Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Disinhibition of the orbitofrontal cortex biases decision-making in obesity

Nature Neuroscience volume 26, pages 92–106 (2023)Cite this article

Subjects

Abstract

The lateral orbitofrontal cortex (lOFC) receives sensory information about food and integrates these signals with expected outcomes to guide future actions, and thus may play a key role in a distributed network of neural circuits that regulate feeding behavior. Here, we reveal a new role for the lOFC in the cognitive control of behavior in obesity. Food-seeking behavior is biased in obesity such that in male obese mice, behaviors are less flexible to changes in the perceived value of the outcome. Obesity is associated with reduced lOFC inhibitory drive and chemogenetic reduction in GABAergic neurotransmission in the lOFC induces obesity-like impairments in goal-directed behavior. Conversely, pharmacological or optogenetic restoration of inhibitory neurotransmission in the lOFC of obese mice reinstates flexible behavior. Our results indicate that obesity-induced disinhibition of the lOFC leads to a failure to update changes in the value of food with satiety, which in turn may influence how individuals make decisions in an obesogenic environment.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Diet-induced obesity induces deficits in satiety-induced devaluation.
Fig. 2: Diet-induced obesity induces deficits in CTA.
Fig. 3: Obesity reduces inhibitory tone in the lOFC and disinhibits principal output neurons.
Fig. 4: Disinhibition of the lOFC impairs devaluation.
Fig. 5: A GAT-1 transporter inhibitor reverses diet-induced impairment in inhibitory tone of obese mice.
Fig. 6: Increasing GABAergic tone in the lOFC of obese mice rescues goal-directed behavior.
Fig. 7: Increasing GABAergic drive via optogenetic stimulation of GABAergic interneurons restores GABAergic inhibition of lOFC pyramidal neurons.
Fig. 8: Optogenetically restoring inhibitory drive in the lOFC of obese mice rescues goal-directed behavior.

Similar content being viewed by others

Data Availability

Source data are provided with this paper and are also available on Figshare at https://doi.org/10.6084/m9.figshare.21330999.

References

  1. Apovian, C. M. Obesity: definition, comorbidities, causes, and burden. Am. J. Manag Care 22, s176–s185 (2016).

    Google Scholar 

  2. Obesity and Overweight (World Health Organization, 2021); https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight

  3. Swinburn, B., Sacks, G. & Ravussin, E. Increased food energy supply is more than sufficient to explain the US epidemic of obesity. Am. J. Clin. Nutr. 90, 1453–1456 (2009).

    CAS  Google Scholar 

  4. Cornell, C. E., Rodin, J. & Weingarten, H. Stimulus-induced eating when satiated. Physiol. Behav. 45, 695–704 (1989).

    CAS  Google Scholar 

  5. Krebs, J. R. The gourmet ape: evolution and human food preferences. Am. J. Clin. Nutr. 90, 707S–711S (2009).

    CAS  Google Scholar 

  6. Gortmaker, S. L. et al. Changing the future of obesity: science, policy and action. Lancet 378, 838–847 (2011).

    Google Scholar 

  7. Cohen, D. A. Neurophysiological pathways to obesity: below awareness and beyond individual control. Diabetes 57, 1768–1773 (2008).

    CAS  Google Scholar 

  8. Horstmann, A. et al. Slave to habit? Obesity is associated with decreased behavioural sensitivity to reward devaluation. Appetite 87, 175–183 (2015).

    Google Scholar 

  9. Reichelt, A. C., Morris, M. J. & Westbrook, R. F. Cafeteria diet impairs expression of sensory-specific satiety and stimulus-outcome learning. Front Psychol. 5, 852 (2014).

    Google Scholar 

  10. Ongür, D. & Price, J. L. The organization of networks within the orbital and medial prefrontal cortex of rats, monkeys and humans. Cereb. Cortex 10, 206–219 (2000).

    Google Scholar 

  11. Gremel, C. M. & Costa, R. M. Orbitofrontal and striatal circuits dynamically encode the shift between goal-directed and habitual actions. Nat. Commun. 4, 2264 (2013).

    Google Scholar 

  12. Gremel, C. M. et al. Endocannabinoid modulation of orbitostriatal circuits gates habit formation. Neuron 90, 1312–1324 (2016).

    CAS  Google Scholar 

  13. Lu, M. T., Preston, J. B. & Strick, P. L. Interconnections between the prefrontal cortex and the premotor areas in the frontal lobe. J. Comp. Neurol. 341, 375–392 (1994).

    CAS  Google Scholar 

  14. Price, J. L. Definition of the orbital cortex in relation to specific connections with limbic and visceral structures and other cortical regions. Ann. N. Y. Acad. Sci. 1121, 54–71 (2007).

    Google Scholar 

  15. Lichtenberg, N. T. et al. Basolateral amygdala to orbitofrontal cortex projections enable cue-triggered reward expectations. J. Neurosci. 37, 8374–8384 (2017).

    CAS  Google Scholar 

  16. Ramírez-Lugo, L., Peñas-Rincón, A., Ángeles-Durán, S. & Sotres-Bayon, F. Choice behavior guided by learned, but not innate, taste aversion recruits the orbitofrontal cortex. J. Neurosci. 36, 10574–10583 (2016).

    Google Scholar 

  17. Wright, S. M. & Aronne, L. J. Causes of obesity. Abdom. Imaging 37, 730–732 (2012).

    Google Scholar 

  18. Vainik, U., García-García, I. & Dagher, A. Uncontrolled eating: a unifying heritable trait linked with obesity, overeating, personality and the brain. Eur. J. Neurosci. https://doi.org/10.1111/ejn.14352 (2019).

