THE E w UPTAKE 05 HEXAVALENT CHROMIUM av emvmkomas. uvm AND mnw TESSUE ar- THE TURTLE cn'avsemvs 3915 Thai: for The: Dome o! PIT. D. MICHIGAN STATE UNIVERSITY Jack Russel? Hofiferi I962 This is to certify that the thesis entitled The _I__rl Vitro Uptake of Hexavalent Chromium by Erythrocytes, Liver and Kidney Tissue of the thle Chrysemy_s Picta presented by Jack Russell Hoffert has been accepted towards fulfillment of the requirements for 23.2...— degree inJhXSinng ?m 6. Gm”, Major professor Date February 27, 1962 0-169 LIBRARY Michigan State University amummyWWW L. 3 129 PLACE ll RETURN BOX to remove thle checkout from your record. TO A OID FINES return on or before dete due. DATE DUE DATE DUE DATE DUE ; ‘ R- A ilti‘:v5: MSU le An Affinnettve Action/Equal Opportunity lnetltulon WW1 THE l§_VITRO UPTAKE OF HEXAVALENT CHROMIUM BY ERYTHROCYTES, LIVER AND KIDNEY TISSUE OF THE TURTLE CHRYSEMYS PICTA BY Jack Russell Hoffert AN ABSTRACT OF.A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology and Pharmacology 1962 / Approved by @12th (j w'l’wox/ ABSTRACT THE l§_VITRO UPTAKE OF HEXAVALENT CHROMIUM BY ERYTHROCYTES, LIVER AND KIDNEY TISSUE OF THE TURTLE CHRYSEMYS PICTA by Jack Russell Hoffert Blood samples from the turtle Chrysemys picta were taken by direct cannulation of the left aortic arch. The packed heparinized cells were washed three times with isotonic Ringer-phosphate-glucose buffer (pH 7.4). The uptake of chromium by cells was rapid initially, probably related to the amount of surface binding; this was followed by uptake at an exponential rate but without reaching an asymptotic value by 4 hours. The exponential uptake is related to the rate of diffusion through the cell membrane and to binding (either enzymatic or physical) to the cytoplasmic proteins. Cell homogenates showed a linear uptake over a 4 hour incu- bation period. The binding of chromium to the cell surface or to the cytoplasmic proteins does not appear to be dependent upon metabolic activity since experiments using NaF, NaCN, Na azide and NaAsO2 (10'3m) had no effect on uptake of hexavalent chromium by the erythrocytes or liver and kidney slices. Nucleated erythrocytes when placed in 20 p.p.m. chromium for 1/2 hour at 200C. showed a significant Jack R. Hoffert decrease in their electrophoretic mobility. This indi- cates a reduction in the zeta potential due to binding of chromium at the cell surface. Fractionation and isolation of the erythrocyte components has shown that chromium binds to the haemoglobin as well as to the cell ghosts and nuclei. THE IN VITRO UPTAKE OF HEXAVALENT CHROMIUM BY ERYTHROCYTES, LIVER AND KIDNEY TISSUE OF THE TURTLE CHRYSEMYS PICTA BY Jack Russell Hoffert A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Physiology and Pharmacology 1962 ii AC KN OWLE DGME NT The author wishes to express his sincere appreciation to Dr. P. O. Fromm, Department of Physiology and Pharma— cology, for his assistance which contributed greatly to this work. The author is also indebted to his fellow graduate students for their technical assistance throughout the study. In addition the writer is indebted to the United States Public Health Service, National Institute of Health, for funds in support of this work. iii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . . 4 Inorganic Chemistry of Chromium 4 Metal-Protein Compounds 7 Interaction of Chromium Compounds with Biological Fluids 11 Toxicology of Chromium 15 Distribution and Retention of Chromium in the Organism 23 Nucleated Erythrocytes 29 MATERIALS AND METHODS . . . . . . . . . . . . . . . 31 Experimental Animal 31 Tissue Preparation 34 Radioactive ISosope Methodology 38 Electrophoresis 4O Cataphoresis Apparatus 43 Paper Chromatograms 47 Preparation of Hemoglobin and Globin 47 Autoradiography 49 Cellular Respiration 52 RESULTS . . . . . . . . . . . . . . . . . . . . . . 54 In Vitro Chromium Accumulation . . . . . . . . . 54 Electrophoretic Mobility Temperature Distribution of Na CrSlO in the Nucleated Erythrocyte 2 4 Effects of Metabolic Inhibitors on Uptake of Na2Cr5104 Electrophoresis and Paper Chromatograms Histoautoradiography Effect of Exposure to NaZCrO4 on the 002 Distribution of Chromium Following Intra- peritoneal Injection DISCUSSION . . . . . . . . . . . . . . . . . . SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . REFERENCES CITED . . . . . . . . . . . . . . . . . APPENDICES A. PHOTOMICROGRAPHS AND AUTORADIOGRAMS OF VARIOUS TISSUES FROM CHRYSEMYS PICTA . . . . . . . . B. EFFECT OF METABOLIC INHIBITORS ON THE INLVITRO UPTAKE OF NaZCr5104 . . . . . . . . . . . . C. PREPARATION OF TISSUES FOR RADIOAUTOGRAPHS . iv Page 60 61 62 65 65 75 77 78 82 96 98 107 138 140 TABLE LIST OF TABLES Effect of Different Media on Uptake of Na CrSlO by Nucleated Erythrocytes Electrophoretic Mobility of Nucleated Erythro- cytes When Tagged with NaZCrO4 . Influence of Temperature on the Uptake of NaZCrSlO4 by NUcleated Erythrocytes . . . Effect of Metabolic Inhibitors on Uptake of NaZCrSlO4 by Turtle Tissue . . . . . Effect of Exposure to Na2CrO4 on the 002 of Turtle Tissue . . . . . . . . . . . . . Organ-Body Weight Relationships in Chrysemys picta . Distribution of Chromium in Various Tissues Following Intraperitoneally Injected NaZCr5104. 2 4 PAGE 55 57 63 68 76 79 8O LIST OF FIGURES Ventral View of the Turtle Heart . . . . . . Apparatus for Electrophoresis . . . . . . . Northrup—Kunitz Cataphoresis Apparatus . Multiple-Unit Constant—Pressure Micro- respirometer . . . . . . . . . . . . . . Uptake of NaZCrSlO4 by Nucleated Erythrocytes. Distribution of Na2Cr5104 in the Nucleated Erythrocyte . . . . . . . . . . . . . . . . . Filter Paper Electrophoresis with Super- imposed Radioautograph . . . . . . . . . . . Cellulose Acetate Electrophoresis Strips . . . APPENDIX B Normal Liver and Gall Bladder Section from the Turtle . . . . . . . . . . . . . . . . Normal Kidney Tissue from the Turtle . . . . . Negative Test for Hemosiderin . . . . . . . Liver Section Showing a Positive Bleaching Reaction for Melanin . . . . . . . . . . . . . Chronic Multiple Abscesses of the Turtle Neck. Liver Section Showing Area of Lymphatic Infiltration . . . . . . . . . . . . . . . Section of Aorta Taken Near the Heart . . . Intravascular Parasite Located in a Coronary Artery . . . . . . . . . . . . . . . . . . . . vi Page 35 41 44 50 59 66 69 71 108 110 112 114 116 118 120 122 Figure Gross Specimen of Turtle Heart Following 6 Days of Daily Heart Punctures . . . . . Lung Section Showing Smooth Muscle Radioautograph of Lung Tissue . . . . . Mesentery from Area of Stomach . Radioautograph of Mesentery Section Shown in Figure L . . . . . . . . . . . . Radioautograph of Nucleated Erythrocyte Tagged with Cr51 . . . . . . . . . . . Radioautograph of Small Intestine Showing No Localization of the Radioactivity from Cr51. vii Page 124 126 128 130 132 134 136 INTRODUCTION It is generally believed that pure metallic chromium is biologically inert and exerts no harmful effects on living tissues. Nearly all of the data available indicate that the chromium III (trivalent) compounds probably do not produce any serious damage to body tissues. In con- trast, chromium VI (hexavalent) compounds exert an ex- tremely irritating, corrosive and, under some circumstances, toxic action on body tissues. The harmful effects may be due to the oxidizing ability of the compound or general properties similar to those associated with heavy metals. The factors involved in chromium accumulation and excretion have been described as has work on the transport and labeling of non-nucleated erythrocytes with hexavalent chromium-51. The application of the tagged cell technique in clinical practice is well documented. Measurements of total cell volume, blood volumes, plasma volumes and red cell life spans have all been accomplished using different forms of chromium-51. Basic differences exist between the nucleated and non— nucleated erythrocytes. The nucleus is indirectly indicated as the controller of the oxidative processes of the cell. Upon losing the nucleus in the normoblastic phase of hema— topoiesis the non—nucleated red blood cell also loses its mitochondria. Approximately 80-85% of the dry weight of nucleated and 95% of the dry weight of the non-nucleated red blood cells is made up of metabolically inert haemo— globin (Ponder, 1948). For this reason it would be pre- dicated that the red cell might have a different total metabolic scheme than that reported for somatic cells. In general, the Q0 of the erythrocyte is about 20 times lower 2 than that of somatic tissue. The avian nucleated red blood cell has a 002 some 10 times that reported for the non- nucleated red blood cell. In general, the permeability of red cells to anions like Cl- is extremely high in comparison with general ionic permeability. Most of the energy for active ion transport by non—nucleated red cells is derived from anaerobic gly- colysis, whereas energy for active transport in nucleated erythrocytes is derived mainly from aerobic metabolism involving the tricarboxylic acid cycle. Because of the many differences in the permeability and metabolic activity of the two types of cells the possi- bility of tagging nucleated erythrocytes with chromium was investigated and the results compared with data for non-nucleated red cells. The establishment of a suitable procedure for tagging nucleated RBC's will permit the investigation of many physiological parameters of the blood and blood vascular systems of lower vertebrates. The lg vitro binding of chromium to the nucleated red cell of the painted turtle was compared with the chromium-51 distribution in the liver and kidney following ip_vitro incubation. These studies have given some insight into the possible routes of excretion as well as areas of con— centration of chromium in body tissues. LITE RATURE REVIEW Inorganic Chemistry 9§_Chromium Chromium has been known only since 1766 when Legmann first isolated "crocoite," the natural lead chromate (PbCrO4), from the Plomb range d§.Siberie. L. N. Vanquelin in 1797 proposed that this ore from Siberia was in actu- ality a new element and he named it "chroma,' derived from the Greek word for color. The metal itself has no color, but most of its salts are colored. Chromium, atomic num— ber 24, is a transition metal of the first long period, group VI of the periodic table and is characterized by its ability to function both as a nonmetal or metal. The chemical atomic weight is 52.01. Four isotopes of chromium occur naturally and have the following mass numbers and relative occurrence in nature: 50 (4.31%), 52 (83.76%), 53 (9.55%) and 54 (2.38%). The atomic radius of the chromium atom is 1.24 to 1.35 A. Chromium has many points of similarity with vanadium, molybdenum, manganese and tungsten and shows discontinuous changes in several properties which occur at temperatures 0 of 37 C., Young's modulus, internal friction, restivity and coefficient of thermal expansion (Sully, 1954). . . 