Abstract
The phenotype Hp 2-1 of haptoglobin has been previously associated with increased risk of Parkinson disease (PD) and with serum iron abnormalities in PD patients. Tobacco smoking has been consistently observed in epidemiology studies to be inversely related to PD risk, with mechanisms that remain uncertain. We recently observed that the protective effect of smoking on PD risk is stronger among subjects of haptoglobin Hp 2-2 and Hp 1-1 phenotypes, and weaker among subjects of haptoglobin Hp 2-1 phenotype. In this PD case–control study, we investigated whether tobacco smoking was associated with changes in serum haptoglobin and ferritin concentration that depended on haptoglobin phenotype among 106 PD patients and 238 controls without PD or other neurodegenerative disorders. Serum ferritin concentration, serum haptoglobin concentration, haptoglobin phenotype, and smoking data information of cases and controls were obtained. Differences in haptoglobin and ferritin concentration by smoking status and pack-years of smoking were calculated as well as regression between pack-years and haptoglobin and ferritin concentration, and the effect of haptoglobin phenotype on these parameters. Tobacco smoking was associated with an elevation in serum haptoglobin concentration, especially among healthy controls of haptoglobin Hp 2-2 phenotype, and with an elevation in ferritin concentration especially among PD patients of haptoglobin Hp 2-1 phenotype. These findings suggest that an elevation in haptoglobin concentration, preferentially among subjects of haptoglobin Hp 2-2 phenotype, could be a contributing factor to the protective effect of smoking on PD risk.
Keywords: Tobacco smoking, Haptoglobin concentration, Ferritin concentration, Haptoglobin phenotype, Parkinson’s disease
Introduction
Haptoglobin (Hp) is a serum protein that binds free hemoglobin and allows scavenging of free hemoglobin by macrophages (Bowman and Kurosky 1982; McCormick and Atassi 1990; Kristiansen et al. 2001). Because of a genetic polymorphism of the alpha chain of the Hp gene (Langlois and Delanghe 1996), there are three common Hp phenotypes in the population, Hp 1-1, Hp 2-1, and Hp 2-2, which exhibit profound structural and functional differences: Hp 1-1 molecules have the smallest molecular weight, and Hp 2-2 molecules have the highest molecular weight, up to 900 kDa (Langlois and Delanghe 1996). Hp 2-2 complexes have lower Hb-binding capacity than Hp 1-1 and Hp 2-1 (Javid 1965), and also have lower Hb-scavenging power than Hp 2-1 and Hp 1-1 as a consequence of their lower ability to reach extra-vascular fluids, due to their higher molecular mass (Langlois and Delanghe 1996; Melamed-Frank et al. 2001). Hp polymorphism also affects serum Hp concentration, since subjects with Hp 1-1 phenotype have the highest levels of serum Hp concentration, while subjects with Hp 2-2 phenotype have the lowest levels (Langlois and Delanghe 1996). Hp 2-1 phenotype was previously found to be associated with a significantly increased risk of Parkinson disease (Costa-Mallen et al. 2008). We also recently observed that abnormally low serum iron levels in PD patients are present predominantly among subjects of Hp 2-1 phenotype, while abnormalities in PD patients of Hp 2-2 and Hp 1-1 phenotype are less pronounced (Costa-Mallen et al. 2015).
A very consistent epidemiologic finding in PD research is a reduced risk of PD, by about 50 %, for smokers compared with non-smokers (Ritz et al. 2007). Lifetime cigarette smoking was also found to be associated with reduced Lewy body accumulation (Tsuang et al. 2010). Selective survival of non-smoking PD cases does not account for the protective effect of cigarette smoking (Morens et al. 1996), and several hypotheses have been advanced over the years to explain the strong protective effect of smoking on PD risk. For example, smoking causes inhibition of monoamine oxidase B, the main metabolizing enzyme of dopamine (Fowler et al. 1996). However, none of the current hypotheses are completely supported by epidemiological findings. Second-hand smoke was recently found to confer reduced PD risk with a magnitude similar to the effect of active smoking (Searles et al. 2012), suggesting a biological effect of cigarette smoke rather than a personality trait being predisposing both to smoking avoidance and PD risk. Haptoglobin phenotype is a modifying factor for the effect of smoking on PD risk, as tobacco smoking exerts a protective effect on PD risk for subjects of Hp 2-2 and Hp 1-1 phenotype, but not for subjects of Hp 2-1 phenotype (Costa-Mallen et al. 2015), with mechanisms not elucidated.
It is noteworthy that tobacco smoking causes an increase in hematocrit (Kurata 2006) and higher circulatory iron. This effect was postulated to be due to the induced hypoxia and to the increased red blood cell turnover (Young and Moss 1989; Northrop-Clewes and Thurnham 2007). Ghio et al. (2008) reported that serum iron and ferritin levels among smokers were higher compared with never smokers.
Excess of iron deposition in the brain has been shown to be able to cause oxidative stress and contribute to the neurodegenerative processes typical of PD (Halliwell 2001; Berg et al. 2002; Youdim et al. 2004). Changes in systemic iron metabolism could potentially affect the levels of brain iron. We, therefore, hypothesized that smoking-related changes in iron metabolism may underlie the inverse association of smoking with PD by ultimately leading to lower levels of iron in the brain.
In addition, haptoglobin can exert neuroprotection by scavenging free hemoglobin (Zhao et al. 2009), and an increase in haptoglobin concentration itself could potentially be neuroprotective by reducing oxidative stress (Alayash et al. 2013). We, therefore, explored whether tobacco smoking was associated with higher Hp concentration in PD cases and controls, and if this effect was present preferentially in subjects of Hp 2-2 phenotype.
Methods
Study participants
Informed consent was obtained from all individual participants included in the study. All study procedures were approved by the Institutional Review Boards of Bastyr University, the University of Washington, and the Veterans Affairs Puget Sound Health Care System (VAPSHCS). Study participants were recruited in the Puget Sound area of Washington State between 2010 and 2015. Recruitment sources for the PD patients were the Washington State Parkinson Disease Registry (WPDR), the Michael J. Fox Foundation Fox Trial Finder, Bastyr University campus, the Bastyr Center for Natural Health (BCNH), Senior Centers, and referral from collaborating Neurologists in the Seattle area.
