Where There’s Smoke (and Stress), Is There Fire?
Risk V: The data are mixed, but two risk factors could offer insights into MS
Cigarettes and excess stress lurk in the shadows of MS, but their role in the disease remains elusive. Exposure to either one seems to be associated with the most common form of MS, relapsing-remitting disease, and scientists are struggling to understand why. (They know less about connections between stress or smoking and other types of MS.) Heavy smokers acquire MS more frequently than do nonsmokers, but the danger in taking a drag now and then is harder to measure, and therefore its risk remains unknown. Studying the effects of stress poses challenges for researchers because it arises from a sense of being threatened—and a situation that makes one individual want to flee might seem harmless to another. Investigators must distinguish stressful events from a person’s reaction to those events. Some individuals take family feuds in stride, for example, whereas others agonize.
Studies that find a dose-dependent effect of cigarettes strengthen the case against smoking. The longer a person has smoked a pack per day, the greater the chance that individual has of developing the disease (Handel et al., 2010). The link between MS and secondhand smoke is shakier, as is a connection between MS and occasional smoking. “We don’t know much about party smokers,” says epidemiologist Trond Riise of the University of Bergen in Norway.
The case against excess stress finds mixed support. Last year, Riise and his colleagues reported no connection between MS onset and a history of stressful events, such as physical or sexual abuse in childhood (Riise et al., 2011). To draw its conclusion, the team combed through responses to questionnaires completed by 94,185 female nurses participating in the second Nurses’ Health Study—a long-term epidemiological study established in the United States in 1989 that aimed to reveal factors contributing to MS and many other diseases. On the other hand, an earlier review that summarized 30 years’ worth of observational studies concluded that, compared to people without MS, those who acquire the disease had experienced a greater number of stressful events (Artemiadis et al., 2011). A drawback to this synthesis is that the types of stress varied tremendously among studies. For example, one study rated family problems whereas another focused on severe threats.
To complicate matters further, most studies measure stress based on self-reports, which are notoriously prone to bias, says Artemios Artemiadis, a physician and researcher at the National & Kapodistrian University of Athens who focuses on stress. A patient who has recently learned that he or she has MS might remember the past more negatively than someone who has received a clean bill of health. “I’ve had patients with MS who say, ‘I got MS because my business wasn’t going well,’ ” Artemiadis says. “I think anxiety is an important potential risk factor, but the problem is, we don’t have a stress-o-meter, a gold standard of measuring stress.” A good way to determine whether stress helps cause MS, Artemiadis says, is to assess the MS rate among a large cohort of people exposed to something unarguably stressful before a portion of them acquire the disease. A 2004 study concluded that parents who unexpectedly lost a child were at higher risk of developing MS than those who had not (Li et al., 2004). The strength of the work lies in the fact that the data do not rely on self-reports. However, the authors note that without information on lifestyle, health outside of MS, and family history—information they did not collect—confounding factors might make their results difficult to interpret. At least one study whose analysis included some of these characteristics showed no support for the notion that stress contributes significantly to MS onset, at least in women (Riise et al., 2011). Prospective gathering of quantitative stress indicators such as cortisol, catecholamines, heart rate, and blood pressure poses significant logistical problems, given that MS is rare in the general population.
Validated mechanisms that explain how smoking or stress lead to MS would further bolster the idea that they increase risk for the disease. Hypotheses that relate smoking to inflammation abound, mainly from studies on other illnesses. For example, nicotine can weaken the blood-brain barrier, according to some reports. “If the barrier allows inflammatory substances in the blood to leak through to the brain, [nicotine] might help trigger MS,” Riise says. The drug might require a partner, however: According to one study, use of tobacco snuff does not seem to pose a risk of MS (Hedström et al., 2009). Another potential mode of action involves the respiratory infections that result from smoking. These illnesses promote inflammation in peripheral tissues; if they have similar effects in the central nervous system, they might foster MS (Handel et al., 2010). A third possibility revolves around nitric oxide, a component of cigarette smoke, which might promote demyelination (Smith et al., 1999). Finally, smoking can cause chemical groups to stick to DNA and thus change gene activity in ways that lead to cancer. MS researchers suggest that a similar process could be involved in MS (Burrell et al., 2011, and “Nature, Nurture, and What’s in Between”).
