Introduction

Asthma is a disorder characterized by (1) recurring symptoms of wheezing, chest tightness, and/or shortness of breath, (2) pathologic evidence of airway inflammation, and (3) physiologic changes in bronchial hyperresponsiveness and airflow obstruction. It is the most common chronic illness in childhood and a major cause of morbidity in adults, affecting 4–17 % of children and 7–10 % of adults in the United States and, overall, nearly 30 million Americans (Yun et al. 2012).

Over several decades, asthma has evolved from being recognized as a single disease to include a diverse group of phenotypes with dissimilar natural histories, pathophysiologies, and responses to treatment (Gauthier et al. 2015; Haldar et al. 2008). It is assumed that there are distinctive molecular pathways underlying these disparate conditions, but they remain to be defined.

To provide simplicity, one can extrapolate Occam’s razor to asthma and propose that there is one cause underlying the numerous phenotypes of this disease and that the responsible molecular pathway is a deficiency of iron, absolute and/or functional, in the lung tissues. In this review, iron homeostasis in asthmatic conditions is delineated. Associations between iron homeostasis and (1) recognized determinants of asthma (Table 1), (2) co-morbidity of asthma, and (3) therapeutic interventions in asthma are demonstrated. Finally, possible relationships between the pathophysiologic events producing asthma and iron homeostasis are described.

Table 1 Determinants of asthma
Full size table

Iron homeostasis

As a result of a favorable oxidation–reduction potential and its abundance in nature, iron performs a wide range of biological functions. This metal is therefore an essential micronutrient required for virtually every aspect of normal cell function, and living systems must have iron to survive. However, water-soluble ferrous ion (Fe2+) was effectively removed from the Earth’s crust after its reaction with atmospheric oxygen produced by photosynthesis. Ferric ion (Fe3+) remained but at concentrations inadequate to meet the requirements for life; the concentrations of Fe3+ soluble in water at physiologic pH values is about 10−18 M while that required for life approaches 10−6 M. To support life, greater quantities of metal were needed. This challenge to living systems was achieved by (1) the chemical reduction of Fe3+ to Fe2+ (i.e. ferrireduction) with its subsequent import and utilization and (2) the complexation of Fe3+ with chelators coupled to receptors for uptake of the complex, which rendered the metal bioavailable. In addition to solubility-limited availability, the metal presents a potential for oxidative stress related to iron-catalyzed generation of radicals. Such reactivity mandates that iron availability be tightly regulated and, as a result, life normally exists at the interface of iron-deficiency and iron-sufficiency.

Cellular iron homeostasis is maintained by a coordinated expression of proteins involved in the import, export, storage, and utilization of this metal (Hentze et al. 2004). Post-transcriptional control mediated by iron-regulatory proteins (IRP) is essential (Wallander et al. 2006). IRP1, the cytosolic counterpart of mitochondrial aconitase, is a bifunctional protein that, through [4Fe–4S] cluster assembly/disassembly, shifts from the aconitase to the IRP1 form in response to decreased intracellular iron concentrations (Kennedy et al. 1992). Accordingly, iron levels regulate the RNA-binding capacity of IRP1. With diminished availability of the metal, IRP1 binds to cis-acting mRNA moieties termed iron-responsive elements to impact the expression of proteins involved in import, export, storage, and utilization. Mechanisms include stabilization of the mRNA of the divalent metal transport 1 importer (DMT1) and transferrin receptor 1 to promote translation and increase their expression while also suppressing the synthesis of the storage protein ferritin and the exporter ferroportin-1 (Mims and Prchal 2005).

The serum ferritin level is currently accepted as the best indicator of stored and total body iron (Kruger et al. 2012). It is assumed that there is an association between the total body iron and that concentration of metal available to cells, tissues, and living systems. Normally, concentrations of iron inadequate to support the function of cells, tissues, and living systems are reflected by levels of serum ferritin, serum iron, transferrin, transferrin receptor, and other proteins involved in iron homeostasis. Insufficient iron can result from both an absolute and a functional deficiency of the metal (Lopez et al. 2015). In an absolute deficiency, iron concentrations are lacking as a result of either decreased intake or an elevated loss; provision of the metal can improve such a condition. A functional deficiency frequently develops from the introduction of an inappropriate chelator that binds host iron, decreasing concentrations available to the cell (e.g. a particle) (Ghio et al. 2013). In the presence of such an inappropriate chelator, a functional deficiency may exist despite total iron being either normal or even elevated.

The host response to a deficiency in iron (both absolute and functional) includes an increased expression of importers in an attempt by affected cells to reverse the loss of requisite metal. Concomitantly, a disruption of iron homeostasis with metal deficiency in a living system necessitates prioritization of the available iron with potential biochemical and physiological consequences (Guiang et al. 1997). With deficiency, the distribution of the metal can favor specific tissues with red blood cells possibly assuming priority over others. Mechanisms by which prioritization of iron delivery to tissues is regulated have not been fully delineated. Iron concentrations in the non-heme tissues (e.g. skeletal muscle and heart) can subsequently appear to be sacrificed at the expense of red cell iron incorporation (McKay et al. 1983; Zamora et al. 2016). Such a propensity of red blood cells to accumulate iron in preference to other tissues may reflect that 80–90 % of the body’s TfRs in the human normally reside on erythroid cells (Beguin 1992). However, blood hematocrit can also decrease before reductions in brain iron suggesting that, at least in older animals, brain iron can be spared at the expense of the erythron (Dallman et al. 1975). Accordingly, while anemia can reflect iron deficiency in all tissues of a living system, metal concentrations in peripheral tissues can be diminished prior to any loss in red blood cells. It can be assumed that if there is anemia present, the risk for iron deficiency in tissues other than red blood cells may be increased.

Iron concentrations in serum samples from asthmatics

Measurement of indices of iron homeostasis in serum is the most direct manner of demonstrating metal deficiency in asthma. Serum iron concentrations demonstrated no differences in comparisons between healthy subjects and asthmatic patients (Vural et al. 2000). Plasma iron levels were also reported to be elevated in asthmatics (Ekmekci et al. 2004; Kocyigit et al. 2004). More recently, using data from the 2007 to 2010 National Health and Nutrition Examination Survey (NHANES) including serum ferritin, serum soluble TfR, and TfR/log ferritin, and iron stores among women (aged 20–49 years) in the United States, indices of iron homeostasis were associated with asthma (Brigham et al. 2015). Specifically, serum ferritin greater than 76 ng/mL was associated with decreased odds of lifetime asthma, of current asthma and of asthma attack/episode in the past year. In addition, increases in tissue iron need and decreases in body iron, (represented by lower sTfR and lower sTfR/log ferritin, respectively) correlated with decreases in forced expiratory volume in one second, supporting an influence of tissue and body iron on lung function.

Demographic determinants of asthma

Age and gender

The incidence and prevalence of asthma demonstrate an inverse relationship with age; both are greatest in young children and less in late adolescence and adulthood (Dodge and Burrows 1980; Singleton et al. 2006). More than 50 % of patients with childhood asthma enter clinical remission by puberty. Regarding the relationship between total body iron and age, the healthy newborn has relatively high concentrations of stored iron. During the first few months of life, when milk (which is low in iron) is the primary source of nutrition, the stockpiled metal supports growth of the newborn. Subsequently, iron stores are mobilized after birth and decrease with duration of nursing. With the introduction of iron-rich solid foods, the concentration of total metal concentration increases, and an invariable accumulation of iron continues through life (Fig. 1). The observed incidence and prevalence of asthma demonstrate an inverse relationship with total iron concentrations, supporting the relationship between asthma and iron deficiency. The remission of asthma with aging can reflect augmented iron availability as stored concentrations (e.g. serum ferritin) increase with age, and deficiency decreases. Comparable to measures of asthma incidence and prevalence, bronchial hyperresponsiveness decreases with age again reflecting elevated metal concentrations among older individuals (Burrows et al. 1995; Le Souef et al. 1995).

Fig. 1
figure 1

Serum ferritin concentrations increase with age in both males and females. This association of iron with age and gender can support an inverse relationship of the metal with asthma

Full size image

Similar to age, gender is a major determinant of both asthma and iron status. Males can have a higher prevalence of asthma in infancy and early childhood (Almqvist et al. 2008; Schatz and Camargo 2003); asthma is diagnosed 1.5 times more frequently in young boys than in young girls (Dodge and Burrows 1980). Regarding hospital admissions for asthma, males are admitted for asthma nearly twice as often as age-identical females in the 0- to 5-year-old and 6- to 10-year-old age groups (Skobeloff et al. 1992). With the onset of puberty, the prevalence of asthma becomes higher in females and will remain higher throughout the remainder of life (Dodge and Burrows 1980; Melgert et al. 2007; Postma 2007). While the prevalence of asthma in boys (7.7 %) was similar to girls (7.4 %) at 11.1 years of age, the prevalence was 6.2 and 4.3 % in females and males respectively at 16.3 years of age (Vink et al. 2010). Similar to prevalence, the rates of emergency visits, hospitalizations, and health care utilization for asthma are higher in adult females than males (Debley et al. 2004; Lee et al. 2006). For hospital admissions, the female-to-male ratio is nearly 3:1 among those between 20 and 50 years of age, while females over 50 years of age are admitted for asthma at a rate of about 2.5:1 when compared with their age-equivalent male counterparts (Skobeloff et al. 1992). Similarly, there is evidence suggesting that airway hyperreactivity is more prevalent in women than in men in the general population (Leynaert et al. 1997; Manfreda et al. 2001).

These disparities in asthma measures between the genders correspond to the observation that serum ferritin levels in males up to 11 years of age are lower than in females, reflecting less total iron in males (Emond et al. 1996; Milman and Kaas Ibsen 1984) (Fig. 1). Following puberty and throughout adulthood, females will have lower iron concentrations (Fig. 1). These lower concentrations of stored iron among females correlate with the disparities between the genders in asthma incidence/prevalence and measures of severity, reinforcing the proposed relationship of asthma with iron homeostasis.

Ethnicity

Disparities in the incidence and prevalence, indices of morbidity and mortality, and treatment response of asthma have been identified among ethnic groups (Drake et al. 2008). Asthmatics with African-American ancestry have particularly severe disease (Oliveti et al. 1996; Rumpel et al. 2012). It has been proposed that these differences reflect dissimilarities in smoking, exposures to environmental tobacco smoke and air pollution, compliance with and response to therapies, urbanization, psychosocial stress, economic and educational advantages, and availability of health insurance. However, corresponding to an increased risk for asthma, African-American ethnicity has been associated with alterations in iron homeostasis (Li et al. 2015; McLaren et al. 2012). Diminished body and serum iron levels have been observed among African-Americans, supporting the inverse association between asthma and metal availability (Chambers et al. 2006; Pfeiffer et al. 2013).

Physiologic determinants of asthma

Prenatal and perinatal factors

Asthma diagnoses are increased among newborns with low birth weight and those born preterm (Been et al. 2014; Mu et al. 2014). Size at birth is associated with the occurrence of asthma not only in the neonate but also in later life (Barker et al. 2013; Gregory et al. 1999; Lu et al. 2012). A gender difference has been observed in the effect of birthweight on the prevalence of asthma in childhood and even among adults, with females showing a disproportionate effect (Brostrom et al. 2013). Similar to birthweight, preterm delivery predicts asthma with an approximately fourfold increase in the incidence for individuals born prematurely (Beaudoin et al. 2013; Landry et al. 2011). A negative dose–response association has been noted between gestational age and the purchase of prescription asthma medication in infancy and childhood (Damgaard et al. 2015). In another study, children born preterm were observed to have a higher risk of asthma compared to term children at ages 0–5 and 6–9 years (He et al. 2015). Preterm babies also have a higher prevalence of abnormally low pulmonary function and bronchial hyperresponsiveness at 21 or 22 years of age (Islam et al. 2015; Landry et al. 2016; Vollsaeter et al. 2015).

Increased risks for asthma among those born preterm and with low birth weight correspond to changes in iron homeostasis and deficiency of the metal. The requirement for iron during pregnancy exceeds that in the non-pregnant state and increases through gravidity approximating 0.8, 4–5, and >6 mg in the first, second, and third trimesters respectively (Bothwell 2000; Cetin et al. 2011). The additional metal is used for (1) expansion of the maternal erythrocyte mass, (2) replacement of maternal blood lost at delivery, (3) creation of placenta, and (4) fetal growth. In later pregnancy, the amount of iron that can be absorbed from the diet is frequently less than that needed. Therefore, a woman must enter pregnancy with iron stores in excess of 300 mg if requirements of the fetus are to be met. A minority of women may have this reserve (Baker et al. 2010). Reflecting this lack of reserve, maternal serum ferritin concentration decreases during pregnancy (Puolakka 1980); in one study, serum ferritin decreased from 31.9 ng/mL in the first trimester to 10.3 ng/mL in the third trimester (Shields et al. 2011). Subsequently, indices of iron homeostasis in the pregnant woman (e.g. hemoglobin levels, transferrin saturation, and serum ferritin concentrations) reflect reduced stores. Non-hereditary, non-hemolytic anemia is the most common complication of pregnancy with rates around 9.3 % (Bruce et al. 2008). However, among low-income minority populations in the United States, iron-deficiency anemia can affect approximately 27 % of pregnancies in the third trimester (American College of Obstetricians and Gynecologists 2008; Scholl 2005).

The fetus normally stockpiles relatively high concentrations of iron acquired from the mother; this transfer is especially effective during the last trimester of the pregnancy (Fig. 2). As a result of the requirement for iron transfer, variability in maternal stores will influence that of the newborn infant, the child, and perhaps the adult (Kaneshige 1981; Puolakka et al. 1980). Maternal serum iron and fetal iron levels positively correlate suggesting that the fetus can extract iron in amounts proportional to the levels available in the mother (Hokama 1999; Singla et al. 1979). Fetal iron reserves are dependent on maternal iron reserves and, in turn, the mother’s iron intake (Kelly et al. 1978; Milman et al. 1991; Takala et al. 2010). This correlation between indices of maternal and newborn iron status reveals a considerable vulnerability of the fetus to iron deficiency during intrauterine life. With limited stores, a lack of a sufficient dietary intake of the metal, and an inability to regulate absorption by the gastrointestinal tract, iron deficiency can result (Collard 2009). Subsequently, maternal anemia will be a major contributor to both iron deficiency and anemia among infants (Liu et al. 2015).

Fig. 2
figure 2

Serum ferritin levels in full term and premature children. Diminished levels of available iron in the premature infant increase the risk for asthma

Full size image

Along with anemia, low birth weight, and preterm birth, asthma can be included in the spectrum of disease associated with maternal-fetal iron deficiency (Fig. 3). A birth cohort study found an inverse association between umbilical cord iron levels and later onset wheeze (Shaheen et al. 2004). Maternal anemia during pregnancy, reflecting diminished iron availability in both the mother and the fetus, was associated with wheeze and asthma in children (Triche et al. 2011). Reduced maternal iron status during pregnancy (even as early as in the first trimester) was directly related to increased wheeze and atopic sensitization and decreased lung function in children up to 10 years of age (Nwaru et al. 2014).

Fig. 3
figure 3

The overlapping relationships of maternal-fetal iron deficiency with anemia, asthma, low birth weight, and prematurity. Positive correlations between anemia, asthma, low birth weight, and prematurity support a diminished availability of iron as an underlying cause of each

Full size image

Regarding anemia in the neonate, term newborns demonstrate a decline in hemoglobin after birth reflecting inadequate quantities of iron (Ferri et al. 2014). In infants, iron deficiency anemia most frequently occurs 3 months after iron stores are depleted, which is typically at 6–9 months of age (Shakur et al. 2010).

Iron is an absolute requirement for cell proliferation, and its deficiency leads to apoptosis/cell death (Hileti et al. 1995; Yu et al. 2007). Comparable to cultured cells, iron stimulates growth in living systems including humans. Beginning pregnancy with non-depleted iron stores increases infant birthweight while maternal anemia diminishes birthweight (Levy et al. 2005; Ribot et al. 2012; Singla et al. 1997). Mild to moderate iron deficiency in the mother, even without anemia, contributes to lower iron reserves in the fetus and to lower birth weight (Bhargava et al. 1991). Birth weight is also associated with decreased levels of iron as indicated not only by maternal hemoglobin and hematocrit but also by ferritin (Rasmussen 2001). Newborns of non-anemic mothers supplemented with iron demonstrate greater birthweights (and elevated serum ferritin concentrations) (Aranda et al. 2011; Wise 2013). Small for gestational age infants have lesser total iron stores (blood ferritin) as compared to appropriate for age infants at birth (Mukhopadhyay et al. 2011). Birth weight can significantly correlate with the hemoglobin concentration at 6 months of age (Gunnarsson et al. 2004; Shakur et al. 2010). Greater doses of iron supplementation impact higher birth weights in newborns (Ribot et al. 2013). The associations of low birth weight with asthma likely reflect a relationship of both with concentrations of available iron.

The understanding of the biological mechanisms that control the timing of delivery is incomplete. Thus the relationship of preterm delivery with iron homeostasis is poorly understood. In both the mother and the fetus, decreased indices of iron availability correspond with preterm delivery. Maternal anemia is associated with a shorter gestation period and it is plausible that the frequently observed association between maternal hemoglobin/hematocrit with birth weight are caused by a shorter gestation (Allen 2001; Gunnarsson et al. 2004; Shakur et al. 2010). Ferritin concentrations are positively associated with gestation duration with iron supplementation effecting longer pregnancies (Hemminki and Rimpela 1991; Hemminki and Starfield 1978). Plasma iron concentrations measured in the umbilical cord blood of babies rise with gestational age (Scott et al. 1975). Similarly, fetal iron stores reflected by umbilical cord serum ferritin can almost triple from 63 to 171 µg/L between 23 and 41 weeks gestation (Siddappa et al. 2007) (Fig. 2). Premature birth is similarly associated with iron deficiency in the first 6 months of life, and iron supplements can reduce this risk (Gorten and Cross 1964; Lundstrom et al. 1977; Uijterschout et al. 2015) (Fig. 2). With higher doses of iron supplementation, there is less iron deficiency anemia and fewer preterm deliveries and low birth weights in newborns (Ribot et al. 2013). Reflecting iron stores, mean hemoglobin in preterm infants was lower than mean hemoglobin in full-term infants with values of 13.4, 14.5, 15.0, and 16.8 g/dL at 26–30, 28, 32, and 37–40 weeks gestational age, respectively. Similarly, premature children demonstrate lower levels of ferritin (Takala et al. 2010). The correlation of premature delivery to asthma likely reflects an association of both with available iron.

Diurnal variation

Nocturnal worsening of symptoms of asthma (e.g. dyspnea and cough) is frequent and can be accompanied by elevations in airway inflammation, airflow limitation, and airways hyperresponsiveness (Spengler and Shea 2000; Sutherland 2005). As many as 75 % of asthmatic subjects are awakened by asthma symptoms at least once per week with approximately 40 % experiencing nocturnal symptoms on a nightly basis. Both healthy and asthmatic individuals show the lowest levels of forced expiratory volume in one second at approximately 4 AM, but the latter demonstrate increased circadian variations in pulmonary function indices. Most deaths related to asthma symptoms occur during the night (Tough et al. 1998). Nocturnal asthma has been attributed to allergen exposure in the bedroom, gastroesophageal reflux, a drop in body temperature with airway cooling, and a diminished efficacy of medications. However, human serum iron levels also demonstrate diurnal variation with nadirs in availability being in the early morning. Thus serum iron levels parallel the increase in asthma symptoms, physiologic changes, and deaths (Cao et al. 2012; Ridefelt et al. 2010). Tissue iron levels in animal models reveal a comparable diurnal variation further supporting a participation of iron homeostasis in this form of asthma (Unger et al. 2009; Unger et al. 2014).

Menstruation

Peri-menstrual worsening of asthma has been documented in a significant number of asthmatic women. It is more commonly observed in severe disease and has been correlated with both aspirin sensitivity and lower baseline pulmonary function (Eliasson et al. 1986; Rao et al. 2013; Vrieze et al. 2003). Menstruation can function as a contributing factor in the development of episodes of near fatal asthma with significantly more near fatal asthma episodes observed on the first day of menstruation than on the remaining days (Martinez-Moragon et al. 2004). Changes in pulmonary function and bronchial hyperreactivity show associations with the menstrual cycle in both healthy and asthmatic populations (Dratva et al. 2010; Farha et al. 2009; Jeon et al. 2009). Supporting an affiliation of peri-menstrual worsening of asthma with decreased iron availability, women 18–44 years of age with regular menstrual cycles showed cyclical changes in serum concentrations of hemoglobin, iron, and ferritin as well as transferrin saturation with the lowest values observed during menses (Kim et al. 1993).

Pregnancy

Asthma is a common condition during pregnancy. Worldwide the prevalence of asthma among pregnant women is on the rise, and pregnancy leads to a worsening of asthma for many (Murphy and Gibson 2011). Asthma control is most frequently impacted in the latter stages of pregnancy. Despite iron absorption rising in pregnancy, the amount that can be absorbed from the diet is frequently less than the iron requirements (Bothwell 2000). The serum ferritin concentration drops during pregnancy reflecting mobilization of iron from maternal stores to meet the increased demands of the fetus. In addition, maternal asthma is associated with poor pregnancy outcomes including increased risks for low birthweight and preterm delivery. Iron deficiency can be seen as the common risk factor leading to all these outcomes (Demissie et al. 1998; Murphy et al. 2011).

Exercise

Exercise-induced bronchoconstriction is airway hyperresponsiveness with exercise and occurs frequently among athletes even in the absence of a diagnosis of asthma or any respiratory symptoms. In general, exercise-induced bronchoconstriction affects 8–17 % of the population of individuals without asthma, 30–70 % in elite athletes, and up to 90 % in the population of individuals with asthma (Anderson 2012; Burr et al. 2006; Randolph 2009; Weiler et al. 2007). Airway hyperresponsiveness is more common among athletes relative to those individuals in the general population (Langdeau et al. 2000; Sue-Chu 2012). Exercise-induced bronchoconstriction is often the first manifestation of asthma; there are a significant number of individuals with no history of asthma that respond to vigorous exercise with a mild degree of exercise-induced bronchoconstriction (Godfrey and Bar-Yishay 1993). Exercise-induced asthma was initially described in cross-country skiers, but studies demonstrate that asthma is more common in all trained athletes relative to control subjects (Helenius et al. 1998). Exercise-induced asthma is especially common in long-distance runners and elite swimmers (Helenius et al. 1997).