Controls were recruited from the Bastyr University campus, the BCNH, Senior Centers, and advertisement over local websites. Seventeen PD cases and 12 controls for a total of 29 participants included in this study were enrolled in the Parkinson’s Genetic Research Study (PaGeR) (Zabetian et al. 2005).
All PD cases recruited met the “UK Parkinson’s Disease Society Brain Bank clinical diagnostic criteria” (UKBB) for PD (Gibb and Lees 1988). Only patients meeting UKBB clinical diagnostic criteria were included in the study.
For all PD cases, medical records were obtained from each patient’s neurologist and were reviewed by a second neurologist (S-C Hu) to determine whether each patient met UKBB criteria. In addition to the exclusion criteria as from the UKBB, the presence of more than one relative with PD also constituted exclusion criteria for the PD patients, as determined from a screening telephone interview before the study visit.
Controls were subjects free of PD or other neurodegenerative diseases, as determined from subjects’ interviews and medical records reviews. For the 12 controls from the PaGeR study, no additional medical history exclusion criteria were imposed on control eligibility.
Exclusion criteria for the PD cases and controls recruited from all other sources other than the participants recruited from the PaGeR were: the presence of current active cancer under treatment, chronic hepatitis, and HIV sero-positivity, as determined from participants’ interviews and medical records reviews. Participants who had an acute illness or fever at the time of the visit were rescheduled for a different date when the acute illness had subsided.
The control group was frequency matched to cases by age in 10-year categories.
Data collection
Blood samples were collected by venipuncture from each study participant; serum was separated, split in different aliquots, and frozen at −70 °C. One aliquot was submitted to the LabCorp Clinical laboratory for tests of serum haptoglobin (Hp) concentration and ferritin concentration. Hp concentration was expressed as mg/dl, and ferritin concentration was expressed as ng/ml. Hp phenotyping was performed on serum samples for all subjects, with a method as previously described (Costa-Mallen et al. 2008). Briefly, each serum sample was mixed with purified hemoglobin (Hb) solution, allowing Hp in the serum sample to bind to Hb. The Hp–Hb complexes were subjected to non-denaturing acrylamide gel electrophoresis on 4–20 % gradient gels at 70 V for 16 h, followed by peroxidase staining for bands visualization.
For the 29 participants from the PaGeR study, ferritin concentration could be determined for all participants, whereas Hp concentration could only be determined for 7 participants, because of lack of availability of sufficient sample for Hp concentration tests; as a result, ferritin concentration was collected for a total of 344 participants, 106 PD cases and 238 controls, while Hp concentration was collected for a total of 322 participants, 91 PD cases and 231 controls.
Demographic and smoking history information was obtained from all study participants, including age, gender, race, number of cigarettes a day smoked, and years smoked. Participants were considered “ever smokers” if they had smoked at least 100 cigarettes during their lifetime. Participants were considered as “past” smokers if they were ever smokers but were not currently smokers at the time of the visit. Participants who were not current smokers but smoked at least 100 cigarettes in the past were considered as past smokers. Pack-years of smoking were calculated by multiplying packs/day of cigarettes smoked by the number of years smoked.
Data analysis
Age- and gender-adjusted differences in mean serum Haptoglobin and serum ferritin were calculated between never smokers and ever smokers, and among never smokers, past smokers, and current smokers by one-way analysis of covariance (ANCOVA). This analysis was performed for the overall study population, and separately for the three Hp phenotype groups. Separate analyses were also performed by PD case/control status.
Linear regression was used to test the relation between pack-years of smoking and serum Hp concentration, and serum ferritin concentration, in models where age and gender were included as covariates. Standardized beta (B) coefficients of linear regression between pack-years of smoking and Hp concentration and ferritin concentration were calculated. Linear regression was used also to test for interaction between never/past/current smoking status and Hp phenotype on ferritin concentration and Hp concentration levels, in models where case/control status, gender, and Hp phenotype were used as fixed factors, age was used as covariate, and either serum ferritin or serum Hp concentration was the dependent variable.
All analyses were performed in the overall study population independently of ethnicity. In separate analyses limited to non-Hispanic white subjects, the results did not change materially, and thus, only data for the entire study population are presented.
All data analysis was performed with SPSS 19.0 software.
Results
The study population consisted of 344 subjects, 106 PD cases and 238 controls, for whom ferritin concentration and Hp phenotype were determined and smoking status information was collected.
Hp concentration could be measured for a total of 322 participants, of which 91 PD cases and 231 controls. The demographics and smoking status of the participants are shown in Table 1.
Table 1.