  19. Rossi, M. A. & Yin, H. H. Methods for studying habitual behavior in mice. Curr. Protoc. Neurosci., 60, 8291–8299 (2012).

  20. Swanson, A. M., Allen, A. G., Shapiro, L. P. & Gourley, S. L. GABAAα1-mediated plasticity in the orbitofrontal cortex regulates context-dependent action selection. Neuropsychopharmacology 40, 1027–1036 (2015).

    CAS  Google Scholar 

  21. Lau, B. K. et al. Obesity-induced astrocyte dysfunction impairs heterosynaptic plasticity in the orbitofrontal cortex. Cell Rep. 36, 109563 (2021).

    CAS  Google Scholar 

  22. Thompson, J. L. et al. Obesity-induced structural and neuronal plasticity in the lateral orbitofrontal cortex. Neuropsychopharmacology 42, 1480–1490 (2017).

    CAS  Google Scholar 

  23. Oginsky, M. F. & Ferrario, C. R. Eating ‘junk food’ has opposite effects on intrinsic excitability of nucleus accumbens core neurons in obesity-susceptible versus -resistant rats. J. Neurophysiol. 122, 1264–1273 (2019).

    CAS  Google Scholar 

  24. Ohliger-Frerking, P., Horwitz, B. A. & Horowitz, J. M. Serotonergic dorsal raphe neurons from obese zucker rats are hyperexcitable. Neuroscience 120, 627–634 (2003).

    CAS  Google Scholar 

  25. Quirk, M. C., Sosulski, D. L., Feierstein, C. E., Uchida, N. & Mainen, Z. F. A defined network of fast-spiking interneurons in orbitofrontal cortex: responses to behavioral contingencies and ketamine administration. Front Syst. Neurosci. 3, 13 (2009).

    Google Scholar 

  26. Glykys, J., Mann, E. O. & Mody, I. Which GABA(A) receptor subunits are necessary for tonic inhibition in the hippocampus? J. Neurosci. 28, 1421–1426 (2008).

    CAS  Google Scholar 

  27. Bonin, R. P., Martin, L. J., MacDonald, J. F. & Orser, B. A. Alpha5GABAA receptors regulate the intrinsic excitability of mouse hippocampal pyramidal neurons. J. Neurophysiol. 98, 2244–2254 (2007).

    CAS  Google Scholar 

  28. Pavlov, I., Savtchenko, L. P., Kullmann, D. M., Semyanov, A. & Walker, M. C. Outwardly rectifying tonically active GABAA receptors in pyramidal cells modulate neuronal offset, not gain. J. Neurosci. 29, 15341–15350 (2009).

    CAS  Google Scholar 

  29. Bright, D. & Smart, T. Methods for recording and measuring tonic GABAA receptor-mediated inhibition. Frontiers in Neural Circuits 7, 193 (2013).

    CAS  Google Scholar 

  30. Drasbek, K. R. & Jensen, K. THIP, a hypnotic and antinociceptive drug, enhances an extrasynaptic GABAA receptor-mediated conductance in mouse neocortex. Cereb. Cortex 16, 1134–1141 (2006).

    Google Scholar 

  31. Anastasiades, P. G., Marlin, J. J. & Carter, A. G. Cell-type specificity of callosally evoked excitation and feedforward inhibition in the prefrontal cortex. Cell Rep. 22, 679–692 (2018).

    CAS  Google Scholar 

  32. Lee, V. & Maguire, J. The impact of tonic GABAA receptor-mediated inhibition on neuronal excitability varies across brain region and cell type. Front Neural Circuits 8, 3 (2014).

    Google Scholar 

  33. Miyamae, T., Chen, K., Lewis, D. A. & Gonzalez-Burgos, G. Distinct physiological maturation of parvalbumin-positive neuron subtypes in mouse prefrontal cortex. J. Neurosci. 37, 4883–4902 (2017).

    CAS  Google Scholar 

  34. Chen, Y.-W., Fiscella, K. A., Bacharach, S. Z. & Calu, D. J. Effect of cafeteria diet history on cue-, pellet-priming-, and stress-induced reinstatement of food seeking in female rats. PLoS ONE 9, e102213 (2014).

    Google Scholar 

  35. Tantot, F. et al. The effect of high-fat diet consumption on appetitive instrumental behavior in rats. Appetite 108, 203–211 (2017).

    Google Scholar 

  36. Ducrocq, F. et al. Decrease in operant responding under obesogenic diet exposure is not related to deficits in incentive or hedonic processes. Obes. (Silver Spring) 27, 255–263 (2019).

    CAS  Google Scholar 

  37. Friend, D. M. et al. Basal ganglia dysfunction contributes to physical inactivity in obesity. Cell Metab. 25, 312–321 (2017).

    CAS  Google Scholar 

  38. Beeler, J. A., Faust, R. P., Turkson, S., Ye, H. & Zhuang, X. Low Dopamine D2 receptor increases vulnerability to obesity via reduced physical activity not increased appetitive motivation. Biol. Psychiatry https://doi.org/10.1016/j.biopsych.2015.07.009 (2015).

  39. O’Neal, T. J., Friend, D. M., Guo, J., Hall, K. D. & Kravitz, A. V. Increases in physical activity result in diminishing increments in daily energy expenditure in mice. Curr. Biol. 27, 423–430 (2017).