51 . . Radioactive chromium, Cr , is usually dispensed for routine clinical use as sodium chromate, Na2Cr5104. Chromium—51 has a half life of 27.8 days,disintegrating primarily by K capture with transmutation to vanadium with an 8% emission of 0.32 mev gamma. There is no alpha or beta emission (Beierwaltes, Johnson and Solari, 1957). Chromium complexes and multivalent compounds complicate both the qualitative and quantitative analytical procedures. The oxidation of chromium III to chromium VI can be effected by relatively weak oxidants, such as the ferric ion, iodine and oxygen. Anhydrous chromium trichloride is insoluble in water but has a flaky structure with high surface area on which contaminating O and H20 vapor are 2 readily absorbed. CrCl3 is readily oxidized in air by: + __. + 4CrCl3 O2 2Cr203 6C12 Marshall (1960) showed that absorption of molecular oxygen by chromous salts provides one of the most efficient methods of producing anaerobic conditions. In the hexavalent state the salts of chromium may be considered as derived from the hypothetical chromic acid H2CrO4. Hopkins (1942) has presented a basic outline of the major chemical forms of this element as follows: V 1 Cl R t' Representative Ionic a ence ass eac ion Salt Nature 2 Chromous Basic CrCl2 Cation 3 Chromic weakly basic CrCl3 Cation 3 Chromite Weakly acidic NaCrO2 Anion 6 Chromate Acidic NaZCrO4 Anion 6 Dichromate Acidic Na2CrO7 Anion Udy (1956) showed that H2CrO4 is an acid and dissociates in two steps as follows HCO H++HCO- (1) 2r4—"’ r4 _. + = HCrO :=H + CrO4 (2) In dilute solutions reaction #1 is practically complete to the right. constant is 6 X 10'-7 H 0 gives: 2 Cr2 Reaction #2 is reversible and the ionization Solid dichromate when dissolved in + H O—.- 2 2H2CrO4 Usable methods of volumetric determination depend almost entirely upon the reduction of chromium valence from VI to III and reverse oxidation from a valence of III to VI as follows: — + HCrO + 7H + Cr2 + + 14H + +++ 3e «p-CI‘ +++ 3e-—~2Cr + 4 H20 + O 7 H2 According to Udy (1956) the usual method of oxidation in a basic medium is to use H202 or permanganate. In an acid solution the oxidation is by persulfate or perchloric acid. Reduction is accomplished by ferrous sulfate or potassium iodide. Of the many forms of chromium encountered in the above reactions, only the trivalent and hexavalent states show enough stability to exist in biological systems. Trivalent chromium is found in solutions at pH values below 7, while the hexavalent form is found in pure solution in much more alkaline media. Trivalent chromium, depending on the pH, exists as pure solutions, complex colloidal suspensions or flocculent precipitates. Metal—Protein Compounds A11 proteins are capable of binding metal ions. The stability and extent of the binding of the metal—protein compounds is varied greatly by the nature of the metal ion, the nature of the protein and the net charge on the pro- tein. According to Gustavson (1958), the transition group elements generally possess the greatest affinity for pro- teins and therefore form the most stable compounds. Posi- tively charged polynuclear chromium complexes such as ++ Cr/ Ocr\ fr4ci\so/ are held very firmly to protein and little if any disso- ciation is detectable. The chromium complexes are potent cross linking agents and in this way lead to formation of stabilized protein structures. The binding of basic chromium sulphate by collagen is important in the tanning of leather. Chromed hide resists the action of boiling water and proteolytic enzymes. Gustavson (1956) has assembled evidence which points to the carboxylate groups of collagen as the principal ligand groups to coordinate with Cr III. The anionic chromium complexes are attached to the e—amino groups of the lysine residues and the arginine residues but the complexes formed are not as stable as the normal carbo— hydrate type. Cationic chromium fixation to collagen causes its isoelectric point to shift towards higher pH values (pH 5 to pH 7.5) due to partial inactivation of carboxyl ions of the zwitter ionic protein. Non-ionic chrome fixa- tion doesn't affect the isoelectric point of the protein. Gustavson (1958) blocked 90% of the carboxylate groups in the collagen by esterification with methanol and O.lN HCl. The esterification caused a decrease in the degree of Cr III binding to protein. Gurd and Wilcox (1956) point out that after the primary binding of Cr III,presumably further coordination occurs on aging by which other protein groups are brought into linkage with the metal,and perhaps further hydrolysis of the metal complexes will lead to larger aggregates of metal atoms. The protein chains are tied together in highly stable chelate structures. The chelate structures which are formed by Cr III have donor atoms of oxygen. Such complexes exchange their donor groups with great reluctance. The bonds to the metal are considered to be largely covalent. Metallic cations are found intimately associated with proteins throughout biological systems. Most metallic cations which are found in living cells are found in bound form. Gurd and Wilcox (1956) point out that the trivalent metal ions, A1+++, V+++, Cr+++ and Fe+++ all bind tightly to oxygen in preference to nitrogen and have a strong tendency to form polynuclear complexes at low pH values. The side-chain groups such as carboxyl, imidazole, or sulfhydryl are usually of much more importance in binding the metal ions than the terminal amino and carboxyl groups. The positive and negative charges normally found on 10 proteins form a net oriented field which will hinder or accelerate the approach of the metal ion. Naismith (1958) believes that anionic chromium complex (potassium dioxalatohydroxoaquochromiate) forms cross— linkages between the electronegative carboxyl side chains, with coordination links being involved with collagen and the chromium complex. This has not been proven as yet because of the denaturation of the protein during attempts to identify its reactive groups. Gustavson (1956) looks at the tanning process as an electroneutralization of the charges of the carboxyl groups into the chromium complex, which results in the formation of coordinate-covalent bonds of great stability between the carboxyl and the chromium atoms of the complex. In an excellent review of interaction of Cr III and Cr VI with proteins Grogan and Oppenheimer (1955) state that much of the "protein binding" as determined by dialysis techniques may represent the formation of chromium olate complexes that are practically nondiffusible and approach the colloidal state. This is particularly true of the chromic chloride. The CrCl3 when injected into rats and rabbits was distributed in a pattern similar to that of small sized, slow clearing colloids. 11 The Interaction 9§_Chromium Compounds with Biological Fluids Hexavalent chromium remains soluble, whereas trivalent chromium is relatively insoluble at normal body pH. Michael (1939) reported that anionic hexavalent chromium precipi— tated plasma proteins on the acid side of their isoelectric point and cationic trivalent chromium on the alkaline side. This precipitation is reversible and proteins are not changed. The denaturation of proteins by chromium salts on the acid side of their isoelectric point has a marked tem- perature coefficient, proceeding rapidly at body temperature and therefore may have physiological significance. Quanti- tative estimations of the protein-chromium precipitation showed that the protein combines in a definite stoichiometric proportion. Cr VI is reported to be reduced to Cr III by contact with beef plasma albumin at pH 7 (Michael, 1939). In spite of the fact that traces of chromium are almost universally present in plants and animals there is no evidence at present to indicate what role, if any, chromium plays in either plant or animal physiology. Gordon and Thompson (1936) studied the connection be- tween tanning action and opsonization by using washed guinea pig leucocytes and staphylococci in saline. CrCl3, chrome 12 alum and potassium bichromate all acted as opsonins in inducing phagocytosis. The tanning reactions described above will occur on the surface of the organisms and bring about changes which allow for phayocytosis by the leuco- cytes (Gordon and Thompson, 1937). Studies by Horecker, Statz and Hegness (1939) showed that aluminum ions in very small concentrations accelerated the aerobic oxidation of succinate (succinic dehydrogenase cytochrome system). Trivalent chromium can be substituted for aluminum in accelerating the aerobic oxidation. The action of aluminum and chromium can not be satisfactorily interpreted. Titanium with an atomic number of 22 and vanadium with an atomic number of 23 have been shown to have varying actions on certain biological oxidations. Bernheim and Bernheim (1939) investigated the effects of elements with numbers from 24 to 28 by addition of their salts to a suspension of rat liver. With the exception of manganese they had very little effect on O uptake. Mn (M/20,000) 2 caused definite inhibition of 02 uptake of rat liver but not of kidney and brain. In a study by Meldrum and Roughton (1934) the properties of carbonic anhydrase were not affected by CrC13. 13 The activation of phosphoglucomutase by metal ions was reported by Stickland (1949). The greatest activity of phosphoglucomutase was found only in the simultaneous presence of three active ingredients: hexosediphosphate, Mg++, and a second metal (A1+++, Cr+++, Pb+++, Fe+++ or CeH+). Consideration of the affinities then suggested that if the two-metal activation has any physiological importance then Cr+++ and Mg+++ are the metals concerned. Perlman (1945) added K2Cr207 to Aerobacter aerogenes and found that levels of 10 ug./1. resulted in an increased synthesis of pantothenic acid. Chromium alone increased the synthesis of biotin. Herrman and Speck (1954) found that the chromate- treated homogenates showed decreased extractibility of nucleic acids. The effect of chromium is specific insofar as treatment of tissues with other acids such as trichlor- acetic acid, metaphosphoric acid, picric acid or Bouin's solution doesn't decrease extractibility of the nucleic acids. It seems probable that a small portion of chromate is reduced as is evidenced by the greenish color of the residue and in this state it forms complexes with nucleic acids or nucleoproteins, which are more resistant to the decomposition of trichloracetic acids. Both RNA and DNA 14 at pH 7 are precipitated immediately by CrCl Two moles 3. of chromium and 1 mole of acid react. Cr VI compounds at pH 7 had no precipitating effect. Wacker and Vallee (1959) recently isolated from beef liver a ribonucleoprotein containing 0.1% chromium. A 20,000X increase in chromium concentration occurred in the ribonucleoprotein fraction as compared to the whole liver homogenate. The chromium may play some part in the function of the nucleic acids either in maintenance of structure or specific sites of binding between proteins and nucleic acids. The high chromium content of liver was first noted in connection with glutamic dehydrogenase fraction from beef liver. Investigation into a wide variety of phyla and species (Wacker and Vallee, 1959) showed that all of them contain large amounts of some metal ions in the nucleic acids. Curran and Azarnoff (1961) presented data that showed that manganese markedly increases the incorporation of radioactivity (Cl4 labeled acetate):hfix>cholesterol and slightly into fatty acids. Chromium greatly stimulates incorporation into both cholesterol and fatty acids. Schwarz and Mertz (1961) showed that chromium III is necessary for the maintenance of a normal glucose tolerance, 15 which appears to indicate that it may be a bioelement essential for mammalian organisms. As such, the impair- ment of glucose tolerance due to Cr III deficiency does not constitute a very serious danger to animals. However, the phenomenon resembles the disturbances of tolerance in diabetes and other diseases. Mertz, Roginski and Schwarz (1960), using fat bodies of rats, showed that at low levels of insulin the supplementation of chromium greatly increased the uptake of glucose. Toxicology g: Chromium Hexavalent chromium is a very strong oxidizing agent and for this reason it