Portal de Periódicos da CAPES

(S)Pot on Mitochondria: Cannabinoids Disrupt Cellular Respiration to Limit Neuronal Activity

2017; Cell Press; Volume: 25; Issue: 1 Linguagem: Inglês

10.1016/j.cmet.2016.12.020

1932-7420

Tibor Harkany, Tamas L. Horváth,

Mitochondrial Function and Pathology

Classical views posit G protein-coupled cannabinoid receptor 1s (CB1Rs) at the cell surface with cytosolic Giα-mediated signal transduction. Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar instead place CB1Rs at mitochondria limiting neuronal respiration by soluble adenylyl cyclase-dependent modulation of complex I activity. Thus, neuronal bioenergetics link to synaptic plasticity and, globally, learning and memory. Classical views posit G protein-coupled cannabinoid receptor 1s (CB1Rs) at the cell surface with cytosolic Giα-mediated signal transduction. Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar instead place CB1Rs at mitochondria limiting neuronal respiration by soluble adenylyl cyclase-dependent modulation of complex I activity. Thus, neuronal bioenergetics link to synaptic plasticity and, globally, learning and memory. Irrespective of a cell's degree of specification within any mammalian tissue, its O2-respiring mitochondria are the sources of bioenergy. Mitochondria originate from bacteria (Taanman, 1999Taanman J.W. Biochim. Biophys. Acta. 1999; 1410: 103-123Crossref PubMed Scopus (17) Google Scholar) with their circular DNA coding for 37 prokaryote gene homologs. However, mitochondria contain ∼1,500 proteins in mammals (Chacinska et al., 2009Chacinska A. Koehler C.M. Milenkovic D. Lithgow T. Pfanner N. Cell. 2009; 138: 628-644Abstract Full Text Full Text PDF PubMed Scopus (986) Google Scholar), which are encoded by the nuclear genome, synthesized in the cytosol, and imported through at least 4 pathways into mitochondria (Chacinska et al., 2009Chacinska A. Koehler C.M. Milenkovic D. Lithgow T. Pfanner N. Cell. 2009; 138: 628-644Abstract Full Text Full Text PDF PubMed Scopus (986) Google Scholar). An unexpected variety of steroid, neuropeptide, and neurotransmitter receptors, including those for glucocorticoids, estrogen, androgens, thyroid hormone, prostaglandins, N-methyl-D-aspartate, and acetylcholine (Lee et al., 2008Lee J. Sharma S. Kim J. Ferrante R.J. Ryu H. J. Neurosci. Res. 2008; 86: 961-971Crossref PubMed Scopus (43) Google Scholar), is present in mitochondria to modulate cellular respiration. Endocannabinoids, like prostaglandins, are arachidonic acid-derived signal lipids whose intracellular pools can serve as metabolic intermediates to satisfy the cell's metabolic demands (Nomura et al., 2011Nomura D.K. Morrison B.E. Blankman J.L. Long J.Z. Kinsey S.G. Marcondes M.C. Ward A.M. Hahn Y.K. Lichtman A.H. Conti B. Cravatt B.F. Science. 2011; 334: 809-813Crossref PubMed Scopus (527) Google Scholar). Endocannabinoids bind type 1 and 2 cannabinoid receptors (CB1R/CB2R) in many organ systems. CB1R and CB2R are members of the superfamily of GPCRs and considered as cell-surface signaling proteins. While a decade-long consensus suggested strict cell-type-specific expressional rules for CB1R (being "neuronal") and CB2R (considered "glial") in the brain, the use of sophisticated genetics shifted this view (Han et al., 2012Han J. Kesner P. Metna-Laurent M. Duan T. Xu L. Georges F. Koehl M. Abrous D.N. Mendizabal-Zubiaga J. Grandes P. et al.Cell. 2012; 148: 1039-1050Abstract Full Text Full Text PDF PubMed Scopus (334) Google Scholar), favoring receptor heterogeneity as a determinant of signal integration and outcome. Besides cell-surface GPCRs, endo- and phytocannabinoids (from Cannabis sp.) can also target intracellular receptors or binding sites: N-acylethanolamines activate nuclear peroxisome proliferator-activated receptors, and anandamide recognizes an intracellular binding site in TRPV1 channels (van der Stelt et al., 2005van der Stelt M. Trevisani M. Vellani V. De Petrocellis L. Schiano Moriello A. Campi B. McNaughton P. Geppetti P. Di Marzo V. EMBO J. 2005; 24: 3026-3037Crossref PubMed Scopus (204) Google Scholar). The exclusive cell-surface targeting of CB1R is the dogma that Giovanni Marsicano and his colleagues challenged in a series of thought-provoking studies in the recent past (Bénard et al., 2012Bénard G. Massa F. Puente N. Lourenço J. Bellocchio L. Soria-Gómez E. Matias I. Delamarre A. Metna-Laurent M. Cannich A. et al.Nat. Neurosci. 