Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The functional significance and evolution of the coronary circulation in sharks Cox, Georgina Kimberly 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_february_cox_georgina.pdf [ 4.09MB ]
Metadata
JSON: 24-1.0223106.json
JSON-LD: 24-1.0223106-ld.json
RDF/XML (Pretty): 24-1.0223106-rdf.xml
RDF/JSON: 24-1.0223106-rdf.json
Turtle: 24-1.0223106-turtle.txt
N-Triples: 24-1.0223106-rdf-ntriples.txt
Original Record: 24-1.0223106-source.json
Full Text
24-1.0223106-fulltext.txt
Citation
24-1.0223106.ris

Full Text

THE FUNCTIONAL SIGNIFICANCE AND EVOLUTION OF THE CORONARY CIRCULATION IN SHARKS   by   Georgina Kimberly Cox    B.Sc. Hons. The University of New Brunswick, 2006 M.Sc. The University of British Columbia, 2010    A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE  REQUIREMENTS FOR THE DEGREE OF     DOCTOR OF PHILOSPHY   in   THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES   (Zoology)   THE UNIVERSITY OF BRITISH COLUMBIA   (Vancouver)   December 2015   © Georgina Kimberly Cox 2015    ii Abstract   The coronary circulation first appeared in the chordate lineage in cartilaginous fishes where it perfuses the entire myocardium, just like in birds and mammals but unlike in most teleost fishes. Yet, despite the pivotal position of elasmobranchs in the evolution of the coronary oxygen supply, the functional significance of their coronary circulation has never been investigated. Elasmobranchs are of special interest because of the morphological arrangement of their cardiomyocytes, which has resulted in the majority of the ventricular myocardium having access to oxygen from both a coronary supply and the venous blood returning to the heart. In order to determine the relative contribution of the coronary oxygen supply to cardiovascular function, I measured coronary blood flow (CBF) in the sandbar shark, Carcharhinus plumbeus, and leopard shark, Triakis semifasciata, while manipulating cardiovascular status using pharmacological approaches and in vivo temperature changes, respectively. By exploring inter- and intra-individual variation in cardiovascular variables I show that coronary blood flow is directly related to heart rate in both bradycardic (R2= 0.6, P <0.001) and trachycardic (R2= 0.8, P <0.001) states, as it is in mammalian hearts. I suggest that changes in heart rate are related to changes in coronary vascular resistance in elasmobranchs. As I found that 3-4% of cardiac output is directed towards the myocardium (~0.07% of body mass), a methodology was developed to quantify the degree of the coronary vascularization within the spongy and compact tissues of the shark myocardium.  iii Using this methodology, coronary vascularity and vessel morphology of six species of shark were compared to explore the functional relationships between coronary morphology and physiological phenotypes across species. I further determined that the coronary circulation showed evidence of having evolved in response to different thermal regimes across species. Together the results of this thesis show that the coronary circulation in sharks plays a crucial role in myocardial oxygen delivery to the entire heart, something not previously appreciated, and that habitat temperature plays a role as a selective agent in shaping coronary morphology.     iv Preface    A version of Chapter 2 was submitted as Cox GK, Brill R, Bonaro K and Farrell AP “Determinants of coronary blood flow in the sandbar shark, Carcharhinus plumbeus”. I designed and conducted all experiments, carried out all analyses and wrote the manuscript under the supervision of Dr. AP Farrell. Dr. R Brill provided laboratory space and experimental animals. Under my guidance K Bonaro, an undergraduate student, helped monitor shark vitals during the experiment. All authors provided editorial input to the manuscript.  A version of Chapter 3 will be submitted as Cox GK, Wegner NC, Lai NC and Farrell AP. Coronary blood flow in leopard sharks, Triakis semifasciata, during acute warming. I designed and conducted all experiments, carried out all analyses and wrote the manuscript under the supervision of Dr. AP Farrell. Dr. NC Wagner provided laboratory space and experimental animals. Dr. NC Lai assisted in several surgeries. All authors provided editorial input to the manuscript.  A version of Chapter 4 was submitted as Cox GK, Kennedy G. and Farrell AP. Morphological arrangement of the coronary vasculature in a shark (Squalus sucklei) and a teleost (Oncorhynchus mykiss). I designed the methodology, conducted the perfusions and tissue preparation, carried out all analyses and wrote the manuscript under the supervision of Dr. AP Farrell. Under my guidance G Kennedy, an undergraduate student, assisted in the staining and quantification of vessel dimensions in the trout. All authors provided editorial input to the manuscript.   v Chapter 5 included 6 shark species obtained while collaborating with Drs. JF Seffensen, N. C. Wegner and R Brill. I designed the methodology, conducted the perfusions and tissue preparation, carried out all analyses and wrote the manuscript under the supervision of Dr. AP Farrell. Dr. M Mandic provided expertise on the phylogenetic independent contrast analysis.  All experiments in this dissertation were approved by the UBC animal care Committee (A11-0355, A09-0008).     vi Table of Contents  Abstract ........................................................................................................................................ ii Preface ......................................................................................................................................... iv Table of Contents ..................................................................................................................... vi List of Tables ............................................................................................................................. ix List of Figures ............................................................................................................................. x List of Abbreviations ............................................................................................................. xv Acknowledgements ...............................................................................................................xvi Dedication ............................................................................................................................. xviii Chapter 1: General introduction ......................................................................................... 1 1.1 Oxygenation of the myocardium ............................................................................................ 1 1.1.1 The luminal circulation ....................................................................................................................... 1 1.1.2 The coronary circulation .................................................................................................................... 2 1.1.3 Evolutionary patterns of the coronary circulation ................................................................... 4 1.2 Optimization of myocardial oxygen delivery .................................................................... 6 1.2.1 Physiological modifications influencing oxygen convection ................................................. 6 1.2.2 Morphological modifications influencing oxygen diffusion ................................................... 7 1.3 Sharks as model organisms ................................................................................................... 10 1.4 Thesis objective and hypothesis ......................................................................................... 10 Chapter 2: Determinants of coronary blood flow in the sandbar shark, Carcharhinus plumbeus ....................................................................................................... 15 2.1 Summary ...................................................................................................................................... 15 2.2 Introduction ................................................................................................................................ 16 2.3 Methods: ....................................................................................................................................... 19 2.3.1 Animals................................................................................................................................................... 20 2.3.2 Surgical procedures ........................................................................................................................... 20 2.3.3 In vivo protocol: Intravascular administration of adrenergic and cholinergic drugs ............................................................................................................................................................................. 23 2.3.4 In vitro protocol: Coronary artery perfusion ........................................................................... 23 2.3.5 Calculations and statistics .............................................................................................................. 24 2.4 Results: ......................................................................................................................................... 27 2.4.1 Baseline status: In vivo cardiovascular variables .................................................................. 27 2.4.2 Effects of extravascular compression on coronary blood flow .......................................... 28 2.4.3 Pharmacological stimulation of cardiovascular status. ...................................................... 29 2.4.4 In vitro perfused hearts ................................................................................................................... 30  vii 2.5 Discussion: .................................................................................................................................. 31 2.5.1 Cardiovascular status at baseline and in response to drug injection .............................. 31 2.5.2 Determinants of coronary blood flow ......................................................................................... 33 2.5.3 Relative importance of the coronary and luminal oxygen supplies in sandbar shark hearts ................................................................................................................................................................. 37 2.5.4 Conclusions ........................................................................................................................................... 39 Chapter 3: Coronary blood flow in leopard sharks, Triakis semifasciata, during acute warming ........................................................................................................................ 52 3.1 Summary ...................................................................................................................................... 52 3.2 Introduction ................................................................................................................................ 53 3.3 Methods ........................................................................................................................................ 55 3.3.1 Animals................................................................................................................................................... 55 3.3.2 Surgical procedures ........................................................................................................................... 56 3.3.3 In vivo temperature challenge. ..................................................................................................... 58 3.3.4 In vitro coronary artery perfusion ............................................................................................... 59 3.3.5 Calculations and statistics .............................................................................................................. 60 3.4 Results........................................................................................................................................... 61 3.4.1 Temperature effects on baseline in vivo cardiovascular status ........................................ 61 3.4.2 In vitro coronary artery perfusion ............................................................................................... 62 3.5 Discussion .................................................................................................................................... 62 3.5.1 Cardiovascular status in response to increasing water temperature ............................. 62 3.5.2 Determinants of coronary blood flow ......................................................................................... 64 3.5.3 Conclusions ........................................................................................................................................... 65 Chapter 4: Morphological arrangement of the coronary vasculature in a shark (Squalus sucklei) and a teleost (Oncorhynchus mykiss) ........................................... 71 4.1 Summary ...................................................................................................................................... 71 4.2 Introduction ................................................................................................................................ 72 4.3 Methods ........................................................................................................................................ 75 4.3.1 Animals................................................................................................................................................... 75 4.3.2 Perfusion fixation of the coronary vasculature ....................................................................... 76 4.3.3 Tissue sectioning and staining ...................................................................................................... 77 4.3.4 Imaging and morphological measurements ............................................................................ 78 4.3.5 Statistics................................................................................................................................................. 79 4.4 Results........................................................................................................................................... 80 4.4.1 Interspecific difference between trout and dogfish sharks ................................................. 80 4.4.2 Intraspecific in differences in vascularization within the dogfish myocardium ......... 81 4.5 Discussion .................................................................................................................................... 82 4.5.1 Methodology ......................................................................................................................................... 82 4.5.2 Morphometrics .................................................................................................................................... 84 4.5.3 In vitro coronary perfusion rates in trout and dogfish ........................................................ 87 4.5.4 Conclusions ........................................................................................................................................... 88  viii Chapter 5: Linking the morphology of the coronary circulation with physiology and temperature across shark species ................................................. 101 5.1 Summary .................................................................................................................................... 101 5.2 Introduction .............................................................................................................................. 101 5.3 Methods ...................................................................................................................................... 103 5.3.1 Animals................................................................................................................................................ 103 5.3.2 Tissue perfusion, fixation, sectioning, staining, imaging and morphological measurements ............................................................................................................................................. 104 5.3.3 Calculations and Statistical analysis ........................................................................................ 104 5.4 Results......................................................................................................................................... 106 5.4.1 Intraspecific variation in morphological phenotypes ........................................................ 106 5.4.2 Interspecific variation in morphological phenotypes ........................................................ 107 5.4.3 Interspecific comparison between morphological and physiological phenotypes .. 108 5.4.4 Interspecific comparison between morphological phenotypes and temperature ... 110 5.5 Discussion .................................................................................................................................. 111 5.5.1 Myocardial oxygen supply in shark species ........................................................................... 111 5.5.2 Functional links between morphological and physiological phenotypes ................... 112 5.5.3 Morphological phenotypes and environmental temperature ......................................... 115 5.5.4 Conclusions ........................................................................................................................................ 117 Chapter 6: General discussion and conclusions........................................................ 126 6.1 Role of the coronary circulation in sharks .................................................................... 126 6.1.1 Coronary blood flow rates in relation to cardiac function ............................................... 127 6.1.2 Morphological phenotypes of the coronary circulation .................................................... 129 6.1.3 Balancing oxygen perfusion and diffusion in functioning hearts. ................................. 130 6.2 Evolution of the coronary circulation ............................................................................. 132 6.3 Future directions .................................................................................................................... 134 Bibliography .......................................................................................................................... 138     ix List of Tables   Table 2. 1: Variation in baseline cardiovascular variables for 10 anaesthetized sandbar sharks. ..................................................................................................................................... 37  Table 2. 2: Cardiovascular responses to acetylcholine and adrenaline in anaesthetized sandbar sharks (N= 10 except for CBF measurements where N=6).  Values are means ± sem. ................................................................................................................... 41  Table 3. 1: Mean (± SEM) in vitro coronary artery blood flow and resistance at given filling pressures .................................................................................................................................... 66  Table 4. 1: Average in vitro coronary perfusion rate and resistance at physiological input pressures in dogfish and rainbow trout. ......................................................................... 90  Table 4. 2: Average vascularity and vessel dimensions in the compact and spongy myocardial tissue of dogfish and rainbow trout. ..................................................................... 91  Table 5. 1: Average vascularity, vessel dimensions and  inter-vascular distances in 6 species of shark. ................................................................................................................................. 118  Table 6. 1: Metabolic rate, P50 and average inter-vascular distance in four species of shark. ...................................................................................................................................................... 137         x List of Figures  Figure 2. 1: Representative traces of cardiac output and heart rate in anaesthetized sandbar sharks during A baseline, B) following acetylcholine 10-6 M injection C) a type I response to an adrenaline 10-6 M injection and D) a type II response to adrenaline 10-6 M injection. Each trace is 1 min. ..................................................................... 43  Figure 2. 2: Representative traces of simultaneous recordings of baseline cardiac output, coronary flow, ventral aortic pressure and heart rate in an anaesthetized sandbar shark for 8 seconds. Dashed vertical lines represent the beginning and end of systole determined using ventral aortic pressure. ............................................................ 44  Figure 2. 3: Representative traces of simultaneous recordings of baseline cardiac output, coronary flow and ventral aortic pressure in an anaesthetized sandbar shark for one cardiac cycle. .......................................................................................................................... 45  Figure 2. 4: The relationships between mean cardiovascular variables with coronary flow for individual anaesthetized sandbar sharks (colours denote Fish ID). Relationships for total cardiac output (A), heart rate (B), stroke volume (C), ventral aortic pressure (D), cardiac power output (E). R2 and P values for linear regression appear at the top of each panel. Each data point represents an averaged (bars = s.e.m.) baseline value for one fish, and 2 additional values from periods where the animal had returned to a steady state for greater than 3 minutes following drug injections. ................................................................................................................................................ 46  Figure 2. 5: The relationship between coronary flow and coronary resistance (A), and heart rate and coronary resistance (B) between individual anesthetized sandbar sharks. R2 and P values for each relationship appear at the top of each graph. Each data point represent an individual fish, averaged (bars = s.e.m.) for the baseline value and 2 additional values from periods where the animal had returned to a steady state for greater than 3 minutes following drug injections. Different colours denote Fish ID (N=6). ......................................................................................................................... 47  Figure 2. 6: Relationship between coronary flow and total cardiac output (A), heart rate (B), stroke volume (C), ventral aortic pressure (D), and cardiac power output (E) in anesthetized sandbar sharks with a linear regression. N=6 with R2 and P values at the top of each graph. Different symbols denote different fish ID. Different colours denote different drug states with red indicating acetylcholine (bright red = 10-6 M concentration), green indicating adrenaline and blue indicating baseline and post saline injection values. The grey lines in panel b illustrate the association  xi between heart rate and CBF in sharks that altered heart rate by >10 bpm (N=5; one shark did not vary heart rate by 10 bpm). ................................................................................. 48  Figure 2. 7: The relationship between coronary flow and coronary resistance (A), heart rate and coronary resistance (B). R2 and P values for each relationship appear at the top of each graph. Different symbols denote different fish ID. Colours denote different drug states with red indicating acetylcholine, green indicating adrenaline and blue indicating baseline and post saline injection values. Data from post-acetylcholine 10-6 M injection was removed from this analysis as it was the only concentration to have significant effects on cardiovascular variables and has been shown to have a differential effect on dorsal and ventral aortic pressure, thus coronary resistance could not be estimated. ............................................................................ 49  Figure 2. 8: The relationship between coronary flow and coronary input pressure in the perfused sandbar shark heart (A) and the relationship between coronary flow and coronary artery resistance in vivo (; N= 6) and the perfused heart prep (; N=5) at different input pressures (B). Data for acetylcholine 10-6 M were not included. .................................................................................................................................................. 50  Figure 2. 9: The linear relationship between heart rate and mean coronary blood flow for sandbar sharks in comparison to the heart rates and left ventricular mean coronary blood flows of other vertebrates at normal body temperatures. Mammalian regression equations were taken from summarized data compiled by Duncher and Bache (Review, 2011; Fig. 4). ............................................................................... 51  Figure 3. 1:  Mean (± SEM) relative changes in coronary blood flow with temperature (A) and absolute changes in ventral aortic pressure (B) and heart rate (C) over a 10°C acute temperature increase in leopard sharks. The pink area indicates data collected at temperatures that are above the normal environmental temperature range for leopard sharks (>24 °C). An asterisk indicates a significant difference from 20°C. The numbers below the data points 20°C indicate the number of sharks represented for each cardiovascular variable until the pink zone when some experiments were terminated due to the appearance of cardiac arrhythmias 67  Figure 3. 2: Representative trace of an arrhythmic event in a leopard shark (size = 14 kg) at 28°C. ....................................................................................................................................... 68  Figure 3. 3: Representative traces of coronary blood flow, ventral aortic pressure and heart rate at 20°C and 30°C in a leopard shark (size = 12 kg). .................................. 69  xii  Figure 3. 4: The relative change in coronary blood flow with a temperature-mediated increase in heart rate in leopard sharks. The pink data points indicates data collected at temperatures that are above the normal environmental temperature range for leopard sharks (>24 °C). n= 6 unless otherwise indicated. .. 70  Figure 4. 1: Cross sections of perfused fixed blood vessels in myocardial tissue in the rainbow trout (A, B: O. mykiss) and Pacific dogfish (C, D, E, F: S. suckleyi) at 400x magnification. (A) compact ventricular myocardium of trout, (B) spongy ventricular trabeculae of trout, (C) compact ventricular myocardium of dogfish, (D) ventricular trabeculae of dogfish, (E) conal myocardium of dogfish, (F) atrial trabeculae of dogfish. The black scale bar is 50 µm. .......................................................................................... 92  Figure 4. 2: Corrosion cast and micro-CT images of the blood vessels in the dogfish heart.  A) A micro CT-image of a Microfill cast of the coronary circulation around the junction of the conus arteriosus (con) and ventricle (vent). B) A Mercox cast of a the microcirculation in a single trabecula of ventricular spongy myocardium. C) A micro-CT image of a cross-section through the ventricle and atrium (atr) of dogfish heart. D) A micro-CT image if a cross-section through the ventricle and atrium of a trout .......................................................................................................................................................... 93  Figure 4. 3: Average vessel area and vessel wall thickness or the myocardium of dogfish. Different symbols represent different individual sharks (n=4) and different colours denote different cardiac tissues. .................................................................................... 94  Figure 4. 4: Frequency distribution  of vessel area for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts. .................................................................. 95  Figure 4. 5: Frequency distribution of inter-vascular distance for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts. ........................... 96  Figure 4. 6: Frequency distribution of vessel diameter for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts. ................................................... 97  Figure 4. 7: A comparison of the vascularity of spongy myocardium of the atrium and ventricle of dogfish (N=4). Number of vessels is presented as a function of trabecular area in the ventricular spongy (R2 = 0.7; P <0.001) and atrial spongy (R2 = 0.3; P < 0.001) myocardium. Note there are a number of trabeculae with a small area that had no blood vessels. ....................................................................................................... 98   xiii Figure 4. 8: A comparison of the vascular density of spongy myocardium of the atrium and ventricle of dogfish (n=4).   Number of vessels is presented as a function of trabecular area. Note there are a number of trabeculae with a small area that had no blood vessels. .................................................................................................................................. 99  Figure 4. 9: Frequency distribution of inter-vascular distance to the edge atrial and ventricular trabeculae of dogfish (n=4). ................................................................................... 100  Figure 5. 1: Species specific mean inter-vascular distance, vessel diameter and vascular density in the conal compact (A-C), ventricular compact (D-F), ventricular spongy (G-I) and atrial spongy (J-L) heart tissue. Different symbols represent the means of the different myocardial tissue types; ☐ = conal compact,  = ventricular compact,  = ventricular spongy,  = atrial spongy. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1). Statistically significant differences between species (P ≤ 0.05) are denoted by different letters. Data is presented as mean ± s.e.m with specie arranged based increasing maximal environmental temperature from left to right. ..................................................................................................... 119  Figure 5. 2: The relationship between mean vascular density and mean inter-vascular distances (A, B) and between inter-vascular distance and vessel diameter in spongy (A, C) and compact myocardial tissues (B, D) for hearts of 6 species of elasmobranch. Data is presented as mean ± s.e.m. ............................................................... 120  Figure 5. 