Description of the Chemical Senses of the Florida Manatee, Trichechus Manatus Latirostris, in Relation to Reproduction

Home , Lingual papillae, Vandenbergh effect, Whitten effect

DESCRIPTION OF THE CHEMICAL SENSES OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS, IN RELATION TO REPRODUCTION

MEGHAN LEE BILLS

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

2011

To my best friend and future husband, Diego Barboza: your support, patience and humor throughout this process have meant the world to me

ACKNOWLEDGMENTS

First I would like to thank my advisors; Dr. Iskande Larkin and Dr. Don Samuelson.

You showed great confidence in me with this project and allowed me to explore an area outside of your expertise and for that I thank you. I also owe thanks to my committee members all of whom have provided valuable feedback and advice; Dr. Roger Reep, Dr.

David Powell and Dr. Bruce Schulte.

Thank you to Patricia Lewis for her histological expertise. The Marine Mammal

Pathobiology Laboratory staff especially Drs. Martine deWit and Chris Torno for sample collection. Thank you to Dr. Lisa Farina who observed the anal glands for the first time during a manatee necropsy. Thank you to Astrid Grosch for translating Dr. Vosseler‟s article from German to English. Also, thanks go to Mike Sapper, Julie Sheldon, Kelly

Evans, Kelly Cuthbert, Allison Gopaul, and Delphine Merle for help with various parts of the research. I also wish to thank Noelle Elliot for the chemical analysis.

Thank you to the Aquatic Animal Health Program and specifically: Patrick

Thompson and Drs. Ruth Francis-Floyd, Nicole Stacy, Mike Walsh, Brian Stacy, and

Jim Wellehan for their advice throughout this process. To Dr. Bob Bonde, Cathy Beck and everyone at USGS, thank you not only for your help with my project but those of sirenian researchers everywhere. I also owe much gratitude to Sally O‟Connell who kept me on track with the paperwork and requirements of a PhD student and Dr. Charles

Courtney who has provided support and advice for the achievement of my career goals.

Thank you very much to the facilities that allowed me to work with their captive manatees. Specifically: Marilyn Margold at the South Florida Museum‟s Parker

Aquarium, Joe Gaspard of Mote Marine Laboratory, Dr. Andy Stamper and Kim O‟Dell of Disney‟s Epcot, The Living Seas.

Finally, I would like to thank my Mom, Dad and brother for their encouragement and support not only during my PhD but for my entire college career. I would also like to thank Heather Maness, Maggie Hunter, Gretchen Henry, Shannon Skevakis, Noel

Takeuchi and Jen McGee for their feedback, critiques and friendship. I also could not have been successful without the love and support of my fiancé, Diego Barboza.

Funding was provided through the Florida Fish and Wildlife Conservation

Commission, Whitney Marine Laboratory for Marine Bioscience and the University of

Florida Aquatic Animal Health Program, the University of Florida College of Veterinary

Medicine Consolidated Faculty Research Development Grant and Sigma Xi-Grant in Aid of Research. Research completed under US Fish and Wildlife permit #: MA067116-1 and UF IACUC Study #: 200902684.

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS ...... 4

LIST OF TABLES ...... 8

LIST OF FIGURES ...... 9

LIST OF ABBREVIATIONS ...... 12

ABSTRACT ...... 15

CHAPTER

1 INTRODUCTION ...... 17

Chemoreception ...... 17 Olfaction ...... 17 Vomeronasal Organ ...... 20 Taste ...... 21 Chemical Communication ...... 25 Pheromones ...... 26 Mammalian Reproductive Chemical Communication ...... 31 Paenungulata ...... 33 Aquatic Mammals ...... 37 Olfaction ...... 38 Vomeronasal Organ ...... 42 Taste ...... 43 The Florida Manatee ...... 47

2 ANAL GLANDS OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS: A POTENTIAL SOURCE OF CHEMOSENSORY SIGNALS ...... 52

Background ...... 52 Materials and Methods...... 54 Light Microscopy ...... 55 Transmission Electron Microscopy ...... 55 Results ...... 56 Light Microscopy ...... 58 Electron Microscopy ...... 61 Discussion ...... 61

3 TASTE BUDS IN THE ORAL CAVITY OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS, WITH ESTROGEN RECEPTOR Β WITHIN THE LINGUAL ROOT TASTE BUDS ...... 66

Background ...... 66 Materials and Methods...... 70 Gross and Microanatomy ...... 71 Taste Bud Size and Quantification ...... 72 Immunohistochemistry...... 74 Transmission Electron Microscopy ...... 76 Results ...... 76 Estrogen Receptor β ...... 80 Ultrastructure ...... 83 Discussion ...... 84

4 THE NASAL CAVITY AND OLFACTORY EPITHELIUM OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS ...... 91

Background ...... 91 Materials and Methods...... 94 Light Microscopy ...... 96 Immunohistochemistry...... 97 Results ...... 97 Light Microscopy ...... 100 Immunohistochemistry...... 103 Discussion ...... 104

5 MALE FLORIDA MANATEE, TRICHECHUS MANATUS LATRIROSTIS, BEHAVIORAL RESPONSE TO FEMALE MANATEE URINE FROM DIFFERENT REPRODUCTIVE TIMEPOINTS ...... 109

Background ...... 109 Methods ...... 112 Results ...... 117 Discussion ...... 124

6 CONCLUSIONS ...... 127

APPENDIX:CHEMICAL ANALYSIS OF FEMALE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS URINE ...... 134

LIST OF REFERENCES ...... 140

BIOGRAPHICAL SKETCH ...... 156

LIST OF TABLES

Table page

2-1 Samples used for anal gland analysis...... 54

3-1 Taste bud samples collected for gross and histological analysis and fixed in 10% NBF ...... 71

3-2 Samples used for estrogen receptor β analysis ...... 75

3-3 Taste bud and papillae number within manatee tongue roots...... 78

3-4 Taste bud size within manatee tongue roots ...... 79

3-5 Measured length of foliate papillae from the tongue roots of two manatees...... 80

3-6 Amount of estrogen receptor β reaction, average of two observers‟ ratings, within the epidermis of the Florida manatee tongue root...... 82

3-7 Amount of estrogen receptor beta reaction, average of two observers‟ ratings, within the taste buds of the Florida manatee...... 83

4-1 Samples collected for nasal cavity and olfactory epithelium evaluation ...... 95

4-2 The olfactory thickness of various manatees from different areas of the olfactory system as represented by different block numbers...... 100

5-1 Male manatees assessed with female manatee urine...... 115

5-2 Ethogram of behavior categories performed by captive male West Indian manatees exposed to urine of female manatees...... 116

A-1 Manatee urine samples used for chemical analysis...... 135

LIST OF FIGURES

Figure page

1-1 Transduction of olfactory information during inhalation ...... 18

1-2 The transduction pathways of the VNO and MOE within a hamster...... 21

1-3 Transduction of signals to and within the brain during taste with food in the oral cavity...... 24

2-1 Adult female (MEC0983), collecting duct (CD) approaching the anorectal junction...... 56

2-2 Ventral view of a male manatee calf with an inset of the current working diagram of anal gland anatomy...... 57

2-3 Anal glands (AG) of male fetus F-MSW03160 completely ventral to the canal. . 58

2-4 Collecting duct (CD) and its contents ...... 59

2-5 Collecting duct (CD) joining anal integument as an excretory duct...... 59

2-6 Various histologic stains of adult manatee anal glands...... 60

2-7 Smooth muscle actin stain of adult female LPZ102765 demonstrating the location of myoepithelial cells...... 60

2-8 TEM of manatee anal glands ...... 61

3-1 Midsagittal view of manatee head ...... 72

3-2 Tongue root location and foliate papillae on side of root. Scale bars represent approximately 1 cm...... 73

3-3 Taste bud measurements...... 73

3-4 Typical foliate papillae with taste buds 20X and magnification of taste bud at 250X...... 74

3-5 Taste buds of the soft palate of LPZ102639...... 77

3-6 Apparent taste buds, indicated by arrows, of the nasal valve of MNW0905 ...... 77

3-7 Estrogen receptor β staining within the lingual root of an adult manatees ...... 81

3-8 Electron micrographs of manatee taste buds...... 84

3-9 Mid-sagittal section of a manatee head with a transverse section demonstrating the groove between tongue and teeth (arrow)...... 86

4-1 Head of a male manatee, MNW0905, used for nasal examination ...... 95

4-2 Section of manatee head, LPZ102765, from which olfactory and nasal samples were collected...... 96

4-3 Measurement of olfactory thickness...... 97

4-4 Manatee head with sections dividing the general area of bandsaw cuts made on a perinatal male, MEC0853...... 98

4-5 Cross sections of the nasal cavity of a perinatal male Florida manatee, MEC0853 ...... 99

4-6 Cross section of mucous producing cells from adult female, MNW0901, H&E 20X...... 100

4-7 Mucous glands of adult female, MNW0901, at approximately section 2 of Figure 4-4...... 101

4-8 Respiratory epithelium within the nasal cavity of the Florida manatee...... 101

4-9 Cross section from adult female, MNW0901, demonstrating olfactory epithelium ...... 101

4-10 Section of manatee head along sagittal plane and corresponding epithelial lining within Ethmoturbinates II and III, and dorsal nasal turbinate (D) ...... 102

4-11 Diagram representing the turbinate system and location of epithelial types within the manatee head ...... 103

4-12 OMP staining of olfactory epithelium from an adult female, LPZ102765 ...... 104

5-1 Urinary hormone levels of Lorelei, a female Florida manatee...... 113

5-2 Urinary hormone levels of Charlotte, a female Florida manatee ...... 113

5-3 Urinary hormone levels of Sara, a female Florida manatee...... 114

5-4 Behavior of Snooty in 2009 in response to female urine from different reproductive time points...... 119

5-5 Behavior of Lou in 2011 in response to female urine from different reproductive time points...... 120

5-6 Behavior of Snooty in 2011 in response to female urine from different reproductive time points ...... 121

5-7 Behavior of Vail in 2011 in response to female urine from different reproductive time points...... 122

5-8 Behavior of male manatees to female urine from different reproductive time points in 2010...... 123

6-1 Overall chemosensory receptor organs within the head of the manatee ...... 129

A-1 A PCA plot demonstrating separation between group A-anestrous urine on the left and group B-estrous urine on the right...... 137

A-2 A PLS-DA analysis demonstrating separation in group A-anestrous urine on top and group B-estrous urine on bottom...... 137

LIST OF ABBREVIATIONS

AG Anal Glands

AEC Aminoethylycarbazole

BG Bowman‟s Glands

BM Basement Membrane

CD Collecting Duct

CE Columnar Epithelium

CM Centimeters

CN Cranial Nerve

COD Cause of Death

CRL Crown Rump Length

D Dorsal nasal turbinate

E1S Estrogen

FSH Follicle Stimulating Hormone

G Goblet cells

H&E Hematoxylin and Eosin

HLA Human Leucocyte Antigen

HMDB Human Metabolome Database

HPLC High Performance Liquid Chromatography

KEGG Kyoto Encyclopedia of Genes and Genomes

LH Luteinizing Hormone

M Middle nasal turbinate

Mi Microvilli

MALDI-TOF Matrix Assisted Laser Desorption/Ionization-Time of Flight

MHC Major Histocompatibility Complex

MMCD Madison-Qingdao Metabolomics Consortium Database

MOE Main Olfactory Epithelium

MYO Myoepithelial cell

NB Nerve Bundle

NBF Neutral Buffered Formalin

NG Nasal Glands

NM Not Measured

NST Nucleus of the Solitary Tract

OE Olfactory Epithelium

OMP Olfactory Marker Protein

PAS Periodic Acid Schiff

PCA Principle Components Analysis

PCB Polychlorinate Biphenyls

PG Progesterone

PLS-DA Partial Least Squares-Discriminant Analysis

QMP Queen Mandibular Protein

SCC Solitary Chemosensory Cell

SE Squamous Epithelium

T1R Taste Receptor 1

T2R Taste Receptor 2

TEM Transmission Electron Microscopy

TL Total Length

TOF Time-of-Flight

TP Taste Pore

V Vein

V1R Vomeronasal Receptor 1

V2R Vomeronasal Receptor 2

VNO Vomeronasal Organ

µm micrometer

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

DESCRIPTION OF THE CHEMICAL SENSES OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS, IN RELATION TO REPRODUCTION

Meghan Lee Bills

December 2011

Chair: Iskande V. Larkin Cochair: Don A. Samuelson Major: Veterinary Medical Sciences

The use of taste and smell is well documented in terrestrial mammals and aquatic non-mammalian species but has never been examined in fully aquatic mammals. It is thought that in mysticetes, pinnipeds, and sirenians there is olfactory detection at the water surface whereas detection within the water is mediated by taste buds. Anecdotal evidence indicates that the fully aquatic, endangered Florida manatee, Trichechus manatus latirostris, is capable of chemically sensing a female manatee in estrus. This project sought to define the chemical sensing abilities of the manatee and included a characterization of chemosensory receptive and transmission anatomy, behavioral assessment of male reaction to female manatee urine, and chemical analysis of that urine.

A series of anatomic descriptions were completed including gross and microscopic examinations of the anal glands (n=11), taste buds (n=9) and olfactory epithelium (n=9) of the manatee. Each analysis included a representative sample of varying ages, sexes and causes of death. The anal glands were comparable to the apocrine intramuscular scent glands of dogs and emptied at the anorectal junction. Olfactory epithelium covers

a majority of the three ethmoturbinates and a small portion of the septum. The examination of the mouth corroborated the presence of taste buds of the tongue root.

The location of taste buds may allow for transport of chemicals without continuous exposure to the external environment. Immunohistochemical staining for estrogen receptor β within the lingual root was completed on 23 animals of varying ages (10 males and 13 females). There was consistent staining within the epidermis and apparent staining within the taste buds of 18 animals.

Behavioral analysis assessed a total of five adult males using urine from three different adult females. The study used a double blind, continuous sampling method in which an observer recorded the start time of individual behaviors before and after sample addition. The samples included salt or freshwater, estrous or anestrous urine.

Two males exhibited a chemosensory response to at least one estrous sample from each of the three females. The urine that evoked a behavioral response was analyzed using reverse phase high performance liquid chromatography (HPLC) coupled to an

Agilent time-of-flight (TOF) mass spectrometer with a positive electrospray ionization. A principle components and partial least squares-discriminant analysis of the ionic peaks showed a significant separation between 64 ions within the estrous and anestrous urine.

The Florida manatee was the first fully aquatic marine mammal to provide evidence of chemical signaling both behavioral, male reaction to female urine, and anatomic, anal glands. Manatees possess potential receptors for these signals however the primary mechanism for perception of them has not yet been determined.

Understanding chemical sensing in manatees provides insight into the yet undefined chemosensory abilities of other fully aquatic mammal species.

CHAPTER 1 INTRODUCTION

Chemoreception

Within the field of sensory systems, the area of chemoreception has had limited study and a majority of current research has focused on laboratory species. All animals studied have developed the ability to discriminate between small variations in chemical structure and the signal blends used for communication can be extremely complex.

Chemoreception varies from other senses because unlike vision or hearing the subject cannot scan the environment but must come into direct contact with the chemicals. For these reasons, the senses of taste and smell are more difficult to investigate compared to the other senses and there are many aspects of the interactions between chemicals and their receptors that are not well understood. However, the use of chemical signals is ubiquitous among aquatic and terrestrial animals and vital to locating food, evading predators and reproducing successfully.

Olfaction

Of the sensory systems, odors are one of the most complex forms of stimuli to delineate and there are still many facets of the ligand-receptor bond that are not yet understood. Not only is the chemical signal itself important in the processing of olfactory information but also the simultaneous integration of the chemical‟s position in space, whether in air or water, is equally important (Finger et al. 2000).

The relaying of olfactory information to the brain begins with heptahelical G- protein coupled receptors in the peripheral processes of bipolar neurons in the nasal cavity. The centrally directed processes of these neurons collect as the olfactory nerve filaments of cranial nerve I and terminate in the olfactory bulb (Ache and Zhainazarov

1995) (Figure 1-1). Genetic data indicate that there are 400-1000 different olfactory receptors in mammals (Parmentier et al. 1992, Ressler et al. 1994) whereas only 50-

100 appear to exist in fish (Ngai et al. 1993, Barth et al. 1996).

Figure 1-1. Transduction of olfactory information during inhalation. Air flows indicated by dashed and dotted lines AM, amygdala; LOFC, lateral orbitofrontal cortex; MOFC, medial orbitofrontal cortex; OB, olfactory bulb; OC, olfactory cortex; OE, olfactory epithelium. Adapted from Shepherd 2006.

Each olfactory receptor is encoded for by single-exon genes, which are distributed in clusters across multiple chromosomes (Finger et al. 2000). There are currently two known receptor classes termed type I and type II, which evolved from common ancestral genes. The absence of class I receptors in terrestrial mammals and birds and their existence in fish indicates that these receptors are responsible for

detection of water-soluble odorants. Type II receptors are thought to be specialized for volatile odorant detection. This is supported by the presence of nonfunctional pseudogenes of type II receptors, which are found in the genome of the striped dolphin,

Stenella coeruleoalba. This correlates with the loss of olfaction as ancestral dolphins entered an aquatic environment. The absence of type I receptors indicates that odontocetes do not have the ability to perceive aquatic odorants (Freitag et al. 1998).

These conclusions have been supported by numerous publications (Marchand et al. 2004, Levasseur et al. 2007) and although more research is needed on this topic, the categorization of type I and II receptor classes is the current working model. The differences in the amount and type of receptors between aquatic and terrestrial species indicates an adaptation to environment because aquatic animals encounter a smaller number of water-soluble molecules compared to the variety of hydrophobic, volatile compounds available to terrestrial animals.

For both type I and II receptors, many non-neuronal factors can affect how a signal is coded and subsequent processing of information in the brain. For example, once an odor has reached the mucus or sensilar fluid of the nasal cavity it can be

“presented” to a receptor by odorant-binding proteins, or it can be cleared by odorant- degrading proteins. A receptor can only detect one type of signal but any given signal may bind to many receptors. In addition, once a receptor has been bound by an odorant it cannot send another action potential until that odorant has been removed, and this process may require an odorant degrading molecule (Finger et al. 2000).

After the signal from the main olfactory epithelium (MOE) has synapsed on the olfactory bulb, the second synapse occurs on one of many higher olfactory centers

located in the cortex. Olfactory signals most commonly project to the piriform cortex.

The limbic system has an olfactory component that includes portions of the piriform cortex, amygdala, and hippocampus (Finger et al. 2000). Comparative studies have correlated decline in olfactory use and olfactory bulb size within a species with a decrease in the size of the limbic system (Reep et al. 2007).

Higher olfactory centers then project to subcortical and neocortical areas or can cycle back to the olfactory epithelium. The direct link from olfactory receptors to the olfactory bulb allows for almost immediate response to stimuli (Finger et al. 2000), which is useful in life threatening situations as well as in a competitive reproductive setting. In addition to the MOE, there is a specialized olfactory organ, the vomeronasal organ (VNO) that is used for pheromone detection (Wyatt 2006).

Vomeronasal Organ

The predominant proposed use of the VNO is for reproductive signal reception.

The VNO projects to the accessory olfactory bulb, which then projects to regions of the amygdale distinct from the targets of the MOE (Finger et al. 2000). There are two known receptor types of the VNO; type 1 (Dulac and Axel 1995) and type 2 (Herrada and Dulac

1997, Matsunami and Buck 1997) (V1R and V2R respectively). Both are seven transmembrane g-protein coupled receptors (Halpern et al. 1995). The direct link from the accessory olfactory bulb to the amygdala and subsequently to subcortical pathways

(Figure 1-2) allows for a faster and less consciously controlled reaction than traditional olfaction. In the past, this complete separation of the projections of the main and olfactory bulbs indicated that the VNO transmits different information than the MOE, in most cases, pheromonal social cues (Finger et al. 2000). However, it has become

evident that there is much overlap between the signals and resultant effects mediated by the VNO and MOE (Brennan and Zufall 2006, Wang et al. 2007).

Figure 1-2. The transduction pathways of the VNO and MOE within a hamster. Text and Image © Dr. Michael Meredith and Neuroscience Program FSU.

The VNO is thought to have evolved from an extra fold that appears in the olfactory bulb of archaic fish species. Over evolutionary time, this fold evolved into a separate structure with divergent pathways from that of traditional olfaction and is most developed in terrestrial vertebrates. However, in many reptiles that evolved from a terrestrial to an aquatic existence, it is apparent that the VNO is not present or is vestigial. This indicates that species which transitioned from a terrestrial to an aquatic environment have not developed a VNO for use in aquatic systems (Bertmar 1981,

Eisthen 1992). In addition, birds and primates, including humans, do not have a functional VNO (Meisami and Bhatnagar 1998).

Taste

Taste reception is predominantly associated with food preference and is a combination of several senses including texture, olfaction and the detection of spice or

chemesthesis. This combination of senses and the ability to respond to a wide variety of signals ranging from large lipid molecules to small simple chemical compounds such as salt makes taste unique compared to other senses. Taste buds are the primary end organ for taste in vertebrates and are found predominantly on the tongue in various papillae as well as in other areas of the mouth including the soft palate, upper esophagus and epiglottis. There are five known types of taste. Salty and sour are detected through ion channels. Sweet, bitter and umami are detected through G protein coupled receptors.

There are two genetically distinct taste receptors: type 1 (Li X et al. 2002) and type 2 (Adler et al. 2000, Matsunami et al. 2000) (T1R and T2R respectively). In general, there are more T2R genes, also known as bitter taste receptors than T1R, sweet and umami or protein receptors. There is a correlation between what an animal eats and the number of T1R and T2R genes. For example, herbivores such as cows have more bitter receptors than carnivores such as dogs, most likely to protect them from toxins in vegetation (Nei et al. 2008).

In mammals, taste buds are composed of at least four types of taste cells that each appears to be responsible for various forms of signal reception. This varies according to the organism. These cells can be distinguished by their morphological characteristics including shape and cytoplasm color. Type I cells appear to be glial-like: they support other cells and provide ion gradient control for the sour and salty senses.

Type II cells contain receptors for sweet, bitter and umami transductions. Type III cells are an intermediary between receptors and nerve fibers and respond broadly to sweet, salty, sour, bitter and umami signals that are picked up by type I and II cells. The

function of type IV cells is not known but they are located a distance from the pore of the taste bud (Suzuki 2007, Chaudhari and Roper 2010).

Taste buds synapse on axons of the facial (CN VII), glossopharyngeal (CN IX) or vagus (CN X) nerve for transmission of signals to the central nervous system. These nerves transmit information from rostral (CN VII), intermediate (CN IX), and caudal (CN

X) portions of the oral cavity, respectively. Cranial nerves VII, IX and X synapse on the nucleus of the solitary tract (NST). Axons from the rostral, gustatory division of the NST terminate in the parabrachial nuclei, which then project to limbic structures including the lateral hypothalamus, central nucleus of the amygdale, and the bed nucleus of the stria terminalis (Figure 1-3). Portions of the limbic system are implicated in taste perception

(Finger et al. 2000), but there are currently no comparative data on changes in limbic and taste CNS structures as has been reported in relation to the olfactory system.

In addition to taste buds there are solitary chemosensory cells (SCC) that are similar to taste cells but are not located within taste buds. In aquatic species these cells have been located throughout the exterior body or in olfactory epithelium. In both aquatic and terrestrial invertebrates and vertebrates these cells are located in the oral cavity or within respiratory epithelium, and along the trachea (Finger et al. 2003;

Sbarbati and Osculati 2003). SCC‟s have also been located within the body and internal organs, and depending on location are responsible for a multitude of sensory reflexes

(Hofer et al. 1996; Hofer and Drenckhahn 1998).

Like olfactory receptors, SCC‟s require an aqueous environment and so contain either a thick mucus, electron dense substance or are located in a fluid medium such as an aquatic environment. SCC‟s synapse onto distal processes of remote ganglion cells

and are innervated by any cutaneous nerve including spinal and trigeminal (Whitear

1992). In an aquatic environment, these cells are typically involved in predator avoidance and habitat recognition responses, although they have also been implicated in food assessment (Kotrschal et al. 1997). Within terrestrial mammals they are used for airway protection and reaction to other abnormal changes in chemical composition of the digestive tract (Hofer et al. 1996; Hofer and Drenckhahn 1998; Finger et al. 2003)

Figure 1-3. Transduction of signals to and within the brain during taste with food in the oral cavity. Air flows indicated by dashed and dotted lines; dotted lines indicate air carrying odor molecules. ACC, accumbens; AM, amygdala; AVI, anterior ventral insular cortex; DI, dorsal insular cortex; LH, lateral hypothalamus; LOFC, lateral orbitofrontal cortex; MOFC, medial orbitofrontal cortex; NST, nucleus of the solitary tract; OB, olfactory bulb; OC, olfactory cortex; OE, olfactory epithelium; PPC, posterior parietal cortex; SOM, somatosensory cortex; V, VII, IX, X, cranial nerves; VC, primary visual cortex; VPM, ventral posteromedial thalamic nucleus. Adapted from Shepherd 2006.

In general, terrestrial and aquatic taste systems are very similar. In both systems, taste buds are located inside the mouth. However, there are examples from aquatic species including catfish barbels and lip taste buds in which taste buds are located outside the oral cavity. The connections from the receptor cells to the central nervous system are also similar. In fish, those taste receptors outside the mouth that are used for food localization synapse on the facial nerve whereas those inside the mouth that trigger swallowing are mediated by the vagus nerve (Finger 1997).

In summary, the signals perceived by the olfactory epithelium, VNO and taste buds can result in changes to behavior and physiology. Signals given off by an animal and perceived by another can be used for communication. In higher order terrestrial vertebrates, olfaction is the predominant form of reception, while in lower order aquatic vertebrates it is taste. It is not known if chemical communication occurs through taste in terrestrial vertebrates (Wyatt 2006).

Chemical Communication

Chemical communication can be used within and between species. Functions include reproduction, alarm, aggregation/recruitment, scent marking/territorial behavior, or social organization. The effect of climate change and pollutants on an animal‟s ability to detect its environment including predators, prey, or mates has had limited research

(Finger et al. 2000, Wyatt 2006). In recent years it has been found that increased temperatures and pollutants such as oil (Temara et al. 1999) and copper (Sandahl et al.

