Autofluorescence imaging captures heterogeneous drug response differences between 2D and 3D breast cancer cultures.

M C Skala,T M Cannon,A T Shah

doi:10.1364/boe.8.001911

Abstract

Two-photon microscopy of cellular autofluorescence intensity and lifetime (optical metabolic imaging, or OMI) is a promising tool for preclinical drug development. OMI, which exploits the endogenous fluorescence from the metabolic coenzymes NAD(P)H and FAD, is sensitive to changes in cell metabolism produced by drug treatment. Previous studies have shown that drug response, genetic expression, cell-cell communication, and cell signaling in 3D culture match those of the original in vivo tumor, but not those of 2D culture. The goal of this study is to use OMI to quantify dynamic cell-level metabolic differences in drug response in 2D cell lines vs. 3D organoids generated from xenograft tumors of the same cell origin. BT474 cells and Herceptin-resistant BT474 (HR6) cells were tested. Cells were treated with vehicle control, Herceptin, XL147 (PI3K inhibitor), and the combination. The OMI index was used to quantify response, and is a linear combination of the redox ratio (intensity of NAD(P)H divided by FAD), mean NADH lifetime, and mean FAD lifetime. The results confirm that the OMI index resolves significant differences (p<0.05) in drug response for 2D vs. 3D cultures, specifically for BT474 cells 24 hours after Herceptin treatment, for HR6 cells 24 and 72 hours after combination treatment, and for HR6 cells 72 hours after XL147 treatment. Cell-level analysis of the OMI index also reveals differences in the number of cell sub-populations in 2D vs. 3D culture at 24, 48, and 72 hours post-treatment in control and treated groups. Finally, significant increases (p<0.05) in the mean lifetime of NADH and FAD were measured in 2D vs. 3D for both cell lines at 72 hours post-treatment in control and all treatment groups. These whole-population differences in the mean NADH and FAD lifetimes are supported by differences in the number of cell sub-populations in 2D vs. 3D. Overall, these studies confirm that OMI is sensitive to differences in drug response in 2D vs. 3D, and provides further information on dynamic changes in the relative abundance of metabolic cell sub-populations that contribute to this difference.

Highlights

Many promising new drugs fail in clinical trials, highlighting a need for more robust preclinical screens that eliminate these drugs before costly clinical trials [1]
This study aims to verify that Optical metabolic imaging (OMI) is sensitive to these known differences in drug response between 2D and 3D cultures, by conducting experiments in breast cancer cell monolayers in parallel to previous experiments conducted in 3D breast cancer tumor organoid cultures [15]
The OMI index decreases in cells that are responding to a drug, and the OMI index increases or does not change in cells that are not responding to a drug

Summary

Introduction

Many promising new drugs fail in clinical trials, highlighting a need for more robust preclinical screens that eliminate these drugs before costly clinical trials [1]. Differences in the microenvironment of 2D and 3D cultures, including unique molecular localization patterns and ECM composition, trigger different signal transduction cascades and result in different cell behaviors [5]. These differing behaviors in 2D vs 3D cultures include apoptotic sensitivity, kinase phosphorylation, and regulation of the activity of receptors involved in neoplastic transformation, such as estrogen receptor (ER) and human epidermal growth factor receptor 2 (HER2) [4]. It is advantageous to include other, non-malignant cells representative of the heterogeneous cell populations in in vivo tumors, such as stromal cells, as they are known regulators of proliferation, quiescence, and drug sensitivity in nearby malignant cells [10]

Objectives

Methods

Results

Conclusion