VEGF/VEGFR-1 but not to VEGF/VEGFR-2 interactions.

In contrast, phosphorylated VEGFR-2 links with Shc, Grb2 and Nck, but also with

phosphatases of SHP-1 and SHP-2, and initiates the phosphorylation of the MAPK

cascade through the stimulation of Raf VEGF has also been found to induce the tyrosine

phosphorylation and recruitment of focal adhesion kinase (Ortega et al. 1999).

Angiogenesis can occur in endothelial cells during events such as tumor

progression, diabetic retinopathy, and rheumatoid arthritis. Local angiogenesis may be

due to the release of soluble mediators by the involved tissues. At this point there is a









switch in the phenotype of the quiescent endothelial cell to an activated phenotype. The

endothelial cells are then able to respond to mitogenic signals (Ortega et al. 1999). There

is a larger increase in the upregulation of genes modulated by VEGFR-2 than with

VEGFR-1. These induced genes include growth factors, protease inhibitors, receptors,

transcription factors, and actin-binding proteins (Yang et al. 2002). At this point,

mitogenic growth factors are released and this changes the activated phenotype to an

angiogenic phenotype. Interactions between angiogenic growth factors and their

receptors provide the signals for cell migration, proliferation, and differentiation to form

new vessels (Ortega et al. 1999).

The normal cornea is usually an avascular structure. If blood vessels are present

within the cornea, this is referred to as corneal angiogenesis or vascularization and is

considered a pathological condition that can affect visual acuity. The causes of corneal

vascularization are varied and in humans can include infections (such as trachoma and

herpes), immunologic processes, alkali bums, toxic and nutritional deficiencies, and the

wearing of contact lenses. In experimental settings, corneal vascularization can be

induced through chemical, microbiological, and physical injuries (such as burns or

implantation of pellets containing a wide range of chemicals), nutritional deficiencies,

and immunological reactions (Klintworth 1991).

Rabbits have often been used to study experimentally induced corneal

vascularization. After an injury of some type has occurred in the cornea, cell migration

and mitosis are involved in healing of the epithelial corneal surface. If abrasion occurs,

cell migration takes place almost immediately. In order to return the epithelium to its

usual thickness, cell division follows. Cell migration can heal small injuries within 24









hours. If a larger portion of the anterior epithelium is injured, the limbus is the main

source for production of epithelial stem cells. Endothelial cells are not able to respond as

quickly to cell loss. In the absence of new cell formation, the presence of too few

endothelial cells may result in edema, which can result in corneal disease (Samuelson

1999).

The stromal edema that occurs with corneal vascularization aids in the growth of

blood vessels in the cornea and this is an inflammatory process. There are many

inflammatory mediators that are involved in this process including histamine, serotonin,

ADP, prostaglandins, many cytokines, free radicals, and immune aggregates. However,

several factors can help reverse the process including epinephrine, cortisone, and some

drugs such as antihistamines and glucocorticoids (Klintworth 1991).

When an injury of the cornea occurs, stroma edema takes place and neutrophils

move to the source of the injury to release enzymes and chemotaxic agents. Epithelial

wound healing begins within one hour of the injury and wing layer cells flatten and slide

next to the injury site in order to cover the area within 24 to 96 hours. The epithelium

returns to its full thickness by the action of basal stem cells of the limbus. Proliferation

of the stromal fibroblast takes place after 24 hours (Ahmadi and Jakobiec 2002).

Leukocytes move in the extravascular space within 24 hours and move through the

corneal stroma to the site of injury. The first new vessels appear after 27 hours.

Advancing new vessels remain permeable to fluid, plasma proteins, and smaller proteins.

Forming corneal blood vessels do not have firm intercellular complexes; whereas, well-

formed corneal vessels in advanced stages of wound healing do not leak. Once new

vessels have developed, the basal lamina breaks down allowing the endothelial cells to









move into the extracellular matrix for further vessel formation. Progression of these

endothelial cells at the end of the developing new blood vessels extends towards the

damaged area (Klintworth 1991). The movement of endothelial cells, which elongate and

form solid sprouts developing lumen, follows this. Two sprouts form a loop and blood

flows, and the process is repeated from the top of the loops (Pepose and Ubels 1992).

New blood vessels can emerge within 33 hours after injury. A network of vessels is

formed that becomes linked and these vessels then begin to circulate (Klintworth 1991).

Objectives

This study will examine several aspects of corneal vascularization in the Florida

manatee, Trichechus manatus latirostris. The purpose of the first part of this study is to

describe more clearly corneal vascularization by examining the architecture through

three-dimensional reconstruction in order to find possible patterns in size, distribution,

and location of blood vessels that may be related to age, gender, and environment.

Secondly, immunohistochemistry will be performed to determine if there is a presence or

absence of vascular endothelial growth factor (VEGF) receptors-1 and -2 in the corneas

of two manatees varying in age. This will be done in order to try and further explain

what is occurring in the corneas at a biochemical level.














CHAPTER 2
CORNEAL VASCULARIZATION IN THE FLORIDA MANATEE AND THREE-
DIMENSIONAL RECONSTRUCTION

Introduction

The eye is an amazing organ that allows organisms the ability to interpret light-

based information about their environment. Normal ocular functions are, in part,

supported by temporary and permanent vascular beds found throughout the developing

and mature eye. During embryonic development, temporary vascular beds are associated

with the lens. Posteriorly, the lens is supplied with blood from the hyaloid artery system

(Drohan et al. 2002), which consists of an extensive arterial, arteriolar and capillary

network throughout the vitreous body (Los et al. 2000). The hyaloid artery moves toward

the lens from the optic nerve head and provides a network of vessels that covers the

posterior side of the lens (tunica vasculosa lentis). The lens is also supplied anteriorly

with blood from vascular branches that form the pupillary membrane and are fed by long

ciliary arteries (Drohan et al. 2002) and are drained by vessels in the area of the future

ciliary body (Los et al. 2000). The vessels in the tunica vasculosa lentis connect with the

pupillary membrane (Strek et al. 1993). Later in gestation the hyaloid artery involutes,

resulting in a transparent remnant called the canal of Cloquet that leads to the lens and

becomes avascular (Drohan et al. 2002). The regression of the hyaloid artery begins

when cellular components of the blood vessel walls break down (Los et al. 2000).

In addition to temporary vascular beds, permanent vascular beds can also be found

in the eye. The ciliary body is the site for one of the key vascular structures occurring in









the eye by the major arterial circle (MAC). In most terrestrial mammals, the ciliary body

is supplied by the MAC, which originates from combinations of the anterior ciliary artery

and the long posterior ciliary arteries (Morrison et al. 1987). The most extensive vascular

bed within the ciliary body occurs within the ciliary processes and is fed by arterioles

derived from the MAC. In rats and guinea pigs the blood vessels within the ciliary

processes are concentrically parallel and travel posteriorly emptying into the choroidal

veins, forming a vascular pattern that is similar to that of primates (Morrison and Van

Buskirk 1984). The ciliary process in cats and dogs is supplied by one arteriole that also

travels posteriorly and sends out capillary branches that drain outwardly into venous

sinuses. Sheep, goats, pigs, and cows have ciliary processes that receive blood from

many arterioles and veins that empty into the choroidal circulation (Morrison et al. 1987).

In a study conducted by Miles (2002) examining the angioarchitecture of the ciliary

body of different whale species, similar vascular patterns were observed in the ciliary

body of all species. It was found that a single arteriole located in the apical area of the

ciliary body supplied the capillary bed in the bottlenose dolphin. In three other species

examined, Globicephela, K. simus, and K. breviceps, several arterioles supplied the

capillaries of each ciliary process. Other vessels in the ciliary body of the whale species

examined were hypothesized to aid in venous outflow, including many small vessels

located at the base of the processes, which collateralized with vessels responsible for

collecting aqueous humor. In addition to the smaller veins, larger veins were also found

that aided in the movement of blood of the iridal and ciliary processes to the choroidal

venous outflow system.









