THE BULLETIN EDITORIAL COMMITTEE

Editor Dr. David S. Miller Editorial Assistant Michael P. McKernan

Dr. David S. Miller, Chair

Dr. J.B. Claiborne

Dr. John H. Henson

Dr. Harmut Hentschel

Dr. Karl Karnaky

Dr. Eva Kinne-Saffran

Dr. Robert L. Preston

Published by the Mount Desert Island Biological Laboratory July 2002 $10.00 THE BULLETIN VOLUME 41, 2002

Mount Desert Island Biological Laboratory Salisbury Cove, Maine 04672

TABLE OF CONTENTS

Introduction ii Adrian Hogben Memorial v David Opdyke Memorial vii David Peakall Memorial ix Report Titles x Reports 1-109 Officers and Trustees 111 Scientific Personnel 114 Summer Fellowship Recipients 121 Seminars, Workshops, Conferences, Courses 123 Publications 130 Author Index 132 Species Index 134 Keyword Index 135 THE MOUNT DESERT ISLAND BIOLOGICAL LABORATORY

RESEARCH AND EDUCATION IN THE BIOLOGY OF MARINE ANIMALS

INTRODUCTION The Mount Desert Island Biological Laboratory (MDIBL) is an independent non-profit biological station located on the north shore of Mount Desert Island, overlooking the gulf of Maine about 120 miles northeast of the Portland near the mouth of the Bay of Fundy. The island, well known for Acadia National Park, provides a variety of habitats including shallow and deep saltwater, a broad intertidal zone, saltwater and freshwater marshes, freshwater lakes and streams, forests and meadows. The Laboratory is the largest cold water research facility in the Eastern United States, and its unique site provides an outstanding environment for studying the physiology of marine and freshwater flora and fauna. During 2001, the scientific personnel included 55 principal investigators and 95 associates, representing 68 institutions in 23 states and four European countries.

HISTORY AND ORGANIZATION MDIBL was founded in 1898 at South Harpswell, Maine by J.S. Kingsley of Tufts University. Its present site at Salisbury Cove was donated by the Wild Gardens of Acadia, and relocation was completed in 1921. The Wild Gardens of Acadia, a land-holding group headed by George B. Dorr and John D. Rockefeller, Jr., who was instrumental in the founding of Acadia National Park. In 1914, the Laboratory was incorporated under the laws of the State of Maine as a non-profit scientific and educational institution. Founded as a teaching laboratory, MDIBL is now a center for marine research and education that attracts investigators and students from across the U.S. and around the world. Since the pioneering work of H.W. Smith, E.K. Marshall and Roy P. Forster on various aspects of renal and osmoregulatory physiology of local fauna, the Laboratory has become known worldwide as a center for investigations in electrolyte and transport physiology, developmental biology, electrophysiology and marine molecular biology. The Mount Desert Island Biological Laboratory is owned and operated by the Board of Trustees and Members of the Corporation; at present, there are 430 members. Officers of the Corporation - Chair, Vice-Chair, Director, Secretary, Treasurer, Clerk - and an Executive Committee are elected from among the Trustees. The Chair and Executive Committee oversee and promote long range goals of the Laboratory. The Director, with the aid of a full-time Administrative Director, staff and a Scientific Advisory Committee is responsible for implementing the scientific, educational and public service activities of the Laboratory. NIEHS TOXICOLOGY CENTER Introduction: The Center for Membrane Toxicity Studies (CMTS), an NIEHS Marine and Freshwater Biomedical Sciences Center was established at the Mount Desert Island Biological Laboratory (MDIBL) in 1985. The purpose of this Center has been to involve a group of internationally recognized investigators, who are primarily experts in mechanisms of epithelial transport, to study the biological effects of environmental pollutants on cell and membrane transport functions. The primary emphasis of this research effort has been to elucidate the mechanisms of toxicity of environmental pollutants at the cellular and molecular level, using novel aquatic models developed at this laboratory. The focus of the research programs of the Center has broadened in the last several years from the more narrow objective of identifying the molecular targets for the effects of heavy metals (or metal compounds) on cell functions, to include the effects of a broader range of environmental toxicants (including marine toxins) and the mechanisms by which the organism takes up and eliminates a wide range of xenobiotics and environmental pollutants. However, the concept that a “membrane lesion” accounts for the cellular toxicity of many environmental toxins still remains as a paradigm. Research Cores: The Center consists of 3 highly integrated research cores or themes consisting of: 1) Signal Transduction, 2) Ion and Cell Volume Regulation, and 3) Xenobiotic Transport and Excretion. Investigators in the Signal Transduction Core are examining the basic mechanisms concerning the cell’s signaling response to changes in its external environment, particularly as related to environmental stress, heavy metal exposure, marine toxins and environmental estrogens. Investigators in the Ion and Cell Volume Regulation Core are interested in determining the fundamental mechanisms by which cells regulate their cell volume, internal pH and secretory functions and how these processes are disturbed by environmental influences. Work in this Core has considerable overlap with the Signal Transduction Core. Investigators in the Xenobiotic Transport and Excretion Core are examining the processes that are used by various epithelial tissues such as the liver and kidney to take up and excrete drugs and xenobiotics and other toxic compounds that enter from the environment and to study the effects of toxicants on this process. Investigators in this Core also interact with investigators working in the other two Cores. Facilities Cores: The Center provides for 5 facility cores for Center investigators. These include: 1) an Animal Core that is responsible for the acquisition, and maintenance of the many marine species available to investigators at this Center; 2) an Instrumentation Core that maintains the basic laboratory equipment that investigators would not otherwise be able to easily bring to the laboratory (a fully equipped cell culture and molecular biology facility is also part of this core); 3) a Cell Isolation, Culture and Organ Perfusion Core that provides isolated cells and tissues from marine species to Center investigators; 4) an Electrophysiology Core that maintains equipment for basic electrophysiologic measurements as well as an oocyte injection facility; and 5) an Imaging Core that maintains and operates a confocal fluorescent microscope as well as providing other imaging technology including epifluorescence and video-enhanced microscopy. Community Outreach and Education Program: The Center’s outreach program involves community education on water monitoring programs. This is directed primarily at high school and college students in the immediate area of the laboratory. However, an extensive summer research educational program includes high school students from both regional and national sites, the latter emphasizing minority student education as well as college and postdoctoral fellowship training. Pilot Projects: The Pilot project program provides support for investigators who are interested in pursuing a new projected related to environmental toxicology in one or more of the Center’s Research Cores. The purpose of these Pilot grants is to obtain preliminary data to facilitate new grant submissions. Grants are awarded competitively and successful applicants receive up to $10,000/season. APPLICATIONS AND FELLOWSHIPS Research space is available for the entire summer season (June 1 - September 30) or a half-season (June 1 - July 31 or August 1 - September 30). Applications for the coming summer must be submitted by February 1st each year. Investigators are invited to use the year-round facilities at other times of the year, but such plans should include prior consultation with the MDIBL office concerning available facilities and specimen supply. A number of fellowships and scholarships are available to research scientists, undergraduate faculty and students, and high school students. These funds may be used to cover the cost of laboratory rent, housing and supplies. Stipends are granted with many of the student awards. Applicants for fellowships for the coming summer research period are generally due in January. For further information on research fellowships, please contact: Dr. Patricia H. Hand Administrative Director Mount Desert Island Biological Laboratory P.O. Box 35 Salisbury Cove, Maine 04672 Tel. (207) 288-3605 Fax. (207) 288-2130 [email protected] Students should contact: Michael McKernan Director of Education and Conferences [email protected] ACKNOWLEDGEMENTS The Mount Desert Island Biological Laboratory is indebted to the National Science Foundation and National Institutes of Health for substantial support. Funds for building renovations and new construction continue to permit the Laboratory to expand and upgrade its research and teaching facilities. Individual research projects served by the Laboratory are funded by private and government agencies, and all of these projects have benefited from the NSF and NIH grants to the Laboratory. For supporting our educational initiative, MDIBL acknowledges the Cserr/Grass Foundation, Milbury Fellowship Fund, American Heart Association – ME, NH, VT Affiliate, Blum/Halsey Fellowship, Stanley Bradley Fund, Stan and Judy Fund, Bodil Schmidt-Nielsen Fellowship Fund, Maine Community Foundation, NSF - Research Experience for Undergraduates, the Hearst Foundation and many local businesses and individuals. Adrian Hogben (1921 - 2001)

Adrian Hogben was a highly valued colleague, mentor and friend to many at MDIBL. Born on Nov. 12, 1921 in Buckinghamshire, England; he received the majority of his education and scientific training in the U.S. Adrian earned a BS from the University of Wisconsin in 1941 and the MD from the same institution in 1943. In 1950 he completed the Ph.D. at the University of Minnesota. After an internship at Philadelphia General Hospital in 1944 he served for two years in the U.S. Army medical Corps and then as a fellow in medicine at the Mayo Clinic.

Adrian’s scientific career, that established him as an international leader in the field of membrane transport, was launched during a two-year stay in the lab of Professor Hans Ussing in Copenhagen. There he made the pivotal discovery that the gastric mucosa of the frog actively secreted chloride, perhaps the first definitive demonstration of active chloride transport by any cell membrane (PNAS 37:393-395, 1951). The behavior of the gastric mucosa and the relation of chloride movement to acid secretion was to occupy him for the remainder of his scientific career.

When Adrian first became an investigator at MDIBL in 1957 he had already served on the NIH staff in the Section on Kidney and Electrolyte Metabolism (1951 - 57) and had been selected as Chair of the Department of Physiology at the George Washington University Medical School (1957 -- 61). Adrian's work at MDIBL followed in the great tradition of Smith and Marshall as he searched for marine models to use in the solution of problems in medical physiology. Within a few years he had published in both Nature and Science describing unique properties of the teleost swim badder and the elasmobranch stomach. During his 44 years of association with MDIBL Adrian worked to recruit scientists and to contribute to the scientific development of junior investigators. During a major portion of this time he also served as Professor and Chair of the Department of Physiology and Biophysics at University of Iowa College of Medicine.

Throughout his career Adrian was engaged in Science, both as an individual activity and a collective enterprise. He served on major peer review panels for NIH and NSF as well as numerous editorial boards. He served MDIBL as a member of the Executive Committee and the Scientific Advisory Committee. While at George Washington he even taught a course in physiology on television (WTOP-TV)!

Adrian’s wry sense of humor was always at the ready during casual conversation or vigorous is scientific exchange. The MDIBL community will miss him and always remember his unflagging support and love for the lab; the sort that has sustained this island institution for over 100 years.

David C. Dawson, Ph.D. Oregon Health Sciences University David F. Opdyke (1915 - 2001)

David Opdyke, a long time research scientist at the Mount Desert Island Biological Laboratory(MDIBL) and friend of the laboratory in many capacities, died April 2, 2001 at the age of 85. Born in 1915 in Montpelier, Ohio, Dave graduated from Heidelberg College and received his Ph.D. from Indiana University where he held faculty appointments there and at Case Western Reserve School of Medicine where he worked with the famous Carl Wiggers pursuing his life long interest in cardiovascular physiology. Dr. Opdyke left academia for a five year period beginning in 1951 to work for the Merck Institute for Therapeutic Research. He enjoyed the science and international travelling involved, but missed the contact with students which he so enjoyed.

This vacuum in his life was soon filled when he became the first faculty member hired by the Seton Hall Medical School in New Jersey. Dave was instrumental in the formation and success of this school which evolved into the University of Medicine and Dentistry of New Jersey (UMDNJ). As the first Department Chair of Physiology at UMDNJ, Dave formed a cohesive group of leading cardiovascular researchers who in turn graduated many fine scientists and physicians. During his years at UMDNJ Dave personally mentored scores of graduate students and over his lifetime authored over 100 scientific publications. After retiring from his New Jersey post, Dr. Opdyke continued part time teaching at East Carolina State University Medical School in North Carolina until he was 77 years old.

The happiest times in Dave Opdyke’s life were spent at MDIBL where he arrived every spring with his wife Betty, children David and Nancy and always a graduate or undergraduate student to study endocrine controls of the cardiovascular system in dogfish shark. His first year at the laboratory was1969. In his introductory seminar in Dahlgren Hall, Dave paced across the then trampoline floor and in his characteristic gravely voice humbly told a room full of nephrologists and renal experts that he was “just a cardiovascular physiologist”. In the same unassuming way in the ‘80’s and early ‘90’s Dave managed the Laboratory endowment fund and with wise investing, single-handedly helped it grow almost forty-fold. Dave served as Trustee of the Laboratory and Chair of the Finance Committee during his working years, but perhaps his most famous role, which he assumed after retirement, was that of “Dr. Shark”. Every Wednesday afternoon, dressed in his Mount Desert Island “map” short sleeved shirt, Dave would enthrall and enchant the many children and adults who came on tour to see the marine animals at the MDIBL touch tank. He loved to pick up the sharks and other specimens and tell about their unique features and how they were useful to science.

Frenchman Bay itself held a fascination for Dave Opdyke. Many a sunny summer afternoon would find him standing straight up through the mid hatch in his motor boat, “Summer Thing”, speeding across miles of splashing waves to explore all the hidden nooks and crannies in the Bay. In later years he had a small Cape Islander made which was large enough to accommodate an overnight for himself and Betty on their more elaborate exploration adventures.

Dave represented the true essence of the Laboratory – rigorous research and teaching, dedicated service, and an unfaltering love for the surrounding beauty of the place. His Laboratory family misses him.

Barbara Kent, Ph.D. Mount Desert Island Biological Laboratory David B. Peakall (1931 - 2001)

David B. Peakall died suddenly and unexpectedly following surgery on August 18, 2001. David was trained as a chemist obtaining a Ph.D. in Physical Chemistry in 1956 from the University of London and a D.Sc. from the same institution in 1979 for his thesis on The Ecological Effects of Pollutants.. His long and distinguished research career demonstrates the successful melding of his interest and training in chemistry with a life- long interest in ornithology. His work was conducted at Department of Anatomy of the Upstate Medical Center in Syracuse, NY, Laboratory of Ornithology and the Section of Ecology and Systematics, Division of Biological Sciences at Cornell University, the Canadian Wildlife Service, where he was Chief of Wildlife Toxicology and at MDIBL from 1972-1985.

David’s work focused specifically on understanding how environmental pollutants affect bird populations and more generally on identifying ways to use wildlife species to monitor the health of the environment (biomarkers) and environmental health. David’s team worked collaboratively with Canadian Wildlife Service biologists across Canada on such issues as reproductive declines in falcons, effects of mercury on common loon reproduction, impacts of forest spraying on songbird populations, effects of acid precipitation, impacts of dioxin in pulp mill effluent on great blue heron reproduction, and the effects of contaminants on fish eating birds particularly in the Great Lakes. David conceived the idea of swapping eggs between “clean” and “dirty” colonies to isolate the impact on parental behavior from that of embryo toxicity. In the late 1960s and 1970s, David’s work was instrumental in determining how persistent DDT pollution caused egg shell thinning that decimated populations of some species of birds. A good portion of this work was done at MDIBL in collaboration with Bill Kinter and David Miller, the result being several publications in Science and Nature. This highly successful collaborative effort was one of the first substantial programs in environmental toxicology at the lab. David and MDIBL collaborators then turned their attention to the sub-lethal effects of ingested crude oil on sea birds. In combined field and laboratory studies, they found small oral doses caused disruption of osmoregulation, nutrient utilization, growth and nesting behavior. They also identified the components of the oil that were most toxic, these being ones that were concentrated as spilled oil weathered.

David authored over 140 scientific papers and book chapters as well as four books. However, he will be remembered best by those who knew him for his hospitality, his love of a good cup of tea, fine wines and good food. He cared passionately about our world and its inhabitants. He was a fine scientist, a loving and devoted husband and father, and a great friend.

David M. Miller, Ph.D. National Institute of Environmental Health Sciences REPORT TITLES

Reports preceded by an asterisk were prepared by investigators funded by the NIEHS Center for Membrane Toxicity Studies at the Mount Desert Island Biological Laboratory

IONIC REGULATION

* Silva, P., C. Sighinolfi, J. Richards, R. Hays, K. Spokes and F. Epstein. Colchicine does not prevent stimulation of shark rectal gland by VIP or CNP 1

* Epstein, F., C. Sighinolfi, J. Richards, R. Hays, K. Spokes and P. Silva. Effect of nocodazolone on stimulation of shark rectal gland by VIP and CNP 3

Fellner, S. and L. Parker. A Calcium sensing receptor (CaSR) modulates the function of rectal gland artery (RGA) and tubules (RGT) in Squalus acanthias 5

Evans, D., J. Roeser and J. Stidham. Natriuretic peptide hormones do not inhibit NaCl transport across the opercular skin of the killifish, Fundulus heteroclitus 7

Evans, D., J. Roeser and J. Stidham. Endothelin and nitric oxide interact to inhibit NaCl transport across the opercular skin of the killifish, Fundulus heteroclitus 8

Evans, D., R. Roeser and J. Stidham. Characterization of the receptor mediating the prostaglandin-induced inhibition of NaCl transport across the opercular skin of the killifish, Fundulus heteroclitus 9

Pelis, R. and J.L. Renfro. Active sulfate secretion by the intestine of winter flounder (Pleuronectes americanus) 10

Morrison-Shetlar, A., S. Edwards and J.B. Claiborne. Molecular identification and cloning of an NHE-2 like isoform from the gills of the dogfish shark (Squalus acanthias) 11

Smith, C., H. Skaggs and R. Henry. Salinity adaptations in the euryhaline green crab, Carcinus maenas 12

Mickle, J., K. Zarella, L. Petell, A. Lankowski, M. Clay and J. Dranoff. Identification and characterization of a novel P2Y nucleotide receptor in Fundulus heteroclitus. 15

Towle, D. and P. Peppin. Na+/K+/2Cl- cotransporter mRNA expression in the blue crab Callinectes sapidus measured by real-time quantitative PCR 17

Koomoa, D.L., M. George and L. Goldstein. Skate (Raja erinacea) anion exchanger, sAE1, expression in Xenopus laevis oocytes 18 Kormanik, G. Inhibition of glutamine synthetase increases branchial ammonia excretion by the dogfish, Squalus acanthias 19

Edwards, S., A. Morrison-Shetlar and J.B. Claiborne. Molecular identification of Na+/H+ exchanger cDNA in the gills of the euryhaline mummichog (Fundulus heteroclitus) 20

Hair, N., S. Edwards, A. Morrison-Shetlar and J.B. Claiborne. Relative expression of mRNA for NHE-2 in the gills of the longhorned sculpin, Myoxocephalus octodecimspinosus 21

Weakley, J., S. Sligh, A. Morrison-Shetlar and J.B. Claiborne. Effect of salinity alterations and respiratory acidosus on acid-base transfers in the mummichog (Fundulus heteroclitus) 23

Guizouarn, H., D. Myers and L. Goldstein. Cloning of sAE3 anion exchanger of skate (Raja erinacea) erythrocyte 25

* Karnaky, K., E. Milner, J.N. Forrest, Jr. and L. Forte. Guanylin / guanylate cyclase signaling in the intestine of dogfish shark (Squalus acanthias) and American eel (Anguilla rostrata) 26

Mickle, J., C. Blaschak, C. Yurk and G. Cutting. Localization of GFP-KFCFTR to the plasma membrane of mammalian epithelial cells 29

King, S., S. Pedersen, R. Rigor and P. Cala. Partial sequence of the sodium hydrogen exchanger from winter flounder red blood cells 31

XENOBIOTIC TRANSPORT

* Cai, S.Y., C. Soroka, N. Ballatori and J. Boyer. Molecular identification and tissue distribution of a multidrug resistance associated protein (Mrp2) isolated from the liver of the little skate (Raja erinacea) 33

* Notenboom, S., R. Masereeuw, F. Russel and D. Miller. Measurement of xenobiotic induced nitric oxide production in killifish (Fundulus heteroclitus) renal proximal tubules 35

* Bauer, B., G. Fricker and D. Miller. Regulation of xenobiotic efflux pumps in killifish (Fundulus heteroclitus) brain capillaries 36

* Henson, J., S. Kolnik, S.Y. Cai, N. Ballatori and J. Boyer. Localization of endogenous xenobiotic transporters in isolated clusters of polarized skate hepatocytes 37 MOLECULAR TOXICOLOGY

* Maples, N., J. Peterson and L. Bain. Reproductive impacts of arsenic exposure in mummichogs mediated by Mrp1 39

* Baldwin, W. and J. Roling. Development of two subtractive cDNA libraries from winter flounder 40

* Villalobos, A., D. Miller, C. Dehm and J. Larry Renfro. Structural and biochemical changs in blood-CSF barrier of Squalus acanthias in response to heat shock and zinc exposure 41

* Kullman, S. and D. Hinton. Toxicant induced differential gene expression and production of an aquatic gene array 43

* Barnes, D., J.N. Forrest, Jr., J. Wise, Sr., and R. Winn. Derivation of continuous marine cell lines for model systems in toxicology and cell biology 45

* Wise, J., Sr., P. Antonucci, S. Holt, L. Elmore, J.N. Forrest, Jr., J. Boyer and B. Bryant. Establishing cell lines in four marine animals: bowhead whale (Baelaena mysticetus), beluga whale (Delphinapterus leucas), little skate (Raja erinacea), and spiny dogfish (Squalus acanthias) 48

* Dowd, B., A. Bewley and B. Forbush. A low dose of mercury inhibits phosphorylation of the Na+-K+-Cl- cotransporter in the rectal gland of the dogfish shark, Squalus acanthias 50

COMPARATIVE BIOCHEMISTRY

Henry, R. A repressor substance of carbonic anhydrase induction is present in the eyestalks of euryhaline, but not stenohaline, crustaceans 52

Townsend, K., C. Spannings-Pierrot, D. Hartline, S. King, R. Henry and D. Towle. Expression of crustacean hyperglycemic hormone (CHH) mRNA in neuroendocrine organs of the shore crab Carcinus maenas 54

* Althoff, T., E. Kinne-Saffran, J. Luig, H. Schütz and R. Kinne. Sodium-D-glucose transport in Squalus acanthias kidney and intestine: Functional and molecular differences 56

Crockett, E. and R.P. Hassett. Enriched preparations of plasma membranes from zooplankton 57

Chakravarty, D., A. Kohn, R. Greenburg and D. Kültz. Fundulus heteroclitus 14-3-3.a inhibits an endogenous chloride channel in Xenopus laevis oocytes 60 Musch, M. and L. Goldstein. Activation of SYK tyrosine kinase is not required for oligmoerization of little skate (Raja erinacea) erythrocyte band 3 61

Koob, T. and M. Koob-Edmunds. Egg capsule proteins from four species of Northwestern Atlantic skates 62

Preston, R., I. Sud, V. Williams, A. Budhai and A. Bryant. Racemase activities in the tissues of marine mollusks 65

Weinman, E. and D. Steplock. Localization of NHERF-1, NHERF-2, and CFTR in the rectal gland of the spiny dogfish 67

* Hartzell, H.C. and R. Winn. Calcium transients during the first and second mitotic divisions of the fertilized medaka egg 71

* Ballatori, N., D. Seward, G. Fricker, M. Runnegar, J. Henson, D. Miller and J. Boyer. Retention of structural and functional polarity in cultured skate (Raja erinacea) hepatocytes 73

* Seward, D., R. Anderson, C. Bennet, J. McCoy, S.Y. Cai, J. Boyer and N. Ballatori. ß- Actin mRNA expression is markedly upregulated wheras its protein levels are unchanged in primary skate hepatocyte cultures from the little skate, Raja erinacea 75

Lankowski, A., M. Emerson, J. Mickle, J. Riordan, G. Cutting and B. Stanton. Trafficking of CFTR in killifish (Fundulus heteroclitus) opercular membrane: response to elevations in cAMP and adaptation to seawater 77

* Epstein, F., P. Silva, K. Spokes and R. Hays. CFTR and the myosin V motor protein are on the same transport vesicles in the rectal gland of Squalus acanthias 79

* Peters, K., J.N. Forrest, Jr., C. Marino and R. Frizzell. Fractionation of shark rectal gland cells for identification of CFTR trafficking compartments 81

Sasse, P., L. Cleemann, B. Fleischmann, J. Hescheler and M. Morad. Subcellular localisation of intracellular calcium stores in embryonic mouse cardiomyocytes 84

Ämmälä, C., I. Dukes, L.P. He, C. Rhodes, L. Philipson and L. Cleemann. Visualization of secretory granules in pancreatic islets and ß-cells using TIRF and confocal microscopy 88

Degtiar, V., E. Jones, M. Chau, L. Cleemann and M. Morad. TIRF imaging of focal Ca2+ release in voltage-clamped atrial and ventricular myocytes 92 COMPARATIVE PHYSIOLOGY

Evans, D. and M. Kozlowski. Characterization of endothelin, nitric oxide, prostaglandin E, and natriuretic peptide receptors in the bulbus arteriosus of the eel, Anguila rostrata 96

Evans, D. Direct observation of blood flow in the gill of the eel, Anguila rostrata, by videomicroscopy 97

Findlay, K., K. Ensign and G. Kidder III. Fundulus numbers in Northeast Creek vary with tide and salinity 99

Lenz, P., D. Hartline and A. Hower. Calanus finmarchicus: Kinematics of the pereiopods during an escape 100

Wellner, M., J. Litteral, T. Kirsch, P. Waldron, M. Elger, H. Hentschel and H. Haller. Isolation of genes expressed in developing nephrons in the adult kidney of Squalus acanthias 101

* Weber,G., M. Bewley, J. Pena and J.N. Forrest, Jr. Functional expression of shark VIP receptor in Xenopus oocytes: activation of CFTR chloride channels 103

* Motley, W, S. Forrest, R. Williams, S. Decker and J.N. Forrest, Jr. Effects of procaine on secretagogue stimulated chloride secretion in the rectal gland of the dogfish shark, Squalus acanthias 105

* Decker, S., S. Forrest, W. Motley, R. Williams and J.N. Forrest, Jr. Peptide histidine isoleucineamide stimulates chloride secretion in the perfused shark rectal gland 108

* Forrest, S., A. Pelletier, S. Decker and J.N. Forrest, Jr. Staurosporine and cytochalasin D do not inhibit the chloride secretory response to constant infusions of shark C-type natriuretic peptide 109 COLCHICINE DOES NOT PREVENT STIMULATION OF SHARK RECTAL GLAND BY VIP OR CNP.

Patricio Silva,1 Christopher Signolfi,2 Jana Richards,3 Richard Hays,4 Katherine Spokes,5 and Franklin H. Epstein.5 Departments of Medicine, 1Temple University Hospital, Philadelphia, PA 19140, 4Albert Einstein College of Medicine, Bronx, NY 10461, 5Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, MA 02215. 2Brewer High School, Brewer, ME 04412. 3Bucksport High School, Bucksport, ME 04421.

Previous work from our laboratory has suggested that the actin cytoskeleton is involved in the stimulation of chloride secretion induced by C-type natriuretic peptide (CNP) in the rectal gland of Squalus acanthias, while the stimulation induced by vasoactive intestinal peptide (VIP) is independent of it (Silva, P. et al. Bull MDIBL 39:5-7, 2000 and Bull MDIBL 40:27-8, 2001), In the present experiments we examined the role of the microtubular component of the cytoskeleton using colchicine to disrupt the intracellular microtubules.

Shark rectal glands were perfused as described in Silva, P, et al. (Methods Enzymol 192:754-66, 1990). The secretion of chloride was stimulated with either VIP or CNP, given as a bolus infusion over a period of one minute at the dose calculated to expose the perfused gland to 10-7M for VIP, and 5x10-7M for CNP. In all experiments the glands were perfused with procaine 10-2M to prevent the release of VIP from nerves within the rectal gland. Colchicine was used at a concentration of 2.5 x 10-5M or 10-4M throughout the experiment. After a control period of perfusion consisting of three 10-min collections, during which a stable basal secretory rate was established, a bolus of 10-7M VIP (Sigma Chemical Co.) or 5 x 10-7M C-type natriuretic peptide (Sigma Chemical Co.) was given over one min, without altering the rate of gland perfusion. Collections were continued at ten minute intervals. Chloride in the rectal gland secretion was measured by amperometric titration using a Buchler-Cotlove chloridometer.

The results are summarized in Figures 1 and 2. Perfusion with 25 µM or 100 µM colchicine did not inhibit the stimulation of the perfused rectal glands by CNP or by VIP.

800 Figure 1. Effect of colchicine on stimulation of perfused glands by CNP. Columns represent the average of 3 ten minute collection periods ± 600 standard error of the mean. CNP was given over one minute as a bolus at a final concentration of -7 5 x 10 M. n=5 for Basal and CNP, n=8 for 400 colchicine 20 µM, n=4 for colchicine 80 µM, and n=12 for both colchicine doses combined,

Colchicine did not inhibit stimulation by CNP Chloride secretion (µEq/h/g) 200 when present in the perfusion at a concentration of 20 or 80 µM. 0 Basal CNP CNP + colchi- CNP + colchi- CNP + colchi- cine 20 µM cine 80 µM cine 20 & 80 1200

Figure 2. Effect of colchicine on stimulation of 1000 perfused glands by VIP. Columns represent the

mean ± standard error of the mean of three 800 consecutive 10 minute periods before (Basal) and -7 after the infusion over 1 minute of 10 M VIP 600 (final concentration). Colchicine 20 or 80 µM did

not inhibit stimulation by VIP. n=6 for basal and 400 VIP, n=6 for 20 µM colchicine, n=4 for 80 µM Chloride secretion (µEq/h/g) colchicine, and n=10 for 20 and 80 µM 200 colchicine.

0 Basal VIP VIP + colchi- VIP + colchi- VIP + colchi- cine 20 µM cine 80 µM cine 20 & 80

These results suggest that, as contrasted with nocodazole, colchicine at the doses used in these experiments does not inhibit the secretion of chloride stimulated by CNP or VIP. Supported in part by MDIBL and NIEHS P30 ESO3828-16. EFFECT OF NOCODAZOLE ON STIMULATION OF SHARK RECTAL GLAND BY VIP AND CNP.

Franklin H. Epstein,1 Christopher Signolfi,2 Jana Richards,3 Richard Hays,4 Katherine Spokes1, and Patricio Silva.5 1Department of Medicine, Harvard Medical School and Beth Israel Deaconess Medical Center, Boston, MA 02215 4 Department of Medicine, Albert Einstein College of Medicine, Bronx, NY 10461 2Brewer High School, Brewer, ME 04412 3Bucksport High School, Bucksport, ME 04421. 5 Department of Medicine, Temple University Hospital, Philadelphia, PA 19140.

The rectal gland of Squalus acanthias can be stimulated to secrete chloride by two endogenous hormones, vasoactive intestinal peptide(VIP) and C-type natriuretic peptide (CNP). These agonists entrain different intracellular second-messenger cascades. VIP activates adenylate cyclase and protein kinase A, whereas CNP activates guanylate cyclase and probably protein kinase C. Direct stimulation of perfused glands by CNP is blocked by cytochalasin D, which interferes with actin filament organization, and by ML-7, which inhibits myosin light chain kinase, but these agents do not inhibit chloride secretion stimulated by VIP. It seems reasonable therefore to hypothesize that the actin cytoskeleton plays a larger role in transducing stimulation of the rectal gland induced by CNP than by VIP.

In the present experiments we attempted to explore the role of the microtubular portion of the cytoskeleton in chloride transport stimulated by VIP and CNP, by using the pharmacological agent nocodazole (Sigma Chemical Co). Nocodazole binds and disrupts intracellular microtubules in a reversible way, so that it is feasible to study its effect during applications and after washout.

Isolated rectal glands of Squalus acanthias were perfused with oxygenated shark Ringer's solution at pH 7.6 as previously described (Silva, P. et al. Methods Enzymol 192:754-66, 1990). Procaine, 20 mM, was added to all perfusion media in order to block the release of VIP from nerves within the gland, an action of natriuretic peptides that might otherwise complicate evaluation of the direct effect of CNP on rectal gland epithelial cells. In experiments with nocodazole the concentration of this compound in the perfusate was 20 µM. Duct fluid was collected at 10 minute intervals in small tared plastic centrifuge tubes and the volume measured every 10 minutes by weighing. The concentration of chloride in the duct fluid was estimated by amperometric titration.

An initial thirty minutes of control perfusion (three collection periods) allowed the gland to reach a stable basal state. At the end of this control period a 1 ml bolus of either VIP (Sigma Chemical Co.) or C-type natriuretic peptide (Sigma Chemical Co.) was perfused directly into the arterial catheter over 1 min. After three ten minute periods, the perfusion solution was changed to one without nocodazole. After an additional 30 minutes of washout, another identical bolus of VIP or CNP was injected into the rectal gland artery and duct secretion was collected for 3 final ten minute periods. It was thus possible to estimate the effect of VIP and CNP on the rectal gland secretion with and without nocodazole, and after 30-60 minutes of nocodazole washout. The results are summarized in Figures 1 and 2. Perfusion with 20 µM nocodazole completely inhibited direct stimulation of the perfused rectal glands by CNP, and the inhibition was not relieved by nocodazole washout. Nocodazole partially inhibited, but did not prevent, stimulation by VIP, and this inhibition was reduced by nocodazole washout. Nocodazole consistently reduced the flow of perfusate to isolated glands, by and average of 32±4%, and the flow returned to normal levels during washout. This action is consistent with the known vasoconstrictive effect of microtubule depolymerization in mammalian vascular smooth muscle cells (Johns, D.G. et al. J. Biomed. Sci., 2000, 7: 1-43).

These results strengthen the hypothesis the the cytoskeleton is involved in the immediate stimulation of chloride secretion by the shark rectal gland. Disruption of the microtubular network inhibits secretion stimulated by CNP and VIP but the inhibition is more pronounced with CNP-induced stimulation. Supported in part by MDIBL and NIEHS P30 ESO3828- 16.

1250

Figure 1 . Effect of nocodazole on stimulation of 1000 perfused glands by CNP. Columns represent the average of 3 ten minute collection periods ± standard error of the mean. CNP was given over one minute as a bolus at a 750 final concentration of 5 x 10-7M. n= 5 for all columns.

Nocodazole completely inhibited stimulation by CNP 500 when present in the perfusion at a concentration of 20 µM, and after 30 minutes of washout. Values are mean±SEM. Chloride secretion (µEq/h/g) 250

0 Basal CNP CNP + CNP post nocodazole nocodazole

1000 Figure 2 . Effect of nocodazole on stimulation of perfused glands by VIP. Columns represent the mean ± standard error of the mean of three consecutive 10 750 minute periods before (Basal) and after the infusion over 1 minute of 10-7M VIP (final concentration). Nocodazole partially inhibited stimulation by VIP, (p<0.01) and the 500 inhibition was relieved during washout (p< 0.02). n=6 for basal and VIP and n=8 for VIP plus nocodazole and VIP post-nocodazole. Values are mean±SEM. 250 Chloride secretion (µEq/h/g)

0 Basal VIP VIP + VIP post nocodazole nocodazole

A CALCIUM SENSING RECEPTOR (CaSR) MODULATES THE FUNCTION OF RECTAL GLAND ARTERY (RGA) AND TUBULES (RGT) IN SQUALUS ACANTHIAS

Susan K. Fellner1 and Laurel Parker2 1 Department of Cell and Molecular Physiology, University of North Carolina at Chapel Hill, 27599. 2 University of Maine, Orono, ME.

Blood flow to the RGA in conscious sharks varies from less than 1% in some animals to 2-7% of the cardiac output in others, suggesting a pattern of intermittent blood flow (Kent and Olson, Am. J. Physiol. 243:R296-R303, 1982). Shuttleworth (J. Exp. Biol. 103:193-204, 1983) has shown that α-adrenergic catecholamines cause vasoconstriction of the RGA and that agents known to stimulate RGT secretion of salt (Cyclic AMP, vasoactive intestinal peptide (VIP)) abolished the vasoconstriction. Subsequent studies in the shark, Scyliorhinus canicula L., perfused at in vivo pressures showed that reducing perfusion to one third of control values caused a marked decline in sodium excretion by the gland (J. Exp. Biol. 125:373-384, 1986). Previous work from our laboratory (Fellner, 2+ MDI Bull. 40:88-90, 2001) showed that cytosolic calcium ([Ca ]i) of RGA responded in 2+ 2+ a dose-dependent fashion to changes in extracellular calcium ([Ca ]e) and that Ca entry was not affected by the L-channel blocker, nifedipine. Thus, in the current study we 2+ investigated the mechanism(s) by which changes in [Ca ]e might influence calcium signaling in not only in RGA and but also in RGT. In particular, we tested the hypothesis that a calcium sensing receptor (CaSR) might be present in both and contribute to intermittency of function of the rectal gland.

2+ We prepared RGA and RGT and measured [Ca ]i with fura-2 ratiometric analysis as previously described (MDI Bull. 40:88-90,2001). Rectal gland tubules were teased from thin slices of rectal gland from which the capsule had been removed. RGA segments with intact endothelium were < 0.1 mm in size. To investigate the possibility 2+ that a CaSR is responsible for the sensitivity of RGA and RGT to [Ca ]e, we employed a variety of agonists known to activate the CaSR. Agonist activation of a CaSR stimulates mobilization of Ca2+ from the endo- or sarcoplasmic reticulum (ER, SR) followed by Ca2+ entry through store-operated calcium entry (SOC) channels (Brown, EM, et al. Nature. 366:575-580,1993). Gadolinium, a specific inhibitor of SOC, is an agonist of the 2+ CaSR. RGA segments in nominally calcium-free shark Ringer’s had a baseline [Ca ]i of 165 ± 26 nmol/L which rose to 263 ± 27 nmol/L following the addition of Ca2+ (N = 10, 3+ 2+ P <0.01). Subsequent addition of Gd further increased [Ca ]i to 314 ± 23 nmol/L (P <0.05), suggesting that Gd3+ stimulates additional mobilization of Ca2+ from the SR. The response to Gd3+ is blunted compared to that achieved by other CaSR agonists likely because of the inhibitory effect on Ca2+ entry via SOC.

Spermine and other polyamines are agonists for the CaSR (Quinn, S.J. Am. J. Physiol. 273:C1315-23, 1997). Figure 1 compares the response of RGT in calcium-free Ringer’s to Ca2+ (5 mmol/L) followed by the addition of spermine (0.3 mmol/L) or Gd3+ (0.3mmol/L).

2+ 2+ The increment in [Ca ]i over baseline following the addition of Ca was similar 2+ in both studies (*, P < 0.01), but spermine further increased [Ca ]i by 235 ± 94 nmol/L (**, P <0.01) compared to only 60 ± 21 nmol/L for Gd3+ ( #, P <0.05). These 700 data show that RGT, like RGA, respond to external Ca2+ with a step-wise 600 ** 2+ i increase in [Ca ]i and that both ] 500 3+ 2+ spermine and Gd are agonists for a 400 CaSR in the tubules. As well, the * 3+ 300 # diminished response to Gd compared * to spermine may derive from the [Ca Increase in from Baseline 200 2+ inhibition of Ca entry via SOC by 100 3+ Gd . 0 2+ 2+ 2+ Ca Ca + spermine Ca Ca2+ + Gd3+

2+ Figure 1. Increment in [Ca ]i in RGT in calcium-free Ringer’s to external Ca2+ (5 mmol/L) followed by the addition of spermine (0.3 mmol/L) or Gd3+ (0.3mmol/L).

Because the extracellular domain of the CaSR interacts with polyvalent cations, it was hypothesized that activation of the receptor might occur through the screening of charged side chains of acidic or basic aminoacids. Furthermore, if ionic strength were increased by the addition of salts to the extracellular environment, the ability of polycations to trigger the CaSR should be decreased (Quinn, JS. et al. J. Biol. Chem. 273:19579-19586, 1998). In the isolated perfused RGT of Squalus acanthias, cAMP cases a biphasic response, the second of which arises from stimulation of the Na-K/2Cl cotransporter (Greger,R. et al. Pfluger’s Arch. 438:165-174). Addition of NaCl (150 mmol/L) to the perfusate caused cell shrinkage and a relative increase in cell chloride concentration. In mammalian ascending limb of Henle cells, activation of the CaSR results in diminished cAMP production and increased degradation (Ferreiera, M. J. Biol. Chem.273:15192-15102). Thus, inhibition of the CaSR by increasing ionic strength should result in an increase in cAMP and Cl transport.

We propose that the function of a CaSR in shark RGA and RGT is to inhibit tonically blood flow to the gland and to inhibit salt secretion by the tubules during non- feeding periods. When the shark ingests food and seawater, the resultant increase in salt concentration (ionic strength) of the blood and interstitial spaces would then inhibit the CaSR resulting in disinhibition of RGA vascular contraction and reversal of inhibition of salt secretion. Such a control mechanism would permit the animal to efficiently recruit the function of the rectal gland only during periods of feeding.

Future studies will be directed at studying the effect of alterations of ionic strength on salt secretion in the isolated perfused rectal gland and on calcium signaling in cells of RGA and RGT. Supported by a MDIBL New Investigator Award. Laurel Parker was supported by a National Science Foundation fellowship.

NATRIURETIC PEPTIDE HORMONES DO NOT INHIBIT NaCl TRANSPORT ACROSS THE OPERCULAR SKIN OF THE KILLIFISH, FUNDULUS HETEROCLITUS

David H. Evans1, Jennifer M. Roeser1, James D. Stidham2 1Department of Zoology, University of Florida, Gainesville, FL 32611 2Department of Biology, Presbyterian College, Clinton, SC 29325

As in mammals, fish natriuretic peptides (NPs) are secreted from the heart and produce a variety of cardiovascular effects (e.g., Takei, Y. Int Rev Cytol 194: 1-66, 2000). In addition, NPs inhibit drinking, as well as intestinal ionic uptake in teleosts (Loretz, C.A. Am Zool 35: 490-502. 1995. Interestingly, it is not clear whether NPs are actually natriuretic or anti- natriuretic in fishes (e.g., Loretz, C.A.and Pollina, C. Comp Biochem Physiol A 125: 169-87, 1995). It is generally accepted (e.g., above reviews, and McCormick, S.D. Am. Zool., in press, 2002) that NPs stimulate gill salt extrusion, largely due to a single study that demonstrated that a -7 truncated peptide (atriopeptin II) stimulated slightly (20% at 10 M) the short circuit current (Isc) across the opercular skin of the killifish (Scheide, J. and Zadunaisky, J. Am. J. Physiol. 254: R27-32, 1988). The fact that receptors are present and stimulate the production of cyclic GMP in fish gill tissue (e.g., Donald, J. et al. Am. J. Physiol. 267: R1437-1444, 1994) supports a role of the gills in NP responses, but the effect could be on vascular rather than epithelial tissue. Karnaky has been unable to confirm the stimulation of the opercular skin Isc (pers. comm.), so we decided to examine the effects of two natriuretic peptides (eel ANP and porcine CNP) on this system.

The experimental protocol was as described previously (Evans, D.H. et al. Bull. MDIBL 39: 17, 2000), with increasing concentrations of ANP or CNP (10-10 to 3 x 10-7 M) added to the basolateral side of pairs of opercular skin, mounted in an Ussing chamber. No changes in resistance were seen for any of the treatments. Data are expressed as % of initial Isc (CNP/ANP), or % of Isc before the SRX was added.

Neither fish ANP (eel) nor porcine CNP (both of which are vasoactive in fishes; e.g., Evans, D.H. J. Exp. Zool. 289: 273-284, 2001) produced a significant change in the Isc generated by the killifish opercular skin in our hands (Table 1; mean % control ± SE, N = 7-8). Importantly, SRX reduced the Isc approximately 50% when added at the end of the CNP or ANP concentration curve, showing that the tissue retained it's usual responsiveness to the ETB receptor agonist (Evans, D.H. et al., this volume). 10-10M 3x10-10M10-9 M 3x10-9M10-8 M 3x10-8M10-7 M 3x10-7M10-7 M SRX pCNP 100±1 100±2 100±3 99±4 100±4 104±3 106±3 104±2 45±5 eANP 98±1 98±2 97±3 97±3 101±4 105±5 108±6 111±5 52±8

Our data demonstrate clearly that neither eel ANP nor porcine CNP affect the Isc across the opercular skin, which is quite sensitive to the ET agonist, SRX. (Supported by NSF IBN- 0089943 to DHE and an REU supplement to JMR from this grant, as well as NSF DBI-9820400 to the MDIBL) ENDOTHELIN AND NITRIC OXIDE INTERACT TO INHIBIT NaCl TRANSPORT ACROSS THE OPERCULAR SKIN OF THE KILLIFISH, FUNDULUS HETEROCLITUS

David H. Evans1, Jennifer Roeser1, James D. Stidham2 1Department of Zoology, University of Florida, Gainesville, FL 32611 2Department of Biology, Presbyterian College, Clinton, SC 29325

Our initial studies demonstrated that both ET and NO could produce measurable inhibition of the short-circuit current (Isc) across the killifish operculum (Evans, D.H. et al. Bull. MDIBL 39: 17, 2000), the standard model for the teleost chloride cell (e.g., Evans, D.H. et al. J. Exp. Zool. 283: 641-652, 1999). To investigate this putative control system further, we have determined if the ET and NO effects are concentration-dependent, and whether there is any interaction between the two pathways.

