2523222 52222 22522622235222 925222223. 32. 222W? 2:22.23 2:25 2* 222cm 22,-.22 This is to certify that the thesis entitled METABOLIC AND FUNCTIONAL ACTIVITIES 0F PHAGOCYTIC CELLS DURING GALACTOSEMIA presented by William John Litchfield has been accepted towards fulfillment of the requirements for Ph. D. degree in Biochemistry Max/4% Major professor Date /Q/L/7€ 0-7639 2. V A worm: my *5 -7 nfimsm 800K BINDERY INC. l'RRARY BINDERS lil‘. staunch}. HICIIGIJ ABSTRACT METABOLIC AND FUNCTIONAL ACTIVITIES OF PHAGOCYTIC CELLS DURING GALACTOSEMIA By William John Litchfield To account for enhanced susceptibility to infection among infants deficient in galactose—l-phosphate uridylyl- transferase (EC 2.7.7.10), the acute effects of D—galactose on metabolic and functional activities of phagocytic cells involved in host defense were investigated. Human and guinea pig polymorphonuclear leukocytes (PMN) when incu- bated in medium containing 30 mM galactose displayed sub- stantially less killing of Escherichia coli than when in- cubated in medium with 5 mM glucose. Impaired bactericidal activity was dependent upon galactose concentration but could be partially averted by supplementing the galactose- containing medium with 15 mM glucose. Phagocytic activi- ties of guinea pig PMN and peritoneal macrophages were assayed by following ingestion of 32P-labelled E. coli and were also depressed by elevated galactose. ’ The inhibitory action of galactose on phagocyte func- tion was further investigated in normal and galactosemic chicks by monitoring the in.vitro killing of E. coli by William John Litchfield leukocytes and the in vivo clearance of colloidal 125l— labelled bovine serum albumin (125I-BSA) from the circula- tion. Elevated levels of galactose (30 mM) impaired the bactericidal activities of both normal and galactosemic leukocytes. However, the latter cells were more suscep— tible to the galactose dependent inhibition. Phagocytic l2SI-BSA, indexes, obtained from data on the clearance of were calculated to be 0.0553 and 0.0297 for normal and galactosemic chicks, respectively. Chicks fed a control diet displayed a logarithmic clearance of colloid with postinjection time, whereas this relationship was not as apparent when galactosemic chicks were employed. Moreover, galactose impaired phagocytic functions of both circulating leukocytes and tissue-fixed macrophages as well as the overall development of the reticuleondothelial system. Although galactose and glucose were transported into PMN, competition between these two hexoses for cell entry was eliminated as a mechanism of galactose toxicity. Neither galactose nor fructose competed with [G—SH]2— deoxyglucose for uptake by guinea pig PMN, whereas compe- titive inhibition was observed when either glucose or mannose were employed. Transport of [G-BHJZ-deoxyglucose proceeded in vitro with Kb of 1.8 mM and Vmax of 0.67 nmole min-1 per 106 cells. This uptake was temperature dependent, and phosphorylation of the 6-position was necessary for intracellular concentration. Uncouplers of oxidative phosphorylation did not alter 2-deoxyglucose uptake. William John Litchfield However, uptake was sensitive to inhibitors of glycolysis and could be inhibited by pre-incubating cells with 2 mM iodoacetate, 40 mM fluoride, or 30 mM galactose. The latter observation indicated that galactose could interfere with glucose entry by blocking its phosphorylation rather than competing with its uptake. Significant decreases in intracellular levels of adeno- sine 5'-triphosphate and in levels of glycolytic intermedi- ates were found to explain the galactose dependent loss of phagocytic activity. These decreases were attributed to (i) the presence of a futile adenosine triphosphatase cycle, and (ii) the accumulation of galactose intracellularly with subsequent inhibition of glycolysis. To account for the impaired bactericidal activity of PMN during galactosemia, a number of biochemical parameters associated with the oxygen-dependent killing mechanism were investigated. Galactose did not affect either oxygen con— sumption or extracellular hydrogen peroxide formation. However, the ability of PMN to reduce extracellular cyto- chrome 3 via superoxide anion was significantly impaired. The latter action could not be accounted for by decreased superoxide generation but could be attributed to oxidation of ferrocytochrome g,by galactose radical. Formation of this carbohydrate radical was observed in two chemical systems, and its presence in PMN was supported by data on cellular reduction of nitroblue tetrazolium, formate oxi- dation, and chemiluminescence. These observations suggested William John Litchfield that galactose could impair the killing of bacteria by reac- ting with hydroxyl radical and subsequently scavenging superoxide anion. METABOLIC AND FUNCTIONAL ACTIVITIES OF PHAGOCYTIC CELLS DURING GALACTOSEMIA By William John Litchfield A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1976 ACKNOWLEDGEMENTS The author wishes to express sincere gratitude to Professor William W. Wells for his enthusiasm, guidance and financial support during the course of this research. I would also like to thank each of my committee members, Drs. Robert Moon, Pamela Fraker, Allen Morris, Steven Aust, and N. E. Tolbert for their honest criticisms and helpful suggestions. Finally, I wish to thank my loVing wife Marilyn for her constant encouragement and under- standing. ii TABLE OF CONTENTS Page LIST OF TABLES. . . . . . . . . . . . . . . . . . . vii LIST OF FIGURES . . . . . . . . . . . . . . . . . . ix LIST OF ABBREVIATIONS . . . . . . . . . . . . . . . xi INTRODUCTION. 1 Organization . 1 Rationale and Objectives 1 Literature Survey. 3 References 8 CHAPTER I. INHIBITORY ACTION OF D-GALACTOSE ON ~ PHAGOCYTE METABOLISM AND FUNCTION. . . . . . 13 Abstract . . . . . . . . . . . . . . . . . 13 Introduction . . . . . . . . . . . . . . . 14 Materials and Methods. . . . . . . . . . . 15 Phagocyte preparations . . . . . . . . . 15 Bacteria . . . . . . . . . . . . . . . . 17 Phagocytic and bactericidal assays . . . l7 20xidation of glucose and galactose . . . 19 Metabolite assays. . . . . . . . . . . . 21 Results. . . . . . . . . . . . . . . . . . 22 DiscuSSion O O O O O O O O O O O O O O O I 26 References . . . . . . . . . . . . . . . . 50 iii CHAPTER Page II. INHIBITORY ACTION OF GALACTOSE ON PHAGO- CYTES FROM NORMAL AND GALACTOSEMIC CHICKS. . . 42 Abstract . . . . . . . . . . . . . . . . . . 42 Introduction . . . . . . . . . . . . . . . . 45 Materials and Methods. . . . . . . . . . . . 44 Animals and materials. . . . . . . . . . . 44 Phagocyte preparations . . . . . . . . . . 44 Bacteria . . . . . . . . . . . . . . . 45 Bactericidal assays. . . . . . 45 Intravascular clearance of colloidal lZSI-BSA . . . . . . . . . . . . . . . . . 46 RES development. . . . . . . . . . . . . . 48 Results. . . . . . . . . . . . . . . . . . . 48 Effect of galactose on bactericidal activity . . . . . 48 Effect of galactose. on clearance of colloid. . . . . . . . . . . . . . . . 49 RES development. . . . . . . . . . . . . . 50 Discussion . . . . . . . . . . . . . . . . . 51 References . . . . . . . . . . . . . . ... . 55 III. HEXOSE TRANSPORT IN HUMAN AND GUINEA PIG POLYMORPHONUCLEAR LEUKOCYTES . . . . . . . . . 60 Abstract . . . . . . . . . . . . . . . . . . 60 Introduction . . . . . . . . . . . . . . . . 61 Materials and Methods. . . . . . . . . . . . 62 Materials. . . . . . . . . . . . . . . . . 62 Cell preparations. . . . . . . . . . . . 62 Oxidation of carbohydrate. . . . . . . . . 65 Transport of carbohydrate. . . . . . . . . 64 Levels of ATP. . . . . . . . . . . . . . . 66 Results. . . . . . . . . . . . . . . . . . . 67 Oxidation of carbohydrates to 14002. . . . 67 Exclusion of sucrose from PMN. . . . . 67 Time course for 2-deoxyglucose uptake. . . 68 iv CHAPTER Page Uptake of 2-deoxyglucose versus cell concentration. . . . . . . . . . . . . . 68 Kinetics of uptake . . . . . 69 Effects of heterologous carbohydrates on uptake. . . . . . . . . . . . . 70 ATP levels . . . . 72 Effects of heterologous carbohydrates on uptake by human PMN . . . . . . . . . 72 Discussion . . . . . . . . . . . . . . . . . 75 References . . . . . . . . . . . . . . . . . 80 IV. EFFECT OF CARBOHYDRATE ON THE OXYGEN- DEPENDENT KILLING MECHANISM OF POLYMORPHO— NUCLEAR LEUKOCYTES . . . . . . . . 93 Abstract . . . . . . . . . . . . . . . . . . 95 Introduction . . . . . . . . . . . . . . . . 94 Materials and Methods. . . . . . . . . . . . 95 Animals and materials. . . . . . . . . . . 95 Cell preparations. . . . . . . 96 Extracellular reduction of cytochrome c. . 96 Cellular reduction of nitroblue tetra-— zolium . . . . . . . . . . . . . . .2. . 97 Cellular release of hydrogen peroxide. . . 98 Cellular formate oxidation . . . . . . . . 100 Cellular oxygen consumption. . . . . 100 Methional assay for hydroxyl radicals. . . 101 Oxidation of hexoses by hydroxyl radicals. 105 Polyol formation by PMN. . . . . . . . . . 104 PMN chemiluminescence. . . . . . . . . . . 105 Results. . . . . . . . . . . . . . . . . . . 106 Extracellular reduction of cytochrome c. . 106 Cellular reduction of nitroblue tetra-— zolium . . . . . . 107 Cellular release of hydrogen peroxide. . . 107 Cellular formate oxidation . . . . . . . . 108 Cellular oxygen consumption. . . . . . . 108 Methional assay for hydroxyl radicals. . . 109 Oxidation of hexoses by hydroxyl radicals. 111 Polyol formation in PMN. . . . . . . . . . 112 PMN chemiluminescence. . . . . . . . . . . 115 CHAPTER Page Discussion. . . . . . . . . . . . . . . . . . 115 References. . . . . . . . . . . . . . . . . . 124 SUMMARY. . . . . . . . . . . . . . . . . .'. . . . . . 142 References. . . . . . . . . . . . . . . . . . 147 APPENDIX . . . . . . . . . . . . . . . . . . . . . . . 148 vi TABLE Chapter I. II. III. IV. Chapter I. Chapter I. II. III. IV. LIST OF TABLES Page I Effect of galactose on bactericidal activ— ity of human polymorphonuclear leukocytes. . 52 Effect of galactose on bactericidal activ- ity of guinea pig polymorphonuclear leukocytes . . . . . . . . . . . . 55 Influence of carbohydrate on phagocytosis of latex particles by guinea pig peritoneal macrophages. . . . . . . . . . . . . 54 Conversion of galactose to glucose by guinea pig PMN as indicated by the specific activity of [1-14 C]D-glucose . . . . . . . . 55 Effect of galactose on levels of adenine nucleotides and glycolytic intermediates in guinea pig peritoneal macrophages . . . . 56 II Response of RES to injection of E. coli. . . 56 III Conversion of 14C- labelled carbohydrate to 4CO2 by guinea pig PMN. . . . . . . . . . 82 Exclusion of [U- l4C]sucrose from guinea pig PMN. . . . . . . . . . . . . . . . . . . 85 Effect of carbohydrate and insulin on up- take of [G—3 H]2— —deoxyg1ucose by guinea pig PMN. . . . . . . 84 Effect of metabolic inhibitors on rates of [G—3H]2-deoxyglucose uptake and phosphory— lation by guinea pig PMN . . . . . 85 Effect of iodoacetate and galactose on ATP levels in guinea pig PMN . . . . . . . . 86 vii TABLE VI. Eff ct of carbohydrate on uptake of [G— H]2-deoxyglucose by human PMN. . Chapter IV I. Carbohydrate effect on cytochrome 2 reduction by PMN . II. Effect of carbohydrate on reactions involving superoxide anion radical . III. Effect of carbohydrate on hydrogen peroxide formation by PMN. . . . IV. Carbohydrate effect on l4C-formate oxida- tion by PMN. . . . . . . . . . . V. Effect of carbohydrate on oxygen consump- tion by PMN. . . . . . . . Appendix I. Analysis of carbohydrates in sera. II. Acid phosphatase activity associated with guinea pig PMN . III. Distribution of hexosaminidase and peroxi— dase in guinea pig PMN . . . . viii Page 87 127 129 130 151 152 148 149 150 FIGURE Chapter 1. 5a. 5b. Chapter 1. Chapter 1. LIST OF FIGURES I Semilogarithmic plot of E. coli growth in high galactose media. . . . . . . . . . Inhibitory effect of galactose on phago— cytosis of 32P—labelled E. coli by guinea pig PMN . . . . . . . . . . . . . . . . . Inhibitory effect of galactose on phago- cytosis of 32P-1abelled E. coli by guinea pig peritoneal macrophages. . . . . . . Effect of galactose on oxidation of [l-14 C] glucose and [6-14 C]glucose by guinea pig PMN . . . . . . . . Phase contrast photomicrograph of guinea pig peritoneal macrophages. . . . . . Bright-field fluorescence photomicrograph. of guinea pig exudate cells stained with acridine orange II Effect of galactose on bactericidal activity of normal and galactosemic chick leukocytes in vitro. . . Intravascular clearance of colloidal lZSI—BSA- in normal and galactosemic chicks Effect of galactose diet on spleen and liver develOpment in chicks . . . III Time- -course for uptake of 5.0 mM 2- -deoxy- glucose by guinea pig PMN . . . . . . Uptake of 2—deoxyglucose versus cell con- centration. . . ix Page 37 38 39 4O 41 41 57 58 59 88 89 FIGURE Chapter 1. Appendix 1. Dependence of 2-deoxyg1ucose uptake and 5-O-methy1glucose uptake on levels of external homologous carbohydrate. Inhibition of 2— deoxyglucose transport by glucose and mannose . . . . . . . Temperature dependence of 0.2 mM 2- deoxyglucose uptake and 0.1 mM 5- 0- -methyl— glucose uptake. . . . . . . . . . IV Effect of carbohydrate and pre-incubation time on nitroblue tetrazolium reduction by PMN . . . . . . . . . . . . . . . . . . Effect of galactose concentration on nitroblue tetrazolium reduction by PMN. Relationship between sc0poletin fluores- cence and nanomoles of hydrogen peroxide Effect of polyols on hydroxyl radical dependent ethylene formation from methional. Enzymatic generation of hydroxyl radicals . Effect of carbohydrates and polyols on hydroxyl radical dependent ethylene forma- tion from methional. Non—enzymatic genera- tion of hydroxyl radical. . . . . . . . . Identification of hexonic acids as products of the reaction between carbohydrates and hydroxyl radicals in vitro. . Identification of galactitol in homogenates of PMN after incubation with 50 mM galactose. Accumulation of intracellular polyol during incubation of PMN with elevated levels of carbohydrate. . . Effect of carbohydrate on PMN chemilumines— cence . . . . . . . . . . . . . . Effect of superoxide dismutase on the decay of acridine orange fluorescence in phago- cytes . . . . . . . Page 90 91 92 133 134 135 136 137 158 139 140 141 151 AMP, ADP, ATP A0 BSA CGD CPM EDTA G6PDH i.p. KRPS NADH MPO PMN PNP-NAG RES SD SEM SOD LIST OF ABBREVIATIONS adenosine 5'-mono-, di-, or triphosphate acridine orange bovine serum albumin chronic granulomatous disease counts per minute ethylenediaminetetraacetic acid glucose—6-phosphate dehydrogenase intraperitoneal Krebs—Ringer phosphate solution reduced nicotinamide adenine dinucleotide myeloperoxidase polymorphonuclear leukocytes para-nitrophenyl-N'-acetylglucosamine reticuloendothelial system standard deviation standard error of the mean superoxide dismutase xi INTRODUCTION Organization This dissertation is divided into four chapters, each of which is in a form acceptable for publication in a bio- chemical journal. Chapter I has been published in Infegr tion and Immunity under the authorship of William J. Litch- field and William W. Wells (1976) 15, 728-754. Chapter II and III have been submitted to Infection and Immunity and Journal of Biological Chemistry, respectively. Chapter IV is in preparation to be submitted to the latter journal. The first chapter is presented with the permission of the American Society for Microbiology, COpyright holder. An appendix containing supplementary data is included at the end of the text. Rationale and Objectives Frequent occurrences of bacterial infection among patients with poorly controlled diabetes mellitus (1,2) and of Escherichia coli septicemia leading to death among infants with galactose—l-phosphate uridylyltransferase deficiency (5,4,5) have been well documented but poorly understood. Recent reports concerning host defense among diabetics demonstrated that the ability of polymorphonuclear 2 leukocytes (PMN) to ingest and destroy bacteria was depres— sed during hyperglycemia (6,7,8). This action was pri— marily attributed to elevated plasma glucose and was not a result of insulin or opsonin deficiencies (9). These ob- servations suggested that a similar type of impairment could occur during galactosemia. However, there was no pertinent information about the function of PMN during this condition, and leukocytes from galactosemic infants were not readily available. To test ‘this hypothesis, studies were therefore con- ducted on PMN isolated from the peripheral blood of normal human and guinea pig sources. Levels of galactose employed in these studies (1 to 50 mM) were comparable to those en- countered during the pathological condition (10,11), and it was assumed that any deleterious effect of galactose in these cells would be more pronounced in PMN of galactosemic origin. This assumption was supported by comparing the ef— fects of galactose upon bactericidal activity of leukocytes from normal and galactosemic chicks. All assays of bacteri- cidal activity were performed in vitro using serum—treated E. coli, whereas phagocytic assays were performed under 52 similar conditions using P-labelled E. coli or polystyrene latex particles. Extrapolation of results from these exper- iments to the in vivo state was facilitated by observing the inhibitory action of galactose upon the in vivo clearance of colloidal BSA from the circulation of galactosemic chicks. The biochemical basis of galactose toxicity in 3 phagocytes was explained after investigating metabolic para- meters associated with the physiological functions. More- over, effects of galactose upon phagocytic activity were correlated with cellular dependence upon glycolysis, whereas effects upon bactericidal activity were correlated with the oxygen-dependent killing mechanism. Literature Survey The ability of phagocytic cells such as PMN, mono- cytes, and macrophages to ingest and destroy a wide variety of microorganisms is of particular importance to the protec- tion of a host against infection (12,15,14,15,l6). This physiological relationship is exemplified by enhanced sus- ceptibility to infection observed during a number of clini- cal disorders which predominantely affect phagocytosing cells.. These disorders include leukemia (17,18,19), chronic granulomatous disease (CGD)(16,20), glucose—6-phosphate dehydrogenase (G6PDH) deficiency (21,22), myeloperoxidase (MPO) deficiency (25), and Chediak-Higashi syndrome (24,25). It is significant that three, and perhaps four, of these disorders affect the oxygen dependent killing mechanism of PMN, whereas the latter syndrome affects phagocyte degranu- lation. In CGD and G6PDH deficiency the ability of PMN to phagocytese microorganisms is not impaired (16,21,22). However, killing of catalase—positive bacteria such as E. 921; and Staphylococcus aureus is significantly reduced (3,16,21,22). These observations indicate that phagocytosis 4 and intracellular killing are two distinct physiological functions in PMN and that each function may be explained in separate biochemical terms. Nevertheless, since intracel- lular killing depends upon previous bacterial uptake, both of these functions are impaired during disorders of phago- cytosis per se. These disorders can result from opsonin difficiencies in which phagocytes do not properly recognize bacteria (14) and also from a number of nutritional states in which the phagocytic cells may be metabolically impaired (26). Enhanced susceptibility to infection, primarily by encapsulated bacteria (14), is thus associated with de- creased opsonization in various abnormalities of the com- plement system (27) and of the properdin system among neo- nates (28) and patients with sickle cell anemia (29). In- fection is further associated with metabolically impaired phagocytes during vitamin deficiency (26), protein- calorie kwashiorkor (50), and increased glucose or fructose ingestion (8). Moreover, the latter observations indicate that phagocytosis is a "finely tuned" process which may undergo daily fluctuations in vivo depending upon blood glucose and other nutrient levels. This is supported by impaired phagocytic activity during hyperglycemia (6,7,8) and during starvation (50,51). Since the initial demonstration by Sbarra and Karnovsky that phagocytosis in PMN could be inhibited by agents which antagonize glycolysis (52), a number of studies have been published in support of the hypothesis that phagocytosis 5 depends upon energy supplied by this metabolic pathway (55,54,55,56). This hypothesis holds for other phagocytes such as macrophage (56,57,58) and disputes the earlier view of Penn that phagocytosis proceeds without metabolic ex— penditure (59). While phagocytosis in PMN is not altered by inhibitors of oxidative metabolism such as antimycin A, cyanide, and dinitrophenol, this function is very sensitive toiodoacetate and fluoride (52,33) which simultaneously inhibit cellular lactic acid production (52) and depress intracellular levels of ATP (40). In addition, phagocytosis can be impaired by inhibitors of glucose transport such as Neethylmaleimide and cytochalasin B (41,42). The above hypothesis is further supported by observations that PMN are not gluconeogenic (45,44) and that these cells contain relatively few mitochondria (52). Bactericidal activity, although dependent upon phago- cytosis for supply of substrate, is a separate process which can be accounted for in different biochemical terms. This function is associated with a number of toxic and bacteriostatic agents which are released into the phagoly- sosome during degranulation. Agents such as acid, lysozyme, lactoferrin, and cationic proteins contribute synergisti- cally to this function (45,46). However, killing of bac- teria is mainly an oXidative process in PMN (45,47). This process is proceeded by dramatic increases in oxygen con- sumption (52,58), in hexose monophosphate shunt activity (52,48), and in enzyme activities of glutathione reductase 6 (49,50) and of NADH oxidase (51,52) and NADPH oxidase (55, 54). Killing is significantly reduced in the absence of catalase (59), superoxide dismutase (59,60) and benzoate (59). These studies indicate the involvement of hydrogen peroxide, superoxide anion, and hydroxyl radical, while it is further suggested that singlet oxygen may participate (61,62). PMN are known to produce measurable levels of hydrogen peroxide (65,64,65) and superoxide anion (66,67), which may be bactericidal by themselves (60,68) or in com- bination with myelOperoxidase plus halide (45,69). Sig- nificant decreases in killing capacity are found in phago- cytes from patients with MPO dysfunction (25) and from patients with CGD or with G6PDH deficiency (16,21,22). In the latter deficiencies, NADH oxidation and hydrogen peroxide formation are not stimulated during phagocytosis. Although glucose metabolism in normal and diabetic phagocytes has been vigorously studied (43,44,70,7l), much less attention has been given to the fate of other sugars. Normal PMN can convert l4C-labelled galactose, mannose, and fructose into labelled glycogen (72,75). However, when PMN are incubated with [2-14C] galactose, there is less randomi- zation of 140 in the glucose molecules of glycogen than when [2-14C] glucose is employed (72). This indicates that galactose follows a more direct pathway to glycogen and that the following pathway described by Lelior (74) and Kalckar et a1. (75) is present: ATP ADP galactose ‘\\1/, ~¢~ galactose-l—phosphate (l) UDP-glucose glucose-l-phosphage galactose-l—phosphate :::>?