Reduced GABA transmission onto ventral tegmental area dopamine neurons underlies vulnerability to a mouse model of Anorexia Nervosa

doi:10.21203/rs.3.rs-5037203/v1

Download PDF

Article

Reduced GABA transmission onto ventral tegmental area dopamine neurons underlies vulnerability to a mouse model of Anorexia Nervosa

https://doi.org/10.21203/rs.3.rs-5037203/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Anorexia nervosa (AN) has the highest mortality among psychiatric diseases. Hyperactivity is a persistent symptom and alteration of mesolimbic dopamine transmission has been linked to the development and maintenance of the disease and of hyperactivity. However, whether local mesolimbic neurocircuit plasticity is causally involved remains unclear. Especially the role of local GABA control over dopamine neurons, a powerful regulator of the dopamine system, in an AN context is unresolved. We hypothesize that combining caloric restriction with exercise alters dopamine transmission via GABA disinhibition that, in turn, facilitates the expression of maladaptive behaviors such as hyperactivity.

Therefore, we characterized the impact of the activity-based anorexia (ABA) model on the plasticity of the dopamine reward system using ex-vivo electrophysiology coupled with optogenetic manipulations. Ventral tegmental area dopamine (VTA_DA) neurons displayed a higher firing frequency in ABA-exposed animals compared to control mice. This coincided with reduced GABAergic transmission on VTA_DA neurons, at least in part attributable to decreased excitability of local VTA GABA (VTA_GABA) neurons. Restoring the excitability of VTA_GABA neurons via chemogenetic activation rescued mice from starvation, by decreasing running wheel activity.

In summary, we found that the anorexic state leads to dysregulation of VTA_GABA transmission on VTA_DA neurons that reinforces maladaptive behaviors such as hyperactivity. We uncovered a new mechanism linked to the disturbed dopamine system in ABA-exposed animals, identifying a hitherto unknown role of decreased local GABAergic control over VTA dopamine neuron output.

Biological sciences/Neuroscience/Diseases of the nervous system

Biological sciences/Neuroscience/Reward

Anorexia nervosa (AN) is a metabo-psychiatric eating disorder characterized by a fear of gaining weight and self-starvation, leading to life-threatening weight loss (1, 2). Up to 80% of AN patients engage in excessive exercise and displaying restlessness (3, 4). Hyperactivity interferes with weight gain and predicts relapse and chronicity of AN and therefore represents a major barrier to recovery (5, 6).

The dopamine (DA) system has been implicated in the neurobiology underlying AN since the 70s (7) and has been strengthened by recent clinical evidence of dysfunction of the reward system in AN (8–12). Recovered AN patients exhibit reduced levels of a DA metabolite in cerebral spinal fluid (13) and increased DA D2/D3 receptor binding in the ventral striatum (10), although the latter was not confirmed in another study (14). Furthermore, AN patients display altered reward processing associated with differential responses in the ventral striatum depending on the rewarding stimulus (15), reduced VTA / substantia nigra to nucleus accumbens (NAc) functional connectivity (16) and increased responsiveness of reward circuits to food stimuli (11). An altered effective connectivity was found in AN patients between the hypothalamus and the DA reward system when anticipating a sweet reward (12). These findings indicate alterations in DA signaling in AN, but the precise nature and mechanism underlying these alterations remain unknown.

Animal studies support the link between the DA reward system and the regulation of energy homeostasis. DA transmission from ventral tegmental area (VTA) is involved in energy homeostasis by regulating both exercise and feeding behaviors (17–22). The activity-based anorexia (ABA) model is typically used to better understand the neurobiology of AN and evaluate potential treatment targets (for review see (23, 24)). Manipulation of dopamine transmission in the ABA model facilitated the development of behaviors that resemble AN symptoms, such as excessive exercise and weight loss (for review see (25)). Mice without the dopamine transporter gene (DAT knockouts) are hyperdopaminergic and more susceptible to lose weight in the ABA model due to accelerated development of excessive running (26). Dopamine receptor 2 (D2) antagonists, reduced locomotor activity and prevented “self-starvation” in rats (27, 28). Overexpression of D2 receptors in ventral striatum increased hyperactivity and the susceptibility to lose weight in the ABA model (29) and an imbalance between D1- and D2-expressing neuron activities in the ventral striatum leads to anorexia-like phenotype (22). These rodent data align with patients with AN displaying hyperactivity in particular during the acute phase of the disease (3). Strikingly, it is still unclear how environmental factors that drive negative energy balance, such as food restriction and the availability to exercise, impact on DA transmission.

Although alterations have been observed in the DA system of AN patients and manipulations of dopamine transmission have been shown to impact on AN-like symptoms in animals exposed to the ABA model, there remains a gap in understanding the specific alterations occurring in the dopamine system. Alterations in intrinsic properties of DA neurons may be causal to changed DA transmission, and GABA transmission has been proposed as a key player in ABA vulnerability and maladaptive behaviors. Individuals with the greatest expansion of the GABAergic synapses on hippocampal and prefrontal pyramidal neurons were the most resilient to ABA exposure (30). Moreover, running performances has been linked to disinhibition of VTA DA neurons (31), and lesion/inhibition of VTA GABAergic neurons generates mania-like behaviors (32). In a study using deep brain stimulation in the VTA, suppression of hyperactivity was attributed to increased VTA GABA activity in a mouse model of schizophrenia (33). Therefore, GABAergic inhibition of the dopamine system emerges as a candidate mechanism, given its potent regulatory influence on dopamine neuron output (34, 35). To date, the impact of ABA-exposure on VTA_DA neurons and the involvement of GABA transmission in DA alterations is unknown. We hypothesize that hyperactivity of the midbrain DA system is involved in hyperactivity and body weight loss through local GABA disinhibition of VTA_DA neurons. Therefore, we assessed 1) the consequences of the ABA model on DA neurons in the VTA, 2) the GABAergic origin of the VTA_DA neuronal alterations and 3) the influence of modulating GABA neurons on the vulnerability to the ABA model. This work contributes to a better understanding of the interactions between energy balance regulation, the newly-described VTA_GABA alterations and its impact on VTA_DA neurons. Ultimately, it leads to a deeper comprehension of AN etiology and suggests that targeting the VTA_GABA system has therapeutic potential.