    Google Scholar 

  40. Sun, H., Yan, J., Sun, B., Song, L. & Yan, J. Taste sensitivity to sucrose is lower in outbred Sprague-Dawley phenotypic obesity-prone rats than obesity-resistant rats. Biochem. Biophys. Res. Commun. 489, 155–163 (2017).

    CAS  Google Scholar 

  41. Ostlund, S. B. & Balleine, B. W. Orbitofrontal cortex mediates outcome encoding in Pavlovian but not instrumental conditioning. J. Neurosci. 27, 4819–4825 (2007).

    CAS  Google Scholar 

  42. Panayi, M. C. & Killcross, S. Functional heterogeneity within the rodent lateral orbitofrontal cortex dissociates outcome devaluation and reversal learning deficits. eLife 7, e37357 (2018).

    Google Scholar 

  43. Parkes, S. L. et al. Insular and ventrolateral orbitofrontal cortices differentially contribute to goal-directed behavior in rodents. Cereb. Cortex 28, 2313–2325 (2018).

    Google Scholar 

  44. Bissonette, G. B., Schoenbaum, G., Roesch, M. R. & Powell, E. M. Interneurons are necessary for coordinated activity during reversal learning in orbitofrontal cortex. Biol. Psychiatry 77, 454–464 (2015).

    Google Scholar 

  45. Dingess, P. M. et al. Consumption of a high-fat diet alters perineuronal nets in the prefrontal cortex. Neural Plast. 2018, 2108373 (2018).

    CAS  Google Scholar 

  46. Reichelt, A. C., Hare, D. J., Bussey, T. J. & Saksida, L. M. Perineuronal nets: plasticity, protection, and therapeutic potential. Trends Neurosci. 42, 458–470 (2019).

    CAS  Google Scholar 

  47. Furlong, T. M., Jayaweera, H. K., Balleine, B. W. & Corbit, L. H. Binge-like consumption of a palatable food accelerates habitual control of behavior and is dependent on activation of the dorsolateral striatum. J. Neurosci. 34, 5012–5022 (2014).

    CAS  Google Scholar 

  48. Alonso-Caraballo, Y. et al. Sex specific effects of ‘junk-food’ diet on calcium permeable AMPA receptors and silent synapses in the nucleus accumbens core. Neuropsychopharmacology 46, 569–578 (2021).

    CAS  Google Scholar 

  49. Richardson, N. R. & Roberts, D. C. Progressive ratio schedules in drug self-administration studies in rats: a method to evaluate reinforcing efficacy. J. Neurosci. Methods 66, 1–11 (1996).

    CAS  Google Scholar 

  50. Herman, M. A. & Roberto, M. Cell-type-specific tonic GABA signaling in the rat central amygdala is selectively altered by acute and chronic ethanol. Addiction Biol. 21, 72–86 (2016).

    CAS  Google Scholar 

Download references

Acknowledgements

We would like to acknowledge the Hotchkiss Brain Institute optogenetic core facility and the advanced microscopy facility for their technical support. 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, Piikuni, Kainai First Nations), the Tsuut’ina and the Stoney Nakoda (including the Chiniki, Bearspaw and Wesley First Nations). The City of Calgary is also home to Metis Nation of Alberta, Region III Funding; this research was supported by a Koopmans Research Award, Mathison Centre for Research and Education Neural Circuits research grant, Canadian Institutes of Health Research operating grant (CIHR, FDN-147473) and a Canada Research Chair Tier 1 (950-232211) to S.L.B. L.N. was supported by postdoctoral awards from Les Fonds de la Recherche en Sante du Quebec, Alberta Innovates Health Solutions and CIHR. L.T.S. was supported by a Harley Hotchkiss Doctoral Scholarship in Neuroscience and Alberta Graduate Excellence Doctoral Scholarship.

Author information

Author notes

  1. These authors contributed equally: Lauren T. Seabrook, Lindsay Naef

Authors and Affiliations

  1. Department of Physiology and Pharmacology, Hotchkiss Brain Institute, The University of Calgary, Calgary, Alberta, Canada

    Lauren T. Seabrook, Lindsay Naef, Corey Baimel, Allap K. Judge, Tyra Kenney, Madelyn Ellis, Temoor Tayyab, Mataea Armstrong, Min Qiao & Stephanie L. Borgland

  2. Department of Psychology, The University of British Columbia, Vancouver, British Columbia, Canada

    Stan B. Floresco

Authors
  1. Lauren T. Seabrook
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lindsay Naef
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Corey Baimel
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Allap K. Judge
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Tyra Kenney
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Madelyn Ellis
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Temoor Tayyab
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Mataea Armstrong
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Min Qiao
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Stan B. Floresco
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Stephanie L. Borgland
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

L.T.S. and C.B. performed electrophysiological experiments. L.T.S., L.N., A.K.J., T.K., M. E. and T. T. performed behavioral experiments. L.T.S. and M.Q. performed immunohistochemistry experiments. L.T.S, L.N., C.B., M.A. and S.L.B. analyzed the data. S.B.F. provided the Med Associates code. L.T.S., L.N. and S.L.B. wrote the manuscript and all authors edited it.

