44" 44444qu ."x 4 .4" 44 i . ‘ 4 :1} 44. 4 .4 '4 44:1 7 4',“ 4'44. 4' 44 44.4 ,4'4 .4 . 44 IIIII '. . g 444444 "44.4444." ‘4- 4 4 '4‘ ‘ 444 I” 7,444. 4,,4, MM 4 4 . 4 ,4 . 4 ,. '4,444 II 444' ‘"44"‘4'44'4'4 “ 4,44 4 .4444 ‘4'." 4 ’4': - 4 . ' ._" '2- .-’£ - a, -—~~__. ._ w';:_ 1". 44 W M. i- " _ . ‘an' .- IIo . t‘ u an, - f ‘ A . u, on _ , -— . , I I II , 4444444 4 4444.4 4 4. :4. " .. 4.....44 44434.: .4 '4 . . ,- 4444,444444,44 4,4,4, 44 444.. 4444 44 . ' 44444444444444444" » 4,, I,,44 4 4- . .I“I 4444.4." 4444444 4 4,» 4444' . 4', . 444 444,444 ,. 4 . ~ 444 4 ; ,4 I, .‘ III I I III ,II; ,III I, 4,444 ' 44"4444“,:4 435:}. 411;, 4444444444444 “444 L34 9% 44I4I44 I ,,4.4 4% 4 i,4.4 . 1,344.44 {4444, III, 44 ,4, ,,4', 4,4,. .444" ,,4 4444 4444, "44 3,4, 4444444'4",.4,I,44 I44. .44.: I 4473: 4. 444,44, 4,, .4 4444,,I,,,44"' 4444.4,4414144444'4" 4' 44444,,j444, 4.414344 5.4 ,444. .4444.“ I44 4444,4444, 444,4 444,1,4,4444,;,44'44 444%,, , ,, ”44444144, I, III, 4444444444444,“ 44,,44IIII 44444144444444.4044 IIIIIIIIIIIPIIrfiIgfi: I ,4, ,,, 44' I,I,,,44,"" .4444 4,444, 4'44 '4 4 4 , 4,, l4444'Il 4'444'4'44'4’444'44'444 4444444444 .44.."4444 444 " 4444 '4 4‘“ ' 4' 544 "44"44'4" "4 4 .4444'444'4' 44444444 4 444' 444444'4'444 4 4 ' ,. ' 4 44 4' '4', 44 I .44. . .4 x 44"". 4 44444 4'4 44. .4 4 '44 J .1444 t f 44444444444444444 ,4 4 44 " . 4 44‘. ,b4444yfi4fi4444,,‘-I,,I44n4. 5' 1 ‘4444 '4' E ‘4, 44,1444,4444,'444'.,4444,{4% 'I .4, H5; . .4 , 4 , fl,.4444,,,4,,,4,4:‘.,,,, ,4444'4 444544444? .. I, .. 4' ‘Iz’I'4'l-4I I' 4 4444' i 0 . .I ‘2": 4,: h,“ 3‘:th p.355? 4544.. ‘ .l 0'].- - {792414444 4 ,l '444I4I III ”44'" 4,, ”44444414" 4444, 4,, v ' . «,4 I ::é. t;— ._ ~_ _~_ .444. 4,444.44 . 4 44444441444444444 4 4" , '4'44'44U 44444444444444 4444 '4 44444 4,. 4' 444"" 4444444 444,44 I4,,:4 '4I4'I4|,, 444 H44 IIIII L? 44. ,. 44 " 44"" H" '4 444444444 I, 4 44444.44. 44444444444444". 4444444444444”44414444414444.4444‘44444444 ' ll/I/II/II/l/II/Ill/{IllI/ll/I/I/III/llfll/Illllllllllllllll 3 1533A RY ' Efiichig an State Uaherstty j This is to certify that the thesis entitled CLEARANCE AND KILLING 0F CANDIDA ALBICANS IN THE PERFUSED MOUSE LIVER presented by Lee Robert Schwocho has been accepted towards fulfillment of the requirements for M.S. damein M1crob1ology and Public Health was Qflmflf. D. Major fessor Date 3/34/67/ / / 0-7639 Q OVERDUE FINES: L-‘L 25¢ per day per iteu 1 (51’4“ ‘ gamma LIBRARY MATERIALS: Place in book return to remove charge from circulation records vi“. CLEARANCE AND KILLING 0F CANDIDA ALBICANS IN THE PERFUSED MOUSE LIVER by Lee Robert Schwocho A Thesis Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Microbiology and Public Health 1981 ABSTRACT CLEARANCE AND KILLING OF CANDIDA ALBICANS IN THE PERFUSED MOUSE LIVER By Lee Robert Schwocho Hepatic interactions of two 9. albicans isolates with perfused mouse livers were characterized and compared in normal and glucan- treated mice. Normal livers, in the absence of serum, trapped greater than 90% of both isolates and killed greater than 20% of isolate I and approximately 15% of isolate II. Silica treatment abolished killing and decreased trapping suggesting that candidicidal activity of the liver is mediated by Kupffer cells. Phenylbutazone had no effect. Immune serum, but not normal serum, enhanced trapping and killing of isolate I but not isolate II. Liver hypertrophy was evident in mice treated with glucan, but no enhanced candidicidal activity was observed in the absence of humoral factors. Specific immune serum and normal calf serum increased killing of both isolates in glucan stimulated livers suggesting a requirement for serum opsonin in facilitating glucan enhanced killing. Specific immune serum potentiated the greated increase in killing. Glucan treatment in conjunction with immune serum increased killing of isolate I to approximately 40% and isolate II to greater than 33%. D-mannose, but not D-glucose or D-mannitol, impaired trapping of both isolates in livers of normal mice. Together, the data suggest that Lee Robert Schwocho hepatic trapping of Candida albicans involves phagocytic events as well as interactions of the yeasts with surface receptors on sinusoidal cells. Cumulatively, the results support the role for the liver in restricting hematogenous dissemination of g; albicans in the infected host. To Anne for her constant encouragement and unending support ii ACKNOWLEDGMENTS I would like to thank Dr. R. 0. Moon for serving as my graduate chairman. for review of my manuscript and for his concern with my pro- fessional, as well as personal growth. A special thanks is extended to Dr. E. S. Beneke, Dr. A. L. Rogers, and Dr. M. Patterson for serving on my graduate committee and for their dedication to teaching. Because of these educators my stay at Michigan State University has been a reward- ing and enriching experience. Thanks also to Ruth A. Vrable for her technical assistance and capacity to keep the lab running smoothly. To Dr. E. Werner for her sense of humor, for laughing at my jokes when no one else did, and for keeping the coffee pot full. Thanks to Bob, Dace, Ellen and Bryon for making the lab a friendly and usually interesting place to work. Special thanks to Jim for his friendship and assistance throughout my studies. To CG and the gang for providing a refuge, a place to play, and good company to play with. TABLE OF CONTENTS LIST OF TABLES .......................... INTRODUCTION ........................... LITERATURE REVIEW ........................ Candida albicans and candidiasis .............. Pathogenicity and virulence of C. albicans ......... Host defense mechanisms directed against C. albicans . . . . The reticuloendothelial system and Kupffer cells ...... Vascular clearance of C. albicans ............. Liver perfusions: micFobicidai and trapping mechanisms Glucan and phenylbutazone ................. MATERIALS AND METHODS ...................... Animals .......................... Candida albicans ...................... Glucan preparation ..................... Glucan treatment ...................... Chemicals and reagents ................... In situ liver perfusions .................. Statistical analysis .................... RESULTS ............................. Trapping and killing of C. albicans I and II by normal per- fused mouse livers .................... Effects of silica treatment and phenylbutazone on trapping. and killing of C. albicans I and II by perfused mouse livers .......................... Effects of normal calf serum and Candida- -specific bovine immune serum on trapping and killing of.g_. albicans I and II by normal livers .................... Comparison of liver weights from normal and glucan-treated m1ce ........................... Effects of glucan treatment on trapping and killing of g, albicans I and II by perfused mouse livers ........ Effects of normal calf serum and Candida- -specific bovine immune serum on trapping and killing of C. albicans I and II by livers from glucan- —treated animals ......... Effects of mannose, glucose, and mannitol on trapping and killing of C. albicans I and II by normal perfused mouse livers .......................... Page vi 42 Page DISCUSSION ............................ 51 APPENDICES ............................ 57 Appendix A: Effects of g, Parvum and glucan treatments on carbon clearance in mice ................. 57 Appendix 8: Quantitative comparison of the percent of viable g, albicans recovered from liver homogenates and blended liver homogenates ............... . . 59 LITERATURE CITED ......................... 61 LIST OF TABLES Table Page 1 Trapping and killing of C. albicans I and II by normal pere fused mouse livers ...................... 40 2 Effects of silica treatment and phenylbutazone on trapping and killing of C. albicans I and II by perfused mouse livers . . . 41 3 Effects of normal calf serum and Candida specific bovine immune serum on trapping and killing of C. albicans I and II by normal mouse livers .................... 43 4 Comparison of liver weights following glucan administration . 44 5 Trapping and killing of C. albicans I and II by glucan treated mouse livers ......................... 45 6 Effects of normal calf serum and Candida specific bovine immune serum on trapping and killing of C, albicans I and II by glucan treated mouse livers ................ 46 7 Effects of mannose, glucose, and mannitol on trapping and killing of C, albicans I by normal mouse livers ....... 48 8 Effects of mannose, glucose, and mannitol on trapping and killing of g, albicans II by normal mouse livers ....... 49 A1 Effects of C. parvum and glucan treatments on carbon clearance in m1ce ...................... 58 BI Quantitative comparison of the percent of viable C. albicans recovered from liver homogenates and blended liver homogenates 60 vi INTRODUCTION Knowledge of the candidicidal properties of the macrophage is scant. Some investigators have suggested that ingestion of Candida albicans by macrophages may facilitate dissemination of the organism rather than impede infection (88, 93, 150, 179). Much of the recent work on the candidicidal ability of macro- phages has centered on jg_vitro studies with peripheral, peritoneal, and alveolar macrophages, while little attention has focused on the liver Kupffer cell. Sawyer et al. (142), using a rat liver perfusion model demonstrated that normal livers, in the absence of serum, were capable of trapping, but not killing perfused C. albicans. Neither trapping nor killing was improved by the addition of humoral factors, but killing was enhanced by treating rats with Corynebacterium parvum vaccine. Trapping involved both phagocytic and non-phagocytic para- meters. The major objective of this thesis was to evaluate the trapping and candidicidal properties of the perfused mouse liver. Preliminary studies demonstrated that normal livers were capable of killing 9, albicans with great efficiency. The role of the Kupffer cell in mediat- ing killing was established by the use of the macrophage-specific toxin, crystalline silica (54). Silica-poisoned livers, depleted of Kupffer cells, were unable to kill perfused yeast. A second objective was to determine if non-immune and specific-immune serum could enhance hepatic clearance and killing. Additionally, activation of the reticu- loendothelial (RE) system by glucan, a potent macrOphage stimulant (33-40, 126) was attempted to determine whether hepatic tissue could be non-specifically enhanced in its ability to kill cleared Candida. Final studies focused on C, albicans adherence in the perfused liver. Diamond and Krzesicki (31) and Warr (168) have described mannose (and related compounds) impaired adherence of Candida species to neutrophils (PMN) and macrophages in 11339, Additionally, Freidman and Moon (54) and Sawyer et al. (142), have described the trapping of bacteria and yeast in the perfused liver and noted that both microorganisms asso- ciated extensively with endothelial cellscfl’the sinusoids. Conse- quently, preliminary studies were performed using mannose and related compounds in an attempt to block adherence of g, albicans in the liver and impair trapping. LITERATURE REVIEW Candida albicans and candidiasis Candida albicans is frequently an indigenous human inhabitant often colonizing the mucous membrances of the alimentary tract, respiratory tract, and urogenital tract. It appears only transiently on the skin (26, 46). Rarely causing disease in healthy individuals, .9. albicans is an opportunistic pathogen which can cause disease under adverse or abnormal conditions (1, 14). The most common manifestations of infection are superficial lesions of the mucous membranes, espe- cially of the mouth and vagina. It may also manifest as an allergic phenomenon. Less common, but more severe forms of the disease include chronic mucocutaneous and systemic involvement (44, 114, 117). As an opportunistic pathogen C, albicans causes disease in the compromised and debilitated host. A multitude of factors has been identified as contributing to man's susceptibility. Elevated glycogen and glucose concentrations in body secretions, associated with hormonal imbalances, are thought to specifically contribute to the incidence of g, albicans infections in pe0ple with diabetes (76) and during preg- nancy (43). Trauma (106), drug addiction (88), and concomittant bac- terial infection (75) are other, common potential causes underlying C, albicans infections. Among hospitalized patients iatrogenic procedures often result in infection (3, 48). Surgery and broad spectrum antibiotic therapy have been implicated in altering normal flora, thereby allowing C, albicans an access for colonization (48). Additionally, some antibio- tics can inhibit the synthesis of anti-Candida antibodies (19) as well as, the phagocytic and metabolic activity of granulocytes (22, 79). Sulfonamides at pharmacologically active concentrations jg_vjyg_inhibit the myeloperoxidase (MPO)-dependent antimicrobial system of human PMN jg_vitro (79). Chan and Balish (22) have reported that phagocytosis of C, albicans by human PMN is retarded by 1 ug amphotericin B, a drug commonly used to treat systemic Candida infections (102). The use of broad spectrum antibiotics also causes an increase in the number and severity of lesions provoked by g, albicans in humans (146). Corticos- teroid therapy decreases resistance to infection, persumably by inhibit- ing the release of lysosomal enzymes into C, albicans containing phagosomes (48). The two major forms of severe human illness caused by C. albicans are chronic mucocutaneous candidiasis (CMCC) and severe disseminated candidiasis (44). Chronic mucocutaneous candidiasis is a rare disease characterized by persistent Candida infection of skin, hair, nails, mucous membranes and regularly associated with a cell-mediated immune system defect (42, 44, 139). Clinical and experimental evidence sug- gests that: (i) tolerance exists to a specific Candida antigen, (ii) a specific T-lymphocyte defect exists such that there is a basic fail- ure in antigen recognition or mediator production, or (iii) monocyte dysfunction precludes the possibility of an effective immune response. The most common clinical and laboratory manifestations indicative of a cell-mediated immunodeficiency associated with CMCC are: (i) cutaneous anergy to Candida antigens, (ii) reduced, or absence of, lymphocyte transformation in response to Candida antigens, and (iii) reduced, or complete suppression of, production of leukocyte migration inhibitory factor after stimulation with Candida antigens. Because other host defense mechanisms are intact, notably granulocyte function, the infec- tion does not become systemic (44, 139). Pe0ple afflicted with chronic granulomatous disease (CEO) or MPO-deficiency are highly susceptible to the disseminated form of can- didiasis (81, 82, 85). The MPO-HZOZ-halide system of human PMN and monocytes is the predominate means by which these phagocytic cells kill ingested C, albicans (60, 80, 82). The exact mechanism by which the MPO-HZOZ-halide system generates its anti-candidicidal effect is uncer- tain, but MPO and H202 are essential (8, 80, 81). Myeloperoxidase defi- ciency is characterized by a paucity of complete lack of the enzyme, whereas monocytes and PMN from individuals with C60 fail to generate H202 (80-82, 153-155). Lehrer and Cline (82) were the first to demon- strate that phagocytic leukocytes from patients with MPO-deficiency or CGD are capable of phagocytizing C, albicans normally, but cannot kill them. Candida albicans infections complicating neoplastic and myelopro- liferative disease have contributed significantly to the mortality of cancer patients in the past twenty years and the incidence of such infections in these patients continues to rise. Evas et al. (48), in a review of 2,517 autopsy reports, between the years 1960 to 1964, veri- fied 109 demonstrable fungal infections of internal viscera by g, albicans. In a more extensive case review, between the years 1963 to 1975, Myerowitz et al. (114), reported that the incidence of dissemi- nated candidiasis associated with leukemia rose from 1.1 cases/year prior to 1971, to 6.8 cases/year between 1971 and 1975. These investi— gators and others (14, 129) attribute the increased occurrence of dis- seminated candidiasis to prolonged granulocytopenia; induced either