Samuel Pleasure, MD, PhD

Narcolepsy, an interesting disorder where patients exhibit excessive sleepiness, cataplexy, and sleep paralysis, has long fascinated neuroscientists. Important genetic studies demonstrated the importance of hypocretin/orexin receptors in pedigrees of narcoleptic dogs and goats, and revealed the central role for these receptors and the neurons that express them in the hypothalamus in the induction of sleep and maintenance of arousal. However, most patients with narcolepsy, unlike the dog models, don’t have mutations in these receptors or a strict Mendelian inheritance of narcolepsy.

Other studies established, long before the involvement of hyocretin receptors was known, that there is an inherited basis for narcolepsy in humans, but one that it is associated with a particular HLA haplotype (

), suggesting a role for autoimmunity in narcolepsy. Interestingly, in patients with narcolepsy who have been autopsied, there is a loss of the cells expressing the hypocretin receptors in the hypothalamus (unlike in the receptor mutation cases where the cells are there but the receptors aren’t). This suggests that some sort of immune process leads to the loss of these cells in susceptible individuals. But what is the inciting event that leads to this autoimmune process?

Several observations over the last few years have indicated that one of the inciting events may be infection with influenza or vaccination with particular influenza vaccines. Finally, 

has pretty much nailed down this hypothesis with some molecular detail.  In this paper the authors demonstrate that there is sequence homology between one of the proteins expressed by the H1N1 influenza virus, and also in the H1N1 vaccine that was used for a time in Europe, and the human hypocretin-2 receptor. This is the specific receptor previously demonstrated to be associated with narcolepsy. The authors also showed that antibodies in the serum of patients who were administered this vaccine and who developed narcolepsy bind to this receptor expressed in cell culture. This finding provides a satisfying end to the associations previously found between an “epidemic” of narcolepsy in Scandinavia after H1N1 vaccination and also with the previously noted seasonal association with narcolepsy in China.

What remains to sort out is whether all cases of narcolepsy (other than cases with Mendelian inheritance) are post-infectious (or post-vaccination) and whether there are viruses other than H1N1 that are associated. Studies in this area are important because it may be possible to design vaccines that will avoid this problem entirely in the future and also to determine who is at risk for this type of unfortunate post-infectious/vaccination complication.

Articles

Observations over the last few years have indicated that an inciting event of narcolepsy may be infection with influenza or vaccination with particular influenza vaccines.

Narcolepsy and the Flu

It has long been clear that patients with extensive head trauma have a greatly increased risk of subsequent neurodegenerative disease, especially tau-based traumatic encephalopathy in the frontotemporal dementia (FTD) spectrum. This is being appreciated as sadly too common in ex-boxers and American football players as well as others with repeated head trauma and concussions. There has also been quite a bit of evidence that some patients with these risks also develop other neurodegenerative diseases, eg, Parkinson disease. However, clear population associations have been difficult to dissect, making it challenging to determine true risks and to begin possible therapeutic interventions early in the process. One of the key questions for physicians in the community is to determine whether less significant head trauma in older patients has a clear association with the onset of Parkinson disease. A new study from Kristine 

 presents data from a major effort to examine this question.

In this study the authors examined a very large cohort of patients (aged 55 and older) that presented to the ER from 2005-2006 either with traumatic brain injury (TBI) or other trauma without brain injury. They then examined the risk of these patients being diagnosed with Parkinson disease by the end of follow up in 2011, and compared the risks of these two cohorts, controlling for a large array of potentially confounding variables. The idea of comparing TBI vs non-TBI trauma patients is interesting and was designed to decrease the likelihood of “reverse causation” – the idea that a patient with early Parkinson disease might be more likely to fall and suffer trauma. Using this approach and controlling for variables, the authors found that TBI patients were 44% more likely to develop Parkinson disease in the ensuing several years.

The implications of this study are broad. Previous work has been a bit unclear about the degree of risk that brain trauma confers for Parkinson disease and this study helps to settle that question. In fact, this study should clear the way to studying particular groups of patients to determine whether those at increased risk of Parkinson disease after trauma have specific characteristics (genetic or environmental) that would allow them to be identified. This would open the way to prevention trials with neurotrophic or neuroprotective agents. The failure of these measures at this point may be due in part to their application only after patients begin to have overt signs of disease. By being able to identify those at increased risk after a defined event, it seems that a new era of preventive trials would have a better chance of being useful.

Our blogger discusses a recent study investigating the likelihood of head trauma in older patients leading to the onset of Parkinson disease and the implications of the findings.

Head Trauma and the Risk of Parkinson Disease

Although developmental myelination is not complete in the cortex until early adulthood in mammals, including mice and men, there is abundant evidence that myelination can continue in the adult brain. There has been much speculation as to the function of this adult myelination in the cortex. One important feature of this adult capacity is the response to myelin injury, which allows robust remyelination. This capacity is inadequate to allow full repair in diseases like multiple sclerosis, and this is an area of very active research in many labs. What, though, is the function of continued adult myelination in the normal undiseased brain? A recent study from Richardson and colleagues has shed remarkable light on this process and how important it appears to be.

The authors have been studying the dynamic production and function of the cells that generate oligodendrocytes in the adult brain (oligodendrocyte progenitor cells – OPCs) for many years. OPCs make up 5% of all cells in the adult cortex and comprise the majority of cells that divide in the adult cortex. Observations over recent years have shown that these cells divide and generate mature oligodendrocytes in the adult mammalian brain in a fairly continuous way with a gradual fall with aging (much as new neuron production falls off).

