Novel Synthetic Approaches for Bisnaphthalimidopropyl (BNIP) Derivatives as Potential Anti-Parasitic Agents for the Treatment of Leishmaniasis

Leishmaniasis is a neglected parasitic disease that is widely seen in more than 60 countries worldwide, including Turkey and its subcontinental region. There are several chemotherapy agents for the treatment of leishmaniasis, including pentavalent antimonials—i.e., sodium stibogluconate (Pentostan) and meglumine antimoniate (Glucantim), pentamidine, conventional amphotericin B deoxycholate, miltefosine, paramomycin (aminosidine), and liposomal amphotericin B. However, these therapies are usually unsatisfactory due to dose-limiting toxicity issues and limited efficacy. Furthermore, resistance gained by parasites endangers future success of these therapies. Addressing these issues, the development of novel drugs with high efficacy has a vital importance. Latest studies have shown that bisnaphthalimidopropyl (BNIP) derivatives display high activity against Leishmaniasis parasites by selectively targeting parasitic sirtuin proteins and interacting with DNA. Despite the promising anti-parasitic activity, the low solubility and toxicity on human macrophages are the limitations to overcome. This study describes the new synthesis strategies for existing—i.e., BNIPDaoct and BNIPDanon—and novel BNIP derivatives differing in respect of their alkyl linker chain lengths. The new synthesis approach provides certain advantages compared to its existing alternatives reported in the literature. The proposed methodology does not only decrease the number of synthesis steps and production time of BNIPDaoct and BNIPDanon, but also provides higher yields, thereby making the synthesis highly cost-effective.


Introduction
Considered as a neglected parasitic disease, Leishmaniasis is widely seen in more than 60 countries worldwide, including Turkey and its subcontinental region, affecting more than 1.5 million people with an annual death rate reaching up to 100,000 people [1,2]. It is usually transmitted by sandfly carrying protozoan kinetoplastid parasite to humans. Infective forms of this parasite are called metacyclic promastigotes that are taken up by macrophages at the vertebrate host, such as diaminononane (BNIPDanon), but also provides higher yields, thereby making the synthesis highly cost-effective. Our paper focuses on medicinal chemistry of the molecules. With the advantage of simplicity in synthesis, the proposed strategy may widen the studies on the use of BNIP in medicine.

Chemistry
All oxygen-and moisture-sensitive reactions described herein were performed in glassware that was oven dried (110 • C, 12 h) then flame dried (N 2 stream) immediately prior to use. Reactions were conducted under an atmosphere of nitrogen, using pre-dried septa. The tips of cannulae were flame-dried under a stream of dry nitrogen gas prior to use. Most of the reaction solvents such as dimethylformamide (DMF), tetrahydrofuran (THF), dichloromethane (DCM), and acetonitrile (ACN) were purchased as sure seal dry solvents. All other reagents and solvents were purified according to standard convention. Unless indicated otherwise, silica gel chromatography (SGC) was performed on 70-240 mesh according to routine published procedures. Analytical thin-layer chromatography (TLC) was performed on glass-backed 250~plates by visualizing with anisaldehyde and phosphomolybdic acid. 1 H Spectra were recorded at 500 MHz and 13 C spectra at 125 MHz, both on a Bruker AM 400 instrument (Bruker, Karlsruhe, Germany). Mass spectra were recorded on a Shimadzu mass spectrometry (MS) system equipped with electron spray ionization (ESI) source operated in positive mode (Kyoto, Japan). Mass spectra were recorded on a Shimadzu mass spectrometry (MS) system. In all synthetic methods, appropriate reference methods were used. All other chemicals were purchased from Sigma Aldrich (Germany) and used as supplied. All melting points were taken on a Stuart melting point apparatus SMP30 (Staffordshire, UK).

