Abstract
Dopamine (DA) neurons in the ventral tegmental area (VTA) respond to motivationally relevant cues, and circuit-specific signaling drives different aspects of motivated behavior. Orexin (ox; also known as hypocretin) and dynorphin (dyn) are coexpressed lateral hypothalamic (LH) neuropeptides that project to the VTA. These peptides have opposing effects on the firing activity of VTADA neurons via orexin 1 (Ox1R) or kappa opioid (KOR) receptors. Given that Ox1R activation increases VTADA firing, and KOR decreases firing, it is unclear how the coreleased peptides contribute to the net activity of DA neurons. We tested if optical stimulation of LHox/dyn neuromodulates VTADA neuronal activity via peptide release and if the effects of optically driven LHox/dyn release segregate based on VTADA projection targets including the basolateral amygdala (BLA) or the lateral or medial shell of the nucleus accumbens (lAcbSh, mAchSh). Using a combination of circuit tracing, optogenetics, and patch-clamp electrophysiology in male and female orexincre mice, we showed a diverse response of LHox/dyn optical stimulation on VTADA neuronal firing, which is not mediated by fast transmitter release and is blocked by antagonists to KOR and Ox1R signaling. Additionally, where optical stimulation of LHox/dyn inputs in the VTA inhibited firing of the majority of BLA-projecting VTADA neurons, optical stimulation of LHox/dyn inputs in the VTA bidirectionally affects firing of either lAcbSh- or mAchSh-projecting VTADA neurons. These findings indicate that LHox/dyn corelease may influence the output of the VTA by balancing ensembles of neurons within each population which contribute to different aspects of reward seeking.
Significance Statement
The mesolimbic dopamine (DA) system is known to play a crucial role in motivation and reward learning and receives neuromodulatory input from the lateral hypothalamus (LH). We show that optical stimulation of the orexin (ox)-containing LH input in the VTA releases both ox and dynorphin (dyn) to bidirectionally alter VTADA firing. Furthermore, ox and dyn differentially modulate firing of DA inputs to the basolateral amygdala, whereby dyn predominates, or to the nucleus accumbens which is sensitive to both neuromodulators. Our findings contribute to a more comprehensive understanding of the neuromodulatory effects of coreleased LH ox and dyn on the VTADA system.
Introduction
Reward and reinforcement processes drive motivated behaviors and are essential for survival and are guided by the ventral tegmental area (VTA; Berridge and Robinson, 1998; Bass et al., 2013; Wang et al., 2015; Morales and Margolis, 2017). Dopamine (DA) neurons in the VTA integrate multiple inputs to encode signals influencing motivated behaviors through diverse projections, including projections to the nucleus accumbens (NAc; Beier et al., 2019) and amygdala subnuclei, such as the basolateral amygdala (BLA; Lammel et al., 2008; Morel et al., 2022). VTADA projections to the NAc and BLA are anatomically segregated, such that DA neurons do not send collaterals to both regions (Beier et al., 2015; Baimel et al., 2017).
The VTA also receives input from orexin (ox; also known as hypocretin) neurons of the lateral hypothalamus (LH). LHox-containing fibers project to the VTA and make close appositions to VTADA neurons (Peyron et al., 1998; Fadel and Deutch, 2002; Baldo et al., 2003). While retrograde labeling from the VTA reveals a significant LHox input to the VTA (González et al., 2012), other studies have indicated that LHox neurons synapse onto ∼5% of VTADA and GABA neurons, even though there are many ox-containing dense core vesicles within the VTA (Balcita-Pedicino and Sesack, 2007). Ox interacts with orexin 1 (Ox1R) and 2 (Ox2R) receptors expressed in the VTA (Marcus et al., 2001; Narita et al., 2006), which are thought to primarily couple to Gq proteins as activation of Ox1R increases intracellular calcium (Uramura et al., 2001) and endocannabinoids (Tung et al., 2016) and increases firing of VTADA neurons (Korotkova et al., 2003; Baimel et al., 2017). Ox microinjected in the VTA can increase DA release in the NAc (Vittoz and Berridge, 2006; España et al., 2011). This is consistent with the findings that optogenetic activation of the LHox input to the VTA promotes DA release in the NAc. This occurs under phasic release conditions, likely resulting from the stimulation of glutamatergic afferents to the VTA (Thomas et al., 2022). Thus, ox strengthens the activity and output of dopaminergic neurons that project to the NAc. As such, ox action in the VTA is linked with motivational processes (Tyree and de Lecea, 2017). In particular, ox signaling in the VTA increases motivation for highly salient rewards such as addictive drugs or energy dense foods (Borgland et al., 2009; Thompson and Borgland, 2011).
Oxs are colocalized with dynorphin (dyn) in ∼95% of neurons (Chou et al., 2001; Li and van den Pol, 2006). These peptides are also coexpressed within dense core vesicles (Muschamp et al., 2014), suggesting that they are coreleased. Dyn is the endogenous ligand of Gi/o-coupled kappa opioid receptors (KORs), which are expressed on somatodendrites of VTADA neurons (Abraham et al., 2022). Activation of KORs in the VTA inhibits firing (Margolis et al., 2003, 2006; Ford et al., 2006), suppresses excitatory input to VTADA neurons (Margolis et al., 2005), and reduces DA release in the NAc (Robble et al., 2020). Thus, given the opposing action of ox and dyn, how does corelease of ox and dyn alter the activity of VTADA neurons? When both ox and dyn are applied to VTADA neurons that are responsive to either agonist individually, there is no net effect on the firing rate, suggesting that the opposing effects of each peptide effectively cancel one another out (Muschamp et al., 2014). However, few dopaminergic neurons are simultaneously responsive to both KOR and Ox1R activation (Muschamp et al., 2014). Electrical stimulation of miniature brain slices containing LH ox neurons released dyn, measured with an enzyme-linked immunosorbent assay using a dyn antibody (Li and van den Pol, 2006). Exogenous application of ox or dyn to the LH (Li and van den Pol, 2006) or the VTA (Baimel et al., 2017) produces opposing effects on firing. However, it is unknown how subpopulations of VTADA neurons that participate in distinct circuits differentially respond to the neuromodulatory influence of LHox/dyn neurons. We combined circuit tracing, optogenetics, and whole-cell patch-clamp recordings to investigate (1) if the LHox/dyn input to the VTA can corelease ox and dyn to modulate VTADA neuronal firing and (2) if VTADA neuronal projections to the lateral shell of the NAc (lAcbSh), the medial shell of the NAc (mAcbSh), or the BLA are differentially modulated by LHox/dyn input. We hypothesized that photoactivating the LHox/dyn input to the VTA induces activation or inhibition of distinct projection-defined subpopulations of VTADA neurons via ox or dyn, respectively.
Materials and Methods
Subjects
Adult male and female orexincre mice (Postnatal Day 60–90) were originally obtained from the Yamanaka lab at the University of Tokyo (Inutsuka et al., 2014) and bred locally at the University of Calgary Clara Christie Center for Mouse Genomics. Mice were group-housed (3–5 same sex per cage) with ad libitum access to water and food. Mice were housed in ventilated cages in a temperature (21 ± 2°C) and humidity-controlled (30–40%) room on a 12 h reverse light/dark cycle (lights on at 8:00 A.M.). Experiments were performed during the animal's light cycle. All experimental procedures adhered to ethical guidelines established by the Canadian Council for Animal Care and animal use protocols approved by the University of Calgary Animal Care and Use Committee (AC21-0034).
