Smenamide A Analogues. Synthesis and Biological Activity on Multiple Myeloma Cells

Smenamides are an intriguing class of peptide/polyketide molecules of marine origin showing antiproliferative activity against lung cancer Calu-1 cells at nanomolar concentrations through a clear pro-apoptotic mechanism. To probe the role of the activity-determining structural features, the 16-epi-analogue of smenamide A and eight simplified analogues in the 16-epi series were prepared using a flexible synthetic route. The synthetic analogues were tested on multiple myeloma (MM) cell lines showing that the configuration at C-16 slightly affects the activity, since the 16-epi-derivative is still active at nanomolar concentrations. Interestingly, it was found that the truncated compound 8, mainly composed of the pyrrolinone terminus, was not active, while compound 13, essentially lacking the pyrrolinone moiety, was 1000-fold less active than the intact substance and was the most active among all the synthesized compounds.


Introduction
Marine sponges, together with their symbiotic microorganisms, have proven to be a rich source of skeletally new substances [1][2][3], which have often inspired novel strategies in anticancer drug discovery. Targeted cancer therapies consist of "drugs" which interfere with specific molecules necessary for tumor growth and progression. A primary goal of these therapies is to fight cancer cells with more precision without hitting normal cells. These drugs are classified into monoclonal antibodies, directed against antigens expressed on the neoplastic cell surface, and small molecules, usually designed to interfere with protein targets [4].
Smenamides A (1) and B (2) (Figure 1) are highly functionalized peptide/polyketide substances isolated by our group in 2013 from the Caribbean sponge Smenospongia aurea [5]. They have proven to be interesting for their structural features, such as the unusual N-methylacetamide western terminus, the dolapyrrolidone eastern terminus, typical of dolastatin-15 (3), a potent antimitotic agent derived from Dolabella auricularia [6], and the chlorovinyl functional group, common to some cyanobacterial metabolites, such as jamaicamides (4-6, Figure 1), isolated from Lyngbiamajuscula [7]. The only difference between the two smenamides resides in the configuration of the C-13/C-15 double bond positioned close to the middle part of the polyketide portion of the molecule. It has been speculated that this could determine a different overall shape and, as a consequence, the different biological behavior observed for smenamides [5].  (4)(5)(6). Configuration at C-16 in smenamide A as determined by synthesis [8].
Smenamides have proven to be active in blocking the proliferation of the Calu-1 cancer cell line at nanomolar concentrations, working, however, with a different mechanism of action. Smenamide A, more interestingly, acts with a clear pre-apoptotic mechanism proving to be the more promising as a lead compound. It is worth stating that the configuration at C-16 in both smenamides remained unassigned in the original study due to the limited amount of the natural substances available. In a recent study, a chiral protocol strategy aimed at the total synthesis of the smenamide family was designed, starting from commercially available S-citronellene, a cheap starting material [8]. Two stereoisomers of smenamide A, namely ent-smenamide A and 16-epi-smenamide A (7, Figure 2), were synthesized. This synthetic effort allowed us to determine the C-16 configuration of smenamide A as R (Figure 1), as well as to develop a flexible synthetic route towards this class of substances.
In the present study, the antiproliferative activity of 16-epi-smenamide A has been evaluated on multiple myeloma (MM) cell lines. MM is a clonal plasma cell malignancy accounting for approximately 13% of all hematological cancers [9]. It originates from post-germinal centre B cells that accumulate somatic hypermutation and immunoglobulin heavy-chain class switching [8]. Several novel agents have been introduced into clinical practice but, after an initial response, most patients relapse or progress with a treatment-refractory disease [10]. For this reason, MM still proves to be incurable for most patients. In this scenario, it is necessary to develop new agents targeting novel pathways relevant for the MM cells, thereby increasing the range of available therapies.
In addition to 16-epi-smenamide A, the eight simplified synthetic analogues 8-15 ( Figure 2) have also been synthesized. They were conceived as "functional-analogues" of smenamide A, incorporating some of the potential activity-determining structural features of the natural product. They were easily prepared thanks to the flexible nature of the previously developed synthetic route, with the aim of probing the importance of the main structural features of the smenamides, that is, the pyrrolinone, chlorovinyl and N-methylacetamido functional groups. In this paper, we illustrate a case-example of the application of this strategy to the design and study of functional-analogues of complex natural lead compounds.

