Important legal notice
   
Contact   |   Search   
RTD info logoMagazine on European Research Special Issue - April 2005   
Top
 HOME
 TABLE OF CONTENTS
 EDITORIAL
 In Brief

Download pdf de en fr


BIOLOGY
Title  The nuts and bolts of remembering

A mass of information is constantly bombarding the brain. It must select what it needs and record what it deems to be useful for subsequent recall. But how does such a complex process work? Neurologists now believe they have identified the principal elements involved: a neurotransmitter, namely glutamate; a mechanism, known as long-term potentiation (LTP); and a region of the brain, the hippocampus. But how they interact remains a mystery.

Hippocampus neurons. The hippocampus plays a major role in memorising recent events and is part of the limbic system (or paleocortex) that is the site of emotions and memory processes.  © INSERM/A.Represa-Bermejo
Hippocampus neurons. The hippocampus plays a major role in memorising recent events and is part of the limbic system (or paleocortex) that is the site of emotions and memory processes.
© INSERM/A.Represa-Bermejo
It was 110 years ago, at the Royal Society in London, that the renowned Spanish neuroanatomist Santiago Ramón y Cajal came up with a revolutionary hypothesis: the learning process facilitated the expansion and growth of what he referred to as ‘protuberances’ that interconnect the neurons and which we today call synapses. The hypothesis was revolutionary because it offered a solution to the paradox of how the brain – anatomical studies of which had shown stable and permanent neuronal circuits – is able to remember experiences that are by nature fleeting. To paraphrase Heraclitis for whom “you never bathe twice in the same river”,   Ramón y Cajal essentially suggested that we never think twice with the same brain, as each new experience lived changes the neuronal circuits of which it is composed. 

Synaptic plasticity
Lacking experimental arguments, this visionary idea remained speculative until 1973. That was when the Swede Timothy Bliss and the Norwegian Terje Lømo showed that, in rabbits, brief high-frequency stimulation of a neuronal pathway sending sensory information from the cortex to the hippocampus produced a lasting improvement in synaptic activity: neurons targeted in the hippocampus acquired an increased sensitivity to any subsequent stimulation. The most remarkable feature of this form of ‘plasticity’ is that although produced in just a few tenths of a millisecond the changes produced in the synapses remain for weeks if not months.

This discovery, known as long-term potentiation (LTP), sparked considerable enthusiasm in the scientific community. However, the actual link between this synaptic plasticity and the learning and memory processes had yet to be established. In the 1980s, a number of laboratories turned their attention to simple forms of associative learning in rats, such as the association between a sound and a mild electric shock. After a period of conditioning, the animal reacted to the sound alone in exactly the same way as if it had received the electric shock. Thus, as the learning process progresses so the efficiency of the synaptic transmission in the hippocampus circuits increases. However, these data have a correlation value only and are not proof of a relationship of cause and effect. Pharmacology and genetics subsequently provided the answers that electrophysiology could not. 

Glutamate transmission
Research teams headed by Edvard Moser and Menno Witter have shown that direct connections exist between the entorhinal cortex and the CA1 region of the hippocampus and are sufficient to feed the hippocampus place cells. Here: marker visualisation of interaction by neurons passing from the CAI zone of the hippocampus (not visible, located to the right of the image) to the soma (shown as a crescent) in the C1 zone.
Research teams headed by Edvard Moser and Menno Witter have shown that direct connections exist between the entorhinal cortex and the CA1 region of the hippocampus and are sufficient to feed the hippocampus place cells. Here: marker visualisation of interaction by neurons passing from the CAI zone of the hippocampus (not visible, located to the right of the image) to the soma (shown as a crescent) in the C1 zone.
In the late 1980s, the Richard Morris group at Edinburgh University showed that when rats are administered an NMDA-type glutamate receptor blocker (a very common neurotransmitter in the hippocampus), which blocks the plasticity of the synapses without disturbing the transmission of neuronal messages from another receptor, the animals are no longer able to perform a spatial navigation task, such as finding their way in a maze. As the blocker doses increase so the synaptic plasticity is reduced and memory lapses increase. Genome engineering technologies have since lent new arguments to support the role of this ‘glutamatergical transmission’ by the NMDA receptor in LTP Induction. In transgenic mice, in which the neurons in certain areas of the hippocampus do not express the receptor, LTP is eliminated and the animals show major deficiencies in spatial memory. 

Edvard Moser, a neurobiologist at Trondheim University (NO) and coordinator of the European Nappy (Network analysis of hippocampal memory processing) project, explains that: “it is now largely accepted that the LTP and its opposite, Long-Term Depression (LTD), are key mechanisms in storing information in the central nervous system. LTP also seems to play a very different role in the hippocampus than in the spinal cord. Researchers are currently focusing on understanding the operations performed by given neuronal circuits, by virtue of LTP and LTD. These operations serve to link the neurons into cellular assemblies and to modify their entry/exit relations.” 