is difficult to believe that it could exist in the living cell as such. In the presence of organic matter, whether in the living cell, the plasma or alimentary canal reduction of Cr VI is inevitable. It is possible that hexavalent chromium might upset the fine balance of the oxidation-reduction enzymes and hydrogen carriers, but such an effect needs investigation (Water Research News, 1960). The most striking chronic and acute toxic effects have been noted largely with the Cr VI compounds and have been attributed to the high solubility, diffusibility under physiological conditions, escharotic, and oxidizing 16 properties (Grogan and Oppenheimer, 1955). The apparent innocuousness of Cr III compounds is probably due to their insolubility in biological fluids. The lack of activity of chromium in animal metabolism has become an asset to workers studying such varied sub— jects as feed utilization, blood volume, and tissue half life. Chromium compounds may well prove to be valuable tools in the study of animal and plant physiology. Udy (1956) reports only a slight stimulatory response in plants with low levels of chromium and therefore the lack of evidence for nutritional requirement of this element would suggest that there is little justification for incorporation of special chromium containing materials in fertilizers. The latest standard for drinking water (U. S. Public Health Service, 1946) specifies a maximum of 0.05 p.p.m. by weight of hexavalent chromium. No standards for tri- valent chromium compounds in food or water have been published. Partington (1950) cites many cases of suicidal poisonings with oral doses of potassium dichromate. Acute gastro-enteritis, blood and albumin in the urine, vomiting, and gastric hemorrhage all occur within 10 minutes after the poisoning. 17 The retention of dichromate and H SO by glassware, 2 4 as studied by Henry and Smith (1946) revealed as much as 0.01 ug./ml. chromium after 10 rinsings in distilled water. A.95% inhibition of enzyme urease is reported for a 1-10 ug./ml. chromium solution. In synthetic media 10 Ug./m1. chromium is very toxic to the growth of Staphylococcus aureus. A ten-fold increase in the chromium concentration is needed to obtain equivalent inhibition when the organ- ism is cultured in a nutrient broth. This is undoubtedly due to binding of heavy metal ions by the nutrient broth. Because of the extreme difficulty of ridding glassware of dichromate after cleaning in cleaning solution, Henry and Smith (1946) deemed it advisable to use another method for cleaning all glassware that was to be used in experiments with living cells or enzyme isolations. Richards (1936) found that only 50% of Amblystoma hatched normal larvae when the bichromate concentration was 0.001 ug./m1. A diatom Raphidium was injured by 1.0 ug./ml. of bichromate. 0.1 ug./ml. was sufficiently toxic to invalidate experiments made with yeast cells Oscillatoria. Cann, webster and Johnson (1932) found milk contained up to 0.5 p.p.m. chromium from storage tanks. They found that chromium in the food even to the extent of 100 p.p.m. 18 had no deleterious effects on the general health and repro— duction capacity of the rat. Orally administered chromium is not retained by the rat but is promptly and completely eliminated by the gastro-intestinal tract. Akatsuka and Fairhall (1934) reported chromic salts such as chromic carbonate and chromic phosphate are not poisonous to cats whether introduced through the digestive tract or the respiratory system. The derivatives of chromic acid (lead chromate) certainly exhibit toxic properties. Samitz and Pomerantz (1958) showed that chromates can be absorbed by the skin of guinea pigs:and no evidence that nickel salts were absorbed. Sodium lauryl sulfate produced local irritation on the skin. MacKenzie EE.§L- (1958) administered Cr VI and Cr III in the drinking water of rats. The Cr VI was absorbed about 9X more than the Cr III. Drinking water containing 25 p.p.m. chromium produced no toxic symptoms in one year. No differences were found between the controls as to water intake, food consumption, weight gain or blood picture. Oral ingestion of Cr VI compounds may lead to intense irritation of the gastro—intestinal tract resulting in violent epigastric pain, nausea, vomiting, severe diarrhea, and hemorrhages. Severe poisoning can be produced in 19 animals by intravenous and subcutaneous injections of these compounds. Gross and Heller (1946) found that when potassium and zinc chromate were added to the feed or drinking water of animals, the chromium was excreted in the feces in an insoluble complex form, possibly bound with proteins. Potassium dichromate was reported to be more readily ab- sorbed than CrCl3 when injected subcutaneously in animals. Kidney lesions in man and animals have been reported more frequently than any other type of systemic damage. The nephritis involves chiefly the tubules, with destruc- tion of the epithelium. Singh and Singh (1946) experi- mented with CrCl3 ip vitro and indicated that the toxic doses resulted in tonic contraction of unstriated muscle, while lowered concentration led to inhibition. MacNider (1924) stated that the chromate induced renal injury is primary tubular and affects particularly the epithelium of the convoluted tubules. Late in such a nephritis the glomeruli participate in the injury. A rapid reduction in urine formation occurred along with the inability of the kidney to eliminate phenolsulphonphtha- lein, and a retention of chlorides, blood urea, non- protein nitrogen and creatinine. The chromate nephritis 20 is associated with the development of an acid intoxication which is the result of the retention phenomenon. Hunter and Raberts (1933) reported the following order as experimental nephrotoxins: uranium nitrate, mercuric chloride and potassium bichromate. Because of the local irritating and corrosive properties and violent gastro- intestinal manifestations, bichromate is not too good. They believe that chromium is a pure tubular poison, of the first and middle division of the proximal convoluted tubules in the monkey. Ophuls (1911) reported sediment in the uriniferous tubules, with hyperemic kidney, albuminuria and necrosis of epithelium following subcutaneous injec- tions of potassium bichromate in the guinea pig. He also found that the rabbit was more susceptible than the guinea pig to experimental nephritis produced by subcutaneous injections of chromates. Palmieri and Mele (1960—1961) reported that injections of 10-20 mg./kg. doses of Cr (804)3 into the dog produces 2 conspicuous lowering of the blood pressure. The dose of 100 mg. caused a rapid and marked hypotension which, following convulsions, terminated in the death of the ani— mal. In all the animals there was a diminution of the 21 amplitude of cardiac contraction. The reduction of ampli— tude is probably from an action directed on the myocardial fibers similar to action of mercury, cobalt and vanadium. The most common effects of chromium compounds on industrial workers are those which result from direct con- tact of the skin with the chromates or with chromic acid. The skin reactions which are of two different types--chrome ulcers and chrome dermatitis--have been described frequently in the medical literature. No cases of cancer of the skin have been reported as a result of exposure to chromium or any of its compounds. Common effects which result from the inhalation of chromate dust or chromic acid mist are 1flceration.and perforation of the nasal septum. Broncho— genic carcinoma is believed by most authorities to be attributed to the carcinogenic action of hexavalent chromium. Some believe the carcenogenic property of the chromate dust is due to its irritating effect. Others believe that the chromates are not carcinogenic pe£_§§J but that they act on some of the organic substances in the body,leading to the production of a carcinogenic agent. Amounts of chromium needed for cancer production are not known and negative results on attempts to produce lung cancer in animals with chromium have been reported many times. 22 Hueper (1958) states that the carcinogenic effect of chromium might depend upon an adequate and prolonged release of biologically more active hexavalent chromium from pul- monary deposits of inhaled biologically rather inert tri- valent chromium compounds. The total available epidemio— logic, chemical and experimental evidence, nevertheless, favors the concept that chromium represents the carcino- genic agent. It has been demonstrated by Baetjer, Damron and Budacz (1959) that men employed in the chromate chemi— cal manufacturing industry have higher rates of broncho- genic carcinoma than those in a comparable control group. Since the mechanism of cancerization is thought to involve sublethal genetic transformation or mutations, the site of such changes should be in the cell nucleus, pro- ducing direct action of these agents on the genetic nucleoprotein materials of the cell. Grogan (1958) showed that repeated intravenous administration of Cr VI resulted in penetration of chromium into the platelets and leuko- cytes as well as the erythrocytes. Repeated I.V. doses also resulted in Cr VI appearing in the nucleus of erythrocytes. Fairhall (1957) stated that the death rate due to cancer of the respiratory system among exposed employees 23 was 21.8%. This is 16X the expected rate. Distribution and Retention 9: Chromium ig_the Organism Baetjer, Damron and Budacz (1959) in an extensive paper described the distribution and retention of chromium in men and animals. I.P. injections of Cr VI in the guinea pig results in more chromium being excreted and more appeared in the blood than with the intratracheal route. With the trivalent form the higher blood level was due to a greater plasma value. No chromium was found in the bones one to three days after the intratracheal injection of chromic chloride or up to 90 days after injection with sodium chromate. The lungs retained trivalent chromium not only to a greater extent than the hexavalent form but also for a much longer period. When water-soluble chromates were injected intratracheally, only about 15% of the in— jected dose was recovered from the lungs. Twenty per cent was found in the blood and another 5% in the liver, kidneys and spleen altogether. Slices of human lung bind Cr VI to a lesser extent than lung tissue of guinea pigs. Both lung and liver show the same binding for CrCl with no species 3 differences. Lungs do not contain any components which 24 bind chromium to a greater extent than other tissues and lung of man does not bind Cr III differently than lungs of other animals tested. wennesland _t_§l, (1957a) reported a slight accumula— tion of Cr51 in the spleen and lung tissues of the rabbit. Wennesland §t_§l, (1957b) found no evidence of Cr51 accumu— lation in spleens of dogs killed with sodium pentobarbital. The longer the storage time of the tagged RBC the greater will be the excess of Cr51 found in the lung, liver and spleen, particularly if the cells have been stored in saline. This is believed to be a function of the reticulo- endothelial system. Sequestration of cells in the lung may be analogous to trapping in the organ of leukocytes or macromolecules. The pulmonary circulation was found capable of removing and delivering vast quantities of leuko- cytes from the blood, as well as storing and destroying them. Bierman _§_§l. (1956) also reported that the cells may go through the pulmonary epithelium into the alveoli and be lost via sputum or G.I. tract if the sputum is swallowed. They may be destroyed here (lungs) and form granulocytes