ABSTRACT
An experimental micellar formulation of 1:1.5 amphotericin B-sodium deoxycholate (AMB:DCH 1:1.5) was obtained and characterized to determine its aggregation state and particle size. The biodistribution, nephrotoxicity, and efficacy against pulmonary aspergillosis in a murine model were studied and compared to the liposomal commercial formulation of amphotericin B after intravenous administration. The administration of 5 mg/kg AMB:DCH 1:1.5 presented 2.8-fold-higher lung concentrations (18.125 ± 3.985 μg/g after 6 daily doses) and lower kidney exposure (0.391 ± 0.167 μg/g) than liposomal commercial amphotericin B (6.567 ± 1.536 and 5.374 ± 1.157 μg/g in lungs and kidneys, respectively). The different biodistribution of AMB:DCH micelle systems compared to liposomal commercial amphotericin B was attributed to their different morphologies and particle sizes. The efficacy study has shown that both drugs administered at 5 mg/kg produced similar survival percentages and reductions of fungal burden. A slightly lower nephrotoxicity, associated with amphotericin B, was observed with AMB:DCH 1:1.5 than the one induced by the liposomal commercial formulation. However, AMB:DCH 1:1.5 reached higher AMB concentrations in lungs, which could represent a therapeutic advantage over liposomal commercial amphotericin B-based treatment of pulmonary aspergillosis. These results are encouraging to explore the usefulness of AMB:DCH 1:1.5 against this disease.
KEYWORDS: aspergillosis, amphotericin B, deoxycholate, pulmonary concentrations, efficacy, nephrotoxicity
INTRODUCTION
The increase in immunosuppressive agents and the extensive usage of corticosteroids in chronic obstructive pulmonary disease have led to a rising prevalence of invasive mycoses such as aspergillosis in recent years (1, 2). The initial site of most fungal infections such as aspergillosis in humans is primarily the lungs, with hematogenous dissemination to the spleen, kidneys, liver, and brain as the disease progresses. Therefore, high initial pulmonary concentrations of antifungal drugs are needed to avoid the dissemination of the pathogens that make the infection even more difficult to treat. Liposomal amphotericin B (LAMB) is the “gold standard” in aspergillosis therapy (1, 3). Several studies have demonstrated that the large particle sizes of different AMB formulations are related to high nephrotoxicity (4–7). Commercial dimeric deoxycholate (DCH)-amphotericin B (D-AMB) shows a high percentage of small particles (56.2 ± 4.3 nm) and a small percentage of large particles (around 4.0 μm) (6), the latter related to nephrotoxicity (4–7). LAMB, the reference commercial formulation, with a polyaggregated AMB form, shows a small particle size (around 100 nm) that enhances antifungal efficacy and diminishes drug toxicity (8), while other lipid complexes marketed as Abelcet (ABLC) present polyaggregated AMB with a particle size of 1.6 to 11 μm (7). This larger particle size of the ABLC formulations favors a greater pulmonary distribution but also increases their nephrotoxicity compared to LAMB (4).
However, the high costs of commercial lipid formulations limit their use. On the other hand, the less expensive commercial dimeric deoxycholate-amphotericin B (D-AMB) has lower antifungal efficacy and causes serious side effects (9). Therefore, various new amphotericin B formulations such as emulsions, liposomes, and microspheres have been developed to increase efficacy and decrease side effects (10). In previous studies, it has been observed that polyaggregated systems of AMB with sodium deoxycholate are a safer and less toxic form than D-AMB, with the same proportion of DCH, when administered intravenously (i.v.) (6, 11). The AMB distribution to organs with high presence of macrophages and organs of the reticuloendothelial system like liver and spleen is related to formulation parameters such as the aggregation state and particle size (11, 12). After the parenteral administration of LAMB, the highest concentrations appeared in organs of the reticuloendothelial system, while lower drug concentrations appeared in other organs such as lungs (13). Thus, a lung concentration of 1.44 μg/g was obtained 24 h after the administration of a 5-mg/kg of body weight dose of LAMB (14). These probably insufficient pulmonary concentrations make it necessary to search for new AMB formulations in order to guarantee higher efficacy against dangerous pulmonary pathogens like Aspergillus spp.
The aim of this project is to develop a lung-specific delivery system of AMB with a high pulmonary distribution and a low nephrotoxicity. A low-renal-dissemination pulmonary aspergillosis model was selected in order to correlate nephrotoxicity results with the new AMB formulation.
RESULTS
AMB formulation characteristics.
The aggregation state of amphotericin B in the experimental formulations was evaluated by measuring UV-visible absorbance. A standard AMB formulation using methanol as a solvent (M-AMB) and a formulation with no surfactant in water for injection (AMB:DCH 1:0) were used as reference formulations and compared with the deoxycholate-containing formulation (AMB:DCH 1:1.5). The absorption spectrum of M-AMB showed four high pronounced peaks at 353, 372, 390, and 414 nm. The absorption spectrum of AMB:DCH 1:0 showed two faint peaks at 386 and 403 nm.
The absorption spectrum of the AMB:DCH 1:1.5 formulation in water showed a shoulder around 330 nm and some faint peaks at higher wavelengths (393, 407, and 424 nm). The absorbance values at different wavelengths of this deoxycholate-containing formulation were clearly different from the previously observed ones in the spectrum of M-AMB. Furthermore, AMB:DCH 1:1.5 showed 3-fold-higher absorbance values than the formulation with no DCH (AMB:DCH 1:0).
The particle sizes of the experimental (AMB:DCH 1:1.5) and reference (AMB:DCH 1:0) formulations were determined and expressed as mean particle size (in nanometers) ± standard deviation (SD). Both formulations in water presented polydispersity indexes of less than 0.6.
The presence of sodium deoxycholate in the AMB:DCH 1:1.5 formulation significantly (P < 0.001) decreased the particle size (404.9 ± 1.7 nm) versus the formulation without surfactant, AMB:DCH 1:0 (514.8 ± 34.2 nm).