by chemotherapeutic agents as part of the treatment regimen or caused by the disease itself. In the severely immunosuppressed individual excess colonization of the gastrointestinal (GI) tract by C. albicans, facilitated by the use of antibiotics, is increasingly becoming a major portal of entry lead- ing to dissemination (152). Injury to the mucosal epithelium of the GI tract, due to direct cytotoxicity of chemotherapeutic agents, direct penetration of the organism, or both results in invasion of submucosal vessels and eventual hematological dissemination throughout the body (113, 114). Severe histopathology and inflammation accompanying disse- minated candidiasis in humans has been observed in the skeletal system (25), joints (111), myocardium (52), epidermis (15), cerebrum (87), meninges (9), eye (75, 103), urinary tract (61), kidneys (24, 114), liver and spleen (113, 114). Myerowitz et al. (114) have suggested that the increasing involvement of the liver and Spleen may be compli- cated by depressed macrophage clearance in these two major RE organs. Klein and Watanakunkron (75) studied the fate of 85 episodes of hospital acquired candidemia in 77 patients and noted that the patients followed one of four clinical courses: (i) spontaneous resolution (42.8%), (ii) development of endophthalmitis following apparent resolution (5.1%), (iii) severe illness requiring appropriate therapy (31.4%), and (iv) no therapy resulting in death from candidemia (20.7%). Pathogenicity and virulence of C. albicans Despite the reports of extensive organ involvement the pathogene- sis of systemic g, albicans infections is poorly understood. Causes which have been specifically proposed as contributing to mortality include embolization (138), toxemia (66), uremia (174), pancreatic damage (179), myocardial damage (116), and pulmonary hypersensitivity (175). Nevertheless, the kidney has generally been assumed to be the major target organ; mortality thus attributed to subsequent renal patho- logy (93, 131, 133). In a recent study of experimental candidiasis in mice, however, Leunk and Moon (86) observed fundamental differences in the pathogenesis of the disease following i.v. administration of dif- ferent doses. A low dose challenge (1.0 x 106) of C. albicans resulted in a non-fatal or slowly progressive renal infection, whereas a high dose challenge (4.5 x 106) invariably resulted in a rapid toxemic-like death. Montes and Nilbour (101) were the first to describe C, albicans as both an intracellular and extracellular parasite. In infected tis- sue it exists as a dimorphic fungus, growing in both yeast and mycelial forms (31, 47, 49, 65, 150, 179). Experimental animal studies have implicated both the yeast (47, 93, 165) and mycelial (65, 179) forms as being the more virulent. Iannini et al. (65) proposed that the enhanced pathogenicity of the mycelial form1was related to its being preferentially trapped in peripheral capillary beds, thereby preventing delivery to phagocytic cells of the liver and spleen. In a more criti- cal comparison of pathogenicity, however, Mardon et al. (93) reported that yeast phase C. albicans provoked a more lethal response in mice, as measured by mortality rates, than comparable numbers of mycelial forms. Sprippoli and Simonetti (157) were able to enhance the virulence of both forms innoculated i.p. in mice, but not i.d. in rabbits by simultaneous administration of tetracycline. The effects of tetracy- cline were dose dependent. Vaughn and Weinberg (165) have correlated virulence of g, albicans with jn_yivg copper concentrations. Copper in physiological concentrations suppressed the filamentation of blasto- spores in yitrg_and potentiated the pathogenesis and growth of mycelial and yeast forms in mice. Evidence has also accumulated which suggests that aberrations in iron metabolism by the host may enhance the viru- lence of g, albicans. Elin and Nolff (45) have reported that high serum iron concentrations facilitates the growth of Candida and injec- tions of iron enhance the lethality of the organism for mice. Iron unsaturated lactoferrin is fungastatic in human serum, but the effect is negated by the addition of iron (19). Although g, albicans is primarily an opportunistic pathogen, it appears to be invasive in both the compromised and healthy host, capa- ble of passing through mucous membrane barriers and gaining access to the vascular compartment. Candida albicans was found to pass unchanged through the GI wall of a healthy volunteer following oral administra- tion, resulting in fungemia and funguria. Fortunately, the yeast was cleared in three hours by the host's defenses (48). IQ. albicans asso- ciated phospholipase activity has been implicated in aiding this inva- sive process, since epithelial cell membranes contain approximately 60% phospholipid (125). Young (179), Stanley and Hurley (150), Mardon et al. (93), and Maisch and Calderone (88) have implicated phagocytic cells of the RE system in aiding the dissemination of C. albicans. These investigators have observed germ tube and pseudomycelial formation by C, albicans following phagocytosis by mouse and rabbit macrOphages. Ingested yeast remained dormant for approximately an hour, after which they began to germinate. By four hours macrophage integrity had been disrupted and macrophages were consumed by dense mycelial growth. Sasada and John- ston (140) have proposed that the ability of C, albicans to survive after being phagocytized related to its ability to limit macrOphage oxidative metabolism. These authors found that phagocytized g, parap- .--r m1: q sillosis stimulated a more vigorous oxidative burst and superoxide (02') release than did phagocytized C. albicans. Macrophage metabolic changes correlated with enhanced killing of Q._parapsillosis. Cutler (28) has pr0posed that the degree of chemotactic activity elicited by g, albicans may have general relevance in regard to the invasiveness of different isolates. Candida albicans induced chemotaxis has been described for the guinea pig (28) and human PMN (126). Human PMN chemotaxis required complement activation. Viable C, albicans or twelve hour culture filtrates were chemotactic for guinea pig PMN by themselves. Cutler (28) ran exhaustive studies on the chemotactic ability of eight different strains of C. albicans and discovered that the only strain without demonstrable chemotactic activity was isolated from a case of severe disseminated candidiasis. The other isolates were from superficial infections or healthy volunteers. Host defense mechanisms directed against C. albicans Delineation of host defense mechanisms directed against 9, albi- cans have relied on extrapolation from clinical observations in humans and laboratory experimentation in animals. A recent review of protec- tive mechanisms against C, albicans has been published by Rogers and 10 Balish (135). Protective roles have been described for an intact humoral response (56, 89, 94, 106), an intact cell-mediated immune response (42, 44, 100, 139), and PMN (2, 31, 32, 60, 82, 133). Phago- cytic cells of the RE system (distinct from thymus-dependent activated macrophages) have been ascribed both a protective role (49, 81, 85, 89, 173) and a detrimental role (already described) in combating C. albicans infections. A Moser and Domer (106) in assessing the role of antibody-mediated immunity during systemic candidiasis in mice selectively eliminated the B-cell arm of immunity by cyclophosphamide (CY) treatment. Following cutaneous immunization with viable C, albicans, CY-treated mice, though retaining an intact T-cell arm of immunity possessed depressed levels of circulating antibodies and were markedly more susceptible to a subse- quent.i.v. challenge than untreated mice. Additionally, CY-treated mice were unable to confine the spread and multiplication of g. albicans from the cutaneous innoculation. Giger et al. (56) reported similar results in an earlier study. Maita et al. (89) have proposed that antibodies are important only as an adjunct to phagocytosis and killing. ‘However, what role if any, antibodies play in conferring protection in humans remains to be determined. Patients with Candida disease typi- cally possess normal or elevated levels of circulating and secretory antibodies against Candida antigens (94, 139). The relationship between thymus-dependent cell-mediated immunity and CMCC has already been discussed, but the role of thymus-dependent cell-mediated immunity against systemic candidiasis remains controver- sial. Miyake et al. (100) have reported that protective immunity against systemic cnadidiasis can be transferred from subcutaneously 11 vaccinated mice to normal recipients with lymphoid cells, but not serum. However, this protective effect appeared to be only partially effective and was not evident until late in infection. Pearsall et al. (122) have reported on the existence of a lymphokine capable of reducing the number of viable g, albicans jn_yjtrg, In contrast, Rogers and Balish (133) have proposed that an intact cell-mediated immune system is not required for defense against disseminated candidiasis and may actually be associated with decreased murine resistance (132). Rogers et al. (132) reported that congenitally athymic nude mice possessed a greater capacity to prevent the growth of i.v. administered C. albicans in the kidney and clear the microbe from the liver than their phenotypically magnum . un- normal litter mates. Furthermore, thymus reconstituted mice were just as susceptible as normal mice to systemic infection. The role of the PMN in combating systemic Candida infections has been suggested by clinical observations that patients with MPO-defi- ciency or CGD are readily susceptible to infection. Lehrer and Cline (82) compared the phagocytic and intracellular killing abilities of PMN from normal individuals and patients suffering from MPO-deficiency and C60. Neutrophils from normal volunteers phagocytized 100% of a 107 dose of g. albicans after ten minutes and killed 29% after one hour. Although phagocytosis of a similar inoculum was complete after ten minutes, PMN from patients with MPO-deficiency or CGD failed to kill the ingested yeast after one hour. Subsequent investigations by others have substantiated these findings (32, 80, 85, 95). The high incidence of disseminated candidiasis in patients suf— fering from neoplastic diseases, rendered neutropenic by the disease or therapy, has provided additional evidence in support of the PMN as a 12 primary defense against systemic candidiasis. Laboratory confirmation has come from Johnson et al. (71). They demonstrated that L1210 leuke- mia cells rendered mice neutropenic, suppressed the inflammatory reac- tion normally elicited by C, albicans, and enhanced the susceptibility of leukemic mice to i.v. challenge. Histological evidence in support of the PMN has been provided by Rogers and Balish (133). These authors compared renal histology in normal and BCG activated mice over a 21 day period, following i.v. challenge with viable C, albicans. They found the predominate cellular infiltrate in both groups of mice to be PMN. Neutrophil accumulation was strikingly evident by day seven and persisted throughout the period of observation. Pearsall and Lagunoff (121) described similar results for a thigh muscle Candida induced lesion in mice. A modest infiltra- tion of other cell types, notably macrophages and a few lymphocytes, preceded the inital PMN response. In assessing the role of the macrophage in compating g, ablicans infections it is necessary, and often difficult, to differentiate between 'inate' macrophages and macrophages which have been 'sensitized' through the T-cell mediated arm of immunity. This is especially true, since experimental animals and humans are routinely exposed to Candida antigens or similar antigens very early in life (135). Presently, con- troversy exists over whether the macrophage can effectively kill C, albicans and what role it may play in containing infection. Arai et al. (2), similar to the results obtained by others (93, 150, 179), reported that rabbit alveolar macrOphages from normal and immunized animals were incapable of killing ingested g, albicans jg_ vitro. Phagocytizing macrophages were eventually destroyed by 13 proliferation and pseudomycelial formation. Maisch and Calderone (88) also reported similar results for rabbit blood monocytes jg_vivo. In contrast, Lehrer et al. (83) reported just the opposite. They found that alveolar macrophages from both normal and Mycobacterium butrycium stimulated rabbits were capable of killing ingested C, albicans effec- tively. Resident macrophages killed 28% of the ingested yeast after one hour incubation and stimulated macrophages killed 33% under similar [ .-'-_- conditions. Additionally, killing was not confined to alveolar macro- phages. Resident and stimulated peritoneal macrophages were also capa- ble of killing ingested yeast, but not as efficiently as alveolar macrophages. Some resident peritoneal macrophages were disrupted by mycelial formation after prolonged incubation. Evron (49) reported that peritoneal macrophages from mice sensitized with viable C. albicans, but not from mice sensitized with heat killed cells in incomplete Freunds adjuvant killed ingested C. albicans and limited to a degree germina- tion and myceliam formation. Killing and inhibition were not 100% effective. Maita et al. (89) used resident, BCG activated, and PHA- induced lymphokine-stimulated mouse peritoneal macrophages to study the candidicidal properties of these cell populations in vitro. Resident macrophages were unable to kill ingested g, albicans in the absence of homologous immune serum. Bacille Calmette Guerin and lymphokine- stimulated macrophages were candidicidal by themselves. Lymphokine- stimulated macrOphages possessed the greatest degree of phagocytic and candidicidal activity. Sasada and Johnston (140) compared the candidi- cidal activity of BCG and LPS elicited mouse peritoneal macrophages to resident peritoneal macrophages and reported results similar to Maita et al. (89). 