pubs.acs.org/jmc Article Expansion of the 4‑(Diethylamino)benzaldehyde Scaffold to Explore the Impact on Aldehyde Dehydrogenase Activity and Antiproliferative Activity in Prostate Cancer Ali I. M. Ibrahim, Elisabet Batlle, Smarakan Sneha, Rafael Jiménez, Raquel Pequerul, Xavier Parés, Till Rüngeler, Vibhu Jha, Tiziano Tuccinardi, Maria Sadiq, Fiona Frame, Norman J. Maitland, Jaume Farrés,* and Klaus Pors* Cite This: J. Med. Chem. 2022, 65, 3833−3848 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Aldehyde dehydrogenases (ALDHs) are overex- pressed in various tumor types including prostate cancer and considered a potential target for therapeutic intervention. 4- (Diethylamino)benzaldehyde (DEAB) has been extensively reported as a pan-inhibitor of ALDH isoforms, and here, we report on the synthesis, ALDH isoform selectivity, and cellular potencies in prostate cancer cells of 40 DEAB analogues; three analogues (14, 15, and 16) showed potent inhibitory activity against ALDH1A3, and two analogues (18 and 19) showed potent inhibitory activity against ALDH3A1. Significantly, 16 analogues displayed increased cytotoxicity (IC50 = 10−200 μM) compared with DEAB (>200 μM) against three different prostate cancer cell lines. Analogues 14 and 18 were more potent than DEAB against patient-derived primary prostate tumor epithelial cells, as single agents or in combination treatment with docetaxel. In conclusion, our study supports the use of DEAB as an ALDH inhibitor but also reveals closely related analogues with increased selectivity and potency. ■ INTRODUCTION The aldehyde dehydrogenase (ALDH) superfamily of 19 provided no information on isoform selectivity.14 Many studies indicate that DEAB is a reversible and broad inhibitor of different human isoforms is a group of NAD(P)+-dependent several ALDH isoforms, which is a confounding factor when enzymes that catalyze several cellular processes, including using the Aldefluor assay.15,16 The complexity of the ALDH detoxification of endogenous and exogenous aldehydes and interaction is evident by recent work, which has shown that biosynthesis of retinoic acid (RA), which is a modulator of DEAB is also a substrate for ALDH1A1 and ALDH3A1,15 an stem cell (SC) differentiation.1−3 Recent interest in ALDHs irreversible inhibitor for ALDH1A2 and ALDH2, and neither a emerges from their roles linked to cancer cell proliferation, substrate nor inhibitor for ALDH1L1 and ALDH4A1, while it differentiation, and survival4,5 and potential as markers of has been shown to be metabolized by ALDH1A3, ALDH1B1, tumor-initiating cells or cancer SCs (CSCs).6 and ALDH5A1.17 Additionally, DEAB has been shown to Identification of CSCs is frequently carried out using the covalently bind to ALDH7A1, in which the DEAB-enzyme Aldefluor assay, which includes the addition of 4- complex was successfully cocrystallized with the cofactor (diethylamino)benzaldehyde (DEAB), a small-molecule inhib- NAD+.18 It has been proposed that the structural features of itor that is used as a control to identify subpopulations of cells the amino acid residues at the ALDH active sites are the main with high ALDH expression (ALDHhigh) and with SC-like factors determining whether DEAB is a substrate or an properties.4 DEAB is a pan-ALDH inhibitor, which has been inhibitor.17 Moreover, the ALDH−DEAB binary structure has found to delay differentiation of CSCs,7,8 also showing been hypothesized to be stabilized by resonance arrangement, potential in combination treatments with other drugs.9−13 Early work showed that 4-(dipropylamino)benzaldehyde was Received: August 2, 2021 more potent than DEAB as a reversible inhibitor against mouse Published: February 25, 2022 and human ALDH1, with variable inhibitory effects according to the selected substrate. 14 In the same study, 4- (dimethylamino)benzaldehyde and 4-(dibutylamino)- benzaldehyde showed low binding affinity to ALDH1 but © 2022 American Chemical Society https://doi.org/10.1021/acs.jmedchem.1c01367 3833 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article which is initiated and supported by the amine electron donors Scheme 1. Nomenclature of DEAB Analogues and General at the para position to the carbonyl group.17 Synthesis Scheme; Reaction Starts by a Nucleophilic ALDH1A1 and 1A3 are highly expressed in SC-like Substitution at the Fluorine (A) or Hydroxyl Group (B) of subpopulations and several cancer types.8,19 We have recently the Substituted Benzaldehydea reported ALDH1A1 and 1A3 isoform expression in different prostatic tissue-derived cell lines (normal, benign, and malignant) and patient-derived primary prostate tumor epithelial cells and shown potential in inhibiting these for therapeutic intervention.20 Given the utility of DEAB in both identifying CSC populations21 and potential in treating such ALDHhigh-expressing populations,8 we decided to generate a new library of analogues that could be used to explore the DEAB scaffold to unravel key features. Here, we report on the synthesis, biological evaluation against ALDH1A1, 1A3, and 3A1, and antiproliferative activity as single agents against a panel of prostate cancer (PCa) cell lines. We also assessed selected agents for their potential in combination treatment with docetaxel. ■ RESULTS AND DISCUSSION Chemistry. DEAB analogues were synthesized using either aliphatic or aromatic nucleophilic substitution one-step reactions. The nucleophilic aromatic substitution of fluorine by the desired secondary amine is shown in Scheme 1A. The presence of the aldehyde linked para to the fluorine group is likely facilitating the nucleophilic attack through electronic arrangement within the aromatic ring.22−24 The aldehyde is chemically reactive toward amines; however, as imines are less likely to be produced by secondary amines, the products were generally afforded in good yields (33−100%). Compounds 16, 33, and 35 were synthesized by the aliphatic nucleophilic substitution reaction between the phenolic hydroxyl group and isopropyl iodide, as shown in Scheme 1B.25 All DEAB analogues were synthesized, purified, and characterized by 1H NMR, 13C NMR, and HRMS as described in the Experimental Section. Inhibitory Effect of DEAB Analogues against ALDH Isoforms. Previous studies have established the importance of a Standard conditions: (a) secondary amine, DMF, K2CO3, 25−100 ALDH1 isoforms: 1A126 and 1A327 in CSCs and 3A128 in °C and (b) 2-iodopropane, DMF, K2CO3, 25−90 °C, 6 h. drug resistance. Accordingly, we decided to explore a new library of DEAB analogues for interaction with these isoforms by generating structure−activity relationships (SAR) useful in Accordingly, we captured full data analysis for compounds informing DEAB properties critical to biological activity. 14, 26, 29, and DEAB, and IC50 values were calculated for the The inhibition screening and IC50 values were assayed using best fits, demonstrating the latter two compounds to be nearly the conditions described in the Experimental Section. The IC50 equipotent with DEAB (IC50 = 0.48 ± 0.06 μM), while value for compound 7 was measured at a saturating substrate compound 14 was ∼15-fold less effective in inhibiting concentration with ALDH3A1 because the same remaining ALDH1A1 activity (Table 1). In general, those compounds activity was observed in inhibition screening both near the Km with a better inhibition profile had dipropyl, diethyl, or a value and at the saturating substrate concentration. Exper- pyrrolidine as an R group, the latter being the best, especially imental values are shown as the mean ± SE. IC50 values and compared to those with other groups such as morpholine or represent the concentration of the compound that decreases methyl-piperazine. Among those compounds that provided 50% the enzyme activity determined in the absence of the satisfactory results, unsubstituted analogues at the meta inhibitor. ND = not determined. position (X group) yielded better inhibition results (measured ALDH1A1. Given the presence of the aldehyde group and as remaining activity at 10 μM compound) than those the evidence that DEAB has been reported to be a slow substituted with halogen atoms, that is, DEAB (m-H, 2%) ALDH1A1 substrate, we decided to perform both inhibition versus 13 (m-Br, 24%), pyrrolidine 4 (m-H, 10%) versus 12 and substrate studies. Most compounds displayed lower (m-Cl, 17%), or dipropylamine 1 (m-H, 12%) versus 14 (m-Br, inhibitory potency against ALDH1A1 compared with DEAB 45%) (Figure S1). In fact, 14 yielded an IC50 of 7.08 ± 0.70 and were evaluated at an initial dose of 10 μM (Figure S1). μM, approximately 15-fold less active when compared to Due to the significant substrate activity shown with many DEAB as an inhibitor for ALDH1A1 (IC50 = 0.48 ± 0.06 μM). compounds, even much higher than that with DEAB (Figure Contrary to ALDH1A3 and 3A1, the most potent ALDH1A1 S2), it was difficult to assess their inhibitory properties. inhibitors were analogues bearing an electron-donating group: 3834 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article Table 1. IC50 Values of the Most Potent DEAB Analogues against ALDH1A1, ALDH1A3, and ALDH3A1 Isoforms 26 (m-CH3, IC50 = 0.80 ± 0.16 μM) and 29 (m-OCH3, IC50 = compounds to be superior to DEAB in inhibiting ALDH1A3 0.88 ± 0.05 μM) (Table 1 1). enzymatic activity with hexanal as a substrate (Table 1 and ALDH1A3. From the initial one-dose (10 μM) evaluation Figure S1). (Figure S1), the selection of compounds was based not only on There are several interesting SARs that reveal trends of inhibitory properties but also on their ability to act as inhibitory properties. Analogues bearing a methyl group at the substrates for the other two ALDH isoforms. Some meta position to the aldehyde and an aliphatic moiety at the compounds were shown to display very low IC50 values para position reveal that the modulation of the latter from regarding ALDH1A3 (Table 1), yet its high percentage of diethyl (DEAB) or dipropyl (1) to constrained heterocycles substrate activity for ALDH1A1 and/or ALDH3A1 would based on pyrrolidine (26) or piperidine (15) rings increases complicate their further investigation and use as inhibitors. For the inhibitory effect: 15 (IC50 = 0.29 μM) > 26 (IC50 = 1.15 example, 16 displayed the lowest IC50 value of all compounds μM) > DEAB (IC50 = 10.4 μM) ≈ 1 (only tested at 10 μM, in the library for ALDH1A3 (0.26 μM), yet it was shown to be Figure S1). Analogues incorporating para-positioned piper- a good ALDH3A1 substrate (Figure S2). Similarly, compounds azine 24 and especially morpholine 27 also showed inhibitory 7 and 14 revealed comparable IC50 values (0.55 and 0.66 μM, capacity decreasing ALDH1A3 activity down to 45 and 10% at respectively); however, 7 presented a significantly higher 10 μM dose, respectively. Pyrrolidine 12 with a chlorine ALDH1A1 substrate activity compared to that of 14 (17.9 vs installed at the meta position to the aldehyde was almost twice 3.14%). This result may be due to chlorine being a smaller as potent (IC50 = 0.31 μM) as the propyl analogue 7 (IC50 = atom compared to bromine, and thus, steric hindrance might 0.55 μM), further substantiating the presence of a heterocycle explain why compound 7 can be better accommodated into the at the para position for enhanced potent ALDH1A3 inhibitory active site. Nonetheless, ALDH1A3 studies revealed many activity. Interestingly, 34 was the only compound with an NO2 3835 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article electron-withdrawing group that appeared effective at inhibit- tends to increase, whereas the Vmax value barely changes. This ing ALDH1A3, with an IC50 = 0.27 μM. Comparison of propyl behavior best fits with a competitive type of inhibition. For analogues 7 (m-Cl) and 14 (m-Br) revealed similar IC50 values these compounds, Ki values are all below 1 μM, in good (0.55 and 0.63 μM, respectively), indicating that the atomic agreement with the IC50 values. The calculated value for 14 volume and electronegativity of these halogens are not critical was Ki = 0.46 ± 0.15 μM, which indicates that this compound for ALDH1A3 inhibition. is an excellent inhibitor for