Search

In silico approaches identified 1, N-(6-((4-bromo- benzyl)amino)hexyl)-3,5-bis(trifluoromethyl)benzene sulfonamide, as a potential inhibitor of the S100A2-p53 protein-protein interaction, a validated pancreatic cancer drug target. Subsequent cytotoxicity screening revealed it to be a 2.97 μM cell growth inhibitor of the MiaPaCa-2 pancreatic cell line. This is in keeping with our hypothesis that inhibiting this interaction would have an anti-pancreatic cancer effect with S100A2, the validated PC drug target. A combination of focused library synthesis (three libraries, 24 compounds total) and cytotoxicity screening identified a propyl alkyl diamine spacer as optimal; the nature of the terminal phenyl substituent had limited impact on observed cytotoxicity, whereas N-methylation was detrimental to activity. In total 15 human cancer cell lines were examined, with most analogues showing broad-spectrum activity. Near uniform activity was observed against a panel of six pancreatic cancer cell lines: MiaPaCa-2, BxPC-3, AsPC-1, Capan-2, HPAC and PANC-1. In all cases there was good to excellent correlation between the predicted docking pose in the S100A2-p53 binding groove and the observed cytotoxicity, especially in the pancreatic cancer cell line with high endogenous S100A2 expression. This supports S100A2 as a pancreatic cancer drug target.

In silico screening predicted 1 (N-(4-((4-(3-(4-(3-methoxyphenyl)-1H-1,2,3-triazol-1-yl)propyl)piperazin-1-yl) sulfonyl)-phenyl)acetamide) as an inhibitor of the S100A2-p53 protein-protein interaction. S100A2 is a validated pancreatic cancer drug target. In the MiaPaCa-2 pancreatic cell line, 1 was a ∼50 μM growth inhibitor. Synthesis of five focused compound libraries and cytotoxicity screening revealed increased activity from the presence of electron withdrawing moieties on the sulfonamide aromatic ring, with the 3,5-bis-CF3 Library 3 analogues the most active, with GI50 values of 0.91 (3-ClPh; 13 i; BxPC-3, Pancreas) to 9.0 μM (4-CH3 ; 13 d; PANC-1, Pancreas). Activity was retained against an expanded pancreatic cancer cell line panel (MiaPaCa-2, BxPC-3, AsPC-1, Capan-2, PANC-1 and HPAC) and the normal cell line MCF10A (breast). Bulky 4-disposed substituents on the terminal phenyl ring enhanced broad spectrum activity with growth inhibition values spanning 1.1 to 3.1 μM (4-C(CH3 )3 ; 13 e; BxPC-3 and AsPC-1 (pancreas), respectively). Central alkyl spacer contraction from propyl to ethyl proved detrimental to activity with Library 4 and 5.5- to 10-fold less cytotoxic than the propyl linked Library 2 and Library 3. The data herein was consistent with the predicted binding poses of the compounds evaluated. The highest levels of cytotoxicity were observed with those analogues best capable of adopting a near identical pose to the p53-peptide in the S100A2-p53 binding groove.

Lead (Z)-N-(4-(2-cyano-2-(3,4-dichlorophenyl)vinyl)phenyl)acetamide, 1 showed MCF-7 GI50 =30 nM and 400-fold selective c.f. MCF10A (normal breast tissue). Acetamide moiety modification (13 a-g) to introduce additional hydrophobicity was favoured with MCF-7 breast cancer cell activity enhanced at 1.3 nM. Other analogues were potent against the HT29 colon cancer cell line at 23 nM. Textbook SAR data was observed in the MCF-7 cell line, in an MTT assay, via the ortho (17 a), meta (17 b) and para (13 f). The amino alcohol -OH moiety was pivotal, but no stereochemical preference noted. But, these data did not fit our homology modelling expectations. Aberrant MTT ((3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide) screening results and metabolic interference confirmed by sulforhodamine B (SRB) screening. Interfering analogues resulted in 120 and 80-fold CYP1A1 and CYP1A2 amplification, with no upregulation of SULT1A1. This is consistent with activation of the AhR pathway. Piperidine per-deuteration reduced metabolic inactivation. 3-OH / 4-OH piperidine analogues showed differential MTT and SRB activity supporting MTT assay metabolic inactivation. Data supports piperidine 3-OH, but not the 4-OH, as a CYP substrate. This family of β-amino alcohol substituted 3,4-dichlorophenylacetonitriles show broad activity modulated via the AhR pathway. By SRB analysis the most potent analogue was 23 b, (Z)-3-(4-(3-(4-phenylpiperidin-1-yl)-2-hydroxypropoxy)phenyl)-2-(3,4-dichlorophenyl)-acrylonitrile.

