w, .L .‘.‘,~ u: T‘ a.‘ h' . l .1. 1‘ .-.~':q,.‘. A. . '5' .... 1%. i A ammun- . _ . g. UPTA'KE- msmsunou AND mconponmou or 59R 1A 7sz AND BLOOD OF , RAIE’TBOW TROUT (Saimo gja Hdflefit) ~ . ,Di‘sjslenafi on for the Degree of Ph. D MECHEGAN STATE UNIVERSTTY RICHARD E. WALTER ' 3975 w. . ‘ . 7 _ ‘ . E r . This is to certify that the thesis entitled Uptake, Distribution, and Incorporation of 59Fe in Tissue and Blood of Rainbow Trout (Salmo gairdneri) presented by Richard L. Walker has been accepted towards fulfillment of the requirements for Ph . D . degree in Physiology BM 0 . $310M“. Major professor Date February 14; 1975 0-7639 ~t "'5’ ' 3mm 97 E" HUAB & SUNS' 800K BINDERY INC. LIBRARY BINDEHS ‘ SPRINGPOFT. NICHTGI'I J ,“ l £7 <:) ABSTRACT *A V) UPTAKE, DISTRIBUTION, AND INCORPORATION \ OF 59Fe IN TISSUE AND BLOOD OF RAINBOW TROUT (Salmo gairdneri) BY Richard L. Walker There is little information in the literature pertain- ing to iron metabolism and erythrocyte production in fish. This study was designed to evaluate the storage iron facili— ties in various tissues and to trace the distribution of radioiron in tissues and blood following an intraperitoneal (i.p.) injection of 59F e. Iron deficiency anemia was in- duced in an experimental group of rainbow trout in order to measure its effect on red blood cell production and mobil- ization of storage iron. Forty percent of the blood volume of 48 experimental rainbow trout was removed in 4 separate bleedings over a seven day period. On the day of the last bleeding these fish and an additional 48 controls were injected, i.p., with 59Fe (1 uCi/lOO g). Six control and six experimental fish were killed on days 1, 2, 4, 8, ll, 16, 23, and 30 following the injection. The percent of initial injected 59Fe, total iron concentration, and specific activity were measured in the liver, spleen, head kidney, pyloric caeca, intestine, Richard L. Walker and skeletal muscle. Samples Of fecal material, bile and urine were also collected for analysis of the 59Fe content. Plasma iron, total iron binding capacity, hematocrit, hemoglobin, and red blood cell 59Fe incorporation were measured. Most of the 59Fe was absorbed from the peritoneal cavity within 24 hrs. after the i.p. injection. Equilibrium between the plasma 59Fe pool and that Of the tissue was established by day 8. Experimental fish RBC 59Fe content increased to 70-80% of the initial injected dose by day 16 compared to 50% in the controls. This was attributed to the difference in reticulocyte count which was 10-12% for the bled and 2-3% for control fish. The rate that iron is in- corporated into hemoglobin by immature red cells is much slower (about half) than the rate of RBC 59Fe uptake, thus, iron is temporarily stored in the cytoplasm. The iron for hemoglobin formation was Obtained from liver iron stores which dropped from 12% to less than 1% of the initial in- jected dose by day 16. Total iron concentration in liver decreased from 200 to less than 100 ug Fe/g. The decrease in liver iron may have stimulated iron absorption by the intestine and pyloric caeca. There is evidence for a feed- back mechanism mediated by transferrin. UPTAKE, DISTRIBUTION, AND INCORPORATION OF 59Fe IN TISSUE AND BLOOD OF RAINBOW TROUT (Salmo gairdneri) BY Richard L: Walker A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology 1975 ACKNOWLEDGMENTS I thank Drs. P. O. Fromm and J. R. Hoffert for their advice and guidance and for assistance in preparing the manuscript. Thanks are also given to Drs. L. F. Wolterink, R. A. Bernard and H. E. Johnson for their review of the manuscript and helpful suggestions. I am grateful to Esther Brenke for her invaluable technical assistance during all phases of this study. Above all I wish to thank Harold Bergman for his advice, concern and encouragement and for the time spent working on the computer programs. Thanks are given to other students who assisted in the data collection. I am grateful to Dr. Fromm for financial support through the Michigan Agricultural Experiment Station, Project 122, and the EPA Grant NO. R-80lO34. ii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O I O O O O O 0 LIST OF FIGURES O O O O O O O I O O O O O O O O O O INTRODUCTION AND LITERATURE REVIEW . . . . . . . . Control of Iron Absorption. . . . . . . . . . . Luminal Factors Controlling Iron Absorption. Mucosal Cell Iron Absorption . . . . . . . . Active Transport and Enzymatic Control . . . Non—Ferritin Intermediates . . . . . . . . . Serosal Absorption of Iron . . . . . . . . . Plasma Iron . . . . . . . . . . . . . . . . . . Transferrin . . . . . . . . . . . . . . . . . Structure and Chemical Nature. . . . . . . . Delivery of Iron to the Reticulocyte . . . . Erythropoiesis. . . . . . . . . . . . . . . . . Erythropoietic Tissue. . . . . . . . . . . . Erythropoietic Stimulants. . . . . . . . . . Erythropoietic Depressants . . . . . . . . . Erythropoietic Stimulating Factor and Renal Erythropoietic Factor . . . . . . . . . . Immature Red Blood Cell Iron Uptake . . . . . . Red Cell Life Span . . . . . . . . . . . . . Storage Iron. . . . . . . . . . . . . . . . . . Ferritin and Hemosiderin . . . . . . . . . . Liver and Spleen as Iron Storage Organs. . Iron Storage in Intestine, Muscle and Kidney Summary. . . . . . . . . . . . . . . . . . . RESEARCH RATIONALE O C O O O O O C O O O O I O O O 0 MATERIALS AND METHODS. . . . . . . . . . . . . . . Experimental Animals. . . . . . . . . . . . . . Experimental Protocol . . . . . . . . . . . . . Bleeding Technique. . . . . . . . . . . . . . . Tissue sample I O O O O O O O O O O O O O O O 0 iii 33 33 33 34 34 TABLE OF CONTENTS--continued Page Bile sample 0 O O O O O I O O O O O O O G O O O 35 Measuring 59Fe Activity . . . . . . . . . . . . 36 Hematocrit, Hemoglobin and Reticulocyte Counts. 37 Plasma Iron, UIBC and TIBC. . . . . . . . . . . 38 Tissue Total Iron Analysis. . . . . . . . . . . 39 BlOOd VOlU-Ine I O O O O O O O O O O O O O O I O O 40 Urine Collection. . . . . . . . . . . . . . . . 41 Statistical Analyses. . . . . . . . . . . . . . 44 RESULTS 0 O O O O O O O O O O O O O 0 O O O O O O O 45 General Health of Animals . . . . . . . . . . . 45 Tissue Data Interpretation. . . . . . . . . . . 45 Hot, Hb and Reticulocytes . . . . . . . . . . . 46 Comparison of Tissue Specific Activity. . . . . 49 Plasma. C I C O O O I O O I O O O O O O O O O O 49 Plasma Iron, UIBC and TIBC. . . . . . . . . . . 53 Head Kidney and Kidney. . . . . . . . . . . . . 55 Liver O O C O O I C 0 O O O O O O O O O O O O O 58 Spleen. C O O C O I O O I O O O I O O O O O O O 61 Pyloric Caeca . . . . . . . . . . . . . . . . . 61 Intestine O O 0 O O O O C O O C O O O O O O O O 67 MuSCle. . O O O O C O O O O, O O C O O O O O O O 70 Bile VOlmne O O O O C O O O I O O O O O O O O O 70 Percent Initial Dose in Bile. . . . . . . . . . 73 Intestinal Contents and Urine . . . . . . . 73 Whole Blood and Whole Animal 59Fe Activity. . . 76 DISCUSSION 0 O O O O O O O O O O O O O O O O O O O 80 Future Research . . . . . . . . . . . . . . . . 94 CONCLUSIONS 0 O O O O O O O 9 O O Q 0 O O O O O O O 98 APPENDICES . O O O O O C O O C O O O O O O O O O O 100 A. PHYSIOLOGICAL SALINES. . . . . . . . . . . . 100 B. PLASMA IRON DETERMINATION. . . . . . . . . . 101 C. UNSATURATED IRON BINDING CAPACITY. . . . . . 103 D. TISSUE IRON ANALYSIS . . . . . . . . . . . . 105 59 E. PERCENT INITIAL INJECTED Fe, TOTAL IRON CONCENTRATION, AND SPECIFIC ACTIVITY OF TISSUE AND BLOOD . . . . . . . . . . . . . . 107 LITERATURE CITED . . . . . . . . . . . . . . . . . 112 iv TABLE LIST OF TABLES Tissue blood volumes and tissue wet weight/ body weight ratios for 250-300 g rainbow trout Hematocrit, hemoglobin, and percent reticulo- cytes of blood samples from control and experi— mental fish. . . . . . . . . . . . . . . . . . Tissue and plasma mean specific activity ranked from highest to lowest on each sampling day of the study . . . . . . . . . . . . . . . Plasma iron, UIBC and TIBC for experimental and control fish . . . . . . . . . . . . . . . Mean, standard error, and number of observa- tions (X i_SE(N)) of percent initial injected 59Fe, total iron concentration, and specific activity of tissue, plasma, and whole blood of control and bled fish on days l,2,4,8,ll 16, 23, and 30 following i.p. injection of SéFe. Page 46 48 50 54 107 LIST OF FIGURES FIGURE 1. Flow diagram of iron metabolism in mammals showing the distribution of iron to various pools following absorption from the lumen of the intestine. . . . . . . . . . . . . . . . 2. Urine collection apparatus . . . . . . . . . 3. Percent initial injected 59Fe (A), total iron concentration (B), and specific activity (C) of plasma from experimental and control trout on days l,2,4,8,ll,l6,23, and 30 following i.p. injection of 59Fe . . . . . . . . . . . 4. Percent initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the head kidney of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 59Fe . . . . . . . . . 5. Percent initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the livers of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 59Fe . . . . . . . . . 6. Percent of initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the spleens of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 5 Fe . . . . . . . . . 7. Percent initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the pyloric caeca of experimental and con- trol trout on days l,2,4,8,ll,l6,23, and 30 following i.p. injection of 59Fe . . . . . . vi Page 43 52 57 60 63 65 LIST OF FIGURES--continued 8. 10. ll. 12. Percent initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the intestine of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 5 Fe. . . . . . . . . . Percent initial 59Fe injected (A), total iron concentration (B), and specific activity (C) of the muscle of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 5 Fe. . . . . . . . . . Bile volume (A), and percent of initial in- jected 59Fe in 1 ml bile (B), and in fecal material (C) from control and experimental trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 5 Fe. . . . . . . . . . Percent initial injected 59Fe in whole blood (A) and in the whole animal (B) for control and experimental trout on days 1, 2,4, 8,11, ESP ,23, and 30 following i. p. injection of Clearance of 59Fe from the plasma of trout following an i.p. injection of this isotOpe . vii Page 69 72 75 78 83 INTRODUCTION AND LITERATURE REVIEW There is probably more information known about the metabolism of iron than any other atom or molecule which composes living organisms (Garby and Vuille, 1967). Iron is an essential element involved in the physiological functions of oxygen transport and cellular respiration, and it is present in hemoglobin, myoglobin and cytochromes. From measurements made on human subjects and rats, an over- whelming volume of literature has resulted which deals with iron absorption, the mode of iron transport via the plasma, the utilization and conservation of iron, storage, and excretion. A normal, adult, 70 kg man contains 3-5 9 of iron, 55% of which is in the form of hemoglobin, and 10-20% as myoglobin (Moore, 1958). The remainder is stored in the liver, spleen, kidneys and bone marrow, or is in transit within the plasma. Approximately 0.6-1.5 mg of iron is absorbed daily from the diet via the duodenum and upper jejunum, and 0.5-1.0 mg is excreted. Iron is recycled within the body from destroyed red blood cells and returned via the bone marrow as new red cells with very little iron excreted in the urine and feces. Since the details of iron metabolism in lower vertebrates are largely unknown a basic outline of iron metabolism in mammals (Figure 1) will be used as a model for comparison with the iron cycle as it exists in rainbow trout. The fate of inorganic iron will be traced from ingestion to absorption and eventual in- corporation into the hemoglobin of red blood cells, and/or storage as ferritin or hemosiderin in the liver and spleen. Control of Iron Absorption Luminal Factors Controlling Iron Absorption Iron exists in the ferric and ferrous states in in- gested food, but it is the ferrous iron which is more easily absorbed (Figure 1). One of the most widely studied and effective iron reducing agents is ascorbic acid. This and other organic acids including succinic, lactic, pyruvic, and citric enhance the absorption of both organic and in- organic iron by acting not only as reducing agents, but also as iron chelators (Van Campen, 1974; Dowdle gt al., 1960; Jacobs et al., 1966). Although there is no complete agreement as to the importance of simple sugars such as fructose, lactose, sucrose, and glucose in promoting iron absorption and retention, a majority of the reports speak of the advantages these simple carbohydrates offer. All ten of the essential amino acids are effective in increas- ing iron uptake. Figure 1. Flow diagram of iron metabolism in mammals show- ing the distribution of iron to various pools following absorption from the lumen of the intestine. Ferric ions are attached to trans- ferrin in the plasma after iron crosses the mucosal cell. Transferrin delivers iron to the liver and spleen, the main storage organs, which contain ferritin and hemosiderin, two iron stor- age proteins. Iron is also transferred to the sites of red blood cell production which are mammalian bone marrow and fish head kidney. The majority of iron is in the form of hemoglobin as Fe+2. Following hemoglobin catabolism by the reticulo-endothelial system, iron is transferred to storage in the liver or spleen, or returned to the sites of red blood cell production. Less than 0.5% of the total iron is lost to the urine or feces. INTESTINAL LUMEN ++ Fe ’\ MUCOSAL CELL J— FERRITIN-Fe+++ HEMOSIDERIN-Fer++ A +++ ‘?/////,RPLASNA TRANiFERRIN- e STORAGE SITES:-—4~SITES 0F RED BLOOD CELL PRODUCTION (ERYTHROPOIESIS) E.G. LIVER HEAD KIDNEY - E.G. TROUT SPLEEN BONE MARRON - E.G. MAMMALS FERRITIN—Fe+++ HEMOSIDERIN-Fe+++ V HEMOGLOBIN - Fe++ TRANSFERRIN TRANSFERRIN +++ +++ Fe~ Fe W HEMOGLOBIN CATABOLISM +++ Fe l LOSS IN URINE, FECES Figure l Among the factors which depress the absorption of iron from the lumen of the intestine are phosphates and phytates. When complexed with iron these compounds reduce and almost completely block iron absorption. Callender (1964) dis- cusses two other circumstances which result in depressed iron absorption; achlorhydria and secretion of pancreatic juice. Achlorhydria is associated with inadequate HCl pro- duction, which leads to incomplete breakdown of foodstuffs and decreased liberation of simple iron salts. Iron remains bound to complex organic molecules which are not absorbed but are lost with the feces. Pancreatic juice may serve more as an intraluminal regulator of excessive iron absorp- tion, and therefore play a beneficial role. The action of pancreatic juice in decreasing iron absorption is not en- tirely clear, but it is believed that the bicarbonate con- tent of the secretion is the factor which complexes with iron and decreases its availability for absorption (Van Campen, 1974). Mucosal Cell Iron Absorption Primary control mechanisms for iron absorption by in- testinal mucosal cells are unknown, as are the mechanisms for transfer of information to the control system (Van Campen, 1974). There are several theories regarding mucosal cell iron absorption but there is still lack of agreement as to which is correct. One of the earliest and most debated theories for con- trol of mucosal iron absorption was the "mucosal block theory" proposed by Hahn et_al, (1943) which was later sup- ported by Granick (1946). The theory essentially states that intestinal mucosal cells contain the primary mechanisms for control of iron absorption, and that the absorptive process may be blocked by the saturation of the absorptive mechanism with iron. Granick (1946) maintained that the build—up of ferritin in the mucosal cell blocked iron ab- sOrption. The theory that the concentration of ferritin in the mucosal cell is the controlling factor in iron ab- sorption was later supported by Conrad and Crosby (1963) and Crosby (1965), but with some modification of the theory proposed by Granick. They maintained that iron overload in the diet could overwhelm the absorptive mechanism and result in increased iron absorption. But in support of the mucosal block theory, Conrad and Crosby nOted that iron deficient animals had a greater absorptive capacity than animals with abnormally high concentrations of iron in the plasma. Crosby (1965) hypothesized that the mucosal cell "ferritin apparatus" is responsive to plasma iron concentration, thus providing a way of relating to the intestinal epithelial cells the body iron status. Radioiron given intravenously (i.v.) to an iron deficient animal does not appear in the villous epithelial cells, and iron given by mouth is readily absorbed. However, radioiron administered i.v. to - iron-loaded patients, with higher than normal plasma iron concentrations, appeared in the intestinal epithelial cells via the "back door" with the result that orally administered radioiron was not absorbed. It is believed that the radio- iron in the villous epithelial cells is incorporated into ferritin, thus becoming part of the ferritin apparatus. Crosby further states that ferritin may capture an intracel- lular iron-carrier molecule which is responsible for move- ment of iron from the lumen to the serosal surface of the musocal cell. In this manner a small amount of ferritin could block the absorption of a large amount of ingested iron. The mucosal block theory is rejected by Schade (1972)‘ who found that a decrease in intestinal iron uptake was not correlated with an increase in intestinal ferritin. Schade induced inflammation in a group of rats by giving an intra- muscular injection of turpentine which resulted in a reduc- tion in the intestinal absorption of iron. By measuring intestinal ferritin and 59Fe activity following oral admin- istration of radioiron, Schade found that there was little or no decrease in ability of the