UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The Bulbus arteriosus of tuna : Form and function Braun, Marvin Herbert 2001

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

Item Metadata

Download

Media
831-ubc_2001-0331.pdf [ 7.92MB ]
Metadata
JSON: 831-1.0089971.json
JSON-LD: 831-1.0089971-ld.json
RDF/XML (Pretty): 831-1.0089971-rdf.xml
RDF/JSON: 831-1.0089971-rdf.json
Turtle: 831-1.0089971-turtle.txt
N-Triples: 831-1.0089971-rdf-ntriples.txt
Original Record: 831-1.0089971-source.json
Full Text
831-1.0089971-fulltext.txt
Citation
831-1.0089971.ris

Full Text

T H E BULBUS ARTERIOSUS O F TUNA; F O R M AND FUNCTION by M A R V I N H E R B E R T B R A U N B.Sc. (Hon.), Un ivers i ty o f B r i t i s h Co lumb ia , 1996 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department o f Zoo l o g y W e accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A July 2001 © M a r v i n Herbert Braun, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT The bulbus arteriosus, the fourth chamber of the teleost heart, is very similar in structure and function to blood vessels. However, when the inflation properties of the bulbus are compared with those of an artery, a striking difference appears. The pressure-volume loops of arteries are J-shaped, indicating an increasing stiffness with increasing distension. The bulbus pressure-volume curve can best be described as an r-J curve, with an initial sharp rise followed by long plateau phase, and a final upturn following the plateau. This suggests that the bulbus has a high initial stiffness that decreases as volume increases before rising again at high distensions. To analyse the cause of this inflation behaviour, 1 utilised light microscopy, electron microscopy, mechanical testing of bulbar and arterial walls, and in vivo recordings of diameter and pressure changes of bulbi. The results show that the answer to the bulbus question is multifaceted. The steep initial rise in pressure is due to the geometry of the bulbus. When empty, the bulbus lumen is very small. According to the Law of Laplace, a small radius requires a large pressure increment in order for expansion to occur. As the lumen radius increases, the pressure increments decrease in size. The extremely compliant plateau phase of the bulbar inflation is due to the specialised material properties of the bulbar wall. The combination of a high elastin/collagen ratio, lack of elastic lamellae and specialised elastin result in a decreased elastic modulus (stiffness) within the bulbar wall. Once on the plateau phase of the bulbar inflation, this decreased stiffness allows the bulbus to expand a tremendous amount for small changes in pressure. The tensile tests show that the bulbar material exhibits J-shaped changes in stiffness when stretched and, despite the low modulus, at high extensions, the bulbar material can become quite stiff, resulting in the final rise in stiffness seen in inf lat ion. The inf lat ion behaviour o f the bulbus a l lows to it funct ion as an exceptional pressure and vo lume reservoir, m im i c k i n g the Windkesse l effects o f a much longer arterial tree. i n TABLE OF CONTENTS A B S T R A C T i i T A B L E O F C O N T E N T S iv L I S T O F F I G U R E S v i i L I S T O F E Q U A T I O N S x v i A C K N O W L E D G E M E N T S x v i i I N T R O D U C T I O N 1 C i rcu lat ion 1 Problems o f the Heart 1 Haemodynamic Solutions 2 Haemodynamic Solutions in F i sh 4 The Bu lbus Arteriosus: location and gross morphology 5 Bu lbu s Arteriosus: functions 8 Bu lbu s Arter iosus vs. Artery: wa l l structure 9 Bu lbus Arter iosus vs. Artery: elastin 11 Bu lbus Arter iosus vs. Artery: functional differences? 13 Or ig in o f the Bu lbus Arteriosus: cardiac or arterial? 13 Goal s o f the Study 15 M A T E R I A L S A N D M E T H O D S 17 Pressure V o l u m e Loop s 17 B l o c k i n g o f Smooth Mu sc l e and Denaturation o f Co l l agen 21 Pressure-Vo lume L o o p s o f M o d i f i e d B l u e M a r l i n B u l b i 22 Stress-Strain Exper iments on D i f ferent Parts o f the B u l b i 22 Tensi le Tests on Isolated Pieces o f the B u l b i 27 V i deo D imens iona l Ana lys i s f rom Y e l l o w f i n Tuna Showing 28 D imens iona l Changes o f the B u l b i dur ing in vivo and in situ Pressure Recordings Statistics 31 H i s to logy 33 E lect ron M i c r o s copy 33 R E S U L T S 34 1. Gross Mo rpho logy 34 2. L i gh t M ic roscopy: ye l l owf in tuna 34 W a l l Structure 34 iv B l o o d Supply o f the Bu lbus 51 Bu lbar Lumen at H i g h and L o w Pressure 57 3. E lect ron M i c ro scopy : ye l l owf i n tuna 57 4. Inflations 67 In vitro Bu lbus Arter iosus 67 In situ Bu lbus Arter iosus 70 In vitro Arter ies 73 Smooth Mu sc l e Inactivation and Co l l agen 80 Denaturation Sectioned Bu lbu s Arter iosus 80 Inside-out Bu lbus Arter iosus 83 Dissected Bu lbu s Arter iosus 83 In situ Inf lation o f Bu lbus Arter iosus us ing V D A 85 5. Tensi le Test ing 88 Y e l l o w f i n Tuna Bu lbus and Ventra l Ao r t a R ings 88 B l ue M a r l i n Bu lbu s and Ventra l A o r t a R ings 95 Y e l l o w f i n and B i geye Tuna Bu lba r Segments 100 6. In Situ and In Vivo Recordings: ye l l owf i n tuna 103 Relat ionship between D iameter and V o l u m e 115 D I S C U S S I O N - 140 1. The Prob lem The Bu lbus Inf lat ion Curve 140 Problems with Tradit ional Ana lyses 142 2. Inflation Mechan ics : the sharp rise 143 Effects o f Shape 144 Effect o f W a l 1 Mo rpho l o g y 144 Effects o f Interactions Be tween E last in, Co l l agen 146 and Smooth Mu s c l e The L a w o f Lap lace 147 3. Inflation Mechan ics : material properties 152 Bu lbus vs. Artery 152 Advent i t ia l Co l lagen 153 Elast in 155 Lamel lae in Whales: l ow modulus and high 156 compl iance 4. Inflation Mechanics : the f inal rise in stiffness 158 5. Funct ions o f the Bu lbus Arter iosus 161 V o l u m e Reservoir 162 Consequences o f Comp l i ance 166 Pressure Reservoir 168 Funct ions in Other Species 170 Or ig in o f the Tuning: smooth muscle 171 Tun ing o f the Bulbus: lumen size 173 Benef i t o f Bu lba r Locat ion 173 Cont ro l l i ng the Bu lba r Properties 174 v Nervous Contro l o f the Bu lbu s 174 Ho rmona l Contro l o f the Bu lbu s 175 Bu lbar Contro l o f the Card iovascular System 176 Funct ions o f the Long i tud ina l E lements 177 6. Things to do 180 7. Summary 181 R E F E R E N C E S 185 vi LIST OF FIGURES 1. A. Schematic compar ison between the circulat ions o f mammals and fish. 7 B. Pressure and f l ow traces f rom a ye l l owf i n tuna (Thunnus albacares). (Jones et 7 al., 1993) C. D i ag ram o f the heart o f a ra inbow trout. (Randal l , 1970) 7 D. r-shaped pressure-volume curve vs. J-shaped pressure-volume curve. 7 2. Set-up used for inf lat ing bu lb i dur ing static pressure-volume experiments. 19 3. The protocol ut i l i zed to separate the bulb i o f ye l l owf i n tuna and blue mar l in into 23 posterior, middle and anterior segments. 4. The set-up used for uniaxial extensions o f bulbar tissue loops. 25 5. M e t h o d used for f ind ing the values o f thickness and in it ia l length for bulbar 26 loops. 6. Set-up for the video dimensional analysis ( V D A ) . 29 7. The set-up o f the V D A study in the pericardial cavity o f a ye l l owf i n tuna. 30 8. H o w the V D A w i ndow translates diameter changes into voltage. 32 9. A. The bulbus arteriosus o f a ye l l ow fin tuna 36 B. A ye l l owf i n tuna bulbus opened to reveal the longitudinal elements with in 36 C. The bulbus arteriosus o f an albacore tuna 36 D. The bulbus arteriosus o f a blue marl in 37 E. A blue marl in bulbus opened to reveal the longitudinal elements w i th in 37 F. The bulbus arteriosus o f a pomfret 37 10. L ight microscopy. Cross sections f rom the bulbus arteriosus o f a ye l l ow f i n tuna showing the wa l l structure changing f rom anterior to posterior A. M o s t anterior cross section 40 B. Section posterior to ( A ) 40 vii C. Section posterior to (B) 40 D. M o s t posterior section 40 11. L i gh t microscopy. Compar i son o f cross sections f rom the midd le and posterior o f the ye l l owf i n tuna bulbus arteriosus A. The midd le o f the bulbus 42 B. The posterior o f the bulbus 42 12. L i gh t microscopy. A series taken f rom a single cross section through the wa l l o f a ye l l owf i n tuna bulbus arteriosus A. The outer media 43 B. The transition between the outer and inner media 43 C. The longitudinal elements 43 13. L i gh t microscopy. Compar i son o f the wa l l morphology at the anterior and posterior o f the ye l l owf in tuna bulbus arteriosus A. A cross section through the anterior end o f the bulbus arteriosus 46 B. A cross section through the posterior end o f the bulbus arteriosus 46 14. L i g h t microscopy. Cross sections showing the different orientations o f elastin w i th in the w a l l o f the ye l l owf i n tuna bulbus arteriosus A. The transition between the inner and outer media 48 B. Long i tud ina l and radial elements 48 15. L i g h t microscopy. Compar i son between cross sections f rom the anterior bulbus arteriosus and ventral aorta o f the ye l l owf i n tuna A. The anterior bulbus arteriosus 50 B. The ventral aorta 50 16. L i gh t microscopy. Long i tud ina l sections o f the posterior end o f the ye l l ow f i n tuna bulbus arteriosus A. The outer edge o f the media reveals a layer o f smooth muscle 53 viii B. C lose up o f the smooth muscle f rom (A ) 53 17. L i gh t microscopy A. Long i tud ina l sections f rom the posterior end o f the ye l l ow f i n tuna bulbus 55 arteriosus B. Cross section f rom the posterior end o f the ye l l owf i n tuna bulbus arteriosus 55 C. Cross section f rom the middle o f the ye l l owf in tuna bulbus arteriosus 55 18. L ight microscopy. A. Cross section o f a b lood vessel in the adventit ia o f the ye l l ow f i n tuna bulbus 56 arteriosus B. Long i tud ina l section o f a b lood vessel in the adventit ia o f the ye l l ow f in tuna 56 bulbus arteriosus 19. L i gh t microscopy. A. Long i tud ina l section f rom a ye l l owf in tuna bulbus arteriosus f i xed in an 58 empty state. B. Long i tud ina l section f rom a ye l l owf in tuna bulbus arteriosus f i xed wh i l e 58 inflated. 20. E lect ron micrograph. Cross section f rom a ye l l owf i n tuna bulbus arteriosus 59 showing the transition between the adventit ia and the media. 21. E lect ron micrograph. Cross section f rom a ye l l owf i n tuna bulbus arteriosus 61 showing the col lagenous adventitia. 22. E lect ron micrograph. Cross section f rom the media o f a y e l l ow f i n tuna bulbus 62 arteriosus indicat ing c i rcumferent ia l ly arranged smooth muscle cells and elastin f ibr i ls. 23. E lect ron micrograph. Cross section f rom the media o f a y e l l ow f i n tuna bulbus 64 arteriosus indicat ing longitudinal ly arranged smooth muscle cel ls and elastin f ibr i ls. 24. E lect ron micrograph. Cross section through a layer o f smooth muscle cells in 65 the media o f a ye l l owf i n tuna bulbus arteriosus. 25. E lect ron micrograph. Cross section o f the luminal surface f rom the bulbus 66 ix arteriosus o f a ye l l owf in tuna. 26. E lect ron micrograph. Cross section o f a longitudinal element f rom the bulbus 68 arteriosus o f a ye l l owf in tuna. 27. E lec t ron micrograph. Cross section f rom the ventricular end o f a y e l l o w f i n tuna 69 bulbus arteriosus showing the luminal surface. 28. Pressure-volume loops f rom the bulbi o f y e l l ow f i n tuna, sail f ish, blue marl in, 71 striped marl in, albacore tuna, mahimahi tuna, and pomfret. 29. Compar i son o f the pressure-volume loops f rom the bu lb i o f y e l l ow f i n tuna 72 inflated w i th and without an intact pericardium. 30. A . Pressure-volume loop f rom ye l l owf i n tuna ventral aorta max ima l l y inf lated 75 to 100 u i B. Pressure-volume loop from ye l l owf in tuna ventral aorta max ima l l y inf lated 75 to 25 u.1. C. Compar i son o f the inflations in ( A ) and (B) after normal izat ion. 75 31. A. Pressure-volume loop for a blue mar l in ventral aorta. 77 B. Compar i son o f the ventral aorta pressure-volume loops f rom ye l l ow f i n tuna 77 and blue marl in. 32. A. Pressure-volume loop f rom the dorsal aorta o f a ye l l owf i n tuna. 79 B. Pressure-volume loop f rom the dorsal aorta o f a ye l l owf i n tuna parasit ized by 79 the larval cestode, Dasrynchus talismani. C. Compar i son between the pressure-volume loops f rom a parasit ized dorsal 79 aorta, an unparasitized dorsal aorta and a bulbus arteriosus. 33. Compar i son o f the pressure-volume loops f rom fresh bu lb i before and after undergoing 2 treatments: a tissue bath containing a 10" 5 M solut ion o f nicardipine to inactivate the smooth muscle and a high temperature bath to denature the collagen. A. B u l b i f rom ye l l owf in tuna 81 B. B u l b i f r om ra inbow trout 81 34. Pressure-volume loops for segments cut f rom a blue mar l in bulbus arteriosus. 82 x 35. A. Pressure-volume loop f rom an inside-out blue mar l in bulbus arteriosus. 84 B. Pressure-volume loop f rom a normal blue mar l in bulbus arteriosus. 84 36. Compar i son o f pressure vo lume loops f rom the bulbi o f blue mar l in after layers 86 o f the wa l l were removed. 37. A. Pressure-volume loops created in situ using the ye l l ow f i n tuna prep f rom the 87 V D A set up B. Pressure-diameter loops created in situ using the ye l l ow f i n tuna prep f rom 87 the V D A set up. 38. Schematic drawing showing the different layers o f the bulbar wa l l and their 89 respective thicknesses. 39. A. Stress-strain curves compar ing the adventit ia and outer media o f y e l l ow f i n 90 tuna bulbi. B. Modulus-s t ra in curves compar ing the adventit ia and outer media o f y e l l o w f i n 90 tuna bulbi. 40. A. Stress-strain curves compar ing the anterior, middle and posterior portions o f 92 ye l l ow f i n tuna bulbi. B. Modulus-s t ra in curves compar ing the anterior, middle and posterior portions 92 o f ye l l ow f i n tuna bulbi. C. Stress-strain curves compar ing the ventral aorta, anterior bulbus, midd le 93 bulbus and posterior bulbus o f ye l l owf i n tuna. D. Modulus - s t ra in curves compar ing the ventral aorta, anterior bulbus, midd le 93 bulbus and posterior bulbus o f ye l l owf in tuna. 41. A. Stress-strain curves compar ing the anterior, middle and posterior portions o f 96 blue mar l in bulbi. B. Modulus-s t ra in curves compar ing the anterior, midd le and posterior portions 96 o f blue marl in bulbi. C. Stress-strain curves compar ing the ventral aorta, anterior bulbus, midd le 97 bulbus and posterior bulbus o f blue marl in. D. Modulus - s t ra in curves compar ing the ventral aorta, anterior bulbus, midd le 97 bulbus and posterior bulbus o f blue marl in. xi 42. Modulus - s t ra in curves compar ing the ventral aortas f rom ye l l ow f i n tuna and 98 blue marl in. 43. Modulus - s t ra in curves compar ing the anterior, midd le and posterior segments 99 f r om the bu lb i o f ye l l owf i n tuna and blue marl in. 44. A. Stress-strain curves compar ing different segments f rom the bulb i o f 102 ye l l ow f i n and b ig eye tuna. B. Modulus-stra in curves compar ing different segments f rom the bu lb i o f 102 ye l l ow f i n and b ig eye tuna. 45. A. Record ings o f p rox imal b lood pressure and bulbus diameter f rom a y e l l o w f i n 104 tuna. B. Compar i son o f static and dynamic pressure-diameter loops f rom the bulb i o f 104 ye l l owf i n tuna. 46. A. Record ings o f prox imal b lood pressure and bulbus diameter f rom a ye l l ow f i n 107 tuna. B. D y n a m i c pressure-diameter loops generated by plott ing pressure against 107 diameter for the highl ighted heartbeat in (A). 47. A. Record ings o f p rox imal b lood pressure and bulbus diameter f r om a y e l l o w f i n 108 tuna. B. Dynamic pressure-diameter loops generated by plott ing pressure against 108 diameter for the highl ighted heartbeat in (A). 48. A. Record ings o f p rox imal b lood pressure and bulbus diameter f rom a y e l l o w f i n 109 tuna. B. D ynam i c pressure-diameter loops generated by plott ing pressure against 109 diameter for the highl ighted heartbeat in (A). 49. A. Recordings o f prox imal b lood pressure and bulbus diameter f rom a y e l l o w f i n 110 tuna. B. Dynamic pressure-diameter loops generated by plott ing pressure against 110 diameter for the highl ighted heartbeat in (A). 50. A. Record ings o f prox imal b lood pressure and bulbus diameter f rom a y e l l o w f i n 111 tuna. xii B. D y n a m i c pressure^diameter loops generated by plott ing pressure against 111 diameter for the highl ighted heartbeat in ( A ) 51. D y n a m i c b lood pressure and diameter changes f rom the bulb i o f y e l l ow f i n tuna. A. F o l l o w i n g a bradycardia 113 B. A t mean systol ic pressure 113 C. A t very h igh pressure 114 D. A t very h igh pressure 114 52. A. Compar i son o f pressure-volume loops f rom static, in situ inf lat ions o f 116 ye l l ow f i n tuna bulbi. B. Compar i son o f pressure-diameter loops f rom the same bu lb i as (A). Anter io r 116 and posterior refer to where on the bulbus the V D A w i n d o w was centred. C. D iameter plotted against vo lume for the bulbus measured at the anterior end. 117 D. D iameter plotted against vo lume for the bulbus measured at the posterior 117 end. 53. A . Pressure-volume loop f rom a static, in situ inf lat ion o f a y e l l ow f i n tuna 119 bulbus arteriosus. B. Pressure-diameter loop for the bulbus in (A ) 120 C. D iameter plotted against vo lume for the same bulbus as (A ) 120 54. Sti l ls f r om the V D A analysis showing bulbar d imensional changes between 121 diastole and systole 55. Consequences o f the bulbus length changes on the V D A analysis. 123 56. The method used to model the expansion o f the bulbus. 124 57. A. The relationship between diameter and length in a hypothetical cone where 127 length and diameter increased by an equivalent amount as vo lume increased. B. The ratio o f length strain to diameter strain plotted against vo lume for the 127 hypothetical cone in (A). C. The relationship between diameter and length in a hypothetical cone where 128 xiii the increases in length decreased as vo lume increased. D. The ratio o f length strain to diameter strain plotted against vo lume for the 128 hypothetical cone in (C). 58. The ratio o f length strain to diameter strain plotted against vo lume for actual 129 bulbar inflations. 59. A. Record ings o f p rox imal b lood pressure and bulbus diameter f rom a ye l l ow f i n 131 tuna during normal beating. B. The vo lume changes w i th in the bulbus dur ing the beating in ( A ) 131 60. A. Record ings o f p rox imal b lood pressure and bulbus diameter f rom a ye l l ow f i n 132 tuna during normal beating. B. The vo lume changes with in the bulbus dur ing the beating in ( A ) 132 61. A. Record ings o f p rox imal b lood pressure and bulbus diameter f rom a ye l l ow f i n 133 tuna wh i ch experienced a slight bradycardia. B. The vo lume changes wi th in the bulbus dur ing the beating in ( A ) 133 62. A. Record ings o f prox imal b lood pressure and bulbus diameter f rom a ye l l ow f i n 135 tuna recently injected with saffan. B. The vo lume changes wi th in the bulbus dur ing the beating in ( A ) 135 63. P r ox ima l b lood pressure and bulbus diameter f rom ye l l ow f i n tuna 136 64. A . P r ox ima l b lood pressure and bulbus diameter f r om a y e l l o w f i n tuna. 138 B. A long trace containing the recordings seen in (A ) 138 65. A. Reco rd ing o f prox imal b lood pressure in a ye l l owf i n tuna. • 139 B. Record ing o f bulbus diameter over the same time per iod as (A). 139 66. A. The typica l r-shaped pressure-volume loop o f the bulbus arteriosus. 141 B. The less common r-J shaped pressure-volume loop o f the bulbus arteriosus. 141 67. A. No rma l i z ed pressure-volume loops f rom a ye l l owf i n tuna bulbus arteriosus 150 and ventral aorta. xiv B. The dimensional changes a bulbus and an artery wou ld undergo f rom zero 150 pressure to diastol ic pressure. 68. A . No rma l i z ed pressure vo lume loops f rom a ye l l owf i n tuna bulbus arteriosus 165 and ventral aorta B. Compar i son o f the diameter changes experienced by a bulbus arteriosus and 165 an artery dur ing inf lat ion f rom diastole to systole. x v LIST OF EQUATIONS 1 Strain (e) = L / L n where L is the extension and L n is the unstressed (or ig inal) length. 2. Stress (o) = f/Ao(l +e c) where f is measured force, A n is the in it ia l cross-sectional area and e c is the mean value o f strain over wh i ch the stress has been calculated. Stress was expressed as true stress and was calculated assuming constant vo lume 3. Incremental modulus ( E j n c ) = (1 - v 2 ) Aa/Ae c (Be rge l , 1961) where v is Po i s son ' s ratio for the artery or bulbus. Po i s son ' s ratio is the negative quotient o f c i rcumferent ia l strain and longitudinal strain. In most soft tissues, Po i s son ' s ratio is very close to 0.5 (Bergel, 1961). 4. V o l u m e ( V ) = l/37rr 2l where r = radius and 1 = length 5. Average wa l l stress in a th in-wal led cyl inder. Stress (a) = Pr/h where P = pressure, r - radius and h - thickness 6. Stress at a specif ic point in the wa l l o f a th ick -wal led cyl inder. o~ m a x = pi(a 2 + b 2 )/(b 2 -a 2 ) where a = inner radius, b = outer radius, and p; equals interior pressure. (Harvey, 1974) 7. L a w o f Laplace. T = P * R where T = tension, P = pressure and R = radius. xvi ACKNOWLEDGEMENTS I'd l ike to thank my supervisors D a v i d Jones and John Gosl ine. It was a pr iv i lege to work w i th and learn f rom two such distinguished men. Du r i n g the course o f my degree, they taught me much more than biology. 1 sincerely appreciate Dave for taking the t ime to expose me to opera, darts, soccer and the finer things in l ife. W h i l e John exposed me to garl ic, lots and lots o f garlic. Thank you for mak ing my degree so rewarding and enriching. I'd also l i ke to thank R i cha rd B r i l l for a l l ow ing me to work at his faci l i t ies in Hono lu l u . W i thout his participation, much o f the work wou ld never have been completed. F ina l l y , I'd l i ke to thank my fami l y and friends for their support throughout this endeavour. xvii INTRODUCTION Circulation A l l organisms require c i rculat ion o f nutrients and fuel and uti l i se d i f fus ion as the pr imary dispersal mechanism for br ing ing fuel to cells. F o r much o f evolut ionary history, organisms have maintained an eff ic ient distr ibution o f nutrients and fuels by max im i s i ng d i f fus ion. However , as animals became larger and more active, re ly ing on s imple d i f fus ion for the relay ing o f necessary substances across large distances to the tissues became increasingly problematic. T o compensate, vertebrates and several large, active invertebrates (the cephalopods) (Shadwick and Gos l ine, 1985; Shadwick et al., 1987; Shadwick and N i l s son, 1990) developed closed c irculatory systems. Instead o f requir ing d i f fus ion for the movement o f substances through the body as in an open c irculatory system, these animals ut i l i sed pressure gradients to propel f l u id (blood) through ho l l ow tubes (b lood vessels) that branched throughout the body. A s the branching increases in complex i ty, the tubes become smaller unti l the distances separating the cells f r om b lood become smal l enough to a l low quick d i f fus ion o f oxygen and nutrients. Problems of the Heart W h i l e the advent o f the closed circulatory system a l lowed animals to greatly increase in size and complex i ty, it introduced new problems due to the use o f a pulsati le pump (heart) to push f lu id through tubes. The beating o f a heart results in periods where the f l ow is either stopped or reversed, l im i t ing the perfusion o f the terminal capi l lar ies and creating a large impedance (resistance to f low). These problems increase both the wo rk o f the heart and the overal l inef f ic iency o f the circulatory system. 1 Haemodynamic Solutions The entire field o f haemodynamics is an answer to a single question: how do different organisms deal w i th the problems inherent.in a pulsati le pump? The general solut ion to the problem o f a pulsati le heart is to use elastic arteries that can expand dur ing systole, store energy and reco i l dur ing diastole, returning the stored potential energy and mainta in ing forward f low. However , several variations on this general theme occurred dur ing evolut ion. In the simplest or Windkesse l model, the aorta and major arteries are considered to be a single elastic reservoir l i nk ing the heart in series w i th the peripheral resistance. Despite a pulsati le i n f l ow into the system f rom the heart, the out f low into the periphery is relat ively smooth. This is due to the mechanical properties o f the arteries wh i ch a l lows them to expand and recoi l elast ical ly w i th each pulse o f b lood ejected f rom the heart. The arteries expand dur ing systole and recoi l dur ing diastole, maintaining b lood f l ow through the periphery. However , one o f the basic tenets o f the Windkesse l is that all pressure changes occur s imultaneously throughout the system, in effect, the pressure wave is transmitted at an inf inite velocity, and the whole system inflates in synchrony. (McDona l d , 1974). W h i l e this condit ion occurs in ectothermic vertebrates (Gibbons and Shadwick, 1991), mammals and birds have more compl icated haemodynamics. W i t h a l ow heart rate, the wavelength is very long, and therefore transmiss ion t ime o f the pulse through the arteries is a negl ig ib le proportion o f the cardiac cycle. However , when the pressure wavelength is less than eight to ten times the length o f the arterial tree, there is an abrupt shift in haemodynamics f rom a Windkesse l to a wave 2 transmiss ion system (M i lnor , 1982). Therefore, mammals and birds are more appropriately model led as wave-transmiss ion systems because the arterial pressure waves o f homeotherms are propagated over distances approaching their wavelengths. The pressure pulse is no longer considered to move inf in i te ly fast but has a f in ite transmiss ion t ime through the arterial system and arrives later at more distal sites. Th i s results in complex wave propagation effects such as pulse ampl i f icat ion, distort ion and reflections. Th i s system is p r imar i l y due to the much higher heart rates o f the mammals and birds when compared with s imi lar ly sized ectotherms. The benefits o f the wave transmission system occur when the length o f the arterial tree is matched to the heart rate so that the heart sits roughly 0.25 wavelength f rom a major ref lect ing site (a branch point or capi l lary bed). This results in the reflected pressure wave being 180 degrees out o f phase w i th the incident wave at the heart. The two waves destructively interfere, lower ing the pressure amplitude at the heart. A t the periphery, however, the two waves are in phase and sum, generating h igh ly osc i l latory pressure pulses. Under the same conditions, the flow waves are reflected in an opposite manner; summing at the heart and being out o f phase at the periphery. Th i s generates a h ighly osc i l latory out f low f rom the heart wh i le mainta in ing a much smoother flow through the periphery. A major benefit o f this " t uned " transmiss ion l ine is that it lowers the impedance o f the entire system, thereby lower ing the work o f the heart and increasing eff ic iency. A t the heart, the combinat ion o f a small pressure pulse w i th a large flow pulse results in a l ow impedance. Therefore, the heart is effect ively uncoupled f rom the terminal resistance by wo r k i n g against l ow impedance. 