Five focused libraries of pyrimidine-based dynamin GTPase inhibitors, in total 69 compounds were synthesised, and their dynamin inhibition and broad-spectrum cytotoxicity examined. Dynamin plays a crucial role in mitosis, and as such inhibition of dynamin was expected to broadly correlate with the observed cytotoxicity. The pyrimidines synthesised ranged from mono-substituted to trisubstituted. The highest levels of dynamin inhibition were noted with di- and tri- substituted pyrimidines, especially those with pendent amino alkyl chains. Short chains and simple heterocyclic rings reduced dynamin activity. There were three levels of dynamin activity noted: 1-10, 10-25 and 25-60 μM. Screening of these compounds in a panel of cancer cell lines: SW480 (colon), HT29 (colon), SMA (spontaneous murine astrocytoma), MCF-7 (breast), BE2-C (glioblastoma), SJ-G2 (neuroblastoma), MIA (pancreas), A2780 (ovarian), A431 (skin), H460 (lung), U87 (glioblastoma) and DU145 (prostate) cell lines reveal a good correlation between the observed dynamin inhibition and the observed cytotoxicity. The most active analogues ( 31a , b ) developed returned average GI 50 values of 1.0 and 0.78 μM across the twelve cell lines examined. These active analogues were: N 2 -(3-dimethylaminopropyl)- N 4 -dodecyl-6-methylpyrimidine-2,4-diamine ( 31a ) and N 4 -(3-dimethylaminopropyl)- N 2 -dodecyl-6-methylpyrimidine-2,4-diamine ( 31b ).

To exploit the interaction of the aryl hydrocarbon receptor (AhR) pathway in developing breast cancer specific cytotoxic compounds, we examined the breast cancer selectivity and the docking pose of the AhR ligands (Z)-2-(2-aminophenyl)-1 H -benzo[ de ]isoquinoline-1,3(2 H )-dione (NAP-6; 5 ) and 10-chloro-7 H -benzo[ de ]benzo[4,5]imidazo[2,1- a ]isoquinolin-7-one (10-Cl-BBQ; 6 ). While the breast cancer selectivity of 5 in vitro is known, we discuss the SAR around this lead, and show for the first time using phenotypic cell line screening and the MTT assay that 6 also presents with breast cancer selectivity , notably in the triple negative (TN) receptor breast cancer cell line MDA-MB-468, the ER+ breast cancer cell lines T47D, ZR-75-1 and the HER2+ breast cancer cell line SKBR3 (GI 50 values of 0.098, 0.97, 0.13 and 0.21 µ M, respectively). Indeed, 6 is 55-fold more potent in MDA-MB-468 cells than the normal MCF10A breast cells (GI 50 of 0.098 vs 5.4 µ M) and more than 130-fold more potent than in cell lines derived from pancreas, brain and prostate (GI 50 of 0.098 vs 10-13 µ M). Molecular docking poses of 5 and 6 together with analogue synthesis and phenotypic screening show the importance of the naphthalene moiety, and an ortho-disposed substituent on the N -phenyl moiety for biological activity.

Knoevenagel condensation of 3,4-dichloro- and 2,6-dichlorophenylacetonitriles gave a library of dichlorophenylacrylonitriles. Our leads (Z)-2-(3,4-dichlorophenyl)-3-(1H-pyrrol-2-yl)acrylonitrile (5) and (Z)-2-(3,4-dichlorophenyl)-3-(4-nitrophenyl)acrylonitrile (6) displayed 0.56±0.03 and 0.127±0.04 μm growth inhibition (GI50 ) and 260-fold selectivity for the MCF-7 breast cancer cell line. A 2,6-dichlorophenyl moiety saw a 10-fold decrease in potency; additional nitrogen moieties (-NO2 ) enhanced activity (Z)-2-(2,6-dichloro-3-nitrophenyl)-3-(2-nitrophenyl)acrylonitrile (26) and (Z)-2-(2,6-dichloro-3-nitrophenyl)-3-(3-nitrophenyl)acrylonitrile (27), with the corresponding -NH2 analogues (Z)-2-(3-amino-2,6-dichlorophenyl)-3-(2-aminophenyl)acrylonitrile (29) and (Z)-2-(3-amino-2,6-dichlorophenyl)-3-(3-aminophenyl)acrylonitrile (30) being more potent. Despite this, both 29 (2.8±0.03 μm) and 30 (2.8±0.03 μm) were found to be 10-fold less cytotoxic than 6. A bromine moiety effected a 3-fold enhancement in solubility with (Z)-3-(5-bromo-1H-pyrrol-2-yl)-2-(3,4-dichlorophenyl)acrylonitrile 18 relative to 5 at 211 μg mL-1 . Modeling-guided synthesis saw the introduction of 4-aminophenyl substituents (Z)-3-(4-aminophenyl)-2-(3,4-dichlorophenyl)acrylonitrile (35) and (Z)-N-(4-(2-cyano-2-(3,4-dichlorophenyl)vinyl)phenyl)acetamide (38), with respective GI50 values of 0.030±0.014 and 0.034±0.01 μm. Other analogues such as 35 and 36 were found to have sub-micromolar potency against our panel of cancer cell lines (HT29, colon; U87 and SJ-G2, glioblastoma; A2780, ovarian; H460, lung; A431, skin; Du145, prostate; BE2-C, neuroblastoma; MIA, pancreas; and SMA, murine glioblastoma), except compound 38 against the U87 cell line. A more extensive evaluation of 38 ((Z)-N-(4-(2-cyano-2-(3,4-dichlorophenyl)vinyl)phenyl)acetamide) in a panel of drug-resistant breast carcinoma cell lines showed 10-206 nm potency against MDAMB468, T47D, ZR-75-1, SKBR3, and BT474. Molecular Operating Environment docking scores showed a good correlation between predicted binding efficiencies and observed MCF-7 cytotoxicity. This supports the use of this model in the development of breast-cancer-specific drugs.

