Discovery of Indeno[1,2-c]quinoline Derivatives as Potent Dual Antituberculosis and Anti-Inflammatory Agents

A series of indeno[1,2-c]quinoline derivatives were designed, synthesized and evaluated for their anti-tuberculosis (anti-TB) and anti-inflammatory activities. The minimum inhibitory concentration (MIC) of the newly synthesized compound was tested against Mycobacterium tuberculosis H37RV. Among the tested compounds, (E)-N′-[6-(4-hydroxypiperidin-1-yl)-11H-indeno[1,2-c]quinolin-11-ylidene]isonicotino-hydrazide (12), exhibited significant activities against the growth of M. tuberculosis (MIC values of 0.96 μg/mL) with a potency approximately equal to that of isoniazid (INH), an anti-TB drug. Important structure features were analyzed by quantitative structure–activity relationship (QSAR) analysis to give better insights into the structure determinants for predicting the anti-TB activity. The anti-inflammatory activity was induced by superoxide anion generation and neutrophil elastase (NE) release using the formyl-l-methionyl-l-leucyl-l-phenylalanine (fMLF)-activated human neutrophils method. Results indicated that compound 12 demonstrated a potent dual inhibitory effect on NE release and superoxide anion generation with IC50 values of 1.76 and 1.72 μM, respectively. Our results indicated that compound 12 is a potential lead compound for the discovery of dual anti-TB and anti-inflammatory drug candidates. In addition, 6-[3-(hydroxymethyl)piperidin-1-yl]-9-methoxy-11H-indeno[1,2-c]quinolin-11-one (4g) showed a potent dual inhibitory effect on NE release and superoxide anion generation with IC50 values of 0.46 and 0.68 μM, respectively, and is a potential lead compound for the discovery of anti-inflammatory drug candidates.


Introduction
Tuberculosis (TB) is an infectious disease caused by Mycobacterium tuberculosis which primarily affects the lungs, causing an intense local inflammatory response that is critical to the pathogenesis of tuberculosis [1,2]. Therefore, TB is a chronic inflammatory condition in which regulatory and pro-inflammatory processes occur that contribute to severe lung pathology, leading to lung tissue necrosis and the promotion of mycobacterial dissemination and transmission [3]. Anti-inflammatory drugs, especially corticosteroids, are currently used for adjunctive therapy in most severe life-threatening forms of tuberculosis, such as meningitis and pericarditis [4]. The importance of developing new drugs with dual anti-inflammatory and antimycobacterial activities is amplified by the emergence of multi-drug resistant (MDR) strains and the global human immunodeficiency virus (HIV) pandemic [5][6][7]. The anti-leprosy drug clofazimine [8] exhibited dual anti-inflammatory and antimycobacterial activities. However, its potency is not sufficient, motivating the design and search for novel agents.

Results and Discussion
The anti-TB activities of indeno[1,2-c]quinoline derivatives are summarized in Table 1.