Experimental model and subject details

In all experiments, naïve young adult female mice were used (18–30g, 2–5 months). Before experiments started, all mice were group-housed in eurostandard type II cages (n = 2–6, Tecniplast) in a 12:12 light/dark cycle (lights on at 07:00 a.m.) at 22 ± 2°C (60–65% humidity) and properly handled. Unless otherwise specified, animals had access to ad libitum water and lab chow (3.61 kcal/g, CRM(E), 801730, Special Diets Services) and cage enrichment. The following mouse lines were used in this study: C57Bl6J (Jax#664), Pitx3-GFP (Jax#041479) and VGAT-Cre (Jax#028862) mice. Animals were bred in house but originated from the Jackson Laboratory. Homozygous VGAT-cre females were crossed to homozygous or heterozygous Pitx3-GFP males in order to generate double transgenic animals. Experiments were approved by the Animal Ethics Committee of Utrecht University and the Dutch Central Authority for Scientific Procedures on Animals (CCD#AVD1150020198686), and were conducted in agreement with the Dutch law (Wet op de Dierproeven, 2014) and the European regulations (Guideline 86/609/EEC).

Activity-Based Anorexia (ABA) model and ad libitum running/feeding setup

Mice were single housed and randomly attributed, based on body weight (0.01gr resolution), to one of three groups: wheel running (WR), activity-based anorexia (ABA) and food restricted (FR). To reach stable running wheel activity (RWA), all mice were habituated for 10 days with ad libitum access to food and a running wheel in the unlocked (WR and ABA groups) or locked (FR) configuration. At ABA0, food was removed from FR and ABA mice 3 h after the start of the dark cycle. On all following days, food was provided for the first 3h of the dark cycle. For mice in the WR group, food was measured at dark onset daily, but not removed. Body weight was measured in all mice immediately prior to the start of the dark cycle (before food presentation). RWA was continuously collected and analyzed using a Cage Registration Program (Department Biomedical Engineering, UMC Utrecht, The Netherlands). Food anticipatory activity (FAA) was defined as RWA recorded during the last 6h of the light phase (unless otherwise indicated). We defined as humane endpoint (HEP) more than 20% loss of initial body weight that was not recovered following food exposure. The 20% loss of initial body weight was used as cut off for survival analyses. Mice were habituated to intraperitoneal (i.p.) injections (100ul of saline/10g of mice) for at least three days before the onset of the behavioral paradigm (ABA 0). In Fig. 4, the habituation phase described above was followed by an ad libitum protocol (for both food and running) in order to assess the dose-dependent effect of CNO on feeding, RWA and variation of body weight.

CNO administration

For chemogenetic experiments, Clozapine N-oxide (CNO) dihydrochloride (HB6149, Hello Bio) was dissolved in saline and injected i.p. at different doses (from 0.04mg/kg to 2mg/kg) with a volume of 100ul/10gr of mice. For chemogenetic inhibition, CNO was injected once a day, 5 hours before dark onset, at 0.5mg/kg. For chemogenetic activation, CNO was injected twice a day in order to have a better coverage of the period in which CNO was in the system (5 hours before dark onset and after feeding) at 0.25mg/kg. The measurements of body weight and food intake were performed at dark onset and after feeding. FAA was defined as the 5 hours following CNO injections.

For CNO dose determination in the ad libitum feeding / running protocol, different CNO doses were randomized and injected at dark onset daily (ranging from 0.25 to 2mg/kg for inhibition and from 0.04 to 0.625mg/kg for activation). The measurements of body weight and food intake were performed at injection time and 3 hours later. RWA activity was measured as the sum of revolutions performed during these 3 hours.

Stereotaxic Surgeries for Viral Injections

Stereotaxic injections of viral vectors were performed with a stereotaxic apparatus (UNOB.V. model 68U025) as previously described (36). Mice were injected bilaterally with a volume of 300nl per side at an injection rate of 100nl/min with pAAV5-hSyn-DIO-hM4D(Gi)-mCherry (44362-AAV5, AddGene), pAAV-hSyn-DIO-hM3D(Gq)-mCherry (44361-AAV5, AddGene), pAAV-EF1a-double floxed-hChR2(H134R)-mCherry-WPRE-HGHpA (20297-AAV5, AddGene) or pAAV-hSyn-DIO-mCherry (50459-AAV5, AddGene) at titer 3.3*10^12gc/ml in the VTA (AP:-3.2, ML: 1.6, DV: -4.8 from bregma, angle 15°, Paxinos and Franklin). Experiments were performed 3–6 weeks after AAV stereotaxic injection. Injection sites were checked in all animals by preparing sagittal sections of 250µm for electrophysiology or 50µm after chemogenetic in vivo experiments. As in all mice the VTA was hit, but often with some variable spread to surrounding GABA neurons, we refer to these neurons as midbrain GABA neurons.