The ciliary body vasculature has also been investigated in the Florida manatee by

Natiello (1999). It was found that the ciliary process in the manatee contains many

capillaries, partly similar to that observed in cetaceans and pinnipeds. Abundant blood

vessels were observed on the outside folds that lead into the choroidal vasculature. The

major arterial circle (MAC), found at the base of the iris, fed the ciliary body's

vasculature. The anterior ciliary artery and two long posterior ciliary arteries formed the

MAC. Venous outflow of the ciliary body consisted of a unique bifurcated system,

suggesting an interesting vascular dynamic associated with aqueous humor production.

The choriocapillaris, located in the posterior uvea, provides another example of a

permanent vascular bed in the eye, especially for animals with a tapetum lucidum. Blood

vessels penetrate the tapetum at right angles and are parallel to the incident light in order

to supply the choriocapillaris. The blood vessels go straight through the tapetum so as

not to interfere with light reflection. In carnivores such as the dog, cat, ferret, and grey

seal, the tapetum under the choriocapillaris is cellular (tapetum cellulosum) and indents

into the RPE so that the tapetum has a flat, reflective surface (Braekevelt 1986). With

some species of animals that have a tapetum fibrosum, such as the sheep, the

choriocapillaris is not indented into the RPE (Braekevelt 1986b).

The vasculature of the tapetum in most domestic species consists of medium-sized

vessels in the dorsal section of the choroid. Variations in the reflective color of the

tapetum can result from choroidal blood vessels. Many small vessels pass through the

tapetal layer and form the choriocapillaris, which is a single layered capillary bed. The

lumen of the choriocapillaris is wide, which allows red blood cells to go through easily

(Samuelson 1999). In the posterior pole of the choriocapillaris, alternating feeding









arterioles and draining venules are found (Alm 1992). The area including the RPE,

choriocapillaris and Bruch's membrane is responsible for a number of functions

including moving metabolites to photoreceptors, stabilization of the structural design of

outer segments of the photoreceptors, storing vitamin A for visual pigments, and

phagocytosis and lysosomal break down of photoreceptor outer segments (Braekevelt

1983).

Bhutto and Amemiya (2001) conducted a study to investigate the architecture of

the rat choroid by using scanning electron microscopy of vascular corrosion casts. In this

study, it was found that the choriocapillaris viewed from the retinal side formed a

network of capillaries that ranged in diameter and appeared to resemble a dense

honeycomb pattern of vessels. Additionally, an irregular pattern of vessels was also

observed. The two different vascular patterns in the choriocapillaris were evenly

distributed throughout the choroid except in the peripheral area. When viewed from the

posterior side, the choriocapillaris' vascular pattern resembled a palm-like pattern that

ended at the ora serrata. Of particular interest is that vision was not impeded even with a

dense network of blood vessels forming the choriocapillaris.

Permanent vascular beds can also be found in the mammalian retina. In order for

normal function to occur in a retina over 160 [Lm in thickness, ocular blood flow must be

strictly maintained. The central retinal vein controls blood flow from the retina and is

formed from retinal veins meeting on the optic nerve head. The central retinal vein has

two stems at the disc that connect while they are still in the optic nerve (Ruskell 1998).

Retinal perfusion for most mammals is carried out by the choroidal vessels, which are

found between the outer retina and sclera (Steinle et al. 2000). The vascular needs of the









outer retina are also met by choroidal circulation. Species that have a highly vascularized

inner retina usually have a wide range of vessel patterns and these vascular patterns are

determined by density of neural cells in the ganglion cell layer forming an area centralis

(Dunlop et al. 1997). Capillaries within the retina have been found to be very thin in

diameter and contain fewer corpuscular elements than the choriocapillaris (Ninomiya and

Kuno 2001). Major blood vessels in the retina avoid areas of high neural cell density.

By doing this, visual acuity is not negatively affected and many species of animals

display a similar retinal vascular pattern (Dunlop et al. 1997).

In some snake and lizard species, there is a unique ocular structure called the

spectacle or brille, which is a fused eyelid that forms a clear covering over the cornea. In

some species of snakes the spectacle is about the same size as the cornea, and in other

species the spectacle is much larger than the cornea and can even cover a portion of the

head (Sivak 1977). Walls (1942) originally believed that a sebaceous material entered

between the spectacle and cornea to act as a lubricant. Walls also thought that if a

spectacle had a flattened curvature, then the refractive involvement of the cornea would

be decreased. Sivak (1977) found that the spectacle essentially replaced the cornea as a

refractive component and that the spectacle plays the major refractive role.

Mead (1976) injected microsilicone into the vessels of the spectacle and found that

the spectacle was indeed vascularized, having vessels that transversed the stromal layer

of the spectacle. Visualization, using a slit lamp at a power of 32 X, was possible and

showed small vessels with transparency in the walls of the vessels. The vessels were

only visible because of the reflection of red blood cells that were circulating through

them. It was also found that there was a vertical arrangement of vessels in the spectacle









of members of some snake families. The most interesting finding of this study was that

the spectacle maintained a vascular pattern and still retained a high level of transparency.

The normal healthy cornea, unlike other regions such as the retina and

choriocapillaris, is an example of an ocular structure that is normally avascular.

However, if blood vessels are present within the cornea, this is referred to as corneal

angiogenesis or vascularization and is considered a pathological condition that affects

visual acuity. The causes of corneal vascularization are varied and in humans can include

infections (such as trachoma and herpes), immunologic processes, alkali burns, toxic and

nutritional deficiencies, and the wearing of contact lenses. Except for contact lenses,

identical causes have been observed in domestic species. In experimental settings,

corneal vascularization can be induced similarly through chemical, microbiological, and

physical injuries, nutritional deficiencies, and immunological reactions (Klintworth

1991).

Rabbits have often been used to study experimentally induced corneal

vascularization. After an injury of some type has occurred in the cornea, cell migration

and mitosis are involved in healing of the epithelial corneal surface. If abrasion occurs,

cell migration takes place almost immediately. In order to return the epithelium to its

usual thickness, cell division follows. Cell migration can heal small injuries within 24

hours. If a larger portion of the anterior epithelium is injured, the limbus is the main

source for production of epithelial stem cells. Endothelial cells are not able to respond as

quickly to cell loss. In the absence of new cell formation, existing endothelial cells

compensate for the loss. Too few endothelial cells are unable to provide adequate

deturgescence, resulting in corneal edema (Terry and Ousley 2003; Vemuganti et al.









2002). The cornea is a relatively dehydrated tissue being made up of 75 to 85% water.

Deturgescence is the state of dehydration in the cornea and is carried out by the

endothelium and epithelium (Samuelson 1999).

The stromal edema that occurs with corneal vascularization aids in the growth of

blood vessels in the cornea and this is an inflammatory process. There are many

inflammatory mediators that are involved in this process including histamine, serotonin,

ADP, prostaglandins, many cytokines, free radicals, and immune aggregates (Klintworth

1991). Angiogenesis is the biological process in which new blood vessels form from

existing vasculature (Nicosia and Villaschi 1999; Battegay 1995; Folkman 1995; Schultz

and Grant 1991). Angiogenesis occurs in almost every tissue and is a necessary process

that is critical in normal conditions such as development, reproduction, healing of

wounds, bone repair, ischemic heart disease, ischemic peripheral vascular disease, and

diabetic retinopathy (Battegay 1995; Adamis et al. 1994). Pathological conditions in

which angiogenesis can occur include cancer, rheumatoid arthritis, trauma, tumor growth

and metastazation, or infection (Schultz and Grant 1991).

