)JI'LJ A) - I ' u“: 14“ ‘ .W' v t' f" r vivfuu' vvvu-u,(, AN Nip} . :41 M _ (a. .I "f ‘ 1 ,.,, v ,l . u.,-..,.,. « ...... .- i): ,- ‘l w M91711." . n- .H. ‘vlu‘l 51., .- u‘n? 7 v - q I ’1 h ' ‘ V " P‘I' ;L{“ h'. (K I" ‘ r:°;a=w-s ‘— .c~ d h\“¢’\‘-"’- .4.'4. . l.."0 wS'S ”140% Ill/HUM!!!Ill!“Willi/HIM!(llIllNH! _»LlBRARY3 Michigan State University This is to certify that the _ dissertation entitled DEVELOPMENT OF LIPOSOMAL AMPHOTERICIN B BEARING ANTICANDIDAL ANTIBODY AND USE OF THIS PREPARATION IN THE THERAPY OF A MURINE MODEL OF CANDIDIASIS presented by Duane Bus 5 ell Ho spenthal has been accepted towards fulfillment of the requirements for _ l Doctor of Philosophx degree in Medical Weology 4K? 1 Major professo Date 15 Feb 1989 ”VIII: n- Arr........'... 1 ' "1 IA ~ I - . 0.12771 MSU LlBRARlES n » RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. A - ,o,’ n we -., I, r ~+ " 4'_l‘ {at u "“9! J.) DEVELOPMENT OF LIPOSOMAL AMPHOTERICIN B BEARING ANTICANDIDAL ANTIBODY AND USE OF THIS PREPARATION IN THE THERAPY OF A MURINE MODEL OF CANDIDIASIS BY Duane Russell Hospenthal A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1989 3Qx4'% If -1 ABSTRACT DEVELOPMENT OF LIPOSOMAL AMPHOTERICIN B BEARING ANTICANDIDAL ANTIBODY AND USE OF THIS PREPARATION IN THE THERAPY OF A MURINE MODEL OF CANDIDIASIS BY Duane Russell Hospenthal Systemic candidiasis and other disseminated mycoses account for appreciable morbidity and mortality in patients with hematologic malignancies and other debili— tating conditions. The treatment of choice for most of these mycotic infections is the antifungal agent, ampho- tericin B (AMB). While AMB has been shown to produce a wide range of antifungal effects, it has also been called the most toxic antimicrobial agent in current usage today. Due to this toxicity, new formulations have been explored for the deliver of this drug. One of the most interesting of these approaches has been the encapsulation of AME in phospholipid vesicles or liposomes. Liposomal amphotericin B (LAMB) was produced in this study by a reverse-phase evaporation procedure. Employing this procedure, LAMB was also produced which bore antibody specific to Candida albicans om its surface (LAMB-Ab). This new formulation, LAMB-Ab, was shown to possess external, g; albicans specific antibody. Toxicity of LAMB-Ab and LAMB to human erythrocytes in vitgg was much less than that of free AMB (fAMB). Anticandidal activity of these three compounds was comparable ig_vi££g. Therapeutic effect of LAMB-Ab was compared to other AMB containing preparations 1 vivo in a murine model of systemic candidiasis. In this model, mice infected by intraperitoneal injection received therapy in both prophylactic and treatment studies. LAMB-Ab improved the survival rates of mice over LAMB, which itself improved these rates over fAMB. Therapy with additional prepara— tions provided evidence that increased survival afforded by LAMB-Ab was due to the attachment of g; albicans specific antibody. LAMB-Ab produced similar results in the therapy of a model of candidiasis initiated via intravenous inocula- tion. Ihn this second model of disseminated murine candidiasis, the infection was followed and characterized by organ homogenization counts of viable yeasts. Targeting of liposomal AMB via the attachment of anticandidal antibody was shown to be experimentally possible in this study. This targeting, in addition to the reduced toxicity of liposome encapsulation, may allow therapeutic use of amphotericin B to become more safe and effective in the future. ACKNOWLEDGEMENTS I would like to express my sincere gratitude to all those who provided support and guidance to me in this undertaking. To my committee, special thanks for their ongoing guidance of this project from start to finish. To Dr. Alvin Rogers, I wish to extend my thanks for serving as uw'rmajor professor, for guiding my introduction into the world of research, and for all his help in medical mycology along the way. I wish to express my appreciation to Dr. Everett Beneke for all his help and special insight in this project and all the other problems which I brought to him. Also special thanks to Dr. Beneke for serving as my major professor during the studies which culminated in the awarding of my Master of Science degree. Special thanks to Dr. Gary Mills for help in getting this project off the ground through his help and insight iii bio- chemistry and pmoper research technique. In addition I would like to express my indebtedness to Dr. Karen Klomparens for her wisdom in the art of electron ndcroscopy and the help of her staff at the Center for Electron Optics. Thanks to Dr. Ronald Patterson for his guidance in the field of immunology. iv In addition to thanks bestowed my committee, I would like to extend my appreciation to Karl Gretzinger for his assistance in the murine study and to Dr. Dennis Gilliland for his interpretation of the statistical method which was used. Finally, I wish to express my undying love and appreciation to my wife Carol, and thank her and my parents, brothers, sister, and the rest of my family for their patience and support over all the years. TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O O O O O O 0 LIST OF FIGURES O O O O 0 O O O O O O O O O O O O O 0 INTRODUCTION 0 O O O O O O O O O O O O O O O O I O 0 LITERATURE REVIEW . . . . . . . . . . . . . . . . . . Candida albicans and Candidiasis . . . . . . . . Amphotericin B and Antifungal Therapy . . . . . . Liposomal Technology . . . . . . . . . . . . . . Liposomal Amphotericin B . . . . . . . . . . . . Targeting of Liposomes . . . . . . . . . . . . . ARTICLE I Development of amphotericin B liposomes bearing antibody specific to Candida albicans . . . . . . ARTICLE II Effect of attachment of anticandidal antibody to the surfaces of liposomes encapsulating amphotericin B in the treatment of murine candidiasis . . . . . . . . . . . . . . . . . . . ARTICLE III Treatment of a murine model of systemic candidiasis with liposomal amphotericin B bearing antibody to Candida albicans . . . . . . . . . . SUMMARY 0 O O O O O O O C O O O O O O O O O O O O O BIBLIOGRAPHY C O O O O O O O O O O O O O O O O O O 0 vi Page vii viii 12 24 3O 38 43 73 LIST OF TABLES Table Page Article I 1 The cytotoxic effect of free AMB, LAMB, and LAMB—Ab on a 2% human red blood cell suspension after incubation for 45 m at 37C 0 O O O O O O O O O O O O O O O O O O O 72 Article II 1 Mice surviving systemic candidiasis after treatment or prophylaxis with AMB preparations . . . . O O O O O O O O O O O O 87 vii Figure 1 Article I la lb 2a 2b Article I 1 LIST OF FIGURES Chemical structure of amphotericin B . . . . Transmission electron micrograph of LAMB-Ab (bar = 0 02 um) o o o o o o o o o o 0 Electron micrograph of LAMB (bar = 0.2 um) . Micrograph of LAMB-Ab reacted with Candida albicans (bar = 20 um) . . . . . . . . . . . Fluorescent micrograph of Candida albicans reacted with palmitic acid modified antibOdy (Ab-P) o o o o o o o o o o o o o o The effect of antibody-bearing amphotericin B-encapsulating liposomes (LAMB-Ab), liposomal amphotericin B (LAMB) and free amphotericin B (fAMB) on the growth of Candida albicans . . . . . . . . . . . . . . I Effect of liposome—encapsulated AMB bearing antibody specific to C. albicans on the survival of mice infected with g; albicans. One day after this infection, the animals were treated with 0.6 mg of AME per kg . . . Effect of liposome-encapsulated AMB bearing antibody specific to C. albicans on the survival of mice infected with C; albicans One day prior to infection, each animal was given a prophylactic dosage of 0.6 mg of AMB per kg . . . . . . . . . . . . . . . . . . . viii Page 14 64 66 68 70 71 88 89 LIST OF FIGURES, CONT'D. Figure Page Article III 1 Candida albicans present in the kidneys and livers of untreated mice sacrificed during the first 7 days of infection . . . . . . . 107 2 The effect produced by anticandidal antibody bearing liposomal AMB on the survival of mice infected with Candida albicans . . . . 108 ix I NTRODUCT I ON Fungal disease in its wide diversity of forms has plagued humans throughout the ages. From athlete's foot (tinea pedis) to valley fever (coccidioidomycosis), fungi have assailed the human species in numerous presentations. The severity of mycoses range from the cosmetic nuisance of tinea versicolor to the often life—threatening systemic mycoses. Of all fungal disease afflicting humans, Candida albicans, a yeast usually present as normal flora, is by far the most common etiologic agent. Treatment of systemic fungal infection has long been unreliable and even dangerous to the human host. Prior to the introduction of amphotericin B in the early 1960's, treatment for the majority of disseminated mycoses was virtually nonexistent. The discovery of amphotericin B in 1956 heralded the introduction of this polyene antifungal agent which, to date, is the drug of choice for most disseminated mycoses including aspergillosis, candidiasis, coccidioidomycosis, cryptococcosis, histoplasmosis, auui mucormycoses. Although it has remained the most reliable and widely used systemic antifungal jJ) the past twenty five years, amphotericin B has many toxic properties 2 which make this antifungal a difficult and often dangerous drug to administer. Due to the toxic properties of this compound, researchers iJi the past tn“) decades have searched for drugs to replace amphotericin B or methods to deliver this antibiotic in a less toxic form. The new family of azole antifungals, including ketoconazole, itraconazole, and fluconazole have shown promise in the replacement of amphotericin B in some applications. Semisynthetic amphotericin derivatives are also being studied due to their reduced toxicities. While final analysis of these new antifungals remains incom- plete, amphotericin B remains the drug of choice for all life-threatening systemic mycoses. Adjuvarnz therapies with other compound, such as S-fluorocytosine, hydro- cortisone sodium succinate, aspirin, diphenhydramine, prochlorperazine, sodium bicarbonate, mannitol, etc., have shown reduction in some, but not all aspects of amphoter- icin B treatment toxicity. Presently, the most promising attempt to reduce the toxic effect of amphotericin B (M1 the host, is by the encapsulation of this antifungal compound in phospholipid liposomes. Liposome incorporation of amphotericin B, though still largely experimental, has been shown to greatly enhance the therapeutic effect of the drug by allowing much larger dosages to be employed with dramatically reduced toxic effect. Currently, under a FDA Investigational New Drug 1 permit, liposomal amphotericin B therapy is producing excellent results in human hematologic malignancy patients with disseminated mycoses. Antibody targeting of other liposomal compounds has been shown in laboratory studies to aid in the site- specific delivery of these compounds. Targeted delivery of amphotericin B to specific sites (HE infection could lead to increased efficacy and decreased toxicity to the host tissue and therefore to the host. The goal of this project was to produce such liposomes and test the therapeutic efficacy of these vesicles. Production of liposomes which encapsulate amphotericin B and bear antibody specific to C; albicans on their surfaces was the first objective of this project. These lipsomes, once formulated, were compared to both commer- cial and liposomal forms of amphotericin B. Comparison of these formulations included observing their 12 vitro effect on the growth of C; albicans and their toxic effect on human red blood cells. The therapeutic effect