—§:: UDP-galactose 2 UDP-galactose ) :>> UDP-glucose (3 UDP-glucose derived from l4C—labelled galactose could then enter glycogen synthetase or be converted to glucose-l- phosphate via UDP—glucoseInnsphosphatase. In contrast, [2-140] glucose is initially phosphorylated by hexokinase (71), and the 14C is more likely to transverse the glycoly- tic and pentose cycle reactions before conversion to glyco- gen (72). Galactose-l-phosphate and UDP-galactose have been isolated from leukocyte homogenates (76), and the activities of both galactokinase (eq. 1) and galactose—l- phosphate uridylyltransferase (eq. 2) have been measured in PMN (77). The latter enzyme is not detectable in phago- cytes from galactosemic patients (77,78,79), and its levels are depressed in cells from asymptomatic heterozy- gotes (77,80). 10. ll. 12. 13. 14. 15. 16. 8 References Bybee, J.D., and Rogers, D.E. (1964) J. Lab. Clin. Med. g3, 1-13 Goto, Y., Sato, S., and Masuda, M. (1974) Tohoku J. Exp. Med. 112, 339—353 Kelley, S., Burns, J., and Desjardins, L. (1974) Am. J. Epidemiol. 99, 8-15 Quan-Ma, R., Wells, H.J., Wells, W.W., Sherman, F.E., and Egan, T.J. (1966) Am. J. Dis. Child. 112, 477- 478 Nadler, H.L., Inouye, T., and Hsia, D.Y.Y. (1969) In Galactosemia p. 129 (Hsia, D.Y.Y., ed.) Charles C. Thomas, Springfield, IL Johnson, J.E. (1970) In Diabetes Mellitus pp. 734- 745 (Ellenberg, M., and Rifken, K., eds.) MacGraw- Hill, New York Bagdade, J.D., Root, R.K., and Bulger, R.J. (1974) Diabetes 2 , 9-15 Sanchez, A., Reeser, J.L., Lau, H.S., Yahika, P.Y., Willard, R.E., McMillan, P.J., Cho, S.Y., Magie, A.R., and Register, U.D. (1975) Am. J. Nutr. 26, 1180-1184 Drachman, R.H., Root, R.K., and Wood, W.B.,Jr. (1966) J. Exp. Med. 12 , 227—240 Medline, A., and Medline, N.M. (1972) Can. Med. ASSOC. J. 107, 877-878 Penington, J.S., and Prankerd, T.A.J. (1958) Clin. SCio i1, 385-391 Mudd, S., McCutcheon, M., and Leuke, B. (1954) Physiol. Rev. 14, 210-275 Berry, L.J., and Spies, T.D. (1949) Medicine 28, 259- 249 Stossel, T.P. (1974) N. Engl. J. Med. 290, 855—859 Saba, T.M. (1970) Arch. Intern. Med. 126, 1051-1052 Baehner, R.L. (1972) Pediatr. Clin. North Am. 1 , 935-956 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 9 Rosner, F., Valmont, I., Kozinn, P.J., and Caroline, L. (1970) Cancer 25, 855-842 Skeel, R.T., Yankee, R.A., and Henderson, E.S. (1971) J. Lab. Clin. Med. 11, 975-984 Strauss, R.R., Paul, B.B., Jacobs, A.A., Simons, C., and Sbarra, A.J. (1970) Cancer Res. 59, 480-488 Holmes, B.H., and Good, R.A. (1971) In Immunobiology: Current Knowledge of Basic Concepts in Immunology and Their Clinical Aspects pp. 55-56 (Good, R. A., and Fisher, D.W., eds.) Sinauer Associates, Inc., Stan- ford, Conn. Cooper, M.R., DeChatelet, L.R., McCall, C.E., LaVia, M.F., Spurr, C.L., and Baehner, R. (1972) J. Clin. Invest. 52, 769-778 Baehner, R.L., Johnson, R.B., Jr., and Nathan, D.G. (1972) J. Reticuloendothelial Soc. 22, 150-169 Lehrer, R.T., and Cline, M.J. (1969) J. Clin. Invest. 8, 1478-1488 Root, R.K., Rosenthal, A.S., and Balestra, D.J. (1972) J. Clin. Invest. 1, 649-655 Stossel, T.P., Root, R.K., and Vaughn, M. (1972) N. Engl. J. Med. 286, 120-125 Gontzea, I. (1974) Nutrition and Anti-Infectious Defence S. Karger, Basel, Switzerland Rosen, E.S., Alper, C.A., and Janeway, C.A. (1974) In Hematology of Infancy and Childhood (Nathan, D.G., and Oski, F.A., eds.) W.B. Saunders Co., Philadelpha Stossel, T.P., Alper, C.A., and Rosen, E.S. (1975) Pediatrics 22, 134-137 Johnston, R.B., Jr., Newman, S.L., and Struth, A.G. (1973) N. Engl. J. Med. 288, 805-808 Douglas, S.D., and SchOpfer, K. (1974) Clin. Exp. Immunol. 21, 121-128 Berry, L.J., Davis, J., and Spies, T.D. (1945) J. Lab. Clin. Med. 29, 684-694 Sbarra, A.J., and Karnovsky, M.L. (1959) J. Biol. Chem. 234, 1355-1362 33. 34. 35. 56. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 10 Karnovsky, M.L., and Wallach, D.E.H. (1961) J. Biol. Chem. 256, 1895-1901 Karnovsky, M.L. (1962) Physiol. Rev. 42, 145-168 Karnovsky, M.L. (1974) In Progress in Immunology II Vol. 4 pp. 85-95 (Brent, L., and Holborrow, J., eds.) North-Holland Publishing Co., Amsterdam Karnovsky, M.L., Simons, S., Glass, E.A., Shafer, A.W., and Hart, P.D. (1970) In The Mononuclear Phagocytes pp. 105-120 (Van Furth, R. ed.) F.A. Davis Co., Philadelphia Oren, R., Farnham, A.E., Saito, K., Milofsky, E., and Karnovsky, M.L. (1965) J. Cell. Biol. 21, 487-498 Karnovsky, M.L., Lazdins, J., and Simmons, S.R. (1975) In Monocuclear Phagocytes In Immunity, Infectionyyand Pathology pp. 425-458 (Van Furth, R., ed.) Blackwell Scientific Publications, London Fenn, W.O. (1921) J. Gen. Physiol. 5, 575-593 Mazur, M.T., and Williamson, J.R. (1976) Fed. Proc. 25: 279 Gee, J.B.L., Khadwala, A.S., and Bell, R.W. (1974) J. Reticuloendothel. Soc. l5, 394-405 Axline, S.G., and Reaven, E.P. (1974) J. Cell. Biol. ég. 647-659 Noble E.P., Stjernholm, R.L., and Ljungdahl, L. (1961) Biochim. Biophys. Acta 4 , 593—595 Stjernholm, R.L., Burns, C.P., and Hohnadel, J.H. (1972) Enzyme l , 7-51 Klebanoff, S.J. (1975) In The Phagocytic Cell in Host Resistance pp. 45-56 (Bellanti, J.A., and Dayton, D.H., eds.) Raven Press, New York Klebanoff, S.J., and Hamon, C.B. (1975) In Mononuclear Phagocytes In Infectioninmmunology, and Pathology pp. 507-527 (Van Furth, R., ed.) Blackwell Scientific Publications, London Sbarra, A.J., Selvaraj, R.J., Paul, B.B., and Mitchell, Jr., G.W. (1975) In The Reticuloendothelial System pp. 57-48 (Rebuck, J.W., Berard, G.W., and Abell, M.R., eds.) Williams & Wilkins Co., Baltimore 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 65. 64. 65. ll Beck, W.S. (1958) J. Biol. Chem. 252, 271-285 Reed, P.W. (1969) J. Biol. Chem. 244, 2459—2464 Strauss, R.R., Paul, B.B., Jacobs, A.A., and Sbarra, A.J. (1969) Arch. Biochem. 155, 265-271 Baehner, R.L., Gilman, N., and Karnovsky, M.L. (1970) J. Clin. Invest. 29, 692-700 Cagan, R.H., and Karnovsky, M.L. (1964) Nature 204, 255-256 Iyer, G.Y.N., and Quastel, J.H. (1967) Can. J. Biochem. Physiol. 2;, 427—434 Rossi, F., and Zatti, M. (1964) Br. J. Exp. Pathol. i2, 548-559 Holmes, B., Page, A.R., Windhorst D.B., Quie, P.G., White, J.G., and Good, R.A. (1968) Ann. N.Y. Acad. Sci. 155, 888-901 Mandell, G.L. (1974) Infect. Immun. 9, 557-541 Ripley, R.J., and Sbarra, A.J. (1967) J. Bacteriol. 2i, 1417-1424 Ripley, R.J., and Sbarra, A.J. (1967) J. Bacteriol. 24. 1425—1430 Johnston, Jr., R.B., Keele, Jr., B.B., Misra, H.P., Webb, L.S., Lehmeyer, J.E., and Rajagopalan, K.V. (1975) In The Phagocytic Cell in Host Resistance pp. 61-75 (Bellanti, J.A., and Dayton, D.H., eds.) Raven Press, New York Yost, Jr., F.J., and Fridovich, I. (1974) Arch. Bio- chem. Biophys. 161, 395-401 Krinsky, N.I. (1974) Science ;§§, 565-565 Allen, R.C., Stjernholm, R.L., and Steele, R.H. (1972) Biochem. Biophys. Res. Commun. 21, 679-684 Paul, B., and Sbarra, A.J. (1968) Biochim. Biophys. Acta 156, 168-178 Root, R.K., Metcalf, J., Oshino, N., and Chance, B. (1975) J. Clin. Invest. 55, 945—955 Iyer, G.Y.N., Islam, D.M.F., and Quastel, J.H. (1961) Nature 1 2, 535-541 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 12 Babior, B., Kipnes, R., and Curnutte, J. (1975) J. Clin. Invest. 52, 741-744 Drath, D.B., and Karnovsky, M.L. (1975) J. Exp. Med. 141, 257-262 Fridovich, I. (1976) In Free Radicals.in Biology Vol. I pp. 259-277 (Pryor, W.A., ed.) Academic Press, New York Klebanoff, S.J. (1968) J. Bacteriol. 25, 2151-2158 Esman, V. (1972) Enzyme l , 52-55 Beck, W.S. (1958) J. Biol. Chem. 252, 251-270 Stjernholm, R.L., and Noble, E.P. (1961) J. Biol. Chem. 2 6, 5093-3096 Esman, V., Noble, E.P., and Stjernholm, R.L. (1965) Acta Chem. Scand. 22, 1672-1676 Lelior, L.F. (1951) Arch. Biochem. Biophys. 55, 186 Kalckor, H.M., Braganca, B., and Munch-Peterson, A. (1955) Nature 172, 1058 Klant, N., and Schucher, R. (1965) Can. J. Biochem. Physiol. 1, 849-858 Tedesco, T.A., and Mellman, W.J. (1969) J. Clin. Invest. 28, 2590-2597 Weinberg, A.N., Herring, B., Johnson, P., and Field, J.B. (1960) Clin. Res. 8, 27 Weinberg, A.N., and Segal, S. (1960) Science 1 2, 1015 Mellman, W.J., Tedesco, T.A., and Baker, L. (1965) Lancet 1, 1595-1596 CHAPTER I INHIBITORY ACTION OF D—GALACTOSE ON PHAGOCITE METABOLISM AND FUNCTION Abstract To account for enhanced susceptibility to infection among galactosemics, the acute effects of D-galactose on metabolic and functional activities of phagocytic cells in vitro were investigated. Human and guinea pig poly- morphonuclear leukocytes (PMN) when incubated in medium containing 50 mM galactose displayed substantially less killing of Escherichia coli than when incubated in medium with 5 mM glucose. Impaired bactericidal activity was de- pendent upon galactose concentration but could be partially averted by supplementing the galactose-containing medium with 15 mM glucose. Phagocytic activities of guinea pig PMN and peritoneal macrophages were assayed by following ingestion of 32P-labelled E. coli and were also depressed by elevated galactose. Galactose was readily epimerized to glucose by resting PMN, and this conversion was stimu— lated by phagocytosis. Incubation of macrOphages in the presence of galactose resulted in depletion of intracel— lular levels of adenosine 5'—triphosphate as well as other metabolites. 13 14 Introduction Although bacterial infection (5,22) and E. coli septicemia (9,17) are frequently reported among infants deficient in galactose—l-phosphate uridylyltransferase (EC 2.7.7.10), little is known about the functional integ- rity of the host defense mechanism during galactosemia. Among patients fed diets restricted in galactose content both cellular and humoral immune systems are believed to be functional (15), but there is no pertinent information about the function of these systems in patients during galactose toxicity. In the present communication, we suggest that in- creased susceptibility to infection may result from im- paired phagocyte function during galactosemia. We demon— strate that elevated levels of galactose in vitro exert direct inhibitory effects upon the bactericidal activities of human and guinea pig polymorphonuclear leukocytes (PMN) and upon the phagocytic activities of guinea pig PMN and peritoneal macrophages. We also present data that indicate that the action of galactose on phagocytes is not primarily osmotic but may be attributed to changes in the intracel- lular levels of adenosine 5'—triphosphate (ATP) or other metabolites (results of preliminary studies were reported in Fed. Proc. 55:578, 1975). 15 Materials and Methods Phagocyte preparations. Human blood samples were provided from 10 healthy adult volunteers through the courtesy of the American Red Cross Regional Blood Bank. Guinea pig blood was obtained from adult male guinea pigs (Connaught Laboratories, Ltd., Willowdale, Ontario) fed a commercial diet and water with 0.04% L-ascorbate ad libitum. Guinea pig serum, fetal calf serum, Medium 199, and Hanks balanced salt solution were purchased from Grand Island Biological Corp., Grand Island, N.Y. D—glucose and D—galactose were purchased from Mallinckrodt Chemical Works and Sigma Chemi— cal Co., respectively, of St. Louis, Mo. Human and guinea pig PMN were isolated from 10 ml samples of venous blood supplemented with 100 U of heparin (grade I, Sigma) per m1 and 2 ml aliquots of 6% Dextran 250 (Pharmacia Fine Chamicals, Uppsala, Sweden) in phos— phate buffered saline, pH 7.4. After one hour of sedi- mentation at room temperature, leukocyte-rich supernates were decanted and centrifuged at 250 xg for 5 min. Pre- parations were washed once with 0.87% NH 01 to lyse re— 4 maining erythrocytes (2), and leukocytes were suspended in Krebs-Ringer phosphate solution without calcium, pH 7.4, containing 10% autologous serum. Cell suspensions were gassed with 95:5 mixture of 0 002, adjusted to the 2: appropriate carbohydrate concentration and incubated for 1 hour at 57°C prior to assaying for phagocytic or bac- tericidal activities. Amounts of D—glucose used as l6 supplement were calculated after determining by gas chromatography the levels of carbohydrate contributed by sera (24). Levels of glucose in sera varied from 4 to 6 mM, whereas levels of galactose were too low to be quantitated (less than 0.05 mM). Guinea pig peritoneal macrophages were obtained 4 days after injecting, intraperitoneally, 5 ml of sterile 1% caseinate in saline. Exudates were obtained by flushing peritoneal cavities with Hanks balanced salt solution and centrifuged as described above. Macrophages suspended in Medium 199 plus 20% guinea pig serum were allowed to spread on plastic culture dishes for 2 hours at 57°C and were rinsed twice with fresh medium. Cells were cultured aerobically in this medium with appropriate concentrations of carbohydrate for 2 hours prior to use. Cell viability was determined by exclusion of 0.04% trypan blue.‘ Differen- tial cell counts, performed on Wright stained smears, showed homogeneities of human PMN, guinea pig PMN, and macrophage pOpulations to be between 70 and 85%, greater than 95%, and greater than 98%, respectively. Photomicrographs were taken of guinea pig peritoneal macrophages using a Universal model Zeiss microscope with phase optics (Figure 5a). The same microscOpe equipped with a BG-12 fluorescence excitation filter, as well as an 0G-550 emission filter, was employed to take bright-field fluorescence photomicrographs of guinea pig exudate cells (Figure 5b). The latter cell pOpulation was stained for 17 50 min with 50 uM acridine orange. Both photographs are of cells in the presence of Krebs-Ringer phosphate solu- tion, pH 7.4, with 5.0 mM glucose. Bacteria. E. coli grown aerobically at 57°C in thiogly- colate broth (Baltimore Biological Laboratory, Cockeys- ville, Md.) were harvested in late log phase and concen- trated using a 0.45 pm membrane filter (Millipore Corp., Bedford, Mass.). Bacteria were washed extensively with Krebs-Ringer phosphate solution and suspended in autolo- gous sera or guinea pig sera for 15 min before exposure to phagocytes. Radioactive bacteria were prepared in the same manner, except that 10 uCi of carrier-free [32P] orthophosphate (New England Nuclear Corp., Boston, Mass.) was included per ml of growth medium. Growth of E. coli in 6.0 m1 suspensions supplemented with 10% fetal calf serum and carbohydrate was monitored by following absorbance at 620 nm using a Klett-Summerson colorimeter. Phagocytic and bactericidal assays. Phagocytic activities, expressed as the average number of bacteria ingested per phagocyte, were determined by monitoring the uptake of viable 32P-labelled E. coli into guinea pig PMN in sus— 6 cells/ml) and into macrophages in pensions (6.2 x 10 monolayer on glass cover slips (9.1 x 105 cells/cover slip). Using a method similar to that described by Lentz and DiLuzio (11), 2.0 ml aliquots of PMN suspension were l8 placed in siliconized Erlenmeyer flasks (25 ml) and in- cubated at 57°C in a shaking Dubnoff water bath. Upon addition of E. coli (164 cpm per 106 bacteria), 0.2 ml volumes were removed from each flask at 20, 60, and 120 min. Prior to removal of each aliquot, cells were thorough- ly resuspended by briefly shaking each flask with a Vortex mixer. Volumes were transferred to 5.0 ml siliconized centrifuge tubes, and PMN were washed three times by re- peated suspension and centrifugation at 250 xg for 10 min. Pellets of washed cells were directly dispersed in 10 ml of Bray's solution and counted using a Beckman CPM-100 liquid scintillation counter. Cover slips containing peritoneal macrophages were re— moved at the same time intervals and at 180 min from cul— ture media containing labelled E. coli (558 cpm per 106 bacteria). Cover slips were rinsed three times with fresh media and transferred directly into scintillation vials with Bray's solution. All assays were performed in tripli- cate on cells pooled from three guinea pigs. Phagocytic activities were also determined by observ- ing the uptake of polystyrene latex particles by peritoneal macrOphages. Cells were cultured for 4 hours at 57°C in the presence of Medium 199 supplemented with 20% fetal calf serum. MacrOphages were allowed to adhere to cover slips, and at one hour prior to assay, cover slips were transferred to the same medium containing elevated carbo- hydrate. Polystyrene latex particles were then added in 19 a 50 fold particle to cell ratio, and uptake was observed after an additional hour using a Zeiss microscope with phase Optics. Both the number of phagocytosing cells and the number of particles ingested per cell were determined. Killing of bacteria by human and guinea pig PMN was assayed by exposing 1.0 ml cell suspensions (1.5 x 107 PMN/ml) in sterile Erlenmeyer flasks to serum-treated bacteria and incubating at 57°C in a shaking water bath. Aliquots of 0.2 ml, removed immediately and at 120 min, were serially diluted with sterile distilled water and then mixed with sterile 4% Trypticase soy agar plus 0.5% glucose in petri dishes at 45 to 50°C. After incubating each dish at 57°C for 24 hours, the number of viable bacteria per ml was determined by counting the number of colonies per dish and multiplying by the corresponding dilution factor. Each experiment was performed in tripli- cate using two plates per dilution with cells pooled from eight human donors or from at least three guinea pigs. Oxidation of glucose and galactose. [1-140] glucose, [6514C] glucose, and [1-140] galactose with specific activi— ties of 48.2, 55.8, and 45.0 mCi/mmol, respectively, were purchased from New England Nuclear Corp. Conversion of each substrate to 14CO2 was performed in sealed Erlenmeyer flasks at 57°C. Each reaction, run in triplicate, was initiated by adding 2.4 uCi of labelled carbohydrate with the apprOpriate amount of non-isotOpic carrier in 0.5 m1 20 of Krebs-Ringer phosphate solution to 0.5 ml suspensions 6 to 15.5 x 106 cells) with 20% guinea pig of PMN (5.7 x 10 serum. After one hour of incubation, suspensions were acidified to release bound CO2 by adding 0.2 ml of 20% H2804 and incubating at room temperature for an additional hour. Evolved CO2 was collected from the start of each experiment on folded strips of Whatman 5MM paper (1 x 5cm) with 0.2 ml of 20% KOH in plastic centerwells. Centerwells were transferred to scintillation vials with 10 ml of Bray's solution and counted. Oxidation of carbohydrate by phagocytosing cells was determined in the same manner, except that a 20 fold (bacteria per PMN) excess of serum- treated E. coli (heat-killed for 15 min at 90°C) was in— cluded in the 0.5 ml aliquot of labelled carbohydrate. Conversion of galactose to glucose by resting and phagocytosing PMN was determined by comparing the specific 14 C] glucose remaining after incubation 6 activities of [1- of cells (12.5 x 10 PMN) in media containing either 1.0 mM [1-140] glucose or [1-140] glucose with 10 mM galactose. After a 5 hr incubation at 37mc, cells were centrifuged from the medium, and neutral sugars were isolated from the supernants by adding 1.0 ml volumes of 0.5 N Ba(0H)2 and 5% SnSO4 (21). Specific activities were determined by converting labelled glucose to glucose-6—phosphate, separating it by paper chromatography, and dividing the counts per minute in this fraction by the number of micro- moles (5). 