Design and Synthesis of Novel Hybrid Molecules against Malaria

The effective treatment of malaria can be very complex: Plasmodium parasites develop in multiple stages within a complex life cycle between mosquitoes as vectors and vertebrates as hosts. For the full and effective elimination of parasites, an effective drug should be active against the earliest stages of the Plasmodium infection: liver stages (reduce the progress of the infection), blood stages (cure the clinical symptoms), and gametocytes (inhibit the transmission cycle). Towards this goal, here we report the design, the synthetic methodology, and the characterization of novel hybrid agents with combined activity against Plasmodium liver stages and blood stages and gametocytes. The divergent synthetic approach allows the access to differently linked primaquine-chloroquine hybrid templates in up to eight steps.


Introduction
Malaria is still one of the most menacing infectious diseases worldwide with estimated 207 million clinical cases and 627,000 death cases (WHO; [1]). The treatment of malaria is complicated by the increasing development of resistance to the currently used medicaments and by the pathogen's unique biological characteristics [2]. The intracellular and unicellular Plasmodium parasites [3] develop within a complex life cycle between mosquitoes as vectors and vertebrates as hosts [4]. Plasmodium sporozoites are transmitted through the bite of anopheles mosquitoes to humans [5,6] and are transported by the bloodstream to the liver cells. Therein, they develop into preerythrocytic forms [6] and finally transform into erythrocytes infecting [4,7] merozoites [8]. These merozoites mature into schizonts, followed by the rupture of red corpuscles and the release of thousands of merozoites that reinfect erythrocytes and in the end continue the circle of asexual reproduction [9]. A few merozoites develop into gametocytes that are transmitted back to the female anopheles mosquitoes during the blood meal [10], finally completing the malaria transmission cycle [6]. All clinical symptoms such as fever, anemia, splenomegaly [9], and neurological impairment are associated with the parasitic blood stages [7], and the severity of symptoms depends on the patients' immune status [9].
Among the human pathogenic malarial parasites, Plasmodium falciparum is the most dangerous one [11], responsible for more than 90% of all death cases [12]. Unlike P. falciparum, Plasmodium vivax and Plasmodium ovale have a particular characteristic: they produce hypnozoites [11,13]. The hypnozoites remain quiescent in the liver cells [14] for several weeks up to several years until their activation causes a relapse without a new infectious bite [6]. Plasmodium parasites have another peculiar characteristic: their surface antigens are different in many parts in each developmental stage. With this complex life cycle, in each stage the parasites can be considered as a different organism, although all stages are attributed to the same genome [15]. This complicates the treatment of malaria that should be active against as many stages as possible; thus finding new effective compounds will help current therapies.
Taking this into account, we have reported here the design, synthesis, and characterization of novel hybrid molecules consisting of the well-known antimalarial drugs 2 International Journal of Medicinal Chemistry primaquine (1) and chloroquine (2). The most active hybrid compounds of our study are potential new drug templates for the treatment of malaria and show activities against the Plasmodium liver stages and blood stages and against gametocytes [16,17]. The synthetic methodology uses a divergent synthetic approach to differently linked primaquinechloroquine hybrid templates, resulting in several molecules based on one key intermediate (12).
The new hybrids reported in this work showed good to excellent biological activity against the liver stages (P. berghei), blood stages (P. falciparum, strains 3D7, Dd2, and K1), and gametocytes (P. falciparum); the activity is better than the activity of the parent drugs primaquine (1) and chloroquine (2) combined. The detailed biological results are part of a more comprehensive study described and published in our patent application [16,17].

Basic Considerations for the Design of Hybrid
Molecules against Malaria. Sporozoites and liver stages are fascinating targets for the development of new drugs. Effective drug molecules against the earliest stages would decrease or even avoid the occurrence of blood stages and, consequently, of clinical symptoms (prophylactic effectiveness) [14]. In addition, active compounds against gametocytes would fully eradicate the parasites, thereby efficiently preventing the transmission cycle [14]. However, more than 90% of the current drug research projects worldwide aim at targeting the blood stages [6]. At present, there is no medicament available equally active against all stages of the life cycle and against all Plasmodium species [18]. In order to improve the malaria treatment, the World Health Organization recommended in 2001 to combine antimalarial drugs and to avoid the monotherapeutic treatment [19]. Combination therapy may overcome pharmacodynamic disadvantages [20], but resistance may develop when the drugs have different half-life values or have in general long half-lives, thus reaching subtherapeutic blood concentrations [21][22][23][24]. Additionally, patients often do not comply with complex treatment schemes of multiple medicaments [25].
