Society’s solution to energy requirements is to generate power remotely, then transmit it over vast distances. In contrast, biological systems adopt a local approach, generating power wherever it is needed. Perhaps nowhere is this more evident than in neurons, where mitochondria, power plants of the cell, are dispatched to presynaptic nerve terminals at the tips of distant axons. But what about postsynaptic terminals, which sprout all over the dendritic tree? Are they also provided with local power generators? In tomorrow’s Cell, Morgan Sheng and colleagues at the Picower Center for Learning and Memory at MIT demonstrate how distribution and redistribution of mitochondria in dendritic arbors can impact synaptic plasticity, a prerequisite for learning and memory. The results add another twist to the link between mitochondria and neurodegenerative disorders, such as Alzheimer and Parkinson diseases (see ARF related news story and ARF related story).

First author Zheng Li and colleagues investigated mitochondrial maneuverings in hippocampal dendrites by using time-lapse confocal fluorescence microscopy to track the organelles. Consistent with previous research, the authors found that mitochondria were localized mainly in dendritic shafts, with only a small fraction (five to 10 percent) turning up in the stubby protrusions, or spines, that harbor postsynaptic terminals. However, when Li and colleagues stimulated these neurons (using potassium chloride in the case of cultured neurons and electrodes for hippocampal slices), they recorded redistribution of the mitochondria and about a twofold increase in the numbers of mitochondria in dendritic spines. In the case of the hippocampal slices, the additional mitochondria only turned up near sites of stimulation. More distant spines in the same neuron were not affected, indicating that the response is a local one.

Might this redistribution have functional significance? To test this, the authors enlisted the help of two proteins, dynamin-related protein 1 (Drp1) and optic atrophy 1 (OPA1). Together, these proteins have been found to exert almost complete control over alterations to mitochondrial morphology and distribution in other cells. When Li expressed dominant-negative variants of Drp1 and OPA1 in hippocampal neurons, he found that the numbers of mitochondria in the cell body were about the same as in wild-type cells, but numbers in dendrites dropped by about half. Perhaps even more dramatic was the fact that, when cells expressed the dominant-negative Drp1 and OPA1, the numbers of dendritic spines fell by about 80 and 55 percent, respectively. In contrast, overexpression of normal Drp1 or OPA1 resulted in an increase in dendritic spines. This finding “…implies that mitochondria are not only required but limiting for the formation and/or maintenance of synapses,” noted the authors.

Dynamics of dendritic mitochondria
Using fluorescent confocal microscopy, Morgan Sheng and colleagues have tracked mitochondria (labeled with red fluorescing marker DsRed2) in dendritic arbors (fluorescing green due to expression of an enhanced version of yellow fluorescent protein). In this image, overexpression of Drp1, a protein involved in mitochondrial distribution and morphology, leads to a dramatic increase in the number of mitochondria (which appear yellow against the green background) in dendritic shafts and spines. [Image courtesy Morgan Sheng and Cell Press.]

But what about synaptic plasticity, such as the ability of neurons to generate synapses in response to excitatory stimuli? Are mitochondrial dynamics crucial for this phenomenon? To test this, the authors measured the number of synapses in hippocampal cells that were stimulated with potassium chloride. Nine hours after this treatment, wild-type neurons had about a 40 percent increase in synapses; neurons expressing the dominant-negative Drp1 saw no extra synapses; and cells overexpressing Drp1 saw an almost twofold increase.

Together, the data indicate that not only are mitochondria shunted to areas where dendrites are being stimulated, but also that the growth of new synapses is dependent on mitochondrial dynamics. Examining this further, the authors found that the motility of mitochondria was increased by about 75 percent when neurons were stimulated, and decreased by about the same amount when inhibited. Basal motility was unaffected by the expression of the dominant-negative Drp1, but changes elicited by excitation or inhibition were abolished.

“Our findings point to a high level of dynamism in dendritic mitochondrial distribution, which is regulated by synaptic activity and correlated with synapse morphogenesis,” write the authors. It is also worth noting that mitochondrial dynamics may be related to the microtubule-associated protein tau, which is the major component of neurofibrillary tangles, a hallmark of Alzheimer disease (see ARF related news story).—Tom Fagan


  1. Mitochondria in Alzheimer Disease: Pulling the Plug on the Synapse
    Zheng Li and colleagues provide compelling data and discussion to confirm the growing suspicion that proper mitochondrial structure and function are requisite for synaptic operation. Indeed, it is not only mitochondrial function that is important, but also the location of these organelles. Through the use of transfected dynamin family GTPases (Drp1 and OPA1), which regulate mitochondrial fission and distribution, Morgan Sheng’s group demonstrate that the localization of presynaptic mitochondria are a limiting factor in synapse creation and maintenance. In addition, their data suggest that neuronal activity also contributes to region-specific mitochondrial allocations (Li et al., 2004).

    While the role of mitochondria in the dendritic spine and synapse remain open for definitive explorations, their ATP-manufacturing and calcium-regulating properties may support the necessary microenvironment for spine and synapse activity. However, as the authors remind us, knowledge about mitochondrial movement and co-related synapse activity is sparse at best.

    Synaptic loss and mitochondrial dysfunction are both associated with Alzheimer disease (AD) (Hirai et al., 2001) and connecting the two, as shown by this present work, is attractive. In addition, the neuronal cytoskeleton, responsible for intracellular organelle transport, is also defective in AD pathogenesis (Cash et al., 2003) and likely plays a synergistic role in affecting the intracellular location of mitochondria. As with other paradigms in AD, such defects eventually reach a critical mass and lead to occult pathology and disease (Zhu et al., 2004).


    . The importance of dendritic mitochondria in the morphogenesis and plasticity of spines and synapses. Cell. 2004 Dec 17;119(6):873-87. PubMed.

    . Mitochondrial abnormalities in Alzheimer's disease. J Neurosci. 2001 May 1;21(9):3017-23. PubMed.

    . Is Alzheimer's disease a mitochondrial disorder?. Neuroscientist. 2002 Oct;8(5):489-96. PubMed.

    . Alzheimer's disease: the two-hit hypothesis. Lancet Neurol. 2004 Apr;3(4):219-26. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Aβ Production Linked to Oxidative Stress
  2. Protection Against Parkinson’s—How the DJ Changes Station
  3. Tau Kinase Clears Microtubules—Keeps Axonal Transport on Track

Further Reading

Primary Papers

  1. . The importance of dendritic mitochondria in the morphogenesis and plasticity of spines and synapses. Cell. 2004 Dec 17;119(6):873-87. PubMed.