21 Metabolite assay_. Aliquots of peritoneal macrOphages (ap— proximately 5.0 x 107 cells) were cultured aerobically in plastic culture dishes at 57°C in the presence of Eagle basal medium (Grand Island Biological Co.) with 10% fetal calf serum and either 5.0 mM glucose or 5.0 mM glucose with 50 mM galactose. After 2 hr, cells in the monolayer were removed and centrifuged as described above. Pellets of cells were immediately frozen by placing centrifuge tubes in a bath of dry ice and is0pr0panol. Cells were stored at -80°C until homogenizing for 10 min at 4°C with 5 volumes of 5 N perchloric acid. Homogenates were centri- fuged at 5000 xg for 10 min to remove precipitated protein, and supernates were neutralized with a mixture of 2 N KCH, 0.4 M imidazole, and 0.4 M K01. Levels of protein were determined in each homogenate by the method of Lowry §t_al, (15), using bovine serum albumin (fraction V, Sigma) as a protein standard. Lactate was quantified spectrOphoto- metrically using lactate dehydrogenase (EC 1.1.1.27) and by monitoring the reduction of acetylpyridine adenine dinucleotide (Boehringer Mannheim Corp., New York, N.Y.) at 566 nm (6). All other soluble metabolites were determined fluorometrically employing pyridine nucleotide-coupled enzyme reactions as previously described (12). Lactate dehydrogenase, pyruvate kinase (ECEL7.1.40), and adenylate kinase (EC 2.7.4.5) were used sequentially to determine levels of pyruvate, adenosine 5'—diphosphate and adenosine 5'-mon0phosphate, whereas glucose-6—phosphate dehydrogenase 22 (EC 1.1.1.49) and hexokinase (EC 2.7.1.1) were used se- quentially to determine glucose-6-phosphate and ATP. Glyceraldehyde-5-phosphate, dihydroxyacetone phosphate, and fructose—l,6-bisphosphate were quantified using glycerol- 5-phosphate dehydrogenase (EC 1.1.1.8), triosephosphate isomerase (EC 5.5.1.1), and fructose-bisphosphate aldolase (EC 4.1.2.15), sequentially. All enzymes were purchased from Sigma except the latter three, which were obtained from Boehringer. Results The action of galactose upon bactericidal activity of phagocytes was observed with PMN from both human and guinea pig sources. When serum-treated E. coli and human PMN were incubated together in medium containing 5 mM glucose, a substantial decrease in the number of viable E. coli oc- curred over a 2 hr period (Table I, 0.005% survival). In- cubation of bacteria without phagocytes resulted in bac— terial growth (187% survival), and incubation of bacteria with PMN, but without added carbohydrate, resulted in sig— nificantly less bactericidal activity (97.4% survival). The use of 50 mM galactose as the sole carbohydrate source also failed to support normal killing of E. coli. More- over, in the presence of 5.0 mM glucose, 50 mM galactose severly inhibited bactericidal activity (Table I, 54.2% versus 0.005% survival); however, this inhibition could be almost completely averted by increasing the glucose 23 concentration from 5.0 mM to 15 mM in the presence of 50 mM galactose (0.052% survival). Similarly, when bac- teria and guinea pig PMN were incubated together (Table II), the inhibitory action of galactose in the presence of 5.0 mM glucose was found to be concentration dependent. This action could not be attributed to hyperosmolarity, because the impaired state was again prevented by elevating the glucose level to 15 mM, and morphological changes in PMN associated with hypertonicity (20) were not observed. Viability of PMN was not altered by incubation with galac— tose during these assays, and although some cell clumping did occur this was kept to a minimum by employing sili- conized glassware, and using Krebs—Ringer phosphate solu— tion without calcium. Since the systems under study in Tables I and II were dynamic, depending not only upon phagocytic and killing processes but also upon the rates of bacteria growth, it became necessary to establish whether galactose could be affecting the growth of E. coli. For this purpose, growth curves were investigated under similar conditions but without PMN (Figure 1). High levels of galactose slightly impaired the growth of organisms, and this effect could not be prevented by increasing glucose levels. Thus, im- pairment of both PMN function and bacterial growth occurred in the presence of galactose. It should be noted that the assay systems under study in Tables I and II were of similar design but differed 24 with respect to the species of PMN and the type of serum employed. The number of bacteria added to each system also varied, and this probably accounts for observed dis- crepancies between the percentage of survival of E. coli in these tables at comparable carbohydrate levels (54.2 versus 112% survival in the presence of 5 mM glucose plus 50 mM galactose with human and guinea pig PMN, respectively). The extent to which galactose affects phagocytosis was determined by monitoring the uptake of 32P-labelled E. coli into guinea pig PMN and peritoneal macrophages. PMN in- cubated in medium containing 5 mM glucose displayed most phagocytosis, whereas cells incubated in similar media with 15 to 50 mM galactose displayed 87 or 76% of this activity, respectively (Figure 2). Effects of galactose on phago- cytic activity were more pronounced when employing macro- phages (Figure 5), where the same relative amounts of galactose decreased uptake to 67 and 47%, respectively. Although inhibition of phagocytosis contributes to the loss of bactericidal activity, the extent of this inhibition did not appear sufficient to account for all the impairment in killing capacity. Visual observation of phagocytosing macrOphages showed that both the number of polystyrene latex particles in- gested per phagocytosing cell and the number of phagocy- tosing cells were lower in the presence of 50 mM galactose (Table III). Addition of 10 mM glucose to the medium con- taining 50 mM galactose appeared to partially relieve this 25 depression. However, higher levels of glucose without galactose were also inhibitory. Moreover, the number of ingested particles varied greatly (from 1 to 50) between phagocytosing cells. To further investigate losses of phagocytic and bac— tericidal function, the action of galactose upon meta- bolic processes associated with these functions were studied. Effects of galactose upon the oxidation of [1-140] glucose and [6-140] glucose by guinea pig PMN are presented in Figure 4. Although 10 mM galactose reduced the amount of 14CO evolved from [1-140] glucose by both 2 resting and phagocytosing cells, a 2.5 fold increase in glucose oxidation associated with phagocytosis was rela- tively unaffected. Evolution of 14CO from [6-140] glu- 2 cose by resting cells was 50 fold less than that evolved from [1-140] glucose and was impaired by galactose to a greater extent. Thus, either galactose inhibited glucose transport, and/or its oxidation, or galactose was converted to glucose in sufficient amounts to lower its specific radioactivity. The latter explanation appears correct (Table IV). When resting or phagocytosing PMN were incubated for 5 hours in media with 1.0 mM labelled glucose, the specific activi- ties of residual glucose were essentially equal. Addition of 10 mM galactose to either medium, however, resulted in decreased specific activities. That phagocytosing cells converted more galactose to glucose than resting cells was 26 suggested by these data and further verified by monitoring the evolution of 14CO from [1-140] galactose. Resting and 2 phagocytosing cells incubated in media with 1.0 mM labelled galactose oxidized 0.59 and 1.70 nmole of galactose per hour per 106 cells, respectively. These oxidation rates were approximately 10 fold less than those for glucose (data not shown) and appeared to be insufficient for normal bactericidal action of PMN when 50 mM galactose was pro- vided (Table I). Peritoneal macrophages cultured for 2 hr in basal medium (Eagle) containing 5 mM glucose, 10% fetal calf serum, and 50 mM galactose showed significant decreases in intracellular ATP (P<:0.005) as well as elevations in levels of adenosine 5'-diphosphate and adenosine 5'-monophosphate (Table V). Cell viability was lower under these conditions, and levels of all glycolytic intermediates, with the excep- tion of pyruvate, were depressed. Discussion The present results demonstrate that elevated levels of D-galactose inhibit the phagocytic activities of guinea pig PMN and peritoneal macrophages as well as the bacteri- cidal activities of guinea pig and human PMN. Since levels of galactose employed in this study (5 to 50 mM) are ob— served in infants deficient in galactose-l-phosphate uri— dylyltransferase (14,16), it is reasonable to assume that phagocyte function is also impaired in these patients when 27 blood galactose levels remain elevated. Such impairment of phagocyte function could severely compromise the host defense system and may represent an underlying cause for increased susceptibility to E. coli infection observed among galactosemics. The effects of galactose upon phagocytes are prob- ably very similar to the effects of other carbohydrates previously implicated in impairing phagocyte function. Sanchez §t_§1. (18) reported that elevated levels of blood glucose depressed phagocytic activity of human PMN toward Staphylococcus epidermidis, and Bagdade and co-workers (1) found that the ability of human PMN to ingest and destroy type 25 pneumococcus was impaired by increasing glucose levels to 55 mM. In studying the correlation between sus— ceptibility to infection and poorly controlled diabetes mellitus, Drachman §t_§l. (4) followed the ingestion of type 25 pneumococcus by rat PMN and found that phagocytosis was depressed both in vivo by hyperglycemia and in vitro in crowded cell suspensions by 45 mM concentrations of glucose, fructose, sucrose, mannose, xylose, and arabinose. Noting previous reports that solute concentrations above 40 milliosmolal inhibited phagocyte function (20), Drachman concluded that the action of elevated carbohydrate was primarily osmotic. Our results, however, indicated that the action of galactose upon phagocytes was not related to osmolality, since the inhibitory effect of 50 mM galactose upon 28 bactericidal activity could be averted by elevating glu- cose levels to 15 mM, and since the morphological changes in PMN and macrophage associated with hypertonicity were not seen. Correlations between impaired activity and PMN viability, as well as between activity and cell clumping, were also not observed. Since phagocytosis by PMN (19), as well as by macro- phage (8), is known to depend largely upon energy supplied by glycolysis, suppression of this metabolic pathway may represent a primary action of galactose. Depressed levels of intracellular ATP and glycolytic intermediates were found in guinea pig peritoneal macrophages when incubated with elevated levels of galactose (Table V), and similar results were reported to occur in other tissues during galactose toxicity (5). Suppression of glycolysis could arise from either of the following mechanisms: (i) an inhibition of glucose transport into the phagocyte, or (ii) an accumulation of galactose, or one of its metabolites, intracellularly with subsequent inhibition of one or more glycolytic enzymes. However, both galactose and galactose- l-phosphate are concentrated by leukocytes (7,25), and we have recently found that glucose, but not galactose, com- petitively inhibits the uptake of[G—3H]2—deoxyglucose into guinea pig PMN (see Chapter III). Thus, competition for transport is an unlikely explanation for galactose action in these cells. The rapid conversion of galactose to glucose in 29 phagocytes (Table IV) can be accounted for by the combined action of the Lelior pathway enzymes and either a specific or nonspecific phosphatase. This pathway, while function- ing to dilute the glucose label in our oxidation experi— ments, was stimulated by phagocytosis and may have con- tributed to depletion of intracellular ATP via a futile adenosine triphosphatase action. Similar cycles operating during galactose neurotoxicity have been previously re- ported (10). 10. 11. 12. l5. 140 150 30 References Bagdade, J.D., Root, R.K., and Bulger, R.J. (1974) Diabetes 2 , 9-15 Bigley, R.H., and Stankova, L. (1974) J. Exp. Med. 122, 1084-1092 Donnell, G.N., Bergren, W.R., and Ng, W.G. (1967) Biochem. Med. ;, 29—55 Drachman, R.H., Root, R.K., and Wood, W.B. (1966) J. Exp. Med. 124, 227-240 Granett, S.E., Kozak, L.P., McIntyre, J.P., and Wells, W.W. (1972) J. Neurochem. l2, 1659—1670 Holzer, H., and Soling, H. (1965) In Methods of Enzymatic Analysis pp. 275—277 (Bergmeyer, H.V., ed.) Academic Press, New York, N.Y. Kalant, N., and Schucher, R. (1962) Can. J. Biochem. Physiol. 4;, 849-858 Karnovsky, M.L., Simons, S., Glass, E.A., Shafer, A.W., and Hart, P.D. (1970) In The Mononuclear Phagocytes pp. 105—120 (Von Furth, R., ed.) F.A. Davis 00., Philadelphia, Pa. Kelly, S.J., Burns, J., and Desjardins, L. (1974) Am. J. Epidemiol. 22, 8-15 Kozak, L.P., and Wells, W.W. (1971) J. Neurochem. ;§, 2217-2228 Lentz, P.E., and DiLuzio, N.R. (1974) In Methods in Enzymology pp. 647-655 (Fleisher, S., and Packer, L., eds.) Academic Press, New York, N.Y. Lowry, 0.H., and Passonneau, J.V. (1972) In A Flexible System of Enzymatic Analysis pp. 146-149, 151-153, and 167-168 Academic Press, New York, N.Y. Lowry, 0.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) J. Biol. Chem. 125, 265—275 Medline, A., and Medline, N.M. (1972) Can. Med. Assoc. J. 107, 877-878 Nadler, H.L., Inouye, T., and Hsia, D.Y.Y. (1969) In Galactosemia pp. 127-159 (Hsia, D.Y.Y., ed.) Charles C. Thomas, Springfield, Ill. 16. 17. 18. 19. 20. 21. 22. 25. 24. 51 Pennington, J.S., and Prankerd, T.A.J. (1958) Clin. Sci. l1, 585-391 Quan-Ma, R., Wells, H.J., Wells, W.W., Sherman, F.E., and Egan, T.J. (1966) Am. J. Dis. Child. 112, 477-478 Sanchez, A., Reeser, J.L., Law, H.S., Yahiku, P.Y., Willard, R.E., McMillian, P.J. Cho, S.Y., Magie, A.R., and Register, V.D. (1973) Am. J. Clin. Nutr. 22, 1180-1184 Sbarra, A.J., and Karnovsky, M.L. (1959) J. Biol. Chem. 254, 1555-1562 Sbarra, A.J., Shirley, W., and Baumstark, J.S. (1965) J. Bacteriol. §§, 306-313 Somogyi, M. (1945) J. Biol. Chem. lgg, 69—73 Stiehm, E.R. (1975) Am. J. Dis. Child. lgg, 458-445 Tedesco, T.A., and Mellman, W.J. (1969) J. Clin. Invest. 4§, 2590-2597 Wells, w.w. (1974) In Clinical Biochemistry pp. 951- 945 (Curtius, H.C., and Roth, M., eds.) vol. 2 Walter de Gruyter, New York, N.Y. 52 TABLE I. Effect of galactose on bactericidal activity of human polymorphonuclear leukocytesa Carbohydrate Content Average Viable of Medium (mM) . coli/m1 Percent Glucose Galactose PMN 0 Min. 120 Min. Survival 5.0 0 + 3.50 x 108 1.81 x 104 0.005 5.0 0 — 1.22 x 108 2.29 x 108 187.0 0 0 + 4.81 x 108 4.68 x 108 97.4 0 30 + 3.39 x 108 3.25 x 108 96.0 5.0 30 + 3.63 x 108 1.24 x 108 34.2 15.0 30 + 3.87 x 108 1.95 x 105 0.052 aHuman PMN (1.3 x 107 cells) were incubated at 57°C in 1.0 ml of media containing 10% autologous sera and the indi— cated levels of carbohydrate. were added after one hour of preincubation. ment was performed in triplicate. Serum-treated bacteria Each experi- 33 TABLE II. Effect of galactose on bactericidal activity of guinea pig polymorphonuclear leukocytes8 Carbohydrate Content Average Viable of Mediumij)» E coli/ml Percent Glucose Galactose PMN 0 Min. 120 Min. Survival 5.0 0 + 2.02 x 106 1.72 x 104 0.571 5.0 0 - 2.93 x 106 4.06 x 106 139.0 5.0 7.5 + 3.24 x 106 1.32 x 106 4.07 5.0 15.0 + 2.42 x 106 9.70 x 106 40.1 5.0 30.0 + 2.49 x 106 2.80 x 106 112.0 15.0 30.0 + 2.62 x 106 4.13 x 104 1.58 aGuinea pig PMN (1.5 x 107 cells) were incubated at 57°C in 1.0 ml of media containing 10% guinea pig sera and the indicated levels of carbohydrate. Serum-treated bacteria were added after one hour of preincubation. Each experi— ment was performed in triplicate. 34 TABLE III. Influence of carbohydrate on phagocytosis of latex particles by guinea pig peritoneal a macrophages Carbohydrate Content % Phagocytosing Particles Ingested of Medium2(mM) Cellsb Per Phagocytosing Glucose Galactose CellC 5.0 0 90.9 8.0 i 0.8 (37) 10.0 0 88.5 4.8 i 0.4 (46) 20.0 0 85.7 4.5 i 0.4 (47) 30.0 0 80.9 4.5 i 0.4 (47) 5.0 10.0 76.3 4.3 i 0.4 (95) 5.0 30.0 74.8 3.3 i 0.3 (50) 10.0 30.0 89.0 5.4 i 0.7 (34) aPeritoneal macrophages were cultured aerobically for 4 hr at 57°C in the presence of Medium 199 supplemented with 20% fetal calf serum. One hr prior to assay dishes were adjusted to the appropriate carbohydrate concentration. To commence assay, a 50 fold particle to cell excess of polystyrene latex particles were added. Counting was per- formed by phase contrast microscOpy at 800x, one hr after particle addition. bPercent of macrophages that have ingested at least one particle. CAverage number of particles ingested per phagocytosing cell i SEM. Brackets indicate the number of cells ob- served. 35 TABLE IV. Conversion of galactose to glucose by guinea pig PMN as indicated by the specific activity of [l-l4C]D-glucosea Carbohydrate Content Cell Glucose Specific Activity of Medium (mM) State After 5 Hours with PMN Glucose Galactose (cpm x 10-5) i SEM 1.0 O Resting 5.05 i 0.55 1.0 O Phagocytosing 5.08 i 0.55 1.0 10 Resting 2.07 i 0.16 1.0 10 Phagocytosing 1.45 i 0.29 aPMN (12.5 x 106 cells) were incubated at 37°C in 1.0 ml of media containing 10% guinea pig sera, 2.0 uCi of [1-140] D-glucose, and the indicated amounts of carbohydrate. After 5 hours, specific activities of extracellular glucose were determined in triplicate. - .LA 97) [ I {—4 m 36 TABLE V. Effect of galactose on levels of adenine nucleo- tides and glycolytic intermediates in guinea pig peritoneal macrophages8 Intracellular Levels (nmoles/mg protein i SEM) After Cells were Incubated With: Metabolite 5 mM Glucose 5 mM Glucose + P 50 mM Galactose ATP 6.88 i 0.21 2.16 i 0.12 0.005 ADPb 0.73 2.80 —— AMPb 0.80 4.47 __ Glucose-6-phos- phate 0.95 i 0.02 0.70 i 0.01 0.01 Fructose-1,6- diphosphate 2.69 i 0.17 2.54 i 0.06 NS Dihydroxyacetone phosphateb 0.48 Below 0.05 ‘ —- Glyceraldehyde—5- phosphate 0.45 Below 0.05 -— Pyruvateb 1.76 2.06 —— Lactate 545 i 12.2 406 i 4.9 0.01 Decrease in Cell Viability during Incubation 2.8% 16.0% aMacrophages (5.0 x 10'7 cells) were incubated at 57°C for 2 hours in basal medium (Eagle) with 10% fetal calf serum and the indicated levels of carbohydrate. Levels of sig— nificance (P) were calculated by the Student's two-tailed t-test and are given for values determined in triplicate for three separate experiments; NS refers to means not signifi- cantly different. bValues are the means of triplicate determinations from one study. 0.100(- ABSORBANCE 620 nm 0.003 Figure 1. 0.031 0.010 37 o 8.0 12 HOURS OF GROWTH Semilogarithmic plot of E. coli growth in high galactose media. Bacteria were grown at 57°C in 6.0 m1 of KRPS with 10% fetal calf serum and a) 5 mM lucose, b) 5 mM glucose and 15 mM galactose, c 5 mM glucose and 50 mM galactose, or d) 15 mM glucose and 50 mM galactose. 58 3.0 15 3 UPTAKE / CELL 5.0 Figure 2. l I so 150 TIME (min) Inhggitory effect of galactose on phagocytosis of .P-labelled E. coli by guinea pig PMN. 20 Labelled bacteria were added at zero time to PMN suspensions in media with 10% guinea pig sera, and 0, 15, or 50 mM galactose. Uptake/ cell represents the average number of bacteria ingested per phagocyte. Each point is the mean of triplicate determinations and bar line indi- cate S.E.M. 59 3.0 - Figure 5. u: G) 7.5 UPTAKE / CELL 20 so 150 160 TIME (min) Inhibitory effect of galactose on phagocytosis of 32P—labelled E. coli by guinea pig peritoneal macrophages. Labelled bacteria were added at zero time to macrophage monolayers in media with 10% guinea pig sera and 5.0 mM glucose with 0, 15, or 50 mM galactose. Uptake/cell represents the average number of bacteria ingested per phagocyte. Each point is the mean of triplicate determina- tions and line bars indicate S.E.M. 4O 3.0 103 CPM 1.0 I Figure 4. - 0.03 0.02 e 0.01 I'//////////////////////////A +'///////////////////A IW +7////A Effect of galactose on oxidation of[1-14C]glu- cose andl6-14C]glucose by guinea pig PMN. PMN were incubated for 1 hr in media with 10% guinea pig sera and either 1.0 mM glucose (-) or 1.0 mM glucose with 10 mM galactose (+) Evolved l4002, expressed as 103 CPM per 106 cells, was determined for both resting (r) and phagocytosing (p) PMN. Each experiment was performed in triplicate on cells from each of 5 guinea pigs. Bar lines equal S.E.M. Figure 5a. 'Phase contrast photomicrograph of guinea pig macrophages. Figure 5b. Bright—field fluorescence photomicrograph of guinea pig exudate cells stained with acridine orange. CHAPTER II INHIBITORY ACTION OF GALACTOSE 0N PHAGOCYTES FROM NORMAL ANDCEIACTOSEMIC CHICKS Abstract The inhibitory effect of galactose on phagocyte func— tion was investigated in normal and galactosemic chicks by monitoring the in vitro killing of Escherichia coli by leukocytes and the in vivo clearance of colloidal 1251- labelled bovine serum albumin (125I—BSA) from the circula— tion. Elevated levels of galactose (50 mM) impaired the bactericidal activities of both control and galactosemic leukocytes. However, the latter cells were more susceptible to the galactose dependent inhibition. Galactosemic leuko- cytes displayed near normal bactericidal activity when as- sayed in vitro under simulated normal conditions. Phagocytic indexes, obtained from data on the clear- ance of colloidal 125I-BSA, were calculated to be 0.0555 and 0.0297 for control and galactosemic chicks, respectively. Chicks fed a control diet displayed a logarithmic clearance of colloid with postinjection time, whereas this relation- ship was not as apparent when galactosemic chicks were employed. Moreover, galactose impaired phagocytic functions 42 a3 XT, Q 45 of both circulating leukocytes and tissue-filled macrOphages as well as the overall development of the reticuloendothel- ial system. Introduction Previous studies have demonstrated that elevated levels of galactose, as encountered during galactosemia, are in— hibitory to phagocyte function in vitro (Chapter I). These studies were performed with elicited guinea pig macrophages and with polymorphonuclear leukocytes isolated from the peripheral blood of normal guinea pigs or human donors. These studies were not performed with phagocytes from galactosemic sources. Therefore, data regarding galactose action in galactosemic cells and in vivo were not obtained. To acquire this information, experiments were designed using leukocytes from galactosemic chicks and using a standard test of reticuloendothelial system (RES) function in these animals. Major objectives of the present research are threefold: l) to determine whether the killing of E. coli by leukocytes of normal and galactosemic animals is impaired to similar extents by elevated levels of galactose, 2) to test whether galactosemia impairs in vivo phagocytic activity of tissue- fixed macrOphages, as judged by the intravascular clearance of colloidal 125I-labelled bovine serum albumin (125I—BSA), and 5) to ascertain whether experimentally induced galacto- semia in developing Leghorn cockerels is a good model S W.,...u A VI 44 system for studying the effects of galactose upon human phagocytes. Materials and MethodS- Animals and materials. Day-old Leghorn cockerels (Gallus domesticus) were obtained through the generosity of MacPherson Hatchery and Rainbow Trail Hatchery of Ionia and St. Louis, Michigan, respectively. The basal 2 diet of Rutter gt_§;. (l) was fed ad libitum to both control and galactosemic chicks. However, in the latter case, this diet contained either 50% or 50% (w/w) D-galactose sub- stituted for an equal amount of cerelose (D—glucose mono— hydrate). Chicks were kept in brooders at 52°C. Adult chicken serum, trypan blue, and Hank's balanced salt solution were purchased from Grand Island Biological 00., Grand Island, N.Y., whereas all other reagents were obtained from Sigma Chemical Co., St. Louis, Mo. Radio- active iodine (NalZSI), as well as other materials neces- sary for protein labelling, were kindly provided by Dr. Pamela Fraker of Michigan State University. Phagocyte2preparations. Leukocytes were isolated from 10 ml samples of whole blood obtained by cardiac puncture of ether anestetized chicks. Samples were pooled from either six control chicks or eight galactosemic chicks that had been fed the 50% galactose diet for 48 hours. Blood was supplemented with heparin (100 U/ml) and with dextran 45 (1.2% w/v). After 1 hour of sedimentation at room tempera- ture, leukocyte—rich supernatants were decanted and centri- fuged at 250 xg for 5 min. Preparations were washed once with 0.87% NH4Cl to lyse remaining erythocytes, and leuko- cytes were suspended in Hank's balanced salt solution con- taining 10% adult chicken serum. Cell suspensions were gassed with a 95:5 mixture of 02: 002, adjusted to the ap- propriate carbohydrate concentration and incubated for 1 hour at 57°C prior to assaying for bactericidal activity. Amounts of glucose and galactose used as supplements were calculated after determining by gas chromatography the levels of carbohydrate Contributed by adult chicken sera (see Appendix). Levels of galactose were too low to be quantitated (less than 0.05 mM). Bacteria. A phage resistant strain of E. coli, designated E. coli B/r (ara'), was kindly provided by Dr. R. Anderson of Michigan State University. Bacteria were grown aero- bically at 57°C in thioglycolate broth and were harvested in late log phase using a 0.45 pm membrane filter. Bac- teria were washed extensively with Krebs-Ringer phosphate solution and were suspended in adult chicken serum for 15 min before exposure to leukocytes. Bactericidal assays. Killing of bacteria by chick leuko- cytes was assayed by exposing 1.0 ml cell suspensions in 8 sterile Erlenmeyer flasks to 10 serum-treated bacteria and incubating at 58°C in a shaking water bath. Aliquots 46 of 0.2 ml, removed immediately and at 120 min, were seri— ally diluted with sterile water and then mixed with sterile 4% Trypticase soy agar plus 0.5% glucose in petri dishes at 45 to 50°C. After incubating each dish at 37°C for 24 hours, the number of viable bacteria per ml was determined by counting the number of colonies per dish and multiply- ing by the corresponding dilution factor. Each assay was performed in triplicate using two plates per bacterial dilution and using between 4.9 x 107 and 6.4 x 107 leuko- cytes per m1. Quantities of leukocytes were determined by counting cells in a hemocytometer using phase optics. Leukocyte viability was measured by the exclusion of 0.04% trypan blue (2) . Lntravascular clearance of colloidal 125I-BSA. Radioac— tiVe BSA (1.88 x 105 cpm/11g) was prepared according to the method of Fraker and Speck (5) by exposing 100 11g of BSA 0 M potassium iodide to 15.5 )lCi of in 100 pl of 6.7 x 10’:L N531251 in the presence of 0.4 ug tetrachloroglycoluril. Iodination was performed for 5 min at 4°C and was termi— hated by decanting the reaction solution into test tubes Without tetrachloroglycoluril. Solutions were extensively die«:Lyzed against borate-saline buffer, pH 8.2, and finally agaimst Krebs-Henseleit bicarbonate solution, pH 7.4. Aliquots of protein precipitated with 15% (w/v) tri- chlOrcacetic acid prior to and after dialysis contained 54-7% and 99.2% of the soluble radioactivity, respectively. 