Corresponding author

Correspondence to Stephanie L. Borgland.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Neuroscience thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Body weights and glucose tolerance in lean and obese food restricted or ad libitum fed mice.

a) Mice fed a low-fat diet on ad libitum (lean ad libitum n = 8), or food restricted (lean food restricted n = 9) feeding schedules weighed less than mice fed a high-fat diet on ad libitum (obese ad libitum n = 9) and food restricted (obese food restricted n = 8). Two-way RM ANOVA: Feeding schedule effect: F (1, 29) = 1.682, p = 0.2048, Diet effect: F (1, 29) = 92.08, P < 0.0001****, Diet and feeding schedule interaction: F (1, 29) = 2.771, p = 0.1068. We had an a priori hypothesis that mice fed a high-fat diet would gain more body weight. Sidak’s multiple comparisons comparing food restricted in lean and obese mice P < 0.0001**** and comparing lean and obese ad libitum p < 0.0001****. Data are presented as mean values + /− SEM. b) Mice fed a high-fat diet (obese ad libitum n = 8 and obese food restricted n = 8) had reduced insulin sensitivity than mice fed a low-fat diet (lean ad libitum n = 8 and lean food restricted n = 9) as indicated by blood glucose (nM/L) concentrations following an intraperitoneal injection of 20% D-glucose solution. Three-way RM ANOVA mixed effects model: time × diet × feeding schedule interaction, F (11, 153) = 1.962, p = 0.0359*, diet × feeding schedule interaction: F (1, 153) = 0.4945, p = 0.4830. time × feeding schedule interaction: F (11, 153) = 2.935, p = 0.4830, time × diet interaction F (11, 153) = 10.72, p < 0.0001****, feeding schedule effect, F (1, 15) = 1.910, p = 0.1872, diet effect, F (1, 153) = 27.63, p < 0.0001****, time effect F (11, 165) = 90.76, p < 0.0001****. Data are presented as mean values + /− SEM. c) Mice fed a high-fat diet (obese ad libitum n = 8 and obese food restricted n = 8) had reduced baseline insulin sensitivity than mice fed a low-fat diet (lean ad libitum n = 8 and lean food restricted n = 9) indicated at time 0 before 20% glucose injection. Two-way RM ANOVA: Schedule effect: F (1, 29) = 0.1033, p = 0.7503, Diet effect F (1, 29) = 10.73, p = 0.0027**, Schedule and diet interaction: F (1, 29) = 0.05216, p = 0.8210. Data are presented as mean values + /− SEM. d) Mice fed a high-fat diet (obese ad libitum n = 8 and obese food restricted n = 8) have reduced insulin sensitivity than mice fed a low-fat diet (lean ad libitum n = 8 and lean food restricted n = 9) indicated by area under the curve. Two-way RM ANOVA: Schedule effect: F (1, 29) = 2.152, p = 0.1531, Diet effect: F (1, 29) = 27.62, p = <0.0001****, Schedule and diet interaction: F (1, 29) = 0.5155, p = 0.4785. We had an a priori hypothesis that mice fed a high-fat diet would have prolonged blood glucose levels. Sidak’s multiple comparison test comparing lean and obese mice with food restricted p = 0.0058** and ad libitum p = 0.0005*** feeding schedules. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 2 Satiety-induced devaluation and context generalization.

a) Lean (n = 60) and obese (n = 60) mice were trained to lever press (RR5, RR10, and RR20 training) for sucrose prior to behavioral testing on the devaluation task (included cohorts from selective satiety, context generalization and flavor preconditioning experiments). Two-way RM ANOVA: diet effect F (1, 118) = 17.04, P < 0.0001****, Schedule of reinforcement effect (RR5 vs. RR10 vs. RR20): F (1.310, 154.6) = 18.26, P < 0.0001****, Diet × schedule of reinforcement interaction: F (2, 236) = 7.112, p = 0.0010. A Dunnett’s multiple comparison test compared RR10 and RR20 schedule of reinforcement to RR5 was performed to see if mice were escalating their lever presses from RR5. Lean RR5 vs RR10 P < 0.0001****, Lean RR5 vs RR20 p = 0.0006***, Obese RR5 vs RR10 p = 0.0073**, Obese RR5 vs RR20 p = 0.0046**. There is no significant difference in slopes. Simple linear regression: F = 3.678. DFn = 1, DFd = 356, p = 0.0559, but there was a significant difference in the slope intercepts F = 40.26. DFn = 1, DFd = 357, P < 0.0001****. b) Satiety-induced devaluation procedure. c) In the devaluation by selective satiety task, mice with long term exposure to a 60% high-fat (obese n = 31) diet gained more weight than mice with long term exposure to a 10% low-fat (lean n = 35) diet. Unpaired T-test: t (64) =10.88, p < 0.0001****. Data are presented as mean values + /− SEM. d) There was no difference in the consumption of sucrose (ml) of lean (n = 35) and obese (n = 31) during the 3-hour pre-feeding period prior to devaluation (devalued condition). Unpaired t-test: t (64) =0.6408, p = 0.5239. Data are presented as mean values + /− SEM. e) Lean (n = 35) mice devalued the outcome as indicated by decreased lever presses in the devalued compared to the valued condition. Obese mice (n = 31) were not sensitive to devaluation as indicated by comparable lever presses in the valued and devalued conditions. RM two-way ANOVA: devaluation effect, F (1,64) =7.8, p = 0.0067**, diet group × devaluation interaction, F (1,64) =5.8, p = 0.0187*, Sidak’s multiple comparisons test, lean: p = 0.0006***, obese: p = 0.9561. Data are presented as mean values + /− SEM. f) Obesity decreased the revaluation index ((lever presses valued – lever presses devalued)/ (lever presses valued + lever presses devalued)). Unpaired t-test: lean (n = 35), obese (n = 31), t (64) =4.109, p = 0.0001***. Data are presented as mean values + /− SEM. g) Total number of lever presses (valued+devalued) of obese mice (n = 31) compared to lean mice (n = 35) during the devaluation task. Unpaired t-test: t (64) =1.886, p = 0.0638. Data are presented as mean values + /− SEM. h) In the context generalization satiety-induced devaluation task, mice with long term exposure to a 60% high-fat (obese n = 16) diet gained more weight than mice with long term exposure to a 10% low-fat (lean n = 12) diet. Unpaired T-test: t (26) =10.18, p < 0.0001****. Data are presented as mean values + /− SEM. i) There was no difference in the consumption of sucrose (ml) of lean (n = 12) and obese (n = 16) mice during the pre-feed. Unpaired T-test t (26) =0.3258, p = 0.7472. Data are presented as mean values + /− SEM. j) In the sensory-specific satiety devaluation task, mice with long term exposure to a 60% high-fat (obese n = 13) diet gained more weight than mice with long term exposure to a 10% low-fat (lean n = 12) diet. Unpaired T-test: t (23) =6.068, p < 0.0001****.

