REGULATION OF BLOOD FLOW THROUGH A THE ISOLATED- PERFUSED GILLS OF RAINBOW TROUT: EFFECTS OF VASOACTIVE AGENTS ON FUNCTIONAL SURFACE AREA MM for the Degree of PA. 0. MICHIGAN STATE UNIVERSITY. f HAROLD I... BERGMAN 1973 L I B R A R Y Michigan State University g amam'a av T‘? HUAS & SUNS' BOOK BINDERY INC. LIBRQRY amocns- srnlueroar. ulculen §. 4R“ A :3 ’ mg!um;1112::thWM W16 W 200 A 777 ABSTRACT REGULATION OF BLOOD FLOW THROUGH THE ISOLATED-PERFUSED GILLS OF RAINBOW TROUT: EFFECTS OF VASOACTIVE AGENTS ON FUNCTIONAL SURFACE AREA BY Harold L. Bergman Published studies on gill anatomy and effects of vaso— active agents on vascular resistance in teleost gills have prompted speculation about the physiological significance of different blood flow paths through this organ. In this study the influx of l4C-urea, a passively diffusing molecule, was used to indicate the relative functional respiratory surface area of isolated-perfused rainbow trout gills. Perfusion of vasoactive agents in these preparations significantly altered both l4C-urea influx and branchial vascular resistance. 9 5 An increase in norepinephrine perfusion from 10- to 10— M increased l4C-urea influx 4.6-fold, while an epinephrine increase from lo—7 to 10—5M caused a 5T6—fold increase in marker influx. Both catecholamines produced an overall decrease in branchial vascular resistance, but sometimes only after a transient increase. Either a or 8 adrenergic 1 blockade halved the catecholamine effect on 4C—urea influx while blocking the appropriate increase (a response) or Gas ) 1" i." ‘ I" K. 4", x I ‘ ‘ A 'I ,4 Harold L. Bergman decrease (8 response) in branchial vascular resistance. Perfusion of pharmacological d or B adrenergic stimulants produced effects which mimicked the B or a blocked catechol- amine results, respectively. Acetylcholine, when increased from 10”8 to 10—6M, decreased l4C-urea influx to 1/5 of its control value, while causing a marked increase in branchial vascular resistance. Data presented in this thesis supports the contention that gill functional respiratory surface area is controlled by both neural and hormonal mechanisms. REGULATION OF BLOOD FLOW THROUGH THE ISOLATED-PERFUSED GILLS OF RAINBOW TROUT: EFFECTS OF VASOACTIVE AGENTS ON FUNCTIONAL SURFACE AREA BY Harold L: Bergman A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Fisheries and Wildlife 1973 DEDICATION To Annette, Jill and Peter ACKNOWLEDGMENTS I thank Drs. P. O. Fromm and H. E. Johnson for guidance and support during this study and throughout my stay at Michigan State University. Thanks are also due the other members of my committee, Drs. T. G. Bahr and S. D. Aust, for their review of the manuscript and suggestions for its improvement. I extend special thanks to Dr. K. R. Olson for many valuable ideas and hours of stimulating discussion. I am indebted to Esther Brenke for her valuable technical assis- tance and for typing the rough drafts of this thesis, and to Dr. J. R. Hoffert and K. R. Olson for the photographic work. Dr. Frank Kutyna offered many helpful suggestions, and fellow graduate students in the Physiology and the Fisheries and Wildlife Departments gave constructive criticism. They are all to be thanked. I am grateful for financial support from the EPA Train— ing Grant No. T-90033l and for support from EPA Grant No. R-801034 to Dr. Fromm. TABLE OF CONTENTS LIST OF TABLES O O O O O 0 O 0 O O O 0 LIST OF FIGURES . . . . . . . . . . . INTRODUCTION. 0 O O O O O O O 0 O O 0 MATERIALS AND METHODS . . . . . . . . Experimental Animals . . . . . . Experimental Apparatus . . . . . Gill Dissection and Cannulation. Perfusion Solutions. . . . . . . Experiment Protocols . . . . . . l4C-Urea Influx Experiments Perfusion Pressure Experiments. Data Treatment and Statistics. . 4C-Urea Influx Data. . . . Perfusion Pressure Data . . RESULTS 9 0 O O O 0 O O O O 9 O H O 0 General Techniques . . . . . . . C—Urea InflUX. o o o o o o o o Perfusion Pressure . . . . . . . DISCUSSION. 0 0 0 O 0 O O O O O 0 6 0 Functional Surface Area. . . . . Perfusion Pathways . . . . . . . e 0 Hormonal and Neural Control Mechanisms CONCLUSIONS 0 O O O O O O O I O O O 0 LITERATURE CITED. . o . . . . . . . . APPENDIX. 0 O O 0 6 O O O O O 0 G 0 6 iv Page vi 10 10 10 18 25 27 27 29 30 30 35 36 36 37 51 67 67 7O 73 78 79 82 LIST OF TABLES TABLE Page 1. Vasoactive drugs and hormones used in l4C—urea g influx and perfusion pressure experiments. . . . 26 g; 2. Protocols for l4C—urea influx experiments. . . . 28 ‘2? 2w y. 2.1 .L‘“ ' 4 "If" t‘ LIST OF FIGURES FIGURE Page 1. Diagram of the gill perfusion apparatus. . . . . 12 2. Photograph of the gill perfusion apparatus . . . l4 3. Gill arch dissection steps . . . . . . . . . . . 21 4. Results from a typical epinephrine experiment. . 33 5. The effect of the a adrenergic agonist, phenyl— ephrine, on l4C-urea influx. . a . . . . . . . . 39 6. The effect of the 8 adrenergic agonist, isopro— terenol, on 14C- -urea influx. . . . . . . . . . 41 7. The effect of norepinephrine on l4C-urea influx. 43 8. The effect of adrenergic blockade on norepine- phrine induced C-urea influx . . . . . . . . . 46 14 9. The effect of epinephrine on C-urea influx . . 