The effect of serum albumin on amphotericin B aggregate structure and activity.

PURPOSE: Mild heat treatment of Fungizone (FZ, an amphotericin B:deoxycholate preparation) leads to a new self-associated form (HFZ) that demonstrates improved therapeutic index in vivo. The origin of the improvement may lie in the differential stability in the presence of serum proteins. The purpose of this study is to assess the effect of human serum albumin (HSA) on the structure and stability and in vitro channel forming ability of these two preparations against model fungal and mammalian membrane vesicles. METHODS: Kinetic absorption and CD spectroscopy were used to assess the kinetic and equilibrium stability of the characteristic amphotericin B complexes in the presence of HSA. Kinetic fluorescence spectroscopy of pyranine entrapped in model fungal and mammalian membrane vesicles was used to measure the cation-selective channel forming ability of HZ and HFZ delivered from HSA. RESULTS: It is shown that FZ is rapidly converted from its aggregated form to a protein-bound monomer in the presence of HSA, whereas HFZ demonstrates greater stability by persisting as a stable inactive aggregate. Fluorescence measurements of ion currents show that HSA attenuates the membrane-activity of both preparations. However, the activity of both HFZ and FZ remains significant against ergosterol-containing membranes. This is the first direct measurement of the intrinsic channel forming abilities of these amphotericin B preparations in the presence of serum proteins. CONCLUSION: These data provide a mechanistic rationale for the similar efficacy and lower toxicity of HFZ.

Heat treatment of amphotericin b modifies its serum pharmacokinetics, tissue distribution, and renal toxicity following administration of a single intravenous dose to rabbits.

The purpose of this investigation was to determine the serum pharmacokinetics, tissue distribution, and renal toxicity of amphotericin B (AmpB) following administration of a single intravenous dose (1 mg/kg of body weight) of Fungizone (FZ) and a heat-treated form of FZ (HFZ) to New Zealand White female rabbits. FZ solutions were heated at 70 degrees C for 20 min to produce HFZ. Blood samples were obtained before drug administration and serially thereafter. After collection of the 48-h blood sample, each rabbit was humanely sacrificed and the right kidney, spleen, lungs, liver, and heart were harvested for AmpB analysis. Serum creatinine levels were measured before and 10 h after drug administration. AmpB concentrations in the serum and tissues were analyzed using high-performance liquid chromatography. FZ administration to rabbits resulted in a greater-than-50% increase in serum creatinine concentrations compared to baseline. However, HFZ administration resulted in no difference in serum creatinine concentrations compared to baseline. The AmpB area under the concentration-time curve (AUC) after HFZ administration was significantly lower than the AmpB AUC in rabbits administered FZ. However, AmpB systemic total body clearance was significantly greater in rabbits administered HFZ than in rabbits administered FZ without any differences in volume of distribution at steady state. Kidney tissue AmpB concentrations, although not significantly different, were greater in rabbits administered FZ than in rabbits administered HFZ. Likewise, lung and spleen AmpB concentrations, although not significantly different, were greater in rabbits administered FZ than in rabbits administered HFZ. However, liver AmpB concentrations were significantly lower in rabbits administered FZ than in rabbits administered HFZ. No significant differences in heart AmpB concentration between rabbits administered FZ and those given HFZ were found. These findings suggest that the pharmacokinetics, tissue distribution, and renal toxicity of AmpB are modified following administration of HFZ. HFZ could be an improved low-cost AmpB drug delivery system that has a potentially higher therapeutic index than FZ.

Heat-induced superaggregation of amphotericin B modifies its interaction with serum proteins and lipoproteins and stimulation of TNF-alpha.