Slice Preparation, Cell-attached and Whole-Cell Patch Clamp Recording

Mice were anesthetized with isoflurane at 12 p.m. (corresponding to dark onset) and then rapidly decapitated. The brain was quickly removed and coronal slices (250µm) were prepared as described previously (36). In Pitx3-GFP mice, GFP-positive dopaminergic cells were recorded within the VTA. In VGAT-cre/Pitx3-GFP mice injected with cre-dependent AAV-ChR2-mCherry, GFP+/mCherry- dopaminergic cells were recorded within the VTA. In VGAT-cre/Pitx3-GFP mice injected with cre-dependent AAV-mCherry, mCherry+/GFP- GABAergic cells were recorded within the VTA.

For cell-attached recordings of dopaminergic neurons, we used filtered extracellular medium in the glass pipette and pipette resistance during recording was between 20-300MΩ. Recordings were made in voltage clamp gap free mode. Cells were clamped at 0mV and kept in this configuration for at least 10min. To record basic intrinsic properties of VTA_DA and VTA_GABA neurons, a potassium-gluconate internal solution was used and cells were held in current-clamp mode in the absence of any current injections. A series of depolarizing current steps of 800ms starting at -105pA with 15pA increment were then applied. IPSCs were recorded with a cesium-chloride internal solution in the presence of 10µM CNQX (6-cyano-7-nitroquinoxaline-2,3-dione, Tocris 1045) and 50µM D-AP5 (D-2-amino-5-phosphonovalerate, Tocris 0106). Neurons were voltage clamped at -70mV. Recordings of spontaneous (s)IPSCs were performed at least 10min after patch for 1min. PPR recordings of VTA-driven GABA_AR-mediated currents were performed in the presence of 10µM CNQX and 50µM D-AP5. GABA_AR mediated currents were measured in response to optostimulation of VTA_GABA terminals using two pulses with 50, 100 and 200ms inter-pulse interval. The PPR was calculated by dividing the average amplitude of the evoked synaptic response (eIPSC) to the 2nd pulse by the amplitude of the eIPSC to the 1st pulse. Light pulses: 470 nm, 0.1 ms; 1-10mW. Data were analyzed offline using Clampfit 11.2 (MolecularDevices).

Quantification and statistical analysis

Student’s t-tests (unpaired and paired) or equivalent non-parametric tests when data were not normality distributed (Mann-Whitney and Wilcoxon matched-pairs signed rank test) were used when two groups were compared. Welch’s correction was applied on unpaired t-test when variances were not equal. One-way ANOVA test or equivalent non-parametric tests when normality failed (Kruskal-Wallis test) were used when 3 groups were compared. Two-way ANOVA tests were used when 2 factors were present (Two-way repeated-measure ANOVA when one of the factors was a repeated-measure). Sidak’s post hoc test was applied when ANOVA showed a significant interaction or a significant main effect. Pearson's Correlation Coefficient was used to measures the strength of the linear relationship between two variables. Log rank Mantel Cox test was used to compare survival curves. All statistical data were obtained using GraphPad Prism 7 (Graphpad Software). Statistical significance was *p < 0.05, **p < 0.01, ***p < 0.001 (#p < 0.1 was used as an indication of a trend). All data are presented as means ± SEM. Details of the statistical analysis per figure are summarized in Tables S1 and S2 (main and supplementary figures respectively).

Mice exposed to the ABA model lose weight and develop excessive running wheel activity

We exposed female mice to the ABA model. The ABA protocol consisted of multiple days with a 3h feeding period at dark onset, combined with unlimited access to a running wheel (Fig. 1A). Compared to FR (exposed to food restriction but with no wheel access) and WR (running wheel access and no food restriction) controls, ABA mice lost more body weight (Fig. 1B) while they ate a similar amount of food as FR mice and both groups ate less than WR controls (Fig. 1C). During the 5 days of ABA exposure, increased light phase RWA develops (Fig. 1D). This maladaptive behavior, defined as food anticipatory activity (FAA), occurs the hours before food exposure at dark onset. Overall, ABA and wheel running (WR) animals display similar 24h running distances (Fig. 1.E). ABA mice decrease dark phase running activity, especially when food is available (Fig. 1D), while FAA develops as running shifts to the light phase (Fig. 1F). We found that body weight was positively associated with food intake (Fig.S1A) and negatively with FAA (Fig.S1B). Moreover, the animals displaying the highest RWA activity before ABA starts were more vulnerable to develop FAA and to lose body weight (Fig.S1C-D), similarly to what has been previously found (37). Thus, in the ABA model the combination of reduced food intake and access to a running wheel leads to negative energy balance and produce AN-like symptoms.

ABA exposure leads to increase VTA_DA neuronal activity

For electrophysiological cell-attached patch-clamp recordings on ex vivo brain slices, we harvested the Pitx3-GFP female brains at ABA5 just before dark onset, i.e. after FAA occurs in ABA animals and before food exposure for ABA and FR groups. (Fig. 2A). ABA exposure did not affect the proportion of spontaneously active VTA_DA neurons (Fig. 2B), known to display a regular “pace-making” activity (38). However, the firing frequency of the spontaneously active VTA_DA neurons was increased in ABA-exposed animals compared to WR and FR groups (Fig. 2C-D and S2B). Thus, ABA-exposed animals display an increased frequency of VTA_DA neuronal activity compared to controls.