3 The benefits o f the tuned wave transmission system are a by-product o f the high heart rates associated w i th endothermy. In order for ectotherms to achieve the quarter wavelength frequency, heart rate wou ld have to greatly increase. Shadwick et al. (1987), found that an octopus wou ld have to increase its heart rate 10-fold in order to achieve the uncoup l ing found in mammals and birds, an increased energy cost that is l i ke ly higher than any benefits received f rom the lowered impedance. Haeinodynaniic Solutions in Fish M o s t studies o f cardiovascular dynamics have occurred in terrestrial organisms; few studies have looked at haemodynamics in fish. In ye l l ow f i n tuna, Jones el al. (1993) suggested that the combinat ion o f large size and high heart rates (> 100 min" 1 ) w o u l d a l l ow for s ignif icant transmission effects. However , no evidence o f wave transmissions was found. Instead, when Four ier series o f pressure and f l o w waveforms were examined, there was a good match w i th theoretical and mathematical Windkesse l models. I f the large and cardiovascular ly robust tunas are Windkessels, then the combinat ion o f small size and l ow heart rate makes other fish cardiovascular systems Windkesse l s as we l l . D u e to the nature o f the fish c irculatory system, it's un l i ke ly that any f i sh cou ld possess anything but a Windkesse l . In fish, the ratio o f arterial length to wavelength is much smaller than the 0.25 required for a tuned transmission l ine because o f the short ventral aorta separating the heart f rom the gil ls. Indeed, the short artery separating the heart f rom the g i l l s even seems inadequate for proper Windkesse l functions, as it does not possess enough compl iance to effect ively depulsate and maintain f low. 4 Birds , mammals ( M c D o n a l d , 1974), amphibians (Gibbons and Shadwick, 1991) and cephalopds (Shadwick et al., 1987; Shadwick and N i l s son, 1990) make use o f a long arterial tree to attenuate pressure waves sent out by the heart. In fish, the large resistance o f the g i l l capi l lary bed is separated f rom the heart by a short ventral aorta (Figure 1 A) . Th i s creates a potential problem because the ventral aorta is too short to effect ively depulsate pressure and maintain f lows. It is not long enough for the ref lect ion effects seen in higher vertebrates, and it does not provide sufficient capacitance to act as a Windkesse l . However , recordings o f pressure and f l ow f rom the prox imal aorta o f teleosts (Johansen, 1962; Stevens et al. 1972; Jones et al, 1974; Satchel 1, 1991; Jones et al, 1993) show a s ignif icant amount o f smoothing, reveal ing the presence o f a large elastic element in the teleost c i rculat ion (Figure IB) . The smoothness o f teleost f lows is in marked contrast wi th those f rom the prox imal aorta o f a bird, mammal or amphibian, where h ighly disturbed, and often reversed, f l ow is the norm (McDona l d , 1974). In mammals, posit ive arterial f l ow occurs for less than one-third o f the cardiac cyc le (Spencer and Greiss, 1962), wh i le posit ive f l ow in the ventral aorta exists for more than three-quarters o f the cardiac cyc le in f i sh (Johansen, 1962). The surpris ing smoothness o f f l o w and pressure in teleosts is due to the bulbus arteriosus. The Bulbus Arteriosus: location and gross morphology The bulbus arteriosus is the fourth chamber o f the teleost heart and is distal to the ventricle. It is located wi th in the pericardial cavity and separates the ventricle f rom the ventral aorta (Figure 1C). A l l bulbi are swol len prox imal l y and taper distal ly into the 5 Figure 1. A. Schematic comparison of the circulations of mammals and fish. The highlighted portions are considered equivalent (modified from Schmidt-Nielson, 1975). B. Pressure and flow traces from a yellowfin tuna (Thunnus albacares). Ventral aortic pressure (PVA). Dorsal aortic pressure (PDA), Ventricular pressure (Pv), and Ventral aortic flow (QVA) (Jones et ai., 1993). C. The heart of the rainbow trout showing the four chambers: sinus venosus, atrium, ventricle and bulbus arteriosus (from Randall, 1970). D. r-shaped inflation curve vs. J-shaped inflation curve M a m m a l 7 ventral aorta, however, w i th in these parameters, a large variety o f shapes occur. The funct ional s ign i f icance o f the different shapes is unknown (Santer, 1985). L i kew i se , the internal shape o f the bulbus shows a great deal o f var iab i l i ty between species. The lumen o f the bulbus can be smooth, ridged or trabeculated. The trabeculae can be "spongy" or separated into discrete longitudinal and radial elements. In one species o f deep-sea Macroiiridae, the prox imal port ion o f the lumen contains a carti laginous inner tube separate f rom the outer, spongy trabeculae (Greer -Walker et al., 1995). U n l i k e the other chambers o f the heart (sinus venosus, artrium and ventricle), the bulbus arteriosus contains no cardiac muscle and is not actively contractile. It is an elastic tissue that contains the same three components as arteries: elastin, col lagen and smooth muscle. Bulbus Arter iosus: functions? Researchers have described the bulbus as a Windkesse l for nearly seventy years (von Skraml ick, 1935, cited in Priede, 1976; Mott , 1950; Satchell, 1971; Stevens et al 1972; L i c h t and Harr is, 1973; Jones et al. 1974; Priede, 1976; Farre l l , 1979; Watson and Cobb, 1979; Ben jam in et al. 1983; Ben jamin et al. 1984; Santer, 1985; Bushne l l et al. 1992; Jones el al. 1993). M u c h o f the evidence for the bulbus's funct ion come f rom the recordings o f pressure and f l ow in the prox imal aorta o f teleosts. A s mentioned previously, the f l ows are much smoother than those f rom a s imi lar location in a bird or mammal . Pressure and f l ow recordings f rom teleosts are also much smoother than those f rom elasmobranchs (Satchell and Jones, 1967; Metca l fe and But ler, 1982), wh ich lack a bulbus arteriosus. Deta i led analyses o f pressure and f l ow traces recorded f rom teleosts 8 have also resulted in predictions regarding the exact effects o f the bulbus arteriosus. Stevens and Randa l l (1967) calculated that the bulbus is capable o f tak ing up 2 5 % o f the stroke vo lume when smoothing f lows, and Pr iede (1976) and Stevens et al. (1972) suggest that bulbus contributes 2 5 % and 2 9 % o f post-systol ic ventral f low, respectively. W h i l e many researchers have stated that the bulbus acts to depulse pressure and maintain f l ow, only a few (L i ch t and Harr is, 1973; Priede, 1976; Farre l l , 1979; Bu shne l l et al., 1992) have looked at whether the bulbus actually functions in the manner ascribed to it. Unfortunately, the studies were rudimentary, l imited to cursory analyses o f static inf lat ions and gross morphology. H o w the bulbus arteriosus m im ic s the effects o f a much longer arterial system has never been explained. Bulbus Arteriosus vs. Artery: wall structure W h i l e the functional aspects o f the bulbus arteriosus have been neglected, the morpho logy has been extensively studied. Whether the bulbus is cardiac or arterial in nature is an open question (Benjamin et al, 1983), however, due to some obvious s imilarit ies, arterial nomenclature is used to describe the bulbar morphology. A s in arteries, the bulbar wa l l is composed o f three layers: an int ima composed o f a single layer o f endothelial cells, a thick media pr imar i ly composed o f elastin and smooth muscle and a col lagenous adventit ia surrounded by an outer layer o f mesothel ial cells. The endothelial cel ls can be squamous, co lumnar or cuboidal (Leknes, 1981; Ben jamin et al. 1983, Ben jamin el al., 1984; Leknes, 1985). Ben jamin (1982) and Leknes (1981) discovered that several species o f teleosts had cells that stained strongly wi th per iodic ac id - Sh i f f s ( PAS ) , indicat ing the presence o f carbohydrates. However, after 9 examin ing 80 different species, Ben jamin el al (1984) demonstrated that P A S pos it iv i ty is not a trait common to al l teleosts. The endothelial cells also contain membrane-bound bodies that can have a low, medium or high electron density (Leknes, 1981), and there seems to be a correlation w i th the number o f membrane-bound bodies and P A S pos i t iv i ty (Benjamin et al., 1984). There is microscop ic evidence for their discharge into the bulbar lumen (Benjamin et al., 1983), but the contents o f the membrane-bound bodies is unknown. Except for the layer o f mesothelial cells, the adventit ia is almost entirely col lagen (Benjamin et al., 1983; Bushhe l l et al., 1992) and is thought to l imit bulbar strain (Priede, 1976; Raso, 1993; lcardo etal, 1999a; Icardo et al., 1999b). No t all bu lb i receive a b lood supply Or innervation, but in those that do, the b lood vessels and large nerve bundles are conf ined to the adventitia (Watson and Cobb, 1979). Smaller nerve bundles only penetrate a few microns into the media. The media forms 90-95 % o f the bulbus and is composed o f smooth muscle sparsely distr ibuted wi th in an elastin f ramework (L icht and Harr is, 1973; Watson and Cobb, 1979; Icardo et al, 2000). However , exceptions do exist, as the bulbus o f Pleuronectes platessa contains no smooth muscle (Santer and Cobb, 1972). It is in the media where the morphologies o f bu lb i and arteries diverge. In the arteries o f mammals, concentric laminae o f elastin are interspersed w i th layers o f smooth musc le and col lagen fo rming elastic units cal led lamellae (C lark and G lagov, 1985). The lamel lar unit is the basic unit o f the arterial wa l l . B i gge r arteries have more units and smaller arteries have fewer (C lark and Glagov, 1985), and, with a few exceptions (Gos l ine and Shadwick, 1996), the lamellar unit appears to have been conserved dur ing 10 mammal ian evolut ion (Wo l i n s ky and Glagov, 1967). The structure o f the ventral aorta o f teleosts is not so constant. In trout, smooth muscle is arranged both c i rcumferent ia l ly and long i tud ina l ly (Seraf ini-Fracass ini et al, 1978). E last ic laminae can be found throughout the ventral aorta o f carp, eel, ra inbow trout and ye l l owf in tuna (Seraf in i -Fracass in i et al, 1978; Leknes, 1986, Bushne l l et al, 1992). However , in many other species o f f ish, elastic laminae only rarely occur between muscle layers and are general ly l imi ted to the subendothelial space (Seraf ini-Fracass ini et al, 1978; Leknes, 1986). In bu lb i , the smooth muscle cells are only loosely arranged w i th respect to the long axis o f the vessel. In contrast to arteries, elastin laminae are not found in the bu lb i o f teleosts. Bu l ba r elastin occurs in a loose meshwork o f f ibr i l s ( Seraf ini-Fracass ini et al, 1978; I sokawa et. al, 1990; Icardo etal, 2000). Bulbus Arteriosus vs. Artery: elastin The elastin laminae in mammals are made o f elastic fibers. These elastic f ibers are composed o f two components: amorphous elastin and elastin-associated mic ro f ib r i l s (Cleary and G ibson, 1983). The microf ibr i l s are not thought to be load-bearing but serve as a scaffold upon wh i ch elastin is deposited, fo rming the elastic fiber. In teleost aorta, the elastic f iber is quite different. Elastin-associated microf ibr i l s are rarely observed ( I sokawa et al, 1990) and the elastin is not found as an amorphous component wi th in the elastic fibers. Instead, elastin occurs as 25 nm f ibr i l s (Seraf in i -Fracass in i el al, 1978; I sokawa et al 1988) woven at random into a framework; although the f ibr i l s do show some c ircumferent ia l alignment. The fibri ls, when aligned, occas ional ly coalesce, and resemble the amorphous elastin o f mammals (Seraf ini-Fracass ini el al, 1978; I sokawa et 11 al. 1988). In the bulbus, the f ibr i l s are less ordered, never coalescing. Th i s may be due to the lack o f laminae wi th in the bulbar wa l l . Despite the large differences between the forms o f mammal ian and teleost elastin, s imilar it ies do occur at a higher magnif icat ion. The 25 nm f ibr i l s in the ventral aorta and bulbus can be resolved into 2.5 nm pr imary f i laments w h i c h are indist inguishable in dimensions f r om those present in bov ine l igament preparations (Seraf ini-Fracass ini etal, 1978). E las t in differs w ide ly in amino acid compos i t ion between vertebrate groups. Compared w i th mammal ian elastin, teleost elastin shows a high degree o f variance. It is composed o f a very different set o f amino acids, being enriched in g l yc ine and polar residues and depleted in desmosines and val ine (Sage and Gray, 1977). Salmd'nid elastin shows the lowest cross- l ink ing content, and, despite having analogous behavior under stress, the Young ' s modulus o f the Sa lmonid elastin is only ha l f that o f bov ine (Seraf in i -Fracass in i et al, 1978). These molecular differences are reflected in large differences between the staining characteristics o f bulbar and bov ine elastins ( L i ch t and Harr i s , 1973). D i f ferences also occur between elastins in the bulbus arteriosus and ventral aorta wi th in a single species. Carp appear to have two b iochemica l variants o f arterial elastin. One is concentrated in the bulbus arteriosus and is soluble in a lka l i and acid. The other is found in the ventral aorta (possibly w i th the bulbar variant) and, l ike mammal ian elastin, is a lka l i and acid insoluble. Staining characteristics ( L i ch t and Harr is, 1973) also reveals changes between the two. 12 Bulbus Arteriosus vs. Artery: functional differences? When an artery is inflated, the resulting pressure vo lume loop is described as " J -shaped" (F igure ID). However , when a bulbus is inflated, the pressure vo lume loops is very different, fo rming an r-shaped loop (L icht and Harr is, 1973; Priede, 1976; Bushne l l el al., 1992). This curve is unique. In arteries, the J-shaped stiffness is the result o f elastin and col lagen wo rk i ng in conjunct ion as a stra in- l imit ing system (Wa inwr i gh t et al., 1976). When the col lagen in the arterial wa l l is recruited to prevent excessive expansion, the material becomes stiffer as the elastic modulus increases and, as a result, the wa l l stress begins to increase sharply. B u l b i also possess elastin, smooth muscle and col lagen arranged in a s imi lar manner. Yet, when inflated, the bulbus appears to be very st i f f in i t ia l ly, f o l l owed by a "p lateau" in wh i ch it becomes very compliant. Origin of the Bulbus Arteriosus: cardiac or arterial? Santer, 1985, describes the question surrounding the bu lbus ' s or ig in as unresolvable: is it cardiac or arterial? Those in favour o f a cardiac or ig in suggest that, because o f its locat ion with in the pericardial cavity, it is, by def in i t ion, part o f the heart (Priede, 1976), derived f rom the conus arteriosus is elasmobranches. The conus arteriosus contains cardiac muscle and act ively contracts during each cardiac cycle. However , the bulbus has neither cardiac muscle nor is it actively contractile. A l though it is different f rom arteries in a number o f ways, its compos i t ion is much closer to arteries than to the conus arteriosus. In fact, it appears that the conus has not become the bulbus but has been replaced by the bulbus. In teleosts, a ring o f muscle occurs at the junct ion between the bulbus and the ventr ic le that is thought to be the remnant o f the conus (Bushnel l el al, 13 1992). The remnant o f the conus, separating the bulbus f rom the ventricle, clearly indicates that the bulbus is a more recent structure that has g rown posteriorly towards the heart. The various maladies wh i ch the bulbus is susceptible to are all t yp i ca l l y arterial. M ine ra l i z a t i on o f the bulbus arteriosus was found in adult ra inbow trout (He ide l et al. 1997), arteriosclerosis was found in steelhead trout (Kubasch and Rourke, 1990), aging changes in the heart o f guppies included deposit ion o f col lagen fibers in the bulbus arteriosus (Woodhead, 1984) and a dissecting aneurysm was discovered in a st ick leback (Benjamin, 1985). Bu l bu s wa l l design is different f rom that o f a mature artery. However , morpho log ica l data shows that it is s imi lar to the ventral aorta o f a larval teleost. Leknes (1986) describes the larval ventral aorta as having muscle cel ls that are r ich in membrane-associated ribosomes, and containing a wel l -deve loped Go l g i apparatus surrounded by membrane-bound, electron dense bodies. The muscle cells are surrounded by elastin f ibr i l s , and no elastin lamellae are seen in up to 21-day-old larva. W h i l e the elastin may not have been l inked into lamellae, some al ignment o f the free elastin f ibr i l s d id occur. The embryon ic artery even possessed endothelial covered protusions o f variable height and w idth that extended into the lumen; possible precursors to longitudinal elements. Th i s descr ipt ion o f the larval aorta is very s imi lar to what is found w i th in the bulbus and a l lows the formulat ion o f the hypothesis that the bulbus is a neotenic structure, very s imi lar to a larval aorta. I f a portion o f the aorta were to g row into the pericardial cavity, many o f the selection pressures normal ly operating on arteries wou ld be l ifted. W i t h the danger o f over expansion no longer a serious threat, selection on that segment o f aorta 14 cou ld favor increasing distensibi l ity. Mod i f i ca t i on s a l l ow ing the bulbus to become an effective Windkesse l ( increasing elast in-col lagen ratio, loss o f lamellae, l imi ted c i rcumferent ia l ly arranged collagen) wou l d result in a selective advantage. There is funct ional as we l l as morpholog ica l evidence suggesting that the bulbus ' or ig in is arterial. It appears to be fundamental ly separated f rom the rest o f the heart in how it responds to pressure loads. In trout subjected to pressure overloads in vivo and in vitro, protein synthesis increased in the atria and ventricle without any change in the bulbus (Hou l ihan et al, 1988). C la rk and Rodn i c k (1999) also show that in trout wi th a w ide range o f R V M s (relative ventr ic le mass) due to vo lume and pressure overloads, neither bulbus mass nor compl iance changed as R V M increased. Th i s evidence strongly suggests that bulbar and ventricular functions are not closely l inked. The overwhe lming evidence appears to show that the bulbus, despite being w i th in the per icardial cavity, is arterially derived. Goals of the Study These experiments were designed to examine the bulbus inf lat ion curve w i th specif ic regard to its causes. Despite the tremendous amount o f var iabi l i ty between bu lb i (shape, wa l l morphology, elastin-smooth muscle interactions, innervation), the goal was to f ind a set o f rules that could explain the unique pressure-volume loops in a l l bulb i . Once the causes o f the unique bulbus arteriosus P -V loop were understood, 1 hoped to address several related questions: 1. Is the pressure-volume loop generated during static inflations an accurate representation o f what occurs in the animal? 1 2. D o the special inf lat ion dynamics impact how the bulbus performs its W indkes se l funct ion? 3. H o w does the bulbus affect the phys io logy o f the fish? MATERIALS AND METHODS Pressure Volume Loops The experiments were performed on bulbi isolated f rom Thunnus albacares (yel lowfin tuna), Thunnus alalunga (albacore tuna), Thunnus obesus (bigeye tuna), Coryphaena hipporus (mahimahi), Taractichthys steindachneri (pomfret), Makaira nigricans (blue marlin), Tetrapturus audax (striped marlin), Istiophorus platypterus (sailfish) and Oncorhynchus mykiss (rainbow trout). Ra inbow trout were held in outdoor tanks at the Univers i ty o f Br i t i sh Co lumb ia at temperatures that ranged f rom 4-10 ° C. Ye l l ow f i n tuna were held in large outdoor tanks at the Nat ional Mar ine Fisheries Service K e w a l o Research Fac i l i ty in Hono lu lu , H I , U S A . The water temperature in the holding tanks was 25° C. The other species o f fish were obtained f rom the fish auction in Hono lu l u and had been caught by local fishermen. Ra i nbow trout were ki l led wi th a b low to the head. Fresh ye l lowf in and bigeye tuna bulbi were obtained f rom animals which had died during the course o f other cardiovascular experiments. After death, an incision slightly posterior to the gills was made on the ventral surface, exposing the pericardium. The pericardium was slit in the midline. The length o f the bulbus arteriosus was measured in situ. The bulbus arteriosus was removed by cutting posterior to the ventriculo-bulbar junct ion and anterior to the bulbo-aort ic junct ion and placed in saline. T w o ye l lowf in tuna bulbi, however, were tested in situ w i th the pericardium intact. F o r the large bulbi obtained f rom species at the auction (yel lowfin tuna, albacore, blue marlin, striped marlin, sailfish, mahimahi, and pomfret), a different approach was 17 uti l ized. The bulbus was removed through a lateral incis ion underneath the operculum. This technique was possible due to the large size o f these species, and, more importantly, it prevented damage to the saleable portions o f the fish. Mechan ica l testing o f b lood vessels was l imited to ye l lowf in tuna and blue marlin arteries. Sections o f blue marlin ventral aortas were dissected away f rom fish being sold at the Hono l u l u fish auction. Ye l l ow f i n tuna aortas were dissected f rom freshly k i l led animals. D u e to the morphology o f the dorsal aorta, removing an intact segment for testing was problematic. The vessel is tightly bound to the spinal column, and our attempts to separate the dorsal aorta f rom the spine resulted in leaks. Therefore, all dorsal aortas were inflated in situ. F o r two yel lowf in tuna, the dorsal aortas were heavily parasitized by the larval cestode Dasrynchus talismani (B r i l l , 1987). The bulbus was double cannulated using P E tubing wi th flared ends, and ligatures were placed behind the flare around the bulbo-ventricular and bulbo-arterial junctions. The anterior cannula was attached to a pressure transducer, whi le the posterior cannula was attached to a syringe fi l led wi th R inger ' s solution. The bulbi f rom smaller fish (rainbow trout, some ye l lowf in tuna) were held in a tissue bath at their in vivo length. The bulbi o f the large tuna, marlin, sailfish and mahimahi were not held at any specific lengths because we were unable to measure their in vivo size during removal through the lateral incision. D u e to the large difference in sizes o f the bulbi, no single size o f tubing or infusion syringe was used. Instead, the sizes o f tubing or syringes were picked to best fit the size o f the bulbus. F o r the tubing, the size that most closely approximated the diameter o f the ventral aorta was chosen, and for the syringe, we chose the size that was big enough to exceed 18 •D C CO E >-T3 "D 0 CO iS CD 1 • CO CO _Q CO CO < CD co E c £ g CD .£ co i - (]) <u o E g> § 8 i s CO c S i CD Q. =5 7~ CO co sz o>l-5 | _Q D) £ I I II *** "O — CD CO 3 CO _ Ct 3 ~ E 8 M T> TJ * C (D £ a ) P = -C O i ? 2 § 19 the capacity o f the bulbus, whi le still being small enough so that accurate measurements o f small volumes could be made. F o r the ra inbow trout, the temperature o f the R inger ' s was maintained at 10-15° C, whi le for the tropical fish the R inger ' s was left at r oom temperature, 25° C. The arteries were also cannulated in both ends, w i th one end attached to a pressure transducer and the other to an infusion syringe (Figure 2). A measured volume o f f lu id was then injected into the bulbus or artery, and the resultant pressure signal was amplified and recorded using either Labtech No tebook or Dasylab software on a computer (Figure 2). Cycles o f inflation and deflation were performed until consistent results were seen. Precondit ioning usually required five to ten cycles for the bulbus and three to five cycles for the arteries. These initial cycles were discarded. E a ch experiment consisted o f 8-15 trials, and results f rom any trials in which a loss o f more than 5 % o f the injected saline occurred were not used. A f te r precondit ioning, the data were recorded and plotted as pressure in mm o f mercury versus vo lume in ml or p i . T o a l low comparison between different sized fish or different vessel lengths, injected volume was normalized to the maximum vo lume o f the vessel, resulting in plots o f pressure versus a unitless volume with a maximum o f one. The max imum volume to which the bulbi and arteries were normalized was established experimentally. Performing similar inflations on a number o f different bulbi and arteries a l lowed us to ascertain when they appeared to be near bursting. Us ing both the pressures reached during inflations and the appearance o f the bulbi at these high pressures, we wou ld decide at which point to stop adding fluid. G o i n g past this point often resulted in 20 failure o f the preparation; either a rupture o f the tissue wou ld result, or the high pressures wou l d begin forc ing fluid out around the cannulae. Blocking of Smooth Muscle and Denaturation of Collagen A l l bulbi used in these experiments came f rom freshly k i l led ye l lowf in tuna (n=4) and ra inbow trout (n=6) and were mounted in a tissue bath ful l o f R inger ' s solution. A f te r pre-condit ioning the bulbus, inflation-deflation cycles were performed which were considered to be indicative o f the normal pressure-volume behaviour. The bulbus was then placed in a 10"5 M solution o f the C a 2 + channel blocker, nicardipine, for 10-15 minutes, whi le the interior o f the bulbus was fi l led with the same nicardipine solution. The bulbus was again precondit ioned, fo l l owed by several inflation-deflation cycles. N icard ip ine is light sensitive, so these trials were performed in dim light and the bath was covered with a black cloth. F i sh col lagen has thermally-sensitive inter- and intramolecular crossl inks that denature at quite l ow temperatures. Rose et al (1988) showed that the denaturation temperature o f halibut collagen (a cold-water fish) was 17 degrees Celsius, whi le the denaturation temperature o f big-eye tuna (a warm-water fish) was 31° C. The protoco l for all fish involved placing the bulbus in water that ranged f rom 60-100° C for 30-45 minutes. The bulbus was then cooled down to experimental temperatures (10-15° C for the trout, 25° C for the ye l lowf in tuna). Fo l l ow ing this, the bulbus was again precondit ioned before performing a series o f inflation-deflation loops. The data f rom different fish were averaged, and the different pressure-volume loops (1 .normal, 2.smooth muscle b locked, 3.collagen denatured and smooth muscle removed) were plotted on the same gr id and compared. 21 Pressure-Volume Loops of Modified Blue Marlin Bulbi The marlin bulbus arteriosus is long and thin. This makes it amenable to a number o f studies not easily performed on bulbi f rom other fish. The bulbi f rom two blue marlin were cut into anterior and posterior sections. Pressure-volume loops were done on each section. B l ue marlin bulbi (n=5) were turned inside-out to dissect away tissue. The order in wh ich the tissues were removed was 1. longitudinal elements, 2. remainder o f inner layer, 3. media. The wa l l was sequentially removed until only a very thin layer o f the outer bulbar wal l remained. Af ter each removal o f tissue, the marlin bulb was turned right way out and pressure-volume loops were performed. Fou r blue marlin bulbi were left inside-out and inflated. The normal protocols for generating pressure-volume loops were fo l lowed. Stress-Strain Experiments on Different Parts of the Bulbus Arteriosus (anterior, middle and posterior rings) La rge bulbi f rom yel lowf in tuna (n=6) and blue marlin (n=4) were obtained f rom the fish auction in Hono lu lu . Three loops were cut f rom each bulb: an anterior, middle and posterior loop (Figure 3). Thickness, w idth and diameter were measured w i th calipers to an accuracy o f +/- 0.2 mm. In addition to the rings consisting o f the entire wal l , tests were also conducted on rings consisting o f the adventitia or outer medial layers o f the bulbus. Ventra l aortic loops f rom blue marlin and ye l lowf in tuna were also tested. 22 CO o CD n CO CO 13 _Q 0) O CD > I c m .2 c S I 2 5 | i co E c CD oj co •° £ T3 H -C O <0 </, CO CD It — CO 5 1 .Q _ 3 .2 «o I t» CU 2 £ (0 3 CO «-CL "o CU CO CD 2 i— N C = o> CO = 1 ,1 o O CO (D O t - .-= 2-2 | o. a) a; c — _, co .55 H -a x • c .<S «"> co c £ CD 3 I i 2 M l U. E 3 CD 23 The loops underwent uniaxial force-extension tests. These tests were conducted using a custom built " s t retch ing " machine (Figure 4). The loops o f tissue were mounted over two L-shaped stainless steel bars and placed in a 20-25° C saline bath. The ends o f the stainless steel bars were filed down to a combined diameter o f 1.2 mm. One o f the bars was attached to the bath whi le the other was attached to a moveable crosshead through a force transducer. The force-extension tests were performed by s low cycl ic stretching v ia the crosshead at a constant velocity o f 6 cm min" 1. Precondit ioning o f the loops was achieved by performing at least f ive preliminary cycles until the force-extension behaviour o f the tissue was consistent. Strain (e) is the ratio o f the change in length divided by the initial length and was calculated by the formula e = L/Lo (1) where L is the extension and L 0 is the unstressed (original) length. Since the loops became flattened, once they were mounted, they were considered to be two parallel sheets o f tissue, each having a length one half the circumference o f the loop. L ( ) cou ld then be calculated f rom the measurements o f radius and wall-thickness o f a r ing (Figure 5). The midwal l radius was used to calculate circumference, as suggested by L i l l i e et al. (1994). Stress (a ) was expressed as true stress and was calculated assuming constant vo lume as CT = f/Ao( l+ e c ) (2) 24 CD C o CD E c • m i "GO 0 o to CO £ to t3 0 3 C CO £ to > <D ^ C C D . 3 (U p ° £ o E • 2 l: C " to c ,« ° o <5 as _ §" E ^ di .£ o S; u= o 5 0 5 o w o) = r -o a) IS « a5 £ o Q . CD CO CD to "2 CD to 55 o ? I | m to | | « CO 03 CD O > O E O ~ CO c CD CD CO 0 3 CD E co ^ ro .2 += " c ^ o t_ to .£ £ to to _ i _ m I I I l e t o p CO t U 1 0 = J= 5= O) CD LL - Q 25 wall thickness Figure 5. When calculating stress and strain for the bulbar loops, values of thickness and initial length (L0) are required. L0 is the length of the flattened loop, measured at the midwall, as described in Lillie et al. (1994). 26 where f is measured force, Ao is the initial cross-sectional area and £ c is the mean value o f strain over wh ich the stress has been calculated. Circumferential stiffness was calculated as the incremental modulus (E; n c). E i n c = (1- v2) Ao7Ae c (Be rge l , 1961) (3) where v is Po i s son ' s ratio for the artery or bulbus. Po i s son ' s ratio is the negative quotient o f circumferential strain and longitudinal strain. In most soft tissues, Po i s son ' s ratio is very close to 0.5 (Bergel, 1961) and for the purposes o f our calculations, we accepted this value. The stress-strain curves were smoothed using TableCurve. f ableCurve is a curve f itt ing program that generates a number o f different equations to describe the curve. Once an appropriate fit has been found, the equation can be used to generate the exact dependant value o f any desired independent variable (the y-value for any input x). This a l lowed the comparison o f different stress-strain curves at the same values o f strain. Tensile Tests on Isolated Pieces of the Bulbus Arteriosus Strips o f tissue were dissected f rom the bulbi o f ye l lowf in and bigeye tuna. The tissue was cut into a rectangular shape, and the dimensions were measured using calipers to an accuracy o f +/- 0.2 mm. One end was glued transversely to a piece o f metal attached to a micromanipulater. The other end o f the tissue was attached to a Grass force transducer (model F T 0 3 ) to record forces produced as the material was stretched. 