M. A. Ghannoum

International Conference for the Development of a Consensus on the Management and Prevention of Severe Candidal Infections

Because of the rapidly increasing incidence of serious candidal infections, a consensus conference of 22 investigators from the United States, Europe, and Japan was held to discuss strategies for the prevention and treatment of deep-organ infections caused by Candida species. Commonly asked questions concerning the management of candidal infections were selected for discussion by the participating investigators. Possible answers to the questions were developed by the investigators, who then voted anonymously for their preferences. In certain instances, unanimity or a strong consensus was the result. In all cases, the full spectrum of responses was recorded and is presented in this report. The forms of candidal infection addressed included candidemia, candiduria, hepatosplenic candidiasis (chronic systemic candidiasis), candidal endophthalmitis, and candidal peritonitis. Prevention and treatment strategies were considered for patients who have undergone surgery, for neutropenic and nonneutropenic patients, and for patients who have undergone bone marrow and solid organ transplantation. The therapeutic roles of amphotericin B (standard and lipid formulations) and the azoles were considered.

Crude oil and hydrocarbon-degrading strains of Rhodococcus rhodochrous isolated from soil and marine environments in Kuwait

Soil and marine samples collected from different localities in Kuwait were screened for microorganisms capable of oil degradation. Both fungi and bacteria were isolated. The fungal flora consisted of Aspergillus terreus, A. sulphureus, Mucor globosus, Fusarium sp. and Penicillum citrinum. Mucor globosus was the most active oil degrading fungus isolated. Bacterial isolates included Bacillus spp. Enterobacteriaceae, Pseudomonas spp., Nocardia spp., Streptomyces spp.,and Rhodococcus spp. Among these Rhodococcus strains were the most efficient in oil degradation and, relatively speaking, the most abundant. Bacterial and fungal isolates differed in their ability to degrade crude oil, with Rhodococcus isolates being more active that fungin in n-alkane biodegradation, particularly in the case of R. rhodochrous. In addition to medium chain n-alkanes, fungi utilized one or more of the aromatic hydrocarbons studied, while bacteria failed to do so. R. rhodochorous KUCC 8801 was shown by GLC and post-growth studies to be more efficient in oil degradation than isolates known to be active oil degraders.

Studies on the Anticandidal Mode of Action of Allium sativum (Garlic)

The mode of action of aqueous garlic extract (AGE) was studied in Candida albicans. The minimum inhibitory concentration (MIC) of AGE against six clinical yeast isolates ranged between 0.8 and 1.6 mg ml-1. Scanning electron microscopy and cell leakage studies showed that garlic treatment affected the structure and integrity of the outer surface of the yeast cells. Growth of C. albicans in the presence of AGE affected the yeast lipid in a number of ways: the total lipid content was decreased; garlic-grown yeasts had a higher level of phosphatidylserines and a lower level of phosphatidylcholines; in addition to free sterols and sterol esters, C. albicans accumulated esterified steryl glycosides; the concentration of palmitic acid (16:0) and oleic acid (18:1) increased and that of linoleic acid (18:2) and linolenic acid (18:3) decreased. Oxygen consumption of AGE-treated C. albicans was also reduced. The anticandidal activity of AGE was antagonized by thiols such as L-cysteine, glutathione and 2-mercaptoethanol. Interaction studies between AGE and thiols included growth antagonism, enzymic inhibition and interference of two linear zones of inhibition. All three approaches suggest that AGE exerts its effect by the oxidation of thiol groups present in the essential proteins, causing inactivation of enzymes and subsequent microbial growth inhibition.

High-temperature hydrocarbon degradation by Bacillus stearothermophilus from oil-polluted Kuwaiti desert

Kuwaiti desert samples contaminated with crude oil contained Bacillus stearothermophilus strains capable of growth on crude oil as a sole source of carbon and energy, obligately at high temperature. No thermophilic oil utilizers were present in water samples collected from the Arabian Gulf. Most of the desert strains had an optimum temperature of 60°C and grew best on pentadecane (C15), hexadecane (C16) and heptadecane (C17). n-Alkanes with shorter and longer chains, n-alkenes, and aromatic hydrocarbons were less readily utilized.

Pathogenicity determinants of Candida.