intestinal epithelial cells to absorb iron in the group of rats with inflammation, but there was a significant decrease in the transfer of iron from the intestine to the body. Although there was an in- crease in intestinal iron due to the reduction in iron transport, there was no increase in intestinal ferritin content. Schade attributes any increase in intestinal ferritin, which other researchers have found, to intestinal build-up of iron as a result of disruption of iron trans- port. In other words, intestinal ferritin concentration is not the primary controlling factor in the transport mechan- ism, but is secondarily a result of iron build—up. Schade offers no information, nor does he speculate on the nature of the primary controlling factor in iron absorption. Active Transport and Enzymatic Control Active transport and enzymatically controlled reactions which are described by Michaelis-Menten kinetics have been hypothesized as having importance in the iron absorption process (Manis and Schacter, 1964; Jacobs e£.al., 1966). It is now accepted that iron metabolism involves uptake by the mucosal epithelial cell, transport across the mucosal cell, and transport out of the mucosal cell. Manis (1973) presents evidence for an enzymatically controlled reaction + +2 to Fe 3 for the oxidation of Fe within the cell. He pro- poses that this ferroxidase reaction may be a means of regulating iron absorption since iron must be in the triva- lent form before entering the plasma or sequestered by the mucosal cell as part of the trivalent iron pool. The impor- tance of the mucosal cell trivalent iron pool (which may be ferritin) in the control of iron absorption has already been considered. Dowdle EE.El° (1960) demonstrated active transport of 59Fe against a concentration gradient using everted gut sacs. Dependence on oxidative metabolism for generation of phosphate bond energy, and demonstration of saturation kinetics all point to active transport in the absorption of iron by gut segments taken just posterior to the pyloris of rats. Non-Ferritin Intermediates Another hypothesis proclaims the importance of non- ferritin iron-protein intermediates in the absorption of iron. Halliday and Powell (1973) discovered the presence of three non-ferritin intermediates while studying the absorp- tion of iron by isolated rat mucosal cells. They described a two-component curve showing an initial, rapid phase of iron uptake in the first two minutes followed by a slower uptake over the next 30 minutes of exposure to medium of 59Fe ferric chloride and aScorbic acid. Following frac- tionation of the cells and elution, they discovered the three intermediates present during the initial rapid phase of incorporation. These intermediates may be iron-trans— porting molecules which carry iron across the cell and perhaps to the iron-free protein, apoferritin, the precursor of ferritin. The slower uptake period was dominated by the 10 presence of ferritin with diminished quantities of the intermediate iron-carrying proteins. Serosal Absorption of Iron Iron has been administered intramuscularly (i.m.), intravenously (i.v.), orally, and intraperitoneally (i.p.) with about the same results; namely, appearance of the majority of the iron in red blood cells. In an attempt to measure appearance of iron in intestinal ferritin following i.p. injection of iron—dextran, Thirayothin and Crosby (1962) found iron concentrated in the core of the intestinal villi just above the lamina propria. When iron is adminis— tered orally the concentration is greatest at the surface of the intestinal villi. The explanation for the appearance near the lamina propria is the sequestration of iron from the peritoneal cavity by phagocytes which enter the intes- tinal serosa by diapedesis. The iron engulfed by the phagocytes is incorporated into the villi epithelial cells and the fate of this iron is similar to that absorbed from the lumen of the gut. The iron-laden phagocytes may also cross the epithelial cell and enter the lumen of the gut to be excreted as excess iron (Crosby, 1965). The majority of iron administered i.p. is absorbed via the lymphatic system and delivered to the venous circu- lation via the thoracic duct (Thirayothin and Crosby, 1962). Iron entering the plasma is distributed to the body stores 11 and red blood cells in the same manner as iron administered orally, i.v., and i.m. Plasma Iron Iron is moved across the mucosal cell to the plasma pool in accordance with the body iron requirements. The transport into the plasma may involve an active process, but there is a lack of information regarding this step in the . . . . . + iron metabolism pathway. Free, ionic iron (Fe+2 and Fe 3 ) does not exist in the plasma; it is bound to an iron-carrier protein known as transferrin. The process of attachment of iron to the transferrin molecule probably occurs at special receptor sites on the vascular surface of the mucosal cell. There is little information available regarding the mechan- ics of transfer of iron from the mucosal cell to transferrin, but the process probably involves a redox reaction because mucosal-Fe+2 must be oxidized to Fe+3 (Figure l). The standard plasma iron concentration range in humans is 120 to 146 ug Fe/lOO ml (Ramsay, 1958). Average values for fish are 25 ug/lOO ml in carp (Field et_al., 1943) and 61 ug/lOO ml in the tench (Hevesy gt al., 1964), both well below the average iron concentration in human plasma. The plasma iron pool of humans represents about 0.1% of the total body iron, and the half-life of iron in plasma is quite short, about 90 minutes (Ramsay, 1958). This figure 12 was determined by measuring 59Fe removal from the plasma following i.v. injection. Most plasma iron is sequestered by the erythropoietic tissue and incorporated into hemo- globin by immature red blood cells. The daily turnover of plasma iron in humans is about 20-40 mg/day, which is approximately the amount required for daily hemoglobin syn- thesis. There is a slight diurnal variation in plasma iron concentration, being highest in early morning and lowest during the afternoon. The difference is usually no greater than 15 ug/lOO m1. Transferrin Structure and Chemical Nature Transferrin, the beta-globulin protein which serves as the vehicle for iron transport in the plasma, has a molecu- lar weight of about 86,000 and has two separate iron bind- ing sites/molecule, each capable of binding one atom of ferric iron (Fletcher and Huehns, 1968; Bearn and Parker, 1964). The transferrin concentration in human plasma (0.24- 0.28 g/100 ml) is such that the number of iron binding sites exceeds the amount of iron in the plasma; the transferrin pool is usually only 30-40% saturated with iron depending on the erythropoietic activity of the bone marrow and the overall iron status of the individual (Laurell, 1951). This figure represents the normal plasma iron content of 13 120-140 ug Fe/lOO ml in humans and 25-60 ug Fe/100 ml in fish. The total iron binding capacity is about 300-360 ug Fe/lOO m1 plasma, but only under pathological conditions (such as hemochromatosis) will the per cent saturation of the transferrin pool reach 100%. Total iron binding capacity usually does not change even though plasma iron concentration fluctuates diurnally. Abnormally high trans- ferrin levels, resulting in iron binding capacities above 400 ug/lOO ml, are usually an indication of chronic iron deficiency (Laurell, 1958). Subnormal transferrin levels (below 260 ug Fe/lOO ml) indicate impaired synthesis of plasma protein. Iron is very tightly bound to transferrin; at physio- logical pH the equilibrium constants for the two iron bind- ing sites are around 1030 (Fletcher and Huehns, 1968). Strong chelating agents such as ethylenediamine tetraacetic acid (EDTA) will not remove iron from transferrin. However, iron is easily removed from the transferrin at receptor sites in immature red blood cells and storage areas in the liver and spleen. Delivery of Iron to the Reticulopyte Attachment of iron-loaded transferrin molecules to special receptors on the reticulocyte and the subsequent transport of iron across the cell surface have been thoroughly studied (Jandl gt gt., 1959; Jandl and Katz, 14 1963; Katz, 1965). It is known that transferrin molecules saturated with ferric iron have a much greater affinity for the receptor sites on the reticulocyte cell membrane than transferrin lacking one or both atoms of iron (Jandl and Katz, 1963; Fletcher and Huehns, 1968). During transfer of one atom of ferric iron from transferrin to an immature red blood cell, the transferrin molecule remains at the sur« face of the cell for approximately one minute. Upon loss of one iron atom, the transferrin molecule is quickly rev placed by another transferrin molecule carrying the full complement of two atoms of ferric iron. After completing the transfer of iron, transferrin less one atom of iron circulates back to intestinal mucosal cells or to storage sites in the liver and spleen for reloading. Jandl and Katz (1963) were among the first to provide evidence for the presence of transferrin receptors on the surface of reticulocytes. Their experiments involved the 1311, 59Fe), the meta- use of doubly-labeled transferrin ( bolic inhibitors cyanide and dinitrophenylphenol, and enzymatic alteration of the cell membrane with trypsin. The 131I label was used to follow attachment of transferrin to the reticulocyte surface, while 59Fe was used to measure reticulocyte iron incorporation. They found that alteration of the cell surface with trypsin prevented attachment of transferrin and, subsequently, blocked iron absorption. Dilute concentrations of the metabolic inhibitors did not 15 block attachment of transferrin to the cell surface, but they did prevent transfer of iron across the cell membrane. In an excellent summary of these experiments, Katz and Jandl (1964) conclude that transferrin attaches to a speci— fic, allosteric receptor site on the reticulocyte cell surface and that transfer of iron to the cell is an active process. Mature red blood cells do not retain these re- ceptors, therefore they are unable to incorporate iron carried by the transferrin molecule, and contain functional hemoglobin with iron which cannot be exchanged with that of other intracellular or extracellular iron pools. Fielding gt gt. (1969) suppOrt the findings of Katz and Jandl (1964) and have demonstrated the importance of sulfhydryl radicals in the incorporation of iron from trans- ferrin. The addition of the sulfhydryl inhibitor, p-hydroxy- mercuribenzoate (PMB) which acts specifically at the cell surface, to a medium containing 59Fe-transferrin and reticu- locytes resulted in a significant decrease in reticulocyte iron absorption. They concluded that sulfhydryl groups in the cell membrane are involved in the process of reticulocyte iron uptake, but they were unsure of whether the sulfhydryl groups were involved with attachment of transferrin to specific receptor sites, or with the dissociation of the transferrin-iron complex after adhesion to the cell surface, or both. Later, Edwards and Fielding (1971) reported that dilute concentrations of PMB did not affect transferrin l6 attachment to the reticulocyte surface, but it was effec- tive in reducing iron uptake by the reticulocyte. They concluded that sulfhydryl groups are active in dissociation of iron from its bond with transferrin after transferrin has attached to the allosteric receptor site. In summary, transferrin is a beta-globulin of molecu- lar weight 86,000 which serves as the iron-carrier protein in plasma and is responsible for the transport of ferric iron from the intestinal mucosa to iron storage areas in the liver and spleen, and to the erythropoietic tissues. Iron is transferred to immature red blood cells for incor- poration into hemoglobin. The transfer process involves attachment of the transferrin molecule to allosteric recep- tor sites On the reticulocyte surface and the removal of one atom of ferric iron which is transported across the surface of the cell by active means. Sulfhydryl radicals play an important role in the dissociation of iron from transferrin and absorption by the reticulocyte. The whole process is completed within one minute, and the transferrin molecule, minus one iron atom, is replaced by another mole- cule with two iron atoms. Transferrin reenters the circu- lation and repeats the cycle with attachment to the receptor sites in the intestine. l7 Etythropoiesis Erythropoietic Tissue In mammals, birds, and reptiles the bone marrow is the main erythropoietic (red blood cell producing) organ. Generally speaking, fish lack sufficient bone marrow for erythrocyte production and, therefore, must rely on some other tissue. The kidney, and particularly the head kidney in rainbow trout (Figure l), is the main organ or erythro- poiesis in teleost fish (Topf, 1953). The renal intertubu- 1ar tissue contains lymphoid tissue which is the origin of the small lymphoid hemoblasts, the progenetors of the erythrocyte line (Catton, 1957; Klontz gt gt., 1969). Head kidney tissue appears microscopically similar to red bone marrow, and the spongelike structure of the tissue re- sults in a complex blood flow pattern through the interstices between the tubules (Catton, 1957). New red cells enter the circulation in the head kidney, along with a few immature red cells or reticulocytes. The number of immature red cells in the peripheral circulation varies with the erythropoietic activity and immediate need for additional red cells. The erythropoietic activity is in turn controlled by various external and internal stimuli. Erythropoietic Stimulants Stimulants of erythropoiesis include bleeding, hypoxia, injection of phenylhydrazine, and an increase in serum 18 calcium. Among these, bleeding is the most effective and widely investigated stimulant. In experiments with turtles, Altland and Thompson (1958) found high reticulocyte counts (10-20% of red cell population) which lasted for at least 21 days after a series of bleedings. Hirschfeld and Gordon (1965b) noted an increase in reticulocyte numbers and erythroblast 59Fe incorporation 14 days following removal of 20% of the blood volume of turtles. Rosse gt gt. (1963) report similar findings in frogs after removal of 30% of the blood volume and also in birds after removal of 25% of the blood volume (Rosse and Waldman, 1966), using thymidine-2-14C incorporation by peripheral immature red blood cells as an indicator of erythropoietic activity. Zanjani gt gt (1969) used bleeding as a means of stimulating erythropoiesis in fish, and Finch gt gt. (1959) employed bleeding as the technique for the study of kinetics of erythropoiesis in rabbits. In summary, bleeding has been reported as an effec- tive means of increasing red blood cell production in nearly all vertebrates that rely on hemoglobin or oxygen transport. Hypoxia, an effective erythropoietic stimulant in homeotherms such as rats and birds, does not appear to affect red cell production in poikilotherms. Altland and Parker (1955) artificially decreased the P02 of turtle blood, and Observed no increase in hematocrit or hemoglobin concentra- tion. Rosse gt gt. (1963) reported a similar result in frogs exposed to hypoxic conditions, and they speculate that 19 the increase in erythropoietic activity due to bleeding must be mediated by a mechanism different from the response to hypoxia. The only positive erythropoietic response, increased hemoglobin concentrations and red cell counts, to lowered ambient P0 was found in goldfish which were acclimated to low Oxygen concentrations for several days (Prosser gt_gt,, 1957). This resultant increase in oxygen carrying capacity may be a reason why goldfish, and other members of the carp family, are able to live in oxygenvpoor water. Since trout require fast—flowing, highlyeoxygenated water for survival, it is doubtful that such a response to hypoxia (i.e., a significant increase in hemoglobin concen« tration) occurs in these teleosts (Dawson, 1933). Drugs, such as phenylhydrazine (which causes hemolysis of red cells), have been used to artificially stimulate erythrOpoiesis in several species, including fish. Finch gt_gt. (1959) compared the erythropoietic effects of bleed- ing and phenylhydrazine injection (i.p.) in rabbits, and found relatively little difference in the erythropoietic response to both stimulants. Tambourin gt_gt, (1973) found phenylhydrazine injections were effective in producing high plasma levels of erythropoietin (see next section) in mice. Smith gt gt. (1971) report that a single injection of phenyl— hydrazine (12.5 Ug/g body weight) into Chinook salmon pro— duced severe anemia within 10 days, at which time hematocrits and hemoglobin concentrations had been reduced to 1—5% of 20 normal. Red cell hemolysis was indicated by the appearance Of cell-free hemoglobin in the fin rays and by histological changes in liver, spleen, and kidney tissue. However, blood parameters returned to normal levels 95 days after the injection, indicating an increase in erythropoietic activity following the induced anemia. Serum calcium levels may also affect erythropoiesis. Such is the case in rats as demonstrated by Perris and Whitfield (1971). After calcium chloride or parathyroid hormone injection they found an increase in the mitotic activity of the bone marrow leading to increased reticulo- cyte production and 59Fe incorporation by immature cells in the peripheral blood. Following parathyroidectomy, there was a decrease in plasma calcium and erythropoiesis, but this situation could be reversed if calcium chloride or parathyroid hormone were administered. In nephrectom- ized rats, which lack the ability to produce erythropoietin, administration of calcium and/or parathyroid hormone, stimulated mitosis in bone marrow erythropoietic cells without secondarily initiating erythropoietin release. Perris and Whitfield speculate that there may be two mechan- isms involved in the control of erythropoiesis in the rat. Calcium may serve in the regulation of non—specific mitosis in the bone marrow cells, while erythropoietin may have a highly specific differentiating action in bone marrow which results in increased erythrocyte production. 