Striatal fiber photometry reflects primarily non-somatic activity

Abstract

Fiber photometry recordings are commonly used as a proxy for neuronal activity, based on the assumption that increases in bulk calcium fluorescence reflect increases in spiking of the underlying neural population. However, this assumption has not been adequately tested. Here, using endoscopic calcium imaging in the striatum we report that the bulk fluorescence signal correlates weakly with somatic calcium signals, suggesting that this signal does not reflect spiking activity, but may instead reflect subthreshold changes in neuropil calcium. Consistent with this suggestion, the bulk fluorescence photometry signal correlated strongly with neuropil calcium signals extracted from these same endoscopic recordings. We further confirmed that photometry did not reflect striatal spiking activity with simultaneous in vivo extracellular electrophysiology and fiber photometry recordings in awake behaving mice. We conclude that the fiber photometry signal should not be considered a proxy for spiking activity in neural populations in the striatum.Significance statementFiber photometry is a technique for recording brain activity that has gained popularity in recent years due to it being an efficient and robust way to record the activity of genetically defined populations of neurons. However, it remains unclear what cellular events are reflected in the photometry signal. While it is often assumed that the photometry signal reflects changes in spiking of the underlying cell population, this has not been adequately tested. Here, we processed calcium imaging recordings to extract both somatic and non-somatic components of the imaging field, as well as a photometry signal from the whole field. Surprisingly, we found that the photometry signal correlated much more strongly with the non-somatic than the somatic signals. This suggests that the photometry signal most strongly reflects subthreshold changes in calcium, and not spiking. We confirmed this point with simultaneous fiber photometry and extracellular spiking recordings, again finding that photometry signals relate poorly to spiking in the striatum. Our results may change interpretations of studies that use fiber photometry as an index of spiking output of neural populations.

Elevated fear responses to threatening cues in rats with early life stress is associated with greater excitability and loss of gamma oscillations in ventral-medial prefrontal cortex

“…LFPs reflect aggregate voltage in a region, and given the density of dendritic arbors relative to somas, these changes in voltages in a region may biased towards reflecting afferent inputs to a region, via depolarization and hyperpolarization of dendrites receiving those signals. Support for this perspective was recently provided in models of calcium transient activity with GCaMP sensors in the dorsal striatum (Legaria et al, 2021). Given this, one hypothesis consistent with our data is that vmPFC in ELS animals is lacking relevant feedback on the efficacy of extinction learning, and this information may be provided via gamma band oscillations.…”

Therapeutic deep brain stimulation disrupts movement-related subthalamic nucleus activity in parkinsonian mice

“…Fiber photometry signals may increase in response to greater firing in the population as a whole, or in response to increased synchronization across the population. Recent work comparing neuronal firing and fiber photometry (with simultaneous multi-unit electrophysiology and fiber photometry) in the striatum suggests that while an initial phase of calcium signal correlates well with firing, fiber photometry signals are also more prolonged, which may represent dendritic calcium influx associated with backpropagation ( Legaria et al, 2021 ). For a combination of reasons, fiber photometry signals are slow and often lag changes in spiking.…”

Enhanced food motivation in obese mice is controlled by D1R expressing spiny projection neurons in the nucleus accumbens

“…Recordings of bulk calcium activity can include subthreshold fluctuations in both dendritic and somatic calcium 43 . To test whether increases in D1 SPN calcium activity reflected increased excitatory drive onto D1 SPNs , we performed whole cell recordings of spontaneous excitatory events in the presence of tetrodotoxin (TTX), which allowed us to record quantal release in the absence of action potential-mediated vesicular release (mini excitatory post-synaptic currents, or mEPSCs).…”