Early vessel formation occurs through a process called vasculogenesis, in which

endothelial cells separate and grow in an avascular tissue, and then unite to form a

primitive tubular network. The aorta and major veins are examples of this early network

of vessels. Interconnected branching patterns of vessels that are characteristic of mature

vasculature form through angiogenic remodeling. At this stage, endothelial cells join

together securely with supporting cells to form mature vessel walls (Yancopoulos et al.

2000). New blood vessels form after the first capillary tubes have developed through

angiogenesis (Nicosia and Villaschi 1999). These vessels sprout into previously









avascular tissue as a result of endothelial cell migration and proliferation, and then further

develop (Yancopoulos et al. 2000; Nicosia and Villaschi 1999). This process is referred

to as angiogenic sprouting and is responsible for vascularizing some tissues throughout

normal development (i.e., neural tube and retina) (Yancopoulos et al. 2000). Mitosis of

endothelial cells increases the sprout and a small lumen develops due to the curve within

each cell. Eventually a loop is formed and blood flow occurs when two hollow sprouts

join at the ends (Schultz and Grant 1991).

When an injury of the cornea occurs it is accompanied with stromal edema and

eventually corneal angiogenesis as neutrophils move to the source of the injury to release

enzymes and chemotaxis agents. Epithelial wound healing begins within one hour of the

injury and wing layer cells flatten and slide next to the injury site in order to cover the

area within 24 to 96 hours. The epithelium returns to its full thickness by the action of

basal stem cells of the limbus. Proliferation of the stromal fibroblast takes place after 24

hours (Ahmadi and Jakobiec 2002). Leukocytes move in the extravascular space within

24 hours as well as traveling through the corneal stroma to the site of injury. The first

new vessels appear after 27 hours. Advancing new vessels remain permeable to fluid,

plasma proteins, and smaller proteins. Forming corneal blood vessels do not have firm

intercellular complexes, whereas well-formed corneal vessels in advanced stages of

wound healing do not leak. Once new vessels have developed, the basal lamina breaks

down at specific sites, allowing the endothelial cells to move into the extracellular matrix

for further vessel formation. Progression of these endothelial cells at the end of the

developing new blood vessels extends toward the damaged area (Klintworth 1991). The

movement of endothelial cells, which elongate and form solid sprouts developing lumen,









follows this. Two sprouts form a loop and blood flows, and the process is repeated from

the top of the loop (Pepose and Ubels 1992). New blood vessels can emerge within 33

hours after injury. A network of vessels is formed that become linked and these vessels

then begin to circulate (Klintworth 1991).

An investigation into the corneal anatomy of the Florida manatee (Trichechus

manatus latirostris) conducted by Samuelson et al. (1997) consistently found the

presence of blood vessels within the corneas of eight eyes that were examined

histologically. The number, size, and depth of the vessels varied among eyes. In all the

specimens examined, no signs of infection or injury were present. The interference of

corneal vascularization on the manatee's ability to see, i.e., visual hindrance, had not

been assessed.

Within the last century, manatee eye anatomy and vision has not been studied

extensively. Early studies conducted by Walls (1942) suggested that manatees have

limited visual capabilities and acuity. Other research conducted supported Walls' theory

that manatee eyes were adapted for low light conditions and contained very few ganglion

cells, as well as no mechanism for accommodation, thus limiting vision (West et al. 1991;

Piggins et al. 1983). However, behavioral research conducted by Hartman (1979)

suggested that vision is the sensory mechanism in which manatees rely on the most for

receiving information about their environment. More recent research carried out by Mass

et al. (1997) found that the ganglion-cell distribution in the retina did not appear to be

even and was diverse across the retina with higher ganglion cell densities near the center

of the retina suggesting a limited visual resolution. These anatomical findings were









consistent with behavioral studies conducted by Bauer et al. (1999) and Colbert et al.

(1999) where they also found that manatees had a limited visual ability.

We propose as our central hypothesis that corneal vascularization in the Florida

manatee affects the vision of these animals. The amount of vascularity from collected

specimens will be determined. The patterns) and amount of corneal vascularization

from previously collected specimens will be determined by light microscopic three-

dimensional reconstructions of the vascular elements found in the corneas from 22

deceased manatees. Correlations of the amount of vascularity with factors including age,

gender, time of year, and location where the animals were found will be made. This

study will lay the foundation for determining the extent to which corneal vascularization

exists in manatees and for understanding the potential impact of this unique condition on

their ability to survive.

Materials and Methods

Specimen Collection

Florida manatee eyes were collected from Dr. Gordon Bauer at New College in

Sarasota, Florida. Dr. Bauer received the eyes from Florida Marine Research Institute

(FMRI) in St. Petersburg, Florida. Twenty-six eyes from 22 individuals were used and

the eyes were taken from necropsied animals over a period of 5 years. Each eye was

labeled with an identification number given to it by FMRI. Information including

gender, size, county where the animal was found, and time when the animal was found

was recorded. Manatee eyes were placed in fresh 10% neutral buffered formalin solution

(Luna 1968).









Preparation of Eyes

All extraneous tissue surrounding the globe was removed using a scalpel (see

figure 2-1). Individual corneas were measured in millimeters from dorsal to ventral and

medial to lateral sections. An additional measurement was taken from the cornea to the

optic nerve. After measurements were taken and recorded, the corneas were removed

from the globe by cutting 1 mm from the limbus into the sclera with a sharp razor. A

notch was cut into the limbus at 12:00 in the dorsal section of each cornea in order to

know orientation during histological examination. The remainder of the eye was stored

and the cornea was placed in a tissue cassette, labeled and then returned to the 10%

neutral buffered formalin solution until processing.

Histology

In preparation for embedding, the corneas were dehydrated by placing them in the

following ethyl alcohols for an hour each: 70%, 80%, two changes of 95%, followed by

three changes of 100%. All dehydration steps were prepared by using ethyl alcohol. The

tissue was then cleared with two changes of xylene. The tissues were infiltrated with two

changes of Fisher Tissue Prep at 570 Celsius. The tissues were then embedded using

Fisher Tissue Prep T565 in metal molds and allowed to harden.

Using a Reichert-Jung 2030 microtome, the embedded corneas were cut in serial

dorsal/ventral sections at a thickness of 10 microns. The tissue ribbons were floated out

on a 600 Celsius water bath and then placed on 3 x 1" x 1 Superfrost microscope slides

(Fisher Scientific, Pittsburgh, Pennsylvania) with three sections per slide. Each slide was

labeled with the identification number of the animal, order of section, and orientation of

tissue. Approximately 35 slides and 105 sections per eye where cut. After cutting was









completed, the tissues were placed in a slide rack and into an oven at 600 Celsius

overnight to dry. After the slides had dried completely, they were deparaffinized and

then stained using a Masson's Trichrome Stain (Humason 1972). When staining was

completed, the slides were coversliped using synthetic xylene based mounting media, Eu-

kitt.

Three-Dimensional Reconstruction

Three-dimensional reconstructions of corneal vasculature where carried out using

the Light Microscopy facility in the McKnight Brain Institute at the University of Florida.

Slides that had been histologically processed were placed on a Ludl-motorized scanning

stage of an AxioPlan 2 Zeiss Microscope. A Sony model DXC 970MD camera was used

to take individual pictures of each serial section in sequential order. Imaging Research

Incorporated (MCID) created the computer software used to produce the images.