14 The results of these investigations support the concept that macrophages are capable of killing 9, albicans and point to a possible role for the macrophage in controlling infection. They also suggest that macrophage stimulation by the cell-mediated arm of immunity and possibly antibody may potentiate the process. Lehrer (81) and Leijh et al. (85) have reported on the ability of human periperal blood monocytes to ingest and kill g, albicans. After 60 minutes incubation in yjtrg human monocytes ingested 96% of an ino- culum of g, albicans and killed 50% of the ingested yeast (85). Human monocytes, like PMN, from patients with MPO-deficiency and C60 are impaired in their ability to kill 9, albicans. Phagocytosis is normal, but the yeast are not killed (21, 81). Fundamental to the anti-Candida pr0perties of phagocytic cells are recognition, phagocytosis, and intracellular killing. Recognition is an energy independent event. Phagocytosis and intracellular killing are energy dependent events (20). The energy for phagocytosis is derived primarily from glycolosis. The energy for intracellular kill- ing is derived primarily from oxidative metabolism (153—156). Recognition and phagocytosis of Candida species by granulocytes can be potentiated by specific antibody directed against the cell wall (2, 18, 44, 60, 85, 89). The cell wall of C. albicans is a multilay- ered structure (41, 124) composed of glucans, mannans, N-acetylglucosa- mine, and some glycoproteins (23, 180, 181). The outer layer is composed almost exclusiVely of mannans; the primary antigenic determi- nats of the cell wall being different alpha-1,2-glycosidic linkages of the mannan hexose moeities (96). Accordingly, antibodies induced by intact cells are directed against the mannans of the cell wall. 15 Hasenclever and Mitchell (58) used these agglutinating antibodies to divide C, albicans into two serological groups; group A and group B. Group A contains all the antigenic structures of group 8, plus an addi- tional antigen or antigens. Considerable cross-reactivity also exists among different Candida species. Human 190 is an effective C} albicans opsonin. Normal serum contains substantial titers of 'natural' Candida specific antibodies (44). Diamond et al. (32) have reported that direct interactions between human PMN and C, albicans pseudohyphae occurs in the absence of 1 serum, but serum factors (IgG) facilitate pseudohyphae damage. Arai et al. (2) concluded that the opsonization potential of immune serum was species specific. Rabbit Candida specific-immune serum enhanced the phagocytic indices of rabbit alveolar macrophages for C. albicans, but not guinea pig PMN. Mouse Candida specific-immune serum enhanced phago- cytosis of C, albicans by resident and activated mouse peritoneal macrophages and in contrast to the results of Arai et al. also enhanced the candidicidal potential of resident macrophages (89). Several investigators (50, 82, 85) have reported the necessity for a heat-labile serum factor for maximum phagocytosis of C, albicans by human PMN and monocytes. In the absence of fresh serum phagocytosis was essentially absent. Opsonic activity was abolished by heating at 56°C for 30 minutes, suggesting a role for complement in the process. Serum opsonins have also been reported to facilitate the phagocytosis of other pathogenic fungi besides Candida Species (10, 18, 161). Recent evidence has accumulated for the existence of a carbohy- drate(s) specific surface receptor capable of promoting adherence and phagocytosis of Candida species by direct surface-surface interactions 16 with phagocytic leukocytes. Binding of an organic glycoprotein and synthetic glycoconjugate to murine alveolar macrophages was inhibited by yeast mannan (149). Because the nature of the synthetic glycoconju- gate was known, Stahl (149) prOposed that the surface receptor recog- nized either glucose or mannose moieties. In a later study, Narr (168) demonstrated that resident mouse alveolar macrophages bound and ingested g, kurzgj_jn_yitrg.with a high degree of efficiency in the absence of serum. Binding was inhibited by D-mannose, D-glucosamine, horse-radish peroxidase, and beta-glucouronidase. It was not effected by D-mannitol, D-glucose, D-galactose, or L-fucose. Pretrypsinization of alveolar macrophages also prevented binding of yeast cells. Further evidence in support of a receptor was reported by Diamond and Krzesicki (31). These investigators demonstrated that C. albicans pseudohyphae were capable of adhering to human PMN in the absence of serum. Binding was inhibited by Candida mannans, but not D-mannose, dextran, chitin, Con A, or highly charged amino acids. Pretreatment of pseudohyphae with chymotrypsin or PMN with trypsin also impaired binding. Additionally, UV light induced damage to pseudohyphae promoted the release of a protein-polysaccharide complex which also blocked adherence to human PMN. Narr (168) and Diamond and Krzesicki (31) have postulated that the surface receptor specifically recognized mannose moieties. Slight biochemical or antigenic differences between C, kruggi_and C, albicans pseudohyphae have not been ruled out in explaining the differences in mannose-sensitive adherence between the two. Phagocytic and microbicidal mechanisms of phagocytic leukocytes have been reviewed by Stossel (153-155) and Barboir (8). The primary microbicidal (and candidicidal) mechanism employed by phagocytes is the l7 oxygen-dependent MPO-HZOZ-halide system originally described by Kleban- off (74). Following activation by the appropriate stimulus two cellular events important in the microbicidal process occur: degranulation and initiation of the respiratory burst (8). Degranulation involves the fusion of the phagosome with cytoplasmic granules. The granules contain lytic enzymes and other materials associated in degradation and killing. Granular contents are discharged into the vesicle containing the micro- organism (153-155). Respiratory burst describes a metabolic pathway responsible for generating microbicidal agents, via the reduction of oxygen (8). The metabolic events incorporated in the respiratory burst include augmented oxygen consumption, enhanced hexose-monOphosphate- shunt activity, and increased production of 02' and H202 (8, 156, 164). The exact mechanism(s) whereby the MPO-HZOZ-halide system exerts its microbicidal effect is uncertain, but thought to proceed by one of three mechanisms (8, 80, 81). In the presence of H202 MPO catalyzes the halogenation of the microbial cell wall with death resulting from the loss of integrity of the halogenated surface (74). Cl', 1', and Br" have been used to halogenate different microbes jn_yitrg, but it is believed that Cl' is the physiological substance, since it is the most abundant halide in the cell (8). Lehrer (81) has challenged the vali- dity of this hypothesis, however. Methimazode, isoniazid, and amino- triazode can inhibit the halogenation of C. albicans by normal monocytes without impairing killing. A second mechanism proposes that MPO and H202 catalyze the decarboxylation and deamination of microbial cell wall amino acids, resulting in disruption of the cell surface and death (21). A correlate 18 to this hypothesis is that free amino acids are decarboxylated generat- ing microbicidal free-aldehydes (153-155). The most recent hypothesis proposes that the MPO-HZOZ-halide sys- tem kills by means of singlet oxygen, the reactive species being OH' and/or 02' (8). Presently, this hypothesis appears the most likely, but evidence remains inconclusive. MP0 is present in, and therefore considered, the primary candidi- cidal mechanism of the mouse (140), rat (81), and guinea pig (2) granulocyte but not the rabbit macrophage (83). Lehrer et al. (83) discovered that the candidicidal activity of rabbit macrophages was not impaired by agents inhibitory to the MPO-system of human monocytes, but was inhibited by other agents which were ineffective in those cells. Their conclusion was that the candidicidal mechanism of the rabbit macrOphage was uniquely different from the MPO-HZOZ-halide system of human granulocytes. Human PMN (80) and monocytes (81) in addition to the MPO-system possess a second MPO-independent candidicidal mechanism. This second mechanism is ineffective against C. albicans, but lethal for_§. parapsillosis and g, pseudotropicalis. It is functional in normal granulocytes under anaerobic conditions and MPO-deficient and C00 impaired granulocytes also. Several cationic-like proteins have been isolated from human, rabbit, and guinea pig granulocytes which may account for the nature of this second mechanism (84). The reticuloendothelial system and Kupffer cells Much of the information regarding the candidicidal properties of macrophages has come from_yn vitro studies. Little attention has been focused on the functional role or the candidicidal activity of the 19 intact RE system during systemic candidiasis. Nevertheless, some infor- mation has accumulated suggesting a beneficial role for the RE system in combating deep Candida infections. Rogers and Balish (134) were able to confer protection in germ-free rats, from i.v. challenge with viable C, albicans, by stimulating the RE system with incomplete Freunds adjuvant. Bird and Sheagren (12) reported enhanced RE function in mice systemically infected with C. albicans, but failed to correlate «bfl‘J‘ enhanced function with protection. Trnovec et al. (163) also provoked enhanced RE function in mice by treating with various C. albicans frac- tions and concluded that properties of the RE system played a signifi- cant role in limiting the spread of C. albicans. However, they too failed to prove it. Cells of the RE system originate from mesenchymatous tissue and include fixed tissue macrophages of the spleen, lymph nodes, bone marrow, alveolar macrophages of the lung, microglial cells of the cen- tral nervous system, liver Kupffer cells, and blood monocytes (72, 158, 170). MacrOphages of the RE system are not end cells. Under normal conditions they proliferate with low frequency (98, 166), however, it is generally believed that different tissue macrOphages, at one time or another, are derived from bone marrow promonocytes via the blood mono- cyte (33, 115, 169). MacrOphages of the RE system represent one of the most important short term defenses against invading microorganisms, comprising the major cellular barrier against microorganisms in the bloodstream (29, 99, 151). Liver Kupffer cells represent the largest group of fixed tissue microphages in mammals, and on the basis of functional and numerical superiority play a major role in the physiological events of 20 the RE system (4, 35, 123). Kupffer cells account for 38% of total liver cells and 50% of total RE cells (62). The cytochemistry of Kupffer cells is typical of macrophages. Unlike monocytes, endogenous peroxidase activity is located in the nuclear envelope and endoplasmic reticulum and not cytoplasmic granules (33). Acid phosphatase and other lytic enzymes are contained in lyso- somes (169). Kupffer cells reside in hepatic sinusoids where they are anchored to the fenestrated endothelium by cytoplasmic lamellopodia and filopo- dia (54, 107, 142). They protrude well into the sinusoidal lumen and are covered with numerous invaginations, blebs, and ruffles which con- tribute significantly to the surface area exposed to the blood (107, 141). Such extensive surface morphology aids the Kupffer cell in sca- venging particulate matter from the blood (20, 130). Blood enters the liver via the portal vein and to a lesser extent the hepatic artery. Portal venules branch out from the portal vein and empty into liver sinusoids at the periphery of liver lobules. Blood perculates through the sinusoids, which radiate at right angles from the portal venules; generally running parallel to each other. Sinusoids empty into central venules near the center of the lobules. Arterioles of the hepatic artery empty into sinusoids distal from the sinusoidal-central venule junction. Central venules converge on the central vein, which eventually empties into the inferior vena cava via the hepatic vein. Particulate matter is first deposited in the sinu- soids along the periphery and later towards the center of the lobules (13, 158). Certain cells in the sinusoidal wall exert a sphincter control over the lumen, which acts to regulate blood flow through the 21 sinusoids (13). Altered flow rates influence the uptake of inert and viable particulate matter by the liver (5, 6, 63). Vascular clearance of C. albicans Bloodstream clearance of C, albicans has been characterized by various investigators. Sawyer et al. (142) and Iannini et al. (65) categorized the bloodstream clearance of g. albicans blastospores into two phases: an initial rapid clearance followed by a slower decrement phase. Greater than 90% of the injected yeast were cleared in the first five minutes. The entire inoculum was recovered in RE organs as viable cells thirty minutes after administration. Hepatic recovery accounted for the highest percentage of recovered organisms (142). Iannini et al. (65) and Evans and Mardon (47) studied the vascular clearance of both blastospores and mycelial phase C, albicans. Both .groups reported that blastospores were preferentially sequestered in the liver, but noted different locations of preferential trapping of myce- lial forms. Iannini and his group reported that mycelial forms were preferentially cleared from the bloodstream of rabbits in peripheral capillary beds. Evans and Mardon observed that mycelial forms were pre- ferentially cleared from the blood of mice by the lungs. Whether the differences reflect the use of different experimental animals or differ- ent routes of injection is not known. Baine et al. (7) have reported different tissue localization of yeast phase C, albicans in rabbits, depending on the route of injection. Prior immunization with heat killed cells had no effect on trapping of either form in rabbits (65). Viability of both forms decreased in the liver and the lungs of mice twelve days after injection. Disappearance from the lungs was faster 22 than in the liver (47). Spores, cell walls, and glucan of C, albicans preferentially localized in the liver after i.v. injection into mice (97). Electron micrographs of liver tissue revealed that viable C, albicans were phagocytized by Kupffer cells within fifteen minutes after injection. Phagocytosis was proceeded by step-wise degradation of the cell wall and eventual death of the yeast. Glucan of the cell wall persisted, however, and induced granuloma formation four days after injection. Granulomas were characterized by large numbers of macro- phages and lymphocytes, but few PMN. Liver perfusions: microbicidal and trapping mechanisms The perfused liver provides an excellent opportunity to study factors involved in hepatic clearance and killing of microorganisms. Jeunet et al. (67-70) have used the perfused liver to study phagocyto- sis and the RE blockade. Moon et al. (104) have demonstrated that the perfused liver approximates jfl_yjyg_events and is capable of distin- guishing between microbial trapping and microbial killing functions of the liver. Manwaring and Coe (91) did liver perfusions using rabbits and pneumococci. Pneumococci suspended in Ringers-Locke solution were not removed by normal livers. Media supplemented with 1 to 10% normal rabbit serum had no effect, but 1% immune serum potentiated 100% removal, after multiple passes. The activity in immune serum was heat stable, 60°C for thirty minutes suggesting the removal was mediated by antibody. Manwaring and Fritschen (92) performed similar studies in dogs using E. coli, Staph aureus, and B, anthracis and got similar results, however, some trapping was evident in the absence of humoral 23 factors. Additionally, different bacteria manifested different degrees of retention. In the absence of serum S. coli and Staph adreus were trapped with greater efficiency than S, anthraCis. Nardlaw and Howard (167) and Jeunet et al. (67) reported similar results using a number of different bacteria in rats. S, meletensis and S. typhosa were readily removed by the perfused liver in the absence of specific anti- body or plasma opsonin. Clearance was not improved by specific immune serum. In contrast, S, abortus was removed by the perfused liver only .