Nostalgia: the similarities between immunological and neurological memory

This article introduces a series of reviews covering Neuroimmunology appearing in Volume 248 of Immunological Reviews

The components of this volume of Immunological Reviews are devoted to the subject of neuroimmunology. In this preface, I reflect on the concept of memory, a function that is critical for both the immune system and the nervous system. The nervous and immune systems are tuned to react to external stimuli, and both are imbued with the profound capacity to recall earlier events, what we term as memory. It is memory for past antigens that allows the immune system to respond effectively to microbial challenges through neutralization of most microbes before they can cause harm. Diseases of memory are associated with some profound clinical pathology in both the immune system and nervous system. Some of these syndromes have conceptual similarities, so here I compare diseases involving mistaken identity, and those involving retrograde amnesia. Such disorders are encountered in human immune biology and neurology.

Immune memory is triggered by not only an encounter with a portion of an antigen that binds to an affinity matured antibody or to a T-cell receptor but also there are additional more subtle clues that orchestrate a complex recall or memory response. When recalling a microbial memory, the immune system has a complex set of sensors that can discern subtle structural motifs in the microbial world, so-called pathogen-associated molecular patterns (PAMPs) (1, 2). These molecular cues trigger memory as they signify a previous encounter, eliciting innate immunity and augmenting the adaptive immune responses.

Likewise, neurological memory enables us to recall and then to respond effectively to many complex stimuli. With some similarity to immune memory, neurological memory can also be triggered by subtle molecular cues, often an olfactory stimulus. Rather famously in literature, in Proust’s Remembrance of Things Past (3), a sweet, spongy cake, known as a madeleine, dipped in tea triggered a complex cascade of childhood memories. Thus, both immune memory and neurological memory involve a transformation of the initial molecular stimulus, into a complex response. As Proust himself wrote in his famous work, ‘Remembrance of things past is not necessarily the remembrance of things as they were’ (3). This transformation of a subtle cue into a complex response determined by past events is a common feature in both immune memory and neurological memory, as described elsewhere.

In the immune system, we retain, sometimes with the help of secondary or ‘booster’ immunizations, life-long immunity to many devastating and lethal microbial infections. We may ultimately lose our immunity to chicken pox, varicella virus, and this puts us at risk for shingles. A ‘booster’ vaccine against varicella taken after age 60 can largely prevent shingles. With age, we also lose memories, and the various dementias are the most common diseases associated with this reality, with Alzheimer’s the most well known.

In our brains, we retain precious memories of events and emotional states in the past. Neurological memory sculpts our personalities, provides us with the ability to perform complex pieces of music, ride a bike, and carry out arithmetic calculations. Immune memory efficiently leads to neutralizing responses to deadly microbes after an initial encounter, even though certain features on that microbe may have changed or mutated. Fortunately from a controlled immunization in childhood, we have the capacity to neutralize a wide variety of microbial threats throughout our lives.

What do immune memory and neurological memory actually have in common, other than the concept given to us in language, with the shared word ‘memory’? Is memory at the various types of neurological synapses at all similar to memory in the immune system? Are changes at chemical synapses in the brain at all similar to the adaptive gene rearrangements inherent in the diverse immunoglobulin and the T-cell receptors found on memory T and B cells? The answer, at least as we know in the early part of the 21st century, is that neurological and immunological memory have very little in common on a molecular level. There is some information on at least some molecules that play a role in neurological and immunological memory. A wonderful example is brain-derived neurotrophic factor that aids learning and memory and is also produced by T cells that help quell brain immunity (4). There are undoubtedly many other molecules to be discovered that play central roles in the two processes given the same name, memory-immunological and neurological. Beyond shared molecules, it is perhaps instructive to consider the many functional similarities when one looks at pathological disorders of these two parallel systems: immune and neurological memory.

