Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2021 Feb 25;6(9):6424–6437. doi: 10.1021/acsomega.1c00104

Physicochemical Profiling and Comparison of Research Antiplasmodials and Advanced Stage Antimalarials with Oral Drugs

Amritansh Bhanot 1, Sandeep Sundriyal 1,*
PMCID: PMC7948433  PMID: 33718733

Abstract

graphic file with name ao1c00104_0007.jpg

To understand the property space of antimalarials, we collated a large dataset of research antiplasmodial (RAP) molecules with known in vitro potencies and advanced stage antimalarials (ASAMs) with established oral bioavailability. While RAP molecules are “non-druglike”, ASAM molecules display properties closer to Lipinski’s and Veber’s thresholds. Comparison within the different potency groups of RAP molecules indicates that the in vitro potency is positively correlated to the molecular weight, the calculated octanol–water partition coefficient (clog P), aromatic ring counts (#Ar), and hydrogen bond acceptors. Despite both categories being bioavailable, the ASAM molecules are relatively larger and more lipophilic, have a lower polar surface area, and possess a higher count of heteroaromatic rings than oral drugs. Also, antimalarials are found to have a higher proportion of aromatic (#ArN) and basic nitrogen (#BaN) counts, features implicitly used in the design of antimalarial molecules but not well studied hitherto. We also propose using descriptors scaled by the sum of #ArN and #BaN (SBAN) to define an antimalarial property space. Together, these results may have important applications in the identification and optimization of future antimalarials.

1. Introduction

Different studies have estimated costs between 2.8 and 9.8 billion dollars to bring a new drug to the market.1,2 At the same time, the failure rate in the clinical studies remains high, that is attributable, inter alia, to poor bioavailability and safety.3 Involvement of such high stakes in the drug discovery calls for diligent efforts in understanding the factors affecting “druglikeness” or “druglike properties.”

Although the specific definition of “druglikeness” is debatable,4 broadly, it is akin to optimum oral bioavailability. The latter depends on the aqueous solubility, absorption, permeation, metabolic stability, and transporter-mediated efflux of a molecule, spawned by its interactions with several biomolecules and biomembranes in vivo. Thus, like the drug action, druglikeness is also a function of a molecule’s chemical structure or physicochemical profile. Consequently, on average, orally available drugs represent an amalgamation of optimum physicochemical properties required to interact favorably with the human physiology.

One of the earliest attempts to understand the influence of molecular properties on druglikeness was undertaken by Lipinski et al. with the introduction of the rule of five (Ro5).5 Lipinski’s Ro5 proposes a cutoff for four molecular properties, namely, molecular weight (MW < 500 Da), calculated partition coefficient (clog P < 5), hydrogen bond acceptor (HBA < 10), and hydrogen bond donor (HBD < 5), to assess the druglike characteristics. Molecules within the recommended limits of at least two of these four descriptors are expected to have good permeation and absorption, resulting in good oral bioavailability. Despite several limitations6 and criticisms,4,7 the Ro5 still seems to be a useful criterion to eliminate non-ideal molecules in the early phase of drug discovery.8,9 Since the seminal work by Lipinski, the role of other structural descriptors such as aromatic rings (#Ar),10−12 the fraction of sp3 carbon (Fsp3),13 topological polar surface area (TPSA),14,15 distribution coefficient (log D),16−18 and the number of rotatable bonds (#RB)14 has also been recognized to affect the “developability” of a molecule. Some authors have proposed using score-based and other quantitative druglikeness metrics instead of the rules with hard cutoffs.17,19 Additionally, mapping of compound optimization trajectories primarily based on the ligand efficiency and ligand lipophilic efficiency has also been recommended for successful drug hunting.20−22

Overall properties of drugs are also ought to be governed by the nature of their target. Indeed, drugs targeting different protein classes (such as kinases, nuclear hormone receptors, and proteases) possess variable properties.20,23 This is because of these targets’ distinct binding pockets requiring unique molecular size ranges, lipophilicity, ionization, or H-bonding capacity. For instance, compared to marketed oral drugs, orally available anticancer protein kinase inhibitors are larger, more lipophilic, and more complex.24 Similarly, orally used anti-infective drugs have higher MW, low lipophilicity, and greater HBA/HBD and ring counts.25−27 Specific properties may also be required to access a particular tissue/organ/organelle where a biological target might be residing, exemplified by the drugs acting on the central nervous system (CNS). In general, CNS drugs are smaller, lipophilic, and unionized than other drugs as there is a need for these molecules to overcome the blood–brain barrier.28,29 Thus, a target- or organ-specific chemical space exists within a broad oral drug space. Optimal properties required by a molecule to interact with its biological target (for in vitro potency) may or may not be orthogonal to those required for desirable oral bioavailability. Therefore, understanding oral drug space in a particular biological target context may provide useful insights that guide the drug design for the given target/biological endpoint.

Malaria is an infectious disease caused by the Plasmodium species belonging to the Apicomplexan phylum, spread by the mosquito bite. Malaria mostly affects tropical and sub-tropical populations, with children and pregnant women being the most vulnerable groups.30 Owing to the spread of resistant strains of the parasite, treatment of malaria involves a multi-drug regimen. In this context, the artemisinin combination therapy is regarded as a “gold standard,” and it is used as the first-line therapy in malaria treatment. However, resistance is spreading against artemisinin at alarming levels31−35 and may lead to devastating outcomes.36 These concerns have prompted large-scale high throughput screening (HTS) campaigns against Plasmodium falciparum, resulting in a large amount of data for retrospective learning and prospective predictions. These efforts are mostly based on phenotypic whole-cell assays against asexual or sexual stages of the parasite and have produced novel leads and clinical candidates.37−39 Despite the tremendous efforts in the past decade,40 only two small antimalarial molecules, tafenoquine and arterolane, have been approved in the past 20 years.

The antiplasmodial molecules act through different targets residing in different organelles such as parasite cell membranes, mitochondria, apicoplasts, food vacuoles, and the cytoplasm. Since the parasite inhabits host red blood cells (RBCs), the molecules active in antiplasmodial phenotypic assays must cross at least three membrane barriers. The latter consists of the host RBC membrane, parasitophorous vacuolar membrane (PVM), and parasite plasma membrane.41,42 Such a permeability barrier may impose specific properties to the active set of molecules compared to the inactive ones in these assays. For instance, large-scale phenotypic HTS by GlaxoSmithKline (GSK) found hit molecules to be larger and more lipophilic compared to the average source compound collection.43 Thus, it would be interesting to perform a systematic analysis of the property space of research antiplasmodial (RAP) molecules with differential potencies. Such comparison may reveal the key chemical descriptors important for allowing permeation across host/parasite lipoidal membranes or target engagement. However, cellular permeation alone is not enough to achieve optimum oral bioavailability properties. Also, research molecules are known to differ from clinical candidates and drugs in terms of physicochemical properties.21,44,45 Therefore, a comparison is required between RAP and the advanced stage antimalarial (ASAM) molecules with the proven in vivo oral bioavailability and efficacy. Such a comparison among RAP and ASAM would help map the trajectory as initial antimalarial hit advances from the discovery stage to the lead stage.20 To further characterize the antimalarial property space, comparison with other oral drugs is also required.

We collated and studied the average properties of the above-stated datasets. The results reveal interesting differences and similarities in RAPs, ASAMs, and oral drugs in the property space. Furthermore, we have characterized an antimalarial property space that may facilitate the identification of new antimalarial molecules.

2. Results and Discussion

2.1. Data Collection and Data Analysis

We used readily available open-source tools and resources for the data collection and the analysis. The set of RAP molecules was collated from the ChEMBL database, one of the largest collections of biologically active compounds reported in the medicinal chemistry literature.46,47 Additionally, the results of several phenotypic HTS campaigns against P. falciparum have also been deposited in ChEMBL by pharmaceutical companies like GSK.43 While compounds disclosed from such large screens may not be ideal for further development,48 such data may be used for the physicochemical profiling of antimalarials. Nevertheless, to ensure the quality of activity data, we have included compounds tested at multiple concentrations against the parasite with known IC50/EC50 values. Although several of these molecules are tested in different labs under different assay conditions, such heterogeneous data are acceptable for the qualitative comparison of bioactivities.49 These molecules were classified into different potency classes (HA, MA, and IN, see Methodology) to observe the effect of various properties on the in vitro potency. The HA dataset is the largest one since mostly successful results are reported in the literature. Due to the same reason, the IN class was found to have comparatively few molecules, and hence, the latter was topped up with the inactive molecules reported by GSK-Tres Cantos Antimalarial Set (TCAMS) screening (see Methodology).43

In addition to the marketed antimalarials, the ASAM set consists of antimalarials currently undergoing clinical trials and molecules considered “leads” with promising efficacy and oral bioavailability in animal studies.37,38 Thus, the difference between the RAP and ASAM molecular properties may indicate the influence these properties have on the “developability” of antimalarials.

For this study, the “oral drug” is defined as a small molecule (MW < 900 Da) currently approved by a regulatory body for oral administration to treat or prevent any disease in humans. The set of oral drugs was obtained from the DrugCentral50 database, which consists of drugs approved not only by the US FDA but also by the regulatory agencies in Europe, Japan, and other countries. The library was further updated with the recently approved drugs by the US FDA (till July 2020). Consequently, our library of oral drugs is extended (total 1954) in comparison to the recently compiled set of 750 oral drugs used for property profiling.4,8 The latter is limited to the oral drugs approved till 2017 by the US FDA.

While Lipinski suggested the cutoff of 500 Da for the MW, some authors have suggested that the actual limit for the MW may be higher for the orally absorbed drugs,27,51,52 prompting us to use a cutoff of 900 Da for the collation of all datasets. Using these criteria, the final datasets of IN, MA, HA, ASAM, and oral drugs consist of 7365, 6620, 10,557, 66, and 1954 molecules, respectively. Some of the molecules are present in more than one category. For example, several oral drugs are also part of the IN dataset. Similarly, currently marketed antimalarials are part of both ASAM as well as oral drug datasets.

The open-source program RDKit was used to calculate the MW, HBA, HBD, Fsp3, #RB, #Ar, and heteroaromatic ring count (#HetAr), while DataWarrior was used for the computation of the clog P, TPSA, carboaromatic ring count (#CarboAr), aromatic nitrogen count (#ArN), and basic nitrogen count (#BaN). Comparison among the different categories of molecules was performed using various statistical parameters and hypothesis tests employed earlier in similar studies.4,5,10,14 Given the skewness and kurtosis in the data (see Supporting Information), the Kruskal–Wallis test was employed for hypothesis testing, in addition to the t-test.4 Certain properties like clog P, HBA, HBD, and TPSA are known to be correlated with the MW.14 Hence, property trends were monitored for both large (MW > 500 Da) and small (MW < 500 Da) molecules within the given category (Supporting Information, Figures S3–S13). Expectedly, most of the molecules (∼80%) in our complete dataset belong to the latter class (Supporting Information, Table S2). The large molecules in the ASAM class possess only eight molecules, and hence, the results for this category should be interpreted with caution.