2012; 15: 558-564Crossref PubMed Scopus (370) Google Scholar, Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar) by placing this GPCR into neuronal mitochondria. They localized CB1Rs to the outer membrane of mitochondria ("mtCB1R"), and signal transduction analysis revealed reduced cyclic-AMP concentrations and subsequent inhibition of protein kinase A (PKA) and complex I activities (Bénard et al., 2012Bénard G. Massa F. Puente N. Lourenço J. Bellocchio L. Soria-Gómez E. Matias I. Delamarre A. Metna-Laurent M. Cannich A. et al.Nat. Neurosci. 2012; 15: 558-564Crossref PubMed Scopus (370) Google Scholar) leading to reduced neuronal respiration. Moreover, they showed that cell-impermeant CB1R antagonists and agonists (hemopressin and biotinylated HU210, respectively) only partially modify endocannabinoid-dependent synaptic plasticity, implying the participation of mtCB1Rs through altering energy availability in situ (Bénard et al., 2012Bénard G. Massa F. Puente N. Lourenço J. Bellocchio L. Soria-Gómez E. Matias I. Delamarre A. Metna-Laurent M. Cannich A. et al.Nat. Neurosci. 2012; 15: 558-564Crossref PubMed Scopus (370) Google Scholar). Now, their most recent study (Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar) critically advances existing knowledge on mtCB1R signaling by identifying intricate mechanistic details. From in silico predictions, they reveal that a 22 amino acid-long sequence in its N terminus (N1–22) significantly increases the probability of mitochondrial CB1R targeting and import. In elegant rescue experiments using reintroduction of full-length versus N-terminally-truncated CB1Rs on a Cnr1−/− background, they show critical differences in neuronal respiration upon stimulation with synthetic cannabinoid agonists and phytocannabinoids. Most notably, they find that exclusion of CB1Rs from mitochondria blocks cannabinoid-induced amnesia. Through detailed signal transduction analysis (Figure 1), a case for cannabinoid-induced mtCB1R activation releasing Gαi proteins, inhibition of soluble adenylyl cyclase, and PKA-dependent phosphorylation of the complex I NDUFS2 subunit is presented. Since both constitutively active PKA and phosphomimetic NDUFS2 prevent cannabinoid-induced amnesia (Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar), mtCB1Rs are identified as gatekeepers of bioenergetic pathways whose activity is indispensable for the activity-dependent coordination of neuronal network output. The findings by Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar certainly provoke a variety of questions, many of which could individually revise chapters in the exciting history of (endo-)cannabinoid research. First, they posit that an internal 22-residue N-terminal sequence drives CB1Rs to the mitochondria. The most common protein import pathway in mitochondria uses peptide targeting "presequences," which fall outside the N-terminal protein coding region, and are recognized by the translocase of the outer membrane (TOM) complex (Chacinska et al., 2009Chacinska A. Koehler C.M. Milenkovic D. Lithgow T. Pfanner N. Cell. 2009; 138: 628-644Abstract Full Text Full Text PDF PubMed Scopus (986) Google Scholar). An alternative mechanism relies on non-cleavable signals in the protein sequence, including N-terminal signal-anchor sequences recognized by a putative insertase of the mitochondrial outer membrane (Mim1) (Chacinska et al., 2009Chacinska A. Koehler C.M. Milenkovic D. Lithgow T. Pfanner N. Cell. 2009; 138: 628-644Abstract Full Text Full Text PDF PubMed Scopus (986) Google Scholar). Considering the localization of, and N-terminal prediction for, mtCB1Rs, the latter mechanism might be favored, yet its molecular principles remain unknown (Figure 1). Notably, Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar predicted ∼40%–50% probability for the mitochondrial import of CB1Rs. While this is strikingly higher than that of CB2Rs ( 95%). Thus, one prevailing question is whether the likelihood of CB1R recruitment to mitochondria could depend on cell states, neuronal activity, or pathologies. Furthermore, the process of CB1R sorting toward mitochondria versus cell surface seems to emerge as a rate-limiting step to generate functionally distinct intracellular CB1R pools. Historically, epitope tagging of the N terminus of the CB1R was preferred given concerns toward C-terminal modifications influencing G protein binding, signaling, recycling, and the recruitment of interacting proteins. Since the proportion of CB1Rs imported into mitochondria is low (Bénard et al., 2012Bénard G. Massa F. Puente N. Lourenço J. Bellocchio L. Soria-Gómez E. Matias I. Delamarre A. Metna-Laurent M. Cannich A. et al.Nat. Neurosci. 