The incidence of infections due to Candida albicans and other related species has increased in recent years. A number of factors have contributed to this, e.g. the use of a wide range of potent antibacterial and immunosuppressive therapeutic agents and the increased incidence of immune‐deficiency diseases such as AIDS. Pathogenicity determinants which confer virulence on C. albicans, and other Candida species to a lesser extent, have been reviewed. These include factors related to species and strains, adherence, dimorphism, toxin and enzyme production and cell surface composition. This review clearly shows that C. albicans virulence is a function of a multiplicity of factors working jointly to overcome the host defences. A lack or debility in any of these parameters will reflect negatively on its infectivity and make it difficult for Candida to establish itself, particularly in a healthy individual.

In vitro determination of optimal antifungal combinations against Cryptococcus neoformans and Candida albicans.

There is currently no rapid, reliable, and reproducible in vitro technique to describe the growth-inhibitory interactions of antifungal drug combinations over a wide range of drug concentrations. We have developed a microdilution plate assay that was used to determine optimal drug combinations and concentrations of one-, two-, and three-drug regimens of amphotericin B (AmphB), fluconazole (FLU), and 5-fluorocytosine (5FC) for growth inhibition of three isolates each of Cryptococcus neoformans and Candida albicans. These growth inhibition data were then used in a multifactorial design technique to (i) generate contour and surface response plots to aid visual interpretation and (ii) develop mathematical equations describing the growth responses of the fungi to a wide range of antifungal concentrations and ratios. Our data indicated that (i) antifungal drug-drug interactions affecting yeast growth are complex functions of the drugs used in combination, their absolute concentrations, and also their relative (proportional) concentrations; (ii) AmphB-FLU combinations had additive effects against C. albicans over wide concentration ranges for each agent but were indifferent (i.e., were less than additive) in their inhibitory effect on C. neoformans; (iii) other two-drug combinations (FLU-5FC or AmphB-5FC) had indifferent effects on the growth of both fungi; and (iv) three-drug combinations (AmphB-FLU-5FC) showed an additive inhibitory effect on the growth of both C. albicans and C. neoformans. The finding that no antagonism was observed in combinations employing AmphB and FLU in this in vitro model is of critical importance since it argues against the current theoretical concept, based on the individual drugs mode of action, of antagonism between these two drugs. These microdilution techniques provide a method to determine rational regimens of antifungal agents in multidrug combinations for future testing to correlate in vitro activity with in vivo response. The use of this approach has made the evaluation of complex antifungal drug-drug interactions possible and provided important new information to the evolving field of antifungal drug combination.

Thrombin-induced rabbit platelet microbicidal protein is fungicidal in vitro.

Platelet microbicidal protein (PMP) is released from platelets in response to thrombin stimulation. PMP is known to possess in vitro bactericidal activity against Staphylococcus aureus and viridans group streptococci. To determine whether PMP is active against other intravascular pathogens, we evaluated its potential fungicidal activity against strains of Candida species and Cryptococcus neoformans. Anionic resin adsorption and gel electrophoresis confirmed that the fungicidal activity of PMP resided in a small (approximately 8.5-kDa), cationic protein, identical to previous studies of PMP-induced bacterial killing (M.R. Yeaman, S.M. Puentes, D.C. Norman, and A.S. Bayer, Infect. Immun. 60:1202-1209, 1992). When assayed over a 180-min period in vitro, the susceptibilities of these fungi to PMP varied considerably. Generally, Candida albicans strains (mean survival, 33.5% +/- 6.9% [n = 6]) as well as isolates of Candida glabrata (mean survival, 50.8% +/- 2.9% [n = 2]) were the most susceptible to killing by PMP, while Candida guillermondii and Candida parapsilosis were relatively resistant to PMP-induced killing. Compared with C. albicans, C. neoformans was relatively resistant to the fungicidal activity of PMP, with a mean survival among the isolates studied of 77.4% +/- 12.4% (n = 6). Against C. albicans, PMP-induced fungicidal activity was time dependent (range, 0 to 180 min), PMP concentration dependent (range, 10 to 150 U/ml), and inversely related to the fungal inoculum (range, 5 x 10(3) to 1 x 10(5) CFU/ml). Scanning electron microscopy of PMP-exposed C. albicans and C. neoformans cells revealed extensive surface damage and collapse, suggesting that the site of PMP fungicidal action may directly or indirectly involve the fungal cell envelope. Images