of antibody—bearing liposomal amphotericin B was compared to that produced by the liposomal drug without antibody and the free unencapsulated antifungal in two murine models of systemic candidiasis. Disseminated candidiasis models employed in these studies were pmoduced via either intravenous or intraperitoneal injection of g; albicans. LITERATURE REVI EW Candida albicans and Candidiasis Candidiasis is an acute or chronic, superficial or systemic infection produced by members (Hi the genus Candida. Species of this genus considered the most virulent. thn order of decreasing virulence) include Candida albicans, g; tropicalis, g; stellatoidea, C; krusei anui g; parapsilosis (Hurley, 1980). Candida albicans is by far the most common species of the genus which produces candidiasis, accounting for greater than ninety percent of infections (Roberts et. al., 1984). Q; albicans is the only member of the genus which is thought to regularly produce a fatal disease in both humans and animals (Mourad and Friedman, 1961). In addition to being the cause of most cases of candidiasis, C; albicans is the etiologic agent for the vast majority of all fungal infections. This organism is the cause of over eighty percent of all mycoses (Lopez-Berestein et. al., 1983). CL albicans is a dimorphic fungus belonging to the form-phylum Deuteromycota, the form-class Blastomycetes, and the form-family Cryptococcaceae. Commonly this fungus occurs as emu asporogenous yeast which reproduces by 5 budding, but it can also form hyphae, pseudohyphae and chlamydoconidia. The telomorph state of this fungus is currently believed to belong in the basidiomycete genus Leucosporidium (Rippon, 1988). Candida albicans is considered an obligate animal saprotroph which can reside as normal flora in the throat, buccal mucosa, intestine, or vagina (Carmo-Sousa, 1969). The incidence of the yeast in normal humans as flora has been reported to be from ten to fifty percent (Seelig and Kozinn, 1982). Taschdjian et. al. (1973) reported isolation of Candida from normal subjects in the order of twenty tx>zfifty nine percent from the alimentary tract and eleven to seventeen percent from the vagina. Isola- tion of C; albicans from the vagina of pregnant women or women on oral contraceptives was approximately double that of normal healthy females. Odds (1988) derived similar numbers from compiling many isolation studies and weighing them by number of subjects studied. In this study he reported mean frequencies of recovery of g; albicarm; in the mouths, feces and vaginas of normal subjects as 17.7, 15.0 and 12.7 percent, respectively. Mean recovery of this yeast from hospitalized patients was calculated to hue 40.6 percent from the oral cavity and 26.4 percent in the feces. Women patients with vaginitis and without vaginitis had mean isolation rates of 25.9 and 17.8 percent, respectively. In a recent study, 6 Burford-Mason et. al. (1988) reported.aui oral carriage rate of thirty percent in healthy subjects. This group also presented a correlation of Candida carriage with blood group O and the non-secretion of blood group antigens. Rippon (1988) states that C; albicans comprises a small, yet constant population in the normal alimentary tract. 9; albicans is also common flora in bird and mammal alimentary tracts (Carmo-Sousa, 1969; Odds, 1988). C. albicans has the reported ability to survive on skin, sand, water, and food, and thus has no problem transfer- ring between hosts (Rippon, 1988). Candidiasis is EH1 opportunistic infection which can affect almost any area of the body. The mycosis may be localized as a cutaneous infection of the skin or nails, as a mucocutaneous infection of the mouth, throat, bronchi, lungs, vagina, or gastrointestinal tract, or it can present as a systemic infection as a septicemia, endocarditis, or meningitis. Candida species can also cause allergic disease in their hosts. Rippon (1988) cites five conditions in which Candida can leave its normal flora niche and enter into a pathogenic interaction with its host. These conditions are: 7 I1. Extreme youth. During the establishment of normal flora Candida may overgrow and produce infections such as thrush and diaper rash. 2. Physiologic change. Factors affecting a change in the yeast's environment which allow its over- growth. These include changes such as those seen in pregnancy, steroid therapy, and endocrine dys- functions such as diabetes. 3. Prolonged administration of antibiotics. The removal or alteration of other flora which sup- press Candida can lead to candidiasis. 4. General debility and the constitutionally inadequate patient. This category includes a wide range of host problems from slight avitaminosis to severe immune defects. Almost any disease, defect, or therapy which can allow a breach of the host's immunologic defense can lead to candidal infection. 5. Iatrogenic and break-barrier. Medical pro- cedures which are invasive, as well as any trauma, can allow a portal to invasion for Candida species. 8 Systemic infection by g; albicans is the most life— H threatening of the candidiases. These severe mycoses have _-.—.1.—~-‘ < become all too common in postoperative and immunosuppres- '1 sed patients. The prevalence of disseminated candidiasis '-‘ t’ has come from a rare occurrence prior to 1960 to an important hospital-acquired infection. In a recent study Candida was reported to be the fifth most common isolate from blood 1J1 nosocomial septicemias and the forth most common isolate from all blood cultures (Edwards et. al., 1978). Reingold et. al. (1986) reported a fifty percent increase in candidiasis acquired in the hospital and/or by immunocompromised patients between 1976 and 1980-1982. This increase in prevalence is believed to be coincident with the increased use of antibiotics, immunosuppres- sants, hyperalimentation fluids, catheters, invasive monitors, heroin abuse and surgical procedures including organ transplants and prosthetic heart valve replace- ments. Reingold et. al. (1986) also attributed the rise in candidiasis to the increasing life spans now afforded to immunocompromised patients by modern therapy. Candidi- asis is also prevalent and not a rare cause of fatality in burn victims. In a study reported by MacMillan et. al. (1972) 15 of 385 children admitted to a burn unit died of candidiasis. C; albicans is also a cause of mortality in myeloproliferative auui other organic immunocompromising disorders. It has been reported that over twenty percent 9 of leukemia patients and thirteen percent of lymphoma patients die of candidal infection (Feld et. al.,1974; Inagaki et. al., 1974). Systemic or deep candidiasis is an infection which may present as a focal involvement of an organ system or as a disseminated disease. [Hsseminated candidiasis is used widely as"a Synonym of systemic candidiasis due to the -nu. fact that the majority of these infections are initiated via a hematogenohs transmission of the yeast (Odds, 1988). The exceptions to this rule are infections of the respira- tory, digestive and urogenital tracts which are believed to be initiated directly by candidal flora. These infections are considered by many to belong in the mucocu- taneous group of Candida infections due to their origin and location. Systemic infection is believed to be most commonly produced by the spread of C; albicans by the blood from the gastrointestinal tract euni barrier-break procedures, especially intravenous catheters (Roberts et. al., 1984). t;; albicans has been documented to infect almost every organ and body compartment. The fungus can be the cause of meningitis, cerebral candidiasis, endocar- ditis, myocarditis, pericarditis, peritonitis, endophthal- mitis, arthritis, osteomyelitis, pyelonephritis, cortical renal infection, as well as septicemia (candidemia) (Emmons et. al., 1977; Odds, 1988; Rippon, 1988; Seelig and Kozinn, 1982). 10 Candidal septicemia and disseminated candidiasis are the most overwhelming forms of the systemic disease and can lead to rapidly fatal outcomes. These infections are usually confined to severely compromised patients, especially those with hematological malignancies. Candi- demia can lead to rapid death directly due to shock and coma (Dennis et. al., 1968; Stone, 1974), or via dissemi- nated intravascular coagulation or other disseminated pathologies (Phillippidis et. al., 1971). The heart and kidneys are the most common sites of disease in dissemi- nated candidiasis with cerebral, meningeal, bone and joint involvement occurring less often (Seelig and Kozinn, 1982). The kidney In“; been reported to be the primary target organ of time genus Candida in disseminating infections (Louria et. al., 1962). Pathology of all the organs affected include multiple microabscesses and a predominance of neutrophils (Kauffman and Jones, 1986). In murine studies deaths from high inoculum counts are associated with interstitial myocarditis and those with decreased inoculum are related to renal failure (Ryley et. al., 1988). qucute disseminated candidiasis closely resembling the human disease is produced by the intravenous or intraperi- toneal injection of C; albicans into mice or rabbits. Guinea pig or rats may also be used, but with a decreased reproducibility of results (Odds, 1979). Intravenous 11 inoculation of g; albicans leads to the production of lesions in the kidneys, lungs, liver, heart, brain, spleen, and other organs in mice (Louria et. al., 1963). Injection of 106 colony-forming units via the caudal veins of a mice leads to one hundred percent mortality within a week (Adriano and Schwarz, 1955). At lower inoculation doses, only 9; albicans recovered in the kidneys has been shown to be producing a progressive infection (Hurley and Winner, 1963). Rogers and Balish (1976) reported that g; albicans persisted in the lungs, spleen and liver of mice for up to thirteen days after infection, but the yeast did not multiply within these organs. This group supported results of Louria et. al. (1963) that indicated that the kidneys were the only organ in which this chronic disease progresses in this model. Kidney involvement in the murine model has been shown to be produced asymmetrically involving the invasion by C; albicans of the renal tubular lumen. Accompanying this kidney involvement is a decrease in kidney function auui a marked diuresis (Ryley et. al., 1988).‘ ReSults simular to these are produced in mice immmaperitoneally employing inoculum at a ten-fold increase over that of the intravenous model (Young, 1958). Intraperitoneal injection of g; albicans produces an almost immediate transformation from a yeast to a filamentous form followed by growth within the body-3 12 cavity. The filamentous form, thought to be produced to evade phagocytic cells, then invades abdominal organs, primarily the pancreas. By twenty four hours, C; albicans enters the blood vessels of the pancreas and is dissem- inated tn) other parts of the body including the kidneys (Young, 1958). As in the intravenous models tine kidney has been shown to be the only organ in which infection persists.*f “Amphotericin B and Antifungal Therapy Therapy of systemic mycoses dates back only as far as 1903 when tflue beneficial effect of potassium iodide was cfiscovered (Drouhet, 1970). This therapy proved to be only beneficial against most, but not all cases of sporotrichosis. It wasn't until almost fifty years later that antifungal chemotherapy made its next advance. In 1949, Hazen and Brown (1951) discovered nystatin, which they originally named fungicidin. Nystatin, the metabolic product of a actinomycete, heralded the discovery of more than 60 other antibiotics produced by the actinomycetes which are called polyenes (Drouhet, 1970). Of this large number of polyene antibiotics discovered in the early 1950's (including candicidin, ascosin, eulicin, trichi— mycin and amphotericins A and B), only amphotericin B {moved to be absorbed well enough and have a toxicity which was acceptable enough to be used systemically 13 (Hildick-Smith et. al., 1964). 