2.2. Comparison of the Globally Approved Oral Drugs with the FDA-Approved Drugs

Since our library consists of globally approved oral drugs, we compared it with the recently reported set of FDA-approved oral drugs. Most of the physicochemical properties can be computed unambiguously except for log P, which displays variable results based on the algorithm and computational programs used.4 For this study, we used the open-source Actelion clog P algorithm implemented in the DataWarrior program,53 which recognizes 368 atom types contributing toward the final value. This algorithm has been shown to outperform many other programs when tested on a dataset of 96,000 compounds.54 Moreover, a satisfactory correlation was observed for Actelion clog P versus the experimental log P (0.882) and Actelion clog P versus StarDrop clog P (0.935) (Supporting Information, Figures S1 and S2) for the set of 452 drugs compiled by Shultz.4

Despite the different compilation criteria and clog P algorithms, our extended set and the FDA-approved oral drugs show comparable 90th and 10th percentile values for important physicochemical properties (Table 1). The 90th percentiles for all properties, except for MW, are within Lipinski’s cutoffs for both the libraries. Consequently, 786 out of 1954 drugs (∼91%) in our library pass the Ro5. The drop in the 90th percentile of MW in our dataset (519.0 Da) in comparison to that of the FDA drugs (552.2 Da) may be due to the applied MW cutoff of 900 Da in the former case. Moreover, 90th percentiles of the TPSA and #RB of both libraries are also close to the limits proposed by Veber et al. for optimum bioavailability.14 The 90th and 10th percentiles for #Ar and Fsp3 descriptors are also identical for both libraries. The comparison of FDA-approved oral drugs before 19974 with the combined oral drugs demonstrates slight inflation in MW, clog P, and TPSA descriptors, in line with the earlier reports.4,21,25

Table 1. Comparison of 90th and 10th Percentiles of Various Molecular Propertiesa.

molecular property oral drugs (N = 1954) ASAM (N = 66) HA (N = 10,557) MA (N = 6620) IN (N = 7365)
MW 519.0 (204.2),b 552.2 (197.0),c 470.3 (171.2) 500.9 (253.6) 588.8 (296.6) 568.7 (277.8) 567.1 (242.4)
clog Pd 4.85 (−0.87),b 4.80 (−0.36),c 4.65 (−0.64) 5.54 (0.34) 6.43 (1.50) 5.78 (1.50) 5.64 (0.20)
HBA 9 (2),b 10 (2),c 10 (2) 8 (3) 9 (3) 9 (3) 10 (3)
HBD 4 (0),b 4 (0),c 4 (0) 4 (0) 4 (0) 3 (0) 4 (0)
TPSA 145.1 (27.4),b 143.3 (29.0),c 139.8 (21.3) 118.3 (43.2) 125.0 (33.1) 129.1 (33.5) 142.3 (34.9)
#Ar 3 (0),b 3 (0),c 3 (0) 4 (0) 4 (1) 4 (1) 4 (0)
#RB 10 (1),b 11 (1),c 10 (1) 9 (1) 12 (2) 11 (2) 12 (1)
Fsp3 0.78 (0.13),b 0.78 (0.13),c 0.83 (0.08) 0.94 (0.12) 0.71 (0.09) 0.59 (0.07) 0.87 (0.07)
Open in a new tab
a

The values in brackets represent 10th percentiles.

b

The 90th percentile values of all oral drugs (N = 750) approved by the US FDA for the period 1900–2017. Taken from the Supporting Information of ref (4).

c

The 90th percentile values for oral drugs (N = 341) approved before the proposal of the Ro5, that is, for the period 1900–1997. Taken from the Supporting Information of ref (4).

d

clog P values were calculated using the DataWarrior program for our dataset, while the StarDrop program was used in ref (4).

For charged molecules, log D may be more relevant than clog P as a measure of lipophilicity. This is evident from the application of the property forecast index and AbbVie multi-parametric scoring function in judging compound quality.16−18 However, our analysis is limited to clog P and other key properties as we do not have access to any commercial software for log D calculation. Also, to our knowledge, no open-source program is available to compute log D for an extensive database such as the one used in this study.

Overall, our oral drug library is updated with the most recent approvals and conforms to the property space of druglike compounds reported by other authors.

2.3. MW and clog P

MW has been shown to affect oral absorption, especially of hydrophilic drugs. The latter are mostly absorbed through paracellular spaces or cell junctions, which have a restricted size of 3–6 Å in humans. This is supported by the distinctive absorption kinetics of polar drugs observed in different species of animals and is attributable to the variation in the paracellular pore size.55 Nevertheless, Lipinski’s cutoff of MW may also arise from the limited number of large molecules pursued in drug discovery due to the challenges associated with their synthesis4,56 or due to the limit imposed by other descriptors correlated to MW.14,57 Lipophilicity affects the cellular uptake and oral absorption by influencing dissolution and partitioning of a drug into the lipid bilayer.

In RAP molecules, the mean (and median) MW increases with increasing antiplasmodial activity (Table 2, Figure 1). The HA and MA categories display significantly higher MW than the IN class. This trend is also visible in 90th and 10th percentile values for MW for these classes. The average MW of the ASAM group is lower (389.9 Da) compared to that of the HA class (432.1 Da) but higher than that of the oral drugs (357.2 Da), results statistically significant according to the t-test but not the Kruskal–Wallis test. The 90th percentile for the MW of the ASAM molecules (500.9 Da) is almost identical to the threshold of 500 Da suggested by Lipinski.

Table 2. Comparison of Mean/Median of Molecular Properties among the Different Categories of Molecules.

  oral drugs (N = 1954) ASAM (N = 66) HA (N = 10,557) MA (N = 6620) IN (N = 7365)
molecular property mean (median) mean (median) mean (median) mean (median) mean (median)
MW 357.2 (339.5) 389.9 (391.4) 432.2 (418.4) 408.8 (393.5) 384.0 (361.4)
clog P 2.23 (2.43) 3.07 (3.13) 3.94 (3.40) 3.60 (3.60) 2.90 (3.00)
HBA 5.48 (5) 5.60 (5.5) 5.80 (6) 5.68 (5) 5.82 (5)
HBD 1.90 (2) 2.10 (2) 1.70 (1) 1.52 (1) 1.85 (1)
TPSA 80.38 (72.34) 75.90 (73.11) 74.68 (69.30) 77.44 (71.44) 82.00 (73.12)
#Ar 1.70 (2) 2.12 (2) 2.73 (3) 2.65 (3) 1.97 (2)
#CarboAr 1.13 (1) 1.23 (1) 1.76 (2) 1.78 (2) 1.38 (1)
#HetAr 0.52 (0) 0.89 (1) 0.96 (1) 0.86 (1) 0.59 (0)
#RB 5.09 (4) 4.72 (4) 6.59 (6) 5.94 (5) 5.86 (5)
Fsp3 0.432 (0.4) 0.423 (0.375) 0.355 (0.310) 0.314 (0.285) 0.407 (0.363)
#BaN 0.62 (1) 1.09 (1) 1.01 (1) 0.66 (0) 0.71 (0)
#ArN 0.60 (0) 1.16 (1) 1.10 (1) 1.03 (0) 0.75 (0)
Open in a new tab

Figure 1.

Figure 1

Open in a new tab

Boxplots for the MW, clog P, HBA, and HBD properties for different sets of molecules. The mean values are given in bold above each boxplot and represented by the red line within the boxes. The yellow dots represent outliers.

Like MW, mean and 90th percentile values for clog P also show a steady increase from IN to MA to HA categories (Figure 1), suggesting a positive correlation between lipophilicity and antiplasmodial activity in phenotypic assays. However, like MW, the clog P of ASAM molecules also converges back to lower values while maintaining a statistically higher average than that of the oral drugs, as per the t-test. The trend is maintained for both low and high MW categories of RAP molecules, with average clog P showing an increase with increasing potency (Supporting Information, Figure S3).

In summary, bulkier and lipophilic molecules tend to show potent in vitro antiplasmodial activity, which agrees with GSK-TCAMS screening results.43 This means that in the currently used antiplasmodial phenotypic assays, membrane permeability of molecules is not adversely affected by their large size or high lipophilicity. Nevertheless, to advance these molecules in the antimalarial pipeline, MW and clog P must be optimized toward Lipinski’s thresholds.

RAP molecules’ high permeability despite their bulky and lipophilic nature may result from their facilitated transport via parasite-induced new permeation pathways. The latter allows the entry of diverse molecules within the infected RBC.58 For instance, the plasmodial surface anion channel linked to the clag gene family59 induced on the infected RBC membrane can carry large and lipophilic molecules.60 Once inside the infected RBCs, these molecules may further cross the PVM, which itself contains several non-selective channels to carry bulky molecules.61−63 The high lipophilicity may also allow molecules to partition within the lipid portion of the biological membrane, thus enabling passive diffusion.64 However, such “obese”65 molecules are likely to exhibit low solubility, extensive metabolism, and P-glycoproteins-mediated efflux, preventing their progression to the ASAM category, which explains the relatively lower MW and clog P averages of the latter class.

2.4. HBA and HBD

The HBA (sum of O and N atoms) and HBD (sum of NH and OH groups) are important parameters that determine the overall polarity and H-bonding capacity of a molecule. These two descriptors also affect the aqueous solubility,66,67 a prerequisite for oral absorption.

The HA and MA molecules show a significantly higher average HBA but lower HBD compared to the oral drugs (Table 2, Figure 1). The ASAM molecules display averages for HBA and HBD that do not differ statistically to those of oral drugs. However, the 90th percentile for both descriptors complies with Lipinski’s thresholds for all the categories of molecules. The larger antimalarial molecules (MW > 500 Da) consistently display lower HBA and HBD than the oral drugs, and the averages decrease with increasing in vitro potency. Overall, for small molecules (MW < 500), the HBA and HBD do not seem to have a noticeable influence on the antimalarial activity, but lower values for these descriptors are preferred for larger molecules.

2.5. TPSA and #RB

Veber and co-workers demonstrated TPSA and #RB descriptors to be the better predictors of oral bioavailability in comparison to the Ro5 with their undisclosed dataset (N = 1100).14 The molecules with a TPSA ≤ 140 Å2 and #RB ≤ 10 were found to have good oral bioavailability in rat models. In DataWarrior, TPSA is calculated using the original approach of Ertl et al., which was also adopted by Veber and co-workers.68 Our dataset of oral drugs (N = 1954) displays comparable results with the 90th percentiles of 144.7 Å2 and 10 for the TPSA and #RB, respectively (Table 1).

In RAP molecules, mean TPSA decreases (Figure 2) with an increase in potency (IN = 81.99 Å2, MA = 77.44 Å2, and HA = 74.68 Å2), with 90th percentiles also showing the same trend (Table 1). While the HA and MA molecule averages are significantly lower than that of the oral drugs (as per the t-test but not the Kruskal–Wallis test), statistical difference is not observed either between the HA/MA versus ASAM (75.9 Å2) or ASAM versus oral drugs (80.4 Å2). Interestingly, ASAM molecules display the lowest 90th percentile (mean = 118.3 Å2) among all categories. This trend of TPSA variation is exhibited by small and large molecules, albeit differences are more dramatic in the latter case (Supporting Information, Figure S6). Inclusively, these results suggest that a lower TPSA is advantageous for both in vitro and in vivo antimalarial activity, especially for the larger molecules. Presumably, higher polarity negatively affects permeability across the multiple membranes that an antimalarial molecule must cross.41

Figure 2.

Figure 2

Open in a new tab

Boxplots for the TPSA, #RB, #Ar, and #CarboAr properties for different sets of molecules. The mean values are given in bold above each boxplot and represented by the red line within the boxes. The yellow dots represent outliers.