Source data

Extended Data Fig. 3 Sucrose reinforced devaluation and progressive ratio.

a) Devaluation procedure. b) When sucrose is delivered during the devaluation test, lean mice (n = 8) devalued the sucrose reward (reduced lever presses in the devalued condition), while obese mice (n = 13) did not (similar number of lever presses during valued and devalued condition). Two-way RM ANOVA: devaluation effect: F (1, 19) = 5.283, p = 0.0331, diet effect: F (1, 19) = 0.9864, p = 0.3331, diet × devaluation interaction: F (1, 19) = 2.541, p = 0.1274. A priori hypothesis was that sated lean mice would devalue the sucrose reward. Sidak’s multiple comparisons test: lean valued vs. devalued p = 0.0454*, obese valued vs. devalued p = 0.8193. Data are presented as mean values + /− SEM. c) Revaluation index of lean (n = 8) and obese (n = 13) mice in the devaluation task. Unpaired t-test: t (19) =2.481, p = 0.0226*. Data are presented as mean values + /− SEM. d) Total lever presses (valued + devalued) were not altered by diet (lean (n = 8) vs. obese (n = 13)). Unpaired t-test t (19) =0.9932, p = 0.3331. Data are presented as mean values + /− SEM. e) Weights of lean (n = 15) and obese (n = 11) mice in the progressive ratio test. Unpaired T-test: t (24) =13.25, p < 0.0001****. Data are presented as mean values + /− SEM. f) Number of lever presses of lean (n = 15) and obese (n = 11) mice during progressive ratio. Unpaired t-test: t (24) =1.521, p = 0.1412. Data are presented as mean values + /− SEM. g) Number of sucrose reinforcers received by lean (n = 15) and obese (n = 11) mice during progressive ratio. Unpaired t-test: t (24) =1.282, p = 0.2122. Data are presented as mean values + /− SEM. h) Breakpoint obtained by lean (n = 15) and obese (n = 11) mice during progressive ratio. Unpaired t-test: t (24) =1.234, p = 0.2293. Data are presented as mean values + /− SEM. i) Lean (n = 15) but not obese mice (n = 16) learn the contingency change when rewards given regardless of lever presses. Two-way ANOVA: Diet effect F (1, 29) = 1.575, p = 0.2195, Contingency effect: F (1, 29) = 19.97, p = 0.0001 ***, Diet × contingency interaction F (1, 29) = 1.734, p = 0.1982. We had an a priori hypothesis that lean mice would devalue and obese mice would not. Sidak’s multiple comparison comparing lean non degraded (ND) vs contingency degraded (CD) day p = 0.0007 ***, obese non degraded and contingency changed day p = 0.0612. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 4 Obese mice have continued impairment of devaluation 7d after diet removal.

a) Weights of mice before and after 7 days removal of diet following exposure to either the low-fat (n = 9) or high-fat diet (n = 8). RM Two-way ANOVA: Diet effect F (1, 15) = 37.03, p < 0.0001****, Time effect F (1, 15) = 13.08, p = 0.0025**, Diet × time interaction F (1, 15) = 4.772, p = 0.0452. Sidak’s multiple comparisons test: lean: pre and post diet removal p < 0.0001****, obese: pre and post diet removal. p < 0.0001****. Data are presented as mean values + /− SEM. b) Lean (n = 9) and obese (n = 8) mice were trained to lever press (RR5, RR10, and RR20 training) for sucrose prior to behavioral testing on the devaluation task. Two-way RM ANOVA: Diet effect F (1, 15) = 16.28, p = 0.0011**, Training effect F (2, 30) = 2.892, p = 0.0710, Diet × training interaction F (2, 30) = 0.3876, p = 0.6820. Simple linear regression: Slopes are equal F (1,47) = 0.5650, p = 0.4560 and Y intercepts are different F (1,48) = 26.70, P < 0.0001 ****. Data are presented as mean values + /− SEM. c) There was a difference in the consumption of sucrose (ml) of lean (n = 9) and obese (n = 8) during the 3-hour pre-feeding period prior to devaluation (devalued condition). Unpaired t-test: t(1f) =2.341, p = 0.0335*. Data are presented as mean values + /− SEM. d) Lever presses during devaluation of lean (n = 9) and obese mice (n = 8). RM two-way ANOVA: devaluation effect, F (1, 15) = 4.178, p = 0.0589, diet effect: F (1, 15) = 5.038, p = 0.0403 *, diet group × devaluation interaction, F (1, 15) = 11.97, p = 0.0035. Sidaks multiple comparisons lean valued vs devalued p = 0.0023, obese valued vs devalued p = 0.5726. Data are presented as mean values + /− SEM. e) Obesity decreased the revaluation index ((lever presses valued – lever presses devalued)/ (lever presses valued + lever presses devalued)). Unpaired t-test: lean (n = 9), obese (n = 8), t (15) =2.478, p = 0.0256*. Data are presented as mean values + /− SEM. f) Total number of lever presses (valued + devalued) of obese mice (n = 8) compared to lean mice (n = 9) during the devaluation task. Unpaired t-test: t (15) =2.245, p = 0.0403*. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 5 Pyramidal neuron excitability in lean and obese mice on ad libitum or food restricted feeding schedules.