Initially each section was digitized at a magnification of 25 X (2.5-Zeiss-Plan/NEO

Floar). The sections then were aligned based on landmarks (e.g., blood vessels or the

outline of the cornea) that were common with the previous section. The number of

images taken for each cornea varied from 16 to 48 sections (each section being 10

microns in thickness). Once all images per cornea were taken, the blood vessels were

outlined in red and the corneal tissue was outlined in yellow for each section using a

mouse driven color-coded screen cursor. The MCID program has an alignment feature

that allows the user to align the sections optically. Once the images of each cornea were

stacked and aligned, the MCID program then created a three-dimensional cube. The cube

could be rotated 360 degrees, as well as display the outlined structures (blood vessels and

corneal outline) within the stacked images that comprised each cornea. Other factors









including transparency, lighting, and viewing orientation could be manipulated allowing

for several images to be taken of each cornea. Due to these functions on the MCID

program, detailed maps of the spatial arrangement of blood vessels in the cornea could be

made. Three maps were made for each cornea that included a face-on view, a dorsal or

ventral view, and a lateral or medial view. For each map that was constructed, an image

was made that included both outlined features and an additional map was also made that

included only the blood vessels present in the cornea. The MCID program was also able

to calculate the volume and area of outlined features in each cornea, which allowed us to

analyze and measure the vascularization as a percent of the total volume of tissue

examined.

After examination of the maps constructed at 25 X, three additional corneal maps

were constructed at 200 X (20X-Zeiss-Plan/APO Chromat). Two of the corneas were

highly vascularized and one had a low percentage of vascularization. Each cornea was

divided into dorsal, ventral, medial and lateral quadrants. The same process was carried

out within each quadrant as previously described, but at 200 X instead of 25 X.

However, instead of outlining all the corneal tissue, only the stroma was outlined in order

to visualize vasculature that was passing from the stroma to the anterior epithelium. This

process of increasing the magnification and separating the stroma and anterior epithelium

was done in order to examine the vascular patterns of smaller, feeder vessels passing

between the two cornea layers, as well as determine the vascularization by these blood

vessels as a percent of the total volume. Additionally, the diameter of blood vessels at 25

X and 200 X was calculated in microns using a Nikon Labophot-2 microscope.









Statistics

One-way ANOVAs were carried out using the SAS computer program based on

blood vessel as a percent of the total volume in each cornea calculated for 20 individuals.

Four different parameters were analyzed including gender (male versus female), age class

(calf, juvenile, or adult), coast found (east versus west), and season found (Spring,

Summer, Fall, or Winter).

Results

Anatomical Description

The stroma of the cornea stained a light blue to medium blue color. Within the

stroma, blood vessels can be seen that are either red in color due to red blood cells or lack

color and can be seen because of their hollowed shape (Figures 2-2 through 2-6). The

border surrounding the blue stroma is stained a deep red to purple and is the anterior

epithelium. In Figures 2-2 and 2-3 vessel penetration between the two layers can be

easily observed. As Figures 2-2 through 2-6 illustrate, the vessels found in the cornea

vary in shape, size, and location. Diameters of these vessels were measured at 25 X in

microns using several specimens including MNW-9016, MNW-9114, MNW-9022, and

MSE-9105. The average vessel diameter for MNW-9016 was 60 am, 75 am for MSW-

9114, 82.1 am for MNW-9022, and 121.4 am for MSE-9105. The overall average for

vessel diameter at 25 X was 84.6 am.

Orientation (left eye, right eye, dorsal, ventral, lateral, or medial) of the manatee

cornea as used for three-dimensional reconstruction can be viewed in Figure 2-7 and the

actual reconstructions of manatee corneal tissue at 25 X are represented in Figures 2-8

through 2-33. In Figures 2-8 through 2-33 A, a face-on view of both the corneal tissue

and vasculature can be seen. The tissue present represents approximately one-fourth of









the entire cornea. If the corneal tissue is removed, then only the vessels of the cornea can

be viewed (Figures 2-8 through 2-33 B). This allows for better visualization of the

vasculature present without it being obscured by corneal tissue. Figures 2-8 through 2-33

C represent a dorsal or ventral view of each tissue with the vessels present. The view is

dorsal or ventral depending on where most vasculature was observed through three-

dimensional reconstructions. In Figures 2-8 through 2-33 D, only the vasculature within

the cornea can be seen, which allows observations to be made of vascular patterns as they

pass through the tissue. In Figures 2-8 through 2-33 E and F, the same applies as with

Figures 2-8 through 2-33 C and D. However, the views represent a medial or lateral

(horizontal) perspective of the tissue and vasculature (Figures 2-8 through 2-33 E) or just

the vasculature within the corneal tissue (Figures 2-8 through 2-33 F). If left or right

orientation of an individual eye was not determined, then the medial or lateral perspective

was labeled horizontal instead.

The three-dimensional reconstructions of the corneal vascularization have the

appearance of spaghetti-like entangled networks. Tissues with less vascularization make

up the majority of the samples (Figures 2-10, 2-11, 2-14, 2-16, 2-18 through 2-30, and 2-

33) and appear to have randomly scattered vessels across the area of the tissue. No

apparent patterns can be seen with the vasculature and the vessels were found throughout

all parts of the cornea. Corneal tissues that are more highly vascularized (Figures 2-8, 2-

9, 2-12, 2-13, 2-15, 2-17, 2-31, and 2-32) have greater amounts of blood vessels that can

be seen in the reconstructions and cover more area of tissue. These tissue specimens

resemble the less vascularized tissue specimens in that they appear random in nature and

no apparent pattern can be observed.









It should be noted that the most highly vascularized cornea (Figure 2-17) had an

interesting vessel formation. This cornea appeared to contain a jumble of vessels that had

no discernable pattern. Vessels were seen criss-crossing throughout the entire area

represented by the three-dimensional reconstruction. Additional corneas of interest were

those of the fetus (Figures 2-32 and 2-33). A network of vessels was observed in both

corneas of the fetus, possibly indicating a developmental process. The vascularization in

the fetal corneas resembled that of the other corneas in that the vessels appeared random

in nature with no apparent pattern.

Additionally, three corneas were examined at a higher magnification of 200 X. The

tissue chosen for these examinations included MNW-9016, MSW-9114, and TM-8619

(left) because MNW-9016 and MSW-9114 were the two most highly vascularized

corneas and TM-8619 (left) was one of the least vascularized corneas. The area of

interest at 200 X was along the junction between the stroma and anterior epithelium of

the cornea. On the histologically processed slides, the stroma and anterior epithelium

were very similar in appearance, however; red blood cells were observed in the anterior

epithelium and not in the stroma. Additionally, the anterior epithelium stained a dark red

color and the stroma usually stained a light purple or blue color. The area between the

stroma and anterior epithelium was of interest because vessels not only came close to the

anterior epithelium via the stroma, but also penetrated into it without any apparent

connective tissue elements (see Figures 2-2 through 2-6).

Three-dimensional reconstructions were also carried out with these corneas at 200

X in order to examine the smaller, feeder vessels present in the stroma and anterior

epithelium. Four quadrants (dorsal, medial, ventral, and lateral) where mapped out for









each cornea and vasculature within each quadrant was examined. A larger number of

vessels were observed in the tissues examined at 200 X (Figures 2-34 through 2-41). The

figures were arranged in a similar pattern as with those at 25 X. However, instead of

being six images representing each cornea from various angles, there are eight images of

MNW-9016 (Figures 2-34 and 2-35), eight images of MSW-9114 (Figures 2-36 and 2-

37) and sixteen images of TM-8619 (Figures 2-38 through 2-41) because only one

quadrant was reconstructed for each cornea at 25 X and two to four quadrants were

reconstructed at 200 X. The vasculature at 200 X is similar to that at 25 X in that the

vessels do not form patterns and seem randomly distributed throughout the corneal area.