Surgical procedures
All orexincre mice (mice expressing the cre recombinase in cells expressing the pre–pro-ox peptide; Inutsuka et al., 2014) received bilateral infusions of either channelrhodopsin [“ChR2”; AAV2/8-EF1a-DIO-hChR2(H134R)-mCherry; Neurophotonics, Centre de Recherche CERVO] or control (“mCherry”; AAV2/8-hSyn-DIO-mCherry; Neurophotonics) virus. Mice were anaesthetized with isoflurane gas (5% for induction, 1–2% for maintenance) and secured in a stereotaxic frame (David Kopf Instruments). All measurements were made relative to the bregma for viral infusions. Viral injections were performed using a microinjector (Nano-inject II; Drummond Scientific Company). Each mouse received six infusions into the LH (100 nl per infusion, 23.1 nl/s), three in each hemisphere [anteroposterior (AP), −1.35; mediolateral (ML), ±0.9; dorsoventral (DV), −5.2, −0.5, 1, −5.0] for a total of 300 nl per hemisphere. After each infusion, the microinjector was left in place for 3 min to allow diffusion of virus away from the needle tip. After all infusions were complete in one hemisphere, the microinjector was left in place for an addition 5 min, 500 μm dorsal of the final injection location to allow diffusion of the virus through the brain tissue. Red RetroBeads (max excitation at 530 nm/max emission at 590 nm; 200 nl; Lumafluor) were infused bilaterally into the lAcbSh (from the bregma: AP, +1.425 mm; ML, ±1.75 mm; DV, −4.25 mm), the mAcbSh (from the bregma: AP, +1.65 mm; ML, ±0.5 mm; DV, −4.6), or the BLA (from the bregma: AP, −1.0 mm; ML, ±3.1 mm; DV, −5.3 mm). Injection sites were confirmed in all animals by preparing the coronal section of the lAcbSh or mAcbSh and in horizontal sections of the BLA. All mice received pre- and postoperative analgesic (meloxicam 5 mg/kg, subcutaneous) and were returned to their home cages and allowed to recover for 6–8 weeks prior to further experimental procedures. The location of the virus expression was performed post hoc.
Electrophysiology
All electrophysiological recordings were performed in slice preparations from adult male and female orexincre mice 6 weeks after receiving the cre-dependent viral vector containing ChR2. Briefly, mice were anesthetized with isoflurane and transcardially perfused with an ice-cold N-methyl-d-glucamine (NMDG) solution containing the following (in mM): 93 NMDG, 2.5 KCl, 1.2 NaH2PO4.2H2O, 30 NaHCO3, 20 HEPES, 25 d-glucose, 5 sodium ascorbate, 3 sodium pyruvate, 2 thiourea, 10 MgSO4.7H2O, and 0.5 CaCl2.2H2O and saturated with 95% O2–5% CO2. Mice were then decapitated, and brains were extracted. Sections containing the Lumifluor injection site were confirmed visually. Horizontal sections (250 μm) containing the VTA were cut in NMDG solution using a vibratome (VT1200, Leica Microsystems). Slices were then incubated in NMDG solution (32°C) saturated with 95% O2–5% CO2 for 10 min before being transferred to a holding chamber containing artificial cerebrospinal fluid (ACSF; in mM): 126 NaCl, 1.6 KCl, 1.1 NaH2PO4, 1.4 MgCl2, 2.4 CaCl2, 26 NaHCO3, and 11 glucose (32–34°C) equilibrated with 95% O2–5% CO2 for at least 45 min before recording. Slices were transferred to a recording chamber on an upright microscope (Olympus BX51WI) and continuously superfused with ACSF (2 ml/min, 34°C). Cells were visualized on an upright microscope using “Dodt-type” gradient contrast infrared optics, and whole-cell recordings were made using a MultiClamp 700B amplifier (Axon Instruments, Molecular Devices). Recording electrodes (3–5 MΩ) for measuring firing rates were filled with a potassium-d-gluconate internal solution (in mM): 130 potassium-d-gluconate, 4 MgCl2, 10 HEPES, 0.5 EGTA, 10 sodium creatine phosphate, 3.4 Mg-ATP, and 0.3 Na2GTP and 0.2% biocytin.
After breaking into the cell, hyperpolarization-activated cation currents (Ihs) were recorded in voltage-clamp mode using a voltage step to −130 mV to DA neurons voltage clamped at −70 mV. Ih was determined as the change in current between ∼30 and 248 ms after the voltage step was applied. Because most DA neurons ceased firing within 5 min of recording, current-step-induced firing was used for all experiments. For current-step experiments, the membrane potential for each neuron was set to −60 mV by DC injection via the patch amplifier, and a series of five current pulses (250 ms in duration, 5–25 pA apart, adjusted for each cell) were applied every 45 s, where the minimum current amplitude was set for each cell so that the first pulse was subthreshold and did not yield firing. From this series of current steps, we then selected a current step that yielded 3–5 action potentials during the baseline period and used that step for the analysis of peptide effects, as described previously (Baimel et al., 2017).
We optically stimulated LHox/dyn inputs at 30 Hz over a range of durations (10, 20, or 30 s) from a light-emitting diode (LED) blue light source (470 nm) that directly delivered the light path through the Olympus 40× water immersion lens. Once a maximal stimulation was determined, we continued with 30 Hz, 30 s stimulation for further experiments. Both SB334867 and NorBNI (Ox1R antagonist, KOR antagonist; Tocris; 1 µM) were dissolved in 100% DMSO stock solutions and then diluted to their final concentration containing 0.001% DMSO in ACSF and bath applied to slices.
Analysis of action potential firing
Firing data for all neurons were analyzed with the MiniAnalysis program (Synaptosoft). Optical stimulation-induced changes in firing are expressed as a percentage of baseline. For the analysis of the time courses, the firing rate pre- and postoptical stimulation was normalized to the average of the 10 min baseline firing rate. For analyses of effect sizes, the last 2 min prior to the optical stimulations or at the end of the recordings was averaged. To distinguish responders showing a decrease or increase from nonresponders, we used a criterion of a 20% change in the firing rate from the baseline. Responses from neurons of male and female mice were analyzed together. Our pilot data indicated that there were not sufficiently large sex differences in electrophysiological responses, and therefore mice were grouped together as these studies were not sufficiently powered to test for sex differences.
All statistical analyses were performed in GraphPad Prism 9.4.1 (GraphPad). Paired t test was used to compare before and after drug applications. In the case where three or more timepoints were compared, a repeated measures ANOVA was used. In all electrophysiology experiments, sample size is expressed as N/n where “N” refers to the number of cells recorded from “n” animals. Recordings of male and female mice were grouped together due to the limited availability of the mice. All cells were then averaged together and presented as mean ± SEM with individual values overlayed. All significance was set at p < 0.05. The levels of significance are indicated as follows: ****p < 0.0001, ***p < 0.001, **p < 0.01, and *p < 0.05.
Immunohistochemistry and confocal microscopy
The VTA is composed of a heterogeneous collection of cell types, distinguished in part by neurotransmitter content. To determine whether recorded neurons are indeed DAergic, we filled neurons with biocytin while recording and subsequently processed slices for tyrosine hydroxylase (TH). Brain slices from patch-clamp recordings in orexincre mice were fixed overnight in cold 4% paraformaldehyde (PFA) and then stored in phosphate-buffered saline (PBS) until processing. Sections were then blocked in 10% normal goat serum and incubated with mouse anti-TH (1:1,000) for 24 h at room temperature. Alexa Fluor 488 goat anti-mouse (1:400) and DyLight 594-conjugated streptavidin were applied to identify DA neurons tagged with biocytin. Slices were mounted with Fluoromount.