Demographics and characteristics of the study population
| Total | PD cases | Controls | |
|---|---|---|---|
| Number, n | 344 | 106 | 238 |
| Men, n (%) | 167 (48.55 %) | 68 (64.15 %) | 99 (41.60 %) |
| Women, n (%) | 177 (51.45 %) | 38 (35.85 %) | 139 (58.40 %) |
| Age, mean (SD) | 64.18 (10.93) | 67.05 (9.98) | 62.90 (11.10) |
| Men, mean age (SD) | 64.77 (11.72) | 67.40 (9.94) | 62.96 (12.53) |
| Women, mean age (SD) | 63.62 (10.12) | 66.42 (10.42) | 62.86 (10.01) |
| Hoehn and Yahr PD stage (SD) | 2.15 (0.69) | 0 | |
| Men | 2.25 (0.69) | 0 | |
| Women | 1.95 (0.66) | 0 | |
| PD duration (years since diagnosis) | 5.51 (5.28) | ||
| Men | 4.92 (4.94) | ||
| Women | 6.45 (5.71) | ||
| Smoking status | |||
| Never smokers (%) | 174 (50.6 %) | 70 (66.0 %) | 104 (43.7 %) |
| Ever smokers (%) | 170 (49.4 %) | 36 (34.0 %) | 134 (56.3 %) |
| Past smokers (%) | 95 (27.6 %) | 31 (29.2 %) | 64 (26.9 %) |
| Current smokers (%) | 75 (21.8 %) | 5 (4.7 %) | 70 (29.4 %) |
| Years smoked, mean (SD) | 11.43 (15.20 %) | 6.33 (11.46) | 13.70 (16.10) |
| Packs/day, mean (SD) | 0.38 (0.50) | 0.24 (0.47) | 0.44 (0.50) |
| Pack-years, mean (SD) | 9.89 (16.75) | 5.57 (14.77) | 11.82 (17.24) |
| Pack-years past smokers | 14.92 (16.56) | 17.74 (23.10) | 13.55 (12.23) |
| Pack-years current smokers | 26.49 (20.25) | 8.05 (5.57) | 27.81 (20.29) |
| Years since quitting for past smokers (SD) | 29.23 (13.54) | 32.35 (13.04) | 27.79 (13.64) |
| Race | |||
| White | 307 (89.2 %) | 100 (94.3 %) | 207 (87.0 %) |
| African American | 16 (4.7 %) | 2 (1.9 %) | 14 (5.9 %) |
| Asian | 7 (2.0 %) | 2 (1.9 %) | 5 (2.1 %) |
| Native American | 1 (0.3 %) | 1 (0.9 %) | 0 (0.0 %) |
| More than 1 race | 13 (3.8 %) | 1 (1.0 %) | 12 (5.0 %) |
| Serum ferritina (ng/ml), mean (SD) | 110.60 (110.61) | 120.22 (124.44) | 106.31 (103.86) |
| Min–max (ng/ml) | 3–868 | 3–790 | 9–868 |
| Men, ferritin (ng/ml), mean (SD) | 141.94 (135.81) | 148.47 (145.14) | 137.45 (129.58) |
| Min–max (ng/ml) | 3–868 | 3–790 | 9–868 |
| Women, ferritin (ng/ml), mean (SD) | 81.03 (68.109) | 69.68 (41.35) | 84.13 (73.56) |
| Min–max (ng/ml) | 7–634 | 7–176 | 11–634 |
| Serum Hpb (mg/dl), mean (SD) | 131.30 (52.57) | 126.69 (54.70) | 133.11 (51.71) |
| Min–max (mg/dl) | 4–332 | 4–326 | 25–332 |
| Men, Hp (mg/dl), mean (SD) | 126.86 (55.13) | 124.81 (60.08) | 128.04 (52.35) |
| Min–max (mg/dl) | 4–326 | 4–326 | 33–322 |
| Women, Hp (mg/dl), mean (SD) | 135.03 (50.17) | 129.43 (46.41) | 136.53 (51.19) |
| Min–max (mg/dl) | 25–332 | 31–215 | 25–332 |
Ferritin concentration was measured for all the participants, 106 PD cases and 238 controls
Hp concentration was measured for 91 PD cases and 231 controls. For 22 participants, 15 PD cases and 7 controls, Hp concentration could not be measured. Demographic characteristics of the PD cases and controls for whom Hp concentration was measured were very similar to the ones for the overall study population
Serum ferritin concentration results
Serum ferritin was overall significantly higher in men than women (men = 141.94 ng/ml, and women = 81.03 ng/ml, p < 0.001), as expected (Cook et al. 1976).
PD cases had overall slightly higher serum ferritin concentration than controls (PD = 120.22 ng/ml, controls = 106.31 ng/ml); however, this difference was not statistically significant (Table 2). The difference in serum ferritin between PD cases and controls was not significant also when considering men and women separately, or after stratification for Hp phenotype groups (Table 3).
Table 2.
Mean ferritin in PD cases and controls and stratified by gender and Hp phenotype groups
| PD case/control status |
n | Mean ferritin concentration (ng/ml) |
Standard error |
Age-adjusted p value |
Age- and gender-adjusted p value |
|
|---|---|---|---|---|---|---|
| Total subjects | PD cases | 106 | 120.22 | 12.087 | 0.237 | 0.995 |
| Controls | 238 | 106.31 | 6.732 | |||
| Total | 344 | 110.60 | 5.964 | |||
| Women | PD cases | 38 | 69.68 | 6.709 | 0.461 | – |
| Controls | 139 | 84.13 | 6.240 | |||
| Total | 177 | 81.03 | 5.119 | |||
| Men | PD cases | 68 | 148.47 | 17.601 | 0.310 | – |
| Controls | 99 | 137.45 | 13.023 | |||
| Total | 167 | 141.94 | 10.509 | |||
| Hp 1-1 | PD cases | 17 | 111.05 | 26.043 | 0.259 | 0.238 |
| Controls | 37 | 82.08 | 8.587 | |||
| Total | 54 | 91.20 | 10.109 | |||
| Hp 2-1 | PD cases | 50 | 129.86 | 21.17 | 0.487 | 0.849 |
| Controls | 121 | 122.14 | 11.31 | |||
| Total | 171 | 124.40 | 10.09 | |||
| Hp 2-2 | PD cases | 39 | 111.87 | 15.012 | 0.281 | 0.880 |
| Controls | 80 | 93.57 | 9.194 | |||
| Total | 119 | 99.57 | 7.903 |
Table 3.