‘J h-u .J: I‘d-CH in the presence of specific antibody (67). Humoral factors were not necessary for hepatic trapping of S. coli, 3. pyocyana, P, mirabilis, Staph aureus, S, murium, S, cereus, and S, pyogenes. Conversely, two Clostridia species and S, gallimurium suspended in buffered medium passed straight through the liver. Trapping of all bacteria species was enhanced by the addition of normal human serum to the perfusion medium, except for Staph aureus, 9, murium, S, cereus, and S, pyogenes, which was actually reduced (167). In a parallel study, Howard and Nardlaw (63) attributed the opsonic activity of normal serum to speci- fic antibody, complement, and possibly properidin. The opsonic activity of the sera was destroyed by heating (56°C, 30 mins.) or absorption with homologous bacteria, antibody-antigen immune complex, and zymosan. They attributed the removal of bacteria from the perfusion medium to phagocytosis, however Moon et al. (104) have demonstrated that hepatic trapping is not synonomous with phagocytosis. Although trapping does not imply phagocytosis, Friedman and Moon (54) have demonstrated that Kupffer cells are necessary for maximum 10 trapping. Normal nurine livers trapped an average of 63% of a 1 x 10 to 2 x 1010 dose of S, typhimurium suspended in M-199. Trapping of a 24 similar inoculum in Kupffer cell depleted livers was reduced over 30%. From these results they suggested that bacterial trapping is a physical event mediated by surface receptors on endothelial cells or simply due to mechanical restriction of the sinusoids. Mechanical restriction seems unlikely though, since some strains of bacteria, both cocci and bacillus shaped, are capable of passing entirely through the liver with- out being retained (67, 91, 167). None of the studies reported herein have dealt with the relative bactericidal properties of hepatic and humoral factors. Bonventre and Oxman (16) studied the cellular and humoral factors associated with hepatic clearance and killing in rats. The immunological status of the animal had no effect on phagocytosis or killing of Staph aureus, but did effect killing of S, enteritidis. Immune humoral or cellular fac- tors by themselves provoked a limited degree of destruction. Together, they accounted for a 95% reduction in viability of the bacteria. Moon et al. (104) demonstrated that non-immune humoral factors were capable and necessary for killing of S. typhimurium in perfused rat livers. Normal livers killed only 7% of an inoculum suspended in M-199. In the presence of 10% whole rat blood or plasma killing was increased to over 50%. Friedman and Moon (55) further characterized the blood components responsible for hepatic killing of S. typhimurium. Their results showed that viable Kupffer cells and complement, via the alternate pathway, were necessary for killing. They also concluded that specific antibody only enhanced trapping in normal animals. Normal livers in the presence of 5% plasma killed more than 37% of an inoculum of S. typhimurium. Killing of a similar inoculum in livers depleted of Kupffer cells was only 9%. To determine the importance of complement 25 in killing, complement activity was depleted by heating (57° and 50°, 30 mins.), zymosan absorption, chelation with EGTA, and immunoabsorp- tion of C3. All treatments significantly reduced bactericidal activity in the perfused liver. Chelation with EDTA had no effect, which sug- gested a role for the alternate complement pathway, since EDTA specifi- cally impairs the classical pathway. Nhen immune plasma was treated to destroy complement activity bactericidal activity in the liver was reduced, but trapping was increased, emphasizing the role of specific antibody in trapping only. Contradictory to the results of Friedman and Moon, Ruggiero et al. (137) reported a requirement for both classical complement activation and specific antibody in effecting killing of S. 9911 in perfused rat livers of endotoxin tolerant animals. Whether this difference reflects the use of endotoxin tolerant animals or different bacteria is not known. Endotoxin tolerant animals, however, may possess high levels of antibody to S, gglj_which could mask the surface of the bacteria and prevent direct interaction with the alternate complement pathway. Hepatic trapping by the perfused liver is not unique to bacteria, since tumor cells (136) and g, albicans (7, 142) are avidly trapped as well. Mouse liver Kupffer cells manifested a preferential phagocytosis and degradation of different tumor cells in 11.11.9- Adenocarcinoma and melanoma cells were avidly trapped and degraded within 90 minutes of infusion, but lymphosarcoma and leukemia cells were rarely effected. Perfused rabbit livers manifest a strong avidity for g, albicans blastospores (7). Normal rabbit livers trapped an average of 90% of a dose of g, albicans suspended in buffered medium. Addition of 5% nor- mal rabbit serum increased trapping, slightly but significantly, to 26 98%. Trapping in the presence of heated 5% normal rabbit serum was intermediate between the high and low values. Sawyer et al. (142) did 9, albicans liver perfusions in rats. Normal livers cleared greater than 85% of the infused inoculum without any killing. Neither trapping nor killing were enhanced by the addition of 10% homologous whole blood to the perfusion medium. Scanning electron micrographs revealed that S, albicans spores were trapped in liver sinusoids adhering to fenes- trated endothelial cells. Rarely were trapped yeast associated with Kupffer cells. These results differ from those of Meister et al. (97) previously described for whole animals. Enhanced trapping and signifi- cant killing was induced in rats by treatment with g, parygm_vaccine (143). Corynebacterium parvum-treated, perfused livers killed only 5% of an inoculum of g, albicans suspended in M-199, after thirty minutes. Extending the perfusion time to 60 minutes increased killing to over 20%. In the presence of 5% homologous plasma, perfused livers killed greater than 30% of a similar inoculum after 60 minutes. Hepatic trap- ping in g, parygm-treated animals involved both phagocytic and non- phagocytic parameters. Kupffer cells possess complement (C3b) and 19 PC surface recep- tors (64, 144) which may facilitate phagocytic trapping in the presence of humoral factors. In the absence of humoral factors trapping of g, albicans by Kupffer cells may be mediated by the proposed granulocyte, mannose—sensitive surface receptor described by Diamond and Krzesicki (31) and Narr (168). It remains to be determined if non-phagocytic trapping of Q, albicans involves mechanical factors or chemical inter- actions of yeast cell walls with sinusoidal membranes. Day et al. (30) recently reported that hepatic clearance of i.v. administered, iodine- 27 labelled IgMzBSA immune complexes in rats was impaired by pre- or coin- jection of mannan. The report emphasized that hepatic endothelial cell surface receptors responsible for immune clearance may recognize and bind mannose oligosaccharides associated with circulating antibo- dies. These same receptors may mediate non-phagocytic sinusoidal trapping of bacteria and g, albicans as emphasized by Moon et al. (104), Friedman and Moon (54), and Sawyer et al. (142). Glucan and phenylbutazone The RE system can be activated by innumerable different agents including zymosan (127). Riggi and DiLuzio (127) identified glucan as the RE stimulatory agent in zymosan. The glucan of zymosan was first physically characterized by Hassid et al. (59) and found to consist predominately of linear glucopyranose units joined by beta-1,3- glucosidic linkages. The molecular weight of glucan extracted by the method of Hassid et al. is approximately 6,500 daltons. The biological prOperties of glucan have been reviewed by DiLuzio (34, 35). Glucan is a potent stimulant in mice (177), rats (128), humans (90), and even crayfish (147), but not rabbits or dogs (40). Glucan administration in susceptible animals potentiates profound hyperplasia and hypertrophy of RE organs (6, 105, 128), significantly enhances primary and secondary immune responses (105, 177, 37), and increases serum lysozyme levels (77). Glucan administration is pro- ceeded by marked, reversible hypertrOphy and hyperfunction in the liver, lung, and spleen (5, 177) in a time and dose dependent manner (178). Liver weights of mice treated with three daily consecutive i.v. injections of glucan (4 mg/Kg body wt.) increased 11% 24 hours after the last injection and peaked after ten days. Enhanced phagocytosis 28 coincided with increases in liver weight. Carbon clearance was enhanced 24 hours after the last glucan injection and peaked after ten days. Twenty-five days after treatment ceased depressed phagocytic activity associated with reduced liver weights was pronounced. By 30 days liver weights and phagocytic activity returned to normal. The pharmacological effects of glucan are specific for cells of the RE system (39, 128). SulphobromOphthalein clearance from the bloodstream, a function of hepatic parenchymal cells, is not altered following glucan administration (6), though vascular clearance of colloidal carbon (177), RE test lipid emulsion (5), and sheep red blood cells (105) is accelerated. The enhanced vascular clearance associated with glucan administration is due primarily to enhanced Kupffer cell activity (105). The hepatic uptake of foreign RBC by glu- can treated mice, in relation to controls, was increased 140% 30 minutes after injection. After one hour, hepatic activity was still 63% greater than the glucan-treated group. In contrast to enhanced liver uptake, lung removal of foreign RBC in RE hyperfunctional mice decreased 60% at 30 minutes and one hour. Spleen uptake in the glucan- treated group was unaltered from controls at any time period. The hypertrophy and hyperfunction stimulated by glucan treatment reflects the formation of new RE cells and enhanced activity of pre- existing cells, especially Kupffer cells (38, 127, 128). Glucan has no direct effect on circulating opsonin levels, nor are opsonic factors involved in glucan induced alterations of the RE system, though opsonins are required for maximum phagocytosis by isolated RE cells (37). Deimann and Fahimi (33) reported a large influx and differential locali- zation of monocytes and macrophages into rat livers 18 to 48 hours 29 after a single i.v. injection of glucan (30 mg/Kg body wt.). Monocytes preferentially. adhered to the endothelial lining of portal venules, whereas macrophages typically congregated to central venules and adhered. Transmission electron micrographs and cytochemical techniques suggested the existence of transition forms between peripheral blood monocytes and Kupffer cells in the sinusoids. No PMN were evident in any of the liver preps. Glucan administered i.v. is preferentially sequestered in the liver and the most pronounced effects associated with RE organs occur in the liver (5, 35) which may partially explain why enhanced vascular clearance is attributable primarily to increased hepatic uptake. Twenty-four to 48 hours after glucan administration monocytic granulo- mas appear around the periphery of liver lobules. These monocytes are phagocytic and account for a large proportion of particulate uptake. Kupffer cells not only increase in number, but also in appearance. Ribonucleo-protein and endoplasmic reticulum increase in number. Mito- chondria are more numerous and larger. Cyt0plasmic vesicles are also more numerous and many contain clearly discernable glucan particles. Kupffer cells as a whole are also largerand swollen, displaying promi- nent cytoplasmic bulging into sinusoidal lumens. The net effect is to markedly restrict blood flow through the sinusoids (5). Measurement of total hepatic perfusion flow remains normal in glucan-treated animals, but when expressed as perfusion rate per unit of liver mass the flow rate is decreased. Burgaleta et al. (17) have studied the cel- lular effects of glucan on mouse peritoneal macrophages and have reported similar effects. Additionally, glucan treatment increased 3O spreading and adherence to glass, as well as augmenting chemotactic activity by 20 to 50%. Treatment of experimental animals and human volunteers with glu- can has conferred protection against a number of different infectious and cancerous agents. DiLuzio (34) and DiLuzio et al. (36) have reported protection in mice against Shays myelogenous leukemia. Protec- tion in animals has also been reported against ascite sarcoma tumors, L1210 leukemia, and adenosarcoma tumors (119). In clinical trials with humans Mansell et al. (90) have reported necrosis and tumor cell destruction of metastatic lesions in patients with malignant melanoma and pulmonary seated adenocarcinoma. Glucan treatment in animals has conferred protection against experimental infections with Staph aureus (78), Sporotrichum schenkii, Cryptococcus neoformans, Plasmodium berghgji, Mycobacterium leprae (35), and Q. albicans (173). Pretreatment of mice with glucan was necessary to confer protection against systemic g, albicans challenge. Glucan pretreatment enhanced survival time and postponed initial mortality compared to control mice or mice receiving treatment after challenge. Glucan pretreatment also reduced the severity and duration of a subcu- taneously induced g, albicans lesion. Glucan had no effect on experi- mental Toxoplasma gondii infection in mice (35). Glucan treatment rendered rats hyperactive to endotoxin shock and enhanced mortality (27, 162), but reduced tourniquet shock induced mortality (162). Glucan mediated macrophage activation appears to be a non- specific thymus-independent phenomenon (35), however, the exact chemi- cal or physical nature of glucan responsible for activation remains obscure. Fitzpatrick et al. (51) reported that thermal degradation and 31 oxidation of glucan resulted in total loss of RE stimulatory activity. These same authors reported formylation increased and acetylation did not alter the effects of glucan. According to DiLuzio and Riggi (40) sulfation effects the activity of glucan depending on the degree of sulfation (by weight). A low degree of sulfation (0.4%) did not alter the RE stimulatory ability or the degree of liver, lung, and spleen hypertrophy induced by glucan. Eleven percent sulfatidn decreased the hyperphagocytic state compared to native glucan, but was still stimula- tory compared to saline controls. Additionally, pulmonary and liver hypertrophy was not induced by the 11 percent sulfated glucan, but spleen hypertrophy toa lesser extent than native glucan was. DiLuzio et al. (36) contend that the pharmacologically active component of glucan is related to the beta-1,3-glucosidic linked back- bone of the molecule. Minor, beta-1,6-glucosidic linked components have been discovered, but are unable to induce macrophage activation or hypertrophy. Fred and Zaremba (53), however, have reported that the RE response to glucan is more dependent on particle size than on unique chemical structure. Their conclusion is supported by the work of Suzuki et al. (160) who reported that water-soluble glucan had minor effects on macrophage mediated tumor destruction in mice. By compari- son, water-insoluble glucan promoted greater than 88% inhibition and destruction of tumor growth. At variance with the results of Fred and Zaremba and Suzuki et al. is the hypothesis of DiLuzio and Riggi (40) that a specific metabolite of glucan rather than its particulate nature is responsible for RE stimulation. DiLuzio and Riggi reported that water-soluble di- and oligosaccharides of glucan were capable of stimu- lating phagocytic activity without inducing hypertrophy in experimental 32 animals. The results of Suzuki et al. and Riggi and DiLuzio should be compared with caution, however. Lack of tumor destruction by water- soluble glucan-treated macrophages does not preclude the possibility of enhanced phagocytosis. It may be that hyperphagocytosis is inducible by water-soluble glucan, but the full complement of activated macrophage activities requires treatment with particulate glucan. Phenylbutazone (PB), an antiinflammatory drug which inhibits phagocytosis and intracellular killing by different leukocytes (118, 148, 156) is commonly used to evaluate granulocyte-microbe interactions. Leijh et al. (85) demonstrated that 1 mM PB inhibited the intracellular killing of g, albicans by human PMN and blood monocytes, but had no effect on phagocytosis. Lehrer et al. (83) reported that PB had no effect on phagocytosis or intracellular killing of g, albicans by rabbit macrophages and concluded the candidicidal mechanism of rabbit macrophages was different than the MPO-system of human granulocytes. In metabolic studies, Nhitehouse (171) observed that PB uncoupled oxida- tive phosphorylation in isolated rat mitochondria without impairing electron transport. Strauss et al. (156) noted that the drug inhibited 14 and 14C-formate in both resting and phagocy- oxidation of glucose-l-C tizing PMN, suggesting an effect on the hexose-monophophate-shunt. The study also reported that PB inhibited glucose-6-phosphate and 6- phosphogluconate dehydrogenase activity. Kjosen et al. (73) verified that PB inhibits intracellular killing primarily by blocking the hexose-monophosphate-shunt. Odegaard et al. (118) have mused that PB may also block intracellular killing by preventing fusion of phagosomes with lysosomal granules. MATERIALS AND METHODS Animals Male and female CD-1 mice were obtained from the Carworth Farms Division of Charles River Laboratories; Portage Michigan and housed in {- our laboratory. Animals were supplied with standard lab chow and water . ag_libitum. Both male and female mice, 25 to 30 g were used without regard to sex. L L Candida albicans Two different isolates of g, albicans were generously supplied by Drs. Everett Beneke and Alvin Rogers of Michigan State University. For purposes of clarification they have been designated simply as isolate I and isolate II. Isolate I was a fecal isolate from a woman experienc- ing recurrent vaginitis. Isolate II was originally from Hasenclever's lab and known to be serotype 8. Identification was confirmed by assi- milation of glucose, maltose, and sucrose, but not lactose or cello- biose, germ tube formation in fetal calf serum, and chlamydospore production on corn meal agar plus 1% TWeen 80. Stock cultures of Candida were maintained on Sabouraouds dextrose agar (Difco, Detroit, Michigan) at 4°C and trasnferred to fresh agar slants every two weeks. Yeast used for perfusions were grown in 100 ml tryptic soy broth (Difco, Detroit, Michigan) plus 4% dextrose (Fisher Scientific Co., Fair Lawn, N.J.) for 18 to 20 hr at 37°C with constant stirring. Cells were harvested by centrifugation at 850 x g for 10 min 33 34 and washed three times in 0.85% sterile saline. After the third wash, cells were suspended in 50 ml sterile saline and blended in a Haring blender. Haemocytometer counts were done and the yeast adjusted to 107 cells/ml in sterile M-199 (GIBCO, Grand Island, N.Y.) for use in liver perfusions. Glucangpreparation Glucan was prepared by the method of Hassid et al. (59). Basi- cally, bakers yeast (Sacchromyces cervisae, MSU Stores, Annheiser Busch, St. Louis, Mo.) was digested in 2L of 3% NaOH in a warm water bath for 4 hr then allowed to stand at room temperature overnight. The superna- tant was decanted and 2L of fresh 3% NaOH was added to the residue. The mixture was boiled in a water bath for 2 hr then allowed to cool at room temperature overnight. The supernatant was decanted and the residue acidified with approximately 800 ml concentrated HCl. An addi- tional 2L of 3% HCl was added to the residue and heated again for several hours on a warm‘water bath. The supernatant was decanted, the residue washed several times with distilled water, and finally washed with boiling distilled water. After washing, the residue was centri- fuged, decanted and mixed with 1L of alcohol. After standing at room temperature for several days the residue was resuspended in fresh alco- hol, centrifuged, washed with ether, and dried under vacuum. The final product (glucan) was collected and stored at room temperature. In preparation for injection, glucan was suspended in saline (10 mg/ml) and sonicated (MSE 100 watt Ultrasonic Disintegrator, the Dann Co., Cleveland, Ohio) at 22,000 cycles/sec for 30 minutes. After sonication the glucan was sterilized by autoclaving and stored at 4°C. 35 The biological activity of glucan was confirmed by carbon clearance studies (Appendix A). Glucan treatment Two different regimens of glucan treatment were employed. Regimen one consisted of a single i.v. injection of 0.5 mg glucan 2 to 4 days prior to perfusions. Regimen two consisted of the same injection schedule, but waiting 7 to 8 days before perfusions. Chemicals and reagents Phenylbutazone (10 mM) (PB, Sigma Chemical Co., Columbus, Ohio) dissolved in 95% ethanol was added to M-199 to yield a final concentra- tion of 1 mM or 5 mM (pH adjusted to 7.3 by addition of 1N Na0H). The solution was filter sterilized (Millipore Corp., Bedford, Mass., pore size 0.22 um) and stored at 4°C. For perfusion studies, livers were infused with 1 ml of an appro- priate concentration of PB, washed for 20 min with M-199 and infused with yeast. For silica studies, Dorenturp crystalline silica (0012), particle size 5 um or less, was supplied by Dr. Ing M. Reisner, Steinkohlenberg- bauereiw, 43 Essen-Kray, Frillendolfer Strabe 351, w. Germany. Silica prepared by autoclaving in powder form was suspended in sterile saline at a concentration of 20 mg/ml. Prior to injection the suspension was sonicated (Bronsonic Ultrasonic Cleansor, no. 8220, Branson Instruments Co., Sketon, Conn.) and a total of 10 mg were administered i.v. over a three day period. Mice received 3 mg on days one and two and 4 mg on day three. Perfusions were done on day four. 36 For carbohydrate studies mannose and glucose (Sigma Chemical Co., St. Louis, M0.) were dissolved in deionized distilled water at a con- centration of 0.5 g/ml (50% wt./vol.), filter sterilized, and added to sterile M-199 to a 1% or 5% final concentration. Mannitol (Sigma Chemical Co., St. Louis, Mo.) was dissolved at a concentration of 0.25 g/ml (25% wt./vol.), filter sterilized, and added to M-199 to a 1% final concentration. All solutions were stored at 4°C. Experimentally, a cell suspension containing 107 g, albicans/ml was made in M-199 plus the appropriate carbohydrate and stored in an ice bath immediately prior to use. Livers were washed with 30 ml M-199 and immediately prior to infusion of yeast, perfusion of M-199 plus the appropriate carbohydrate was begun. For studies using serum, normal calf serum (NCS, Flow Laborator- ies, Rockville, Md.) was added to a concentration of 5% in M-199, filter sterilized, and stored at -20°C. Bovine immune serum (BIS) specific fOr S, albicans I was obtained as a gift from Mr. James Veselenak. Imnune serum was added to NCS to a 20% concentration. This mixture was added to M-199 to yield a final 1% concentration of immune serum, filter sterilized, and stored at -20°C prior to use. Experiments with serum were performed by the same methods used for carbohydrates. Bovine immune serum and NCS were titered against each 9, albicans isolate using a tube agglutination technique. Normal calf serum was prediluted 1/10 and serial 2-fold dilutions were made in saline. Bovine immune serum was prediluted 1/1000 and serially 2-fold diluted in saline. S, albicans was killed (1% formaldehyde, 60°C, 1 hr.), 7 adjusted to 2 x 10 cells/ml by haemocytometer count, and added to 7 diluted serum to yield 10 cells/test tube. Qualitative pour plates of 37 the killed yeast were done to assure none remained viable. Agglutina- tion titers were read after 24 hr incubation in a 37°C water bath. Normal calf serum showed no agglutination activity against either isolate. Bovine immune serum had an agglutination titer of 1:8000 to Q, albicans I and 1:2000 to g, albicans II. In situ liver perfusions Procedures for liver perfusions have been described elsewhere (104). Briefly, a midline abdominal incision was made and skin and viscera reflected to expose the portal vein and inferior vena cava. Ligatures were placed beneath the portal vein and inferior vena cava above the renal vein. The portal vein was cannulated and secured. The inferior vena cava was snipped below the ligature and perfusion of M-199 begun. Next, a medial midline thoracic incision was made and the rib cage reflected to expose the heart and superior vena cava. The left auricle was snipped and an efferent cannula inserted into the incision, through the superior vena cava, and secured by ligature. The inferior vena cava was tied off above the renal vein and 30 ml of M-199 perfused through the liver to flush it of blood. After washing, 1 ml of 107 S, albicans was slowly and steadily perfused through the liver. Perfusions were done at room temperature and lasted approximately 30 min until 50 ml of effluent were collected. After perfusions were complete the liver was removed, homogenized (Tri-R-Stir-R homogenizer, Model no. S63C, Tri-R Instruments, Rockville Center, N.Y.) in 5 ml saline and adjusted to a final volume of 10 ml. Quantitative pour plates of the liver homogenate were performed to determine the number of yeast trapped in the liver. The effluent 38 was blended and quantitative pour plates were carried out to determine the number of yeast that passed through the liver. Quantitative pour plates of the original inoculum were used to determine the total number of yeast perfused. Recovery of yeast was expressed as a percentage of the total number perfused (100%). Those yeast not recovered in the liver homoge- nate or perfusate (total recovery) were presumed killed in the liver (Appendix 8). They were determined by subtracting the percent total recovery from the percent perfused. The percent trapped represents those yeast not recovered in the effluent and is determined by sub- tracting the percent recovered in the effluent from the percent per- fused. Statistical analysis Data was evaluated by the White rank order test (172). RESULTS Trapping and killing of C. albicans I and II by normal perfused mouse livers Normal livers trapped an average of 93.5% of a 107 dose of p, albicans I on a single pass (Table 1). An average of 70.6% of the yeast were recovered in the liver and 6.5% in the perfusate for a total recovery of 77.1%. It is presumed that the 22.9% of the yeast not recovered were killed by the liver. Normal livers trapped an average of 98.7% of a 107 dose of Q, albicans II on a single pass (Table 1). An average of 83.6% of the yeast were recovered in the liver and 1.2% in the perfusate for a total recovery of 85.9%. The liver killed 15.1% of the yeast. The data indicates there is a slight, but significant dif- ference in the behavior of these two isolates in normal livers. Effects of silica treatment and phenylbutazone on trapping and killing of C. albicans I and II by perfused mouse livers Depleting livers of Kupffer cells, by silica poisoning, produced marked effects on trapping and killing of both 9, albicans I and II. Total recovery of both isolates was greater than 100% indicating no yeast were killed (Table 2). The most striking effects occurred with Q, albicans 1. Recovery from the liver decreased 31.5% and increased in the perfusate 57.3%. Total trapping decreased 54.4%. In all cases the differences are significant. 39 40 Table 1. Trapping and killing of S, albicans I and II by normal perfused mouse livers.a Recovery % Experimental C. albicans I C. albicans II Liver 70.6 :_4.6b 83.6 1 8.9 Perfusate 6.5 :_4.6b 1.2 :_1.8 Total 77.1 :_4.6° 84.9 :_9.2 Killing 22.9 14.5b 15.1 :_9.2 Trapping 93.5 14.6b 98.7 :_1.9 aMean :_standard deviation of at least twelve separate experimental determinations. bP = .001 GP = .01 The results with p, albicans 11, though not as striking, are also significant. The precent recovery from the liver was similar in normal and silica-treated animals, but recovery in the perfusate from silica- treated mice increased 12.4%, an amount approximately equal to killing in normal animals. Trapping in this group decreased 9.7%. Phenylbutazone had no effect on trapping or killing of p, albicans I by perfused livers (Table 2). Total recovery in the presence of 1 mM PB averaged 89.9%. Slightly more than 20% of the yeast were killed. In the presence of 5 mM PB total recovery averaged 95.5% and 22.2% of the yeast were killed. The results are similar to normal values. Incubation of Q, albicans with P8 and perfused through the perfusion apparatus in the absence of livers had no effect on yeast viability (data not shown). 41 .pmmmx mo cowmzmcma o» cowca mco~mu=apzcmca mo cowpmcucmucou mumwcqognam mgu do Lowcg .vowgma amt omega 8 Lm>o .>.w umgmumwcescm .mumpwm mcwppmumzcu as oH we .mcowumcwEmemu qucmewgqum mpmgmgmm xwm pmmmp an mo ao.o am Hoo.o a u PE H sue: ummzmgmn mgm>w40 .mcowmamcma on Peace 8 gap: umpamaa 882:8 cowumw>mu ugmucwum.fl :mmzm m.m H 4.8 m; H :3. 3. H 9: ma H 9% me: .+. Nam 8.8 H 98 8.3%.; --- ~.m H.H.mH o.e.fl.~.- m.m.fl.e.o~ --- c.e.H.m.- accp_w¥ .82 H :2 2 M 3m 3, M Q: m; H 8.2 a; H 32 8.4 H m: :33 88.4.“ m.m~ m.H.H_N.H H.m.w m.e m.m.u ~.oH um.Hfi.H_w.mm 8.8.“ m.o aaamscaaa mangm mammam mammaa wgumam $dfiufi8 94H92 22: anacaaaa-aUL_cm pasaoz ama :5 m uma :5 H aamaaaaa-auPpwm Pasta: Paacaepaaaxm HH acaucnpa .8 H meaaaapa .8 x1xuw>oumm m.mcm>w_ «macs ummzwcma an H“ new H memUPQFm am.mo m:w_wa can mcwaamgp co mconuzapzcmga use pcmsummgp mowpwm mo muumcmm .N mpnmh 42 Effects of normal calf serum and Candida-specific bovine immune serum on trappipg and killingpof C. albicans I and II by normal livers Both Candida isolates were perfused through normal livers in the presence of 5% NCS or 1% BIS. Normal calf serum had no effect on trap- ping or killing by normal livers perfused with either isolate (Table 3). One percent BIS significantly enhanced trapping and killing of p, albicans I perfused through normal livers, even though the actual differences are slight (Table 3). Bovine immune serum increased trap- ping by 5.3% and killing by 8%. Only 1.2% of the dose was recovered in the perfusate. The increase in killing was not artifactual due to immune aggregation of the yeast, since when yeast were incubated in the presence of BIS and perfused in the absence of livers an average of 105.7% of a 107 dose was recovered (data not shown). Bovine immune serum had no effect on trapping or killing of Q. albicans II by normal livers, even though the immune serum did show considerable cross-reactivity (Table 3). Comparison of liver weights from normal andpglucan-treated mice Glucan treatment produced a marked hypertrophy of mouse livers (Table 4). Two to four days following treatment liver weights increased 45% over controls. By seven to eight days livers weighed 58% more than controls. Increased liver weights are significantly greater than controls. 43 .mgm>wp Peace: .m> Ho. 8 m .mcm>w_ Peace: .m> Hoe. am .mcowpmcwscwumu Fmpcmswgmaxm mumgmamm m>8m mo copumw>wc ccmvcmum H.=mmzv .AmmHuz PE mm mzpa moz P5 e .mHm —E Hv Ezgmm mazes? mcw>on xau .Ammfinz FE mm mapa muz FE my Ezgmm upmo meLo: fimn .mcowumcwEmemv Fmpcwewcmaxm mumgmamm x88 pmmmp 88 mo covumw>mc ucmucmum.fl cmmzm momma 88HH88 fiufig %8H888 RTHQS 88H888 23%: emuefl 88+888 Wanna héumdm Weuia 88H88~ 85:; 88H~48 88H888 wmumém La8H888 88H8aa 88H8ga 28h 88H88 afiumg QSHNA aéoumg Qauma 9§H98 afistg Wmufig 88H88~ QQHQQ 88H8g8 wafifi 88H88K 22: 8888 a882 Pascoz 8.88H8 n882 Fasaoz Pa8caewaamx8 88 88882828 .8 , H 8:88P8F8 .8 N xum>oumm 8.8Lm>8_ mmsoe Peace: ma HH 8:8 H mcmurnpm am we mcwppwx new m:_aamcp :o 53:88 8:355? mcw>on 88888888 newncmu 8:8 Eagmm wpmu Peace: mo mpummLM .m m_nm» 44 Table 4. Comparison of liver weights following glucan administration. Liver weightsa (grams) b Normal Glucan treated Glucan treatedC d d 1A4102 2C9105 227id5 aMeanistandard deviation of at least six separate experimental deter- minations. b0.5 mg glucan i.v., single injection, liver weights determined 2-4 days after treatment. c0.5 mg glucan i.v., single injection, liver weights determined 7-8 days after treatment. dp = .001 Effects ofpglucan treatment on trapping and killing of C. albicans I and II bypperfused mouse livers Livers from glucan-treated mice showed no enhancement in killing of Q, albicans I either two to four days or seven to eight days post treatment, or p, albicans II at seven to eight days (Table 5). Trap- ping in glucan-treated animals was also similar to normal animals. Effects of normal calf serum and Candida-specific bovine immune serum on trapping and killing of C. albicans I and II byglivers from glucan-treated animals Trapping of p, albicans I perfused with NCS through glucan- treated livers averaged 93.8% (Table 6). Total recovery declined to 70.2% suggesting a 6.9% increase in killing (22.9% vs. 29.8%; P = 0.05). Trapping of p, albicans II perfused with NCS through glucan- treated livers averaged 99.8% (Table 6), however, total recovery fell to 76.4% and killing increased by 8.5% (15.1% vs. 23.6%; P = 0.01). 45 .8:owu8cwe:mum8 888:858888xm 88888888 m>8m mo :88888>88 8888:888.H :8828 .u:8288m:u 88888 8888 8:888 op :m>88 8:88 8:888:8888 .:o8uumw:8 888:88 ..>.8 :88888 as 8.88 .u:m&p88:8 88888 8888 8:88 88 83» 8:88 8:888:8888 .:888888:8 888:88 ..>._ :88388 as 8.88 .8:owu8:8588888 888:858888xm 88888888 x88 88888 88 mo :88888>m8 8888:888.H :8828 8.8 H 8.88 8.8 H 8.88 8.8 .11. 8.88 8.8 H 8.88 8.8 .8 8.88 85888.: 8888.88 8.88 8.88 8.8.8 8.88 8.88.8.8 8.88 8.88 88:5. 8888.88 8.88 8.88 8.88 8.88 88.88.88 8.8.3.88 8388 8.8.88.8 8.8 8.8.8 8.8 8 8.8 8.8888 8.8.88.8 88883.88 8.8 8... 8.88 8.8 8. 8.88 8.8 .8 8.88 8.8 H 8.88 8.8 .8 8.88 88>: 888 8:8588888 885882 8.888 #:8588888 88 #:8588888 885882 888:858888xm 88 88888888 .8 8 88888888 .8 8 >gm>oumm 8.8:8>8— 88:82 88888:» :88888 as 88 8:8 8 8:888888 am 88 8:88888 8:8 8:888888 .8 88888 46 .8:8>88 885:8: .8> mo. 88 a .8:8>88 885:8: .8> 8o. 8 .8:o888:8e:8888 888:85888ax8 88888888 x88 88888 88 88 :88888>88 8888:888 8.:882 .8:8su88:8 88888 8888 8:888 op :8>88 8:88 8:888:8888 .:88888n:8 888:88 ..>.8 :88888 as 8.88 8.8 .8 8.88 8.8 8 8.88 8.8 8 8.88 88.8 8 8.88 8.8 8 8.88 8.8 .8 8.88 888888.: 88.8 8 8.88 88.8 8 8.88 8.8 8 8.88 88.88 8 8.88 88.8 8 8.88 8.8 .8 8.88 85:88. 