Automated high-content imaging for cellular uptake, from the Schmuck cation to the latest cyclic oligochalcogenides

Recent progress with chemistry tools to deliver into living cells has seen a shift of attention from counterion-mediated uptake of cell-penetrating peptides (CPPs) and their mimics, particularly the Schmuck cation, toward thiol-mediated uptake with cell-penetrating poly(disulfide)s (CPDs) and cyclic oligochalcogenides (COCs), here exemplified by asparagusic acid. A persistent challenge in this evolution is the simultaneous and quantitative detection of cytosolic delivery and cytotoxicity in a high-throughput format. Here, we show that the combination of the HaloTag-based chloroalkane penetration assay (CAPA) with automated high-content (HC) microscopy can satisfy this need. The automated imaging of thousands of cells per condition in multiwell plates allows us to obtain quantitative data on not only the fluorescence intensity but also on the localization in a very short time. Quantitative and statistically relevant results can be obtained from dose–response curves of the targeted delivery to selected cells and the cytotoxicity in the same experiment, even with poorly optimized cellular systems.

Fluorescence cellular imaging was performed using an IXM-C automated microscope from ImageXpress equipped with a Lumencor Aura III with 5 independently selectable solid-state light sources, bandpass filters and 5 objectives (4× to 60×). Sample preparation and washing on 96-well plates was performed using a Plate washer Biotek EL406®. IR spectra were recorded on a Perkin Elmer Spectrum One FT-IR spectrometer (ATR, Golden Gate) and are reported as wavenumbers ν in cm −1 with band intensities indicated as s (strong), m (medium), w (weak). 1 H and 13 C spectra were recorded on a Bruker 500 MHz spectrometer and are reported as chemical shifts (δ) in ppm relative to TMS (δ = 0). Spin multiplicities are reported as a singlet (s), doublet (d), triplet (t) and quartet (q) with coupling constants (J) given in Hz, or multiplet (m). Broad peaks are marked as br. 1 H and 13 C resonances were assigned with the aid of additional information from 1D and 2D NMR spectra (H,H-COSY, DEPT-135, HSQC and HMBC). LC-MS were recorded using a Thermo Scientific Accela HPLC equipped with a Thermo C18 (5 cm × 2.1 mm, 1.9 μm particles) Hypersil gold column coupled with an LCQ Fleet three-dimensional ion trap mass spectrometer (ESI, Thermo Scientific) with a linear elution gradient from 95% H2O/5% CH3CN + 0.1% TFA to 10% H2O/90% CH3CN + 0.1% TFA in 4.0 min at a flow rate of 0.75 mL/min. HRESIMS for the characterization of new S3 compounds was performed on a Xevo G2-S Tof (Waters) and are reported as the mass-percharge ratio m/z calculated and observed.
Complex 25. To a freshly prepared solution of WT streptavidin (1 mL in PBS, 20 μM), 23 (6 μL, 10 mM in DMSO, 3 equiv) was added and the mixture was shaken for 10 min at rt. Then, the complex 25 was concentrated and washed with PBS (2 × 0.5 mL) using Amicon® Ultra 0.5 mL centrifugal filters (cut off: 30 kDa, 10 min, 14.0 krpm). The material was recovered from the centrifugal filter and diluted again with PBS to the required concentration (i.e., 200 μM, 100 μL).

HeLa cells stably expressing the HaloTag-GFP-Mito fusion protein (HGM):
The cells were originally designed by the Chenoweth lab [S4]. They were cultured using the described procedure [S3].

HeLa cells transiently transfected with the GTS-HaloTag-GFP:
The plasmid was a gift from the Hensel group [S5]. HeLa cells were seeded at 5 × 10 4 cells/mL in DMEM + 10% FBS + 1% Pen/Strep on 96-well ibiTreat sterile μ-plates (150 μL per well) and kept at 37 °C with 5% CO2 overnight. On the next day, transient transfection was performed by diluting the cells with The cells were kept at 37 °C with 5% CO2. After 4 h, the medium was exchanged with fresh FluoroBrite DMEM + 10% FBS, and the cells were incubated overnight.

General HC CAPA protocol
As in [S1]. HGM cells were seeded at 8 × 10 4 cells/mL in FluoroBrite DMEM + 10% FBS on μ-plates 96-well ibiTreat sterile and kept at 37 °C with 5% CO2 overnight. The next day, the cells were washed with PBS (3 × 3 mL/well) and the media were exchanged to Leibovitz's (4  Figure S2. Duplicates were performed for each condition.

Optimization of the HC CAPA data analysis in HGM cells
HC CAPA was performed as in Section 4.1, except that 25 was not incubated (only 26 and Hoechst 33342 were added). As in [S1], for each cell, the blue and green channel images are used for the segmentation of the nuclei and the whole cell body, respectively ( Figure S3). Figure S3: Segmentation of the nuclei (blue, right) and the cell body (green, right) using the green channel (left). Scale bar: 50 m.

S8
Three successive filters have then been used for the cell selection process ( Figure S4): 1) Cells positively expressing GFP (GFP channel with minimum intensity threshold).
2) The shape of the nucleus, to eliminate dead or dividing cell (maximum roundness of 0.7, with roundness = height/length).
3) The size of the nucleus, an additional filter for dividing and dead cells (minimum area of 130 µm 2 ). Top-hat transform of the green channel image is used to lower the background and to facilitate the segmentation of the mitochondria, creating the corresponding mask ( Figure S5).  To correlate the GFP and 26 fluorescence, integrated intensity sums for each cell (at least 1500 cells) were plotted and fitted with a linear regression to retrieve the goodness of fit (r 2 ). S10

HC CAPA for transient transfection
HeLa cells were transfected as described in Section 3. HC CAPA was performed as in Section 4.1, except that 25 was not incubated (only 26 and Hoechst 33342 were added). As in [S1], during imaging, samples were kept at 37 °C with 5% CO2. A total of 25 images/well at 40× were recorded, using three channels: blue (

Data analysis of the HC CAPA for transient transfection
For each cell, the blue channel image is used for the segmentation of both nuclei and the whole cell body ( Figure S8). Figure S8: Segmentation of the nuclei (blue, right) and cell body (green, right) using the blue channel (left). Scale bar: 50 m.
Top-hat transform of the green channel image is used to lower the background and to facilitate the segmentation of the Golgi apparatus, creating the corresponding mask.

S13
To correlate the GFP and 26 fluorescence, the integrated intensity sums for each cell (at least 1500 cells) were plotted and fitted with a linear regression to retrieve the goodness of fit (r 2 ). The resulting dependence of the relative intensity values (Irel) from the concentration of 25 (c25) was plotted and fitted with Equation S1 to retrieve the half maximal effective concentration (CP50) value (n is the Hill coefficient).

HC
The cytotoxicity was evaluated by calculating the ratio between the number of selected and nonselected cells and fitted with a linear regression. S14 Figure S12: 1 H NMR (500 MHz, DMSO-d6) spectrum of 23.