Mean ferritin in never smokers, past smokers, and current smokers stratified by Hp phenotype and PD case/control status
| Case/control status and Hp phenotype |
Smoking status |
n | Mean ferritin concentration (ng/ml) |
Standard error |
Age-adjusted p value |
Age- and gender-adjusted p value |
|---|---|---|---|---|---|---|
| Total subjects | Never smokers | 174 | 107.73 | 7.480 | 0.914 | 0.997 |
| Past smokers | 95 | 116.08 | 12.232 | |||
| Current smokers | 75 | 110.30 | 14.533 | |||
| Hp 1-1 | Never smokers | 29 | 93.93 | 15.793 | 0.641 | 0.644 |
| Past smokers | 15 | 90.66 | 18.571 | |||
| Current smokers | 10 | 84.10 | 13.663 | |||
| Hp 2-1 | Never smokers | 88 | 110.30 | 11.141 | 0.233 | 0.383 |
| Past smokers | 53 | 131.00 | 19.125 | |||
| Current smokers | 30 | 154.10 | 33.140 | |||
| Hp 2-2 | Never smokers | 57 | 110.78 | 12.823 | 0.088 | 0.111 |
| Past smokers | 28 | 99.67 | 17.411 | |||
| Current smokers | 34 | 80.67 | 9.616 | |||
| Total PD cases | Never smokers | 70 | 117.18 | 12.172 | 0.569 | 0.530 |
| Past smokers | 31 | 120.22 | 26.848 | |||
| Current smokers | 5 | 162.80 | 107.302 | |||
| PD Hp 1-1 | Never smokers | 14 | 119.07 | 30.236 | 0.566 | 0.651 |
| Past smokers | 3 | 73.66 | 46.548 | |||
| Current smokers | 0 | – | – | |||
| PD Hp 2-1 | Never smokers | 32 | 101.93 | 16.610 | 0.010 | 0.031 |
| Past smokers | 17 | 155.35 | 46.072 | |||
| Current smokers | 1 | 590.00 | – | |||
| PD Hp 2-2 | Never smokers | 24 | 136.41 | 21.671 | 0.032 | 0.088 |
| Past smokers | 11 | 78.63 | 17.456 | |||
| Current smokers | 4 | 56.00 | 13.391 | |||
| Total controls | Never smokers | 104 | 101.37 | 9.458 | 0.932 | 0.771 |
| Past smokers | 64 | 114.07 | 12.842 | |||
| Current smokers | 70 | 106.55 | 13.855 | |||
| Controls Hp1-1 | Never smokers | 15 | 70.46 | 9.524 | 0.149 | 0.163 |
| Past smokers | 12 | 94.91 | 21.023 | |||
| Current smokers | 10 | 84.10 | 13.663 | |||
| Controls Hp 2-1 | Never smokers | 56 | 115.08 | 14.783 | 0.812 | 0.911 |
| Past smokers | 35 | 121.02 | 18.628 | |||
| Current smokers | 30 | 136.63 | 29.635 | |||
| Controls Hp 2-2 | Never smokers | 33 | 92.15 | 15.025 | 0.922 | 0.771 |
| Past smokers | 17 | 113.29 | 26.264 | |||
| Current smokers | 30 | 83.96 | 10.685 |
Statistically significant values at p ≤ 0.05 are shown in bold
Ferritin was higher in subjects of Hp 2-1 phenotype as compared with subjects of Hp 1-1 and Hp 2-2 phenotype overall (Hp 1-1 = 91.20 ng/ml, Hp 2-1 = 124.40 ng/ml, Hp 2-2 = 99.57 ng/ml, age-, gender-, smoking status-adjusted p value = 0.003). To test whether the higher ferritin levels in subjects of Hp 2-1 phenotype were due to an effect of smoking, we stratified by ever/never smoking status, and the higher levels of serum ferritin in subjects of Hp 2-1 phenotype was present among subjects who were ever smokers (Hp 1-1 = 88.04 ng/ml, Hp 2-1 = 139.34 ng/ml, Hp 2-2 = 89.25 ng/ml, p = 0.025), but not among subjects who were never smokers (Hp 1-1 = 93.93 ng/ml, Hp 2-1 = 110.30 ng/ml, Hp 2-2 = 110.78 ng/ml, p = 0.713). Tobacco smoking was associated with an elevation of serum ferritin is subjects of Hp 2-1 Hp phenotype, but not among subjects of Hp 1-1 or Hp 2-2 phenotype. In fact, as shown in Table 3), while there were no significant differences in serum ferritin levels among never–past–current smokers overall, after stratifying by Hp phenotype, serum ferritin levels resulted to be progressively higher from never smokers to past smokers to current smokers among subjects of Hp 2-1 phenotype, especially among PD cases. On the other hand, serum ferritin levels resulted progressively lower from never smokers to past smokers to current smokers among subjects of Hp 2-2 phenotype (Table 3); this interaction between Hp phenotype and never/past/current smoker status was statistically significant (p = 0.036). The three way-interaction between Hp phenotype, never/-past/current smoker status, and PD case/control status,was also statistically significant (p = 0.041).
The interaction between never/past/current smoker status, Hp phenotype, and gender was not statistically significant (p = 0.136).
We tested the correlation between pack-years of smoking and serum ferritin concentration (Table 4) and observed a significant increase in serum ferritin levels with increase in pack-years of smoking among PD cases (standardized beta coefficient = 0.289, p = 0.005). When stratifying by Hp phenotype, pack-years of smoking was correlated with an increase in serum ferritin among PD cases of Hp 2-1 phenotype (standardized beta coefficient = 0.392, p = 0.008), but not among PD cases of Hp 2-2 phenotype, similar to what observed for the association between ferritin and never–past–current smoking status. The interaction between pack-years of smoking and Hp phenotype was statistically significant (p = 0.003).
Table 4.
Correlation between pack-years of smoking and ferritin concentration in the overall study population and stratified by Hp phenotype and PD case/control status
| Subjects | n | Standardized beta coefficient |
p value | Age-adjusted p value |
Age- and gender-adjusted p value |
|---|---|---|---|---|---|
| Total subjects | 344 | 0.110 | 0.041 | 0.045 | 0.130 |
| Hp 1-1 | 54 | 0.053 | 0.706 | 0.616 | 0.769 |
| Hp 2-1 | 171 | 0.192 | 0.012 | 0.015 | 0.062 |
| Hp 2-2 | 119 | −0.055 | 0.554 | 0.557 | 0.527 |
| PD cases | 106 | 0.289 | 0.003 | 0.002 | 0.005 |
| Hp 1-1 | 17 | 0.067 | 0.799 | 0.886 | 0.873 |
| Hp 2-1 | 50 | 0.392 | 0.005 | 0.002 | 0.008 |
| Hp 2-2 | 39 | −0.110 | 0.505 | 0.489 | 0.401 |
| Controls | 238 | 0.049 | 0.455 | 0.514 | 0.924 |
| Hp 1-1 | 37 | 0.149 | 0.378 | 0.301 | 0.434 |
| Hp 2-1 | 121 | 0.084 | 0.359 | 0.486 | 0.787 |
| Hp 2-2 | 80 | −0.006 | 0.958 | 0.994 | 0.824 |
Statistically significant values at p ≤ 0.05 are shown in bold
Among controls, the correlation between pack-years of smoking and serum ferritin levels resulted non-significant overall, as well as for the individual Hp phenotype groups (Table 4).