ALDH1A3. Notably, comparison of diethyl 13 (IC50 = 10.7 μM) and With all observations taken together, it can be concluded diisopropyl 14 (IC50 = 0.63 μM) compounds revealed that compound 14 yielded the most promising inhibition approximately 17-fold difference in capacity to inhibit parameters while also displaying very low activity as a substrate ALDH1A3. This result suggests that the extra two methyl for the ALDH1A1 and ALDH3A1 isoforms (lower than 5% at groups may provide additional nonpolar interactions with the 10 μM for both isoforms). active-site pocket, enough to strengthen binding. Figure S6 ALDH3A1. Compounds 14, 18, and 19 were shown to depicts the comparison of docking poses of 13 with 14 and display IC50 values below 10 μM, with both 18 (IC50 = 1.61 DEAB within the ALDH1A3 binding pocket. Analogue 13 μM) and 19 (IC50 = 1.29 μM) approximately threefold more demonstrated a similar binding orientation as DEAB, potent than DEAB (IC50 = 5.67 μM). The latter two establishing H-bonds with Cys313, Cys314, and Thr315. compounds share similar chemical structures, and the Both the N,N-diethyl side chains of 13 and DEAB established inhibitory activity is most likely related to three structural van der Waals contacts with Ile132 and Leu185 in the binding features: the lipophilicity of the para-substituted functional site of ALDH1A3, whereas analogue 14 was able to form only groups, the side chain flexibility involving free rotations, and one H-bond with Thr315 and maintained van der Waals the meta-substituted functional group. They differ in the para contacts with Ile132 and Leu185. Interestingly, the two extra substitution groups diethylamine and dipropylamine. Com- methyl groups at the side chain of 14 were accommodated pared to other analogues, these groups appear to be better between Trp189 and Leu471 residues, favoring stronger van accommodated in the active site and show more activity than der Waals contacts than 13 and DEAB, thus improving its those compounds with a similar m-NO2 group but with p- binding to ALDH1A3. Therefore, these findings may explain substituted groups of lower lipophilicity (i.e., morpholine 32 or why 13 acted as an ALDH1A1 substrate with almost 60% pyrrolidine 34). activity relative to hexanal, whereas 14 only yielded 4% activity It can be concluded that the NO2 group at the meta position (Figure S2). seems to be providing these compounds with higher inhibitory From all the compounds with a CH3 group at the meta activity. The docking analysis suggested that Tyr115, one of position, compound 15 displayed the most potent inhibitory the ALDH3A1 residues that normally interacts with the CHO properties. Compound 16, containing an isopropoxy group at part of the molecule, presents some interaction with the NO2 the para position, was found to be one of the most potent group as well. This additional binding might be related to the compounds under the conditions investigated (IC50 = 0.26 better inhibitory capacity. μM). Significantly, several new analogues exhibited up to 100- The only structural difference among 18 and 19 is the N- fold higher potency in inhibiting ALDH1A3 as compared to substituted group at the para position to the aldehyde, and DEAB (IC50 = 10.20 ± 2.15 μM) when using hexanal as a although the results of the ALDH3A1 inhibition are very substrate. similar, their ability to act as substrates differs, especially with The type of inhibition and Ki values determined for ALDH1A1 (17% of 18 vs 69% of 19, Figure S2). Other para- compounds 14, 15, and 16 against ALDH1A3 are shown in substituted dipropylamine analogues (1, 7, 14, 17, 18, and 38) Figures 1, S4, and S5, respectively. It can be observed that were evaluated, but only those with electron-withdrawing when increasing the inhibitor concentration, the Km value groups demonstrated the capacity to significantly inhibit ALDH3A1 activity (below 15% remaining activity). Selectivity might be linked to a Gln122 residue, as a site for providing selectivity for ALDH3A1 only.29 Kinetic evaluation of both 18 and 19 revealed them as excellent candidates for ALDH3A1 inhibition. Both com- pounds were further characterized as ALDH3A1 inhibitors by determination of the type of inhibition and Ki value. Data are shown in Figures 2 and 3. Compounds were best fitted to a competitive type of inhibition. The calculated value for Ki was 0.30 ± 0.06 μM for 18 and 0.24 ± 0.04 μM for 19, which indicates that compounds 18 and 19 are excellent inhibitors of ALDH3A1 functional activity. Additionally, we decided to further investigate the role of 18 as an ALDH3A1 substrate (4.21% with 10 μM, Figure S2) to assess whether it would interfere with the inhibition experi- ments. Kinetic analysis of the saturation profile for 18 with Figure 1. Inhibition kinetics of ALDH1A3 by compound 14 at ALDH3A1 was best fitted to the Michaelis−Menten equation various concentrations of inhibitor: −○− 0 μM; −●− 0.25 μM; −□− 2.5 μM; and −■− 5 μM. Hexanal was used as the substrate. modified for substrate inhibition (Figure 4), producing a Km of The values of the kinetic parameters calculated from a fit to the 2.82 ± 0.35 μM and a Ksi of 113 ± 25 μM. For comparison, competitive inhibition equation are Vmax = 0.37 ± 0.01 U/mg; Km = the Km of the best ALDH3A1 substrate, 4-NBA, is 10-fold 16.1 ± 4 μM; and Ki = 0.46 ± 0.15 μM. Results are the mean ± SE of higher, with a Km of 31.0 ± 4.9 μM. Regarding kcat, 4-NBA has duplicate experiments. a 26-fold higher turnover number (359 ± 14 min−1) than 3836 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article compound 18, which has a kcat value of 13.4 ± 0.7 min−1. These findings suggest that 18 performs better as an inhibitor than as an ALDH3A1 substrate. We also performed two different IC50 calculations for compound 18. The aim was to check whether 18 was acting as a substrate during the standard 5-min preincubation time, when NADP+ and the enzyme were added but with no standard substrate in the reaction mixture. One IC50 was calculated by incubating with the cofactor and the other was calculated without it, and no differences were observed (Figure S6), indicating that the function of compound 18 as an inhibitor was not being affected by its role as a substrate under the assayed conditions. Antiproliferative Activity of DEAB Analogues in Figure 2. Inhibition kinetics of ALDH3A1 by compound 18 at Prostate Cancer Cell Lines. Several members of the various concentrations of inhibitor: −○− 0 μM; −●− 0.5 μM; −□− ALDH family of enzymes have been shown to be expressed 1 μM; and −■− 5 μM. 4-Nitrobenzaldehyde (4-NBA) was used as in PCa,30−34 and ALDH1A1 and 1A3 isoforms have been the substrate. The values of the kinetic parameters calculated from a reported to be expressed at higher levels in tumor tissues fit to the competitive inhibition equation are Vmax = 3.59 ± 0.16 U/ compared to benign prostatic hyperplasia and normal mg; Km = 26.92 ± 5.41 μM; and Ki = 0.30 ± 0.06 μM. Results are the prostate.35 ALDHs have also been acknowledged to promote mean ± SE of duplicate experiments. clonogenic and migration cell capabilities in vitro and enhance the metastatic potential in vivo,34,36 while the expression correlates with a higher Gleason score (G8-9) in vivo.20,31,35 We have previously shown the importance of the RA pathway in PCa37,38 and ALDHs as a potential target,20 and in this context, it was suitable to explore the new DEAB library of compounds. Due to their ALDH expression profile (Figure 5), we chose to use a small panel consisting of the three prostate cancer cell lines PC3, DU145, and LNCaP. Figure 3. Inhibition kinetics of ALDH3A1 by compound 19 at various concentrations of inhibitor: −○− 0 μM; −●− 0.5 μM; −□− 1 μM; and −■− 5 μM. 4-NBA was used as the substrate. The values of the kinetic parameters calculated from a fit to the competitive inhibition equation are Vmax = 3.69 ± 0.16 U/mg; Km = 27.81 ± 5.36 μM; and Ki = 0.24 ± 0.04 μM. Results are the mean ± SE of duplicate experiments. Figure 5. Immunoblot analysis was carried out using ALDH1A1, 1A3, and 3A1 specific antibodies in a panel of immortalized PCa cell lines (PC-3, LNCaP, and DU145) and using GAPDH as a control. All compounds were initially assessed using the MTT assay in an initial two-dose point screen (96 h exposure) to identify hit compounds. The results revealed clear dose-dependent trends (Figure S14), which, together with observations from the biochemical screening, led to 17 compounds including DEAB to be further investigated in a five-dose screen. All DEAB analogues displayed IC50 values in the micromolar range (10−200 μM) as measured using the MTT assay. Most compounds showed equipotent antiproliferative activity in PC3 (expressing ALDH1A1 and 3A1) and DU145 (expressing ALDH1A1) cell lines and increased potency against LNCaP Figure 4. Km value for ALDH3A1 using compound 18 as a substrate. Experimental values were fitted to the substrate inhibition equation (expressing ALDH1A3) cells (Table 2); the cellular potency is and the kinetic values were Km = 2.82 ± 0.35 μM, Ksi = 113 ± 25 μM, likely to be multifactorial and not just a direct correlation of and kcat of 13.4 ± 0.7 min−1. Data were the result of duplicate the ALDH isoform expression. Nonetheless, the dipropyl experiments and expressed as the mean ± SE. moiety appeared to provide compound 14 with a higher antiproliferative potency than diethyl-based analogue 13 and piperidine analogue 23 across all three cell lines. These findings were also found for compound 18 with the dipropyl 3837 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article Table 2. Antiproliferative Effect of DEAB Analogues Against evaluated using two different concentrations (50 and 200 μM), PC3, LNCaP, and DU145 PCa Cell Lines and results revealed dose-dependent reduction in percentage cell viability of primary prostate epithelial cultures (Figure 6A− IC50 (μM) Cmpd PC3 LNCaP DU145 1 >200 73 ± 14 190 ± 25 6 172 ± 30 137 ± 14 133 ± 11 7 77 ± 17 10 ± 3 90 ± 27 8 130 ± 9 37 ± 1.4 110 ± 3 13 166 ± 41 47 ± 0.6 123 ± 12 14 47 ± 6 25 ± 1 61 ± 5 15 123 ± 29 61 ± 14 106 ± 13 16 >200 >200 >200 17 106 ± 25 61 ± 18 103 ± 4 18 98 ± 12 31 ± 6 100 ± 13 19 >200 136 ± 13 >200 21 >200 179 ± 14 187 ± 14 22 196 ± 4 82 ± 10 155 ± 32 23 84 ± 6 46 ± 10 84 ± 12 26 >200 172 ± 8 >200 38 75 ± 10 50 ± 3 71 ± 11 DEAB >200 >200 >200 moiety, which showed higher antiproliferative activity than diethyl analogue 19. Therefore, analogues 14 and 18 were shown to be the most promising compounds in agreement with data obtained from the biochemical screening. In Figure 6. Cell viability of primary cells following treatment with 14 addition, compound 14 showed the highest antiproliferative (A), 18 (B), and DEAB (C) as a single or combination treatment activity in both DU145 and PC3 cell lines with IC50 values of with docetaxel. Patient samples assessed included BPH sample H415/ 61 and 47 μM, respectively; this correlated with the 15 and cancer samples H568/15 RM, H431/14 LM, H488/14 RM, biochemical assay IC50 data obtained for ALDH1A1 (7.08 and H517/15 RM. The experiment was carried out in triplicate and μM), ALDH1A3 (0.63 μM), and ALDH3A1 (8.00 μM). values are represented as the mean. Statistical significance was IC50 values represent the concentration of the compound calculated using a paired two-tailed Student’s t-test, in which the mean that decreases cell survival by 50%. Results are expressed in μM of untreated cells was compared with the mean of treated cells, *p < ± SD and derived from at least three independent experiments. 