2007) can have major consequences to the senses of taste and smell. In general, a chemical used for detection, communication and engagement between individuals of the same species is termed a pheromone. These signals can both attract and deter potential mates depending on the age and condition of the animals involved. Chemicals

used for signaling are usually leaked metabolites and so are energetically cheap to produce and broadcast, which is why they are most likely ubiquitous among species to transmit reproductive signals (Bigiani et al. 2005).

Pheromones

The term pheromone comes from the Greek words pherein, to carry, and hormon, to excite. Initially pheromones were considered externally excreted hormones that acted between individuals. However, the chemical compounds that make up pheromones are much more complex than a single steroid hormone. The best way to specify the definition of pheromone is to break it into several categories of chemical signals that represent points on a continuum. The first is the traditional definition of a pheromone as a single chemical compound that elicits a behavioral or physical change in an individual. The second is a pheromone blend that consists of a mixture of structurally related compounds that only cause a change when in the correct ratio and the final category is a mosaic signal in which many compounds are needed in the right proportions to observe an effect (Johnston 1998). To aid in signal detection for terrestrial species, proteins are frequently necessary as accessory components. They can be used as odorant/pheromone binding proteins that transport an odor across the sensilar fluid of the olfactory epithelium or odorant-degrading proteins, which clear odors out of the fluid (Vogt and Riddiford 1981, Pelosi 1994).

There are many different situations in which a pheromone can be used, and depending on the category or type of response elicited, the chemical compounds used can vary. For example, many sex pheromones are considered to have evolved from leaked hormones that the opposite sex uses as a signal of impending reception

(Sorenson and Stacey 1999). Also, the opposite sex can take advantage of the signals

that are used in other areas, such as food perception. For example, the male oriental fruit moth attracts females using plant-derived chemicals that exploit the female‟s response to food (Lofstedt et al. 1989).

Pheromones can have immediate effects on receivers, known as a releaser effects, or elicit longer term physiological changes termed primer effects. The first mammalian pheromonal signals to be discovered were those that have an immediate or releaser effect on the behavior of the receiver such as the expression of a volatile chemical used by female elephants to attract males (Rasmussen 1999). Primer effects include the stimulation of sperm production in fish, which acts by stimulating hormone excretion, or termite caste determination which also acts through hormones, but instead of adults it affects levels of juvenile hormone development (Wyatt 2006). A single pheromone may act as both a releaser and/or primer and so the reactions can be very complex (Sorenson 1996).

Pheromones used by one sex to attract the opposite sex for the purposes of reproduction are known as sex pheromones and most likely evolved through means of sexual selection (Wyatt 2006). Insects have specialized neural circuits, known in moths as the macroglomerular complex, to process sex pheromones. Terrestrial vertebrates use a combination of the vomeronasal organ (VNO) and main olfactory epithelium

(MOE). These are both scent structures that receive chemicals in the environment

(Finger et al. 2000).Sex pheromones convey a combination of information to potential mates including receptivity, species, sex, genotype, and even health. These signals can be used by species with internal fertilization to allow females to choose the most beneficial males such as in the scorpionfly, Panorpa, in which males provide food to

females and broadcast this „gift‟ partially through pheromones (Thornhill 1979). Sex pheromones also are used to coordinate external fertilizers, such as observed in the mass spawning of corals that in part may be triggered by steroid hormone release

(Twan et al. 2006).

In contrast to sex pheromones, in which an individual is attracted to signals generated by the opposite sex, aggregation pheromones incite response to a signal from either sex. These attracting pheromones are important in many invertebrate species both marine and terrestrial. This could be useful for the recruitment of larva such as in barnacles, Balanus amphitrite, which use a species-specific settlement inducement protein complex to attract larva to a particular settlement area (Dreanno et al. 2007). Another form of aggregation pheromone is used by the tree-killing bark beetle, Pityogenes bidentatus, to attract large numbers of the beetles to host trees, thereby overcoming the tree's defenses. To stimulate aggregation, these beetles use pheromones containing methyl and alcohol groups and host tree component chemicals as well as chemicals given off by symbiotic bacteria in the digestive system of the beetles (Byers 1995). On the other hand, species use aggregation pheromones to group together as defense against predation, such as in the California spiny lobster,

Panulirus interruptus (Zimmer-Faust et al. 1985). Another form of aggregation is used by males of a species to respond to male sexual pheromone signal. The signal is initially intended to attract females, but aggregation occurs when other males respond as well (Wyatt 2006).

Alarm signals result in a fight or flight response by an individual which can be in response to predators or invasion of a nest. These signals are not as altruistic as the

name implies. The pheromone given off by a prey species serves to both deter predators and cause a change in behavior by conspecifics to increase potential escape by the signaler. Thus the signal may result in another individual falling prey to a predator because of sudden movement or lack thereof. These signals can be used for multiple purposes. For example, the ant, Formica rufa, uses a combination of formic acid and saturated hydrocarbons to defend and alarm nest mates to danger, recruit nest mates or repel enemies (Lofqvist 1976). The sea anemone, Anthopleura elegantissima, releases an alarm pheromone, anthopleurine, when attacked to cause contraction of the tentacles of surrounding anemones (Howe 1976). Alarm pheromones also serve to make prey unpalatable and many species respond to compounds given off in the blood which indicates an attacked conspecific (Wyatt 2006).

Scent marking uses specialized glandular secretions to mark a territory.

Territoriality is especially effective in social insects because the colony can be in several places at once and therefore can defend a relatively large area. Ants will also defend trails that lead to valuable resources (Holldobler and Wilson 1990). Although some social insects do use pheromones to mark or defend territory this is predominantly a feature of mammals and vertebrates in general (Wyatt 2006). It may also be that the territorial pheromone use of insects and aquatic invertebrates has not been fully examined yet since the scent marking of mammals is such an obvious behavior.

The best example of pheromone use for social organization comes from the honeybee, Apis mellifera. There are several pheromones used by the honey bee to maintain the hive. The most important of these is a signal excreted by the queen known as queen mandibular pheromone (QMP) that is made up of five principal chemical

components (Winston and Slessor 1992). The pheromone is relatively nonvolatile and is incorporated into the colony by messenger bees that pass the message on through antennal contact with other bees in the hive. It indicates that a bee is a member of that nest and inhibits reproduction by other bees. The pheromone is also short lived and is continually produced by the queen. If the queen dies, then the pheromone is no longer available and production of a new queen will begin. This pheromone is considered an honest signal that allows a hive to function efficiently with only one queen. QMP is used by many social insects including other bees, Apis sp., and fire ants, Solenopsis invicta

(Vargo and Laurel 1994).

In all of the pheromones described above it is very important to have a differentiation between species to prevent inappropriate mating, and in individuals to prevent attack on a member of the nest or family. For this reason animals have developed the ability to discriminate between very small variations in chemical structure.

The pheromone blends or mosaics used can be extremely complex (Finger et al. 2000,

Wyatt 2006). The study of the development and evolution of these chemicals and receptors provides fascinating insight into phylogenic relationships as well as divergent and convergent evolution.

The expression of chemical signals can occur through normal processes such as urination and defecation or can involve dedicated structures such as anal glands. The use of chemical signals is ubiquitous among aquatic and terrestrial animals. The type of signals used to broadcast information differs because of the variation in chemical behavior between terrestrial and aquatic environments. Pheromones in terrestrial systems are generally small, volatile compounds such as acetates or acids. Those used

in aquatic systems are also small but are not volatile and are easily transported in water, for example small peptides or steroid hormones (Wyatt 2006). These principles depend on the intended use of the compound, for example the honeybee‟s QMP is nonvolatile because it is passed through contact and not the air. Although generalizations can be made about terrestrial versus aquatic and the purpose of certain chemical signals, the exact composition of those signals is usually very complex and there are still factors of pheromone mixtures and mosaics that are not completely understood today.

Mammalian Reproductive Chemical Communication

Mammals have a wide variety of chemical signals to broadcast their reproductive status to the same and opposite sex. One of the most interesting components of pheromone mediated mate choice in mammals is the ability to detect single locus changes in major histocompatibility complex (MHC) genes. MHC genes, known as human leucocyte antigen (HLA) in humans are a component of the immune system.

Differing smells in human body odor and urine caused by changes in MHC can distinguish parents, siblings, half siblings and 50% of cousins from non-related mates

(Wyatt 2006). The more variation between MHC/HLA in potential mates, the more likely a mating is to occur. In humans, the stronger the variation the more attraction there is between potential mates (Wedekind et al. 1995).

There are several specific effects of male and female excretions that influence the reproduction of female and male mammals. These effects which include the Bruce,

Whitten, Vandenbergh, Lee Boot, and Coolidge effects have been demonstrated both in laboratory animals such as mice and hamsters, and in natural settings. The Bruce effect is a pregnancy block discovered in mice that results from females smelling urine

released by males. The scent of almost any male that did not fertilize the eggs of the female will result in a block of implantation of the fertilized eggs (Bruce 1960). The Lee

Boot effect is a female urinary pheromone mediated signal that inhibits estrus in adult female mice and delays puberty in juvenile female mice. This effect occurs in high density situations where males are limited (Vandenbergh 1999).The Whitten effect results in the induction of estrus in adult females by male pheromones while the

Vandenbergh effect accelerates puberty in young females through the same male pheromone signals (Ma et al. 1999, Novotny et al. 1999). Finally the Coolidge effect results in an animal‟s renewed interest in mating when presented with a novel mate to increase potential variability in genes (Johnston and Rasmussen 1984). Also, in populations that have a dominant and subordinate female the dominant female will generally suppress the subordinate‟s ability to ovulate (Wyatt 2006). In the common marmoset, Callithrix jacchus, this occurs through a combination of olfactory, visual and behavioral cues (Barrett et al. 1990).

As described earlier, the VNO is a specialized structure located in the rostral area of the head directly above the palate (Bertmar 1981). Various behaviors observed in mammals are used to deliver scents to the VNO; the most common is termed flehmen. Flehmen is a physical movement of the lips and tongue that allows for easier access of chemicals to the VNO (Ladewig and Hart 1980). It has been observed in many terrestrial mammals including goat (Ladewig and Hart 1980), horse (Stahlbaum and Houpt 1989), domestic cat (Bland 1979) and dogs (Dawley 1998). The opossum uses a behavior known as snuggling in which the snout is passed over an odor source

several times (Poran et al. 1993). Each mammal has its own method for directing scents to the VNO including flehmen, snuggling or some other behavioral technique.

Until recently the VNO was considered the processing center of pheromones while the MOE was for receiving traditional scents. However, it has recently been discovered that pheromones can be perceived by the MOE and therefore species without a VNO can respond to pheromonal cues (Kelliher 2007, Zufall and Leinders-

Zufall 2007). It may be that in species with a VNO this structure is necessary for the initial sensory perception, but once a response to signal has been established, the MOE is adequate as has been observed in the pig (Dorries et al. 1997).

There are a multitude of behaviors that result from pheromonal cues in mammals. These are essential to the individual broadcasting or responding to the cue in order to perpetuate their genetic material as efficiently as possible. These cues are thought to have evolved as a result of sexual selection over evolutionary time (Sorenson

1996, Rasmussen 1999). They are not unique to mammals, but the variety and complexity of these signals in mammals make the specific chemical combinations and resulting behaviors difficult to study. More research is needed on how those effects observed in laboratory settings correlates with natural conditions, and on the specific composition of each chemical signal. Very few mammal species have been studied in depth for their use of sexual pheromonal cues. Most of these studies have been based in the laboratory with such species as the mouse, rat or hamster. By comparison, the elephant is the predominant example of pheromonal cue use in a wild species.

Paenungulata

One of the most interesting phylogenetic relationships is contained in the clade

Paenungulata which includes three extant orders of mammals: Proboscidea

(elephants), Sirenia (manatees/dugongs), and Hyracoidea (hyrax) (Simpson 1945). To understand these relationships, it is important to examine the animal‟s evolutionary development and current common features. The genetic relatedness of these orders has been exhibited in hemoglobin sequences (Kleinschmidt et al. 1986) and chromosome painting (Pardini et al. 2007), while their common diet, vibrissae (Reep and Bonde 2006), lingual structure (Yoshimura et al. 2008), and many other physical characteristics are observable.

Despite the large differentiation between an aquatic and terrestrial lifestyle the manatee and elephant have a surprising number of characteristics in common. The most obvious is the herbivorous diet, which indicates similar taste structures to allow for detection of nutrients or toxins in their food. They also have a similar mother-calf pair bond in which the mother feeds the calf from axial mammary glands until it is about two years of age (Marmontel 1995, Emanuelson 2006). However, unlike manatees, elephants have a tightly knit social structure in which the calf is cared for by other adult females in the herd known as allomothering and has contact with its siblings and other calves. Elephants also remain in the herd well beyond weaning; for their entire lives if female and up to 20 years if male (Lee 1987). Manatee calves remain with their mother for 2-3 years and although they follow similar migratory routes do not overtly socialize with the mother after weaning (Marmontel 1995).

Although social herds are common in elephants and not in manatees, there is a similarity in the dispersed male-female social groups of elephants and the complete lack of male-female sociality except during reproduction in manatees. This absence of continuous sociality results in reduced male-male competition and, therefore, males

project their status and condition to females using other methods. Elephants tend to move in a home range that changes according to rainfall, vegetation or reproductive receptivity with a variation of 34 km2-6400 km2 depending on the herd (Schulte 2006).

The manatee predominantly migrates to warm water sites in the winter and vegetation dense areas in the summer with about 12% remaining resident in a site year round. The manatee is mostly reacting to changes in temperature for its migration, but vegetation and, perhaps, mate localization also affect migration (Deutsch et al. 2003).

In elephants the temporal glands on the side of the male elephant‟s head carry musth signal, which in combination with behavior indicates condition and position in the hierarchy (Schulte and Rasmussen 1999). A component of this broadcast signal is the volatile chemical frontalin. As an elephant ages, it develops differing ratios of the enantiomers or mirror image of the chemical structure for frontalin which initiates increased interest from dominant females. Therefore, an older and higher condition male will have more mating access to dominant females than the younger males

(Greenwood et al. 2005). These temporal glands and the musth produced in them are unique to the elephant, although many other species use glandular excretions to express their reproductive status. Manatee males may use anal or other glands to express their condition and females may broadcast their reproductive state to attract males.

To detect signals expressed in urine or other excretions, elephants have a distinct flehmen response in which the elephant touches a sample with the trunk and then places the trunk tip in their mouth, exposing the opening of the VNO to the sample of concern (Rasmussen et al. 1982). This adaptation allows the elephant to direct

scents of interest directly to their VNO and this results in more control over the scents on which an individual may focus. The manatee has also been known to exhibit flehmen-like movement of their muzzle but absence of a VNO indicates that this is likely for another purpose.

Elephants use infrasonic communication and have excellent hearing and vision

(Langbauer 2000), while manatees have limited near sighted vision and adequate hearing. However, the manatee‟s sense of touch is extremely sensitive and most likely their predominate form of sensory perception (Reep et al. 2002, Reep and Bonde

2006). Their sensitivity to touch may enable infrasonic sound detection like the elephant but there is no direct evidence to support this.

The lack of social interaction between female manatees precludes them from using signals such as the elephant‟s female-female bonding pheromones (Rasmussen

1999) or the extensive vocalizations that keep the herd organized and functioning

(Langbauer 2000). Also, the Asian elephant (Elephas maximus) has matriarch led herds that utilize information passed from generation to generation on food and water location

(Rasmussen 1999). Although the female manatee remains with her calf for 2-3 years and does pass on information about food and water, the breakup of this bond does not allow for the continued learning that is found in elephants. Therefore, the need to taste salt water gradients and differences in food sources is greater in the manatees than elephants because of the decreased time to acquire learned sources from the mother.

This would indicate a higher preference of taste for manatees while elephants have a much higher emphasis on smell.

The elephant has a highly developed VNO and olfactory system (Johnson and

Rasmussen 2002, Gobbel et al. 2004) unlike the manatee which does not have a VNO and has a reduced olfactory system (Lowell and Flanigan 1980, Mackay-Sim 1985,

Reep et al. 2007). During reproductive periods male elephants use their VNO and MOE to detect chemical signals given off by the females in urine that indicate receptivity.

These signals can include information indicating species, sex, individuality, signals to offspring and other females, and a quantitative amount of pheromone signal directed at males (Rasmussen 1999). The use of smell for these purposes by the manatee is unknown but has been hypothesized (Moore 1956, Hartman 1979, Rathbun and O‟Shea

1984) as has the possibility of an additional chemical sense such as taste or external receptors (Vosseler 1924, Hartman 1979).

Very little chemical communication research has been completed on members of the Paenungulata taxon other than the elephant. The hyrax is known to have a similar set of vibrissae to the manatee that could be used for both detecting their environment or communication (Sale 1970). The hyrax also has a vomeronasal organ (Stobel et al.

2010) and a gland on the dorsocaudal area of their back that has been linked to reproduction and attention from the young (Olds and Shoshani 1982). Unlike the manatee the hyrax live in family groups and unlike the elephant these are usually male lead, similar to rodents. In general, the importance of chemical communication to elephant reproduction and social hierarchy is the example by which future studies of

Sirenian and Hyracoidean research should follow.

Aquatic Mammals

Chemical senses were the first sense to appear in early evolution during the period of time when all species occupied an aquatic environment. Taste and smell

developed within those individuals and are evident in varying forms within archaic and derived aquatic species today. Variation in sensory use can occur for a variety of reasons including status in the food hierarchy, temperature, humidity, rainfall, or the medium in which a species lives. The largest effect on the function of the chemical senses is water. An aqueous as compared to terrestrial environment affects how chemicals are distributed and available for reception. Therefore, the greatest differences in taste and smell organ structure and function can be observed between aquatic and terrestrial species. This is most evident in the decreased or absent sense of smell observed in marine mammals (Freitag et al. 1998).

Olfaction

The pinniped olfactory system is less developed than terrestrial carnivores

(Oelschlager and Oelschlager 2002; Reep et al. 2007, Montie et al. 2009) and phocids and walrus, Odobenus rosmarus, have a reduced olfactory bulb compared to otariids

(Harrison and Kooyman 1968). However, this is based on preliminary examination of a limited number of pinniped brains and more detailed descriptions are needed because these results tend to contradict behavioral observations (Kowalewsky 2006).

For pinnipeds, olfaction is most likely important in mother pup interactions

(Phillips 2003) and perhaps foraging. Their ability to follow a direct tract to foraging grounds, even though the position of those locations change within a large area, led researchers to believe that chemoreception was being used. A study using two harbor seals, Phoca vitulina, indicated that they can detect picomolar concentrations of dimethyl sulphide, a chemical expressed in highly productive foraging areas of the ocean (Kowalewsky 2006). Also, territorial marking is thought to occur in the ringed seal, Phoca hispida, using facial glands. The combination of alcohols and lipids within

the samples indicate that they can be deposited on surfaces and detected as volatiles through olfaction (Ryg et al. 1992).

The trend towards a reduced olfactory bulb in phocids and walrus may be linked to the amount of time they spend with their young. For example, the hooded seal,

Cystophora cristata, a phocid, weans its pup in four days and does not leave them during this time (Bowen et al. 1985); whereas, the South American fur seal,

Arctocephalus australis, nurses for up to twelve months and will leave its pup for days at a time while foraging. Behavioral observations indicate that fur seals smell their pup upon return and before nursing to identify them (Phillips 2003). Walrus nurse at sea and are the only pinniped known to forage with their young (Fedak et al. 2002), which would preclude the necessity of reunions following foraging. Therefore, olfaction is likely less important in phocid and walrus mother-pup interactions but their general use of olfaction should not be dismissed. No research has focused on pinniped use of chemoreception in the aquatic environment for the purposes of territoriality or reproduction although anecdotal evidence indicates this may be the case (Reeves et al. 2002).

Anatomic evidence indicates that odontocetes do not have olfactory capabilities.

The few odontocete species examined do not have an accessory and olfactory bulb, olfactory tract or anterior olfactory nucleus as adults (Oelschlager and Buhl 1985,

Marino 2007, Oelschlager et al. 2008a). The regression of the odontocete olfactory system has been demonstrated through histologic examination of 13 harbor porpoise,

Phocoena phocoena, embryos and fetuses. At the 28.6 mm stage the olfactory fibers were fewer and the olfactory foramina in the cribiform plate smaller than at the 24 mm stage. As the fetus develops, the distinction between olfactory fibers and the terminal

nerve becomes more difficult to identify and at 167 mm total length the cells are in rapid regression (Oelschlager and Buhl 1985). Therefore, it appears that the bulb may have been present in early evolutionary lineages of the delphinids and is lost in present day species during embryonic development (Oelschlager and Buhl 1985). In addition to the regression of the olfactory neuroanatomy, one histological study characterizing the nasal complex of fifteen harbor porpoise, Phocoena phocoena, found no evidence of olfactory epithelium (Prahl et al. 2009). There are no published studies on the behavioral ability of the odontocetes to react to olfactory cues.

In addition to the anatomic evidence, the investigation of the odontocete genome for olfactory receptor genes corroborates gradual evolutionary loss of olfaction in odontocetes. Using eight species of toothed whales, a multigene tree of 115 olfactory receptor sequences from a diverse array of class II olfactory receptor paralogues was defined (McGowen et al. 2008). Several independent pseudogenization events support the theory of the loss of an olfactory sense over evolutionary time. This genomic and anatomic evidence in addition to the lack of behavioral accounts of odontocetes using olfaction indicates that between the Miocene and present day these species lost the ability to smell. This is invariably due to the lifestyle and environment of these animals, in addition to the energetic cost required to maintain an olfactory system.

Within the baleen whales, Mysticetes, Oelschlager (1992) examined embryos and early fetuses from three different species; blue, Balaenoptera musculus, fin,

Balaenoptera physalus, and humpback, Megaptera novaeangliae whale. Unlike

Odontocetes, the olfactory bulb, tract, and epithelium of the Mysticetes did not regress and was well formed and evident in a 105mm crown-rump length (CRL) fin whale fetus

(Oelschlager 1992). In addition to the embryologic evidence, of adult brains from 11

Bowhead whales, Balaena mysticetus, (Duffield et al. 1992) and a single Humpback

(Breathnach 1955) contained olfactory peduncles, nerve fibers and small holes in the cribiform plate of the ethmoid bone, which would allow the olfactory nerve to pass into the brain. In a recent study, the Bowhead whale olfactory bulb was located and measured 0.13% of the brain weight and 51% of the olfactory receptor genes were intact (Thewissen et al. 2011).

In addition to the embryologic evidence, adult brains of 11 Bowhead whales,

Balaena mysticetus, (Duffield et al. 1992) and a single humpback (Breathnach 1955) demonstrate the presence of olfactory peduncles and nerve fibers but no olfactory bulb.

The inability to find the bulb is presumed to be due to the size and condition of the available brains; its presence is conjectured because of small holes in the cribiform plate of the ethmoid bone which would allow the olfactory nerve to pass into the brain

(Duffield et al. 1992, Breathnach 1955). In a recent study the olfactory bulb was located and measured 0.13% of the brain weight. In addition, 51% of the olfactory receptor genes were intact (Thewissen et al. 2011)

Anecdotal behavioral evidence and a highly reduced but intact olfactory system indicate that Mysticetes retain some sense of airborne smell (Oelschlager 1989). Baleen whales spend a majority of their time at or near the surface to feed, unlike their toothed counterparts that dive deep to locate prey (Berta et al. 2006). Therefore, the use of airborne scents to locate groups of zooplankton or other prey as well as perhaps locate signals given off by predators or mates would correspond with the anatomic evidence of the Mysticetes. It would be interesting to see if there is a variation in olfactory size

between those Mysticetes that skim off of the surface such as the bowhead and those that feed on invertebrates from bottom sediment like the gray whale (Berta et al. 2006).

Berta et al. (2006) state that Sirenians have a larger olfactory bulb than cetaceans and pinnipeds; however, Reep et al. (2007) demonstrated that manatees have a smaller olfactory system than pinnipeds. This study did not include comparisons to dugongs or cetaceans and it would be interesting to complete the comparison using representatives of these taxa.

Taking into account the habitat of the respective marine mammals that sirenians would be expected to have a larger olfactory bulb than Mysticetes but smaller than pinnipeds. This comparison correlates with the aquatic lifestyle of Sirenians and with what has been observed in other marine mammals. In this regard, a portion of the brain is still dedicated to the processing of scent (Reep et al. 2007), and since the manatee has limited opportunity to take in airborne scents, e.g., only during a breath, then this center may be adequate for the manatee to process important scent information during feeding, copulating, and simultaneous breathing with other animals.

Vomeronasal Organ

There are limited anatomic studies examining the olfactory capabilities of pinnipeds. Although it is widely reported that they possess a VNO (Kowalesky et al

2006, Mackay-Sim et al. 1985, Meisami and Bhatnagar 1998) no references have an examination of the pinniped VNO or its development. Most commonly referenced information on the VNO in pinnipeds include a note on the pinniped olfactory bulb but do not mention a vomeronasal organ or accessory olfactory bulb (Harrison and Kooyman

1968, Oelschlager and Oelschlager 2002). In addition, an examination of the brain of a live Californian sea lion using magnetic resonance imaging gave no indication of a

vomeronasal organ or accessory olfactory bulb, although olfactory structures were documented (Montie et al. 2009).

During an examination of lateral olfactory tract fibers, in relation to the accessory olfactory bulb, an accessory olfactory structure was observed in the Alaskan fur seal,

Callorhinus ursinus, Steller‟s sea lion, Eumetropias jubata and California sea lion,

Zalophus californicus. No accessory olfactory tract was found in the Harbor seal, Phoca vitulina (Switzer et al. 1980). It appears that pinnipeds do have an accessory olfactory bulb and therefore a VNO but more research is needed to compare the development, size and functionality of that organ to those in other well studied mammals.