AMB biodistribution to kidneys and lungs.
The study of AMB biodistribution consisted of administration of a single and multiple doses of AMB in uninfected and immunosuppressed mice in order to evaluate the AMB concentrations reached in lungs and kidneys.
AMB concentrations in renal and lung tissues 24 h after a single dose of AMB:DCH 1:1.5 or LAMB formulations administered at 5 mg/kg are shown in Fig. 1. The LAMB formulation showed AMB concentrations in kidneys 15 times greater than those of AMB:DCH 1:1.5 (P < 0.01). However, AMB concentrations reached in lung tissues with the AMB:DCH 1:1.5 formulation at 24 h were significantly higher (P < 0.01) than the ones obtained with LAMB treatment (9.173 ± 0.498 versus 2.527 ± 0.386 μg/g, respectively). After 6 days of treatment, renal concentrations of AMB showed an important cumulative effect (Fig. 1A). However, low kidney levels of AMB were reached with the AMB:DCH 1:1.5 formulation after 6 days of treatment (0.391 ± 0.167 μg/g). Thus, AMB renal concentrations for LAMB were 15-fold higher than those of the AMB:DCH 1:1.5 formulation (P < 0.01). Also, lung concentrations of AMB showed a cumulative effect with both AMB:DCH 1:1.5 and LAMB formulations (Fig. 1B). A higher concentration in lung tissue was observed with the AMB:DCH 1:1.5 formulation at a dose of 5 mg/kg (18.125 ± 3.985 μg/g) compared with a LAMB formulation dose of 5 mg/kg/day (6.567 ± 1.536 μg/g).
FIG 1.
Mean and standard deviation of AMB concentrations (micrograms per gram) in (A) kidneys and (B) lungs of immunosuppressed mice (n = 6 mice/group) intravenously treated with 1 or 6 daily doses of 5 mg/kg of AMB:DCH 1:1.5 or LAMB. LAMB treatments with 1 or 6 daily doses were significantly different from AMB:DCH 1:1.5 (P < 0.01).
Drug nephrotoxicity.
One of the most important problems with amphotericin B is its nephrotoxicity. Therefore, the renal toxicities of all AMB new experimental formulations should be tested and compared with those of commercial formulations. In this first preclinical study of toxicity, immunocompetent mice were used.
Serum nephrotoxicity biochemical parameters such as creatinine and blood urea nitrogen (BUN) were tested after six daily doses of each treatment at therapeutic or elevated AMB doses (5 and 20 mg/kg/day). Table 1 shows the serum creatinine and BUN levels. All creatinine concentrations obtained with experimental and commercial formulations were slightly higher than the one obtained with the control group. These creatinine values were similar after the administration of therapeutic or high doses of AMB. The commercial LAMB formulation presented the highest creatinine concentrations.
TABLE 1.
Creatinine and BUN concentrations in uninfected mice 24 h after 6 daily doses of the indicated amphotericin B formulation group
| Formulation | Dose (mg/kg) | Mean ± SD concn (mg/dl) ofa: |
|
|---|---|---|---|
| Creatinine | BUN | ||
| Control | 0.18 ± 0.008 | 18.29 ± 0.609 | |
| LAMB | 20 | 0.23 ± 0.034 | 18.160 ± 0.963 |
| 5 | 0.24 ± 0.018 | 18.29 ± 0.609 | |
| AMB:DCH 1:0 | 20 | 0.22 ± 0.028 | 17.27 ± 0.875 |
| 5 | 0.23 ± 0.005 | 17.73 ± 1.276 | |
| AMB:DCH 1:1.5 | 20 | 0.20 ± 0.011 | 18.20 ± 0.748 |
| 5 | 0.21 ± 0.005 | 19.88 ± 0.762 | |
Creatinine values after LAMB treatments at 20- and 5-mg/kg doses were significantly different from control group (P < 0.01 and P < 0.001, respectively). Creatinine values after AMB-DCH 1:0 treatments at 20- and 5-mg/kg doses were significantly different from control group (P = 0.035 and P < 0.01, respectively). Creatinine values after AMB-DCH 1:1.5 treatments at 20- and 5-mg/kg doses were not significantly different from control group (P = 0.055 and P = 0.057, respectively). Creatinine values after LAMB treatments at 20- and 5-mg/kg doses were significantly different from AMB-DCH 1:1.5 (P = 0.005 and P = 0.007, respectively). BUN values after all treatments were not significantly different from control group (P > 0.05).
Histopathological studies of renal tissue of all mice treated with 6 daily doses of AMB at 5 mg/kg/day showed tubular and glomerular morphological abnormalities compared to the control group, which maintained the normal morphology (Fig. 2). These changes in renal tissue of mice treated with experimental and commercial AMB formulations include extensive areas of inflammatory cells (surrounding tubules and glomeruli), extracapillary proliferative glomerulonephritis, and acute tubular epithelial damage with the presence of tubular necrosis, cell debris detachment, swelling, and tubular dilation. The renal sample of the LAMB formulation also showed protein casts in renal tubules. The photomicrographs of mouse kidney sections regarding higher AMB doses (20 mg/kg/day) showed a small increase in renal tissue damage (data not shown) when this dose was administered compared with the same formulations given at 5 mg/kg/day.
FIG 2.
Histopathology of kidneys from nonimmunosuppressed, uninfected mice following intravenous treatment with AMB formulations at 6 daily therapeutic doses of 5 mg/kg/day. (A) Appearance of normal renal tissue in a kidney from the control group. (B) Tissue from mice treated with AMB:DCH 1:1.5. (C) Tissue from mice treated with AMB:DCH 1:0. (D) Tissue from mice treated with commercial formulation (LAMB). Arrows indicate inflammatory cells, arrowheads show tubular epithelial damage, and an asterisk indicates intraluminal protein casts. Size bars, 100 μm.
Drug efficacy.