Octahydroepoxyisoindole analogues of norcantharidin were accessed through a Diels-Alder reaction of an amine-substituted furan with maleic anhydride and subsequent reduction of the bicyclo[2.2.1]heptene olefin. Despite retention of the carboxylate and the ether bridgehead known to impart cytotoxic activity to norcantharidin, none of these analogues displayed notable cytotoxicity against the 11 cell lines examined: HT29 (colon), MCF-7 (breast), A2780 (ovarian), H460 (lung), A431 (skin), Du145 (prostate), BE2-C (neuroblastoma), SJ-G2 and U87 (glioblastoma), MIA (pancreatic), and SMA (spontaneous murine astrocytoma). The incorporation of an amino-substituted system post-synthesis of norcantharidin afforded facile access to 14 acid/amide-substituted norcantharidin analogues. Of these, only four displayed sufficient activity at the initial 25 μm compound screening dose to warrant full evaluation of growth inhibition. Common to these analogues was the presence of a 4-biphenyl moiety, and in particular 3-(2-(furan-2-ylmethyl)-3-(4-biphenylamino)-3-oxopropylcarbamoyl)-7-oxabicyclo[2.2.1]heptane-2-carboxylic acid (13 c) and 3-(2-(pyrrole-2-ylmethyl)-3-(4-biphenylamino)-3-oxopropylcarbamoyl)-7-oxabicyclo[2.2.1]heptane-2-carboxylic acid (24) displayed high levels of cytotoxicity, returning GI50 values of 15 nm (HT29) to 2.9 μm (U87) and 17 nm (SMA) to 2.8 μm (U87), respectively. These are the most cytotoxic norcantharidin analogues reported to date.

A series of 28 norcantharidin (NorC)-inspired analogues were accessed via a robust two-step Ugi intramolecular Diels–Alder (IMDA) sequence. Four analogues displayed whole-cell cytotoxicity equipotent to that of NorC and cisplatin against a number of cancer cell lines and a normal breast cell line (MCF10A). Notably, (3S,3aS,6R)-2-benzyl-7-methyl-N-(naphthalen-2-yl)-1-oxo-1,2,3,6-tetrahydro-3a,6-epoxyisoindole-3-carboxamide (trans-27) displayed superior whole-cell activity against breast (MCF-7, GI=2.9 μm) and colon (HT29, GI=6.4 μm) cancer cell lines relative to the control (cisplatin), which elicited respective GIvalues of 6.5 and 11.3 μm against the aforementioned cell lines. This analogue also displayed improved activity relative to NorC across the breast (MCF-7, GI=2.9 μm; NorC GI=7.5 μm), ovarian (A2780, GI=2.2 μm; NorC GI=4.4 μm), and neuroblastoma (BE2-C, GI=2.2 μm; NorC GI=3.7 μm) cancer cell lines. Structure–activity relationship (SAR) investigations demonstrated that retention of sphybridized connections within the tetrahydroepoxyisoindole carboxamide scaffold is crucial, as aromatization to a phenolic functionality decreased activity, whereas removal of a single olefin bond abolished cytotoxicity. Nonetheless, with respect to the latter, use of crotonic acid as opposed 2-butynoic acid in the Ugi-IMDA sequence imparted a significant improvement to diastereoselectivity, with the cis/trans isomer ratio shifting from ≈1:1.2 to ≈0.5:9.5.

Cantharidin (1) and its derivatives are of significant interest as serine/threonine protein phosphatase 1 and 2A inhibitors. Additionally, compounds of this type have displayed growth inhibition of various tumour cell lines. To further explore both of these inhibition pathways, a number of amide-acid norcantharidin analogues (15-26) were prepared. Compounds 23 and 24, containing two carboxylic acid residues, showed good PP1 and PP2A activity, with IC(50) values of approximately 15 and approximately 3 mum, respectively. Substituted aromatic amide analogues 45, 48, 49, 52, 53, and 54 also displayed good PP1 and PP2A inhibition, with IC(50) values in the range of 15-10 microM (PP1) and 11-5 microM (PP2A). However, bulky ortho substituents on the aromatic ring caused the aromatic ring to be skewed from the NCO planarity, leading to a decrease in PP1 and PP2A inhibition. A number of analogues, 20, 22, 25 and 46, showed excellent tumour growth inhibition, with 46 in particular being more potent than the lead, norcantharidin 2.