Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, November 2007, p. 4133-4140, Vol. 51, No. 11
0066-4804/07/$08.00+0 doi:10.1128/AAC.00669-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Portland Veterans Affairs Medical Center, Portland, Oregon 97239,1 Department of Chemistry, Portland State University, Portland, Oregon 97207,2 Oregon Health and Science University, Portland, Oregon 97201,3 Department of Biological Sciences, Old Dominion University, Norfolk, Virginia 235294
Received 22 May 2007/ Returned for modification 25 July 2007/ Accepted 27 August 2007
|
|
|---|
|
|
|---|
![]() View larger version (10K): [in a new window] |
FIG. 1. Chemical structures of selected antimalarials and chemosensitizers.
|
While the mechanism of chemosensitization is not fully understood, recent studies suggest that mutations in the P. falciparum CQ resistance transporter (PfCRT) protein, particularly amino acid substitutions at position 76, may play key roles in the mode of action of verapamil (14, 18, 37). Structure-activity profiling and three-dimensional quantitative structure-activity relationship (QSAR) studies by Bhattacharjee and colleagues revealed a pharmacophore with critical features for potent CQ-chemosensitizing activity, which consists of two aromatic hydrophobic groups and a hydrogen bond acceptor site at the side chain, preferably on a nitrogen atom (8, 9, 25).
Our previous work described functionalized tricyclic xanthones that exert their antimalarial activities by accumulation in the acidic digestive vacuole of the parasite and formation of soluble complexes with heme (29, 30, 33-35, 50, 65). Here, we switched to the acridone nucleus to facilitate the attachment of a suitable R group (e.g., alkyl amine) at the 10-N position for chemosensitization function (Fig. 2). A further motivation for switching to the acridone system is for the design of a dually functional agent, with structural modifications to facilitate binding to heme and to the central nitrogen atom to introduce chemosensitization. The present paper focuses attention on functionalizing the acridone nucleus for the chemosensitization phenomenon. A series of novel 10-N-substituted acridones (Fig. 3), bearing alkyl side chains (with lengths ranging from two to eight carbons) with tertiary amino groups at the terminal position, were designed, synthesized, and evaluated for the ability to enhance the potencies of quinoline drugs against MDR malaria parasites.
![]() View larger version (7K): [in a new window] |
FIG. 2. Design pathway of acridone chemosensitizers.
|
![]() View larger version (8K): [in a new window] |
FIG. 3. Chemical structures of profiled 10-N-substituted acridone derivatives.
|
|
|
|---|
Synthesis of 10-N-substituted acridones. Methods for the chemical syntheses of 10-N-substituted acridones are illustrated in Fig. 4. The two-carbon chain derivative was prepared from 9(10H)-acridone by refluxing with 2-(diethylamino)ethyl chloride hydrochloride in anhydrous acetone in the presence of K2CO3. N-alkylation of the ring nitrogen with longer side chains (n > 2) was achieved in two steps. First was the reaction of 9(10H)-acridone with 1,n-chlorobromoalkanes in the presence of potassium carbonate in anhydrous acetone under reflux conditions to yield the corresponding N-10-(n-chloroalkyl)-acridones. Following isolation of this intermediate, the N-10-(n-chloroalkyl)-acridones were stirred with diethylamine in dimethyl sulfoxide in the presence of NaI, leading to formation of the desired alkylamino-substituted acridones. The final product was purified by crystallization or column chromatography and analyzed for identity and purity by proton nuclear magnetic resonance (NMR) and combustion analysis prior to use.
![]() View larger version (8K): [in a new window] |
FIG. 4. Synthetic routes for 10-N-substituted acridone derivatives.
|
In vitro drug susceptibility testing in P. falciparum. In vitro antimalarial activity was determined by a malaria SYBR Green I-based fluorescence (MSF) method described previously by Smilkstein et al. (54) with slight modification (66). Stock solutions of each test drug were prepared in sterile distilled water at a concentration of 10 mM. The drug solutions were serially diluted with culture medium and distributed to asynchronous parasite cultures on 96-well plates in quadruplicate in a total volume of 100 µl to achieve 0.2% parasitemia with a 2% hematocrit in a total volume of 100 µl. Automated pipetting and dilution were carried out with a programmable Precision 2000 robotic station (Bio-Tek, Winooski, VT). The plates were then incubated for 72 h at 37°C. After incubation, 100 µl of lysis buffer with 0.2 µl/ml SYBR Green I (54, 66) was added to each well. The plates were incubated at room temperature for an hour in the dark and then placed in a 96-well fluorescence plate reader (Spectramax Gemini-EM; Molecular Diagnostics) with excitation and emission wavelengths at 497 nm and 520 nm, respectively, for measurement of fluorescence. The 50% inhibitory concentration (IC50) was determined by nonlinear regression analysis of logistic dose-response curves (GraphPad Prism software).