3:  The relationship between inter-vascular distances and edge distances in the spongy myocardial tissues for different shark species. Values for the ventricular  atrial  () tissue types are plotted for each individual and this relationship is represented by the grey dashed line. The relationship between species specific mean values () is represented by the solid line. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1) ...................................................................................... 121  Figure 5. 4: The relationships between mean edge distance (A) and mean inter-vascular distance (B) with area of the largest avascular trabeculae in either the atrial ventricular () spongy myocardial tissues across shark species. Solid line represents the relation with atrial myocardium. Dotted line represents relationship with ventricular myocardium. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1). ............................................................................................................................. 122  Figure 5. 5: Mean inter-vascular distances and vascular densities in four species of shark in relation to their respective mean ventral aortic blood pressures (A- D), cardiac outputs (E -H), metabolic rates (I -L) and arterial oxygen transport capacity  xiv (M-P) in spongy and compact myocardial tissue.  Measurements of mean ventral aortic pressure were taken from chapters 2 and 3 for the sandbar shark and leopard shark, respectively.  Estimated cardiac output calculated in chapter 2 for the sandbar shark was also used. Additional literature values for mean ventral aortic pressure (Johansen et al., 1966; Lai et al., 1997), cardiac output (Lai et al., 1990b; Lai et al., 1997; Scharold et al., 1988) and metabolic rate (Brett and Blackburn, 1978; Dowd et al., 2006; Scharold et al., 1988; Sepulveda et al., 2007) were included where available for the species. The metabolic rates used were measured between 16- 21°C in all sharks aside from the dogfish, that was measured at 10°C. Data is presented as mean ± s.e.m.............................................................................................................. 123  Figure 5. 6:  The relationship between P50, IVD and mean myocardial vascular density in the spongy (A,B) and compact (C,D) myocardial tissues of 6 species of elasmobranch. Linear regressions (dashed lines) and significant phylogenetically independent contrast (PIC) correlations (solid lines) are plotted on species data that are presented as mean ± s.e.m. P50 values were obtained from the literature for sandbar sharks (Brill et al., 2008), leopard sharks (Lai et al., 1990a) and dogfish (Lenfant and Johansen, 1966). Other values were attained through personal communication with Diego Bernal (mako and blue sharks) and Neill Herbert (Greenland shark). ............................................................................................................................. 124  Figure 5. 7: The relationship between mean myocardial phenotypes of the coronary circulation and maximal environmental temperature in the compact (A-C) and spongy (D-F) myocardial tissue types across 6 shark species.  Linear regressions (dashed lines) and significant  phylogenetically independent contrast (PIC) correlations (solid lines) are plotted on species data that are presented as mean ± s.e.m. Maximal temperatures were collected from active and passive acoustic tracking studies (Hight and Lowe, 2007) in addition to gill net and longline surveys (Merson and Jr, 2001) in addition to CTD scans of the oceanographic regions where sharks were captured (personal communication with NOAA and the survey crew of the Dana). .............................................................................................................................................. 125    xv List of Abbreviations   ACh  acetylcholine AD  adrenaline ANOVA analysis of variance °C  degree Celsius CBF  coronary blood flow CPO  cardiac power output ECG  electrocardiogram IVD  inter-vascular distance Mv  ventricular mass sem  standard error of the mean     xvi Acknowledgements   The completion of this thesis could not have been accomplished without the help and support of my mentors, friends and family.  First, I would like to express my gratitude towards my PhD and MSc Supervisor Dr. Tony Farrell - who has provided constructive feedback on all of my scientific endeavours for over ¼ of my life. Thank you for always pushing me to reach my potential every step of the way, even when I resisted. I am grateful for the patience, advice and wisdom you have imparted to me over the years and I am sure I am a better scientist and person for it.   I consider myself very fortunate to have had Bob Shadwick, Colin Brauner and Bill Milsom my PhD thesis committee. Not only do I admire you all as great scientists but your feedback, enthusiasm and encouragement to “think broadly” has been a great asset to my thesis and also my personal development. Another thank you goes out to Jeff Richards, whose words of support and encouragement went a long way in helping me overcome many obstacles during my thesis.   Over the years my fellow Farrellites have been a source of great support and encouragement. Thank you all! A special thanks to my “lab moms” Erika Eliason and Linda Hanson for their support and for being excellent sounding boards, even from across the content or across campus. Chris Wilson, Adam Goulding and Sabine Laguë were there with me for many years through the ups and downs of my PhD and provided help and support along the way for which I am grateful.   Within the Zoology department there are many friends external to my lab that played enormous roles in my personal and scientific life. To my COMPHY cohort - I do not believe that anyone could wish for a better group of people to go through grad school with. Those with me from the beginning of my MSc to the end of my PhD include the following. Milica Mandic whose friendship not only enriched my life scholastically but also personally. Her clear way of thinking and analytical abilities greatly added to my thesis. I could not have hoped for a more generous and caring person to celebrate the thesis highs with and battle through the thesis lows with. Thank you for always being there. Gigi Lau’s work ethic was always something to aspire to. Her patience and determination to teach me math eventually forced me into being better at converting units and her friendship through it all provided great comfort when it was needed most. Mike Sackville’s uncanny ability to come up with great ideas and discuss them in such an open and positive way made him an ideal academic brother to have. Ben Speers-Roesch’s enthusiasm for sharks was so contagious it, at times when I needed it, renewed my own enthusiasm. Matt Regan’s engaging way of approaching topics and theories made him a joy to work and chat with over the years.   xvii Those that joined me during my PhD included Till Harter, the ideal my officemate and friend who listened and supported me. Yvonne Dzal, the perfect combination of hard worker and drinker, your friendship in and outside of the lab was just awesome. Katelyn Tovey, thanks for the support and getting me to the gym. Taylor Gibbons, your positive attitude always increased general well being, and always made me feel much better. Emily Gallagher thanks for being unrelentingly positive, you will be missed greatly. Other special mentions go out to Rush Dillion, Carla Crossman, Rebecca Stephen and Stephanie Avery-Gomm for just being great and supportive.  Thank you friends for not only having an incredibly positive impact on my life but also enriching my experience at UBC beyond measure – and of course for being quality drinking partners when everything seemed to be going to “not a great place”.  Your friendships are some of the most valuable things I take from time at UBC.  Last, but not at all least, thank you to my family. My mum Jacqueline Cox’s encouragement and unwavering confidence were crucial to my success. My dad Roger Cox inspired me to study the natural world and his confidence in me, his feedback and willingness to read everything I asked him to were great assets to me throughout my degrees.  My brother Graham Cox’s technical support made it possible to actually write this thesis. And Littlefoot, as always, thank you for agreeing with all my ideas. Thank you Mniaro’s for your calming presence.   To my partner Seth Rudman, an exorbitant amount of thank you for the years of love and support  -not to mention free statistical consultations.     xviii Dedication          To my family, who is responsible for my love, respect, and curiosity about the natural world.   1 Chapter 1: General introduction   1.1 Oxygenation of the myocardium    All vertebrates have an obligate requirement for oxygen and are thus dependent upon a system that transports oxygen from the respiratory surfaces to the body tissues at a rate that meets metabolic demand. Central to this oxygen delivery system is the assiduously beating heart.  Although exposure to various agents of selection has resulted in vertebrate hearts differing in size, shape, tissue composition, power generating capacity and metabolic demands, they are all united by their common need for a continuous supply of energy to function. This energy is ultimately provided by oxidative phosphorylation. As a functioning heart is fundamental to survival in vertebrates, it becomes clear that the myocardial oxygen supply is of paramount importance to survival too.  In this regard, vertebrate hearts show a remarkable diversity in the form of myocardial oxygen delivery during development and as adults (Sedmera and Wang, 2012). Even so, this diversity can be simplified to two basic oxygen delivery routes: luminal and coronary.  1.1.1 The luminal circulation  The Luminal circulation refers to the venous blood returning to the heart, where it surrounds the endocardial lining of the myocardium. The rate at which  2 oxygen is delivered via this circulation is calculated as the product of cardiac output and venous oxygen content. During early development the luminal circulation in all vertebrates is the sole oxygen source because even if a coronary circulation develops, it does so later (Tomanek, 2012).   Hearts that continue to depend solely on the luminal oxygen supply route always have a spongy myocardium. The spongy myocardial architecture consists of a network of interconnecting and branching trabeculae that results in a high myocardial surface area in contact with venous return (fig. 1.1). Some teleosts, for example, appear to have exclusively avascular sheet-like trabeculae (Pieperhoff et al., 2009) that ensures a short diffusion distance for gas exchange and a large surface area that is in contact with the luminal blood (fig. 1.1).   However, not all spongy myocardia are entirely devoid of blood vessels. Vascularized trabeculae are reported in species belonging to the holocephalians, elasmobranchs, crocodilians and mammals (fig. 1.1) (Durán et al., 2015; Goo et al., 2009; Kohmoto et al., 1997; Tota, 1989).  Even so, the outer annulus of the cylindrical/elliptical shaped trabeculae (fig. 1.1) of mammals is avascular, apparently dependent on the luminal supply, while the interior of the trabeculae contains variable numbers of coronary vessels relating to individual trabeclum diameter (Goo et al., 2009).  1.1.2 The coronary circulation   The coronary circulation can be defined as a vascular network devoted to delivering oxygenated arterial blood to the myocardium of at least one cardiac  3 chamber within the pericardium. The rate at which oxygen is delivered via this circulation can be calculated as the product of coronary blood flow and arterial oxygen content. While all adult vertebrates retain some degree of spongy myocardium, albeit minimal in birds and mammals, many other species develop hearts with an outer circumference of more closely packed and highly organized muscle fibers that surround the interior spongy myocardium (Pieperhoff et al., 2009). This outer layer is referred to as the compact myocardium. As the compact layer of the myocardium becomes thicker, oxygen diffusion from the luminal supply cannot suffice for an oxygen supply. Thus, a coronary circulation is required to oxygenate the myocytes of compact myocardium that have become functionally isolated from the luminal oxygen supply as a result of too large a diffusion distance relative to oxygen demand. Consequently, all species with compact myocardial tissue in either the cardiac outflow tract (conus arteriosus) or ventricle are dependent on a coronary circulation to some degree (Axelsson and Farrell, 1993; Farrell et al., 2012).   Across vertebrate taxa species possessing compact myocardial morphology may or may not have an extensive spongy myocardium, which may or may not be vascularized by the coronary circulation (fig. 1.1). Dependence on the coronary circulation for maintaining cardiac function in these hearts appears dependent on various factors including the cardiac morphology, physiology and the extent of coronary vascularization (Axelsson and Farrell, 1993; Brady and Dubkin, 1964; Farrell et al., 2012; Gamperl et al., 1994a; Gamperl et al., 1995; Juhasz-Nagy et al., 1962; Kohmoto et al., 1997).  4  1.1.3 Evolutionary patterns of the coronary circulation  The first vertebrate lineage to develop coronary circulation was the cartilaginous fishes, Chondrichthyes. In elasmobranchs the coronary circulation permeates the conal and ventricular compact, which comprises anywhere from ~5% to 40% of the ventricular mass (Santer and Greer Walker, 1980; Tota, 1989) and further extends into the trabeculae of the spongy myocardium to an unknown degree (De Andrés et al., 1990; De Andrés et al., 1992; Muñoz-Chápuli et al., 1994; Tota, 1983). Although these primarily spongy hearts have vascularized trabeculae, the relative contribution of the coronary circulation to myocardial oxygen delivery in these ancient vertebrates is completely unknown.    Coronary function in the bony fishes has been reviewed previously (Axelsson, 1994; Davie and Farrell, 1991a; Franklin and Axelsson, 1994). Like the elasmobranchs, bony fish that have penetrating coronary vessels (~50% of species; Farrell et al, 2012) vary with regard to the extent of compaction (~16% to 74%; Santer and Greer Walker, 1980) and vascularization (Foxon, 1950; Grant and Regnier, 1926; Poupa and Lindström, 1983; Santer, 1985; Santer and Greer Walker, 1980; Tota, 1983). However, the majority of bony fish have avascular trabeculae (Farrell et al., 2012; Gamperl and Farrell, 2004; Pieperhoff et al., 2009). Possessing avascular trabeculae is also typical of non-crocodilian reptiles and amphibians who, like the cartilaginous and bony fishes, retain > 50% spongy myocardium in their hearts (Brady and Dubkin, 1964; Farrell et al., 2012). While a high degree of compaction is typically associated with athletic species, the possession of a  5 primarily spongy myocardium appears to convey a degree of independence from the coronary circulation.  For example, complete coronary ligation does not kill salmon, trout or rattlesnakes, (Farrell and Steffensen, 1987; Gamperl et al., 1994a; Hagensen et al., 2008; Steffensen and Farrell, 1998), whereas even partial blockage of the mammalian coronary circulation can be fatal (Berne, 1964; Duncker and Bache, 2008; Feigl, 1983; Rubio and Berne, 1975; Tomanek, 2012). The dependence of the mammalian heart on its coronary circulation was recognized over a century ago by Porter (1898) “the continued coordinated contractions of the mammalian heart are impossible in the absence of a supply of blood to the cardiac muscle”.  This is also likely true for birds because they too have very little spongy myocardium,  ~1% of myocardial mass(Faraci, 1986; Faraci and Fedde, 1986; Faraci et al., 1984; Faraci et al., 1985) .   It is clear that both the morphology and physiology of the coronary circulation, along with myocardial tissue composition, varies within and across vertebrate taxa. Despite this, we do not know if the variation that we see in extant taxa is the result of natural selection.  If it is in fact adaptive, the selective pressures that have given rise to this variation remain unknown. One way to begin to answer these questions is to determine how the morphology and physiology of the coronary circulation can be modified in order to optimize oxygen delivery in response to various environmental and physiological stressors.    6 1.2 Optimization of myocardial oxygen delivery   When oxygen demands vary, there are two primary ways in which the coronary circulation may accomplish the task of optimizing oxygen delivery to the working myocardium. Physiological changes can maximize blood flow and thus the rate of oxygen convection to the tissues. Alternatively, or in concert, morphological changes can minimize diffusion distance and maximize the rate of oxygen diffusion into the tissues.    1.2.1 Physiological modifications influencing oxygen convection   There are numerous physiological adaptations within the myocardial tissue that influence the rates of oxygen uptake, transport and usage. The majority of these adaptations, however, rely on the continued convection of blood through the coronary circulation. The rate of this convection is positively related to coronary blood pressure and inversely related to coronary resistance.  As mean arterial blood pressure is usually tightly controlled, changes in coronary blood flow are primarily regulated by varying vascular resistance, which is accomplished by modulating vascular smooth muscle tone (Duncker and Bache, 2008).  In accordance with Poiseuille’s law, which states that vessel radius is positively related to flow to the fourth power, small changes in vascular smooth muscle tone, and thus vessel diameter, have large impacts on coronary flow rate.   Though vessel dimensions appear to have the most significant effects on vascular resistance, increases in blood viscosity caused by increases in temperature  7 or haematocrit can also increase resistance and negatively impact flow rates (Pries et al., 1992).  However, if myocardial oxygen extraction across vertebrates is similar to that of mammals (i.e., routinely between 70 – 80; Restorff et al., 1977; Tomanek, 2012), increasing the oxygen carrying capacity of the blood through elevations in haematocrit may offset the negative impacts of elevating blood viscosity. The balance between carrying capacity and viscosity appears different across species and even within species in response to various environmental or physiological stressors (Brill et al., 2008; Snyder, 1971; Wells and Weber, 1991; Windberger et al., 2003). Regardless, the ability to rapidly alter coronary blood pressure or vascular resistance in response to stressors ensures that the coronary oxygen supply can be modified in concert with fluctuations in myocardial demand.   1.2.2 Morphological modifications influencing oxygen diffusion  Although the rate of convection is determined by the coordinated control of cardiac function and vascular smooth muscle tone, oxygenation of the tissues is ultimately limited by the spatial organization of supply vessels. In muscles, it is essential that oxygen supply vessels be arranged in a way that optimizes oxygen diffusion distances without infringing upon the cross-sectional area required by muscle tissue to generate adequate force. Additionally, the rate of oxygen convection through a tissue must be balanced such that the rate of oxygen delivery matches metabolic demand, while ensuring adequate capillary residency time for oxygen to diffuse from the blood into the tissue.   8  A major factor influencing the rate of oxygen diffusion from the capillaries to the tissue is oxygen diffusion distance. Assuming vascular diameter remains constant, increases in capillary density are likely associated with reduced diffusion distances, and thus increases in oxygen delivery rates. Previous studies have linked acclimation temperature, exercise performance and hypoxia tolerance with variations in capillary density across several vertebrate taxa.   Temperature-induced increase in whole animal metabolic rate can trigger concurrent changes in cardiac function and thus myocardial oxygen demand in ectothermic vertebrates. The 40% increase in myocardial capillary density observed in rainbow trout (Oncorhynchus mykiss) acclimated to 11°C and 18°C (Egginton and Cordiner, 1997) was likely a response to a temperature-induced increase in myocardial oxygen demand. Environmental temperature is less likely to induce changes in muscle capillarity in birds and mammals as most maintain a constant body temperature.   Exercise training, however, does increase myocardial oxygen demand in endotherms. Interestingly, the mammalian response to an exercise-induced increase in myocardial oxygen demand varies slightly from that of temperature-induced increases in teleost fish. The morphological changes to exercise training in mammals results in an initial increase in myocardial capillary density, however, this is only temporary and capillary density returns back to routine levels after several weeks of training (White et al., 1998). Capillary diameter, however, increases significantly over weeks of exercise training allowing for an increased rate of blood flow (White et al., 1998).   9  Hypoxia has also been shown to induce increased morphological changes to capillary beds in birds and mammals. For example, increases in muscle capillarity have been linked to acclimation and adaptation to high altitude in both birds and mammals (León-Velarde et al., 1993; Mathieu-Costello et al., 1998; Oelz et al., 1986). This hypoxia induced increase in myocardial capillarity, however, is not a response to an increased tissue oxygen demand, as it is in temperature-acclimated fish, or exercising mammals. It is more likely that increasing capillary density during hypoxia offsets the decreased oxygen diffusion rate caused by the reduction in the oxygen partial pressure gradient in hypoxia.    While capillarity and vessel morphology within species is plastic, these variables respond comparatively slower than physiological changes such as increasing blood flow. Of course, changes in capillarity and coronary blood flow are not mutually exclusive. Morphological changes cannot only affect oxygen diffusion rates but also reset the upper and lower limits of various physiological variables, such as oxygen convection rates. For example, the increase in capillary diameter in response to exercise training in mammals allows for an increased coronary blood flow (Duncker and Bache, 2008; White et al., 1998). Whether these specific physiological and morphological plastic responses mimic the more long-term changes in heritable genetic variation that begets adaptation has yet to be determined.     10 1.3 Sharks as model organisms    Sharks are an ideal model organism in which to investigate the relationship between various coronary morphologies and environmental variation.  While all sharks have a coronary circulation, there is substantial variation in the morphology of the coronary artery (De Andrés et al., 1990; De Andrés et al., 1992; Muñoz-Chápuli et al., 1994).  Unlike their bony relatives (the teleost fish), the coronary circulation of sharks perfuses both the spongy and compact tissues, making sharks potentially more dependent on the coronary circulation for myocardial oxygen than their bony relatives. Additionally, sharks have adapted to a staggeringly wide range of habitats that vary in many environmental variables, such as temperature. Sharks also show substantial variation in athleticism and metabolic rates (Bernal et al., 2003a; Bernal et al., 2003b; Carrier et al., 2012; Hamlett, 1999).  Furthermore, two clades of shark have developed endothermy (Bernal et al., 2001; Block and Finnerty, 1994), creating an additional axis of variation to investigate. Given all this variation, sharks represent an ideal model system to discover relationships between cardiac phenotypes and environmental and physiological stressors.    1.4 Thesis objective and hypothesis  The primary goals of this thesis were to determine the relative contribution of the coronary circulation to the myocardial oxygen supply in sharks and characterize the physiological responses of the coronary circulation to variations in cardiovascular function. To do this I had to develop a methodology that allowed me  11 to quantify the variation in the morphology of coronary circulations among disparate shark species. Finally, I investigated whether variation in morphological characteristics of the coronary circulation are linked to environmental and physiological stressors that would suggest a pattern of adaptation.  Specifically, chapter 2 aims to test the hypothesis that coronary blood flow in the sandbar shark (Carcharhinus plumbeus) is positively related to myocardial oxygen demand and also determine the relative contribution of the coronary circulation to myocardial oxygen delivery. This was accomplished by simultaneously measuring coronary blood flow, ventral aortic pressure and partial cardiac output while pharmacologically varying cardiovascular function in anesthetized sharks. This chapter also examined how coronary blood flow rates responded to variation in coronary input pressures in spontaneously beating excised shark hearts in order to further assess the determinants of coronary blood flow in sharks.  Following from the results of chapter 2, chapter 3 sought to determine whether the relationships between coronary blood flow and various cardiovascular variables were maintained in un-anesthetized Leopard sharks (Triakis semifasciata) exposed to an acute to temperature increase. The goal of this chapter was to characterize the physiological responses of the coronary circulation to an environmental temperature stress. This chapter tested the hypothesis that temperature-induced increases in heart rate and cardiac oxygen demand would be met by concurrent increases in coronary blood flow and was the first study to measure coronary blood flow during a temperature challenge in any fish.    12  Because the results from chapters 2 and 3 revealed that the shark myocardium received substantial blood flow via the coronary circulation, the goal of chapter 4 was to investigate the degree of myocardial vascularization in the spongy myocardium. The specific objectives of chapter 4 were to develop a reliable methodology for the quantification of myocardial vessels within compact tissue of the conus arteriosus and ventricle, as well as within the spongy myocardial tissues of the ventricle and atrium. I used this methodology to compare cardiac vascularity of a representative shark, the Pacific spiny dogfish shark (Squalus sucklei) and a teleost, the rainbow trout (Oncorhynchus mykiss).  I tested the hypothesis that the rainbow trout would have a higher myocardial vascular density than the dogfish, as the trout heart does more work by producing higher arterial blood pressures and faster heart rates then the dogfish.  Within the shark heart I tested the hypothesis that the ventricular compact myocardium would have a higher vascular density than that of the spongy myocardium, due to the potential for the luminal circulation to aid in oxygen delivery to the trabeculae.   In chapter 4, the methodology developed in chapter 3 was used to examine the morphology of the coronary circulation in the hearts of 5 additional species of sharks including the sandbar shark (C. plumbeus), leopard shark (T. semifasciata), Greenland shark (Somniosus microcephalus), blue shark (Prionace glauca) and mako shark (Isurus oxyrinchus). The first objective of this chapter was to determine if there were specific morphological phenotypes associated with various morphological or physiological phenotypes across taxa.  Specifically, I aimed to investigate whether species with higher metabolic demands increased the rate of  13 oxygen diffusion or perfusion to the myocardium, using measurements of vascular density, inter-vascular distances and vessel diameter.  The second objective was to explore the relationship between morphological phenotypes of the coronary circulation and temperature, which may be a key driver of cardiovascular evolution due to its relationship with metabolic rate (Gillooly et al., 2001). Specifically, I tested whether shark species that are exposed to higher maximal environmental temperatures have specific morphological characteristics that would indicate an increased capacity for myocardial oxygen delivery.   The general outcome of this thesis was a) the first functional understanding of the significance of the coronary artery to cardiovascular function in sharks, b) the development of a methodology that allows for the quantification of the complex morphology of coronary circulation in the myocardium of fish and c) insight into how selective pressures may have produced various coronary morphological phenotypes across sharks. In doing so this thesis not only contributes a new understanding of coronary function in a lineage of ancient vertebrates, but also gives novel insight into the evolution of the coronary circulation.     14   Figure 1.1 Graphic representations of the current understanding of the organization of coronary circulation relative to the cardiac muscle types (outer compact and inner spongy myocardium) for a hagfish heart (A), representative teleost heart with a vascularized compact myocardium (B), a representative shark heart with vascularized compact and spongy myocardium (C) and a representative mammalian heart with a vascularized compact and spongy myocardium (D). Panel E shows the trabecular arrangements of a sockeye salmon (Oncorhyncus nerka) modified from (Pieperhoff et al., 2009). Panel F shows the trabecular arrangements of a dogfish shark (Squalus sucklei). The hagfish heart spongy myocardium (A) was modified from (Grimes et al., 2010) and the interior of spongy the myocardium of B was modified from (Pieperhoff et al., 2009).    15 Chapter 2: Determinants of coronary blood flow in the sandbar shark, Carcharhinus plumbeus1   2.1 Summary  Although sharks and teleosts have similar proportions of their hearts comprised of compact and spongy myocardial tissue, the coronary circulation of sharks perfuses a higher proportion of the heart when compared to teleosts. This study investigates the relative contribution of the coronary circulation to myocardial oxygen delivery by simultaneously measuring coronary blood flow (CBF), ventral aortic blood pressure and cardiac output while manipulating cardiovascular function using pharmacological approaches. By exploring inter- and intra-individual variation in cardiovascular variables, I discovered only a weak relationship between CBF and cardiac power output (R2= 0.1; P =0.01). Interestingly, I discovered CBF had a significant and direct relationship with heart rate (R2= 0.6; P <0.001) that was likely modulated by changes in coronary vascular resistance. With 3-4% of cardiac output being directed through the coronary circulation, my results indicate that the coronary circulation plays a major role in myocardial oxygen delivery both routinely and as heart rate increases.                                                                 1 A version of this chapter has been submitted as Cox GK, Brill R, Bonaro K and Farrell AP Determinants of coronary blood flow in the sandbar shark, Carcharhinus plumbeus.  16 2.2 Introduction  Coronary artery morphology and physiology have been well studied in mammals and the overall consensus is that even brief perturbations in coronary blood flow can have lethal consequences (Berne, 1964; Duncker and Bache, 2008; Tomanek, 2012). This is due to both the highly aerobic nature of the heart, which meets 99% of routine ATP needs through oxidative phosphorylation, and the high routine myocardial oxygen extraction (70-93%) (Tomanek, 2012). As a result, any increase in myocardial oxygen demand is met primarily (80%) through concurrent increases in coronary blood flow (CBF) (Restorff et al., 1977; Tomanek, 2012). In mammals, most increases in myocardial oxygen demand, and thus CBF, are typically associated with increases in heart rate (60%), contractility (20%), and left ventricular work (20%)(Duncker and Bache, 2008). Factors affecting CBF in other vertebrates, such as bony fish species, have been sparsely investigated using in vivo, in vitro and in situ perfused heart experiments and have been summarized previously (Axelsson, 1994; Davie and Farrell, 1991a; Franklin and Axelsson, 1994). In bony fishes, cardiac oxygen use and CBF have been tightly linked with cardiac power output (CPO; the product of cardiac output and mean aortic blood pressure) (Driedzic et al., 1983; Gamperl et al., 1995; Graham and Farrell, 1990). However, to the best of my knowledge, there are no studies on the factors effecting CBF in any cartilaginous fishes, which are the most ancient lineage of vertebrates to have evolved a coronary circulation (Durán et al., 2015; Santer, 1985; Tota, 1989). Indeed, only one in vivo study of coronary blood  17 flow has ever been reported in an elasmobranch (two anesthetised school sharks, Galeorhinus australis), and the results from this study suggested that there are similarities in cardiac and coronary artery function between sharks and mammals (Davie and Franklin, 1993).  Elasmobranch fishes are of special interest because while all extant species have a coronary circulation, similar to that of birds and mammals, the cardiac muscle is not arranged predominantly as compact myocardium. Most vertebrate taxa, including elasmobranchs, some teleost fishes, reptiles, birds, and mammals have ventricles comprised of both compact and spongy myocardium (fig.1.1). Compact myocardium is composed of closely packed cardiomyocytes that form the outer circumference of the heart to a variable thickness depending on the species (Farrell et al., 2012; Santer, 1985). While mammalian and avian hearts are 99% compact myocardium (Farrell et al., 2012; Santer, 1985), the compact myocardium of elasmobranchs and teleosts typically comprises <50% of the ventricular mass (Mv) and thus are mainly composed of spongy myocardium (Santer, 1985).  In contrast to the compact myocardial tissue, the cardiomyocytes of the spongy myocardium are arranged in a highly branched network of interconnecting beams called trabeculae (fig. 1.1). This cardiac morphology ensures a high surface area to muscle volume ratio. Thus, the mainly spongy ventricular morphology also allows the majority of the shark and teleost hearts to potentially obtain oxygen from the venous blood passing through the heart, its luminal myocardial oxygen supply. Indeed, the ancestral condition for the chordate heart is that this luminal oxygen supply (the product of cardiac output and venous oxygen content) is the only  18 oxygen source for the heart (Farrell et al., 2012; Tomanek, 2012). In fact, a large proportion of vertebrates (perhaps half of the extant teleost species and most amphibian species), as well as all extant cyclostome fishes have avascular hearts that depend solely on this luminal supply (Farrell et al., 2012). However, the trabeculae of elasmobranchs are reported to have a coronary supply (Durán et al., 2015; Tota, 1989). Regardless of whether a heart has a coronary oxygen supply, luminal oxygen supply, or both, it is essential that myocardial oxygen supply and demand be matched. As mammals increase cardiac output, and thus myocardial oxygen demand, during exercise primarily by increasing heart rate, a positive linear relationship is found between heart rate and CBF (Duncker and Bache, 2008; Tomanek, 2012). In comparison, CBF in bony fish, mainly salmonids, has a linear relationship with CPO (Gamperl et al., 1995). The relationship between heart rate and CBF observed in mammals may not be as prevalent in teleosts and elasmobranchs as they have the ability to modulate cardiac output by changing stroke volume by a large amount as well as changing heart rate (Farrell, 1991). This may be beneficial for their mainly spongy hearts because modulation of stroke volume allows fish to maintain lower heart rates during exercise, increases myocardial stretch, and prolongs luminal blood residency time, thus improving myocardial access to the oxygen remaining in the venous blood. I hypothesized that the same relationships between CPO and CBF observed in teleost fishes exists in the sandbar shark (Carcharhinus plumbeus). What remains unclear is the relative contributions of the coronary versus luminal oxygen supply  19 routes in hearts of elasmobranchs, where the heart is comprised mainly of spongy myocardium (85%), yet the trabeculae receive a coronary blood supply. The aim of my study was to quantify CBF in an elasmobranch over a range of heart rates and CPOs to gain insight into the determinants of the CBF and the relative contribution of the coronary circulation to the myocardial oxygen supply. Because I was able to measure CPO and CBF directly and simultaneously in sandbar sharks, it was possible to make the first estimate of the coronary oxygen supply in an elasmobranch. The sandbar shark, Carcharhinus plumbeus is well suited to measure CBF directly with a flow probe because it is one of the few shark species in which the coronary arteries converge into a single main coronary artery before penetrating the pericardium and then dividing into multiple coronary arteries on the surface of the cardiac outflow tract, the muscular conus arteriosus.  2.3 Methods:   All animal capture, housing, and experimental procedures were approved by the College of William and Mary and the University of British Columbia Animal Care and Use Committees (protocol numbers: IACUC-2014-05-13-9549-rwbril and A11-0355, respectively), and followed all applicable laws and regulations.  Individuals that were just large enough to accommodate the smallest Transonic flow probe on the single coronary artery were used. To transmit sufficient signal to the flowmeter, this miniature flow probe has a very short lead, which therefore limited all measurements to anaesthetized fish. Pharmacological  20 interventions were used to manipulate cardiovascular status of the anaesthetized fish.   2.3.1 Animals Experiments were conducted at the Virginia Institute of Marine Science Eastern Shore Laboratory (Wachapreague, VA, USA). Sandbar sharks (N=10) of both sexes (65-85 cm; 1.60 ± 0.41 kg; range 1.8-3.0 kg) were captured from the tidal lagoon system surrounding this facility using hook-and-line fishing gear, transported to the laboratory, and held in a shore-side circular tank (~8 m in diameter and 2 m deep) supplied with flow-through seawater from the adjacent lagoon (temperature 25-27°C, salinity 34-36‰). Animals were held at a natural photoperiod although the tank was shaded with black mesh for protection from direct sunlight. Fish were fed three times a week with cut pieces of Atlantic menhaden (Brevoortia tyrannus), but not for at least 24 h prior to experimentation.   2.3.2 Surgical procedures Individual sharks were anaesthetized with an injection of ketamine HCl (100 mg ml-1, Henry Schein, Dublin, OH, USA) into the caudal vein (dose rate of 0.35 ml kg-1), a dose that maintained anaesthesia throughout the entire experiment. The surgical plane of anaesthesia was verified when sharks failed to respond to a tactile stimulus (tail pinch), after which they were moved into the laboratory and placed ventral side up in a custom designed V-board. They were covered with wet towels  21 and their gills were irrigated with aerated seawater. A 3-4 cm midline incision was made anterior to the coracoid bar through the skin and coracomandibularis muscle, and the ventral aorta was carefully exposed with blunt dissection through the connective tissue of anterior jaw muscles. Due to the anatomical constraint common to all elasmobranchs (an immediate bifurcation of the ventral aorta into the 4th and 5th afferent gill arteries directly anterior to the pericardium) and position of the single coronary artery (proximal to the 3rd afferent gill arteries), total cardiac output could not be measured directly in sandbar sharks without disrupting the pericardium, the coronary arteries and the conal myocardium. Instead, an ultrasonic transit-time perivascular blood flow probe (2.5 mm PSB; Transonic Systems Inc., Ithaca, NY, USA) was positioned around the ventral aorta anterior to the 3rd afferent gill artery to estimate cardiac output by measuring blood flow to the anterior gill arches (Taylor et al., 1977). Ventral aortic pressure was measured via a catheter fashioned from polyethylene tubing (PE-60; IntramedicTM, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) that was inserted into the ventral aorta between the two most anterior afferent gill arteries. Retrograde perfusion of the major coronary arteries on the conus arteriosus with blue dye revealed that the coronary arteries merged into one coronary vessel prior to penetrating the pericardium ventral to the branch point of the 3rd afferent gill arteries. After careful dissection, a second perivascular flow probe (0.5 mm PSB; Transonic Systems, Ithaca, NY, USA) was placed around the single coronary artery to measure CBF (N=6). Variation in coronary morphology meant that CBF was measured in only one branch of the hypobranchial artery near the convergence of the hypobranchial arteries in two  22 other sharks and is reported as such, or it was not possible to implant the coronary flow probe (N=2).  Following surgery, the shark was moved to the experimental tank. At this time, the catheter was connected to a pressure transducer (model DPT-6100, VBM Medizintechnik, Kirchseeon, Germany) that had been calibrated against a static column of seawater and referenced to the water level in the experimental tank. A 4-channel bridge amplifier (Somedic, Hörby, Sweden) amplified signals from the blood pressure transducer. The flow probes were connected to two TS420 Perivascular Flow Modules (Transonic Systems, Ithaca, NY, USA). Cardiovascular variables were monitored on-line and stored for subsequent analysis using a Power Lab Unit (ADInstruments, Castle Hill, Australia) connected to a laptop computer running LabChart Pro software (v.7.3.7; ADInstruments). The shark was held ventral side up on the v-board at a water level that covered the gills. Irrigation of the gills with oxygenated seawater was maintained throughout the experiment. At the end of the in vivo study, the shark was euthanized with an overdose of sodium pentobarbital (~350 mg kg-1 injected IV, Schering-Plough Animal Health, Union, NJ, USA). The heart was then excised, rinsed clean of blood, and placed in an elasmobranch physiological saline solution (concentrations in mM: 360 Urea, 280 NaCl, 90 TMAO, 12 KCl, 10 CaCl2·2H20, 5 MgCl·6H20, 5 glucose, 4.5 NaHCO3, 0.5 NaH2PO4, pH 7.8) for in vitro coronary artery perfusion (see below).    23 2.3.3 In vivo protocol: Intravascular administration of adrenergic and cholinergic drugs After recording baseline cardiovascular variables for 1 h, acetylcholine (ACh) and adrenaline (AD) were used to induce reversible changes in heart rate in order to measure the effects on CBF. Drugs were slowly infused via the indwelling ventral aorta cannula (i.e., pre-gill) in a concentrated form with the goal of achieving final blood concentrations of 1x10-8 M, 1x10-7 M and 1x10-6 M (assuming a 5% blood volume) using a 0.25 ml kg-1 carrier volume of heparinized saline. Following the injections of ACh, heart rate was allowed to return to steady state before injecting the next concentration (< 5min for 1x10-8 M and 1x10-7 M ACh, but up to 1 h for 1x10-6 M ACh). Injection of the AD concentrations followed at 5 to 10 min apart. Peak drug responses were based on heart rate (fig. 2.1).  2.3.4 In vitro protocol: Coronary artery perfusion The excised perfused heart preparation followed that used previously for salmon (Farrell and Graham, 1986). The largest coronary artery on the conus arteriosus was cannulated with polyethylene tubing (PE-60) that was tipped with smaller tubing (PE-10), and secured to the ventral aorta with 4-0 braided silk suture thread. Initial perfusion with heparinized saline at a pressure of 40 cm H2O (1 cm H2O = 0.1 kPa) rapidly cleared the coronary vessels of blood. Collateral arterial connections on the conus arteriosus ensured perfusion of the majority of the conal coronary circulation downstream of the cannulation site. Once cleared of blood, the other main coronary branches were occluded upstream of the cannulation site to  24 prevent backflow from any severed coronary arteries on the conus arteriosus. Large diameter PE tubing was also secured in the severed ventral aorta and in the sinus venosus close to its junction with the atrium. The ventral aortic cannula was filled with saline and raised to a static pressure head of 30 cm H2O to distend the conus arteriosus. The atrial cannula was similarly set to a stable pressure head of 3 cm H2O to distend the cardiac chambers. Preparation time was 20-25 min. Coronary flow was measured as the time required for 0.4 ml of saline to enter the coronary circulation. Flow rate was then re-measured as a function of stepwise increases in coronary artery perfusion pressure (40, 50, 60, and 70 cm H2O).   2.3.5 Calculations and statistics Heart rate was determined from the ventral aorta blood flow trace as the number of systolic peaks per minute. Blood flow through the ventral aorta anterior to the 3rd afferent gill arteries was converted to cardiac output by assuming equal vascular resistance and blood flow to each gill arch [i.e., cardiac output = measured blood flow ∙ 0.4-1].  Corrections for cardiac output due to probe placement in elasmobranchs has been discussed previously (Lai et al., 1989; Taylor et al., 1977). Stoke volume was calculated as the estimated cardiac output  ∙  heart rate-1. Mean ventral aortic pressure was calculated using LabChart Pro software (v.7.3.7; ADInstruments) software that averaged all blood pressure values (100/s sampling rate) between systole and diastole from the blood pressure recording and was used in the calculation of CPO. Estimates of cardiac power output (in mW) were calculated from the estimated cardiac output ∙ measured mean ventral aortic  25 pressure ∙ 0.098 (Davie and Franklin, 1992; Farrell et al., 1992). Dorsal aortic pressure was estimated as 0.7 ∙ mean ventral aortic pressure to account for the loss in blood pressure across the branchial circulation and was based on previous simultaneous measurements of dorsal and ventral aortic blood pressure in epaulette sharks (Hemiscyllium ocellatum) and spotted catsharks (Scyliorhinus canicula) (Short et al., 1979; Stensløkken et al., 2004; Taylor et al., 1977). Dorsal aortic blood pressure was used in the calculation of coronary vascular resistance. Coronary vascular resistance and systemic vascular resistance were calculated from estimated dorsal aortic blood pressure ∙ CBF-1 and measured ventral aortic pressure ∙  ?̇?𝑏-1 (respectively) and are reported as cm H2O min g Mv min-1.  However, coronary resistance was not calculated after ACh (1x10-6 M; the only concentrations to elicit changes cardiovascular variables) and AD injections because of the known effects of these drugs on branchial vascular resistance (Bushnell et al., 1992; Short et al., 1979; Stensløkken et al., 2004; Taylor et al., 1977), thereby altering the assumed pressure drop across the branchial circulation. In vitro coronary resistance was calculated by dividing coronary input pressure by the flow rate of saline through the cannula. The %CBF was calculated from the measured CBF divided by the estimated cardiac output. The beginning and end of systole was determined using the ventral aortic pressure channel recording (fig. 2.2). Ventricular systole is defined in the mammalian literature as the portion of the cardiac cycle when the ventricles are contracting. The systolic period can be measured accurately from intraventricular pressures and ECG traces, which measure the electrical activity of contraction and relaxation. In the absence of ECG traces or intraventricular pressure measurements,  26 aortic blood flow or pressure measurements can also be used to estimate the systolic time period, but this results in a slight underestimation as the short period of isovolumetric ventricular contraction (and decreased ejaculation) is omitted. Further, an overestimate is introduced in determining the end of systole as ventricular diastole begins prior to the cessation of blood flow and before the dichrotic notch in the aortic blood pressure. This is due to the time it takes for the ventricular pressure to fall below that in the aorta, thus triggering valve closure. Sharks, however, differ from mammals in that their outflow tract (conus arteriosus) is surrounded by cardiac muscle, which contracts after the ventricle starts contracting. The effect of this conal contraction is prolonged blood flow through the ventral aorta during ventricular diastole. A major anatomical difference in this contractile outflow tract is the many tiers of flap valves (at least 3;  Satchell and Johnes, 1967) present within the shark’s conus arteriosus. The effect of conal contraction is to postpone the closure of the upper tiers of valves until the intrapericardial pressure, generated by ventricular ejection, decays to a lower value. Given that my whole experimental approach was predicated by not disrupting the pericardial wall, I used ventral aortic pressure to estimate ventricular systole. As the conus arteriosus is unable to generate equivalent pressure as the ventricle, peak ventral aortic pressure was used the marker for the end of systole/beginning of diastole in this study. As increases in pressure and blood flow in the ventral aorta appeared to occur in concert (fig. 2.2), the ventral aortic pressure trace was also used to mark the end of diastole/beginning of systole. Therefore my approach likely underestimated the systolic time period. Reported values are mean values (± s.e.m.)  27 for 10 sharks unless stated otherwise, and six sharks for CBF measurements. Statistically significant differences (𝑃 ≤ 0.05) among the control (baseline), drug responses and recovery values were tested using a one-way, repeated measures ANOVA followed by a Holm-Sidak post-hoc test in SigmaStat (v10.0.1, Aspire Software International, Ashburn, VA, USA). The relationships between cardiovascular variables were tested by regression in SigmaStat, which was also used to produce the curves.  2.4 Results:  2.4.1 Baseline status: In vivo cardiovascular variables  Mean ventricular mass (Mv) was 0.070% ± 0.003 of shark body mass.  The baseline cardiovascular status for the 10 anaesthetized sandbar sharks is reported in Table 2.1.  Mean cardiac output was 17.5 ml min-1 kg-1 with a heart rate of 42 bpm and a stroke volume of 0.44 ml beat-1 kg-1.  Mean ventral aortic pressure averaged 55.8 cm H2O, which reflected a systemic resistance of 3.5 cm H2O min kg min-1. As a result, CPO averaged 2.2 mW g-1 Mv.  Under these conditions, CBF averaged 0.78 ml min-1 g-1 Mv, or 3.0% of cardiac output, with an estimated coronary resistance of 58.8 cm H2O min g Mv min-1.  The individual variation in cardiovascular status was used to explore associations with CBF. A linear relationship between heart rate and CBF explained 60% of the variability (R2 = 0.6) and approached statistical significance (P= 0.06) (fig. 2.4B). Notably, %CBF was significantly higher in individuals with higher  28 baseline heart rates (R2=0.3, P<0.001). However, individual variability in CBF was not significantly related to cardiac output, stroke volume, ventral aortic pressure or CPO (fig. 2.4). Figure 2.5 shows that the 3.5-fold variation in individual CBF values was primarily determined by coronary resistance (R2 = 0.90; P =0.01), which in turn was inversely related to heart rate (R2 = 0.76; P =0.02).  2.4.2 Effects of extravascular compression on coronary blood flow Blood flow in the ventral aorta and coronary artery were both continuous during the cardiac cycle of anaesthetised sandbar sharks. The phasic relationships between baseline cardiac output, CBF, ventral aortic pressure and heart rate are illustrated in figure 2.2. It appears that extravascular compression first starts to reduce coronary blood flow during isovolumetric contraction (fig. 2.2, 2.3) with CBF showing a nadir during early systole. CBF then starts to increase early in ventricular diastole, which is followed by a reduction in cardiac output. Conal contraction occurs early during ventricular diastole and this is associated with CBF momentarily leveling off or even slightly decreasing (figs. 2,2 and 2.3). The conal valves close with conal contraction, cardiac output falls to zero and peak CBF occurs at this time (fig. 2.2). As a result, 77% of CBF occurred during ventricular diastole under baseline conditions. The percentage of the cardiac cycle spent in systole ranged from 18 to 38% across all heart rates including pharmacological changes described below. As expected, the time spent in systole significantly increased with heart rate (R2=0.5, P<0.001) and yet CBF was related to heart rate and coronary resistance was inversely related to heart rate.  29   2.4.3 Pharmacological stimulation of cardiovascular status. A sham saline injection did not significantly change cardiovascular variables from baseline values (data not shown). Also, following drug injections, cardiovascular status recovered to baseline status (Table 2.2) with the exception of heart rate in two individuals after the maximum ACh injection. The peak cardiovascular responses to ACh and AD are summarized in Table 2.2.  ACh produced a dose-dependent decrease in cardiac output by significantly slowing heart rate and decreasing stroke volume with 1 x 10-6 M ACh. Mean ventral aortic pressure did not change significantly in response to ACh. Nevertheless, CPO decreased by 2.5-fold and systemic resistance increased by over 3-fold with 1 x 10-6 M ACh. In response to 1 x 10-6 M ACh, CBF was more than halved and even reversed briefly during systole. However, the percentage of cardiac output going to the coronary artery was unchanged by ACh.  Injection of AD produced no overall effect on cardiac output, heart rate, stroke volume, CPO or systemic resistance (Table 2.2). Mean ventral aortic pressure decreased significantly with 1 x 10-7 M AD, but without changing the percentage of cardiac output going to the coronary artery. These mean responses to 1 x 10-6 AD masked two phenotype responses for heart rate (fig. 2.1). The Type I phenotype (N=5) was an average 15% increase in heart rate offsetting a 17% decrease in stroke volume, whereas heart rate was unresponsive in the Type II phenotype (N=5).   30 Given the large shifts of cardiovascular status following ACh injections, the individual associations between CBF and cardiovascular variables were extended to include the peak responses to ACh and AD (figs. 2.6 and 2.7). Again, a significant positive relationship (P<0.001) explained 60% of the variability between heart rate and CBF (fig. 2.6 B). The powerful association between heart rate and CBF was illustrated in those fish that altered heart rate by >10 bpm (N=5; one shark did not vary heart rate by 10 bpm), each of which showed positive linear relationships (fig. 2.6 B grey lines). In addition, weak but significant relationships were apparent between CBF and both cardiac output and CPO, which was likely due to the effect of heart rate on both variables because neither ventral aortic pressure nor stroke volume were significantly related to CBF (fig. 2.6). Again, following the removal of data from post-acetylcholine 10-6 M injection (the only concentration to have significant effects on cardiovascular variables), individual variation in CBF was significantly associated with variability in coronary resistance (fig. 2.7 A) and coronary resistance was inversely associated with heart rate (fig. 2.7 B). Both associations explained more than 50% of the individual variability.  2.4.4 In vitro perfused hearts Increasing coronary input pressure from 40 to 70 cm H2O increased CBF in perfused sandbar shark hearts (fig. 2.8 A). Despite a more than 2-fold individual variability in the initial CBF at 40 cm H2O (difference), CBF had a significant (inverse and exponential) association with coronary resistance (fig. 2.8 B) at each perfusion  31 pressure. Moreover, these in vitro curves paralleled the in vivo relationship between CBF and coronary resistance (fig. 2.8 B).  2.5 Discussion:  My results provide the first characterization of in vivo CBF in an elasmobranch without compromising cardiac function by opening the pericardium.  Furthermore, I provide the first measurements of total CBF in an elasmobranch and its relationship to simultaneous measurements of other cardiovascular variables. These experiments revealed a strong relationship between CBF and heart rate. While this result is consistent with the mammalian literature (Duncker and Bache, 2008), it is inconsistent with our hypothesis that a strong relationship should exist between CBF and CPO, as discovered in salmon ( R2>0.8; Davie and Farrell, 1991a; Gamperl et al., 1995). The relationship between CBF and heart rate reflected a strong relationship between heart rate and the estimated coronary resistance. Furthermore, CBF was tightly linked to coronary resistance in vivo and in vitro over the range of pressures explored here. This questions the assumption that CPO is tightly linked to CBF independently of heart rate in elasmobranchs.  2.5.1 Cardiovascular status at baseline and in response to drug injection Aside from heart rate, these are the first cardiovascular variables recorded for the sandbar shark. Mean ventral aortic pressure was similar to that measured in a related species, the black tipped reef shark (Carcharhinus melanoptera; ~50 cm  32 H2O) during tonic immobilization (Davie et al., 1993). Heart rate (42 bpm) was lower than that previously measured (~65 bpm) at similar temperatures in un-anaesthetized, chemically immobilized sandbar sharks (Dowd et al., 2006). This likely resulted in cardiac output (17.5 ± 1.8 ml min-1 kg-1) and CPO (2.2 ± 0.3 mW g-1 Mv) being lower in this study than previous measurements in other species of awake elasmobranchs at similar temperatures (Butler and Metcalfe, 1988; Speers-Roesch et al., 2012). Baseline CBF measured in sandbar sharks (0.78 ± 0.13 ml min-1 g-1 Mv) was 20% higher than the earlier partial CBF measurements in anaesthetized school shark at ~25°C, and 4-times higher than CBF in coho salmon (Oncorhynchus kisutch) (where only 30 - 40% of the ventricle is perfused) at 10°C (Axelsson and Farrell, 1993; Davie and Franklin, 1993).  CBF in the sandbar shark can be compared to left ventricular CBF in mammals at similar heart rates (fig. 2.9). Interestingly, the left ventricular CBF in resting ponies and horses at heart rates of 50 bpm are almost identical to the mean CBF in sandbar shark at 50 bpm (fig. 2.9) (Duncker and Bache, 2008; Manohar, 1987). The apparently higher oxygen requirement of the mammalian heart (a higher CPO) and the lower mechanical efficiency, ~12% for the trabeculae of mammals (Taberner et al., 2011) versus  ~20% for the relatively spongy hearts of sharks and teleosts, (Davie and Franklin, 1992; Graham and Farrell, 1990;), may be compensated by mammals having a higher haematocrit when compared with ectothermic sharks (Butler and Metcalfe, 1988; Windberger et al., 2003).  The observed variability in baseline cardiovascular function among individual sharks may reflect an anaesthetic effect. Although ketamine has been  33 shown to be an effective anaesthetic for sharks (Smith, 1992; West et al., 2008), it can produce large variation between individuals in depth of anaesthesia as well as elevated blood pressure in humans at an identical injection concentration (Blake and Korner, 1981; Grant et al., 1983). Similar to mammals (Blake and Korner, 1981; Komatsu et al., 1995), steady ventilatory movements existed when the sharks were anaesthetized. Additionally, if there was a pressor effect of anaesthesia in these sharks, the expected pressor responses to drug injections may have been attenuated. For example, responses to 1 x 10-6 M ACh produced only cardiac depression, which was expected (Chopin and Bennett, 1995; Taylor et al., 1977), and AD produced only a weak tachycardia in only half of the fish. Similar individual variation in the magnitude of responses to AD injection has been observed previously in un-anesthetised trout and elasmobranchs (Chopin and Bennett, 1995; Gamperl et al., 1994a; Gamperl et al., 1994b; Wood and Shelton, 1980).  2.5.2 Determinants of coronary blood flow The metabolic oxygen need of the mammalian heart is largely (~80%) related to the mechanical work done during contraction, and only ~20% for general tissue maintenance (Duncker and Bache, 2008; Yaku et al., 1993). When the mammalian myocardial oxygen demand is elevated by exercise, concurrent increases in CBF account for 80% of the increase in myocardial oxygen supply (Duncker and Bache, 2008). Coho salmon similarly display a concurrent increase in CBF in proportion to CPO (R2 = 0.81) during swimming challenges (Axelsson and Farrell, 1993; Gamperl et al., 1995). By pharmacologically modifying cardiac output,  34 and utilizing the variations in CPO and CBF within individual sandbar sharks, I detected a significant but weak (R2= 0.12) relationship between CPO and CBF (fig. 2.6). This was contrary to our hypothesis as the observed >4-fold range in CPO was not associated with proportional changes in CBF in the sandbar sharks. However, this relationship could be stronger in un-anesthetised sharks and warrants further investigation even though the challenge will be to use even larger sharks so that a flow probe with a much longer lead can be used.  