The purpose of the present study was to examine the influence of heat-induced superaggregation of Amphotericin B (AmB) in the Fungizone (FZ) formulation on its interaction with human serum components and relate this to reduced toxicity. Whole serum distribution studies showed that a significantly lower percentage of AmB from HFZ was recovered in the high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglyceride-rich lipoprotein (TRL) fractions and a greater percentage recovered in the lipoprotein-deficient plasma (LPDP), though the majority of both preparations were recovered in LPDP. Circular dichroism (CD) and difference absorption spectroscopy were used to determine the stability of FZ and heat-treated FZ (HFZ) in the presence of HDL, LDL, serum, and albumin. The CD studies indicate that the "core" aggregate of HFZ is more stable in the presence of HDL and LDL, whereas the FZ is less stable and more dynamic with the core aggregate dissociating to a greater extent in the presence of either purified lipoprotein. Absorption studies with whole serum and purified albumin suggest that FZ aggregates are far less stable in the presence of albumin than HFZ and that interaction with serum albumin is a dominant feature for both drug preparations. HFZ also has a different effect on the cytokine response in vitro. Studies using THP-1 human monocytes show that HFZ provokes a smaller release of tumor necrosis factor (TNF)-alpha than FZ. This cytokine may be associated with the unpleasant side effects of AmB. These findings suggest that heat-induced superaggregation of AmB alters its interaction with HDL, LDL, serum proteins, and monocytes, and these findings may be important in explaining the reduced toxicity of the superaggregated form of AmB.

Activity and kinetics of dissociation and transfer of amphotericin B from a novel delivery form.

Recently it has been demonstrated that moderate heat treatment of Amphotericin B/deoxycholate solutions (HAmB-DOC ) leads to a therapeutically interesting supramolecular rearrangement that can be observed by significant changes in light scattering, CD, and absorbance. In this study, we continue the investigation of the physical properties of this new form by evaluating the activity and kinetics of dissociation and dispersion of HAmB-DOC and AmB-DOC in saline, serum, and in model mammalian or fungal lipid biomimetic membrane vesicles. Stopped-flow spectrophotometry combined with singular value decomposition (SVD) and global analysis were used to resolve the components of this process. The dissociation kinetics for both states are complex, requiring multi-exponential fits, yet in most cases SVD indicates only two significant changing species representing the monomer and the aggregate. The kinetic mechanism could involve dissociation of monomers from coexisting spectroscopically similar but structurally distinct aggregates or sequential rearrangements in supramolecular structure of aggregates. Rate constants and amplitudes of dissociation from aggregates to monomer in buffer, whole serum, 10% cholesterol, and ergosterol membrane vesicles are generally greater for AmB-DOC, demonstrating its greater kinetic instability. In addition, at comparable low concentrations, HAmB-DOC and AmB-DOC are nearly equally active at promoting cation selective permeability in ergosterol-containing membranes; however, HAmB-DOC is much less active against mammalian mimetic cholesterol-containing vesicles, despite a higher level of self-association, supporting previous observations that there exists a specific "toxic aggregate" structure.

Lipid and stress dependence of amphotericin B ion selective channels in sterol-free membranes.

The idea that amphotericin B (AmB) may not require sterols to form ion selective channels has recently been criticized on the grounds that egg phospholipids commonly used in experiments may contain small amounts of sterol which associate with AmB to form AmB/sterol pore channel structures. It was recently shown in this laboratory that modest osmotic stress can enhance the formation of AmB channels in sterol-free egg phosphatidylcholine (eggPC) membranes. We have tested AmB's ability to form ion channels/defects in synthetic palmitoyl oleoyl (POPC), dieicosenyl (DEPC) and natural eggPC osmotically stressed large unilamellar vesicles (LUV) using pyranine fluorescence detected ion/H+ exchange. These sterol-free POPC LUV exhibit greatly increased sensitivity to cation selective AmB channel formation when osmotically stressed; even more than eggPC. Under these stressed conditions, AmB activity was observed at [AmB]/POPC ratios as low as 3.5x10(-4), corresponding to about 34 AmB molecules/vesicle. DEPC vesicles were almost completely unresponsive, demonstrating a strong bilayer thickness dependence. These results prove conclusively that AmB can form sterol-free channels and do so within therapeutic concentration ranges (>0.5-10x10(-6) M) in a stress-dependent manner. This phenomenon may allow us to use osmotic stress changes in simple model systems to spectroscopically isolate and characterize the thus-far elusive AmB channel forming aggregate. In addition, this stress dependence may be responsible for the potentiation of renal toxicity of AmB in the ascending branch of the loop of Henle which is under greatest osmotic stress.

Osmotic stress sensitizes sterol-free phospholipid bilayers to the action of Amphotericin B.