Serum Hp concentration results
Serum Hp concentration was similar in men and women (Hp in men = 126.86 mg/dl, Hp in women 135.03 mg/dl, p = 0.165). As shown in Table 5), serum Hp concentration was not significantly different in PD cases and controls overall, or after stratification by gender and Hp phenotype.
Table 5.
Mean Hp concentration in PD cases and controls, and stratified by gender and Hp phenotype groups
| PD case/control status |
n | Mean Hp concentration (mg/dl) |
Standard error |
Age-adjusted p value |
Age- and gender-adjusted p value |
|
|---|---|---|---|---|---|---|
| Total subjects | PD cases | 91 | 126.69 | 5.735 | 0.322 | 0.435 |
| Controls | 231 | 133.11 | 3.403 | |||
| Total | 322 | 131.30 | 2.930 | |||
| Women | PD cases | 37 | 129.43 | 7.631 | 0.559 | – |
| Controls | 138 | 136.53 | 4.358 | |||
| Total | 175 | 135.03 | 3.793 | |||
| Men | PD cases | 54 | 124.81 | 8.177 | 0.597 | – |
| Controls | 93 | 128.04 | 5.429 | |||
| Total | 147 | 126.86 | 4.548 | |||
| Hp 1-1 | PD cases | 16 | 165.88 | 9.773 | 0.792 | 0.713 |
| Controls | 35 | 159.63 | 6.468 | |||
| Total | 51 | 161.59 | 5.355 | |||
| Hp 2-1 | PD cases | 44 | 133.36 | 6.509 | 0.309 | 0.389 |
| Controls | 117 | 143.42 | 4.705 | |||
| Total | 161 | 140.67 | 3.860 | |||
| Hp 2-2 | PD cases | 31 | 97.00 | 10.996 | 0.411 | 0.583 |
| Controls | 79 | 106.10 | 5.257 | |||
| Total | 110 | 103.54 | 4.872 |
Consistently with what was previously shown (Langlois and Delanghe 1996), serum Hp concentration was higher for subjects of Hp 1-1 phenotype, intermediate for subjects of Hp 2-1 phenotype, and lowest for subjects of Hp 2-2 phenotype (serum Hp in Hp 1-1 = 161.59 mg/dl, in Hp 2-1 = 140.67 mg/dl, in Hp 2-2 = 103.54 mg/dl, p < 0.001).
Serum Hp concentration was not different between PD cases and controls overall (PD = 126.69 mg/dl, and controls = 133.11 mg/dl, age- and gender-adjusted p = 0.435)(Table 5). No significant differences in Hp concentration between PD cases and controls were present also after stratification by gender or Hp phenotype (Table 5).
Hp concentration resulted lower in PD patients than controls among subjects who were ever smokers (PD = 112.64 mg/dl, controls = 139.12 mg/dl, age- and gender-adjusted p value = 0.032). Among never smokers, instead, serum Hp concentration was not different between PD cases and controls (PD = 132.94 mg/dl, controls = 125.10 mg/dl, age- and gender-adjusted p value = 0.294. The interaction between case/control status and ever/never smoking status on Hp concentration was statistically significant (p = 0.016).
Hp concentration resulted significantly higher among current smokers as compared with past smokers and never smokers in the overall study population (p = 0.016) (Table 6). The difference in Hp concentration between never/past/current smokers was strongest among subjects of Hp 2-2 phenotype (p = 0.001) as compared with subjects of Hp 2-1 and Hp 1-1 phenotype. The interaction between Hp phenotype and never/past/current smoking status on Hp concentration levels was statistically significant (p < 0.001).
Table 6.
Mean Hp concentration in never smokers, past smokers, and current smokers, for the overall study population, and stratified by Hp phenotype and PD case/control status
| Case/control status and Hp phenotype |
Smoking status |
n | Mean Hp concentration (mg/100 ml) |
Standard error |
Age-adjusted p value |
Age- and gender-adjusted p value |
|---|---|---|---|---|---|---|
| Total subjects | Never smokers | 162 | 128.15 | 4.198 | 0.018 | 0.016 |
| Past smokers | 88 | 123.86 | 5.300 | |||
| Current smokers | 72 | 147.47 | 6.053 | |||
| Hp 1-1 | Never smokers | 27 | 156.30 | 7.542 | 0.068 | 0.065 |
| Past smokers | 14 | 166.07 | 9.576 | |||
| Current smokers | 10 | 169.60 | 12.837 | |||
| Hp 2-1 | Never smokers | 83 | 141.08 | 5.413 | 0.532 | 0.459 |
| Past smokers | 48 | 130.94 | 6.719 | |||
| Current smokers | 30 | 155.10 | 9.083 | |||
| Hp 2-2 | Never smokers | 52 | 92.88 | 6.735 | 0.001 | 0.001 |
| Past smokers | 26 | 88.08 | 7.430 | |||
| Current smokers | 32 | 133.41 | 9.404 | |||
| Total PD cases | Never smokers | 63 | 132.94 | 7.305 | 0.215 | 0.212 |
| Past smokers | 26 | 109.23 | 8.163 | |||
| Current smokers | 2 | 157.00 | 52.000 | |||
| PD Hp 1-1 | Never smokers | 14 | 164.71 | 10.899 | 0.690 | 0.357 |
| Past smokers | 2 | 174.00 | 24.000 | |||
| Current smokers | 0 | – | – | |||
| PD Hp 2-1 | Never smokers | 29 | 138.62 | 9.046 | 0.284 | 0.279 |
| Past smokers | 15 | 123.20 | 7.352 | |||
| Current smokers | 0 | – | – | |||
| PD Hp 2-2 | Never smokers | 20 | 102.45 | 15.105 | 0.940 | 0.994 |
| Past smokers | 9 | 71.56 | 9.660 | |||
| Current smokers | 2 | 157.00 | 52.000 | |||
| Total controls | Never smokers | 99 | 125.10 | 5.064 | 0.008 | 0.005 |
| Past smokers | 62 | 130.00 | 6.582 | |||
| Current smokers | 70 | 147.20 | 6.134 | |||
| Controls Hp1-1 | Never smokers | 13 | 147.23 | 10.206 | 0.063 | 0.071 |
| Past smokers | 12 | 164.75 | 10.798 | |||
| Current smokers | 10 | 169.60 | 12.837 | |||
| Controls Hp 2-1 | Never smokers | 54 | 142.41 | 6.809 | 0.396 | 0.319 |
| Past smokers | 33 | 134.45 | 9.190 | |||
| Current smokers | 30 | 155.10 | 9.083 | |||
| Controls Hp 2-2 | Never smokers | 32 | 86.91 | 5.555 | <0.001 | <0.001 |
| Past smokers | 17 | 96.82 | 9.665 | |||
| Current smokers | 30 | 131.83 | 9.657 |
Statistically significant values at p ≤ 0.05 are shown in bold
Among control subjects, Hp concentration resulted progressively higher from never smokers, to past smokers to current smokers (p = 0.005), and this increase in Hp concentration associated with smoking was strongest among subjects with Hp 2-2 phenotype (p < 0.001), while it was not present among subjects of Hp 2-1 phenotype (Table 6). Among PD cases, the differences in Hp concentration between never/past/current smokers were overall less pronounced than in controls, and not statistically significant, either as a main effect or for the different Hp phenotype groups.
