GPCR-Based Dopamine Sensors-A Detailed Guide to Inform Sensor Choice for In vivo Imaging

Abstract

Understanding how dopamine (DA) encodes behavior depends on technologies that can reliably monitor DA release in freely-behaving animals. Recently, red and green genetically encoded sensors for DA (dLight, GRAB-DA) were developed and now provide the ability to track release dynamics at a subsecond resolution, with submicromolar affinity and high molecular specificity. Combined with rapid developments in in vivo imaging, these sensors have the potential to transform the field of DA sensing and DA-based drug discovery. When implementing these tools in the laboratory, it is important to consider there is not a 'one-size-fits-all' sensor. Sensor properties, most importantly their affinity and dynamic range, must be carefully chosen to match local DA levels. Molecular specificity, sensor kinetics, spectral properties, brightness, sensor scaffold and pharmacology can further influence sensor choice depending on the experimental question. In this review, we use DA as an example; we briefly summarize old and new techniques to monitor DA release, including DA biosensors. We then outline a map of DA heterogeneity across the brain and provide a guide for optimal sensor choice and implementation based on local DA levels and other experimental parameters. Altogether this review should act as a tool to guide DA sensor choice for end-users.

Keywords: behavior; dopamine; drug screening; fiber photometry; fluorescent biosensor; genetically encoded; in vivo fluorescent imaging; neuromodulator; pharmacology.

Figure 1

In vivo applications of dLight1 and RdLight1 dopamine (DA) sensors in mice. Graphs show normalized fluorescent responses (dFF). (a) dLight1 variants were validated in multiple imaging and experimental modalities: (i) dual-color fiber photometry of nucleus accumbens (NAc) cells using the red-shifted calcium sensor jRGECO1a [77] and local DA release using dLight1.1 following unpredictable shock exposure (dotted line); (ii) fiber photometry imaging of DA release using dLight1.1 in the NAc following optogenetic stimulation (5–20 Hz, 2 seconds) of DA cell bodies in the ventral tegmental area (VTA) using the red-shifted opsin ChrimsonR; (iii) two-photon imaging of DA release (dLight1.2) (top) across heterogeneous sites in motor cortex (M1/M2) across 17 μM large regions of interest (= red square ROIs, bottom), here showing ROIs responsive to locomotion/reward expectation vs. rest (green vs. orange) at the Go cue (vertical line) and increases in fluorescence upon reward delivery (dark green) but not reward omission (light green) (population data) (middle). (b) RdLight1 validations: (i) dual-color fiber photometry of VTA terminals in the NAc using the green axon-targeted calcium sensor axon-GCaMP6f [52] and local DA release using RdLight1 following unexpected, audible reward deliveries (dotted line) and (ii) fiber photometry imaging of DA release using RdLight1 in the NAc following optogenetic stimulation (1–32 pulses; 2 seconds) of DA cell bodies in the VTA using the green opsin ChR2.

Figure 2

Distribution of norepinephrine (NE) (left hemispheres) and dopamine (DA) (right hemispheres) neurotransmitter levels as measured by enzyme isotope biochemistry assays in micropunches of the rat brain. The red fill pattern indicates the percentage of NE and DA relative to the highest measured value. NE: very high = greater than 64.0; high = 48.1–64.0; moderate = 32.1–48.1; low = 16.0–32.1; very low= less than 16 ng/mg protein. DA: very high = greater than 83.6; high = 62.7–83.6; moderate = 41.8–62.7, low = 20.9–41.8; very low= less than 20.9 ng/mg protein. Of note, the distribution pattern of DA and NE in the dorsal striatum and NAc is dramatically distinct—with DA content being very high (80–100%), while NE content is very low (less than 20%). Au1: primary auditory cortex; BNST: Bed nucleus of the stria terminalis; CPu: caudate-putamen (= dorsal striatum); CRB: cerebellum; DR: dorsal raphe; ENT: entorhinal cortex; GPe: globus pallidus externus; HPC: hippocampal formation; Hyp: Hypothalamus; LS: lateral septum; M1: primary motor cortex; mPFC: medial prefrontal cortex; MS: medial septum; NAc: Nucleus accumbens; OT: olfactory tubercle; PVT: paraventricular nucleus of the thalamus; SN: substantia nigra; SNc: SN compacta; TeA: temporal association cortex; VP: ventral pallidum; VTA: ventral tegmental area. For detailed brain region annotations see the original image source [130]. This Figure was modified with permission from Björklund & Hökfelt (1984), Handbook of Chemical Neuroanatomy, Vol. 2–Part 1. © Elsevier Science Publishers B.V. (1984), Amsterdam, Netherlands [130].

Figure 3

Graphical representation of previously reported basal dopamine concentrations measured via intracerebral microdialysis experiments performed in rats. Coronal brain sections from AP + 3.24 mm to AP -8.04 mm (relative to bregma) are depicted. The colored circles indicate the reported positions of the microdialysis probes and the fill color of the circles is color-coded to represent the values of the reported basal dopamine concentrations. Of note, the striatal basal dopamine concentrations are amongst the largest in rat brain (brain sections AP + 1.08 mm and AP + 0.24 mm). Importantly, the basal dopamine levels do not represent phasic dopamine release since they are predominantly measured during baseline at rest and are a result of dialysate collection times in the range of several minutes. For detailed information on the previous studies reported in this analysis please consult Supplementary Table S1. This figure was modified with permission from Paxinos & Watson (2005), The rat brain in stereotaxic coordinates, 5th edition. © Elsevier Academic Press (2005), Burlington, MA 01803, USA [213].

Figure 4

Concentration-response curves to determine affinity (K_d) and dynamic range (dFF_max) of dopamine (DA) biosensors in vitro. (a) In a first step, HEK293 cells are transfected with the DA sensor plasmid of interest. (b) After 2 days of sensor expression, sensor fluorescence is measured in response to DA titrations (10⁻³ to 10³ nM) which are perfused into the bath during time-lapse confocal imaging with 488 nm (dLight1, YdLight1) or 561 nm (RdLight1) light. Time-lapse images of fluorescent cells before and after addition of increasing DA concentrations are obtained and used to calculate the signal to noise ratio (SNR) and average fluorescence intensity (F). (c) dFF is calculated using the following equation: [F(peak)–F(basal)]/F(basal) where F(basal) and F(peak) are the averaged fluorescence intensity of 10 frames before and after, addition of a given DA concentration, respectively. Maximal dFF values for a given concentration are plotted and a single-site-specific binding equation is used to fit the data points and determine the affinity (K_d value) and dFF_max of the sensor for this ligand (titration data shown in panel c are from References [58,59]). For further details on the protocol, see Reference [99].

Figure 5

Sensor properties to consider for sensor choice: (a) Local dopamine (DA) levels can be a useful measure to guide sensor choice based on ligand affinity. It is generally recommended to look at sensor affinity as a first criterion and to match it with the local expected concentration of DA in the brain ROI, using high and very high affinity variants for regions with sparse innervation and medium/low affinity variants for regions with dense innervation; here we classify existing DA sensors based on their affinity category (affinity K_d and dynamic range dFF_max are noted in brackets). (b) Other sensor properties should be considered (see also Table 1 for exact values) and their relative importance will depend on the individual experimental parameters: (i) dynamic range (if possible at least 250–300%), (ii) molecular specificity (affinity for DA should be far greater than affinity for NE in brain regions with dual DA/NE innervation), (iii) on and off kinetics (should be as short as possible in assays where high temporal resolution is required), (iv) basal brightness (may not matter for fiber photometry but high brightness may improve identification of small cell compartments in 1- and 2-photon imaging), (v) color spectrum (3-colors available; important when multiplexing with opsins, sensors or photopharmacology) and (vi) molecular scaffold (3-scaffolds available; important when multiplexing with (photo)pharmacology: if drug has affinity for the DA receptor scaffold, signal may be affected; this property can however be harnessed in drug discovery experiments where sensor fluorescence can be used as a specific readout of DA receptor subtypes activation in vitro or in vivo).

Figure 6

Molecular scaffold as a double-edged sword for pharmacology and drug discovery: (a) In vitro assays: [Top]: multiplex imaging of drug efficacy at distinct dopamine (DA) receptor subtypes in HEK293 cells expressing either RdLight1 (= DRD1 scaffold) or dLight1.5 (= DRD2 scaffold) in the same culture dish. [Middle]: representative images of HEK293 cells and their average fluorescent responses (dFF) to individual drugs. Scale bars, 10 μm. [Bottom]: Simultaneously measured fluorescence responses of both sensors during bath application of DRD1 drugs: A77636: 100 nM, SCH23390: 10 μM and DRD2 drugs: Quinpirole (Quin): 10 μM, Sulpiride: 400 nM [59]. (b,c) In vivo assays: photometry imaging of drug efficacy at DRD1 (b) and DRD2 (c) receptors in mice expressing dLight1.1 (= DRD1 scaffold) or GRAB-DA1m (= DRD2 scaffold) in the NAc or striatum following optogenetic (opto) stimulation of DA neurons in the midbrain and i.p. injection of drugs: SCH23390: 0.25 mg/kg, Eticlopride: 2 mg/kg (figures were reused with permission from References [58,60]). (d) Pharmacological properties of sensors as a result of their parent receptor scaffold as a double-edged sword. Etic: Eticlopride. “+”: agonists, “−“: antagonists. GBR-12909: DA reuptake inhibitor.

Figure 7

Piloting sensor use in the laboratory can follow these four steps: (a) First, optimal sensor expression in the region or cell of interest should be achieved by optimizing viral injections parameters, often based on protocols previously established in the laboratory for other viral vectors. After a minimum of 2 weeks of incubation, sensor expression can be evaluated as shown here for dLight1.1 in the nucleus accumbens (NAc) [58]. If experiments are expected to last more than 2 months, additional tests can be performed to verify that sensor expression does not induce cell death (e.g., caspase-3 staining) or inflammation (e.g., staining for reactive microglia or astrocyte markers), nor does it affect the basal properties of the cells of interest (membrane potential can be measured in slice physiology). (b) When piloting sensor use in a new brain region with possible sparse innervation, maximal sensor responses should be determined using optogenetic stimulation of dopamine (DA) neurons (e.g., measure NAc DA release using dLight1.1 after ventral tegmental area (VTA) DA neuron ChrimsonR stimulation [58], as shown here). Imaging of DA sensors should be performed using the imaging modality of choice and in vivo kinetics specific for this brain ROI determined by stimulating DA neurons at increasing frequencies (e.g., 5–20 Hz). (c) Pharmacological ligands can be used to validate DA sensors in vivo, e.g., (i) by using reuptake inhibitors e.g., GBR-12909 to increase release or (ii) by using DA receptor antagonists specific to the sensor’s parent receptor to decrease [58] (shown here) or abolish (see Reference [226]) fluorescent signals and verify that transients reflect DA release, e.g., using the DRD1 antagonist SCH-23390. (d) In a last step, sensors can be validated using classical behavioral events known to induce DA release in order to verify optimal detection of DA transients in response to physiological stimuli (e.g., measure dLight1.1 responses in the NAc after free sucrose consumption [58], as shown here). Note that native transients will be several-fold lower than after optogenetic stimulations, see Reference [58]). Users should also verify that sensor on/off- kinetics are compatible with the behavioral task of choice. Further information can be found in Reference [99].