We have tested the ability of Amphotericin B to form ion channels/defects in osmotically stressed large unilamellar vesicles (LUV) using pyranine fluorescence detected ion/H+ exchange. We found that sterol-free LUV exhibit greatly increased sensitivity to AmB channel formation in the soluble oligomer state (> 0.5 microM) under modestly hypoosmotic conditions (< 100 delta mosM). These vesicles are completely insensitive under isoosmotic conditions. The related antibiotics, Amphotericin B methyl ester and Nystatin showed almost no activity under hypoosmotic conditions in the absence of sterol. This difference may be attributable to differences in solution oligomeric states. Experiments with KCl and CaCl2 internal buffers demonstrate that these sterol-free AmB membrane disruptions are highly selective for monovalent cations (K+) over anions (Cl-), ruling out massive lysis or unselective membrane defects caused by osmotic pressure. Thus, AmB seems to be acting as a 'molecular harpoon', an expression coined to describe substances which can selectively target osmotically stressed, strained or highly curved membranes. These results may provide a rationale for AmB's reported anti-HIV activity and reported activity against sterol-free small unilamellar vesicles (highly curved membranes) as well as the reduced activity of liposomal drug delivery systems toward cholesterol-containing and sterol-free membranes (fewer soluble oligomers).

Na+, K+ and Cl- selectivity of the permeability pathways induced through sterol-containing membrane vesicles by amphotericin B and other polyene antibiotics.

Membrane diffusion potentials induced by amphotericin B (AmB), amphotericin B methyl ester (AmE), N-fructosyl AmB (N FruAmB) and vacidin, an aromatic polyene antibiotic, in ergosterol- or cholesterol-containing egg yolk phosphatidylcholine large unilamellar vesicles (LUV), were measured in various media, in order to determine the relative selectivity of Na+, K+, Cl- and other ions in these environments. Changes in the membrane potential were followed by fluorescence changes of 3,3'-dipropylthiadicarbocyanine (diS-C3-(5)). Subtle changes in intercationic selectivity were monitored by measuring biionic potentials, using the fluorescent pH sensitive probe pyranine. In all the cases studied, the intercationic selectivity of the permeability pathways induced by the four antibiotics was weak compared to that of specific biological channels, though distinct differences were noted. With AmB the selectivity appeared to be concentration dependent. Above 5 x 10(-7) M, the sequence determined for sterol-free small unilamellar vesicles (SUV) and cholesterol-containing SUV and LUV, Na+ > K+ > Rb+ > or = Cs+ > Li+ (sulfate salts), corresponded closely to Eisenman selectivity sequence number VII. At 5 x 10(-7) M and below the selectivity switched from Na+ > K+ to K+ > Na+. In contrast, Li+ was the most permeant ion for AmB channels in the presence of ergosterol. The selectivity between Na+ or K+ vs. Cl- varied with the antibiotic. It was very strong with vacidin at concentrations below 5 x 10(-7) M, smaller with AmB, nil with AmE and N FruAmB. The selectivities observed were antibiotic, concentration and time dependent, which confirms the existence of different types of channels.

Temperature effects on the aggregation state and activity of amphotericin B.

Amphotericin B (AmB) in aqueous solution becomes less aggregated as temperature is increased as measured by CD spectroscopy. Concomitantly, AmB's ability to induce K+ leakage from cholesterol-containing large unilamellar vesicles (LUV) decreases 4-fold per 10 degrees C increase. In contrast, ergosterol-containing vesicles show no clear trend in their temperature sensitivity to AmB action. These data strongly support a model of AmB action where drug aggregates and not monomers are the major channel-inducing species against cholesterol- but not ergosterol-containing membranes.

A kinetic method for measuring functional delivery of amphotericin B by drug delivery systems.

The human toxicity of amphotericin B can be considerably reduced by associating the drug with liposomes of varying lipid compositions. Some lipid compositions are much more effective than others. We show that a simple kinetic fluorescence assay using pyranine as an indirect probe of amphotericin-induced K+ currents may be used to study different liposomal drug delivery systems in vitro. We find that lipid mixtures composed of DMPC/DMPG/amphotericin at a 7:3:1 mole ratio show very slow functional delivery with a preference for ergosterol over cholesterol-containing membrane vesicles. On the other hand, amphotericin delivered from egg phosphatidylcholine liposomes lead to 100-fold increases in K+ leakage at one-fifth the amphotericin concentration of the 7:3:1 system. The egg phosphatidylcholine system as well as micellar amphotericin also show a slight selectivity towards cholesterol-containing vesicles over ergosterol. These results are consistent with previous clinical and in vitro cellular studies and this technique may prove valuable in screening of other delivery systems.