This study was performed in immunosuppressed and infected animals. Figure 3A shows survival rates of untreated animals and animals treated with 5 mg/kg AMB:DCH 1:1.5 or LAMB. All animals from the control group succumbed within 8 days postchallenge. Both treatments AMB:DCH 1:1.5 and LAMB significantly increased the survival of the animals, with survival rates of 42.8% and 50.0% in comparison to the control group (P = 0.077). However, no significant differences were found between treated groups (P > 0.05).
FIG 3.
(A) Survival of mice (n = 8 mice/group) challenged intranasally with 20 μl of A. fumigatus (5 × 104 conidia/ml) and then intravenously treated for 6 days with 5 mg/kg placebo (5% dextrose), 5 mg/kg LAMB, and 5 mg/kg AMB:DCH 1:1.5. (B) Mean percentage of weight change in uninfected immunosuppressed mice (n = 8/group) treated with 5 mg/kg LAMB and 5 mg/kg AMB:DCH 1:1.5 and in infected mice (n = 8/group) treated with 5 mg/kg LAMB and 5 mg/kg AMB:DCH 1:1.5. P < 0.05 in comparison to the control group.
Body weights showed a drop in all animals (Fig. 3B).This pattern was followed by a continuous drop in weight in control animals, in contrast to a progressive rise in weight of the treated groups. Infected mice treated with the AMB:DCH 1:1.5 and LAMB formulations at doses of 5 mg/kg showed similar percentages of weight loss, as well as weight recovery starting on days 8 and 6 postinfection, respectively.
Infection showed viable fungal elements in most of the lungs from control and treated animals, with no significant differences between groups (P > 0.05). Fungal burden in the group treated with the experimental AMB:DCH 1:1.5 formulation showed the lowest values (log10 2.58 ± 2.130 CFU/g lung tissue), with 40% of mice displaying nondetectable CFU in this tissue. However, contrary to our expectations, all mice in the group receiving LAMB had high CFU per gram values (log10 4.45 ± 0.40 CFU/g lung tissue). Infection showed low spread to kidneys in control animals, but high clearance was observed in treated animals, with undetectable CFU in 80% of animals receiving AMB:DHC 1:1.5 and 100% of those receiving LAMB.
No CFU were recovered from kidneys at the end of the experimental time (i.e., 12 days after infection in treated animals), in contrast to the results in lungs, where a few animals treated with AMB:DHC 1:1.5 and LAMB showed low fungal burdens with no significant differences between them (P = 0.371) (data not shown). Important changes in eradication percentages for each treatment group between days +8 and +12 were observed. Thus, the group that received 5 mg/kg of the AMB:DCH 1:1.5 formulation presented similar eradication results in lung tissue 12 days after infection to those obtained on day 8. The group that received 5 mg/kg/day of the LAMB formulation showed an important increase in eradication percentages between days +8 and +12 (3 log10).
DISCUSSION
The aggregation state study of the reference M-AMB formulation showed peaks corresponding to the monomeric form of AMB (15). However, the absorption spectrum of the AMB:DCH 1:1.5 formulation was clearly different from the previously observed spectrum of the monomeric form and did not show the characteristic high peak of the dimeric form (15). Therefore, this spectrum of experimental AMB:DCH formulation was indicative of the presence of the amphotericin B polyaggregated form, which usually presents a lower nephrotoxicity than D-AMB (6, 9). AMB:DCH 1:1.5 showed higher absorbance values than the reference formulation with no deoxycholate (AMB:DCH 1:0). These higher absorbance values of the AMB:DCH 1:1.5 formulation may be related to the polyaggregated AMB form in a small-particle-size micellar system induced by the surfactant action of the sodium deoxycholate (16, 17).
Particle size measurements were performed in triplicate, and the results are represented as the mean ± standard deviation. Polydispersity indexes were similar to the ones presented by other authors in micellar formulations (6). In previous studies, the polyaggregated formulation AMB:DCH 1:0 showed a larger particle size (1,280 ± 216.0 nm) when using a shorter agitation time (2 min) (6). Therefore, stirring conditions have been proven to be a key issue in order to obtain small polyaggregated AMB systems, as other authors have also reported (17).
The AMB:DCH 1:1.5 polyaggregated formulation presented a small standard deviation when its particle size was studied, and this low variability should be attributed to the stabilizing effect of DCH by means of ionic interactions with AMB (5). Thus, in this AMB formulation, sodium deoxycholate acts as a stabilizing agent inhibiting particle growth, so the surfactant effect is critical in AMB nanoparticle systems (16, 17). The use of surfactants as stabilizing agents has been reported previously by several authors (6, 18).
Finally, it is noteworthy that the particle size of AMB in this polyaggregated preparation was very different from the one observed with the commercial AMB formulation (LAMB), which has been reported as about 100 nm by other authors and therefore may be related to a different organ distribution (19).
The AMB concentrations obtained in this work with the studied formulations showed different AMB distributions to kidneys and lungs, as other authors have previously observed (13). Our results suggest that particle size and surface morphology of colloidal and liposomal systems may influence the distribution of AMB to different organs. Formulations with small particle size (LAMB) presented high concentrations in organs like the kidney, while the new polyaggregated AMB:DCH formulation with a larger particle size led to a greater lung distribution.
At the renal level, the new polyaggregated formulation developed in this work showed smaller AMB concentrations in kidneys than the ones observed with LAMB (20, 21) after either 1 or 6 days of treatment (5 mg/kg/day). The increase in AMB renal concentrations observed with the DCH formulation may be related to the influence of the hydrophilic sodium deoxycholate surfactant on the distribution of micelles. However, due to the important renal effects of AMB formulations (11), a toxicity study was performed with the experimental DCH formulation, employing LAMB and AMB:DCH 1:0 formulations as a reference, to observe the possible influence of AMB and DCH on nephrotoxicity.
The serum nephrotoxicity biochemical parameter study showed that BUN values for all the formulations were similar to those of the control group at both doses studied (5 and 20 mg/kg) and similar to those obtained in different treatments with LAMB (22). Therefore, the high AMB renal concentrations after LAMB treatment did not translate into pathological BUN values.
