High Content Image Analysis of Cellular Responses of the Murine J774A.1 Cell Line and Primary Human Cells Alveolar Macrophages to an Extended Panel of Pharmaceutical Agents

Abstract

Introduction: In vitro screening of macrophages for drug-induced effects, such as phospholipidosis, is useful for detecting potentially problematic compounds in the preclinical development of oral inhaled products. High-content image analysis (HCIA) is a multi-parameter approach for cytotoxicity screening. This study provides new insights into HCIA-derived response patterns of murine J774A.1 cells and primary human alveolar macrophages (hAM).

Methods: Several compounds were compared with reference groups (cationic amphiphilic drugs and apoptosis inducers) at different concentrations (0.01 to 10 µM). After incubation, cells were stained with fluorescence markers and HCIA was performed (Cytation™ 5 Cell Imaging System). Ten parameters were analysed: non-adherent cells, increased or reduced mitochondrial activity, membrane permeability, cell area, nuclear area, polynucleated cells, vacuole area, neutral and phospholipid content. A new system of response categorisation was developed for data analysis.

Results: Murine J774A.1 cells exhibited a drug-induced response pattern that was distinct to the corresponding pattern of hAM cells. Comparison with the literature revealed that primary cells (rat or human origin) have similar response patterns, while cell lines (mouse, rat or human) exhibited a different response pattern. Hierarchical clustering revealed toxicologically aligned clusters of compounds, suggesting potential use for understanding mechanisms of drug effects in cell lines and primary cells.

Conclusions: Valuable information for selecting a suitable cell type for HCIA screening of macrophage responses to drug compounds is provided. All cell types were suitable for screening drug-induced phospholipidosis. Still, human primary alveolar macrophages responded differently to drug treatment compared to macrophage cell lines and may be required to evaluate broader response-patterns and mechanisms of toxicity.

Keywords: cationic amphiphilic drugs; drug-induced phospholipidosis; high-content image analysis; human primary alveolar macrophages.

Fig. 1

Images of fluorescently labelled (A) J774A.1 and (B) hAM comparing untreated (left panel) and amiodarone-treated (right panel) cells. Scale bar: 100 μm. The number of adherent cells (including box and whisker plots of quartiles, median, maximum and minimum values) are depicted for (C) untreated J774A.1 and (D) untreated hAM cells. The remaining nine HCIA parameters are shown in a separate diagram using box and whisker plots. Box limits depict the range of the central 50% of the data, with a line marking the median value. Whiskers show the minimum and maximum values. The data was calculated from n = 20 experiments (with six technical replicates each) for J774A.1 cells (passage numbers #4–30) and 5 different hAM donors (D1-5; with six technical replicates each).

Fig. 2

Data distribution (box and whisker plots) of the percentage of abnormal J774A.1 cells following 48 h incubation of a low and high dose of (A) amiodarone, (B) budesonide, (C) campothecin, (D) tiotropium, (E) indacaterol and (F) amikacin. Box limits depict the range of the central 50% of the data, with a line marking the median value. Whiskers show the minimum and maximum values from n = 6 experiments with different passage numbers. Note: Treatment of J774 with 5 µM amiodarone resulted in 98.3–100% elevated phospholipid staining. Due to the narrow distribution of the data, the box and whiskers in graph (A) are small but present in the top right corner.

Fig. 3

Categorisation of dose-dependent responses in J774A.1 cell populations incubated 48 h with increasing concentrations of test compounds grouped according to pharmaceutical class: (A) CADs, (B) antibiotics, (C) beta-agonists, and (D) anticholinergics. Categorisation values were derived from the data distribution of percentage of abnormal cells in J774.A1 cell populations from n = 6 experiments with different passage numbers. The numbers on the left y-axis denote the compound dose in µM. Compounds are listed in order of increasing lipophilicity (based roughly on log D) within the pharmaceutical classes. Individual values for all data points are listed in the supplementary information.

Fig. 4

Categorisation of dose-dependent responses in J774A.1 cell populations incubated 48 h with increasing concentrations of test compounds grouped according to pharmaceutical class: (A) apoptosis inducers, (B) corticosteroids, (C) miscellaneous pharmaceutical classes. Categorisation values were derived from the data distribution of percentage of abnormal cells in J774.A1 cell populations from n = 6 experiments with different passage numbers. The numbers on the left y-axis denote the compound dose in µM. Compounds are listed in order of increasing lipophilicity (based roughly on log D values) within the pharmaceutical classes. Individual values for all data points are listed in the supplementary information.

Fig. 5

Influence of (A) CADs, (B) apoptosis inducers and (C) beta-agonists on the mean percentage of abnormal cells in hAM cell populations (n = 5 hAM donors). The numbers on the left y-axis denote the compound dose in µM. Compounds are listed in order of increasing lipophilicity (based roughly on log D values) within the pharmaceutical classes. Individual values for all data points are listed in the supplementary information.

Fig. 6

Hierarchical cluster map of (A) J774A.1 cells and (B) hAMs treated with selected compounds (5 µM). The heatmap values represent the mean increase in the percentage of abnormal cells. The color-coded squares next to each compound name indicate the pharmaceutical class (Table I).

Fig. 7

Hierarchical cluster map of (A) J774A.1 cells and (B) hAMs treated with selected compounds (10 µM). The heatmap values represent the intensity of the response. The color-coded dots next to each compound name indicate the pharmaceutical class (as defined in Table I).