Compounds 7-15
16-epi-smenamide A (7, Figure 2) is the C-16 epimer of the natural smenamide A (1, Figure 1). It was synthesized starting from S-citronellene using the chiral protocol previously reported. 16-epi-smenamide A was tested on SKM-M1 and RPMI-8226 cells, two MM cell lines, showing it to be able to reduce cell viability in a dose-dependent way at nanomolar concentrations (see Section 2.2). We demonstrated that 16-epi-smenamide A, despite possessing the opposite configuration at C-16, retains the potent antiproliferative activity shown by the natural compound, smenamide-A, thus suggesting that this configuration does not affect the nature of its activity. Therefore, as a working hypothesis for the design of simplified analogues of 16-epi-smenamide A (7), this compound was hypothetically disconnected into two main building blocks, corresponding to the polyketide and the peptide moieties. To probe the importance of the main structural features of smenamides, eight "functional-analogues" of 16-epi-smenamide-A were prepared. In particular, the truncated compound 8, retaining the C1-C18 portion of smenamide A, was synthesized to investigate the role of the pyrrolinone moiety. Compounds 9-12, in turn, represent the simplified C15-C27 polyketide portion and retain only the chlorovinyl and N-methylacetamide functional groups. They also served to investigate the role of the geometric isomerism around the C20/C21 double bond. The modulation of the polarity within the 9/11 and 10/12 pairs was achieved by acetylation. Ester 13, only lacking the pyrrolinone moiety, was prepared to simulate the entire polyketide portion, while compound 14 and its acetyl-derivative 15 allowed us to investigate the role of the chlorine atom. In fact, it is well known that the presence of halogens in natural products is important for the modulation of the biological activity [11,12], as previously reported.
Ketone 18 ( Figure 4) is a versatile intermediate to access 16-epi-smenamide analogues. It was easily prepared from commercially available S-citronellene, as depicted in Figure 4, and used as the starting material to obtain the seven analogues 9-15 by the introduction of the chlorovinyl, methylene and α,β-unsaturated ethyl ester functionalities ( Figure 5). Thus, the Wittig olefination of 18 gave the two isomeric chlorovinyl derivatives 19 and 20 in a 3:2 ratio in favor of 19, which could be separated by column chromatography.   Deprotection of both 19 and 20 with tetrabutylammonium fluoride (TBAF) in tetrahydrofuran (THF) afforded alcohols 9 and 10, respectively, whose acetylation with Ac 2 O/pyridine gave the corresponding acetyl derivatives 11 and 12, respectively.
The methylene derivatives 14 and 15 were prepared by Wittig olefination of 18 using methylenetriphenylphosphorane ( Figure 6). In particular, the first obtained product 22 was deprotected with TBAF in THF to give the desired alcohol 14 whose acetylation with Ac 2 O/pyridine finally afforded the acetyl-derivative 15. All synthesized compounds were tested on RPMI-8226 cell lines, as described in Section 2.2. Compound 8, lacking the great part of the polyketide moiety, was not active at all. As for the truncated polyketide compounds 9-15, it was shown that only compound 13, essentially lacking the pyrrolinone terminus, retained a certain degree of activity. In particular, a 1000-fold decreased EC50 value resulted, compared to the intact parent substance 7. Equally, neither alcohols 9 and 10 nor the corresponding acetates 11 and 12, not the dechlorinated analogues 14 and 15 showed significant activities. On the other hand, when the activity data of compounds 9-12 are compared with those of 13, it is evident that the α,β-unsaturated ethyl ester function plays a role in the activity. In addition, even if it seems that the pyrrolinone terminus does not represent a crucial functional part of the molecule, its absence reduces the activity of 13 suggesting that it, or the entire C1-C15 unsaturated moiety, may be equally important for the full activity of smenamides, conferring rigidity to the molecule, possibly needed to exert the activity. However, these data alone do not allow us to speculate about the importance of the chlorine atom as well as of the configuration of the C20/C21 double bond on the activity.