To check for colocalization of ChR2 expression and ox in LH, mice were deeply anesthetized with isoflurane and transcardially perfused with PBS and then with 4% PFA. Brains were dissected and postfixed in 4% PFA at 4°C overnight and then switched to 30% sucrose. Coronal frozen sections were cut at 30 µm using a cryostat. Ten percent goat serum was applied to block nonspecific binding for 1 h. Sections were then incubated with primary antibody rabbit anti-ox 1:500 (Phoenix Pharmaceuticals, H-003-30) and chicken red fluorescent protein (to enhance mCherry reporter expression) 1:2,000 (Rockland, 600-901-379) in 1% BSA for 1 h at room temperature followed by incubation with secondary antibody Alexa Fluor 488 goat anti-rabbit and Alexa Fluor 594 goat anti chicken 1:400 for 1 h.
To check for the colocalization of RetroBeads, biocytin and TH in VTA-patched slices, 10% goat serum was applied to block nonspecific binding for 1 h. Sections were then incubated with primary antibody mouse anti-TH (1:1,000) in 1% BSA for 24 h at room temperature followed by incubation with secondary antibody Alexa Fluor 647 goat anti-mouse (1:400) and Alexa Fluor 488 streptavidin (1:200) for 2 h.
All images were obtained at 10× on an Olympus scanner microscope (Olympus Canada) and at 20× on a Nikon Eclipse C1si confocal microscope (Nikon Canada). Cell count in LH was quantified with ImageJ at 20×.
Results
Optical stimulation of LHox/dyn inputs to VTADA neurons bidirectionally modulates firing
To study the effect of LHox/dyn photoactivation on VTADA neuronal activity, we injected AAV2/8-EF1a-DIO-hChR2(H134R)-mCherry or control (“mCherry”; AAV2/8-hSyn-DIO-mCherry) virus into the LH of orexincre mice (Fig. 1A–C). An 81 ± 12% of LH ox neurons expressed the reporter for ChR2, and 91 ± 3% of ChR2-expressing neurons were ox+. Because dyn is expressed in ox-containing neurons, and no other neurons in the LH, this manipulation is specific to the LHox/dyn population. We recorded the response of VTADA neurons, identified by electrical characteristics and post hoc TH staining, after optical stimulation of LHox/dyn inputs in the VTA, using whole-cell patch-clamp electrophysiological recordings (Fig. 1D–F). Following the baseline recording, we applied optical stimulation at a frequency of 30 Hz with increasing duration, which has been shown to effectively modulate the firing of VTADA neurons previously (Thomas et al., 2022). If an increase in firing frequency occurred in response to optical stimulation, we applied the Ox1R antagonist SB334867 (1 µM, 15 min), as ox is known to increase firing of VTADA neurons (Korotkova et al., 2003; Baimel et al., 2017; Thomas et al., 2022). Conversely, if a decrease in firing occurred, we administered the KOR antagonist NorBNI (1 µM, 15 min), as dyn is known to decrease firing of VTADA neurons (Margolis et al., 2003; Baimel et al., 2017). We did not apply an antagonist if no change in firing occurred. Optical stimulation of LHox/dyn inputs to VTADA neurons did not alter the firing rate in mCherry mice (baseline: 100 ± 0.03%; 30 Hz, 10 s: 106 ± 4%; 30 Hz, 20 s: 114 ± 4%, 30 Hz, 30 s: 112 ± 5%; n/N = 17/13; Fig. 1G), suggesting that the increasing duration of the 473 nM LED does not alter firing on its own. In contrast, optical stimulation of LHox/dyn inputs to VTADA neurons of ChR2-expressing mice produced diverse responses. In 37% of VTADA neurons, there was a duration-dependent increase in firing (baseline: 104 ± 4%; 30 Hz, 10 s: 121 ± 10%; 30 Hz, 20 s: 143 ± 10%; 30 Hz, 30 s: 151 ± 10%; SB34867: 131 ± 5%; n/N = 7/5; Fig. 1H,I–K). Increased firing was significantly different from baseline at 30 Hz, 20 s (p = 0.035) and 30 Hz, 30 s (p = 0.016), but not after application of SB334867 [p = 0.57; RM one-way ANOVA: F(2.23, 13.4) = 6.52, p = 0.0091; Fig. 1J]. In 47% of VTADA neurons, there was a duration-dependent decrease in firing (baseline: 99 ± 0.5%; 30 Hz, 10 s: 73 ± 6%; 30 Hz, 20 s: 59 ± 7%; 30 Hz, 30 s: 56 ± 4%; NorBNI: 92 ± 16%; n/N = 9/5; Fig. 1H). Decreased firing was significantly different from baseline at 30 Hz, 10 s (p < 0.0001), 30 Hz, 20 s (p < 0.0001), and 30 Hz, 30 s (p < 0.0001), but not after application of NorBNI [p = 0.57; RM one-way ANOVA: F(4, 32) = 19.26, p < 0.0001; Fig. 1L,M]. A subset of VTADA neurons (16%) showed no difference in firing (Fig. 1H,I).
Optical stimulation of LHox/dyn inputs in the VTA produces a diverse response on the firing rate of VTADA neurons. A, Schematic of viral strategy. Orexincre mice were infused with either channelrhodopsin [AAV2/8-EF1a-DIO-hChR2(H134R)-mCherry; “ChR2”] or control virus (AAV2/8-hSyn-DIO-mCherry; “mCherry”). LH, lateral hypothalamus; VTA, ventral tegmental area. B, Coronal brain sections mapping viral infection areas within the LH (−1.34 mm in AP plane relative to bregma). C, Immunohistochemical demonstration of cre-dependent expression of ChR2-mCherry (red, top), ox (green, middle), and a merged image of ox and mCherry (ChR2 reporter) within the LH (bottom). Scale bar, 50 µm. D, Experimental timeline of electrophysiology experiments with repeated optical stimulation. E, Example image of LHox fibers expressing ChR2 within the VTA of orexincre mice (merged image of VTADA neuron labeled with tyrosine hydroxylase (TH, green) and LHox fibers expressing ChR2 (mCherry reporter, red). F, VTADA neuronal identification by colabeling of biocytin (red, left) and TH (green, middle) or the merged image of VTADA neuron colabeled with TH and biocytin (right). Scale bar, 25 µm. G, Time course of evoked firing before and after 30 Hz photoactivation of LHox/dyn inputs in VTADA neurons (10, 20, or 30 s) of mCherry-expressing control mice. H, Time course of evoked firing before and after 30 Hz photoactivation of LHox/dyn inputs on VTADA neurons (10, 20, or 30 s) of ChR2-expressing mice. I, Distribution of responses of VTADA neurons to LHox/dyn photostimulation. J, Bar graph of averaged firing responses (% baseline) of the 2 min prior to the next optical stimulation for VTADA neurons that increased their firing to optical stimulation of LHox/dyn inputs. Individual responses are overlayed. K, Sample traces of evoked action potentials that increased frequency before and after optical stimulation of LHox/dyn inputs. L, Bar graph of averaged firing responses (% baseline) from 2 min prior to the next optical stimulation for VTADA neurons that decreased their firing to optical stimulation of LHox/dyn inputs. M, Sample traces of evoked action potentials that decreased frequency before and after optical stimulation of LHox/dyn inputs.