Odontocetes do not have a VNO. There is no evidence of development in embryos or fetuses in harbor porpoise (Oelschlager and Buhl 1985) nor is there any evidence in adult harbour porpoise (Marino et al. 2001). Mysticetes do not have a VNO or evidence of a vomeronasal nerve during development or as adults (Oelschlager

1989). In Sirenia, the Florida manatee does not possess a VNO (Mackay-Sim et al.

1985) and there is no literature examining the dugong. However, both mysticetes

(Thewissen et al. 2011) and sirenians (Mackay-Sim et al. 1985) have a small olfactory bulb. In sheep and pigs, the female is still attracted to the male and still has a luteinizing hormone surge even if the VNO has been severed (Wyatt 2006). This indicates that marine mammals that have olfactory capabilities may use these to detect and communicate with potential mates even though their VNO is absent.

Taste

Examination of a California sea lion, Zalophus californianus, indicates that pinnipeds have a sensitivity to sour, bitter, and salty but not sweet (Friedl et al. 1990).

Umami was not tested although they are most likely sensitive because of their diet.

pinnipeds are able to detect saltwater gradients within a 4% variance at levels of 30 parts per thousand as determined through a go/no-go paradigm of two harbor seals

(Sticken and Dehnhardt 2000). According to Yoshimura et al. (2002), the California sea lion has taste buds on vallate and fungiform papillae. Although behavioral evidence indicates that pinnipeds do have a sense of taste, more definitive evidence of taste bud location, structure, density and size is needed.

In a single sample from a young, 1.5-year-old, male walrus, it was observed that there are very few (140 or less) but large (75 X125 µm) taste buds on the walrus tongue

(Kastelein et al. 1997). This would place the walrus‟ sense of taste intermediate between the harbor seal with the greatest amount and the California sea lion and grey seal with the least (Sonntag 1923). The walrus uses a suction feeding method, and so it is likely that mechanoreception is more important than gustation. However, captive walruses do exhibit selectivity in food choice and so may be using taste under certain circumstances (Kastelein et al. 1997)

Bottlenose dolphins have taste buds in pits at the base of their tongue; these pits develop from circumvallate papillae that are present in neonates (Kuznetzov 1990).

Also, they have a highly developed gustatory nucleus of the thalamus (Kruger

1959).Go/no-go paradigms using a single adult captive male have indicated that bottlenose dolphins can respond to the four common tastants; salty, sour, sweet and bitter at an order of magnitude below human capabilities. Another study using a similar paradigm demonstrated that a single 3-5 year old male dolphin had a threshold of

.0173M for a sour tastant and 1.36X10-5M for bitter, comparable to human thresholds

(Nachtigal and Hall 1984). Umami has not been tested.

Several Russian researchers have hypothesized that because of the limited number and specialized structure of dolphin taste receptors they may use taste for chemical communication (Friedl et al. 1990). However, no form of chemical communication has been tested for using dolphins.

Other Odontocete species have been examined for taste buds with mixed results. Harbour porpoises (Phocoena phocoena), Amazon River dolphins (Inia geoffrensis), and beluga whales (Delphinapterus leucas) were discovered to have a novel microvilli structure in the papillae, but no taste buds were observed (Kuznetzov

1990). No taste buds were found in the Indus river dolphin (Plantanista gangetica),

(Arvy and Pilleri 1970), La Plata dolphin (Pontoporia blainvillei), (Yamasaki et al. 1976), the sperm whale (Physeter macrocephalus), or the long finned pilot whale

(Globicephala melas) (Pfeiffer et al. 2001). However, the examination of the pilot whale did not include the tongue root, the area where most taste buds are located in cetaceans.

Taste buds have been located in a newborn male Stejneger‟s beaked whale,

Mesoplodon stejnegeri (Shindo et al. 2008) and striped dolphins, Stenella coeruleoalba, of varying sex and age (Komatsu and Yamasaki 1980). In Baiji river dolphins, Lipotes vexillifer, taste buds were located with fewer found in a young adult female than a calf and none in an older adult female (Li 1983). The conflicting evidence of taste bud presence in Odontocetes could be due to a number of factors. The small size of the tongue and relatively higher density of taste buds in fetal and newborn dolphins enables easier identification of taste buds in young animals than the limited number of taste

buds and larger tongue size of adults. Another possibility could be a regression of taste buds post weaning.

These examinations involved small sample sizes and limited comparisons between age and sex. There were also many instances of taste bud localizations in one species‟ sample and none in another. Therefore, the true extent of individual species taste ability and use is difficult to determine currently but in general, odontocetes have the ability to taste and may be using this in food assessment or mate recognition. A combination of behavioral and anatomic analyses of taste buds on the bottlenose dolphin (Friedl et al. 1990) provide the most reliable and useful results, although not feasible with most species.

There are no formal published reports on the use of taste or presence of taste buds in mysticetes. It is assumed that taste is used by mysticetes, perhaps more than odontocetes, since the use of baleen allows for food to be present in the mouth for a longer period of time, giving the animal the chance to expel any undesirable food.

The dugong has taste buds in pits of the lateral margins of the tongue (Yamasaki et al. 1980). They live in a marine environment and so may also use taste for freshwater detection and food palatability, but it is not known if their chemoreception is used during reproductive behavior. The Florida manatee has taste buds in the foliate papillae of the tongue root (Levin et al. 2002). In general, Sirenian taste buds are more numerous than those of cetaceans or pinnipeds (Yamasaki et al. 1980).

There has been no published examination looking for taste buds or solitary chemosensory cells in any area besides the tongue in any marine mammal. It would be advantageous to have some kind of sense organ outside the mouth to assess

palatability of food, locate mates, and determine salt water gradients, especially for those species living in a marine environment in which salt balance is a concern.

Of the approximately 125 known marine mammals of the world, 22 are listed under the endangered species act as threatened or endangered. Over the past 60 years, three marine mammals have gone extinct including the most recent Baiji River

Dolphin in China. For each of these species, the cause of their extinction or vulnerability is in some way human related, including pollution, vessel strikes, and bycatch (Marine

Mammal Protection Act 2010). Considering the precarious situation that these animals are in and the known influences of environmental contaminants on chemoreception, it is vital to learn as much as possible about their biology in order to decrease any negative effects of pollution and/or climate change. An endangered species that provides the opportunity to explore the use of taste in smell in a fully aquatic mammal is the Florida manatee.

The Florida Manatee

The endangered Florida manatee, Trichechus manatus latirostris, presents a unique opportunity to study the use of chemoreception by aquatic mammals. The animals are available in Florida in the wild, as live animals maintained in captivity, which are available for behavioral assays, and as postmortem carcasses. Furthermore, it is one of many marine mammals suspected of using chemosensation for reproductive purposes (Rathbun and O‟Shea 1984, Reynolds and Odell 1991, Hartman 1979, Moore

1956).

The reproductive behavior of the Florida manatee, Trichechus manatus latirostris, is not well understood due to the difficult nature of studying reproduction in a low visibility, aquatic environment of a sparsely populated endangered species. In

addition, the manatee has limited social interactions, which include mother/calf bonds and mating herds in which many males follow a receptive female for up to four weeks

(Hartman 1979, Reep and Bonde 2006). It is the formation of these mating herds that is of interest, as it has been observed that males will travel appreciable distances to track females in estrus. Although the manatee may mate at any time of the year, peak sexual activity occurs during the warm spring and summer months when manatees have dispersed from their warm water refuge sites (Reep and Bonde 2006). Therefore, it is assumed that there is a signal, presented by the female, which alerts males to her receptiveness.

Although this signal could be visual or auditory, the manatee has limited vision and is reported to be near sighted, having vision only slightly better than river dolphins

(Supin et al. 2001). In addition, manatee vocalization generally occurs in mother and calf communication although it has been observed in mating herds (Hartman 1979).

Their vocalizations consist of short chirps or squeals in the range of 3-24 kHz (Nowacek et al. 2003) and would most likely be lost in the estuarine environment of the manatee over long distances. A more likely source of this signal would be chemical cues given off by the female.

There have been a number of documented behavioral observations of the manatee indicating that the female does produce chemical signals. Among the earliest came from Dr. Vosseler, a German veterinary practitioner who was responsible for the care of a male and female manatee in captivity together. He noted that when the female was in estrus the male would swim at the bottom of the tank moving his nose and lip, both lateral and medial parts, so that the inner lip was exposed. In addition, the male

would always find the locations along the bottom of the pool where the female‟s genitals had contacted (Vosseler 1924). Moore (1956) noted a male manatee “exploring” with his muzzle a female manatee‟s genital area. Hartman (1979) discussed the prevalence of mouthing between manatees and hypothesized about their chemoreceptive abilities.

Furthermore, there has been documentation of manatees in Crystal River, FL rubbing each year on the same submerged logs and stones. These actions were observed to occur most frequently around the genitals, eyes, arm pits, and chin. It was hypothesized that females, which rubbed more frequently than males, were leaving a chemical signal of their reproductive state (Rathbun and O‟Shea 1984).

The sum of the behavioral accounts indicate that a female manatee may be able to broadcast her reproductive state utilizing secretions and/or excretions. The most likely source of this signal is urine, as this is the broadcast used most frequently by mammals (Wyatt 2006). Other potential sources include anal or vaginal glands

(Marmontel 1988). A steroid or peptide derivative secreted in the mucus of the vagina could stick to those rubbing posts and tanks on which manatees have been observed rubbing.

Anatomic evidence also indicates that the use of taste and/or smell for reproduction is feasible. The Florida manatee has taste buds in foliate papillae located on the side of the tongue root (Levin and Pfeiffer 2002 ) and most likely use taste for detecting salt water gradients or seeking specific nutrients from vegetation (Reep and

Bonde 2006). Their herbivorous diet increases their vulnerability to toxins that can be detected by a strong bitter taste (Nei 2008) which is generally the function of foliate papillae (Collings 1974). Touch is probably their predominant sense for food detection

but taste is likely supplementary. It has long been thought that males use chemoreception to find female manatees in estrus and so it is not known if they use their sense of taste or smell or a combination to do this (Levin and Pfeiffer 2002, Reep and Bonde 2006). In addition, it has been hypothesized that the manatee possess modified taste or smell receptors on the lips or roof of the mouth due to their behavior of mouthing each other and objects (Vosseler 1924, Hartman 1979).

Although the VNO does generally play a role in pheromone detection, the MOE is an equally important part of the signal detection and capable of pheromone reception without the VNO (Wang et al. 2007). This may have direct implications for the Florida manatee, especially with regard to animal husbandry. In laboratory mice the MOE was implicated in long range detection while the VNO was used during close contact estrus detection (Achiraman et al. 2010). The male manatee may detect female presence from long distances using chemical signaling and within close contact using touch. Therefore, the manatee‟s MOE (Mackay-Sim 1985) could be the site of these signal transductions and would explain the continued presence of an olfactory bulb (Mackay-Sim 1985). In addition, it has been hypothesized that the manatee may possess modified taste or smell receptors on the lips or roof of the mouth due to their behavior of mouthing each other and objects (Vosseler 1924, Hartman 1979).

The Florida male manatees‟ innate ability to find a female in estrus in addition to their social actions when in a mating herd indicates that they use chemosensation. This could be through taste and/or smell receptors and it is not clear which pathway is most used. The combination of manatee availability relative to other marine mammals in

addition to the behavioral evidence warrants an in depth examination of their olfactory and taste capabilities.

Through histological examination of suspected sites of chemoreception and signal transmission in addition to behavioral observation of the reaction of captive male manatees to female manatee urine, a thorough analysis of the manatees‟ chemoreceptive abilities is possible. This will allow for conservation efforts to target those areas of the environment necessary for the highest fecundity. This investigation provides insight into the manatee‟s chemosensory capabilities especially with respect to reproduction. Concurrently, an appreciation of the use of chemoreception in fully aquatic mammals has been documented for the first time.

CHAPTER 2 ANAL GLANDS OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS: A POTENTIAL SOURCE OF CHEMOSENSORY SIGNALS

Background

Anal glands have been identified in a variety of terrestrial and aquatic species

(McColl 1967) with detailed anatomic description in a limited number of animals

(Montagna and Parks 1948, Eglitis and Eglitis 1961, Greer and Calhoun 1966). Anal glands have been observed in all species examined, however, in humans the glands are proportionally smaller than in other animals and appear to have either limited or no function (Eglitis and Eglitis 1961, McColl 1967). All glands from the tissue surrounding the anus and anal canal are a combination of apocrine and sebaceous secretory units and can be found anywhere along the anal canal or in perianal tissue (Eglitis and Eglitis

1961). Many of the glands have been modified from sweat glands. They can be found singly, with many openings into the canal such as in pigs, Sus scrofa (McColl 1967) or the various secretions can collect in an anal sac and be secreted through a single large opening such as in the cat, Felis domesticus (Greer and Calhoun 1966). This duct can end in a nipple, as observed in the black-tailed prairie dog, Cynomys Ludovicianus

(Jones and Plakke 1981) or as a single opening commonly observed in dogs, Canis familiaris (Montagna and Parks 1948).

Secretion by anal glands appears to occur through a combination of myoepithelial cells surrounding the adenomeres (Atoji et al. 1998) and the position of the glands between internal and external sphincter muscles (Montagna and Parks,

Jones and Plakke 1981). Therefore, the secretion can be through both voluntary and involuntary muscle movement.

Anal glands are generally accepted to be used for the expression of reproductive chemical signals but secretions can also be used to identify species, sex, or produce odors for defense or subduing prey (Eglitis and Eglitis 1961). Anal glands have never been associated with a function other than those related to chemosensation. Animals that use anal glands for reproductive communication typically produce varying amounts and types of secretions in relation to their reproductive cyclicity (Gorman et al. 1978).

Previous research on anal glands of terrestrial animals demonstrated their use for territorial marking in canines (Donovan 1969), the beaver, Castor fiber, (Rosell and

Schulte 2004), and the Eurasian river otter, Lutra lutra (Gorman et al. 1978). Although the beaver and river otter inhabit both terrestrial and aquatic environments they appear to use these secretions only on land (Rosell and Schulte 2004, Gorman et al. 1978). In fish, anal glands are used by males of the Blenniidae family for reproduction. Several species of male blennids have a pair of anal glands found on either side of the anal fins and behavioral testing has demonstrated female mate choice is based on the presence of these glands and their contents (Barata et al. 2008). The function of mammalian anal glands in an aquatic environment has never been examined.

The Florida manatee, Trichechus manatus latirostris, is an endangered, fully aquatic, herbivorous mammal in the order Sirenia. During a routine necropsy of a young male manatee there appeared to be glands near the anus. This initiated an examination of this region for glands. Anal glands have never been described in the manatee or any other fully aquatic mammal although there has been reference to perianal glands with ducts emptying directly into the water of the Black Sea bottlenose dolphin (Kuznetsov

1974) but no further detail was provided. This, in addition to a reference to mucous anal

glands in the common seal, Phoca vitulina, (McColl 1967) are the only published accounts of anal glands in cetaceans or pinnipeds.

Manatees will be the first fully aquatic mammal to have their anal glands morphologically characterized in detail. The goal of this study is to determine the structure of the anal glands and generate functional hypotheses through histochemical analyses and comparison to other species.

Materials and Methods

Samples were collected from freshly dead manatees that died from boat strikes, cold stress or natural causes and were brought to the state of Florida‟s Marine Mammal

Pathobiology Laboratory in St. Petersburg, FL. Eleven animals, six female and five male, ranging from a fetus to a large adult were examined (Table 2-1). All animals were in good body condition except for three chronically ill females injured by watercraft that were emaciated: MEC0983, LPZ102765, and LPZ102654.

Table 2-1. Samples used for anal gland analysis. NBF=Neutral Buffered Formalin. All information is available from the Marine Mammal Pathobiology Laboratory Manatee Mortality Database1. Adults were measured as a straight length while the fetuses were crown-rump length. Animal Sex Age Fixation Cause of Death Date of Death Number Class/Length (cm) SWFTm0826 F Adult/261 10%NBF Chronic Watercraft Injury October 2008 LPZ102654 F Adult/261 10%NBF Chronic Watercraft Injury December 2008 MEC0983 F Adult/275 10%NBF Chronic Watercraft Injury July 2009 MNE0939 F Adult/290 10%NBF Acute Watercraft Injury August 2009 LPZ102765 F Adult/267 10%NBF Chronic Watercraft Injury August 2010 2%Glutaraldehyde Unfixed/Frozen MSE0910 F Calf/188 10%NBF Cold Stress January 2009 LPZ102639 M Adult/309 10%NBF Chronic Watercraft Injury November 2008 MNW0905 M Juvenile/210 10%NBF Natural March 2009 MSE0943 M Juvenile/205 10%NBF Cold Stress April 2009 M-187 M Fetus/30 10%NBF Aborted Fetus February 1980 F-MSW03160 M Fetus/39 10%NBF Acute Watercraft Injury October 2003 1Manatee mortality database available online at: http://research.myfwc.com/manatees/

Tissue surrounding the anal canal to several centimeters below the anorectal junction was removed. All samples but one were fixed in 10% Neutral Buffered Formalin

(NBF). From one animal, four one square centimeter sections of anal gland were removed and frozen fresh for lipid staining, while three one cm by one cm by two mm samples were fixed in 2% glutaraldehyde - 0.1 M phosphate buffered solution for transmission electron microscopy (TEM). The rest of the tissue was fixed in 10% NBF.

Light Microscopy

The anal tissue samples were examined grossly using a dissecting microscope and measurements taken on the size and distribution of the anal glands. Samples of the glands and ducts were embedded in paraffin and sectioned at 5-30µm. The 5µm sections were stained with hematoxylin and eosin (H&E) to outline the general anatomy of the tissue, periodic acid Schiff (PAS) to stain glycogen rich material and identify mucous rich secretions, or Masson‟s trichrome to differentiate between muscle fibers and the collagen of connective tissue. Sections were mounted on superfrost plus slides and stained using Santa Cruz Smooth Muscle Actin (HUC1-1) with an Invitrogen

Histostain plus kit (AEC) to identify myoepithelial cells. The 20-30µm sections were stained with hematoxylin and eosin. Extremely thick sections of approximately 120µm were cut on a slab microtome and stained with hematoxylin or trichrome to determine the extent of the duct system. Unfixed frozen samples of the anal glands were cut at 6

µm using a cryostat and stained with oil red O to identify lipids.

Transmission Electron Microscopy

sections between 70-90 nm were sectioned using an ultramicrotome and mounted on a

100 mesh Formvar coated grid. The sections were stained with uranyl acetate and

Reynold‟s lead citrate and examined with a transmission electron microscope (Hitachi H

7000).

Results

All of the manatees in this study, regardless of age or gender had diffuse, large, apocrine glands at the anorectal junction. The anal glands were found within both the internal and external sphincter muscles of the anal canal between 3 and 7 cm cranial to the anal sphincter. They were diffuse and found in clusters that cover a large area, reaching 3-5 centimeters in total length and 2-3 centimeters at the widest point. They appeared to have a tapered shape, forming two diamond patterns between the sphincter muscles cranial to and at the anorectal junction. They were composed of aggregated glands contained within individual fascia that emptied into several collecting ducts (Figure 2-1) ranging in size from 3-170 µm in diameter. The opening of the duct was not readily visible grossly because of the convoluted folding of the integument

(Figure 2-2).

Figure 2-1. Adult female (MEC0983), collecting duct (CD) approaching the anorectal junction. Scale bar equals 0.5cm.

Figure 2-2. Ventral view of a male manatee calf with an inset of the current working diagram of anal gland anatomy. Scale bar approximates 1 cm. A) anal glands, B) external sphincter muscle, C) internal sphincter muscle, D) Anal Canal, E) Anorectal junction, F) Rectum, G) Excretory duct, H) Anal orifice.

Both males and females in the study had extensive anal glands. Those glands collected during fall and winter months were smaller with less obvious secretory production such as in LPZ102654 which had small, compact glands without a discernable duct system. Large, adult animals tended to have larger glands and one female, MEC0983, had the largest and most obvious duct system. Like adults, the two male fetuses, estimated mid-term gestation, had two sets of glands, one on each side of

the canal that tapered from caudal to cranial orientation. However, unlike adults, one fetal specimen, (F-MSW03160), had glands that were completely ventral to the anal canal (Figure 2-3) while in the other fetus (M-187) they were on either side, cranial and caudal to the canal.

Figure 2-3. Anal glands (AG) of male fetus F-MSW03160 completely ventral to the canal. From cross-sectional cut across the anal canal. Scale bar equals 0.5cm.

Light Microscopy

The anal glands of the examined manatees consisted of branched tubules

(Figure 2-4), which collected in large bilayered ducts that emptied directly into the anal canal (Figures 2-4 and 2-5). These ducts ranged in diameter from 3-170 µm, increasing as they approached the canal. The glands and ducts are embedded in both smooth and skeletal muscle. At the secretory portion the epithelium was simple cuboidal (Figure 2-

6). The strong eosinophillic nature of the glands and luminal contents (Figure 2-6A) indicated a protein ladened substance. The glands produced a mucous rich secretion as can be observed in the PAS stain (Figure 2-6B) while the oil red O stain indicated that lipids are present in the secreting cells and lumen(Figure 2-5C). Trichrome staining

demonstrated the encasing of the glands in muscle followed by connective tissue as they progressed toward the lumen of the anal canal (Figure 2-5D). The myoepithelial cells surrounding the adenomeres can be seen in the anti-smooth muscle actin antibody stained tissue (Figure 2-7).

Figure 2-4. Collecting duct (CD) and its contents. H&E of adult female MEC0983. Note the large size and branching of the duct as well as its lining consisting of two cell layers. Scale bar equals 100µm.

Figure 2-5. Collecting duct (CD) joining anal integument as an excretory duct. H&E stain of anal glands of MEC0983. Note the Proximity of nerve bundles (NB). Scale bar equals 100µm.

Figure 2-6. Various histologic stains of adult manatee anal glands. Scale bars equal 100 µm. A. Adult female, MEC0983 H&E. Single layer cuboidal cells visible. Positive eosin reaction within the luminally stored secretions indicates presence of protein. B. PAS stain of adult female, SWFTm0826. The deep purple reaction within the glands indicates mucus while the small clears spots within and surrounding the purple stain indicates other material. C. Adult female, LPZ102765. Oil red O stain. The brightly red reaction reveals lipid rich material. D. Adult female, SWFTm0826. Masson trichrome. Red stains muscle and glandular cells, while green stains connective tissue. The glands are encased in connective tissue and surrounded by muscle.

Figure 2-7. Smooth muscle actin stain of adult female LPZ102765 demonstrating the location of myoepithelial cells. Scale bar equals 100µm.

Electron Microscopy

Transmission electron microscopy demonstrated apical caps separating from the cells into the lumen (Schaumburg-Lever and Lever 1975). Cells that appeared to be actively secreting had fewer organelles apically and changed in morphology and number of microvilli compared to other cells. Myoepithelial cells, often with indented nuclei, formed an incomplete basal layer associated with the secretory tubules (Figure

2-8B).

Figure 2-8. TEM of manatee anal glands. The apocrine secretion is evident because of the Apical Caps (AC). A) Microvilli (M), Scale bar 5 µm B) Scale bar 10µm, a myoepithelial cell indicated (MYO).

Discussion

While the sample size was small and did not allow for statistical analysis, there were no apparent differences between males and females. However, seasonal differences are suggested by the smaller size and less productivity of those samples collected during fall and winter months. The samples collected from adult animals in spring and summer months were the largest and had the most secretory material within the cells and lumen. This included two animals, MEC0983 and LPZ102765, which were chronic boat strike victims and were emaciated at the time of necropsy. The males

examined were young, or the samples were collected in late fall (see Table 1) thus in all males examined the glands appeared smaller and had less secretory material than in spring and summer females. Further comparison of specimens of various ages and sex should determine whether gender differences exist and further demonstrate seasonal variation.

Manatees do not have distinct breeding seasons but do tend to mate during the warmer spring and summer months throughout the Southern United States (Reep and

Bonde 2006). The suggestion of a seasonal change between those animals which died in colder versus warmer months indicates that the animals may be using their glands to communicate with mates. This is similar to the behavior observed in the Eurasian river otter (Gorman et al. 1978). However, the otter deposits its secretions on land, while the manatee would be depositing into an aqueous environment.

The large size, productivity, and change with season of these anal glands indicate that they are in use and may play important roles for the manatee. The large number of glands and storage ducts observed in close proximity to the epidermis suggests that substantial quantities of secretory material can be released at any given time. The location of the glands between muscle layers allows for secretion of contents during defecation. It may be that manatees deposit the secretions into the feces and the buoyant nature of the lipids allow for transport in the water; whereas the mucous would allow secretions to remain on submerged objects for longer periods of time. It is not known if the secretions can be actively expressed without defecation. The fact that the portion of the sphincter with which this gland is associated contains skeletal muscle suggests that it could.

Anal fistulas and cancerous tumors associated with anal glands are a problem in cats (Greer and Calhoun 1966) and dogs (Williams et al. 2003) and have been documented in humans (McColl 1967). Of all glands examined in these manatees, none showed evidence of impaction or infection. The relatively small and well protected emptying ducts indicate that infections typically observed in other mammals are of limited concern to the manatee. However, attention should be paid to anal gland health during post-mortem evaluation of manatees to determine prevalence of these conditions.

The size of the fetuses examined here indicate that both were gestationally mid- term animals, which in manatees would be approximately 6 months of prenatal development (Rathbun et al. 1995). In humans the development of anal glands begins in the early second trimester (fourth month) (Eglitis and Eglitis 1961). Therefore, the development of these glands appears to follow similar embryologic development as observed in humans. The manatee fetal glands appear to develop at or near the ventral surface and may move dorsally to surround the anal canal. However, more specimens are required to corroborate this observation and demonstrate the origination of gland development.

The separation of apical caps into the lumen of secretory units demonstrates that these glands are apocrine (Figure 2-8A). Unlike the anal glands of all previously examined animals such as the dog (Montagna and Parks 1948), cat (Greer and

Calhoun 1966), and 20 various species examined by McColl (1967), the manatee has only apocrine and no sebaceous glands. This indicates that the glands are used primarily for signal transmission (Eglitis and Eglitis 1961). In addition, there is no

reference to anal glands for which the presumed purpose is not the expression of chemical signals. Assuming that these structures produce one or more signals, the manatee should possess the ability to perceive these signals, whether by olfaction or taste. The manatee does possess taste buds (Levin and Pfeiffer 2002) and olfactory epithelium (Mackay-Sim et al. 1985); however, the extent to which these systems are used is unknown. Behavioral and comprehensive anatomic examination of these systems is needed to determine how the manatee perceives chemical signals.

