7v“ : .‘fivam'mm‘v.§1‘(mfi,a‘*;‘v {{‘(d‘QfJ-gmo ‘1‘.-~-Q-vfi-ow.._.m’ ‘ ’-«1-.‘.._"-.~‘_- ‘o--.‘ —.-.--—--~- . A“. 3N VITRO STUDY OF ERYTHROPOIESiS EN: RAINBOW TROUT SALMO GAIRDNERI) usme 59Fe01 Thesis for the Degree of M. 8. HIGH! - TATE UNIVERSITY E CHAR!) LAURENCE WALKER 1972' fi‘wfi w—V ‘w‘vv‘r‘v . ., . . ' —.\ . r ‘A. r ’ 0v. "’ r .. -/I - . _ . 0' ' -.vo ' "'o , — , ,.- ’ ‘- - r I. . ' .o , . ..r , ‘ ' - ' ‘ o a-. , "'-9- t -- .. , _ ’ “' "9 '-c- I. .. . _ ‘ _ 7.. . . - "' .g.-,_. , , l - 'r-,.,,‘_,.’. . . ..q 0 cf.-. fl- ' _. r...-— "o . .-l-...,, _ .‘.".. - vv 0 ..v ..{.¢.o-":'., 7 " ’ ' r - .- "' ""' ovio-tr " -vpaoa pvoorrvv..'.".‘. Vt Wet; o'.'_’_.'."‘ "" .I-..'O"“'~.' '1' h ‘ '0- ' ' "“ - a. ' ° -'~"-"’.'_'!: 11"“..‘ L. 1111111111111111111111111111111111111111111111111 11111111111 1:13:32 3 1293 10402 University ABSTRACT IN VITRO STUDY OF ERYTHROPOIESIS IN RAINBOW TROUT 51§SALMO GAIRDNERI) USING FeCl3 By Richard Laurence Walker Trout red cells were incubated in vitro withngeCl3 for varying lengths of time. The cells were washed three times with phosphate buffered saline (PBS) and the amount of activity remaining was expressed as pgnge/g hemoglobin (Hb). Red cells from normal trout were found to incorporate O. 005 pg/g Hb/48 hr via two rate constants representing attachment of radioiron to the red cell mem- brane and incorporation into intracellular storage areas or heme. Erythrocytes incorporated 1. 5 times more radioiron when suspended in plasma than when suspended in PBS. Trout red cells incorporated 6 to 7 times as much iron as dog red cells for the same time period. Erythrocytes from trout bled 20 to 30% of their whole blood volume incorporated (over a 48 -hour period) 30 times as much activity as control trout red cells. The increase in iron incorporation was attributed to the increase in reticulocytes in the peripheral Richard Laurence Walker circulation. Reticulocytes stained with New Methylene Blue N were apparent as early as 5 days following hemorrhage of 20 to 30% whole blood volume, and counts remained high at day 20 as shown by the significant increase in radioiron incorporation from 0.080 Pgnge/g Hb on day 1 to 0.350 Pgnge/g Hb on day 20. IN VITRO STUDY OF ERYTHROPOIESIS IN RAINBOW TROUT (SALMO GAIRDNERI) USING 59FeC13 By Richard Laurence Walker A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Physiology 1972 DEDICATION I would like to dedicate this thesis to Dr. Ronald O. Kapp and Dr. Arlan Edgar, Department of Biology at Alma College, whose interest and encouragement made my graduate education possible. ii ACKNOWLEDGMENTS The author wishes to acknowledge Dr. Paul Fromm and Dr. Jack Hoffert for assistance in preparing the manuscript and for advice concerning experimental design. Thanks are also extended to Dr. Howard E. Johnson for reviewing the manuscript and Dr. Lester F. Wolterink for help in analyzing the data. Recognition is given to Kenneth R. Olson, David L. Norton and other graduate students who offered valuable advice during the course of the project, and to Esther Brenke for her help. This research was supported in part by the Michigan Agricultural Experiment Station, Project 122. iii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODU C TION LITERATURE REVIEW . Origin of Erythrocytes Theories of Human Red Cell Production Theories of the Red Cell Line in Teleosts Red Cell Maturation Erythropoietic Stimulation Hypoxia . Hemorrhage . Starvation . . Erythropoietic Response Erythropoietic Stimulating Factor (ESF) Species Specificity of ESF. . Role of the Spleen in Erythropoiesis Transferrin Heme Synthesis Iron Storage in Red Cell Iron Pool . . Effect of Multivalent Ions on Erythropoiesis . Lead Cobalt RESEARCH RATIONALE MATERIALS AND ME THODS Experimental Animals Bleeding Technique . iv Page vi . vii 10 10 10 14 15 16 16 19 22 23 29 3O 31 32 32 33 35 37 37 38 Hematocrit, Hemoglobin and Red Cell Counts Radioiron and Isotope Technique . Dialysis . . Uptake Studies . . Radioiron Uptake from Plasma and PBS Radioiron Uptake by Dog and Trout RBC' 8 Uptake Equilibrium Curve . . Reticulocyte Study RESULTS Dialysis . Uptake Study. . Radioiron Uptake from Plasma and PBS Dog vs. Trout . Reticulocyte Study . . Bled vs. Non -bled Uptake Study DISCUSSION . Dialysis . . Uptake Studies . . ’ Effect of Hemorrhage . Application to Fisheries and Environmental Studies Procedures for Monitoring Immature RBC Numbers in Fish APPENDIX A: SOLUTIONS AND STAINS APPENDIX B: EQUATIONS AND CALCULATIONS . Page 38 40 42 43 44 45 45 45 48 48 50 50 51 52 69 73 73 73 75 79 80 90 91 Table LIST OF TABLES Page Activity in counts / minute (cpm) and % total activity incorporated (% TA) after 48 hours of dialysis of blood from control and bled trout against 59Fe -labeled PBS at 13° C. . . . . . . 49 Results of dialysis of 59Fe -labeled whole blood against non -labe1ed blood after 24 hours of incubation at 13° C. . . . . . . . . . . . . . . . 50 Half time (T%) and y—intercept for components I and II (Appendix B) of 59Fe incorporation by trout RBC' 3 suspended in plasma and PBS overa24hourperiodat13°C. . . . . . . . . . .. 51 Half time and y-intercept for components I and II of trout and dog RBC 59Fe incorporation over a 24 hour period at 13° and 37° C. . . . . . . . 52 Hematocrit, hemoglobin, red cell count, 24 hour Fe incorporation by RBC' 3, and % reticulocytes of blood samples from hemorrhagedtrout................ 55 Half time and yeintercept of components I and II for 59Fe uptake by bled and control troutRBC's...................70 vi Figure 10. LIST OF FIGURES Diagram showing the fate of the hemocytoblast (free stem cell) after Klontz et al. (1963) . Maturation scheme for development of rainbow trout erythrocytes (Klontz et a1. , 1963) . Proposed scheme for formation of erythro- poietin (ESF) Semi -log plot of components I and II for trout and dog RBC Fe incorporation . Hematocrit, hemoglobin and red cell counts at various bleeding periods in rainbow trout . 59Fe uptake by trout RBC' 3 in pg59Fe/g Hb/ 24hours Relation of radioiron incorporation to percent reticulocytes among RBC' 3 in rainbow trout Control RBC sample stained with New Methylene Blue N (1000X magnification) RBC sample stained with New Methylene Blue N from a trout 13 days following the removal of 20% of the blood volume . Sample of RBC' 3 stained with New Methylene Blue N from trout 20 days after. removal of 20% of their blood volume (1000X magnification) . vii Page 18 54 57 60 62 64 64 66 Figure 11. 12. 13. 14. 15. 16. BBC sample stained with New Methylene Blue N from trout 20 days after‘the removal of 20% of the blood volume An example of trout mature erythrocytes with elongate cell form, dark —staining nucleus and agranular cytoplasm (2500X magnification) RBC sample stained with New Methylene Blue N from a trout 20 days after removal of 20% of the blood volume . Semi -log plot of components I and II of 59Fe uptake rates for control and bled trout Example of an 59Fe uptake equilibrium curve Example of calculation and semi -log plot of components I and II of a two -component uptake curve . viii Page 66 68 68 72 94 96 INTRODUC TION The production of red cells by some organ or tissue and their appearance in the peripheral circulation is termed erythro- poiesis. Hemorrhage and hypoxia are two stresses which can initiate erythropoiesis in homeotherms. Immature red blood cells increase in number in the peripheral circulation in response to a lowering of blood PO and/or a decrease in blood volume. 2 The physiology of the red blood cell of aquatic species such as rainbow trout may be influenced by the presence of pollu- tants in the water such as insecticides, herbicides, polychlorinated biphenyls, and the heavy metals mercury, chromium, lead, cobalt, etc. Lead and cobalt are known to affect the oxygen -carrying capacity of the red cell. Jandl et a1. (1959) state that lead inter- rupts the biosynthesis of heme within the immature red cell, while cobalt (Co+2 or Co+3) in the ionic state is more tightly bound to immature red cells than iron. Also trivalent Co is not displaced from transferrin by Fe+3, thus disrupting the transport of iron to the red cell. Mercury and other heavy metal ions may have a similar effect on the oxygen transport mechanism, thus stimulating the production and release of immature red cells in the peripheral circulation. Therefore toxicants as they are present in aquatic environments may affect the erythropoietic activity of aquatic animals. If this is actually the case it should be possible to measure changes in the erythropoietic activity by measuring the number of radiolabeled immature red cell forms, such as reticulocytes or polychromatocytes, present in the peripheral circulation. Radio- iron (59Fe) incorporation by immature red cells may be a useful parameter or monitor in the detection of the effects of pollutants on red cell production in aquatic animals. This precipitated the study of red cell production rate and the feasibility of measuring erythropoiesis in trout with the use of 59FeCl3. LITERATURE REVIEW Much of the literature on red blood cells pertains to mammalian species. Some authors have attempted to study teleost red cell production and maturation, but most find it difficult due to the inability to distinguish many of the stages in teleost red cell maturation. The information reviewed here pertains to either mammalian or lower vertebrate forms. Or_igin of Erythrocytes Theories of Human Red Cell Production The final or myeloid period of hematopoiesis begins during the fifth month of human fetus development (Wintrobe, 1961), and it is initiated with the establishment of placental circulation. At first the liver is the chief organ of erythropoiesis, while leukocyte development occurs in the bone marrow. Soon the bone marrow becomes hematopoietically active and the function of the liver as the site of red cell production disappears. Myeloid cells in bone marrow are involved in hematopoiesis, the production of blood cells of all types. Growth of myeloid cells leads to reduction of the mesenchymal cells of the liver and other tissues to a very scant reticular stroma which remains throughout life with its erythropoietic potentialities intact (Wintrobe, 1961). Multipotential cells of the mesenchyme are found in bone marrow, the intralobular capillaries of the liver, the venous sinuses of the spleen and lymph nodes, in the walls of the intestine, in serous membranes, and possibly in the adrenals. Together they form the reticulo -endothelial system. Wintrobe (1961) has reviewed the two major theories of blood cell production, the monophyletic and polyphyletic theories, and cites several references which deal with the subject of blood cell origin from myeloid cells and the reticulo —endothelial system. The classical monophyletic theory proposes that lymphocytes of lymphatic tissue are identical with primitive blood cells and are totipotential. With the proper stimulus erythrocytes, leukocytes, granulocytes and monocytes are said to all arise from lymphocytes. The polyphyletic theory essentially states that formation of erythrocytes and various forms of leukocytes arise from specific blast cells instead of from a common progenator. There are a number of corollaries to this theory which deal with specific origins of granulocytes, agranulocytes and erythrocytes. Maximow (1924) proposed a common stem cell which differs in size and structure from lymphocytes and is capable of producing both red and white blood cells. Doan et a1. (1932) proposed two independent lines for blood cell development: erythrocytes from endothelial cells of the bone marrow and white cells from reticular cells. Jordan and Speidel (1924) studied stem cells in elasmobranchs and teleosts and found them to be indistinguishable morphologically from. lymphocytes and capable of developing into red or white blood cells. Theories of the Red Cell Line in Teleosts Most of the theories of teleost red cell origin can be clas - sified as monophyletic in nature. A stem cell or hemoblast is proposed as the common ancestor of both red and white cells. Yoffey' s (192 9) data supports the theory of Jordan and Speidel (1924); and he states that the lymphoid tissue of fishes, composed of small, round cells with large nuclei and little or no cytoplasm, gives rise to all cellular components of blood. Duthie (1939) found the main center of blood cell formation to be in the kidney where erythro- cytes and leukocytes develop from intertubular tissue; however, he states that the lymphoid centers in the submucous layer of the intestine and mesentery are mainly for leukocyte formation. Jakowska (1956) stated that a large hemoblast was the first identifiable form in hematopoietic tissue or in the circulating blood of teleosts. She proposed