Potassium-selective amphotericin B channels are predominant in vesicles regardless of sidedness.

Amphotericin B (AmB) is a membrane-active antibiotic which has been shown to increase ion and small molecule permeability in a variety of model and biological membrane systems. A major mechanistic model, based on BLM systems, proposes that amphotericin forms barrellike pores with cholesterol which are cation selective when added to one side of the membrane and anion selective when added to both sides. We have tested this hypothesis on small and reverse-phase large unilamellar vesicles (SUV and REV) with and without cholesterol. The method used to measure K+, Cl-, and net ion currents is based on ion/H+ exchange detected by the entrapped pH probe pyranine. We find that AmB forms channels which have net selectivity for K+ over Cl- regardless of sidedness or sterol content in SUV. REV with 10% cholesterol also show net K+ selectivity with double-sided addition. Differences are noted between cholesterol- and non-sterol-containing vesicles consistent with at least two separate modes of action: (1) cholesterol-containing SUV form some larger diameter pores which allow the passage of larger ions especially when added to both sides; (2) SUV without sterol form pores which are still K+ over Cl- selective, but larger ions do not pass. The latter mode of action precludes a sterol/pore type of model but not necessarily a barrellike model consisting only of amphotericin molecules.(ABSTRACT TRUNCATED AT 250 WORDS)

Amphotericin B and Nystatin show different activities on sterol-free vesicles.

It has generally been assumed that the polyene antibiotics Nystatin and Amphotericin B cause membrane damage by the same mechanism. However, using kinetic fluorescence methods we have found that AmB and Nystatin have very different activities on sterol-free dioleoyl phosphatidylcholine and egg phosphatidylcholine small unilamellar vesicles. At very low AmB concentrations (less than 1/1000 lipids in egg phosphatidylcholine) significant K+ permeability enhancement is observed. However, even at very high Nystatin to lipid ratios (1/100) very little K+ current is induced, particularly in dioleoyl phosphatidylcholine vesicles. The novel technique described here uses a K+/H+ exchange mechanism to detect minute transmembrane K+ currents by monitoring internal membrane vesicle pH changes with pyranine.

Spectral evidence for cytochrome o in the brown membrane of Halobacterium halobium.

The brown membrane domain of Halobacterium halobium is a developmental precursor of the purple membrane and it contains a b-type cytochrome in addition to bacterio-opsin. In this report we provide spectroscopic evidence that the majority of the cytochrome content is a halobacterial cytochrome o. This cytochrome has absorption spectral properties in the oxidized, reduced, and CO liganded states which are characteristic of cytochrome o. The CD spectra show a complex bilobed pattern in the Soret spectral region which reflects the similarity of the heme environment to those of other b-type cytochromes involved in electron transport. We have also demonstrated a positive cooperativity of CO binding which, combined with CD spectral results, suggests two interacting heme moieties per cytochrome o. Size exclusion HPLC of solubilized brown membrane preparations shows a heme b containing protein with an Mr 42,000-46,000. The cytochrome may be easily separated from bacterio-opsin using a hydroxyapatite elution method on Triton X-100 solubilized preparations. The relative ease with which brown membrane disks may be oriented in high optical quality films should make the brown membrane a valuable model system for future spectroscopic investigations of cytochrome o.

A selective cholesterol-dependent induction of H+/OH- currents in phospholipid vesicles by amphotericin B.