Place and direction cells
The Nappy project, which receives EU funding of €1.83 million under the Fifth Framework Programme(1), is a good example of this shifting focus of neurobiologists.

Study of the effects of nicotine in the hippocampus of rats. The expression of the PSA-NCAM protein and production of newneurones both have an impact on the plasticity of the synapse, learning and memory. It is noticeable that the number of new neurones (dark core) is much greater in the control group (left) than in those exposed to nicotine (right). © INSERM/N.Abrous & P;V.Piazza
Study of the effects of nicotine in the hippocampus of rats. The expression of the PSA-NCAM protein and production of newneurones both have an impact on the plasticity of the synapse, learning and memory. It is noticeable that the number of new neurones (dark core) is much greater in the control group (left) than in those exposed to nicotine (right). © INSERM/N.Abrous & P;V.Piazza
Study of the effects of nicotine in the hippocampus of rats. The expression of the PSA-NCAM protein and production of newneurones both have an impact on the plasticity of the synapse, learning and memory. It is noticeable that the number of new neurones (dark core) is much greater in the control group (top) than in those exposed to nicotine (bottom).
© INSERM/N.Abrous & P;V.Piazza
To appreciate its importance, it must be viewed in the context of earlier research carried out by John O'Keefe and John Dostrovsky, both of University College London (UK), who discovered that certain hippocampus cells, while silent most of the time, fired off sudden electrical signals when the animal found itself at certain places in the environment. The cells in question are place cells, whose electrical activity is linked to the animal’s position in space. Their fields of activity, that is the zone in which the firing is intense, are established within a few minutes of entering a new environment and, once established, can last for several weeks.

Subsequently, direction cells – corresponding to the orientation of the animal’s head – were discovered in several regions connected to the hippocampus. A network linking place cells and direction cells could, therefore, be the support for a certain kind of spatial memory. Until recently, it was believed that this network was the entorhino-hippocampic loop that links the entorhinal cortex to the hippocampus, where the information is processed by a succession of synapses associating three precise regions: the gyrus dentatus and the two regions of the cornu ammonis, known as CA3 and CA1. 

It is this excessively simple concept that the research carried out by Moser and his colleagues, published in the journal Science in 2002, has called into question. In co-operation with the Menno Witter group at the Free University of Amsterdam (NL), they demonstrated that direct connections exist between the entorhinal cortex and the CA1 region of the hippocampus, and that these are sufficient to feed the hippocampus place cells. More than that, they showed that place cells already exist in the entorhinal cortex, even if their activity profiles are different. Conversely, destruction of the entorhinal cortex by a toxin disturbs the response fields of the place cells in the CA1 region. Rather than being a simple conveyor of information, the entorhinal cortex is, thus, revealed as a genuine centre for the pre-processing of spatial information. 

Mobilising memory
And where does memory fit into in all this? “A major aspect of our project is to develop new behavioural paradigms that would make it possible to distinguish experimentally between the various types of episodic memory, such as the memory of what event occurs when and where,” explains Edvard Moser. It is precisely this latter aspect that Moser and his colleagues now hope to investigate by means of a learning test involving associated pairs. In these tests, that are not unlike those used in clinical psychology to detect amnesia, a rat has to learn how to memorise the association of two successive stimuli – a smell and a place – in order to find food. In 2003, Richard Morris and his team had shown in Nature that glutamatergic transmission in the hippocampus is vital to learning a new odour/place pair but not to remembering a pair learned previously.

Thanks to new technology developed by the British firm Axona Ltd, a partner in the project, which permits the wireless recording of the simultaneous activity of several neurons in an waking animal, the consortium is now hoping to use the test developed by Morris to assess its principal hypothesis, namely that when learning a new activity the entire hippocampus is reactivated when the memory is mobilised. This is an ambitious hypothesis that, according to Moser, could provide “a new understanding of the function of the hippocampus, with the emphasis on the storing and recall of memories rather than spatial learning.”

(1) LTP has already been the subject of 17 research projects under the Fourth and Fifth Framework Programmes.


Printable version

Features 1 2 3 4 5
  Science is embedded in culture and history
  Inside the memory machine
  DNA, life’s memory
  The nuts and bolts of remembering
  From forgetfulness to dependency


  TO FIND OUT MORE  
 
  • Nappy project (Network analysis of hippocampal memory processing)
  •  

      CONTACTS  
     
  • Edvard Moser – Norwegian University of Science and Technology – Centre for the Biology of Memory, Trondheim
  •  


       
      Top
    Features 1 2 3 4 5

    TO FIND OUT MORE

    • Nappy project (Network analysis of hippocampal memory processing)

    CONTACTS

    • Edvard Moser – Norwegian University of Science and Technology – Centre for the Biology of Memory, Trondheim