Summary: In this paper, they are aiming to make a fast, sensitive and specific optically-controllable protein interaction. At first, they note that there are a number of photoactive proteins that have been engineered for use as photoswitchable actuators, including the LOV2 domain of phototropin 1, Vivid (VVD) and a bunch of others (cryptochrome 2, FKF1, UVR8, EL222). Point blank, I know nothing of any of these so I really can't comment on how this fits in with the field (maybe I'll try to come back to this later).
In this work, they focus on Vivid (VVD), from Neurospora crassa. It's one of the smallest proteins, uses FAD as a chromophore (which is ubiquitous in eukaryotic cells), and homodimerizes when blue light is applied. However, it's got drawbacks.
- It homodimerizes. Imagine you want to make gene A active when you apply light. You split gene A in half (A* and A') and make and express two fusion proteins: VVD-A* and VVD-A'. When you shine blue light, you're as likely to match A* with A' (its correct partner) as with A* (not its correct partner).
- It's slow. After you stop the blue light, it takes 3-4 hours before the dimers separate back into monomers.
They set out to fix these two problems. First, they made two VVD proteins, one with a modified interface containing two new arginine residues (and thus positively charged), and one containing a new aspartate and glycine residue (and thus negatively charged). They call these pMag and nMag, and tag each with half of a split firefly luciferase. pMag on its own doesn't respond to light in COS-7 cells, and neither does nMag, but when coexpressed, they generate at least a 10-fold increase in luciferase luminescence when blue light is applied.
The interaction is now specific, so they then turn to speed. nMag and pMag dissociation has a 2-hour time constant. Based on previous work suggesting residues that affect the photocycle of VVD, they make two isoleucine to valine changes in nMag and pMag (producing nMagFast2 and pMagFast2), which reduces the dissociation time constant to 25 seconds.
However, this reduced the overall dimerization as well, so they made a few more modifications in another area expected to affect the photocycle, which they term nMagHigh1 and pMagHigh1. nMagHigh1-pMagHigh1 are slower to dissociate than the original VVD, but produce a much stronger signal (the data suggest much more dimerization in the dark, and in the light).
At this point, I got a little confused, because I immediately jumped to (luminescence in blue light)/(luminescence in dark) as the appropriate metric, and they instead just quote numbers about luminescence in blue light. I agree this is important since it would be nice to not have to fry my cells with blue light in order to drive some protein activity, but it seems to me like ON:OFF luminescence ratio is a far more important parameter that somehow they never show in a plot.
I suppose thus the table of figures of merit I would like to see for various pairs would include: 1) ON:OFF ratio, 2) time constant 3) basal dimerization magnitude in dark.
After testing a handful of their engineered pairs, they settle on pMagFast2-nMagHigh1 as a fast (< 1 minute time constant) and reasonably sensitive and specific pairing (eyeballing it from their plot, its ON:OFF ratio is probably about 10, though its dark dimerization is at least several-fold higher than for pMag-nMag or nMagFast2-pMagFast2 (hard to tell based on the plot).
To test their constructs, they express nMagHigh1-GFP-CAAX and pMagFast2-iRFP test in COS-7 cells. The CAAX tag localizes nMagHigh1 to the membrane, and brief light irradiation makes the iRFP move to the membrane on a timescale of ~ 1.5 seconds, with a dissociation timescale of ~7 seconds. As a control, they show that pMagFast2 does not localize to the membrane in the absence of nMagHigh1.
They then show that they can use a split PI3 kinase. First, they show that by again tethering one of their magnets to the membrane, the other one moves to the membrane. Second, by imaging AktPH (a "PI3K biosensor") in a third color channel, they show that it also moves to the membrane after blue light induction, suggesting a functional PI3 kinase.
They then show that focal induction of PI3 kinase yields local actin polymerization (as measured by lifeact-mCherry [lifeact is apparently a peptide that binds to F-actin]), formation of lamellopodia, and 'ruffles' in the membrane. Continuous light at a single point near the membrane for 10-20 minutes induces lamellopodia and expansion of the cell boundary nearby (and contraction at the other side of the cell).
Overall, very cool technique.
Questions I still have:
- How leaky is this? Their pairs that drive luciferase the most during blue light also drive luciferase luminescence in the dark at levels higher than some other variants (e.g. pMag-nMag) when illuminated! What practical consequence does this have? E.g. how much extra PI3 phosphorylation is happening in the dark?
- Could you use this to drive gene expression in a graded manner? (e.g. could I precisely control exactly how much transcription I want just by varying the optical intensity?) Update: looks like this has been done in 2012 using VVD to drive dimerization of the gal4 dna-binding domain monomer.
- In the paper, they show 8 mutant proteins (4 p-type, 4 n-type) based on changes at 6 positions, and they test 8 pairs. Is there more potential for larger-scale optimization? What if you could exhaustively screen large libraries of mutants for the three qualities I listed above? (ON:OFF ratio, dark dimerization,
- Is wavelength tuning possible? Or does the use of FAD significantly restrict the wavelengths possible?