For mammals, Duncker and Bache (2008) determined that increases in myocardial oxygen demand, and thus CBF, during exercise result primarily from the increased oxygen requirements for elevating heart rate (60%), followed by contractility (20%) and left ventricular work (20%). Furthermore, due to baroreceptor feedback on heart rate and vascular tone, aortic pressure is typically maintained at a relatively constant level with large increases in myocardial oxygen demand mainly associated with increases in cardiac output and not arterial blood pressure. Together this implies that by primarily modulating heart rate in mammals, which increases myocardial oxygen consumption, there should be, and is, a strong relationship between CBF and heart rate (Duncker and Bache, 2008). In contrast, teleosts and elasmobranchs modulate stroke volume to a greater degree than heart rate during exercise, yet primarily increase heart rate during warming (Clark et al., 2008; Dowd et al., 2006; Driedzic and Gesser, 1994; Eliason et al., 2013; Gamperl et al., 1995; Lai et al., 1989; Steinhausen et al., 2008). However, like mammals, increases in CBF appear crucial for supplying adequate oxygen when heart rate increases. I suggest that the coronary hyperaemia associated with high heart rates  35 likely resulted primarily from a reduction in coronary vascular resistance, which appears to be the most prominent method by which CBF is regulated in vertebrates (Davie and Farrell, 1991a; Duncker and Bache, 2008; Tomanek, 2012). Among the sandbar sharks, CBF increased 4-fold in vivo in conjunction with a 4-fold reduction in coronary resistance over a 30 bpm range in heart rate (fig. 2.7). Yet, in un-anaesthetized mammals, coronary resistance increases 6-14% over a 100 bpm range, and in trout 20-25% over a range of 45 bpm (Farrell, 1987; Feigl, 1983). Furthermore, as heart rate increases, there is a concurrent increase in the proportion of the cardiac cycle spent in systole, which translates to a prolonged extravascular compression of the coronary vasculature, increasing coronary resistance and phasically decreasing CBF. In sandbar sharks, increases in heart rate resulted in the proportion of the cardiac cycle spent in systole increasing from 18 to 38%, values similar to previous estimates of 38% in elasmobranchs and 23% in teleosts (Randall, 1968; Satchell, 1971). Despite extravascular compression which drives maximal CBF into the diastolic phase of the cardiac cycle (fig. 2.2, fig. 2.3), a linear relationship exists between heart rate and coronary resistance (Davie and Franklin, 1993; Farrell, 1987; Gamperl et al., 1995; Khouri et al., 1965; Miyazaki et al., 1990; Scaramucci, G.B., 1695). The immediate response of coronary resistance to changing heart rate is consistent with a metabolic flow regulation, as increasing heart rate is energetically expensive (van de Hoef et al., 2012). Our in vitro results also suggest that there is coronary auto-regulation in sandbar sharks as changes in coronary perfusion pressure (within the physiological range) had little effect on CBF. Even supra-physiological increases in coronary perfusion pressure in vitro only  36 slightly increased CBF, with the majority of the change attributable to changes in coronary resistance (fig. 2.7). This may indicate that auto-regulation of CBF appeared early in vertebrate evolution.  The mechanism by which vascular resistance, and thus CBF, is regulated is likely linked to the partial pressure of oxygen in the myocardial tissue at the end of systole. As the ventricle contracts myocardial oxygen demand is highest and oxygen supply, CBF, moves to its lowest (fig. 2.2). Even across individuals with similar cardiac output, those with a higher routine heart rate (and thus higher systolic period) and a lower stroke volume have higher routine CBF when compared to individuals with a higher stroke volume and lower heart rate (fig. 2.4).  The decrease in myocardial tissue oxygen tension due to the prolonged systole as heart rate increases could trigger vasodilation through a variety of established mechanism that might be transducing using oxygen, carbon dioxide or pH sensing pathways (Feigl, 1983) that trigger the release of a vasoactive agent. Coronary vasodilation in response to exercise in mammals has been found to be regulated by nitric oxide, hydrogen sulphide, ATP-sensitive potassium channels and adenosine (Dombkowski et al., 2005; Ishibashi et al., 1998). Additionally, serotonin and catecholamines relax coronary vascular smooth muscle in trout and elasmobranchs (Costa et al., 2015a; Farrell and Davie, 1991b; Small and Farrell, 1990; Small et al., 1990). In the absence of sympathetic innervation to the shark heart (Nilsson, 1994), autocrine or paracrine mediators of vascular tone such as endothelin, which has a universal constrictor effect, or inorganic compounds such as nitric oxide, may be the  37 mediators of vascular resistance (Chatterjee et al., 2015; Costa et al., 2015a; Small and Farrell, 1990).  Coronary perfusion pressure was estimated from mean ventral aortic pressure and assumed a constant blood pressure loss across the branchial and hypobranchial vessels that precede the coronary arteries. Thus, the blood pressure driving CBF averaged ~40 cm H2O because ventral aortic blood pressure rarely changed with drug injections, although it did vary between individuals (fig. 2.4). Thus, the difference between the in vivo and in vitro curves for coronary resistance and CBF at 40 cm H2O can likely be accounted for by the differences in the kinematic viscosity between saline and blood. However, it is also possible that the assumption of a constant 30% pressure drop was not valid both within a shark and between sharks.  For example, if heart rate varied inversely with branchial resistance, dorsal aortic pressure would decrease during bradycardia and this would decrease the driving pressure for CBF by an unmeasured amount. For this reason, we must be careful of over-interpreting the responses to ACh, which reduced heart rate and can constrict branchial arteries (Stensløkken et al., 2004); Wood 1974), reducing dorsal aortic pressure.  2.5.3 Relative importance of the coronary and luminal oxygen supplies in sandbar shark hearts With around 3% of cardiac output being directed to the myocardium (~0.07% of body mass) via the coronary artery, the heart is a relatively well- 38 perfused organ. Baseline CBF can be estimated to deliver 1.6 μmol O2 min-1 g-1 Mv by assuming arterial oxygen content is similar to that measured in leopard sharks, Triakis semifasciata, (2 mmol O2 L-1; (Lai et al., 1990a). The compact cardiac muscle of the conus arteriousus and ~15% of the ventricle of the sandbar shark heart must rely on this oxygen supply. Oxygen can also be obtained from the lumen of the heart by the spongy myocardium of the atrium and 85% of the ventricle. Using in vitro measurements of myocardial oxygen consumption for teleosts, it has been estimated that even predominantly spongy hearts use <4% of the available luminal oxygen supply (Davie and Farrell, 1991a; Davie and Franklin, 1992; Farrell, 1987; Farrell et al., 1985). Similarly, the luminal oxygen supply far exceeds what is required for maintaining maximal function in the shark hearts because, with a venous oxygen content of 0.9 mmol O2 L-1, this luminal oxygen supply delivers 23 μmol O2 min-1 g-1 Mv routinely or 10 μmol O2 min-1 g-1 Mv when venous oxygen content is reduced to 0.4 mmol O2 L-1 during exercise (Lai et al., 1990a). But, to what degree shark hearts take advantage of this luminal oxygen supply is unknown. In salmonids it is clear that the luminal supply is critical because, unlike sharks, their spongy myocardium lacks coronary vessels. This raises important questions about the vascularization of trabeculae in the spongy myocardium of elasmobranchs, which is studied in subsequent chapters of my thesis. Indeed, the role of coronary perfusion of ventricular trabeculae may be under appreciated given that sharks direct a greater proportion (%CBF) of cardiac output to the heart when compared to salmonids (Cameron, 1975; Davie and Farrell, 1991a). Furthermore, Davie and Farrell (1991) found that, similar to salmon, the perfused dogfish shark (S. acanthias) heart was  39 unable to overcome the effects of luminal hypoxia, yet perfusion of the coronary artery with red cell suspension restored mean cardiac output pressure through a significant increase in heart rate. This indicates that oxygen delivery through the coronary circulation is sufficient to impact overall myocardial function. Our data suggests that sharks take increasing advantage of this coronary supply as heart rates increase. With a high cardiac ejection fraction (80-90%; Franklin and Davie; Lai et al., 1990b) it appears that the luminal oxygen supply in elasmobranchs and teleosts could become limited at higher heart rates, which might help explain the relationship between heart and CBF. Alternatively, the rigid pericardium that surrounds the elasmobranch heart and the resultant vis-a-fronte filling mechanism allow for the coupling of ventricular contraction and atrial filling (Lai et al., 1996). This may reduce the overall energetic cost of modulating stroke volume by reducing the energetic cost associated with atrial filling (Farrell et al., 1988a). This could potentially explain the lack of relationship between stroke volume and CBF and the subsequent weak relationship between CBF and CPO.   2.5.4 Conclusions Changes in CBF in sandbar shark are significantly linked to variation in heart rate and I suggest that this is primarily modulated by coronary artery resistance. While the luminal oxygen supply appears sufficient for increasing stroke volume in sharks, when heart rate becomes elevated the luminal supply appears inadequate to support demand. This may be due to increases in oxygen use by the myocardium in combination with a decrease in luminal blood residency time, a reduction in luminal  40 blood perfusion to the distal spongy myocardium, and/or a decrease in venous blood oxygen pressure. However, the relationship between CBF and heart rate could be stronger in un-anesthetised sharks and warrants further investigation. 41  Table 2. 1: Variation in baseline cardiovascular variables for 10 anaesthetized sandbar sharks. Fish ID Estimated cardiac output (ml min-1 kg-1) Heart rate (bpm) Estimated stroke volume (ml beat-1 kg-1) Mean ventral aortic pressure (cm H2O) CPO (mW g-1 Mv) Estimated systemic resistance (cm H2O min g Mv ml-1) CBF (ml min-1 g-1 Mv) Estimated coronary resistance (cmH2O min g Mv ml-1) CBF (%?̇?𝑏) 1 11.6 54 0.22 58.0 2.1 5.0 - - - 2 13.8 31 0.44 63.6 1.9 4.6 - - - 3 24.5 36 0.68 64.6 4.0 2.6 0.64 70.3 1.7 4 18.7 56 0.33 49.7 2.1 2.7 1.29 27.0 5.0 5 13.7 52 0.26 56.3 1.7 4.1 0.67 59.0 3.7 6 26.8 48 0.55 44.5 2.7 1.7 0.98 31.7 2.6 7 23.5 27 0.88 52.0 2.6 2.2 0.36 101.8 1.2 8 15.6 50 0.31 66.5 2.0 4.3 0.74 63.2 4.0 9 16.4 43 0.38 54.2 2.0 3.3 0.11* - - 10 9.9 26 0.38 49.2 1.3 5.0 0.21* - - Mean ± s.e.m. 17.5 ± 1.8 42 ± 4 0.44 ± 0.07 55.8 ± 2.3 2.2 ± 0.3 3.5 ± 0.4 0.78 ± 0.13 58.8 ± 11.2 3.0 ± 0.6  *Coronary blood flow was measured only in one branch of the coronary artery and these values were not used to calculate mean coronary blood flow. CPO is the estimated cardiac power output, CBF is the measured coronary blood flow and Mv stands for ventricular mass. 42 Table 2. 2: Cardiovascular responses to acetylcholine and adrenaline in anaesthetized sandbar sharks (N= 10 except for CBF measurements where N=6).  Values are means ± sem.  Baseline Acetylcholine  Recovery Adrenaline  10-8 10-7 10-6  10-8 10-7 10-6 Estimated cardiac output (ml min-1 kg-1) 17.5 ± 1.8 18.3 ± 1.9 14.3 ± 1.5 6.3 ± 1.3 *  15.3 ± 1.6 17.5 ± 1.7 16.5 ± 1.4 17.5 ± 1.9  Heart rate (bpm) 42 ± 4 43 ±3 39 ± 3 29 ±4 *  41 ±3 47 ±4 42 ±3 47 ± 4  Estimated stroke volume (ml beat-1 kg-1) 0.44 ± 0.07 0.46 ± 0.07 0.38 ± 0.04 0.22 ± 0.04 *  0.39 ±0.05 0.39 ± 0.06 0.43 ±0.05 0.39 ±0.05  Mean ventral aortic pressure (cmH2O) 55.8 ± 2.3 56.5 ± 2.7 57.2 ± 2.7 58.5 ± 3.0  51.5 ± 3 57.2 ±2 .7 49.3 ± 3.4 * 57.2 ± 4.0 Diastolic 43.4 ± 2.7 48.7 ± 2.4 49.0 ±2.5 50.2 ± 3.0 *  45.4 ± 2.7 51.2 ± 2.7 43.4 ± 3 51.2 ± 3.8 Systolic 57.0 ± 4.3 67.3 ± 3.6 69.1 ± 3.5 73.8 ±3.3 *  59.6 ± 3.7 64.7 ±2.9 57.0 ±4.2 64.7 ±  4.8  CPO (mW g-1 Mv) 2.23 ± 0.3 2.42 ± 0.2 1.95 ± 0.2 0.88 ± 0.2 *  1.8 ± 0.2 2.27 ± 0.2 1.8 ± 0.2 2.27 ± 0.3  Systemic resistance (cm H2O min gMv ml-1) 3.5 ± 0.3 3.4 ± 0.5 4.3 ± 0.5 11.9 ± 1.7*  3.9 ± 0.6 3.6 ± 0.6 3.4 ± 0.5 3.8 ± 0.7  CBF (ml min-1 g-1 Mv) 0.78 ± 0.13 0.81 ± 0.15 0.68 ± 0.17 0.35 ±0.14*  0.60 ± 0.10 0.61 ± 0.12 0.62 ±0.11 0.64 ±0.08  CBF (%?̇?𝑏) 3.0 ± 0.6 2.9 ± 0.5 3.2 ± 0.7 4.0 ± 1.6  3.1 ± 0.7 3.1 ± 0.7 2.9 ± 0.6 3.0 ± 0.7  Coronary resistance (cm H2O min gMv ml-1) 58.8 ± 11.2 - - -  67.34 ± 12.6 - - -   * Indicates a significant difference from the baseline value. CPO is the estimated cardiac power output, CBF is the measured coronary blood flow and Mv stands for ventricular mass.   43  Figure 2.1: Representative traces of cardiac output and heart rate in anaesthetized sandbar sharks during baseline (A), following acetylcholine 10-6 M injection (B) a type I response to an adrenaline 10-6 M injection (c) and a type II response to adrenaline 10-6 M injection (D). Each trace is 1 min.     44   Figure 2. 1: Representative traces of simultaneous recordings of baseline cardiac output, coronary flow, ventral aortic pressure and heart rate in an anaesthetized sandbar shark for 8 seconds. Dashed vertical lines represent the beginning and end of systole determined using ventral aortic pressure.   45  Figure 2.2: Representative simultaneous recordings of baseline cardiac output, coronary flow and ventral aortic pressure in an anaesthetized sandbar shark during a cardiac cycle.      46    Figure 2. 3: The relationships between mean cardiovascular variables with coronary flow for individual anaesthetized sandbar sharks (colours denote Fish ID). Relationships for total cardiac output (A), heart rate (B), stroke volume (C), ventral aortic pressure (D), cardiac power output (E). R2 and P values for linear regression appear at the top of each panel. Each data point represents an averaged (bars = s.e.m.) baseline value for one fish, and 2 additional values from periods where the animal had returned to a steady state for greater than 3 minutes following drug injections.     47   Figure 2. 4: The relationship between coronary flow and coronary resistance (A), and heart rate and coronary resistance (B) between individual anesthetized sandbar sharks. R2 and P values for each relationship appear at the top of each graph. Each data point represent an individual fish, averaged (bars = s.e.m.) for the baseline value and 2 additional values from periods where the animal had returned to a steady state for greater than 3 minutes following drug injections. Different colours denote Fish ID (N=6).    48   Figure 2. 5: Relationship between coronary flow and total cardiac output (A), heart rate (B), stroke volume (C), ventral aortic pressure (D), and cardiac power output (E) in anesthetized sandbar sharks with a linear regression. N=6 with R2 and P values at the top of each graph. Different symbols denote different fish ID. Different colours denote different drug states with red indicating acetylcholine (bright red = 10-6 M concentration), green indicating adrenaline and blue indicating baseline and post saline injection values. The grey lines in panel b illustrate the association between heart rate and CBF in sharks that altered heart rate by >10 bpm (N=5; one shark did not vary heart rate by 10 bpm).   Heart rate (bpm)10 20 30 40 50 60Total cardiac output (ml/ min kg)0 10 20 30Coronary flow (ml/ min g Mv)0.00.20.40.60.81.01.21.41.6Stroke volume (ml/beat kg)0.0 0.2 0.4 0.6 0.8 1.0Ventral aortic pressure (cmH2O)30 40 50 60Coronary flow (ml/ min g Mv)0.00.20.40.60.81.01.21.41.6Cardiac power output (mW/g Mv)0 1 2 3 4 5R2 = 0.2P = <0.001R2 = 0.6P = <0.001R2 = 0.0P = 0.62R2 = 0.0P = 0.22R2 = 0.12P = 0.01A B CD E 49   Figure 2. 6: The relationship between coronary flow and coronary resistance (A), heart rate and coronary resistance (B). R2 and P values for each relationship appear at the top of each graph. Different symbols denote different fish ID. Colours denote different drug states with red indicating acetylcholine, green indicating adrenaline and blue indicating baseline and post saline injection values. Data from post-acetylcholine 10-6 M injection was removed from this analysis as it was the only concentration to have significant effects on cardiovascular variables and has been shown to have a differential effect on dorsal and ventral aortic pressure, thus coronary resistance could not be estimated.    50   Figure 2. 7: The relationship between coronary flow and coronary input pressure in the perfused sandbar shark heart (A) and the relationship between coronary flow and coronary artery resistance in vivo (; N= 6) and the perfused heart prep (; N=5) at different input pressures (B). Data for acetylcholine 10-6 M were not included. 51   Figure 2. 8: The linear relationship between heart rate and mean coronary blood flow for sandbar sharks in comparison to the heart rates and left ventricular mean coronary blood flows of other vertebrates at normal body temperatures. Mammalian regression equations were taken from summarized data compiled by Duncher and Bache (Review, 2011; Fig. 4).    Heart rate (bpm)0 50 100 150 200 250 300Cor onary Blood fl ow (ml/min g)-1012345DogSwineHorseHumanSandbar shark y = 0.021x - 0.27y = 0.016x - 0.30y = 0.023x - 0.32y = 0.012x - 0.38y = 0.016x - 0.61 52 Chapter 3: Coronary blood flow in leopard sharks, Triakis semifasciata, during acute warming2   3.1 Summary Heart rate and CBF in the leopard shark increases with temperature such that there is a significant linear relationship between the two. The increase in CBF with heart rate is similar to what has been previously observed in mammals during intense exercise and is likely due to both a reduction in vascular resistance in the coronary circulation and an increase in coronary perfusion pressure based on the findings in the previous chapter.  However, CBF appears to plateau at temperatures above 27°C, the temperature at which cardiac arrhythmias were observed in the larger sharks in this study. The development of arrhythmias at high temperatures in the larger individuals indicates that, like teleosts, larger individuals have lower temperature tolerances when compared to smaller conspecifics.                                                                 2 A version of this chapter will be submitted as Cox GK, Wegner NC, Lai NC and Farrell AP. Coronary blood flow in leopard sharks, Triakis semifasciata, during acute warming.   53 3.2 Introduction  Ectothermic vertebrates increase tissue oxygen demand when environmental temperature rises and this demand is usually met by an increased arterial oxygen convection via increases in heart rate (Lillywhite et al., 1999). While many underlying factors may potentially limit upper thermal tolerance, it is clear that heart rate and cardiac output cannot increase indefinitely with temperature.  This chapter explores the possibility that a limitation to myocardial oxygen supply contributes to the inability of the heart to work faster at supra-optimal temperatures and that coronary blood flow (CBF) is related to heart rate. As the hearts of most ectothermic vertebrates consist mainly of avascular spongy myocardium, it has long been assumed that their primary oxygen supply is derived from the direct diffusion of oxygen from venous blood within the ventricle (Davie and Farrell, 1991a). Thus, during acute warming when venous oxygen partial pressure falls as a result of increased oxygen extraction at the tissues (Eliason et al., 2013), a decrease in luminal oxygen supply could limit cardiac function, as proposed for teleosts (Davie and Farrell, 1991a). This is not because there is insufficient oxygen content in the luminal supply, but rather because the venous partial pressure of oxygen is too low to drive an adequate diffusion rate. The reason for this is because the partial pressure of oxygen in the blood that drives oxygen delivery is low in venous blood (the high blood oxygen affinity in elasmobranchs being a contributing factor) (Butler and Metcalfe, 1988). Jones (1986) calculated that a minimal luminal oxygen tension of 1.3 kPa was required for adequate rates of  54 oxygen extraction by mammalian myocardial cells (Jones, 1986). Many species of elasmobranch have routine venous oxygen tensions below this critical threshold, and thus should be unable to extract sufficient oxygen from the luminal supply (Bushnell et al., 1982; Lai et al., 1990a; Piiper et al., 1977; Taylor and Barrett, 1985). In a wide range of ectothermic vertebrates, e.g., holocephalians, elasmobranchs, scombrid teleosts and crocodilians (Durán et al., 2015; Kohmoto et al., 1997; Tota, 1989), the coronary circulation brings arterial blood directly from the gills/lungs to the ventricular trabeculae. Therefore, these predominantly spongy hearts have two potential oxygen supply routes, one venous and one arterial. Although the hearts in these species must be less dependent upon their luminal oxygen supply, the functional significance of arterial coronary oxygen supply during acute warming is unknown for any of these ectotherms. Nevertheless, the observation of a strong relationship between CBF and heart rate shown for sandbar sharks in chapter 2 leads to the prediction that CBF would increase with heart rate in response to increases in environmental temperature.  However, only bradycardic state and routine states were investigated in chapter 2. Therefore, the aim of this study was to investigate the role of CBF during an acute temperature increase in unanesthetized elasmobranchs. While positive relationships between temperature and heart rate have been observed in several species of teleost fish, amphibians and reptiles (Bennett, 1972; Clark et al., 2008; Eliason et al., 2011; Eliason et al., 2013; Lillywhite, 1987; Lillywhite et al., 1999; Seebacher, 2000; Taylor, 1931), there is a distinct lack of information on this relationship in elasmobranchs with predominately spongy hearts and a coronary  55 supply. The combined results of several studies however, reveal a positive relationship between heart rate and temperature in small-spotted catsharks, Scyliorhinus canicula, acclimated to water temperatures from 7 to 17°C (Butler and Metcalfe, 1988; Butler and Taylor, 1975; Short et al., 1979). I hypothesized that the positive relationship between CBF and heart rate found in sandbar sharks (Chapter 2) would exist when heart rates were elevated above routine values in the leopard shark (Triakis semifasciata, Girard).  The leopard shark is well suited for this study. Active and passive acoustic tracking of mature female leopard sharks show that they preferentially aggregate in near shore shallow embayments where sea floor water temperature averages 21.8°C but can reach 26°C (Hight and Lowe, 2007). During the early afternoon average body temperature is 19.5°C, but ranges from 18.0 to 25.9°C in summer. Most leopard sharks spend up to several hours daily at 24°C before moving into cooler waters at night (Hight and Lowe, 2007). Thus, leopard sharks experience rapid changes in environmental water temperatures (Hight and Lowe, 2007).  3.3 Methods  3.3.1 Animals Female leopard sharks (N=8) (11.7 ± 0.8 kg) were captured off the coast of La Jolla, San Diego using hook-and-line fishing gear, transported to the laboratory, and held in a shore-side circular tank (~ 4 m in diameter and 1.5 m deep) supplied with  56 flow-through sea (temperature 19-20°C, salinity 34-36‰). Leopard sharks of this size were needed to resolve CBF with Doppler flow probes. Animals were held at a natural photoperiod and fasted for at least 24 h prior to experimentation. This study was approved by the University of British Columbia Animal Care and Use Committees (A11-0355), and followed all applicable laws and regulations.   3.3.2 Surgical procedures Individual sharks were anaesthetized by immersion in 20°C seawater and tricaine methanesulfonate (MS- –1; Sigma, St Louis, MO, USA). The surgical plane of anaesthesia was verified when sharks failed to respond to a tactile stimulus (tail pinch), after which they were moved into the laboratory and placed ventral side up in a custom designed V-shaped board. They were covered with wet towels and their gills were irrigated with aerated seawater containing a lower dose of MS-222 (50 mg l-1). A 4-6 cm midline incision was made anterior to the coracoid bar through the skin and coracomandibularis muscle, and the ventral aorta was carefully exposed with blunt dissection through the connective tissue of anterior jaw muscles. Due to anatomical constraints (the immediate bifurcation of the ventral aorta and afferent gill arteries directly anterior to the pericardium) and position of the coronary artery, total cardiac output cannot be measured directly in leopard sharks without disrupting the pericardium and the conal myocardium. Instead, an ultrasonic transit-time perivascular blood flow probe (2.5 - 3.0 mm PSB; Transonic Systems Inc., Ithaca, NY, USA) was positioned around the ventral aorta  57 anterior to the 3rd afferent gill artery to estimate cardiac output by measuring blood flow to the anterior gill arches. Although the PSB Transonic probe was of suitable size for the vessel and the lead was sutured to a coracomandibularis muscle adjacent to the ventral aorta in order to secure the probe’s position on the vessel, it was found that the Transonic probe outputs greatly under-reported cardiac output based on previous studies (Lai et al., 1989). Furthermore, several traces appeared to be blunted. As this may have resulted from the difference in probe calibration temperature and the experimental temperature range used in this study, thus values for cardiac output and stroke volume were deemed unreliable and are not reported here. Ventral aortic pressure was measured via a heparinized catheter fashioned from polyethylene tubing (PE-50; IntramedicTM, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) that was occlusively implanted in the 3rd afferent gill artery. The ventral aortic cannula remained free from blood clotting in all but one shark and thus pressure was measured in 7 sharks. The coronary arteries originate from the first efferent gill artery and travel posteriorly between the coracobranchial muscles on either side of the ventral aorta before entering the pericardium and dividing on the conus arteriosus. After careful dissection, a Doppler flow probe (0.5-0.7 mm Triton Technology, Inc, San Diego, CA USA) was placed around one of the two lateral hypobranchial arteries to measure half of the total CBF coronary blood flow. Slight variation in coronary artery morphology prevented access in two sharks, reducing CBF measurements to 6 sharks. Following surgery, the shark was moved to the experimental tank (~2m long x 75 cm wide x 75 cm deep) and the gills were irrigated with fresh, aerated seawater  58 at 20°C from the experimental tank until the shark resisted artificial irrigation. Following overnight recovery the catheter was connected to a pressure transducer (model DPT-6100, VBM Medizintechnik, Kirchseeon, Germany) that had been calibrated against a static column of seawater and referenced to the water level in the experimental tank. A 4-channel bridge amplifier (Somedic, Hörby, Sweden) amplified signals from the blood pressure transducer. The transonic flow probe was connected to a TS420 Perivascular Flow Modules (Transonic Systems, Ithaca, NY, USA) and the Doppler probe was connected to CBI Pulsed Doppler Module (Triton Technology, Inc., San Diego, CA USA). Cardiovascular variables were monitored on-line (behind a curtained area to limit disturbances during experimentation) and stored for subsequent analysis using a Power Lab Unit (ADInstruments, Castle Hill, Australia) connected to a laptop computer running LabChart Pro software (v.7.3.7; ADInstruments).   3.3.3 In vivo temperature challenge.  Recordings of stable routine cardiovascular variables were performed for 1 h at 20°C, after which water temperature was increased by increments of 1°C every 30 min to a final temperature of 30°C after 5 h. This was accomplished by circulating heated aerated water from the anterior of the holding chamber into the animal holding area. Oxygen partial pressure of the water was measured during the experiment to ensure that the water within the chamber did not become hypoxic during the experiment. Cardiovascular variables were recorded and monitored continuously during the temperature challenge. Cardiovascular variables were  59 averaged over a 1 min period at the end of each 1°C temperature increment to provide the value for that temperature. The sharks were warmed to 30.1 °C before recovery at the control temperature by immediately flushing the holding tank with fresh 20°C seawater. Arrhythmias were observed in three sharks before 30°C was reached and so the control temperature was restored prematurely. At the end of the exposure, sharks were euthanized with an overdose of tricaine methanesulfonate (MS- –1; Sigma, St Louis, MO, USA) before the heart was excised, rinsed clean of blood, and placed in an elasmobranch physiological saline solution (concentrations in mM: 360 Urea, 280 NaCl, 90 TMAO, 12 KCl, 10 CaCl2·2H20, 5 MgCl·6H20, 5 glucose, 4.5 NaHCO3, 0.5 NaH2PO4, pH 7.8).  The excised heart was then used for in vitro coronary artery perfusion.   3.3.4 In vitro coronary artery perfusion The excised perfused heart preparation followed that used previously for salmon (Farrell and Graham, 1986). The largest coronary artery on the conus arteriosus was cannulated with polyethylene tubing (PE-50) that was tipped with smaller tubing (PE-20), and secured to the ventral aorta with 4-0 braided silk suture thread. Initial perfusion with heparinized saline at a pressure of 30 cm H2O (1 cm H2O = 0.1 kPa) rapidly cleared the coronary vessels of blood. Collateral connections between the various coronary arteries ensured perfusion of the majority of the coronary circulation downstream of the cannulation site. Once cleared of blood, the other main coronary branches were occluded upstream of the cannulation site to prevent backflow from any severed coronary arteries on the conus arteriosus. The  60 total preparation time was 15 min. Coronary flow was measured as the time required for 0.6 ml of saline to enter the coronary circulation at a pressure of 30 and 40 cm H2O, roughly 70 - 80% of the measured in vivo ventral aortic blood pressure.   3.3.5 Calculations and statistics Heart rate (bpm) was determined using systolic peaks from either the coronary blood flow or ventral aortic pressure signals. Ventral aortic pressure (cm H2O) is reported as the mean value from blood pressure traces. Instantaneous blood velocity in a coronary artery was recorded with a Doppler flow probe as an index of CBF. Control = 0%, the relative change assumes that there is equal blood flow in the paired coronary arteries. The coronary artery proved too difficult to cannulate in situ, which prevented calibration of the flow probes to report CBF. The effect of temperature on mean cardiovascular variables was assessed with one-way repeated-measures ANOVA followed by Holm-Sidak post-hoc test to identify individual temperatures that were significantly different from the initial value at 20°C. Statistical significance was set at P ≤ 0.05 and statistical tests were run in SigmaStat (v10.0.1, Aspire Software International, Ashburn, VA, USA). The relationships between cardiovascular variables were described by regression in SigmaStat, which was also used to produce the curves that best fit the data.   61 3.4 Results  3.4.1 Temperature effects on baseline in vivo cardiovascular status  Baseline cardiovascular status at 20°C is summarized in figure 3.1. Heart rate was 39 bpm and mean ventral aortic pressure was 45 cm H2O. Warming by 10°C significantly and linearly increased heart rate (R2 = 0.98, P <0.001; fig. 3.1 C). Heart rate reached 64 bpm at 30°C, resulting in a Q10 of 1.6.  