0.05, **p < 0.01, and ***p < 0.001, combination treatment with Cell line sensitivity after treatment with ALDH-targeting docetaxel. compounds might not only be a reflection of the ALDH expression and functional activity. Previous reports have C). In line with the PCa cell line data, analogues 14 and 18 demonstrated that there is a striking difference in the were more potent than DEAB, a pattern also evident in the metabolic phenotypes of the three cell lines used in this combination with docetaxel (1 nM). To gain further insights study.39 Due to a mitochondrial dysfunction, PC3 and DU145 into the observations obtained from both the biochemical cells have been shown to have an increased glycolytic reliance, assays and chemosensitivity experiments, docking studies were unlike LNCaP which is highly oxidative, which might performed for compounds 14 and 18, as they showed contribute to ALDH activity and compound sensitivity. promising results. Resistance mechanisms might contribute to compound Docking Studies of Compounds 14 and 18 on sensitivity. For example, the PC3 cell line represents the type ALDH1A1, ALDH1A3, and ALDH3A1 Isoforms. Analogues of cancer that is difficult to treat as it is an androgen- 14 and 18 were docked into the three investigated isoforms independent metastatic prostate cancer type.20 Nevertheless, (ALDH1A1, 1A3, and 3A1), with all analysis shown in Figures apart from compound 16, it was noticeable that all analogues S7−S13. In addition, binding modes with DEAB were included selected for chemosensitivity screening displayed more potent to compare its binding affinity to these three analogues, as antiproliferative effects when compared with DEAB. shown in Figure 7. From the docking study of analogue 14 ALDHs have also been linked to chemo- and radioresistance with the ALDH1A1 binding site (Figure 7A), the aldehyde in cancer therapy.40,41 An increased expression of ALDH can oxygen of 14 was found to form an H-bond with Tyr297, and have a chemoprotective effect on cells due to their metabolic the phenyl ring established face-to-face π−π stacking with and detoxifying abilities [reviewed in ref 41]. The taxanes Phe171. The electron-rich m-Br group was found to be in close paclitaxel and docetaxel have been reported to be less effective proximity of Phe171, Tyr297, and Ile304, capable of in ALDH-expressing cells.42−44 Given that docetaxel is the constituting van der Waals contacts with these residues. One most commonly used cytotoxic drug for the treatment of of the propyl groups of the N-substituted side chain established advanced PCa, we next decided to evaluate three compounds van der Waals contacts with Trp178. Compound 14 was found (14, 18, and DEAB) as single agents and in combination with to superimpose on DEAB, maintaining the abovementioned docetaxel against five patient samples (four cancers with protein−ligand interactions. Compound 18 (Figure 7B) Gleason score 7 and one BPH) to investigate the potential of displayed similar protein−ligand interactions to 14 and such a combination in clinical samples. The compounds were DEAB, in the binding pocket of ALDH1A1. However, the 3838 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article Figure 7. Molecular docking of compounds 14 (green) and 18 (magenta) into ALDH1A1 (A,B, PDB ID: 4WPN), ALDH1A3 (C,D, PDB ID: 5FHZ), and ALDH3A1 (E,F, PDB ID: 4H80) binding sites. Best fit binding modes are compared to DEAB (orange). presence of m-NO2 in the place of an m-Br group did not result Figure 7D, did not properly occupy the ALDH1A3 binding in an H-bond interaction within the binding site. site, justifying the poor/inactive potencies against the As depicted in Figure 7C, analogue 14 occupied the binding ALDH1A3 isoform obtained from the biochemical studies. pocket of ALDH1A3 with high affinity through several The presence of the m-NO2 group in 18, as an H-bond intermolecular forces: the aldehyde oxygen formed a H-bond acceptor, completely altered the binding disposition of this contact with Thr315, the m-Br phenyl ring established van der analogue within the enzyme binding site, by forming an H- Waals interactions with Phe182, Trp189, and Leu471, and the bond network at the entrance of the ALDH1A3 binding pocket electronic interactions between bromine and both Phe308 and with Arg139 and Ala473 residues, thus restricting their deeper Ile132 residues significantly contributed to enhanced binding entry into the active site. affinity with the target, which may explain the results obtained Both ligands constituted the fundamental H-bonds with in the ALDH1A3 biochemical studies (Table 1). In addition, Cys243 and Asn114 of ALDH3A1 (Figure 7E,F). In particular, compound 14 was found to be superimposed with the binding compound 18 exhibited a preference binding for ALDH3A1 disposition of DEAB, altogether contributing to the improved when compared with 1A1 and 1A3 isoforms. Apart from H- binding affinity of 14 for ALDH1A3. Analogue 18, as shown in bond contacts, both analogues displayed significant van der 3839 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article Waals interactions in the ALDH3A1 binding pocket; the with a high Gleason score of 7. Although the results are only phenyl ring of 14 was observed to establish an edge-to-face indicative, they do suggest that docetaxel treatment of π−π stacking with Tyr115 and van der Waals contact with the aggressive primary PCa cells might benefit from the inclusion Ile394 side chain. The m-Br group of 14 was found to be in of an ALDH inhibitor such as compound 14 or 18. In uterine proximity of Ile391, thus favoring the electronic interaction, endometrial cancer, paclitaxel treatment has been shown to and one of the propyl of the N-substituted side chain formed a increase the proportion of ALDHhigh cells in clinical samples hydrophobic contact with Tyr65. Figure 7F also shows that the and in spheroids.54 Spheroids are enriched in CSCs, which phenyl ring of ligand 18 formed a face-to-face π−π stacking mainly depend on an enhanced glycolytic metabolic pathway with Tyr315 and the N-substituted side chain provided a van for their proliferation and survival. This glycolytic activation is der Waals contact with Tyr65. In addition, the docking results mediated by the expression of ALDH and its crucial of 14 and 18 were compared with those for DEAB in the downstream effector GLUT1. Interestingly, the combination ALDH3A1 binding pocket (Figure 7E,F). DEAB was super- of ALDH/GLUT1 inhibitors with paclitaxel has been shown to imposed on the binding orientation of 14 within the suppress the proliferation of endometrial cancer cells in a ALDH3A1 binding site, exhibiting identical H-bond and synergistic manner, indicating that ALDH-dependent GLUT hydrophobic interactions. The phenyl ring of DEAB formed activation might be relevant for the maintenance of chemo- an edge-to-face π−π stacking with Tyr115 and van der Waals resistance of CSCs. contacts with Ile394, whereas the ethyl group from one of the In conclusion, further mechanistic studies are required to N-ethyl side chains of DEAB established another van der fully understand how inhibition of ALDH activity might be Waals contact with Tyr65. The phenyl ring of 18 was oriented linked to potentiation of docetaxel treatment in PCa. Given the differently relative to the phenyl ring of 14. well-established capacity of DEAB to inhibit ALDHhigh- ■ CONCLUSIONS DEAB has been shown to be effective in treating aggressive expressing CSCs, it also remains to be explored whether analogues from our DEAB library, such as 14 and 18, can be used as chemical probes to further unravel the significance of ALDHhigh-expressing subpopulations with SC-like propensity ALDH expression in CSCs and possible link to taxane in vitro and features as a key component of the Aldefluor assay resistance. An additional outcome of the present work might used to identify cancer cells with SC properties. In regard to indicate that the Aldefluor assay specificity could be improved the former, it is possible that the inclusion of a therapeutic using compounds 14 and 15 to label ALDH1A3-expressing agent aimed at eradicating prostate CSCs could increase the cells or 18 and 19 for ALDH3A1-expressing cells, given their overall survival rate.19,45−47 In PCa, patients who no longer increased selectivity as compared with DEAB. respond to androgen therapy (ADT) develop an aggressive disease known as castrate-resistant prostate cancer (CRPC), which has high propensity for metastasis and short median ■ EXPERIMENTAL SECTION General Procedures. All materials and reagents were used as survival rates ranging from 12.1 to 27.0 months.48−50 received with no further purification. 4-Fluoro-3-nitrobenzaldehyde, Currently, drugs that are used to treat advanced PCa include 3-cyano-4-fluorobenzaldehyde, 2-bromo-4-fluorobenzaldehyde, and 4- small-molecule-based therapies such as the androgen receptor fluoro-2-methoxybenzaldehyde were purchased from fluorochem, 3- (AR) inhibitor enzalutamide,51 the CYP17A1 inhibitor bromo-4-benzaldehyde, 4-fluorobenzaldehyde, 4-fluoro-3-methoxy- abiraterone acetate,51 and the taxanes docetaxel52 and benzaldehyde, 3-chloro-4-fluorobenzaldehyde 4-fluoro-3-methylben- cabazitaxel.53 Unfortunately, these therapies are rarely curative, zaldehyde, diethylamine, dipropylamine, morpholine, piperidine, necessitating the identification of new molecular targets and/or pyrrolidine, DMSO, and 4-diethylaminobenzaldehyde (DEAB) were development of therapeutic strategies to treat aggressive PCa. purchased Sigma-Aldrich; 1-methylpiperazine and 5-nitrovanillin were purchased from Acros Organics. Chemical reactions were monitored In this study, we wanted to (i) generate a DEAB library to help by analytical thin-layer chromatography using Merck 9385 silica gel understand ALDH isoform selectivity and (ii) investigate the 60 F254 aluminum-backed plates through visualizing the spotted potential of treating PCa cells as single agents and combination plates under ultraviolet (UV) at 254 and 366 nm. Intermediates and with docetaxel. Accordingly, we synthesized a small library of final products were purified by column chromatography using silicagel 40 compounds with a benzaldehyde scaffold and evaluated 60A 40−63 μm. Proton and carbon NMR spectra were analyzed for them against three ALDH isoforms1A1, 1A3, and 3A1 all intermediates and final products on a Bruker AMX400 (400 MHz) known to be expressed in PCa. The results showed that an nuclear magnetic resonance spectrometer. Chemical shifts were electron-donating group (amine or oxygen) at the para reported in parts per million (δ, ppm) downfield from internal position to the aldehyde group is important for activity, TMS. Coupling constants (J) were expressed in Hertz (Hz). High- resolution mass spectra were obtained by the Engineering and particularly when connecting with a lipophilic carbon chain Physical Sciences Research Council (EPSRC) mass spectrometry that participates in van der Waals interactions with ALDH service, Swansea. Melting points were measured with a Gallenkamp active sites. In addition, the presence of an electron- melting point apparatus. All compounds biologically evaluated were withdrawing group at the meta position to the aldehyde also >95% pure by HRMS/HPLC analysis except analogue 40 (82%); was found to increase affinity to the binding site. The HPLC traces for all compounds can be found in the Supporting synthesized compounds showed promising inhibitory proper- Information. ties, notably identifying compounds 14−16, 18, and 19 with General Procedures for the Synthesis of Compounds 1−15, superior activity against ALDH1A3 and 3A1 isoforms when 17−32, 34, 36−38. To a stirred solution of starting aldehyde (1 equiv) in DMF (10 mL), the corresponding amine (6.0 equiv) and compared with DEAB. K2CO3 (2.0 equiv) were added and stirred at (25−100) °C for The antiproliferative activity of the compounds was also hours−days. The reaction was then cooled to room temperature, and evaluated in three PCa cell lines, and interestingly, most DMF volume was reduced by evaporation under vacuum. Water (30 compounds outperformed DEAB in cellular potency. We mL) was then added to the mixture and stirred for 30 min. The further investigated DEAB and analogues 14 and 18 in mixture was then extracted with ethyl acetate (EA) (3 × 20 mL), and combination with docetaxel in primary PCa cells from tumors the organic fractions were combined, washed with water, and dried 3840 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article with MgSO4. EA was then evaporated under vacuum to give the 117.7. HRMS (ESI): calcd for C10H8N2O [M + H]+, 173.0709; product as crude, which was purified by silica gel column found, 173.0705. chromatography, affording the desired targeted compounds. 