Dimorphism-associated Variations in the Lipid Composition of Candida albicans

Yeast and mycelial forms of Candida albicans ATCC 10231, growing together in 12 h and in 96 h cultures, were separated and their lipids were extracted and characterized. The total lipid content of the yeast forms was always lower than that of the mycelial forms. In 12 h cultures the lipids from the two morphological forms consisted mainly of polar compounds, viz, phospholipids and glycolipids. In 96 h cultures both the yeast and mycelial forms accumulated substantial amounts of apolar compounds, mainly steryl esters and triacylglycerols. The mycelial forms were more active than the yeast forms in this respect. Major differences in the lipid composition between the two morphological forms involved the contents of sterols and complex lipids that contain sterols. As a rule, the yeast lipids contained much larger proportions of free sterols than the mycelial lipids. However, the mycelial lipids contained several times more sterols than the yeast forms but bound as steryl glycosides, esterified steryl glycosides and steryl esters. Steryl glycosides and esterified steryl glycosides occurred in yeast lipids only in traces, if at all. The major steryl glycoside in the mycelial forms was unequivocally identified as cholesteryl mannoside. At both phases of growth the apolar and polar lipid fractions from the mycelial forms contained higher levels of polyunsaturated fatty acids (18:2 and 18:3) but lower levels of oleic acid (18:1) than the corresponding fractions from the yeast forms. The lipid content and composition of 12 h and 96 h yeast and mycelial forms of C. albicans KCCC 14172, a clinical isolate, were almost identical with those of C. albicans ATCC 10231.

Sterol compositions and susceptibilities to amphotericin B of environmental Cryptococcus neoformans isolates are changed by murine passage.

Previous studies have shown that sequential isolates from patients with persistent Cryptococcus neoformans meningoencephalitis can vary in sterol composition and susceptibility to antifungal drugs. To investigate the potential of host factors as mediators of this phenomenon, we compared fungal susceptibilities of environmental and clinical isolates from a limited geographic area. Clinical isolates were less susceptible to amphotericin B than environmental isolates. Five environmental isolates were passaged through BALB/c murine hosts; the passaged isolates had changes in sterol composition and reduced amphotericin B susceptibilities relative to those of the parent isolates. In contrast, murine passage of these isolates did not alter their susceptibilities to fluconazole. The results confirm that changes in sterol composition and antifungal susceptibility can occur in vivo as a result of host factors and suggest that human infection can result in selection of variants with reduced susceptibilities to amphotericin B.

Sterol composition of Cryptococcus neoformans in the presence and absence of fluconazole.

Analysis of the sterol compositions of 13 clinical isolates of the pathogenic yeast Cryptococcus neoformans obtained from five patients with recurring cryptococcal meningitis showed that, unlike Candida albicans, the major sterols synthesized by this yeast were obtusifoliol (range, 21.1 to 68.2%) and ergosterol (range, 0.0 to 46.5%). There was considerable variation in the sterol contents among the 13 isolates, with total sterol contents ranging from 0.31 to 5.9% of dry weight. The isolates from the five patients who had relapses had different total sterol contents and compositions in comparison with those of the pretreatment isolates, indicating either that the sterols had been changed by therapy or that the patients were infected with new isolates with different sterol compositions. Growth of the cryptococcal isolates in the presence of subinhibitory concentrations of fluconazole (0.25x the MIC) significantly altered the sterol content and pattern. The total sterol content decreased in nine isolates and increased in four isolates in response to pretreatment with fluconazole. Fluconazole had no consistent effect on ergosterol levels. In contrast, fluconazole caused a decrease in obtusifoliol levels and an increase in 4,14-dimethylzymosterol levels in all isolates. These results indicate extensive diversity in sterol content, sterol composition, and sterol synthesis in response to subinhibitory concentrations of fluconazole in C. neoformans strains. We propose that fluconazole inhibits the sterol synthesis of C. neoformans by interfering with both 14 alpha-demethylase-dependent and -independent pathways. No correlation between the sterol compositions of C. neoformans isolates and their susceptibilities to fluconazole was found.