Storage of information in cultured neurons

Photo Sharing and Video Hosting at PhotobucketCultured neurons seem like ants away from their colony: removed from their parent organ, dissociated from their fellow workers and placed into an unnatural environment. But neurons plated onto a culture dish connect to each other, forming simple neural networks that give rise to spontaneous electrical activity. And, in recent years, researchers have developed culture dishes containing arrays of microelectrodes embedded within them, such that the electrical activity of the cultured neurons can be recorded. These new techniques have revealed the remarkable functional properties of neurons in culture – the cultured networks of dissociated cells can “learn”. In other words, they can modify their initially spontaneous activity into something purposeful, such as controlling a flight simulator or controlling the movements of artificial animals in a virtual environment.

Now, Itay Baruchi and Eshel Ben-Jacob of Tel Aviv University show that networks of cultured neurons can also store information. The image on the left shows their experimental set up. Nerve cells were isolated and cultured on a specialized dish in which microelectrodes are embedded so that the electrical activity of the cells can be recorded. A micropipette was then used to apply picrotoxin to small groups of cells in specific locations on the culture dish. Picrotoxin is a GABA receptor antagonist; its addition to the culture dish therefore suppressed the activity of inhibitory interneurons in the cell culture. As a result, synchronized bursting events (SBEs) – waves of electrical activity with specific patterns in both space and time – were observed in the nerve cell culture.

The cultured neurons “stored” information about the patterns of electrical activity evoked in the network by application of picrotoxin. Several different types of SBEs were evoked in the network, each of which starts at a specific location in the culture dish and is propagated along a specific trajectory. This activity continued as long as picrotoxin was applied; when the drug was removed, the cultured cells returned to their basal activity. But the SBEs could be precisely reproduced later on. If the initial application of picrotoxin to a specified location on the culture dish generated a specific type of SBE, exactly the same pattern of activity could be elicited up to 40 hours later by applying picrotoxin to the same location. Thus, the collective activity of the cultured cells had somehow been “imprinted” within the network.


Baruchi, I. & Ben-Jacob, E. (2007). Towards neuro-memory-chip: Imprinting multiple memories in cultured neural networks. Phys. Rev. E. doi: 10.1103/PhysRevE.75.050901. [Full text]

2 thoughts on “Storage of information in cultured neurons

  1. Pingback: lisa paitz spindler

  2. I´ve been reading “Adavances in network electrophysiology: usin multi-electrode arrays” author, Taketani and Baudry, and this kind of laboratory manipulation of neural tissue is a promising avenue that allow us to study directly the responses of neural cells with the advantage of knowing in first hand its nature, and then to create target drugs or even to create computational interfaces that one day could bring into existence something like a “bionic man”.
    The outline of the post is promising regarding the latter.

Comments are closed.