Biomedical Engineering Reference
In-Depth Information
overcome this issue, we utilized primary cultured dissociated neurons obtained from mouse hippocam-
pus and applied SHG chromophores from the outside of the cell, in the bath solution. The advantage of
this system is that when the density of neurons is low enough, there is little overlap between membrane
structures in the culture dish, which will alter the SHG signal itself and/or voltage sensitivity when the
chromophore is applied in the bath. Indeed, such an uncertainty is one of the major challenges in apply-
ing SHG membrane potential imaging to membrane-rich densely packed samples such as brain slices.
Using very low-density culture neurons, SHG signals could be observed from all over the neuronal
structures including axons and dendrites, which were confirmed by immunocytochemical analysis with
specific markers for axons and dendrites.
When the action potential was generated at the soma by positive current injection, reliable SHG signal
changes were observed along axonal arbors. Again, as SHG imaging is a quantitative imaging technique,
we could compare peak SHG signal changes at the axons and soma and conclude that there is no differ-
ence between voltage deflections at the soma and axons. Furthermore, we realized that the variability
in voltage changes at the axons is very small, not like calcium signals reported previously [35,36]. These
data suggest that the variability in calcium signals reported previously is likely due to the differences in
calcium channels rather than voltage deflection per se.
In addition, we performed different sets of experiments to address the nature of this nonattenuating
propagation of action potential into axonal arbors. Action potentials are generated by the sequential
activations of voltage-gated sodium channel and potassium channels [37]. When axons are endowed
with these channels in enough density, action potentials regenerate themselves on their propagation
direction. In contrast to these regenerating active propagation of membrane potentials, passive propaga-
tion of voltage changes are attenuating in nature. To address to what extent this kind of passive propaga-
tion occurs in the axons, we induced nonregenerative voltage changes at the soma and measured SHG
signal changes in the axons. In sharp contrast to what was observed with action potential, these non-
regenerative membrane potential changes attenuated significantly along the axonal arbors. These data
reveal two types of voltage propagations from soma to axonal arbors in these cultured hippocampal
neurons: nonattenuating regenerative action potential and highly attenuating nonregenerative voltage
fluctuations.
10.6 conclusions and Perspectives
As mentioned earlier, membrane potential measurement by SHG imaging has a unique advantage in that
it is selective for plasma membranes and it can also provide quantitative information that could not have
been obtained from other methods. As this kind of information is crucial in understanding the physiol-
ogy of neurons and other excitable cells, expectations are high for further applications of this technique.
At the same time, the history of membrane potential measurement by SHG is still recent and there-
fore there is still plenty of room for improvement. Especially, at present, the signal-to-noise ratio is still
quite low compared to more established fluorescence-based membrane potential imaging techniques.
Therefore, as has been the case so far, efforts need to be made in parallel to further refine the technique
itself and to apply the technique to physiological questions. For this goal, we believe that collaborative
efforts between biologists and researchers from other fields such as synthetic chemists are crucial. We are
hopeful that such collaborations will help further develop the technique and shed new light on the physi-
ology of neurons and other excitable cells. Finally, we briefly describe other potential applications as well
as two important aspects of researches in considering future developments of the field.
10.6.1 Applications
There are many targets and physiological events for which we have limited knowledge about the mem-
brane potential dynamics. First of all, membrane potential dynamics in small structures in neurons
such as dendritic spines and axons have just begun to be investigated. For example, there still is no direct