What is the purpose of this continued production of myelinating oligodendrocytes and what do they do? Some clues come from studies showing that humans who develop very specialized skills in adulthood, e.g., juggling or musical skills, have increased white matter density in cortical areas that serve the enhanced learned motor functions. This information led the authors to directly test the proposition that motor tasks might increase myelination in mice. They engineered mice that have a genetically regulated switch, allowing them to block the production of new oligodendrocytes in a time controlled way in adulthood (for the molecular details see the actual study). Using these mice the authors were able to show very tight control of production of new oligodendrocytes, and were able to shut off this capacity. The mice looked and appeared normal and their brain morphology was normal.

The authors then took these mice, with oligodendrocyte production shut off, trained them in a complicated motor task (a very complex running wheel), and compared the mice that couldn’t make oligodendrocytes to their littermates that could. They found that the running wheel led to a dramatic increase in oligodendrocyte production in the control littermates, but not in the experimental mice. Remarkably, the mice without oligodendrocyte production were also unable to learn the motor task as well as the control mice. To show that this inability to learn was directly due to the production of oligodendrocytes during the motor learning task, the authors compared mice that were trained prior to having oligodendrocyte generation shut off and found that these mice were able to perform the task fine (because they learned the task earlier). This result excludes a role for continued oligodendrocyte production directly in the performance of the motor task, but instead implies a requirement for the new oligodendrocytes to learn the new task. The implications of this work are far reaching. It appears that continued oligodendrocyte production and fine grain myelination in cortical areas are important for learning. Whether this is true for other types of learning in other brain regions will require more experimentation.

These findings are analogous to the longstanding findings that production of new neurons in the dentate gyrus is important for spatial learning (a different type of task-dependent learning) and neuron production in the olfactory bulb for olfactory discrimination in rodents. The work of Richardson and colleagues adds a potential role for the ongoing production of glia as an important and probably collaborating group of cells that can modulate neural circuitry in a complex way. Therapeutically, it is obviously of interest to continue to understand mechanisms that regulate oligodendrocyte production as well as repair of myelin in various diseases.

Motor skill learning requires active central myelination

. Science. 2014 Oct 17;346(6207):318-322.

What is the function of continued adult myelination in the normal undiseased brain? A recent study sheds remarkable light on this process and its importance.

Myelin Formation and Motor Learning in Adulthood

Developmental forms of motor neuron disease usually are disastrous syndromes that frequently lead to early demise of children due to loss of spinal motor neurons and generalized paralysis. The most common genetic spinal muscular atrophy (SMA) form is caused by altering the dosage of 2 genes, 

 The genetics are complicated by the fact that the 

 gene is expressed widely but at low levels so that loss of the 

 gene is compatible with early embryonic development while in other species deletion of 

 is early lethal (because of the lack of the partially compensating effects of 

This has made it challenging to develop adequate models to devise possible treatment strategies. However, a recent study

 has provided an important step forward and revealed potential approaches that may allow amelioration of this disease at some point in the future.

The key to this new model was to develop an approach that allows decreasing the dosage of the SMN protein in motor neurons in a selective way. The authors used adeno-associated viruses to knock down expression of SMN (using shRNA constructs) in early postnatal piglets and found that this leads to decreased protein expression that is in the same range as that seen in human patients with SMA.

This led to a clinical syndrome in the pigs over the following weeks that closely resembled that seen in patients with SMA. The pigs developed evidence of denervation and motor weakness accompanying pathologic loss of motor neurons. Thus, this appears to be a fairly faithful model of SMA in a large animal.

The authors went on to show that if they first infected the piglets with a virus that expresses human SMN (and that doesn’t get knocked down by the shRNAs), the clinical syndrome could be prevented in the piglets infected with shRNA viruses. This demonstrates that indeed it is the knockdown of SMN in the piglets that is responsible for the clinical syndrome.

More importantly and of potential relevance in patients, the investigators went on to rescue the piglets after the onset of clinical symptoms by expressing human SMN after the induction of the syndrome and showing that this is able to at least partially correct the phenotype-clinically, pathologically, and electrophysiologically.

This study raises reasonable hope that a virus-based expression approach could meaningfully correct the molecular phenotype in patients with severe forms of SMA and perhaps convert their phenotype to less severe SMA phenotypes.

Obviously, many questions remain after this fairly early stage study:

• How long can this correction be made to last?

• In this study, the authors found clear clinical improvement that lasted for a month or 2, but would it be permanent or would repeated treatments to boost expression be necessary?

• Will adeno-associated viruses be the best choice for this approach?

Although this virus-mediated molecular treatment approach may seem extreme, I think that in the context of such a dire disease as severe SMA this type of approach may well find significant traction.

1. Duque SI, Arnold WD, Odermatt P, et al. 

A large animal model of spinal muscular atrophy and correction of phenotype. 

Potential approaches to devising strategies that may allow amelioration of this disease are revealed in a recent study. But questions remain.

Samuel Pleasure, MD, PhD

Articles

Narcolepsy and the Flu

Head Trauma and the Risk of Parkinson Disease

Myelin Formation and Motor Learning in Adulthood

A Big Step Forward for Spinal Muscular Atrophy?