We next investigated if this difference in firing frequency was due to changes in cell excitability (i.e. intrinsic electrical properties). Results are presented in Fig. 2E-I and Fig.S2C-E where whole-cell patch-clamp recordings were performed allowing to access the intracellular compartment of VTA_DA neurons (Fig. 2E). We found that ABA exposure did not affect the intrinsic properties of VTA_DA neurons. Input/output curves, resting membrane potential (RMP), membrane resistance (Rmem), rheobase, action potential (AP) threshold were unchanged (Fig. 2F-I and Fig.S2D-E). To summarize, VTA_DA neurons from ABA-exposed animals display an increased firing frequency compared to control groups, which is not due to alterations of basic membrane properties of VTA_DA cells.

Reduced inhibitory transmission from local VTA GABA neurons to VTA_DA neurons in ABA-exposed animals

As intrinsic properties of VTA_DA cells were unaltered after ABA, the integrity of neuronal inputs they received - influencing neuronal activity by excitation (i.e. glutamatergic transmission) or inhibition (i.e. GABAergic transmission) – was then assessed. Spontaneous glutamatergic excitatory transmission onto VTA_DA neurons was unaltered in ABA animals (Fig.S3A-B) whereas spontaneous GABAergic inhibitory transmission was reduced following ABA exposure (Fig. 3A-B). Specifically, the frequency of spontaneous inhibitory post-synaptic currents (sIPSC) was decreased in the ABA group whereas the average amplitude of sIPSCs was similar between groups (Fig. 3A-B). The difference in IPSC frequency was not found anymore after application of tetrodotoxin (miniature IPSCs, Fig.S3C-D). In ABA-exposed animals, there was a reduction in GABAergic signals onto VTA_DA neurons, likely involving a presynaptic but action potential-dependent mechanism. The GABAergic deficits in the ABA condition were specific to the synapse, as we did not observe any alterations of extrasynaptic GABA-A mediated tonic conductances (Fig.S3E-G).

To further assess presynaptic GABA release, a paired-pulse ratio (PPR) protocol was performed consisting of a paired stimulation of a synapse given in a quick succession with 3 different time intervals (Fig. 3E and S3H). Electrical stimulation of synapses resulted in an overall increased PPR in the ABA group (FigS3I) which is in accordance with a scenario in which the probability of GABA release is reduced when compared to the WR group (39). We predicted that local GABA neurons were involved considering their powerful regulatory function onto dopamine neurons. Therefore, the PPR was performed using optogenetic stimulations on midbrain GABA neurons surrounding VTA_DA cells (Fig. 3D-E). This protocol revealed an increased PPR in the ABA group (Fig. 3F) showing that local midbrain GABA to VTA_DA synapses displayed reduced probability of GABA release. To further corroborate a deficit in local GABAergic VTA interneurons, whole-cell patch-clamp recordings of VTA_GABA neurons were performed (Fig. 3G). VTA_GABA neurons from ABA-exposed animals displayed decreased excitability: RMP (Fig. 3H) and rheobase (Fig. 3J) were found decreased and increased respectively whereas Rmem (Fig. 3I) and AP threshold (Fig. 3K) were unchanged. This results in a significantly shifted current-voltage plot (Fig. 3L) as well as a trend of a decreased ability to trigger action potentials with increasing steps of current injections (Fig. 3M). ABA exposure also affects the proportion of spontaneously active VTA_GABA neurons (Fig. 3N) and therefore reduces their firing frequency compared to WR controls regardless of the chosen experimental unit, the cell (Fig. 3O) or the animal (Fig.S3J). Thus, inhibitory transmission from local VTA_GABA neurons onto VTA_DA neurons is weaker in the ABA group compared to the WR group.

Chemogenetic inhibition of midbrain GABA neurons exacerbates negative energy balance induced by the ABA model while their activation rescues from starvation

To study the causal involvement of ventral midbrain GABA neurons on running wheel activity and food intake, we chemogenetically targeted these neurons. Adeno-associated viruses with Cre-dependent DREADDs (either excitatory hM3Dq or inhibitory hM4Di) were injected in the VTA of vesicular GABA transporter (VGAT)-cre mice, (Fig. 4A and S4A-C).

We first tested whether this GABAergic network regulates food intake and RWAin an ad libitum situation (Fig. 4B). Different doses of CNO were randomly administered every day. Inhibition of midbrain GABA neurons led, for most of the doses used, to decreased food intake (Fig. 4C) as well as increased RWA (Fig. 4D) within the 3h following CNO administration. This led to a mild but significant decrease in body weight expressed as relative body weight within these 3h and compared to the saline injection day (Fig. 4E). In contrast, activating midbrain GABA neurons led to increased food intake and decreased RWA resulting in a significant increase in body weight (Fig. 4F-H) when doses of 0.1-0.25mg/kg were used. However, when midbrain GABA neurons are activated at the highest CNO dose tested, running is drastically suppressed and food intake mildly increased suggesting the induction of motor-based side effects. Based on these results, we identified the optimal CNO doses for both chemogenetic inhibition (0.5 mg/kg) and activation (0.25 mg/kg). Importantly, with these doses, we found no alterations in motor coordination nor in DREADD-Gi animals (Fig.S4D) neither in DREADD-Gq animals (Fig.S4E). Moreover, CNO had no effect in control animals that express the control fluorophore, regardless of the tested dose. To summarize, midbrain GABA neurons are involved in energy balance regulation by acting on both feeding and exercise in an ad libitum situation.