The experimental protocol was as described previously (Evans, D.H. et al., Op. Cit., 2000), with increasing concentrations of ET-1 (mammalian; American Peptide), sarafotoxin S6c (ETB- receptor specific agonist; American Peptide), or the NO donor sodium nitroprusside (SNP, Sigma) added to the basolateral side of the opercular skin, mounted in an Ussing chamber. To avoid volume or carrier effects, the same volume of carrier was added to the apical side of the experimental tissue. In all experiments, tissue resistance was monitored by automatic opening of the circuit briefly every 2 minutes and recording the voltage. No changes in resistance were seen for any of the treatments. Since recent studies on mammalian kidney tubules have demonstrated that much of the ET inhibition of transport is mediated through the release of NO (e.g., Plato, C.F. et al., Am J Physiol Renal Physiol 279: F326-F333, 2000), we tested the effect of the NOS inhibitor, L-NAME on the ET-mediated inhibition.

ET-1, SRX S6c, and SNP all produced concentration-dependent inhibition of the Isc across the killifish opercular skin, with initial, statistically significant inhibitions at 10-10 M for both ET-1 and SRX S6c and 10-8 M for SNP. Addition of 10-4 M L-NAME to opercular skins was associated with a significant, 15%, increase in the Isc (N = 26, p = 0.0001), and the presence of the NOS inhibitor, reduced the subsequent SRX-induced inhibition by 18% (N = 14; p = 0.002). We conclude, therefore, that ET can inhibit opercular skin NaCl transport directly, but approximately 20% of the inhibition is via activation of ETB receptors and production of NO. Importantly, the initial stimulation of the Isc seen after L-NAME was added suggests that NaCl transport across the opercular skin is under tonic, inhibitory control of NO. What cells produce ET and NO in the gill epithelium is unknown as are the roles these interacting signaling axes play in the control of gill transport vs. gill perfusion. Future research will address these interesting questions. (Supported by NSF IBN-0089943 to DHE and an REU supplement to JR from this grant, as well as NSF DBI- 9820400 to the MDIBL) CHARACTERIZATION OF THE RECEPTOR MEDIATING THE PROSTAGLANDIN- INDUCED INHIBITION OF NaCl TRANSPORT ACROSS THE OPERCULAR SKIN OF THE KILLIFISH, FUNDULUS HETEROCLITUS

David H. Evans1, Jennifer M. Roeser1, James D. Stidham2 1Department of Zoology, University of Florida, Gainesville, FL 32611 2Department of Biology, Presbyterian College, Clinton, SC 29325

Our preliminary studies (Evans, D.H. et al. Bull. MDIBL 39: 17, 2000) have shown that, similar to ET-1 and nitric oxide, prostaglandin E (PGE2) inhibits the short-circuit current (Isc) across the killifish opercular skin, the standard model for NaCl extrusion by the fish gill (e.g., Evans, D.H. et al. J. Exp. Zool. 283: 641-652, 1999). Our data confirmed an earlier study, using PGE1 (Eriksson, O., et al. Acta Physiol. Scand 125: 55-66, 1985). To characterize the receptor(s) involved in this response, we examined the concentration dependence of the PGE2-induced inhibition, as well as the effect of receptor-specific agonists, which we have previously used to discriminate between putative PG receptors in the ventral aorta of the shark (Evans, D.H. and Nowacki, J., Bull. MDIBL 40: 15, 2001). PGE2 can bind to at least four receptors, designated EP1-4 (e.g., Wright, D. H. et al., Am J. Physiol. 281: R1343-R1360, 2001).

The experimental protocol was the same as described in the accompanying abstract (Evans et al., this volume), except that increasing concentrations of PGE2, Butaprost (EP2-specific), or Sulprostone (EP1- and EP3-specific) were applied to the basolateral side of the isolated skin, with appropriate corrections for volume and carrier on the apical side. No significant changes in tissue resistance were seen in any of the experiments.

The inhibition of the Isc by PGE2 was concentration dependent, statistically significant at all concentrations at or above 10-10 M and reaching 40% at 10-5 M (N = 6). Application of Butaprost vs. Sulprostone in another series of experiments ( N = 6) produced opposing effects. Inhibition of Isc by sulprostone was apparently concentration-dependent, but only statistically significant at 3x10-7 M and above; stimulation by butaprost was only significant at 3x 10-8 and 10-7 M. Concentrations above that produced stimulation also, but high variability precluded statistical significance (p = 0.06-0.08). In mammalian systems, EP2 stimulates the intracellular production of cyclic AMP, EP1 stimulates IP3, and EP3 inhibits cyclic AMP. It remains to be seen which receptors are expressed and important in the fish gill, but our preliminary data suggest that the local effecter, PGE2 may modulate active NaCl transport across the teleost fish gill epithelium via two opposing receptors. (Supported by NSF IBN-0089943 to DHE and an REU supplement to JMR from this grant, as well as NSF DBI-9820400 to the MDIBL) ACTIVE SULFATE SECRETION BY THE INTESTINE OF WINTER FLOUNDER (PLEURONECTES AMERICANUS)

Ryan M. Pelis and J. Larry Renfro Department of Physiology and Neurobiology, University of Connecticut, Storrs, CT 06269

Marine teleosts must drink seawater to prevent dehydration; thus, the intestine plays an essential role in ionic and osmotic homeostasis. Fluid entering the intestine is isosmotic with plasma due (in part) to desalination in the esophagus (Hirano, T., and Mayer-Gostan, N., Proc. Natl. Acad.Sci. USA 73: 1348-1350,1976). Active absorption of sodium and chloride by the intestine drives water absorption (Skadhauge, E., J. Physiol. 204: 135-158,1969). Divalent ions are also ingested and absorbed in large quantities during drinking. However, the mechanisms of divalent ion transport by the marine teleost intestine are not well understood. In preliminary experiments reported here we have determined the direction and mechanism of sulfate transport by the intestine of winter flounder (Pleuronectes americanus).

Paired Ussing chambers were used to measure unidirectional fluxes of radioactive sulfate 35 2- ( SO4 ) across the intestinal epithelium of winter flounder. Under short-circuited conditions sulfate was actively secreted (blood-to-lumen) at a rate of 7.42 ± 0.63 nmoles x cm-2 x hr-1 (n=23). The transepithelial potential difference (TPD) and resistance (TER) averaged 0.83 ± 0.17 mV (mucosal positive) and 37.1 ± 3.81W x cm2, respectively. Short-circuit current (Isc), a measure of chloride absorption across this tissue was –10.0 ± 1.02 mA x cm-2. The electrical properties (TPD, TER and Isc) of this epithelium are in good agreement with published results (Field, M. et al., J. Membrane Biol. 41: 265-293,1978). Application of sodium cyanide (10 mM) abolished net sulfate transport. Ouabain (0.1 mM) lowered net sulfate secretion to 25 % of the control value by reducing the secretory flux. Taken together, the effects of sodium cyanide and ouabain suggest that active sulfate secretion is metabolically and sodium-gradient dependent.

In the marine teleost renal proximal tubule (Renfro, J.L. and Pritchard, J.B., Am. J. Physiol. 244: F488-F496, 1983) sulfate secretion occurs by anion exchange, and this prompted the examination of similar mechanisms in the intestine. In the present study, net sulfate transport decreased to zero following removal of lumenal chloride and bicarbonate. Removal of lumenal chloride alone lowered net sulfate secretion by 75 % while removal of lumenal bicarbonate increased net secretion by 50 %. The anion exchange inhibitor DIDS (4,4’- diisothiocyanatostilbene-2-2’-disulfonic acid, 0.2 mM) had no effect when applied to the serosal side; however, mucosal application reduced net secretion to 52 % of the control value. Neither serosal nor mucosal DIDS altered TPD, TER or Isc. Net sulfate secretion was reduced to zero in fish fed (i.e. food in stomach) just prior to flux measurements. Feeding decreased unidirectional secretion 33% and increased absorption 100 %. Vasoactive intestinal peptide (VIP, 10-6 M), a gastrointestinal tract hormone known to stimulate pancreatic duct bicarbonate secretion in mammals, reduced flounder intestinal net sulfate secretion 56% through a reduction in secretion and increase in absorption similar to that elicited by feeding.

This preliminary investigation has identified active sulfate secretion by the marine teleost intestine, a process facilitated by a lumenal anion exchanger with greater affinity for chloride than bicarbonate. Feeding and VIP reduce sulfate secretion indicating that this mechanism may be hormonally controlled. Supported by NSF-IBN0078093.

MOLECULAR IDENTIFICATION AND CLONING OF AN NHE-2 LIKE ISOFORM FROM THE GILLS OF THE DOGFISH SHARK (SQUALUS ACANTHIAS).

Alison Morrison-Shetlar, Susan Edwards, J.B. Claiborne Department of Biology, Georgia Southern University, Statesboro GA 30460

The branchial epithelial tissue is the predominant site of acid-base regulation in fish and the sodium-hydrogen exchanger (NHE) and is thought to be involved in ion regulation and acid secretion in the gills of marine species (Claiborne, Physiology of Fishes 179-200, 1998, Edwards, et al. Comp. Biochem. Physiol. Part A, 130:81-91, 2001). There are seven known mammalian isoforms of NHE and one non-mammalian isoform (NHE-β) (Wakabayashi, et al. Physiol. Rev. 77 (1): 51-74, 1997). The objective of this study was to use molecular techniques to identify the presence of the mRNA for NHE in the gill of the dogfish shark (Squalus acanthias), obtain the full-length cDNA sequence, and determine the isoform(s) present in the dogfish shark gills.

The gills of the dogfish shark were excised and total RNA extracted by the method of Chomczynski and Sacchi (Analyt. Biochem.162:156-159, 1987). The purified RNA was used in a 3’/5’ RACE (Generacer kit, Invitrogen) reaction, with gene specific primers (GSP) for the 3’ and 5’ which were designed against known fragments of NHE –2 like dogfish sequences previously obtained (Morrison-Shetlar - unpublished data). The initial RACE reaction often resulted in multiple bands. To obtain single bands of the correct size, up to three nested PCR reactions needed to be performed, each using a GSP generated slightly up or down stream and the USP primer that bound to the synthetic primer in the 3’5’ RACE reaction (Fig. 1).

Figure 1. Nesting of 3' race product using specific primers to obtain single gene transcripts.

The resulting 3’ and 5’RACE products were isolated by gel electrophoresis and following gel extraction, subcloned into pCR 2.1 vector (Invitrogen) and transformed into bacteria. Positive clones were picked and the DNA isolated and sequenced. As new sequence was generated, additional primers were synthesized for sequencing until the complete gene was sequenced.

The full-length 2605 nucleotide shark NHE sequence, incorporating non-translated 5’ and 3’ data, was identified and found to contain an open reading frame encoding 749 amino acids. The sequence was compared to other known mammalian and non-mammalian sequences (GenBank BLAST) a high homology to rat NHE2 (62% identity, 76% similarity) was obtained. To date, this is the first full-length shark NHE-2 like cDNA known.

This research was funded by NSF grants #IBN 9808141 & #IBN 0111073 to JBC & AIMS.

SALINITY ADAPTATIONS IN THE EURYHALINE GREEN CRAB, CARCINUS MAENAS

Catherine Smith1, Hollie Skaggs2, and Raymond Henry2 1University of Illinois High School, Champagne, IL 61801 2Department of Biological Sciences, Auburn University, Auburn AL 36849

Euryhaline marine crustaceans are osmotic and ionic conformers when adapted to high salinity (33 ppt); their hemolymph osmotic and ionic concentrations passively reflect those in the surrounding seawater. Hemolymph osmotic concentration is made up primarily of salts (NaCl), but intracellular osmotic concentration contains high amounts of small organic molecules (free amino acids and quaternary ammonium compounds). These organisms can invade the more dilute waters of the estuary because they possess the ability to activate physiological and biochemical mechanisms that allow them to regulate hemolymph osmotic and ionic concentrations above those in the ambient water. However, these organisms still undergo significant hemodilution during the acute phase of low salinity acclimation, and this can result in tissue water gain and cell swelling. Cell volume is usually readjusted by reducing the intracellular concentration of these free amino acids (FAA) and quaternary ammonium compounds. These are released intact from the cells into the hemolymph and are believed to be metabolized by some as of yet unidentified tissue. Therefore, low salinity adaptation involves a coordinated response between ion regulation and cell volume regulation.

Carcinus maenas, the green shore crab, is a euryhaline marine species that is an osmoconformer at high salinity but that makes the transition to osmoregulation at a salinity of approximately 26 ppt. When transferred from 33 to 12 ppt, the hemolymph undergoes a dilution of about 200 mOsm in the initial 24 hr post-transfer (Table 1). This initiates the cell volume regulatory response, as evidenced by the increase in FAA in the hemolymph, as these substances are released from the intracellular space. The FAA are measured indirectly by assaying for the concentration of total ninhydrin-positive substances, or TNPS. Ninhydrin reacts with ammonia, amino acids, and quaternary ammonium compounds and can be detected colorimetrically. In crustacean tissues, the overwhelming majority of intracellular organic osmotic effectors is known to be the FAA. Hemolymph ammonia, as measured by the phenol-hypochlorite method on deproteinized samples, does not change, indicating that the FAA are neither metabolized prior to release into the hemolymph nor in the hemolymph itself.

Table 1. Time course of changes in hemolymph osmotic concentrations, and concentrations of total ninhydrin positive substances (TNPS) and ammonia in C. maenas acclimated to 33 ppt and transferred to 12 ppt. Mean + SEM (N = 6). T = 12oC.

Treatment Osmolality TNPS Ammonia -1 (mOsm kg H2O) (mM) (uM) 33 ppt acclimated 915 + 5 2.26 + 0.5 179 + 28 33-12 ppt acute transfer T1 hr 905 + 8 2.75 + 0.5 176 + 27 T4 hr 834 + 6 4.14 + 0.7 185 + 15 T12hr 751 + 16 3.84 + 0.5 163 + 33 T24 hr 713 + 29 6.57 + 0.9 192 + 18 T48 hr 696 + 40 7.80 + 0.9 196 + 19 T96 hr 674 + 37 5.11 + 0.3 119 + 14 T7 days 667 + 30 4.40 + 0.3 181 + 23

In all euryhaline marine invertebrates, including crustaceans, FAA are known not to be excreted from the animal; rather, it is believed that high rates of ammonia excretion during low salinity acclimation are indicative of FAA deamination, conservation of the carbon skeleton, and excretion of the amino group. C. maenas has a typical profile of ammonia excretion in response to low salinity exposure (Table 2): a significant increase in the rate of excretion almost immediately after low salinity transfer, peak rates occurring at the same time (4-48 hr) that tissue FAA are being released from cells into the hemolymph.

Table 2. Ammonia excretion in intact C. maenas acclimated to 33 ppt and at various times after transferred to 12 pt. Mean + SEM (N=6). T = 12oC.

Ammonia excretion (n mol gm-1 hr-1) 33 ppt T 2hr T 4 hr T 6 hr T 12 hr T 24 hr T 96 hr T 7day 182 + 40 438 + 80 541 + 120 594 + 160 688 + 120 412 + 30 382 + 30 306 + 50

The most likely site of deamination of FAA is the gill. If the FAA are taken up from the hemolymph by the gills and deaminated intracellularly, the carbon skeleton could be used for energy, and the resulting ammonia could be excreted directly into the ambient water without altering hemolymph ammonia concentrations. Initial evidence for this comes from the comparison of the intracellular FAA pools in leg muscle vs gills. At 33 ppt, the FAA pool in leg muscle is about 4-5 times higher than in either G2 or G9 (Table 3). After transfer to 12 ppt, the FAA pool in leg muscle declines rapidly over 24 hr and stabilizes at new, lower levels thereafter. The decline parallels the increases in hemolymph FAA and ammonia excretion. In contrast, the FAA pool in the gills actually increases during the acute phase of low salinity acclimation, possibly indicating uptake of amino acids from the hemolymph.

Table 3. The concentration of total ninhydrin positive substances (TNPS) in leg muscle, anterior (G2) and posterior (G9) gills of C. maenas acclimated to 33 ppt and transferred to 12 ppt. Mean + SEM (N = 6). T = 12oC.

TNPS (m mol gm-1 wet weight) Treatment Leg G2 G9 33 ppt acclimated 750 + 137 151 + 18 176 + 20 33 to 12 ppt transfer T 6hr 603 + 109 203 + 30 261 + 35 T 12 hr 666 + 102 209 + 31 246 + 37 T 24 hr 460 + 50 294 + 37 370 + 57 T 48 hr 406 + 52 274 + 46 415 + 40 T 96 hr 416 + 32 216 + 62 368 + 39 T 7day 365 + 121 291 + 16 500 + 39 Measurements of ammonia excretion from isolated anterior and posterior gills also supports this idea. Anterior (G4) and posterior (G8) were dissected out of crabs acclimated to 33 ppt, gently blotted free of seawater and hemolymph, and placed in cold, millipore-filtered seawater for 30 min in order to determine tissue ammonia production. At 33 ppt, G3 and G9 had about the same rate of ammonia production (Table 4). Anterior and posterior gills taken from crabs that had been exposed to 12 ppt for 2 hr had elevated rates of ammonia production, consistent with the gill as the central site of amino acid deamination during the cell volume regulatory response. Interestingly, the elevated rate of ammonia production persisted only in G3 after crabs were acclimated to 12 ppt for 14 days.

Table 4. Ammonia production from isolated anterior (G3) and posterior (G8) gills of C. maenas acclimated to 33 ppt and transferred to 12 ppt. Mean + SEM (N = 5). T = 12oC.

Ammonia production (u mol gm-1 hr-1) 33 ppt acclimated 33 - 12 ppt T 2hr 12 ppt acclimated G3 0.472 + 0.07 0.600 + 0.11 0.770 + 0.08 G8 0.504 + 0.06 0.661 + 0.11 0.492 + 0.08

These results strongly suggest that during the process of low salinity acclimation, in addition to the osmotic and ionic regulatory ability of C. maenas, mechanisms of cell volume regulation are also necessary for survival in low salinity. The green crab appears to have a typical invertebrate cell volume regulatory response: cell swelling followed by reduction of the intracellular FAA pool, loss of cellular water, and restoration of cell volume. The FAA pool is reduced by release of amino acids into the hemolymph, and transport via the hemolymph to the gills where they are taken up and deaminated. The resultant ammonia is excreted directly into the ambient water, and the carbon skeleton is conserved.

Supported by NSF IBN 97-27835 to RPH, an NSF REU award (HSS), by an Auburn University Undergraduate Research Fellowship (HSS), and by the Auburn University Fund for Excellence (HSS). IDENTIFICATION AND CHARACTERIZATION OF A NOVEL P2Y NUCLEOTIDE RECEPTOR IN FUNDULUS HETEROCLITUS

John E. Mickle1, Katrina Zarella2, Lydia Petell3, Alexander Lankowski4, Michael Clay5, and Jonathan A. Dranoff6

1Institute of Genetic Medicine, Johns Hopkins Univ. School of Medicine, Baltimore, MD 21287 2College of the Atlantic, Bar Harbor, ME 04609 3Bates College, Lewiston, ME 04240 4Dartmouth College, Hanover, NH 03755 5Sumner Memorial High School, Sullivan, ME 04664 6Liver Ctr & Dept. Med., Sec Digestive Diseases, Yale Univ. Sch. Med., New Haven, CT 06520

Killifish, Fundulus heteroclitus, is a euryhaline estuarine teleost that readily adapts to extreme variation in environmental salinity. This variation creates substantial osmotic stresses that affect cell volume homeostasis, requiring compensatory activation of solute transport. The mechanism by which this process occurs is unknown. To elucidate the underlying mechanisms we sought to identify the molecular components involved in cell volume regulation. Extracellular nucleotides are known to regulate chloride secretion and cell volume in mammalian systems by interaction with specific P2Y cell surface receptors (Dranoff, J.A. et al., Am. J. Physiol. 281:G1059-67, 2001). Thus, we tested the hypothesis that extracellular nucleotides regulate secretion via P2Y receptors in killifish. Accordingly, we examined the expression, distribution and function of P2Y receptors to determine whether these receptors regulate chloride transport in secretory epithelia of killifish.

The expression of P2Y receptors in killifish was examined using a homology cloning approach. Killifish were collected at Northeast Creek (Mount Desert Island, ME), and held in aquaria containing running seawater prior to use. Various tissues were dissected from seawater- adapted killifish, total RNA was extracted, reverse transcribed, and subsequently amplified by PCR using degenerate oligonucleotides designed to detect P2Y receptors P2Y1, P2Y2, P2Y4, and P2Y6. A PCR product of 400 bp was observed upon agarose gel electrophoresis from samples of killifish operculum, gill, eye, brain, intestine, skeletal muscle, and spleen. The product was excised from the gel and sequenced to determine molecular identity. Phylogenetic analysis (BLASTx, http://www.ncbi.nlm.nih.gov/BLAST/) indicates that this novel product is approximately 55% similar to the P2Y subclass P2Y4 from turkey (Figure 1). These experiments demonstrate that killifish expresses mRNA for a single P2Y receptor related to the P2Y4 receptor of higher species, and that this receptor is expressed ubiquitously in killifish.

The cellular distribution of P2Y receptors in killifish was determined by examining ligand- 2+ 2+ mediated cytosolic Ca (Ca i) increases in individual cells of intact opercular epithelium. P2Y receptors act as G protein-coupled receptors that bind extracellular nucleotide ligands such as ATP. Binding of ATP or other nucleotides to P2Y receptors leads to inositol trisphosphate 2+ 2+ (InsP3)-mediated cytosolic Ca (Ca i) increases in cells. To determine whether extracellular 2+ nucleotides activated P2Y-dependent Ca i increases in killifish cells, opercular epithelium was dissected as described previously, and incubated at room temperature for 30 minutes with the Ca2+ indicator fluo-4/AM (Molecular Probes, Eugene, OR) and the mitochondrial indicator mitotracker red (Molecular Probes) to identify mitochondria-rich chloride cells. Fluorescence

Figure 1. The killifish P2Y receptor (KF P2Y) is most closely related to the P2Y4 receptor of higher species. A 400 bp fragment of the putative KF P2Y receptor was obtained using a homology cloning approach and compared with known sequences of cloned P2Y receptors. Molecular analysis was performed using a Clustal algorithm (Lasergene DNAstar, Madison, WI). The KF P2Y gene is closely related to the turkey (t p2y4) and mouse (m p2y4) P2Y4 receptors. h: human, x: xenopus, s: skate.

increases in individual cells were monitored using an Olympus Fluoview confocal imaging system. Opercula were mounted on a specially designed chamber and perfused with a variety of 2+ nucleotides and/or inhibitors. Ca i increases were observed primarily in chloride cells and occurred in response to a variety of nucleotides in an agonist-specific fashion (UTP > UDP > 2+ ATP) typical of cloned P2Y4 receptors. Nucleotide-induced Ca i increases were inhibited by the nucleotidase apyrase and the P2Y receptor inhibitor suramin. Taken together, these experiments demonstrate that chloride cells in killifish opercular epithelium express a P2Y receptor that has pharmacologic properties similar to the P2Y4 receptor of higher organisms.

Nucleotide-induced changes in chloride secretion were assessed by short circuit current (Isc) measurements. Opercular epithelia were mounted in Ussing chambers and transepithelial potential differences were clamped to zero. Changes in basal Isc, transepithelial voltage (Vt) and resistance (Rt) were measured. The addition of isoproterenol (10 µM) to the basolateral membrane stimulated Cl- transport (Isc: 213.16 µAmp/cm2 ± 46.33 SEM, n = 10). Subsequent application of ATP (1 mM) to the apical surface had no appreciable effect, but addition of the same nucleotide to the basolateral surface significantly inhibited Cl- transport (182.14 µAmp/cm2 ± 41.04, p < 0.05, paired Student’s t-Test). The chloride channel inhibitor diphenylamine-2- carboxlic acid (DPC) similarly inhibited isoproterenol-induced increases in Isc (127.23 µAmp/cm2 ± 32.76, p < 0.05). Taken together, these results indicate that extracellular nucleotides exposed to the basolateral surface of opercular epithelium inhibit isoproterenol- induced chloride secretion.

These experiments indicate that killifish express a P2Y receptor related to the P2Y4 receptors of higher species. Moreover, these observations suggest that this receptor is expressed on the basolateral surface of chloride cells, and that ligand-mediated activation of the receptor affects chloride transport in killifish secretory epithelia. Work is in progress to clone the full-length killifish P2Y gene, and to decipher the intricacies of receptor-mediated inhibition of chloride secretion. Support was provided by MDIBL New Investigator Awards to J.E.M. and J.A.D.; REU awards (NSF DBI 9820400) to L.P. and K.Z.; the 2001 Silk Fellowship to K.Z.; a Cystic Fibrosis Foundation Student Traineeship Award to A.L.; and a Hancock County Scholars Award to M.C.

NA+/K+/2CL- COTRANSPORTER mRNA EXPRESSION IN THE BLUE CRAB CALLINECTES SAPIDUS MEASURED BY REAL-TIME QUANTITATIVE PCR

David W. Towle and Paul Peppin Mount Desert Island Biological Laboratory, Salsbury Cove, ME 04672

Euryhaline crabs osmoregulate in low salinities by taking up NaCl across the gill epithelium, driven by basolateral Na++K+-ATPase. Among the candidate apical transporters that may participate in this process is the Na+/K+/2Cl- cotransporter (NKCC), which has been amplified and sequenced from ion-transporting posterior gills of the blue crab Callinectes sapidus (Towle, Am. Zool. 38:114A, 1998; Accession Number AF190129). In the present study, we measured the abundance of NKCC mRNA in a variety of tissues and asked whether acclimation salinity might induce enhanced transcription of NKCC mRNA in posterior gills. Poly-A mRNA in 2 µg of total RNA was reverse transcribed using oligo-dT and SuperScript II reverse transcriptase in a final volume of 20 µl. Cotransporter cDNA in 1-µl aliquots was amplified in the presence of SYBR Green dye using Qiagen Quantitect chemistry and the Stratagene MX4000 Multiplex Quantitative PCR System. A dilution series demonstrated a linear relationship between threshold cycle (Ct) and log10 of template availability (Fig. 1).

Fig. 1. Standardization of

multiplex quantitative A B PCR system with a

template dilution series Dilution: 1 10 (A) showing linear

100 relationship between

Ct 1000 threshold cycle (C , shown t for undiluted template)

and log of template 10 abundance (B).

Analysis of NKCC mRNA abundance in total RNA extracts of tissues from crabs acclimated to 35 ppt salinity revealed the highest expression in antennal gland, with intermediate expression levels in gills and hepatopancreas (Fig. 2). Heart, hypodermis, and skeletal muscle showed negligible levels of NKCC mRNA. Transfer of crabs from 35 to 5 ppt salinity was accompanied by a significant increase in the abundance of NKCC mRNA in gills, particularly in posterior gills. Supported by the National Science Foundation (DBI-0100394).

Fig. 2. NKCC mRNA expression in tissues of Callinectes sapidus acclimated to 35 (all tissues) or 5 ppt (gills only shown) measured by real- time PCR and SYBR green binding, normalized to posterior gill at 5 ppt.

SKATE (RAJA ERINACEA)ANION EXCHANGER, sAE1, EXPRESSION IN XENOPUS LAEVIS OOCYTES.

Dana-Lynn Koomoa, Megan George, Leon Goldstein Brown University, Molecular Pharmacology, Physiology and Biotechnology Department, Providence, RI 02912, USA. Mount Desert Island Biological Laboratory, Salisbury Cove, Maine 04672

Volume regulation is a fundamental property of most cells. In the red blood cells of trout and skates (Raja erinacea), cell swelling activates the release of small organic compounds (e.g. taurine) out of the cell followed by osmotically obligated water, effectively restoring cell volume. The membrane protein AE1 (band 3) has been suggested to be involved in this regulatory volume response in skates erythrocytes (Perlman, D., L. Goldstein. J. of Exper. Zool 283: 725-733, 1999). Previous studies have shown that expression of trout anion exchanger (trAE1) in Xenopus laevis oocytes induces band 3 anion exchange activity and organic osmolyte channel activity (Fi vet et al. EMBO J. 14:5158- 5169,1995). The expression of skate AE1 in Xenopus laevis oocytes has been shown to induce anion exchange activity (H. Guizouarn, personal communication). The purpose of this study was to determine whether expression of skate AE1 in Xenopus oocytes induces organic osmolyte channel activity in these cells.

Total mRNA was isolated and purified from skate erythrocytes, reverse transcribed and the cDNA for skate AE1 was purified. The skate AE1 cDNA was cloned into pGEMTeasy, linearized with SacII and transcribed (Promega). The capped sAE1 cRNA was recovered, resuspended and an aliquot was analyzed by agarose-formaldehyde gel electrophoresis. The oocytes were removed from ice-anesthetized Xenopus laevis and defolliculated by collagenase treatment. Then, they were injected with 50 nL of sAE1 cRNA O (6.0 ng/oocyte) or water. The oocytes were maintained at 18 C in ND96 supplemented with penicillin (10 U/mL) and streptomycin (10 ug/mL). For taurine uptake experiments, 6-10 oocytes were incubated in 0.4 mL gluconate containing 3H-taurine with a specific activity of 40,000 cpm/nmol taurine. For inhibitor studies, 1.0 mM quinine, 0.1 mM DNDS, 0.1 mM niflumic, 0.1 mM NPPB acid or 0.1 mM DIDS were added to the incubation medium. Oocytes were washed in ice cold media, quickly transferred to scintillation vials and 50uL of 20% SDS were added. Liquid scintillation fluid was added to tubes. Tubes were vortexed then placed in liquid scintillation counter to determine the taurine uptake of each oocyte.

When skate AE1 was expressed in Xenopus oocytes, the transport of organic osmolytes (taurine) increased significantly (11.6±2.5 pmol 3H-taurine/oocyte/hr) compared to that of control oocytes (1.6±0.6pmol 3H-taurine/oocyte/hr). Pharmacological blockers (DIDS, DNDS, NPPB, niflumic acid and quinine), known to inhibit the volume activated organic osmolyte channel in skate erythrocytes, were tested on oocytes expressing sAE1. In oocytes injected with sAE1, DIDS, DNDS, NPPB, niflumic acid andquinine inhibited the organic osmolyte channel, decreasing taurine transport significantly, 91%, 67%, 79%, 92% and 89%, respectively . These results suggest that the expression of sAE1 in Xenopus oocytes induces organic osmolyte channel activity and that this channel has pharmacological sensitivity similar to that in the skate red blood cells. Supported by NSF grant IBN-9974350 (to L.G.). INHIBITION OF GLUTAMINE SYNTHETASE INCREASES BRANCHIAL AMMONIA EXCRETION BY THE DOGFISH, SQUALUS ACANTHIAS

Gregg A Kormanik Department of Biology, Univ. North Carolina at Asheville, NC 28804;

Urea synthesis by elasmobranchs uses glutamine synthesized from ammonia via glutamine synthetase (Anderson, P., Science 208:291-293, 1980). Ammonia excretion is low in Squalus embryos (Kormanik, G., J. Exp. Biol. 144:583-587, 1989) and adults, compared to that of seawater teleosts (Wood, C., P. Part and P. Wright, J. Exp. Biol. 198:1545-1558, 1995). We have shown that enzyme ratios favor glutamine synthesis in liver, brain, kidney and gill, and that the gill is the major site for ammonia excretion (Kormanik et al., Bull. MDIBL 37:99-100, 1998; Kormanik et al., Bull. MDIBL 39:102-103, 2000). Ammonia may be retained at excretory surfaces by conversion to the far less permeable glutamine molecule via glutamine synthetase. To determine if glutamine synthetase plays a role in ammonia retention, ammonia and urea excretion via gills and kidney were measured in Squalus acanthias before and after injection of L-methionine sulfoximide, an inhibitor of glutamine synthetase (Kormanik et al., Bull. MDIBL 39:102-103, 2000). Male dogfish (Squalus acanthias) were obtained and prepared as previously described (Kormanik et al., Bull. MDIBL 39:102-103, 2000). Fish (1-2 kg) were placed in covered, aerated 16.5 liter fiberglas boxes with running seawater (14 C.) and an air-lift recirculator, and allowed to recover for 24 hours. Flow to the box was stopped for a 2 hour control period, the box was rinsed (0.5 hrs) and L-methionine sulfoximide (72 mg/kg in 1-2 ml Elasmobranch Ringer's solution) was injected via an indwelling cannula. Flow was again stopped for 2 hour periods, with intermediate rinsing. Water samples representing branchial efflux were collected for ammonia and urea flux measurement and analyzed as previously described (Kormanik et al., Bull. MDIBL 39:102-103, 2000). Date were analyzed using a one-way ANOVA on ranks (Dunn's test). Data are expressed as mean ± SEM. Results of the experiments are shown in Table 1. Branchial excretion rates for ammonia increased significantly over time (p=0.030) while those for urea did not (p=0.325; urea rates not shown). Thus branchial glutamine synthetase may help prevent nitrogen loss at the gill by converting ammonia nitrogen to the less permeable nitrogen carrier glutamine, as well as provide nitrogen for urea synthesis.

+ Table 1. Ammonia (total: NH3 + NH4 ) excretion by the gills of Squalus acanthias before (control) and after injection of the GSase inhibitor, methionine sulfoximine (n =2- 8). Excretion is expressed in micromol kg-1 h-1. Post-injection rates are significantly different from the control period (p=0.030).

|------Post Injection Period (hour)------|

Preinjection 1 2 3.5 4.5 6 7 8.5 24 Control ______

Ammonia 56.7 45.0 103 112 129 224 229 277 213 ± 14 ± 13 ± 29 ± 34 ± 61 ±137 ±64 ±148 ±77

Supported by NSF IBN 9507456 and a UNCA Intramural Grant to GAK. MOLECULAR IDENTIFICATION OF Na+/H+ EXCHANGER cDNA IN THE GILLS OF THE EURYHALINE MUMMICHOG (FUNDULUS HETEROCLITUS). Susan L. Edwards, Alison Morrison-Shetlar, James B. Claiborne Department of Biology, Georgia Southern University, Statesboro, GA, 30460. The current models for branchial acid excretion in fishes include Na+/H+ exchange and the electrogenic excretion of H+ via an H+-ATPase. The predominant route of acid excretion in some freshwater fishes is widely thought to be via the H+-ATPase/Na+ channel system (Sullivan et al., Can. J. Zool. 74:2095-2103, 1996; Perry et al., J. Exp. Biol. 203:459-470, 2000). In contrast, the hyperosmotic environment of seawater provides a situation in which proton efflux, powered by passive inward Na+ exchange diffusion, could be more favorable than the proton pump model of freshwater animals that requires direct ATP hydrolysis. The current marine model for H+ excretion is centered upon Na+/H+ exchange across the apical membrane of the branchial epithelium (Claiborne, In: The Physiology of Fishes. 2nd Edition, ed.: D.H. Evans, CRC Press, 179-200, 1998). We hypothesize that in marine and some brackish water fishes the predominant mechanism of acid excretion is via a Na+/H+ exchanger (NHE). This study utilized RT-PCR techniques to obtain the full-length sequence of NHE cDNA in the gills of the mummichog.

Total RNA from the gills of SW adapted mummichogs (Fundulus heteroclitus) was isolated using the Tri-Reagent methodology (Sigma, St Louis). First strand cDNA was synthesized from gill total RNA with either oligo-dT or GeneRacerª Oligo-dT using Superscriptª II RNase H-reverse transcriptase according to the manufacturer's protocol (Invitrogen). Degenerate oligonucleotide primers based on conserved regions of several vertebrate Na+/H+ exchanger sequences were used to amplify a discrete 750 bp product using mummichog gill cDNA as the template. The initial fragment showed a 60% amino acid homology to the rat NHE2. To isolate the entire mummichog gill NHE cDNA, a 5′ and 3′ RACE protocol (Generacerkit, Invitrogen) was employed. Primers were designed from the initial 750 bp fragment. In the 5′ direction, the cDNA sequence was extended about 850 bp in several independent clones. In the 3′ direction, a 1500 bp cDNA sequence was extended to a poly (A)+ tail also in a number of independent clones. Both the 3′ and 5′ RACE generated fragments overlapped with the original 750 bp fragment. Together, these clones generated a composite cDNA of 2481 nucleotides. The cDNA contained an open reading frame encoding a polypeptide of 717 amino acid residues. In a pairwise comparison the F. heteroclitus gill NHE showed a 58-62% amino acid homology to other known mammalian NHE2 amino acid sequences. A relationship analysis based upon multiple alignment showed that the mummichog NHE was more closely related to the rat NHE2 and NHE4 than other epithelial NHE isoforms (Fig. 1). This research was funded by NSF grants IBN-9808141 & IBN-0111073 to JBC & AIMS.

Figure 1. Relationship tree of NHE isoforms using the derived gill NHE amino acid sequence from the mummichog and other known NHE sequences, GenBank accession numbers: M85299 NHE1 (Rattus norvegicus), L11004 NHE2 (R. norvegicus), M85300 NHE3 (R. norvegicus), M85301 NHE4 (R. norvegicus) & U09274 crab NHE (Carcinus maenas). Bootstrap values are indicated at the nodes. RELATIVE EXPRESSION OF mRNA FOR NHE-2 IN THE GILLS OF THE LONG- HORNED SCULPIN, MYOXOCEPHALUS OCTODECIMSPINOSUS

Nichole L. Hair, Susan L. Edwards, Alison Morrison-Shetlar, and James B. Claiborne Department of Biology, Georgia Southern University, Statesboro, GA 30460

Acid-base homeostasis in fish is accomplished by the transfer of acid-base relevant ions between the aquatic organism and the water (Claiborne, J.B., In: The Physiology of Fishes. 2nd Edition, ed.: D.H. Evans, CRC Press, 179-200, 1998). In marine fish, one of the gill transport proteins thought to drive the excretion of H+ to the water is the Na+/H+ exchanger (NHE) (Yun, C., et al., Am. J. Physiol. 269: G1-11, 1995). We recently described the first full-length NHE-2 like sequence in fish (the long-horned sculpin, Myoxocephalus octodecimspinosus; Gunning, D., et al., Bull. Mt. Desert Is. Biol. Lab. 40: 71-72, 2001). In order to investigate the functional role of NHE-2 in acid-base balance in the sculpin, we have used relative semi-quantitative RT-PCR to measure changes in the expression of mRNA in the gill tissue following an internal acidosis. Concurrent in vivo measurements of net H+ transfers were also made.

Long-horned sculpin (100-300 g) were cannulated and placed in darkened acrylic boxes (1.8-2.1 liter) with running seawater and allowed to recover for 8-12 hours. Following this recovery period, the chamber was closed to running seawater, aerated, and a control H+ flux of 8-12 hours was measured. Subsequent to the pre-infusion period, fish were infused with either 0.1N HCl (2 mmol kg-1) or a comparable volume of distilled water. Fish were sacrificed at h 1, 2, 5, or 8 and a 20 ml sample was collected from the seawater at each of these intervals so that cumulative transfers of H+ between the fish and the water could be calculated as described by Claiborne et al. (J. Exp. Biol. 193: 79-95, 1994). Ambion’s QuantumRNAä Classic 18S Internal Standard kit was used in relative semi-quantitative RT-PCR to determine expression levels of NHE-2 gill mRNA. Specific primers for NHE-2 gill mRNA were developed in the 3' region of the open reading frame (Gunning, D., et al., Bull. Mt. Desert Is. Biol. Lab. 40: 71-72, 2001) to yield an 858 bp product. The ratio of this product to a 488 bp 18S internal standard was then determined. PCR reactions for each animal were run in triplicate and products visualized on ethidium agarose gels. The apparent density of the PCR product bands were digitized for quantitative comparison. The densitometric measurements were used to calculate the mean ratio for all lanes with detectable internal standard expression (NIH Image software).

As we have shown previously (Claiborne, J.B., et al., J. Exp. Biol. 193: 79-95, 1994), sculpin infused with 0.1 N HCl showed a higher net H+ excretion rate (difference calculated between post-infusion and pre-infusion rates) than those fish infused with distilled water (Table 1). Post-infusion excretion rates were significantly higher than their pre-infusion controls in hour 1, 2, and 8 for acid-infused fish, while control fish were only higher than pre-infusion values at hour 1 (data not shown). Gill mRNA expression of the acid-loaded fish was found to be significantly higher in the first two hours following acid infusion when compared to control fish (Table 2). The mean pixel ratio at hour two was some 2.5x higher in the acidotic animals. At hour 5, mRNA expression in these animals remained high and only decreased significantly (p<0.01) in the fish sampled at hour 8. Band densities in the control animals were variable but did not change significantly from each other throughout the experiment. Table 1: Net excretion of H+ for water- and acid-infused sculpin. Each time period represents a separate group of experimental animals. Means + S.E (N). * denotes significance between the respective pre-infusion and post-infusion period for each group (p<0.05, one-tailed Student’s paired t-test).

Hour Control Acid-loaded (mmole kg-1 hr-1) (mmole kg-1 hr-1) 1 1.07 + 0.15 (4)* 1.66 + 0.70 (5)* 2 0.21 + 0.61 (4) 1.23 + 0.45 (5)* 5 0.37 + 0.14 (3) 0.77 + 0.37 (4) 8 0.18 + 0.16 (3) 0.21 + 0.17 (4)*

Table 2: Mean band density ratio of NHE-2 PCR product and 18S control for control and acid- infused sculpin. A higher ratio indicates increased expression of NHE-2 product. Means + S.E (N). * denotes significance increase in experimental group over that of comparable control group (p<0.05, one-tailed unpaired test). † indicates mean ratio significantly different from animals sampled at other time periods within the same treatment.

Hour Control Acid-loaded (band density ratio) (band density ratio) 1 0.39 + 0.11 (4) 0.63 + 0.07 (5)* 2 0.22 + 0.03 (4) 0.55 + 0.10 (5)* 5 0.40 + 0.12 (3) 0.56 + 0.11 (4) 8 0.18 + 0.04 (3) 0.13 + 0.01 (4)†

Post-infusion net H+ excretion increased rapidly in sculpin and 83% of the acid load had been lost to the water in the first hour. As observed previously by Claiborne et al (J. Exp. Biol. 193: 79-95, 1994), sculpin "overexcreted" a total of 3.87 mmol kg-1 by hour 8, nearly twice the infused load. Interestingly, water infusion also induced a significant proton loss (perhaps a volume loading or stress response) which returned to pre-infusion levels by hour two. The pattern of excretion following acidosis was mirrored by the measured expression of gill NHE-2 transcripts, as expression in acid loaded animals were higher than controls in the first two hours and decreased only after 8 hours. These data imply that increased mRNA transcription (and presumably the resulting protein translation) plays a role in the observed net H+ excretion across the gills. The time course was similar to (though perhaps longer lived than) mRNA expression of NHE measured in the gills of the hagfish following an infused acid load (Edwards, S.L., et al., Comp. Biochem & Physiol. 130: 81-91, 2001). Western analysis using heterologous NHE antibodies have demonstrated that the expression of NHE isoforms can increase after one hour of hypercapnic acidosis in Fundulus heteroclitus (Edwards, S.L., et al., FASEB Journal (Abstracts) 15: supplement, pg. 11, LB43, 2001), but it remains to be seen if NHE-2 protein expression in sculpin will follow the time course observed for mRNA transcription. Sculpin specific antibodies are currently being developed to test this. This research was supported by NSF IBN-9808141 and IBN-0111073 to J.B.C. and A.I.M.S. EFFECT OF SALINITY ALTERATIONS AND RESPIRATORY ACIDOSIS ON ACID-BASE TRANSFERS IN THE MUMMICHOG (FUNDULUS HETEROCLITUS)

Jill C. Weakley1, Samuel Sligh2, Alison I. Morrison-Shetlar1 and James B. Claiborne1 1Department of Biology, Georgia Southern University, Statesboro, GA 30460 2Hillside High School, Durham, NC

An elevation in the levels of ambient CO2 rapidly induces respiratory acidosis in fish (Claiborne, J.B. In: The Physiology of Fishes. 2nd Edition, ed.: D.H. Evans, CRC Press, 179-200, 1998). While nonbicarbonate buffering plays a role in the restoration of plasma pH, gill transepithelial exchange of acid-base relevant molecules has been shown to be the major mechanism during compensation in most species (Claiborne, J.B., et al., J. Exp. Zool. 279: 509-520, + + - - 1997). These transfers involve Na /H and Cl /HCO3 exchanges with the surrounding media. In freshwater adapted fish, H+ excretion (thought to be via H+-ATPase; Perry, S.F., et al., J. Exp. Biol. 203: 459-470, 2000) linked to Na+ uptake would be optimal for both ion uptake as well as compensation of acid-base stress, but Na+/H+ exchange would cause an increase in the ionic and osmotic stresses already faced by fish adapted to seawater. The mummichog (Fundulus heteroclitus) is a euryhaline species which inhabits salinities ranging from freshwater to full- strength seawater. Little is known about the gill mechanisms of acid-base balance in this species, but the gill regulatory strategies utilized by freshwater adapted mummichogs appear to be very different from other freshwater fishes to date (Patrick, M.L. and Wood, C.M., Comp. Biochem. Physiol. 122: 445-456, 1999). We have recently shown that proteins immunologically similar to mammalian NHE-3 are expressed at higher levels in seawater adapted mummichogs exposed to external hypercapnia (Wall, B., et al., Bull. Mt. Desert Is. Biol. Lab. 40: 58-59, 2001) and the expression may be dependant on the salinity to which the animals are adapted (Edwards, S.L., et al., FASEB Journal (Abstracts) 15: supplement, pg. 11, LB43, 2001). The aim of this study was to collect baseline data towards understanding the net H+ transport across the gills in this species adapted to various salinities and subjected to a respiratory acidosis.