27 The internal longitudinal elements were tested as intact structures. Complete strips o f the wal l were stretched in both longitudinal and transverse directions. The outer, adventitial layer o f the wal l and the inner, medial layer were dissected free and tested individually. Once force and extension had been recorded, the values o f stress, strain and modulus were calculated as described above. Video Dimensional Analysis (VDA) from Yellowfin Tuna Showing Dimensional Changes in the Bulbus During In Vivo and In Situ Pressure Recordings. Ye l l ow f i n tuna were anaesthetized using benzocaine and placed in a chamois leather cradle wi th the ventral surface up. A hose running aerated sea was placed in the mouth o f the anesthetized fish in order to simulate ram ventilation. A midline incis ion was made along the ventral surface to expose the pericardial cavity. The pericardium was removed and the heart exposed. Du r i ng the experiment, Saffan was injected to maintain anesthesia. Bu lbus pressure was measured through a cannula inserted into the bulbus and connected to a Unonics M o d e l P-106 pressure transducer (Wayland, Massachusetts) (Figures 6, 7). P o p tests established the frequency response o f the system to be 32 H z , wi th damping 0.12 o f crit ical damping (Mi lnor, 1982). L ight anesthesia was maintained using periodic injections o f saffan (3 mg kg" 1 intrarterially). T o decrease the heart rate, water flow over the gills was stopped for several seconds. Changes in the diameter o f the bulbus during systole and diastole were measured using a video dimension analyzer ( V D A , Instrumentation for Phys io logy and Medic ine, 28 0 «+-» Q_ E o o _ § CD C > c 8 cD i f . C 0 ^ " ° O 5= O S i s c E T3 CO c co ••-» CD i— CO CO 2 a. CD co > x: £ eo co r- 3 -C ± 3 CO O) CO CD 3 P 3 C "O ft 42 O O TO CD .2> •^•o to — <u o N C "co x: to CD o CD •g -o "> is s CO S CO 3  C 5 S 2 CO CD •o c CO i — CD N >» CO c CO "co c o CO c 0) E T3 O 0 "D > CO JZ co M to 3 O i » E to co >>:Q CD E o CO 3 « C J2 « CO OJ •g > CO F £ S i— (O CO c to CD S . £ H CD CO CD CD "O =5 • e o CO •g to O & CD > ® CD o CD ^ CO co to •2 E ^ o c £ *- CD 3 Q. Q. JD ? § o a) » £ CO to H -. ~ o <0 T3 to 0) CD CD £ CO O) 3 O C Q) Q. CO •- x r U. CD O *J 3 3 Q. » E 3 O 0 o CO (0 CD r-1 § O T3 > CD • § CD CO CO >> CD CO 3 O CO g 0 CO 1 I N C _>> CO I CD CO CO 2 9 VDA window Figure 7. The set-up of theVDA study in a yellowfin tuna. It shows the bulbus in relation to the ventricle within the pericardial cavity. The catheter used to record pressure changes can be seen on the top of the bulbus. The VDA window is aligned with the bulbus such that the black portion of the window will follow the edges of the bulbus during a heartbeat. 30 model 303). This system consists o f a video camera, a video processor and a monitor. The camera was focused on the bulbus, and the signal fed through the processor. The V D A util izes the video signal to give a D C voltage that is proport ional to the distance between two selected contrast boundaries on the monitor. The V D A was used to track the movement o f the outside surface o f the bulbus as it expanded and contracted during systole and diastole (Figure 8). B y calibrating the voltage generated, dynamic dimensional changes can be recorded. F o r a more in depth explanation o f the V D A , see Fung (1981). Statistics Descr ipt ive statistics (mean and standard error) were calculated for the pressure-volume curves generated by b lock ing the smooth muscle and heating the bulbus, as wel l as for the all the stress-strain curves generated by the various extension protocols. The treatments in the b lock ing and heating study were compared using a two -way repeated measures A N O V A . F o r comparisons between the stress-strain curves for the anterior, middle and posterior segments o f the tuna and marlin bulbi, a three-way A N O V A was performed. Mu l t i p l e comparisons between the different treatments for the pressure-volume and stress-strain curves were performed using a Tukey test. A l l comparisons between groups were performed using SigmaStat 2.0. 31 CO c g o a. CO o ~ OT CO 3 c «* CD CD CD -*—< O _ > CD CD >>£ CD " d C L CD 5-1 CD 8 o >— T3 CD E 9-5 2 c CO CL h-o LU o _J O > CD CD O ) c CD CD 2? CO 5 £ > a -<D • -8 1 CO CD C CO -3 CO c co CO - CD | S | o a) E o co *s £ 3 CD 3 CO co CD E 5 -Q Q £ > 5 CD 5 CO £ o o III 0 0 > E P i ? a CD u. o CO 32 Histology T w o bulbi and ventral aortas f rom yel lowf in tuna were fixed in formal in and sent to Wax- i t H i s to logy Services (Aldergrove, B C , 604-822-1595) to be sectioned and stained. One bulbus was fixed whi le inflated, whi le another was fixed at zero pressure. The sections were stained w i th a Ve rhoe fF s elastic stain (Kiernan, 1998). Some sections were v iewed using a polar iz ing lens for evidence o f birefringent material. Typical ly, only highly ordered structures are birefringent. Col lagen, having a large degree o f internal order, appears very bright through the polar iz ing lens. Elastin, on the other hand, is not ordered and appears dark under the cross-polarized light. Electron Microscopy E lec t ron micrographs were prepared at the Univers ity o f Br i t i sh Co lumb ia E lect ron M i c r o s copy Faci l ity. Bu lbus samples were initially fixed in 2 . 5 % glutaraldehyde in 0.1 M sodium cacodylate ( p H 7.2). Fo l l ow ing f ixation, the samples were washed in 0.1 M sodium cacodylate buffer overnight and post-f ixed in 1 % osmium tetroxide. The samples were then subjected to a graded alcohol dehydration, stained using uranyl acetate and lead and embedded in Epon/Araldite. The blocks were sectioned with a glass knife and v iewed with a Ziess E M IOC. 33 RESULTS 1. Gross Morphology The bulbus arteriosus o f the ye l l owf in tuna is largest just outside the heart and tapers down towards the ventral aorta (Figure 9A) . Internally, it possesses a complex array o f longitudinal elements, elastic chords that arise f rom the ventr icular end as discrete structures. Towards the ventral aorta, the longitudinal elements are much more securely attached to the lumen wa l l , becoming ridges (Figure 9B). Compared w i th the bulbus o f the ye l l owf in tuna, the bulbus arteriosus o f the blue mar l in is much longer and thinner (Figure 9D). It is also much f loppier than that o f the ye l l owf i n . Internally, it also possesses longitudinal elements. M a n y o f the elements are attached to the bottom o f the pocket valve wh i ch closes o f f the bulbus f r om the ventr ic le (Figure 9E). The albacore bulbus is s imi lar to that o f the ye l l owf in tuna (F igure 9C). The bulbus o f the pomfret has a more extreme taper (Figure 9F). Re lat ive to its length, the post venticular swe l l i ng is very large, and the diameter dramatical ly decreases in size down to the ventral aorta. 2. Light Microscopy: yellowfin tuna W a l l S t r u c t u r e F igure 10 shows a series o f cross-sections taken f rom the bulbus arteriosus o f a ye l l ow f i n tuna. The sections were taken anterior to posterior, f rom near the ventral aorta (Figure 10 A ) to near the ventricle (Figure 10D). The bulbus is made up o f three layers: 34 Figure 9. A. The bulbus arteriosus from a yellowfin tuna. Note the coronary artery attached to the dorsal surface. B. A yellowfin tuna bulbus opened to reveal the longitudinal elements (I.e.) within. The longitudinal elements radiate out from locations near the ventriculo-bulbar edge. The radiation points of the bulbus are often the pocket valves that prevent backflow into the ventricle during diastole. C. The bulbus of an albacore tuna. D.The bulbus of a blue marlin. Contrary to the football-shaped bulbus of the tuna, the marlin bulbus is long, thin and cylindrical. E. The bulbus of a blue marlin cut open to reveal the longitudinal elements (I.e.). The I.e. of the marlin also radiate out from the pocket valves of the bulbus. Note that the wall thickness, relative to the length, is much smaller than in the tuna. F. The bulbus of a pomfret. The bulbus of the pomfret is very wide near the ventricle and rapidly decreases in width towards the ventral aorta (v.a.). 35 36 37 adventitia, tunica media and int ima (L i ch t and Harr is, 1973). The adventit ia (a) o f the y e l l o w f i n tuna bulbar wa l l is a thin f ibrous layer, containing mesothel ial cel ls and col lagen. The bulk o f the bulbar col lagen is concentrated in this outer layer. The adventit ia does contain a small amount o f elastin, however it is not as ordered or in as h igh a concentration as in the rest o f the bulbus. In F igure 10 A , one can see the coronary artery (ca) attached to the top o f the bulbus, continuous w i th the adventitia. The majority o f the bulbus wa l l is media (m). The media accounts for over 9 0 % o f the bulbus wa l l in carp (L icht and Harr is, 1973), approximately 9 5 % o f the wa l l in white bass, Mororie chrysops (Rafinesque) (Raso, 1993), and a s imi lar d iv i s ion is seen in ye l l ow f i n tuna. Despite containing bands o f smooth muscle running thoughout, the media is p r imar i l y made up o f elastin. The media is d iv ided into two layers, a dense outer layer contain ing elastin fibers running in a c ircumferential manner and a " l oo se r " inner media in w h i c h the majority o f the elastin fibers are oriented longitudinal ly. Th i s inner med ia contains the longitudinal elements (I.e.), chords o f tissue that run along the inner surface o f the bulbus. The architecture o f the longitudinal elements is most elaborate near the ventr ic le (F igure 10D). Thei r morphology is very irregular, w i th an overa l l " s pongy " appearance. Towards the anterior end o f the bulbus, the longitudinal elements become less complex. They do not project as far into the lumen, and the " s pongy " appearance disappears. They become more f i rm ly attached to the wa l l , and, near where the bulbus jo in s the ventral aorta (Figure 10A), they are little more than ridges along the lumen. The inner layer o f the bulbus, the int ima, is only a single endothelial ce l l layer l in ing the lumen. 38 Figure 10. Sequential cross-sections from the pressure-fixed bulbus arteriosus of a yellowfin tuna going from anterior to posterior. The bulbus wall is separated into three layers: adventitia (a), media (m), and intima. The intima is a single-cell layer, and cannot be resolved on these slides. The innermost portion of the media is broken up into longitudinal elements (le), chords of elastic tissue that run longitudinally through the lumen of the bulbus. A. Most anterior cross-section. The coronary artery (c.a.) can be seen attached to the adventitia. The longitudinal elements are little more than ridges at this point in the bulbus. B. This slide is taken from a section that is posterior to (A), both the I.e. and the lumen size are greatly enlarged. C. This slide is taken from a section that is posterior to (B). The enlarged lumen is becoming filled as the layer of longitudinal elements becomes even larger and more elaborate. D. Most posterior section. The ridged appearance of the longitudinal elements from the anterior sections is replaced with a thick, ornate layer of longitudinal elements that have a spongy texture. Stained with Verhoeff s elastic stain. 39 The different layers w i th in the bulbus arteriosus o f the ye l l ow f i n tuna are not the same throughout. F igure 11 compares cross-sections taken f rom the midd le o f the bulbus (Figure 11 A ) and near the ventricle (Figure 1 IB) . B o t h sections show the adventitia, inner and outer media, but the proportions are different. Wh i l e the size o f the col lagenous adventit ia remains relat ively constant in both sections, the outer media at the posterior end is approximately ha l f the thickness o f the same layer in the middle. Near the ventricle, the compact, c ircular layers o f elastin that make up the outer media decrease in number. Th i s is mirrored by an increase in the size and complex i ty o f the inner media, speci f ica l ly, the longitudinal elements. Under h igh magnif icat ion, the difference between the inner and outer media is more obvious. F igure 12 is sequence o f pictures taken f rom a single cross-section through the bulbar wa l l . The pictures are oriented so that the lumen is to the left and the outer wa l l is to the right. F igure 12A is f rom the outer edge o f the media, 12B the border o f the outer and inner media and F igure 12C, shows longitudinal elements. The outer portion o f the media (Figure 12A, B ) has a thick layer o f c i rcumferent ia l elastin fibers (cf), with the long axes o f the fibers obvious in this transverse section. In F igure 12B, an abrupt transition to longitudinal fibers (If) occurs, mark ing the transition between the inner and outer media. The long axes o f these elastin f ibers cannot be seen, instead, the longitudinal fibers show up as small c ircles, the result o f the transverse section. The longitudinal orientation o f elastin f ibers is maintained throughout the longitudinal elements (le) (F igure 12C). The elastin f ibers in the bulbar wa l l , wh i l e not in the discrete lamina o f arteries, show very specif ic order and internal organization. The clear spaces in the micrographs are l i ke ly cel lular. The bulbus contains 41 A., middle bulbus B . posterior bulbus Figure 11. Cross-sections comparing the middle and posterior bulbus. Underneath the collagenous adventitia (a), the media of the bulbus is divided into two layers, a dense outer media (om) with elastin fibers arranged circumferentially and an inner media (im) with elastin fibers arranged longitudinally. The inner media contains the longitudinal elements (le). A. The middle of the bulbus has a well-defined adventitia, outer media and inner media. B. The posterior of the bulbus also shows an adventitia, outer media and inner media but the proportions are changed. The outer media is half the thickness of the same layer in the middle bulbus (A.). This decrease in the thickness of the outer layer is mirrored by an increase in the size and complexity of the longitudinal elements. Stained with Verhoeffs elastic stain. 42 Figure 12. A series of pictures taken from a single cross-section through the bulbar wall. The pictures are oriented so that the outer edge of the bulbus is on the right and the lumen is on the left. A. The most obvious feature of the outer media is a thick layer of circumferential elastin fibers (cf). B. The transition point between the outer and inner media. The circumferential fibers (cf) give way to the longitudinally arranged elastin fibers (If) of the inner media. C. Within the longitudinal elements (le), the longitudinal orientation of the elastin fibers is maintained. Stained with Verhoeff s elastic stain. 43 a large number o f cel ls such as smooth muscle fibers and assorted fibroblasts. However , the Ve rhoe f f s elastic stain is specif ic to elastic tissue, result ing in the cel lu lar material appearing clear. A s w i th the thickness o f the wa l l , the differences between the outer and inner med ia are not constant along the length o f the ye l l owf i n bulbus. F igure 13 is a compar i son between anterior (Figure 13 A ) and posterior (Figure 13B) cross-sections. In the anterior cross-section, the majority o f the wa l l is composed o f c i rcumferent ia l ly arranged elastin fibers in concentr ic rings (Figure 13 A ) . This pattern is maintained until the media becomes d iv ided into longitudinal elements. A t that point, the fibers become long i tud ina l ly oriented (If). In contrast, the posterior cross-section (F igure 13B) shows two longitudinal layers o f elastin fibers (If) near the outer edge o f the media. Between the longitudinal layers, and for the rest o f the outer media, the elastin f ibers are c i rcumferent ia l ly arranged (cf). A s in the anterior section, these c i rcumferent ia l f ibers make up the majority o f the wa l l and are only replaced w i th longitudinal f ibers near the internal surface. A t that point, the longitudinal elements begin project ing into the lumen space. The ind iv idua l longitudinal elements in the posterior ha l f o f the ye l l ow f i n bulbus are discrete chords, attached to each other and the wal l s w i th radial cords. F igure 1 4 A is a cross-section that shows the transition between the c ircumferential f ibers o f the outer med ia (om) and the longitudinal fibers o f the inner media (im). A longitudinal element (le) seen in F i gure 14B has the typica l smal l circles representing transversely sectioned fibers. However , to the right o f the longitudinal element is an attached structure that 44 Figure 13. Cross-sections comparing the wall morphology at the anterior and posterior sections of the bulbus. A. In the anterior bulbus, the majority of the wall is composed of circumferentially arranged elastin fibers (cf). The circumferential orientation of the elastin is maintained throughout the wall until the media is divided into longitudinal elements. At that point, the elastin takes on a longitudinal orientation (If). B. In the outer media of the posterior bulbus, bands of longitudinally arranged fibers (If) are interspersed between the circumferential fibers (cf). However, the majority of the outer media is composed of circumferentially arranged elastin fibers. At the transition between the outer and inner media, the fiber orientation becomes longitudinal (If). Stained with Verhoeff s elastic stain. 45 A. anterior adventitia t lumen B . posterior adventitia t lumen 46 Figure 14. Cross-sections showing the three orientations of elastin within the wall of a yellowfin tuna bulbus. A. At the transition between the inner media (im) and outer media (om), the circumferential and longitudinal orientations of elastin can be seen. The long axes of the fibers in the outer media identify them as circumferential, while the small circles in the inner media show that the fibers have been transected, indicating a longitudinal orientation. B. The longitudinal elements (le) have the same elastin fiber pattern as the inner media. There are also radial elements (re) that attach the longitudinal elements to the lumen wall. The fiber pattern in the radial elements is neither longitudinal nor circumferential. This indicates that fibers were aligned at an intermediate angle to the plane of the section. Stained with Verhoeffs elastic stain. 47 48 possesses neither longitudinal nor c ircumferential fibers. Th i s is the radial element (re) attaching the longitudinal element to the wa l l . The bulbus arteriosus is def initely not an enlarged artery. F igure 15 demonstrates the disparity between cross-sections f rom a ye l l owf in bulbus (F igure 15 A ) and a cross-section f rom a ye l l owf i n ventral aorta (Figure 15B). These sections were taken f rom tissue separated by less than a cm and are at the same magnif icat ion. There is an obvious difference in w a l l thickness; however, it is w i th in the wa l l where the important differences are seen. The ventral aorta has a fa ir ly typ ica l arterial compos it ion. Layers o f smooth muscle (sm), sheets o f elastin cal led lamellae (el) and col lagen (co) are found in close prox imity, fo rming regular layers throughout the wal l . Smooth muscle jo in s adjacent lamellae, acting as " Spanmuske ln " (Dobr in, 1978). The elastin, co l lagen and smooth muscle fo rm lamel lar units, acting together to l imi t the strain in the aortic wa l l . Co l lagen is found w i th in the media o f the aortic wa l l as we l l as being very abundant throughout the adventitia. The anterior bulbus contains no elastin lamellae. The elastin f ibers (ef) are ordered; however, they are not jo ined into large continuous sheets. In the artery, the elastin lamel lae alternate with layers o f smooth muscle. The majority o f the bulbar wa l l is elastin fibers, wi th layers o f smooth muscle spread throughout the media and col lagen relegated to the adventitia. In this section (Figure 15 A ) , no smooth muscle can be seen, despite the same staining and magnif icat ion as the ventral aorta. However , smooth musc le does occur in the bulbus, hidden by the large amount o f elastin. The bulbar elastin stains very heavi ly w i th V e r h o e f f s elastin stain. W h i l e this results in much o f the smooth muscle being hidden f rom view, layers o f muscle can sti l l 49 Figure 15. Comparison between cross-sections taken from the anterior bulbus (A.) and ventral aorta (B.) of the yellowfin tuna. These sections were separated by less than a cm in the animal. A. The anterior bulbus is heavily stained and shows the circumferential alignments of the elastin fibers (ef) in the outer media. The orientation of the elastin fibers goes from circumferential to longitudinal within the longitudinal elements. B. The wall of the ventral aorta is much thinner than the wall of the bulbus. The ventral aorta does not stain as heavily as the bulbus. The ventral aorta has elastin lamellae (el) separated by the layers of smooth muscle (sm). Collagen (co) is very abundant within the adventitia. Stained with Verhoeff s elastic stain. 50 be seen. The smooth muscle layers are more dense in the outer media. F igure 16 is a longitudinal section o f the posterior bulbus near the ventricle. B e l o w the outer edge o f the bulbus is a layer o f smooth muscle (sm) (Figure 16A). A closer v i ew (F igure 16B) shows that the muscle is sandwiched between two layers o f longitudinal elastin f ibers (If). The musc le layer ends just beyond the top o f the frame in F igure 17 A. It continues down towards the ventr iculo-bulbar opening and appears to attach to a ventr iculo-bulbar va lve (Figure 17A). A posterior cross-section under a po lar iz ing lens (Figure 17B) shows a large amount o f ordered material occur ing near the lumen, wh ich corresponds w i th the ventr iculo-bulbar va lve (v) seen in F igure 17A. The fact that the va lve is h igh ly birefr ingent suggests that it is largely composed o f col lagen. F igure 17C is a cross-section f rom the middle o f the bulbus, again v iewed under polar ized light, and no birefr ingent material occurs in the posit ion o f the valve. The bulbar va lve does not project this far into the bulbus. W h i l e some o f the birefringent material that l ines the bulbar cross-sections is col lagen f rom the adventitial layer, much o f it is an artifact due to the f i x i n g process. Crystals are formed in the solutions and, being highly ordered, appear very bright under polar ized light. Blood Supply of the Bulbus The large ventr ic le o f the ye l l owf in tuna heart performs a great deal o f wo r k and requires a large number o f coronary vessels to supply oxygen and nutrients. W h i l e not act ively contracti le, the bulbus arteriosus also requires oxygen and nutrients in greater vo lumes than can be supplied w i th diffus ion. T o that end, the ye l l ow f i n bulbus possesses an extensive series o f b lood vessels. In addit ion to the large coronary artery that runs 51 Figure 16. Smooth muscle in a longitudinal section from the posterior end of the bulbus. A. Near the outer edge of the media, a layer of smooth muscle (sm) occurs. B. The same section under higher magnification reveals that the smooth muscle is sandwiched between two layers of longitudinal elastin fibers. Stained with Verhoeffs elastic stain. 52 53 Figure 17. A. Longitudinal section from the posterior of the bulbus. The smooth muscle layer (sm) is attached to the pocket valve (V) which separated the bulbus from the ventricle. B. Cross-section from the posterior bulbus. The valve (V) is birefringent under polarizing light, indicating it is composed of a highly ordered material. C. Cross-section from the middle of the bulbus. No birefringence occurs, indicating that the valve does not continue this far into the bulbus. Stained with Verhoeff s elastic stain. 55 Figure 18. A. Cross-section of a blood vessel (bv) in the adventitia of a yellowfin bulbus. The transection of the blood vessel indicates it has a longitudinal orientation. B. Longitudinal section of a blood vessel in the adventitia of a yellowfin bulbus. The transection of the blood vessel indicates it has a circumferential orientation. Stained with Verhoeffs elastic stain. 56 along the dorsal surface, there are a large number o f smaller vessels running in several directions through the bulbus. F igure 18A shows a b lood vessel (bv) transected dur ing a cross-section, indicat ing a longitudinal orientation. The b lood vessel in F igure 18B was transected during a longitudinal section, indicat ing a c irumferential orientation. A s in blue mar l in (Dav ie and Daxboeck, 1984), the majority o f b lood vessels run circumferent ia l ly, branching directly f rom the coronary artery. The vessels appear to be l imi ted to the adventitia, as is the case for most fish species examined (Watson and Cobb 1979). B u l b a r L u m e n at H i g h and L o w pressure The ye l l ow f i n bulbus is filled w i th longitudinal elements (le). A t zero pressure, they almost completely occlude the lumen (Figure 19A). When the bulbus is inflated, the longitudinal elements are pushed.to the side, opening the lumen ( L ) to f l ow f rom the ventr ic le (F igure 19B). 3. Electron Microscopy: yellowfin tuna In cross-section, there is a clear d iv i s ion between the adventit ia and the media (Figure 20) o f the bulbus arteriosus. Where the col lagen o f the adventit ia gives way to the smooth muscle and elastin o f the media, a transition point (arrowheads) is obvious. The layer o f mesothel ial cells (mc) can be seen outside o f the col lagen layer. Th i s ce l l layer is continuous w i th the visceral pericardium. Several nucleated erythrocytes can also be seen in the section. 57 Figure 19. Longitudinal sections from two bulbi. A. A bulbus fixed in an empty state. The longitudinal elements occlude the lumen. B. A bulbus fixed while inflated. The longitudinal elements are pushed to the side, opening the lumen to flow from the ventricle. Stained with Verhoeff s elastic stain. 58 Figure 20. Cross-section. Electron micrograph of the transition between the adventitia and media. The adventitia is almost entirely collagenous. The media is composed of elastin and smooth muscle. A layer of mesothelial cells (mc) can be seen on the outer edge of the adventitia. Stained with uranyl acetate and lead. 59 The col lagen fibers o f the adventit ia are not all oriented in a single plane (Figure 21). Th i s cross-section shows that wh i le the majority o f the large bundles are arranged long i tud ina l ly (If), there are also smaller bundles w h i c h run c i rcumferent ia l ly (cf). The c i rcumferent ia l f iber bundles are often relat ively short. However , the brevity o f the segments may be due to the sectioning process. The orientation o f two segments (*) f rom the upper right o f the micrograph suggests that they were part o f a larger bundle wh i ch had the connect ing region removed by the plane o f the section. Th i s impl ies a wavy mesh o f col lagen bundles. E las t in and smooth muscle make up the majority o f the media (F igure 22). Ve r y l itt le col lagen occurs in the media or along the luminal surface. The smooth muscle cells possess a large number o f p lasmalemmal vesicles (arrows) surrounding much o f the cel l . It has been suggested that these vesicles play a role in pinocytos is and the transfer o f materials out o f the cells (Karnovsky and Shea, 1970, Cas ley-Smith, 1971). Since smooth muscle functions in elastin production (Rucker and T inker, 1977), the p lasmalemmal vesicles may be p lay ing a role in the secretion o f elastin or elastin precursors. However , the actual ves icular funct ion is unknown. The elastin is not arranged into lamellar sheets as in arteries but remains in loose f ibr i ls. The f ibr i l s superf ic ia l ly resembled the g lycoprote in mic ro f ib r i l s associated w i th mammal ian elastin. However , using elastases, elastin stains and molecular analyses, Seraf in i - Fracass in i el al. (1978), Ben jamin et al. (1983) and I sokawa et a/.(1988), have demonstrated that the loose f ibr i l s are elastin. W h i l e the bulbar elastin is not found in concentr ic lamellae, the f ibr i l s do show an orientation suggesting an association w i th the smooth muscle cells. The elastin f ibr i l s in F igure 22 are c i rcumferent ia l ; their long axes 60 Figure 21. Cross-section. Electron micrograph of the collagenous adventitia. The collagen fibers are not arranged in a single direction. The majority of the fibers are arranged longitudinally (If) but there are also bundles of circumferential fibers (cf). Some bundles are truncated by the plane of the section (*). This suggest the bundles have a wavy conformation. Stained with uranyl acetate and lead. 61 Figure 22. Cross-section. Electron micrograph showing the media. Smooth muscle and elastin make up the majority of the media. The elastin is found as loose fibrils. The long axes of the elastin fibrils are aligned with the long axes of the smooth muscle cells, indicating a circumferential arrangement. The smooth muscle cells possess a number of ptasmafemmal vesicles along the edge of the membrane (arrows). Stained with uranyl acetate and lead. 62 f o l l o w the axes o f the smooth muscle. In F igure 23, the muscle cells are go ing into the plane o f the section, indicat ing a longitudinal direction. N o obvious ax ia l a l ignment can be seen for the elastin f ibr i l s , indicat ing that they are also go ing into the plane and have the same orientation as the smooth muscle cells. The " e m p t y " space surrounding the smooth muscle cel ls and elastin is l i ke ly a combinat ion o f g lycoproteins and proteoglycans (Seraf ini-Fracass ini et al., 1978) The smooth muscle cells in the ye l l owf in bulbus are not always sparsely distr ibuted with in the elastin. F igure 24 shows a large number o f smooth muscle cel ls in close prox imity. The cel ls in this muscle layer are not attached, but they do possess projections that may funct ional ly l ink the cells. Layers o f smooth muscle in the bulbus have been described as possessing a spiral orientation (Watson and Cobb, 1979; Yamauch i , 1980) however; it was di f f icu l t to establish an exact orientation o f the muscle layer w i th in the ye l l owf i n tuna. The smooth muscle cells once again contain a large number o f smal l vesicles a long the edges o f the membrane (arrows). The inner layer o f the ye l l owf in tuna bulbus is quite spongy. W h i l e the inner media is also composed o f elastin and smooth muscle, it is less ordered than the more superf ic ial med ia (Figure 25), s imilar to what occurs in Anguilla anguilla (Icardo et al., 2000). N o col lagen occurs along the luminal surface. Endothel ia l cel ls mod i f ied for a secretory role are common along both the luminal surface and the longitudinal elements (Figure 26). These endothelial cel ls contain the p lasmalemmal vesicles found along the margin o f the membrane (arrows) as we l l as much larger vesicles (arrowheads) that are found throughout the cells. A l l f ish species examined show bulbar endothelial cel ls possessing these large membrane-bound vesicles, and they have been described as l ip id 63 Figure 23. Cross-section Electron micrograph showing the media. The long axes of neither the smooth muscle nor the elastin fibrils can be seen. This suggests that they are coming out of the plane of the section and are arranged longitudinally. Stained with uranyl acetate and lead. 64 Figure 24. Cross-section. Electron micrograph showing a rich layer of smooth muscle cells in the bulbar media. The close proximity of the cells and the large number of projections between them suggest they may be functionally linked. However, no junctions can be seen. The smooth muscle cells have plasmalemmal vesicles under the edge of the membrane. Stained with uranyl acetate and lead. 65 Figure 25. Cross-section. Electron micrograph of the luminal surface of the bulbus. This layer of the bulbar wall is "looser" than the outer media. The primary components are elastin fibrils and smooth muscle cells. No collagen can be seen. Stained with uranyl acetate and lead. 66 granules (Santer and Cobb, 1972), moderately electron dense bodies (Leknes, 1980) and specif ic endothelial granules (Benjamin et al, 1983). However , the exact nature o f the material w i th in the vacuoles or their funct ion is sti l l unknown. They make up a large port ion o f the cel l vo lume in the endothelial cells examined (F igure 26), mi r ror ing what has been seen in other species (Benjamin et al, 1983). Co l lagen is l imi ted to the adventit ia except near the ventricle, where a large amount o f co l lagen can be found be low the layer o f endothelial cel ls (ec) (Figure 27). The moderately dense bodies that are characteristic o f the endothelial cel ls can also be seen (arrowheads). The majority o f the col lagen is arranged long i tudinal ly (If), however, other orientations are obvious. Prev ious examinations o f the innervat ion in bulb i suggested that large neurons were conf ined to the adventitia, wi th smal ler axon ic bundles penetrating a few microns into the adventit ia (Watson and Cobb, 1979; Ben jam in el al, 1983). However , this micrograph demonstrates another exception to normal bulbar structure. A neuron can be seen in the midst o f the col lagen several microns be low the endothelial layer. This col lagenous layer l i ke ly corresponds to the ventr iculo-bulbar valve. 4. Inflations In Vitro Bulbus Arteriosus Pressure-volume loops o f the bulbus arteriosus (Figure 28) had the same shaped curve irrespective o f the species o f f ish tested. These curves can best be described as " r -shaped", w i th the dist inguishing feature being a cont inuously decreasing slope. W i t h respect to the bulbar inflations, the r-shaped curve describes a sharp init ial rise in 67 Figure 26. Cross-section. Electron micrograph of a longitudinal element. The longitudinal elements contain endothelial cells modified for secretion. The endothelial cells contain plasmalemmal vesicles underneath the membrane (arrows) as well as larger, electron-dense vesicles scattered throughout the cell (arrowheads). Stained with uranyl acetate and lead. 68 Figure 27. Cross-section. Electron micrograph of the luminal surface of the bulbus. This section comes from near the ventricular end of the bulbus. Large, electron-dense vesicles can be seen in the endothelial cells (arrowheads). A large amount of collagen can be seen below the endothelial cells (ec). The orientation of the collagen fibers is primarily longitudinal (If). A neuron can be seen in this section. Stained with uranyl acetate and lead. 69 pressure f o l l owed by a plateau phase in wh i ch large changes in vo lume result in small changes in pressure. W h i l e the general shapes o f the pressure-volume curves were the same between w ide l y separated species, the curves were not identical but var ied w i th regards to the magnitude o f the pressure required to reach max imum volume. The bulb i o f the ye l l ow f i n tuna had very large increases in pressure (>60 m m H g ) for the init ial injections o f vo lume, wh i l e in other species, such as the mahimahi and deep-sea pomfret, the init ial rise in pressure was on the order o f 20 mm Hg. Res i l ience describes how we l l a material recovers its shape after an inf lat ion-def lat ion cycle. The difference between the areas under the inf lat ion and deflat ion curves is the amount o f energy lost as heat. When this loss is normal ized to the area under the inf lat ion curve, the resulting percentage is known as hysteresis. A h ighly resi l ient material has a l o w hysteresis. The hysteresis values are g iven for the curves in F igure 28. W h i l e in a l l cases, the hysteresis values are quite low, the different species shown in F igure 28 demonstrate a trend in wh i ch the f ish that have the highest operating pressures also have the largest hysteresis. In Situ Bulbus Arteriosus M o s t o f the bulb i were catheterized outside o f the animal or with in an open pericardial cavity. The results o f an expansion in wh i ch the bulbus o f a ye l l ow f i n tuna was catheterized without opening the pericardium are shown in F igure 29. The init ial 70 {/> "(0 <D i _ CD -«-> CO C D T -C O CNJ 0 3 T -O C D Csl" T -00 00 i ro = co ^ £ ra *• i s E n n o E <D a . ro - C F >* co . a co ro fc o . H I H H 00 d CD E <°. 3 o — O > o CM O (BH tutu) ajnssdjd to Q. to 3 o -c C L . O to CD CO ^ 3 ^ J CD CD ^ O ? 2 E co k 5 ° § 8 5 § 3 § C CD C XI 3 I— CD JZ ' CD C O l _ J O CD I-c o ro c CD (0 c n C O c to a> S (0 to to c c -Q 3 XI CD x: E o i _ to CL o o CD E _3 O > i CD L _ 3 to to CD 00 « N CD i _ 3 CO LL t Q . .9- ^ "to o 1 -to to II 5 *= -c o «| •E TJ I s c CO CO E ^ CD x: 3 co 5 E 8 £ ,co <q 5 c S5 § I t 4£ ro _ CD to tO (5 to o 3 CO CD x: to CO to _o O ) CD c CD c 3 O E CO CD x: 4— o CD i_ 3 to (0 CD E CO CO to to CD to >. x: CD O c CD CO CD CD CO 1 CD CD ^ 3 CO > to to CD i _ CD -«-* to 71 O Y E L L O W F I N T U N A B A volume (ml) Figure 29. Pressure-volume loops from bulbi of yellowfin tuna. One was inflated with the pericardium intact and the other was inflated with the pericardium removed. slope o f the in situ inf lat ion was higher than that o f the unrestrained bulbus and resulted in a higher plateau pressure at max imum volume. In Vitro Arteries The pressure-volume loops o f the arteries tested did not show the r-shape o f the bulbar P - V loops. Instead o f a constant decrease, the slopes o f the arterial P - V loops were often cont inuously r is ing. Th i s type o f " J - shaped " curve is more correctly used when descr ib ing the non-l inear mechanical properties (stress, modulus) o f b io log ica l materials. However , for descriptive purposes, the J-shape is used to descibe pressure-volume curves that exhibit an increasing slope. The ventral aorta o f the ye l l owf in tuna showed a J-shaped pressure-volume loop (Figure 30). F o r eighty percent o f the vo lume injected, the pressure increased by 80 mm Hg . It rose another 80 m m H g for the last twenty percent o f vo lume injected. The funct ional consequence o f this inf lat ion behaviour is that the artery becomes increasingly st i f f w i th increasing volume, thereby l im i t i ng expansion. The ye l l ow f i n aorta showed very l itt le hysteresis (10.2%), being a highly resil ient elastic element. L i k e the different sized bulb i , the size o f the vessel tested does not affect the pressure-volume characteristics when max ima l l y inflated. F igure 3 0 A shows a segment o f ventral aorta inflated w i th 100 p i and reaching a pressure o f 160 m m H g ; F i gure 30B shows a segment o f ventral aorta also reaching a pressure o f 160 mm Hg , however, it does so for a vo lume o f 25 p i . When vo lume is normal ized, the curves are very s imi lar (F igure 30C). 73 Figure 30. Pressure-volume loops for yellowfin tuna ventral aortas. A. A portion of ventral aorta inflated with 100 pi. B. A portion of ventral aorta inflated with 25 ul. C. A comparison of the two different sized aortas after they are normalized and plotted on the same axes. The dashed lines indicate the physiological pressure range. The ventral aorta o f the blue marl in d id not show a simple J-shaped pressure-vo lume loop (Figure 31 A ) . The slope o f the curve fe l l before r is ing, result ing in two inf lect ion points. I w i l l ca l l this an r-J curve. Despite this more complex behaviour, w i th in the expected phys io log ica l b lood pressure range, the marl in ventral aorta possessed inf lat ion characteristics that were comparable to those o f the ye l l ow f i n tuna. The mar l in ventral aorta became increasingly st i f f as it was inflated, as evidenced by the rise in slope at high volume. Therefore, in the functional range o f pressures, the ventral aorta o f the blue mar l in and ye l l owf i n tuna behaved s imi lar ly (F igure 3 IB ) . The mar l in ventral aorta also demonstrated a high resi l ience and l ow hysteresis ( 9 % ) . The dorsal aorta o f the ye l l owf i n tuna also has an r-J-shaped pressure-volume loop, as the slope o f the curve showed an init ia l drop (Figure 32A) . A f te r the early in f lect ion point, the vessel exhibited the continuous rise in slope over the b io log ica l pressure range that is o f functional importance for arteries. The dorsal aortas o f two ye l l owf i n tuna were parasitized by the larval cestode Dasrynchus talismani ( B r i l l , 1987). This resulted in the dorsal aortic lumen being occ luded at l ow pressures. The inf lat ion behaviour (Figure 32B) o f these vessels was different than that o f the structurally identical unparasit ized vessels or than that o f the bulbus arteriosus (Figure 32C). A t high pressures, the curve began increasing in slope, however, over much o f the in vivo pressure range, the curve was dist inctly r-shaped. The in it ia l sharp rise and subsequent fa l l in slope was reminiscent o f the bulbar inf lat ions and the f ina l sharp rise in slope was reminiscent o f the inf lat ion behaviour o f arteries. W h i l e the P - V loop f rom the parasitized dorsal aorta (Figure 32B ) showed a very large in it ia l rise in pressure fo l l owed by a compliant plateau; after this plateau, the slope 76 180 - i volume Figure 31. A. Pressure volume loop for a blue marlin ventral aorta. The dashed lines indicate the putative physiological pressure range. B. Comparison between the ventral aorta pressure-volume loops from yellowfin tuna and blue marlin. The volume axes are normalized. The dashed lines indicate the physiological pressure range. 77 Figure 32. A. Pressure-volume loop from the dorsal aorta of a yellowfin tuna. B. Pressure-volume loop from the dorsal aorta of a yellowfin tuna parasitized by the larval cestode, Dasrynchus talismani. C. Comparison between the pressure-volume loops from a parasitized dorsal aorta, an unparasitized dorsal aorta and a bulbus arteriosus. All tissues come from yellowfin tuna. The dashed lines indicate the physiological pressure range. 78 o f the parastized dorsal aorta P -V loop again rose sharply. A t max ima l vo lume, the pressures reached in both the parastized and unparasit ized vessels were s imilar, and the sharp rise in stiffness in the parasit ized curve occured at roughly peak systol ic pressure. Therefore, despite the different inf lat ion pattern o f the parasitized dorsal aortas, over the pressure range o f the tuna, these vessels cou ld sti l l serve their strain l im i t i ng functions. Smooth Muscle Inactivation and Collagen Denaturation W h e n smooth muscle and col lagen were removed or inactivated, the pressure-vo lume curves o f the bulbus changed (Figure 33). Contro l inf lat ions on fresh bu lb i f rom ye l l ow f i n tuna and ra inbow trout show typical r-shaped inf lat ion curves. The addit ion o f 1CT5 M nicardip ine (a C a 2 + channel b locker) changed the curve. A f te r n icardip ine was added to inactivate smooth muscle, the magnitude o f the plateau port ion o f the curve dropped f rom that o f the control level. Heat ing resulted in the bulbus becoming even more compl iant, as the curve generated post-heating was lower than either the smooth musc le -b locked bulbi curves or the control bulbi curves. The results were the same for both the ye l l ow f i n tuna bulbus (Figure 33 A ) and the ra inbow trout bulbus (Figure 33B). The repeated measures two-way A N O V A showed that the treatments had a s ignif icant effect on the P - V curves ( pO .OO l ) . Sectioned Bulbus Arteriosus The pressure-volume loops f rom the anterior and posterior sections o f the blue mar l in bulbus are s imi lar to the P - V loops f rom the who le bulbus (F igure 34). Three o f 80 0 0.5 1 1.5 2 2.5 3 3.5 volume (ml) volume (ul) Figure 33. Comparison of the pressure volume loops from fresh bulbi before and after undergoing two treatments: a tissue bath containing a 10"5 M solution of nicardipine to inactivate the smooth muscle and a high temperature tissue bath to denature the collagen. A. Bulbi from yellowfin tuna (n=4). B. Bulbi from rainbow trout (n=6). 81 Figure 34. Pressure volume loops of segments cut from a blue marlin bulbus. The different segments were 2.2 cm long from the anterior end of the bulbus (2.2cm ANT), 1.3 cm long from the anterior end of the bulbus (1.3cm ANT) and 2.0 cm long from the posterior end of the bulbus (2.0cm POS). 82 the segments (control, anterior 2.2 cm, posterior 2 cm) reached a f inal pressure w i th in the range 80-100 m m H g . The posterior 2 cm section had a steeper curve than the anterior 2.2 cm section. The anterior section, cut down to a size o f 1.3 cm, showed the steepest curve, w i th a peak pressure o f approximately 140 m m Hg . Inside-Out Bulbus Arteriosus W h e n the bulb i o f blue marl in were turned inside out and inflated, the results were pressure vo lume loops with a shape strongly reminiscent o f other bulbus inf lat ions (Figure 35 A ) . Wh i l e the results were not identical to normal marl in bulbus inf lat ions (Figure 35B), the resulting pressure-volume curve is undeniably bulbus- l ike: a sharp in i t ia l rise f o l l owed by a plateau phase Dissected Bulbus Arteriosus The purpose behind turning the blue marl in bulbus inside out was to a l low the removal o f bulbar layers. A f te r several layers f rom the bulbus had been removed and the bulbus was turned right side out, it wou ld no longer hold its shape and col lapsed when empty. Th i s meant that, in it ia l ly, f lu id could be injected into the empty bulbus w i th no concomitant rise in pressure. W h e n the vo lume injected had f i l led the space wi th in the bulbus, addit ional f l u id injections were resisted by the wa l l elements, resulting in a rise in pressure. Inflations began at the point where the bulbus was fu l l and at zero pressure. Since we did not count the f lu id used to f i l l the bulbus in i t ia l ly, the x-axis o f the P - V 83 Figure 35. A. Pressure volume loop from an inside-out blue marlin bulbus arteriosus. B. Pressure-volume loop from a normal blue marlin bulbus arteriosus. loops is, in fact, the change in vo lume o f the bulbus after zero pressure, A V , and not the total vo lume injected. The removal o f the longitudinal elements alone resulted in a small drop in the in i t ia l slope o f the pressure-volume curve but did not alter the plateau slope at high pressure (Figure 36). However , when the inner media was removed as we l l , the curve changed drastically, w i th the in it ia l slope fa l l ing twenty-fold. The discrepancy in in i t ia l slopes resulted in a large difference in the pressures reached at a g iven volume. A n inf lat ion o f 4 m l resulted in a pressure increase o f approximately 10 m m H g in the dissected bulbus, wh i l e the normal curve showed a pressure o f almost 80 m m H g . However , when the vo lume in the dissected bulbus was increased f rom 6 to 8 ml o f f lu id , the slope rose rapidly, result ing in a pressure o f 110 mm Hg . The r-shaped curve o f the bulbus became J-shaped, possessing an in i t ia l l ow slope fo l l owed by a steep rise. A t very large volumes, the typica l bulbar inf lat ion characteristics became artery-l ike. The same behavior occurred in all the bulb i tested in this fashion. A s the longitudinal elements were removed, a drop in the magnitude o f the curve occurred. A f te r the removal o f even more o f the bulbus wa l l , essentially leaving the outer layer, the bulbus curve became artery-l ike. In Situ Inflation of Bulbus Arteriosus using VDA A ye l l ow f i n tuna bulbus f rom a f ish used in the dynamic V D A studies was inf lated post mortem. This f ish had been a part o f a study look ing at the effects o f A N F on the cardiovascular system. In this bulbar inf lat ion (Figure 37), the slope d id not 85 120 i AV (ml) Figure 36. Pressure volume loops from the bulbi of blue marlin. The bulbi have layers of the wall dissected away. The different treatments are: normal (nothing done to the bulbus); w/o I.e. (the longitudinal elements are removed); w/o l.e./inner (the longitudinal elements and inner media are removed). 86 160 -i 140 j 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 diameter ratio Figure 37. A. Pressure-volume loop created in situ using the yellowfin tuna prep from the VDA setup. B. Pressure-diameter ratio created using the yellowfin tuna prep from the VDA setup. The VDA was used to measure the diameter changes. 87 decrease continuously. Instead, at high volumes, the slope began to increase, the result o f an increase in stiffness. Th i s bulbus curve turned up at a relatively l ow pressure. W h i l e other ye l l owf i n bu lb i showed no sign o f increasing stiffness at pressures we l l above 100 mm Hg , this particular spec imen ' s stiffness rose beginning at a pressure o f 80 m m H g . 5. Tensile Testing Yellowfin Tuna Bulbus and Ventral Aorta Rings Af te r per forming several c i rcumferent ia l force-extension tests on r ing samples, we decided to use a value o f thickness wh ich was smaller than the fu l l thickness o f the bulbus w a l l when calculat ing stress. In a number o f force-extension studies the inner media l layer o f the bulbus (with longitudinal elastin fibers) r ipped (F igure 38) w i th no change in the slope o f the force curve being generated, suggesting that this layer o f the bulbus was contr ibuting l itt le mechanical strength to the material. Inc lud ing the thickness o f the inner medial layer when calculat ing stress wou ld result in an underestimation. The inner media makes up approximately 6 0 % o f the wa l l thickness and, therefore, the value we used was approximately 4 0 % o f the fu l l thickness o f the bulbus wa l l . A difference in mechanical properties occurs between the adventit ia and outer media (F igure 39). L oop s o f the respective layers subjected to uniaxia l extensions had J -shaped stress-strain and modulus-strain curves; however, at strains over 1, the adventitial layer generated higher stresses (Figure 3 9A ) and was much stiffer (Figure 39B) than the 88 circumferential longitudinal hooks outer media inner media adventitial layer Figure 38. Schematic drawing showing the different layers of the bulbar wall and their respective thicknesses. Only the thickness of the adventitia and outer media were used to calculate wall stress. 89 A. B. •o o E 2000 n 1800 1600 1400 0. « 1200 3 1000 800 600 400 200 0 in vivo strain \ 0.5 1 strain adventitia outer media 1.5 Figure 39. Circumferential material properties of the adventitia (n=1) and outer media (n=3) of yellowfin tuna bulbi. Data were obtained from uniaxial tensile tests on tissue loops. A. Stress-strain curves B. Modulus-strain curves. The in vivo strain range was calculated using VDA data from yellowfin tuna. Results are ± S.E. 9 0 outer medial layer. A t a strain o f 1.2, the adventitial layer was about 20 times stiffer than the outer media l layer. Despite this large difference in stress and stiffness at strains above one, at strains be low one, the mechanical properties were very s imilar. A t these levels, the outer medial layer actually bears more o f the load in the wa l l than the adventit ia does, due to the differences in the wa l l thickness. The outer media was approximately 10 t imes th icker than the adventit ial layer (adventitial layer: 0.04 cm, outer medial : 0.4 cm). The outer med ia showed a much greater capacity for length changes than the adventitia. W h i l e the adventit ia broke at a strain o f 1.25, the outer media reached strains over 1.7. Th i s large difference in breaking strains is somewhat mis leading, because strain is normal ized to the init ial length o f the material being tested. Since the adventit ial layer had a larger in it ia l diameter than the outer media, the absolute length at wh i ch the adventit ial and medial layers broke was much closer than the breaking strains suggest. The stress-strain curves generated f rom loops o f tissue cut f rom the bu lb i o f y e l l ow f i n tuna (Figure 4 0 A ) and stretched c i rcumferent ia l ly d id not have the " r - shape" wh i ch was characteristic o f the bulbar pressure-volume loops. Instead, the stress-strain curves had the normal J-shaped curve o f many b io log ica l materials, ind icat ing nonl inear stiffness. The slope o f the curve remains l ow until a strain o f 1.2, at wh ich point the stress rapidly increases. The modulus-strain curve (Figure 40B ) shows that the stiffness o f the material rose exponential ly, w i th the major rise beginning at a strain o f 1.2. Init ial ly, the different segments (anterior, middle, posterior) had s imi lar modu l i , but at a strain o f 0.6, the curve representing the midd le begins to separate f rom the other two tissues. The midd le section, for a g iven strain, had a higher stiffness than both the posterior and anterior bulbar sections. The middle segment was not as extensible as the other segments. 91 600 strain Figure 40. Circumferential material properties of different segments of the bulbus (n=6) and ventral aorta (n=3) of yellowfin tuna. Data were obtained from uniaxial tensile tests on tissue loops. A. Stress-strain curves of the anterior, middle, and posterior portions of the bulbus. B. Modulus-strain curves of the anterior, middle and posterior portions of the bulbus. C. Stress-strain curves of the ventral aorta, anterior, middle and posterior portions of the bulbus. D. Modulus-strain curves of the ventral aorta, anterior, middle and posterior portions of the bulbus. The in vivo strain range was calculated using VDA data from yellowfin tun. Results are ± S.E. 92 CO CL in w OJ 4500 4000 3500 3000 2500 2000 1500 1000 500 ~f~^  m dfKciiur i i 3 - B — middle 1 1 P~ 1 nnQtdrinr j in vivo strain • J -x- ventral aorta 0.5 1 strain 1.5 D . co Q. CO 3 3 T3 O E 9000 8000 7000 6000 5000 4000 3000 2000 1000 —•— anterior — H — middle !/n vivo strain ! J - A - posterior j ] if ventral aorta i i * 0 0.5 i 1 i i 1.5 2 strain 93 The anterior and posterior segments both reached strains o f 1.6, wh i le the middle section reached a strain o f 1.4. The different portions o f the bulbus had an effect on the stiffness. The three-way A N O V A showed that the middle port ion was s igni f icant ly different (p<0.001) f rom the anterior and posterior portions o f the bulbus (Figure 40B) . There was also an interaction between the level o f the strain and the port ion o f the bulbus. A t strains be low 0.8, the curves were not s ignif icantly different. A t higher strains, the modulus o f the midd le port ion differed s ignif icant ly f rom that o f the other portions. The levels o f stress and modulus measured in the bulbus, despite us ing the smaller value o f thickness, were sti l l re lat ively l ow when compared w i th those calculated f rom a ventral aorta (Figure 40C). The ventral aorta reached a higher level o f stress (F igure 40C ) and was much stiffer than the bulbus (Figure 40D). The peak modulus o f the ventral aorta was seven times higher than the peak value o f any section o f the bulbus. The ventral aorta also reached its peak levels o f stress and modulus at a lower strain than the bulbus tissues did. The modulus o f the ventral aorta was s ignif icantly different than the modulus for a l l the portions o f the bulbus. The ventral aorta was stiffer than the bulbus over most o f the phys io log ica l range. The in vivo strain range was calculated f rom recordings o f the in vivo beating o f ye l l ow f i n tuna hearts. Fo r both the separated tissue rings (Figure 39) or the intact rings (Figure 40), the material properties o f the ye l l owf in tuna bulbus (elastic modulus and extensibi l i ty) were much greater than what was experienced dur ing normal beating. 94 Blue Marlin Bulbus and Ventral Aorta Rings L i k e the loops f rom the ye l l owf in tuna bulbus, the mar l in bulbus rings also reached large levels o f strain when stretched c i rcumferent ia l ly (F igure 41). The loops cut f r om the anterior o f the bulbus were the most extensible, reaching strains o f 1.5. The posterior loops were the least extensible, w i th a max ima l strain o f 1.2. The strain o f 1.3 reached by the midd le loops places them, appropriately enough, between the other segments w i th respect to extensibil ity. However , un l ike the tuna, there was no difference between the different segments w i th regard to the values o f stress or modulus (Figures 41 A , 41B) . A t a strain o f 1.2 (the largest strain at wh ich there was data for all three segments) there was no s ignif icant difference between the values o f stress or modulus. The mar l in ventral aorta was stiffer than the marl in bulbus (F igure 41C) , however, not to the extent seen in tuna (Figure 40C). The modulus values o f the ventral aorta were three t imes higher than those o f the marl in bulbus at a strain o f one (F igure 41D ) (as opposed to the sevenfold increase in stiffness seen in the tuna ventral aorta seen in F igure 40D). The mar l in aorta was over twice as st i f f as the mar l in bulbus over most o f the phys io log ica l range o f strains (assuming marl in and tuna have a s imi lar range o f strains). F igure 42 shows that, at a strain o f one, the ventral aorta o f the ye l l ow f i n tuna was approximately six times stiffer than the ventral aorta o f the blue marl in. B o t h the marl in ventral aorta and the ye l l ow f i n tuna ventral aorta had max ima l strains o f 1.1. The mechanical properties o f the ye l l owf i n tuna bulbus and the blue mar l in bulbus were much closer than the properties o f their respective arteries. F i gure 43 demonstrates that the modulus values for the entire marl in bulbus complete ly over lapped w i th the values o f the middle o f the ye l l owf in tuna bulbus and were not s ignif icant ly 95 A. 4 5 0 0 0.5 1 1.5 2 strain Figure 41. Circumferential material properties of different segments of the bulbus (n=4) and ventral aorta (n=3) of blue marlin. Data were obtained from uniaxial tensile tests on tissue loops.A. Stress-strain curves of the anterior, middle, and posterior portions of the bulbus. B. Modulus-strain curves of the anterior, middle and posterior portions of the bulbus. C. Stress-strain curves of the ventral aorta, anterior, middle and posterior portions of the bulbus. D. Modulus-strain curves of the ventral aorta, anterior, middle and posterior portions of the bulbus. The in vivo strain range was calculated using VDA data from yellowfin tuna. Results are ± S.E. 1400 -i C . 1200 H strain 9 0 0 0 -8 0 0 0 strain Figure 42. Circumferential material properties of the ventral aortas from yellowfin tuna (n=4) and blue marlin (n=3). The modulus and strain were calculated from uniaxial extensions of artery loops. Results are ± S.E. 98 CM l O c s •+-> (0 LO o o o o CM O O 00 o o C D O O O O CM O O O O O 00 o o CO o o o o « s f CM o CO C g CO c CD X CD H X CO 'c D E o T3 O CD Xi CD x: E o L _ >•— CO c CD E CO CD to 2 to o CL •D C CO 0) T3 CO D O CO o CD c (0 E £ CD -«—< C CO CD O CO CD '•c CD CL O 1 _ CL H CD (0 E To c CD E O b CO CD CO TD C CO CO _D Zi TJ O E CD II c CO E LU <D C/J CD C CO CO tn co 3 II co CO C CO =S CL ~ o c o 2 ^ (Bd>i) sninpouj 3 O) CD U 99 different f r om the anterior and posterior portions o f the bulbus be low a strain o f 1.2. A f te r the strain o f 1, the anterior and posterior section o f the ye l l ow f i n tuna had lower modulus values than those o f the marl in. The anterior and posterior sections o f the tuna bulbus were more compl iant than any area o f the marl in at high strains. A s we did not have any recordings o f a blue mar l in heart beating, we were unable to get an estimate for the in vivo strains experienced by a mar l in bulbus. Yel lowfin and Bigeye Tuna Bu lba r Segments The isolated bulbus segments f rom ye l l owf in and bigeye tuna were roughly rectangular and, instead o f being lengthened on the customized " stretching mach ine " , they were attached to a micromanipulator and extended. The stress-strain data for these extensions are shown in F igure 44A. The same data are plotted as modulus-strain data in F i gure 44B. The obvious th ing f rom both plots is that the tissues cou ld be broken into two general groups: 1, st iff and 2, extensible. A t a strain o f one, the modulus for most o f the tissues fe l l into a range that lay between 50 and 100 kPa. E ven the values gathered using the previous method o f stretching tissue rings (middle layer o f y e l l ow f i n tuna bulbus, stretched c ircumferent ia l ly, Y F , M L c i r c * ) fitted into this range. Di f ferent bulbar wa l l segments were stretched either c i rcumferent ia l ly or longitudinal ly over a large strain and the modulus remained low. Despite the var iabi l i ty inherent in dissecting pieces o f tissue away f r om the wa l l , the different tissues had s imi lar values o f stress and modulus over a wide range o f strains. The exceptions were the two sections o f bulbar outer layer f r om bigeye tuna, stretched longitudinal ly. Compared w i th the other tissues, both pieces 100 Figure 44. Circumferential and longitudinal material properties of segments from the bulbi of yellowfin and bigeye tuna. The data were obtained by stretching rectangular pieces of the bulbar wall using a micromanipulator. YF, OL, circ.= yellowfin tuna, a segment of the outer layer, stretched circumferentially. YF, ML, long = yellowfin tuna, a segment of the middle layer, stretched longitudinally. YF, LE, long = yellowfin tuna, a sement of longitudinal element, stretched longitudinally BE, LE, long = bigeye tuna, a segment of longitudinal element, stretched longitudinally. BE, OL, circ = bigeye tuna, a segment of outer layer, stretched circumferentially. BE, OL, long = bigeye tuna, a segment of outer layer, stretched longitudinally. BE, ML, long = bigeye tuna, a segment of middle layer, stretched longitudinally. YF, ML, circ*. Yellowfin tuna, a segment of middle layer, stretched circumferentially. This piece was stretched as a loop using the Instron stretching machine. A. Stress-strain curves. B. Modulus-strain curves. 101 A. 140 120 100 80 i: 60 40 20 P / B -•— YF, OL, circ f / / / - a - YF , ML, long. _jT / * YF, ILE, luiiy. T a / // ~*~BE'ILE'long-/ / + x /r ~*~ BE> 0l' circ-/ /1p — I — BE2, OL, circ. / j a O ^ l / ^ ^ * ^ B E ' ° L ' l o n g -Jngg^Z/^ m BE2. OL. lonq. J g ^ ^ ^ ^ ^ V ^ ^ -e— BE, ML, long •YF, ML, circ.' 0.5 1.5 strain 800 700 •5- 600 500 400 300 200 100 / - • - YF, OL, circ. / - » - YF, ML, long. / YF, ILE, long. I - x - BE, ILE, long. / - • - BE, OL, circ. I • nr~^  c\\ f\rf -f I Otz£, VJL, circ. / cj x B E . 