88.8 8 8.88 88.8 8 8.88 8.8 8 8.88 88.88 .8 8.88 88.8 .8 8.88 8.8 8 8.88 88.88 88.8 8 8.8 88.8 .8 8.8 88.8 .8 8.8 88.8 .8 8.8 8.8 8 8.8 8.8 .8 8.8 88888888 88888.88 8.88 8.88 8.88 8.88 88.88 .8 8.88 88888.88 8.8.8 8.88 88>: 888 882 885882 888 882 888882 888882888888 88 88888888 .8 8 88888888 .8 8 888>ou8m 8.8:8>88 88:85 8888888 :88888 88 88 8:8 8 8:888888 8m mo 8:88888 8:8 8:888888 :o 5:888 8:8258 8:8>on 88888888 8888:88 8:8 58:88 8888 885:8: mo 8888888 .8 8888» 47 Serum alone was not responsible for the increased killing. When either isolate was incubated with NCS and perfused in the absence of livers total recovery of a 107 dose averaged 109.8% and 118.7% for p, albicans I and II respectively (data not shown). Glucan-treated livers perfused with BIS trapped 99.1% of p. albicans I and 99.7% of Q, albicans II after a single pass. The increase in trapping is significant for isolate I, but not for isolate II (probably because of the already high percentage of S, albicans II trapped by normal livers (Table 6). Comparing normal livers in the absence of serum to livers from glucan-treated animals perfused with BIS recovery of Q. albicans I from the liver declined 12.3% and recovery from the perfusate declined 5.6% suggesting that killing increased 17.9% (Table 6). In all cases the observed differences are significant. The results for Q. albicans II were similar. Compared to normal livers perfused with M-199, total recovery fell to 65.7% and killing increased 19.2% (Table 6). Recovery from the liver decreased 18.2% and from the perfusate 0.9%. All dif- ferences are significant. Effects of mannose, glucose, and mannitol on trappinggand killing of C. albicans I and II by normalpperfused mouse livers Glucose and mannitol had no effect on trapping of either p, albicans I or 11 (Tables 7 and 8). In contrast, livers perfused with 1% mannose trapped only 79.2% of Q, albicans I, a significant decrease of 14.3% (Table 7). In the presence of 5% mannose the percent trapped declined 41.1% (Table 7). This represents an additional decrease of 48 .888>88 88838888 888::85 88 .8> mo. .888>88 88588: .8> A88 Hoe. 88 .8:8888:8588888 888:858888x8 88888888 x88 88888 88 88 :88888>88 8888:888.H :8828 8.8.8 8.88 8.8 8.8.88 888.888 8.88 88.8.8 8.88 8.8.8 8.88 88888888 988€8 #38888 888888 #8898 888888 85:; 9T89: 887898 888888 888888 888888 :88 8.8.8 8.8 8.8 8.8.88 8.88.88.8.8.88 88.8.8 8.88 8.8.8.8.8 888888888 8.8 88.88 8.888 8.88 8.8888 8 8.88 88.8 8 8.88 8.8 8 8.88 88>: 8888::8sno 88888~mro 888::821o 888::8suo 882882 888:82888mxm 88 88 88 88 8 xu8>o88m 88588: 88 8 8:888888 .u 88 8:88888 8:8 8:888888 8.888>88 88:85 :8 8888::85 8:8 .8888888 .888::8s 88 8888888 .8 88888 49 .8888885 88 888 8888882 88 8883888 8828 .888>88 88588: .8> mo. u 88 .88888888588888 8888858888x8 88888888 x88 88888 88 8o 888888>88 88888888.H c8828 8.8 8 8.88 8.8 8 8.88 88.8 8 8.88 88.8 8 8.88 8.8 8 8.88 88.88828 8888.88 8.88888 8.88 88.88 8.88888 8.88888 85:3. 8.8 8 8.88 8.8 8 8.88 8.88 8 8.88 8.8 8 8.88 8.8 8 8.88 88888 8.8 8 8.8 8.8 8 8.8 88 .8 8 8.8 88.8 8 8.3 8.8 8 8.8 388.3888 8.8 8 8.88 8.8 8 8.88 88.88 8 8.88 88.: 8 8.88 8.8 8 8.88 88>: 8888::8s-8 8888:8mr8 8888:88E-8 _ 888::8E-8 882882 8888858888xm 88 88 88 88 8.xu8>888m 88288: 88 88 88888888 .w.8o 8.888>88 88:85 8888888 888 88888888 :8 88888885 888 .888888m .8888885 88 8888888 .w 88888 50 26.7% compared to livers perfused with 1% mannose. Both comparisons are significantly differnet, however, total recovery and killing are the same for mannose perfused and M-199 perfused livers. Declines in trapping were reflected in a decrease in the percent of yeast recovered in the liver and an increased recovery in the perfusate. Recovery from livers perfused with 1% mannose, compared to M-199, declined 17.5% and increased in the perfusate 14.9%. With 5% mannose recovery from the liver fell 36.8%, compared to M-199, and 19.3% compared to 1% mannose. Recovery from the perfusate increased 41.2% and 26.3% compared to M-199 and 1% mannose respectively. A similar effect was observed with Q. albicans II (Table 8), except that perfusion with 5% mannose did not have the same pronounced effects as was observed with isolate I. In the presence of 1% mannose normal livers trapped 13.4% less of a dose of Q. albicans 11. With 5% mannose a reduction of only 6.9% was observed, compared to M-199, but an increase of 6.5% was observed compared to 1% mannose. Recovery from livers perfused with 1% mannose declined 8.6% and from the perfusate increased 13.5%. These values are significantly different, although there was no difference in the percent total recovery or killed. With 5% mannose recovery from the liver fell 11.1%. Recovery from the per- fusate increased 7.1%, but decreased 6.4% compared to M-199 and 1% mannose respectively. The differences between 5% mannose and M-199, but not those between 5% mannose and 1% mannose are significant. DISCUSSION While macrOphages of the RE system represent the major, short term defense against invading microorganisms, their ability to contain 9, albicans infection remains controversial. Evas et al. (48) and Myerowitz et al. (114) have reported increasing hepatic involvement in GI seeded cases of disseminated candidiasis. Liver Kupffer cells represent the largest group of fixed tissue macrophages in mammels (4, 35, 123), and therefore, could be postulated to play a major role in restricting hematogenous spread from the GI tract. There is general agreement in the literature that the liver is the major organ involved in bloodstream clearance of Q, albicans (7, 47, 65, 114, 142). The perfused liver model provides an Opportunity to study hepatic- g, albicans interactions. Sawyer et al. (142) have shown that normal rat livers avidly trapped Q. albicans, but no killing occurred even in the presence of homologous plasma or whole blood. In contrast, when 9, albicans was perfused through normal mouse livers with only M-199 substantial killing of both isolates occurred (Table 1). Additionally, slight but significant differences in trapping and killing were observed. g, albicans I was less avidly trapped, but more efficiently killed than 9, albicans II. Sasada and Johnson (140) recognized dif- ferential killing between 9, albicans and g, parapsillosis by isolated mouse peritoneal macrophages. Enhanced g, albicans survivability correlated with its ability to limit macrophage oxidative metabolism. 51 52 Perhaps different 9, albicans strains possess different abilities to limit macrOphage oxidative metabolism and survive phagocytosis. Killing of Q. albicans, presumably by Kupffer cells, is in agreement with the report of Lehrer et al. (83) which emphasized the candidicidal activity of isolated rabbit macrophages. Resident alveolar macrOphages killed 28% and resident peritoneal macrophages 15% of a similar inoculum of Q, albicans used in this study. Crystalline silica is a macrophage specific toxin (54). Friedman and Moon (54) used the compound to deplete mouse livers of Kupffer cells and subsequently reduced trapping and killing of S, typhimurium in per- fused livers. Sawyer (143) used silica to inhibit killing of g, albicans in macrophage depleted livers of Q, ggryumytreated rats. In agreement with these results, silica treatment abolished killing and reduced trapping of g. albicans in perfused mouse livers (Table 2). Silica mediated abrogation of killing highlights two concepts critical to these perfusion studies. First, it strongly supports the assumption that non-recoverable yeast are killed by the liver. Second, it empha- sizes a role for the Kupffer cell in mediating killing. It could be postulated that hepatic killing was mediated by blood-borne PMN. The fact that silica abolished killing and that livers were thoroughly washed free of blood prior to infusion of yeast argues strongly against this possibility. In vitro studies with isolated Kupffer cells would further clarify if these cells possess candidicidal activity. Sawyer (143) demonstrated that PB completely abrogates killing of g, albicans by perfused rat livers. Killing in the perfused mouse liver was not inhibited by PB (Table 2). Kjosen et al. (73) demon- strated that PB inhibits killing through an effect on the MPO-system. 53 Lehrer et al. (83) could not block killing of g, albicans by rabbit macrophages with addition of PB to incubating cells and concluded the candidicidal activity of rabbit macrophages did not rely on the MPO- system. The same conclusion may be true for mouse Kupffer cells. Additional work using other agents (cyanide, azide, sulfadiazone) inhibitory to the MPO-system is needed, however. Specific immune serum, but not normal serum enhanced trapping and killing of Q, albicans I by normal livers (Table 3). Actual differ- ences, though slight, are significant. Neither NCS nor BIS had an effect on hepatic interactions with g, albicans II (Table 3). The results for isolate I are similar to the reports of Maita et al. (89) and Bonventre and Oxman (16) which emphasized enhanced candidicidal (89) and bactericidal (16) activity of macrophages in the presence of immune serum. It can be postulated that increased trapping in the presence of immune serum is facilitated by enhanced adherence of opso- nized yeast to FC receptors on Kupffer cell surfaces (108-110). Perfu- sions with immune serum in livers treated with pronase or mercaptho- ethanol to cleave receptors and silica-poisoned livers depleted of Kupffer cells would help to delineate the mechanism. Why BIS had no effect on killing of g, albicans II by normal livers is puzzling, especially since immune serum exhibited a strong degree of cross-reactivity and increased the candidicidal activity of glucan-treated livers (Table 6). A single dose of glucan enhanced carbon clearance (Appendix A) and increased liver weights (Table 4). These results indicate that a single dose of glucan can stimulate at least some of the effects typi- cally associated with multiple doses of the drug (5, 6, 105, 127, 54 177). Enhanced candidicidal activity was not evident in livers from glucan-treated animals in the absence of serum (Table 5). Increased killing was demonstrated in glucan-treated livers in the presence of serum (Table 6). The addition of either NCS or BIS to the perfusion medium potentiated killing of both isolates. Bovine immune serum pro- duced the greatest increase in killing. Together, the results suggest that serum is required for manifestation of glucan enhanced killing of g; albicans by perfused livers. Sawyer (143) reported similar results accentuating a requirement for plasma opsonin for manifestation of increased candidicidal activity of g, paryum_stimulated rat livers. Two points critical to an explanation of glucan enhanced hepatic killing are necessary. Glucan administration is proceeded by a massive influx of mononuclear cells into the liver (33). It has been observed that this influx is largely responsible for the hypertrophy and hyper- function associated with glucan treatment (38, 127, 128), however, on an individual cellular basis, mononuclear phagocytes isolated from glucan-treated livers manifest a decreased phagocytic potential (38). Additionally, although serum opsonins are not involved in glucan induced alterations of the RE system per se, they are required for _ maximum phagocytosis by isolated RE cells (37, 38, 105). It can be envisaged that the absence of enhanced g, albicans killing by glucan- treated livers in the absence of serum reflects an inability of new phagocytic cells to phagocytize the yeast. It should be noted that an hepatic PMN influx has never been described following glucan adminis— tration, therefore subsequent killing should not be attributed to them. Contrary to the reports of Sawyer et al. (142) that hepatic trap- ping of g, albicans by normal rat livers involves predominately 55 non—phagocytic parameters, the high degree of killing in normal mouse livers (Table 1) and the reduction in trapping caused by silica treat- ment (Table 2) suggests that trapping in mice includes phagocytosis by Kupffer cells. This is consistent with the report of Meister et al. (97) which noted that viable g, albicans was avidly phagocytized and subsequently degraded by mouse Kupffer cells following injection into whole animals. Kupffer cell phagocytosis in perfused livers has also been described for cancer cells (136) and foreign RBC (112). Trapping does not appear to be exclusive of non-phagocytic events, however. Consistent with previous reports (54, 143) Kupffer cells were necessary for maximum trapping, but some trapping was still evident in Kupffer cell depleted livers. Trapping, but not killing of Q. albicans was also impaired by perfusion with D-mannose, but not D-glucose or D-mannitol (Tables 7 and 8). Previous reports have described the existence of a mannose sensi- tive receptor which facilitates the adherence of different Candida species to different granulocytes jg_yitrg, Binding was impaired by yeast mannan (33) and mannose (168). If the adherence of g, albicans to mouse Kupffer cells is mediated by a similar receptor it would seem likely that killing in the presence of mannose would be reduced. However, it is premature to reject the idea of such a receptor. More conclusive results could probably be obtained by perfusion studies with mannan. Sawyer et al. (142) have described the trapping of g, albicans in normal rat livers. Yeast became 'log jammed' in sinusoidal spaces and appeared to be adhering to endothelial cells of the sinusoidal wall. Another possible explanation for the mannose sensitive trapping of g, 56 albicans reported here is that there are endothelial cell surface receptors which promote adherence through interactions with the yeast surface. Evidence recently reported by Day et al. (30) support this concept. They observed that hepatic clearance of IgMzBSA immune com- plexes in rat livers was impaired by pre- or coinjection of mannan. There are great disparities in the degree of trapping between 9, albicans I and II in silica-treated livers (Table 2) and livers perfused with mannose (Tables 7 and 8). Further investigation is needed to explain this phenomenon. It may be that receptors, if present, have different affinities for the surface of the different isolates. Physi- cal differences between the two surfaces may also present steric hin- drances. Especially puzzling is why silica treatment reduced trapping of C, albicans I to a much greater extent than C, albicans II. Silica treatment reduced trapping and increased the percent recovery from the perfusate of isolate II by an amount approximately equal to normal killing. This is not true for isolate I. If, in addition to chemical and phagocytic interactions, phsycial restriction is involved in trap- ping, morphological features of g, albicans II may restrict its passage through the liver and account for a large proportion of trapping. Morphological differences between the two isolates suggests this may be true. Under the light microsc0pe C, albicans I consisted primarily of discrete, individual, elliptical cells. Individual colony forming units of C. albicans II were composed of elongated, multi-budding units and therefore larger. APPENDICES APPENDIX A EFFECTS OF C, PARVUM AND GLUCAN TREATMENTS 0N CARBON CLEARANCE IN MICE Carbon clearance studies were performed in Harlan ICR female mice weighing 20 to 25 g and done by the method of Biozzi et al. (11). Clearance rates were compared between normal, 9, parvumftreated, and glucan-treated animals. 