48 10. The effect of adrenergic blockade on epinephrine induced l4C- -urea influx. . a . . . . . . . . . . 50 11. The effect of acetylcholine on l4C-urea influx . 53 12. The effect of acetylcholine on l4C—urea influx in the presence of epinephrine . . a . . . . . . 55 13. Typical effects of norepinephrine on perfusion pressure in the absence and presence of adrener- gic blockers . . . . . . . . . . . . . . . . . . 58. 14. Typical effects of epinephrine on perfusion pres— sure in the absence and presence of adrenergic blockers . . . . . . . . . . . . . . . . . . . . 60 15. Typical effects of adrenergic agonists on perfu- sion pressure. . . . . . . . . . . . . . . . . . 63 vi LIST OF FIGURES--Continued Page l6. 17. Typical effect of acetylcholine on perfusion pressure. . . . . . . . . . . . . . . . . . . . 63 The effect of various drugs on acetylcholine induced perfusion pressure changes. . . . . . . 65 vii INTRODUCTION It is widely speculated that teleosts can regulate func- tional respiratory surface area by adjusting blood flow pat- tern through the gills. But the existence of some described blood pathways has been disputed, and the physiological mechanisms which regulate blood flow through different path- ways remain unclear. Evidence suggesting functional regulation of blood path- ways includes the 5 to 10-fold increase in oxygen uptake reported for several species of fish during exercise (Saunders, 1962; Brett, 1964; Stevens and Randall, 1967a), and the nega- tive Na+ balance which accompanies increased oxygen uptake in freshwater rainbow trout (Wood and Randall, 1973). The significance of such a regulatory mechanism is that fish could I) maximize gas exchange during periods of increased activity by perfusing most or all of the respiratory blood pathways, and 2) restrict the undesirable effects of salt and water movements by limiting blood flow to fewer respiratory paths during periods of rest. The general anatomical features of gills from several teleost species have been described by Hughes and Grimstone (1965), Newstead (1967), and Morgan and Tovell (1973). The anatomical surface area of each gill arch is greatly increased by its subdivision into two successively smaller units, the filaments and secondary lamellae. The somewhat flattened filaments extend out from the gill arch in two rows. These rows, or hemibranchs, can be extended out to intercept the water which flows through the gill slits on each side of the arch. Numerous platelike secondary lamellae are attached by their edges and extend out from the top and bottom faces of each filament. The secondary (respiratory) lamellae are the functional respiratory units. Basically, each lamella consists of two epithelial sheets which are held together by pillar cells. The lacunar spaces between the pillar cells are large enough to allow passage of red blood cells. Histological studies have revealed similar circulatory anatomy in the common eel (Steen and Kruysse, 1964), coho salmon (Newstead, 1967) and rainbow trout (Richards and Fromm, 1969). In these species blood flow from the afferent filamental artery to the efferent filamental artery was reported to be through at least two pathways: 1) the flat lacunar secondary lamellae (respiratory pathway), and 2) a central sinus in the filament body (non—respiratory shunt pathway). Steen and Kruysse also observed an additional non- respiratory pathway in the eel, consisting of a direct con- nection between the afferent and efferent filamental arteries at each filament tip. Respiratory blood flow through the secondary lamellae is thought to be further subdivided be- tween a preferential path around the free margin of each lamella and the lacunar paths amongst the pillar cells (Hughes and Grimstone, 1964; Newstead, 1967; Skidmore and Tovell, 1972). In a recent report (Morgan and Tovell, 1973) the non- respiratory filamental sinus pathway in the rainbow trout was not confirmed. Filamental sinuses were observed but were assigned a lymphatic function only, since no red blood cells and no openings from sinuses to filamental arteries were seen. As an alternative to the filamental sinus shunt pathway, the authors supported a recruitment mechanism proposed earlier by Hughes (1972). This proposal suggests that during periods of low oxygen demand respiratory blood flow is directed to the secondary lamellae on the proximal end of the filaments only. During periods of heightened activity when gas exchange demands are increased, additional lamellae toward the distal filament tip are successively recruited to increase respira- tory blood flow. Whether respiratory blood flow is modulated by use of filamental shunts or lamellar recruitment, some mechanism must function to adjust vascular resistances in the appropriate pathways. Hence, a number of workers have sought histological evidence for the presence of vascular smooth muscle in gills. .l,,”$"zlf‘ter‘ "In”; ‘ (um 14"; Morgan and Tovell (1973) reported the presence of a continuous layer of muscle surrounding the endothelial layer of afferent and efferent lamellar arterioles, corroborating observations by Richards and Fromm (1969). Neither study revealed muscular elements