0 L . l o n 9-/ jf~r^^~*~-^r^ B E Z O L l o n a / n T ^ f ^ J ^ T ^ BE, ML, long, ^ ^ ^ ^ t * ? ^ ^ * ^ - H - YF, ML, circ* i i i i 0.5 1.5 strain 102 become very st i f f at a l o w strain, reaching a modulus over 400 kPa at a strain o f 0.5. The two general classes o f tissues were also very different in their abi l it ies to change in length. The outer layer, when stretched longitudinal ly, was l imited to a strain o f 0.5, the other tissues al l reached strains we l l over 1. The strains reached by the long i tudinal ly stretched outer layers f rom bigeye tuna bulb i matched we l l w i th the strains reached in vivo in y e l l ow f i n tuna. The max imum, in vivo longitudinal strain reached by any ye l l ow f i n tuna bulbi was 0.48. Th i s matching between in vivo and in vitro data suggests that it is va l i d to make conclus ions on ye l l owf in tuna bu lb i based on data f rom bigeye tuna bulbi.. 6. In Situ and In Vivo Recordings: yellowfin tuna The v ideo dimensional analyser ( V D A ) f o l l owed the wal l s o f the y e l l o w f i n tuna bulbus dur ing both systole and diastole and a l lowed me to map the d imens ional changes associated w i th each heartbeat. The single heartbeat in F igure 45 A shows the difference between the d imensional and pressure changes. The rapid increase in pressure resulted in a sharp increase in diameter. However , wh i le the systol ic pressure o f 70 mm H g gradual ly dec l ined to 30 mm H g , the decrease in diameter occurred differently. There was a sharp init ial fa l l in diameter wh i le the pressure was high (>50 mm Hg) , f o l l owed by a smoother decl ine that more closely mirrored the changes in pressure. B y plott ing pressure against diameter for the heartbeat in F igure 45A, a pressure-diameter (P-D) loop was generated (Figure 45B) , showing the inf lat ion behaviour o f the bulbus under in vivo conditions. When this dynamic pressure-diameter loop was 103 80 70 _ 60 3 x | 50 I 4 0 a 30 20 10 0 p r e s s u r e 1.2 1.18 1.16 1.14 a 1.12 1 0 " 1 . 1 ^ 1.08 3. 1.06 1.04 1.02 1 0.98 1.05 1.1 1.15 diameter ratio 1.2 1.25 1.3 Figure 45. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Comparison of static and dynamic pressure-diameter loops. The dynamic trace was created by plotting pressure against diameter for the heartbeat in (A). The dynamic trace is superimposed on a pressure-diameter curve created by a static inflation. 104 compared w i th a pressure-diameter loop produced using the static inf lat ion technique, the dynamic and static behaviours matched we l l . In both cases, the slope in i t ia l ly rose sharply, f o l l owed by a leve l ing o f f as the bulbus reached the plateau phase o f the inf lat ion. There was a s ignif icant hysteresis in the dynamic loop, indicat ive o f a v i scous element in the bulbar wa l l . F o r the plateau, the inf lat ion curve was higher than the def lat ion curve, ind icat ing a loss o f energy. However , during the in it ia l rise o f the r-curve, the inf lat ion and deflat ion curves crossed over, indicat ing energy added. The bulbus is non-contracti le; therefore, the posit ive work loop was most l i ke ly an artifact o f the changing length o f the bulbus. The change in the length o f the bulbus could be clearly seen at the end o f the deflation. W h i l e pressure continued to drop, the diameter o f the bulbus increased, indicat ing wider, upstream segments entering the f ie ld o f v i ew o f the v ideo camera. The dynamic P-D loop o f F igure 45B is for a typical beat and demonstrates the bulbar behaviour over the pressure range o f 30-70 mm Hg. However , by l ook ing at beats cover ing the pressure range 20-160 mm Hg , the entire inf lat ion curve o f the bulbus can be recreated: the in it ia l steep rise, the plateau and the f inal steep rise at large inf lat ions and pressures. A t the l ow end o f the pressure range, the bulbus is operating on the steep part o f the inf lat ion curve and small changes in vo lume result in large, rapid changes in pressure. In F igure 46A , the small, low-pressure, heart beat wh ich is boxed generates a very steep P-D loop (Figure 46B). A s the f o l l ow ing beat begins reaching a normal pressure (70 m m Hg) , the bulbus seems to reach the distensible plateau region o f the inf lat ion curve. 105 Over a normal pressure range o f 30-80 m m H g (Figure 47), the P-D loop resembles that seen in F igure 45. The curve is very steep until approximately 50 mm H g , after wh i ch it begins to flatten out as the bulbus enters the plateau region. F o r an even higher pressure range o f 65-90 m m H g , the bulbus inf lat ion appeared to be shifted almost entirely on the plateau (Figure 48). The difference between the inf lat ion and def lat ion curves decreased as the pressure increased along w i t h the heart rate. F o r a very rapid heart beat over the pressure range 85-95 mm H g (F igure 49), the P-D loop shows the bulbus inf lat ing entirely on the plateau, as the loop is hor izontal, w i th very l itt le vert ical component. There is also very l itt le difference between the inf lat ion and def lat ion curves. The final stage o f the static P - V loops was a steep rise f o l l ow i ng the plateau. In static inflations, this only occurred in bulbi that were extraordinari ly stretched by large vo lumes due to high pressures, hormonal manipulat ion or dissections. In F i gure 50, the heart f r om a fish was beating w i th a systol ic pressure o f 170 mm H g and a diastol ic pressure o f 115 mm Hg . The P-D loop f rom this heart beat (Figure 50B) shows that wh i le the bulbus operate on the plateau for much o f the beat, above 140 m m Hg , the bulbus stiffness increases, creating a J-shaped curve after the plateau. The previous Figures have shown that P-D loops generated f rom ind iv idua l heartbeats have the same sections as the P -V loops derived f rom static inf lat ions (a sharp in i t ia l rise in pressure, a plateau phase and a final sharp rise in stiffness under condit ions o f h igh inf lat ion). B y examin ing several traces, one can see the special properties that the different sections g ive the bulbus. 106 20 10 -0 1 1 1 1 1 1 1 0.92 0.94 0.96 0.98 1 1.02 1.04 1.06 diameter (cm) Figure 46. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Dynamic pressure-diameter loop generated by plotting pressure against diameter for the highlighted heartbeat in (A). Arrows pointing right indicate the inflation; arrows pointing left indicate the deflation. 107 90 0.00 90 80 -I 70 60 50 40 30 20 10 f 0.94 5 B 0.98 0.96 5 0.92 ~ 0.9 0.88 0.86 0.20 0.40 0.60 time (s) 0.80 1.00 0.86 0.88 0.9 0.92 0.94 diameter (cm) 0.96 0.98 Figure 47. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Dynamic pressure-diameter loop generated by plotting pressure against diameter for the heartbeat in (A). Arrows pointing right indicate the inflation; arrows pointing left indicate the deflation. 108 Figure 48. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Dynamic pressure-diameter loop generated by plotting pressure against diameter for the heartbeat in (A). Arrows pointing right indicate the inflation; arrows pointing left indicate the deflation. 109 Figure 49. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Dynamic pressure-diameter loop generated by plotting pressure against diameter for the heartbeat in (A). Arrows pointing right indicate the inflation; arrows pointing left indicate the deflation. 110 Figure 50. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. Dynamic pressure-diameter loop generated by plotting pressure against diameter for the heartbeat in (A). Arrows pointing right indicate the inflation; arrows pointing left indicate the deflation. I l l In F igure 51 A , the largest pressure pulse was generated in i t ia l ly (*), w i th subsequent pulses becoming smaller, wh i l e peak pressure increased. Conversely, the smallest increase in diameter occurred during the largest change in pressure (*) and successive beats resulted in larger diameter changes. The first fluid injection after the bradycardia had a larger impact on the pressure than those that fo l lowed. The beating pattern shown in F igure 5 1 A is equivalent to the sharp in it ia l rise o f the static inf lat ion tests. Immediately after saffan injection, the hearts o f the tuna wou ld often deviate f rom the normal beating patterns, speeding up, increasing pressure or both. In a fish injected w i th saffan (F igure 5 IB ) , the lowest diastol ic pressure was 50 m m H g (after a long interbeat interval), wh i l e the highest pressure was 155 mm Hg . The beating pattern was irregular, however, it did demonstrate several functions o f the bulbus. Du r i n g the diastol ic decl ine, there was an occasional smal l pressure " b l i p " (arrows). These smal l increases in pressure (1-5 mm Hg) , were mirrored by large changes in the diameter o f the bulbus. Often these diameter changes were nearly as large as those associated w i th pressure changes that were twenty times or more larger. In that high-pressure range (100-120 m m Hg) , the bulbus was very sensitive to pressure changes, as it was dur ing the plateau stage o f the static inflations. However , at pressures greater than 130 mm H g , systol ic diameter did not vary a great deal, regardless o f the size o f the pressure pulse. In F igure 51C, the highl ighted beats (arrowheads) had peak systol ic pressures o f 130, 171 and 167 mm Hg , however, the systol ic diameter o f the bulbus varied only s l ightly (1.077-1.095 cm). The pressure sensit ivity o f the bulbus has disappeared. A t l ow pressure, it was the bulbus diameters 112 Figure 51. Proximal blood pressure and bulbus diameter from yellowfin tuna. A. Following a bradycardia, the largest pressure pulse is generated by the smallest stroke volume (*). B. At mean systolic pressure, the bulbus is very sensitive to small changes in pressure (arrows). 113 200 1.15 10 15 time (s) 20 25 Figure 51. Proximal blood pressure and bulbus diameter from yellowfin tuna. C. At very high pressure, the bulbus diameter becomes insensitive to large fluctuation in pressure (arrowheads). D. At very high pressure, despite the variability in blood pressure, the diameter changes in the bulbus are very similar. The bulbus is insensitive at high pressure. 114 that showed large var iab i l i ty f r om beat to beat, however, at very h igh pressure, the fluctuations in size are d imin i shed (Figure 5 ID). Despite erratic beating that resulted in peak systol ic pressure fluctuations o f over 50 mm Hg , the beat-to-beat diameter showed very l itt le variation. Peak systol ic diameter remained w i th in the range 0.92-0.96 cm wh i l e the systol ic-diastol ic diameter changes were 0.18-0.23 cm. Relationship between Diameter and Volume The dynamic P -D loops (Figures 45-50) have the same features as the static P - V loops and, in descr ib ing them, 1 have made the assumption that bulbar diameter is an accurate indicat ion o f bulbar volume. The val id i ty o f this assumption can be checked by close examinat ion o f static inflations. Static r-shaped pressure-loops were generated using the in situ V D A preparations f o l l ow ing the dynamic experiments. B y s imultaneously measuring the diameter changes due to each injection o f f lu id , it was also possible to create a plot o f pressure vs. diameter (Figure 52B) wh i ch was remarkably s imi lar to the r-shaped pressure-volume loops discussed above. W h e n diameter and vo lume are plotted against each other, it is obvious that the relationship between diameter and vo lume is linear in both the anterior (Figure 52C) and posterior (Figure 52D) portions o f the bulbus. B y performing a linear regression on the curves in F igures 52C and 52D, one can f ind a cal ibrat ion curve descr ibing the interactions between injected vo lume and diameter. V i sua l l y , the regression fits the curves extremely we l l , an assessment conf i rmed by the R 2 values. 115 0 0.2 0.4 0.6 0.8 1 volume (ml) B 90 • 80 -70 a> 60 i E g 50 £ 3 40 W CO £ CL 30 20 10 -0 • 0 0.05 0.1 0.15 0.2 0.25 diameter strain Figure 52. A. Pressure-volume loops from static, in situ inflations of bulbi from yellowfin tuna. Anterior and posterior refer to where on the bulbus the VDA window was centered. B. Pressure-diameter strain loop for the same bulbi as in (A). The diameter strain was calculated using diameter data from the VDA. C. Diameter plotted against volume for the bulbus measured at the anterior end. A linear regression was run on this plot and the solution is shown. D. Diameter plotted against volume for the bulbus measured at the posterior end. A linear regression was run on this plot and the solution is shown. 116 0.52 i c . volume (ml) 117 However , under certain circumstances, the relationship was not as precise. Occas ional ly , the bulbus was inflated to the point where its stiffness began to rapidly increase, f o rm ing a "J-shaped" curve (Figures 37, 53). When this happens, the relationship between diameter and vo lume changes (Figure 53C). Once the bulbus reaches the " J " part o f the slope (for F igure 53C, this happens at 1 ml injected), a difference between the inf lat ion and deflat ion curves results. A single l inear regression can no longer describe the inf lat ion-def lat ion cycle. Th i s change is l i ke ly due to differences in the longitudinal and c ircumferent ia l strains. Regardless, under normal beating, the regression equations generated f rom static inf lat ions w i l l a l l ow one to get a good estimate o f the vo lume w i th in the bulbus dur ing normal beating. D iameter and vo lume have a l inear relationship. F o r Figures 45-50, the pressure-diameter curves are good approximations o f the dynamic pressure-volume relationships w i th in the bulbus. Unfortunately, this fortuitous linear relationship raises an unsettl ing question: how can it poss ibly be? In geometry, the relationship between vo lume and diameter is not linear. I f the bulbus is model led as a cone w i th vo lume " V " then its vo lume is g iven as V = l/37ir 2l (4) where r = radius, 1 = length and vo lume is proportional to radius 2. S ince radius is I /2 proport ional to diameter, this demonstrates that diameter is proport ional to vo lume Therefore, i f the bulbus were truly a cone, the plots o f diameter vs. vo lume in F igures 5 2 C - D wou ld not be linear. However , the bulbus is not a s imple cone. A s the bulbus inflated, it also lengthened (F igure 54) and, therefore, moved in the v ideo field o f v iew. Since the V D A 118 180 i volume (ml) Figure 53. A. Pressure-volume loop from a static, in situ inflation of a bulbus from a yellowfin tuna. The VDA was centered on to the posterior portion of this bulbus. B. Pressure-diameter ratio loop for the same bulbus as in (A). The diameter ratio was calculated using diameter data from the VDA. C. Diameter plotted against volume for the bulbus. A linear regression was run on this plot and the solution is shown. 119 diameter ratio Figure 54. These stills from the VDA analysis show the differences in bulbus size between diastole and systole. Length, as well as width, changes with each beat of the heart. 121 windows were f ixed, the point that was fo l l owed at the start o f inf lat ion was not the point being f o l l owed at the end. The consequence o f this was a consistent underestimation o f the actual diameter changes as narrower, more distal portions o f the bulbus were being " pu shed " into the f ie ld o f v i e w (Figure 55). Because the length o f the bulbus was increasing (preventing the v ideo camera f rom f o l l ow i ng a s ingle point o f the bulbus), a more sophisticated model o f pressure-vo lume relationships w i th in the bulbus was required. Therefore, I ut i l ized the method outl ined in F igure 56 to model the behaviour o f the bulbus, in the hopes o f exp la in ing the observed l inear relationship between vo lume and diameter. U s i ng equation (4) and arbitrary values o f height (h) and volume, an in it ia l value o f radius (r) is obtained. Increasing the vo lume and the length (H) results in a new value for the radius at the bottom o f the cone (R). Us ing tr igonometry, the radius o f the new (r1) cone at the length o f the or ig inal cone (h) can be calculated. D i v i d i n g the new radius by the in it ia l radius results in a radius ratio. F i gure 5 7 A shows a plot o f diameter vs. vo lume for a hypothetical cone for wh ich the increase in vo lume was matched w i th an equivalent increase in length (i.e Av/Al = 10). Th i s model shows a linear relationship at low volumes, but at higher vo lumes the plot is dist inctly non-linear. When the ratio between longitudinal strain and diameter strain was plotted against vo lume (Figure 57B), the ratio increased. The model was modi f ied so that the length changes decreased in magnitude as the vo lume increased. In essence, we gave the cone a J-shaped longitudinal stiffness. Th i s model resulted in a l inear relationship between diameter and vo lume (Figure 57C), 122 m r-E 1 CO <-E M o •g 2 CD E CTJ O > 0) o CO CTJ • Mm " D CD' x : o ) CD > > 0) o 00 CO o c CTJ 0) "co o Q . C D x: c o • CO o Q. 3 CD XJ H i • -I—» u> o II LL Q. 123 124 s imi lar to the results o f the in situ bulbar inflations (Figure 54). When the ratio between longitudinal strain and diameter strain was plotted against vo lume (F igure 57D), the ratio decreased. The plot in F igure 57C resembles the diameter-volume curve actually found in the bulbus, and the model used to generate the curve resembles the bulbus in several important ways. The cone was model led to resist longitudinal expansion more strongly as the vo lume increased, a feature o f the bulbus (Figure 40). F igure 44 demonstrated that the bulbus is stiffer longitudinal ly than c ircumferent ia l ly, a feature o f the model in F igure 57D. I f the same axes are used to plot the inf lat ion o f a bulbus, the same trend occurs (F igure 58). A s the vo lume in the bulbus increased, the ratio o f length strain to the diameter strain decreased. The decreasing slope was found in both the bulbus and the model. The unexpected relationship between diameter and vo lume in the bulbus is a coinc idence result ing f rom the interactions o f the l inear and radial expansions. The f luke result ing in the l inear relationship between bulbar diameter and vo lume a l lowed a more detailed analysis o f the in vivo funct ioning o f the bulbus. Qual i tat ively, the l inear relationship between diameter and vo lume a l lows the inference that a change in diameter is due to an equivalent change in volume: i f one heart beat results in a bulbus diameter change twice as large as another, then twice as much f lu id entered the bulbus dur ing that beat. Quantitatively, the fact that a linear regression c losely describes the interaction o f diameter and vo lume (Figure 54) a l lows an analysis o f the vo lume into and out o f the bulbus w i th each beat. D u e to the differences in d imens ion a long the bulbus, the dynamic measurements and static measurements (which generate the cal ibrat ion equation) must be taken f rom the 125 Figure 57. A. Using the formulas in Figure 50, this data was calculated for a hypothetical cone. The length and diameter of this cone increased by an equivalent amount as volume increased (ie. Av/AI = 10). B. The ratio of length strain to diameter strain plotted against volume for the hypothetical cone in (A). C. Using the formulas in Figure 50, this data was calculated for a hypothetical cone. For this model, the increases in length decreased as the internal volume grew. (ie. as volume increased, the circumferential expansion became relatively larger than the longitudinal expansion). D. The ratio of length strain to diameter strain plotted against volume for the hypothetical cone in (C). 126 127 0 500 1000 1500 2000 2500 3000 3500 4000 4500 A volume 0 J 1 1 1 1 1 1 1 —i i 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 volume (ml) Figure 58. The ratio of length strain to diameter strain plotted against volume for actual bulbus inflations. 129 same locations along the bulbus. F o r this reason, cal ibrat ion curves could not be calculated for all recordings o f pressure and diameter. F igures 59 and 60 are examples o f typ ica l beating patterns for the anaesthetized y e l l o w f i n tunas we worked with. F igure 5 9 A shows diameter and pressure measured at a point near the middle o f the bulbus. A t a heart rate o f 1 H z , pulse pressure was approximately 40 m m H g w i th a peak systol ic pressure o f 70 mm Hg . D iameter changes were about 0.1 cm, go ing f rom 0.7 cm to 0.8 cm. F igure 53B demonstrates how those diameter changes translate into the vo lume wi th in the bulbus. The 0.1 cm diameter change resulted in bulbar vo lume vary ing f rom 0.3 ml to 0.8 ml. F igure 60 is a trace f rom a different ye l l owf in tuna. The heart was beating near 1 H z , wh i le pulse pressure was approximately 45 mm H g and the diameter changes were approximately 0.2 cm (Figure 60A) . The vo lume changes were s imi lar to those in F igure 59, go ing f rom 0.2 m l to 0.8 ml. W h e n the b lood pressure and heart rate fa l l , however, bulbus vo lume also falls. In F i gure 61, the heart was beating normal ly during the first 12 seconds, after wh i ch it began to s low f rom a rate o f 1.2 H z to 0.8 Hz . Bu lbus diameter and internal vo lume began to fa l l , and, between 20 and 25 seconds, the heart appeared to miss several beats, result ing in long diastol ic periods. Du r i ng those periods, bulbar vo lume fe l l near zero. However , even at those l ow internal volumes, the pressure remained at 20 mm Hg . The fact that a very smal l internal vo lume still resulted in a relatively large pressure suggests that the bulbus was operating on the in it ia l , steep section o f the r-shaped inf lat ion curve at this point. 130 a 53' 3 CD sr 3 Figure 59. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. The volume changes within the bulbus during the beating in (A). 131 A. U) x E E 3 (0 o I— a B. 0.9 0.8 0.7 0.6 g 0.5 a E 0.4 3 O > 0.3 0.2 0.1 0 W 3 4 5 6 time (s) 10 Figure 60. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. The volume changes within the bulbus during the beating in (A). 132 120 A. 100 o x £ E ¥ 3 (0 80 60 40 20 B. 1.2 -1 1 -0.8 1 lum 0.6 o > 0.4 0.2 -o -— pressure — diameter 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 Q. 3 CD O *» 3 Figure 61. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna. B. The volume changes within the bulbus during the beating in (A). 133 F igure 6 2 A shows a V D A recording o f the ventr icular end o f the bulbus f rom a f i sh shortly after an inter-arterial inject ion o f saffan. The heart was beating extremely fast (4 H z ) and pulse pressure is approximately 15 m m H g (80-95 m m Hg) . Despite the small pulse pressure, the changes in diameter (0.08 cm) were nearly as large as in the previous figures. Th i s suggests that the bulbus was behaving as i f it were on the plateau phase o f the r-shaped curve, where small changes in pressure generate large changes in volume. Indeed, the vo lume changes (0.6 ml - 1 ml) seen in F igure 62B were nearly the same size as those i i i F igures 59, 60, lending more evidence to the idea that the bulbus was funct ion ing on the plateau o f its inf lat ion curve. Under these condit ions o f heart rate and pressure, the bulbus maintained a large internal volume, even dur ing diastole. W h i l e us ing diameter changes to estimate vo lume changes wi th in the bulbus a l lowed a quantitative analysis o f the beating heart with in the fish, in some cases, circumstances occurred wh ich precluded anything but an educated guess at what was occurr ing w i th the bulbus. The lengthening o f the bulbus dur ing inf lat ion generated the l inear relationship between diameter and volume. However , wh i le the bulbus lengthened as it f i l l ed , the f l ip side o f this occurrence was that the bulbus shortened as it emptied. The shortening o f the bulbus " p u l l s " a wider, posterior portion o f the bulbus into the field o f v iew. T w o examples o f this phenomenon can be seen in F igure 63, as measured near the ventral aorta (63 A ) and ventricle (63B) o f the same fish. Du r i n g these shortening events, i f the l inear regression equations were applied, the bulbus wou l d appear to fill w i th b lood because it increased in diameter as pressure fel l . W h i l e the shortening events could be predicted by the occurrence o f a long diastol ic period, occas ional ly f luctuations in bulbus diameter occurred for no apparent 134 A. 120 100 x° 80 E E m £ a 60 40 20 0 1 2 3 4 5 6 time (s) B. 1.2 0.8 o | 0.6 o > — pressure diameter 1.06 1.04 1.02 a 1 | I 0.98 Z o 3. 0.96 0.94 0.92 0.9 10 0.4 0.2 1 1 1 1 1 1 1 1 1 1 0 1 2 3 4 5 6 7 8 9 10 time (s) Figure 62. A. Recordings of proximal blood pressure and bulbus diameter from a yellowfin tuna recently injected with saffan. B. The volume changes within the bulbus during the beating in (A). 135 100 0.85 Figure 63. A-B Proximal blood pressure and bulbus diameter from yellowfin tuna. 136 reason. In F igure 64A, the peak systolic pressure remained near 70 mm H g wh i l e the size o f the bulbus constantly decreased. The entire trace containing F igure 6 4 A can be seen in F i gure 64B. Over the first ten seconds, despite the very consistent pressure trace, diastol ic diameter o f the bulbus cont inual ly decreased, f rom 0.66 to 0.55 cm. The removal o f water f l ow over the tuna 's g i l l s (black stars) resulted in a decrease in heart rate and peak systol ic pressure rose dur ing this per iod by about 20 mm H g . A s expected, there was a concomitant increase in the absolute diameter o f the bulbus during the bradycardia. F o l l o w i n g the resumption o f ram vent i lat ion (white stars), decreases in the bulbus diameter occurred, fa l l i ng f rom an in i t ia l d iastol ic size o f 0.69 cm to a l ow o f 0.58 cm. W h i l e the changes dur ing the bradycardia can be explained, much o f the size change in the bulbus occurred before any manipulat ion o f heart rate. Throughout the trace, changes in relative bulbus diameter (diastole to systole) remained near 0.1 cm. The recurr ing changes in bulbus diameter occurred a number o f t imes in several f ish. F i gure 65 is an example o f regular beating over 80 seconds showing how the bulbus diameter undergoes large rhythmic f luctuations wh i le the pressure remains very constant. Note: the s low-down in heart rate at the 45-second mark occurred spontaneously. 137 0.85 0.65 0.75 » a 3' 3 o O 3 a S" 3 s 3 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 time (s) Figure 64. A. Proximal blood pressure and bulbus diameter from yellowfin tuna. B. A longer trace containing the recordings seen in (A.). The black stars indicate the time when ram ventilation of the tuna was ceased and the white stars indicate when ventilation was renewed. 1 3 8 A. I £ E © co co as 0.95 10 20 30 40 50 60 time (s) 70 80 10 20 30 40 time (s) 50 60 70 80 Figure 65. A. Recording of proximal blood pressure in a yellowfin tuna. B. Recording of bulbus diameter over the same time period as (A). 139 DISCUSSION 1. The Problem The bulbus inflation curve The bulbus arteriosus has a unique pressure-volume curve (F igure 66); much different than the J-shaped inf lat ion that is the norm for arteries. The normal P - V curve for a bulbus is r-shaped, w i th a steep init ial rise in pressure for a small change in vo lume (1) , a plateau dur ing wh i ch smal l changes in pressure result in large changes in vo lume (2) , and, in a few instances, a f inal sharp rise in pressure f o l l ow ing the plateau (3). F i nd i n g out the causes o f and reasons behind this curve was the goal o f this research. 1 hoped to understand the nature o f the curve, what structures cause the bulbus to inflate in this manner, and how it is used funct ional ly w i th in the animal. T o that end, 1 examined the morpho logy o f the bulbus through light and electron microscopy, d iscovered the material properties through inf lat ions o f bu lb i and tensile tests on isolated tissue sample and measured in vivo d imensional changes. Studies o f arteries have revealed that since stress is roughly analogous to pressure and strain is related to the internal volume, the plots o f stress and strain have the same shape as plots created using pressure and volume. The stra in- l imit ing design o f arteries results in J-shaped pressure-volume and stress-strain curves (Roach and Bur ton , 1957) and the bulbus also possesses J-shaped material properties (Figures 39-41), however, the bulbus arteriosus pressure-volume curve (Figure 28) is r-shaped. The paradox o f the strange bulbus P -V loop is that wh i le the bulbus appears to have a high in i t ia l stiffness that decreases as the internal vo lume increases, the material properties o f isolated bulbar 140 0 0.2 0.4 0.6 0.8 1 volume (ml) 180 i 0 0.5 1 1.5 2 2.5 volume (ml) Figure 66. The pressure-volume loops of the bulbus arteriosus. A. The typical r-shaped curve. A sharp, initial rise in pressure (1) followed by a compliant plateau phase (2). B. Less common bulbus inflation curve. An r-J curve. A sharp, initial rise in pressure (1), a compliant plateau phase (2) and another sharp rise in pressure (3). 