We next chemogenetically targeted ventral midbrain GABA neurons in ABA-exposed animals (Fig. 5A-B). As shown in Fig. 5C-F, animals expressing the inhibitory DREADD-Gi and the control group were exposed to ABA and injected with 0.5 mg/kg CNO every day 5h before food exposure (when most of the animals start expressing FAA). Inhibiting midbrain GABA neurons increased the ABA-induced body weight loss (Fig. 5C), compared to ABA-controls. There were no changes in daily food intake or total RWA (Fig. 5D and S5C). However, light phase RWA and FAA were increased in the DREADD-Gi group (Fig. 5E and S5A). Thus, inhibition of midbrain GABA neurons accelerated the development of FAA leading to an increased vulnerability to the ABA model (Fig. 5C and 5F).

In Fig. 5G-J, activation of midbrain GABA neurons with 0.25 mg/kg CNO twice a day was protective against body weight loss (Fig. 5G). Daily measurements of food intake or total RWA did not reveal any differences (Fig. 5H and S5D) but FAA was drastically suppressed (Fig. 5I and S5B). These data suggest that excessive running that develops along ABA exposure can be suppressed by increasing midbrain GABA neuronal activity, resulting in a better survival to the ABA model (Fig. 5G and J). Thus, restoring the excitability of midbrain GABA neurons via chemogenetic activation reduced body weight loss by decreasing running wheel activity. Conversely, chemogenetic inhibition increased vulnerability to the ABA model by increasing FAA.

We identified a VTA neural circuit underlying hyperactivity and excessive weight loss in an animal model for anorexia nervosa. ABA-exposed mice display increased activity of VTA_DA neurons co-occurring with a decreased GABA release probability and reduced neuronal excitability of local VTA_GABA cells, the second likely responsible for the first. These results support that the midbrain dopamine system undergoes plasticity under the condition of restricted feeding and voluntary exercise. Restoring activity of GABA neurons by chemogenetics decreases vulnerability to the ABA model by suppression of excessive running and reduction of body weight loss. We identified a novel mechanism explaining how negative energy balance leads to dysregulation of the midbrain dopamine system that in turn reinforces maladaptive behaviors such as hyperactivity.

Disinhibition of the dopamine system may be a key candidate for the establishment of negative energy balance. GABA (dys)functions in AN and GABAergic drugs have hardly been studied in AN therapy (40, 41). Alterations of GABA transmission were linked to hyperactivity, decreased food intake and anxiety levels in AN rodent models (42), including within the VTA (32, 43). Decreased GABA levels were found in the hypothalamus of ABA rats and indirect rescue of GABA levels (with the neuropeptide kisspeptin) increased food intake (44). Within amygdala circuits, ABA-exposure compromised inhibition [42]. In another study, increased inhibition in rodent hippocampal pyramidal neurons was found protective against the ABA-induced weight loss by indirectly suppressing hyperactivity (45). Therefore, ABA vulnerability may partly rely on GABAergic alterations and increased GABA transmission may rescue maladaptive behaviors such as excessive exercise (30). To our knowledge, the present study is the first to provide evidence for reduced VTA_GABA transmission being involved in AN-like symptoms, such as hyperactivity, likely due to increased VTA_DA activity.

The results obtained in this study are in accordance with the hypothesized relationship between the hyperdopaminergic function and the regulation of energy expenditure (46): low dopamine function leads to restrictive behaviors (conserve energy/exploit environment) whereas higher dopamine function drives expansive behaviors (expend energy/explore environment). This is supported by a growing number of studies manipulating dopamine transmission in ABA-exposed animals and showing that increased dopamine transmission leads to higher vulnerability to body weight loss (22, 26, 29, 47). Running wheel activity has been shown to increase striatal DA release and induce a hyperdopaminergic state (48, 49). Chemogenetic activation of VTA_DA neurons and neurons projecting from the VTA to the NAc leads to hyperlocomotion (17), further supporting the idea that activation of the DA reward system and hyperactivity are causally linked. Surprisingly, one study found that activating VTA to NAc neurons prevented and rescued body weight loss (by increasing food intake) induced by the ABA model (50), but in that study non-dopaminergic projection neurons were likely also targeted. Recently, a biphasic model of AN has been proposed with 1) a hyperdopaminergic state in the acute phase, accompanied by symptoms such as hyperactivity and 2) a chronic low dopamine function resulting in behavioral rigidity (25). This model supports the clinical observations of low DA metabolite levels and increased D2 binding in recovered AN patients. That said, the mechanisms and brain circuits linking AN-like symptoms and the dopamine system are still poorly understood. In this context, our data suggest that reduced VTA_GABA neuron activity is causal to negative energy balance by increasing dopamine function leading to the development of maladaptive behaviors such as hyperactivity.

The ABA paradigm is a well-established rodent model of AN (23, 51) and the development of FAA is a hallmark and intriguing phenomenon. We hypothesized that FAA, and more generally hyperlocomotion, could occur because of an urge to exercise originating from alterations of VTA neuronal activity. Indeed, in patients with AN, the hyperactivity that develops despite negative consequences may originate from alterations at the motivational level (3, 52) and future research would be needed to study the competition between motivation to eat and motivation to exercise that the ABA model cannot fully disentangle (53). Nonetheless, this study suggests that neural circuits in the VTA, known to be involved in other forms of compulsive/addictive behaviors (54), may participate to the development of excessive exercise in eating disorders.