We hypothesize that the manatee uses its anal glands for chemical communication. As the manatee is not a territorial animal, marking territory is not likely a use for the secretions. Also, the manatee is an herbivore that has no natural predators and therefore would not use the secretion as defense or to subdue prey. Finally, it has long been hypothesized that these aquatic mammals use chemical signaling for communication and specifically in the broadcast of reproductive signals (Rathbun and

O‟Shea 1984). Anecdotal evidence of this communication includes documentation of manatees in Crystal River, FL rubbing genitals/anus, eyes, arm pits, and chin on the same submerged logs and stones each year. It has been proposed that females, which rubbed more frequently than males, were leaving a chemical signal of their reproductive state (Rathbun and O‟Shea 1984). Additional wild manatee observations include a male manatee “exploring” with his muzzle a female manatee‟s genital area (Moore 1956).

In the captive setting, a German veterinary practitioner responsible for the care of a male and female manatee inhabiting the same tank, noted that when the female was in estrus the male would swim at the bottom of the tank and always find the locations along the bottom where the female‟s genitals had contacted (Vosseler 1924). The

rubbing of the genital/anal area, which in females are centimeters apart, suggests that a signal is contained in the secretions/excretions of these areas that can be perceived by males. The presence of mucous in anal gland and vaginal secretions (Marmontel 1988) may increase retention time for these signals on rocks, logs or other surfaces.

Although staining can give an indication of whether a secretion contains mucous, lipids, proteins, etc… the stain does not identify the specific chemicals within that secretion. To determine the purpose of the signals produced and released by these glands, research is needed to determine the exact chemical composition of the gland secretion. Since collection through the ducts from a live or freshly dead animal is not practical due to the internal design of the duct system a novel chemical analysis method known as matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) will be used to analyze glands from animals of various sexes and ages to determine which chemicals are most prevalent in the most productive glands. Through a better understanding of what chemicals are released by the manatee‟s anal gland there will be a better indication of what type of signal is used to express reproductive status, health, or other parameters to conspecifics.

CHAPTER 3 TASTE BUDS IN THE ORAL CAVITY OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS, WITH ESTROGEN RECEPTOR Β WITHIN THE LINGUAL ROOT TASTE BUDS

Background

The sense of taste is mediated by the chemosensory organ‟s taste buds, which are located predominantly in the tongue but can be found throughout the oral cavity and epiglottis. Within mammals, taste buds are generally located in papillae. There are four types of papillae found on the tongue: fungiform, filiform, foliate, and circumvallate or vallate. Except for filiform papillae, which aid in mastication (Iwasaki 2002), taste buds can be found within all or none of these papillae depending on species (Hosley and

Oakley 1987). In general fungiform papillae are mushroom shaped and found towards the apex and sides of the rostral portion of the tongue. Filiform papillae are long „v‟ shaped cones lacking taste buds and located throughout the dorsal surface of the tongue. Foliate papillae are parallel grooves containing large concentrations of taste buds and generally located at the lateral base of the tongue. Circumvallate (vallate) papillae are dome shaped and if present are found on the dorsal surface of the tongue forming a row (Iwasaki 2002).

Depending on the location of the taste buds within the oral cavity, they synapse on one of three cranial nerves: the facial (CN VII), glossopharyngeal (CN IX) or vagus

(CN X) (Finger et al. 2000). In terrestrial animals, gustation is a combination of senses including taste, smell, touch, temperature and chemesthesis or spice. In aquatic species, the contribution of volatile odorants to the sense of taste is minimized or absent and in fish the lateral olfactory system contributes aqueous chemicals to food detection

(Kotrschal 2000). However, in marine mammals, which have no or reduced olfactory

bulbs (Lowell and Flanigan 1980), the sense of taste is likely to have an increased importance to food localization, reproduction and communication.

The exact transduction mechanisms of taste are not well understood. There are four cell types that mediate taste within the taste bud. Type I cells are glial like and support other cells. Cell type II are receptors for sweet, bitter and umami transductions.

Type III cells are an intermediary between receptors and nerve fibers, responding broadly to salty and sour ion gradients. The function of type IV cells is not known but they are located near the base of the taste bud and are morphologically Merkel-like

(Finger 2005, Suzuki 2007). The main neurotransmitter from taste cells to the gustatory nerves is ATP (Finger et al. 2005); however, other regulators are thought to play a role in taste transduction including estrogen (Toyoshima et al. 2007). Within the oral cavity estrogen receptor β also is a potential regulator of breathing (Behan and Thomas 2005), mucous and salivary gland function (Valimaa et al. 2004) or reproduction (Cheek et al.

1998).

Within terrestrial mammals, taste buds have been documented consistently and quantitative measurements taken in a few species. The rhesus monkey, Rhesus macaque, has an estimated 8,000-10,000 taste buds within all papillae of the tongue and half of these are within the foliate papillae (Bradley et al. 1985). In the Syrian golden hamster, Mesocricitus auratus, there are 230 taste buds within the foliate papillae, which represent about 30% of the total taste buds within the hamster‟s head.

Rats, Rattus norvegicus, have an average of 1265 taste buds with approximately 455 of those in foliate papillae (Miller and Smith 1984). Within the porcine tongue fungiform papillae have significantly fewer taste buds than vallate papillae with an average 3.9

taste buds per dorsal fungiform papillae and 22.1 taste buds per lateral papillae compared to the 733 taste buds per vallate papillae (Mack et al. 1997).

In an aquatic environment, taste buds are well documented in fish and especially in specialized structures near the mouth of the fish such as barbels (Kotrschal 2000).

However, information on taste bud location and density in aquatic mammals is limited.

Within semi-aquatic species such as those in the suborder Pinnipedia there are fewer taste buds than their terrestrial counterparts and they appear to have sensitivity to sour, bitter, and salty but not sweet (Friedl et al. 1990). Umami, the property of a savory or protein rich food, was not tested but pinnipeds are thought to be sensitive because of their high protein diet. Pinnipeds are able to detect salt water gradients within a 4% variance at levels of 30 parts per thousand (Sticken and Dehnhardt 2000). According to

Yoshimura et al. (2002) and (2007), the California sea lion, Zalophus californianus californianus, and spotted seal, Phoca largha, have taste buds on vallate and fungiform papillae. The walrus, Odobenus rosmarus, has taste buds few in number (140 or less) but large in size (75 X125 µm) on the tongue (Kastelein et al. 1997). This would place the walrus‟ sense of taste intermediate between the harbor seal with the greatest taste and the California sea lion and grey seal, Halichoerus grypus, with the least taste

(Sonntag 1923). There are no other reports on the number or distribution of taste buds within Pinnipedia.

Bottlenose dolphins, Tursiops truncatus, have taste buds in pits at the base of their tongue that develop from circumvallate papillae present in neonates (Kuznetzov

1990). They also have a highly developed gustatory nucleus of the thalamus (Kruger

1959). There are conflicting reports on how well bottlenose dolphins can respond to the

four common tastants; salty, sour, sweet and bitter. One study demonstrated dolphins respond at an order of magnitude below human capabilities (Kuznetzov 1990), while another measured a threshold of 0.0173M for a sour tastant and 1.36X10-5 for bitter, near human thresholds, in a 3-5 year old male dolphin (Nachtigal and Hall 1984).

Umami has not been tested.

Taste buds have been located in a newborn male Stejneger‟s beaked whale,

Mesoplodon stejnegeri (Shindo et al. 2008) and striped dolphins, Stenella Coeruleoalba, of varying sex and age (Komatsu and Yamasaki 1980). In a Baiji river dolphin calf,

Lipotes vexillifer, taste buds were located with fewer found in a young adult female and none in an older adult female (Li 1983). These examinations involved small sample sizes and limited comparisons between age and sex. Therefore, the true extent of individual species taste ability and use is difficult to determine currently, but in general odontocetes have the ability to taste and may be using this in food assessment or mate recognition. There are no formal published reports on the use of taste or presence of taste buds in mysticetes.

Within the order Sirenia, the dugong, Dugong dugon, has taste buds in pits of the lateral margins of the tongue (Yamasaki et al. 1980), and the Florida manatee,

Trichechus manatus latirostris, has taste buds in foliate papillae located on the side of the tongue root (Levin and Pfeiffer 2002 ). In both species it is thought that taste is integral for freshwater detection and food palatability, but it is not known if their chemoreceptive use translates to communication of reproductive status (Yamasaki et al.

1980; Reep and Bonde 2006). In general, sirenians‟ taste buds are more numerous than cetaceans or pinnipeds (Yamasaki et al. 1980).

The Florida manatee is an endangered species for which increased understanding of their physiology, reproduction and nutrition will help direct management decisions. In manatees, touch is probably the predominant sense for food detection but taste is likely supplementary (Reep and Bonde 2006). Their herbivorous diet increases their vulnerability to toxins, which can be detected by a strong bitter taste

(Nei 2008) and would indicate a need for good taste detection. In addition, it has long been thought that males use chemoreception to find female manatees in estrus and so it is not known if they use their sense of taste or smell or a combination to do this (Levin and Pfeiffer 2002, Reep and Bonde 2006).

This study sought to characterize the manatee taste buds and compare the cellular components, number and size to other well studied species. In addition, the presence of estrogen receptor β was assessed and compared within lingual taste buds and epidermis.

Materials and Methods

Samples for gross and histological analysis (Table 3-1) were collected from manatees that died within 24 hours of sample collection (i.e., freshly dead). They died from boat strikes, cold stress or natural causes and were brought to the state of

Florida‟s Marine Mammal Pathobiology Laboratory in St. Petersburg, FL. Entire head samples were obtained for collection of tongue roots, portions of the soft palate and from one animal the valve at the base of the nasal passage was available for collection.

All samples but one were fixed in 10% Neutral Buffered Formalin (NBF). From one animal three sample cubes, one cm by one cm by two mm, were blocked and fixed in 2% glutaraldehyde - 0.1 M phosphate buffered solution for transmission electron microscopy (TEM), the rest of the tissue was fixed in 10% NBF.

Table 3-1. Taste bud samples collected for gross and histological analysis and fixed in 10% NBF. TL-Total Length and is measured in cm. COD-cause of death. *From LPZ102639 three samples were collected for fixation in 2% glutaraldehyde. Animal Number Sex/TL/COD Sample Collected MEC0853 M/119/Perinatal Tongue Root MSE0943 M/205/Natural Tongue Root MNW0905 M/210/Natural Tongue Root Soft Palate Nasal Valve MSE0909 M/262/Cold Stress Tongue Root Soft Palate SWFTm0820 M/287/Chronic Watercraft Tongue Root Soft Palate LPZ102639 M/309/Chronic Tongue Root* Soft Palate MSE0910 F/188/Cold Stress Tongue Root Soft Palate MNW0901 F/212/Cold Stress Tongue Root Soft Palate SWFTm0826 F/261/Chronic Watercraft Tongue Root Soft Palate

Gross and Microanatomy

The tongue root and oral cavity were examined grossly using a dissecting microscope (Figure 3-1). Areas of the oral cavity with papillae structures or pits were removed to assess the presence of taste buds. Tongue roots were removed for taste bud counts and measurements. For those tongue roots that were complete, papillae counts were taken over the entire structure. Sections of the tongue root containing foliate papillae, soft palate, and nasal valve were taken for histological staining. The samples were embedded in paraffin and sectioned at 5µm. The 5µm sections were stained with hematoxylin and eosin (H&E) to outline the general anatomy of the tissue.

Dorsal

B C

N.P.

T Rostral

Figure 3-1. Midsagittal view of manatee head; T-tongue, N.P.-nasal passage, B.C.-brain cavity. Note the location of tongue.

Taste Bud Size and Quantification

Tongue roots were removed and if complete, foliate papillae counted (Figure 3-

2). The five micrometer H&E sections were examined and taste bud measurements taken for any taste bud for which the pore was visible. A total of ten manatees had between one and four taste buds with visible pores. This procedure ensured that the measurement was taken at the center of the taste bud and that an individual taste buds was not measured more than once. Two measurements were taken; vertically, from the taste pore to the basement membrane and horizontally from either end of the taste bud wall at the widest point (Figure 3-3).

The number of taste buds within papillae was calculated. For these counts taste buds on either side of the papillae were counted at a magnification of 250X to differentiate taste buds (Figure 3-4). Some animals had multiple papillae that were counted and multiple sections within the same papillae. Distances between the sections

were between 10 and 100µm and so there was potential for the same taste bud to be counted more than once. The average number of taste buds within papillae was calculated from twenty-one manatees representing male and females of various ages and causes of death. The length of the foliate papillae was measured using a micro- caliper on the tongue roots of LPZ102766 and LPZ102939.

Dorsal

Rostral

Figure 3-2. Tongue root location and foliate papillae on side of root. Scale bars represent approximately 1 cm.

Figure 3-3. Taste bud measurements. Made using Leika photographic software at 400X. Both vertical and horizontal measurements were taken.

Figure 3-4. Typical foliate papillae with taste buds 20X and magnification of taste bud at 250X. Note ability to differentiate the four separate taste buds at 250X.

Immunohistochemistry

A total of twenty-three samples were used for estrogen receptor β staining and were obtained from archived tongue root tissue blocks that had been collected from all freshly dead manatees since 2002 (Table 3-2). Additionally, those samples collected for examination of taste bud location were used for estrogen receptor β staining.

Specimens included males and females that died at or near the time of birth, perinatal, as well adult male and female manatees that had various causes of death.

Sections were mounted on superfrost plus slides and incubated at a dilution of

1:100 overnight with Millipore Anti-Estrogen Receptor β with a Vector Vectastain

Universal quick kit followed by an Immpact AEC peroxidase substrate kit. Due to the

large number of specimens, twenty-three, the slides were divided into four groups and run separately. Within each group there was a negative control without antibody and a consecutive section from MSW03143 run to control for consistent staining between groups. Staining intensity was assessed by two independent investigators using a blinded analysis on a scale of 0-3 with 0 being no reaction and 3 representing intense reaction. Both taste buds and epidermis were scaled. The observer‟s scales were averaged and those averages compared between groups using a student‟s t-test.

Table 3-2. Samples used for estrogen receptor β analysis. TL-Total Length and is measured in cm, COD-cause of death. Animal Number Sex/TL/COD MSW0830 F/88/Perinatal MSE0630 F/121/Perinatal MEC0470 F/124/Perinatal MSW0601 F/170/Chronic Watercraft MSE0910 F/188/Cold Stress SWFTm0806b F/192/Cold Stress MSW0316 F/198/Cold Stress MEC0616 F/199/Cold Stress MEC0633 F/238/Acute Watercraft LPZ102120 F/259/Chronic Watercraft MSE0504 F/261/Acute Watercraft MSW03143 F/263/Red Tide LPZ102331 F/292/Chronic Watercraft MEC0831 M/95/Perinatal MEC0629 M/97/Perinatal MNE0633 M/116/Perinatal MEC0469 M/120/Perinatal LPZ102331 M/201/Chronic Watercraft SWFTm0802b M/213/Cold Stress MSE0632 M/255/Chronic Watercraft MEC0608 M/276/Unknown LPZ102639 M/309/Chronic Watercraft LPZ102203 M/XXX/Chronic Watercraft

Transmission Electron Microscopy

Samples fixed in glutaraldehyde were cut into small portions (2 mm3) and post fixed in 2 % osmium tetroxide for 2 hours at room temperature. They were dehydrated in grades of alcohol and embedded in an epoxy (Epon-Araldite) plastic mixture. Ultrathin sections between 70-90 nanometers were sectioned using an ultramicrotome and mounted on a 100 mesh Formvar coated grid. The sections were stained with uranyl acetate and Reynold‟s lead citrate and examined with a transmission electron microscope (Hitachi H 7000).

Results

This study confirmed that in the manatee tongue, taste buds are located in the foliate papillae. In addition, we found taste buds in one of the seven manatees examined within the soft palate, directly above those located in the tongue (Figure 3-5) and observed what appeared to be taste buds in the valve at the junction of the nasopharynx and oropharynx (Figure 3-6). The average number of taste buds observed in a single cross-section of the foliate papillae was 12±6.4. Within the right tongue root there was an average of 27±4 foliate papillae and in the left side 24±4.9 (Table 3-3).

The average size of the manatee taste buds was measured at 60µmX49µm (Table 3-4).

The average papilla length was 1.13±0.65mm (Table 3-5). Taking into account the length of the foliate papillae, the width of the taste buds and the number of taste buds observed within a cross-section of the papillae, the number of taste buds ranged from

77 taste buds/papilla to 520 taste buds/papillae. Using these estimates and the number of papillae within the tongue root, the total estimated number of taste buds within a manatee tongue root was between 3,233 and 31,140.

Figure 3-5. Taste buds of the soft palate of LPZ102639. Scale bar equals 100µm.

Figure 3-6. Apparent taste buds, indicated by arrows, of the nasal valve of MNW0905. Scale bar equals 100µm.

Table 3-3. Taste bud and papillae number within manatee tongue roots. TL-Total Length and is measured in cm, COD-cause of death, NM-Not Measured. Animal Taste Number Number Number Animal sex/TL/COD Buds/Papillae Papillae Right Papillae Left MSW0830 F/88/Perinatal 13 NM NM MEC0831 M/95/perinatal 10 NM NM 24 MEC0629 M/97/Perinatal 10 NM NM 11 MEC0853 M/119/Perinatal 19 NM NM MEC0469 M/120/Perinatal 15 NM NM MSE0630 F/127/perinatal 26 NM NM MSE0910 F/188/Cold Stress 11 NM 22 17 19 18 19 14 19 16 15 9 MSE0909 M/175/cold stress 9 28 26 9 8 7 10 11 SWFTm0806b F/192/Cold Stress 14 NM NM 11 MSW03176 F/198/Cold Stress 25 NM NM 11 LPZ102106 M/201/Chronic 9 NM NM Watercraft MSE0943 M/205/Cold Stress 7 21 17 5 MNW0905 M/210/Natural 6 NM NM 9 MNW0901 F/212/Cold Stress NM NM 18 SWFTm0802b M/213/Cold Stress 18 NM NM MEC0633 F/238/Acute 3 NM NM 13 MSW0696 M/251/natural 17 NM NM MSE0632 M/255/Chronic 7 NM NM Watercraft 12

Table 3-3. Continued Taste Number Number Animal Number Animal sex/TL/COD Buds/Papillae Papillae Right Papillae Left SWFTm0826 F/261/Cold Stress 6 NM NM 7 5 4 6 4 9 3 6 MSW03143 F/263/red tide 34 NM NM LPZ102765 F/267/Chronic NM 29 24 Watercraft LPZ102941 F/272/Chronic NM 32 29 Watercraft MEC0680 M/276/Unknown 9 NM NM MSW0332 M/307/Red Tide 12 NM NM LPZ102939 NM 27 29

Average ± S.D. 12±6.4 27±4 24±4.9

Table 3-4. Taste bud size within manatee tongue roots. TL-Total Length and is measured in cm, COD-cause of death. Animal Number Animal sex/TL/COD Taste bud vertical size Taste Bud Horizontal (µm) Size (µm) SWFTm0802b M/213/Cold Stress 72 71 MEC0633 F/238/Acute 87 67 MEC0629 M/97/Perinatal 45 31 MSW03176 F/198/Cold Stress 73 51 MEC0831 M/95/Perinatal 64 38 41 33 MSE0630 F/127/Perinatal 63 39 37 40 38 26 SWFTm0826 F/261/Cold Stress 84 58 93 55 73 30 55 55 MSW0696 M/251/Natural 61 73 70 60 MEC0853 M/119/Perinatal 57 53 47 58 58 76 44 47 MSW0332 M/307/Red Tide 44 40 66 44 42 40 Average ± S.D. 60±16 49±14

Table 3-5. Measured length of foliate papillae from the tongue roots of two manatees. The overall average length was 1.13±0.65mm. Foliate Papillae LPZ102939 LPZ102765 Left side Right side Left Side Right Side 1 1 0.5 0.5 1 2 0.5 1 0.5 2 3 0.5 1 0.5 0.5 4 0.5 1 1 1.5 5 1 0.5 1 1 6 1.5 0.5 1.5 1.5 7 1 0.5 2 1 8 1 1 1.5 1 9 0.5 0.5 0.5 1 10 0.5 0.5 1 0.5 11 0.5 1 1 2 12 0.5 2 1 0.5 13 0.5 2 0.5 1 14 0.5 1 2.5 1 15 0.5 1.5 2.5 1.5 16 0.5 2 3 1.5 17 0.5 2 1 1.5 18 1 1 4 2 19 1 1.5 1 2 20 0.5 1 1 1.5 21 0.5 0.5 1 0.5 22 1 2 1 0.5 23 1 1.5 2.5 0.5 24 1 1 1 1 25 1 2 NA 1 26 1 3 NA 1.5 17 1 0.5 NA 1 28 1.5 NA 1 29 1 NA NA 2 Average 0.79 1.20 1.38 1.19

Estrogen Receptor β

Immunohistochemical staining for estrogen receptor β demonstrated the presence of these receptors within the epidermis and taste buds of the manatee (Figure 3-5).

There was a significantly greater amount of reaction in epidermis of the tongue root

compared to the taste buds themselves (p = 0.0007, student‟s t-test). Therefore, the comparisons have been split into epidermis and taste bud groups (Tables 3-6 and 3-7).

B A

C D

Figure 3-7. Estrogen receptor β staining within the lingual root of an adult manatees. A) taste buds and B) epidermis of an adult male manatee, MEC0608 and C) taste buds and D) epidermis of an adult female, MSW03143.

Table 3-6. Amount of estrogen receptor β reaction, average of two observers‟ ratings, within the epidermis of the Florida manatee tongue root. TL-Total Length and is measured in cm, COD-cause of death. Animal Number Sex/TL/COD Nuclear Staining Y/N Amount of Stain (0-3) MSW0830 F/88/Perinatal N 1.5 MSE0630 F/121/Perinatal Y 2.5 MEC0470 F/124/Perinatal N 1.5 MSW0601 F/170/Chronic Watercraft N 2 MSE0910 F/188/Cold Stress Y 2 SWFTm0806b F/192/Cold Stress Y 2 MSW0316 F/198/Cold Stress N 1.5 MEC0616 F/199/Cold Stress Y 3 MEC0633 F/238/Acute Watercraft Y 3 LPZ102120 F/259/Chronic Watercraft N 1 MSE0504 F/261/Acute Watercraft Y 2 MSW03143 F/263/Red Tide Y 3 LPZ102331 F/292/Chronic Watercraft Y 2.5 MEC0629 M/97/Perinatal N 3 MNE0633 M/116/Perinatal Y 1.5 MEC0469 M/120/Perinatal Y 2 LPZ102331 M/201/Chronic Watercraft N 1.5 SWFTm0802b M/213/Cold Stress Y 1 MSE0632 M/255/Chronic Watercraft N 1 MEC0608 M/276/Unknown N 1.5 LPZ102639 M/309/Chronic Watercraft Y 3 MEC0831 M/95/Perinatal N 2 LPZ102203 M/XXX/Chronic Watercraft Y 1.5

All manatees had estrogen receptor β reaction localization in the cytoplasm, but only 17 out of 23 manatees examined had some sort of nuclear staining; nine of which had staining in the taste buds. There was no variation in age, sex or COD as to whether there was or wasn't nuclear staining. Five of the manatees had nuclear staining in both epidermis and taste buds, eight had staining in epidermis only, and four had staining in taste buds only. Perinatal taste buds were significantly more stained in males than females (p = 0.015, student‟s test) but almost exactly the same in adults (p = 1,

student‟s t-test). The epidermis of adult females were more strongly stained (2.2) than males (1.8) but not significantly (p = 0.12, student‟s t-test).

Table 3-7. Amount of estrogen receptor beta reaction, average of two observers‟ ratings, within the taste buds of the Florida manatee. TL-Total Length and is measured in cm, COD-cause of death. Animal Number Sex/TL/COD Nuclear Staining Y/N Amount of Stain (0-3) MSE0630 F/121/Perinatal N 0.5 MSW0601 F/170/Chronic Watercraft N 1 MSE0910 F/188/Cold Stress N 0.5 SWFTm0806b F/192/Cold Stress N 0.5 MSW0316 F/198/Cold Stress Y 1.5 MEC0616 F/199/Cold Stress Y 1.5 MEC0633 F/238/Acute Watercraft N 1.5 LPZ102120 F/259/Chronic Watercraft N 0.5 MSE0504 F/261/Acute Watercraft N 2 MSW03143 F/263/Red Tide Y 2 LPZ102331 F/292/Chronic Watercraft Y 1.5 MSW0830 F/88/Perinatal N 0.5 MNE0633 M/116/Perinatal Y 1.5 MEC0469 M/120/Perinatal N 1.5 LPZ102331 M/201/Chronic Watercraft Y 1.5 SWFTm0802b M/213/Cold Stress Y 0.5 MSE0632 M/255/Chronic Watercraft N 0.5 MEC0608 M/276/Unknown Y 1.5 LPZ102639 M/309/Chronic Watercraft N 2 MEC0831 M/95/Perinatal Y 1.5 MEC0629 M/97/Perinatal N 2.5 LPZ102203 M/XXX/Chronic Watercraft N 1.5

Ultrastructure

Examination by transmission electron microscopy demonstrated the location of the taste pore and basement membrane as well as the four types of taste cells: I, II, III, and

IV. Type I contain a darker nucleus and cytoplasm while type II has the lighter nucleus and cytoplasm. Type III can be identified by their dark granules and type IV were found near the basement membrane (Figure 3-8).

I III

I II II II TP IV BM

A Figure 3-8. Electron micrographs of manatee taste buds. A) The beginning of the taste pore of the bud, Taste cell I, II and taste pore (TP), scale bar equals 10µm. B) the base of a taste bud including basement membrane (BM) and each of the four taste cells; I, II, III, IV, scale bar=20 µm.

