Materials and methods
Culturing and treatment of the worm strain
Caenorhabditis elegans wild type Bristol N2 strain was cultured at 20 °C on nematode growth media (NGM) agar plates seeded with the Escherichia coli strain OP50. For microfluidic experiments, the E. coli strain HT115 was suspended in S-medium at a concentration of 1.4 × 109 cells/ml. Worms were exposed to perovskite compounds from L3 stage the day of the experiment. The injected media and perovskite solutions were freshly prepared the day of the experiment as follows. MAPbI3 and MASnI3 were suspended in S-medium of 50 mg/ml. After filtration at 0.22 um, this stock solution was diluted in bacteria-containing S-medium to 50, 100 and 200 ug/ml.
Microfluidics experiments
The microfluidics device fabricated and tested by Cornaglia et al.(Cornaglia et al., 2016) contains 8 separated channels and allows for testing 8 biological conditions during one experiment. Each channel is subdivided into 6 connected chambers offering the possibility for biological replicates. The whole setup was placed within an inverted microscope (Axio Observer, Zeiss) equipped with two illumination systems: (i) a precisExcite High-Power LED illumination system (Visitron, Puchheim, Germany) for brightfield imaging. The microscope had a motorized xy-stage and the automated imaging process was controlled using VisiView Premier Image acquisition software (Visitron, Puchheim, Germany). Images were acquired every 20 minutes during one week through a Hamamatsu Orca-ER CCD camera (Hamamatsu, Solothurn, Switzerland).
Data analysis and statistics
The recorded images allowed for survival studies, growth measurements, and fertility assessment. Image processing was performed using Fiji software(Schindelin et al., 2012) (version 2.0.0-rc-15/1.49k). For Survival, we plotted the time point where we observed a dead worm in each condition. Worm areas were measured by processing time-lapse brightfield pictures as follows;
Each frame was first converted to a binary image by applying a threshold to the full stack of time-lapse images and transforming it into a set of binary masks. Each stack of masks was then analysed using the “particle analysis” Fiji plugin, which allows directly extracting area values for each picture in the stack. For fertility assessment, we tracked the sequence where the first egg was laid, and then we probed the time point where the first larva emerged. The experiment was repeated 3 times independently using new batches of media, perovskites and by inverting the position of the conditions in the device channels. We analysed each dataset alone by averaging the sextuplicate of each condition and measuring the standard deviation (S.D.) using One-way ANOVA test followed by a Tukey-Kramer post-hoc test (non-treated vs. treated conditions).