a) Representative recordings of action potentials observed at 100pA, 300pA and 500pA current steps from lOFC pyramidal neurons of lean and obese mice on ad libitum or food restricted feeding schedules. b) Obese mice fed high-fat diet on ad libitum (n = 3) and food restricted (n = 3) weight more than lean mice on ad libitum (n = 3) and food restricted (n = 3) feeding schedule. RM Two-way ANOVA: Diet effect F (1, 8) = 34.16, p = 0.0004***, food schedule effect F (1, 8) = 12.93, p = 0.0070**, Diet × food schedule interaction F (1, 8) = 0.2985, p = 0.5997. We had an a priori that obese mice would weigh more than lean mice regardless of food schedule. Sidak’s multiple comparisons test: ad libitum lean vs obese p = 0.0039**, food restricted lean vs obese 0.0113*. Data are presented as mean values + /− SEM. c) Diet-induced obesity increased the excitability of lOFC pyramidal neurons as indicated by frequency-current (F–I) plots of number of action potentials at current injections from 0pA to 500pA of lean food restricted (n = 14 cells/3 mice), lean ad libitum (12 cells/3 mice) and obese food restricted (n = 14 cells/3 mice) and obese ad libitum (n = 12 cells/3 mice). RM 3-way ANOVA: current step effect F (20, 520) = 509.4, p < 0.0001****, diet effect: F (1, 436) = 12.28, p = 0.0005***, food schedule effect: F (1, 26) = 4.450, p = 0.0447*, current × diet interaction: F (20, 436) = 13.45, p < 0.0001****, current × food schedule interaction: F (20, 436) = 1.792, p = 0.0193*, diet × food schedule interaction: F (1, 436) = 0.1139, p = 0.7359, current × diet × food schedule interaction F (20, 436) = 0.9759, p = 0.4906. Data are presented as mean values + /− SEM. d) Diet-induced obesity (obese food restricted n = 14 cells/3mice, obese ad libitum n = 12 cells/3mice) increases the firing slope compared to lean mice (lean food restricted n = 14 cells/3mice, lean ad libitum n = 12 cells/3 mice) in both food restricted and ad libitum feeding schedules. Mean excitability slope (slope of linear regression from individual cells, x=current step y=number of action potentials) of lOFC pyramidal neurons. Two-way RM ANOVA: Diet effect: F (1, 48) = 20.85, P < 0.0001, food schedule effect: F (1, 48) = 0.9912, p = 0.3244, diet × food schedule interaction: F (1, 48) = 1.106, p = 0.2983. Based on previous excitability data we had an a priori hypothesis that obese mice would have hyperexcitable pyramidal neurons. Sidaks multiple comparisons ad libitum lean vs obese p = 0.0007***, food restricted lean vs obese p = 0.0254*. Data are presented as mean values + /− SEM. e) The resting membrane potential in lean (lean food restricted n = 14 cells/3mice, lean ad libitum n = 12 cells/3 mice) and obese mice (obese food restricted n = 14 cells/3 mice, obese ad libitum n = 12 cells/3 mice) was not altered based on food schedule. Two-way RM ANOVA: Diet effect: F (1, 48) = 6.144, p = 0.0168*, Feeding schedule effect: F (1, 48) = 0.04941, p = 0.8250, Diet × feeding schedule interaction: F (1, 48) = 0.2927, p = 0.5910. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 6 Action potential characteristics of lOFC neurons from lean and obese mice.