3-Chloro-4-(dipropylamino)benzaldehyde (7). Starting re- General Procedures for the Synthesis of Compounds 16, agents were 3-chloro-4-fluorobenzaldehyde and dipropylamine, 33, and 35. To a stirred solution of the starting aldehyde (1 equiv) in processed at 100 °C for 24 h. The crude was purified by column DMF (10 mL), iodopropane (4.0 equiv) and K2CO3 (2 equiv) were chromatography using EA/PE (0.1:10) to afford 33.1% of the title added and stirred at 90 °C for 6 h. The reaction was then cooled to product as a yellow oil. Rf = 0.42 (EA/PE, 1:10). 1H NMR (400 room temperature, and DMF volume was reduced by evaporation MHz, CDCl3): δ 9.73 (s, 1H), 7.75 (d, 1H, J = 1.9 Hz), 7.58 (dd, 1H, under vacuum. Water (30 mL) was then added to the mixture and J = 8.4, 1.9 Hz), 6.99 (d, 1H, J = 8.4 Hz), 3.21−3.05 (m, 4H), 1.58− stirred for 30 min. The mixture was then extracted with EA (2 × 20 1.38 (m, 4H), 0.80 (t, 6H, J = 7.4 Hz). 13C NMR (101 MHz, CDCl3): mL), and the organic fractions were combined, washed with water, δ 189.7, 153.8, 132.9, 130.1, 128.7, 127.7, 121.2, 53.8, 20.6, 11.4. and dried with MgSO4. EA then evaporated under vacuum to give the HRMS (ESI): calcd for C13H18ClNO [M + H]+, 240.1150; found, product as crude, which was then purified by silica gel column 240.1150. chromatography, affording the desired targeted compounds. 3-Chloro-4-(piperidin-1-yl)benzaldehyde (8). Starting re- 4-(Dipropylamino)benzaldehyde (1). Starting reagents were 4- agents were 3-chloro-4-fluorobenzaldehyde and piperidine, processed fluorobenzaldehyde and dipropylamine, processed at 100 °C for 48 h. at 100 °C for 24 h. The crude was purified by column The crude was purified by column chromatography using 5% EA in chromatography using EA/PE (0.2:10) to afford 65.0% of the title petroleum ether (PE), affording 56% of the title product as a yellow product as a yellow oil. Rf = 0.08 EA/PE (0.2:10). 1H NMR (400 oil. Rf = 0.26 (EA/PE, 1:10). 1H NMR (400 MHz, CDCl3): δ 9.72 (s, MHz, CDCl3): δ 9.84 (s, 1H), 7.86 (d, 1H, J = 1.9 Hz), 7.71 (dd, 1H, 1H), 7.72 (d, 2H, J = 8.9 Hz), 6.67 (d, 2H, J = 8.9 Hz), 3.40−3.25 (t, J = 8.2, 1.9 Hz), 7.11 (d, 1H, J = 8.2 Hz), 3.22−3.08 (m, 4H), 1.86− 4H, J = 7.2 Hz), 1.78−1.53 (m, 4H), 0.98 (t, 6H, J = 7.4 Hz). 13C 1.70 (m, 4H), 1.65 (m, 2H). 13C NMR (101 MHz, CDCl3): δ 189.9, NMR (101 MHz, CDCl3): δ 190.0, 152.5, 132.2, 124.8, 111.0, 53.0, 155.6, 132.1, 130.9, 129.6, 128.3, 120.0, 52.2, 25.9, 24.1. HRMS 20.3, 11.4. HRMS (ESI): calcd for C13H19NO [M + H]+, 206.1539; (ESI): calcd for C12H14ClNO [M + H]+, 224.0837; found, 224.0832. found, 206.1535. 3-Chloro-4-morpholinobenzaldehyde (9). Starting reagents 4-(Piperidin-1-yl)benzaldehyde (2). Starting reagents were 4- were 3-chloro-4-fluorobenzaldehyde and morpholine, at processed fluorobenzaldehyde and piperidine, processed at 100 °C for 24 h. The 100 °C for 24 h. The crude was purified by column chromatography crude was purified by column chromatography using EA/PE using EA/PE (0.5:10) to afford 87% of the title product as a yellow (0.05:10) to afford 66.7% of the title product as a yellow solid. Rf solid. Rf = 0.24 (EA/PE, 2:10). mp 83 ± 1 °C. 1H NMR (400 MHz, = 0.13 (EA/PE, 1:10). mp 70 ± 1 °C. 1H NMR (400 MHz, CDCl3): CDCl3): δ 9.88 (s, 1H), 7.89 (d, 1H, J = 1.9 Hz), 7.76 (dd, 1H, J = δ 9.66 (s, 1H), 7.64 (d, 2H, J = 8.9 Hz), 6.81 (d, 2H, J = 8.9 Hz), 3.32 8.3, 1.9 Hz), 7.12 (d, 1H, J = 8.3 Hz), 3.98−3.80 (m, 4H), 3.27−3.12 (m, 4H), 1.59 (m, 6H). 13C NMR (101 MHz, CDCl3): δ 190.3, (m, 4H). 13C NMR (101 MHz, CDCl3): δ 189.9, 154.2, 132.1, 131.7, 155.1, 132.0, 126.2, 113.3, 48.4, 25.3, 24.3. HRMS (ESI): calcd for 129.7, 128.5, 119.9, 66.8, 51.1. HRMS (ESI): calcd for C11H12ClNO2 C12H15NO [M + H]+, 190.1226; found, 190.1221. [M + H]+, 226.0629; found, 226.0630. 4-Morpholinobenzaldehyde (3). Starting reagents were 4- 3-Chloro-4-(4-methylpiperazin-1-yl)benzaldehyde (10). fluorobenzaldehyde and morpholine, processed at 100 °C for 48 h. Starting reagents were 3-chloro-4-fluorobenzaldehyde and N-methyl- The crude was purified by column chromatography using 10% EA in piperazine, processed at 100 °C for 24 h. The crude was purified by petroleum ether, affording 56.6% of the title product as a yellow solid. column chromatography starting with CH2Cl2 and gradually Rf = 0.47 (EA/PE, 1:1). mp 63 ± 1 °C. 1H NMR (400 MHz, CDCl3): increasing CH3OH to final mixture of CH3OH/CH2Cl2 (0.5:10), affording 80.5% of the title product as a yellow solid. Rf = 0.20 δ 9.73 (s, 1H), 7.70 (d, 2H, J = 8.9 Hz), 6.85 (d, 2H, J = 8.9 Hz), 3.78 (CH3OH/CH2Cl2, 0.5:10). mp 42 ± 1 °C. 1H NMR (400 MHz, (t, 4H, J = 4.9 Hz), 3.27 (t, 4H, J = 4.9 Hz). 13C NMR (101 MHz, CDCl3): δ 9.76 (s, 1H), 7.77 (d, 1H, J = 1.9 Hz), 7.63 (dd, 1H, J = CDCl3): δ 190.6, 155.2, 131.8, 127.6, 113.5, 66.5, 47.3. HRMS (ESI): 8.3, 1.9 Hz), 7.02 (d, 1H, J = 8.3 Hz), 3.15 (m, 4H), 2.55 (m, 4H), calcd for C11H13NO2 [M + H]+, 192.1019; found, 192.1015. 2.29 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 189.8, 154.5, 132.0, 4-(Pyrrolidin-1-yl)benzaldehyde (4). Starting reagents were 4- 131.4, 129.7, 128.3, 120.0, 54.9, 50.6, 46.0. HRMS (ESI): calcd for fluorobenzaldehyde and pyrrolidine, processed at 100 °C for 48 h. C12H15ClN2O [M + H]+, 239.0944; found, 239.0946. The crude was purified by column chromatography using EA/PE 3-Chloro-4-(diethylamino)benzaldehyde (11). Starting re- (0.5:10) to afford 87.5% of the title product as a yellow solid. Rf = agents were 3-chloro-4-fluorobenzaldehyde and diethylamine, pro- 0.31 (EA/PE, 2:10). mp 92 ± 1 °C. 1H NMR (400 MHz, CDCl3): δ cessed at 55 °C for 48 h. The crude was purified by column 9.73 (s, 1H), 7.74 (d, 2H, J = 8.8 Hz), 6.60 (d, 2H, J = 8.8 Hz), 3.40 chromatography using EA/PE (0.1:10) to afford 65.9% of the title (t, 4H, J = 6.6 Hz), 2.07 (t, 4H, J = 6.6 Hz). 13C NMR (101 MHz, product as a yellow oil. Rf = 0.60 (EA/PE, 2:10). 1H NMR (400 CDCl3): δ 190.3, 151.9, 132.2, 124.9, 111.4, 47.8, 25.4. HRMS (ESI): MHz, CDCl3): δ 9.73 (s, 1H), 7.76 (d, 1H, J = 1.9 Hz), 7.59 (dd, 1H, calcd for C11H13NO [M + H]+, 176.1070; found, 176.1065. J = 8.5, 1.9 Hz), 6.99 (d, 1H, J = 8.5 Hz), 3.24 (q, 4H, J = 7.1 Hz), 4-(4-Methylpiperazin-1-yl)benzaldehyde (5). Starting re- 1.05 (t, 6H, J = 7.1 Hz). 13C NMR (101 MHz, CDCl3): δ 189.8, agents were 4-fluorobenzaldehyde and 1-methylpiperazine, processed 153.5, 132.7, 130.2, 128.8, 127.8, 121.0, 45.7, 12.5. HRMS (ESI): at 90 °C for 24 h. The crude was purified by column chromatography calcd for C11H14ClNO [M + H]+, 212.0837; found, 212.0835. using (CH3OH/CH2Cl2, 0.5:10) to afford 96.7% of the title product 3-Chloro-4-(pyrrolidin-1-yl)benzaldehyde (12). Starting re- as a yellow solid. Rf = 0.49 (CH3OH/CH2Cl2, 1:10). mp 74 ± 1 °C. agents were 3-chloro-4-fluorobenzaldehyde and pyrrolidine, processed 1 H NMR (400 MHz, CDCl3): δ 9.73 (s, 1H), 7.70 (d, 2H, J = 8.9 at 60 °C for 24 h. The crude was purified by column chromatography Hz), 6.87 (d, 2H, J = 8.9 Hz), 3.40−3.31 (m, 4H), 2.58−2.45 (m, using EA/PE (0.5:10) to yield 81.9% of the title product as a yellow 4H), 2.31 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 190.3, 155.0, oil. Rf = 0.54 (EA/PE, 2:10). 1H NMR (400 MHz, CDCl3): δ 9.61 (s, 131.8, 127.0, 113.5, 54.6, 47.0, 46.1. HRMS (ESI): calcd for 1H), 7.66 (d, 1H, J = 1.9 Hz), 7.50 (dd, 1H, J = 8.6, 1.9 Hz), 6.66 (d, C12H16N2O [M + H]+, 205.1335; found, 205.1331. 1H, J = 8.6 Hz), 3.52 (t, 4H, J = 6.6 Hz), 1.97−1.77 (m, 4H). 13C 4-(1H-Imidazole-1-yl)benzaldehyde (6). Starting reagents were NMR (101 MHz, CDCl3): δ 189.3, 150.7, 133.9, 129.5, 127.1, 119.5, 4-fluorobenzaldehyde and 1H-imidazole, processed at 100 °C for 24 115.3, 51.2, 25.8. HRMS (ESI): calcd for C11H12ClNO [M + H]+, h. The crude was purified by column chromatography using 210.0680; found, 210.0678. (CH3OH/CH2Cl2, 0.1:10) to afford 38.9% of the title product as a 3-Bromo-4-(diethylamino)benzaldehyde (13). Starting re- pale-yellow solid. Rf = 0.29 (CH3OH/CH2Cl2, 0.05:10). mp 161 ± 1 agents were 3-bromo-4-fluorobenzaldehyde and diethylamine, pro- °C. 1H NMR (400 MHz, CDCl3): δ 10.05 (s, 1H), 8.05−7.99 (m, cessed at 55 °C for 48 h. The crude was purified by column 3H), 7.59 (d, 2H, J = 8.5 Hz), 7.38 (s, 1H), 7.26 (s, 1H). 13C NMR chromatography using EA/PE (0.1:10) to yield 72.6% of the title (101 MHz, CDCl3): δ 190.6, 141.7, 135.4, 135.0, 131.6, 131.2, 121.1, product as a yellow oil. Rf = 0.43 (EA/PE, 1:10). 1H NMR (400 3841 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article MHz, CDCl3): δ 9.74 (s, 1H), 7.98 (d, 1H, J = 2.0 Hz), 7.65 (dd, 1H, 4-(Diethylamino)-3-methoxybenzaldehyde (20). Starting re- J = 8.3, 2.0 Hz), 7.02 (d, 1H, J = 8.3 Hz), 3.20 (q, 1H, J = 7.1 Hz), agents were 4-fluoro-3-methoxybenzaldehyde and diethylamine, 1.03 (t, 1H, J = 7.1 Hz). 13C NMR (101 MHz, CDCl3): δ 189.8, processed at 100 °C for 5 days. The crude compound was purified 155.1, 136.0, 131.2, 129.2, 122.2, 119.2, 46.0, 12.4. HRMS (ESI): by column chromatography starting with EA/PE (0.1:10) and calcd for C11H14BrNO [M + H]+, 256.0332; found, 256.0333, gradually increasing to EA/PE (0.5:10) to yield 60.4% of the title calculated for C11H14BrNO [M + H+2]+, 258.0311; found, 258.0309. compound as a yellow oil. Rf = 0.26 (EA/PE, 1:10). 1H NMR (400 3-Bromo-4-(dipropylamino)benzaldehyde (14). Starting re- MHz, CDCl3): δ 9.70 (s, 1H), 7.30 (dd, 1H, J = 7.9, 1.5 Hz), 7.27 (d, agents were 3-bromo-4-fluorobenzaldehyde and dipropylamine, 1H, J = 1.5 Hz), 6.80 (d, 1H, J = 7.9 Hz), 3.81 (s, 3H), 3.26 (q, 4H, J processed at 80 °C for 24 h. The crude was purified by column = 7.0 Hz), 1.05 (t, 6H, J = 7.1 Hz). 13C NMR (101 MHz, CDCl3): δ chromatography using EA/PE (0.05:10) to yield 48.5% of the title 190.6, 151.7, 146.1, 128.9, 126.6, 117.3, 109.6, 55.6, 45.7, 12.7. product as a yellow oil. Rf = 0.45 (EA/PE, 1:10). 1H NMR (400 HRMS (ESI): calcd for C12H17NO2 [M + H]+, 208.1332; found, MHz, CDCl3): δ 9.73 (s, 1H), 7.97 (d, 1H, J = 1.9 Hz), 7.64 (dd, 1H, 208.1330. J = 8.3, 1.9 Hz), 7.02 (d, 1H, J = 8.3 Hz), 3.17−3.04 (m, 4H), 1.54− 3-Bromo-4-(4-methylpiperazin-1-yl)benzaldehyde (21). 1.38 (m, 4H), 0.79 (t, 6H, J = 7.4 Hz). 13C NMR (101 MHz, CDCl3): Starting reagents were 3-bromo-4-fluorobenzaldehyde and N-methyl- δ 189.8, 155.4, 136.2, 131.0, 129.2, 122.2, 118.8, 54.0, 20.4, 11.5. piperazine, processed at 100 °C for 24 h. The crude was purified by HRMS (ESI): calcd for C13H18BrNO [M + H]+, 284.0645; found, column chromatography starting with CH2Cl2 and followed by a 284.0645, calculated for C13H18BrNO [M + H+2]+, 286.0624; found, gradual increase of polarity to (CH3OH/CH2Cl2, 0.5:10) to yield 286.0620. 93.0% of the title product as a yellow oil, which was solidified under 3-Methyl-4-(piperidin-1-yl)benzaldehyde (15). Starting re- vacuum. Rf = 0.13 (CH3OH/CHCl2, 0.5:10). mp 55 ± 1 °C. 1H agents were 4-fluoro-3-methylbenzaldehyde and piperidine, processed NMR (400 MHz, CDCl3): δ 9.74 (s, 1H), 7.95 (d, 1H, J = 1.9 Hz), at 100 °C for 24 h. The crude was purified by column 7.67 (dd, 1H, J = 8.3, 1.9 Hz), 7.02 (d, 1H, J = 8.3 Hz), 3.21−3.10 chromatography using EA/PE (0.1:10) to yield 64.7% of the title (m, 4H), 2.61−2.49 (m, 4H), 2.29 (s, 3H). 13C NMR (101 MHz, product as a yellow oil. Rf = 0.46 (EA/PE, 1:10). 1H NMR (400 CDCl3): δ 189.6, 155.8, 135.3, 131.9, 130.2, 120.4, 118.7, 54.8, 51.0, MHz, CDCl3): δ 9.77 (s, 1H), 7.60 (s, 1H) 7.58−7.53 (m, 1H), 6.94 46.0. HRMS (ESI): calcd for C12H15BrN2O [M + H]+, 283.0441; (d, 1H, J = 8.1 Hz), 2.95−2.78 (m, 4H), 2.25 (s, 3H), 1.71−1.59 (m, found, 283.0443. [M+2 + H]+, 285.0420; found, 285.0419. 4H), 1.53 (dt, 2H, J = 10.9, 5.6 Hz). 13C NMR (101 MHz, CDCl3): δ 3-Bromo-4-morpholinobenzaldehyde (22). Starting reagents 191.5, 158.6, 132.5, 132.0, 130.4, 129.3, 118.5, 52.5, 26.3, 24.3, 18.6. were 3-bromo-4-fluorobenzaldehyde and morpholine, processed at HRMS (ESI): calcd for C13H17NO [M + H]+, 204.1383; found, 100 °C for 48 h. The crude was purified by column chromatography 204.1381. using CH2Cl2 to yield 71.5% of the title product as a yellow solid. Rf = 4-Isopropoxybenzaldehyde (16). Starting reagents were 4- 0.20 (Ch3OH/CH2Cl2, 0.5:10). mp 96 ± 1 °C. 1H NMR (400 MHz, hydroxybenzaldehyde and 2-iodopropane. The crude was purified by CDCl3): δ 9.78 (s, 1H), 8.00 (d, 1H, J = 1.8 Hz), 7.71 (dd, 1H, J = column chromatography using EA/PE (0.1:10) to yield quantitatively 8.2, 1.9 Hz), 7.03 (d, 1H, J = 8.3 Hz), 3.93−3.71 (m, 4H), 3.18−3.02 the title compound as a pale-yellow oil. Rf = 0.38 (EA/PE, 1:10). 