The hybrid concept gained importance in the last decade. The synthesis of hybrid molecules of two or more [26] established drugs (full or partial molecules) and of natural product structures was used more and more and resulted in synergistic effectiveness, especially against resistant organisms by these new structures [27]. A hybrid drug has a single pharmacokinetic profile, easy to predict and control [28,29], thus superior to a standard combination therapy [30]; hybrid drugs are absorbed, distributed, metabolized, and excreted at one single rate [20]. With hybrid drugs there is no competition for plasma protein binding as in the case of single drugs [20], thus reducing the risk of drug interactions [31]. The ratio of the single drugs is determined by the hybrid structure, and doses cannot be as flexibly administered as in the case of single drugs [32]. In addition, pharmacokinetic characteristics can be controlled by the linkage moiety [33]. For a successful hybrid drug, the activities of each single drug should be within the same concentration range to prevent the activity of only one moiety [20]. In conclusion, hybrid molecules can offer the advantages of a combination therapy along with improved pharmacokinetic profiles and potential activity against resistant strains by these new structures but have the disadvantage of a less flexible administration.
The use of established drugs can result in a rapid therapeutic progress with low economical risk [34,35]. To avoid the vicious circle of a Plasmodium infection, an ideal drug would have good activity against sporozoites (prophylactic activity) [36], preerythrocytic liver stages (causal-prophylactic) [37,38], blood stages (suppressive-prophylactic and therapeutic activity), and gametocytes [39,40] as well as against hypnozoites to avoid relapses [41].
The drugs primaquine (1) and chloroquine (2) have been successfully used in combination for the weekly chemoprophylaxis against malaria during the Vietnam War [42][43][44]. Both are approved and low-priced drugs [45], and they are active against most stages of the Plasmodium infection in humans. Primaquine (1) has a very short half-life [41,46] and accumulates potentially harmful metabolites [47,48]. However, primaquine has been safe and an effective prophylactic agent for nonpregnant women and travelers with a normal glucose-6-phosphate dehydrogenase level [42]: in endemic malaria regions up to 20% of the patients on average suffer from such a deficiency [49]. Chloroquine has a long terminal half-life with a higher risk to develop resistant strains under subtherapeutic concentrations [50,51] and is a safe medicament in therapeutic doses [52]. In addition, primaquine can reduce the chloroquine efflux transport (a resistance mechanism) by binding to the PfCRT channels [53] and support the activity of chloroquine [54].
The goal of our study was to design and synthesize compounds active against different Plasmodium infection stages by the combination of structural motives of both compounds in one hybrid molecule.

Chemistry.
Hybrid molecules of primaquine (1) and chloroquine (2) pharmacophores were designed and synthesized with different types of linkage: with an authentic linker part, without a linkage moiety, with an elongated piperazine diamide or diamine linker bond, and with an aromatic-type linkage (Figures 1 and 2) [16,17].

Primaquine-Chloroquine Hybrid Molecules with
Authentic Linkage. The first dual molecules were synthesized with an authentic linkage part in order to avoid additional structural moieties on the pharmacodynamic and pharmacokinetic properties of the hybrid compounds. Earlier examples of primaquine derivatives showed that the linkage of primaquine can be performed via the primary amine without loss of activity [55,56]. Therefore, hybrid structures of primaquine (1) and a chloroquine moiety (3) were synthesized with a linker using original side chain of primaquine, with a pharmacophore ratio of 1 : 1 and 1 : 2 of primaquine (1) to chloroquine (2). Since a Buchwald-Hartwig amination protocol leads merely to low yields (a decomposition of the palladium  catalyst was observed), a nucleophilic substitution reaction of primaquine (1) and 4,7-dichloroquinoline was performed at 120 ∘ C under neat conditions to obtain the hybrid molecules (rac)-5 (82% yield) and (rac)-6 (77%, Scheme 1) [57].