9, parxgm_treatment consisted of a single i.v. injection of 0.35 mg g, paryum_vaccine (Burroughs Wellcome Co., Research Triangle, N.C.) ten days before carbon clearance studies. Glucan treatment consisted of a single i.v. injection of 0.5 mg glucan two days before carbon clearance studies. Carbon clearance was done as follows: 5 mg colloidal carbon (Pelikan carbon suspension C/11/143/1, Gunther Wagner, Hanover, Germany) was injected i.v. Blood samples were drawn from the retro-orbital plexus after 2 and 15 min intervals and 0.05 ml samples were lysed in 4 ml of 0.1% Na2003. The concentration of carbon was determined photometrically using an Hitachi Perkin Elmer spectrophotometer with tungsten light at wavelength 650 nm. The equa- tion, K (phagocytic index)=(logCl-logcz)/(t2-t1), where C represents the blood colloidal carbon and t time, was used to determine the phago— cytic index. To calculate the biological half-life (t%)’ the equation ty=0.301/K was used. 2 In normal mice K equals 0.053 which is equivalent to a half-life of 6.32 min (Table 1). In g, parvum treated mice K equals 0.12 which 57 58 is equivalent to a half-life of 2.76 min. In glucan treated mice, K equals 0.14 which is equivalent to a half-life of 2.26 min. Both the phagocytic index (K) and the biological half-life (t%) for g, parvum: treated and glucan-treated mice were significantly different compared to normal mice suggesting non-specific stimulation of the reticuloendo- thelial system. Table A1. Effects of g, parvum and glucan treatments on carbon clearance in mice. Phagocytic Biological Treatment Index Half-life (mins.) P Normal 0.053 6.32 -- _(_:_. m 0.120 2.76 P=.05 Glucan 0.140 2.26 P=.05 aMean value from at least six separate experimental determinations. APPENDIX B QUANTITATIVE COMPARISON OF THE PERCENT OF VIABLE C, ALBICANS RECOVERED FROM LIVER HOMOGENATES AND BLENDED LIVER HOMOGENATES A major criticism of perfusion experiments is the presumption that non-recoverable organisms are killed, i.e. it is difficult to prove killing without a corpse. To help verify that the percent of C. albicans recovered from the liver homogenate actually reflects true values and the percent of Candida presumed killed is real and not arti- factual due to aggregation of the yeast an additional control was per- formed. Normal CD-1 mice were perfused with both isolates of Candida. After homogenization the total liver homogenate was diluted 1:10 in 90 ml saline and blended (approximately 30 sec) in a Waring blender. Quantitative pour plates of the blended-homogenate were compared to values obtained by homogenization of livers only (Table 1). The results obtained from this procedure indicate that homogenization of livers was sufficient to disrupt any microaggregates of yeast, if present, and supports the presumption that killing is real and not artifactual. There is no significant difference in the calculated values for any parameter between the two methods. At first glance, there would appear to be marked differences in the percentages of C. albicans II recovered between the two methods. However, the discrepancy can be explained by the wide range of values derived from different animals. 59 60 Table Bl. Quantitative comparison of the percent of viable C, albicans recovered from liver homogenates and blended liver homogenates.a Recovery§% C. albicans I C. albicans II Homogenize Homogenize & Homogenize Homogenize & Experimental only blend only; blend Liver 70.6 :_4.6 77.3 i 4.3 83.6 :_8.9 90.8 1_4.4 Perfusate 6.5 i 4.6 3.3 :_0.9 1.2 :_1.9 0.7 :_1.4 Total 77.1 :_4.6 80.5 :_4.7 84.9 :_9.2 91.5 :_4.6 Killing 22.9 :_4.6 19.5 :_4.7 15.1 i_9.2 8.5 :_4.6 Trapping 93.5 :_4.6 96.7 :_0.9 98.7 :_1.9 98.8 :_1.4 aMean :_standard deviation of at least six separate experimental determinations. LITERATURE CITED IO. 11. LITERATURE CITED Ahearn, D. G. 1978. Medically important yeast. Ann. Rev. Microbiol. .32: 59—68. Arai, T., Y. Mikami, K. Yokoyama. 1977. Phagocytosis of C. albicans by rabbit alveolar macrophages and guinea pig neutro- phils. Sabouraudia. 15; 171-177. Arnon, R. G., and R. Ehrlich. 1977. Systemic candidiasis follow- ing candiac surgery: An improved outlook. South. Med. J. 19; 585-587. Aschoff, L. 1924. Retriculoendothelial system. pp. 1-33. In; Lectures in pathology. Harper and Row. New York. Ashworth, C. T., N. R. DiLuzio, and S. J. Riggi. 1963. A morpho- logical study of the effects of RE stimulation upon hepatic removal of minute particles from the blood of rats. Exp. Mol. Pathol. (Suppl.). 1; 83-103. Assidao, C. B., J. P. Filkins, and J. J. Smith. 1964. Metabolic and surface factors governing phagocytosis in the perfused rat liver. J. RES. 1; 393-404. Baine, W. B., M. G. Koeing, and J. S. Goodman. 1974. Clearance of C. albicans from the bloodstream of rabbits. Infect. Immun. _1_o_ :“14—‘7—20- 14 5. Barboir, B. M. 1978. Oxygen-dependent microbial killing by phagocytes. NEJM. 228; 659-666. Bayer, A. S., J. E. Edwards, Jr., J. S. Seidel, and J. B. Guze. 1976. Candida meningitis. Reports of severe cases and review of the English language literature. Medicine (Balt.). .55: 477-486. Beaman, L., and C. A. Holmberg. 1980. In vitro response of alveolar macrophages to infection with—Coccidiodes immitis. Infect. Immun. 28; 594-600. Biozzi, G., B. Benacerraf, and B. N. Halpern. 1953. Quantitative study of the granulopectic activity of the reticuloendothelial system. II. A study of the kinectics of the granulopectic activity of the RES in relation to the dose of carbon injected: Relationships between the weight of the organs and their acti- vity. Br. J. Exp. Pathol. 34; 441-457. 61 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 62 Bird, 0. C., and J. N. Sheagren. 1970. Evaluation of RES phagocy- tic activity during systemic C. albicans infection in mice. Proc. Soc. Exp. Biol. Med. .133: 34-37. Block, E. H., and R. S. McCluskey. 1977. Biodynamics of phagocy- tosis: An analysis of the dynamics of phagocytosis in the liver by in vivo microsc0py. pp. 21-32. In; E. Wisse and D. L. KnoEE (eds.). Kupffer cells and other liver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. Bodey, G. P. 1966. Fungal infections complicating acute leukemia. J. Chronic Dis. 12; 667-687. Bodey, G. P., and M. Luna. 1974. Skin lesions associated with disseminated candidiasis. JAMA. 222; 1466-1468. Bonventre, P. F., and E. Oxman. 1965. Phagocytosis and intracel- lular disposition of viable bacteria by the isolated perfused rat liver. J. RES. _2: 313-325. Burgaleta, C., M. C. Territo, S. G. Quan, and D. W. Golde. 1978. Glucan activated macrophages: Functional characteristics and surface morphology. J. RES. gg; 195-204. Calich, L., G. Vera, T. L. Kipnis, M. Mariano, C. F. Neto and W. Dias DaSilva. 1979. The activation of the complement system by Paracoccidiodes brasiliensis jg_vitro: Its opsonic effect and possible significance for an in vivo model of infection. Clin. Immunol. Immunopathol. .12: 20-30. Caroline, L., F. Rosner, and P. J. Koznin. 1969. Elevated serum iron, low unbound transferrin and candidiasis in adult leukemia. Blood. 34; 441-451. Carr, I. 1977. The Kupffer cell--an overview. pp. 355-364. In; E. Wisse and D. L. Knook (eds.). Kupffer cells and other liver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. Cech, P., H. Stadler, J. J. Widman, A. Rohner, and P. A. Meischer. 1979. Leukocyte myeloperoxidase deficiency and diabetes melli- tus associated with g. albicans liver abscess. Am. J. Med. 66; 149-153. Chan, C. K., and E. Balish. 1978. Inhibition of granulocyte phagocytosis of C. albicans by amphotericin B. Canad. J. Microbiol. 24: 363-364. Chattaway, F. W., S. Shenolikar, J. O'Reilly, and A. J. E. Barlow. 1976. Changes in the cell surface of dimorphic forms of C. albicans by treatment with hydrolytic enzymes. J. Gen. Micro- biol. .95: 335-347. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 63 Chatterjee, S. N., M. Fiala, J. Weiner, J. A. Stewart, B. Stocey, and N. Warner. 1978. Primary cytomegalovirus and opportunistic infections. Incidence in renal transplant recipients. JAMA. 249; 2446-2449. Chimel, H., M. H. Grieco, and R. Zickel. 1973. Candida osteomye- litis: Report of a case. Am J. Med. Sci. 69; 299-304. Conant, N. F., D. T. Smith, R. D. Baker, and J. L. Callaway. 1971. Candidiasis. pp. 325-364. Manual of clinical mycology. 3rd. ed. W. B. Saunders Co. Philadelphia, Penn. Crafton, C. G., and N. R. DiLuzio. 1969. Influence of altered RE function on vascular clearance and tissue distribution of Salmonella enteritidis endotoxin. Proc. Soc. Exp. Biol. Med. 115; 596-601. Cutler, J. E. 1977. Chemotactic factors produced by C. albicans. Infect. Immun. 18: 568-573. David, J. R. 1975. MacrOphage activation by lymphocyte mediators. Red. Proc. ‘34; 1730-1736. Day, J. F., R. W. Thornburg, S. R. Thorpe, and J. W. Burgess. 1980. Carbohydrate mediated clearance of AbzAg complexes from the circulation: The role of high mannose oligosaccharides in hepatic uptake of IgMzAg complexes. J. Biol. Chem. 225; 2360-2365. Diamond, R. D., and R. Krzesicki. 1978. Mechanisms of attachment of neutrophils to g, albicans pseudohyphae in the absence of serum and subsequent damage to pseudohyphae by microbicial pro- cesses of neurophils jg_vitro. J. Clin. Invest. .61: 360-369. Diamond, R. D., R. Krzesicki, and J. Wellington. 1978. Damage to pseudohyphal forms of g. albicans by neutrOphils in absence of serum jg_vitro. J. Clin. Invest. 61; 349-359. Deimann, W. and H. D. Fahimi. 1979. The appearance of transition forms between monocytes and Kupffer cells in the liver of rats treated with glucan: A cytochemical and ultrastructural study. J. Exp. Med. 149; 883-895. DiLuzio, N. R. 1975. Pharmacology of the RES-Accent on glucan. pp. 412-421. In; S. M. Reichard, M. R. Escobar, and H. Fried- man (eds.). The RES in health and disease: Adv. Exp. Med. Biol. vol. 73A. Plenum Press. New York. 1977. Influence of glucan on macrophage structure and function: Relation to inhibition of hepatic metastases. pp. 397-406. In: E. Wisse and D. L. Knook (eds.). Kuppfer cells and other TTver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 64 DiLuzio, N. R., R. McNamee, E. Jones, S. Lassoff, W. Sear and E. 0. Hoffmann. 1976. Inhibition of growth and dissemination of Shay myelgenous leukemic tumor in rats by glucan and glucan activated macrophages. pp. 397-413. In: H. Friedman, M. R. Escobar, and S. M. Reichard (eds.). THE RES in health and disease: Adv. Exp. Med. Biol. vol. 738. Plenum Press. New York. DiLuzio, N. R. and H. S. Morrow. 1971. Comparative behavior of soluble and particulate antigens and inert colloids in RE sti- mulated and depressed mice. J. RES. 9; 273-287. DiLuzio, N. R., J. C. Pisano, and T. M. Saba. 1970. Evaluation of the mechanism of glucan-induced stimulation of RES. J. RES. 7: 731-742. _ DiLuzio, N. R. and S. J. Riggi. 1964. The relative participation of hepatic parenchymal and Kupffer cells in the metabolism of chylomicrons. J. RES. 1; 248-263. . 1970. The effects of laminarin, sulfated glucan, and oligosaccharides of glucan on RE activity. J. RES. .8: 465-473. Djaczenko, W. and A. Cassone. 1972. Visualization of new ultra- structural components in the cell wall of C. albicans with fixation containing TAPO. J. Cell Biol. 52; 186-190. Djawari, D., 0. P. Hornstein, J. Gross, and W. Meinhoff. 1977. Defects of phagocytosis and intracellular killing of C, albicans by granulocytes in patients with familiar and non-familiar chronic mucocutaneous candidiasis. Arch. Dermatol. Res. 269: 159-161. Drake, T. E. and H. I. Maibach. 1973. Candida and candidiasis. I. Cultural indications, epidemiology, and pathogenesis. Post- grad. Med. .53: 83-88. Edwards, J. E., R. I. Lehrer, E. R. Stiehm, T. J. Fischer, and L. S. Young. 1978. Severe candidal infections: Clinical perspectives, immune defense mechanisms, and current concepts of therapy. Ann. Intern. Med. .89: 91-106. Elin, R. and S. Wolff. 1973. Effect of pH and iron concentration on growth of Candida albicans in human serum. J. Inf. Dis. .127: 705-708. Emmons, C. W., C. H. Binford, J. P. Utz, and K. J. Kwon-Chung. 1973. Candidiasis. pp._325-364. Manual of clinical mycology. 3rd ed. W. B. Saunders Co., Philadelphia, Penn. Evans, 2. A. and D. N. Mardon. 1977. Organ localization in mice challenged with atypical Candida albicans strain and pseudohy- phal variant. Proc. Soc. Exp. Biol. Med. .155: 234-238. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 65 Evas, P., M. J. Goldstein, and P. Sherlock. 1972. Candida infec- tion of the gastrointestinal tract. Medicine. 51; 367-379. Evron, R. 1980. In vitro phagocytosis of Candida albicans by peritoneal mouse macrophages. Infect. Immun. 28; 963-971. Ferrante, A. and Y. H. Thong. 1979. Requirement of heat-labile opsonin for maximal phagocytosis of g, albicans. Sabouraudia. 17: 293-297. Fitzpatrick, F. W., L. J. Haynes, N. J. Silver and F. J. DiCarlo. 1964. Effects of glucan derivatives upon phagocytosis by mice. J. RES. 1; 423-428. Franklin, W. G., A. 8. Simon, and T. M. Sodeman. 1976. Candida myocarditis without valvulitis. Am. J. Cardiol. 38; 924-928. Fred, R. K., T. M. Zaremba, and J. G. Harris. 1967. The tissue distribution of intraveneously administered 14C-glucan. J. RES. 4; 437. (Abstract). Friedman, R. L. and R. J. Moon. 1977. Hepatic clearance of Salmonella typhimurium in silica-treated mice. Infect. Immun. 16; 1005-1012. 1980. Role of Kupffer cells, complement, and specific antibody in the bacteriacidal activities of perfused livers. Infect. Immun. 22; 152-157. Giger, D. K., J. E. Domer, and J. T. McQuitty, Jr. 1978. Experi- mental murine candidiasis: Pathological and immune responses to cutaneous inoculation with C. albicans. Infect. Immun. .19: 499-509. Halpern, B. H., A. R. Perrat, G. Biozzi, C. Stiffel, D. Mouton, J. C. Morard, Y. Bouthilier, and C. Decreusfond. 1963. Stimu- lation de 'activite phagocytaire du systeme reticuloendothelial provoquee par Corynebacterium parvum. J. RES. 1; 77-96. Hasenclever, H. F. and W. 0. Mitchell. 1961. Antigenic studies of Candida. I. Observations of two antigenic groups in Candida albicans. J. Bacteriol. .82; 570-573. Hassid, W. Z., M. A. Joslyn, and R. M. McCready. 1941. The mole- cular constitution of an insoluble polysaccharide from yeast, Sacchromyces cerevisiae. J. Am. Chem. Soc. .63: 295-298. Hilger, A. E. and D. L. Danley. 1980. Alteration of polymorpho- nuclear leukocyte activity by viable Candida albicans. Infect. Immun. .27: 714-720. Holt, R. J. and A. Azima. 1978. The emergence of C. albicans resistant to miconazole during the treatment of urinary tract candidosis. Infection. 6; 198—199. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 66 Howard, J. G. 1970. The origin and immunological significance of Kupffer cells. pp. 178-199. In; R. van Furth (ed.). Mononu- clear phagocyte. F. A. Davies Co., Philadelphia, Penn. Howard, J. G. and A. C. Wardlaw. 1958. The Opsonic effect of nor- mal serum on the uptake of bacteria by the RES. Perfusion studies with isolated rat liver. Immunol. I; 338-352. Huber, H., S. 0. Douglas, and H. H. Fudenberg. 1969. The IgG receptor: An immunological marker for the characterization of mononuclear cells. Immunol. II; 7-21. Iannini, P. B., G. D. Arai, and F. M. LaForce. 1977. Vascular clearance of blastospore and pseudomycelial phase Candida albicans. Sabouraudia. I5; 201-205. Iawata, K., K. Uchida, K. Hamajima, and A. Kawamura. 1974. Role of canditoxin in experimental Candida infection. Jpn. J. Med. Sci. Biol. ‘21; 130-133. Jeunet, F. S., W. A. Cain, and R. A. Good. 1968. Differential recognition of Brucella organisms by Kupffer cells: Studies with isolated perfused livers. Proc. Soc. Exp. Med. .Igg: 187-190. 1969. Reticuloendothelial function in the isolated per- fused liver. III. Phagocytosis of Salmonella typhosa and Brucella melitensis and the blockage of7the RES. J. RES. 6; 391-410. Jeunet, F. S. and R. A. Good. 1967. Reticuloendothelial function in the isolated perfused liver. I. Study of rates of clear- ance, role of a plasma factor, and the nature of the RE blockade. J. RES. 4; 351-369. 1969. Reticuloendothelial function in the isolated perfused liver. II. Phagocytosis of heat aggregated bovine serum albumin. Demonstration of two components in the blockade of the RES. J. RES. 6; 94-107. Johnson, J. A., B. H. S. Lau, R. L. Nutter, J. M. Slater, and C. E. Winter. 1978. Effect of L1210 leukemia on the susceptibility of mice to C. albicans infection in mice. Infect. Immun. I9; 146-151. Karnovsky, M. L. D. Drath, and L. Lazdins. 1975. Biochemical aspects of the function of the RES. In; S. M. Reichard, M. R. Escobar, and H. Friedman (eds.). The RES in health and disease. Adv. Exp. Med. Biol. vol. 73A. Plenum Press. New York. Kjosen, 8., H. H. Bassol, and C. 0. Solberg. 1976. Influence of phenylbutazone on leukocyte glucose metabolism and function. J. RES. ‘20: 447-455. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 67 Klebanoff, S. J. 1967. Iodination of bacteria: A bacteriacidal mechanism. J. Exp. Med. I26; 1063-1078. Klein, J. J. and C. Watanakunakron. 1979. Hospital acquired fun- gemia. Its natural course and clinical significance. Am. J. Med. 62; 51-58. Knight, L. and J. Fletcher. 1970. Growth of Candida albicans in saliva: Stimulation of glucose associated with antibiotics, corticosteroids and diabetes mellitus. J. Infect. Dis. I26; 371-377. Kokoshis, P. L. and N. R. DiLuzio. 1979. Serum lysozyme: an index of macrophage function. J. RES. .26: 85-100. Kokoshis, P. L., D. L. Williams, J. A. Cook, and N. R. DiLuzio. 1978. Increased resistance to Staph aureus infection and enhancement of serum lysozyme activity by glucan. Science. I66: 1340-1342. Lehrer, R. I. 1971. Inhibition by sulfanamides of the candidici- dal activity of human neutrophils. J. Clin. Invest. 66; 2498-2505. 1972. Functional aspects of a second mechanism of can- didicidal activity of human neutrophils. J. Clin. Invest. 6I; 2566-2572. . 