Studying Synaptic Connectivity and Strength with Optogenetics and Patch-Clamp Electrophysiology

Abstract

Over the last two decades the combination of brain slice patch clamp electrophysiology with optogenetic stimulation has proven to be a powerful approach to analyze the architecture of neural circuits and (experience-dependent) synaptic plasticity in such networks. Using this combination of methods, originally termed channelrhodopsin-assisted circuit mapping (CRACM), a multitude of measures of synaptic functioning can be taken. The current review discusses their rationale, current applications in the field, and their associated caveats. Specifically, the review addresses: (1) How to assess the presence of synaptic connections, both in terms of ionotropic versus metabotropic receptor signaling, and in terms of mono- versus polysynaptic connectivity. (2) How to acquire and interpret measures for synaptic strength and function, like AMPAR/NMDAR, AMPAR rectification, paired-pulse ratio (PPR), coefficient of variance and input-specific quantal sizes. We also address how synaptic modulation by G protein-coupled receptors can be studied with pharmacological approaches and advanced technology. (3) Finally, we elaborate on advances on the use of dual color optogenetics in concurrent investigation of multiple synaptic pathways. Overall, with this review we seek to provide practical insights into the methods used to study neural circuits and synapses, by combining optogenetics and patch-clamp electrophysiology.

Keywords: CRACM; brain slices; connectivity; dual color optogenetics; optogenetics; patch-clamp electrophysiology; plasticity; synapses.

Figure 1

Methods to assess synaptic connectivity in neural circuits. (A) Ionotropic synaptic transmission assessed with optogenetic approaches, using single pulse stimulation. Example of GABA_AR synaptic transmission. Top: Using a patch pipette solution with a low-chloride ion concentration results, at slightly depolarized voltage clamping holding potentials such as −50 mV, in outward GABA_AR transmission (i.e., Cl⁻ influx through GABA_ARs) after optogenetic stimulation of GABAergic axons. Bottom: Instead, using a patch pipette solution with a high-chloride ion concentration results in inward GABA_AR transmission (i.e., Cl⁻ efflux through GABA_ARs) after optogenetic stimulation of GABAergic axon terminals. (B) Metabotropic synaptic transmission assessed with optogenetic approaches, stimulating with trains of pulses. Example of GABA_BR synaptic transmission. Top: When using a patch pipette potassium-based solution (no cesium), trains of optogenetic stimulation, at slightly depolarized voltage clamping holding potentials such as −50 mV, can induce slow kinetic outward GABA_BR transmission (i.e., K⁻ efflux through GABA_BR-regulated potassium channels like GIRKs. Bottom: When using a patch pipette solution with a high cesium internal content, the cesium ion blockade of potassium channels like GIRKs interferes with the ability to detect metabotropic GABA_BR signaling after optogenetic GABA terminal stimulation. (C) Whether an optogenetically evoked synaptic input onto a patched neuron is monosynaptic can be determined with the pharmacological combination of tetrodotoxin (TTX) and 4-aminopyridine (4-AP). In the presence of TTX, monosynaptic feedforward input onto the patched neuron should be salvageable by 4-AP administration. (D) Instead, if optogenetic stimulation indirectly leads to an observed synaptic current into the patched cell (e.g., an outward current due to disynaptic inhibition), such a polysynaptic current will not be observed after TTX + 4-AP application. (E) Optogenetic stimulation may lead to combinations of mono- and polysynaptic activity onto the patched neuron, which can also be separated via the combination of TTX + 4-AP.

Figure 2

Methods to assess synaptic strength and properties in neural circuits. (A) Using optogenetic stimulation at excitatory synapses AMPA-NMDA ratios can be calculated to gauge synaptic strength, by normalizing AMPAR-mediated to NMDAR-mediated currents (I-AMPAR; I-NMDAR). This can be done in different ways, requiring conditions to separate the two currents from each other. A₁: Clamping the cell at −60 mV allows for determination of I-AMPAR at peak values (without I-NMDAR contributions due to its block by Mg²⁺ ions under these conditions, top left), and then clamping the cell at +40 mV allows for determination of I-NMDAR peak values (alleviating Mg²⁺ block, bottom left). In this case, the I-NMDAR value will require pharmacological separation from I-AMPAR. A₂: Instead of using pharmacological separation between I-AMPAR and I-NMDAR at +40 mV, it is also possible to evaluate I-NMDAR at a delayed timepoint after stimulation (e.g., 100 ms), where I-AMPAR has decayed, but I-NMDAR has not due to its slower kinetics. A₃: I-AMPAR and I-NMDAR can also both be taken, at peak values, at +40 mV requiring pharmacological isolation of one of the two components to separate them. (B) Optogenetic interrogation of input-specific AMPAR subunit composition (changes) can be done by assessing current/voltage (I/V) relationships of the receptor. Top: Typical AMPARs, containing RNA-edited GluA2 subunits, conduct equally well inwardly (i.e., net cation influx into the cell) as they do outwardly (i.e., net cation efflux out of the cell). Thus, with a reversal potential at 0 mV, those AMPARs would conduct about 1.5× (60/40) more inward current at −60 mV compared to +40 mV. Bottom: GluA2-lacking AMPARs (which can conduct calcium) instead exhibit inward rectification (i.e., larger inward currents than outward currents) mainly due to polyamine block of the receptor at depolarized membrane states under which outward currents would occur. (C) Optogenetics can be used to evaluate the synaptic paired pulse ratio (PPR), often indicative of presynaptic release properties. PPR is calculated by giving two (or more) pulses in quick succession, generating two separate postsynaptic currents, and dividing the amplitude of the second by the first response. PPRs can reveal whether synapses are facilitating (paired pulse facilitation; PPF; top) or depressing (paired pulse depression; PPD; bottom). (D) The coefficient of variance (CV), often reformulated as 1/CV²⁺ for practical reasons, can be informative about presynaptic properties such as release probability and number of release sites. CV itself is calculated by dividing the standard deviation of postsynaptic amplitudes by the mean of the amplitudes. (E) Under conditions where ACSF calcium ion concentration is replaced by an equimolar (or higher) concentration of strontium ions, synaptic transmission becomes more asynchronous and ‘quantal-like’. Optogenetic synapse stimulation under strontium conditions allows approximation of the quantal size at specific synapses.

Figure 3

Dual color optogenetics to assess properties of multiple synapses in a neural circuit. (A) When combining expression of a blue light sensitive opsin (e.g., ChR2 or Chronos) with a more red-shifted opsin (e.g., Chrimson) in distinct presynaptic sources, it becomes possible to stimulate separate synaptic inputs in a slice, by stimulating the tissue with blue light wavelengths vs. red-shifted ones. (B) Control experiments are required to assess the extent of spectral cross over. Left: A situation where only the red-shifted opsin (Chrimson) has been expressed, and the tissue is stimulated with either blue or orange wavelengths. Orange wavelengths indeed activate the axon terminals causing synaptic responses. However, medium to strong blue light irradiance will also (unintentionally) stimulate the red-shifted Chrimson. Instead if low irradiance blue light is used, this does not sufficiently activate Chrimson and no ‘off-target’ synaptic transmission occurs. Right: The opposite scenario where only the ‘blue light’ opsin is expressed (Chronos). Typically orange light stimulation will not activate this opsin and no ‘off-target’ synaptic transmission occurs. Instead, the opsin has sensitivity to blue light, even with low irradiance. Thus, with low blue light stimulation intensities, spectral crosstalk can be minimized. (C) An alternative approach to separate contributions of (left) blue opsin-expressing vs. Chrimson-expressing nerve terminals, without minimizing stimulation intensities. Right: A prolonged pulse of orange light (~605 nm) will typically activate (pulse onset) but then subsequently inactivate (during pulse) Chrimson-expressing nerve terminals. Thus, if at the offset of the orange light pulse (with Chrimson-terminal inactivation) a blue light pulse is given, it activates ‘blue opsin’ expressing terminals without concurrent contributions of the (blue-light sensitive) Chrimson expressing ones.