UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of midbrain tegmentum in the coordination of episodic breathing in carp (Cyprinus carpio) and… O’Neill, Angela Elizabeth 2005

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

Item Metadata

Download

Media
831-ubc_2005-0292.pdf [ 15.95MB ]
Metadata
JSON: 831-1.0091995.json
JSON-LD: 831-1.0091995-ld.json
RDF/XML (Pretty): 831-1.0091995-rdf.xml
RDF/JSON: 831-1.0091995-rdf.json
Turtle: 831-1.0091995-turtle.txt
N-Triples: 831-1.0091995-rdf-ntriples.txt
Original Record: 831-1.0091995-source.json
Full Text
831-1.0091995-fulltext.txt
Citation
831-1.0091995.ris

Full Text

The Role of the Midbrain Tegmentum in the Coordination of Episodic Breathing in Carp (Cyprinus carpio) and Rainbow Trout (Oncorhynchus my kiss) By Angela Elizabeth O'Neill B.Sc, The University of British Columbia, 2001 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (ZOOLOGY) THE UNIVERSITY OF BRITISH COLUMBIA April 2005 © Angela Elizabeth O'Neill, 2005 Abstract Because respiratory needs m a y vary due to act ivi ty leve l or oxygen avai lab i l i ty , vertebrates must adjust their breathing to meet these changing levels o f respiratory dr ive. In vertebrates, the basic respiratory rhy thm is generated i n the medul la ; however , this rhy thm is modi f i ed b y both sensory feedback and input f rom higher centres to produce the broad range o f vertebrate breathing patterns. In one such pattern, c o m m o n l y produced dur ing periods o f reduced respiratory dr ive by some species i n every vertebrate class, breaths are organized into groups (episodes) separated by periods wi thout breathing (apneas). In carp, a site i n the dorsal mesencephal ic tegmentum, just ventrolateral to the oculomotor nucleus, appears to terminate apneas by in i t ia t ing breathing episodes (Juch and B a l l i n t i j n , 1983). In order to test whether this site is necessary for the product ion o f episodic breathing, I les ioned this midbra in tegmental site i n decerebrate/spinalectomized carp (Cyprinus carpio) and trout (Oncorhynchus mykiss) us ing stereotaxic microinject ions o f O . O l m M ka in ic ac id . C a r p normal ly breathe ep isodica l ly i n n o r m o x i a and hyperoxia , w h i l e trout normal ly breathe cont inuously , on ly breathing ep isod ica l ly i n extreme hyperoxia . Decerebrate/spinalectomized carp and trout breathed normal ly compared to intact f i sh i n n o r m o x i a and hyperoxia , but not i n severe hypox ia . In i t ia l ly , the ka in i c ac id p rov ided a tonic st imulus w h i c h exci ted breathing; however , this st imulatory effect dissipated after approximately ninety minutes, when the ka in ic ac id started k i l l i n g neurons. In both carp and trout, l es ion ing this midbra in tegmental site altered the breathing pattern, increasing breathing frequency w h i l e decreasing ampli tude such that total vent i la t ion remained unchanged. T h e ka in ic ac id microinject ions e l iminated apneas i n fifty percent o f carp (seven o f fourteen) and reduced the occurrence o f apneas i n a further twenty-nine percent (four o f fourteen). H i s t o l o g i c a l analysis tentatively suggested that apneas were e l iminated o n l y i n carp for w h i c h the ka in ic ac id les ioned the midbra in tegmental site. O n l y four o f nine trout breathed episodica l ly , even i n hyperoxia , and o f these four trout, ka in ic ac id e l iminated apneas i n one trout and reduced the occurrence o f apneas i n two others. T h i s midbra in tegmental site does not no rmal ly influence the total l eve l o f respiratory dr ive , but does regulate breathing pattern, decreasing frequency and increasing ampli tude. T h i s site also appears to regulate episodic breathing i n carp and trout; however , it is s t i l l unclear whether this site is essential to the product ion o f this episodic pattern. i i Table of Contents Abstract i i Tab le o f Contents i i i L i s t o f Tables v i i L i s t o f Figures v i i i L i s t o f Abbrev ia t ions x i i i A c l m o w l e d g m e n t s x v Chapter 1: Introduction 1 Teleost Respi ra t ion: A n Introduction 1 Phases o f Respi ra t ion i n Teleost F i s h 1 M u s c l e s o f Respi ra t ion i n the Trou t 2 Innervation o f the Respiratory M u s c l e s 5 L o c a t i o n and Discharge Patterns o f Respiratory Neurons i n the Teleost M e d u l l a 6 Respiratory Rhy thms Produced i n the Isolated Teleost M e d u l l a 6 L o c a t i o n o f respiratory neurons i n the teleost medu l l a 7 Discharge Patterns o f Respiratory Neurons 11 Respiratory R h y t h m Generators 12 Prev ious ly proposed sites for rhy thm generators 12 M u l t i p l e Respi ra tory R h y t h m Generators i n the E a r l y Deve lopmenta l Stages o f Vertebrates 13 M u l t i p l e Respira tory R h y t h m Generators i n Other Vertebrates 15 Exper imenta l Ev idence for M u l t i p l e R h y t h m Generators i n Teleosts 18 T h e Quanti ty and L o c a t i o n o f Respi ra tory R h y t h m Generators i n the T r o u t M e d u l l a 19 Brea th ing Patterns in Teleosts , 19 i i i W h a t is Brea th ing Pattern ? 19 W h y Breathe E p i s o d i c a l l y ? 21 Brea th ing Patterns i n R a i n b o w Trout and C a r p 23 T h e Influence o f Per iphera l Feedback f rom C r a n i a l Nerves on Respi ra t ion 25 T h e R o l e o f the M i d b r a i n i n Brea th ing Pattern 26 Respiratory Input f rom H i g h e r Centres 26 Exper imenta l Ev idence for M i d b r a i n Involvement i n Brea th ing R h y t h m i n Teleosts 27 M i d b r a i n Involvement i n Brea th ing Pattern i n Other Vertebrates 28 T h e R o l e o f the Mesencepha l ic Tegmentum i n Brea th ing Pattern i n R a i n b o w Trout and Carp : Hypothes is 31 Chapter 2: D o Decerebrate/Spinalectomized F i s h Breathe N o r m a l l y ? 34 Introduction 34 Me thods 35 Exper imen ta l A n i m a l s 35 Instrumentation and Surgery 35 Decerebrat ion and Spina lec tomy 37 M e a s u r i n g Respiratory Signals 38 Exper imen ta l P ro toco l : Series 1 4 0 Exper imenta l P ro toco l : Series 2 41 Da ta A n a l y s i s 41 Resul ts 43 Series T. Respi ra t ion i n Intact, Decerebrate, and Decerebrate /Spinalectomized T rou t 43 Series 2: Brea th ing i n Decerebrate /Spinalectomized Trout and Carp i n N o r m o x i a and H y p e r o x i a 51 D i scus s ion 59 Intact Trout i n N o r m o x i a 59 Decerebrate Trout i n N o r m o x i a 59 Decerebrate /Spinalectomized Trou t i n N o r m o x i a 60 Decerebrate /Spinalectomized Carp i n N o r m o x i a 60 Intact Trou t i n H y p o x i a 61 Decerebrate Trout in H y p o x i a 62 i v Decerebrate /Spinalectomized Trout i n H y p o x i a 62 Intact Trout in H y p e r o x i a 63 Decerebrate Trout i n H y p e r o x i a 63 Decerebrate /Spinalectomized T rou t i n H y p e r o x i a 64 Decerebrate /Spinalectomized Carp i n H y p e r o x i a 64 Conc lu s ions 65 Chapter 3: T h e R o l e o f the J A B Site i n Coord ina t ing Respiratory Pattern 66 Introduction 66 Me thods 69 Instrumentation and Surgery 69 M i c r o i n j e c t i o n o f K a i n i c A c i d 71 M e a s u r i n g Respiratory Signals 75 Exper imenta l P ro toco l 76 H i s t o l o g y 77 D a t a A n a l y s i s 78 Results ._. 80 K a i n i c A c i d Mic ro in jec t ions 80 C a r p - Brea th ing Frequency 81 C a r p - Brea th ing A m p l i t u d e 82 Carp - To ta l Ven t i l a t ion 83 Carp - O v e r a l l Response 85 Trou t - Brea th ing Frequency 90 Trout - Brea th ing A m p l i t u d e 90 Trou t - To ta l Ven t i l a t ion 91 Trou t - O v e r a l l Response 93 H i s t o l o g y 98 D i s c u s s i o n 103 Cr i t ique o f Me thods : 103 T i m e Course o f K a i n i c A c i d Effects 106 Decerebrate /Spinalectomized Carp Breathe N o r m a l l y 107 v T h e Exc i ta to ry Phase: K a i n i c A c i d Mic ro in jec t ions i n Ca rp In i t ia l ly Stimulate Brea th ing 108 T h e T o x i c Phase: Effects i n Ca rp o f the K a i n i c A c i d Mic ro in jec t ions O v e r T i m e 108 L e s i o n i n g the I A B Site Affects Ep i sod i cBrea th ing i n Carp 110 Decerebrate/Spinalectomized Trout Breathe N o r m a l l y I l l T h e Exc i ta to ry Phase: K a i n i c A c i d Mic ro in jec t ions i n Trout In i t ia l ly Stimulate Brea th ing 112 T h e T o x i c Phase: Effects O v e r T i m e o f K a i n i c A c i d Mic ro in jec t ions i n Trout 113 K a i n i c A c i d Mic ro in jec t ions into the M i d b r a i n Tegmentum A p p e a r to Af fec t E p i s o d i c Brea th ing i n Trou t 114 Conc lus ions 114 References : .116 A p p e n d i x 124 v i List of Tables Table A . 1: Brea th ing ampli tude ( c m H 2 0 ) o f intact, decerebrate, and decerebrate/spinalectomized trout i n normoxia , h y p o x i a , and hyperox ia 124 Table A . 2 : Brea th ing frequency (breaths/min) o f decerebrate/spinalectomized carp before and after injection o f ka in ic ac id into the mesencephalic tegmentum, i n both normal and l o w levels o f respiratory dr ive 124 Table A . 3 : Brea th ing ampli tude ( cmHaO) o f decerebrate/spinalectomized carp before and after injection o f ka in ic ac id into the mesencephalic tegmentum, i n both normal and l o w levels o f respiratory dr ive 125 Table A . 4 : Brea th ing frequency (breaths/min) o f decerebrate/spinalectomized trout before and after inject ion o f ka in ic ac id into the mesencephal ic tegmentum, i n both normal and l o w levels o f respiratory dr ive 125 Tab le A . 5 : Brea th ing ampli tude ( c m H 2 0 ) o f decerebrate/spinalectomized trout before and after injection o f ka in ic ac id into the mesencephal ic tegmentum, i n both normal and l o w levels o f respiratory dr ive . 126 v i i List of Figures Figure 1.1: T h e phases o f the respiratory cyc le i n trout (Hughes and Shel ton, 1962) 2 F igure 1.2: A schematic d iagram o f the muscles o f the respiratory pump and their relat ion to the bones o f the palatal and opercular complexes i n the trout (Ba l l in t i jn and Hughes , 1965) 4 F igure 1.3: A sagittal v i e w o f the branchia l arches o f the carp, showing both the internal and external branchia l arch levator muscles (Ba l l in t i jn and Punt, 1985) 5 F igure 1.4: T h e locat ion o f respiratory neurons i n the teleost medul la , as w e l l as their major interconnections, projected onto the dorsal surface o f the bra in (Ba l l in t i jn , 1982) 9 F igure 1.5: L o c a t i o n o f several respiratory neuron populat ions i n the trout medul la , projected onto the surface o f a sagittal section (Bamford , 1974B) 10 Figure 1.6: A sagittal schematic o f the embryonic mouse h indbra in showing its d iv i s i on into eight rhombomeres (For t in et al., 2000) 14 F igure 1.7: Representative traces o f the episodic breathing pattern and the continuous breathing pattern i n the decerebrate/spinalectomized carp 20 F igure 1.8: Representative breathing traces f rom a decerebrate/spinalectomized trout, demonstrating that breathing frequency can be reduced from normal levels by either s l o w i n g the continuous breathing pattern or by organ iz ing breaths into episodes separated by apneas 21 F igure 1.9: Representative respiratory traces showing the spectrum o f breathing patterns, f rom continuous to episodic , i n the decerebrate/spinalectomized carp 24 v i i i Figure 1.10: A cross-section o f the carp midbra in , showing a bilateral site i n the mesencephal ic tegmentum (the J A B site) that appears to initiate episodes dur ing episodic breathing (Juch and Ba l l i n t i j n , 1983) 28 Figure 1.11: A representative f ict ive breathing trace, recorded f rom the root o f the vagus nerve o f an isolated bul l f rog (Rana catesbeiana) brainstem-spinal co rd , showing that transecting jus t caudal to the optic ch iasma converts the episodic f ict ive breathing pattern to a continuous one ( R e i d et al, 2000) 30 F igu re 2 .1: Photograph o f a trout instrumented w i t h impedance probes, an opercular cavi ty pressure cannula , and a bucca l cavi ty pressure cannula 36 F igure 2.2: Photograph o f a fu l ly instrumented, decerebrate/spinalectomized trout c l amped into the experimental tank w i t h the stereotaxic device and he ld i m m o b i l e by body sponges, w i th a secondary photograph o f the dorsal v i e w o f a decerebrate/spinalectomized trout brain 38 F igure 2.3: Brea th ing frequency (breaths/min) o f intact trout i n no rmox ia , hypox ia , and hyperox ia 44 F igure 2.4: Representative breathing traces o f an intact trout i n normoxia , hypox ia , and hype rox ia 45 F igure 2.5: Percent change i n breathing ampli tude f rom n o r m o x i a o f intact trout i n no rmoxia , hypox ia , and hyperox ia 46 F igure 2.6: Percent change i n total vent i lat ion f rom n o r m o x i a i n intact trout i n no rmoxia , h y p o x i a , and hyperox ia 47 F igure 2.7: Brea th ing frequency (breaths/min) o f intact, decerebrate, and decerebrate/spinalectomized trout i n no rmoxia , hypox ia , and hyperox ia 49 i x Figure 2.8: Percent change i n the breathing ampli tude o f decerebrate and decerebrate/spinalectomized trout i n no rmox ia , hypox ia , and hyperox ia f rom intact, no rmox ic trout 50 F igure 2.9: Percent change i n total vent i la t ion o f decerebrate and decerebrate/spinalectomized trout i n no rmoxia , hypox ia , and hyperox ia f rom intact, no rmox ic trout 51 F igure 2.10: Percent change i n breathing frequency, ampli tude, and total vent i lat ion i n decerebrate/spinalectomized trout i n n o r m o x i a and hyperox ia 53 F igure 2.11: Representative respiratory traces compar ing a decerebrate/spinalectomized trout breathing cont inuously i n n o r m o x i a but ep isodica l ly i n hyperox ia to a decerebrate/spinalectomized trout breathing cont inuously i n both n o r m o x i a and hype rox ia . . . .54 F igure 2.12: Percent change i n breathing frequency, ampli tude, and total vent i lat ion i n decerebrate/spinalectomized carp i n n o r m o x i a and hyperox ia 56 Figure 2.13: Representative respiratory traces compar ing continuous, episodic , and w a x i n g and wan ing breathing patterns i n decerebrate/spinalectomized carp 57 F igure 2.14: T h e percentage o f decerebrate/spinalectomized trout and carp breathing ep i sod ica l ly i n n o r m o x i a and hyperox ia 58 F igure 3.1: D i a g r a m o f a sagittal v i e w o f the trout bra in wi th the approximate transection sites o f the decerebration and spinalectomy ( M o d i f i e d f rom M e e k and Nieuwenhuys , 1997) 70 F igure 3.2: Photograph o f a decerebrate/spinalectomized carp, instrumented wi th a bucca l and an opercular cannula, i n the experimental tank wi th the stereotaxic device 70 F igure 3.3: Photograph o f the dorsal v i e w o f a decerebrate/spinalectomized carp bra in .71 Figure 3.4: T h e approximate locat ion o f the ka in ic ac id inject ion sites marked on the dorsal surface o f a decerebrate/spinalectomized carp bra in 72 F igure 3.5: T h e approximate loca t ion o f the ka in ic ac id injection sites marked on the dorsal surface o f a decerebrate/spinalectomized trout bra in 73 F igure 3.6: Photograph o f the microinjec t ion apparatus: inject ing ka in ic ac id f rom a four-barreled, glass micropipette into the midbra in o f a decerebrate/spinalectomized t rou t . . . . . . . . . 75 F igure 3.7: Representative breathing trace showing responses to ka in i c ac id microinject ions into three different sites i n the mesencephal ic tegmentum o f a decerebrate/spinalectomized carp. .82 F igure 3.8: Percent change in breathing frequency, ampli tude, and total vent i la t ion i n decerebrate/spinalectomized carp f rom control n o r m o x i a after sham and ka in ic ac id microinject ions into the mesencephal ic tegmentum i n both n o r m o x i a and hyperox ia 84 F igure 3.9: Percent o f decerebrate/spinalectomized carp that breathed ep isod ica l ly i n no rmox ia and hyperox ia both before and after microinject ions o f ka in ic ac id into the mesencephalic tegmentum 87 Figure 3.10: Representative breathing trace f rom a decerebrate/spinalectomized carp for w h i c h ka in ic ac id microinject ions into the mesencephal ic tegmentum el iminated the episodic breathing pattern 88 F igure 3.11: Percent change i n breathing frequency, ampli tude, and total vent i la t ion f rom control no rmox ia i n the E E and E R subsets o f decerebrate/spinalectomized carp 89 F igure 3.12: Percent change i n breathing frequency, ampli tude, and total vent i la t ion i n decerebrate/spinalectomized trout f rom control n o r m o x i a after sham and ka in ic ac id microinject ions into the mesencephalic tegmentum at both normal and l o w levels o f respiratory dr ive 92 x i Figure 3.13: Percent o f trout that breathed i n episodes i n no rmox ia and hype rox ia before and after microinject ions o f ka in ic ac id into the mesencephal ic tegmentum 95 F igure 3.14: Representative breathing trace f rom the on ly decerebrate/spinalectomized trout for w h i c h ka in ic ac id microinject ions complete ly e l iminated the episodic breathing pattern 96 F igure 3.15: Percent change i n breathing frequency, ampli tude, and total vent i lat ion f rom cont ro l n o r m o x i a i n the subset o f decerebrate/spinalectomized trout that on ly breathed cont inuously and the subset that breathed ep isod ica l ly 97 F igure 3.16: L o c a t i o n o f the ka in ic ac id inject ion sites relative to the J A B site i n trout 99 Figure 3.17: K a i n i c ac id microinject ions that d i d not les ion tissue w i t h i n the J A B site d i d not el iminate apneas i n a representative carp 100 F igure 3.18: K a i n i c ac id microinject ions that d i d les ion tissue i n the J A B site e l iminated apneas i n a representative carp 101 F igure 3.19: K a i n i c ac id microinject ions d i d not les ion tissue w i t h i n the J A B site and s l ight ly increased breathing frequency i n both representative trout 102 x i i List of Abbreviations A C S F - A r t i f i c i a l Cerebra l -Sp ina l F l u i d A d d . a.p.o. - muscularis adductor arcus palatini et operculi A d d . m d . - muscularis adductor mandibulae A M P A - ( ± ) - a - A m i n o - 3 - h y d r o x y - 5 - m e t h y l i s o x a z o l e - 4 - p r o p i o n i c ac id A N O V A - A n a l y s i s o f Va r i ance C A - commissure ansulate C a C l 2 - c a l c i u m chlor ide C b - corpus cerebe l lum C e r - cerebel lum C L - c le i thrum bone D . o . - muscularis dilator operculi G A B A - y- A m i n o butyric ac id H m d . - hyomand ibu la bone H y . - h y o i d bone H y . h y . - muscularis hyohyoideus III m - oculomotor nucleus (also n O C ) I X - glossopharyngeal nerve DC m - glossopharyngeal motor nucleus J A B site - dorsal mesencephalic tegmental site described by Juch and B a l l i n t i j n (1983) i n carp K C 1 - potassium chlor ide KH2PO4 - potassium dihydrogen orthophosphate L C - lobus caudalis cerebel l i L e v . h.a.p. - muscularis muscularis levator hyomandibulae et arcus palatini L . j . - l ower j a w bone L L - lemniscus lateralis N a C l - sod ium chlor ide N a 2 H P 0 4 - sod ium phosphate n G - nucleus glomerulosus x i i i n L V - nucleus lateralis va lvulae N M D A - N-methyl-D-aspartate n O C - oculomotor nucleus (also III m) O p . - operculum O p L - optic lobe P - muscle propriocept ive input Pal .pt - palato-pterygoid bone P B S - phosphate buffered solut ion PO2 - part ial pressure o f d isso lved oxygen P.hy . - muscularis protractor hyoidei Q u . - quadrate bone R f - ret icular formation R L - recessus lateralis ( infundibul i) Sth. - s ty lohyal bone Sthy. - muscularis sternohyoideus t - tegmental respiratory neurons T O - tectum op t i cum V - t r igeminal nerve V C b - v a l v u l a cerebel l i V d - descending t r igeminal nucleus V m - t r igeminal motor nucleus V e 4 - fourth ventricle V I I - facia l nerve V I I a - anterior facia l motor nucleus V I I i - intermediate facia l nucleus VTI m - facia l motor nucleus V I I p - posterior facia l motor nucleus X - vagus nerve (or vagal sensory input) X m - vagal motor nucleus Acknowledgments Firs t and foremost, thank y o u to m y supervisor, B i l l M i l s o m . B i l l , your extensive knowledge o f comparat ive phys io logy is truly impress ive , as is your abi l i ty to go for months w i t h m i n i m a l sleep, consuming on ly c innamon buns and coffee. T h a n k y o u for your guidance and support. Thank y o u for your enthusiasm when things went w e l l and for your patience and understanding when things went terribly wrong . Thank you for o rgan iz ing the many lab meetings, retreats, pot lucks , hikes , sk i trips, and other adventures that make the M i l s o m lab such a fun and interesting group to w o r k wi th . Thank you for teaching me many useful sk i l l s and for encouraging me to think. Thank y o u for g i v i n g me the freedom to make mistakes, and for ba i l i ng me out when I got i n over m y head. B i l l , w o r k i n g wi th y o u has been amazing . T h a n k you , and thank y o u again. Thank y o u to m y commit tee members, T r i s h Schul te and C o l i n Brauner , for their excel lent advice and questions. Thank y o u to m y thesis defense examiners, Jef f Richards and T o n y Far re l l , for their great questions and suggestions. T h a n k y o u to a l l the students o f the M i l s o m lab. T h a n k y o u for be ing m y colleagues, offering me excel lent advice and suggestions to improve m y project. E v e n more important ly, thank y o u for be ing m y friends. Y o u r support and kindness mean so m u c h to me. Thank you , L i e n e k e M a r s h a l l , for m a k i n g that crazy summer running experiments i n a construct ion zone bearable, for our fun adventures i n N o v a Scot ia , and for sharing your w i s d o m and friendship. T h a n k you , Joanna P ie rcy , for your endless patience f i x i n g m y computer problems, for be ing m y g y m buddy, and for entertaining me wi th your loveable l izards . Thank you , A n d r e a Corco ran , for a lways cheer ing me up when things went wrong . Thank y o u , Char i s sa F u n g , for four years o f great adventures and for always k n o w i n g just when I most needed a hug. Thank y o u , Ca ta l ina Reyes , for dragging me out to have fun f rom t ime to t ime. Thank y o u , C o l i n Sanders, for teaching me h o w to care for m y f ish and for p rov id ing great conversations when m y experiments ran late and everyone else i n the b u i l d i n g had gone home. Thank you , G l e n n Tattersall , B e t h Z i m m e r , and L i s a Sk inner for your advice and support dur ing m y first summer o f experiments. T h a n k y o u , Jon Chatburn , for your endless supply o f jokes and dancing frogs, and for your excel lent advice. Thank y o u to the new M i l s o m i t e s , B a r b Ga jda and E m i l y C o o l i d g e , for x v l aughing at m y terrible jokes , and G r a h a m Scott, for a lways asking good questions, whatever the topic o f conversat ion. Thank you , Janice M e i e r , Crys t a l Brauner , and K i m B o r g , for keeping our lab running smoothly and for the great luncht ime conversations. Thank y o u to C a r o l P o l l o c k , K a t h y N o m m e , G r e g B o l e , G o r d o n M c l n t y r e , Celeste Leander , T o d d Harper , W a y n e G o o d e y , L y n n N o r m a n , and C h i n S u n for your support, guidance and friendship. Teach ing b io logy w i t h y o u has been an incredible experience. Thank y o u for he lp ing me to be a better teacher. Thank y o u to a l l the fantastic staff at Z o o l o g y , i nc lud ing K a t h y Gorkof f , A l l i s o n Barnes , Scott Usher , B r u c e G i l l e sp i e , H e n r y C h u n g , and A r t h u r Vanderhors t for keeping the department running smoothly and for a l l your help and patience. Thank y o u to m y wonderful f ami ly . T h a n k you to m y parents, M i k e and N a n c y O ' N e i l l , for your endless love and support. Thank y o u for be l i ev ing i n me, even dur ing those times w h e n I no longer d id . Thank you , D a d , for phon ing me every single day w h i l e I was t ry ing to f in i sh m y thesis, and for a lways m a k i n g me smile . T h a n k you , M o m , for our great talks that a lways left me feel ing happier. Thank you , Ka t i e O ' N e i l l , m y amazing sister, for your humour , love , and a l l your c lever surprises. Thank y o u to m y N a n a , Bet ty A n n a n d , for a l l our fantastic adventures, i nc lud ing the great trip to Ireland and our upcoming trip to C h i n a , and for your love and support. Thank y o u to D a d , Ka t i e , N a n a , and R o n and B e v Foster for t ravel ing a l l the way f rom V a n c o u v e r Island to attend m y defense. T o m y entire f ami ly , y o u are m y strength and m y inspirat ion. I love you . Thank y o u to A n g e l a and M a n f r e d Steininger, for our Sunday dinners and excel lent conversations. Thank you , A n g e l a , for reading m y entire thesis and for attending m y defense. T o m y wonderful M i l e s Steininger, thank y o u for he lp ing me to figure out m y statistics problems and for offering me great advice w i t h m y work . Thank y o u for supporting me w i t h love , laughs, and snuggles. T h a n k y o u for m a k i n g each day an adventure. M i l e s , y o u are truly amazing . I l ove y o u so m u c h . F i n a l l y , thank you to Rufus , the litt le b lack and white cat w h o kept me company w h i l e I wrote this thesis. Thank you , Rufus , for b i t ing m y toes, k n o c k i n g over m y pi les o f papers, h id ing m y pens under the dishwasher, and leaping up on m y lap and demanding attention when I was t ry ing to focus. Y o u might be a litt le monster, but y o u s t i l l make me smi le . x v i Chapter 1: Introduction Teleost Respiration: An Introduction Phases of Respiration in Teleost Fish J T o breathe, a f ish must irrigate its g i l l s , w h i c h are the site o f gas exchange. A teleost f ish such as the ra inbow trout generates a nearly continuous f l o w o f water over its g i l l s by suck ing water i n through its mouth to the two opercular cavit ies, where the g i l l s are located, and forc ing it out through its opercular flaps. A typica l respiratory c y c l e is composed o f two major phases: expansion and compress ion. Firs t the mouth opens and the bucca l and opercular cavit ies expand i n vo lume , creating a negative pressure that sucks water i n through the mouth (Hughes and Shel ton, 1962). Because the pressure in the opercular cavit ies is a lways lower than that i n the bucca l cavi ty dur ing the expansion phase, water is sucked through the bucca l cavi ty into the opercular cavit ies, where it f lows over the g i l l s (Hughes and Shel ton, 1962). T o start the compress ion phase, the mouth closes and the bucca l and opercular cavit ies are compressed, creating h igh pressure that forces the water to f l o w over the g i l l s and out through the opercular slits (Hughes and Shel ton, 1962). D u r i n g compress ion, t h e h u c c a l cavi ty pressure is higher than that o f the opercular cavities un t i l the very end o f this phase, when the opercular flaps c lose (Hughes and Shel ton, 1962). T h e bucca l cavi ty pressure reaches its peak before the pressure i n the opercular cavit ies does. Because o f this staggering o f pressure m a x i m u m s between the cavit ies, at the very end o f the compress ion phase, the pressure i n the opercular cavit ies is br ief ly higher than that i n the bucca l cavi ty (Hughes and Shel ton, 1962). T h i s b r i e f per iod when the opercular flaps are complete ly c losed is no rmal ly the on ly t ime dur ing the respiratory cyc l e when water current is not act ively produced. Throughout this work , I used the pressure changes o f the bucca l cavi ty to determine the beg inn ing o f each respiratory phase. H o w e v e r , it is important to remember when discuss ing the phases o f respiration that the pressure changes i n the opercular cavit ies w i l l a lways lag s l ight ly behind those o f the bucca l cavi ty . F igure 1.1 shows the changes i n pressure seen dur ing the phases o f the respiratory cyc le i n a ra inbow trout (Hughes and Shel ton, 1962). 1 Phases of the Respiratory Cycle Mouth Movement Opercular C Movement O Buccal Valve Q L -Opercular C Valve O +1.0 Pressure in cm water -1.0 Buccal cavity positive + 1 Q Differential pressure in cm water Buccal cavity negative -1,0 Seconds Figure 1.1: Phases of the respiratory cycle in trout The movements of the mouth and opercular flaps (C indicates closed while O indicates open) are related to the pressures of the buccal and opercular cavities (measured in cm of water). For the differential pressure trace, positive pressures indicate that the buccal cavity pressure is greater than that of the opercular cavities. (Hughes and Shelton, 1962) Muscles of Respiration in the Trout T h e bucca l cavi ty and opercular cavit ies are mechanica l ly coup led by the hyomandibu la r bone, w h i c h connects the palato-pterygoid bone o f the bucca l cavi ty to the opercular bone o f the opercular cavit ies (Ba l l in t i jn and Hughes , 1965). Therefore, changing the vo lume o f one o f these cavities indi rec t ly affects the vo lume o f the others (Ba l l in t i jn , 1982). Because o f this c o u p l i n g effect, the contract ion o f any muscle connected to the hyomand ibu la or the h y o i d arch w i l l alter the vo lume o f both the bucca l cavi ty and the opercular cavit ies (Ba l l in t i jn , 1982). Consequent ly , the trout is able to operate its respiratory pump us ing on ly three muscles: the muscularis muscularis levator hyomandibulae et arcus palatini, the muscularis adductor arcus 2 palatini et operculi, and the muscularis adductor mandibulae (Ba l l in t i jn and Hughes , 1965). D u r i n g l o w intensity breathing, these three essential pump muscles produce the typica l respiratory c y c l e as fo l lows . Firs t , the muscularis adductor mandibulae closes the mouth by adduct ing the lower j a w (Bal l in t i jn and Hughes , 1965). Nex t , the muscularis adductor arcus palatini et operculi adducts the opercu lum and compresses both the bucca l cavi ty and the opercular cavit ies (Bal l in t i jn and Hughes , 1965). T h i s compress ion greatly increases the pressure i n a l l three cavit ies, forc ing water over the g i l l s and out through the opercular flaps into the environment (Ba l l in t i jn and Hughes , 1965). B y adducting the operculum, w h i c h is connected v i a the mandibulo- interopercular l igament to the lower j a w , the muscularis adductor arcus palatini et operculi also indi rec t ly opens the mouth (Ba l l in t i jn and Hughes , 1965). F i n a l l y , the muscularis muscularis levator hyomandibulae et arcus palatini abducts the palato-pterygoid bone, expanding the bucca l cavi ty (Ba l l in t i jn and Hughes , 1965). Because o f the mechan ica l connect ion between the bucca l cavi ty and the opercular cavit ies, this muscle also indi rec t ly abducts the operculum, increasing the vo lume o f the opercular cavit ies (Ba l l in t i jn and Hughes , 1965). T h i s increase i n bucca l and opercular vo lume generates a negative pressure that draws water through the bucca l cavi ty , into the opercular cavit ies, and over the g i l l s (Ba l l in t i j n and Hughes , 1965). A s w o u l d be expected f rom the phys ica l actions o f these respiratory muscles , the muscularis levator hyomandibulae is active dur ing the expansion phase o f breathing w h i l e the muscularis adductor mandibulae and muscularis adductor arcus palatini et operculi are active dur ing compress ion (Bamford , 1974B) . T h e muscles o f the respiratory pump and the bones o f the sku l l i n the trout are il lustrated i n F igure 1.2 (Ba l l in t i jn and Hughes , 1965). W h e n exposed to either h y p o x i a or hypercapnia, trout increase their depth o f breathing, meaning that they suck i n an increased vo lume o f water per respiratory cyc le (van D a m , 1938). T o increase the magnitude o f expansion, and thus the vo lume o f water avai lable for gas exchange dur ing each respiratory cyc l e , the trout ut i l izes other muscles , i n c l u d i n g the muscularis dilator operculi, muscularis hyohyoideus, muscularis sternohyoideus, and muscularis protractor hyoidei (Ba l l in t i jn and Hughes , 1965). E v e n dur ing quiet breathing, however , the muscularis protractor hyoidei, muscularis dilator operculi, and muscularis sternohyoideus can sometimes show l o w levels o f act ivi ty (Bamford , 1974B) . B y on ly extensively recrui t ing these four muscles dur ing h igh intensity breathing, however , when i t needs to provide the g i l l s w i t h larger vo lumes o f water for gas exchange, the trout decreases the energy required to power its respiratory pump when at rest, thus p rov id ing a h igh l eve l o f respiratory eff ic iency (Bal l in t i jn , 1972). 