However, the most obvious difference between 200 X reconstructions versus 25 X

reconstructions is that the vascular patterns at 200 X occupy more of the volume and are

far more noticeable. Additionally, vessels could be observed penetrating the anterior

epithelium from the stroma with the higher magnification reconstructions. This level of

detailed vascularization could not be detected at 25 X. The diameters of blood vessels at

200 X in MNW-9016, MSW-9114, and TM-8619 (L) were also measured. The average

vessel diameter for MNW-9016 was 29.3 [am, 27.1 [am for MSW-9114, and 26.8 [am for

TM-8619 (L). The overall average for vessel diameters at 200 X was 27.3 am.

Morphometry

Using the MCID computer program implemented on a M5 system (Imaging

Research, Inc.) at the Light Microscopy Facility in the McKnight Brain Institute,

vascularization as a percent of the total volume was calculated for each manatee cornea

examined. The area of the corneal tissue covered in the three-dimensional reconstructed

images at 25 X represented approximately one-fourth of the total corneal area present.

Photographing and imaging the entire corneal tissue area at 25 X was not possible due to









screen size limitations. The reconstructed image that was performed in each specimen

represented the most vascularized area in the cornea. However, 25 X was selected

because this magnification allowed the maximum area to be represented for each

specimen while still being able to observe vasculature.

The vasculature volumes calculated for each manatee cornea ranged from 0.1% to

1.2 %, with average blood vessel volume coverage of 0.3%. The following values for

each manatee cornea, according to its identification number (see Table 2-1), represent the

volume of vascularization based on the total area of the cornea examined at 25 X.

Additional volumes were calculated for three corneas that were further examined at 200

X (MNW-9016, MSW-9114, and TM-8619 left) in order to better view the vasculature of

smaller, feeder vessels between the stroma and anterior epithelium. The following values

for each manatee cornea, according to its identification number (see Table 2-2), represent

the volume of vascularization based on the total area of the cornea examined in four

separate quadrants at 200 X (dorsal, ventral, medial, and lateral or side quadrants if the

eye was not labeled left or right).

Statistics

The statistics were calculated using the SAS computer program by means of a one-

way ANOVA. The effect of gender (male versus female), age class (calf, juvenile, or

adult), coast found (east versus west), and season found (Spring, Summer, Fall, or

Winter) based on blood vessel as a percent of the total volume (which was calculated

using the MCIP computer program) of tissue was examined in 20 individuals. The

unknown and fetus samples were not used in calculating statistics because no additional

information was available for these samples. A one-way ANOVA was used because little

variation was seen among the separate factors analyzed. In the case of three individuals









where both eyes were examined, the average between the two eyes was taken and used as

the volume value for comparing within the statistics (MSW-9137=0.3%, TM-8619=0.2%,

and TM-8630=0.2%). There was no significant gender effect found (Fi,18=1.34,

P=0.2622); no significant age class effect found (F2,17=0.82, P=0.4588); no significant

coast effect found (Fi,1s=0.07, P=0.7996); and no significant season effect found

(F3,16=0.30, P=0.8259).

Table 2-1. Information on manatee tissue used and volume of vasculature at 25 X
Manatee I.D. % Date Season Length Class Sex County
number volume found found (cm) found


MNW-9016 0.7 9-28-90 Fall 320 Adult M Sarasota
MNW-9022 0.6 11-16-90 Fall 147 Calf F Manatee
MNW-9111 0.1 4-8-91 Spring 264 Adult M Dixie
MNW-9116 0.1 6-6-91 Summer 210 Calf M Collier
MSE-9101 0.5 1-13-91 Winter 335 Adult M Broward
MSE-9105 0.6 2-24-91 Winter 323 Adult M Palm
Beach
MSE-9107 0.3 4-12-91 Spring 232 Calf M Broward
MSE-9115 0.5 8-9-91 Summer 292 Adult M Martin
MNE-9131 0.1 10-31-91 Fall 348 Adult M Duval
MSW-9114 1.2 5-22-91 Spring 340 Adult F Lee
MSW-9131 0.1 9-14-91 Fall 292 Adult M Charlott
MSW-9137 0.3 10-31-91 Fall 289 Adult M Okeech-
obee
MSW-9141 0.3 11-28-91 Fall 252 Juvenile M Lee
MSW-9143 0.1 4-17-86 Spring 260 Juvenile M Lee
TM-8619 0.2 6-22-86 Summer 219 Calf F Collier
TM-8630 0.2 4-6-91 Spring 246 Juvenile F Volusia
UCF-9114 0.2 10-26-91 Fall 293 Adult F Brevard
UCF-9137 0.4 10-30-91 Fall 220 Calf M Brevard
UCF-9138 0.4 12-27-91 Winter 296 Adult F Brevard
UCF-9141 0.3 1-13-91 Winter 306 Adult M Brevard
Unknown 0.7 NA NA NA NA NA NA
Fetus 0.5 NA NA NA NA NA NA


e









Table 2-2. Volume of vasculature at 200 X
Manatee I.D. Dorsal Medial Ventral Lateral Side
number quadrant quadrant quadrant quadrant quadrant
MNW-9016 5.36% 5% 0.44% NA --
MSW-9114 1.28% -- 1.31% -- 2.8%
TM-8619 4.93% 0.13% 0.15% 4.59% --

Discussion

The three-dimensional reconstructions of the Florida manatee corneas produced

interesting results. In the 26 corneas examined from 22 individuals, vascular beds were

present in every cornea. This finding varies little from previous studies conducted by

Samuelson et al. (1997 and 1994). The vascularization occurred primarily in the stroma

with some extension of vessels into the anterior epithelium. There were no large vessels

found in the cornea and vascularization was not heavily extensive throughout. The

vascularization that was present was found throughout, but the area where vessels were

occurring most commonly varied among individuals and even between eyes from the

same animal. The general pattern of vascularization throughout the cornea was highly

irregular with criss-crossing vessels that extended through several layers in the stroma

and anterior epithelium. The reconstructions of vessels found at 25 X (see Figures 2-8

through 2-33) best resembles that of a "spaghetti-like network" with no specific design

(Figures 2-2 through 2-6).

No major differences in density or size were found in the general pattern of

vascularization reflecting a morphological pattern with few outliers. The average amount

of vascularization present in the corneas examined in this study was 0.3%. Only two

animals examined contained appreciably greater amounts of vascularization in their

corneas with 1.2% (MSW-9114, Figure 2-17) and 0.7% (MNW-9016, Figure 2-8). This

overall pattern in MSW-9114, with it being the highest vascularized cornea examined,









did not differ greatly from the other corneas studied. It only contained more vessels. The

only common factor between the two animals with the highest amount of vascularization

found was that they were adults. The animals were not found in the same year or season,

were different sexes, and were found in different counties. However, the counties where

the animals were found were in close proximity to each other; Lee and Sarasota counties

are only separated by Charlotte County on the southwest coast of Florida. Interestingly,

the third highest amount of vascularization recorded was a calf at 0.6% (MNW-9022),

suggesting that age of the animals does not play a critical role in determining the amount

of vascularization present in their corneas. However, this calf was found in Manatee

County, which is located directly north of Sarasota County. While, this finding may

indicate that some environmental influence found in the water in that general area may be

playing a role in the higher amounts of vascularization found in these corneas, another

individual (MSE-9105) with 0.6% of its cornea vascularized was found in Palm Beach

County, which is directly across the state on the southeast coast of Florida. Based on its

location, the influence of the environment may not play an important role in determining

vascularizing factors after all. Overall, Table 2-1 shows little variation in the percent of

vascularization found in the corneas of the manatees examined. Nonetheless, several

differences are seen in such factors as date found, age of the animal, size of the animal,

and location where the animal was found. Statistically, these differences were not found

to be significant.