Primaquine-Chloroquine Hybrid Molecules without
Linkage. Additional derivatives consisting of the aromatic pharmacophore were prepared by a direct linkage of both motives in a ratio of 1 : 1 and 1 : 2 of primaquine (4) to chloroquine (3) moiety.

Divergent Synthetic
Scheme 2: Synthesis of the hybrid compounds 9 and 10. Reagents and conditions: (a) H 2 (g), Pd/C, MeOH (dry), 25 ∘ C; (b) 4,7-dichloroquinoline (0.7 equivalents), neat, 120 ∘ C; (c) 4,7-dichloroquinoline (3 equivalents), neat, 120 ∘ C. different ways: through one or both nitrogen atoms and through amide or amine bonds. By using a divergent synthetic route, six hybrid molecules (13 to 18) were obtained starting from the same educt piperazine (11, Figure 2) in up to eight steps. The synthesized hybrid molecules differ in their number of nitrogen atoms and in their type of bond (amide or amine bond) that determines their basic molecular properties. The dual molecules consist at least of two quinoline nitrogen atoms, of two anilinic nitrogen atoms, and of either one or two basic tertiary piperazine amine functionalities (14, 16, and 18) or of one and two nonbasic tertiary piperazine amide functions (13,15, and 17). Thus, the compounds differ in the possibility to be enriched by protonation in acidic compartments (described for chloroquine [58]), thus influencing the bioactivity against the parasitic stages.
All piperazine-linked hybrid molecules contained a C3 spacer between the primaquine pharmacophore and the piperazine part. The required acyl chloride 19 for the spacer part was produced in 63% yield by the reflux of 3-bromopropionic acid with thionyl chloride [59]. The sequence started with piperazine, which was first protected by a Boc group using Boc 2 O in DCM (dry) at 0 ∘ C to obtain the monoprotected piperazine 20 in 78% yield. Compound 20 was further converted into the bromine derivative 21 in 91% yield by slowly dropping 19 into a solution of 20 in DCM (dry) at temperatures from 0 ∘ C to 25 ∘ C (Scheme 3) [60].
6-Methoxy-8-aminoquinoline (4) was deprotonated using NaH in DMF (dry) at 0 ∘ C and was allowed to react with the bromine-substituted linker part 21 at temperatures from 0 ∘ C up to 25 ∘ C to obtain compound 23 in good yields (74%) and the disubstituted side product 24 in 17% yield. Performing the Boc deprotection of amine 23 using TFA, the free amine and key intermediate 12 of the divergent route was obtained in 98% yield and was linked to 4,7-dichloroquinoline at 120 ∘ C under neat conditions to give hybrid 13 in 62% (Scheme 3). The key intermediate 12 was reduced by LiAlH 4 in THF (dry) to obtain compound 25 in low yields of 30% (Scheme 4). Multiple reasons for the low reduction yields of primaquine analogs have been described earlier [61]: formation of numerous soluble and stable complexes of the primaquine moiety with the reducing agent LiAlH 4 at room temperature [61], which complicates the detection of the complete consumption of the starting material during the reaction. Furthermore, a splitting of the 8-amine-carbon bond by reducing agent LiAlH 4 and a partial reduction of the pyridinic moiety to the corresponding tetrahydroisoquinoline might have caused low yields [61]. Conversion of 25 with 4,7dichloroquinoline at 120 ∘ C under neat conditions gave compound 14 in 41% yield.
The synthesis of hybrid molecules with a further elongated C3-piperazine-C3 linkage started with the introduction of the additional C3 part to the key intermediate 12, which was allowed to react with the acyl chloride 19 at −20 ∘ C to give amide 26 in 69% yield (Scheme 5), and the two additional regioisomers as well as the corresponding elimination product of amide 26. The azide function was introduced to 26 using NaN 3 to obtain the azide 27 in 92% yield, followed by the reduction to the amine 28 in 81% yield using the Staudinger reaction. The linkage of amine 28 with 4,7dichloroquinoline gave the dual molecule 15 or the disubstituted hybrid 17 (depending on the number of equivalents of 4,7-dichloroquinoline) in low yields of 30% (15) and 28% (17), respectively.
Starting from the azide 27, two further hybrid molecules (16,18) were obtained. Azide 27 was reduced using LiAlH 4 in 34% yield to amine 29 with two tertiary and basic piperazine nitrogen atoms that might influence the bioactivity values by enrichment in acidic compartments. The free amine function of compound 29 was linked with 4,7-dichloroquinoline to the monosubstituted 16 and the disubstituted 18 in 17% and 23% yield, respectively (Scheme 6).

Synthetic Route to Hybrid Molecule 30 with an
Aromatic-Type Linkage. The influence of the linkage composition on the bioactivity values was further investigated, and a hybrid molecule with a nonbasic, plain aromatic linker moiety was designed and synthesized. The distance between the linkage positions of the primaquine pharmacophore and of the chloroquine pharmacophore moieties corresponded approximately to the distance in the genuine hybrid molecule 5. The number of nitrogen atoms was not changed, but the lipophilicity was expected to be higher than that for the genuine compound 5 due to the formal para-xylene part.
The first attempts to synthesize 30 from the less expensive chloroquine moiety were not successful. Therefore, the synthesis was started from the primaquine building block.
The protecting group free protocol started with the commercially available 1,4-benzenedimethanol, which was converted by the use of 1,2-dibromotetrachloroethane and PPh 3 to the monobrominated product 31 in 63% yield and to the side product, the double brominated compound 32, in 35% yield (Scheme 7). The introduction of the azide function by NaN 3 gave 99% of compound 33, followed by the tosylation with p-toluenesulfonyl chloride to compound 34 in 71% yield and by the linkage to the primaquine motif 4 yielding compound 35 (41%). The key intermediate 35 was also obtained by the linkage of the brominated derivative 31 with 6-methoxy-8-aminoquinoline (4) in 68% yield, followed by the azide introduction using DPPA and DBU in toluene (dry) at room temperature in 75% yield [62]. The first route over four steps resulted in the key intermediate 35 with an overall yield of 18%; the second route over three steps resulted in intermediate 35 with a yield of 32%.
Compound 35 was reduced to amine 37 (95%) by Staudinger conditions; the amine 37 was finally linked to the second pharmacophore moiety using 4,7-dichloroquinoline at 120 ∘ C under neat conditions, yielding the target hybrid molecule 38 in 43%. 2.3. Discussion. Hybrid 5 showed a significant effect on the morphology of the liver and on the total number of liver stage parasites (P. berghei), which were reduced by a concentration significantly lower than that of primaquine (1) required for the same activity. All additional hybrid molecules showed an influence on the size of liver stages in comparison with the parasite wildtype and on the number of liver stages after 24 hours (Scheme 8).
Amide 13 compared with amine 14 showed a decreased number of liver stages and a greater influence on the growth with a smaller diameter of the parasitic stages. The number of liver stages and the parasite growth were slightly decreased using the elongated amide 15 compared with the 14 with a second C3 linker; in comparison with 15, the amine 16 showed a stronger increase in the number of parasites, whereas the diameter did not alter significantly. The introduction of the second elongating linker part as well as the variation of the basic molecule properties of the four investigated molecules (13 to 16) had very little influence on the activity against the number and the diameter of parasites. The introduction of two chloroquine motives resulted in molecules that were less effective against the growth compared with the previous monosubstituted derivatives: amide 17 and the corresponding amine 18 showed a comparable weak effect on the growth of parasites, but after 24 hours 17 and 18, as well as 16 and 38, showed the lowest number of liver stages. In this case as well, a modification of the basic molecule property values did not influence the activity against liver stages.
Almost all of the investigated hybrid molecules showed improved activity against blood stages of P. falciparum (3D7, Dd2, and K1). Against strain 3D7, hybrid molecule 5 showed lower activity than chloroquine, but 5 was still very active at a concentration lower than 1 M, also in comparison with primaquine and to the chloroquine motif 3. The primaquine moiety 4 was not active. The combination of both the chloroquine and primaquine motifs 3 and 4 showed activity in the range of the chloroquine motif alone and hence had no additional or synergistic effect on each other. Also, the equimolar combination of primaquine (1) and chloroquine (2) showed activity in the concentration range of chloroquine.
Against strain Dd2, hybrid 5 showed a slightly improved activity compared with 3D7. Chloroquine (2) was less active against Dd2 than against 3D7, and primaquine (1) had only a weak effect. The primaquine motif 4 showed no activity, but the chloroquine motif 3 surprisingly showed a better activity compared with 3D7. This was also confirmed by the equimolar combination of both motives 3 and 4. The equimolar combination of chloroquine (2) and primaquine (1) showed enhanced activity compared with only chloroquine (2), thus hinting the chloroquine resistance-reversing effect of primaquine (1) [54].
Against strain K1, the equimolar combination of chloroquine (2) and primaquine (1) showed a better activity than against the strain Dd2. Chloroquine (2) showed a comparable activity in comparison with Dd2, and the order of efficacy of primaquine (1) was 3D7 < Dd2 < K1. Also, the hybrid compound 5 showed the best activity against the K1 strain. The order of activity of compound 5 was 3D7 ≈ Dd2 < K1 (five times higher than Dd2 and 3D7). This higher activity against K1 strain than against the other strains can be probably due to the resistance-reversing effect of primaquine [54], combined with the better activity of primaquine on its own. The  primaquine moiety 4 was inactive against K1, but surprisingly the chloroquine motif 3 showed its best activity against K1. The combination of both motifs 3 and 4 showed effectiveness against K1 in the same range as 3 against K1 due to presence of the chloroquine motif. Both compounds 9 and 10 showed good activities against all strains but were slightly less active than the first representative 5. The derivative 6 showed improved bioactivities in the submicromolar range and had a similar linkage as compound 5 originating from the primaquine side chain but with a pharmacophore ratio of primaquine to chloroquine moiety of 1 : 2. This might be due to not only the number of pharmacophores but also an increased lipophilicity correlating with an enhanced membrane permeability, an increased basicity with a higher enrichment in acidic compartments, or an increased planar, aromatic system capable of shielding the surface of hemozoin from further growth.
The piperazine-linked amide 13 and the amine 14 showed the lowest activities against the blood stages. Only 14 showed very good activity against strain K1 (lower than 1 M). The low activity of both these molecules could be due to the lack of a strong basic center such as the tertiary nitrogen atom, like in compound 16.
Compounds 15 and 16, elongated piperazine derivatives with a second C3 linker, showed improved activities against 3D7 and Dd2 whereby the diamine 16, with higher basic properties, was significantly active against strain K1. This might be influenced by a higher total flexibility of compounds 15 and 16 due to two C3 linkers; the basic properties of the diamide 15 and the diamine 16 might not be the cause of their higher International Journal of Medicinal Chemistry activity, since they are both significantly more effective against all strains compared with hybrid 5. A higher ratio of primaquine to chloroquine motifs of 1 : 2 resulted in a slightly higher activity for 17 and 18 compared with the very good activities of the derivatives 15 and 16.
A combination of linkage properties and the pharmacophore ratio, such as for compound 6, seemed to influence the activity more than just the ratio of primaquine and chloroquine by itself (compounds 15, 16 compared to 17, 18). Also compound 30 with an aromatic-type moiety as linkage and a comparable distance for the pharmacophore moieties as in compound 5 but with higher lipophilicity properties and a possible higher hemozoin shielding (30) showed very good activities, lower than 0.1 M against all strains.
In conclusion, hybrid molecules 5, 6, 15-18, and 30 all showed excellent activity values against the tested strains 3D7, Dd2, and K1 of P. falciparum with concentrations lower than 0.1 M: ten times higher than the definition of a hit substance required in drug development [68]. These compounds also showed activity in the concentration range of equimolar combination of substances primaquine (1) and chloroquine (2) or even better.
The inhibition of the development of blood stages to gametocytes and the death of gametocytes are important to block the vicious circle of transmission, reinfection, and spread of resistant organisms [69,70]. Although a gametocidal activity is required for a successful hybrid molecule, the compounds that are active against blood stages should not induce the formation of gametocytes [71], which can be generally induced by medicaments in subtherapeutic concentrations, since the parasite is able to vary the number of gametocytes to ensure the survival [72,73].
The influence of primaquine (1) on gametocytaemia was used as positive control. As expected, chloroquine (2) showed a neglectable influence. The hybrid molecules all influenced the development of gametocytes. The best activity was obtained for the linker-free compounds 9 and 10, the C3linked piperazine derivative 14, and the first representative 5. Compound 6, the amide 13, the diamide 15, the diamine 16, the diamide 17, the diamine 18, and compound 30 were all comparably active. Interestingly, the linkage moiety, the basicity, and the pharmacophore ratio seemed to barely influence the activity against gametocytes.

Conclusion.
We presented in summary eleven hybrid molecules of primaquine (1) and chloroquine (2) synthesized in eight or less synthetic steps, investigated for their activity against the Plasmodium liver stages, blood stages, and gametocytes (Scheme 8). In conclusion, we showed that it is possible to design and synthesize hybrid drug molecules that have good to excellent bioactivity values against different stages of the Plasmodium infection. We also analyzed the influence on the development of liver stages in hepatocytes (P. berghei), the activity against blood stages of three different strains (3D7, Dd2, and K1 of P. falciparum), and their efficacy against the maturation of gametocytes (P. falciparum) [16,17]. In general, it is important to highlight that hybrid molecules often do not follow structural activity relationship rules of the single components. The most promising structure of the synthesized compounds was the hybrid molecule 30 with the highest activity against the number and the diameter of liver stages, against blood stages of 3D7, Dd2, and K1 (lower than 0.1 M), and against gametocytes. This novel class of hybrid molecules shows activity against all stages of the Plasmodium infection in humans and also against resistant strains [16,17].

Experimental Section
3.1. General Information. All solvents were distilled before use. Commercially available materials were used without further purification, purchased by Sigma-Aldrich. Thin layer chromatography was carried out using silica gel 60 254 or alumina with fluorescent indicator. The compounds were detected by fluorescence quenching at 254 nm, fluorescence at 356 nm, or staining with iodine or ninhydrin. Flash chromatography was performed using silica gel (20-63 mesh), deactivated silica gel (20-63 mesh; 7.5% ammonia), or ICN neutral or basic alumina, deactivated with 15% H 2 O. NMR spectra were obtained on a Bruker DMX 600 apparatus and are reported in ppm relative to internal solvent signal with coupling constants ( ) in Hertz (Hz). Spectra were usually obtained at 25 ∘ C. EI mass spectrometry was carried out on a Finnigan MAT 8200; ESI-HRMS was measured on a Bruker Daltonik micrOTOF-focus. (6). A total of 197.9mg (0.76 mmol) of the free base of primaquine (1) and 477.6 mg (2.41 mmol) of 4,7-dichloroquinoline were heated at 120 ∘ C under neat conditions for 7.5 hours. The reaction mixture was allowed to cool down, followed by the suspension of the residue in a solvent mixture of dichloromethane and small amounts of methanol and by the addition of some drops of ammonia solution to alkalize the mixture. The solvent mixture was removed under reduced pressure; the residue was suspended in dichloromethane, filtered (Celite), and concentrated.

N 1 ,N 4 -Bis(7-chloroquinolin-4-yl)-N 4 -(6 -methoxyquinolin-8 -yl)pentane-1,4-diamine
Purification with flash column chromatography on deactivated silica gel (gradient elution starting with petroleum ether/ethyl acetate 1 : 1 and followed by 1 : 2) yielded compound 6 (340.7 mg, 77%) as yellow crystals.      (4). A suspension of the commercially available 6-methoxy-8-nitroquinoline (7, 1.830 g, 8.96 mmol) and Pd/C (183.0 mg, 10% m/m) in MeOH (dry, 50 mL) was stirred under hydrogen atmosphere at room temperature for 2 hours. The reaction mixture was filtered (Celite) and concentrated, and the residue was purified by flash column chromatography on deactivated silica gel  (9). A total of 49.5 mg (0.28 mmol) of 6-methoxy-8-aminoquinoline (4) and 38.3 mg (0.19 mmol) of 4,7dichloroquinoline were allowed to react at 120 ∘ C under neat conditions for 9 hours. The mixture was allowed to cool down; the residue was suspended in dichloromethane and methanol and finally alkalized with a small amount of ammonia solution. The solvent mixture was removed under reduced pressure; the residue was suspended in dichloromethane, filtered (Celite), and concentrated. Purification on deactivated silica gel (petroleum ether/ethyl acetate 5 : 1) gave product 9 (60.5 mg, 95%) as beige crystals.    (10). 6-Methoxy-8-aminoquinoline (4, 62.1 mg, 0.36 mmol) and 4,7-dichloroquinoline (212.8 mg, 1.07 mmol) were heated at 120 ∘ C under neat conditions for 9 hours. The reaction mixture was suspended in dichloromethane; methanol was added, and the mixture was alkalized by the addition of ammonia solution. The solvent mixture was removed under reduced pressure; the residue was suspended in dichloromethane, filtered (Celite), and concentrated. Purification by flash column chromatography on deactivated silica gel (gradient elution starting with petroleum ether/ethyl acetate 6 : 1, followed by petroleum ether/ethyl acetate 5 : 1) yielded compound 10 (45.      (20) [74]. A solution of Boc 2 O (5.471 g, 25.07 mmol, 0.5 mM) in DCM (dry, 50 mL) was slowly dropped to a solution of piperazine (4.323 g, 50.14 mmol, 0.4 mM) in DCM (dry, 125 mL) at 0 ∘ C for 70 min and was stirred for 1 hour at 0 ∘ C. The solvent was removed under reduced pressure, and the obtained colorless residue was suspended in 75 mL of water. The double protected and colorless compound 22 was filtered off; the filtrate was alkalized using aqueous K 2 CO 3 solution and exhaustively extracted with diethyl ether. The combined organic extracts were dried (MgSO 4 ), filtered, and concentrated. Purification with flash column chromatography on deactivated silica gel (DCM/MeOH 10 : 1) and recrystallisation (diethyl ether) gave 20 (3. (21). A solution of 3-bromopropionyl chloride (19,176.0 mg, 1.03 mmol) was slowly dropped to a suspension of the monoprotected piperazine 20 (95.7 mg, 0.51 mmol) and of sodium acetate (105.7 mg, 1.29 mmol) in DCM (dry, 3 mL) at 0 ∘ C. After completed addition, the reaction mixture was allowed to warm up to room temperature, and the full conversion was determined by thin layer chromatography (deactivated silica gel DCM/MeOH 10 : 1). The excess of sodium acetate was filtered off (Celite) and all volatile substances were removed under reduced pressure. Purification   (23). NaH (782.6 mg of a 55% oily dispersion, 17.93 mmol) was added in portions to a solution of 6-methoxy-8-aminoquinoline (4, 1.041 g, 5.97 mmol) in DMF (dry, 30 mL) under nitrogen atmosphere at 0 ∘ C. The reaction mixture was stirred for 30 min at 0 ∘ C, followed by the addition of the bromine derivative 21 in portions. The mixture was stirred for 1 hour at 0 ∘ C, allowed to warm up to room temperature, and stirred for further 18 hours. The excess of NaH of the reddish brown suspension was carefully hydrolyzed by a small amount of water. The solvent DMF was azeotropically removed using toluene under reduced pressure; the residue was suspended in dichloromethane and filtered (Celite). The filtrate was concentrated and purified by flash column chromatography on silica gel (ethyl acetate 100%). The obtained light yellowish foam was recrystallized (diethyl ether/petroleum ether) and gave compound 23 (1.838 g, 74%) as light yellow crystals.   (24).  (12). TFA (3.2 mL, 0.04 mmol) was added to a yellowish solution of the Boc protected derivative 23 (352.4 mg, 0.85 mmol) in DCM (12 mL) at 25 ∘ C, leading to a change of color to an intense orange color. The reaction mixture was stirred for 1 hour at 25 ∘ C, followed by the addition of aqueous NaHCO 3 solution (decolorization of the solution) and the exhaustive extraction using dichloromethane. The combined organic extracts were dried (MgSO 4 ) and filtered, and all volatile components were removed under reduced pressure. Purification with flash column  (13). Compound 12 (71.7 mg, 0.23 mmol) and 4,7-dichloroquinoline (45.5 mg, 0.23 mmol) were allowed to react at 120 ∘ C under neat conditions for 4.5 hours. The residue was suspended in methanol, the reddish brown mixture was alkalized using ammonia solution (change of color to yellow), and all volatile components were removed under reduced pressure. Purification by flash column chromatography on deactivated silica gel (ethyl acetate 100%) gave compound 11 (67.5 Methoxy-N-(3 -(piperazin-1 -yl)propyl)quinolin-8amine (25).