Ox neurons also release glutamate (Rosin et al., 2003). Optogenetic stimulation of LH inputs is likely to induce the concurrent release of other neurotransmitters expressed in ox neurons. We were unable to evoke AMPAR EPSCs from optical stimulation of LHox/dyn terminals in the VTA (Extended Data Fig. 2-1), even though optical stimulation of LHox/dyn inputs to the VTA can potentiate electrically evoked NMDARs (Thomas et al., 2022). To eliminate potential effects of fast transmission on the activity of VTADA neurons, evoked firing of VTADA neurons (n = 12 cells from nine mice) was recorded in the presence of synaptic blockers including AP5 (50 mM), DNQX (10 mM), and picrotoxin (100 mM), antagonists of NMDA receptors, AMPA receptors, and GABAA receptors, respectively (Fig. 2A,B). As before, we observed that VTADA neurons were either activated [41% of neurons, RM one-way ANOVA: F(4, 16) = 3.58, p = 0.029; Fig. 2B,C] or inhibited [41% of neurons, RM one-way ANOVA: F(4, 16) = 3.74, p = 0.025; Fig. 2B,C] by optical stimulation of LHox/dyn inputs, suggesting that these changes in VTADA neuron firing are mediated by postsynaptic peptidergic modulation. A subset of neurons were unchanged by optical stimulation (16%; Fig. 2B,C). Increased firing from 30 Hz, 30 s stimulation was significantly different from baseline (p = 0.0093; baseline: 3.6 ± 0.4 APs; 30 Hz, 30 s: 6.6 ± 0.8 APs; SB34867: 5.2 ± 0.9 APs; n/N = 5/3; Fig. 2D), and this change was inhibited by the Ox1R antagonist SB334867 [1 µM; p = 0.13, RM one-way ANOVA: F(1.516, 6.063) = 14.7, p = 0.0059; Fig. 2D,E]. Furthermore, decreased firing from 30 Hz, 30 s stimulation was different from baseline (p = 0.037; baseline: 3.8 ± 0.4 APs; 30 Hz, 30 s: 1.8 ± 0.4 APs; NorBNI: 4.2 ± 0.5 APs; n/N = 5/4) and was blocked by the KOR antagonist NorBNI [1 µM; p = 0.82, RM one-way ANOVA: F(1.764, 7.054) = 7.0, p = 0.023; Fig. 2F,G]. Because 30 Hz, 30 s optical stimulation of LHox/dyn inputs to the VTA produced the largest response in either direction, we used this stimulation in the subsequent experiments. Furthermore, all subsequent experiments were conducted in the presence of synaptic blockers.
Synaptic inhibitors did not block LHox/dyn photostimulation-induced changes in firing of VTADA neurons. A, Schematic describing the strategy for optical stimulation of LHox/dyn inputs on VTADA neuronal firing in the presence of synaptic blockers including DNQX, AP5, and picrotoxin. Optical stimulation of LHox/dyn inputs on VTADA did not induce AMPA oEPSCs; see Figure 2-1 for more details. B, Time course of evoked firing before and after 30 Hz photoactivation of LHox/dyn inputs in VTADA neurons. C, Distribution of responses of VTADA neurons to LHox/dyn photostimulation. D, Number of action potentials during baseline, after 30 Hz, 30 s optical stimulation, and after SB334867 (1 µM) application in VTADA neurons that increased their response to LHox/dyn optical stimulation. E, Sample traces of evoked action potentials that increased frequency before and after optical stimulation of LHox/dyn inputs. F, Number of action potentials during baseline, after 30 Hz, 30 s optical stimulation, and after NorBNI (1 µM) application in VTADA neurons that decreased their response to LHox/dyn optical stimulation. G, Sample traces of evoked action potentials that decreased frequency before and after optical stimulation of LHox/dyn inputs.
Figure 2-1
Optical stimulation of LHox/dyn inputs does not produce AMPA EPSCs in the VTA. A) Diagram of parameters used to optically stimulate AMPA EPSCs recorded at -70 mV in the presence of picrotoxin in the VTA. B) Percent change in response post optical stimulation compared to pre-stimulation baseline. C) Example sweep from a neuron recorded before and after optical stimulation. Download Figure 2-1, TIF file.
Altered firing of VTADA neurons upon photoactivation is due to activation of Ox1R and KOR
We next wanted to confirm that optical stimulation of LHox/dyn inputs to the VTA was releasing ox or dyn leading to activation of their receptors, Ox1R or KOR, respectively expressed in the VTA. In the presence of SB334867 (1 µM), LHox/dyn optical stimulation decreased VTADA firing in all neurons [baseline: 100 ± 1%; 30 Hz, 30 s (percent of baseline): 64 ± 5%; NorBNI: 96 ± 6%; n/N = 7/4; Fig. 3A]. In SB334867, action potentials decreased from baseline (3.7 ± 0.4) to 2.1 ± 0.3 after 30 Hz, 30 s stimulation of LHox/dyn inputs. This effect was washed off by NorBNI [3.6 ± 0.4 Aps; RM one-way ANOVA: F(1.54, 9.14) = 23.54, p = 0.0004, Dunnett's: baseline vs 30 Hz, 30 s, p = 0.0015; baseline vs NorBNI, p = 0.96; Fig. 3B,C]. We next recorded VTADA neurons in the presence of NorBNI (1 µM), an antagonist that activates signaling pathways that induce a long-lasting suppression of receptor activity and thus is not washed out (Bruchas et al., 2007). LHox/dyn optical stimulation increased VTADA firing in five neurons (baseline: 99 ± 0.6%; 30 Hz, 30 s: 126 ± 8%; SB334867: 99.4 ± 0.6%; n/N = 5/3) but had no effect in two neurons (Fig. 3D). In NorBNI, action potentials increased from baseline (3.9 ± 0.3) to 4.7 ± 0.5 after 30 Hz, 30 s stimulation of LHox/dyn inputs. This effect was washed off by SB334867 [3.9 ± 0.3 Aps; RM one-way ANOVA: F(2, 12) = 10.8, p = 0.0021, Dunnett's: baseline vs 30 Hz, 30 s, p = 0.0032; baseline vs SB334867, p > 0.999; Fig. 3E,F]. We next recorded VTADA neurons in the presence of both SB334867 and NorBNI. There was no change in evoked firing of VTADA neurons after LHox/dyn optical stimulation (baseline: 103 ± 2%; 30 Hz, 30 s: 105 ± 5%; n = 6/5; Fig. 3G). In SB334867 and NorBNI, there was no difference in action potential number between the baseline, (3.2 ± 0.2), after 30 Hz, 30 s stimulation of LHox/dyn inputs (3.3 ± 0.2), and wash [3.2 ± 0.3; RM one-way ANOVA: F(1.24, 6.34) = 0.55, p = 0.53, Fig. 3H,I]. This provides evidence that (1) LHox/dyn stimulation in the VTA leads to ox and dyn release in the VTA and that (2) increased VTADA firing after LHox/dyn optical stimulation is mediated by Ox1R signaling, whereas decreased VTADA firing is mediated by KOR signaling. Importantly, these effects on firing were found to be independent of synaptic glutamate or GABA release.
Altered firing of VTADA neurons upon photoactivation of LHox/dyn inputs is due to activation of Ox1Rs and KORs. A, All VTADA neurons decreased firing after photoactivation of LHox/dyn inputs to the VTA in the presence of synaptic blockers and the Ox1R antagonist SB334867 (1 µM). B, Bar graph of evoked action potentials in the presence of SB334867 and synaptic blockers before and after optical stimulation of LHox/dyn inputs and after washout with NorBNI (1 µM). C, Sample traces of evoked action potentials in the presence of SB334867 and synaptic blockers during baseline, after optical stimulation, and after NorBNI treatment. D, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in VTADA neurons preblocked with NorBNI (1 µM). E, LHox/dyn photoactivation increased the firing of VTADA neurons preblocked with NorBNI. F, Sample traces of evoked action potentials before and after optical stimulation. G, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in VTADA neurons preblocked with SB334867 and NorBNI (1 µM). H, LHox/dyn photoactivation had no effect on the firing of VTADA neurons preblocked with SB334867 and NorBNI. I, Sample traces of evoked action potentials before and after optical stimulation.