Results and Discussion
The anti-TB activities of indeno[1,2-c]quinoline derivatives are summarized in Table 1.
The same structure-activity relationships were observed for C9-methoxy derivatives in which compounds 4a-4c were inactive, while their respective C6-hydroxypiperidine derivative 4d-4g exhibited a potent dual inhibitory effect on NE release and superoxide anion generation. Among them, 9-methoxy-6-(3-piperidinemethano-1-yl)-11H-indeno[1,2-c]quinolin-11-one (4g) was the most potent dual inhibitor of NE release and superoxide anion generation, with IC 50 values of 0.46 and 0.68 µM, respectively. Compound 5 was only weakly active, while compound 6 was inactive indicated the reduction of the keto group is unfavorable. The keto group of 3d can be converted into its oxime derivatives, compounds 7 and 8, and retain comparable anti-inflammatory activity. However, the semicarbazone derivatives, compounds 9 and 10, selectively inhibit superoxide anion generation with IC 50 values of 3.64 and 2.41 µM, respectively. The benzohydrazide derivative 11 was capable of inhibiting superoxide anion generation with IC 50 value of 1.67 µM but was inactive in the inhibition of NE release. (E)-N -(6-(4-hydroxypiperidin-1-yl)-11H-indeno[1,2-c]quinolin-11-ylidene)isonicotino hydrazide (12) was found to be a potent dual inhibitor of NE release and superoxide anion generation, with IC 50 values of 1.76 and 1.72 µM, respectively, approximately equal potent to the positive LY294002.
To provide a better understanding of the relationship between the structure descriptors and anti-TB activities, a quantitative structure-activity relationship (QSAR) model was constructed to analyze important features and evaluate prediction performance. First, a total of 13 chemicals with determined MIC values as shown in Table 1 were randomly divided into a training dataset for model construction and a test dataset for independent test of the constructed model. The training and test datasets consist of 9 (12, 11, 3g, 3c, 3e, 3a, 4d, 4a and 5) and 4 (3d, 3f, 3b and 4e) chemicals, respectively. Subsequently, each chemical was transformed into a vector of 1648 descriptors representing its structure features. The sequential feature selection algorithms developed by our group [30,31] were applied to simultaneously select a minimum number of relevant descriptors maximizing the R 2 performance and develop multiple regression models for QSAR analysis. Please note that only the three most important features were considered due to the small number of chemicals in the dataset. In the QSAR analysis, pMIC values were utilized in that the MIC values in µg/mL were converted toµM/mL and then converted to −log MIC (pMIC). The model is expected to be useful for chemicals within the pMIC range from 1.24 to 2.67. The final QSAR model with high performance with R 2 = 0.96, R 2 adj = 0.94, root-mean-squared error (RMSE) = 0.085, and Q 2 = 0.86 could well explain the variation of pMIC values representing the anti-TB activity. The constructed QSAR model and three important descriptors are shown in Table 3. SCH-6 (simple chain, order 6) epresents the Chi chain descriptor of order 6 [32] that directly correlates with pMIC values with a high correlation coefficient of 0.83 for the 13 chemicals. CrippenLogP is the Crippen's LogP descriptor [33]. The hmax (representing the maximum hydrogen E-State value in a molecule) belongs to the descriptors of atom type electrotopological state [34][35][36]. Table 2. Anti-inflammatory activities of indeno [1,2-c]quinoline derivatives in formyl-Lmethionyl-L-leucyl-L-phenylalanine (fMLF)-activated human neutrophils (IC 50 in µM) a . 1.67 ± 0.87 >10 12

Compounds
1.72 ± 0.23 1.76 ± 0.49 LY294002 b 1.36 ± 0.33 2.21 ± 0.45 a Concentration necessary for 50% inhibition (IC 50 ); Results are presented as mean ± SEM (n = 3). b LY294002 (a phosphatidylinositol-3-kinase inhibitor) was used as a positive control for superoxide anion generation and elastase release. While CrippenLogP and hmax show only a mild negative correlation to pMIC values with correlation coefficient values of −0.13 and −0.01, respectively, for the 13 chemicals, the combination of all three descriptors largely enhances the prediction performance of the QSAR model with a 15% improvement on R 2 . The observed and predicted pMIC values on the training dataset are shown in Table 4. To further evaluate the prediction ability of the QSAR model, the model was applied to predict the 4 chemicals in the test dataset. As shown in Table 5, the predicted and observed values for chemicals with exact MIC values are well correlated with a high correlation coefficient of 0.98. Altogether, the good performance of the QSAR models shows the importance of the three descriptors for determining the anti-TB activity. The model could be further tested and improved when more data are available.

General
Melting points were determined on a Electrothermal IA9100 melting point apparatus and are uncorrected. Nuclear magnetic resonance ( 1 H and 13 C) spectra were recorded on a Varian Gemini 200 spectrometer or Varian-Unity-400 spectrometer. Chemical shifts were expressed in parts per million (δ) with tetramethylsilane (TMS) as an internal standard and coupling constant (J) in hertz (Hz). Thin-layer chromatography was performed on silica gel 60 F-254 plates purchased from E. Merck and Co. IR spectra were measured using a FTIR Perkin-Elmer system-2000 spectrometer (Perkin-Elmer, Waltham, MA, USA). The elemental analyses were performed in the Instrument Center of National Science Council at National Cheng-Kung University using Heraeus CHN-O Rapid EA (Heraeus Co., Harau, Germany), and all values are within ± 0.4% of the theoretical compositions.