associated with the filamental sinus. In an earlier study, Newstead (1967) found no muscle tissue in the region of afferent or efferent lamellar arterioles, but did report what appeared to be muscle filaments in the pillar cells. He, therefore, attributed a contractile function to the pillar cells, Supporting a view held by Hughes and Grimstone (1965). This hypothesis was further strengthened by the demonstration that these pillar cell fibers were actomyosin—like proteins (Bettex-Galland and Hughes, 1972), and that the pillar cells were innervated (Gilloteaux, 1969). Although physiological regulation of gill blood pathways is unclear, it is generally thought to involve hormonal and/or neural inhibition and stimulation of vascular smooth muscle. The approach taken by fish physiologists to elucidate these regulatory mechanisms has, of course, been influenced by knowledge about similar systems in higher vertebrates. Vascular resistance in mammals is controlled by three main chemical transmitters: epinephrine (adrenaline), norepine- phrine (noradrenaline), and acetylcholine. Epinephrine and norepinephrine, catecholamines released from the adrenal medulla and peripheral chromaffin cells, increase resistance in some vascular beds and decrease resistance in others. Ahlquist (1948) postulated that the opposite responses resulted from presence of two different receptors for cate- cholamines. He suggested that a-adrenergic receptors mediated excitory responses (vasoconstriction), while B-adrenergic receptors mediated inhibitory responses (vasodilation). The user-W = w neurohumoral transmitters found in higher vertebrates include acetylcholine and, again, norepinephrine. Norepinephrine it“, w, .- 7.0,;13-‘2‘ is the transmitter released at postganglionic sympathetic nerve endings, while acetylcholine is found at all pregan- glionic autonomic nerve endings and all postganglionic parasympathetic nerve endings. Acetylcholine is also the transmitter at motor nerve synapses with skeletal muscle. Two distinct actions have been attributed to acetylcholine (Dale, 1914) and have also been explained by two different receptor types. The receptors at postganglionic parasympa- thetic nerve ends could be blocked by muscarine or atropine and were termed muscarinic receptors, while receptors at autonomic ganglia and skeletal neuromuscular junctions could be blocked by nicotine and were called nicotinic receptors. Physiologists have assumed that regulation of vascular resistance is similar in fish and higher vertebrates, and they have tested the effects of cholinergic and adrenergic drugs and hormones (as well as many other chemicals) on branchial resistance. Some studies have demonstrated changes in branchial resistance by monitoring dorsal and ventral aortic blood pressures of intact fish during injection of various vasoactive agents (Chester Jones, et 31., 1967; Reite, 1969). In studies where only dorsal or ventral aortic blood pressure changes were reported (Mott, 1951; Randall and Stevens, 1967), interpretations about branchial vascular resistance are difficult. Undoubtedly, the injected vasoactive agents affected cardiac output and systemic vascular resistance as well as vascular resistance in the gill, and the pressure responses would reflect a combination of all these effects. Results from experiments with isolated-perfused gills are much more easily interpreted. Vasoactive agent effects on branchial resistance have been detected by observing changes in perfusion pressure (Reite, 1969) or flow rate (Keys and Bateman, 1932, Ustlund and Fange, 1962; Rankin and Maetz, 1971; Randall et 31., 1972) in these isolated preparations. The above studies revealed that branchial vascular resistance was reduced by catecholamines and increased by acetylcholine. Using two different methods, Steen and Kruysse (1964) and Richards and Fromm (1969) observed that acetylcholine in- creased filamental sinus blood flow and epinephrine increased secondary lamellar blood flow. Steen and Kruysse also reported increased oxygen uptake after injection of epiner phrine in Vivo. ' '«1',‘:,,":-'P'go){{f|.i’;:u ,‘ The effects of epinephrine and norepinephrine, coupled with a report that circulating concentrations of these catecholamines are elevated during exercise (Nakano and Tomlinson, 1967), support the hypothesis that blood flow in the gills is at least partly under hormonal control. The reports that norepinephrine and acetylcholine affect branchial vascular resistance also suggest the possibility of autonomic nervous control of blood flow pattern through the gill. The existence of sympathetic and parasympathetic nerve trunks to the gill (Nicol, 1952), and histological demonstration of nerve endings on the pillar cells and afferent and efferent arteries (Gilloteaux, 1969) add impetus to the neural control argument. There have been no reports, however, showing a direct effect of autonomic nerve stimulation on the resistance of any vascular bed in fish (Campbell, 1970). Existence of a mechanism for adjusting blood flow pattern to regulate functional surface area of the gill has been sup- ported by the anatomical and physiological evidence given above. The