Temporal characteristics of endogenous LHox/dyn corelease on DA neuronal activity
Next, we investigated the temporal characteristics of LHox/dyn-mediated modulation of VTADA neuronal activity. Following LHox/dyn optical stimulation, VTADA neurons that increased or decreased firing were identified and analyzed separately. Increased firing of VTADA neurons induced by optical stimulation of LHox/dyn inputs peaked 4 min after stimulation and returned to baseline levels within 25 min (baseline: 99 ± 0.5%; 30 Hz, 30 s: 148 ± 11%; return to baseline, 114 ± 9%; n/N = 8/6). We then tested if a second optical stimulation could evoke peptide release. This subsequent stimulation again led to an increase in firing of VTADA neurons with a peak 4 min after stimulation (second optostim, 139 ± 10%). Both first and second peaks after optical stimulation were significantly different to their respective baselines [RM two-way ANOVA: F(1, 7) = 10.17, p = 0.015, Tukey's multiple comparisons tests: baseline vs 30 Hz, 30 s: p = 0.0003; second baseline vs second optostim: p = 0.003; Fig. 4A–C). Conversely, decreased VTADA firing induced by LHox/dyn optical stimulation exhibited a peak in response 8 min after stimulation and prolonged inhibition that did not return to baseline after 40 min (baseline: 100 ± 0.8%; 30 Hz, 30 s: 57 ± 8%; n/N = 8/5; Wilcoxon matched-pairs signed rank test: baseline vs optical stimulation, p = 0.0156; Fig. 4D–F). When we administered a second optical stimulation 15 min after the first stimulation in VTADA neurons that decreased firing, we observed a small but significant further decrease in response compared with the second baseline [Fig. 4G–I; RM two-way ANOVA: F(1, 9) = 15.00, p = 0.0038, Tukey's multiple comparisons tests: baseline vs 30 Hz, 30 s: p = 0.04; baseline 2 vs optostim 2: p = 0.047]. In summary, our data reveal a significant difference in the duration of excitatory and inhibitory responses to LHox/dyn optical stimulation in the VTA, such that inhibition of VTADA neurons persists, whereas activation of VTADA neurons is more transient.
Temporal dynamics of VTADA neuronal responses to 30 Hz LHox/dyn optical stimulation. A, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in VTADA neurons that increased firing. A second photoactivation of LHox/dyn produced a repeated increase in firing of VTADA neurons. B, Bar graph of evoked action potentials before and after the first and second optical stimulations of LHox/dyn inputs to VTADA neurons. C, Sample traces of evoked action potentials during baseline, after first and second optical stimulations. D, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in VTADA neurons that decreased firing. E, Bar graph of evoked action potentials before and after optical stimulation of LHox/dyn inputs to VTADA neurons that decreased firing. F, Sample traces of evoked action potentials during baseline and after optical stimulation. G, Time course of evoked firing before and after two photoactivations of LHox/dyn inputs in VTADA neurons that decreased firing. H, Bar graph of evoked action potentials before and after the first and second optical stimulations of LHox/dyn inputs to VTADA neurons that decreased firing. I, Sample traces of evoked action potentials during baseline, after the first and second optical stimulations.
Photoactivation of LHox/dyn terminals VTA DA has diverse effects on evoked firing based on projection target
The VTADA system is heterogeneous and is increasingly thought about in terms of anatomically and functionally distinct subnetworks (Watabe-Uchida et al., 2012). VTADA neurons project to different regions on the basis of their localization along the ML axis (Lammel et al., 2008; Beier et al., 2015, 2019). We next tested if the distinct firing responses induced by optical stimulation of LHox/dyn inputs segregate by dopaminergic projection target. Therefore, in orexincre mice expressing ChR2 in ox neurons, we recorded from VTADA neurons retrogradely labeled from the subregions of the NAc or the BLA using fluorescent beads. We first targeted the NAc by injecting red Lumifluor RetroBeads, a retrograde tracer, in two subnuclei: the IAcbSh (Fig. 5A–C) and the mAcbSh of orexincre mice expressing ChR2 in LHox/dyn neurons (Fig. 6A–C). We confirmed the RetroBead injection sites (Figs. 5C, 6C, 7C), as well as the TH expression in the VTA neuron projecting to the lAcbSh (Fig. 5A), mAcbSh (Fig. 6A), or BLA (Fig. 7A). We also compared the electrophysiological characteristics of VTADA neurons with known projections (Extended Data Fig. 5-1). As reported previously (Lammel et al., 2011; Baimel et al., 2017), VTADA neurons that project to the lAcbSh have larger hyperpolarization-activated current (Ih) than those projecting to the mAcbSh or the BLA (Kruskal–Wallis test, p = 0.0079) with significant differences between VTADA neurons that project to the lAcbSh and mAcbSh (p = 0.016) or BLA (p = 0.024; Dunn's multiple comparison test; Fig. 5-1A). There was also a significant difference between groups on capacitance (Kruskal–Wallis test, p = 0.0009), with VTADA neurons that project to the lAcbSh having a larger capacitance, reflecting larger cell size, than those projecting to the mAcbSh (p = 0.0012) or the BLA (p = 0.0135, Dunn's multiple comparison test; Fig. 5-1B). Input resistance was also different between groups (Kruskal–Wallis test, p = 0.0175), with a significant difference between VTADA neurons that project to the lAcbSh and the BLA (p = 0.02, Dunn's multiple comparison's test; Fig. 5-1C). Taken together VTADA neurons projecting to the lAcbSh have larger Ih current and capacitance and smaller input resistance than those projecting to the mAcbSh or the BLA.
Photoactivation of LHox/dyn inputs preferentially activates most lAcbSh-projecting VTADA neurons. A, Images of VTA sections with TH-positive neurons (purple), biocytin (green), and retrograde-labeled (red) neurons from the lAcbSh. Scale bar, 50 µm. B, Schematic of viral strategy C, Example of RetroBead injection site in the lAcbSh of orexincre mice. Scale bar, 1 mm. D, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in lAcbSh-projecting VTADA neurons. E, Evoked action potentials before and after optical stimulation of LHox/dyn to lAcbSh-projecting VTADA neurons and in the presence of SB334867 for VTADA neurons that increased their firing to LHox/dyn stimulation. LAcbSh-projecting VTADA neurons have different electrophysiological properties than BLA- or mAcbSh-projecting VTADA neurons. See Figure 5-1 for more details. F, Sample traces of evoked action potentials before and after optical stimulation for VTADA neurons that increased firing to photostimulation of LHox/dyn inputs. G, Evoked action potentials before and after optical stimulation of LHox/dyn to lAcbSh-projecting VTADA neurons and in the presence of NorBNI for VTADA neurons that decreased their firing to LHox/dyn stimulation. H, Sample traces of evoked action potentials before and after optical stimulation for VTADA neurons that decreased firing to photostimulation of LHox/dyn inputs. I, Distribution of responses of lAcbSh-projecting VTADA neurons to LHox/dyn photostimulation. J, Time course of evoked firing of lAcbSh-projecting VTADA neurons preblocked with SB334867 and NorBNI. K, Sample traces of evoked action potentials of lAcbSh-projecting VTADA neurons before and after optical stimulation in the presence of SB334867 and NorBNI.
Figure 5-1
lAcbSh-, mAcbSh-, or BLA-projecting VTADA neurons have different intrinsic electrophysiological properties. A) HCN current B) capacitance and C) input resistance of lAcbSh- (open bars), mAcbSh- (shaded bars)- and BLA- (filled bars) projecting VTADA neurons recorded from ChR2 orexincre mice. Download Figure 5-1, TIF file.
