Two photon spectroscopy and microscopy of the fluorescent flavoprotein, iLOV

Abstract

LOV-domains are ubiquitous photosensory proteins that are commonly re-engineered to serve as powerful and versatile fluorescent proteins and optogenetic tools. The photoactive, flavin chromophore, however, is excited using short wavelengths of light in the blue and UV regions, which have limited penetration into biological samples and can cause photodamage. Here, we have used non-linear spectroscopy and microscopy of the fluorescent protein, iLOV, to reveal that functional variants of LOV can be activated to great effect by two non-resonant photons of lower energy, near infrared light, not only in solution but also in biological samples. The two photon cross section of iLOV has a significantly blue-shifted S0 → S1 transition compared with the one photon absorption spectrum, suggesting preferential population of excited vibronic states. It is highly likely, therefore, that the two photon absorption wavelength of engineered, LOV-based tools is tuneable. We also demonstrate for the first time two photon imaging using iLOV in human epithelial kidney cells. Consequently, two photon absorption by engineered, flavin-based bio-molecular tools can enable non-invasive activation with high depth resolution and the potential for not only improved image clarity but also enhanced spatiotemporal control for optogenetic applications.

Live‐Cell Copper‐Induced Fluorescence Quenching of the Flavin‐Binding Fluorescent Protein CreiLOV

“…Therefore, the larger the lifetime of the donor protein, the higher the sensitivity of the FRET‐based biosensor. The fluorescence lifetimes of several FbFPs were reported and generally ranged from >3 to 6 ns . For comparison, GFP‐based fluorescent proteins rarely have lifetimes that are >4 ns .…”

“…For comparison, GFP‐based fluorescent proteins rarely have lifetimes that are >4 ns . Here, we used time‐correlated single‐photon counting to measure an average fluorescence lifetime for CreiLOV of 4.50±0.09 ns (Figure A, B), which is a little shorter than iLOV (5.27 ns), but longer than photostable iLOV (phiLOV2.1, 3.36 ns) and free FMN (4.38 ns) …”

Photoactivatable CaMKII induces synaptic plasticity in single synapses

“…Second, while glutamate uncaging experiments require the removal of extracellular Mg 2+ or depolarization of neurons to release Mg 2+ from NMDARs 2 , paCaMKII can be activated in physiological conditions such as the mouse brain in vivo. Third, the wavelength window of MNI-glutamate uncaging is up to 800 nm 2 , while that of paCaMKII is up to~980 nm (i.e., FMN absorption) 32 . Because a longer wavelength has a superior penetration depth, this is advantageous for sLTP induction in deep tissues.…”

“…Because a longer wavelength has a superior penetration depth, this is advantageous for sLTP induction in deep tissues. Fourth, since the two-photon cross section of LOV2 (0.5-0.9 GM in 800-900 nm) 32 is higher than that of MNI-glutamate (0.06 GM at 720 nm, the most widely used wavelength for efficient photolysis) 58,59 , weaker laser power (i.e., lower cell toxicity) can be used for sLTP induction. Together, these advantages facilitate the manipulation of synaptic plasticity under physiological conditions with low phototoxicity.…”

“…(a.a.), which is smaller than most other sensors (i.e., CRY2/CIBN 498/170 a.a., Dronpa 257 a.a., PhyB 908 a.a., and UVR8 124 a.a.). Furthermore, to our knowledge, while the two-photon cross sections of Dronpa, PhyB, and UVR8 have not been determined, it is known that the light-absorbing cofactor of LOV2, flavin mononucleotide (FMN), has a relatively large two-photon cross section (0.5-0.9 GM in 800-900 nm) 32 . It is larger than that of flavin adenine dinucleotide, the light-absorbing cofactor for CRY2 (0.02-0.04 GM in 800-900 nm) 33 .…”