Fundulus heteroclitus (3.86 ± 0.36 g; mean ± S.E., n=31) were caught in minnow traps in Northeast Creek, on Mount Desert Island, and acclimated in 10 gallon tanks for seven to 10 days. Fish were placed in one of three acclimation salinities: freshwater (FW: dechlorinated tap water + 1% seawater; ~10 mOsm), isoosmotic "brackish" water (BW: seawater diluted with dechlorinated tap water; ~325mOsm), or seawater (SW: ~930 mOsm). Aeration was accomplished from bubbling and filtering of the tanks, except the seawater tank which received running seawater. Following acclimation, animals were weighed, placed in individual plastic experimental chambers (total volume 150 mls) and a 13 hr control period with normal aeration was begun. Water samples (8 ml) were collected at the start and end of the control period. The water was then flushed with a fresh 150 ml volume of the same salinity and the air bubbled into the chamber was either maintained (normocapnia series) or modified to a 1% CO2 in air mixture (hypercapnia series) from a GF-2 Gas Mixing Flowmeter (Cameron Instruments). In most cases, gases were bubbled into the fish chamber through small plastic pipette tips to avoid potential leaching of buffers from air stones and the confounding effect on water titration analysis which this caused (unpublished data). Over the 23 hour experimental period (water was flushed at hour 8 to prevent high ammonia levels (Claiborne, J.B., et. al., J. Exp. Biol. 193: 79-95, 1994) the media was periodically sampled (hr 0, 1, 2, 4, 8, 23). Volumetric titration of 2 ml portions of the water samples to a pH of 4.00 (FW samples: 0.025N HCl; BW and SW samples: 0.1N HCl), analysis for total ammonia, and the calculation of net H+ transfers between the fish and the water (∆H+ in mmol kg-1 h-1) were calculated as described by Claiborne et al. (op. cit.).

As shown in Table 1., a significant increase in ∆H+ excretion was measured in hypercapnic fish adapted to both FW and BW (measured over the first 8 hours of the experiment) when compared to control fish in the same salinities. High variability in the control SW fish clouded the resolution of the SW group, but paired analysis of the pre-hypercapnic control period (-0.31 ± 0.29 mmol kg-1 h-1) with the post-hypercapnic rates (0.67 ± 0.28 mmol kg-1 h-1), indicated a significant increase in ∆H+ excretion was induced by hypercapnia in this group as well (paired t-test, one- tailed; p<0.05).

Table 1. ∆H+ transfer rates between the fish and the water (over 8 hours) in Fundulus heteroclitus adapted to freshwater (FW), isoosmotic water (BW) and seawater (SW). The Net ∆H+ is calculated as the mean difference between control and hypercapnic ∆H+ rates for each group and represents the net increase in acid transfers from the animals during hypercapnia. Mean ± S.E. (N). One-tailed, unpaired t-test between groups. * indicates p < 0.05.

Normocapnia Hypercapnia Net ∆H+ (mmol kg-1 h-1) (mmol kg-1 h-1) (mmol kg-1 h-1) FW -0.47 ± 0.26 (5) 0.22 ± 0.08 (5) 0.70 * BW 0.12 ± 0.07 (5) 0.52 ± 0.05 (5) 0.40 * SW 0.17 ± 0.63 (6) 0.67 ± 0.28 (5) 0.50 (ns)

As has been demonstrated in several marine and freshwater species, hypercapnia induces a net increase in H+ efflux in Fundulus adapted to all three salinities. This allows compensation of internal acidosis during the hypercapnic exposure (see review by Claiborne, op. cit.). Because of the ∆ + - ∆ + control H uptake (or HCO3 loss) in the FW animals, the net H rate induced by hypercapnia was the highest in this group. At the same time, the magnitude of ∆H+ transfers was the lowest in the FW while the BW and SW rates were similar (both increasing by ~4x over control), perhaps indicating the impact of external counter ion availability on the absolute rate of transport -1 - (Claiborne, op. cit.). Interestingly, the normocapnic FW fish lost ~9.15 mmol kg of HCO3 over - the 23 hour period. We also noted high rates of net HCO3 loss in long-horned sculpin (Myoxocephalus octodecimspinosus) and the gulf toadfish (Opsanus tau) following transfer to low salinities (Claiborne, et al., op. cit.; Claiborne, J.B., et al., J. Fish Biol. 56: 1539-1544, 2000), though the present values are more than double that of the euryhaline toadfish. Similar net H+ uptake rates have also previously been reported in another study on this species (Patrick and Wood, op. cit.). It is unclear as to the root cause of the acid uptake, but the imbalance does imply that the mummichog is not in acid-base homeostasis when exposed to the most dilute salinity used in this study.

Immunological detection of Na+/H+ exchangers in the gills of this species has shown that NHE expression increases within one hour of 1% hypercapnic exposure (Wall et. al., op. cit.). Clearly, the mummichog is able to dramatically increase H+ transfer to the water following hypercapnia. We speculate that NHE driven transfers are responsible for the acid-base compensation observed in vivo. This research supported by NSF IBN-9808141 and IBN-0111073 to J.B.C. and A.I.M.S, NSF REU DBI-9820400 to JCW and a Hancock County Fellowship to SS. CLONING OF sAE3 ANION EXCHANGER OF SKATE (RAJA ERINACEA) ERYTHROCYTE

Hélène Guizouarn1, Daniela Myers2 and Leon Goldstein2 1 Laboratoire Jean Maetz, Physiologie des Membranes Cellulaires, 284 chemin du Lazaret, 06230 Villefranche/Mer, France. 2Division of Biology and Medecine, Brown University, Providence, RI 02912, USA.

The anion exchangers family consists of integral membrane proteins of ubiquitous - - occurence catalyzing exchange of Cl and HCO3 . Up to now, four different genes are known˚: AE1, AE2, AE3 and AE4 coding for multiple forms of anion exchangers that are specifically expressed in different tissues. Associated with their ubiquitous occurence, anion exchangers are involved in many different processes such as cellular and organismal aging, neurological diseases and structural integrity of the cell. In fish erythrocytes the anion exchanger is proposed to mediate the volume-sensitive loss of taurine and other small organic compounds. Cloning of tAE1, the anion exchanger of trout erythrocytes, and expression of this protein in Xenopus oocytes led to the conclusion that it is able to form an anion channel permeable to taurine as well as to some organic solutes (choline, urea and sorbitol) and inorganic cations (Na+ and K+). These permeabilities are activated in trout red cell in response to a decrease in intracellular ionic strength (Fi vet et al. EMBO J. 14˚:5158-5169,˚1995; Guizouarn et al. J. Physiol.˚535˚:2, 497-506, 2001). Volume regulation of hyposmotically swollen skate red cells induces activation of a taurine permeability that was also shown to involve the anion exchanger (Musch et al˚; J. Biol. Chem. 274˚: 7923-7928, 1999). To investigate the role of skate AE in swelling-induced taurine pathway, AE cloning in skate red cells was initiated two years ago (MDIBL Bulletin, 38,p.23,2000). Surprisingly, we found three different isoforms of AEs expressed in skate erythrocytes: AE1, AE2˚, and AE3. The present paper reports the cloning of sAE3 isoform.

The membrane spanning domains of sAE3 were amplified by PCR using cDNA derived from skate erythrocytes. RACE PCR (Rapid amplification of cDNA ends by PCR) done on cDNA of skate erythrocytes failed to give specific products. Thus it was decided to clone the 5 and 3 ends of sAE3 in a cDNA library of skate brain hypothezing that this tissue expresses predominantly the AE3 isoform. Skate brain poly-A RNA purified by affinity chromatography (Qiagen kit, Valencia CA 91355 ) were retrotranscribed in cDNA using the Marathon amplification kit (Clontech, Palo Alto CA 94303) procedure. 3 RACE PCR gave a main product of about 600 bp coding for the C terminal part of sAE3, cloned in pGEMT-easy vector (Promega, Madisson WI 53711-5399).To get the N-terminal part of sAE3, four different 5 RACE PCR runs were needed, none of the 5 RACE PCR done being able to amplify directly in one piece the 2100 coding bp corresponding to the N-terminal cytoplasmic domain of AE3. sAE3 isoform is about the same size as sAE2 with the same organization˚: a large N-terminal cytoplasmic domain (about 700 aa) and a membrane spanning domain of about 500 aa with a long connecting loop between spans 5 and 6. It differs from sAE1 which has a shorter N-terminal cytoplasmic domain (400 aa) and a spanning domain devoid of the long connecting loop between spans 5 and 6. Future experiments are aimed at determining the physiological functions of the three isoforms. Supported by NSF grant IBN 9974350. GUANYLIN/GUANYLATE CYCLASE SIGNALING IN THE INTESTINE OF DOGFISH SHARK (SQUALUS ACANTHIAS) AND AMERICAN EEL (ANGUILLA ROSTRATA)

Karl J. Karnaky, Jr.1, Emily Milner2, John N. Forrest, Jr.3, and Leonard R. Forte4 1Department of Cell Biology and Anatomy and the Marine Biomedical and Environmental Sciences Program, Medical University of South Carolina, Charleston, SC 29425 3Department of Medicine, Yale University New Haven, CT, 06510 2Presbyterian College, Clinton, SC, 29325 4Truman VA Hospital and Dept of Pharmacology, Univ. of Missouri, Columbia, MO 65212

Guanylin peptides activate membrane receptor-guanylate cyclase (R-GC) and regulate the intestinal secretion of Cl-, bicarbonate and fluid in mammals via the cGMP signaling pathway. Guanylin-like peptides and/or R-GC-receptors have been identified in all vertebrate classes including freshwater fish (Forte, Reg. Peptides. 81:25-39, 1999). Existence of this regulatory pathway in rectal glands of the dogfish shark is uncertain. Boluses of guanylin or E. coli heat-stable enterotoxin (ST) modestly stimulated Cl- secretion in perfused rectal glands (Silva et al., Bull. MDIBL 36:53-54, 1997; Silva et al., Bull. MDIBL 37:73, 1998). In contrast, guanylin does not stimulate Cl- secretion in perfused tubules isolated from rectal glands (Greger et al., Pflugers Arch 438:15-22, 1999). Treatment of primary cultures of rectal gland epithelial cells with ST elicits small increases in cGMP levels, whereas atriopeptin stimulates 40-fold increases in cGMP levels (Kennedy et al., Bull. MDIBL 30,102-103, 1991). In addition, guanylin and ST did not stimulate Cl- secretion in cultured rectal gland epithelial cells raised on collagen-coated filters (Kennedy et al., op. cit.; Karnaky et al., Bull. MDIBL 38, 70-71, 1999; Forrest et al., Bull. MDIBL 41, 122-124, 2001). We also found that 1 µM ST does not stimulate cGMP production in rectal gland slices in vitro (Karnaky et al., Bull. MDIBL 38, 70-71, 1999). Recently we showed that treatment of isolated glands with 1µM uroguanylin, guanylin or ST by continuous perfusion did not stimulate Cl- secretion, whereas forskolin markedly stimulates Cl- secretion (Forrest et al., ob cit). In contrast, we discovered that treatment of mucosa isolated from dogfish shark proximal intestine with E. coli ST elicits significant increases in tissue cGMP (Karnaky et al., Bull. MDIBL 41, 51-52, 2001), suggesting that a guanylin-cGMP pathway is present. Moreover, cDNAs encoding eel uroguanylin, zebrafish guanylin (Comrie et al., Biochem. Biophys. Res. Comm. 281,1078-1085, 2001) and both eel and medaka forms of R-GCs have been identified (Mantoku et al., J. Biochem. 125, 476-486, 1999; Comrie et al., Comp. Biochem. Physiol. B 129:575-586, 2001). We summarize herein our recent findings of the biological effects of uroguanylin and ST to stimulate cGMP production in the intestinal mucosa from S. acanthias and A. rostrata and also report the primary structure of uroguanylin expressed in the American eel, A. rostrata.

Intestines were obtained from dogfish sharks, S. acanthias, and eels, A. rostrata, from standard suppliers at the MDIBL. Mucosae from the first 2 cm of intestine were isolated and washed 3 times in either shark Ringer’s or teleost Ringer’s solutions and then resuspended in 1-2 ml of the same buffers containing 1 mM isobutylmethylxanthine to inhibit cGMP hydrolysis. About 50-100 mg of mucosa was incubated for 40 minutes at room temperature (18 oC) in 0.2 ml of buffers containing either vehicle or human uroguanylin and E. coli ST peptides. Tissue cGMP was measured by radioimmunoassay and expressed as pmol cGMP per mg protein.

A cDNA encoding preprouroguanylin was synthesized using intestinal RNA isolated from A. rostrata with a homology cloning technique that uses degenerate oligonucleotide primers to amplify uroguanylin cDNAs by RT-PCR. The cDNAs were cloned, sequenced and the amino acid sequence of prouroguanylin deduced. The primary structure of uroguanylin active peptides from A. rostrata and A. anguilla are identical, but substantially different than the zebrafish guanylin and the mammalian forms of guanylin and uroguanylin (Fig 1).

10 µM uroguanylin or 1 µM E. coli heat stable enterotoxin (ST) elicits ~10-fold increases in cGMP accumulation in the intestinal mucosa from dogfish, with maximal stimulation of cGMP levels occurring by 40 min (data not shown). The EC50 for stimulation of cGMP was ~1- 3 nM for ST and ~100 nM for uroguanylin (Fig. 2A). Likewise, treatment with 1 µM ST elicited ~16 fold increases in cGMP in the intestinal mucosa from eels (Fig. 2B). A guanylin-like peptide isolated from shark intestine stimulates cGMP accumulation in T84 human intestinal cells, thus guanylin/uroguanylin from S. acanthias also activates a human form of R-GC (data not shown).

We conclude that the guanylin-cGMP signaling pathway is present in the intestine of both elasmobranchs and teleosts. Whether both guanylin and uroguanylin genes are present in teleosts and cartilaginous fish is a fundamental question. A guanylin peptide has been identified in zebrafish and a uroguanylin peptide has now been elucidated in two different eel species. We plan to continue our efforts to clone cDNAs encoding guanylin-like peptides from eel and dogfish intestine to address the fundamental question of whether both guanylin and uroguanylin genes occur in both elasmobranchs and teleosts. A physiological role for the guanylin-cGMP signaling pathway in teleosts may involve osmoregulation. Comrie et al., (ob cit) reported that intestinal expression of uroguanylin mRNAs are up-regulated in both non-migratory “yellow” eels and the more sexually mature, migratory “silver” eels following acclimation to a seawater environment. Cloning the cDNAs encoding uroguanylin provides a useful probe to explore the expression of uroguanylin in the intestine of A. rostrata in relation to adaptation to fresh versus seawater conditions in further studies at the MDIBL. We also plan to continue investigating the actions of guanylin peptides on cGMP production and ion transport in the eel intestine.

GUANYLIN Opossum S H T C E I C A F A A C A G C Human P G T C E I C A Y A A C T G C Zebrafish V D V C E I C A F A A C T G C

UROGUANYLIN Opossum Q E D C E L C I N V A C T G C Human N D D C E L C V N V A C T G C L Eel (A. anguilla) P D P C E I C A N A A C T G C L Eel (A. rostrata) P D P C E I C A N A A C T G C L

E. coli ST ST-human strain N Y C C E L C C N P A C A G C Y

Figure 1. Primary structures of guanylin and uroguanylin peptides.

Figure 2. Intestinal cGMP responses to uroguanylin and/or E. coli ST in dogfish and eel. Data for the dogfish (A) are expressed as the mean ± SEM of triplicate assays from one intestinal mucosa preparation and the data for eel (B) represents the mean ± SEM (n=5 replicates) for cGMP accumulation in mucosal preparations from two different animals. E.coli ST concentration in the experiment with eel intestines is 1 µM. Error bars are small for controls in (B).

Supported by MDIBL New Investigator Awards, NIH grant DK 34208 and NIEHS grant P30-ES3828. EM was a recipient of a Research Experience for Undergraduates Fellowship from the NSF (DBI-9820400). Publication Number 197 of the Grice Marine Biological Laboratory of the University of Charleston. LOCALIZATION OF GFP-KFCFTR TO THE PLASMA MEMBRANE OF MAMMALIAN EPITHELIAL CELLS

John E. Mickle, Carrie Blaschak, Catherine Yurk and Garry R. Cutting

The McKusick-Nathans Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Baltimore, MD 21287.

Cystic fibrosis is a life-limiting lung disease that affects approximately 30,000 people in the United States. The disease is caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR). The most common mutation, ∆F508, alters folding of the native protein such that CFTR no longer reaches its targeted location, the apical membrane, where it would otherwise mediate chloride secretion across airway epithelia (Mickle et al. Am J Hum Genet 66:1485-1495, 2000). Not surprisingly, many CF gene therapies aim to shuttle the ∆F508 mutant protein to the apical membrane so that function can be restored. However, the mechanisms that facilitate CFTR trafficking to this location are not clear. To elucidate these mechanisms we employed a comparative approach. CFTR has been identified in numerous species, the most divergent form is from killifish, Fundulus heteroclitus. Since killifish (kf)CFTR also localizes to the apical membrane of native epithelia (Mickle, et al., MDIBL Bulletin 39:75-76, 2000; Lankowski et al. Submitted), we hypothesized that the trafficking mechanism is evolutionarily conserved. Accordingly, we expressed kfCFTR in polarized mammalian epithelial cells and assessed protein localization.

To distinguish kfCFTR expression from any trace of endogenous CFTR upon heterologous expression in mammalian epithelial cells, the cDNA of green fluorescent protein (GFP) (provided by N. Muzyczka, Zolotukhin et al., J Virol 70:4646-4654, 1996) was conjoined to full- length kfCFTR cDNA (provided by T. Singer, Singer et al., Am J Physiol 43:C715-C723, 1998). A successive subcloning strategy was utilized because suitable restriction sites were unavailable to readily subclone these two cDNAs in tandem. A PCR product that fused GFP cDNA to the first 21 nucleotides of kfCFTR cDNA was generated using oligonucleotide primers that incorporated an EcoRI and AgeI restriction site onto the 5’ and 3’ termini, respectively. The product was then ligated into the TA cloning vector pCR2.1 (Invitrogen, Carlsbad, CA), and sequenced to verify fidelity. Once affirmed, the PCR insert was excised with EcoRI and AgeI, size separated from vector DNA by gel electrophoresis, extracted and column purified for ligation to kfCFTR cDNA. Similar manipulation of the plasmid containing kfCFTR cDNA yielded two cDNA products: a 1.2 Kb fragment generated by AgeI digestion and a 4.3 Kb segment cut by AgeI on the 5’ terminus and EcoRI on the 3’ end. To foster ligation efficiency, a three-way ligation reaction was performed with the excised PCR insert, the 4.3 Kb fragment and the expression vector pcDNA3.1(-) (Invitrogen), which had been linearized with EcoRI and dephosphorylated to promote intermolecular assembly. Subsequent transformation and growth in DH5α subcloning efficiency cells (Invitrogen), plasmid DNA extraction, and restriction enzyme analyses confirmed proper construction. To complete the process, the three-way construct was linearized with AgeI, dephosphorylated, ligated to the 1.2 Kb fragment, and transformed as previously described. Restriction analyses and DNA sequencing established that the final product was assembled properly.

To determine whether GFP-kfCFTR trafficks to the apical membrane of mammalian epithelial cells, the full-length GFP-kfCFTR cDNA construct was transiently expressed in MDCK II cells using standard transfection protocols (Lipofectamine Plus Reagent, Invitrogen). o Cells were cultured for three days at 37 C and 5% CO2 with supplemented media (DMEM + 10% FBS + 0.375% NaHCO3, Invitrogen); then, fixed, permeabilized, and stained with dyes for F-actin (rhodamine phalloidin, Molecular Probes, Eugene, OR) and nuclei (DAPI, Sigma, St. Louis, MO). Fluorescent images were captured with a Zeiss LSM 410 confocal microscope. GFP alone is expressed throughout the cytoplasm, but GFP-kfCFTR colocalizes with F-actin to the plasma membrane (Figure 1), particularly at the apical and lateral surfaces. In contrast, GFP- human CFTR localization is primarily apical in MDCK cells (Moyer, et al., J Clin Invest 104:1353-1361, 1999). This suggests that kfCFTR can interact sufficiently with mammalian orthologs for membrane localization, but that specific apical localization is precluded by differences in kfCFTR sequence. These differences are currently being exploited to identify key factors involved in apical targeting. To confirm that these initial observations reflect trafficking in a polarized system, the transepithelial resistance is also being measured and other membrane markers are also being used to localize kfCFTR in MDCK II cells. Support was provided by a MDIBL New Investigator Award to J.E.M.

A B C D

10µM 10µM 10µM 10µM

Apical Apical Apical Apical G E F H

10µM Basolateral 10µM Basolateral 10µM Basolateral 10µM Basolateral

Figure 1. Heterologous expression of killifish CFTR in mammalian epithelial cells. Confocal imaging reveals that F-actin is concentrated at the plasma membrane (A) while GFP (B) is expressed throughout the cytoplasm in the same MDCK II cell. In contrast, F-actin (C) and GFP-kfCFTR (D) co-localize at the plasma membrane. Cross- sectional reconstructions of serial scans through the cells presented in panels A-D show nuclei (E) and cytoplasmic GFP expression (F). Likewise, nuclei are shown for comparison (G), but expression of GFP-kfCFTR (H) localizes to the apical and lateral membrane of MDCKII cells

PARTIAL SEQUENCE OF THE SODIUM HYDROGEN EXCHANGER FROM WINTER FLOUNDER RED BLOOD CELLS.

Scott A. King, Stine F. Pedersen, Robert R. Rigor, and Peter M. Cala Dept. of Human Physiology, School of Medicine, UCD, One Shields Ave., Davis, CA 95616

Regulatory Volume Increase (RVI) in response to osmotic perturbation is a well- documented phenomenon in many different species and cell types. One of the first descriptions of this response was in the winter flounder (Pseudopleuronectes Americanus) red blood cells (RBCs) at the Mount Desert Island Biological Laboratory (Cala, Bull. MDIBL 13:20-25, 1973). In response to hypertonic stress, the winter flounder RBCs were shown to exhibit regulatory volume increase (RVI) following shrinkage by gaining Na, Cl and osmotically obligated water, returning the cell to normal volume (Cala, J Gen. Physiol., 69:537-52 1977). Several membrane proteins have been linked to RVI in different cell types and species. These include the type 1 Na/H exchange (NHE1) and the Na-K-2Cl co-transporter (Cala, In: Chloride Channels and Carries, In Nerve, Muscle and Glial Cells, New York: Plenum Press, 67-81, 1990). Since NHE1 has been detected in other teleost fish RBCs (Borgese, Proc Natl Acad Sci 89(15): 6765-9, 1992), we revisited the winter flounder RBC studies to determine if NHE1 activity is responsible for RVI. Last year we demonstrated that the RBC Na influx is accompanied by obligatory proton efflux consistent with NHE1mediated RVI (Rigor, Bull. MDIBL 40:68-70, 2001). Expression of NHE1 in winter flounder RBCs was confirmed using antibodies specific to NHE1 (Rigor, Bull. MDIBL 40:68-70, 2001). Western blots stained with antibodies to NHE1, but not NHE3, detected a protein running at approximately 110 kD on SDS-PAGE, comparable to the molecular weight of NHE1 from other species. In this study, we also demonstrated that although the transport properties of the winter flounder NHE1 are similarly to NHE1 from other species, the protein is somewhat unique in that it is unaffected by the common NHE1 inhibitors, amiloride, EIPA and HOE694 (Rigor, Bull. MDIBL 40:68-70, 2001). Our lab has previously shown that the Amphiuma NHE1 also differs from human in its sensitivity to the inhibitor HOE694 (McLean, Am. J. Physiol. 276: C1025-37, 1999). To investigate the molecular basis for species-specific NHE1 inhibitor sensitivity, we have begun sequencing the winter flounder NHE1. Since NHE1 is fairly homologous across species, a comparison of the winter flounder sequence to the human and Amphiuma may identify the regions necessary for inhibitor function.

In these experiments, winter flounder blood was taken from the caudal vein by puncture with a heparin containing syringe. RBCs were isolated by centrifugation and washed twice with isotonic saline. Messenger RNA (mRNA) was purified from the RBCs using the PolyTract A system (Promega, Madison, WI). Briefly, cells were lysed in a strong denaturing solution to inhibit RNase activity. Biotinylated oligo dT primers were added to the lysate and annealed to the mRNA. The mRNA complex was then purified using streptavidin coupled magnetic beads. DNA was synthesized from the mRNA using avian myeloblastosis virus reverse transcriptase (Promega) and degenerate primers based on homologous regions of the trout, amphiuma and human NHE1 (forward, ttacagaattcgtgcagggcatgacaattcggc and reverse, cttctcgaggtgtggcccggggtcacttaggc) were used to amplify the C-terminal region of the winter flounder NHE1. Using this approach, we isolated an approximately 1-kb DNA segment from winter flounder RBCs that corresponds to the carboxy terminal tail of NHE1. We began sequencing from this region since it is fairly homologous between species. We have, so far, sequenced 788 nucleotides of the winter flounder NHE1 carboxyl terminus (Fig. 1), corresponding to amino acids 511 to 790 of the Amphiuma NHE1.

Figure 1: Partial nucleic acid sequence of the winter flounder NHE1 carboxy terminus. ggtggagcttcttgcagtgaagaagaagaaggagagcaaacgatcaataaacgaggagattcacacacagttcctggatcatctgctcacaggcatcgaagacatctg tggacattacggacaccatcactggaaagacaagctgaaccgcttcaacaagtcctacgtgaagaagtggctgatcgcgggggaacgctcctccgagcctcagctca tctccttctacaacaagatggagatgaagcaggccatgatgctggtggagagcgggagcgccgccaagctgccctccatcgtatcctccgtctccatgcagaacattca gcaaaagggtccgaccagaggacgagcgataccgagcatctccaagagtcgagaggcggagatcagaaagatcctgagagcaaacctgcagaagacaagacag aggcttcgctcctacagccgacacgacctgatgatcgatcccttcgaggacaatgtgagtgaggttcgcttcaggaagcagagggtggagatggagaggaggatgag tcactatctcacggtccctgccaatcgccaggaaacccctccagtgaggaaagtctgttttgagccagaacaccaggtttatacgtacgacactgagggcgagagccc cagaggtcgagcagctcagactcatcccagccctcctgacgccatcggcctgatgaacgaagcgaccccgaggcccaatcagagcagcgcactgcgagatgaagg tgaaacggagctgagggcgaaagcacccgacgaccaggagg

The winter flounder amino acid sequence was 75% homologous to NHE1 from human and Amphiuma proximal to amino acid 700. After this point, the homology between the winter flounder and other species was lost except for sporadic short segments (Fig. 2).

Figure 2: Amino acid comparison of amphiuma, human, and winter flounder NHE1. 510 550 AMPHI MTIRPLVELL AVKKKQETKR SINEEIHTQF LDHLLTGIED HUM MTIRPLVDLL AVKKKQETKR SINEEIHTQF LDHLLTGIED WIN FLOU VELL AVKKKKESKR SINEEIHTQF LDHLLTGIED 551 600 AMPHI ICGHYGHHHW KDKLNRFNKK YVKKCLIAGE RSTEPQLIAF YHKMELKQAI HUM ICGHYGHHHW KDKLNRFNKK YVKKCLIAGE RSKEPQLIAF YHKMEMKQAI WIN FLOU ICGHYGHHHW KDKLNRFNKS YVKKWLIAGE RSSEPQLISF YNKMEMKQAM 601 650 AMPHI ELVESGGLGR IPSAVSTVSM QNIQPKAKPT DRFIPALSKV KEEEIRKILR HUM ELVESGGMGK IPSAVSTVSM QNIHPKSLPS ERILPALSKD KEEEIRKILR WIN FLOU MLVESGSAAK LPSIVSSVSM QNIQQKGPTR GRAIPSISKS REAEIRKILR 651 700 AMPHI TNLQKTRQRL RSYNRHTLVA DPYEEAWNQM LLRRQKAHQL EQRMNNYLTV HUMAN NNLQKTRQRL RSYNRHTLVA DPYEEAWNQM LLRRQKARQL EQKINNYLTV WIN FLOU ANLQKTRQRL RSYSRHDLMI DPFEDNVSEV RFRKQRV.EM ERRMSHYLTV 701 750 AMPHI PAHKMDSPTM TRARVGSNPM AYEPKANIRD LPTITIDPA. ..SPESVDIV HUM PAHKLDSPTM SRARIGSDPL AYEPK.. .ED LPVITIDPAS PQSPESVDLV WIN FLOU PANRQETPPV RKVCFEPEHQ VYTYDTEGES .PRGRAAQTH PSPPDAIGLM 751 800 AMPHI NEEKK...... SLPTE KEEEEEEGIV MTAKEPPSPG TDDVFTPGAG HUM NEELKGKVLG LSRDPAKVAE EDEDDDGGIM MRSKETSSPG TDDVFTPAPS WIN FLOU ...... A TPRPNQSSAL RDEGETE... LRAK...... APDDQE

We are continuing to sequence the winter flounder NHE1 and hope to have the complete gene shortly. We are also developing chimeras of human and winter flounder NHE1, to help isolate the regions of NHE1 that are responsible for inhibitor sensitivity. In the process, we will also determine those regions required for ion translocation. This work was funded by NIH grant HL21179 and The Salisbury Cove Research Fund. MOLECULAR IDENTIFICATION AND TISSUE DISTRIBUTION OF A MULTIDRUG RESISTANCE ASSOCIATED PROTEIN (Mrp2) ISOLATED FROM THE LIVER OF THE LITTLE SKATE, RAJA ERINACEA

Shi-Ying Cai, Carol J. Soroka, Ned Ballatori and James L. Boyer

Liver Center, Department of Medicine, Yale University School of Medicine, New Haven, CT 06520 and the University of Rochester School of Medicine, Rochester, NY 14642

Members of the MRP subfamily of ABC transporters function as multidrug export pumps for a variety of organic conjugates and are widely expressed. One member of this family, human MRP2 and rodent Mrp2 is localized to the apical membrane of hepatocytes where it functions to extrude a diverse group of compounds including bilirubin-glucuronide, glutathione and glutathione conjugates. We have previously described an ATP-dependent GSH and glutathione S-conjugate transporter in isolated plasma membrane vesicles from skate liver (Rebbeor et al. AJP 279:G417-G425, 2000). These studies support in-vivo and in-vitro studies from our laboratory in the free swimming skate (Raja erinacea) and isolated skate liver and hepatocyte cultures that demonstrate biliary excretion of Mrp2 substrates.

To identify this transporter at the molecular level, we developed degenerate oligonucleotide primers that hybridized with two highly conserved regions in Mrp2s from human, rabbit, mouse, C.elegans and yeast. Using RT-PCR, a 410 band fragment was amplified from both skate liver and kidney total RNA. DNA sequencing confirmed that this fragment coded for a portion of the Mrp2 protein. A 32P labeled fragment was prepared and used to screen a skate liver cDNA library. Colonies were isolated that contained up to 4 kb of the Mrp2 gene. After 5'RACE, a full-length skate Mrp2 gene was identified, which contains 5870 bp and encodes 1564 AA, with 197 bp 5’UTR, and 1033 bp 3’UTR. Genbank BLASTing indicates that this cDNA is an ABC transporter with the highest identities corresponding to human MRP2 and rabbit Mrp2 (56% each), mouse and rat Mrp2s (54%), and C.elegans Mrp2 (44%). Figure 1 illustrates a phylogenetic tree for the Mrp2 family. Computer modeling predicts 17 transmembrane domains, 2 ATP binding cassettes, 2 N-glycosylation sites, 6 PKC phosphorylation sites, and one tyrosine phosphorylation site in skate Mrp2. An apical targeting sequence, TAL has also been found at the C-terminal of sMrp2. Northern-blot analysis in multiple skate tissues showed that sMrp2 is a 6 kb transcript and is expressed in liver, kidney, and intestine. A rabbit polyclonal antibody was raised to a C-terminal peptide of sMrp2 and detects a 180 kD band in skate liver vesicle membrane in a Western-blot. Peptide competition experiments completely eliminate the staining of this 180 kD band. Immunofluorescence studies localize sMrp2 to the canalicular membrane of skate liver, the apical membrane of proximal convoluted tubes of the skate kidney, and the apical membrane of enterocytes in the skate small intestine. Mutation sites in the hyperbilirubinemic Dubin-Johnson Syndrome in man and the TR- /Groningen/EHBR rats are all conserved from this evolutionary sMrp2 precursor.

These findings indicate that a member of the Mrp2 subfamily is highly expressed in skate liver at the canalicular apical excretory domain. It is likely that this transporter is responsible for the canalicular excretion of a variety of organic anions (BSP, bilirubin, biliverdin, carboxyfluorescein diacetate, lucifer yellow) that we have demonstrated previously are excreted in high concentrations in bile of this marine vertebrate. Supported by DK 25636(to JLB), DK48823 (to NB) and ES 03828 from the USPHS

human mrp2 rabbit mrp2 mouse mrp2 rat mrp2 skate mrp2 human mrp1 human mrp3 human mrp6 c.elegans mrp2 yeast mrp2 human mrp4 human mrp5 64.6

60 50 40 30 20 10 0

Figure 1. Phylogenetic tree of MRP2/Mrp2s and other human MRPs. MEASUREMENT OF XENOBIOTIC INDUCED NITRIC OXIDE PRODUCTION IN KILLIFISH (FUNDULUS HETEROCLITUS) RENAL PROXIMAL TUBULES

Sylvia Notenboom1, Rosalinde Masereeuw1, Frans G.M. Russel1 and David S. Miller2 1Department of Pharmacol. and Toxicol., University Medical Centre, Nijmegen, The Netherlands 2Laboratory of Pharmacol. & Chem., NIH/NIEHS, Research Triangle Park, NC 27709

In killifish renal proximal tubules, two ATP-driven xenobiotic export pumps (Mrp2 and p-glycoprotein) are rapidly regulated by endothelin (ET) acting through an ETB receptor, NO synthase (NOS) and protein kinase C (PKC; Masereeuw et al, Mol. Pharm. 57:59-67, 2000; Notenboom et al, Am. J. Physiol, 282:F458-464, 2002). Several nephrotoxicants, radiocontrast agents, aminoglycoside antibiotics and heavy metal salts, induce ET release from the tubules and “hijack” this signaling pathway (Terlouw et al, Mol. Pharm. 59:1433-1440, 2001; Notenboom et al, Am. J. Physiol, op. cit.; Terlouw et al, submitted). The participation of NOS and NO in this signaling pathway is of particular interest, because of their role in hypoxia and chemical nephrotoxicity (Kone, Am. J. Kidney Dis. 30:311-333, 1997). Here we demonstrate directly generation of NO by killifish tubules after exposure to ET-1 or nephrotoxicants.

NO production was measured using 4,5-diaminofluorescein (DAF), which increases fluorescence upon exposure to NO. For experiments, tubules were loaded for 1 h in medium containing 10 µM DAF-2 diacetate (Molecular Probes). This non-fluorescent derivative is membrane permeant and upon entering cells is hydrolyzed to DAF. After loading, tubules were transferred to confocal chambers containing medium without (control) and with effectors. After 5 min, confocal images of tubules were acquired and saved. In controls, average tubule fluorescence was low and constant, but it increased about 5-fold when tubules were exposed to the NO generator, sodium nitroprusside. Roughly the same magnitude of increase in fluorescence (5-8-fold) was also found for tubules exposed to the nephrotoxicants, diatrizoate, CdCl2, amikacin and gentamicin, at concentrations that reduce transport mediated by Mrp2. Exposure to ET-1 also generated NO and its production was blocked by the NOS inhibitor, L-NMMA. Finally, NO production induced by gentamicin was blocked by the ETB receptor antagonist, RES-701-1. Together, these results demonstrate that ET-1 and the nephrotoxicants rapidly stimulated NO production, most likely through activation of the ETB receptor and NOS.

Additional experiments have shed light on signaling by NO. In many systems, NO signals through cyclicGMP and PKG. In killifish tubules, transport on Mrp2 was particularly sensitive to 8-bromo-cyclicGMP (50% reduction at 500 nM), whereas transport on the classical organic anion system was insensitive (no reduction at 10 µM). In initial experiments, KT-5823, a PKG-selective inhibitor, blocked in part the effects of 8-bromo-cyclicGMP on Mrp2-mediated transport, indicating NO signaling through cyclicGMP and PKG. However, we recently found that cyclic nucleotides compete for transport on Mrp2 in shark rectal gland tubules (Miller et al, Am. J. Physiol. 282:R774-781, 2002). Thus, it is also possible that in killifish renal tubules cyclicGMP could reduce Mrp2-mediated transport through competition; this possibility will be investigated in the near future. Supported by the Dutch Kidney Foundation and the MDIBL CMTS (ES 03828). REGULATION OF XENOBIOTIC EFFLUX PUMPS IN KILLIFISH (FUNDULUS HETEROCLITUS) BRAIN CAPILLARIES

Bjorn Bauer1, Gert Fricker1 and David S. Miller2 1Institut für Pharmazeutische Technologie und Biopharmazie, D-69120 Heidelberg, FRG 2Laboratory of Pharmacology & Chemistry, NIH/NIEHS, Research Triangle Park, NC 27709

The brain capillary endothelium is a formidable barrier to the entry of xenobiotics into the central nervous system (CNS). This barrier protects the CNS from toxic chemicals, but also denies entry to therapeutics, e.g., chemotherapeutics. Miller et al (Mol. Pharm. 58:1357-1367, 2000) developed a simple but powerful experimental system in which to study drug transport across intact brain capillaries. It consists of freshly isolated capillaries, fluorescent substrates and confocal imaging. This system was used to follow bath to lumen transport of selected fluorescent xenobiotics in capillaries from rat and pig where transport was found to be mediated by two ATP-driven drug export pumps, p-glycoprotein and Mrp2. One shortcoming of the system was the limited capillary viability. We recently validated a new and long-lived comparative model for studying drug transport across the blood-brain barrier (Miller et al, Am. J. Physiol., 282:R191- R198, 2002). Using brain capillaries isolated from two poikilotherms, a teleost (killifish) and an elasmobranch (dogfish shark, Squalus acanthias) we found, as in mammals, evidence for involvement of two ATP-driven xenobiotic export pumps, p-glycoprotein and Mrp2, in transport from CNS to blood. As is mammals, both transporters were localized to the luminal membrane of the endothelial cells. The present report documents initial attempts to identify mechanisms that alter transport function in killifish brain capillaries.

Accumulation (60 min) of fluorescent substrates in individual killifish capillary lumens was measured using confocal microscopy and quantitative image analysis. Substrates used were sulforhodamine 101 free acid for Mrp2 and a fluorescent verapamil derivative for p- glycoprotein. Cell to lumen transport by both drug efflux pumps was reduced by treatments that elevated intracellular cyclicAMP, cyclicGMP and nitric oxide (NO) and that activated protein kinase C (PKC). A PKC-selective inhibitor, blocked the effects of PKC activation. The polypeptide hormone, endothelin-1 (ET-1), at low nanomolar concentrations, also reduced transport. ET-1 appeared to act through an ET-B receptor and at least in part through NO synthase and NO. ET-1 did not act through PKC, PKA or PKG, suggesting that additional hormones could acutely alter transport function by signaling through parallel pathways. None of the treatments that reduced luminal accumulation of the fluorescent substrates increased passive permeability to small fluorescent dextrans, suggesting that reduced xenobiotic efflux pump activity rather than increased tight junctional permeability underlies the effects observed.

The present initial results indicate that at least four parallel signaling pathways (ET-NO, PKC, PKA, PKG) regulate Mrp2 and p-glycoprotein function in brain capillary endothelial cells. Since these xenobiotic export pumps are believed to be important constituents of the “active” blood-brain barrier, understanding how signaling is turned on and off could be of value in designing treatments to strengthen a damaged barrier and to overcome the barrier and deliver therapeutics to the CNS. Supported in part by a travel grant from the Maren Foundation to BB, by DFG FR1211/8-1 to GF and by the MDIBL CMTS (ES 03828). LOCALIZATION OF ENDOGENOUS XENOBIOTIC TRANSPORTERS IN ISOLATED CLUSTERS OF POLARIZED SKATE (RAJA ERINACEA) HEPATOCYTES

John H. Henson1, Sarah E. Kolnik1, Shi-Ying Cai2, Ned Ballatori3 and James L. Boyer2 1Dept. of Biology, Dickinson College, Carlisle, PA 17013 2Liver Center, Yale Univ. School of Medicine, New Haven, CT 06520 3Dept. of Environmental Medicine, Univ. Rochester School of Medicine, Rochester, NY 14642

Isolated skate hepatocytes retain their structural and functional integrity as clusters of polarized cells arranged around a central bile canaliculus. This in vitro model has served as an excellent system for the characterization of xenobiotic transport mechanisms given that the secretion of fluorescent transport substrates can be monitored in living cells using confocal microscopy. Recent studies have examined the expression and function of the transport proteins Bsep (bile salt excretory protein) and Mrp2 (multidrug resistance protein 2) in these cells using antibody probes generated against the mammalian homologues of these proteins. More recently, PCR-based methods have allowed for the generation of peptide antibodies against the endogenous skate forms of Bsep and Mrp2 as well as the organic anion transporting polypeptide Oatp (Cai et al., Bull. MDIBL 39:1-2, 2000; Cai et al., Bull. MDIBL 40:83-85, 2001). The objective of the present study was to study the cellular and subcellular distributions of these endogenous transporters and to begin to examine their expression during long term culture. In addition preliminary studies were done to characterize transporter regulation in these cells. Particular attention was paid to the possibility that cAMP may be instrumental in regulating Mrp2 localization as has been shown with mammalian hepatocytes (Roelofsen et al., J. Cell Sci. 111:1137-1145, 1998).

The isolation and culture conditions for skate hepatocytes have been described previously (Ballatori et al., Bull. MDIBL 37:85-86, 1998). The cells were allowed to settle onto poly-L- lysine coated coverslips and then fixed in either ice cold methanol or acetone. The cells were then rehydrated in PBS and stained with the appropriate peptide-specific rabbit polyclonal antibodies against skate Mrp2, Bsep, or Oatp, followed by labeling with goat anti-rabbit secondary antibodies conjugated with fluorescent compounds. Additional characterization of transporter localization was performed by double labeling the skate cells for filamentous actin using fluorescent phalloidin or staining nuclei using the DNA dye propidium iodide. Fluorescently labeled cells were viewed using a 40X (1.0 NA) water-immersion objective lens on an Olympus Fluoview laser scanning confocal microscope. For the Mrp2 regulatory studies cultures of skate hepatocyte clusters were exposed to dibutrylcAMP and the transport of the Mrp2-specific substrate sulforhodamine was analyzed using a Zeiss 510 multiphoton imaging system. These same cultures were then processed for immunofluorescent staining with anti-skate Mrp2.

Figure 1 shows that skate Mrp2 displays an apical membrane concentration typical of that seen with this transporter in mammalian cells. In addition faint punctate fluorescent objects are visible in the cytoplasm of the skate hepatocytes and these may correspond to vesicle-associated skate Mrp2. The distribution of skate Oatp appears opposite to that of skate Mrp2, given that skate Oatp is concentrated on the basolateral margins of the cells and appears excluded from the apical membranes which form the bile canaliculus (arrow). Bsep distribution appears to combine elements of each of these other transporters since it concentrates in the apical region but clearly also displays some basolateral labeling. This is unexpected and may reflect cross reactivity or the fact that all of the transporter has not been trafficked to the apical membrane. All three of the endogenous transporters appeared to maintain their polarized localization pattern in skate clusters kept in culture for up to seven days. The results of the Mrp2 cAMP regulation studies suggested that cAMP may alter Mrp2 localization patterns and canalicular geometry, however these results are very preliminary and need to be substantiated through additional experimentation.

Figure 1: Immunofluorescent labeling of skate Mrp2, Oatp and Bsep in freshly isolated skate hepatocyte clusters.

Analysis of the distribution and function of endogenous xenobiotic tranporters is an important step in the continuing characterization of the novel experimental system represented by isolated skate hepatocyte clusters. These cells have advantages over isolated mammalian hepatocytes, including maintenance of structural and functional polarity and long term viability, and as such provide an exceptional in vitro model for determining xenobiotic transport mechanisms. Future experiments will concentrate on determining the possible regulation of xenobiotic transporters in the skate hepatocyte system. Additional cAMP-based studies will be carried out along with experiments aimed at determining if transporter expression and/or localization is impacted by exposure to environmental toxins.

Special thanks to Dr. David Miller for performing the sulforhodamine transport analysis. Supported by grants DK25636 and ES03828 from the NIH and a Howard Hughes Medical Institute Undergraduate Biological Science Education Program Grant to Dickinson College. REPRODUCTIVE IMPACTS OF ARSENIC EXPOSURE IN MUMMICHOGS MEDIATED BY MRP1 Nikki L. Maples, Janis S. K. Peterson, and Lisa J. Bain Department of Biological Sciences, University of Texas at El Paso, El Paso, TX 79968

Arsenic is a common pollutant of both surface and ground water, and can be found in relatively high concentrations in drinking water, and is considered to be a human carcinogen, although the exact mechanism of action is not well understood. The new drinking-water standard for arsenic has come under much scrutiny from congress, the scientific community, and the public recently. Thus, knowledge of the mode of arsenic toxicity, as well as the potential for alterations in human reproductive function, needs to be further investigated. This is best done in a species such as the mummichog (Fundulus heteroclitus) that is easily manipulated, breeds in a synchronous pattern, and one in which the developing embryos can be readily monitored. Recently, investigators have suggested that arsenic may work through an estrogenic mode of action, as rats exposed to an environmentally relevant drinking water level of arsenic had a significant reduction in plasma levels of estrogen, along with decreased ovarian, uterine, and vaginal weights (Chattopadhyay, S., et al., J. Toxicol. Sci. 24:425-431, 1999). We hypothesize that the part of the mechanism of arsenic toxicity may be due to the multidrug resistance-associated protein-1 (MRP1/ABCC1) and its ability to transport both conjugates of arsenic as well as 17b-estradiol glucuronide, which will result in an a decrease in estradiol levels, an increase in oxidative stress, and ultimately causes lowered reproductive fitness.

Adult mummichogs were collected near MDIBL and half were exposed to 0.4ppm sodium arsenate for four days. The fish were placed in clean 70% seawater, allowed to spawn, and the total number of eggs produced, their viability and hatchability, and the survival of the hatchlings up to four days was examined. The total number of eggs produced did not differ between the groups, but the viability of the eggs was reduced by almost half in the tanks where both parents were exposed to arsenic. The failure of seemingly viable embryos to hatch was increased 2-fold in tanks where either one of the parents or both the parents were exposed. The survival of the hatchlings to four days was not different between any of the groups. However, it does appear that arsenic exposure is causing some reproductive and developmental impacts in these organisms. Next, serum estradiol and testosterone levels, and hepatic MRP1 levels were examined in the adults. None of the levels differed significantly between exposure groups. Glutathione S-transferase activity and total glutathione concentration was examined in both the adults and the hatchlings as a measure of conjugative ability and potential oxidative stress. Neither the enzyme activity nor glutathione levels was different in either the adults or the hatchlings. We are currently examining differential gene expression in the adults and the hatchlings in an attempt to discern the mechanism of action of arsenic. We hope that these techniques will reveal groups of genes up- or down-regulated that play an important role in arsenic’s toxicity, with particular interest in its ability to alter reproductive function. (Supported in part by the CMTS grant P30 ES03828-16). DEVELOPMENT OF TWO SUBTRACTIVE cDNA LIBRARIES FROM WINTER FLOUNDER.

William Sandry Baldwin and Jonathan Adam Roling Biological Sciences, University of Texas at El Paso, El Paso, TX 79968

Chromium is a major contaminant at several Superfund sites including Shipyard Creek in Charleston, SC, Pacific Sound Resources in Elliott Bay and Puget Sound, Seattle, WA, as well as Portsmouth Naval Shipyard in Kittery, ME. These marine and estuarine areas contaminated with chromium are primarily contaminated with the trivalent or hexavalent forms, both of which tend to accumulate in the sediments. Therefore, winter flounder (Pseudopleurinectes americanus) was used as the sentinel species to examine the effects of chromium on alterations in gene expression in liver. Furthermore, winter flounder is a carnivorous species that may also biomagnify chromium from its food, as well bioconcentrate chromium from the water and sediments. Humans also consume large quantities of flatfish, including winter flounder and halibut, which are sediment dwellers.

Three winter flounder per group were injected with either 0.15N saline, 400µg/kg Cr(III), or 25µg/kg Cr(VI) in 0.15N saline. Each flounder weighed between 300 and 360 grams. Twenty-four hours following injections, winter flounder were euthanized with MS-222 and the livers were excised. Half of the livers were used to make cytosol and microsomes and the other half was extracted for RNA using TriZol Reagent (Life Technologies, Gaithersburg, MD). Total RNA was used to obtain mRNA for subtractive hybridization, which was done using the Clontech PCR-SelectTM cDNA subtraction Kit (Palo Alto, CA). Subtractive clones were obtained and individual cDNA clones were amplified by PCR and spotted onto nylon filters to make 96-well microarrays. Microarrays have demonstrated several potentially differentially expressed genes due to Cr(VI) and a few that may be differentially expressed due to Cr(III).

For Cr(VI), we have partially sequenced several genes that are putatively differentially expressed, including glutathione S-transferase alpha3, a non-selenium glutathione peroxidase, protein elongation factor-2, carboxypeptidase, cysteine proteinase inhibitor, and complement regulatory plasma protein (ApoH). Several other genes (ESTs) have no or considerably weaker similarities based on GenBank. Primers have been made to ApoH, elongation factor-2, GSTα and glutathione peroxidase. RT-PCR indicates that the GST is down-regulated about 40% and glutathione peroxidase is down-regulated more than 80%. GSH levels are doubled, indicating that exposure to Cr(VI) causes oxidative stress. RT-PCR has shown only a slight change in the regulation of elongation factor-2 by Cr(VI). Furthermore, data from our arrays recently confirmed by real-time PCR, indicates that an EST fragment isolated during subtractive hybridization is reduced 90%.

For Cr(III), we have partially sequenced two putative cysteine proteases, and a putative oxidoreductase/periplasmic zinc transporter, however, we have not confirmed differential expression of any of these genes. Interestingly, Cr(III) treated HepG2 human liver hepatoma cells have demonstrated alterations in cysteine proteases and several proteinase inhibitors using cDNA and confirmed by RT-PCR. Further work may demonstrate similar results in vivo with flounder liver. (Supported in part by the CMTS grant P30 ES03828-16). STRUCTURAL AND BIOCHEMICAL CHANGES IN THE BLOOD-CSF BARRIER OF SQUALUS ACANTHIAS IN RESPONSE TO HEAT SHOCK AND ZINC EXPOSURE

Alice R.A. Villalobos1, David S. Miller2, Cynthia Dehm, and J. Larry Renfro3 1Department of Environmental Medicine, University of Rochester, Rochester, NY 14642 2Laboratory of Pharmacology and Chemistry, NIEHS-NIH, Research Triangle Park, NC 27709 3Department of Physiology and Neurobiology, University of Connecticut, Storrs, CT 06269

The choroid plexus epithelium forms the blood-cerebrospinal fluid (CSF) barrier and ac- tively removes metabolic wastes and xenobiotics from CSF, which is contiguous with the ex- tracellular fluid (ECF) bathing the neural and glial cells of the central nervous system. Thus, modulation of the barrier’s morphology and transport capacity may compromise homeostatic regulation of the brain’s ECF and, in turn, the activity of neural networks. Others have shown that thermal and chemical stresses may modulate both physical and functional properties of other transporting epithelia, such as renal proximal tubule (Brown, M.A. et al., PNAS 89:3246-3250, 1992). Structural and biochemical effects of heat stress and zinc on the blood-CSF barrier were evaluated in isolated IVth choroid plexus of spiny dogfish shark (S. acanthias). Induction of heat shock protein-70 (HSP70) in control and stressed tissues was determined by immunoblot analysis, and changes in morphology and ultrastructure were assessed by scanning and transmis- sion electron microscopy.

The morphology and ultrastructure of freshly dissected choroid plexus and control tissues maintained in medium M199 modified for elasmobranch tissues (Valentich, J.D. et al., Am. J. Physiol. 260:C813-C823, 1991) for 7.5 h at 13.5˚C (i.e., daily peak temperature of the holding tank) appeared virtually identical; immunoblot analysis showed no induction of HSP70. The tips of the clavate-shaped microvilli at the apical (CSF-facing) surface are normally ~0.3 µm in maximum transverse width and ~0.08 µm in minimum transverse width (Table 1). However, following mild heat stress (MHS, 18.5˚C, 3 h) with recovery (1.5 h, 13.5˚C) the microvilli had expanded to roughly three times the normal size with no change in microvillus length.

Table 1. Effects of temperature stress and zinc on dimensions of clavate microvilli in shark IVth choroid plexus. Dimensions Control, 18.5C,3h: 23.5C, 1h: (µm) Unexposed 13.5 C, 7h 13.5C, 1.5h 13.5C, 1.5h 50 µM Zn, 6h

Microvillus 1.17 1.77* 1.12 1.62* 1.82* length ±0.029 ±0.120 ±0.036 ±0.074 ±0.110

Maximum 0.30 0.33 0.88* 1.28* 0.96* transverse ±0.033 ±0.050 ±0.096 ±0.080 ±0.095

Minimum 0.07 0.09 0.22* 0.20* 0.25* transverse ±0.008 ±0.005 ±0.029 ±0.026 ±0.037 Values are mean ± SEM of n = 6 measurements. Determinations were done on scan- ning electron micrographs (20,000 X). *Significantly different form control at P < 0.05. Transmission electron micrographs revealed that the various intracellular structures and apical tight junctional complexes were unchanged, but the enlarged microvilli were multiple lay- ers of fused membranes and myeloid-like whorls. In separate tissues, MHS with recovery in- duced HSP70 approximately 3-fold. By comparison, severe heat stress (SHS, 23.5˚C, 1 h) with 1.5 h recovery resulted in variable enlargement of microvilli (Table 1); some areas were indistin- guishable from control tissue, while others resembled MHS tissues. In a separate set of SHS tis- sues there was minimal induction of HSP70. Six-hour exposure to 25 or 50 µM ZnCl2 at 13.5˚C with 1.5 h recovery in zinc-free buffer resulted in 3-fold distension of the microvilli, identical to that described for MHS (Table 1). In a separate experiment, tissues were incubated for 12 h in zinc-free tissue culture medium or for 6 h with 50 µM ZnCl2 and 0 or 6 h in zinc-free medium. Immediately after zinc exposure, levels of HSP70 were comparable to controls. However, at 6 h post-exposure, HSP70 accumulation had increased 3-fold.

These findings suggest that physical and chemical stresses, such as hyperthermia or heavy metals alter blood-CSF barrier morphology. These structural changes are accompanied by induction of HSP70 accumulation, a hallmark of the cellular stress response. However, the se- verity of a given stress may limit the cell’s adaptive responses. This may explain in part the at- tenuated morphologic changes and minimal induction of HSP70 with severe heating. Whether such changes in morphology or HSP70 accumulation are accompanied by or directly coupled to changes in solute transport remains to be determined.

This work was funded in part by NSF-IBN980616; NINDS-NS39452S; NIEHS- ES10439; NIEHS-ES01247; NIEHS-ES03828; NSF-IBN9604070; NSF-IBN0078093. Ms. Dehm was supported in part by NSF-DBI-9820400. ARA Villalobos was supported by a New Investigator Award. TOXICANT INDUCED DIFFERENTIAL GENE EXPRESSION AND PRODUCTION OF AN AQUATIC GENE ARRAY

Seth W. Kullman, and David E. Hinton. Nicholas School of the Environment, Duke University, Durham, North Carolina 27708

Most toxic substances released into the environment either accidentally or by deliberate discharge eventually become aquatic contaminants. A concerted effort has thus been made to develop toxicity assays that precisely reflect the impact of chemical exposure on aquatic organisms, aquatic ecosystems and ultimately human health. Recent advances in toxicogenomics have resulted in the development of high throughput gene analysis technology that could literally revolutionize toxicity monitoring. We propose to take advantage of these developments to create DNA arrays for high throughput aquatic exposure assessment. The fundamental basis of these assays is to monitor altered gene expression in aquatic organisms following exposure to environmental toxicants. Because gene expression is a sensitive end point, DNA array technology is ideally suited to identify toxicity following acute and chronic exposures. DNA arrays can screen thousands of genes simultaneously, eliminate the time and cost of traditional biomarker analysis and aid in determining chemical mechanisms of action. Two specific objectives were identified for the development and characterization of these arrays: (1) obtain differentially expressed, toxicant responsive genes following exposure of medaka fish (Oryzias latipes), to prototypic classes of environmental toxicants including poly-halogenated hydrocarbons (2,3,7,8-tetrachlorodibenzo-p-dioxin) and environmental estrogens (17-a - ethinylestradiol); (2) array differentially expressed genes and identify unique gene markers of toxicity and characteristic gene expression profiles. Ideal gene markers will be quantifiable, display a high degree of sensitivity (i.e. detectable at low concentrations), have a reasonable degree of accuracy (i.e. responsive over a range of concentrations), and exhibit limited antagonism by secondary stressors. These studies were initiated to test the validity of using DNA arrays to rapidly screen changes in gene expression in response to aquatic toxicant exposure.

The major goal of this project is to develop the use of DNA-arrays as a quantitative tool to assess exposure and effect of environmental toxicants in the aquatic environment. To accomplish this aim, toxicant-induced gene markers for prototypic classes of environmental pollutants were to be identified by suppressive subtractive hybridization. This is a highly sensitive method with a proven track record for isolating novel and rare differentially expressed transcripts. Differentially expressed genes are represented as a library of subtracted clones, which are subsequently arrayed on nylon membranes and characterized for usefulness as biomarkers of toxicity.

Young adult, three month old, sexually mature, medaka were exposed to a single dose of 2,3,7,8-tetrachlorodibenzo-p-dioxin, or 17-a -ethinylestradiol by i.p. injection for 48 hrs. In this phase, concentrations of each toxicant were chosen that are sub-lethal, yet sufficient to initiate transcription of responsive genes. Environmental conditions were rigorously maintained by partial submersion of exposure containers into a tabletop tank system with a 16 hr light/8 hr dark photoperiod and constant temperature of 25˚C. Exposure duration was determined by testing a subset of treated fish for induction of well characterized biomarkers for dioxin, for EE2 exposure. Livers were collected form a subset of exposed and control fish at 0, 12, and 48 hrs and western blot analysis was conducted for induction of CYP1A1 and vitellogenin proteins. Each analysis revealed a single, time dependent, protein band of appropriate molecular weight in exposed animals. Total RNA was subsequently isolated from the remaining fish (48 hrs after dosing), cDNA was synthesized, and libraries were constructed for each exposure and control population. Forward and reverse subtractive hybridization was performed from exposure condition. Initially, cDNAs from treated fish were used as tester DNA and non-treated cDNA as driver to generate subtracted libraries that are enriched for genes that have been induced (forward subtraction). A second round of SSH was performed using cDNA from control fish as tester DNA and cDNA from treated fish as driver to generate a subtracted library enriched for genes that are repressed (reverse subtraction). The resulting cDNA libraries were subsequently cloned into a T/A cloning vector and transformed in to E. coli DH5a cells. Individual colonies were plated on LB agar plates, selected, and inoculated into replicate 96 well plates for growth, storage and preservation. Significant progress was made on aim 1 of this research project resulting in the identification of several differentially expressed genes from both estrogen and dioxin treated animals. Gene sequences are currently being sequenced and compared to existing data bases for identity. Once sequencing is complete, aim two will be addressed.

Small aquarium fish species including medaka are increasingly being used as vertebrate models in various fields of biology including toxicology, genetics, and developmental biology. These organisms are suitable for use in genomic studies due to their small genome size, and ease of application in forward genetics. The medaka fish have become a standard model organism in aquatic toxicity studies. Their transparent chorions, reproductive capacity, external gender characteristics, and sensitivity to toxicants are but a few of the characteristics that have made them an excellent model for toxicity and toxicogenomic studies. With the emerging technology in gene expression analysis, biomarker applications in applied toxicology can be greatly improved. Genome expression technologies are becoming increasingly reliable and accessible and it is now feasible to generate global mRNA expression profiles for whole genomes following stimuli such as exposure to toxicants. Measurement of global mRNA expression is of particular importance in disciplines like toxicology, where toxic phenotypes are produced by altered expression of multiple genes. Interest in the application of DNA- array analysis for toxicity studies is rapidly growing and has proven useful for defining mechanisms of toxicity and classifying toxicants based on characteristic gene markers and expression profiles. Compared to contemporary standardized bioassays, it is expected that DNA-array assessment of global mRNA expression will greatly improve the accuracy and efficiency of toxicity screening. It is believed that this technology will have a broad base of applications pertaining to many aspects of toxicology including assessment of toxicity in drug development, elucidation of mechanism of action, prediction of disease status, determining occupational exposure and monitoring pollutant exposure in the environment. This work was funded in part by the North Carolina Biotechnology Center NCBC#2001-ARG-0030. DERIVATION OF CONTINUOUS MARINE CELL LINES FOR MODEL SYSTEMS IN TOXICOLOGY AND CELL BIOLOGY

David W. Barnes1, John N. Forrest, Jr.2, John P. Wise, Sr.3 and Richard N. Winn4 1National Stem Cell Resource, ATCC, Manassas, VA 20012 2Department of Internal Medicine, Yale University School of Medicine, New Haven, CT,. 06510 3Environmental and Genetic Toxicology, Schools of Medicine and Public Health, Yale University, New Haven, CT, 06510 4Warnell School of Forest Resources, University of Georgia, Athens, GA 30602

In recent years, methods have been developed for establishing cell lines from freshwater and marine fish (Collodi, P. et al., Cell Biol. and Tox. 8:43-61, 1992; Bradford, C. et al., Mol. Marine Biol. and Biotech. 6:270-288, 1997; Buck, C. et al., Marine Biotech. 3:193-202, 2001). These lines have provided models for both molecular developmental biology and toxicology (Collodi, P. et al., Xenobiotica, 24:487-493, 1994; Singh, N.N. et al. Marine Biotech, 3:27-35, 2001). Many of these cultures retain normal karyotype and at least partial differentiated function. The cultures do not show growth crisis, unlike mammalian cells cultured in conventional media. The goal of this project is to derive continuous cell lines for model systems used at MDIBL. One of these is the dogfish shark (Squalus acanthias) rectal gland (Frizzell, R.A., Physiol Rev. 79(Suppl):S1-2, 1999; Aller, S.G. et al., Am. J. Physiol. 276:C442-9, 1999; Waldegger, S. et al., Pflugers Arch. 437:298-304, 1999). Other models include skate (Raja erinacea) hepatocytes (Rebbeor, J.F. et al., Am. J. Physiol. Gastrointest. Liver Physiol. 279:G417- 25, 2000; Ballatori, N. et al., Am. J. Physiol. Gastrointest. Liver Physiol. 278:G57- 63.2000) and killifish (Fundulus heteroclitus) tissues (Hossler, F.E. et al., J. Morphol. 185:377-86,1985; Mickle, M. and Cutting, G. R., Med Clin North Am. 84:597-607, 2000; Moyer, B.D. et al., J. Clin. Invest.104:1353-1358, 2000).

The methods used to establish fish cell cultures rely on the replacement, reduction or supplementation of the usual bovine serum supplement with a variety of purified peptide growth factors, nutritional supplements and other proteins specific for the cell type (Helmrich, A. and Barnes, D. Meth. Cell Biol. 59:29-38, 1999). For instance the peptides epidermal and fibroblast growth factor in combination with extracts of trout embryo, shark egg yolk and serum from trout or eel have proved effective in specific cases. Attention is also given to optimization of the basal nutrient medium for fish cell culture. Conveniently, cultures often can be weaned from some of the more exotic supplements, once dependably proliferating cells have been derived. In some cases it is necessary to alter the substratum or passaging procedure or provide conditioned medium from other established fish lines to allow derivation of functional cultures.

Primary cultures of dogfish shark rectal gland were examined for growth responses to a wide range of peptide hormones and other potential effectors of proliferation in both primary and secondary cultures. Passaging protocols were compared, using a variety of hydrolases (eg., trypsin, collagenase, hyaluronidase) and procedures, and it was found that a simple increase in temperature of about 5 degrees C for the period of the enzyme treatment improved survival of passaged cells. This may be related to a protective alteration of the membrane through lipid phase changes for the period of enzyme treatment, and is not simply the result of increased enzyme activity at the higher temperature. Extended incubation of primary cultures revealed that a low percentage of proliferating fibroblastic cells are present in the cultures. A range of substratum treatments was assessed for improvements in cell adhesion, and none were found better than plating the cells on a collagen II gel, identical to that used in the primary cultures.

A combination of dibutryl c-AMP (1 mM) and forskolin (10 microM) increased cell survival and markedly altered the morphology of the cells, leading to a more compact appearance. Cyclic AMP is recognized to be involved in signaling for transport phenomena in these cells, and may also have a role in cell survival and proliferation. A similar phenomenon is seen with some mammalian kidney cell cultures, in which cAMP influences both cell proliferation and expression of differentiated physiological function. Because of the naturally slow growth rate of the cells, results have been limited, and considerably more insight will be required to establish multipassage proliferating cultures of rectal gland cells. Cells from skate liver also were initiated as primary cultures and have been maintained with evidence of metabolic activity over more than six months. The cultures have been best maintained as multicellular aggregates rather than populations of individual, autonomous cells.

Attempts were made to develop cell lines from a transgenic medaka strain carrying a bacteriophage lambda cII transgene target (Winn R. et al., Proc. Natl. Acad. Sci. U S A 97:12655-60, 2000). This system provides a means for quantitative assessment of some classes of mutations and assay of genotoxic activity of purified environmental toxins and mixtures of known or unknown toxins. Actively proliferating cultures of liver and gill cells from the transgenic medaka were obtained using variations of media developed for zebrafish. Similar results were also obtained with cultures from Fundulus spleen and ovary. The transgenic medaka liver cells were successfully passaged and behaved much like zebrafish liver cells in culture. Unfortunately, these cultures were lost to contamination, but it is likely that the cultures can be reproducibly initiated.

A permanent, reproducible source of experimental material and the ability to manipulate both the environment of the system and the genetics of the cells in vitro will provide additional tools for exploitation of the unique properties of the shark rectal gland). This and the other culture systems also will be useful for heavy metal or other toxicity studies (Forrest, J.N. et al., J Exp Zool. 279:530-6, 1997) and will facilitate examination of xenobiotic metabolism and transport (Miller, D.S. et al., Am. J. Physiol.. 275:R697-705, 1998). The culture of cells from the transgenic medaka will allow comparison of mutagenic activity for genotoxins in vivo and in vitro, as well as providing a system for identifying modulators of mutagenesis and mechanisms of action. The ability to transfect and select fish cells in vitro also allows the use of these cultures in studies of molecular biology and mechanisms of action of physiological and hormonal mediators of cellular behavior. Supported in part by a New Investigator Award to D.B from NIEHS P30-ES3828 (Center for Membrane Toxicology Studies) and P40 RR 15452 to D.B. (National Stem Cell Resource). ESTABLISHING CELL LINES IN FOUR MARINE ANIMALS: BOWHEAD WHALE, (BAELAENA MYSTICETUS), BELUGA WHALE (DELPHINAPTERUS LEUCAS), LITTLE SKATE (RAJA ERINACEA), AND SPINY DOGFISH (SQUALUS ACANTHIAS)

John Pierce Wise, Sr., Peter. G. Antonucci, Shawn E. Holt, Lynne Elmore, John N. Forrest Jr., James Boyer and B. H. Bryant Yale University School of Medicine New Haven, CT 06520

The long-term objective of this research is to develop models of metal-induced carcinogenesis in cells from marine animals for comparison with human models. This is important because several metals are established carcinogens and both human and marine animals are heavily exposed to metals. Currently, there are no models and little or no data concerning the genotoxic effects of metals in marine animals. Thus, the short-term objective of this research is to focus on one metal of significant public health concern (hexavalent chromium (Cr(VI)), and establish the necessary cell lines to develop this model.

Skate hepatocytes were obtained from Dr. James Boyer. Shark rectal gland epithelial cells were obtained from Dr. John Forrest. Beluga skin fibroblasts were obtained from the Mystic Aquarium in Mystic, Connecticut and bowhead kidney cells were obtained from Thomas Goodwin at NASA in Houston., Texas. The approach was to immortalize the cells with hTERT the catalytic subunit of telomerase, E6/E7 the oncoproteins produced by human papilloma virus and large t-antigen the oncoprotein produced by Simian virus 40. Each immortalizing vector contained the immortalizing gene and a drug resistance gene. Thus, cells expressing the immortalized gene were selected based on their resistance to geneticin or neomycin. Once immortalized the cell lines were to be derived from single cells and analyzed for genotoxicity after Cr(VI) exposure.

The project presented many technical challenges. The culture and reagents of the shark and skate cells require additional salts and urea to maintain a high osmolality. Because of the need and relevance of these cell lines for the laboratory as a whole, the decision was to focus on these cells first. Sharks cells were successfully passage twice using standard mammalian tissue culture methods. Further, shark cells and tissues were found to endogenously express high levels of shark telomerase, which suggests that trying to immortalize the cells with human telomerase is not the right approach. Accordingly, future experiments will focus on optimizing the cell culture conditions and immortalizing the cells with SV40 large T antigen and the E6/E7 oncoproteins.

The culture of skate cells provided different challenges. Cultures would succumb to bacterial infections after several days. Some success was found as one culture was maintained for 60 days without contamination. Skate cells were not successfully passaged. Like the shark, skate cells and tissues were also found to endogenously express high levels of skate telomerase, which suggests that trying to immortalize the cells with human telomerase is not the right approach. Accordingly, future experiments will focus on optimizing the cell culture conditions and immortalizing the cells with SV40 large T antigen and the E6/E7 oncoproteins. Important steps were accomplished with the project regarding the culture of shark and skate cells. The next steps are: 1) define new culture media for the skate; 2) optimize the protocol for passaging the shark cells; 3) infect the shark, skate and whale cells; 4) begin defining culture variables for the killifish; and 5) evaluate the genotoxicity of Cr(VI) in each. The last step will be to compare any observed genotoxicity with data concerning the genotoxicity of Cr(VI) in immortalized human cells that has already been determined by the Principle Investigator.

This project was supported by Mount Desert Island biological Laboratory and a New Investigator Award from the Center for Membrane Toxicity Studies to John Pierce Wise, Sr. and grant to the Center from NIEHS (NIEHS P30 ESO3828-15). A LOW DOSE OF MERCURY INHIBITS PHOSPHORYLATION OF THE NA-K-CL COTRANSPORTER IN THE RECTAL GLAND OF THE DOGFISH SHARK, SQUALUS ACANTHIAS.

Brian F.X. Dowd, Arnaud Bewley, and Biff Forbush Department of Cellular and Molecular Physiology, Yale University, New Haven, CT 06520

The Na-K-Cl (NKCC1) cotransporter is essential for epithelial salt and water secretion as well as for the maintenance of cell volume and [Cl] in a wide variety of animal cells. We and others have previously shown that Hg inhibits the function of the NKCC1 in shark rectal gland at concentrations near 25 µM (Jacoby et al.,Am. J. Physiol., 277, C684-92, 1999; Kinne-Saffran and Kinne, Biochim. Biophys. Acta., 151, 442-451, 2001). On the other hand, short circuit current in rectal gland cell monolayers is inhibited at Hg concentrations below 10 µM, and this effect has been attributed to an effect on shark CFTR (Ratner et al; Bull MDIBL 37:20-22, 1998). When forskolin is used as a stimulus, NKCC1 is activated by phosphorylation in response to ionic changes secondary to CFTR activation, and it would thus be expected that NKCC1 activation would be very sensitive to Hg. In contrast, NKCC1 is also activated by hypertonicity, but this does not involve CFTR, and might not be expected to be Hg sensitive. Here we report the effect of inorganic mercury on the phosphorylation state of NKCC1. The results do not fit the simple hypothesis above, but give new insights into the action of Hg on regulation of cellular processes. They also suggest that up-regulation of the Na-K-Cl cotransporter may be the critical point at which mercury affects secretion in the rectal gland.

These experiments were accomplished in isolated rectal gland tubule cells, using a polyclonal antibody (anti-P-NKCC) that detects the cotransporter when it is phosphorylated at T184 and T189 (Flemmer et al; Bull. MDIBL 38:80-2, 1999). Shark rectal gland tubules were prepared as described previously. NKCC1 was activated in the tubule cells either by addition of 50 µM forskolin or by shrinkage in 575 mM sucrose, and HgCl2 was added either in a preincubation, concurrent with activating additions, or concurrent with lysis of the cells with 1% Triton. All experiments were terminated by addition of 1M H3PO4/1% Triton and the samples were analyzed by dot blotting and development with anti-P-NKCC.

We found that preincubation with inorganic Hg completely prevents phosphorylation of NKCC1 in rectal gland tubule cells. With a 4% cytocrit, the K0.5 for this effect of mercury was about 35 µM. We found however that by reducing the cytocrit we proportionately reduced the K0.5 for mercury, demonstrating that Hg absorption by the tubule cells is a Figure 1. The effect of inorganic Hg on phosphorylation of NKCC1. These are individual determinations of NKCC1 phosphorylation activated either by forskolin or hypertonic medium as marked in tubules at 0.06% cytocrit. Representative of results of three experiments. limiting phenomenon. At very low cytocrits (0.06%), K0.5 was found to be 1-2 µM and it appears from the available dataset that this is the limiting value for Hg sensitivity of cotransporter activation.

We examined cotransporter phosphorylation with a hypertonic stimulus as well as with forskolin. To our surprise, both modes of activation were inhibited by Hg at 1-4 µM and the K0.5s were indistinguishable from one another in internally paired experiments (Fig. 1). This finding strongly supports a model in which a single inhibited process occurs after the point of convergence of the two regulatory stimuli; a likely candidate is the cotransporter-kinase, which is yet unidentified.

Since hypertonic stimulation of NKCC1 does not involve CFTR, these findings demonstrate that Hg inhibition of NKCC1 phosphorylation is not due to inhibition of CFTR. On the contrary, we note that NKCC1 activation is necessary for transepithelial ion transport to occur, and for generation of short-circuit current in an Ussing chamber system. Thus we propose that the Hg inhibition of short-circuit current measured by Forrest and coworkers (Ratner et al; op cit) in rectal gland tissue culture is most likely due to inhibition of NKCC1 activation, as well as to direct inhibition of CFTR as proposed.

In the course of these experiments it was noted that while mercury inhibits activation of NKCC1 by hormonal and hypertonic stimuli, by itself it actually increased activation of NKCC1 in the resting state. The K0.5 of this effect was >40 µM, but was difficult to determine, since maximal activation was not reached even at 100 µM Hg. A similar observation has been made for the activation of JNK by Hg in LLC-PK1 cells, where the phosphorylation of JNK is increased by exposure to 10 µM Hg (Matsuoka, M. et al. Toxicol. Sci. 53: 361-368, 2000). As to the mechanism of the activation of NKCC1, it is possible that it is achieved (a) by inhibition of the phosphatase which dephosphorylates NKCC1, (b) activation of the kinase, or (c) an effect upstream of NKCC1 regulatory phosphorylation/ dephosphorylation.

To investigate the possibility that the relevant phosphatase is inhibited by Hg (a, above), the rate of dephosphorylation of NKCC1 was examined following solubilization of the rectal gland cell with 1% Triton. This method was used previously to demonstrate that PP1 is the phosphatase that is involved in NKCC1 dephosphorylation (Darman et al, J. Biol. Chem., 276, 34359-34362, 2001). In the current experiments it was found that Hg strongly inhibited dephosphorylation of NKCC1 with a K0.5=28±2 (n=3) without preincubation and K0.5 = 15 µM (two of three experiments) following a 10 min preincubation. Presuming that the somewhat different dose-dependence is due to the broken vs. intact cell difference, this result provides a clear explanation for activation of NKCC1 by Hg in the resting state. This work was supported by NIH DK 47661 and NIEHS P30-ES 3828. A REPRESSOR SUBSTANCE OF CARBONIC ANHYDRASE INDUCTION IS PRESENT IN THE EYESTALKS OF THE EURYHALINE CRAB, CARCINUS MAENAS , BUT NOT THE STENOHALINE CRAB, CANCER IRRORATUS

Raymond P. Henry Department of Biological Sciences, Auburn University, Auburn, AL 36849

Carbonic anhydrase (CA) activity is induced tenfold (from approximately 120 to 1,200 -1 -1 umol CO2 mg protein min ) in the posterior, ion transporting gills (G7-9) of the euryhaline green crab, CARCINUS MAENAS , after transfer from 33 to 10 ppt salinity (Henry et al., Bull. Mt. Desert Island Biol. Lab. 38:55, 1999). This induction is a result of salinity-mediated gene activation and increased expression of CA mRNA (Gehnrich et al., Bull. Mt. Desert Island Biol. Lab. 40:114-115, 2001). The increase in mRNA expression occurs at 24 hours after low salinity transfer and immediately precedes the initial increase in CA activity. Induction of CA activity is enhanced by removal of the major endocrine complex of the crab, the eyestalk. Eyestalk ablation (ESA) resulted in a doubling of CA activity in posterior gills in crabs acclimated to 32 ppt and not given a low salinity stimulus, and it potentiates the salinity-stimulated response by about 20% (Henry et al., Bull. Mt. Desert Island Biol. Lab. 39:21-22, 2000). Anterior gills served as control tissues since CA activity in this tissue does not respond to either low salinity or ESA.

These data served as the basis for hypothesizing the presence of a repressor substance in the eyestalks of C. maenas acclimated to high salinity, which inhibits CA expression and keeps CA activity at baseline levels. The effect of this putative repressor is either reduced or removed upon exposure to low salinity, allowing CA induction to occur. Furthermore, injection of extracts taken from the eyestalks of crabs acclimated to 33 ppt, inhibits CA induction by 50% in intact (untreated) crabs transferred from 33 to 10 ppt (Henry, Bull. Mt. Desert Island Biol. Lab. 40:35-36, 2001). To test this idea further, a series of experiments were designed to determine if the putative repressor was present in eyestalks of C. maenas after low salinity acclimation and if this putative compound was also present in a stenohaline, osmotic and ionic conforming species, Cancer irroratus.

C. maenas and C. irroratus were collected from Frenchman s Bay, maintained at 33 ppt in running seawater, and fed mussels. C. maenas were transferred directly to a 75 gallon recirculating aquarium of 10 ppt salinity either untreated (control) or after treatment with ESA. After 4 and 7 days post-transfer, CA activity was measured in anterior (G3) and posterior (G7) gills. At 4 days after transfer to low salinity, there was a doubling of CA activity in G7 and no change in G3 (Table 1). ESA enhanced this increase by about 50%. At 7 days post-transfer, CA -1 -1 activity in G7 was 1,241 umol CO2 mg protein min (a sixfold increase in this case), and this was potentiated by about 20% by treatment with ESA (Table 1).

In a second set of experiments, C. maenas were acclimated to 10 ppt for a period of 14 days. At that point, crabs were either left untreated (controls) or subjected to ESA. Both groups were left at 10 ppt for another 7 days and then assayed for CA activity in G3 and G7. There was no change in CA activity in G3, and there was no significant difference in activity in G7 between the control and the ESA-treated crabs (Table 1). ESA, which potentiates CA induction in the acute phase of low salinity acclimation, has no effect on CA activity after acclimation is complete. This suggests that whatever repressor is present in the eyestalk at high salinity is gone once the crab has acclimated to low salinity, as removal of the eyestalk has no further effect on CA activity.

In a third set of experiments, C. maenas acclimated to 33 ppt were then transferred to 10 ppt and given a daily injection of eyestalk extract taken from either C. maenas or C. irroratus, both also acclimated to 33 ppt. Crabs were injected through the arthroidial membrane covering the hemolymph sinus at the base of the walking leg. Two eyestalks were homogenized in 0.5 ml of filtered seawater and centrifuged at 10,000 g for 10 min at 4oC. Each crab was given a 400 ¼l injection of the supernatant immediately before transfer to 10 ppt and daily for a period of 7 days thereafter, at which point CA activity in G3 and G7 was measured. Eyestalk extract from C. maenas reduced CA induction by 50% compared to untreated crabs (Table 1). However, eyestalk extract from C. irroratus had no effect on salinity-stimulated CA induction in the gills of C. maenas.

Table 1. Effects of salinity transfer, eyestalk ablation, and injection of eyestalk extracts on CA activity in anterior (G3) and posterior (G7) gills of the green crab, Carcinus maenas.

Injections were performed daily for a period of 7 days. CA activity reported as umol CO2 mg protein-1 min-1. Mean + SEM (N). T = 12oC.

CA Activity Treatment Anterior (G3) Posterior (G7) 33 ppt acclimated 97 + 12 (11) 210 + 16 (11) 33-10 ppt transfer, untreated, 4 days 99 + 17 (6) 422 + 64 (6) 33-10 ppt transfer, ESA, 4 days 124 + 11 (6) 612 + 72 (6) 33-10 ppt transfer, untreated, 7 days 71 + 13 (6) 1,241 + 76 (6) 33-10 ppt transfer, ESA, 7 days 90 + 5 (7) 1,465 + 152 (7) 10 ppt acclimated, untreated, 7 days 88 + 13 (5) 1,983 + 155 (5) 10 ppt acclimated, ESA, 7 days 106 + 18 (4) 1,706 + 227 (4) 33-10 ppt transfer, untreated, Carcinus eyestalk injected 61 + 5 (10) 593 + 53 (10) 33-10 ppt transfer, untreated, Cancer eyestalks injected 123 + 4 (7) 1,244 + 93 (7)

These results confirm and extend the original hypothesis of the presence and action of a carbonic anhydrase repressor in the eyestalk of C. maenas. The effects of this substance are removed quickly after exposure to low salinity, as the greatest potentiation in CA induction occurs during the acute phase of low salinity acclimation. In green crabs acclimated to low salinity, the substance and/or its effects are absent, as evidenced by ESA having no effect on CA induction in crabs acclimated to 10 ppt. Additionally, this putative repressor appears to be absent from stenohaline crabs (e.g., C. irroratus) that do not possess the mechanism for CA induction in low salinity.

Supported by NSF IBN 97-27835.

EXPRESSION OF CRUSTACEAN HYPERGLYCEMIC HORMONE (CHH) mRNA IN NEUROENDOCRINE ORGANS OF THE SHORE CRAB CARCINUS MAENAS

Kristy Townsend1, Céline Spanings-Pierrot2, Daniel K. Hartline3, Shawna King4, Raymond P. Henry5, and David W. Towle6 1University of Maine, Orono 2Université Montpellier II, France 3University of Hawaii, Manoa 4Mount Desert Island High School, Bar Harbor, ME 5Auburn University, AL 6Mount Desert Island Biological Laboratory, Salsbury Cove, ME

Neurosecretory cells that produce crustacean hyperglycemic hormones (CHHs) have been identified in the sinus gland/X organ (SG) complex of the eyestalk, in the pericardial organ (PO), and in the gut of crabs. Two CHHs have been characterized at the molecular level in the shore crab Carcinus maenas, one form that predominates in the sinus gland/X organ and a second form found in the pericardial organ (Dircksen et al., Biochem. J. 356: 159-170, 2001). SG-CHH and PO-CHH are identical in the first 40 amino acids in the N-terminal region; the C- terminal 33 or 32 amino acids differ. cDNA sequencing suggests that a segment is spliced out of the SG-CHH precursor, thus altering the recognized translation stop site. SG-CHH is thought to regulate blood glucose, enzyme secretion by the hepatopancreas, and Y-organ ecdysteroid production but PO-CHH exhibits none of these properties (Dircksen et al., 2001).

Extracts of sinus gland and pericardial organ are known to enhance Na++K+-ATPase activity and uptake of Na+ in isolated perfused crab gills (Sommer and Mantel, J. Exp. Zool. 248: 272-277, 1988; Spanings-Pierrot et al., Gen. Comp. Endocrinol.119: 340-350, 2000). Strong evidence suggests that the active agent in sinus gland is a CHH-related peptide (Spanings-Pierrot et al., 2000) but it is not likely to be SG-CHH itself. Because of the potential regulatory role of CHHs with respect to gill ion transport, we have initiated a molecular study directed toward producing recombinant CHHs for in vitro studies with isolated gills. In this initial report, we examined the expression of CHHs in neurosecretory tissues of the shore crab Carcinus maenas, using reverse transcription and the polymerase chain reaction.

Table 1. Primer sequences and expected product sizes for amplification of crustacean hyperglycemic hormone cDNAs from sinus gland/X-organ (SG) and pericardial organ (PO) of the shore crab Carcinus maenas.

Primer Sequence (3’-5’) Expected size: SG Expected size: PO XPF2 cga ccg tgc tct gtt caa tga ctt 192 bp 317 bp XPR2 tcc tgc caa cca tct gta cct ttc PF3 aca ccg ata tta gga tca act gc None 455 bp PR3 cat tac ata agc tcg tta ccc aga PF4 gca gca ggg atg gac tta aag gat None 223 bp PR4 ttg gct tgt aga gtg atg gag ggt

Pericardial organs were identified by vital staining with methylene blue. Forming a coarse meshwork of vertical bars and horizontal trunks, the pericardial organ lies over the openings of the branchiopericardial veins, partially attached to the wall of the pericardial cavity. Sinus gland/X-organ complexes dissected from eyestalks and isolated pericardial organs were quickly immersed in guanidine isothiocyanate denaturing solution and total RNA extracts were prepared (Chomczynski and Sacchi, Anal. Biochem. 162: 156-159, 1987). Poly-A-containing mRNAs were reverse transcribed using oligo-dT and SuperScript II reverse transcriptase. The resulting cDNA mixture served as template in polymerase chain reactions that contained forward and reverse primers based on published CHH sequences from C. maenas (Dircksen et al., 2001) (Table 1).

Amplification of CHH cDNA from sinus gland/X-organ and pericardial organ by RT- PCR revealed a surprising outcome. Primers XPF2 and XPR2 that were designed to amplify both SG and PO CHH sequences did so successfully, as determined by gel electrophoresis (Fig. 1). Primer pairs PF3/PR3 and PF4/PR4 produced the expected-size products with PO as template, but also amplified a PO-like CHH from sinus gland/X-organ. These results suggested that SG contains the mRNA for both forms of CHH. Direct sequencing of the PCR products confirmed these speculations.

Template: SG PO PO PO SG SG Fig. 1. Amplification of CHH 1200 bp cDNAs from sinus gland/X- organ (SG) and pericardial organ 800 bp (PO) of the shore crab Carcinus maenas. Primer combinations are indicated below the 400 bp photograph of the ethidium- stained electrophoretic gel and refer to the primers listed in 200 bp Table 1. Size standards were electrophoresed in the left lane. Primers: XPF2 XPF2 PF3 PF4 PF3 PF4

XPR2 XPR2 PR3 PR4 PR3 PR4

Extracts of both sinus gland/X-organ and pericardial organ are known to stimulate Na+ uptake and Na++K+-ATPase activity of perfused gills (Sommer and Mantel, 1988; Spanings- Pierrot et al., 2000). Pericardial organ is shown here not to contain mRNA coding for SG-type CHH, but both the sinus gland/X-organ and pericardial organ contain mRNA coding for PO-type CHH. Our results, combined with earlier observations by other investigators, thus indicate that the ionoregulatory agent in both sinus gland and pericardial organ may be the “PO-type” CHH. Supported by a National Science Foundation grant and Research Experiences for Undergraduates Supplement (IBN-9807539) to DWT.