Photoactivation of LHox/dyn inputs bidirectionally affects the neuronal firing of mAcbSh-projecting VTADA neurons. A, Images of VTA sections with TH-positive neurons (purple), biocytin (green), and retrograde-labeled (red) neurons from the mAcbSh. Scale bar, 50 µm. B, Schematic of viral strategy C, Example of RetroBead injection site in the mAcbSh of orexincre mice. Scale bar, 1 mm. D, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in mAcbSh-projecting VTADA neurons. E, Evoked action potentials before and after optical stimulation of LHox/dyn to mAcbSh-projecting VTADA neurons and in the presence of SB334867 for VTADA neurons that increased their firing to LHox/dyn stimulation. F, Sample traces of evoked action potentials before and after optical stimulation for VTADA neurons that increased firing to photostimulation of LHox/dyn inputs. G, Evoked action potentials before and after optical stimulation of LHox/dyn to mAcbSh-projecting VTADA neurons and in the presence of NorBNI for VTADA neurons that decreased their firing to LHox/dyn stimulation. H, Sample traces of evoked action potentials before and after optical stimulation for mAcbSh-projecting VTADA neurons that decreased firing to photostimulation of LHox/dyn inputs. I, Distribution of responses of mAcbSh-projecting VTADA neurons to LHox/dyn photostimulation. J, Time course of evoked firing of mAcbSh-projecting VTADA neurons preblocked with SB334867 and NorBNI. K, Sample traces of evoked action potentials of mAcbSh-projecting VTADA neurons before and after optical stimulation in the presence of SB334867 and NorBNI.
Photoactivation of LHox/dyn inputs to BLA-projecting VTADA primarily decreases firing. A, Images of VTA sections with TH-positive neurons (purple), biocytin (green), and retrograde-labeled (red) neurons from the BLA. Scale bar, 50 µm. B, Schematic of viral strategy C, Example of RetroBead injection site in the BLA of orexincre mice. Scale bar, 1 mm. D, Time course of evoked firing before and after photoactivation of LHox/dyn inputs in BLA-projecting VTADA neurons. E, Evoked action potentials before and after optical stimulation of LHox/dyn to BLA-projecting VTADA neurons and in the presence of SB334867 for VTADA neurons that increased their firing to LHox/dyn stimulation. F, Sample traces of evoked action potentials before and after optical stimulation for VTADA neurons that increased firing to photostimulation of LHox/dyn inputs. G, Evoked action potentials before and after optical stimulation of LHox/dyn to BLA-projecting VTADA neurons and in the presence of NorBNI for VTADA neurons that decreased their firing to LHox/dyn stimulation. H, Sample traces of evoked action potentials before and after optical stimulation for BLA-projecting VTADA neurons that decreased firing to photostimulation of LHox/dyn inputs. I, Distribution of responses of BLA-projecting VTADA neurons to LHox/dyn photostimulation. J, Time course of evoked firing of BLA-projecting VTADA neurons preblocked with SB334867 and NorBNI. K, Sample traces of evoked action potentials of BLA-projecting VTADA neurons before and after optical stimulation in the presence of SB334867 and NorBNI.
We next characterized the evoked firing responses of VTADA neurons to optical stimulation of LHox/dyn inputs. We found that 55% of VTADA neurons projecting to the lAcbSh increased firing to optical stimulation of LHox/dyn inputs, whereas 36% decreased firing (Fig. 5D–I). Of the 11 DA-lAcbSh neurons, six neurons increased firing [baseline: 3.5 ± 0.2 APs; 30 Hz, 30 s: 4.8 ± 0.3 APs; SB334867: 3.8 ± 0.4 APs; n/N = 6/5 mice, RM one-way ANOVA: F(2, 15) = 4.72, p = 0.0256, Dunnett's: baseline vs 30 Hz, 30 s p = 0.018, baseline vs SB334867: p = 0.69; Fig. 5E,F]. Increases in firing were blocked by SB33864 (Fig. 5E), suggesting increases in firing were mediated by Ox1R signaling. Four of 11 neurons decreased firing in response to optical stimulation of LHox/dyn inputs [baseline: 4 ± 0.4 APs; 30 Hz, 30 s: 2.7 ± 0.5 APs; NorBNI: 3.5 ± 0.5 APs; n/N = 4/2 mice, RM one-way ANOVA: F(1.92, 5.78) = 11.4, p = 0. 01; Dunnett's: baseline vs 30 Hz, 30 s: p = 0.025, baseline vs NorBNI: p = 0.28; Fig. 5G,H]. Decreased firing was blocked by NorBNI (Fig. 5G), suggesting that this was mediated by KOR signaling. One lAcbSh-projecting VTADA neuron exhibited no change in firing (Fig. 5D,I). In the presence of Ox1R and KOR antagonists, there was no change in firing of VTADA neurons after LHox/dyn optical stimulation (Fig. 5J,K), suggesting changes in firing were mediated by peptide release.
The 30 Hz optical stimulation of LHox/dyn inputs in the VTA had a similar bidirectional effect on mAcbSh-projecting VTADA neurons. Four out of 13 (31%) VTADA neurons recorded increased firing in response to LHox/dyn stimulation [baseline: 3.5 ± 0.3 APs; 30 Hz, 30 s: 5.0 ± 0.4 APs; SB334867: 4.0 ± 0.4 APs; n/N = 4/2 mice, RM one-way ANOVA: F(1.39, 4.19) = 9.0, p = 0.034, Dunnett's: baseline vs 30 Hz, 30 s p = 0.05, baseline vs SB334867: p = 0.56; Fig. 6E,F]. Additionally, the stimulation led to a decrease in firing activity for 7 out of 13 (54%) recorded neurons [baseline: 4.1 ± 0.3 APs; 30 Hz, 30 s: 2.0 ± 0.3 APs; NorBNI: 3.6 ± 0.3 APs; n/N = 7/5 mice, RM one-way ANOVA: F(1.35, 8.1) = 66.3, p < 0.0001; Dunnett's: baseline vs 30 Hz, 30 s p = 0.0003, baseline vs NorBNI: p = 0.28; Fig. 6G,H]. The increase in firing was reversed by SB334867 (Fig. 6E,F), whereas the decrease in firing was reversed by NorBNI (Fig. 6G,H). Two of 13 cells had no response (15%; Fig. 6D,I). There was no change in firing after optical stimulation of LHox/dyn inputs to VTADA neurons in the presence of both KOR and Ox1R antagonists (Fig. 6J,K), confirming that bidirectional changes in firing are mediated by LH ox or dyn.
We next examined the response of BLA-projecting VTADA neurons to LHox/dyn optical stimulation. The majority of BLA-projecting VTADA (8 of 12; 67%) are inhibited by optical stimulation of LHox/dyn inputs [baseline: 3.9 ± 0.2 APs; 30 Hz, 30 s: 2.7 ± 0.2 APs; NorBNI: 3.4 ± 0.2 APs; RM one-way ANOVA: F(1.89, 15.06) = 31, p < 0.0001, Dunnett's: baseline vs 30 Hz, 30 s, p = 0 < 0.0001; baseline vs NorBNI: p = 0.06; Fig. 7D,G,H]. This inhibition of firing was blocked by NorBNI (Fig. 7D,G). Optical stimulation increased firing in only 3 of 12 (25%) BLA-projecting VTADA neurons although this did not reach statistical significance [baseline: 3 ± 0.6 APs; 30 Hz, 30 s: 4.3 ± 0.7 APs; SB334867: 3.3 ± 0.7; RM one-way ANOVA: F(1.0, 2.0) = 13.0, p = 0.069; Fig. 7D,E,F,I]. One of 12 neurons recorded exhibited no change (Fig. 7D,I). Finally, there was no change in firing after optical stimulation of LHox/dyn inputs to BLA-projecting VTADA neurons in the presence of both KOR and Ox1R antagonists (Fig. 7J,K). Taken together, 30 Hz LHox/dyn photoactivation inhibited a large proportion of BLA-projecting VTADA neurons but fewer lAcbSh-projecting or mAcbSh-projecting VTADA neurons. Overall, these results indicate that the majority of VTADA neurons activated by LHox/dyn stimulation project to the lAcbsh, whereas the majority of inhibited neurons project to the BLA or the mAcbSh. These experiments identify how firing of lAcbSh-, mAcbSh-, or BLA-projecting VTADA neurons can be tuned by corelease of neuropeptides from the LH in a physiological state, such that LHox dominates in the VTADA–lAcbSh projection, whereas LHdyn dominates in the VTADA–mAcbSh and BLA projections.
Discussion
Here, we demonstrated that optical stimulation of LHox/dyn inputs has distinct modulatory effects on the activity of projection target-defined VTADA neurons. We showed that 30 Hz optical stimulation of LHox/dyn produced both ox and dyn neuromodulation of VTADA firing. LHox/dyn stimulation-induced increased firing was blocked by the Ox1R antagonist, whereas LHox/dyn stimulation-induced decreased firing was blocked by the KOR antagonist. When both peptide receptors were blocked, no change in firing occurred after LHox/dyn stimulation, suggesting that changes in firing are mediated by peptide release into the VTA. Furthermore, optical stimulation of LHox/dyn inputs increased firing in lAcbSh-projecting VTADA neurons, but mostly inhibited firing in the mAcbSh and BLA projections. These results suggest that corelease of ox and dyn may balance ensembles of VTADA within each projection to influence DA output.
Several whole animal behavioral studies have demonstrated optogenetic stimulation of behavioral responses consistent with peptide release (Adamantidis et al., 2007; Atasoy et al., 2012; Jego et al., 2013; Kempadoo et al., 2013). However, the temporal dynamics of the release and the postsynaptic neurons upon which this neuromodulatory action occurs is not clear. This can be addressed in brain slice preparations where optical stimulation frequency and duration can be controlled and responses to fast acting amino acid neurotransmitters or coreleased peptides can occur. The LHox/dyn input to the VTA presents a unique preparation where only a small proportion of terminals synapse onto VTA neurons, but there is a high density of peptide-containing dense core vesicles within the VTA, allowing for neuromodulatory action from peptide release (Balcita-Pedicino and Sesack, 2007). While single brief (2–5 ms) light pulses are sufficient to evoke neurotransmitters from synaptic vesicles as voltage-dependent calcium channels are highly coupled to synaptic vesicles in the active zone, these protocols may be insufficient to release neuropeptides (Bruns and Jahn, 1995; Leenders et al., 1999; Südhof, 2012). The release of neuropeptides typically requires a higher frequency and longer duration of depolarization to allow for sufficient calcium-dependent mobilization of dense core vesicles to the plasma membrane (Weisskopf et al., 1993; Leenders et al., 1999; Muschol and Salzberg, 2000). This may be related to the location of dense core vesicles away from the active zone (Bruns and Jahn, 1995; van den Pol, 2012). We found that optical stimulation at 30 Hz with a 5 ms pulse width for either 10, 20, or 30 s could alter the firing rate of VTADA neurons. Importantly, this occurred in the absence of amino acid-mediated synaptic transmission and was blocked by antagonists for Ox1R and KOR, suggesting neuropeptide release into the VTA. Consistent with this, we found that optogenetically stimulating glutamatergic LH ox neurons did not produce AMPA excitatory postsynaptic currents, but potentiated electrically evoked NMDA currents, suggesting a neuromodulatory role (Thomas et al., 2022). Thus, we demonstrate here a bidirectional neuromodulatory effect of LH ox and dyn on VTADA firing.
Ox neurons have intrinsic features that promote long-lasting firing activity (Burt et al., 2011). Ox neurons are in an intrinsically depolarized state largely due to the constitutively active cation current by transient receptor potential C channels (Cvetkovic-Lopes et al., 2010). This depolarized state maintains neurons near their firing threshold leading to sustained spontaneous firing. In vivo recordings suggest that ox neurons exhibit slow (<10 Hz) tonic discharges during wakefulness (Takahashi et al., 2008; Hassani et al., 2009). These mechanisms are likely important for the physiological functions of ox neurons, which require prolonged output to maintain a wakeful state. However, ox neurons can also follow faster frequencies likely important for the release of neuropeptides from dense core vesicles. For example, action potentials of LHox/dyn neurons can efficiently follow optogenetic stimulation at 30 Hz (Thomas et al., 2022) up to 50 Hz frequencies (Adamantidis et al., 2007). Several local mechanisms might contribute to this elevated firing activity. Ox can be somatodentrically released which activates Ox2Rs to open nonselective cation channels which can depolarize ox neurons or increase presynaptic glutamate release (Li et al., 2002). Furthermore, ox neurons receive numerous glutamatergic inputs which outnumber inhibitory synapses that may also contribute to sustained depolarized states required for neuropeptide release (Horvath and Gao, 2005). Finally, ox neurons use astrocyte-derived lactate as an energy substrate to maintain spontaneous firing and the excitatory action of glutamatergic transmission (Parsons and Hirasawa, 2010). Circumstances in which these sustained frequencies (and presumed ox release) may occur are in response to arousing, motivationally relevant situations. Recordings of presumed ox neurons in freely moving rats found that ox neurons have tonic activity during active waking, grooming, and eating, but have rapid firing during periods of adaptive behaviors, such as exploration, play, and predation (Mileykovskiy et al., 2005; Wu et al., 2011). Highly arousing events, such as stress and reward seeking, may also sufficiently increase sustained firing to promote neuropeptide release, which may be required to engage monoamine neuromodulatory systems to either escape or take advantage of the opportunity (Giardino and de Lecea, 2014). Thus, ox release may occur in response to salient, motivationally relevant events to coordinate adaptive behaviors (reviewed in Mahler et al., 2014).
Application of the Ox1R antagonist SB334867 inhibited the excitatory effects of LHox/dyn stimulation. Notably, ox can interact with both Ox1R and Ox2R that are expressed within the VTA (Marcus et al., 2001; Narita et al., 2006). The affinity of SB334867 is 40 nM at Ox1R which is 50-fold selective over Ox2R (Smart et al., 2001); however a bath application of 1 mM likely has inhibitory action at both Ox1R and Ox2R. Ox1R and Ox2R are typically expressed postsynaptically on VTADA and some GABAergic neurons (Fadel and Deutch, 2002; Balcita-Pedicino and Sesack, 2007), although some reports have demonstrated a presynaptic action of ox A in the VTA (Borgland et al., 2009). Although few Gq-coupled receptors are expressed presynaptically, Ox1 and Ox2 receptors are known to be promiscuous in their G-protein alpha subunit coupling (Kukkonen and Leonard, 2014). KORs are also expressed on both somatodendrites of VTADA neurons but mediate their postsynaptic effects via opening a G-protein-coupled inwardly rectifying potassium conductance via activation of Gi/o-coupled KORs. KORs can also be expressed at terminals within the VTA, such that dyn can suppress both excitatory and inhibitory synaptic transmission onto VTADA neurons (Margolis et al., 2005; Ford et al., 2006). Importantly, in these experiments, photostimulation of LHox/dyn activated or inhibited firing in the presence of synaptic blockers, thus negating possible presynaptic or indirect activation. Taken together, neuromodulatory action of LHox/dyn photostimulation of VTADA neurons was mediated by postsynaptic somatodendritic Ox1R or KOR activation.
