Bioorganic & Medicinal Chemistry Letters 24 (2014) 1752–1757

Contents lists available at ScienceDirect

Bioorganic & Medicinal Chemistry Letters journal homepage: www.elsevier.com/locate/bmcl

A potential CARB-pharmacophore for antineoplastic activity: Part 1 Zbigniew J. Witczak a,⇑, Tomasz Poplawski b,⇑, Anna Czubatka b, Joanna Sarnik b, Pawel Tokarz c, Adam L. VanWert a, Roman Bielski a a

Department of Pharmaceutical Sciences, Nesbitt School of Pharmacy, Wilkes University, 84 W. South Street, Wilkes-Barre, PA 18766, USA Department of Molecular Genetics, University of Lodz, Lodz 90-236, Poland c Department of Organic Chemistry, University of Lodz, Lodz 91-403, Poland b

a r t i c l e

i n f o

Article history: Received 30 December 2013 Revised 10 February 2014 Accepted 13 February 2014 Available online 23 February 2014 Keywords: CARB-Pharmacophore Thio-sugars Anhydro-sugars 2-C-Methylene-myo-inositol oxide Cancer cell lines

a b s t r a c t Diverse functionalized representatives of various classes of sugars, such as thio-, anhydro-, and sulfamido-sugars and myo-inositol oxide, were synthesized and assessed for cytotoxicity against human cancer cell lines (A549, LoVo, MCF-7 and HeLa). The inositol oxide (4) was more active against MCF-7 cells (i.e., an estrogen-dependent breast cancer line), whereas all 3 sulfur-containing compounds showed strongest activity against A549 cells (i.e., a lung adenocarcinoma line). We propose to use a concept of functional ‘CARB-pharmacophores’ when evaluating a potential for the compounds’ general antineoplastic activity. Future studies will determine the reasons for cell-type specificity of these compounds. The thio-sugar motif appears to be a promising lead for future developments. Ó 2014 Elsevier Ltd. All rights reserved.

Early studies of carbohydrate-binding protein interactions unveiled a new era of research on their roles in biological systems. Since then, many additional studies have proved a number of important structure–activity relationships for carbohydrate-based therapeutics. These studies clearly demonstrated that the diversity and complexity of carbohydrates is permissive for carrying out a wide range of biological functions. Research on carbohydrates is now undergoing considerable growth and promises to be a major source of drug-discovery leads.1–10 One of the important concepts in the development of new carbohydrate-based therapeutics has been coined the ‘sugar code’.11 The ‘sugar code’ can broadly be defined as the constellation of pharmacophores12 and functional groups that lend specific roles to carbohydrates in an organism. The fundamental factor dictated by the ‘sugar code’ is the affinity of a carbohydrate for its target, usually a protein.13 However, when considering carbohydrates studied as potential therapeutics one must note that the number of employed potentially effective functional groups is very limited. Compounds carrying such groups will usually belong to the category of natural product-derived according to Wilson and Danishefsky.14 There have been relatively many studies looking for novel, biologically

⇑ Corresponding authors. Tel.: +1 570 4084276; fax: +1 570 408 4299 (Z.J.W.); tel.: +48 42 6354486; fax: +48 42 6354484 (T.P.). E-mail addresses: [email protected] (Z.J. Witczak), tomasz. [email protected] (T. Poplawski). http://dx.doi.org/10.1016/j.bmcl.2014.02.036 0960-894X/Ó 2014 Elsevier Ltd. All rights reserved.