All AMB formulations induced higher creatinine values than the control group. Similar increases in creatinine levels were observed with other polyaggregated AMB formulations at elevated doses (15 mg/kg) (23), but an increase in creatinine values was observed in the LAMB group compared to AMB:DCH 1:1.5 at both doses studied (5 and 20 mg/kg). These high creatinine values were possibly related to the high AMB redistribution into the kidneys observed with LAMB (22, 24). It is particularly noteworthy that the experimental AMB:DCH 1:1.5 formulation did not show pathological creatinine values, and no significant difference (P > 0.05) from the control group was observed. However there were significant differences between creatinine values after administration of the LAMB and AMB:DCH 1:1.5 formulations at therapeutic and elevated AMB doses (5 and 20 mg/kg/day) (P < 0.001 and P < 0.01, respectively).
Nevertheless, all renal tissue samples of mice treated with all the formulations after 6 daily doses of 5 or 20 mg/kg/day presented inflammatory cells and tubular epithelial damage, while protein casts were only observed in renal tissue of mice treated with LAMB. Thus, all preclinical nephrotoxic studies with experimental AMB formulations should include histopathological studies, due to the fact that serum nephrotoxicity biochemical parameters (creatinine and BUN values) would be insufficient to confirm prompt renal damage.
Furthermore, the renal histopathological study has shown the existence of renal damage (Fig. 2) when AMB renal concentrations are ≥0.372 μg/g, since this was the AMB renal concentration of mice treated intravenously (i.v.) with 5 mg/kg/day of AMB:DCH 1:1.5 for 6 days (Fig. 1). On the other hand, when renal damage was compared between DCH formulations (AMB:DCH 1:0 and AMB:DCH 1:1.5), it was observed that the presence of DCH in AMB formulations did not increase the nephrotoxicity after 6 daily doses of 5 and 20 mg/kg/day (Table 1 and Fig. 2).
The nephrotoxicity values described for the AMB:DCH 1:1.5 formulation for 6 days at high doses (20 mg/kg) were related more to the presence of large particles at this dose than to its high renal AMB concentration. These results are consistent with the nephrotoxicity values after prolonged treatments observed in AMB formulations with large particle size (25). However, nephrotoxicity values described with LAMB formulation are related more to a high AMB renal concentration. However, it must be emphasized that while the LAMB formulation presented 15-fold-higher AMB renal concentrations than the DCH formulations, a 15-fold greater nephrotoxicity was not observed.
Lung AMB concentrations reached with the experimental AMB:DCH 1:1.5 formulation at a dose of 5 mg/kg were higher than the ones obtained with LAMB. Differences in the particle sizes between the AMB:DCH 1:1.5 and LAMB formulations may explain the differences observed in the distribution of AMB to the lungs because a larger particle size will allow a faster opsonization by macrophages (11). This may also explain the high AMB concentrations reached with both experimental AMB formulations in organs rich in macrophages like the lungs (14). These lung concentrations would be adequate to treat aspergillosis (MIC, 0.5 to 8 μg/ml) (26). Furthermore, when the experimental AMB:DCH 1:1.5 formulation was administered, the threshold pulmonary concentration (>4 μg/g) required for therapeutic efficacy of LAMB treatment in a murine model of pulmonary aspergillosis (27) was exceeded. The pulmonary AMB concentrations obtained with LAMB after multiple doses were similar to the ones obtained with different AMB formulations by other authors (27, 28). The AMB:DCH 1:1.5 formulation had the greatest cumulative effect on lung tissue compared to the LAMB formulation (25).
Treatments consisting of AMB:DCH 1:1.5 and the gold standard formulation (LAMB) have shown equivalent efficacies in terms of survival and reduction of tissue burden, despite the AMB levels being higher in lungs of animals receiving AMB:DCH than in those treated with LAMB. This may be due to the immunomodulatory effect of the drug carrier since it has been demonstrated that empty liposomes show efficacy against invasive pulmonary aspergillosis (IPA) modeled in corticosteroid-immunosuppressed mice. The role of empty liposomes, as well as DCH, in the progression and outcome of the infection deserves to be further explored (29). In addition, comparable decreases in the percentages of weight loss in the different groups were observed. A similar relationship between survival times and weight loss results has been previously observed with different AMB aspergillosis treatments (7, 30).
Our pulmonary model has demonstrated high pulmonary affectation with viable fungal elements at day 8 postinfection, with low renal affectation suggesting that the A1160 strain has poorly disseminated to other organs at 8 days postchallenge (31). The absence of nephrotoxicity after treatment allows us to consider that nephrotoxicity results in uninfected mice could be adequate to compare AMB:DCH 1:1.5 and LAMB at doses of 5 mg/kg. Previous studies showed similar BUN values between infected and uninfected mice given LAMB and ABLC treatments (22).
The high mortality displayed by the infected control group was due to the fungal burden (32, 33). In addition, the tissue burden study performed 8 days postinfection showed great fungal clearance in lungs of those animals receiving 5 mg/kg of AMB:DCH 1:1.5, while all mice receiving 5 mg/kg of LAMB showed slightly higher levels of CFU. These results were consistent with the lung concentration data obtained in uninfected mice. Probably, the largest particle size of the AMB:DCH 1:1.5 formulation produced a faster biodistribution to lung tissue (9.173 ± 0.498 μg/g at 24 h) and a greater fungal burden clearance (34). The fast drug clearance of lung tissue for the AMB:DCH 1:1.5 formulation might explain the decrease in survival rate (day +8) (Fig. 3). Nevertheless, LAMB at doses of 5 mg/kg had low lung concentration values after the initial dose in uninfected mice (2.527 ± 0.386 μg/g at 24 h). This low initial biodistribution could be related to an increase in fungal burden during the first stage of the infection in this group (27). Doses of the LAMB formulation of 10 mg/kg would be required in order to obtain a significant reduction in A. fumigatus growth during the first stage of the infection (27, 35).