141 rings (F igure 40, 41) indicate that the bulbus has J-shaped stiffness. The bulbar material becomes increasingly st i f f as it is stretched. Problems with Traditional Analyses In many studies o f arterial mechanics, inflations are ut i l i zed as the pr imary source o f data. F r o m the inf lat ion data, a series o f analyses results in values o f stress, strain and modulus that come direct ly f r om the properties o f the entire structure instead o f isolated fragments o f material. However , we were unable to perform theses analyses for a number o f reasons. U s ing inf lat ion data to analyze the material properties o f an artery requires certain assumptions. W h i l e some o f those assumptions, such as the Poisson 's ratio for b io log ica l materials, are va l id for the bulbus, others are not. Arteries general ly have a w a l l thickness o f between 5 - 1 0 % o f their radius. A s such, when stress in the wa l l is calculated using values o f internal radius and pressure, the differences between the forces acting on the inner and the outer wa l l are considered neglible. Therefore, equations such as a = Pr/h (5) (P = pressure, r = radius and h = thickness) a l l ow the calculat ion o f wa l l stress f rom inf lat ion data. This equation a l lows the calculat ion o f the average wa l l stress in th in-wa l led cyl inders. However , the bulbus has a wa l l thickness that can be anywhere f rom 1 0 % to 8 0 % o f the radius, and for th ick -wal led cyl inders, the variation in stress between the outer and 142 inner surfaces becomes significant. Therefore, a special th i ck -wa l led cy l inder solut ion is required. < W = P,(a 2 + b 2)/(b 2-a 2) (Harvey, 1974) (6) where a = inner radius, b = outer radius, and pi equals interior pressure was developed to help deal w i t h the stress inequality in a thick wa l led pressure vessel. I f the values o f stress obtained w i t h the th i ck -wa l l fo rmula are compared w i th those obtained us ing the th in-wa l l or average stress formula, a large difference between the two values occurs as the wa l l thickness increases. F o r this reason, the use o f the thin wa l l f o rmu la to get a measure o f average stress greatly underestimates the value o f stress along the int ima. However , us ing the th ick wa l l f o rmu la does not appreciably s impl i f y the analysis. The thick wa l l f o rmu la was designed for vessels w i th a un i fo rm wa l l thickness. The wa l l thickness o f the bulbus varies a long its length and changes f rom beat to beat. Therefore, the compl icated shape and internal anatomy o f the bulbus precludes the use o f inf lat ion data to derive its mechanica l properties. 2. Inflation Mechanics: the sharp rise The sharp in it ia l rise in the r-shaped inf lat ion curve o f the bulbus is very unusual. A large in it ia l stiffness that decreases during expansion is counter- intuit ive when examin ing normal b lood vessels. However , the bulbus is not a " no rma l " b lood vessel. It possesses an unusual shape, special ized external and internal morphology, and unique molecu lar components. A n y one o f these exc lus ive ly bulbar features may be responsible for the in i t ia l rise in pressure seen in bulbar inflations. 143 Effects of Shape Di f ferent teleosts have differently shaped bulb i (Santer, 1985). They can be pear-shaped or round, long and thin or short and wide. Yet , the shape o f the bulbus appears to be irrelevant to the shape o f the inf lat ion curve. The blue mar l in and the ye l l ow f i n tuna have drast ical ly different bulbar shapes. The ye l l owf in tuna bulbus is c lass ica l ly bu lb-shaped and quite f i rm, wh i l e the blue mar l in bulbus is long, thin and l imp (F igure 9). Despite their disparity in gross morphology, inf lat ions o f the two bulb i result in nearly identical pressure-volume loops (Figure 28) w i th respect to the in it ia l slope and pressure level o f the plateau. The shape o f the bulbus is due more to the shape o f the f i sh than to inf lat ion properties. The marl in is a long, thin f i sh w i th a long, thin pericardial cavity; therefore, the marl in bulbus is also long and thin. B y the same token, the footbal l shaped tuna has a relat ively short, round bulbus. Effect of Wall Morphology W a l l morpho logy is very complex in bu lb i ( L i ch t and Harr i s , 1973, Priede, 1975, Watson and Cobb, 1979, Benjamin, 1982, Icardo etal. 1999a, Icardo etal. 1999b, Icardo et al. 2000), w i th var ious species possessing different orientations o f elastin, smooth musc le and col lagen w i th in the bulbar wa l l . In tuna, col lagen is p r imar i l y relegated to the outer surface o f the bulbus in longitudinal and circumferent ia l layers (F igure 21). The elastin and smooth muscle layers are more complex, having longitudinal, c i rcumferent ia l and radial orientations (Figures 12, 14, 22, 23) running throughout the media. The orientations o f elastin and smooth muscle at the anterior o f the bulbus are different than their orientation at the posterior (Figure 13). These changes in the wa l l design are 1 4 4 reflected in the material properties, wh i ch are not the same throughout the bulbus o f the ye l l ow f i n tuna (F igure 40) and blue mar l in (Figure 41). In arteries, maintenance o f mechanica l properties o f vessels results f rom the correct arrangement o f smooth muscle cells and extracel lular fibrous proteins in the wa l l (Cenacchi et al., 1995). The al ignment o f the bulbar materials was a poss ib i l i ty as a cause o f the r-shaped curve. W h i l e interesting, the internal morpho logy is not the cause o f the r-shaped bulbar inf lat ion. Internal morpho logy varies w ide l y among different teleosts (L i cht and Harr is, 1973, Priede, 1975, Watson and Cobb, 1979, Benjamin etal., 1983, Icardo etal. 1999, Icardo et al. 2000) wh i le the P -V loops remain similar. Smooth muscle, vascular izat ion and innervation vary w ide l y between fish, often associated w i th activity. F o r example, the bu lb i o f tuna (Figure 18, 24), marl in (Dav ie and Daxboeck, 1984) and ra inbow trout (Santer, 1985), al l have h ighly vascular ized and muscular bu lb i , wh i l e the pla ice (Watson and Cobb, 1979), carp (L icht and Harr is, 1973) and st ickleback (Santer and Scarborough, 1983) do not. I f a universal phenomenon such as the r-shaped P - V loop was solely due to the specif ic locations o f the elastin, col lagen and smooth muscle layers, more s imi lar i ty in morpho logy between species wou ld be expected. Furthermore, radical ly disrupting the internal morphology does not cause the r-shaped curve to become J-shaped. When inf lat ing them inside out destroyed the internal design o f blue mar l in bulbi, the P -V loops remained r-shaped (Figure 35). W h e n marl in bulbi were sectioned before inf lat ion, separating the anterior end f rom the posterior end, r-shaped P - V loops (Figure 34) sti l l resulted. A l though the different segments d id not possess identical inf lat ion curves, cons ider ing the large differences in morphology, the s imi lar i ty between the curves is remarkable. 145 W a l l design does affect inflations, as the posterior, bulbus segments reached higher peak pressures than s imi lar ly sized anterior segments. The fact that differential stiffness and strain do occur along the length o f the bulbus (Figures 40, 41) may a l l ow a un i f ied inf lat ion in a structure where radius and wa l l thickness are not constant. I f the bulbus were made o f a single material w i th identical material properties throughout, certain portions wou l d preferential ly inflate, much l i ke a long bal loon. Effects of Interactions between Elastin, Collagen and Smooth Muscle W h i l e the location o f elastin, col lagen and smooth muscle w i th in the bulbar wa l l are not crucial for creating the r-shaped curve, interactions between the three proteins may be. E las t in and col lagen behave differently depending on how they are organized, and novel connections and orientations a l l ow for novel properties. W h i l e co l lagen is ord inar i ly much stronger than elastin, the connections between elastin composites in the l igaments o f bird 's w ings are so strong that the point o f fai lure usual ly occurs in the col lagenous sections ( B r o w n et al. 1994). The bulbus arteriosis is made o f the same elastin, col lagen and smooth muscle as an artery; however, the internal organization o f these materials is quite different. B u l b i lack lamel lar units (Watson and Cobb, 1979). Instead o f lamel lar sheets o f elastin, bulbar elastin occurs as loose f ibri ls. M o s t bulbi have one or more layers o f spiral ly arranged smooth muscle (Santer., 1985) running through the media wh i ch is c losely al igned w i th the loose elastin structure (Figure 22, 23). In the ye l l owf in tuna, the orientation o f the elastin f ib r i l s d i rect ly f o l l ows that o f the smooth muscle cells. Ben jamin et al. (1983) found that the smooth muscle and elastin fibers in the st ickleback bulbus were attached. 146 W h i l e w e were unable to demonstrate this, the corresponding al ignments o f the elastin and muscle f ibers (Figure 22, 23) suggest that they could be attached in some manner. Bu lba r co l lagen is almost exc lus ive ly conf ined to the outer layer o f the bulbus, f o rm ing a thin, f ibrous adventit ia (F igure 20). The exception to adventit ial col lagen was the r ich band o f subendothelial col lagen at the ventr icular-bulbar junct ion (F igure 27). Th i s co l lagen almost certainly corresponds to the va lve seen in F i gure 17. Co l lagen bundles are found in longitudinal clusters along the ventr icular border o f the ventr icular-bulbar va lve in white bass (Morone chrysops) (Raso, 1993) and are the basis o f the valves in the conus arteriosus o f elasmobranchs (Hamlett et al, 1996). E last in, smooth muscle and col lagen are in a special arrangement wi th in the bulbus and, as such, may be interacting to a l low the r-shaped inf lat ion. The chemica l b l ock i ng o f the smooth muscle and the denaturation o f the col lagen resulted in mod i f ied P - V loops (Figure 33). When smooth muscle and col lagen were prevented f rom part ic ipat ing in the bulbus inf lat ion, the result was a large fa l l in the level o f the plateau f rom the control value. The fa l l was not nearly as pronounced f o l l o w i n g the removal o f col lagen as it was f o l l ow ing the removal o f smooth muscle, suggesting that the smooth musc le plays a more important role in the inf lat ion o f the bulbus than the col lagen does. W h i l e the P -V inf lat ion curve was affected by these modif icat ions, it remained r-shaped. The L a w of Laplace The answer to part o f the bulbar mystery lies w i th in the relationship between the size o f the bulbus lumen at peak systole and the size when the bulbus is empty. The external diameter o f the bulbus is much larger than that o f the ventral aorta. However , the 147 th ick bulbar wa l l , combined w i th the longitudinal elements, results in a lumen that is relat ively small. A t the end o f diastole, when the bulbus is empty, the lumen is nearly occ luded by the longitudinal elements. A t peak systole, the longitudinal elements have been pushed out to the wal ls, resulting in a large open channel for f l o w through the bulbus (F igure 19). Th i s change in lumen radius results in the in it ia l sharp rise in the pressure-volume loop because o f a relationship between pressure and radius k n o w n as the L a w o f Lap lace. The L a w o f Lap lace states that T = P * R (7) where T is tension, P is pressure and R is radius. Tens ion is the deforming force that a l lows expansion. In essence, this states that it takes a larger pressure to inflate a small cy l inder than a large cy l inder (assuming the two structures have s imi lar wa l l thicknesses and material properties). This law explains why structures such as capi l lar ies, a lveol i and plant cel ls are capable o f withstanding high pressures without bursting. Thei r t iny radi i turn a large pressure into a very smal l tensile force. The L a w o f Lap lace is responsible for the large rise in pressure that generates the r-shaped curve (Figure 67 A ) . Regardless o f the outer dimensions, when compared w i th an artery, a bulbus has a very smal l lumen (Figure 67B ) at l o w pressures. The small internal radius results in a negl ig ib le tension in the bulbar wa l l . F o r a th in-wal led artery w i th the same outer diameter, tension rises rapidly. F o r the example in F igure 67B, the tension in i t ia l ly created in the artery is over four times larger than in the bulbus. The large tension in the artery causes it to expand a great deal at a relatively l ow pressure. Arter ies general ly expand 4 0 - 5 0 % when pressurized to phys io log ica l ranges ( M c D o n a l d , 1974). In the ye l l ow f i n tuna bulbus, the same pressure is reached for a fract ion o f the 148 Figure 67. A. Normalized pressure-volume loops from a yellowfin tuna bulbus arteriosus and ventral aorta. The r-shaped bulbus inflation allows a large increase in pressure for a small change in internal volume. The J-shaped aorta inflation requires a large amount of internal volume before it reaches a significant volume. The arrows indicate the volume required to reach the indicated pressure. B. The dimensional changes a bulbus and an artery would undergo from zero pressure to diastolic pressure. At zero pressure, the bulbus lumen is relatively smaller than the artery's, resulting in a much lower tension during inflation. The low tension in the bulbus requires a large pressure to allow expansion. A large pressure can be generated in for a small volume. The bulbus (wall thickness 80% of diameter) undergoes a strain of 10%. The larger lumen of the artery results in a high initial tension and a large amount of expansion occurs at a relatively low pressure. The artery (wall thickness 5% of diameter) undergoes a strain of 40%. D = external diameter d = lumen diameter (Values of strain for the bulbus were calculated using the VDA. Values of strain for the proximal aorta came from McDonald, 1974). 150 volume. G o i n g f rom zero pressure to phys io log ica l general ly on ly requires a strain o f around 1 0 % . In bulb i , the smal l lumen requires a large pressure in order to generate a large tension and a l l ow expansion. However , as the lumen grows, tension increases, and the pressure increments needed to open the lumen further decrease, generating the r-shaped curve. Ev idence for this interaction was seen when the internal layers were removed f rom mar l in bulb i (F igure 36). A s the effective internal radius increased, the pressure increments required for expansion decreased. Th i s resulted in a drop in the level o f the plateau to the point where the inf lat ion became J-shaped. A s the wal l s were dissected away to increase lumen size, an interaction between the wa l l thickness and wa l l stress occurs. A c co rd i n g to equation (5), as wa l l thickness decreases, stress rises, wh i ch causes a large rise in the stiffness and creates a J-shaped curve. The ye l l ow f i n tuna dorsal aortas that were infected w i th parastic cestodes (Figure 32b) revealed themselves to be a natural experiment and demonstrated the universal appl icab i l i ty o f the L a w o f Laplace. W h i l e a normal dorsal aorta has a pressure-volume loop wh i ch is essentially J-shaped (Figure 32a), the P -V loop o f the parastized aortas, despite having the same lamellar units o f col lagen, elastin and smooth muscle as the unparasit ized specimens, were quite different. Instead o f the gradual ly increasing J -shaped slope, the parasitized vessel required a large init ial pressure increment to open the vessel for f low. Just as the bulbus required a large pressure increment to push the longitudinal elements against the inner wa l l , the parasitized dorsal aorta required a large pressure gradient to push the long thin larval cestodes aside and increase the lumen diameter. 151 However , despite the s imilar it ies between the bulbus and parastized aorta wi th respect to the in it ia l pressure increments, at large expansions, the parastized aorta showed the large rise in stiffness characteristic o f arteries wh i l e the stiffness o f the bulbus remained l ow (Figure 32c). The sharp init ial increases in pressure dur ing the inf lat ions o f both the bulbus and the parasitized aorta are due to the L a w o f Laplace, but at high expansions, the parasitized aorta behaves as an aorta whi le the bulbus curve remains r-shaped. Another mechanism is at work, wh i ch results in the h ighly compl iant plateau phase o f the bulbus. 3. Inflation Mechanics: material properties Bulbus vs. Artery The mechanical properties o f a bulbus are not the same as an aorta, despite being composed o f the same materials. In both ye l l owf in tuna (Figure 40) and blue mar l in (Figure 41), the modulus o f the ventral aorta is several t imes higher than the modulus o f the bulbus over the entire phys io log ica l range and beyond. The bulbus is also much more extensible than the ventral aorta. These differences in material properties ref lect fundamental differences in the wa l l design. Contrary to what occurs in many f ish, (Leknes, 1986), the ye l l owf i n ventral aorta has the typical arterial design (Dobr in, 1978) o f concentr ic elastic lamellae separated by layers o f smooth muscle and col lagen (F igure 15). A s mentioned previously, bulbar elastin is found, not in lamellae (F igure 15), but in loose f ibr i l s ( L i ch t and Harris, 1973, Serafini-Fracassini, 1978, Ben jamin etal. 1983, Icardo et al, 2000) (Figure 22). Compared w i th the amount o f elastin in the bulbus, 152 smooth muscle is relatively scarce and col lagen is pr imar i l y relegated to the thin adventit ia (F igure 20). The lack o f lamellar sheets, together w i th the uncoup l ing o f col lagen f rom elastin, is pr imar i l y the cause o f the l ow modulus. Advent i t ia l Collagen The adventit ia is a loose, f ibrous layer, not as dense as the outer media, and composed pr imar i l y o f col lagen (Figures 20, 21). The c i rcumferent ia l ly arranged col lagen bundles o f the adventit ia appear quite wavy under the electron microscope (Figure 21). The l ow init ial slope o f the stress-strain curve may be the result o f merely straightening out the fo lded col lagen bundles before actually pu l l i ng on the fibers themselves. Ben jamin et al. (1983), Raso (1993), Icardo et al. (1999a), Icardo et al. (1999b) and Icardo et al (2000) have suggested that the adventit ia is pr imar i ly responsible for l im i t i ng the radial distension o f the bulbus. This is unl ike ly, however, because the adventit ia has a modulus that is s imi lar to that o f the outer media over the phys io log ica l range (Figure 39). Th i s l o w circumferent ia l modulus is partly due to the fact that the col lagen in the adventit ia is predominantly oriented to prevent longitudinal rather than radial expansion (Figure 21). The material properties bear this out; the outer surface is only stiffer than the inner layer when stretched in the longitudinal d irect ion (Figure 44). The informat ion that the adventit ia and outer media possess s imi lar c i rcumferent ia l modu l i , coupled w i th the fact that the adventit ia is much thinner than the media, shows that the adventit ia does not play as large a role in l im i t i ng the expansion o f the bulbus as was prev ious ly speculated. W h i l e the common model o f elastin-collagen interactions suggests that col lagen is the strain-l imiter, B r o w n et al. (1994) showed that elastin composites are, in fact, capable o f 153 becoming nearly inextensible w i th a high rupture strength. In the bulbus, the adventit ia does not appear to l imit radial expansion unti l very large strains are reached. In fact, the col lagen in the bulbus is arranged in such a manner that radial distensibi l i ty, rather than strain- l imitat ion appears to be the goal. B e y o n d the fact that the col lagen is pr imar i l y conf ined to the adventitial layer, the e la s t inxo l lagen ratio is very high. In ra inbow trout, the e las t inxo l lagen ratio is about 14 (Seraf ini-Fracass ini et al, 1978), compared w i th 1.5 in mammal ian prox imal aorta (MacDona ld , 1974). The higher the e la s t inxo l lagen ratio, the more compl iant the structure. L i c h t and Harr i s (1973) found the bulbus to be th i rty-two times more distensible than the human thoracic aorta over the same pressure range. The majority o f the adventitial col lagen is arranged long i tudinal ly (F igure 21) and, therefore, does not strongly resist c ircumferential expansion. W i t h one exception, when various pieces and layers o f the bulbus were stretched longitudinal ly or c i rcumferent ia l ly, there was a remarkable consistency o f material properties (F igure 44). Compared w i t h the other tissues, the outer layer became very st i f f at a l ow strain when stretched in the longitudinal direction. The outer layer was l imited to a strain o f 0.5 when stretched longitudinal ly, wh i le the other tissues reached strains greater than 1. This m a x i m u m strain, derived f rom the tensile tests, agrees wi th what was recorded during in vivo measurements w i th the V D A . The max imum in vivo longitudinal strain reached by any bulbus was 0.48. These results support the hypothesis that the col lagen in the adventit ia is pr imar i l y oriented to resist longitudinal rather than c i rcumferent ia l expansion. 154 Elast in The bulbus is composed pr imar i ly o f elastin (Figures 10, 15). Est imates o f the actual amount o f elastin w i th in the bulbus are as high as 7 0 % (Seraf in i -Fracass in i , 1978) to 90%) ( L i ch t and Harr is, 1973). U n l i k e mammal ian elastin, teleost elastin is not found in elastic fibers composed o f an amorphous elastin component and elastin-associated g lycoprote in microf ibr i l s . Elast in-associated microf ibr i l s have been described in some teleost arteries, however, their occurrence is rare ( Isokawa, 1990). M o r e importantly, teleost arterial elastin is almost exc lus ive ly found in a f ib r i l l a r form, an elastin morpho logy shared by the bulbus. The elastin f ibr i l s found throughout the bulbus have been described only in fish, and have been compared to the g lycoprote in mic ro f ib r i l s associated w i th amorphous elastin to make elastic fibers in vertebrates (Yamauch i , 1980). W i t h the exceptions o f lampreys ( DeMon t and Wright, 1993) and hagfish (Wright, 1984), these g lycoprote in microf ib r i l s are not considered to be load-bearing in vertebrate arteries. The aortas o f these pr imi t i ve vertebrates contain no elastin, little col lagen and are pr imar i l y composed o f microf ibr i l s . The microf ibr i l s found in cyelostome arteries appear to be direct ly homologous to the elastin associated microf ibr i l s o f vertebrate elastic fibers ( D e M o n t and Wr ight, 1993). The lack o f tensile strength is not a detriment in the l o w -pressure arterial systems o f the agnathans. W h i l e bulbar microf ibr i l s superf ic ia l ly resemble the g lycoprote in mic ro f ib r i l s o f mammal i an elastic fibers, Serafini-Fracassini et al. (1978), Ben jamin et al. (1983) and I sokawa et al. (1988) have def in i t ive ly demonstrated, using elastases, elastin stains and molecu lar analyses, that the loose f ibr i l s are elastin. The benefits o f the f ib r i l design vs. the more traditional lamel lar structures can be seen in the diameter changes the bulbus 155 undergoes throughout a cardiac cycle. The bulbus experiences considerable radial expansion dur ing each heart beat, and Benjamin et al. (1983) has suggested that the lamel lar sheets wou ld be unable to manage the large-scale length changes required. E i ther the sheets wou l d tear at high strains or the fo ld ing necessary at l ow strains w o u l d disrupt the structure o f the bulbar wal l . The elastin f ibr i ls, by benefit o f being unattached to anything but ground substance, are able to slide past one another dur ing regular expansion without damage. Lamellae in Whales: low modulus and high compliance However , the problems suggested by Ben jamin et al. (1983) could be remedied by lamel lar modif icat ions wh i ch are relatively minor compared to the systemic changes wh i ch have occured in the bulbus. Longer lamellae, wh ich wou ld require a larger extension before becoming taut, wou ld el iminate the problems o f damage at large strains. Pack i ng the lamellae at l ow strains is un l i ke ly to be important, as the inner bulbar wa l l is already thrown into numerous folds and trabeculae. F i n whales have changed their arterial mechanics through s imi lar modifcat ions o f the lamel lar unit (Gos l ine and Shadwick, 1996). In whales, the prox imal aorta is expanded to fo rm an aortic bulb, capable o f large expansions. A t pressures above phys io log ica l , the aortic bulb st i l l expands without large increases in stiffness, very different f r om the normal behaviour o f arteries. The modif icat ions in the f in whale bulb include: increased lamel lar numbers resulting in a decreased tension per lamel lar unit, a high e la s t inxo l l agen ratio, thicker elastin layers and more diffuse col lagen. These changes result in vessel wa l l stress being resisted predominantly by elastin, even at high 156 strains. Ve r y different properties occur in the thoracic aorta o f the fin whale. The thoracic aorta has a very thin wa l l , result ing in fewer lamellae than w o u l d be expected, therefore, each lamel lae is under more tension than normal lamellae. T o combat this, the elastin layers are supported by relatively th ick col lagen layers wh i ch are al igned to combat c i rcumferent ia l expansion. The fin wha le shows that modi f icat ion o f the lamel lar number and structure a l lows a range o f properties f rom very st i f f to very distensible. Contrary to the assertions o f Ben jam in et al. (1983), the lamellar structure wou ld not prevent the piscine bulbus f rom being a very compl iant, highly extensible structure. Indeed, many o f the features that make the aortic bulb o f the fin whale compl iant are also found in the bulbus arteriosus: h igh elast in-col lagen ratio, thicker elastin layers and more diffuse col lagen. Despite s imi lar designs in the gross morphology o f the whale bulb and the teleost bulbus, a b i g difference in the material properties exists. The modulus o f the fin whale aortic bulb is at least twice as high as that o f the bulbus over a w ide range o f strains, often much higher (Shadwick and Gos l ine, 1994). This suggests that the f ib r i l l a r design o f the bulbus does not combat the dangers o f large expansions but lowers the elastic modulus o f the material. The f ibr i l s o f the bulbus arteriosus, and the result ing lack o f cohes ive structure, a l l ow a reduced stiffness. It is un l i ke l y that the l ow modulus o f the bulbus is due merely to the gross structure o f the elastin f ibr i ls. Bu lba r elastin is chemica l l y distinct f r om that o f mammal i an elastin ( C h o w et al., 1989; Spina et al, 1979) and even f rom elastin in the teleost ventral aorta (L icht and Harr is, 1973). Bu lbus elastin is also soluble in hot a lkal i ( L i cht and Harr is, 1973), suggesting that it is less hydrophobic than other elastins. In fact, 157 when examined, the elastin f rom teleosts possesses less hydrophobic amino acids than other elastin variants (Seraf ini-Fracass ini, 1978; Sage, 1982). The increase in polar amino acids may decrease the modulus, as much o f the recoi l in elastin is dr iven by hydrophobic interactions. Therefore, the answer to the question about the bulbus ' s extens ib i l i ty on the plateau o f the curve is multifaceted. The arrangement o f the col lagen, the structure o f the elastin and the chemica l nature o f the elastin al l combine to a l low large vo lume changes w i th in the bulbus for small pressure changes dur ing the plateau phase o f the inf lat ion. 4. Inflation Mechanics: the final rise in stiffness It has been established that the inf lat ion o f the bulbus arteriosus can be broken into two different sections. The init ial rise in the bulbar P - V loops is due to the gross morphology o f the bulbar lumen wh i le the extreme bulbar compl iance on the plateau o f the P - V loop is due to the l ow modulus and high extensibi l i ty o f the specia l ized bulbar wa l l . However , there is a third element to bulbar inflations wh i ch has been ignored.. Under certain condit ions, the r-shaped curve shows a final upturn that can only be due to a large increase in stiffness. Du r i ng the inflations o f bulb i w i t h wa l l layers removed (Figure 36), the final curve showed a large increase in stiffness. In fact, the P - V loop was highly reminiscent o f an artery 's J-shaped loop. A final large increase in stiffness was also seen in the P -V loops recorded in situ using the V D A (Figure 37, 55). Tens i le tests show the cause o f the final rise in stiffness o f the bulbus. F igures 39-44 demonstrate an obvious increase in stiffness when isolated tissues were subjected to 158 high strains. Th i s suggests that, at very high strains, the adventit ia may play a role in strain- l imitat ion because the col lagenous adventitial layer is much stiffer than the outer media l layer above strains o f one (Figure 39). However, at strains be low one, the mechanica l properties are very similar. A t these levels, the outer medial layer actually bears more o f the load in the wa l l than the adventit ia does, due to differences in the w a l l thicknesses (adventitial layer: 0.04 cm, outer medial layer: 0.4 cm). On l y at high strains does the rise in adventit ial stiffness result in the majority o f the load being borne by this outer, col lagenous layer. Tissue samples encompass ing the entire wa l l demonstrated the same phenomena (Figure 40-41). A t in vivo strains, the slopes o f the stress-strain and modulus-strain curves are low. The large increase in stiffness does not occur until strains in excess o f one, and the chances o f the bulbus reaching such a large strain under normal condit ions are s l im. Record ings o f a beating heart f rom an anaesthetized ye l l ow f i n tuna suggest that for the major ity o f beats analyzed, the beat-to-beat variation in strain is on the order o f 0.3-0.4 between systole and diastole. The stroke vo lume o f y e l l ow f i n tuna is in the range o f 0.57-0.95 m l k g - 1 (Jones et al. 1993). In one o f our in situ tests we inflated the bulbus f rom 1.2 kg ye l l ow f i n tuna w i th 2 ml o f f lu id. Despite this except ional ly large stroke volume, the bulbus on ly reached a circumferential strain o f 0.47. Tens i le tests on isolated loops f rom ye l l owf i n tuna showed that, independent o f posit ion, the materials had s imi lar properties at strains be low 0.8 (Figure 40). S ince the in vivo strains are be low that level, the different portions o f the bulbus have the same mechanical properties under normal cardiovascular conditions. S imi la r properties were seen in blue mar l in bulbar tissue. W i t h no recordings o f a blue marl in heart beating, we 159 were unable to get an estimate for the in vivo strains experienced by a mar l in bulbus. However , the s imi lar it ies between both the respective ecologies o f tuna and mar l in and the material properties o f their bulb i suggest that the mar l in bulbus experiences a range o f strains s imi lar to that found in the ye l l owf in tuna. In that case, the marl in bulbus wou ld be in no danger o f fai lure during normal operations. The study in wh i ch sections o f the wa l l were sequentially dissected away (F igure 36) is obv ious ly unnatural. B y remov ing tissue layers, the normal arrangements o f tissue are destroyed. Ordinar i ly , the media holds the adventit ia in a relatively loose arrangement. W i t h this tissue arrangement, the adventit ia does not become taut unti l the media stretches to the point where all the " s lack " in the adventit ia is taken up. When the media was removed f rom the wa l l , the bulbus structure effect ively col lapsed and saline cou ld be added without a corresponding rise in pressure. Therefore, as we added vo lume to reach zero pressure, the bulbus became preloaded and the adventit ial " s lack " was taken up. This resulted in the adventit ia becoming taut at a l ow Av. The outer layer experienced a very large in it ia l volume, reached a large strain at a l ow pressure and showed the large rise in stiffness seen in the isolated tissue loops. I also encountered inflations o f who le bulb i in wh i ch the outer col lagen layer was recruited to prevent over-expansion (Figures 37, 55). The reason was pharmacolog ical . The bulbus that was inflated in situ and showed a large amount o f c i rcumferent ia l stiffness (F igure 37) came f rom a fish wh i ch been used in an A N F experiment. A m o n g the phys io log ica l effects o f A N F in teleosts is the abi l i ty to relax vascular smooth muscle (Ac ie rno et al., 1991). The bulbar smooth muscle was relaxed, causing the bulbus to be except ional ly compl iant and a l l ow ing it to extend more than usual for a g iven pressure. 