active compounds belonging to aminosugars, sugars equipped with the sulfate group, and modified with other particular functionalities.14 It is very rare that carbohydrates carrying particular biologically important groups, for example, an epoxy, are considered to be potential therapeutics. Clearly, the ‘Big Pharma’ is not interested in this kind of the research.14 In previous studies we designed several functionalized carbohydrates that are biologically active as a-fucosidase and a-glucosidase inhibitors,15,16 and potentially as anticancer therapeutics.17,18 In continuation of our studies on diverse pharmacophores we want to propose the concept of ‘functional carb-pharmacophores’ (FCP) as a novel approach to testing the biological response from functionalized sugar derivatives. We consider FCPs ‘carbohydrate containing specific chemical functional groups that render these compounds potentially therapeutically useful’. The sugar ‘functional carb-pharmacophores’ selected for the present study are marked in red and depicted in Figure 1. 5-Thio-D-glucose (1) and 6-thio-D-fructopyranose (2) represent a category of rather rare monosaccharides containing a hemi-thioacetal functionality. Note that in both compounds the sulfur atom comprises a part of the ring, and thus, the labile (‘hydroxyl’) part of the hemiacetal contains an oxygen atom. The difference between the two thio-sugars is that moving the hydroxymethyl group from the position C-5 to C-1 of 5-thio-D-glucose (1) (aldose) produces 6-thio-b-D-fructopyranose (2) (ketose). This particular

1753

Z. J. Witczak et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1752–1757

Functional CARB-Pharmacophore motifs selected for the study Thio-sugar motif

Anhydrosugar motif

Sulfamidosugar motif

OH

S

OH

O

OH

H 3CO 2S HN

O HO

O

OH OH

HO

OH OH 1,6-anhydro-5- C -hydroxymethylα-L-altro-pyranose

1

3

5-thio-D-glucose

S HO

OH

HO

OH OH 1,5-anhydro-6-deoxy-6-methanesulfamido-D-glucitol 5

O OH

OH

HO

OH HO

OH 2

OH 4

6-thio-β-D-fructopyranose

OH 2-C -methylene-myo-inositol oxide

Figure 1. The functional groups of diverse sugar motifs.

modification explains the observed difference in pharmacological response from both thiosugars. These compounds are significantly different from thioglucose, for example, where the hemi-thioacetal part contains an SH group and the oxygen atom is a component of the ring. Since the thioglucose-like compounds have been studied extensively we selected sugars with the S-ring thioacetal moiety. The second motif we selected is a unique anhydro-sugar motif, where both oxygen atoms forming the acetal belong to the fivemembered (anhydro) ring and one of these atoms belongs to the pyranose ring as well. The saccharide is additionally equipped with a hydroxymethyl group at the C-5 position. The other anhydrocompound we selected differs substantially from the first anhydro-compound. It is an exocyclic epoxide coupled to the inositol moiety. It is hoped to simulate the motif of exocyclic monosaccharide epoxides. Finally, we chose a motif constructed from a carbohydrate derivative with a methanesulfonamido group. Formally, the compound is not a reducing sugar, but 1,5-glucitol, since the hydroxyl group from the C-1 has been removed by reduction. The compound is equipped with another relatively rare (at least for carbohydrate derivatives) but possibly consequential functionality a sulfonamido group at the primary (C-6) position. The lipophilicity of a drug candidate can have a dramatic impact on a variety of pharmacokinetic characteristics, including its ability to be absorbed after oral administration and passively diffuse across biological membranes, so that it may reach its site of action. Typically, the more lipophilic a drug, the better is its absorption following oral administration, provided its dissolution rate is adequate. This characteristic is particularly relevant to carbohydratebased drugs, as they tend to be more hydrophilic in nature. Lipophilicity of free and specifically functionalized carbohydrates at various positions varies dramatically, and depends on the chemical characteristics and P-value of associated functional groups. However, it is well known that a strong lipophilicity will lead to poor solubility in gastric and intestinal fluids. Thus, establishing a useful hydrophilic–lipophilic balance is critical for development of carbohydrate based therapeutics. An increase in lipophilicity often enhances the cytotoxic effect of molecules,19–22 likely as a result of more rapid passive permeation through the cell membrane, and perhaps sometimes more efficient non-polar interactions with a target protein.

