Towards photoacoustic neuroimaging in mice: a chemigenetic approach for acoustogenic calcium sensors

Photoacoustic imaging is an attractive imaging modality for in vivo applications such as neuroimaging, able to image large fields of view, deep in tissue, at high resolution. However, robust molecular reporters are required for this modality, to label specific targets of interest and visualise dynamic biochemical events. Various genetically-encoded or synthetic reporters have been implemented, but suffer from poor optical properties or lack of labelling specificity. Here, we report the development of novel “chemigenetic” contrast agents and calcium sensors for photoacoustic imaging, based on synthetic dyes and the self-labelling HaloTag protein. I designed and synthesised new photoacoustic reporters based on quenched near-infrared rhodamine-like derivatives, tailored for use with HaloTag to exploit the open-closed equilibrium of the lactone ring. These were characterised in vitro for high photoacoustic signal and turn-on upon binding to HaloTag. The most promising ligands were used for the design of calcium indicators, with the HaloTag-based HaloCaMP protein, which show an absorption-modulated change in calcium-dependent photoacoustic signal. Cell permeability of these ligands was confirmed by efficient labelling of live cells cytosolically expressing HaloTag. Our resulting “acoustogenic” calcium sensors show large turn-ons in response to calcium, up to 8-fold, high photoacoustic signal and photostability, outperforming existing far-red sensors for this modality such as NIR-GECO1 in tissue-mimicking phantoms. Ex vivo brain slices with neuronal expression of HaloTag, labelled with our ligands, could be consistently visualised via photoacoustic tomography. In vivo labelling experiments in mice, however, showed mixed results with the occasional specific labelling of HaloTag-expressing neurons suggesting that the bioavailability of our ligands needs to be improved. In summary, we developed a novel approach for the design of photoacoustic reporters, with the first chemigenetic probes for this modality. Our acoustogenic dyes and first-generation photoacoustic calcium sensors, show superior in vitro performance to existing sensors, and we can label mice brain tissues to produce a localised, strong PA signal. To reach our long-term goal of whole-brain neuroimaging in mice, we have started protein engineering to improve the dynamic range, calcium binding affinity, and used established methods to assess dye bioavailability to optimise this key parameter for future in vivo applications. In general, with this hybrid design, we hope to stimulate photoacoustic probe and sensor development which can enable this powerful modality to uncover its true potential in the field of neuroimaging and beyond.