Altogether, our data brings new insights on the impact of running and food restriction on the reward system, allowing to better understand the neuronal mechanisms involved in energy balance regulation. Interestingly, Beeler and Burghardt recently hypothesized that the combination of caloric restriction with exercise, triggers an escalating spiral of increasing dopamine reinforcing weight-loss behaviors (25). Our results support this hypothesis. The results found in this study also suggest for the first time that a discrete neuronal population, namely VTA_GABA neurons, is directly contributing to hyperactivity and favor restlessness. Indeed, ABA-exposed mice expend more energy because they may be “restless” and activation of VTA_GABA neurons decreases excessive locomotor activity. This is particularly relevant in the context of AN since restlessness is a symptom of AN (4, 55) and our results may contribute to a new therapeutic avenue to explore. Overall this study contributes to further understanding of how changes in VTA neuronal activity and neural circuit interact to impact on the susceptibility to develop anorectic behavior in the ABA model.

COMPETING INTERESTS

The authors declare no competing interests.

AUTHOR CONTRIBUTIONS

R.A. and F.D. designed experiments with the contribution of F.M. F.D. performed behavioral, chemogenetic and ex vivo electrophysiology experiments. E.B., L.D., J.H and M.V. helped with behavioral experiments. F.D., I.W.D., M.L and K.K. performed stereotactic surgeries. F.D. performed analysis on all experiments. F.D., R.A. and F.M. wrote the manuscript with input from all authors.

ACKNOWLEDGEMENTS

We would like to thank the entire Adan and Meye Labs for discussions and critical reading of the manuscript. This research was supported by the Dutch Research Council NWO ALWOP.137, OCENW.M.22.111 and Gravitation grant 024.004.012; to RAH and the Fyssen Fundation; to FD.