1H (m, 4H). 13C NMR (101 MHz, CDCl3): δ 189.8, 155.6, 135.5, 132.3, NMR (400 MHz, CDCl3): δ 9.87 (s, 1H), 7.95−7.68 (m, 2H), 7.09− 130.2, 120.4, 118.9, 66.8, 51.5. HRMS (ESI): calcd for C11H12BrNO2 6.84 (m, 2H), 4.68 (Sep, 1H, J = 6.1 Hz), 1.38 (d, 6H, J = 6.1 Hz). [M + H]+, 270.0124; found, 270.0126. [M+2 + H]+, 272.0104; found, 13 C NMR (101 MHz, CDCl3): δ 190.8, 163.2, 132.1, 129.5, 115.6, 272.0103. 70.3, 21.9. HRMS (ESI): calcd for C10H12O2 [M + H]+, 165.0910; 3-Bromo-4-(piperidin-1-yl)benzaldehyde (23). Starting re- found, 165.0906. agents were 3-bromo-4-fluorobenzaldehyde and piperidine, processed 4-(Dipropylamino)-3-methoxybenzaldehyde (17). Starting at 100 °C for 24 h. The crude was purified by column reagents were 4-fluoro-3-methoxybenzaldehyde and dipropylamine, chromatography using EA/PE (0.1:10) to yield 84.6% of the title processed at 100 °C for 5 days. The crude was purified by column product as a yellow oil. Rf = 0.33 (EA/PE, 1:10). 1H NMR (400 chromatography using EA/PE (0.1:10) to yield 65.9% of the title MHz, CDCl3): δ 9.81 (s, 1H), 8.03 (d, 1H, J = 1.9 Hz), 7.73 (dd, 1H, compound as a yellow oil. Rf = 0.12 (EA/PE, 0.5:10). 1H NMR (400 J = 8.3, 1.9 Hz), 7.07 (d, 1H, J = 8.3 Hz), 3.18−2.99 (m, 4H), 1.85− MHz, CDCl3): δ 9.68 (s, 1H), 7.28 (dd, 1H, J = 8.7, 1.7 Hz), 7.26 (d, 1.68 (m, 4H), 1.62 (m, 2H). 13C NMR (101 MHz, CDCl3): δ 189.8, 1H, J = 1.7 Hz), 6.75 (d, 1H, J = 8.7 Hz), 3.80 (s, 3H), 3.25−3.09 (m, 157.1, 135.4, 131.5, 130.1, 120.4, 118.9, 52.7, 25.9, 24.0. HRMS 4H), 1.59−1.40 (m, 4H), 0.80 (t, 6H, J = 7.4 Hz). 13C NMR (101 (ESI): calcd for C12H14BrNO [M + H]+, 268.0332; found, 268.0333. MHz, CDCl3): δ 190.5, 151.4, 146.4, 128.5, 126.7, 116.9, 109.9, 55.6, [M+2 + H]+, 270.0311; found, 270.0310. 54.2, 20.8, 11.5. HRMS (ESI): calcd for C14H21NO2 [M + H]+, 3-Methyl-4-(4-methylpiperazin-1-yl)benzaldehyde (24). 236.1645; found, 236.1640. Starting reagents were 4-fluoro-3-methylbenzaldehyde and N- 4-(Dipropylamino)-3-Nitrobenzaldehyde (18). Starting re- methylpiperazine, processed at 100 °C for 6 days. The crude was agents were 4-fluoro-3-nitrobenzaldehyde and dipropylamine, pro- purified by column chromatography starting with CH2Cl2 and cessed at 25 °C for 1 h. The crude was purified by column gradually increasing polarity to CH3OH/CH2Cl2 (0.5:10) to yield chromatography using CH2Cl2 to yield 96% of the title product as a 71.4% of the title product as a pale yellow solid. Rf = 0.420 (CH3OH/ dark yellow solid. Rf = 0.44 (EA/PE, 2:10). mp 42 ± 0.5 °C. 1H NMR CH2Cl2, 1:10). mp 68 ± 0.5 °C. 1H NMR (400 MHz, CDCl3): δ 9.82 (400 MHz, CDCl3): δ 9.78 (s, 1H), 8.18 (d, 1H, J = 2.0 Hz), 7.84 (s, 1H), 7.63 (s, 1H), 7.60 (m, 1H), 7.02 (d, 1H, J = 8.0 Hz), 2.98 (t, (dd, 1H, J = 8.9, 2.0 Hz), 7.12 (d, 1H, J = 8.9 Hz), 3.28−3.12 (m, 4H, J = 4.78 Hz), 2.56 (s, 4H), 2.33 (s, 3H), 2.30 (s, 3H). 13C NMR 4H), 1.71−1.52 (m, 4H), 0.87 (t, 6H, J = 7.4 Hz). 13C NMR (101 (101 MHz, CDCl3): δ 191.32, 157.19, 132.43, 131.97, 130.90, 129.31, MHz, CDCl3): δ 188.7, 148.9, 139.0, 132.2, 130.4, 125.8, 119.7, 53.7, 118.54, 55.21, 50.92, 46.06, 18.52. HRMS (ESI): calcd for 20.7, 11.2. HRMS (ESI): calcd for C13H18N2O3 [M + H]+, 251.1390; C13H18N2O [M + H]+, 219.1495; found, 219.1497. found, 251.1388. 4-(4-Methylpiperazin-1-yl)-3-nitrobenzaldehyde (25). Start- 4-(Diethylamino)-3-Nitrobenzaldehyde (19). Starting reagents ing reagents were 4-fluoro-3-nitrobenzaldehyde and N-methylpiper- were 4-fluoro-3-nitrobenzaldehyde and dirthylamine, processed at 25 azine, processed at 25 °C for 1 h. The crude was purified by column °C for 1 h. The crude was purified by column chromatography using chromatography using CH2Cl2 followed by (CH3OH/CH2Cl2, CH2Cl2 to yield 91.7% of the title product as a dark yellow solid. Rf = 0.1:10) to yield quantitatively the title compound as a yellow solid. 0.29 (EA/PE, 2:10). mp 50 ± 0.5 °C. 1H NMR (400 MHz, CDCl3): Rf = 0.24 (CH3OH/CH2Cl2, 0.5:10). mp 101 ± 1 °C. 1H NMR (400 δ 9.77 (s, 1H), 8.14 (d, 1H, J = 2.0 Hz), 7.83 (dd, 1H, J = 8.9, 2.0 MHz, CDCl3): δ 9.82 (s, 1H), 8.23 (d, 1H, J = 2.0 Hz), 7.90 (dd, 1H, Hz), 7.10 (d, 1H, J = 8.9 Hz), 3.32 (q, 4H, J = 7.1 Hz), 1.20 (t, 6H, J J = 8.7, 2.0 Hz), 7.13 (d, 1H, J = 8.7 Hz), 3.32−3.16 (m, 4H), 2.64− = 7.1 Hz). 13C NMR (101 MHz, CDCl3): δ 188.7, 148.1, 138.9, 2.47 (m, 4H), 2.34 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 188.7, 132.3, 130.2, 125.8, 119.2, 46.0, 12.4. HRMS (ESI): calcd for 149.4, 139.8, 133.2, 129.7, 127.7, 119.9, 54.4, 50.5, 45.9. HRMS C11H14N2O3 [M + H]+, 223.1077; found, 223.1074. (ESI): calcd for C12H15N3O3 [M + H]+, 250.1191; found, 250.1192. 3842 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article 3-Methyl-4-(pyrrolidin-1-yl)benzaldehyde (26). Starting re- 4-Morpholino-3-Nitrobenzaldehyde (32). Starting reagents agents were 4-fluoro-3-methylbenzaldehyde and pyrrolidine, pro- were 4-fluoro-3-nitrobenzaldehyde and morpholine, processed at 25 cessed at 100 °C for 48 h. The crude was purified by column °C for 1 h. The crude was purified by column chromatography using chromatography using first EA/PE (0.1:10) before increasing to EA/ CH2Cl2 and gradual increase to CH3OH/CH2Cl2 (0.1:10) to yield PE (0.2:10) to yield 70.4% of the title compound as a dark yellow oil. quantitatively the title compound as a dark yellow oil. Rf = 0.35 (EA/ Rf = 0.10 (EA/PE, 0.5:10). 1H NMR (400 MHz, CDCl3): δ 9.64 (s, PE, 4:10). 1H NMR (400 MHz, Acetone): δ 9.93 (s, 1H), 8.31 (d, 1H), 7.51−7.48 (m, 1H), 7.48−7.45 (m, 1H), 6.64 (d, 1H, J = 9.0 1H, J = 1.9 Hz), 8.03 (dd, 1H, J = 8.7, 1.9 Hz), 7.42 (d, 1H, J = 8.7 Hz), 3.44−3.28 (m, 4H), 2.34 (s, 3H), 1.94−1.80 (m, 4H). 13C NMR Hz), 3.80 (t, 4H, J = 4.7 Hz), 3.26 (t, 4H, J = 4.7 Hz). 13C NMR (101 (101 MHz, CDCl3): δ 191.5, 154.4, 134.1, 129.9 127.1, 124.7, 113.8, MHz, CDCl3): δ 189.9, 150.0, 141.1, 134.0, 129.7, 129.2, 121.1, 66.9, 51.0, 25.7, 22.0. HRMS (ESI): calcd for C12H15NO [M + H]+, 51.6. HRMS (ESI): calcd for C11H12N2O4 [M + H]+, 237.0875; 190.1226; found, 190.1221. found, 237.0879. 3-Methyl-4-morpholinobenzaldehyde (27). Starting reagents 4-Isopropoxy-3-methoxybenzaldehyde (33). Starting reagents were 4-fluoro-3-methylbenzaldehyde and morpholine, at 100 °C for 6 were 4-hydroxy-3-methoxybenzaldehyde and 2-iodopropane. The days. The crude was purified by column chromatography using first crude was purified by column chromatography using CH2Cl2 to EA/PE (0.5:10) followed by EA/PE (1:10) to yield 45.9% of the title yield quantitatively the title compound as a yellow oil. Rf = 0.18 product as a yellow solid. Rf = 0.08 (EA/PE, 1:10). mp 67 ± 1 °C. 1H (CH2Cl2). 1H NMR (400 MHz, Acetone): δ 9.86 (s, 1H), 7.50 (dd, NMR (400 MHz, CDCl3): δ 9.81 (s, 1H), 7.62−7.59 (m, 1H), 7.58− 1H, J = 8.2, 1.9 Hz), 7.43 (d, 1H, J = 1.9 Hz), 7.12 (d, 1H, J = 8.2 7.62 (m, 1H), 7.00 (d, 1H, J = 7.9 Hz), 3.91−3.64 (m, 4H), 3.02− Hz), 4.75 (sept, 1H, J = 6.1 Hz), 3.89 (s, 3H), 1.36 (d, 6H, J = 6.1 2.82 (m, 4H), 2.29 (s, 3H). 13C NMR (101 MHz, CDCl3): δ 191.4, Hz). 13C NMR (101 MHz, Acetone): δ 191.2, 154.0, 151.5, 131.1, 156.9, 132.6, 132.2, 131.3, 129.4, 118.5, 67.1, 51.6, 18.5. HRMS 126.6, 114.4, 111.0, 71.7, 56.1, 22.2. HRMS (ESI): calcd for (ESI): calcd for C12H15NO2 [M + H]+, 206.1176; found, 206.1173. C11H14O3 [M + H]+, 195.1016; found, 195.1013. 3-Methoxy-4-(piperidin-1-yl)benzaldehyde (28). Starting re- 3-Nitro-4-(pyrrolidin-1-yl)benzaldehyde (34). Starting re- agents were 4-fluoro-3-methoxybenzaldehyde and piperidine, pro- agents were 4-fluoro-3-nitrobenzaldehyde and pyrrolidine, processed cessed at 100 °C for 48 h. The crude was purified by column at 25 °C for 1 h. The crude was purified by column chromatography chromatography using first EA/PE (0.1:10) followed with gradual using CH2Cl2 to yield quantitatively the title compound as a yellow increase up to EA/PE (2:10) to yield 70.4% of the title compound as solid. Rf = 0.20 (EA/PE, 2:10). mp 118 ± 1 °C. 1H NMR (400 MHz, a yellow oil. Rf = 0.50 (EA/PE, 2:10). 1H NMR (400 MHz, CDCl3): Acetone): δ 9.81 (s, 1H), 8.19 (d, 1H, J = 2.0 Hz), 7.87 (dd, 1H, J = 8.9, 2.0 Hz), 7.13 (d, 1H, J = 8.9 Hz), 3.33 (t, 4H, J = 6.5 Hz), 2.05 (t, δ 9.75 (s, 1H), 7.32 (dd, 1H, J = 8.1, 1.8 Hz), 7.28 (d, 1H, J = 1.8 4H, J = 6.5 Hz). 13C NMR (101 MHz, acetone): δ 189.3, 146.5, Hz), 6.89 (d, 1H, J = 8.1 Hz), 3.84 (s, 3H), 3.11−2.96 (m, 4H), 137.3, 132.6, 130.7, 125.4, 117.5, 51.5, 26.2. HRMS (ESI): calcd for 1.74−1.60 (m, 4H), 1.53 (m, 2H). 13C NMR (101 MHz, CDCl3): δ C11H12N2O3 [M + H]+, 221.0926; found, 221.0932. 191.0, 152.2, 148.6, 130.4, 126.6, 117.4, 109.1, 55.6, 51.5, 26.0, 24.3. 4-Isopropoxy-3-methoxy-5-Nitrobenzaldehyde (35). Starting HRMS (ESI): calcd for C13H17NO2 [M + H]+, 220.1332; found, reagents were 4-hydroxy-3-methoxy-5-nitrobenzaldehyde and 2- 220.1330. iodopropane. The crude was purified by column chromatography 3-Methoxy-4-(pyrrolidin-1-yl)benzaldehyde (29). Starting using EA/PE (0.5:10) with a gradual increase to EA/PE (3:10) to reagents were 4-fluoro-3-methoxybenzaldehyde and pyrrolidine, yield 25.8% of the title compound as a pale yellow solid. Rf = 0.41 processed at 100 °C for 48 h. The crude was purified by column (EA/PE, 3:10). mp 71 ± 0.5 °C. 1H NMR (400 MHz, Acetone): δ chromatography using first EA/PE (0.1:10) and then EA/PE (0.5:10) 9.85 (s, 1H), 7.76 (d, 1H, J = 1.8 Hz), 7.63 (d, 1H, J = 1.8 Hz), 4.88− to yield 57.9% of the title compound as a yellow solid. Rf = 0.25 (EA/ 4.76 (sept, 1H, J = 6.1 Hz), 3.93 (s, 3H), 1.15 (d, 6H, J = 6.1 Hz). 13C PE, 0.1:10). mp 38 ± 0.5 °C. 1H NMR (400 MHz, CDCl3): δ 9.60 (s, NMR (101 MHz, Acetone): δ 190.4, 155.3, 146.9, 145.5, 132.6, 1H), 7.23 (dd, 1H, J = 8.0, 1.7 Hz), 7.20 (d, 1H, J = 1.8 Hz), 6.50 (d, 118.7, 114.8, 78.1, 57.2, 22.6. HRMS (ESI): calcd for C11H13NO5 [M 1H, J = 8.0 Hz), 3.74 (s, 3H), 3.44 (t, 4H, J = 6.7 Hz), 1.84 (t, 4H, J = + H]+, 240.0872; found, 240.0869. 6.7 Hz). 13C NMR (101 MHz, CDCl3): δ 190.0, 148.7, 145.1, 128.1, 3-Nitro-4-(piperidin-1-yl)benzaldehyde (36). Starting reagents 126.4, 112.6, 109.7, 55.7, 50.6, 25.5. HRMS (ESI): calcd for were 4-fluoro-3-nitrobenzaldehyde and piperidine, processed at 25 °C C12H15NO2 [M + H]+, 206.1176; found, 206.1174. for 1 h. The crude was purified by column chromatography using 3-Methoxy-4-(4-methylpiperazin-1-yl)benzaldehyde (30). CH2Cl2 to yield quantitatively the title compound as a yellow oil. Rf = Starting reagents were 4-fluoro-3-methoxybenzaldehyde and N- 0.31 (EA/PE, 2:10). 1H NMR (400 MHz, acetone): δ 9.75 (s, 1H), methylpiperazine, processed at 100 °C for 48 h. The crude was 8.12 (d, 1H, J = 2.0 Hz), 7.82 (dd, 1H, J = 8.7, 2.0 Hz), 7.23 (d, 1H, J purified by column chromatography using first CH3OH/CH2Cl2 = 8.7 Hz), 3.14−3.07 (m, 4H), 1.67−1.44 (m, 6H). 13C NMR (101 (0.1:10) with gradual increase of CH3OH to CH3OH/CH2Cl2 MHz, acetone): δ 189.7, 150.6, 140.7, 133.6, 130.0, 128.1, 121.1, 52.4, (0.5:10) to yield 66.9% of the title product as a yellow solid. Rf = 26.4, 24.4. HRMS (ESI): calcd for C12H14N2O3 [M + H]+, 235.1083; 0.28 (CH3OH/CH2Cl2, 0.5:10). mp 78 ± 1 °C. 1H NMR (400 MHz, found, 235.1084. Acetone): δ 9.85 (s, 1H), 7.47 (dd, 1H, J = 8.1, 1.7 Hz), 7.40 (d, 1H, 4-(Diethylamino)-2-methoxybenzaldehyde (37). Starting re- J = 1.7 Hz), 7.03 (d, 1H, J = 8.1 Hz), 3.93 (s, 3H), 3.21 (t, 4H, J = 4.7 agents were with 4-fluoro-2-methoxybenzaldehyde and diethylamine, Hz), 2.51 (t, 4H, J = 4.7 Hz), 2.27 (s, 3H). 13C NMR (101 MHz, processed at 100 °C for 48 h. The crude was purified by column acetone): δ 191.2, 153.1, 148.3, 131.7, 126.5, 118.1, 110.8, 56.0, 55.9, chromatography using EA/PE (1:10) to yield 59.4% of the title 50.6, 46.4. HRMS (ESI): calcd for C13H18N2O2 [M + H]+, 235.1441; compound as yellow solid. Rf = 0.20 (EA/PE, 2:10). mp 101 ± 1 °C. found, 235.1439. 