anatomical evidence is somewhat tenuous, however, because the proposed filamental sinus pathway has not been confirmed in recent observations, and the recruitment mechanism is still quite speculative with little evidence to support it. The physiological evidence taken as a whole offers the strong- est support for the functional surface area hypothesis. However, alternate explanations can be offered for each physio- logical observation when taken alone. Increased oxygen uptake during exercise or following epinephrine injection might be accounted for by increased cardiac output and ventilation volume; negative Na+ balance during exercise could result from altered kidney function; vasoactive agent—induced changes in branchial vascular resistance reflect only the sum of changes in flow resistances through the gill and provide no informa- tion about resistance changes in the pathways that have been proposed. Therefore, no single piece of physiological evidence has been strong enough by itself to confirm the hypothesis. A critical series of experiments is needed to provide strong evidence that either confirms or denys the functional surface area hypothesis. Such evidence could be obtained by using isolated—perfused gills to simultaneously measure the effects of vasoactive agents on branchial vascular resistance and functional surface area. Aside from the basic research interests in this problem, there is a growing need to understand gill structure and func- tion for a more practical reason. The Vital functions of fish gills are known to be adversely affected by various aquatic pollutants. We cannot hope to understand or evaluate these abnormal physiological conditions, if we incompletely compre- hend gill function in healthy fish. The objectives of the present study were to 1) perfect an isolated-perfused gill technique which could be used for 3 to 4 hour-long experiments under conditions resembling, as closely as possible, those found in yiyg, 2) confirm or deny regulation of functional surface area of the teleost gill, and 3) determine the nature of physiological control mechan- isms which could be responsible for regulating functional surface area. In this study the influx of l4C—urea was used as a rela- tive measure of gill functional surface area. Urea is not metabolized in rainbow trout gill tissue (K. R. Olson, personal communication) and is not known to be actively transported by teleost gills. The method is based on the assumption that diffusional influx of l4C-urea is limited principally by the extent of secondary lamellae perfusion-— i.e., the functional surface area of the gill available for diffusional influx of the markero 2‘2}? . umwmpu\ TM ,1, MATERIALS AND METHODS Experimental Animals Rainbow trout (Sglmg gairdneri), 200 to 300 grams, were obtained from the Michigan Department of Natural Resources hatchery in Grayling, Michigan. At Michigan State University, the fish were held in flowing dechlorinated tap water at 10-12°C under a controlled photoperiod of 16 hours light per day. Animals were fed a maintenance diet of EWOS 159 salmon pellets (Astra—Ewos, SBdertalje, Sweden), but were starved one week prior to use. All experiments were conducted between January and December, 1973. Experimental Apparatus Two gill arches could be perfused simultaneously in the apparatus shown in Figures 1 and 2. Perfusion solutions were delivered from polyethylene bottles fitted with glass tubes glued into the bottle bases. Silicone rubber tubing (1.5 mm i.d., 3.5 mm o.d.) coupled the bottles to 4-way stopcocks which allowed perfusing solutions to be conveniently switched during an experiment without introducing air bubbles into the system. A multichannel peristaltic pump (Brinkman Instru- ments, Inc., Westbury, N.Y.) pumped solutions through the gills. lO ll .mnmospmcmnu OHDwmmHm ARV “mxooomoum mmslv Amy “Hmccmco :¢= ca mmoup .Hmccmco :4: CH OHSmmmum coflmsmumm .mcHSD Howomaaoo cofluomum .Hmccmco =m= CH mndmmmnm conSmnmm "Eouuon 0p mo» Eonm mcflpmmn Uuoowu cmmummaom mmmnw Amv “mcoHDSHOm mchSMHOm Ammv “maze oflpHopmflumm Ammv “Hmpwfiocmfi COHDMHQHHMO whammmsm Aumv “cmmnmwaom mmmnw va “HODMHSESOOM ucsoo most Honomaaoo coflpomnm Aoomv “HODOOHHOO cofluomum Aumv “mumpcsoo gone UHHDUOHOODOQm ADV “smegmao scamsmumm =m= ca noun Haaw Ame “Hmccmco coflmswnmm =m= cfl some HHHO Amv .mcofluomccoo Decompomam Dcmmmummn mmcfla pmcmmp “mQOHDomccoo paw mamccmco scamsmnmm ucmmmummn mmcfla tflaom .mSDMHmmmm coflmsmumm HHHm mcp mo EMHmMHQII.H musmflm H mnsmfim 12 13 .msumsmmmm counmsmfinmm HHHm 9.3 m0 cmmumoeongIJ stmHm l4 Figure 2 15 Pump output could be regulated over a wide range, but in practice only two pump settings were used: either 0.25 or 0.50 ml per minute. The lower flow rate was used during cannulation, but before an experiment began the flow was increased to the faster rate. Since the same size pump tubing was used for both perfusion channels, the flow rate was identical, or very nearly so, to each gill arch. Of the several types of peristaltic pump tubing which were tried, Tygon (0.8 mm i.d., 2.4 mm o.d.) gave the most consistent flow and wear characteristics. To maintain calibrated flow rates, however, the pump tubing was replaced after every third experiment. From the pump to each gill arch, perfusion solutions flowed through 10 cm silicone rubber tubing (1.0 mm i.d., 3.0 mm o.d.), a "T" connector, 10 cm polyethylene tubing (PE 60), and an afferent cannula made from a dulled, cut-off 20 gauge needle. The silicone rubber tubing between the pump and gill arch dampened the pulsatile flow in a manner analogous to the "Windkessel effect" of the conus arteriosus and ventral aorta. Efferent flow was through a 20 gauge cannula, 32 cm PE 60 tubing, a drop counter assembly, and into sample vials in an Isco model 563 fraction collector (Instrumentation Special- ties Co., Lincoln, Neb.). Different drop counter assemblies were used for the two perfusion channels. In the "A" channel, 16 a Grass model DCA-l drOp counter adaptor (Grass Instrument Co., Quincy, Mass.) was modified by gluing a 2 cm square acrylic plastic block to the top of the adaptor. Holes had been drilled to the center of the block from two adjacent faces so that flow entered a side face of the block and left through the bottom hole which communicated with the hole in the Grass drop counter adaptor. Tubes made from syringe needles were glued into and protruded from the holes in the side (16 gauge tube) and bottom (20 gauge tube) faces of the plastic block. The efferent polyethylene tubing could be firmly inserted into the side face tube and the bottom face tube facilitated delivery of uniformly sized drops. The drops then fell through the light beam of a Grass PTTl photoelectric transducer. After stepwise amplification by a Grass 5P1K preamp and a Grass 5E driver amp, the transducer signal drove the recorder pen on one channel of a Grass model 5D polygraph. Each pen deflection represented one drop. In the "B" perfusion channel, the drop counter assembly consisted of another drilled plastic block which was clamped in place above an Isco model 600 photoelectric drop counter. The drop counter signal then drove the fraction collector, which was set to turn the tube reel to the next sample after 64 drOps had been collected. A signal from the fraction collector control unit was interfaced through a relay switch (120 volt AC, series 200 relay) to the signal marker input of 17 the Grass polygraph. Thus, each turn of the fraction collec- tor, representing collection of 64 drops from the "B" perfusion channel, deflected a pen in the signal marker channel of the Grass polygraph. During all experiments, the gill arches were immersed in baths which consisted of 350 m1 of 1% non-nutrient Ringer solution (Appendix) in rectangular glass staining dishes. For l4C-urea influx experiments, enough l4C-urea (Inter- national Chemical and Nuclear Corp., Irving, Ca1if., specific activity equalled 56.3 mc/mM) was added to each bath to obtain about 104 dpm/ml. This represented a urea concentration of less than 1.0 uM/l. The baths were continuously stirred with Teflon coated magnetic stirring bars. A sheet of polyurethane foam, 2 cm thick, insulated the baths from heat produced by the magnetic stirring motors. The gill arches were suspended in the baths with gill holders which consisted of a ring stand and movable metal arms terminated with alligator clips° These clips were clamped onto the metal cannulae which protroded from each side of the gill arches. When the gill arches were properly placed in the baths, the vertical distance from the efferent cannula to the drop counter assembly was 20 cm. The pressure on the efferent side of the gill, therefore, was equivalent to 15 mm Hg, l8 and mimicked the systemic resistance normally present in the intact animal. To monitor perfusion pressures, the "T" connectors between the pump and gill arches were connected with PE 60 tubing to Statham P23AC pressure transducers (Statham Trans— ducers, Inc., Hato Rey, Puerto Rico). Since the perfusion pressures were measured between pump and gill, they were analogous to measurement of ventral aortic pressures in an intact fish. A pressure calibration manometer was also con— nected to the transducers with PE 60 tubing. The gill arches, manometer base, and transducers were all in the same hori- zontal plane to facilitate accurate calibration and measure— ment of pressures. After stepwise amplification by a Grass 5P1K preamp and a Grass 5E driver amp, each transducer signal drove a recorder pen in each of two channels of the Grass 5D polygraph. The entire experimental apparatus was assembled in a cold room which was maintained at 11:2°C. Although the air temperature fluctuated through this 9 to 13°C cycle every 20 to 30 minutes, the perfusion and bath solution temperatures did not vary appreciably. Gill Dissection and Cannulation Fish were netted, stunned with a sharp blow to the cranium, and immediately decapitated just posterior to the 19 opercula. The head was placed in a beaker of non-nutrient Ringer solution (Appendix), which contained 2 USP units sodium heparin/ml (Organon, Inc., W. Orange, N.J.). The heart, which continued to beat, pumped this solution through the gills clearing them of blood. After about 15 minutes the head was removed from the beaker, and the ventricle of the heart was quickly cut