The fungal burden results obtained on day +12 confirm that this A. fumigatus infection model by pulmonary inhalation poorly disseminated to kidneys (33). Differences in lung tissue CFU values for AMB:DCH 1:1.5 and LAMB treatment may be related to the AMB concentrations due to the different particle sizes between the two formulations (11, 15). A decrease in CFU for LAMB formulation at the end of the study (+12 day) was related to a smaller particle size and morphological characteristic of the liposome form of AMB (30, 36). Thus, small spheres of commercial AMB formulation (LAMB) remain in the lungs 12 to 48 h after administration (7, 28). These results from different formulations suggest that their morphology and particle size could affect the pharmacokinetic/pharmacodynamic characteristics of the different AMB formulations.
Nevertheless, a limitation of this study, performing the pharmacokinetic comparisons between uninfected mice and infected mice is the fact that the AMB concentrations that would be reached in infected mice would be higher than the ones obtained in this study in uninfected mice (22, 27). Thus, the similar efficacies observed with AMB:DCH 1:1.5 and LAMB would be related to concentrations higher than 4 μg/g in infected animals (the AMB concentration associated with a significant reduction of lung fungal burden) with both formulations and not only the concentrations observed with uninfected animals (Fig. 1) in the first 24 h (27).
In conclusion, the micellar polyaggregated AMB:DCH 1:1.5 formulation showed a 2.8-fold-higher AMB concentration in lungs than the commercial LAMB formulation after 6 doses of 5 mg/kg and a slightly lower nephrotoxicity due to the formulation characteristics. The improved pulmonary efficacy/toxicity ratio allows the micellar AMB:DCH 1:1.5 formulation to be considered a good alternative to LAMB liposomes for the treatment of pulmonary aspergillosis. Future studies with infected mice will be performed to evaluate the pharmacokinetic/pharmacodynamic efficacy of treatments with the experimental AMB:DCH 1:1.5 and commercial LAMB formulations on their own compared to treatments combining different doses of these formulations.
MATERIALS AND METHODS
Drugs.
Amphotericin B (AMB) (PubChem CID 5280965) was supplied by Bristol Myers Squibb (Barcelona, Spain), and LAMB (Ambisome) was supplied by UCB-Pharma (Brussels, Belgium).
Preparation of formulations.
The experimental micellar formulation was prepared with AMB and DCH at a ratio of 1:1.5 (wt/wt) drug to carrier. The preparation method was as follows. DCH (PubChem CID: 23668196) purchased from Fluka-Biochemika (Bucks, Switzerland) was dissolved at 18.75 mg in 50 ml of water for injection, and then 12.5 mg of AMB was added, and the mixture was vortexed for 4 min at 2,400 rpm. These suspensions were diluted in 5% dextrose solution, as necessary, vortexed for 4 min at 2,400 rpm, and stored in darkness at 25°C. For the characterization of the aggregation state of the formulations, an AMB standard solution (M-AMB) was required. This formulation was prepared as follows. AMB (10.0 mg) was dissolved in 500 ml of methanol and vortexed for 4 min at 2,400 rpm. The M-AMB formulation was diluted in 5% dextrose solution, as necessary.
Characterization of AMB formulations.
Different AMB:DCH formulations in water (AMB:DCH 1:0 and AMB:DCH 1:1.5) and the M-AMB formulation in methanol were spectrophotometrically analyzed by UV absorption to determine the aggregation state of AMB. A scanning spectrum of each formulation was recorded by a Shimadzu UV-1700 spectrophotometer (Kyoto, Japan) between 300 and 450 nm. Each sample was analyzed in triplicate.
The particle size of AMB:DCH formulations in water was analyzed by Microtrac S-3500, (Microtrac, Inc., Montgomeryville, PA). Mean size (in nanometers) was determined based on size distribution in number. Each sample was analyzed in triplicate.
Animals.
Four-week-old OF-1 male mice (Criffa S.A., Barcelona, Spain) weighing between 27 and 32 g were used in all experiments. All animals were housed under standard conditions with water and food ad libitum. All animal care procedures were supervised and approved by the Complutense University Animal Welfare and Ethics Committee as well as the Universitat Rovira i Virgili Animal Welfare and Ethics Committee.
AMB biodistribution to kidneys and lungs.
This study was performed in immunosuppressed uninfected mice in order to evaluate the AMB concentrations in kidneys and lungs after a single dose or 6 daily doses of AMB (5 mg/kg/day).
Four groups of 6 animals each were immunosuppressed by subcutaneous injection of cortisone acetate (125 mg/kg) every 3 days (37). Five days after immunosuppression began, animals received a single dose of AMB:DCH 1:1.5 or LAMB both administered intravenously (i.v.) at doses of 5 mg/kg or in multiple doses administered once daily (QD) at doses of 5 mg/kg for 6 days.
To determine the concentrations of AMB in kidney and lung tissues, mice were euthanized 24 h after the last administration. Organs were aseptically removed and frozen until used. Then organs were homogenized in 0.5 ml of water for injection, and the total AMB concentration in tissue was determined by high-performance liquid chromatography (HPLC) using a modification of previously published assays (38). Briefly, all HPLC assays were performed using the modular system Jasco (Tokyo, Japan), which consisted of an automatic sample injection (A5-2050), a high-pressure pump (PU-1580), and a UV detector (1575), operating at 406 nm. Analyses were performed on a Hypersil BDS C18 column (5 μm, 250 mm by 4.6 mm). The column was isocratically eluted at a flow rate of 1 ml/min with 40:4.3:55.7 (vol/vol/vol) acetonitrile-acetic acid-water mobile phase. The injection volume was 100 μl. Each sample was analyzed in triplicate.