1 H NMR (400 MHz, Acetone): δ 9.98 (s, 1H), 7.45 (d, 1H, J = 8.9 3-Methoxy-4-morpholinobenzaldehyde (31). Starting re- Hz), 6.23 (dd, 1H, J = 8.9, 2.1 Hz), 6.10 (d, 1H, J = 2.1 Hz), 3.78 (s, agents were 4-fluoro-3-methoxybenzaldehyde and morpholine, 3H), 3.37 (q, 4H, J = 7.1 Hz), 1.07 (t, 6H, J = 7.1 Hz). 13C NMR processed at 100 °C for 48 h. The crude was purified by column (101 MHz, acetone) δ 186.0, 165.0, 154.8, 130.3, 115.0, 105.1, 93.6, chromatography using first EA/PE (0.5:10) then gradually increase of 55.7, 45.2, 12.8. HRMS (ESI): calcd for C12H17NO2 [M + H]+, EA to EA/PE (2:10) to yield 68.4% of the title compound as a pale- 208.1332; found, 208.1329. yellow solid. Rf = 0.59 (EA/PE, 1:1). mp 96 ± 1 °C. 1H NMR (400 4-(Dipropylamino)-2-methoxybenzaldehyde (38). Starting MHz, acetone): δ 9.72 (s, 1H), 7.35 (dd, 1H, J = 8.1, 1.8 Hz), 7.27 reagents were 4-fluoro-2-methoxybenzaldehyde and dipropylamine, (d, 1H, J = 1.8 Hz), 6.90 (d, 1H, J = 8.1 Hz), 3.79 (s, 3H), 3.62 (t, processed at 100 °C for 48 h. The crude was purified by column 4H, J = 4.6 Hz), 3.13 (t, 4H, J = 4.6 Hz). 13C NMR (101 MHz, chromatography using EA/PE (1:10) to yield 43.7% of the title acetone): δ 191.3, 153.1, 148.0, 132.0, 126.4, 118.0, 110.8, 67.4, 56.0, compound as yellow solid. Rf = 0.38 (EA/PE, 2:10). mp 92 ± 1 °C. 51.2. HRMS (ESI): calcd for C12H15NO3 [M + H]+, 222.1125; found, 1 H NMR (400 MHz, acetone) δ 9.97 (s, 1H), 7.44 (d, 1H, J = 8.9 222.1123. Hz), 6.22 (dd, 1H, J = 8.9, 2.1 Hz), 6.08 (d, 1H, J = 2.1 Hz), 3.77 (s, 3843 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article 3H), 3.28 (t, 4H, J = 7.6 Hz), 1.62−1.43 (sextet, 4H, J = 7.5 Hz), 0.81 lack of sensitivity when its low substrate concentration was used in the (t, 6H, J = 7.4 Hz). 13C NMR (101 MHz, acetone) δ 186.0, 164.9, assay. For ALDH1A1, 5 μM hexanal (Sigma) was used. For 155.3, 130.3, 115.0, 105.3, 93.8, 55.7, 53.2, 21.2, 11.5. HRMS (ESI): ALDH1A3, 10 and 250 μM hexanal (Sigma) were used. For calcd for C14H21NO2 [M + H]+, 236.1645; found, 236.1643. ALDH3A1, 4-NBA (Sigma) was used at 31 and 250 μM. All tert-Butyl 4-(4-formyl-2-nitrophenyl)piperazine-1-carboxy- substrates were prepared in the corresponding assay buffer at a late (39). To a solution of 4-fluoro-3-nitrobenzaldehyde (1 equiv) in concentration of 2 mM and further diluted to reach the final acetone (20 mL), pyridine (1 mL) and 1-Boc-piperazine (1.2 equiv) concentrations required per experiment. The concentration of the were added and stirred at room temperature for 4 h. The solvent was enzyme was kept from 50- to 100-fold lower than that of the substrate then evaporated under vacuum to yield a dark-yellow oil, which was for all enzymatic assays. purified by column chromatography using EA/PE (1:10) to afford Enzyme Kinetics with DEAB Analogues as Substrates. 39.4% of the title compound as a yellow solid. Rf = 0.19 (EA/PE, According to their structure with a carbonyl group, compounds 2:10). mp 141 ± 1 °C. 1H NMR (400 MHz, acetone): δ 9.94 (s, 1H), were tested for their substrate properties against ALDH1A1, 8.33 (d, 1H, J = 2.0 Hz), 8.03 (dd, 1H, J = 8.9, 2.0 Hz), 7.44 (d, 1H, J ALDH1A3, and ALDH3A1. For this assay, the conditions were the = 8.9 Hz), 3.76−3.43 (m, 4H), 3.43−3.15 (m, 4H), 1.48 (s, 9H). 13C same as previously described except using the DEAB analogues as NMR (101 MHz, acetone): δ 189.91, 155.04, 150.11, 141.03, 133.88, substrates instead of hexanal or 4-NBA at a concentration of 10 μM. 129.70, 129.18, 121.35, 80.08, 51.03, 44.23, 28.52. HRMS (ESI): Results are expressed as the percentage of activity at 10 μM versus the calcd for C16H21N3O5 [M + H]+, 336.1559; found, 336.1560. activity at 10 μM of their standard substrate (hexanal or 4-NBA). 3-Nitro-4-(piperazin-1-yl)benzaldehyde (40). To a solution of Values are expressed as the mean ± SE. 39 (1 equiv) in CH2Cl2 (15 mL), trifluoroacetic acid (5 mL) was To calculate the Km value of compound 18 as a substrate for added and stirred at room temperature for 2 h. CH2Cl2 was then ALDH3A1, several concentrations of compound 18 were used. evaporated under vacuum to give a dark-yellow oil, which was purified Experimental values were fitted to the adaptation of the Michaelis− by column chromatography using (CH3OH/CH2Cl2, 0.5:10) to yield i 2y K m + [S ] + jjjj [S ] z vmax ·[S ] zzz quantitatively the title compound as a yellow-orange solid (82%). Rf = Menten equation for substrate inhibition v = and 0.38 (CH3OH/CH2Cl2, 1:10). mp 144 ± 1 °C. 1H NMR (400 MHz, k Ki { acetone) δ 9.98 (s, 1H), 8.38 (d, 1H, J = 1.9 Hz), 8.10 (dd, 1H, J = shown as the mean ± SE. 8.6, 1.9 Hz), 7.56 (d, 1H, J = 8.6 Hz), 3.71−3.61 (m, 4H), 3.61−3.50 Determination of the Kinetic Constants (IC50 and Ki). In (m, 4H), 2.10 (s, 1H). 13C NMR (101 MHz, acetone) δ 190.19, order to determine the IC50 values, reaction rates were determined at 149.65, 141.82, 134.47, 130.55, 129.34, 122.25, 48.60, 43.97. HRMS various concentrations of inhibitor at a fixed concentration of (ESI): calcd for C11H13N3O3 [M + H]+, 236.1030; found, 236.1031. substrate. As substrates for ALDH1A1, ALDH1A3, and ALDH3A1, Compound 40 was obtained in 82% purity. 5 μM hexanal, 6 μM hexanal, and 31 μM 4-NBA, respectively, were Purification of Recombinant Human ALDHs and Enzymatic used. The IC50 values were calculated by nonlinear fitting of the Assays. Human ALDH1A1, ALDH1A3, and ALDH3A1 were cloned obtained data to a sigmoidal plot using GraFit 5.0 (Erithacus and recombinantly expressed from the pET-30 Xa/LIC vector. software), with the following 4-parameter equation i y range 1 + jjj x z Protein purification was achieved by affinity chromatography on a zz y= s + background , where y is the specific activity, x is the k IC50 { nickel-charged chelating Sepharose Fast Flow 5-mL column (His Trap column, Cytiva), which specifically binds the protein due to its inhibitor concentration, background is the minimum y value, range is N-terminal (His)6 tag, using an Ä KTA FPLC system (Cytiva), as the fitted uninhibited value minus the background, and s is the slope previously described.55 Enzymes including the His tag were stored at factor. Values are expressed as the mean ± SE. −80 °C in 20 mM Tris/HCl and 0.5 M NaCl, pH 8.0, until use. Activity assays to determine the type of inhibition and Ki value Reaction buffers were as follows: ALDH1A1 was assayed in 50 mM were performed using various substrate and inhibitor concentrations HEPES, 0.5 mM EDTA, and 0.5 mM DTT, pH 8.0; ALDH1A3 was maintaining the same conditions as for the IC50 experiments and using assayed in 50 mM HEPES, 30 mM MgCl2, and 5 mM DTT, pH 8.0; GraFit 5.0 for data processing. The data of enzymatic activities at and ALDH3A1 was assayed in 50 mM Tris−HCl and 5 mM DTT, pH 8.0. Activity under standard conditions was measured fluorimetri- different inhibitor concentrations were fitted to the Michaelis− cally at 25 °C using a Cary Eclipse (Varian) fluorimeter to follow the Menten equation to determine the values of Km and Vmax. Next, results were fitted to the equations for competitive, given by i y Vmax·[S ] / jjj1 + [I ] z z purification procedure and to check the enzyme concentration before each kinetic experiment. The fluorescence of NADH was charac- k Ki z { i [I ] y [S ] + K m·jjj1 + zz Vmax·[S ] Ki z terized at 460 nm with excitation at 340 nm, and 5 μM NAD(P)H v= ; noncompetitive, given by v = ; k { K m + [S ] was added as an internal standard to obtain the absolute reaction i [I ] y rates.55 Standard activity was measured at saturating concentrations of K m + [S ]·jjj1 + zz Vmax·[S ] Ki z uncompetitive, given by v = ; and mixed, given by k { substrate using 30 μM hexanal (ALDH1A1), 250 μM hexanal (ALDH1A3), or 250 μM 4-NBA (ALDH3A1). NAD+ was 500 μM i [I ] y zz + ijjj1 + [I ] yzzz[S ] K m·jjj1 + Vmax·[S ] .56 The type of inhibition was selected based Ki z for ALDH1A1 and ALDH1A3, and NADP+ was 1 mM for v= ALDH3A1. k { k K i′ { Inhibition Screening. All compounds tested were dissolved in on the lowest error value. Parameters were expressed as the mean ± DMSO and assayed at a final concentration of 1% (v/v) DMSO. SE. Single-point measurements of enzymatic activity at 10 μM inhibitor Docking studies of compounds 14 and 18 on ALDH1A3 and were performed for the 40 DEAB analogues against the three isoforms ALDH3A1 isoforms. (ALDH1A1, ALDH1A3, and ALDH3A1). For the initial screening, (a) Protein and ligand preparation: The three X-ray complexes the enzymatic activity was measured in 96-well plates (final volume of ALDH1A3, ALDH3A1, and ALDH1A1 were downloaded from the 200 μL) in a Victor 3 Multilabel Plate Reader (Perkin Elmer), by Protein Data Bank with their PBD IDs 5FHZ,57 4H80,58 and monitoring the fluorescence of the NAD(P)H produced during the 4WPN,59 respectively. The cocrystallized ligands, ions, and water reaction (excitation at 340 nm and emission at 460 nm). Alternatively, molecules were removed from the X-ray complexes and H-bonds; the enzymatic activity was measured in a Varian Cary 400 UV−vis missing residues were added to the protein with the aid of protein spectrophotometer, by monitoring the increase in the absorbance of preparation wizard of Maestro. All the compounds were drawn using NAD(P)H at 340 nm (ε = 6.22 mM−1·cm−1) or in a Cary Eclipse the Build panel of Maestro and subjected to a conformational search Varian fluorimeter, as described above. The inhibition screening was of 1000 steps in a water environment (using the generalized-Born/ preferably performed at two substrate concentrations (near the Km surface-area model) through Macromodel software. A Monte Carlo value and at a saturating substrate concentration), except for algorithm with the MMFF and a distance-dependent dielectric ALDH1A1, which was only tested at substrate saturation due to a constant of 1.0 was applied while using Macromodel.60 3844 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article (b) Molecular docking: All docking calculations were carried out CO2 for 24 h. DEAB, 14, and 18 were prepared in DMSO at a stock on the X-ray structure of human ALDH1A1 in complex with selective concentration of 200 mM. Cells were treated with 100 μL of ALDH inhibitor 1-{[1,3-dimethyl-7-(3-methylbutyl)-2,6-dioxo-2,3,6,7-tetra- inhibitors at concentrations of 50 and 200 μM as single treatment. hydro-1H-purin-8-yl]methyl}piperidine-4-carboxamide (PDB ID: Combination treatments included 100 μL of 50 μM ALDH inhibitor 4WPN), X-ray structure of human ALDH1A3 in complex with RA + 1 nM docetaxel and 100 μL of 200 μM ALDH inhibitor + 1 nM and NAD+ (PDB ID: 5FHZ), and X-ray structure of human docetaxel. Cells were also treated with 100 μL of 1 nM docetaxel only. ALDH3A1 in complex with selective inhibitorN-[4-(4-methylsul- 1 nM docetaxel was chosen since it was the IC50 when other primary fonyl-2-nitroanilino)phenyl]acetamide (PDB ID: 4H80). Glide 5.0 samples were analyzed in our laboratory. Control wells included blank with the standard precision (SP) method61 was used for docking of all (media only) and untreated cells (DMSO only). Each experiment was compounds on the three X-ray structures. performed in triplicate, and cell seeding density was the same in all GLIDE 5.0. The binding site was defined by a rectangular box of 10 wells; therefore, changes in treated cells were compared to untreated Å along the x, y, and z axes centered on the ligand. The option of DMSO control cells. Plates were returned to the incubator for a 72 h imposing a maximum value to the number of atoms that a ligand may incubation before further processing. Alamar Blue solution was added have (when docked) was deactivated. Thus, all the ligands were at 10% of total sample volume, and plates were incubated at 37 °C for docked independently from the number of their atoms, whereas the 1−4 h before absorbance was analyzed in a plate reader.64 Alamar GLIDE defaults were used for the remaining parameters. The Blue has an excitation wavelength of 530−560 nm and an emission GlideScore fitness function is based on Chemscore but includes a wavelength of 590 nm. Total % cell viability/cell survival was steric-clash term and adds buried polar terms to penalize electrostatic calculated by dividing the absorbance of the treated sample by the mismatches and modifications on other