Results and Discussion
This study reports the synthesis of existing (i.e., compounds 3 and 4) and novel BNIP analogs (i.e., compounds 5, 6, 7 and 8) that may possess potential anti-parasitic activity. BNIPDaoct (compound 3) and BNIPDanon (compound 4) that were developed by Lin and Pavlov [37] were taken as the starting point of the study. The new derivatives differ from these in respect of their alkyl linker chains. Accordingly, the intermediate compounds, compounds 1a and 1b, were synthesized as illustrated in Figure 1; and compound 2 was synthesized in the CDI and butanediol in THF as illustrated by Figure 2.  Different from the synthesis procedure described in detail in Oliveira et al. [16], where a five-step synthesis was designated, BNIPDaoct was synthesized in our approach within two steps ( Figure 3). According to the previous approach ( Figure 3B), first the naphthalimidopropanol compound was obtained by reaction of 1,8-naphthalic anhydride and 3-aminopropanol in the presence of DBU base and DMF solvent, affording the production of target compound, i.e., compound 1b, with 95% yield [16]. This intermediate was then reacted with p-toluenesulfonyl chloride for a tosylation reaction in pyridine solvent at 0 • C temperature to afford O-tosylpropylnaphthalimide compound with 53% yield. The 1,8-diaminooctane compound was activated by treatment with mesitylene chloride at room temperature in the pyridine solvent to give N 1 ,N 8 -dimesityloctane compound in 70% yield ( Figure 3B). Dimesitylated intermediate was coupled with O-tosylpropylnaphthalimide compound in the next step. The synthesized N 1 ,N 8 -dimesityloctane, and the O-tosylpropylnaphthalimide were used as starting material of the N-alkylation reaction. BNIPDaoct compound was then obtained with 85% yield by deprotection reaction, where anhydrous dichloromethane solvent and hydrobromic acid/glacial acetic acid reagent were used to remove mesyl protecting groups in the backbone of the compound, and thereby fully protected polyamine derivatives were obtained. With each single-reaction yield given, the five-step synthesis of BNIPDaoct in Oliveira et al. [16] occurs within an overall yield of 30%.
In contrast to this previous approach, the synthesis of BNIPDaoct (compound 3) was achieved within two steps through our new approach. Accordingly, the substitution reaction was first carried out as indicated in Figure 3A. The substitution reaction between 1,8-naphthalic anhydride and 1,3-diaminopropane at 75 • C in ethanol as solvent provided the desired intermediate compound 1a In the second and final step, compound 1a and 1,8-dibromoctane were let into reaction in the presence of DBU base in MeOH solvent to produce BNIPDaoct (compound 3) with 60% yield ( Figure 3A). Two-step BNIPDaoct synthesis occurred with an overall yield of 48%, which is better than the yield achieved in Oliveira et al. [16].
The lack of three additional chemical steps and higher yield clearly provides cost effectiveness for the synthesis of not only BNIPDaoct, but also other BNIP derivatives such as BNIPDanon (compound 4) and compounds 5 and 7. As an observation, it is also important to state that, for the synthesis of compound 1a, at first, we attempted the reaction conditions with organic solvent such as dry DMF and organic bases such as DBU and DIEA, but this synthesis method was later changed due to extremely low yields. Therefore, adaptation to the outlined approach not only shortened the synthesis, but also overcame the challenges in the synthesis of BNIP analogs. 13 C NMR analysis of compound 3 (BNIPDaoct) and compound 4 (BNIPDanon) is presented in Table 1. Compound 4 (BNIPDanon) is the homologated version of compound 3 (BNIPDaoct) that has additional CH 2 moiety in the linker portion. Thus, again applying the two-step chemical approach described in Figure 3A, BNIPDanon (compound 4) was successfully synthesized with 50% yield ( Figure 4A).  Compound 5 was obtained as a result of the amide reaction of the amine group of the first step product, compound 1a, with subaric acid ( Figure 4A). Compound 6 was obtained by amide coupling procedure with the amine groups at both ends of the diamine using 1-8 diaminooctane in the amide reaction, where compound 1c, 3-(1,3-dioxo-1H-benzo[de]isoquinolin-2(3H)-il)propanoic acid (C 15 H 11 NO 4 ), was used as the starting material ( Figure 4C).
For the synthesis of compound 8, firstly we attempted the conditions same as in the synthesis of compound 7 prior to carbonation of compound 1b with CDI, followed by the reaction of 1,4-diaminobutane. However, this method was abandoned due to long four-step synthesis and low yields in the reactions. For that reason, compound 1b was synthesized initially as described in Figure 1, which was then followed by a reaction with 1,4-diisocyanatobutane, affording compound 8 with 70% yield ( Figure 4B).
Although the synthetic feasibility of these novel BNIP analogs looked very straightforward, several challenges were faced during the synthesis, mainly due to the highly hydrophobic characters of BNIPs and their initial or intermediate compounds. These challenges were overcome through the innovated synthesis approach introduced in this study ( Figure 3A). Linker modifications are seen as the prominent solution to enhance the solubility of BNIP analogs, which may potentially improve the activity of the compounds. Herein, this study reports a new and more feasible chemistry approach for the synthesis of both existing and novel BNIP analogs containing various substituents as drug-like moieties. While the yields of the production seem to vary depending on the structure of the analogs, the synthetic feasibility of these molecules evidently provides insights about the novel two-step synthesis approach.

Conclusions
In conclusion, this study introduces a concise and flexible approach for the assembly of a wide variety of BNIP-scaffold analogs. Novel BNIP analogs-including compounds 5-8-are designed in order to improve the solubility and the activity of the existing molecules, i.e., BNIPDaoct (compound 3) and BNIPDanon (compound 4). Since 8 or 9 carbon linkers were the optimum linkers based on the preliminary assay results against L. infantum promastigotes [16], linker modification analogs of these BNIP molecules were firstly designed. Accordingly, functional groups such as amides, ureas, carbametes, reverse amides, and reverse carbametes were added to the carbon chain backbone of BNIPDaoct and BNIPDanon. Although the chemistry reported by Kong Thoo Lin and Pavlov (2000) looked very straightforward [37], some serious issues related with low yield and solubility obliged alteration of the synthesis strategy during the study, which rewardingly resulted in development of a more cost-effective and faster route for the synthesis of BNIP molecules, as presented for the first time in this study. Two-step synthesis of BNIP derivatives stands as an alternative approach to the synthesis procedure reported in the literature [16,37].
A salient feature of the presented approach is avoidance of longer steps, as well as higher costs, for synthesis. This route is complementary to known methodologies [16,37] and should be applicable for the preparation of many interesting compounds hosting a substituted with different linker motifs. Thus, modifications to the linker chain, as well as on the aryl groups in the BNIP derivatives, might show some improvements in its intrinsic biological activities against Leishmaniasis, as well as cancer, which will be the topic of our future studies. The cellular activity of newly synthesized BNIP derivatives is currently being investigated on L. infantum parasites and their results will be reported separately in the future.