To examine how chronic and acute stress might affect MS, researchers have subjected mice with a disease that resembles MS (autoimmune encephalomyelitis, or EAE) to various insults (see “Animal Arsenal”). Stressors include water deprivation, immobilization and isolation, crowding, and electric shocks. Delivering one of these treatments for more than an hour per day for more than five consecutive days constitutes “chronic” stress, according to the researchers who use this model. “Acute” stress typically lasts for less than an hour per day and for less than 5 days. In general, chronic stress alleviates EAE in mice, whereas acute stress exacerbates it (Heesen et al., 2007).
No one knows why acute and chronic stress exert opposite effects on EAE. One hypothesis involves mast cells, which play a role in immunity, allergies, wound healing, and other host defense mechanisms. Acute—and not chronic—stress appears to activate these cells. And experiments in pigeons as well as rodents suggest that mast cells increase the permeability of the blood-brain barrier, whose breakdown has been implicated in MS (Esposito et al., 2001, and “Gate Crashers”). In contrast, chronic stress might protect against EAE by desensitizing mice; rodents exposed to chronic stress had an abnormally subdued response to an injection of the inflammatory cytokine IL-1β (Stefferl et al., 2001).
The different mechanisms underlying acute and chronic stress in rodents with EAE might not exactly match what occurs in humans because the states aren’t as clearly defined in real life as they are in the lab. “We can say a flood is an acute stressor,” Artemiadis explains, “and the loss of your home because of the flood, a chronic stress.” Whether these translate into lab-induced zaps and prods over days and hours remains elusive.
Researchers have hypothesized that the part of the neuroendocrine system that controls reactions to stress, the HPA (hypothalamic-pituitary-adrenal) axis, might overreact in people who acquire MS. That’s an attractive hypothesis because the HPA axis can influence the immune system, and its malfunctions could lead to abnormal inflammation. At the moment, however, little—if any—evidence suggests that HPA problems lead to MS in humans, Artemiadis says. A 2007 report encourages investigators to approach this hypothesis by comparing the gene-expression profiles in patients who’ve recently been diagnosed with MS to those of people without the disease (Heesen et al., 2007). Anomalous activity of HPA-related genes might suggest that the alteration occurs early in the disease, and possibly before it begins.
Whether the HPA axis boosts the risk of MS remains murky, but once a person has the disease, quantities of HPA-related compounds are elevated (Gold et al., 2005). Furthermore, severe disease tends to correlate with unusually strong HPA responses. This perturbation appears to be associated with disease activity, as it diminishes in periods of remission (Gold et al., 2005). Observations of the HPA axis in MS patients might reveal little about whether the network contributes to susceptibility, Artemiadis cautions, because the immune system acts differently once a person has the disease. Furthermore, people who are in the middle of a relapse are likely to feel more stress than they otherwise would, he adds.
People already have plenty of reasons to refrain from smoking and reduce excess stress, but evidence is mounting that MS offers another. The mechanisms by which these factors act might lend insights into the disease. Perhaps, for example, information about how the nitric oxide in cigarettes triggers demyelination will point toward molecules or pathways that could help researchers decipher why one smoker acquires MS and another gets off scot-free.
Although Artemiadis says “we really have no clue” when it comes to the route stress might take to causing MS, he also urges investigators not to drop stress as a suspect. Both stress and smoking have proved difficult cases to solve, but ones worthy of pursuit.
Key open questions
- Does growing up around smokers increase a person’s risk of MS?
- Do biomarkers of stress, such as cortisol levels, increase a person’s risk of MS?
- Does smoking contribute to MS development due to excess inflammation? If so, how might such inflammation be treated to prevent MS?
- Do acute and chronic stress alter a person’s risk of MS in divergent ways?
Previous article in series: "Nature, Nurture, and What's in Between"
Next article in series: "Moving Toward Convictions"