Photobleaching and Photoactivation

A few months ago, we purchased a Rapp Optoelectronics galvo-scanning system, along with 405 and 473 nm lasers from Vortran Laser Technology, to provide a photoactivation and photobleaching system for our high speed widefield system. This system is capable of photoactivating or photoconverting any protein that is switched by 405nm light (which is most of them) and photobleaching GFP, while simultaneously acquiring in the GFP, RFP, and Cy5 channels. The entire system is controlled through Micro-Manager.

Today, with help from Nico Stuurman, we took it out for a spin. Nico provided Drosophila S2 cells with either mEos2 or GFP-tubulin as well as help getting everything set up correctly. Here are two videos demonstrating what it can do. Both were acquired on a widefield microscope with a 100x / 1.4 NA objective.

Photoconversion of mEos2-labeled tubulin in the spindle of a Drosophila S2 cell. The video is sped up 20-fold from real time.
Photobleaching of GFP-labeled tubulin in a Drosophila S2 cell. First, a GFP aggregate is bleached, which rapidly recovers. Second, nearby microtubules are bleached, which do not recover over the same time scale. The acquisition doesn't stop during the bleaching, so you can see the bright flash as the bleaching laser is turned on. The video is sped up 1.5x from real time.

We also learned some interesting things about photoconverting mEos2. At high peak intensities, the protein bleaches instead of converting, so to get good conversion the trick is to use low intensities for longer periods of time. The movie above was converted using 0.5 mW laser power (in a ~1 μm spot), rastered over the area 12 times, for a total exposure time of about 1 second. Higher powers for shorter periods lead to bleaching, so there seems to be some nonlinear process at work.

GFP bleaching in the second video was done with 0.5 mW illuminating a single spot for 500 ms. For bleaching larger areas or more rapidly, you will likely need more power. At higher power levels the 405 nm laser bleaches mEos2 very efficiently, so it seems likely that it could be used to bleach other fluorescent proteins as well, though we have not tested this.

In all cases, laser powers refer to the power specified in the laser software. The actual power reaching the sample is about 20% of the power out of the laser.