21 Erythropoietic Depressants The most common depressants of erythropoiesis are starvation and hyperoxia. In turtles starvation lowers plasma iron to as much as 1/3 normal concentration, but otherwise the peripheral hemograms of starved and normal fed turtles are about the same, i.e., hematocrit, hemo- globin and red cell counts are very similar (Hirschfeld and Gordon, 1965b). Starvation for 6 weeks prior to bleed- ing prevented an increase in erythroblast production and, thus, depressed the erythropoiesis which was seen in normal fed turtles subjected to the same bleeding schedule. Thus, the nutritional status of the animal may exert a consider- able effect on erythropoietic activity, and what may appear to be a normal fed animal (according to hematocrit and hemoglobin) may actually be erythropoietically impaired. Zanjani gt gt. (1969) found that starvation significantly reduced erythropoiesis in fish, but, unlike the turtle, the hemoglobin, hematocrit and red cell counts were also significantly reduced. Even though the effects of starva- tion appear to depress erythropoiesis in turtles subjected to bleeding, the nutritional status of starved blue gourami (Zanjani gt_gt., 1969) did not seem to hamper increased red blood cell production in response to bleeding. It is obvious from the above discussion that the nutritional status of the experimental animals should be considered in investigations of erythropoiesis. The inhibition of erythropoiesis by hyperoxia was demonstrated by Fletcher gt_gt. (1973). Using parabiotic rats, they found that induced hyperoxia in one partner by exposure to 95-100% oxygen resulted in depression of normal red blood cell production in the other partner. They attributed the depression to a humoral factor produced in the hyperoxic partner which inhibits erythropoiesis. It has been speculated that erythropoiesis may be controlled by a set of hormones which induce erythropoietic activity during hypoxia and depress erythropoiesis during hyperoxia. Etythropoietic Stimulating Factor and Renal Erythrppoietic Factor It now appears that erythropoiesis is under the con— trol of a humoral factor, erythropoietin or erythropoietic stimulating factor (ESF), which acts as a stimulant to red blood cell production. ESF is released into the plasma in response to anemia, hypoxia, hemorrhage and other erythro- poietic stimuli. It acts at the bone marrow or head kidney to increase both the rate of red blood cell production and their release into the peripheral circulation. Gordon (1959) has written an excellent review of erythropoietin which includes 248 references concerning the origin, physical characteristics, and actions of this hormone. Erythropoietin (a glyCOprotein with a molecular weight of about 60,000) is formed in the plasma from a combination of liver globulin and a substance secreted from the kidneys, 23 renal erythropoietic factor (REF). Zanjani gt_gt, (1967) have shown that REF extracts from kidneys produces ESF when added to dialyzed rat serum and they presented evidence to show that REF is not related to angiotensin or renin. Little is known about the structure or mechanism of action of REF other than it may be activated by renal cyclic-AMP through a protein kinase (Rodgers gt gt., 1974). Erythropoietin appears to be species specific. Mammalian ESF will not increase erythropoiesis in birds, reptiles or fish, nor are any ESF fractions from these lower vertebrates effective in mammals. Zanjani gt gt. (1969) experimented with ovine ESP and found that it had no erythro- poietic effects in fish, but anemic fish plasma did stimu- late red blood cell production when injected in fish of the same species. Immature Red Blood Cell Iron Uptatg The number of immature red blood cells in the peripheral circulation is normally quite small (1% red cell population in mammals). But, after subjection to bleeding, hypoxia or any erythropoietic stimulant, there is a substantial in- crease in the peripheral immature red cell population. This increase initiates mobilization of storage iron for hemo- globin synthesis. The reticuloendothelial (RE) cells in the liver and spleen, which remove old or damaged red blood 24 cells from circulation, store iron sequestered from the catabolism of hemoglobin. Release of this iron from RE cells is regulated by the erythropoietic activity. Using heat-denatured erythrocyctes containing 59Fe hemoglobin, the movement of iron from the RB cells to the immature red blood cells in the peripheral circulation has been traced (Lipschitz gt gt., 1971). Following i.p. administration of 59Fe there is a direct correlation between the number of immature red blood cells in the peripheral circulation and the 59Fe activity of the blood (Walsh gt gt., 1949; Jensen gt gt., 1953). As stated previously, iron is delivered to the imma- ture red blood cell directly from transferrin at specific receptor sites on the cell membrane. There may be a second means of iron incorporation by immature red blood cells. For example, in animals with erythropoietically active bone marrow, iron (as ferritin) may be pinocytized by erythro- blasts from RE cells in the bone marrow. Whether this is also the case in the fish head kidney is not known. Regardless of the means of Obtaining iron, it is stored in the immature red blood cells as ferritin and eventually as ferruginous micelles in mitochondria (Bessis and Breton- Gorius, 1959; Kaplan gt gt., 1954). During hemoglobin formation the mitochondria release the ferruginous micelles to the cytoplasm and iron is then incorporated into the hemoglobin molecule. The ferritin clusters and iron-laden 25 mitochondria disappear from the cytoplasm upon completion of hemoglobin formation. Red Cell Life Span Hevesy gt gt. (1964) injected tench i.m. with glycine- 2-l4c and found the red blood cell life span to be 150-200 days. There is evidence that the red cell life span fluc- tuates with changing environmental parameters such as atmospheric pressure and temperature, and with the animals' metabolic rate. Mice red blood cells formed during exposure to 0.5 atmospheres had a shorter half-life (39.3 days) than those of control mice (46.9 days) (Abbrecht and Littell, 1972). Although hypoxia due to decreased atmospheric pres- sure apparently has an effect on mammalian red cell life span, it is doubtful that hypoxic conditions affect poikilo- term red cell longevity since no erythropoietic response to lowered environmental PO has been observed in these 2 animals. However, temperature and metabolic rate are both considered factors which influence pOlkilotherm red cell life span. Altland and Brace (1962) found the red cell life Span to be between 600 and 800 days in turtles and 700-1,400 days in toads. The long life span is attributed to the low metabolic rate of these animals. There is evidence that environmental temperature affects red cell life span in frogs (Cline and Waldmann, 1962). At 24-26°C the life span is approximately 200 days, but this value is 26 increased significantly in frogs exposed to 4°C. Perhaps the poikilothermic frogs are less reliant on hemoglobin for oxygen transport at such low temperatures. The arctic ice fish, for example, does not rely at all on red blood cells and hemoglobin for oxygen transport. This fish lives in water that is just above the freezing point, and under such conditions the metabolic rate is low enough that the amount of oxygen transported in the dissolved state in the plasma is sufficient to meet the oxygen requirements of the tissues. There is currently no information avail- able concerning changes in ambient temperature and meta- bolic rate as they affect red cells in rainbow trout. In summary, the main erythropoietic organ in trout appears to be the head kidney. Erythropoiesis is controlled by the plasma ESF concentration which is increased in re- sponse to blood loss, or hemolytic anemia. Hypoxia does not appear to be an ESF stimulant in poikilotherms; however, there is some indication that members of the carp family respond to hypoxia with an increase in circulating red blood cells. Starvation, which is an effective erythropoietic depressant in turtles, may not hamper erythropoiesis in fish as evidenced by the increased number of 59Fe incorpor- ating immature red blood cells in the peripheral circulation of the blue gourami following bleeding. However, the nutritional states of the animal should be considered in any study of erythropoiesis. 27 Iron is mobilized from the storage areas, mainly in the liver and spleen, and transported to immature red blood cells for eventual incorporation into hemoglobin. Prior to hemoglobin formation in immature red cells, iron is Stored as ferritin in the cytoplasm and mitochondria. The concentration of ferritin gradually decreases as hemo— globin formation nears completion. ' The life span of red blood cells in fish has been measured and is estimated to be at least 150 days. Red cell survival is subject to changes in ambient temperature and the animal's metabolic rate. Although hypoxia appears to result in decreased life span of newly formed red cells in mammals, there is no evidence of shortened red cell survival in poikilotherms subjected to hypoxia. Storage Iron Iron is transported to depots from the intestine via transferrin or is sequestered from old and dying red cells. The liver and spleen are the main storage organs which con— tain the iron storage proteins ferritin and hemosiderin (Figure 1). Other organs and tissues such as the intestine, kidney and skeletal muscle contain trace amounts of ferritin and hemosiderin. Hemoglobin, the main source for storage iron, is sequestered by RB cells and recycled through the storage iron pool befofe‘being reutilized (Sanchez-Medal 28 gt gt., 1970; Cook gt_gt,, 1974). Time in storage and quantity of iron stored vary considerably with the iron status of the animal. Anemia due to blood loss severely reduces the iron storage by stimulating erythropoietic activity. Iron overload (e.g., hemochromatosis) results in a significant increase in the iron content of the spleen and liver. The following discussion is concerned with ferritin and hemosiderin, and the storage of iron in the liver and spleen. Ferritin and Hemosiderin Ferritin, the most common form of stored iron, is a water-soluble crystalline protein with a molecular weight of 860,000. It has two components, an iron—free protein, apoferritin, and micelles of a colloidal hydrated iron oxide-phosphate complex (Harrison, 1964). About 17-23% of the dry weight of ferritin is iron. Apoferritin exists in the intestinal mucosal cells and parenchymal and RB cells of the liver and spleen. During absorption and storage, ferric ions are attached to the apoferritin molecule to form the crystalline structure, ferritin. Reduction of the ferric iron of ferritin to the ferrous state by biological reducing agents results in the separation of the iron from apoferritin. Hemosiderin is composed of protein and non—protein organic constituents of variable nature and quantity 29 (Harrison, 1964). The structure is ill—defined, but hemo- siderin is known to be rich in ferric hydroxide and ferric phosphates. Iron composes abOut 8 to 45% of the weight of hemosiderin, depending on the nature of organic constitu- ents. It has been suggested that hemosiderin may contain ferritin and apoferritin molecules. Whatever the composi— tion, hemosiderin serves as a secondary iron storage com— pound, which normally appears in small amounts in liver and spleen. Liver and Spleen as Iron Storage Organs The cell types containing high iron concentrations are the RE cells of the liver and spleen, and the liver paren- chymal cells. Liver ferritin is chiefly stored in the parenchymal cells (hepatocytes) and some ferritin and hemo- siderin is present in the RE cells (Kupffer cells). Paren— chymal ferritin is formed almost exclusively from iron delivered by plasma transferrin, while Kupffer cells sequester iron from the catabolism of senescent red blood cells (Cook EE.El-r 1972). Some 93% of liver iron is stored in parenchymal cells and 7% in Kupffer cells (Cook gt_gt., 1974). In hemolytic anemia and other diseases resulting in iron overload, the excess iron is quickly put into storage as hemosiderin in the Kupffer cells (VanWyk gt_gt., 1971). This storage is easily verified in tissue sections of liver in which hemosiderin is discernible as numerous blue 30 clusters following Perl's reaction for hemosiderin iron (Pearse, 1961). Iron is slowly removed from hemosiderin and placed in storage as ferritin to re-establish the normal ferritin:hemosiderin ratio in the liver. The spleen, which maintains the largest iron concen- tration of any organ, is mainly composed of RE cells which contain both ferritin and hemosiderin. The ferritin con- tent is relatively constant but hemosiderin content, like that of the liver Kupffer cells, varies in relation to the iron status of the animal. Hemosiderin has been detected in fish spleens by Grover (1968), and Yu gt gt. (1971) who noted a direct relation between spleen hemosiderin con- centration and erythropoietic activity. After subjecting blue gourami to bleeding Yu gt gt. (1971) discovered a reduction in the hemosiderin content of the spleen which they attributed to mobilization of storage iron for in- creased hemoglobin synthesis in immature red blood cells. Iron Storage in Intgstine, Muscle and Kidney Trace amounts of iron are detectable as ferritin and hemosiderin in the intestinal mucosa, skeletal muscle and kidney. Appearance of iron in these tissues is most noticeable during iron overload following hemolysis, or injection of high concentrations Of iron dextran. Intestinal mucosa, which is important in regulation of iron absorption, 31 is significant in terms of total iron storage under normal conditions. Summary Ferritin and hemosiderin are the two iron storage come pounds; ferritin being the more prevalent and stable of the two forms. Hemosiderin operates as the temporary iron storage form when excessive amounts of iron overload the body. The liver and spleen are the two main iron storage organs; the spleen having the highest iron concentration. Parenchymal cells in the liver contain ferritin and store almost all the iron found in the liver. The RE or Kupffer cells, of the liver contain some iron as hemosiderin and ferritin sequestered from methemoglobin of lysed red blood cells. The spleen contains many RE cells which destroy old red cells and concentrate iron as hemosiderin and ferritin. Other storage areas in the intestinal mucosa, skeletal muscle and kidney are relatively insignificant. RESEARCH RAT IONALE This project was originally designed to study the effects of various water-borne pollutants on fish erythro- cyte production and iron metabolism. A number of para- meters were chosen for investigation, including red blood cell 59Fe incorporation and tissue 59Fe distribution. However, there was very little basic information pertain- ing to iron metabolism in poikilotherms such as rainbow trout. Therefore, it was necessary to establish a baseline for iron metabolism in fish before using these animals in pollution experiments. The objectives of the study were: 1. To follow uptake and distribution of intraperi- toneally injected 59Fe in blood, plasma and tissues of rainbow trout, 2. To measure the total iron content of various tis- sues and establish their importance in iron storage, and 3. To evaluate the effects of iron deficiency induced by bleeding on: a) red blood cell production, b) the utilization of iron in storage, and c) the plasma iron and transferrin concentrations. 32 MATERIALS AND METHODS Experimental Animals Rainbow trout (Salmo gairdneri) weighing 250-300 g were obtained from Midwest Fish Farm Enterprises at Gladwin, Michigan. They were transported to Michigan State University in 80 gallon galvanized metal tanks lined with non-toxic paint. The tanks were contained within insulated boxes fitted with agitators for aeration. At the university fish were held in 120 gallon fiberglass tanks (Frigid Units, Toledo, Ohio) equipped with aeration stones and flowing water supply (2 liters/min) and held at 13 : 1°C under 14 hours of light and 10 hours darkness per day. The water was filtered through activated charcoal to remove particulate iron and chlorine. Experimental Protocol The fish were divided into an experimental and a con- trol group; each containing 50 animals. Both groups were fed a maintenance diet (100 g feed/10 kg fish) of EWOS salmon pellets (Astra-Ewos, SOdertélje, Sweden) three times weekly. The iron content of the food was 300-400 ug Fe/g 33 34 food. A total of 40% of the blood volume was removed from each experimental fish in four bleedings (10% blood volume/ bleeding) over a seven day period, allowing 48 hrs between bleedings. On the day of the last bleeding, both experi- mental and control fish received an intraperitoneal injec- tion of 0.9% saline solution (Appendix A) containing 1 uCi 59FeC13/100 9 fish (Amersham-Searle, Chicago, Ill.) and having a specific activity of 5 uCi/ug Fe. Six experimen- tal and six control fish were sacrificed on days 1, 2, 4, 59 8, ll, 16, 23 and 30 following the Fe injection. Bleeding Technique Fish were bled from the caudal vein at a point just posterior to the anal fin using a heparinized syringe fitted with a 1% inch long 21 gauge needle. The needle was inserted through the ventral surface into the hemal arch, and the plunger was pulled back slowly until the desired amount of blood was obtained. Tissue Sample After the removal of 4-5 ml of whole blood the fish was killed by a sharp blow to the head and weighed on a Torbal Torsion Balance (Torsion Balance Co., Clifton, N.J.). The following tissue samples were removed (the average 35 sample weights are in parentheses): liver (0.5 g), spleen, head kidney, and middle third of the kidney (0.3 g), muscle (0.5 g), pyloric caeca (0.5 g), and intestine (0.5 g). After removal of the intestinal contents, the gut was rinsed with 0.9% saline and a 1 cm length of midgut was taken as the intestine sample. The muscle tissue sample was removed from the flank just dorsal to the lateral line. Tissue sample wet weights were measured using a Roller- Smith Balance (Roller-Smith Co., Bethleham, Pa.) and intes- tinal contents were tare weighed on a Mettler Model B5 Balance (Mettler Instrument Corp., Princeton, N.J.). The tissues and intestinal contents were stored at 0°C in 12 x 75 mm glass culture tubes (Scientific Products, 59 McGraw Park, Ill.) for future total iron and/or Fe analyses. Bile Sample Before removing the liver, the gall bladder was located and a 1 m1 syringe fitted with a 25 gauge needle was used to withdraw as much of the bile as possible. The bile vol- ume was measured and placed in a glass culture tube for 59 .measurement of Fe activity. 