Mean ventral aortic pressure was unchanged over the temperature range, although the five sharks that reached 30°C had atypical ventral aortic pressure traces that showed a rapid drop in pressure during diastole (fig. 3.3). The three largest sharks (13 – 14.5 kg) developed cardiac arrhythmias prior to reaching 30°C, preceded by a decrease in mean ventral aortic pressure (fig. 3.2).  CBF (N=5) was significantly and linearly related with temperature (R2 = 0.82, P<0.001; fig. 3.1 A) and the majority of CBF occurred mainly during diastole independent of the increase in heart rate with temperature (fig. 3.3). However, CBF variability increased considerably at temperatures above 25°C (fig 3.1 A). With heart rate and CBF both increasing with temperature, the two variable were significantly and linearly related (R2 = 0.8, P < 0.001, fig. 3.4). Nevertheless, CBF tended to plateau at 27°C, the temperature at which arrhythmias were first observed in the largest shark. Furthermore, CBF decreased at 25°C prior to any further increase with temperature in the majority of individuals (fig. 3.1 A).   62 3.4.2 In vitro coronary artery perfusion Increasing coronary input pressure from 30 to 40 cm H2O did not produce a significant increase in coronary flow rate (0.13 ± 0.02 and 0.17 ± 0.03 ml min-1 g-1 Mv, respectively) or change coronary artery resistance (260 ± 40 and 279 ± 46 cm H2O min g Mv ml-1, respectively (table 2.1).   3.5 Discussion   My results provide the first characterization of in vivo CBF in an un-anaesthetized elasmobranch as well as the in vivo CBF in response to an acute temperature change in any fish species. Acute warming was used to elevate heart rate above baseline levels to investigate the role of cardiac work rate on CBF. Consistent with my hypothesis, there was a strong linear relationship between heart rate and CBF between 20°C and 30°C. Thus, my Ph.D. thesis has shown a tight relationship between heart rate and CBF for both bradycardic (Ch2) and tachycardic states in sharks.   3.5.1 Cardiovascular status in response to increasing water temperature Baseline ventral aortic pressure in 9 -14.5 kg leopard sharks was slightly higher than that reported for smaller (0.6 -2.7 kg; 63 cm H2O) leopard sharks (Lai et al., 1990b) and similar to the dogfish species Scyliorhinus canicula (~54 cm H2O) (Taylor et al., 1977). Heart rate was ~ 25% lower in the larger leopard sharks when compared with smaller leopard sharks (Lai et al., 1989; Scharold et al., 1988), which  63 might be expected because routine heart rate decreases with increasing body size (Dowd et al., 2006; Schmidt-Nielsen, 1984).  The Q10 over this temperature range was similar to that measured in juvenile sandbar sharks (Q10 = 1.8) over a similar temperature range (Dowd et al., 2006). The Q10 for heart rate in teleost fish usually ranges from 1.3 – 3.0 depending on the species and temperature range tested (Gamperl, 2011) and a Q10 of 2.3 has been measured in varanid lizards (Bennett, 1972). In vivo studies with teleosts have shown that once heart rate has reached its peak, a cardiac arrhythmia can develop with acute warming to or slightly below the critical thermal maximum for the species (Casselman et al., 2012; Clark et al., 2008; Eliason et al., 2013). Here I found that the three largest sharks developed cardiac arrhythmias prior to completing the temperature challenge, consistent with Clark et al. (2008). Thus, cardiac rhythm is disrupted at extremely warm temperatures in teleosts, amphibians, reptiles (Biörck and Johansson, 1955; Rutskina et al., 2009) and now sharks. Beyond a potential influence of body size, my study provides fresh insight into the failure of the heart during acute warming. Leopard sharks, which routinely experience habitat temperatures below < 24°C developed atypical ventral aortic pressure traces as well as cardiac arrhythmia at 30°C and below. Additionally, the plateau for CBF further indicates that an upper thermal limit for leopard sharks was approached in the present study, thus suggesting that the peak heart rate and coronary flows recorded in this study may be maximum values for leopard sharks.   64 3.5.2 Determinants of coronary blood flow  Consistent with chapter 2 and mammalian studies (Duncker and Bache, 2008), changes in heart rate were strongly linked with CBF in sharks.  The coronary circulation in both sharks and mammals perfuse all myocardial tissue. Here ventral aortic pressure did not change during warming, so unless gill resistance decreased and dorsal aortic pressure increased, the increase in CBF would be a result of reduced coronary vascular resistance, which is the mechanism suggested in Ch 2. Indeed, vascular resistance may be linked to the observed decrease in CBF in every individual at ~25°C (fig. 3.1 A) through the effects of extravascular compression. Force-frequency relationships in the myocardium of elasmobranchs show that there is an initial increase in tension development with frequency, however after an apex is reached force declines sharply with increasing frequency (Driedzic and Gesser, 1988). If contractile force is maximized in the leopard shark heart at ~50 bpm (fig. 3.4), it is possible that an associated increase in extravascular compression at this heart rate will increase coronary resistance such that CBF is decreased despite maximal vasodilation. The subsequent decrease in force (and extravascular compression) that occurs in other sharks at high heart rates may explain the sharp increase in CBF at heart rates above 50bpm (fig. 3.4), and thus at temperatures over 26°C.  Temperature has been shown to alter the potency of vasoactive agents acting on coronary arterioles in rainbow trout (Oncorhynchus mykiss), e.g., the dilatory effect of adenosine on coronary arterioles increases with temperature (Costa et al., 2015a). However, speculation on possible mechanisms that alter coronary  65 resistance during the warming of leopard sharks is impossible without further study because the responses to acetylcholine, catecholamines, angiotensin, serotonin, adenosine and bradykinin have been shown to vary with drug concentration, vessel size, age, temperature and species (Costa et al., 2015a; Farrell, 1987; Farrell and Davie, 1991a; Farrell and Davie, 1991a; Small and Farrell, 1990; Small et al., 1990). Nevertheless, because the shark heart lacks sympathetic innervation (Nilsson, 1994), autocrine or paracrine mediators of vascular tone such as endothelin, which has a universal constrictor effect, or inorganic compounds such as nitroprusside that appears to have a universal dilatory effect, may be important modulators (Chatterjee et al., 2015; Costa et al., 2015a; Small and Farrell, 1990).  The decrease in gill resistance and the increase in dorsal aortic pressure observed previously with acclimation temperature in S. canicula (Butler and Metcalfe, 1988; Butler and Taylor, 1975; Short et al., 1979) can now be viewed in a different light. Warm acclimation could be seen as a resetting of the driving pressure for CBF in sharks. Therefore, studies of CBF after thermal acclimation is warranted. 3.5.3 Conclusions Heart rate and CBF in the leopard shark increase linearly with temperature and with each other. The dependence of CBF on heart rate is likely due to a reduction in vascular resistance in the coronary circulation. However, CBF for the leopard shark appears to plateau at temperatures above 27°C, the temperature at which cardiac arrhythmias appear in larger sharks.     66  Table 3. 1: Mean (± SEM) in vitro perfused coronary artery blood flow and resistance at given filling pressures Coronary input pressure (cm H2O) Coronary flow (ml min-1 kg-1) Coronary artery resistance (cm H2O min g Mv ml-1) N 30 0.13 ± 0.02 261 ± 40 6 40 0.17 ± 0.03 279 ± 46 7    67   Figure 3. 1:  Mean (± SEM) relative changes in coronary blood flow with temperature (A) and absolute changes in ventral aortic pressure (B) and heart rate (C) over a 10°C acute temperature increase in leopard sharks. The pink area indicates data collected at temperatures that are above the normal environmental temperature range for leopard sharks (>24 °C). An asterisk indicates a significant difference from 20°C. The numbers below the data points 20°C indicate the number of sharks represented for each cardiovascular variable until the pink zone when some experiments were terminated due to the appearance of cardiac arrhythmias   Temperature(C)20 22 24 26 28 30Heart rate (bpm)020406080Ventral aortic pr essure (cm H20)102030405060* ** ** **ABCR2 = 0.8P < 0.00187577646 55R2 = 0.98P < 0.001***Coronary blood flow (% change)01002003003 68   Figure 3. 2: Representative trace of an arrhythmic event in a leopard shark (size = 14 kg) at 28°C.  69   Figure 3. 3: Representative traces of coronary blood flow, ventral aortic pressure and heart rate at 20°C and 30°C in a leopard shark (size = 12 kg).    70 Heart rate (bpm)30 40 50 60 70Coronary blood flow (% change)020406080100120140160180R2 = 0.8P < 0.001544 Figure 3. 4: The relative change in coronary blood flow with a temperature-mediated increase in heart rate in leopard sharks. The pink data points indicates data collected at temperatures that are above the normal environmental temperature range for leopard sharks (>24 °C). n= 6 unless otherwise indicated.     71 Chapter 4: Morphological arrangement of the coronary vasculature in a shark (Squalus sucklei) and a teleost (Oncorhynchus mykiss)3   4.1 Summary   Given that my previous chapters have shown an unexpectedly high proportion of cardiac output being directed to the coronary circulation in shark species and that CBF is greatly influenced by HR, as in mammals, my prediction is that the coronary circulation in the spongy trabeculae of sharks is better developed than previously thought. To test this idea, I developed a methodology in this chapter that allowed for the quantification of vascularity, vessel morphology and estimations of oxygen diffusion distances. Although vascularity was significantly higher in the compact myocardium of trout when compared to dogfish, inter-vascular distances were similar. This was due to the significantly larger vessel diameter in dogfish, which corresponds to their larger red blood cell size when compared to trout.  In contrast to rainbow trout, the spongy myocardial tissue of the dogfish does not appear to rely greatly on the luminal circulation for oxygen as measurements of vascularization in the spongy myocardial tissues exceed that of the compact tissues in dogfish.                                                                3 A version of this chapter will be submitted as Cox GK, Kennedy G. and Farrell AP. Morphological arrangement of the coronary vasculature in a shark (Squalus sucklei) and a teleost (Oncorhynchus mykiss)  72 4.2 Introduction  A functioning heart is critical for survival in all vertebrates. To function properly, the heart needs a near continuous supply of oxygen due to a limited anaerobic capacity.  For birds and mammals, the oxygen supply to the condensed compact myocardium, which comprises the majority of the heart, comes primarily from a dedicated coronary circulation, which has its origin as the first branch off the systemic aorta.  Even the myocardial trabeculae, which function in preventing the inversion of the bicuspid and tricuspid valves (Goo et al., 2009; Tomanek, 2012), have coronary vessels despite direct contact with luminal blood, which is absent in the compact myocardium.  The vascular anatomy of fish highlights the importance of the coronary circulation because it originates as the first branch of the anterior most efferent gill artery. This makes the heart one of the first organs to receive oxygenated arterial blood, as in mammals. Despite this anatomical similarity qualitative descriptions of the coronary vasculature of fishes suggest a different coronary architecture within the heart. Foremost, there are two functional oxygen delivery routes to the myocardium: the coronary circulation and the luminal blood (Davie and Farrell, 1991a). In teleost fish, these two routes are associated with different myocardium muscle arrangements: coronary arteries are associated with the outer compact myocardium and luminal blood with the inner spongy trabeculae. The spongy trabeculae are devoid of coronary vessels in most species apart from tuna. In fact, rarely does a teleost heart have a ventricle with < 50% spongy myocardium (Santer  73 and Greer Walker, 1980), which means most of the ventricular myocardium relies mainly on the luminal oxygen supply (Davie and Farrell, 1991a). The hearts of many teleost species even lack compact myocardium and the associated coronary circulation, relying entirely on the luminal oxygen supply. Chondrichthyans are of special interest because they are the most ancient vertebrates to have developed coronary arteries but have a predominately spongy ventricular myocardium in all species studied to date.  Only the cardiac muscle of the conus arteriosus is entirely compact myocardium.  Nevertheless, in contrast to teleosts and fetal mammals, the spongy as well as the compact myocardium of the chondrichthyan heart receives coronary vessels (Helle et al., 1983; Santer and Greer Walker, 1980; Tota, 1989).  This is true even in species that are suggested to have secondarily lost their compact myocardium (Durán et al., 2015). Thus, while chondrichthyan compact myocardium of the ventricle and conus arteriosus has a coronary oxygen supply, the spongy myocardium of the atrium and ventricle has two potential oxygen supply routes, the coronary and luminal. The relative importance of these two routes is unknown, but based on my in vitro and in vivo findings in chaptesrs 2 &3 an unusually large proportion of cardiac output is directed to the shark coronary circulation. Furthermore, the CBF seems to be very responsive to changes in heart rate and coronary resistance.   The anatomical arrangement of myocardial tissue and gross arrangement of the major coronary arteries are well documented in a variety of elasmobranch species (subclass of chondrichthyes).  The theme is highly species variable (Muñoz-Chápuli et al., 1994).  Qualitatively, characterization of elasmobranch major  74 coronary arteries dates back to the 1800’s (Carazzi, 1905; De Andrés et al., 1990; De Andrés et al., 1992; Grant and Regnier, 1926; Marples, 1936; Muñoz-Chápuli et al., 1994; Parker, 1887; Parker and Davis, 1899; Pavesi, 1874; Tota, 1989; Tota et al., 1983) and includes 34 species (23 genera), 21 species of sharks being particularly well documented (De Andrés et al., 1990; De Andrés et al., 1992). Lacking, however, are quantitative morphological measurements of the coronary macro- or microvasculature in any elasmobranch.  Qualitative and quantitative morphological observations of vasculature are critical to properly understand the structural limits of gas exchange in animals. Indeed, the normal anatomical arrangement and function of the adult coronary circulation has been extensively studied in mammals (Tomanek, 2012). In contrast, the only other taxa represented in the quantitative literature for coronary microvasculature are two species of bird (Scott et al., 2011) and one species of teleost fish (Egginton and Cordiner, 1997), These studies revealed similar capillary density for low altitude birds and mammals, while rainbow trout have about half the vascular density.  This dearth of information is likely due in part to the variable coronary microvascular arrangements in non-avian and non-mammalian vertebrates, and the difficulty of applying quantitative methods to these more complex coronary arrangements.  Therefore, the objective of the present study was to develop a reliable methodology for the quantification of myocardial vessels in both the trabecular and compact myocardium of fish hearts. These methods were then used to compare cardiac vascularity for a representative shark, spiny dogfish (Squalus sucklei), and  75 the only teleost fish for which data already exists, rainbow trout (Oncorhynchus mykiss).  I hypothesized that (1) the degree of myocardial vascularization in dogfish will be higher in the ventricular compact than the ventricular spongy myocardium due the lack of luminal oxygen supply; (2) vascularization will be highest in the cardiac chambers that do the most work, i.e., the ventricle followed by the conus arteriosus and then the atrium and (3) that rainbow trout heart will have a greater capillary density in the compact myocardium of the ventricle because it routinely performs more work with a higher blood pressure with a faster heart rate (Sandblom et al., 2009; Stevens and Randall, 1967; Thorarensen et al., 1996).    4.3 Methods   4.3.1 Animals Pacific spiny dogfish (S. suckleyi; body mass (𝑀𝑏) ranging from 0.8 to 1.2 kg) were caught in English Bay off the coast of West Vancouver, B.C., Canada. They were held in 2,000 L fiberglass tanks with flow-through, aerated seawater at 9-11oC.  Rainbow trout (O. mykiss; 𝑀𝑏 ranging from 0.8 – 1.5 kg) obtained from Miracle springs hatchery (Mission, B.C., Canada) were housed in 2,000-liter fiberglass tanks with flow-through freshwater at temperature of 9-11oC. All procedures were approved by the Animal Care Committee of the University of British Columbia and conducted in accordance with the Canadian Council on Animal Care guidelines.   76 4.3.2 Perfusion fixation of the coronary vasculature  Fish were euthanized by an overdose of tricaine methane sulphonate (MS222). Once fish were unresponsive to tactile stimulus (tail pinch) 1 mL of 1000 IU/ml heparin (Sigma-Aldrich Co. LLC., Oakville, ON, Canada) was injected into the caudal vein and allowed to circulate until ventilation ceased. Fish were then placed on a surgery table ventral side up and the entire heart was removed and placed in water-jacketed beaker containing heparinized physiological saline at 10oC. The coronary arteries in the dogfish originate from the anterior efferent gill arteries and pass posteriorly between the coracobranchial muscles on either side of the ventral aorta before entering the pericardium and dividing on the conus arteriosus. A major branch of the coronary artery was cannulated with polyethylene tubing (PE-60; IntramedicTM, Becton, Dickinson and Company, Franklin Lakes, NJ, USA) that was tipped with smaller tubing (PE-10) and secured at the anterior end of the conus arteriosus to ensure maximal perfusion of the coronary vasculature. The lateral connections between major conal coronary arteries allowed for the perfusion of the entire conus, ventricle, atrium and sinus venosus of the dogfish heart. In trout the coronary artery was cannulated anterior to its bifurcation on the outflow tract (bulbus arteriosus) with a tapered PE-10 tipped cannula.  The coronary artery was perfused with heparinized physiological saline at 10oC. The saline recipes for the respective species were identical to those used in previous in vitro studies on working excised hearts (Davie and Farrell, 1991b; Farrell, 1987) with the addition of papaverine (10 μmol L-1; Sigma-Aldrich Co. LLC., Oakville, ON, Canada) in order to elicit full vasodilation (Farrell, 1986). Coronary artery input pressures were  77 maintained at 30 cm H2O for dogfish and 40 cm H2O for trout, which were intended to be close to the respective in vivo blood pressure in the dorsal aorta. When the effluent from the cardiac outflow tract and the coronary arteries was visually clear of red blood cells, perfusion rate was measured by timing the entry of 0.4 ml of saline into the coronary circulation. Afterwards perfusion was switched to 4% buffered paraformaldehyde until the heart became pale and rigid, after which the entire heart was then transferred to fresh 4% buffered paraformaldehyde. For histology, hearts were dehydrated over several days to 70% EtOH. One additional heart from both the dogfish and trout were perfused with a vascular corrosion compound rather than the fixative. A Mercox (Ladd Research, Williston, Vermont, USA) vascular cast allowed analysis of the larger vessels after tissue digestion in KOH, whereas a MICROFIL® (Flow Tech, Inc., Carver, Massachusetts, USA) vascular cast allowed for non-destructive micro-CT imaging using a Scanco µCT-100 (Scanco Medical, Brüttisellen, Switzerland) with an isotropic voxel size of 17.2 µm.   4.3.3 Tissue sectioning and staining In dogfish (N=4), representative fixed tissue samples were taken after separation of the heart into the compact myocardium of the conus arteriosus, and ventricle, as well as the spongy myocardium of the ventricle and atrium.  These samples were trimmed before embedding in paraffin wax and sectioned (5 µm thickness; Wax-it Histology Services Inc. Vancouver, B.C. Canada). An effort was made to section the tissue perpendicular to myocyte orientation, especially in the trabeculae. Five non-sequential sections were selected from each of the 4 tissues  78 and placed on glass slides. The sections were then stained with Verhoeff’s elastic-tissue stain and counterstained in van Gieson’s stain (Preece, 1972). In trout (N=4), representative fixed tissue samples of compact and spongy myocardium of the ventricle were removed as one piece. Tissue sections were obtained in the same manner as above.   4.3.4 Imaging and morphological measurements Tissue sections were microscopically examined at 400x magnification with a Zeiss Axio cam MRC attached to a Zeiss Axio Scope.A1 (Zeiss, Germany). The area analysed in each section was greatly enhanced by stitching together multiple images of the sections in Photoshop (Adobe v.11.0). Selection of regions chosen for digitization contained predominantly circular vessel cross-sections, similar to methods suggested in Gerdes (2015) (Gerdes, 2015) who also suggested rejection of regions with signs of incomplete perfusion, such as red blood cells, collapsed vessels or tissue damage (fig. 4.1). Selection of vessel cross sections was especially difficult for conal tissue (figs. 4.1 A, C, E and 4.2). Longitudinal and oblong vessels were distinguished from a tissue separation artefact by the presence of a stained endothelium. Images were analyzed using Volocity Image Analysis software (PerkinElmer Inc.). Vessel diameter was measured as the shortest distance across a vessel cross section. Inter-vascular distances (IVD) were measured as the distance across muscle tissue between the vessel walls of two adjacent vessel cross-sections. In the ventricular and atrial spongy myocardial tissue, the distance from the edge of the trabeculae to the walls of the closest internal vessels is referred to as the edge  79 distance. Vascular density was calculated as the number of microvessels per mm2. Vascular density of the ventricular and atrial spongy trabeculae was calculated as the number of vessels per cross-sectional area of trabeculae. I use the term vascular density instead of capillary density because many microvessels with vessel walls of < 2 µm ranged in diameter from ~10 µm to 60 µm (with corresponding areas of ~100 µm2 to 2800 µm2) (fig. 4.3), diameters that far exceeded a typical capillary diameter.  Vessels with walls <2 µm were include in vascular density measurement (Pittman, 2011; Tsai et al., 2003).   4.3.5 Statistics Between species comparisons of vascular variables in the ventricular compact myocardial tissue were tested using a Students t-test in Sigmastat (Aspire Software International, Ashburn, VA, USA). Single-factor analysis of variance (ANOVA) with Holm-Sidak post-hoc test was used to compare between dogfish myocardial tissue types. The statistical program “R” was used to create the graphs. Statistical significance was set at p ≤ 0.05 and all values are reported as means ± sem.    80 4.4 Results  4.4.1 Interspecific difference between trout and dogfish sharks The gross morphology of the dogfish and trout coronary vasculature differed in both the proportion and type of heart tissue types perfused by the coronary circulation (fig. 4.2 C, D). The dogfish coronary circulation supplied the spongy tissue of the ventricle and atrium, as well as the compact tissue of the ventricle and conus arteriosus (fig. 4.2 C). In contrast, the coronary circulation of the trout supplied only the compact myocardium of the ventricle (fig. 4.2 D). Although micro-CT scans highlighted the presence of vessels within the cardiac tissues, the technique did not resolve all of the microvasculature and thus only illustrated that dogfish and trout varied with regard to tissue types. Nevertheless, similar coronary flow rates and coronary vascular resistances were obtained when the coronary circulations of the dogfish and trout were perfused in vitro with saline using a physiological input pressure relevant for each species (table 4.1). As the perfusate was switched from saline to 4% buffered paraformaldehyde, the frequency of contraction progressively decreased until the hearts typically arrested in diastole.  The vascular density and vessel area differed significantly in ventricular compact myocardium between dogfish and rainbow trout (fig. 4.1; table 4.2). Trout had a significantly greater vascular density and a smaller average vessel area compared with dogfish (table 4.2; fig. 4.4). As a result, IVD distance within the ventricular compact was similar for the two species, but dogfish had greater variation in IVD (table 4.2; fig. 4.5). Microvessels in ventricular compact  81 myocardium were no closer than 2.9 µm in both species, and no further apart than 32 µm in the trout and 41 µm in the dogfish. Mean vessel diameter did not differ significantly (trout = 8.8 µm; dogfish = 10 µm), but trout had a smaller variation in vessel diameter and area (fig. 4.4 and 4.6).  The vascular density and vessel area in the compact myocardium of the dogfish conus arteriosus was also significantly lower than the ventricular compact of the trout (table 4.1).   4.4.2 Intraspecific in differences in vascularization within the dogfish myocardium Neither vascular density nor vessel dimensions differed significantly between the compact of the conus arteriosus and ventricle of the dogfish (table 4.1). The orientation of coronary microvessels perfusing the compact tissue of the dogfish heart was highly variable. Figure 4.2 illustrates the orientation of the coronary vessels within the spongy trabeculae as well as its overall orientation in the compact muscle of the conus and anterior ventricle. Many of the dogfish coronary microvessels ran diagonally or perpendicular to muscle fibre orientation in the conus arteriosus, and it was thus impossible to choose sections of sufficient area for analysis that were completely free of longitudinal or oblong vessel sections for that tissue (figs. 4.1 C, E and 4.2 A). In contrast, coronary micro-vessel orientation was aligned well with muscle fibre orientation in the spongy tissue of the dogfish (figs.  4.1 D,F, 4.2 B). Despite anastomoses between adjacent microvessels, the majority of coronary vessels ran in parallel within the ventricular spongy trabeculae (fig. 4.2 B). Vascular density differed significantly between the  82 ventricular spongy and compact tissues, however vessel area did not. While vascular density was similar for ventricular and atrial spongy myocardium, vessel area was significantly greater for atrial compared with ventricular trabeculae despite a similar vessel wall thickness (table 4.2; fig. 4.3). Average internal IVDs of atrial and ventricular trabeculae, which ranged in cross-sectional area from ~ 300 µm2 to ~ 50000 µm2 (figs. 4.7 and 4.8), did not differ from average edge distances (fig. 4.9). Atrial and ventricular trabeculae were avascular if cross-sectional area was < 3500 µm2 and < 2700 µm2, respectively (figs. 4.7, 4.8). The number of vessels increased with increasing cross-sectional area of trabeculae (fig. 4.7), but vascular density was maintained (fig. 4.8).  4.5 Discussion  4.5.1 Methodology  This is the first quantitative evaluation of coronary microvascular morphology in the heart of any elasmobranch and only the second study for a fish species.  Moreover, the data generated here for rainbow trout can be compared directly with earlier data for the same species (Egginton and Cordiner, 1997), which used a methodology applicable to a regular capillary architecture, such as that found in skeletal muscle and in the condensed compact myocardium of birds and mammals (Egginton and Cordiner, 1997; Gerdes, 2015; Scott et al., 2011).  My measurements of vascular density in the compact myocardium of trout did not differ  83 significantly from previous measurements of the capillary density in trout (Egginton and Cordiner, 1997) (Table 4.2).  Thus, I am confident the methodology used in this paper can be used to examine the vascular architecture of the highly branched vascularized trabeculae. Furthermore, this methodology can be adopted for other species of sharks, fish and reptiles in which the trabeculae contain coronary vessels (Durán et al., 2015; Kohmoto et al., 1997; Tota, 1989).  Where applicable, the methods used here followed the suggestions of Gerdes (2015) for improving mammalian cardiac morphometric data. The selection of tissue regions with minimal tissue separation and containing primarily circular vessel cross-sections free of red blood cells indicate adequate perfusion which increases result consistency and reduces error (Gerdes, 2015). Differences in the contraction state of the heart, however, can contribute to variation in vascular density observed between individuals of the same species (Gerdes, 2015). If individuals varied in contraction state at the time of fixation, vascular density could be slightly over or underestimated, with more contracted hearts having lower vascular densities due to the capillaries being further apart. Although this can be corrected for in mammals, using resting sarcomere length, such data are unavailable for the species studied here. The trout and dogfish hearts in this study all appeared to arrest in diastole. Following the perfuse fixation and sectioning, tissue shrinkage of up to 30% can result from dehydration in ethanol. This is unavoidable for tissue staining procedures (Gerdes, 2015; Preece, 1972), but a constant for histological practices.  Therefore, my comparisons between species and tissues are valid because all tissues were treated in the same manner. Although other methods for  84 estimating capillarity (Hudlicka et al., 1992), an advantage of my methodology is that it estimated vessel dimensions as well as vascularity and oxygen diffusion distances of dilated vessels across species. Several mathematical models exist that integrate morphometric data to quantitatively estimate tissue oxygen delivery (Egginton, 1990).  