that this large hemoblast is a stem cell which can be either leukogenic or erythrogenic and give rise to all types of blood cells in the teleost. Klontz et al. (1963), in an unpublished manuscript, also support the totipotential free stem cell theory. They state that there is no justification in making a distinction between a reticulum cell and an endothelial cell. In theirscheme the hemocytoblast, which is derived from a reticulo -endothelial cell, gives rise to all blood cells in rainbow trout (Figures 1 and 2). Large and small lymphoid hemoblasts arise from the hemocytoblast and are pluri- potential. The large lymphoid hemoblast produces granulocytes, especially neutrophils and macrophages, while the small lymphoid hemoblast forms erythrocytes, thrombocytes and lymphocytes (Figure 1). The proerythroblast is known to undergo mitosis and is the direct descendant of the small lymphoid hemoblast (Figure 2). Following the proerythroblast are the erythroblast, in which mitosis may occur, the early polychromatocyte, with no noticeable mitosis, and the mid -polychromatocyte, where chromatin is compacted into thick, radially arranged strands. The late polychromatocyte is the youngest cell in the series to appear in the circulatory system Hemocytoblast (free stem cell) / \ Small Lymphoid Large Lymphoid hemoblast hemoblast \ / Granuloc te Lymphocyte flrocyte y Thrombocyte ia circulating macrophage Figure 1. --Diagram showing the fate of the hemocytoblast (free stem cell) after Klontz et al. (1963). Erythrocytic series Hemocytoblast (free stem cell) 1 Small lymphoid hemoblast Proerythroblast 1 Erythroblast Early - Polychromatocyte Mid - Polychromatocyte Late - Polychromatocyte Reticulocyte Erythrocyte Figure 2. --Maturation scheme for development of rainbow trout erythrocytes (Klontz et al. , 1963). in small numbers. No nuclear membrane is visible at this stage and mitosis does not appear to occur. The reticulocyte is next in order of maturation and has an oval or elongate, compacted nucleus. The ribosomal RNA inside the cytoplasm stains dark blue with New Methylene Blue N. This characteristic is important in dis- tinguishing reticulocytes from mature erythrocytes when reticulocyte counts are made. A mature erythrocyte is characterized as having an elongate, compact nucleus, and mitotic division is not apparent. However, fish on certain diets have a high incidence of amitosis, the significance of which is undetermined. In mitosis (or direct cell division) there is first a simple cleavage of the nucleus without change in the structure (such as the formation of chromosomes) followed by the division of the cytoplasm. Recently a form of amitotic division of erythrocytes has been discovered in eight species of avians and two poikilothermic species (Engelbert and Young, 1970 a and b). Deutsch and Engelbert (1970) describe this type of red cell formation as sac -like protubera- tions or nuclear blebs which enlarge and then differentiate into red blood cells. Chromatin is shed into the extensions of the nuclear membrane which eventually come free of the cytoplasm of the mother cell and form sacs. These sacs are attached to the mother cell by fine threads which eventually break, setting the sacs free 10 as clone cells. The clone cell first appears as a naked nuclear mass which later develops cytoplasm, RNA and hemoglobin. Red Cell Maturation Yuki (1957) proposed that fluctuations of the blood cell types are usable as a biological indicator in studies of fish physiology. He noted, however, that studies based on this idea never advanced because, as Jakowska (1956) and Klontz et al. (1963) found, differen- tial identification of blood cells in fish is very difficult due, in part, to the fact that even in normal fish undifferentiated blood cells are found in the peripheral circulation. Yuki reports that such a state resembles the beginning of the third stage of hemopoiesis in the fetus of higher vertebrates. Deutsch and Engelbert (1970) mention Jordon (1938) and Andrew' s (1965) observations that red blood cell prolifera- tion and maturation take place largely in circulating blood of teleosts. Despite the common appearance of immature red cell forms in the peripheral circulation, Altland and Brace (1962) estimated the mean erythrocyte life span in the turtle to be between 600 and 800 days. Erythropoietic Stimuli Hypoxia According to Rosse and Waldmann (1966) the erythropoietic response to environmental hypoxia appears only among homeotherms. 11 Erythropoiesis in birds and mammals may be due to the fact that other physiological adjustments, such as increased respiratory exchange, increased circulatory rate, decreased body temperature, or shift to anaerobic metabolism are not enough to compensate for the increased oxygen deficit caused by hypoxia. Production of the erythropoietic stimulating factor, and therefore erythropoiesis, is controlled by the balance between tissue oxygen supply and the requirements of the body for oxygen. Jandl et al. (1959) state that maturation of the red cell and incorporation of iron into heme in vitro are somewhat inhibited, rather than stimulated by anoxia. Iron uptake by reticulocytes is least under anaerobic conditions and increases slightly with increas- ing oxygen tensions. There is no evidence of a direct anoxic stimulus to red cell metabolism in terms of the ability of immature red cells to take up iron, to incorporate iron into heme and to mature morpho— logically. In other words, low ambient PO 2 in reticulocytosis but the uptake of iron by these immature forms may cause an increase does not result in a concomitant increase in the hemoglobin concen- tration. Fried, Johnson, and Heller (1970) report an increase in the rate of erythropoiesis and an increase in the hematocrit of rodents after exposure to hypoxia. Erythropoietin levels reach a peak after 12 8 hours of exposure to hypoxia but then fall to barely detectable levels after 72 hours. They hypothesize a feedback inhibition where low blood PO triggersthe release of erythropoietin which 2 in turn results in elevated red cell production and increased hematocrit (Hct). The increase in reticulocytes and Hct then may cause the noticeable drop in blood erythropoietic stimulating factor (ESF) level. Therefore elevation of blood ESF titers is the initial response to hypoxia. Rosse and Waldman (1966) state that poikilotherms, such as frogs and turtles, do not increase their rate of erythropoiesis when exposed to reduced ambient PO . Altland and Parker (1955) 2 noted this to be true in box turtles exposed to low P02 for 53 days at 25, 000 ft. altitude where they reported no (significant difference in Hct or Hb values, erythrocyte or leukocyte concentration of control or exposed turtles. There was no evidence of erythroblastosis in spleen smears. This confirmed the findings of Sokolov (in Altland and Parker, 1955) that fish, frogs, toads and turtles show no increase in erythrocytes when exposed to a barometic pressure of 200 mm Hg (P = 42 mm Hg) for 7 days. Bonnet(in Altland and Parker, 1955) O 2 also found that exposure of fish and frogs to a pressure of 10 to 20 mm Hg for 1 to 2 hours daily for 6 days failed to stimulate erythropoiesis. Evidence in favor of hypoxia stimulating erythro- poiesis in poikilotherms is reported by Gordon (1935) who found 13 exposure of the salamander (Necturus maculosus) to 330 mm Hg pressure for 7 weeks resulted in an increase in numbers of immature red cells and hematocytoblasts in the peripheral circula - tion. Rosse, Waldmann, and Hull (1963) found that in frogs, sublethal levels of hypoxia do not stimulate erythropoiesis as measured by the incorporation of thymidine-2 - 14C into peripheral erythrocytes. No reticulocytosis is seen in response to acute or chronic anoxia in the turtle, frog or lizard and it appears that environmental hypoxia does not grossly alter the oxidative metabolic rate of the frog. However, chronic hypoxia may induce polycythemia in newts and fish. Prosser et al. (1957) noted that acclimation of the goldfish to low PO lowered the oxygen consumption from an average of 2 0.239 to 0. 163 ml 02/g/hr and shifted the critical PO from approxi- 2 mately 3. 1 to 1. 5 ml 02/ liter. This shift may have resulted from an increase in Hb concentration in the blood. The Hb values of control goldfish averaged 9. 8 g/100 ml, compared to 11. 9 for fish exposed to 2. 0 ml 02/ liter. The statistical significance of these two values was not high (0. 1 E p E 0. 2) but they concluded that acclimation of the goldfish to low PO results in an increase in 2 circulating Hb. They also noted a 30% increase in the normal RBC count after acclimation of fish to O. 92 ml 02/liter. 14 Hemorrhage Rosse et al. (1963) noticed that even though sublethal hypoxia did not stimulate erythropoiesis, hemorrhage of 1/ 3 of the blood volume of frogs resulted ina 70-fold increase in erythropoiesis as measured by incorporation of thymidine -2 -14C into erythrocytes in the peripheral circulation. They concluded that either the erythropoiesis -stimulating response of bleeding resulted from some effect other than tissue hypoxia, or that hypoxia induced by anemia is somehow different from that induced by changes in the environ- mental PO . Somehow, the poikilothermic animal makes adjustments 2 to environmental hypoxia that are not made to anemia. This is contrary to what is believed to be the physiological response of mammals to hypoxia and hemorrhage. Altland and Thompson (.1958) observed that a marked reticulocytosis was induced in turtles by repeated bleedings. High reticulocyte counts persisted for 3 weeks following hemorrhage, and 4 months were required for restoration of normal erythrocyte values. Following hemorrhage of turtles, Hirschfeld and Gordon (1965) noted decreases in RBC and plasma volumes, Hct, Hb and serum iron concentration. An indication of the cellular response to bleeding was the appearance of radioiron -labeled erythroblasts in the circulation 2 days after 59Fe administration, signifying 15 erythropoiesis, and the release of immature RBC' 8 into the peripheral circulation. Zanjani et al. (1969) reported that after bleeding, RBC regeneration rapidly occurred in gourami, a tropical fish. By as early as the fifth day following hemorrhage in both fed and starved fish, considerably greater values were obtained for radioiron incorporation into circulating erythrocytes, as compared to non— bled fish. Starvation Lack of food causes a significant decrease in the ability of poikilothermic vertebrates to respond to hypoxia or hemorrhage. Zanjani et al. (1969) in their study of humoral factors influencing erythropoiesis in the blue gourami, found that although total blood volume was not altered significantly, RBC counts, Hb, Ht and percentage incorporation of radioiron into circulating RBC' 3 were all lowered due to the lack of food. Full replacement of the diet completely reversed the inhibitory effects of food deprivation on erythropoiesis in the gourami. Hirschfeld and Gordon (1965) found that starved —bled turtles were incapable of increased erythropoiesis as measured by 59Fe incorporation by reticulocytes. Erythroblast levels were not elevated in those turtles subjected to starvation and bleeding. l6 Starvation of turtles for 6 weeks prior to hemorrhage prevented increases in benzidine -staining (used to detect Hb) and also radioiron incorporation by erythroblasts, both of which increase in fed -bled turtles. Starvationled to a decrease in serum iron concentration, and these researchers warn that the inability of some workers to detect changes in erythropoiesis in lower vertebrates following application of stimuli effective in mammals may be due to an inadequate nutritional state. Erythropoietic Response Erythropoietic Stimulating Factor (ESF) Summarized below is the excellent review of erythropoiesis and the origin and function of ESF by Gordon (1959). Erythropoietin, "Carnot serum, " hematopoietine, plasma erythropoietic stimulating factor, plasma erythropoietic factor (EPF) and erythropoietic stimulant are all terms referring to ESF, which has been classified as a hormone that is specifically responsible for increased RBC . production. ESF is a glycoprotein (M.W. 60, 000) formed by the combi- nation of the Renal Erythropoietic Factor (REF) with a globular protein produced by the liver. REF is produced exclusively by the kidney in dogs; but in rats, rabbits and humans ESF is found in the 17 blood even after removal of the kidneys. REF is secreted by cells in the glomeruli in response to hypoxia and at the same time the liver increases its production of the specific globulin which reacts with REF (Figure 3). Combination of the two occurs in the plasma to form ESF which in turn increases the production of proerythro— blasts by the erythropoietin-sensitive stem cells. Zanjani et a1. (1967) in a study of the role of ionic and vasoactive agents on erythropoietin formation detected the existence of an erythropoietic inhibitor. Exogenous ESF was inactivated when mixed with kidney homogenates, and it was speculated that REF con- taining extracts may have an enzyme which in the presence of a serum -borne metal ion, destroys newly formed ESF under in vitro conditions. ESF levels may be restrained due to the presence of an ESF -inhibitor which destroys circulating erythropoietin and main- tains a normal plasma ESF concentration. Angiotensinase, the enzyme which destroys angiotensin 11, requires Ca+2 for its effective- ness and was thought to be the ESF -inactivating factor in the REF- containing extracts. However, the presence of Ca+2 and angioten- sinase failed to activate destruction of ESF and actually stimulated ESF production in vitro. Zanjani et al. (1967) concluded that REF required Ca+2 at a lower concentration for ESF production than does the erythropoietin inactivator. 