Locus Coeruleus Optogenetic Light Activation Induces Long-Term Potentiation of Perforant Path Population Spike Amplitude in Rat Dentate Gyrus

Abstract

Norepinephrine (NE) in dentate gyrus (DG) produces NE-dependent long-term potentiation (NE-LTP) of the perforant path-evoked potential population spike both in vitro and in vivo. Chemical activators infused near locus coeruleus (LC), the source of DG NE, produce a NE-LTP that is associative, i.e., requires concurrent pairing with perforant path (PP) input. Here, we ask if LC optogenetic stimulation that allows us to activate only LC neurons can induce NE-LTP in DG. We use an adeno-associated viral vector containing a depolarizing channel (AAV8-Ef1a-DIO-eChR2(h134r)-EYFP-WPRE) infused stereotaxically into the LC of TH:Cre rats to produce light-sensitive LC neurons. A co-localization of ~62% in LC neurons was observed for these channels. Under urethane anesthesia, we demonstrated that 5-10 s 10 Hz trains of 30 ms light pulses in LC reliably activated neurons near an LC optoprobe. Ten minutes of the same train paired with 0.1 Hz PP electrical stimulation produced a delayed NE-LTP of population spike amplitude, but not EPSP slope. A leftward shift in the population spike input/output curve at the end of the experiment was also consistent with long-term population spike potentiation. LC neuron activity during the 10 min light train was unexpectedly transient. Increased LC neuronal firing was seen only for the first 2 min of the light train. NE-LTP was more delayed and less robust than reported with LC chemo-activation. Previous estimates of LC axonal conduction times suggest acute release of NE occurs 40-70 ms after an LC neuron action potential. We used single LC light pulses to examine acute effects of NE release and found potentiated population spike amplitude when a light pulse in LC occurred 40-50 ms, but not 20-30 ms, prior to a PP pulse, consistent with conduction estimates. These effects of LC optogenetic activation reinforce evidence for a continuum of NE potentiation effects in DG. The single pulse effects mirror an earlier report using LC electrical stimulation. These acute effects support an attentional role of LC activation. The LTP of PP responses induced by optogenetic LC activation is consistent with the role of LC in long-term learning and memory.

Keywords: dentate gyrus; hippocampus; locus coeruleus; long-term potentiation; norepinephrine; optogenetic; perforant path; short-term potentiation.

Figure 1

(A) Left: dopamine-β-hydroxylase (DBH) staining of locus coeruleus (LC) neurons. Middle: enhanced yellow fluorescent protein (eYFP) expression in TH:Cre LC neurons. Right: merge image of both markers. Arrow indicates an example of occlusion of DBH staining by strong eYFP expression in a TH:Cre neuron. (B) Graph of average co-localization of the channel expression marker eYFP⁺ in LC neurons sampled in three LC sections from 19 rats. (C) Coronal LC sections from two rats illustrating eYFP⁺ expression in putative tanycyte cells and processes at the medial ventricular non-neuronal LC margin. White arrows indicate tanycyte exemplars. Note another example of medial eYFP⁺ cells and processes in the merge panel in (A) under the white arrow stem. Cer = cerebellum with autofluorescing granule cells. IV = adjacent 4th ventricle.

Figure 2

Schematic representations of LC with optrode placements designated by symbols. Blue/yellow symbols represent optrode placements that successfully produced light activation of the recorded cells. Red symbols indicate optrodes that did not produce light activation of recorded cells. Sites at which long-term potentiation (NE-LTP) experiments were conducted are indicated by stars. The dashed lines indicate the outlines of LC in each respective section.

Figure 3

Light activation of anatomically identified LC neurons. (A) Oscilloscope recording showing LC units during baseline and during light pulse train activation. Note activation continuing between pulses and the recruitment of additional units. (B) LC unit waveform cut by Datawave algorithm (inset) and firing pattern plotted as a histogram before, during and after a 10 s light activation. (C) Group histograms for LC and control units during a 5 s light activation protocol. (D) Group histograms for LC and control units during a 10 s light activation protocol. Single asterisk, p < 0.05; triple asterisks, p < 0.001.

Figure 4

Histogram for LC units during the 10 min light activation protocol in 1-min bins. Inset: histogram for the first 10 s of light activation in the 10 min period. Note similarity to previous 10 s light data. *p < 0.05.

Figure 5

LTP of population spike for LTP rats with consistent waveforms (n = 4). Inset shows an average waveform during baseline and at the end of the potentiation period from one subject. Note absence of EPSP slope change. The dotted lines indicate light activation. The first three denote 5 or 10 s periods and the last denotes the onset and offset of the 10 min activation period. The LC optrode was adjusted to optimize unit recording for light activation prior to 0.

Figure 6

Input output curves for increasing perforant path currents prior to light activation LTP baseline and after the conclusion of the experiment. (A) A leftward shift for population spike occurs after the experiment for population spike amplitude once threshold is exceeded. (B) There is no change in EPSP slope input/output over the same period.

Figure 7

Paired pulse effects at three interstimulus interval (ISI). Example paired pulse waveforms for each interval at baseline are included. Bars = 2 mV, 10 ms. (A) At 30 ms ISI prior to the experiment (Baseline) there was no clear inhibition. This was similar at the end of the experiment (Post-Light) and also seen when P1 sizes were matched by lowering the current (Matched). (B) At 70 ms ISI the expected potentiation occurred land this was unchanged at the end of the experiment. (C) At 200 ms ISI the expected inhibition was observed, but again there was no effect on this paired pulse inhibition of the light manipulation and the enduring NE-LTP.

Figure 8

(A) Acute single population spike potentiation when a 50 ms LC light pulse is delivered 40–50 ms prior to the perforant path pulse. This single pulse effect is consistent with conduction times for LC fibers. (B) Examples of modulatory effects of single LC light pulses preceding PP input at each of the four intervals tested. Bars = 2 mV; 5 ms.