However, these models only estimate the maximal capacity for oxygen delivery. This is because functional characteristics, such as local oxygen tensions, are difficult to assess. Here I use IVD and edge distances as a measure of oxygen diffusion distance, which are easy measurements to compare across species that are related to the rate of oxygen diffusion.   4.5.2 Morphometrics  While vascularization in the ventricular and atrial trabeculae of shark hearts is well established (Tota, 1989), an unexpected discovery here was the impressive level of coronary vascularity. Indeed, vascularity was 383 ± 72 mm2 and 352 ± 73 mm2 for the ventricular and atrial trabeculae, respectively. In contrast to my hypothesis, higher vascular densities were found in the spongy tissues when compared to the compact tissues. The functional implication of this is that the shark heart may primarily depend on the coronary artery for its myocardial oxygen delivery to the conus arteriosus, the entire ventricle, atrium and sinus venous.  This conjecture is further supported by the measurements of IVD and edge distances, which are both estimates of oxygen diffusion distances. While teleost and sharks both have primarily spongy ventricles, the spongy myocardium of teleosts is  85 typically arranged into avascular sheets, which not only enhances the surface area to volume ratio in contact with the luminal supply but also decreases maximal diffusion distances (Pieperhoff et al., 2009). This is different from the tubular and highly vascularized trabeculae of dogfish, where their radii would otherwise exclude an effective luminal supply of oxygen based on the maximum IVD measured in this study (fig. 4.5 and 4.7) (Santer, 1985). The lack of statistically significant (P < 0.05) relationship between trabecular edge distances and trabecular IVDs indicates that the luminal supply plays a small role in routine myocardial oxygen delivery to only the outer annulus (20 µm) of the trabeculae (fig. 4.8). This is inconsistent with what has been observed in mammals where the capillary distances from the edge of the trabeculae significantly exceed IVDs within the trabeculae (Goo et al., 2009). Similar to mammals, however, dogfish trabeculae that are sufficiently small are avascular. The radii of these trabeculae do not exceed that of the maximal IVDs (41 µm) measured within the dogfish myocardium, indicating that there is a maximal effective oxygen diffusion distance within the dogfish heart regardless of if oxygen is diffusing from the coronary or luminal oxygen supplies (figs. 6, 7). Furthermore this means that as trabecular size increases so does its dependence on the coronary circulation. Assuming circular trabeculae, the range of trabeculae diameter calculated using trabeculae area measured in this study is similar to that measured previously (Santer, 1985) and indicates a representative spread of trabecular size is presented in this paper. Although vascular density is higher in the trout, larger vessels in the dogfish myocardium result in the species having similar IVDs and thus oxygen diffusion  86 distances. Minimal IVD in the trout compact and the myocardial tissues of the dogfish heart is 2.9 µm, indicating that this is likely the minimal distance possible (fig. 4.4). Maximal distance also appears to be similar within the trout and dogfish myocardial tissues. This suggests that oxygen diffusion through the myocardium of both teleosts and elasmobranchs is such that tissue further than ~20 µm from an oxygen source does not receive oxygen at a rate that can fuel maximal cardiac oxygen demand. Both the variation in inter-capillary distances and maximal inter-capillary distances in the compact myocardium of mammalian hearts are typically smaller. However, average inter-capillary distances in rats exposed to sea level oxygen tensions are 17 µm, similar to the average IVDs in tissues of the dogfish myocardium (Bourdeau-Martini et al., 1974; Henquell et al., 1977).   The differences in vessel diameter, which result in similar IVD between trout and dogfish, correspond to the diameters of red blood cells in the respective species (Table 4.2) (Egginton and Cordiner, 1997). The most common diameter vessel measured in the trout myocardium is ~9 µm (table 4.2; fig. 4.5) and this corresponds well with the diameter of a teleost red blood cell (~8 µm) (Nikinmaa, 2012). The most common vessel diameter in the dogfish myocardium is ~12 µm (table 4.2; fig 5) across all myocardial tissues, which again corresponds to the size of an elasmobranch red blood cell (~12 µm) (Nikinmaa, 2012). Although vessel diameter was not found to be significantly different between the two species, vessel area was significantly larger for dogfish tissues in comparison trout.    87 4.5.3 In vitro coronary perfusion rates in trout and dogfish The coronary circulation in trout is supplied by a single artery, which can be cannulated prior to it dividing to the cardiac outflow tract (bulbous arteriosus). The in vitro coronary flow rates and resistances in this artery measured here compare well to previously measured flows  in excised, beating trout hearts (Farrell and Graham, 1986). In dogfish, it was only possible to cannulate one of the possible 2-3 major branches of the coronary artery located on the conus arteriosus. However, coronary flow rates and resistances measured in this study were similar to in vivo measurements of total coronary flow and coronary resistance in anaesthetized sandbar sharks (Tables 4.1, 2.1 and 2.2). This similarity, in addition to red blood cell clearance in the dogfish coronary vasculature, indicates that there are many anastomoses between the major branches of the coronary artery within conal tissue that allow for flow distribution throughout the dogfish heart (fig. 4.9).  Although coronary flow rates between trout and dogfish are similar with regard to absolute perfusion rates and coronary perfusion rates (table 4.1), this is not indicative of similar maximum myocardial oxygen demands per gram of ventricular tissue. This is because the coronary circulation is perfusing a smaller proportion of the heart in the trout (40%) when compared to the dogfish (100%). Furthermore, although the myocardial oxygen demand (~ 0.25 µL O2 s-1 g-1 Mv) required to generate 1 mW of power at 15°C (Davie and Franklin, 1992; Graham and Farrell, 1990) is similar in the coronary of ligated hearts of trout and dogfish, these values also do not indicate similar myocardial oxygen demands. This is because the estimates of myocardial oxygen consumption were calculated from the drop in  88 oxygen content of cardiac perfusate traveling through lumen of the excised hearts. When the proportion of the myocardium in contact with the cardiac perfusate is taken into account myocardial oxygen consumption in the trout is estimated to be almost double that of the dogfish per gram of heart.  The higher oxygen demand of the trout myocardium in comparison to the dogfish myocardium may be due to higher routine heart rates in trout. Rainbow trout held at 10°C have heart rates of 50 bpm, twice that of similarly sized dogfish at 10°C (20 bpm; Sandblom et al., 2009; Thorarensen et al., 1996). As heart rate is related to increases in myocardial oxygen demand and CBF, species with higher routine heart rates should have higher routine myocardial oxygen demands (Duncker and Bache, 2008; Tomanek, 2012). The difference in heart rate corresponds well with the measurements for vascular density. This is consistent with my hypothesis that vascular density should be higher in the ventricular compact of trout hearts, because they generate more work with a higher blood pressure and have a faster heart rate when compared to dogfish hearts.  However, studies on additional species are required to more adequately address the question of whether the coronary supply is related to heart rate, blood pressure, or other physiological and environmental variables.   4.5.4 Conclusions The unexpected finding that the spongy myocardium of the dogfish ventricle is likely supplied with oxygen predominately from the coronary circulation has a major implication for our current view of the evolution of the coronary supply in  89 mammals. Although the morphology of the coronary artery differs across vertebrates it appears cartilaginous fish were the first to utilize the coronary artery to oxygenate the entire myocardium. Furthermore, the over two-fold higher vascular densities in the trout compact when compared to that of the dogfish may indicate that vascular density is linked with heart rate or pressure generation across vertebrate species.    90 Table 4. 1: Average in vitro coronary perfusion rate and resistance at physiological input pressures in dogfish and rainbow trout. Species Absolute perfusion rate  (ml min-1) Coronary Perfusion rate (ml min-1 kg-1) Coronary input pressure (cm H2O) Coronary resistance (cm H2O min kg ml-1) Dogfish 0.92 ± 0.23 0.45 ± 0.13 30 76 ± 24 Trout 0.89 ± 0.13 0.49 ± 0.03 40 110 ± 25  Values are mean  ±  s.e.m for N= 4    91 Table 4. 2: Average vascularity and vessel dimensions in the compact and spongy myocardial tissue of dogfish and rainbow trout. Species Chamber  Tissue Vascular density (#/total area mm2) Inter-vascular distance (μm) Vessel diameter (μm) Vessel area (μm2) Dogfish atrium spongy 352 ± 73bc 15.4 ± 0.8 13.1 ± 1.4 227.4 ± 21.0 c ventricle spongy 383 ± 72b 17.3 ± 0.2 10.0 ± 0.7 131.6 ± 8.7 ab conus compact 197 ± 63a 16.8 ± 1.4 13.2 ± 1.2 194.2 ± 24.3 bc ventricle compact 226 ± 45ac 17.1 ± 2.4 11.3 ± 1.1 170.3 ± 20.3 bc Trout ventricle compact 766 ± 108d 12.6 ± 1.0 8.8 ± 0.4 117.0 ± 10.7a Troutŧ ventricle compact 1194 ± 178d - - -   ŧData taken from Eggington and Cordiner 1997 Values are mean ± sem for N= 4 Statistically significant differences (P<0.05) within columns are denoted by different letters     92   Figure 4. 1: Cross sections of perfused fixed blood vessels in myocardial tissue in the rainbow trout (A, B: O. mykiss) and Pacific dogfish (C, D, E, F: S. suckleyi) at 400x magnification. (A) compact ventricular myocardium of trout, (B) spongy ventricular trabeculae of trout, (C) compact ventricular myocardium of dogfish, (D) ventricular trabeculae of dogfish, (E) conal myocardium of dogfish, (F) atrial trabeculae of dogfish. The black scale bar is 50 µm.   93   Figure 4. 2: Corrosion cast and micro-CT images of the blood vessels in the dogfish heart.  A) A micro CT-image of a Microfill cast of the coronary circulation around the junction of the conus arteriosus (con) and ventricle (vent). B) A Mercox cast of a the microcirculation in a single trabecula of ventricular spongy myocardium. C) A micro-CT image of a cross-section through the ventricle and atrium (atr) of dogfish heart. D) A micro-CT image if a cross-section through the ventricle and atrium of a trout  A B C D vent con vent vent atr atr  94   Figure 4. 3: Average vessel area and vessel wall thickness or the myocardium of dogfish. Different symbols represent different individual sharks (n=4) and different colours denote different cardiac tissues.   95   Figure 4. 4: Frequency distribution of vessel area for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts.  0.0000.0030.0060.0090.0120 300 600 900 1200Vessel area (m m)FrequencyTissueDogfish Atrial spongyDogfish Ventricular compactDogfish Ventricular spongyDogfish Conal compactTrout compact 96    Figure 4. 5: Frequency distribution of inter-vascular distance for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts.  0.0000.0250.0500.0750 10 20 30 40 50Intervascular distance (m m)FrequencyTissueDogfish Atrial spongyDogfish Ventricular compactDogfish Ventricular spongyDogfish Conal compactTrout compact 97    Figure 4. 6: Frequency distribution of vessel diameter for the various myocardial tissues of dogfish (n=4) and rainbow trout (n=4) hearts.  0.000.050.100.150 10 20 30 40 50 60Vessel diameter (m m)FrequencyTissueDogfish Atrial spongyDogfish Ventricular compactDogfish Ventricular spongyDogfish Conal compactTrout compact 98   Figure 4. 7: A comparison of the vascularity of spongy myocardium of the atrium and ventricle of dogfish (N=4). Number of vessels is presented as a function of trabecular area in the ventricular spongy (R2 = 0.7; P <0.001) and atrial spongy (R2 = 0.3; P < 0.001) myocardium. Note there are a number of trabeculae with a small area that had no blood vessels.   99   Figure 4. 8: A comparison of the vascular density of spongy myocardium of the atrium and ventricle of dogfish (n=4).   Number of vessels is presented as a function of trabecular area. Note there are a number of trabeculae with a small area that had no blood vessels.   100    Figure 4. 9: Frequency distribution of inter-vascular distances and edge distances in the atrial and ventricular trabeculae of dogfish (n=4).  0.000.020.040.0610 20 30Distance (m m)FrequencyTrabeculaAtrial edge distancesAtrial internal distancesVentricular edge distancesVentricular internal distances 101 Chapter 5: Linking the morphology of the coronary circulation with physiology and temperature across shark species   5.1 Summary   Using the methodology developed in chapter 4, the hearts of five additional species of elasmobranch were examined in order to investigate functional relationships between their coronary morphology and their physiological phenotype across these species. I further sought to determine if coronary morphology showed evidence of having evolved in response to different thermal regimes. Results indicate that the spongy and compact tissues differed with regard to vascular density and inter-vascular distance across species. It appears that tissues and species are subjected to different combinations of selective pressures with temperature potentially playing some role as an agent of selection to shape coronary morphology.  5.2 Introduction  Elasmobranchs live in a wide range of environments and have diverse life history strategies, yet all are known to have a coronary circulation (Tota, 1989). However, the degree to which sharks rely on their coronary circulation for  102 myocardial oxygen delivery is largely unknown. Determining the relative importance of the coronary circulation to myocardial oxygen supply in sharks could illuminate the evolutionary history of myocardial oxygen delivery for derived vertebrate groups.  Here I use morphological measurements of vascular density, inter-vascular distances (IVD) and vessel size to investigate the importance of the coronary circulation in myocardial oxygen delivery across six shark species.  Rate of oxygen delivery can be influenced through variation in morphological and/or physiological phenotypes via changes in oxygen perfusion and diffusion rates. However, the degree of variation in morphological traits may be restricted by physiological parameters and vice versa. Therefore exploring the relationship between morphological and physiological variables gives an insight into the co-variation between traits and ultimately provide inferences to their ability to evolve independently. Here I seek to determine if significant relationships between variables indicate functional relationships between coronary morphology and myocardial oxygen demand and supply.    A phylogenetically corrected correlation between environmental variables and phenotypes allows some inference of the selective pressures that shape phenotypic evolution. An environmental variable that has been shown to have a strong effect on the evolutionary trajectory of ecothermic organisms is temperature, and specifically temperature may be a key driver of cardiovascular adaptation due to its relationship with metabolism and heart rates (Gillooly et al., 2001; Lillywhite et al., 1999). A previous study on temperature acclimation in rainbow trout  103 (Oncorhynchus mykiss) demonstrated that a higher acclimation temperature leads to an increase in myocardial capillary density (Egginton and Cordiner, 1997) despite a smaller relative ventricular mass (Gamperl and Farrell, 2004). Antarctic ice fish, which experience extremely low environmental temperatures, have abnormally large vessel diameters relative to other fishes (Sidell and O’Brien, 2006). Yet, little is known about how temperature may affect the evolution of capillary density and other cardiovascular phenotypes. In this chapter I explore the relationship between morphological phenotypes of the coronary circulation and temperature to determine if sharks show evidence of having evolved morphologically in response to different thermal regimes.   5.3 Methods  5.3.1 Animals Shark species were captured using hook-and-line fishing gear.  The species were the Pacific spiny dogfish sharks (Squalus sucklei), leopard sharks (Triakis semifasciata) and sandbar sharks (Carcharhinus plumbeus). Species captured using longline fishing gear with bated hooks included the blue shark (Prionace glauca), mako sharks (Isurus oxyrinchus) and Greenland sharks (Somniosus microcephalus). The hearts from sandbar sharks (N= 5), leopard sharks (N =5) and dogfish sharks (N=4) were selected from animals used in chapters 2, 3 and 4. Greenland sharks (N  104 = 5) were 337 - 464 kg and averaged 407 ± 37 kg while mako sharks (N = 5) ranged from 11 - 97 kg and averaged 39 ± 13 kg and the blue shark (N = 1) was 25 kg.   5.3.2 Tissue perfusion, fixation, sectioning, staining, imaging and morphological measurements Following euthanization of the sharks, coronary artery perfusion for fixation through to the imaging of cardiac sections and subsequent collection of morphological measurements were conducted in an identical fashion to that reported in the methods section of chapter 4 and using the same criteria and assumptions. Morphological variables included in this study were vascular density, inter-vascular distance (IVD), vessel diameter, vessel area and trabecular edge distances. These morphological measurements were collected for the myocardial tissues types (N=4); the conal compact, ventricular compact, ventricular spongy and atrial spongy for all species (N = 6).   5.3.3 Calculations and Statistical analysis  Significant differences in morphological measurements (vascular density, IVD, vessel area and diameter) between the myocardial tissue types (conal compact, ventricular compact, ventricular spongy, and atrial spongy) within a species were determined using a single-factor analysis of variance (ANOVA) with a Holm-Sidak post-hoc test performed on log-transformed data. A student’s t-test was performed on log-transformed data in order to detect any significant differences in edge  105 distances between the atrial and spongy tissues within a species. A two-way ANOVA with a Holm-Sidak post-hoc test was used to detect significant differences in mean vascular density, IVD, vessel area and diameter between species, tissues and the interaction between them.  As edge distances did not vary significantly within a species the edge distances from both the ventricular and atrial spongy tissues were averaged to calculate mean spongy edge distances for each species.  Mean spongy vascular density, IVD and vessel diameter for each species were calculated similarly due to their similar tissue morphology. Mean compact tissue values were calculated by averaging conal and ventricular compact tissue values for vascular density, IVD and vessel diameter and area. A surrogate for the atrial oxygen transport capacity was calculated as the product of cardiac output and arterial oxygen content.  Shark species varied considerably in body mass.  Therefore, all morphological variables were regressed against mass to determine the significant size effects on coronary morphology. For the morphological variables that were significantly related to body mass, residuals were used to remove the effect of size from the relationships. Regressions between morphological and physiological phenotypes were determined using the least squares regression analysis in SigmaStat 10.0  (Aspire Software International, Ashburn, VA, USA). A phylogenetic independent contrast (PIC) analysis using PDAP:PDTREE (v. 1.16) in Mesquite (Maddison and Maddison 2004) was used to remove the effect of phylogeny from the regressions. The shark phylogeny used for PIC was based on 595 species of  106 elasmobranch using the NADH2 mitochondrial gene supplied by Gavin Naylor (Naylor et al., 2012) . Statistical significance was set at P ≤ 0.05 and all values are reported as means ± s.e.m.   5.4 Results  5.4.1 Intraspecific variation in morphological phenotypes Tissue specific mean values for vascular density, IVD, vessel diameter and area as well as edge distances are listed in table 1 for the 6 shark species. Vascular density varied significantly between tissue types within the hearts of all species apart from the mako shark, which had uniform vascular density across all tissue types (table 5.1). IVD, vessel diameter and edge distance (distance from the edge of the trabeculae to the nearest internal vessel within the trabeculae) did not vary between tissue types within species and there was no significant difference between IVD and edge distances in the atrial or ventricular spongy tissues within species. Although vessel area did vary significantly between tissue types within the hearts of 3 species (table 5.1), vessel diameter and area co-varied (P<0.001, R2=0.96). As vessel area and vessel diameter related similarly to other phenotypes across taxa vessel diameter was chosen to represent vessel size for subsequent analysis.    107 5.4.2 Interspecific variation in morphological phenotypes Tissue specific mean values for IVD, vessel diameter and vascular density within myocardial tissue types across taxa are shown in figure 5.1 (A-O). IVD showed a significant difference between species (DF=5, F=13.95, P<0.001), with that for the Greenland sharks having 33% greater mean IVD than sandbar sharks (fig 5.1). Different tissues also showed significant variation in IVD (DF=3, F=4.83, P=0.005), with the spongy tissues having higher IVDs than compact tissues as determined by a Holm-Sidak post-hoc test (fig. 5.1). There was no significant interaction between species and tissues with regard to IVD or between IVD and species body mass.  Vessel diameter showed a significant effect of species (DF=5, F=23.49, P<0.001) but not tissue type (DF=3, F=0.901, P=0.446) or the interaction between species and tissue (DF=15, F=1.289, P=0.237) (fig. 5.1). Post-hoc analysis determined that vessel diameters in the Greenland shark were significantly larger, while sandbar sharks had significantly smaller vessel diameters when compared to other shark species (fig. 5.1).  Vessel diameter in the spongy myocardium, but not the compact myocardium, was significantly related to species body mass (R2 = 0.7, P=0.052).  Measures of vascular density showed a significant difference between species (DF=5, F=26.85, P<0.001) and tissue types (DF=3, F=4.55, P=0.007) (fig. 5.1). There was also a significant interaction between species and tissue type (DF=15, F=5.22, P<0.001) that was driven primarily by a greater difference between  108 species in vascular density of the spongy tissues (fig. 5.1).  Vascular density in the spongy myocardium, but not the compact myocardium, was significantly related to species body mass (R2 = 0.7, P = 0.047).  Post-hoc tests revealed that the spongy and compact tissues differed with regard to IVD and vascular density with the compact tissue having shorter IVD and higher vascular densities when compared to spongy tissues. However, there was no significant effect of tissue type on vascular diameter. As a result the conal and ventricular compact tissues were combined in order to generate mean compact tissue values for morphological measurements. Similarly, the ventricular and atrial spongy tissues were combined into mean spongy tissue values for morphological measurements. These mean values for the combined spongy tissues and compact tissues were used in the subsequent analysis.  5.4.3 Interspecific comparison between morphological and physiological phenotypes  Across species, variation in vascular density was not significantly related to IVD in either the spongy or compact tissues (fig. 5.2 A, B) even when corrected for species size. Vessel diameter increased significantly with IVD for both tissue types (fig. 5.2 C, D), however, when the relationship was corrected for species body mass the relationship between spongy vessel diameter and IVD was no longer significant (R2 = 0.2, P = 0.330). Mean spongy IVD was however, significantly positively related to edge distances (fig. 5.3). Additionally, both spongy edge distance and IVD were  109 significantly and positively related to the size of the largest avascular trabeculae in spongy myocardial tissues across species (fig. 5.4). Species with higher average IVD and edge distances also possessed the largest avascular trabeculae, with atrial trabeculae attaining a larger size than avascular ventricular trabeculae (fig. 5.4). Neither IVD, vessel diameter (data not shown) nor vascular density in spongy and compact tissue were significantly related to either mean ventral aortic blood pressure (fig. 5.5 A-D), cardiac output (fig. 5.5 E-H), whole animal metabolic rate (fig. 5 I-L), or a surrogate for arterial oxygen transport, the product of cardiac output and arterial oxygen content (fig. 5.5 M-P) across the four shark species for which data were available. However, a number of trends were observed across ectothermic shark species (N=3, excluding the regionally endothermic mako shark). There was a trend towards increasing vascular density and decreasing IVD in response to higher mean ventral aortic pressures and metabolic rate across the ectothermic species in the compact tissues (fig. 5.5 C, D and G, H). Additionally, mean spongy IVD in the ectothermic species appeared to increase with cardiac output while vascular density decreased (fig. 5.5 E, F).  A significant relationship was found between P50, the partial pressure of oxygen at which 50% of haemoglobin is saturated, and the mean vascular density as well as the IVD of spongy tissues across species. This relationship was significant when corrected for the phylogenetic relationship among species using PIC (fig. 5.6 A,B). However, this relationship was not observed in the compact tissues (fig. 5.6 C,D).  110  5.4.4 Interspecific comparison between morphological phenotypes and temperature Mean IVD, vascular density and vessel diameter in both the spongy and compact tissues were related to species-specific maximum environmental temperatures in figure 5.6 (A-D). Species that encounter higher maximum environmental temperatures have significantly shorter IVD’s and smaller vessel diameters within the compact tissues, even when corrected for the phylogenetic relationship among species using PIC (fig. 5.7 A, C). Although IVD was not related to maximal temperature in the spongy tissues, the relationship between vessel diameter and maximum environmental temperature was significant even when corrected for species weight (fig. 5.7 D, F). However, following PIC analysis, this relationship only approached significance (P=0.058; fig. 5.6 F).  Vascular density in the spongy and compact tissue was not related to maximum environmental temperature (fig. 5.7 B, E). However, when the relationship between spongy vascular density and temperature was corrected for body mass using residuals the relationship was significant (R2 = 0.7, P = 0.051).   111 5.5 Discussion  5.5.1 Myocardial oxygen supply in shark species Combined, the results of this study indicate that sharks rely primarily on the coronary circulation for myocardial oxygen delivery. This conclusion is supported by the extensive vascular density observed for all myocardial tissue types in six shark species. Thus, despite shark hearts being primarily composed of ventricular spongy tissue, the role of the luminal oxygen supply appears limited to providing the outer annulus of the spongy trabeculae and relatively few number of sufficiently small trabeculae. Significant correlations between IVD, edge distances and the area of the largest avascular trabeculae across species may indicate that there are species-specific differences in maximal oxygen diffusion distances from both the luminal and coronary supply. Furthermore, species with larger IVD and edge distances also possessed the largest avascular trabeculae.   As the coronary circulation between these shark species differ significantly in morphological phenotypes, perhaps in response to differing selective pressures, I was able to both investigate possible functional linkages between morphological and physiological phenotypes, and explore possible drivers of evolution within the coronary circulation.    112 5.5.2 Functional links between morphological and physiological phenotypes Various morphological phenotypes of the coronary circulation can affect the rate of myocardial oxygen delivery by affecting the rate of tissue oxygen perfusion, diffusion or both. Such morphological phenotypes of the coronary circulation as vascular density, IVD and edge distances are thought to be representative of oxygen diffusion distances. Higher capillary densities are usually associated with shorter inter-capillary distances and thus oxygen diffusion distances (Egginton et al., 1988; Hudlicka et al., 1988; Krogh, 1919a; Krogh, 1919b). In contrast, larger vessel sizes are usually associated with higher perfusion rates (White et al., 1998) as small changes in vessel diameter can cause large changes in flow rates (Poiseuille’s law).  Contrary to the expected inverse relationship between vascular density and IVD, the analysis showed that these phenotypes were not significantly related to each other across species (fig. 5.2). This likely resulted from IVD’s increasing with mean vessel diameter across shark species. As IVD is a proxy for oxygen diffusion distances, the corresponding increase in vessel diameter may potentially help to offset the increase in diffusion distance (resulting from both an increase in IVD and vessel diameter) through a possible increase in mean blood flow by reducing coronary vascular resistance. Thus, as diffusion distance increases, so does the rate of arterial oxygen transport to the myocardium. Alternatively, increasing IVD with vessel diameter may be necessary to prevent a “swiss cheese effect” - having large holes (vessels) reduces the amount of cheese (muscle) per unit area. As the reduction in area available for myocardial muscle could negatively impact force  113 production, increasing IVD with diameter could serve to ensure the maintenance of cardiovascular function while increasing oxygen perfusion.  As variations in one or more morphological phenotypes may significantly impact myocardial oxygen demand or delivery rates, a general hypothesis is that the rate of myocardial oxygen delivery is ultimately affected by both the structural design of the coronary circulation and the myocardial tissue. Indeed, myocardial tissue type was significantly linked to the morphology of the coronary circulation with regard to tissue specific differences in IVD and vascular density, with IVDs being significantly shorter in compact tissue types when compared to spongy tissue types across species (table 1). This finding may indicate a difference in tension generation between the compact and spongy tissues. The compact myocardium is likely required to generate higher tensions as a result of its outermost position in the heart and outflow tract (Laplace’s law). Consequently, oxygen demand in the compact may exceed that of the spongy and thus shorter IVDs could serve to increase oxygen delivery rates to the compact tissue.   