47 Colloidal 125I-BSA was prepared by heat denaturing 5.0 m1 of a 1.0% BSA solution (4) containing the above labelled material. The solution was adjusted to pH 10.0 by adding 0.2 N NaOH, heated to 79°C for 20 min, and cooled on ice. The pH was further reduced to the isoelectric point of BSA by adding 0.2 N HCl, and the colloidal suspension was centrifuged at 22,000 xg for 15 min at 4°C. The resulting pellet was washed twice and resuspended in 1.5 ml of Krebs- Henseleit bicarbonate solution containing 150 U of heparin. Particle diameters ranged from 0.1 to 5 pm as judged by visual observation with a Zeiss microscOpe. 125I-BSA was Intravascular clearance of colloidal determined using 10 day old chicks fed either a control diet or a 50% galactose diet from 4 days of age. Each chick was anestetized with diethyl ether, and the thoracic cavity was surgically opened to expose the heart. An injection of colloidal 125I-BSA (50 ul/loo gm chick weight) was immedi- ately given via a 25 gauge needle into the left ventricle of the heart, and 0.1 ml aliquots of whole blood were re- moved at 1.0 min intervals from the right ventricle. Aliquots of blood were placed directly into vials and counted for 10 min with a Beckman Biogamma scintillation spectrometer set for 1251 counting. Levels of 125I—BSA remaining in the blood at each time were determined for 10 control chicks and 10 galactosemic chicks. Whole body, spleen and liver weights were noted for each animal, and subsequently, the distribution of counts in each liver and v..- .47 A Kuw~ :. 48 spleen was determined. Control chicks used in this experiment displayed blood glucose levels of 12.5 i 1.2 mM, whereas galactosemic chicks showed blood glucose and galactose levels of 14.9 i 2.6 mM and 18.4 i 2.6 mM, respectively. The latter chicks were demonstrating galactose neurotoxic behavior. RES development. Spleens and livers were surgically removed from control chicks and galactosemic chicks fed a 50% galac— tose diet from 7 days of age. Organs were washed once with sterile saline, blotted dry, and immediately weighed using an analytical balance. In a separate experiment, chicks were placed on each diet at 4 days of age and immunologi- cally stressed by injecting, i.p., 107 viable E. coli. Organ weights were determined at 12 days of age and com— pared to the weights of organs from unstressed animals. At least six chicks were employed in each group. Results Effect of galactose on bactericidal activity. The inhibi— tory action of galactose upon the bactericidal activity of chick leukocytes in vitro is illustrated in Figure l. Leukocytes from 9 day old chicks, fed a control diet, killed the most serum treated E. coli during a two hour incubation (8.5% bacterial survival), while leukocytes from chicks, fed a 50% galactose diet, displayed near normal bactericidal activity in the absence of galactose (15.4% bacterial 49 survival). Substantially lower activities, however, were found in both groups when cells were incubated and assayed with 50 mM galactose. Killing of E. coli by galactosemic cells was greatly impaired in this case (85.5% bacterial survival), whereas killing by control cells was impaired to a lesser extent (56.2% bacterial survival). The presence of 50 mM galactose did not affect the growth of E. coli nor the viability of chick leukocytes under these conditions. Effect of galactose on clearance of colloid. The effect of galactose in vivo upon intravascular clearance of colloidal 125I-BSA is shown in Figure 2. Ten day old chicks fed a control diet displayed a logarithmic clearance of colloidal protein with postinjection time. However, this relation- ship was not as apparent from the data on galactosemic chicks. The former data was easily fit to the following empirical formula for RES clearance (5): _ -kt C .. CO 10 where C equalled CPM/ml blood postinjection, CO equalled CPM/ml blood at zero time, t represented postinjection time, and k was the global phagocytic index, a measure of intravascular clearance. Thus, values of k were equiva- lent to the Opposite of the slope of the logarithmic form of this equation (log C = —kt + log 00) and were directly obtained from the lines in Figure 2. Both lines in this 5O figure were obtained by the method of least squares. Control chicks and galactosemic chicks displayed global phagocytic indexes (k) of 0.0555 and 0.0297, respectively. Whole body weights of galactosemic chicks (57.8 i 1.9 gm) were significantly lower than the weights of control chicks (87.5 i 2.57 gm) during this experiment. Similarly, the weights of galactosemic livers (2.02 i 0.06 gm) and spleens (48.6 i 0.4 mg) were substantially lower than the weights of control livers (5.45 i 0.15 gm) and spleens (99.0 i 0.6 mg). These data in combination with values for k permitted calculation of corrected phagocytic indexes (a) by the following equation (5): W k1/3 GK: w + L + S where W, L, and S corresponded to whole body, liver, and spleen weights, respectively. Corrected phagocytic in- dexes for control chicks and galactosemic chicks were 0.558 and 0.299, respectively. By the termination of each intravascular clearance study (approximately 6 min postinjection), livers and spleens of control chicks acquired 28.2% and 1.85%, respec- tively, of the total injected colloidal 125I-BSA. On the other hand, galactosemic livers and spleens sequesterred 25.4% and 1.85% of the dose, respectively. RES development. The effect of galactose diet on development of liver and spleen is illustrated in Figure 5. Chicks on control diet showed normal increases in spleen and liver 51 weights, whereas chicks on 50% galactose diet displayed ab— normal weight increases. Significant differences between these groups were apparent by 14 days of age. Whole body weight was also lower in galactosemic chicks (data not shown). However, liver and spleen weights were depressed to a greater extent. Table I shows that spleen weight of control chicks was responsive to E. coli injection. Increases in spleen weight were not found, however, with galactosemic chicks, and in- creases in liver weights of either group were not observed following the bacterial stress. Discussion Although bactericidal activities of leukocytes from both control and galactosemic chicks were inhibited by 50 mM galactose in vitro, activities of leukocytes from the latter source were affected to a greater extent (Figure 1). These observations demonstrated that the inhibitory actions of galactose were not limited to phagocytes of mammalian origin (Chapter I) and that the bactericidal functions of galacto- semic cells were previously compromised in vivo. The latter finding was not, however, a result of permanent cellular damage since near normal bactericidal activity was observed by galactosemic cells in the absence of galactose. These observations strongly suggested that galactosemic cells contained elevated levels of galactose and/or its metabolites which as glycolytic antagonists could inhibit "f t. (I) 52 phagocytosis (6) and which as free radical scavengers could impair oxidative bactericidal activity (7). Moreover, both galactose and galactitol were observed in galactosemic leukocytes (data not shown), and the diffusion of these agents from galactosemic cells could be a mechanism for the restoration of bactericidal activity. Since levels of galactose-l—phosphate uridylyltrans- ferase (EC 2.7.7.10) are low in both control and galacto- semic chick leukocytes (8), the existence of a futile ATPase cycle converting galactose to glucose (Chapter I) is unlikely in these cells. Nevertheless, a similar ATPase cycle, involving the cyclic phosphorylation and dephosphory- lation of galactose, is found in galactosemic chick brain (9) and could be present in galactosemic chick phagocytes. Such a cycle might occur in phagocytes by the action of either a specific or non-specific phosphatase upon galactose— l-phosphate. Clearance of colloidal 125I-BSA from the circulation of control chicks (Figure 2) obeyed an exponential function commonly employed to define RES activity in other species (5,10). Intravascular clearance by galactosemic chicks, on the other hand, did not follow such a clear relationship, and this was probably a result of experimental error rather than a delay in the onset of phagocytosis. Global phago- cytic indexes for RES function of control and galactosemic chicks were calculated to be 0.0555 and 0.0297, respectively. However, since these animals were substantially different in 55 in both size and RES organ weight, the corrected phagocytic indexes of 0.558 and 0.299, respectively, were better measures of in vivo phagocytic activity by tissue-fixed macr0phages. These experiments were limited by three major assumptions: 1) that colloid was not entrapped by capil- laries, 2) that Opsonins were not necessary or limiting for 125I-BSA phagocytosis, and 5) that anesthesia and anoxia affected both control and galactosemic chicks equally. Although spleens of control and galactosemic chicks accumulated the same levels of 125I—BSA (1.85% of the dose), livers of these animals acquired 28.2% and 25.4% of the dose, respectively. These observations indicate that at 10 days of age chick RES relies upon macr0phages of liver rather than of spleen and that the action of galac- tose regarding the RES is predominantly upon cells of the former organ. Table I demonstrates, however, that spleen weight in control chicks is sensitive to injection of E. coli. This response is not a measure of RES function per se but is related to an influx or proliferation of im- munocompetent cells (11). Thus, inhibition of this response during galactosemia suggests that cellular immunity may also be depressed in vivo. Moreover, the present results indicate that experi- mentally induced galactosemia in chicks may be a good model system for studying the effects of galactose on human phagocytes. Chick leukocytes, like those from galactosemic infants, are deficient in galactose-l-phosphate 54 uridylytransferase (l2) and, like normal human leukocytes, are functionally impaired by elevated levels of galactose. The impaired activity of RES cells in galactosemic chicks, as well as the retarded RES organ develOpment, further suggest that these chicks, like galactosemic infants (15), are more susceptible to bacterial infection. On the other hand, human leukocytes are morphologically quite different from chick leukocytes (14), and human RES function depends to a greater extent upon the macrophages of liver and spleen (5). In addition, galactosemic infants commonly display enlarged livers and spleens, whereas these symp- toms were not observed in galactosemic chicks. 10. ll. 12. 13. 14. 55 References Rutter, W.J., Krichevsky, P., Scott, H.M., and Hansen, R.G. (1953) Poultry Sci. 32, 706—715 Phillips, H.J. (1973) In Tissue Culture Methods and Applications pp. 406—408, Academic Press, New York, N.Y. Fraker, P.J., and Speck, Jr., J.C. (1976) Anal. Biochem. in preparation Taplin, G.V., Johnson, D.E., Dore, E.K., and Kaplan, H.S. (1964) J. Nucl. Med. 2, 259—275 Saba, T.M. (1970) Arch. Intern. Med. 126, 1051—1052 Sbarra, A.J., and Karnovsky, M.L. (1959) J. Biol. Chem. 244, 1355-1562 Johnston, Jr., R.B., Keele, Jr., B.B., Misra, H.P., Webb, L.S., Lehmeyer, J.E., and Rajagopalan, K.V. (1975) In The Phagocytic Cell in Host Resistance pp. 61-75 (Bellanti, J.A., and Dayton, D.H., eds) Raven Press, New York, N.Y. Granett, S.E., Kozak, L.P., McIntyre, J.P., and Wells, W.W. (1972) J. Neurochem. £2, 1659—1670 Kozak, L.P., and Wells, W.W. (1971) J. NeuroChem. i2, 2217-2228 Halpern, B.N., Biozzi, G., Benacerraf, B., Stiffel, C., and Hillemand, B. (1956) C.R. Soc. Biol. 150, 1507—1511 Simonsen, M. (1962) In Progress in Allergy Vol. 6 pp. 549, S. Kargel, New York, N.Y. Medline, A., and Medline, N.M. (1972) Can. Med. Assoc. J. 107, 877-878 Donnell, G.N., Bergren, W.R., and Ng, W.G. (1967) Biochem. Med. l, 29-55 Sturkle, P.D. (1965) in Avian Physiology, 2nd. ed., pp. ll—l8, Cornell University Press, Ithaca, N.Y. 56 Table I. Response of RES to injection of E. coli.8 Diet Injection Spleen Weighti Liver Weight mg i SEM gm i SEM Control — 117 i 22.6 5.50 i 0.45 Control + 182 i 22.8 5.79 i 0.84 Galactose - 46.7 i 15.6 1.89 i 0.54 Galactose + 49.4 + 15.6 1.79 i 0.54 aSix chicks were placed on each diet at 4 days of age and sacrificed eight days later. Each injected chick re— ceived 107 viable E. coli. Galactose diet was 50% (w/w) galactose. I00 80 .l <1 2 > 60 (I D (I) 32 4o 20 0 Figure l. 57 —. “—fl-‘--—-R \ \. ‘\~\ _ \. ‘~\~ ‘\\ ‘\ \ \\ ‘\ \ \o _ \ . ‘\ \xik \. \“x \“x \. - '\ \. ‘0) P l 1 l l I O 60 I20 TIME (mm) Effect of galactose on bactericidal activity of normal (0) and galactosemic ( o ) chick leukocytes in vitro. E. coli were added at zero time to cell sus- pensions incubated with 5.0 mM glucose ) or 5.0 mM glucose and 50 mM galactose ( ------ ). 58 KI 0.0297 K8 0.0553 l l l l 2.0 - E ‘\ E 0 LS '- m 0 2 < a: < M] .J 0 LG *- 0 Figure 2. LC 2.0 3.0 4.0 TIME (min) Intravascular clearance of colloidal 125I-BSA in normal and galactosemic chicks. Normal chicks and alactosemic chicks are represented by (a? and ( o ), respectively. 59 SPLEEN 03”)- 01K)- (0 E 2 am- CE I— ' ‘ ' I: o LIVER “J 7.00- 3 500- 3.00- . , LOO" l 1 I I0 I5 20 AGE (days) Figure 5. Effect of galactose diet on spleen and liver development in chicks. Normal chicks and galactosemic chicks are represented by ( O ) and ( o ), respectively. CHAPTER III HEXOSE TRANSPORT IN HUMAN AND GUINEA PIG POLYMORPHONUCLEAR LEUKOCYTES Abstract The prOperties of 2—deoxyglucose and 5-O-methyl- glucose transport in human and guinea pig polymorphonuclear leukocytes (PMN) were investigated. Uptake of [G-BH] 2-deoxyglucose by guinea pig PMN proceeded in vitro with a 1 Km and Via of 1.8 mM and 0.67 nmoles min- per 106 cells, x respectively. This system was competitively inhibited by glucose and mannose but was not significantly affected by galactose, fructose or 5-0—methylglucose. Maximal uptake of 2—deoxyglucose occurred at 41°, and phosphorylation of the 6-position was necessary for its intracellular concen- tration. This process, while not altered by uncouplers of oxidative phosphorylation, was sensitive to inhibitors of glycolysis. Preincubation of cells with 2 mM iodoacetate for 50 min significantly reduced the uptake of 2-deoxy- glucose and the intracellular levels of adenosine-5'- triphosphate without decreasing cell viability. [Methyl-3H]5-0-methylglucose, on the other hand, was not phosphorylated after entry, and uptake was insensitive 6O 61 to the presence of either heterologous hexoses or meta— bolic inhibitors. These and kinetic results indicated that uptake of 5-0-methylglucose by guinea pig PMN occurred by simple diffusion, whereas uptake of 2-deoxyglucose occurred by facilitated diffusion with subsequent phosphorylation. Similar transport systems were found in PMN isolated from human peripheral blood. Introduction Phagocytosis and intracellular killing of microorgan— isms are two primary functions of neutrOphilic polymor- phonuclear leukocytes (PMN). These activities are of par- ticular importance to the protection of a host against infection. However, in certain disorders, such as diabetes mellitus (1,2) and galactosemia (5), the capacity of PMN to ingest and destroy bacteria is impaired. Previous studies indicate that this impairment directly results from ele- vated levels of plasma carbohydrate (1-5) and is not an effect of insulin or opsonin deficiencies (1). Since phagocytosis is sensitive to inhibitors of gly- colysis (4) and glucose transport (5), the deleterious ef- fect of galactose on PMN function could largely result from a competitive inhibition of glucose transport. To test this suggestion, we undertook a detailed study of sugar transport in human and guinea pig PMN. 62 Materials and Methods Materials. All radioisotopes were purchased from New England Nuclear Corp. with the exception of [l-l4C]fructose which was obtained from Amersham Searle Corp. All carbo- hydrates were of the Q—configuration and were purchased from Sigma Chemical Co., Mallinckrodt Chemical Works, and Nutritional Biochemical 00. Other reagents were primarily from Sigma with the following exceptions: iodoacetic acid, Matheson Coleman and Bell Manufacturing Chemists; potassium cyanide, Baker Chemical 00.; p-chloromercuribenzoate and cytochalasin B, Calbiochem; phlorizin, Nutritional Bio— chemical 00.; N—ethylmaleimide, Aldrich Chemical 00.; U-80 regular Iletin insulin, Eli Lilly and 00.; dextran 250, Pharmacia Fine Chemicals, Inc.; guinea pig serum and Hank's balanced salt solution, Grand Island Biological 00.; and polystyrene latex particles (1.1 u dia.), Dow Diagnos- tics. All glassware was siliconized with a 1% solution of Siliclad, Clay Adams. Cell preparations. PMN was isolated from adult male guinea pigs (Connaught Laboratories, Ltd.) fed a commercial diet and water with 0.04% L—ascorbic acid, ad libitum. Guinea pigs were injected i.p with 5.0 m1 of sterile 1% caseinate in saline, and exudates were removed 15 to 20 hours later by flushing peritoneal cavities with Hank's balanced salt solution. Exudate cells composed of greater than 95% PMN were pooled and centrifuged at 250 xg at room temperature 65 for 5 minutes. Cells were washed and resuspended in Krebs- Ringer phosphate solution without calcium, pH 7.4. Human PMN were prepared from 10 ml samples of venous blood taken from 15 healthy adult volunteers. Samples were pooled, supplemented with 1.2% dextran 250, and allowed to sediment for one hour at room temperature. Leukocyte-rich supernatants were removed and centrifuged as described above. To lyse remaining erythrocytes, pellets of leuko- cytes were twice suspended and incubated for one hour at 57° in 200 ml of 0.015 M Tris, pH 7.2, containing 0.75% NH4Cl. After the second incubation, cells were again centrifuged and were suspended in 5.0 ml of fresh auto- logous plasma. Cells in plasma were then applied to 20 ml of siliconized glass beads (0.5 mm dia.), and PMN were isolated as described by Rabinowitz (6). Human PMN of at least 90% purity as judged by visual count were washed and resuspended in Krebs-Ringer phosphate solution without calcium, pH 7.4. Cell viability was determined by exclu- sion of 0.04% Trypan blue (7). Oxidation of carbohydrate. Conversion of l4C-labelled car— bohydrate to 14CO2 by guinea pig PMN was performed in sealed 25 ml Erlenmeyer flasks as previously described (4). Cell 6 PMN) were incubated for 1 hour at 57° suspensions (7 x 10 in 1.0 ml aliquots of Krebs-Ringer phosphate solution con- taining 10% guinea pig serum and 1.0 mM labelled carbo- hydrate, 0.25 uCi. Reactions were stOpped by adding 0.2 ml 64 of 7.5N H to the cells, and evolved CO2 was collected 2304 on 1 x 5 cm folded strips of Whatman 5MM paper with 0.2 ml of 5.75N KOH in plastic centerwells. Labelled CO2 was absorbed from the start of each experiment and for 1 hour after acidifying the medium. Centerwells were transferred directly to scintillation vials with 10 m1 of Bray's solu- tion and counted. Oxidation of carbohydrate by phagocytosing cells was determined in the same manner except that a 20—fold (particle/PMN) excess of polystyrene latex particles was included in each flask. Purity of each l4C-labelled carbohydrate was determined to be greater than 99% by ascending paper chromatography with 7:5 (v/v) ethanol: 1M ammonium acetate. Chromatograms were subsequently monitored with a Packard model 7201 chromatogram scanner. Transport of carbohydrate. Uptake of 3H—labelled carbo- hydrate was determined by monitoring the incorporation