Despite a lack of detailed information that might suggest that the two types of memories share any common molecular mechanisms, there are some intriguing similarities between disorders of memory in the immune and nervous system. As the gifted writer and eminent neurologist Oliver Sacks has written, ‘Constantly, my patients drive me to question, and constantly my questions drive me to patients’ (5). I compare here certain neurological disorders of memory with some conditions where immunological memory has failed. I compose this preface with a tip of my metaphorical ‘hat’ towards Professor Sacks. These clinical anecdotes about pathological conditions of memory, both the neurological and immunological forms of memory, serve to illuminate the many rich intersections of these concepts in the fields of neuroscience and immunobiology.

Molecular mimicry and mistaken identity in neurology and immunology

A classic neurologic counterpart of mistaken identity is exemplified by the syndrome of prosopagnosia, where an individual can no longer identify common objects like a face, but instead is able to recognize a familiar face by alternative cues, or features that are associated with the face, like the hat in Oliver Sacks’ classic, The Man Who Mistook his Wife for a Hat (5; Fig. 1). This fascinating neurological phenomenon shares features with autoimmune diseases manifest when there is a failure of recognition of self. Prosopagnosia can be seen after a stroke or tumor and often involves damage in the fusiform gyrus in the brain. The fusiform gyrus connects the temporal and occipital lobes with the hippocampus, a structure so vital to memory. Lesions here are associated with this rare and shocking behavioral deficit, where mistaken identity is so profound that a husband might mistake his wife of half a century for a hat! Failure to recognize self and non-self leads to complex autoimmune diseases, as described in many articles in this volume.

Many immunologists have elaborated how the immune system can distinguish self from non-self. Sir McFarlane Burnet in his 1960 Nobel Lecture wrote, ‘How can an immunized animal recognize the difference between an injected material like insulin or serum albumin from another species and its own corresponding substance?’ (6) He continued, ‘If a cell or clone is limited to one or two patterns, then it is practical to postulate that any carrying either one or two self-reactive patterns is eliminated, leaving only clones carrying patterns corresponding to configurations not present in the body. This is the form taken by the clonal selection theory and provided two patterns is adopted as the usual number for a diploid somatic cell, it provides a reasonable interpretation of the facts’ (6).

The concept of ‘Clonal Selection’ has dominated how we have tried to understand self versus non-self recognition over the past half century. Yet, we now are fully aware that a vast number of structures have shared chemical structures that are identical between the microbial world and our own self-constituents. A competing idea, ‘Moelcular Mimicry’, one not mutually exclusive with Clonal Selection, accounts for how autoimmune diseases develop. The concept of molecular mimicry accounts for how the immune system can mistake the identity between the shared structures of what we call ‘self’ and those in the realm of ‘non-self’, mostly constituents of the microbial world (Fig. 1). Thus, chemical structures on β-hemolytic streptococcus are shared with chemical features on the basal ganglia of the brain. Infection with streptococcus can lead to the choreoathetosis, termed Sydenham’s Chorea or St. Vitus’s dance, seen a few weeks after such a streptococcal infection. Indeed antibodies to streptococcus are cross-reactive with antigens in the basal ganglia, and this cross-reaction may be at the center of the pathogenesis of Sydenham’s Chorea (7, 8). Similar structures shared between Campylobacter jejuni and gangliosides cause an autoimmune condition called Guillain–Barre syndrome, resulting in autoimmune inflammation of the peripheral nerves. Anti-GD1 and GD3 antibodies targeting regions with similar atomic contours found on Campylobacter jejuni and also present in the peripheral nerve axon are the basis for this disease (9, 10).

The autoimmune immune response to carbohydrates in Guillain–Barre and in poststreptococcal disorders like Sydenham’s Chorea allows us to make a bridge to neurological memories. Immune recognition of bacterial carbohydrates is exploited in the design of conjugate vaccines where bacterial polysaccharides are coupled to protein carriers (11). These vaccines have proven highly effective in providing lifelong immune memory to pneumococcus and influenza. This recognition of the carbohydrate component of an antigen is reminiscent of how neurological memory can be triggered by an encounter with that ‘sweet Madeleine dipped in tea’ made famous in Proust’s Remembrance of Things Past (3).

