High-resolution phenotypic profiling of natural products-induced effects on the single-cell level

Abstract

Natural products (NPs) are highly evolved molecules making them a valuable resource for new therapeutics. Here we demonstrate the usefulness of broad-spectrum phenotypic profiling of NP-induced perturbations on single cells with imaging-based High-Content Screening to inform on physiology, mechanisms-of-actions, and multi-level toxicity. Our technology platform aims at broad applicability using a comprehensive marker panel with standardized settings streamlined towards an easy implementation in laboratories dedicated to natural products research.

Figure 1. Broad-spectrum High-Content Screening technology platform providing high-resolution cytological profiling.

(a) A combination of 14 cellular markers informs on compound-induced perturbations, including all major organelles as well as components of important regulatory pathways. (b) The generation of cytological profiles involves fluorescent staining of cellular targets, imaging, and subsequent quantitative analysis of selected features from 500–1000 valid objects (i.e., cells). An overall number of 134 cellular features is extracted to generate high-resolution cytological profiles. These are reduced to 20 ‘core’ features for straight-forward analysis and convenient visualization. (c) Cytological profiles composed of 20 core features of three natural products are shown as polar plots, including two flavones (1 = 3,3′,4′,5,5′,7,8-Heptamethoxyflavone; 2 = 3′,4′,5,6,7-Pentahydroxy-3-methoxyflavone) and a derivative of podophyllotoxin (3). Bar plots indicate percentages of cells in each phase of the cell cycle: Low = damaged or apoptotic cells (low DNA content); 2N = cells in G0/G1 phase; 2N_4N = cells in S phase; High = cells in M phase and higher ploidy.

Figure 2. Prediction and validation of biological targets/mechanism of action (MOA) of antineoplastic compounds.

(a) Hierarchical clustering of cytolological profiles composed of 20 core features of a group of 26 anti-neoplastic compounds reveals various distinct functional sub-groups that clearly reflect different molecular targets and MOAs. Colors indicate positive (yellow) or negative (blue) deviation from the mean of untreated control cells (value = 1). Full cytological profiles were used for clustering (left). Cytological profiles showing a reduced set of 20 core markers are displayed for easier visualization of affected processes and a vertical grey bar was used to separate between regulatory (i.e. NFkB, p53, and caspase 9 activation) and other cellular markers. Colors indicate positive (yellow) or negative (blue) deviation from the mean of untreated control cells (value = 1). The dendrogram depicts distances between individual cytological profiles based on Spearman rank correlation. (b) Four candidate natural products and three reference compounds were tested for phosphorylation of H2AX in a cell based assay to assess DNA double strand breaks caused by topoisomerase II poisons (control shows no effects). (c) Detection of phospho-H2AX in HeLa cells treated with vehicle control (non-treated) or with 20 μM of compound 4 (see d) for 6 hours. (d) Structures of selected natural products and reference compounds of the topoisomerase cluster from (a): derivative of 10-hydroxycamptothecin (4), Norsecurinine (5), Dorsmanin E (6), 2-(3-Acetoxy-10,13-dimethyl-2,3,4,7,8,9,10,11,12,13,14,15,16,17-tetradecahydro-1H-cyclopenta[a]-phenanthren-17-yl)-2-hydroxypropyl acetate (7), (S)-(+)-Camptothecin, beta-Lapachone, and Genistein.

Figure 3. High-resolution cell-based structure-activity profiling.

(a) Hierarchical clustering of cytological profiles of a group of 16 closely related derivatives of podophyllotoxin reveals highly similar patterns of activity on a set of markers as well as pronounced individual differences. Full cytological profiles were used for clustering (left). Cytological profiles showing a reduced set of 20 core markers are displayed for easier visualization of affected processes and a vertical grey bar was used to separate between regulatory (i.e. NFkB, p53, and caspase 9 activation) and other cellular markers. Colors indicate positive (yellow) or negative (blue) deviation from the mean of untreated control cells (value = 1). The dendrogram depicts distances between individual cytological profiles based on Pearson correlation. (b) Comparison of cell cycle profiles (bar plots) and cytological profiles (heat map) of two chlorinated derivatives distinguished only by the position of chlorine in the phenyl ring. Cell cycle profiles indicate percentages of cells in each phase of the cell cycle: Low = damaged or apoptotic cells (low DNA content); 2N = cells in G0/G1 phase; 2N_4N = cells in S phase; High = cells M phase and higher ploidy. (c) Comparison of cytological profiles (upper bar plots) and cell cycle profiles (lower bar plots) of three brominated derivatives distinguished only by the position of bromine in the phenyl ring. Most pronounced differences of the three compounds (i.e., 10, 11, 12) in the cytological profiles are represented in the colored bar plot. Dashed line indicates control treatment levels (i.e., 1). Cell cycle profiles indicate percentages of cells in each phase of the cell cycle: Low = damaged or apoptotic cells (low DNA content); 2N = cells in G0/G1 phase; 2N_4N = cells in S phase; High = cells in M phase and higher ploidy. Chemical structures show podophyllotoxin and a scaffold of the derivatives of podophyllotoxin that were used in this study.