Anti-Mycobacterium Activity
Primary screening is conducted against M. tuberculosis H 37 Rv (ATCC 27294) in BACTEC 12B medium (BD Microbiology Systems, Franklin Lakes, NJ, USA) using a broth microdilution assay to determine the actual minimum inhibitory concentration (MIC) using the Microplate Alamar Blue Assay (MABA) [37]. The MIC is defined as the lowest concentration effecting a reduction in fluorescence of 90% relative to controls. Vero cell (normal cell) viability is assessed on the basis of cellular conversion of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), into a formazan product using the Promega Cell Titer 96 Non-radioactive Cell Proliferation Assay (Promega, Madison, WI, USA).

Superoxide Generation and Elastase Release
Superoxide generation and elastase release were carried out according to the procedures described previously [38]. Neutrophils (6 × 10 5 /mL) were equilibrated at 37 • C for 2 min and incubated with compounds for 5 min. Neutrophils were then activated by fMLF (100 nM) in the pretreatment of cytochalasin B (1 µg/mL for superoxide generation and 0.5 µg/mL for elastase release) for 10 min. Superoxide anion production was assayed by monitoring the superoxide dismutase-inhibitable reduction of ferricytochrome C. Elastase release was performed using MeO-Suc-Ala-Ala-Pro-Valp-nitroanilide as the elastase substrate.

Quantitative Structure-Activity Relationship (QSAR)
Structure features of 1648 1D and 2D descriptors and PubChem fingerprints for chemicals were firstly generated by using the PaDEL-Descriptor [39] software (http://padel.nus.edu.sg/software/ padeldescriptor). The calculations are mainly based on the Chemistry Development Kit [40] with some additional descriptors and fingerprints including atom-type electrotopological state descriptors, McGowan volume, molecular linear free energy relation descriptors, ring counts, count of chemical substructures, binary fingerprints and count of chemical substructures. It has been extensively utilized in many QSAR studies such as non-genotoxic hepatotoxicity prediction [41] and inhibitor design [42]. Sequential feature selection algorithms developed by our group were subsequently applied to simultaneously identify important features and build multiple [30,31] regression models. For each run, the descriptor with highest R 2 performance was iteratively appended into the multiple regression model. To avoid overfitting problems, only descriptors with a maximum correlation coefficient to selected descriptors ranging from −0.5 to 0.5 were considered. Due to the small number of chemicals, only the best three descriptors were considered. The descriptor set with the highest R 2 performance is utilized to construct the final multiple regression model for QSAR analysis. The measurements of R 2 , R adj 2 , root-mean-squared error (RMSE), and leave-one-out cross-validated Q 2 were applied to evaluate the performance of the QSAR model.

Conclusions
We have synthesized certain indeno [1,2-c]quinoline derivatives for anti-TB and anti-inflammatory evaluations. Among them, (E)-N -[6-(4-hydroxypiperidin-1-yl)-11H-indeno[1,2-c]quinolin-11-ylidene]isonicotinohydrazide (12), exhibited significant activities against the growth of M. tuberculosis (MIC values of 0.96 µg/mL) with a potency approximately equal to that of INH. Compound 12 has also demonstrated a potent dual inhibitory effect on NE release and superoxide anion generation with IC 50 values of 1.76 and 1.72 µM, respectively. These results indicated that compound 12 is a potential lead compound for the discovery of dual anti-TB and anti-inflammatory drug candidates. In addition, 6-[3-(hydroxymethyl)piperidin-1-yl]-9-methoxy-11H-indeno[1,2-c]quinolin-11-one (4g) showed a potent dual inhibitory effect on NE release and superoxide anion generation with IC 50 value of 0.46 and 0.68 µM respectively, and is a potential lead compound for the discovery of anti-inflammatory drug candidates. Further studies on the structural optimization and the molecular pharmacological mechanism of compounds 12 and 4g are ongoing.