SODIUM-D-GLUCOSE TRANSPORT IN SQUALUS ACANTHIAS KIDNEY AND INTESTINE: FUNCTIONAL AND MOLECULAR DIFFERENCES

Thorsten Althoff1,2, Eva Kinne-Saffran2, Jutta Luig2, Hendrike Schütz2, and Rolf Kinne2 1Philipps-Universität Marburg, Marburg, Germany 2Department of Epithelial Cell Physiology Max-Planck-Institute for Molecular Physiology, Dortmund, Germany

In a variety of species it has been shown that D-glucose absorption in the kidney and intestine is mediated by sodium-D-glucose cotransport systems (SGLT). Interestingly, biodiversity of cellular functions does not require a large number of functionally different units; it can be managed by a limited number of closely related molecules like in the SGLT family. To date, there are two major members, SGLT1 and SGLT2. SGLT1 is characterized by a sodium to glucose stoichiometry of 2 Na+ to 1 D-glucose and a high affinity to D-glucose and galactose whereas SGLT2 has a one to one stoichiometry and a relatively low affinity for the sugars.

In functional studies using isolated brush border membrane vesicles we have previously obtained evidence that the sodium-D-glucose cotransporter present in the shark kidney belongs to the SGLT2 type transporter (Kipp, H. et al. Am. J. Physiol. 273:R134-R142, 1997). Further support for this assumption was provided by the successful cloning of a SGLT2 type transporter from this tissue (Kinne, R.K.H. at. al. unpublished data). The current investigation is aimed to define the molecular nature of the transporter responsible for absorption of sugar in the intestine of the shark.

As a first step we studied the tissue distribution of the renal SGLT2 in the shark by isolating mRNA, transcribing it into cDNA and measuring the number of copies of SGLT2 using specific primers and real time PCR. About 2000 copies of SGLT2 per nanogram of cDNA were found in shark kidney mRNA whereas no significant signal could be detected in mRNA isolated from intestinal mucosal scrapings. In order to exclude an accidental degradation of the mRNA during its isolation we also tested the samples for the presence of ß-actin mRNA. Using the appropriate primer for this message significant high signals for ß- actin were obtained both in renal and intestinal samples.

Therefore, we turned again to functional studies using brush border membrane vesicles isolated from intestinal mucosal scrapings in order to define the intestinal D-glucose transporter protein by determining its apparent affinity for D-glucose and its relative inhibition by D-galactose and D-xylose. These experiments showed that the apparent Km for D-glucose was about 70 times lower in the intestine than in the kidney (0.027 mM versus 1.9 mM) and the inhibition by D-galactose was 2.5 fold larger (73% versus 29% at 0.1 mM D- glucose and 5 mM D-galactose). There was also a significant difference in the inhibitory potency of xylose.

Taken together these results suggest that in the shark different transporters mediate the sodium dependent D-glucose absorption in kidney and intestine. In the kidney predominantly SGLT2 is involved whereas in the intestine - as in the skate (Althoff, Th. et al. unpublished data) and in many other species – a sodium-D-glucose cotransporter belonging to the SGLT1 type within the SGLT family facilitates the transport across the brush border membrane. ENRICHED PREPARATIONS OF PLASMA MEMBRANES FROM ZOOPLANKTON

Elizabeth L. Crockett and R. Patrick Hassett Department of Biological Sciences Ohio University, Athens, OH 45701

Zooplankton are a central link in marine foodwebs. Marine zooplankton are dominated by copepods (Class Crustacea), which are arguably ‘the most numerous multicellular organisms on earth’ (Mauchline, J., Adv. Mar. Biol. 33: 1-710, 1998). Knowledge of how temperature and nutrition affect the biology of these animals is key to understanding factors that limit zooplankton growth. In our ongoing study of how cholesterol and its sterol precursors may limit growth in zooplankton we plan to define the coupling between dietary and membrane-specific requirements for cholesterol. Cholesterol is an essential component of the plasma membranes of animals and is a requisite for animal growth (Siperstein, M.D. , J. Lipid Res. 25: 1462-1468, 1984). Since the vast majority of membrane-associated cholesterol is localized to the plasma membrane (Lange, Y., et al., J. Biol. Chem. 264: 3786-3793, 1989) it is necessary for us to have a working preparation of plasma membranes from zooplankton. The results presented herein characterize an enriched preparation of plasma membranes from copepods and the brine shrimp, Artemia franciscana.

Copepods (primarily Calanus finmarchicus, Pseudocalanus spp., and Temora longicornis) were collected in Maine from Frenchman Bay and offshore of the Damariscotta River. Partially sorted plankton samples (with larger zooplankton and detritus removed) were held in 10- and 20-liter containers in a flowing seawater tank until needed for membrane preparations. Water changes (50%) and food additions (300 ml from a batch culture of the diatom Thalassiosira weissflogii) were performed daily. Prior to preparing membranes, the copepods were concentrated and detritus, dead animals, and unwanted species were removed from the sample. A subsample was saved for estimating the composition of the sample, with weights estimated from literature values. Artemia franciscana were cultured in the laboratory at Ohio University at 20°C and 30 ppt salinity on an artificial, yeast-based diet (Roti-Rich, Florida Aqua Farms).

Density-gradient centrifugation was used to prepare an enriched fraction of plasma membranes. Whole animals were first homogenized in 9 volumes of 325 mM sucrose, 1mM EDTA, 25 mM HEPES (pH 7.6 at 25oC) with 3 passes in a 35 ml Potter Elvehjem homogenizer powered by a variable speed drill. Aliquots of the homogenate were kept on ice for marker enzyme analyses or frozen immediately in liquid nitrogen for subsequent analyses of marker enzymes (see below). After centrifuging for 10 minutes in fixed angle rotor (Sorvall SS-34) at 1,400gmax, supernatants were centrifuged for one hour at 93,000gmax (Beckman 70Ti). Pellets were recovered and resuspended in homogenization medium before layering onto 18% (v/v) Percoll (in a HEPES-buffered sucrose solution). A self-generated gradient was established at 37,000gmax for 25 minutes (Beckman 70Ti). The uppermost band in the gradient (corresponding to the lowest density) was recovered by pumping a saturated sucrose solution to the bottom of the centrifuge tube and collecting the band from the top. After resuspending the band, Percoll was removed and membrane material pelleted by centrifugation at 93,000gmax (Beckman 70Ti) for 60 minutes. The final pellet was resuspended in 25 mM HEPES with homogenization in a 2 ml Ten-Broeck ground glass homogenizer.

Activities of marker enzymes associated with plasma and intracellular membranes were measured according to standard protocols in order to determine enrichment factors for membrane fractions. No detectable activity could be measured for γ-glutamyltranspeptidase (copepods and Artemia) or alkaline phosphodiesterase (Artemia), two enzymes that have been used previously as markers for the plasma membrane. Plasma membrane markers 5’-nucleotidase (5’NT) and Na+/K+-ATPase (NKA), and the ER marker NADPH-cytochrome c reductase (NCCR) were assayed on material which had been previously frozen. Because of the lability upon freezing of the mitochondrial membrane marker cytochrome c oxidase (CCO), this marker was assayed on freshly prepared homogenates and membrane fractions the day of the preparation. Protein contents were determined using the bicinchoninic acid method.

Table 1. Enrichment factors and recoveries. Values are average ± 1 standard deviation. n = number of replicates. ______n 5'NT NKA CCO NCCR ______Enrichment (-fold) copepod 6 2.8 ± 1.2 8.5 ± 3.9 1.7 ± 0.4 2.6 ± 0.2 Artemia 4 4.2 ± 1.3 7.5 ± 0.9 no data no data

Recoveries (%) copepod 6 3.7 ± 1.5 10.8 ± 6.1 Artemia 4 4.5 ± 0.2 8.4 ± 1.7

Enrichment factors and recoveries of plasma membrane markers are similar for copepods and Artemia (Table 1). Although there is a positive and significant correlation between enrichment factors for NKA and 5’NT (Figure 1) enrichments for NKA are, on average, 3-fold greater (copepods) and 1.8-fold greater (Artemia) than enrichments for 5’NT. Similarly, recoveries are always greater for NKA than for 5’NT (Table 1). These results indicate 1) the membrane domains where the two enzymes are localized are different (and the preparation favors the recovery of the plasma membrane domain in which NKA is associated) and/or 2) there may be a soluble form of the enzyme as has been demonstrated in vertebrates (Zimmerman, H., Biochem. J. 285: 345-365, 1992). Intracellular membrane contamination of the preparation arises primarily from the endoplasmic reticulum. NCCR is enriched in the membrane fraction 2.6-fold while CCO is enriched less than 1.7-fold. 6 Figure 1. Enrichment of 5'nucleotidase vs + + 5 r=0.958, sig. p<0.05 Na /K -ATPase. Open 5 circles are preparations 1 4 of Artemia; closed circles represent copepods. Labels 3 r=0.825, sig. p<0.05 6 adjacent to copepod 2 symbols identify 2 4 preparation number. 3 5'Nucleotidase Enrichment (-fold) Lines fit by linear 1 regression. 4 6 8 10121416 Na+/K+-ATPase Enrichment (-fold)

The variability in enrichment factors for NKA is different in preparations made with either copepods or Artemia. Coefficient of variation (V) for the enrichment factor of NKA is 46% for copepods while V for Artemia is only 12%. Because our study requires plasma membrane preparations that are consistent we are currently trying to identify the source(s) of this variation. One possibility is differences in species composition among preparations. Preparations varied from a mixture of Calanus and Temora (preps 1,2, and 5) to predominantly Calanus (prep 6, 79% Calanus by weight) to predominantly Pseudocalanus (preps 3 and 4 , 70% and 87% respectively). Another factor is the condition of the copepods, as less active, and thus possibly damaged or otherwise unhealthy, copepods were removed during sorting to varying degrees among preparations. In the future we plan to modify our collection and sorting techniques to ensure consistency among preparations, which we anticipate will reduce the variability in NKA enrichment.

This research is supported by a research challenge grant from Ohio University and NSF OCE 0117132. FUNDULUS HETEROCLITUS 14-3-3.a INHIBITS AN ENDOGENOUS CHLORIDE CHANNEL IN XENOPUS LAEVIS OOCYTES

Devulapalli Chakravarty, Andrea Kohn, Robert Greenberg, and Dietmar Kültz The Whitney Laboratory University of Florida, St. Augustine, FL 32080

We have previously cloned a phospho-adapter protein called 14-3-3.a from Fundulus heteroclitus gill epithelium. Furthermore, we have shown that this protein and its corresponding gene are induced during transfer of fish from seawater to fresh water. 14-3-3 proteins bind to other proteins when these are phosphorylated on Serine or Threonine and are important modulators of such phospho-proteins and signaling complexes. Thus, we hypothesize that 14-3-3.a plays a key role in controlling the adaptive remodeling of fish gill epithelium that takes place during salinity change. Here we report that 14-3-3.a inhibits an endogenous Ca2+-activated chloride channel (CaCl) in Xenopus oocytes. 14-3-3.a cRNA was transcribed from pGEMT vector using T7 RNA polymerase. Fifty nanoliter of 14-3-3.a cRNA were injected into oocytes and the current generated by a calcium- activated chloride channel that is endogenously expressed in Xenopus oocytes was monitored by whole cell recording after a 48 h incubation period. A control group, in which Xenopus oocytes were injected with 50 nL water was analyzed in parallel to the 14-3-3.a-injected oocytes. The results of this experiment demonstrate that fish 14-3-3.a inhibits a chloride channel (Fig. 1).

Since a CFTR-type chloride channel mediates salt extrusion across gill epithelium in seawater fish and 14-3-3.a is down-regulated under such conditions we interpret these data as evidence for a potential role of 14-3-3.a in the regulation of salt transport across teleost gills. This work was supported by the Salisbury Cove Research Fund (DK) and NSF MCB-0114485 (DK). ACTIVATION OF SYK TYROSINE KINASE IS NOT REQUIRED FOR OLIGMOERIZATION OF LITTLE SKATE (Raja erinacea) ERYTHROCYTE BAND 3 Mark W. Musch1 and Leon Goldstein2 1The Martin Boyer Laboratories, Department of Medicine, The University of Chicago, Chicago, IL 60637 2Department of Molecular Pharmacology, Physiology, and Biotechnology, Brown University, Providence, RI 02912 Upon volume expansion, skate erythrocytes, like many other cells, swell and then undergo loss of cell solutes to accomplish a regulatory volume decrease (RVD). Water from the cell must accompany these effluxed solutes, thereby accomplishing the volume decrease. The solutes which are lost are numerous, but include electrolytes (K and Cl), sugar alcohols (sorbitol), and non-metabolized amino acids (taurine and _-alanine). The efflux pathways for these solutes may be numerous and are currently unresolved. Is skate erythrocytes, the anion exchange protein band 3 appears to play an important role in the formation/regulation of the volume activated efflux pathway for taurine. First, pharmacologic inhibition of band 3 with stilbenes inhibits swelling-activated taurine efflux. Secondly, species which lack band 3 (hagfish and lamprey) do not demonstrate a swelling-activated taurine efflux. In the skate erythrocyte, a number of biochemical events occur upon volume expansion which involve band 3. Notably, band 3, which may normally be present in the membrane as a dimer, is found to a much greater degree in a tetrameric form. Additionally, the interaction of band 4.1 with band 3 decreases greatly upon volume expansion and ankyrin binding to band 3 is of greater affinity. The cytoplasmic N- terminal domain of band 3 also becomes tyrosine phosphorylated to a greater degree, suggesting that tyrosine kinases may play a role in these events. Of note, two tyrosine kinases, syk and lyn, become activated in the skate erythrocyte after volume expansion and the syk inhibitor piceatannol nearly completely blocks the swelling-activated taurine efflux. The question arises whether syk activation is required for oligomerization of band 3. To address this, cells were pretreated with piceatannol and then volume expanded. Band 3 oligomerization was then determined in these cells using our previously published technique (JBC 269:19683-19686, 1994). Whereas only 10% of the band 3 in the isoosomtic erythrocyte is in the tetrameric form, nearly 40% is found so 5-15 min after volume expansion. This increase in tetrameric form of band 3 was not inhibited by preincubation with piceatannol, strongly suggesting that tyrosine kinase activity is not required for oligomerization. To determine if oligomerization is the signal for induction of tyrosine kinase activation was addressed by incubating cells under isoosmotic conditions with pyridoxal-5-phosphate or DNDS. These agents cause oligmerization of band 3 to a lesser extent than hypotonic exposure (resulting in volume expansion). Both PLP and DNDS were able to activate syk, demonstrating that oligomerization appears to be an important signal in activation of tyrosinekinases and subsequent steps in the activation of the taurine efflux. At present, we do not know those amino acid motifs in skate band 3 which mediate this effect, but as skate band 3 has been recently been cloned, mutagenesis studies and expression in a heterologous system may allow us to address this question.

Supported by NSF grant IBN-9974350 (LG) and DK47722 (MWM) EGG CAPSULE PROTEINS FROM FOUR SPECIES OF NORTHWESTERN ATLANTIC SKATES

Thomas J. Koob and Magdalena Koob-Emunds Skeletal Biology Section, Shriners Hospital for Children, Tampa, FL 33612

Oviparity has evolved at least four times in cartilaginous fish (Dulvy and Reynolds, Proc. Royal Soc. Lond. B 264, 1309-1315, 1997). All chimaeroids are oviparous, species in six orders of sharks are oviparous, and all species in the order Rajiformes, the skates, are oviparous. Common to all cartilaginous fish is an oviducal or shell gland that produces materials that encapsulate fertilized eggs. The basic morphology of shell glands is remarkably similar in cartilaginous fish, although the relative size and complexity depends on the reproductive mode (Hamlett et al., J. Exp. Zool. 282, 399-420, 1998). However, what comprises capsules in these diverse taxa is not known. In fact, capsule composition has been studied in only two species, Leucoraja erinacea (Koob and Cox, Environ. Biol. Fishes 38, 151-157, 1993) and Scyliorhinus canicula (Luong et al., Biochem. Biophys. Res. Comm. 250, 657-663, 1998). The present study was undertaken to compare the protein composition of egg capsules from four skate species.

Partially formed egg capsules were collected from Leucoraja erinacea (little skate), Leucoraja ocellata (winter skate), Malacoraja senta (smooth skate), and Amblyraja radiata (thorny skate). The recently secreted, untanned region of each capsule was solubilized in SDS/PAGE gel sample buffer containing β-mercaptoethanol (GSB). We have previously shown that this method completely solubilizes the untanned capsule proteins (Koob and Cox, Environ. Biol. Fishes 38, 151-157, 1993). The extracts were analyzed by electrophoresis on 4-20% linear gradient SDS/PAGE gels. In order to determine whether the mechanism responsible for capsule sclerotization involving catechol based polymerization similar in Leucoraja erinacea (Koob and Cox, op.cit.) occurs in other species, tyrosine and Dopa contents in specimens from a partially formed M. senta capsule were measured by amino acid analysis. Changes in tyrosine and Dopa contents were correlated with the extent of capsule polymerization by assessing solubility of the capsule proteins in GSB from specimens adjacent to those in which tyrosine and Dopa were measured.

Freshly oviposited egg capsules of the four species differ in size, form and color (Fig. 1). L. erinacea capsules are brownish green with golden hues. L. ocellata capsules are deep green. Both M. senta and A. radiata are dark brown. Differences in color indicate distinct chemistries associated with composition or the polymerization mechanism.

Egg capsules of the four species are composed of a similar array of structural proteins ranging in size from 50 to 5 kDa (Fig. 2). However, the apparent molecular weights and the relative amounts of the major proteins differ among the capsules. Common to three of the capsules (L.erinacea, M. senta and A. radiata) is a group of major proteins of relatively small size: a doublet at approximately 11 kDa and a smaller protein of lesser relative abundance. Other similarities include a 40 – 42 kDa protein present in all capsules and several proteins between 21 and 36 kDa. The protein composition of L. ocellata capsule differs significantly from that of the others. The major low molecular weight proteins (10 – 11 kDa) are absent or significantly larger (15 to 20 kDa). Leucoraja erinacea

Leucoraja ocellata

Amblyraja radiata Malacoraja senta

Figure 1. Egg capsules of the little skate (Leucoraja erinacea), winter skate (Leucoraja ocellata), thorny skate (Amblyraja radiata), and smooth skate (Malacoraja senta). The scale bar in each photograph represents 2 cm.

MW (kDa) 1234 lane 1. L. erinacea 2. L. ocellata 3. M. senta 4. A. radiata 55.4 –

36.5 – 31.0 –

21.5 – 14.0 – 6.0 –

Figure 2. SDS/PAGE analysis of egg capsule proteins solubilized from newly secreted, untanned egg capsule removed from the lumen of the shell gland in four skate species. To examine the polymerization mechanism in M. senta, amino acid compositions were obtained on 8 sequential specimens down the length of a partially formed capsule. Amino acid analysis showed that tyrosine contents at 0.54 moles/mole glycine were highest in the freshly secreted, untanned portion of the capsule (specimens 1 and 2; Fig. 3). In successive specimens, tyrosine content declined, reaching its lowest level (0.39 moles /mole glycine) in the oldest, tanning specimen (#8). Dopa was not detectable in untanned specimens 1 and 2. Dopa contents increased in successive specimens, reaching highest levels in specimen 8 (0.12 moles/mole glycine). There was a direct relationship between the loss in tyrosine and the increase in Dopa, indicating tyrosine was hydroxylated forming Dopa. The increase in Dopa correlated with a decrease in solubility of the egg capsule proteins (Fig. 3).

tyrosine Dopa

0.55 0.2

0.45 0.1 Dopa (moles/mole glycine) tyrosine (moles/mole glycine)

0.35 0 12345678 untanned Spec imen tanning

Figure 3. Tyrosine and Dopa contents in a partially formed egg capsule of Malacoraja senta. The dorsal wall of the capsule was sectioned into eight equivalent specimens from the newly secreted, white portion in the lumen of the shell gland (#1 specimen) to the oldest, tanning region (#8) in the uterus. Specimens were lyophilized, weighed, hydrolyzed in 6 N HCl at 108oC for 24 hr, dried in vacuo, and the amino acid composition determined. The amount of tyrosine and Dopa in each specimen was normalized to moles of glycine. The inset SDS/PAGE gel shows the relative solubility of the capsule proteins. Adjacent specimens of the capsule were placed directly in GSB containing β-mercaptoethanol, extracted for 24 hr, then electrophoresed on 4-20% SDS/PAGE and stained with Coomassie blue.

Further characterization of the egg capsule proteins in these species will be necessary to establish chemical and structural relationships. Nevertheless, these observations suggest similarities in several major structural proteins, particularly the small proteins which we have previously shown in L. erinacea capsules are unusual in that they are composed of 50% glycine and 20 –25% tyrosine. Differences in the protein composition of the capsules may reflect functional adaptation related to capsule properties. Current studies are underway to determine whether the mechanical properties, permeability, and durability in the marine environment differ. Despite the difference in protein composition, the polymerization chemistry appears similar in L. erinacea and A. senta, and therefore may be analogous in all skate egg capsules. Funded by the Shriners Hospitals for Children, project 8610.

RACEMASE ACTIVITIES IN THE TISSUES OF MARINE MOLLUSKS

Robert L. Preston1, Ishani Sud2, Velvet Williams2, Alexandra Budhai 3 and Amanda S. Bryant1 1Department of Biological Sciences Illinois State University, Normal, IL. 61790-4120 2Hillside High School, Durham, NC. 27707 3Brooklyn Technical High School, Brooklyn, NY. 11217

We have shown previously that D-amino acids occur in the free amino acid pools in tissues of marine invertebrates in diverse phyla (Preston, R.L. Comp. Bioch. Physiol. 87B: 55-62, 1987) and are transported and metabolized by invertebrate tissues (Preston, R.L. Comp. Bioch. Physiol. 87B: 63-71, 1987). The principal metabolic pathway interconverts D- and L-amino acids via equilibrative enzymes called racemases (Preston, R.L., et al., Bull. MDIBL 36: 86, 1997). We focused a substantial portion of our previous studies on the biochemical and molecular characterization of alanine racemase which commonly occurs in many species (Preston, R.L., et al., Bull. MDIBL 36: 86, 1997; Preston, R.L., et al., Bull. MDIBL 40: 116, 2001). In this paper we report the detection of a putative threonine racemase for the first time and as well as identifying the presence of alanine and serine racemases in three species of mollusk.

Animals used in these experiments were collected in Salsbury Cove, ME and maintained in running seawater aquaria. D-Amino acid content and racemase activity were measured using a coupled enzyme assay. Standard assay conditions used to detect neutral D-amino acids were: tissue extract (50 ml), tetrasodium pyrophosphate (NaPP) buffer saturated with the chromophoric peroxidase substrate, o-dianisidine (50 mM NaPP, pH 8.5; 110 ml), D-amino acid oxidase (0.06 units; 20 ml) and horseradish peroxidase (0.02 mg; 20 ml) and water, 50 ml. The reaction was run at room temperature in a 96 well microplate and the product formation (o-dianisidine dimer) measured at 490 nm using an ELISA plate reader (Bio-Tek Instruments, Winooski, VT). A variety of positive (containing added D- amino acid) and negative controls were also included. The absorbance was read every 15 min initially and at longer intervals for up to 10 hours. For all measurements 4 to 8 replicates were done. The racemase assay used a similar procedure except that 200 mM L-threonine, L-alanine or L-serine was added to the extract. This assay can detect neutral D-amino acids at concentrations from 50 mM or higher but the detection of racemases depends on stimulating D-amino acid formation from L-amino acids in excess of that detected in initial baseline measurements.

Table 1 shows the results from analysis of D-amino acid content and serine, threonine and alanine racemase activity in three species of mollusks. The “activity Ratio” shown in Table 1, [A490 each condition/A490 (extract only control)], compares the A490 of the control (extract only) condition to endogenous D-amino acid or racemase activity. A “Ratio” significantly >1.0 indicates the presence D- amino acid or racemase activity. In the slipper shell, Crepidula fornicata, both high concentrations of neutral D-amino acids (activity Ratio = 4.85) and vigorous alanine racemase activity was detected (activity Ratio = 14.6). It is usually the case that high D-amino acid content and vigorous racemase activity occur together but this is not always the case since the limit of detection of the coupled enzyme assay is about 50 mM and the concentrations present in some species may be below this limit (Preston, R.L., et al., Bull. MDIBL 40: 116, 2001). In the foot muscle of Stimpson’s whelk, Colus stimpsoni, significant D-amino acid content (activity Ratio = 1.81), and high serine racemase activity (activity Ratio = 5.74) was detected. In the gill tissue of the blue mussel, Mytilus edulis, D-amino acids were detected (activity Ratio = 4.5) as well as threonine racemase (activity Ratio = 8.25).

Table 1: Racemase activities in marine invertebrate tissue extracts*.

Species Cepidula fornicata Colus stimpsoni Mytilus edulis Whole foot muscle gill

A490/5hr** SE Ratio*** A490/5hr** SE Ratio*** A490/5hr** SE Ratio*** Extr. Only 0.041 0.002 0.031 0.001 -- 0.008 0.001 -- Total D-AA 0.199 0.008 4.85++ 0.056 0.002 1.81++ 0.038 0.001 4.5++ Ala Racemase 0.600 0.014 14.6++ ------Ser Racemase ------0.178 0.008 5.74++ ------Thr Racemase ------0.066 0.002 8.25++

* See text above for explanation of the coupled enzyme assay. All samples contained NaPP buffer pH 8.5 with saturating concentrations of the chromophoric reagent, o-dianisidine. Abbreviations: Extr. Only = tissue extract in NaPP buffer; Total D-AA = endogenous D-AA measured under conditions that do not stimulate racemase activity; Ala Racemase. = tissue extract plus 200 mM L-alanine; Ser Racemase. = tissue extract plus 200 mM L-serine; Thr Racemase. = tissue extract plus 200 mM L-threonine. A small amount of D-amino acid contamination is sometimes present in some commercially available L-amino acid reagents. Racemase measurements were corrected for this by subtraction of the mean of control values for wells containing only L- alanine, serine or threonine, as appropriate.. ** The absorbance at 490 nm was read against a NaPP blank for 197 to 591 min incubation period and the rate

(A490/5hr) was calculated. Long incubations are typically required for accurate detection of racemase activity and the 5 hr time period was a typical experimental time required.

*** Ratio = [A490 each condition/A490 (extract only control)]. The denominator compensates for the intrinsic color of the extract and the assay enzymes. An activity Ratio significantly >1.0 indicates the presence D-amino acid or racemase activity. Note: The symbol (++) indicates that a statistical comparison (t-test) of the absorbance values for each demarcated experimental condition compared with the respective control conditions (extract only) are significantly different from the control at the p< 0.05 or better (n = 4 or 8).

As far as we are aware this is the first report of threonine racemase activity in marine invertebrate tissues and in eukaryotic tissues. Alanine racemase is quite widely distributed in marine invertebrate tissues and only recently has serine racemase been detected in these organisms as well (Preston, R.L., et al., Bull. MDIBL 40: 117-118, 2001). Since threonine is a close structural analogue of serine, caution should be exercised in assuming this is a racemase different from the serine racemase, although our preliminary data suggests it is different. These data are also consistent with the fact that each amino acid appears to have a specific selective racemase as is the case in bacteria where racemases have been quite well characterized (Adams, E., Adv. Enzymol. Relat. Areas Mol. Biol. 44: 69-138, 1976). We are in the process of molecular characterization of the alanine racemase from Mya arenaria and plan to extend this sort of analysis to other racemases (Preston, R.L., et al., Bull. MDIBL 40: 116, 2001). (Amanda Bryant was an NSF-REU Fellow (DBI-9820400); Ishani Sud and Velvet Williams were Burroughs-Wellcome Fellows; Alexandra Budhai was a Hancock County Scholar funded in part by the SETH Program). LOCALIZATION OF NHERF-1, NHERF-2, AND CFTR IN THE RECTAL GLAND OF THE SPINY DOGFISH.

Edward J. Weinman and Deborah Steplock Department of Medicine, Department of Physiology, University of Maryland School of Medicine; Medical Service, Department of Veterans Affairs Medical Center, Baltimore, MD, 21201.

Recent studies in mammalian epithelial tissues have indicated that many membrane transporters, channels, and receptors do not exist as isolated proteins but rather are part of multiprotein complexes. Such complexes serve to link membrane proteins to reactive signaling molecules, cytoskeletal elements, and other effector proteins. In 1993, we reported the requirement for an additional protein co-factor in cAMP mediated inhibition of the renal brush border Na+-H+ exchanger (NHE3) (Weinman, E.J. et al., J. Clin. Invest. 92:1781-1786, 1993.). Subsequently, we isolated, cloned, and characterized a protein called the Na+-H+ Exchanger Regulatory Factor (NHERF-1) (also called EBP 50) (Weinman, E. J. et al., J. Clin. Invest. 95:2143-2149, 1995.). NHERF-1 is a modular protein containing two tandem protein-protein interactive PSD-95/Dlg/ZO-1 (PDZ) domains and an ezrin-radixin-moesin-merlin binding domain in the C-terminus. NHERF-1 functions to link together a signal-complex of proteins including ezrin, cAMP dependent protein kinase (PKA), and actin. This pre-formed signal-complex transduces cAMP signals in the renal proximal tubule by facilitating the phosphorylation of NHE3 and the down-regulation of its activity (Zizak, M.. et al., J. Biol. Chem. 274: 24753-24758, 1999.). A second member of this family of proteins was cloned from a yeast 2-hybrid screen using the C-terminus of NHE3 as bait and named NHERF-2 (also called E3KARP and TKA-1) (Yun, C.H..et al., Pro. Nat. Acad. Sci. USA 94:3010- 3015, 1997.). NHERF-2 also contains two PDZ domains and an ezrin-radixin-moesin-merlin binding domain. NHERF-2, like NHERF-1, functioned to facilitate the formation of a signal- complex of proteins that down-regulated NHE3 in response to cAMP in a heterologous expression system. Subsequent studies demonstrated that the NHERF proteins interact with a variety of transporters and ion channels, membrane receptors, transcription factors,and other signaling molecules. The potential biologic roles of NHERF-1 and NHERF-2 has been reviewed in detail in recent publications (Weinman,,E.J. et al., Am. J. Physiol. Renal Physiol. 279: F393-399, 2000., Shenolikar, S. et al., Am. J. Physiol. Renal Physiol. 280:F389-F395, 2001., Weinman, E.J. et al., J. Clin. Invest. 108: 185-186, 2001., Voltz, J.W. et al., Oncogene 20:6309-6314, 2001.).

While NHERF-1 and NHERF-2 share a number of similarities in their structure and in their putative physiologic roles, they are not identical. NHERF-1 is a phosphoprotein. In-vitro, it can be phosphorylated by PKA but in-vivo G Protein Receptor Kinase 6A and cyclin dependent kinase Cdc2 appear to be the endogenous kinases for this protein (Hall,R.A. et al., J. Biol. Chem. 274:24328-24334, 1999, He, J. et al., J. Biol. Chem. 276:41559-41565, 2001.). NHERF-2, on the other hand, has not been demonstrated to be a phosphoprotein. Moreover, the binding targets of the PDZ domains of both proteins are overlapping but not identical. Finally, recent studies in the rat kidney using antibodies that distinguished NHERF-1 from NHERF-2 indicated a discrete nephron distribution of the two proteins (Wade, J.B. et al., Amer. J. Physiol. Cell Biol. 280: C192- 198, 2001.). NHERF-2 is expressed in glomeruli, in the capillaries of the cortex and medulla of the kidney, and in principal cells. By contrast, NHERF-1 is expressed in the proximal convoluted tubules and in intercalated cells. In the rat kidney, no cells were identified that expressed both proteins. These finding suggested that each protein subserved a unique physiologic role in the kidney.

One of the more intriguing off-shoots of the isolation of the NHERF proteins was the recognition that it binds with high affinity to the Cystic Fibrosis Transmembrane Regulator (CFTR) (Wang, S. et al., FEBS Lett 427:103-108, 1998., Hall, R.A. et al., Proc Natl Acad Sci USA 95:8496-8501, 1998., Short, D.B. et al., J. Biol. Chem. 273:19797-19801, 1998.). It was initially proposed that NHERF functioned as a chaperon for CFTR from the golgi to the plasma membrane. More recent experiments have postulated that NHERF serves as a membrane retention signal for CFTR (Moyer, B.D. et al., J. Clin. Invest. 104:1353-1361.). It has also been suggested that NHERF facilitates the dimerization of CFTR and, as a consequence, full expression of chloride channel activity (Raghuram, V. et al. Proc. Natl. Acad. Sci. U S A. 98:1300-5.). These two proposal are not mutually exclusive and similar associations have been suggested to apply to the relation between other PDZ proteins and other transporters, channels, and receptors. To date, studies of the interaction between NHERF and CFTR have employed in-vitro binding assays and heterologous expression systems. Study of the relation between these two proteins in native tissues has proven difficult. The rectal gland of the spiny dogfish shark, Squalus acanthias, is a salt-secreting organ consisting of a morphologically homogeneous cell type containing a CFTR-like chloride channel in the apical membrane and a Na-K-2Cl transporter in the basal membrane (Forrest, J.N., Kidney Int. 49:1557-1562, 1996.). In response to agonists and hormones that increase intracellular cAMP, chloride secretion is increased over 10 fold in these glands (Lehrich, R.W. J. Clin. Invest. 101:737- 745, 1998.). Given the association between NHERF and cAMP signaling in the regulation of NHE3 in the mammalian kidney, and the association between CFTR and NHERF, we elected to study the potential role of the NHERF proteins in the shark rectal gland. In addition, we sought to determine if NHERF-1 or NHERF-2 co-localized with CFTR in this tissue.

Shark rectal glands were harvested from pithed male dogfish sharks and perfused in vivo using elasmobranch Ringer’s solution containing 270 mM NaCl, 4 mM KCl, 3 mM MgCl2, 2.5 mM CaCl2, 1 mM KH2PO4, 8 mM NaHCO3, 350 mM urea, 5 mM glucose, and 0.5 mM Na2 SO4 equilibrated to pH7.5 by bubbling with 99% O2 and 1% CO2.. Glands were perfused for 30 min in the absence or presence of 10 nM Vasoactive Intestinal Peptide (VIP). Control and experimental glands were then perfusion fixed for 20 to 40 min using a solution containing 3% sucrose, 4% paraformaldehyde, 300 mM Hepes, pH 7.2, and 0.1% glutaraldehyde. The perfusion fixed glands were prepared for confocal microscopy by preparing 8-10 :m sections from a region 2-3 cm from the distal tip of the gland. Sections were treated with either 1% SDS or 6M guanidine for 10 min to unmask antigenic sites. Similar localizations were obtained with either treatment. Sections were incubated with primary antibody overnight at 4o C. After washing, they were incubated with secondary antibody for 2 hr at 4o C. Appropriate species-specific antibodies coupled to Alexa 488 or 568 dyes (Molecular Probes, Eugene, OR) were diluted 1:100. Sections were examined with a Zeiss LSM410 confocal microscope.

CFTR antibody was obtained from Chemicon International, Inc. This polyclonal antibody was generated from a peptide fragment representing amino acids 386 to 412 of human CFTR, a region conserved in shark CFTR. NHERF-1 antibody was a polyclonal antibody to full-length recombinant rabbit NHERF-1. The NHERF-2 antibody was generated in chicken to amino acids 271 to 285 of human NHERF-2 encompassing a region where NHERF-1 and NHERF-2 are most dissimilar. Preliminary Western immunoblots using lysates of shark rectal gland indicated that each antibody identified a single band representing a protein of appropriate size; CFTR=160-180 kDa, NHERF-1=55 kDa, and NHERF-2=50 kDa..

NHERF-2 was found in relatively high abundance in cells lining the sinusoidal spaces (Figure 1). No staining was seen in the tubular cells. Treatment of the glands with VIP did not alter the distribution of NHERF-2 (not shown). By contrast, NHERF-1 was localized to the apical membrane of the rectal gland cells (Figure 2, panel A). It was not detected in the basolateral side of the rectal gland cells or in the vascular spaces. As previously reported, CFTR was expressed in the apical membrane of these cells (Figure 2, panel B). In cells that labeled positive for CFTR, NHERF-1, but not NHERF-2, was detected in a similar distribution (Figure 2, panel A). Merging of the images, however, suggested that while NHERF-1 and CFTR were present in the apical membrane, they did not co-localize in unstimulated glands. By contrast, in cells treated with VIP for 30 min, the merged color images indicated co-localization of NHERF-1 and CFTR.

Thus, these experiments provide preliminary evidence that proteins recognized by antibodies to mammalian NHERF-1 and NHERF-2 Figure 1. react with proteins of similar size in shark rectal gland. The tissue distribution of NHERF-1 and NHERF-2 in shark rectal gland differ. NHERF-2 is distributed to cells that line the vascular sinusoidal spaces; perhaps representing

Figure 2. endothelial like cells. This result would appear to be comparable to the expression of NHERF-2 in the vascular structures of the rat kidney. NHERF-1, on the other hand, was detected in the apical membrane of the rectal gland cells. This pattern of distribution is also mirrored in the rat kidney where NHERF-1 is expressed in the brush border membrane of the proximal convoluted tubule. Although NHERF-1 and CFTR were present in a similar apical membrane location in control rectal glands, the merged color images did not indicate co-localization of the two proteins. By contrast, after agonist stimulation, CFTR and NHERF-2 co-localized by immunocytochemical analysis.

These experiments did not address directly the physiologic role of the NHERF proteins in the shark rectal gland. The results, however, are consistent with the interpretation that CFTR and NHERF-1 reside in separate microdomains in or near the apical membrane of the rectal gland cells in the basal state and that cAMP induces movement of CFTR to a site where it associates with NHERF-1. The movement of CFTR from a subapical location to the membrane would be consistent with the findings of Forrest and co-workers ( (Lehrich, R.W. J. Clin. Invest. 101:737-745, 1998.). By analogy to mammalian studies, it would be postulated that when CFTR is translocated to the apical membrane, NHERF-1 serves to retain and/or correctly orient the CFTR channels to maximally facilitate the secretion of chloride. Clearly, additional studies will be required to fully understand the role of NHERF-1 in the trafficking and function of CFTR. The role of NHERF-2 in endothelial cells is current unknown. These preliminary studies, however, indicate that the shark rectal gland would be a useful tissue for study of the NHERF proteins and the relation between NHERF-1 and CFTR.

The authors appreciate the intellectual input and technical assistance of Drs. John Forrest and Patricio Silva as well as the members of their laboratories. Drs. James B. Wade and Shirish Shenolikar also provided valuable input. This work was supported by grants from the National Institutes of Health, DK 55881, Research Service, Department of Veterans Affairs, and by a New Investigator Award from the Salisbury Cove Research Fund.

CALCIUM TRANSIENTS DURING THE FIRST AND SECOND MITOTIC DIVISIONS OF THE FERTILIZED MEDAKA EGG

H. Criss Hartzell 1 and Richard N. Winn 2 1 Department of Cell Biology, Emory University School of Medicine, Atlanta, Ga 30322-3030 2 Aquatic Biotechnology & Environment Laboratory, University of Georgia, Athens, GA 30602

Recently, considerable evidence has emerged that calcium ions play an important role in the regulation of mitosis (e.g.; Whitaker M. Larman MG. Calcium and mitosis. Seminars in Cell & Developmental Biology 12, 53-58: 2001). Cell cycle changes in cytosolic free Ca were first observed in 1986 by Roger Tsien’s laboratory (Science 233, 886-889: 1986), where it was shown that Ca rises at the onset of anaphase. These and other studies have led to the hypothesis that calcium transients regulate the progression of the cell cycle. However, the hypothesis remains incompletely investigated and not universally accepted because calcium transients are not always observed during mitosis, especially in mammalian cells in culture. To obtain more insight into the role of calcium in cell cycle, we examined calcium transients dividing eggs of the Medaka fish (Oryzias).

Fertilized eggs prior to the first mitotic division were collected from female Medaka fish early in the morning immediately after fertilization. The eggs were microinjected with ~ 1 nl of the fluorescent Ca-sensitive dye fluo-4 (10 mM of the free acid in Ca-free water). The microinjection pipet was pulled from 1.5 mm diameter glass capillary and injection was controlled by air pressure delivered to the back of the pipet. Injection was monitored on a fluorescent dissecting microscope. The eggs were then oriented on the stage of an Olympus confocal microscope and Fluo-4 fluorescence was imaged in a plane predicted to be perpendicular to the plane of the first cleavage furrow. Images were recorded once very minute.

The figure below plots the pixel intensity measured in a region of cytosol about 10 mm below the plasma membrane parallel to the cleavage furrow. “Begin” and “end” refer to the

160

140

120

100

80

Pixel Intensity 60

40 begin end

20 0 20 40 60 Time (min) onset completion of formation of the cleavage furrow. Approximately 10 min prior to the onset of the first cytokinesis (which probably corresponded to the onset of anaphase), the cytosolic Ca concentration began to rise and Ca oscillations were recorded. Initially, the oscillations were not localized to the cleavage furrow, but were global in nature. As cytokinesis progressed, the oscillations localized to the site of the cleavage furrow. Ca concentration at the cleavage furrow remained elevated for 10-15 minutes after cytokinesis was completed, but during this time Ca oscillations were smaller in amplitude. After 15 min, the Ca concentration returned to basal levels similar to those recorded at prophase.

In several cases, we were able to observe the eggs through 2 or more cycles of cell division. Interestingly, the increase in Ca at the cleavage furrow was most dramatic during the first division and was either greatly diminished or absent in subsequent divisions. This suggests that local Ca transients may play a role in the determination of the axis of embryo. This may explain why Ca transients are not always observed during the cell cycle: perhaps the Ca transients are not essential for cell cycle progression but are important in determining the axis of cytokinesis. Additional studies are planned to investigate these questions.

Supported by a New Investigator Award from MDIBL and NIH grant GM60448. RETENTION OF STRUCTURAL AND FUNCTIONAL POLARITY IN CULTURED SKATE (RAJA ERINACEA) HEPATOCYTES

Ned Ballatori1, David J. Seward1, Gert Fricker2, Maria Runnegar3, John H. Henson4, David S. Miller5, and James L. Boyer6 1Dept. of Environ. Med., University of Rochester School Medicine, Rochester, NY 14642 2Institut fur Pharmazeutische Technologie und Biopharmazie, D-69120 Heidelberg, Germany 3Department of Medicine, University of Southern California, Los Angeles, CA 90033 4Department of Biology, Dickinson College, Carlisle, PA 17013 5Laboratory Pharmacology and Chemistry, NIH/NIEHS, Research Triangle Park, NC 27709 6Dept. of Medicine and Liver Center, Yale Univ. School of Medicine, New Haven, CT 06520

Isolated liver parenchymal cells in culture have proven to be invaluable for addressing a multitude of biochemical, physiological, and toxicological questions (Alpini et al., Hepatology 20:494-514, 1994). Hepatocytes have been isolated from many species, including marine and freshwater fish (Baski and Frazier, Aquatic Toxicol. 16:229-256, 1990). Cultured hepatocytes offer a number of advantages over in vivo or isolated perfused liver models, including elimination of systemic effects, better control of environmental conditions, reduced variability between experiments, simultaneous and repeated sampling in a single experiment, and more efficient use of animals, reagents, and time. Short-term cultured hepatocytes closely mimic the functions of cells in the intact liver and provide a useful model for studying acute effects of drugs and xenobiotics, whereas primary monolayer cultures allow for measurement of long-term effects (days to weeks). However, one drawback of most hepatocyte culture models is that the cells begin to dedifferentiate and lose transport polarity immediately after isolation. In contrast, our previous experience with short-term cultures of hepatocytes isolated from the little skate indicates that these cells largely retain hepatobiliary polarity for at least 8 h (Miller et al., Am. J. Physiol. 270:G887-G896, 1996; Smith et al., J. Exp. Zool. 241:291-296, 1987; Henson et al., J. Exp. Zool. 271:273-284, 1995). When compared with mammalian hepatocytes, skate hepatocytes are easier to isolate and maintain in short-term culture, and they exhibit high viability, stability, and maintenance of normal hepatocyte functions (Smith et al., J. Exp. Zool. 241:291-296, 1987; Henson et al., J. Exp. Zool. 271:273-284, 1995). Skate hepatocytes have been used to evaluate mechanisms of hepatic metabolism, detoxification, membrane transport, and cell volume regulation (e.g., Ballatori et al., Toxicol. Appl. Pharmacol. 95:279-291, 1988; Am. J. Physiol. 267:G285-G291, 1994; Mol. Pharmacol. 48:472-476, 1995; Ballatori and Boyer, Am. J. Physiol. 262:G451-G460, 1992).

The goal of the present study was to develop conditions for long-term culture of skate hepatocytes that allow for the retention of membrane polarity. Skate hepatocytes were isolated by a mild collagenase perfusion technique and incubated at 12oC in serum-free elasmobranch Ringer solution. Pilot experiments compared some enriched culture media, and in particular a modified DMEM/F12 medium, to a simple elasmobranch Ringer solution (both media supplemented with 0.1 mM dexamethasone and penicillin/streptomycin, 100 U: 0.1 mg/ml). Surprisingly, only small differences in the overall preservation of metabolic and structural integrity were found. However, the enriched medium had one major shortcoming, it was highly conducive to the growth of bacteria and other parasites that were occasionally present in the livers of these wild animals and that were inadvertently co-isolated with the hepatocytes. In contrast, these organisms were unable to thrive in elasmobranch Ringer solution containing penicillin/streptomycin. Thus, all subsequent studies were carried out using elasmobranch Ringer as the culture medium.

Isolated skate hepatocytes contained numerous lipid-filled vesicles, which presumably provide their major source of metabolic energy, and most of cells were present as clusters of 3-12 hepatocytes surrounding a bile canaliculus, rather than as single cells. The number of single cells in the preparation could be increased by lengthening the time of collagenase perfusion. These cells and cell clusters readily attached to untreated plastic culture dishes, and formed a loose, network-type three dimensional structure when dexamethasone was present in the medium. However, attachment to the culture dishes was only moderately strong, particularly for the larger hepatocyte clusters, given that only a small fraction of the total surface area of these clusters was in contact with the surface of the dish. In contrast to plastic dishes, attachment to glass cover slips was weak, although it was markedly improved by coating the cover slips with poly-L- lysine, or with type I or type IV collagen. The effects of the collagen or poly-L-lysine coatings was transient, however, as the cells gradually detached from treated glass over 2-4 days in culture.

Trypan blue and propidium iodide exclusion were measured on days 0-7 of culture, and were found to be >98%. During this time interval the cells maintained high intracellular concentrations of K+, ATP and reduced glutathione (GSH), and high ratios of ATP/ADP and GSH/GSSG. Cultured skate hepatocyte clusters also maintained a relatively normal morphology and a polarized distribution of apical and basolateral membrane domains. Polarity was confirmed both morphologically, using antibodies raised against the skate canalicular bile salt transporter, Bsep, and antibodies to the canalicular multispecific organic anion transporter, Mrp2, and functionally, by monitoring secretion of the fluorescent organic anions NBD-taurocholate, a Bsep substrate, and fluorescein-methotrexate, an Mrp2 substrate, into the bile canalicular spaces. Quantitative RT-PCR analysis revealed that mRNA levels of the hepatic transporters Oatp, Bsep, and Mrp2 declined only slowly over the 7-day period to levels of 70, 35 and 40% of initial values, respectively. Examination of microtubules, actin filaments and cytokeratin filaments by immunofluorescence microscopy revealed that the cytoskeleton maintained a largely polarized distribution during the culture period.

These studies indicate that in contrast to mammalian hepatocytes, isolated skate hepatocyte clusters retain polarity of structure and function in culture, and should provide an excellent system for investigating long-term effects of drugs and xenobiotics on hepatobiliary functions. (Supported by National Institute of Health Grants ES03828, ES01247, DK34989, DK25636, DK48823, NSF DBI 9820400, and by the Burroughs Wellcome Foundation). b-ACTIN mRNA EXPRESSION IS MARKEDLY UPREGULATED WHEREAS ITS PROTEIN LEVELS ARE UNCHANGED IN PRIMARY HEPATOCYTE CULTURES FROM THE LITTLE SKATE, RAJA ERINACEA

David J. Seward1, Rachel E. Anderson2, Catherine S. Bennet3, Jesse McCoy4, Shi-Ying Cai5, James L. Boyer5, Ned Ballatori1 1Dept. of Environmental Medicine, Univ. of Rochester School of Med., Rochester, NY 14642 2Harvard College, University Hall, Cambridge, MA 02138 3Choate Academy, Wallingford, CT 06492 4Durham H.S., Durham, NC 5Dept. of Medicine and Liver Center, Yale University School of Med., New Haven, CT 06520

b-Actin is widely considered cell biology’s quintessential housekeeping gene. This protein, a ubiquitous structural component of the cytoskeleton, is constitutively expressed in all cell types. Its expression levels within a particular cell type are stable, predictable and internally consistent and can thus be used to normalize RNA and protein data for other genes. To assess whether b-Actin might be used as a housekeeping gene in our primary skate hepatocyte cell culture model we measured its mRNA and protein levels over time in culture.

Hepatocytes from male skates (~1kg) were isolated by mild collagenase perfusion and cultured on standard plastic 6-well plates in a simple elasmobranch Ringer solution (containing in mM, 270 NaCl, 4 KCl, 2.5 CaCl2, 3 MgCl2, 0.5 Na2SO4, 1 KH2PO4, 8 NaHCO3, 350 urea, 5 D-glucose and 5 Hepes/Tris, pH 7.5), supplemented with dexamethasone (0.1 mM) and penicillin/streptomycin (100 U: 0.1 mg/ml). Cells were kept in culture for 7 days with total RNA and protein samples isolated on days 0, 1, 3 and 7. Total RNA for each time point was isolated using Qiagen’s RNeasy Mini Kit for RNA isolation from tissue culture. Preceding quantitative RT-PCR analysis, total skate hepatocyte RNA, 2 mg, was subjected to DNase I treatment using GibcoBRL Life Technologies Deoxyribonuclease I, Amplification Grade. Quantitative RT-PCR was conducted on a Rotor-Gene 2000 Real-Time Cycler employing Taqman chemistry and GibcoBRL Life Technologies SuperScript One-Step RT-PCR with Platinum Taq. A partial skate b-Actin sequence was cloned by degenerative PCR. Gene specific primers and Taqman probes were designed to that skate b-Actin sequence using Perkin-Elmer’s Primer Express computer software. mRNA levels for the b-Actin gene were assessed by measuring the copy number in 10ng of total cell culture RNA. Day 0 values were assigned a value of 1 and successive time points were expressed as a ratio to day 0 values. Total cell protein was isolated via acid precipitation with homogenization buffer containing 5% PCA, 1mM EDTA, 1mM PMSF, and mammalian protease inhibitor cocktail 5ml/ml (Sigma P-8340). Cells were lysed completely in the homogenization buffer and then spun at high speed for 1 minute to pellet protein. The protein pellet was resuspended in a buffer containing 10 mM Tris (pH 7.4), 1 mM Na- orthovanadate, 0.1 mM leupeptin and 10% SDS. Final protein concentrations were determined using the Lowry protein assay. Protein samples were prepared for SDS-PAGE analysis by boiling in BioRad’s Laemmli Sample Buffer containing b-mercaptoethanol. 50mg of total protein were separated on BioRad precast Tris-HCl polyacrylamide gradient gels (4%-20%) for 2 hours at 90-100V. The proteins were transferred to PVDF membranes where they were exposed to an anti-Actin monoclonal antibody (ICN, Clone: C4, cat# 691002). This antibody reacts with a conserved N-terminal region found in all actin monomers (Lessard, J.L., Cell Motility and the Cytoskeleton, 10: 349-362, 1988). Reactive bands were visualized using an anti-mouse IgG1 secondary antibody conjugated to HRP followed by chemiluminescence detection. Blots were analyzed and relative protein levels determined using a Kodak Image station. As with the mRNA data, day 0 protein values were assigned a value of 1 and successive time points were expressed as a ratio to day 0 values.

b-Actin mRNA expression increased 2.0±0.3 fold from day 0 to day 1. By day 3 levels had elevated to 3.8±0.5 fold over day 0 controls. On day 7, b-Actin mRNA levels had risen 6.4±1.2 fold over initial levels. Interestingly, these changes in transcript expression appeared to have no effect on b-Actin protein levels.

Figure 1. Relative -Actin b 8 Expression in Primary Skate Hepatocyte Cultures Over Time. 7 b-Actin mRNA Total RNA and protein from skate hepatocytes in culture were isolated 6 b-Actin Protein over a 7-day time course. Using gene specific Taqman probes and Expression 5 primers, relative mRNA expression 4 levels for b-Actin were calculated -Actin for each time point. By day 7, b- b 3 Actin mRNA levels were 6.4±1.2 fold higher than those measured on 2 day 0. Protein expression levels Relative remained relatively constant over 1 the 7-day culture period. All 0 values are mean±standard error. 0 1 Days 3 7

The reasons for the marked change reported in b-Actin mRNA expression are not fully understood. A recent paper by Otsu et al. (Cell Physiol. Biochem. 11:33-40, 2001) reports a similar phenomenon in primary rat hepatocyte cultures. Their data indicate a general trend in which liver specific gene mRNAs are downregulated while constitutively active genes, such as b-Actin, are upregulated. Their work suggests that the absence of an extracellular matrix could play an integral role in both the transcriptional regulation as well as the overall stability of gene transcripts. Our observations support these previous findings; however, because much of the skate genome has yet to be cloned we cannot assess whether this is a general trend in the expression of constitutive versus liver specific genes. Furthermore, it is unclear whether or not the increase in b-Actin transcript level is due to an increase in transcriptional activity or simply to an increase in transcript stability. The data presented by Otsu et al. in rat hepatocytes suggest the transcription rate for b-Actin is stable and as a result they speculate that increased transcript stability is the most likely explanation for the observed increases in b-Actin mRNA in their experiments. Despite the observed increases in b-Actin transcript in our skate hepatocyte RNA, western blot data indicate that b-Actin protein levels remain relatively constant. Thus, although b-Actin may still be useful for normalizing protein data, its role as a benchmark for evaluating changes in mRNA expression in primary culture models may require reassessment. (Supported by National Institute of Health Grants ES03828, ES01247, DK34989, DK48823, NSF DBI 9820400, and by the Burroughs Wellcome Foundation). TRAFFICKING OF CFTR IN KILLIFISH (FUNDULUS HETEROCLITUS) OPERCULAR MEMBRANE: RESPONSE TO ELEVATIONS IN CAMP AND ADAPTATION TO SEAWATER

Alexander Lankowski1, Maude Emerson2, John E. Mickle3, John R. Riordan4, Garry R. Cutting3 and Bruce A. Stanton1 1Department of Physiology, Dartmouth Medical School, Hanover, NH 03755 2Groton School, Groton, MA 01450 3Institute of Genetic Medicine, Johns Hopkins University School of Medicine, Baltimore, MD 21287 4Mayo Graduate School of Medicine, Mayo Clinic, Scottsdale, AZ 85259.

CFTR is a cAMP-activated Cl channel that mediates transepithelial Cl transport in a variety of mammalian epithelia including airway, intestine, pancreas, sweat duct and the opercular membrane of killifish. It is well accepted that stimulation of CFTR-mediated Cl secretion by cAMP involves protein kinase A-mediated phosphorylation of CFTR. In addition, in some but not in all epithelial cells, cAMP also stimulates CFTR-mediated Cl secretion by stimulating the insertion of CFTR into the plasma membrane from a cytoplasmic vesicular pool (Jilling, et al., Int Rev Cytol. 172:193-241, 1997). However, with the exception of studies in shark rectal gland (Lerich et al., J. Clin. Invest. 101:737-745, 1998), studies on CFTR trafficking have been conducted in immortalized cells in culture and/or heterologous cell lines over-expressing CFTR. The biological significance of CFTR trafficking in these models is unknown. Accordingly, the goals of this project were to: (1) Develop an in vivo model to study CFTR trafficking and (2) Test the hypothesis that physiological stimuli enhance CFTR mediated Cl secretion by inducing the insertion of CFTR into the plasma membrane from a cytoplasmic vesicular pool. To this end, studies were conducted on killifish opercular membrane, an epithelium that expresses high levels of endogenous CFTR (Marshall and Bryson. Comp. Biochem. Physiol. 119A(No1):97-106, 1998, Singer et al., Am J Physiol. 274:C715-23, 1998). Elevations in cAMP and an increase in the salinity of the swim water stimulate CFTR mediated Cl secretion by the operculum by an undefined mechanism.

Killifish were collected from Northeast Creek (Bar Harbor, ME) and held in aquaria containing running seawater at the MDIBL (August, 2001) or artificial seawater (Instant Ocean) at the Dartmouth Medical School (September, 2001 to January, 2002). Fish were kept in seawater for at least two weeks to acclimate. Adaptation to fresh water was achieved by gradually (~1 hr) reducing the aquaria salinity to 10% and maintaining the fish in 10% seawater for two weeks. Subsequently, the water was changed gradually (~1 hr) to dechlorinated tap water. Fish were maintained in fresh water for at least two weeks before the fish were returned to 100% seawater. The amount of CFTR in the plasma membrane of the opercular membrane was determined by cell surface biotinylation and SDS-PAGE (Loffing et al. Am. J. Physiol. 275: C913-920, 1998) using a monoclonal antibody that recognizes shark and killifish CFTR (60.1.2: J. Riordan, unpublished). Opercular membranes were isolated and mounted in Ussing chambers and the CFTR Cl current was measured as described (Ernst, et al., J. Cell Biol. 87:488-497, 1980). Similar results were obtained at the MDIBL and Dartmouth. In fish adapted to 100% seawater, isoproterenol (10 mM) increased CFTR mediated Cl secretion across opercular membranes from 153 to 237 mA/cm2 (15 minutes) and dramatically increased the amount of CFTR in the plasma membrane by 4.1-fold as determined by western blot analysis and densitometry (Figure 1). A cocktail of compounds that increase cAMP levels (db-cAMP, 500 mM; IBMX, 0.1mM and Forskolin, 1 mM) elicited a similar result. A chronic change in salinity from 100% seawater to freshwater reduced CFTR protein expression in cell lysates and the plasma membrane and reduced CFTR Cl secretion to ~0 mA/cm2. 24 hours after the aquaria water was changed from fresh to 100% seawater CFTR Cl currents increased from ~0 to 93 mA/cm2 and the amount of CFTR in cell lysates and the plasma membrane increased dramatically. Taken together, these data demonstrate that the opercular membrane of killifish is an excellent in vivo model to study CFTR trafficking and that physiological stimuli enhance CFTR mediated Cl secretion, in part, by inducing the insertion of CFTR into the plasma membrane from a cytoplasmic vesicular pool. (Supported by a MDIBL New Investigator Award to B.A.S., a NSF REU program DBI 9820400 award to M.E., a MDIBL New Investigator Award to J. E. M, and a Cystic Fibrosis Foundation award to A. L.).

Figure 1. Isoproterenol stimulates CFTR- mediated Cl secretion in killifish opercular membranes, in part, by increasing the amount of CFTR in the plasma membrane. Representative Western blot of CFTR in the plasma membrane (lanes 1 and 2) and whole cell lysates (lanes 3 and 4) in killifish opercular membrane treated with isoproterenol (Iso: 10-5 M) for 15 minutes or vehicle (Con). CFTR was detected with monoclonal antibody 60.1.2 (1:1,000) and an anti-mouse HRP secondary antibody (1:5,000) by enhanced chemiluminescence (ECL: Pierce). Similar results were obtained in a total of three experiments. In each experiment the left opercular membrane was treated with isoproterenol and the right opercular membrane of each fish was treated with vehicle (four fish/experiment). 50 mg of protein was loaded in each lane. Overexposure of the blot reveals CFTR in the plasma membrane of control membranes (not shown). CFTR AND THE MYOSIN V MOTOR PROTEIN ARE ON THE SAME TRANSPORT VESICLES IN THE RECTAL GLAND OF SQUALUS ACANTHIAS

Franklin H. Epstein,1 Patricio Silva,2 Katherine Spokes,1 and Richard M. Hays3 Departments of Medicine, 1Harvard Medical School and Beth Israel Deaconess Medical Center, Boston MA 02215, 2 Temple University Hospital, Philadelphia, PA 19140, and 3Albert Einstein College of Medicine, Bronx, NY 10461

Chloride secretion by the shark rectal gland is sharply increased by vasoactive intestinal peptide (VIP) and C-type natriuretic peptide (CNP). These agents act partly or wholly by increasing chloride transport through the cystic fibrosis transmembrane conductance regulator (CFTR) in the apical membrane of the rectal gland epithelial cell. The increase in chloride secretion produced by CNP is significantly reduced by cytochalasin D (Silva, P. et al., Bull. Mt. Desert Island Bio. Lab. 40: 27-28,2001), indicating that an intact actin cytoskeleton is required for the activation of chloride transport by this hormone.. This finding, in addition to a shift in immunostaining of CFTR towards the apical membrane in glands stimulated by cAMP agonists or genistein (Lehrich, R.W., et al., J. Clin. Invest. 101: 737-745, 1998), supports the view that trafficking of CFTR to the apical membrane is the major mechanism involved in the stimulation of chloride secretion. In the case of CNP, trafficking would appear to take place along subapical actin filaments.

The "motors" for vesicle movement along actin filaments are members of the family of unconventional myosins (Wu, X. and Hammer, J. A. Curr. Opin. Cell Biol.12: 42-51, 2000). Of the unconventional myosins, myosin V is a leading candidate for vesicle transport because of its long "legs", which facilitate processive movement along the actin helix (Walker, M.L. et al., Nature 405: 804-807, 2000).

We have employed immunogold electron microscopy to identify subapical vesicles carrying CFTR in rectal gland epithelial cells, and a double immunogold labeling protocol to determine whether CFTR-carrying vesicles also have myosin V motors. Rectal glands were fixed by perfusion with 4% paraformaldehyde, 0.1% glutaraldehyde, and sections were cut from the gland and stored in shark Ringer-paraformaldehyde-glutaraldehyde solution at 4 degrees. They were embedded in epon and sectioned at the Analytic Imaging Center at the Albert Einstein College of Medicine. The etched epon method, an antigen retrieval method, which optimizes visualization of structure and tissue antigenicity (Rocken, C., and Roessner, A. J. Histochem. Cytochem. 47: 1385-1394, 1999) was then employed on the sections. Etching was done with sodium ethoxide, and the sections were then microwaved in sodium citrate buffer. A monoclonal anti-shark CFTR antibody (60.1.2) provided by Dr. John Riordan was used to detect CFTR, and a polyclonal anti- myosin V antibody provided by Dr. John A. Hammer was used to detect myosin V in the sections.

Fig. 1 shows heavy arrays of actin filaments running from the apical microvilli to the terminal web, and a vesicle whose wall contains CFTR. Fig. 2 shows myosin V in Western blots of lysates of rectal gland and Hela cells. Figures 3 and 4 show vesicles doubly labeled for CFTR (15 nm gold) and myosin V (10 nm gold). 1 2 Fig. 1 Etched epon section showing actin filaments (arrowheads) and a vesicle labeled for CFTR (arrow).

Fig. 2. Western blots on lysates of rectal gland (left) and Hela cells (right) against myosin V. The heavy bands at the top are in the starting wells. Hash marks are 218, 131, 43 and 33 kDa.

3 4

Figs. 3 and 4. Double immuno- gold staining of vesicles for myosin V (6nm particles, small arrowheads) and CFTR (10 nm particles, large arrowheads).

Our studies show that vesicles are abundant in the actin-rich subapical region of the epithelial cell. Some of these vesicles carry CFTR, and, with the double immunogold method, some are shown to carry myosin V as well. We would conclude that myosin V may well be the motor that moves CFTR in its final stage of transport to the apical membrane. Co-immunoprecipitation studies to support the possibility of myosin V association with CFTR vesicles, and in vitro studies of vesicle movement along actin filaments, will help to further characterize this system.

Supported by an MDIBL New Investigator Award to RH. FRACTIONATION OF SHARK RECTAL GLAND CELLS FOR IDENTIFICATION OF CFTR TRAFFICKING COMPARTMENTS

Kathryn W. Peters1, John N. Forrest, Jr.2, Christopher R. Marino3 , and Raymond A. Frizzell 1 1Department of Cell Biology and Physiology, University of Pittsburgh, School of Medicine, Pittsburgh, Pennsylvania 15261 2Department of Medicine, Yale University, School of Medicine, New Haven, Connecticut 06510 3Department of Medicine, University of Tennessee, Memphis, Tennessee 38163

The shark (Squalus acanthias) rectal gland (SRG) is composed of secretory tubules in which the cystic fibrosis transmembrane conductance regulator (CFTR) is highly expressed. Sev- eral studies suggest this model is useful for studying translocation of CFTR from its site of synthe- sis to its functional location in the plasma membrane. A membrane vesicle fraction, distinct from the basolateral membrane, was isolated using a sucrose gradient and differential centrifugation (Dubinsky and Monti, Am J. Physiol. 251: C721-C729, 1986). Antibody labeling studies have demonstrated that CFTR translocates and concentrates at the apical membrane in response to cAMP stimulation of chloride secretion (Lehrich et al., J. Clin. Invest. 101: 737-745, 1998). We have observed SNARE proteins, e.g. vamp, snap, and syntaxin 1A in SRG (Peters et al., MDIBL Bull. 40: 55-57, 2001) thus components of the membrane fusion machinery are present. Accordingly, SRG should al- low us to evaluate our hypothesis that CFTR and VAMP-2 are present in cy- toplasmic vesicles under basal condi- tions and that upon stimulation, CFTR is depleted from this compartment. Rectal glands were removed and perfused under basal or stimulated (10 µM forskolin) conditions (Lehrich et al., 1998). Tissues were homog- enized and subjected to a series of dif- ferential centrifugations (Dubinsky and Monti, 1986). The 33,000g pellet was suspended in iodixanol (OptiPrep) to a 15% final concentration, then centri- fuged in a vertical rotor (Vti65.2) for Figure 1. Shark rectal glands were perfused under three hours at 50,000 rpms. Under basal (A) or stimulated (B) conditions and subjected these conditions a self forming gradi- to a series of differential centrifugations. The ent is generated which should allow 33,000g pellet was suspended in a final concentration separation of organelles and cytoplas- of 15% iodixanol and centrifuged in a vertical rotor to mic vesicles. This method was chosen form self-generated gradients. After electrophoresis because the hyperosmotic concentration and transfer, proteins were subjected to chemilumi- of sucrose is believed to coincide with nescence in which R-3195 anti-CFTR polyclonal the density of some intracellular mem- antibodies were utilized. Figure 2. Samples were prepared as described for Figure 1B. Chemiluminescence was con- ducted using VAMP-2 mono- clonal antibodies (Synaptic Systems). branes, e.g. GLUT4-containing vesicles, making it more difficult to resolve distinct fractions (Hashiramoto and James, Mol. Cell. Biol. 20: 416-427, 2000). Twenty-four 200 µl fractions were collected from the top of the tube, proteins were resolved by SDS-PAGE, then immunoblotted using R-3195 antibodies generated against the C terminus of CFTR (French et al., J. Clin. Invest. 98: 1304-1312, 1996). Immunoblot of gradient fractions from a basal gland (Figure 1A) reveals that the majority of CFTR is in the lighter fractions (1 and 3-5) and in the heaviest fractions (22-24). In contrast, in the stimu- CFTR lated gland, CFTR is observed in fractions of a medium density (8-11). Confirmation of cellular compartments associated with these fractions awaits immunoblots with organelle specific mark- ers. Interestingly, VAMP-2, a vesicle associated protein, that we VAMP-2 have shown previously to co-immunoprecipitate with CFTR as - well as cause a potentiation of Cl currents when co-expressed with CFTR in oocytes (Peters et al., Amer. J. Physiol. 277: C174- C180, 1999), is present in these fractions (Figure 2). Figure 3. Schematic The co-sedimentation of CFTR and VAMP-2 suggests that representation of it may represent a vesicle population of CFTR en route to the inside out cytoplas- plasma membrane. If this hypothesis is correct, CFTR would be mic vesicles showing oriented (Figure 3) such that cytoplasmic sites would be acces- orientation of CFTR sible to antibodies for immunoprecipitation of these vesicles in and VAMP-2. buffers without detergents. We began to answer this question using immunoisolations. The starting material for our initial stud- ies was the 33,000g pellet which was suspended in either HES (20mM HEPES, 1 mM EDTA, 250 mM sucrose, pH 7.4; Hashiramoto and James, 2000) or HE buffer (100 mM mannitol, 5 mM K+-HEPES, pH 7.6; Dubinsky and Monti, 1986). Approximately 300 µg of the sample was suspended in buffer, commercially available monoclonal antibodies generated against the C ter- minus of CFTR (Chemicon), were added, and samples were rotated for several hours at 4 oC.

Figure 4. The 33,000g pellet from a non-perfused gland was utilized in immunoisolations. Monoclonal antibodies raised against the C terminus of CFTR (Chemicon) were used in conjunction wtih Protein A to precipitate the im- mune complex. Proteins were observed through chemilu- minescence using 3195 antibodies for the immunoblot. Lane 1--positive control; Lane 2--immunoisolation using HEPES/mannitol buffer; Lane 3--immunoisolation in HEPES/EDTA/sucrose buffer. µ  These IgG2 antibodies have a high affinity for Protein A; therefore, 50 l Dynabeads Protein A was added so that immune complexes could be precipitated. After several washes, proteins were resolved by SDS-PAGE and immunoblotted with R-3195 polyclonal antibodies. As seen in Fig- ure 4, either band B (lane 2) or band C (lane 3) appears to be resolved with HM or HES buffer, respectively. Similar results were obtained when R-3195 polyclonal antibodies were used for both the immunoprecipitation and immunoblot (data not shown). The yield in this procedure is noticeably low and future experiments will include optimization of sample and antibodies. These results represent the beginning of a project aimed at identification of proteins in CFTR trafficking compartments. Future studies will include marker analyses of the 24 fractions to identify which organellar or subcellular fractions are represented. The integrity of vesicles will be confirmed through electron microscopy. Immunoisolations will be performed on fractions containing CFTR and VAMP-2. We will attempt to identify other proteins in these vesicles through silver staining and protein sequencing. These studies were supported by NIH DK 56490. SUBCELLULAR LOCALISATION OF INTRACELLULAR CALCIUM STORES IN EMBRYONIC MOUSE CARDIOMYOCYTES

Philipp Sasse1, Lars Cleemann2, Bernd K Fleischmann1, Jürgen Hescheler1, Martin Morad2

1University of Cologne, Robert-Koch-Str. 39, Köln, 50931 Germany, 2Department of Pharmacology, Georgetown University, Washington, DC 20007

During murine gestation the heart is seen 8.5 dpc (days postcoitum) as a contracting tube that in the following days undergoes rapid morphological and cellular developments both prior to and following birth at 21 dpc. Postnatal development of adult excitation-contraction (E-C) coupling in ventricular cardiomyocytes includes formation a t-tubular system that forms dyadic junctions with sarcoplasmic reticulum (SR) thereby allowing function coupling between L-type Ca2+ channels and Ca2+ release channels (ryanodine receptors, RyRs). Before this transition, 2+ contraction may be activated directly by Ca current (ICa), which is the major excitatory current during early (11-13 dpc) murine cardiogenesis (Davies et al., 1996, Circ. Res. 78:15-25). It is also known, however, that mRNA of cardiac RyR (RyR-2) is present during early rat embryogenesis (Gorza et al., 1997, J. Mol. Cell. Cardiol. 29:1023-1036) and that mutant mouse embryos lacking this protein die ~10 dpc (Takeshima et al., 1998, EMBO Journal. 17:3309- 3316). It is possible therefore that RyRs and intracellular Ca2+ stores may play different roles during embryogenesis than following the post-natal maturation of the heart. To explore this possibility we examined Ca2+ signaling in mouse (Mus musculus) embryonic heart cells during early (10 dpc) and late (16-18 dpc) stages of embryonic development using rapid (240 frames per second) 2-dimensional confocal microscopy.

Pregnant mice at the desired stage of gestation were killed by cervical dislocation and the hearts of the embryos were dissected and enzymatically digested using 1 mg/ml collagenase B in solution also containing (in mM): NaCl 120, KCl 5.4, MgSO4 5, Na-pyruvate 5, glucose 20, taurine 20, HEPES 10 (pH 6.9 adjusted with NaOH). The dissociated cells were plated onto gelatin-coated glass slips and maintained in an incubator in standard DMEM (Dulbecco’s modified medium) containing 20% fetal calf serum. The cells attached to the glass surface and started spontaneous contractions within 12 hours and were examined after 2-3 days at 37 oC using confocal fluorescence microscopy (Noran Inc.) following loading of the cells with the permeant Ca2+-indicator dye Fluo-3AM (2 µM, 12 min, 37 °C). The standard Tyrode’s solution contained (in mM): NaCl 137, KCl 5.4, CaCl2 2-5, MgCl2 1, HEPES 10, glucose 10, pH 7.4 (adjusted with NaOH). Cells were generally activated by rapid application (<50 ms) of a depolarizing solution containing 70-100 mM K+ (iso-osmolar replacement of K+ for Na+ or KCl- addition) or a solution with 5-10 mM caffeine. The sarcolemma and internal membranes were visualized by staining with di-2-ANEPEQ. Some cells were voltage-clamped in the whole cell configuration using Cs+ instead of K+ in the external solution and a dialyzing solution containing (in mM): NaCl 15, Cs+-aspartate 110, TEA-Cl 15, HEPES 20 Mg-ATP 5, cAMP 0.2, EGTA 2, K5-Fluo-3 2, (pH 7.2 adjusted with CsOH). Spontaneously beating cells were selected for experiments. To evoke Ca2+ signals in embryonic cells we used a depolarizing K+-rich solution to trigger electrical activity and activate Ca2+ current while caffeine (5-10 mM) was targeted at

0 ms 20.8 ms 58.3 ms 375 ms

A

B

C Edge Center Nucleus

2 =

D 0 F F/ ∆

KCl KCl Caffeine time, [1 s]

3 150 E Early F , 0 Late se ] F/F a

-1 ∆ * 2 100 ele t, [s -r

2+ )/d

N a 0 -release, 1 50 2+ d(F/F te of C Ca

a

R 0 0 edge center nucleus edge center nucleus edge center nucleus edge center nucleus KCl Caffeine KCl Caffeine Figure 1. Ca2+-signaling in mouse cardiomyocytes isolated early (10 dpc) or late (16-18 dpc) during embryonic development. Panels A to D show results from a single cell preloaded with Flou-3 AM. Panels A and B show samples of confocal images recorded at 240 Hz during rapid application of a K+-rich solution (panel A) and a solution with 5 mM caffeine (panel B) at the times indicated following the onset of Ca2+ release (0 ms). Panel C: Confocal image of sarcolemma and internal membranes stained for 1 min with 5 µM di-2-ANEPEQ. N indicates nucleus. Scale bar is 10 µm. Panel D. Time course of local Ca2+-transients measured at the edges and center, and in the nucleus during two brief KCl depolarizations followed by 2+ exposure to 5 mM caffeine. Panel E: Comparison of local Ca transients (maximal ∆F/Fo) produced by KCl depolarization (KCl) or 5 mM caffeine, at edge, center or nucleus of cardiac cells from early (10 dpc, n = 12- 16) or late (16-18 dpc, n = 8) stages of development. F: Comparison of the rate of rise of local Ca2+ transients (maximal d(F/Fo)/dt) under the same conditions. The * indicates significance at the p<0.05 level.

RyRs to directly induce SR Ca2+-release. Panels A to D in Figure 1 show typical responses recorded in a single embryonic cell in the late stage of development (16-18 dpc). Panel A shows selected frames of Ca2+-dependent Fluo-3 fluorescence recorded during KCl depolarization. The first frame, labeled 0 ms, shows the onset of the Ca2+ transient occurring over wide areas. Based on subsequent staining of external and internal membranes with di-2-ANEPEQ (panel C) it appears that the initial Ca2+ signal occurs both at surface of the cell and in the interior, but not in the nucleus (upper right, panel A: 0 ms, 20.8 ms; N in panel C). In the following frames (20.8 ms, 58 ms), the Ca2+ signals got stronger, invaded the nucleus (58 ms), and then faded away (375 ms). The time-course of the Ca2+ signals detected at the edges (edge) of the cell, in central regions (center), and in the nucleus was plotted in panel D for two consecutive KCl depolarization and for the caffeine-induced Ca2+ release shown in detail in the confocal images in panel B. In this cell the caffeine-induced response was first seen (0 ms) in the center of the cell, from where it spread throughout the cell. The caffeine signal was much stronger than the KCl-induced rise in Ca2+. The nuclear response was again delayed (frame 20.8 ms), but then displayed a strong afterglow (375 ms), typical of cardiac cells. The strong caffeine-induced Ca2+_ signal suggest the presence of prominent SR-Ca2+ stores gated by RyRs. Considering the lack of t-tubules in embryonic cells, it was surprising that Ca2+ signals appeared at many locations (panel A: 0 ms) throughout spontaneously beating and KCl-stimulated cells from one frame to the next (<4.167 ms), i.e. too fast and too uniform to be mediated by diffusion of Ca2+.

To validate and expand these observations we performed similar experiments using myocytes harvested early (10 dpc) and late (16-18 dpc) during embryonic development. Figure 1E compares the maximal amplitude of the Ca2+ signals in different regions at the cell at the indicated stages of development. The caffeine induced-responses (maximal ∆F/F0) did not change in different regions and during development suggesting that the amount of releasable SR Ca2+ did not change, and that the response was of sufficient duration to allow equilibration between different regions. The smaller KCl-induced responses, on the other hand, showed a trend towards larger Ca2+ signal in late development both at the edges and centers of cells, but not in the nuclei. To probe the local Ca2+ signals occurring prior to regional equilibration, we 2+ measured the maximal rate of rise of Ca (d(F/F0)/dt, panel F) and found that the enhancement of the Ca2+ signal during developmental was more pronounced, especially in the central region of cells (*, p <0.05), and was seen with both KCl and caffeine (panel E). Furthermore, it was a consistent finding that both the amplitude and rate of rise of Ca2+ signals, were of similar amplitude in the center and at the edges of cells.

A strikingly different picture was seen in voltage-clamped embryonic cardiomyocytes dialyzed with 1 mM Fluo-3 and 2 mM EGTA, to buffer intracellular Ca2+ and limit its diffusion (Fig. 2). In such experiments we found that the activation of a small Ca2+ current by depolarization to 10 mV was accompanied by Ca2+ signals that were strong near the surface membrane, but were equally weak at center and nucleus. The suppression of the central Ca2+ signal in dialyzed whole cell clamped myocytes suggests that the Ca2+ buffers may block the Ca2+-induced Ca2+ release mediated by RyRs.

Experiments were typically terminated by staining the cellular membranes by rapid perfusion of a solution containing the voltage-sensitive and moderately lipophillic dye

B C A 10 mV -60 mV

A p

50 Edges D E

= 1 0 F/F Center ∆ Nulcleus N

100 ms

Figure 2. Ca2+ current and local Ca2+ transients in an embryonic cardiac cell dialyzed with 1 mM Fluo-3 and 2 mM EGTA. Panel A: Time course of Ca2+ current and local Ca2+ transients (Edges (ROI0+ROI3), center (ROI1), nucleus (ROI2)) activated by an 80 ms depolarizing voltage-clamp pulse from –60 to 10 mV. Panel B and C show confocal images of the Ca2+ signals in frames recorded 12.5 and 25 ms, respectively, following deopolarization. Panel D and E show confocal images of membranes after staining with di-2-ANEPEQ for 10 s (panel D) and 60 s (panel E). (Membrane capacitance 79 pF.)

di-2-ANEPEQ. This compound stained the surface membrane within seconds (Fig. 2D) and penetrated to internal membranes within a minute (Fig 1 C, Fig. 2E) revealing extensive internal membrane structures, at times clustered around the nucleus. Consistent with ultrastructural observation, we found perinuclear origin of Ca2+ releases and Ca2+ sparks in some cells.

The finding of equally large caffeine-induced Ca2+ transients at 10 and 16-18 dpc suggests that development of RyR sensitive Ca2+ stores is an early event in murine caridogenesis. These Ca2+ stores appear to cluster around the nucleus, but are also widely distributed throughout the cells (cultured 1-3 days), and appear to be activated rapidly (<4-10 ms) to release significant amounts Ca2+ both during KCl-depolarization and spontaneous beating. It is critical to note that the rate of Ca2+ release increased from the early to the late stage of embryonic development suggestive of growth-dependent organization of SR. The structural organization and activation of this Ca2+ pool can hardly be the same as in mature ventricular cardiomyocytes (dyadic SR 2+ Ca release triggered by ICa) since the embryonic cells lack not only t-tubules but also the striation indicative of a regular myofilament array. The lack of t-tubules and sensitivity to Ca2+ buffers suggests some similarity to Ca2+ signaling events in atrial cells. We did not, however, observe the pronounced delays in the central Ca2+-transient often associated with atrial E-C coupling.

We conclude that intracellular Ca2+ stores play a significant and increasing critical role in cardiac E-C coupling fairly early in embryonic development. The gating and modulation of these stores remains to be determined, but it is apparent that they differ from that observed in atrial and ventricular myocytes of the adult mammalian heart. Supported by NIH RO1 16152 and a grant to Philipp Sasse from the Maine Affiliate of the American Heart Association.

VISUALIZATION OF SECRETORY GRANULES IN PANCREATIC ISLETS AND β-CELLS USING TIRF AND CONFOCAL MICROSCOPY

Carina Ämmälä1, Iain D. Dukes1, Li-Ping He2, Christopher Rhodes3, Louis Philipson4, and Lars Cleemann2

1Glaxo Smith Kline Research Institute, RTP, NC 27709 2Dept. of Pharmacology, Georgetown University, Washington, DC 20007 3Pacific Northwest Research Inst., University of Washington, Seattle, WA 98122 4Dept. of Medicine, University of Chicago, Chicago, IL 60637

Insulin, a key hormone regulating glucose homeostasis, is released from β-cells in the pancreatic islets of Langerhans through Ca2+-dependent regulated exocytosis of membrane bound insulin-containing secretory granules. The total number of granules present in a single β- cell has been estimated to about 13,000 (Dean, 1973, Diabetologia 9:115-119), which can be divided into at least three functionally distinct populations, the reserve, the readily releasable (RRP), and the immediately releasable pool (IRP) (Eliasson et al., 1997, J. Physiol. 503:399- 412). Granules in the two releasable pools are viewed as docked to the plasma membrane. The RRP requires ATP-dependent priming of the granules before fusion can occur whereas the IRP constitutes primed granules that can be rapidly released in an ATP- and temperature-independent manner in response to an elevation in intracellular Ca2+. The sized of the IRP has been estimated to about 50 granules. To study secretion of granules located in the IRP we used confocal and total internal reflection fluorescence (TIRF) microscopy to monitor the distribution and movement of fluorescence labeled secretory granules. The confocal technique allowed us to detect granules throughout the β-cell while TIRF microscopy has the ability to focus on an extremely thin surface layer (< 0.15 µm) and therefore might be useful to monitor docked vesicles that fuse with the membrane in the process of exocytosis. The fluorescent techniques were also used to detect Ca2+ signals (Fluo-3) and membrane staining (di-2-ANEPEQ).

Experiments were performed using either cultured mouse (Mus musculus) pancreatic β- cells and islets or β-TC3 cells, an insulinoma cell line. The vesicles were labeled using either quinacrine or adenovirus-mediated infection of a syncollin-green fluorescent protein (Syn-GFP) gene construct that targets expression of the Syn-GFP protein to secretory granules. Freshly o dissociated cells and islets were cultured at 37 C for 1 to 4 days following viral infection. During subsequent experimentation at room temperature the cells were superfused with a control solution containing: 137 mM NaCl, 4.7 mM KCl, 2.6 mM CaCl2, 1 mM MgCl2, 10 mM Hepes and 2 mM glucose. To trigger exocytosis increases in intracellular Ca2+ was achieved by rapidly (<50 ms) changing from the control to a depolarizing with iso-osmolar substitution of K+ for Na+ (45 mM KCl, 100 mM NaCl), or a solution to which 10 mM caffeine had been added. At room temperature, these manipulations would trigger the release of granules present in the IRP.

Figure 1A shows a confocal image of β-TC3 cells stained with quinacrine that is known to rapidly partition into the acidic environment of the secretory vesicles. This dye staining produced several bright spots of fluorescence separated by ~1-2 µm. Subsequent staining of the

A A B

Figure 1. Confocal fluorescence images β-TC3 cells. Panel A: clustering of secretory granules labeled with quinacrine. Panel B: Subsequent stronger staining of cell membrane with di-2-ANEPEQ. The contours of cells detected in panel B are shown as solid white lines in panel A.

surface membrane with di-2-ANEPEQ (Fig. 1B) allowed us to determine the number and outline of cells and to place the quinacrine-fluorescence at one end of each cell (opposite to the nucleus). Confoal sectioning revealed ~20-100 fluorescence hot-spots in each cell, so that each hot-spot may be produced by a cluster of more than 100 secretory granules. Considering the resolution of the confocal technique ((∆x, ∆y, ∆z) ~ (0.5 µm, 0.5 µm, 1 µm)) and density of vesicles, it is unlikely that this method would be able to detect single vesicles. The diameter of individual granules is 0.2-0.4 µm.

We therefore proceeded to use TIRF microscopy to focus on the space just below the membrane of β-cells and islets cultured on laminin-treated glass cover slips. The shallow depth of penetration from the surface of the glass cover slip may provide superior resolution (∆x, ∆y, ∆z) ~ (0.5 µm, 0.5 µm, 0.15 µm)), but only if the cells adhere closely to surface. Staining of the cell membrane with di-2-ANEPEQ generally produced a well-defined, stable and fairly uniform fluorescence within the entire bright field image suggesting that the entire lower membrane surface was in close contact with the glass cover slip.

Cells were incubated with the Ca2+-indicator dye Fluo-3AM (5 µM, 30 min) to ascertain elevated intracellular Ca2+ levels through activation of voltage gated Ca2+ channels by KCl depolarization or release of Ca2+ from intracellular stores triggered by extracellular application of caffeine. Figure 2A shows that the Ca2+-induced TIRF fluorescence in β-TC3 cells was also fairly uniform within each cell but varied greatly from cell to cell. A fluorescent bead is seen in the right side of the panel. Beads that were added to the perfusion chamber settled on the surface of the cover slip within a few minutes and served to find the optimal focal plane and monitor the stability of the laser light (488 nm) used for fluorescence excitation. Figure 2C shows cellular Ca2+ transients produced promptly by rapid application of a depolarizing solution with 45 mM KCl. These Ca2+ signals were strong in some cells, but smaller or absent in others including some were bright cells (cell # 1, Fig. 2BC) that may be Ca2+-overloaded. The TIRF images obtained with di-2-ANEPEQ and Fluo-3, suggest that the cells adhere well to the laminin-treated cover slips and that that some cells generate large and fast Ca2+ transients.

A C 45 mM KCl 1.5

1

e

nc

e 1.0

c 2

es

B

fluor 3

3

0.5 4 4 2 * Fluo- 1

3 0 500 1000 1500 Time, [ms]

Figure 2. Detection of Ca2+-depenAdent Fluo-3signals in β-TC3 cells using TIRF microscopy. Panel A: Image of cells and a fluorescent bead. Panel B identifies the cells by number and shows the location of a fluorescent bead (*). Panel C: Ca2+-transients in cells exposed to 45 mM KCl for 1200 ms. The TIRF images were recorded at 734 Hz, but plotted after averaging over 8 frames.

The TIRF images obtained by fluorescence labeling of vesicles with GFP were quite different in that they contained a number of bright spots (Fig. 3CD) about as bright as the fluorescent beads, of which three are seen towards the right side of the panels. The spots produced by staining of vesicles Bcould often be resolved at distances of 1-2 µm (Fig. 3BCD) as in the confocal sections (Fig. 1A). At other time they were found along contours too close together to be clearly distinguished. In cells that were stimulated repeatedly over a period of minutes, we observed that the intensity of the bright* spots changed significantly, sometimes as much as 2 or 3 fold. Fig. 3E shows how the intensities of three bright spots changes over time so that point #2 went from being the weakest (panel C) to being the strongest (panel D).

Some well defined bright spots at distinct fixed locations showed repeated abrupt changes in fluorescence (5-10%) during brief (1 sec) activation by KCl depolarization (Fig. 3F). These changes may occur synchronously at different points, but it was also noticeable that the synchronized signals varied from spot to spot, and that a strong signal at one point (trace #1, last pulse), might be completely absent at a neighboring point (trace #2).