So, strategically functionalizing carbohydrates to increase their lipophilic nature may improve their oral bioavailability and hopefully increase their antineoplastic effects as well. Our ‘functional carb-pharmacophore’ concept is also highly dependent on the lipophilicity value of test compounds and consequently the determined water solubility is of utmost importance. The calculated log P and C log P values for compounds 1–5 along with three representative sugars (glucose, sucrose, trehalose) are listed in Table 1. Although all compounds studied here are hydrophilic, they bear a broad range of log P values. As Table 1 shows the most water soluble compounds are 4 and 5, and the least soluble are 1 and 2. This range is useful for assessing this parameter’s role in cytotoxicity. Molecular models of all functionalized derivatives (1–5) with a putative common pharmacophore were constructed using Accelrys’ software (Fig. 2) as described below. Pharmacophore generation: The five compounds were used to generate the Common Feature Pharmacophores with Accelrys’ Discovery Studio according to the HipHop Catalyst strategy (Fig. 2) as described previously.23,24 The following protocol was used: (1) generation of conformational models for each compound, (2) assignment of each molecule as active (default, i.e., a Principal value of 2) (3) Common Feature Pharmacophore generation using the 3D conformation models in step 1. The four features that were included in the pharmacophore query were hydrogen-bond acceptor, hydrogen-bond donor, hydrophobic, and ionizable groups. The molecular model data for pharmacophore geometry for compounds 1 and 2 (with thio-sugar motifs) clearly indicate that the sulfur atom orientation is outside hydrogen bond donors and Table 1 Values of log P and C log P Compound 1 2 3 4 5 glucose sucrose trehalose

Log P 1.65 1.36 2.38 3.22 3.6 3.24 3.7 5.48

C Log P

Lit. Reference

2.1557 1.9984 1.5973 1.1214 1.1515 22 22 22

1754

Z. J. Witczak et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1752–1757

Figure 2. Shared molecular features of the functional carb-pharmacophores. Accelrys’ Common Feature Pharmacophore Generation was performed as described in methods. Colored spheres represent common features of the five compounds. Hydrogen-bond donor groups were abundant in all compounds, and the ‘consensus’ hydrogen bond donor arrangement is indicated with purple spheres. A hydrogen-bond acceptor group (green) was also common to all compounds, with the ‘consensus’ position indicated with green spheres.

acceptors. Other functional groups –OH at C-6 for 1 and –OH at C-4 for 2 are inside green sphere and are mapped as hydrogen bond acceptors, whereas remaining –OH groups in both compounds 1 and 2 are mapped as hydrogen bond donors inside purple spheres. That particular distinction suggests that the ‘targeted pharmacophore’ can be constructed as a triangle by connecting a green sphere to a purple sphere and a sulfur atom as depicted in Scheme 1. The compiled model pharmacophore for the thio-sugar motif is consistent with our earlier data17 on inhibition of cell proliferation and viability study of (1–4)-S-thiodisaccharides and their sulfoxides and sulfones.18

Scheme 1. Schematic geometry of ‘targeted pharmacophore’ for thio-sugar motif.