An unexpected aspect of this study was the unique opportunity to study the cornea

of a manatee fetus. Reconstructions of the left and right corneas of this individual

showed that both contained vascularization, but with varying amounts. This interesting









developmental case allowed us to examine endogenous versus exogenous factors that

may affect vascularization. Interestingly, the irregular vascular patterns observed in the

other manatees were also observed in the fetus. This observation suggests that the

vascularization is stimulated in the animal during development and not through

exogenous factors such as the environment, especially when this animal was never

exposed to an aquatic environment outside that of its mother's uterus. The findings in the

fetus suggest that corneal vascularization in the Florida manatee may be an evolutionary

adaptation process. Other areas that could be investigated to further explore this

possibility include examining other vascular patterns throughout the body that may be

unusual and/or unique to the manatee. It is also possible that some change in organ

development occurs that alters vascular patterns. For example, a recently discovered

low-density receptor-related lipoprotein called Lrp5 has been found to provide dual

function in very different organ systems when there is a defect that occurs in it. The Lrp5

protein functions as a Wnt coreceptor. Wnt proteins help control many developmental

processes such as mesoderm induction, cell fate determination, limb patterning, joint

formation, and organogenesis. Lrp5 binds to Wnt proteins in osteoblasts and Lrp5 is

needed for normal deterioration of embryonic vasculature in the eye. Lrp5 and Wnt

proteins work in conjunction to regulate osteoblast proliferation and function, and eye

vascularization during late development and after birth. Recent studies have discovered

that Lrp5 becomes inactivated in human patients with osteoporosis-pseudoglioma

syndrome and is mutated in patients that have high bone mass syndrome. Mice were

used to further investigate phenotypes that occurred when the Lrp5 protein was deficient.

It was found that they displayed two phenotypes; a low bone mass density caused by a









reduction in bone formation and they retained their hyaloid vasculature in the eye

throughout life (Kato et al. 2002). Currently it is not known if manatees have an altered

Lrp5 protein. However, further investigations into correlations between the Lrp5 protein

and other vascular components may prove to be very insightful. Manatees do have a high

bone density so it may be possible that the corneal vascularization occurring is a by

product of their high bone density. However, it should be noted that no hyaloid

vasculature has been observed in the adult manatee eye. This may rule out the possibility

of an Lrp5 deficiency, but there are other aspects of the manatee genome that have not

been explored which may be causing the vascularization observed in the cornea.

Three corneas were reconstructed at a higher magnification of 200X (Figures 2-34

through 2-41), as opposed to 25X, to demonstrate angiogenic patterns of small blood

vessels and their association and penetration between the stroma and anterior epithelium.

This was a very interesting find because this penetration of vessels between layers does

not usually occur without the apparent presence of connective tissue (see Figures 2-2

through 2-6). By having been divided into quadrants based on its anatomical position,

essentially the entire cornea was evaluated. Vascularization was not uniform throughout

the quadrants of each individual examined (see Table 2-2). Vascularization was present

regionally and varied in location between the three corneas reconstructed. The area

covered by vessels present at this magnification was greater than that of vessel coverage

at the lower magnification. This was most likely due to the ability to better visualize and

trace smaller vessels at the higher magnification. Even though the vessels did cover more

area in the cornea at this magnification, they were smaller in size and helped demonstrate

the amount of vascularization. These vessels may be too small to interfere with vision.









As with the larger vessels observed at 25X, the smaller vessels at 200X did not have a

regular pattern. In the future, it may be of interest to reconstruct all corneas in this study

at 200X in order to determine if any repeating patterns do arise. However, the vascular

pattern at this magnification currently appears to be random in nature. The most

interesting aspect of this part of the study was the presence of the vessel penetration

between the stroma and anterior epithelium. To the best of our knowledge, this

occurrence has not been documented previously.

Our original hypothesis that blood vessels would interfere with vision was not

supported because the vessels observed did not appear to be large enough in either size or

amount to interfere with vision. Vascular beds throughout the eye occur under normal

conditions without interfering with light transmission. One such example is the

choriocapillaris. The choriocapillaris vascular structure is different from that observed in

the manatee corneal vascular structure in that the choriocapillaris is only a single layer of

broad to narrow small vessels (Bhutto and Amemiya 2001; Buggage et al. 1996), whereas

the corneal vasculature observed penetrated several layers and continued into adjacent

anterior epithelial cells as seen through tangential sections. However, a corrosion cast

study conducted on Wistar Kyoto rats investigating the choriocapillaris did find some

interesting vascular patterns that are comparable to those found in the manatee cornea.

The vascular pattern of the choriocapillaris in these rats viewed from the retinal side near

the posterior pole was a nonhomogeneous complex of capillaries that differed in diameter

size and was composed of a thick honeycomb pattern of vessels. However, near the

peripheral areas of the choriocapillaris, a more regular pattern was formed that was

elongated and palm-like in structure. This peripheral palm-like vascular pattern found in









the choriocapillaris terminated at the ora ciliaris retinae (Bhutto and Amemiya 2001).

Even though no regular vascular patterns were found in the peripheral sections of the

manatee cornea, the irregular patterns found in the peripapillary area of the

choriocapillaris appear to be similar to that found in the manatee corneal vascular

structure. Irregular vascular patterns in the choriocapillaris have also been found in the

rhesus monkey. It was found in these mammals that the posterior pole of the

choriocapillaris also had a nonhomogeneous structure consisting of distinct lobular

segments (Flower 1993).

The choriocapillaris and tapetum lucidum work in conjunction to enhance vision

for many species of animals. The vasculature of the tapetum in most domestic species

consists of medium-sized vessels in the dorsal section of the choroid. Many vessels pass

through the tapetal layer and form the choriocapillaris. The tapetum inserts among

branching vessels found in the choroid and the choriocapillaris (Samuelson 1999). The

vessels that pass through the tapetum and supply the choriocapillaris in species such as

the cat, dog, ferret, and grey seal, do so at right angles to the tapetum. Following this

pattern, the blood vessels travel parallel to the incident light, which interferes with the

reflective function of the tapetum as little as possible (Braekevelt 1986). This does not

seem to be occurring in the vessels found in the manatee cornea. Instead, the vessels are

found in any orientation and direction.

It is important to note that even though the vascular patterns of the manatee cornea

and choriocapillaris of some areas of various species are similar in that they both have

irregular patterns; they most likely share more differences than similarities in vascular

structure. The choriocapillaris is only a single layer of vessels, whereas the corneal









vascular extends through several layers and even penetrates into adjacent cells. Still, it is

important to note that the choriocapillaris does have a relatively high surface area of

vessel coverage, which does not interfere with light transmission, thus affecting vision

(Buggage et al. 1996). If the vascular patterns in the choriocapillaris are disrupted for

any reason, it can result in insufficient removal of waste that is generated by the RPE

cells causing an accumulation of additional wastes at Bruch's membrane (Cao et al.

1998). This information regarding the choriocapillaris suggests that not only are the

vascular patterns in and around the choriocapillaris important for not interfering with

vision in animals with tapeta, but also needed for normal retinal functions to occur.

Perhaps the vasculature in the corneas of the manatees observed is more

comparable to the vasculature of the retinal vessels. Vascular patterns in the retina can be

found in the center and peripheral areas. In the outer retina, vascular demands are met by

choroidal circulation. Species of animals that have vascularized inner retinas have a wide

range of blood vessel patterns, and these patterns are most often determined by the

density distributions of neural cells in the ganglion cell layer. In areas of high cell

density, large capillary beds are found. For example, in the domestic cat (Felis

domesticus), the squirrel monkey (Saimiri sciureus), and human a high concentration of

capillaries are found in the area centralis and fovea where high cell densities are found.

Large blood vessels that are found in the retina differ from retinal capillaries in that they

will not be found in areas of high cell density. This feature allows visual acuity to be

maintained without compromise by allowing light to be transmitted without interference

(Dunlop et al. 1997).