iflate-like bodies." The effects of tempera- ture on the uptake of colloidal CrPO by isolated rat liver 4 (Brauer, Leong and Halloway, 1957) has a 010 for the 25 overall reaction rate equal to 1.92 between ZT%2.and 380C. The uptake can be represented as an irreversible reaction between the colloid micelles and an effectively unlimited fixed active phase, presumably the surface of'UuaKupffer cells. Knoll and Fromm (1960) found that in rainbow trout exposed to 2.5 mg. Cr/l. in tap water accumulation of chromium in concentrations exceeding the environment in the spleen, posterior gut, pyloric caece, stomach, and kidney. All these organs were believed to be correlated with excretion. The major route of entry of hexavalent chromium into the trout is probably the gills. They stated that the nucleated fish erythrocytes behave dif- ferently than the mammalian red blood cells in that they showed little uptake of Cr VI. Ingrand (1961) concluded that the distribution of Cr51 in the mouse after I.V. injection of radioactive sodium chromate obtained by autoradiography showed a localization in the liver, in the spleen and in the skeleton at the level of the zones of osseus repair. Ehrlich mouse ascites carcinoma cells were labeled with radiochromate and the valence state of the chromium and the intracellular distribution studied by Rajam and Jackson 26 (1958). Seventy-one per cent of the total intracellular Cr51 was associated with the soluble cytoplasmic fraction; 52% was associated with the ethanol precipitated protein fraction. Hexavalent chromium appeared to be intra— cellularly reduced to the trivalent state. The cell mem— brane of the tumor cell is impermeable to intracellular Cr III ions. The stability of this intracellular label is not dependent on protein binding alone. Gray and Sterling (l950a, l950b),in their classical papers on the methodology of chromium tagging of red cells and plasma proteins of the human red cells, stated that: (1) Anionic hexavalent form of the isotope labeled red cells while the cationic trivalent form bound firmly to the plasma proteins. Ninety per cent uptake occurred by human cell in two hours when done ip_yit£g_in saline. (2) The site of tagging of the erythrocytes with Cr VI appears to be on the globin portion of the haemoglobin. (3) Haemoglobin demonstrated a significantly greater capacity for Cr III than for Cr VI. (4) It was suggested that the anionic hexavalent chromium diffuses through the red cell membrane and is bound by the haemoglobin. (5) There is probably a reduction of Cr VI to Cr III before this binding takes place. 27 The preceding discussion of the pharmacology of chromium—51 makes it clear that chromium forms a stable and nontoxic label for red blood cells if anionic hexa- valent chromium is used. Cationic trivalent chromium is a good label for plasma proteins. The rationale used for determining plasma volume, hematocrit, total blood volume or total red blood cell mass is presented by Beirwalters, Johnson and Solari (1957). The Council of Pharmacy and Chemistry of the A.M.A. (1955) stated that sodium radio chromate (Cr51) has not been shown to produce any signi— ficantly deleterious effects on normal erythrocytes either radioactive or metallic. The maximum dose at any one time was set at 390 uc. per person. Ebaugh _t_al, (1953) reported a 90% uptake of Cr VI by human cells at 390C. in 5 minutes using an ACD incuba- tion solution. ACD solution was given as trisodium citrate = 1.65 gm., citric acid = 0.60 gm., dextrose = 1.84 gm., diluted to 75 ml. A 50% uptake occurred in five minutes at 260C. compared to only a 10% uptake at 1.8OC. in five minutes with human RBC. They found no change in concentration of chromium salt ranging from 0.25 to 9.5 ugm. salt/ml. whole blood. At lower pHs higher uptake of 51 Cr was reported. Mollison and Veall (1955) pointed out 28 the possibility that some Cr51 might be bound to the small amount of plasma protein adhering to the surface of the red cell. The uptake is better with packed cells in ACD than with whole blood mixed with ACD. Von Ehrenstein and Zacharias (1958) reported that heparin decreases Cr51 uptake and that for human cells ACD was the best medium. Ca++ ions diminish the uptake of chromate. Ca++ ions also diminish the binding of Cr by haemoglobini_n y_i_t_r_o_. It is believed that the Ca++ is affecting other structures in addition to the cell membrane. Direct measurement of the rate of elution of chromium from red cells has been done by Smith and Krivit (1960). The average elution per day ig_yi§£g_of normal mature red cells was 0.2%. This low rate of elution was said to be due to the use of a phosphate—glucose-buffer solution in place of the normal saline solution. The addition of glucose decreased the rate of autochemolysis. Davson and Reiner (1942) stated that the permeability of the RBC to anions (C1_) is extremely rapid in compari— son to general ionic permeability. Possibly the speciali— zation of the membrane has occurred to permit the rapid diffusion of the negative ions. The predominantly lipid membrane represents an effective and selective barrier to 29 diffusion so that the cell may maintain an internal environ- ment different from its surrounding medium. Excretion of intravenously injected chromium in the dog is by way of the urine (Collins, 1958). Excretion in bile and feces is negligible. Glomerular filtration and tubular reabsorption are the two mechanisms involved in the renal handling of unbound chromium. Collins also found evidence for tubular excretion of dialyzable chromium, but stated this was probably of minor importance. I vivo reduction occurs after intravenous injection of Cr VI because no Cr VI is found in the urine. Results of this work indicate that chromium is excreted at least in part in organic combinations. Nucleated Erythrocytes Lyman (1945) showed that there was an anti—hemolytic action of calcium in the blood of the snapping turtle and that the blood can not be oxalated without inducing complete hemolysis. Dilution with NaCl also induces hemolysis. In . ++ . . other speCies of turtles Ca ions are of less importance as an antihemolytic agent. The calcium ion and other ions can alter the permeability of the cell membranes. For example, chromium acts by precipitating the protein of the cell wall 30 and the action is not comparable to that of calcium. Hamkid and Ferguson (1940) found hemolytic action of fluorides on certain dogfish nucleated erythrocytes. Oxalates and arsenates produced similar effects. All of the above chemicals removed free magnesium and calcium which in turn altered the permeability of the RBC. Maizels (1956) reported that the African tortoise has erythrocytes which transport Na+ and K+ and the energy for this transport is derived from oxidative metabolism. They cannot maintain their relative impermeability to Na+ and K+, unless calcium is present in the external medium. Metabolism and cation transport in human erythrocytes are based on glycolysis. 31 MATERIALS AND METHODS Experimental Animal The common painted turtle (Chrysemys picta) was selected as an experimental animal for the following reasons: (1) their morphology, gross anatomy and physiology have been explored, (2) sexual differences are relatively simple, (3) they are readily available during any time of the year, and (4) they are easily kept under laboratory conditions. The animals were obtained from a commercial supply house and stored at 120C. in stainless steel tanks containing water to a depth of 5 inches. One week prior to use as experimental animals they were kept at 260 C. in 26 gallon glass aquaria, each of which contained a wire platform just above the water level. Fresh tap water was placed in the tanks once every week just after feeding. Chrysemys picta were observed to feed only under water. The painted turtle is reported as omniverous in its natural environment but under the experimental condition they would eat only meat scraps. Leafy green vegetables were never consumed. The turtles were all at least 4-5 inches long and therefore considered to be mature. Only those turtles that were healthy and showed no signs of pathology other than that noted below were used in this study. 32 The general condition of the turtles was not affected by a slight infestation of leeches which were found cling- ing to the shell and skin. Like many of the lower verte- brates, turtles are parasitized by all of the major classes of protozoans but the relationships between these parasites and definite turtle diseases are not known. Many times during histological examination, examples of vascular parasitism were observed (see figure H), however, there was no indication of any tissue or organ pathology. Of interest was the high degree of pigmentation found in the liver. At first the pigment appeared to be hemosiderin, but by histo— chemical tests using Turnbull's Blue Method for hemosiderin (Gridley, 1960), negative results were obtained (see figure C). The technique using the bleaching procedure by Gridley (1960) revealed the pigment to be melanin (see figure D). In a general review of the literature it was found that melanin pigmentation of organs occurs very frequently in lower animals. The liver appeared, on gross examination, to be light red-brown, often pale brown. Histochemical tests revealed slight infiltration of adipose tissue and the presence of glycogen. In many cases the urinary bladder of the turtle was observed to contain considerable sediments. This has been reported to be without symptomatology (Kaplan, 1957). 33 One bacterial infection was encountered but was felt not to be of serious consequence. The infection first showed up as a white-brown film or paste around the head and the eyes. This, in advanced stages, spread over the neck and, in very advanced cases, was observed on the front limbs. In severe cases, before treatment, complete closure of the eyes resulted from build-up of pus and exudate. A routine bacteriological examination revealed that the animals were infected with Aeromonas hydrgphila. This organism was found resistant to Bacitracin, Erythromycin and Penicillin and susceptible to Aureomycin, Neomycin, Terramycin tetracycline etc. The infection was treated by adding, to the water in the holding tanks, injectable Liquamycin manufactured by Pfizer Laboratories. Liquamycin contains 10 mg./ml. oxytetracycline, of which 50 ml. were added to 150 liters of water once a week until the infection was cleared up. Breed, Murray and Smith (1957) stated that A. hydrophila was a 3 u motile red-shaped pathogenic bacterium found in frogs, salamanders, fish, mice, guinea pigs, rabbits and snakes. It causes hemorrhagic septicemia. The source prob- ably was the holding tanks, kept in the same room, used for the storage of frogs which were infected with red leg. 34 Optimum temperature for this organism was reported at 100 C.-150C. which included the holding temperature. Periodic treatment in sunlight(u1traviolet)light has been reported by Kaplan (1957) to aid in combating this type of bacterial infection. A few animals were found with tumors located on the neck. The pathology report from the Department of Pathology stated the tumor was a chronic, multiple abscess (figure E). Predominate tissue changes revealed that this lesion is characteristic of an abscess of long duration. A well developed connective tissue capsule around the abscess was found and some bacteria were also present. Figures F-J show additional histological sections from the turtle. Eggsue Preparation The animals were killed with a heavy blow to the head, the plastron removed, and blood withdrawn by direct canula- tion of the left aortic arch (see figure 1). After removal of the pericardium a ”bull—dog” clamp was placed as far distal as possible but still able to clamp the right aortic arch, innominate, left aortic arch and the pulmonary arteries, thus diverting the blood from the somatic circulation. 