Similar to what we observed for ever/past/current smoking status, serum Hp concentration was significant correlated with pack-years of smoking in the overall population (standardized beta coefficient = 0.157, p = 0.003) (Table 7). When stratifying by Hp phenotype, the correlation between Hp concentration and pack-years of smoking was stronger for subjects of Hp 2-2 phenotype (standardized beta coefficient = 0.324, p = 0.001), and non-significant for subjects of Hp 2-1 and Hp 1-1 phenotype (Table 7).
Table 7.
Correlation between pack-years of smoking and Hp concentration in the overall study population and stratified by Hp phenotype and case/control status
| Subjects | n | Standardized beta coefficient |
p value | Age-adjusted p value |
Age- and gender-adjusted p value |
|---|---|---|---|---|---|
| Total subjects | 322 | 0.157 | 0.005 | 0.005 | 0.003 |
| Hp 1-1 | 51 | 0.144 | 0.312 | 0.279 | 0.380 |
| Hp 2-1 | 161 | 0.134 | 0.091 | 0.104 | 0.064 |
| Hp 2-2 | 110 | 0.324 | 0.001 | 0.001 | 0.001 |
| PD cases | 91 | 0.019 | 0.855 | 0.887 | 0.928 |
| Hp 1-1 | 16 | −0.063 | 0.818 | 0.394 | 0.631 |
| Hp 2-1 | 44 | 0.074 | 0.633 | 0.541 | 0.522 |
| Hp 2-2 | 31 | −0.113 | 0.544 | 0.555 | 0.574 |
| Controls | 231 | 0.188 | 0.004 | 0.005 | 0.002 |
| Hp 1-1 | 35 | 0.254 | 0.141 | 0.130 | 0.181 |
| Hp 2-1 | 117 | 0.136 | 0.145 | 0.155 | 0.101 |
| Hp 2-2 | 79 | 0.446 | <0.001 | <0.001 | <0.001 |
Statistically significant values at p ≤ 0.05 are shown in bold
The correlation between pack-years of smoking and Hp concentration was significant among control participants overall (standardized beta coefficient = 0.188, p = 0.002), and in particular for control subjects of Hp 2-2 phenotype (standardized beta coefficient = 0.446, p < 0.001), but not among PD cases (Table 7), consistently to what observed for the association between Hp concentration and never/-past/current smoking status.
Discussion
In this case–control study of PD, we investigated possible effects of tobacco smoking that could be at the base of the consistently observed inverse association between smoking and PD risk (Morens et al. 1996; Ritz et al. 2007; Tsuang et al. 2010; Searles et al. 2012). Accordingly, we tested whether tobacco smoking was associated with changes in serum proteins that are implicated in regulation of iron metabolism and protection from oxidative stress, and tested serum ferritin and serum Hp concentration in PD cases and controls.
Our results for ferritin are consistent to those previously reported by Ghio et al. (2008), that smoking was associated with elevated serum ferritin concentration. We observed that the elevation in serum ferritin associated with pack-years of smoking was significant among PD patients, and not significant among controls. Notably, the association between higher serum ferritin levels and smoking was present among PD cases of Hp 2-1 phenotype, but not PD cases of Hp 2-2 phenotype. In fact, among patients of Hp 2-2 phenotype, smoking was associated with an opposite effect, lower levels of serum ferritin. The interaction between Hp phenotype and smoking on serum ferritin levels was significant, indicating that the Hp phenotype of as subject influences the effect of tobacco smoking on ferritin levels.
In spite of ferritin levels being higher among men than women, as expected, there were no significant differences by gender on the effect of smoking on ferritin levels, and no significant interaction between gender, Hp phenotype, and smoking status on ferritin levels.
Serum ferritin levels usually correlate with body iron status (Jacobs and Worwood 1975), but in conditions of inflammation, ferritin levels can also increase in the absence of iron overload (Torti and Torti 1994; Cohen et al. 2010).
The main change in brain ferritin in PD, especially in the early stages of the disease, consists in a decrease in the concentration of L-ferritin, while an increase in H-ferritin occurs in later stage of PD progression (Koziorowski et al. 2007; Connor et al. 1995). The decrease in L-ferritin in the SN of PD patients suggests that the process of degeneration starts with a higher availability of free iron, which is released from the ferritin shell. An increase in brain non-ferritin labile iron may be responsible to neurodegenerative processes in PD patients (Wypijewska et al. 2010). How serum ferritin correlates with brain iron levels in general has still not been completely elucidated. In the study of Walter et al. (2012), serum iron was inversely correlated with brain iron in PD patients. It is still not known whether serum ferritin levels may affect brain iron binding proteins and iron metabolism to affect brain levels of L-ferritin and H-ferritin.
If indeed elevated serum ferritin levels were a contributing factor to increased brain free iron and brain oxidative stress, the fact that smoking was associated with higher serum ferritin levels among PD cases with Hp 2-1 phenotype, but not Hp 2-2 phenotype, may also contribute to the lack of protective effect of smoking on PD risk for subjects with Hp 2-1 phenotype that we have previously described (Costa-Mallen et al. 2015).
