1. Report this comment

    Andrew Belmont said:

    I applaud this editorial about phototoxicity. This is a problem my laboratory has struggled with for many years now. We are studying long-range interphase chromosome movements and we have found through painful experience that the mechanism underlying this phenomenon is exquisitely sensitive to light.

    Indeed, it took one of my students more than 2 years to recognize and then overcome this problem, as partially described in a 2006 Current Biology publication (Long-range directional movement of an interphase chromosome site. Chuang CH, Carpenter AE, Fuchsova B, Johnson T, de Lanerolle P, Belmont AS. Curr Biol. 2006 Apr 18;16(8):825-31. PMID: 16631592 [PubMed – indexed for MEDLINE]).

    Briefly, we were looking at the movement of a tagged chromosomal site from the nuclear periphery to the nuclear interior after targeting a transcriptional activator to this site. We discovered this effect based on statistical analysis of fixed cell populations. When we went to the microscope the student spent years trying to visualize the actual motion with no success. To make a long story short, it turned out that a single exposure from our microscope system killed this long-range movement, despite the cells continuing to progress through the cell cycle under the same imaging conditions with hundreds of exposures.

    Using the change in nuclear distribution from periphery to interior in fixed cell populations as our assay, we were able to produce a 400 fold reduction in this phototoxicity problem simply by changing our filters and light sources. Surprisingly the solution was to introduce inexpensive glass filters on top of our interference filters as a solution- presumably there are small imperfections leading to leakage of UV and near UV light with standard interference filters. This is described in the supplemental data of the cited paper. Fortunately we had a simple assay that we could compare to fixed cell populations that never saw light.

    For us this entire and very interesting biological phenomenon of directed interphase chromosome movements disappeared with phototoxicity under what most people would consider minimal irradiation.

    Several years ago we switched to the Applied Precision OMX platform, where with laser monochromatic excitation and EMCCD cameras we have been able to significantly decrease the phototoxicity effect and extend our observations. We are about to submit a new manuscript describing chromosome movements up to 5-6 microns directed towards nuclear speckles when we transcriptionally activate a Hsp70 transgene. But even in this study we still see this movement decrease in frequency as we increase our exposure totals. We suspect investigators with conventional systems would miss this long-range chromosome movement entirely.

    Finally, I like your suggestion about measuring cell cycle parameters as a simple test of phototoxicity. Many people using live cell imaging use the ability of cells to progress through mitosis as a quality check, but this is very different from showing that there is no actual perturbation in cell cycle parameters. One assay my laboratory has explored is using the time delay between the end of S phase and mitosis as an assay for phototoxicity. We found that before an overt block in mitosis was observed we could observe a lengthening of the G2 period, as measured from multiple cells in live experiments using a GFP-PCNA expressing cell line.

    But I agree with you that this is not sufficient. Ideally, what I like to see in our experiments is some type of observable- directly related to the scientific phenomenon we are studying- that we can measure in fixed populations that have never been exposed to light. We can then compare this to the same observable as measured in our live cell imaging experiments.

    Thank you for highlighting this problem.