of both tritium counts and [U-l4C]sucrose counts into pellets of cells. Sucrose, which was shown to be excluded from entering PMN, was employed in each assay to correct for trapped 3H-carbohydrate present in the extracellular space. Such corrections were made by multiplying [U—l40]sucrose counts in the pellets by the ratio of tritium to [U-14C] sucrose counts in the supernatants (8). Counts attributable to intracellular 3H-carbohydrate were then calculated by subtracting extracellular tritium counts from total tritium Ch O 7.9 65 counts in each pellet. Unless otherwise stated, reactions employed approxi- mately 4 x 106 PMN suspended in 0.5 ml of Krebs-Ringer phosphate solution, pH 7.4. Suspensions Were incubated at 57° for 50 min prior to adding 0.5 ml volumes Of buffer containing 0.5 p01 of [G—BH]2-deoxyglucose, 0.05 uCi of [U-l4C]sucrose, and various amounts of nonisotOpic 2- deoxyglucose or inhibitors. After addition of labelled carbohydrate, cells were further incubated for 5 min at the reaction temperature. Since uptake as well as loss Of labelled 2-deoxyglucose was found to be negligible at temperatures below 4°, reactions were stopped by placing tubes with cells on ice and centrifuging at 5000 xg at 0° for 5 min. Supernatants were immediately aspirated, and pellets were transferred directly into scintillation vials with 10 m1 Of Bray's solution. Pellets of cells and supernatant fractions were counted with a Beckman CPM 100 liquid scintillation counter set for double label counting. Uptake Of [methyl—3H]5—0—methylglucose was determined by the same procedure. Quantities of PMN in each assay were determined by counting cells in a hemocytometer. Cell counts were per- formed at least in quadruplicate with a Universal model Zeiss microsc0pe using phase Optics. Intracellular levels of phosphorylated 2-deoxyglucose were measured after the uptake experiments were performed. Scintillation fluid from each vial was dried, and 66 carbohydrate was removed from each napthalene residue by extracting with 20 m1 of distilled water. Extracts were concentrated under a stream of nitrogen and were spotted on Whatman 5MM paper for ascending chromatography. After development for 15 hours with 7:5 (v/v) ethanol: 1M am- monium acetate, Chromatograms were cut into segments cor- responding to fast and slow migrating bands of radioac- tivity. Phosphorylated 2-deoxyglucose associated with the latter segments and free 2-deoxyglucose plus sucrose present in the former segments were eluted with buffer into separate scintillation vials, dried, and counted as described above. In some experiments, phosphorylated 2- deoxyglucose was precipitated by Somogyi's method (9), and levels were determined by calculating the difference between counts attributable to total and free 2—deoxy— glucose. Results of both procedures were comparable. Levels of ATP. PMN (15.2 x 106 cells/assay) were incubated either with or without inhibitors at 57° in 1.0 ml aliquots of Krebs-Ringer phosphate solution, pH 7.4. After 50 min Of incubation, cells were centrifuged for l min at 5700 xg. Pellets were immediately frozen by immersing tubes in liquid nitrogen, and cells were stored at -80° until homo- genizing with 0.21 m1 of 5N perchloric acid. Homogenates were centrifuged at 5000 xg for 10 min to remove precipi— tated protein, and supernates were neutralized with 0.18 ml of 2N KOH: 0.4M imidazole: 0.4M K01. Levels of ATP were 67 determined in each supernate by monitoring the reduction of NADP at 540 nm in the presence of excess glucose-6-phosphate dehydrogenase (EC 1.1.1.49) and hexokinase (EC 2.7.1.1) (10). Results Oxidation of carbohydrates to 14CO . Table I shows the 2 1400 levels Of produced by guinea pig PMN during a 1 hour 2 incubation at 57° with various l4C-labelled carbohydrates. Values obtained with [1-14C]glucose were higher than those Obtained with other carbohydrates, and participation of the hexose monOphosphate shunt in this process is indicated by comparing values for the oxidation of [1-14C]glucose to those Of [6-14C]glucose. Addition of latex particles, which stimulates hexose monOphosphate shunt activity (11), en- hanced 14CO 2 production by the following extents: with glucose, 2.6 fold; mannose, 2.6 fold; galactose, 2.8 fold; and fructose, 1.5 fold. Increases in 14C0 evolution during 2 phagocytosis were not Observed when l4C—labelled 2-deoxy— glucose, 5-O-methylglucose, or sucrose were employed. The latter carbohydrates were not readily converted to 14CO2 by PMN, and the values Obtained in each of these cases could be attributed to traces (less than 1%) of labelled contamina- tion as judged by paper chromatography. Exclusion of sucrose from PMN. That [U—l4C]sucrose was not taken up by PMN is illustrated in Table II. Radioactivity associated with pellets of cells did not change significantly 68 over the course of a two hour incubation. Similar values for pellet-associated radioactivity were also found when cells were preincubated for 10 min with 5.0 mM nonisotOpic sucrose before the addition of equal amounts of labelled sucrose. This eliminated the possibility that sucrose rapidly entered PMN before the initial measurement and sup- ported Our contention that pellet-associated radioactivity indicated the amount Of trapped extracellular medium. Time course for 2-deoxyglucose uptake. Data in Figure 1 indicate the time course for uptake of 5.0 mM [G-3HJ2— deoxyglucose into guinea pig PMN. Uptake was essentially linear with time for the first ten minutes, after which a maximal value Of 4.8 nmoles/106 cells was reached. In order to determine whether this value represented a concen- tration Of sugar in either its free or phosphorylated form, intracellular levels of 2-deoxyglucose were calculated upon dividing uptake by the intracellular water space of 0.42 ul/ 106 PMN (12). An approximate 2-fOld concentration Of sugar then became apparent (Figure 1, right ordinate). Intra- cellular 2-deoxyglucose was 69.1—84.9% phosphorylated. Uptake of 2-deoxyglucose versus cell concentration. Before data from separate experiments could be pooled, we determined that uptake was linear with cell concentration and that "crowding effects," as discussed later, were not present. ‘FLgure 2 indicates the former relationship for cell concen- 6 ‘tmrtions up to 5.6 x 10 PMN/ml when 0.2 mM [G-BHJ2- 69 deoxyglucose was employed. Subsequent experiments using this and other levels of 2—deoxyglucose showed linearity up to approximately 8 x 106 PMN/ml. Kinetics Of uptake. Values illustrated in Figure 5 were determined by stopping cell incubations after 5 min and measuring the penetration of labelled 2—deoxyglucose or 5-0- methylglucose under conditions Of varying substrate concen- trations. Initial velocities for uptake were therefore not directly measured but were calculated by dividing the aver- age nmoles of sugar/106 cells by the 5 min time interval. Unlike the uptake of 5—O—methfiglucose, penetration Of 2— deoxyglucose clearly followed saturation type kinetics and yielded a Km and Vma of 1.8 mM and 0.67 nmoles min—l/lO6 x cells, respectively, when analyzed by the method of Lineweaver and Burk (15). Data obtained with 5-0-methyl- glucose gave an apparent Km and Vmax of 68 mM and 21 nmoles min-l/lO6 cells, respectively. However, the latter values could not be verified using a Hofstee type analysis (14). Effects of heterologous carbohydrates on uptake. When PMN suspensions were incubated with either 4.0 mM glucose or mannose together with 0.2 mM [G-BHJ2—deoxyglucose, (Table III), the uptake of labelled sugar was significantly im— paired. However, inhibition by similar levels Of galactose and fructose was not observed. Addition of 5-0-methyl- glucose or insulin had no significant effects on 2- 70 deoxyglucose uptake, and the presence of insulin did not increase the inhibition by glucose. In similar experiments where 5-O-methylglucose uptake was monitored, no inhibitors of uptake were found, and insulin was also without effect (data not shown). Experiments were then conducted using various 2- deoxyglucose levels at fixed concentrations of glucose or mannose (Figures 4a and 4b, respectively). Both glucose and mannose appeared to be competitive inhibitors of 2-deoxy- glucose uptake; the former giving a Ki of 2.67 i 0.52 mM and the latter of 2.28 i 0.55 mM. All lines in Figure 4 were fit to the data by the method of least squares using a weighted computor program as previously described (15). Effects of temperature and metabolic inhibitors. As seen in Figure 5, the velocity Of 0.2 mM [G-BH]2-deoxyglucose penetration rose significantly with increasing temperature until 41°. Beyond this point velocity decreased and was particularly sensitive to temperatures between 48° and 55°. In contrast, the rate Of 0.1 mM [methyl-3H]5-O—methyl- glucose uptake did not change significantly during similar experiments. The dependence of 2-deoxyglucose uptake upon tempera- ture suggested that this process relied, in part, upon metabolic activity. To test this suggestion, PMN were preincubated for 50 min at 57° with various metabolic inhibitors, and the rates of [G-BH]2-deoxyglucose uptake, 71 as well as phosphorylation, were measured (Table IV). In the presence Of 2.0 mM flfloacetate rates Of both uptake and phosphorylation were significantly reduced to 55% and 11% Of the control values, respectively. Preincubation of cells with either 40 mM sodium fluoride or 0.2 mMIimkEcetate re- sulted in similar degrees of inhibition in both uptake and phosphorylation. However, uncouplers of oxidative phos- phorylation, such as antimycin A, potassium cyanide, and dinitrophenol, did not show significant effects. Phlorizin, N-ethylmaleimide, and p-chloromercuribenzoate produced slight reductions Of 2-deoxyglucose uptake, but these were only significant when the former two agents were employed. In subsequent experiments, cytochalasin B at 0.5 pg/ml and 1.0 ug/ml also inhibited uptake by 30.3% and 35.7%, respec- tively. These differences were significant at the P<:0.0l level. Uptake of [methyl-3H]5-0-methylglucose was not af- fected by any of the above metabolic inhibitors (data not shown). A number of other agents were preincubated with cells for 50 min but were without significant effects on [G-BH] 2-deoxyglucose uptake. These included 10 mM sodium barbital, 5 mM theophylline, 5 mM caffeine, 4 mM myO-inositol, 5 mM colchicine, 1 mM CAMP, 1 mM dibutryl CAMP, 1 mM cGMP, and 1 mM dibutryl cGMP. Preincubation with polystyrene latex particles (200:1 particles: cell) and 0.5 mU of insulin, resulted in higher (127% and 145% Of control, respectively) but not significantly different, uptake rates. Effects of 72 insulin were not altered by substituting Krebs—Henseleit bi- carbonate solution for Krebs—Ringer phosphate solution, nor did this substitution affect [methyl—3H]3—0-methy1g1ucose uptake. However, rates for [G-BHJ2-deoxyglucose uptake were 50% lower when the former medium was employed. Dialyzed 10% guinea pig serum did not affect the uptake of either glucose analog. ATP levels. Intracellular levels of ATP in guinea pig PMN were significantly lowered by the presence of 2.0 mM iodo— acetate (Table V). Conditions during the incubation of cells were identical with those employed during the inhibitor ex- periments (Table IV). Effects of heterologous carbohydrates on uptake by human EMN. The uptake Of 0.2 mM [G-BH]2—deoxyglucose by human PMN was approximately 2 fold greater than the uptake by guinea pig PMN (0.227 vs 0.100 nmoles min-l/IO6 cells). Despite this difference, inhibitors of uptake for human PMN (Table VI) were very similar with those for guinea pig PMN (Table III). Galactose, fructose, and 5-0-methylglucose did not significantly impair 2-deoxyglucose uptake when present at a 20:1 inhibitor to substrate ratio, whereas glucose and mannose resulted in significant inhibition. Mannose in this case was a slightly better inhibitor than glucose, and some inhibition did occur in the presence Of 50 mM galactose, but the latter was not significant. 75 Discussion Although the conversion of [1-140] glucose to 14CO 2by PMN has been frequently reported (16,17), studies on the oxidation Of other carbohydrates to 002 have not been previously made. Our Observations (Table I) demonstrate that [1-140] mannose, [1—140] galactose, and [1—140] fruc- l4 tose are also metabolized to CO2 and that this oxidative metabolism is stimulated by phagocytosis. The degree Of 14 stimulation of [l- 0] glucose oxidation is comparable to values previously reported (16), and the rate Of oxidation by phagocytosing cells agrees with the 20 nmoles hr-l/lO6 cells Observed by Stjernholm §£_gl.(l8). That a constant degree of stimulation (2.6-2.8 fold) occurred, when either labelled glucose, mannose or galactose (but not fructose) were employed, strongly suggests that the former.three hexoses are converted into a common intermediate, i,e., glucose-6-phosphate, prior to oxidation by the hexose monophosphate shunt. Support for this suggestion is three- fold: 1) PMN contain a complete Leloir pathway and are Capable of converting galactose to glucose-6-phosphate (5,19), 2) l4C—labelled mannose is converted to lactate, Via mannose—6—phosphate and fructose-6-phosphate, with the same labelling pattern as lactate derived from l4C-labelled glucose (20), and 5) PMN are not gluconeogenic and are not able to convert fructose-l—phosphate or fructose-l, 6- bisphosphate to glucose—6-phosphate (21). 74 The inability of guinea pig PMN to transport or oxi- dize l4 C-sucrose (Table II) is in agreement with results on human PMN described by Englhardt and Metz (l2) and with Ob- servations on rabbit alveolar macrOphages described by Gee §£_§l. (5). Both of these studies similarly employed the exclusion of sucrose to correct for extracellular space during hexose transport experiments. The use Of this tech- nique also corrects for a non—specific type of uptake en- countered during phagocytosis. Esman refers to this as "piggy-back" phagocytosis or the "concomitant engulfment of extracellular medium with the phagocytosis of particles" (22). Such non-specific uptake increases the insulin space in pellets Of leukocytes by 4% to 6% (25) and, thus, could result in elevated values for 2-deoxyglucose uptake if l4C-sucrose was not also employed. When 5 mM [G-BH]2-deoxyglucose is presented to guinea pig PMN, maximal intracellular levels of label approach 11.4 mM after 40 min of incubation (Figure 1). Since 69% to 85% of this label is phosphorylated, the maximal intra- cellular levels of free 2-deoxyglucose can be calculated to range from 1.7 mM to 5.5 mM. These values suggest that entry of 2—deoxyglucose occurs in the free form and that phosphorylation is necessary for 2-deoxyglucose concentra- tion. These results are in accord with those of Esman (22), who found free intracellular glucose in PMN at low external glucose concentrations, and also with those of Luzzatto and Leoncini (24), who concluded that the rate of carbohydrate 75 utilization was slower than the rate Of carbohydrate entry. While data in Figure l eliminate an active transport system for 2-deoxyglucose uptake, those from Figure 5 and Figure 4 rule out simple diffusion and indicate a facili- tated transport mechanism. Uptake of 2—deoxyglucose dis- played saturation-type kinetics and was competitively in- hibited by glucose and mannose which is suggestive of a common carrier mediated transport system. Similar inhibi- tion of 2—deoxyglucose uptake by glucose occurs in rabbit alveolar macrOphages (5), and inhibition by glucose and mannose, but not galactose and fructose, is reported for rat diaphram muscle (25). Studies on guinea pig lymph node cells by Helmreich and Eisen (26) also imply a common car- rier mechanism for hexose transport; however, this mechanism appears to be more selective in PMN, since our results did not show competitive inhibition by fructose or 5-0-methyl— glucose. Further evidence for facilitated diffusion of 2-deoxy- glucose arises from the temperature dependence of [G-3H] 2-deoxyglucose uptake (Figure 5) and from the effects of metabolic inhibitors upon this process (Table IV). Since hexose phosphorylation is required for sequestering Of label, inhibition of phosphorylation by depleting intracellular ATP would be expected to lower the diffusion gradient for free 2-deoxyglucose and, thus, impair the rate of 2—deoxyglucose entry. We Observed significantly lower rates of entry when guinea pig PMN were incubated at temperatures below 57°, 76 when an isotonic bicarbonate solution was substituted for Krebs-Ringer phosphate solution, and when cells were pre- incubated with 40 mM sodium fluoride or 2.0 mM iodoacetate. Pre—incubation with the latter inhibitor not only impaired uptake and phosphorylation to the greatest extents but also significantly decreased the intracellular levels of ATP (Table V). These results, together with the lack of sig— nificant inhibition by potassium cyanide, antimycin A, and dinitrOphenol, further demonstrate that active glycolysis, and not oxidative phosphorylation, is the primary source of ATP in PMN. This agrees with previous Observations that PMN contain few mitochondria and that inhibitors of gly— colysis can depress phagocytic function (4). The lack of an in vitro insulin effect on either 2— deoxyglucose uptake or an inhibition of uptake by glucose (Table III) is consistent with our findings that the rate of carbohydrate entry was always greater than the rate Of phos- phorylation. However, this also implies that phosphoryla- tion of 2-deoxyglucose is not affected by the presence of insulin. Similar results were Obtained by Beck (27), who could not find insulin sensitivity in leukocyte hexokinase, by Englhardt and Metz (12), who did not observe an insulin effect upon glucose transport into human PMN, and by Helmreich and Eisen (26), who were unable to demonstrate an insulin effect upon glucose uptake into guinea pig lymph node cells. On the other hand, Luzzatto (28) observed an increase in xylose penetration into leukocytes at a 77 non-physiological level of insulin (1 U/ml), and Klant and Schucher (29) found an increased disappearance of glucose when leukocytes were incubated with 0.5 U/ml of insulin. Levels Of insulin employed in our studies (0.5 mU/ml) were slightly higher than physiological and did not affect either 2—deoxyglucose or 5-O-methyglucose uptake. The uptake Of [methyl-3H]5-0-methylglucose in our ex- periments was remarkably different from the uptake of [G-BH] 2—deoxygluc0se in three respects: 1) 5-0—methylglucose was not concentrated nor phosphorylated after entry, 2) uptake was insensitive to the presence of heterologous carbohydrate, metabolic inhibitors, and to varying temperature, and 5) up- take failed to follow saturation-type kinetics. Whereas kinetic parameters were determined for this glucose analog using a Lineweaver—Burk type analysis (Figure 5b), these values were substantially different from those for 2— deoxyglucose and could not be verified by the method of Hofstee (14). Since both Km and Vmax for 5-0-methylglucose uptake tend towards infinity, it is apparent that transport of this analog in PMN occurs by simple rather than facili- tated diffusion. Although decreases in cell metabolism, as well as glucose uptake, have been noted in concentrated PMN suspen— sions, we did not encounter this during our experiments. This so—called "crowding effect" has been attributed to changes in extracellular pH as a result of lactate accumula— tion and to the subsequent inhibition Of phosphofructokinase 78 activity (22). Although Englhardt and Metz (12) Observed this in their experiments with 7.6 x 107 PMN/ml during a 1 hour incubation, our experiments avoided this effect by utilizing fewer cells and shorter (5 min)-incubation times. Therefore, lactic acid was not allowed to build up, and linearity between 2-de0xyglucose uptake and cell concentra- tion was established (Figure 2). This method also circum- vented any problems associated with excessive 2—deoxyglu— cose-6-phosphate accumulation which could non-competitively inhibit the hexokinase reaction (24). Similarities between the uptake of 2-deoxyglucose by guinea pig PMN and the uptake by human PMN isolated from peripheral blood are apparent from comparing values in Table III to those in Table VI. Although the rate of uptake with human cells was approximately 2 fold greater than the rate with guinea pig cells at 57°, uptake in both cases was substantially inhibited by the presence Of glucose or man- nose. Galactose, fructose, and 5—0-methylglucose at 4 mM did not significantly impair 0.2 mM 2—deoxyglucose entry into either cell type. However, uptake was slightly de- creased when human PMN were incubated with 50 mM galactose. This latter effect was not Observed with guinea pig PMN under similar conditions but was found, together with lower ATP levels (Table V), after a 50 min pre—incubation with elevated galactose. It therefore appears that galactose may impair 2—deoxyglucose uptake by blocking its phosphory- lation rather than by competing with its uptake. Although 79 competition in this case with human PMN cannot be eliminated per se, it seems unlikely that elevated levels of galactose, as encountered during galactosemia, could impair phagocyte function by competing with glucose uptake. Thus, impaired phagocytic activity in the presence of galactose may be primarily attributed to other factors, such as i) the pres- ence of a futile adenosine triphosphatase cycle, or ii) the intracellular accumulation of galactose or one of its meta- bolites with subsequent inhibition of glycolysis (5). 10. 11. 12. 130 14. 15. 16. 80 References Drachman, R.H., Root, R.K., and Wood, W.B. (1966) J. Exp. Med. 124, 227-240 Sanchez, A., Reeser, J.L., Law, H.S., Yahiku, P.Y., Willard, R.E., McMillian, P.J., Cho, S.Y., Magie, A.R., Register, V.D. (1973) Am. J. Clin. Nutr. 2Q, 1180- 1184. Litchfield, W.J., and Wells, W.W. (1976) Infect. Immun. 15, 728-754 Sbarra, A.J., and Karnovsky, M.L. (1959) J. Biol. Chem. 244, 1555-1562 Gee, J.B.L., Khandwala, A.S., and Bell, R.W. (1974) J. Reticuloendothel. Soc. 15, 594-405 Rabinowitz, Y. (1964) Blood 22, 811-828 Phillips, H.J. (1975) In Tissue Culture Methods and Applications pp. 406-408, Academic Press, New York Khandwala, A.S., and Gee, J.B.L. (1974) Biochem. Pharmacol. 