The anhydro-sugar motif pharmacophore can be constructed similarly with complete distinction of a hydrogen bond acceptor (green sphere) and a hydrogen bond donor inside purple sphere with oxygen atom of anhydro rings as the third point of the triangle geometry of targeted pharmacophore. Finally, the sulfamidosugar motif pharmacophore is consistent with previous assignment as hydrogen bond acceptor (green sphere) and hydrogen bond donors (purple sphere) with sulfone moiety as the remaining corner of the triangle geometry of targeted pharmacophore. All selected derivatives were synthesized in our laboratory according to our earlier published protocols or literature methods: 5-thioglucose25 (1), 6-thio-fructopyranose26 (2), 1,6-anhydro-5C-hydroxymethyl-a-altro-pyranose27 (3), 2-C-methylene-myo-inositol oxide28 (4), 1,5-anhydro-6-deoxy-6-methane-sulfamido-Dglucitol29 (5). Whereas, literature reports on biological activity of 5-thioglucose30 (1), revealed its specific activity31 as ATP depleting agent there are no literature reports regarding biological activity of 6thio-fructopyranose (2), and 1,6-anhydro-5-C-hydroxymethyl-aaltropyranose (3) and 1,5-anhydro-6-deoxy-6-methane-sulfamido-D-glucitol (5). Interestingly, the 2-C-methylenee-myo-inositol oxide derivative (4) with anhydro-sugar moftiff selected for this study is a well known week inhibitor of phosphatidyl-inositol-specific phospholipase C (PI-PLC).28 None of the selected derivatives (1–5) have ever been screened against cancer cells viability and this is the first report of their biological activity and effect on cancer cell viability. To evaluate the biological properties of examined sugars (1–5) we used a set of cancer cell lines presented in Table 2. We choose cell lines that represent various types of cancer—hormone-dependent, hormone-independent and blood cancers. Those representative sets of cancer cell lines are commonly used in cytotoxicity studies and are well described including their genetic profile.

1755

Z. J. Witczak et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1752–1757 Table 2 Characteristics of examined cell lines Cell line

Tissue/disease

Type

Culture properties

No. of seeded cells/well

A549 LoVo MCF-7 HeLa K562

Lung/carcinoma Colon/Dukes’ type C, grade IV, colorectal adenocarcinoma Mammary gland-breast/adenocarcinoma Cervix/adenocarcinoma Bone marrow/chronic myelogenous leukemia

Hormone-independent Hormone-independent Hormone-dependent Hormone-dependent Blood cancer

Adherent Adherent Adherent Adherent Suspension

1  104 2.5  104 5  104 1  104 2  104

MCF-7, LoVo and A549 cells were cultured in DMEM (Gibco), HeLa cell line in F-12K medium (Gibco) and K562 in RPMI 1640 medium (Gibco) supplemented with fetal bovine serum (BII) and penicillinstreptomycin mix (Sigma). Cells were incubated in humidified incubator at 37 °C, 5% CO2 and passaged every 2–3 days at 80% confluence. Adherent cell lines were plated into a 96-well tissue culture plates for 12 h. prior to experiment. Number of seeded cells per well are shown in a Table 2. Cytotoxicity of studied compounds was analyzed with colorimetric assay - Cell Counting Kit-8 (CCK-8) from Dojindo (Sigma). This assay allows sensitive colorimetric assays for the determination of the number of viable cells in the cytotoxicity assays The detection sensitivity is higher than any other tetrazolium salts such as MTT, XTT or MTS.32,33 At 80% confluence, cells were subcultured into the 96-well plates. After the monolayer of cells became formed for 12 h, cells were treated with a range of concentrations of compounds (2 mM, 4 mM, 6 mM and 8 mM). After the 12-h treatment, we renewed the serum-free medium containing 10% of CCK-8 solution and cells were incubated in a CO2 incubator for 3 h. The CCK-8 assay assessed cell viability by measuring the enzymatic reduction of WST-8 by cellular dehydrogenases to an orange formazan product that is soluble in tissue culture medium. The amount of formazan produced is directly proportional to the number of living cells. The optical density (OD) at 450 nm was determined using a microplate reader. Percentage of cell activity was calculated as: 100  (S B)/(K B) = viability%, where S is the absorbance of treated sample, B is the blank and K is the absorbance of control. All experiments were repeated at least three times, and the data were presented as mean ± SD and analyzed by one-way ANOVA using Statistica software. The P value of

A potential CARB-pharmacophore for antineoplastic activity: part 1.

Diverse functionalized representatives of various classes of sugars, such as thio-, anhydro-, and sulfamido-sugars and myo-inositol oxide, were synthe...
1MB Sizes 4 Downloads 3 Views