Discussion

The estimate of between 77 and 520 taste buds/papillae falls under what has been observed in other species circumvallate papillae, which are similar in structure and location to foliate papillae (Davies et al. 1979, Mack et al. 1997). On the tongue of the bovine, Bos taurus, the number of taste buds per papillae had extreme variation. Some papillae had fewer than 200 taste buds/papilla and one papilla had 1,786 taste buds with an average of approximately 800 taste buds/papilla. Porcine tongues had between

583-875 taste buds/circumvallate papilla (Mack 1997).

The large variation of taste bud estimate between 3,233 and 31,140 is due to the small sample size and limited examination of any single animal. Future experiments should focus on a more standardized examination of each tongue root and including histological sectioning throughout the length of the papillae, such as in the examination of the porcine tongue (Mack et al. 1997). The actual number of taste buds within the

manatee tongue is likely between the above estimates which would put it at a similar estimate as the bovine tongue which has between 14,765 and 21,691 taste buds

(Davies et al. 1979). Another animal with a similar range is the rhesus monkey with estimated 8,000-10,000 taste buds (Bradley et al. 1985). The manatee has far more taste buds than the Syrian golden hamster, Mesocricitus auratus, at 723 taste buds within the hamster‟s head or the rat which have an average of 1,265 taste buds (Miller and Smith 1984). Although exact counts are not given, the manatee appears to have more taste buds per papilla than the hippopotamus, Hippopotamus amphibius amphibius (Yoshimura et al. 2009) and hyrax, Procavia capensis (Yoshimura et al.

2008).

The concentration of taste buds within the foliate papillae of the tongue root may be correlated to food mastication and the transportation of the resultant mixture of water and food particles. There are narrow grooves on either side of the manatee‟s tongue that lead from the palate to the root of the tongue where taste buds are located (Figure

3-9). This anatomical design would allow for the mixture to be tasted before the bolus was transported to the back of the mouth. It would also allow for quicker expulsion of non-desired food material.

The position of taste buds within the foliate papillae may correlate with the herbivorous diet of the manatee. In humans, there is a strong response to the highly bitter compound quinine hydrochloride on the vallate papillae, which are comparable to foliate, at the base of the tongue as compared to the fungiform papillae on the front of the tongue (Collings 1974). It is likely that similar taste receptors are located in the foliate papillae at the base of the tongue of hyrax and manatee as these are herbivorous

species. Through behavioral studies examining manatees sense of bitter taste as well as assessment of bitter receptors, T2R‟s, within the manatee genome and immunohistochemical staining for these receptors within the tissue, confirmation of this hypothesis would be made.

Dorsal

Dorsal NP R o st BC ra NP l C a R T u o T d st al ra l TR

Figure 3-9. Mid-sagittal section of a manatee head with a transverse section demonstrating the groove between tongue and teeth (arrow). T-tongue, NP- nasal passage, BC-brain cavity, TR-tongue root.

The isolation of taste buds in manatees‟ foliate papillae may be explained by their habitat and diet. Differences in taste bud number and location between two similar families of fish, the Blennidae and Gobiidae, indicate that differing foraging strategies and diet can affect chemosensory structure distribution (Fishelson and Delarea 2004).

Both fish families have similar body types and live in the same ecological niche of coral reefs but have differing diets. An examination of the taste buds on the lips and mouth of these fish demonstrated differences. In gobies the number of taste buds increased with

age and therefore growth rate, reaching a total of 7500 taste buds. These were situated mostly on the lips and towards the front of the oral cavity. Blennies were observed to have many fewer taste buds, no more than 1000, with most of them located on the posterior part of the mouth. The amount and distribution of these taste buds correlated with the respective fish‟s diet. Blennies were determined to be much less selective than gobies and gobies fed predominantly by suctioning in their prey from bottom sediment and so depend on taste instead of visual cues (Fishelson and Delarea 2004). These results indicate that among fish and perhaps all species there is a correlation between diet type, feeding strategy and the use of taste.

Quantification of soft palate taste buds of the manatee is impractical due to their scarcity and difficulty to locate. Taste buds located on areas other than the tongue have had limited study and there is much debate on what their purpose. A current hypothesis that has received the most attention has been that epiglottal taste buds, which would correlate with those found on the manatees valve, aid in the breathing response to ensure that animals do not aspirate liquid or food particles (Bradley et al. 1980). This hypothesis is supported by behavioral data in which newborn piglets and lambs were subjected to water and milk placement on their laryngeal surface. Their breathing was monitored and it was found that they went through an apneic period that persisted to death if allowed (Downing and Lee 1975). It is also hypothesized that soft palate taste buds aid in nipple attachment in young animals (El-Sharaby et al. 2001) and provide supplementary taste information in all animals (Miller and Spangler 1982). Therefore, soft palate taste buds may be more prevalent in younger manatees.

The cellular structure of the taste buds in manatees is similar to other species.

Each cell type is represented. The size of manatee taste buds is comparable to other mammals (Sonntag 1923). They are larger than both young and adult rats, 46µm,

(Mistretta and Baum 1984) but smaller than the walrus 75 X125 µm (Kastelein et al.

1997). The size of the taste bud however, does not necessarily indicate the amount of taste transduction. The number and distribution of cells within individual taste buds are more indicative of taste ability. When rat and mouse taste buds were compared it was found that the rat taste bud is larger in volume than the mouse but the mouse has a significantly larger density of taste cells within the taste bud (Ma et al. 2007). Therefore, the sense of taste is not solely dependent on how many taste buds there are but how many cells are present to detect chemicals. A similar calculation of manatee taste bud density and taste cell number would allow for a comparison with the traditional laboratory species of rat and mouse.

The estrogen receptor β staining within the tongue root and taste buds of the manatee in all ages and sexes indicates that it has a universal purpose and has an important function within the oral cavity. The absence of nuclear staining in some animals may have been a consequence of inadequate fixation. The samples were collected over a series of years by various people and although a protocol was in place for limited fixation followed by immediate embedding, this may not have always been the case. The significant difference in perinatal taste bud positive reaction between males and females requires more samples to verify. The epidermis of adult females reacted more positively (2.2) than that of adult males (1.8) but not significantly (p= 0.12).

It has been documented that estrogen is an important hormone in the production of mucous and saliva, although the exact pathways are not known (Valimaa et al. 2004).

Not only is the production of mucous and saliva important to the transduction of taste but estrogen itself may be a component of taste transduction (Toyoshima et al. 2007).

Although in an aquatic environment the manatee would not be as reliant on saliva to transport chemicals, the mucous would be important to maintain chemicals within the region of taste buds for an adequate amount of time to be sensed without being cleared away by external water. This likely also explains the location of taste buds at the interior of the buccal cavity. Any taste buds closer to the front of the mouth would be more susceptible to clearing by water.

Another potential function for estrogen receptors within the mouth of the manatee is for the reproductive communication. Within the goldfish it is well documented that males respond strongly to varying levels of 17β-estradiol (Bjerselius et al. 2001). This is one of the most common estrogens and is the main ligand for estrogen receptor β.

Manatee hormones vary with reproductive cycling similarly to other species and they may have evolved the ability to detect these variations in chemicals similar to the goldfish (Sorenson and Stacey 1999).

Regardless of the function(s) for estrogen receptors in the mouth of the manatee there is concern about chemicals that may mimic estrogen in the environment. The impact of these endocrine disrupting chemicals could include the blocking of estrogen‟s effects or causing an unnecessary increase in their function. Estrogen receptors are of special concern in reference to mimicking chemicals and although all animals are sensitive to estrogens, those that reside in an aquatic setting are especially vulnerable.

Although not demonstrated in aquatic mammals, there are several examples of aquatic vertebrates and invertebrates that are affected physiologically and behaviorally by estrogen mimicking compounds (Cheek et al. 1998). Biodegradation products of detergents have direct interaction with fish estrogen receptors (White et al. 1994) and when exposed, male fish have developed significantly higher levels of vitellogenin, an egg yolk protein (Purdom et al. 1994). Also, polychlorinated biphenyls (PCBs) have been demonstrated to have feminizing effects on developing male red eared sliders in the same way as exposure to estradiol (Bergeron et al. 1994). Additional chemicals that have been implicated in disruption of estrogen receptors include phytoestrogens, bisphenolics (from plastic degradation) and organochlorine pesticides (Shanle and Xu

2011).

The distribution and amount of taste buds within the oral cavity of the manatee indicate that the sense of taste is important to the manatee. The presence of estrogen receptor β is also interesting and with further study may increase the understanding of the use of estrogen as a chemical signal in mammals in general. Research is also needed to evaluate the impacts of pollutants which are common within the manatees‟ environment on the sense of taste.

CHAPTER 4 THE NASAL CAVITY AND OLFACTORY EPITHELIUM OF THE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS

Background

The nasal cavity of mammals is the site of temperature and humidity control, filtration, and chemical detection of inspired air. The epithelium: squamous, respiratory and olfactory, is responsible for these functions. Squamous epithelium is found in the rostral most nasal passage and protects the other epithelium from large obstructions and in many species contains hair as an added filter. The respiratory epithelium has nasal associated lymphatic tissue to support immune response, and mucous producing goblet cells that facilitate warming of the air, control of humidity, and removal of inhaled particulate material. The olfactory epithelium is typically located in the dorsocaudal portion of the turbinates and is composed of receptors that project their signals through the basement membrane to form cranial nerve one (Gross et al. 1982).

Mammals have a series of turbinates that originate from various bones including the ethmoid, nasal and maxillary. These turbinates serve to increase the surface area within the nasal cavity to provide access to the epithelium for temperature control, humidity control and chemical detection. The maxillary and nasal turbinates have been associated with respiratory epithelium while the ethmoturbinates are where olfactory epithelium can be found (Kavoi et al. 2010, Stobel et al. 2010, Van Valkenburgh et al.

2011).

Of the various functions of the nasal cavity, olfaction is the most variable between species with regard to the amount and location of olfactory epithelium.

Variation occurs for several reasons including habitat, diet and use of chemical communication. For example, the dog, Canis familiaris, a carnivore, has a great density

of olfactory receptors and a thicker olfactory epithelium than sheep, Ovis aries, an herbivore (Kavoi et al. 2010). Domestic dogs have a wide variation in the surface area of olfactory epithelium, 169.46 cm2 in the German Shepherd compared to 26.89 cm2 in the Pekingese. However, this is much higher than humans 10-12.5cm2 and rabbits 7.4-

10.2 cm2 (Adams 1972).

The most extreme example of variation with habitat is within the, odontocetes

(Oelschlager and Buhl 1985), fully aquatic mammals which do not have olfactory capabilities. Other fully aquatic mammals; the mysticetes (Oelschlager 1989) and sirenians (Mackay-Sim et al. 1985) do appear to maintain olfactory capabilities.

The Odontoceti (also known as toothed whales) that have been examined do not have an accessory and main olfactory bulb, olfactory tract or anterior olfactory nucleus as adults (Oelschlager and Buhl 1985, Marino 2001, Oelschlager et al. 2008a). It appears that the bulb may have been present in early evolutionary lineages of the delphinids and is lost in present day species during embryonic development

(Oelschlager and Buhl 1985). In addition to the regression of the olfactory neuroanatomy, a histological study characterizing the nasal complex of fifteen harbor porpoises, Phocoena phocoena, found no evidence of olfactory epithelium (Prahl et al.

2009). There are no published studies on the behavioral ability of odontocetes to react to olfactory cues.

In addition to the anatomical evidence, the investigation of the odontocete genome for olfactory receptor genes corroborates gradual evolutionary loss of olfaction

(McGowen et al. 2008). Several independent pseudogenization events support the theory of the loss of an olfactory sense over evolutionary time. This genomic and

anatomic evidence in addition to the lack of behavioral accounts of odontocetes using olfaction indicates that between the Miocene and present day these species lost the ability to smell. This is invariably due to the lifestyle and environment of these animals in addition to the energetic cost required to maintain an olfactory system.

Within the baleen whales, Mysticeti, Oelschlager (1992) examined embryos and early fetuses from three different species: blue, Balaenoptera musculus, fin,

Balaenoptera physalus, and humpback, Megaptera novaeangliae whales. Unlike

Odontocetes, the olfactory bulb, tract, and epithelium of the Mysticetes did not regress and was well formed and evident in a 105mm crown rump length (CRL) fin whale fetus

(Oelschlager 1992). In addition to the embryologic evidence, of adult brains from 11 bowhead whales, Balaena mysticetus, (Duffield et al. 1992) and a single humpback

(Breathnach 1955) contained olfactory peduncles, nerve fibers and small holes in the cribiform plate of the ethmoid bone, which would allow the olfactory nerve to pass into the brain. In a recent study, the bowhead whale olfactory bulb was located and measured 0.13% of the brain weight and 51% of the olfactory receptor genes were intact (Thewissen et al. 2011).

Anecdotal behavioral evidence and a highly reduced but intact olfactory system indicate that mysticetes retain some sense of airborne smell (Oelschlager 1989). Baleen whales spend a majority of their time at or near the surface to feed, unlike many of their toothed counterparts that dive deep to locate prey (Berta et al. 2006). Therefore, the use of airborne scents to locate groups of zooplankton or other prey as well as potentially locate signals given off by predators or mates would correspond with the anatomic evidence of the mysticetes. It would be interesting to see if there is a variation

in olfactory size between those mysticetes that feed near the surface, such as the bowhead, and those that feed on invertebrates from bottom sediment like the gray whale, Eschrichtius robustus.

Within the order Sirenia, the Florida manatee, Trichechus manatus latirostris, is a fully aquatic mammal that inhabits marine and freshwater areas and is an herbivore.

The manatee exhibits the behavior of sniffing in captivity. Unlike odontocetes but similar to mysticetes, the manatee possesses a small olfactory bulb (Wiseman and Reep Pers.

Comm., Mackay-Sim et al. 1985, Reep et al. 2007). Although the manatee does not have a vomeronasal organ (VNO), it does have what appears to be olfactory epithelium

(MacKay-Sim et al. 1985). No further study has been reported on manatee olfaction.

This study seeks to characterize the manatee olfactory epithelium and turbinate system and compare this to other mammals. It is expected that the manatee does possess olfactory epithelium as indicated by Mackay-Sim et al. (1985) but that there is less than observed in terrestrial mammals.

Materials and Methods

Samples were collected from manatees that that died within 24 hours of sample collection (i.e., freshly dead). They were killed by boat strikes, cold stress or natural causes and were brought to the state of Florida‟s Marine Mammal Pathobiology

Laboratory in St. Petersburg, FL. From eight animals; either the entire head or half of a mid-sagittal cut of the head were collected and fixed in 10% Neutral Buffered Formalin

(NBF) for 24-48 hours (Table 4-1). The head was then cut on a bandsaw into approximately 1.5 cm thick sections along either the transverse or horizontal plane

(Figure 4-1). The sections of nasal cavity and turbinates were then decalcified for five to

seven days in Cal-Ex decalcifying solution and cut into smaller samples approximately

two square centimeters for placement in histology cassettes.

Dorsal

Caudal Rostral

A B

Figure 4-1. Head of a male manatee, MNW0905, used for nasal examination. Cut in the A) transverse plane, right side, and B) horizontal plane, left side.

Table 4-1. Samples collected for nasal cavity and olfactory epithelium evaluation. TL- Total Length and is measured in cm, COD-cause of death. Animal Number Sex/TL/COD/Age Class Plane of Section/Side of Head MSE0910 F/188/Cold Stress/Calf Transverse/Left MNW0901 F/212/Cold Stress/Calf Transverse/Right Transverse/Left SWFTm0826 F/261/Chronic Watercraft/Adult Transverse/Right Transverse/Left MEC0853 M/119/Perinatal/Newborn Calf Transverse/Left Transverse/Right MNW0905 M/210/Natural/Calf Horizontal/Right Transverse/Left MSE0909 M/262/Cold Stress/Adult Horizontal/Right Horizontal/Left SWFTm0820 M/287/Chronic Watercraft/Adult Transverse/Left Transverse/right LPZ102639 M/309/Chronic/Adult Transverse/Left

From one animal, an adult female (LPZ102765) that was euthanized after complications from a chronic boat strike, the entire head was perfused with 4% paraformaldehyde and the brain collected for another study. The head was then cut at 2 cm‟s off-midline along the sagittal plane to expose the lateral portion of the nasal cavity

and turbinates. The head was post-fixed in 10% NBF for 18 hours and nasal and olfactory epithelium collected for histological assessment (Figure 4-2).

Dorsal

Caudal

Figure 4-2. Section of manatee head, LPZ102765, from which olfactory and nasal samples were collected.

Light Microscopy

Samples were embedded in paraffin and sectioned with a rotary microtome at

5µm. The 5µm sections were stained with hematoxylin and eosin (H&E) to outline the general anatomy of the tissue. A few samples were stained with periodic acid Schiff

(PAS) to stain glycogen rich material and identify mucous rich secretions and their secreting cells. Each stained slide was examined and compared to the location of the nasal passage from which it was collected to identify the location of olfactory epithelium.

When possible the olfactory epithelial thickness was measured from basal lamina to apical surface using the Leica Application suite v. 3.5.0 digital micrometer (Figure4-3).

Figure 4-3. Measurement of olfactory thickness.

Immunohistochemistry

From the female manatee LPZ102765, five µm sections were mounted on superfrost plus slides and incubated at a dilution of 1:1000 for one hour with Wako

Olfactory Marker Protein (OMP) antibody using a Vector Vectastain Universal quick kit followed by an ImmPACT AEC peroxidase substrate kit. The location and density of olfactory epithelium was compared to the areas within the nasal passage from which the samples were collected.

Results

Squamous epithelium lined the most rostral area of the nasal cavity with occasional vibrissae for the first few centimeters of the canals. Respiratory epithelium began midway between section 1 and 2 on Figure 4-4 and continued along the nasal

passage and both the middle and dorsal nasal turbinates. It included a majority of goblet and mucous producing glands as well as sections of large veins. The olfactory epithelium was found throughout the ethmoturbinates and bordering septum. Grossly, it is a dark yellow, brown color that can be differentiated by both color and texture from the lighter and thicker respiratory epithelium. For identification of turbinate structure the transverse sections were most elucidative (Figures 4-4 and 4-5). The horizontal sections helped to identify the type of epithelium within the turbinates.

6 5 4 3 2

Figure 4-4. Manatee head with sections dividing the general area of bandsaw cuts made on a perinatal male, MEC0853.

C A B

1 1

2 2

D 3 D

D E F

Figure 4-5. Cross sections of the nasal cavity of a perinatal male Florida manatee, MEC0853. Each transverse section correlates to a segment from Figure 4-4 A) section 1, B) section 2, C) section 3, D) section 4, E) section 5, and F) section 6. Ethmoturbinates 1-3 where olfactory epithelium can be found and the (M) Middle nasal turbinate and (D) dorsal nasal turbinate that contain respiratory epithelium. Scale bar approximates one centimeter.

Light Microscopy

Throughout section one and a portion of section two of Figure 4-4, the epithelium progressed from squamous to ciliated columnar with extensive mucous glands and goblet cells (Figure 4-6). A majority of the epithelium in the ventral portion of sections two, three, and four was heavily embedded with goblet cells and mucous glands like those in Figures 4-7 and 4-8. Within the ethmoturbinates of sections five and six olfactory epithelium was predominant (Figures 4-9, 4-10) and found in a small section of the caudal most portion of the dorsal nasal turbinate (Figure 4-11). The average thickness of the olfactory epithelium was 54.79 µm (Table 4-2). Sample size was too low to determine any statistical variation with animal sex or age.

Table 4-2. The olfactory thickness of various manatees from different areas of the olfactory system as represented by different block numbers. Animal ID- tissue block number Olfactory Thickness (µm) LPZ102765-21 77.355 LPZ102765-22 60.59 LPZ102765-23 48.84 LPZ102765-24 43.88 LPZ102639-23 44.58 MSE0909-48 61.895 MSE0909-42 58.79 SWFTm0826-37 56.655 MEC0853-30 40.565

Average 54.79

Figure 4-6. Cross section of mucous producing cells from adult female, MNW0901, H&E 20X. Taken at approximately section 3 of Figure 4-4.

100

A B

Figure 4-7. Mucous glands of adult female, MNW0901, at approximately section 2 of Figure 4-4. A) 400X, H&E and B) 250X, PAS.

NG V A B

Figure 4-8. Respiratory epithelium within the nasal cavity of the Florida manatee. A) perinatal male, MEC0853, at approximately section 3 of Figure 4-4 at 250X with staining of goblet cells (G) using PAS and B) an H&E of an adult male, SWFTm0820, at approximately section 2 of Figure 4-4 with veins (V) and Nasal Glands (NG), scale bar = 500µm.

Figure 4-9. Cross section from adult female, MNW0901, demonstrating olfactory epithelium. H&E, 100X, approximately section 6 of Figure 4-4.

101

OE OE

BG BG A B Dorsal

II R

Caudal III

CE SE eE eE G eE

C D V eE Figure 4-10. Section of manatee head along sagittal plane and corresponding epithelial lining within Ethmoturbinates II and III, and dorsal nasal turbinate (D). All stained with H&E, Bowman‟s Glands (BG), Olfactory Epithelium (OE), Squamous Epithelium (SE), Columnar Epithelium (CE), Goblet Cells (G), Vein (V); A) Olfactory epithelium at 250X, B) Olfactory epithelium at 100X, C) Transitional epithelium with scale bar 100µm, D) Respiratory epithelium at 250X.

102

Figure 4-11. Diagram representing the turbinate system and location of epithelial types within the manatee head, T=tongue. The middle nasal turbinate (M) is a small extension of bone at the cranial part of the nasal cavity while the dorsal nasal turbinate (D) extends across the entire dorsal portion of the cavity. Ethmoid turbinates (I, II, and III) as well as a small portion of the caudal nasal turbinate are where olfactory epithelium can be found.

Immunohistochemistry

Localization of OMP was successful but not consistent. Only one block of epithelium identified as olfactory with H&E reacted to the OMP antibody but both the receptors and neurons stained darkly (Figure 4-12). There was excessive background stain. The process degraded the tissue and caused breaks and tears that may have impaired the ability to see the stain within some samples.

103

A B

C D

Figure 4-12. OMP staining of olfactory epithelium from an adult female, LPZ102765. A) Olfactory epithelium and nerve fibers with scale bar 100 µm, B) and C) Olfactory receptors scale bar at 50 µm, D) nerve bundle stained with OMP and scale bar of 100 µm.

Discussion

Manatees have far fewer and larger turbinates than found in animals with a highly developed sense of smell, macrosmatic, such as cats, dogs and bears but a similar turbinate structure as observed in microosmatic species with a poor sense of smell, such as humans and rabbits (Adams 1972). Unlike other examined aquatic mammals such as pinnipeds, the manatee does not have highly developed maxilloturbinates which are used for heat and gas exchange (Van Valkenburgh et al.

2011). This is likely because manatees live in a warm water environment and do not dive deep. However, the manatee does have extremely large veins, which increase the

104

amount of blood present for heat exchange. The large venous system likely acts as a thermoregulatory system depending on the need of the animal. Also, the dense mucous producing glands, including goblet cells, are another factor that would enhance gas exchange within the nasal cavity. Other marine mammals likely have similar morphology within the venous system of the lamina propria of the respiratory epithelium.

The turbinate structure of the manatee is most similar to the hyrax, its closest evolutionary relative. Within the ethmoid region, its three turbinates are in a similar position to those of the hyrax. The hyrax does have a vomeronasal organ (VNO) (Stobel et al. 2010), which the manatee lacks (Mackay-Sim et al. 1985). Of the other close evolutionary species to the manatee, the elephant has five and the aardvark nine ethmoid turbinates that are especially convoluted compared to the manatee and hyrax.

The elephant and aardvark also have a VNO. As discussed by Stobel et al. (2010) the structure of the ethmoid region is not indicative of placement within evolutionary systematics. However, the structure does support that the manatee and hyrax are microosmatic compared to the macrosmatic elephant and aardvark (Stobel et al. 2010).

The location of olfactory epithelium within the ethmoturbinates supports what was observed by Mackay-Sim et al. (1985). The manatee olfactory epithelium does react to

OMP, although not consistently. This inconsistency may be due to the fixation process, tissue preparation or may reflect a variation in OMP expression within the manatee.

Future experiments should use optimal preservation for IHC and include multiple animals to verify the antibody binding. The antibody does bind to the olfactory neurons and therefore could be used in the brain to verify regions of olfactory transduction. The density of olfactory receptors reacting to the OMP is less than other species as is the

105

thickness, 54.71 µm, compared to dogs, 72.5 µm and sheep 56.8 µm. The greatest length of ethmoturbinate, ~3 cm is less than both sheep ~3.4 cm and dog ~4.3 cm

(Kavoi et al. 2010). Future research should calculate the area that is covered by olfactory epithelium to compare to what is known about other vertebrate species. The best methods would take advantage of digital systems that allow for accurate measurements of structures.

As expected, the manatee has less olfactory epithelium than most terrestrial mammals, including sheep, dogs, and humans. However, there are still substantial amounts of receptors, especially considering the decreased amount of exposure time to odorants. Therefore, the question remains of why manatees have maintained their ability to smell while other fully aquatic species, odontocetes, have not.

The manatee does masticate food at the surface occasionally, which would allow signals to be transported to the olfactory receptors. Although not a social species, the manatee maintains strong mother/calf bonds for the first few years of life and forms large mating herds (Hartman 1979). The mother and calf will breathe simultaneously and this likely helps form a bond. Manatees in a mating herd spend more time at the surface than during other activities, which would suggest communication through olfactory signals. Also, during cold winter months, manatees congregate within warm water refuges. They breathe in unison while thermoregulating and this would present an opportunity to smell each other. It is unlikely that the manatee transports chemicals to the olfactory epithelium underwater because of the valve that separates the nasal passage from the oropharynx. However, it is possible that retronasal smelling could occur similar to how terrestrial mammals perceive flavor.

106

To determine how well the manatee is able to detect odorants and to determine whether retronasal exposure of odorants is possible, behavioral experiments would be the most clarifying. Go/no go trials have been completed with captive manatees examining their ability to see, hear and touch (Reep and Bonde 2006). A similar study design using odorants typical of the manatee‟s environment and applied to both the nares and mouth would begin to reveal the extent of olfactory ability.

In addition to behavior experiments, examination of the manatee‟s olfactory receptor genes could indicate the range of odorants they are able to detect. Genetic data indicate that there are 400-1000 different olfactory receptors in mammals

(Parmentier et al. 1992, Ressler et al. 1994), while only 50-100 exist in fish (Ngai et al.

1993, Barth et al. 1996). There are currently two known receptor classes characterized type I and type II that evolved from common ancestral genes. The absence of class I receptors in terrestrial mammals and birds and the existence of these receptors in fish indicate that these are responsible for detection of water-soluble odorants. Type II receptors are found in terrestrial species and so are thought to be specialized for volatile odorant detection. In addition, nonfunctional pseudogenes of type II receptors are found in the striped dolphin, Stenella coeruleoalba, genome, which correlates with the loss of olfaction as dolphins entered an aquatic environment (Freitag et al. 1998).

An examination of the manatee‟s transcriptome would indicate those genes that are actively producing RNA. One would expect manatees to have type II receptors and a similar olfactory receptor gene repertoire as the bowhead whale in which 51% of the olfactory receptor genes are intact (Thewissen et al. 2011). This would imply that the manatee is able to detect a smaller number of odorants than terrestrial mammals but of

107

those odorants they detect, the concentrations needed would likely be the same or less than terrestrial mammals as indicated by the amount of epithelium present.

These data indicate that the manatee is capable of smelling. For what purpose the manatee retains a sense of smell is yet to be determined. Through additional examination of the manatee genome and behavior the extent of olfactory use can be ascertained.

108

CHAPTER 5 MALE FLORIDA MANATEE, TRICHECHUS MANATUS LATRIROSTIS, BEHAVIORAL RESPONSE TO FEMALE MANATEE URINE FROM DIFFERENT REPRODUCTIVE TIMEPOINTS

Background

Reproductive chemical signaling is used by many vertebrate species ranging from large terrestrial elephants, Loxodonta africana, to small aquatic goldfish Carassius auratus auratus. Terrestrial mammals including the elephant use highly volatile compounds as pheromones (Rasmussen 1999); whereas, in an aquatic environment a chemical signal is typically soluble and of a size small enough to be dispersed (Wyatt

2006). The use of chemical cues by fully aquatic mammals has never been verified although numerous anecdotal accounts exist (Vosseler 1924, Hartman 1979, Lowell and Flanigan 1980).

Mammals use a wide variety of chemical signals to broadcast their reproductive status to the same and opposite sex. These signals can both attract and deter potential mates depending on the age and condition of the animals involved. Chemicals are energetically cheap to produce and broadcast, which is why they are most likely ubiquitous among species for the transmission of reproductive signals (Bigiani et al.

2005). Very few mammal species have been studied in depth for their use of sexual pheromonal cues. Laboratory species including the mouse, Mus musculus, rat, Rattus norvegicus and hamster, Mesocricetus auratus, have been examined in depth while the elephant is the predominant example of the wild use of pheromonal cues by a mammal.

These species use a combination of excreted signals in urine and glandular secretions. The elephant has temporal glands that express the male‟s reproductive status to females and other males through a combination of chemicals commonly

109

known as musth. These excretions contain many chemicals including the pheromones frontalin and acetones (Schulte and Rasmussen 1999). Both of these represent the typical volatile odorants used as a pheromone by terrestrial species and in fact a specific acetone, (Z)-7-dodecen-l-yl acetate is used by both moths and elephants to express reproductive receptivity (Rasmussen et al. 1996). In non-mammalian aquatic species it has been found that two types of signals are used, either steroid derivatives as with the goldfish (Dulka et al. 1987), or peptide pheromones as with the newt,

Cynops pyrrhogaster (Kikuyama et al. 1995).

The Florida manatee, Trichechus manatus latirostris, is a close evolutionary relative to the elephant but its reproductive behavior is not well understood due to the difficult nature of studying a sparsely populated endangered species in a low visibility, aquatic environment. In addition, the manatee has limited social interactions. These include mother/calf bonds that last between two and three years (Hartman 1979) as well as mating herds in which many males follow a receptive female for up to four weeks

(Rathbun et al 1995, Larkin 2000). It is the formation of these mating herds that is of interest, as it has been observed that males will travel considerable distances to track females in estrus, similar to the elephant. Although the manatee may mate at any time of the year, peak sexual activity occurs during the warm spring and summer months when manatees have dispersed from their warm water refuge sites (Rathbun et al. 1995

Reid et al. 1995). Therefore, it is assumed that there is a signal given off by the female which alerts males to her receptiveness.

Although this signal could be visual or auditory, manatees have limited vision and are reported to be near sighted, having vision only slightly better than river dolphins

110

(Supin et al. 2001). In addition, manatee vocalization generally occurs in mother and calf communication although it has been observed in mating herds (Hartman 1979).

Their sound consists of short chirps or squeals in the range of 3-24 kHz (Nowacek et al.

2003) and would most likely be lost in the estuarine environment of the manatee over long distances. A more likely source of this signal would be chemical cues given off by the female.

There have been a number of documented behavioral observations of the manatee suggesting that females produce chemical signals detectable by males.

Among the earliest came from Dr. Vosseler, a German veterinary practitioner who was responsible for the care of a male and female manatee housed within the same pool together. He noted that when the female was in estrus the male would swim at the bottom of the tank moving his nose and lateral and medial parts of the lip so that the inner lip was exposed. In addition the male would always find the locations along the bottom of the pool where the female‟s genitals had contacted (Vosseler 1924). Moore

(1956) noted a male manatee “exploring” with his muzzle a female manatee‟s genital area. Hartman (1979) discussed the prevalence of mouthing between manatees and hypothesized about their chemoreceptive abilities.

The body of the behavioral accounts indicates that a female manatee is able to broadcast her reproductive state using secretions and/or excretions. The most likely source of this signal is urine, as this is the broadcast used most frequently by mammals

(Wyatt 2006). To date, there have been no obvious glands detected on the manatee that could provide these secretions outside of the female reproductive tract (Don

Samuelson Pers Comm, Robert Bonde Pers Comm). The gut transit time of

111

approximately one week (Larkin et al. 2007) would not be conducive to signal transmission through feces, while it could explain the long time spent in mating herds.

According to current anatomical evidence, the Florida manatee, Trichechus manatus latirostris, has taste buds in the foliate papillae of the tongue root (Levin et al.

2002) and a reduced olfactory bulb (Mackay-Sim et al. 1985, Reep et al. 2007) but no vomeronasal organ (Mackay-Sim et al. 1985). Anecdotally, the manatee appears to be able to distinguish between different salt concentrations when locating freshwater and may also use taste in their food choice along with tactile stimuli. In addition, it has been hypothesized that the male manatee, has the chemoreceptive ability to track a female in estrus (Reep and Bonde 2006)

This project was the first experimental assessment of the use of chemical signaling for reproduction by any marine mammal and specifically the manatee. This study will use male reaction to female urine to elucidate the chemosensory ability of the manatee and the potential location of reception for these signals. We expect a behavioral response that may include reproductive or chemosensory behaviors from the male manatees to female manatee urine.

Methods

Urine samples came from four captive female manatees that were trained for urine collection (Horikoshi 2004).The reproductive state of the female manatee urine samples used, whether estrus or anestrus, was determined using urinary hormone levels as measured with radioimmunoassays run by the endocrinology laboratory of the

Smithsonian National Zoological Park (Figures 5-1,5-2, and 5-3). Antibodies for estradiol, progesterone, luteinizing, and follicle stimulating hormone were used to indicate reproductive cyclicity. Hormonal analysis determines the phase of estrus for a

112

particular female at the time of collection and the behavioral tests specify which samples contain a signal of biological interest to the male.

Figure 5-1. Urinary hormone levels of Lorelei, a female Florida manatee. Luteinizing hormone (LH), Follicle stimulating hormone (FSH), Estrogen (E1S) and Progesterone (PG) were all measured with radioimmunoassays. The red boxes denote estrus peaks and the black box anestrus periods.

Figure 5-2. Urinary hormone levels of Charlotte, a female Florida manatee. Luteinizing hormone (LH), Follicle stimulating hormone (FSH), Estrogen (E1S) and Progesterone (PG) were all measured with radioimmunoassays. The red boxes denote estrus peaks and the black box anestrus periods.

113

Four series of trials were run with a total of five males assessed and three different female‟s urine from seven different dates (Table 5-1). None of the males assessed had ever been exposed to or mated with a reproductively active female. Each set of trials had a slightly different technique for applying urine to the tank and observing the males due to facility arrangement. In general, the study used a double blind, continuous sampling method (Martin and Bateson 2007), recording the start time of each individual behavior as outlined on the ethogram (Table 5-2). This chart was developed from preliminary observations and previous studies examining manatee and elephant behavior (Bagley et al. 2006, Gomes et al. 2008, Horikoshi-Beckett and

Schulte 2006, King and Heinen 2004).

Figure 5-3. Urinary hormone levels of Sara, a female Florida manatee. Luteinizing hormone (LH), Follicle stimulating hormone (FSH), Estrogen (E1S) and Progesterone (PG) were all measured with radioimmunoassays. The red boxes denote estrus peaks and the black box anestrus periods.

114

The male manatee(s) was isolated in the medical pool and observed for three minutes for respiratory rate and behavior without contact. Following baseline collection a burette was opened containing 5-10ml of one of the following: salt or deionized water

(positive controls in a fresh water or saltwater enclosure, respectively), non-estrous or estrous urine. The solution was released over one minute for the ten milliliter samples and over thirty seconds for the five milliliter samples. Using the ethogram, the observer that was blind to the sample recorded from the opening of the burette both the respiratory rate and the start time for each behavior continuously over five minutes. The behavior was videotaped for retrospective analysis and all trials at a given facility were completed within a one week time frame. If an interesting behavior was observed, recording continued for an additional 5-10 minutes.

Table 5-1. Male manatees assessed with female manatee urine. Snooty was isolated while the other males were observed simultaneously. Facility Male(s) Male Current Dates trial run Female Urine- Age/Age at initial Estrous/Anestrous captivity South Florida Snooty 63/Birth August 9th-14th, Lorelei Museum 2009 5-21-03/7-25-03 Parker August 15th-22nd, Lorelei Aquarium 2011 5-21-03/7-22-03 Sara 6-2-05/4-18-05 Mote Marine Hugh 27/Birth November 15th-19th, Lorelei Laboratory 2010 7-11-03/7-2-03 Lorelei Buffett 24/Birth 5-3-03/4-5-03 Lorelei 6-22-03/7-18-03 Disney Lou ~20/~a year July 22nd-26th, 2011 Charlotte Epcot's the 4-10-05/10-13-05 Living Seas Sara Vail ~20/~a year 4-4-05/9-18-05

115

Table 5-2. Ethogram of behavior categories performed by captive male West Indian manatees exposed to urine of female manatees. Developed from Bagley et al. 2006. Behavior categories Definition specific behaviors Active Circular movement (CM) Moves around enclosure in a circular motion Circular Pectoral (CP) Moves around enclosure in a circular motion using pectoral flippers. Interaction at Gate (GI) Nose to gate with or without animals on the other side. Lift (LI) Removes upper body from tank using pectoral flippers Bottom Swim (BS) Swimming along bottom of enclosure Continuous Roll (CR) Swimming and rolling continuously Emerge (EM) Lifts head out of water Gate Interaction (GI) Swimming at gate separating medical pool and main enclosure Chemosensory Search Opens and moves lips to allow water passage at bottom of enclosure Roll (RO) Turns body completely within water column-only chemosensory to Snooty Sniff (SN) Breaths in through nostrils rapidly Continuous Roll Corner(C Roll C) Rolls continually in corner at which sample applied Search and Coprophagia Searching behavior followed immediately by (Search/Cap) ingestion of fecal Inactive Surface Rest (SR) No movement at water surface Bottom Rest (BR) No movement at tank bottom Nose to Wall (NW) Presses nose to wall Submerged (SU) Submerges below the surface after activity Column Rest (Col R) No movement within water column Surface Rest at Gate (SR-GI) No movement at gate separating medical and main pool Socialize Socialize Both (SO-Both) Both Lou and Vail Participate in Socialization Socialize Lou (SO-Lou) Lou attempting to socialize with no reciprocation from Vail Socialize Vail (SO-Vail) Vaile attempting to socialize with no reciprocation from Lou Other Other (OT) Displays a behavior not defined by the ethogram

116

The trials occurred at the same time of day to decrease time of day effects.

There was a 25-45 minute break between each of the three assays within a given day depending on the size of the tank and constraints on animal availability. The protocol of three trials (positive control, estrous and anestrous urine) presented once a day was repeated over the course of five days. Only following complete data analysis that included verification of behaviors on video was the researcher informed about the identity of the samples added for an individual assay.

Variation from the general protocol included the first series of data collection with

Snooty in August 2009 were collected using both a positive (salt water) and negative

(tank water) control. Behavior was observed for fifteen minutes instead of three prior to adding sample to the water and observed for twenty minutes instead of five after the sample was added to the water. Only two trials of the four (negative control, positive control, anestrous and estrous urine) were run during a day with a separation of at least one hour to allow tank filtration. Each of the four trials was repeated three times over the course of six days.

Hugh and Buffett could not be isolated in the medical pool and so were called back to the pool at the beginning of both pre and post sample addition. Also during a few of the trials pipettes were used for sample addition instead of burettes: November

15th 2010 observing Hugh and Buffett, July 23rd and 24th observing Lou and Vail, and

August 16th and 17th observing Snooty.

Results

Positive responses were observed from two male manatees, Snooty (Figure 5-4) and Lou (Figure 5-5), to female manatee estrous urine and little to no reaction to anestrous urine. Reaction to estrous urine included more time spent below the surface

117

exploring the bottom of the tank in a behavior termed searching. In the case of Snooty the behavior included rolling, which was not observed in any other trials during the

August 2009 series of trials. Specifically, Snooty responded to the addition of estrous urine during one of the three trials in which it was added. Snooty spent 359 seconds of the total 900 seconds searching and rolling. This in comparison to the salt water sample reaction in which he spent 150 seconds searching and did not roll at all.

Lou responded to the addition of estrous urine during four out of the five trials in which it was added. Lou‟s data were calculated as a percent of time spent on a behavior. During the first estrous sample he spent 47% of his time searching and rolling beneath the drip. On the second day, following estrous urine addition Lou spent 34% of his time searching. On the third day he spent 24% of his time searching and on the fourth day 39% of his time searching and rolling beneath the drip. The reaction on the final day of trials was not considered a response because Lou had exhibited searching behavior accounting for 27% of his time during the previous water trial therefore; the 8% of time spent searching following estrous sample addition was not considered a genuine reaction.

Neutral responses to the estrous urine were observed during the Snooty August

2011 trials (Figure 5-6). The male manatee Vail, who is housed with Lou and was assessed simultaneously as him, also had a neutral response to the urine. He spent a majority of his time swimming in large circles and rolling or interacting with Lou. He also exhibited coprophagia, which accounts for most of the chemosensory response outlined in Figure 5-7. Neither Hugh nor Buffet exhibited a response to the urine (Figure 5-8).

Three out of the five days of trials were spent begging for food or attention.

118

Figure 5-4. Behavior of Snooty in 2009 in response to female urine from different reproductive time points. A) Summary of time spent on activities deemed „Active‟, „Rest‟, and „Chemosensory‟ B) Exact time spent on all observed behaviors. „Active‟ includes: Circular Pectoral Swimming (CP), Gate Interaction (GI), Lift (LI), Emerge (EM), Swim and Movement (MVMT). Rest includes: Stop, Submerged (SU), Surface Rest (SR), Bottom Rest (BR), and Nose to Wall (NW). Chemosensory includes: Roll (RO), Search, and Sniff (SN).

119

Figure 5-5. Behavior of Lou in 2011 in response to female urine from different reproductive time points. A) Summary of time spent on activities deemed „Active‟, „Rest‟, and „Chemosensory‟ B) Exact time spent on all observed behaviors. „Active‟ includes: Pectoral Swimming (PS), Bottom Swimming (BS), Rise, Continuous Roll (C.Roll), Roll, and Gate Interaction (GI). Rest includes: Surface Rest at Gate (SR-GI), Bottom Rest (BR), Surface Rest (SR), and Column Rest (Col. R). Chemosensory includes: Search, and Continuous Roll Corner (C.Roll C). Socialize includes: Socialization for both (SO-Both), Socialization from Lou (SO-Lou) and Socialization from Vail (SO- Vail).

120

Figure 5-6. Behavior of Snooty in 2011 in response to female urine from different reproductive time points. A) Summary of time spent on activities deemed „Active‟, „Rest‟, and „Chemosensory‟ B) Exact time spent on all observed behaviors. „Active‟ includes: Circular Pectoral Swimming-Clockwise (CP-CW), Circular Pectoral Swimming-Counter Clockwise (CP-CCW),Gate Interaction (GI), Circular Pectoral Aggressive Swimming (CP-Ag), Lift (LI), Swim and Bouncing Aggressive (B/AG), Circular Movement (CM). Rest includes: Stop, Emerge (EM), Submerged (SU), Surface Rest (SR), Bottom Rest (BR), and Nose to Wall (NW). Chemosensory includes: Roll (RO), Search.

121

Figure 5-7. Behavior of Vail in 2011 in response to female urine from different reproductive time points. A) Summary of time spent on activities deemed „Active‟, „Rest‟, and „Chemosensory‟ B) Exact time spent on all observed behaviors. „Active‟ includes: Pectoral Swimming (PS), Bottom Swimming (BS), Rise, Continuous Roll (C.Roll), Roll, and Gate Interaction (GI). Rest includes: Surface Rest at Gate (SR-GI), Bottom Rest (BR), Surface Rest (SR), and Column Rest (Col. R). Chemosensory includes: Search and Search followed by coprophagia (Search/Cap). Socialize includes: Socialization for both (SO-Both), Socialization from Lou (SO-Lou) and Socialization from Vail (SO-Vail).

122

Figure 5-8. Behavior of male manatees to female urine from different reproductive time points in 2010. A) Hugh and B) Buffett.

123

Discussion

Snooty is the oldest known manatee at 63 years old and has been in captivity his entire life. He has never mated or been in captivity with a reproductively active female: however, he is housed occasionally with female calves that are being rehabilitated before release. During the 2009 trial in which there was a heightened reaction he was housed with a female that was approaching her release date and was nearly the size of a juvenile. He was also exhibiting behaviors outside of the study period that were reproductive in nature. However, during the 2011 trials he was not exhibiting these behaviors and was not housed with a larger female calf. This may explain his response in 2009 but not 2011. Regardless, his response indicates that there is an innate chemosensory trigger for male manatees that can be found in female manatee estrous urine.

Lou has been in captivity since he was a calf and also has never mated with or been in the same enclosure as a reproductively active female. His behavior corroborates the results of Snooty and his younger age may explain the more consistent reactions to the estrous urine. Also, due to time constraints on the availability of Lou and

Vail there were only 25-30 minute breaks between trials. This may explain the continued chemosensory behavior of Lou in assays following estrous urine input.

Vail, also in captivity since he was a calf and never mated, exhibited chemosensory behavior that was predominantly coprophagic, which is common in his daily behavior. However, Vail‟s use of searching to locate the fecal samples that he then consumed validates the use of the term search to locate a chemical of interest. The lack of response from Vail to the estrous urine is likely due to his submissive nature within

124

the hierarchy of the housing. Lou is the dominant male and this is an indication of why

Vail did not react.

Hugh and Buffett, both born in captivity and never mated, are trained to react to stimuli and be rewarded. They spent a majority of the trial time begging for food or attention from the researchers and viewing public. They also could not be isolated in the medical pool which made observations difficult because the entire main enclosure could not be observed from a single vantage point. This is why only their location and general behavior are reported.

All of these males have been in captivity their entire adult lives and have never mated or been exposed to a reproductive female. Therefore, these results are preliminary and should be repeated with wild males or those that are in captivity but are known to have mated or been in the wild as reproductive adults.

Despite the naiveté of the males, it is apparent that male manatees are capable of detecting females in estrus. The fact that Snooty at 61 reacted to estrous urine after having never mated indicates how strong this instinct is and it is likely that if this is repeated with reproductively active males the results would be more conclusive.

The behavioral reaction occurs entirely underwater, indicating that the chemicals are brought to receptors through aqueous transmission. This could be done through retronasal delivery from the oral cavity through the caudal portion of the nasal tract to the olfactory epithelium or the taste buds could be involved in the sensing of the chemical(s) of interest within the urine. There also could be a mechanism not yet understood. Researchers have previously alluded to the idea that marine mammals use chemoreceptors that may vary from terrestrial mammals (Oelschlager 2008b, Friedl et

125

al. 1990, Kuznetsov 1990). The manatee presents an opportunity to study this further and determine if their taste buds, olfactory epithelium, or other chemoreceptors are responsible for this behavioral response.

Through the examination of the manatee‟s reproductive behavior and sensory physiology a further understanding of the manatee‟s chemoreceptive ability is possible.

This will aid in determination of environmental factors that may affect the manatee‟s fecundity, such as excess copper, a known negative influence on other aquatic animal‟s chemoreceptive abilities (Sandahl et al. 2007). In addition, with an increased understanding of the physiology of manatee chemoreception and its manifestation as behavioral changes, a more extensive understanding of its role in wild manatee reproduction is possible.

126

CHAPTER 6 CONCLUSIONS

Experimentation on an endangered marine mammal is difficult for many and logistical reasons. Despite this the Florida manatee has provided an opportunity to document for the first time chemoreceptive systems and behavioral responses of a fully aquatic mammal in relation to chemical communication. These results will help to direct future reproductive studies of the manatee and other marine mammals.

The male Florida manatee has demonstrated behaviorally that it responds to chemical cues in the urine of females. The chemical analysis of this urine has demonstrated a significant difference in 64 atomic weights of chemicals present in estrous compared to anestrous urine (Appendix). Through further studies these masses will be analyzed to determine the most likely metabolite candidates for causing the behavioral responses recorded in this study. Then those chemicals can be isolated for more critical analysis techniques including the use of tandem mass spectrometry to identify a more complete representation of the composition of those chemicals. This is the first step to identifying a chemical compound that is expressed and perceived by a fully aquatic marine mammal.

The expression of these chemicals has been demonstrated in urine but is likely also found in the anal glands. The suggestion of a seasonal change in gland size and productivity indicates that manatees may be using their glands to communicate with mates. This is similar to the behavior observed in the river otter, Lutra lutra (Gorman et al. 1978). However, the otter deposits its secretions on land, while the manatee is depositing into an aqueous environment, therefore the chemicals are likely very different

127

in composition. This is the first description of these glands in a fully aquatic mammal but they are likely found in many other marine mammal species.

The mode of reception for these chemical signals is unknown. Their behavioral reaction appears to be entirely underwater, indicating that the chemicals are brought to receptors through aqueous transmission. This could be through retronasal delivery to the olfactory epithelium or the taste buds could be involved in the sensing of the chemical(s) of interest within the urine (Figure 6-1). There also could be a mechanism not yet understood. Researchers have previously alluded to the idea that marine mammals use chemoreceptors that may vary from terrestrial mammals (Friedl et al.

1990, Kuznetsov 1990, Oelschlager 2008b). These have included allusions to potential small, non-taste bud receptors on the tongue (Oelschlager 2008b) or receptors found external to the oral cavity (Kuznetsov 1990).

As demonstrated, the manatee has ample taste buds to perceive a chemical signal but the transmission of pheromonal cues has never been shown to take place through taste. However, the transmission of the chemicals retronasally to the olfactory epithelium would be difficult because of the valve at the base of the nasal canal. The manatee would have to have excellent control of this valve to allow aqueous solutions to pass from the oropharynx and through the nasopharynx to the olfactory epithelium at a minimum volume of water. The manatee does have olfactory epithelium and has demonstrated the ability to sniff at odorants of interest above the water but the manatee does not open their nares underwater and so direct input through the nares is unlikely.

128

Figure 6-1. Overall chemosensory receptor organs within the head of the manatee. Note the foliate papillae and location of taste buds within the soft palate as well as, the apparent taste buds at the orthonasal valve. Arrows represent potential airflow for olfactory detection, both through the nares (1) and retronasal delivery (2).

The undescribed mode of chemical reception could be related to the solitary chemosensory cell (SCC). These cells are used by terrestrial mammals in the nasal canal and trachea to perceive and respond to irritants including bacteria (Lin et al. 2008,

Tizzano et al. 2010, Tizzano et al. 2011). They are found within the skin along the entire length of the body in many fish species and are used by fish to detect signals given off by conspecifics (Kortschal 1995). As the manatee is an aquatic mammal it is possible that the SCC‟s serve both purposes represented by terrestrial mammals and fish.

Further research is needed as to the location and innervation of SCC‟s of the Florida manatee.

129

It appears that the manatee uses some type of sex pheromone but it is also possible that they have an aggregation pheromone as well. Mating herds are composed of anywhere from a few to 12 or more male manatees pursuing a single female

(Hartman 1979, Reid et al. 1995). Males are likely attracted by sound to these herds and by the signal released by the female during estrus; however, other males may also be releasing a signal that over time has evolved into an aggregation pheromone used as and eavesdropping signal. Manatees likely do not use territorial pheromones, alarm pheromones, or pheromones for social organization.

Within mammals, specific effects that are observed in relation to reproduction include the Bruce, Whitten, Vandenbergh, Lee Boot, and Coolidge effects. Any of these effects that require a high density of animals such as the Lee Boot, which inhibits estrus in females by other females (Vandenbergh 1999), are likely not occurring in wild manatee populations. The Bruce effect is a pregnancy block that results when females smell urine released by males that did not fertilize the egg (Bruce 1960). This would be counterproductive in a mating herd because so many unsuccessful males are present.

The Whitten effect results in the induction of estrus in adult females by male pheromones while the Vandenbergh effect accelerates puberty in young females through the same male pheromone signals (Ma et al. 1999, Novotny et al. 1999). These effects could occur in manatees but would be difficult to assess. Finally the Coolidge effect results in an animal‟s renewed interest in mating when presented with a novel mate to increase potential variability in genes (Johnston and Rasmussen 1984). Female manatees are generally pursued during mating herds and will become evasive during

130

them (Hartman 1979, Reid et al. 1995). This behavior does not indicate an interest in a male as would be expected during the Coolidge effect.