2 , 1781-1786 Somogyi, M. (1945) J. Biol. Chem. 160, 69—73 Lowry, 0.H., and Passonneau, J.V. (1972) In A Flexible System of Enzymatic Analysis pp. 151-152, Academic Press, New York Baehner, R.L. (1975) In The Phagocytic Cell in Host Resistance (Bellanti, J.A., and Dayton, D.H. eds.) pp. 175-195, Raven Press, New York Englhardt, A., and Metz, T. (1971) Diabetologica 1, 145-151 Lineweaver, H., and Burk, D. (1954) J. Am. Chem. Soc. 6, 658-666 Hofstee, B.H.J. (1960) Science 1 1, 59 Dunne, C.P., Gerlt, J.A., Rabinowitz, K.W., and Wood, W.A. (1973) J. Biol. Chem. 248, 8189-8199 Karnovsky, M.L. (1974) In Progress in Immunology II (Brent, L., and Holborrow, J., eds.) Vol. 4, PP. 85-95, North-Holland Publishing Co., New York 17. 18. 19. 20. 21. 22. 25. 24. 25. 26. 27. 28. 29. 81 Karnovsky, M.L. (1962) Physiol. Rev. 42, 145-168 Stjernholm, R.L., Burns, C.P., and Hohnadel, J.H. (1972) Enzyme l , 7-51 Klant, N., and Schucher, R. (1965) Can. J. Biochem. Physiol. 44, 849-858 ‘ Esman, V., Noble, E.P., and Stjernholm, R.L. (1965) Acta Chemica Scandinavica 42, 1672-1676 Noble, E.P., Stjernholm, R.L., and Ljungdahl, L. (1961) Biochim. Biophys. Acta 42, 595-595 Esman, V. (1972) Enzyme l , 52—55 Berger, R.R., and Karnovsky, M.L. (1966) Fed. Proc. 22, 840-845 Luzzatto, L., and Leoncini, G. (1961) J. Biochem. 42, 249-257 Kipnis, D.M., and Cori, C.F. (1959) J. Biol. Chem. 234, 1958-1965 Kelmreich, E., and Eisen, H.N. (1959) J. Biol. Chem. 234, 1958-1965 Beck, W.S. (1958) J. Biol. Chem. 2 2, 251-270 'Luzzatto, L. (1960) Biochem. Biophys. Res. Commun. 2, 402-406 Klant, N., and Schucher, R. (1962) Can. J. Biochem. Physiol. 42, 899-905 82 Table 1. Conversion of l4C-labelled carbohydrate to 14CO2 by guinea pig PMNa l4CO2 produced Carbohydrate Resting Phagocytosing P nmoles/hour/lO6 cells 1 S.D. [l-l4CJGlucose 7.16 i 1.57 18.5 i 0.4 0.01 [l-l4C]Mannose 2.06 i 0.16 5.42 i 0.55 0.01 [1-14C]Ga1actose 0.60 i 0.14 1.70 i 0.10 0.01 [1—1401Bructose 0.26 i 0.02 0.34 i 0.01 0.02 [6-14C]Glucose 0.14 i 0.02 [U—l4C]2-deoxyglucose 0.11 i 0.02 0.11 i 0.01 ns [U-l4C]5-O-methy1- glucose 0.01 0.01 ns [U-l4C]sucrose 0.05 0.04 0 ns aCells (7 x 106 PMN) were incubated for 1 hour at 57° in 1.0 m1 aliquots Of buffer containing 10% guinea pig serum and 1.0 mM labelled carbohydrate, 0.25 uCi. CO2 was collected from the start of each experiment on strips Of Whatman 5mm paper saturated with KOH and for 1 hour after acidifying medium. Each experiment was performed in triplicate. S.D. refers to standard deviation. Significance levels, P, are given for comparison of values from resting cells to values from cells incubated with polystyrene latex particles; ns indicates no significance. 85 Table II. Exclusion of [U-l4C]Sucrose from guinea pig PMNa Incubation time [U-l4C]Sucrose in pellet P counts/min i S. D. O 914 i 114 2 879 i 84 ns 50 854 i 84 ns 60 1002 i 251 ns 120 914 i 181 ns aPMN (1.18 x 106 cells/assay) were incubated at 57° with buffer containing 10% autologous serum and 1.0 mM [U-l40] sucrose, 0.25 uCi. After each time point tubes were placed in ice and immediately centrifuged for 5 min at 5000 xg at 4°. Pellets were rinsed with cold buffer and suspended in scintillation fluid. Each time point was performed in triplicate: ns indicates no significance between values at a given time and values at 0 time. 84 Table III. Effect of carbohydrate and insulin on uptake [G-SHJ2-deoxyglucose by guinea pig PMNa Additions % Uptake P None 100 i 15.1 Insulin 95.7 i 4.9 ns Glucose 42.5 i 10.1 0.001 Glucose + insulin 44.5 i 5.6 0.001C Mannose 46.9 i 4.2 0.001 Galactose 95.8 i 7.5 ns Fructose 95.8 i 6.8 ns 5—O-methylglucose 87.0 i 7.1 ns 3PMN were incubated at 37° in 1.0 ml of buffer. All addi- tions were made at 0 time. whereas levels of carbohydrate and insulin were 4.0 mM and P indicates levels of signifi— cance between values with additions and values without ad— 0.5 mU/ml, respectively. ditions: ns indicates no significance. bUptake min/106 cells: 2-deoxyglucose was 0.2 mM, rate without additions was 0.142 i 0.019 nmoles/ CNO significance between values obtained with glucose and values Obtained with both glucose and insulin. 85 Table IV. Effect of metabolic inhibitors on rates of [G-BH] 2—deoxyglucose uptake and phosphorylation by guinea pig PMNa Additions Uptake Phosphorylation nmoles min-l/IO6 PMN None 0.100 Iodoacetate 0.2 mM 0.057C Iodoacetate 2.0 mM 0.033b Fluoride 40 mM 0.064C Antimycin A 1.0 ug/ml 0.086n8 Cyanide 6.0 mM 0.086ns Dinitrophenol 1.0 mM 0.096nS Phlorizin 10.0 mM 0.069e N-ethylmaleimide 0.2 mM 0.070d Chloromercuribenzoate 0.2 mM 0.081nS 0 0. 0 0 .064 018 .007 .017 .042 .058 aPMN (2.81 x 106 cells/assay) were preincubated with or without additions for 50 min at 57°. [G-3H]2-deoxyglucose was added to a final concentration Of 0.2 mM, and uptake rates were measured. performed 5 times, and levels Of significance between values with additions and values without additions are indicated by: b) 0.005, c) 0.01, d) 0.02, e) 0.05, and ns (no significance). After this period Each assay was 86 Table V. Effect of iodoacetate on ATP levels in guinea pig PMNa Additions ATP levels P nmoles/106 cells i S.D. None 1.52 i 0.46 Iodoacetate 2.0 mM 0.41 i 0.55 0.05 Galactose 50.0 mM 0.78 i 0.27 ns aPMN (15.2 x 106 cells) were incubated with or without additions for 50 min at 57°. After incubation cells were immediately centrifuged at 57° for l min at 5700 xg. Pellets were quickly forzen by immersing tubes in liquid nitrogen. Cells were stored at -80° until assay; see "Experimental Procedures" for assay conditions. P refers to levels of significance between values with additions and values without additions; ns indicates no signifi- cance o 87 Table VI. Effect of carbohydrate on uptake Of [G-BH]2- deoxyglucose by human PMNa Additions % Uptake i S.D.. P None 100 i 20.1 Mannose 4.0 mM 59.1 i 7.8 0.02 Glucose 4.0 mM 50.8 i 4.6 0.02 Fructose 4.0 mM 78.2 i 9.9 ns Galactose 4.0 mM 78.2 i 6.6 ns Galactose 50.0 mM 64.5 i 11.9 0.05 5-0-methylglucose 4.0 mM 98.8 i 8.8 ns aPMN (0.848 x 106 cells/ml) were incubated in 1.0 ml buffer at 57°. All additions were made at 0 time. [G-BH]2-deoxy- glucose was 0.2 mM, and uptake rate without additions was 0.227 i 0.046 nmoles/min/lo6 cells: N25. P refers to levels of significance between values with additions and values without additions; ns indicates no significance. 88 UPTAKE (nmoles/Ioscells) (613' I I5 206 (mM) PO 1 J. C) INTRACELLULAR l l I l 1 l O 20 40 60 TIME (min.) Figure l. Time-course for uptake of 5.0 mM 2-deoxyglu- cose by guinea pig PMN. Assays were performed in quadruplicate using 7 x 106 PMN/experiment at 57°C. Line bars indicate S.D. Intracellular concentrations (right ordinate) are derived from uptake values (left ordinate), see text. 89 2.5 A 2.0 U) 2 O E 5 LS LL! 1‘: F— 0_ LO 3 0.5 0 Figure 2. l l l I LO 2.0 3.0 4.0 IO6 cells Uptake Of 2-deoxyglucose versus cell con- centration. Uptake was measured 5 min after adding 0.2 mM labelled 2-deoxyglucose. Line bars indicate S.D. of three experiments. 90 VELOCITY (nmoles / min. / IO6 cells) 1 l l ' Figure 5. 0.5 LO LS 2.0 [Substrate] mM l/[Substra'o] mM.I Dependence of 2-deoxyglucose uptake (0) and 5-0-methylglucose uptake ( O ) on levels of external homologous carbohydrate. Velocity is plotted (A) against external substrate concentration, and these values are replotted (B) according to Lineweaver and Burk. Each point is a mean Of values from 5 experiments that were performed in triplicate. Line bars indicate S.D. I / VELOCITY 91 l2-» I BLD' . 41)“ .////”” __‘¢E‘ZZ;;”I 1 1 1 I t2- I - u 81)” 4.0 - . / Figure 4. 0 (15 L0 L5 2&) I / [Substrate] mM-l Inhibition of 2-deoxyglucose transport by glu- cose and mannose. Lineweaver-Burk plots for inhibition by glucose (A) and mannose (B) were generated from data ob- tained in 4 and 5 experiments, respectively; each of which were performed in triplicate. Levels of inhibitor are (0) none, (0) 0.25 mM, (A) 0.49 mM, (:1) 0.98 mM, and (I) 1.96 mM. 92 0." 0.l0 ' 0.00 ' VELOCITY (nmoles lmin. / :06 cans) 0.02 " H PP N l 1 l 20 30 40 50 TEMPERATURE (‘0) Figure 5. Temperature dependence of 0.2 mM 2—deoxyglucose uptake (0 ) and 0.1 mM 5—0-methylglucose uptake (0). Line bars indicate S.D. from 5 experiments. CHAPTER IV EFFECT OF CARBOHYDRATE ON THE OXYGEN—DEPENDENT KILLING MECHANISM OF POLYMORPHONUCLEAR LEUKOCYTES Abstract To account for the impaired bactericidal activity of polymorphonuclear leukocytes (PMN) during galactosemia and hyperglycemia, a number Of biochemical parameters associ- ated with the oxygen-dependent killing mechanism of guinea pig PMN were investigated. These included the formation of superoxide anion, hydrogen peroxide and hydroxyl radical, as well as the consumption of oxygen and chemiluminescence. Incubation Of PMN in medium containing 50 mM galactose substantially impaired the extracellular reduction of cyto— chrome g. This action was prevented by the addition of catalase. However, this impairment could not be adequately explained on the basis of either decreased superoxide anion production or increased hydrogen peroxide generation. Fur— ther experiments demonstrated 1) that hexoses and polyols, which are accumulated by PMN, effectively competed with methional for reaction with hydroxyl radical, and 2) that hydroxyl radical oxidized hexose to hexonic acid. These 95 94 results indicated the participation of carbohydrate radicals and suggested that such radicals were formed by PMN. The formation of carbohydrate radicals by PMN could account for oxidation Of extracellular cytochrome g and could be dele- terious to normal bactericidal activity. Introduction Phagocytosis and microbicidal activity are two distinct functions of polymorphonuclear leukocytes (PMN). The former is an endergonic process relying upon energy supplied from glycolysis (1), whereas the latter is mainly oxidative in character depending upon the production of reduced oxygen intermediates. Bactericidal activity is impaired under anaerobic conditions (2,5) and can be partially inhibited by incubating leukocytes with catalase (4), superoxide dismutase (4,5) or benzoate (4). Hydrogen peroxide (6,7,8) and super- oxide anion (9,10) are formed by PMN, and production of hydroxyl radical (4,11), as well as singlet oxygen (4,12, 15), has been proposed. Considerable evidence suggests that the generation of these agents is linked to the univalent reduction of oxygen by a cyanide-insensitive NADH oxidase (14,15) and that this system is related to the activity of the hexose monophosphate shunt (16,17). In this regard, absence Of extracellular glucose has been shown to decrease formate oxidation, and thus hydrogen peroxide production, in PMN (8). However, effects of elevated levels of hexose upon the formation of hydrogen peroxide and other reduced 95 forms of oxygen have not been previously determined. These effects could account for the impaired bactericidal activ- ity of PMN during galactosemia and hyperglycemia and are described herein. Materials and Methods Animals and materials. Adult male guinea pigs were Obtained from Elm Hill Laboratories, Cambridge, Mass., and fed a com- mercial diet and water with 0.04% ascorbic acid, ad libitum. Xanthine, scopoletin, phenazine methosulfate, nitroblue tetrazolium, methional, heparin (grade I), NADH (grade III), xanthine oxidase (EC 1.2.5.2), horse radish peroxidase (EC 1.11.1.7), and cytochrome g (type III) were purchased from Sigma Chemical CO. Hydrogen peroxide, catalase (EC 1.11.1.6), superoxide dismutase (EC 1.15.1.1), and potassium cyanide were purchased from Mallinckrodt Chemi- cal Works, Worthington Biochemical Co., Miles Laboratories, Ltd., and J.T. Baker Chemical 00., respectively. All car- bohydrates and polyols were purchased from Sigma, Mallinc— krodt, or Nutritional Biochemical Co. Guinea pig serum and l4C-labelled formate were purchased from Grand Island Biological CO. and New England Nuclear Corp., respectively. Dichlorodiacetylfluorescein and triethylene diamine were products of Eastman Organic Chemicals. Xanthine oxidase and catalase were routinely passed through 1 x 50 cm columns of Sephadex G-10 to remove pre- servatives, and gel filtration was similarly used to free 96 ferrocytochrome g from excess ascorbate. Stock solutions of hydrogen peroxide were prepared immediately before use by diluting 50% H202 2000 fold with distilled water. Con- centrations were then determined by recording the 250 nm absorbance and employing the molar extinction coefficient Of 81 M.1 cm"1 (6). Cell preparations. Guinea pig PMN and macrOphages were isolated from casein injected peritonea as described in Chapter III and Chapter 1, respectively. Concentrations of cells in suspension were determined using a hemocytometer under phase optics. Each cell count was performed in quad- ruplicate. Extracellular reduction of cytochrome c. Reactions in- volving the reduction Of ferricytochrome g by superoxide anions were investigated by using a method similar to that reported by Babior §§_§;. (9). Both resting and phagocy- tosing PMN (0.88 x 107 to 1.06 x 107/assay) were pre- incubated at 57°C for 2 hr in 1.0 ml of KRPS containing 10% guinea pig serum. Cell suspensions were then diluted with one volume of ferricytochrome 9 solution (152 uM) and divided into two 1.0 ml aliquots. One aliquot was placed on ice to serve as a reaction blank, while the other was incubated for an additional hour at 57°C. Cells along with polystyrene latex particles were removed by centrifuging each aliquot at 5600 xg for 5 min, and levels of reduced cytochrome g were determined in each cell supernate by 97 monitoring the absorbance at 550 nm. Various levels of carbohydrate were added prior to the pre-incubation period, while other agents including superoxide dismutase (0.2 mg/ ml) and catalase (0.5 mg/ml) were added along with cytochrome g. All assays were performed in quadruplicate on cells pooled from at least two guinea pigs. Phagocytosing cells received a 50 fold (particle to cell) excess Of polystyrene latex particles. Effects of carbohydrate upon the reduction of cyto- chrome g by superoxide anion were also studied in separate experiments without cells. Superoxide was generated enzy- matically in 1.1 ml of 0.05 M sodium phosphate buffer, pH 7.8, containing: 50 pM xanthine, 100 M EDTA, 10 pM ferricytochrome g, 27 mM carbohydrate, and 5.2 mU Of xanthine oxidase. Reactions were performed 6 to 8 times, and rates of cytochrome 9 reduction at 550 nm were deter- mined using a Gilford model 5500 spectrophotometer in the general kinetic—l program mode. Cellular reduction of nitroblue tetrazolium. Reduction of nitroblue tetrazolium to blue formazan was studied in sus- pensions Of resting and phagocytosing PMN. Cells (4.9 x 106/assay) were pooled from at least two guinea pigs and were incubated at 57°C in KRPS containing 10% autologous guinea pig serum. In one experiment the length of pre- incubation time was varied prior to stimulating cells with polystyrene latex particles. Whereas, in a second experiment 98 pre-incubation time was kept constant for 1 hr, while car— bohydrate concentrations were varied. In both cases stimulated cells were incubated at 57°C for an additional 50 min in the presence of nitroblue tetrazolium (140 uM), and reactions were stOpped by placing cell suspensions on ice. Cells were centrifuged at 5000 xg for 5 min, super- nates were decanted, and formazan was extracted from pellets of cells with 6.0 ml of pyridine. Formazan was quantitated by its absorption at 515 nm. Effects of carbohydrate upon the reduction of nitro- blue tetrazolium by superoxide anion were also determined in separate experiments without cells. Superoxide was generated non—enzymatically in 1.1 ml of KRPS containing: 78 uM NADH, 25 uM phenazine methosulfate, and 50 pM nitro- blue tetrazolium. Reactions were performed 6 to 8 times at room temperature using a Gilford model 5500 spectro— photometer in the general kinetic-l program mode. Cellular release of hydrogen peroxide. Rates of hydrogen peroxide release from PMN were quantitated by monitoring the decrease in scOpoletin fluorescence upon oxidation of scOpoletin by peroxide and horse radish peroxidase (7). PMN (0.50 x 10'7 to 1.25 x 107/assay) were incubated for 5 hr at 57°C in KRPS, pH 7.4. Cells were contained in 0.5 m1 dialysis bags and were dialyzed against 2.5 ml of the same buffer during incubation and for one hour at 4°C following incubation. Peroxide levels were determined by adding 0.1 99 ml of each dialysate to 1.9 ml of KRPS containing 2 uM sc0poletin and l ug/ml horse radish peroxidase. Reactions were completed after 45 min at room temperature, and fluorescence emission was then measured between 450 nm and 470 nm upon excitation with 515 nm plus 566 nm light. Since this system was standardized in the absence Of cells by adding known amounts of hydrogen peroxide (Figure 5), decreases in sc0poletin fluorescence were directly related to increases in peroxide generation. When phagocytosing PMN were employed, a 200 fold (particle: cell) excess of polystyrene latex particles was included in each dialysis bag. In similar experiments, peroxide levels were determined by monitoring the fluorescence of diacetylfluorescein after its oxidation by peroxide and horse radish peroxidase (18). This method required one-fifth the level of hydrogen perox— ide, as well as 0.5 uM leuko—diacetylfluorescein, prepared by exposing its dichlorO-derivative tO 0.01 N NaOH for 50 min (19). Although linear standard curves were found with both hydrogen donors, variations in fluorescence were sub— stantially less when sc0poletin was employed. PMN in all assays were pooled from at least four guinea pigs and were washed extensively with isotonic buffer to remove contaminating ascorbate and glutathione. Preliminary experiments performed on cells without washing, as well as cells without dialysis, gave inconsistent results. 100 l4C-labelled Cellular formate oxidation. Conversion of formate to 14CO2 by PMN was performed in 25 m1 siliconized Erlenmeyer flasks fitted with gas-tight stOppers and plastic centerwells. Cells (2.07 x 107/assay).were incubated for one hour at 57°C in 1.0 ml Of KRPS, pH 7.4, containing 1.0 uCi of l4C—labelled formate, diluted with non-isotopic car- rier to 2 mM. Glucose was present in all assays at a level of 5 mM, whereas galactose or potassium cyanide were in- cluded where apprOpriate at 50 mM or 5 mM levels, respec- tively. Reactions were stOpped by introducing 0.2 ml of 7.5 N H2804 to each cell suspension, and evolved CO2 was collected in plastic centerwells on 1 x 5 cm folded strips of KOH-saturated Whatman 5MM paper. Liberated CO2 was absorbed from the start of each experiment and for one hour after acidifying the medium. Centerwells were directly transferred to scintillation vials containing 10 ml of Bray's solution and counted. Each assay was performed in triplicate on cells pooled from at least two guinea pigs. Cellular oxygen consumption. Rates of oxygen consumption by resting and phagocytosing PMN (0.50 x 107 tO 0.97 x 107/ assay) were measured using a Yellow Springs Institute model 55 biological oxygen monitor with a Clark oxygen electrode. PMN were placed in siliconized glass chambers containing 5 ml of KRPS plus 10% guinea pig serum, and cells were pre— incubated with carbohydrate for 2 to 5 hrs at 57°C. Cell suspensions were then agitated with magnetic stirrers to 101 insure a continuous supply Of dissolved oxygen, and the rate of consumption was recorded over a 20 min period. Monitoring was briefly interrupted for the introduction of polystyrene latex particles (200 particles/cell) and re- cording was resumed for another 20 min. Each experiment employed cells from one guinea pig and was repeated 5 to 4 times on subsequent days. Since rates of consumption were recorded in ul/hr/cell number, the general gas law was ap- plied tO calculate the actual concentrations of oxygen consumed under these conditions. Methional assay for hydroxyl radicals. Interactions be- tween carbohydrates and hydroxyl radicals were determined by monitoring the inhibition ofthylene production from the hydroxyl radical dependent degradation of methional (20). Hydroxyl radicals were generated either enzymatically as a result of the xanthine oxidase reaction or non-enzymatically by a Fenton-type hydroxylating system without ascorbic acid. The former system included 1.0 mM methional, 0.2 mM xan— thine, O.l mM EDTA, and 150 mU xanthine oxidase in 1.0 ml of 50 mM potassium phosphate buffer, pH 7.8. Whereas, the latter system contained 1.0 mM methional and 1.0 mM H202 in 1.0 ml of the same buffer. Addition of ferrous iron was not necessary in the latter system since iron levels in phosphate buffer were sufficient to promote univalent peroxide reduction in the presence of dissolved oxygen. All reactions were performed at room temperature in 25 ml 102 siliconized Erlenmeyer flasks fitted with gas-tight stoppers. Gas above each reaction mixture was sampled with a gas-tight syringe and was analyzed for ethylene by injecting 1.0 ml aliquots into a Hewlett Packard model 440 gas chromato- graph equipped with a 6 foot glass column of Chromosorb 102. Column temperature and flame ionization detector temperature were maintained at 90°C and 220°C, respectively. Nanomoles of ethylene present in a flask at any given time (t) were calculated using a computer program based upon the following equation: nmolest = (Rs/Rstd) x (VZ/Vl) + nmolest_l where R8 = detector response with sample (cm) R = detector response with ethylene standard std (cm/nmole) V = volume of flask (ml) <: H 1 volume of injection (m1) In separate experiments, ethylene production from 1.0 mM methional was also investigated using cell suspensions of PMN (2.45 x 107/assay) and of macrophage (1.41 x 107/ assay) in 1.0 ml volumes of KRPS. These reactions were per- formed at 57°C, and a number of controls were run simul- taneously. Controls included: buffer alone, buffer plus polystyrene latex particles, and buffer plus killed cells.' Cells were killed both by heating to 90°C for 15 min and by a repetitive freeze—thaw procedure. 