'The other polyenes discovered during this era have been clinically limited to the therapy of cutaneous and mucocutaneous mycotic infections. Amphotericin 8 stands out in the polyene antibiotics as the only member of this group which can be readily administered into the body fluids in a therapeutic concentration without causing the host great injury (Hildick-Smith et. al., 1964)." Amphotericin B is currently, and has been for the past twenty five years, the drug of choice for all life- threatening mycoses. In systemic fungal infections, it is the most reliable and widely used antifungal agent (Graybill and Craven, 1983). Amphotericin B is currently recommended in the treatment of the aspergillosis, blastomycosis, disseminated candidiasis, coccidioido- mycosis, cryptococcosis, trichosporonosis, histoplasmosis, mucormycosis, and systemic sporotrichosis (Beneke et. al., 1984). In the treatment of systemic candidiasis, Seelig and Kozinn (1982) name this antifungal agent the most effective drug available. .‘_J;Amphotericin B was isolated and first described by Gold et. al. in 1956 as a metabolic product of Streptomyges nodosus M4575, a soil actinomycete isolated in Venezuela“ Synonyms for this antifungal include Amphozone, Fungizone, Fungilin and Ampho-Moronal. 14 Amphotericin B is 21 924.11 molecular weight polyene with a chemical formula of C47H73N017 as depicted in Figure 1 (Merck & Co., Inc., 1983). Amphotericin B is character- ized by having a macrolide ring closed by the formation of an internal ester and bearing eleven hydroxyl groups (Medoff and Kobayashi, 1980). The antifungal compound has both hydrophobic and hydrophilic aspects to its structure and is very insoluble in water. The seven double bounds of amphotericin B form the hydrophobic part, while the CH3 0 H W cam Figure 1. Chemical Structure of Amphotericin B deroxyl groups and the mycosamine moiety form the hydrophilic portion (Medoff et. al., 1983). In the commercial form, Fungizone, amphotericin E3 is delivered with deoxycholate to allow suspension of the drug in saline. The deoxycholate allows the amphotericin B to be delivered in micelles formed by this compound. OH 15 Amphotericin B, both in the commercial fornieuui in its native form, appears as a ytfliow compound which is sensitive to inactivation by exposure to air auui light (Merck & Co., Inc., 1983). The reported effect of amphotericin B, along with the other polyene antifungals, is to promote leakage of cellular components leading to cell death. This effect is dependent on the attachment of these polyene compounds to sterols jJIICell membranes (Medoff and Kobayashi, 1980). Amphotericin B produces both a fungicidal and fungistatic response depending on the concentration delivered to the fungus (Brajtburg et. al., 1980). At low concentrations, amphotericin B causes small pores or channels to leak small components such as potassium and magnesium ions, and is reversible. The damage is more severe and irreversible at higher concentrations, causing the fungicidal effect. This effect has also been shown in human erythrocytes with their release of potassium ions when exposed tn) low concentrations of the drug and hemoglobin release at increased amounts of amphotericin B (Teerlink et. al., 1980). Recently it has been shown that amphotericin B produces its lytic and lethal effects on Candida albicans by oxidative damage (Sokol-Anderson et. 211., 1986). Antifungal activity of amphotericin B comes from the increased affinity of the compound for ergosterol, which is found in fungal cell membranes, over cholesterol, the 16 common mammalian cell membrane sterol (Kotler-Brajtburg et. al., 1974). In addition to these primary effects, amphotericin B has also been reported to effect specific membrane enzymes of fungal cells and act as an immuno- adjuvant toward the host. Surarit and Shepherd (1987) have shown greater than seventy five percent inhibition of Candida albicans cell membranes enzymes ATPase, glucan synthase, adenyl cyclase and 5'-nucleotidase 1 situ. Vecchiarelli et. al. (1986) reported an 12 yiyg augmenta- tion of resistance to g; albicans provided by amphotericin B which correlated i3 yitgg with increased anticandidal activity of macrophages from mice in the study. Amphoter- icin B has also been shown to activate macrophages i2 vitro to kill bacteria (Lin et. al., 1977), parasites (Olds et. al. 1981), and tumor cell lines (Chapman and Hibbs, 1978). At higher concentrations the antifungal becomes toxic to cells and causes decrease in chemotaxis of neutrophils and phagocytosis and killing in macrophages (Hauser and Remington, 1982). Although amphotericin B has been shown to preferen- tially bind to ergosterol, it also binds to cholesterol found in mammalian cell membranes. This single fact provides much of the answer as to how and why amphotericin B causes the many toxic effects with which it is associ- ated. Human patients treated with.tflue antifungal drug present a wide range of acute and chronic toxic effects. 17 The most common of these are the association of chills and fever with the intravenous infusion and the varying degree of nephrotoxicity seen after this treatment (Medoff and Kobayashi, 1980; Pratt, 1977; Speller, 1980). Short term effects begin usually within four to six hours after the beginning of infusion (Graybill and Craven, 1983). These can commonly include vomiting, hypotension and delirium in addition tn) fever and chills. Other symptoms which occasionally occur are nausea, abdominal pain, headache, anorexia, phlebitis and rarely cardiac arrhythmias (Pratt, 1977). Treatment over an extended period (ME time with slow infusion rates is performed to avoid the acute cardiotoxicity of amphotericin B which can progress to cardiovascular collapse and death. After a few weeks of therapy a reversible normocytic, normochromic anemia appears in most patients. This is due to a suppression of erythrocyte production (Brandriss et. al., 1964). In rare instances leukopenia and thrombocytopenia will accompany this anemia. Other rare toxic manifestations include hepatic dysfunction and allergic reactions. The major toxic effects of amphotericin B therapy are those which affect the kidney. Glomerular filtration rate has been shown to be decreased approximately forty percent in almost all patients treated with amphotericin B (Medoff et. al., 1983). During treatment, nephrotoxicity is monitored via blood urea nitrogen and creatinine serum 18 levels. In one study of amphotericin B therapy, ninety three percent of patients had elevated blood urea nitrogen levels and eighty three percent had elevated creatinine values (Butler et. al., 1964). Histopathology of amphotericin B treated kidneys show tubular degenerative changes with intratubular and interstitial calcium deposits (Wertlake et. al., 1963). Renal acidosis (nu) also be present in individuals being treated with ampho- tericin B (McCurdy et. al., 1968). This acidosis is believed t1) be associated with the nephrocalcinosis and renal loss of potassium ions seen in these patients. The combination of these renal effects of the drug lead to varying degrees of permanent kidney damage. This damage may be severe enough to cause renal failure and even death in the patient before amphotericin B (2“) eradicate the fungal disease being treated (Butler et. al., 1964). Even with its toxic effects, amphotericin B remains the single most reliable drug in the treatment of most I life-threatening mycoses (Taylor et. al., 1982)."; Many .“ other systemic antifungals have been introduced since amphotericin B in 1956, yet at this time these drugs have either proven to be less effective or are still experi- mental. These other antifungal compounds include the pyrimidine analogue 5-fluorocytosine, the azoles (e.g. 19 ketoconazole, itraconazole and fluconazole), semisynthetic derivatives of amphotericin B and chitin-syntheshs inhibitors. Flucytosine (S-fluorocytosine) is a synthetic antifungal agent effective against many disease-causing yeasts and the dematiaceous fungi which cause chromo- blastomycoses. It is believed to work by inhibiting protein synthesis after incorporation into RNA” 'The use of flucytosine is limited by the large number of organisms which are resistant or become resistant to this antifungal during treatment (Roberts et. al., 1984). 5-fluoro- cytosine can produce dose—related thrombocytopenha and neutropenia, as well as nausea, diarrhea and rash. In current usage, flucytosine is the recommended drug for chromoblastomycosis and cryptococcosis (in combination with amphotericin B) (Beneke et. al., 1984). The azoles are a group of synthetic compounds which appear to work as antifungal agents by inhibiting ergosterol biosynthesis. Clinical resistance to the azoles is a rarity, as is clinical resistance to amphotericin B and the other polyenes (Hitchcock et. al., 1987; Smith et. al., 1986). The azoles currently used against systemic fungal infections are miconazole, ketoconazole, itraconazole and fluconazole. Miconazole is poorly absorbed orally, intravenously produces side effects including phlebitis, pruritus, nausea, fever or 20 chills and rash, and has had very limited clinical testing (Heel et. al., 1980). This antifungal agent also has been reported to cause hyponatremia and hematological abnormalities (Stevens, 1977). For these reasons and the advent of ketoconazole, which has a longer luilf-life iri serum and is less toxic (Brass et. al., 1982), miconazole is regarded as a second choice drug (Roberts et. al., 1984). Ketoconazole is the most widely used of the systemically effective azoles. It is recommended in the treatment of chronic and disseminated candidiasis, chromoblastomycosis, paracoccidioidomycosis, and some forms of coccidioidomycosis (Beneke et. al., 1984). Some serious side effects of this antifungal include hepatitis, gynomastia and impotence. Other toxic effects include rash, nausea, anorexia and gastrointestinal disturbances (Graybill and Craven, 1983). These effects are either rare (hepatitis is estimated to occur in 1:10,000 patients) (n: mild in their presentations as compared to fluconazole or amphotericin B (Roberts et. al., 1984). For this reason ketoconazole is often the first choice in the treatment of systemic mycoses that are not life- threatening. This reduced toxicity and the fact that ketoconazole is orally prescribed has also led to it being widely studied as a possible prophylaxis for fungal infection in neutropenic cancer patients (Meunier- Carpenter, 1984; Young, 1982). New additions to the azole 4‘ I / 21 family of antifungal agents, itraconazole and fluconazole, are presently classified as FDA Investigational New Drugs in the United States. Itraconazole and fluconazole are effective against a wider range of fungi and are better absorbed than ketoconazole. Itraconazole has been shown to be absorbed eight times better than ketoconazole and to be one hundred times more active against Aspergillus strains (Marichal et. al., 1985). Both itraconazole and fluconazole are cleared more slowly than ketoconazole and thus therapeutic levels are easier'tx) reach with these drugs in the patient (Graybill and Ahrens, 1984). Fluconazole has also been