160 Th i s bulbus showed a large increase in stiffness at 80 m m H g , wh i l e ordinary bu lb i reached pressures over 100 mm H g without any large changes in stiffness. F o r a g iven pressure, the ANF - a f f e c t ed bulbus reached much higher levels o f strain than d id an ordinary bulbus, pushing the wa l l material onto the steep part o f the modulus-strain curve and dramat ica l ly increasing the stiffness o f the bulbus. 5. Functions of the Bulbus Arteriosus The bulbus arteriosus is a Windkessel . It depulsates and smoothes pressure and f l ow, protecting the g i l l vasculature and prolonging f l ow dur ing diastole. The effect it has on both b lood vo lume and b lood pressure are undeniable. Yet, how a relat ively short bulbus m im i c s the effects o f longer arterial trees o f terrestrial vertebrates has never been explained. K n o w l e d g e o f the inf lat ion mechanics provides the first steps towards this understanding. However , to make inferences based on the in vitro inf lat ion curve, it is vital that the bulbus behave s imi lar ly in vivo. That the bulbus does, indeed, behave the same in vivo as it does in vitro is graphica l ly shown by Figures 45-50. F igure 45 shows close matching between a single beat and a single inf lat ion. B l o o d in i t ia l ly entering the bulbus results in a large jump in the pressure, fo l l owed by a stage in the inf lat ion during wh i ch the bulbus becomes much more compl iant and a l lows large vo lume changes for small rises in pressure. In fact, al l three features o f the bulbus inf lat ion can be recreated f rom actual beats. The Lap lac i an relationship at small volumes that results in large pressure changes for small vo lumes can be seen in Figure 46. The compl iance o f the bulbus near mean systol ic pressure is 161 obvious in F igures 48, 49. A n d the f inal rise in stiffness occurs at a pressure o f 160 mm H g in F i gure 50. Volume Reservoir That the bulbus acts as a vo lume reservoir is obvious to anyone who has seen a teleost heart beating. The bulbus is capable o f both expanding to store cardiac output and reco i l ing elast ical ly to return the stored f lu id to the circulat ion. In f i sh w i t h long diastol ic periods, it is crucial to have a reservoir capable o f mainta in ing f l ow over the gil ls. In l ing cod, Ophiodon elongatus, b lood f l ow in the ventral aorta due to the elastic rebound o f the bulbus arteriosus represents about 2 9 % o f total cardiac output (Randal l , 1968). When contracted, the bulbus vo lume is smaller than a single stroke volume. However , estimates o f 25%> (Priede, 1976) to 200-300 % o f stroke volume (Bushnel l et al. 1992) show that the bulbus is capable o f ho ld ing a very large amount o f b lood. M y studies suggest that the capacity o f the ye l l l owf i n tuna bulbus is in excess o f the high value. The plateau phase o f the bulbar inf lat ion aids in its role as a vo lume reservoir. Over a phys io log ica l pressure range o f 25 mm Hg , the bulbus can ho ld and return 8 0 % o f its vo lume (Figure 68A) , a A v o f 0.8. A n artery, over the same range, shows a A v o f 0.05, wh i ch is only a fract ion o f the capacitance o f the bulbus. In vivo, the large changes in internal vo lume that ye l l owf i n tuna bulb i experience can be clear ly seen in F igure 60, where bulbus vo lume ranges f rom 0.2 to 0.8 ml throughout the cardiac cycle. Du r i n g an inf lat ion cycle, these vo lume changes the bulbus experiences wo r k out to c i rcumferent ia l strain changes o f 3 0 - 4 0 % (Figure 68B). This behaviour is in stark contrast to the normal funct ion ing o f arteries, wh i ch typ ica l l y experience c i rcumferent ia l strains o f 2 - 7 % during 162 an inf lat ion cyc le (McDona l d , 1974). These large differences between the behaviour o f arteries and bulb i i l lustrate two different means to the same end. B o t h bulbi and arteries are designed to increase the capacitance in the c irculatory system in order to depulsate and attenuate f l ows and pressures. However , as can be seen in F i gure 68, arteries are not very compliant. Capacitance in arteries is achieved through length. E ven a relatively inextensible tube can prov ide s ignif icant capacitance i f o f suff icient length. Teleosts lack the luxury o f a long arterial tree separating the heart f rom the gi l ls. Instead, capacitance is increased w i th the bulbus and its r-shaped inflation. The tremendous compl iance o f the bulbus on the plateau o f its P -V loop results in a large A v over the phys io log ica l pressure range and a l lows a relatively short bulbus greatly increase the capacitance o f the teleost arterial system. Observations o f the in vivo beating o f the heart show that the bulbus funct ions in an inflated state (Figures 59-62), and, f o l l ow ing a long bradycardia, I wou ld witness the bulbus " pump " itself up. T o reach a size where the f l ows in and out o f the bulbus were in balance often required 2-3 beats (Figure 63). I f a bradycardia were induced, causing the bulbus to empty, the same inf lat ion pattern wou ld be repeated. This suggests that the bulbus normal ly maintains a large vo lume o f blood. Due to the vo lume o f b lood stored in the bulbus, it is un l ike ly that the bulbus wou ld completely empty (Figures 59-62). Even dur ing a long diastol ic period, the f lu id reserves in the bulbus a l low posit ive f l o w to occur. L i c h t and Harr i s (1973) found the bulbus o f the carp, Cyprinus carpio, to be 32 X more distensible than the human thoracic aorta over a s imi lar pressure range. Because o f the extreme compl iance o f the bulbus over the plateau region o f its inf lat ion curve, smal l 163 Figure 68. A. Comparison of the volume changes (Av) experienced by a bulbus arteriosus and ventral aorta over a physiological pressure range. The thick lines indicate the bulbar volume changes and the thin lines indicate the ventral aortic changes. B. Comparison of the diameter changes experienced by a bulbus arteriosus (b.a.) and an artery during inflation from diastole to systole. The large strain (35%) experienced by the bulbus mirrors the large volume changes seen in (A). The artery undergoes a strain of 5%, showing the relatively low compliance seen in (A.). (Values of strain for the bulbus were calculated using the VDA. Values of strain for the proximal aorta came from McDonald, 1974). 165 changes in pressure result in large changes in volume. A t pressures near or above mean arterial pressure, the bulbus is h ighly responsive to the slightest change in pressure, expanding to a relatively large degree (Figure 5 IB ) . Consequences of Compliance The compl iance o f the bulbus makes it very sensitive to changes in cardiac output and peripheral resistance. Th i s was demonstrated in many o f our in vivo recordings showing bu lb i w h i c h dramatical ly changed in size, often in a cyc l i ca l manner (Figures 64^65). F luctuat ions in the size o f the bulbus not due to bradycardia or arrythmia cou ld result f r om an increased cardiac output or a change in the downstream resistance o f the g i l l vasculature. Random diameter changes occurr ing beat to beat are l i ke ly due to changes in cardiac output. The smooth, cyc l i ca l diameter changes, repeating over a period o f several heart beats (Figure 64, 65), suggest that a variation in the g i l l resistance is the culprit. A s f i sh breathe, the opening and c los ing o f the opercula changes the pressure in the buccal cavity (Hughes and Shelton, 1958; Randa l l et al., 1965; Ho le ton and Jones, 1975; Stevens and Randa l l , 1967) affecting the resistance in the c irculatory system. The g i l l capi l lar ies wou ld experience the increased pressure, and impedance wou l d increase. Opercular movements are un l ike ly to be the source o f the resistance changes because tuna are obligate ram venti lators and have a very l imi ted capacity for buccal pumping. However , Bushne l l and B r i l l (1992) have shown that tuna adjust the size o f their gape, contro l l ing the amount o f water f l ow over the gills. The i r data also impl ies that tuna can 166 rearrange b lood f l ow w i th in the g i l l s to br ing about s ignif icant modi f icat ions o f vascular resistance. Resistance changes in the tuna c i rculat ion cou ld also be the result o f sympathetic innervation. Waves o f weak contraction were seen running up and down the length o f the anesthetized f i sh and this f i c t ive sw imming could be responsible for the cyc l i c a l resistance changes. Whatever the source, changes in resistance cause c y c l i c changes in bulbar size. However , the size changes in the bulbus are mirrored by minute changes in pressure. B y expanding, the compl iant bulbus damps the resistance changes, and, as a consequence, the heart does not experience increased impedance. The compl iance that a l lows the bulbus to perform as an exceptional vo lume reservoir also protects it f rom over-expansion. The very fact that the bulbus can expand to such a large degree l imits the poss ibi l i ty o f rupture under normal conditions. However , extreme condit ions, such as a combinat ion o f increased g i l l resistance and cardiac output, might be a danger to the bulbus. Fortunately, the phys io logy o f the fish makes the occurrence o f those circumstances extremely unl ikely. Card iac output in fish can be increased in a number o f ways: catecholamine release, increased temperature, increased venous return, and release o f vagal inh ib i t ion (Randal l , 1968), and the best way to qu i ck l y increase g i l l resistance is to stop ventilation. However , in tuna, when water f l ow is removed, and g i l l resistance increases, the fish automatical ly experiences a bradycardia, decreasing cardiac output and removing the danger to the bulbus. I f g i l l resistance can be increased another way (ie. severe vasoconstr ict ion), the bulbus may begin to enlarge. However , there are l imits, both external and internal, w h i c h l im i t bulbar over-expansion. Its posit ion wi th in the pericardial cavity precludes over-167 expansion because as the bulbus increases in size, it w i l l interfere w i th the f i l l i n g o f the atr ium and cardiac output w i l l decrease. T o increase stiffness and prevent rupture at very high pressures, adventit ial col lagen can be recruited. A t high pressures (>140), bulbar stiffness greatly increases, the sensitivity is lost, and pressures o f 130 m m H g and 160 m m H g result in very s imi lar changes in vo lume (Figure 51C, 5 ID). Pressure Reservoir The effects o f the bulbus on pressure are undeniable. In the prox imal out f low tract, teleosts have a much smoother and slower decl ine in diastol ic pressure than elasmobranchs (Satchell and Jones, 1967; Metca l fe and Butler, 1982), amphibians, birds, or mammals (Johansen, 1962; Spencer and Greiss, 1962; M c D o n a l d , 1974). The bulbus's abi l i ty to act as a pressure reservoir, l ike its funct ion as a vo lume reservoir, stems f rom its inf lat ion dynamics. The sharp in it ia l rise o f the curve is the key. D u e to the Lap lac i an relationship between the lumen radius and the pressure required for inf lat ion, even a small stroke vo lume w i l l result in a large j ump in pressure (Figure 67). W h e n empty, the bulbus becomes " p r imed " w i th the first heart beat, regardless o f cardiac output. I f the cardiac output is high, a large pressure head w i l l sti l l be maintained, but the compl iance o f the bulbus a l lows it to expand and "absorb" excess f lu id , protecting the g i l l vasculature f rom damage by preventing large rapid, changes in pressure and f low. E v e n when the cardiac output is low, the bulbus w i l l maintain b lood f l o w through the g i l l s at a high pressure. F o l l o w i n g a long diastol ic period, the first heart beat w i l l have a larger effect on pressure than any f o l l ow i ng beats (Figure 51 A ) . The benefits o f the bulbar design is that it a l lows the bulbus to behave s imi lar ly under both high and l ow cardiac outputs. 168 In ra inbow trout, the bulbus is most compl iant at the systol ic pressure (C lark and Rodn ick , 1999), and the same occurs in ye l l owf i n tuna. Du r i ng systole, increasingly large changes in vo lume result in relatively small pressure increases. The compl iant plateau o f the bulbus that a l lows it to serve as a vo lume reservoir also a l lows the bulbus to " s t o re " pressure. Once the heart has reached its "operating pressure", the properties o f the bulbus a l l ow it to remain at that level, despite large increases in volume. B y keeping the pressure afterload relat ively constant, the heart 's wo rk is decreased. Du r i ng diastole, the plateau a l lows the bulbus to maintain a high pressure wh i l e its internal vo lume is decreasing. Prev ious ly , I said the bulbus has a large A v over a small pressure range, a id ing in its funct ion as a vo lume reservoir. The f l ip side o f this occurrence is that the bulbus maintains a very constant (relatively high) pressure over a large decrease in pressure. The bulbus can lose 8 0 % o f the vo lume and pressure w i l l on ly fa l l by a smal l amount (Figure 68). K e e p i n g b lood go ing through the gi l l s is o f vital importance to teleosts, especial ly those w i th l ow heart rates. I f the bulbus were designed with the J-shaped inf lat ion characteristics o f an artery, the heart wou ld have a much more d i f f icu l t task. A n artery-l ike bulbus wou l d need to be almost completely f i l l ed in order to reach a high pressure, at wh ich point a rapid increase in stiffness wou l d occur (Figure 68). W o r k i n g against very r ig id wal l s w o u l d increase the w o r k o f the heart. Dur ing diastole, the artery-l ike bulbus wou l d also be problematic, as a small amount o f f lu id loss wou ld result in a large drop in the pressure head fo rc ing b lood though the gi l ls. M u c h o f the diastol ic period wou ld occur at a l ow pressure. The special modif icat ions o f the bulbus a l l ow it to act, in effect, as another contracti le chamber o f the heart. On ly instead o f us ing cardiac muscle to generate force, 169 the bulbus morpho logy a l lows it to make use o f the energy f rom the heartbeat to maintain the movement o f the b lood. The init ia l port ion o f the r-shaped curve qu i ck l y transforms the energy f r om the beat into a pressure gradient, wh i le the plateau keeps and returns the stored energy at a constant pressure level. B y maintain ing a pressure gradient, the bulbus extends the proport ion o f the cardiac cyc le dur ing w h i c h b lood f l ows into the g i l l s (Randal l , 1968; Stevens et al. 1972). Functions in Other Species These functions o f the bulbus are not restricted to the high performance tuna and marl in. The r-shaped inflations are a characteristic o f al l teleosts that we have looked at. What does change f rom species to species is the level o f the plateau (Figure 28). In fact, there is a correlation between the b lood pressure o f each fish and the pressure at wh ich its P - V loop plateaus. Y e l l o w f i n tuna, sai l f ish and marl in have high (100 m m H g ) plateaus, wh i l e deep-sea fish l i ke the pomfret and inactive fish l ike carp (L icht and Harr i s , 1973) have curves that plateau at a much lower level (30 m m Hg) . In ra inbow trout, the pressure at wh i ch the bulbus becomes most compl iant (the plateau) is very c lose to the mean systol ic b l ood pressure (C lark and Rodn ick , 1999). The plateau level set by the in it ia l rise in pressure is not random but corresponds to the b lood pressure range experienced by the fish. E a c h species's bulbus appears to be "tuned" to work at a specif ic pressure. Origin of the Tuning: smooth muscle 170 The or ig in o f the tuning is varied. Co l lagen and smooth muscle both affect the level o f the plateau o f the P - V curve (Figure 33). W h i l e the relative amount o f bulbar col lagen may be s imi lar in different species o f teleosts, the amount o f smooth muscle and vascular izat ion does vary to a large degree. Santer and Cobb (1972) and Watson and Cobb (1979) found no smooth muscle and very l itt le vascular izat ion when descr ib ing the bulbus o f the sedentary plaice. Tuna (Figures 18, 24) and marl in (Dav ie and Daxboeck, 1984) bu lb i contain a s ignif icant amount o f smooth muscle and are h ighly vascular ized. F o r f i sh w i th l ow b lood pressures, having a bulbus made o f elastin and col lagen seems to be suff ic ient to generate the proper pressure. However , f ish that have a higher performance cardiovascular system require smooth muscle for proper tuning o f the bulbus. The P - V loops in F igure 27 support this hypothesis. The hysteresis level decreases f r om f ish to f i sh as the pressure level o f the plateau drops. Hysteresis is general ly associated w i th v iscous elements, wh i ch absorb energy as they deform. In arterial wal l s , the culprit is smooth muscle (Dobr in, 1978). In the bulbus, as hysteresis decreases, so too does the amount o f smooth muscle. Bu l ba r smooth muscle is capable o f contracting to increase the stiffness o f the wa l l . Farre l l et al. (1979) showed that the bulbi f rom ra inbow trout could be made to relax when exposed to hypox ia and adrenergic st imulat ion and to contract under chol inerg ic st imulation. The ye l l owf i n tuna bulbus is innervated (Figure 27) and contains a large amount o f smooth muscle wi th obvious orientations (F igure 22, 23). Wh i l e y e l l ow f i n tuna smooth muscle cel ls do not show any obv ious connections, their close p rox imi ty to one another (Figure 24) may make them funct ional ly l inked, a id ing in smooth contractions. Watson and Cobb (1979) found the smooth muscle in a variety o f 171 fish to be jo ined into a " cab le - l i ke electrical syncyt ium" . Th i s design a l lows smooth musc le deep in the media to be excited by neural st imulation in the periphery. Gap junct ions have been clear ly ident i f ied in a number o f species (Watson and Cobb, 1979), and in species where gap junct ions are rare (ie. st icklebacks) the muscle possesses " bu t ton - l i ke " contacts wh i ch l ink adjacent smooth muscle cells (Benjamin et al. 1983). Because only the outer muscle cells have any contact w i th adventitial nerves, electrotonic coup l ing a l lows synchronized contractions. Bu l ba r smooth muscle is v i ta l ly important to the funct ion ing o f the bulbus. It is innervated (Watson and Cobb, 1979), responds to a variety o f pharmacolog ica l and environmental s t imul i (Farrel l, 1979) and causes large changes in the inf lat ion charcteristics o f the bulbus when it is prevented f rom contracting (F igure 33). This is in direct contrast to previous speculation that smooth muscle o f the bulbus merely served to produce elastin and had no structural role ( L i ch t and Harr is, 1979). The importance o f the smooth muscle to the inf lat ion properties o f the bulbus shows that fresh tissue must be ut i l ized when performing studies i nvo l v i ng the bulbus. Smooth muscle w i l l die in a matter o f hours, wh i le elastin and col lagen maintain their properties much longer. F o r this reason, the mahimahi P - V loop in F i gure 28 may not accurately reflect in vivo behaviour. When obtaining samples f r om the H o n o l u l u fish market, the fishermen w o u l d occas ional ly br ing in fish several days old. The l o w pressure level, combined w i th the small hysteresis suggests that the mahimahi, a high-performance fish, no longer had l i v i ng smooth muscle. The plateau o f the albacore P -V loop is somewhat lower than the literature value o f b lood pressure (70 m m H g ) for an albacore tuna ( La i et al., 1987). A l though we cannot rule out the poss ibi l i ty that the bulbar smooth 172 muscle was beginning to deteriorate, the elevated pressures in the literature may be due to the excited state o f the fish, wh i ch had been recently caught at sea. Tuning of the Bulbus: lumen size Adjus t ing lumen size is another method whereby the bulbus can be tuned to the respective systol ic b lood pressure o f the fish. The bulb i o f high-performance f i sh l i ke the tuna, marl in, sa i l f i sh and trout possess longitudinal elements, wh i ch effect ively occ lude the lumen at l ow pressures. When the longitudinal elements are removed (F igure 36), the plateau level o f the P -V loop drops. Th i s s imple experiment demonstrates the effectiveness o f the longitudinal elements in elevating pressure. F i sh that do not possess a high-pressure arterial system (ie carp, L i ch t and Harr is, 1973; plaice, Santer and Cobb, 1972; st ickleback, Santer and Scarborough, 1983; white-blooded antarctic fish, Chionodraco hamatus, Icardo et al., 1999a; red-blooded antarctic fish, Trematomus bernacchii, Icardo et al., 1999b; deep-sea macrourids, Greer Wa l ke r et al., 1985) lack longitudinal elements. Instead, their lumen surfaces can have ridges, folds, trabeculae or be smooth. In these fish, the lumen is un l ike ly to be completely occ luded at l o w pressure. The larger lumen diameter results in a smaller init ial pressure increment and the bulbus is tuned to work at a lower average systol ic pressure. Benefit of Bulbar Location A model study performed by Campbe l l et al (1981) showed that a large compl iance located directly outside the heart is more effective at decreasing peak systol ic pressure than a compl iance located anywhere else in the circulat ion. Since the majority o f the heart 's wo r k is invo lved in generating tension (Jones, 1991), lower ing the tension-173 t ime integral o f the heart by decreasing the peak systol ic pressure translates into large energy savings. Therefore, the posit ion o f the bulbus in the teleost c i rculat ion, just distal to the heart, makes it o f great importance for increasing the overal l e f f ic iency o f the f i sh cardiovascular system. Controlling the Bulbar Properties A s seen in F igures 51, 59, 62, the compl iance o f the bulbus arteriosus a l lows the cardiovascular system a great deal o f f lex ib i l i t y when deal ing wi th var ied pressures and cardiac outputs. Despite a variety o f different beating regimes: bradycardia, tachycardia and arrythmia, pressure remained consistent. However , having an extremely compl iant structure in series w i th the heart results in a serious problem. A n y effort to raise b lood pressure w o u l d be countered by the large compl iance. The bulbus is designed to deal w i th this possible problem. Its smooth muscle tone can be modi f ied in a variety o f ways to adjust bulbar compl iance. Innervation (Watson and Cobb, 1979) a l lows changes on a very fast t ime scale wh i l e the various endocrine b ind ing sites (Farrel l , 1979; K i n e el al, 1991; Waugh et al., 1995; Cer ra el al., 1996, Cer ra el al, 1997) on the bulbus also a l low the modulat ion o f bulbar properties. Nervous Control of the Bulbus Bu lba r innervation varies f rom species to species (Watson and Cobb, 1979) but is usual ly relegated to the adventitia, where it can initiate contraction o f the smooth muscle. Contrary to what has been found in most f i sh (Watson and Cobb, 1979), the neurons in ye l l ow f i n tuna (F igure 27) are not confined to the adventit ia but also occur in the 174 subendothelial layer, innervating muscle surrounding the bulbo-ventr icular valve. La rge amounts o f smooth muscle surround the valve in the bulbus (F igure 16), and in many teleosts, there is a r ing o f cardiac muscle at the junct ion between the ventr ic le and bulbus arteriosus (Sanchez-Quintana and Hur le, 1987; Sanchez-Quintana et al, 1996). Rad ia l insertions o f myocard ia l f ibers into the bulbo-ventr icular muscle f iber r ing have been described. Du r i n g systole, the orientation o f the fibers may contribute to active opening o f the bulbo-ventr icular or i f ice (Sanchez-Quintana and Hur le , 1987; Sanchez-Quintana et al., 1996). Ac t i ve opening o f the aortic va lve occurs in mammals (Thubr ikar et al, 1980), and pressure and f l ow recordings f rom teleosts suggest active valves are invo lved (Bushnel l et al., 1992). In teleosts, separation o f the aortic and ventr icular pressure traces occurs at, or soon after, peak pressure (Jones et al, 1992), wh i le passive va l v i ng wou ld result in separation o f the pressure traces during the decrease o f ventr icular pressure. Hormonal Control of the Bulbus W h i l e the bulbus can be affected by chol inerg ic and adrenergic agonists (Farrel l , 1979), it also has b ind ing sites for tachykin ins (Waugh et al, 1995), bradykinins (Con lon etal, 1995), and A N P ( K i m etal, 1991; Ac ie rno etal, 1991; Cer ra et al, 1992; Cer ra et al, 1996, Cer ra et al, 1997). K i n i n s have a variety o f functions depending on dose, release site and the neuronal makeup o f the target tissue (Onor i el al, 1999). In f ish, the effects o f intra-arterial injections o f bradyk in in d i f fer markedly. Na t i ve bradyk in in produces a transient hypertension in cod (Platzack and Con lon, 1997) and a complex triphasic pressor-depressor-pressor response in ra inbow trout (Olson el al, 1997). Tachyk in i n general ly acts as a vasoconstrictor, however, its functions also vary w i th 175 regard to receptor type and location (Onor i et al., 1999). In ra inbow trout intestine, tachyk in in caused contractions o f the smooth muscle as we l l as st imulat ion o f cho l inerg ic and serotonergic neurons (Jensen and Ho lmgren, 1991). In arteries, tachyk in in occas ional ly caused contractions, but the overal l funct ion is unknown (Kagstron and Ho lmgren , 1998). Natr iuret ic peptides also show mixed results in f ish. W h i l e A N P is general ly a hypotensive agent in mammals, its actions in f i sh are numerous. In ra inbow trout, A N P decreased dorsal and ventral aortic pressure, venous pressure, cardiac output, stroke vo lume and g i l l resistance. A N P also increased systemic resistance, heart rate and vascular compl iance (Olson el al., 1997). In At lant ic cod, A N F produced tachycardia as we l l as decreases in both ventral and dorsal aortic pressure (Ac ie rno et al. 1991). In freshwater eels (Oudit and But ler, 1995), A N P reduced dorsal aortic pressure, cardiac output and stroke vo lume wh i le not changing systemic resistance. The widespread actions o f both the k in ins and natriuretic peptides in fish are sti l l being unravel led. These hormones appear to have direct effects on smooth muscle as we l l as indirect effects through catecholamine release. What is clear is that they are a powerfu l means o f modulat ing the cardiovascular system. The large amount o f h igh and l ow aff in i ty receptors possessed by the bulbus for these different control hormones a l lows it to generate large changes. I f the bradykinins and A N P only increased the compl iance o f the bulbar smooth muscle and the tachykinins only reduced the compl iance, increased control o f vascular pressure, impedance, venous return, and cardiac output wou ld result. Bulbar Control of the Cardiovascular System 176 The central pos it ion o f the bulbus in the c i rcu lat ion makes it a pr ime candidate to be a modulator for the rest o f the cardiovascular system. There is mount ing evidence that the bulbus serves an endocrine function. The bulb i f r om a l l fish species examined show endothel ial cel ls possessing large membrane-bound vesicles (Benjamin et al., 1983). These vesicles have been described as l ip id granules (Santer and Cobb, 1972), moderately electron dense bodies (Leknes, 1980) and specif ic endothel ial granules (Benjamin et al., 1983). In tuna, they make up a large port ion o f the cel l vo lume in the endothel ial cel ls examined (Figure 26), mirror ing what has been seen in other species (Benjamin el al., 1983). These endothelial cells have been seen releasing their ves icular contents into the b lood stream (Benjamin et al. 1983). W h i l e the exact nature o f the contents is unclear, one can hypothesize that they contain one o f the many vasoactive peptides that the bulbus is known to possess ( K i m el al, 1991; Siharath et al., 1995). N o t only does the bulbus possess receptor sites for A N P , it has also been shown to contain levels o f A N P on a par with the atria ( K i m et al, 1991). Increased stretch in endothelial cells cou ld result in A N P release and a decrease in cardiac output. The lumen cells are in a pr ime pos it ion to transduce changes in the forces coming f rom the heart or increased impedance f rom the gil ls. These changes cou ld be manifested as increased or decreased stretch or shear, and hormones could be released in response. Functions of the Longitudinal Elements The funct ion o f longitudinal elements in teleosts has never been properly explained. Pr iede (1976) contended that the longitudinal elements acted as struts, supporting the bulbus wa l l dur ing inf lat ion cycles. H e suggested that, due to the thick 177 bulbar wa l l , the inner layers were subjected to much larger strains than outer layers. Th i s prob lem o f strain distr ibution wou ld be solved by the longitudinal elements, as they d id not undergo the same circumferent ia l strains o f the wa l l . However , this hypothesis is w rong for a number a reasons. F i gure 44 shows that the longitudinal elements have no greater stiffness or strength compared w i th the other materials in the wa l l . Long , extensible, " f l o p p y " struts wou ld be o f l ittle use in a functional role. There is also no problem due to the d i f fer ing strains between the various layers. The bulbus is not made o f an isotropic material. D i f ferent amounts and orientations o f elastin, col lagen and smooth muscle in the adventit ia and media generate a host o f material properties. F igure 3 9 shows that the bulbus was designed to accommodate the differences in strain between the inner and outer layers. The medial layer breaks at a much higher strain than the adventitia. W h e n the bulbus expands, the media experiences a larger absolute strain than the adventitia. However , the media has a higher breaking strain than the adventitia. Des i gn ing the bulbus w i th anisotropic properties, speci f ica l ly a higher breaking strain for the media, prevents early fai lure. The bulb i o f al l f i sh are subject to large expansions, wh i le longitudinal elements seem to be l imited to high performance fish. A n y funct ional explanation o f the longitudinal elements wou ld need to take into account the b io logy o f these species. These fish ( inc lud ing tuna, marl in, sai lf ish, trout, salmon) have a suite o f adaptations a l l ow ing them to maintain act iv ity far longer than the average teleost. That these species also possess longitudinal elements suggests that the longitudinal elements are also an adaptation towards a high-performance life. W e have seen that when the longitudinal 178 elements are removed, the magnitude o f the pressure-volume curve drops (Figure 36). W h e n the longitudinal elements fill up the lumen dur ing diastole (F igure 19), the bulbus requires a higher pressure gradient for inflation. Since bulb i are tuned to wo r k at the operating pressure o f the fish, longitudinal elements may be necessary to generate the higher pressure requirements o f those high performance fish w h i c h possess them. B y ut i l i z ing the system o f longitudinal elements, the relaxed bulbus possesses a small out f low tract. Du r i ng inf lat ion, the compliant, easily stretched longitudinal elements are pushed out to the sides o f the vessel wa l l , greatly increasing the lumen size. W h e n inflated, the bulbus is shaped l ike an aneurysm, and in some respects, the bulbus is s imi lar to an aneurysm. Aneurysms have been shown to damp osci l latory f l o w and act as vo lume reservoirs (Jain, 1963). Yet, the bulbus does not seem to suffer f rom the problems o f an aneurysm, specif ical ly, rupture or thrombose format ion due to recirculat ion o f b lood (Jain, 1963; Scherer, 1973). In an aneurysm, wh i l e a center core o f f l u id f l ows through, a large amount o f b lood becomes trapped and recirculates, eventually c lott ing and entering the b lood stream as a thrombose (Scherer, 1973). M o d e l studies have demonstrated that the higher the velocity o f the central core, the smaller the core becomes, thereby increasing the amount o f recirculat ing f l u id (Scherer, 1973). Since the higher performance fish have higher cardiac outputs, the b lood is under a higher velocity, and the amount o f fluid recirculat ing wou ld increase. Th i s potential ly fatal situation is prevented by the longitudinal elements. The longitudinal elements m in im i ze the amount o f "deadspace" w i th in the bulbus when relaxed. They also decrease the space in wh i ch f lu id can recirculate by their posit ion along the wa l l dur ing systole. In an autoradiographic study on the c i rculat ion o f the lungfish, Johansen and H o i (1968) 179 witnessed the b lood leav ing the trabeculated heart in distinct laminar flow channels. The longitudinal elements o f the lungf ish lumen help " g u i de " the b lood through the bulbus wh i l e mainta in ing laminar f low. In fish that are l i ke ly to encounter recirculat ion problems w i th in the bulbus (ie. fish w i th a large cardiac output), the longitudinal elements wou l d aid in gu id ing the flow channels through the bulbus and into the ventral aorta. 