Studies using scanning electron microscopy (SEM) investigating vascular patterns

in the retina of rats have shown small stem arterioles that are thin in diameter and are

nearly straight. These arterioles branch and meet with capillaries that go on to form

massive plexuses and sparse vascular networks in the center and periphery of the retina.

The capillaries found in the retina are thin in diameter, measuring 3 to 4 am (Ikebe et al.

2001). Additionally, it has been found in rats using vascular corrison casts that side

branching of arterioles is the main vascular pattern observed in the arterial tree of the

retina. This vascular pattern most likely allows the best flow of materials to the most

superficial layers of the retina (Pannarale et al. 1996).

The corneal vessels appear to be most similar to retinal vessels in that they both

form multiple layers. The vasculature of the retina allows light to pass through with

minimal interference, especially towards the center of the retina where large vessels are

rarely found under normal conditions. Larger vessels are more commonly found in the

peripheral areas of both the retina and cornea where there is minimal impedance of light

transmission. The importance of retinal blood vessel structure is especially noticeable

when there is some type of abnormality. Normal choroidal vasculature is necessary for

normal retinal function and if the vasculature is compromised for any reason, the result

can be dysfunction or death of photoreceptors. This can be a sign of a larger problem

such as diabetic retinopathy (Cao et al. 1998). Other physiologic changes that may occur

in retinal vasculature patterns under abnormal conditions can result in diseases such as

glaucoma and age-related macular degeneration. Disturbances most commonly noted

with these disorders include a change in ocular pressure and metabolic changes in

autoregulation of oxygenated blood delivery to the retina (Yazdanfar et al. 2003).









An additional vascular pattern occurring under normal conditions in reptilian eyes

is in the spectacle. The spectacle is a fused eyelid that forms a clear covering over the

cornea and is the same size or larger than the cornea. Microsilicone injected into the

spectacle found that the structure was vascularized and that the vessels were in a vertical

arrangement in some snake species (Mead 1976). The most interesting finding of this

study was that the spectacle maintained a vascular pattern and still retained a high level

of transparency. These vessels seem to be similar to those found within the manatee

cornea in that they are both small and similar in size throughout the area. However, not

much work has been conducted on the spectacle of reptilian species, so it is hard to make

any definite comparisons between the manatee cornea and reptilian spectacle until more

research has been carried out.

In the normal, healthy cornea, light passes through unimpeded. The stroma, which

makes up the majority of the cornea, has an arrangement of fibrils in such a fashion that

does not interfere with light transmission (Freund et al. 1995). However, conditions such

as disease, injury, trauma, or clinically induced malformations can change the balance

within the cornea leading to inflammation or eventually, more serious problems. Such

side effects occur in many species of animals under pathological conditions (Klintworth

1991).

In human ophthalmology there is a need for immediate and quick diagnosis and

medical aid in response to injury or disease causing corneal vascularization. At the

General Eye Service of the Massachusetts Eye and Ear Infirmary, 35 of 845 patients

(4.14%) who came in for an eye examination had some form of corneal vascularization.

This 4.14% of patients represents 1.4 million Americans with overwhelming infections









that can cause blindness. The primary causes of corneal vascularization in humans result

from chlamydial infections (affecting 400 million people worldwide and blinding 6

million), onchocerciasis infections (affecting 50 million people and blinding 1 million),

herpes simplex eye infections (affecting 500,000 in the United States), and the wearing of

contact lenses (affecting 125,000 to 470,000 people in the United States). This

information suggests that corneal vascularization is a major contributor to eye disease in

humans (Lee et al. 1998).

An additional condition that can occur in humans and cause vascularization in the

cornea is homocystinuria. Homocystinuria is an autosomal-recessive disorder of amino

acid metabolism that affects several organ systems including the skeletal system, central

nervous system, vascular systems, and the eyes. Even though corneal opacities with

homocystinuria are not common, 8 out of 90 eyes examined with homocystinuria

contained corneal opacities. Eyes that were found to have corneal opacities caused by

homocystinuria were found with deep, central scarring and opacification associated with

vascularization caused by long-term corneal edema and dislocation of the lens (Rao et al.

2002). Such disorders can most likely be ruled out in the manatee because pathology

reports of the necropsied animals examined did not indicate major problems occurring in

organ systems caused by genetic problems nor were dislocated lens and edema observed

in the manatee cornea during gross and histological examination. Therefore, it is most

likely safe to postulate that this genetic disorder or others like it are not responsible for

the vascularized corneas found in the Florida manatees examined, especially because

research investigating genetic disorders has not found much if any evidence to indicate

such problems exist. Currently the only known genetic disorder to occur in manatees is a









defect called congenital ectrodactyly, which is a malformation that is characterized by

one or more phalanges missing from one or more digits (Watson and Bonde 1986).

However, more research is needed in this area of study to draw any conclusive results

suggesting that corneal vascularization in the Florida manatee may be a possible side

effect from some other physiological factor.

More conclusive results can be drawn from other research carried out on various

animal species investigating comeal vascularization and its impact on the surrounding

tissue, which indicates that vascularization observed in the manatee cornea is not caused

by a pathological condition such as injury, disease or trauma. When injury occurs to the

cornea, often the epithelium will contain goblet cells and the stroma becomes

vascularized leading to a deficiency in vision. Corneal function will then be interfered

with resulting in recurring corneal erosions, decrease in vision caused by uneven corneal

surfaces, weakened tensile strength, and ineffectual barrier function (Lee et al. 1998).

While not all of these factors could be examined at the histological level with the

manatee cornea, no comeal erosions were found.

Mice with induced corneal vascularization caused by cauterization were found to

have several symptoms not found in the manatee such as edema and opacity occurring in

all specimens examined, along with a loss of epithelial and endothelial cells and a large

amount of infiltrating neutrophils and macrophages into the stroma (Sonoda et al. 1998).

Rats that were used to study inflammation associated with induced corneal

vascularization through krypton laser bums demonstrated different histological

observations than those found in the manatee. After the laser burns were performed on

the rats, epithelial wounding of the limbus was found along with an increase in the









number of cells around the limbal vessels. Additionally, infrequent inflammatory cells

were observed and vascularization eventually advanced towards the center of the cornea

leading to the vascularization of the entire cornea including the deep stroma (Kvanta et

al. 2000).

Induced corneal vascularization in rabbits through alkali injury also resulted in

trauma not observed in manatee corneas. Damage to all cellular components including

the epithelium, keratocytes, and endothelium was observed light microscopically. Edema

in the stroma caused it to increase by 35% and on the edge of the stroma, damaged cells

and nuclear debris were seen. Towards the peripheral portion of the stroma, fibroblast

and inflammatory cells were found. Additionally, the structure of the stroma was

disrupted so that lamellae separated and collagen fibrils lost normal parallel arrangement

leading to large spaces between the fibrils or fibrils that were almost touching (Huang et

al. 2001).

Dogs can be afflicted by corneal vascularization through such disorders as

spontaneous chronic corneal epithelial defects (SCCED). This defect causes erosions in

the cornea that are characterized by sheets of loosely arranged epithelium that can take up

to six months to heal. Similar erosions can also be found in humans that take long

periods of time to repair and are often associated with afflictions such as traumatic

abrasions, epithelial membrane dystrophy, anterior stromal dystrophies, and neurotrophic

keratitis. In dogs with SCCED, the basement membrane of the anterior epithelium will

detach from the stroma indicating that the extracellular matrix components in these

animals may have been previously compromised (Bentley et al. 2001). In the manatee

corneas studied, no erosions or detached basement membranes were observed, once again









suggesting that the vascularization found is not an abnormal condition for the Florida

manatee.