shown to be more active than ketoconazole. Troke et. al. (1985) reported an example of the superior performance of fluconazole in a murine model of systemic candidiasis. Most importantly, fluconazole is water soluble and has the ability to reach therapeutic levels in the brain via oral administration (Graybill et. al., 1986). In a recent study of human coccidioidal meningitis patients fluconazole was shown to produce substantial penetration into the cerebrospinal fluid with only minimal toxicity (Tucker et. al.,1988). Initial results of this therapy were encouraging and prompted Tucker et. al. (1988) to state that fluconazole may become the drug of choice for coccidioidal meningitis. The semisynthetic derivatives of amphotericin B include the experimental drugs amphotericin B methyl ester 22 hydrochloride and N-D-ornithyl amphotericin Esinethyl ester. These two derivatives have been shown to be less toxic and more efficacious than amphotericin B in laboratory animals (Parmegiani et. al., 1987). However, both of these compounds have been shown to produce neurologic effects at increased dosages. In subchronic toxicity studies conducted by Massa et. al. (1985), N-D-ornithyl amphotericin B methyl ester proved to cause lxflmvioral and morphological brain damage in dogs at doses of 2.5 and 10 mg of the drug per kilogram of animal. Treatment with large cumulative doses of amphotericin B methyl ester hydrochloride over prolonged periods have also been shown to produce neurologic problems. In clinical studies this antifungal agent produced a distinctive neurologic syndrome and injury to human white matter (Ellis et. al., 1982). I Peptidyl nucleoside antibiotics known to inhibit chitin synthesis are yet another experimental class of antifungal agents being explored. This group of compounds, which includes nikkomycins and polyoxins, have been reported to produce very low or inapparent toxic effects in laboratory mice (Isono et. al., 1965). Nikkomycin, the most potent of the group, has proven itself effective in a murine model of disseminated candidiasis (Becker et. al., 1988). The problem encountered in this study was that when treatment was 23 stopped, the candidal infection reestablished in the nikkomycin-treated mice and ultimately led to their death. Paralleling the search for new, effective antifungal drugs has been the search for safer methods of presenting amphotericin B to the mammalian patient. The two most common approaches to the task of reducing toxicity have been to administer amphotericin B concurrently with other drugs (n: substances or to encapsulate the antifungal in phospholipid vesicles (liposomes). Synergistic effect of amphotericfimiia with 5—fluorocytosine have been shown to reduce the toxicity of amphotericin B by reducing the amount used. This reduction has proven effective in only a few of the systemic mycoses, such as crytococcosis (Speller, 1980). Experimentally, mycolase has been studied as a compound to potentiate amphotericin B treatment (Chalkley et. al., 1985). This mixture of enzymes, which includes a chitinase, glucanases and exoglycosidases, has shown some enhancement to ampho- tericin B therapy. Many compounds have been and/or are presently being used in the attempt to reduce the toxic symptoms of amphotericin B therapy. To reduce chills and fever, hydrocortisone sodium succinate is added to the infusion of the antifungal drug. Premedication with aspirin, diphenhydramine or meperidine hydrochloride has also been employed to control chills and fever (Graybill and Craven, 1983; Medoff and Kobayashi, 1980). Nausea 24 associated with amphotericin B therapy is often treated with prochlorperazine (Medoff et. al., 1983), Gouge and Andriole (1971) have reported reduction in nephrotoxicity in rats co-treated with sodium bicarbonate. Mannitol administered concurrently with amphotericin B has been shown tx> reduce blood urea nitrogen levels to normal in dogs (Hellebusch et. al.,1972). This effect has not proven to reflect any reduction in renal dysfunction in patients treated in this manner (Bullock and Bathona, 1976). The most recently applied method to reduce amphotericin B toxicity has been to encapsulate this compound in liposomes. This encapsulation has been shown to beeffective in the reduction of toxicity of amphoter- icin E3 in a large number of laboratory and clinical studies. A U.S. patent has recently (1988) been issued to Liposome Technology Incorporated of Menlo Park for their formulation of liposomal amphotericin B. Liposome Technology Liposomes (lipid bodies) is the term coined to describe vesicles consisting of phospholipid bilayers assembled ime: closed membrane systems (Bangham, 1980). These phospholipid vesicles were initially used as model systems for biological membranes. Much of the action of amphotericin B and its interaction with sterols has been investigated using liposomes as model membranes (Bolard 25 et. al., 1984; Clejan and Bittman, 1985; Cohen, 1983; Cybulska emu. al., 1986). Recently, liposomes have been explored as potentially important drug delivery systems due to their property to entrap compounds in their internal aqueous or lipid compartments and sequester these compounds from direct contact with the host (Lopez- Berestein, 1986). Liposomes can carry polar drugs in their aqueous compartment and nonpolar drugs in both this and their lipid membrane compartment (Juliano and Stamp, 1979; Stamp and Juliano, 1979). Liposomal encapsulation puevents the reaction of compounds with cellular blood constituents and reduces the clearance rate of many encapsulated substances (Gregoriadis and Neerunjun, 1975). This allows reduction in allergic and toxic reactions caused by unencapsulated forms of many compounds. The contents (Hf liposomes are believed to be transferred to cells directly by fusion or via the endocytosis of the entire liposomes by cells (Scheider, 1985). The actions of liposomes have been shown to be modified by physical and chemical properties of their construction. An example of this is the tendency of negatively charged liposomes to accumlated preferentially in the spleen (Schneider, 1985). Much researdh has been focused on the encapsulation of toxic antitumor and other drugs which have been limited in their usage because of toxicity problems (Juliano and Stamp, 1978). 26 Besides drug delivery, liposome transport of many other substances has been studied. Liposomes have been studied as vaccine carriers to produce immunity to many viruses, bacteria, parasites and various proteins including birth control antigens (Alving, 1987; Ostro, 1987a). The induction of the immune system is another area of exploration. Liposomes incorporating a polyribo- nucleotide have been shown to increase interferon production in cell culture (Straub et. al., 1974). The property of liposomes that lends their ability to interact preferentially with host immune cells is their tendency to accumulate in the reticuloendothellal system of the patient upon intravenous injection (Hsu and Juliano, 1982; Juliano and Stamp, 1975; Kimelberg and Meyhew, 1978; Pagano and Weinstein, 1978; Richardson, 1983). Other compounds carried in liposomes include hormones, such as bovine somatotrophic hormone (to increase milk production) and epithelial growth factor (to aid in wound healing), and tear components (Ostro, 1987). Liposomes are also being used as emollients in cosmetics and as diagnostic tools such as the recently introduced immunoassay test kit for Streptococcus species (Schach, 1987). Liposomes are produced in three classes or categories depending on their relative size and lamellar nature. These three categories are small unilamellar (SUV), large unilamellar (LUV) and multilamellar (MLV) vesicles. These 27 classes have associated with them differing physical properties and thus differing applications. Multilamellar vesicles were the first liposomes produced. Bangham et. al. described the first and most popular method to produce these liposomes in 1965. In this procedure the lipids used are dissolved in an organic solvent. This solvent is then evaporated leaving the lipids as a film on the inner surface of the preparation flask. Drugs or other compounds to be encapsulated are then added to the flask in aqueous solution. The flask is then swirled by hand to produce large MLVs. These large multilamellar liposomes can be~tup to fractions of a ndllimeter in diameter and have the concentric appearance of a sliced onion when observed by electron microscopy (Pagano and weinstein, 1978). These liposomes are very responsive to osmotic gradients and can swell or shrink in response. Large multilamellar liposomes have a capture volume of about 4 ul/mg phospholipid (Poznansky and Juliano, 1984). The second class of liposomes produced were the small unilamellar vesicles. These phospholipid vesicles usually have a diameter in the twenty to fifty nanometer range and single bilayer membranes and aqueous compartments. These small liposomes have the smallest capture volume of time three classes of vesicles, a mere 0.5 ul/mg lipid (Poznansky and Juliano, 1984). SUVs are insensitive to osmotic pressures and display different physical chemical 28 characteristics than MLVs (Pagano and Weinstein, 1978). These microscopic vesicles also present the advantage of an increased half-life in blood over the larger classes of liposomes (Schneider, 1985). These small liposomes are commonly produced by ultrasonic dispersion. This may be accomplished via probe sonication or in an ultrasonic cleaning bath. Other less common methods include forcing phospholipid mixtures under high pressure through a needle or other small orifice, or by preparing lipid mixtures in detergents and then removing the detergents by dialysis. The third and final class of liposome preparations is large unilamellar vesicles. These large, single bilayered membranes are usually between sixty nanometers and several micrometers in diameter. Characteristically these liposomes are two hundred to one thousand nanometers and a capture volume of approximately 14 ul/mg lipid (Poznansky and Juliano, 1984). These liposomes have been produced by the slow hydration of phospholipid film into distilled water and through the injection of an ether solution of lipid into warm aqueous media (Pagano and Weinstein, 1978). The most common preparation method of LUVs is via reverse-phase evaporation as described by Szoka and Papahadjopoulos (1978). In this method material to be encapsulated, in aqueous mixture, is added to phospho- lipids in organic solvent and then the liposomes are 29 formed by the evaporation of these solvents under reduced pressure. Sonication of this product can also be used to produce small unilamellar vesicles. The ability of liposomes to deliver compounds 12 yiyg was first examined by Gregoriadis and Ryman in 1972. In this study liposomes were shown to deliver albumin into the livers of rats.’ Depending on the targeted site of a particular liposome-encapsulated compound, liposomes may be introduced E1 3133 by intravenous, intraperitoneal, intracerebral, or localized injections or by oral admini- stration (Kimelberg and Mayhew, 1978). Local..intramus- cular injections lead to the accumulation of encapsulated material in the lymphatic system associated with the area (Gregoriadis, 1977).. Oral administration of liposomal insulin has been proven to be a possible method to deliver insulin in normal and diabetic rats (Gregoriadis, 1977). Factor VIII has been reported to be absorbed by Hemophilia A patients given this procoagulant orally in liposomes (Schneider, 1985). The most common method, intravenous injection, provides a route to administer compounds to a large area of the host via the circulatory system. Circulating liposomes are preferentially taken up by organs rich in reticuloendothelial cells. Accumulation in liver, spleen, lung and bone marrow is the rule in both mice and humans (Lopez-Berestein et. al., 1984b and 1984c). Liposomes of 30 diameters equal to or smaller than one micron circulate freely in the mammalian patient. Hunt et. al. (1979) reported that those larger tend to accumulate in the lungs of mice. This presents a problem with large liposomes delivering drugs unless the lungs are indeed tflua target organ. Papahadjopoulos (1986) introduced the projected study of liposomal treatment of Pneumocystis pneumonia in acquired immunodeficiency syndrome patients (AIDS) as emu example of when the lungs would be targeted. Liposomes can be altered in size after they are produced by many methods. Multilamellar vesicles in particular are usually large and heterogenous by nature. These liposomes can be fractionated by centrifugation or separated by filtering. Olson et. al. (1979) presented a procedure to make defined size liposomes by a process of extrushxu through polycarbonate membranes. Liposomal Amphotericin B The first study of the encapsulation of amphotericin B in liposomes was conducted by New et. al. (1981). While exploring liposomal leishmaniasis therapies this group discovered that amphotericin B toxicity to tflue host was 6 reduced when delivered in liposomes. In the following year Taylor et. al. (1982) published a study which employed liposomal amphotericin B in the treatment of murine histoplasmosis. Liposomes produced by this group 31 were of the MLV class and included ergosterol. These vesicles were shown to produce a nine-fold reduction in toxicity of amphotericin B. The study also showed that encapsulation increased the tissue concentration and decreased the serum concentration of the drug. Members of this group then went on to explore the use of liposomal amphotericin B in the treatment of murine cryptococcosis (Graybill et. al., 1982). This study indicated liposomal amphotericin B to have reduced toxicity and better localization than the commercial preparation of the antifungal. The liposomes and the disease-producing yeast cells accumulated in.tflua reticuloendothelial system as expected. In what seems to contradict previous studies, liposomal amphotericin B was shown to reduce intracerebral counts of cryptococci following intravenous therapy. Using the same liposome formulation, Ahrens et. al. (1984) explored this novel antifungal therapy in a murine model of candidiasis. Liposomal amphotericin B showed increased efficacy by allowing larger doses tn) be used. At equal doses with the commercial drug preparation, liposomal amphotericin B in this formulation proved less effective. Later studies have reported that liposomes containing sterols have inherent toxic aspects which can diminish the apparent efficacy of amphotericin B encapsulated within them (Juliano et. al., 1985). Hopfer et. al. (1984) reported that with the incorporation of 32 sterols twelve and fifty times as much amphotericin B was needed to kill strains of Candida albicans when liposomes included cholesterol and ergosterol, respectively. Other groups have also studied liposomal amphotericin B in the treatment of candidiasis and leishmaniasis reporting lowered toxicity in all cases (Panosian et. al., 1984; Tremblay et. al., 1984 and 1985). In order in) produce more stable amphotericin B liposomes some groups have prepared polymerized phospholipid vesicles (Mehta et. al., 1986). Unlike the previously mentioned studies, polymerized liposomes produced thus far do not reduce amphotericin B toxicity. The most thoroughly explored and promising encapsula- tion of amphotericin B in liposomes is time formulation introduced by the Lopez-Berestein group of Houston (Lopez- Berestein et. al., 1983; Juliano et. al., 1983). Liposomes produced by this group are nmltilamellar vesicles consisting of a seven to three molar ratio of dimyristoyl phosphatidylcholine (DMPC) and dimyristoyl phosphatidyl-glycerol (DMPG) prepared in a manner similar to the original Bangham et. al. (1965) procedure previously discussed. This formulation has been shown to produce no abnormal findings in tdood chemistry or histology of animals injected with it (Lopez-Berestein et. val., 1983). Liposomal amphotericin B prepared in this manner produced none of the nephrocalcinosis and renal 33 parenchymal edema that the free (commercial) amphotericin B produced in mice. The liposomal drug proved as effective as the free drug in equal concentrations in an murine model of disseminated candidiasis (Lopez-Berestein et. al., 1983). Though the maximum tolerated dose of free drug was 0.8 mg per kilogram of mouse (in this study), 12 milligrams of liposomal amphotericin B per kilogram was well tolerated by the mice. This allowed much larger doses of liposomal drug to be used and thus a higher therapeutic index achieved by the encapsulation of amphotericin B. The pharmacologic effects of this liposome formulation (without amphotericin B) in cancer patients has been examined (Lopez-Berestein et. al., 1984b). Liposomes carrying radiolabel were shown to produce tissue retention predominately in the reticuloendothelial cells of the patients. No toxic side effects were reported and the conclusion was made that effective targeting of drugs to treat pathologic conditions involving organs rich in reticuloendothelial cells could be afforded via liposomes. The additional finding that liposomal amphotericin B accumulated in the reticuloendothelial cells even more in mice infected with candidiasis also kindled hopes of more effective therapy of fungal infection (Lopez-Berestein et. al., 1984c). Many patients afflicted with systemic candidiasis (n: other fungal infections have hematologic 34 malignancies or other underlying conditions which render them neutropenic and immunocompromised. Knowing this, the Lopez-Berestein group next studied the effect of liposomal amphotericin B in the treatment of candidiasis in neutropenic mice, as well as the effect of this compound on immune and other circulatory cells. Liposomal amphotericin B prophylaxis 1J1 neutropenic mice was reported to provide protection against candidi- asis (Lopez-Berestein et. al., 1984). Empty liposomes or free amphotericin B showed no prophylactic efficacy in these studies. In treatment studies of disseminated candidiasis in neutropenic mice liposomal amphotericin B proved again to be more efficacious due to the increase in dosage that this preparation allowed (Lopez-Berestein et. al., 1984a). In addition, liposomal drug therapy was shown in: be effective five days after initiation of the disease while free drug therapy had to be initiated within three days to provide any effect. Liposomal amphotericin B improved survival time and reduced renal impairment in the neutropenic mice. Liposome-encapsulation of amphotericin B has been shown lg YEEEE to greatly reduce the toxic effect of the drug on mammalian cells. The effect of liposomal amphotericin B produced by the Lopez-Berestein group on human blood cell was reported by Mehta et. al. (1984). They showed that while both free and liposomal drug killed 35 C; albicans just as effectively, free amphotericin B lysed erythrocytes at one microliter per milliliter buffer, but the liposomal drug produced no appreciable lysis at one hundred times this concentration. A later study by Juliano et. al. (1987) showed that lack CHE toxicity in these liposomes is due to the use of saturated acyl chain phospholipids. Liposomal amphotericin B produced MHJH] unsaturated acyl chain phospholipids have been shown to be as toxic as the free drug itself. They also pointed cum: that the toxicity or lack of toxicity pucduced by amphotericin B preparations is due to the lipid composition of the membranes that they contact. Juliano et. al. (1987) put forth the theory that amphotericin B is selectively transferred from liposomes to cells of the host and not slowly leaked from the liposomes as previously believed. They purport that this diffusion is regulated by properties of both the liposome and cell membranes. Szoka et. al. (1987) tested the toxicity and organ distribution of various liposomal and free amphotericin B preparations and came to a somewhat different conclusion. They concluded that liposomal preparations of amphotericin B showed no change from the free drug in organ distribution and only lessened or slowed it}; toxic effect. They theorize that liposomal encapsulation reduces toxicity of amphotericin B by slowing its transfer ratelto sensitive target cells. 36 Mehta et. al. (1985a) reported that the 7:3 DMPC to DMPG liposomal amphotericin B formulation reduced the immunosuppressive effects seen in free amphotericin B therapy. The liposomal drug was shown not to reduce macrophage production of superoxide anion or differentia- tion markers 12 ZiEEQ like the free drug does. While free amphotericin B inhibited T cell blastogenesis at high concentrations, this proved not the case for the liposomal form of the antifungal. Both forms of the drug did however inhibit antibody production 12 yiyg. The liposomal amphotericin B formulation of the Lopez- Berestein group is, as stated above, currently being investigated clinically under a FDA Investigational New Drug permit. Results of the irdtial clinical trials of liposomal amphotericin B have been quite promising. The first published results reported remission in eight of twelve patients treated with the liposomal compound (Lopez-Berestein et. al., 1985). Included in this group were seven persons with aspergillosis, three with candidiasis, one with mucormycosis, and one suffering from histoplasmosis. The twelve patients, all immuno— supressed and nine granulocytopenic, were all previous nonresponders to free amphotericin B as well as other antifungals. Liposomal amphotericin B infusion produced mild or moderate temperature increases or chills i1) two patients, but no other toxic responses. In fact, three 37 patients with liver or kidney dysfunction actually showed improvement in the respective affected organs during the course of treatment. In a similar group of patients, all with hematologic malignancies, fungal infections, and previous amphotericin B therapy, liposomal amphotericin B also proved to be efficacious (Shirkhoda et. al., 1986). Six of the eight patients with disseminated mycosis of the liver, most also having spleen involvement, showed improvenmnn: after liposomal therapy. In another report of nine patients with hepatosplenic candidiasis, Lopez-Berestein et. al. (1987) have reported cures produced by the liposomal drug in eight cases. The remaining patient showed improvement during therapy. One of the eight cured patients in this study had a history of anaphylactic reaction to free amphotericin B. Besides the study of liposomal amphotericin B, other liposomal strategies for the therapy of systemic fungal infection have and are currently being explored. These include studies of the encapsulation of other antifungal compounds and studies of combined treatment of liposomal amphotericin B with other liposome-encapsulated materials. The study of liposome-encapsulated nystatin in the treatment of candidiasis has shown that intravenous injection of this drug may be possible (Mehta et. al., 1987 and 1987a). Another study by Mehta et. al. (1985) 38 showed that combined treatment of liposomal amphotericin B and a liposome-encapsulated macrophage activator (6-O-stearoyl- N-acetylmuramyl- L—alpha-aminobutyryl-D- isoglutamine) provided an additive effect as compared to either of these compounds alone. This macrophage activator itself has been shown to be an effective prophylactic liposomal therapy against candidiasis in mice (Fraser-Smith et. al., 1983; Lopez-Berestein et. al., 1983a). These results may also be reflected in human macrophage activation. Mehta et. al. (1982) showed that macrophage activator delivered in liposomes was phago— cytized by human peripheral blood monocytes lg vitro. Targeting of Liposomes Liposomes delivered intravenously are localized in the reticuloendothelial system after a period of circulation in the plasma. The duration of this circulation is dependent upon various characteristics of each particular liposome preparation including size, composition and surface charge (Gregoriadis et. al., 1985 and 1977). Small liposomes remain in the circulatory system for much longer periods than large multilamellar vesicles (Gregoriadis, 1977). During the circulatory life of the liposomes, a chance for specific targeting of these vesicles to specific cells exists. Employing this idea, and using various molecules on the surface of liposomes, 39 researchers have explored this theoretical concept of specific delivery. Molecules which have been bound to the surface of liposomes include glycoproteins, glycolipids, antibodies auui other cytophilic molecules. Initial results from these studies indicate that specific liposomal targeting is indeed a realistic approach to deliver compounds to cells in proximity of the circulatory system. Glycolipid targeting of liposomes to the liver of mice has been reported by Nozawa et. al. (1986). Galactose— containing ligands on liposome surfaces were reported to be highly specific for hepatocytes. Bachhawat and Dasgupta (1986) also targeted hepatocytes with the glycolipid, monosialoganglioside with success. A glycolipid sulfatide has been shown by Yagi and Naoi (1986) to allow liposomes access across time blood-brain barrier to the brain. Polysaccharide-coated liposomes have been shown to be effective in the treatment of Legionnaire's disease in guinea pigs (Sunamoto, 1986). Sunamoto states that this use of sisomycin encapsulated in O-palmitoylamylopectin- coated liposomes is the first successful treatment of a bacterial infection by targeted liposomal drug. Antibody and fragments of antibody are by far the most studied of the liposome-targeting molecules. A multitude of methods have been devised to add active antibody to the 40 surface of liposomes. The simplest of these methods, detailed by Gregoriadis and Neerunjun in 1975, involves adding antibody to the aqueous solution to be encapsula- ted. Huang and Kennel (1979) later reported that the sonication employed in this procedure was responsible for the external expression of a proportion of the antibody. This group showed that sonication of preformed liposomes with antibody led to the externalization of some antibody with retained activity. Covalent coupling of antibody was reported in 1979 by Torchlin et. al. This group "activated" preformed liposomes containing Indium-111 chloride with glutaralde- hyde. These "activated" liposomes were then allowed to react with antimyosin antibody overnight. These liposomes were shown to cause the localization of Indium-111 chloride in the infarcted area of a dog's heart, the targeted site. Leserman (1981) found fault with both of the above mentioned formulations of antibody—targeted liposomes. He stated that.tflua sonication insertion method was too inefficient and also likely to produce Fc-mediated binding. He was unable to repeat the glutaraldehyde formulation results in his laboratory. The method introduced by Leserman involved covalently modifying both the amine end of the antibody and the phosphatidylethanol- amine used in the liposomes. This modification added the 41 cross-linking reagent N-hydroxy succinimidyl -1 -3- (2 pyridilyldithio) propionate, also denoted SPDP, to both of these components. Following production of liposomes with SPDP linked phosphatidylethanolamine, modified antibody is added and allowed to react forming the covalent connection of the two. Specificity of these liposomes was reported by Leserman et. al. (1983) in a study involving localization of these vesicles in mice. The study also reported that antibody bound to liposomes failed to induce antiidiotypic responses. The localiza- tion of this preparation of targeted liposomes yum; also studied by Gregoriadis et. al. (1985) with similar results. The addition of F(ab')2 to liposomes by oxidation- reduction reaction has been reported by Heath et. al. (1980). 'Fhis preparation entailed first oxidizing the surface of preformed liposomes with periodate, followed by the addition of antibody fragment and the reducing agent sodium cyanoborohydride. Binding of liposomes tn) human erythrocytes increased two hundred times when this method was employed to add conjugated antierythrocyte F(ab')2 to these vesicles. This method was also reported to be effective ill the binding of whole antibody to liposomes (Heath et. al., 1984). Recent work performed by this group shows a shift from this formulation to the prepara- tion presented by Leserman above using SDSP cross-binding. 42 Employing this new formulation, they have shown 13 yitgg that the toxic antitumor agent methotrexate-gamma- aspartate can be delivered safely and specifically to cancer cell lines by antibody-directed liposomes (Heath et. al., 1983; Paphadjopoulos et. al., 1985). Huang et. al. (1981) have presented another way of attaching antibody to liposomes. Their method involves the use of previously modified antibody and no oxidization of the liposomes. The advantage of this method, put forth by Huang et. al. (1982), is that the liposomes, and more importantly the often chemically fragile liposomal contents, are not subjected to the potential damage that the previously described coupling methods may produce. In this formulation antibody is covalently modified by the addition of fatty acid residues to its Fc end (Huang et. al., 1980). The modified antibody is then allowed to react with slightly destabilized liposomes produced by reverse-phase evaporation (Shen et. al., 1982). The product of this reaction is then stabilized by the removal, by dialysis, of the destabilizing emulsifier and residual organic solvents. The result of this process is the formulation of liposomes which safely encapsulate materials and externalize a specific immunoglobulin. Development of Amphotericin B Liposomes Bearing Antibody Specific to Candida albicans Duane R. Hospenthal, Alvin L. Rogers and Gary L. Mills Published in Mycopathologia 101: 37-45 (1988) 43 44 Development of Amphotericin B Liposomes Bearing Antibody Specific to Candida albicans. Duane R. Hospenthall, Alvin L. Rogerslrz, and Gary L. Millsl Departments of 1Botany and Plant Pathology, 2Microbiology and Public Health and Medical Technology Program, Michigan State University, East Lansing, MI 48824, USA Keywords: amphotericin B, Candida albicans, liposome Summary. Liposomes expressing external antibody specific for Candida albicans auui encapsulating amphotericin B were developed and characterized in this study. Antibody was first modified by the covalent attachment of palmitic acid residues. ZLiposomes were produced by reverse-phase evaporation and modified antibody was incorporated into these liposomes via the hydrophobic interaction between the palmitic acid and the phospholipids composing the liposomes. The liposomes were characterized as to the amount of amphotericin B by spectroscopy and for the 45 puesence of antibody by protein analysis and secondary immunolabeling by fluorescent and electron microscopic methods. Immunogold labeling showed that the antibody was being expressed externally on the liposomes in the electron microscopic studies and the specificity of these liposomes for g; albicans was observed by secondary immunofluores-cence. 'Address for offprints: Duane R. Hospenthal, Department of Botany and Plant Pathology, Michigan State University, East Lansing, MI 48824, USA' Introduction Candida albicans is the causative organism of greater than eighty percent of all fungal infections. It is responsi- ble for thirteen percent of fatal infections in lymphoma patients and over twenty percent of those in leukemia patients (3). .Although several new antifungal agents are being used, amphotericin B remains the drug of choice for most systemic fungal infections including candidiasis. Until recently the use of this antifungal agent has been limited due to its acute and chronic toxicity. However, it has now been established that encapsulation of amphotericin B in liposomes can reduce the toxic effects of the drug without decreasing its efficacy (6,8). The best characterized and most promising results reported are 46 from a particular phospholipid liposome formulation (7:3 molar ratio of dimyristoyl phosphatidylcholine to dimyristoyl phosphatidylglycerol) (6,8,9,22). This formulation has been tested successfully in laboratory animals (12,13,14, 16) as well as iriaa group of hematologic malignancy patients with systemic mycoses which previously have shown no response to amphotericin B treatment (11,19). The results of these and other studies show a large potential for future use of liposomal amphotericin B treatment. Liposomes containing drugs other than amphotericin B and nmdified with antibody have also been shown to In; useful in the specific delivery of these drugs and other macromolecules to antigenic sites on cell and tissues (5). For example, the cytotoxic action of actinomycin D encapsulated in antibody bearing liposomes is selective towards only those cells bearing the corresponding antigen (4). An important prerequisite for producing liposomes which are directed by antibody is to attach the immunoglobulins to the phospholipid vesicles without damaging or losing the substance which is being encapsu- lated. Withthis in mind we have adopted a procedure in which palmitic acid residues are covalently bound to antibody and then these modified immunoglobulins are inserted into the surface of preformed liposomes (7,18). 47 Materials and Methods Antibody modification Adsorbed rabbit antisera to g; albicans (Difco, Detroit, MI) was reacted with N-hydroxysuccinimide ester of palmitic acid (NHSP) to produce antibody with covalently bound palmitic acid residues (Ab-P) as previously described by Huang et al (7,18). The NHSP was prepared by the method of Lapidot et a1 (10) with 40 uCi 1-14c- palmitic acid per mmol palmitic acid included to follow the reacticnu. Equal molar concentrations of N-hydroxy- succinimide and palmitic acid were reacted in the presence CM? 10 mice were treated with a single injection two days after the initiation of the infection with g_._ albicans and observed 21 days for survival. Liposome and free drug preparations were incubated 20 min at 37 C prior to injection into the caudal veins of the mice to further reduce the toxicity of the liposomal compounds (Szoka et al., 1987). Treatment dosages of 0.6 anui 1.2 mg AMB/kg mouse of fAMB, LAMB, and LAMB-Ab were compared. These dosages were selected due to the fact that the maximum tolerated dose of fAMB has been reported by Lopez-Berestein et al. (1983) to be 0.8 mg AMB/kg in the mouse size used in this study. 