In Vitro Evaluation of Activity on Multiple Myeloma Cell Lines
In order to study the in vitro effects of 16-epi-smenamide A (7) and its synthetic analogues 8-15, inner salt] assays were performed on SKM-M1 and RPMI-8226, MM cell lines, to evaluate their effects on cell viability. Compound 7, tested at increasing concentrations (10-300 nM) for 48 h, was shown to reduce cell viability in both MM cell lines in a dose-dependent way ( Figure 7). More than 50% of viability reduction was observed between 30 and 50 nM concentration. EC 50 for compound 7 was calculated as 44 nM in SKM-M1 cells, and 24 nM in RPMI-8226 cells, after 48 h of treatment.  Further investigation of the cell death mechanism was carried out using compound 13 (at 1 and 5 µM) to treat RPMI-8226 cells. Control experiments were carried out with dimethylsulfoxide (DMSO) as vehicle control, or with untreated cells. After 72 h of treatment, Annexin-V fluorescein isothiocyanate (FITC)/propidium iodide (PI) analyses were performed to evaluate whether the cytotoxic activity of compounds 13 was related to apoptosis induction. Data obtained showed that a significant increase of apoptotic cells at both concentrations occurred when cells are treated with compound 13 (5% of increase at 1 µM respect to control (* p < 0.05) and 66% at 5 µM (*** p < 0.001) (Figure 10a,b)). Moreover, compound 13 was able to significantly decrease the number of cell in G0/G1 phase and increase those in S phase at both concentrations (Figure 10c,d).

Functional Tests
RPMI-8226 cell line was treated with 1 µM and 5 µM of compound 13 or with DMSO vehicle or not treated for 72 h (cell density 3×10 5 cells/mL) and used in: -Apoptosis assay Apoptosis of RPMI-8226 was evaluated by cytometric analysis of Annexin V and PI-stained cells using fluorescein isothiocyanate (FITC) Annexin V Apoptosis Detection kit I (Becton Dickinson, BD, Franklin, NJ, USA) [16]. Samples were prepared following the manufacturer's instructions; stained cells were acquired using NAVIOS flow cytometer (Beckman Coulter, Brea, CA, USA) and analyzed by Kaluza software (Beckman Coulter). 10,000 events were acquired for each samples; single positive for Annexin V and double positive for Annexin V and PI cells were interpreted as signs of early and late phases of apoptosis respectively. Percent of apoptotic cells was obtained from the sum of early and late apoptosis.
-Cell cycle analysis After treatment RPMI-8226 cells were fixed in cold ethanol 70% for 1 h, then labeled with PI (Sigma Aldrich, St. Louis, MO, USA)/RNase A (EuroClone S.p.a., Pero, MI, Italy) staining solution for 30 min. Samples were acquired by NAVIOS flow cytometer and analyzed by Kaluza software (Beckman Coulter). 10,000 events were acquired for each sample.

Statistical Analysis
Statistical significance was determined using a paired t test by GraphPad Prism. All error bars represent the standard deviation (SD) of the average.

Conclusions
This study adds new knowledge about the antiproliferative activity and the possible role of smenamides, chlorinated peptide/polyketide substances originally isolated from the Caribbean sponge Smenospongiaaurea, as lead compounds in anticancer drug research. Our results have shown that the configuration at C-16 slightly affects the activity, since the 16-epi-analogue 7 was still active at nanomolar concentrations. Interestingly, it has been found that the truncated compound 8, containing the pyrrolinone terminus, was not active while compound 13, composed of the intact C12-C27 portion, retained the activity, even though its EC50 value was 1000 times smaller compared with the parent 16-epi-smenamide 7. In addition, compound 13 was able to block the cell cycle at the G0/G1 phase. It is worth noting that smenothiazoles [17], biogenetically related but structurally different from smenamides, possess the same activity. This study provides the basic knowledge needed to design simplified and synthetically easily accessible analogues that could target MM cells.