by entering the pericardial sac from its exposed caudal end. This prevented air emboli from being pumped into the gills during the dissection procedure. The pectoral fins were cut off at this time, so that they would not interfere with later steps in the dissection. With the head held by Kelly forceps clamped to the upper jaw, an operculum was pulled out, one blade of heavy scissors was inserted anteriorly until the tip protruded from the mouth, and the lower jaw was cut (Figure 3A). The operculum was removed by cutting along a line from its dorsal connection to a point just dorsal to the eye and on through to the tip of the upper jaw (Figure 3B). Removal of the remaining operculum was similarly started by cutting the lower jaw, but before the dorsal opercular cut was made the anterodorsal part of the head was cut away just anterior to the origin of the first pair of gill arches in the roof of the buccal cavity (Figure 3C). The remaining operculum was then trimmed off along its dorsal connection. 20 Figure 3.-—Gill arch dissection steps. See text for discussion. Figure 3 22 Grasping the exposed anterior and posterior ends of the spinal column between the thumb and forefinger, the cranium was trimmed away from the gill basket with fine tipped scissors. One tip of the scissors was inserted between the cranium and the dorsal origin of the first pair of gill arches and the connecting tissue trimmed back for about one centimeter. The gill basket then tended to hang away from the cranium, stretching the posterior buccal cavity epithelium (Figurelfln. The epithelial tissue was then cut as far as possible on both sides exposing more of the tissue between the base of the cranium and the esophagus. The incision was deepened by continuing to alternately cut esophageal-cranial connective tissue and buccal epithelium until the anterior end of the head kidney was exposed. The cranium was completely removed by cutting through the head kidney and the body wall on each side. With the tongue held between the thumb and forefinger, the tips of 15 cm scissors were inserted on either side of the basibranchial bone between the ventral origins of the first and second gill arches (Figure 3E). By pushing the tongue down toward the tips of the scissors the basibranchial bone was cut with no damage to the filaments of the second gill arch. The gill basket was then set down so that the cut eSOphagus and pectoral fin bases rested on the table and the ventral aSpect of the animal faced forward. The first pair 23 of gill arches were cut away one arch at a time by inserting the scissors between the dorsal origins of the first and second arches on one side and cutting diagonal to the midline of the animal (Figure 3F). This cut was more easily made if the tongue was pulled forward so that the first arch was straightened out and pulled against the scissors. After turning the gill basket 180°, the basibranchial bone between the second and third pair of arches was cut in the same way as the earlier cut between the first and second pair (Figure 3G). To facilitate insertion of the scissor tips through the gill slits for this cut, a forefinger was placed behind the point where the cut was to be made and used to help push the gills onto the scissors. By pushing forward on the Ventral side of the animal and pushing down on the scissor handles the second pair of arches bent upward lifting the filaments of the second arch away from the cutting edge of the scissors. After making the cut, the gill basket was again rotated 180°, and the cut ventral end of the second arches was held away from the third arches with teethed forceps. Damage to the arch could be avoided by biting the forceps into the two exposed ends of the cut basibranchial bone rather than the soft tissue around it. With the ventral end of the arches held up and away from the third arches, the scissors were inserted to cut away the dorsal ends. These two cuts were 24 made in the same way as the earlier cuts to remove the first pair of arches. With the second pair of gill arches now dissected com- pletely free, fine tipped 10 cm scissors were used to separate the two arches (Figure 3H). The arches were held up with forceps so that the sharp blade of the scissors could be in- serted into the caudal end of the cut ventral aorta. The Ventral side of the Vessel was then laid open exposing the openings to the left and right afferent branchial arteries. Taking care that one afferent branchial artery could be seen on each side of the scissors blade, the dorsal side of the ventral aorta and the basibranchial bone were cut longitudinal- 1y. This cut separated the two gill arches and they were ready for cannulation. For cannulation, a gill arch was placed in a watch glass and covered with non-nutrient Ringer solution to prevent drying. With the pump delivering about 0.25 ml perfusion solution per minute, the afferent cannula was inserted down the afferent branchial artery. A single knot was tied around the arch with number 3 cotton suture to hold the cannula in place. Throughout these procedures the pressure recording for the arch was closely watched as an indicator of flow blockage. If cannulation