98 Statistical analysis The 21 day survival patterns of mice treated with the AMB preparations and untreated controls were compared at each dose employing a generalized Wilcoxon test (Gehan, 1965). Results Canidiasis Model The amount of viable Q; albicans in the kidneys, liver, and spleen of untreated mice was examined each day through the first 7 days of the study. Figure 1 shows the results of viability counts in the kidneys and livers of these mice. Spleen homogenates of all mice in the study pmoduced less than seven colonies (usually zero) upon plating of the smallest dilution (1:5) and thus were not included in this figure. Mice which died on days 6 and 7 of the infection had 9; albicans concentrations similar to the mice sacrificed on these days, approximately 104 cfu in the liver and 106 cfu in the kidneys per mouse. The size and gross morphology of kidneys at similar days of the infection (in both matched pairs and between mice) were quite varied, even among tflua treated mice. Individual kidneys were atrophied, enlarged with or without purulent discharge, nodular with granulomatous 99 material, or normal appearing. Liver and spleen gross morphology showed much more consistency in this study. Therapy of candidiasis The combined 21 day survival pattern of two studies of mice treated with 0.6 mg AMB/kg is shown in Figure 2. All amphotericin B formulations improved the survival of mice over the control at this dose (P < 0.001). LAMB—Ab and LAMB increased the survival over the free drug (P < 0.001 and P < 0.03, respectively). Survival distribution between the LAMB and the LAMB-Ab groups of mice was also significant (P < 0.007). Additional mice, which were infected and treated parallel to this study, were sacrificed at days 7 and 14 of the infection. At day 7, all three groups of treated mice (fAMB, LAMB, and LAMB-Ab) produced no recoverable cfu from their spleens or livers, auui all had similar recovery values of 104 cfu in their kidneys. Day 7 colony counts in the untreated mice are incorporated into the data presented in Figure 1. Fourteen days after infection, LAMB treated mice had 107 cfu in their kidneys and 105 cfu in their livers. At this time, mice treated with LAMB-Ab had recoveries of 107 cfu and 104 cfu in their kidneys and livers, respectively. Due to the survival rates of the untreated and fAMB groups (Figure 2), mice from these groups did not survive to be sampled at day 14. 100 Surviving mice from one 0.6 mg AMB/kg study were examined at day 42 for residual candidal infection. Of this group, which consisted of 1 fAMB, 2 LAMB, and 5 LAMB- Ab mice, all the organs were negative for growth except for two of the LAMB-Ab mice. These two mice had livers which grew 103 cfu and kidneys which grew 106 cfu. Experimental therapy studies at the dose of 1.2 mg AMB/kg proved, as previously seen in our peritoneal murine candidiasis model (Hospenthal et al., 1989), to show little separation of the two liposomal compounds. The 21 day survival of mice at this dose was 64.7% for LAMB-Ab and 61.1% for LAMB. The survival of the LAMB—Ab and LAMB groups did prove to better than the fAMB groups (P < 0.02 and P < 0.03, respectively). Discussion Liposomal amphotericin B which bears anticandidal antibody, LAMB-Ab, was shown to produce increased survival in mice infected with Candida albicans at a dose (0.6 mg AMB/kg) which is below the maximum tolerated dose of Fungizone in these mice. Our formulation of LAMB at this dose also produces an increase in survival over the free drug (fAMB), though much less than that of the LAMB-Ab treatment. Murine survival studies of these liposomal amphotericin B compounds at a dosage of 1.2 mg AMB/kg 101 also show these compounds to be more effective than fAMB. LAMB at this dose increases the survival rate of mice to that of LAMB-Ab. LAMB-Ab at this dose shows no change in survival rate over that which it produces at 0.6 mg AMB/kg. The authors feel this may be due to problems associated with the 90% nonspecific protein content of the antisera which is used in the preparation of LAMB-Ab. Future studies employing a monoclonal anticandidal antibody will be neccessary to explore this detail. The organ homogenization study of this model of systemic candidiasis shows that at the time when untreated mice begin to die from this infection, six days after infection, they normally have 103 cfu in their livers and 106 cfu in their kidneys. The normal survival of these mice is 6 to 17 days post infection. Mice sacrificed in the 0.6 mg AMB/kg LAMB and LAMB-Ab groups at 14 days produced cfu counts which averaged 10 times greater than those produced by the untreated control. This may indicate that mice surviving longer, may be more resistant to the infection, and yet may harbor large numbers of the yeast. The encapsulation of amphotericin B in liposomes has been proven to reduce the toxic effect of this antifungal in numerous studies (Juliano et al., 1987; Mehta et al., 1984; Shirkhoda et al., 1986; Szoka et al., 1987). In these studies, this reduction in toxic effect has allowed 102 increasingly larger doses of AMB to be employed, and thus has improved the clinical effectiveness of this drug. In our studies, we have shown that our formulation, LAMB-Ab, has increased the survival of mice with disseminated candidiasis as compared to fAMB when administered in single, low dosages. In previous studies we have shown that this compound is less toxic than fAMB when administered to noninfected mice or incubated with human red blood cells. Dissemination of fungal cells which are susceptible to AMB is thought to be largely via the circulatory system. If our compound, or other prepar- ations, can produce antibody-antigen targeting of drugs to infectious agents, then perhaps lower dosages of drugs with toxic side effects can be employed successfully in the treatment of these diseases. The combination of this targeting and the reduction in toxicity that liposomes provide could improve the efficacy of many toxic compounds, including that of anmhotericin B in the treatment of life-threatening mycoses. Acknowledgement This research was supported in part by a grant from Michigan Health Care Education and Research Foundation, Inc. (grant number: 036-SAP/88—05) under its Student Awards Program. 103 References Bannatyne R M, Cheung R 1977 Susceptibility of Candida albicans to amphotericin B and amphotericin B methyl ester. Antimicrobial Agents and Chemotherapy 12:449- 450. Bodey G P 1984 Candidiasis in cancer patients. American Journal of Medicine 77:13-19. Gehan E A 1965 A generalized Wilcoxon test for comparing arbitrarilgr single-censored samples. ‘Biometrika 52:203-217. Graybill J R, Craven P C 1983 Antifungal agents used in systemic mycoses: activity and therapeutic use. Drugs 25:41-62. Heath T D, Montgomery J A, Piper J R, Papahadjopoulos D 1983 Antibody-targeted liposomes: increase in specific toxicity of methotrexate-gamma-aspartate. Proceedings of the National Academy of Sciences USA 80:1377-1381. 104 Hospenthal D R, Rogers A L, Beneke E s 1989 Effect of attachment of anticandidal antibody to the surfaces of liposomes encapsulating amphotericin B in the treatment of murine candidiasis. Antimicrobial Agents and Chemotherapy 33:16-18. Hospenthal D R, Rogers A L, Mills G L 1988 Development of amphotericin B liposomes bearing antibody specific to Candida albicans. Mycopathologia 101:37-45. Juliano R L, Grant C W M, Barber K R, Kalp M A 1987 Mechanism of the selective toxicity of amphotericin B incorporated into liposomes. Molecular Pharmacology 31:1-11. Juliano R L, Lopez-Berestein G, Hopfer R, Mehta R, Mehta K, Mills K 1985 Selective toxicity and enhanced therapeutic index of liposomal polyene antibiotics in systemic fungal infections. Annals of tine New York Academy of Sciences 446:390—402 Kirsh R, Goldstein R, Tarloff J, Parris D, Hook J, Hanna N, Bugelski P, Poste C; 1988 An emulsion of amphotericin B improves the therapeutic index when treating sysytemic murine candidiasis. Journal of Infectious Diseases 158:1065-1070. 105 Leserman L D, Machy P, Devaux C, Barbet J 1983 Antibody- bearing liposomes: targeting in vivo. Biology of the Cell 47:111-116. Lopez—Berestein G, Bodey G P, Frankel L S, Mehta K 1987 Treatment of hepatosplenic candidiasis with liposomal amphotericin Bu .Journal of Clinical Oncology 5:310— 317. Lopez-Berestein G, Fainstein V, Hopfer R, Mehta K, Sullivan M P, Keating M, Rosenblum M G, Mehta R, Luna M, Hersh E M, Reuben J, Juliano R L, Bodey G P 1985 Liposomal amphotericin B for the treatment of systemic fungal infections in patients with cancer: a preliminary studyu .Journal of Infectious Diseases 151:704-710. Lopez-Berestein G, Mehta R, Hopfer R L, Mills K, Kasi L, Mehta K, Fainstein V, Luna M, Hersh E114, Juliano R 1983 'Treatment enui prophylaxis of disseminated infection due to Candida albicans in mice with liposome-encapsulated amphotericin B. Journal of Infectious Diseases 147:939-945. 106 Mehta R, Lopez—Berestein G, Hopfer R, Mills K, Juliano R L 1984 Liposomal amphotericin B is toxic to fungal cells but not to mammalian cells. Biochimica et Biophysica Acta 770: 230-234. Efidrkhoda A, Lopez-Berestein G, Holbert J M, Luna M A 1986 Hepatosplenic fungal infections: CT and pathological evaluation after treatment with liposomal amphotericin B. Radiology 159:349-353. Szoka F C, Milholland D, Barza M 1987 Effect of lipid composition and liposome size on toxicity and in vitro fungicidal activity of liposome-intercalated amphotericin B. Antimicrobial Agents and Chemotherapy 31:421-429. 107 w .b. 01 o~ ‘1 1 l 1 l _l Log“, Colony Forming Units 53 1 Days Figure l. Candida albicans present in the kidneys (O) and livers (A) of untreated mice sacrificed during the first 7 days of infection. Animals were infected with 3 X 105 cfu of the yeast. Yeast concentrations less than 10910 cfu of 2 were below the sensitivity of the study design. 108 100 Percent Survival b 0‘ co 9 0 9 to O as O 7 l4 2] Days Figure 2. The effect produced by anticandidal antibody bearing liposomal AMB on the survival of mice infected with Candida albicans. Two days after initiation of the infection the animals were treated with a single dose of 0.6 mg AMB/kg as LAMB—AB (I), LAMB (A), fAMB (0), or not treated (V) . SUMMARY The initial aim of this project was to determine if the toxicity of amphotericin B could be reduced and the therapeutic effect increased by the encapsulation of this polyene antifungal agent in phospholipid vesicles which bear antibody to etiologic agents of deep mycoses. The fungal agent choosen for this project was Candida albicans, time most common of the fungal pathogens which infect humans. The liposomal amphotericin B which bore anticandidal immunoglobulins was tested both lg ylllg, and lg ylyg in two murine models of candidiasis. Formulation of the liposomal drug bearing antibodies was influenced primarily by work of Lopez—Berestein et. al. (1984; 1983a) in the encapsulation of amphotericin B within liposomes and that of Huang et. al. (1982; 1981; 1980) and Shen et. al. (1982) in the modification and incorporation of antibody onto the surfaces of liposomes produced by reverse-phase evaporation. Combining features of liposome technology presented by these groups with various original modifications, including 109 110 methodology to produce much smaller vesicles, liposomes encapsulating amphotericin B and bearing external antibody to g; albicans were produced. In Article I, formulation and lg ylggg testing of this preparation, LAMB—Ab, was reported. This preparation was visualized and shown to bear external antibody by the use of transmission electron microscopy with secondary immunogold labeling. Secondary immunofluorescence was employed to visualize that the antibody which was externalized on LAMB-Ab was still active against 9; albicans at the end of preparation. Toxicity and antifungal