The average for the #RB descriptor increases (Figure 2) significantly with increasing antiplasmodial potency, especially for the smaller molecules (Supporting Information, Figure S7), with the HA class displaying the highest mean of 6.6. In contrast, the ASAM molecules are relatively rigid, with fewer #RB (mean = 4.7) comparable to that of the oral drugs (mean = 5.09), suggesting that high flexibility or a greater RB count is not detrimental to the in vitro antiplasmodial activity. Nevertheless, lower #RB averages of ASAMs and oral drugs as compared to that of the HA molecules confirms the importance of lower flexibility for overall oral bioavailability and agrees with the observation of Veber et al.(14)

2.6. #Ar and Type of Rings

The number and nature of rings in a molecule can influence a molecule’s physicochemical properties, ultimately influencing its clinical success.10−12 The high attrition rate of molecules with a higher #Ar may be due to the low water solubility, high protein binding, and non-specific binding with other proteins leading to undesired effects. One crucial implication of the high content of aromatic carbons is the inhibition of human ether-à-go-go-related gene (hERG) channels that may lead to cardiotoxicity.10

On average, the HA molecule possesses more aromatic rings (mean = 2.73) than the MA (mean = 2.65) and IN (mean = 1.97) categories (Figure 2). However, the mean #Ar falls significantly in ASAM molecules (mean = 2.12) compared to that of HA but is still greater than that of oral drugs (mean = 1.66). These results are statistically significant, as indicated by both the t-test and the Kruskal–Wallis test. The 90th percentile for #Ar in ASAM and RAP molecules is 4, a unit higher than that of the oral drugs. This trend for #Ar also seems to be equally important for the large and small molecules (Supporting Information, Figure S8), suggesting aromaticity to be a key determinant for both in vitro and in vivo antimalarial activity.

The structurally related carboaromatic and heteroaromatic rings (e.g., phenyl vs pyridine) display distinct values of lipophilicity, polarity, conformation preference, and H-bond capability.69 Consequently, the carboaromatic ring count (#CarboAr) and heteroaromatic ring count (#HetAr) descriptors have varying influences on a molecule’s pharmacokinetics and pharmacodynamic profile.12

While the mean #CarboAr of the ASAM category (1.23) is not different from that of the oral drugs (1.13), the #HetAr mean of bioavailable ASAMs (0.89) is significantly higher as compared to the mean of oral drugs (0.52). Additionally, for active molecules (HA and MA), both #CarboAr and #HetAr are significantly higher than those of the IN class. A similar trend is observed in both the high and low MW class of compounds (Supporting Information, Figures S9 and S10). These observations suggest that while aromaticity is vital for both in vitro and in vivo antimalarial activity, there is a need to limit the #CarboAr to advance the antimalarial molecules toward clinical application. These observations agree with an earlier study, which suggests that a higher #CarboAr has a more substantial detrimental effect on a compound’s developability than a higher #HetAr.12

Given the importance of nitrogen-containing heterocycles in drug discovery,70 in general and for antimalarial drug discovery,71−73 in particular, we analyzed the aromatic nitrogen count (#ArN) in all compounds.

The #ArN is positively correlated with in vitro potency, as apparent from the mean #ArN for HA (1.11), MA (1.03), and IN (0.75) categories (Figure 3). The ASAM molecules display a mean value of 1.17 for the #ArN, which is significantly higher than that of the oral drugs (0.60) but not that of the HA class. However, the #ArN seems to be more critical for low MW compounds than for the bulkier molecules (Supporting Information, Figure S11). All antimalarial categories also display the 90th percentile of 3 for the #ArN, one unit higher than that of the oral drugs. The proportion of molecules possessing at least one N-heteroaromatic ring also increases with an increase in vitro antiplasmodial activity (HA > MA > IN) and attains the highest value of ∼62% for ASAM molecules (Supporting Information, Table S2), much higher than that of the oral drugs (∼30%). In contrast, no noticeable trend was observed for the aliphatic N-heterocycle, with all molecules showing similar percentages (Supporting Information, Table S2). Amongst all aromatic N-heterocycles, the quinoline ring appears most frequently in antimalarial molecules, followed by pyridine and pyrimidine (Supporting Information, Table S3). The latter two rings are also prevalent in oral drugs, an observation in line with the earlier reports.11,74

Figure 3.

Figure 3

Open in a new tab

Boxplots for the #HetAr, #ArN, #BaN, and Fsp3 properties for different sets of molecules. The mean values are given in bold above each boxplot and represented by the red line within the boxes. The yellow dots represent outliers.

Together, these observations suggest that higher content of the #ArN is favorable for the in vitro and in vivo antimalarial activity, and N-heterocyclics have a high probability of advancing in the antimalarial discovery pipeline.

A high content of the #CarboAr and #HetAr in antimalarials may be attributed to the historical success of quinoline-based antimalarials. The quinoline ring consists of two aromatic rings, one CarboAr and one HetAr. After several years, quinoline derivatives are still being pursued as antimalarials due to the synthetic tractability, cost-effectiveness, and ability of quinoline-based molecules to retain activity against chloroquine-resistant Plasmodium strains.75 Consequently, the antimalarial literature is replete with quinoline and related heterocyclic molecules.71 Additionally, several hybrid molecules76,77 and bisquinoline molecules78−81 are reported to possess potent antiplasmodial activity. The former consists of 4-aminoquinoline pharmacophore in conjunction with other heterocycles, while the latter contains two quinoline rings (total four #Ar) attached with a variable linker. Such bulky molecules are represented more in the RAP molecules, resulting in a higher #Ar and a high MW in these sets of molecules (vide supra).

Most of the quinoline and related N-heterocycle-based antimalarials target the hemozoin formation inside the parasite food vacuole. The latter is an essential process carried by the parasite to detoxify heme resulting from the hemoglobin degradation.82−85 The quinoline and related cyclic scaffolds foster π-stacking interactions with the porphyrin’s pyrrole rings, thereby inhibiting the nucleation and growth of the hemozoin crystals.84,86−89

In the context of target engagement, aromatic nitrogen can also act as an HBA and may dramatically improve potency, as observed for Plasmodiuml-lactate transporter (PfFNT) inhibitors.90 Another reason for the prevalence of N-heteroaromatics in the antimalarial design might be the emergence of Plasmodium kinases as drug targets.91−93 The majority of kinase inhibitors target the adenosine triphosphate (ATP)-binding pocket of the kinases, and several N-heterocycles mimic the adenosine ring of ATP. Another speculation may be that flatness might be favorable for transporting these molecules across the RBC or parasite membranes mediated by hitherto undiscovered transporters.

2.7. Fraction of sp3 Carbons (Fsp3)

The fraction of sp3 hybridized or tetrahedral carbons (Fsp3), calculated as the ratio of sp3 carbons to total carbons, is another critical physicochemical descriptor.13 The oral drugs are known to attain a higher Fsp3 in comparison to the clinical candidates. The higher Fsp3 correlates well with improved solubility and lower melting points, factors likely to improve oral bioavailability.13,94,95 This observation is confirmed with our compiled library of oral drugs and ASAM molecules, both of which possess higher 90th percentile values and averages for Fsp3 than that of RAP molecules (Tables 1 and 2). The Fsp3 and #Ar descriptors show an overall negative correlation (r = −0.632 for all molecules), yet the ASAM molecules possessing a higher 90th percentile for the #Ar also display a higher 90th percentile for the Fsp3, in comparison to that of the oral drugs. Among RAP molecules, MA and HA categories possess a significantly lower Fsp3 (Figure 3) than the ASAM/oral drugs categories; a trend also observed in both small and bulkier molecules (Supporting Information, Figure S12). Together with the #Ar (vide supra), the results of Fsp3 analysis reemphasizes that the flat structure of antimalarials may have a positive influence on the in vitro potency. Nonetheless, as suggested for other orally available drugs,13 a higher Fsp3 is advantageous to advance antimalarials in the drug discovery pipeline.

2.8. Basicity

In DataWarrior, the basic nitrogen count (#BaN) can be used to estimate the molecule’s basicity. The #BaN descriptor is based on a set of empirical rules rather than the computation of pKa values. In addition to the amine groups, nitrogen atoms in certain heterocycles, such as quinoline, pyridine, and imidazole, are counted as BaN depending on the other ring substituents. However, only a weak correlation (0.328) is observed between the #BaN and #ArN in the dataset, suggesting the two descriptors to be orthogonal.

On average, all categories of antimalarial molecules possess a higher #BaN (Figure 3); however, statistical significance is displayed only in the case of oral drugs versus ASAM and oral drugs versus HA categories. Among RAP molecules, HA molecules possess a significantly higher #BaN than MA and IN molecules. These trends are also found to apply to small and large molecules (Supporting Information, Figure S13). Also, #BaN 90th percentiles are consistently higher for all antimalarials than that of oral drugs. This observation confirms the importance of basic character for antimalarial molecules for in vitro as well as in vivo activity.

The presence of basic centers in the form of amines or ArN (as in aminoquinolines) is known to improve the in vitro antiplasmodial potency of molecules acting through diverse mechanisms.96−102 This is especially true for the antimalarials that target hemozoin formation within the parasite digestive vacuole (DV).84,103−105 The basic molecules ionize inside the acidic contents of the parasite’s DV106 leading to their entrapment and high intravacuolar concentration.104,106 The basic side chain and ring nitrogen of quinoline antimalarials are also proposed to make crucial interaction with the heme.84,98,102,107 In fact, the initial ionic interaction between the protonated nitrogen of the chloroquine side chain and heme carboxylate may be required to bring these together for further binding.107 The slightly lower cytoplasmic pH of the parasite105 might also be the driving force for the internalization of the basic molecules.105 In summary, the basicity of antimalarials may be necessary for both target binding and distribution within the parasite and its acidic DV. Also, the basic nitrogen centers are often added during lead optimization to improve solubility and metabolism, resulting in improved bioavailability.69,100 This justifies the highest proportion of the #BaN and #ArN in ASAM molecules compared to that of RAP molecules. One notable exception is the artemisinin class of antimalarials,108 which lack ArN or BaN in their structure and yet possess high in vitro and in vivo potency.

2.9. Antimalarial Property Space

One of this study’s objectives was to probe if an antimalarial space may be defined within the broad oral drug space. According to the widely cited Lipinski’s and Veber’s rules, the majority of the compounds in all categories conform to the drug-like space (Table 3). Expectedly, oral drugs and ASAM molecules show higher compliance for both rules. However, the combination of various fundamental properties used in these rules (MW, clog P, HBA, HBD, TPSA, and #RB) did not reveal a property space typical to that of antimalarial molecules. Given the importance of the #BaN and #ArN in antimalarials (vide supra), we hypothesized that these structural features might be used as a scaling factor to differentiate antimalarials from other molecules. Thus, a new descriptor, the sum of #BaN and #ArN (SBAN), was defined and used to scale various Lipinski’s and Veber’s properties by taking the latter’s ratio to the factor of 1 + SBAN.