was successful, initial pressure rarely exceeded 70 mm Hg and usually fell to below 50 mm Hg within several minutes. In a good preparation the pressure 25 did not increase when the cannula was knotted in place. To insert the efferent cannula it was usually necessary to use 7 to 10X magnification with a stereoscopic dissecting scope. Black pigmentation spots on the efferent branchial artery facilitated its location, once the surrounding muscle had been carefully teased apart. Holding one edge of the cut end of the artery with number 3 Dumont tweezers, the cannula was inserted. After confirming adequate flow in the efferent tube, the cannula was tied in place in the same way as the afferent cannula. Both cannulae were withdrawn to within several mm of the knots, the arch was suspended in the bath and the cannulae clamped in place with the gill holders. After connecting the efferent tube to its drop counter assembly, the arch was ready for use. Perfusion Solutions Perfusion solutions consisted of vasoactive drugs and/or hormones (Table 1), added to a glucose-Ringer solution (Appendix). The Ringer solution was prepared fresh daily from crystalline glucose and stock solutions of the inorganic constituents. Before vasoactive agents were added, the Ringer solution was vacuum filtered through a 0.22 pm Millipore filter. The solution was then vigorously shaken to assure atmospheric equilibration. 26 Table l.--Vasoactive drugs and hormones used in l4C—urea influx and perfusion pressure experiments. Drug generic and Text Manufacturer trade name abbreviation or source . l . . Atropine 504 --—- Sigma Chemical Co. St. Louis, Mo. Acetylcholine Cl ACH Sigma Chemical Co. St. Louis, Mo. Epinephrine HCl EPI Wolins Farmingdale, N.Y. Hexamethonium C1 ~--- Dr. Frank Kutyna MSU, E. Lansing, Mich. Isoproterenol HCl IPT Winthrop Laboratories Isuprel HCl New York, N.Y. Norepinephrine bitartrate NEPI Winthrop Laboratories Levophed bitartrate New York, N.Y. Phenoxybenzamine HCl POB Smith, Kline & French Labs. Dibenzyline HCl Philadelphia, Pa. . l . Phentolamine -e-- CIBA Pharmaceutical Co. Regitine Summit, N.J. Phenylephrine HCl PEP Winthrop Laboratories Neo-synephrine New York, N.Y. Propranolol HCl PROP Ayerst Laboratories, Inc. Inderal HCl 1 Reserpine Serpasil New York, N.Y. CIBA Pharmaceutical Co. Summit, N.J. l . . . Used only in perquion pressure experiments. 27 Experiment Protocols Usually, two gill arches were perfused simultaneously with the drug or hormone concentration being varied in the experimental arch and either not added or maintained at a constant level in the control arch. l4 . . C-Urea Influx Experiments: These experiments were. designed to measure the effect of vasoactive drugs and hor- mones on perfusion pressure and influx of l4C-urea. Changes in these parameters are taken to reflect 1) changes in the degree of respiratory lamellae perfusion, and 2) concomitant changes in functional surface area of the gill available for diffusional influx of l4C-urea. The experiment protocols are shown in Table 2. After cannulation, gill arches were perfused for an equilibration period of about one hour before experiments were begun. During this time, arches were perfused with the same solutions that were to be used in the initial period of the experiment. Except when adrenergic blocking drugs were used, the same treatments were applied to both control and experimental arches through the initial period. Starting with the initial period, twenty fractions were collected during each experimental period. In the early studies, all even numbered fractions were collected and counted for l4C-urea activity. Later it was evident that fewer samples would be adequate, and thereafter every fifth fraction was collected Table 2.--Protocols for 14 28 C—urea influx experiments. Experiment Arch1 Dru concentrations2 during different periods3 EquIIIbratIon & Initial Second Third Fourth Fifth a Agonist C None None None None --- )3 None 10'5M PEP 10—4M PEP 10'3M PEP —-- B Agonist C None None None None —-— E None 10'7M IPT 10'6M IPT 10'5M IPT --- Epinephrine c 10'7M Epi 10'7M Epi 10'7M Epi 10'7M Epi --- E 10'7M Epi 10'6M Epi 10'5M Epi 10'7M Epi --— Epinephrine c io'7M Epi 10'6M Epi 10'5M Epi ——— -—— a Block E 10‘7M Epi 10'6M Epi 10‘ M Epi --- --- + PCB + P03 + POB . . -7 -6 . -5 . Epinephrine C 10 M Epi 10 M Epi 10 M Epi --- --— a Block 10'7M Epi 10’ M Epi 10'5M Epi —-- -—- + Prop + Prop Pr Epinephrine C 10‘7M Epi lO-GM Epi 10 5M Epi --— --- Double Block4 E 10'7M Epi 10_6M Epi 10 5M Epi + P05 + POB + POB --- --- + Prop + Prop + Prop . . -9 . -9 . -9 —9 -9 Norepinephrine C 10 M Nepi 10 M Nepi 10 M Nepi 10 M Nepi 10 M Nepi a 10'9M Nepi io'BM Nepi 10'7M Nepi lO-GM Nepi 10'5M Nepi Norepinephrine C lO-gM Nepi lO-BM Nepi lO-7M Nepi 10_6M Nepi lO—SM Nepi a Block E 10'9M Nepi 10-8M Nepi 10'7M Nepi 10_6M Nepi 10'5M Nepi + PCB + POB + P08 + P08 + POE Norepinephrine C 10—9M Nepi lO-BM Nepi 10-7M Nepi IO-GM Nepi lO-SM Nepi 8 Block E 10'9M Nepi 10'8M Nepi 10'7M Nepi 10‘6M Nepi 10’5M Nepi + Prop + Prop + Prop + Prop + Prop Norepinephrine 10-9M Nepi lO-BM Nepi 10_7M Nepi 10_6M Nepi lO-SM Nepi Double Block E 10'9M Nepi 10’8M Nepi 10‘7M Nepi 10'6M Nepi io'SM Nepi + P08 + POB + POB + POB + POB + Prop + Prop + Prop + Prop + Prop Acetylcholine l C None None None None --- E None 10_8M Ach 10'7M Ach 10'6M Ach --- Acetylcholine 2 c 10'5M Epi 10'5M Epi lO-SM Epi io'SM Epi -—- E 10' M Epi 10'5M Epi 10' M Epi 10_5M Epi --- -8+ -7+ -6 No Ach 10 M Ach 10 M Ach 10 M Ach 1C = control arch, E = experimental arch 2Full names and sources of drugs shown in Table 1. 3Time of experimental periods: Equilibration = 1 hour, other periods = about 40 min. 4Blocking drug concentrations maintained ag constant level for entire experiment. a Blocker = Phenoxybenzamine (POB) at 10' M B Blocker = Propranolol (Prop) at 10‘5M. Double Block = a + 8 block 29 for counting. At the end of a few experiments under each protocol, the vasoactive substance perfused in the experi- mental arch was restored to the control concentration. This was done to verify return of perfusion pressure and l4C-urea uptake to control values. Perfusate was collected directly into vials containing liquid scintillation cocktail (Appendix). During each experi- ment, 3 to 5 successive 100 pl samples of the l4C-urea baths were also taken for counting. All perfusate and bath samples were counted within 48 hours on a Mark 1 liquid scintillation counter (Nuclear-Chicago Corp., Des Plaines, Ill.). Using variably quenched l4C standards prepared by Nuclear—Chicago Corp., the channels—ratio method (Bush, 1963) was used to convert counts per minute (cpm) for each sample to disintegra- tions per minute (dpm). Average background was determined by counting periodic vials containing only scintillation cock- tail and was subtracted from all sample counts. Perfusion Pressure Experiments: Although most of the perfusion pressure data were collected in the l4C-urea influx experiments described above, other experiments were conducted in which only pressure responses were measured, thus, no l4C—urea was added to the baths and no perfusate fractions were collected. These experiments were conducted to determine the effect of several blocking drugs on vasoactive hormone- induced perfusion pressure changes seen in the studies 30 described above. Usually, the blocking drug was perfused in the experimental arch and the effect of a vasoactive hormone on perfusion pressure was tested in both control and experi— mental arches. In some cases the blocking agent was removed from the perfusion fluid to determine if control responses were restored. Data Treatment and Statistics 14 . . . C-Urea Influx Data: The l4C—urea influx in any experi— ment was determined not only by treatment effects but also by such variable factors as the size of the gill arch, the per- centage of gill arch perfused due to length of cannulae insertion, the extent of blood clotting in the respiratory lamellae, and bath l4C-urea activity. To separately test the effect of vasoactive substances on l4C-urea influx it was necessary to eliminate sources of variability not due to treat- ment effects. The l4C—urea count data from individual gill arches were, therefore, converted to percentages of the control l4C-urea counts from the initial experimental period for that arch. The l4C-urea sample activities from the initial experi- mental period were averaged, and the average was defined as 100% of initial activity. The activities of individual samples collected in later experimental periods were then converted to a percentage of this mean initial activity: 31 A %I= _—Sx100 AI % I = percent of initial activity = sample activity in dpm A = mean initial activity in dpm Because high l4C-urea activity in the baths constituted an essentially infinite source of l4C—urea for diffusion into the gill, it was not necessary to correct for the l4C-urea removed from the bath with the perfusion fluid. Occasionally, however, a gill arch leaked enough perfusion fluid into its bath during an experiment to significantly dilute the bath l4C—urea. In severe cases the experiment was abandoned, but where possible a correction factor was calculated from the reduction in bath sample counts. This was done by plotting the fraction of bath activity remaining versus time. The correction factor for any sample was then the fraction of bath activity remaining at the time the perfusate sample was collected. These factors were divided into the perfusate sample % I values to account for progressive reduction of bath l4C-urea available for diffusion into the gill arch. For each experiment the percent of initial activity values were plotted for both the experimental and control arches. The plots from a typical experiment are shown in Figure 4. Typically, the percent of initial activity increased 32 .meHpoon some Hopsmfiflnmmxm How mwfl>fluom HoHpHCH mo pcmouom .m.x “mcoflp loony some Honucoo cH mpfi>flpoo HMHDHQH mo scooped .moaonflo ammo .wmnoum HHHm Avon now mufl>flpoo deepens Homecoo one mucwmmumon mafia HMDGONHHOS UHHOm was .psonmsoncp Shaoa mos comm Houpcoo one CH coeumupcmocoo ocflncmmcflmm .m>oem csocm AN noumv some Hopcmfiflummxm CH macepmnpcoocoo ocflucmmcflmm Spas mpoflnmm Houcoaflnmmxm mumocflamp mmcfla pocmmp HMOflpHm> .pcoeflnmmxo ocflusmmCHmm amoemwu o Scum mpasmomll.v mnsmfim 33 v ousmflm 35.: ME... zo_._.<._:zz