Alamar Blue assay for mammalian cell viability. The general cytotoxic effects of acridone derivatives on host cells were assessed by functional assay as described previously (1, 66, 69), using murine splenic lymphocytes induced to proliferate and differentiate by concanavalin A. Splenic lymphocytes isolated from C57BL/6J mice were washed twice in RPMI 1640 medium and resuspended in complete RPMI containing 10% fetal bovine serum, 50 µg/ml penicillin/streptomycin, 50 µM ß-mercaptoethanol, and 1 µg/ml concanavalin A. Cells (100 µl/well) then were seeded into 96-well flat-bottom tissue culture plates containing drug solutions (100 µl) serially diluted with complete culture medium to a final cell density of 2 x 105 per well. The plates were then incubated for 72 h in a humidified atmosphere at 37°C and 5% CO2. An aliquot of a stock solution of resazurin (Alamar Blue, prepared in 1x phosphate-buffered saline) was then added at 20 µl per well (final concentration, 10 µM), and the plates were returned to the incubator for another 24 h. After this period, the fluorescence in each well was measured in a Gemini EM plate reader with an excitation wavelength at 560 nm and an emission wavelength at 590 nm. IC50s were determined by nonlinear regression analysis of logistic concentration-fluorescence intensity curves (GraphPad Prism software).
Evaluation of the in vitro chemosensitizing activity of acridones. For drug interaction studies, a fixed subinhibitory concentration (500 nM) of the selected acridone was combined with the pre-serially diluted quinoline-containing drug. The effects of the acridones on the activities of quinoline antimalarials were assessed by comparing concentration-response curves for quinoline alone and in the presence of the selected acridone. The data were transformed and expressed as the response modification index (RMI). The RMI is defined as the ratio of the IC50 of drug A in the presence of drug B to the IC50 of drug A alone, where drug A is the quinoline antimalarial and drug B is the selected acridone. An RMI of 1.0 indicates no change in the IC50 of drug A upon addition of drug B. An RMI of less than 1.0 represents chemosensitization (including possible synergy), and an RMI of more than 1.0 represents antagonism (42). This method is widely used in the field as an initial screening tool for chemosensitization effect, mainly due to its ease of operation.
Effects of PfCRT mutations on chemosensitization activities of KF-A6 on CQ and QN. Details of the single-step selection of the P. falciparum pfcrt mutant lines 106/176I, 106/176N, and 106/176T have been described previously by Cooper et al. (14). The IC50s of drugs alone or in combination (e.g., CQ/KF-A6, CQ/verapamil, QN/KF-A6, and QN/verapamil) were determined by a modified MSF method in which 1% parasitemia and 1% hematocrit were used.
Fixed-ratio isobologram analysis of drug interactions between CQ and acridones.
For definitive determination of synergy, selected acridones were tested in combination with CQ using a modified fixed-ratio method described by Fivelman et al. (20). After determination of the IC50s for selected drugs, stock solutions were prepared with each drug at concentrations such that the final concentration in our 96-well drug susceptibility assay after four or five twofold dilutions approximated the IC50. If we call these stock solutions drug A and drug B, then six final stock solutions were prepared from this initial stock: drug A alone, drug B alone, and volume-volume mixtures of drugs A and B in the following ratios: 4:1, 3:2, 2:3, and 1:4. Twofold dilutions of each of the six final stock solutions were performed robotically across a 96-well plate in quadruplicate. Subsequent steps were typical of the standard drug susceptibility methods as described above. Initial data analysis yielded the intrinsic dose-response curve for each drug alone and four different fixed-ratio combination dose-response curves, with corresponding IC50s. The fractional inhibitory concentrations (FICs) were then calculated by the following formulas: FIC (A) = IC50 of drug A in combination/IC50 of drug A alone; FIC (B) = IC50 of drug B in combination/IC50 of drug B alone; FIC index = FIC (A) + FIC (B). The isobolograms were constructed by plotting a pair of FICs for each combination of CQ and the selected compound. Traditionally, an interpretation of a straight diagonal line (FIC index = 1) on the isobologram indicates a purely additive effect between the two drugs. A concave curve below the line (FIC index < 1.0) indicates synergy of the combination, while a convex curve above the line (FIC index > 1.0) indicates antagonism. However, conventions have been adopted that define a FIC index of
0.5 as synergy, a FIC index of
4.0 as antagonism, and a FIC index between 0.5 and 4.0 as no interaction or indifference (2, 41). Isobologram analysis with fixed-ratio combinations is a more sophisticated and disciplined method to establish the definitive determination and the degree of synergy (or antagonism) for drug interaction.