Table 3. Number of Molecules Compliant to the Specified Guidelines.

guidelines oral drugs (N = 1954) ASAM (N = 66) HA (N = 10,557) MA (N = 6620) IN (N = 7365)
Lipinski’s Ro5 1786 (91%) 59 (89%) 8713 (83%) 5777 (87%) 6451 (88%)
Veber’s rule 1647 (84%) 61 (92%) 8693 (82%) 5609 (85%) 5933 (81%)
guideline-1 s-TPSA 5 to 65; s-RB ≤ 6; s-HBA ≤ 5; s-HBD ≤ 2 1338 (68%) 60 (91%) 8234 (78%) 4823 (73%) 4924 (67%)
guideline-2 s-TPSA 5 to 65; s-RB ≤ 6; s-HBA ≤ 5; s-HBD ≤ 2; MW ≥ 235 1114 (57%) 60 (91%) 8090 (77%) 4639 (70%) 4418 (60%)
Open in a new tab

Interestingly, the property space described by the resulting scaled descriptors (s-MW, s-clog P, s-HBA, s-HBD, s-TPSA, and s-RB) revealed the confinement of the ASAM molecule to a narrow region within the broad druglike space. Two typical examples of azithromycin and albitiazolium are represented in Figure 4. Azithromycin is overtly bulky (MW 749 Da) and highly polar (TPSA 180 Å2) due to several HBAs. However, the SBAN value of 2 in azithromycin results in lowered s-HBA and s-TPSA, pushing it to the antimalarial space with other ASAM molecules (Figure 4A vs4B). Similarly, highly flexible albitiazolium (#RB = 17) also relocates to the antimalarial space upon using scaled descriptors (Figure 4C vs4D).

Figure 4.

Figure 4

Open in a new tab

Plots showing ASAM molecules (red circles) in the property space. The bulky and polar azithromycin (A) converges to the antimalarial space defined by guideline-2 (B) when the corresponding scaled descriptors, s-TPSA and s-HBA, are used. Similarly, highly flexible and polar albitiazolium (C) also moves to the antimalarial space (D) following the application of the scaled descriptors, s-TPSA and s-RB.

These results encouraged us to propose guidelines or thresholds based on the scaled descriptors to characterize an antimalarial property space, particularly for the ASAM class. Based on the importance of individual properties and various plots between different scaled descriptors, we focused on s-TPSA, s-RB, s-HBA, and s-HBD. We found guideline-1 based on s-TPSA (5–65 Å2), s-RB (≤6), s-HBA (≤5), and s-HBD (≤2) to be more selective for antimalarial molecules (ASAM, HA, and MA) than for other oral drugs and IN molecules. 91% of ASAM, 78% of HA, and 73% of MA molecules comply with guideline-1, while only 68% of oral drugs and 67% of IN class are included in the same region (Table 3; Figure 5). Adding a threshold of MW ≥ 235 Da (guideline-2) further resulted in improved selectivity for the ASAM compared to that of the oral drugs and IN category. These results highlight the importance of the SBAN-scaled descriptors in defining the antimalarial drug space. Interestingly, all six ASAM molecules not complying with guideline-2 have high TPSA with no (or only one) SBAN count, resulting in their exclusion with the application of only the s-TPSA threshold (60–65 Å2). Five of these are natural products (or natural product analogue) displaying high polarity owing to the presence of carboxylic (artesunate and CDRI 9778), phosphonic (fosmidomycin), or phenolic/enolic (tetracycline and doxycycline) acidic moieties (Supporting Information, Figure S14). The dapsone, on the other hand, possesses two aromatic amines that are not considered “basic” by the DataWarrior program.

Figure 5.

Figure 5

Open in a new tab

Plot portraying the percentage of molecules in compliance with specific guidelines. While most of the molecules in each category pass Lipinski’s and Veber’s rules, the thresholds based on the scaled descriptors (guideline-1 and guideline-2) are more selective for antimalarials.

Several oral drugs and their close analogues have been shown to have potent activity against P. falciparum in several repurposing studies.109−113 As a result, out of the 1114 oral drugs picked by the guideline-2, 186 are already part of either RAP or ASAM libraries. Interestingly, of the remaining 928 oral drugs, 516 (∼56%) show the SkelSpheres53 descriptors-based similarity of 0.7 or more to at least one HA or MA class of molecules. These observations further raise confidence in the use of guideline-2 for defining the antimalarial property space and offer a testable hypothesis. For example, it would be interesting to systematically evaluate these drugs, some of which are recently approved, against the parasite.

3. Conclusions

In summary, this work provides insights into the average property space of RAP and ASAM molecules, vis-à-vis oral drugs using readily available open-source cheminformatics tools. The RAP molecules are significantly “obese”65,114 and belong to “flatland”.13 These research molecules display a positive correlation between their MW, #Ar, and clog P descriptors and in vitro potency (IN < MA < HA). However, for ASAM molecules, these properties converge close to Lipinski’s Ro5 thresholds while still maintaining higher averages than those of oral drugs. This suggests that overtly higher MW, lipophilicity, and a flat molecular shape may be helpful for the permeability across the RBC/parasite membranes, but lower values are preferred to obtain clinical or lead antimalarials. These observations also highlight the inability of the whole-cell antiplasmodial assays to filter out non-ideal molecules, which is an often-cited advantage of phenotypic assays.

Although a higher #Ar seems to contribute to in vitro potency, druglikeness is maintained only by increasing the #HetAr rather than the #CarboAr.11,12 Similarly, the higher average and 90th percentile of the Fsp3 descriptor in the ASAM and oral drugs than that of RAP molecules reconfirm its influence on clinical success.13 The HBA/HBD descriptors appear to be significant only for the bulky (MW > 500 Da) molecules with lower values favoring antimalarial activity. The lower TPSA and #RB also improve the likelihood of obtaining antimalarials with oral bioavailability, an observation in line with that of Veber et al.(14)

We also recognized that the #ArN and #BaN, the lesser-studied descriptors, are essential elements present in structurally diverse antimalarials, including historically successful aminoquinolines. Both the #ArN and #BaN might be assisting in target engagement and/or distribution within the parasite’s acidic DV. We found a positive correlation between antimalarial activity and the #ArN and #BaN, while oral drugs were revealed to have lower values for these two descriptors. The high #ArN is primarily due to the expansive explorations of quinoline, pyridine, and pyrimidine rings, while amine groups contribute to the high #BaN. Judging by its high frequency in all categories of antimalarial molecules, the quinoline ring is still relevant to the antimalarial drug design.

We also propose the use of properties scaled by the SBAN count to define a region in the property space where the probability of finding druglike antimalarials (ASAMs) seems to be high. Two guidelines specifying the thresholds of scaled descriptors are suggested, albeit natural product-like molecules with acidic functionalities do not appear to conform to this space. In a physiological context, it seems that the SBAN count and other favorable properties assist antimalarials in crossing multiple membrane barriers to reach the intracellular targets of the parasite.41 There is a clear indication that the Plasmodium’s highly evolved transportome41,115−117 interacts with several marketed or advanced-stage antimalarials either as an antimalarial target or as part of a resistance mechanism.115,117 Thus, a family of transporters in the parasitized RBC or PVM may be responsible for transporting antimalarials within the property space identified in this study. This is in line with Kell’s hypothesis, implying that the carrier-mediated cellular uptake of molecules is more common than the diffusion across the phospholipid bilayer.118−121 This is supported by the fact that the majority of drugs display high similarity to natural human metabolites.122 It must be noted, however, that the characterization of Plasmodium’s transporters and their substrate specificities is challenging owing to its complex biology.

Overall, these results may have important implications in future explorations of antimalarial molecules. In general, relatively bulky, lipophilic, and flat molecules consisting of nitrogen scaffolds decorated with amine groups seem to be preferable candidates for antimalarial drug design. The libraries conforming to the proposed antimalarial property space may provide higher hit rates in experimental or virtual HTS studies.

Specific properties, such as MW, may change over time, apparently due to the exploration of newer targets and technologies.4,21,25 This change may also be reflected upon antimalarial leads and drugs.123 Hence, property-based studies such as this should be reassessed as more data emerge in the future.

4. Methodology

4.1. Data Curation

The ChEMBL-26 database124 was searched within the DataWarrior program (version 5.2.1) to obtain the RAP set of molecules.53 The molecules screened in a whole-cell phenotypic assay against “P. falciparum” were imported within DataWarrior where macromolecules (MW > 900 Da) and organometallic compounds were removed. The compounds annotated with definite IC50 or EC50 values (with the qualifier “=”, “∼”; and not “<”, “>”) in nM or μM units were retained, and salt forms were neutralized. The canonical codes were generated using the DataWarrior program, and duplicate molecules were merged. In the case of multiple IC50/EC50 values for the same compounds, the geometric average was calculated. The molecules with a large difference (>10 folds) in its multiple IC50/EC50 values were discarded. The average IC50/EC50 values were used for the classification of RAP molecules into “highly-active” (HA) (IC50/EC50 ≤ 1000 nM), “moderately active” (MA) (IC50/EC50 1001–9999 nM), and “inactive” (IN) (IC50/EC50 ≥ 10,000 nM) classes. The molecules found to be inactive in the HTS study conducted by GSK using the TCAMS43 were also added to the IN set to expand the latter. The data for the TCAMS screening was downloaded from ChEMBL-neglected tropical disease webpage.125 Only small (MW < 900 Da) and non-redundant molecules found to be inactive against both DD2 and 3D7 strains (for which no percentage inhibition data is reported) were retained (total 4351). The ASAM set of molecules was gathered from various literature reports (see Supporting Information, Table S1) and consisted of 33 marketed antimalarials, 19 clinical candidates, and 14 lead molecules. The structure of oral drugs was downloaded from DrugCentral50 database and updated with the new chemical entities approved by the US FDA till July 2020.

4.2. Physicochemical Property Analysis

The physicochemical properties were calculated either using the DataWarrior program or RDKit126 nodes implemented in KNIME platform 4.1.2.127 GraphPad Prism was used for the computation of various statistical parameters (such as mean, median, 90th/10th percentiles, and confidence intervals) and hypothesis testing using the non-parametric one-way analysis of variance (Kruskal–Wallis method). The p-values were calculated at a 95% confidence level. DataWarrior was used to compute boxplots and associated p-values using the t-test and for calculating Spearman correlation coefficients. The frequency of rings for all categories of molecules was carried out by counting the “plain ring systems” using DataWarrior.

4.3. Data and Software Availability

The structures and related data of RAP molecules can be freely obtained from ChEMBL database (https://www.ebi.ac.uk/chembl/). The ASAM set of molecules was manually curated from the recent literature (Supporting Information, Table S1). The set of oral drugs was obtained from DrugCentral database, downloadable from https://drugcentral.org/download. The KNIME platform (v 4.1.2) is freely available at https://www.knime.com/downloads. The RDKit nodes for KNIME are available to download within the KNIME platform. The DataWarrior program (v 5.2.1) is an open-source cheminformatic software downloadable at http://www.openmolecules.org/datawarrior/download.html.

Acknowledgments

The authors acknowledge the Department of Science and Technology, Science & Engineering Research Board (DST-SERB), New Delhi, for financial assistance through the Core Research Grant (CRG/2018/001527). We also thank Navya Bhandaru for assisting in the compilation of the library of oral drugs.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.1c00104.

  • Details of the ASAM category of molecules, distribution of low and high MW compounds, distribution of aromatic and aliphatic N-heterocycles, six most frequently occurring rings in each category, correlation between Actelion clog P versus the experimental log P and Actelion clog P versus StarDrop clog P, distribution of the mean of clog P, HBA, HBD, TPSA, #RB, #Ar, #CarboAr, #HetAr, #ArN, Fsp3, and #BaN among low and high MW compounds within the IN, MA, HA, ASAM, and oral drugs categories, and statistical data (PDF)

The authors declare no competing financial interest.