Collectively, signs of inflammation and concurrent side effects found in humans

and other mammals have not been observed in the vascularized corneas of the Florida

manatee, indicating that the vessels present are not occurring under abnormal conditions.

In summary, every Florida manatee cornea examined (n=26) contained vascularization.

The vessels observed varied in location, amount, and even between corneas of the same

individual. However, no side effects such as edema or erosions were observed in the

surrounding tissue indicating that this may not be an abnormal condition in the corneas of

Florida manatees. Because vascularization was observed in the fetus, corneal

vascularization in the manatee may be a developmental or evolutionary process possibly

occurring as a side effect from some other unknown physiological factor. The

significance of these findings represent the first study known to us in which corneal

vascularization has been found that did not occur because of some type of pathological

condition. Additionally, the size of the vessels found in the corneas indicates that they

are too small to interfere with light transmission and are most likely not affecting

manatees' ability to survive in their different aquatic environments.









































Figure 2-1: Florida manatee eye with ocular tissue and fat removed revealing the eye and
optic nerve. The eye is being compared to a dime for size comparison.




















































Figure 2-2: These histological pictures ofMNW-9111 show (A) a long blood vessel in
the stroma (blue) that approaches the anterior epithelium (purple) at 100X, (B)
blood vessel penetration between the stroma and anterior epithelium at 200X,
and (C) close detail of the vessels present in the stroma at 400X.










































I


Figure 2-3: These pictures were taken ofMSW-9114 of the same blood vessel at varying
magnifications to show the penetration of this vessel into the anterior
epithelium (purple): (A) 200X, (B) 400X, and (C) 1000X.















/B
r 1)




-a


Figure 2-4: (A) The arrangement of smaller vessels near the anterior epithelium (purple)
from the larger trunk vessel in the stroma (blue) at 200X, specimen MSW-
9114. (B) This picture of UCF-9114 shows the smaller vessels (see arrow)
that were outlined within the stroma for three-dimensional reconstruction
(200X).


_ I


OOFI

























\4


Figure 2-5: Pictures from MNW-9016 showing (A) a blood vessel passing through the
stroma (blue) at 200X, and (B) red blood cells present in the vessel as it
penetrates the anterior epithelium (purple) at 1000X.






60















'rA
a






















,ri





Figure 2-6: These pictures were taken from the unknown specimen. (A) Shows the edge
of the limbal vascular arrangement at the peripheral cornea (200X), and (B)
larger vessels that go right up to the anterior epithelium at 200X.




















































Figure 2-7: Anatomical orientation of manatee eyes as they are described in three-
dimensional reconstruction: (A) left eye orientation (face-on view), (B) right
eye orientation (face-on view), (C) dorsal eye orientation, (D) ventral eye
orientation, (E) lateral eye orientation (or horizontal view if left or right is
unknown), and (F) medial eye orientation (or horizontal view if left or right is
unknown).















































Figure 2-8: MNW-9016, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


1 ^D
T


_- -

Aim


F















































Figure 2-9: MNW-9022(R), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).


t


--





































Figure 2-10: MNW-9111, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) medial view of vessels and corneal tissue,
and (F) medial view of only the vessels at 25X (arrow indicates anterior
position).


LI'


Y r

ID;


F






















































Figure 2-11: MNW-9116, (A) face-on on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) horizontal view of vessels and corneal tissue, and
(D) horizontal view of only the vessels at 25X.


at
rB







B


ax


V
S -
lb
r






D



























/rJYI I(II IA



t D


cm





F


Figure 2-12: MSE-9101, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).





















































Figure 2-13: MSE-9105, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) medial view of vessels and corneal tissue,
and (F) medial view of only the vessels at 25X (arrow indicates anterior
position).






68














A B














C D














E F

Figure 2-14: MSE-9107, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


















































-




E F

Figure 2-15: MSE-9115, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).





Figure 2-16: MSE-9131, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


rJ





F


%I






I
























B


Figure 2-17: MSW-9114, (A) face-on on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


-r

rf





F
















































Figure 2-18: MSW-9131(R), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).


/




B


H



Ii


F














































Figure 2-19: MSW-9137(L), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).


I ~ r


I~c


J a-


F






74









1 r



A B










I '


IC













E F

Figure 2-20: MSW-9137(R), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).















































Figure 2-21: MSW-9141, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


SD


s a


't-~





B








































Figure 2-22: MSW-9143, (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) horizontal view of vessels and corneal
tissue, and (F) horizontal view of only the vessels at 25X (arrow indicates
anterior position).


/Y~4


Sii

I


r-.

r



























































Figure 2-23: TM-8619(L), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).


B


C 11~1 D


F






78






























C I
A B













E F


Figure 2-24: TM-8619(R), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) dorsal view of vessels and corneal tissue, (D)
dorsal view of only the vessels, (E) lateral view of vessels and corneal tissue,
and (F) lateral view of only the vessels at 25X (arrow indicates anterior
position).

























































E F

Figure 2-25: TM-8630(L), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) medial view of vessels and corneal tissue,
and (F) medial view of only the vessels at 25X (arrow indicates anterior
position).


B


1 D
















































Figure 2-26: TM-8630(R), (A) face-on view of vessels and corneal tissue, (B) face-on
view of only the vessels, (C) ventral view of vessels and corneal tissue, (D)
ventral view of only the vessels, (E) medial view of vessels and corneal tissue,
and (F) medial view of only the vessels at 25X (arrow indicates anterior
position).


.r


t $


7-




F



















-K >


Figure 2-27: UCF-9114, (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal
view of only the vessels, (E) horizontal view of vessels and corneal tissue, and
(F) horizontal view of only the vessels at 25X (arrow indicates anterior
position).


t1I


EI






F














































Figure 2-28: UCF-9137, (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral
view of only the vessels, (E) horizontal view of vessels and corneal tissue, and
(F) horizontal view of only the vessels at 25X (arrow indicates anterior
position).


4'-

)IA

B


I '


I
a
n a
r



F






83









*1A





A B














C














E F

Figure 2-29: UCF-9138, (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral
view of only the vessels, (E) horizontal view of vessels and corneal tissue, and
(F) horizontal view of only the vessels at 25X (arrow indicates anterior
position).













































Figure 2-30: UCF-9141, (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal
view of only the vessels, (E) horizontal view of vessels and corneal tissue, and
(F) horizontal view of only the vessels at 25X (arrow indicates anterior
position).


1'


D


F




















A


I


Figure 2-31: Unknown, (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal
view of only the vessels, (E) horizontal view of vessels and corneal tissue, and
(F) horizontal view of only the vessels at 25X (arrow indicates anterior
position).


m F


F




















































Figure 2-32: Fetus(L), (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) dorsal view of vessels and corneal tissue, (D) dorsal
view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F)
lateral view of only the vessels at 25X (arrow indicates anterior position).


It


F


B





Figure 2-33: Fetus(R), (A) face-on view of vessels and corneal tissue, (B) face-on view
of only the vessels, (C) ventral view of vessels and corneal tissue, (D) ventral
view of only the vessels, (E) lateral view of vessels and corneal tissue, and (F)
lateral view of only the vessels at 25X (arrow indicates anterior position).


I


4
Xc:'I


B


-~






88

















A B










tzr&-









Figure 2-34: Dorsal quadrant of MNW-9016 at 200X, (A) vasculature and corneal
outline from face-on view, (B) vasculature only from face-on view, (C)
vasculature only from dorsal view, and (D) vasculature only from horizontal
view (arrow indicates anterior position).





















A


Figure 2-35: Medial quadrant ofMNW-9016 at 200X, (A) vasculature and corneal tissue
outline from face-on view, (B) vasculature only from face-on view, (C)
vasculature only from dorsal view, and (D) vasculature only from horizontal
view (arrow indicates anterior position).


_1w