Proust and Long-Term Memory

Jonah Lehrer, in Proust was a Neuroscientist, suggests that Marcel Proust, in his
writing, predicted the, “instability and inaccuracy of [long-term] memory…” [1]. Before
the dawn of the 21st century, neuroscience suggested that memory, valuable pieces of
information, were archived in a structure in the brain, such as the lateral and basal nuclei
of the amygdala.  In 2000, research on rats with fear conditioning and a protein inhibitor
showed that the act of remembrance (reactivation) in fact changed the molecular
underpinnings of the memory by making the memory ‘labile’ once again [2].  Therefore,
new protein synthesis at the synapse was needed to ‘reconsolidate’ the information to
storage for long-term memories.  Therefore, Lehrer argues that the recent neurobiological
understanding of memory and its storage mirror that of Proust: memory is not a stable
entity that can be identically recalled repeatedly, but is instead constantly reformed
through modified synaptic connections every time it is accessed [1].

Biological efforts to understand the basis for memory begins with a comparison of
‘behavioral memory’ and synaptic plasticity as both exhibit short and long-term types [3].  
Synaptic plasticity, the change in the electrical activity across the space dividing synapses
can be divided into mnemonic plasticity that is directly related to short and long-term
memory, metaplastic plasticity that is related to higher order learning functions, and
homeostatic plasticity that may allow for memory, but is not directly related to any
learning or memory function [4].

Research has used this framework to infer Short and Long-Term memory
Formation (STF and LTF respectively) from physiological characteristics of forms of
mnemonic plasticity.  Specifically, observations of synaptic plasticity from the
stimulation of only one pulse of serotonin (5-HT) in Aplysia stimulated short-term
memory that correlates with modifications of existing proteins in the pre and post-
synapse [3].  Physiologically, the influx of serotonin activated protein kinases A and C,
which increased calcium concentration and presynaptic vesicular traffic respectively [5].

Unlike changes to only the plasticity that exist for short-term memory, long-term
memory storage also correlated with the expression of new proteins.  Five repeated
pulses, as opposed to one for STF, correlate with LTF.  Protein kinase A is activated
similar to STF, except the catalytic subunit moves to the nucleus to upregulate ApCREB-
1.  MAP kinases are also activated that similarly localize to the nucleus to remove
repression from ApCREB-2, such that both transcription factors can act on the promoters
of cAMP genes.  ApC/EBP, a transcription factor is also upregulated downstream [5].  
Finally, several epigenetic changes occur, such as histone tail acetylation and DNA
methylation that correspond to changes in synaptic plasticity with LTF [5,6].

The role of CREB, in particular, serves several useful roles in LTF.  For the
initialization of memory storage, results from studies of Aplysia show that CREB
regulates the transcription of new proteins characteristic of LTF such as alpha-CaMKII
[7,8].  The ability for the memory to be reconsolidated after
it has been recalled and made labile again is provided by the local rapamycin-sensitive
translation also provided by a CREB isoform that is homologously found in humans,
mice, and Drosophila [3].  The CREB isoform is only needed during the
‘stabilization phase’ (i.e. 24-72 hrs), outside of which CREB will not effectively
reconsolidate memories [9].  CREB’s duplicitous functions-- both a
transcription factor as well as a mediator of translation; its role in both initiation and
stabilization of long-term memory—in the midst of the rapid changes of synaptic
plasticity may be explained by its prion-like behavior.  As described by the CDC, prions
are proteins prone to misfolding that can cause neurological illnesses such as Mad Cow
disease [12].  A population genetics study of the prion protein gene (PRNP) in humans
demonstrates that the particular mutations in the gene promote a protein phenotype
favorable to LTF, while others promote a phenotype that is deleterious to LTF.  The
authors use these results to imply that the CREB protein, one of the most consistent
actors throughout the process of LTF, may be where PNRP is expressed given its Protean
prion-like behavior [10].

Lehrer’s presentation of the more recent neurobiological model of memory
storage considers the plasticity of the nervous system (e.g. synaptic plasticity) to redefine
memory not as something static to be archived and recalled, but as a constantly evolving
entity that is redefined and rearchived [1]. From a biological perspective, I have concerns
with these observations of the neural basis of memory.  Although Lehrer makes the
argument that the malleability of the CREB isoform is the fundamental key that allows
LTF, the PRNP mutations that cause LTF may in fact act on a host of other proteins in
the synaptic region besides CREB.  Also, the correlative observations of epigenetic
changes opens up many unanswered questions concerning how these epigenetic changes
relate to synaptic gene regulation and plasticity, and the corresponding effects on
memory [11].

