With the ability to measure levels of different mRNA in an individual cell, as well as to measure the movement of organelles such as mitochnodria, we need to consider the 'dynamic state' of these particles as they move and are targeted within neurons. Why? Even though you may know if a specific gene is expressed at higher levels in a certain condition, you need to see if this gene is preferentially targeted to different sites within the neuron. Other components, such as mitochondria, can act as important energy sources for different activities throughout the neuronal processes, as well as the soma. If we can accurately predict how these particles are transported and localized, we will develop better insight onto how these movements can be disrupted in disease or injury.
When we examine individual particles within a cell, we need to consider the stochastic based movement of the single particle. This is in contrast to most predictions at larger length scales, which use continuum descriptions to predict the bulk movement of proteins or particles in solution. As an example, one can model the motion of a particle that is found in neuronal dendrites and axons - the ribonucleoprotein (RNP) complex. The RNP complex is probably very important for neuronal function, since it is a combination of a motor molecule (e.g., kinesin), mRNA species, possibly linker or chaperone proteins between the mRNA and motor molecule, and targeting proteins. One can stain for these RNP complexes within neurons (see the image below as an example), and techniques are being developed to track these complexes in real time. Andrew Taylor in our lab has an interest in this area, and is working in the Eberwine lab to see if we can identify the components that are in these RNPs.
We have developed simulations of how the RNPs move within the neuron, using some information from the literature. We 'assembled' a regular microtubule architecture, similar to the structure that is seen on EM examination of axons. We placed a RNP particle at a random starting position within this microtubule array, and we allowed the particle to freely diffuse, bind and unbind to the individual microtubules, and move under rapid transport when bound to the microtubules. The images of the simulations are shown below - the cross section of the microtubule array (show as open circles), and the path of the diffusing particle (shown in red). If you 'track' the motion of the particle along the length of the axon, you get an episodic motion of the particle (see figure). The motion is not smooth because the motion is a mix of freely diffusing motion, rapid fast motion, and stationary motion. We have used these predictions to examine how the movement of the particles may be affected by the density of the microtubule array (net transport rates slow as the microtubules become more sparse), the binding kinetics of the particle to the microtubules, and the overall transport rate as the number of RNPs increase.
At this level of simulation, there is an opportunity to rapidly explore other transport phenomenon. For example, we have become interested in mitochondrial motility because of its potential role in neuronal apoptosis, as well as its role in recovery after injury. We are tracking the motion of individual mitochondria, which are transported along both the microtubule track and, to a lesser extent, the cortical actin network within the neuron. The fundamental steps are the same as RNP transport - the mitochondria binds and unbinds to the microtubule network via motor molecules, and the particles can freely diffuse when not bound. The significance of the questions in this area are slightly different - we want to learn some of the limiting steps for mitochondrial transport because it may limit how quickly the neuron can recover from an insult, and it may also change the ultimate outcome (survival?) of the neuron to a stimulus.