Homogenized tissue samples were spiked with meloxicam (10 μl at 200 μg/ml). Two extractions of the homogenate aqueous tissue were carried out with methanol (400 μl × 2). After every extraction, the mixture was vortexed (2,500 rpm, 2 min) and then centrifuged (4,000 rpm, 4 min). The supernatant was filtered using a hydrophilic polyvinylidene fluoride (PVDF) filter with a pore size of 0.45 μm. Previous to the homogenization and analysis by HPLC, each organ was weighed, and the results were expressed as micrograms per gram of each organ. Under these conditions, the relative retention times of AMB and the internal standard were 8.5 and 6.1 min, respectively. Tissue AMB concentrations were calculated from linear regression calibration curves of the AMB/internal standard peak height ratio. The linear range in plasma was 0.01 to 10 μg/ml (y = 0.2191x − 0.0026; R2 = 0.9986).
Drug nephrotoxicity.
The drug nephrotoxicity study was performed in immunocompetent uninfected mice. Thirty-five mice were divided into seven groups (n = 5) and treated i.v. with 5 or 20 mg/kg/day of LAMB, AMB:DCH 1:0, AMB:DCH 1:1.5, or placebo (5% dextrose) for 6 days. Blood samples were collected by cardiac puncture 24 h after the sixth treatment. Creatinine and blood urea nitrogen (BUN) were analyzed in serum using a modular AutoAnalyzer Cobas 711 (Roche, Basel, Switzerland).
Histopathological evaluations were also performed on the renal tissues in these seven groups of mice. Kidneys were collected 24 h after the last drug treatment. Tissues were fixed in 4% paraformaldehyde (24 h) and paraffin embedded. Cut tissue sections of 5 μm were mounted on glass slides, rehydrated in water, and stained with hematoxylin and eosin (H&E; Sigma-Aldrich, St. Louis, MO). All tissues were examined, and microphotographs of sections at a ×20 magnification were taken from the stained samples using an inverted IX70 microscope (Olympus, Hamburg, Germany).
Murine model of pulmonary aspergillosis.
A. fumigatus strain A1160 was cultured on peptone-dextrose agar (PDA) and incubated at 37°C for 5 days. A conidial suspension was obtained by flooding the culture plate with 5 ml of saline with 0.05% Tween 20, scraping the fungal growth with a culture loop, and drawing up the resultant suspension with a Pasteur pipette. The suspension was then filtered to remove clumps of hypha or agar, and the inoculum was adjusted by hemocytometer count and serial dilution to the desired concentration. Animals were immunosuppressed by subcutaneous injection of cortisone acetate (125 mg/kg), starting 4 days prior to infection and then every 3 days (37). Infection was performed by nasal instillation of 5 × 104 CFU/mouse in 20 μl prior to anesthesia with inhaled sevoflurane.
Efficacy study.
Groups of 16 animals each received LAMB or AMB:DCH 1:1.5. Both formulations were administered i.v. at a dose of 5 mg/kg QD for 7 days. The control group received placebo (5% dextrose). Eight animals per group were used for CFU determination (+8 days), and the other 8 animals were used for survival follow-up for 12 days.
Twelve days postinfection, the surviving animals from the CFU groups were euthanized by CO2 inhalation followed by cervical dislocation. Kidneys and lungs were aseptically removed, homogenized in 0.9% saline, 10-fold diluted, and placed on PDA plates for determination of CFU per gram (+12 days).
Statistical analysis.
Results related to particle size studies are expressed as mean values ± standard deviation from three measures and tested by the independent-sample t test. Results related to AMB concentrations were tested by the Mann-Whitney test. Fungal burden data were analyzed by the Mann-Whitney test, and survival curves were compared using the log rank test. Statistical differences were performed via one-way analysis of variance (ANOVA) using Minitab 15 (Minitab, Ltd., Coventry, United Kingdom). Tukey's test was used for paired-group comparisons. A P value of <0.05 was considered to indicate statistical significance.
ACKNOWLEDGMENTS
This research was jointly funded by a project of the R + D + i research program (reference SAF2015-66690-R) and a grant from the Complutense University of Madrid and the Administration of the Community of Madrid for the research group 910939. This work was supported by Spanish grants from the Ministry of Economy and Competitiveness ISCIII-FIS (PI14/01195, PI17/00276), cofinanced by FEDER Funds from the European Commission, “A Way of Making Europe,” ISCIII-RETIC REDinREN/RD16/0009/0026 and Comunidad de Madrid B2017/BMD-3686 (CIFRA2-CM).