a) Capacitance is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/3 mice) mice. Two-way ANOVA: Diet effect: F (1, 34) = 0.3277, p = 0.5708, Picrotoxin effect: F (1, 34) = 3.141, p = 0.0853, Diet × picrotoxin interaction: F (1, 34) = 4.573, p = 0.0397*. Data are presented as mean values + /− SEM. b) Resting membrane potential is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8cells/3mice) and obese PTX (n = 9cells/5mice) mice. Two-way ANOVA: Diet effect F (1, 34) = 0.6887, p = 0.4124, Picrotoxin effect: F (1, 34) = 0.007646, p = 0.9308, Diet × picrotoxin interaction F (1, 34) = 0.7791, p = 0.3836. Data are presented as mean values + /− SEM. c) Input resistance is increased in obese (n = 8 cells/3 mice) compared to lean mice (n = 10 cells/3 mice), but no difference in PTX (lean PTX (n = 11 cells/3 mice), obese PTX (n = 9 cells/5 mice)). Two-way ANOVA: Diet effect: F (1, 34) = 5.360, p = 0.0268*, Picrotoxin effect: F (1, 34) = 2.079, p = 0.1585, Diet × picrotoxin interaction: F (1, 34) = 5.340, p = 0.0270 *. Sidak’s multiple comparisons: No PTX lean vs obese mice, p = 0.0062**, PTX lean vs obese, P > 0.9999. Data are presented as mean values + /− SEM. d) Rheobase is decreased in obese (n = 8 cells/3 mice) compared to lean mice (n = 10 cells/3 mice), but no difference in PTX (lean PTX (n = 11 cells/3 mice), obese PTX (n = 9c ells/5 mice)). Two-way ANOVA: Diet effect F (1, 34) = 13.82, p = 0.0007***, Picrotoxin effect: F (1, 34) = 4.043, p = 0.0523, Diet × picrotoxin interaction: F (1, 34) = 4.174, p = 0.0489*. Sidaks multiple comparisons: No PTX lean vs obese mice p = 0.0007***, PTX lean vs obese p = 0.41. Data are presented as mean values + /− SEM. e) Action potential height is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/5 mice) mice. Two-way ANOVA: Diet effect: F (1, 34) = 6.050, p = 0.0191*, Picrotoxin effect: F (1, 34) = 2.359, p = 0.1338, Diet × picrotoxin interaction: F (1, 34) = 0.6723, p = 0.4180. Data are presented as mean values + /− SEM. f) Action potential width is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/5 mice) mice. Two-way ANOVA: Diet effect: F (1, 34) = 0.02203, p = 0.8829, Picrotoxin effect: F (1, 34) = 1.201, p = 0.2807, Diet × picrotoxin interaction: F (1, 34) = 2.864, p = 0.0997. Data are presented as mean values + /− SEM. g) Threshold to fire is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/5 mice) mice. Two-way ANOVA, Diet effect: F (1, 34) = 0.04215, p = 0.8386, Picrotoxin effect: F (1, 34) = 1.252, p = 0.2711, Diet × picrotoxin interaction: F (1, 34) = 0.2539, p = 0.6176. Data are presented as mean values + /− SEM. h) After hyperpolarization height is similar in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/5 mice) mice. Two-way ANOVA: Diet effect: F (1, 34) = 0.5353, p = 0.4694, Picrotoxin effect: F (1, 34) = 0.04374, p = 0.8356, Diet × picrotoxin interaction: F (1, 34) = 1.468, p = 0.2340. Data are presented as mean values + /− SEM. i) After hyperpolarization width is different in no PTX lean (n = 10 cells/3 mice), lean PTX (n = 11 cells/3 mice) no PTX obese (n = 8 cells/3 mice) and obese PTX (n = 9 cells/5 mice) mice. Two-way ANOVA: Diet effect: F (1, 34) = 7.140, p = 0.0115 *, Picrotoxin effect: F (1, 34) = 3.777, p = 0.0603, Diet × picrotoxin interaction: F (1, 34) = 5.648, p = 0.0232 *, Sidaks multiple comparisons: No PTX lean vs obese mice p = 0.0028**, PTX lean vs obese, p = 0.9715. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 7 Inactivation of GABAergic neurons (via hM4d(Gi)) impairs satiety-induced devaluation and conditioned taste aversion.

a) Body weights of mCherry (n = 7) and hM4d(Gi) (n = 9) are similar: Unpaired t-test: t (14) =0.1395, p = 0.8910. Data are presented as mean values + /− SEM. b) Lean mCherry (n = 7) and hM4d(Gi) (n = 9) were trained to lever press on a RR5, RR10 and RR20 schedule. The learning slope is not different between lean and obese mice. Simple linear regression of slopes: F (1,44) = 0.02496, p = 0.8752. There is no difference in the intercepts: F (1,45) = 0.01782, p = 0.8944. Data are presented as mean values + /− SEM. c) Lean mCherry (n = 7) and hM4d(Gi) (n = 9) drink the same amount of sucrose in the pre consumption test. Two-way ANOVA: virus effect: F (1, 14) = 2.549, p = 0.1327, Drug effect: F (1, 14) = 3.064, p = 0.1019, Drug and diet interaction: F (1, 14) = 0.001013, p = 0.9751. Data are presented as mean values + /− SEM. d) Body weights of mCherry (n = 12) and hM4d(Gi) (n = 13) are the same: Unpaired t-test: t (23) =0.7069, p = 0.4867. Data are presented as mean values + /− SEM. e) Lean mCherry (n = 12) and hM4d(Gi) (n = 13) underwent pairing where a novel gelatine flavor was paired with LiCl inducing sickness, or vehicle. 3-way RM ANOVA: Conditioning day effect: F (1.924, 44.25) = 6.360, p = 0.0041**, virus effect: F (1, 23) = 2.344, p = 0.1394, value effect F (1.000, 23.00) = 0.3603, p = 0.5542, day × virus interaction: F (2, 46) = 0.1939, p = 0.8244, day × value interaction: F (1.890, 43.46) = 0.4546, p = 0.6267, virus × value interaction: F (1, 23) = 0.002321, p = 0.9620, day × virus × value F (2, 46) = 1.432, p = 0.2493. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 8 NNC-711 in the lOFC restores devaluation in obesity.