Furthermore, physiological or morphological traits that impact myocardial oxygen consumption or delivery may be linked with various morphological phenotypes of the coronary circulation. I hypothesized that species with higher myocardial oxygen demands would have higher degrees of myocardial vascularization (represented by vascular density) and shorter diffusion distances (represented by IVD). Mean ventral aortic blood pressure, cardiac output and metabolic rate were selected to determine any possible relationships to  114 morphological phenotypes as these have been shown to increase myocardial oxygen consumption across vertebrates (Agnisola et al., 1998; Berglund et al., 1958; Davie and Farrell, 1991a; Davie and Franklin, 1992; Duncker and Bache, 2008; Tomanek, 2012). Although no significant links existed between physiological phenotypes and any morphological variable across species, the non- heterothermic species showed a trend between compact myocardium and mean ventral aortic pressure that was consistent with the hypothesis. An additional trend was observed in the spongy tissues of the ectothermic species such that as cardiac output increased, IVD increased and vascular density decreased (fig. 5.5). As 3-4% of cardiac output (tables 2.1 and 2.2) is directed towards the myocardium and the spongy myocardium makes up the majority of myocardial mass, this increased coronary perfusion with higher cardiac outputs may allow for larger diffusion distances.  The endothermic mako sharks may lie outside of these relationships due their comparatively high arterial oxygen transport capacity (fig. 5.5). Measurements of arterial oxygen content in mako sharks are over twice those found in the other ectothermic shark species (Lai et al., 1990a; Lai et al., 1997; Perry and Gilmour, 1996).  The results were consistent with those in chapters 2 and 3 where coronary blood flow was not significantly linked with mean ventral aortic pressure and only weakly associated with cardiac output. As maximal heart rates in the majority of shark species have not been documented it was not possible to test if morphological phenotypes were correlated to maximal heart rates, as was the case with coronary blood flow (figs. 2.5 and 3.4).    115  There was, however, a significant relationship between vascularization and P50 found for the spongy myocardium but not the compact myocardium. As the spongy myocardium has a relatively high surface area to volume ratio in contact with luminal blood, one would expect this relationship to reflect a functional link between oxygen unloading and oxygen diffusion distance or vascularity. However, the increased myocardial diffusion distance associated with increasing IVD and vessel diameter was not associated with a lower haemoglobin oxygen affinity. Instead, the reverse relationship was found. This link could be an indirect result of both phenotypes responding significantly to an external selective pressure (and thus inadvertently linked), epistatic interaction, linkage disequilibrium or other factors that are beyond the scope of this analysis. Temperature not size appeared to be a key external selective pressure for variation in P50 as they were not significantly related to maximum temperature, even when excluding the heterothermic mako shark.   5.5.3 Morphological phenotypes and environmental temperature Fluctuations in environmental temperature have been linked with variation in both the physiology and morphology of the cardiovascular system (Butler and Metcalfe, 1988; Clark et al., 2008; Driedzic and Gesser, 1994; Eliason et al., 2013; Farrell et al., 1988b) and specifically the coronary circulation (Barron et al., 1987; Costa et al., 2015a; Costa et al., 2015b; Egginton and Cordiner, 1997; Muñoz-Chápuli et al., 1994) in teleosts and elasmobranchs. Yet, there is a distinct lack of  116 information on how temperature may influence the evolution of the coronary circulation. I hypothesized that temperature may act as a selective pressure driving the evolution of the coronary circulation due to the correlation of temperature with coronary flow and heart rate (Ch. 3), blood viscosity (Bentley et al., 1993; Nguyen et al., 2004; Snyder, 1971) and oxygen diffusion rates (Bentley et al., 1993).  As blood flow rates are inversely proportional to blood viscosity (Poiseuille's law) reductions in temperature decrease both oxygen diffusion and perfusion. While significant reductions in IVD may help to offset the reduced oxygen diffusion rates, small increases in vessel size may counteract the reduction in perfusion caused by increased blood viscosity. As IVD and vessel diameter are inversely related the results (fig. 5.7) indicate that changing vessel diameter and increasing perfusion rates are functionally more efficient for modulating myocardial oxygen delivery, even when diffusion distances are concurrently increased (fig. 5. 7).  This study found a significant linear relationship between vessel diameter and maximum environmental temperature in the compact tissue, which suggests that animals inhabiting colder temperatures have larger vessels (fig. 5.7 C).  This phenomenon is also observed in Antarctic ice fish, which have particularly large vessels for a teleost (Sidell and O’Brien, 2006). Yet, this relationship appears driven by the significantly larger vessels of the Greenland shark and may not reflect a linear relationship between temperature and vessel size (fig. 5.7 C). Future work to determine whether increased vessel size is found only in fish inhabiting  117 exceptionally cold environments or whether more moderate changes in environmental temperature cause a similar pattern would be useful.   5.5.4 Conclusions  Morphological phenotypes of the coronary circulation appear to be functionally linked to balance the diffusion and perfusion of oxygen such that overall myocardial oxygen delivery meets myocardial demand. Although measurements of vascular density are thought to be representative of oxygen supply, my results suggest IVD and vessel diameter are more informative variables to use when comparing between taxa. Temperature appears to play some role in acting as selective pressure to shape the overall coronary morphology, however, it does not explain all variation in morphological phenotypes. This suggests that other selective pressures or phenotypic drift may also play a role.    118 Table 5. 1: Average vascularity, vessel dimensions and  inter-vascular distances in 6 species of shark. Species Tissue type Vascular density (# mm-2) Inter-vascular distance (μm) Vessel diameter (μm) Vessel area (μm2) Edge distances (μm) Greenland conal compact 383 ± 51a 21.7 ± 1.8 14.8 ± 2.1 374.9 ± 82.9 a  ventricular compact 271 ± 18 a 20.2 ± 2.2 28.2 ± 9.7 847.4 ± 287.0 b  ventricular spongy 137 ± 32 b 27.1 ± 2.4 19.3 ± 1.6 649.6 ± 77.4 b 32.2  ± 5.5 atrial spongy 65 ± 0 c 32.1 ± 1.3 31.3 ± 5.3 1931.9 ± 985.8 c 44.8 ± 13.7 Dogfish conal compact 197 ± 63 a 16.8 ± 1.4 13.2 ± 1.2 194.2 ± 24.3 bc  ventricular compact 226 ± 45 ac 17.1 ± 2.4 11.3 ± 1.1 170.3 ± 20.3 bc  ventricular spongy 383 ± 72 b 17.3 ± 0.2 10.0 ± 0.7 131.6 ± 8.7 ab 16.1 ± 0.9  atrial spongy 352 ± 73 bc 15.4 ± 0.8 13.1 ± 1.4 227.4 ± 21.0 c 19.7 ± 3.5 Mako conal compact 320 ± 49 11.9 ± 0.5 12.1 ± 0.7 128.1 ± 12.1 a  ventricular compact 320 ± 81 10.0 ± 1.55 9.8 ± 0.2 80.6 ± 9.7 b  ventricular spongy 426 ± 48 13.9 ± 2.7 10.3 ± 0.6 95.7 ± 11.5 ab 14.2 ± 2.5 atrial spongy 309 ± 22 14.1 ± 2.9 14.4 ± 3.8 109.8 ± 9.8 ab 14.2 ± 3.0 Leopard conal compact 450 ± 43a 12.3 ± 1.8 12.2 ± 0.5 141.5 ± 27.0  ventricular compact 528 ± 35a 12.8 ± 1.6 10.4 ± 0.5 99.6 ± 17.1  ventricular spongy 512 ± 28a 17.7 ± 2.5 10.9 ± 0.3 117.2 ± 10.3 15.6 ± 3.1 atrial spongy 288 ± 26b 19.5 ± 3.2 11.3 ± 0.4 157.7 ± 22.3 20.0 ± 3.6 Sandbar conal compact 438 ± 34 a 9.6 ± 0.9 7.9 ± 0.4 76.4 ±2.0  ventricular compact 618 ± 98 ab 11.1 ± 1.8 8.2 ± 0.7 87.3 ± 12.1  ventricular spongy 856 ± 130 b 12.9 ± 1.2 7.7 ± 0.8 83.2 ± 16.4 11.4 ± 1.6 atrial spongy 758 ± 81 ab 12.3 ± 3.1 8.1 ± 0.6 91.2 ± 18.2 13.8 ± 1.6 Blue conal compact 487 13.0 12.9 125.6  ventricular compact 410 14.7 12.4 59.4  ventricular spongy 414 26.8 11.5 174.3 17.7 atrial spongy 250 18.5 9.7 130.9 20.0 Values are mean ± s.e.m. Statistically significant differences within species (P<0.05) are denoted by different letters within columns for that species.  119   Figure 5. 1: Species specific mean inter-vascular distance, vessel diameter and vascular density in the conal compact (A-C), ventricular compact (D-F), ventricular spongy (G-I) and atrial spongy (J-L) heart tissue. Different symbols represent the means of the different myocardial tissue types; ☐  = conal compact,  = ventricular compact,  = ventricular spongy,  = atrial spongy. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1). Statistically significant differences between species (P ≤ 0.05) are denoted by different letters. Data is presented as mean ± s.e.m with specie arranged based increasing maximal environmental temperature from left to right.    120   Figure 5. 2: The relationship between mean vascular density and mean inter-vascular distances (A, B) and between inter-vascular distance and vessel diameter in spongy (A, C) and compact myocardial tissues (B, D) for hearts of 6 species of elasmobranch. Data is presented as mean ± s.e.m.     121   Figure 5. 3:  The relationship between inter-vascular distances and edge distances in the spongy myocardial tissues for different shark species. Values for the ventricular   atrial  () tissue types are plotted for each individual and this relationship is represented by the grey dashed line. The relationship between species specific mean values () is represented by the solid line. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1)  122  Figure 5. 4: The relationships between mean edge distance (A) and mean inter-vascular distance (B) with area of the largest avascular trabeculae in either the atrial (O) or ventricular () spongy myocardial tissues across shark species. Solid line represents the relation with atrial myocardium. Dotted line represents relationship with ventricular myocardium. N = 5 in all species apart from the dogfish (N = 4) and blue shark (N = 1).      123   Figure 5. 5: Mean inter-vascular distances and vascular densities in four species of shark in relation to their respective mean ventral aortic blood pressures (A- D), cardiac outputs (E -H), metabolic rates (I -L) and arterial oxygen transport capacity (M-P) in spongy and compact myocardial tissue.  Measurements of mean ventral aortic pressure were taken from chapters 2 and 3 for the sandbar shark and leopard shark, respectively.  Estimated cardiac output calculated in chapter 2 for the sandbar shark was also used. Additional literature values for mean ventral aortic pressure (Johansen et al., 1966; Lai et al., 1997), cardiac output (Lai et al., 1990b; Lai et al., 1997; Scharold et al., 1988) and metabolic rate (Brett and Blackburn, 1978; Dowd et al., 2006; Scharold et al., 1988; Sepulveda et al., 2007) were included where available for the species. The metabolic rates used were measured between 16- 21°C in all sharks aside from the dogfish, that was measured at 10°C. Data is presented as mean ± s.e.m.   124  Figure 5. 6:  The relationship between P50, IVD and mean myocardial vascular density in the spongy (A,B) and compact (C,D) myocardial tissues of 6 species of elasmobranch. Linear regressions (dashed lines) and significant phylogenetically independent contrast (PIC) correlations (solid lines) are plotted on species data that are presented as mean ± s.e.m. P50 values were obtained from the literature for sandbar sharks (Brill et al., 2008), leopard sharks (Lai et al., 1990a) and dogfish (Lenfant and Johansen, 1966). Other values were attained through personal communication with Diego Bernal (mako and blue sharks) and Neill Herbert (Greenland shark).    125  Figure 5. 7: The relationship between mean myocardial phenotypes of the coronary circulation and maximal environmental temperature in the compact (A-C) and spongy (D-F) myocardial tissue types across 6 shark species.  Linear regressions (dashed lines) and significant phylogenetically independent contrast (PIC) correlations (solid lines) are plotted on species data that are presented as mean ± s.e.m. Maximal temperatures were collected from active and passive acoustic tracking studies (Hight and Lowe, 2007) in addition to gill net and longline surveys (Merson and Jr, 2001) in addition to CTD scans of the oceanographic regions where sharks were captured (personal communication with NOAA and the survey crew of the Dana).   126 Chapter 6: General discussion and conclusions   The general objective of my doctoral research was to investigate the role of the coronary circulation in sharks, the most ancient extant vertebrate lineage with a coronary circulation. To address this general objective I measured the changes in coronary blood supply in response to variation in cardiovascular function. I then developed a methodology that allowed for the quantification of morphological phenotypes to investigate if variation in these phenotypes was linked to physiological phenotypes or environmental temperature in a way that would suggest a pattern of adaptation. The preceding chapters have clearly shown that the coronary circulation in sharks plays a crucial role in myocardial oxygen delivery, not only to the compact myocardium, but to the entire heart.  6.1 Role of the coronary circulation in sharks  Based on the observation of vascularized trabeculae in sharks, I predicted that the form and function of the coronary circulation in sharks would be linked to cardiovascular function across species. This does indeed appear to be the case and variation in coronary morphology and function across species also seems to be adaptive. Below I discuss the functional significance of the coronary circulation within and across shark species using both physiological and morphological data.    127 6.1.1 Coronary blood flow rates in relation to cardiac function  Myocardial contraction was found to exert sufficient extra-vascular compressive force to limit CBF primarily to diastole (figs 2.2 and 3.3). Furthermore, increases in heart rate resulted in reductions in the diastolic time fraction. Nevertheless, CBF was significantly correlated with heart rate such that 60 to 80% of the variation in CBF was linked to changes in heart rate in both bradycardic and trachycardic states respectively (figs. 2.5 B and 3.4). In vivo and in vitro data indicate this relationship is maintained primarily through a significant reduction in coronary resistance with increasing heart rate (figs. 2.4, 2.6 and 2.7). Due to the relationship between acclimation temperature and gill resistance in S. canicula (Butler and Metcalfe, 1988; Butler and Taylor, 1975; Short et al., 1979) it is possible that increases in coronary perfusion pressure contributed to elevating CBF as temperature and heart rate increased in the leopard shark (fig. 3.1 A, C). However, increases in estimated in vivo coronary perfusion pressure and measured in vitro coronary input pressures within the physiological range had insignificant impacts on CBF (fig. 2.7, 2.8 and table 3.1). This indicates there is some degree of auto-regulation in the coronary vasculature.  Although there is a strong positive correlation between heart rate and CBF, variation in stroke volume and ventral aortic pressure was not related to CBF in anesthetized sharks (fig. 2.6 C, D). As a result CBF was only weakly correlated with cardiac output and cardiac power output (fig 2.6 A, E). Assuming that these results were not related to the anaesthesia it is likely the energetic cost of increasing stroke volume, and thus any subsequent increases in cardiac output and power output  128 resulting from increases in stroke volume, is met by the luminal supply. The increased blood residency time and myocardial stretch associated with larger stroke volumes both serve to enhance time for oxygen diffusion and the area of tissue supplied from the luminal circulation. As trabeculae stretch with chamber filling, the accompanying reduction in trabecular diameter increases the proportion of spongy myocardial mass that can be supplied by the luminal oxygen supply. The lack of a significant difference between edge distances and IVDs indicates that the luminal circulation supplies the outer annulus of the trabeculae routinely, and perhaps increases supply with increasing stroke volume (tables 4.2 and 5.1).  Measuring CBF during exercise could elucidate the importance of CBF to myocardial function when stroke volume is the primarily variable for cardiac output. If the end systolic partial pressure of oxygen is indeed what regulates the extent of coronary vasodilation, as suggested in chapter 2, measuring CBF during exercise would reveal a positive relationship between CBF and stroke volume. The increased oxygen demand associated with increasing the number of cross-bridge formations at higher stroke volumes would result in a lower end systolic tissue oxygen pressure. With a near stable heart rate during exercise, the effect of an increased heart rate on CBF would be greatly reduced or removed.  However, measuring CBF, cardiac output and pressure on exercising sharks will have to wait until blood flow recording technology advances to a state that is robust enough to endure the antics and abrasive skin of exercising sharks.    129 6.1.2 Morphological phenotypes of the coronary circulation Although edge distances and IVD did not vary within species, both varied across species (fig. 5.1) maintaining a positive relationship to each other (fig. 5.3). Measurements of IVD and edge distance indicate there is species-specific maximal diffusion distance and thus only sufficiently small trabeculae of the spongy myocardium are dependant of the luminal supply. As trabecular size increases so does the trabecular dependence on the coronary circulation. A trabeculum that is ~5 times wider than the species’ maximal oxygen diffusion distance  (½IVD) is ~80% dependant on the coronary supply. Additionally, there are species-specific differences in maximal avascular trabeculae size that correlated with species-specific differences in edge distances and IVDs across species (fig. 5.4). Together these relationships suggest that maximal diffusion distances exist within species and vary across species. IVDs are further related to vascular diameter and area but not to vascular density (fig. 5.2). This indicates that there are also minimal IVD (figs. 4.5 and 5.2) that ensure muscle area is proportionally larger than vessel area within the heart, thus avoiding the “swiss cheese effect” and maintaining a balance between oxygen supply and demand. As vessel diameter ultimately affects vascular density, it may be more accurate to estimate maximal diffusion distances using vascular dimensions and IVDs rather than vascular density when comparing across species that differ in red blood cell size or average vessel area. This is supported by the observation that both vessel diameter and IVD were significantly related to a selective pressure (maximum temperature) while vascular density appeared unresponsive (fig 5.7).   130 Although there were trends within the ectothermic species linking coronary morphology to functional phenotypes indicative of myocardial oxygen demand or oxygen transport, there were no significant linkages between them (fig. 5.5). Furthermore, the compact and spongy myocardial tissues appeared to differ in their relationships with functional phenotypes and also temperature (fig. 5.7). These differences may result from either positional differences or proximity to the luminal circulation (table 5.1, figs. 5.1, 5.2 and 5.5). For example, across species there was a significant effect of tissue type on IVD such that the compact tissues had significantly shorter IVDs than the spongy tissues (table 5.1, fig. 5.1). In order for the spongy and compact tissues of the ventricle to exert similar pressures on the luminal blood, increased rates of oxygen diffusion may be required to fuel the higher tension generation required by the outer compact tissue. Further variation between the compact and spongy tissue phenotypes in relation to P50 and temperature may result from varying effects on arterial and venous blood properties (figs. 5.6 and 5.7). Even within the spongy tissues there was variation across species with avascular trabeculae in the atrial myocardium attaining larger sizes than avascular ventricular trabeculae (figs 4.7 and 5.4). This supports the idea that differences in oxygen delivery rates vary between chambers in relation to their external work rate.   6.1.3 Balancing oxygen perfusion and diffusion in functioning hearts.   Inferences on the overall function of the coronary circulation can be made from both physiological and morphological studies. However physiological  131 experiments only indicate average responses of the whole heart and morphological studies give insight into only a snap shot of the heart frozen in a particular state of contraction. In reality, the heart repetitively cycles through contraction and relaxation and the effects of this on various tissue types are not well understood. While myocardial contraction will decrease, and even reverse blood flow, in the major vessels of the coronary circulation (on which flow probes are located) it is difficult to delineate the possible variation in flow regimes though the different tissue or vessel types during contraction. While the outer compact tissues appear to constrict coronary vessels as it contracts around the inner spongy, trabeculae shorten during contraction. This shortening likely increases the tortuosity of the internal vessels and increases oxygen diffusion distances from the luminal circulation (edge distances). Increased tortuosity and decreased blood flow rates, however, will increase the time for oxygen diffusion from the coronary circulation. In contrast, trabeculae lengthening that occurs during relaxation/myocardial filling also lengthens the internal coronary vessels and reduces edge distances, thus increasing the myocardial area that can benefit from luminal circulation. Thus, the effects of the coronary and luminal circulation may vary with tissue type and the contractile state of the heart. Indeed it appears that the pattern/ rate of coronary blood flow and coronary morphology can be affected by the rate of contraction (figs. 2.2 - 2.6, 3.3 and 3.4), force of contraction (fig. 3.1 A), work of contraction (figs. 4.7 and 5.4) and the degree of tension generated during contraction (table 5.2; fig. 5.1). If myocardial dependence on the coronary circulation varies with the contractile state, this supports the idea that variation in cardiovascular function, such as the  132 modulation stoke volume or heart rate, will impact the role of the coronary and luminal circulations (fig. 2.6 B, C).   6.2 Evolution of the coronary circulation Although all vertebrates are united by the requirement to balance myocardial oxygen supply and demand, or die, the means by which vertebrate taxa accomplish this vary greatly (Farrell et al., 2012). Looking across vertebrate taxa, the role of the coronary circulation varies from non-existent to essential. Those taxa in which the coronary circulation never developed, or has been secondarily lost, rely on solely on the luminal circulation for this oxygen supply. Many of these species have completely spongy myocardium consisting of sheet-like trabeculae. In contrast, species that rely almost fully on the coronary circulation for myocardial oxygen delivery typically have highly compact hearts with a smaller degree of vascularized spongy tissue. However, sharks are part of the most ancient lineage to develop a coronary circulation and despite maintaining a predominately spongy heart, appear to be primarily dependent on the coronary circulation for oxygen.  Morphological phenotypes of the shark coronary circulation were found to vary with temperature – but other selective pressures almost certainly play a role in shaping coronary morphology and function.  For example, life history differences between species could have profound selective effects on aspects of the coronary morphology.  Studies on natural selection have often shown that extremes are the primary drivers of evolution (Boag and Grant, 1981). In addition, the evolution of  133 the coronary is constrained, as is the evolution of all complex phenotypes, by selective pressures on other aspects of the oxygen delivery system and the pleiotropic nature of the genes that underlie the these phenotypes (Barton, 1990). For example, the mako shark has one of the highest metabolic rates, but also a comparatively low P50 and short IVD (table 6.1).  The low affinity haemoglobin and short oxygen diffusion distances aid oxygen delivery diffusion to the body tissues to support a high metabolic demand and, all other things being equal, also leave venous blood with a higher oxygen tension when it reaches the heart. In contrast the lower metabolic rate of dogfish can be met by the comparatively higher P50 and the longer IVD (table 6.1). These differences also likely reflect adaptations to different habitats and life histories.   Here I found variation in a range of morphological and physiological phenotypes relating to myocardial oxygen supply across sharks.  I also found some evidence that this variation corresponds with differences in the thermal regimes they experience.   As such, our data suggests that the coronary morphology of sharks have evolved in response to local abiotic stressors.  The compact and the spongy myocardium appear to have adapted differently to temperature, perhaps due to differences in the functional linkages with other physiological phenotypes that influence oxygen delivery, such as metabolic rate and P50 (table 6.1).    134 6.3 Future directions  To more fully understand how different selection pressures shape variation in the coronary morphology it would be useful to disentangle some of the abiotic and evolutionary confounds that are present in any across species comparison.  To better understand how temperature variation drives differences in coronary morphology, a population-level study across a temperature gradient would be particularly useful.  A species that exhibits local population endemism in a range of thermal regimes would be ideal for this type of comparison.  A population comparison could also provide some information about the pace of evolution of the coronary circulation, as biogeographic methods could be used to ascertain the date of divergence between populations, which would be useful to known given projected thermal changes over the next 100 years.    Some additional selective pressures that would likely impact myocardial oxygen delivery are hypoxia and athleticism. Environmental oxygen profiles and habitat use data could provide some information on the level of hypoxia that each species often encounters.  Data on the maximum metabolic rate would be particularly useful to understand the relationship between myocardial oxygen demand and the morphological characters that dictate oxygen delivery, as species with higher maximum metabolic rates would need to meet higher oxygen demands. With additional environmental and physiological data I could develop a more complete understanding of the selective pressures that have shaped the coronary morphology.    135 Cardiac phenotypes are not only shaped by the genes that underlie them, plasticity can also play a large role in the phenotypes that determine myocardial oxygen delivery.  To disentangle the relative contributions of genetics and the environment on coronary morphology, an acclimation experiment would be ideal. Acclimation experiments to different temperatures, oxygen concentrations, and exercise regimes would be particularly informative. With knowledge on the amount of plasticity in coronary morphological phenotypes in response to different acclimation regimes, it would be possible to determine if species level differences in morphology stem primarily from plasticity or genetic differences. Developing an understanding of the role of plasticity and genetics in dictating coronary morphological phenotypes is key to understanding how species deal with environmental and life history differences. On a physiological level, the relationship between CBF and CPO requires further investigation on un-anesthetised sharks. These experiments will help to determine if CPO is linked to increases in CBF outside of the dependence of CPO on heart rate in sharks. These experiments would also indicate if the weak relationship found between CBF and estimated cardiac output is affected by stroke volume. In situ experiments on sharks with intact pericardium may help determine if the lack of relationship between stroke volume and CBF is due to the influence of the luminal supply. Along with concurrent measures of myocardial oxygen consumption in situ protocols could be adapted to investigate if coupling between ventricular contraction and atrial filling reduces the energetic cost of modulating stroke volume compared to heart rate. Furthermore, inclusion of myocardial oxygen consumption  136 would provide some information on the degree of metabolic flow regulation through the coronary artery. This could be compared to the degree of auto-regulation in the coronary circulation if coronary input pressure is independently manipulated.  These are just a few directions for future research that will help to better inform us on the evolution and functional significance of the coronary circulation. As I have discovered that the coronary circulation of sharks plays crucial role in myocardial oxygen delivery, this opens the door for comparisons across species that depend primarily on the coronary circulation for myocardial oxygen. Hopefully this thesis will generate new questions addressing how the coronary circulation adapts in response to various environmental and physiological stressors in order to meet myocardial oxygen demand across vertebrate taxa.     137 Table 6. 1: Metabolic rate, P50 and average inter-vascular distance in four species of shark. Species Metabolic Rate (mg O2 kg-1 h-1) P50 (mmHg) Average IVD (μm) Dogfish 49 17 16.7 Leopard 105 15.2 12.5 Sandbar  110 22 15.6 Mako 344 10.6 11.5 P50 values were obtained from the literature for sandbar sharks (Brill et al., 2008), leopard sharks (Lai et al., 1990a) and dogfish (Lenfant and Johansen, 1966). Mako values were attained through personal communication with Diego Bernal. Metabolic rate date was obtained from the literature for sandbar sharks (Dowd et al., 2006), leopard sharks (Scharold et al., 1988), dogfish (Brett and Blackburn, 1978) and mako sharks (Sepulveda et al., 2007).    138 Bibliography   Agnisola, C., Jensen, F. B., Tota, B. and Mustafa, T. (1998). Performance of the isolated rainbow trout heart perfused under self-controlled coronary pressure conditions: effects of high and low oxygen tension, arachidonic acid and indomethacin. J. Comp. Physiol. B 168, 96–104. Axelsson, M. (1994). The coronary circulation: a fish perspective. Braz. J. Med. Biol. Res. Rev. Bras. Pesqui. Medicas E Biol. Soc. Bras. Biofisica Al 28, 1167–1177. Axelsson, M. and Farrell, A. P. (1993). Coronary blood flow in vivo in the coho salmon (Oncorhynchus kisutch). Am. J. Physiol. - Regul. Integr. Comp. Physiol. 264, R963–R971. Barron, M. G., Tarr, B. D. and Hayton, W. L. (1987). Temperature-dependence of cardiac output and regional blood flow in rainbow trout, Salmo gairdneri Richardson. J. Fish Biol. 31, 735–744. Barton, N. H. (1990). Pleiotropic models of quantitative variation. Genetics 124, 773–782. Bennett, A. F. (1972). The effect of activity on oxygen consumption, oxygen debt, and heart rate in the lizards Varanus gouldii and Sauromalus hispidus. J. Comp. Physiol. 79, 259–280. Bentley, T. B., Meng, H. and Pittman, R. N. (1993). Temperature dependence of oxygen diffusion and consumption in mammalian striated muscle. Am. J. Physiol. - Heart Circ. Physiol. 264, H1825–H1830. Berglund, E., Borst, H. G., Duff, F. and Schreiner, G. L. (1958). Effect of heart rate on cardiac work, myocardial oxygen consumption and coronary blood flow in the dog. Acta Physiol. Scand. 42, 185–198. Bernal, D., Dickson, K. A., Shadwick, R. E. and Graham, J. B. (2001). Review: Analysis of the evolutionary convergence for high performance swimming in lamnid sharks and tunas. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 129, 695–726. Bernal, D., Smith, D., Lopez, G., Weitz, D., Grimminger, T., Dickson, K. and Graham, J. B. (2003a). Comparative studies of high performance swimming in sharks II. Metabolic biochemistry of locomotor and myocardial muscle in endothermic and ectothermic sharks. J. Exp. Biol. 206, 2845–2857. Bernal, D., Sepulveda, C., Mathieu-Costello, O. and Graham, J. B. (2003b). Comparative studies of high performance swimming in sharks I. Red muscle morphometrics, vascularization and ultrastructure. J. Exp. Biol. 206, 2831–2843.  139 Berne, R. M. (1964). Regulation of coronary blood flow. Physiol. Rev. 44, 1–29. Biörck, G. and Johansson, B. (1955). Comparative studies on temperature effects upon the electrocardiogram in some vertebrates. Acta Physiol. Scand. 34, 257–272. Blake, D. W. and Korner, P. I. (1981). Role of baroreceptor reflexes in the hemodynamic and heart rate responses to althesin, ketamine and thiopentone anesthesia. J. Auton. Nerv. Syst. 3, 55–70. Block, B. A. and Finnerty, J. R. (1994). Endothermy in fishes: a phylogenetic analysis of constraints, predispositions, and selection pressures. Environ. Biol. Fishes 40, 283–302. Boag, P. T. and Grant, P. R. (1981). Intense natural selection in a population of darwin’s finches (geospizinae) in the galápagos. Science 214, 82–85. Bourdeau-Martini, J., Odoroff, C. L. and Honig, C. R. (1974). Dual effect of oxygen on magnitude and uniformity of coronary intercapillary distance. Am. J. Physiol. -- Leg. Content 226, 800–810. Brady, A. J. and Dubkin, C. (1964). Coronary circulation in the turtle ventricle. Comp. Biochem. Physiol. 13, 119–128. Brett, J. R. and Blackburn, J. M. (1978). Metabolic rate and energy expenditure of the spiny dogfish, Squalus acanthias. J. Fish. Res. Board Can. 35, 816–821. Brill, R., Bushnell, P., Schroff, S., Seifert, R. and Galvin, M. (2008). Effects of anaerobic exercise accompanying catch-and-release fishing on blood-oxygen affinity of the sandbar shark (Carcharhinus plumbeus, Nardo). J. Exp. Mar. Biol. Ecol. 354, 132–143. Bushnell, P. G., Lutz, P. L., Steffensen, J. F., Oikari, A. and Gruber, S. H. (1982). Increases in arterial blood oxygen during exercise in the lemon shark (Negaprion brevirostris). J. Comp. Physiol. 147, 41–47. Bushnell, P. G., Jones, D. R. and Farrell, A. P. (1992). The Arterial System. In Fish Physiology (ed. W.S. Hoar, D. J. R. and A. P. F.), pp. 89–139. Academic Press. Butler, P. J. and Metcalfe, J. D. (1988). Cardiovascular and Respiratory Systems. In Physiology of Elasmobranch Fishes (ed. Shuttleworth, D. T. J.), pp. 1–47. Springer Berlin Heidelberg. Butler, P. J. and Taylor, E. W. (1975). The effect of progressive hypoxia on respiration in the dogfish (Scyliorhinus canicula) at different seasonal temperatures. J. Exp. Biol. 63, 117–130. Cameron, J. N. (1975). Morphometric and flow indicator studies of the teleost heart. Can. J. Zool. 53, 691–698. Carazzi, D. (1905). Sul systema arterioso di Selache maxima e di altri squalidae. Anat. Anz. 26, 63–124.  140 Carrier, J. C., Musick, J. A. and Heithaus, M. R. (2012). Biology of Sharks and Their Relatives, Second Edition. CRC Press. Casselman, M. T., Anttila, K. and Farrell, A. P. (2012). Using maximum heart rate as a rapid screening tool to determine optimum temperature for aerobic scope in Pacific salmon Oncorhynchus spp. J. Fish Biol. 80, 358–377. Chatterjee, S., Fujiwara, K., Perez, N. G., Ushio-Fukai, M. and Fisher, A. B. (2015). Mechanosignaling in the vasculature: emerging concepts in sensing, transduction and physiological responses. Am. J. Physiol. - Heart Circ. Physiol. ajpheart.00105.2015. Chopin, L. K. and Bennett, M. B. (1995). The regulation of branchial blood flow in the blacktip reef shark, Carcharhinus melanopterus (Carcharhinidae: Elasmobranchii). Comp. Biochem. Physiol. A Physiol. 112, 35–41. Clark, T. D., Sandblom, E., Cox, G. K., Hinch, S. G. and Farrell, A. P. (2008). Circulatory limits to oxygen supply during an acute temperature increase in the Chinook salmon (Oncorhynchus tshawytscha). Am. J. Physiol. - Regul. Integr. Comp. Physiol. 295, R1631–R1639. Costa, I. A. S. F., Hein, T. W. and Gamperl, A. K. (2015a). Cold-acclimation leads to differential regulation of the steelhead trout (Oncorhynchus mykiss) coronary microcirculation. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 308, R743–R754. Costa, I. A. S. F., Hein, T. W., Secombes, C. J. and Gamperl, A. K. (2015b). Recombinant interleukin-1ß dilates steelhead trout coronary microvessels: effect of temperature and role of the endothelium, nitric oxide and prostaglandins. J. Exp. Biol. jeb.119255. Davie, P. S. and Farrell, A. P. (1991a). The coronary and luminal circulations of the myocardium of fishes. Can. J. Zool. 69, 1993–2001. Davie, P. S. and Farrell, A. P. (1991b). Cardiac performance of an isolated heart preparation from the dogfish (Squalus acanthias): the effects of hypoxia and coronary artery perfusion. Can. J. Zool. 69, 1822–1828. Davie, P. S. and Franklin, C. E. (1992). Myocardial oxygen consumption and mechanical efficiency of a perfused dogfish heart preparation. J. Comp. Physiol. B 162, 256–262. Davie, P. S. and Franklin, C. E. (1993). Preliminary observations on blood flow in the coronary arteries of two school sharks (Galeorhinus australis). Can. J. Zool. 71, 1238–1241. Davie, P. S., Franklin, C. E. and Grigg, G. C. (1993). Blood pressure and heart rate during tonic immobility in the black tipped reef shark, Carcharhinus melanoptera. Fish Physiol. Biochem. 12, 95–100. De Andrés, A. V., Muñoz-chápuli, R., Sans-Coma, V. and García-Garrido, L. (1990). Anatomical studies of the coronary system in elasmobranchs: I. Coronary arteries in lamnoid sharks. Am. J. Anat. 187, 303–310.  141 De Andrés, A. V., Muñoz-Chápuli, R., Sans-Coma, V. and García-Garrido, L. (1992). Anatomical studies of the coronary system in elasmobranchs: II. Coronary arteries in hexanchoid, squaloid, and carcharhinoid sharks. Anat. Rec. 233, 429–439. Dombkowski, R. A., Russell, M. J., Schulman, A. A., Doellman, M. M. and Olson, K. R. (2005). Vertebrate phylogeny of hydrogen sulfide vasoactivity. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 288, R243–R252. Dowd, W. W., Brill, R. W., Bushnell, P. G. and Musick, J. A. (2006). Standard and routine metabolic rates of juvenile sandbar sharks (Carcharhinus plumbeus), including the effects of body mass and acute temperature change. Fish. Bull. 104,. Driedzic, W. R. and Gesser, H. (1988). Differences in force-frequency relationships and calcium dependency between elasmobranch and teleost hearts. J. Exp. Biol. 140, 227–241. Driedzic, W. R. and Gesser, H. (1994). Energy metabolism and contractility in ectothermic vertebrate hearts: hypoxia, acidosis, and low temperature. Physiol. Rev. 74, 221–258. Driedzic, W. R., Scott, D. L. and Farrell, A. P. (1983). Aerobic and anaerobic contributions to energy metabolism in perfused isolated sea raven (Hemitripterus americanus) hearts. Can. J. Zool. 61, 1880–1883. Duncker, D. J. and Bache, R. J. (2008). Regulation of coronary blood flow during exercise. Physiol. Rev. 88, 1009–1086. Durán, A. C., López-Unzu, M. A., Rodríguez, C., Fernández, B., Lorenzale, M., Linares, A., Salmerón, F. and Sans-Coma, V. (2015). Structure and vascularization of the ventricular myocardium in Holocephali: their evolutionary significance. J. Anat. 226, 501–510. Egginton, S. (1990). Morphometric analysis of tissue capillary supply. In Vertebrate Gas Exchange (ed. Boutilier, D. R. G.), pp. 73–141. Springer Berlin Heidelberg. Egginton, S. and Cordiner, S. (1997). Cold-induced angiogenesis in seasonally acclimatized rainbow trout (Oncorhynchus mykiss). J. Exp. Biol. 200, 2263–2268. Egginton, S., Turek, Z. and Hoofd, L. J. C. (1988). Differing patterns of capillary distribution in fish and mammalian skeletal muscle. Respir. Physiol. 74, 383–396. Eliason, E. J., Clark, T. D., Hague, M. J., Hanson, L. M., Gallagher, Z. S., Jeffries, K. M., Gale, M. K., Patterson, D. A., Hinch, S. G. and Farrell, A. P. (2011). Differences in thermal tolerance among sockeye salmon populations. Science 332, 109–112. Eliason, E. J., Clark, T. D., Hinch, S. G. and Farrell, A. P. (2013). Cardiorespiratory collapse at high temperature in swimming adult sockeye salmon. Conserv. Physiol. 1, cot008. Faraci, F. M. (1986). Circulation during hypoxia in birds. Comp. Biochem. Physiol. A Physiol. 85, 613–620.  142 Faraci, F. M. and Fedde, M. R. (1986). Regional circulatory responses to hypocapnia and hypercapnia in bar-headed geese. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 250, R499–R504. Faraci, F. M., Kilgore, D. L. and Fedde, M. R. (1984). Oxygen delivery to the heart and brain during hypoxia: Pekin duck vs. bar-headed goose. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 247, R69–R75. Faraci, F. M., Kilgore Jr., D. L. and Fedde, M. R. (1985). Blood flow distribution during hypocapnic hypoxia in pekin ducks and bar-headed geese. Respir. Physiol. 61, 21–30. Farrell, A. P. (1986). Cardiovascular responses in the sea raven, Hemitripterus americanus, elicited by vascular compression. J. Exp. Biol. 122, 65–80. Farrell, A. P. (1987). Coronary flow in a perfused rainbow trout heart. J. Exp. Biol. 129, 107–123. Farrell, A. P. (1991). From hagfish to tuna: a perspective on cardiac function in fish. Physiol. Zool. 64, 1137–1164. Farrell, A. P. and Davie, P. S. (1991a). Coronary vascular reactivity in the skate, Raja Nasuta. Comp. Biochem. Physiol. Part C Comp. Pharmacol. 99, 555–560. Farrell, A. P. and Davie, P. S. (1991b). Coronary artery reactivity in the mako shark, Isurus oxyrinchus. Can. J. Zool. 69, 375–379. Farrell, A. P. and Graham, M. S. (1986). Effects of adrenergic drugs on the coronary circulation of Atlantic salmon (Salmo salar). Can. J. Zool. 64, 481–484. Farrell, A. P. and Steffensen, J. F. (1987). Coronary ligation reduces maximum sustained swimming speed in chinook salmon, Oncorhynchus tshawytscha. Comp. Biochem. Physiol. A Physiol. 87, 35–37. Farrell, A. P., Wood, S., Hart, T. and Driedzic, W. R. (1985). Myocardial oxygen consumption in the sea raven, Hemitripterus americanus: the effects of volume loading, pressure loading and progressive hypoxia. J. Exp. Biol. 117, 237–250. Farrell, A. P., Johansen, J. A. and Graham, M. S. (1988a). The role of the pericardium in cardiac performance of the trout (Salmo gairdneri). Physiol. Zool. 61, 213–221. Farrell, A. P., Hammons, A. M., Graham, M. S. and Tibbits, G. F. (1988b). Cardiac growth in rainbow trout, Salmo gairdneri. Can. J. Zool. 66, 2368–2373. Farrell, A. P., Davie, P. S., Franklin, C. E., Johansen, J. A. and Brill, R. W. (1992). Cardiac physiology in tunas. I. In vitro perfused heart preparations from yellowfin and skipjack tunas. Can. J. Zool. 70, 1200–1210. Farrell, A. P., Farrell, N. D., Jourdan, H. and Cox, G. K. (2012). A perspective on the evolution of the coronary circulation in fishes and the transition to terrestrial life. In Ontogeny and Phylogeny of the Vertebrate Heart (ed. Sedmera, D.) and Wang, T.), pp. 75–102. Springer New York.  143 Feigl, E. O. (1983). Coronary physiology. Physiol. Rev. 63, 1–205. Foxon, G. E. H. (1950). A description of the coronary arteries in Dipnoan fishes and some remarks on their importance from the evolutionary standpoint. J. Anat. 84, 121–131. Franklin, C. and Axelsson, M. (1994). Coronary hemodynamics in elasmobranchs and teleosts. Cardioscience 5, 155–161. Gamperl, A. K. (2011). Integrated responses of the circulatory system to temperature. In Encyclopedia of fish physiology: from genome to environment, pp. 1197–1205. San Diego, CA: Academic Press. Gamperl, A. K. and Farrell, A. P. (2004). Cardiac plasticity in fishes: environmental influences and intraspecific differences. J. Exp. Biol. 207, 2539–2550. Gamperl, A., Pinder, A. and Boutilier, R. (1994a). Effect of coronary ablation and adrenergic stimulation on in vivo cardiac performance in trout (Oncorhynchus Mykiss). J. Exp. Biol. 186, 127–143. Gamperl, A., Pinder, A., Grant, R. and Boutilier, R. (1994b). Influence of hypoxia and adrenaline administration on coronary blood flow and cardiac performance in seawater rainbow trout (Oncorhynchus mykiss). J. Exp. Biol. 193, 209–232. Gamperl, A. K., Axelsson, M. and Farrell, A. P. (1995). Effects of swimming and environmental hypoxia on coronary blood flow in rainbow trout. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 269, R1258–R1266. Gerdes, A. M. (2015). How to evaluate and improve the overall quality of cardiac morphometric data. Am. J. Physiol. - Heart Circ. Physiol. ajpheart.00232.2015. Gillooly, J. F., Brown, J. H., West, G. B., Savage, V. M. and Charnov, E. L. (2001). Effects of size and temperature on metabolic rate. Science 293, 2248–2251. Goo, S., Joshi, P., Sands, G., Gerneke, D., Taberner, A., Dollie, Q., LeGrice, I. and Loiselle, D. (2009). Trabeculae carneae as models of the ventricular walls: implications for the delivery of oxygen. J. Gen. Physiol. 134, 339–350. Graham, M. S. and Farrell, A. P. (1990). Myocardial oxygen consumption in trout acclimated to 5°c and 15°c. Physiol. Zool. 63, 536–554. Grant, R. T. and Regnier, M. (1926). The comparative anatomy of the cardiac coronary vessels. Heart 13, 375–388. Grant, I. S., Nimmo, W. S., Mcnicol, L. R. and Clements, J. A. (1983). Ketamine disposition in children and adults. Br. J. Anaesth. 55, 1107–1111. Grimes, A. C., Durán, A. C., Sans-Coma, V., Hami, D., Santoro, M. M. and Torres, M. (2010). Phylogeny informs ontogeny: a proposed common theme in the arterial pole of the vertebrate heart. Evol. Dev. 12, 552–567.  144 Hagensen, M. K., Abe, A. S., Falk, E. and Wang, T. (2008). Physiological importance of the coronary arterial blood supply to the rattlesnake heart. J. Exp. Biol. 211, 3588–3593. Hamlett, W. C. (1999). Sharks, Skates, and Rays: The Biology of Elasmobranch Fishes. JHU Press. Helle, K. B., Miralto, A., Pihl, K. E. and Tota, B. (1983). Structural organization of the normal and anoxic heart of Scyllium stellare. Cell Tissue Res. 231, 399–414. Henquell, L., Odoroff, C. L. and Honig, C. R. (1977). Intercapillary distance and capillary reserve in hypertrophied rat hearts beating in situ. Circ. Res. 41, 400–408. Hight, B. V. and Lowe, C. G. (2007). Elevated body temperatures of adult female leopard sharks, Triakis semifasciata, while aggregating in shallow nearshore embayments: Evidence for behavioral thermoregulation? J. Exp. Mar. Biol. Ecol. 352, 114–128. Hudlicka, O., Egginton, S. and Brown, M. D. (1988). Capillary diffusion distances - their importance for cardiac and skeletal muscle performance. Physiology 3, 134–138. Hudlicka, O., Brown, M. and Egginton, S. (1992). Angiogenesis in skeletal and cardiac muscle. Physiol. Rev. 72, 369–417. Ishibashi, Y., Duncker, D. J., Zhang, J. and Bache, R. J. (1998). ATP-sensitive K+ channels, adenosine, and nitric oxide–mediated mechanisms account for coronary vasodilation during exercise. Circ. Res. 82, 346–359. Johansen, K., Franklin, D. L. and Van Citters, R. L. (1966). Aortic blood flow in free-swimming elasmobranchs. Comp. Biochem. Physiol. 19, 151–160. Jones, D. P. (1986). Intracellular diffusion gradients of O2 and ATP. Am. J. Physiol. - Cell Physiol. 250, C663–C675. Juhasz-Nagy, A., M, S., M, S. and B, V. (1962). Coronary circulation of the tortoise heart. Acta Physiol. Acad. Sci. Hung. 23, 33–48. Khouri, E. M., Gregg, D. E. and Rayford, C. R. (1965). Effect of exercise on cardiac output, left coronary flow and myocardial metabolism in the unanesthetized dog. Circ. Res. 17, 427–437. Kohmoto, T., Argenziano, M., Yamamoto, N., Vliet, K. A., Gu, A., DeRosa, C. M., Fisher, P. E., Spotnitz, H. M., Burkhoff, D. and Smith, C. R. (1997). Assessment of transmyocardial perfusion in alligator hearts. Circulation 95, 1585–1591. Komatsu, T., Singh, P. K., Kimura, T., Nishiwaki, K., Bando, K. and Shimada, Y. (1995). Differential effects of ketamine and midazolam on heart rate variability. Can. J. Anaesth. 42, 1003–1009. Krogh, A. (1919a). The rate of diffusion of gases through animal tissues, with some remarks on the coefficient of invasion. J. Physiol. 52, 391–408.  145 Krogh, A. (1919b). The number and distribution of capillaries in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J. Physiol. 52, 409–415. Lai, N. C., Graham, J. B., Lowell, W. R. and Shabetai, R. (1989). Elevated pericardial pressure and cardiac output in the leopard shark Triakis semifasciata during exercise: the role of the pericardioperitoneal canal. J. Exp. Biol. 147, 263–277. Lai, N. C., Graham, I. B. and Burnett, L. (1990a). Blood respiratory properties and the effect of swimming on blood gas transport in the leopard shark Triakis semifasciata. J. Exp. Biol. 151, 161–173. Lai, N. C., Shabetai, R., Graham, J. B., Hoit, B. D., Sunnerhagen, K. S. and Bhargava, V. (1990b). Cardiac function of the leopard shark, Triakis semifasciata. J. Comp. Physiol. B 160, 259–268. Lai, N. C., Graham, J. B., Bhargava, V. and Shabetai, R. (1996). Mechanisms of venous return and ventricular filling in elasmobranch fish. Am. J. Physiol. - Heart Circ. Physiol. 270, H1766–H1771. Lai, N. C., Korsmeyer, K. E., Katz, S., Holts, D. B., Laughlin, L. M. and Graham, J. B. (1997). Hemodynamics and blood properties of the shortfin mako shark (Isurus oxyrinchus). Copeia 1997, 424–428. Lenfant, C. and Johansen, K. (1966). Respiratory function in the elasmobranch Squalus suckleyi G. Respir. Physiol. 1, 13–29. León-Velarde, F., Sanchez, J., Bigard, A. X., Brunet, A., Lesty, C. and Monge-C, C. (1993). High altitude tissue adaptation in Andean coots: capillarity, fibre area, fibre type and enzymatic activities of skeletal muscle. J. Comp. Physiol. B 163, 52–58. Lillywhite, H. B. (1987). Temperature, energetics, and physiological ecology. In Snakes: Ecology and Evolutionary Biology, pp. 422 – 77. New York, NY: MacMillan Pub. Co. Lillywhite, H. B., Zippel, K. C. and Farrell, A. P. (1999). Resting and maximal heart rates in ectothermic vertebrates. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 124, 369–382. Manohar, M. (1987). Transmural coronary vasodilator reserve and flow distribution during maximal exercise in normal and splenectomized ponies. J. Physiol. 387, 425–440. Marples, B. J. (1936). The blood vascular system of the Elasmobranch fish Squatina swuatina (Linne). Trans. R. Soc. Edinb. 58, 1–3. Mathieu-Costello, O., Agey, P. J., Wu, L., Szewczak, J. M. and MacMillen, R. E. (1998). Increased fiber capillarization in flight muscle of finch at altitude. Respir. Physiol. 111, 189–199. Merson, R. R. and Jr, H. L. P. (2001). Distribution, movements and growth of young sandbar sharks, Carcharhinus plumbeus, in the nursery grounds of delaware bay. Environ. Biol. Fishes 61, 13–24.  146 Miyazaki, S., Guth, B. D., Miura, T., Indolfi, C., Schulz, R. and Ross, J. (1990). Changes of left ventricular diastolic function in exercising dogs without and with ischemia. Circulation 81, 1058–1070. Muñoz-Chápuli, R., De Andrés, A. V. and Dingerkus, G. (1994). Coronary artery anatomy and elasmobranch phylogeny. Acta Zool. 75, 249–254. Naylor, G. J. P., Caira, J. N., Jensen, K., Rosana, K. A. M., Straube, N. and Lakner, C. (2012). Elasmobranch phylogeny: A mitochondrial estimate based on 595 species. In Biology of Sharks and Their Relatives, pp. 31–56. Nguyen, T. T., Biadillah, Y., Mongrain, R., Brunette, J., Tardif, J.-C. and Bertrand, O. F. (2004). A method for matching the refractive index and kinematic viscosity of a blood analog for flow visualization in hydraulic cardiovascular models. J. Biomech. Eng. 126, 529–535. Nikinmaa, M. (2012). Vertebrate Red Blood Cells: Adaptations of Function to Respiratory Requirements. Springer Science & Business Media. Nilsson, C. B. (1994). Comparative Physiology and Evolution of the Autonomic Nervous System. CRC Press. Oelz, O., Howald, H., Prampero, P. E. D., Hoppeler, H., Claassen, H., Jenni, R., Buhlmann, A., Ferretti, G., Bruckner, J. C., Veicsteinas, A., et al. (1986). Physiological profile of world-class high-altitude climbers. J. Appl. Physiol. 60, 1734–1742. Parker, T. J. (1887). On the blood vessels of Mustelus antarcticus; a contribution to the morphology of the vascular systemin the vertebrata. Philos. Trans. R. Soc. B Biol. Sci. 177, 685–732. Parker, G. H. and Davis, F. K. (1899). The blood vessels of the heart in Carcharias, Raja and Amia. Proc. Boston Soc. Nat. Hist. 29,. Pavesi, P. (1874). Contribuzione alla Storia Naturale del genere Selachi. Ann. Mus. Civ. Storia Nat. Genova 6, 5–72. Perry, S. and Gilmour, K. (1996). Consequences of catecholamine release on ventilation and blood oxygen transport during hypoxia and hypercapnia in an elasmobranch Squalus acanthias and a teleost Oncorhynchus mykiss. J. Exp. Biol. 199, 2105–2118. Pieperhoff, S., Bennett, W. and Farrell, A. P. (2009). The intercellular organization of the two muscular systems in the adult salmonid heart, the compact and the spongy myocardium. J. Anat. 215, 536–547. Piiper, J., Meyer, M., Worth, H. and Willmer, H. (1977). Respiration and circulation during swimming activity in the dogfish Scyliorhinus stellaris. Respir. Physiol. 30, 221–239. Pittman, R. N. (2011). Oxygen gradients in the microcirculation. Acta Physiol. 202, 311–322.  147 Poupa, O. and Lindström, L. (1983). Comparative and scaling aspects of heart and body weights with reference to blood supply of cardiac fibers. Comp. Biochem. Physiol. A Physiol. 76, 413–421. Preece, A. (1972). A manual for histologic technicians. Third. Boston, USA: Little, Brown and Company. Pries, A. R., Neuhaus, D. and Gaehtgens, P. (1992). Blood viscosity in tube flow: dependence on diameter and hematocrit. Am. J. Physiol. - Heart Circ. Physiol. 263, H1770–H1778. Randall, D. J. (1968). Functional Morphology of the Heart in Fishes. Am. Zool. 8, 179–189. Restorff, W. von, Holtz, J. and Bassenge, E. (1977). Exercise induced augmentation of myocardial oxygen extraction in spite of normal coronary dilatory capacity in dogs. Pflüg. Arch. 372, 181–185. Rubio, R. and Berne, R. M. (1975). Regulation of coronary blood flow. Prog. Cardiovasc. Dis. 18, 105–122. Rutskina, I. M., Litvinov, N. A., Roshchevskaya, I. M. and Roshchevskii, M. P. (2009). Temperature adaptation of the heart in the grass snake (Natrix natrix L.), common European viper (Vipera berus L.), and steppe viper (Vipera renardi Christoph) (Reptilia: Squamata: Serpentes). Russ. J. Ecol. 40, 314–319. Sandblom, E., Cox, G. K., Perry, S. F. and Farrell, A. P. (2009). The role of venous capacitance, circulating catecholamines, and heart rate in the hemodynamic response to increased temperature and hypoxia in the dogfish. Am. J. Physiol. - Regul. Integr. Comp. Physiol. 296, R1547–R1556. Santer, R. M. (1985). Morphology and innervation of the fish heart. Adv. Anat. Embryol. Cell Biol. 89, 1–102. Santer, R. M. and Greer Walker, M. (1980). Morphological studies on the ventricle of teleost and elasmobranch hearts. J. Zool. 190, 259–272. Satchell, G. H. (1971). Circulation in Fishes. CUP Archive. Satchell, G. H. and Johnes, M. P. (1967). The function of the conus arteriosus in the port jackson shark, Heterodontus portusjacksoni. J. Exp. Biol. 46, 373–382. Scaramucci, G.B. (1695). Theroremata familiaria viros eruditos consulentia, de variis physico-medicis lucubrationibus juxta leges mecanicas. In De motu cordis, pp. 70–81. Urbino, Italy: Apud Joannem Baptistam Bustum. Scharold, J., Lai, N. C., Lowell, W. R. and Graham, J. B. (1988). Metabolic rate, heart rate, and tailbeat frequency during sustained swimming in the leopard shark Triakis semifasciata. Exp. Biol. 48, 223–230. Schmidt-Nielsen, K. (1984). Scaling: Why is Animal Size So Important? Cambridge University Press.  148 Scott, G. R., Schulte, P. M., Egginton, S., Scott, A. L. M., Richards, J. G. and Milsom, W. K. (2011). Molecular evolution of cytochrome c oxidase underlies high-altitude adaptation in the bar-headed goose. Mol. Biol. Evol. 28, 351–363. Sedmera, D. and Wang, T. eds. (2012). Ontogeny and Phylogeny of the Vertebrate Heart. New York, NY: Springer New York. Seebacher, F. (2000). Heat transfer in a microvascular network: the effect of heart rate on heating and cooling in reptiles (Pogona barbata and Varanus varius). J. Theor. Biol. 203, 97–109. Sepulveda, C. A., Graham, J. B. and Bernal, D. (2007). Aerobic metabolic rates of swimming juvenile mako sharks, Isurus oxyrinchus. Mar. Biol. 152, 1087–1094. Short, S., Taylor, E. W. and Butler, P. J. (1979). The effectiveness of oxygen transfer during normoxia and hypoxia in the dogfish (Scyliorhinus canicula L.) before and after cardiac vagotomy. J. Comp. Physiol. 132, 289–295. Sidell, B. D. and O’Brien, K. M. (2006). When bad things happen to good fish: the loss of hemoglobin and myoglobin expression in Antarctic icefishes. J. Exp. Biol. 209, 1791–1802. Small, S. A. and Farrell, A. P. (1990). Vascular reactivity of the coronary artery in steelhead trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. Part C Comp. Pharmacol. 97, 59–63. Small, S. A., MacDonald, C. and Farrell, A. P. (1990). Vascular reactivity of the coronary artery in rainbow trout (Oncorhynchus mykiss). Am. J. Physiol. - Regul. Integr. Comp. Physiol. 258, R1402–R1410. Smith, M. (1992). Capture and transportation of elasmobranchs, with emphasis on the grey nurse shark (Carcharias taurus). Mar. Freshw. Res. 43, 325–343. Snyder, G. K. (1971). Influence of temperature and hematocrit on blood viscosity. Am. J. Physiol. 220, 1667–1672. Speers-Roesch, B., Brauner, C. J., Farrell, A. P., Hickey, A. J. R., Renshaw, G. M. C., Wang, Y. S. and Richards, J. G. (2012). Hypoxia tolerance in elasmobranchs. II. Cardiovascular function and tissue metabolic responses during progressive and relative hypoxia exposures. J. Exp. Biol. 215, 103–114. Steffensen, J. F. and Farrell, A. P. (1998). swimming performance, venous oxygen tension and cardiac performance of coronary-ligated rainbow trout, Oncorhynchus mykiss, exposed to progressive hypoxia. Comp. Biochem. Physiol. A. Mol. Integr. Physiol. 119, 585–592. Steinhausen, M. F., Sandblom, E., Eliason, E. J., Verhille, C. and Farrell, A. P. (2008). The effect of acute temperature increases on the cardiorespiratory performance of resting and swimming sockeye salmon (Oncorhynchus nerka). J. Exp. Biol. 211, 3915–3926.  149 Stensløkken, K.-O., Sundin, L., Renshaw, G. M. C. and Nilsson, G. E. (2004). Adenosinergic and cholinergic control mechanisms during hypoxia in the epaulette shark (Hemiscyllium ocellatum), with emphasis on branchial circulation. J. Exp. Biol. 207, 4451–4461. Stevens, E. D. and Randall, D. J. (1967). Changes in blood pressure, heart rate and breathing rate during moderate swimming activity in rainbow trout. J. Exp. Biol. 46, 307–315. Taberner, A. J., Han, J.-C., Loiselle, D. S. and Nielsen, P. M. F. (2011). An innovative work-loop calorimeter for in vitro measurement of the mechanics and energetics of working cardiac trabeculae. J. Appl. Physiol. 111, 1798–1803. Taylor, N. B. (1931). The relation of temperature to the heart rate of the south african frog (Xenopus dactylethra). J. Physiol. 71, 156–168. Taylor, E. W. and Barrett, D. J. (1985). Evidence of a respiratory role for the hypoxic bradycardia in the dogfish Scyliorhinus canicula L. Comp. Biochem. Physiol. A Physiol. 80, 99–102. Taylor, E. W., Short, S. and Butler, P. J. (1977). The role of the cardiac vagus in the response of the dogfish Scyliorhinus canicula to hypoxia. J. Exp. Biol. 70, 57–75. Thorarensen, H., Gallaugher, P. and Farrell, A. P. (1996). Cardiac output in swimming rainbow trout, Oncorhynchus mykiss, acclimated to seawater. Physiol. Zool. 69, 139–153. Tomanek, R. J. (2012). Coronary Vasculature: Development, Structure-Function, and Adaptations. Springer Science & Business Media. Tota, B. (1983). Vascular and metabolic zonation in the ventricular myocardium of mammals and fishes. Comp. Biochem. Physiol. A Physiol. 76, 423–437. Tota, B. (1989). Myoarchitecture and vascularization of the elasmobranch heart ventricle. J. Exp. Zool. 252, 122–135. Tota, B., Cimini, V., Salvatore, G. and Zummo, G. (1983). Comparative study of the arterial and lacunary systems of the ventricular myocardium of elasmobranch and teleost fishes. Am. J. Anat. 167, 15–32. Tsai, A. G., Johnson, P. C. and Intaglietta, M. (2003). Oxygen gradients in the microcirculation. Physiol. Rev. 83, 933–963. van de Hoef, T. P., Nolte, F., Rolandi, M. C., Piek, J. J., van den Wijngaard, J. P. H. M., Spaan, J. A. E. and Siebes, M. (2012). Coronary pressure-flow relations as basis for the understanding of coronary physiology. J. Mol. Cell. Cardiol. 52, 786–793. Wells, R. M. G. and Weber, R. E. (1991). Is there an optimal haematocrit for rainbow trout, Oncorhynchm mykiss (Walbaum)? An interpretation of recent data based on blood viscosity measurements. J. Fish Biol. 38, 53–65.  150 West, G., Heard, D. and Caulkett, N. (2008). Zoo Animal and Wildlife Immobilization and Anesthesia. John Wiley & Sons. White, F. C., Bloor, C. M., McKirnan, M. D. and Carroll, S. M. (1998). Exercise training in swine promotes growth of arteriolar bed and capillary angiogenesis in heart. J. Appl. Physiol. 85, 1160–1168. Windberger, U., Bartholovitsch, A., Plasenzotti, R., Korak, K. J. and Heinze, G. (2003). Whole blood viscosity, plasma viscosity and erythrocyte aggregation in nine mammalian species: reference values and comparison of data. Exp. Physiol. 88, 431–440. Wood, C. M. and Shelton, G. (1980). Cardiovascular dynamics and adrenergic responses of the rainbow trout in vivo. J. Exp. Biol. 87, 247–270. Yaku, H., Slinker, B. K., Mochizuki, T., Lorell, B. H. and LeWinter, M. M. (1993). Use of 2,3-butanedione monoxime to estimate nonmechanical VO2 in rabbit hearts. Am. J. Physiol. - Heart Circ. Physiol. 265, H834–H842.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0223106/manifest

Comment

Related Items