REFERENCES
- 1.Fortún J, Meije Y, Fresco G, Moreno S. 2012. Aspergillosis. Clinical forms and treatment. Enferm Infecc Microbiol Clin 30:201–208. doi: 10.1016/j.eimc.2011.12.005. [DOI] [PubMed] [Google Scholar]
- 2.Voltan AR, Quindós G, Medina KP, Fusco-Almeida AM, Soares MJ, Chorilli M. 2016. Fungal diseases: could nanostructured drug delivery systems be a novel paradigm for therapy? Int J Nanomedicine 11:3715–3730. doi: 10.2147/IJN.S93105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Walsh TJ, Anaissie EJ, Denning DW, Herbrecht R, Kontoyiannis DP, Marr KA. 2008. Treatment of aspergillosis: clinical practice guidelines of the Infectious Diseases Society of America. Clin Infect Dis 70:327–360. doi: 10.1086/525258. [DOI] [PubMed] [Google Scholar]
- 4.Clemons KV, Schwartz JA, Stevens DA. 2012. Experimental central nervous system aspergillosis therapy: efficacy, drug levels and localization, immunohistopathology, and toxicity. Antimicrob Agents Chemother 56:4439–4449. doi: 10.1128/AAC.06015-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wang Y, Ke X, Voo ZX, Yap SSL, Yang C, Gao S, Liu S, Venkataraman S, Obuobi SAO, Khara JS, Yang YY, Ee PLR. 2016. Biodegradable functional polycarbonate micelles for controlled release of amphotericin B. Acta Biomater 46:211–220. doi: 10.1016/j.actbio.2016.09.036. [DOI] [PubMed] [Google Scholar]
- 6.Moreno-Rodríguez AC, Torrado-Durán S, Molero G, García-Rodríguez JJ, Torrado-Santiago S. 2015. Efficacy and toxicity evaluations of new amphotericin B micelle systems for brain fungal infections. Int J Pharm 494:17–22. doi: 10.1016/j.ijpharm.2015.08.003. [DOI] [PubMed] [Google Scholar]
- 7.Olson JA, Adler-Moore JP, Jensen GM, Schwartz J, Dignani MC, Proffitt RT. 2008. Comparison of the physicochemical, antifungal, and toxic properties of two liposomal amphotericin B products. Antimicrob Agents Chemother 52:259–268. doi: 10.1128/AAC.00870-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Nahar M, Mishra D, Dubey V, Kumar Jain N. 2008. Development, characterization and toxicity evaluation of amphotericin B-loaded gelatin nanoparticles. Nanomedicine 4:252–261. doi: 10.1016/j.nano.2008.03.007. [DOI] [PubMed] [Google Scholar]
- 9.Ishida K, Castro RA, Torrado JJ, Serrado DR, Pereira Borba-Santos L, Pereira Quintella L, de Souza W, Rozental S, Lopes-Bezerra LM. 2018. Efficacy of a poly-aggregated formulation of amphotericin B in treating systemic sporotrichosis caused by Sporothrix brasiliensis. Med Mycol 56:288–296. doi: 10.1093/mmy/myx040. [DOI] [PubMed] [Google Scholar]
- 10.Mariné M, Espada R, Torrado JJ, Pastor FJ, Guarro J. 2009. Efficacy of a new formulation of amphotericin B in murine disseminated infections by Candida glabrata or Candida tropicalis. Int J Antimicrob Agents 34:566–569. doi: 10.1016/j.ijantimicag.2009.07.005. [DOI] [PubMed] [Google Scholar]
- 11.Serrano DR, Hernández L, Fleire L, González-Alvarez I, Montoya A, Ballesteros MP, Dea-Ayuela MA, Miró G, Bolás-Fernández F, Torrado JJ. 2013. Hemolytic and pharmacokinetic studies of liposomal and particulate amphotericin B formulations. Int J Pharm 447:38–46. doi: 10.1016/j.ijpharm.2013.02.038. [DOI] [PubMed] [Google Scholar]
- 12.Souza AC, Nascimiento AL, Vasconcelos NM, Jerônimo MS, Siqueira IM, Santos R-L, Cintra DO, Fuscaldi LL, Pires Júnior OR, Titze-de-Almeida R, Borin MF, Báo SN, Martins OP, Cardoso VN, Fernandes SO, Mortari MR, Tedesco AC, Amaral AC, Felipe MS, Bocca AL. 2015. Activity and in vivo tracking of amphotericin B loaded PLGA nanoparticles. Eur J Med Chem 95:267–276. doi: 10.1016/j.ejmech.2015.03.022. [DOI] [PubMed] [Google Scholar]
- 13.Zhao M, Hu J, Zhang L, Zhang L, Sun Y, Ma N, Chen X, Gao Z. 2014. Study of amphotericin B magnetic liposomes for brain targeting. Int J Pharm 475:9–15. doi: 10.1016/j.ijpharm.2014.08.035. [DOI] [PubMed] [Google Scholar]
- 14.Lewis RE, Albert ND, Liao G, Hou J, Prince RA, Kontoyiannis DP. 2010. Comparative pharmacodynamics of amphotericin B lipid complex and liposomal amphotericin B in a murine model of pulmonary mucormycosis. Antimicrob Agents Chemother 54:1298–1304. doi: 10.1128/AAC.01222-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Espada R, Valdespina S, Alfonso C, Rivas G, Ballesteros MP, Torrado JJ. 2008. Effect of aggregation state on the toxicity of different amphotericin B preparations. Int J Pharm 36:64–69. doi: 10.1016/j.ijpharm.2008.05.013. [DOI] [PubMed] [Google Scholar]
- 16.Faustino C, Serafim C, Ferreira I, Pinheiro L, Calado A. 2015. Solubilization power of amino acid-based Gemini surfactant towards the hydrophobic drug amphotericin B. Colloids Surf A Physicochem Eng Asp 480:426–432. doi: 10.1016/j.colsurfa.2014.11.039. [DOI] [Google Scholar]
- 17.Zu Y, Sun W, Zhao X, Wang W, Li Y, Ge Y, Liu Y, Wang K. 2014. Preparation and characterization of amorphous amphotericin B nanoparticles for oral administration through liquid antisolvent precipitation. Eur J Pharm Sci 53:109–117. doi: 10.1016/j.ejps.2013.12.005. [DOI] [PubMed] [Google Scholar]
- 18.Tan SW, Billa N, Roberts CR, Burley JC. 2010. Surfactant effects on the physical characteristics of amphotericin B-containing nanostructured lipid carriers. Colloids Surf A Physicochem Eng Asp 372:73–79. doi: 10.1016/j.colsurfa.2010.09.030. [DOI] [Google Scholar]
- 19.Jung SH, Lim DH, Jung SH, Lee JE, Jeong KS, Seong H, Shin BC. 2009. Amphotericin B-entrapping lipid nanoparticles and their in vitro and in vivo characteristics. Eur J Pharm Sci 37:313–320. doi: 10.1016/j.ejps.2009.02.021. [DOI] [PubMed] [Google Scholar]