In general, the manatee has a reduced sense of smell but an apparently heightened sense of taste within a localized region, the tongue root, of the oral cavity.

An aquatic environment is more conducive to those chemicals that are perceived by the taste buds. The reduction of olfactory use is evident both anatomically with the loss of olfaction in odontocetes (Oeschlager and Oeschlager 2008b), and genetically with reduced olfactory receptors in fish, 50-100 (Ngai et al. 1993, Barth et al. 1996) versus terrestrial mammals, 400-1000 (Parmentier et al. 1992, Ressler et al. 1994). There also may be a correlation between the type of food an animal eats and their degree of taste.

For example, the herbivorous bovine tongue has many more taste buds (14,765-

21,691) than the omnivorous rhesus monkey (8000-10000): the manatee (3,233-

31,140) may have a similar amount of taste buds to the bovine. However, future studies should calculate density of taste buds within an area to take into account the size of the tongue, or area in which taste buds are located. The manatee likely has the highest concentration of taste buds within a small area than any other species yet examined.

From what has been observed during the behavioral trials it appears that a manatee can sense a relatively small amount of urine, five ml, from up to three meters away. In a natural situation it is likely that manatees can sense the much larger normal urine bouts from distances larger than three meters. They are probably using chemical sensing for distances less than a hundred meters but this is likely the primary initial sense that males use to locate females during traveling.

131

Locating females for copulation is likely a combination of learned migratory patterns, to areas where males have been successful in the past, in addition to encountering acoustic cues from other males and perhaps the receptive female, as well as chemical signatures of both the receptive male and other females. Vision and touch are likely involved in the act of mating but not locating mates or mating herds. From the current research it is unclear which chemoreceptors: olfactory, taste, both or something different such as SCC‟s are responsible for the behavioral response.

Anecdotal evidence in addition to the life history patterns of marine mammals indicates that they are capable of chemoreception for both feeding and reproduction

Taste is most likely used by all marine mammal species for detecting salt water gradients and supplementing their food preference. It may also be used in detecting signals given off by other animals either for detection of prey or mates. Many marine mammal species would be served through the use of taste, solitary chemoreceptors such as those found on fish or some other chemical detection mechanism to aid in the detection of mates and the assessment of food. More comprehensive and comparative research is needed in all genera of marine mammals for both taste and smell but the area that is least researched and has been most hypothesized in these animals is the use of chemoreception for chemical signaling in reproduction. The Florida manatee demonstrates that this is a possibility and further research is needed.

Finally, more research is needed in general on the mechanisms of chemoreception. Comparatively little is known about these complex mechanisms compared to the other senses of sight, audition and touch. It would be beneficial to determine what marine mammals are capable of sensing compared to their terrestrial

132

and aquatic counterparts. Aquatic mammals are an important evolutionary link and would provide information about the development and loss of chemoreception that would aid in overall studies of these senses. The aquatic environment is an ideal medium for the use of chemosensory systems and the use of these by marine mammals correlates with what has long been hypothesized.

133

APPENDIX CHEMICAL ANALYSIS OF FEMALE FLORIDA MANATEE, TRICHECHUS MANATUS LATIROSTRIS, URINE

The first step in identification of pheromones released by the manatee for chemical signaling is analysis of the urine. Similar experiments have included analysis of goldfish pheromones in which six females were utilized to verify the release of the steroid hormone 17α, 20β-dihydroxy-4-pregnen-3-one (Dulka et al. 1987). In an aquatic newt a sample size of fifty was used to identify the peptide pheromone, sodefrin, released by the male red bellied newt (Kikuyama et al. 1995). The smaller sample size of the goldfish was reliant on previous research while the larger sample size of the newt was to collect enough secretion to allow for behavioral assays and analysis. The current study on manatees will require a sample size in between these numbers most likely close to the nine individuals required to determine the identity of the elephant pheromone (Z)-7-dodecen-1-yl acetate (Rasmussen et al. 1996).

To begin the identification of potential chemicals in female manatee urine that may be involved in attracting a male for reproduction a metabolomics approach was used. Metabolomics is the characterization of an organism‟s metabolite profile under certain conditions, for example disease or a natural state such as pregnancy. The metabolome or collection of metabolites can represent an individual as a whole or certain tissues, systems or cells within that organism. It represents both the genomes effect on the system and the natural or external conditions (Rochfort 2005). Through comparison of variation in the metabolome of an organism in differing natural states such as those related to reproduction one can develop an understanding of the changes taking place and potential chemical cues used by conspecifics.

134

A global urine metabolomics approach was used to compare metabolome variation between estrous and anestrous female manatee urine. Urine samples came from four captive female manatees that were trained for urine collection (Table A-1)

(Horikoshi 2004). The anestrous and estrous state were determined using urinary hormone levels as measured with radioimmunoassays. Also, three estrous samples;

Lorelei 5/21/03, Charlotte 4-10-05 and Sara 4-4-05 were confirmed to provoke a behavioral response in male manatees. The chemist running the analysis was not aware of which group of samples were estrous and which were anestrous.

Table A-1. Manatee urine samples used for chemical analysis. Anestrous Estrous Lorelei 7/25/03 Lorelei 5/21/03 Lorelei 6/22/03 Lorelei 6/30/03 Lorelei 5/3/03 Lorelei 7/4/03 Lorelei 7/11/03 Lorelei 5/17/03 Charlotte 10/13/05 Charlotte 5/17/05 Charlotte 5/30/05 Charlotte 4/10/05 Sara 5/22/05 Sara 5/7/05 Sara 4/18/05 Sara 6/2/05 Sara 9/18/05 Sara 9/24/05 Sara 9/13/05 Sara 4/4/05 Holly 8/3/10 Holly 2/3/10

One concern with the urine samples was the 6-8 years storage time and 2-3 known freeze-thaw events of the current samples. However, there have been numerous studies targeting the effects of storage and freeze-thaw on metabolite profiles which indicate that for the purposes of qualitative evaluation, there is minimal effect. In one such experiment the effect of up to nine freeze-thaw events, storage in a -20oC versus a

-80oC freezer and storage for up to six months had no effect on the stability of the urine while significant differences were observed following 48 hours or more of the sample at room temperature (Gika et al. 2008). The changes observed in the urine were

135

quantitative with increases in some and decreases in other well known metabolites.

Therefore, it is likely that with proper storage techniques the impact of bacteria and any freeze-thawing would not affect the results of a qualitative study such as this.

Urine was analyzed using a reverse phase high performance liquid chromatography (HPLC) coupled to an Agilent time-of-flight (TOF) mass spectrometer with a positive electrospray ionization. The samples were run in triplicate. This combination allows for the separation and measurement of molecular weight of metabolite ions within a sample. The mass spectrometer has a mass accuracy of less than two ppm which will enable metabolic database searching for identification of small molecule metabolites and, therefore, will work well with pheromones which have a small molecular weight typically between 80 and 300 (Wyatt 2006).

Following mass measurements, the amount of information generated required multivariate statistical tests to indicate masses of interest. Chemometric analyses including a principal components analysis (PCA) and partial least squares regression were employed and are the most commonly used statistical techniques for metabolomics data. Both of which allow for analysis and modeling of complex datasets, that is, datasets that include multiple points of interest such as species, time of collection, control versus experimental group, etc. These techniques also account for noisy, incomplete and collinear data structure analysis. PCA allows for comparison of datasets in a visual model plane that indicate groupings, trends or outliers among the data. Partial least square regression is similar to PCA but instead of providing a number of different dimensional variables it expresses the predicted and observed variables as a linear regression (Trygg et al. 2007).

136

The peaks, which represent the time that an ion traveled and therefore its molecular weight, were aligned and compared using an unpaired t-test. Out of 1,163 aligned peaks, 64 were considered significant and run in a PCA and partial least squares-discriminant analysis (PLS-DA). The PCA showed nice separation between groups A-anestrous and B-estrous (Figure A-1) urine as did the PLS-DA (Figure A-2).

Figure A-1. A PCA plot demonstrating separation between group A-anestrous urine on the left and group B-estrous urine on the right.

Figure A-2. A PLS-DA analysis demonstrating separation in group A-anestrous urine on top and group B-estrous urine on bottom.

137

We will next focus on masses that are significantly different and found in the estrous urine samples that evoked a behavioral response. Then, to determine which metabolite a mass may represent, we will search for it using one of several available databases. There are thousands of potential metabolites expressed in urine and these vary between species, individuals and according to environmental and body conditions.

Databases listing known metabolites and their chemical properties are the best source of identification currently available. The human has the most studied metabolome with several databases listing metabolites. The human metabolome database (HMDB) has over 2100 human metabolites from urine, blood, and cerebrospinal fluid listed including over 90 descriptive parameters for each metabolite (Wishart et al. 2007). The consortium for metabonomic toxicology, COMET, database is also human specific and includes metabolites representing drug toxicity (Rochfort 2005). Maintained by the

National Magnetic Resonance Facility in Madison, Wisconsin the Madison-Qingdao

Metabolomics Consortium Database (MMCD) pulls information from public databases such as Kyoto Encyclopedia of Genes and Genomes (KEGG) and PubChem to provide over 50 parameters on metabolites from a variety of species (Cui et al. 2008). There is also a bovine database being developed that will provide the most comparable metabolites for manatees.

We would also like to expand our criteria to include more potential chemical masses. Those shown above were present in a least half the samples, but there was already a natural division of the samples in half, it would be more accurate to put the constraint of presence in at least a quarter of the samples. This would mean that it would be found in at least half of the estrous or half of the anestrous samples. Following

138

identification of masses of interest those ions at the mass and retention time will be isolated and analyzed using tandem mass spectrometry. This will give a more accurate description of the components of the chemical.

The separation between the two groups, A-anestrous and B-estrous, is encouraging and indicates that there are metabolites that serve the purpose of signal expression. To identify these metabolites we need a larger sample size, closer to the six or nine individuals used in goldfish (Dulka et al. 1987) and elephant (Rasmussen et al.

1996) analysis. We also need more examples of estrous and anestrous urine, e.g., various dates, to increase the samples available for analysis.

139

LIST OF REFERENCES

Ache, B.W., A.B. Zhainazarov. 1995. Dual second-messenger pathways in olfactory transduction. Current Opinions in Neurobiology. 5:461-466.

Achiraman S., P. Ponmanickam, D.S. Ganesh, G. Archunan. 2010. Detection of estrus by male mice:Synergistic role of olfactory-vomeronasal system. Neuroscience Letters. 477(3):144-148.

Adams D.R., 1972. Olfactory and non-olfactory epithelium in the nasal cavity of themouse, Peromyscus. American Journal of Anatomy. 133:37-50.

Adler E., M.A. Hoon, K.L. Mueller, J.Chandrashekar, N.J.P. Ryba, C.S. Zuker. 2000. A novel family of mammalian taste receptors. Cell. 100(6):693–702.

Arvy L., G. Pilleri. 1970. The tongue of Platanista gangetica and remarks on the cetacean tongue. Investigations on Cetacea Vol. 2. Editor Pilleri G. Pp75-77.

Atoji Y., Y. Yamamoto, Y. Suzuki. 1998. Apocrine sweat glands in the circumanal glands of the dog. The Anatomical Record. 252(3):403-412.

Bagley K.R., T.E. Goodwin, L.E.L. Rasmussen, B.A.Schulte. 2006. Male African elephants, Loxodonta Africana, can distinguish oestrous status via urinary signals. Animal Behaviour. 71:1439-1445.

Barata E.N., R.M. Serrano, A. Miranda, R. Nogueira, P.C. Hubbard, A.V.M. Canario. 2008. Putative pheromones from the anal glands of male blennies attract females and enhance male reproductie success. Animal Behaviour. 75:379-389.

Barth A.L., N.J. Justice, J. Ngai. 1996. Asynchronous onset of odorant receptor expression in the developing zebrafish olfactory system. Neuron. 16:23-34.

Barrett J., D.H. Abbott, L.M. George. 1990. Extension of reproductive suppression by pheromonal cues in subordinate female marmoset monkeys, Callithrix jacchus. Journal of Reproduction and Fertility. 90:411-418.

Behan M., C.F. Thomas. 2005. Sex hormone receptors are expressed in identified respiratory motoneurons in male and female rats. Neuroscience. 130(3):725-734.

Bergeron J.M., D. Crews, J.A. McLachlan. 1994. PCBs as environmental estrogens:turtle sex determination as a biomarker of environmental contamination. Environmental Health Perspectives 102:780-781.

Berta A., J.L. Sumich, K.M. Kovacs. 2006. Marine mammals:Evolutionary biology. 2nd Ed. Academic Press, Oxford, UK.

Bertmar G. 1981. Evolution of the vomeronasal organs in vertebrates. Evolution. 35:359-366.

140

Bigiani A., C. Mucignat-Caretta, G. Montani, R. Tirindelli. 2005. Pheromone reception in mammals. Reviews of Physiology, Biochemistry, and Pharmacology. 155:1-35.

Bjerselius R., K. Lundstedt-Enkel, H. Olsen. I. Mayer, K. Dimberg. 2001. Male goldfishreproductive behavior and physiology are severely affected by exogenous exposure to 17β-estradiol. Aquatic Toxicology. 53(2):139-152

Bland K.P. 1979. Tom-Cat odour and other pheromones in feline reproduction. Veterinary Science Communications. 3:125-136.

Bowen W.D., O.T. Oftedal, D.J. Boness. 1985. Birth to weaning in 4 days: remarkable growth in the hooded seal, Cystophora cristata. Canadian Journal of Zoology. 63:2841-2846.

Bradley R.M., H.M. Stedman, C.M. Mistretta. 1985. Age does not affect numbers of taste buds and papillae in adult rhesus monkeys. The Anatomical Record. 212:246-249.

Breathnach A.S. 1955. The surface features of the brain of the humpback whale (Megaptera novaeangliae). Journal of Anatomy. 89:343-354.

Brennan P.A., F. Zufall. 2006. Pheromonal communication in vertebrates. Nature. 444:308-315.

Bruce HM. 1960. A block to pregnancy in the mouse caused by proximity of strangemales. Journal of Reproductive Fertility.1:96–103.

Byers J.A. 1995. Host-tree chemistry affecting colonization in bark beetles. In Carde R.T. W.J. and Bell, editors. Chemical ecology of insects 2. Chapman and Hall. New York, NY, pp 154-213.

Ceruti M.G., P.V. Fennessey, S.S. Tjoa. 1985. Chemoreceptively active compounds in secretions, excretions and tissue extracts of marine mammals. Comparative Biochemistry and Physiology. 82A(3):505-514.

Chaudhari N., S. D. Roper. 2010. The cell biology of taste. The Journal of Cell Biology. 190(3):285-296.

Cheek A.O., P.M. Vonier, E. Oberdorster, B.C. Burow, J.A. McLachlan. 1998. Environmental signaling: a biological context for endocrine disruption. Environmental Health Perspectives. 106(Supplement 1):5-10.

Collings V.B. 1974. Human taste response as a function of locus of stimulation on the tongue and soft palate. Perceptions and Psychophysics. 16(1):169-174.

Cui Q., I.A. Lewis, A.D. Hegeman, M.E. Anderson, J. Li, C.F. Schulte, W.M. Westler, H.R. Eghbalnia, M.R. Sussman, J.L. Markley. 2008. Metabolite identification via the Madison Metabolomics Consortium Database. Nature Biotechnology. 26:162.

141

Date-Ito A., H. Ohara, M. Ichikawa, Y. Mori, K. Hagino-Yamagishi. 2008. Xenopus V1R vomeronasal receptor family is expressed in the main olfactory system. Chemical Senses. 33:339-346.

Davies R.O., M.R. Kare, R.H. Cagan. 1979. Distribution of taste buds on fungiform and circumvallate papillae of the bovine tongue. The Anatomical Record. 195:443-446.

Deutsch C.J., J.P. Reid, R.K. Bonde, D.E. Easton, H.I. Kockman, T.J. O‟Shea. 2003. Seasonal movements, migratory behavior, and site fidelity of West Indian Manatees along the Atlantic coast of the United States. Wildlife Monographs. 151:1-77.

Donovan C.A. 1969. Canine anal glands and chemical signals (pheromones). Journal of American Veterinary Medical Association. 155(12):1995-6.

Dorries K.M., E. Adkins-Regan, B.P. Halpern. 1997. Sensitivity and behavioral responses to the pheromone androstenone are not mediated by the vomeronasal organ in domestic pigs. Brain Behavior and Evolution. 49(1):53-62.

Downing S.E., J.C. Lee. 1975. Laryngeal chemosensitivity: A possible mechanism of sudden infant death. Pediatrics. 55(5):640-649.

Dreanno C., R.R. Kirby, A.S. Clare. 2007. Involvement of the barnacle settlement- inducing protein complex (SIPC) in species recognition at settlement. Journal of Experimental Marine Biology and Ecology. 351(1-2):276-282.

Duffield D.W., J.T. Haldiman, W.G. Henk. 1992. Surface morphology of the forebrain of the bowhead whale, Balaena mysticetus. Marine Mammal Science. 8(4):354-378.

Dulac C., R. Axel R. 1995. A novel family of genes encoding putative pheromone receptors in mammals. Cell. 83:195–206.

Dulka J.G., N.E. Stacey, P.W. Sorensen, G.J. Van Der Kraak. 1987. A steroid sex pheromone synchronizes male-female spawning readiness in goldfish. Nature. 325:251-253.

Eglitis J.A., I. Eglitis. 1961. The glands of the anal canal in man. The Ohio Journal of Science. 61:65-79.

Eisthen H.L. 1992. Phylogeny of the vomeronasal system and of receptor cell types in the olfactory and vomeronasal epithelia of vertebrates. Microscopy Research and Technique. 23:1-21.

El-Sharaby A., K. Ueda, K. Kurisu, S. Wakisaka. 2001. Development and maturation of taste buds of the palatal epithelium of the rat: histological and immunohistochemical study. Anatomical Record. 263(3):260-268.

142

Emanuelson K. 2006. Neonatal care and hand rearing. In Fowler M.E., S.K. Mikota, editors. Biology, medicine, and surgery of elephants. Blackwell Publishing. Ames, Iowa, pp 233-242.

Fedak M.A., B. Wilson, P.P. Pomeroy. 2002. Reproductive behavior. In: Perrin F.A., B. Wursig, H. Thewissen, editors. Encyclopedia of marine mammals. San Diego: Academic Press. 1015-1027.

Finger T.E. 1997. Feeding patterns and brain evolution in ostariophysean fishes. Acta Physiologica Scandinavica Suppl. 638:59-66.

Finger T.E., W.L. Silver, D. Restrepo. 2000. The neurobiology of taste and smell. Wiley- Liss. New York, NY.

Finger T.E., B. Bottger, A. Hansen, K.T. Anderson, H. Alimohammadi, W.L. Silver. 2003. Solitary chemoreceptor cells in the nasal cavity serve as sentinels of respiration. Proceedings of the National Academy of Sciences. 100:8981–86.

Finger T.E. 2005. Cell Types and lineages in taste buds. Chemical Senses. 30:i54-i55.

Finger T.E., V. Danilova, J. Barrows, D.L. Bartel, A.J. Vigers, L. Stone, G. Hellekant, S.C. Kinnamon. 2005. ATP signaling is crucial for communication from taste buds to gustatory nerves. Science. 310:1496-1499.

Fishelson L., Y. Delarea. 2004. Taste buds on the lips and mouth of some blennid and gobiid fishes: comparative distribution and morphology. Journal of Fish Biology. 65:651-665.

Freitag J., G. Ludwig, I. Andreini, P. Rossler, H. Breer. 1998. Olfactory receptors in aquatic and terrestrial vertebrates. Journal of Comparative Physiology A. 183:635- 650.

Friedl W.A., P.E. Nachtigall, P.W.B. Moore, N.K.W. Chun. 1990. Taste reception in the Pacific bottlenose dolphin (Tursiops truncatus gilli) and the California sea lion (Zalophus californianus). In: Thomas J., R. Kastelein, editors. Sensory abilities of cetaceans.Plenum, New York, pp 447–451.

Gika H.G., G.A. Theodoridis, I.D. Wilson. 2008. Liquid chromatography and ultra- performance liquid chromotagraphy-mass spectrometry fingerprinting of human urine:sample stability under different handling and storage conditions for metabonomics studies. Journal of Chromotagraphy A. 1189(1-2):314-322.

Gobbel L., M.S. Fischer, T.D. Smith, J.R. Wible, K.P. Bhatnagar. 2004. The vomeronasal organ and associated structures of the fetal African elephant, Loxodonta Africana (Proboscidea, Elephantidae). Acta Zoologica Stockholm. 85:41-52.

143

Gomes F.F.A., J.E. Vergara-Parente, S.F. Ferrari. 2008. Behaviour patterns in captive manatees (Trichechus manatus manatus) at Itamaraca Island, Brazil. Aquatic Mammals. 34(4):269-276.

Gorman M.L., D. Jenkins, R.J. Harper. 1978. The anal scent sacs of the otter (Lutra lutra). Journal of Zoology. 186:463-474.

Greenwood D.R., D. Comesky, M.B. Hunt, L.E.L. Rasmussen. 2005. Chemical communication-chirality in elephant pheromones. Nature. 438(7071):1097-1098.

Greer M.B., M.L. Calhoun. 1966. Anal sacs of the cat (Felis domesticus). American Journal of Veterinary Research. 27(118)773-781.

Gross E.A., J.A. Swenberg, S. Fields, J.A. Popp. 1982. Comparative morphometry of the nasal cavity in rats and mice. Journal of Anatomy. 135(1):83-88.

Hartman D.S. 1979. Ecology and Behavior of the manatee (Trichechus manatus) in Florida. Special publication No. 5: The American Society of Mammalogists. Ed. James Layne. 118-120.

Halpern M. 1987. The Organization and Function of the Vomeronasal System. Annual Review of Neuroscience. 10:325-362.

Halpern M., L.S. Shapiro, C.P. Jia. 1995. Differential localization of G proteins in the opossum vomeronasal system. Brain Research. 677:157-161.

Harrison R.J., G.L. Kooyman. 1968. General physiology of pinnipedia. In The behavior and physiology of pinnipeds. Harrison R.J., R.C. Hubbard, R.S. Petterson, C.E. Rice, R.J. Schusterman, editors. Appleton Century Crofts, New York. Pp 212-296.

Herrada G., C. Dulac. 1997. A novel family of putative pheromone receptors in mammals with a topographically organized and sexually dimorphic distribution. Cell. 90:763–773.

Höfer D., B. Puschel, D. Drenckhahn. 1996. Taste receptor-like cells in the rat gut identified by expression of α-gustducin. Proceedings of the National Academy of Sciences. USA. 93: 6631– 34.

Höfer D., D. Drenckhahn.1998. Identification of the taste cell G-protein alpha-gustducin in brush cells of the rat pancreatic duct system. Histochemistry and Cell Biology. 110:303–309.

Holldobler B., E.O. Wilson. 1990. The ants. Berlin: Springer-Verlag.

Horikoshi C.2004. Effect of hormonal and temporal factors on captive female manatee (Trichechus manatus latirostris) behavior. MS Thesis. Georgia Southern University, Statesboro, GA. Pp 117.

144

Horikoshi-Beckett C., B.A. Schulte. 2006. Activity patterns and spatial use of facility by a group of captive female manatees (Trichechus manatus latirostris). Zoo Biology. 25(4): 285-301.

Hosley M., B. Oakley. 1987. Postnatal development of the vallate papilla and taste buds in rats. The Anatatomical Record. 218, 216–222.

Howe N.R. 1976. Behavior of sea anemones evoked by the alarm pheromone anthopleurine. Journal of comparative physiology A: Neuroethology, sensory, neural, and behavioral physiology. 107(1):67-76.

Iwasaki, S. 2002. Evolution of the structure and function of the vertebrate tongue. Journal of Anatomy. 201:1-13.

Johnson E.W., L.E.L. Rasmussen. 2002. Morphological characteristics of the vomeronasal organ of the newborn asian elephant (Elephas maximus). The Anatomical Record. 267:252-259.

Johnston R.E. 1998. Pheromones, the Vomeronasal System and Communication- From Hormonal Responses to Individual Recognition. Olfaction and Taste XII-An international Symposium. 855:333-348.

Johnston R.E., K. Rasmussen. 1984. Individual recognition of female hamsters by males: role of chemical cues and of the olfactory and vomeronasal systems. Physiology and Behavior. 33:95-104.

Jones T.R., R.K. PLakke. 1981. The histology and histochemistry of the perianal scent gland of the reproductively quiescent black-tailed prairie dog (Cynomys ludovicianus). Journal of Mammalogy. 62(2):362-368.

Kastelein R.A., J.L. Dubbeldam, M.A.G. deBakker. 1997. The anatomy of the walrus head (Odobenus rosmarus). Part 5: The tongue and its function in walrus ecology. Aquatic Mammals. 23(1):29-47.

Kavoi B., A. Makanya, J. Hassanali, H. Carlson, S. Kiama. 2010. Comparative functional structure of the olfactory mucosa in the domestic dog and sheep. Annals of Anatomy. 192:329-337.

Kelliher K.R. 2007. The combined role of the main olfactory and vomeronasal system in social communication in mammals. Hormones and behavior. 52:561-570.

Kikuyama S., F. Toyoda, Y. Ohmiya, K. Matsuda, S. Tanaka, H. Hayashi. 1995. Sodefrin:a female attracting peptide pheromone in newt cloacal glands. Science. 267:1643-1645.

Kimchi T., J. Xu, C. Dulac. 2007. A functional circuit underlying male sexual behavior in the female mouse brain. Nature. 448:1009-1014.

145

King J.M., J.T.Heinen. 2004. An assessment of the behaviors of overwintering manatees as influenced by interactions with tourists at two sites in central Florida. Biological Conservation.117:227-234.