1975. The fungacidal mechanisms of human monocytes. 1. Evidence of myeloperoxidase-linked and myeloperoxidase- independent candidicidal mechanisms. J. Clin. Invest. 66; 338-346. Lehrer, R. I. and M. J. Cline. 1969. Interactions of 6; albicans with human leukocytes and serum. J. Bacteriol. 66; 996-1004. Lehrer, R. I., L. G. Ferrari, J. Patterson-Delfield, and T. Sorrell. 1980. Fungacidal activity of rabbit alveolar and peritoneal macrophages against 6; albicans. Infect. Immun. 26; 1001-1008. Lehrer, R. 1., K. Ladra, and R. Hake. 1975. Non-oxidative funga- cidal mechanisms of mammalian granulocytes: demonstration of components with candidicidal activity in human, rabbit, and guinea pig leukocytes. Infect. Immun. ‘II: 1226-1234. Leijh, P. C. J., M. T. van den Barselaar, and R. van Furth. 1977. Kinetics of phagocytosis and intracellular killing of C. albicans by human granulocytes and monocytes. Infect.‘Immun. W -318. * Leunk, R. D. and R. J. Moon. 1979. Physiological and metabolic alterations accompanying systemic candidiasis in mice. Infect. Immun. 26; 1035-1041. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 68 Louria, D. B., D. P. Stiffel, and B. Bennett. 1962. Disseminated monoliasis in the adult. Medicine. II; 307-337. Maisch, P. and R. A. Calderone. 1980. Adherence of 6; albicans to a fibrin-platelet matrix formed In vitro. Infect. Immun. 21; 650-656. Maita, R, K. R. Kamar, and L. N. Mohaptra. 1980. Candidicidal activity of mouse macrophages In vitro. Infect. Immun. 26; 477-482. Mansell, P. W. A., H. Ichinose, R. J. Reed, E. T. Kremetz, R. McNamee, and N. R. DiLuzio. 1975. Macrophage-mediated destruc- tion of human malignant cells In_vivo. J. Nat. Can. Instit. 66; 571-580. Manwaring, W. H. and H. C. Coe. 1916. Endothelial Opsonins. J. Immunol. I; 401-408. Manwaring, W. H. and W. Fritschen. 1923. Study of microbe-tissue affinity by perfusion methods. J. Immunol. .6; 83-89. Mardon, D. N., J. L. Gunn, and E. Robinette, Jr. 1975. Variation in the lethal response in mice to yeast-like and pseudomycelial forms of 6; albicans. Canad. J. Microbiol. 2I; 1681-1687. Mathur, S., J. M. Goust, E. 0. Horger, III, and H. H. Fudenberg. 1977. Immunoglobulin E anti-Candida antibodies and candidiasis. Infect. Immun. .I6: 257-259. Meer, J. W., M. van Der, P. C. J. Leijh, M. van den Barselaar, and R. van Furth. 1978. Functions of phagocytic cells in chronic mucocutaneous candidiasis. Br. Med. J. I; 147-148. Meister, B., B. Heymer, H. Schafer, and 0. Haferkamp. 1977. Role of Candida albicans in granulomatous tissue reactions. 1. In_ vitro degradation of 6; albicans and immunospecificity of split products. J. Infect. Dis. ‘I66: 235-242. 1977. Role of Candida albicans in granulomatous tissue reactions. II. In vivo degradation of 6. albicans in hepatic macrophages of middl J. Infect. Dis. .I66: 235-242. Meuret, G. 1975. Origin, ontogeny, and kinetics of mononuclear phagocytes. pp. 71-81. In; S. M. Reichard, M. R. Escobar, and H. Friedman. (eds.). The reticuloendothelial system in health and disease: Adv. Exp. Med. Biol. vol. 73A. Plenum Press. New York. Miles, A. J. 1952. In: B. N. Halpern. (ed.). Physiopathology of the RES. pp. 108-203. Charles C. Thomas. Springfield, Ill. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 69 Miyake, T. K., K. Takeya, K. Nomoto, and J. Muroka. 1977. Cellu- lar elements in the resistance to Candida infection in mice. 1. Contribution of T-lymphocyte and phagocytes at various stages of infection. Microbiol. Immunol. 2I; 703-725. Montes, L. F. and W. H. Wilbour. 1968. Ultrastructural features of host-parasite relationships in oral candidiasis. J. Bacteriol. .66: 1349-1356. Montgomerie, J. 2. and J. E. Edwards, Jr. 1975. Syngenism of amphotericin B and 5-flurocytosine for Candida species. J. Infec. Dis. I62; 82-86. 1978. Association of infection due to C. albicans with _—dfit753enous hyperalimentation. J. Infect. Dis.'fI6Z: 197-201. Moon, R. J., R. A. Vrable, and J. A. Broka. 1975. In situ separation of bacterial trapping and killing functTdns of the perfused liver. Infect. Immun. I2; 411-418. Morrow, S. H. and N. R. DiLuzio. 1965. The fate of foreign red cells in mice with altered reticuloendothelial function. Proc. Soc. Exp. Biol. Med. _II6; 647-652. Moser, S. A. and J. E. Domer. 1980. Effects of cyclophosphamide on murine candidiasis. Infect. Immun. .22: 376-386. Motta, P. 1977. Kupffer cells as revealed by SEM. pp. 93-102. In: E. Wisse and D. L. Knook. (eds.). Kupffer cells and other TdVer sinusoidal cells. Elseveir/North Holland Press. New York. Hunthe-Kaas, A. C. 1976. Phagocytosis in rat Kupffer cells ill vitro. Exp. Cell Res. 66; 319-327. Munthe-Kaas, A. C., T. Berg, P. 0. Selglen, and R. Seljelid. 1975. Mass isolation and culture of rat Kupffer cells. J. Exp. Med. III; 1-10. Munthe-Kaas, A. C., G. Kaplan, and R. Seljelid. 1976. On the mechanism of internalization of opsonized Kupffer cells In_ vitro. Exp. Cell Res. I66; 201-212. Murray, W. H., M. A. Fialk, and R. B. Roberts. 1976. Candida arthritis: a manifestation of disseminated candidiasis. Am. J. Med. ‘66: 587-595. Muto, M. and T. Fujita. 1977. Phagocytic activities of the Kupffer cell: a SEM study. pp. 109-119. In; E. Wisse and D. L. Knook. (eds.). Kupffer cells and other liver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 70 Myerowitz, R. L. 1978. Ultrastructural observations in dissemi- nated candidiasis. Arch. Pathol. Lab. Med. I62; 506-511. Myerowitz, R. L., G. J. Pazin, and C. M. Allen. 1977. Dissemi- nated chandidiasis: changes in incidence underlying diseases, and pathology. Am. J. Clin. Pathol. '66: 29-38. North, R. J. 1969. The mitotic potential of fixed phagocytes in the liver as revealed during the development of cellular immunity. J. Exp. Med. .I66: 315-323. Oblack, D. J., J. Schwartz, and I. A. Holder. 1978. Biochemical examination of sera during systemic Candida infection in mice. Infect. Immun. I6; 992-998. Odds, F. C. 1979. Candida and candidosis. pp. 1-21. Univer- sity Park Press. Baltimore, Md. Odegaard, A. and J. Lamvik. 1976. The effects of phenylbutazone and chloramphenicol on phagocytosis of radiolabelled C. albicans by human monocytes cultivated In_vitro. Acta. Path. Fficrobiol. Scand. Sec. C. 66; 37-44. Olstad, R. and R. Seljelid. 1980. The cellular reaction in nor- mal and tumor-bearing mice following intraperitoneal glucan injection. Acta. Path. Microbiol. Scand. Sec. C. .66: 97-102. 0tero, R. B., N. L. Goodman, and J. C. Parker. 1978. Predispos- ing factors in systemic and CNS candidiasis: histopathological and cultural observations in the rat. Mycopathol. 66; 113-120. Pearsall, N. and D. Lagunoff. 1974. Immunological responses to 6; albicans. I. Mouse-thigh lesion as a model for experimen- tal candidiasis. Infect. Immun. 6; 999-1002. Pearsall, N. N., J. Sundsmo, and R. S. Wiser. 1973. Lymphokine toxicity fbr yeast cells. J. Immunol. _I: 1444-1446. Pisano, J. C., J. P. Filkins, and N. R. DiLuzio. 1968. Phagocy- tic and metabolic activities of isolated rat Kupffer cells. Proc. Soc. Exp. Biol. Med. (I26: 917-922. Poulain, D., G. Tronchin, J. F. Dubremetz, and J. Biquet. 1978. Ultrastructure of the cell wall of 6. albicans blastospores: study of its constitutive layers by use of a cytochemical tech- nique revealing polysaccharides. Ann. Microbiol. (Paris). I26; 141-153. Price, F. M. and R. A. Cawson. 1977. Ph05pholipase activity in 6; albicans. Sabouraudia. I6; 179-185. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. “56: 660-682. 71 Ray, T. L. and K. D. Wuepper. 1976. Activation of the alternate (pr0peridin) pathway of complement RY.Q- albicans and related species. J. Invest. Dermatol. .62: 700-703. Riggi, S. H. and N. R. DiLuzio. 1961. Identification of a RES stimulating agent in zymosan. Am. J. Physiol. 266; 297. 1962. Hepatic function during RE hyperfunction and *hyperplasia. Nature. 1.9.3; 1292-1294. Rippon, J. W. and D. N. Anderson. 1978. Experimental mycosis in immunosuppressed rabbits. Mycopathol. 66; 91-96. Rogers, 0. E. 1960. Host mechanisms which act to remove bacteria from the bloodstream. Bacteriol. Rev. .26: 50-66. Rogers, T. and E. Balish. 1976. Experimental 6. albicans infec- tion in convential and germ-free rats. Infect. Immun. I6; 33-38. Rogers, T. J., E. Balish, and D. D. Manning. 1976. The role of thymus-dependent CMI in resistance to experimental disseminated candidiasis. J. RES. 26; 291-298. Rogers, T. J. and E. Balish. 1977. The role of activated macro- phages in resistance to experimental renal candidiasis. J. RES. .22: 309-318. 1978. Immunity to experimental renal candidiasis in ——_—rats. .Infect. Immun. _1_9_: 737-740. 1980. Immunity to Candida albicans. Microbiol. Rev. Roos, E. and K. P. Dingemans. 1977. Phagocytosis of tumor cells by Kupffer cells_in vivo and in the perfused mouse liver. pp. 183-190. In: E. Wisse and D. L. Knook. (eds.). Kupffer cells and other TTver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. Ruggiero, G., A. Andreana, and H. H. Fudenberg. 1980. Enhanced phagocytosis and bacteriacidal activity of hepatic RES during endotoxin tolerance. Infect. Immun. 22; 798-803. Salvin, S. B. 1952. Endotoxin in pathogenic fungi. J. Immunol. 66; 89-99. Sams, Jr., W. M., J. L. Jorizzo, R. Snyderman, B. V. Jegasothy, F. E. Ward, M. Weiner, J. G. Wilson, W. J. Yount, and S. B. Willard. 1979. Chronic mucocutaneous candidiasis: immunolo- gic studies of three generations of a single family. Am. J. Med. 62; 948-959. 140. 141. 142. 143. 144. 146. 147. 148. 149. 150. 151. 72 Sasada, M. and R. B. Johnston. 1980. Macrophage microbicidal activity: correlation between phagocytosis-associated oxidative metabolism and the killing of Candida by macrophages. J. Exp. Med. I62; 85-98. Satodate, R., S. Sasou, K. Oikawa, N. Hatakeyama, and S. Kutsura. 1977. Scanning electron microscopical studies on the Kupffer cell in phagocytic activity. pp. 121-129. In; E. Wisse and D. L. Knook. (eds.). Kupffer cells and other liver sinusoidal cells. Elseveir/North Holland Biomedical Press, New York. Sawyer, R. T., R. J. Moon, and E. S. Beneke. 1976. Hepatic clearance of 6, albicans in rats. Infect. Immun. I6; 1348- 1358. Sawyer, R. T. 1978. Hepatic clearance of Candida albicans in normal and Corynebacterium pgrvum treated rats. PhTD. disser- tation. Michigan State University. East Lansing, Mi. Scheiber, A. D. and M. M. Frank. 1972. Role of antibody and com- plement in the immune clearance and destruction of erythrocytes II. Molecular nature of IgG and IgM complement fixing sites and effect of their interactions with serum. J. Clin. Invest. 51: 583-589. Seelig, M. S. 1966. Mechanisms by which antibiotics increase the incidence and severity of candidiasis and alter immunologi- cal defenses. Bacteriol. Rev. 66; 442-459. Soderhill, K. and T. Unestam. 1979. Activation of serum prophe- noloxidase in arthropod immunity: the specificity of cell wall glucan activation and activation by purified fungal proteins of crayfish phenoloxidase. Can. J. Microbiol. 26; 406-414. Solberg, C. 0. 1974. Influence of phenylbutazone on the phago- cytic and bacteriacidal activity of neutrophil granulocytes. Acta. Path. Microbiol. Scand. Sec. B. 62; 258-262. Stahl, D. D., J. S. Rodman, M. J. Miller, and P. H. Schlesinger. 1978. Evidence for receptors mediating binding of glyc0pro- teins, glycoconjugates, and lysosomal glycosidases by alveolar macrophages. Proc. Nat. Acad. Sci. U.S.A. .26: 1399-1403. Stanley, V. C. and R. Hurley. 1969. The growth of Candida spe- cies in cultures of mouse peritoneal macrophages. J. Pathol. 62; 357-366. Stiffel, C., D. Mouton, and G. Biozzi. 1971. Role of the RES in defense against invasion by neoplastic, bacterial, and immuno- competent cells. pp. 305-314. In; N. R. DiLuzio and K. Flem- ming (eds.). The reticuloendothelial system in health and disease: Adv. Exp. Med. Biol. vol. 15. Plenum Press. New York. 73 152. Stone, H. H., C. E. Geheber, and L. D. Kob. 1973. Alimentary tract colonization by Candida albicans. J. Surg. Res. I6; 273-276. ' 153. Stossel, T. P. 1974. Phagocytosis I. NEJM. 266; 717-723. 154. . 1974. Phagocytosis II. NEJM. 266; 774-780. 155. . 1974. Phagocytosis III. NEJM. 266; 833-839. 156. Strauss, R. R., B. P. Benoy, and A. J. Sbarra. 1968. Effect of phenylbutazone on phagocytosis and intravellular killing by guinea pig PMN leukocytes. J. Bacteriol. 66; 1982-1990. 157. Strippoli, V. and N. Simonetti. 1973. Effect of tetracycline on the virulence of Y and M forms of Candida albicans in experi- mentally induced infections. Mycopath. Mycol. Appl. .6I: 65-73. 158. Stuart, A. E. 1970. The reticuloendothelial system. E. and S. Livingstone. London, England. 159. Summers, D. F., A. P. Grollman, and H. F. Hasenclever. 1964. Polysaccharide antigens of Candida cell wall. J. Immunol. .62: 491-499. 160. Suzuki, S., M. Suzuki, H. Hatsukaiwa, H. Sunayama, T. Suzuki, M. Uchiyama, F. Fukuoka, M. Nakanishi, and S. Akiya. 1969. Anti- tumor activity of polysaccharides. III. Growth inhibitory activity of purified mannan and glucan fractions from baker's yeast against sarcoma-180 solid tumor. GANN. 66; 273-277. 161. Swenson, F. J. and T. K. Kozel. 1978. Phagocytosis of Crypto- coccus neoformans by normal and thioglycolate-activated macro- phages. Infect. Immun. .2I: 714-720. 162. Trejo, A. R., C. G. Crafton, and N. R. DiLuzio. 1971. Influence of RE functional alterations on mortality patterns in endotoxin and tourniquet shock. J. RES. 6; 299-306. 163. Trnovec, T. A. Gajodsik, S. Bezek, D. Siki, V. Koprda, M. Zemanek, and V. Faberova. 1977. The effect of fractions isolated from 6; albicans on phagocytic activity of the RES in mice. J. RES. 22; 111-120. 164. Tsan, M. F. 1977. Stimulation of the hexose mon0phophate shunt independent of hydrogen peroxidase and superoxide production in rabbit alveolar macrophages during phagocytosis. Blood. .66: 935-946 165. Vaughn, V. J. and E. D. Weinberg. 1978. Candida albicans dimor- phism and virulence: role of copper. Mycopathol. 66; 39-42. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 74 Volkman, A. 1976. Disparity in origin of mononuclear phagocyte p0pulations. J. RES. I6; 249-268. Wardlaw, A. C. and J. G. Howard. 1958. A comparative survey of the phagocytosis of different species of bacteria by Kupffer cells: perfusion studies with isolated rat liver. Br. J. Exp. Pathol. 66; 113-117. Warr, G. A. 1980. A macrophage receptor for glycoproteins of potential importance of phagocytic activity. Biochem. Bi0phys. Res. Comm. .66; 737-745. Warr, G. A. and V. S. Shjivic. 1974. Origin and division of liver macrophages during stimulation of the mononuclear phago- cyte system. Cell Tis. Kin. .13 559-565. Warick, R. and P. L. Williams. (eds.). Gray's Anatomy. 35th edition. W. B. Saunders Co. Philadelphia, Penn. pp. 712-714. 1973. Whitehouse, M. W. 1964. Uncoupling of oxidative phosphorylation by some arylacetic acids (anti-inflammatory or hyphocholestero- lemic drugs). Nature. .26I: 629-630. Wilcoxon, F. and R. A. Wilcox. 1949. Some rapid approximate statistical procedures. American Cyanimide Co. New York. Williams, 0., J. A. Cook, E. 0. Hoffman, and N. R. DiLuzio. 1978. Protective effect of glucan in experimentally induced candidia- sis. J. RES. 126: 479-490. Winner, H. I. 1956. Immunity to experimental monoliasis. J. Pathol. Bacteriol. .II: 234-237. . 1960. Experimental monoliasis in the guinea pig. J. Pathol. Bacteriol. 26; 420-423. Wisse, E. 1977. Ultrastructure and function of Kupffer cells and other sinusoidal cells in the liver. pp. 33-60. In; E. Wisse and D. L. Knook. (eds.). Kupffer cells and other liver sinusoidal cells. Elseveir/North Holland Biomedical Press. New York. Wooles, W. R. and N. R. DiLuzio. 1963. Reticuloendothelial function and the immune response. Science. I62; 1078-1080. 1964. The phagocytic and proliferative response of the RES following glucan administration. J. RES. I; 160-169. Young, G. 1958. The process of invation and the persistence of Candida albicans injected intraperitoneally into mice. J. Infect. Dis. I62; 114-120. 75 Yu, R., C. Y. Bishop, F. P. Cooper, and H. F. Hasenclever. 1967. 180. Glucans from Candida albicans (serotype B) and Candida parapsillosis. Canad. J. Chem. ‘66: 2264-2268. Yu, R., C. T. Bishop, F. P. Cooper, H. F. Hasenclever, and F. Blanck. 1967. Structural studies of mannan from Candida albicans (serotype A and B), Candida parapsillosis, Candida Steiiatoidea, and Candida tropicallis. Canad. J. Chem. 66; 2205-2211. 181.