- 20.Clark JM, Whitney RR, Olsen SJ, George RJ, Swerdel MR, Kunselman L, Bonner DP. 1991. Amphotericin B lipid complex therapy of experimental fungal infections in mice. Antimicrob Agents Chemother 35:615–621. doi: 10.1128/AAC.35.4.615. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Wasan KM, Grossie VB Jr, Lopez-Berestein G. 1994. Concentrations in serum and distribution in tissue of free and liposomal amphotericin B in rats during continuous intralipid infusion. Antimicrob Agents Chemother 38:2224–2226. doi: 10.1128/AAC.38.9.2224. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Olson JA, Adler-Moore JP, Schwartz J, Jensen GM, Proffitt RT. 2006. Comparative efficacies, toxicities, and tissue concentrations of amphotericin B lipid formulations in a murine pulmonary aspergillosis model. Antimicrob Agents Chemother 50:2122–2131. doi: 10.1128/AAC.00315-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zia Q, Khan AA, Swaleha Z, Owais M. 2015. Self-assembled amphotericin B-loaded polyglutamic acid nanoparticles: preparation, characterization and in vitro potential against Candida albicans. Int J Nanomedicine 10:1769–1790. doi: 10.2147/IJN.S63155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Fernández-García R, de Pablo E, Ballesteros MP, Serrano DR. 2017. Unmet clinical needs in the treatment of systemic fungal infections: the role of amphotericin B and drug targeting. Int J Pharm 525:139–148. doi: 10.1016/j.ijpharm.2017.04.013. [DOI] [PubMed] [Google Scholar]
- 25.Adler-Moore JP, Proffitt RT. 2008. Amphotericin B lipid preparations: what are the differences? Clin Microbiol Infect 14:25–36. doi: 10.1111/j.1469-0691.2008.01979.x. [DOI] [PubMed] [Google Scholar]
- 26.Lass-Flörl C, Alastruey-Izquierdo A, Cuenca-Estrella M, Perkhofer S, Rodriguez-Tudela JL. 2009. In vitro activities of various antifungal drugs against Aspergillus terreus: global assessment using the methodology of the European Committee on Antimicrobial Susceptibility Testing. Antimicrob Agents Chemother 53:794–795. doi: 10.1128/AAC.00335-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lewis RE, Liao G, Hou J, Chamilos G, Prince RA, Kontoyiannis DP. 2007. Comparative analysis of amphotericin B lipid complex and liposomal amphotericin B kinetics of lung accumulation and fungal clearance in a murine model of acute invasive pulmonary aspergillosis. Antimicrob Agents Chemother 51:1253–1258. doi: 10.1128/AAC.01449-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Takemoto K, Yamamoto Y, Ueda Y, Sumita Y, Yoshida K, Niki Y. 2006. Comparative study on the efficacy of AmBisome and Fungizone in a mouse model of pulmonary aspergillosis. J Antimicrob Chemother 57:724–731. doi: 10.1093/jac/dkl005. [DOI] [PubMed] [Google Scholar]
- 29.Lewis RE, Chamilos G, Prince RA, Kontoyiannis DP. 2007. Pretreatment with empty liposomes attenuates the immunopathology of invasive pulmonary aspergillosis in corticosteroid-immunosuppressed mice. Antimicrob Agents Chemother 51:1078–1081. doi: 10.1128/AAC.01268-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Olson JA, Schwartz JA, Hahka D, Nguyen N, Bunch T, Jensen GM, Adler-Moore JP. 2015. Toxicity and efficacy differences between liposomal amphotericin B formulations in uninfected and Aspergillus fumigatus infected mice. Med Mycol 53:107–118. doi: 10.1093/mmy/myu070. [DOI] [PubMed] [Google Scholar]
- 31.Barchiesi F, Santinelli A, Biscotti T, Greganti G, Giannini D, Manso E. 2016. Delay of antifungal therapy influences the outcome of invasive aspergillosis in experimental models of infection. J Antimicrob Chemother 71:2230–2233. doi: 10.1093/jac/dkw111. [DOI] [PubMed] [Google Scholar]
- 32.Lewis RE, Albert ND, Kontoyiannis DP. 2014. Comparative pharmacodynamics of posaconazole in neutropenic murine models of invasive pulmonary aspergillosis and mucormycosis. Antimicrob Agents Chemother 58:6767–6772. doi: 10.1128/AAC.03569-14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Gavaldà J, Martín T, López P, Gomis X, Ramírez JL, Rodríguez D, Len O, Puigfel Y, Ruiz I, Pahissa A. 2005. Efficacy of high loading doses of liposomal amphotericin B in the treatment of experimental invasive pulmonary aspergillosis. Clin Microbiol Infect 11:999–1004. doi: 10.1111/j.1469-0691.2005.01262.x. [DOI] [PubMed] [Google Scholar]
- 34.Groll AH, Lyman CA, Petraitis V, Petraitiene R, Armstrong D, Mickiene D, Alfaro RM, Schaufele RL, Sein T, Bacher J, Walsh TJ. 2006. Compartmentalized intrapulmonary pharmacokinetics of amphotericin B and its lipid formulations. Antimicrob Agents Chemother 50:3418–3423. doi: 10.1128/AAC.00241-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Clemons KV, Espiritu M, Parmar R, Stevens DA. 2005. Comparative efficacies of conventional amphotericin B, liposomal amphotericin B (AmBisome), caspofungin, micafungin, and voriconazole alone and in combination against experimental murine central nervous system aspergillosis. Antimicrob Agents Chemother 49:4867–4875. doi: 10.1128/AAC.49.12.4867-4875.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Adler-Moore JP, Gangneux JP, Pappas PG. 2016. Comparison between liposomal formulations of amphotericin B. Med Mycol 54:223–231. doi: 10.1093/mmy/myv111. [DOI] [PubMed] [Google Scholar]
- 37.Dixon DM, Polak A, Walsh TJ. 1989. Fungus dose-dependent primary pulmonary aspergillosis in immunosuppressed mice. Infect Immun 57:1452–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Corral MJ, Serrano DR, Inmaculada Moreno I, Torrado JJ, Domínguez M, Alunda JM. 2014. Efficacy of low doses of amphotericin B plus allicin against experimental visceral leishmaniasis. J Antimicrob Chemother 69:3268–3274. doi: 10.1093/jac/dku290. [DOI] [PubMed] [Google Scholar]