Effects of acridone derivatives on inhibition of substrate uptake by recombinant hDAT, hSERT, and hNET transporters. The uptake assays employed for this study were adapted from the method of Eshleman et al. (17). Human embryonic kidney (HEK)-human dopamine (hDAT), -serotonin (hSERT), and -norepinephrine (hNET) cells were grown on 150-mm-diameter tissue culture dishes. The medium was removed, and the plates were washed twice with Ca2+- and Mg2+-free phosphate-buffered saline. The cells were gently scraped from the plates in Krebs-HEPES assay buffer, and cell clusters were separated by trituration with a pipette. Aliquots (50 µl) of the suspended cells were added to assay tubes containing drugs and Krebs-HEPES assay buffer in a final volume of 0.5 ml. Competition experiments were conducted with triplicate determinations for each point. After a 10-min preincubation in a 25°C water bath, 3H-labeled neurotransmitter (20 nM [final concentration] of [3H]DA (dopamine), [3H]5-HT (serotonin), or [3H]NE (norepinephrine), 56, 26.9, or 60 Ci/mmol, respectively) was added, and the assay mixture was incubated for 10 min at 25°C. The reaction was terminated by filtration through Wallac filtermat A filters presoaked in 0.05% polyethylenimine, using a Tomtec cell harvester. Scintillation fluid was added to each filtered spot, and radioactivity remaining on the filters was determined using a Wallac 1205 Betaplate scintillation counter. Specific uptake was defined as the difference in uptake observed in the absence and presence of 5 µM mazindol (hDAT and hNET) or 5 µM imipramine (hSERT). Prism software was used to analyze the uptake data.
Effects of acridone derivatives on radioligand binding to histamine receptors. [3H]pyrilamine binding assays were conducted using the modified radioligand binding methods of Janowsky et al. (31). [3H]pyrilamine and recombinant human histamine receptor subtype 1, expressed in CHO cells, were purchased from Perkin-Elmer Life and Analytical Sciences (Boston, MA). [3H]pyrilamine binding assay mixtures contained an aliquot of a membrane preparation (approximately 50 µg protein), drug, and [3H]pyrilamine (5 nM final concentration) in a final volume of 500 µl. Tris-HCl buffer (50 mM; pH 7.4) was used for all assays. Specific binding was defined as the difference in binding observed in the presence and absence of 10 µM hydroxyzine. Membranes were preincubated with drugs at room temperature for 10 min before the addition of [3H]pyrilamine unless otherwise indicated. The reaction mixture was incubated for 90 min at room temperature in the dark and was terminated by filtration through Wallac Filtermat A filters (the filters were soaked in 0.05% polyethyeneimine for 15 min prior to filtration), using a 96-well Tomtec cell harvester. Scintillation fluid (50 µl) was added to each filtered spot, and radioactivity remaining on the filter was determined using a Wallac 1205 Betaplate or 1405 microBeta scintillation counter. Competition experiments were conducted with duplicate determinations for each point. GraphPad Prism software was used to analyze all kinetic and competition binding data.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Intrinsic in vitro antimalarial activities, in vitro cytotoxicities, and biophysical properties of selected chemosensitizers
|
The calculated biophysical properties of these acridones suggest that the compounds are moderately lipophilic at physiological pH, with cLogP values from 3.11 to 5.49 (Table 1). As indicated by pKa values ranging from 9.81 to 10.65 (Table 1), the acridone derivatives with the ionizable alkyl amine side chains, KF-A2, KF-A3, KF-A4, KF-A5, KF-A6, and KF-A8, exist as a mixture of unprotonated and protonated forms at pH 7 (physiological conditions) and predominantly as positively charged forms at pH values close to 5 (approximating conditions of the parasite food vacuole) (7, 16, 23, 24, 26, 36, 68).