secondary terms. The docking absorbance of control and multiplying it by 100. analyses were carried out using the SP method. A total of 50 docking solutions were generated for each ligand, and the top-ranked docking pose was considered as the final pose. The reliability of the docking program: GLIDE 5.0 was assessed by ■ ASSOCIATED CONTENT * Supporting Information sı performing self-docking analysis and calculating the root-mean-square The Supporting Information is available free of charge at deviation (RMSD) between the crystallographic position of the ligand https://pubs.acs.org/doi/10.1021/acs.jmedchem.1c01367. and the ligand’s disposition predicted by docking. The rms_analysis program of the Gold suite was used to calculate the RMSD difference, HPLC traces for all compounds. PDB codes 4WPN considering only the heavy atoms of the ligand. The docking method (ALDH1A1), 5FHZ (ALDH1A3) and 4H80 is able to produce a binding pose within 2.0 Å RMSD of the (ALDH3A1) were used for docking studies of crystallographic disposition, therefore considered as reliable.62 compounds 14 and 18, respectively (ZIP) Immunoblotting. The PCa cell lysates (DU145, LNCaP, and Molecular Formula Strings (CSV) PC3) were prepared using the RIPA buffer. The lysates were briefly sonicated and centrifuged, and the protein concentrations were Additional data include results of inhibition screening as determined using the BCA protein assay kit (Thermo Fisher, CA, % of remaining ALDH activities in the presence of USA). A total of 40 μg of cell lysates was separated using 10% SDS- DEAB analogues, % of substrate activities of DEAB polyacrylamide gel electrophoresis and transferred on to Amersham analogues,IC50 representations of 14 with ALDH1A3 Hybond ECL nitrocellulose membranes (Amersham, QC, Canada). and 18 with ALDH3A1, inhibition kinetics curves of The membranes were blocked using 5% skim milk in phosphate- ALDH1A3 by 15 and 16, IC50 representations of 18 buffered saline with 0.1% Tween-20 for 1 h and then incubated with with ALDH3A1 with and without NADP+, docking specific primary antibodies overnight at 4 °C followed by incubation with HRP-conjugated secondary antibodies for 1 h at room studies of compounds 14 and 18 on ALDH1A1, temperature. Specific protein bands were visualized using an ECL ALDH1A3 and ALDH3A1 isoforms and representations detection kit (Amersham). The primary and secondary antibodies of the preliminary two-dose point antiproliferative data used were rabbit anti-ALDH1A1 (D9J7R, Cell signaling Technology, using PC3 cells and the MTT assay (PDF) ■ UK), mouse anti-GAPDH (6C5, Abcam, Cambridge, UK), rabbit anti-ALDH1A3 (N2C2, GeneTex, California, USA), mouse anti- ALDH3A1 (G-2, Santa Cruz Biotechnology, CA), anti-rabbit HRP AUTHOR INFORMATION (Dako), and goat anti-mouse HRP (Abcam, Cambridge, UK). Corresponding Authors Chemosensitivity Studies. MTT (3-(4, 5-dimethyl-2-thiazolyl)- Jaume Farrés − Department of Biochemistry and Molecular 2, 5-diphenyl-2H-tetrazolium bromide) assay was used to determine Biology, Faculty of Biosciences, Universitat Autònoma de cell viability.63 Immortalized PCa cells (PC-3, Du-145, and LNCaP) Barcelona, Barcelona E-08193, Spain; orcid.org/0000- were seeded in 96-well plates and allowed to adhere overnight at 37 °C, 5% CO2, and 100% humidity. The following day, cells were 0001-9069-3987; Email: jaume.farres@uab.cat treated with the DEAB analogues at appropriate concentrations Klaus Pors − Institute of Cancer Therapeutics, School of ranging from 12.5 to 200 μM. After 96 h of exposure, 200 μL of MTT Pharmacy and Medical Sciences, Faculty of Life Sciences, (Sigma) solution (0.5 mg/mL) was added to each well and incubated University of Bradford, Yorkshire BD7 1DP, U.K.; at 37 °C for 4 h. The formazan crystals were dissolved in DMSO orcid.org/0000-0002-0837-5208; Email: k.pors1@ (Sigma), and the absorbance was read using a microplate reader bradford.ac.uk (Multiskan EX; Thermo Fisher Scientific) at 540 nm. Data analysis was performed using Microsoft Excel 2013 and GraphPad Prism Authors software. Ali I. M. Ibrahim − Institute of Cancer Therapeutics, School of Primary cells cultured at early passage from human prostate tissue Pharmacy and Medical Sciences, Faculty of Life Sciences, biopsies included one BPH sample H415/15−BPH/PSA 0.61/age 74 University of Bradford, Yorkshire BD7 1DP, U.K.; Faculty of and four PCa samples H568/15 RM−Gl7(3 + 4)/PSA 8.5/age 69, Pharmacy, Al-Zaytoonah University of Jordan, Amman H431/14 LM−Gl7(3 + 4)/PSA 14/age 66, H488/14 RM−Gl7(3 + 4)/PSA 12.6/age 60, and H517/15 RM−Gl7(3 + 4)/PSA 4.4/age 65. 11733, Jordan Primary cells were obtained with ethical consent (REC ref 07/ Elisabet Batlle − Institute of Cancer Therapeutics, School of H1304/121) at radical prostatectomy (cancer) and transurethral Pharmacy and Medical Sciences, Faculty of Life Sciences, resection (BPH) and were seeded as previously described37 in 96-well University of Bradford, Yorkshire BD7 1DP, U.K.; plates 5000 cells/well in 100 μL of SCM and incubated at 37 °C, 5% Department of Biochemistry and Molecular Biology, Faculty 3845 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article of Biosciences, Universitat Autònoma de Barcelona, Barcelona NMR, carbon-13 nuclear magnetic resonance; CH2Cl2, E-08193, Spain dichloromethane; CH3OH, methanol; CRPC, castrate-resist- Smarakan Sneha − Institute of Cancer Therapeutics, School of ant prostate cancer; CYP, cytochrome p450; DEAB, 4- Pharmacy and Medical Sciences, Faculty of Life Sciences, (diethylamino)benzaldehyde; DMF, dimethylformamide; University of Bradford, Yorkshire BD7 1DP, U.K. DMSO, dimethyl sulfoxide; DPBA, 4-(dipropylamino)- Rafael Jiménez − Department of Biochemistry and Molecular benzaldehyde; EA, ethyl acetate; GLUT1, glucose transporter Biology, Faculty of Biosciences, Universitat Autònoma de 1; 1H NMR, proton nuclear magnetic resonance; HRMS, high Barcelona, Barcelona E-08193, Spain resolution mass spectrometry; Hz, hertz; IC50, half-maximal Raquel Pequerul − Department of Biochemistry and inhibitory concentration; Ki, inhibitory constant; mp, melting Molecular Biology, Faculty of Biosciences, Universitat point; MHz, megahertz; NAD+ , nicotinamide adenine Autònoma de Barcelona, Barcelona E-08193, Spain dinucleotide; NADH, nicotinamide adenine dinucleotide Xavier Parés − Department of Biochemistry and Molecular hydrogen; 4-NBA, 4-nitrobenzaldehyde; PCa, prostate cancer; Biology, Faculty of Biosciences, Universitat Autònoma de PDB, protein data bank; PE, petroleum ether; RA, retinoic Barcelona, Barcelona E-08193, Spain acid; Rf, retention factor; RMSD, root-mean-square deviation; Till Rüngeler − Department of Biochemistry and Molecular SCs, stem cells; CSCs, cancer stem cells; TLC, thin layer Biology, Faculty of Biosciences, Universitat Autònoma de chromatography ■ Barcelona, Barcelona E-08193, Spain Vibhu Jha − Department of Pharmacy, University of Pisa, REFERENCES 56126 Pisa, Italy (1) Ginestier, C.; Wicinski, J.; Cervera, N.; Monville, F.; Finetti, P.; Tiziano Tuccinardi − Department of Pharmacy, University of Bertucci, F.; Wicha, M. S.; Birnbaum, D.; Charafe-Jauffret, E. Retinoid Pisa, 56126 Pisa, Italy; orcid.org/0000-0002-6205-4069 signaling regulates breast cancer stem cell differentiation. Cell Cycle Maria Sadiq − Institute of Cancer Therapeutics, School of 2009, 8, 3297−3302. Pharmacy and Medical Sciences, Faculty of Life Sciences, (2) Jackson, B.; Brocker, C.; Thompson, D. C.; Black, W.; Vasiliou, University of Bradford, Yorkshire BD7 1DP, U.K.; Cancer K.; Nebert, D. W.; Vasiliou, V. Update on the aldehyde dehydrogen- Research Unit, Department of Biology, University of York, ase gene (ALDH) superfamily. Hum. Genom. 2011, 5, 283−303. Yorkshire YO10 5DD, U.K. (3) Kasimanickam, V. Expression of retinoic acid-metabolizing Fiona Frame − Cancer Research Unit, Department of Biology, enzymes, ALDH1A1, ALDH1A2, ALDH1A3, CYP26A1, CYP26B1 University of York, Yorkshire YO10 5DD, U.K. and CYP26C1 in canine testis during post-natal development. Reprod. Norman J. Maitland − Cancer Research Unit, Department of Domest. Anim. 2016, 51, 901−909. Biology, University of York, Yorkshire YO10 5DD, U.K. (4) Moreb, J. S.; Zucali, J. R.; Ostmark, B.; Benson, N. A. Heterogeneity of aldehyde dehydrogenase expression in lung cancer Complete contact information is available at: cell lines is revealed by Aldefluor flow cytometry-based assay. https://pubs.acs.org/10.1021/acs.jmedchem.1c01367 Cytometry, Part B 2007, 72, 281−289. (5) Muzio, G.; Maggiora, M.; Paiuzzi, E.; Oraldi, M.; Canuto, R. A. Author Contributions Aldehyde dehydrogenases and cell proliferation. Free Radic. Biol. Med. A.I.M.I., J.F. and K.P. designed the study. A.I.M.I. performed 2012, 52, 735−746. all chemical experiments, V.J. and T.T. performed computa- (6) Ma, I.; Allan, A. L. The role of human aldehyde dehydrogenase tional studies, E.B., R.J., R.P. and T.R. provided recombinant in normal and cancer stem cells. Stem Cell Rev. 2011, 7, 292−306. ALDH protein and performed biochemical studies, and S.S., (7) Chute, J. P.; Muramoto, G. G.; Whitesides, J.; Colvin, M.; Safi, M.S., F.F. and N.J.M. performed ALDH expression analysis R.; Chao, N. J.; McDonnell, D. P. Inhibition of aldehyde and compound evaluation in PCa cell lines and primary cells. dehydrogenase and retinoid signaling induces the expansion of The first draft of the manuscript was co-written by A.I.M.I., human hematopoietic stem cells. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 11707−11712. E.B., J.F. and K.P. J.F., X.P. and K.P., funding acquisition. All (8) Mao, P.; Joshi, K.; Li, J.; Kim, S.-H.; Li, P.; Santana-Santos, L.; authors have given approval to the final version of the Luthra, S.; Chandran, U. R.; Benos, P. V.; Smith, L.; Wang, M.; Hu, manuscript. All authors have contributed to editorial improve- B.; Cheng, S.-Y.; Sobol, R. W.; Nakano, I. Mesenchymal glioma stem ments of the first draft. cells are maintained by activated glycolytic metabolism involving Funding aldehyde dehydrogenase 1A3. Proc. Natl. Acad. Sci. U.S.A. 2013, 110, This research was funded by Prostate Cancer UK (grant ref. 8644−8649. no.: S12-027), the Spanish Ministerio de Ciencia e Innovación (9) Croker, A. K.; Allan, A. L. Inhibition of aldehyde dehydrogenase (PID2020-119424RB-I00), Al-Zaytoonah University of Jordan (ALDH) activity reduces chemotherapy and radiation resistance of and UoB International Development Fund Scheme (PhD stem-like ALDHhiCD44(+) human breast cancer cells. Breast Cancer studentship to EBR). RJ is a recipient of a PIF predoctoral Res. Treat. 2012, 133, 75−87. (10) Awad, O.; Yustein, J. T.; Shah, P.; Gul, N.; Katuri, V.; O’Neill, fellowship from Universitat Autònoma de Barcelona. A.; Kong, Y.; Brown, M. L.; Toretsky, J. A.; Loeb, D. M. High ALDH Notes activity identifies chemotherapy-resistant Ewing’s sarcoma stem cells The authors declare no competing financial interest. that retain sensitivity to EWS-FLI1 inhibition. PloS One 2010, 5, ■ ACKNOWLEDGMENTS We thank Maria Butjosa and Jan Julià for their assistance in No. e13943. (11) Wang, W.; Zheng, S.; He, H.; Ge, H.; Saeed, B. N,N- diethylaminobenzaldehyde targets aldehyde dehydrogenase to erad- icate human pancreatic cancer cells. Exp. Ther. Med. 2020, 20, 662− performing inhibition screening and kinetic characterizations. ■ 670. (12) Park, J. W.; Jung, K.-H.; Lee, J. H.; Moon, S. H.; Cho, Y. S.; ABBREVIATIONS Lee, K.-H. Inhibition of aldehyde dehydrogenase 1 enhances the ADT, androgen therapy; ALDH, aldehyde dehydrogenase; AR, cytotoxic effect of retinaldehyde on A549 cancer cells. Oncotarget androgen receptor; BPH, benign prostate hyperplasia; 13C 2017, 8, 99382−99393. 3846 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article (13) Jiménez, R.; Pequerul, R.; Amor, A.; Lorenzo, J.; Metwally, K.; (30) Milone, M. R.; Pucci, B.; Bifulco, K.; Iannelli, F.; Lombardi, R.; Avilés, F. X.; Parés, X.; Farrés, J. Inhibitors of aldehyde Ciardiello, C.; Bruzzese, F.; Carriero, M. V.; Budillon, A. Proteomic dehydrogenases of the 1A subfamily as putative anticancer agents: analysis of zoledronic-acid resistant prostate cancer cells unveils novel Kinetic characterization and effect on human cancer cells. Chem. Biol. pathways characterizing an invasive phenotype. Oncotarget 2015, 6, Interact. 