Schmidt U, Adan R, Böhm I, Campbell IC, Dingemans A, Ehrlich S et al (2016) Eating disorders: the big issue. Lancet Psychiatry 3:313–315
Anorexia Nervosa Genetics Initiative, Eating Disorders Working Group of the Psychiatric Genomics Consortium, Watson HJ, Yilmaz Z, Thornton LM, Hübel C et al (2019) Genome-wide association study identifies eight risk loci and implicates metabo-psychiatric origins for anorexia nervosa. Nat Genet 51:1207–1214
Davis C, Katzman DK, Kaptein S, Kirsh C, Brewer H, Kalmbach K et al (1997) The prevalence of high-level exercise in the eating disorders: Etiological implications. Compr Psychiatry 38:321–326
Casper RC (2022) Restlessness and an Increased Urge to Move (Drive for Activity) in Anorexia Nervosa May Strengthen Personal Motivation to Maintain Caloric Restriction and May Augment Body Awareness and Proprioception: A Lesson From Leptin Administration in Anorexia Nervosa. Front Psychol 13:885274
Carter JC, Blackmore E, Sutandar-Pinnock K, Woodside DB (2004) Relapse in anorexia nervosa: a survival analysis. Psychol Med 34:671–679
Casper RC (2016) Restless activation and drive for activity in anorexia nervosa may reflect a disorder of energy homeostasis. Int J Eat Disord 49:750–752
Barry VC, Klawans HL (1976) On the role of dopamine in the pathophysiology of anorexia nervosa. J Neural Transm 38:107–122
Kontis D, Theochari E (2012) Dopamine in anorexia nervosa: a systematic review. Behav Pharmacol 23:496–515
Cha J, Ide JS, Bowman FD, Simpson HB, Posner J, Steinglass JE (2016) Abnormal reward circuitry in anorexia nervosa: A longitudinal, multimodal MRI study: Abnormal Reward Circuitry in AN. Hum Brain Mapp 37:3835–3846
Frank GK, Bailer UF, Henry SE, Drevets W, Meltzer CC, Price JC et al (2005) Increased Dopamine D2/D3 Receptor Binding After Recovery from Anorexia Nervosa Measured by Positron Emission Tomography and [11C]Raclopride. Biol Psychiatry 58:908–912
Frank GKW, Reynolds JR, Shott ME, Jappe L, Yang TT, Tregellas JR, O’Reilly RC (2012) Anorexia Nervosa and Obesity are Associated with Opposite Brain Reward Response. Neuropsychopharmacology 37:2031–2046
Frank GKW, DeGuzman MC, Shott ME, Laudenslager ML, Rossi B, Pryor T (2018) Association of Brain Reward Learning Response With Harm Avoidance, Weight Gain, and Hypothalamic Effective Connectivity in Adolescent Anorexia Nervosa. JAMA Psychiatry 75:1071
Kaye W (1999) Altered Dopamine Activity after Recovery from Restricting-Type Anorexia Nervosa. Neuropsychopharmacology 21:503–506
Broft A, Slifstein M, Osborne J, Kothari P, Morim S, Shingleton R et al (2015) Striatal dopamine type 2 receptor availability in anorexia nervosa. Psychiatry Res Neuroimaging 233:380–387
Fladung A-K, Grön G, Grammer K, Herrnberger B, Schilly E, Grasteit S et al (2010) A Neural Signature of Anorexia Nervosa in the Ventral Striatal Reward System. Am J Psychiatry 167:206–212
Tadayonnejad R, Majid D-A, Tsolaki E, Rane R, Wang H, Moody TD et al (2022) Mesolimbic Neurobehavioral Mechanisms of Reward Motivation in Anorexia Nervosa: A Multimodal Imaging Study. Front Psychiatry 13:806327
Boekhoudt L, Omrani A, Luijendijk MCM, Wolterink-Donselaar IG, Wijbrans EC, van der Plasse G, Adan RAH (2016) Chemogenetic activation of dopamine neurons in the ventral tegmental area, but not substantia nigra, induces hyperactivity in rats. Eur Neuropsychopharmacol 26:1784–1793
Boekhoudt L, Roelofs TJM, de Jong JW, de Leeuw AE, Luijendijk MCM, Wolterink-Donselaar IG et al (2017) Does activation of midbrain dopamine neurons promote or reduce feeding? Int J Obes 41:1131–1140
Zhu X, Ottenheimer D, DiLeone RJ (2016) Activity of D1/2 Receptor Expressing Neurons in the Nucleus Accumbens Regulates Running, Locomotion, and Food Intake. Front Behav Neurosci 10. https://doi.org/10.3389/fnbeh.2016.00066
O’Connor EC, Kremer Y, Lefort S, Harada M, Pascoli V, Rohner C, Lüscher C (2015) Accumbal D1R Neurons Projecting to Lateral Hypothalamus Authorize Feeding. Neuron 88:553–564
Bond CW, Trinko R, Foscue E, Furman K, Groman SM, Taylor JR, DiLeone RJ (2020) Medial Nucleus Accumbens Projections to the Ventral Tegmental Area Control Food Consumption. J Neurosci 40:4727–4738
Walle R, Petitbon A, Fois GR, Varin C, Montalban E, Hardt L et al (2024) Nucleus accumbens D1- and D2-expressing neurons control the balance between feeding and activity-mediated energy expenditure. Nat Commun 15:2543
Zhang J, Dulawa SC (2021) The Utility of Animal Models for Studying the Metabo-Psychiatric Origins of Anorexia Nervosa. Front Psychiatry 12:711181
Scharner S, Stengel A (2021) Animal Models for Anorexia Nervosa—A Systematic Review. Front Hum Neurosci 14:596381
Beeler JA, Burghardt NS (2022) The Rise and Fall of Dopamine: A Two-Stage Model of the Development and Entrenchment of Anorexia Nervosa. Front Psychiatry 12:799548
Beeler JA, Mourra D, Zanca RM, Kalmbach A, Gellman C, Klein BY et al (2021) Vulnerable and Resilient Phenotypes in a Mouse Model of Anorexia Nervosa. Biol Psychiatry 90:829–842
Routtenberg A, Kuznesof AW (1967) Self-starvation of rats living in activity wheels on a restricted feeding schedule. J Comp Physiol Psychol 64:414–421
Klenotich SJ, Ho EV, McMurray MS, Server CH, Dulawa SC (2015) Dopamine D2/3 receptor antagonism reduces activity-based anorexia. Transl Psychiatry 5:e613–e613
Welch AC, Zhang J, Lyu J, McMurray MS, Javitch JA, Kellendonk C, Dulawa SC (2019) Dopamine D2 receptor overexpression in the nucleus accumbens core induces robust weight loss during scheduled fasting selectively in female mice. Mol Psychiatry. https://doi.org/10.1038/s41380-019-0633-8
Aoki C, Santiago AN (2022) Pathway-specific GABAergic inhibition contributes to the gain of resilience against anorexia-like behavior of adolescent female mice. Front Behav Neurosci 16:990354
Dubreucq S, Durand A, Matias I, Bénard G, Richard E, Soria-Gomez E et al (2013) Ventral Tegmental Area Cannabinoid Type-1 Receptors Control Voluntary Exercise Performance. Biol Psychiatry 73:895–903
Yu X, Ba W, Zhao G, Ma Y, Harding EC, Yin L et al (2021) Dysfunction of ventral tegmental area GABA neurons causes mania-like behavior. Mol Psychiatry 26:5213–5228
Lu C, Feng Y, Li H, Gao Z, Zhu X, Hu J (2022) A preclinical study of deep brain stimulation in the ventral tegmental area for alleviating positive psychotic-like behaviors in mice. Front Hum Neurosci 16:945912
Tan KR, Yvon C, Turiault M, Mirzabekov JJ, Doehner J, Labouèbe G et al (2012) GABA Neurons of the VTA Drive Conditioned Place Aversion. Neuron 73:1173–1183
Beier K (2022) Modified viral-genetic mapping reveals local and global connectivity relationships of ventral tegmental area dopamine cells. eLife 11:e76886
Linders LE, Patrikiou L, Soiza-Reilly M, Schut EHS, van Schaffelaar BF, Böger L et al (2022) Stress-driven potentiation of lateral hypothalamic synapses onto ventral tegmental area dopamine neurons causes increased consumption of palatable food. Nat Commun 13:6898
Aston SA, Caffo BS, Bhasin H, Moran TH, Tamashiro KL (2023) Timing matters: The contribution of running during different periods of the light/dark cycle to susceptibility to activity-based anorexia in rats. Physiol Behav 271:114349
Grace A, Onn S (1989) Morphology and electrophysiological properties of immunocytochemically identified rat dopamine neurons recorded in vitro. J Neurosci 9:3463–3481
Linders LE, Supiot LF, Du W, D’Angelo R, Adan RAH, Riga D, Meye FJ (2022) Studying Synaptic Connectivity and Strength with Optogenetics and Patch-Clamp Electrophysiology. Int J Mol Sci 23:11612
Himmerich H, Lewis YD, Conti C, Mutwalli H, Karwautz A, Sjögren JM et al (2023) World Federation of Societies of Biological Psychiatry (WFSBP) guidelines update 2023 on the pharmacological treatment of eating disorders. World J Biol Psychiatry 24:643–706
Steinglass JE, Kaplan SC, Liu Y, Wang Y, Walsh BT (2014) The (lack of) effect of alprazolam on eating behavior in anorexia nervosa: A preliminary report. Int J Eat Disord 47:901–904
Spadini S, Ferro M, Lamanna J, Malgaroli A (2021) Activity-based anorexia animal model: a review of the main neurobiological findings. J Eat Disord 9:123
Schoukroun F, Befort K, Bourdy R (2024) The rostromedial tegmental nucleus gates fat overconsumption through ventral tegmental area output in male rats. Neuropsychopharmacology. https://doi.org/10.1038/s41386-024-01855-w
Skowron K, Jasiński K, Kurnik-Łucka M, Stach P, Kalita K, Węglarz WP, Gil K (2020) Hypothalamic and brain stem neurochemical profile in anorectic rats after peripheral administration of kisspeptin‐10 using ¹ H‐nmr spectroscopy in vivo. NMR Biomed 33:e4306
Wable GS, Barbarich-Marsteller NC, Chowdhury TG, Sabaliauskas NA, Farb CR, Aoki C (2014) Excitatory synapses on dendritic shafts of the caudal basal amygdala exhibit elevated levels of GABA _A receptor α4 subunits following the induction of activity-based anorexia: α4-Containing Gaba _a rs in Amygdala and Anorexia. Synapse 68:1–15
Beeler JA, Frazier CRM, Zhuang X (2012) Putting desire on a budget: dopamine and energy expenditure, reconciling reward and resources. Front Integr Neurosci 6. https://doi.org/10.3389/fnint.2012.00049
Cai X, Liu H, Feng B, Yu M, He Y, Liu H et al (2022) A D2 to D1 shift in dopaminergic inputs to midbrain 5-HT neurons causes anorexia in mice. Nat Neurosci 25:646–658
Hattori S, Hashitani T, Matsui N, Nishino H (1996) Dynamic regulation of striatal dopaminergic grafts during locomotor activity. Brain Res 710:45–55
Greenwood BN (2019) The role of dopamine in overcoming aversion with exercise. Brain Res 1713:102–108
Foldi CJ, Milton LK, Oldfield BJ (2017) The Role of Mesolimbic Reward Neurocircuitry in Prevention and Rescue of the Activity-Based Anorexia (ABA) Phenotype in Rats. Neuropsychopharmacology 42:2292–2300
Chowdhury TG, Chen Y-W, Aoki C (2015) : Using the Activity-based Anorexia Rodent Model to Study the Neurobiological Basis of Anorexia Nervosa. J Vis Exp 52927
Casper RC (2006) The ‘drive for activity’ and restlessness in anorexia nervosa: Potential pathways. J Affect Disord 92:99–107
Hurel I, Redon B, Scocard A, Malezieux M, Marsicano G, Chaouloff F (2019) Beyond the Activity-Based Anorexia Model: Reinforcing Values of Exercise and Feeding Examined in Stressed Adolescent Male and Female Mice. Front Pharmacol 10:587
Koob GF, Volkow ND (2010) Neurocircuitry of Addiction. Neuropsychopharmacology 35:217–238
Scheurink AJW, Boersma GJ, Nergårdh R, Södersten P (2010) Neurobiology of hyperactivity and reward: Agreeable restlessness in Anorexia Nervosa. Physiol Behav 100:490–495