36 Measuring;59Fe‘Activity 59Fe activity was determined using a Nuclear Chicago Model 1085 dual—channel gamma counter equipped with a Model 8725 scalar/analyzer and NaIeTl crystal (Des Plaines, Ill.). In order to select the two 59Fe photopeaks at 1.098 and 1.289 MEV, the base was set at 500 and the window at 250 with a coarse attenuator setting of 8 and fine attenue 137Cs standard calibration at base ator at 25 (based on the 610, window 100, coarse attenuator 4, fine attenuator 25). The high voltage settings were 900 coarse, 90 fine and the mode selector switch was on WIDE DIFF. These settings resulted in a detector efficiency of 5v10% of the actual dpm. To insure consistency in counting efficiency, radio« iron standards prepared from the stock 59FeCl were counted 3 with each group of tissue and blood samples. All samples were counted for l min and the activity was corrected for background radiation (10 cpm) and for isotope decay from day zero, the time of the i.p. injection of 59Fe. It was important to maintain constant geometry during counting. Therefore, a standard volume of 1 m1 of blood and plasma were counted, and tissue and fecal samples were packed near the bottom of the culture tubes to approximate the same volume. 37 HematocritL Hemoglobin and Reticulocyte Counts Blood was drawn into heparinized capillary tubes (Biological Research, St. Louis, Mo.) and centrifuged (IEC Model MB Microhematocrit Centrifuge, Needham Heights, Mass.) at 11,500 rpm for 2 min. The percentage of packed red blood cells was determined with an IEC Circular Micro- capillary Tube Reader. Hemoglobin determinations were made spectrOphoto- metrically by the cyanmethemoglobin method (Oser, 1965) using Hycel (Houston, Texas) standards and reagents and a Beckman Model DG-B spectrophotometer (Fullerton, Calif.). A standard curve of the optical density (OD) versus concen- tration (g%) was plotted from dilution of the prepared standard. Hemoglobin samples were prepared by adding 20 ul of whole blood to 5 ml of cyanmethemoglobin reagent. Hemoglobin concentrations in gram percent were calculated from equation 1: Concentration (unknown) = OD (unknown Concentration (standardT’ OD (standard) where gram percent anui corresponding OD from the standard curve were used for the concentration and OD of the standard. Reticulocyte (immature red blood cell) counts were made on blood smears prepared by a supra-vital staining technique reported by Lucas and Jamroz (1961). Equal volumes 38 of fresh whole blood and New Methylene Blue N (in 0.9% saline) were mixed and incubated for 20 min at 13°C. To make a blood smear, a small drop of this mixture was then placed on a clean glass slide and drawn across the surface quickly with the edge of a second glass slide applying very light pressure so as not to rupture the cells. The smears were allowed to air dry, and then were examined using the oil immersion objective of a microscope. Reticulocytes were distinguished from mature red cells by the blue staining RNA present in the cytoplasm. There was some difficulty in distinguishing mature from immature red cells because of the RNA remnants present in some mature cells. Therefore, cells were categorized as to their degree of staining. A section of the slide was selected where the cells were evenly spaced. Using an ocular grid a total of 1,000 cells (immature and mature) were counted. The number of reticulocytes was expressed as a percentage of this popu- lation. Plasma Iron, UIBC and TIBC Plasma iron was determined colorimetrically by the method of Plaut gt gt. (1972) as described in Appendix B. The results were expressed in ug Fe/100 ml plasma (ug%). Care was taken to prevent hemolysis during the centrifuga- tion of 3-4 ml of whole blood in an IEC Model Cl Clinical 39 Centrifuge (Needham Heights, Mass.). Plasma from any hemo- lyzed samples was discarded because of the error intro- duced by the high concentration of iron in the hemoglobin of lyzed red cells. The plasma total iron binding capacity (TIBC) was calculated from the plasma iron content and the unsaturated iron binding capacity (UIBC). The method of Kuypers and van Oers (1973) was used for the determination of UIBC of a 0.2 ml sample of plasma (Appendix C). The UIBC assay is based on the incubation of 59Fe-labeled ferric ammonium citrate with the 0.2 ml plasma sample and the incorporation of iron by unsaturated plasma transferrin. Excess iron was removed after incubation by addition of an anion ex- change resin strip (Gallard-Schlesinger Chemical, L.I., 59 N.Y.) and the total Fe activity remaining in the incuba— tion mixture was used in calculating the UIBC in ug%. Tissue Total Iron Anatysis Tissue iron was measured colorimetrically after wet- ashing with a sulfuric-nitric acid mixture (Appendix D). The ashing procedure was partially adapted from the Dow Chemical Company (1970) method for "Determination of Mercury in Fish" and from H. Bergman (personal communication). Tissue weighing from 0.1 to 0.5 g (wet weight) was dissolved in 5 parts concentrated sulfuric acid and one part 40 concentrated nitric acid at 90°C. After digestion for 2-3 hrs in this mixture potassium permanganate was added and the digestion was allowed to continue overnight at room temperature. The digestate was decolorized the following day with hydroxylamine hydrochloride and an aliquot was removed for iron analysis. The colorimetric procedure was the standard bathophenanthroline method for iron reported by Diehl and Smith (1965). This method was sensitive down to 0.04 ppm using the Beckman DB-G spectrophotometer at 540 nm. Blood VOlume 51 Total blood volume was determined by the Cr-RBC method of Conte 22.2l3 (1963). Whole blood from a donor 51Cr (10 uCi/ml blood) at 13°C on fish was incubated with a Scientific Industries (Springfield, Mass.) variable speed rotator set at 6 rpm for 24 hrs. The red blood cells were spun down, washed 3 times with phosphate buffered saline (PBS, Appendix A), and rediluted to the desired Hct (30-40%) with PBS. After measuring the specific activity (cpm/ml RBC), the 51 Cr-labeled RBCs were injected intracardially using a 27 or 28 gauge needle. The syringe was slowly rinsed twice with the blood of the recipient fish to insure delivery of the total aliquot of labeled blood. After 30 41 min a blood sample was drawn from the caudal vein and the specific activity was determined (cpm/ml whole blood). Blood volume was calculated by dividing the total injected activity by the specific activity of the recipient blood. Tissue blood volumes (ml/g) were determined by dividing the tissue sample 51Cr activity (cpm/g) by the specific activity of the whole blood (cpm/ml). Data for calculation of whole blood volume and tissue blood volume were obtained from fish of the same size as used in the iron experiment. The tissue blood volumes were used to correct tissue 59Fe activity and total iron concen- tration for the blood iron content. It was not possible to determine tissue blood volumes by the 51Cr method in the fish used in the iron experiments because of the time factor and the problem of discriminating 51Cr energy peaks from 59Fe. Urine Collection To determine urinary loss of iron fish were catheter— ized and placed in the apparatus seen in Figure 2. Fish were anesthetized with tricaine methanesulfonate (Finquelfip, Ayerst Laboratories, New York), and a PE 90 catheter with flared tip was sutured into the urogenital opening. The fish were then placed in 30 x 6.5 cm glass cylinders sealed with NO. 13 rubber stoppers. The catheter was fed through Figure 2. 42 Urine collection apparatus. Following cathe- terization of the urogenital opening with PE 90 tubing, trout were placed in four 30cm X 6.5cm. cylindrical glass fish chambers (D) connected to a water reservoir. Water was filtered through glass wool (A) before enter- ing the reservoir (B). A constant head pressure was maintained by an overflow tube (C). Aerated water passed from the reservoir by gravity flow to each fish chamber (D). The catheter was passed through a small hole in the rubber stopper and urine was collected in test tubes (E). After leaving the chamber, water was di- rected through a fecal trap before passing down the drain (F). 43 Figure 2 44 a small hole in the rear rubber stopper and connected to a collecting tube. Each cylinder received a constant flow (150 ml/min) of fresh, highly oxygenated water from a reservoir by gravity flow. Water entering the reservoir at 660-700 ml/min was filtered through a column of glass wool to remove particulate iron. An overflow line was connected near the top of the reservoir to keep the head pressure constant. The outflow tubes from each cylinder were connected to side-arm Erlenmeyer flasks where fecal material was collected for 59Fe analysis before the water passed down the drain. Inflow and outflow oxygen concen- trations (mg/l) were checked periodically to make sure the fish received an adequate O2 supply. Statistical Analyses Statistical analyses of the data were done on the M.S.U. Control Data 6500 Computer using a one-way layout analysis of variance for equal or unequal sample size. The means were compared for significant differences by the Student-Newman-Kuels test (Sokal and Rohlf, 1969 a and b) at a = 0.05. RESULTS General Health of Animals The trout used in this study were well-fed and appeared quite healthy as indicated by the amount of fecal material in the gut, the concentration of fat in the peritoneal cav- ity and the hematocrit and hemoglobin concentrations. All but one fish were sexually mature and there were 61 females and 33 males randomly sampled in the study. Tissue Data Interpretation Total blood volume of rainbow trout averaged 3.5 ml/ 100 9 fish as determined by dilution of 51Cr-labeled red blood cells. The blood volumes of the various tissues sampled in the study are given in Table l. Included in this table are the tissue wet weight/body weight ratios for the size trout used in this study. The tissue sample 59Fe content was corrected for tissue blood volume and expressed as percent of the initial injected 59Fe appearing in the total organ or tissue mass. Total iron concentration (corrected for blood) is expressed as ug Fe/g tissue, and specific activity is the concentration of 59Fe in ug/g tis- sue divided by the total iron content in one gram of 45 46 Table l.--Tissue blood volumes and tissue wet weight/body weight ratios for 250-300 g rainbow trout. Blood volumes were determined from dilution of 51Cr-labe1ed RBC injected intracardially.’ Tissue Blood Volume (ml whole blood/g) Tissue Wet Weight (9/100 g body wt.) Liver 0.150 1.103 Spleen 0.230 0.213 Head Kidney 0.250 0.200 Kidney 0.283 0.620 Pyloric Caeca 0.030 1.273 Muscle 0.006 88.746 Intestine 0.023 0.420 tissue, or: ug 59Fe pg Total Iron Standard errors are not included in any of the figures due to the resulting cluttered appearance. But the sample means, standard errors and number of observations for tissue 59 total iron, tissue and blood Fe content, and tissue speci- fic activity are listed by days in Appendix E. Hct, Hb and Reticulocytes The removal of 40% of the blood volume within a 7 day period had a significant effect on the hematocrit, hemo- globin and reticulocyte count of the experimental fish. Bleeding significantly depressed the red cell volume from 47 43 to 14% of the blood volume (Table 2) and reduced the hemoglobin concentration from 8.145 to 2.850g%. The first indication of a significant recovery in hematocrit was 16 days following the last bleeding, however, the hemoglobin concentration did not recover until 30 days after bleeding. The difference in recovery of hematocrit and hemoglobin indicates that the red blood cells lost by bleeding are replaced by immature red blood cells which lack complete hemoglobin molecules. The reticulocyte data show an in- crease in the percent reticulocytes from 2.99 at the last bleeding to 17.64% eleven days later. There is an increase in the per cent reticulocytes to 11.07% on day 8 but this value is not significantly higher than the reticulocyte count on preceding days. The presence of RNA in the cyto- plasm, as seen in the supravital stain, is indicative of hemoglobin formation in these cells as they undergo matura- tion. Recovery from bleeding was complete by day 30 as both hematocrit and hemoglobin were not significantly dif- ferent from the control and the percent reticulocytes of the bled fish had declined to 5%. The increase in hematocrit in the control fish between days 1 and 30 was statistically significant. However, there was no concomitant increase in hemoglobin or percent reticu- locytes within the 30 day time span. 48 .mo.o u a pm moconMMHo uCMOHMHcmHm ¥ .. iooaav.o m sam.h Aooom.a m oo.¢s Houuaoo Amvoom.o + ooo.m Amoams.o + omm.m Amoam.m + o~.mm Hmuamaflummxm om iovva.o m oov.o roomms.o m sma.a looms.~ m sm.mq Houuqoo «Amvmmo.~ + smo.m 4xmissm.o + sam.m .Amosm.~ + om.mm Hmucmeflummxm mm Amvsma.a m ovm.~ roomm¢.o m omo.» Ame~.~ m oo.wm Houucoo .Amvomo.~ + mam.oa «imvaoa.o + mmo.m Asimo.a + so.am Hmucmsaummxm ma Amvmmm.o m om~.~ lovams.o m sm~.m Amvm~.m m mm.mm Houuaoo .Amommm.m + o¢m.aa .Amvoam.o + kflm.m .imoms.a + sm.v~ HmpcmeHumaxm HH Amommm.o m mms.H 1035mm.o.m sas.m imvmm.a m mm.ov Houpcoo imvmms.m + oso.HH .Amomsv.o + ome.¢ .Amvmm.m + mm.om HanngAHmmxm m roomsm.o.m mmm.~ imvmmv.o m has.s isomo.m_m sm.os Houucoo Amooom.a + ooo.m *Amomfim.o + kflm.m .imvmm.a + mm.ma amazoeflummxm v Amines.o m omo.s Amvmm.m m 5H.Hm Houucoo .Amvmmm.o + omm.m .imvwm.fl + sH.vH Hmucmefinmmxm N Amvmms.o m.omm.v Amvmmo.o m nmo.s Amoqv.m m 5H.m~ Houpqoo .imvmms.o + cmm.m .imvmmm.o + msv.m .Amvsa.~ + mm.ma Hangmanummxm H cofiuoofidH umom whoa loavmsm.o.u Hmm.m iomvmmm.o.u m~.mm Hmucmsaummxm nus Aoaoom¢.o H cvs.N Aomoms~.o H amm.m AomommH.H H om.m~ Hmpcmeflummxm cum Aomvomm.o H mma.m Ammvoma.a H mv.s~ Hmucmsflummxm cam Ammvmmm.o_u mva.m immoamo.fl.u ma.mv Hangmaflummxm umH manummam Ausmoummv Aucmoumm Emumv Aucmoummv mmumooHsoflumm afiQOHmOEmm uwnooumfimm msouo hon .szmm‘H com: mum boondocH .mmmm mo cofluomflcfl .m.fl o» Hoanm swam Hmucwfiflnmmxm on» Eonm mmcflomwan mumummmm HDOm ca om>oEmH mm3 mESHo> vocab maon3 may no usmonwm huuom .amflm HmucmEHummxm can Hon» Icoo Eoum mmamamm UOOHQ mo moumooasoflumu poochom new .qflnoamoamn .ufiuooumEmmnu.~ magma 49 Comparison of Tissue Specific Activity Table 3 ranks the tissues and plasma in order from highest specific activity to lowest in experimental and control fish from day l to day 30.- Head kidney had the highest specific activity of any tissue during the first 4 days following the i.p. injection of 59Fe. Bled fish head kidney was about 4 times greater than the next highest tissue and was exceeded only by the plasma. Control fish head kidney was lk—Z times greater than the next highest tissue. After 4—8 days the head kidney fell and liver and spleen assumed the position of highest rank in control and bled fish. The pyloric caeca were intermediate in rank, followed by intestine and muscle which had the lowest speci— fic activity. It was interesting to note that even though the plasma specific activity decreased 4—8 fold within 8 days, it was still higher than any tissue throughout the experiment. The only exception was on day 4 when head kidney Specific activity was higher than plasma in bled fish. Plasma Plasma total iron concentration, percent initial in- jected dose, and specific activity over the 30 day study period are shown in Figure 3. Plasma total iron conCentra- tion averaged 50-60 pg Fe/lOO ml plasma. The apparent increase in total iron concentration from day l to day 30 50 mammam u m mcwummucH u H mHomsz n z Hmanm 006m u xx cmmamm u m Hm>HH u H mowmo n U 0H.0 s 00.0 2 0H.0 2 00.0 2 0H.0 2 00.0 s 00.H 2 00.0 2 0H.H H 00.H 0 00.H H 00.H H 00.0 0 00.0 H 00.0 H 00.0 0 H0.0 o 00.H H 00.0 o 0H.0 0 00.0 H 00.0 0 00.0 o 00.0 H 0H.0 mm 00.0 o 00.0 mm 0H.0 Hm H0.0 mm 00.0 o 0H.0H 0 00.0H o H0.0 m 00.0 mm 00.0 m 00.0 o 00.0 o 0H.00 H 00.00 H H0.0H H 00.0 H 00.0H H 00.0H H 0H.0H H 00.HH H 0H.00 am 00.00 mm 00.00 mm 00.00 0 00.00 m 00.00 m 00.00 m 00.00 0 0H.00 m 00.00H 0 00.000 0 .coo 00.0 2 00.0 H 00.0 H 00.0 2 0H.0 2 00.0 2 00.0 H 00.0 2 00.0 mm 00.0 2 00.0 2 00.0 H 00.0 H 00.0 H 00.0 o 00.H m 0H.0 H 0H.0 mm 00.H H 00.0 mm 00.0 o 00.0H o H0.0 2 00.0 H 0H.0 o H0.0 o 0H.0 o 0H.0 o 00.0 H HH.H0 0 00.00 0 00.0 H H0.H H 00.0 H 00.0 mm 00.0 H 0H.0 0 00.00 H 00.00 H H0.0 o 00.0 0 00.0 0 00.0 0 00.0 m 00.0H am 00.00 m 00.00H 00 00.00 mm 0H.00 0 00.0H m 0H.00 0 0H.00 m 00.00 m 00.0HH Hm 00.00H 0 00.000 0 .mxm 00 >00 00 000 0H 000 HH man 0 >00 0 >00 0 >00 H >00 msouo mIoH x >HH>0004 UHMHommm m .wosum map mo woo mcHHmEmm zoom .mMmm mo coHuommcH .Q.H on HoHHm mmcHUmmHn mumummmm Hoow CH amHm HmucmEHHmmxw may Eoum Um>oE ImH mm3 wESHo> UOOHQ may mo ucmoumm wuuom co ummsoa on ummann EOHM omxcmn >HH>Huom owmaowmm some MEmMHm 6cm mammHBI|.m canoe Figure 3.