Kleinschmidt T., J. Czelusniak, M. Goodman, G. Braunitzer. 1986. Paenungulata: A comparison of the hemoglobin sequence from the elephant, hyrax, and manatee. Molecular Biology and Evolution. 3(5):427-435.

Komatsu S., F. Yamasaki. 1980. Formation of the pits with taste-buds at the ligual root in the striped dolphin, Stenella coeruleoalba. Journal of Morphology. 164(2):107- 119.

Kotrschal K., W.D. Krautgartner, A. Hansen. 1997. Ontogeny of the solitary chemosensory cells in the zebrafish, Danio rerio. Chemical Senses. 22(2):111- 118.

Kotrschal K. 2000. Taste(s) and olfaction(s) in fish: a review of specialized sub-systems and central integration. Pflugers Archives. 439:R178-180.

Kowalesky S., M. Dambach, B. Mauck, G. Dehnhardt. 2006. High olfactory sensitivity for dimethyl sulphide in harbor seals. Biology Letters. 2(1):106-109.

Kruger L. 1959. The thalamus of the dolphin (Tursiops truncates) and comparison with other mammals. Journal of Comparative Neurology. 111:133-194.

Kuznetsov V.B. 1974. A method of studying chemoreception in the Black Sea bottlenose dolphin (Tursiops truncatus). Morphology, physiology and acoustics of marine mammals. Nauka Publ. House, Moscow. 27-45. as cited in Lowell WR, Flanigan WF. 1980. Marine Mammal Chemoreception. Mammal Review. 10(1):53- 59.

Kuznetzov V.B. 1990. Chemical sense of dolphins: Quasi-olfaction. In: Thomas J, Kastelein R, editors. Sensory abilities of cetaceans. Plenum, New York, pp 481- 503.

Ladewig J., B.L. Hart. 1980. Flehmen and vomeronasal organ function in male goats. Physiology and Behavior. 24(6):1067-1071.

Langbauer W.R. 2000. Elephant communication. Zoo Biology. 19:425-445.

Larkin I.V. 2000. Reproductive endocrinology of the Florida manatee, (Trichechus manatus latirostris):Estrous cycles, seasonal patterns and behavior. Phd Disseration. University of Florida, Gainesville,FL. Pp 354.

Lee P.C. 1987. Allomothering among African elephants. Animal Behavior. 35:278-291.

146

Levasseur A., L. Orlando, X. Bailly, M.C. Milinkovitch, E.G.J. Danchin, P. Pontarotti. 2007. Conceptual bases for quantifying the role of the environment on gene evolution: the participation of positive selection and neutral evolution. Biological Review. 82:551-572.

Levin M.J., C.J. Pfeiffer. 2002. Gross and microscopic observations on the lingual structure of the Florida manatee, Trichechus manatus latirostris. Anatomia, Histologia, Embryologia. 31:278-285.

Li X., L. Staszewski, H. Xu, K. Durick, M. Zoller, E. Adler. 2002. Human receptors for sweet and umami taste. Proceedings of the National Academy of Sciences. 99(7):4692–4696.

Li Y. 1983. The tongue of the Baiji, Lipotes vexillifer. Acta Zoologica Sinica. 29:35 –47

Lin W., T. Ogura, R.F. Margolskee, T.E. Finger, D. Restrepo. 2008. TRPM5-expressing solitary chemosensory cells respond to odorous irritants. Journal of Neurophysiology. 99:1451-1460.

Lofqvist J. 1976. Formic acid and saturated hydrocarbons as alarm pheromones for the ant Formica rufa. Journal of Insect Physiology. 22:1331-1346.

Lofstedt C., N.J. Vickers, W.L. Roelofs, T.C. Baker. 1989. Diet related courtship success in the oriental fruit moth, Grapholita molesta (Tortricidae). Oikos. 55:402- 408.

Lowell W.R., W.F. Flanigan. 1980. Marine Mammal Chemoreception. Mammal Review. 10(1):53-59.

Ma W.D., Z.S. Miao, M.V. Novotny. 1999. Induction of estrus in grouped female mice (Mus domesticus) by synthetic analogues of preputial gland constituents. Chemical Senses. 24:289-293.

Ma H., R. Yang, S.M. Thomas, J. C. Kinnamon. 2007. Qualitative and quantitative differences between taste buds of the rat and mouse. BMC Neuroscience. 8:5

Mack A., A. Singh, C. Gilroy, W. Ireland. Porcine lingual taste buds:a quantitative study. The Anatomical Record. 247:33-37.

Mackay-Sim A., D. Duvall, B.M. Graves. 1985. The West Indian manatee (Trichechus manatus) lacks a vomeronasal organ. Brain Behavior and Evolution. 27:186-194.

Marchand J.E., X. Yang, D. Chikaraishi, J. Krieger, H. Breer, J.S. Kauer. 2004. Olfactory receptor gene expression in tiger salamander olfactory epithelium. Journal of Comparative Neurology. 474:453-467.

Marino L. 2007. Cetacean brains: How aquatic are they? The Anatomical Record. 290:694-700.

147

Marmontel M. 1988. The reproductive anatomy of the female manatee Trichechus manatus latirostris (Linnaeus 1758) based on gross and histologic observations. M.S. Thesis. University of Miami, Miami, FL. Pp 91.

Marmontel M. 1995. Age and reproduction in female Florida manatees. In O‟Shea T.J, B.B. Ackerman, H.F. Percival, editors. Population biology of the Florida manatee. Information and Technology report 1. Washington, DC, Pp 98-119.

Martin P., P. Bateson. 2007. Measuring Behavior: An Introductory Guide. 3rd ed. New York: Cambridge University Press.

Matsunami H., L.B. Buck. 1997. A multigene family encoding a diverse array of putative pheromone receptors in mammals. Cell. 90:775–784.

Matsunami H., J.P. Montmayeur, L.B. Buck. 2000. A family of candidate taste receptors in human and mouse. Nature. 404:601–604.

McColl I. 1967. The comparative anatomy and pathology of anal glands: Arris and Gale Lecture delivered at the Royal College of Surgeons of England. Annals of the Royal College of Surgeons of England. 40:36-67.

McGowen M.R., C. Clark, J. Gatesy. 2008. The Vestigial Olfactory Receptor Subgenome of Odontocete Whales: Phylogenetic Congruence between Gene- Tree Reconciliation and Supermatrix Methods. Systems Biology. 57(4): 574-590.

Meisami E., K. Bhatnagar. 1998. Structure and diversity in mammalian accessory olfactory bulb. Microscopy Research and Technique. 43:476-499.

Miller I.J., K.M. Spangler. 1982. Taste bud distribution and innervation on the palate of the rat. Chemical Senses. 7:99-108.

Miller I.J., D.V. Smith. 1984. Quantitative taste bud distribution in the hamster. Physiology and Behavior. 32:275-285.

Mistretta C.M., B.J. Baum. 1984. Quantitative study of taste buds in fungiform and circumvallate papillae of young and aged rats. Journal of Anatomy. 138(2):323- 332.

Montagna W., H.F. Parks. 1948. A histochemical study of the glands of the anal sac of the dog. Anatomical Record. 100(3):297-317.

Montie E.W., N. Pussini, G.E. Schneider, T.W.K. Battey, S. Dennison, J. Barakos, F. Gulland. 2009. Neuroanatomy and volumes of brain structures of a live California sea lion (Zalophus californianus) from magnetic resonance images. The Anatomical Record. 292:1523-1547.

Moore J.C. 1956. Observations of manatees in aggregations. American Museum Novitates. 1811:1-24.

148

Nachtigall P.E., R.W. Hall. 1984. Taste reception in the bottlenose dolphin. Acta Zoologica Fennica. 172:147-148.

Nei M., Y. Niimura, M. Nozawa. 2008. The evolution of animal chemosensory receptor gene repertoires: roles of chance and necessity. Nature Reviews:Genetics. 951- 963.

Ngai J., A. Chess, M.M. Dowling. 1993. Coding of olfactory information: Topography of odorant receptor expression in the catfish olfactory epithelium. Cell. 70:667-680.

Novotny M.V., W.D. Ma, D. Wiesler, L. Zidek. 1999. Positive identification of the puberty-accelerating pheromone of the house mouse: the volatile ligands associating with the major urinary protein. Proceedings of the royal society of London series B-biological sciences. 266:2017-2022.

Nowacek D.P., B.M. Casper, R.S. Wells, S.M. Nowacek, D.A. Mann. 2003. Intraspecific and geographic variation of West Indian manatee (Trichechus manatus spp.) vocalizations. Journal of the Acoustical Society of America. 114(1):66-69.

Oelschlager H.A., E.H. Buhl. 1985. Occurrence of an olfactory bulb in the early development of the harbor porpoise (Phocoena phocoena L.). In: Duncker H.R., Fleischer, editors. Functional morphology in vertebrates. New York: Fischer. Pp 695–698

Oelschlager H.A. 1989. Early development of the olfactory and terminalis systems in baleen whales. Brain behavior and evolution. 34:171-183.

Oelschläger H. A. 1992.Development of the olfactory and terminalis systems in whales and dolphins. In: Doty R. L., D. Müller-Schwarze, editors. Chemical signals in vertebrates VI. New York: Plenum Press; 1992. p. 141-147.

Oelschlager H.A., B. Kemp. 1998. Ontogenesis of the sperm whale brain. The Journal of Comparative Neurology. 399:210-228.

Oelschlager H.A., J.S. Oelschlager. 2002. Brains. In: Perrin F.A., B. Wursig, H. Thewissen. editors. Encyclopedia of marine mammals. San Diego: Academic Press. 133–158.

Oelschlager H.A., M. Haas-Rioth, C. Fung, S.H. Ridgway, M. Knauth. 2008a. Morphology and Evolutionary Biology of the Dolphin (Delphinus sp.) Brain–MR Imaging and Conventional Histology. Brain Behavior and Evolution. 71:68-86.

Oelschlager H.A., J. S. Oelschlager. 2008b. Brains. In: Perrin F.A., B. Wursig, H. Thewissen, editors. Encyclopedia of marine mammals. San Diego: Academic Press. 133–158.

Olds N., J. Shoshani. 1982. Procavia capensis. Mammalian Species. 171:1-7.

149

Pardini A.T., P.C.M. Obrien, B. Fu, R.K. Bonde, F.F.B. Elder, M.A. Ferguson-Smith, F. Yang, T.J. Robinson. 2007. Chromosome painting among Proboscidea, Hyracoidea, and Sirenia: support for Paenungulata (Afrotheria, Mammalia) but not Tehtytheria. Proceedings of the Royal Society B. 274:1333-1340.

Parmentier M,. F.Libert, S. Schurmans, S. Schiffmann, A. Lefort, D. Eggerickx, C. Ledent, C. Mollereau, C. Gerard, J. Perret, A. Grootegoed, G. Vassart. 1992. Expression of members of the olfactory receptor gene family in mammalian germ cells. Nature. 355:453-455.

Pelosi P. 1994. Odorant-binding proteins. Critical Reviews in Biochemical and Molecular Biology. 29:199-228.

Pfeiffer D.C., A. Wang, J. Nicolas, C.J. Pfeiffer. 2001. Lingual ultrastructure of the long- finned pilot whale (Globicephala melas). Anatomia, Histologia, Embryologia. 30:359-365.

Phillips AV. 2003. Behavioral cues used in reunions between mother and pup south American fur seals (Arctocephalus australis). Journal of Mammalogy. 84(2):524- 535.

Prahl S., S. Huggenberger, H. Schliemann. 2009. Histological and ultrastructural aspects of the nasal complex in the harbour porpoise, Phocoena phocoena. Journal of morphology. 270:1320-1337.

Purdom C.E., P.A. Hardiman, V.J. Bye, N.C. Eno, C.R. Tyler, J.P. Sumpter. 1994. Estrogenic effects of effluent from sewage treatment works. Chemical Ecology. 8:275-285.

Rasmussen L.E.L. 1999. Evolution of chemical signals in the Asian Elephant, Elephas maximus: behavioral and ecological influences. Journal of bioscience. 24(2):241- 251.

Rasmussen L.E.L., M.J. Schmidt, R. Henneous, D. Groves, G.D. Daves. 1982. Asian bull elephants:flehman-like responses to extractable components in female elephant estrous urine. Science. 217:159-162.

Rasmussen L. E. L., T.D. Lee, W.L. Roelofs, A.J. Zhang, G.D. Daves. 1996. Insect pheromone in elephants, Nature 379:684.

Rathbun GB, T.J. O‟Shea. 1984. The manatee‟s simple social life: scent marking in an aquatic mammal. The Encyclopedia of Mammals. Ed. David Macdonald. Facts on File: New York. 300-301.

150

Rathbun G.B., J.P. Reid, R.K. Bonde, J.A. Powell. 1995. Reproduction in free-ranging Florida manatees. In:O‟Shea TJ, Ackerman BB, Percival HF, editors. Population Biology of the Florida Manatee. Washington DC: US Department of Interior, National Biological Service, Information and Technology Report No. 1. Pp 135- 156.

Reep R.L., C.D. Marshall, M.L. Stoll. 2002. Tactile hairs on the postcranial body in Florida manatees: A mammalian lateral line? Brain Behavior and Evolution. 59(3):141-154.

Reep R.L., and B.K. Bonde. 2006. The Florida Manatee: biology and conservation. University Press of Florida, Gainesville, FL.

Reep R.L., B.L. Finlay, R.B. Darlington. 2007. The limbic system in mammalian brain evolution. Brain, Behavior and Evolution. 70:57-70.

Reeves R.R., B.S. Stewart, P.J. Clapham, J.A. Powell, P.A. Folkens. 2002. National Audubon Society:Guide to marine mammals of the world. Chanticleer Press, New York, New York.

Ressler K.J., S.L. Sullivan, L.B. Buck. 1994. Information coding the olfactory system: Evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell. 79(7):1245-1255.

Reynolds J.E., D.K. Odell. 1991. Manatees and Dugongs. New York: Facts on File. Pp 41.

Rochfort S. 2005. Metabolomics reviewed:a new „omics‟ platform technology for systems biology and implications for natural products research. Journal of natural products. 68:1813-1820.

Rosell F., B.A. Schulte. 2004. Sexual dimorphism in the development of scent structures for the obligate monogamous Eurasian beaver (Castor fiber). Journal of Mammalogy. 85(6):1138-1144.

Ryg M., Y. Solberg, C. Lydersen, T.G. Smith. 1992. The scent of rutting male ringed seals (Phoca-hispida). Journal of Zoology. 226:681-689.

Sandahl J.F., D.H. Baldwin, J.J. Jenkins, N.L. Scholz. 2007. A sensory system at the interface between urban stormwater runoff and salmon survival. Environmental Science and Technology. 41:2998-3004.

Sarko D.K., and R.L. Reep. 2007. Somatosensory areas of manatee cerebral cortex: histochemical characterization and functional implications. Brain, Behavior and Evolution. 69:20-36.

Sbarbati A., F. Osculati. 2003. Solitary chemosensory cells in mammals? Cell Tissue Organs. 175:51-55.

151

Schulte B.A., L.E.L. Rasmussen. 1999. Musth, sexual selection, testosterone, and metabolites. In Johnston R.E., D. Muller-schwarze, P.W. Sorenson, editors. Advances in chemical signals in vertebrates. Kluwer Academic publishers/Plenum press. New York, NY, pp 383-395.

Schulte B.A. 2006. Behavior and social life. In Fowler M.E., S.K. Mikota, editors. Biology, medicine, and surgery of elephants. Blackwell Publishing. Ames, Iowa, Pp 35-44.

Schaumburg-Lever G., W.F. Lever. 1975. Secretion from human apocrine glands: an electron microscopic study. The Journal of Investigative Dermatology. 64(1):38-41.

Shanle E.K., W. Xu. 2011. Endocrine disrupting chemicals targeting estrogen receptor signaling: identification and mechanism of action. Chemical Research in Toxicology. 24(1):6-19.

Shepherd G.M. 2006. Smell images and the flavour system in the human brain. Nature. 444:316-321.

Shindo J., T.K. Yamada, K. Yoshimura, I. Kageyama. 2008. Morphology of the tongue in a newborn Stejneger‟s beaked whale (Mesoplodon stejnegeri). Okajimas Folia Anatomica Japonica. 84(4):121-124.

Simpson G.G. 1945. The principles of classification and a classification of mammals. Bulletin of the American museum of natural history. 85:1-350.

Sinclair J.G. 1966. The olfactory complex of dolphin embryos. Texas Report on Biology and Medicine. 24:426-431.

Sonntag C.F. 1923. The comparative anatomy of the tongues of the Mammalia. 8. Carnivora. Proceedings of the Zoological Society of London. 93:129-153.

Sorensen P.W. 1996. Biological responsiveness to pheromones provides fundamental and unique insight into olfactory function. Chemical Senses. 21:245-256

Sorenson PW, Stacey NE. 1999. Evolution and specialization of fish hormonal pheromones. In Johnston RE, Muller-schwarze D, Sorenson PW, editors. Advances in chemical signals in vertebrates. Kluwer Academic publishers/Plenum press. New York, NY, Pp 15-48.

Stahlbaum C.C., K.A. Houpt. 1989. The role of Flehmen response in the behavioral repertoire of the stallion. Physiology and Behavior. 45(6):1207-1214.

Sticken J., G. Dehnhardt. 2000. Salinity discrimination in harbour seals: a sensory basis for spatial orientation in the marine environment? Naturwissenschaften. 87:499- 502.

152

Stobel A., A. Junold, M.S. Fischer. 2010. The morphology of the eutherian ethmoidal region and its implications for higher-order phylogeny. Journal of Zoological Systematics and Evolutionary Research. 48(2): 167-180.

Supin A.Y., V.V. Popov, A.M. Mass. 2001. The sensory physiology of aquatic mammals. Boston: Kluwer Academic.

Suzuki T. 2007. Cellular mechanisms in taste buds. The Bulletin of Tokyo Dental College. 48(4):151-161.

Switzer R.C., Johnson J.I., Kirsch J.A.W. 1980. Phylogeny through brain traits: Relation of lateral olfactory tract fibers to the accessory olfactory formation as a palimpsest of Mammalian descent. Brain Behavior and Evolution. 17:339-363.

Temara A., I. Gulec, D.A. Holdway. 1999. Oil-induced disruption of foraging behaviour of the asteroid keystone predator, Coscinasterias muricata (Echinodermata). Marine Biology. 133:501-507.

Thewissen J.G.M., J. George, C. Rosa, T. Kishida. 2011. Olfaction and brain size in the bowhead whale (Balaena mysticetus). Marine Mammal Science. 27(2):282-294.

Thornhill R. 1979. Male pair formation pheromones in Panorpa scorpionflies (Mecoptera:Panorpidae). Environmental Entomology. 8:886-889.

Tizzano M., B.D. Gulbransen, A. Vandenbeuch, T.R. Clapp, J.P. Herman, H.M. Sibhatu, M.E.A. Churchill, W. L. Silver, S.C. Kinnamon, T.E. Finger. 2010. Nasal chemosensory cells use bitter taste signaling to detect irritants and bacterial signals. Neuroscience. 107(7):3210-3215.

Tizzano M., M. Cristofoletti, A. Sbarbati, T. E. Finger. 2011. Expression of taste receptors in solitary chemosensory cells of roden airways. BMC Pulmonary Medicine. 11:3.

Toyoshima K., Y. Seta, T. Toyono, S. Kataoka. 2007. Immunohistochemical identification of cells expressing steroidogenic enzymes cytochrome P450scc and P450 aromatase in taste buds of rat circumvallate papillae. Archives of Histology and Cytology. 70(4):215-224.

Trygg J., E. Holmes, T. Lundstedt. 2007. Chemometrics in metabonomics. Journal of Proteome Research. 6:469-479.

Twan W.H., J.S. Hwang, Y.H. Lee, H.F. Wu, Y.H. Tung, C.F. Chan. 2006. Hormones and reproduction in scleractinian corals. Comparative Biochemistry and Physiology A-Molecular and Integrative Physiology. 144(3):247-253.

153

Valimaa H., S. Savolainen, T. Soukka, P. Silvoniemi, S. Makela, H. Kujari, J.A. Gustafsson, M. Laine. 2004. Estrogen receptor-β is the predominant estrogen receptor subtype in human oral epithelium and salivary glands. Journal of Endocrinology. 180:55-62.

Vandenbergh J.G. 1999. Pheromones, mammals. In Knobil E., J.D. Neill, editors. Encyclopedia of reproduction, Vol 3. New York:Academic Press, pp 563-565.

Van Valkenburgh B., A. Curtis, J.X. Samuels, D. Bird, B. Fulkerson, J. Meachen- Samuels, G.J. Slater. 2011. Aquatic adaptations in the nose of carnivorans:evidence from the turbinates. Journal of Anatomy. 218:298-310.

Vargo E.L., M. Laurel. 1994. Studies on the mode of action of a queen primer pheromone of the fire ant Solenopsis invicta. Journal of Insect Physiology. 40(7):601-601.

Vogt R.G., L.M. Riddiford. 1981. Pheromone binding and inactivation by moth antennae. Nature. 293:161-163.

Vosseler J. 1924. Pflege und Haltung der Seekühe (Trichechus) nebst Beiträgen zu ihrer Biologie, Pallasia 2:216-221.

Wang Z., Nudelman A., Storm D.R. 2007. Are pheromones detected through the main olfactory epithelium? Molecular Neurobiology. 35:317-323.

Watkins W.A., D. Wartzok. 1985. Sensory biophysics of marine mammals. Marine Mammal Science. 1(3):219-260.

Wedekind C., T. Seebeck, F. Bettens, A.J. Paepke. 1995. MHC-dependent mate preferences in humans. Proceedings of the royal society of London. 60(1359):245- 249.

White R., S. Jobling, S.A. Hoare, J.P. Sumpter, M.G. Parker. 1994. Environmentally persistent alkylphenolic compounds are estrogenic..Endocrinology. 135:175-182.

Whitear M. 1992. Solitary chemosensory cells. In: T.J. Hara,editor. Chemoreception in Fishes, 2nd Ed. London:Elsevier Press, pp 103-125.

Williams L.E., J.M. Gliatto, R.K. Dodge, J.L. Johnson, R.M. Gamblin, D.H. Thamm, S.E. Lana, M. Szymkowski, A.S. Moore. 2003. Carcinoma of the apocrine glands of the anal sac in dogs:113 cases (1985-1995). Journal of American Veterinary Medical Association. 223(6):825-831.

Winston M.L., K.N. Slessor. 1992. The essence of royalty-honey-bee queen pheromone. American Scientist. 80(4):374-385.

154

Wishart D.S., D. Tzur, C. Knox, R. Eisner, A.C. Guo, N. Young, D. Cheng, K. Jewell, D. Arndt, S. Sawhney, C. Fung, L. Nikolai, M. Lewis, M. Coutouly, I. Forsythe, P. Tang, S. Shrivastava, K. Jeroncic, P. Stothard, G. Amegbey, D. Block, D.D. Hau, J. Wagner, J. Miniaci, M. Clements, M. Gebremedhin, N. Guo, Y. Zhang, G.E. Duggan, G.D. MacInnis, A.M. Weljie, R. Dowlatabadi, F. Bamforth, D. Clive, R. Greiner, L. Li, T. Marrie, B.D. Sykes, H.J. Vogel, L. Querengesser. 2007. HMDB: the human metabolome database. Nucleic Acids Research. 35:D521-D526.

Wyatt T.D. 2006. Pheromones and Animal Behavior: Communication by Taste and Smell. Cambridge University Press. Cambridge, United Kingdom.

Yamasaki F., H. Satom, T. Kamiya. 1976. The tongue of the Franciscana La-Plata dolphin Pontoporia blainvillei. Okajimas Folia Anatomica Japonica. 53:77-91.

Yamasaki F., S. Komatsu, T. Kamiya. 1980. A comparative morphological study on the tongues of manatee and dugong (Sirenia). The Scientific Reports of the Whales Research Institute. 32:127-144.

Yoshimura K., J. Shindoh, K. Kobayashi. 2002. Scanning electron microscopy study of the tongue and lingual papillae of the California sea lion (Zalophus californianus californianus). The Anatomical Record. 267:146-153.

Yoshimura K. N., J. Shindo, Y Miyawaki, K. Kobayashi, I Kageyama. 2007. Scanning electron microscopic study on the tongue and lingual papillae of the adult spotted seal, Phoca largha. Okajimas Folia Anatomica Japonica. 84(3):83-97.

Yoshimura K, N. Hama, J. Shindo, K. Kobayashi, I. Kageyama. 2008. Light and scanning electron microscopic study on the lingual papillae and their connective tissue of the Cape hyrax Procavia capensis. Journal of Anatomy. 213:573-582.

Yoshimura K., N. Hama, J. Shindo, K. Kobayashi, I. Kageyama. 2009. Light and scanning electron microscopic study on the tongue and lingual papillae of the common hippopotamus, Hippopotamus amphibius amphibius. Anatomical Record. 292(7):921-934.

Zimmer-Faust R.K., J.E. Tyre, J.F. Case. Chemical attraction causing aggregation in the spiny lobster, Panulirus interruptus and its probable ecological significance. Biological Bulletin.169(1):106-118.

Zufall F., and Leinders-Zufall T. 2007. Mammalian pheromone sensing. Current Opinion in Neurobiology. 17:483-489.

155

BIOGRAPHICAL SKETCH

Meghan Lee Bills was born in Oswego, New York in 1982 to parents Deborah

(Koester) Bills and Richard Lee Bills of Fair Haven, NY. Meghan was raised in Fair

Haven, NY and graduated from Red Creek Junior/Senior High School in 2001. She attended college at the University of Delaware during which time she completed a senior thesis analyzing spontaneous ovarian contractions in several fish species under the supervision of Dr. Malcolm Taylor. Meghan graduated with an Honors Bachelor of

Science degree with distinction in 2005. She went on to receive a Master of Science degree in marine biology from Nova Southeastern University in 2007. During her masters she focused on bottlenose dolphin distribution along the Southeast coast of the

United States under the supervision of Dr. Edward Keith. She then received her PhD in

2011 from the University of Florida Aquatic Animal Health Program of the College of

Veterinary Medicine under the supervision of Dr. Iskande Larkin and Dr. Don

Samuelson.

156