There is NO Competing Interest.

DucrocqetalSupplementaryInformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Reduced GABA transmission onto ventral tegmental area dopamine neurons underlies vulnerability to a mouse model of Anorexia Nervosa

Status:

Version 1

Abstract

Figures

INTRODUCTION

METHOD DETAILS

Experimental model and subject details

CNO administration

Stereotaxic Surgeries for Viral Injections

Slice Preparation, Cell-attached and Whole-Cell Patch Clamp Recording

Quantification and statistical analysis

RESULTS

Mice exposed to the ABA model lose weight and develop excessive running wheel activity

ABA exposure leads to increase VTA_DA neuronal activity

Reduced inhibitory transmission from local VTA GABA neurons to VTA_DA neurons in ABA-exposed animals

DISCUSSION

Declarations

COMPETING INTERESTS

AUTHOR CONTRIBUTIONS

ACKNOWLEDGEMENTS

References

Additional Declarations

Supplementary Files

Status:

Version 1

Reduced GABA transmission onto ventral tegmental area dopamine neurons underlies vulnerability to a mouse model of Anorexia Nervosa

Status:

Version 1

Abstract

Figures

INTRODUCTION

METHOD DETAILS

Experimental model and subject details

CNO administration

Stereotaxic Surgeries for Viral Injections

Slice Preparation, Cell-attached and Whole-Cell Patch Clamp Recording

Quantification and statistical analysis

RESULTS

Mice exposed to the ABA model lose weight and develop excessive running wheel activity

ABA exposure leads to increase VTADA neuronal activity

Reduced inhibitory transmission from local VTA GABA neurons to VTADA neurons in ABA-exposed animals

DISCUSSION

Declarations

COMPETING INTERESTS

AUTHOR CONTRIBUTIONS

ACKNOWLEDGEMENTS

References

Additional Declarations

Supplementary Files

Status:

Version 1

ABA exposure leads to increase VTA_DA neuronal activity

Reduced inhibitory transmission from local VTA GABA neurons to VTA_DA neurons in ABA-exposed animals