105 Oxidation of hexoses by hydroxyl radicals. Products of the reaction between carbohydrates and hydroxyl radicals were generated by incubating various levels of glucose or galac- tose in a Fenton-type hydroxylating system without ascor- bate. Reaction solutions of 1.0 ml contained: 5 mM FeSO4, 15 mM EDTA, 5 mM H 02, and l to 50 mM carbohydrate in 40 2 mM potassium phosphate buffer, pH 7.4. Reactions were initiated by adding hydrogen peroxide and were judged as complete when similar solutions without carobhydrate but containing 5 mM aniline gave an opaque dark-brown appear- ance. All solutions were incubated at room temperature and after completion were dried under N2 gas. Products were analyzed by gas chromatography and by ascending paper chromatography on Whatman 5MM paper using a 7:5 (v/v) ethanol: 1 M ammonium acetate solution as deve10per. Gas chromatography was performed on the tri- methylsilyl derivatives Of products using a Hewlett Packard model 5850A gas chromatograph equipped with OV-l and XE-6O columns. Standards included the trimethylsilyl derivatives of each hexose, each corresponding hexonic acid (in both open and closed ring forms), each hexuronic acid, and a variety of pentoses and disaccharides. Derivatives were prepared as previously described (21). Hexonic acid production by PMN was similarly investi- gated by incubating cells pooled from 5 guinea pigs (5.84 x 108/assay) in 1.11 ml Of KRPS containing either 5.0 mM glucose or 5.0 mM glucose with 50 mM galactose. 104 Incubations were run for 6 hr at 57°C and were terminated by adding one volume of 50% trichloroacetic acid while vortexing. Protein was removed by centrifugation at 20,000 xg for 15 min, and supernates were extracted 4 times with 5 volumes of diethyl ether. Neutralized super— nates were then dried under N2 and derivatized for gas chromatography. Standards were similarly treated with trichloroacetic acid. In order to determine whether hydroxyl radicals could contribute to the marked increase in glucose C-l oxidation Observed during phagocytosis, glucose and galactose oxida- tion experiments as described in Chapter III were repeated but in the presence of free radical scavengers. These in— cluded 0.2 mg/ml superoxide dismutase, 0.5 mg/ml cytochrome g, 0.5 mg/ml catalase, and 50 mM sucrose. Polygl formation by PMN. Production of sorbitol and galac— titol by PMN was investigated by incubating cells pooled from 5 guinea pigs in 1 ml aliquots of KRPS containing 50 mM glucose or 50 mM galactose. After incubating cells (1.52 x 107/assay) for varying lengths of time at 57°C, cell suspensions were centrifuged at 5000 xg for l min, and the extracellular media was discarded. Pellets of cells were immediately washed with fresh medium in the absence of carbohydrate and frozen with liquid N2. PMN were stored at —80°C until homogenizing in the presence Of 5 N perchloric acid (see Chapter I, Metabolite assays). 105 Quantitation of polyol levels in cell homogenates was per- formed by gas chromatography Of the trimethylsilyl deriva- tives. Each determination was run at least in duplicate, and similar experiments were performed on Cells incubated without carbohydrate, with 5 mM glucose, or with 50 mM galactose plus polystyrene latex particles. PMN chemiluminescence. Chemiluminescence was monitored with a Beckman CPM-100 liquid scintillation spectrometer operated in the out—of—coincidence summation mode. The instrument was used at room temperature with an Open window setting and at 100% gain. PMN (5.98 x 107/assay) were pooled from four guinea pigs and were incubated in siliconized glass scintillation vials containing 6.0 ml of KRPS. Cells were kept in the dark for eight hours at 57°C, and were stimulated by adding a 20 fold (bacteria: PMN) excess of heat-killed E. coli in 0.4 m1 Of autologous guinea pig serum. Chemiluminescence was monitored every 12 seconds for 2 min before and 4 min after stimulation of cells. Bacteria were grown and heat- killed as described in Chapter I. In separate experiments, superoxide dismutase and triethylenediamine were included at 0.2 mg/ml and 0.1 mM, respectively. 106 Results Extracellular reduction of gytochrome c. Effects of car- bohydrate, superoxide dismutase and catalase upon the re- duction of cytochrome g by PMN (Table I) can be enumerated as follows: 1) reduction was more pronounced when phago- cytosing rather than resting cells were employed, 2) reduc- tion was enhanced by pre-incubating cells with 5.0 mM glucose, 5) reduction was sensitive to superoxide dismutase but not to catalase, 4) reduction was impaired by pre- incubating cells with 50 mM glucose or 50 mM galactose, 5) reduction was also impaired by galactose when added after the pre-incubation period, and 6) the inhibitory action Of galactose did not occur in the presence of catalase. The former three effects indicated that cytochrome reduction was dependent upon superoxide anion release from PMN, whereas the latter three effects suggested that a catalase- sensitive inhibition of cytochrome reduction occurred in the presence of 50 mM carbohydrate. This catalase—sensi- tive impairment could be explained on the basis of extra- cellular cytochrome re—oxidation 1) if production of an oxidizing agent, such as hydrogen peroxide, was enhanced by elevated carbohydrate or 2) if other oxidizing agents, such as hydroxyl radical or a longer-lived carbohydrate radical, were formed. To test these suggestions, we under- took further experiments to determine whether elevated levels of galactose affected hydrogen peroxide generation or oxygen consumption by PMN. 107 Cellular reduction Of nitroblue tetrazolium. Further evi- dence that elevated levels Of carbohydrate affected free- radical type reactions in PMN was Obtained by monitoring the reduction of nitroblue tetrazolium. A time-course study (Figure 1) showed that dye reduction was independent Of pre—incubation time when phagocytosing cells were kept in KRPS with 5.0 mM glucose. However, when these cells were exposed to either 50 mM glucose or 5.0 mM glucose plus 50 mM galactose, the reduction of dye was impaired after 1 hr Of pre-incubation. Data in Figure 2 demonstrated that nitroblue tetrazolium reduction was stimulated during phagocytosis and that impairment of this reduction was de- pendent upon the extracellular galactose concentration. In separate experiments without cells (Table II), neither the superoxide-dependent reduction of cytochrome 9 nor the reduction Of nitroblue tetrazolium were impaired by elevated levels of carbohydrate. These observations indicated that carbohydrate could not directly interact with cellular superoxide. However, it was possible that carbohydrate could lower cellular superoxide production or that other free radicals, if formed, could interact with superoxide anion. Cellular release of hydrogen peroxide. As measured by the oxidation of sc0poletin, levels Of peroxide released from phagocytosing PMN were approximately 2 fold greater than those released from resting PMN (Table III). Catalase, 108 when added to these cell suspensions, substantially lowered the oxidation Of sc0poletin by 82% and 66%, respectively. Release Of hydrogen peroxide from PMN was not significantly affected by external carbohydrate. However, slight in— creases in peroxide formation (up to 157% of the control) were found when resting PMN were incubated with 50 mM glucose or 50 mM galactose. Cellular formate oxidation. Intracellular steady-state levels of hydrogen peroxide were estimated by observing the oxidation Of l4C-labelled formate by endogenous catalase Operating in the peroxidatic mode. Potassium cyanide, which should have completely inhibited this reac- tion, was found to decrease the evolution of radioactivity by only 46%. This suggested that labelled formate was evaporating during the experiment and that values Obtained in the presence of cyanide should be considered as back- ground. After correcting for this background, levels of 14CO2 produced by PMN incubated in 5.0 mM glucose were 1.9 fold greater than those produced by PMN incubated in 5.0 mM glucose plus 50 mM galactose. The difference be- tween these values, however, was not statistically sig— nificant (Table IV). Cellular oxygen consumption. As indicated in Table V, rates of oxygen consumption by PMN were stimulated 2 to 11 fold during phagocytosis of polystyrene latex particles. Significant differences in these rates were not found, 109 however, to result from the incubation of either resting or phagocytosing cells in medium containing elevated levels of carbohydrate. Methional assay for hydroxyl radicals. The production of ethylene from methional during the enzymatic conversion of xanthine to urate (Figure 4) indicated that hydroxyl radi— cal was produced. Since both hydrogen peroxide and super— oxide arise from the xanthine oxidase reaction (22), hydroxyl radical was probably formed secondarily by a Haber Weiss reaction as follows: - . 1 02- + 11202 -——-—> OH + 0H + 02 (eq. 1) Hydroxyl radical then interacted with methional to produce ethylene. However, in the presence of polyol the formation of ethylene was impaired. Since polyol did not interfere with the xanthine oxidase reaction (Table II), these data indicated that galactitol, mannitol and sorbitol were ef- fectively scavenging hydroxyl radical. Data in Figure 4 were Obtained from one representative experiment in which all four reactions were run simultaneously. Subsequent experiments gave similar results for polyol inhibition, but the overall rates of ethylene production were lower. This resulted from the rapid decay in xanthine oxidase activity with time. To avoid such variations in the rate Of hydroxyl radi- cal generation, a simple non-enzymatic system was 110 constructed by adding 1.0 mM H O and 1.0 mM methional to 2 2 50 mM potassium phosphate buffer. This cyclic system (Figure 5) depended upon the following three reactions: H202 + Fe+2——> OH- + OH' + Fe+3 (eq. 2) H202 + 0H° -+> 0H“ + 02‘ + 2H (eq. 5) 024 + Be+3———> 102 + Fe+2 (eq. 4) Ethylene production from the hydroxyl radical depend- ent degradation of methional was thus stimulated by adding more H2O2 or ferrous iron and was inhibited by adding superoxide dismutase or benzoate. Ethylene formation was also impaired by including 40 mM levels Of polyol, hexose or myo—inositol. Each point in Figure 5 was obtained by averaging values from 4 to 6 determinations, and all differ- ences between average control values and values Obtained with inhibitors were significant at the P-15 mM). The former Observation indicated that some glucose may be necessary for maintenance of intra- cellular reducing equivalents through the hexose-monophos- phate shunt and substantiated the results of Iyer §t_§4. (8) that intracellular peroxide production was decreased under these conditions. The latter observation, however, could not be accounted for by decreased hexose—monOphos- phate shunt activity since elevated levels of carbohydrate did not significantly affect glucose C-l oxidation (Chapter I). Moreover, the inhibitory effect of galactose upon cyto- chrome 9 reduction did not require pre—incubation of cells 115 with hexose, and this impairment was not Observed in the presence Of catalase. Thus, galactose acted predominantely upon the cytochrome g assay system, and its reaction de- pended upon the presence Of extracellular hydrogen peroxide. This impairment could be explained on the basis of the fol- lowing model in which cytochrome g was reduced by super- oxide anion and reoxidized by carbohydrate radical (G’): Since PMN are known to produce 02; and H 0 it has 2 27 been proposed (4,11) that hydroxyl radical is formed via the Haber Weiss reaction (eq. 1). Carbohydrates, such as glucose or galactose, could then interact with hydroxyl radical to form carbohydrate radical as follows: GH + OH' —-—> G' + H20 (eq. 7) This reaction has been Observed with glucose (25), mannitol (20,24) and sucrose (25), and our results in Figures 4 and 5 demonstrate that a variety of hexoses and polyols are capable of this reaction. Data from pulse radiolysis studies have shown that complete scavenging of hydroxyl radical can occur with 5 mM glucose (26), and second-order 116 rate constants for reaction of glucose and sucrose with 10 M-1 -1 0H“ have been calculated to be 1 x 10 sec and 2.5 x 109 M71 sec-1 , respectively (25,25). Like hydroxyl radical and mannitol radical (24), carbohydrate radicals could oxidize ferrocytochrome 9 (eq. 6) or react with superoxide anion via the following reaction: G' +0; 2 + H+—» GH + 02 (eq. 8) However, mannitol radical does not compete well with excess ferricytochrome g for reaction with superoxide anion (24), and thus in our system, carbohydrate radicals probably do not affect the reduction Of cytochrome per se. Carbohy- drate radicals could, however, as in the phagolysosome scavenge both hydroxyl radical (eq. 7) and superoxide anion (eq. 8). A major point in this discussion is that catalase did not affect the cytochrome 9 reduction assay when PMN were incubated with 5 mM glucose. This observation has been previously observed (9) and can be explained on the basis of ferricytochrome g and carbohydrate concentrations. As mentioned earlier, leukocyte H202 may be formed by the non- enzymatic dismutation of superoxide anion (4,27): 202: + 2H+——. 11202 + 102 (eq. 9) However, at high concentrations Of ferricytochrome 2, re- duction Of extracellular cytochrome occurs (eq. 5), and 117 the above reaction (eq. 9) is competitively inhibited. As superoxide anion is formed, levels of ferricytochrome 3 decrease, and the production of hydrogen peroxide becomes favored. Although peroxide concentrations may then be adequate to promote hydroxyl radical formation (eq. 1), 5 mM levels of carbohydrate may be insufficient to compete with other agents for reaction with OH'. Such agents could be contributed by serum, and this competition could be over- come by increasing the carbohydrate concentration. That the above model (eq. 5 and eq. 6) operated extra- cellularly was supported by the failure of 50 mM galactose to significantly affect either hydrogen peroxide generation (Table III) or oxygen consumption (Table V). These Observa- tions suggested that extracellular levels of superoxide anion were not impaired and that the action of carbohydrate upon this system must be through cytochrome g reridation. These results also showed that increases in H202 generation could not account for this effect. Rates of extracellular H202 production (Table III) were in the range Of previously reported values (6,7), and rates of oxygen consumption (Table V) were comparable to those reported by Root §£_§4. (7). Both parameters were similarly stimulated by phago- cytosis. Significant decreases in nitroblue tetrazolium reduc— tion, when phagocytosing PMN were incubated with 50 mM glucose or galactose (Figure 1), suggested that hexose or polyol accumulated intracellularly and that this accumulation 118 promoted carbohydrate radical formation (eq. 7). Carbohy- drate radical could have then competed (eq. 8) with nitro- blue tetrazolium for reduction by superoxide anion. These suggestions were supported by 1) the effect Of increasing galactose concentrations upon intracellular nitroblue tetrazolium reduction (Figure 2), 2) the detection of polyol accumulation within PMN (Figures 7 and 8), and 5) the failure of carbohydrate alone to react with superoxide anion (Table II). Nevertheless, a certain amount of caution must be expressed regarding this interpretation since the results in Figures 1 and 2 could also be explained by decreased cell permeability to nitroblue tetrazolium and since it is not clear whether all nitroblue tetrazolium reduction by PMN is superoxide-dependent. Although formazin formation can be inhibited by superoxide dismutase in strictly chemical (28) and enzymatic systems (29), some evidence has been presented that superoxide dismutase does not com- pletely inhibit cellular dye reduction (4,50) and that this reduction is also inhibited by a variety of other proteins (51). However, formazan accumulation is stimulated by phagocytosis (52) and is impaired in leukocytes from pa— tients with chronic granulomatous disease (4), a dysfunc- tion of NADH-oxidase activity (55,54). Our results obtained from studies on cellular formate oxidation (Table IV) and PMN chemiluminescence (Figure 9) were also consistent with the above model. Although rates 119 of l4C-labelled formate oxidation were not significantly de- creased by incubating cells in 50 mM galactose, the lack of significance could largely be attributed to variations re- sulting from labelled formate evaporation. Similar problems were encountered by Iyer §j_24. (8). Average rates of formate oxidation, which depended upon the reaction of H202 with catalase Operating in the peroxidatic mode (55), were lower in the presence of 50 mM galactose. Thus, steady- state levels Of H202 were probably decreased intracellularly. This could be accounted for by carbohydrate radical formation within the phagolysosome (eq. 7), and its subsequent reaction with superoxide anion (eq. 8) to compete with further H202 production (eq. 9). Briggs g§_§4. have recently shown, using a histochemical method for H202 localization, that H202 was formed via NADH-oxidase at the external plasma membrane and that this activity was partially internalized during phagocytosis (50). Therefore, the intercellular reaction between carbohydrate radical and superoxide anion (eq. 8) would not necessarily affect extracellular super— oxide or peroxide levels. This agreed with our data on extracellular peroxide formation (Table III) and indicated that internal hydrogen peroxide was not transferred across the plasma membrane. Such a transfer would not be probable because of the high cellular content of catalase and myelo— peroxidase (56). Chemiluminescence Of phagocytosing PMN (Figure 9) was enhanced by pre-incubating cells with 20 mM or 50 mM 120 galactose. This effect indicated an interaction between galactose and free radicals which resulted in light emis— sion from either excited carbohydrate molecules or some other meta-stable intermediate. Chemiluminescence of PMN has been associated with the production Of singlet oxygen by a number of reactions (4,12,15). These have included the Haber Weiss reaction (eq. 1), the non-enzymatic dismutation of super- oxide anion (eq. 9), the reduction Of H202 by hypochlorite and myeloperoxidase (12,27), and the reaction between hydroxyl radical and superoxide anion (57). Since the latter reaction is analogous to (eq. 8), increased chem- iluminescence could be interpreted as indicating the occur— rence of this reaction when cells are incubated with galac- tose. Benzoate, which scavenges hydroxyl radical, as well as superoxide dismutase and catalase have been shown to partially inhibit PMN chemiluminescence (57). Our results also demonstrated this inhibition with superoxide dismutase; however, the singlet oxygen trapper, triethylenediamine (1,4—diazobicyclo-2,2,2-Octane)(58), was without effect. This observation did not support the above interpretation in regard to the source of chemiluminescence with galactose, and this suggested that excited carbohydrate molecules may be the light emitting species. Allen g£_g4. have also sug- gested that excited carbonyls promote this reaction (59). Excess levels of glucose did not enhance chemilumines- cence by PMN in our studies. This could be a result Of the high rate of glycolysis (l) and thus low levels of glucose 121 accumulation (40). In addition, reduction of glucose to sorbitol was at least 5 fold less than the reduction of galactose to galactitol when comparable levels of galactose were present (Figure 8). A variety of hexoses and polyols competed with methion- al for reaction with hydroxyl radical (Figures 4 and 5). These Observations indicated that the resulting carbohydrate radicals were not as powerful in oxidizing potential as the hydroxyl radical which they replaced. Nevertheless, these radicals were capable of oxidizing ferrocytochrome g and superoxide anion, as previously discussed, and could undergo further oxidation by reacting with a second hydroxyl radi- cal. The latter reaction was observed by incubating glucose and galactose in a Fenton-type hydroxylating system (Figure 6). Since the products of this reaction were gluconic and galactonic acids, respectively, the hydroxyl radical must have abstracted a hydrogen from the hexose C-l position (eq. 10) and secondarily combined with the carbohydrate radical (eq. 11): 0 0 QC-H + 0H° -—+> *C' + H20 (eq. 10) I O 0\ §C° + 0H'——> \C-OH (eq. 11) I I The formation of gluconic acid in this manner has been previously observed upon pulse radiolysis and X -irradia- tion of glucose solutions (26). However, to our knowledge, this was the first observation of hexonic acid production 122 from a Fenton—type hydroxylating system. As evident from our 20% yield of hexonic acid from 1 mM hexose, levels of hydroxyl radical in this system ap- proached 1074M. However, steady-state levels of hydroxyl radical in PMN are probably at least 2 orders of magnitude less at any given period of time. Thus, even though carbo- hydrate radicals may be formed in PMN, the further oxida- tion Of these radicals to hexonic acid is less likely to occur. Side reactions as proposed by McCord g§_§4. for the dismutation of mannitol radical (24) may also be favored, and in this case only minute quantities of hexonic acid would be formed. These levels of hexonic acid may evade detection by gas chromatography. In conclusion, carbohydrate radicals have been demon- strated in two hydroxyl radical generating systems. One Of these systems produced non-biological levels of this radical, while the other produced low steady-state levels of hydroxyl radical comparable to levels associated with phagocytosing PMN. Hexonic acid formation from hexose was found in the former system, whereas carbohydrate radical formation was followed in the latter system by monitoring inhibition of the hydroxyl radical—dependent degradation of methional. These results suggested that carbohydrate radicals were formed by PMN, and indeed, most Of the effects of 50 mM galactose upon parameters of the oxygen-dependent killing mechanism could be explained on this basis. As demonstrated in Chapter I, 50 mM galactose impaired 125 phagocytosis of 32P-labelled E. coli by 24% in guinea pig PMN. However, the-killing Of E. coli was almost completely inhibited (<'1% activity) under these conditions. Since this bacterium is greatly affected by the oxygen-dependent killing mechanism of PMN (27), the impaired bactericidal activity could be accounted for by the reaction of carbo- hydrate radicals with superoxide anion (eq. 8) within the phagolysosome. This suggestion is supported by our data on cellular nitroblue tetrazolium reduction, formate oxi- dation, and chemiluminescence. 10. 11. 12. 