Chemosensitization of CQ in MDR P. falciparum. To examine the abilities of the acridone derivatives to enhance CQ potency against P. falciparum, the IC50s of CQ were determined in combination with 500 nM of the test acridone against the CQR strain Dd2 and the CQS strain D6 (Table 2). At the selected drug concentration, the acridones with a tertiary amine at the ends of the various-length carbon chains exhibited CQ-chemosensitizing activities against an MDR strain of P. falciparum (Dd2) ranging from indifferent (KF-A2) to partial (KF-A3 and KF-A8) to considerable (KF-A4, KF-A5, and KF-A6). In the presence of 500 nM KF-A6, the antimalarial activity of CQ was potentiated >80-fold in the MDR strain (Dd2) to a level (19.0 nM) similar to that observed for the CQS strain, D6 (14.7 nM). As revealed in Table 2, KF-A5 and KF-A6 exhibited a CQ-chemosensitization effect (against the CQR strain Dd2) superior to those of the well-studied chemosensitizers verapamil and chlorpheniramine and comparable to the effect of desipramine at the equivalent drug concentration. There was no apparent enhancement of the sensitivity of CQ against the CQS strain (D6) with either the acridone derivatives or the reference chemosensitizers (Table 2).
|
View this table: [in a new window] |
TABLE 2. In vitro chemosensitizing effects of acridone derivatives and other known chemosensitizers on CQ against P. falciparum
|
Isobolar analysis of drug interactions between CQ and selected chemosensitizers. To assess whether the degree of chemosensitization met the current definition of synergy, in vitro interaction studies between CQ and selected acridone derivatives against both CQR (Dd2) and CQS (D6) parasites were evaluated by the more sophisticated isobolar analysis using the fixed-ratio combination method (Fig. 5). In these studies, significant synergy was observed in the combinations of CQ/KF-A4, CQ/KF-A5, and CQ/KF-A6 against the CQR parasite Dd2, with mean FIC indices of 0.3, 0.27, and 0.25, respectively. The CQ/KF-A4, CQ/KF-A5, and CQ/KF-A6 combinations exhibited superior synergy compared to the reference drug combination CQ/verapamil (FIC index = 0.41) against the CQR parasite Dd2.
![]() View larger version (16K): [in a new window] |
FIG. 5. Isobologram of the in vitro interaction between CQ and chemosensitizers against the MDR strain of P. falciparum, Dd2 (A), and the CQS strain of P. falciparum, D6 (B). The x axis represents the FICs of CQ, and the y axis represents the FICs of selected chemosensitizers. Each point represents a FIC of the respective chemosensitizer and its corresponding FIC of CQ. The diagonal (dashed) line indicates the hypothetical additive drug effect.
|
Chemosensitization to DCQ and QN in MDR P. falciparum. The lead acridone derivative KF-A6 enhances the sensitivity of other quinoline-containing antimalarials, DCQ (a major metabolite of CQ) and QN, in the MDR parasite Dd2 but has no effect on the CQS strain D6 (Table 3). In the presence of 500 nM KF-A6, the IC50 of DCQ against Dd2 was lowered more than 80% and the IC50 of QN against Dd2 was lowered to approximately the same level as that observed for the sensitive strain, D6.
|
View this table: [in a new window] |
TABLE 3. In vitro chemosensitizing effects of KF-A6 on DCQ and QN against P. falciparum
|
Table 4 shows the effect of PfCRT position 76 mutations on the chemosensitizing activity of KF-A6 and verapamil when tested in combination with either CQ or QN. The results reveal mutation-specific changes in intrinsic sensitivity and chemosensitizing activity for all drugs. Verapamil and KF-A6 showed moderate intrinsic antimalarial activity, with micromolar IC50s against the mutant lines 106/176T and 106/176I. In the 106/176N line, however, a much greater intrinsic sensitivity to KF-A6 and verapamil was observed when the drugs were tested alone. The potency of CQ was enhanced by both drugs against all tested mutant lines, and the potency of QN was enhanced by both drugs against the mutant lines 106/176T and 106/176N. KF-A6 was a more potent chemosensitizing agent (for both CQ and QN) than verapamil in these cases. Most significantly, the unique inhibitory activity of verapamil against QN in the 106/176I line was also observed with KF-A6. Against all parasite lines, the chemosensitization patterns of KF-A6 on CQ and QN mirrored that of verapamil.
|
View this table: [in a new window] |
TABLE 4. Effects of PfCRT position 76 mutations on in vitro chemosensitization of CQ and QN by KF-A6 and verapamil in P. falciparum
|
As shown in Table 5, the well-studied chemosensitizers tricyclic antidepressants desipramine and imipramine showed high specificity for blocking the uptake of serotonin and norepinephrine. In contrast, acridone derivatives showed essentially no effect on all three transporters.