a) Weights of lean and obese mice following long-term exposure to either the 10% low fat (lean n = 14) or 60% high-fat (obese n = 16) diet in the devaluation by selective satiety task. Unpaired T-test: t (28) =7.818, p < 0.0001****. Data are presented as mean values + /− SEM. b) Number of lever presses performed on the last two days of RR5, RR10, RR20 training prior to behavioral testing for selective satiety devaluation of lean (n = 14) and obese (n = 16) mice. Two-way RM ANOVA: Diet effect: F (1, 28) = 24.80, P < 0.0001****, training schedule effect: F (1.441, 40.35) = 11.45, p = 0.0005***, diet × training schedule interaction: F (2, 56) = 3.103, p = 0.0527. We expected that both lean and obese mice would escalate their lever presses. Dunnett’s multiple comparison: lean RR5 vs RR10 p = 0.4182, lean RR5 vs RR20 p = 0.0171*, obese RR5 vs RR10 p = 0.0554, obese RR5 vs RR20 p = 0.0102*. Data are presented as mean values + /− SEM. c) Sucrose presconsumption of lean vehicle mice (n = 12), lean NNC mice (n = 12), and obese vehicle mice (n = 14) and obese NNC mice (n = 15). Two-way RM ANOVA mixed effects model: Interaction: F (1, 21) = 0.03347, p = 0.8566, diet effect: F (1, 28) = 0.5022, p = 0.4844, Drug effect: F (1, 21) = 0.007126, p = 0.9335. Data are presented as mean values + /− SEM. d) Total distance traveled (m) of lean (n = 7) and obese (n = 6) mice in a one-hour open field test immediately following lOFC infusions of either vehicle or NNC-711 (NNC-711 10 µM). Two-way RM ANOVA: Interaction: F (1, 11) = 0.1175, p = 0.7383, vehicle vs. NNC-711 effect: F (1, 11) = 0.7986, p = 0.3906, diet effect: F (1, 11) = 16.76, p = 0.0018*. Based on previous literature and because there was a diet effect, we expected that obese mice would travel less distance than lean mice. Sidaks multiple comparison test: vehicle lean vs obese p = 0.0172*, NNC-711 lean vs obese p = 0.0061**. Data are presented as mean values + /− SEM.

Source data

Extended Data Fig. 9 Optogenetic activation of lOFC inhibitory neurons restores devaluation in obesity.

a) Electrophysiology validation of ChR2 expressed in GABAergic neurons of the lOFC. Example traces of current evoked (pA) from a lOFC GABAChR2 neuron to a train of light pulses delivered at 5 Hz for 4 ms (5 mW). b) Electrophysiology validation of ChR2 expressed in GABAergic neurons of the lOFC. Example traces of current evoked (pA) from a lOFC GABAChR2 neuron to a single 500 ms pulse in voltage clamp. This was reliably repeated in several different brain slices from VGATcre mice. c) VGATcre mice exposed to the high-fat diet (n = 7) weighed more than lean mice (n = 8) prior to behavioral testing for devaluation by selective satiety. Unpaired t-test: t (13) =7.531, p < 0.0001****. Data are presented as mean values + /− SEM. d) Lean (n = 8) and obese (n = 7) VGATcre mice were trained to lever press (RR5, RR10, and RR20 training) for sucrose prior to behavioral testing on the devaluation task. Two-way ANOVA mixed effects model: training schedule effect: F (1.504, 19.55) = 9.122, p = 0.0031**, Diet effect: F (1, 13) = 32.23, P < 0.0001****, Schedule of reinforcement and diet interaction: F (2, 26) = 4.331, p = 0.0238 *. Dunnett’s multiple comparisons: Lean RR5 vs RR10 p = 0.9403, lean RR5 vs RR20 p = 0.0662, obese RR5 vs RR20 p = 0.0109*, obese RR5 vs RR20 p = 0.0108*. Data are presented as mean values + /− SEM. e) Lean (lean inactive 589 nm n = 8, lean active 473 nm n = 8) and obese (obese inactive 589 nm n = 6, obese active 473 nm n = 7) mice consumed similar amounts of sucrose prior to the selective satiety procedure. Two-way ANOVA mixed effects model: Diet effect F (1, 13) = 0.02776, p = 0.8702, Stimulation effect F (1, 11) = 1.468, p = 0.2510, diet × stimulation interaction F (1, 11) = 1.468, 0.2510. Data are presented as mean values + /− SEM. f) Total distance traveled (m) of lean (n = 6) and obese (n = 3) mice in a one-hour open field test immediately following photo-stimulation (589 nm inactive or 473 nm active) of lOFC GABAergic neurons. Two-way RM ANOVA: Interaction: F (1, 7) = 0.01176, p = 0.9167, Diet effect: F (1, 7) = 3.399, p = 0.1078 photostimulation effect: F (1, 7) = 0.8146, p = 0.3967. Data are presented as mean values + /− SEM.

Source data

Supplementary information

Supplementary Information

Supplementary Table 1: Statistics table.

Reporting Summary

Source data

Source Data Fig. 1

Excel spreadsheet.

Source Data Fig. 2

Excel spreadsheet.

Source Data Fig. 3

Excel spreadsheet.

Source Data Fig. 4

Excel spreadsheet.

Source Data Fig. 5

Excel spreadsheet.

Source Data Fig. 6

Excel spreadsheet.

Source Data Fig. 7

Excel spreadsheet.

Source Data Fig. 8

Excel spreadsheet.

Source Data Extended Data Fig. 1

Excel spreadsheet.

Source Data Extended Data Fig. 2

Excel spreadsheet.

Source Data Extended Data Fig. 3

Excel spreadsheet.

Source Data Extended Data Fig. 4

Excel spreadsheet.

Source Data Extended Data Fig. 5

Excel spreadsheet.

Source Data Extended Data Fig. 6

Excel spreadsheet

Source Data Extended Data Fig. 7

Excel spreadsheet.

Source Data Extended Data Fig. 8

Excel spreadsheet.

Source Data Extended Data Fig. 9

Excel spreadsheet.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Seabrook, L.T., Naef, L., Baimel, C. et al. Disinhibition of the orbitofrontal cortex biases decision-making in obesity. Nat Neurosci 26, 92–106 (2023). https://doi.org/10.1038/s41593-022-01210-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41593-022-01210-6

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing