Understanding the Flu: Host and Environmental Factors Associated With Susceptibility to Influenza Virus-Induced Disease
Pregnancy increases a woman’s risk of influenza-associated complications because of the physiological changes induced during gestation. During pregnancy, a woman’s heart rate increases, while lung capacity and cell-mediated immunity both decrease during pregnancy (58). Epidemiological data obtained during the 1918 and 1957 influenza pandemics indicated that the risk of influenza-induced mortality was increased in pregnant women (59,60). Additional evidence for increased influenza-related morbidity during pregnancy is found in a study by Neuzil et al wherein influenza infection correlated with increased acute cardiopulmonary hospitalizations in pregnant women (61), while other studies have found that rates of hospitalization are increased in general in influenza-infected versus non-infected pregnant women (62). Case study analysis of pregnant women in the United States infected with the recent 2009 H1N1 influenza viruses also found an increased rate of hospitalization in infected pregnant women over infected members of the general population (63). Additionally, a study investigating the interferon responses to 2009 pandemic H1N1 influenza virus in peripheral blood mononuclear cells (PBMC) found that the PBMC from pregnant women produce significantly less type I and type II interferon in response to viral infection than non-pregnant women (64), suggesting that pregnant women have reduced antiviral immunity to influenza virus infection. Current epidemiological evidence regarding the risks and benefits associated with influenza vaccination during pregnancy are contradictory. While some population studies have found no significant differences between rates of hospitalization and infant morbidity due to influenza illness in vaccinated and unvaccinated pregnant women (65), others find that vaccination against influenza virus provides enhanced protection against influenza-induced disease and hospitalizations during pregnancy (66). Moreover, the effects of influenza viral load on fetal birth defects and psychiatric disorders are highly debated in the literature (67-69).
4. Chronic Respiratory Conditions
Individuals with preexisting medical conditions such as asthma and chronic obstructive pulmonary disease (COPD) are more likely to be hospitalized because of complications from influenza virus infection than infected individuals that do not have underlying medical conditions (70). Research shows that children with asthma and other chronic medical conditions are at enhanced risk for cardiopulmonary hospitalizations, outpatient visits and antibiotic treatment courses because of influenza virus infections (71). Moreover, examination of pediatric patients infected with the novel 2009 H1N1 influenza A virus found that children with asthma were more likely to suffer severe influenza-associated complications such as hospitalization and mortality (72-74). Additionally, respiratory infections, including influenza virus infections, have been associated with longer hospitalizations and impaired lung function in patients with COPD (75). Individuals with asthma and COPD are encouraged to receive influenza virus immunizations to prevent the increased morbidity associated with infection, and this recommendation has been formally endorsed for children with asthma by the Advisory Committee on Immunization Practices since 1964 (76). However, studies examining the role of influenza virus infection and vaccination directly on asthma and COPD exacerbations have conflicted in their results (77,78).
5a. Family functioning
The effect of stress on susceptibility to influenza has been examined using several different types of stressors, primarily using experimental studies in mice. The only epidemiologic study to examine the relationship between stress and influenza virus infection looked at stress related to family functioning (79). The authors found significant increases in the risk of acquiring influenza B virus infection among dysfunctional families, specifically those that were perceived as rigid, chaotic, or enmeshed as compared with balanced families. In addition, family cohesion and adaptability was also associated with an increased risk of influenza virus infection.
5b. Restraint stress
The majority of experimental animal studies on stress and influenza susceptibility have used either restraint stress or exercise-related stress as models for stress exposure. Hermann et al. exposed three different strains of adult mice to continual restraint stress; the mice all exhibited decreased cellular and inflammatory immune responses when compared to control mice. There were no significant effects of stress on influenza-related mortality except in DBA/2 mice, which exhibited a decreased mortality rate in stress-exposed mice (80). Hunzeker et al. (81) examined the effect of restraint stress on natural killer (NK) cell activity and viral replication during mouse-adapted influenza A virus (PR8) infection. The authors found decreased expression of cytokines responsible for NK cell recruitment and function, while there was an increase in expression of anti-viral type I interferon. Age modified the effect of stress on influenza susceptibility in a study by Padgett et al (82), in which aged C57BL/6 mice had increased mortality after infection with mouse-adapted influenza PR8 virus compared to young mice, and restraint stress exacerbated the mortality rate among old mice but not young mice. Avitsur et al (83) found that influenza A virus (PR8)-infected mice subjected to restraint stress had increased expression of pro-inflammatory cytokines, and this effect was more pronounced in females.
5c. Exercise stress
Several studies have been done to examine the effects of moderate and severe exercise stress on susceptibility to influenza. Ilback et al (84) examined the effect of preconditioning exercise or exhaustive exercise on H3N2 influenza- and tularemia-associated mortality in Swiss-Webster inbred mice. Compared to sedentary mice that had not been exposed to any exercise regimen, preconditioned mice had a 25% decrease and exercise exhausted mice had a 33% increase in influenza-related mortality. Neither of the exercise regimens had an effect on tularemia mortality. The effect of severe exercise stress prior to influenza virus infection was examined in a study by Murphy et al (85). Mice were subjected to treadmill running for two hours (or until volitional fatigue) for three consecutive days, immediately followed by intranasal inoculation with influenza virus. Mice exposed to severe exercise stress had significantly increased mortality compared to resting controls, as evidenced by decreased percent surviving and decreased time to death. Exercise-stressed mice also had increased morbidity than sedentary mice, including an increased percentage showing symptoms, a shorter time to developing symptoms, an increased percentage of weight loss and an increased severity of symptoms.
5d. Stress early in life
Not only does stress have immediate effects on health and susceptibility to disease, but stress experienced early in life has been shown to have long term effects on health issues such as chronic diseases and mental health (86-88). Two studies have examined the effects of neonatal stress on susceptibility to influenza A virus infection later in life (89,90). In these studies, neonatal stress was induced by separating mouse pups from their mother for six hours per day for the first 14 days after birth. Mice that had been exposed to neonatal stress and infected with influenza A (PR8) virus as adults had increased viral loads, as well as more rapid kinetics of and increased sickness behavior. Overall, stressed mice had increased expression of pro-inflammatory cytokines following influenza virus infection, although the expression of some cytokines, including IL-1, TNF-α, and IFN-γ, was only elevated in female mice.
5e. Psychosocial stress
A mouse model of social conflict in which male mice are subjected to social reorganization (SRO) was used to evaluate the effects of psychosocial stress on influenza A (PR8) virus infection (91). After a period of initial observation, C57BL/6 inbred mice were classified as either ‘dominant’ or ‘subordinate’, and then dominant mice were switched between cages to create SRO. Mice infected with influenza virus following SRO had significantly higher mortality rates than control mice (56% vs. 11%), and ‘dominant’ mice had the highest mortality rates (86%) compared to ‘subordinate’ mice (48%). Histopathology of lung tissue suggested that the increased mortality was attributed to hyperinflammation in response to the viral infection.
Malnutrition is known to be associated with increased susceptibility to and severity of many infectious diseases (92,93). The effect of malnutrition on the ability to mount an adequate immune response is thought to be the mechanism responsible for this relationship. Some research also suggests that host nutritional status may also affect the virulence of the infectious agent (94-98)(98). Studies examining the effects of nutrition on susceptibility to influenza can be divided into two broad categories; the effects of protein-energy deficiencies and the effects of vitamin and micronutrient deficiencies or supplementation.
6a. Protein-energy deficiencies
6a.i. Protein deprivation
In 1979, Pollett et al. (99) conducted a study to examine the effect of protein-energy malnutrition (PEM), which is a common cause of immune deficiency in malnourished children, on susceptibility to influenza virus infection in mice. Beginning at 18 days after birth, C57BL/6 inbred mice were fed a low-protein diet (4% protein) and were infected intranasally with influenza virus 3-4 weeks later. Mice in the unexposed group were fed a normal diet containing 18-20% protein. Higher mortality rates and significantly decreased anti-influenza virus serum antibody titers were observed among protein-deprived mice. Additionally, viremia was observed in protein-deprived mice but not control mice, and viral loads persisted longer in the lungs of protein-deprived mice. A recent study by Taylor and colleagues further investigated the relevance of PEM to H1N1 influenza A virus susceptibility in mice and found that a low protein diet correlated with enhanced influenza-induced disease following PR8 and 2009 H1N1 virus infections based on its effects on tissue damage, inflammatory cell infiltration, viral persistence and virus-induced mortality (100). Taken together, these studies clearly demonstrate that a low protein diet is a significant susceptibility factor in susceptibility to influenza-induced disease in mice. However, further studies will be required to verify whether PEM has significant effects on influenza-associated illnesses in humans.
6a.ii. Caloric restriction
Dietary caloric restriction (CR) without malnutrition has been shown to extend the life span and preserve immune function of healthy rodents (101,102). Two studies have been done on the effect of caloric restriction on influenza susceptibility. The first study examined the effect of caloric restriction on H1N1 influenza susceptibility in aged mice (103). Aged C57BL/6 inbred mice that had been on a 40% CR diet since the age of 3 months as well as both young and aged mice fed a non-restricted diet were intranasally infected with 3 different doses of PR8 influenza virus. Aged CR mice had higher mortality for all three doses compared to both young and old non-diet restricted mice (100% vs. 40-60%). While non-diet restricted mice began to recover from weight loss at day five post infection, CR mice never recovered from disease-related weight loss and also experienced a 10-fold increase in viral titers in the lung. Based on the results of the first study, the investigators then sought to determine whether CR alone or CR in combination with advanced age was responsible for the inability to fight off influenza A virus (PR8) infection (104). Mice began a 40% CR diet at age 14 weeks and were challenged with influenza virus intranasally at 6 months of age. Again, the CR mice experienced significantly increased mortality and weight loss. In addition, increased viral titers, cellular infiltration and pathology were observed in the lungs of CR mice.
In addition to protein-energy deficiencies, obesity was also proposed to be a possible risk factor for susceptibility to influenza because of the immune dysfunction observed in obese individuals. Prior to the 2009 H1N1 pandemic, the only studies that examined the relationship between obesity and influenza susceptibility were performed in mice. Smith et al (105) evaluated the effect of diet-induced obesity on susceptibility of mice to PR8 influenza A virus infection. C57BL/6 inbred mice were exposed to a high fat, high sucrose diet for 22 weeks and then challenged with influenza virus intranasally. Control mice were fed a low fat, no sucrose diet. Obese mice experienced significantly increased mortality (42% vs. 5.5%) and enhanced lung pathology. Immunologically, obese mice showed reduced expression of antiviral type I interferon and delayed expression of pro-inflammatory cytokines as well as decreased proportions and cytotoxicity of NK cells in the lung. A follow-up study (106) demonstrated that impaired recruitment of dendritic cells to the lungs affects antigen presentation in obese mice infected with influenza A virus.
In the wake of the 2009 H1N1 pandemic, many studies from countries across the globe demonstrated that obesity is a risk factor for hospitalization and death following 2009 H1N1 influenza virus infection, as reviewed by Fezeu et al. (107). Kwong and colleagues (108) examined the relationship between obesity and seasonal influenza in a Canadian cohort. Using the data from population health surveys conducted over 12 influenza seasons, they found that obese individuals were more likely to have been hospitalized for a respiratory illness such as influenza.
6b. Vitamins and micronutrients
Selenium is a trace mineral that is required for the production of many proteins (called selenoproteins), including the antioxidant enzyme glutathione peroxidase (GPx) (109). Dietary selenium is important for both innate and adaptive immune system function (109,110), and it is thought that these effects may be partly mediated by the antioxidant effects of selenium. The effects of selenium deficiency on the susceptibility to disease are a topic of intense investigation, and several experimental studies have been done to explore this relationship in the context of influenza infection. Beck et al (93) investigated the effects of selenium deficiency by exposing mice to a selenium-depleted diet for 4 weeks prior to challenge with a mild strain of H3N2 influenza virus. Selenium-deficient mice had significantly enhanced and longer duration of lung inflammation and increased amounts of cellular lung infiltrates following influenza virus infection. Cytokine profiles indicated a Th1-type immune response in control mice, while selenium-deficient mice exhibited a Th2-type immune response. In a follow-up to this study, the investigators found that viral isolates passaged through selenium-deficient mice caused more severe lung inflammation in infected mice than virus that was passaged through non-selenium deficient mice, indicating a change in virus virulence based on the host nutritional status (111). Interestingly, sequencing of the viral isolates showed that there were very few changes in the hemagglutinin or neuraminidase sequences, but there were many changes in the matrix protein coding sequences, and the changes were conserved among the three different isolates. The most recent study, performed by the before-mentioned group, used the same selenium depletion protocol but they challenged mice with a more virulent strain of influenza A virus, PR8 (98). In this study, selenium-deficient mice had increased survival rates when compared to control mice (50% vs. 100%), although there were no differences in lung pathology or viral titers in the lung between the two groups. Similar to the first study, the cytokine and chemokine profiles indicated a Th1-type immune response in control mice, and Th2-type response in selenium-deficient mice. Unlike experiments using the milder influenza virus strain, there were no mutations in the genomes of virus isolated from the selenium-deficient mice. The investigators attributed the reduced mortality rates to a decreased inflammatory response in the mice.
6b.ii. Antioxidants and vitamin E
One of the effects of influenza virus infection is an increase in the production of reactive oxygen species (ROS). The oxidative stress mediated by ROS is likely to play a role in the damage caused by influenza virus infection (112,113). A decrease in the concentration of antioxidants in the lung and liver has been observed in response to influenza A virus infection with H1N1 PR8 and H3N2 strains, which may be due to the increase in oxidants (113,114). The effects of dietary supplementation with antioxidants on the reduced severity of influenza virus infection have been investigated in two experimental studies. The first study examined the effect of vitamin E, which has known antioxidant effects among other benefits, on the severity of influenza virus infection in young and aged mice (114). Young (4 months) and aged (22 months) mice were fed a diet supplemented with 500 ppm vitamin E for 6 weeks prior to H3N2 influenza virus challenge. Age-matched controls were fed a diet with 30 ppm vitamin E. Vitamin E supplementation reduced lung virus titers on day five post-infection in young mice and on all days in aged mice. Vitamin E supplementation also enhanced NK cell activity in aged mice, and higher serum antibody titers were observed in both young and aged mice. A second study by the same group of researchers examined the effects of supplementation with several other antioxidants and combinations of antioxidants, including vitamin E, glutathione, vitamin E + glutathione, melatonin, and strawberry extract(115). Decreased viral titers in the lung, decreased weight loss, and increased food consumption was seen only in mice supplemented with vitamin E (when compared to the control group); no effects were seen in other groups, indicating that the effect of vitamin E on viral titers and weight loss may not be due solely to its antioxidant properties.
6b.iii. Polyunsaturated fatty acids
Polyunsaturated fatty acids, such as those found in dietary fish oil, are known to have anti-inflammatory properties and have beneficial effects for patients with chronic inflammatory diseases such as rheumatoid arthritis, inflammatory bowel disease, and cardiovascular disease (116-119). The same anti-inflammatory properties can also cause increased susceptibility to and severity of infections that require a strong inflammatory response to clear the infection. The inflammatory response is necessary for clearance of an influenza virus infection, but the inflammation is also responsible for much of the pathology associated with infection. In a study by Byleveld et al (120), BALB/c inbred mice were fed either a diet supplemented with fish oil or a diet supplemented with beef tallow (control group). Mice were fed the experimental diet for 14 days and then challenged with H3N2 influenza virus intranasally. All of the mice recovered from the infection, but the mice on the fish oil diet had increased weight loss and viral titers in the lung, as well as delayed viral clearance. The antibody response was also blunted in fish oil fed mice, with decreased levels of serum IgG and lung IgA and decreased expression of IFN-gamma. Schwerbrock et al (121) conducted a similar study in which C57BL/6 inbred mice were exposed to either a fish oil supplemented diet or a corn oil supplemented diet for 14 days prior to influenza A virus (PR8) challenge. Fish oil fed mice had delayed recovery from infection, increased mortality rates (51% vs. 10%), and a 7-fold increase in viral load. Despite the poorer outcome, the mice had reduced lung pathology, decreased lung neutrophils, NK cells and CD8 cells, and decreased cytokine and chemokine expression in the lung. These results are consistent with an impaired inflammatory response among mice with fish oil supplemented diets, which led to the inability to clear the infection.
6b.iv. Vitamin A
Vitamin A plays an important role in both innate and adaptive immunity and is required for the development and function of many types of lymphoid cells. The role of vitamin A deficiency in influenza susceptibility was examined in a study by Stephensen et al. (122). A diet deprived of vitamin A was fed to pregnant BALB/c dams and subsequent offspring after weaning. At seven weeks of age, the mice were intranasally infected with mouse-adapted H3N2 influenza virus. No differences were observed in survival, viral titers or clearance from the lung. Diminished serum antibody responses and lung IgA titers were seen in vitamin A-deficient mice. After the acute phase of infection, vitamin A-deficient mice had inflammatory lesions in their lungs that progressed to adenomatiod metaplasia, which was not observed in control mice.
The relationship between stressful exercise, quercetin (a flavonoid found in many fruits and vegetables), and influenza susceptibility was examined in a study by Davis et al.(123). ICR inbred mice in quercetin treatment groups were fed quercetin by oral lavage for 7 days prior to viral challenge with influenza A PR8 virus. Mice in severe exercise exposed groups were subjected to 3 consecutive days of treadmill exercise until mice reached volitional fatigue, which was followed by intranasal challenge with influenza virus. Quercetin treatment offset the shortened time to sickness and increased morbidity seen in mice exposed to severe exercise, as well as the severity of symptoms. Quercetin treatment also offset the shortened time to death and mortality rate in mice exposed to severe exercise (74% mortality in severe exercise group; 52% mortality in mice fed quercetin who were exposed to severe exercise). Quercetin treatment alone also reduced influenza-induced mortality (28% mortality in quercetin-fed mice and 50% mortality in placebo non-exercised mice).Continued on Next Page »