Lehrer points to the CREB isoform’s double-identity as the critical neurological
component of long-term memory: it acts as a crucial component of LTF over time, and
also exhibits malleability to adapt to different roles over time.  While Lehrer’s argument
that CREB is the principal neurophysiological component of LTF is convincing, his
argument hinges on the presumption that the rest of the synaptic system remains stable
throughout LTF.  I find this assumption tenuous as the growing information on neuronal
plasticity, specifically synaptic plasticity, implies that the brain and nervous system are
consistently adapting in ways not so clearly delineated throughout development [4].
Therefore, the neurobiology suggests that the notion that one’s memory of his or her first
romantic experience will remain the same memory when one is older is false.  Instead,
memory behaves more as something that is redefined and recreated over time.  Therefore,
the original memory of a romantic experience from childhood is as equally plausible an
explanation for the original occurrence as any other memory recollected throughout later
life.

The story that Lehrer presents also implies a distributive organization to long-
term memories.  These observations have direct implications for a sense of valuation or
hierarchy of memories, and indirect implications for the study of history.  If all memories
are ‘created equal’ as explained by the same CREB protein, then are they valued the same
by our nervous system?  In my life, I have experienced my share of strong memories
versus memories that I would say are not as prevalent.  Therefore, the neurobiological
story of memory seems to be at odds with my personal observations of my own memory.  
Perhaps all ‘boxes’ in the nervous system value memories the same, but the I-function is
the discriminating component that selects what it ‘wants’ to remember.

Similarly, in debates over how one should approach understanding the story of the
past, some believe that the objective experience of the past is not entirely knowable, but
that the story that will always be incorrect since it is dependent on the point of view of
the teller.  In contrast, other historians believe that there is an objective story of the past
that can be discerned among the many different accounts of what happened.  If we are to
assume that Lehrer’s story for the mutability of memory is true, then an objective study
of history is not only very difficult, but impossible, as even the memory of an event will
perpetually be perturbed over the life of the story teller.

Works Cited:

[1] Lehrer, Jonah. Proust was a Neuroscientist. New York: First Mariner Books, 2007

[2] Nader, Karim, Schaefe, Glenn and Joseph LeDoux (2000) “Fear Memories require
protein synthesis in the amygdala for reconsolidation after retrieval.”  Nature.  406: 722-
726.

[3] Si, Kausik, Giustetto, Maurizio, Etkin, Amit, Hsu, Ruby, Janisiewicz, Agnieszka,
Miniaci, Maria, Kim, Joung-Hun, Zhu, Huixiang, and Eric Kandel (2003) “A Neuronal
Isoform of CPEB Regulates Local Protein Synthesis and Stabilizes Synapse-Specific
Long-Term Facilitation in Aplysia.” Cell. 115: 893-904.

[4] Kim, Sang Jeong and David J. Linden, (2007) “Ubiquitous Plasticity and Memory
Storage.” Neuron. 56.

[5] Lee, Yong-Seok, Bailey, Craig H., Kandel, Eric R., and Bong-Kiun Kaang (2008)
“Transcriptional regulation of long-term memory in the marine snail Aplysia.” Molecular
Brain. 1:3.

[6] Miller, Courtney, Campbell, Susan, and J. David Sweatt (2008) “DNA methylation
and histone acetylation work in concert to regulate memory formation and synaptic
plasticity.”  Neurobiology of Learning and Memory. 89:599-603.

[7] Richter, Joel (2001) “Think Globally, translate locally: What mitotic spindles and
neuronal synapses have in common.” PNAS. 98: 13: 7069-7071.

[8] Martin, Kelsey, Casadio, Andrea, Zhu, Huixiang, E, Yaping, Rose, Jack, Chen, Mary,
Bailey, Craig, and Eric Kandel (1997). “Synapse-Specific, Long-Term Facilitation of
Aplysia Sensory to Motor Synapses: A Function for Local Protein Synthesis in Memory
Storage.” Cell. 91. 927-938.

[9] Miniaci, Maria Concetta, Kim, Joung-Hun, Puthanveettil, Sathyanarayanan, Si,
Kausik, Zhu, Huixiang, Kandel, Eric R., and Craig H. Bailey (2008) “Sustained CPEB-
Dependent Local Protein Synthesis is Required to Stabilize Synaptic Growth for
Persistence of Long-Term Facilitation in Aplysia.”Neuron. 59: 1024-1036.

[10] Papassotiropoulos, Andreas, Wollmer, M. Axel, Aguzzi, Adriano, Hock, Christoph,
Nitsch, Roger M., and Dominique J.-F. de Quervain (2005)  “The prion gene is
associated with human long-term memory.”  Human Molecular Genetics. 14:15: 2241-
2246.

[11] Duman, Ronald S. and Samuel S. Newton (2007) “Epigenetic Marking and Neuronal
Plasticity.” Biological Psychiatry. 62: 1-3.

[12] http://www.cdc.gov/ncidod/dvrd/prions/ accessed 14 April 2009.