The present data are consistent with the notion that exocytosis in stimulated β-cells occurs at fixed locations where fluorescence labeled secretory granules produce changes in TIRF fluorescence as they dock at the membrane and then fuse to release their contents. The rapid changes are of the order of 5 to 10 %, while slower changes are much larger. This may indicate that the secretory sites resolved by TIRF microscopy harbor about 10-20 docked vesicles within

A B C D * 3 2 1 3 2 1 3 2 1 *

*

E F 16 1 #1, Control #2, 45 mM KCl 2 3

e 1 c

en 14 sc e 2 r uo F = fl - ∆

P 2

F 12 G

10 3 1234567 Record # time, 1 s

Figure 3. Detection of secretion in a single mouse pancreatic β-cell using Syn-GFP and TIRF microscopy. A: Bright field image of the #6cell about 12 µm in diameter. B: Diagram showing location of “secretory sites” ( ) #2 and fluorescent beads (*). Panels C (record #2) and D (record #6) show changes in fluorescence images also graphed in panel E for the three sites identified by number. Panel F: rapid changes in fluorescence evoked by KCl-depolarization. Note that the traces in this panel were shifted from levels in panel E to appear aligned and in sequence. Each plotted point corresponds to 8 averaged frames recorded at 734 Hz.

that the secretory sites resolved by TIRF microscopy harbor about 10-20 docked vesicles within ~0.15 µm of the membrane. It is tempting to speculate that the docking of a single vesicle produces an abrupt increase in fluorescence, and vesicle fusion an abrupt fall, and that these processes are stochastic but to some degree synchronized by repetitive firing of action potential producing localized Ca2+ signals near the membrane. While such detail may require both technical improvement and more extensive experimentation, the present data make it plausible that TIRF microscopy used in conjunction with Syn-GFP may be developed a much-needed tool to routinely study the control of insulin secretion at the cellular level. We thank Dr. Li Ma for preparing the viral vector for expression of Syn-GFP. Supported by NIH RO1 16152, AHA 808116U, and the American Diabetes Association.

TIRF IMAGING OF FOCAL Ca2+ RELEASE IN VOLTAGE-CLAMPED ATRIAL AND VENTRICULAR MYOCYTES

Vadim Degtiar, Emily Jones, My Chau, Lars Cleemann, and Martin Morad

Georgetown University, 3900 Reservoir Road, NW, Washington, District of Columbia 20007

Contraction of both atrial and ventricular mammalian cardiomyocytes is controlled by Ca2+ signals that are generated by Ca2+-influx via L-type Ca2+ channels (DHP-receptors) and amplified by Ca2+-induced Ca2+ release (CICR) via Ca2+ release channels (ryanodine receptors, RyRs) in the membrane of the sarcoplasmic reticulum (SR). The organization of the SR, however, is quite different: ventricular cells have widely distributed dyadic junctions between SR and t-tubules, while atrial cells, lacking t-tubules, have numerous peripheral couplings at the cell surface and only non-junctional SR and RyR in the cell interior. Confocal measurements of Ca2+ sparks have to served to demonstrate functional implications of these structural differences, but has yet to provide a detailed picture in terms of unitary Ca2+ fluxes. We have therefore developed a new approach to examine differences in subsarcolemmal Ca2+ signaling in atrial and ventricular cells. We use rapid (734 frames/s) total internal reflection fluorescence (TIRF) microscopy, which has the ability to focus on a thin surface layer (~ 0.15 µm) of cells, provided the cells can made to stick to the fluid/glass interface where total internal reflectance takes place.

TIRF microscopy (Axelrod et al., 1982, J. Microscopy, 129:19-28) has been used to image structures and events within a surface layer that is considerably thinner (∆z ≅ 0.15 µm) than the depth of the focal plane (∆z ≅ 0.8 µm) provided by other light microscopic techniques, including confocal microscopy. This technique has been used to study surface layers, and fusion of synaptic vesicles. The basic requirement is that is that an exponentially decaying “evanescent” electromagnetic wave is generated at the glass bottom of a perfusion chamber where cells in sufficiently close contact are subject to fluorescence excitation. The evanescent wave arises within an area where a parallel beam through the underlying glass body is totally reflected due to o its hypercritical angle of incidence (θ > 60 , sinθ > nwater/nglass , nwater ~ 1.33, nglass ~ 1.5). The glass body in question may be a separate prism on an upright microscope, or, on a more convenient inverted microscope, the objective itself if numerical aperture is sufficient for total internal reflection (NA > nwat). Such objectives are now available with NAs of 1.45 to 1.65. We have followed a different approach by designing a system where the first element of a modified Zeiss objective (x40, NA 1.4, oil) combines the geometries of a ball lens and a prism. This allows a compact design, a simple light path, a high angle of incidence (55-80o), and the use of standard optical materials (lens, immersion oil, cover slip). A fiber-coupled laser (HeNe, 10 mW, 488 nm, Omnichrome) was used for fluorescence excitation of Ca2+ sensitive dyes (Fluo-3, Fluo-3AM, Fluo-4) and fluorescent beads (0.2 µm Ø). Emitted light (> 500 nm) was intensified (single micro-channel plate) and recorded with fast CCD camera (128x128 pixels, 8 bits, 734 frames/s, Dalsa).

Our general experimental approach was to examine single voltage-clamped cardiomyocytes while simultaneously recording subsarcolemmal Ca2+ signals using TIRF microscopy. Single atrial and ventricular rat cardiomyocytes were prepared for each day of experimentation by enzymatic dissociation of hearts excised rapidly from deeply anestethized rats (Rattus norvegicus). The cells were plated onto pretreated glass cover slips (25 mm Ø), and were used within 10 hours. Internally dialyzed cells were voltage-clamped (Dagan) in the whole cell configutation using depolarizing pulses to activate Ca2+ current and evoke Ca2+ release. The Cs+-based dialyzing solution typically contained 0.8 mM Fluo-4 and 2 mM EGTA.

It is an important property of the TIRF technique that it is sensitive to minute vertical movements (~ 10 nm) within the evanescent field, so that motion artifact might be superimposed on the Ca2+ signal that are our primary interest. We therefore studied the development of stable adhesions of cardiomyocytes to pretreated glass cover slip. Variations in pretreatment included different methods of cleaning of the cover slips (waeek acid and/or alcohol), and coating with various compounds (collagen, poly-lysine, gelatin, and laminin) applied in a wetting aqueous solution to provide a final density of (1-20 µg/cm2) after drying in a dust free environment at room temperature or ~80 oC. The cells were plated onto the pretreated glass cover slips in different concentration, either immediately after enzymatic dissociation or after stabilizing up to 2 hours, and were then left in Tyrode’s solution up to 10 hours. During this period we checked the survival, stickiness, and areas of adhesion of cells. Stickiness was tested by counting the number of cells that remained in position after switching on the stream from a rapid perfusion “puffer” system. The areas of adhesions, and their stability were evaluated in the TIRF mode after staining the cells with the cell permeant Ca2+-indicator dye Fluo-3nAM. We found that a few stable adhesions in each cell developed over a period of hours (>2 h) and had the largest area (5-20 µm across) when the cells where plated immediately after enzymatic dissociation and when laminin was used as coating. Such areas of adhesion had a well-defined outline that did not change when turning on the stream from the puffer or activating contractions by KCl- depolarization. The other compounds (gelatin, collagen, and polysine) did not promote adhesion to the same extent and/or had adverse effects on cell survival. Laminin was effective in a layer (1-2 µg/cm2) that is estimated to be too thin (10-20 nm) to degrade the TIRF signals. The methods used for the initial washing of the cover slips and the drying of the coating were of little consequence.

We next proceeded to recorded Ca2+ signals in the TIRF mode from voltage clamped cells dialyzed with 0.8 mM Fluo-4 and 2 mM EGTA. Figure 1 compares The Ca2+ currents TIRF Ca2+ signals recorded in an atrial (panels A-C) and a ventricular cells. The bright field images show outlines and patch electrodes. As shown in panels B and E, cell adhesions appear in TIRF images as relatively small bright areas within the outline of the cells. The smaller bright spots around the cells correspond to fluorescent beads used for focusing on the exact bottom of the perfusion chamber. Panels C and F show the Ca2+ currents measured following depolarization to 0 mV in relation to the “global” TIRF Ca2+ signals measured at 734 Hz from the entire areas of adhesion. The current in the ventricular cell (panel F) was clearly contaminated by Na2+ current, but it was a general finding that Ca2+ current in atrial cells was much smaller in atrial than in 2+ ventricular cells. Nevertheless, the Ca signals, measured as ∆F/Fo, were typically larger in atrial cellsthan in ventricular cells and developed developed faster as indicated by the arrows at 50% of peak (10 vs. 15 ms). This finding may indicate that the peripheral couplings in atrial cells are located within reach of the evanescent wave, while Ca2+ from dyadic releases in ventricular cells reach the cell surface only after some diffusion delay.

Atrial cell Ventricular cell

A C D F

B E

Figure1. Comparison of TIRF Ca signals in voltage clamped atrial (panels A-C) and ventricular (p anels D-E) cells. Panels A and D show bright field images with outlines of the cells and the patch electrodes. The outlines of the cells were transferred to panels B and E, which show the areas of adhesion within the cells and several fluorescent beads mainly around the cells. The box in panel B corresponds to the area analyzed in Fig. 2. Panels C and F show recorded Ca2+ currents (top) and the glabal TIRF Ca2+ signals measured within the entire area of adhesion. The scale bars in panels C and D correspond to 1000 pF for the current trace, 0.5 (panel B) or (0.3) panel (F) for the Ca2+ signal, and 20 ms for the common time axis. The atrial cell was depolarized from –60 to 0 mV while the ventricular cell was depolarized from –80 to 0 mV.

When examined in greater details, the ca signals within the areas of adhesion showed distinct patterns. Figure 2 shows consecutive frames recorded at 1.4 ms intervals in an atrial cell (the same as in Fig. 1A-C) immediately after depolarization (#1-5) and ~15 m later (#12). Notice that bright fluorescence spots appear in increasing numbers from one frame to the next. The regularly spaced arrows in frame # 3 point at focal Ca2+ signals that appeared on a longitudinal line with the approximate sarcomere spacing (~ 1.8 µm). This longitudinal alignment was also noticeable at a later time (frame #12) when the patter of focal Ca2+ signals is much denser. It was also common to observe to bserve Ca signal aligned in the transverse direction (#12). Some prominent focal Ca2+ signals can clearly be followed from frame to frame (vertical arrows in frames #3-5) yet do not last nearly as long as the typical Ca sparks (~15-20 ms duration in observed with confocal microscopy (vertical arrows in frames #3-5) yet do not last nearly as long as the Ca2+ sparks typically observed with confocal microscopy (~15-20 ms in rat atrial cardiomyocytesmycytes). It should be noticed, however that we also detected longer lasting focal Ca2+ signal resembling the focal Ca releases observed with confocal microscopy.

Our results acquired with TIRF microscopy, suggest that Ca2+ signaling in the subsarcolemmal spaces of atrial and ventricular myocytes, in addition to the conventional Ca2+ sparks, described by confocal microscopy, is accompanied by brief, focal events. These brief events, possibly “quarks”, appear to reactivate at the same location, unlike the sparks, and often

2+ Figure 2. Focal Ca releases #1 #4 recorded 7in the TIRF mode at 734 Hz immediately after voltage-clamp depolarization of an atrial cardiomyocyte from –60 to 0 mV. The frames #1-5 are numbered consecutively each showing the light accumulated during a 1.4 ms interval. Frame #12 was recorded near the peak #2 #5 of the global Ca2+ signal about 18 ms after depolarization and is reproduced with a different gray scale. Arrows and lines indicate prominent focal Ca2+ signals and their spacing and alignment.

#3 #12

trigger other adjacent (within 1-2 µm) fast events. The probability of both very fast, and “classical” Ca2+ events was markedly increased (comparing to the probability of spontaneous events occurring at –80 or –60 mV) already with moderate depolarizations (-50 to –30 mV), and further increased at 0mV. However, very strong depolarizations (+80 mV) only slightly supported subsarcolemmal Ca2+ signaling, with a very strong signal triggered by the tail-current during membrane repolarization.

In addition to an increase of probability of transient events to occur, depolarization also caused a steady elevation of Ca2+ concentration. The very fast and brief events were also seen in experiments where cells were dialyzed with internal solution containing only Fluo-4, without other exogenous Ca2+ buffers. Addition of EGTA reduced mainly the overall accumulation of Ca2+ in subsarcolemmal space. I.e., the relative slow increase of (F/Fo) during prolonged depolarization was dramatically stronger in experiments with low or no exogenous Ca2+ -buffers added to the internal solution, while transient events were affected to a lesser extent. Despite the smaller Ca2+ channel current density, atrial myocytes revealed stronger changes in the subsarcolemmal [Ca2+], compared to ventricular ones.

We conclude that TIRF microscopy provides detailed information about the subsarcolemmal Ca2+ signaling, that appear to be different isn atrial and ventricular myocytes. The high spatial and temporal resolution of the techniqe reveal in fascinating details a new level of Ca2+ signaling. The detailed analysis of these Ca2+ signals in terms of the Ca2+ fluxes generated by clustered or single DHP- and RyRs remains to be performed. Supported by AHA 9808116U and NIH RO1 16152. CHARACTERIZATION OF ENDOTHELIN, NITRIC OXIDE, PROSTAGLANDIN E, AND NATRIURETIC PEPTIDE RECEPTORS IN THE BULBUS ARTERIOSUS OF THE EEL, ANGUILLA ROSTRATA

David H. Evans and Matthew S. Kozlowski Department of Zoology, University of Florida, Gainesville, FL 32611

The bulbus arteriosus (BA) connects the cardiac ventricle of teleosts to the ventral aorta; it expands during cardiac systole and rebounds, pushing blood into the ventral aorta, during diastole. The result of this “windkessel effect” is a smoothing of blood flow to the delicate gill vessels. Various studies (e.g., Holmgren, S., Acta Physiol. Scand. 99: 62-74, 1977; Watson, A.D. and Cobb, J.L.S., Cell Tiss. Res. 196: 337-346, 1979; Farrell, A., J. Exp. Zool. 209: 169- 173, 1979) suggest that the BA is controlled by a variety of neurotransmitters and paracrines and may be more than a simple windkessel. Our preliminary studies have demonstrated that the BA of the eel, Anguilla rostrata, expresses receptors for endothelin, porcine C-type natriuretic peptide, and prostaglandin E1, and also responds to NO (Evans, D.H. and Harrie, Bull Mt. Desert Isl. Biol. Lab. 38: 31, 1999). The present study was undertaken to determine, more specifically, the receptors expressed in this tissue. We used techniques described previously (e.g., Evans, D.H. and Gunderson, M.P., Am. J. Physiol. 274: R1050-R1057, 1998; Evans, D.H. and Gunderson, M.P. Bull. MDIBL 37: 107, 1998) for the capture and maintenance of eels (Anguilla rostrata) and preparation of vascular rings from the BA. Resting tensions of 500 mg were set and maintained in each species before the sequential addition of putative vasoactive agonists.

The bulbus arteriosus rings constricted when either ET-1 or sarafotoxin S6c (SRX, specific ETB-receptor agonist) were applied, dependent upon the concentration, but the constriction only reached statistical significance at concentrations above 10-7 M (N = 8). This suggests a rather low sensitivity to these ET agonists, but it does indicate the presence of ETB as well as ETA receptors. After BA rings were preconstricted with 10-7 M ET-1, the NO donor sodium nitroprusside dilated the BA vascular smooth muscle in a concentration-dependent manner over the range of 10-8M to 10-4 M (N = 8), with the exception of 10-7 M, where dilation was not statistically significant (p =0.13). Thus, it appears that the intracellular receptor (soluble guanylyl cyclase) for NO is present in this tissue. Similarly, both the prostaglandin EP1/EP3- receptor agonist sulprostone and EP2-receptor agonist butaprost dilated ET-preconstricted rings in a concentration-dependent manner, reaching statistical significance at 3x10-9 M and 10-8 M, respectively. The dilation produced at the maximum concentration (10-6 M) was the same for both agonists (N = 6). These data suggest that at least two types of receptors are present that bind to PGE2 in this tissue. Natriuretic peptides also produced concentration-dependent dilation of ET-preconstricted rings, reaching statistical significance at 3x10-10 M ANP (eel) and 10-7 M for CNP (porcine). Lower concentrations of pCNP appeared to produce dilations (starting at 3x10-9 M), but high variability precluded statistical significance. In addition, the maximal dilation produced at 2x10-7 M was 47% less for CNP than ANP (N = 8). Thus, it appears that both Natriuretic Peptide Receptor-A and NPR-B are present. Our data are preliminary, but it appears that a variety of receptors are expressed in the eel BA, suggesting a complex local and regional control of this vessel. (Supported by NSF IBN-0089943 to DHE and an REU supplement to MSK from this grant, as well as NSF DBI-9820400 to the MDIBL) DIRECT OBSERVATION OF BLOOD FLOW IN THE GILL OF THE EEL, ANGUILLA ROSTRATA BY VIDEOMICROSCOPY

David H. Evans Department of Zoology, University of Florida, Gainesville, FL 32611

In the past few years, we have demonstrated that a variety of vasoactive substances (e.g., adenosine, natriuretic peptides, endothelin, nitric oxide, and prostaglandins) can alter tension in blood vessels from a variety of species of fishes, most notably the ventral aorta that carries blood from the heart to the gills (e.g., J Comp Physiol [B] 162: 179-83, 1992.; J Exp Zool 265: 84-7, 1993; J Comp Physiol [B] 165: 659-64, 1996; Am J Physiol 274: R1050-7, 1998.; J. Exp. Zool. 289: 273-284, 2001; Physiol. Biochem. Zool. 74: 120-126, 2001). In addition, more recent work has shown that at least ET, NO, and PGs also may control active transport across the marine fish gill (Evans, D.H. et al., Bull. MDIBL 39: 17; this volume). Since endothelin has also been shown to alter the pattern of blood flow through individual gill lamellae (Sundin, L., and G. E. Nilsson. J. Comp. Physiol. B 168: 619-623, 1998), we have initiated a research program to visualize and record the effects of these other vasoactive substances on blood flow through the eel gill filaments and lamellae in vivo. We report here the results of preliminary studies (on control eel gills) that demonstrate that videomicroscopy of eel gill blood flow is reasonably easy.

American eels, A. rostrata, were purchased from a local fisherman and maintained in running sea water at 16 ˚C. Fish were anesthetized in 0.1g/l MS222 in sea water and placed, ventral side up, in sufficient aerated sea water to cover the body (containing 0.025g/l MS222). The operculae were cut along the ventral aspect in order to expose the gills and to immobilize the fish (via pins to the underlying paraffin base). Gill blood flow was visualized through a 20X water-immersion lens on an Olympus BX30 reflected light microscope (with polarizing filters) and recorded by an Optronics DEI-750 video camera system, which output was connected via a Dazzle A/D converter to a Macintosh iBook G3 running iMovie. Representative video clips were exported to Quicktime, compressed, and saved as self-looping .jpg files. The clips may be viewed by cutting and pasting the URLs into a browser. If Quicktime is not installed on a specific computer, go to http://www.apple.com/quicktime/download/ for a free copy of Quicktime. The largest clip is 12 MB, which should download in less than one minute on an ISDN line; the others should take ca. 15 sec.

Fig. 1 (http://www.zoo.ufl.edu/dhefish/fishgill.jpg) (copyright by Benjamin Cummings). This diagram from Campbell's "Biology" text illustrates the basic structure of the teleost fish gill. Note that each gill arch contains a stack of gill filaments with numerous lamellae at right angles to the filament. Blood runs along the training edge of the filament in the afferent filamental artery, enters the lamella through the prelamellar arteriole, flows as a sheet through the lamellae (around pillar cells which separate the epithelial sheets of the lamella), exits the lamella via the postlamellar arteriole, and travels back to the gill arch via the efferent filamental artery. The lamellae are the site of gas exchange in fishes.

Clip 1. (http://www.zoo.ufl.edu/dhefish/Filament.1.mov) This shows the distal region of a single filament (on edge) running vertically and numerous lamellae at right angles (horizontal). The flow of blood through the afferent filamental artery is obvious (running upward in the middle of the filament), as is the flow into each lamellae, through the prelamellar arterioles. Flow through individual lamellae is also visible, with somewhat greater flow (darker band of flowing erythrocytes) through the peripheral channel on the edge of the lamellae (especially visible on the first lamella on the left). It is obvious that blood is flowing even through the most distal lamellae of this filament. This total perfusion of all lamellae is in contrast to the conclusions of Booth (J. Exp. Biol. 83: 31-40, 1979) that suggested that only approximately 60% of the lamellae were perfused in the unstimulated trout filaments.

Clip 2. (http://www.zoo.ufl.edu/dhefish/Filament.2.mov) This clip shows a filament from the top, with several lamellae and afferent and efferent filamental arteries visible. As the focus changes, you should be able to see perfusion of a single lamella at the end of the filament.

Clip 3. (http://www.zoo.ufl.edu/dhefish/fil.moving.small.mov) This clip shows a single filament (on edge, from lower right to top left) with numerous lamellae to the right. Notice blood flow through the afferent filamental artery, the prelamellar arterioles, and in the peripheral channels of the many lamellae. At the start of the clip, flow through a postlamellar arteriole is seen at the bottom left of the frame. The filament is moving, presumably as the fish breathes. Note that the flow slows as the filament moves downward. We have no directly measurements, yet, of either ventilation or cardiac output, but we hypothesize that this alteration in perfusion rate may be associated with ventilation to match ventilation and perfusion.

Clip 4. (http://www.zoo.ufl.edu/dhefish/Lamella.mov) This is two lamellae, one in focus, showing the pattern of blood flow around the pillar cells, and through the peripheral channel, in particular in the second lamella, which is out of focus.

Clips 5 and 6. These both show an array of lamellae (mostly from the top) in a single filament from a fish 5 hrs (http://www.zoo.ufl.edu/dhefish/Lam.5hrs.mov) and 8 hrs (http://www.zoo.ufl.edu/dhefish/Lam.8hrs.mov) after the start of the experiment, demonstrating that the preparation is viable for prolonged periods.

Our data show that blood flow in the eel gill is easily visualized with this videomicroscopy system. Our next step is to monitor, and quantify, changes in these patterns elicited by infusion of the vasoactive signaling agents described above. In addition, this system could be utilized to study the effects of hypoxia, water-borne xenobiotics, or acidification on gill perfusion. (Supported by NSF IBN-0089943 to DHE) FUNDULUS NUMBERS IN NORTHEAST CREEK VARY WITH TIDE AND SALINITY

Kathleen Findlay1, Kimberly Ensign1 and George W. Kidder III2 1Dept. of Biological Sciences, Illinois State University, Normal, IL 61709 and 2Mt. Desert Island Biological Lab.

The behavioral osmoregulation hypothesis suggests that estuarian fish such as Fundulus heteroclitus minimize their osmotic work by choosing to inhabit water with an average salinity matching that of their plasma. Under laboratory conditions this seems to be true (Kidder, Bull MDIBL 36:69, 1997), but observations in the wild, while supporting this hypothesis, have hitherto lacked quantitation.

We used an underwater camera with a 50’ cable to monitor the presence of fish in a defined volume of water. A 20 x 20 cm black square was painted on a square piece of wood, and ring stand components were used to support an underwater camera 20 cm from the center of the square. When submerged in the stream with the board vertical and parallel to the flow direction, fish could be identified and counted as they swam through the field of view on the monitor. Any fish seen against the black square was Fundulus abundance therefore present in the 2.6 liter volume 5/30 - 6/4/01 thus defined. The majority of fish seen 60 were Fundulus; the few sticklebacks 50 Mean+SE; N = 5 40 observed were not counted. Salinity and 30 temperature were measured with an 20 Orion 105 salinity meter, with the probe 10 # fish /10 min /10 fish # 0 near the camera location. Current floods 20 16 Our site was south of the Route 3 12 bridge over Northeast (King’s) Creek on 8 Mount Desert Island. This is the site of a 4 Salinity, ppt Salinity, 0 broken dam, with current reversing and -100 -80 -60 -40 -20 0 20 40 60 80 100 salinity changing dramatically during Minutes after high tide at Bar Harbor spring tides, but only a small change in water level. Fundulus were counted for Figure 1. Fundulus abundance measured as described in the 10 minute periods ±90 min around high text, with simultaneous salinity and current observations. All tide at Bar Harbor, as determined from data are mean ± SE for 5 three-hour observation periods. published tide tables.

Figure 1 shows killifish moving upstream before current reversal and returning both with and against the current. Few fish were observed at the time of maximum flood. These data suggest that similar observations farther up and down the stream would show corresponding patterns as fish approach and leave the present site. Future work will use automated fish recording techniques to minimize the effort involved in these experiments.

This work was performed during the Illinois State University course “Marine and Maritime Physiology”. We thank Nature Vision, Inc., for the loan of the Aqua-Vu camera system.

CALANUS FINMARCHICUS: KINEMATICS OF THE PEREIOPODS DURING AN ESCAPE

Lenz, P. H., Hartline, D. K., and Hower, A. Pacific Biomedical Research Center, University of Hawaii at Manoa, Honolulu, HI 96822

Planktonic copepods possess an escape response that removes them from a potential threat at 300 to 1000 body lengths per second with an acceleration of 100 to 200 m.s-2 (Buskey et al. 2002). In order to achieve this, copepods generate a work output of 8 ergs for each power stroke of the pereiopods (Svetlichnyy, Oceanology 27:497-502, 1987; Lenz and Hartline, Mar. Biol. 133:249-258, 1999). This translates into muscular work-production rates of ca. 500 ergs.s-1 (Lenz and Hartline 1999, op cit.), which is among the higher rates in the animal kingdom. In order to understand how this performance is achieved, we studied pereiopod movements using high-speed video (1000 fps), while simultaneously monitoring force production in tethered Calanus finmarchicus.

During the slow swim, C. finmarchicus has the first antennae deployed at 90° to the axis of the body. Locomotion is slow and is produced by the rhythmic beating of the cephalic appendages. The pereiopods, or swimming legs are promoted (held anteriorly directed) and the urosome is usually extended along the axis of the body. During an escape the cephalic appendages retract against the cephalothorax, the urosome is raised dorsally, the first antennae fold against the prosome and the pereiopods produce a metachronal wave of power strokes starting with the 5th pair of legs and ending with the 1st pair (Storch, Berh. Dt. Zool. Ges., Zool. Anz. Suppl. 4:118-129, 1929). During the power stroke, each pereiopod pair moves in an approximately 180° arc from the promoted to the remoted position. The pereiopod pairs produced individual peaks in force, with the exception of the 5th pair (Fig. 1). At 8 to 9 °C, these peaks in force, spaced at 2 to 3 ms, corresponded to the 90 to 100° position of the pair of swimming legs (Fig. 1). For each force transient, the largest forces typically ranged between 40 and 60 dynes and were produced by the 4th, 3rd and 2nd pereiopod pairs. During the power strokes, the pereiopods maximized their surface area through the spreading of the exopods and endopods as well as their setae (Fig. 1). In contrast, during the return stroke, the swimming legs minimized their surface area, thus generating only a weak to undetectable reverse force. The duration of this entire sequence of power and return strokes varied among individuals ranging from 9 to 12 ms at 8 to 9 °C. Supported by NSF grant OCE 99-06223 and Mount Desert Island Biological Laboratory New Investigator Award.

P5 P4 P3 P2 P1 80 Figure 1. Force record from a C. finmarchicus during power strokes of the 60 five pairs of pereiopods. Duration of the force transient is 12 ms with a maximum 40 force of 60 dynes (P3). The vertical lines indicate the times when the individual 20 pereiopods were nearly perpendicular to Force (dynes) the main axis of the body as seen on the 0 corresponding video frames. -20 0 5 10 15 20 Time (ms)

ISOLATION OF GENES EXPRESSED IN DEVELOPING NEPHRONS IN THE ADULT KIDNEY OF SQUALUS ACANTHIAS

1,2 3 1,2 Maren Wellner , Jennifer Litteral , Torsten Kirsch , Patricia Waldron, 1 4 1 Marlies Elger , Hartmut Hentschel , and Hermann Haller 1 Medical School Hannover, Dep. of Nephrology, Hannover, Germany 2 Franz-Volhard-Klinik at the Max-Delbrück Center for Molecular Medicine, Berlin, Germany 3 Mount Desert Island Biological Laboratory, Salisbury Cove, Maine 4 Max-Planck Institute for Molecular Physiology, Dortmund, Germany Elasmobranches are able to develop new nephrons during their life span, and their kidney may serve as a model system for nephron development. We previously showed that the kidney in the adult dogfish revealed four developmental stages as well as mature nephrons. These developmental zones express genes necessary for kidney development. We used tissue dissection and subtractive hybridization to analyze gene expression in developing nephrons within the adult kidney. Total RNA from shark kidney developing and differentiated zones were isolated and cDNA synthesis was performed. Subtractive hybridization allowed us to clone partial cDNAs involved in kidney development. More than 1000 clones were amplified by PCR and the cDNA was spotted onto glass slides. The slides were hybridized with total RNA from developing and mature kidney tissue. The RNA was labeled with Cy 5 and Cy 3 fluorescent dyes as a flip dye experiment. Differentially expressed genes were sequenced and analyzed using public databases. From the subtractive hybridization, we found two candidate genes differentially expressed in developing nephrons. The first cDNA fragment contains an open reading frame of 111 aa and shows 91% identity to the TAF-I of Xenopus laevis. Template activating factor I (TAF-I) was originally identified as a host factor required for DNA replication and transcription of adenovirus genome complexed with viral basic proteins. Purified TAF-I was shown to bind to core histones and stimulate transcription from nucleosomal templates. Human TAF-I consists of two acidic proteins, TAF-I alpha and TAF-I beta, which differ from each other only in their amino-terminal regions. TAF-I plays a role in remodeling higher-order chromatin structure as well as nucleosomal structure through direct interaction with chromatin basic proteins. (Matsumoto K et al., Mol Cell Biol 19, 6940-52, 1999) TAFs interact with transcriptional activators and mediate transcriptional activation. Innumerable transcription factors integrate cellular and intercellular signals to generate a profile of expressed genes that is characteristic of the cellular properties of the cell. Variants of the general transcription factors play specific roles in embryonic development, reflecting the requirements of the developmental gene regulation. (Veenstra and Wolffe, Trends in Biochem. Sciences 26, 665- 671, 2001) A further cDNA fragment contains 209 base pairs and shows homologies (more than 50%) to the sec 61 of Danio rerio. Sec 61 codes for a protein translocation channel. The initial step in the biogenesis of most extra-cellular proteins in eukaryotic cells is their translocation into the endoplasmic reticulum. Translocation occurs through a hydrophilic channel that has been conserved throughout evolution. The channel itself has been proposed to be a passive conduit for polypeptides. Other associated proteins provide the driving force for translocation and determine its directionality. The Sec61ß subunit is required during Drosophila embryonic development. Homozygous mutant embryos die at the end of embryogenesis. (Valcarel et al., J. Cell Science 112, 4389 - 4396, 1999) Using glass slide hybridization, we found cDNAs overexpressed in developing nephrons. A cDNA fragment translated into the corresponding protein shows 65% identity to the UDP- glucuronosyltransferase of macaca fascicularis. Glucuronidation is the major detoxification pathway. The reaction is catalyzed by a family of UDP-glucuronosyltransferases and is involved in conjugation with many endobiotic and xenobiotic substances with glucoronic acid, forming inactive water-soluble intermediates. (Grancharov K et al. Pharmocology and The rapeutics 89, 171 - 186, 2001) Cubero et al. (Dig Dis Sci 46, 2762-2767, 2001) examined the bilirubin UDP- glucuronosyltransferase throughout rat fetal development. They observed an increase in conjugated billirubin within 90 days. Furthermore, we isolated a 498 base pair cDNA fragment showing 57% identity to the human L-kynurenine aminotransferase. This enzyme catalyzes the formation of kynurenic acid. Kynurenic acid has become a standard tool for use in the identification of glutamate-releasing synapses and has been used as the parent for several groups of compounds now being developed as drugs for the treatment of epilepsy and stroke. Beal et al (Brain Res. Dev. Brain Res. 68, 136-139, 1992) observed that in rat fetal brain, kynurenic acid concentrations were increased 5 fold prenatally and reached adult levels at day 7 after birth.

Fig. 1: Protein Alignment of the isolated protein sequence to the Xenopus laevis TAF-I sequence In summary, we isolated cDNAs coding for Template activating factor I, Sec 61, UDP- glucuronosyltransferase and L-kynurenine aminotransferase. The cDNA fragments described play a role in developmental processes. Further experiments like quantitative RT-PCR and in situ- hybridization will confirm the role of the candidate genes and will give us a deeper insight into the molecular mechanisms of nephron development. FUNCTIONAL EXPRESION OF SHARK VIP RECEPTOR IN XENOPUS OOCYTES: ACTIVATION OF CFTR CHLORIDE CHANNELS

Gerhard Weber1,2, Marie Bewley1, John Pena1, John N. Forrest Jr.1. 1Dept. of Medicine, Yale University School of Medicine, New Haven, CT 06510 2 Division of Nephrology, Medical School Hannover, Hannover, Germany

We previously reported the cloning, full length sequencing and molecular characterization of a VIP-like receptor from the shark rectal gland (Pena et al., Bull. MDIBL 40:133-135,2001). The deduced shark VIP receptor (sVIP-R) protein had only 57% identity with mouse VIP-1 receptor and 56% identity with human and rat VIP-1 receptor. This receptor is the likely target for VIP and related hormones that stimulate salt secretion in the rectal gland.

We have now prepared a full-length expression clone of the shark VIP receptor and report here the functional expression of this receptor in Xenopus oocytes using the co-expression of human CFTR as the reporting element. In control experiments, oocytes injected with water or with CFTR cRNA only did not respond to exogenous VIP (100nM). Figure 1 below demonstrates that oocytes co-injected with sVIP-R and hCFTR cRNA respond to 10 nM VIP with an increase in chloride conductance that is reversible following removal of VIP from the bath.

IV-plot from VIPR 08-21-01 oocyte D

12000 G(m) (µS) I (nA) 10000 VIP 10 nM

8000

6000

FR 4000

2000 FR wash out

-1 20 -100 -80 -60 -40 -20 0 20 40 60 Vm (mV) -2000

-4000

-6000

-8000

-10000

Figure 1. Current voltage plot during a 2-s voltage ramp from –120 mV to +60 mV on a representative oocyte expressing shark sVIP-R and hCFTR, before and after exposure to 10 nM VIP.

A dose response to mammalian VIP in sVIR-R and CFTR cRNA co-injected oocytes is shown in figure 2. VIP (1nM) elicited a detectable increase in chloride conductance and the EC50 for VIP was approximately 8 nM. Maximal activation was observed at 80-100 nM VIP. The expressed shark VIP receptor also responds to mammalian pituitary adenylate cyclase activating peptide (PACAP), peptide histidine isoleucineamide (PHI) and secretin. Further dose response studies are needed to determine the relative potency of these agonists.

300 Gm (µS)

200

100

0 -11 -10 -9 -8 -7 -6 VIP dose response (M) Figure 2. Dose response to VIP in Xenopus oocytes expressing shark VIP-R and hCFTR (n=3-5 oocytes each point).

In summary, we have successfully co-expressed the shark rectal gland VIP-like receptor with hCFTR in Xenopus oocytes and show that this receptor is responsive to mammalian VIP, PHI, PACAP and secretin. The cloned receptor is therefore the likely membrane protein mediating VIP and related secretagogue stimulated chloride secretion in the intact rectal gland.

This work was supported by NIH grants DK 34208 and NIEHS P30-ES 3828 (Center for Membrane Toxicity Studies). EFFECTS OF PROCAINE ON SECRETAGOGUE STIMULATED CHLORIDE SECRETION IN THE RECTAL GLAND OF THE DOGFISH SHARK, SQUALUS ACANTHIAS William Motley1, Suzanne J. Forrest3, Raquel Williams2, Sarah Decker1,3, John N. Forrest Jr. 3 1 Middlebury College, Middlebury VT, 05753 2Georgia Southern University, Statesboro, GA 30460 3Dept. of Medicine, Yale University School of Medicine, New Haven, CT 06510

C-type natriuretic peptide (CNP) stimulates chloride secretion in the shark rectal gland (Schofield et al., Am. J. Physiol., 261: F734, 1991; Solomon et al., Am. J. Physiol., 262: R707, 1992). There is evidence that CNP activates chloride secretion directly via a CNP receptor located on the basolateral membrane of the rectal gland epithelial cell (Aller et al; Am. J. Physiol., 276:C442-C449, 1999). It has been proposed that CNP also acts indirectly, by releasing vasoactive intestinal peptide (VIP) from nerves, leading to activation of CFTR via the basolateral VIP receptor. This hypothesis was supported in part by the observation that procaine, a local anesthetic, significantly inhibited CNP stimulated chloride secretion in the perfused shark rectal gland (Silva et al; Am. J. of Physiol., 277: R1725-R1732, 1999) whereas procaine appeared to be without effect on VIP stimulated secretion in earlier studies (Stoff et al; Am. J. Physiol., 255: R212-R216, 1988). In these previous experiments, the secretagogues (VIP and CNP) were given as 1 min bolus infusions and secretion was measured at 10 min intervals. We reasoned that if procaine selectively inhibited putative CNP-stimulated nerve release of VIP, then procaine should have no effect on either VIP or forskolin infused at constant concentrations. Therefore, we perfused rectal glands in vitro in paired experiments conducted by

Effects of 10mM Procaine on VIP (3nM) Stimulated Chloride Secretion

2500

Control 2000 Procaine Added

1500

1000 Chloride Secretion (µEq/h/g) 500

0 010203040506070 Time (min)

Figure 1. Effects of procaine (10mM) added at t=0 on VIP (3nM) stimulated chloride secretion. VIP was added to both groups at t=30. P<0.01 at 36, 37 and 50 min and P<0.05 at other time points. Data are means± standard error for n=6 paired experiments. blinded observers with methods described previously (Lehrich et al., J. Clin. Invest. 101:737-745, 1998). The effects of procaine on constant infusions of two known secretagogues, VIP and forskolin, were examined. In control experiments, after thirty minutes of basal perfusion, the secretagogue was perfused continuously for 30 min, with 1 min measurements from minutes 25-60. In the experimental perfusions, procaine was added in both the basal state and during the secretagogue infusions. The solution containing secretagogue was divided into two aliquots prior to the addition of procaine to ensure identical concentrations of secretagogue in the paired experiments.

Effects of 10mM Procaine on Forskolin (1µM) Stimulated Chloride Secretion

1000.00

900.00 Control 800.00 Procaine Added

700.00

600.00

500.00

400.00

300.00 Chloride Secretion (µEq/h/g) 200.00

100.00

0.00 010203040506070 Time (min)

Figure 2. Effects of procaine (10mM) added at t=0 on forskolin (1µM) stimulated chloride secretion. Forskolin was added to both groups at t=30. P<0.001 at points 36 and 37 min and P<0.01 from 38-40 min. Data are means ± standard error for.n=10 paired experiment

VIP and forskolin stimulated chloride secretion in the shark rectal gland were significantly inhibited by 10-2M procaine. The chloride concentration of the secreted fluid was also significantly reduced by the addition of procaine (data not shown). VIP acts on extracellular G-protein coupled VIP receptors, while forskolin acts at an intracellular site on the catalytic unit of adenylate cyclase. The effects of procaine to block significantly both a receptor agonist (VIP) and a secretagogue acting downstream at an intracellular second messenger site (forskolin) suggest that procaine is a non-specific inhibitor acting at another site(s) or that the anesthetic has a toxic metabolic effect on the shark rectal gland cell. Our experimental results differ from those reported previously by Stoff et al and Silva et al who observed that identical concentrations of procaine were without effect on VIP but inhibited CNP stimulated secretion. Our methods differed from the earlier reports in three regards: use of paired blinded experiments, constant infusion of secretagogues, and measurements at one minute intervals. In other systems, procaine has been shown to inhibit Na-K-ATPase (at 17 mM) (Kutchai et al., Pharmacol Res 43:399-403, 2001), voltage gated sodium and potassium channels (at 60µM) (Brau et al., Anesth Analg 87:885, 1998), calcium stores (Leppanen et al., J Neurophysiol, 78(4):2095-107, 1997), and calcium channels (Yoshino et al; Naunyn-Shmiedebergs Archives of Pharmacology, 353(3):334-41, 1996). These data suggest that procaine cannot be used to specifically inhibit neurotransmitter release in the perfused shark rectal gland. The results are also consistent with the view that CNP acts primarily, and possibly solely, on basolateral guanylate cyclase mediated CNP receptors in the shark rectal gland. This work was supported by NIH grants DK 34208 and NIEHS P30-ES 3828 (Center for Membrane Toxicology Studies). PEPTIDE HISTIDINE ISOLEUCINEAMIDE STIMULATES CHLORIDE SECRETION IN THE PERFUSED SHARK RECTAL GLAND

Sarah Decker1,2, Suzanne J. Forrest1, William Motley2, Raquel Williams3, John N. Forrest Jr. 1 1Dept. of Medicine, Yale University School of Medicine, New Haven, CT 06510 2 Middlebury College, Middlebury VT, 05753 3 Georgia Southern University, Statesboro, GA 30460

Peptide Histidine Isoleucineamide (PHI) is an endogenous ligand that activates VIP receptors in numerous tissues. PHI and VIP share a common precursor peptide in mammals and as we have previously shown, in the dogfish shark. We previously reported the full length cloning and sequence of the PHI hormone gene from shark brain using degenerate PCR and RACE-PCR (Plesch et al, Bull. MDIBL 39: 135-136, 2000). In the present experiments, we examined whether mammalian (porcine) PHI activated chloride secretion in the isolated perfused rectal gland. Glands were perfused with shark Ringers for thirty minutes, then PHI was constantly infused at a concentration from 1 to 10 nM for the next thirty minutes with readings at 1 min intervals. At a concentration of 1 nM, PHI was without effect on chloride secretion. This contrasts with 1nM porcine VIP, which stimulated chloride secretion to approximately 1400 µEqCl/h/g in parallel control experiments (n=4). PHI (2.5nM) elicited a slight increase in chloride secretion above basal values. At concentrations of 5 and 10 nM chloride secretion increased to approximately 800 and 1200 µEqCl/h/g respectively.

Dose Response to PHI

1600 PHI 1nM 1400 PHI 5nM 1200 PHI 10nM PHI 2.5nM 1000

800

600

400 Chloride Secretion (µEq/h/g) 200

0 20 30 40 50 60 Time (min)

Figure 1. Dose response to porcine PHI (1-10 nM) stimulated chloride secretion in the perfused shark rectal gland (n=3-6 glands at each concentration).

These experiments demonstrate that in the in vitro perfused shark rectal gland, mammalian PHI is a chloride secretagogue, although the peptide is less potent than VIP or pituitary adenylate cyclase activating peptide (PACAP). We predict that the potency order of mammalian ligands for the cloned shark VIP receptor will be VIP=PACAP>PHI.

This work was supported by NIH grants DK 34208 and NIEHS P30-ES 3828 (Center for Membrane Toxicology Studies). STAUROSPORINE AND CYTOCHALASIN D DO NOT INHIBIT THE CHLORIDE SECRETORY RESPONSE TO CONSTANT INFUSIONS OF SHARK C-TYPE NATRIURETIC PEPTIDE

Suzanne J. Forrest1, Andrea Pelletier1, Sarah Decker2, John N. Forrest Jr. 1 1Dept. of Medicine, Yale University School of Medicine, New Haven, CT 06510 2 Middlebury College, Middlebury VT, 05753

Silva et al. have proposed that the action of C-type natriuretic peptide in the shark rectal gland may involve protein kinase C and actin filaments because both staurosporin, an inhibitor of protein kinase C, and cytochalasin D, an inhibitor of actin filaments, blunt the response to boluses of CNP (Silva et al., Am. J. Physiol. 277: R1725-R1732, 1999; Silva et al., Bull MDIBL 39: 5-7, 2000). Because bolus injections result in variable and uncertain concentrations of these agents, we carried out experiments with staurosporine and cytochalasin D using constant infusions of secretagogues. Additionally, we performed blinded paired experiments (to control for factors such as temperature and observer bias) and measured chloride secretion rates at one minute intervals to provide greater sensitivity to small changes in secretion.

Staurosporine (10 nM) did not inhibit the chloride secretory response to 1.5 nM VIP (figure 1) or 3 nM CNP perfused at constant concentrations (figure 2). Similarly, cytochalasin D did not inhibit the chloride secretory response to a constant infusion of shark CNP (figure 3).