3 Figure 1.2: A schematic diagram of the muscles of the respiratory pump and their relation to the bones of the palatal and opercular complexes in the trout The abbreviations for the bones are as follows: Pal.pt. = palato-pterygoid, Qu. = quadrate, Hmd. = hyomandibula, Op. = operculum, L.j. = lower jaw, Hy. = hyoid, CI. = cleithrum, Sth. = stylohyal. The abbreviations for the muscles are as follows: D.o. = muscularis dilator operculi, Lev. h.a.p. = muscularis muscularis levator hyomandibulae et arcus palatini, Add. a.p.o. = muscularis adductor arcus palatini et operculi, Add. md. = muscularis adductor mandibulae, P.hy. = muscularis protractor hyoidei, Hy.hy. = muscularis hyohyoideus, Sthy. = muscularis sternohyoideus. (Ballintijn and Hughes, 1965) In addi t ion to the powerfu l muscles o f the respiratory pump, the muscles o f the g i l l arches and g i l l filaments also contract w i th each respiratory cyc l e (Ba l l in t i jn , 1984; B a l l i n t i j n and Punt, 1985). In the carp, the external muscularis branchial arch levators, w h i c h connect the neurocranium to the branchial arches, contract synchronously w i t h the muscularis levator hyomandibulae dur ing the expansion phase o f respirat ion (Ba l l in t i jn and Punt , 1985). These muscles expand the branchial basket by abduct ing the g i l l arches, and also mainta in the even distr ibution o f these arches, keeping the g i l l f i lament curtain continuous to m a x i m i z e the water avai lable for gas exchange (Ba l l in t i j n and Punt , 1985). A d d i t i o n a l l y , these muscles contribute to l ower ing the bucca l cavi ty f loor , thus increasing the vo lume o f water pumped dur ing each respiratory cyc l e (Ba l l in t i jn and Punt, 1985). T h e muscles o f the g i l l arches (the muscularis branchial arch levators) are shown i n F igure 1.3. T h e g i l l f i lament muscles are active dur ing the transition between compress ion and expansion, when the pressure i n the opercular cavit ies is highest (Ba l l in t i jn , 1984). These muscles help to rapid ly re-establish the posi t ive pressure differential between the bucca l and opercular cavit ies dur ing each respiratory c y c l e w h e n the opercular pressure becomes higher than the bucca l pressure (Ba l l in t i jn , 1984). D u r i n g heavy breathing, these muscles are also weak ly active di rect ly f o l l o w i n g m a x i m a l expansion o f the opercular cavit ies (Ba l l in t i jn , 1984). Th i s secondary per iod o f act ivi ty reduces the resistance o f the g i l l s when the opercular pressure is at its lowest , a l l o w i n g more water to f l o w into the opercular cavit ies f rom the bucca l cavi ty , thus preventing any back f low o f water f rom the environment into the opercular cavit ies (Ba l l in t i jn , 1984). Figure 1.3: A sagittal view of the branchial arches of the carp. This diagram shows the five branchial gill arches (1-5), the external muscularis branchial arch levators (el-e4), and the internal muscularis branchial arch levators (il and i2, active only during the cough). The branchial arches are positioned medial to the operculum bone (Op. in Figure 1.2) and so are not normally visible in the intact fish. (Adapted from Ballintijn and Punt, 1985) Innervation of the Respiratory Muscles T h e muscularis adductor mandibulae and muscularis levator hyomandibulae are innervated b y the t r igeminal nerve ( V ) , w h i l e the facia l nerve (VII) innervates the muscularis adductor arcus palatini et operculi (Edgewor th , 1933). T h e glossopharyngeal ( X I ) and vagus 5 ( X ) nerves innervate the g i l l arches and appear to, affect the amplitude o f breathing output (Hughes and Shel ton, 1962). T h e glossopharyngeal nerve innervates the anterior side o f the first g i l l arch wh i l e the vagus nerve innervates the posterior side (Dune l -Erb et al., 1993). Branches o f the vagus innervate the remain ing g i l l arches as w e l l (Dune l -Erb et al., 1993). Because these nerves provide both sensory and motor innervat ion (Dune l -Erb et al., 1993), they must be responsible for d r iv ing the external muscularis branchial arch levators and the g i l l f i lament adductor muscles (Ba l l in t i jn , 1984; B a l l i n t i j n and Punt , 1985). In the teleost medul la , the motor nuc le i o f the t r igeminal , fac ia l , g lossopharyngeal , and vagus nerves are located ventro-lateral to the fourth ventr icle , i n an almost cont inuous c o l u m n ( M e e k and Nieuwenhuys , 1997). The rostral regions o f the fac ia l motor nucleus control the arcus palatini et operculi v i a the facia l nerve (Lui ten , 1976). T h e rostral regions o f the t r igeminal motor nucleus innervate the muscularis adductor mandibulae w h i l e the caudal regions control the muscularis levator hyomandibulae and muscularis dilator operculi v i a the t r igeminal nerve (Lui ten , 1976). C e l l bodies o f the glossopharyngeal motor neurons are found o n the rostral end o f the continuous glossopharyngeal-vagal motor c o l u m n ( K a n w a l and C a p r i o , 1987). T h e c e l l bodies o f the motor neurons that innervate the g i l l arches are found on ly i n the rostral por t ion o f the vagal motor nuc le i ( K a n w a l and C a p r i o , 1987). These vagal motor neurons are arranged i n segments according to w h i c h g i l l arch they innervate ( K a n w a l and C a p r i o , 1987). A l l o f the crania l nerve motor nuc le i contain some neurons that produce acetylchol ine as a neurotransmitter (Eks t rom, 1987). Location and Discharge Patterns of Respiratory Neurons in the Teleost Medulla Respiratory Rhythms Produced in the Isolated Teleost Medulla F o r many years, physiologis ts have studied the product ion o f the respiratory rhy thm i n the teleost brain to determine whether the brainstem is capable o f generating this rhy thm independently or i f input f rom the crania l nerves and higher bra in centres is essential. A d r i a n and Buytend i jk (1931) discovered that the surface o f the isolated goldf i sh bra in produces s low wave depolarizat ions, demonstrating that the bra in is capable o f p roduc ing rhythms wi thout any sensory feedback. These s low wave depolarizat ions produced a rhy thm that approximate ly 6 matched recordings o f respiratory rhythms, leading them to suspect that these signals or iginated i n the respiratory centre and w o u l d have dr iven breathing movements i n the intact go ldf i sh ( A d r i a n and Buytendi jk , 1931). Th i s experiment indicated that the respiratory centre c o u l d poss ib ly generate the respiratory rhy thm wi thout input f rom the crania l nerves; however , i t was s t i l l unclear whether these depolarizations were actually generated by the respiratory neurons. U s i n g microeletrodes, W o l d r i n g and D i r k e n (1951) recorded respiratory discharges di rect ly f rom populat ions o f neurons wi th in the medul la , conf i rming that the brainstem does indeed generate the respiratory rhythm. In 1956, Hukuha ra and O k a d a f ina l ly p roved that the teleost brainstem c o u l d produce the respiratory rhy thm independent o f a l l external input. T h e y comple te ly isolated the medu l l a by severing a l l the crania l nerves and the spinal cord and by transecting the bra in at the rostral border o f the medul la . T h e y then recorded populat ions o f neurons i n that isolated medu l l a that discharged w i t h respiratory rhy thm (Hukuhara and Okada , 1956). In 1962, v o n Baumgar ten and Sa lmoi ragh i conf i rmed that respiratory rhy thm product ion does not depend on proprioreceptive feedback v i a the crania l nerves. They found that when a l l propriocept ive feedback to the respiratory centres i n the medu l l a was r emoved by muscular paralysis , the generation o f respiratory rhy thm cont inued (von Baumgar ten and Sa lmoi ragh i , 1962). Location of respiratory neurons in the teleost medulla Neurons that discharge wi th a respiratory rhy thm have been identif ied i n several m a i n locations w i th in the teleost medul la . F o r example , i n the go ldf i sh , respiratory neurons are located i n two bilateral strips running f rom the caudal edge o f the vaga l lobes up to the rostral edge o f the cerebel lum, at the anterior end o f the medu l l a (Wa ld ron , 1972). These strips are located f rom approximately 0.5 to 1.2 m m lateral to the mid l ine and f rom 2.8 to 5.0 m m ventral to the dorsal surface o f the brain (von Baumgar ten and Sa lmoi ragh i , 1962). H o w e v e r , the highest concentration o f respiratory neurons is found between 3.6 to 4.4 m m ventral to the dorsal b ra in surface (von Baumgar ten and Sa lmoi ragh i , 1962). These respiratory neurons are located i n the motor nuc le i o f the crania l nerves, f rom the t r igeminal to the hypoglossa l , and exist almost exc lus ive ly i n the parvocel lu lar (smal l cel l ) regions, i m p l y i n g that they are interneurons and not motor neurons (von Baumgar ten and Sa lmoi ragh i , 1962). T h e tench medu l l a contains at least 2000 neurons that discharge wi th a respiratory rhy thm (Wald ron , 1972). S i m i l a r to those o f the goldf ish , these neurons are organized into two bilateral strips, lateral to the mid l ine (Wa ld ron , 7 1972). H o w e v e r , near the rostral end o f these strips, ventral to the central por t ion o f the cerebel lum, each bilateral strip appears to diverge into two sub regions (Wald ron , 1972). T h e larger sub reg ion is orientated more dorsal ly and laterally than the smaller sub region, w h i c h is more ventral and runs closer to the mid l ine o f the medu l l a (Wa ld ron , 1972). Respira tory neurons i n the tench medu l l a are located ventral to the motor c o l u m n o f the vagus and glossopharyngeal nerves and i n the region o f the fac ia l and t r igeminal motor nuc le i (Shelton, 1961). In the trout, the respiratory neurons are also found i n two bilateral strips that run f rom approximately 1.5 m m anterior to the obex up to the border between the cerebel lum and the optic lobes (Bamford , 1974B) . M a n y o f these neurons appear to be interneurons, located i n the parvocel lu lar regions o f the motor nuc le i o f the t r igeminal , fac ia l , and glossopharyngeal nerves (Bamford , 1974B) . In the trout, most respiratory neurons appear to be found i n the dorsal ha l f o f the medu l l a (Bamford , 1974B) . H o w e v e r , i n the tench, most sites o f respiratory act ivi ty appear to be i n the ventral ha l f o f the medu l l a (Shelton, 1961). T h i s difference probably reflects the fact that the facia l and t r igeminal motor nuc le i are much more ventral i n the tench brainstem (Shelton, 1961) than i n the trout brainstem ( M e e k and Nieuwenhuys , 1997). A n a t o m i c a l differences such as these often make exact comparisons between different species o f teleosts diff icul t . Howeve r , it is possible to conclude that i n teleosts, respiratory neurons can be found i n a pair o f bilateral strips runn ing f rom the anterior to posterior o f the medu l l a (Ba l l in t i jn , 1982). A n a t o m i c a l regions inc luded i n these strips are the motor nuc le i o f the t r igeminal , fac ia l , g lossopharyngeal , and vagus nerves, as w e l l as the descending t r igeminal nucle i and the reticular format ion (Bal l in t i jn , 1982). The general locat ions o f respiratory neurons i n the teleost medu l l a are shown projected onto the dorsal surface o f the bra in i n F igure 1.4 (Ba l l in t i jn , 1982). Loca t ions o f several respiratory neuron populat ions i n the trout medul la , projected onto the surface o f a sagittal section o f the brain , are shown i n F igure 1.5 (Bamford , 1974B) . 8 Figure 1.4: The location of respiratory neurons in the teleost medulla (projected onto the dorsal surface of the brain.) This particular teleost brain diagram most closely resembles that of the tench. The left side of the diagram shows the main respiratory sites, located in the midbrain ventral to the optic lobes and in the medulla ventral to the cerebellum and facial lobe. The right side of the diagram uses arrows to show the major interconnections between these respiratory neuron populations. The area between the two dashed lines (marked T) is the region necessary to maintain the normal respiratory rhythm (from transection experiments). Abbreviations: m m = oculomotor nucleus, t = tegmental respiratory neurons, Rf = reticular formation, V m = trigeminal motor nucleus, V d = descending trigeminal nucleus, VII m = facial motor nucleus, IX m = glossopharyngeal motor nucleus, VII i = intermediate facial nucleus, Xm = vagal motor nucleus, P = muscle proprioceptive input, X = vagal sensory input. (Ballintijn, 1982) 9 VI Figure 1.5: A sagittal section of the trout medulla showing the location of several respiratory neuron populations. The neuron populations are projected onto the surface of the sagittal section and are represented by black circles. Abbreviations: Cer = cerebellum, Op L = optic lobe, V = trigeminal motor nucleus, VI = abducens motor nucleus, Vila = anterior facial motor nucleus, VIIp = posterior facial motor nucleus, LX-X = glossopharyngeal-vagal motor column. (Adapted from Bamford, 1974B). W i t h i n the respiratory regions o f the tench medul la , neurons that discharge wi th a respiratory rhy thm do not appear to be un i formly distributed (Shel ton, 1961). Nei ther do they appear to be organized into large, discrete nuc le i (Shelton, 1961) or smaller anatomical clusters (Wa ld ron , 1972). Respiratory neurons can be very di f f icul t to locate us ing a stiff wi re electrode; this m a y be because cel ls showing a respiratory rhy thm appear to be randomly scattered throughout the reticular format ion (Hughes and Shel ton, 1962). H o w e v e r , another reason that these cel ls may be diff icul t to identify may be that on ly some o f the total popula t ion o f respiratory neurons may be active dur ing the recording per iod. F o r example , dur ing l o w intensity breathing, the motor neurons that innervate a part icular musc le may alternate their act ivi ty over the course o f several minutes, such that an i n d i v i d u a l motor neuron cou ld either be active or resting at any g iven time (Ba l l in t i jn and A l i n k , 1977). Therefore, even an exhaust ively thorough stiff w i re electrode survey o f the respiratory reg ion c o u l d potent ial ly fa i l to identify a large number o f respiratory neurons. A d d i t i o n a l l y , not even a l l the active respiratory neurons can be identif ied us ing a stiff w i re electrode, since to be recognized as be ing respiratory, a 10 neuron must discharge wi th a rhy thm that can be related to that o f breathing (Hughes and Shel ton, 1962). A l t h o u g h many respiratory neurons do discharge rhy thmica l ly , other neurons o f the respiratory centre may prov ide a tonic exci tat ion or inh ib i t ion and thus w o u l d not be identif ied as respiratory (Hughes and Shel ton, 1962). Therefore, al though we can describe the approximate regions o f the teleost medu l l a that are essential for mainta in ing respiration, our knowledge o f the locat ion o f respiratory neurons w i th in these regions is far f rom complete . A d d i t i o n a l l y , al though any stiff w i re electrode survey o f the medu l l a to identify respiratory neurons w i l l be useful to identify general regions that show respiratory act ivi ty , as w e l l as to catalogue the different types o f rhythmic respiratory act ivi ty , this type o f search cannot be used to conc lus ive ly define respiratory neural act ivi ty. Discharge Patterns of Respiratory Neurons Depend ing on their per iod o f act ivi ty w i t h i n the breathing cyc le , respiratory neurons can be roughly c lass i f ied as compress ion or expansion phase neurons (Hukuhara and Okada , 1959; von Baumgar ten and Sa lmoi ragh i , 1962). H o w e v e r , the variety o f discharge patterns seen i n respiratory neurons can actually be m u c h more complex depending o n the exact timing o f the act ivi ty w i th in the phase (Bamford , 1974B) . A d d i t i o n a l l y , some phase-spanning respiratory neurons are active i n both breathing phases (Wald ron , 1972). In the go ldf i sh , neurons that are neighbours anatomical ly can often discharge wi th very different patterns o f act ivi ty and can be active dur ing complete ly different phases o f the respiratory cyc l e (von Baumgar ten and Sa lmoi ragh i , 1962). Other times, however , neurons i n c lose p rox imi ty to one another can be active dur ing the same phase o f the cyc le , but at different points wi th in the same respiratory phase (von Baumgar ten and Sa lmoi ragh i , 1962). S i m i l a r l y , W a l d r o n found that respiratory neurons i n the goldf i sh and tench are not organized into smal l anatomical clusters w i t h synchronized rhythms, as was prev ious ly hypothesized, but instead found that no particular region showed act ivi ty exc lus ive ly i n one respiratory phase (1972). T h e same results were found i n the trout, where no pattern between the act ivi ty o f a neuron w i t h i n the respiratory c y c l e and its loca t ion i n the brainstem cou ld be determined (Bamford , 1974B) . B a l l i n t i j n and A l i n k also conc luded that respiratory neurons do not appear to be spatially arranged i n groups that fire at the same phase o f the cyc l e ; neurons located side by side may be active dur ing different respiratory phases and have different f i r ing patterns as w e l l (1977). 11 Respiratory Rhythm Generators Previously proposed sites for rhythm generators Because i n the tench transection o f the medu l l a at the caudal edge o f the facia l lobe (the rostral edge o f the exposed fourth ventricle) stops breathing complete ly , Shel ton suggested that neurons i n this region o f the bra in must be responsible for generating the respiratory rhy thm (1959). Af te r compar ing this region to the equivalent area o f the m a m m a l i a n bra in , Shel ton postulated that the neurons c ruc ia l to generating this rhy thm cou ld be located i n the reticular format ion (1959). Neurons f rom the reticular formation prov ide direct input to the facia l and t r igeminal motor nuc le i that control the muscles o f the respiratory pump (Lu i t en and V a n D e r Pers, 1977); p rov id ing support for Shel ton 's postulat ion. B a l l i n t i j n also conc luded that the reticular formation is responsible for respiratory rhy thm generation (1988). H o w e v e r , because damaging tissue anywhere w i th in the reticular format ion d i d not el iminate breathing so l o n g as the magnitude o f destruction was not too large, he conc luded that w i th in the ret icular formation, no one area is essential (Ba l l in t i jn , 1988). Instead, he proposed that the total amount o f funct ional rhy thm generating tissue is more important to mainta in ing the breathing rhy thm than is any part icular locat ion w i t h i n the reticular format ion (Ba l l in t i jn , 1988). T h i s proposal seems to suggest that the reticular formation is homogenous, such that so l ong as a sufficient vo lume o f rhy thm generating tissue remains intact, the respiratory rhy thm w i l l be produced, regardless o f w h i c h region o f the rhythm generator is removed. T h e l og i ca l conc lus ion f rom this proposal is that the teleost medul lar ret icular formation contains on ly one rhy thm generator, possessing a certain leve l o f redundancy. H o w e v e r , anatomical ly , the reticular format ion i n teleosts can be d iv ided into three regions: the median zone, the lateral zone, and the med ia l zone (Nieuwenhuys and Pouwel s , 1983). W h i l e the lateral zone is conf ined to the medul la , both the median and med ia l zones o f the reticular formation extend into the mesencephalon as w e l l ( M e e k and Nieuwenhuys , 1997). The median zone is located along the mid l ine o f the bra in and is also referred to as the raphe nuc le i (Nieuwenhuys and Pouwel s , 1983). The lateral zone, also k n o w n as the nucleus reticularis parvice l lu lar i s , is the smallest reg ion o f the reticular formation and is found between and lateral to the motor nucle i o f the crania l nerves, on the same dorsoventral l eve l o f the 12 medu l l a (Nieuwenhuys and Pouwel s , 1983). T h e largest region o f the reticular formation is the med ia l zone, w h i c h is found i n the ventral part o f the rhombencephalon and the tegmentum mesencephal i o f the mesencephalon (Nieuwenhuys and Pouwel s , 1983). In the rhombencephalon, the media l zone o f the reticular format ion is located just ventral to the motor nuc le i o f the cranial nerves ( M e e k and Nieuwenhuys , 1997). Th i s region contains several populat ions o f extremely large neurons i n addi t ion to neurons o f a more m e d i u m size (Nieuwenhuys and Pouwe l s , 1983). The rhombencephal ic media l zone o f the reticular format ion i n the trout can be further subdivided into a nucleus reticulus superior, medius, and inferior (Nieuwenhuys and Pouwe l s , 1983). T h e nucleus reticulus inferior is located on the same leve l o f the rostrocaudal axis as the glossopharyngeal and vagal motor c o l u m n w h i l e the nucleus reticulus medius and superior are on the same leve l as the facial and t r igeminal motor nuc le i , respect ively (Nieuwenhuys and Pouwel s , 1983). These anatomical d iv i s ions suggest that the reticular format ion is not a homogenous structure. In carp, axons o f large neurons i n the med ia l reticular format ion project onto the t r igeminal and facia l motor nuc le i (Lui ten and V a n D e r Pers, 1977). Therefore, the media l zone o f the reticular format ion cou ld theoretically be rhy thmica l ly exc i t ing the respiratory motor neurons o f the t r igeminal and facia l nerves. In 1972, W a l d r o n suggested that i n the goldf ish and tench, the respiratory rhy thm generator c o u l d be located med ia l to the motor nuc le i o f the fac ia l nerve. S ince the reticular formation is located med ia l to the crania l motor nuc le i (Shelton, 1961), W a l d r o n ' s proposed rhythm generator c o u l d poss ib ly be the nucleus reticulus medius o f the med ia l zone o f the reticular formation. I f the respiratory rhy thm generator is located i n the reticular format ion, then this segmental arrangement o f the med ia l zone into the nucleus reticulus superior, medius, and inferior suggests that the rhythm generator c o u l d also be subdivided into three distinct but connected rhy thm generators. Multiple Respiratory Rhythm Generators in the Early Developmental Stages of Vertebrates In a l l vertebrates, the brainstem originates f rom the rhombencephalon, a vesicle that forms near the rostral end o f the neural tube dur ing embryonic development (Fort in et al., 2000) . D u r i n g the early stages o f development, this tissue is transiently d iv ided rostrocaudally into eight repetitive regions k n o w n as rhombomeres, as is shown i n F igure 1.6 (Fort in et al., 2000). F o r example , i n the c h i c k embryo, these rhombomeres can be found f rom stages 12 to 24 (Fort in et 13 al., 1995). These rhombomeres are organized into odd/even pairs (for example r3r4 or r5r6); the branchia l motor nerves exi t the medu l l a f rom the even numbered rhombomeres (Kin tner and L u m s d e n , 1999) wh i l e the odd numbered rhombomeres eventually develop to produce the adult respiratory rhy thm (Fort in et al., 2000). Therefore, each odd/even pair o f rhombomeres possesses its o w n independent respiratory rhy thm generator (Champagnat and For t in , 1997). Because , dur ing vertebrate development, the ch i ck hindbrain is partit ioned into eight rhombomeres that are organized into odd/even pairs, and because each pair contains its o w n rhy thm generator (Fort in et al, 2000), the c h i c k embryo must then possess at least four respiratory rhy thm generators. Th i s developmental per iod o f hindbrain d iv i s i on into rhombomeres appears to be c o m m o n to a l l vertebrates (Fort in et al., 2000). Therefore, it can be assumed that at some stage o f embryonic development, a l l vertebrates posses mul t ip le respiratory rhy thm generators. H o w e v e r , as adults, do a l l vertebrates retain these mul t ip le rhy thm generators? Figure 1.6: A sagittal schematic of the embryonic mouse hindbrain. The hindbrain is divided into eight rhombomeres (rl to r8), with the odd numbered rhombomeres in black. The trigerninal nerve (5n) emerges from r2, the facial nerve (7n) emerges from r4, and the glossopharyngeal nerve (9n) emerges from r6. (Adapted from Fortin et al, 2000) 14 Multiple Respiratory Rhythm Generators in Other Vertebrates M u l t i p l e rhy thm generators have been identif ied i n several species o f vertebrates, i nc lud ing the adult skate (Hyde , 1904), lamprey (Rova inen , 1983; Russe l , 1984; Thompson , 1985), and frog ( W i l s o n et al., 2002), and the embryonic c h i c k (For t in et al., 1995) and rat (Onimaru and H o m m a , 2003) . B y compar ing the number o f rhy thm generators i n these other vertebrates, and by understanding their phylogenet ic relat ion to teleosts, it is possible to predict the number o f respiratory rhy thm generators expected i n the teleost medul la . In the skate, no one region o f the medu l l a is exc lus ive ly responsible for con t ro l l ing a l l the respiratory movements (Hyde , 1904). T h e skate medu l l a can be isolated f rom both the spinal co rd and the anterior portions o f the bra in by transection and s t i l l produce the respiratory rhy thm (Hyde , 1904). W h e n the respiratory centre i n the medu l l a is transected a long the mid l ine , each side continues to generate the respiratory rhy thm, al though one side may have a s l ight ly different rhy thm than the other, indica t ing that the skate medu l l a contains at least one pair o f bilateral rhy thm generators (Hyde , 1904). A d d i t i o n a l l y , when the medu l l a is transected perpendicular to the mid l ine , between two motor nuc le i o f the respiratory crania l nerves, each isolated ha l f is capable o f p roduc ing its o w n rhy thm (Hyde , 1904). F o r example , when the medu l l a is transected between the motor nuc le i o f the glossopharyngeal and vagus nerves, the spiracle and first g i l l arch (dr iven by the facia l and glossopharyngeal nerves) as w e l l as the remain ing four g i l l arches (driven b y the vagus) continue to make rhy thmic breathing movements (Hyde , 1904). H o w e v e r , the movements o f the spiracle and first g i l l arch sometimes f o l l o w a s l ight ly different rhy thm than those o f the remain ing four g i l l arches (Hyde , 1904). T h i s abi l i ty to generate a respiratory rhy thm independent o f the rest o f the medu l l a demonstrates that the region o f the respiratory centre that controls the spiracle and first g i l l arch and the region that controls the remain ing four g i l l arches are stimulated by separate rhy thm generators. Therefore, the skate medu l l a contains at least two bilateral pairs o f respiratory rhy thm generators. In the lamprey, the motor neurons innervat ing the respiratory muscles o f the branchia l basket are dr iven by pacemaker cel ls ; the s t imulat ion f rom the pacemaker cel ls causes excitatory postsynaptic potentials i n the motor neurons, caus ing them to discharge (Rova inen , 1974). T h e m i d l i n e o f the medul la , at the rostrocaudal l eve l o f the t r igeminal motor nuc le i , generates spontaneous respiratory bursts that occur just pr ior to respiratory motor neuron bursts (Rova inen , 1985). These spontaneous bursts c o u l d be generated by the pacemaker cel ls that stimulate the 15 respiratory motor neurons (Rova inen , 1985), and thus are evidence for the existence o f respiratory rhy thm generators i n the lamprey medul la . W h e n the lamprey medu l l a is transected a long its mid l ine , both sides continue to produce respiratory discharges, but w i th differ ing rhythms ( K a w a s a k i , 1979). T h e abi l i ty o f each ha l f o f the lamprey medu l l a to generate a respiratory rhy thm independently o f the other demonstrates that the lamprey must have at least one pair o f bilateral rhy thm generators (Rova inen , 1985). Spontaneous respiratory bursts have also been recorded i n the facia l , glossopharyngeal , and vagal motor nuc le i i n the caudal medul la , as w e l l as at least three locations near the t r igeminal motor nucleus (Rova inen , 1983; Russe l , 1984; Thompson , 1985). Whether these are the locations o f independent rhy thm generators or are s imp ly relay sites connect ing m o t o n e u r o n s to one central rhy thm generator remains to be determined; however , this evidence suggests that the lamprey may have at least two pairs o f respiratory rhy thm generators. T h e isolated frog brainstem is capable o f p roduc ing the respiratory rhy thm independent o f any input f rom the crania l nerves, spinal cord , or forebrain ( M c L e a n et al, 1995). Th i s abi l i ty demonstrates that the frog medu l l a contains at least one spontaneous rhythm generator. W h e n the frog (Rana catesbeiana) medu l l a is transected along the mid l ine , the two isolated halves continue to produce a respiratory rhy thm independently, demonstrating that the frog medu l l a contains at least one bilateral pair o f respiratory rhy thm generators ( M c L e a n et al, 1995). T w o distinct sites o f respiratory rhy thm generation have been identif ied i n the frog medul la , one located between the facial and glossopharyngeal nerve roots and the other located at the most caudal root o f the vagus nerve ( W i l s o n et al, 2002). B o t h o f these sites are located i n the reticular format ion o f the medu l l a ( W i l s o n et al., 2002). W h e n these two sites are separated b y transection i n the isolated brainstem, both continue to discharge w i t h a respiratory rhy thm, demonstrating that there are at least two bilateral pairs o f respiratory r h y t h m generators i n the frog medu l l a ( W i l s o n et al, 2002). These two sites function together as a paired osci l lator to generate the respiratory rhy thm ( W i l s o n et al, 2002) . Interestingly, the loca t ion o f these rhy thm generators is not constant throughout the l i fe t ime o f the frog Rana catesbeinana: dur ing the metamorphosis f rom tadpole to adult, the loca t ion o f the lung rhy thm generator appears to migrate f rom the caudal to the rostral brainstem (Torgerson et al, 2001.) Therefore, i n the intact adult frog, the more anterior rhy thm generator appeals to control lung venti lat ion w h i l e the more posterior rhy thm generator appears to control bucca l vent i lat ion ( W i l s o n et al, 2002) . 16 In the c h i c k embryo, respiratory motor rhythms can be recorded f rom the t r igeminal , facia l , glossopharyngeal , and vagus nerves as early as stage 24 (approximately 3.5 days), and this act ivi ty increases as the c h i c k embryo continues to develop (Fort in et al, 1994). D u r i n g stages 24 to 36 (approximately 3.5 to 10 days), when the medu l l a is no longer segmented, each motor nucleus is paired to its o w n rhy thm generator and thus is capable o f generating a respiratory rhy thm i f isolated by transection f rom the rest o f the h indbra in (Fort in et al, 1995). In the intact c h i c k embryo, these rhy thm generators communicate through connections between segments and across the mid l ine to produce a coordinated rhy thm (For t in et al, 1995). These data indicate that the b i rd medu l l a also contains mul t ip le respiratory rhy thm generators, at least dur ing the ear ly stages o f embryonic development. In the embryonic mouse (from 10.5 to 11.5 days after fert i l izat ion) the motor neurons o f the vagus, glossopharyngeal , fac ia l , and t r igeminal nerves appear to generate rhythms independently (Abad ie et al, 2000), ind ica t ing that mammals , at least i n the ear ly stages o f development, m a y also possess mul t ip le rhy thm generators. In adult mammals such as the cat, the pre-Botz inger complex i n the ventral respiratory group o f the medu l l a is capable o f spontaneously discharging w i t h respiratory rhy thm when isolated f rom the rest o f the medu l l a (Smi th etal, 1991). A p p r o x i m a t e l y twenty-f ive percent o f neurons discharging w i t h respiratory rhy thm i n the pre-Botz inger complex appear to be pacemaker cel ls (Smi th et al, 1991). T h e pre-Bo tz inge r complex is located i n approximately the same region o f the medu l l a (ventral to the area between the motor nuc le i o f the facia l and glossopharyngeal nerves) as the l ung rhy thm generator o f the frog ( W i l s o n et al, 2002). Neurons i n the pontine respiratory group (dorso-med ia l to the t r igeminal nerve root) have also been suggested as a site o f rhy thm genesis (St.John, 1998; F e l d m a n and M c C r i m m o n , 1999). In the newborn rat, both the pre-Botz inger complex and the para-facial respiratory group (located ventrolateral to the fac ia l nucleus, close to the ventral surface o f the medul la) appear to generate respiratory rhy thm (On imaru and H o m m a , 2003). Neurons i n these two regions appear to interact, generating and regulat ing the respiratory rhy thm as a coup led osci l la tor (Onimaru and H o m m a , 2003; M e l l e n et al, 2003) . These two respiratory rhy thm generating sites correspond i n locat ion to the sites o f respiratory rhy thm generation i n the frog ( W i l s o n et al, 2002) In both species, one site o f the coupled osci l la tor is active dur ing the pre-inspiratory phase o f each breath wh i l e the other site is active dur ing the inspiratory phase o f each breath; w h i l e the inspiratory sites are inhibi ted by opio ids , the pre-inspiratory sites are not (Vas i l akos et al, 2005). 17 In mammals such as the mouse, the respiratory centres are reorganized dur ing development, s imi la r to the reorganizat ion o f respiratory centres i n the medu l l a o f the deve lop ing frog: the s imple segmental arrangement o f respiratory rhy thm generators i n the embryo is gradual ly converted to the more complex arrangement o f discrete respiratory nuc le i i n the adult ( reviewed i n Champagnat and For t in , 1997). Therefore, it appears that i n both mammals and amphibians, the embryonic arrangement o f segmental respiratory rhy thm generators is altered i n the adult fo rm. H o w e v e r , at least i n some species, the adults do retain mul t ip le respiratory rhy thm generators. Experimental Evidence for Multiple Rhythm Generators in Teleosts Because a l l vertebrates appear to possess mul t ip le respiratory rhy thm generators dur ing embryonic development, and at least some groups appear to retain this feature into adulthood, I w o u l d expect that adult trout medul la c o u l d also contain mul t ip le rhy thm generators. Cons ide r ing the phylogenet ic evidence that both the adult skate and frog possess mul t ip le respiratory rhy thm generators, I w o u l d expect that the teleosts, w h i c h fa l l between the elasmobranches and the amphibians, should also have mul t ip le generators. Indeed, experimental evidence for mul t ip le respiratory rhy thm generators i n the tambaqui, a neotropical teleost, (Sundin et al, 2000) supports this hypothesis. In most previous studies o f respiration i n teleosts, the expansion and compress ion movements o f the bucca l and opercular cavities are coup led together w i t h the same rhythm, as can be seen i n F igure 1.1 (Hughes and Shel ton, 1962). H o w e v e r , the bucca l and opercular rhythms i n quiet ly breathing, intact tambaqui (Colossoma macropomum), a neotropical teleost, can vary independently o f each other (Sundin et al., 2000). Somet imes the bucca l rhy thm is more rapid than the opercular rhy thm, va ry ing f rom a coup l ing ratio o f 1:1 to 1:1.6 (Sundin et ai, 2000) . These bucca l and opercular movement rates also respond differently to respiratory s t imulat ion: in t roducing N a C N into the inspi red water can increase the bucca l rhythm whi l e l eav ing the opercular rhythm unaffected (Sundin et al., 2000). Because o f this occas ional asynchronous variat ion i n the tambaqui bucca l and opercular rhythms, S u n d i n et al. (2000) suggested that the bucca l and opercular movements are cont ro l led by two independent central rhy thm generators that are normal ly but not a lways entrained to create a coupled rhythm. 18 The Quantity and Location of Respiratory Rhythm Generators in the Trout Medulla T h e ret icular formation is the proposed site o f the respiratory rhy thm generator i n vertebrates (Ba l l in t i jn , 1988). Because the neurons o f the med ia l reticular format ion o f the trout medu l l a can be d iv ided anatomical ly into three regions (Nieuwenhuys and Pouwe l s , 1983), I propose that trout have at least three bi lateral respiratory rhy thm generators. T h e rostral rhy thm generator w i l l be located i n the nucleus reticulus superior o f the media l reticular format ion. T h i s proposed rhy thm generator w o u l d most l i k e l y dr ive the motor neurons o f the t r igeminal nerve, w h i c h innervate the muscularis adductor mandibulae and muscularis levator hyomandibulae muscles o f the respiratory pump. The midd le rhythm generator w i l l be located i n the nucleus ret iculus medius wh i l e the caudal rhy thm generator w i l l be located i n the nucleus reticulus inferior. T h e midd le rhy thm generator w o u l d dr ive the motor neurons o f the facia l nerve, w h i c h innervate the muscularis adductor arcus palatini et operculi o f the respiratory pump. F i n a l l y , the caudal rhy thm generator w o u l d dr ive the motor neurons o f the glossopharyngeal and vagus nerves, w h i c h innervate the muscles o f the g i l l arches and filaments. In the intact f ish , these three rhy thm generators, al though distinct, w o u l d normal ly interact to produce a coordinated breathing cyc le . H o w e v e r , these sites should each be capable o f p roduc ing a respiratory rhy thm independently i f isolated ( W i l s o n et ai, 2002). Breathing Patterns in Teleosts What is Breathing Pattern ? T h e basic breathing rhy thm produced by the rhy thm generators i n the medu l l a is modi f i ed to produce the necessary total vent i la t ion to satisfy a l l respiratory needs. A s respiratory dr ive changes, teleost f i sh can vary their total vent i lat ion by increasing or decreasing their breathing frequency, ampli tude, or both (Hughes and Shel ton, 1962). H o w e v e r , they can also vary total vent i la t ion by changing the timing components o f breathing pattern. These dictate the w ay i n w h i c h breaths are organized over time. In mammals , the most c o m m o n l y seen breathing pattern is cont inuous breathing ( M i l s o m , 1991), where one breath fo l lows another w i th litt le or no pause i n between. Howeve r , there are other ways to arrange breaths over t ime. Some vertebrates, such as the bu l l f rog {Rana catesbeiana), normal ly breathe wi th a pattern o f s ingle or 19 double breaths that are randomly distributed ( M i l s o m , 1991). The T o k a y gecko (Gekko gecko) also normal ly breathes randomly, w i t h single breaths or groups o f several breaths separated by apneas o f variable length ( M i l s o m , 1984). Ano the r breathing pattern seen i n vertebrates such as carp (Lumsden , 1996), turtles (Vi t a l i s and M i l s o m , 1986), and hibernating ground squirrels ( M i l s o m et al., 1997) is the episodic pattern. In the episodic breathing pattern, breaths are organized into groups (episodes) that are separated by periods without breathing (apneas). F igure 1.7 compares a representative trace o f the continuous and episodic breathing patterns i n decerebrate/spinalectomized carp. A l t h o u g h the duration o f the apneas and episodes, as w e l l as the number o f breaths per episode, can be variable, this pattern differs f rom the random breathing pattern i n that episodic breathing is a lways c lear ly rhy thmic . Continuous Episodic 1 cmH 2 0 20 seconds Figure 1.7: Episodic and continuous breathing patterns in the carp. These breathing traces were recorded from the pressure changes in the buccal cavity of a decerebrate/spinalectomized carp. It is possible to change breathing pattern wi thout changing overa l l breathing frequency, as can be seen i n F igure 1.8. S l o w , continuous breaths can y i e l d the same total vent i la t ion and overa l l breathing frequency as more rap id breaths grouped into episodes because the apneas i n the episodic breathing pattern reduce the overa l l breathing frequency. 