18 Kidney Liver é— Hypoxia ___9 V V REF Globulin Erythropoietin Erythropoietin - sensitive stem > Proerythroblasts cells Figure 3. --Proposed scheme for formation of erythropoietin (Ganong, 1 96 9). 19 Fried et al. (1970) noted a rise in rat plasma erythropoietin levels to a peak after 8 hours of hypoxia. ESF levels were barely detectable at 72 hours and upon injection of erythropoietin no decrease in the rate of ESF clearance‘was noticed. Therefore it is likely that production of ESF decreased even though the erythro- poietic response to hypoxia was not concluded. ESF initiates the erythropoietic response, but apparently the continued high plasma ESF titers are not necessary and may be inactivated after delivering the initial stimulus to the stem cells. Species Specificity of ESF Several authors have attempted to stimulate erythropoiesis by injecting one species with anemic plasma or ESF —containing plasma of another species. Rosse et al. (1963) report that the serum of bled frogs increases the incorporation of thymidine -2 -14C into erythrocytes of frogs, but anemic frog serum does not stimulate erythropoiesis in the polycythemic mouse. Serum taken from anemic humans containing large amounts of erythropoietin does stimulate RBC production in the polycythemic mouse, but does not do so in the frog. Rosse concludes that stimulating factors in frogs and mammals are chemically different. Erythropoietin (ESF) from mammalian serum requires a protein structure and sialic acid for biological activity. Rosse and 20 Waldman (1966) discovered that bird ESF has a similar protein structure but lacks sialic acid since it is not inactivated with sialidase. This is why hypoxic quail serum and anemic chicken serum stimu- late erythropoiesis in quail but fail to do so in mice, while anemic human serum which increases erythropoiesis in mice is without effect in polycythemic quail. Erythropoiesis in birds and mammals is induced by the same stimuli, but is mediated by different protein humoral factors. Goldberg et al. (1956), while studying the biosynthesis of heme by avian erythrocytes in vitro, discovered that the erythro- poietic factor which might exist in anemic chicken serum does not act directly on heme synthesis. Two chickens were bled daily for one week, while two others were injected with phenylhydrazine, resulting in severe anemia and increased reticulocytosis. The plasma from the anemic chickens was added to a hemolysate solution prepared from lyzed red cells which had been centrifuged and the stroma removed. Plasma was also taken from two normal chickens and added to the hemolysate as a control. Goldberg found that the uptake of radioiron into the heme portion of the red cell was no greater in the presence of extracts of anemic bird plasma than in the presence of extracts from normal birds. 21 Zanjani et al. (1969) report erythropoiesis in gourami after administration of anemic fish plasma. The addition of 0. 15 ml of plasma obtained from anemic gourami to fish with suppressed erythropoiesis, due to starvation, resulted in a highly significant increase in the rate of 59Fe incorporation into the immature red cells. Plasma obtained from normal fish had minimal erythropoietic activity. Administration of mammalian ESF to starved fish also caused a significant increase in radioiron incorporation into circu- lating red blood cells, but the erythropoietic effects did not occur until large doses of ESF were used. The first clear response was obtained only when 0. 80 International Reference Preparation (IRP) units of ESF were administered (16 units/100 g body wt. ). Krzymowska et al. (1960) showed that plasma concentrates from anemic sheep stimulated erythropoiesis in carp. In this regard, the reported failure of mammalian ESF to stimulate erythropoiesis in birds and frogs may have been due to the use of insufficient amounts of this factor. When hypoxia -induced polycythemic mice were injected with anemic gourami plasma no erythropoietic response occurred (Zanjani et al. , 1969). The fish plasma was "toxic" to mice even when very dilute amounts were injected. An in vitro study was conducted by Zanjani et al. (1969) to determine if anemic fish plasma and mammalian ESF stimulated 22 erythropoiesis in the gourami by direct effect on the circulating erythrocytes. While gourami reticulocytes incorporated radioiron in vitro, no stimulatory effect on this process occurred upon the addition of anemic gourami plasma or mammalian ESF to the RBC samples. In fact, red cells from starved gourami accumulated radioiron in vitro to a smaller degree than cells obtained from normally fed fish. Role of the Spleen in Erythropoiesis Yoffey (1929) describes the spleen of fish as a capsule consisting of a single layer of epithelium with a thin stratum of fibrous tissue underneath, and pulp consisting of spongy cellular reticulum through which blood percolates. The arteries of the spleen of most teleosts terminate by dividing into three or four short, thick -walled capillaries which open directly in the spaces of the pulp reticulum. In his review, Gordon (1959) points out that splenectomy prevents the appearance of increased reticulocytosis in animals subjected to hemorrhage, and speculates that the spleen may be one of the sites of ESF production. Meints (1969) reported an increase in spleen size and a shift of red cell maturation from the spleen to the peripheral circulation in frogs injected with phenylhydrazine, an erythropoietic agent. The spleen of fish may 23 be a storage and production site for immature red cells and leukocytes (Duthie, 1939). In cases of severe hemolytic anemia the spleen maybe forced to release more erythrocytes and play a larger role in manufacture of red cells. The bone marrow in mice normally has the greatest responsibility for red cell production, the spleen having only a minor role. Hemorrhage and hypoxia produce a marked increase in splenic radioiron incorporation while the increase in bone marrow radioiron uptake is not as great (Bozzini et al. , 1970). Splenectomy in a normal adult mouse pro- duces a mild anemia and total circulating red cell volume is decreased by about 15%. Bozzini concludes that red cell production in bone marrow is normally not sufficient in the mouse to meet the demand for erythrocytes and requires the assistance of extra- medullary erythropoiesis, the spleen being the secondary site of erythropoiesis along with liver andlymph tissue. Transferrin When immature red cells which lack iron are present in increased numbers in the plasma, the condition is referred to as reticulocytosis. Iron may be obtained by simple equilibrium between ferrous and ferric ions in the extracellular fluid and the intracellular fluid of the red cell or from an iron-binding protein 24 molecule (transferrin) present in the plasma which acts as a carrier of iron atoms. Jandl and Katz (1963) state that immature red cells acquire most of their iron for Hb formation by the second method. Radioactive iron has been shown to be taken up directly from the plasma or from solutions of purified transferrin by reticulocytes both in vitro and in vivo. After injection of transferrin -bound radioiron, radioactivity is present in bone marrow within a few minutes. Transferrin is a B1 globular protein (M.W. 86, 000) with two separate iron binding sites, each capable of binding one atom of ferric iron (Jandl et al. , 1959; and Fletcher and Huehns, 1968). The half -life of transferrin in humans is approximately 12 days. There are more transferrin iron -binding sites than atoms of iron in normal plasma, thus no free iron is present. This is not unex- pected since even small quantities of free iron are toxic. Transferrin iron is very tightly bound to the protein molecule and only very concentrated EDTA solutions will remove this iron. Binding proteins have been found in lamprey and fish blood, but it is not known whether they have one or two iron-binding sites per molecule (Fletcher and Huehns, 1968). Iron is transferred directly from the iron-binding protein to specific iron -binding receptors on the cell surface of reticulocytes 25 (Jandl et al. , 1959). Iron transfer is an energy -dependent process (Wintrobe, 1961); however, it is improbable that transfer is directly dependent upon oxidative mechanisms since uptake of 59Fe is only moderately diminished by anoxia in vitro. Uptake is stimulated by the presence of glucose and inhibited by agents which block inter- mediary glucose metabolism (Jandl et al- , 1959). Anaerobic glycolysis appears to be the main source of the energy required from. the transfer process. The specific membrane receptor sites are present only in developing red cells and after hemoglobin synthesis is complete, they are lost (Fletcher and Huehns, 1968). Not only is there an iron carrier protein in the plasma, but also one inside the immature red cell. The ease with which red cell pre- cursors release iron from transferrin, compared with the difficulty of achieving this in vitro with the use of EDTA at physiological pH, suggests that the attachment of transferrin to the receptor site produces a change in conformation of a specific protein inside the cell (Fletcher and Huehns, 1968). To summarize the transfer process, ferric iron is bound to transferrin in the plasma and is transported to specific receptor sites on the reticulocyte membrane where it is actively transported to the inside of the cell by a process involving a change in the structure of a specific iron—carrying pro- tein inside the cell. 26 Upon completion of the transfer process the transferrin molecule, minus one atom of iron, detaches from the reticulocyte membrane receptor site. A new transferrin molecule, with both iron binding sites saturated, then attaches to the vacant receptor site and the transfer process is repeated. Jandl and Katz (1963) report a specific dynamic equilibrium between cell attached trans - ferrin and free transferrin for attachment to the immature red cell. The rate of transferrin detachment is much the same as the rate of attachment, indicating that free transferrin exchanges quite rapidly with cell -attached transferrin, the half-time of exchange being approximately one minute. To illustrate the importance of transferrin in formation of hemoglobin, comparison was made between incorporation of 59Fe by immature and mature RBC' 3 from a saline solution and from plasma containing transferrin. Jensen et al. (1953) found that uptake of iron by duck erythrocytes suspended in 0. 15 M NaCl was approxi- mately 35% less than that of cells suspended in plasma. The decrease in iron incorporation was due to the reduction in heme -bound iron, the nonheme 59Fe fraction being the same for both media. Jandl et al. (1959) and Walsh et al. (1949) reported that both mature and immature washed red cells took up iron from a saline solution at a greater rate than from plasma containing transferrin. 