Another example of mistaken identity occurs in the memory loss associated with antibodies to NMDA receptors. Such antibodies are expressed in ovarian teratomas and provide the basis for one of the fascinating and rare autoimmune paraneoplastic syndromes associated with psychiatric disorders and memory loss. Dodel’s review in this volume (12) describes such antibodies. Anti-NMDA receptor antibodies are often found in systemic lupus erythematosus and may underlie some of the psychiatric and memory disturbances seen in lupus. This is described in the chapter by Diamond (13).

Loss of immunological memory and transient global amnesia

Transient global amnesia and Korsakoff’s psychosis are two examples where loss of memory occurs in a retrograde fashion. This means that memory for recent events is lost and memory of the more distant past is retained to a much greater degree than recall for more recent events. Transient global amnesia often follows ischemia in the basilar circulation of the brain. The amygdala and hippocampus are often profoundly affected. Korsakoff’s psychosis results from thiamine deficiency and is sometimes accompanied with other neurological manifestations, including paralysis of eye muscles, loss of balance and confusion, so-called Wernicke–Korsakoff syndrome.

With some similarity memory can be lost with various disease states. In human immunodeficiency virus (HIV) infection, there is profound loss of memory cells with ensuing loss of the ability to combat infections for which we have previously been immune. The massive loss of CD4⁺ T memory cells in HIV and simian immunodeficiency virus (SIV) is a consequence of the destruction of memory T cells from the virus (14). Like the locus for memory loss in transient global amnesia, the locus for memory loss in SIV resides in CCR5 CD4⁺ T cells (14). One of the remarkable aspects of memory is that in both the immune system and the nervous system, memory may truly reside in certain cell populations, even in specific locations. In the immune system, where transplantation is common in various diseases, loss of immune function certainly occurs following bone marrow transplantation, for example. People with food allergies may lose them after a transplant. Gain of immune function also occurs with the development of a particular allergic state, like food allergy, from a donor of the transplant. Thus, the gain of function and loss of function for immune memory is a very real phenomenon (15). Immune memory and even ‘horror autotoxicus’ can occur when the immune system recognizes a portion of a self-molecules that does not normally reside in the thymus (16). We have not yet seen a parallel experiment performed with gain or loss of neurological memory, but I would predict that this may be coming soon with the era of stem cell transplants to the nervous system.

This volume contains a wealth of chapters dealing with the milieu in which immunity to the nervous system flourishes. The chapters by Axtell and Raman (17), Goverman (18), Kuchroo (19), and Mayo and Weiner (20) describe the molecular basis for autoimmunity to neurological tissues. Rivest (21) shares how immune responses are modulated by the neuroendocrine axis. Other articles by Ransohoff (22) and Kawakami (23) describe the remarkable mechanisms that allow immune cells to home to the brain. Fugger (24), Hauser et al. (25), and Hafler (26) describe the intricate genetic regulation of brain immune interactions in pathological conditions like multiple sclerosis. Overall, the field of neuroimmunology is well covered in this volume. It will be exciting to see how the fields of neuroscience and immunology continue to illuminate those common processes like memory that are vital in each system.

Although in 2012 there is little known about common mechanisms and molecules that underlie the process of memory in the immune system and nervous system, we may someday realize there is much more than merely conceptual connections between the physiological processes that govern immune and neurological memory. Much more needs to be learned. There are concepts like the synapse that have been studied in the neurosciences and immunology. Despite major differences between neurological and immune synapses, over the past decade there has been an elucidation of certain shared molecules and common cell biological features (27, 28). Like our understanding of immune and neurological synapses, we may begin to understand shared features between immune and neurological memory. After all, there are actual synapses between the nerves and immune organs, and there is even evidence that memory processes are at work at these synapses that mark the interface between these disparate physiological systems (29).