20 Normal Frequency: Continuous Low Frequency: Continuous Low Frequency: Episodic 1cmH20 10 sec Figure 1.8: The episodic breathing pattern can be used to reduce overall breathing frequency. Breathing frequency can be reduced from normal levels (Normal Frequency: Continuous) by simply slowing the continuous breathing pattern (Low Frequency: Continuous) or by organizing breaths into episodes separated by apneas (Low Frequency: Episodic). In both cases the overall breathing frequency has been halved (from approximately 60 breaths/min to approximately 30 breaths/min) compared to the normal frequency. These breathing traces were recorded from the pressure changes in the opercular cavity of a rainbow trout. Why Breathe Episodically ? A t least some species i n a l l vertebrate classes have been found to breathe ep isodica l ly . M a m m a l s , birds, and most water-breathing f ish tend to breathe cont inuously dur ing normal levels o f respiratory dr ive , but many species change their breathing pattern to an episodic one when respiratory dr ive is reduced ( M i l s o m , 1991). F o r example , mammals such as the golden-mantled ground squirrel (Spermophilus lateralis) breathe ep isod ica l ly dur ing hibernat ion ( M i l s o m et al., 1997) w h i l e northern elephant seals breathe ep isodica l ly dur ing sleep ( M i l s o m et al., 1996). Ech idnas (Tachyglossus aculeatus) also breathe episodica l ly when hibernat ing; however , this episodic breathing pattern can be converted back to continuous breathing wi th exposure to hypercapnia ( N i c o l and Ande r son , 2003). F i s h such as the ra inbow trout normal ly breathe cont inuously, but when exposed to increased levels o f d isso lved oxygen they w i l l often breathe i n episodes (Randa l l and Jones, 1973; W o o d and Jackson, 1980). E v e n birds such as the ch icken , at least dur ing embryonic development, have been found to breathe ep isod ica l ly (For t in et al., 1995). M i c e also breathe ep isod ica l ly dur ing embryonic development (Abad ie et al., 1999). M a n y reptiles and amphibians, as w e l l as some f i sh , breathe ep isodica l ly dur ing normal 21 levels o f respiratory dr ive and on ly breathe cont inuously dur ing periods o f increased respiratory dr ive ( M i l s o m , 1991). F o r example , f ish such as the carp normal ly breathe ep isodica l ly , but w i l l breathe cont inuously when exposed to h y p o x i a ( L o m h o l t and Johansen, 1979; Juch and B a l l i n t i j n , 1983; L u m s d e n , 1996). A l l i ga to r s (Alligator mississippiensis) and caimans (Caimen sclerops) also normal ly breathe ep isodica l ly (Nai feh et al., 1971), as do turtles such as Pseudemys scripta (V i t a l i s and M i l s o m , 1986). Bu l l f rogs (Rana catesbeiana) no rmal ly tend to breathe i n single or double breaths distributed randomly over t ime; however , when respiratory dr ive is increased by exposure to h y p o x i a or hypercapnia they tend to group their breaths into regular ly-spaced episodes ( M i l s o m , 1991). That episodic breaming has been conserved i n some species i n a l l classes o f vertebrates suggests that this breathing pattern is somehow advantageous. What , then, is the advantage o f breathing episodica l ly ? Brea th ing , l i ke any other mechanica l act ivi ty , consumes energy. F o r each breath, energy is required to overcome the mechanica l resistance o f the breathing apparatus. H o w e v e r , the amount o f w o r k done i n each breath varies depending on the t idal vo lume and frequency o f breathing and the specific mechanica l properties o f the breathing apparatus. Therefore, for any g iven species, there should be some opt imal combina t ion o f breathing frequency and ampli tude that satisfies the respiratory requirements o f the an imal at each l eve l o f respiratory dr ive w h i l e m i n i m i z i n g the w o r k o f breathing ( M i l s o m , 1984). A s respiratory dr ive decreases, so too does the op t imal combina t ion o f breathing frequency and ampli tude for m i n i m i z i n g the w o r k o f breathing; however , total vent i la t ion can on ly decrease to a certain leve l before the vo lume o f water or air inspi red is insufficient to overcome the dead space o f the breathing apparatus ( M i l s o m , 1984). Therefore, once the t idal vo lume o f breaths reaches a certain m i n i m u m , to mainta in efficient breathing dur ing any further reductions i n respiratory dr ive the breathing frequency alone should decrease. F o r example , the T o k a y gecko (Gekko gecko) tightly regulates its t idal vo lume at approximately twice the vo lume o f the dead space o f its lungs and b ronch ia l tubes so that at least ha l f o f the air inspired is avai lable for a lveolar respiration ( M i l s o m , 1984). F o r breaths o f this ideal ampli tude, breathing faster or s lower than the op t imum frequency consumes unnecessary energy ( M i l s o m and V i t a l i s , 1984). H o w e v e r , what i f respiratory dr ive falls to a l eve l where the breathing frequency required to meet the respiratory needs is less than the op t imum frequency required to m i n i m i z e the cost o f breathing ? Decreas ing the breathing frequency w o u l d actually increase the w o r k required for each breath. One solut ion to this d i l e m m a is to change the breathing pattern f rom a continuous one to an episodic one. B y 22 breathing episodica l ly , it is possible to decrease the overa l l breathing frequency but s t i l l mainta in the o p t i m u m frequency for m i n i m i z i n g the w o r k o f respiration. T h i s compromise is possible because wi th in each episode the instantaneous frequency can be maintained at the opt imal l eve l w h i l e the overa l l breathing frequency is s t i l l reduced because o f the apneas between these episodes. T h e turtle, Pseudemys scripta, appears to use episodic breathing to m i n i m i z e the energy consumed by respiration: to increase or decrease the overa l l breathing frequency, the turtle s i m p l y increases or decreases the duration o f apneas between episodes w h i l e w i t h i n the episodes themselves the instantaneous frequency and t idal vo lume are mainta ined at the op t imal levels predicted to m i n i m i z e the w o r k o f breathing (Vi ta l i s and M i l s o m , 1986). Thus breathing w i t h an episodic breathing pattern appears to m i n i m i z e the w o r k o f breathing dur ing periods o f l o w respiratory dr ive. Breathing Patterns in Rainbow Trout and Carp Trout normal ly breathe cont inuously and vary their total vent i lat ion by increasing the frequency or ampli tude o f these continuous breaths; however , when they experience very l o w levels o f respiratory dr ive , such as i n hyperoxia , they can breathe ep isod ica l ly (Randa l l and Jones, 1973; W o o d and Jackson, 1980). Ca rp , on the other hand, no rmal ly breathe i n episodes, al though they w i l l convert to a continuous breathing pattern when their respiratory dr ive is h i g h enough, such as when they are exposed to h y p o x i a ( L o m h o l t and Johansen, 1979; Juch and B a l l i n t i j n , 1983; Glass et al., 1991; L u m s d e n , 1996). H o w e v e r , al though breathing pattern can be general ized at each leve l o f respiratory dr ive as either episodic or continuous, breathing pattern should more accurately be described as a spectrum that ranges f rom continuous to episodic breathing. A s respiratory dr ive increases, the duration o f apneas decreases such that the breathing pattern gradual ly changes f rom episodic to continuous. W h e n the breathing pattern contains apneas that are short and i r regular ly spaced, the breaths are not grouped into clear episodes, g i v i n g a pattern o f "messy episodes" that has a higher overa l l frequency than the episodic breathing pattern but a lower overa l l frequency than the continuous one. A l t h o u g h this "messy ep i sod ic" pattern is often purely a transit ional state between the continuous and episodic breathing patterns, it can sometimes persist for substantial periods o f time dur ing intermediate levels o f respiratory dr ive and thus is wor th ment ioning . W a x i n g and w a n i n g (also k n o w n as the Cheyne-Stokes pattern) is another intermediate breathing pattern that can also persist for l o n g 23 periods o f time, especial ly i n hyperoxic f ish (Dejours et al, 1977). A l t h o u g h this pattern is continuous because there are no apneas, it differs f rom regular continuous breathing because breathing ampli tude and frequency increase and decrease i n a rhythmic pattern. F igure 1.9 shows the spectrum o f breathing patterns o f carp. The spectrum o f breathing patterns i n the carp is s imi la r to that o f the tambaqui (Colossoma macropomum) , w h i c h also appears to range f rom continuous to w a x i n g and wan ing to episodic ( R e i d et al, 2003). Clear Episodes —441-4— Weakly Episodic Waxing and Waning Continuous 1 cmH20 30 seconds Figure 1.9: The spectrum of breathing patterns in carp, from continuous to episodic. These representative traces were recorded from the pressure changes in the buccal cavity of a decerebrate/spinalectomized carp. 24 The Influence of Peripheral Feedback from Cranial Nerves on Respiration Respira tory motor output can be modi f i ed by peripheral feedback f rom sensory receptors innervated by the crania l nerves. F o r example , feedback f rom g i l l arch proprioceptors can modi fy the breathing rhy thm dur ing quiet respiration (Satchell , 1959), and can even entrain the breathing rhy thm when the g i l l arch osci l la t ions are ar t i f ic ia l ly manipulated (de G r a a f and Roberts , 1991). These g i l l arch proprioceptors sense the relative pos i t ion o f the g i l l arch dur ing respirat ion; their tonic discharge frequency increases dur ing g i l l arch abduction and decreases dur ing adduction (de G r a a f and B a l l i n t i j n , 1987). Proprioceptors i n the first g i l l arch are sensitive to mechanica l s t imulat ion o f the g i l l f i laments, or rakers (Bur leson and M i l s o m , 1993). Mechanoreceptors located i n the first g i l l arch provide feedback v i a the glossopharyngeal nerve whereas the last four g i l l arches provide feedback v i a the vagus nerve (Satchel l , 1959; D u n e l - E r b et al., 1993). T h i s feedback is transmitted f rom the glossopharyngeal and vagus nerves to the respiratory motor neurons v i a the intermediate facia l nucleus (Satchel l , 1959). Feedback f rom the g i l l arches v i a the glossopharyngeal and vagus nerves also appears to affect the ampli tude o f breathing output (Hughes and Shel ton, 1962). In the tench, after transection o f the vagus and glossopharyngeal nerves, the ampli tude o f respiratory movements increased for a per iod o f one or two hours before returning to pre-transection levels (Shel ton, 1959). Th i s temporary increase suggests that i n the intact animal , feedback f rom the g i l l arches v i a the glossopharyngeal and vagus nerves regulates breathing by cont ro l l ing the ampli tude o f breaths, perhaps to prevent excessive stretching o f the cartilage segments o f the g i l l arches. H o w e v e r , i f this feedback is r emoved , alternate control mechanisms can apparently assert themselves after a transit ion per iod o f several hours. Such alternative control mechanisms c o u l d include sk in stretch receptors and other mechanoreceptors i n tissues surrounding the j a w that are st imulated b y respiratory movements and send their feedback to the t r igeminal nucleus (Juch, 1981). T h e breathing rhy thm can also be inf luenced by feedback f rom peripheral chemoreceptors, such as oxygen receptors (Daxboeck and Ho le ton , 1978) and carbon d iox ide receptors (Hughes and Shel ton, 1962). F o r instance, oxygen receptors i n the first g i l l arch o f trout, innervated b y the glossopharyngeal nerve, respond to h y p o x i a and cyanide (Bur leson and M i l s o m , 1993; B u r l e s o n and M i l s o m , 1995). In the traira (Hoplias malabaricus), h y p o x i a stimulates oxygen chemoreceptors i n the first g i l l arch, causing an increase i n both breathing ampli tude and frequency (Sundin et al., 1999). In the tambaqui (Colossoma macropomum), 25 oxygen receptors on a l l o f the g i l l arches are stimulated by hypox ia , result ing i n an increase i n breathing frequency (Sundin et al, 2000) . A d d i t i o n a l l y , carbon d iox ide and/or p H chemoreceptors i n the g i l l arches o f the tambaqui responded to hypercarbia to stimulate an increase i n breathing frequency (Sundin et al, 2000). The first g i l l arch o f teleosts contains chemoreceptors that can sense external oxygen levels ( in the inspired water) as w e l l as chemoreceptors that can sense internal oxygen levels ( in the b lood) ; however , i t is s t i l l unclear whether the remain ing g i l l arches have on ly external oxygen chemoreceptors or whether they have both ( reviewed i n M i l s o m , 1996). A l t h o u g h the exact ' locat ion o f a l l o f these chemoreceptors and their feedback pathways has yet to be determined, most i f not a l l o f them are l i k e l y innervated by the glossopharyngeal or the vagus nerves. Teleosts do not appear to have central chemoreceptors ( reviewed i n M i l s o m , 1996). A l t h o u g h sensory feedback can certainly influence breathing rhy thm and pattern i n intact animals , it is not essential for the product ion o f the episodic breathing pattern. F o r example , the bul l f rog (Rana catesbeiana) isolated brainstem-spinal co rd preparation can produce an episodic f ic t ive breathing pattern even though it is c lear ly isolated f rom a l l peripheral sensory feedback ( R e i d and M i l s o m , 1998). Therefore, a l l the inputs necessary for the generation o f the episodic breathing pattern must be contained w i t h i n the bra in itself. The Role of the Midbrain in Breathing Pattern Respiratory Input from Higher Centres T h e respiratory rhy thm generated i n the medu l l a is inf luenced by inputs f rom more rostral regions o f the bra in , i nc lud ing sites i n the midbra in and cerebel lum. F o r example , i n carp, a popula t ion o f cerebellar neurons is poss ib ly responsible for integrating the movements o f a l l c rania l muscles , i nc lud ing those that power the respiratory pump (Ba l l in t i j n et al, 1979). Project ions o f several o f these neurons were traced back to the descending t r igeminal nucleus, w h i c h receives propriocept ive signals f rom the respiratory muscles; these neurons were found to be sensitive to forced mechanica l movement o f the bucca l and opercular cavit ies, as w e l l as to the respiratory rhy thm (Bal l in t i jn et al, 1979). In goldf ish , the cerebel lum c o u l d poss ib ly have a tonic inh ib i tory effect on the respiratory neurons i n the medu l l a (von Baumgar ten and 26 Salmoi ragh i , 1962). H o w e v e r , i n the tench, complete r emova l o f the cerebel lum has litt le to no effect on respiration (Shelton, 1959). Other regions o f the bra in are. responsible for integrating sensory inputs w i t h respiratory movements ; for example , i n carp, neurons i n the tectum opt icum show respiratory rhy thm, w h i c h c o u l d poss ib ly be to correct v i sua l images for the mino r distortions o f the retina caused by breathing (Ba l l in t i jn et al., 1979). It is possible that feedback f rom these sensory integrative regions c o u l d also affect the respiratory centre. In the carp, r emov ing the mesencephalon can alter the breathing pattern (Juch and B a l l i n t i j n , 1983), however , i n the tench, transecting the optic lobes has on ly a m i n i m a l effect on the breathing rhy thm (Shel ton, 1959). Experimental Evidence for Midbrain Involvement in Breathing Rhythm in Teleosts A l t h o u g h the isolated teleost medu l l a is capable o f p roduc ing the basic respiratory rhy thm without any peripheral or higher inputs (Hukuhara and Okada , 1956), r emova l o f the midbra in appears to alter the respiratory pattern, suggesting that sites i n the mesencephalon p lay an important role i n generating the breathing pattern. In 1983, Juch and B a l l i n t i j n ident i f ied a respiratory site i n the dorsal mesencephalic tegmentum o f carp that discharged just pr ior to each breathing episode. W h e n this site was e lect r ica l ly stimulated dur ing an apnea, i t ini t iated a breathing episode; when stimulated cont inuously this site increased the duration o f breathing episodes, but not indefini te ly (Juch and B a l l i n t i j n , 1983). W h e n stimulated w i t h short, repetitive electr ical shocks, this site entrained the breathing rhythm, indica t ing that it has a powerfu l effect on the respiratory rhy thm generators o f the medu l l a (Juch and B a l l i n t i j n , 1983). In fact, horse radish peroxidase ( H R P ) injections to this site showed that it is direct ly connected to the reticular formation i n the medu l l a (Juch and B a l l i n t i j n , 1983). Neurons i n this site discharged w i t h a higher frequency when oxygen uptake was l imi ted , suggesting that they respond to oxygen chemoreceptor input by increasing the frequency o f breathing episodes when l o w oxygen levels increase respiratory dr ive (Juch and B a l l i n t i j n , 1983). F i n a l l y , when l igh t ly anesthetized, carp no longer breathe ep isodica l ly (Juch and B a l l i n t i j n , 1983), m u c h l i ke anesthetized caimans {Caiman sclerops) (Nai feh etal., 1971b). W h e n carp are anesthetized, this dorsal mesencephalic tegmental site no longer discharges rhy thmica l ly (Juch and B a l l i n t i j n , 1983), again suggesting that it is i n v o l v e d i n generating episodic breathing. In conc lus ion , these results suggest that this mesencephal ic tegmental site plays an important role i n generating the episodic breathing 27 pattern, ending each apnea by ini t ia t ing a breathing episode and regulating the frequency o f breathing episodes i n response to the leve l o f respiratory dr ive (Juch and Ba l l i n t i j n , 1983). Th i s site is located approximately 3.5 to 4.5 m m ventral to the dorsal surface o f the optic lobes, just ventrolateral to the oculomotor nucleus and ventromedial to the nucleus lateralis valvulae (Juch and Ba l l i n t i j n , 1983), as can be seen i n Figure 1.10. The locat ion o f the site (or sites) that terminates a breathing episode and initiates each apnea is s t i l l u n k n o w n , however . Figure 1.10: Cross section of the carp midbrain showing a site in the mesencephalic tegmentum that appears to initiate episodes during episodic breathing. The site is shown in blue and is located approximately 3.5 to 4.5 mm below the dorsal surface of the optic lobes, just ventrolateral to the oculomotor nucleus. This cross section is approximately 1 mm rostral to the rostral border of the cerebellum. Abbreviations: CA = commissure ansulate, Cb = corpus cerebellum, LL = lemniscus lateralis, nG = nucleus glomerulosus, nLV = nucleus lateralis valvulae, nOC = oculomotor nucleus, TO = tectum opticum, V Cb = valvula cerebelli. (modified from Juch and Ballintijn, 1983) Midbrain Involvement in Breathing Pattern in Other Vertebrates In many other species o f vertebrates, descending inputs f rom the midbra in have also been found to influence the breathing pattern. F o r example , transecting just caudal to the optic chiasma (just rostral to the optic lobes) i n the isolated brainstem-spinal co rd o f the bul l f rog (Rana catesbeiana) changes the Act ive breathing pattern s ignif icant ly ( R e i d et al, 2000). W i t h the brainstem intact to the leve l o f the optic chiasma, the fictive breathing pattern is predominant ly composed o f breaths organized into episodes wi th on ly a very occas ional single 28 breath; after transecting caudal to the optic ch iasma the pattern consists o f predominant ly single breaths, as can be seen i n F igure 1.11 ( R e i d et al., 2000). T h i s change i n breathing pattern suggests that a midbra in site at the l eve l o f the optic ch iasma i n bullfrogs provides a descending input to the respiratory rhy thm generators o f the medu l l a that groups breaths into episodes ( R e i d et ai, 2000) . In decerebrate, unid i rec t ional ly venti lated bul lfrogs, transecting the midbra in at the l eve l o f the rostral or mid-opt ic tectum has on ly a transitory effect on the episodic breathing pattern but changes the overa l l respiratory frequency, suggesting that sites i n these regions play a role i n con t ro l l ing respiratory dr ive but do not direct ly control the breathing pattern (Gargag l ion i et al., i n preparation). H o w e v e r , transecting the caudal optic tectum permanently modif ies the episodic breathing pattern to one o f continuous breaths wh i l e l eav ing the overa l l breathing frequency unchanged, suggesting that sites i n the caudal midbra in are necessary to generate the episodic breathing pattern (Gargagl ion i et al., i n preparation). E lec t r i ca l s t imulat ion experiments also indicate that the caudal midbra in m a y be i n v o l v e d i n generating the episodic breathing pattern i n bul lfrogs, suggesting that the anteroventral tegmentum, the magnocel lu lar nucleus o f the torus semic icul la r i s , and the optic tectum or the nucleus i s thmi c o u l d poss ib ly produce an input that inhibi ts the respiratory rhy thm generators o f the medul la , ending each breathing episode (Gargag l ion i et al, i n preparation). These studies c o u l d not, however , identify any sites that appeared to p lay a role i n grouping breaths into episodes or ini t ia t ing episodes (Gargag l ion i et al, i n preparation). H o w e v e r , a more recent study on the isolated brainstem-spinal cord o f the bu l l f rog has shown that al though transecting the isthmus rhombencephal i f rom the medu l l a dramat ical ly alters the episodic breathing pattern, it does not el iminate episodes altogether (Chatburn, 2004). R e m o v i n g the isthmus reduces the frequency o f breathing episodes, the spontaneous breathing frequency wi th in episodes, and the distr ibution o f episodes over t ime, thus m a k i n g the duration and arrangement o f episodes m u c h more prolonged and random than i n the normal episodic breathing pattern (Chatburn, 2004). T h i s evidence suggests that al though the isthmus is c lear ly important for generating the normal episodic breathing pattern, i t is not essential for the product ion o f episodes (Chatburn, 2004). In conc lus ion , i n bul l f rogs, al though the caudal midbra in is not absolutely essential for the product ion o f episodes, it is c r i t i ca l for generating the normal episodic breathing pattern. 29 30 s Figure 1.11: Effects on the fictive breathing pattern of transecting the isolated bullfrog brainstem-spinal cord just caudal to the optic chiasma. This trace shows the integrated neural discharge recorded from the root of the vagus nerve of an adult bullfrog (Rana catesbeiana) isolated brainstem-spinal cord preparation. The black arrow indicates the time of transection. (Reid et al., 2000) Perfusing tadpole isolated brainstems w i t h baclofen, a G A B A B receptor agonist, converts the episodic f ict ive breathing pattern into a pattern o f most ly single breaths (Straus et al., 2000) . Th i s result indicates that pathways dependent on G A B A B receptors appear to p lay an important role i n cluster ing breaths into episodes (Straus et al., 2000). In the ca iman (Caimen sclerops), transection o f the midbra in at the caudal border o f the optic lobes had no effect on the episodic breathing pattern, however , transecting the medu l l a at the leve l o f the rostral border o f the nucleus laminaris converted the breathing pattern to one o f single breaths (Naifeh et ai, 1971a). These results suggest that i n reptiles, the sites con t ro l l ing the episodic breathing pattern are located i n the rostral region o f the medul la , just caudal to the caudal border o f the optic lobes, a region s imi lar to the pons o f birds and mammals ( M i l s o m et ai, 2004). Interestingly, the episodic breathing pattern i n caimans can also be converted to a pattern o f single, isolated breaths wi th l ight anesthesia (Naifeh et al., 1971b). In mammals , the pons also appears to p lay an important ro le i n respiration. F o r example , i n cats, a site i n the rostrolateral tegmentum at the pontomesencephalic border controls the duration o f inspirat ion by ending inspiratory act ivi ty i n each respiratory c y c l e (Fung and St. John, 1994). The duration o f expirat ion is contro l led by a more caudal pontine site w h i c h is located i n the nucleus parabranchialis medial is and K o l l i k e r - F u s e nucleus (Fung and St. John , 1994). E lec t r i ca l s t imulat ion o f the rostral pons alters the respiratory rhy thm and causes voca l iza t ion i n decerebrate cats ( Y a m a n a k a et al, 1993), suggesting that this region alters the breathing pattern to integrate respiration wi th other activit ies. B i l a t e ra l lesions i n the 30 pneumotaxic system o f the rostral pons o f cats anesthetized w i t h pentobarbital can even cause episodic breathing (Webber and Speck, 1981). In rats, the rostral pons also appears to be i n v o l v e d i n ending inspirat ion dur ing each respiratory cyc l e ; it does this v i a an N M D A receptor-mediated mechan i sm (Fung et al., 1994). The lateral parabranchial and K o l l i k e r - F u s e nuc le i o f the rostral pons appear to terminate inspira t ion i n the golden-mantled ground squirrel (Spermophilus lateralis), also v i a a process mediated by N M D A receptors (Harr is and M i l s o m , 2003) . A l t h o u g h the pons o f mammals certainly appears to be i n v o l v e d i n episodic breathing, it is not necessary for the product ion o f episodes; rat and hamster isolated brainstem-spinal cords can produce f ict ive breathing episodes even wi thout the pons (Corcoran , A and M a r s h a l , L . personal communica t ion) . The Role of the Mesencephalic Tegmentum in Breathing Pattern in Rainbow Trout and Carp: Hypothesis C a r p tend to breathe episodica l ly when respiratory dr ive is normal or l o w , on ly swi t ch ing to a continuous breathing pattern as respiratory dr ive increases (Juch and B a l l i n t i j n , 1983; L u m s d e n , 1996). Trout , however , no rmal ly breathe cont inuously , on ly swi tch ing to an episodic breathing pattern when respiratory dr ive decreases substantially (Randa l l and Jones, 1973; W o o d and Jackson, 1980). Therefore, I w o u l d expect that when no rmox ic , the majori ty o f trout should breathe cont inuously w h i l e the majority o f carp should breathe episodica l ly . W h e n hyperoxic , a l l carp should breathe episodica l ly , as should at least some trout. I hypothesize that descending inputs and sensory feedback act on the respiratory rhy thm generators i n the medu l l a to produce these breathing patterns i n carp and trout. Because o f evidence f rom previous studies on other vertebrate species, I predict that the caudal midbra in is i n v o l v e d i n coordinat ing this episodic breathing pattern. Based o n the evidence that a mesencephal ic tegmental site i n carp is i n v o l v e d i n ending each apnea by s t imulat ing a breathing episode (Juch and B a l l i n t i j n , 1983), I conc lude that this midbra in site plays an important role i n the product ion o f the episodic breathing pattern. M y goa l i n these studies was to microinject ka in ic ac id into the dorsal mesencephal ic tegmental site d iscovered by Juch and B a l l i n t i j n (1983), w h i c h I w i l l refer to as the J A B site, to investigate the effects o f destroying i t o n the episodic breathing pattern o f carp and trout. B y measur ing respiratory signals i n n o r m o x i a and hyperoxia , both before and after the injections, I attempted to 31 determine the effects o f this site on the breathing patterns o f these f ish w i th both normal and l o w levels o f respiratory drive. Because this site appears to end each apnea by s t imulat ing breathing episodes i n ep i sod ica l ly breathing carp (Juch and B a l l i n t i j n , 1983) but is not necessary for continuous breathing (Hukuhara and Okada , 1956), I expected that destroying i t w o u l d affect episodic but not cont inuous breathing. Therefore, I proposed that destroying this dorsal mesencephal ic tegmental site i n cont inuously breathing f ish should have no effect o n respiration. Howeve r , I proposed that destroying this site i n ep isodica l ly breathing f ish w o u l d either el iminate apneas or lengthen the duration o f apneas. M y first hypothesis was that destroying the J A B site w i t h ka in ic ac id w o u l d el iminate apneas, convert ing the episodes to continuous breathing. W h e n carp are l ight ly anesthetized, they do not breath ep isodica l ly and the J A B site does not discharge rhy thmica l ly (Juch and B a l l i n t i j n , 1983.) S i m i l a r l y to carp, ca iman (Caiman sclerops) normal ly breathe ep isodica l ly , but when l igh t ly anesthetized breathe i n single, isolated breaths (Nai feh et al, 1971b); ca imen also produce this breathing pattern o f single, isolated breaths after transection o f the rostral medul la , just rostral to the l eve l o f the nucleus laminaris (Nai feh et al, 1971a). Th i s evidence suggests that anesthesia inhibi ts the J A B site i n carp, and its equivalent i n ca imen , thus e l imina t ing episodic breathing. Al te rna t ive ly , it is possible that anesthesia inhibi ts both episodic breathing and the J A B site independently. H o w e v e r , because I have found no evidence suggesting that teleosts breathe ep isodica l ly without the midbra in and because the caudal midbra in appears to be essential i n generating the normal episodic breathing pattern i n amphibians ( R e i d et al, 2000; Chatburn, 2004; Ga rgag l i on i et al, i n preparation), I propose that i n teleosts, the J A B site is necessary to generate the normal episodic breathing pattern. Therefore, destroying the J A B site i n ep isodica l ly breathing carp and trout should el iminate the episodic breathing pattern. M y second hypothesis was that i f destroying the J A B site d i d not el iminate apneas, it w o u l d instead lengthen the duration o f the apneas i n the episodic breathing pattern. Juch and B a l l i n t i j n (1983) found that cont inuously s t imulat ing the J A B site shortened apneas by in i t ia t ing breathing episodes, but d i d not prevent apneas entirely. Because s t imulat ing the J A B site shortened the duration o f apneas, it m a y also, i n the intact f ish , shorten apneas f rom the duration they w o u l d be i f they were regulated exc lus ive ly by sensory feedback. I f the ini t ia t ion o f breathing episodes were regulated on ly by sensory feedback and not by a rhy thmic st imulus f rom 32 the midbra in , then the length o f apneas should increase i n length dependent on the l eve l o f d isso lved oxygen available i n the water. A l t h o u g h carp already adjust the duration o f apneas depending on the l eve l o f avai lable d isso lved oxygen (Juch and B a l l i n t i j n , 1983; L u m s d e n , 1996), destroying the J A B site should enhance this effect, lengthening the durat ion o f apneas i n both ep i sod ica l ly breathing carp and trout. In conc lus ion , I propose that destroying the J A B site should either e l iminate episodes or lengthen the duration o f apneas i n ep isodica l ly breathing carp and trout w h i l e not affecting respirat ion i n cont inuously breathing carp and trout. 33 Chapter 2: Do Decerebrate/Spinalectomized Fish Breathe Normally ? Introduction T o study the role o f the mesencephalic tegmentum on the breathing pattern o f trout and carp, I required a f ish w i t h an exposed midbra in that w o u l d remain motionless i n a stereotaxic device but that w o u l d s t i l l breathe wi th the same rhy thm as an intact f ish . Therefore, m y in i t i a l goa l was to create such a preparation. It was essential that these experiments be performed on l i v i n g animals , animals that were either anaesthetized or decerebrate so that they d i d not suffer pain . Anesthet ics , however , can interfere w i t h nerve function (Bur leson and M i l s o m , 1993), and w i t h the generation and modi f ica t ion o f respiratory rhy thm and pattern. In fact, trout anesthetized w i t h 6 0 m g / L benzocaine do not breathe at a l l . H o w e v e r , some fish do continue to breathe even when anesthetized; for example , tench l igh t ly anesthetized wi th urethane s t i l l breathe (Hughes and Shel ton, 1962), as do channel catfish anesthetized w i t h 20 m g / L o f M S - 2 2 2 (Bur leson and Smi th , 2001). I found that some carp anesthetized wi th 4 5 m g / L o f benzocaine w o u l d also occas ional ly make weak breathing movements; however , this breathing was abnormal , both i n frequency and pattern. In fact, carp do not breathe ep isodica l ly when exposed to even l ight doses o f anesthetic (Juch and B a l l i n t i j n , 1983), ind ica t ing that al though anesthesia does not totally inhib i t breathing, i t interferes w i th the product ion o f normal breathing patterns. Therefore, because m y goal was to study breathing patterns, I chose to investigate decerebrate fish. T o establish whether decerebrate trout and carp breathed normal ly , I needed to compare the breathing o f decerebrate fish to that o f intact fish. Because any movements o f the fish dur ing the experiment w o u l d interfere w i t h microinject ions w i th in the bra in , I also required a fish that w o u l d remain comple te ly para lyzed i n a l l regions caudal to the respiratory pump. A l t h o u g h it w o u l d have been possible to paralyze the fish us ing an intramuscular injection o f tubocurarine (1.5 mg/kg) (Juch and B a l l i n t i j n , 1983), the paralysis w o u l d not have been selective, rendering the respiratory pump ineffective also, so I chose to spinalectomize the fish to select ively paralyze the body caudal to the pectoral fins (Juch 34 and B a l l i n t i j n , 1983). Therefore, I also needed to compare the breathing rhythm o f spinalectomized f ish to that o f intact f i sh to ensure that they were not s igni f icandy different. Thus , I compared breathing i n intact trout to that i n both decerebrate and decerebrate/spinalectomized trout. Because I d i d not want to subject the trout to pa in , I c o u l d not study spinalectomized f ish without decerebrating them. T o compare the respiratory rhythms o f these f ish under different levels o f respiratory dr ive , I studied each f ish at each leve l o f surgery (intact, decerebrate, and decerebrate/spinalectomized) i n normoxia , hypox ia , and hyperox ia (Series 1). I also compared breathing i n decerebrate/spinalectomized trout and carp i n n o r m o x i a and hyperox ia (Series 2) to literature values for intact f ish. Methods Experimental Animals Trout o f both sexes were obtained f rom a l o c a l hatchery and raised i n an outdoor tank wi th a constant f l o w o f dechlorinated V a n c o u v e r water. Ca rp o f both sexes were obtained f rom a commerc i a l f isherman i n the Okanagan reg ion (Br i t i sh C o l u m b i a ) and kept i n a second outdoor tank, also wi th a constant f l ow o f dechlorinated V a n c o u v e r water. T h e water temperature i n both tanks was kept at approximately 10 degrees Ce ls ius . The f ish were fed three times w e e k l y wi th commerc i a l l y purchased pellets. I on ly performed surgeries on fish that had been fasted for at least 24 hours. T h e trout var ied i n weight f rom 200 g to 800 g (mean o f 611 ± 67 g) w h i l e the carp weighed f rom 240 g to 560 g (mean o f 402 ± 25 g). Instrumentation and Surgery T o instrument each fish, I first anesthetized it w i t h 6 0 m g / L benzocaine (dissolved i n 9 5 % ethanol) i n the c i rcu la t ing water. W h i l e the fish was anesthetized, I ar t i f ic ia l ly ventilated it by insert ing a hose into its mouth so that water f l owed cont inuously over its g i l l s . I p laced the fish i n a foam dish to support its body dur ing surgery. I instrumented the fish w i th both an opercular cavi ty pressure cannula and a bucca l cavi ty pressure cannula. T o cannulate the opercular cavi ty , I punctured a hole i n one opercular flap and inserted a piece o f P E 1 0 0 polyethylene tubing into the cavi ty , f lar ing the inside end such that it c o u l d not be pu l l ed out. I then secured the cannula 35 to the outside of the opercular flap using surgical thread. I connected the far end of the cannula to a Deltran IV disposable pressure transducer (Utah Medical Products Inc.). To cannulate the buccal cavity, I punctured a hole through the snout of the fish, then inserted a piece of flared polyethylene tubing (PE240) dorsally through the hole such that the flared end of the tube was firmly anchored against the roof of the mouth. I fed the pressure cannula, a 60cm piece of polyethylene tubing (PE100), ventrally through this larger tube into the buccal cavity and punctured the end with several smaller holes to make it more sensitive to pressure changes. To hold the cannula in place within the PE240 tubing, I tied a piece of surgical thread firmly around the dorsal end of the PE240 tube where it protruded from the snout of the fish. I connected the far end of the cannula to a second Deltran IV disposable pressure transducer (Utah Medical Products Inc.). Additionally, I instrumented fish in the Series 1 experiments with impedance probes. I sutured an impedance probe from a UFI model 2991 Impedance Converter onto both opercular flaps, beside the outermost edge. Because the probes are directly across from each other, they generate a current between them that flows through the head of the fish. The tissues of the fish provide a resistance (impedance) to this current, which the probes measure. As the fish moves its opercular flaps, the distance between the probes increases and decreases, causing the impedance to the current to also change. The probes measure these changes in impedance and can thus be used to record the frequency and relative amplitude of the movements of the opercular flaps. A photograph of a trout instrumented with both impedance probes and buccal and opercular pressure cannulae is shown in Figure 2.1. Figure 2.1: Photograph of a trout (Series 1) instrumented with impedance probes and an opercular cavity pressure cannula, as well as a buccal cavity pressure cannula. 