1200 VIP(1.5 nM)+Staurosporine(10 nM) n=4

1000 VIP(1.5 nM) n=4

800

600

400

200

0 0102030405060 Time (minutes)

Figure 1. Chloride secretory response to 1.5 nM VIP added at time=30 in the presence and absence of staurosporine (10nM) added at time=0. All perfusate also contained BSA 0.1mg/ml. 2000 CNP(3nM) n=6

1750 CNP(3nM)+Staurosporine 1500 (10nM) n=6 1250 1000 750 500 250 0 0102030405060 Time (minutes)

Figure 2. Chloride secretory response to 3 nM sCNP added at time=30 in the presence and absence of staurosporine (10nM) added at time=0. All perfusate also contained BSA 0.1mg/ml.

3000 2800 2600

2400 2200 CNP + CytochalasinD 2000 CNP 1800 1600

1400 1200 1000 800 600

400 200 0 010203040506070 Time(minutes) Figure 3. Chloride secretory response to 3 nM CNP added at time=30 in the presence and absence of cytochalasin D (1µM) added at time=0. All perfusate also contained BSA 0.1mg/ml. n=3 paired experiments.

In summary, using paired blinded experiments, one minute measurements and constant infusion of secretagogues and inhibitors, we did not observe effects of either staurosporine or cytochalasin D on CNP stimulated chloride secretion in the perfused gland. These studies raise the possibility that the action of CNP may not involve protein kinase C or actin filaments. Other techniques, such as antisense morpholino oligomers, may be necessary to resolve these possible protein-protein interactions.

This work was supported by NIH grants DK 34208 and NIEHS P30-ES 3828 (Center for Membrane Toxicology Studies). MDIBL REGISTER PAST PRESIDENTS PAST DIRECTORS

Dr. John S. Kingsley 1910-1922 Dr. Ulrich Dahlgren 1920-1926 Dr. Harold D. Senior 1922-1926 Dr. Herbert V. Neal 1926-1931 Dr. William Proctor 1926-1927 Dr. William H. Cole 1931-1940 Dr. Hermon C. Bumpus 1927-1932 Dr. Roy P. Forster 1940-1947 Dr. Warren H. Lewis 1932-1937 Dr. J. Wendell Burger 1947-1950 Dr. Ulrich Dahlgren 1937-1946 Dr. Warner F. Sheldon 1950-1956 Dr. Dwight Minnich 1946-1950 Dr. Raymond Rappaport 1956-1959 Dr. William H. Cole 1950-1951 Dr. Alvin F. Rieck 1959-1964 Dr. Homer W. Smith 1951-1960 Dr. William L. Doyle 1964-1967 Dr. Eli K. Marshall 1960-1964 Dr. Charles E. Wilde 1967-1970 Dr. Roy P. Forster 1964-1970 Dr. H. Victor Murdaugh 1970-1975 Dr. William L. Doyle 1970-1975 Dr. Richard M. Hays 1975-1979 Dr. Jack D. Myers 1975-1978 Dr. Leon Goldstein 1979-1983 Dr. Charles E. Wilde 1978-1979 Dr. David H. Evans 1983-1992 Dr. Raymond Rappaport 1979-1981 Dr. David C. Dawson 1992-1998 Dr. Bodil Schmidt-Nielsen 1981-1985 Dr. John N. Forrest, Jr. 1998- Dr. Franklin H. Epstein 1985-1995

CHAIR OF THE BOARD Dr. James L. Boyer 1995-

2001-2002 OFFICERS Chair, Board of Trustees Dr. James L. Boyer Vice Chair Dr. Rolf K.H. Kinne Director Dr. John N. Forrest, Jr. Secretary Dr. John H. Henson Treasurer Mr. Richard J. Wright Clerk Nathaniel I. Fenton, Esq.

EXECUTIVE COMMITTEE DIRECTOR’S ADVISORY COMMITTEE Dr. James L. Boyer, Chair Dr. John N. Forrest, Jr., Chair Dr. Edward J. Benz Dr. Edward J. Benz Dr. John N. Forrest, Jr., Ex Officio Dr. James L. Boyer Dr. Patricia H. Hand, Ex Officio Dr. Garry R. Cutting Dr. Barbara Kent, Ex Officio Dr. David H. Evans Dr. Rolf K.H. Kinne Dr. Biff Forbush, III Mrs. Emily Leeser Dr. Raymond A. Frizzell Mr. Richard Wright Dr. Patricia H. Hand, Ex Officio Dr. Barbara Kent Dr. David S. Miller Dr. J. Larry Renfro Dr. David W. Towle Administrative Director Dr. Patricia H. Hand TRUSTEES

Class of 2002

James L. Boyer, M.D. Mr. Alan B. Miller Ensign Professor of Medicine Business Finance & Restructuring Chief, Div. of Digestive Diseases Weil, Gotshal & Manges LLP Yale University School of Medicine New York, New York

John N. Forrest, Jr., M.D. Frank Moya, M.D. Professor, Dept. of Internal Medicine Chairman Emeritus Yale University School of Medicine Department of Anaesthesiology Mount Sinai Medical Center; Rolf Kinne, M.D., Ph.D. Chairman & President Director, Max-Planck Institute Frank Moya, M.D. & Assoc., P.A. Fuer Molekulare Physiologie Miami, Florida Dortmund, Germany

Jessica Lewis, M.D. Mr. Richard J. Wright Prof. Emeritus, Department of Medicine D.P. Investments Corp. University of Pittsburgh; New York, New York Salisbury Cove, Maine and Pittsburgh, Pennsylvania

Class of 2003

Edward J. Benz, Jr., M.D. Mrs. Emily Leeser William Osler Professor and Chair Mount Desert, Maine and Department of Medicine New York, New York Professor of Molecular Biology & Genetics Johns Hopkins University School of Medicine Mrs. Edith Rudolf Mount Desert, Maine and Raymond A. Frizzell, Ph.D. New York, New York Professor and Chairman Department of Cell Biology & Physiology Ms. Susan Scherbel School of Medicine Hancock Point, Maine and University of Pittsburgh New York, New York

Neil Smith, M.D. Rockport, Maine Class of 2004

Biff Forbush III, Ph.D. Barbara Kent, Ph.D. Professor and Director of Graduate Studies Mount Desert Island Biological Laboratory Dept. of Cellular and Molecular Physiology Yale University School of Medicine

John H. Henson, Ph.D. Associate Professor Department of Biology Dickinson College

Class of 2005

J.H. Michael Agar Spencer Ervin West Tremont, ME Bass Harbor, ME

Terence Boylan Murk-Hein Heinemann, M.D. Rhinebeck, NY Associate Professor of Opthamology Associate Attending Opthamologist Franklin H. Epstein, M.D. Cornell University Medical College William Applebaum Professor of Medicine Beth Israel Hospital Harvard Medical School SCIENTIFIC PERSONNEL

Principal Investigators Associates

Lisa Bain, Ph.D. N. Maples Assistant Professor J. Peterson Department of Environmental Toxicology Clemson University

William Baldwin, Ph.D. J. Roling Assistant Professor Department of Environmental Toxicology Clemson University

Ned Ballatori, Ph.D. R. Anderson Associate Professor of Toxicology C. Bennet Department of Environmental Medicine J. McCoy University of Rochester School of Medicine D. Seward

David Barnes, Ph.D. Director National Stem Cell Resource American Type Cell Culture

Edward J. Benz, Jr., M.D. K. Barnes President A. Blakaj Dana Farber Cancer Institute

Nancy Berliner, M.D. Professor of Internal Medicine and Genetics Department of Internal Medicine Yale University School of Medicine

James L. Boyer, M.D. Ensign Professor of Medicine Chief, Division of Digestive Diseases Director Liver Center Yale University School of Medicine

Peter M. Cala, Ph.D. W. Gochberg Professor and Chairman S. King Dept. of Human Physiology R. Rigor School of Medicine Z. Zhuang University of California, Davis Gloria V. Callard, Ph.D. B. Lutton Professor of Biology J. St. George Boston University

James B. Claiborne, Ph.D. S. Edwards Professor of Biology N. Hair Georgia Southern University S. Sligh J. Weakley

Lars Cleemann, Ph.D. V. Degtiar Assistant Professor of Pharmacology P. Sasse Georgetown University S.H. Woo

Elizabeth L. Crockett, Ph.D. R.P. Hassett Assistant Professor J. Collins Department of Biological Sciences Ohio University

Gary R. Cutting, M.D. Professor of Pediatrics and Medicine Aetna/US Healthcare Professor of Medical Genetics Director, DNA Diagnostic Laboratory Institute of Genetic Medicine Johns Hopkins University School of Medicine

David C. Dawson, Ph.D. Professor and Chair Department of Physiology and Pharmacology School of Medicine Oregon Health Sciences University

James Denegre, Ph.D. New Investigator The Jackson Laboratory

Jonathan A. Dranoff, M.D. M. Clay Associate Research Scientist L. Petell Yale University School of Medicine K. Zarella

Marlies Elger, Ph.D. Research Scientist Innese Medizin/Nephrologie Medizinische Hochschule Hannover Franklin H. Epstein, M.D. J. Richard William Applebaum Professor of Medicine M. Schnermann Beth Israel Hospital C. Sighinolfi Harvard Medical School K. Spokes

David H. Evans, Ph.D. J. Catches Professor and Chair of Zoology M. Kozlowski University of Florida J. Roeser

Susan K. Fellner, Ph.D. L. Parker Research Professor Department of Cellular and Molecular Physiology University of North Carolina at Chapel Hill

Bliss Forbush III, Ph.D. A. Bewley Professor and Director of Graduate Studies B. Dowd Department of Cellular and Molecular Physiology I. Gimenez Yale University School of Medicine M. Pedersen

John N. Forrest, Jr., M.D. A. Bewley Professor of Medicine S. Decker Director of Student Research S. Forrest Department of Internal Medicine W. Motley Yale University School of Medicine J. Pena D. Philpotts G. Weber R. Williams M. Zikusoka

Leonard R. Forte, Ph.D. Professor and Senior Career (VA) Scientist Department of Pharmacology Missouri University School of Medicine Harry S. Truman Memorial Veterans' Hospital

Raymond A. Frizzell, Ph.D. K. Peters Professor and Chair Department of Cell Biology and Physiology University of Pittsburgh School of Medicine

Leon Goldstein, Ph.D. M. George Professor E. Hawrot Department of Molecular Pharmacology D.L. Koomoa Physiology & Biotechnology M. Musch Brown University Helene Guizouarn, Ph.D. D. Myers Chargée de Rechuche CNRS Laboratoire Jean Maetz Laboratorie des Physiologie des Membranes Cellulaires

Hermann Haller, M.D. J. Litteral Professor of Medicine B. Lauterbach Dept. of Nephrology M. Loehn Hannover Medical School A. Sheldon P. Waldron M. Wellner

Criss Hartzell, Ph.D. Professor Department of Cell Biology Emory University

Richard M. Hays, M.D. Investigator and Professor of Medicine Department of Medicine Albert Einstein College of Medicine

Raymond P. Henry, Ph.D. C. Booth Professor H. Skaggs Department of Biological Sciences C. Smith Auburn University

John H. Henson, Ph.D. S. Kolnik Visiting Professor V. Trabosh School of Biological Sciences University of East Anglia

Hartmut Hentschel, Ph.D. Research Scientist Innese Medizin/Nephrologie Medizinische Hochschule Hannover

Karl J. Karnaky, Ph.D. E. Milner Associate Professor Department of Cell Biology and Anatomy Medical University of South Carolina George W. Kidder III, Ph.D. Instrumentation Officer Senior Scientist Mount Desert Island Biological Laboratory

Rolf K.H. Kinne, M.D., Ph.D. T. Althoff Director S. Rosin-Steiner Max-Planck-Institute for Molecular Physiology Dortmund, Germany

Evamaria Kinne-Saffran, M.D., Ph.D. Senior Scientist Max-Planck-Institute for Molecular Physiology Dortmund, Germany

Thomas J. Koob, Ph.D. L. Koob-Edmunds Section Chief, Skeletal Biology J. Long Shriners Hospital for Children

Gregg A. Kormanik, Ph.D. Professor and Chair Department of Biology University of North Carolina - Asheville

Seth W. Kullman, Ph.D. Assistant Research Professor Environmental Sciences and Policy Nicholas School for the Environment and Earth Sciences Duke University

Dietmar Kültz, Ph.D. S. Kültz Assistant Professor of Physiology The Whitney Laboratory University of Florida

Petra H. Lenz, Ph.D. D. Hartline Associate Research Professor A. Hower Békésy Laboratory of Neurobiology Pacific Biomedical Research Center University of Hawaii

John E. Mickle, Ph.D. Instructor Department of Pediatrics Institute of Genetic Medicine Johns Hopkins University School of Medicine David S. Miller, Ph.D. B. Bauer Research Physiologist C. Breen Laboratory of Pharmacology and Chemistry S. Notenboom NIH/NIEHS

Martin Morad, Ph.D. Professor of Pharmacology and Medicine Department of Pharmacology Georgetown University Medical Center

Alison I. Morrison-Shetlar, Ph.D. Associate Professor of Biology Director of Faculty Development Department of Biology Georgia Southern University

Robert L. Preston, Ph.D. A. Budhai Professor of Physiology A. Bryant Department of Biological Sciences I. Sud Illinois State University V. Williams

J. Larry Renfro, Ph.D. R. Pelis Professor of Physiology Department of Physiology & Neurobiology University of Connecticut

Patricio Silva, M.D. Chief, Section of Nephrology Temple University Hospital

Bruce A. Stanton, Ph.D. A. Lankowski Professor M. Emerson Department of Physiology Dartmouth Medical School

James D. Stidham, Ph.D. Professor of Biology Presbyterian College

David W. Towle, Ph.D. S. King Senior Scientist P. Peppin Mount Desert Island Biological Laboratory K. Townshend Alice R.A. Villalobos, Ph.D. C. Dehm Assistant Professor Department of Physiology and Neurobiology University of Connecticut

Edward J. Weinman, M.D. T. Chiapparelli Professor of Medicine D. Steplock Division of Nephrology University of Maryland School of Medicine

Richard N. Winn, Ph.D. Associate Professor School of Forest Resources University of Georgia

John Wise, Ph.D. P. Antonucci Assistant Professor A. Bidulescu Department of Epidemiology and Public Health B. Bryant Yale University School of Medicine S. Holt K. Schumm E. Tang 2001 SUMMER FELLOWSHIP RECIPIENTS

Hancock County Scholars Recipients: Mentors:

Catherine Bennet, Choate Rosemary Hall Ned Ballatori, Ph.D. James L. Boyer, M.D. Michael Clay, Sumner Memorial High School Jonathan Dranoff, M.D. William Gochberg, Cretin Durham Hall Peter M. Cala, Ph.D. Shawna King, MDI High School David W. Towle, Ph.D. Jana Richards, Bucksport High School Franklin H. Epstein, M.D. Kyle Schumm, Charter Oak High School John Wise, Ph.D. Abigail Sheldon, C.E. Jordan High School Hermann Haller, M.D. Katherine Smith, University Laboratory High School Raymond Henry, Ph.D.

Community Environmental Toxicology Laboratory:

Melissa Brown, MDI High School Janet Redman Meghan Burgess, MDI High School Janet Redman Joshua Collins, MDI High School Elizabeth Crockett, Ph.D. R. Patrick Hassett, Ph.D. Amber Howard, MDI High School Janet Redman Jennifer Reynolds, MDI High School Janet Redman

Durham High School Recipients:

Jesse McCoy, Hillside High School Ned Ballatori, Ph.D. James L. Boyer, M.D. Samuel Sligh, Hillside High School J.B. Claiborne, Ph.D. Allison Morrison-Shetlar, Ph.D. Ishani Sud, Hillside High School Robert L. Preston, Ph.D. Velvet Williams, Hillside High School Robert L. Preston, Ph.D.

Secondary Education through Health:

Alexandra Budhai, Brooklyn Technical High School Robert L. Preston, Ph.D.

Trinity High School Recipients:

William Motley, Trinity High School John N. Forrest, Jr., M.D. Denise Philpotts, Trinity High School John N. Forrest, Jr., M.D.

Maine Research Internships for Teachers and Students (MERITS):

Cynthia Dehm, Central High School Alice Villalobos, Ph.D. Chris Sighinolfi, Brewer High School Franklin H. Epstein, M.D. Undergraduate Student Fellowships (Supported by NSF):

Peter Antonucci, Southern CT State University John Wise, Ph.D. Amanda Bryant, Illinois State University Robert L. Preston, Ph.D. Bronwyn Bryant, Brown University John Wise, Ph.D. Justin Catches, University of Florida J.B. Claiborne, Ph.D. Sarah Decker, Middlebury College John N. Forrest, Jr., M.D. Maude Emerson, Harvard University Garry Cutting, M.D. Bruce A. Stanton, Ph.D. Matthew Kozlowski, University of Florida David A. Evans, Ph.D. Emily Milner, Presbyterian College Karl Karnaky, Ph.D. Laurel Parker, Bodowin College Susan K. Fellner, Ph.D. Lydia Petell, Bates College Jonathan Dranoff, M.D. Jennifer Roeser, University of Florida David A. Evans, Ph.D. Kristy Townsend, University of Maine David A. Towle, Ph.D. Patricia Waldron, Grinnell College Hermann Haller, M.D. Jill Weakley, Georgia Southern University J.B. Claiborne, Ph.D. Raquel Williams, Georgia Southern University J.B. Claiborne, Ph.D. Katrina Zarella, College of the Atlantic Jonathan Dranoff, M.D.

Stanley Bradley/Stan and Judy Fellowship:

Rachel Anderson, Brown University Ned Ballatori, Ph.D. James L. Boyer, M.D. 2001 SEMINARS

Seminars preceded by an asterisk were presented by investigators supported by the NIEHS Center for Membrane Toxicity Studies at the Mount Desert Island Biological Laboratory

Morning Transport Seminars

*July 2 “Mechanisms and control of sulfate regulation in flounder” J. Larry Renfro, Ph.D., Professor of Physiology, Department of Physiology and Neurobiology, University of Connecticut

July 9 “Volume regulatory ion flux pathways: Conflicting demands and coordinated control” Peter M. Cala, Ph.D., Professor and Chair, Department of Human Physiology, School of Medicine, University of California, Davis

July 16 “How do cells regulate their osmolyte channels?” Leon Goldstein, Ph.D., Professor and Vice Chair, Department of Physiology, Brown University

*July 23 “Motoring to the membrane” Richard M. Hays, M.D., Professor of Medicine, Albert Einstein College of Medicine

July 30 “Paracrines and gill transport” David H. Evans, Ph.D., Professor and Chair, Department of Zoology, University of Florida

August 6 “Ion channel remodeling in cardiac hypertrophy and failure” Martin Morad, Ph.D., Professor of Pharmacology and Medicine, Georgetown University Medical Center

*August 13 “CFTR: What’s it really like inside the pore?” David C. Dawson, Ph.D., Professor and Chair, Department of Physiology and Pharmacology, Oregon Health Sciences University

*August 20 “PDZ proteins and CFTR trafficking” Bruce A. Stanton, Ph.D., Professor, Department of Physiology, Dartmouth Medical School

*August 27 Of strings and SNAREs: Two tails of CFTR trafficking” Raymond A. Frizzell, Ph.D., Professor and Chair, Department of Cell Biology and Physiology, School of Medicine, University of Pittsburgh

Friday Noon Brown Bag Seminars

July 6 Introductory 5 minute talks by MDIBL Principal Investigators to summarize summer research projects July 13 Remainder of introductory 5 minute talks by MDIBL Principal Investigators

July 27 “Ten years of CF in ten minutes” John Mickle, Ph.D., Instructor, Institute of Genetic Medicine, Johns Hopkins University School of Medicine

*“P2Y receptors and signaling” Jon Dranoff, M.D., Assistant Professor of Medicine, Yale University School of Medicine

“Sodium-hydrogen exchangers in gill” Susan Edwards, Ph.D., Visiting Assistant Professor, Biology Department, Georgia Southern University

Wednesday Evening Public Seminars

July 3 “Molecular physiology of membrane transport in crab gills” David W. Towle, Ph.D., MDIBL Senior Scientist and Director of the Marine DNA Sequencing Center

*July 11 THE TWENTIETH WILLIAM B. KINTER MEMORIAL LECTURE - “Steroid and xenobiotic receptors – how a cell senses its chemical environment” Ronald M. Evans, Ph.D., Professor, Gene Expression Laboratory, The Salk Institute for Biological Studies, LaJolla, California

July 18 THE SEVENTH HELEN F. CSERR MEMORIAL LECTURE – Beyond the Pale: Membrane Biogenesis and Stability in Anemic and Pigment-Deficient Mice” Luanne L. Peters, Ph.D., Staff Scientist/Program Director, Center for Mouse Models of Heart, Lung, Blood and Sleep Disorders, The Jackson Laboratory

*July 25 THE ELEVENTH THOMAS H. MAREN MEMORIAL SEMINAR - “The Scent of an Ion: Calcium Sensing from Fish to Man” Steven C. Hebert, M.D., Professor and Chairman, Department of Cellular and Molecular Physiology, Yale University School of Medicine.

*July 30 THE SIXTH LEONARD SILK MEMORIAL LECTURE – “A Question of Intent” David A. Kessler, M.D., Dean; Professor of Pediatrics, Medicine, and Public Health, Yale University School of Medicine

*August 8 THE NINETH JOHN W. BOYLAN MEMORIAL LECTURE – “The Dependency of Human Health on the Health of Other Species” Eric Chivian, M.D., Director, Center for Health & the Global Environment; Assistant Clinical Professor of Psychiatry, Harvard Medical School. In 1980, Dr. Chivian co- founded International Physicians for the Prevention of Nuclear War, which was the recipient of the 1985 Nobel Peace Prize August 15 “Molecular Basis for the Selectivity of Paracellular Transport” James Anderson, Ph.D., M.D., Chief, Section of Digestive Diseases; Professor, Internal Medicine and Cell Biology, Yale University School of Medicine

*August 22 “Structural basis for alpha-neurotoxin recognition of nicotinic acetylcholine receptors” Edward Hawrot, Ph.D., Upjohn Professor in Pharmacology; Professor of Medical Science; Chair, Department of Molecular Pharmacology, Physiology & Biotechnology, Brown University

Special Seminars and Presentations

*July 20, 1:00 p.m. “Regulation of Drug Export Pump function in Renal Proximal Tubule” David S. Miller, Ph.D., Research Physiologist, Laboratory of Phrmacology and Chemistry, NIH, National Institute of Environmental Health Sciences

July 20, 8:00 p.m. “Ancient Perspectives on Future Climate” Daniel Schrag, Ph.D., Professor of Earth and Planetary Sciences, Harvard University

August 2, 7:30 p.m., “Future Science: Scientists of Tomorrow at Work Today” Public seminar featuring presentations by local and national high school and college students participating in exciting hands-on research training programs at MDIBL and the Jackson Laboratory

August 7, 1:00 p.m. “A Cell-Mediated Gene Transfer Approach to the Production of Transgenic and Knockout Lines of Fish” Paul Collodi, Ph.D., Associate Professor, Department of Animal Sciences, Purdue University

August 28, 4:00 p.m. “A conversation with Harold Varmus, M.D.” Informal question and answer session with Harold Varmus, M.D., the President of Memorial Sloan Kettering; former Director of the National Institutes of Health; recipient of the 1989 Nobel Prize in Medicine.

All Campus Ethics Seminar Series “Responsible Conduct in Research”

July 10 Session #1 - hosted by J.B. Claiborne, Ph.D., Professor of Biology, Georgia Southern University, and David H. Evans, Ph.D., Professor and Chair, Department of Zoology, University of Florida

July 17 Session #2 – hosted by Susan Edwards, Ph.D., Visiting Assistant Professor, Georgia Southern University, and John Mickle, Ph.D., Instructor, Institute of Genetic medicine, johns hopkins University School of Medicine July 24 Session #3 – hosted by Jonathan Dranoff, M.D., Assistant Professor of Medicine, Yale University School of Medicine, and Alice Villalobos, Ph.D., Assistant Professor, Department of Environmental Medicine, University of Rochester School of Medicine

2001 WORKSHOPS AND CONFERENCES Workshop *May 4-6 Generation of New Marine Cell Lines and Transgenic Species Generation of New Marine Cell Lines Co-Chairs: J. Larry Renfro, Ph.D., University of Connecticut, MDIBL; Dr. John Wise, Ph.D., Yale University School of Medicine, MDIBL "Fish cell culture: advantages and limitations", David Barnes, Ph.D., American Type Cell Culture "Development and use of fish cell lines", Neils Bols, Ph.D., University of Waterloo "Endothelial cells isolated from teleosts", Rita Garrick, Ph.D., Fordham University "Two and three dimensional marine mammal cell culture models for potential assessment of marine environmental toxicology", Thomas Goodwin, Ph.D., NASA "Telomerase expression as a tool to create continuously proliferating normal cell lines", Shawn Holt, Ph.D., Medical College of Virginia Plenary Speaker — Allan Teasley, Ph.D., National Institute of Environmental Health Sciences Generation of New Transgenic Marine Species Co-Chairs: Thomas Chen, Ph.D., University of Connecticut; Richard Winn, Ph.D., University of Georgia "Pollution inducible transgenes in zebrafish: a work in progress", Michael Carvan, Ph.D., University of Wisconsin — Milwaukee "Recent advances in production of transgenic aquatic organisms and its biotechnological applications", Thomas Chen, Ph.D., University of Connecticut "Development of genetic tools for genomic analyses and transgenesis in fish", Perry Hackett, Ph.D., University of Minnesota "Atlantic coast marine models for transgenic research in fish", Robin Wallace, Ph.D., University of Florida "Development of transgenic fish models for biomedical and environmental research: myths, realities, and opportunities", Richard Winn, Ph.D., University of Georgia Breakout Sessions "Strategies for New Marine Cell Culture Lines", Co-Chairs: J. Larry Renfro, Ph.D., University of Connecticut, MDIBL; John Wise, Ph.D., Yale University School of Medicine, MDIBL "Strategies for New Marine Transgenic Species", Co-Chairs: Thomas Chen, Ph.D., University of Connecticut; Richard Winn, Ph.D., University of Georgia "Strategies for Improving Resources/Collaboration/Funding", Co-Chairs: John Stegemen, Ph.D., Woods Hole Oceanographic Institute; John N. Forrest, Jr., M.D., Yale University School of Medicine, MDIBL

Symposia * July 11-12 EIGHTH ANNUAL MOUNT DESERT ISLAND BIOLOGICAL LABORATORY (MDIBL) ENVIRONMENTAL HEALTH SCIENCES SYMPOSIUM. Sponsored by the NIEHS Center for Membrane Toxicity Studies at the MDIBL, the Kinter Memorial Lectureship Fund, and the MDIBL Intracellular Sensors of Environmental Chemicals Keynote Speaker — "Steroid and Xenobiotic Receptors – how a cell senses its chemical environment" Ronald M. Evans, M.D., The Salk Institute "Dioxin, clocks and oxygen: Prototype signals of a nuclear sensor superfamily", Christopher A. Bradfield, Ph.D., McArdle Laboratory, Univ. of Wisconsin, Madison "Evolution and function of the aryl hydrocarbon (AhR) receptors", Mark Hahn, Ph.D., Woods Hole Oceanographic Institution "Use of genetically altered mice to define the function of xenobiotic receptors", Christopher J. Sinal, Ph.D., National Cancer Institute "PXRs, CARs, and PPARs: Broad specificity nuclear receptors", Steven A. Kliewer, Ph.D., Glaxo Wellcome "Mechanisms and pathophysiology of the PPARs", Janardan K. Reddy, M.D., Northwestern University "Membrane transporters as xenobiotic sensors: BSEP and its regulation by FXR", James L. Boyer, M.D., Yale University School of Medicine and MDIBL "The MRP and MDR efflux transporters", John Riordan, Ph.D., The Mayo Clinic, Scottsdale and MDIBL "The evolution of estrogen receptors", Peter Thomas, Ph.D., University of Texas at Austin "Xenoestrogens and mammalian estrogen receptors", Tim Zacharewski, Ph.D., Michigan State University "Estrogen receptors in fish", Gloria Callard, Ph.D., Boston University and MDIBL * July 24 Biology of Killifish: Fundulus heteroclitus Organizer – John Mickle, Ph.D., Johns Hopkins University School of Medicine

Opening Remarks: “MDIBL and Fundulus heteroclitus” John N. Forrest, Jr., M.D., Yale University School of Medicine; Director, MDIBL

Substrate secretion “Opercular epithelium/chloride secretion” Karl J. Karnaky, Ph.D. Medical University of South Carolina

Cytoplasmic signaling pathways “Phosphoprotein regulation” Dietmar Kültz, Ph.D., The Whitney Laboratory, University of Florida

Membrane proteins “Xenobiotic transporters MRP2, Pgp” David S. Miller, Ph.D., NIH/NIEHS

Ion channels NaK2Cl – Biff Forbush, Ph.D., Yale University School of Medicine NHE – J.B. Claiborne, Ph.D., Georgia Southern University CFTR – John Mickle, Ph,.D., Johns Hopkins University School of Medicine

Extracellular signaling pathways “Nucleotide receptors” Jonathan A. Dranoff, M.D., Yale University School of Medicine

“Paracrine control in gill” David H. Evans, Ph.D., University of Florida

Behavioral osmoregulation George W. Kidder III, Ph.D., MDIBL

Closing Remarks: Karl J. Karnaky, Ph.D.

2001 COURSES

May 27-June 9 Illinois State University, Physiology of Marine and Maritime Animals Course Directors - George Kidder, Ph.D. and Robert Preston, Ph.D. June 2-9 University of Pittsburgh School of Medicine, Intensive Pedagogical Experience in Laboratory Research Course Directors - Raymond Frizzell, Ph.D. and Mark Zeidel, M.D. June 11-16 Yale University School of Medicine, Intensive Pedagogical Experience in Laboratory Research Course Directors - John N. Forrest, Jr., M.D. and David Kessler, J.D., M.D. *June 16-22 Third Annual Intensive Course in Quantitative Fluorescent Microscopy Course Director - Simon C. Watkins, Ph.D., University of Pittsburgh School of Medicine *June 20 The Polymerase Chain Reaction Protocol—mini course for high school biology students Faculty - David Towle, Ph.D., Jennifer Litteral, Michael McKernan, Christine Smith June 25-26 Marine Physiology and Molecular Biology for high school summer fellows Faculty - J.B. Claiborne, Ph.D. and Allison Morrison-Shetlar, Ph.D. September 5-10 Biomedical Exchange Program Forum and International Academy of Life Sciences Meeting Course Director - Hilmar Stolte, M.D. PUBLICATIONS

Ballatori, N., Connolly, G.C., Boyer, J.L. Elasmobranch hepatocytes. In: The Hepatocyte Review. Editors: Berry MN, Edwards AM. Kluwer Academic Publishers, 2000:37-47.

Cai, S-Y., Wang, L., Ballatori, N., Boyer, J.L. Bile salt export pump is highly conserved during vertebrate evolution and its expression is inhibited by PFIC type II mutations. Am J Physiol Gastrointest Liver Physiol 281: G316-G322, 2001.

Cai, S-Y., Wang, L., Ballatori, N., Boyer, J.L. Molecular identification and functional characterization of the bile export pump (Bsep) in a marine vertebrate – evidence that this liver-specific transporter is highly conserved in evolution. In: Biology of Bile Acids in Health and Disease, Falk Symposium 120. Editors: van Berge Henegouwen GP, Keppler D, Leuschner U, Paumgartner G, Stiehl A. Kluwer Academic Publishers, 2001:113-115.

Edwards, S ., Cl aibor ne, J. B. , Morr ison- Shet lar , A. I. , and Toop, T . Expressi on of Na+/ H+ exchanger mRNA in the gil ls of the Atl ant ic hagfi sh (Myxi ne gl ut inosa) i n r esponse to metabol ic acidosis. Comp. Bi ochem. & Physiol . 130: 81-91, 2001.

Hammond, C.L. , Lee, T.K. , and Ballatori, N. Novel roles for glut athione in gene expression, cell death, and membrane transport of organic solutes. J. Hepatology 34: 946- 954, 2001.

Karnaky, K. J., Jr. Teleost chloride cell tight junctions: environmental salinity and dynamic structural changes. In: Tight Junctions (2nd edition). (Ed. by M. Cereijido and J. Anderson) CRC Press, pp. 445-458, 2001.

Kinne-S aff ran, E. and Kinne, R.K.H. Inhibition by mercuric chloride of Na-K-2CL cotr anspor t act ivi ty in rect al gland pl asm a mem brane vesicles i sol at ed from Squalus acanthi as. Bi ochimica et Bi ophysica Acta 1510: 442-451, 2001.

Ki shida M, Call ard GV. Distinct cytochr ome P 450 ar om atase isof orm s i n zebr afi sh (Dani o reri o) br ain and ovary are differ ent ial ly progr am med and estrogen regulat ed during early development. Endocrinol ogy 142:740-750, 2001.

Ki shida M, McLellan M, Mir anda JA, Call ard GV. Est rogen and xenoest rogens upregul ate the brain ar om atase isof orm (P 450ar omB) and pert urb markers of earl y development in zebrafi sh (Danio rerio ). Comp Physiol Biochem 129/2- 3:261-268, 2001.

Liu, J., Chen, H., Miller, D.S., Saavedra, J.E., Keffer, L.K., Johnson, D.R., Klaasen, C.D., and Waalkes, M.P. Overexpression of glutathione S-transferase Pi and multidrug resistance transport proteins is associated with acquired tolerance to inorganic arsenic. Mol. Pharmacol., 60:303-309, 2001. Miller, D.S. Nucleoside phosphonate interactions with multiple organic anion transporters in renal proximal tubule. J. Pharmacol. Exp. Therap., 299:567-574, 2001.

Sweet, D.H., Miller, D.S. and Pritchard, J.B. Ventricular choline transport: A role for organic cation transporter 2 expressed in choroid plexus. J. Biol. Chem, 276:41611- 41619, 2001.

Terloux, S. Masereeuw, R., Russel, F.G.M., and Miller, D.S. Nephrotoxicants induce endothelin release and signaling in renal proximal tubules: effects on drug efflux. Mol. Pharmacol., 59:1433-1400, 2001.

Towl e, D.W., R. S. Paul sen, D. Weihrauch, M. Kordylewski, C. Salvador, J. -H. Li gnot, and C. Spanings-P ierrot. 2001. Na++K+-ATPase in gills of the blue cr ab Callinectes sapidus : cDNA sequencing and sal ini ty-relat ed expression of A-subunit mRNA and protei n. J. E xp. Bi ol . 204: 4005-4012.

Qu, Z. and H.C. Hartzell. Functional geometry of the permeation pathway of Ca- activated Cl channels inferred from analysis of voltage-dependent block. J. Bio. Chem. 276, 18423-18429, 2001.

Wang, W., Seward, D.J., Li, L., Boyer, J.L., and Ballatori, N. Expression cloning of two genes that together mediate organic solute and steroid transport in the liver of a marine vertebrate. PNAS 98: 9431-9436, 2001.

Weihrauch, D., A. Zi egler , D. Si ebers, and D. W. Towl e. Molecul ar characteri zat ion of V- type H+-AT Pase (B- subunit ) i n gil ls of euryhal ine crabs and it s physiological rol e in osmoregulatory ion uptake. J. Exp. Biol . 204: 25-37, 2001.

Werner, A., and Kinne, R.K.H. Evolution of the Na-P i cotr anspor t system s. Am. J. Physiol . - Regul atory, I ntegrati ve, and Comparat ive P hysiol ogy 280: R301-312, 2001. AUTHORS

Althoff, Thorsten 56 Findlay, Kathleen 99 Anderson, Rachel 75 Fleischmann, Bernd 84 Antonucci, Peter 48 Forbush, Biff 50 Baldwin, William 40 Forrest, John N., Jr. 26, 45, 48 Bain, Lisa 39 81, 103 Ballatori, Ned 33, 37, 73 105, 108 75 109 Barnes, David 45 Forrest, Suzanne 105, 108 Bauer, Bjorn 36 109 Bennett, Catherine 75 Forte, Leonard 26 Bewley, Arnaud 50 Fricker, Gert 36, 73 Bewley, Marie 103 Frizzell, Raymond 81 Boyer, James 33, 37, 48 George, Megan 18 73, 75 Goldstein, Leon 18, 25, 61 Bryant, Amanda 65 Greenburg, Robert 60 Bryant, Bronwyn 48 Guizouarn, Hélène 25 Budhai, Alexandra 65 Hair, Nicole 21 Cai, Shi-Ying 33, 37, 75 Haller, Hermann 101 Cala, Peter 31 Hartline, Daniel 54, 100 Carina, Ammälä 88 Hartzell, H. Criss 71 Chakravarty, Devulapalli 60 Hassett, R. Patrick 57 Chau, My 92 Hays, Richard 1, 3, 79 Claiborne, J.B. 11, 20, 21, He, Li-Ping 88 23 Henry, Raymond 12, 52, 54 Clay, Michael 15 Henson, John 37, 73 Cleeman, Lars 84, 88, 92 Hentschel, Hartmut 101 Crockett, Elizabeth 57 Hescheler, Jürgen 84 Cutting, Garry 77 Hinton, David 43 Decker, Sarah 105, 108 Holt, Shawn 48 109 Hower, Amy 100 Degtiar, Vadim 92 Jones, Emily 92 Dehm, Cynthia 41 Karnaky, Karl 26 Dranoff, Jonathan 15 Kidder, George 99 Dowd, Brian 50 King, Scott 31 Dukes, Iain 88 King, Shawna 54 Edwards, Susan 11, 20, 21 Kinne, Rolf 56 Elger, Marlies 101 Kinne-Saffran, Eva 56 Elmore, Lynn 48 Kirsch, Torsten 101 Emerson, Maude 77 Kohn, Andrea 60 Ensign, Kimberly 99 Kolnik, Sarah 37 Epstein, Franklin 1, 3, 79 Koob, Thomas 62 Evans, David 7, 8, 9, 96, Koob-Edmunds, Magdalena 62 97 Koomoa, Dana-Lynn 18 Fellner, Susan 5 Kormanik, Gregg 19 Kozlowski, M.S. 96 Silva, Patricio 1, 3, 79 Kullman, Seth 43 Skaggs, Hollie 12 Kültz, Dietmar 60 Sligh, Samuel 23 Lankowski, Alexander 15, 77 Smith, Katherine 12 Lenz, Petra 100 Soroka, Carol 33 Litteral, Jennifer 101 Spannings-Pierrot, Celine 54 Luig, Jutta 56 Spokes, Katherine 1, 3, 79 Maples, Nikki 39 Stanton, Bruce 77 Marino, Christopher 81 Steplock, Deborah 67 Masereeuw, Rosalinde 35 Stidham, James 7, 8, 9 McCoy, Jesse 75 Sud, Ishani 65 Mickle, John 15, 77 Towle, David 17, 54 Miller, David 35, 36, 41 Townsend, Kristy 54 73 Villalobos, Alice 41 Milner, Emily 26 Waldron, Patricia 101 Morad, Martin 84, 92 Weakley, Jill 23 Morrison-Shetlar, Alison 11, 20, 21, Weber, Gerhard 103 23 Weinman, Edward 67 Motley, William 105, 108 Wellner, Maren 101 Musch, Mark 61 Williams, Raquel 105, 108 Myers, Daniella 25 Williams, Velvet 65 Notenboom, Sylvia 35 Winn, Richard 45, 71 Parker, Laurel 5 Wise, John 45, 48 Pedersen, Stine 31 Zarella, Katrina 15 Pelis, Ryan 10 Pelletier, Andrea 109 Pena, John 103 Peppin, Paul 17 Petell, Lydia 15 Peters, Kathryn 81 Peterson, Janis 39 Philipson, Louis 88 Preston, Robert 65 Renfro, J. Larry 10, 41 Rhodes, Christopher 88 Richards, Jana 1, 3 Rigor, Robert 31 Riordan, John 77 Roeser, Jennifer 7, 8, 9 Roling, John 40 Runnegar, Maria 73 Russell, Franz 35 Sasse, Philipp 84 Schuetz, Hendrike 56 Seward, David 73, 75 Sighinolfi, Christopher 1, 3 SPECIES

Amblyraja radiata (thorny skate) 62 Amphiuma means (two-toed amphiuma) 31 Anguilla rostrata (American eel) 26, 96, 97 Artemia franciscana (brine shrimp) 57 Baelaena mysticetus (bowhead whale) 48 Calanus finmarchicus (copepod) 57, 100 Callinectes sapidus (blue crab) 17 Cancer irroratus (rock crab) 52 Carcinus maenas (green shore crab) 12, 52, 54 Colus stimpsoni (Stimpson's whelk) 65 Crepidula fornicata (slipper snail) 65 Delphinapterus leucas (Beluga whale) 48 Fundulus heteroclitus (killifish, mummichog) 7, 8, 9, 15, 20, 23, 35, 36, 39, 45, 60, 77, 99 Leucoraja erinacea (little skate) 18, 25, 33, 37, 45, 48 61, 62, 73, 75 Leucoraja ocellata (winter skate) 62 Malacoraja senta (smooth skate) 62 Mus musculus (mouse) 84, 88 Mytilus edulis (blue mussel) 65 Myoxocephalus octodecimspinosus (longhorned sculpin) 21 Oryzias latipes (medaka) 43, 71 Pseudopleuronectes americanus (winter flounder) 10, 31, 40 Rattus norvegicus (Norway rat) 92 Squalus acanthias (spiny dogfish) 1, 3, 5, 11, 19, 26, 41, 45, 48, 50, 56, 67, 79, 81, 101, 103, 105, 108 109 Temora longicornis (copepod) 57 Xenopus laevis (African claw-toed frog) 60 KEY WORDS

ABC transporter 33 Guanylin 25 Acid-base 20 Gut 10 balance 23 H+ transfers 23 Ammonia 19 Heat shock protein 41 Anion exchange 10, 25 Hepatocyte culture 73, 75 Arsenic 39 Hormonal control 54 ß-Actin expression 75 Intestine 56 Blood brain barrier 36 Imaging 71 Blood CSF barrier 41 Kidney 56 Blood flow 97 Kinematics 100 Bulbus arteriosus 96 Membrane polarity 73 Calcium 71 Metals 41, 48 Calcium sensing receptor 5 Hypercapnia 23 Carbonic anhydase 52 Hyperthermia 41 cCMP 26 Liver function 73 cDNA 20 Mitosis 71 Cell cycle 71 Molecular cloning 11 Cell lines 45 mRNA 21 Cell volume regulation 12 Mrp 1 39 CFTR 67, 77 Mrp 2 33, 35, 36 103 Na+ H+ exhanger 31 Channel 18 NaCl transport 7, 8, 9 Chloride secretion 15, 105, 108 Na+/K+-ATPase 57 Chromium 40, 48 Natriuretic peptides 7 CNP 1, 3, 109 Nephron 101 Colchicine 1 Neuroendocrine 54 Cotransporter 17 NHE 20, 21, 23 Crab 17 NHERF-1 67 Crustaceans 12, 52, 54 NHERF-2 67 Cyclic GMP 35 Nitric oxide 8, 35, 96 Cytochalasin D 109 Nocodazolone 3 Eel 96 Nucleotide signaling 15 Endothelin 8, 96 Opercular skin 7, 8, 9 Epithelium 15 Operculum 77 Escape 100 Osmolyte 18 Evolution 33 Osmoregulation 12, 52, 54 Excretion 19 99 Expression studies 103 Perfusion 97 Force 100 P-glycoprotein 36 Forskolin 105 PHI 108 Gene expression 17, 40 Plasma membranes 57 Gill 17, 20, 21 Primary cell culture model 75 97 Procaine 105 Glucose transporter 56 Prostaglandins 9 Purinoceptor 15 Quantitative PCR 21 Racemase 65 Receptor 9 Rectal gland 5 Red cell 25 Regulatory volume increase 31 Renal proximal tubule 35 Reproduction 39 Salinity 12, 99 SGLT family 56 Shark intestine 25 Skate hepatocytes 37 Shark rectal gland 105, 108 109 Spermine 5 Sodium-hydrogen exchanger 10 Staurosporine 109 Subtractive hybridization 40 Sulfate 10 Telomerase 48 Taurine 18 Tide 99 Trafficking 77 Underwater camera 99 Uroguanylin 26 Videomicroscopy 97 VIP 1, 3, 10, 105 VIP receptor 103 Volume regulation 18, 25 Xenobiotic transport 37 Xenopus oocytes 103