Severe alterations of the mitochondrial machinery involved in energy generation lead to a group of progressive and usually fatal pathologies collectively known as primary mitochondrial disease (MD), affecting 1:5000 births. Currently, there is no cure and the treatments available are mostly ineffective. MD predominantly affects high energy-requiring organs such as the brain. However, not all neurons are equally vulnerable to MD, but rather show a striking anatomical and cellular specificity. The mechanisms conferring neuronal resistance or vulnerability to MD are currently unknown. To date, research in mitochondrial disease has been hindered by the high degree of variability in disease progression and severity in human patients, and the intrinsic heterogeneity of mitochondria. Furthermore, model systems both in vivo and in vitro to date have failed to identify the mechanisms underlying the anatomical and cellular specificity of these pathologies. Hence, if one could possess the ability to identify, purify and compare cellular and mitochondrial changes from affected and healthy neurons, a better and more significant insight on the biochemical and functional changes associated with MD would be obtained. To address this issue, two ground-breaking approaches were developed to define the molecular basis of neuronal susceptibility to mitochondrial disease. These approaches have the potential to propose new therapeutic targets for MD and are easily applicable to other pathologies associated with mitochondrial dysfunction such as diabetes or neurodegenerative processes.