36 Decerebration and Spinaleciomy T o decerebrate the f ish, I m o v e d it to the experimental tank, a 5 0 c m x 10cm x 10cm b lack plastic box , and c lamped it into the stereotaxic device (Nar ishige Scient i f ic Instrument Laboratory , Japan), f i x i n g its head i n place by the orbi tal r idges. W h i l e secured i n this tank, the mouth and g i l l s o f the f ish were complete ly submersed i n a steady f l o w o f aerated water; however , its dorsal surface was exposed to air to prevent water f rom entering the surgical inc is ions . I p laced sponges on either side o f the f ish to stabilize the rest o f its body and restrict any spontaneous movements o f the s w i m m i n g muscles that might force its head out o f the stereotaxic device. T o ventilate the f ish w h i l e it was unconscious, I directed the water f l ow i n the experimental tank into its mouth wi th a hose such that water f l owed cont inuously over its g i l l s . I cut away the sk in and tissue o n the dorsal surface o f the head to expose the sku l l . U s i n g a M A C A N electrosurgerical unit ( M V - 8 ) to m i n i m i z e b leeding , I r emoved the muscle tissue direct ly caudal to the sku l l , expos ing the vertebral c o l u m n . I then cut into the s k u l l and removed the ove r ly ing fat layer to expose the bra in f rom the frontal lobes back to the caudal border o f the cerebel lum, leav ing the meninges intact. I transected the bra in between the telencephalon and the tectum mesencephalon us ing the electrosurgical unit to prevent excessive b l o o d loss, then removed the forebrain wi th forceps and cotton. T o keep the bra in tissue nourished, as w e l l as to prevent desiccat ion, I submersed the bra in i n ar t i f ic ia l cerebrospinal f l u id ( A C S F ) containing, i n g / L , 7.3 N a C l , 0.26 K C 1 , 0.39 C a C l 2 , 4.3 glucose, 20.5 sucrose, and 0.14 N a 2 H P 0 4 , p H balanced to approximately 7.4 (Bur ton , 1975). I observed the transection wi th a microscope to ensure that no v i s ib le remnants o f the forebrain remained. 37 Figure 2.2: Photograph of a decerebrate/spinalectomized trout clamped into the experimental tank with the stereotaxic device and held immobile by body sponges. This trout is instrumented with buccal and opercular pressure cannulae but not impedance probes. The optic lobes and cerebellum are visible, as is the cotton just rostral to the optic lobes, used for clotting blood after the decerebration. A four-barreled glass micropipette (see Chapter 3) is positioned at the caudal border of the optic lobes, just lateral to the midline. To the right is a labeled picture of the dorsal view of a decerebrate/spinalectomized trout brain (the optic lobes are rostral and the spinal cord is caudal). The white scale bar shows 10 mm. T o spinalectomize the f ish , I exposed its spinal cord by cutting through the dorsal surface o f the vertebral co lumn . I transected the spinal cord approximately 2 to 3 vertebrae caudal to the sku l l . I first made this cut wi th a scalpel , then went over i t again wi th the electrosurgerical blade to ensure that the cord was comple te ly severed, as w e l l as to m i n i m i z e bleeding. Af ter pack ing the exposed region w i t h cotton to promote b l o o d clot t ing, I submersed the spinal cord i n A C S F to prevent desiccat ion. Figure 2.2 shows a decerebrate/spinalectomized trout c lamped into the experimental tank wi th the stereotaxic device. Measuring Respiratory Signals T o measure breathing in the f ish , I recorded the changes i n pressure i n the bucca l and opercular cavit ies . These measurements a l l owed me to observe both the frequency o f respiration and the relative ampli tude o f breaths. I measured the changes i n bucca l and opercular cavi ty 38 pressure us ing Del t ran I V disposable pressure transducers (Utah M e d i c a l Products Inc.) as described above. I ampl i f ied and filtered the bucca l pressure s ignal us ing a G o u l d U n i v e r s a l A m p l i f i e r set to direct current w i th the h igh cutoff set at 1 0 H z and the l o w cutoff set at the dc offset. I ampl i f ied and filtered the opercular pressure s ignal us ing a G o u l d Integrating A m p l i f i e r set o n direct mode. In Series 1 experiments, I also measured the movements o f the opercular flaps us ing a U F I M o d e l 2991 Impedance Conver ter as described above. T o ampl i fy and filter the impedance s ignal I used a G o u l d D C A m p l i f i e r w i t h the l o w pass filter set at 5 H z . I col lected a l l these ampl i f ied , fi l tered signals us ing a D I - 7 2 0 Series data acquis i t ion system (Dataq Instruments) at a sample rate o f 500 samples/second and v i ewed them us ing W i n D a q P r o software (Dataq Instruments). T o calibrate the pressure signals for ampli tude, I measured k n o w n pressures o f water i n a ver t ical c o l u m n , elevated such that 0 c m t b O was l eve l w i th the water surface o f the experimental tank. T h e impedance converter does not direct ly measure the movements o f the opercular flaps, but rather measures the changes i n resistance between the opercular flaps o f the fish, a value that includes such variables as the thickness o f muscle and bone between the probes and thus varies for each fish tested, both because o f i nd iv idua l variat ion and because o f sl ight differences i n the pos i t ion o f the probes on each animal . Therefore, I c o u l d not calibrate the impedance signals precisely for amplitude. Instead, I compared a l l impedance ampli tude values for each ind iv idua l fish to those o f that same fish w h i l e intact and no rmox ic , us ing each fish as its o w n control to generate relative ampli tude values. T o measure breathing at different levels o f respiratory dr ive , I var ied the l eve l o f the oxygen d isso lved i n the water. I generated no rmox ic water by bubb l ing air through the c i rcu la t ing water i n a gas exchange c o l u m n ; normoxic water had a d isso lved oxygen part ia l pressure (PO2) o f approximately 115 m m H g . I bubbled pure nitrogen through the gas exchange c o l u m n to create h y p o x i c water that s tabi l ized at a PO2 o f approximately 15 m m H g after about ten minutes. T o generate hyperoxic water, I bubbled pure oxygen through the gas exchange c o l u m n ; after about ten minutes, the water s tabi l ized at a PO2 o f approximately 500 m m H g . I measured the water PO2 us ing a T y p e E 5 0 4 6 PO2 electrode i n a type D 6 1 6 Thermostatted C e l l and the P H A 9 3 4 PO2 module o f a P H M 7 1 M k 2 acid-base analyzer (Radiometer, Denmark) . 39 Experimental Protocol: Series 1 Series 1 experiments compared the breathing o f decerebrate and decerebrate/spinalectomized trout to that o f intact trout. Af te r instrumenting the trout w i t h impedance probes and bucca l and opercular pressure cannulae, I swi tched the water supply f rom benzocaine-infused water to p la in dechlorinated tap water and m o v e d the trout into the experimental tank. O n c e the trout recovered f rom the anesthesia such that i t spontaneously made rhythmic breathing movements , I r emoved the hose f rom its mouth , c l amped d o w n the tank l i d , and covered it w i th a towel so that it c o u l d fu l ly recover f rom the anesthetic i n a dark, quiet environment. A p p r o x i m a t e l y two hours later, when the trout had fu l ly recovered f rom the anesthetic, I recorded its opercular cavi ty pressure and opercular flap movement for thirty minutes i n normoxia . I then exposed the trout to thirty minutes o f hypox ia , f o l l o w e d b y thirty minutes o f hyperoxia , wh i l e measuring its response by recording opercular cavi ty pressure and opercular flap movement . F i n a l l y , I bubbled air through the gas exchange c o l u m n , returning the water to no rmox ic levels , and recorded its opercular cavi ty pressure and opercular flap movements for another thirty minutes. I calculated a l l breathing frequency and ampli tude values f rom measurements recorded twenty-five minutes after changing the l eve l o f d isso lved oxygen to a l l ow the trout t ime to adjust to each new leve l o f oxygen avai labi l i ty . I made these measurements on the intact trout w h i l e it was unrestrained i n the experimental tank, w i th the l i d c lamped d o w n to prevent escape as w e l l as to provide a dark, quiet environment. Once I had f inished recording breathing i n the intact trout, I reanesthetized i t w i t h 60 m g / L benzocaine d issolved i n the c i rcu la t ing water, c l amped i t into the stereotaxic device i n the experimental tank, exposed its brain , and decerebrated it. Af te r comple t ing the decerebration, I swi tched the water c i rcula t ion to the fresh water supply and observed the trout unt i l it started spontaneously breathing, at w h i c h point I r emoved the hose f rom its mouth and a l lowed i t to continue breathing without assistance. I covered the tank and a l lowed the trout to recover for approximately two hours. W h e n the trout had recovered f rom the surgery and anesthetic, I measured its breathing i n normoxia , hypox ia , and hyperox ia as described for the intact f ish, w i th one except ion. O n c e the trout was decerebrate, I kept it restrained i n the stereotaxic device to prevent the bra in f rom being exposed to water, so I col lected a l l the breathing measurements for the decerebrate and decerebrate/spinalectomized trout f rom fish he ld i n the stereotaxic device. 40 Afte r recording breathing i n the decerebrate trout, I spinalectomized it as described above. I covered the tank for at least ten minutes after the spinalectomy to a l low the f i sh to recover f rom the shock o f the surgery. Because the trout was already'decerebrate and cou ld not suffer pa in , I d i d not need to anesthetize it for the spinalectomy as I had for the previous surgeries. Therefore, it d i d not have to recover f rom anesthetic and so d i d not require as m u c h time to recover f rom the spinalectomy as f rom the previous surgeries. I determined that i t had fu l ly recovered f rom the surgery when the breathing signals s tabi l ized to a steady leve l . O n c e the pressure and impedance signals s tabi l ized, I recorded breathing i n the decerebrate/spinalectomized trout i n no rmoxia , h y p o x i a , and hyperoxia , as described for the decerebrate trout. Experimental Protocol: Series 2 Series 2 experiments compared the breathing o f decerebrate/spinalectomized trout and decerebrate/spinalectomized carp i n n o r m o x i a and hyperoxia . Af te r instrumenting the f ish , I immedia te ly c l amped it into the stereotaxic device i n the experimental tank as described above. I decerebrated the f ish , then immedia te ly spinalectomized it. Af ter comple t ing these surgeries, I swi tched the water c i rcula t ion to fresh water, covered the f ish , and a l l owed it to recover overnight. Af te r the f ish had recovered f rom the surgery and anesthetic, I measured the changes i n bucca l and opercular pressure for at least ha l f an hour to establish a baseline for the quiet ly breathing, no rmoxic f ish . I then exposed the f ish to hyperox ia for ninety minutes. I measured the in i t i a l respiratory response to hyperox ia by recording breathing dur ing the first f ive minutes o f hyperoxia . Af te r ninety minutes o f bubb l ing pure oxygen I again measured breathing to determine the respiratory effects o f pro longed exposure to hyperoxia . I then returned the f ish to no rmox ia , and after thirty minutes measured breathing again to ensure that pro longed hyperoxic exposure d i d not have a last ing effect on respiration. Data Analysis I measured breathing for each f ish w h i l e i t was intact, decerebrate, and decerebrate/spinalectomized (Series 1) and when it was exposed to each l eve l o f d isso lved 41 oxygen (both Series 1 and 2) so that I c o u l d direct ly compare respiration i n each state for each ind iv idua l . Therefore, each fish in its in i t i a l state served as a control wi th w h i c h I c o u l d compare that same fish after each experimental procedure. I v i e w e d and analyzed a l l s ignal traces us ing the W i n D a q W a v e f o r m B r o w s e r and A d v a n c e d C O D A S software package (Dataq Instruments Inc.). W h e n h igh frequency noise obscured the breathing traces, I fi l tered them us ing the A d v a n c e d C O D A S software package (Dataq Instruments Inc.) m o v i n g averages filter, set to take the average o f fifty samples. B y averaging every fifty samples, the m o v i n g averages filter r emoved the h igh frequency noise by conver t ing the or ig ina l breathing trace o f 500 samples per second to 10 samples per second. F o r each experimental step, I analyzed a representative trace o f at least 30 seconds, longer when breathing was irregular or episodic , for the frequency and the ampli tude o f breaths. F o r Series 1 experiments I col lec ted a l l breathing frequency and ampli tude data f rom the impedance traces as these p rov ided the cleanest signals. The data col lected f rom the pressure traces d i d not differ f rom the data col lected f rom the impedance traces, however . Fo r Series 2 experiments I co l lec ted a l l breathing frequency and ampli tude data f rom the bucca l pressure traces as these signals were cleaner than the opercular pressure signals. F o r both series, because o f the large ind iv idua l variat ion i n the baseline breathing ampli tude o f the fish, I converted each ampli tude measurement into the percent change from the in i t i a l , no rmoxic ampli tude measurement, us ing each fish as its o w n cont ro l . I calculated the percent change i n breathing frequency i n the same way . T o calculate total vent i la t ion, I mu l t ip l i ed these percent change i n frequency values w i t h the corresponding percent change i n ampli tude values for each fish, then d iv ided b y 100, to g ive a percent change i n total vent i lat ion value that combined breathing frequency and amplitude. I expressed a l l measurement values as the mean ± the standard error o f the mean. I no rmal ized a l l percent change data us ing the f o l l o w i n g modi f i ed arcsine transform function, x<1; y = ^ arcsin (x-1) + 1 l x>1; y=4x + 1" 4 2 where x represents each ind iv idua l data point and y represents the transformed data point . Da ta points less than 1 (100%) were arcsine transformed whi l e data points greater than or equal 42 to 1 were l inear ly transformed. I compared the breathing frequency, ampli tude, and total vent i la t ion i n each experimental condi t ion for the f ish i n each experimental series us ing a O n e W a y Repeated Measures A n a l y s i s o f Var iance ( A N O V A ) (SigmaStat 2.0) w i th an a lpha value o f 0.05. W h e n the A N O V A results indicated a significant difference, I determined w h i c h groups were s ignif icant ly different us ing an A l l Pa i rwise M u l t i p l e C o m p a r i s o n Procedure T u k e y Test or a M u l t i p l e Compar i sons versus C o n t r o l G r o u p Bonfe r ron i t-test, us ing the in i t i a l , no rmox ic measurements as the cont ro l group (SigmaStat 2.0). W h e n the data fai led the normal i ty requirements for a parametric test, I used a F r i edman Repeated Measures A N O V A on R a n k s (SigmaStat 2.0) fo l l owed by an A l l Pa i rwise M u l t i p l e C o m p a r i s o n Procedure T u k e y Test (SigmaStat 2.0) to analyze the data. Results Series 1: Respiration in Intact, Decerebrate, and Decerebrate/Spinalectomized Trout Intact, no rmox ic trout had a mean respiratory frequency o f approximately 72.5 ± 3.5 breaths/min. T h i s frequency increased s ignif icant ly i n h y p o x i a (p=0.046) to a mean o f approximately 80.5 ± 2.0 breaths/min and decreased s ignif icant ly i n hyperox ia (p<0.001) to a mean o f approximately 59.3 ± 4.0 breaths/min as is shown i n F igure 2.3. Twenty- f ive minutes after be ing returned to normoxia , the trout d i d not breathe s ignif icant ly differently than they d i d dur ing the in i t ia l no rmoxic test per iod. Occas iona l ly dur ing hyperox ia trout w o u l d start to breathe episodica l ly , w i t h periods o f breathing separated by b r i e f apneas, as can been seen i n F igure 2.4. In normoxia , intact trout had a mean breathing ampli tude o f approximately 2.6 ± 0 .8 CIT1H2O. Because these trout had such h igh variat ion i n breathing ampli tude between ind iv idua ls i n no rmoxia , ranging f rom 0.2 to 5.8 c m H 2 0 , it was more accurate to analyze the breathing ampli tude data as percent change f rom control i n n o r m o x i a for each i nd iv idua l trout than to analyze the absolute values. (The absolute values for breathing ampli tude are shown i n Tab le A . l i n the Append ix . ) H y p o x i c trout showed a tendency to breathe w i t h a larger breath ampli tude than normoxic trout (120 ± 19 % o f normoxic values), w h i l e hyperoxic trout had a smaller breath ampli tude (64 ± 5 % o f no rmoxic values), as can be seen i n F igure 2.5; however , these trends were not significant (p= 0.514 and p=0.067, respect ively) . In h y p o x i a , the trout 43 showed a tendency to increase their total vent i la t ion to 132 ± 26 % o f no rmoxic levels, as can be seen i n Figure 2.6; however , this trend was not significant (p=0.36). In hyperoxia , however , intact trout s ignif icant ly (p<0.05) decreased their total venti lat ion to 53 ± 5 % o f their normoxic levels . W h e n in i t i a l ly exposed to hypox ia , almost a l l trout struggled intensely for several minutes, taking large gulps o f water and thrashing their tails. Some trout cont inued to struggle for most o f the hypox ic exposure wh i l e others ca lmed and became s t i l l after the first few minutes o f hypox ia . Figure 2.3: Breathing frequency (breaths/min) of intact trout in normoxia, hypoxia, and hyperoxia. Dissolved oxygen levels were approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. The asterix (*) indicates breathing frequencies that are significantly different from that of initial normoxia. Trout were given twenty-five minutes to adjust to each level of oxygen availability before measurements were recorded, and were tested in normoxia both before (initial) and after (final) hypoxic and hyperoxic exposure. Sample size of nine trout. 44 normoxia hypoxia M hyperoxia 1cmH20 10 See Figure 2.4: Representative breathing traces of an intact trout in normoxia, hypoxia, and hyperoxia. Breathing traces were recorded from the pressure changes in one of the opercular cavities. Dissolved oxygen levels were approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. Fish were given at least twenty-five minutes to adjust to each level of oxygen availability before these traces were recorded. 45 Normoxia Hypoxia Hyperoxia Normoxia (initial) (Final) Figure 2.5: Percent change in breathing amplitude of intact trout from initial normoxia. Dissolved oxygen levels were approximately 115mmHg in normoxia. 15mmHg in hypoxia, and 5(X)rnmHg in hyperoxia. Trout were given twenty-five minutes to adjust to each new level of oxygen availability before measurements were recorded, and were tested in normoxia both before (initial) and after (final) hypoxic and hyperoxic exposure. The breathing amplitude of each trout in hypoxia, hyperoxia, and final normoxia were compared to the breathing amplitude of that same trout in initial normoxia, so that each trout served as its own control. Values were calculated from impedance measurements. Sample size of seven trout. 46 Normoxia Hypoxia Hyperoxia Normoxia (Initial) (Final) Figure 2.6: Percent change in total ventilation from normoxia in intact trout Dissolved oxygen levels were approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. Trout were given twenty-five minutes to adjust to each new level of oxygen availability before measurements were recorded, and were tested in normoxia both before (initial) and after (final) hypoxic and hyperoxic exposure. The total ventilation of each trout in hypoxia, hyperoxia, and final normoxia were compared to the breathing amplitude of that same trout in initial normoxia, so that each trout served as its own control. The asterix (*) indicates a total ventilation value that is significantly different than that of the trout in initial normoxia. Sample size of seven trout. B o t h decerebrate and decerebrate/spinalectomized trout showed a higher breathing frequency in no rmox ia than d id intact trout (Figure 2.7); however , on ly decerebrate/spinalectomized trout breathed s igni f icandy faster than intact trout (p=0.013). The breathing frequency o f normoxic decerebrate trout was approximately 79.4 ± 3 . 7 breaths/min wh i l e that o f decerebrate/spinalectomized trout was approximately 88.9 ± 3.8 breaths/min. Decerebrate trout decreased their breathing amplitude by 47 ± 11 % from that o f intact trout (p<0.001), wh i l e decerebrate/spinalectomized trout decreased their breathing amplitude by 59 ± 5 % (p<0.001) (Figure 2.8). Decerebrate trout showed a 4 2 ± 15 % decrease i n total vent i lat ion (p=0.008) compared to intact trout w h i l e decerebrate/spinalectomized trout showed a 51 ±9 % decrease (p=0.002) (Figure 2.9.) 47 Intact trout increased breathing frequency f rom n o r m o x i a (72.5 ± 3 . 5 breaths/min) to h y p o x i a (80.5 ± 2.0 breaths/min), w h i l e h y p o x i c decerebrate trout d i d not s igni f icandy change breathing frequency compared to normoxia . Decerebrate/spinalectomized trout decreased breathing frequency (p<0.05) from 88.9 ± 3.6 breaths/min i n n o r m o x i a to 57.9 ± 8.3 breaths/min i n h y p o x i a (Figure 2.7). H y p o x i a d id not s igni f icandy alter breathing ampli tude i n intact, decerebrate, or decerebrate/spinalectomized compared to normoxic values (Figure 2.8) (p=0.843, p=0.769, and p=0.08, respectively). Brea th ing ampli tude for decerebrate (p=0.015) and decerebrate/spinalectomized (p=0.001) trout i n h y p o x i a was s ignif icant ly lower than breathing ampli tude i n hypox ic intact trout. W h i l e intact and decerebrate hypox ic trout d i d not alter total vent i la t ion f rom no rmox ic values (p=0.66 and p=0.902, respectively), decerebrate/spinalectomized h y p o x i c trout decreased total vent i la t ion by 49 ± 8% from n o r m o x i a (p=0.037). In hypox ia , both decerebrate (p=0.024) and decerebrate/spinalectomized (p<0.001) trout had s ignif icant ly lower total vent i lat ion than h y p o x i c intact trout. A s d i d the intact trout, decerebrate trout struggled intensely dur ing the first few minutes o f their exposure to hypox ia , taking large gulps o f water and thrashing their tails. Af te r this in i t i a l response, many o f the trout ca lmed for the remainder o f the hypox ic exposure wh i l e others cont inued to struggle per iod ica l ly throughout hypox ia . Decerebrate/spinalectomized trout also responded to the in i t i a l exposure to h y p o x i a b y taking large gulps o f water, however , because they were spinalec tomized they were unable to thrash their tails. Decerebrate and decerebrate/spinalectomized hyperoxic trout d i d not alter their breathing frequency compared to no rmox ic values (Figure 2.7). S i m i l a r l y , intact, decerebrate, and decerebrate/spinalectomized trout d i d not alter breathing ampli tude i n hyperox ia compared to no rmox ic values (p=0.061, p=0.421, and p=0.254, respectively) (Figure 2.8). In hyperoxia , decerebrate/spinalectomized trout had a s ignif icant ly l ower breathing ampli tude (p=0.025) than hyperoxic intact trout; however , hyperoxic decerebrate trout d i d not s ignif icant ly alter breathing ampli tude (p=0.193) compared to hyperoxic intact trout. A l t h o u g h intact trout s ignif icant ly decreased total venti lat ion i n hyperoxia , (p<0.05), decerebrate and decerebrate/spinalectomized trout d id not (Figure 2.9). Nei ther decerebrate nor decerebrate/spinalectomized trout i n hyperox ia had a s ignif icant ly different total vent i lat ion than hyperoxic intact trout (p=0.237). Brea th ing patterns were not not iceably different between the three preparations. F o r example , dur ing hyperoxia , at least one o f the eight trout tested breathed ep isodica l ly i n a l l o f the three surgical stages (not shown). 48 c £ >» o c <D D D" 0 I L . CD 100 80 4 60 4 40 H 20 0 • Normoxia 1 • Hypoxia • Hyperoxia Intact Decerebrate Decerebrate/ Spinalectomized Figure 2.7: Breathing frequency (breaths/min) of intact, decerebrate, and decerebrate/spinalectomized trout in normoxia, hypoxia, and hyperoxia. Dissolved oxygen levels in were approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. Each trout was tested in normoxia, hypoxia, and hyperoxia while intact, decerebrate, and decerebrate/spinalectomized. The asterix (*) indicates a breathing frequency that is significantly different from the intact trout at the same oxygen level. The cross (*) indicates a breathing frequency that is significantly different from normoxia at the same surgical level. Sample size of eight trout. 49 Normoxia Hypoxia Hyperoxia Intact Decebrate Decerebrate/ Spinalectomized Figure 2.8: Percent change in the breathing amplitude of decerebrate and decerebrate/spinalectomized trout in normoxia, hypoxia, and hyperoxia from intact, normoxic trout Dissolved oxygen was approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. Breathing amplitude of each decerebrate and decerebrate/spinalectomized trout was compared to the breathing amplitude of that same trout when intact, so that each trout served as its own control. The asterix (*) indicates a breathing frequency that is significantly different from the intact trout at the same oxygen level. Sample size of seven trout. 50 s 180 -. C5 • —i • 160 -c 140 -> 13 120 -• o H 100 -fl 80 -DC fl C5 6 0 -p f l u 4 0 -w M i 2 0 -0 -Normoxia Hypoxia Hyperoxia Intact Decerebrate Decerebrate/ Spinalectomized Figure 2.9: Percent change in total ventilation of decerebrate and decerebrate/spinalectomized trout in normoxia, hypoxia, and hyperoxia from intact, normoxic trout. Dissolved oxygen was approximately 115mmHg in normoxia, 15mmHg in hypoxia, and 500mmHg in hyperoxia. Total ventilation of each decerebrate and decerebrate/spinalectomized trout was compared to the breathing amplitude of that same trout when intact, so that each trout served as its own control. The asterix (*) indicates a breathing frequency that is significandy different from the intact trout at the same oxygen level. The cross (*) indicates a breathing frequency that is significantly different from normoxia at the same surgical level. Sample size of seven trout. Series 2: Breathing in Decerebrate/Spinalectomized Trout and Carp in Normoxia and Hyperoxia Decerebrate/spinalectomized trout had a mean respiratory frequency o f approximately 85.0 ± 4.6 breaths/minute. Brea th ing frequency decreased by 15 ± 6 % i n the first five minutes o f hyperoxia (p=0.044); however , after ninety minutes o f hyperoxia , breathing frequency was no longer s ignif icant ly different f rom that in no rmox ia (p=0.172). In in i t ia l hyperoxia , breathing 51 frequency was approximately 73.3 ± 7.9 breaths/minute w h i l e i n pro longed hyperox ia breathing frequency was approximately 77.4 ± 10.0 breaths/minute. A m p l i t u d e decreased by 46 ± 5 % i n in i t i a l hyperox ia (p<0.001) and by 56 ± 5 % i n prolonged hyperox ia (p<0.001). To ta l vent i la t ion decreased by 55 ± 4 % i n the first f ive minutes o f hyperox ia (p<0.001) and by 58 ± 6 % after ninety minutes o f hyperox ia (p<0.001). Th i r ty minutes after be ing returned to no rmox ia , breathing frequency, ampli tude, and total vent i la t ion were not s ignif icant ly different than before hyperoxic exposure. F igure 2.10 shows the percent change f rom in i t i a l no rmox ic values for breathing frequency, ampli tude, and total venti lat ion. In normoxia , a l l nine trout breathed cont inuously, w h i l e i n both in i t i a l and pro longed hyperoxia , two o f the nine trout breathed episodica l ly w h i l e the remain ing seven cont inued to breathe cont inuously. In in i t i a l hyperoxia , f ive o f the nine trout showed w a x i n g and wan ing breathing for several minutes w h i l e i n pro longed hyperoxia , four o f the nine trout s t i l l occas iona l ly breathed wi th the w a x i n g and wan ing pattern, alternating wi th continuous breathing. F igure 2.11 compares a representative trace o f a trout breathing cont inuously i n n o r m o x i a but ep isodica l ly i n hyperox ia to a representative trace o f a trout breathing cont inuously i n both n o r m o x i a and hyperoxia . 52 Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Breathing Frequency Breathing Amplitude Total Ventilation Figure 2.10: Percent change in breathing frequency, amplitude, and total ventilation in decerebrate/spinalectomized trout in normoxia and hyperoxia. Po2 values were approximately 115mmHg in normoxia and 500mmHg in hyperoxia. Measurements were made for each trout in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. The breathing measurements of each trout in initial and prolonged hyperoxia and final normoxia were compared to the breathing measurements of that same trout in initial normoxia, so that each trout served as its own control. * indicates a measurement that is significandy different from the initial normoxic value. Sample size of 9 trout. 53 Prolonged Hyperoxia Prolonged Hyperoxia 1 cmH 20 20 seconds Figure 2.11: Representative respiratory traces comparing a decerebrate/spinalectomized trout breathing continuously in normoxia but episodically in hyperoxia (A) to a decerebrate/spinalectomized trout breathing continuously in both normoxia and hyperoxia (B). These traces were generated from buccal pressure signals for both trout, filtered using the Advanced CODAS software package (Dataq Instruments Inc.) moving averages filter. 54 Decerebrate/spinalectomized carp had a mean respiratory frequency o f approximately 50.6 ± 4.0 breath/minute. Th i s frequency decreased by 28 ± 9 % i n in i t i a l hyperox ia (p<0.001) and by 40 ± 7% i n prolonged hyperox ia (p<0.001). In in i t i a l hyperox ia , breathing frequency was approximately 37.5 ± 5.2 breaths/minutes wh i l e i n pro longed hyperox ia breathing frequency was 31.6 ± 4.3 breaths/minute. Th i r ty minutes after be ing returned to normoxia , breathing frequency was not s ignif icant ly different than before the hyperoxic exposure. In the first f ive minutes o f hyperoxia , breathing ampli tude decreased by 30 ± 8 % (p=0.129) w h i l e after ninety minutes it decreased by 33 ± 9 % (p=0.042). A m p l i t u d e in n o r m o x i a was not s ignif icant ly different after hyperoxic exposure. Tota l vent i lat ion decreased by 56 ± 4 % i n in i t ia l hype rox ia (p<0.05) and by 63 ± 5 % i n pro longed hyperox ia (p<0.05). Th i r ty minutes after be ing returned to normoxia , total vent i la t ion was not s ignif icant ly different than before hyperoxic exposure. F igure 2.12 shows the percent change f rom in i t i a l no rmox ic values for breathing frequency, ampli tude, and total vent i la t ion. In no rmox ia , 35.7 % o f the decerebrate/spinalectomized carp breathed ep isodica l ly w h i l e 14.0 % breathed w i t h a wax ing /wan ing breathing pattern. In the first f ive minutes o f hyperoxia , 71.4 % o f the carp breathed ep isod ica l ly w h i l e after ninety minutes o f hyperoxia , 92 .9% o f the carp were breathing in episodes. The proport ion o f carp breathing ep isod ica l ly i n hype rox ia was s ignif icant ly higher than the propor t ion o f carp breathing ep isod ica l ly in n o r m o x i a (p=0.006). Th i r ty minutes after be ing returned to normoxia , 42.9 % o f the carp breathed episodica l ly . In both n o r m o x i a and hyperoxia , many o f the carp alternated between several breathing patterns, ranging f rom clear episodes to "messy" episodes to w a x i n g and wan ing to continuous breathing. F igure 2.13 shows representative traces o f breathing pattern for three carp i n both n o r m o x i a and hyperoxia . F igure 2.14 compares the percentage o f decerebrate/spinalectomized carp and trout breathing ep isodica l ly i n no rmox ia and hyperoxia . 55 c3 •— | 120 © '•+3 •<PH c E o gfe 40 fi 53 Q c o -Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Breathing Frequency Breathing Amplitude Total Ventilation Figure 2.12: Percent change in breathing frequency, amplitude, and total ventilation in decerebrate/spinalectomized carp in normoxia and hyperoxia. Po2 values were approximately 115mmHg in normoxia and 500mmHg in hyperoxia. Measurements were made for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. The breathing measurements of each carp in initial and prolonged hyperoxia and final normoxia were compared to the breathing measurements of that same carp in initial normoxia, so that each fish served as its own control. * indicates a measurement that is significantly different from the initial normoxic value. Sample size of 14 carp. 56 Normoxia Prolonged Hyperoxia B Normoxia Prolonged Hyperoxia Normoxia Prolonged Hyperoxia 1 cmH20 20 seconds Figure 2.13: Representative respiratory traces comparing breathing patterns in decerebrate/spinalectomized carp. Carp A breathed continuously in normoxia but episodically in hyperoxia while carp B breathed episodically in both normoxia and hyperoxia. Carp C breathed with a waxing and waning pattern in normoxia and alternated between episodes and a waxing and waning breathing pattern in hyperoxia. These traces were generated from buccal pressure signals, filtered using the Advanced CODAS software package (Dataq Instruments Inc.). - t 57 m 100 C3 © 09 fl S3 n -C 09 «PN o 0) es +«> fl <y PN Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Figure 2.14: The percentage of decerebrate/spinalectomized carp and trout breathing episodically in normoxia and hyperoxia. Fish were classified as breathing episodically at each level of dissolved oxygen if they breathed in discrete episodes separated by clear apneas (lasting for at least the time required for three normal breaths) and maintained this breathing pattern for at least one minute or more of the representative trace analyzed for that level of dissolved oxygen. The Po2 was approximately 115 mmHg in normoxia and 500 mmHg in hyperoxia. The initial hyperoxia measurement was taken during the first five minutes of hyperoxia while the prolonged hyperoxia measurement was taken after ninety minutes of hyperoxic exposure. The second normoxia measurement was taken thirty minutes after the fish was returned to normoxia. Note that no trout breathed episodically under normoxic conditions. The asterix (*) indicates that the proportion of fish breathing episodically is significandy different than the proportion of fish breathing episodically in normoxia. Sample size of 14 carp and 9 trout. 