27 Approximately 90% of the 59Fe adsorbed from saline remains in the stroma or membrane fraction upon hemolysis and only 10% is in the heme portion. However, when placed in plasma containing the iron -binding protein only the immature forms were able to incorporate the iron. Upon hemolysis of these cells approximately 40 to 90% of the 59Fe was recovered in the heme fraction and 10 to 60% was associated with the reticulocyte mem- branes. Addition of a metallic ion chelator, EDTA, to a cell suspension of iron -binding protein and washed red cells did not affect the transfer of iron from the iron -binding protein to reticulo- cytes until the EDTA/Fe ratio exceeded 10. Free iron in saline was completely blocked from entering reticulocytes and mature red cells alike with the addition of EDTA. Jensen et al. (1953) compared the uptake of radioiron by human and duck erythrocytes in vitro. They discovered that cells from normal human subjects take up negligible quantities of iron, whereas in blood from a patient with hemolytic anemia, containing 33% reticulocytes, appreciable quantities of iron are taken up. Finch ( 1957) states that all immature red cells through reticulocytes are able to take up iron from plasma while mature erythrocytes do not. As determined by autoradiographs, human pronormoblasts and basophilic normoblasts have the greatest ability to assimilate 28 iron. Jandl et al. (1959), in their study. on ironitransfer from serum. iron -binding protein to human reticulocytes, found that the incorporation of 59Fe increases with increasing reticulocyte numbers and confirmed Finch' 3 statement that mature red cells do not acquire iron from transferrin. According to Jandl andKatz (1963), reticulocytes have 100 times the affinity for transferrin that mature RBC' 5 have. Walsh et al. (1949) added support to this hypothesis in their study on heme synthesis in vitro by immature erythrocytes. They state-while blood containing less than 1% reticulocytes took up no measurable quantity of radioiron, uptake was demonstrated repeately in blood with a high-reticulocyte content. That radio- activity-was localized in immature cells was shown by correlation of reticulocyte count and radioactivity in various fractions of this blood separated by the albumin floatation technique. Jandl et al. (1959) found that not only does 59Fe incorpora- tion increase with increasing reticulocyte numbers but incorporation is also related to the amount of iron -binding protein present in the plasma. Increasing amounts of radio -labeled iron -binding protein (IBP) (33% saturated with iron) were added to a series of incubation vials containing a constant percentage of reticulocytes. Radioiron incorporation increased in proportion to the iron -binding protein concentration until a concentration of 160 11M was exceeded; thereafter, 29 further increase in 59Fe IBP resulted. in no further incorporation by the reticulocytes. Percentage saturation of the iron-binding sites on transferrin is another important factor in iron incorporation by reticulocytes. Iron transfer from iron -binding protein to immature erythrocytes in vitro is reported to be decreased when. the percentage saturation of transferrin drops below 30% (Finch, 1957). Jandl et al. (1959) report when transferrin is less than 20% saturated, iron uptake of reticulocytes diminishes despite an amount of free iron in the system well in excess of that utilizable by reticulocytes. Only when the transferrin iron -binding capacity was exceeded did free iron non- specifically attach to red cells. Jandl and Katz (1963) report that reticulocytes have a greater affinity for iron-saturated transferrin than to the iron -unsaturated form. Fletcher and Huehns (1968) state that iron is taken up by reticulocytes from molecules contain- ing two atoms of iron at roughly 10 times the rate that it is taken up from molecules carrying only one atom. They speculate that this is a result of a difference in shape or charge of the transferrin molecules. Heme Synthesis After transfer of iron to the carrier protein inside the reticulocyte, the iron accumulates in the particulate matter of the 30 cell, the stroma, mitochondria and microsomes (Wintrobe, 1961). Upon release from the particulate matter the iron combines with a non -hemoglobin protein. The iron can be incorporated into a hemo- globin molecule as the link between the porphyrin, heme, and the protein molecule, globin, or it can remain permanently as a non- heme fraction within the erythrocyte. Jensen et al. (1953) found that approximately 85% of the iron entering avian erythrocytes becomes irreversibly incorporated-into heme and only 15% is held within the cell as the non —heme RBC iron pool. Equivalent values have been reported for human RBC' 3 by Jandl et al. (1959). Jensen reports that duck erythrocytes are normally supplied with a reserve of Hb precursor, substances such as glycine and preformed por- phyrins, therefore little additional material other than iron is required for hemoglobin synthesis. Based on the hypothesis of Neuberger (1951), that globin synthesis parallels the rate of por— phyrin synthesis, the assumption is made that all newly formed heme is converted to hemoglobin. Iron Storage in Red Cell Kaplan, Zuelzer and Mouriguard (1954), in developing a stain for nonhemoglobin iron, noticed the presence of iron granules in the cytoplasm of normal red cells. If the presence of this iron is 31 not due to the-degradation of Hb, two possibilities exist: either the iron granules represent an intracellular depot for heme synthesis, or an excess of iron entered the cell and was not utilized in Hb formation. Kaplan explains that the appearance of iron granules in normoblasts prior to maturation, and the evidence of adequate Hb synthesis, might indicate that the first of these hypotheses is true. The fact that mature red cells, produced under the condition of iron deficiency anemia, do not contain stainable iron suggests utilization of iron granules during the maturation process. Kaplan concludes that stainable nonhemoglobin iron in the red cell pre- cursors is a normal phenomenon, and might be used in the recogni- tion of immature red cells. Iron Pool Iron from catabolized red cells is not lost by excretion from the body, but, for the most part, is retained in an iron pool. Some authors have reported that iron released from catabolized hemoglobin is re -utilized in preference to other sources of iron for hemoglobin synthesis. However, Sanchez -Medal, Duarte, and Labordini (197 0) found that re -utilization of 59Fe -labeled hemo- globin released from radio -labeled red cells decreased by dilution in an iron pool. They gave daily intraperitoneal injections of 32 59Fe -labeled hemoglobin to two dogs recovering from anemia (due to hemorrhage). Over a period of 14 days only 20. 8% of the 9Fe -hemoglobin was re -utilized, accounting for only 12. 9% of the total iron present in the hemoglobin produced during this period. Thus most (87%) of the iron for hemoglobin synthesis was derived from iron stores. Ross (1946) suggested that once iron is introduced into the body it enters the iron pool and is utilized in a uniform fashion. Sanchez -Medal et al. (1970) concluded that plasma Hb is rapidly catabolized, mainly in the liver, and iron is released and handled in the same manner as iron absorbed through the gut. Effect of Multivalent Ions on Erythropoiesis Lead Goldberg et al. (1956) mention that ionic lead inhibits the biosynthesis of heme by interrupting some intermediate state in heme formation such as the conversion of protoporphyrin to heme. Jandl et al. (1959) state that lead suppresses incorporation of 59Fe by reticulocytes through inhibition of heme synthesis. Incorporation of 59Fe into Hb is almost entirely prevented, but RBC radioactivity is only slightly decreased. This indicates that a very concentrated amount of radioactivity must be found somewhere near the red cell membrane. Jandl found that the increase in stromal 59Fe incorporation 33 increases tremendously whenleadis present within the red cell, and there is no evidence of iron accumulation in the microsomes or mitochondria, the temporary storage-areas for iron before it is incorporated into Hb and nonhemoglobin protein. Cobalt Cobalt is reported to stimulate the erythropoietic response to hypoxia in mammals. Miller andeale (1970) found that daily intraperitoneal injections of cobalt chloride (4 to 8 mg/kg) depresses oxygen consumption in rats and may affect their metabolic response to cold. The stimulating effect of cobalt on the erythropoietic centers is probably caused by the brief periods of tissue hypoxia following each injection. Jandl et al. (1959) mentionthat, like most multi- valent metals, cobalt combines with the membranes of normal washed RBC' 3; however, unlike multivalent iron or chromium, cobalt is taken up more by reticulocytes than by adult cells. Trivalent cobalt (Co-H3) had 1. 47 times the affinity for reticulocytes and l. 38 times the affinity for mature RBC' 3 than divalent cobalt (Co+2), thus the more highly charged cations are more extensively bound by red cell membranes. When iron -binding protein, less than half saturated with 60Co, is placed in a vial with a reticulocyte -rich red cell suspension, 60Co incorporation by reticulocytes is from 5 to 12 times greater than incorporation by mature red cells. Uptake 34 of 60Co from 60Co-transferrin in vitro by reticulocytes is approximately 10% of the uptake of 59Fe from 59Fe -transferrin at the same specific activity, which demonstrates the preference for transferrin —bound iron by the immature red cells. However, once Co+3 is bound to transferrin it is not displaced by Fe+3. Cobalt will remain bound .to transferrin as long as there is a suitable oxidizing agent present to maintain the trivalent state. The erythro- poietic effect of cobalt may be due to the inability of red cells to obtain iron for Hbr synthesis. A form of tissue hypoxia, as proposed by Miller and Hale (1970), may result from the substitution of cobalt for iron on the transferrin molecule. The erythropoietic response to tissue hypoxia is not seen in poikilotherms, and cobalt administra - tion to turtles did not cauSe erythropoiesis (Altland and Thompson, 1958). RESEARCH RATIONALE Since reticulocytes and polychromatocytes are present in the peripheral circulation as forms lacking the essential iron, it should be possible to provide the necessary iron required to complete the synthesis of Hb and convert the reticulocytes into mature erythrocytes. This can be accomplished by adding radio- iron (59FeC13) to the RBC pool. The radioactive tracer will be incorporated into the Hb structure of the reticulocyte as ferrous iron. The rate of radioiron incorporation into the RBC' 3 will be used as an indication of the circulating level of reticulocytes. The greater the number of immature forms in the blood, the greater the amount of 59Fe incorporated. The objectives of this study were: 1. To determine if trout red cells could incorporate 59Fe in vitro. 2. To determine if red cells would retain this label when dialyzed against a substance of equal binding capacity. 35 36 To establish the importance of transferrin or iron -binding protein in the plasma. To measure the kinetics of 59Fe incorporation by red cells. To determine the effect of hemorrhage on red cell production. To determine the practicality of use of red cell 59Fe incorporation as a monitor of the erythropoietic effects of environmental pollution on animal physiology. MATERIALS AND METHODS Experimental Animals Rainbow trout (Salmo gairdneri) weighing 150 -200g were obtained from the Michigan Department of Natural Resources fisheries research station at Grayling, Michigan. They were trans- ported to Michigan State University in an 80 gallon galvanized metal tank lined with non-toxic paint. The tank was contained within an insulated box fitted with a telethermometer and an agitator for aeration. The water was maintained at 13 :1: 2° C during transit. At the university fish were held in 120 gallonifiberglass tanks (Frigid Units, Toledo, Ohio) equipped with aeration stones and flow- through water supply (2 liters/min.) and held at 13 :1: 1° C under 14 hours of light and 10 hours darkness per day. The water was filtered through activated charcoal to remove iron and chlorine. Fish were divided into two groups: one starved and the other fed (100g Feed/10 kg of Fish) commercial trout pellets (3/16 inch, Zeigler Bros. Feed Mills Inc. , Gardners, Penn.) every other day for the duration of the study. Fungal problems, such as 37 38 fin rot, were controlled by cleaning tanks once or twice weekly and treating infected fish with a solution of malachite green. Ble eding Technique Fish were bled from the caudal vein at a point just posterior to the anal fin using a syringe fitted with a 1% inch long 21 gauge needle. The syringe and needle were rinsed with sodium heparin (1, 000 USP units / ml) to prevent clotting. The needle was inserted through the ventral surface into the haemal. arch, and the plunger was pulled back slowly until the desired amount of blood was obtained. This method is a quick and easy way to obtain 2 - 3 ml of blood, with the majority of fish surviving. Hematocrit, Hemoglobin and Red Cell Counts Blood was drawn into heparinized capillary tubes (Biological Research, St. Louis, Mo.) and centrifuged (IEC Model MB Micro- hematocrit Centrifuge, Needham Heights, Mass.) at 11, 500 rpm for 2 minutes. 