35 Figure l Ventral View of the Turtle Heart 36 - . ‘¥> Rt. aortic ' \‘(—_'Iu aortic arch arch \ - p1 ‘(_ u monary ' artery Innominate Rt. atrium . L. atrium Ventricle 37 A 22 gauge needle attached to a 6 inch piece of tygon tubing served as a cannula and blood was allowed to flow into a graduated centrifuge tube containing heparin as an anticoagulant. An average sample of 7 m1. whole blood was centrifuged for 10 minutes at 2000 r.p.m. and the straw-yellow plasma removed using a micro-suction apparatus. The standard tests for bilirubin (Method of Malloy and Evelyn, from Hawk, Oser and Summerson, 1947) were negative; therefore, it was concluded the yellow color was not caused by exces- sive destruction of erythrocytes or hepatic biliary unal— function. The plasma pigment was not identified. The cells were washed three times by resuspending them in 15 ml. Ringer's phosphate solution (see below), mixed and centrifuged. The washed cellsmere then diluted to the desired concentration. Ringer's Solution (cold blooded) NaCl 6.5 gm. KCl 0.1 gm. CaC12* 0.2 gm. Add glass distilled water to 1000 ml. *Dissolved first and then added to salt solution. Phosphate Buffer (pH 7.4) 80.4 ml. M/15 NaZHPO4 19.6 ml. M/15 KH2P04 38 Ringer Phosphate Ten volumes of Ringer's solution are added to 1 volume of M/15 phosphate buffer, pH 7.4 The tissue was removed and rinsed three times in Ringer's phosphate solution and tissue slices 0.5 mm. thick were used in all ig_vitro studies. Wet weights were to the nearest 0.1 mg. Radioactive Isotope Methodology Chromium-51 as CrSlCl3 in HCl solution was obtained from Oak Ridge National Laboratory in shipments of 15 mc. each. A representative shipment would have a concentration of about 80: 10% mc./ml. with a specific activity of 355707 mc./gm., total chromium content of 0.226 mg./ml. and the normality of the HCl solvent = 1.95. Conversion to hexavalent chromium was by the method as given by Schiffman, 1957. Approximately 50 ug. of chromium— 51 are made basic with 2 m1. of 6N NaOH and 1 ml. of 3% H202 added. The solution is then heated to 1100C. for 5 minutes, then boiled for 1 hour, thus allowing for the peroxide to be driven off. After 1 hour,0.2 m1. 6N NaOH were added while the solution was still hot. If bubbles wenagiven off this indicated that the peroxide had not all 39 been boiled off, therefore additional boiling was carried out until no excess H202 was present. The solution was next neutralized by addition of approximately 2.0 ml. 6N HCl. It was cooled and the hexavalent Na CrO 2 4 was ready to use. It was found that if during the neutralization that back titration was necessary, the chromium—51 might be precipitated out of solution due to the high ionic strength of the solvent. The detection apparatus consisted of the following components: a two—inch thallium treated NaI scintillation detector assembly, Model D85; a radiation analyzer Model 1810; a Model 183 scaling unit and a well counter——all manufactured by the Nuclear Instrument and Chemical Cor- poration, Chicago, Illinois. In all cases each individual experiment was done at constant geometry which was adjusted for best possible resolution. Counting errors were at the 5% level or lower. The activity in counts per minute (c.p.m.) was expressed as per unit mass. No attempt was made to analyze the chromium content of the tissue in order to calculate specific activi- ties. In most cases the chromium content approached the Ininimal detectible concentration using the microchemical analytical methods . 40 Because essentially only gamma radiation is counted by a scintillation tube, no correction for self—absorption was found necessary. Absorption of Cr—51 was found by Knoll (1959) to be insignificant under the geomentry as used in this work. Electrophoresis Figure 2 shows the apparatus for electrophoresis de- signed by the author and made of 1/4 inch plastic. It is versatile enough to perform regular filter paper electro— phoresis, starch gel electrophoresis and micro—electrophoresis with Oxoid Cellulose Acetate Electrophoresis Strips obtained from Consolidated Laboratories Inc., Chicago Heights, Illinois. The base of the tank is divided into four buffer com- partments. The two central ones, contained the electrodes of platinum wire. The two corresponding outer chambers are connected to the inner chambers by several holes drilled through the buffer baffles. Strips are suspended across the tank and filter paper wicks then complete the circuit into the buffer solutions. Strip holders are short rectangular plastic bars. A tight fitting cover insures efficient vapor saturation of the apparatus. The whole apparatus is placed in a refrigerator at 40C. during a run. 41 Figure 2 Apparatus for Electrophoresis 43 For protein separation the cellulose acetate strips were saturated with barbitone buffer, pH 6.8 Sodium diethyl barbitone 10.0 gm. Sodium acetate (hydrated) diluted to 1 liter 6.5 gm. A Hamilton Microliter Syringe is used to apply a sample to the strip, already impregnated with buffer, and then the strip is placed in the apparatus. At least 150 V. should be applied and the current should not exceed 0.4 ma. per cm. width. The D.C. power supply used was a constant vol- tage Heath Kit #PS-3. After an appropriate time interval the strips are removed and dried by being suspended in a hot air oven for 10-30 minutes at 80°C.-1000C. They are next stained by floating on the surface of a 0.2% Ponceau-S stain in 3% trichloracetic acid for 10 minutes. In the case of poor penetration of the stain a few drops of methanol are used to wet the back of the strip, which is then restained. Destaining was accomplished by washing in several changes of 5% aqueous acetic acid. CataphoresislApparatus The general phenomenon, called electrokinetics, resides in the motion of two electrically charged surfaces or bodies relative to each other. It is important to note 44 Figure 3 Northrup-Kunitz Cataphoresis Apparatus 46 that electric mobility is due to charges at or near the surface of the particles or molecules. ‘It is a function of the surface charge density, i.e., the number of charges or charged groups per unit area of the surface. Velocity of motion of particles in an electric field is, among other factors, constant; a function of (l) the particle's own charge, (2) current strength (potential or gradient), and (3) resistance to motion offered by the medium (viscosity). The net charge on a protein membrane is due essentially to ionization of carboxyl and hydroxyl groups. The magnitude of the charge depends upon the pH of the solution. Electrophoretic mobility, u, is defined experimentally, as the distance (in microns) transversed by the particle under observation per u in time (sec.) with a voltage gradient (v./cm.). It follows that u has the dimension of u/sec./v./cm. The apparatus used was the Northrup—Kunitz Cataphoresis Apparatus (figure 3), purchased from the Arthur H. Thomas Company, Philadelphia. Power supply was the Heath Kit PS-3. Procedure for calibrating the apparatus was taken from Technological Service, A. H. Thomas Company, 1955. 47 Paper Chromatograms A simplified polychromatic technique as developed by Moffat and Lytle (1959) was used to study the amino acid composition of protein hydrolysates. In all cases the solvent system used was butyl alcohol-acetic acid-water (4:1:5). Whatman No. I filter paper in strips 2.5 cm. wide were used in unidimensional descending chromatographic chambers. Strips were developed for 10—16 hours, the time needed for solvent front to migrate 20 or more cm. The strips were then removed, dried in an oven for 5 minutes at 104OCL-1100C, sprayed with the ninhydrin-cupric nitrate indicator, and placed in an oven for 1.5 to 2 minutes at 105°C. The N—CN indicator consisted of two solutions I + II. Solution I contains 0.2% ninhydrin (anhydrous 1,2,3 triketohydrindene) in 50 ml. of absolute ethyl alco- hol, 10 ml. of glacial acetic acid, and 2 ml. of 2,4,6 collidine. Solution II was a 1.0% solution of cupric nitrate trihydrate in absolute alcohol. Solutions I + II are combined in a ratio of 25 to 1.5 just before using. Preparation Qleemoglobin and Globin A stock solution of HbO2 is prepared from red cells washed as stated above. White, Beaven and Ellis (1956) 48 give the following method for preparation of hemoglobin solutions. The washed cells are mixed with one volume of water and repeatedly frozen and thawed until laked. The stroma is separated by high-speed centrifugation at 15—20X 103 G for one hour at 5°C. This concentrated solution (approximately 10 g./100 ml.) keeps well at 5°C. in small full bottles. Lemberg and Legge (1949) formulated the method used in the preparation of globin. Washed, laked corpuscles (HbO2 solution from above) was cooled to 0°C. and then added gradually to a ten—fold volume of acetone containing 1%»hydrochloric acid also cooled to 0°C. The mixture is allowed to stand for two or three minutes and then is filtered; the acid mixture is washed several times with suction and allowed to dry. All operations are carried out at low temperatures. Block and Bolloing's (1951) article on the preparation of hydrolysate of proteins gives the following methods. A sample of protein containing 1.6 mg. of nitrogen is hydro- lysed under reflux with 10 ml. of 1N HCl for 20 hours. The excess HCl is removed by evaporation to dryness lg yaggg at 35°C. or on the stream bath, and the resulting thin film of amino acid hydrochlorides is placed in a vacuum desiccator 49 over soda lime for 24 hours or longer. The hydrolysate is then taken up in warm water, filtered, again evaporated to dryness, and finally taken up in exactly 1 ml. 10% isopropanol. This solvent is used because it is an ef- fective preservative and yet does not cause esterification under these conditions. Autoradiography Histoautoradiographs were prepared in hopes of locali- zing chromium-51 within cells and organ systems. The excellent book by Boyd (1955) gave many ideas on how to accomplish this end. Chromium-51 has as its only major radiation a weak gamma ray of 0.32 mev energy. For this reason a fairly sensitive film must be used. Also a film of fine, closely packed grains is necessary for the fine resolution needed with histoautoradiography. For the histological sections the most sensitive of the Kodak nuclear tracking emulsions (NTB-3) was first utilized. Unfortunately this material combines its sensi— tivity to gamma radiation with its sensitivity to virtually all other types of radiation; thus it was observed to pick up undesirable background too rapidly. 