activity of LAMB-Ab lg ylggg was compared with the free drug (fAMB) and the liposomal drug without antibody (LAMB). No toxicity to human red blood cells was apparent at concentrations of 100 ug/ml of the two liposomal preparations whereas fAMB lysed these erythro- cytes at a concentration of 12.5 ug/ml. The anticandidal effect (HE these three compounds was essential the same, with LAMB-Ab showing a slight advantage over LAMB and fAMB. Testing of LAMB-Ab in an 1 vivo mouse model was the topic of Article II. Also examined in this article was the effect which was due to the attachment of the antibody and that due to the specificy of this immunoglobulin. In this intraperitoneal-initiated model of candidiasis LAMB- Ab proved to be more effective at lower dosages than fAMB 111 or LAMB in both treatment and prophylaxis of murine candidiasis. Survival rates of mice treated with liposomal amphotericin B preparations containing unat- tached specific antibody or with attached nonspecific antibody in,this therapy model supported tine conclusion that the improved survival in LAMB—Ab treated mice was due to specific, attached immunoglobulin. Toxicity of the liposomal compound was tested grossly 1 vivo in this study. No toxicity, in the form of visual signs or death, was seen with the injection of 2.0 mg AMB/kg LAMB—Ab in uninfected mice. A new mouse model was examined in Article III. In this study, a intravenous—initiated murine model of candidiasis was characterized by the counting of viable g; albicans in the spleen, liver, and kidneys of the infected mice. Preliminary results in this model showed at a treatment dosage of 0.6 mg AMB/kg, LAMB-Ab was more effective than fAMB and LAMB. All of these compounds had virtually identical counts of viable yeast in their organs at similar days of the infection. Though these numbers were substantially different than the control mice, they did not reflect the difference in survival afforded by the three compounds. At the larger dose of 1.2 mg AMB/kg, LAMB-Ab was still more effective than the other two preparations, but showed no significant increase over that which it produced at 0.6 mg AMB/kg. The fact that the 112 antibody preparation used in formulating these liposomes contained only 10% specific 9; albicans antibody, is believed to be of some importance here. Assuming that the average antisera contains about 80% albumin, this protein could be causing an deleterious effect at higher dosages o f LAMB-Ab . To improve the effect of LAMB-Ab in future projects, and better ascertain this effect, the usetcxf monoclonal antibody would seem to be the next logical step. Through this use, with proper controls, one could more fully understand the effect produced by binding antibody to liposomal amphotericin B. Labeling of this antibody could also prove useful in following the localization of LAMB-Ab in the murine model. Ultimately, LAMB-Ab produced with polyclonal antibody (perhaps a carefully selected mixture of monoclonal antibodies) should be thoroughly examined. The goal being to produce an anticandidal formulation with a wide spectrum of action against the many strains of C; albicans. Using the monoclonal antibody—bearing liposomal ammmotericin B, the first step should be to test this compound an: previously used dosages, 0.6 and 1.2 mg AMB/kg, to compare this formulation to those cited in this dissertation. Along with this, more extensive organ counts of viable yeasts are needed to further explore the effect of LAMB—Ab on the mouse model. Increasingly higher 113 dosages also deem exploration to establish tine toxicity and efficacy of this compound and the dose at which actual cure is effected. Multiple dosing therapies also merit studyu .Since clinically, free amphotericin B is given in multiple dosages over an extended period of time, this should also be examined in the murine model. Extending the murine model to more closely match the clinical reality should also include studies with neutropenic mice. In the clinical setting a vast number of the life-threatening mycoses which are treated with amphotericin B are in patients with hematologic malignan- cies and other compromised conditions. Another possible future application of liposomal amphotericir113 bearing antibody could easily be the preparation of this compound with specific antibodies to the other mycoses treated with the antifungal drug. The possibility of LAMB-Ab containing a mixture of differing antifungal immunoglobulins could also be explored. The final and most broad possibility is the application of this technology to encapsulate any compatible drug and bear any antibody which is specific to antigen which would improve the targeting of that drug. The production and preliminary testing of LAMB-Ab has shown promise and success in the reduction of toxicity of amphotericin B and the increase in efficacy of the 114 antifungal compound. Though much is left that could (u: should be done, the groundwork of this research has been established and the path set. BIBLIOGRAPHY Adriano, S. M. and J. Schwarz. 1955. Experimental moniliasis in mice. Am. J. Pathol. 31: 859-873. Ahrens, J., J. R. Graybill, P. C. Craven and R. L. Taylor. 1984. Treatment of experimental murine candidiasis with liposome—associated amphotericin B. Sabouraudia 22: 163-166. Alving, C. R. 1987. Liposomes as carriers for vaccines. 13114. J. Ostro [ed]. Liposomes. From Biophysics to Therapeutics. Marcel Dekker,Inc., New York, pp. 195- 218. Bachhawat, B. K. and P. Dasgupta. 1986. Approadh to tissue targeting of drugs and proteins using glycolipid liposomes. In K. Yagi [ed]. Medical Application of Liposomes. Japan Scientific Societies Press, Tokyo, pp. 81—90. Bangham, A. Eh 1980. Development of the liposome concept. In G. Gregoriadis and A. C. Allison [eds]. Liposomes in Biologlcal Systems. John Wiley & Sons, New York, pp. 1-24. Bangham, A. D., M. M. Standish and J. C. Watkins. 1965. Enffusion of univalent ions across the lamellae of swollen phospholipids. J. Mol. Biol. 13: 238-252. Becker, J. M., S. Marcus, J. Tullock, D. Miller, E. Krainer, R. K. Kahare and F. Naider. 1988. Use of the chitin-synthesis inhibitor nikkomycin to treat disseminated candidiasis in mice. J. Infect. Dis. 157: 212-214. Bendayan, M. 1984. Protein A-gold electron microscopic inmmnocytochemistry: Methods, applications, and limitations. J. Electron Microscopy Technique 1: 243-270. 115 116 Bannatyne, R. M. and R. Cheung. 1977. Susceptibility of Candida albicans to amphotericin B and amphotericin B methyl ester. Antimicrob. Agents Chemother. 12: 449- 450. Beneke, E. S., J. W. Rippon and A. L. Rogers. 1984. Human Mycoses. 8th ERL. The Upjohn Co., Kalamazoo, Michigan, 88pp. Bodey, G. P. 1984. Candidiasis in cancer patients. Am. Bolard, J., M. Cheron and J. Mazerski. 1984. Effect of surface curvature on the interation of single lamellar phospholipid vesicles with arommatic and nonaromatic heptaene antibiotics (vacidin A and amphotericin B). Biochem. Pharmacol. 33: 3675-3680. Brajtburg, J., G. Medoff, G. S. Kobayashi and S. Elberg. 1980. Influence of extracellular K+ or Mg2+ on the stages of the antifungal effects of amphotericin B and filipin. 131.]. D. Nelson and C. Grassi [eds]. Current Chemotherapy and Infectious Disease. Vol. II. Amer. Soc. Microbiol., Washington, DC, pp. 970-972. Brandriss, M. W., S. M. Wolff, R. Moores and F. Stohlman. 1964. Anemia induced by amphotericin B. J. Am. Med. Assoc. 189: 663-665. Brass, C., J. N. Galgiani, T. F. Blaschke, R. deFelice, R. A. O'Reilly and D. A. Stevens. 1982. Disposition of ketoconazole, an oral antifungal, in humans. Antimicrob. Agents Chemother. 21: 151—158. Bullock, W. E. and D. Bathena. 1976. Can mannitol reduce amphotericin B nephrotoxicity? Double-blind study and description of a new vascular lesion in the kidneys. Antimicrob. Agents Chemother. 10: 555-563. Burford-Mason, A. P., J. C. P. Weber and J. M. T. Willoughby. 1988. Oral carriage of Candida albicans, ABO blood group and secretor status in healthy subjects. J. Med. Vet. Mycol. 26: 49-56. Butler, W. T., J. E. Bennett, D. W. Alling, P. T. Wertlake, J. P. Utz and G. J. Hill. 1964. Nephrotoxicity of amphotericin B; Early and late effects in 81 patients. Ann. Int. Med. 61: 175-179. Carmo-Sousa, L. do. 1969. Distribution of yeasts in nature. In A. H. Rose and J. S. Harrison [eds ]. The Yeasts. Vol. 1. Academic Press, London, pp. 79-105. 117 Chalkley, L. J., A. P. J. Trinci and A. M. S. Pope. 1985. Effect of mycolase and amphotericin B on Candida albicans and Candida pseudotropicalis lg vitro and lg v1vo. Sabouraudia 23: 147-164. Chapman, H. A. and J. B. Hibbs. 1978. Modulation of macrophage tumorcidal capability by polyene antibiotics: Support for membrane lipid as a regulatory determinant of macrophage function. Proc. Natl. Acad. USA 75: 4349-4353. Clejan, S. and R. Bittman. 1985. Rates of amphotericin B and filipin association with sterols. J. Biol. Chem. 260: 2884-2889. Cohen, B. E. 1983. The nature of the channels induced by amphotericin B in liposomes. In A. D. Bangham [ed]. Liposome Letters. Academic Press, London, pp. 127-135. Cybulska, B., M. Herve, E. Borowski and C. M. Gary-Bobo. 1986. Effect of the polar head structure of polyene macrolide antifungal antibiotics on the mode of permeabilization of ergosterol- and cholesterol- containing lipidic vesicles studied by 31P-NMR. Mol. Pharmacol. 29: 293-298. Dennis, D. L., C. G. Peterson and W. S. Fletcher. 1968. Candida septicaemia in the severely traumatized patient. J. Trama 8: 177-185. Difco Laboratories. 1984. Bacto Candida albicans antisera. In Difco Manual. 10th Edition. Difco Laboratories Inc., Detroit, MI, pp. 196-198. Drouhet, EL 1970. Basic Mechanisms of antifungal chemotherapy. In J. P. Utz [ed]. Modern Therapy. Vol. 7: Treatment of the Systemic Mycoses. Harper & Row, New York, pp.539-564. Edwards, (I. E., Jr., R. I. Lehrer, E. R. Stiehm, T. J. Fischer and L. S. Young. 1978. Severe candidal infections: Clinical perspective, immune defense mechanisms, and current concepts of therapy. Ann. Intern. Med. 89: 91-106. Ellis, W. G., R. A. Sobel and S. L. Nielsen. 1982. Leukoencephalopathy in patients treated with amphotericin B methyl ester. J. Infect. Dis. 146: 125-137. 118 Emmons, C. W., C. H. Binford, J. P. Utz and K. J. Kwon- Chung. 1977. Medical Mycology. 3rd Edition. Lea & Febiger, Philadelphia, 592 pp. Feld, R., G. P. Bodey, V. Rodriguez and M. Luna. 1974. Cause of death in patients with malignant lymphoma. Am. J. Med. Sci. 268: 97-106. Fraser-Smith, E. B., D. A. Eppstein, M. A. Larsen and T. R. Matthews. 1983. Protective effect of a muramyl dipeptide analog encapsulated in or mixed with liposomes against Candida albicans infection. Infect. Immun. 39: 172-178. Gehan, E. A. 1965. A generalized Wilcoxon test for comparing arbitrarily single-censored samples. Biometrika 52: 203-233. Gold, W., H. A. Stout, J. F. Pagano and R. Donovick. 1956. Amphotericins A and B, antifungal antibiotics produced by a streptomycete. Antibiotics A. 1956: 579-586. Gouge, T. H. and V. T. Andriole. 1971. An experimental model of amphotericin B nephrotoxicity with renal acidosis. J. Lab. Clin. Med. 78: 713-726. Graybill, J. R. and J. Ahrens. 1984. R 51211 (itraconazole) therapy of murine cryptococcosis. Sabouraudia 22: 445-453. Graybill, J. R. and P. C. Craven. 1983. Antifungal agents used in systemic mycoses: Activity and therapeutic use. Drugs 25: 41-62. Graybill, J. R., P. C. Craven, R. L. Taylor, D. PL Williams and W. E. Magee. 1982. Treatment of murine cryptococcosis with liposome-associated amphotericin B. J. Infect. Dis. 145:748-752. Graybill, J. R., S. H. Sun and J. Ahrens. 1986. Treatment.