1 The percentage of packed red cells was determined with an IEC Circular Microcapillary Tube Reader. Hemoglobin determinations were made spectrophotometri- cally by the cyanmethemoglobin method (Oser, 1965) using standards and reagents (Hycel, Houston, Tex.) and a Bausch and Lomb Spectronic 20 Colorimeter (Rochester, New York). A standard 39 curve of optical density (OD) and per cent transmittance (%T) versus concentration (g%) was plotted from dilutions of the prepared standard. Hemoglobin solutions were prepared by adding 20 pl of whole blood to 5 ml of cyanmethemoglobin reagent. Optical densities for each corresponding %T read from the spectrophotometer were calculated using equation 1: OD=log1T (1) Hemoglobin concentrations in gram% were calculated from equa- tion 2: Concentration (unknown) O D (unknown) Concentration (standard? = O D (standard) (2) where gram% and corresponding 0 D from the standard curve were used for the concentration and O D of thestandard. Red cell counts were made using a blood diluting pipette and an A.O. Spencer hemocytomer (Buffalo, New York). Hendrick' s diluting solution (Appendix A) was described by Hesser (1960) as satisfactory for trout blood, whereas Gower' s and Hayem' s diluting solutions result in distortion of the red cells of salmonid species. [To obtain a red cell count, blood is drawn up to the 0. 5 mark on the blood -diluting pipette followed by the addition of Hendrick' 5 solution to the 101 mark and then placed on a vibrating shaker for 2 —3 minutes. 40 Before filling the counting chambers of the hemocytomer, 5 or 6 drops of fluid were removed from the pipette to ensure delivery of an accurately diluted mixture. Samples were counted in two grids and the average RBC count of five secondary squares of each grid were calculated and multiplied by 50, 000 to give the number of RBC' 3 per cmm of blood. Area of 1 secondary square = 0.04 mm2; depth = 0. 1 mm; volume = 0. 004 mm3 (cmm) 1 cmm 0. 004 cmm X 200 = 50, 000 Pipette dilution = 1:200. Radioiron and Isotope Technique The radioiron (as 59Fe C13 in HCl) was obtained from Abbott Laboratories (North Chicago, 111.). Iron59 has a half-life (T%) of 45 days and is a strong gamma emitter with photopeaks at 0. 191, 1. 097 and 1. 289 MEV. The primary stock solution was acidic with a pH of 1. 5 to 2. 5; therefore a dilute solution was pre- pared by adding from 0. 1 to 0. 4 ml of 59FeCl3 to a glass vial (depending on desired concentration) and diluting up to 2. 0 ml with PBS (Appendix A). The pH was maintained by the PBS after adjusting to 7. 2 —7. 6 with 10% Na OH, which was the optimum range for uptake of 59Fe reported by Jandl et al. (1959). A known amount of 59Fe PBS was drawn into a microburet syringe and added directly 41 to each blood sample. A Scientific Industries (Springfield, Mass.) Variable Speed Rotator set at 6 rpm was used to keep the blood sample mixed during the incubation periods. Red cell uptake of 9Fe is temperature dependent (Jandl et al. , 1959; and Jensen et al. , 1953); therefore, all experiments with trout blood were run at 13° C while dog blood samples were maintained at 37° C to insure the proper "in vivo" range of metabolic activity. After incubation samples were removed, radioactivity measured, and the cells washed three times with 1 ml of PBS. Radioactivity of the packed RBC mass after washing was measured and per cent 59Fe incorpora- tion was calculated from equation 3: counts/minute after wash counts/minute before wash 9Fe incorporation = X 100 (3) All counting was done with a Nuclear Chicago (Des Plaines, Ill.) Nal Crystal Well Scintillation Detector. Counting efficiency was estimated at 15 to 20%. The sealer/analyzer (Model 8725) was set to count above 900 KEV, thus counting all photoelectric energy at or above this level which includes the two highest 59Fe photopeaks. This setting was preferred due to the 50% reduction in background counts which resulted when the lower (900 KEV) photo energies were deleted. Although background was reduced, photoelectric energy due to 59Fe gamma radiation was not significantly reduced; therefore, 42 the Rs/ Rb ratio (uncorrected sample rate/background rate) was increased and the necessary counting time shortened, resulting in a more accurate quantitative estimate of the 59Fe present. Dialysis Cellulose dialysis tubing (A. H. Thomas, Phil. , Pa.) with a :1— inch diameter and having an average pore diameter of 4. 8 mp was prepared by boiling 2% inch strips in distilled water, washing 3 times with distilled water and then storing in PBS. One end was tied off with heavy thread, then one solution was added by means of a syringe and needle through the open end. The open end was tied off and the dialysis sacs were placed in disposable glass culture tubes containing the second solution. The 4 ml culture tubes were stoppered and placed on a variable speed rotator at 6 rpm to ensure thorough mixing of each solution. To determine the uptake of 59Fe, whole trout blood was dialyzed against PBS containing radioiron. Two ml of PBS con- taining 0.265 11c 59Fe were added to each culture tube. Approxi- mately 0. 5 to 1.0 ml of pooled whole blood from starved trout was added to a series of dialysis sacs and 0.25 ml of blood pooled from several fish bled 30% of their blood volume a month previously was added to another series of sacs. Sacs were placed in the culture tubes and incubated for 48 hours. At the end of this period the sacs 43 were removed and equal volumes of the saline and whole blood were counted and the percent incorporation of 59Fe by RBC' 3 was determined. In a second experiment labeled and non -labeled whole blood samples were dialyzed against each other to determine how tightly the radioiron was bound by blood. One ml aliquots of whole blood which had been incubated with 59Fe for 48 hours were injected into dialysis sacs and dialyzed against 1. 5 ml of pooled whole blood. After 24 hours incubation the dialysis sacs were removed and 0.2 ml samples of the two blood pools were taken and counted. In the third experiment dialysis sacs containing 0.265 11c 59Fe and 1 ml of plasma were placed into culture tubes containing 1 ml of PBS. After 24 hours incubation the activity in each solution was determined. Acetone was then added to the plasma to precipitate protein. Following centrifugation both supernatant and precipitate were counted and the per cent of activity remaining in each was calculated . Uptake Studies Studies were conducted (a) to determine if plasma has a significant effect on the rate and amount of activity incorporated by the red cells and (b) to test the hypothesis that a peripheral blood 44 sample from a trout should incorporate more iron than a mammalian blood sample due to a greater number of immature forms in the circulation of the trout. In both studies, 8 to 10 ml samples of pooled blood were incubated with 59Fe for various times in a 10 ml microburet syringe attached to a variable speed rotator at 6 rpm. Triplicate 0.2 ml samples of whole blood were removed at various time intervals from 15 minutes to 48 hours, Hct and Hb determined, and the amount of radioactivity measured before washing. The RBC' 3 from each sample were then washed with PBS and centrifuged three times. The counts/minute after washing was converted to 11g 59Fe/g Hb based on the average g% Hb for the pooled blood sample (Appendix B). Jensen et al. (1953) state that in duck red cells no significant in- crease in hemoglobin concentration occurred with the incorporation of iron, thus 59Fe uptake can be accurately expressed in terms of 1.1g 59Fe / g Hb. Radioiron Uptake from Plasma and PBS Ten ml of pooled whole trout blood was divided equally into two tubes and one was centrifuged for 5 minutes at setting 4 on the IEC Clinical Centrifuge. After the plasma had been removed, an equal volume of PBS was added. After the addition of 0.265 11c 45 of radioiron to each sample, they were incubated in microburet syringes for varying lengths of time. The rate and amount of radio- iron incorporation as ug 59Fe/g Hb was determined for both samples. Radioiron Uptake by Dog and Trout RBC' 3 A study was designed to test whether nucleated red blood cells from the peripheral circulation of rainbow trout would incor- porate radioiron to a greater extent than a sample of mammalian red cells. Ten ml of blood was removed from the femoral vein of a normal dog and incubated at 37° C in vitro with 0.265 11c of 59Fe. A 10 ml sample of blood from 5 trout, which had been starved for at least One month, was pooled and incubated at 13° C with 0. 265 11c 59 Fe. Samples were removed at various time intervals, washed, and the amount and rate of radioiron incorporation determined. Uptake Equilibrium Curve The concentration of 59Fe within the RBC measured over a 24- to 48 -hour period was plotted as an equilibrium curve. A two -component analysis of this curve is given in Appendix B. Reticulocyte Study A study was conducted to investigate the effect of hemor- rhage on the number of immature red cell forms in the peripheral 46 circulation of trout and the incorporation of 59Fe by red cells. Fish were kept in the large storage tanks for several weeks and were fed commercial trout pellets every other day. Feeding was continued on a regular basis throughout the duration of the study. Fed trout were in better physical condition than starved trout and were better able “to withstand the stress of bleeding. On the first day each trout was tagged through the lower jaw with a piece of small plastic tape inscribed with a code number and weighed. It was necessary to anesthetize the fish before tagging or weighing by placing them into a solution of MS -222 prepared by adding 0.2 to 0.3 g of the anesthetic to 7 to 8 liters of water. Ten per cent of the total blood volume (estimated to be 3. 5% of the total wet weight) was removed on day one, another 10% on day 3 and then 0. 5 ml blood samples were taken every other day for one week and again at 2 and 3 weeks. Hematocrit, hemoglobin, red cell counts and 59Fe incorporation were measured for each sample obtained. The Student -Newman -Kuels test for multiple comparison for significant difference among means of unequal sample sizes was used to analyze the data obtained (Sokal and Rohlf, 1969a and b). Blood from control and bled fish were compared using New Methylene Blue N Stain for reticulocytes (Appendix A). The 47 red cells were stained by placing a few drops of whole blood into a small glass tube containing about 1 ml of stain and allowed to sit for 5 minutes. After shaking the tube, two drops were removed, placed on a clean glass slide and coverslipped. Dry blood smears could not be used since the dye is a supravital stain. A series of slides were prepared and photomicrographs made using K6hler illumination. Twenty -eight days after initial hemorrhage an uptake study was conducted using 4 ml of pooled trout blood to which 1. 535 11c 59Fewere added. RBC 59Fe incorporation was measured over a 24 hour period and plotted as an equilibrium curve (Appendix B). From the 2 -component analysis the amount and rate of RBC radio- iron incorporation were compared with the control values obtained from pooled blood samples of trout subjected to at least one month of starvation. RESULTS Dialysis From the dialysis experiment it was evident that whole blood was able to take up and retain 59Fe. (See Table 1.) Follow- ing 48 hours of dialysis, whole blood was found to contain from 54 to 62% of the total radioactivity. After separation and three wash- ings with PBS, the packed red cells of the control group retained 31 to 36% while those of the bled group retained 57% of the activity contained in whole blood before washing. In another experiment it was found that trout blood retained radioiron even after 24 hours dialysis against non -labeled blood. Table 2 shows that the non -labeled blood gained no activity from the radiolabeled blood; all activity remained within this sample and virtually no activity was detected in the unlabeled blood. When radiolabeled trout plasma was dialyzed against PBS for 24 hours, 90% of the activity remained in the plasma. Acetone was then added to precipitate the protein in the plasma; and after centrifugation, 90% of the original plasma activity was found in the protein precipitate. 48 49 a .3. man .3 o .8 Sm .N o .3. m3 .2 mg .m 88 a .3. an .H o .3. 3“ .N o .3. 3w .2 is .4 85 o .2" mom s .3. «5 .m m .3. gm .N 28 .v aofinou m .8 ofi .N a .8 mg .m m .3 Sad 42 .2 3.5.80 cmm 5.1 . ms v a3 m .mu you m was... 89 cmm 2:. .5 £8 2 AS. .a ammo a as a. tea. m? e. macaw Ego 130% 883 Big, 5.5 803 22:5 mmm $350393 8:88 28 96am cog one 5 nmfl Hoe.“ @968“: mm? @839, p003 30:? mo “soc pom Esme? .O on“ em mmm @36an mmmm pmfimmw 30.5 cm?