Supplementary Material

ao1c00104_si_001.pdf (4.1MB, pdf)

References

  1. Wouters O. J.; McKee M.; Luyten J. Estimated Research and Development Investment Needed to Bring a New Medicine to Market, 2009-2018. JAMA, J. Am. Med. Assoc. 2020, 323, 844–853. 10.1001/jama.2020.1166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. DiMasi J. A.; Grabowski H. G.; Hansen R. W. Innovation in the Pharmaceutical Industry: New Estimates of R&D Costs. J. Health Econ. 2016, 47, 20–33. 10.1016/j.jhealeco.2016.01.012. [DOI] [PubMed] [Google Scholar]
  3. Dowden H.; Munro J. Trends in Clinical Success Rates and Therapeutic Focus. Nat. Rev. Drug Discovery 2019, 18, 495–496. 10.1038/d41573-019-00074-z. [DOI] [PubMed] [Google Scholar]
  4. Shultz M. D. Two Decades under the Influence of the Rule of Five and the Changing Properties of Approved Oral Drugs. J. Med. Chem. 2019, 62, 1701–1714. 10.1021/acs.jmedchem.8b00686. [DOI] [PubMed] [Google Scholar]
  5. Lipinski C. A.; Lombardo F.; Dominy B. W.; Feeney P. J. Experimental and Computational Approaches to Estimate Solubility and Permeability in Drug Discovery and Development Settings. Adv. Drug Delivery Rev. 1997, 23, 3–25. 10.1016/s0169-409x(96)00423-1. [DOI] [PubMed] [Google Scholar]
  6. Petit J.; Meurice N.; Kaiser C.; Maggiora G. Softening the Rule of Five—Where to Draw the Line?. Bioorg. Med. Chem. 2012, 20, 5343–5351. 10.1016/j.bmc.2011.11.064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Abad-Zapatero C. A Sorcerer’s Apprentice and The Rule of Five: From Rule-of-Thumb to Commandment and Beyond. Drug Discovery Today 2007, 12, 995–997. 10.1016/j.drudis.2007.10.022. [DOI] [PubMed] [Google Scholar]
  8. Tinworth C. P.; Young R. J. Facts, Patterns, and Principles in Drug Discovery: Appraising the Rule of 5 with Measured Physicochemical Data. J. Med. Chem. 2020, 63, 10091–10108. 10.1021/acs.jmedchem.9b01596. [DOI] [PubMed] [Google Scholar]
  9. Leeson P. D. Molecular Inflation, Attrition and the Rule of Five. Adv. Drug Delivery Rev. 2016, 101, 22–33. 10.1016/j.addr.2016.01.018. [DOI] [PubMed] [Google Scholar]
  10. Ritchie T. J.; Macdonald S. J. F. The Impact of Aromatic Ring Count on Compound Developability—Are Too Many Aromatic Rings a Liability in Drug Design?. Drug Discovery Today 2009, 14, 1011–1020. 10.1016/j.drudis.2009.07.014. [DOI] [PubMed] [Google Scholar]
  11. Ritchie T. J.; Macdonald S. J. F.; Peace S.; Pickett S. D.; Luscombe C. N. The Developability of Heteroaromatic and Heteroaliphatic Rings—Do Some Have a Better Pedigree as Potential Drug Molecules than Others?. MedChemComm 2012, 3, 1062–1069. 10.1039/c2md20111a. [DOI] [Google Scholar]
  12. Ritchie T. J.; MacDonald S. J. F.; Young R. J.; Pickett S. D. The Impact of Aromatic Ring Count on Compound Developability: Further Insights by Examining Carbo- and Hetero-Aromatic and -Aliphatic Ring Types. Drug Discovery Today 2011, 16, 164–171. 10.1016/j.drudis.2010.11.014. [DOI] [PubMed] [Google Scholar]
  13. Lovering F.; Bikker J.; Humblet C. Escape from Flatland: Increasing Saturation as an Approach to Improving Clinical Success. J. Med. Chem. 2009, 52, 6752–6756. 10.1021/jm901241e. [DOI] [PubMed] [Google Scholar]
  14. Veber D. F.; Johnson S. R.; Cheng H.-Y.; Smith B. R.; Ward K. W.; Kopple K. D. Molecular Properties That Influence the Oral Bioavailability of Drug Candidates. J. Med. Chem. 2002, 45, 2615–2623. 10.1021/jm020017n. [DOI] [PubMed] [Google Scholar]
  15. Whitty A.; Zhong M.; Viarengo L.; Beglov D.; Hall D. R.; Vajda S. Quantifying the Chameleonic Properties of Macrocycles and Other High-Molecular-Weight Drugs. Drug Discovery Today 2016, 21, 712–717. 10.1016/j.drudis.2016.02.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Young R. J.; Green D. V. S.; Luscombe C. N.; Hill A. P. Getting Physical in Drug Discovery II: The Impact of Chromatographic Hydrophobicity Measurements and Aromaticity. Drug Discovery Today 2011, 16, 822–830. 10.1016/j.drudis.2011.06.001. [DOI] [PubMed] [Google Scholar]
  17. Degoey D. A.; Chen H.-J.; Cox P. B.; Wendt M. D. Beyond the Rule of 5: Lessons Learned from AbbVie’s Drugs and Compound Collection. J. Med. Chem. 2018, 61, 2636–2651. 10.1021/acs.jmedchem.7b00717. [DOI] [PubMed] [Google Scholar]
  18. Fullam E.; Young R. J. Physicochemical Properties and Mycobacterium Tuberculosis Transporters: Keys to Efficacious Antitubercular Drugs?. RSC Med. Chem. 2021, 12, 45–56. 10.1039/d0md00265h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Bickerton G. R.; Paolini G. V.; Besnard J.; Muresan S.; Hopkins A. L. Quantifying the Chemical Beauty of Drugs. Nat. Chem. 2012, 4, 90–98. 10.1038/nchem.1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Young R. J.; Leeson P. D. Mapping the Efficiency and Physicochemical Trajectories of Successful Optimizations. J. Med. Chem. 2018, 61, 6421–6467. 10.1021/acs.jmedchem.8b00180. [DOI] [PubMed] [Google Scholar]
  21. Leeson P. D.; Springthorpe B. The Influence of Drug-like Concepts on Decision-Making in Medicinal Chemistry. Nat. Rev. Drug Discovery 2007, 6, 881–890. 10.1038/nrd2445. [DOI] [PubMed] [Google Scholar]
  22. Hopkins A. L.; Groom C. R.; Alex A. Ligand Efficiency: A Useful Metric for Lead Selection. Drug Discovery Today 2004, 9, 430–431. 10.1016/s1359-6446(04)03069-7. [DOI] [PubMed] [Google Scholar]
  23. Vieth M.; Sutherland J. J. Dependence of Molecular Properties on Proteomic Family for Marketed Oral Drugs. J. Med. Chem. 2006, 49, 3451–3453. 10.1021/jm0603825. [DOI] [PubMed] [Google Scholar]
  24. Adrian G.; Marcel V.; Robert B.; Richard T. A Comparison of Physicochemical Property Profiles of Marketed Oral Drugs and Orally Bioavailable Anti-Cancer Protein Kinase Inhibitors in Clinical Development. Curr. Top. Med. Chem. 2007, 7, 1408–1422. 10.2174/156802607781696819. [DOI] [PubMed] [Google Scholar]
  25. Leeson P. D.; Davis A. M. Time-Related Differences in the Physical Property Profiles of Oral Drugs. J. Med. Chem. 2004, 47, 6338–6348. 10.1021/jm049717d. [DOI] [PubMed] [Google Scholar]
  26. Gualtieri M.; Baneres-Roquet F.; Villain-Guillot P.; Pugniere M.; Leonetti J.-P. The Antibiotics in the Chemical Space. Curr. Med. Chem. 2009, 16, 390–393. 10.2174/092986709787002628. [DOI] [PubMed] [Google Scholar]
  27. Macielag M. J.Chemical Properties of Antimicrobials and Their Uniqueness. In Antibiotic Discovery and Development; Dougherty T. J., Pucci M. J., Eds.; Springer: New York, 2012; pp 793–820. [Google Scholar]
  28. Wager T. T.; Chandrasekaran R. Y.; Hou X.; Troutman M. D.; Verhoest P. R.; Villalobos A.; Will Y. Defining Desirable Central Nervous System Drug Space through the Alignment of Molecular Properties, in Vitro ADME, and Safety Attributes. ACS Chem. Neurosci. 2010, 1, 420–434. 10.1021/cn100007x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Doan K. M. M.; Humphreys J. E.; Webster L. O.; Wring S. A.; Shampine L. J.; Serabjit-Singh C. J.; Adkison K. K.; Polli J. W. Passive Permeability and P-Glycoprotein-Mediated Efflux Differentiate Central Nervous System (CNS) and Non-CNS Marketed Drugs. J. Pharmacol. Exp. Ther. 2002, 303, 1029–1037. 10.1124/jpet.102.039255. [DOI] [PubMed] [Google Scholar]
  30. World Health Organization . WHO Malaria Report 2019. https://www.who.int/publications-detail/world-malaria-report-2019 (accessed Oct 23, 2020).
  31. Woodrow C. J.; White N. J. The Clinical Impact of Artemisinin Resistance in Southeast Asia and the Potential for Future Spread. FEMS Microbiol. Rev. 2017, 41, 34–48. 10.1093/femsre/fuw037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Müller O.; Lu G. Y.; Von Seidlein L. Geographic Expansion of Artemisinin Resistance. J. Travel Med. 2019, 26, taz030. 10.1093/jtm/taz030. [DOI] [PubMed] [Google Scholar]
  33. Imwong M.; Suwannasin K.; Kunasol C.; Sutawong K.; Mayxay M.; Rekol H.; Smithuis F. M.; Hlaing T. M.; Tun K. M.; van der Pluijm R. W.; Tripura R.; Miotto O.; Menard D.; Dhorda M.; Day N. P. J.; White N. J.; Dondorp A. M. The Spread of Artemisinin-Resistant Plasmodium Falciparum in the Greater Mekong Subregion: A Molecular Epidemiology Observational Study. Lancet Infect. Dis. 2017, 17, 491–497. 10.1016/s1473-3099(17)30048-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Hassett M. R.; Roepe P. D. Origin and Spread of Evolving Artemisinin-Resistant Plasmodium Falciparum Malarial Parasites in Southeastc Asia. Am. J. Trop. Med. Hyg. 2019, 101, 1204–1211. 10.4269/ajtmh.19-0379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Ashley E. A.; Dhorda M.; Fairhurst R. M.; Amaratunga C.; Lim P.; Suon S.; Sreng S.; Anderson J. M.; Mao S.; Sam B.; Sopha C.; Chuor C. M.; Nguon C.; Sovannaroth S.; Pukrittayakamee S.; Jittamala P.; Chotivanich K.; Chutasmit K.; Suchatsoonthorn C.; Runcharoen R.; Hien T. T.; Thuy-Nhien N. T.; Thanh N. V.; Phu N. H.; Htut Y.; Han K.-T.; Aye K. H.; Mokuolu O. A.; Olaosebikan R. R.; Folaranmi O. O.; Mayxay M.; Khanthavong M.; Hongvanthong B.; Newton P. N.; Onyamboko M. A.; Fanello C. I.; Tshefu A. K.; Mishra N.; Valecha N.; Phyo A. P.; Nosten F.; Yi P.; Tripura R.; Borrmann S.; Bashraheil M.; Peshu J.; Faiz M. A.; Ghose A.; Hossain M. A.; Samad R.; Rahman M. R.; Hasan M. M.; Islam A.; Miotto O.; Amato R.; MacInnis B.; Stalker J.; Kwiatkowski D. P.; Bozdech Z.; Jeeyapant A.; Cheah P. Y.; Sakulthaew T.; Chalk J.; Intharabut B.; Silamut K.; Lee S. J.; Vihokhern B.; Kunasol C.; Imwong M.; Tarning J.; Taylor W. J.; Yeung S.; Woodrow C. J.; Flegg J. A.; Das D.; Smith J.; Venkatesan M.; Plowe C. V.; Stepniewska K.; Guerin P. J.; Dondorp A. M.; Day N. P.; White N. J. Spread of Artemisinin Resistance in Plasmodium Falciparum Malaria. N. Engl. J. Med. 2014, 371, 411–423. 10.1056/nejmoa1314981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Conrad M. D.; Rosenthal P. J. Antimalarial Drug Resistance in Africa: The Calm before the Storm?. Lancet Infect. Dis. 2019, 19, e338–e351. 10.1016/s1473-3099(19)30261-0. [DOI] [PubMed] [Google Scholar]
  37. Ashley E. A.; Phyo A. P. Drugs in Development for Malaria. Drugs 2018, 78, 861–879. 10.1007/s40265-018-0911-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tse E. G.; Korsik M.; Todd M. H. The Past, Present and Future of Anti-Malarial Medicines. Malar. J. 2019, 18, 93. 10.1186/s12936-019-2724-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Okombo J.; Chibale K. Recent Updates in the Discovery and Development of Novel Antimalarial Drug Candidates. MedChemComm 2018, 9, 437–453. 10.1039/c7md00637c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Burrows J. N.; Duparc S.; Gutteridge W. E.; Hooft Van Huijsduijnen R.; Kaszubska W.; Macintyre F.; Mazzuri S.; Möhrle J. J.; Wells T. N. C. New Developments in Anti-Malarial Target Candidate and Product Profiles. Malar. J. 2017, 16, 26. 10.1186/s12936-016-1675-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Basore K.; Cheng Y.; Kushwaha A. K.; Nguyen S. T.; Desai S. A. How Do Antimalarial Drugs Reach Their Intracellular Targets?. Front. Pharmacol. 2015, 6, 91. 10.3389/fphar.2015.00091. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Goldberg D. E.; Zimmerberg J. Hardly Vacuous: The Parasitophorous Vacuolar Membrane of Malaria Parasites. Trends Parasitol. 2020, 36, 138–146. 10.1016/j.pt.2019.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Gamo F.-J.; Sanz L. M.; Vidal J.; De Cozar C.; Alvarez E.; Lavandera J.-L.; Vanderwall D. E.; Green D. V. S.; Kumar V.; Hasan S.; Brown J. R.; Peishoff C. E.; Cardon L. R.; Garcia-Bustos J. F. Thousands of Chemical Starting Points for Antimalarial Lead Identification. Nature 2010, 465, 305–310. 10.1038/nature09107. [DOI] [PubMed] [Google Scholar]
  44. Wenlock M. C.; Austin R. P.; Barton P.; Davis A. M.; Leeson P. D. A Comparison of Physiochemical Property Profiles of Development and Marketed Oral Drugs. J. Med. Chem. 2003, 46, 1250–1256. 10.1021/jm021053p. [DOI] [PubMed] [Google Scholar]
  45. Tyrchan C.; Blomberg N.; Engkvist O.; Kogej T.; Muresan S. Physicochemical Property Profiles of Marketed Drugs, Clinical Candidates and Bioactive Compounds. Bioorg. Med. Chem. Lett. 2009, 19, 6943–6947. 10.1016/j.bmcl.2009.10.068. [DOI] [PubMed] [Google Scholar]
  46. Bento A. P.; Gaulton A.; Hersey A.; Bellis L. J.; Chambers J.; Davies M.; Krüger F. A.; Light Y.; Mak L.; McGlinchey S.; Nowotka M.; Papadatos G.; Santos R.; Overington J. P. The ChEMBL Bioactivity Database: An Update. Nucleic Acids Res. 2014, 42, D1083–D1090. 10.1093/nar/gkt1031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Gaulton A.; Bellis L. J.; Bento A. P.; Chambers J.; Davies M.; Hersey A.; Light Y.; McGlinchey S.; Michalovich D.; Al-Lazikani B.; Overington J. P. ChEMBL: A Large-Scale Bioactivity Database for Drug Discovery. Nucleic Acids Res. 2012, 40, D1100–D1107. 10.1093/nar/gkr777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ekins S.; Williams A. J. When Pharmaceutical Companies Publish Large Datasets: An Abundance of Riches or Fool’s Gold?. Drug Discovery Today 2010, 15, 812–815. 10.1016/j.drudis.2010.08.010. [DOI] [PubMed] [Google Scholar]
  49. Dimova D.; Stumpfe D.; Bajorath J. Systematic Assessment of Coordinated Activity Cliffs Formed by Kinase Inhibitors and Detailed Characterization of Activity Cliff Clusters and Associated SAR Information. Eur. J. Med. Chem. 2015, 90, 414–427. 10.1016/j.ejmech.2014.11.058. [DOI] [PubMed] [Google Scholar]
  50. Ursu O.; Holmes J.; Bologa C. G.; Yang J. J.; Mathias S. L.; Stathias V.; Nguyen D.-T.; Schürer S.; Oprea T. DrugCentral 2018: An Update. Nucleic Acids Res. 2019, 47, D963–D970. 10.1093/nar/gky963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Pye C. R.; Hewitt W. M.; Schwochert J.; Haddad T. D.; Townsend C. E.; Etienne L.; Lao Y.; Limberakis C.; Furukawa A.; Mathiowetz A. M.; Price D. A.; Liras S.; Lokey R. S. Nonclassical Size Dependence of Permeation Defines Bounds for Passive Adsorption of Large Drug Molecules. J. Med. Chem. 2017, 60, 1665–1672. 10.1021/acs.jmedchem.6b01483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Doak B. C.; Over B.; Giordanetto F.; Kihlberg J. Oral Druggable Space beyond the Rule of 5: Insights from Drugs and Clinical Candidates. Chem. Biol. 2014, 21, 1115–1142. 10.1016/j.chembiol.2014.08.013. [DOI] [PubMed] [Google Scholar]
  53. Sander T.; Freyss J.; von Korff M.; Rufener C. DataWarrior: An Open-Source Program for Chemistry Aware Data Visualization and Analysis. J. Chem. Inf. Model. 2015, 55, 460–473. 10.1021/ci500588j. [DOI] [PubMed] [Google Scholar]
  54. Mannhold R.; Poda G. I.; Ostermann C.; Tetko I. V. Calculation of Molecular Lipophilicity: State-of-the-Art and Comparison of Log P Methods on More than 96,000 Compounds. J. Pharm. Sci. 2009, 98, 861–893. 10.1002/jps.21494. [DOI] [PubMed] [Google Scholar]
  55. Van de Waterbeemd H.; Smith D. A.; Beaumont K.; Walker D. K. Property-Based Design: Optimization of Drug Absorption and Pharmacokinetics. J. Med. Chem. 2001, 44, 1313–1333. 10.1021/jm000407e. [DOI] [PubMed] [Google Scholar]
  56. Ertl P.; Schuffenhauer A. Estimation of Synthetic Accessibility Score of Drug-like Molecules Based on Molecular Complexity and Fragment Contributions. J. Cheminf. 2009, 1, 8. 10.1186/1758-2946-1-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Egan W. J.; Merz K. M.; Baldwin J. J. Prediction of Drug Absorption Using Multivariate Statistics. J. Med. Chem. 2000, 43, 3867–3877. 10.1021/jm000292e. [DOI] [PubMed] [Google Scholar]
  58. Desai S. A. Why Do Malaria Parasites Increase Host Erythrocyte Permeability?. Trends Parasitol. 2014, 30, 151–159. 10.1016/j.pt.2014.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Nguitragool W.; Bokhari A. A. B.; Pillai A. D.; Rayavara K.; Sharma P.; Turpin B.; Aravind L.; Desai S. A. Malaria Parasite Clag3 Genes Determine Channel-Mediated Nutrient Uptake by Infected Red Blood Cells. Cell 2011, 145, 665–677. 10.1016/j.cell.2011.05.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Cohn J. V.; Alkhalil A.; Wagner M. A.; Rajapandi T.; Desai S. A. Extracellular Lysines on the Plasmodial Surface Anion Channel Involved in Na+ Exclusion. Mol. Biochem. Parasitol. 2003, 132, 27–34. 10.1016/j.molbiopara.2003.08.001. [DOI] [PubMed] [Google Scholar]
  61. Desai S. A.; Rosenberg R. L. Pore Size of the Malaria Parasite’s Nutrient Channel. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2045–2049. 10.1073/pnas.94.5.2045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Nyalwidhe J.; Baumeister S.; Hibbs A. R.; Tawill S.; Papakrivos J.; Völker U.; Lingelbach K. A Nonpermeant Biotin Derivative Gains Access to the Parasitophorous Vacuole in Plasmodium Falciparum-Infected Erythrocytes Permeabilized with Streptolysin O. J. Biol. Chem. 2002, 277, 40005–40011. 10.1074/jbc.m207077200. [DOI] [PubMed] [Google Scholar]
  63. Desai S. A.; Krogstad D. J.; McCleskey E. W. A Nutrient-Permeable Channel on the Intraerythrocytic Malaria Parasite. Nature 1993, 362, 643–646. 10.1038/362643a0. [DOI] [PubMed] [Google Scholar]
  64. Refsgaard H. H. F.; Jensen B. F.; Brockhoff P. B.; Padkjær S. B.; Guldbrandt M.; Christensen M. S. In Silico Prediction of Membrane Permeability from Calculated Molecular Parameters. J. Med. Chem. 2005, 48, 805–811. 10.1021/jm049661n. [DOI] [PubMed] [Google Scholar]
  65. Hann M. M. Molecular Obesity, Potency and Other Addictions in Drug Discovery. MedChemComm 2011, 2, 349–355. 10.1039/c1md00017a. [DOI] [Google Scholar]
  66. Schultes S.; De Graaf C.; Berger H.; Mayer M.; Steffen A.; Haaksma E. E. J.; De Esch I. J. P.; Leurs R.; Krämer O. A Medicinal Chemistry Perspective on Melting Point: Matched Molecular Pair Analysis of the Effects of Simple Descriptors on the Melting Point of Drug-like Compounds. MedChemComm 2012, 3, 584–591. 10.1039/c2md00313a. [DOI] [Google Scholar]
  67. Withnall M.; Chen H.; Tetko I. V. Matched Molecular Pair Analysis on Large Melting Point Datasets: A Big Data Perspective. ChemMedChem 2018, 13, 599–606. 10.1002/cmdc.201700303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Ertl P.; Rohde B.; Selzer P. Fast Calculation of Molecular Polar Surface Area as a Sum of Fragment-Based Contributions and Its Application to the Prediction of Drug Transport Properties. J. Med. Chem. 2000, 43, 3714–3717. 10.1021/jm000942e. [DOI] [PubMed] [Google Scholar]
  69. Pennington L. D.; Moustakas D. T. The Necessary Nitrogen Atom: A Versatile High-Impact Design Element for Multiparameter Optimization. J. Med. Chem. 2017, 60, 3552–3579. 10.1021/acs.jmedchem.6b01807. [DOI] [PubMed] [Google Scholar]
  70. Vitaku E.; Smith D. T.; Njardarson J. T. Analysis of the Structural Diversity, Substitution Patterns, and Frequency of Nitrogen Heterocycles among U.S. FDA Approved Pharmaceuticals. J. Med. Chem. 2014, 57, 10257–10274. 10.1021/jm501100b. [DOI] [PubMed] [Google Scholar]
  71. Kalaria P. N.; Karad S. C.; Raval D. K. A Review on Diverse Heterocyclic Compounds as the Privileged Scaffolds in Antimalarial Drug Discovery. Eur. J. Med. Chem. 2018, 158, 917–936. 10.1016/j.ejmech.2018.08.040. [DOI] [PubMed] [Google Scholar]
  72. Kaur K.; Jain M.; Reddy R. P.; Jain R. Quinolines and Structurally Related Heterocycles as Antimalarials. Eur. J. Med. Chem. 2010, 45, 3245–3264. 10.1016/j.ejmech.2010.04.011. [DOI] [PubMed] [Google Scholar]
  73. Chugh A.; Kumar A.; Verma A.; Kumar S.; Kumar P. A Review of Antimalarial Activity of Two or Three Nitrogen Atoms Containing Heterocyclic Compounds. Med. Chem. Res. 2020, 29, 1723–1750. 10.1007/s00044-020-02604-6. [DOI] [Google Scholar]
  74. Ertl P.; Jelfs S.; Mühlbacher J.; Schuffenhauer A.; Selzer P. Quest for the Rings. In Silico Exploration of Ring Universe to Identify Novel Bioactive Heteroaromatic Scaffolds. J. Med. Chem. 2006, 49, 4568–4573. 10.1021/jm060217p. [DOI] [PubMed] [Google Scholar]
  75. Parhizgar A. R.; Tahghighi A. Introducing New Antimalarial Analogues of Chloroquine and Amodiaquine: A Narrative Review. Iran. J. Med. Sci. 2017, 42, 115–128. [PMC free article] [PubMed] [Google Scholar]
  76. Feng L. S.; Xu Z.; Chang L.; Li C.; Yan X. F.; Gao C.; Ding C.; Zhao F.; Shi F.; Wu X. Hybrid Molecules with Potential in Vitro Antiplasmodial and in Vivo Antimalarial Activity against Drug-Resistant Plasmodium Falciparum. Med. Res. Rev. 2020, 40, 931–971. 10.1002/med.21643. [DOI] [PubMed] [Google Scholar]
  77. Nqoro X.; Tobeka N.; Aderibigbe B. Quinoline-Based Hybrid Compounds with Antimalarial Activity. Molecules 2017, 22, 2268. 10.3390/molecules22122268. [DOI] [PMC free article] [PubMed] [Google Scholar]
  78. Vennerstrom J. L.; Ellis W. Y.; Ager A. L.; Andersen S. L.; Gerena L.; Milhous W. K. Bisquinolines. 1. N,N-Bis(7-Chloroquinolin-4-Yl)Alkanediamines with Potential against Chloroquine-Resistant Malaria. J. Med. Chem. 1992, 35, 2129–2134. 10.1021/jm00089a025. [DOI] [PubMed] [Google Scholar]
  79. Liebman K. M.; Burgess S. J.; Gunsaru B.; Kelly J. X.; Li Y.; Morrill W.; Liebman M. C.; Peyton D. H. Unsymmetrical Bisquinolines with High Potency against P. Falciparum Malaria. Molecules 2020, 25, 2251. 10.3390/molecules25092251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Kondaparla S.; Agarwal P.; Srivastava K.; Puri S. K.; Katti S. B. Design, Synthesis and in Vitro Antiplasmodial Activity of Some Bisquinolines against Chloroquine-Resistant Strain. Chem. Biol. Drug Des. 2017, 89, 901–906. 10.1111/cbdd.12914. [DOI] [PubMed] [Google Scholar]
  81. Vennerstrom J. L.; Ager A. L.; Dorn A.; Andersen S. L.; Gerena L.; Ridley R. G.; Milhous W. K. Bisquinolines. 2. Antimalarial N,N-Bis(7-Chloroquinolin-4- Yl)Heteroalkanediamines. J. Med. Chem. 1998, 41, 4360–4364. 10.1021/jm9803828. [DOI] [PubMed] [Google Scholar]
  82. Egan T. J. Haemozoin Formation. Mol. Biochem. Parasitol. 2008, 157, 127–136. 10.1016/j.molbiopara.2007.11.005. [DOI] [PubMed] [Google Scholar]
  83. Fong K. Y.; Wright D. W. Hemozoin and Antimalarial Drug Discovery. Future Med. Chem. 2013, 5, 1437–1450. 10.4155/fmc.13.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  84. Weissbuch I.; Leiserowitz L. Interplay between Malaria, Crystalline Hemozoin Formation, and Antimalarial Drug Action and Design. Chem. Rev. 2008, 108, 4899–4914. 10.1021/cr078274t. [DOI] [PubMed] [Google Scholar]
  85. Combrinck J. M.; Mabotha T. E.; Ncokazi K. K.; Ambele M. A.; Taylor D.; Smith P. J.; Hoppe H. C.; Egan T. J. Insights into the Role of Heme in the Mechanism of Action of Antimalarials. ACS Chem. Biol. 2013, 8, 133–137. 10.1021/cb300454t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  86. Olafson K. N.; Nguyen T. Q.; Rimer J. D.; Vekilov P. G. Antimalarials Inhibit Hematin Crystallization by Unique Drug–Surface Site Interactions. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 7531–7536. 10.1073/pnas.1700125114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  87. Sullivan D. J. Quinolines Block Every Step of Malaria Heme Crystal Growth. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 7483–7485. 10.1073/pnas.1708153114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  88. Buller R.; Peterson M. L.; Almarsson Ö.; Leiserowitz L. Quinoline Binding Site on Malaria Pigment Crystal: A Rational Pathway for Antimalaria Drug Design. Cryst. Growth Des. 2002, 2, 553–562. 10.1021/cg025550i. [DOI] [Google Scholar]
  89. Solomonov I.; Osipova M.; Feldman Y.; Baehtz C.; Kjaer K.; Robinson I. K.; Webster G. T.; McNaughton D.; Wood B. R.; Weissbuch I.; Leiserowitz L. Crystal Nucleation, Growth, and Morphology of the Synthetic Malaria Pigment β-Hematin and the Effect Thereon by Quinoline Additives: The Malaria Pigment as a Target of Various Antimalarial Drugs. J. Am. Chem. Soc. 2007, 129, 2615–2627. 10.1021/ja0674183. [DOI] [PubMed] [Google Scholar]
  90. Walloch P.; Henke B.; Häuer S.; Bergmann B.; Spielmann T.; Beitz E. Introduction of Scaffold Nitrogen Atoms Renders Inhibitors of the Malarial L-Lactate Transporter, PfFNT, Effective against the Gly107Ser Resistance Mutation. J. Med. Chem. 2020, 63, 9731–9741. 10.1021/acs.jmedchem.0c00852. [DOI] [PubMed] [Google Scholar]
  91. Kappes B.; Doerig C. D.; Graeser R. An Overview of Plasmodium Protein Kinases. Parasitol. Today 1999, 15, 449–454. 10.1016/s0169-4758(99)01527-6. [DOI] [PubMed] [Google Scholar]
  92. Doerig C.; Billker O.; Haystead T.; Sharma P.; Tobin A. B.; Waters N. C. Protein Kinases of Malaria Parasites: An Update. Trends Parasitol. 2008, 24, 570–577. 10.1016/j.pt.2008.08.007. [DOI] [PubMed] [Google Scholar]
  93. Cabrera D. G.; Horatscheck A.; Wilson C. R.; Basarab G.; Eyermann C. J.; Chibale K. Plasmodial Kinase Inhibitors: License to Cure?. J. Med. Chem. 2018, 61, 8061–8077. 10.1021/acs.jmedchem.8b00329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  94. Ward S. E.; Beswick P. What Does the Aromatic Ring Number Mean for Drug Design?. Expert Opin. Drug Discovery 2014, 9, 995–1003. 10.1517/17460441.2014.932346. [DOI] [PubMed] [Google Scholar]
  95. Thomas V. H.; Bhattachar S.; Hitchingham L.; Zocharski P.; Naath M.; Surendran N.; Stoner C. L.; El-Kattan A. The Road Map to Oral Bioavailability: An Industrial Perspective. Expert Opin. Drug Metab. Toxicol. 2006, 2, 591–608. 10.1517/17425255.2.4.591. [DOI] [PubMed] [Google Scholar]
  96. Large J. M.; Birchall K.; Bouloc N. S.; Merritt A. T.; Smiljanic-Hurley E.; Tsagris D. J.; Wheldon M. C.; Ansell K. H.; Coombs P. J.; Kettleborough C. A.; Whalley D.; Stewart L. B.; Bowyer P. W.; Baker D. A.; Osborne S. A. Potent Inhibitors of Malarial P. Falciparum Protein Kinase G: Improving the Cell Activity of a Series of Imidazopyridines. Bioorg. Med. Chem. Lett. 2019, 29, 509–514. 10.1016/j.bmcl.2018.11.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  97. Tsagris D. J.; Birchall K.; Bouloc N.; Large J. M.; Merritt A.; Smiljanic-Hurley E.; Wheldon M.; Ansell K. H.; Kettleborough C.; Whalley D.; Stewart L. B.; Bowyer P. W.; Baker D. A.; Osborne S. A. Trisubstituted Thiazoles as Potent and Selective Inhibitors of Plasmodium Falciparum Protein Kinase G (PfPKG). Bioorg. Med. Chem. Lett. 2018, 28, 3168–3173. 10.1016/j.bmcl.2018.08.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  98. Natarajan J. K.; Alumasa J. N.; Yearick K.; Ekoue-Kovi K. A.; Casabianca L. B.; De Dios A. C.; Wolf C.; Roepe P. D. 4-N-, 4-S-, and 4-O-Chloroquine Analogues: Influence of Side Chain Length and Quinolyl Nitrogen PKa on Activity vs Chloroquine Resistant Malaria. J. Med. Chem. 2008, 51, 3466–3479. 10.1021/jm701478a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  99. Hameed P S.; Solapure S.; Patil V.; Henrich P. P.; Magistrado P. A.; Bharath S.; Murugan K.; Viswanath P.; Puttur J.; Srivastava A.; Bellale E.; Panduga V.; Shanbag G.; Awasthy D.; Landge S.; Morayya S.; Koushik K.; Saralaya R.; Raichurkar A.; Rautela N.; Roy Choudhury N.; Ambady A.; Nandishaiah R.; Reddy J.; Prabhakar K. R.; Menasinakai S.; Rudrapatna S.; Chatterji M.; Jiménez-Díaz M. B.; Martínez M. S.; Sanz L. M.; Coburn-Flynn O.; Fidock D. A.; Lukens A. K.; Wirth D. F.; Bandodkar B.; Mukherjee K.; McLaughlin R. E.; Waterson D.; Rosenbrier-Ribeiro L.; Hickling K.; Balasubramanian V.; Warner P.; Hosagrahara V.; Dudley A.; Iyer P. S.; Narayanan S.; Kavanagh S.; Sambandamurthy V. K. Triaminopyrimidine Is a Fast-Killing and Long-Acting Antimalarial Clinical Candidate. Nat. Commun. 2015, 6, 6715. 10.1038/ncomms7715. [DOI] [PMC free article] [PubMed] [Google Scholar]
  100. Masch A.; Nasereddin A.; Alder A.; Bird M. J.; Schweda S. I.; Preu L.; Doerig C.; Dzikowski R.; Gilberger T. W.; Kunick C. Structure-Activity Relationships in a Series of Antiplasmodial Thieno[2,3-b]Pyridines. Malar. J. 2019, 18, 89. 10.1186/s12936-019-2725-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  101. Lavrado J.; Cabal G. G.; Prudêncio M.; Mota M. M.; Gut J.; Rosenthal P. J.; Díaz C.; Guedes R. C.; Dos Santos D. J. V. A.; Bichenkova E.; Douglas K. T.; Moreira R.; Paulo A. Incorporation of Basic Side Chains into Cryptolepine Scaffold: Structure-Antimalarial Activity Relationships and Mechanistic Studies. J. Med. Chem. 2011, 54, 734–750. 10.1021/jm101383f. [DOI] [PubMed] [Google Scholar]
  102. Egan T. J.; Hunter R.; Kaschula C. H.; Marques H. M.; Misplon A.; Walden J. Structure-Function Relationships in Aminoquinolines: Effect of Amino and Chloro Groups on Quinoline-Hematin Complex Formation, Inhibition of β-Hematin Formation, and Antiplasmodial Activity. J. Med. Chem. 2000, 43, 283–291. 10.1021/jm990437l. [DOI] [PubMed] [Google Scholar]
  103. Tam D. N. H.; Tawfik G. M.; El-Qushayri A. E.; Mehyar G. M.; Istanbuly S.; Karimzadeh S.; Tu V. L.; Tiwari R.; Van Dat T.; Nguyen P. T. V.; Hirayama K.; Huy N. T. Correlation between Anti-Malarial and Anti-Haemozoin Activities of Anti-Malarial Compounds. Malar. J. 2020, 19, 298. 10.1186/s12936-020-03370-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  104. Kaschula C. H.; Egan T. J.; Hunter R.; Basilico N.; Parapini S.; Taramelli D.; Pasini E.; Monti D. Structure–Activity Relationships in 4-Aminoquinoline Antiplasmodials. The Role of the Group at the 7-Position. J. Med. Chem. 2002, 45, 3531–3539. 10.1021/jm020858u. [DOI] [PubMed] [Google Scholar]
  105. Yayon A.; Cabantchik Z. I.; Ginsburg H. Identification of the Acidic Compartment of Plasmodium Falciparum-Infected Human Erythrocytes as the Target of the Antimalarial Drug Chloroquine. EMBO J. 1984, 3, 2695–2700. 10.1002/j.1460-2075.1984.tb02195.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  106. Kuhn Y.; Rohrbach P.; Lanzer M. Quantitative PH Measurements in Plasmodium Falciparum-Infected Erythrocytes Using PHluorin. Cell. Microbiol. 2007, 9, 1004–1013. 10.1111/j.1462-5822.2006.00847.x. [DOI] [PubMed] [Google Scholar]
  107. Bachhawat K.; Thomas C. J.; Surolia N.; Surolia A. Interaction of Chloroquine and Its Analogues with Heme: An Isothermal Titration Calorimetric Study. Biochem. Biophys. Res. Commun. 2000, 276, 1075–1079. 10.1006/bbrc.2000.3592. [DOI] [PubMed] [Google Scholar]
  108. O’Neill P. M.; Posner G. H. A Medicinal Chemistry Perspective on Artemisinin and Related Endoperoxides. J. Med. Chem. 2004, 47, 2945–2964. 10.1021/jm030571c. [DOI] [PubMed] [Google Scholar]
  109. Teixeira C.; Vale N.; Pérez B.; Gomes A.; Gomes J. R. B.; Gomes P. “Recycling” Classical Drugs for Malaria. Chem. Rev. 2014, 114, 11164–11220. 10.1021/cr500123g. [DOI] [PubMed] [Google Scholar]
  110. Chong C. R.; Chen X.; Shi L.; Liu J. O.; Sullivan D. J. A Clinical Drug Library Screen Identifies Astemizole as an Antimalarial Agent. Nat. Chem. Biol. 2006, 2, 415–416. 10.1038/nchembio806. [DOI] [PubMed] [Google Scholar]
  111. Da Cruz F. P.; Martin C.; Buchholz K.; Lafuente-Monasterio M. J.; Rodrigues T.; Sönnichsen B.; Moreira R.; Gamo F.-J.; Marti M.; Mota M. M.; Hannus M.; Prudêncio M. Drug Screen Targeted at Plasmodium Liver Stages Identifies a Potent Multistage Antimalarial Drug. J. Infect. Dis. 2012, 205, 1278–1286. 10.1093/infdis/jis184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  112. Pazhayam N. M.; Chhibber-Goel J.; Sharma A. New Leads for Drug Repurposing against Malaria. Drug Discovery Today 2019, 24, 263–271. 10.1016/j.drudis.2018.08.006. [DOI] [PubMed] [Google Scholar]
  113. Kaiser M.; Mäser P.; Tadoori L. P.; Ioset J. R.; Brun R.; Sullivan D. J. Antiprotozoal Activity Profiling of Approved Drugs: A Starting Point toward Drug Repositioning. PLoS One 2015, 10, e0135556 10.1371/journal.pone.0135556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  114. Gleeson M. P.; Hersey A.; Montanari D.; Overington J. Probing the Links between in Vitro Potency, ADMET and Physicochemical Parameters. Nat. Rev. Drug Discovery 2011, 10, 197–208. 10.1038/nrd3367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  115. Martin R. E. The Transportome of the Malaria Parasite. Biol. Rev. 2020, 95, 305–332. 10.1111/brv.12565. [DOI] [PubMed] [Google Scholar]
  116. Cowell A. N.; Istvan E. S.; Lukens A. K.; Gomez-Lorenzo M. G.; Vanaerschot M.; Sakata-Kato T.; Flannery E. L.; Magistrado P.; Owen E.; Abraham M.; LaMonte G.; Painter H. J.; Williams R. M.; Franco V.; Linares M.; Arriaga I.; Bopp S.; Corey V. C.; Gnädig N. F.; Coburn-Flynn O.; Reimer C.; Gupta P.; Murithi J. M.; Moura P. A.; Fuchs O.; Sasaki E.; Kim S. W.; Teng C. H.; Wang L. T.; Akidil A.; Adjalley S.; Willis P. A.; Siegel D.; Tanaseichuk O.; Zhong Y.; Zhou Y.; Llinás M.; Ottilie S.; Gamo F.-J.; Lee M. C. S.; Goldberg D. E.; Fidock D. A.; Wirth D. F.; Winzeler E. A. Mapping the Malaria Parasite Druggable Genome by Using in Vitro Evolution and Chemogenomics. Science 2018, 359, 191–199. 10.1126/science.aan4472. [DOI] [PMC free article] [PubMed] [Google Scholar]
  117. Meier A.; Erler H.; Beitz E. Targeting Channels and Transporters in Protozoan Parasite Infections. Front. Chem. 2018, 6, 88. 10.3389/fchem.2018.00088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  118. Dobson P. D.; Kell D. B. Carrier-Mediated Cellular Uptake of Pharmaceutical Drugs: An Exception or the Rule?. Nat. Rev. Drug Discovery 2008, 7, 205–220. 10.1038/nrd2438. [DOI] [PubMed] [Google Scholar]
  119. Kell D. B. What Would Be the Observable Consequences If Phospholipid Bilayer Diffusion of Drugs into Cells Is Negligible?. Trends Pharmacol. Sci. 2015, 36, 15–21. 10.1016/j.tips.2014.10.005. [DOI] [PubMed] [Google Scholar]
  120. Kell D. B.; Oliver S. G. How Drugs Get into Cells: Tested and Testable Predictions to Help Discriminate between Transporter-Mediated Uptake and Lipoidal Bilayer Diffusion. Front. Pharmacol. 2014, 5, 231. 10.3389/fphar.2014.00231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  121. Kell D. B.; Dobson P. D.; Oliver S. G. Pharmaceutical Drug Transport: The Issues and the Implications That It Is Essentially Carrier-Mediated Only. Drug Discovery Today 2011, 16, 704–714. 10.1016/j.drudis.2011.05.010. [DOI] [PubMed] [Google Scholar]
  122. O’Hagan S.; Swainston N.; Handl J.; Kell D. B. A ‘Rule of 0.5’ for the Metabolite-Likeness of Approved Pharmaceutical Drugs. Metabolomics 2015, 11, 323–339. 10.1007/s11306-014-0733-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  123. Charman S. A.; Andreu A.; Barker H.; Blundell S.; Campbell A.; Campbell M.; Chen G.; Chiu F. C. K.; Crighton E.; Katneni K.; Morizzi J.; Patil R.; Pham T.; Ryan E.; Saunders J.; Shackleford D. M.; White K. L.; Almond L.; Dickins M.; Smith D. A.; Moehrle J. J.; Burrows J. N.; Abla N. An in Vitro Toolbox to Accelerate Anti-Malarial Drug Discovery and Development. Malar. J. 2020, 19, 1. 10.1186/s12936-019-3075-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  124. CHEMBL26, release date March 2020; 10.6019/CHEMBL.database.26.
  125. GSK-TCAMS Dataset. https://chembl.gitbook.io/chembl-ntd/downloads/deposited-set-1-gsk-tcams-dataset-20th-may-2010 (accessed Oct 3, 2020).
  126. Landrum G.; et al. RDKit: Open-Source Cheminformatics. 2006, http://www.rdkit.org (accessed on September 1, 2020).
  127. Berthold M. R.; Cebron N.; Dill F.; Gabriel T. R.; Kötter T.; Meinl T.; Ohl P.; Sieb C.; Thiel K.; Wiswedel B.. KNIME: The Konstanz Information Miner. Studies in Classification, Data Analysis, and Knowledge Organization (GfKL 2007); Springer, 2007. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao1c00104_si_001.pdf (4.1MB, pdf)

Data Availability Statement

The structures and related data of RAP molecules can be freely obtained from ChEMBL database (https://www.ebi.ac.uk/chembl/). The ASAM set of molecules was manually curated from the recent literature (Supporting Information, Table S1). The set of oral drugs was obtained from DrugCentral database, downloadable from https://drugcentral.org/download. The KNIME platform (v 4.1.2) is freely available at https://www.knime.com/downloads. The RDKit nodes for KNIME are available to download within the KNIME platform. The DataWarrior program (v 5.2.1) is an open-source cheminformatic software downloadable at http://www.openmolecules.org/datawarrior/download.html.

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES