There are electrophysiological and optical methods to investigate neuronal activity in vivo – both of them have their advantages and drawbacks. If you want to record with high temporal precision and deep inside the brain your method of choice is sticking electrodes deep into the brain and record LFPs. Detailed analysis of these LFPs then reveals distinct patterns of neuronal spiking activity that can be segregated into different sources (your individual neurons), but you are practically blind to 1. where you are and then 2. which cells you are recording from. If you want to “see” the neurons you are recording from you will probably prefer calcium imaging methods. Activity of single neurons loaded with a synthetic dye or expressing genetically encoded calcium indicators (GECIs) can be observed and the specificity can additionally be increased if a subpopulation of cells (e.g. specific interneuron types) expresses a fluorescent marker protein. This only accounts for superficial layers of the cortex or the olfactory bulb since even multi photon microscopy doesn’t allow imaging deeper cortical – let alone subcortical structures (side note: there are these papers where people implant prisms and fibers to “dig” deep into the brain – they are all massively invasive though). So what do you do if you want to identify activity of distinct projections from one regions to another deep in the brain? Well, chances are you are quite lost. Calcium imaging doesn’t reach these depths and good luck identifying/segregating distinct LFP signatures that serve your analysis purpose!
In their Cell paper, Gunaydin et al. combine a range of interesting techniques to investigate neuronal dynamics in the ventral tegmental area (VTA) to nucleus accumbens (NAc) projections during social interactions in mice. They use an optical fiber that excites and records GCAMP5g signals in the VTA and NAc. Specifically they conduct an experiment where they inject a floxed GCAMP5 construct in the VTA of TH(Tyrosin-Hydroxylase)-cre mice and image axonal projections with an optical fiber implanted in the NAc (a technique they call photometry). Although this approach is essentially “blind” to the specific anatomy of the field in view at the fiber ending, it allows recording of projections – and importantly only projections from one specific region to another to be examined in the behaving animal on a millisecond time scale. They combine this approach in a separate set of experiments with a optogenetical approach: by injecting a WGA-cre virus in the (postsynaptic) cells in the NAc they can retrogradely label only those cells that are connected to the NAc in the VTA and that – in their experiment – express floxed Halo- or Channelrhodopsins. Thereby they can either inhibit or activate specific projections from the VTA to NAc without much side effects on other projections from the VTA (check this Südhof paper for more WGA-cre madness). Their last set of experiments makes use of an interesting tool for specific (optical) control of GPCRs -in this case D1Rs: They use a method introduced in a 2009 Nature paper to engineer basically a ChR with its intracellular domains replaced by the intracellular domains of the D1R (Dopamine 1 receptor). Shining light on cells expressing this mutant GPCR leads to rising intracellular cAMP levels as would be expected from the activation of D1Rs in the cell membrane. Taken together their paper uses a range of very useful techniques to investigate circuit dynamics in vivo. As they conclude: “This ability to directly measure activity of projections between brain regions provides a new source of data on dynamics of information flow.”
Video below is Supplementary Video 4:Fiber photometry recording from VTA-NAc projections during exposure to a novel social target. Red bar in upper-left corner indicates dF/F synchronized with the animal’s behavior. Note the increases in GCaMP fluorescence as the test mouse actively investigates the target mouse [...]