Bioluminescent Genetically Encoded Glutamate Indicator for Molecular Imaging of Neuronal Activity

Background

The development of optical biosensors allows scientists to study neuronal activity by investigating changes in membrane voltage or neurotransmitter concentrations. However, most of these sensors are based on fluorescence and thus require an excitation light source. This presents some problematic constraints on optical biosensor technology, for example (i) limited imaging depth in the brain due to light scattering, (ii) cell damage by the laser beam, (iii) need for invasive hardware implantation such as fibres for excitation, and (iv) photobleaching of fluorescent indicators. One solution to circumvent these problems is the use of bioluminescence (Fig. 1B).

Bioluminescence, or “biologically” induced light, is produced by enzymes such as luciferase, which is found in fireflies, beetles, worms and marine creatures. Luciferases oxidize their substrate luciferin or coelenterazine (CTZ) to a luminescent molecule (coelenteramide) leading to the emission of light (1). In recent years, researchers have developed numerous synthetic luciferase variants. These include split luciferases, which are “turned on” when two halves come together, for example in response to an analyte. Although bioluminescence is already used for imaging purposes, such as visualizing calcium dynamics (2), no bioluminescent neurotransmitter indicator has been described to date.

In this preprint, the authors aimed to develop a genetically encoded bioluminescent neurotransmitter indicator. Petersen et al. chose glutamate as the analyte, since it is one of the most abundant neurotransmitters in the central nervous system, and is important for movement, behaviour, pain perception and mental health. Furthermore, the previously developed glutamate-sensitive construct SuperGluSnFr, which is based on the glutamate binding protein Glt1, is a perfect example of a sensor construct compatible with a split-luciferase, and demonstrates the wide applicability of the genetic optimisation techniques used in this study (3) (Fig. 1A).

Results

To develop a bioluminescent genetically-encoded neurotransmitter indicator for glutamate, Petersen et al. took a multistep screening approach. Initially, they designed three constructs with three previously described split luciferase variants, which flank the truncated periplasmic Glt1. The constructs have a N-terminal secretion signal domain and a C-terminal membrane anchor (Fig. 1A).

These three variants – named BLING (BioLuminescent Indicator of the Neurotransmitter Glutamate) 0.1, 0.2 and 0.3 – were transfected into HEK cells for testing. After sequential addition of luciferase substrate CTZ and glutamate to the extracellular milieu, the bioluminescent signal was measured with a plate reader. BLING 0.2 was found to be the brightest sensor, and thus its structure was further optimised by varying the 3-amino acid segment around the Glt1 region. For this linker optimization, Petersen et al. created a library consisting of about 400 variants with different linkers containing A, S or P (Ala, Ser, Pro). Screening these constructs revealed a variant they named BLING 1.0 as the sensor with the brightest response to 1 mM glutamate (Fig. 2B), which could efficiently report glutamate concentrations in a dose-dependent manner in a 96-well assay format.

Using live cell bioluminescence microscopy, BLING 1.0 reported physiologically relevant conditions (as low as 1 µM glutamate) at the single cell level. Furthermore, in comparison with established fluorescent neurotransmitter indicators such as iGLuSnFr and GcaMP6m, BLING1.0 as well as BLING0.2 showed a significantly better response to 1 mM glutamate in bulk measurements.

Why we like this preprint

This preprint is a wonderful example of how to use pre-existing resources to create something new. It’s great to see how known constructs can be combined and optimised to help evolve an underdeveloped technique: bioluminescence; and with a pathologically important application: deep brain imaging. Furthermore, BLING 1.0 has multiple possible applications beyond recording neuronal activity, for example in drug screening of fluorescent compounds, where screening with a fluorescent readout is incompatible, or as activators of light-sensitive proteins, with future applications in the treatment of neurological disorders. Above all, it can serve as a model for the development of various other bioluminescent neurotransmitter indicators that use other analytes, such as glycine or GABA. It is also worth mentioning here that the authors have made BLING 1.0 readily available to the scientific community – so if you are interested you can test it in your own setting.

Questions to the authors

1. Have you already tried to use BLING in vivo?
2. You describe the use of Furimazine and hCTZ. Are these two different molecules? Which gives the best signal and is the most biocompatible?
3. What is the time resolution compared to other methods such as iGluSnFr, Dlight or GRAB-DA?
4. Is this bioluminescence signal strong enough to be observed through several millimetres/ centimetres of tissue?
5. BLING 1.0 outperforms the fluorescent reporters iGluSnFr and GcaMP6m, which perform less well in bulk measurements. Do you also expect BLING 1.0 to be superior at the single cell level? Is bioluminescence commonly stronger than fluorescence emission? Is the weak signal observed in the fluorescent probes due to their poor adaption to the HEK cell system used (quantum yields are comparable to fluorophores)?
6. What was the structural change between BLING 0.2 and BLING 1.0? Why do you suspect this improved the activity so much?

References

• Syed, Aisha J., and James C. Anderson. “Applications of bioluminescence in biotechnology and beyond.” Chemical Society Reviews(2021).
• Granatiero, Veronica, et al. “Using targeted variants of aequorin to measure Ca2+ levels in intracellular organelles.” Cold Spring Harbor Protocols1 (2014): pdb-prot072843.
• Hires, Samuel Andrew, Yongling Zhu, and Roger Y. Tsien. “Optical measurement of synaptic glutamate spillover and reuptake by linker optimized glutamate-sensitive fluorescent reporters.”Proceedings of the National Academy of Sciences11 (2008): 4411-4416.

Tags: bioluminescence, biosensor, brain imaging, glutamate, glutamate sensor, luciferase, neurology, neuronal activity, optical biosensor

doi: https://doi.org/10.1242/prelights.29903

Read preprint (No Ratings Yet)