2012; 15: 558-564Crossref PubMed Scopus (370) Google Scholar, Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar), one may wonder whether fluorescent signals in this organelle were previously overlooked or taken as "background." More importantly, N-terminal tagging will change the relative position of the N-terminal 1–22 sequence, likely rendering it dysfunctional in any chimeric protein. Thus, live-cell imaging studies (e.g., neurite outgrowth, cell survival) using N-terminal tags might impinge upon the control of cellular bioenergetics and lead to erroneous experimental outcomes if left uncompensated. Another exciting finding of the Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar report is the cascade of phosphorylation steps leading to the control of complex I, particularly its NDUFS2 subunit. GPCRs are thought to operate in multimers: CB1Rs can co-assemble with, e.g., dopamine, serotonin, and adenosine receptors and undergo G protein switching. Cell-surface CB1Rs are known to sequester Gα, thus rendering other GPCRs signal incompetent (Vásquez and Lewis, 1999Vásquez C. Lewis D.L. J. Neurosci. 1999; 19: 9271-9280Crossref PubMed Google Scholar). Considering the many other receptors localized to mitochondria, it will be interesting to resolve if signaling interactions at the receptor or G protein levels could modulate respiratory output. Another degree of complexity comes from regional differences in the brain since in hypothalamic neurons CB1R-controlled mitochondrial respiration was enhanced (Koch et al., 2015Koch M. Varela L. Kim J.G. Kim J.D. Hernández-Nuño F. Simonds S.E. Castorena C.M. Vianna C.R. Elmquist J.K. Morozov Y.M. et al.Nature. 2015; 519: 45-50Crossref PubMed Scopus (276) Google Scholar) rather than suppressed as seen in hippocampal neurons (Hebert-Chatelain et al., 2016Hebert-Chatelain E. Desprez T. Serrat R. Bellocchio L. Soria-Gomez E. Busquets-Garcia A. Pagano Zottola A.C. Delamarre A. Cannich A. Vincent P. et al.Nature. 2016; 539: 555-559Crossref PubMed Scopus (249) Google Scholar). In the hypothalamus, mtCB1R action was associated with UCP2, which itself is tied to the acute control of complex behaviors (Hermes et al., 2016Hermes G. Nagy D. Waterson M. Zsarnovszky A. Varela L. Hajos M. Horvath T.L. Mol. Metab. 2016; 5: 415-421Crossref PubMed Scopus (17) Google Scholar). Whether and how the differential effects of mtCB1R signaling can be reconciled at the signaling and cellular levels and how these relate to specific neuronal output patterns will require further analysis. Besides endocannabinoids, Δ9-tetrahydrocannabinol (THC) also modulates mtCB1Rs. Thus, it is appealing, and of chief human relevance, that episodic amnesia upon cannabis smoking might arise from THC limiting neuronal respiration: "switching off" neuronal activity and erasing synaptic plasticity. Nevertheless, the effects of mtCB1Rs on postmitotic and stationary adult neurons might be less than under developmental conditions. Cell division, migration, and neurite outgrowth by far surpass energy demands of any mechanism studied so far. Consequently, reliance on mtCB1Rs might predominate during both pre- and postnatal developmental periods. As such, mitochondria accumulate at the active growing ends of microtubules, and these processes are particularly sensitive to CB1R modulation. Therefore, future studies on mtCB1Rs might uncover molecular substrates by which even episodic cannabis use during pregnancy or lactation could permanently modify neuronal connectivity, glial support, and sensitivity to secondary insults in the offspring's nervous system. CB1Rs are among the most abundant GPCRs in the human body. CB1Rs are harnessed for therapeutic benefit in lung, immune system, bone, muscle, eye, skin, and pancreas. Linking cannabinoid signaling with cellular bioenergetics shows particular promise to resolve existing therapeutic shortcomings related to the control of secretory processes, cell renewal, cell motility, and resilience against the many environmental stressors modern humans experience during our ever-expanding lifespan. It will therefore be crucial that conceptual advances and ideas in this field of mitochondria research are adequately matched with technological advances and the development of quality-controlled tools to ensure enduring success. We thank Dr. E. Keimpema for his constructive remarks. This work was supported by the Swedish Research Council (VR-2013-3080, T.H.), Hjärnfonden (T.H.), the Novo Nordisk Foundation (T.H.), the European Research Council (Secret-Cells; ERC-2015-AdG-695136, T.H.), and NIH grant AG051459 (T.L.H.). T.H. is funded by GW Pharmaceuticals on research projects unrelated to the topic of this paper preview.