50 Figure 4 Multiple-unit Constant—pressure Microrespirometer. Photograph Courtesy of E. P. Reineke. m... .1: l 1 i l 1- j 52 Next, both the Kodak Autoradiographic Stripping Plates AR-lO and AR-50 were used (available from Kodak Limited, London, England), but only the AR—50 proved satisfactory. See Appendix for detailed description of the techniques used in application, development and staining the histo- autoradiograms. For the localization of the chromium-51 on the filter paper following electrophoresis or paper chromatography, one of the high speed x-ray films was employed. Kodak No-Screen Medical X-ray Film produced satisfactory reso— lution. Development of the plates followed the directions set forth in ”The Fundamentals of Radiography" by Eastman- Kodak Company (1960). Cellular Respiration Figure 4 shows the apparatus used to measure the oxygen consumption of erythrocytes, liver and kidney tissues of the turtle. The operation of the "multiple-unit constant- pressure microrespirometer," designed by E. P. Reineke, is described in detail by Reineke (1961). The center well of a standard Warberg flask contains potassium hydroxide (0.1 ml. of 10% KOH) and a filter paper wick. The tissue is placed on ice after removal from the 53 animal. One hundred mgms. (wet weight) of tissue slices are placed in 2 ml. of Ringer's phosphate with 50 mg./m1. glucose contained in the main flask. The tissue was sliced to 0.5 mm. thickness with a Stadie Riggs hand microtome. Samples of each tissue were dried to constant weight at 950C. to determine the percentage dry matter content. Erythrocytes were counted with a hemocytometer and the metabolic results based on oxygen consumed per 109 cells. All tissues were run at 370C. with constant agitation. Liver and kidney tissue maintained their oxygen consumption for over 1.5 hours. The erythrocytes held a constant meta- bolic rate for over 3 hours. After a normal control value for Q0 was obtained for each flask, a known 2 amount of chromate was added from the side arm. 002 measurements continued and calculated values compared with the control values. 54 RESULTS Ip_Vitro Chromium Accumulation A number of physical factors have been reported to affect the ip_vitro rate and total amount of chromium-51 uptake by the non-nucleated erythrocytes. “Ethin the physiological temperature range, an increase in tempera- ture will increase the uptake rate. Some anticoagulants are reported to affect chromium accumulation and an ACD solution (trisodium citrate, citric acid and dextrose) is the recommended medium for chromium tagging of human cells. The undesirable reducing effects of glucose on hexavalent chromium would make it seem desirable to keep the level of glucose as low as possible. It is also reported that glucose is needed for maximum uptake of hexavalent chromium by the non—nucleated erythrocytes. Heparin was the most effective anticoagulant to use on turtle blood and it did not interfere with the uptake of chromium by the nucleated erythrocyte. Concentrations of chromium-51 ranging from 0.25 to 9.5 ug./ml. have been shown not to affect the uptake rate by non-nucleated erythrocytes. The effects of different media on the accumulation of 51 hexavalent chromium (NaZCr 04) by the nucleated erythrocytes TABLE 1 EFFECT OF DIFFERENT MEDIA ON UPTAKE OF BY NUCLEATED ERYTHROCYTES 55 Na Cr510 Treatment % Uptakea 0.6% NaCl 0.6% NaCl + 0.5 mg. heparin/10 ml. 0.6%»NaC1 + 0.5 mg. heparin/10 ml. + 1 mg. Ca++/ml. ++ 0.6% NaCl + 1 mg. Ca /ml. Ringer's solution + 0.5 mg. heparin/10 ml. 12.9 13.5 25.3 25.5 16.2 a2 hours; 21.50C; phosphate buffer pH 7.4 b . . . . . Rapid agglutination, slight hemolySis 56 were investigated. Washed cells were added to 0.6% NaCl and the uptake at the end of two hours at 21.50C. was determined and compared with the total amount of activity in the media. The media were modified by addition of reagents as shown in table 1. After the two hour incuba— tion period the cells were removed and washed 3 times with 0.6% NaCl solution. Washed, tagged cells were then placed in plastic containers for counting in the well scintilla- tion counter. In both cases where Ca++ ions were added (table 1) to the media, very evident agglutination occurred within 25 minutes. Statistical analysis revealed that cells in Ringer's solution had a greater uptake than those in NaCl alone. Heparin had no significant effect on the chromium—51 uptake. Figure 5 summarizes the results of comparison of the basic Ringer phosphate buffer media with ACD and Ringer phosphate glucose media. In both experiments (#1 and #2) addition of glucose at a concentration of 50 mg.% (normal blood level for the tur- tle). no statistically significant effect was observed at the 5% level. Use of the recommended ACD solution, adjusted to the osmotic concentration of turtle erythrocytes resulted 57 TABLE 2 ELECTROPHORETIC MOBILITY OF NUCLEATED ERYTHROCYTES WHEN TAGGED WITH Na CrO 2 4 Chromium Mobility (u/sec./cm./volt) Animal No. ::::::;a (mg.Cr VI/lJ Control Treated + + d Turtle 24 20 2.507 _ 0.387 1.765 _ 0.484 Turtleb 24 20 1.668 t 0.564 1.892 t 0.407 + + d Turtle 23 20 2.079 _ 0.326 1.578 _ 0.305 + + d Turtle 23 20 1.566 _ 0.254 1.383 _ 0.213 Turtle 23 20 1.280 i 0.248 1.123 t 0.181C + + d Turtle 23 20 1.606 _ 0.444 1.209 - 0.275 Turtle 23 10 2.446 1 0.385C + C Turtle 23 20 1.968 _ 0.333 + C Turtle 23 40 1.734 _ 0.265 aIncubated for 30 minutes at 200C; then washed. bStored in Ringer phosphate glucose for 24 hrs. at 2°C., before using. c . . . 5% Significance. d1% significance. 58 Uptake of Na 2 Cr5 10 4 Figure 5 by Nucleated Erythrocytes % Uptake 601' 50d 40 l 30 20- 10«- 59 MUOL‘DID' A B C D E Experiment #1 Ringer Ringer Ringer Ringer ACD phosphate phosphate phosphate phosphate A B C D E Experiment #2 + 50 mg./100 cc. glucose + glucose + 10_3M NaF + glucose + 10‘3M NaCN 60 in a statistically significant decrease in chromium accumu— lation compared to uptake from the Ringer phosphate glu- cose media. Two of the common metabolic inhibitors were added to the incubation media at a concentration of 10—3M. At this concentration complete inhibition of enzyme systems affected would normally be expected. NaF (10-3M) had no effect on chromium accumulation in either experiment. NaCN, on the other hand, did cause a significant decrease in chromium—51 accumulation. In the tubes containing NaCN, modification of the cell wall took place because of the accumulation of free hemoglobin indicating hemolysis. At the end of the 2 hour incubation period cells in the Ringer's phosphate media contained a higher percentage protein—bound chromium than those in the ACD media. Electrophoretic Mobility The electrophoretic mobility of nucleated erytyrocytes was measured at a temperature of 210C.in Ringer phosphate glucose buffer at pH 7.4 (table 2). Values are expressed as u/sec./cm./volt and were all made at the level of 0.201 times the depth of the cell. At this level automatic cor- rection for electroendosmotic flow is made. 61 Control values indicate that the erythrocyte's membrane has a net negative charge or zeta potential. Incubation of the cells with NaZCrO4 for 30 minutes at 21°C. followed by washing, then resuspension in fresh buffer, always caused a significant decrease in the electrophoretic mobility. The nucleated erythrocytes, being more dense (i.e., higher specific gravity), tend to settle or sink in the Ringer phosphate buffer much faster than human erythro— Cytes. For this reason one must be very careful to measure the velocity of only those cells that are at the proper level in the electrophoresis cell. Many artificial methods of increasing the viscosity of the medium were tried, including addition of sucrose, albumin, gum acacia etc., but no satisfactory media could be found. In most cases when the viscosity was high enough to significantly slow the settling of the cells the osmotic balance of the media was upset, causing hemolysis. Temperature Results of the effects of temperature on the uptake rate are presented in table 3. It was found that the slope of the lines, fitted by the method of least squares, have 62 values that are not significantly different from zero for temperatures of 20C. and 540C. The coefficient of corre— lation also was not significantly different from zero for these temperatures. The values for a (i.e. the Y inter- cept at time zero) show a generally increasing value with increasing temperature. The QlO (temperature coefficient) was calculated using the equation: log 010 = (lO/tl-t2)(log kl/kZ) where k is the rate at temperature 1 1 k2 is the rate at temperature 2 Distribution of Na2CrSlO4 in the Nucleated—Erythrocytes The procedure for this section was a modification of that presented by Rajam and Jackson (1958). Cells pre- pared as presented above were incubated at 370C. and samples removed and cells washed after 4, 60 and 120 minutes. The cells were hemolyzed by freezing and thawing. After hemolysis a sample was centrifuged, the cell walls and nuclei removed, and their chromium content determined. It was found that on hemolysis the soluble cytoplasmic proteins were released and the cell wall then closed in around the INFLUENCE OF TEMPERATURE ON THE UPTAKE OF Na2Cr 0 TABLE 3 63 51 4 BY NUCLEATED ERYTHROCYTES Temp. 1 l 0C. No. a b rxyz Q10 5 5 3234 73* 0.68* 14 5 3392 174 0.96 ( 2-14) 1.21 20 5 4495 241 0.96 (14-20) 1.17 37 5 2970 783 0.99 (20-37) 1.20 54 5 6630 36* 0.20* 1 Y = a + bx 2 . . . CoeffiCient of correlation * Not significantly different from zero 64 nucleus. The soluble proteins were precipitated by addi- tion of ethanol (final concentration = 50%). The preci- pitated proteins in each case were separated by centri— fugation at 900 G for 15 minutes and the chromium—51 content determined as before. The protein—free supernatant was counted to determine the total free (non-protein bound) chromium. A portion of this final non—protein supernatant was used for the purpose of determining the valance state of the intracellular chromium. To the supernatant 1 ml. of a 5% CrCl3 solution (non-radioactive) was added and mixed. The Cr+++ and the Cr51+++ ions reach equilibrium and both ions are precipi- tated with ammonium hydroxide. This precipitate was washed with distilled water, removed by centrifugation and counted. The data are summarized in figure 6. Values reported for cell walls, bound chromium, and Cr III were obtained from 6 observations each. The other data were calculated by subtraction from an assumed value of 100% for the total activity. Statistical analysis revealed that the activity on the cell walls did not change over a 2 hour period. The amount of free Cr III decreased and the amount of bound Cr increased, both at the 5% level of significance. 65 Effects p§_metabolic Inhibitors pp Uptake of NaZCr51O4 by Turtle Tissue Various inhibitors were added to a Ringer phosphate medium containing cells and chromium and their effects on the hexavalent chromium uptake were determined. Five observations were made over a period of eight hours. Table 4 gives a summary of the effects of these treatments on the slopes of the fitted lines. Those treatments having a significant effect are so indicated in the table. The NaCN (10—3M) treatment caused hemolysis and agglu- tination of the erythrocytes. A concentration of 10-3M is considered to be high enough to inhibit the major per- centage of the affected enzyme systems. Electrophoresis and Paper Chromatograms In figure 7 several of the filter paper electrophoresis strips of hexavalent and trivalent chromium with a radio— autograph superimposed are shown. The dark bands located on strips #5 and #6 are caused by the exposure of the over- laying x-ray plate. All other strips are either CrCl or 3 NaZCrO4 samples. The non-radioactive chromium ions were present in high enough concentrations to be detected by their own natural colors. Strips #2, #3, #4 and #7 are all of trivalent chromium. Strips #1, #6 and #9 were of Figure 6 . . 5 . Distribution of Na2Cr 1O4 in the Nucleated Erythrocyte .m> Hawk HHOUAHV H> Holey .m> HHH uoxmv no mmumxev.m> so ocsomxmv. UHEmmHmouxoxmv .58 03 oo m oma om m oma om m e l N l IJ. N N o v 4 .r1 6 m m . F r.. . m .l: m M! EFL. m a mL OH om on 13 IEQOL % TABLE 4 68 EFFECT OF METABOLIC INHIBITORS ON UPTAKE 51 OF NaZCr 04 BY TISSUE . . Eff t Tissue Inhibitors Media ec on Slope . 1 Kidney R.P. Control . 2 Kidney R.P.G. N.E. Kidney Boiled (4 min.) R.P.G. Decrease** . 1 Liver R.P. Control Liver R.P.G N.E. Liver Boiled (4 min.) R.P.G. Decrease** Erythrocytes R.P.M. Control Erythrocytes Saponin R.P.G. N.E Erythrocytes HgCl (lo—3M) R.P.G. N.E —3 Erythrocytes NaAzide(10 M) R.P.G. N.E Erythrocytes NaCN (10—3M) R.P.G. Decrease* 1 2 0.5 mm. tissue slices at 370C. N.E. — no significant effect **1% level of significance *5% level of significance 69 Figure 7 Filter Paper Electrophoresis with Superimposed Radioautograph. For explanation see text. 7v. 71 Figure 8 Cellulose Acetate Electrophoresis Strips. See text for explanation. l'r . iasv' ‘ " (V 137; .' i;~-3’V£§ 4‘ -‘ 73 hexavalent chromium. The buffer system was Ringer' phosphate solution pH 7.4 It was stated in the methodology section that with addition of excess salts during the conversion of tri— valent chromium to hexavalent chromium a colloidal suspension was formed. Strips #5 and #6 are of this colloidal solution. It was found that this suspension could not be redissolved in distilled water or Ringer's solution. It appears that a small portion of the activity is in the form of hexavalent chromium, but a larger portion is in some other form that has no charge at pH 7.4. In conversions in which the formation of a colloid had not 'taken place the total radioactivity was in the form of Na2Cr5104. Figure 8 shows examples of cellulose acetate electro- phoresis strips. Strips #1 and #2 are of hexavalent and trivalent chromium respectively. The polarity and origin of each strip is indicated. Strips #3 and #5 are of hemo— globin solutions isolated as stated in the methods sections. Both of these strips were run for 2 hours (6 ma.; 200 volts; pH 8.6) and the strips were not stained. Strips #4 and #6 are identical except they were stained. Two definite bands 74 were resolved and further examination revealed that the chromium-51 activity, determined by x—ray film, was located in both fractions. Chemical separation of the globin protein from the haem portion of the hemoglobin molecule showed that the chromium-51 was all located on the globin. Strips #7, #8, #9, #10, #11 and #12 are of normal turtle plasma. Staining procedure and buffers used have been described above. No tagging of plasma proteins was attempted. Paper Chromatograms were run on both the hydrolysates of globin and hemoglobin with intentions of locating the sites of chromium binding. Twenty amino acid standards were prepared as directed by Moffat (1959). Rf values were calculated for all standards which were run concurrently with hydrolysates. The polychromatic stain fades in two weeks so immediate reading is necessary. Both globin and hemoglobin hydroly- sates revealed 17 definite amino acids and the hemoglobin hydrolysates had several other spots that were not identi- fiable. These spots were probably associated with iron. The Chromatograms derived from hydrolysaines of hemo- globin and globin, tagged with chromium-51, were placed on 75 x-ray film and exposed for varying periods of time. In all cases one major band of activity appeared which was not associated with any amino acids. Paper Chromatograms of hexavalent and trivalent chromium gave Rf values identical to the Rf values of the heavy concentration of radioactivity observed in both hydrolysates. Histoautoradiography, Figures K, M, N and 0 show the typical resolutions of the historadiograms made with AR50 film. AR50 has large grain size and this, coupled with the necessity of long exposure (over three months), results in a loss of resolu— tion. The films were exposed during the summer months of 1961. The fallout levels in the state, caused by the Rus- sian tests, were at the highest recorded level; therefore, the background activity was also very high. The most successful autographs were of tissues removed 6 days following in vivo injection of Na2Cr5104. Figure N shows a typical erythrocyte. In most cases the activity was so high that the image of the red cell was very poor. Figure 0 gives a View of the cross section of the small intestine and shows no accumulation of chromium in any of the many tissue layers photographed. Figure L shows a 76 TABLE 5 EFFECT OF EXPOSURE TO Na CrO ON THE 2 4 Q02 OF TURTLE TISSUE Cr VI 9 ' . . .X 0 Tissue No (mg. Cr/l) 002(u1 /hr 1 cells) Erythrocyte 106 Control 20.88 I 0.828 (S.E.) Erythrocyte 21 20 20.30 I 1.849* Erythrocyte 19 200 10.79 i 1.941** 002(ul./hr.X mg. dry wt.) Liver 93 Control 0.93 i 0.069 Liver 46 20 0.94 f 0.097 Liver 35 200 0.61 i 0.039* Note: Liver dry wt. = 23.0 : 0.847% of the wet wt. 1 ‘k 5% signifi **- 37OC.; Ringer's phosphate pH 7.4 + 50 mg./100 cc. glucose cance of difference from zero 1% significance of difference from zero 77 normal H & E pMotomicrograph of the mesentery of the stomach area; a large artery and vein are very prominent. The radioautograph from the same block reveals a high chromium concentration located near the loose connective and lymphoid tissue below the artery. Figure J of the normal turtle lung reveals the unusual fact that this tissue normally has much smooth muscle. Figure K is a radioautograph of lung tissue showing many areas of chromium accumulation. These high density spots were not found in the background areas; however, associa— tion of the activity with any type of tissue was not possible. Effect of Exposure to Na2Cr04 on the 002 of Turtle Tissues The procedure used to measure the oxygen consumption of turtle tissues has been described. The data for 002 for the red cells were based on ul. O2 consumed per 109 cells. The 002 for liver tissue was based on mg. dry weight calculated from the percentage dry weight times the wet weight. Liver dry weight was found to be 23.0 I 0.8% of the wet weight. Of additional interest is the fact that the meta- bolically active tissue of the turtle represents 65.7% of the total body weight, the shell being 34.3 f 0.9% of the 78 body weight. Table 5 presents the data on the effects of chromium VI on the 002. Distribution 9f_Chromium Following Intraperitoneal Injection of NaZCr5104 Five turtles were given chromium—51 as Na2Cr5104 intra- peritoneally and six days later the animals were sacrificed; tissues and organs were removed and weighed. Tissue samples were washed in Ringer solution, placed in 10 ml. plastic counting flasks and counted in the well counter. Table 6 gives the percentage body weight for each of the organs collected. The results of a comparison of organ chromium levels to the blood levels of chromium are presented in table 7. Only two tissues, the kidney and liver, have activities higher than the blood level. The total blood activity was divided so that 16% of the chromium was in the cells and 84% of the total chromium-51 was in the plasma. Na22 was used in an attempt to measure the extracellular volume of the tissues removed in the study. Na22 was injected into the animal and one hour later tissues were removed and counted in the well counter. Calculation of 2 "Na 2 space" followed the example of Manery and Bale (1941). TABLE 6 ORGAN-BODY WEIGHT RELATIONSHIPS IN CHRYSEMYS PICTA 79 Tissue Per Cent Body weight S. Liver 4.6% to. Stomach 1.8 1 Intestine 3.3 0 Lungs 1.1 0 Spleen 0.3 0 Heart 0.4 0 Testes 0.5 0 Oviduct 2.4 0 Pancreas 0.3 0 Kidney 0.4 0 Eggs 80 TABLE 7 DISTRIBUTION OF CHROMIUM IN VARIOUS TISSUES FOLLOWING 5 INTRAPERITONEALLY INJECTED Na2Cr lO4 __organ c.p.m./gm. Na22 "Space” Tissue No' -blood c.p.m./m1. Blood1 5 1.00 2.80 : 0.04(S.D. cells) Kidney 5 2.11 44.80 i 5.10 Liver 5 1.88 27.19 i 3.65 Lungs 5 0.52 56.60 I 9.99 Testes 3 0.44 Spleen 5 0.38 10.80 i 8.06 Bile 5 0.37 Pancreas 5 0.34 Intestine 5 0.32 Urine 4 0.30 Oviduct 2 0.30 Stomach 5 0.28 Heart 5 0.25 47.90 i18.50 Muscle 5 0.20 20.93 i 3.46 l . . . . . Chromium location 6 days after injection Blood cells Plasma 16% of total blood activity 84% of total blood activity 81 Calculation of the Na22 space of the liver permitted calculation of the total cellular and extracellular chromium in the liver. On a per gram basis the hepatic cell did not seem to be able to accumulate chromium to a greater extent than the nucleated erythrocyte. 82 DISCUSSION The nucleated erythrocyte, in many ways, was found to metabolize chromium differently from that predicted on the basis of previous work with human anucleated erythrocytes. Human erythrocytes show a high uptake of hexavalent chromium when incubated in a buffered solution of tri- sodium citrate, citric acid and dextrose (ACD solution). ++ , , , The removal of Ca ions from the media seems to increase the uptake of hexavalent chromium by these cells. Addition ++. . of l mg./m1. of Ca ions Wlll decrease uptake as much as . ++ . 50%. It is well known that removal of Ca ions from whole blood (i.e., during treatment with oxalate or citrate for prevention of coagulation) results in no adverse changes in the human erythrocyte (Von Ehrenstein and Zucharias, 1956). . . ++ It is reported (Maizels, 1956) that the removal of Ca ions from blood of the African tortoise will lead to lysis. Lyman (1946) stated that one can not dilute turtle blood with NaCl or oxalate without inducing complete hemolysis; . ++ . . however, as little as 0.0017M Ca ion can prevent this - . ++ , hemolySis. Ferguson (1940) reported again that Ca ions were needed in the blood of the dogfish, turtle and snakes 83 in order to prevent hemolysis. Treatment with fluorides and arsenates produces hemolysis by removal of magnesium and calcium which in turn alters the permeability of the nucleated erythrocyte. In general, calcium ions will decrease the permeability of the cell membrane to cations. Other ions such as chromium were reported by Lyman (1945) to prevent hemolysis of nucleated red cells, in media of low Ca++ ions, by a process of precipitation of the pro— teins of the corpuscle walls, and therefore its action was not comparable to that of calcium. Results from this work suggest that, in some way, the calcium ion is functional in controlling the permeability of the cell wall to hexavalent chromium. The small amount of Ca++ present in the Ringer's solution is sufficient to demonstrate this control. At higher calcium ion concentrations the membrane is more drastically altered as shown by the marked agglutina— tion. Associated with this agglutination is a marked increase in the uptake of hexavalent chromium. This uptake is probably a membrane phenomenon. Northrop and Freunad (1923) describe electrolyte control of agglutination. Because removal of Ca++ ions is not a practical means of controlling coagulation of turtle blood the possible use of 84 heparin was investigated. VOn Ehrenstein and Zacharias (1958) have reported that heparin decreased the uptake of chromium by 1/3 in the case of the anucleated cells. In the present study the accumulation of chromium by the turtle erythrocyte was shown to be unaffected by the presence of heparin, indicating differences in the mechan- ism of transport of cations and anions by the two types of red cells. Under the conditions identical to those used in this study the human cell is reported to accumulate 90% of the available chromium in 30 minutes while turtle cells accumu- 1ated only 30%. The turtle cells show a very rapid uptake which is completed within 10 minutes, followed by a much slower uptake that is linear with time and shows little or no decrease after 13 hours. The first rapid phase has been associated with the accumulation and binding to the surface of the cells. The second phase has been associated with the accumulation of chromium in the cytoplasm, a slower process governed by transfer through the cell wall. The accumulation of chromium by the nucleated erythro- cyte of the turtle does not seem to be dependent on the presence of glucose. One would therefore suspect that the process might be one of simple diffusion. not requiring 85 metabolic energy. A similar situation is reported by Saltman, Kisken and Bellinger (1956). They have inves- tigated the mechanism of iron transport in mammalian systems and found that the uptake of iron by the rat liver slice was independent of metabolic energy and yet could lead to the establishment of an apparent concentration gradient of the metal within the cell. Iron and chromium are metallic ion capable of forming chelated compounds and in some way this property is tied up with the unusual properties of their transport kinetics. Of the metabolic inhibitors investigated, only NaCN caused a decrease in the uptake of hexavalent chromium. The cyanide treatment caused marked hemolysis which may be associated with an interruption of the metabolic energy pool supplying the energy requiring process neces- sary for maintenance of cell membrane and its permeability. In some way the cell becomes modified by cyanide treatment causing a change in the reactive groups on the membrane surface or by a more indirect approach by way of the effects on the energy metabolism of the cells. Na fluoride, azide and arsenite result in inhibition of different enzyme systems involved with active transport of ions (Maizels, 1951), but none had any significant effect 86 on chromium accumulation by turtle erythrocytes. The study of the distribution and valence state of chromium in the cell was approached from many different directions. Separation of soluble cytoplasmic proteins from the cell walls and nuclei proved to be a major prob— lem. The turtle erythrocyte was found to be very resistant to hemolysis and once the cellular breakdown was accom- plished the liberated proteins would immediately clot. Very dilute solutions of these cytoplasmic proteins were able to form thin gelatinous masses. Removal of Ca++ ions or treatment with heparin did not prevent this coagulation. Similar results (Maizels, 1954) have been reported for the turkey and chicken red cell. Because of this problem the isolation of clean cells ghosts, free from cytoplasmic proteins such as hemoglobin, was very difficult. There is also an indication that hemoglobin is an integral part of the stroma material. All workers have found that their stroma preparations contain this protein, and have found great difficulty in removing it (Prankerd, 1961). It appears possible that hemoglobin may be important in main- taining the structural stability of the cell membrane. The great affinity of chromium for hemoglobin thus gives a feasible explanation for the relatively high accumulation on the cell stroma. 87 The living cell can be looked upon as a metal binding agent which has both cytoplasmic and surface binding sites. In many cases the cell surface represents a temporary barrier to metallic ions. As the metal ion moves into the cell it must first come in contact with the so—called metal binding sites located in the cell wall. A process then exists whereby the binding sites can give up the metal ion to deeper agents that will bind the metal with more stability. The rapid accumulation of chromium may be associated with saturation of the transient binding sites located in the cell wall. The saturated metal binding sites then may serve as "a conveyor belt” to carry additional ions into the cytoplasmic pool. Many different metabolic substances can pass very rapidly through the many types of membrane pores. Other substances like metallic ions are transported much slower. Christensen (1961) stated that the transport of metal ions may well occur more as a result of sequential complex1ng on the mobile chelating agents and mobile metal complexes than by the free movement of the metal ions themselves. The metal ion may therefore enter the cell in combination with some carrier substance. The carrier may be removed l 88 by a metabolic process in the cytoplasm or it may remain firmly attached. The metal then in the case of chromium, is transferred to a permanent binding site on the soluble proteins. The concentration of hemoglobin is greater than other proteins so a larger percentage of chromium is associated with this protein. Chromium was found only on the globin portion of the hemoglobin molecule. Correlations between the exposed chelating sites (transient binding sites of the cell wall) and the sites functioning in the transport of other materials may exist. When Mn++ ions or Mg++ ions are fixed on yeast cells the process of glucose uptake is enhanced. This may indicate that the metal-binding site is part of the glucose trans— port mechanism. Addition of uranyl ions (Christensen, 1961) which fill the superficial sites inhibits the trans— port of glucose. Accordingly Stein (1958) showed that metallic ions bind to N—terminal histidine, which is part of the glycerol transport site and Mertz (1961) showed a relation between glucose transport and chromium ions. If the above hypotheses were true then accumulation of metallic ions by an apparently non—energy requiring process, but still against a concentration gradient may occur. The rate of uptake would be dependent upon the 89 transport of the material across the cell wall, that is, its rate of interaction with the transient binding sites. The large accumulation of trivalent chromium after 5 minutes incubation with hexavalent chromium means that there is an intracellular reduction of chromium VI to chromium III which occurs at a rapid rate. The disappear— ance of free intracellular chromium III could be by oxi— dation to chromium VI or removal by binding to cytoplasmic proteins. At temperature extremes of 20C. and 540C. no signifi- cant chromium accumulation occurred with time. At other temperatures, between these extremes a linear uptake of chromium was observed over an 8 hour period. The uptake rate increased with temperature,reaching a maximum rate at about 40°C. The Q10 values calculated for several temper— ature ranges indicated that the process of chromium accumulation is probably similar to simple diffusion. Brauer, Long and Halloway (1957) calculated Q values 10 for the overall reaction rate for the uptake of CrPO4 radiocolloid by isolated rat liver. They found that over the temperature range of 200C. to 350C. the Q10 was sub- stantially constant and equal to 1.92. This agrees with data on the red cell of the turtle. The reaction of 90 radiocolloid with the liver is believed to be an irrever— sible reaction between the colloid and an effectively unlimited fixed active phase, presumably the surface of the Kupffer cells. Under the conditions existent in the living body the possibility of conversion of soluble chromate to colloidal forms of chromium can not be ruled out. This type of sur- face attraction based on surface charges may then play an important role in chromium accumulation. The temperature effect can be explained in one of two ways: (1) Because the rate of simple diffusion is directly related to the molecular motion, i.e. kinetic energy, of the diffusing molecules, one would predict an increased rate of diffusion with an increased temperature. (2) The temperature elevation may cause an increase in the avail- able reaction sites (transient metal binding sites) resulting in an increased influx of the material. The effects noted at the extreme temperature ranges were probably caused by alteration in the physical proper- ties of the proteins. Boiling of liver and kidney tissue slices also resulted in a decrease in the chromium accumu- lation. This effect is probably due to the denaturation of the membrane proteins and corresponding changes in membrane permeability. 91 It was also noted that increased temperature caused a general increase in the "Y" intercept of the fitted lines. Therefore it can be concluded that the rapid accumulation of Cr51 on the cell membrane is accelerated by increasing temperatures. This may be due to a greater availability of surface reaction sites. The cataphoretic velocity (i.e. electrophoretic mobil- ity) is not influenced by either size or shape of the microscopic particle suspended in an electrolyte-containing medium. The differences in red cell velocities of many different species are given by Abramson (1929) and he believes these differences are almost certainly represen- tative of surface constituents of the different cells. The influence of salts on membrane potentials and cata- phoretic potentials is discussed by Loeb (1923). The net charge of a cell is dependent only on the ionization of the chemical groups making up the cell mem— brane. Chromium tagging significantly decreases the nega- tive charge. Trivalent cations such as Fe+++ and A1+++ are also very effective in reducing the negative zeta potential of proteins (Oliver and Barnard, 1924). In the present studies chromium VI may have been reduced at or on the surface of the cell, then bound to the cell wall proteins 92 An increase in the chromium concentration of the medium was shown to decrease the net negative charge on turtle erythrocytes. In an attempt to compare the metabolism of chromium by the nucleated erythrocytes with liver and kidney slices, it was found that tissue slices present the difficult problem of distinguishing extracellular from cellular fixation, and identifying the barriers that retard metal entrance into cells. Indications of this work on turtle tissues are that the chromium enters liver and kidney cells probably in the same manner that it enters the red cell. The assumption was made that 1.0 ml. of whole blood is equivalent to 1.0 gm. of wet tissue. Taking the blood activity as 1.0 it was found that only the kidney and the liver had activities above that of the blood. Liver activity undoubtedly is associated with the phagocytic activity of the Kupffer cells, and their removal of col— loidal chromium. The ability of the kidney to accumulate chromium above blood levels has been reported many times previously, where chromium, in some way is said to inter— fere with the transport mechanism of the first portion of the proximal convoluted tubule. 93 Historadioautographs indicated that the radioactive chromium is located in the hepatic cells as well as in the extracellular spaces. There is a possible indication that chromium-51 may be concentrated in the lumen of the tubular network, however, because of the relatively poor resolution (with AR50) no definite statement as to the exact intra- cellular localization of chromium can be made. The only other tissue showing an appreciable accumu— lation of chromium was the lungs. Histologically the lungs of the turtle appear to have a normal alveolar epithelium. This is infiltrated with fibers and sheets of smooth muscle. The muscle tissue appears to be connected to the alveoli by areas of loose connective tissue. Radioauto- graphs revealed many scattered areas of chromium accumula— tion, but no positive localization of the activity in any specific tissue was possible. The many areas of chromium activity may be related to lymphatic activity. Circulating leukocytes having the power to engulf col- loidal particles by phagocytosis are known to exist. These cells could take up and concentrate large amounts of chromium. Labeled leukocytes could travel to the lungs or they might originally be located in small areas throughout the thoracic area, resulting in the patchy areas of chromium 94 accumulation as observed in the lungs of the turtle. Bierman, Kelly and Cordes (1956) studied the leukocyte concentration entering and leaving the lungs of man and found a very great capacity of the lungs to hold leukocytes, causing an arterio-venous