—*Percent initial injected of plasma from experimental and control trout and 30 follow- on days 1, ing i. exp., 2, 4, 51 8, 11 p. injection of $9 the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to i.p. injection of point represents the mean for 6 fish (day 30 n=5). Significant differences (P< 0.05) 59Fe (A), total iron concentration (B), and specific activity (C) 16, Fe. 23' Forty percent of 59Fe. Each among control and experimental means are listed There were no significant differences among the total iron concentration means. below. A Exp. Con. Exp. Con. Day Day Day Day Day 1 2 l FJH WL‘H. ‘K‘K‘fl. Days 4 Days Days Days Days 4 2 2 2 “‘ I I 8 8 4 4 4 I I I ll, 11, 8: 8. 8, 16, 23 11' ll, ll, l6, l6, 16' 30 23, 30 30 30 52 3O 3 2 u 6 _ A IN _ M D .. .. S H P. N. A 8 x 0 H E C 4 W. muo_ x >._._>_._.o< U_u.__om.n_m . O O u 3 3 . . . 3 3 A 2 B 2 6v. mm IA D in I .8 8 uuuuu 4. 4 u, \ .m P. d . I1 1 u 1 J < 1 2 5 5 O 0 O 0 3. m L m. 0. w 7 6 5 4 m 0 m awkomHi. ommm 45.52. 3 .x. <2m<4m .8 OO_\o.._ 0: DAY Figure 3 53 was not significant in either the control or experimental fish, nor were there significant differences between these two groups on any of the sample days. In contrast to total iron, the percent of initial injected 59Fe was maximal on day 1 for controls (3.2%) and experimentals (1.7%). For both groups it decreased significantly to 0.3% by day 4 before leveling off at 0.25% from day 8 to day 30. Speci- fic activity decreased 80-90% from day 1 to day 4 and then remained constant from days 4 to 30 in both experimental and control fish. There was no significant difference between experimental and control specific activity or per- cent of initial injected dose. -The half-time for initial clearance of 59Fe from plasma was 24 hrs. Plasma Iron, UIBC and TIBC Although plasma iron showed no significant change, UIBC and TIBC did increase around day 16 in both experi- Imantal and control groups and remained significantly higher than days 1-11 for the remainder of the study (Table 4). This indicates an increase in transferrin concentration in the plasma of both groups after day 11. Paired comparisons showed no significant differences between mean values of controls and experimentals on any given day. The values for TIBC were unusually high in these fish in comparison to human plasma TIBC (300-400 ug%) and values obtained in 54 A000.00_m 0.000 H000.00 m 0.000 A000.0 H 0.00 Houucoo A000.00 + 0.H00 i00H.00 + 0.0H0 i000.0H+ 0.00 HmpcmaHHmmxm 00 A000.00 m 0.000 H000.00 m 0.000 A000.0Hm 0.00 Houpsoo A000.00 + 0.000 H000.00 + 0.000 A000.0 + 0.00 HmpcmeHHmmxm 00 A000.H0 m 0.000 A000.00 m 0.0H0 A000.0 m 0.00 Houucoo A000.00 + 0.0H0 A000.00 + 0.000 A000.0 + 0.00 HmucmsHHmmxm 0H A000.00 m 0.000 H000.00 m 0.000 i00H.0 m 0.00 Houucoo H000.00 + 0.000 AO0H.00 + 0.0H0 A000.0 + 0.00 HmucmsHHmmxm HH A000.00 m_0.000 A00H.00.m 0.000 A000.0_m 0.00 Houpzoo A000.00 + 0.H00 H000.00 + 0.000 A000.0 + 0.00 HmucmeHHmmxm 0 H000.00 m 0.000 A000.00_m 0.000 A000.0 m 0.00 Houucoo A000.0H + H.000 “000.0H + 0.000 A000.0 + H.00 HmpcmsHumem 0 H000.H0 m 0.000 A000.00 m 0.000 i000.0 m 0.00 Houpaoo A000.00 + 0.000 A000.00 + 0.0H0 H000.0 + 0.00 HmucmeHHmmxm 0 A000.00 m 0.000 A000.00 m 0.000 A000.0 m 0.00 Houucoo A000.00 + 0.H00 H000.00 + 0.000 A000.0 + 0.00 HmacmeHummxm H coHuomhdH umom mmmo A0 00.0 m H.00 HmucmeHummxm 000 A0 00.00” 0.00 HmucmeHHmmxm 0H0 A0 00.0 H 0.00 HmucmaHHmmxH 000 A0H00.0 + 0.00 HmucoEHHmmxH 00H 00H0mmHm AHE 00H\00 000 HH8 00mem 000 AH5 00H\00 000 II omHH omHo aouH mammHm @0000 000 .mcmmE Hmuswe I IHHmmxm pom Houusoo cmmzumn mmocmeMMHo “CMOHchmHm o: me3 mnge .szmm + c002 on» mum omGSHocH .mmmm mo HOHuomncH .m.H on HOHHQ mmchmmHn mumummmm HDOM CH anm HmucmEHHmmxw on» EOHM ©m>OEmH mm: mEsHo> oooHn 00033 on» mo pamoumm apnom .cmHm Houucoo pom HousmEHHomxm How umHB can ume .COHH MEmMHmln.0 mHnms 55 an earlier study (250-300 09%). Human plasma, which was used as a standard, was within the range for normal human TIBC. Therefore, the high averages for fish plasma TIBC are probably not due to experimental error. Although there was no statistical difference in plasma iron concentration with time, there was a trend toward an increase which coincided with the significant increase in UIBC on days 16-30. This may indicate mobilization of storage iron be— tween days 11 and 30, which would correspond with the de- crease in tissue specific activity during the same period. Head Kidney and Kidney The head kidney had the highest specific activity of all tissues sampled between days 1 and 4 (Figure 4 and Table 3). This rapid uptake of 59 Fe by erythropoietic tis- sue in the head kidney was followed by an equally rapid reduction (80 to 90%) in radioiron content from day 2 to day 8. The radioiron content of head kidney and kidney tissue was below 2% of the initial injected dose for the remainder of the study. Except for the initial drop in iron concentration in the experimental head kidneys seen on days 2 and 4 the total iron content did not change significantly during the 30 day period. There were no significant differ- ences between head kidney total iron concentration in con- trols and experimental animals nor were there differences in 56 Figure 4.--Percent initial 59Fe injected (A), total iron concentration (B) and specific activity (C) of the head kidney of experimental and control trout on days 1, 2, 4, 8, ll 16, 23, and 30 following i.p. injection of é9Fe. Forty percent of the blood volume of the experimental fish was removed over a seven day period prior to injec- tion of 59Fe. Tissue total iron concentration, and 59Fe activity were corrected for tissue blood volume. Each point represents the mean for 6 fish (Day 30 exp., n=5). Significant dif- ferences (P<:0.05) among control or experimental means are listed below. A Exp. Day 1 # Days 2,4,8,ll,16,23,30 Day 2 # Days l,8,ll,l6,23,30 Day 4 # Days 1,8,11,16,23,30 Con. Day 1 # Days 8,11,16,23,30 Day 2 # Days 8,11,16,23,30 Day 4 # Days 8,11,16,23,30 B Exp. Day 1 # Days 2,4,16,23 C Exp. Day 1 # Days 2,4,11,16,23,30 Day 2 # Days l,4,8,ll,l6,23,30 Day 4 # Days 1,2,8,11,16,23,30 Con. Day 1 # Days 8,11,16,23,30 Day 2 # Days 4,8,11,16,23,30 as of INITIAL 59Fe INJECTED 57 HEAD KIDNEY A DAY i24 8 W DA n; 23 2&3 Y Figure 4 704 30‘ ES "l A ‘1- l6 23 30 DAY EXP. - ---~ CON. 8 o-——-o 58 percent of initial injected dose or specific activity. Total iron concentration and specific activity were not measured in the remaining kidney tissue. Liver As seen in Figure 5 there are some significant differ— ences between experimental and control fish liver 59Fe content, total iron concentration and specific activity. As early as day 11 the percent of initial injected 59Fe had decreased significantly in the experimental fish liver and by day 23 the total iron concentration of the experimental liver was lower than the control (Figure 5B). The specific activities for the experimental group were significantly different from control values at days 16-30. This indicates the importance of the liver as an iron storage depot, a reserve which was substantially depleted when erythropoiesis increased. The high specific activity and percent initial dose relative to other tissue such as caeca, intestine and muscle is an indication that absorbed iron is initially placed in storage in the liver as ferritin or hemosiderin. Total iron, percent initial dose, and specific activity re- mained fairly constant from days 1-30 in control fish. 59 Figure 5.—-Percent initial 59Fe injected (A), total iron concentration (B) and specific activity (C) of the livers of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 following i.p. injection of 59Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to the injection of 59Fe. Tissue total iron concentration and 59Fe activity were corrected for the tissue blood volume. Each point repre- sents the mean for 6 fish (Day 30 exp., n=5). In paired comparisons of experimental and control means, significant differences were found at ZP< 0.05 (*). Significant differences (P<<0.05) among experimental means are listed below. Control means were not significantly different. A Exp. Day 2 # Days 16,23,30 B Exp. Day 1 # Days 2,4,8,ll,l6,23,30 C Exp. Day 2 # Days 8,11,16,23,30 Day 4 # Days l,8,ll,l6,23,30 x of INITIAL 59Fe INJECTED 400. ISSUE 01 O O 119 Fug/9 T O O IOO« 60 IE4 {3 II 0 I6 2'3 3'0 AY Figure 5 SPECIFIC ACTIVITY x IC'4 9’ 9‘ 9 9‘ L" — A LIVER EXP. = ---* CON. = -——- 61 Spleen There was no significant differences between values for control and experimental spleens in terms of total iron content, 59Fe incorporation, or specific activity. A sharp increase in percent of initial dose of 59Fe was noted (Figure 6A) in the spleen from day 1 ($1.75%) to day 2 (34.5%) followed by a gradual decrease in activity to day 23 (0.5%). The rise in activity from day 23 (0.5%) to day 30 (1.25%) is not significant but may be indicative of a general increase in splenic activity after day 23. Specific activity showed similar changes with time (Figure 6B) as percent of initial dose. Total iron concentration ranged from 900 ug Fe/g spleen to as low as 225 ug Fe/g and averaged 300-400 ug Fe/g for experimental and control spleens. There were no significant differences in spleen total iron content among controls and experimentals due to the large variances in the data. Pyloric Caeca Following a peak in 59Fe concentration in the experi- mental fish at day 4 post injection, there is a reduction in radioiron in the pyloric caeca to 0.5% of the initial in- jected activity by day 8. Less than 0.2% of the activity remained at 23 days following 59Fe injection (Figure 7A). 62 Figure 6.-—Percent of initial 59Fe injected (A), total iron concentration (B) and specific activity (C) of the spleens of experimental and control trout on days l,2,4,8,ll,l6,23, and 30 follow- ing i.p. injection of 59Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to injection of 9Fe. Tissue total iron concentration and 59Fe activity were cor- rected for the tissue blood volume. Each point represents the mean for 6 fish (Day 30 exp., n=5). Significant differences (P< 0.05) among control or experimental means are listed below. A Exp. Day 2 # Days l,4,8,ll,l6,23,30 Day 4 # Days 2,23 C Exp. Day 2 # Days l,11,16,23,30 Day 4 # Days l,11,16,23,30 Con. Day 2 # Days l,4,8,ll,l6,23,30 % of INITIAL 59Fe INJECTED 63 SPLEEN 9001 A ESSUE o 8' Q 9 (lie /g o 9 "9 I00- Tea a II o (6 2'3 3'0 AY SPECIFIC ACTIVITY x Io'4 I24 E; II o Is 23 30 AY Figure 6 EXP. = --- CON. = -—- Figure 7.--Percent initial 59Fe injected (A), total iron 64 concentration (B) and specific activity (C) of the pyloric caeca of experimental and control trout on days 1,2,4 8,11,16,23, and 30 following i.p. injection of 59Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedin s over a 7 day period prior to the injection of 5 Fe. Tissue total iron con- centration and 59Fe activity were corrected for the tissue blood volume. Each point represents . the mean for 6 fish (Day 30 exp. n=5). In paired I comparisons of experimental and control means, I significant differences were found at P<<0.05 (*). Significant differences (P< 0.05) among control or experimental means are listed below. A Exp. Day 4 # Days 8,23,30 B Exp. Day 2 # Days 4,8,11,16 C Con. Day 4 # Days 16,23,30 as of INITIAL 59Fe INJECTED pg Fe/g TISSUE PYLORIC CAECA 65 I24 6 II D I6 23 AY 50 Figure 7 66 This decrease differs from that shown by pyloric caeca tissue from controls whose content ranged from 2.67 to 0.57% of the initial injected activity. Due to variability there was no significant difference in control caecal activity from day 1 through day 30. Fluctuation in the total iron concentration (pg Fe/g tissue) of the control caeca was not statistically signifi- cant. The average concentration was about 35 pg Fe/g caecum. Except for the high iron content of 52 Hg Fe/g caecum on day 2, the experimental fish caeca did not differ in iron concentration (30-35 ug/g) from day l to day 30, nor were the daily averages statistically different from the controls. The specific activity of the control caeca is signifi- cantly higher than the experimental caeca at day 23. This is a reflection of the significant loss of 59Fe from the caeca of experimental fish by day 23 which was not accom- panied by change in the total iron content. The significant difference in the control multiple comparison is due to low 59 total iron concentration and a high Fe activity on day 1. Comparison of the other control means shows no significance and generally reflects the lack of difference among the 59Fe activity and total iron content. 67 Intestine The variation in intestine percent 59Fe incorporation, total iron content and specific activity are shown in Figure 8. Control specific activity decreases significantly from day 1 to day 30 and reflects the decrease in the 59Fe content of this organ from 0.5 to 0.1% of the initial in- jected activity. The experimental group intestinal 59Fe activity was lower than the controls at day 23 and the specific activity was lower than the controls at day 30. Total iron content averaged 30-40 ug Fe/g intestine for ex- perimental and control from days 2 to 30 but the variance in individual sampling periods was considerable as illustrated on day l where the range was 245 to 16 ug Fe/g intestine. As with the caecum this variance in total intestinal iron had a considerable effect on the statistical significance of the specific activity when comparing daily averages within the control and experimental groups and between the two groups at various sampling periods. Overall it appears that the specific activity and 59Fe content of the experimental groups was lower than the control, especially from days 16-30. It appears that the caecum and intestine have the same pattern of 59Fe incorporation following i.p. administra- tion of this isotOpe. 68 Figure 8.--Percent initial 59Fe injected (A), total iron concentration (B) and specific activity (C) of the intestine of experimental and control trout on days 1,2,4 8,11,16,23, and 30 following i.p. injection of é9Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to injection of 59Fe. Tissue total iron concen- tration and 59Fe activity were corrected for the tissue blood volume. Each point represents the mean for 6 fish (Day 30 exp., n=5). In paired comparisons of experimental and control means, significant differences were found at P‘<0.05 (*). Significant differences (P<:0.05) among control means are listed below. The experimental means were not significantly different. C Con. Day 1 # Days 2,4,8,11,l6,23,30 69 V I8 23 DAY EXP - --- CON -.—- wIMym .0 .n e s .0 .0 2 N E filo—x >._._>_._.o< UEGNQm N .0. .0. v. __ m A “a _ LEM D I24 6 II Vi V I22: 3 II 30 IS 23 DAY V s I'I I24 6. 5. A 3 2 I. nu nv 00.0 nu nv autumn; ammo ._<_._.._Z_ B x. Figure 8 70 Muscle The fate of 59Fe in muscle is shown in Figure 9. There was a considerable decrease in 59Fe content and specific activity in the muscle of experimental and control fish. Clearance of radioiron was essentially complete by day 8. Differences in specific activity or percent of the initial injected dose between experimental and control means were not significant. Total iron content remained 20-35 pg Fe/g muscle in control fish, but varied from 143 to 12 pg Fe/g muscle in the experimental fish. However, there was essentially no statistical difference in total iron content between experimental and control groups nor among the sample means. Although the total iron content of muscle (pg/g) was similar to that of intestine and pyloric caeca, the specific activity of the muscle was 19 times lower. Muscle contained a higher percent of the initial dose which is attributed to the fact that the total muscle mass constitutes some 88% of the body weight. Bile Volume Bile volume in the gall bladder varied significantly from day 1 to day 30. The plot in Figure 10A displays the rhythmic change in volume which is significantly lower on days 8 and 16 in both experimental and control fish. 71 Figure 9.—-Percent initial 59Fe injected (A), total iron concentration (B) and speCific activity (C) of the muscle of experimental and control trout on days 1,2,4s ,1l,l6,23, and 30 following i.p. injection of Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to the injection of 59Fe. Tissue total iron concentration and 59Fe activity were corrected for the tissue blood volume. Each point repre- sents the mean for 6 fish (Day 30 exp., n=5). Significant differences (P< 0.05) among control or experimental means are listed below. A Exp. Day 1 # Days 8,11,16,23,30 Con. Day 1 # Days 8,16,23,30 Day 2 # Days 8,16,23,30 B Exp. Day 1 # Days 2,4,8,11,16,23,30 C Exp. Day 2 # Days l,4,8,ll,l6,23,30 Con. Day 1 # Days 16,30 Day 2 # Days 8,16,23,30 72 .3 C 2 i i .6 IM _ D s a .H P N. .e 0 0 |||||||| III‘ .4 Vflu... mmW..--....--...I.I.........1 - . . "w. 0 O 8 6 4 2 O 8 6 4 2 6 2 I I I I I E on_x >._._>_._.U< 2&0me a :w s u U _ M A .73 B 0 0 .6Y IA i o H Ann! vs III: .4 NI! ".0. I I fi Is 23 30 DAY Figure 9 I24 s I) I (u IIIIIII u m m... m m m w m m OMPUMfiE on—mm 45.52. 