Serum Hp concentration in the study subjects with Hp 1-1 phenotype was the highest, while subjects with Hp 2-1 phenotype had intermediate Hp levels, and subjects with Hp 2-2 phenotype had the lowest levels: this is a well-known functional effect of the haptoglobin polymorphism, and has been consistently observed (Langlois and Delanghe 1996). We identified an effect of tobacco smoking on serum Hp concentration, as pack-years of smoking was significantly correlated with an increase in serum Hp concentration in the overall study population. To the best of our knowledge, this is the first time that Hp concentration is observed to directly correlate with pack-years of smoking.
Our results also demonstrate that Hp phenotype modifies the effect of smoking on serum Hp concentration. In fact, tobacco smoking resulted in an elevation in serum Hp concentration in subjects of Hp 2-2 phenotype, but not Hp 2-1 phenotype.
Hp has hemoglobin-binding and hemoglobin-scavenging activity (Javid 1965; Langlois and Delanghe 1996; Melamed-Frank et al. 2001; Zhao et al. 2009; Alayash et al. 2013), and an increase in serum Hp concentration could lead to better removal and scavenging of hemoglobin and potentially contribute to lower oxidative stress and PD risk. For these reasons, we believe that a smoking-induced increase in Hp concentration over the lifetime could potentially be a contributing factor to the protective effect of smoking on PD risk that has been consistently observed in epidemiology studies. Since tobacco smoking in this study resulted associated with an elevation in serum Hp concentration among subjects with Hp 2-2, but not Hp 2-1 phenotype, this could potentially be an explanation of the lower risk of PD for subjects with Hp 2-2 phenotype that has been previously observed (Costa-Mallen et al. 2008).
Serum Hp concentration was only slightly and non-significantly lower in PD cases than controls in this study, overall, but it was significantly lower in PD patients than controls among ever smokers: this was because of the association of smoking with an elevation of Hp concentration that occurs specifically among controls, especially of Hp 2-2 phenotype. The correlation between smoking and Hp concentration was much stronger for controls than PD patients. In this respect, we need to consider that a higher percentage of current-smokers were present among controls than PD patients, and this may be a contributing factor to the fact that the elevation of Hp concentration associated with smoking was not statistically significant among PD patients. Further studies on larger populations of PD patients and controls will be needed to determine if the effect of smoking on Hp concentration is present more strongly among healthy control subjects than PD patients.
We postulate that the increase in red blood cells turnover due to hypoxia in smokers (Young and Moss 1989; Northrop-Clewes and Thurnham 2007) may be responsible for an increase in hemoglobin release in the bloodstream, that may ultimately lead to the elevation of Hp and ferritin levels. In our study, smoking was associated with an elevation in Hp concentration especially in controls of Hp 2-2 phenotype and an elevation in ferritin especially in PD patients of Hp 2-1 phenotype, indicating that Hp phenotype influences the effect of smoking on these serum proteins levels.
The heterozygous Hp 2-1 phenotype conferred features that were different than both homozygous Hp 1-1 and Hp 2-2 phenotypes with respect to the effect of smoking on Hp concentration and ferritin concentration. Similarly, we had previously observed that the Hp 2-1 phenotype conferred association with PD (Costa-Mallen et al. 2008), and lower levels of serum iron in PD (Costa-Mallen et al. 2015). The phenomenon where the heterozygous phenotype behaves differently than both homozygous phenotypes is known as heterosis. Heterosis has been observed for association of Hp phenotype with other diseases, such as motor neuron disease, in addition to PD (Fröhlander and Forsgren 1988). In fact an excess of the Hp 2-1 phenotype as compared with both Hp 1-1 and Hp 2-2 phenotypes was observed among motor neuron disease patients (Frölander and Forsgren 1988). At the molecular level, it is not known how the heterozygous Hp 2-1 confers features that are different from both homozygous phenotypes. Interestingly, many other human polymorphic genes display heterosis, for example, the dopamine D2 receptor (DRD2) gene, where carriers of the heterozygous TaqIA 12 genotype display lower DRD2 receptor density than carriers of both homozygous TaqIA 11 and TaqIA 22 genotypes (Comings and MacMurray 2000).
In conclusion, our study identifies associations between tobacco smoking and serum haptoglobin levels and ferritin levels that depend on the haptoglobin phenotype. Further studies will be needed to determine the mechanisms of the observed associations, as well as the temporal patterns of effects after smoking cessation, and the possible role of an elevation of haptoglobin levels in protection from PD risk.
Acknowledgments
The authors would like to thank all the individual study participants. This study was funded by grants from NIH (Grants # R21 NS070202, P50 NS062684, R01 NS065070), the Department of Veteran Affairs (Grant # 1I01BX000531), and Bastyr University (Faculty Seed Grants # 3 and # 4 2009-2011).
Footnotes
Conflict of interest Dr. Costa-Mallen reports no conflict of interest and has been funded by grants from NIH and Bastyr University. Dr. Zabetian reports no conflict of interest and has been funded by grants from the American Parkinson Disease Association, Department of Veterans Affairs, NIH, the Parkinson’s Disease Foundation, and a gift from the Dolsen Foundation. Dr. Shu-Ching Hu reports no conflict of interest and has been funded by grants from NIH. Dr. Agarwal reports no conflict of interest, she reports consulting for Teva Pharmaceutical, Cynapsus, US WorldMeds, UCB, Lundbeck, Impax, and has received research funding from the following companies and institutions: AbbVie Inc., Adamas, Amarantus Bioscience, Biotie, Astellas, Auspex, CHDI Foundation, Chelsea Therapeutic, Civitas, Cynapsus, EvergreenHealth, HSG, inVentive Health, Intec, Lunbeck, Massachusetts General Hospital, Merk & Co., Merz, NIH, Northwestern University, Omeros, Osmotica, Pfizer, PRA, PSG, Solstice, Teva, US WorldMeds. Dr. Checkoway reports no conflict of interest and has been consulting for Alcoa, Inc. (Pittsburgh), NIEHS, Exponent Corp (Los Angeles), Ohio State University, Electric Power Research Institute, (Palo Alto); he reports royalties from Elsevier Publishing Co. and Oxford University Press, and has been funded by grants from NIEHS and the National Multiple Sclerosis Society.