15. 14. 15. 124 References Sbarra, A.J., and Karnovsky, M.L. (1959) J. Biol. Chem. 244, 1355-1362 Holmes, B., Page, A.R., Windhorst, D.B., Quie, P.G., White, J.G., and Good, R.A. (1968) Ann. N.Y. Acad. Sci. 155, 888—901 Mandell, G.L. (1974) Infect. Immun. 2, 557-541 Johnston, Jr., R.B., Keele, Jr., B.B., Misra, H.P., Webb, L.S., Lehmeyer, J.E., and Rajagopalan, K.V. (1975) In The Phggocytic Cell in Host Resistance pp 61-75 (Bellanti, J.A., and Dayton, D.H., eds.) Raven Press, New York, N.Y. Yost, Jr., F.J., and Fridovich, I. (1974) Arch. Biochem. Biophys. 161, 595-401 Paul, B., and Sbarra, A.J. (1968) Biochim. Biophys. Acta 156, 168-178 Root, R.K., Metcalf, J., Oshino, N., and Chance, B. (1975) J. Clin. Invest. 33, 945-955 Iyer, G.Y.N., Islam, D.M.F., and Quastel, J.H. (1961) Nature 192, 555-541 Babior, B., Kipnes, R., and Curnutte, J. (1973) J. Clin. Invest. 22. 741-744 Drath, D.B., and Karnovsky, M.L. (1975) J. Exp. Med. 141, 257-262 Salin, M.L., and McCord, J.M. (1975) J. Clin. Invest. 29. 1319—1323 Krinsky, N.I. (1974) Science 3gp, 363-365 Allen, R.C., Stjernholm, R.L., and Steele, R.H. (1972) Biochem. Biophys. Res. Commun. 41, 679-684 Baehner, R.L., Gilman, N., and Karnovsky, M.L. (1970) J. Clin. Invest. 42, 692-700 Cagan, R.H., and Karnovsky, M.L. (1964) Nature 204, 255-256 l6. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 125 Noseworthy, Jr., J., and Karnovsky, M.L. (1972) Enzyme l , 110-131 Paul, B.B., Strauss, R.R., Jacobs, A.A., and Sbarra, A.J. (1970) Infect. Immun. I, 338-344 Black, M.J., and Brandt, R.B. (1974)-Anal. Biochem. 5g, 246-254 Keston, A.S., and Brandt, R. (1965) Anal. Biochem. 1.1.. 1-5 Beauchamp, C., and Fridovich, I. (1970) J. Biol. Chem. 245, 4641-4646 Sweeley, C.C., Bentley, R., Makita, M., and Wells, W.W. (1963) J. Am. Chem. Soc. §2, 2497 Fridovich, I. (1970) J. Biol. Chem. 232. 4053—4057 Davies, J.V., Griffiths, W., and Phillips, G.O. (1965) In Pulse Radiolysis p. 181 (Ebert, M., ed.) Academic Press, New York, N.Y. McCord, J.M., and Fridovich, I. (1973) Photochem. Photobiol. I], 115—121 Ward, J.F., and Myers,er., L.S. (1965) Radiation Res. g5, 483 Phillips, G.O., Griffiths, W., and Davies, J.V. (1966) J. Chem. Soc. (B) l94-2OO Klebanoff, S.J. (1975) In The Phaggcytic Cell in Host Resistance pp. 45-56 (Bellanti, J.A., and Dayton, D.H., eds.) Raven Press, New York, N.Y. Nishikimi, M., Rao, N.A., and Yagi, K. (1972) Biochem. Biophys. Res. Commun. fig, 849—854 Beauchamp, C., and Fridovich, I. (1971) Anal. Biochem. g5, 276—287 Briggs, R.T., Drath, D.B., Karnovsky, M.L., and Kar- novsky, M.J. (1975) J. Cell. Biol. g1, 566—586 Amano, D., Ka osaki, Y., Usui, T., Yamamoto, S., and Hayaishi, 0. $1975) Biochem. BiOphys. Res. Commun. §§, 272—279 Holmes, B., Page, A.R., and Good, R.A. (1967) J. Clin. Invest. fig, 1422—1428 33. 34. 35. 36. 37. 38. 39. 40. 126 Babior, B.M., Curnutte, J.T., Hull, W.E., and Kipnis, E.S. (1973) J. Clin. Invest. 52, 5a Baehner, R.L., and Karnovsky, M.L. (1968) Science 162, 1277-1279 Chance, B. (1949) Acta Chem. Scand. 1, 236-267 Klebanoff, S.J. (1972) In The Molecular Basis of Electron Trans ort pp. 275-295 (Woessner, J.F., and Hfiijing, F., egs.5 Academic Press, New York, N.Y. Webb, L.S., Keele, Jr., B.B., and Johnston, Jr., R.B. (1974) Infect. Immun. 2, 1051-1056 Foote, C.S., Denny, R.W., Weaver, L., Chang, Y., and Peters, J. (1971) Ann. N.Y. Acad. Sci. 171, 139-145 Allen, R.C., Yevich, S.J., Orth, R.W., and Steele, R.H. (1974) Biochem. BiOphys. Res. Commun. go, 909- 917 Englhardt, A., and Metz, T. (1971) Diabetologica 1, 143-151 127 macamnwmmSm Hamo .nOHPwH>mU vsmvnwpm mopwoflcnw pm 809% UmHoom mHHmo no meOHH95hdwsd ma coasomsmm who: whommm HH< mm: mopmnhmmfim Hamo Ga maohsoopho woodcmm ngpo asp mafins .mnman sowpomop w mm mbhom ow moa so voomam mm: poSwHHm mso .mmHm mmqfism 039 pmmoa .8: 0mm Pm woMOpHnoa .OOHM pa Aficg HmHOHpHccw am How covandoqfi mmz .mposvflawafi o.H 039 oqu cmuflbflu cam A21 mnav m m80h£oophoassom Mo madao> a Spa: cmPSHHU soap who: .asnmm wan mmnflsw KOH Spas .¢.H mm .QOHPSHom mpmsmmosm sowmflmlmpmsm me He o.H as an m sow cosm pm ccecpsceflucsm memes Asmmmc\soa x co.a op see u mm.ov 22mm mHHmo H0H\sdo£ \cmosvos o waohnoovho mmaos: spas mucepecee com.m H mo.m eem.o m mo.o cmceccacc o o.m mm.H + cH.e cmmacpco 0.0m o.m ma.m m H.0H mm.o m mm.e cmmameco o o.m emm.m + mm.m cme.o + me.a mom 0 o.m cce.a m NN.H cmm.o H ma.o 0.0m o.m cco.H + mm.m o.mH o.m Hm.e m mm.c cce.o H mc.o o 0.0m mm.o H mo.m I o o.ma mm.m H o.oa Hm.o H He.m o o.m cm.H + mm.m me.o + Hm.m o o Azsv ~29“ omsflmophoommnm mcfipmom mmopomamw mmoofiaw .m.m H o oaosnoopmo moaprSOQHIohm cc posse maceeecee wzzm an Goaposuoh o oaosnoopmo so poommo mpchhflonsmo .H magma 128 .mmoosaw 28m SH mHHmo SPHS mosam> 809% Amo.o.vmv pmmsmmMHw hapsmoflgfimmflmw .moaoapsmm Nmpma ososmpmhaom Mo mmooxo cao% Om m wm>fiooos mHHoo wmfimopmoommgmo .HHm>HPommmos .28 om cam .Ha\ma m.o .Ha\ms m.o Pm enommsm ohms omopomfimm cam .mmmHMPmo .Amomv ommpsamfiv mwflxohmmSmn Accseeeccov H cance 129 Table II. Effect of carbohydrate on reactions involving superoxide anion radicala Nitroblue tetrazolium Cytochrome c reducedb reductionC . 2 Addition nmoles/min A517 nm/mln x 10 None 1.49 i 0.07 0.95 i 0.05 Galactose 1.61 i 0.21 1.08 i 0.05 Galactitol 1.52 i 0.15 0.96 i 0.06 Sorbitol 1.55 i 0.18 1.05 i 0.06 Mannitol 1.55 i 0.18 0.99 i 0.06 Xylitol 1.57 i 0.14 aEach reaction was performed 6 to 8 times using a Gilford model 3500 spectrOphotometer in the general kinetic-1 pro- All values represent average lepes i standard gram mode. deviation. All carbohydrate levels were 27 nM. bSuperoxide was generated enzymatically in 1.1 ml of 0.05 M sodium phosphate buffer, pH 7.8, containing: 50 uM xanthine, 100 uM EDTA, 10 uM ferricytochrome g, and 3.2 mU xanthine oxidase. CSuperoxide was generated non—enzymatically in 1.1 m1 of Krebs-Ringer phosphate solution, pH 7.4, containing: 78 pM NADH, 2} uM phenazine methosulfate, and 50 pM nitroblue tetrazolium. 130 Table III. Effect of carbohydrate on hydrogen peroxide formation by PMNa nmoles H 02 produced/hour/lO7 cells 2. Additions : S.D. Glucose Galactose b (mM) (mM) Other Resting Phagocytosing o o 4.52 i 1.16 5.0 o 4.61 i 0.88 9.21 i 0.19 50.0 o 6.22 i 1.17 9.55 i 0.54 5.0 50.0 5.55 i 1.59 9.01 i 0.75 5.0 o Catalasec 1.54 i 1.09 1.61 i 1.56 aPMN (0.50 x 10'7 to 1.23 x 107/assay) were incubated for 3 hr in Krebs-Ringer phosphate solution, pH 7.4, at 37°C. Cells were contained in 0.5 m1 dialysis bags and dialyzed against 2.5 m1 of the same buffer during incubation. Peroxide levels were determined in dialysates by measuring the oxidation of s00poletin by peroxidase. Each value is a mean of quadruplicate determinations performed on cells from four guinea pigs. S.D. indicates standard deviation. bPhagocytosing cells were exposed to a 200 fold (particle: cell) excess of polystyrene latex particles. CCatalase was present at 0.3 mg/ml. 131 Table IV. Carbohydrate effect on l4C-formate oxidation by PMNa 14 cpm released nmoles CO2 Additions : S.D. hr.l/107 cells Glucose Galactose (mMz (mM) 5.0 0 18.2 i 1.47 7.6 5.0 30.0 14.4 i 1.57 4.1 5.0b o 9.86 : 1.11C o.o aPMN (2.07 x 107/assay) were incubated for 1 hr at 37°C in 1.0 ml of KRPS, pH 7,4, containing 1.0 pCi l4C—formate. Each assay was performed in triplicate on cells pooled from two guinea pigs. S.D. refers to standard deviation. bKCN was present at 5 mM. cSignificantly different (P<70.005) from values obtained with cells incubated with 5 mM glucose. 132 Table V. Effect of carbohydrate on oxygen consumption by PMNa nmoles 02 consumed/hour/10'7 cells Carbohydrate (mM) : S.D. Glucose Galactose Resting Phagocytosingb 5.0 0 127 i 45.2 701 i 455 5.0 10.0 80.4: 24.5 901 i 454 5.0 20.0 156 i 56.5 865 i 579 5.0 50.0 114 i 54.8 521 i 268 50.0 0 156 i 49.1 5540 aPMN were incubated from 2 to 3 hours at 37°C in Krebs- Ringer phosphate solution, pH 7.4, with 10% guinea pig serum. Oxygen consumption was measured using a Yellow Springs Institute model 53 biological oxygen monitor with a Clark oxygen electrode. Each experiment employed cells from one guinea pig, and each value is a mean of tripli- cate or quadruplicate experiments. S.D. refers to standard deviation. bPhagocytosing cells were exposed to a 200 fold (particle: cell) excess of polystyrene latex particles. CSingle determination. 133 O.l5- ”‘7 1 E 6 I) 3 1 vi 0 Z < a: 0: 8 m 0.0 4 5T o ‘ ' ' O LO 2.0 TIME (hours) Figure 1. Effect of carbohydrate and pre-incubation time on nitroblue tetrazolium reduction by PMN. PMN were incubated with 5.0 mM glucose (0 ), 30 mM glucose ( o ), or 30 mM galactose (I) for the indicated times, prior to assay for NBT reduction. All three groups were assayed at zero time. 134 resting phagocytosing I 7| 7 0.3 b 8' 8;. 0.2 ‘- 3 l \ Q - § § S \ i 1 \ 5 \ l l mM Glucose: 5 5 5 5 mM Galactose: 0 10 20 30 Figure 2. Effect of galactose concentration on nitroblue tetrazolium reduction by PMN. Both resting and phagocytosing cells were in- cubated for 1 hr in the appropriate media prior to assay. Reduction is indicated by increased absorbance at 515 nm. ’ 135 21) 40 6x) 8‘) 8.0- U ‘2’ ,u 61)” (J ‘8’ g «.0P .1 u. 21)" 3 O l 1 P l l 1 l 0 0A1 ()8 L2 nmoles H,0, Figure 3. Relationship between sc0poletin fluorescence and nanomoles of hydrogen peroxide. A FLUORESCENCE 136 Control 4.0 — 0 C 2 Galactitol ; 3.0 - IT] Nonnltol Sorbitol E 2.0 - O E c I.O - 0 ‘ ‘ ' ' 0 IO 20 30 40 TIME (min) Figure 4. Effect of polyols on hydroxyl radical dependent ethylene formation from methional. Enzymatic generation of hydroxyl radicals. All polyol levels were 40 mM. 137 Control 4.0 - Fructou Galactose 2.0 "' Glucooo / Monnooo Control Ethylene Coloctitol Sorbitol Monnitol nmobs Control 4.0 "' myo- lnooltofi o / Bonzooto / Suporoxido Dlornutou o .4- __ l l I O 4.0 8.0 '2 I5 TIME (min) Figure 5. Effect of carbohydrates and polyols on the hydroxyl radical dependent formation of ethylene from meth- ional. Non—enzymatic generation of hydroxyl radicals. All carbohydrate and polyol levels were 40 mM. Benzoate and superoxide dismutase were 5 mM and 0.2 mg/ml, respectively. 138 m I (D I Z I O l l I m I “J I a: I 3 I 4 I a: I I9 ' 2 A C U u] 4 '— m G m U l I l l O 50 IOO 0 5O IOO 'HME (mHL) Figure 6. Identification of hexonic acids as products of the reaction between carbohydrates and hydroxyl radi- cals in vitro. Products are represented in (A) and (C), while standards of gluconic acid and galactonic acid are re resented in (B) and (D), respectively. Peaks are (1 glucono—i—lactone, (2) gluconic acid (open form), (3) galactono—X— lactone, and (4) galactonic acid (open form). 139 RESPONSE DETECTOR KIg ‘ 4 I ‘ l I Figure 7. 4.0 8 .O O 4.0 8.0 TIME (mln.) Identification of galactitol in homogenates of PMN after incubation with 30 mM galactose. Standards are in (A), and cell homogenate in (B). Peaks are (l)o<-methylmannoside, (2) galactitol, (3) galactose, and (4) glucose. |.20 0.80 0.40 POLYOL (nmoloo / Io‘ collo) Figure 8. 140 2.0 TIME (hr. 4.0 I -‘ 3.0 '20 " |.0 Accumulation of intracellular polyol during incubation of PMN with elevated levels of carbo— hydrate. PMN were incubated in 30 mM galactose ( o ) or 30 mM glucose (0) media. (mM) POLYOL INTRACELLULAR l2.0 141 2 o. \_\ \ C) °\ v: \ O - \\\ \- \_\ \\_ .\"-\\ O l l L l l I l 0 LG 2.0 3.0 TIME (min.) Figure 9. Effect of carbohydrate on PMN chemiluminescence. Heat-killed bacteria were added at zero time to cell suspensions pre-incubated in media con- taining (oI) 5 mM glucose, ( o ) 30 mM glucose, (:1) 5 mM glucose plus 20 mM galactose, or (I) 5 mM glucose plus 30 mM galactose. SUMMARY Although phagocytic cells from galactosemic patients were not employed in the present research, the preceding results indicate that these cells would exhibit depressed phagocytic and bactericidal activities in the presence of galactose. Decreased killing of E. coli was observed under galactosemic conditions in vitro using normal leukocytes from peripheral blood of human (p. 32), guinea pig (p. 33), and chick (p. 57) sources. This response was impaired to a greater extent when leukocytes frbm galactosemic rather than normal chicks were employed. In addition, galactose was found to impair phagocytic activities of normal guinea pig PMN (p. 38) and peritoneal macrOphages (p. 39) as well as the intravascular clearance of colloidal 125I—BSA by chicks (p. 58). These results strongly suggest that impaired phagocyte function may be the underlying cause of suscep— tibility to E. coli infection among galactosemic infants. There are, however, many aspects of host defense which were not investigated and which could also contribute to in— creased infection. These include (i) the permeability of the intestinal lining to E. coli, (ii) the capacities of the complement, properdin, and humoral antibody systems, (iii) the capacity of the cellular immune system, and (iv) the susceptibility of E. coli to antibiotics in the presence 142 143 of elevated blood galactose. The inhibitory action of galactose upon phagocytosis per se can be accounted for by decreased intracellular ATP. Levels of ATP were found to be lower in guinea pig PMN (p. 86) and peritoneal macrophages (p. 36) during cul— ture of cells with 30 mM galactose. These decreases could be attributed to two basic inhibitory mechanisms. The first, and most likely in galactosemic phagocytes, is the accumulation of galactose or one of its metabolites intra- cellularly which could subsequently impair glycolysis by inhibiting one of the glycolytic enzymes. It is known that both galactose and galactose-l-phosphate accumulate in galactosemic PMN (1), and data with macrophages (p. 36) suggest that fructose-l,6-bisphosphate aldolase may be in- hibited. Decreases in ATP could further result in lower levels of glucose uptake (p. 85) and additionally impair glycolysis. The second mechanism by which galactose could lower ATP is present in guinea pig PMN (p. 35) and depends upon the conversion of galactose to free glucose. This pathway requires ATP (2,3) and could complete with the energy re- quiring steps leading to phagocytosis. That this conversion is stimulated by 2 fold during phagocytosis (p. 35) supports this suggestion. However, stimulation could also indicate an increase in intracytoplasmic phosphatase activity. Since PMN do not contain glucose-6-phosphatase (4,5), the presence of another enzyme either of lysosomal or cytoplasmic origin 144 is indicated. Cohn and Hirsch have shown a 2 to 3 fold in- crease in both soluble acid phosphatase and alkaline phos- phatase activities during phagocytosis (6). Since these enzymes could retain some activity in the pH range of the cytoplasm, one or both of these could be responsible for the dephosphorylation of glucose-l-phosphate or glucose- 6-phosphate in PMN during galactose loading. This sugges- tion is further supported by evidence indicating that lyso- somal fragility is increased during galactose toxicity in other tissues (7,8). However, data on the redistribution of acid phosphatase in PMN (Appendix, Table II) do not demonstrate any significant increase in free activity, or extracellular activity, as a result of incubation with galactose. Therefore, dephosphorylation of hexose-phos- phates in PMN may occur as a normal process, although ap— parently wasteful in terms of energy conservation. Cyclic phosphorylation and dephosphorylation of triose—phosphates in PMN (9) concurs with this suggestion. A third mechanism for depleting intracellular ATP by competition between glucose and galactose for cell entry does not occur (Chapter III). Effects of galactose upon bactericidal activity (p.433) were much more pronounced than effects upon phagocytic activity (p. 36) in guinea pig PMN. These observations indicate that galactose predominately affects the former physiological function in these cells which is mainly oxi- dative in character (10,11). An inhibitory action of 145 galactose upon the oxygen—dependent killing mechanism is therefore suggested, and this is supported by data on PMN reduction of nitroblue tetrazolium, formate oxidation, and chemiluminescence as discussed in Chapter IV. Moreover, intracellular galactose and galactitol did not impair oxy- gen consumption (p. 132), hexose monOphosphate shunt activity (pp. 35 & 40), or extracellular peroxide forma— tion (p. 130) in these cells; even though levels of ATP were depressed (p. 86). These apparently conflicting re— sults can be reconciled by considering the versatility of the hexose monophosphate shunt. Less than 2% of glucose carbon transverses this pathway in resting PMN, whereas up to 17% enters the shunt during phagocytosis (5). In the presence of galactose, the proportion of glucose-6-phos- phate entering this pathway could increase to compensate for lower substrate levels. Production of galactose radical or galactitol radical by PMN can explain most of the effects of galactose ob- served upon the oxygen—dependent killing mechanism. These radicals are probably very similar to mannitol radical, which has also been proposed in PMN to account for the protection of leukocytes against suicide during phagocytosis (12). Since this deleterious effect was also prevented by superoxide dis- mutase and catalase the scavenge of hydroxyl radical by mannitol was indicated (12). Photographs of the decay of acridine orange fluores- cence, induced by blue light and oxygen (Appendix, Figure l), 146 demonstrate that superoxide anion is involved in this pro- cess. Pre—incubation of cells with superoxide dismutase prevented loss of some red, lysosome-associated (l3) fluorescence. However, pre-incubation of cells with 30 mM galactose facilitated this decay (data not shown). Since this process likely involves the photochemical production of high superoxide anion fluxes, peroxidation of lysosomal membranes may occur (14) and result in leakage of acridine orange from lysosomes. That galactose enhances this pro- cess suggests involvement of galactose radical in promoting lipid peroxidation, perhaps through reacting with super— oxide (eq. 8, Chapter IV) to form singlet oxygen. Alterna- tively, this could be explained by increased lysosomal fragility, but data in Tables I and II of the Appendix do not agree. 10. 11. 12. 13. 14. 147 References Klant, N., and Schucher, R. (1963) Can. J. Biochem. Physiol. 41, 849-858 Lelior, L.F (1951) Arch. Biochem. Biophys. 33, 186 Kozak, L.P., and Wells, W.W. (1971) J. Neurochem. 18, 2217-2228 Noble E.P., Stjernholm, R.L., and Ljungdahl, L. (1961) Biochem. BiOphys. Acta 4 , 595—595 Stjernholm, R.L., Burns, C.P., and Hohnadel, J.H. (1972) Enzyme 1 , 7-31 Cohn, Z.A., and Hirsch, J.G. (1960) J. Exp. Med. 112, 1015-1022 Blosser, J.C., and Wells, W.W. (1972) J. Neurochem. 12, 1539-1547 Schroeder, H., Lawler, J.R., and Wells, W.W. (1974) J. Nutr. 104, 943-951 Esmann, V., Noble, E.P., and Stjernholm, R.L. (1965) Acta Chem. Scand. 1 , 1672-1676 Klebanoff, S.J. (1975) In The Phagocytic Cell in Host Resistance pp. 45-56 (Bellanti, J.A., and Dayton, D.H., eds.) Raven Press, New York Sbarra, A.J., Selvaraj, R.J., Paul, B.B., and Mitchell, Jr., G.W. (1975) In The Reticuloendotheial System pp. 57—48 (Rebuck, J.W., Berard, 0.w., and Abell, M.R., eds.) Williams & Wilkins, Baltimore Salin, M.L., and McCord, J.H. (1975) J. Clin. Invest. 2Q, 1319-1323 Koenig, H. (1973) In Lysosomes in Biology and Pathology pp. 111-162 (Dingle, J.T., and Fell, H.B., eds.) North-Holland Publishing Co., New York Allison, A.C., Harrington, J.S., and Birbeck, M. (1966) J. Exp. Med. 124, 141 APPENDIX 148 Table I. Analysis of carbohydrates in sera8 Concentration (mM) Carbohydrates Fetal Calfb HumanC Guinea Pigb Chickenb Glucose 7.83 2.55 6.00 12.4 Fructose 5.48 — _ _ Sorbitol 1.42 — _ - myo—Inositol 0.72 — _ _ Galactose - — - _ aAliquots of sera (0.2 ml) were diluted with 5 volumes of triple distilled water and treated with 0.4 ml of 0.3 N Ba(0H)2 and of 0.5% ZnSO4. Supernates were then taken to dryness under N2 in the presence of 40 pgAEmethyl-mannoside and later treated with 0.1 m1 of TMS reagent. Each deriva- tized sample was then analyzed 3 times by gas chromatog- raphy, and the values represent means of these determina- tions. (-) indicates levels less than 0.01 mM. bObtained from Grand Island Biological Company. CObtained from North American Biologicals, Inc. 149 Table II. Acid phosphatase activity associated with guinea pig PMNa Extracellular Intracellular Additions releaseb distributionC Glucose Galactose 6 (mM) (mM) mU[hr[10 cells % Free 5.0 0 0.70 i 0.22 14.0 i 1.90 5.0 30.0 0.72 + 0.11 17.8 + 3.60 aPMN (2.79 x 106/assay) were incubated at 37°C for 8 hr in Krebs—Henseleit bicarbonate solution without phosphate, pH 7.4. Acid phosphatase was determined in 0.1 M acetate buffer pH 5.0, containing 0.2% (w/v) Triton—X-100 and 0.05 M123-glycerol phosphate. Each determination was performed 3 to 5 times. I b 6 Enzyme activity released into cell supernate/hr/IO cells. lmU = 1 pmole POZB/min. CRatio of enzyme activities in 22,000 xg supernate to that in 400 xg supernate of cell homogenates. 150 Table III. Distribution of hexosaminidase and peroxidase activity in guinea pig PMNa Hexosaminidaseb PeroxidaseC % Free % Free Carbohydrate (mM) i S.D. i S.D. Glucose Galactose 5.0 0 42.0 i 7.3 22.1 i 4.0 5.0 30.0 46.3 i 2.2 19.8 i 13.6 30.0 0 37.7 i 5.9 23.9 i 10.8 aPMN (1.7 x 107/assay) were incubated for 2 hours at 37°C in Krebs—Ringer phosphate solution, pH 7.4. Distribution of enzyme activity (% Free) was determined by dividing enzyme activity in the 22,000 xg supernates by that in the 400 xg supernates of cell homogenates. Each value was determined 5 times on cells pooled from 4 guinea pigs. S.D. refers to standard deviation. bHexosaminidase activity was determined in 50 mM citrate buffer, pH 4.3, with 0.1% Triton X-100 and with 5 MM PNP- NAG. CPeroxidase activity was determined in 0.1 M phosphate buf— fer, pH 7.0, with 0.T% Triton X-100 by monitoring the ini- tial velocity of guaiacol (0.31 mM) oxidation in the presence of 0.11 mM hydrogen peroxide. 151 Effect of superoxide dismutase on the decay of acridine orange fluorescence in phagocytes. Cells were incubated with 5 mM glucose before exposure to blue light for 1 min (A) and 13 min (B). Cells were incubated with 5 mM glu— cose plus 0.2 mg/ml superoxide dismutase before exposure to light for l min (C) and 13 min (D). Total incubation time prior to staining was 5 hr at 37°C. Staining was for 30 min at the same temperature with 50 uM acridine orange in the dark. IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 456