The increased prevalence of asthma among athletes has been attributed both to an intense respiratory heat and water exchange and to an increased penetration of allergens and pollutants into the lower airways (Boulet and O'Byrne 2015). Exercise is also associated with an acute decrease in serum iron and chronic reductions in iron stores manifested by a “sports anemia”. With exercise a decrement in iron concentration can occur quickly in athletes with serum iron falling 45 % within 15 min of the completion of an ironman triathlon race, 46 % following a 12-h cross-country ski race, and 39 % following a 24-h race (Ginsburg et al. 1996; Pattini et al. 1990). Among untrained individuals, serum iron similarly decreases within 24 h after a high-intensity exercise bout (Smith and Roberts 1994). With repeated aerobic exercise, there can be changes in many measures of iron homeostasis including serum iron, ferritin, transferrin, and transferrin saturation (Auersperger et al. 2013; Di Santolo et al. 2008; Mainous and Diaz 2009; Wilkinson et al. 2002). Accordingly, athletes have a greater risk of iron depletion and anemia; this is especially true for females (Fallon, 2004; Landahl et al. 2005; Ostojic and Ahmetovic 2008; Reinke et al. 2012). Nearly one-fifth of recreational athletes have anemia and a third have an iron deficit (Di Santolo et al. 2008). Changes in iron homeostasis are considered a result of hemodilution, increased destruction of erythrocytes, depressed iron absorption, and increased iron loss in sweat and through the gastrointestinal tract. Comparable to other phenotypes, exercise-induced asthma corresponds with decrements in concentrations of stored and available iron.

Pathologic determinants of asthma

Infections

Respiratory infections are considered to both impact the development of asthma and worsen existing asthma. Respiratory tract infections caused by the viruses, Chlamydophila and Mycoplasma have been postulated to have significant roles in the pathogenesis of asthma (Busse et al. 2010; Guilbert and Denlinger 2010). For patients either at risk of asthma or with existing asthma, viral respiratory tract infections can affect the disease. Studies also invoke atypical bacterial infections, Mycoplasma pneumoniae and Chlamydia pneumoniae, as factors in both acute exacerbation and chronic asthma (Martin 2006). New evidence has shown that wheezing episodes early in life due to human rhinoviruses are a major risk factor for the diagnosis of asthma at age 6 years. Viral respiratory tract infections, especially those caused by human rhinoviruses, are associated with asthma exacerbations. Symptoms can persist for up to 3 months following an acute infection. Respiratory viruses are associated not only with asthmatic exacerbations but also with the response to methacholine challenge in both humans and animal models (Cheung et al. 1995; Little et al. 1978; Sorkness et al. 1994). Post-viral bronchial hyperreactivity syndrome is a common complication of respiratory tract viral infections (Ostransky and Blais 1991; Xepapadaki et al. 2005).

Infections disrupt iron homeostasis in cells, animal models, and humans. In cells, infection promotes an iron deficiency (Fillebeen and Pantopoulos 2013). In animal models, serum iron concentrations decrease following infection reflecting a diminished metal availability (Brosnahan et al. 2012; Heegaard et al. 2000; Konijn and Hershko 1977). Similarly, infections in humans impact iron homeostasis with elevations in serum ferritin and TfR while serum iron decreases (Butensky James et al. 2009; Kim et al. 2013). Decrements in serum iron may be one of the earliest responses to infection in humans (Hoppe and Hulthen 2007). Comparable to levels in the serum, iron availability in the lung is predicted to diminish with infection, and it is proposed that this decrease leads to asthmatic exacerbation. While less studied, this disruption in iron homeostasis is observed following viral infections. Iron is needed for viral replication. Therefore, by ensuring the infected cell is iron replete, a virus favors its own growth (Drakesmith and Prentice 2008). Conversely, reducing iron availability to viruses should benefit the host. Such a reaction has been documented with lower serum iron, percent transferrin saturation, hemoglobin, and hematocrit with mumps and chickenpox virus infections (Cemeroglu and Ozsoylu 1994; Weinberg 1996). This hypoferremia following virus infections is expected to increase the risk for asthma.

Acute phase response

During an acute infection, controlling the initial proliferation and dissemination of microbials can be critical in determining the outcome for the host. Decreasing concentrations of requisite iron available to the pathogen is an effective host defense mechanism employed to limit such replication. Hepcidin coordinates this decreased metal availability by binding to ferroportin and triggering internalization and degradation of the exporter. This results in decreased iron uptake by enterocytes and reduced metal transport by macrophages (Schmidt 2015). Accordingly, during infection serum iron concentrations can be reduced to about 30 % of normal, a physiological change known as the hypoferremia of infection, which limits iron availability for microbes but also leads to increased iron storage in macrophages of the reticuloendothelial system. This same pathway participates in the inflammatory response to numerous agents and stressors (e.g. particles, surgery, and autoimmune disease) and is included in the acute phase response. This can result in an anemia of inflammation (both acute and chronic). A purpose of the acute phase response can be to influence iron homeostasis thereby minimizing the challenge to the living system. The acute phase response can generate an iron deficiency with serum iron decreasing to a nadir by 48 h (Solis et al. 2006). Complete recovery of serum iron concentrations to normal may necessitate more than 45 days. A similar evolution is observed in the parameters dependent on iron levels including the red cell count, serum hemoglobin level, and hematocrit. Finally, the acute phase response includes changes in the expression of numerous genes and proteins involved in the iron regulatory pathway not only in the liver but also in extra-hepatic tissues (Sheikh et al. 2007). Subsequently, iron concentrations can be diminished in serum and throughout the body. After intramuscular injection of turpentine to initiate an inflammatory response in an animal model, levels of iron were decreased in extra-hepatic tissues (e.g. brain) (Malik et al. 2011). During this response, TfR-1 increased in the extra-hepatic tissues reflecting a relative deficiency of iron. While the effects of inflammatory stimuli have not been investigated in the lung and airways, it is likely that the acute phase response will trigger a hypoferremia in these tissues also. The relationship between the acute phase response and iron homeostasis can account for the association between non-specific inflammation and asthma. The prolonged duration of the acute phase response can also assist in understanding why recovery from an asthmatic exacerbation can be delayed.

Endotoxin

Endotoxin exposure can increase the risks of wheeze and asthma in children (Rennie et al. 2012). The timing of exposure may be of significance since endotoxin exposure in early life has also been proposed to be protective against developing childhood asthma (Campo et al. 2006). Endotoxin exposure is also associated with increased risk for wheeze and greater asthma severity later in life (Celedon et al. 2007; Perzanowski et al. 2006). Household endotoxin exposures were associated with asthma symptoms, current asthma medication use, and wheezing (Thorne et al. 2005). A recent study demonstrated that elevated endotoxin concentrations in house dust were associated with current wheezing, exercise-induced wheeze, and use of prescription medication for wheezing (Thorne et al. 2015).

Endotoxin alters iron homeostasis in both cells and living systems resulting in a hypoferremia between 6 and 24 h after exposure of animal models and humans (Duvigneau et al. 2008; Kemna et al. 2005; Krijt et al. 2006; Orro et al. 2004; Vernooy et al. 2005; Zhang et al. 2005). Some portion of the decrease in iron, but not the entirety, may reflect an acute phase response following endotoxin exposure (Plessers et al. 2015). This decreased iron availability will lead to an increased risk for asthma and its worsening.

Antibiotics

Cohort studies demonstrate a greater risk of persistent wheeze and asthma in early childhood with prenatal antibiotic treatment (Benn et al. 2002; Jedrychowski et al. 2006). A dose response relationship between the number of antibiotic courses and the risk of either wheeze or asthma has been demonstrated (Jedrychowski et al. 2006; McKeever et al. 2002). In addition, substantial obstructive reactions occur in some asthmatic subjects following the inhalation of specific antibiotics (e.g. gentamicin) (Dally et al. 1978). Such reactions appear to be non-immunological in nature.

Antibiotics impact iron homeostasis. Several antibiotics are recognized to function directly as iron chelators to produce an anti-bacterial effect (Badal et al. 2015; Moon et al. 2013; Sobke et al. 2012). Others appear to trigger changes in the expression of iron transport genes, which then contribute to an anti-microbial effect (Ferrandiz and de la Campa 2014; van Delden et al. 2013). Still other antibiotics display activity against microbes in a manner that has not been defined in detail but depends on interference with the host iron homeostasis (Fiori and Van Dijck 2012; Lounis et al. 2003). The impact of antibiotics on iron homeostasis suggests a participation in metal availability, which influences asthma.

Vaccinations

Vaccinations have been associated with exacerbations of asthma (Hassan et al. 1992). Both hypoferremia and anemia have been shown to follow vaccination (Olivares et al. 1989). In addition an acute phase reaction is evident among vaccinated animals. This reaction includes decreased serum iron and reduced metal availability (Andersen et al. 2012; Stokka et al. 1994). In turn, the hypoferremia, anemia, decreased serum iron, and reduced metal availability after vaccination could affect asthma.

Smoking and environmental tobacco smoke

Smoking is a major contributor to asthma and its exacerbations. Smoking is associated with acute decrements in pulmonary function (Kougias et al. 2013; Miller and Sproule 1966; Sobol et al. 1977). Endpoints (including spirometric indices, airway resistance, and conductance) have revealed worsening after smoking a single cigarette. Furthermore, smoking by itself increases airway hyperreactivity (O’Connor et al. 1989; Sunyer et al. 1997; Tashkin et al. 1992). These acute effects of smoking cigarettes are greater among asthmatic individuals relative to non-asthmatics (Papaioannou et al. 2010).

There is an association of maternal cigarette smoking with asthma in the newborn with maternal smoking increasing the risk for (1) physician-diagnosed asthma during both childhood and adolescence and (2) diminished airway function at the end of the first year of life (Hollams et al. 2014; Lau et al. 2002; Neuman et al. 2012; Xepapadaki et al. 2009). There is also support for a relationship between maternal smoking and the other asthma risk factors of low birth weight and pre-term births (Bjerg et al. 2011; Cox et al. 2013; Prabhu et al. 2010); the loss in birth weight is calculated to be 19 g for each cigarette smoked per day (Adriaanse et al. 1996).

Comparable to smoking, exposure to environmental tobacco smoke increases the incidence of wheeze and asthma in children and young people (Dezateux et al. 1999; James et al. 2005; Stein et al. 1999). Maternal exposure to environmental tobacco smoke during pregnancy has also been implicated in low birth weight, preterm delivery, and asthma of the newborn (Burke et al. 2012; Windham et al. 2000).

The effects of cigarette smoking and ETS (particle-related exposures) on iron availability are consistent with effects of other forms of particulate matter (PM). Functional groups at the particle surface can complex cell metal (e.g. carboxylic and phenolic groups in cigarette smoke particles) (Ghio et al. 2013). Accordingly, cell iron is lost to the particle (Fig. 4). In response to a reduction in intracellular iron, the cell generates superoxide (a ferrireductant which can participate in the uptake of the metal) and upregulates both importers (e.g. DMT1) and storage (i.e. ferritin) proteins in an attempt to reacquire requisite iron. This particle exposure results in an altered iron homeostasis with an absolute accumulation of the metal in the cell but diminished concentrations of available iron. Accordingly, smoking and environmental tobacco smoke impact iron homeostasis to increase asthma.

Fig. 4
figure 4

Particles and iron sequestration of cell iron. Functional groups at the surface of the particle sequester cell iron including mitochondrial and nuclear sources. The result of particle exposure is an altered iron homeostasis, an accumulation of the metal in the cell, and a functional deficiency

Full size image

Iron in the developing fetus is accumulated against a concentration gradient. Despite some capacity to tolerate maternal iron deficiency, many stresses will impact the availability of the metal in the developing fetus. There are statistically significant negative correlations between maternal smoking and infants’ total body iron (Pateva et al. 2015). The number of cigarettes smoked per day correlates negatively with infants’ ferritin; newborn infants of women who smoke during pregnancy have lower iron stores than those of non-smoking mothers. Maternal smoking during pregnancy decreases the ferritin concentration in umbilical cord blood (94 and 163 ng/ml in smoking and non-smoking respectively) and placenta (Chelchowska and Laskowska-Klita 2002). Accordingly, more packs per day and more days of smoking during pregnancy lead to diminished iron available to the fetus, and this can increase asthma risk.

Air pollution

Human exposure to air pollution increases morbidity related to asthma (Friedman et al. 2001). Exposure to ambient air pollution is associated with increased asthma symptoms, exacerbations, and a decline in lung function in children (Gent et al. 2009; Guarnieri and Balmes 2014; O’Connor et al. 2008). Asthmatic exacerbations increase with proximity to point sources of air pollution (Loyo-Berrios et al. 2007; Meng et al. 2007). While measures of both indoor and outdoor PM can correlate with asthmatic worsening, the latter demonstrate a greater effect (Koenig et al. 2005). Air pollution particle exposure correlates with decrements in pulmonary function, and elevations in ambient particle exposure increase bronchodilator use among asthmatics (Rabinovitch et al. 2006; Trenga et al. 2006). In addition to exacerbations of asthma, several studies have associated exposures to air pollution particles with the development of asthma. A prospective birth cohort of 4146 children followed from birth to 2 years of age showed a significant correlation between a physician diagnosis of asthma in the first year of life and ambient air PM levels (Brauer et al. 2002). Corresponding to the biological effects of other particles (e.g. cigarette smoking and ETS), maternal exposure to air pollution particles has also been implicated in low birth weight and preterm delivery (Dadvand et al. 2013; Stieb et al. 2012). Comparable to ambient air PM, there are also studies, which support a role for diesel exhaust particles in causing asthma (Heinrich and Wichmann 2004). Exposure to polyaromatic hydrocarbons in diesel exhaust particles increases the risk for childhood asthma (Karimi et al. 2015). Finally, higher prenatal PM2.5 exposure at mid-gestation was associated with asthma development in boys by age 6 years (Leon Hsu et al. 2015). Animal studies show that perinatal exposure results in persistent airway hyperresponsiveness (Fanucchi et al. 2006; Manners et al. 2014; Mauad et al. 2008).

The biological effect of an air pollution particle is dependent on complexation of host iron by the particle with diminished metal availability (Fig. 4) (Ghio et al. 2015a). Polycyclic aromatic hydrocarbons, found in air pollution particles, and metabolized products can alter iron homeostasis by complexing cell iron (Schreinemachers and Ghio 2016). Components of air pollution other than particles (e.g. ozone) have similarly been associated with a disruption in iron homeostasis (Ghio et al. 2014; Ghio et al. 2007). These data support a relationship between asthma, exposure to air pollutants, and iron deficiency.

Particles other than those included in cigarette smoke and air pollution

Asthma can be associated with exposure to numerous, disparate particles. The weight of indoor dust samples taken in the home correlated with asthma (Elliott et al. 2007). Total suspended particles from a pre-harvest sugar cane burning had an acute effect on asthma admissions, starting at day one and remaining elevated for four days (Arbex et al. 2007). Likewise, exposures to particles other than cigarette smoking and air pollution particles are associated with decrements in pulmonary function and bronchial hyperreactivity (Ames et al. 1982; Lapp et al. 1972; Sundblad et al. 2002). Regarding transgenerational asthma, cooking with a wood stove during pregnancy, yet another particle-related exposure, is similarly associated with both asthma in children (as well as low birthweights) (Siddiqui et al. 2008).

All particle exposures are associated with a disruption in iron homeostasis (Ghio et al. 2015a, 2013). Surface functional groups will complex lung iron following inhalation and deposition (Fig. 4). While total iron will increase, the available metal will decrease. A functional deficiency in iron results, and this will impact the risk for asthma.

Occupational exposures (non-particulate)

There are numerous work-related exposures, which are associated with asthma. Comparable to other asthma phenotypes, individuals with several types of occupational asthma demonstrate a disruption in iron homeostasis and a functional deficiency of the metal. Inhalation of isocyanates is a leading cause of occupational asthma. These compounds have been demonstrated to disrupt iron homeostasis (Hur et al. 2008; Kim et al. 2010). Occupational exposure to various metals other than iron can also present as asthma or exacerbation of asthma (Daenen et al. 1999; Fernandez-Nieto et al. 2006; Irsigler et al. 1999; Malo et al. 1993; Sauni et al. 2010; Schneider et al. 2012; Wittczak et al. 2008). Exposure to metals other than iron has been shown to disrupt normal iron homeostasis in cells with a resulting functional deficiency (Ghio et al. 2015b). Exposure to cotton dust is associated with a loss in FEV1 across a work shift (i.e. byssinosis or “scheduled asthma”). Such decrements in pulmonary function with resulting symptoms of shortness of breath and wheezing have been observed to be more frequent among women who have diminished iron levels relative to males (Beck et al. 1982; Velazquez et al. 1991). Cotton is comprised of cellulose, a known iron chelator (Phillips and Fernandez 1981; Platt and Clydesdale 1984). Similar to particles, inhaled cotton dust is predicted to react with and accordingly diminish available iron in the respiratory tract. This effect has been observed as the formation of ferruginous bodies in the lung following cotton dust exposure (Spencer 1985).

Inhaled iron chelators

Direct exposure to compounds recognized to chelate iron can initiate respiratory symptoms and changes in pulmonary function consistent with asthma. Prominent among these compounds are citric acid and ethylenediaminetetraacetic acid (EDTA). Citric acid is diagnostically employed in cough provocation tests (Kastelik et al. 2002). Citric acid also causes increased bronchial responsiveness and bronchoconstriction (Ricciardolo et al. 1999). Among patients, there was an increase in airflow resistance reaching a maximum within 30 s after citric acid inhalation. This resistance returned to control values within one to three minutes (Simonsson et al. 1967). Exposure to EDTA, an iron chelator previously used as a preservative in inhalers, was considered responsible for an increase in airflow resistance associated with worsening of asthma (Brophy 1989). Complexation of airway iron by both citrate and EDTA is likely to result in a deficiency of iron leading to asthmatic symptoms and loss of pulmonary function.

Obesity

Obesity is a risk factor for asthma in all demographic groups; it is related more strongly to non-atopic (non-eosinophilic) than atopic airways disease (Raj et al. 2014). The pathophysiology underlying the association between obesity and asthma is poorly defined (Mohanan et al. 2014). Potential mechanisms of asthma in obese individuals could involve changes in the mechanical properties of the respiratory system, adipokine-mediated inflammation, chronic inflammation, gastro-esophageal reflux, and complications from sleep-disordered breathing (Ali and Ulrik 2013).

Iron homeostasis is disrupted by obesity. Obesity alters distribution of iron both at the cellular and tissue levels (Orr et al. 2014). Analysis employing public-use data files from the continuous NHANES (1999–2006) survey suggests a negative relationship between levels of blood iron content and individual body mass index after controlling for other individual characteristics (Neymotin and Sen 2011). This inverse correlation between serum iron and body mass index is observed despite elevations in serum iron with the same measure (Gillum 2001). Iron deficiency and anemia are frequent findings in subjects with progressed stages of obesity (Aigner et al. 2014; Nead et al. 2004). Obesity-associated inflammation is linked to disruptions in iron homeostasis (Cheng et al. 2012). In obese school-aged children, both iron status and inflammation are improved by weight reduction (Gong et al. 2014). The deficiency of iron can participate in the mechanistic pathway leading to asthma associated with obesity.

Maternal overweight increases the likelihood of childhood wheeze more than threefold (Kozyrskyj and Pawlowski 2013). This also correlates with a diminished availability of iron in the newborns of obese mothers and is likely a factor in transgenerational asthma (Phillips et al. 2014).

Anemias including hemoglobinopathies

The worldwide prevalence of anemia in 2010 was 32.9 % with over 2 billion people affected (i.e. almost one-third of the world was anemic) (Kassebaum et al. 2014). In a prospective-cohort study, anemia was found to be a risk factor for asthma among 200 children aged 2–18 years (Ramakrishnan and Borade 2010). Anemic children were 5.75 times more susceptible to asthma attacks when compared with nonanemic children, and iron deficiency was the most common cause (Lopez et al. 2015).

Sickle cell disease can commonly be associated with asthma (Anim et al. 2011; Miller and Gladwin 2012). The criteria for a physician diagnosis of asthma are not well defined in a patient with sickle cell disease. It is unclear whether sickle cell disease and asthma are distinct or overlapping co-morbid conditions (Field and DeBaun 2009). Accordingly, the true prevalence of co-existence is frequently underestimated. Significant overlap does exist in the clinical presentations of an episode of acute chest syndrome and asthma exacerbation. Supporting the concept that sickle cell disease and asthma can be the presentation of a singular process, there is evidence of a higher than expected prevalence of airway hyperreactivity and obstructive lung disease among patients with sickle cell disease without a diagnosis of asthma. Over 50 % of children with sickle cell disease show a positive response to methacholine while the prevalence of airway hyperreactivity is 18 % in children in the general population (Strunk et al. 2008).

Both iron deficiency anemia and iron deficiency have been documented in cohorts of patients with hemoglobinopathies including sickle cell anemia (Mohanty et al. 2008; Palma-Carlos et al. 2005). Sixty-seven percent of subjects with sickle cell anemia and 26 percent with sickle cell trait had evidence suggesting iron insufficiency and, following provision of iron, an improvement in hemoglobin levels can be observed in both sickle cell disorders. In addition, an acute phase reaction has been demonstrated to impact patients with sickle cell anemia (Singhal et al. 1993). Integral to the acute phase reaction, iron will be diminished reflecting a further decrease in availability of this metal. The lack of sufficient iron levels in hemoglobinopathies will increase the risk for asthma.

Autoimmune disease

The prevalence of bronchial hyperreactivity and asthma is increased in autoimmune diseases (Shen et al. 2014). While bronchial hyperreactivity was present in 6.5 % of the healthy controls, 25 % of the systemic sclerosis patients without associated Sjögren’s syndrome, 42.2 % of those with primary Sjögren’s syndrome, and 50 % of those with systemic sclerosis with associated Sjögren’s syndrome were reactive to methacholine (Gudbjornsson et al. 1991; La Corte et al. 1995; La Corte et al. 1992; Potena et al. 1990). Similarly, methacholine responsiveness was elevated among patients with rheumatoid arthritis (55 % arthritics vs. 16 % controls) (Hassan et al. 1994). Airflow obstruction is also significantly increased in these rheumatoid arthritis patients compared with controls.

A disruption in iron homeostasis has been found in autoimmune disease (Baynes et al. 1987; Blake et al. 1980; Palermo et al. 1986). Despite evidence of systemic iron deficiency, iron concentrations are elevated in involved synovium among patients diagnosed with rheumatoid arthritis (Blake et al. 1984; Giordano et al. 1991; Ogilvie-Harris and Fornaiser 1980). Iron deficiency and anemia are frequently associated with disease reflecting either an absolute or a functional insufficiency of the metal (Masson, 2011; van Steenbergen et al. 2013; Vanarsa et al. 2012). Intravenous iron has been employed in treating autoimmune disease with improvement of anemia and decreases in markers of inflammation (Duthie 1985). It is likely that this iron deficiency participates in respiratory disease related to autoimmune disease.

Surgery and critical illness

Surgical procedures can be complicated by asthma (Tirumalasetty and Grammer 2006). Iron deficiency develops post-surgically with serum concentrations and transferrin saturation reaching a nadir 48 h following the procedure (Solis et al. 2006). Other measured parameters reflect a disruption in iron homeostasis with decrements in red cell count, hemoglobin, and hematocrit. Transferrin-soluble receptors increase late in the post-operative course (days 30 and 45). There is not a complete recovery to normal iron homeostasis even after 45 days. Similarly, asthma and abnormalities in iron metabolism can be observed among patients with critical illness. Bronchial hyperreactivity is common among survivors of acute respiratory distress syndrome (Simpson et al. 1978). Those individuals diagnosed to have critical illness (e.g. acute respiratory distress syndrome) can demonstrate significant disruption of iron homeostasis along with an acute phase reaction, which negatively impacts iron availability (Ghio et al. 2003; Headley et al. 1997). The abnormal iron homeostasis after surgery and critical illness can contribute to asthma.

Medications

Certain medications are associated with increased asthma (Covar et al. 2005). The mechanistic pathway by which these medications trigger asthma is proposed to vary. Aspirin-exacerbated respiratory disease is characterized by hypersensitivity, asthma, and chronic rhinosinusitis with nasal polyps after use of aspirin and non-steroidal anti-inflammatory drugs. Aspirin-exacerbated respiratory disease affects 10–29 % asthmatic patients. Similarly, an asthma-like condition with cough develops in approximately 10 % of the patients treated with angiotensin converting enzyme (ACE) inhibitors, and the medication must be discontinued in approximately 50 % of these individuals (Overlack 1996). Beta-blockers can also be associated with bronchoconstriction and worsening of asthma (Short et al. 2012).

These medications, which affect asthma also impact iron import, accumulation, and release by cells (Komarov et al. 2006; Mak et al. 2006). Aspirin use is associated with diminished total iron stores reflected by lower serum ferritin concentrations (Fleming et al. 2001). ACE inhibitors decreased plasma iron concentration within 1–2 h after administration (Skrzypczak et al. 2010). Use of beta-blockers decreases iron availability (Komarov et al. 2006). These disruptions in iron homeostasis can lead to a deficiency of the metal thus increasing risk for asthma.

Stress

Psychosocial stress is linked to the expression of asthma (Cohen et al. 2008; Lange et al. 2011; Wright 2011). The poor are disproportionately affected by such stress and its associated asthma (Rosenberg et al. 2014). Psychosocial stress can similarly affect an acute phase response to diminish iron availability (de Rooij et al. 2007; Maes et al. 1996). The hypoferremia associated with the acute phase response is proposed to then impact asthma.

Co-morbidity of asthma

Co-morbid conditions of asthma can be associated with an absolute and a functional iron deficiency supporting a relationship of these diseases with a disruption in homeostasis of this metal. These co-morbid conditions can include eczema, urticaria, celiac disease, inflammatory bowel disease, migraine headaches, restless leg syndrome, and pulmonary hypertension.

Epidemiology has associated both eczema and urticaria with asthma (Ronchetti et al. 2002; Silverberg and Hanifin 2013). Comparable to wheeze consistent with asthma, umbilical cord blood concentrations of iron can inversely correlate with the later onset of eczema (Shaheen et al. 2004). Patients with atopic eczema have significantly lower levels of serum ferritin relative to controls (David et al. 1990). Low serum iron levels are also frequently observed among patients with chronic idiopathic urticaria while the restoration of normal serum iron is associated with either remission or clinical improvement (Guarneri et al. 2014; Wu et al. 2015).

The diagnosis of celiac disease increases the risk for asthma (Ludvigsson et al. 2011). Celiac disease is rare among individuals without iron deficiency (Karaman et al. 2016; Murray et al. 2013). Elevation of the total iron binding capacity is associated with an increased risk of celiac disease (Letner et al. 2015). Inflammatory bowel disease is a second gastrointestinal disease linked with asthma (Brassard et al. 2015). While not well defined, a disruption in iron homeostasis is obvious in inflammatory bowel disease with significant iron deficiency and anemias observed (Sukumaran et al. 2014; Wiskin et al. 2012).

Epidemiologic investigation supports migraine and asthma as co-morbid disorders (Chen and Leviton 1990; Davey et al. 2002; Wang et al. 2010). Patients with iron deficiency anemia have a high frequency of migraine headaches suggesting a relationship between disruption in the homeostasis of this metal and migraine headaches (Gur-Ozmen and Karahan-Ozcan 2015; Pamuk et al. 2015; Vukovic-Cvetkovic et al. 2010).

Current medication use among those with restless leg syndrome includes asthma/allergy drugs suggesting a possible association between the two (Pearson et al. 2008). Restless leg syndrome is a frequent sleep disorder that is linked to disturbed iron homeostasis. Among patients with iron deficiency anemia, the prevalence of clinically significant restless leg syndrome was 23.9 %, which is nine times higher than in the general population (Allen et al. 2013). Restless leg syndrome is more common during pregnancy (when iron deficiency is common) than in the general population, occurring at a 2–3 times higher prevalence, and intravenous iron can improve symptoms of restless leg syndrome during this time (Picchietti et al. 2012).

Finally, asthma and pulmonary arterial hypertension can be observed as co-morbidity. These two conditions share important pathological features including inflammation, smooth muscle contraction, and remodeling (Said et al. 2010). In both diseases there is increased resistance in airways and pulmonary arteries, respectively, due to airway and pulmonary vasoconstriction, smooth muscle constriction, and thickening of the walls caused by proliferation of smooth muscle and other cells (i.e. remodeling). Remodeling in asthma can include an increase in airway wall thickness, fibrosis, smooth muscle mass and vascularity, as well as abnormalities in extracellular matrix composition. Muscularization and remodeling of smaller pulmonary arteries are essential pathological lesions in pulmonary arterial hypertension. These shared pathological features between asthma and pulmonary arterial hypertension are unlikely to be strictly coincidental. Rather, they probably reflect a single underlying mechanism that is common to both diseases. The two diseases can also demonstrate overlapping clinical presentations (Achouh et al. 2008; Hayes 2007). Iron deficiency is common in patients with idiopathic pulmonary arterial hypertension and can also be associated with disease severity and poor clinical outcome (Decker et al. 2011; Rhodes et al. 2011). There can be a beneficial response among patients diagnosed with pulmonary arterial hypertension who are treated with oral iron (Ruiter et al. 2011). Iron availability modifies pulmonary arterial pressure and pulmonary vascular responses to hypoxia (Smith et al. 2008). In addition, hypoxic pulmonary hypertension may be attenuated by iron supplementation and exacerbated by iron depletion (Smith et al. 2009).

The relationship between these co-morbid conditions and disruptions of iron homeostasis support a role for diminished availability of the metal in asthma.

Treatment of asthma

Treatment for asthma can include pharmacologic agents, which impact iron homeostasis both directly and indirectly. Catecholamines regulate cellular iron homeostasis and deliver requisite iron to cells and organelles (Tapryal et al. 2015; Zhang et al. 2013). Inhaled therapies (corticosteroids alone or combined with beta-2 agonists) are a mainstay in asthma and reduce inflammation (Antoniu 2010; Tang et al. 2010). Through such a reduction in systemic inflammation, a normalization of iron availability is predicted.

The direct provision of iron has been demonstrated to impact animal models of asthma and measures of airway hyperreactivity in humans. Iron supplementation in a well-established mouse model of ovalbumin-driven allergic asthma resulted in a significant decrease in airway eosinophilia, while systemic iron injections led to a significant suppression of allergen-induced airway eosinophilia, Th2 cytokine release, and airway hyperreactivity compared to placebo (Maazi et al. 2011). Iron supplementation in iron-deficient women led to resolution of chronic cough and bronchial hyperreactivity further supporting an association of asthma and iron deficiency (Bucca et al. 2012).

Pathophysiologic events underlying asthma

Disruptions in iron homeostasis with decreased concentrations of available metal can initiate pathophysiologic events culminating in asthma. Such events include (1) inflammation and (2) muscle contraction involved in bronchoconstriction and obstruction.

Regarding the participation of iron in inflammation, changes in the availability of this metal activate numerous pathways that coordinate inflammation (Nakagawa et al. 2014). Cellular iron deficiency and its associated oxidative stress affect an activation of specific kinases including p38, JNK, ERK1, and ERK2 (Choi et al. 2007; Fan et al. 2011; Huang et al. 2007; Yu and Richardson 2011). Increasing the cell concentration of available iron (Ghio et al. 2013) can diminish phosphorylation of kinases following exposure to inflammatory agents. Comparable to kinase phosphorylation, diminished cell iron concentrations correspond to an activation of pro-inflammatory transcription factors including NF-κB, AP-1, HIF-α, and CREB (Fan et al. 2011; Lee et al. 2008; Lu et al. 2009; Varesio et al. 2010; Yuan et al. 2011). Inflammatory mediator release also increases with diminished availability of iron (Choi et al. 2007; Jeong et al. 2005; Lee et al. 2008; Varesio et al. 2010). The biological effects after exposure to inflammatory agents have been demonstrated to be related to iron homeostasis (Driscoll et al. 2001; Ghiazza et al. 2011).

Iron levels specifically affect the balance between and the intensity of Th1 and Th2 arms of the immune response with availability of the metal impacting the Th2 response (Gordeuk et al. 2009). Diminished iron availability during the development of the fetal immune system may result in a pattern of persistent Th2 dominance (Weigert et al. 2015). Th2 cells participate with M2 cells in reparative reactions including asthma (Martinez et al. 2009). The differentiation of these inflammatory cells can reflect iron homeostasis (Corna et al. 2010). Eosinophilia is an endpoint commonly employed as an index of risk for asthma. Iron deficiency can also correspond with eosinophilia (Weigert et al. 2015). Finally, decreased iron availability can lead to elevation of blood levels of IL-4 which participates in the Th2 response (Naderi et al. 2013; Nyakeriga et al. 2005).

Muscle contraction demonstrates a relationship with iron concentration. Negative chronotropic and inotropic effects were observed with inclusion of iron in monolayer cultures of ventricular myocytes (Moreb et al. 1988). Iron reduces peak-developed tension and the maximal rate of tension development in isometrically contracting left atrial strips and right ventricular papillary muscles (Artman et al. 1984a). Iron depresses myocardial contractility as evidenced by dose-related declines in contractile function in isolated isometrically contracting rabbit left atrial strips and right ventricular papillary muscles (Artman et al. 1982, 1984a, 1984b). In contrast, anemic infants with decreased iron concentrations showed increases in cardiac contractility, stroke volume, and left ventricular output period (Cambonie et al. 2007). Such enhanced contractility (left ventricular performance) in anemia can also be observed among adults (Florenzano et al. 1984). Changes in muscle contractility associated with anemia reverse with blood transfusion (Cambonie et al. 2007). In skeletal muscle, iron exposure is associated with diminished force in a strength test (Reardon and Allen 2009). Finally, elevated iron availability impacts venous dilation, possibly through a negative ionotropic effect (Reissman et al. 1955; Robotham and Lietman 1980; Whitten and Brough 1971, 1968).

Conclusions

A relationship between iron deficiency and asthma is supported by direct measures of the metal, evaluation of determinants (demographic, physiologic, and pathologic) of the disease, analysis of co-morbid conditions, comprehension of therapies, and analysis of the pathophysiological events. The determinants are predicted both to interact in lowering iron and to demonstrate synergy in impacting asthma (e.g. exercise and gender both influence iron homeostasis and are anticipated to contribute to an even greater risk for asthma among female athletes) (Fallon 2004; Landahl et al. 2005; Ostojic and Ahmetovic 2008; Reinke et al. 2012). A recognition of the association between iron deficiency and asthma suggests preventive and therapeutic interventions. The prevention of iron deficiency may necessitate preclusion of exposure to specific compounds, which can either function as or be metabolized to inappropriate iron chelators. Several of these compounds associated with disruption of iron homeostasis and asthma have found use in restricting the growth of microbes and plants (e.g. pesticides, antibiotics, and preservatives). Unfortunately, it appears that these same agents can impact the human in an identical manner to disrupt iron homeostasis, produce a metal deficiency, and initiate disease. While exclusion of such compounds from the environment of the lower respiratory tract can be recommended, systemic prohibition might be required. It is unclear if elimination of such exposures is possible.

The relationship between iron deficiency and asthma reveals a potential for treatment with iron. Treatment with oral supplementation can be tested. In addition, intravenous iron might be of benefit especially since it has been employed in the treatment of other diseases in which deficiency exists (e.g. restless leg syndrome) (Lieske et al. 2015). In these disease states, a threshold effect of ferritin (i.e. iron storage) has been suggested. In restless leg syndrome, iron supplementation to achieve ferritin levels of at least 50 ng/ml has been proposed (Kantor et al. 2003; Leschziner and Gringras 2012; Trost et al. 2006). It should be noted that intravenous iron is not always successful in reversing a deficiency (Hogan et al. 2015; Liles 2012). With iron deficiency associated with an acute phase response, intravenous replacement frequently fails (El-Khatib et al. 2006). However, inhaled iron presents a therapeutic approach to treating asthma as an iron deficiency unique to the lung. The inhaled iron could be presented as salts (e.g. iron sulfate) or nanoparticles (e.g. iron oxide), which can be tolerated by humans (Kleinman et al. 1981). While it is expected that some forms of asthma (e.g. asthma associated with active infections) will not be amenable to such an approach addition of ancillary iron has the potential to provide a new treatment for asthma.