— vcw 35:00 89d 6003 mo 39$va mo mason we zeta 2.8 cm; cowwpoaaoofi bravo 133 K. can 333 355833500 5 1,533.34: .H 3an 50 Table 2. -—Results of dialysis of 59Fe -labeled whole blood against non -labeled blood after 24 hours of incubation at 13° C. Activity of labeled Activity of original Percent activity blood after dialysis non -labeled blood remaining in blood (cpm) after dialysis (cpm) after dialysis 4, 7 52 0 100 5, 104 0 100 5, 194 2 99 Uptake Study Radioiron Uptake from Plasma and PBS Trout red cells suspended in plasma incorporated 1. 4 to 1.6 times as much radioiron as RBC' 3 suspended in PBS (Table 3 and Appendix B). Trials A and B are experiments using different pooled blood samples in which the y-intercept of component I represents the amount of activity incorporated between time zero and equilibrium. The y-intercept values of component I for plasma and PBS were 0. 0086 and 0.004 Pgnge/g Hbfor trial A, and 0.0125 and 0.0082 1.1g59Fe/g Hb for trial B, a ratio of 1. 4 to 1. 6 in favor of iron incorporated by RBC' 3 suspended in plasma. The rate of iron incorporation was not dependent on the type of medium, since the half times were not different for plasma 51 or PBS. The half times measured for components I and II in trials A and B were similar. Table 3. ——Half time (T%) and y-inter.cept for components I and 11 Fe incorporation by trout RBC' 8 suspended in plasma and PBS over a 24 hour period at (Appendix B) of 59 13° C. Trial Medium Component T§ (hours) P); -51n;:r}cge§11) A Plasma I 2,. 10 0.0086 A Plasma 11 35. 17 0. 0013 A PBS I 1. 45 O. 0049 B Plasma I 4. 91 0.0125 B Plasma 11 19.78 0.0013 B PBS I 5. 37 0. 0082 B PBS II 16.60 0.0014 Dog vs. Trout Trout RBC' s incorporated 6 or 7 times as much activity than dog RBC' 5 could incorporate over a 24 hour period (Table 4). The trout component I y-intercept value was 0. 0051 compared to 0.0007 Pg59Fe/g Hb recorded for the dog between time zero and equilibrium, illustrating a much greater uptake of iron by the trout RBC' 3 within the 24 hour period (Figure 4). 52 Table 4. --Half time and -intercept for components I and II of trout and dog RBC 9Fe incorporation over a 24 hour period at 13C) and 37° C. _1_ y-intercept Sample Component T2 (hours) (Pg59 Fe lg Hb) Trout I 2. 97 0. 0051 Trout II 18. 84 0.0044 Dog I 18.06 0.0007 Iron incorporation occurred 6 times faster, initially, with trout RBC' 8 compared to the slower uptake of 59Fe by dog RBC' 3. Half times (T%) of componentl illustrate the difference in rate: a T}; of 3 hours for trout compared to an 18 hour T% for dog RBC' 3. Near equilibrium there was no difference in uptake rate between trout and dog; both were within a T313 range of 18 to 19 hours. Reticulocyte Study The hematological data in Table 5 show the effects of repeated bleedings on rainbow trout. The decrease in hematocrit from 38 to as low as 11 is evident in Figure 5. The numbers of RBC' 3 in millions/cmm dropped from 1. 113 to 0. 363 and the Hb fell from 8. 5 g% to as low as 2. 2 g% 20 days after the initial hemor- rhage. There was high correlation coefficient, +0. 988, between the decrease in red cell count and g% Hb over the 20 day period. 53 Figure 4. -— Semi -log plot of components I and II for trout and dog RBC Fe incorporation. The combined equation for components I and II of trout RBC Fe incorporation is . 693 t.I . 693 til CT = 00051 e 2'97 + 0.0044e ”'84 Dog RBC 59Fe incorporation was a one -component process expressed by the equation _ .693 t 18.05 CT = 0.0007 8 The y-intercept value, representing the amount of 59Fe incorporated in 24 hours, for trout RBC' 3 is 6 to 7 times greater than the value measured for dog RBC' s. 0.01 - 0.005 4 0.004 - - Ct o§oa one? firm? $3 3 5 H8 m .0 can 6 ES H when Gorgon po>o§ou mm? CECE 0839, p003 3053 05 .5 ”Eco .Hoa 3:038 .395 nommnuuofion 89d moaaamm @003 .3 moamoogaofion ca can 6 .Omm can HHoHHwHOQHooHHH mm .30: Va #550 Zoo pop 5303253 #:0038031- .m 3an mm 56 .mLmQEHE Z cam .m .m .Hom m 3an mmm .mHmHmEmHmQ @003 mmmfi 9:5me 8 $83on 803 £93 DE: 8 8 8388 8m; woos Ho H8 m .o .3 28 .3 .2 .m i3 :0 .m mac so Emma cam H man no 09582 mwB 25:8 @003 30:? Emu .Hma HEB 430.5 30252 5 263mm wcHnmmB mHHoHHm> E 9550 :mo 9: ES :HnonoEm: .HHHoonEmm: .m mstHm 57 Red cell count (millions/cmm) o o o o o o o o l.‘ to m H CD b L0 0 H H; H' H' o' o' o' :5 l l l L 1 l i l |_| g% Hemoglobin ' C CD 00 I.‘ to L0 0‘) N O I I l I 1 n 1 l l g :3. .5. L. .o oo o o -;_-: m 3 '33 w - >, cd 0 U E as 8 E 'o :3 Q Q) Q) G) 0 B: m m U ‘N >a as Q \\\\\\\\\Igg >: at LO Q 2‘3 :0 T Lu l\\\ \\\\ \wm ' >3 CU Ha l\\\\\\\&\\\ Day 3 R\\\\\\\\\\\\\ Day 1 40- 35' 30.. J: c; N N (%) 11.1001811er 15. 10- o 58 The incorporation of 59Fe by RBC' 8 during 24 hours of incubation rose from 0.080 at the initial bleeding to 0.367 Pgnge/ g Hb 28 days later (Figure 6). A substantial increase in percentage reticulocytes present in the blood occurred between the initial day of hemorrhage, 1. 04%, and 20 days later-when 40. 44% of the red cells were reticulocytes. As illustrated in Figure 7, the increase in pg59Fe/g Hb was related to the rise in reticulocytes with a cor- relation coefficient of +0. 804. The results of the Student -Newman -Kuels test for multiple comparison among the mean values of Hct, Hb, RBC, and 24 hour 9Fe incorporation are summarized in Appendix B. The photomicrographs in Figures 8 to 13 illustrate the increase in reticulocytes in the peripheral circulation. Figure 8 is typical of a control blood sample stained with New Methylene Blue N. There was an abundance of mature erythrocytes in all blood samples taken on the first day of the experiment. The cyto- plasm appears shaded with occasional granules of blue -staining ribosomal RNA while the nucleus appears as a dark, compact oval. The appearance of dark -staining circular immature red cell forms in the blood smears of all hemorrhaged trout was notice- able on day 13 (Figure 9). These reticulocytes appeared as compact masses of dark -staining material. The cytoplasm was filled with 59 .pousmmoa mcHHHHmHHHmL 3339.“ Ho HHHHHOHHHw o5 Hocm mmHnH 5H3 moSHH m bosom? 933 m .UmHm 93. .U omH Hm mason wm .HoH ommm 5H3 HoonnHHoHHH mum? :8 m .3 6003 3053 .Ho moHQEmm .930: $3an gommmmn— 5 m .UmHm 305 .3 333: when u u .w ousmHm mm .93 on .93 ”H .93 m .95 m. .39 H .93 x L. Hr n W l . 15V 255 H 1r :8 265 H 0 6 ll .fi :8 «mod H a: H.405 H ll 5 as .o H in . Ach QMO O H O mhflmwh 67.: .m .m H * . oo.o 10H.o lmH.o loN.o 1mm.o lom.o lmm.o low.o 3d 'SJ 3 a q frZ/QH / 369 . .mHoQEHHHH Z Him .m .m .HoH m 3an wow .oHQEmm @003 m5 HHH mmtnooHHHoHHoH Ho .8385: on... com: Hampamaop mm? aoHHmHoauoocH mnH m 35 mHHHHonpHHH .vow .o+ mm? HcoHoHHHmoo :oHHwHoHHoo SF .305. .502:ku HHH m .UmHm m: Em monooHHHoHHoH «coo Hon 8 :oHHmHoauoocH conHoHan Ho :oHHmHom: H. onsmHnH 62 % reticulocytes among RBC' s 6.33 we; H H H H H o oo.o m I rmod oHi on .o mm 21 AH .c m J a II! 8 cm. Iowa m . w .V L m mmI I immds omI :36 mm 1 memooHsoHHoH axe ‘ umm .o 930: wgnm .33;li O a Spam oHV ._ rowio 63 Figure 8. --Control RBC sample stained with New Methylene Blue N (1000X magnification). Mature erythrocytes are abundant in this photomicrograph with distinct nucleus and cytoplasm. The cytoplasm appears shaded with occasional granules of dark -staining ribosomal RNA while the nucleus appears as a dark, compact oval. Figure 9. --RBC sample stained with New Methylene Blue N from a trout 13 days following the removal of 20% of the blood volume. A. Clump of mature erythrocytes . B. and C. Reticulocyte in its early stages. The nucleus is circular and fills the whole cell; very little ' cytoplasm is evident. 64 0 .Q.‘ § " o C O. . ._ Iou . ~ 3' __ 0 FIGURE 8 FIGURE 9 - 65 Figure 10. -- Sample of RBC' 3 stained with New Methylene Blue N from trout 20 days after removal of 20% of their blood volume (1000X magnification). A. Mature erythrocytes. B. Reticulocytes with elongate cell shape and cytoplasm filled with dark -staining ribosomal RNA. Figure 11. —-RBC sample stained with New Methylene Blue N from trout 20 days after the removal of 20% of the blood volume. A. Mature erythrocyte. B. Reticulocyte with elongate cell shape and cytoplasm filled with dark -staining ribosomal RNA. C. Older reticulocyte in the transition stage to a mature erythrocyte. The nucleus is compact and the cytoplasm has less dark -staining RNA. 66 5*? ‘1'” f ' i ' i u ‘. B 3~& ~ 5 - .. . '5‘ O U s a.“ I A 3' ' 2&- 9 Ion I—-I - fl - FIGURE IO 0 4'. v.00; . V «0 5‘ “g '25:; ‘5'. a“- ' . ; I. ’ 2.1.x FIGURE 11 q 67 Figure 12. --An example of trout mature erythrocytes with elongate cell form, dark -staining nucleus and agranular cyto- plasm (2500X magnification). Figure 13. --RBC sample stained with New Methylene Blue N from a trout 20 days after removal of 20% of the blood volume. A. Mature erythrocytes. B. Early reticulocytes. C. Late reticulocyte, FIGURE I3 69 blue -staining ribosomal RNA which accounted for the abundance of granular material. Eight per cent of the red blood cells were reticulocytes. By day 20 the concentration of immature forms present in the peripheral circulation had increased to 40. 4%. The smaller, elongate cells of Figures 10 and 11 are believed to be older reticulo- cytes just beginning the transition to mature erythrocytes. Figure 12 is an example of mature erythrocytes with elongate cell form, dark -staining nucleus, and agranular cytoplasm. The immature forms in Figure 13 illustrate the scant cytoplasm of the early reticulocyte with circular cell membrane, granular cyto- plasm and a nucleus which composes most of the cell. The late stage of reticulocytosis was noted by the presence of elongate cell forms, filled with dark -staining ribosomal RNA of the endoplasmic reticulum . Bled vs. Non -bled Uptake Study In the uptake study using blood pooled from 4 fish 28 days after initial bleeding, RBC' 8 of bled fish incorporated approximately 30 times more activity as RBC' 3 from control trout, based on the y-intercept values of component I (Table 6 and Figure 14). The initial rate of uptake was 1% times greater for the bled trout with a T% of 1. 8 hours compared to a 2. 9 T% for control trout. 70 Table 6. --Half time and y-intercept of components I and II for 59Fe uptake by bled and control trout RBC' 8. Twenty per cent of the whole blood volume was removed from fish in the bled group on day 1 of the experiment. Sample A Component T% (hours) (Pg-531:3}:pr Control I 2, 97 0.0051 Control II 18. 84 0. 0044 Bled I 1. 83 0.1547 Bled II 23. 42 0. 1442 71 Figure 14. —- Semi-log plot of components I and II of 59Fe uptake rates for control and bled trout. Bled trout were hemorrhaged 10% whole blood volume on day 1 and on day 3, and O. 5 m1 of blood was removed on day 5, 13, 20, and 28. The equation for component I and II for the bled group was _.693 t1 _ .693 tn CT = 0.1547 e 1'83 + 0.1442 e 23'” The equation for the control group was .693tI .693 t -— - 11 CT = 0.0051 e 2'97 + 0.0044e ”'84 The y-intercept values indicate that RBC' 3 from bled trout incorporated 30 times as much activity as those from control trout after 24 hours of incubation at 13° C. See Appendix B. .05 m 59 Ceq : Ct Pg FG/g Hb .005 ‘ 72 Component II Bled 'I'I‘Ol-l’c Component I Bled Trout Component 11 Control Trout Component I Control Trout I 2.0 3.0 4.0 50 1. 0 Time (hours) Figure 14 DISCUSSION Dialysis The results of the dialysis experiments coincide with those of Walsh (1949) who reported that immature human red cells were capable of assirnilating iron and synthesizing heme in vitro and Shemin, London, and Rittenberg (1950) who reported similar find- ings with nucleated red cells from ducks. The fact that iron was tightly bound to whole trout blood is supported by J andl et al. (1959), who found that radioiron was absorbed onto the membranes of red cells and was not easily removed even with the use of EDTA, an effective chelator. Fletcher and Huehns (1968) attributed the tight binding to the presence of transferrin in the plasma of human blood. The presence of an iron -binding protein in teleost plasma was reported by Utter, Ames, and Hodgins (1970) in an electrophoretic study of transferrin in coho salmon plasma. Uptake Studies Evidence from the uptake studies indicates that the iron- binding protein plays a significant role in iron incorporation, since 73 74 trout red cells suspended in plasma incorporated 1. 5 times as much 59Fe as those suspended in PBS. This agrees with Jensen et a1. (1953), who found that duck red cells suspended in plasma took up 1. 5 times more iron than cells suspended in 0. 15M NaCl. J andl et al. (1959) made similar observations on heme which had been isolated from human reticulocytes suspended in plasma and saline. They attributed the difference between plasma and saline to be the fact that immature red cells irreversibly incorporate iron only from transferrin. The two -component uptake curves for fish RBC' s were the result of two processes occurring during iron incorporation. Component I is believed to represent the binding of 59Fe to the stroma or red cell membrane. This appeared to be a fairly rapid process involving the transport of iron from transferrin to a specific site of attachment on the membrane. Component II may represent the iron incorporation by the storage areas on the mitochondria or into the heme (Schemin et a1. , 1950). This may involve an iron receptor in the cytoplasm which could transport the iron to either site in a relatively slow process. A two —component equilibrium curve was noted by Jensen et al. (1953), who reported the rate of iron incorporation to be more rapid within the first 6 hours of incubation and then decreased as equilibrium was approached at 24 hours. 75 Data reported here indicate that the presence of reticulocytes and other immature forms in the peripheral circu- lation is responsible for the increased amount and rate of iron uptake. Walsh et al. (1949) demonstrated this to be true for human blood where samples containing less than 1% reticulocytes took up no measurable quantity of radioiron compared to the amount of 59Fe uptake in blood from patients with pernicious anemia. The results of the dialysis experiment with blood from control and bled trout indicated a possible correlation between the effects of bleeding and radioiron incorporation. Observations made during the long —term reticulocyte study confirmed this to be true. An excellent correla- tion has been noted between the amount of iron uptake and the number of reticulocytes present in the peripheral circulation of ducks (Jensen et al. , 1953). Jandl et a1. (1959) report similar findings in experiments with samples of peripheral blood of humans. Effects of Hemorrhage That hemorrhage is an excellent erythropoietic stimulus in trout is supported by data obtained in the bleeding experiment. Removal of between 20 and 30% of the blood volume over a 5 —day period and subsequent weekly removal of 0. 5 m1 (5 - 10%) of blood resulted in drastic changes in Hct, Hb, RBC count, per cent 76 reticulocytes and 59Fe uptake. The mean values for these parameters on the initial day of bleeding were generally higher than those reported by Schiffman and Fromm (1959); however, this may be due to seasonal variation. The sharp fall in all three parameters 5 days after the initial bleeding was in direct contrast to the results of Zanjani et al. (1969) in their study of the effects of bleeding on erythropoiesis in the gourami. They reported that red cell regeneration after removal of 25 to 30% of the whole blood volume occurred rapidly and, as early as 5 days following bleeding, considerably greater values were obtained for Hct, Hb, and RBC count in bled as compared to non ~b1ed fish. Such a rapid recovery did not occur in rainbow trout. A significant increase in circulating immature red cell forms was noted in rainbow trout after the initial removal of 20 to 30% of the whole blood volume. In experiments with turtles, Altland and Thompson (1958) first noticed a marked increase in reticulocytes 17 days after the initial bleeding and reported that high'reticulocyte counts continued for at least 21 days and as long as 50 days in some instances. In another study with turtles, Hirsch- feld and Gordon (1965) did not find significant differences in reticulocyte counts of control and bled turtles until after 2 weeks of "repeated bleedings. " Thus the increase in reticulocytes appearing 77 in the peripheral circulation is a delayed, long -term response in turtles. This was also true for rainbow trout, since the high reticulocyte count on day 20 indicated that the increased release of immature forms into the circulation was still in progress. Hemorrhage resulted in a significant increase in 59Fe incorporation by trout red cells within 5 days of the initial bleed - ing. Hirschfeld and Gordon (1965) found a significant increase in the levels of 59Fe -1abeled turtle erythroblasts at 6 days following initial hemorrhage; and Zanjani et a1. (1969) report considerably greater values in radioiron incorporation by gourami RBC' 3 as early as 5 days following bleeding. The increase in 59Fe uptake was noted before the appearance of trout reticulocytes, although the two are related. The more accurate parameter used to measure erythropoiesis is believed to be 59Fe incorporation, due in part to the difficulty in distinguishing reticulocytes from mature red cells, both of which stain with New Methylene Blue N. Hirschfeld and Gordon (1965) refer to the inaccuracy of reticulocyte counts and indicate that better measurements of erythropoietic activity can be made with the use of radioiron and autoradiographs. From the results observed. in the experiments previously described, it is concluded that hemorrhage has a direct influence upon the mechanism for re -establishing red cell volume in rainbow 78 trout. The decrease in red cell volume, volume % O and perhaps 2. blood pressure probably triggers the production and output of a hormone such as ESF which is distributed via the circulation to target tissues for the stimulation and increase of red cell produc- tion. These erythrocyte production and storage centers include the kidney and spleen (Duthie, 1939). With the increase in the produc- tion of polychromatocytes, as outlined by Klontz et al. (1963), the ESF titers decrease due possibly to destruction by an enzyme or antibody, or due to a feedback inhibition mechanism to the sites of ESF production, the kidney and liver. Red cell precursors enter the peripheral circulation and undergo maturation (Andrew, 1965; and Jordon, 1938). They receive the iron necessary to complete the molecules of hemoglobin from transferrin present in the plasma. After hemoglobinization, or maturation, the red cells are able to carry oxygen to the cells of the body in exchange for carbon dioxide. Upon destruction of the red cell, hemoglobin is broken down and iron is released to storage pools or to transferrin. The process from initial stress to the output of immature red cell forms into the circulation appears to be variable, but in rainbow trout it may be the matter of 5 days to a week before the response can be detected. 79 Application to Fisheries and Environmental Studies Results of the hemorrhage experiments in this study indicate that erythropoiesis and the appearance of red cells in the peripheral circulation of fish can be measured accurately by RBC 59Fe incor- poration and can be easily supplemented with reticulocyte and RBC counts, hematocrit and hemoglobin determinations. Determination of these parameters should have merit in measuring the erythro— poietic effect of environmental pollution and in studies of the general health of hatchery fish. In a study of the effects of heavy metals on fish, informa- tion on histological changes in the erythropoietic tissues could be supplemented with determinations of the number of immature erythrocytes present in the circulating blood by the methods out- lined above. Pollutants such as heavy metal ions may affect the oxygen carrying capacity of blood by blocking heme synthesis or binding to transferrin, eventually resulting in tissue hypoxia. As an initial response to oxygen deficiency, RBC storage areas in the spleen may be emptied, causing an increase in the number of circulating immature forms, which could be detected histologically and by 59Fe uptake studies. Release of cells from the storage areas may be insufficient compensation and, as a result, after a period 80 of time values for circulating reticulocytes and RBC 59Fe uptake may decrease. The erythropoietic centers would then be stimulated and again this response could be checked histologically and any new production and release of RBC should be reflected in increased 59Fe uptake by RBC' 3. If the sites of red ..cell production are destroyed by heavy metals, the number of circulating immature RBC' s would gradually decrease following spleen RBC depletion, as would RBC 59Fe uptake. A decrease in oxygen -carrying capacity due to substitution of heavy metal ions for iron combined with destruction of erythro- poiesis centers would eventually cause death. Release of immature red cells from storage areas would represent a transient compensa- tion for tissue hypoxia. Procedures for Monitoring Immature RBC Numbers in Fish Monitoring of the effects of pollutants or disease on the physiology of fish blood can be done by the procedures outlined here. Hematological measurements are efficient, accurate and require little expensive equipment. Blood samples may be taken at any monitoring station and transported to a laboratory for the following analyses: 81 Hematocrit, Hemoglobin and RBC counts. Repeated checks of these basic hematological parameters will give an indi- cation of the overall well -being of the experimental animals. Determination of RBC 59Fe incorporation. Data for groups of fish could be obtained by making determinations on a pooled sample of blood, whereas, using only 0.2 ml of blood from individuals provides the opportunity of studying this parameter in individual fish over a period of time. This determination is relatively simple but does require use of a gamma ray detector. Reticulocyte counts. Althoughdifficulty was encountered in distinguishing reticulocytes from mature cells using New Methylene Blue N, measurements of immature RBC forms can be a valuable parameter in stress physiology along with information on RBC 59Fe uptake. Reticulocyte determinations are tedious, but modification of the staining technique could eliminate some of the drudgery. Hirsch- feld and Gordon (1965) state that the benzidine -Wright staining technique for hemoglobin is a more sensitive method for measuring immature forms in the circulation than the supravital technique using New Methylene Blue N. Lucas and Jamroz (1961) have published an atlas of avian 82 hematology which contains several color illustrations of developing nucleated red cells from their origin to matura- tion using the benzidine -Wright and supravital stains. This reference may be valuable in identification of immature fish RBC' 5 and it contains outlines of the various staining procedures. Autoradiographic techniques have also been suggested by Hirschfeld and Gordon (1965) as a sensitive measure of immature red cells in circulating blood. Smears of radio- labeled RBC' 3 can be prepared, dipped in Kodak NT 33 nuclear track emulsion, exposed at 4° C for 10-14 days and developed in Kodak D-19. SUMMARY AND CONCLUSIONS Red cells from control trout were capable of incorporating 30 to 36% of the 59Fe present in whole blood and red cells from bled fish bound as much as 57% of the available radioiron over a 48-hour period. Trout erythrocytes incorporated 1. 5 times more radioiron when suspended in plasma than when suspended in PBS. An iron -binding protein functionally similar to human trans - ferrin is present in the plasma and plays a significant role in iron incorporation. There were two rate constants for iron incorporation by trout erythrocytes. It is suggested that the initial fast uptake, with a half -time of 3 hours, represents the transfer of iron to the red cell membrane and the slower uptake, with a T% of 19 hours, represents the incorporation of iron into heme or some intra- cellular storage site. 83 84 The average 59Fe uptake by control trout red cells was 0. 005 ug/g Hb between time zero and equilibrium (48 hours). Red blood cells from bled trout incorporated 30 times more activity than control trout red cells between time zero and equilibrium (48 hours). Normal trout red cells incorporated 6 to 7 times the amount of activity taken up by dog red cells for the same time period. Hemorrhage of 20 to 30% of the total blood volume caused a sharp decrease in hematocrit, g% hemoglobin and red cell count. Iron incorporation increased from O. 080 to 0. 367 pg59Fe/g Hb between the initial bleeding and day 28. There was a direct correlation (Corr. Coef. = 0.804) between per cent reticulo- cytes and 59Fe incorporation by the red cells. It appears that 59Fe is a very accurate and useful tool for measuring erythropoiesis in trout and exceeds the accuracy of reticulocyte counts as an indicator of hematological stress, since it is difficult to distinguishimmature from mature red cells using New Methylene Blue N stain. BIB LIOGRAP HY BIBLIOGRAPHY Altland, P. D., and K. C. Brace. 1962. Red cell life span in the turtle and toad. Am. J. Physiol., 203:1188-1190. Altland, P. D., and M. Parker. 1955. Effects of hypoxia upon box turtle. Am. J. Physiol., 180:421-427. Altland, P. D. , and E. C. Thompson. 1958. Some factors affect- ing blood formation in turtles. Proc. Soc. Exp. Biol. Med. , 99:456-459. Andrew, W. 1965. Comparative Hematology. Grune and Stratton, New York. 188p. Bozzini, C. E., M. E. Rendo, F. C. H. Devoto, and C. E. Epper. 197 0. Studies on medullary and extramedullary erythro- poiesis in the adult mouse. Am. J. Physiol., 219:724-728. Deutsch, M., and V. E. Engelbert. 1970. Erythropoiesis through clone cell formation in peripheral blood of white sucker, Catostamus commersoni. Can. J. Zool. , 48:1241-1250. Doan, C. A. 1932. Current views on the origin and maturation of the cells of the blood. J. Lab. and Clin. Med. , 17:887 -898. Duthie, E. S. 1939. The origin, development and function of the blood cells in certain marine teleosts: I. Morphology. J. Anat. , (London) 73:396 -412,. Engelbert, V. E., and A. D. Young. 1970a. Erythropoiesis in peripheral blood of seven species of New Zealand and one specie of Canadian birds. Can. J. Zool. , 48:227-230. Engelbert, V. E. , and A. D. Young. 1970b. Erythropoiesis in peripheral blood of tuatara (Sphanodon punctatus) and turtle (Malaclemys terrapin). Can. J. Zool., 48:209—212. 85 86 Finch, C. A. 1957. The role of ironin hemoglobin synthesis. Conference on Hemoglobin (May 2 -3), United States Govern- ment, Washington, D.C. 95-99p. Fletcher, J., and E. R. Huehns. 1968. Function of transferrin. Nature, 218:1211-1214. Fried, W., C. Johnson, and P..Heller. 1970. Observation of regulation of erythropoiesis during prolonged periods of hypoxia. Blood, 36:615-616. Ganong, W. F. 1969. Review of Medical Physiology. IV. Lange Medical Publications. Los Altos, California. 628p. Goldberg, A. , H. Ashenbrucke, G. Cartwright, and M. Wintrobe. 1956. Studies on biosynthesis of heme in vitro by avian erythrocytes. Blood, 9:821 -833. Gordon, A. S. 1935. Effect of low pressure on the blood picture of Necturus maculosus. Proc. Soc. Exp. Biol. Med. , 32:820-822. Gordon, A. S. 1959. Hemopoietine. Physiological Reviews, 39: 1-40. Hesser, E. F. 1960. Methods for routine fish hematology. Pro- gressive Fish-Culturist, 22:164-171. Hirschfeld, W. J., and A. S. Gordon. 1965. The effect of bleed— ing and starvation on blood volumes and peripheral hemo- gram of the turtle, Pseudemys scripta elegans. Anatomical Record, 153:317-323. ' J akowska, S. 1956. Morphology and nomenclature of cells of blood of teleosts. Rev. d' Hematologic, 11:519-539. Jandl, J., J. Inman, R. Simmons, and D. Allen. 1959. Transfer of iron from serum iron -binding protein to human reticulo- cytes. J. Clin. Invest., 38:161-185. Jandl, J. , and J. Katz. 1963. The plasma -to -cell cycle of trans- ferrin. J. Clin. Invest., 42:314-326. 87 Jensen, W., H. Ashenbrucker, G. Cartwright, and M. Wintrobe. 1953. The uptake 3 vitro of radioactive iron by avian erythrocytes. J. Lab. Clin. Med., 42:833-846. Jordon, H. E. 1938. Comparative hematology. in: Handbook of Hematology: II. Edited by H. Downey. Hafner, New York. Jordan, H. E., and C. C. Speidel. 1924. Studies on lymphocytes: 11. The origin, function and fate of the lymphocytes in fish. J. Morphol., 38:529-546. Kaplan, E. , W. Zuelzer, and C. Mouriguard. 1954. Sideroblasts: A study of stainable nonhemoglobin iron in marrow normo- blasts. Blood, 9:203-213. Klontz, G.,» W. Yasutaka, J. Wales, L. Ashley, and C. Smith. 1963. The application of hematological techniques to fishery research. Unpublished manuscript. Western Fish Disease Laboratory, Seattle, Washington. Krzymowska, H., I. Iwanska, and A. Winnicki. 1960. The effect of the erythropoietic factor from sheep on erythropoiesis in carps (Cyprinus carpio L). Acta Physiol. Pol. , 11:425. Lucas, A. M., and C. Jamroz. 1961. Atlas of Avian Hematology. United States Dept. of Agriculture, Washington, D. C. 271p. Maximow, A. A. 1924. Relation of blood cells to connective tissues and endothelium. Physiol. Rev. , 4:533-563. Meints, R. 1969. Erythropoietic activity in Rana pipiens: The influence of severe hemolytic induced anemia on spleen and peripheral blood activities. Comp. Biochem. Physiol. , 30 : 383 -389. Miller, A. T., and D. M. Hale. 1970. Effect of cobalt on oxygen consumption of rats at various ambient temperatures. Arch. internat. Physiol. Bioch. , 78:475-479. Neuberger, A. 1951. Studies in mammalian red cells. In Ciba Foundation Conference on Isotopes in Biochemistry, G. E. W. _Wolstenholme, (Ed.), The Bla'kiston Co. , Philadelphia, Pennsylvania. 288p. 88 Oser, B. L. (Ed.), 1965. Hawk's Physiological Chemistry. McGraw -Hill, New York, New York. 1096p. Prosser, C. L., L. M. Barr, R. Pine, and C. Lauer. 1957. Acclimation of goldfish to low concentrations of oxygen. Physiol. Zool. , 30:137 -141. Ralph, P. H. 1941. The histochemical demonstration of hemoglobin in blood cells and tissue smears. Stain Tech. , 16:105-107. Ross, J. F. 1946. The metabolism of inorganic and hemoglobin iron. J. Clin. Invest., 25:933. Rosse, W. , and T. Waldmann. 1966. Factors controlling erythro- poiesis in birds. Blood, 27:654-661. Rosse, W. , T. Waldmann, and E. Hull. 1963. Factors stimulating erythropoiesis in frogs. Blood, 22:66 -71. Sanchez -Medal, L. , L. Duarte, and J. Labordini. 1970. Hemolysis and erythropoiesis. VI. A comparative study of the utiliza- tion of hemoglobin iron and transferrin iron by the erythro— poietic tissue. Blood, 35:721-726. ‘ Schiffman, R. H., and P. O. Fromm. 1959. Measurement of some physiological parameters in rainbow trout. Can. J. Zool., 37:25-32. Shemin, D., I. London, and D. Rittenberg. 1950. The synthesis of protoporphyrin in vitro by red blood cells of the duck. J. Biol. Chem. , 183:757 -758. Sokal, R. R., and F. J. Rohlf. 1969a. Biometry. W. H. Freeman and Company, San Francisco, California. 776p. Sokal, R. R. , and F. J. Rohlf..1969a- Statistical Tables. W. H. Freeman and Company, San Francisco, California. 253p. Utter, F. M., W. E. Ames, and H. O. Hodgins. 1970. Trans- ferrin polymorphism in coho salmon (Oncorhynchus kisutch). J. Fish. Res. Ed. Canada, 27:2371-2373. 89 Walsh, R., E. Thomas, S. Chow, R. Fluharty, and C. Finch. 1949. Iron metabolism. Heme synthesis _i_n vitro by immature erythrocytes. Science, 110:396 -398. Wintrobe, M. 1961. Clinical Hematology. 5th Edition. Lea and Febiger, Philadelphia, Pennsylvania. 1186p. Yoffey, J. M. 1929. A contribution to the study of the comparative histology and physiology of the spleen, with reference chiefly to its cellular constituents. J. Anat. , 63:314-344. Yuki, R. 1957. Blood cell constituents in fish. I. Peroxidase staining of the leucocytes in rainbow trout (Salmo irideus). Bulletin of Faculty of Fisheries, Hokkaido University, 8:36 -44. Zanjani, E., J. Contrera, G. Cooper, A. S. Gordon, and K. Wong. 1967. Renal erythropoietic factor: Role of ions and vaso- active agents in erythropoietin formation. Science, 156: 1367 -1368. Zanjani, E., M. Yu, A. Perlmutter, and A. S. Gordon. 1969. Humoral factors influencing erythropoiesis in the fish blue gouramiuTrichggjlster trichopterus. Blood, 33:573-581. APPENDICES follows. APPENDIX A SOLUTIONS AND STAINS Hendrick' 3 Blood Diluting Solution (Hesser, 1960) Sodium sulfate 10.0 gm Sodium chloride 2. 5 gm Sodium citrate 1. 5 gm Glacial acetic acid 50. 0 ml Distilled water to make 500. 0 ml Phosphate Buffered Saline Solution (PBS) One liter of a 280 mOSm PBS solution was prepared as Sodium chloride 7. 83 gm Potassium chloride 0. 30 gm Magnesium sulfate 0. 14 gm Potassium phosphate monobasic 0. 46 gm Sodium phosphate dibasic 2. 02 gm Distilled water to make 1000 ml New Methylene Blue N Stain for Reticulocytes (Wintrobe, 1961) New Methylene Blue N 0. 5 gm Potassium oxalate 1. 6 gm Distilled water 100 ml 90 APPENDIX B EQUATIONS AND CALCULATIONS Calculation of pgnge/ g Hb The amount of radioactivity (pc) added to the total blood sample was converted to pg Fe using the specific activity (pc/pg) measured for the stock solution of 59FeC13. Equation 4 was used to calculate the pg of 59Fe in each aliquot of blood before washing. totalJlg 59Fe added to total blood sample = pg59Fe (4) number of aliquots removed aliquot Counts/minute before wash were converted to cpm/pg59Fe by equation 5, cpm/aliquot before wash _ cpm before wash (5) pg 59Fe/ aliquot pg59Fe and the counts/ minute after waSh were converted to pg59Fe incorporated by cpm after wash = Pgnge incorporated (6) cpm before wash/pg 59Fe 91 92 The number of pg59Fe/g Hb was calculated from the pg59Fe incorporated and the average g% hemoglobin measured for the pooled blood sample as shown in equations 7 and 8. g Hb/100 ml blood X ml blood/aliquot = g Hb/aliquot (7) 59 pg Fe incorporated _ 59 g Hb/aliquot of blood — pg Fe/g Hg (3) Mathematical Analysis of Uptake Equilibrium Curve In all the uptake studies the concentration of 59Fe within the red cells was plotted on arithmetic paper as pg59Fe/g Hb on the ordinate and time on the abscissa. The equilibrium curve which resulted from such a plot demonstrated a rapid increase in iron incorporation within the first 4 hours of incubation which leveled off into a rather slow uptake (Figure 15). A point near the tail -end of the curve (at the time of last sample) was labeled the equilibrium concentration, Ceq' The concentration, pg59Fe/g Hb, at each sample period was subtracted from the equilibrium concentration, and the difference, Ceq - Ct’ was plotted on semi -log paper with time on the abscissa (Figure 16). The parabolic shape of the plot indicated that there were two components in the uptake curve: one component represented the initial uptake period from time zero to 93 1:00 oh m m 0mm oHHH chomoHaoH Um O .HHHHHHHQHHHHHHoo Hm :oHHmficoo .9550 gHunHHHHHHoo 0x53: ommm HHm .Ho oHQEmNmII .mH oHHHMHnH 94 NH A .35 35H. 3H m b 2 8%E -m 'V‘ Sod uoH.o 3d ImH.o B a QH / .569 vom.o .mm.o uom.o 95 Figure 16. -- Example of calculation and semi-log plot of components I and II of a two -component uptake curve. 0.05 0.002. Observed values of Ceq - C \ \\ Extrapola ted values from L Component II 96 _ -k11t C1:11 ‘ C011 9 Component 11 Component I = Plot of Observed - Extrapolated Values of C - C eq t C11 : Col e-klt _ 'kIt 'knt Ctotal — CoI e + CoII e I 1% 2 4 6 8 10 1'2 1'4 1'6 1'8 23 22 Time (hrs. ) Figure 16 97 about 4 hours, and the second component represented the uptake between 4 hours and equilibrium. A linear regression analysis was made of the second component with the use of an Olivetti, Under- wood Programma 101 computer, and the Ceq - Ct values were extrapolated to time zero. The y-axis values obtained from the extrapolation of Component II were subtracted from the observed values of Component I for the corresponding time, t. A linear regression analysis of the difference was computed and the slope, y-intercept, and half-time of the difference were used to plot Component I (equation 9). C t = C0 e"kt (9) Where: C1: = concentration at time t Co = concentration at time zero (y -intercept) e = base of natural logarithm k = rate constant = 0.693/T% = slope t = time in hours T-§- = half—time or time necessary to reach% the concen- tration at any time t Components I and II were expressed by equations 10 and 11. - -k1t CtI _ C01 e (10) 98 C = C ’kHt tII 011 e (11) Equation 12 is an expression of the total activity incorporation in the two -component analysis and represents the activity acquired by two separate uptake processes of the red cell. C e—kIt + C e.kIIt (12) Ctotal 2 01 011 Student -Newman -Kuels Test for Comparison of Multiple Means Among Unequal Sample Sizes (Sokal and Rohlf, 1969a and b) The observed range value must be higher than the LSR test statistic to be significantly different at the (X = 0.05 level. All least significant range (LSR) values are significantly different except those marked with an asterisk. 99 Hematocrit LSR Test Statistic Day 1 3 5 13 1 ___ 3 4. 455 --- 5 5. 616 4. 808 --- 13 7.056 6.555 5.568 --- 20 8.224 7.867 7.373 6.680 Observed Range Values Day 1 3 5 13 1 3 7.891 —-— 5 15.792 7.901 ___ 3 24.114 16.223 8.322 ___ 0 26.268 18.796 10.895 2.273* Hemoglobin LSR Test Statistic Day 1 3 5 13 1 ___ 3 1.086 --- 5 1.302 1.142 -—- 13 1.752 1.651 1.378 --- 20 1.927 1.871 1.704 1.617 Observed Range Values Day 1 3 5 13 1 ___ 3 3. 183 --- 5 3. 842 0.659* --- 13 5.279 2.096 1.437 --- 20 6.503 3.220 2.661 1224* 20 20 20 20 100 Red Blood Cell Count .Day 1 1 ___ 3 0.133 5 0.165 13 0.205 20 0.245 iDay 1 1 ___ 3 0.283 5 0.493 13 0.635 20 0.750 pg 59Fe/g Hb iDay 1 1 _-_ 3 0.080 5 0.096 13 0.126 20 0.143 28 0.156 IDay 1 1 ___ 3 0.071* 5 0.100 13 0.203 20 0.270 28 0.287 LSR Test Statistic 3 5 13 0.141 --- 0.190 1.163 --- 0.234 0.218 0.196 Observed Range Values 3 5 13 0.210 —-- 0.352 0.142* --- 0.467 0.257 0.115* LSR Test Statistic 3 5 ' 13 0.085 --- 0.120 0.100 --- 0.140 0.127 0.118 0.154 0.145 0.146 Observed Range Values 3 5 13 0.029* —-- 0.132 0.103 --- 0.199 0.170 0.067* 0.216 0.187 0.084 20 20 20 0.127 20 0.017* 28 28 "7'11 1111 1717111711111 1“