2019, 306, 123−130. 5324−5341. (14) Russo, J.; Chung, S.; Contreras, K.; Lian, B.; Lorenz, J.; Stevens, (31) Li, T.; Su, Y.; Mei, Y.; Leng, Q.; Leng, B.; Liu, Z.; Stass, S. A.; D.; Trousdell, W. Identification of 4-(N,N-dipropylamino)- Jiang, F. ALDH1A1 is a marker for malignant prostate stem cells and benzaldehyde as a potent, reversible inhibitor of mouse and human predictor of prostate cancer patients’ outcome. Lab. Invest. 2010, 90, class I aldehyde dehydrogenase. Biochem. Pharmacol. 1995, 50, 399− 234−244. 406. (32) Wang, S.; Zhou, X.; Liang, C.; Bao, M.; Tian, Y.; Zhu, J.; (15) Zhou, L.; Sheng, D.; Wang, D.; Ma, W.; Deng, Q.; Deng, L.; Zhang, T.; Yang, J.; Wang, Z. ALDH1A3 serves as a predictor for Liu, S. Identification of cancer-type specific expression patterns for castration resistance in prostate cancer patients. BMC Cancer 2020, active aldehyde dehydrogenase (ALDH) isoforms in ALDEFLUOR 20, 387. assay. Cell Biol. Toxicol. 2019, 35, 161−177. (33) Yan, J.; De Melo, J.; Cutz, J.-C.; Aziz, T.; Tang, D. Aldehyde (16) Moreb, J. S.; Ucar, D.; Han, S.; Amory, J. K.; Goldstein, A. S.; dehydrogenase 3A1 associates with prostate tumorigenesis. Br. J. Ostmark, B.; Chang, L.-J. The enzymatic activity of human aldehyde Cancer 2014, 110, 2593−2603. dehydrogenases 1A2 and 2 (ALDH1A2 and ALDH2) is detected by (34) van den Hoogen, C.; van der Horst, G.; Cheung, H.; Buijs, J. Aldefluor, inhibited by diethylaminobenzaldehyde and has significant T.; Pelger, R. C. M.; van der Pluijm, G. The aldehyde dehydrogenase effects on cell proliferation and drug resistance. Chem. Biol. Interact. enzyme 7A1 is functionally involved in prostate cancer bone 2012, 195, 52−60. metastasis. Clin. Exp. Metastasis 2011, 28, 615−625. (17) Morgan, C. A.; Parajuli, B.; Buchman, C. D.; Dria, K.; Hurley, (35) Le Magnen, C.; Bubendorf, L.; Rentsch, C. A.; Mengus, C.; T. D. N,N-diethylaminobenzaldehyde (DEAB) as a substrate and Gsponer, J.; Zellweger, T.; Rieken, M.; Thalmann, G. N.; Cecchini, mechanism-based inhibitor for human ALDH isoenzymes. Chem. Biol. M. G.; Germann, M.; Bachmann, A.; Wyler, S.; Heberer, M.; Spagnoli, Interact. 2015, 234, 18−28. G. C. Characterization and clinical relevance of ALDHbright (18) Luo, M.; Gates, K. S.; Henzl, M. T.; Tanner, J. J. populations in prostate cancer. Clin. Cancer Res. 2013, 19, 5361− Diethylaminobenzaldehyde is a covalent, irreversible inactivator of 5371. ALDH7A1. ACS Chem. Biol. 2015, 10, 693−697. (36) van den Hoogen, C.; van der Horst, G.; Cheung, H.; Buijs, J. (19) Pors, K.; Moreb, J. S. Aldehyde dehydrogenases in cancer: an T.; Lippitt, J. M.; Guzmán-Ramírez, N.; Hamdy, F. C.; Eaton, C. L.; opportunity for biomarker and drug development? Drug Discov. Today Thalmann, G. N.; Cecchini, M. G.; Pelger, R. C. M.; van der Pluijm, G. High aldehyde dehydrogenase activity identifies tumor-initiating 2014, 19, 1953−1963. and metastasis-initiating cells in human prostate cancer. Cancer Res. (20) Quattrini, L.; Sadiq, M.; Petrarolo, G.; Maitland, N. J.; Frame, 2010, 70, 5163−5173. F. M.; Pors, K.; La Motta, C. Aldehyde Dehydrogenases and Prostate (37) Frame, F. M.; Pellacani, D.; Collins, A. T.; Maitland, N. J. Cancer: Shedding Light on Isoform Distribution to Reveal Druggable Harvesting Human Prostate Tissue Material and Culturing Primary Target. Biomedicines 2020, 8, 569. Prostate Epithelial Cells. Methods Mol. Biol. 2016, 1443, 181−201. (21) Marcato, P.; Dean, C. A.; Giacomantonio, C. A.; Lee, P. W. K. (38) Rivera-Gonzalez, G. C.; Droop, A. P.; Rippon, H. J.; Tiemann, Aldehyde dehydrogenase Its role as a cancer stem cell marker comes K.; Pellacani, D.; Georgopoulos, L. J.; Maitland, N. J. Retinoic acid down to the specific isoform. Cell Cycle 2011, 10, 1378−1384. and androgen receptors combine to achieve tissue specific control of (22) Li, D.; Zhang, Q.; Wang, X.; Li, S.; Zhou, H.; Wu, J.; Tian, Y. human prostatic transglutaminase expression: a novel regulatory Self-assembly of a series of thiocyanate complexes with high two- network with broader significance. Nucleic Acids Res. 2012, 40, 4825− photon absorbing active in near-IR range and bioimaging applications. 4840. Dyes Pigments 2015, 120, 175−183. (39) Higgins, L. H.; Withers, H. G.; Garbens, A.; Love, H. D.; (23) Canfield, B. K.; Kuzyk, M. G.; Hightower, S. E.; Li, A. D. Q. Magnoni, L.; Hayward, S. W.; Moyes, C. D. Hypoxia and the Use of quadratic electroabsorption for measurement of the hyper- metabolic phenotype of prostate cancer cells. Biochim. Biophys. Acta polarizability ss of asymmetric molecules. J. Opt. Soc. Am. B 2005, 22, 2009, 1787, 1433−1443. 723−734. (40) Cojoc, M.; Peitzsch, C.; Kurth, I.; Trautmann, F.; Kunz- (24) Yongpruksa, N.; Pandey, S.; Baker, G. A.; Harmata, M. Schughart, L. A.; Telegeev, G. D.; Stakhovsky, E. A.; Walker, J. R.; Benzothiazines in organic synthesis. Synthesis of fluorescent 7-amino- Simin, K.; Lyle, S.; Fuessel, S.; Erdmann, K.; Wirth, M. P.; Krause, M.; 2,1-benzothiazines. Org. Biomol. Chem. 2011, 9, 7979−7982. Baumann, M.; Dubrovska, A. Aldehyde Dehydrogenase Is Regulated (25) Amslinger, S.; Lindner, S. Limno-CP: A Natural-Product- by beta-Catenin/TCF and Promotes Radioresistance in Prostate Inspired 5-Aryl-3(2H)-furanone as Scaffold for a Library of alpha- Cancer Progenitor Cells. Cancer Res. 2015, 75, 1482−1494. Modified Enones. Synthesis 2011, 2011, 2671−2683. (41) Ibrahim, A. I. M.; Sadiq, M.; Frame, F. M.; Maitland, N. J.; (26) Tomita, H.; Tanaka, K.; Tanaka, T.; Hara, A. Aldehyde Pors, K. Expression and regulation of aldehyde dehydrogenases in dehydrogenase 1A1 in stem cells and cancer. Oncotarget 2016, 7, prostate cancer. J. Cancer Metastasis Treat. 2018, 4, 44. 11018−11032. (42) Tanei, T.; Morimoto, K.; Shimazu, K.; Kim, S. J.; Tanji, Y.; (27) Marcato, P.; Dean, C. A.; Pan, D.; Araslanova, R.; Gillis, M.; Taguchi, T.; Tamaki, Y.; Noguchi, S. Association of breast cancer Joshi, M.; Helyer, L.; Pan, L.; Leidal, A.; Gujar, S.; Giacomantonio, C. stem cells identified by aldehyde dehydrogenase 1 expression with A.; Lee, P. W. K. Aldehyde dehydrogenase activity of breast cancer resistance to sequential Paclitaxel and epirubicin-based chemotherapy stem cells is primarily due to isoform ALDH1A3 and its expression is for breast cancers. Clin. Cancer Res. 2009, 15, 4234−4241. predictive of metastasis. Stem Cells 2011, 29, 32−45. (43) Ajani, J. A.; Wang, X.; Song, S.; Suzuki, A.; Taketa, T.; Sudo, K.; (28) Suwala, A. K.; Koch, K.; Rios, D. H.; Aretz, P.; Uhlmann, C.; Wadhwa, R.; Hofstetter, W. L.; Komaki, R.; Maru, D. M.; Lee, J. H.; Ogorek, I.; Felsberg, J.; Reifenberger, G.; Köhrer, K.; Deenen, R.; Bhutani, M. S.; Weston, B.; Baladandayuthapani, V.; Yao, Y.; Honjo, Steiger, H.-J.; Kahlert, U. D.; Maciaczyk, J. Inhibition of Wnt/beta- S.; Scott, A. W.; Skinner, H. D.; Johnson, R. L.; Berry, D. ALDH-1 catenin signaling downregulates expression of aldehyde dehydrogen- expression levels predict response or resistance to preoperative ase isoform 3A1 (ALDH3A1) to reduce resistance against chemoradiation in resectable esophageal cancer patients. Mol. Oncol. temozolomide in glioblastoma. Oncotarget 2018, 9, 22703−22716. 2014, 8, 142−149. (29) Parajuli, B.; Fishel, M. L.; Hurley, T. D. Selective ALDH3A1 (44) Schott, A. F.; Landis, M. D.; Dontu, G.; Griffith, K. A.; Layman, inhibition by benzimidazole analogues increase mafosfamide R. M.; Krop, I.; Paskett, L. A.; Wong, H.; Dobrolecki, L. E.; Lewis, M. sensitivity in cancer cells. J. Med. Chem. 2014, 57, 449−461. T.; Froehlich, A. M.; Paranilam, J.; Hayes, D. F.; Wicha, M. S.; Chang, 3847 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848 Journal of Medicinal Chemistry pubs.acs.org/jmc Article J. C. Preclinical and clinical studies of gamma secretase inhibitors with and virtual screening studies. J. Chem. Inf. Model. 2014, 54, 2980− docetaxel on human breast tumors. Clin. Cancer Res. 2013, 19, 1512− 2986. 1524. (63) Mosmann, T. Rapid colorimetric assay for cellular growth and (45) Maitland, N. J.; Collins, A. T. Prostate cancer stem cells: a new survival: application to proliferation and cytotoxicity assays. J. target for therapy. J. Clin. Oncol. 2008, 26, 2862−2870. Immunol. Methods 1983, 65, 55−63. (46) Carnero, A.; Garcia-Mayea, Y.; Mir, C.; Lorente, J.; Rubio, I. T.; (64) Hirst, A. M.; Simms, M. S.; Mann, V. M.; Maitland, N. J.; LLeonart, M. E. The cancer stem-cell signaling network and resistance O’Connell, D.; Frame, F. M. Low-temperature plasma treatment to therapy. Cancer Treat Rev. 2016, 49, 25−36. induces DNA damage leading to necrotic cell death in primary (47) Wang, T.; Shigdar, S.; Gantier, M. P.; Hou, Y.; Wang, L.; Li, Y.; prostate epithelial cells. Br. J. Cancer 2015, 112, 1536−1545. Shamaileh, H. A.; Yin, W.; Zhou, S.-F.; Zhao, X.; Duan, W. Cancer stem cell targeted therapy: progress amid controversies. Oncotarget 2015, 6, 44191−44206. (48) Grasso, C. S.; Wu, Y. M.; Robinson, D. R.; Cao, X.; Dhanasekaran, S. M.; Khan, A. P.; Quist, M. J.; Jing, X.; Lonigro, R. J.; Brenner, J. C.; Asangani, I. A.; Ateeq, B.; Chun, S. Y.; Siddiqui, J.; Sam, L.; Anstett, M.; Mehra, R.; Prensner, J. R.; Palanisamy, N.; Ryslik, G. A.; Vandin, F.; Raphael, B. J.; Kunju, L. P.; Rhodes, D. R.; Pienta, K. J.; Chinnaiyan, A. M.; Tomlins, S. A. The mutational landscape of lethal castration-resistant prostate cancer. Nature 2012, 487, 239−243. (49) Harris, W. P.; Mostaghel, E. A.; Nelson, P. S.; Montgomery, B. Androgen deprivation therapy: progress in understanding mechanisms of resistance and optimizing androgen depletion. Nat. Clin. Pract. Urol. 2009, 6, 76−85. (50) Drazer, M. W.; Stadler, W. M. The Role of Testosterone in the Treatment of Castration-Resistant Prostate Cancer. Cancer J. 2016, 22, 330−333. (51) Scott, L. J. Enzalutamide: A Review in Castration-Resistant Prostate Cancer. Drugs 2018, 78, 1913−1924. (52) Tannock, I. F.; de Wit, R.; Berry, W. R.; Horti, J.; Pluzanska, A.; Chi, K. N.; Oudard, S.; Théodore, C.; James, N. D.; Turesson, I.; Rosenthal, M. A.; Eisenberger, M. A.; Investigators, T. A. X. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N. Engl. J. Med. 2004, 351, 1502−1512. (53) Tsao, C.-K.; Cutting, E.; Martin, J.; Oh, W. K. The role of cabazitaxel in the treatment of metastatic castration-resistant prostate cancer. Ther. Adv. Urol. 2014, 6, 97−104. (54) Mori, Y.; Yamawaki, K.; Ishiguro, T.; Yoshihara, K.; Ueda, H.; Sato, A.; Ohata, H.; Yoshida, Y.; Minamino, T.; Okamoto, K.; Enomoto, T. ALDH-Dependent Glycolytic Activation Mediates Stemness and Paclitaxel Resistance in Patient-Derived Spheroid Models of Uterine Endometrial Cancer. Stem Cell Rep. 2019, 13, 730−746. (55) Pequerul, R.; Vera, J.; Giménez-Dejoz, J.; Crespo, I.; Coines, J.; Porté, S.; Rovira, C.; Parés, X.; Farrés, J. Structural and kinetic features of aldehyde dehydrogenase 1A (ALDH1A) subfamily members, cancer stem cell markers active in retinoic acid biosynthesis. Arch. Biochem. Biophys. 2020, 681, 108256. (56) Copeland, R. A. Evaluation of Enzyme Inhibitors in Drug Discovery: A Guide for Medicinal Chemists and Pharmacologists, 2nd ed.; John Wiley & Sons, Inc., 2013; pp 57−124. (57) Moretti, A.; Li, J.; Donini, S.; Sobol, R. W.; Rizzi, M.; Garavaglia, S. Crystal structure of human aldehyde dehydrogenase 1A3 complexed with NAD+ and retinoic acid. Sci. Rep. 2016, 6, 35710. (58) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data Bank. Nucleic Acids Res. 2000, 28, 235−242. (59) Morgan, C. A.; Hurley, T. D. Characterization of two distinct structural classes of selective aldehyde dehydrogenase 1A1 inhibitors. J. Med. Chem. 2015, 58, 1964−1975. (60) Schrödinger. MacroModel; Schrödinger, LLC: New York, NY, 2009. (61) Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: a new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening. J. Med. Chem. 2004, 47, 1750−1759. (62) Tuccinardi, T.; Poli, G.; Romboli, V.; Giordano, A.; Martinelli, A. Extensive consensus docking evaluation for ligand pose prediction 3848 https://doi.org/10.1021/acs.jmedchem.1c01367 J. Med. Chem. 2022, 65, 3833−3848