58 Discussion M y goa l for these two series o f experiments was to create a decerebrate/spinalectomized f ish on w h i c h I c o u l d study the role o f the mesencephalic tegmentum i n the product ion o f episodic breathing patterns. Ca rp breathe ep isod ica l ly w i t h normal as w e l l as l o w levels o f respiratory dr ive ( L o m h o l t and Johansen, 1979; Juch and B a l l i n t i j n , 1983; Glass et al, 1991; L u m s d e n , 1996) w h i l e trout o n l y breathe ep isodica l ly w i t h extremely l o w levels o f respiratory dr ive (Randa l l and Jones, 1973; W o o d and Jackson, 1980). I wanted a decerebrate/spinalectomized f ish that w o u l d do the same. In the first series o f experiments, I compared the breathing o f decerebrate and decerebrate/spinalectomized trout to intact trout i n no rmox ia , hypox ia , and hyperoxia . I studied the effects o f h y p o x i a on intact, decerebrate, and decerebrate/spinalectomized trout purely out o f interest; because trout do not breathe ep isod ica l ly when hypox ic (Randa l l and Jones, 1973; W o o d and Jackson, 1980), i t was not necessary for me to understand the effects o f h y p o x i a on the decerebrate/spinalectomized trout. In the second series o f experiments, I quantif ied breathing i n decerebrate/spinalectomized trout and carp i n both n o r m o x i a and hyperoxia , then compared these values to those obtained for intact trout and carp f rom the literature. Intact Trout in Normoxia In normoxia , intact trout breathed cont inuously w i t h a frequency o f 72.5 ± 3.5 breaths/min, consistent w i th measurements f rom other studies (Cameron and D a v i s , 1970; R a n d a l l and Jones, 1973; W o o d and Jackson, 1980; Perry et al, 1992; Perry and G i l m o u r , 1996). Therefore, I can conclude that these intact trout breathed normal ly when compared to intact trout o f previous studies and were a v a l i d control for a study o f the effects o f decerebration and spinalectomy. Decerebrate Trout in Normoxia Decerebrate trout breathed s l ight ly more rap id ly than intact trout, but w i th s ignif icant ly decreased ampli tude. Th i s w o u l d suggest that the cerebrum may prov ide input to the breathing 59 centres o f the medu l l a that modif ies breathing pattern. W h e n removed, it leads to a decrease i n breathing amplitude and an increase in breathing frequency. T h e decrease i n breathing ampli tude more than compensated for the increase i n frequency, leading to a significant decrease i n total vent i lat ion compared to intact trout. Decerebrate trout breathed cont inuously i n no rmox ia , consistent w i t h intact trout i n this and i n previous studies (Randa l l and Jones, 1973; W o o d and Jackson, 1980). Therefore, decerebration does not appear to have a significant effect on breathing frequency or the continuous breathing pattern, suggesting that decerebrate trout w o u l d be acceptable for s tudying breathing pattern w i t h normal levels o f respiratory dr ive. Decerebrate/Spinalectomized Trout in Normoxia A l t h o u g h decerebrate/spinalectomized trout breathed s ignif icant ly faster than intact trout i n no rmox ia , they also s ignif icant ly decreased breathing ampli tude such that total vent i la t ion was also s ignif icant ly lower . Decerebrate/spinalectomized trout breathed cont inuously i n no rmox ia , i n agreement w i th previous f indings for intact trout (Randa l l and Jones, 1973; W o o d and Jackson, 1980). Therefore, I can conclude that al though decerebrate/spinalectomized trout breathe wi th l ower total vent i lat ion than intact trout, they show a normal continuous breathing pattern i n n o r m o x i a and w o u l d thus be acceptable for s tudying breathing pattern i n animals w i t h normal levels o f respiratory drive. Decerebrate/Spinalectomized Carp in Normoxia Decerebrate/spinalectomized carp had a breathing frequency o f 50.6 ± 4.0 breaths/minute i n no rmoxia . Brea th ing frequency o f no rmox ic , intact carp is approximately 20 breaths/minute (Takeda, 1990; Glass et al, 1991; L u m s d e n , 1996). A t least some o f this difference can be expla ined by differences i n PO2 levels ; no rmox ic PO2 i n the current study was approximately 115 m m H g w h i l e i n the previous studies it var ied f rom about 130 m m H g (Takeda, 1990) to 160 m m H g (Lumsden , 1996). H o w e v e r , even consider ing this difference i n avai lable oxygen levels , decerebrate/spinalectomized carp appear to breathe twice as rap id ly as intact carp. In fact, no rmox ic decerebrate/spinalectomized carp had a comparable breathing frequency (approximately 50 breaths/minute) to hypox ic intact carp (Glass et ah, 1991). In normoxia , on ly 35.7 % o f the decerebrate/spinalectomized carp breathed episodica l ly , w h i l e i n previous studies, 60 intact carp d i d not appear to breath cont inuously at PO2 levels higher than approximately 50 m m H g (Lomho l t and Johansen, 1979; Glass et al., 1991). These results suggest that decerebrate/spinalectomized carp experienced a higher leve l o f respiratory dr ive than intact carp. Juch and Ba l l i n t i j n (1983) found that sp ina lec tomiz ing carp just caudal to the skul l d id not affect respiration, al though they d i d not provide measurements o f breathing frequency or ampli tude to substantiate this, nor d i d they decerebrate their f ish. Th i s suggests, as it also d i d i n trout, that perhaps the cerebrum provides some source o f inhib i tory input to the respiratory centres that depresses respiratory dr ive , or at least breathing frequency, al though wi thout further studies this is pure ly speculative. Intact Trout in Hypoxia In hypox ia , al though the trout increased their breathing frequency s ignif icant ly , ampli tude was not s ignif icant ly different f rom no rmox ic values. Thus , al though intact trout showed a tendency to increase their total vent i la t ion f rom normoxic values, this trend was not significant. T h i s response to h y p o x i a was m u c h less robust than expected. Past studies have shown that trout typ ica l ly show a strong response to hypox ia , s ignif icant ly increasing their total vent i la t ion (Smi th and Jones, 1978; K i n k e a d and Perry, 1991; Perry and G i l m o u r , 1996). W h y then d i d these trout not show a stronger response to h y p o x i a ? I exposed these fish to a PO2 o f approximately 15 m m H g , w h i c h is a very severe hypox ia . A l t h o u g h a more severe h y p o x i a should theoretically induce a more pronounced response, after a certain leve l o f oxygen deprivat ion the phys i ca l performance o f the respiratory pump may become l imi ted . Trou t hemoglob in is on ly fifty percent saturated (P50) when water PO2 reaches 14 m m H g (Lumsden , 1996), severely reducing the oxygen del ivery to the tissues. Therefore, a h y p o x i a o f 15 m m H g c o u l d be severe enough to substantially l i m i t respiratory performance. Ca rp show a s imi lar response to progressive severe h y p o x i a ; they also increase breathing frequency unt i l a c r i t ica l PO2 (approximately 10 m m H g ) b e l o w w h i c h further reductions i n PO2 lead to a decrease i n breathing frequency (Glass et al., 1991). A d d i t i o n a l l y , severe h y p o x i a (water PO2 less than 45 m m H g ) can trigger the release o f catecholamines into the bloodstream (Perry et al., 1992), w h i c h i n turn w i l l depress vent i la t ion i n trout ( K i n k e a d and Perry, 1991; Perry et al., 1992). Therefore, the trout i n this study most l i k e l y showed such a weak response to h y p o x i a because the h y p o x i a was so severe. 61 Decerebrate Trout in Hypoxia Decerebrate trout, m u c h l i ke intact trout, showed a tendency to increase breathing ampli tude and total vent i la t ion when exposed to h y p o x i a , however , these trends also were not significant. U n l i k e intact trout, decerebrate trout d i d not increase breathing frequency i n h y p o x i a compared to i n normoxia . Intact trout typ ica l ly increase total vent i la t ion by increasing breathing ampli tude but not frequency ( K i n k e a d and Perry, 1991; Per ry and G i l m o u r , 1996), however , the overa l l weak response o f decerebrate trout to h y p o x i a most l i k e l y reflects the severity o f the h y p o x i a tested. Decerebrate/Spinalectomized Trout in Hypoxia Decerebrate/spinalectomized trout c lear ly showed a different response to h y p o x i a than d i d intact and decerebrate trout. W h i l e h y p o x i c intact trout increased breathing frequency, ampli tude, and total vent i la t ion, hypox ic decerebrate/spinalectomized trout decreased breathing frequency, ampli tude, and total vent i lat ion. W h y ? Sensory feedback to the respiratory centres i n the medu l l a should not have been decreased by the spinalectomy (Bamford , 1 9 7 4 A ; N i l s s o n , 1984; B u r l e s o n and M i l s o m , 1993; B u r l e s o n and M i l s o m , 1995). A l s o , w h i l e the degree o f invasiveness o f this surgery was greater, I kept decerebrate/spinalectomized trout a l ive for three or four days post-surgery and they s t i l l breathed normal ly i n no rmox ia , suggesting that even over several days there was no appreciable deterioration o f these f ish . It is possible that the decerebrate/spinalectomized trout suffered f rom a m i l d anemia; some trout tested m a y have lost up to f ive percent o f their total b l o o d vo lume dur ing the decerebration and spinalectomy. A f ive percent reduct ion i n b l o o d vo lume should not no rmal ly have such a severe effect on breathing performance. Previous studies suggest that i n normoxia , even severely anemic trout (b lood hematocrit reduced f rom 22.8 ± 1.5 % to 3.8 ± 0.9 %) show no significant change i n breathing frequency or ampli tude (Cameron and D a v i s , 1970) and i n the current study, i n n o r m o x i a and hyperox ia this m i l d anemia d i d not appear to affect breathing performance. H o w e v e r , a m i l d anemia c o u l d perhaps have had a more pronounced effect i n severe hypox ia , c o m p r o m i s i n g respiratory muscle performance further. 62 Intact Trout in Hyperoxia In hyperoxia , total vent i la t ion i n intact trout decreased s igni f icandy (by approximately 53 ± 5 % ) , consistent w i th previous studies ( W o o d and Jackson, 1980; K i n k e a d and Perry, 1991). In hyperoxia , the decrease i n total vent i la t ion was p r imar i l y caused by a decrease i n breathing frequency. A l t h o u g h breathing ampli tude tended to decrease i n hyperoxia , this trend was not significant. K i n k e a d and Perry (1991) found that the hyperoxic decrease i n total vent i la t ion was due to a decrease i n both breathing frequency and ampli tude w h i l e W o o d and Jackson (1980) found that the hyperoxic decrease i n total vent i lat ion was p r imar i ly due to a decrease i n breathing amplitude. T h e reasons for these differences i n pattern o f change are not clear. A p p r o x i m a t e l y twenty percent o f the intact trout breathed ep isod ica l ly dur ing hyperox ia w h i l e the rest cont inued to breathe cont inuously; this is consistent w i t h previous studies on intact trout ( W o o d and Jackson, 1980). In conc lus ion , these intact trout breathed normal ly i n hype rox ia when compared to intact trout i n previous studies, suggesting that they are a v a l i d control for studies o f the effects o f decerebration and spinalectomy at l o w levels o f respiratory dr ive. Decerebrate Trout in Hyperoxia Decerebrate trout tended to decrease their breathing frequency, ampli tude, and total vent i la t ion when exposed to hyperoxia , however these trends were not significant. T h i s unexpectedly smal l response m a y reflect the experimental pro tocol . Because the trout were first exposed to thirty minutes o f severe hypox ia , they may have accumulated an oxygen debt i n their tissues that c o u l d have kept their respiratory dr ive s l ight ly elevated. T h i s oxygen debt, combined wi th the possible minor anemia described above, c o u l d perhaps have caused this weak response to hyperoxia . T o fu l ly test the effects o f hyperox ia o n decerebrate trout, it w o u l d have been more useful to expose the fish direct ly to hype rox ia after no rmox ia without m a k i n g them hypox ic first. In the Series 2 experiments, where trout were exposed to hyperox ia wi thout first be ing exposed to severe hypox ia , decerebrate/spinalectomized trout decreased their total vent i la t ion s ignif icant ly , as discussed be low. 63 Decerebrate/Spinalectomized Trout in Hyperoxia Decerebrate/spinalectomized trout decreased their total vent i la t ion s ignif icant ly when exposed to hyperox ia (Series 2), when not first exposed to thirty minutes o f severe h y p o x i a (Series 1). W i t h i n the first f ive minutes o f exposure to hyperoxia , the decerebrate/spinalectomized trout had already reduced total vent i la t ion s ignif icant iy. Af te r ninety minutes o f hyperoxic exposure, total vent i la t ion had decreased by approximately 58 ± 6 % ; W o o d and Jackson (1980) found a s imi lar decrease o f approximately 60 % i n total vent i la t ion i n hyperoxic intact trout. In the current study, decerebrate/spinalectomized trout reduced their total vent i la t ion p r imar i ly b y decreasing breathing ampli tude, i n agreement w i t h results for hyperoxic intact trout i n previous studies ( W o o d and Jackson, 1980; K i n k e a d and Perry , 1991). A p p r o x i m a t e l y thirty percent o f decerebrate/spinalectomized trout breathed ep isodica l ly dur ing pro longed exposure to hyperox ia wh i l e the rest cont inued to breathe cont inuously . T h i s is also consistent w i th results for intact trout i n this and previous studies ( W o o d and Jackson, 1980). Therefore, I can conclude that decerebrate/spinalectomized trout breathe normal ly i n hyperoxia , and w o u l d thus be acceptable for s tudying breathing pattern at reduced levels o f respiratory dr ive. Decerebrate/Spinalectomized Carp in Hyperoxia Decerebrate/spinalectomized carp s ignif icant ly decreased their total vent i la t ion i n hyperox ia w i t h i n the first f ive minutes o f exposure and maintained this reduced leve l o f respirat ion throughout the entire ninety minutes. Intact carp also decrease total vent i lat ion by approximately 65 % dur ing prolonged exposure to hyperox ia (Takeda, 1990). In the current study, decerebrate/spinalectomized carp decreased their total vent i la t ion b y reduc ing both breathing frequency and ampli tude, al though previous studies found that intact carp decreased total vent i la t ion p r imar i ly by decreasing breathing frequency, showing no significant reduct ion i n breathing ampli tude (Takeda, 1990; L u m s d e n , 1996). A l m o s t a l l (approximately 93 percent) o f the decerebrate/spinalectomized carp breathed ep isodica l ly i n hyperoxia , consistent w i th previous studies (Lumsden , 1996). Those ind iv idua ls that were already breathing ep i sod ica l ly i n n o r m o x i a cont inued to breathe ep isodica l ly i n hyperoxia ; these carp reduced their total 64 vent i la t ion by decreasing the number o f breaths per episode as w e l l as the frequency o f episodes, again consistent w i t h previous studies (Lumsden , 1996). Thus , I can conclude that, apart f rom increased breathing frequency, decerebrate/spinalectomized carp breathe normal ly i n hyperox ia and w o u l d therefore be acceptable for s tudying breathing pattern at reduced levels o f respiratory dr ive. Conclusions In conc lus ion , decerebrate/spinalectomized trout breathe normal ly i n both n o r m o x i a and hyperox ia ( W o o d and Jackson, 1980; K i n k e a d and Perry, 1991) and w o u l d therefore be acceptable for use i n future studies on breathing pattern i n f ish w i th normal or l o w levels o f respiratory dr ive . Decerebrate/spinalectomized carp breathed more rapid ly than intact carp i n both n o r m o x i a and hyperoxia (Takeda, 1990; Glass et al, 1991; L u m s d e n , 1996), however their overa l l response to hyperox ia was consistent w i th that o f intact carp (Takeda, 1990). M o s t important ly, al though o n l y some decerebrate/spinalectomized carp breathed ep isod ica l ly i n normoxia , almost a l l breathed ep isod ica l ly i n hyperoxia . Therefore, I can conclude that hyperoxic decerebrate/spinalectomized carp w o u l d be acceptable for use i n future studies o f episodic breathing. 65 Chapter 3: The Role of the JAB Site in Coordinating Respiratory Pattern Introduction In vertebrates, respiratory needs vary depending on condit ions such as oxygen avai labi l i ty or act ivi ty l eve l . Therefore, vertebrates must adjust their breathing to meet these changes in respiratory dr ive. T o vary total vent i la t ion, vertebrates can adjust their breathing pattern, increasing or decreasing both the frequency and ampli tude o f breathing and changing the w ay i n w h i c h breaths are arranged over t ime. A l t h o u g h the most c o m m o n l y seen type o f breathing pattern i n mammals is continuous breathing ( M i l s o m , 1991), where one breath fo l lows another w i t h on ly a m i n i m a l pause i n between, vertebrates actually produce a wide range o f breathing patterns. In this study, I investigated episodic breathing patterns, where breaths are organized into groups (episodes) separated by periods wi thout breathing (apneas). Some species f rom a l l classes o f vertebrates breathe episodica l ly . F o r example , reptiles such as alligators (Alligator miss issippiensis) (Nai feh etal., 1971) and turtles (Pseudemys scripta) (V i t a l i s and M i l s o m , 1986) normal ly breathe i n episodes. A m p h i b i a n s such as bullfrogs (Rana catesbeiana) breathe ep isodica l ly when respiratory dr ive is elevated but not extremely h i g h ( M i l s o m , 1991). Some f ish , such as carp, no rmal ly breathe i n episodes ( L o m h o l t and Johansen, 1979; Juch and B a l l i n t i j n , 1983; L u m s d e n , 1996), w h i l e others, such as trout, o n l y breathe ep isodica l ly w h e n respiratory dr ive is extremely l o w (Randa l l and Jones, 1973; W o o d and Jackson, 1980). D u r i n g embryonic development, birds such as chickens breathe i n episodes (Fort in et al., 1995). E v e n mammals such as golden-mantled ground squirrels (Spermophilus lateralis) ( M i l s o m et al, 1997) and echidnas (Tachyglossus aculeatus) ( N i c o l and Ander son , 2003) breathe ep isodica l ly when hibernating. Because the episodic breathing pattern is so h igh ly conserved across a l l vertebrate classes, it seems l i k e l y that this breathing pattern is adaptive. Prev ious studies suggest that the episodic breathing pattern helps to reduce the mechanica l cost o f breathing when respiratory dr ive is l o w ( M i l s o m , 1984; V i t a l i s and M i l s o m , 1984; V i t a l i s and M i l s o m , 1986). 66 T h e basic respiratory rhythm produced i n the medu l l a is modi f i ed b y both chemosensory feedback and input f rom higher centres to produce the episodic breathing pattern. H o w e v e r , the episodic breathing pattern can be produced without either input ( R e i d and M i l s o m , 1998), suggesting that a l l the mechanisms necessary for p roduc ing this breathing pattern are located i n the bra in itself. In many vertebrates, sites i n the midbra in have been impl ica ted i n the cont ro l o f the episodic breathing pattern. F o r example , i n the bu l l f rog (Rana catesbeiana), the caudal midbra in appears to p lay an important role i n p roduc ing the normal episodic breathing pattern ( R e i d et al., 2000; Chatburn , 2004; Ga rgag l i on i et al., i n preparation). In reptiles such as the ca iman (Caimen sclerops), sites near the rostral border o f the medu l l a appear to produce the episodic breathing pattern (Naifeh et al., 1971a), w h i l e i n mammals the pons also appears to be i n v o l v e d i n control o f breathing pattern (Webber and Speck, 1981; F u n g and St .John, 1994; Har r i s and M i l s o m , 2003). In carp, a teleost f ish , a site i n the dorsal mesencephal ic tegmentum appears to p lay a role i n generating the episodic breathing pattern, ending each apnea by promot ing the next breathing episode (Juch and B a l l i n t i j n , 1983). I w i l l refer to this dorsal mesencephal ic tegmentum site as the J A B site i n honour o f its discoverers, Juch and B a l l i n t i j n . In this study, m y goal was to investigate the J A B site i n carp, as w e l l as its equivalent i n trout, to determine whether it was necessary to produce the episodic breathing pattern i n these f ish . Ca rp normal ly breathe episodica l ly , swi tch ing to continuous breathing on ly when their respiratory dr ive is increased (Lomho l t and Johansen, 1979; Juch and B a l l i n t i j n , 1983; L u m s d e n , 1996). Converse ly , trout no rmal ly breathe cont inuously , swi tch ing to episodic breathing on ly when their respiratory dr ive is extremely l o w (Randa l l and Jones, 1973; W o o d and Jackson, 1980). I measured breathing pattern i n decerebrate/spinalectomized carp and trout i n both n o r m o x i a and hyperox ia to establish a baseline o f normal breathing patterns at both normal and reduced levels o f respiratory drive. T h e n , to determine the effect o f r emov ing the J A B site on the episodic breathing pattern, I attempted to destroy this site w i t h microinject ions o f ka in ic ac id and measured the effect o f this on breathing pattern at normal and l o w levels o f respiratory dr ive. Because the J A B site is proposed to initiate each episode o f breathing, ending each apnea (Juch and B a l l i n t i j n , 1983), but is not necessary for continuous breathing (Hukuhara and O k a d a , 1956) I predicted that destroying it w i th ka in ic ac id should alter the episodic breathing pattern w h i l e hav ing no effect on the continuous breathing pattern. M o r e speci f ica l ly , I predicted that destroying the J A B site w i t h ka in i c ac id w o u l d either el iminate apneas or lengthen their duration. 67 M y first hypothesis was that destroying the J A B site w i th ka in ic ac id w o u l d el iminate apneas, conver t ing the episodic breathing pattern to continuous breathing. W h e n carp are l igh t ly anesthetized, they do not breath ep i sod ica l ly and the J A B site does not discharge rhy thmica l ly (Juch and B a l l i n t i j n , 1983.) S i m i l a r l y to carp, ca iman (Caiman sclerops) no rmal ly breathe ep isodica l ly , but when l igh t ly anesthetized breathe i n single, isolated breaths (Nai feh et al, 1971b); ca imen also produce this breathing pattern o f single, isolated breaths after transection o f the rostral medul la , just rostral to the l eve l o f the nucleus laminar is (Nai feh et al, 1971a). T h i s evidence suggests that anesthesia inhibi ts the J A B site i n carp, and its equivalent i n ca imen, thus e l imina t ing episodic breathing. O f course, it is also possible that anesthesia inhibi ts both episodic breathing and the J A B site independently. H o w e v e r , because I have found no evidence suggesting that teleosts breathe ep isod ica l ly wi thout the midbra in and because the caudal midbra in appears to be essential i n generating the normal episodic breathing pattern i n amphibians ( R e i d et al, 2000; Chatburn , 2004; Ga rgag l i on i et al, i n preparation), I hypothesized that i n teleosts, the J A B site is necessary to generate the normal episodic breathing pattern. Therefore, I predicted that destroying the J A B site i n ep i sod ica l ly breathing carp and trout should el iminate the episodic breathing pattern. M y second hypothesis was that i f destroying the J A B site d i d not el iminate apneas, it w o u l d instead lengthen the duration o f the apneas i n the episodic breathing pattern. Juch and B a l l i n t i j n (1983) found that cont inuously s t imulat ing the J A B site shortened apneas by in i t ia t ing breathing episodes, but d id not prevent apneas entirely. Because the J A B site is capable o f shortening apneas when stimulated, it suggests that the J A B site might also shorten apneas normal ly f rom the duration they w o u l d be i f they were regulated exc lus ive ly b y sensory feedback. I f the ini t ia t ion o f breathing episodes were regulated solely by sensory feedback and not by a rhy thmic st imulus f rom the midbra in , then the length o f apneas should increase i n length as the leve l o f d isso lved oxygen avai lable i n the water increased. Increasing oxygen avai labi l i ty already increases the length o f apneas (Lumsden , 1996), poss ib ly because the J A B site is sensitive to feedback f rom oxygen receptors, s t imulat ing more episodes when the l eve l o f d isso lved oxygen i n the inspi red water is lowered (Juch and B a l l i n t i j n , 1983). H o w e v e r , destroying this site should enhance this effect. Therefore, I predicted that i f destroying the J A B site d i d not e l iminate episodes, it w o u l d lengthen the duration o f apneas i n ep isodica l ly breathing carp or trout when compared to apneas before the destruction o f the J A B site. 68 Methods Instrumentation and Surgery I obtained and maintained the trout and carp as described i n Chapter 2 (pg.35). I anesthetized each f ish w i t h 60 m g / L benzocaine (in 9 5 % ethanol) d i sso lved i n dechlorinated water, then recorded its body mass and length. D u r i n g surgery I venti lated the f ish by insert ing a hose into its mouth to f lush its g i l l s w i t h aerated water containing 45 to 60 m g / L benzocaine to keep i t anesthetized. (To keep trout o f less than about 400 grams anesthetized, I required approximately 60 m g / L benzocaine, w h i l e for larger trout and carp I reduced the concentration to approximately 45 m g / L ) . I instrumented the trout w i t h both an opercular cavi ty pressure cannula and a bucca l cavi ty pressure cannula , as described i n Chapter 2 (pg.35). F igure 2.1 shows a trout instrumented wi th cannulae i n both its bucca l and opercular cavit ies. O n c e the f ish was instrumented, I c l amped it into the stereotaxic device (Nar ishige Scient i f ic Instrument Labora tory , Japan) in the experimental tank and exposed its b ra in as described i n Chapter 2 (pg.37). I then removed the cerebrum b y transecting the bra in at the rostral edge o f the optic lobes, us ing a M A C A N electrosurgerical unit ( M V - 8 ) to m i n i m i z e b leeding. T o keep the bra in tissue nournished, and to prevent dessication, I submersed the bra in i n ar t i f ic ial cerebrospinal f l u i d ( A C S F ) as described i n Chapter 2 (pg.37). O n c e the f ish was decerebrate, I transected the spinal cord at approximately 3 m m caudal to the obex, as described i n Chapter 2 (pg.37). F igure 3.1 shows the approximate locat ions o f the transection sites for the decerebration and spinalectomy on a d iagram o f the trout brain . F igure 3.2 shows a decerebrate, spinalectomized carp c lamped into the experimental tank w i t h the stereotaxic device and Figure 3.3 shows a labeled photograph o f the decerebrate/spinalectomized carp brain. 69 Figure 3.1: Diagram of a sagittal view of the trout brain with the approximate transection sites of the decerebration and spinalectomy. Transection sites are shown in red. Abbreviations: V = trigeminal nerve, VII = facial nerve, IX = glossopharyngeal nerve, X = .vagus nerve. (Modified from Meek and Nieuwenhuys, 1997) Figure 3.2: Photograph of a decerebrate/spinalectomized carp clamped into the experimental tank with the stereotaxic device. The carp is instrumented with a buccal and an opercular pressure cannula. The optic lobes, cerebellum, vagal lobes, and spinal cord are exposed. Cotton was used to control bleeding after the decerebration and is visible just rostral to the optic lobes. 70 Figure 3.3: Photograph of the dorsal view of a decerebrate/spinalectomized carp brain. Microinjection of Kainic Acid K a i n i c ac id (2-Carboxy-3-carboxymethyl -4- i sopropenylpyrro l id ine) is a glutamate agonist that k i l l s neurons b y exci ta t ion-induced apoptosis. K a i n i c ac id destroys neuron c e l l bodies but not axons o f passage (Bender and Ba ize r , 1984). I microinjected ka in ic ac id into the midbrains o f carp and trout to destroy the dorsal mesencephalic tegmental site d iscovered by Juch and Ba l l in t i j n (1983) to determine its effect on episodic breathing. Juch and Ba l l in t i j n (1983) described the locat ion o f this site, w h i c h I w i l l refer to as the J A B site, as being approximately 3.5 to 4 . 5 m m be low the dorsal surface o f the optic tectum, just ventrolateral to the oculomotor nucleus and ventromedial to the nucleus lateralus valvulae (ventrolateral to the fourth ventricle) i n carp. F igure 1.10 shows the approximate locat ion o f the J A B site i n carp. I calculated approximate coordinates for the J A B site i n carp, and its equivalent i n trout, by determining its pos i t ion relative to the oculomotor nucleus and nucleus lateralus valvulae us ing a map o f the internal brain landmarks ( M e e k and Nieuwenhuys , 1997). U s i n g these brain maps, I then determined the pos i t ion o f the J A B site relative to landmarks on the dorsal surface o f the carp and trout brains ( M e e k and Nieuwenhuys , 1997). K a i n i c ac id , when injected i n a bolus o f 75 n l , diffuses i n a ver t ical radius o f approximate ly 0 . 2 m m f rom the injection site (a total diameter o f 0 .4mm) and a hor izontal radius o f approximately 0 .7mm (a total diameter o f 1.4mm) (Sundin et al., 2003); therefore, I calculated 7 1 injection coordinates such that, once the ka in ic ac id diffused out f rom each injection site, the entire J A B site w o u l d be affected. T o cover the fu l l vo lume o f the J A B site, I injected ka in ic ac id at two bilateral injection sites, one caudal and one rostral. In carp, the caudal site was approximately 0 . 6 m m rostral to the anterior border o f the cerebel lum whi le the rostral site was 1.2mm rostral to the anterior border o f the cerebel lum; i n trout the caudal site was approximately 0 .7mm and the rostral site 2 . 0 m m rostral to the anterior border o f the cerebel lum. These injection sites were approximately 0.6 m m lateral to the mid l ine i n carp and 1.0mm lateral to the mid l ine i n trout. A t each injection site, I injected ka in ic ac id at three depths, separated by approximately 0.4 m m . In carp these injection sites were approximately 3.7, 4.1, and 4 . 5 m m beneath the dorsal surface o f the bra in w h i l e i n trout they were approximately 4.8, 5.2, and 5 .6mm beneath the dorsal surface. Figure 3.4 shows the approximate locat ion o f the injection sites o n the dorsal surface o f a decerebrate/spinalectomized carp bra in wh i l e Figure 3.5 shows the injection sites on the dorsal surface o f a decerebrate/spinalectomized trout brain. T o account for variat ion i n bra in size between ind iv idua l f ish, I measured major brain features such as the length and wid th o f the optic lobes, then used these measurements to scale m y coordinate calculat ions from the brain maps to fit that part icular brain. Va r i a t i on i n brain size between indiv iduals o f the same species was m i n i m a l when compared to their differences i n body mass and length. Figure 3.4: The approximate location of the injection sites marked on the dorsal surface of a decerebrate/spinalectomized carp brain. The injection sites are shown in blue while the black lines represent the landmarks (the midline and the rostral border of the cerebellum) used to determine the location of the injection sites. 72 Figure 3.5: The approximate location of the injection sites marked on the dorsal surface of a decerebrate/spinalectomized trout brain. The injection sites are shown in blue while the black lines represent the landmarks (the midline and the rostral border of the cerebellum) used to determine the location of the injection sites. So that I could inject kainic acid, ACSF (control), or marker dye at any given set of stereotaxic coordinates, I manufactured four-barreled micropipettes from fused four-barreled glass capillary tubing with an external tube diameter of 1.2mm and an internal tube diameter of 0.6 mm ( A - M Systems Inc.) I pulled these fused glass tubes on a PE-2 electrode puller (Narishige Scientific Instrument Laboratory, Japan), then broke off the tips under a dissecting microscope to produce a tip diameter of approximately 0.01mm. I mounted the micropipette on an SM-15 model Universal Electrode Carrier (Narishige Scientific Instrument Laboratory, Japan) for accurate micromanipulation during stereotaxic injections. At all three depths of each injection site I injected approximately 75 nl of 0.01 mM kainic acid dissolved in ACSF using a General Valve Corporation picospritzer n. To ensure that the correct volume of fluid was injected at each site I monitored the meniscus of the liquid in the micropipette using a dissecting microscope fitted with an ocular micrometer. I calibrated the picospritzer for each individual micropipette by calculating the cylindrical volume of the pipette tube and determining the height change in the position of the meniscus for each injection that corresponded to a volume of approximately 75 nl. To enable identification of the injection sites during histological analysis, I marked the rostral, dorsal, right site and the caudal, ventral, left 73 site w i t h a 75 n l injection o f 1% pontamine sky blue dye d isso lved in A C S F w i t h 0.01 m M ka in i c ac id (Hs ieh et al., 2000). T o control for the effects o f the injections, I injected approximately 75 n l o f A C S F into the same inject ion sites as for the ka in ic ac id . F igure 3.6 shows the microin jec t ion setup, w i th the tip o f a four-barreled micropipette inserted into the caudal midbra in o f a trout. Because ka in ic ac id is a glutamate agonist, micro in jec t ing i t onto a respiratory site should in i t i a l ly cause a clear respiratory response. A s I made each ka in ic ac id microinjec t ion , I moni tored the breathing o f the f ish to determine i f it responded. A typ ica l response to the ka in i c ac id consisted o f an immediate , substantial exci tat ion o f breathing frequency, ampli tude, and total vent i lat ion. To ta l vent i lat ion remained s ignif icant ly elevated even thirty minutes after the microinject ions . In compar ison , the control injections o f A C S F on ly s l ight ly st imulated respiration, and this exci tat ion dissipated almost immedia te ly . These immediate respiratory responses to the ka in ic ac id microinject ions were very site-sensitive; adjusting the loca t ion o f an inject ion site b y on ly 0 . 3 m m i n any direct ion c o u l d make the difference between a strong respiratory response and no response at a l l . Therefore, i f there was no clear respiratory response to the ka in ic ac id microinject ions, I w o u l d recalculate the J A B site coordinates and the corresponding injection coordinates and microinject ka in ic ac id into these new coordinates. H o w e v e r , it was rarely necessary to do these coordinate adjustments, ind ica t ing that the ka in i c ac id microinject ions were normal ly either direct ly i n the J A B site or were i n close p r o x i m i t y to the J A B site. 74 Figure 3.6: Photograph of the microinjection apparatus. A decerebrate/spinalectomized trout is clamped into the experimental tank by the stereotaxic device. A four-barreled, glass micropipette is inserted into the midbrain of the trout near the caudal border of the optic lobes. The micropipette is held by a micromanipulator, used to position the micropipette tip at the desired injection site using stereotaxic coordinates determined by landmarks on the dorsal surface of the brain. A dissecting microscope is used to observe the position of the micropipette tip more accurately. Measuring Respiratory Signals T o measure the breathing o f each fish, I recorded the changes i n pressure o f the buccal cavi ty and one o f the two opercular cavit ies. F r o m these measurements I was able to calculate the frequency o f respiration and the relative ampli tude o f each breath. T o measure bucca l and opercular pressure, I used Del t ran I V disposable pressure transducers (Utah M e d i c a l Products Inc.). I ampl i f ied and filtered the bucca l pressure signal us ing a G o u l d Un ive r sa l A m p l i f i e r set to direct current w i t h the l o w cutoff set at the dc offset and the h i g h cutoff set to 10 H z . I ampl i f i ed and filtered the opercular pressure s ignal us ing a G o u l d Integrator A m p l i f i e r set on direct mode. 75 T o col lec t and record these signals I used a D I - 7 2 0 Series data acquis i t ion system (Dataq Instruments) at a sample rate per channel o f 500 samples/second and v i e w e d them using W i n D a q P r o software (Dataq Instruments). T o calibrate the pressure signals for ampli tude, I used a ver t ical glass c o l u m n to generate k n o w n pressures o f water. Th i s glass c o l u m n was adjusted such that 0 c m H 2 0 was leve l w i t h the water surface o f the experimental tank. T o compare breathing at normal and decreased levels o f respiratory dr ive , I measured breathing frequency and ampli tude dur ing exposure to hyperox ia as w e l l as normoxia . T h e d isso lved oxygen content (PO2) o f the water i n no rmox ia was approximately 115 m m H g w h i l e i n hype rox ia it was approximately 500 m m H g . I measured the PO2 o f the water us ing a T y p e E 5 0 4 6 P 0 2 electrode i n a type D 6 1 6 Thermostatted C e l l and the P H A 9 3 4 P o 2 module o f a P H M 7 1 M k 2 acid-base analyzer (Radiometer, Denmark) . Experimental Protocol Afte r a l l o w i n g the decerebrate/spinalectomized f ish at least two hours to recover f rom instrumentation and surgery, I measured its breathing frequency and ampli tude i n normoxia . I recorded these no rmox ic respiratory signals for thirty minutes to establish a baseline for normal breathing, then exposed the fish to hyperoxia . I measured breathing dur ing the first five minutes o f hyperox ia to determine the in i t i a l response to increased d isso lved oxygen content, then again after ninety minutes o f hype rox ia to determine the response to a more pro longed hyperoxic exposure. F i n a l l y , I measured the breathing signals thirty minutes after returning the fish to n o r m o x i a to determine i f the hyperoxic exposure had any last ing effects o n breathing. These breathing measurements i n n o r m o x i a and hyperox ia were the control measurements to w h i c h I compared a l l subsequent breathing measurements. Throughout this entire per iod I also moni tored the pressure signals to determine the breathing pattern o f the fish at each leve l o f oxygen avai labi l i ty . Af ter comple t ing the control measurements, I measured the dimensions o f the optic lobes and scaled m y stereotaxic micro in jec t ion coordinates accordingly to speci f ica l ly fit that i nd iv idua l fish. I then made sham injections o f 75 n l o f A C S F into each o f the injection sites w h i l e recording the respiratory response. W h e n the breathing signals s tabi l ized, at least thirty minutes after the sham injections, I measured breathing frequency and ampli tude i n normoxia . I 76 then exposed the f ish to hyperox ia and recorded the in i t i a l and prolonged response, as described above. F i n a l l y , I returned the f ish to n o r m o x i a and, after thirty minutes, measured its breathing again. A s described above, I c lassif ied the breathing pattern as continuous, w a x i n g and wan ing , episodic , or some combina t ion o f these patterns throughout the durat ion o f the experiment. O n c e I had recorded a l l the measurements f rom the sham injected f i sh , I injected 75 n l o f 0.01 M ka in ic ac id into each o f the inject ion sites, us ing the exact same coordinates as for the sham injections. I also injected the rostral, dorsal , r ight site and the caudal , ventral, left site w i th a 75 n l injection o f 1% pontamine sky blue dye d i sso lved i n A C S F w i t h 0.01 m M ka in i c ac id to enable ident if icat ion o f the region o f inject ion sites dur ing h is to logica l analysis (Hs ieh et al., 2000). Af te r a l l o w i n g the f ish at least thirty minutes to recover f rom the injections, when its breathing had stabi l ized but was s t i l l exci ted b y the ka in i c ac id , I measured its breathing frequency and ampli tude i n n o r m o x i a and hyperox ia as described above. Af te r comple t ing a l l measurements for the ka in i c ac id injected f ish , I euthanized i t and preserved its midbra in for h i s to log ica l analysis. Histology I transected the rostral ha l f o f the midbra in and removed it, then dissected out the caudal ha l f o f the midbra in by transecting 1 m m caudal to the rostral border o f the cerebel lum. I r insed the bra in tissue for f ive minutes i n f ish phosphate buffered solut ion ( P B S ) w i t h a p H o f approximately 7.7, containing, i n g / L , 0.2 K C 1 , 0.2 K H 2 P 0 4 , 8.0 N a C l , and 2.16 N a 2 H P 0 4 . I then preserved the tissue i n 4 % formaldehyde d isso lved i n P B S for forty-eight hours at 4 degrees Ce l s ius . Af te r the tissue was preserved, I r insed i t again i n P B S at r o o m temperature. T o cryoprotect the tissue, I then submersed i t i n 20 % sucrose i n P B S for forty-eight hours at 4 degrees Ce l s ius . I froze the preserved, cryoprotected tissue at approximately - 2 7 degrees Ce l s ius and mounted i t i n a b l o c k o f O p t i m a l Cu t t ing Temperature ( O C T ) C o m p o u n d (Tissue-T e k ) . U s i n g a M I C R O M International Cryostat ( H M 5 0 5 E ) , I s l i ced the midbra in tissue into 50 u m sections. F o r the fourteen experimental carp and nine experimental trout, I then mounted these tissue samples onto slides and del ivered them to W a x - I t H i s t o l o g y for h i s to logica l staining. Unfortunately, the midbra in sections f rom these f ish were accidently destroyed dur ing the staining process by Wax- I t H i s t o l o g y technicians. Therefore, to determine the approximate 77 reg ion o f the midbra in destroyed by the ka in ic ac id injections, I microinjected two carp and two trout w i th ka in ic ac id , us ing the same methods as i n the experiments. I shipped the midbra in sections, free-floating i n 4 % formaldehyde i n plastic v ia ls , to F D Neurotechnologies , U S A . T o detect neurodegeneration, F D Neurotechnologies treated the free-f loat ing midbra in sections w i t h an F D N e u r o S i l v e r K i t I, w h i c h stains degenerating neural bodies, axons, and terminals b lack. They then mounted the stained midbra in sections on slides and returned them to me for analysis. Unfortunately, many o f the midbra in sections were poor ly reconstructed dur ing the mount ing process, so i n some cases the precise locat ion o f certain sites and structures was diff icul t to accurately identify. I d i d not use these extremely damaged midbra in sections for h i s to logica l analysis. T o determine the ka in ic ac id injection sites, I identif ied the areas o f neural degeneration stained b lack f rom the F D N e u r o S i l v e r K i t I i n the midbra in sections o f the two representative f ish for each species. I determined the loca t ion o f the damaged neural tissue relative to internal landmarks such as the oculomotor nucle i and the fourth ventricle. Data Analysis F o r each f ish I col lec ted and analyzed breathing frequency and ampli tude data f rom each experimental phase f rom a bucca l pressure trace that best represented the breathing dur ing that phase. F o r the in i t i a l n o r m o x i a data, I selected a representative trace f rom the last f ive minutes o f n o r m o x i a before I exposed the f ish to hyperoxia . F o r the in i t i a l hyperox ia data, I selected a representative trace from the first f ive minutes o f hyperoxic exposure. F o r the prolonged hyperox ia data, I selected a representative trace f rom the f ive minutes direct ly after ninety minutes o f hyperoxic exposure. F i n a l l y , for the data for n o r m o x i a after hyperoxic exposure, I selected f ive minutes o f data after the f ish had been returned to n o r m o x i a for thirty minutes. These representative traces var ied i n length f rom thirty seconds to ten minutes depending on the frequency and pattern o f breathing. F o r cont inuously breathing f ish , I analyzed at least one hundred breaths; thirty seconds o f data was often sufficient when this cont inuous breathing was c a l m and regular. H o w e v e r , for f ish breathing ep isodica l ly , I analyzed at least ten episodes or one hundred breaths (whichever was longer) whenever possible . F o r f ish w i t h very l o w breathing frequencies, I needed to analyze f ive to ten minutes o f the trace to obtain sufficient breathing data. 78 I c lass i f ied a f ish as breathing ep isodica l ly i f it grouped its breaths into discrete episodes separated by clear apneas lasting for at least the time required for three normal breaths (approximately two to four seconds). I c lassif ied a f ish as breathing ep isod ica l ly dur ing each experimental stage i f it breathed ep isodica l ly (meeting the cri teria described above) for at least one minute dur ing the representative trace I analyzed for that stage. H o w e v e r , to determine i f the ka in ic ac id injections had el iminated episodic breathing in a f ish , I w o u l d examine the entire breathing trace for each stage. I calculated the proport ion o f trout and o f carp breathing ep isod ica l ly i n each experimental phase. I compared the proport ion o f ind iv idua ls breathing ep isod ica l ly i n each experimental phase us ing the Chi-square or Fisher E x a c t Test (SigmaStat 2.0). I v i ewed , filtered, and analyzed a l l s ignal traces us ing the W i n D a q W a v e f o r m B r o w s e r and A d v a n c e d C O D A S software package (Dataq Instruments Inc.). Because o f the large i nd iv idua l var ia t ion i n the baseline breathing ampli tude o f the f ish, I converted each ampli tude measurement into percent change f rom the in i t i a l , no rmox ic ampli tude measurement, us ing each f ish as its o w n control . (Because I measured breathing for each f ish at each phase o f the experimental p ro tocol , I c o u l d direct ly compare respiration at each phase for each ind iv idua l f ish and thus determine the effects o f each stage o f the experimental procedure. T h i s a l lowed me to use each f ish i n its in i t i a l , no rmox ic state as its o w n control .) I calculated the percent change i n breathing frequency i n the same way . T o calculate total vent i la t ion, I mu l t ip l i ed these percent change i n frequency values wi th the corresponding percent change i n ampli tude values for each f ish , then d iv ided by 100, to g ive a percent change i n total vent i lat ion value that combined breathing frequency and amplitude. I expressed a l l measurement values as the mean ± the standard error o f the mean. I normal ized a l l percent change data us ing the modi f i ed arcsine transform described i n Chapter 2 (1 and 2 on pg.42). Da ta points less than 1 (100%) were arcsine transformed w h i l e data points greater than or equal to 1 were l inear ly transformed. I compared the transformed breathing frequency, ampli tude, and total vent i lat ion data f rom each experimental phase us ing a One W a y Repeated Measures A n a l y s i s o f Var i ance ( A N O V A ) (SigmaStat 2.0) w i th an a lpha value o f 0.05. W h e n the A N O V A results indicated a significant difference, I determined w h i c h groups were s ignif icant ly different us ing an A l l Pa i rwi se M u l t i p l e C o m p a r i s o n Procedure T u k e y Test or a M u l t i p l e Compar i sons versus C o n t r o l G r o u p Bonfe r ron i t-test, using the in i t i a l , no rmox ic measurements as the control group (SigmaStat 2.0). W h e n the data fai led the 79 normal i ty requirements for a parametric test, I used a F r i edman Repeated Measures A N O V A on R a n k s (SigmaStat 2.0) fo l l owed by an A l l Pa i rwi se M u l t i p l e C o m p a r i s o n Procedure T u k e y Test (SigmaStat 2.0) to analyze the data. Results Kainic Acid Microinjections T h e ka in ic ac id microinject ions caused a strong excitatory reaction i n both carp and trout, s t imulat ing an increase i n both breathing ampli tude and frequency almost immedia te ly . H o w e v e r , the reactions were very site-sensitive. A t each injection site, the three inject ion depths were on ly separated by a ver t ical distance o f 0 .4mm; however , al though injecting ka in ic ac id at one depth might e l ic i t a large excitatory reaction, injecting ka in ic ac id at another depth i n the same site might e l ic i t on ly a smal l response or no response at a l l . F igure 3.7 shows a representative breathing trace for a carp dur ing ka in ic ac id microinject ions at the three different depths at one injection site. K a i n i c ac id microin jec t ion at the first depth caused an almost immediate m i l d excitatory response w h i l e microin jec t ion at the second depth caused on ly a smal l and transient response. M i c r o i n j e c t i o n at the third depth, however , caused a powerfu l excitatory response, greatly s t imulat ing both breathing amplitude and frequency. Therefore, the effects o f ka in ic ac id injection on respiration c o u l d vary hugely i f the depth o f injection was altered by on ly a few tenths o f a mi l l imeter . S i m i l a r l y , adjusting the locat ion o f an injection site by on ly 0 . 3 m m laterally, med ia l ly , rostral ly, or caudal ly c o u l d make the difference between a strong respiratory response and no response at a l l . T h i s evidence suggests that the J A B site i n both carp and trout is a discrete and h igh ly loca l i zed region o f the mesencephal ic tegmentum. Therefore, to destroy the J A B site w i th ka in ic ac id microinject ions, the locat ion o f the injection sites w o u l d need to be extremely accurate. Because o f this, i n each experiment, i f the ka in ic ac id injections d i d not e l ic i t a strong respiratory response, I assumed that I had not hit a respiratory site. In these rare cases I w o u l d recalculate the J A B coordinates and injection sites for that part icular bra in and microinject ka in ic ac id into these corrected coordinates. O n l y i n a few f ish o f each species did-the in i t i a l ka in ic ac id injections e l ic i t no response, and i n a l l o f these cases, when I recalculated the injection coordinates and injected the ka in ic ac id again, the f ish showed a strong respiratory response to the injections. 80 In compar i son to the ka in ic ac id injections, the sham injections ( A C S F ) on ly e l ic i ted a weak excitatory respiratory response. W h i l e the exi tat ion o f respiration by the ka in ic ac id microinject ions persisted for at least thirty minutes, respirat ion was stimulated for less than one minute b y the sham microinject ions. Carp - Breathing Frequency In normoxia , control decerebrate/spinalectomized carp had a mean breathing frequency o f approximately 50.6 ± 4.0 breaths/min. D u r i n g the first f ive minutes o f hyperoxia , breathing frequency decreased (p<0.001) to 37.5 ± 5.2 breaths/min. Af te r ninety minutes o f hyperoxia , breathing frequency was 31.6 ± 4.3 breaths/min, s ignif icant ly lower (p<0.001) than i n no rmox ia . Th i r ty minutes after returning to normoxia , breathing frequency recovered to 49.8 ± 3.8 breaths/min. T h e sham injections d i d not s ignif icant ly affect breathing frequency i n carp. Af te r the sham injections, breathing frequency decreased f rom 47.2 ± 2.0 breaths/min i n n o r m o x i a to 39.7 ± 4 . 1 breaths/min i n in i t i a l hyperox ia (p=0.045). Af ter ninety minutes o f hyperoxia , breathing frequency was 31.7 ± 4.1 breaths/min, s ignif icant ly s lower than i n n o r m o x i a (p<0.001). Brea th ing frequency fu l ly recovered to 46.0 ± 2.6 breaths/min thirty minutes after be ing returned to normoxia . T h e kain ic ac id injections s ignif icant ly increased breathing frequency i n n o r m o x i a (p=0.024), in i t i a l hyperox ia (p<0.001), and prolonged hyperox ia (p=0.008). H o w e v e r , thirty minutes after be ing returned to normoxia , breathing frequency was no longer s ignif icant ly higher than before the ka in ic ac id injections (Figure 3 . 8 A ) . Th i r ty minutes after the ka in ic ac id injections, the carp had a mean breathing frequency o f approximately 66.9 ± 4.8 breaths/min i n no rmox ia . D u r i n g in i t i a l hyperoxia , breathing frequency on ly decreased s l ight ly (p=0.095) to 59.3 ± 6.4 breaths/min, but decreased signif icant ly (p=0.010) to 50.6 ± 6.3 breaths/min after ninety minutes o f hyperoxia . Th i r ty minutes after be ing returned to no rmox ia , breathing frequency recovered to approximately 54.6 ± 4.6 breaths/min. 81 10 cmH20 1 min Figure 3.7: Representative breathing trace showing responses to kainic acid microinjections into three different depths of the same injection site in the mesencephalic tegmentum of a decerebrate/spinalectomized carp. This trace was generated from changes in buccal cavity pressure. Each arrow indicates an injection of 75 nl of 0.01 M kainic acid. Breathing before the first arrow is of the frequency and amplitude normally seen in normoxia. Respiratory responses to kainic acid injection are highly site-specific and range from extreme excitation to inhibition to no response at all. Carp - Breathing Amplitude In normoxia , control decerebrate/spinalectomized carp had a mean breathing ampli tude o f approximately 0.8 ± 0 . 1 c m H 2 0 . H o w e v e r , because these carp had such h igh variat ion i n breathing amplitude between ind iv idua ls i n normoxia , ranging f rom 0.3 to 2.0 c m H 2 0 , it was more accurate to analyze the breathing ampli tude data as percent change f rom control i n n o r m o x i a for each i nd iv idua l carp than to analyze the absolute values. (The absolute values for breathing ampli tude are shown i n Table A . 3 i n the A p p e n d i x . ) In prolonged hyperoxia , breathing ampli tude decreased (p=0.042) by 34.0 ± 9.0 percent f rom normoxia , w h i l e thirty minutes after return to normoxia , breathing ampli tude had recovered to 94.9 ± 10.3 percent o f the in i t i a l no rmox ic values (Figure 3 .8B) . 82 The sham injections d i d not s ignif icant ly affect breathing ampli tude i n carp. Af te r the sham injections, breathing ampli tude remained at 94.4 ± 8.2 percent o f the in i t i a l control no rmox ic values. D u r i n g in i t i a l hyperox ia , ampli tude decreased (p=0.01) to 65.4 ± 6.7 percent o f no rmox ic control values w h i l e after ninety minutes o f hyperoxia , ampli tude remained s ignif icant ly lower than i n n o r m o x i a (p=0.004), at 63.1 ± 7.6 percent o f no rmox ic control values. Th i r ty minutes after return to normoxia , breathing ampli tude recovered to 90.5 ± 8.6 percent o f no rmox ic control values. T h e ka in ic ac id injections d id not s ignif icant ly affect breathing ampli tude i n carp. Af te r the ka in ic ac id injections, breathing ampli tude increased s l ight ly to 109.3 ± 12.0 percent o f no rmox ic control values. D u r i n g the first f ive minutes o f hyperoxia , breathing ampli tude decreased (p<0.001) to 61.1 ± 7.9 percent o f no rmox ic control values w h i l e i n pro longed hyperoxia , ampli tude remained signif icant ly lower than i n n o r m o x i a (p<0.001) at 49.6 ± 5.6 percent o f no rmox ic control values. Th i r ty minutes after return to no rmox ia , breathing ampli tude remained s ignif icant ly lower than i n n o r m o x i a (p=0.008), recover ing to on ly 76.3 ± 8.9 percent o f no rmoxic control values. Carp - Total Ventilation D u r i n g in i t i a l hyperoxia , control decerebrate/spinalectomized carp decreased (p<0.05) total vent i la t ion to 43.6 ± 3.9 percent o f no rmoxic values w h i l e i n pro longed hyperoxia , total vent i la t ion decreased (p<0.05) to 36.7 ± 4.9 percent o f no rmoxic values. Af te r thirty minutes o f no rmoxia , total vent i la t ion recovered to 92.9 ± 10.5 percent o f in i t i a l no rmoxic values (Figure 3 .8C.) T h e sham injections d i d not s ignif icant ly affect total vent i lat ion. In n o r m o x i a after the sham injections, total vent i lat ion remained at 91.2 ± 8.4 percent o f no rmox ic cont ro l values. 83 Figure 3.8: Percent change in breathing frequency, amplitude, and total ventilation in decerebrate/spinalectomized carp from control normoxia after sham and kainic acid microinjections into the mesencephalic tegmentum at both normal and low levels of respiratory drive. Po2 levels were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency (A), amplitude (B), and total ventilation (C) were measured for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum and were compared to the normoxic control values for that same carp such that each fish served as its own control. The asterix (*) indicates a value that is significantly different than the corresponding control value. The cross (*) indicates a value that is significandy different from normoxia at the same experimental phase (Control, Sham, or Kainic Acid). Sample size of 14 carp. 84 D u r i n g in i t i a l hyperoxia , total vent i la t ion decreased (p<0.001) to 49.2 ± 4.6 percent o f no rmox ic cont ro l values and i n prolonged hyperoxia , total vent i la t ion was 35.9 ± 3 . 9 percent o f no rmox ic control values, also s ignif icant ly l ower than i n n o r m o x i a (p<0.001). Th i r ty minutes after return to normoxia , total vent i lat ion recovered to 84.6 ± 8 . 5 percent o f no rmox ic control values. T h e ka in ic ac id injections s ignif icant ly increased total vent i la t ion i n in i t i a l n o r m o x i a (p=0.025) and in i t i a l hyperox ia (p=0.017) but not i n pro longed hyperox ia or f ina l no rmox ia . Af te r the ka in ic ac id injections, i n no rmoxia , total vent i la t ion increased to 152.7 ± 21.9 percent o f no rmox ic control values w h i l e i n in i t ia l hyperoxia , total vent i lat ion decreased (p<0.001) to 69.4 ± 9.4 percent o f no rmoxic control values but was s t i l l s ignif icant ly elevated compared to the corresponding cont ro l values. D u r i n g pro longed hyperoxia , total vent i la t ion remained s ignif icant ly lower than i n n o r m o x i a (p<0.001), at 54.8 ± 9.9 percent o f no rmox ic control values. Th i r ty minutes after return to normoxia , total vent i lat ion recovered to 87.5 ± 14.3 percent o f no rmox ic control values, remain ing s ignif icant ly l ower than i n n o r m o x i a on ly thirty minutes after the ka in i c ac id injections (p=0.004). Carp - Overall Response In normoxia , approximately 36 percent o f decerebrate/spinalectomized carp breathed ep isod ica l ly w h i l e the remain ing carp breathed cont inuously (Figure 3.9). Af te r f ive minutes o f hyperoxia , approximately 71 percent o f carp breathed ep isodica l ly wh i l e after ninety minutes o f hyperoxia , approximately 93 percent o f carp (thirteen out o f fourteen) breathed i n episodes. T h e propor t ion o f carp breathing i n episodes was s ignif icant ly higher i n hyperox ia than i n n o r m o x i a (p=0.006). Th i r ty minutes after be ing returned to no rmox ia , 43 percent o f carp breathed episodica l ly . Af te r the sham microinject ions , approximately 36 percent o f carp breathed ep isodica l ly i n normoxia . In the first f ive minutes o f hyperoxia , approximately 64 percent o f carp breathed ep isod ica l ly wh i l e after ninety minutes o f hyperox ia approximately 93 percent o f carp breathed i n episodes. Af te r the sham injections, the propor t ion o f carp breathing ep isodica l ly was s ignif icant ly higher i n hype rox ia than i n no rmox ia (p=0.006), just as it was for the control carp. A p p r o x i m a t e l y 29 percent o f carp breathed i n episodes thirty minutes after be ing returned to 85 normoxia . C l e a r l y the sham microinject ions d i d not substantially decrease the number o f carp breathing episodica l ly . Af ter the ka in ic ac id microinject ions, no carp breathed episodica l ly i n n o r m o x i a in i t i a l ly , and thirty minutes after be ing returned to no rmox ia , on ly one o f the fourteen carp breathed i n episodes. In in i t i a l hyperox ia , approximately 27 percent o f carp breathed episodica l ly w h i l e 36 percent breathed i n episodes after ninety minutes o f hyperoxia . The propor t ion o f carp breathing ep isod ica l ly i n hyperox ia was st i l l s ignif icant ly higher (p=0.041) than i n n o r m o x i a after the ka in ic ac id injections. H o w e v e r , the ka in ic ac id microinject ions comple te ly e l iminated breathing episodes in ha l f o f the carp tested. In a further 29 percent o f carp, the ka in ic ac id microinject ions reduced the occurrence o f breathing episodes, wh i l e i n the remain ing 21 percent, the microinject ions had no apparent effect on the episodic breathing pattern. T h e ka in ic ac id injections s ignif icant ly decreased the proport ion o f carp breathing ep isod ica l ly i n n o r m o x i a (p=0.041), in i t ia l hyperox ia (p=0.023), and prolonged hyperox ia (p=0.002) (Figure 3.9). In ha l f o f the fourteen carp, the ka in ic ac id injections complete ly e l iminated episodic breathing; I w i l l refer to these seven carp as the E E (episodes el iminated) subset. I w i l l refer to the seven carp for w h i c h the ka in ic ac id injections d i d not el iminate episodic breathing as the E R (episodes remained) subset. F igure 3.10 shows a representative breathing trace for a carp i n the E E subset. T h i s part icular carp breathed cont inuously i n no rmox ia i n a l l phases o f the experiment; however , some o f the E E subset o f carp also breathed ep isodica l ly i n n o r m o x i a before the ka in ic ac id injections. In hyperoxia , the E E subset o f carp breathed ep isodica l ly for the control and sham injection phases o f the experiment. H o w e v e r , after the ka in i c ac id injections, this subset breathed cont inuously throughout the fu l l ninety minutes o f hyperoxia , showing that the ka in ic ac id microinject ions had comple te ly e l iminated episodic breathing. A l t h o u g h frequency s l ight ly increased i n hyperox ia after the ka in ic ac id injections because these carp were no longer breathing in episodes, a sl ight decrease in breathing ampli tude compensated for the change i n frequency, l eav ing total vent i la t ion unchanged. Interestingly, the E E subset o f carp showed some differences in respiratory responses compared to the E R subset o f carp. The breathing frequency response o f the two subsets is almost ident ica l (Figure 3.11 A ) . H o w e v e r , thirty minutes after the ka in ic ac id injections, the E E subset decreased breathing ampli tude i n no rmox ia and hyperox ia wh i l e the E R subset increased breathing amplitude (Figure 3.1 I B . ) Brea th ing ampli tude i n n o r m o x i a thirty minutes after the ka in ic ac id injections (p=0.015), as w e l l as i n in i t i a l hyperox ia after the ka in ic ac id injections 86 (p=0.027) was signif icant ly lower i n the E E subset o f carp compared to the E R subset. T w o hours after the microinject ions, i n hyperoxia , the E E subset reduced breathing amplitude more substantially than d id the E R subset, al though this difference was not significant. Because o f this difference i n breathing ampli tude, the total venti lat ion o f the E R subset was also s ignif icant ly higher than that o f the E E subset i n no rmox ia thirty minutes after the microinject ions (p=0.028) (Figure 3 .11C.) To ta l venti lat ion was s l ight ly higher i n the E R subset than the E E subset for hyperox ia and f inal no rmoxia ; however , these trends were not significant. T h i s difference suggests that for the E E subset, after the ka in ic ac id injections, the increase i n breathing frequency caused by the loss o f episodes was compensated for by the decrease in breathing ampli tude such that total vent i lat ion d i d not s ignif icant ly increase. _ Control Sham Kainic Acid Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Figure 3.9: Percent of carp that breathed in episodes in normoxia and hyperoxia before and after microinjections of kainic acid into the mesencephalic tegmentum. Carp were classified as breathing episodically in each stage of the experimental protocol if they breathed in episodes separated by clear apneas (lasting for the time required for at least three breaths) for at least one minute or more in that stage. These protocol stages were Normoxia, Initial Hyperoxia (the first five minutes of hyperoxia), Prolonged Hyperoxia (after 1.5 hours of hyperoxia), and Normoxia (30 minutes after being returned to normoxia). Carp were tested in these four protocol stages in three states: Control (initially), Sham (after microinjections of ACSF), and Kainic Acid (after microinjections of kainic acid). The Po2 in normoxia was approximately 115 mmHg while the Po2 in hyperoxia was approximately 500 mmHg. The asterix (*) indicates that the proportion of carp breathing episodically is significantly different from the proportion of carp breathing episodically in the corresponding control situation. The cross (*) indicates that the proportion of carp breathing episodically is significantly different from the proportion of carp breathing episodically in normoxia for the same experimental state. Sample size of 14 carp. 87 Control Normoxia Control Hyperoxia Sham Normoxia Sham Hyperoxia Kainic Acid, Normoxia Kainic Acid Hyperoxia 1 c m H 2 0 3 0 seconds Figure 3.10: Representative breathing trace from a decerebrate/spinalectomized carp for which kainic acid microinjections eliminated the episodic breathing pattern. These breathing traces were recorded from pressure changes in the buccal cavity. Po2 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. 88 Figure 3.11: Percent change in breathing frequency, amplitude, and total ventilation from control normoxia in the E E and E R subsets of decerebrate/spinalectomized carp. Po 2 levels were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency (A), amplitude (B), and total ventilation (C) were measured for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham, not shown), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum and were compared to the normoxic control values for that same carp such that each fish served as its own control. The kainic acid injections completely eliminated episodic breathing for carp in the EE subset, while in the ER subset, the kainic acid injections did not completely eliminate episodic breathing. The asterix (*) indicates a value that is significandy different than the corresponding control value. The cross (*) indicates a value that is significandy different from normoxia at the same experimental phase (Control or Kainic Acid). The phi (<|>) indicates a value that is significantly different than the corresponding value for the EE subset. Sample size of 6 EE carp and 7 ER carp. 89 Trout - Breathing Frequency In normoxia , the control decerebrate/spinalectomized trout breathed w i t h a frequency o f approximately 85.0 ± 4.6 breaths/min. D u r i n g in i t i a l hyperoxia , breathing frequency decreased (p=0.044) to 73.3 ± 7.9 breaths/min. H o w e v e r , after ninety minutes o f hyperoxia , breathing frequency increased to 77.4 ± 10 breaths/min, such that i t was not longer s ignif icant ly s lower than i n no rmox ia (p=0.172). Th i r ty minutes after return to no rmox ia , breathing frequency recovered to 87.6 ± 5.1 breaths/min (Figure 3 . 1 2 A ) . T h e sham injections d id not s ignif icant ly affect breathing frequency i n trout. Af te r the sham injections, trout i n no rmox ia breathed at approximately 85.7 ± 5 . 6 breaths/min, w h i l e i n in i t i a l hyperox ia they breathed signif icant ly s lower (p=0.028), at 68.7 ± 5 . 8 breaths/min. D u r i n g pro longed hyperoxia , however , breathing frequency increased to 75.0 ± 7.3 breaths/min, such that it was no longer s ignif icant ly s lower than i n n o r m o x i a (p=0.192). Th i r ty minutes after return to normoxia , breathing frequency recovered to 85.6 ± 5 . 4 breaths/min. T h e ka in ic ac id injections s ignif icant ly increased breathing frequency i n in i t i a l n o r m o x i a (p=0.049), in i t i a l hyperox ia (p=0.027), and prolonged hyperox ia (p=0.04), but not i n f ina l no rmoxia . In normoxia , thirty minutes after the ka in ic ac id injections, breathing frequency was 99.0 ± 6.4 breaths/min. D u r i n g in i t ia l hyperoxia , breathing frequency decreased to 89.5 ± 7 . 1 breaths/min, however this trend was not significant. Brea th ing frequency increased to 100.6 ± 10.3 breaths/min i n prolonged n o r m o x i a and recovered to 98.1 ± 9.3 breaths/min thirty minutes after return to normoxia . Trout - Breathing Amplitude C o n t r o l decerebrate/spinalectomized trout i n n o r m o x i a had a mean breathing ampli tude o f approximately 1.4 ± 0.2 c m H 2 0 . H o w e v e r , because trout breathing ampli tude i n no rmox ia had such a h igh l eve l o f variat ion between ind iv idua ls , ranging f rom 0.5 to 2.5 c m H 2 0 , i t was more accurate to analyze the breathing ampli tude data as percent change f rom the no rmox ic control for each i nd iv idua l trout than to analyze the absolute values. (The absolute values for breathing amplitude are shown i n Table A . 5 in the A p p e n d i x . ) Brea th ing ampli tude decreased (p<0.001) to 53.8 ± 5.0 percent o f no rmox ic values i n in i t i a l hyperox ia and further decreased (p<0.001) to 44.2 ± 4.5 percent o f no rmox ic values i n pro longed hyperoxia . T h i r t y minutes after 90 return to normoxia , breathing ampli tude increased to 111.1 ± 6 . 9 percent o f in i t i a l no rmoxic values (Figure 3 .12B.) T h e sham injections had no effect on breathing ampli tude i n decerebrate/spinalectomized trout. In normoxia , thirty minutes after the sham injections, breathing ampli tude was 94.9 ± 8.3 percent o f no rmox ic control values. D u r i n g in i t i a l hyperoxia , breathing ampli tude decreased (p<0.001) to 51.4 ± 5.7 percent o f no rmoxic control values and after ninety minutes o f hyperoxia , breathing ampli tude further decreased to 42.1 ± 6 . 1 percent o f no rmox ic control values, remain ing s ignif icant ly l ower than i n n o r m o x i a (p<0.001). Th i r ty minutes after return to no rmox ia , breathing ampli tude recovered to 89.1 ± 9.2 percent o f no rmoxic control values. Af ter the ka in ic ac id injections, breathing amplitude showed a tendency to increase i n in i t i a l n o r m o x i a and hyperoxia ; however , these trends were not s ignif icant (p=0.358 and p=0.098, respectively.) In no rmox ia , thirty minutes after the ka in ic ac id injections, breathing ampli tude increased to 123.4 ± 23.0 percent o f no rmoxic control values. D u r i n g in i t ia l hyperox ia , breathing ampli tude decreased (p=0.011) to 70.4 ± 10.0 percent o f no rmox ic control values. Af te r ninety minutes o f hyperoxia , breathing ampli tude further decreased to 48.3 ± 6.3 o f no rmoxic cont ro l values, r emain ing s ignif icant ly lower than i n n o r m o x i a (p<0.001). Th i r ty minutes after return to no rmox ia , breathing ampli tude recovered to 89.3 ± 1 1 . 3 percent o f no rmox ic control levels , s t i l l s ignif icant ly l ower (p=0.02) than i n n o r m o x i a thirty minutes after the ka in ic ac id injections. Trout - Total Ventilation Decerebrate/spinalectomized control trout i n in i t i a l hype rox ia decreased their total vent i la t ion (p<0.001) to 45.0 ± 4.2 percent o f no rmox ic levels , w h i l e i n p ro longed hyperox ia they further reduced their total vent i lat ion to 41.9 ± 6.4 percent o f no rmox ic levels . Th i r ty minutes after return to normoxia , total vent i lat ion recovered to 114.4 ± 7.3 percent o f in i t i a l no rmoxic levels (Figure 3 .12C.) T h e sham injections d i d not s ignif icant ly affect total vent i la t ion i n these trout. In normoxia , thirty minutes after the sham injections, total vent i lat ion remained at 97.7 ± 10.2 percent o f no rmox ic control levels . D u r i n g the first f ive minutes o f hyperoxia , total vent i la t ion was reduced (p<0.001) to 42.7 ± 6.5 percent o f no rmox ic control values, w h i l e after ninety minutes o f hyperoxia , total vent i la t ion was further reduced to 37.7 ± 6.7 percent o f no rmoxic 91 Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia • • • • Control • • • • Sham Kainic Acid Figure 3.12: Percent change in breathing frequency, amplitude, and total ventilation in decerebrate/spinalectomized trout from control normoxia after sham and kainic acid microinjections into the mesencephalic tegmentum at both normal and low levels of respiratory drive. P02 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency (A), amplitude (B), and total ventilation (C) were measured for each trout in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum and were compared to the normoxic control values for that same trout such that each fish served as its own control. The asterix (*) indicates a value that is significantly different than the corresponding control value. The cross (*) indicates a value that is significantly different from normoxia at the same experimental phase (Control, Sham, or Kainic Acid). Sample size of 9 trout. 92 control values. Th i r ty minutes after return to normoxia , total vent i lat ion recovered to 91.2 ± 11.0 percent o f no rmox ic control values. T h e ka in ic ac id injections s ignif icant ly increased total vent i la t ion i n in i t i a l hyperox ia (p=0.04) but not at any other l eve l o f respiratory dr ive. Th i r ty minutes after the ka in ic ac id injections, i n no rmox ia , total vent i la t ion increased to 148 ± 3 1 . 9 percent o f no rmox ic control values. V a r i a t i o n i n total venti lat ion between i nd iv idua l trout was higher dur ing this phase than at any other time dur ing the experiment. D u r i n g in i t ia l hyperoxia , total vent i la t ion decreased (p=0.004) to 74.7 ± 12.4 percent o f no rmox ic control values, w h i l e i n pro longed hyperoxia , total vent i la t ion further decreased to 57.6 ± 9.4 percent o f no rmoxic cont ro l values. Th i r ty minutes after return to normoxia , total vent i la t ion recovered to 109.5 ± 19.6 percent o f no rmoxic control values, s t i l l s ignif icant ly lower than i n no rmox ia thirty minutes after the ka in ic ac id injections (p=0.033). Trout - Overall Response Decerebrate/spinalectomized trout breathed cont inuously for most o f the experiment. N o trout breathed ep isodica l ly i n n o r m o x i a at any stage o f the experiment. In in i t i a l hyperoxia , on ly 22.2 percent (two out o f nine) o f the control trout breathed episodica l ly . A l t h o u g h several more trout breathed ep isodica l ly dur ing the transition between n o r m o x i a and hyperoxia , they on ly cont inued this behaviour for approximately twenty seconds, on ly p roduc ing two or three clear episodes, and therefore d id not meet the cri ter ia for be ing classif ied as breathing ep isodica l ly i n in i t i a l hyperoxia . E v e n after ninety minutes o f hyperoxia , on ly 33.3 percent (three out o f nine) o f the control trout breathed ep isod ica l ly (Figure 3.13.) O f the two trout that breathed e p i s o d i c a l l y i n in i t i a l hyperoxia , on ly one cont inued to breathe ep isodica l ly i n pro longed hyperoxia , and one trout that breathed cont inuously i n in i t i a l hyperox ia breathed ep isod ica l ly i n pro longed hyperoxia . Therefore, out o f nine trout tested, on ly four ever breathed episodica l ly . T h e proport ion o f trout breathing ep isodica l ly i n hype rox ia as compared to n o r m o x i a was not s ignif icant ly different (p=0.149). T h e sham injections had no effect on the number o f trout breathing ep isodica l ly i n each experimental phase. N o trout breathed ep isodica l ly i n normoxia . O n l y 22.2 percent o f the trout breathed ep isodica l ly i n in i t i a l hyperox ia and 33.3 percent o f trout breathed ep isodica l ly i n 93 prolonged normoxia . T h e proport ion o f trout breathing ep isodica l ly i n hyperox ia as compared to n o r m o x i a was not s ignif icant ly different (p=0.149). Af ter the ka in ic ac id injections, however , on ly 11.0 percent o f trout breathed ep isodica l ly i n in i t i a l hyperox ia and 22.2 percent o f the trout breathed ep isod ica l ly i n p ro longed hyperoxia . Therefore, the ka in ic ac id injections on ly comple te ly e l iminated episodic breathing i n one trout o f the four that breathed episodica l ly . Af ter the ka in ic ac id injections, the proport ion o f trout breathing ep isodica l ly d id not change s ignif icant ly (p=0.289) compared to the control . F igure 3.14 shows representative breathing traces for the one trout i n w h i c h the ka in ic ac id injections complete ly e l iminated episodic breathing. In normoxia , this trout breathed cont inuously i n a l l phases o f the experiment; however , after the ka in ic ac id injections it increased both breathing frequency and ampli tude, leading to a substantial increase i n total vent i lat ion. In hyperoxia , this trout breathed ep isodica l ly i n the control and sham inject ion phases; however , after the ka in ic ac id injections, it breathed i n a rapid continuous pattern. Therefore, the ka in ic ac id injections appear to have strongly stimulated breathing i n this trout. T h e subset o f trout that on ly breathed cont inuously responded differently to the ka in ic ac id microinject ions than d id the subset o f trout that breathed episodica l ly . F o r example , the subset o f trout that breathed ep isodica l ly s ignif icant ly increased breathing frequency after the ka in ic ac id injections i n pro longed hyperox ia (p=0.014); however , after the in i t i a l respiratory exci ta t ion f rom the ka in ic ac id had dissipated, the subset o f cont inuously breathing trout d i d not show a significant increase i n breathing frequency i n any experimental phase (p=0.281) (Figure 3 . 1 5 A ) . The subset o f cont inuously breathing trout d id not show a s ignif icant ly different breathing ampli tude response to the ka in ic ac id injections compared to the ep i sod ic ia l ly breathing subset o f trout (Figure 3 .15B) . T h e total venti lat ion response to the ka in ic ac id injections o f the cont inuously breathing subset o f trout is less pronounced that that o f the ep i sod ica l ly breathing subset o f trout, al though this difference is not significant (Figure 3 .15C.) F o r the trout that on ly breathed cont inuously , the ka in ic ac id injections d i d not have a significant effect on breathing pattern. 94 A c i d PU Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Figure 3.13: Percent of trout that breathed in episodes in normoxia and hyperoxia before and after microinjections of kainic acid into the mesencephalic tegmentum. Trout were classified as breathing episodically in each stage of the experimental protocol if they breathed in episodes separated by clear apneas (lasting for the time required for at least three breaths) for at least one minute or more in that stage. These protocol stages were Normoxia, Initial Hyperoxia (the first five minutes of hyperoxia), Prolonged Hyperoxia (1.5 hours of hyperoxia), and Normoxia (30 minutes after being returned to normoxia). Trout were tested in these four protocol stages in three states: Control (initially), Sham (after microinjections of ACSF), and Kainic Acid (after microinjections of kainic acid). The Po2 in normoxia was approximately 125 mmHg while the Po2 in hyperoxia was approximately 500 mmHg. Note that there were no significant differences in the proportion of trout breathing episodically in any phase of the experiment. Sample size of 9 trout. 9 5 Control Normoxia Control Hyperoxia Sham Normoxia Sham Hyperoxia Kainic Acid Normoxia Kainic Acid Hyperoxia 1 ciriHjO 30 seconds Figure 3.14: Representative breathing trace from the only decerebrate/spinalectomized trout in which kainic acid microinjections eliminated the episodic breathing pattern. These breathing traces were recorded from pressure changes in the buccal cavity. Po2 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Note the dramatic increase in both breathing frequency and amplitude in normoxia and hyperoxia after the kainic acid injections. 96 180 -I w a s 160 • s a-140 • o •-1*. 120 • _= 100 =x 80 1 Ch 60 — a 40 o i i 20 i> a. 0 -A * T *iT I L <t>l J A l *B I *B Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Control Episodic C Normoxia Initial Prolonged Normoxia Hyperoxia Hyperoxia Figure 3.15: Percent change in breathing frequency, amplitude, and total ventilation from control normoxia in the subset of decerebrate/spinalectomized trout that only breathed continuously and the subset that breathed episodically. Po2 levels were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency (A), amplitude (B), and total ventilation (C) were measured for each trout in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham, not shown), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum and were compared to the normoxic control values for that same trout such that each fish served as its own control. Note that no value was significantly different than the corresponding control value. The cross (*) indicates a value that is significandy different from normoxia at the same experimental phase (Control or Kainic Acid). The asterix (*) indicates a value that is significantly different than the corresponding control value. The phi (0) indicates a value that is significantly different than the corresponding value for the episodic subset. Sample size of 4 trout. 97 Histology K a i n i c ac id was microinjected bi lateral ly at three depths into both a rostral and a caudal injection site i n carp and trout. These inject ion sites and the J A B site are shown relative to internal anatomical landmarks o f the trout midbra in i n F igure 3.16. D u r i n g sect ioning o f the midbra in tissue o n the M I C R O M International Cryostat ( H M 5 0 5 E ) , I observed that the rostrocaudal diffusion diameter o f the pontamine sky blue microinjected w i t h the ka in i c ac id was approximately 1.5 to 2 .0mm. In the two representative carp, a l l o f the tissue regions damaged by the ka in ic ac id microinject ions were discrete and h igh ly loca l i zed . The ka in ic ac id lesions were ventral and lateral to the fourth ventr icle; more speci f ica l ly , they were ventral and med ia l to the oculomotor nucleus. T h e lesions extended approximately 1.0mm rostrocaudally, 0 . 1 m m mediolateral ly , and 0 . 8 m m dorsoventral ly. T issue damage was v is ib le on both sides o f the mid l ine , however , the extent o f damage on each side var ied rostrocaudally, ind ica t ing that the bilateral injections m a y have been s l ight ly rostrocaudally asymmetr ica l . In one o f the two representative carp, the ka in i c ac id d id not les ion tissue wi th in the J A B site on both sides o f the mid l ine . T h i s carp breathed ep isodica l ly both before and after the ka in i c ac id microinject ions (Figure 3.17). H o w e v e r , i n the second o f the two representative carp, the ka in ic ac id les ioned tissue wi th in the J A B site on both sides o f the mid l ine . T h i s carp breathed ep isodica l ly before the microinject ions but i n single, isolated breaths after the microinject ions (Figure 3.18). In both carp, the ka in ic ac id microinject ions in i t i a l ly st imulated breathing. In the two representative trout, the tissue regions damaged by the ka in ic ac id microinject ions were also h igh ly loca l i zed and discrete. T h e ka in ic ac id lesions were ventral and lateral to the fourth ventricle and ventral and media l to the nucleus lateralis valvulae. A s i n carp, tissue damage was v i s ib le on both sides o f the mid l ine , a l though the extent o f the damage on each side var ied rostrocaudally, again perhaps indica t ing a sl ight rostrocaudal asymmetry i n the injection sites. T h e lesions extended approximately 1.0mm rostrocaudally, 0 . 1 m m mediolateral ly , and 0 . 8 m m dorsoventral ly, s imi lar to the extent o f the lesions i n carp. In both o f the two representative trout, the ka in ic ac id lesions appeared to be s l ight ly ventral to the J A B site. B o t h trout breathed cont inuously before and after the ka in ic ac id microinject ions (Figure 3.18). In both trout, the ka in i c ac id microinject ions in i t i a l ly st imulated 98 breathing, increasing breathing frequency, ampli tude, and total vent i lat ion. Brea th ing frequency was s t i l l s l ight ly elevated ninety minutes after the microinject ions. rostral caudal rostral Figure 3.16: Location of the kainic acid injection sites relative to the JAB site in trout. (Upper) Sagittal-section of a trout brain (modified from Nieuwenhuys et al., 1997) showing the location of the JAB site (blue) and the rostral and caudal injection sites (red lines). The three injections depths are marked as black circles on these red lines. (Lower) Schematic diagram of cross-sections of a trout midbrain at the level of the rostral (left) and caudal (right) injection sites. Nuclei are yellow and the JAB site is blue. The bilateral injections sites are indicated by vertical red lines, and the three injection depths are marked as black circles on these lines. Abbreviations: Cb = corpus cerebellum, LC = lobus caudalis cerebelli, nLV = nucleus lateralis valvulae, nOC = oculomotor nucleus, RL = recessus lateralis (infundibuli), TO = tectum opticum, V Cb = valvula cerebelli, Ve 4 = fourth ventricle. 99 Figure 3.17: Kainic acid microinjections that did not lesion tissue within the JAB site did not eliminate apneas in a representative carp. Cross-section of the midbrain, at the caudal injection site, of the representative carp for which the kainic acid injections did not eliminate apneas. The caudal injection site is located at approximately 0.6mm rostral to the anterior border of the dorsal surface of the cerebellum. Tissue was sliced into 50 um sections and stained with an FD NeuroSilver Kit I to detect neurodegeneration. The black or dark brown regions of tissue (marked with arrows) were damaged by kainic acid. (Upper right) Schematic diagram of the caudal injection site in carp showing the JAB site (blue) and the approximate sites of tissue destroyed by the kainic acid injections (dark red) in this carp. Nuclei are shown in yellow. Abbreviations: CA = commissure ansulate, Cb = corpus cerebellum, LL = lemniscus lateralis, nLV = nucleus lateralis valvulae, nOC = oculomotor nucleus, TO = tectum opticum, V Cb = valvula cerebelli, Ve 4 = fourth ventricle. (Lower) Plots of interbreath intervals (in seconds) before (left) and after (right) the kainic acid injections. Breathing is episodic both before and after the microinjections, as indicated by the high level of variation in the duration of the interbreath intervals. 100 0 10 20 30 40 50 60 70 80 90 100 Breath Number Before Kainic Acid 30 25 20 15 10 5 0 0 10 20 30 40 50 60 70 80 90 100 Breath Number After Kainic Acid Figure 3.18: Kainic acid microinjections that did lesion tissue in the JAB site eliminated apneas in a representative carp. (Upper left) Cross-section of the midbrain, at the caudal injection site, of the caip for which the kainic acid microinjections eliminated apneas. The caudal injection site is located at approximately 0.6mm rostral to the anterior border of the dorsal surface of the cerebellum. Tissue was sliced into 50 um sections and stained with an FD NeuroSilver Kit I to detect neurodegeneration. The black or dark brown regions of tissue (marked with arrows) were damaged by kainic acid. (Upper right) Schematic diagram of the caudal injection site in carp showing the JAB site (blue) and the approximate sites of tissue destroyed by the kainic acid injections (dark red) in this carp. Nuclei are shown in yellow. Abbreviations: CA = commissure ansulate, Cb = corpus cerebellum, LL = lemniscus lateralis, nLV = nucleus lateralis valvulae, nOC = oculomotor nucleus, TO = tectum opticum, V Cb = valvula cerebelli, Ve 4 = fourth ventricle. (Lower) Plots of interbreath intervals (in seconds) before (left) and after (right) the kainic acid injections. Breathing is episodic before the microinjections, as indicated by the high level of variation in the duration of the interbreath intervals, while after the microinjections, breathing is continuous, as indicated by the relative consistency of the duration of the interbreath intervals. 101 0.8 0.7 0.6 0.5 0.4 0.3 0.2 20 40 60 80 100 Breath Number Before Kainic Acid 120 0.8 (fl 1 0.7 & n r o.6 • | 0.5 1 0.4 fi -0.2 ! 03 *&*&^\<^*&&:Y& » • • • • 20 40 60 80 100 120 Breath Number After Kainic Acid Figure 3.19: Kainic acid microinjections did not lesion tissue within the JAB site and slightly increased breathing frequency in both representative trout (Upper left) Cross-section of a representative trout midbrain at the caudal injection site, located at approximately 0.7mm rostral to the anterior border of the dorsal surface of the cerebellum. Tissue was sliced into 50 um sections and stained with an FD NeuroSilver Kit I to detect neurodegeneration. The black or dark brown regions of tissue (marked with arrows) were damaged by kainic acid. (Upper right) Schematic diagram showing the JAB site (blue) and the approximate sites of tissue destroyed by the kainic acid injections (dark red) at the caudal injection site in trout. Nuclei are shown in yellow. Abbreviations: Cb = corpus cerebellum, LC = lobus caudalis cerebelli, nLV = nucleus lateralis valvulae, nOC = oculomotor nucleus, TO = tectum opticum, V Cb = valvula cerebelli, Ve 4 = fourth ventricle. (Lower) Plots of interbreath intervals (in seconds) before (left) and after (right) the kainic acid injections. Breathing is continuous both before and after the microinjections, as indicated by the consistency of the duration of the interbreath intervals. The average interbreath interval decreased (breathing frequency increased) after the microinjections. 102 Discussion Critique of Methods Estimates for the diffusion radius o f ka in i c ac id f rom an injection site vary depending o n the source. M o s t studies estimate the diffusion radius o f ka in ic ac id f rom the injection site us ing injections o f the same vo lume o f dye or fluourescent microspheres. F o r example , i n one study where the diffusion o f ka in ic ac id was based on diffusion o f fluourescent microspheres, a bolus o f 75n l generated a stained reg ion o f approximately 0 . 4 m m i n diameter ver t ica l ly and 1.4mm i n diameter hor izonta l ly (Sundin et al., 2003). Ano the r study found that 7 0 n l microinject ions o f dye into agar ge l on ly generated a stained reg ion o f 0 .15mm i n diameter w h i l e microinject ions o f 2 0 0 n l o n l y stained a reg ion o f 0 . 2 2 m m i n diameter ( H s i e h et al., 1998). I chose to base m y estimations o f ka in ic ac id diffusion on the results f rom a study where ka in i c ac id was injected into the general v iscera l nucleus o f channel catfish (Sundin et al., 2003) as this study most c lose ly resembled m y o w n . A t each injection site, I injected 7 5 n l o f O . O l m M ka in ic ac id at three depths, separated b y approximately 0 .4mm, g i v i n g a total o f 225n l o f O . O l m M ka in ic ac id (2.25pmol) at each injection site. Because a 75n l bolus o f ka in ic ac id w i l l diffuse approximately 0 .2mm ver t ica l ly f rom the injection site (a total diameter o f 0 .4mm) (Sundin et al., 2003) , the ka in ic ac id injected at the three depths for each site should have diffused dorsoventral ly to cont inuously cover a ver t ical c o l u m n o f 1.2mm. A d d i t i o n a l l y , at each depth, the ka in ic ac id should also have diffused approximately 0 . 7 m m hor izonta l ly (a total diameter o f 1.4mm) (Sundin et al, 2003). In each f i sh , I injected ka in ic ac id i n two bilateral sites, one rostral and one caudal , separated b y 0 .6mm. F r o m these two inject ion sites, the ka in ic ac id should have diffused both rostrocaudally, to cont inuously cover a plane o f approximately 2 . 0 m m i n diameter, and mediolateral ly , to cont inuously cover a plane o f approximately 1.4mm i n diameter. Therefore, I predicted that the total reg ion o f tissue damaged by the ka in ic ac id microinject ions should have been a vo lume extending 1.2mm dorsoventral ly, 2 . 0 m m rostrocaudally, and 1.4mm mediola tera l ly f rom the central coordinates o f the J A B site (approximately 3 .4mm 3 ) , on both sides o f the mid l ine . T h i s predicted vo lume o f tissue damage should have fu l ly encompassed the J A B site, destroying it. Indeed, the pontamine sky blue injected w i t h the ka in ic ac id diffused rostrocaudally f rom the inject ion sites to cover a diameter o f approximately 1.5 to 2 .0mm, w h i c h 103 is consistent w i t h m y predic t ion, as w e l l as w i th the diffusion o f ka in i c ac id and fluourescent microspheres microinjected into the general v iscera l nucleus o f the channel catfish (Sundin et al, 2003) . H o w e v e r , f rom the h is to logica l analysis o f the two representative carp and trout, it appears that the ka in ic ac id d i d not cause tissue damage i n as large a region as predicted. Because not a l l midbra in sections surv ived the si lver staining process intact, a l l estimates o f the extent o f tissue damaged by ka in ic ac id are conservative and approximate. E v e n consider ing this, however , i n both carp and trout the rostrocaudal spread o f tissue damage was on ly ha l f o f what was predicted (1 .0mm i n diameter instead o f 2 . 0 m m i n diameter.) S i m i l a r l y , i n both carp and trout, the dorsoventral spread o f tissue damage was on ly approximately seventy percent o f what was predicted (0 .8mm i n diameter instead o f 1.2mm i n diameter), and the mediolateral spread o f tissue damage was very m i n i m a l , on ly approximately ten percent o f what was predicted (0 .1mm i n diameter instead o f 1.4mm i n diameter.) It appears that the predict ions for the vo lume o f tissue damaged by the ka in ic ac id microinject ions were overestimates. Interestingly, the vo lume o f tissue lesioned by the ka in ic ac id more c lose ly agrees wi th the diffusion o f dye i n agar (Hs ieh et al, 1998) than wi th the dif fus ion o f dye i n the general v iscera l nucleus o f the channel catfish (Sundin et al, 2003). Unfortunately, few studies that use ka in ic ac id to les ion bra in tissue actually report the exact dimensions o f the lesions, on ly p rov id ing the locat ion o f the lesions relative to the targeted sites. H o w e v e r , i n one study that d i d report ka in ic ac id les ion size, 150nl o f I m M ka in ic ac id microinjected into the fastigial nucleus o f the cerebellar deep nuc le i o f the rat created lesions o f approximately 0.4 to 0 . 6 m m i n diameter ( X u et al., 2001) ; the size o f the ka in ic ac id lesions i n m y study (approximately 0.1 to 1.0mm i n diameter) is rough ly consistent wi th this f ind ing . One possible explanat ion for the smaller l e s ion vo lume than estimated based on previous studies is that I microinjected a substantially lower amount o f ka in ic ac id . I injected approximately 2 .25pmol o f ka in i c ac id into each injection site, w h i l e i n other studies the total amount o f ka in ic ac id injection per site ranged f rom lOOpmol ( X u and Frazier , 2000; X u et al., 2001) and 2 3 4 p m o l (Zhang et al, 2004) to as h igh as 4 .8nmol (Hs ieh et al, 1998). Howeve r , I based m y predict ions o n a study i n w h i c h lOOnl o f O . O l m M ka in ic ac id (a total o f l p m o l ) was injected per site, successfully les ion ing the target region (Sundin et al, 2003). A d d i t i o n a l l y , because the ka in ic ac id lesions i n m y study were roughly the same size as i n a study where 104 lOOpmol o f ka in i c ac id was injected per site ( X u et al., 2001), this discrepancy i n concentration does not appear to expla in the smaller region o f the damage that was caused by the ka in ic ac id . H o w e v e r , the ka in ic ac id may have diffused further than is indicated by the size o f the lesions. M y estimations for ka in ic ac id les ion size were based on the diffusion o f ka in ic ac id as measured by the diffusion o f fluorescent microspheres and not f rom actual les ion size (Sundin et al., 2003) . The concentration threshold o f ka in ic ac id necessary to stimulate cel ls m a y be m u c h lower than the concentration necessary to k i l l cel ls . Therefore, even though the regions les ioned by the ka in ic ac id were smaller than predicted, the ka in ic ac id m a y have st imulated neurons w i t h i n a greater vo lume o f tissue, perhaps encompassing the entire predicted region. Indeed, the pontamine sky blue injected w i t h the ka in i c ac id diffused out f rom the inject ion site to stain a region wi th a rostrocaudal diameter o f approximately 1.5 to 2 .0mm, suggesting that the ka in ic ac id may have inf luenced neurons w i t h i n a s imi la r vo lume o f tissue even though it on ly les ioned a m u c h smaller vo lume o f tissue. E a c h time I microinjected ka in ic ac id into a f ish , I moni tored its respiration to determine i f the ka in ic ac id exci ted breathing. Because I was able to excite breathing wi th microinject ions o f ka in ic ac id , an exci to toxic substance, into the midbra in tegmentum o f every f i sh I tested, the ka in ic ac id must have reached a site that was capable o f s t imulat ing respiration. M y assumptions were that this midbra in site that stimulated breathing was the J A B , and that i f the ka in ic ac id had in i t i a l ly st imulated this site, then i t w o u l d eventual ly les ion it. H o w e v e r , the threshold concentration o f ka in ic ac id required to k i l l neurons m a y have been greater than the concentration required to excite but not to k i l l neurons. Therefore, it is possible that the ka in ic ac id may have st imulated the J A B site and exci ted breathing wi thout do ing any damage to the site. In the one carp for w h i c h ka in ic ac id bi la teral ly les ioned the J A B site, apneas were e l iminated, conver t ing the episodic breathing pattern into a continuous one. In the carp for w h i c h the ka in ic ac id d id not bi la teral ly les ion the J A B site, apneas were not e l iminated. In both representative trout, the ka in ic ac id d i d not les ion the J A B site and the continuous breathing pattern was not altered. A l t h o u g h this h i s to logica l data is unfortunately incomplete , it suggests that for those f ish where the ka in ic ac id microinject ions had no effect on episodic breathing, the injections may have missed the J A B site w h i l e i n carp where the microinject ions e l iminated apneas, the injections lesioned the J A B site. C lea r ly , wi thout further experimentat ion, this suggestion is on ly speculative. 105 Interestingly, the ka in ic ac id lesions i n both respresentative trout are c lear ly ventral to the J A B site; however , i n carp, the ka in ic ac id lesions are on the same dorsoventral plane as the J A B site. T h i s discrepancy is most l i k e l y due to the difference between carp and trout midbra in anatomy. In the carp, the tectum op t i cum is supported by the va lvu la cerebel l i , w h i c h are located direct ly ventral to the tectum opt icum. In the trout, however , the v a l v u l a cerebel l i are located more ventral ly, and therefore do not provide phys ica l support to the tectum opt icum. In an intact trout, this lack o f mechanica l support w o u l d not be problematic because the tectum op t i cum w o u l d be supported by cerebral-spinal f l u id ; however , dur ing the ka in ic ac id microinject ions, it was necessary to remove m u c h o f this cerebral-spinal f l u i d to obtain a clear v i e w o f the dorsal surface o f the brain. Because the depth o f the injection sites was calculated based on the distance f rom the dorsal surface o f the midbra in to the J A B site, any sagging o f the tectum op t i cum w o u l d direct ly translate to an injection site that was ventral to the target site. T h i s error highl ights the major p rob lem o f determining internal injection sites based o n external landmarks i n a tissue as soft and mal leable as the brain. C l ea r ly h is to logica l analysis o f each i nd iv idua l tested w o u l d be necessary to ensure that the injections actually les ioned the target site. A l t h o u g h that l eve l o f thorough h is to logica l analysis was intended for this study, for reasons beyond m y cont ro l this was not possible. Time Course of Kainic Acid Effects K a i n i c ac id is an exci to toxic glutamate agonist that k i l l s neuron ce l l bodies but not axons (Bender and Ba i ze r , 1984). T h e two phases o f action o f this exci to toxic substance are thus the excitatory phase, when neurons are stimulated, and the toxic phase, when neurons are k i l l e d . Therefore, microinject ions o f ka in ic ac id into the J A B site should in i t i a l ly excite, then k i l l the neurons o f this site. I r emoved and preserved the midbrains o f the four representative f ish ninety minutes to two hours after injecting the ka in ic ac id ; lesions were c lear ly v i s ib le i n a l l four midbrains , indica t ing that after ninety minutes, the ka in ic ac id was i n its toxic phase. T h i s f ind ing agrees w i t h those o f previous studies, w h i c h found that i n rats, the duration o f the in i t i a l excitatory per iod can last f rom thirty (Zhang et ai, 2004) to s ixty ( X u and Frazier , 2000) to ninety minutes ( X u et al, 2001) before f ina l ly destroying the affected neurons. F r o m this evidence, I can conclude that thirty minutes after the ka in ic ac id injections, when I recorded breathing i n n o r m o x i a and in i t ia l hyperoxia , the neurons c o u l d s t i l l be i n the excitatory phase o f 106 the ka in ic ac id . H o w e v e r , two hours after the ka in ic ac id microinject ions, when I recorded breathing i n prolonged hyperoxia , and two and a ha l f hours after the ka in ic ac id injections, when I again recorded breathing i n normoxia , the excitatory phase o f the ka in ic ac id should have most ly dissipated and the toxic phase should have begun. Decerebrate/Spinalectomized Carp Breathe Normally In normoxia , decerebrate/spinalectomized carp had a breathing frequency o f 50.6 ± 4.0 breaths/minute, w h i c h is over twice that o f no rmox ic , intact carp (approximately 20 breaths/minute) (Takeda, 1990; Glass et al, 1991; L u m s d e n , 1996). In fact, no rmox ic decerebrate/spinalectomized carp had a comparable b reaming frequency (approximately 50 breaths/minute) to h y p o x i c intact carp (Glass et al., 1991). O n l y 35.7 % o f the normoxic decerebrate/spinalectomized carp breathed episodica l ly ; however , i n previous studies, intact carp d i d not appear to breath cont inuously at PO2 levels higher than approximately 50 m m H g ( L o m h o l t and Johansen, 1979; Glass etal., 1991). Therefore, decerebrate/spinalectomized carp appear to experience a higher leve l o f respiratory dr ive than do intact carp. Sp ina lec tomiz ing carp just caudal to the sku l l does not affect respirat ion (Juch and B a l l i n t i j n , 1983), suggesting that the cerebrum normal ly depresses respiratory dr ive , or at least breathing frequency, al though without further studies this is purely speculative. Howeve r , decerebrate/spinalectomized carp responded normal ly to hyperoxia , decreasing total vent i la t ion b y approximately 63.3 ± 4.9 percent, consistent w i t h the decrease i n total vent i lat ion o f intact carp i n hyperox ia (approximately 65 %) (Takeda, 1990). A d d i t i o n a l l y , a l l but one o f the decerebrate/spinalectomized carp breathed ep isodica l ly i n hyperoxia , consistent w i th previous studies (Lumsden , 1996). Decerebrat ion and spinalectomy d i d not affect the proport ion o f carp breathing ep isodica l ly i n hyperox ia when compared to intact carp, w h i c h is consistent w i th previous studies w h i c h found that decerebration d i d not affect the distr ibution o f breathing pattern f rom continuous to episodic i n dogfish (Roberts and B a l l i n t i j n , 1988) or i n tambaqui ( R e i d et al., 2003). O v e r a l l , these results suggest that al though decerebrate/spinalectomized carp have a higher respiratory dr ive than do intact carp, they respond normal ly when respiratory dr ive is decreased by hyperoxia . 107 The Excitatory Phase: Kainic Acid Microinjections in Carp Initially Stimulate Breathing The ka in ic ac id microinject ions in i t i a l ly st imulated respiration, almost immedia te ly increasing breathing frequency, ampli tude, and total venti lat ion. T h i s excitatory respiratory response is consistent w i th that o f other studies i n w h i c h ka in i c ac id was injected into bra in sites that excite breathing when e lect r ica l ly stimulated. F o r example , lOOnl injections o f 2 . 3 4 m M ka in ic ac id into the facia l nucleus o f rats dramat ical ly increased both breathing frequency and ampli tude o f phrenic nerve discharge (Zhang et al, 2004). S i m i l a r l y , lOOnl injections o f I m M ka in ic acid into the nucleus gigantocel lular is o f rats in i t i a l ly increased tidal vo lume and minute venti lat ion ( X u etal., 2001) wh i l e lOOnl injections o f I m M ka in ic ac id into the fastigial nucleus o f rats also in i t i a l ly st imulated breathing frequency ( X u and Frazier , 2000). T h i s in i t ia l excitatory respiratory response was expected, as ka in ic ac id is a glutamate agonist that stimulates neurons. The J A B site is k n o w n to influence the respiratory sites o f the medu l l a and can be used to entrain the breathing rhy thm w i t h phasic electr ical s t imulat ion or to initiate and pro long breathing episodes wi th b r i e f (100 to 1500ms) tonic s t imulat ion (Juch and B a l l i n t i j n , 1983). T h i s evidence suggests that tonica l ly exc i t ing the J A B site w i th ka in ic ac id in i t i a l ly increases breathing frequency b y shortening apneas and p ro long ing episodes. H o w e v e r , thirty minutes after the microinject ions , the ka in ic ac id s ignif icant ly increased breathing frequency even i n cont inuously breathing carp, suggesting that a tonic exci ta t ion o f the J A B site can elevate breathing frequency not jus t b y shortening apneas i n the episodic pattern but also by decreasing the interbreath interval i n the continuous pattern. Because the J A B site normal ly discharges before each breath, even when carp are breathing cont inuously (Juch and Ba l l an t i jn , 1983), it is not surprising that heav i ly exc i t ing the J A B site should increase breathing frequency i n cont inuously breathing carp as w e l l as i n ep isodica l ly breathing carp. T h i s evidence suggests that normal ly the J A B site may be i n v o l v e d i n regulat ing breathing frequency. The Toxic Phase: Effects in Carp of the Kainic Acid Microinjections Over Time In no rmoxic decerebrate/spinalectomized carp, thirty minutes after the ka in i c ac id microinject ions , breathing frequency and total vent i lat ion were s t i l l s ignif icant ly higher than control levels . H o w e v e r , two and a ha l f hours after the ka in ic acid injections, both breathing frequency and total vent i lat ion had returned to control levels i n normoxia . Af ter the first five 108 minutes o f hyperox ia (thirty-five minutes after the injections), breathing frequency and total vent i la t ion remained s ignif icant ly higher than control levels for in i t i a l hyperoxia . H o w e v e r , al though breathing frequency remained s ignif icant ly faster after ninety minutes o f hype rox ia (a total o f two hours after the ka in ic ac id injections), breathing ampli tude decreased, reducing total vent i la t ion to control levels for prolonged hyperoxia . Therefore, a l though the ka in ic ac id microinject ions in i t i a l ly exci ted breathing, this effect had most ly dissipated two hours after the injections. T h i s response to ka in ic ac id over time is consistent w i t h other studies where ka in ic ac id was microinjected into sites k n o w n to stimulate breathing. F o r example , ka in i c ac id microinject ions i n the facia l nucleus o f rats stimulated neurons for approximately thirty minutes before f ina l ly destroying them (Zhang et al, 2004). S i m i l a r l y , ka in ic ac id microinject ions into the nucleus gigantocel lular is o f rats in i t i a l ly st imulated neurons, increasing t idal vo lume and minute vent i la t ion; ninety minutes after the injections, these respiratory variables had returned to cont ro l levels ( X u et al, 2001). L i k e w i s e , ka in ic ac id microinject ions into the fastigial nucleus o f rats in i t i a l ly st imulated breathing frequency; after s ixty minutes, breathing had returned to normal levels ( X u and Frazier , 2000). The h i s to log ic ia l evidence o f tissue lesions ninety minutes to two hours after the ka in ic ac id injections suggests respiration returned to control levels not on ly because the ka in ic ac id was no longer exc i t ing the neurons but also because i t had k i l l e d at least some o f these neurons. Therefore, the return o f respiration to control levels after the ka in ic ac id had les ioned the J A B site suggests that al though the J A B site excites breathing when exci ted w i t h a chemica l tonic st imulus, it does not normal ly influence the total l eve l o f respiratory dr ive. Interestingly, al though total vent i la t ion had returned to control levels i n p ro longed hyperoxia , two hours after the ka in ic ac id microinject ions, breathing frequency remained signif icant ly elevated compared to control hyperoxic levels , wh i l e ampli tude was substantially but not s ignif icant ly reduced f rom control hyperoxic levels . N o r m a l l y hyperoxic carp reduce total vent i la t ion by decreasing both breathing frequency and ampli tude. Af te r the ka in ic ac id had les ioned the J A B site, however , breathing frequency remained at no rmoxic levels , ind ica t ing that the hyperoxic carp on ly reduced total vent i lat ion b y decreasing ampli tude. Therefore, al though total vent i la t ion was not affected by the les ion ing o f the J A B site, the respiratory pattern was affected when respiratory dr ive was reduced. In no rmox ia , after the J A B site was lesioned, al though total vent i lat ion remained at control levels , breathing frequency was s t i l l s l ight ly 109 increased w h i l e ampli tude was s t i l l s l ight ly decreased. Th i s evidence suggests that the J A B site m a y also affect the respiratory pattern when at normal levels o f respiratory dr ive ; however this effect is certainly less pronounced than when respiratory dr ive is reduced. Therefore, the J A B site m a y normal ly p lay a role i n ba lanc ing the relative contributions o f breathing frequency and ampli tude to total vent i la t ion, decreasing breathing frequency w h i l e increasing breathing ampli tude. T h i s role is very s imi lar to that o f the pontine respiratory group i n mammals , except that the pontine respiratory group normal ly increases frequency w h i l e decreasing ampli tude ( N g a i and W a n g , 1957; Tang , 1967; St .John et al, 1972). In conc lus ion , these results suggest that the J A B site does not normal ly influence the total l eve l o f respiratory dr ive , but that it does regulate breathing pattern, decreasing frequency and increasing amplitude. Lesioning the JAB Site Affects EpisodicBreathing in Carp K a i n i c ac id microinject ions e l iminated apneas i n fifty percent o f carp. W e r e the apneas e l iminated because the ka in ic ac id microinject ions damaged a site necessary for generating episodic breathing or because the ka in ic ac id elevated respiratory dr ive to such an extent that the carp were no longer able to meet their f ic t ive metabol ic needs by breathing ep isodica l ly? A l t h o u g h the ka in ic ac id in i t i a l ly exci ted breathing, this effect had dissipated two hours after the ka in ic ac id injections, when the ka in ic ac id was i n its toxic phase and total vent i lat ion had returned to control levels . T h i s evidence indicates that two hours after the microinject ions, respiratory dr ive had also returned to control levels , suggesting that the J A B site plays a role i n regulat ing the episodic breathing pattern i n carp. Howeve r , w h y then were apneas e l iminated for on ly fifty percent o f carp? It is possible that the ka in ic ac id microinject ions on ly e l iminated apneas i n the carp for w h i c h the ka in ic ac id les ioned tissue i n the J A B site, w h i l e carp for w h i c h the ka in ic ac id d i d not les ion tissue i n the J A B site cont inued to breathe episodica l ly . T h i s poss ib i l i ty w o u l d i m p l y that the J A B site is essential for generating the episodic breathing pattern, and that damaging it w i t h ka in ic ac id w o u l d therefore complete ly el iminate apneas. The h is to logica l data showed that i n the carp for w h i c h the microinject ions e l iminated the episodic breathing pattern, the ka in ic ac id les ioned tissue i n the J A B site, w h i l e i n the carp for w h i c h the microinject ions d i d not el iminate episodic breathing, the ka in ic ac id d id not bi la teral ly les ion tissue i n the J A B site. These results are 110 consistent w i th the hypothesis that the J A B site is essential for episodic breathing. T h i s hypothesis is also consistent w i t h evidence f rom previous studies on both carp (Juch and B a l l i n t i j n , 1983) and ca iman (Nai feh et al, 1971a; N a i f e h et al, 1971b). H o w e v e r , because the h is to logica l evidence is incomplete , it is imposs ib le to fu l ly support this hypothesis wi thout further research. In pro longed hyperox ia after the ka in ic ac id injections, breathing frequency was elevated to the l eve l o f breathing frequency i n control no rmoxia . O n e explanat ion for this increase i n breathing frequency is that once that J A B site was lesioned, the carp c o u l d no longer breathe ep isodica l ly , and thus the absence o f apneas caused breathing frequency to increase. T h i s explanat ion w o u l d be consistent w i t h the hypothesis, ment ioned above, that the J A B site is essential for p roduc ing episodic breathing. H o w e v e r , it is also possible that i f the J A B site normal ly decreases breathing frequency and increases ampli tude, when i t was lesioned, breathing frequency increased to such a leve l that apneas were s imp ly unnecessary, not imposs ib le , to produce. The involvement o f the J A B site i n regulat ing the episodic breathing pattern may therefore s imp ly be an extension o f its role i n regulat ing breathing frequency. H o w e v e r , because the J A B site is capable o f entraining the respiratory rhy thm and normal ly discharges before each episode as w e l l as before each i nd iv idua l breath (Juch and B a l l i n t i j n , 1983), i t is also possible that the J A B site normal ly both decreases breathing frequency and, when respiratory dr ive is l o w , further reduces frequency b y p roduc ing the episodic breathing pattern. In conc lus ion , al though the J A B site is c lear ly i n v o l v e d i n regulat ing the episodic breathing pattern, it is currently unclear whether it is essential for the product ion o f the episodic breathing pattern or whether it s imp ly a l lows the expression o f the episodic breathing pattern by decreasing breathing frequency. Decerebrate/Spinalectomized Trout Breathe Normally Decerebrate/spinalectomized trout breathed cont inuously i n normoxia , i n agreement w i t h previous f indings for intact trout (Randa l l and Jones, 1973; W o o d and Jackson, 1980). In normoxia , decerebrate/spinalectomized trout had a breathing frequency o f 85.0 ± 4.6 breaths/min, w h i c h is faster than that o f intact trout f rom previous studies (Cameron and D a v i s , 1970; R a n d a l l and Jones, 1973; W o o d and Jackson, 1980; Per ry et al, 1992; Per ry and G i l m o u r , 1996). H o w e v e r , because decerebrate/spinalectomized trout also decreased breathing ampli tude, 111 total vent i la t ion was not increased by this e levat ion i n breathing frequency. Decerebrate trout also decreased breathing amplitude and increased breathing frequency compared to intact trout (Chapter 2), suggesting that i n trout, the cerebrum may normal ly p rov ide input to the respiratory centres o f the medu l l a that increases breathing ampli tude and decreases frequency. These f indings agree wi th those for decerebrate/spinalectomized carp, w h i c h also breathed more rapid ly than d i d intact carp f rom previous studies (Takeda, 1990; Glass et al, 1991; L u m s d e n , 1996). T a k e n together, this evidence suggests that i n teleost f ish , the cerebrum may normal ly depress breathing frequency and elevate breathing amplitude. In hyperoxia , decerebrate/spinalectomized trout decreased their total vent i la t ion by approximately 58 ± 6 % ; W o o d and Jackson (1980) found a s imi la r decrease o f approximately 60 % i n total vent i lat ion i n hyperoxic intact trout. A d d i t i o n a l l y , consistent w i t h intact trout f rom previous studies ( W o o d and Jackson, 1980; K i n k e a d and Perry, 1991), decerebrate/spinalectomized trout reduced their total vent i la t ion p r imar i ly b y decreasing breathing ampli tude. A p p r o x i m a t e l y thirty percent o f decerebrate/spinalectomized trout breathed ep isodica l ly dur ing pro longed exposure to hyperox ia w h i l e the majori ty cont inued to breathe cont inuously , consistent w i t h intact trout i n a previous study ( W o o d and Jackson, 1980). T h i s is also consistent w i t h previous studies that found that decerebration d i d not affect the dis t r ibut ion o f breathing pattern f rom episodic to continuous i n dogfish (Roberts and B a l l i n t i j n , 1988) or tambaqui ( R e i d et al., 2003). These results suggest that decerebrate/spinalectomized trout breathe normal ly i n both no rmox ia and hyperoxia . The Excitatory Phase: Kainic Acid Microinjections in Trout Initially Stimulate Breathing The ka in ic ac id microinject ions stimulated breathing frequency, ampli tude, and total vent i la t ion almost instantaneously in decerebrate/spinalectomized trout. Trou t showed the same immediate excitatory response to the ka in ic ac id injections as d i d carp. A s discussed for carp, this excitatory respiratory response is expected and is consistent w i th results f rom previous studies ( X u and Frazier , 2000; X u et al, 2001 ; Z h a n g et al, 2004) . F o r the subset o f trout that breathed episodica l ly , thirty minutes after the ka in i c ac id microinject ions , breathing frequency, ampli tude, and total vent i la t ion were a l l s t i l l substantially elevated. T h i s response agrees wi th the response seen i n carp for breathing frequency and total vent i la t ion; however , for carp, breathing ampli tude was no longer elevated thirty minutes after 112 the microinject ions . Nonetheless, the f ind ing that a tonica l ly exc i t ing the J A B site increased breathing frequency i n both carp and trout suggests that normal ly the J A B site m a y regulate frequency i n both species o f f ish. H o w e v e r , thirty minutes after the microinject ions , this exci tatory effect had almost comple te ly dissipated for the subset o f trout that on ly breathed cont inuously: breathing frequency, ampli tude, and total vent i la t ion were unchanged f rom cont ro l values. It is possible that i n trout that do not breathe episodica l ly , the J A B site is less inf luent ia l ; however , it is also possible that a lower concentration o f ka in ic ac id diffused to the J A B site i n this subset o f trout than d i d i n the episodic subset o f trout. The Toxic Phase: Effects Over Time of Kainic Acid Microinjections in Trout In decerebrate/spinalectomized trout, thirty minutes after the ka in ic ac id microinject ions, breathing frequency and total vent i la t ion were s t i l l s ignif icant ly elevated i n normoxia . T w o and a ha l f hours after the ka in ic ac id microinject ions, when the ka in ic ac id was i n its toxic phase, breathing frequency was s t i l l substantially elevated; however , a reduction i n breathing ampli tude compensated for the increased frequency, returning total vent i la t ion to control levels . In the first f ive minutes o f hyperoxia , thir ty-five minutes after the microinject ions , breathing frequency and total vent i la t ion were s t i l l s ignif icant ly elevated. Af te r ninety minutes o f hyperoxia , two hours after the ka in ic ac id microinject ions , breathing frequency was s t i l l s ignif icant ly elevated, however , total vent i la t ion had returned to control levels because o f a decrease in breathing ampli tude. T h i s response to the toxic phase o f the ka in ic ac id is very s imi la r to that seen i n carp, where total vent i la t ion also returned to control levels i n both hyperox ia and n o r m o x i a because a decrease i n breathing ampli tude compensated for an increase i n breathing frequency. C lea r ly , as for carp, because total vent i la t ion returned to cont ro l levels i n trout when the ka in ic ac id entered its toxic phase, the J A B site does not normal ly inf luence the total l eve l o f respiratory dr ive. H o w e v e r , as for carp, this site does appear to be i n v o l v e d i n regulat ing breathing pattern. These results suggest that i n both carp and trout, the J A B site normal ly decreases breathing frequency and increases breathing ampli tude. T h e role o f this site appears to be consistent w i th that o f the pontine respiratory group i n mammals , except that the pontine respiratory group has the opposite effect on breathing, increasing frequency w h i l e decreasing ampli tude ( N g a i and W a n g , 1957; Tang , 1967; S t J o h n etal., 1972). 113 Kainic Acid Microinjections into the Midbrain Tegmentum Appear to Affect Episodic Breathing in Trout T h e ka in ic ac id microinject ions affected the episodic breathing pattern i n three o f the four trout that breathed episodica l ly . H o w e v e r , the ka in ic ac id on ly comple te ly e l iminated apneas i n one o f these three trout. O n e poss ib i l i ty is that for this one trout, enough ka in ic ac id diffused into the J A B site to k i l l the neurons there, w h i l e for the two trout where ka in i c ac id on ly decreased the occurrence o f apneas, the concentration o f ka in ic ac id that diffused into the J A B site was o n l y sufficient to k i l l a fraction o f the neurons there. H o w e v e r , wi thout further h i s to logica l evidence, this is purely speculative. Because so few trout breathed ep isodica l ly , i t w o u l d be di f f icul t to co l lec t a sufficient sample size to make any defini t ive conclus ions about the effects o f the J A B site on episodic breathing. T h e episodic subset o f trout responded s imi la r ly to carp i n the toxic phase o f the ka in ic ac id . In both carp and trout, the toxic phase o f the ka in ic ac id increased breathing frequency but reduced amplitude such that total vent i la t ion remained unchanged. Therefore, i n both carp and trout, al though the toxic phase o f the ka in ic ac id altered breathing pattern, it d i d not appear to elevate respiratory drive. H o w e v e r , i n the one trout for w h i c h the microinject ions e l iminated apneas, ka in ic ac id increased breathing frequency, ampli tude, and total vent i lat ion substantially. E v e n after the excitatory effects o f the ka in ic ac id should have dissipated, total vent i la t ion remained elevated. A d d i t i o n a l l y , the pattern o f frequency and ampli tude i n hyperox ia was unusual and erratic (Figure 3.14). Because the on ly trout for w h i c h the microinject ions e l iminated episodic breathing showed such an unusual breathing pattern even two hours after the ka in ic ac id microinject ions, it is d i f f icul t to draw any conclus ions about the effects o f the J A B site on the episodic breathing pattern i n trout. Conclusions In conc lus ion , decerebrate/spinalectomized carp and trout breathe normal ly i n n o r m o x i a and hyperoxia . M y results f rom these f ish suggest that i n both carp and trout, the J A B site does not no rmal ly influence the total l eve l o f respiratory dr ive. In both carp and trout, the J A B site is c lear ly i nvo l ve d i n regulat ing breathing pattern, normal ly decreasing breathing frequency and increasing breathing ampli tude. The J A B site is also i n v o l v e d i n regulat ing episodic breathing; 114 however , it is unclear whether i t s imp ly a l lows the expression o f this pattern by decreasing breathing frequency or whether it act ively produces this pattern. A l t h o u g h these results are consistent w i th the hypothesis that the J A B site is essential for the product ion o f the episodic breathing pattern, wi thout further research, this is purely speculative. 115 References A b a d i e V , Champagnat J , For t in G (2000) Branch iomoto r activities i n mouse embryo . N e u r o Repor t 11: 141-145. A d r i a n E D , Buy tend i jk F J J (1931) Potent ial changes i n the isolated brain stem o f the goldf i sh . J P h y s i o l 71 : 121-135. B a l l i n t i j n C M (1982) N e u r a l control o f respiration i n fishes and mammals , pp. 127-140 In: Exogenous and endogenous influences o n metabol ic and neural control . A d d i n k A D F , Spronk N . ed. Permagon Press, N e w Y o r k . B a l l i n t i j n C M (1984) The respiratory funct ion o f g i l l f i lament muscles i n the carp. Resp i r P h y s i o l 60: 59-74. B a l l i n t i j n C M (1988) E v o l u t i o n o f central nervous control o f vent i la t ion i n vertebrates, pp. 3-27 In: T h e neurobio logy o f the cardio-respiratory system. T a y l o r E W (ed.) Manches ter Un ive r s i t y Press, Manchester . B a l l i n t i j n C M , A l i n k G M (1977) Identif ication o f respiratory motor neurons i n the carp and determination o f their f i r ing characteristics and interconnections. B r a i n Res 136: 261-276. B a l l i n t i j n C M , Hughes G M (1965) The muscular basis o f the respiratory pumps i n the trout. J E x p B i o l 43 : 349-362. B a l l i n t i j n C M , Le i t en P G M , Juch P J W (1979) Respiratory neuron act ivi ty i n the mesencephalon, diencephalon, and cerebel lum o f the carp. J C o m p P h y s i o l 133: 131-139. B a l l i n t i j n C M , Punt G J (1985) G i l l arch movements and the funct ion o f the dorsal g i l l arch muscles i n the carp. Resp i r P h y s i o l 60: 39-57 B a m f o r d O S ( 1 9 7 4 A ) O x y g e n reception i n ra inbow trout (Salmo gairdneri). C o m p B i o c h e m P h y s i o l 4 8 A : 69-76. B a m f o r d O S (1974B) Respira tory neurones i n ra inbow trout (Salmo gairdneri). C o m p B i o c h e m P h y s i o l 4 8 A : 77-83. Bende r D B , B a i z e r JS (1984) Anterograde degeneration i n the superior co l l i cu lus f o l l o w i n g ka in ic ac id and radiofrequency lesions o f the macaque pulvinar . J C o m p N e u r o l 228:284-298. B u r l e s o n M L , M i l s o m W K (1993) Sensory receptors i n the first g i l l arch o f ra inbow trout. R e s p i r P h y s i o l 9 3 : 9 7 - 1 1 0 . . , B u r l e s o n M L , M i l s o m W K (1995) Cardio-vent i la tory control i n ra inbow trout: I. Pha rmaco logy 116 o f branchia l , oxygen-sensi t ive chemoreceptors. Resp i r P h y s i o l 100: 231-238. B u r l e s o n M L , Smi th R L (2001) Centra l nervous control o f g i l l f i lament muscles i n channel catfish. Resp i r P h y s i o l 126: 103-112. B u r t o n R F (1975) R i n g e r solutions and phys io log ica l salines. W r i g h t - Scientechnica, B r i s t o l . C a m e r o n J N , D a v i s J C (1970) Gas exchange i n ra inbow trout (Salmo gairdneri) w i t h va ry ing b l o o d oxygen capacity. J F i s h Res B d Canada 27: 1069-1085. Champagnat J , For t in G (1997) P r i m o r d i a l respiratory-l ike rhythm generation i n the vertebrate embryo. Trends N e u r o s c i 20: 119-124. Chatburn J (2004) Respiratory pattern format ion i n the bu l l f rog (Rana catesbieana). M S c Thes is , Un ive r s i t y o f B r i t i s h C o l u m b i a . D a x b o e c k C , H o l e t o n G F (1978) O x y g e n receptors i n the ra inbow trout, Salmo gairdneri. C a n J Z o o l 56: 1254-1259. de Graa f P J F , B a l l i n t i j n C M (1987) Mechanoreceptor act ivi ty i n the g i l l s o f the carp. II. G i l l arch proprioceptors. R e s p i r P h y s i o l 69: 183-194 de G r a a f P J F , Roberts B L (1991) Entrainment o f the breathing rhy thm o f the carp by imposed osc i l la t ion o f the g i l l . J exp B i o l 155: 93-102. Dejours P , T o u l m o n d A , Truchot J P (1977) The effect o f hyperoxia on the breathing o f marine fishes. C o m p B i o c h e m P h y s i o l 5 8 A : 409-411 . D u n e l - E r b S, B a i l l y Y , Laurent P (1993) Pattern o f g i l l innervat ion i n two teleosts, the perch and the trout. C a n J Z o o l 71 : 18-25. E d g e w o r t h F H (1933) T h e crania l muscles o f vertebrates. Cambr idge Un ive r s i t y Press, Cambr idge . E k s t r o m P (1987) Dis t r ibu t ion o f chol ine acetyltransferase-immunoreactive neurons i n the bra in o f a cyp r in id teleost (Phoxinus phoxinus L . ) . J C o m p N e u r o l 256: 494-515. F e l d m a n J L , M c C r i m m o n D R (1999) Neu ra l control o f breathing, pp. 1063-1090 In: Fundamenta l neuroscience. Z i g m o n d M J , B l o o m F E , Land i s S C , Roberts J L , Squire L R (eds) A c a d e m i c Press, Toronto . Fo r t i n G , Champagnat J , L u m s d e n A (1994) Onset and maturation o f branchio-motor activities i n the c h i c k hindbrain. NeuroRepor t 5: 1149-1152. Fo r t i n G , Fusao K , L u m s d e n A , Champagnat J (1995) R h y t h m generation i n the segmented h indbra in o f ch i ck embryos. J P h y s i o l 486.3 : 735-744. For t in G , Jungbluth S, L u m s d e n A , Champagnat J (1999) Segmental specif icat ion o f G A B A e r g i c 117 i nh ib i t ion dur ing development o f h indbra in neural networks. N a t N e u r o s c i 10: 873-877. Fo r t i n G , de l T o r o E D , A b a d i e V , Guimaraes L , Fou tz A S , Denavi t -Saubie M , R o u y e r F , Champagnat J (2000) Genet ic and developmental models for the neural control o f breathing i n vertebrates. Resp P h y s i o l 122: 247-257. F u n g M - L , St. John W M (1994) Separation o f mul t ip le functions i n venti latory control o f pneumotaxic mechanisms. Resp i r P h y s i o l 96: 83-98. F u n g M - L , W a n g W , St. John W M (1994) Involvement o f ponti le N M D A receptors i n inspiratory termination i n rats. Resp i r P h y s i o l 96: 177-188. Ga rgag l i on i L H , M e i e r J T , B r a n c o L G S , M i l s o m W K (in preparation) M o d u l a t i o n o f respiratory pattern by midbra in sites i n anuran amphibians. Glass M L , R a n t i n F T , V e r z o l a R M M , Fernandes M N , K a l i n i n A L (1991) Cardio-respiratory synchronizat ion and myoca rd ia l function i n hypox ic carp, Cyprinus carpio L . J F i s h B i o l 39: 143-149. Har r i s M B , M i l s o m W K (2003) Apneus i s fo l lows disrupt ion o f N M D A - t y p e glutamate receptors i n vagotomized ground squirrels. Resp i r P h y s i o l N e u r o b i 134: 191-207. H s i e h J H , C h a n g Y C , S u C K , H w a n g J C , Y e n C T , C h a i C Y (1998) A single minute les ion around the ventral respiratory group i n medu l l a produces fatal apnea i n cats. J A u t o n N e r v Syst 7 3 : 7 - 1 8 . H s i e h J H , C h u T C , Y e n C T , C h a i C Y (2000) Effects o f chemica l l es ion ing the dorsal or rostral ventrolateral medu l l a o n glutamate-induced cardiovascular actions o f the pontine gigantocel lular tegmental f ie ld and lateral tegmental f ie ld o f the cat. A c t a Z o o l o g i c a T a i w a n i c a 11(1): 17-32. Hughes G M , Shel ton G (1962) Respiratory mechanisms and their nervous control i n f ish . A d v C o m p P h y s i o l B i o c h e m 1: 275-364 Hukuha ra T , O k a d a H (1956) O n the automatici ty o f the respiratory centres o f the catfish and crucian carp. Jap J P h y s i o l 6: 313-320. H y d e I H (1904) L o c a l i z a t i o n o f the respiratory centre i n the skate. A m J P h y s i o l 10: 236-258. Juch P J W (1981) Mechanoreceptor signals processed by mesencephal ic t r igeminal neurons i n the carp. J C o m p P h y s i o l 141: 157-162. Juch and B a l l i n t i j n (1983) Tegmenta l neurons con t ro l l ing medul la ry respiratory centre act ivi ty i n the carp. Resp i r P h y s i o l 51 : 95-107. K a n w a l J S , C a p r i o J (1987) Centra l projections o f the glossopharyngeal and vagal nerves i n the channel catfish, Ictalurus punctatus: C lues to differential processing o f v iscera l inputs. J C o m p N e u r o l 264: 216-230. 118 K a w a s a k i R (1979) Brea th ing rhythm-generation i n the adult lamprey Entosphenus japonicus. J p n J P h y s i o l 29: 327-338. K in tne r C , L u m s d e n A (1999) N e u r a l induc t ion and pattern formation, pp. 417-450 In: Fundamenta l neuroscience. Z i g m o n d M J , B l o o m F E , L a n d i s S C , Roberts J L , Squire L R (eds) A c a d e m i c Press, Toronto . K i n k e a d R , Per ry S F (1991) T h e effects o f catecholamines o n vent i la t ion i n ra inbow trout dur ing h y p o x i a or hypercapnia . Resp i r P h y s i o l 84: 77-92. L o m h o l t J P , Johansen K (1979) H y p o x i a acc l imat ion i n carp - h o w it affects 0 2 uptake, vent i la t ion, and 0 2 extraction f rom water. Resp i r P h y s i o l 52: 38-49. L u i t e n P G M (1976) A somatotopic and functional representation o f the respiratory muscles i n the t r igeminal and facia l motor nuc le i o f the carp (Cypr inus carpio L . ) . J C o m p N e u r o l . 166: 191-200. L u i t e n P G M , V a n D e r Pers J N C (1977) T h e connections o f the t r igeminal and facia l motor nuc le i i n the bra in o f the carp (Cyprinus carpio L . ) as revealed by anterograde and retrograde transport o f H R P . J C o m p N e u r o l 174: 575-590. L u m s d e n A L (1996) Oxygen-sens i t ive chemoreceptors and cardioventi latory control i n carp. M S c Thesis , Un ive r s i t y o f B r i t i s h C o l u m b i a . M c L e a n H A , K i m u r a N , K o g o N , Perry S F , Remmers J E (1995a) F i c t i ve rhy thm i n the isolated brainstem o f frogs. J C o m p P h y s i o l A 176: 703-713. M e e k J , N ieuwenhuys R (1997) Holosteans and Teleosts. pp. 759-937 In: T h e Cent ra l Ne rvous Sys tem o f Vertebrates. N ieuwenhuys R , T e n Donke laa r H J , N i c h o l s o n C . eds. Springer V e r l a g , N e w Y o r k . M e l l e n N M , J anczewsk i W A , B o c c h i a r o C M , F e l d m a n J L (2003) Op io id - induced quantal s l o w i n g reveals dual networks for respiratory rhy thm generation. N e u r o n 37: 821-826. M i l s o m W K (1984) The interrelationship between pu lmonary mechanics and the spontaneous breathing patter i n the T o k a y l i za rd , Gekko gecko. J E x p B i o l 113: 203-214. M i l s o m W K (1991) Intermittent breathing i n vertebrates. A n n u R e v P h y s i o l 53: 87-105. M i l s o m W K (1996) C o n t r o l o f breathing i n f ish: ro le o f chemoreceptors. pp 359-377 In: P h y s i o l o g y and biochemis t ry o f the fishes o f the A m a z o n . V a l A L , A l m e i d a - V a l V M F , R a n d a l l D J (eds) I N P A , Manaus , B r a z i l . M i l s o m W , Cas t e l l i n i M , Harr i s M , Cas te l l in i J , Jones D , Berger R , B h a r m a S, R e a L , Cos ta D (1996) Effects o f h y p o x i a and hypercapnia o n patterns o f sleep-associated apnea i n elephant seal pups. A m J P h y s i o l 271 : R 1 0 1 7 - R 1 0 2 4 . 119 M i l s o m W K , Chatburn J , Z i m m e r M B (2004) Pont ine influences on respiratory control i n ectothermic and heterothermic vertebrates. Resp i r P h y s i o l 143: 263-280. M i l s o m W K , Har r i s M B , R e i d S G (1997) D o descending influences alternate to produce episodic breathing? Resp i r P h y s i o l 110: 307-317. M i l s o m W K , V i t a l i s T Z (1984) Pu lmonary mechanics and the w o r k o f breathing i n the l i za rd , Gekko gecko. J E x p B i o l 113: 187-202. N a i f e h K H , Hugg ines S E , H o f f H E (1971a) Effects o f brainstem section on respiratory patterns o f c rocod i l i an reptiles. Resp i r P h y s i o l 13: 186-197. N a i f e h K H , Hugg ins S E , H o f f H E (1971b) Study o f the cont ro l o f c rocod i l i an respiration by anaesthetic dissect ion. Resp i r P h y s i o l 12: 251-260. N g a i S H and W a n g S C (1957) Organiza t ion o f central respiratory mechanisms i n the brain stem o f the cat: loca l iza t ion by s t imulat ion and destruction. A m J P h y s i o l 190: 343-349. N i c o l S, A n d e r s o n N A (2003) C o n t r o l o f breathing i n the echidna (Tackyglossus aculeatus) dur ing hibernat ion. C o m p B i o c h e m P h y s i o l A 136(4): 917-925. N ieuwenhuys R , Pouwe l s E (1983) The brain stem o f Ac tonopte ryg ian fishes, pp. 25-87 In: F i s h Neurob io logy : V o l u m e 1: B r a i n stem and sense organs. Northcut t R G , D a v i s R E eds. Un ive r s i t y o f M i c h i g a n Press, A n n A r b o r . N i l s s o n S (1984) Innervation and pharmacology o f the g i l l s . Pp . 185-227 In: F i s h P h y s i o l o g y : V o l u m e X : G i l l s Part A : A n a t o m y , gas transfer, and acid-base regulat ion. H o a r W S , R a n d a l l D J eds. A c a d e m i c Press Inc., Toronto . O n i m a r u H , H o m m a I (2003) A nove l funct ional group for respiratory rhy thm generation i n the ventral medul la . J Neurosc ience 23(4): 1478-1486. Perry S F , G i l m o u r K M (1996) Consequences o f catecholamine release on venti lat ion and b l o o d oxygen transport dur ing h y p o x i a and hypercapnia i n an elasmobranch (Squalus acanthias) and a teleost (Oncorhynchus mykiss). J E x p B i o l 199: 2105-2118. Per ry S F , K i n k e a d R , Fr i tsche R (1992) A r e c i rcula t ing catecholamines i n v o l v e d i n the cont ro l o f breathing i n fishes? R e v F i s h B i o l 2:65-83. R a n d a l l D J , Jones D R (1973) T h e effect o f deafferentation o f the pseudobranch on the respiratory response to h y p o x i a and hyperox ia i n the trout (Salmo gairdneri). Resp i r P h y s i o l 17: 291-301. R e i d S G , M e i e r J T , M i l s o m W K (2000) T h e influence o f descending inputs on breathing pattern formation i n the isolated bu l l f rog brainstem-spinal cord . Resp i r P h y s i o l 120: 197-211. R e i d S G , M i l s o m W K (1998) Respira tory pattern format ion i n the isolated bul l f rog (Rana catesbeiana) brainstem-spinal cord . Resp i r P h y s i o l 114: 239-255. 120 R e i d S G , S u n d i n L , F l o r i n d o L H , R a n t i n F T , M i l s o m W K (2003) Effects o f afferent input on the breathing pattern con t inuum i n the tambaqui (Colossoma macropomum). Resp P h y s i o l N e u r o b i 136 (1): 39-53. Roberts B L , B a l l i n t i j n C M (1988) Sensory interactions w i t h central generators dur ing respiration i n the dogfish. J C o m p P h y s i o l A 162: 695-704. R o v a i n e n C M (1974) Respiratory motoneurons i n lampreys. J C o m p P h y s i o l 94: 57-68. R o v a i n e n C M (1983) A t r igeminal component o f the central pattern generator for respiration i n the adult lamprey. Soc N e u r o s c i A b s t r 9: 541 . R o v a i n e n C M (1985) Respiratory bursts at the mid l ine o f the rostral medu l l a o f the lamprey. J C o m p P h y s i o l A 157: 303-309. R u s s e l l D F (1984) Respiratory neurons near the t r igeminal nucleus i n lampreys. Soc N e u r o s c i A b s t r 10: 754. Satchel l G H (1959) Respiratory reflexes i n the dogfish. J E x p B i o l 36: 62-71 Shel ton G (1959) The respiratory centre i n the tench (Tinea tinea L . ) I. T h e effects o f bra in transection on respiration. J E x p B i o l 36: 191-202. Shel ton G (1961) The respiratory centre i n the tench (Tinea tinea L . ) II. Respi ra tory neuronal act ivi ty i n the medu l l a oblongata. J E x p B i o l 38: 79-92. Smi th J C , El lenberger H , B a l l a n y i K , Rich te r D W , F e l d m a n J L (1991) P re -Botz inger complex : a brainstem reg ion that m a y generate respiratory rhy thm i n mammals . Science W a s h D C 254: 726-729. Smi th F M , Jones D R (1978) L o c a l i z a t i o n o f receptors causing hypox ic bradycardia i n trout (Salmo gairdneri). C a n J Z o o l 56: 1260-1265. St.John W M , Glasser R L , K i n g R A (1972) R h y t h m i c respiration i n awake vagotomized cats w i t h chronic pneumotaxic area lesions. Resp i r P h y s i o l 15: 233-244. St .John W M (1998) Neurogenesis o f patterns o f automatic venti latory act ivi ty . P r o g N e u r o b i o l 5 6 ( 1 ) : 97-117. Straus C , V a s i l a k o s K , W i l s o n R J A , O s h i m a T , Zel ter M , Derenne J -Ph . , S i m i l o w s k i T , W h i t e l a w W A (2003) A phylogenet ic hypothesis for the o r ig in o f h iccough . B i o E s s a y s 25: 182-188. Straus C , W i l s o n R J A , Tezenas du M o n t c e l S, Remmers J E (2000) B a c l o f e n eliminates cluster l ung breathing o f the tadpole brainstem, i n vi t ro . N e u r o s c i L e t 292 (1): 13-16. Sund in L I , R e i d S G , K a l i n i n A L , R a n t i n F T , M i l s o m W K (1999) Card iovascu la r and respiratory 121 reflexes: the t ropical f ish , traira (Hopl ias malabaricus) 0 2 chemoresponses. Resp i r P h y s i o l 116: 181-199. Sund in L I , R e i d S G , R a n t i n F T , M i l s o m W K (2000) B r a n c h i a l receptors and cardiorespiratory reflexes i n a neotropical f ish, the tambaqui (Co los soma macropomum) . J E x p B i o l 203: 1225-1239. S u n d i n L , Turesson J , B u r l e s o n M (2003) Indentification o f central mechanisms v i ta l for breathing i n the channel catfish, Ictalurus punctatus. Resp i r P h y s i o l 138: 77-86. Takeda T (1990) Ven t i l a t ion , cardiac output and b l o o d respiratory parameters i n the carp, Cyprinus carpio, dur ing hyperoxia . Resp i r P h y s i o l 81: 227-240. T a n g P C (1967) B r a i n stem control o f respiratory depth and rate i n the cat. Resp i r P h y s i o l 3:349-366. T h o m p s o n K J (1985) Organiza t ion o f inputs to motoneurons dur ing f ict ive respirat ion i n the isolated lamprey brain. J C o m p P h y s i o l A 157: 291-302. Torgerson C S , G d o v i n M J , Remmers J E (2001) Sites o f respiratory rhythmogenesis dur ing development i n the tadpole. A m J P h y s i o l 280: R 9 1 3 - R 9 2 0 . van D a m L (1938) O n the u t i l i za t ion o f oxygen and regulat ion o f breathing i n some aquatic animals. Dissertat ion, Gron ingen . V a s i l a k o s K , W i l s o n R J A , N a o f u m i K , Remmers J E (2005) A n c i e n t g i l l and l ung oscil lators m a y generate the respiratory r h y t h m o f frogs and rats. J N e u r o b i o l 62: 369-385. V i t a l i s T Z , M i l s o m W K (1986) M e c h a n i c a l analysis o f spontaneous breathing i n the semi-aquatic turtle, Pseudemys scripta. J E x p B i o l 125: 157-171. v o n Baumgar ten R , Sa lmoi ragh i G C (1962) Respiratory neurones i n the goldf ish . A r c h i ta l B i o l 100:31-47. W a l d r o n I (1972) Spat ial organizat ion o f respiratory neurones i n the medu l l a o f tench and goldf ish . J E x p B i o l 57: 449-459. W e b b e r C L , Speck D F (1981) Exper imen ta l b io t per iodic breathing i n cats: effects o f changes i n PIo2 and P I C 0 2 . Resp i r P h y s i o l 46: 327-344. W i l s o n R J A , V a s i l a k o s K , Har r i s M B , Straus C , Remmers J E (2002) Ev idence that ventilatory rhythmogenesis i n the frog involves two dist inct neuronal oscil lators. J P h y s i o l 540 (2) 557-570. W o l d r i n g S, D i r k e n M N J (1951) U n i t act ivi ty i n the medu l l a oblongata o f fishes. J E x p B i o l 28(2): 218-220. W o o d C M , Jackson E B (1980) B l o o d acid-base regulat ion dur ing environmental hype rox ia i n the r a inbow trout (Salmo gairdneri). Resp i r P h y s i o l 42: 351-372. 122 X u F , Frazier D T (2000) M o d u l a t i o n o f respiratory motor output by cerebellar deep nuc le i i n the rat. J A p p l P h y s i o l 89: 996-1004. X u F , Z h o u T , G i b s o n T , Frazier D T (2001) Fas t ig ia l nucleus-mediated respiratory responses depend on the medul la ry gigantocel lular nucleus. J A p p l P h y s i o l 91 : 1713-1722. Y a m a n a k a Y , Sakamoto T , W a d a K , Naka juma Y (1993) Ac t iv i t i e s o f the intralaryngeal muscles dur ing e lectr ical ly induced voca l iza t ion i n decerebrate cats. N e u r o s c i Res 17: 77-81 . Z h a n g C , Y a n H , L i C , Z h e n g Y (2004) Poss ib le involvement o f the facia l nucleus i n regulat ion . o f respiration i n rats. N e u r o s c i Le t t 367: 283-288. 123 Appendix Experimental Stage Mean SEM (cmH20) (cmH20) Intact Normoxia 2.6 0.8 Intact Hypoxia 4.1 0.6 Intact Hyperoxia 1.3 0.4 Intact Final Normoxia 1.6 0.3 Decerebrate Normoxia 1.0 0.4 Decerebrate Hypoxia 1.7 0.8 Decerebrate Hyperoxia 0.6 0.1 Decerebrate Final Normoxia 0.7 0.1 Decerebrate/Spinalectomized Normoxia 0.9 0.3 Decerebrate/Spinalectomized Hypoxia 0.5 0.2 Decerebrate/Spinalectomized Hyperoxia 0.8 0.3 Decerebrate/Spinalectomized Final Normoxia 1.0 0.3 Table A.l: Breathing amplitude (cmH20) of intact, decerebrate, and decerebrate/spinalectomized trout in normoxia, hypoxia, and hyperoxia. P02 values were approximately 125 mmHg in normoxia, 15rnrnHg in hypoxia, and 500 mmHg in hyperoxia. Amplitude values were recorded from the pressure changes in one opercular cavity of each trout. Sample size of 7 trout. Experimental Stage Mean SEM (breaths/min) (breaths/min) Control Normoxia 50.6 4.0 Control Initial Hyperoxia 37.5 5.2 Control Prolonged Hyperoxia 31.6 4.3 Control Normoxia 49.8 3.8 Sham Normoxia 47.2 2.0 Sham Initial Hyperoxia 39.7 4.1 Sham Prolonged Hyperoxia 31.7 4.1 Sham Normoxia 46.0 2.6 Kainic Acid Normoxia 66.9 4.8 Kainic Acid Initial Hyperoxia 59.3 6.4 Kainic Acid Prolonged Hyperoxia 50.6 6.3 Kainic Acid Normoxia 54.6 4.6 Table A.2: Breathing frequency (breaths/min) of decerebrate/spinalectomized carp before and after injection of kainic acid into the mesencephalic tegmentum at both normal and low levels of respiratory drive. Po2 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency was measured for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum. Sample size of 14 carp. 124 Experimental Stage Mean SEM (cmH20) (cmH20) Control Normoxia 0.83 0.13 Control Initial Hyperoxia 0.56 0.09 Control Prolonged Hyperoxia 0.53 0.09 Control Normoxia 0.75 0.12 Sham Normoxia 0.75 0.12 Sham Initial Hyperoxia 0.50 0.07 Sham Prolonged Hyperoxia 0.47 0.07 Sham Normoxia 0.81 0.19 Kainic Acid Normoxia 0.85 0.12 Kainic Acid Initial Hyperoxia 0.50 0.10 Kainic Acid Prolonged Hyperoxia 0.39 0.07 Kainic Acid Normoxia 0.61 0.14 Table A.3: Breathing amplitude (cmH20) of decerebrate/spinalectomized carp before and after injection of kainic acid into the mesencephalic tegmentum at both normal and low levels of respiratory drive. Po2 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing amplitude was measured for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum. Sample size of 14 carp. Experimental Stage Mean SEM (breaths/min) (breaths/min) Control Normoxia 85.0 4.6 Control Initial Hyperoxia 73.3 7.9 Control Prolonged Hyperoxia 77.4 10.0 Control Normoxia 87.6 5.1 Sham Normoxia 85.7 5.6 Sham Initial Hyperoxia 68.7 5.8 Sham Prolonged Hyperoxia 75.0 7.3 Sham Normoxia 85.6 5.4 Kainic Acid Normoxia 99.0 6.4 Kainic Acid Initial Hyperoxia 89.5 7.1 Kainic Acid Prolonged Hyperoxia 100.6 10.3 Kainic Acid Normoxia 98.1 9.3 Table A.4: Breathing frequency (breaths/min) of decerebrate/spinalectomized trout before and after injection of kainic acid into the mesencephalic tegmentum at both normal and low levels of respiratory drive. Po2 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing frequency was measured for each trout in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum. Sample size of 9 trout. 125 Experimental Stage Mean SEM (cmH20) (cmH20) Control Normoxia 1.40 0.24 Control Initial Hyperoxia 0.74 0.14 Control Prolonged Hyperoxia 0.62 0.12 Control Normoxia 1.62 0.34 Sham Normoxia 1.33 0.27 Sham Initial Hyperoxia 0.79 0.20 Sham Prolonged Hyperoxia 0.59 0.15 Sham Normoxia 1.39 0.35 Kainic Acid Normoxia 1.57 0.24 Kainic Acid Initial Hyperoxia 0.94 0.15 Kainic Acid Prolonged Hyperoxia 0.65 0.13 Kainic Acid Normoxia 1.08 0.17 Table A. 5: Breathing amplitude (cmH20) of decerebrate/spinalectomized trout before and after injection of kainic acid into the mesencephalic tegmentum at both normal and low levels of respiratory drive. P02 values were approximately 125 mmHg in normoxia and 500 mmHg in hyperoxia. Breathing amplitude was measured for each carp in normoxia, the first five minutes of hyperoxia (initial hyperoxia), after 1.5 hours of hyperoxia (prolonged hyperoxia), and 30 minutes after being returned to normoxia. These values were measured initially (Control), after microinjections of ACSF (Sham), and after microinjections of kainic acid (Kainic Acid) into the mesencephalic tegmentum. Sample size of 9 trout. 126 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0091995/manifest

Comment

Related Items