The effect of amphotericin B on the proton/hydroxide permeability of small unilamellar vesicles has been investigated by using potential-dependent paramagnetic probes. Amphotericin B at 1-10 molecules/vesicle causes a modest 4-8-fold increase in the background H+/OH- permeability of egg phosphatidylcholine (egg PC) vesicles. However, in the presence of cholesterol, amphotericin B promotes a dramatic increase in the H+/OH- permeability of more than 2 orders of magnitude. Surprisingly, this is not observed in vesicle membranes containing ergosterol. In membranes composed of 5-15 mol% ergosterol, amphotericin B is even less effective at promoting H+/OH- currents than in pure egg PC vesicles. The K+ current promoted by amphotericin B in vesicles formed from egg PC and from egg PC plus cholesterol or ergosterol was measured. No significant sterol dependence was found for the K+ current. These results strongly suggest that different mechanisms, or amphotericin B/sterol complexes, are responsible for the induction of H+/OH- and K+ currents. These results have important implications for understanding the therapeutic and toxic effects of amphotericin B.

Time-dependent binding of paramagnetic and fluorescent hydrophobic ions to the acetylcholine receptor from Torpedo.

In receptor-rich vesicles isolated from Torpedo, paramagnetic or fluorescent phosphonium ions bind to both the acetylcholine receptor (AcChR) and the receptor membrane. When added to receptor vesicles, two to three phosphoniums undergo a slow time-dependent binding to the AcChR. The presence of agonist increases the rate but not the extent of binding of the alkylphosphonium nitroxides. Approximately one phosphonium per receptor can be displaced by the addition of saturating concentrations of the high-affinity histrionicotoxin derivative isodihydrohistrionicotoxin or by the addition of phencyclidine or quinacrine mustard. In addition, preincubation of the receptor with these channel blockers prevents approximately one phosphonium from binding to the receptor. When a series of alkyltriphenylphosphonium ions was studied, it was found that the rate of phosphonium binding to the receptor decreased with increasing probe hydrophobicity. This appears to be a function of the partitioning of the probe between membrane and aqueous phases. The phosphonium ions used here promote desensitization of the receptor, as judged by the binding rate of the fluorescent agonist NBDA-C5-acylcholine or alpha-bungarotoxin. Preincubation of the receptor with isodihydrohistrionicotoxin virtually eliminates the phosphonium-mediated desensitization. The rates of the phosphonium-mediated desensitization also appear to be dependent upon the phase partitioning of the probe. These results strongly suggest that the binding sites for the phosphonium ion (and the high-affinity histrionicotoxin blocking site) are accessible only through the aqueous phase. The phosphonium binding and agonist-induced transitions observed here are not observed with a negative hydrophobic ion probe, or a negative surface amphiphile, indicating that modifications in membrane electrostatics do not contribute to the observed changes.(ABSTRACT TRUNCATED AT 250 WORDS)

A test of discreteness-of-charge effects in phospholipid vesicles: measurements using paramagnetic amphiphiles.

A new series of negatively charged, paramagnetic alkylsulfonate probes was synthesized and can be used to measure both the internal and the external surface potentials of model membrane systems. We tested for discreteness-of-charge effects in lipid membranes by comparing the surface potentials, estimated by use of these negatively charged amphiphiles, with that of a series of positively charged alkylammonium nitroxides in charged membranes. From the partitioning of these probes, the membrane surface potential was estimated in phosphatidylcholine membranes containing either phosphatidylserine or didodecyldimethylammonium bromide. The surface potentials, estimated with either positive or negative probes, were identical, within experimental error, in either positive or negative membranes, and they were well accounted for by a simple Gouy-Chapman-Stern theory. This symmetry, with respect to the sign of the charge, indicates that discreteness-of-charge effects are not significant in determining the potential-sensitive phase partitioning of these probes in model membranes. Thus, despite the fact that charge on membranes is discrete, models that assume a uniform density of charge in the plane of the membrane adequately account for the potentials measured by these amphiphilic probes.

Biotransformation of Cannabidiol to Cannabielsoin by Suspension Cultures of Cannabis sativa and Saccharum officinarum.

Suspension cultures of CANNABIS SATIVA L. and SACCHARUM OFFICINARUM L. have been shown to biotransform cannabidiol to Cannabielsoin under normal growth conditions. Both Cannabielsoin C-1 stereoisomers were produced. Mass spectral and chromatographic evidence were used in this identification. This is the first reported biologically catalyzed synthesis of Cannabielsoin as well as the first biotransformation of any cannabinoid by plant tissue culture systems. Possible mechanisms and significance of this conversion are discussed.