The time courses for the LHox/dyn neuromodulatory effect on VTADA neurons were different for either peptide. LHox/dyn optical stimulation produced a sustained decrease in firing. Because this effect could be washed off with NorBNI, it suggests that dyn lingers in the slice possibly due to decreased metabolic peptidases for dyn that may be washed out with continuous superfusion of the slices. Exogenous application of dyn (5 min) to VTA slices in the presence of peptidase inhibitors, captopril and beestatin, also produced a lasting depression of VTADA firing that did not return to baseline 15 min after washout (Baimel et al., 2017). This is in contrast to a small molecule KOR agonist, U69593, that has transient effects in the VTA (Margolis et al., 2003), suggesting that the lasting effect is unlikely to be due to receptor kinetics. Notably, dyn application in the absence of peptidase inhibitors to ox neurons also produced a transient inhibition of firing (Li and van den Pol, 2006). Taken together, we speculate that the lasting effects of dyn release on VTADA firing are due to poor washout and decreased metabolism of the peptide in the slice preparation. In contrast, the effects of ox were shorter lasting, with a significant decay in response 20 min after application, likely due to faster degradation or washout of ox. Additionally, the differential effects of dyn and ox on VTADA firing can be attributed to their distinct signaling mechanisms (Bruchas and Chavkin, 2010; Kukkonen and Leonard, 2014). Exogenous application of ox (5 min) also has a longer lasting increase of VTADA firing, elevated 15 min after application (Baimel et al., 2017). Notably, we were able to evoke presumed ox release for a second time in VTA slices. This suggests that there are sufficient dense core vesicles to mobilize for repeated release, which may have implications for how LHox/dyn neurons can signal in regions with high density of vesicles. The smaller inhibition with a second stimulation of LHox/dyn is likely because of a ceiling effect of KOR activation.
Previous work has demonstrated that exogenous application of ox and dyn modulates nonoverlapping DAergic circuits originating from the VTA to tune dopaminergic output (Baimel et al., 2017). Specifically, exogenous application of ox potentiates firing of VTADA neurons that project to the lAcbSh, but not the BLA, whereas exogenous dyn inhibits firing in subpopulations of DA neurons in both mAcbSh and lAcbSh and inhibits most BLA-projecting DA neurons (Baimel et al., 2017). While LH ox neurons provide the only source of ox to the VTA (Peyron et al., 1998), dyn is also expressed in the substantia nigra, prefrontal cortex, ventral and dorsal striatum, central amygdala, and dorsal raphe (Watson et al., 1982; Weber et al., 1982; Fallon and Leslie, 1986; Healy and Meador-Woodruff, 1994; Abraham et al., 2022). The dyn-expressing ventral striatal and dorsal raphe neurons project to the VTA and supply a dyn modulatory input in addition to that of LHox/dyn neurons (Shippenberg et al., 2001; Abraham et al., 2022). Given that LHox/dyn neurons are activated during arousing situation, such as stress or motivationally advantageous opportunities, it is important to determine how neuropeptides from this particular input can tune selective VTA originating circuits. We found that, like exogenous application, LH dyn decreased the activity of the majority of BLA-projecting VTADA neurons (Baimel et al., 2017). BLA DA can shape attention-related learning signals and is involved in encoding identity specific cue memories (Esber et al., 2012; Sias et al., 2024). Thus, a suppression of this input during periods when LHox/dyn neurons are activated may reduce formation of these memories, although further research is required to test this hypothesis.
We found that LHox/dyn activation of NAc-projecting VTADA neurons had differential effects dependent on the NAc shell subregion. Whereas the majority of VTADA neurons that projected to the lAcbSh were activated by LHox/dyn stimulation, VTADA neurons projecting to the mAcbSh were primarily inhibited. However, both projections had a large proportion of neurons that also responded to dyn, consistent with studies using exogenously applied dyn (Ford et al., 2006; Baimel et al., 2017). This differentiation may arise, to some extent, from differences in the expression of Ox1R or KOR on VTA neurons projecting to the mAcbSh versus the lAcbSh. However, conclusive evidence for this hypothesis is yet to be established. Taken together, one reason why ox and dyn may be coexpressed and coreleased is to simultaneously modulate different VTADA ensembles that can then influence their downstream projections.
VTADA neurons exhibit heterogeneity in their axonal projections, electrophysiological characteristics, and various molecular features. However, the functional consequences of this diversity on behavior remain poorly elucidated. DA projections from the VTA to the NAc play an important role in motivated behaviors, reinforcement learning, and reward processing (Hamid et al., 2016; Kim and Kaang, 2022). Recent studies have shown the contribution of VTA projections to the amygdala in encoding state-specific motivational salience (Lutas et al., 2019), regulating approach/avoidance behavior toward threats (Miller et al., 2019), and have established the role of VTA projections in modulating BLA activity during aversive conditioning (Tang et al., 2020). An interesting example of how VTADA projections to the NAc or the amygdala are influenced in response to nicotine administration revealed that a nicotine injection induces opposing responses in two distinct subpopulations of VTADA neurons (Nguyen et al., 2021). Specifically, a majority of lateral VTADA projections to the NAc are activated, while a considerable majority of medial VTADA neurons with axons projecting to the amygdala are inhibited after systemic nicotine (Nguyen et al., 2021). Nguyen et al. (2021) also demonstrated that both rewarding and anxiogenic effects of nicotine exposure occur simultaneously and are conveyed by distinct subpopulations of VTADA neurons, such that inhibition of amygdala-projecting VTADA neurons mediates anxiety-like behavior and their activation prevents the anxiogeneic effect of nicotine, while activation of NAc-projecting VTADA neurons likely mediates the reinforcing effect of nicotine (Nguyen et al., 2021). Future research should examine how the concurrent engagement of two circuits with opposing messages could compete to produce specific functional outcomes and whether an imbalance between the two could lead to brain disorders such as addiction.
In summary, we demonstrate that BLA-projecting VTADA neurons were predominantly inhibited in response to LHox/dyn optical stimulation, whereas lAcbSh-projecting and mAcbSh-projecting VTADA neurons were bidirectionally modulated. Thus, it would be predicted that corelease of LH ox and dyn might balance ensembles of VTADA neurons within different projections to influence their final output. Given that ox signaling in the VTA drives motivated reward seeking, we speculate that the effects of ox and dyn corelease in the VTA may act to coordinate dopaminergic output to bias activity toward projection targets such as the NAc, which are critical for effort-driven reward seeking (Correa et al., 2002), while dampening activity in other circuits that are less essential in these tasks.
Footnotes
We acknowledge the Hotchkiss Brain Institute advanced microscopy facility. This research was performed at the University of Calgary which is located on the unceded traditional territories of the people of the Treaty 7 region in Southern Alberta, which includes the Blackfoot Confederacy (including the Siksika, Piikani, and Kainai First Nations), the Tsuut’ina, and the Stoney Nakoda (including the Chiniki, Bearspaw, and Goodstoney First Nations). The City of Calgary is also home to the Metis Nation of Alberta, Region III. This work is supported by a Mathison Centre for Research and Education Neural Circuits research grant, Tier 1 Canada Research Chair (950-232211) and National Science and Energy Research Council grant (Discovery Grant: RGPIN-2023-03428 to S.L.B.).
The authors declare no competing financial interests.
- Correspondence should be addressed to Stephanie L. Borgland at s.borgland{at}ucalgary.ca.
This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed.