6. Things to Do Despite the many answers this study has del ivered, as many questions have been raised. The largest one relates to the bulbar elastin. Structurally, it appears to be unique. Its lack o f an amorphous component (Yamauch i , 1980; Seraf in i -Fracass in i et al, 1978; Sp ina et al., 1979; I sokawa et al., 1990) and its resulting fibrillar morphology deserve a much closer inspection. Chemica l l y , it is special. It stains differently, both in co lour and density than expected and is soluble in both acids and bases. The special chemica l and mechanica l properties are l ike ly due to its special amino acid content. Compared w i th mammal ian elastin, ra inbow trout bulbar elastin is enriched in hydroph i l i c amino acids and g lyc ine, depleted in alanine and val ine and has a lower cross- l ink ing content (Sage, 1982), reducing its hydrophobic ity and resulting in a decreased modulus (Seraf in i -Fracass in i et al., 1978). These features make it chemica l l y distinct, not just f r om the elastin found in other genera (Sage, 1982) but also f rom elastin w i th in the ventral aorta (L icht and Harr i s , 1973). Th i s is unusual because most animals have a single elastin gene. Th i s begs the question: is bulbar elastin genetical ly distinct f r om other fish elastin or are the differences the result o f processing? E i ther poss ibi l i ty is as l ike ly. Post transcriptional 180 processing occurs on a variety o f different molecules in the cel l , wh i l e having fami l ies o f structural proteins is also quite common (ie. col lagen, C o x et al, 1984; silk, Guerette et al, 1996; and proteoglycans, Sandell, 1984). What is required to answer some o f these questions is a molecular analysis: mak ing a c D N A l ibrary f rom the bulbus, us ing elastin probes to pul l out any homologous sequences, P C R to increase the amount o f material f o l l owed by sequencing to establish the gene identity. Once the genetic code is complete, a compar ison w i th the amino acid content o f the bulbar elastin wou ld help answer some o f the questions. 7. Summary Teleost are able to maintain f l ows and depulsate f lows. Contrary to the situation in mammals and birds (Spencer and Greiss, 1962; M c D o n a l d , 1974), f l o w in the prox imal aorta o f teleosts is almost always posit ive and s ignif icant ly depulsated (Bushnel l el al, 1992). These effects on pressure and f l ow occur despite the fact that f ish lack a long arterial tree separating the heart f rom the distal resistance. The answer to this haemodynamic mystery is the bulbus arteriosus. In teleosts, a large proportion o f the compl iance in the cardiovascular system has been loca l i zed w i th in a single short bulb. Hav i n g a large central compl iance lowers the tension-t ime integral o f the heart, increasing its eff ic iency. The large prox imal compl iance also buffers the heart f rom the large distal impedance o f the gi l ls. Th i s design has been recapitulated in d i v ing mammals l i ke whales and seals that shut down their lungs and reorganize their c i rculat ion dur ing long dives, result ing in a large distal impedance. Seals create a large central compl iance by changing the geometry 181 o f the aortic arch, fo rming an aortic bulb (Shadwick and Gos l ine, 1995). In the more h igh ly der ived whales, not only have geometric changes occurred in the p rox ima l aorta, but changes in the material properties o f the arch have also occurred, lower ing the modulus and increasing the compliance. However , it is in the teleosts that this design achieves its apogee. The bulbus has a specia l ized and complex morphology that al lows all bulbar sections, despite differences in size and thickness, to inflate in the same manner. N o t only is the morpho logy o f the bulbus special ized, it also possesses a unique elastin. Bu l ba r elastin is structurally and chemica l l y different than other elastins, resulting in a very l o w modulus o f elasticity. The combinat ion o f the special ized bulbar morphology and chemistry results in a very special pressure-volume relationships. W h i l e arteries have J-shaped P - V loops (Figure 67), the bulbus, despite being composed o f the same materials, has an r-J shaped curve (F igure 66B ) w i th a steep in it ia l rise, a compl iant plateau phase and another steep rise. The J-shaped P -V loop o f arteries can be d iv ided into two regions. The l ow in it ia l slope is due to the rubber- l ike elastin, wh i le the steep f inal slope is due to the high stiffness o f col lagen. The P -V loop o f the bulbus can be s imi lar ly d iv ided. The high init ial slope is due to the geometry o f the lumen. The interactions between radius, pressure and tension require large in it ia l pressure gradients for f l ow through the bulbus to occur. The second section o f the bulbus inf lat ion, the plateau, occurs because o f the extensive modi f icat ions seen in the bulbus wa l l . The large e las t inxo l lagen ratio, l imited amount o f col lagen arranged cirumferential ly, the lack o f lamellae, and the l ow hydrophobic i ty o f the elastin itself, al l combine to lower stiffness, increase extensibi l i ty and a l l ow eff ic ient 1 8 2 recoi l . The third port ion o f the bulbus inf lat ion, the f ina l rise in stiffness, only occurs under extraordinary circumstances. W h i l e the modulus o f bulbus material is much lower than in an artery, at large volumes, the wa l l material is shifted to the steep port ion o f the J-shaped modulus-strain curve, and, combined w i th the decreasing thickness o f the wa l l , the overal l stiffness o f the bulbus increases rapidly. The properties o f the bulbus in f lat ion a l l ow it to be an exceptional pressure and vo lume reservoir. The compl iance o f the bulbus on the plateau a l lows it to hold and return large volumes. Once reaching mean systol ic pressure, smal l changes in pressure cause large vo lume changes as the bulbus expands to hold b lood during systole and contracts to maintain f l o w dur ing diastole. M o s t teleosts show a great deal o f var iab i l i ty in their stroke volume. This situation should result in large fluctuations in pressure and resistance, but the compl iance o f the bulbus buffers the cardiovascular system, attenuating and damping the pulses. The pressure-volume dynamics o f the bulbus also a l l ow it to wo r k exceedingly we l l as a pressure reservoir. Smal l stroke vo lume or large, the sharp in it ia l rise o f the inf lat ion brings the bulbus up to "operat ing pressure". The compl iant plateau phase o f the inf lat ion also a l lows the bulbus to "store" pressure. A s F igure 67 demonstrates, the bulbus maintains a consistent pressure over a large vo lume increase dur ing systole and a large vo lume decrease dur ing diastole. The bulbus has a large effect on the rest o f the cardiovascular system through its pos it ion and mechanical properties. However , it may have a more active role, releasing vasoactive substances in response to changes in f low, pressure and resistance. Whether or not the bulbus transduces forces w i th in the cardiovascular system, its abi l i ty to respond to 183 neural, humora l and hormonal st imul i is fact (Farrel l, 1979; Watson and Cobb, 1979; A c i e r no et al, 1991; Siharath et al, 1995; Cer ra et al, 1997). The variety o f different vectors by w h i c h the bulbus can be mod i f ied shows that it is not merely a s imple Windkesse l , passively smoothing f lows and pressures. Rather, the bulbus is a dynamic vessel, capable o f changing its dimensions and mechanical properties f r om beat to beat, and affect ing the entire cardiovascular system. 184 REFERENCES Ac ie rno , R., Axe l s son, M . , Tota, B. and N i l s son, S. (1991). Hypotens ive effect o f atrial natriuretic factor ( A N F ) in the atlantic cod, Gadus morhua. Comp. Biochem. Physiol. 99C, 11-14. Benjamin , M . , Norman, D., Santer, R.M., and Scarborough, D. (1983). H i s to log ica l , h i s tochemical and ultrastructural studies on the bulbus arteriosus o f the st ick le-backs, Gasterosteus aculeatus and Pungitiuspungitius (Pisces: Teleostei)../. Zoo., Lond. 200, 325-346. Benjamin , M . , No rman, D., Scarborough, D. and Santer, R. M . (1984). Carbohydrate-contain ing endothel ial cel ls l i n ing the bulbus arteriosus o f teleosts and conus arteriosus o f elasmobranches (Pisces). J. ZooL, Loud. 202, 383-392. Benjamin , M . (1985). A dissecting aneurysm in the bulbus arteriosus o f a brook st ickleback, Cu laea inconstans.,/ Comp. Pathol. 95(1), 37-43. Berge l , D. H. (1961). The static elastic properties o f the arterial wa l l . .7. Physiol., Lond. 156, 445-457. B r i l l , R. W . (1987). O n the standard metabol ic rates o f tropical tunas, inc lud ing the effect o f body size and acute temperature change. Fishery Bulletin. 85(1), 25-35. B r o w n , R. E., Butler, J. P., Rogers, R. A. and Le i th , D. E. (1994). Mechan i ca l connections between elastin and collagen. Connective Tissue Research 30, 295-308. Bur ton, A. C. (1954). Re lat ion o f structure to funct ion o f the tissues o f the wa l l o f b lood vessels. Am. J. Physiol. 34(4), 619-641. Bushne l l , P.G., Jones, D.R., and Farre l l , A.P. (1992). The arterial system. In: Fish Physiology, vo l . 1 2A ( ed . W.S. Hoar, D.J. Randal l , and A.P. Farrel l ) , pp.89-120. A cadem i c Press, N e w York . Bushne l l , P. G. and B r i l l , R. W. (1992). Oxygen transport and cardiovascular responses in sk ipjack tuna (Katsuwonuspelamis) and ye l l owf i n tuna (Thunnus alhacares) exposed to acute hpoxia../. Comp. Physiol. B 162, 131-143. Campbe l l , K B . , Rhode, E.A., Cox, R.H. Hunter, W.C. and Noorderg raa f A. (1981). Funct iona l consequences o f expanded aortic bulb: a model study. Am. J. Physiol. 240, R200-R210. Cas ley-Smith, J.R. and Ch in , J.C. (1971) The passage o f cytoplasmic vesicles across endothelial and mesothelial cells. Journal of Microscopy. 93(3), 167-189. 185 Cenacch i , G., Gu iducc i , G., Paquine l l i , G., Garg iu lo, M . , Degani , A., Stella, A., D ' A d d a t o , M . , Spina, M . and Mar t i ne l l i , G. N. (1995). The morpho logy o f elastin in non-speci f ic and inf lammatory abdominal aortic aneurysms: A comparat ive transmission, scanning and immunoelectronmicroscopy study../. Submicrosc. Cytol. Pathol. 27(1), 75-81. Cerra, M . C , Canonaco, M and Tota, B (1992). A quantitative autoradiographic study o f 1 2 51 atrial natriuretic factor in the heart o f a teleost fish (Conger conger). J. Exp. Zool. 263, 215-219. Cerra, M . C , Canonoco, M . , Take i , Y . and Tota, B. (1996). Character izat ion o f natriuretic peptide b ind ing sites in the heart o f the eel, Anguilla anguilla. J. Exp. Zool. 275, 27-35. Cerra, M . C , Canonaco, M . , Ac ierno, R. and Tota, B. (1997). D i f ferent b ind ing activit ies o f A - and B - type natriuretic hormones in the heart o f two Antarct ic teleosts, the red-blooded Trematomus bernacchii and the hemoglobinless Chiondraco hamatus. Comp. Biochem. Physiol. 118A(4), 993-999. Chow, M . , B o y d , C. D., Iruela-Arispe, M . , Wrenn, D. S., Mecham, R. and Sage, E. H . (1989). Character izat ion o f elastin protein and m R N A fo rm sa lmonid fish (Oncorynchus kisutch). Comp. Bioch. Physiol. 93B (4), 835-845. Clark, J. M . and Glagov, S. (1985). Transmural organizat ion o f the arterial media: the lamel lar unit revisited. Arteriosclerosis. 5, 19-34. Clark, R. J. and Rodn ick , K. 1. (1999). Pressure and vo lume overloads are associated with ventr icular hypertrophy in male ra inbow trout. Am. J. Physiol. 277(46), R 9 3 8 -R946. Cleary, E. G. and G ibson, M . A. (1983). Elast in-associated microf ib r i l s and mic ro f ib r i l l a r proteins. International Review of Connective Tissue Research. 10, 97-209. Con lon , J.M., Platzack,B., Marra, L.E., Youson, J.H. and Olson, K.R. (1995). Isolation and b io log ica l act iv ity o f (Trp5)bradykin in f rom the p lasma o f phylogenet ica l ly ancient fish, the bowf in and the longnosed gar. Peptides. 16(3), 485-489. Con lon , J .M. (1999). B radyk i n i n and its receptors in non-mammal ian vertebrates. Regulatory Peptides. 79(2-3), 71-81. Cox , G. N., Kramer, J. M . and Hi rsh, D. (1984). Number and organizat ion o f col lagen genes in Caenorhabditis elegans. Molecular and Cellular Biology. 4(11) 2389-2395. 186 Dav ie , P. S. and Daxboeck, C. (1984). Anatomy and adrenergic pharmacology o f the coronary vascular bed o f Pac i f i c blue mar l in (Makaira nigricans). Can. J. Zool. 62(9), 1886-1888. D e M o n t , M . E. and Wright, G. M . (1993) E last ic arteries in a pr imi t i ve vertebrate: mechanics o f the lamprey ventral aorta. Experientia. 49, 43-46. Dob r i n , P. B. (1978). Mechan i ca l properties o f arteries. Physiological reviews. 58(2), 397-460. Farre l l , A.P. (1979). The windkessel effect o f the bulbus arteriosus in trout. J. Exp. Zool. 209, 169-173. Fung, Y . C . (1981). Biomechanics: Mechanical Properties of Living Tissues. Springer-Ver lag. N e w York . Gos l ine, J. M . and Shadwick, R. E. (1996). The mechanical properties o f f in wha le arteries are explained by novel connective tissue designs. J. Exp. Biol. 199, 985-997. Gibbons, C. A . and Shadwick, R. E. (1991). C i rcu latory mechanics in the toad Bufo marinus. II. Haemodynamics o f the arterial windkessel../. Exp. Biol. 158, 291 -306. Greer Walker , M . , Santer, R. M . , Benjamin, M . and No rman, D. (1985). Heart structure o f some deep-sea f ish (Teleostei-Macrouridae).,/. Zool, Lond. 205, 75-89. Guerette, P. A., Ginz inger, D. G., Weber, B. H. and Gos l ine, J. M . (1996). Science. 275(5258), 112-115. Hamlett , W. C , Schwartz, F. J., Schmeinda, R. and Cuevas, E. (1996). Anatomy, histology, and development o f the cardiac va lvu lar system in elasmobranches../. Exp. Zool. 275(2-3), 83-94. Harvey, 1974. Theory and Design of Modern Pressure Vessels. V a n Nostrand Re inho ld, N e w Yo rk . He ide l , J. R., LaPatra, S. E. and Gi les , J. (1997). M inera l i za t ion o f the bulbus arteriosus in adult ra inbow trout (Oncorhynchus mykiss). Journal of Veterinary Diagnostic Investigation. 9(2), 213-216. Ho le ton , G. F. and Jones, D. R. (1975). Water f l ow dynamics in the respiratory tract o f the carp (Cyprinus carpio L ) . J. Exp. Biol. 63, 537-549. 187 Hou l i han , D. F., Agn i so la , C , Lyndon, A. R., Gray, C. and Hami l ton , N. M . (1988). Prote in synthesis in a fish heart: responses to increased power output. J. Exp. Biol. 137, 565-587. Hughes, G. M . and Shelton, G. (1958). The mechanism o f g i l l vent i lat ion in three freshwater teleosts. Icardo, J. M . , Co lvee, E. and Tota, B. (1999a). Bu lbus arteriosus o f the Antarct ic teleosts. I. The white-b looded Chionodraco hamatus. Anatomical Record. 254(3), 396-407. Icardo, J. M . , Co lvee, E., Cerra, M . C. and Tota, B. (1999b). Bu lbus arteriosus o f the Antarct ic teleosts. II. The red-blooded Trematomus bernacchii. Anatomical Record. 256(2), 116-126. Icardo, J. M . , Colvee, E., Cerra, M . C. and Tota, B. (2000). L i gh t and electron microscopy o f the bulbus arteriosus o f the European eel {Anguilla anguilla). Cells Tissues Organs. 167(2-3), 184-198. Isokawa, K., Takagi , M , and Toda Y. (1988). Ultrastructural cytochemistry o f trout arterial f ibers as elastic components. The Anatomical Record. 220, 369-375. I sokawa, K., Takagi , M . , and Toda, Y . (1990). Ultrastructural and cytochemica l study o f elastic fibers in the ventral aorta o f a teleost, Anguilla japonica. The Anatomical Record. 226, 18-26 Jain, K.K. (1963). Mechan i sm o f rupture o f intracranial saccular aneurysms. Surgery. 347-350. Jensen, J. and Ho lmgren , S. (1991). Tachyk in ins and intestinal mot i l i ty in different f ish groups. General and Comparative Endocrinology. 83(3), 388-396. Johansen, K. (1962). Card iac output and pulsatile aortic f l ow in the teleost, Gadus morhua. Comp. Biochem. Physiol. 7, 169-174. Johansen, K. and H o i , R. (1968). A radiological study o f the central c i rcu lat ion in the lungfish, Protopterus aethiopicus../. Morph. 126, 333-348. Jones, D. R., Lang i l le , B. L., Randal l , D. J. and Shelton, G. (1974). B l o o d f l o w in dorsal and ventral aortas o f the cod, Gadus morhua. Am. J. Phys. 226(1), 90-95. Jones, D.R. (1991). Card iac energetics and the design o f vertebrate arterial systems. In: Efficiency and Economy in Animal Physiology (ed. R W . B lake) , pp. 159-168. Cambr idge Un iver s i ty Press, Cambridge, England. 188 Jones, D R . , B r i l l , R.W., and Bushne l l , P.G. (1993) Ventr icu lar and arterial dynamics o f anaesthetised and sw imming tuna../. exp. Biol. 182, 97-112. Kags t rom, J. and Ho lmgren, S. (1998). Ca lc i ton in gene-related peptide ( C G R P ) , but not tachykinins, causes relaxation o f smal l arteries f rom the ra inbow trout gut. Peptides. 19(3), 577-584. Ka rnov s k y M . J . and Shea, S.M. (1970). Transcapi l lary transport by pinocytosis. Microvascular Research. 2(4), 353-360 K ie rnan , J. A . (1998). Histological and histochemical methods: theory and practice. Butterworth-Heinemann, Oxford . K i m , S. Z , Cho, K. W., K i m , S. H , Ryu, H , Hwang, Y . H. and Lee, G. Y . (1991). Identif icat ion o f immunoreact ive atrial natriuretic peptide in bulbus arteriosus o f freshwater teleostean fish. Comp. Biochem. Physiol. 100B (3), 575-578. Kubasch , A . and Rourke, A. W. (1990). Arter ioscleros is in steelhead trout, Oncorhynchus mykiss (Walbaum): a developmental analysis. J. Fish Biol. 37, 65-69. L a i , N . C , Graham, J. B., L o w e l l , W. R. and Laurs, R. M . (1987). Per icard ia l and vascular pressures and b lood f l ow in the albacore tuna, Thunnus alalunga. Exp. Biol. 46, 187-192. Leknes, 1. L. (1980). Ultrastructure o f atrial endocardium and myocard ium in three species o f Gadidae (Teleostei). Cell Tissue Res. 210, 1-10. Leknes, I. L. (1981). O n the ultrastructure o f the endothel ium in the bulbus arteriosus o f three teleostean species. J. Suhmicrosc. Cytol. 13(1), 41-46. Leknes, I. L. (1985). A scanning electron microscopic study on the heart and ventral aorta in a teleost. Zool Am. Jena 214, 142-150. Leknes, I. L. (1986). Ultrastructural and ultrahistochemical studies on the ventral aorta in larvae o f a teleost, Poecilia reticulata. Anat. Am. Jena 161, 43-51. L i ch t , J. H . and Harr is, W. S. (1973). The structure, compos i t ion and elastic properties o f the teleost bulbus arteriosus in the carp, Cyprinius carpio. Comp. Biocem. Physiol. 46A, 699-70. L i l l i e , M.A., Chalmers, G .W.G. and Gos l ine, J. M . (1994). The effects o f heating on the mechanica l properties o f arterial elastin. Connective Tissue Research. 31(1), 23-35. M c D o n a l d , D. A. (1974). Blood Flow in Arteries, second edition. The W i l l i a m s and W i l k i n s Company, Ba l t imore. 189 Metca l fe , J.D. and Butler, P.J. (1982). Differences between directly measured and calculated values for cardiac output in the dogfish: a cr i t ic i sm o f the F i c k method. ./. Exp. Biol. 99, 255-268. M i l n o r , W.R. (1982). Hemodynamics. The W i l l i a m s and W i l k i n s Company, Ba l t imore. Mo t t , J. C , (1950). Rad io log i ca l observations on the cardiovascular system in Anguilla anguilla. J. Exp. Biol. 27, 324-333. Olson, K.R., Conk l i n , D.J., Weaver, L . Jr., Duff, D.W., Herman, C.A., Wang , X . and Con lon , J M . (1997). Cardiovascular effects o f homologous bradyk in in in ra inbow trout. Am. J. Physiol. 272(4 Pt 2), R l 112-1120. Onor i , L., Agg io , A. and Taddei, G. (1999). Regulatory role o f tachyk in ins on intestinal propuls ive activity. Italian Journal of Gastroenterology & Hepatology. 31(4), 346-351. Oudit, G. Y . and Butler, D. G. (1995). Cardiovascular effects o f arginine vasotocin, atrial natriuretic peptide, and epinephrine in freshwater eels. Am. J. Physiol. 268, R1273-R1280. Platzack, B. and Con lon , J .M. (1997). Pur i f icat ion, structural characterization, and cardiovascular activ ity o f cod bradykinins. Am. J. Physiol. 272(2 Pt2), R710-717. Pr iede, I. G. (1976). Funct ional morphology o f the bulbus arteriosus o f ra inbow trout (Salmo gairdnen1 Richardson)../. Fish Biol. 9, 209-216. Randa l l , D.J., Smith, L.S. and Brett, J R . (1965). Dorsa l aortic b lood pressures recorded f rom the ra inbow trout (Salmo gairdneri). Can. J. Zool. 43, 863-871. Randa l l , D. J. (1968). Funct ional morphology o f the heart in fishes. Am. Zoologist. 8, 179-189. Raso, D. S. (1993). Funct ional morphology o f lamin in, col lagen type IV, col lagen bundles, elastin, proteoglycans in the bulbus arteriosus o f the white bass, Morone chrysops (Rafinesque). Can. J. Zool. 71(5), 947-952. Roach , M . C. and Burton, A. C. (1957). The reason for the shape o f the distensibi l i ty curves o f arteries. Can. J. Biochem. Physiol. 35, 181-190. Rose, C , Kumar , M . , Manda l , A. B. (1988). A study o f the hydrat ion and thermodynamics o f warm-water and cold-water fish collagens. Bioch. J. 249(1), 127-133. Rucker, R. B. and Tinker, D. (1977). Structure and metabol ism o f arterial elastin. International Review of Experimental Pathology. 17, 1-47. 190 Sage, E. H. and Gray, W. R. (1977). Evo lu t i on o f elastin structure. Advances in Experimental Medicine and Biology. 79, 291-312. Sage, H. (1982). Structure-function relationships in the evolut ion o f elastin../. Exp. Derm. 79, 146s-153s. Sanchez-Quintana, D. and Hur le , J. M . (1987). Ventr icu lar myocard ia l architecture in marine fishes. Anat. Rec. 217, 263-273. Sanchez-Quintana, D., Garc ia -Mart inez, V., C l iment, V . and Hur le , J. M . (1996). M y o c a r d i a l f ibre and connective tissue architecture in the f ish heart ventricle. J. Exp. Zool. 257, 112-124. Sandell, L. J. (1984). In situ expression o f col lagen and proteoglycan genes in notochord and dur ing skeletal development and growth. Microscopy Research and Technique 28(6), 470-482. Santer, R. M . (1985). Mo rpho l ogy and innervation o f the f ish heart. Adv. Anal. Embryol. Cell Biol. 89, 1-102. Santer, R. M . and Cobb, J. L. S. (1972). The fine structure o f the heart o f the teleost (Pleuronect.es platessa L ) . Z. Zellforsch. 131, 1-14. Satchell, G. H. and Jones, D. R. (1967). The funct ion o f the conus arteriosus in the Port Jackson shark, Heterodontus portusjacksoni. J. Exp. Biol. 46, 373-382. Satchell, 1971. Circulation in Fishes. A t the Un ives i t y Press, Cambr idge Satchell, G. H. (1991). Physiology and Form of Fish Circulation. Cambr idge Un iver s i t y Press, Cambr idge, England. Scherer, P.W. (1973). F l o w in ax isymmetr ica l glass model aneurysm. .7. Biomechanics. 6, 695-700. Schmidt -N ie l son, K. (1989). Animal Physiology: Adaption and Environment. 3'd Ed., Cambr idge Un ivers i ty Press, Cambridge. Seraf ini-Fracass ini, A., F i e ld , J. M . , Spina, J., Garbia, S. and Stuart, R.J. (1978). The morpho log ica l organizat ion and ultrastructure o f elastin in the arterial wa l l o f trout (Salmo gairdneri) and salmon (Salmo sa/ar). J. Ultras/rue. Res. 65, 1-12. Shadwick, R. E. and Gos l ine, J. M . (1985). Mechan i ca l properties o f the octopus aorta. ./ Exp. Biol. 114, 259-284. Shadwick, R. E., Gosl ine, J. M . , and M i l s o m , W. K. (1987). Arter ia l haemodynamics in the cephalopod mol lusc, Octopus dofleini. J. Exp. Biol. 130, 87-106. 191 Shadwick, R. E. and N i l s son , E. K. (1990). The importance o f vascular elasticity in the c i rculatory system o f the cephalopod Octopus vulgaris. J. Exp. Biol. 152, 4 71 -484. Shadwick, R. E. and Gos l ine, J. M . (1994). Arter ia l mechanics in the fin wha le suggest a unique hemodynamic design. Am. J. Physiol. 267, R805-R818. Shadwick, R. E. and Gos l ine, J. M . (1995). Arter ia l windkessels in marine mammals. J. Exp. Biol, 243-252. Siharath, K., N i sh ioka , R.S. and Bern, H.A. (1995). In vitro product ion o f IGF -b i nd i ng proteins ( I G F B P ) by various organs o f the striped bass, Morone saxatalis. Aquaculture. 135(1-3), 195-203. Spina, M . , Garbisa, S., F ie ld , J. M . and Seraf ini-Fracass ini, A. (1979). The salmonid elastic fibril: A n investigation o f some chemical and phys ica l parameters. Archives of Biochemistry and Biophysics. 192(2), 430-437. Spencer, M .P . and Greiss, F.C. (1962). Dynamic s o f ventr icular ejection. Circulat. Res. 10, 274. Stevens, E. D. and Randa l l , D. J. (1967). Changes in b lood pressure, heart rate and breathing rate during moderate sw imming act iv ity in ra inbow trout. J. Exp. Biol. 46,307-315. Stevens, E. D , Benn ion, G. R., Randal l , D. J., Shelton, G. (1972). Factors affecting arterial pressures and b lood f l ow f rom the heart in intact, unrestrained l ingcod, Ophiodon elongatus. Comp. Biochem. Physiol 43A, 681-695. Thubr ikar, M , No lan , S.P., Bosher, L P . and Deck, J.D. (1980). The cyc l i c changes and structure o f the base o f the aortic valve. American Heart Journal. 99(2), 217-224. von Skraml ick, E. (1935). Uber der kreis lauf bei den fischen. Erghn. Biol. 11, 1-30. Wa inwr ight , S.A., Biggs, W.D., Currey, J.D., and Gos l ine, J .M. (1976). Mechanical. Design in Organisms. Pr inceton Un iver s i ty Press, N e w Jersey. Watson, A .D. and Cobb, J.L.S. (1979). A comparative study on the innervation and vascular izat ion o f the bulbus arteriosus in the teleost fish. Cell Tissue Research. 196, 337-346. Waugh, D., Graff, K.E., Platzack, B., Youson, J.H., Olson, K.R. and Con lon , J .M. (1995). Isolation, local izat ion, and cardiovascular activ ity o f tachykin ins f r om the stomach o f the bowf i n Amia calva. Am. .1. Physiol. 269(3 P t 2), R565 -R571 . 192 Woodhead, A .D. (1984). A g i n g changes in the heart o f a poec i l i i d fish, the guppy Poecilia reticulates. Experimental Gerontology. 19, 383-391. Wo l i n s k y , H . and Glagov, S. (1967). The lamel lar unit o f aortic media l structure and funct ion in mammals. Circulation Res. 20, 99-111. Wr ight , G. M . (1984). Structure o f the conus arteriosus and ventral aorta in the sea lamprey, Petromyzon marinus, and the At lant ic hagfish, Myxine glutinosa: microf ibr i l s , a major component. Can../. Zool. 62(12), 2445-2456. Yamauch i , 1980. F ine structure o f the fish heart. In: Hearts and Heart-Like Organs, vo l . 1. (ed. G. H. Bourke) , pp. 119-144. Academic Press, N e w Yo rk . 193 

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.831.1-0089971/manifest

Comment

Related Items