|
View this table: [in a new window] |
TABLE 5. Inhibition of [3H]neurotransmitter uptake in HEK-hDAT, HEK-hSERT, and HEK-hNET cells and inhibition of [3H]pyrilamine binding to histamine H1 receptor
|
|
|
|---|
The discussion of possible mechanisms of action for acridones and chemosensitizers in general to reverse quinoline resistance undoubtedly needs to start with the antimalarial mode of action of the quinolines and the mechanism of resistance to these drugs in malaria parasites. Unfortunately, despite years of use and study, both mechanisms remain unresolved. However, it is widely accepted that the site of action for CQ is the acidic food vacuole, where CQ accumulates via acid trapping. It has also been proposed that CQ inhibits the formation of hemozoin, the detoxification product formed upon hemoglobin degradation, leading to heme-induced parasite death. CQ resistance appears to be the result of reduced drug accumulation in the food vacuole, leading to decreased drug access to the target heme (19), although various models for the explanation of this phenomenon are still under debate (12, 21, 44, 45). While specific P. falciparum quinoline-resistant phenotypes are likely to be multigenic in origin (15), the most important determinant of CQR is conferred by point mutations in a putative transporter protein, PfCRT, located in the food vacuole membrane of the parasite, with one amino acid change at position 76 playing a critical role (14, 15, 18, 60). Mutations within PfCRT, particularly changes from a charged amino acid residue (lysine, K76) to an uncharged residue (such as threonine [76T], asparagine [76N], or isoleucine [76I]), seem to be important not only in the acquisition of resistance to quinoline antimalarials (e.g., by allowing efflux of diprotic CQ), but also in the mechanism of resistance reversal actions for chemosensitizers (14, 15, 18, 37, 61, 62).
Similarly to the proposed mechanism of action for verapamil, the acridone chemosensitizers may compete for the CQ binding site in PfCRT and reduce CQ efflux from the food vacuole by a charge-repulsion effect, replacing the lost positive charge of K76 with a protonated amino group. Presumably, the weakly basic acridone derivatives (such as KF-A4, KF-A5, and KF-A6) diffuse across biological membranes of the parasite in the unprotonated form, and once they enter the acidic food vacuole, they are rapidly trapped via weak base properties (acid trapping) as ionized forms. In the acidic vacuole, these acridones may alter the pH of the environment or interfere with the efflux of CQ, restoring the efficacy of CQ and other quinoline antimalarials. While the effect of position 76 mutations on the chemosensitizing activity of KF-A6 offers evidence for a direct interaction with the PfCRT protein, the absence of chemosensitizing activity from KF-A6-Cl (lacking the protonable terminal tertiary amine group) further supports our hypothesis. Given the ease of chemical synthesis of 10-N-substituted acridones, additional studies to explore the relationship of structural modifications of acridone derivatives and their chemosensitization activities, most importantly in PfCRT mutants, could help us to further understand the molecular basis of chemosensitizers and possibly provide greater insight into the resistance mechanism of quinoline antimalarials.
A major problem with existing antimalarial chemosensitizers is their neurological side effects at the high doses required to achieve optimal chemosensitization effects. The dose-limiting toxic effects of the antidepressants and antihistamines are mediated by their modes of action, which involve the inhibition of neurotransmitter uptake by the biogenic amine transporters and competition with histamine for receptor sites on effector cells, respectively (17, 52, 56). We believe it is important to show that novel tricyclic chemosensitizers without these effects are possible, and therefore, we have adopted a screening model to assess the effects of drugs on the uptake of biogenic amines by recombinant hDAT, hSERT, and hNET transporters, as well as on binding to human histamine receptor H1. These well-established high-throughput assays represent a valuable aid for the assessment of chemosensitizers, as well as for down-selection of candidate compounds in antimalarial-drug development. The absence of effect on the uptake of biogenic amines or binding to the histamine receptor distinguishes these acridones from other known chemosensitizers (i.e., desipramine, imipramine, and chlorpheniramine) and demonstrates that chemosensitization potency is achievable without the undesirable collateral sequelae.
Chemosensitizers clearly remain of great importance in mechanistic studies, but the challenges of putting them into clinical use reach far beyond safety obstacles. The idea of restoring CQ usefulness is still enticing for the following reasons: (i) CQ remains the least expensive and most readily available antimalarial in Africa (22), (ii) CQ is safe to use in pregnant women and children, and (iii) withdrawal of CQ resistance in some areas raises the possibility of a CQ comeback (38). However, challenges remain, as a chemosensitizer would provide assistance to CQ only in combination therapy, and hence, a "cocktail" with an additional partner drug may be required, making it less efficient, more costly, and thus impractical for malaria treatment where resources are greatly limited. In recognition of this predicament and the need for combination therapy in the age of increasing drug resistance, studies are under way to investigate the feasibility of combining chemosensitization utility with intrinsic antimalarial potency in the same acridone molecule.
Published ahead of print on 10 September 2007. ![]()
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»