A mighty (ochondrial) fight?

Disputes about scientific findings unfold following a well-rehearsed script. One researcher (let’s call her Dr. X) makes an intriguing new observation and manages to get it past the refereeing process of a top journal. The study is published and Dr. X happily moves on to her next project. But she is interrupted by an annoying piece of news e a report from the lab of another scientist (let’s call him Dr. Y) flatly contradicts her. Making things worse, Dr. Y is a real expert in the particular set of techniques used by Dr. X, so much so that his opinion cannot be simply tossed away as irrelevant. But Dr. X believes in her data, so she sets out to vindicate them. She runs new, even more stringent tests and replicates successfully her previous work. Will the new experiments convince Dr. Y? Probably not (unless, of course, he was directly involved in running them). And what about other scientists? Will they be convinced? Friends and family will probably take sides, but the majority will more wisely wait for the dust to settle. If Dr. X stands by her story and others reproduce her findings, the field will accept them and the debate will be eventually forgotten. The wait can be uncomfortable, especially for the parties involved, but this is how science normally works e scientists tend to adopt theories by making conscious or (more often) unconscious statistical assessments of the overall evidence available to them. An article [1] and two back-to-back commentaries published in the present issue of Molecular Metabolism present an interesting experimental controversy, which will likely follow the standard course outlined above. The apple of discord was tossed two years ago by the unexpected finding, made in the laboratory of Giovanni Marsicano at the INSERM in Bordeaux, that CB1-type cannabinoid receptors may be present on the membranes of neuronal mitochondria, and that pharmacological activation of these receptors may decrease the conversion of biochemical energy from nutrients into ATP [2]. This result was surprising enough to warrant a News and Views piece aptly titled ‘Do cannabinoids reduce brain power?’ [3]. But why the surprise? Because the standard view of CB1 receptors is that they are localized, like other G protein-coupled receptors, to the surface of neuronal and glial cells, where they wait for exogenous chemicals (like D-tetrahydrocannabinol in marijuana) or endogenous neurotransmitters (like anandamide and 2-arachidonoylglycerol) to activate them. According to this view, the mitochondrion is not a proper place for a cannabinoid receptor to be hanging around. Yet, Marsicano and collaborators felt that they