008 mammfi. O\on_ o: 73 Percent Initial Dose in Bile The percent initial injected 59Fe appearing in bile is shown in Figure 10B. There was a general decline in activity from day l to day 11 in controls and to day 16 in experimentals followed by an increase in activity to day 30. Although there were no statistical differences in activity among the sample days for control or experimental fish bile, the apparent differences in activity were intriguing. The percent initial injected activity was never above 0.01%, but the increase from 0.001% (days 11 and 16) to 0.004 and 0.0065% on day 30 may indicate loss of some iron to the bile 59 after destruction and desquamation of Fe—labeled red blood cells on days 23 and 30. Intestinal Contents and Urine After an initial peak in 59Fe content in the feces at 3 to 4% of the initial injected dose on day 1, the activity rapidly drops to less than 0.02% by day 2 (Figure 10C). From day 2 to day 30 the 59Fe content remains below 0.02- 0.01% in both experimental and control fish. Intestinal contents from three fish on day l were extremely radioactive (10,000 cpm) which accounts for the relatively high per- centage of initial dose seen in the intestinal contents on that day. The initial peak may be due to accidental 74 Figure 10.vai1e volume (A), and percent of initial injected 59Fe in 1 ml. bile (B) and in fecal material (C) from control and experimental trout on days l,2,4,8,ll,l6,23 and 30 follow- ing i.p. injection of 5 Fe. Forty percent of the blood volume of the experimental fish was removed in 4 separate bleedings over a 7 day period prior to injection of 9Fe. Each point represents the mean for 6 fish (Day 30 exp., n=5). Bile volume in experimental fish is significantly different from control fish on day 2 (P‘<0.05 = *). Significant differences (P< 0.05) among control or experimental means are listed below. A Exp. Day 8 # Days 2,4,23,30 Day 16% Days 2,4,23 Con. Day 4 # Days 1,2,8,16 Day 11% Days 1,2,8,16 Day 23% Days 1,2,8,16 Day 20% Days 1,2,8,16 ml BILE % INITIAL/ml BILE xIO’2 75 45( FECES l 346:) 5 :I g I} C '_ . T fl Z3541 I24 8 u I6 23 30 a: 3 DAY 0) | ‘0 I _.I3.OflIv S a :05 z s -\ at \ I24 8ll I6 23 30 DAY EXP-—-— CON-~—— I24 6 n I6 23 3O DAY Figure 10 76 injection of the 59Fe saline solution directly into the lower intestine. The radioactivity of urine within 4 days of the injec- tion of 59Fe was hardly above background. Therefore it was concluded that insignificant amounts of iron (less than 0.1%) were excreted via the urine. 59 Whole Blood and Whole Animal Fe Activity Figure 11A&B illustrates the percent of initial in- jected dose incorporated by the blood and that recovered in the whole animal. Radioiron incorporated by the total blood volume was essentially that taken up by the red cells since the plasma contained less than 1% of the injected dose by day 4. Peak activity was reached by day 16 when blood from experimental fish contained an average of 80% of the initial injected activity and the controls averaged 55%. This difference was significant and the controls re- mained significantly lower on day 23. The activity appeared to decrease from maximum in both groups but this drop was not significant. The increase in percent reticulocytes corresponds to the increase in 59Fe activity in the experi- mental fish blood. As stated previously, the reticulocyte count is significantly higher in bled fish than in the con- trols on days 11, 16 and 23. 77 Figure 11.-—Percent initial injected 59Fe in whole blood (A), and in the whole animal (B) for control and experimental trout on days 1, 2, 4, 8, ll, 16, 23, and 30 following an i.p. injection of Fe. Percent reticulocytes (—A-) are also shown in A for control and experimental fish. Forty percent of the blood volume was removed from experimental fish in 4 separate bleedings over a 7 day period prior to injection of 59Fe. Each point represents the mean for 6 fish (day 30 exp., n=5). In paired comparisons of experi- mental and control means significant differences were found at Pn<0.05 (*). Significant differ- ences (P‘z0.05) among control and experimental means are listed below. A (percent of initial 59Fe injection) Exp. Day 1 # Days 8,11,16,30 Day 2 # Days 11,16 Con. Day 1 # Days 8,11,16,23,30 Day 2 # Days 8,11,16,23 Day 4 = Day 16 (percent reticulocytes) Con. Day 1 # Day 23 Con. Day 1 # Days 11,16 ..... YYYYYY ...... %of INITIAL Fe INJECTED 78 WHOLE BLOOD .24 {I} I2 5. é I'I I6 23 \ 30 DAY WHOLE ANIMAL I: u I u u 213 30 I24 8 ”CAI? EXP -- .-—-. CON.-- -——~ Figure 11 79 The percent activity in the whole animal, which is the sum of the activity in all tissues sampled and in the total blood volume, reached equilibrium at 70-80% of the injected dose in both groups by day 2. This indicates 59Fe injected i.p. was complete that the absorption of within 2 days after the injection, and almost all of the activity was accounted for in tissues sampled in this study. The majority of the activity appeared in the red blood cell volume. There was no significant change in total body activity from day 2 to 30 in either experimental or con- trol fish. DISCUSSION The main objective of this study was to follow the metabolism of radio—labeled iron in fish for a 30 day period following i.p. injection. Previous studies (Hevesy gt 31,, 1964; Hirschfeld and Gordon, l965a,b; Meints and Carver, 1972) have indicated that i.p. injection of 59Fe was an appropriate and effective means of administering iron to lower vertebrates. In the study reported here, 59Fe was readily absorbed from the peritoneal cavity and 70-90% of the injected dose was recovered in various iron pools. Absorption may have been via the lymph system which, in mammals, removes extracellular fluid from the peritoneal cavity and delivers it to the venous circulation. Although there is little information pertaining to the lymphatic system in teleosts, it does exist in eels. In this species fluid collected by the lateral cutaneous and hemal lymphatic vessels is delivered to a caudal lymph heart which pumps it into the caudal vein (Randall, 1970). Thirayothin and Crosby (1962) speculate that lymphatic absorption is one of two pathways for the assimilation of iron in rats following i.p. injection. The second is via absorption of iron-laden phagocytes across the intestinal serosa leading to the 80 81 appearance of iron in the lamina propria, and, eventually, in the venous circulation of the intestines. Regardless of the mode of absorption, i.p. injection of 59Fe proved to be an efficient and simple way of introducing this tracer into the body iron pools of rainbow trout. The appearance of 59Fe in plasma was very rapid. By the time the first sample of plasma was drawn for analysis 1 day after the i.p. injection, the majority of the initial injected dose had been absorbed from the peritoneal cavity and cleared from the plasma (Figures 3 and 12). The clear- ance of 59 Fe from fish plasma is mathematically described by a logarithmetic function which may be divided into the three components (Figure 12) perhaps comparable to those charac- terized for human plasma by Pollycove (1964) and Finch gt El- (1965). The first component represents the initial clearance of 59Fe-labeled‘transferrin from the plasma into the tissue extravascular spaces which occurs within 24 hrs of adminis- tration. The second component represents the return of 59Fe-transferrin to the plasma pool, a process which is re- sponsible for the change in the slope of the clearance curve between days 2 and 16 following injection. The removal of 59Fe from the plasma by the erythropoietic and iron storage sites is a gradual process which is represented by the third or equilibrium component of plasma iron clearance between days 16 and 30. 82 Figure 12.--Clearance of 59Fe from the plasma of trout following an i.p. injection of this isotope. The observed clearance is the culmination of three independently occurring processes described by the equation: _ 0 Ct — C e l + Cze 2 + C3e 3 where Ct = concentration of 59Fe in the plasma at time t 0 0 0 _ . . . . . Cl,C2,C3 — time zero speCific actiVity of components 1, 2, and 3 kl,k2,k3 = rate constants for components 1, 2, and 3 e = base of the natural logarithms t = time in days The half times for clearance are: ”-Tl—‘Te—g‘adldayforcl 1 = 693 _ .693 = T1 —E;—- 7231 3 days for C2 .693 .693 I: =-——————-—= T1 —k__ .0059 116.7 days for C3 I 000 SPECIFIC ACT. x IO 83 I000. «\ j\ C IO; 3 i c . 2 CI I24 8 II I6 23 30 DAY Figure 12 84 59 Movement of Fe into the extravascular spaces of vari- ous tissues is evident from the initial 59Fe peaks in tissue specific activity seen two days after i.p. injection. Clearance of 59Fe from muscle was rapid, indicating return 59Fe-laden transferrin to the plasma of most all of the after circulation through the extravascular spaces. In con- trast, the gradual decline in specific activity of tissues such as liver, spleen and caecum may be indicative of bind- ing of 59Fe to receptor sites for storage in these tissues. The rapid appearance of 59 Fe in the head kidney and the high specific activity of this tissue indicate that much of the plasma 59Fe is removed by the erythropoietic tissue. Pollycove (1964) reported that as much as 90% of the iron leaving the plasma enters the "erythropoietic labile pool" and less than 5% enters the iron storage pool in humans. In this light, it was not surprising to see a high concentra- tion of 59 Fe in the head kidney on day 2 following the i.p. injection. The percent of initial dose in the remaining kidney was about the same as the head kidney, but the con- centration of 59Fe (cpm/g) was 1.5 to 2 times greater in the head kidney where the majority of the erythropoietic tissue is centered. The activity in the remaining kidney may be attributed to the existence of some erythropoietic tissue in the middle third of the kidney (Klontz et al., 1965). 85 The rapid decrease in head kidney 59Fe activity was probably caused by an increase in erythrocyte radioiron in- corporation (Figure llA). Similar findings were reported by Hevesy §£_§1. (1964) for the tench and Hirschfeld and Gordon (1965a) for turtles. There was also an excellent correlation between the erythrocyte iron incorporation and the number of reticulocytes present in the peripheral circu- lation. Walsh et 31. (1949) and Jensen et 31. (1953) were among the first to recognize this correlation in other species. The output of immature erythrocytes by the head kidney of the bled fish was not significantly higher than the controls until 11 days after bleeding. This difference in reticulocyte population after day 11 was undoubtedly re- sponsible for the significant difference noted between con- trol and experimental red blood cell 59Fe incorporation on days 16 and 23. Even though the increased reticulocyte population was responsible for increased RBC 59 Fe uptake in the bled fish, the RBC iron was not immediately incorporated into heme. This conclusion is supported by the fact that the hematocrit of bled fish recovered by day 16 but the hemoglobin concen- tration did not return to normal until day 30. Smith et a1. (1971) reported similar results with salmon following in- jection of phenylhydrazine, a hemolytic agent. Phenylhydra- zine-induced anemia was more severe, reducing hematocrit and hemoglobin concentrations by 95%. Recovery required 95 days, 86 at which time hematocrit had returned to normal, but hemOv globin concentration was still below normal. Heme pro— duction, as reported by Hevesy et_al. (1964), is much slower than 59Fe incorporation. In their study on tench, they found that 5 9 Fe-heme did not approach equilibrium until 20 days post 59Fe injection, unlike total erythrocyte radio— iron content which approached equilibrium at day 10. Iron was incorporated into the erythrocyte faster than it could be utilized for heme synthesis. This suggests that non- heme iron is stored in the cytoplasm of immature fish erythrocytes prior to heme formation. A similar situation in immature human red cells was reported by Kaplan §E_§1, (1954) who discovered the presence of stainable non-hemo- globin iron in the cytoplasm. They attributed the presence of this non-heme iron to a normal phase of intracellular storage during the process of hemoglobin synthesis. Bessis and Breton-Gorius (1959) have identified the storage form as ferritin and ferruginous micelles in human erythroblasts. A non-heme iron pool has also been demonstrated in avian erythrocytes (Jensen 33 a1., 1953). It is possible that a similar situation exists in fish red blood cells. Iron enters the red cell and is stored as ferritin until it is placed into the heme moiety. It is also conceivable that not all the iron stored in the red cell is converted to heme as the cell matures. 87 There may be a small fraction of nonvheme iron which is exchangeable with iron in the plasma. Hevesy gt_al. (1964) noted a decrease in total erythrocyte 59Fe content with time which they attributed to such an exchange. Although non-heme 59Fe was replaced by unlabeled plasma iron, the heme-59Fe activity did not decrease during the same period. Jensen et 31. (1953) noted that when iron is incorporated into the heme moiety it is irreversibly bound, therefore no exchange can take place with either the cytoplasmic non- heme iron or with plasma iron. Any decrease in the total hemoglobin-59Fe content of the red blood cell is attributed to red cell death. Therefore, in this study the decrease noted in 59Fe activity of the red blood cells after day 16 must be attributed to exchange of erythrocyte non-heme iron with unlabeled plasma iron. Jensen et_§1, (1953) found that :mature avian erythrocytes were also capable of non-heme iron exchange. Therefore, the exchange seen in fish erythrocytes is probably not limited to young red cells and may partially 59Fe by red blood cells from the account for the uptake of control fish. In summary, the incorporation of 59Fe by red blood cells probably represents two physiological processes occur- ring simultaneously; i.e., the incorporation of iron for hemoglobin formation by immature cells and the exchange of iron between red cell cytoplasmic stores and the plasma. 88 The latter is not restricted to immature or recently mature erythrocytes, but may also take place across the cell mem- brane of older red blood cells. Liver, spleen and pyloric caeca all exhibited a peak in specific activity between day 2 and day 4 post injection followed by a sharp decrease by day 8. This may represent the initial circulation of 59Fe through the extravascular spaces and its reentry into the plasma pool. In fact, the 59Fe in liver and cyclic reappearance and disappearance of caecum at 6-7 day intervals may be the result of movement of iron into and out of the interstitial fluid of these organs. The liver of control fish was the only tissue in which the percent of original dose remained above 12% throughout the entire 30 day period of the experiment. The liver was 59 the major organ for Fe storage and was the only tissue in which there were significant differences between control and experimental 59 Fe content, specific activity, and total iron concentration. Control liver specific activity was highest of any tissue after day 8 (Table 3), however, this was not the case for the experimental group, in which liver specific activity fell below that of the spleen. The significant de- crease in total iron concentration, etc., coincided with the increase in blood 59Fe content for experimental fish above that of controls. This indicates that in bled fish more iron was mobilized from the liver iron stores for incorporation 89 into the immature red cells. No other tissue displayed such a unique response to the increase in demand for iron. Following the apparent flux of 59Fe into and out of the splenic extravascular fluid between day l and 4, the 59Fe approached equilibrium with the splenic iron storage pool between days 4 and 16. The percent initial dose ap- peared to decrease on day 23, however, there was no sig- nificant difference between experimental and control specific activity or total iron concentration. Splenic iron stores were not significantly reduced due to bleeding. In contrast, Yu et_al. (1971) report a decrease in splenic hemosiderin bodies within 24-48 hrs following bleeding, indicating the mobilization of splenic iron for erythro- poiesis. They report that spleens of bled fish contained significantly less hemosiderin than controls. A decrease in splenic iron content of the magnitude noted by Yu et_al. (1971) was not evident in the present study. Attempts were made to section the spleens and stain for hemosiderin, but there was considerable interference with the specificity of the stain due to the presence of ceroid, an unsaturated lipid. W00d and Yasutake (1956) and Grover (1968) report similar problems with the histochemical stain for splenic hemosiderin in rainbow trout and bluegills. The pyloric caeca and intestinal 59Fe content, like that of the spleen, approached equilibrium between days 4 and 11. There was considerable variance in the gut activity 90 during the study which may have been due in part to faulty i.p. injection. However, the difference in control and experimental 59Fe content on day 23 may indicate an in- creased rate of removal of iron from storage sites in bled fish. Neither caecal tissue nor intestine ever contained more than 3% of the initial dose and the total iron concen- tration was 5—10 times less than the iron concentration in the liver and spleen. Therefore, intestine and caeca are relatively unimportant in terms of iron storage. The fact that 59Fe content decreases in the intestine and caeca in response to bleeding may not have been significant in terms of supplying red blood cells with iron, but was significant in that it may have favored an increase in iron absorption by the gut. This decrease in 59Fe content coincides with the initial decrease in experimental liver total iron con- centration. In this experiment, it is logical that a sig- nificant decrease in iron in a major storage organ such as the liver might trigger an increase in rate of intestinal absorption of iron to replace that removed by bleeding. The mucosal block theory proposed by Granick (1946) states that there is an inverse correlation between the mucosal cell iron content and the amount of iron absorbed by the intestine. Thus, when mucosal cell iron content is low, there is an increase in intestinal iron absorption from the lumen of the gut. Whether or not the results reported here support the mucosal block theory is questionable. Even though 91 radioiron content decreased in the intestine and caeca of bled (iron-depleted) fish, the total iron content did not change significantly. Also, intestinal iron absorption was not measured during the study, therefore there is no way of knowing if the decrease in radioiron content of the intes- tine and caeca had any effect on iron absorption. Except for the initial peak in radioactivity in the intestine and caeca due to extravascular iron flux, one would expect to . . . 59 see an immediate decrease in Fe activity in these tissues from bled animals. According to Crosby (1965) and Van Campen (1974) there should be essentially no radioiron in the mucosal cells after day 2. ‘Although the percent ini- tial dose in either tissue was below 1% by day 8, the radioiron content was not significantly depleted until 16-23 days after the i.p. injection of 59Fe. Since these animals were fed every other day, the iron content of the food (300-400 ug Fe/g) and the rate of iron absorption may have been enough to fulfill the demand for iron in bled fish. There was no apparent decrease in total iron content of any tissue within 16 days of the last bleed- ing. Therefore, the absorptive mechanism may have responded to the initial bleeding with an increase in iron absorption 7 days prior to the i.p. injection of 59Fe. The increased absorption may have been enough to supply the head kidney with iron for immature red cells. There may have been no 92 need for a further increase in the absorptive process in the gut until there was significant mobilization of liver iron. Apparently the mobilization of intestinal and caecal iron was necessary only after liver iron stores were sub- stantially depleted (<100 pg Fe/g liver). This suggests a feedback mechanism between liver iron content and the con- trol of iron absorption. Although there is no proof, mediation of the feedback mechanism may be via the transferrin concentration in the plasma. Transferrin is a beta-l globulin which is probably formed in the liver where virtually all a and B globulin synthesis occurs (Davidsohn and Henry, 1969). One might suspect that a decrease in liver iron concentration might trigger a mechanism respOnsible for increasing plasma transferrin concentration. Hevesy gt_§1, (1964) report normal fish liver iron concentration to be 425 i 192 ug Fe/g liver with a mean TIBC of 250 ug%. In the present study, control liver iron concentration never exceeded 250 ug Fe/g. However, TIBC was abnormally high (>400 ug%) throughout the study. When the liver iron concentration of bled fish decreased below 100 ug/g there appeared to be a corresponding increase in TIBC (above 700 09%). The TIBC for controls increased during the same period but not to the same extent. The differences between control and experi- mental TIBC are not clear-cut, thus existence of a functional correlation between liver iron concentration and TIBC is 93 pure speculation. If the relationship is real, then the increase in TIBC in response to decreased liver iron may trigger the intestinal absorptive mechanism by removing the mucosal block. This would allow increased absorption of iron from the lumen of the gut. The role of muscle, with the lowest specific activity and iron concentration, was insignificant in the overall metabolism of iron in rainbow trout. The initial 59Fe flux reflects movement into and out of the extravascular fluid of muscle with no evidence of storage or utilization of 59Fe in myoglobin formation. The slight differences between control and experimental 59Fe Content on days 11 and 23 were also insignificant. The loss of 59Fe in the urine, feces and bile was negligible. Only 4 or 5 fish out of 96 lost significant amounts of 59 Fe to the feces and this was probably due to accidental injection directly into the lower gut. In fact, 70-90% of the initial dose could be accounted for in the tissues of experimental and control fish 11 days after the i.p. injection. There was a slight but insignificant de- crease in whole animal 59Fe content after day 16. This decrease corresponds to an increase in bile radioactivity during the same period. The increase in percent initial dose in the bile after day 16 is statistically insignificant, but the upward trend may signify the appearance of 59Fe from denatured red blood cells. Normally, nearly all of the iron 94 from catabolized hemoglobin enters the reticuloendothelial cells of the spleen as described (for fish) by Yu et_al, (1971) and Smith et 31. (1971). However, a small amount of iron may appear with bilirubin in the bile. Future Research Several interesting questions have been raised as a result of this study. The answers to these questions re- quire the investigation of additional variables in the study of iron metabolism. Isolation of hemoglobin, transferrin, and RBC non-heme iron by electrophoretic techniques would have improved this study. Analytical determination of fer- ritin and histological investigation of hemosiderin bodies should have been included to substantiate some of the con- clusions and speculations. However, measurement of these parameters was not realistic in a study of this magnitude. Now that there is some basic information established about the metabolism of iron in trout, the finer points can be investigated. The fate of iron in red blood cells should be studied to elucidate the possible presence of iron storage proteins in the cytoplasm and to determine the rate of in- corporation of this iron into heme. This would require fractionation of the cell and isolation of the heme and non- heme fractions. 95 The relationship between transferrin concentration in the plasma and liver iron content should be studied. One might induce iron deficiency through severe bleeding over one or two weeks, remove the liver, determine the trans- ferrin production in_yit£9, and compare this to production by suitable control liver samples. Another important aspect is intestinal absorption of iron. Interference with the absorptive process will cer- tainly affect iron metabolism on a chrOnic scale. An in- depth study of absorptive rates and the absorptive mechanism should be conducted under normal and iron deficient condi- tions. This might involve cannulation and perfusion of major intestinal arteries and veins to investigate the ef- fects of supra and subnormal concentrations of plasma iron and transferrin on the rate of iron absorption across the mucosa. The existence of an erythropoietic stimulating factor in fish is now under investigation. Zanjani 32 21. (1969) reported the presence of a factor in the plasma of anemic fish which initiates increased red blood cell production in recipient fish. This factor has been isolated from human urine, and there are hemagglutination and bioassay pro- cedures for determining human plasma ESF concentration. However, fish ESF has not been isolated. The characteriza- tion of and assay for fish ESF would be an interesting and 96 valuable area of research. With the development of an assay procedure, one could measure changes in fish plasma ESF levels in response to various stimuli known to increase erythropoiesis in mammals, Such as decreased blood PO and 2 hemolytic anemia. Prosser et_al. (1957) found that goldfish could be acclimated to an environmental oxygen content as low as 2 ml OZ/liter of water and they increased their hemo— globin concentration. It would be interesting to measure the plasma ESF titer in goldfish adapted to low oxygen and compare them with rainbow trout after exposure to low en- vironmental dissolved oxygen. Trout are much more sensitive to hypoxia than goldfish or carp; perhaps due to a lack of erythropoietic response (in trout) to low environmental 02 concentration. The ineffectiveness of hypoxia to stimulate erythropoiesis has been reported in turtles (Altland and Parker, 1955; Altland and Thompson, 1958) and in frogs (Rosse gt 31., 1963). Red blood cell 59Fe incorporation appears to be the most sensitive parameter for the study of the effects of environmental pollutants on fish iron metabolism and erythro- cyte production. Zanjani et_al, (1969), Hevesy gt 21. (1964) and Walker (1972) have shown that fish red blood cells can be successfully incubated in yitrg at room temperature. The rate and amount of 59Fe uptake by RBCs can easily be 59 determined. It is possible to show differences in Fe incorporation which are directly related to the percentage 97 of immature RBCs present in the circulating blood. One problem encountered in in yitrg studies is the maintenance of the iron concentration of the medium used in all incu- bations. One way to circumvent the problem is to prepare a medium of hydroxymethyl‘aminomethane (TRIS) buffered physiological saline containing a known iron concentration and a specific amount of transferrin which approximates the concentration normally found in fish plasma (0.2-0.3 g transferrin/100 ml). The specific activity of this medium is important in the calculation of iron incorporation by the red blood cells. No one parameter by itself is significant in terms of the physiological response of an animal to stress, therefore, hemoglobin, hematocrit and reticulocyte counts should be monitored in addition to RBC 59 Fe incorporation. Plasma iron concentration and TIBC are also important indicators of iron metabolism and ferrokinetics of fish. CONCLUSIONS In this study, the uptake and distribution of 59Fe to various pools was monitored in control and iron deficient rainbow trout following i.p. injection of the tracer. Total iron concentration and specific activity of tissues and organs were measured over a 30 day post injection period. 1. Within 24 hours most of the radioiron was absorbed from the peritoneal cavity, probably via the lymphatic system. Upon entering the plasma, radioiron is assimilated by transferrin. 59Fe-labeled transferrin probably enters the extravascular fluid of tissues, circu- lates, and reenters the plasma. During this process some 59Fe is released from the transferrin and be- comes incorporated into tissue iron storage sites. The erythropoietic response to bleeding is slow since it was 11 days after the last bleeding before reticulocyte counts were significantly higher in bled fish than in the controls. The incorporation of 59Fe by red blood cells prob- ably represents two physiological processes occurring simultaneously; the incorporation of iron for the 98 99 purpose of hemoglobin formation and the exchange of iron between cytoplasmic stores and the plasma as shown by Hevesy gt gt. (1964). The liver, spleen and head kidney are the major iron storage organs. Intestine, pyloric caeca and muscle are insignificant in terms of iron storage. The liver iron pool is the primary source for iron utilized in hemoglobin formation. There may be a feedback mechanism between liver iron stores and the mucosal cells of the intestine. When liver iron concentration decreases intestinal ab— sorption of iron increases. The mediator of this system may be transferrin. The spleen contains the highest total iron concen- tration but turnover of the splenic iron stores appears quite slow in contrast to liver iron stores. Red blood cell 59Fe incorporation, per cent reticulo- cytes, hematocrit and hemoglobin are the best indi- cators of erythropoiesis. Liver iron content and plasma transferrin concentration are the primary indicators of iron metabolism. APPENDICES APPENDIX A PHYSIOLOGICAL SALINES 0.9 g% Saline 9.0 g Sodium chloride diluted to 1 liter (280-290 mOSm) Phosphate Buffered Saline (PBS) One liter of a 280 mOSm PBS solution was pre- pared as follows: Sodium chloride 7.82 g Potassium chloride 0.30 9 Magnesium sulfate 0.14 9 Potassium phosphate monobasic 0.46 g Sodium phosphate dibasic 2.02 g Dilute with distilled watertx>make 1000 ml 100 APPENDIX B PLASMA IRON DETERMINATION Method of Plaut _e_t_ El. 1972 REAGENTS Protein Precipitating Reagent: Trichloroacetic acid (TCA) 25.0 g Distilled water 250 m1 Concentrated HCl 4.17g Thiourea 1.67g Dilute with distilled water to make 500 ml Store at room temperature. Color Reagent: Sodium acetate 35' 3-(2-pyridy1 -5,6,dipheny1-l,2,4 triazene 200 mg (Ferrozine , Hach Chemical, Ames, Iowa) Dilute with distilled water to make 100 ml Standard (stock): 1000 ug/ml Use pure (99%) iron wire (Baker Chemical Co., Phillipsburg, N.J.) Polish with sand paper or emery cloth Weigh out 1 g and dissolve in 5 ml of concen- trated HCl Dilute to 1 liter with iron free distilled water Working Standard: From this stock solution of 1 g/L make a set of standards above and below the expected range of plasma iron concentration: 101 102 10 pg/lOO ml 50 pg/lOO m1 100 pg/100 m1 250 pg/100 ml Procedure: Use iron free glassware in every step. This is done by soaking in 5 N HCl overnight and rinsing with iron free water. 1. Add 3 ml of the protein precipitating reagent to 1 m1 of plasma, working standards, and blank (iron free water). Mix tubes thoroughly and allow to stand for 10 min. Centrifuge for 10 min at 3,000-6,000 rpm. Transfer all of the supernatant to a separate tube and discard the precipitate. Centrifuge for 5 min at 3,000-6,000 rpm or until the supernatant is completely clear of any remain- ing white precipitate. Transfer 3 ml of supernatant to another tube, add 0.6 ml of the color reagent and mix. Zero the spectrophotometer and adjust the blank to read 100% T at 560 nm. Read the standards and the unknown plasma samples. From a plot of the standard concentrations vs. the optical density, calculate the unknown concentrations in pg Fe/100 ml plasma. APPENDIX C UNSATURATED IRON BINDING CAPACITY (UIBC) Method of Kuypers and vanOers (1973) REAGENTS Ferric-Isotope Solution: (Ferric ammonium citrate) 1. Dissolve 414.3 mg of ferric ammonium sulfate-12 H O in each ml of 0.1 N HCl. Pipet 1 ml of this solu- tion into a centrifuge tube and add 30-40 pCi 59FeCl . 3 2. Add 0.5 m1 of concentrated ammonium hydroxide to precipitate the iron. Centrifuge and wash the precipitate with 5% ammonium hydroxide. 3. Dissolve precipitate in 2 ml of iron free water containing 100 mg of citric acid. Dilute to 90 ml with 0.05 M tris buffer (2-amino-2-(hydroxymethyl)- 1,3-propanediol) and adjust the pH to 7.0 with 5% ammonium hydroxide. Dilute to 100 ml with iron free water. 4. Store at 4°C. 5. Determine the iron concentration by the bathophen- anthroline method (Appendix D). Anion Exchange Resin: A 22 x 24 cm sheet of Neptonanion exchange resin was obtained from Gallard-Schlesinger Chemical Co., New York (Product No. S 43000). The sheet was cut into 35 x 7 mm strips which were stored at room temperature in 0.5 M TRIS (pH 7.5). Do not refrigerate. 103 104 Glassware: l. 2. All glassware was made iron free (Appendix B) before use. 12 x 75 mm glass culture tubes used in the incuba- tion tubes were coated with Siliclad ® (Clay—Adams, Inc., New York). Procedure: 1. Pipet 0.2 m1 of plasma or 0.2 ml of 0.9 g% saline blank and 0.2 ml of the ferric—isotope solution into a culture tube and mix continuously for 2 hrs on the Variable Speed Rotator at room temperature. Add a resin strip and 2 ml of 0.05 M Tris buffer (pH 7.5) and continue the incubation for 6 hrs to absorb excess ferric~isotope solution. Determine 59Fe activity in the culture tube, and then remove the resin strip and recount the tube. The UIBC is computed from the following equation: A = 59Fe activity incorporated by 0.2 m1 plasma = cpm in plasma culture tube after removing resin strip minus the cpm in the blank tube after removing resin strip. B = ferric-isotope solution specific activity = cpm/pg Fe C = pg Fe incorporated by 0.2 m1 plasma = wlib’ UIBC = C x 500 = pg Fe/lOO ml plasma APPENDIX D TISSUE IRON ANALYSIS Adopted from the Methods of Dow Chemical (1970), Harold Bergman (personal communication) and Diehl and Smith (1965) Tissue Digestion Procedure: 1. 2. 3. All glassware used was iron free (Appendix B). Tare weight tissues in Pyrex brand micro-Kjeldahl flasks (A. H. Thomas Co., Philadelphia, Pa.) Add 5 m1 of concentrated HCl and 1 m1 of concen- trated HNO to the tissue and to five 1 m1 iron standards 3nd a distilled H20 blank. Place the flasks in a shaking water bath at 90°C and digest tissues for 2-3 hrs. Remove the flasks from the water bath and, after cooling, add 15 m1 of 6% potassium permanganate. Allow the tissue oxidation to continue overnight at room temperature. The following morning, decolorize excess permangan- ate by the drop by drop addition of 10% hydroxyl- amine hydrochloride. Any samples which are not clear or which contain undigested tissue are discarded. This occasionally is evident in large samples of intestine with high lipid contents. 105 106 IRON DETERMINATION (Diehl and Smith, 1965) REAGENTS Color Reagent: Sodium salt of 4,7—diphenyl—l,lOvphenanthrolinedisulv fonic acid (Bathophenanthroline, sulfonated sodium salt) 1.5 mM solution Reducing Agent: Ascorbic acid 1 M solution Buffer: Sodium acetate 4.5 M solution 1. Filter the 4.5 M sodium acetate solution through No. 1 (7 cm) Whatman filter paper using a Bfichner funnel. 2. Add the chelating resin, Chelex 100 (BioRad Labs, Richmond, Calif.) to the filtrate and mix for 203 hrs. Allow the resin to settle, carefully decant the sodium acetate solution and filter to remove any resin beads. PROCEDURE Dilute the digestates to 40 ml in the micro-Kjeldahl flasks with iron free water. Using iron free glassware, transfer 4 ml of digestate to a 25 ml volumetric flask. Add 3 m1 of 1 M ascorbic acid, 5 m1 of the color reagent, and 6 ml of the 4.5 M sodium acetate buffer to each 25 ml volumetric flask in the order as given. Dilute to volume with iron free water, mix and allow 30 min for color development. After zeroing the spectrophotometer, adjust the blank to read 100% T at 540 nm. Unknowns and standards are read as percent T and converted to O.D. Iron concentrations are expressed in pg Fe/g tissue. 7 0 l pwscHucoo 10000.0m00.0 A0000.OM0H.0 A00H0.0m00.0 A000m00 A0000.HMOH.HH “0000.0Hm00.00 .aoo x0000.0+00.0 H0000.0+0H.0 A0000.0+00.0 A000 +00 x0000.0+0H.0H 10000.0 +00.00 .mxm 00 H00HH.0MH0.0 x0000.OM0H.0 10000.0M0H.0 x000.M00 x0000.mm00.0 10000.0 M00.00 .coo H0000.0+0H.0 x0000.0+0H.0 A0000.0+00.0 H000 +00 H000H.0+0H.00 x000H.0 +00.00 .mxm 00 H00HH.oM0H.0 x0000.OM00.0 “0000.0M00.0 H000 M00 A0V0H.AM00.0H H000H.0hM0H.00 .coo A0000.0+00.0 A0000.0+0H.0 A000H.0+00.0 H0VH +00 A0000.H+00.H0 A0000.0 +H0.00 .mxm 0H 10000.0m00.0 A0000.OM00.0 x0000.0M00.0 H000 M00 A0000.AM0H.0H 10000.0HW0H.00 .qoo A000H.0+00.0 A0000.0+0H.0 H00H0.0+00.0 .H000 +00 H0000.0+00.00 A0000.0H+00.00 .mxm HH H0000.0M00.0 A0000.OM00.0 x0000.0m00.0 _ A000 M00 A0000.AM0H.HH x0000.0HM00.00 .coo H0000.0+0H.0 A00HH.0+00.0 x0000.0+00.0 . A000 +00 A0000.0+00.00 10000.0H+00.H0 .mxm 0 H000H.0M00.0 10000.0m00.0 x0000.0M00.0 A000 mH0 A0000.Hm00.0 x0000.0HM00.00 .coo i000H.0+00.0 H0000.0+00.0 A0000.0+00.0 A000 +00 A0000.0+00.0H x0000.0 +00.00 .me 0 A0VH0.eM00.0 A000H.0m00.0 A00H0.0M00.0 H000 mH0 A0000.HM00.0 A0VHH.0HM00.00 .coo H0VHH.0+00.0 H0000.0+0H.H A0000.0+00.0 i000 +0H A0000.0+00.0H “0000.0 +00.00 .mxm 0 x0000.OM00.0 A0000.0M00.0 10000.0M00.0 A000.M00 x0000.HM00.0 H0VH0.0Hw00.0H .coo x0000.0+00.0 H0000.0+00.H A0000.0+00.0 i000 +0H A0000.H+00.0 H0000.0 +00.0H .mxm H NIOH x HE mammHm Aucmoummlmv Aucmoummv 0mm COOHm 0H0£3 .mwouo COHuomhmH \ucmoumm OHHm :H uswonmm am pom CH unmoumm CH unmoumm umom >00 m H0>0 mmCHpmen mumummmm HDOM cH UO>OEOH was anm HmucofiHpmmxm m0 masHo> COOHQ 0:» MO 9000 .0000 Ho :oHuochH .m.H 0cHonHoH 00 000 .00.0H.HH.0.0.0.H 0000 co HmHm 00Hn 000 Houucoo mo cooHn 0H0£3 0cm .mEmmHm .msmmHu m0 auH>Huom OHMHommm 6cm .coHumuucmocoo GOHH Hmuou Iumm wuuom .GOHuomncH .m.H map on HOHHm COHHOQ >00 cm>mm .Om UmuomflcH HOHuHcH unmoumm m0_wzvmm.H xm mGOHum>H0mno m0 Hmnfisc cam .HOHHO cumvcmum,.ccszI.m OHQMB mm m NHQZMQAfl 108 UmscHucoo H0000.0M00.H H000H.Hum00.0 A0VHAM000 A0000.AM0H.0H A000 ”M00 A0000.OM00.0 .coo H0000.0+00.0 H000H.H +H0.H H00H0+00 H0000.0+00.H A0V0H +00 x00HH.0+00.0 .mxm 00 x00H0.QM00.0 A000H.0HM00.0H x0000M00H x0000.mM00.0H H000 ”M00 H0000.omH0.0 .aao x0000.0+00.0 10000.0 +00.0 1000 +00 x0000.0+00.0 x000 +0H A000H.0+00.0 .020 00 H0000.OM00.0 H00H0.0HM00.0H A000AM0HH 10000.0M00.0H x00H0Hm00 H0000.OM0H.0 .coo H0000.0+00.0 x0000.0 +00.H H0VHH+H0 A00H0.0+00.0 x000 +00 A0000.0+00.0 .mxm 0H H0000.0M00.H 10000.0HM0H.0H H000HM0H0 10000.0m00.00 x000HHM00 “0000.0M00.0 .coo H0000.0+00.0 A0H00.0 +00.0 A0000+H0H. A0000.0+H0.0 H000 +00 H0000.0+00.0 .mxm HH H0000.om00.H A0000.0Hm00.HH H000HM0H0 H0000.AMOH.HH x000 HmH0 H0000.0M00.0 .coo A0000.0+00.0 A00H0.0 +00.0 AOVHO+00H A0000.0+00.0 A0VH0 +00 H0000.0+00.0 .me 0 H0000.HM00.0 H000H.0HM0H.00 A0000M00H H0000.AM00.0H A000 ”M00 10000.0MH0.0 .coo A0000.H+H0.0 A00H0.HH+00.00 A0000+00H A0000.0+00.0H H000H +00 H0000.0+00.0 .mxm 0 H00H0.Hm00.0H x00H0.00m00.00 H0000M0HH H000H.0M00.0H x0000wm00H A0000.OM0H.0 .coo H0000.0+00.0H x0000.0 +00.00 A0000+00H H0000.0+00.0H x0000 +00H A0000.0+00.0 .mxm 0 x0000.HM0H.0 A0000.0HmH0.0H H000mw00H H0000.AM00.0H H0000HM000 10000.0m0H.0 .aoo H00HH.0+00.0 H00H0.0 +00.0 A0000+000 x0000.0+00.0 A0VH0H+000 A0000.0+00.0 .mxm H NOCUHM H0>HH WIOH H0>HH H0>HH mamMHm mIOH AHEV .INTOHO COHHOOfldH CH ucwopmm 00.uo< .omcm m\0h on cH usmoumm 00.u0< .ommm OESH0> OHHm “mom mama pmscHHGOOIIm mHnma 109 pmscHucoo H0000.HUM0H.0 H0VHHM0HH A000H.OM00.0 x0000.HmH0.0 A00H0HM000 A0000.0m0H.H .coo H0000.0 +00.0 i00H0+0HH x00H0.0+H0.0 A0000.H+00.0 A0000H+000 A0000.0+0H.H .me 00 H0000.0dm00.0 H0000M00H H00H0.HM00.0 A00H0.HM00.H H0000HM000 A0000.0M00.0 .coo H000H.0 +0H.0 i000 +00 A00H0.0+H0.0 A0000.H+00.0 H0000 +000 H000H.0+00.0 .mxm 00 H0000.0Uw00.0 A0V0HM00H H0000.0M00.0 x0000.HM00.0 i00H0Hm000 x0000.QM00.H .coo H0000.H +00.0 H000 +00 H000H.0+00.0 x0000.H+00.0 H0000 +HH0 A0000.0+00.H .mxm 0H H0000.0HM0H.0 A000Hw00H H0V0H.OM00.0 x0000.HM0H.0 H0000HM0H0 A000H.OM00.H .coo x0000.0 +00.0 A0000+H0H H0000.0+00.0 H0000.0+00.0 H0VH0H+000 A0000.0+00.0 .mxm HH H0000.H mH0.0 x0000m00H H0000.0m00.H A0000.Hm00.0 0000 M000 A0000.0.I.H00.H .08 10000.0 +00.0H A0000+H0H H0V0H.0+00.0 A0000.0+0H.0 H0VH0 +000 A0000.0+H0.H .mxm 0 H00H0.0 m0H.00 A0V0AM00H H0000.OM0H.0 x0000.Hw00.0 AOVOHHmH00 x0000.OM00.0 .coo H0000.00+00.0HH A000H+00 H00H0.0+H0.0 A0V0H.0+HH.H0 A00000+000 H0000.0+0H.0 .mxm 0 00000.00M00.00 A0000M00 H0000.Hm00.0 HO0HH.0MOH.0H x0000mm000 x0000.HM00.0 .coo A0000.00+00.00H H0000+00 x00HH.H+00.0 H0000.0+00.00 H00H0 +000 H0000.0+00.0 .000 0 “0000.00M00.00 H0000M00H H0000.HM00.0 A0000.HM00.0 H0000NM000 H0000.OM00.H .coo i000H.0 +00.00 A0000+000 A0000.H+00.0 H0000.0+00.H H0000H+HH0 A0000.0+00.H .000 H mOCCHM mIOH IchcHM .m MOGCHM cmmm cmmHmm mIOH ammamw cmOHmm macaw GOHHOONQH x .uo¢ .ommm m\0m on cH usmoumm x .uod .Ommm m\0m m: CH unmonmm umom mama I concHuCOOII.m OHQMB 110 COHGHUGOU H0000.oM0H.0 x000um00 H0000.HMH0.0 A000H.HMH0.0 A000HM00 A0000.QM00.0 .000 H0000.0+00.0 H000 +H0 H00H0.0+H0.0 H000H.0+0H.0 .000 +00 H0000.0+00.0 .020 00 H000H.OM00.0 H000HM0H H0000.HM00.0 A000H.HM00.0 .000HM00 A0000.OM00.H .aoo H0000.0+00.0 H000 +00 H0000.0+00.0 A0000.0+H0.0 .000 +00 H00HH.0+0H.0 .mxm 00 H0000.0w0H.0 H000HM00 H0000.mw00.0 .00H0.HM00.0 A000HM00 H000H.0w00.0 .aoo H0000.0+00.0 H000 +0H H0000.0+00.0 H0000.o+0H.0 H000 +00 H0000.0+H0.0 .mxm 0H H000H.OM00.0 H000HM00 H0000.HM00.0 00000.0M00.0 A000Hw00 H00H0.OM0H.0 .qoo H0000.0+00.0 H000H+00 H000H.H+00.H .0000.0+0H.0 i000 +00 H0000.0+00.H .000 HH x0000.0M0H.0 H000Hw00 “0000.0M00.H H0000.mw00.0 .000HM00 H0000.HM00.H .qoo A0000.0+0H.0 A000 +00 H0000.0+00.H H0000.0+00.0 .000 +00 H000H.0+00.0 .mxm 0 A0000.0M00.0 A000Hw00 H0000.HM00.0 H0000.Hm00.0 H00Hflm00 H0000.OM00.H .qoo H000H.0+00.0 H000 +00 H0000.0+0H.0 H00HH.0+00.0H A000 +00 H0000.H+0H.0 .020 0 H00HH.0M00.H H000HMH0 H0000.HM00.0 .000H.0M00.0 H000Hm00 H0000.HM00.0 .noo H0000.0+H0.0 H000 +0H H0000.H+00.0 H0000.H+0o.0 .000 +00 H0000.0+00.H .mum 0 H0000.em00.0 H000Hw00 H000H.0M00.0H H0000.HM00.0H .000 M00 x0000.OM00.0 .qoo H0000.0+00.0 H0000+00H H0000.0+00.0H H0000.H+H0.0 .000 +00 A0000.0+00.H .mwm H mHomDZ mIOH mHomsz mHomDS 00000 mIOH 00000 mommo .mdonu nOHHUONGH x .uo< .owmm m\0m m: 0H unmoumm x .u0¢ .ommm m\om m: nH unwoumm umom mama eosfiuaooI. 0 030.0 lll H0000.CM00.0 i00H0.0+00.0 H0VH0.OM00.0 HovHo.o+vo.o H00 000.0M00.0 A00H0.0+00.0 H00H0.OM00.0 H0000.0+00.0 H00H0.OM00.0 A00H0.0+00.0 10000.0mH0.0 H0000.0+0H.0 H00H0.0MHH.0 A0000.0+0H.0 A000H.AM0H.0 H0000.0+H0.0 030.0 m0H.H A000H.0 +0H.0 0000.0 M00.H H0000.0 +00.0 0000.0MHOH A0000.0 +00.0 10000.0 M00.H H0000.H +00.0 1003.0 M000 x0000.0 +00.0 x0000.Hm00.0 A0000.H +00.0 10000.0 M000 AO0H0.H +00.0 HO0H0.0mu00.00 “0000.0.H00.0 A000 HmH0 A000 +H0 000 M00 HO0H +00 A000 ”M00 A0000 +00 A000 “M00 H000 +00 00 0H M00 H000 +00 000 M00 Hmva +00 1000 HM00 H000 +00 x000 “MOH HO0HHH+000 H0000.OM00.0 i00H0.0+H0.0 H0000.OM0H.0 x0000.0+00.0 x0000.OM00.0 A0000.0+00.0 A0000.CM0H.0 A000H.0+00.0 A0000.OMH0.0 x00H0.0+00.0 “0000.0w00.0 H0V0H.0+00.0 A0000.OM00.0 A0000.0+00.0 H000H.OM00.0 A0000.o+00.o «COO .0me e GOO .mxm .COO .mxm .Coo .mxm .Coo .mxm .Coo .mxm .COU .mxm .COU .mxm om mm 0H HH H mmomm CH quonmm OCHummuCH 0I0H x .060 .0000 mCHumouCH m\mm m: mCprouCH CH HCOUHOC .mCOHo COHuouflmH umom mhmn COCCHHCOOII.m OHQMB LITERATURE CITED LITERATURE CITED Abbrecht, P. H., and J. K. Littell. 1972. Erythrocyte life-span in mice acclimated to different degrees of hypoxia. J. Appl. Physiol., 32:443-445. 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 the box turtle. Am. J. Physiol., 180:421-427. Altland, P. D., and E. C. Thompson. 1958. Some factors affecting blood formation in turtles. Proc. Soc. Exp. BiOlo Medo' 99:456—459. Bearn, A. G., and W. C. Parker. 1964. Some observations on transferrin, p. 60-72.’ £g_F. Gross (ed.) Iron metabolism. Springer-Verlag, Berlin. Bessis, M. C., and J. Breton-Gorius. 1959. Ferritin and ferruginous micelles in normal erythroblasts and hypo- chromic hypersideremic anemias. Blood, 14:423-432. Callender, S. T. 1964. Digestive absorption of iron, p. 89-102. IS F. Gross (ed.) Iron metabolism. Springer— Verlag, Berlin. Catton, W. T. 1957. Blood cell formation in certain teleost fishes. Blood, 6:39-60. Cline, M. J., and T. A. Waldmann. 1962. Effects of tempera- ture on erythropoiesis and red cell survival in the frog. Am. J. Physiol. 203:401-403. Conrad, M. E., and W. H. Crosby. 1963. Intestinal mucosal mechanisms controlling iron absorption. Blood, 22: 406-415. ’ Conte, F. P., H. H. Wagner, and T. 0. Harris. 1963. Measurement of blood volume in the fish (Salmo gairdneri gairdneri). Am. J. Physiol., 205:533-540. 112 113 Cook, J. D., C. Hershko, and C. A. Finch. 1972. Storage iron kinetics. 1. Measurement of the cellular dis- tribution of 59Fe in the rat liver. J. Lab. Clin. Med., 80:613-623. Cook, J. D., C. Hershko, and C. A. Finch. 1974. Storage iron kinetics. ‘IV. Cellular distribution of ferritin iron stores in rat liver. Proc. Soc. Exp. Biol. Med., 145:1378-1381. Crosby, W. H. 1965. The control of iron metabolism. Ser. Haematol., 6:66-71. Davidsohn, I., and J. B. Henry (eds.). 1969. Clinical diagnosis by laboratory methods. W. B. Saunders Co., Philadelphia. 1308 pp. Dawson, A. B. 1933. The relative numbers of immature ery- throcytes in the circulating blood of several species of marine fishes. Biol. Bull., 64:33-43. Diehl, H., and G. F. Smith. 1965. The iron reagents: bathophenanthroline, bathophenanthroline disulfonic acid, 2,4,6-tripyridy1-s-triazine, phenyl-Z-pyridyl ketoxime. G. Frederick Smith Chemical Co., Columbus, Ohio. 64 pp. Dow Chemical Co. 1970. Determination of mercury in fish (atomic absorption spectrophotometric method), method Dowdle, E. B., D. Schachter, and H. Schenker. 1960. Active transport of 59Fe by everted segments of rat duodenum. Am. J. Physiol., 198:609—613. Edwards, S. A., and J. Fielding. 1971. Studies of the effect of sulphydryl and other inhibitors on reticulo- cyte uptake of doubly-labeled transferrin. Brit. J. Haematol., 20:405-416. Field, J. B., C. A. Elvehjem, and C. Juday. 1943. A study of the blood constituents of carp and trout. J. Biol. Chem., 148:261-269. Fielding, J., S. A. Edwards, and R. Ryall. 1969. Effect of sulphydryl inhibition on the uptake of transferrin- bound iron by reticulocytes. J. Clin. Path., 22:677- 679. 114 Finch, C. A., M. L. Hanson, and D. M. Donohue. 1959. Kinetics of erythropoiesis. A comparison of response to anemia induced by phenylhydrazine and by blood loss. Am. J. Physiol., 197:761-764. Finch, C. A., F. Hosain, E. H. Morgan, G. Marsaglia, E. Giblett, and R. S. Hillman. 1965. The ferrokinetic approach to anemia. Ser. Haematol., 6:30-40. Fletcher, J. W., N. I. Gallagher, M. A. Warnecke, and R. M. Donati. 1973. Erythropoiesis and hyperoxia. Proc. Soc. Exp. Biol. Med., 144:569-574. Fletcher, J. W., and E. R. Huehns. 1968. Function of trans- ferrin. Nature, 218:1211—1214. Garby, L., and J-C. Vuille. 1967. The iron pools, p. 361- 380. £3 P. E. Bergner, and C. C. Lushbaugh (eds.) Compartments, pools, and spaces in medical physiology. U. S. Atomic Energy Commission. 521 pp. Gordon, A. S. 1959. Hemopoietine. Physiol. Rev., 39:1-40. Granick, S. 1946. Ferritin: increase in protein apoferri- tin iJI gastrointestinal mucosa as direct response to iron feeding. Function of ferritin in regulation of iron absorption. J. Biol. Chem., 164:737-746. Grover, J. H. 1968. Hemosiderin in bluegill spleens. Trans. Amer. Fish. Soc., 97:48-50. Hahm, P. E., W. F. Bale, J. F. Ross, W. M. Balfour, and G. H. Whipple. 1943. Radioactive iron absorption by gastrointestinal tract. Influence of anemia, anoxia and antecedent feeding. Distribution in growing dogs. J. Exper. Med., 78:169-188. Halliday, J. W., and L. W. Powell. 1973. The use of sus- pensions of isolated rat mucosal cells to study mechan- isms of iron absorption. Clin. Chim. Acta, 43:267-276. Harrison, P. M. 1964. Ferritin and hemosiderin, p. 40-72. I2.F' Gross (ed.) Iron metabolism. Springer-Verlag, Berlin. Hevesy, G., D. Lockner, and K. Sletten. 1964. Iron metabol- ism and erythrocyte metabolism in fish. Acta Physiol. Scand., 60:256-266. 115 Hirschfeld, W. J., and A. S. Gordon. l965a. Some effects of bleeding on the iron metabolism of normal and of starved turtles. J. Exp. 2001., 160:263-270. Hirschfeld, W. J., and A. S. Gordon. 1965b. The effect of bleeding and starvation on blood volumes and peripheral hemogram of the turtle, Pseudemys scripta elegans. Anat. Rec., 153(3):317-323. Jacobs, P., T. H. Bothwell, and R. W. Charlton. 1966. Intestinal iron transport: Studies using a loop of gut with an artificial circulation. Am. J. Physiol., 210:694-700. Jandl, J., J. Inman, R. Simmons, and D. Allen. 1959. Transfer of iron from serum iron-binding protein to human reticulocytes. J. Clin. Invest., Jandl, J. and J. Katz. 1963. transferrin. 38:161-185. The plasma-to-cell cycle of J. Clin. Invest., 42:314-326. Jensen, W., H. Ashenbrucker, G. Cartwright, and M. Wintrobe. 1953. The uptake in vitro of radioactive iron by avian erythrocytes. J. Lab. Clin. Med., 42:833-846. Kaplan, E., W. Zuelzer, C. Mouriquand. 1954. Sideroblasts: A study of stainable nonhemoglobin iron in marrow normoblasts. Blood, 9:203-213. Katz, J. H. 1965. The delivery of iron to the immature red cell: A critical review. Ser. Haematol., 6:15-29. Katz, J. H., J. H. Jandl. 1964. The role of transferrin in the transport of iron into the developing red cell, p. 103-120. IE_F. Gross(ed.)Iron metabolism. Springer- Verlag, Berlin. Klontz, G. W., W. T. Yasutake and T. J. Parisot. 1965. Virus diseases of the western United States. III. Immunopathological Aspects. Ann. N. Y. Acad. Sci., 126:531-542. Klontz, G., W. Yasutake, J. Wales, L. Ashley, and C. Smith. 1969. Application of hematological techniques to fishery research. Unpublished manuscript. Western Fish Disease Laboratory, Seattle, Washington. Kuypers, A. M. J., and R. J. M. van Oers. 1973. A simple method for the semi—micro determination of unsaturated iron binding capacity (UIBC). Z. Klin. Chem. Klin. Biochem., 11:172-173. 116 Laurell, C. B. 1951. What is the function of transferrin in plasma? Blood 6:183-187. Laurell, C. B. 1958. Iron transportation, p. 8-23. lg R. O. Wallerstein and S. R. Mettier (eds.) Iron in Clinical Medicine, U. Calif. Press, Berkley. Lipschitz, D. A., M. 0. Simon, S. R. Lynch, J. Dugard, T. H. Bothwell, and R. W. Charlton. 1971. Some fac- tors affecting the release of iron from reticuloendo- thelial cells. Brit. J. Haematol. 21:289-303. Lucas, A. M. and C. Jamroz. 1961. Atlas of Avian Hema- tology, United States Dept. of Agriculture, Washington. Manis, J. 1973. Ferrous iron oxidation by intestinal mucosa: possible role in mucosal iron metabolism. Proc. Soc. Exp. Biol. Med., 144:1025-1029. Manis, J., and D. Schachter. 1964. Active transport of iron by intestine: mucosal iron pools. Am. J. Physiol., 207:893-900. Meints, R. H. and F. J. Carver. 1972. Erythropoietic activity in Rana pipiens--the influence of nutrition and season on spleen and peripheral blood activities. Comp. Biochem. Physiol. 42A:511-519. Moore, C. V. 1958. An outline of iron metabolism, p. 5&7, l2 R. O. Wallerstein and S. R. Mettier (eds.) Iron in Clinical Medicine, U. Calif. Press, Berkley. Oser, B. L. (ed.). 1965. Hawk's Physiological Chemistry. McGraw-Hill Book Co., New York. 1472 pp. Pearse, A. G. E. 1961. Histochemistry: Theoretical and applied. Little, Brown and Co., Boston. 998 pp. Perris, A. D. and J. F. Whitfield. 1971. Calcium homeo- stasis and erythropoietic control in the rat. Can. J. Physiol. Pharmacol. 49:22-35. Plaut, D., E. Miller, J. MacQueen and R. Blanchard. 1972. Direct and indirect assays of serum iron with a sensa- tive new color reagent. Am. J. Med. Tech. 38:82-89. Pollycove, M. 1964. Iron kinetics, p. 148-184, lg F. Gross (ed.) Iron metabolism. Springer—Verlag, Berlin. 117 Prosser, C. L., L. M. Barr, R. D. Pine, and C. Y. Lauer. 1957. Acclimation of goldfish to low concentrations of oxygen. Physiol. 2001., 30:137-141. Ramsay, W. N. M. 1958. Plasma iron, p. 1-39. £3 H. Sobotka and C. P. Stewart (eds.) Advances in clinical chemistry. Academic Press, New York. Randall, D. J. 1970. The circulatory system, p. 133—172. £2 W. S. Hoar, and D. J. Randall (eds.) Fish Physiology IV. Academic Press, New York. Rodgers, G. M., J. W. Fisher, and W. J. George. 1974. Activation of the renal erythropoietic factor by a cyclic AMP-dependent protein kinase. Proc. Soc. Exp. Biol. Med., 145:1207-1211. Rosse, W., T. Waldman, and E. Hull. 1963. Factors stimu- lating erythropoiesis in frogs. Blood, 22:66-72. Rosse, W., and T. Waldman. 1966. Factors controlling erythropoiesis in birds. Blood, 27:654-661. Sanchez-Medal, L., L. Duarte, and J. Labardini. 1970. Hemolysis and erythropoiesis. VI. A comparative study of the utilization of hemoglobin iron and transferrin iron by the erythropoietic tissue. Blood, 35:721-726. Schade, S. G. 1972. Normal incorporation of oral iron into intestinal ferritin in inflammation. Proc. Soc. Exp. Biol. Med., 139:620-622. . Smith, C. E., L. R. McLain, and W. S. Zaugg. 1971. Phenyl- ' hydrazine-induced anemia in Chinook salmon. Toxicol. Appl. Pharmacol., 20:73-81. Sokal, R. R., and F. J. Rohlf. 1969a. Biometry. W. H. Freeman and Co., San Francisco. 776 pp. Sokal, R. R., and F. J. Rohlf. 1969b. Statistical tables. W. H. Freeman and Co., San Francisco. 253 pp. Tambourin, P. E., F. Wendling, O. Gallien-Lartigue, and D. Huaulme. 1973. Production of high plasma levels of erythropoietin in mice. Biomed. Express, 19:112-116. Thirayothin, P., and W. H. Crosby. 1962. The distribution of iron injected intraperitoneally. Evidence of sero- sal "absorption" by the small intestine. J. Clin. Invest., 41:1206-1210. 118 Topf, W. 1953. Uber die blutbildung und die blutbildungs- statten beim karpfen (Cyprinus carpio L.). 2001. Anzeiger, 150:91-104. Van Campen, D. 1974. Regulation of iron absorption. Fed. Amer. Soc. Exp. Biol. Proc., 33:100-105. Van Wyk, C. P., M. Linder-Horowitz, and H. N. Munro. 1971. Effect of iron loading on non-heme iron compounds in different liver cell populations. J. Biol. Chem. 216:1025-1031. Walker, R. L. 1972. I2 vitro study of erythropoiesis in rainbow trout (Salmo gairdneri) using 59FeCl3. M. S. Thesis. Michigan State Univ. 100 pp. Walsh, R., E. Thomas, S. Chow, R. Fluharty, and C. Finch. 1949. Iron metabolism. Heme synthesis tg vitro by immature erythrocytes. Science, 110:396-398. Wood, E. M., and W. T. Yasutake. 1956. Ceroid in fish. Am. J. Path. 32:591-603. Yu, M. L., C. W. Kiley, D. A. Sarot, and A. Perlmutter. 1971. Relation of hemosiderin to erythropoiesis in the blue gourami, Trichogaster tricopterus. J. Fish. Res. Ed. Canada, 28:47-48. Zanjani, E., J. Contrera, G. Cooper, A. S. Gordon, and K. WOng. 1967. Renal erythropoietic factor: role of ions and vasoactive 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 gourami, Trichogaster tricopterus. Blood, 33: "IIIIIIIIIIIIIIIIIIIII'