Compliance with ethical standards
Ethical approval All procedures performed in studies involved in human participants were in accordance with the ethical standards of the institution and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
References
- Alayash AI, Andersen CB, Moestrup SK, Bülow L. Haptoglobin: the hemoglobin detoxifier in plasma. Trends Biotechnol. 2013;31:2–3. doi: 10.1016/j.tibtech.2012.10.003. doi: 10.1016/j.tibtech.2012.10.003. [DOI] [PubMed] [Google Scholar]
- Berg D, Roggendorf W, Schröder U, Klein R, Tatschner T, Benz P, Tucha O, Preier M, Lange KW, Reiners K, Gerlach M, Becker G. Echogenicity of the substantia nigra: association with increased iron content and marker for susceptibility to nigrostriatal injury. Arch Neurol. 2002;59:999–1005. doi: 10.1001/archneur.59.6.999. [DOI] [PubMed] [Google Scholar]
- Bowman BH, Kurosky A. Haptoglobin: the evolutionary product of duplication, unequal crossing over, and point mutation. Adv Hum Genet. 1982;12:189–261. doi: 10.1007/978-1-4615-8315-8_3. [DOI] [PubMed] [Google Scholar]
- Cohen LA, Gutierrez L, Weiss A, Leichtmann-Bardoogo Y, Zhang DL, Crooks DR, Sougrat R, Morgenstern A, Galy B, Hentze MW, Lazaro FJ, Rouault TA, Meyron-Holtz EG. Serum ferritin is derived primarily from macrophages through a nonclassical secretory pathway. Blood. 2010;116:1574–1584. doi: 10.1182/blood-2009-11-253815. doi: 10.1182/blood-2009-11-253815. [DOI] [PubMed] [Google Scholar]
- Comings DE, MacMurray JP. Molecular heterosis: a review. Mol Genet Metab. 2000;71:19–31. doi: 10.1006/mgme.2000.3015. [DOI] [PubMed] [Google Scholar]
- Connor JR, Snyder BS, Arosio P, Loeffler DA, LeWitt P. A quantitative analysis of isoferritins in select regions of aged, parkinsonian, and Alzheimer’s diseased brains. J Neurochem. 1995;65:717–724. doi: 10.1046/j.1471-4159.1995.65020717.x. [DOI] [PubMed] [Google Scholar]
- Cook JD, Finch CA, Smith NJ. Evaluation of the iron status of a population. Blood. 1976;48:449–455. [PubMed] [Google Scholar]
- Costa-Mallen P, Checkoway H, Zabeti A, Edenfield MJ, Swanson PD, Longstreth WT, Jr, Franklin GM, Smith-Weller T, Sadrzadeh SM. The functional polymorphism of the hemoglobin-binding protein haptoglobin influences susceptibility to idiopathic Parkinson’s disease. Am J Med Genet B Neuropsychiatr Genet. 2008;147B:216–222. doi: 10.1002/ajmg.b.30593. [DOI] [PubMed] [Google Scholar]
- Costa-Mallen P, Zabetian CP, Agarwal P, Hu SC, Yearout D, Samii A, Leverenz JB, Roberts JW, Checkoway H. Haptoglobin phenotype modifies serum iron levels and the effect of smoking on Parkinson disease risk. Parkinsonism Relat Disord. 2015;21:1087–1092. doi: 10.1016/j.parkreldis.2015.07.006. doi: 10.1016/j.parkreldis.2015.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Fowler JS, Volkow ND, Wang G-J, Pappas N, Logan J, MacGregor R, Alexoff D, Shea C, Schlyer D, Wolf AP, Warner D, Zezulkova I, Cilento R. Inhibition of monoamine oxidase B in the brains of smokers. Nature. 1996;379:733–766. doi: 10.1038/379733a0. [DOI] [PubMed] [Google Scholar]
- Fröhlander N, Forsgren L. Haptoglobin phenotype in motor neuron disease. J Neurol Neurosurg Psychiatry. 1988;51:440–442. doi: 10.1136/jnnp.51.3.440. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ghio AJ, Hilborn ED, Stonehuerner JG, Dailey LA, Carter JD, Richards JH, Crissman KM, Foronjy RF, Uyeminami DL, Pinkerton KE. Particulate matter in cigarette smoke alters iron homeostasis to produce a biological effect. Am J Respir Crit Care Med. 2008;178:1130–1138. doi: 10.1164/rccm.200802-334OC. doi: 10.1164/rccm.200802-334OC. [DOI] [PubMed] [Google Scholar]
- Gibb WR, Lees AJ. A comparison of clinical and pathological features of young- and old-onset Parkinson’s disease. Neurology. 1988;38:1402–1406. doi: 10.1212/wnl.38.9.1402. [DOI] [PubMed] [Google Scholar]
- Halliwell B. Role of free radicals in the neurodegenerative diseases: therapeutic implications for antioxidant treatment. Drugs Aging. 2001;18:685–716. doi: 10.2165/00002512-200118090-00004. [DOI] [PubMed] [Google Scholar]
- Jacobs A, Worwood F. Ferritin in serum. Clinical and biochemical implications. N Engl J Med. 1975;292:951–956. doi: 10.1056/NEJM197505012921805. [DOI] [PubMed] [Google Scholar]
- Javid J. The effect of haptoglobin-polymer size on hemoglobin binding capacity. Vox Sang. 1965;10:320–325. doi: 10.1111/j.1423-0410.1965.tb01396.x. [DOI] [PubMed] [Google Scholar]
- Koziorowski D, Friedman A, Arosio P, Santambrogio P, Dziewulska D. ELISA reveals a difference in the structure of substantia nigra ferritin in Parkinson’s disease and incidental Lewy body compared to control. Parkinsonism Relat Disord. 2007;13:214–218. doi: 10.1016/j.parkreldis.2006.10.002. [DOI] [PubMed] [Google Scholar]
- Kristiansen M, Graversen JH, Jacobsen C, Sonne O, Hoffman HJ, Law SK, Moestrup SK. Identification of the haemoglobin scavenger receptor. Nature. 2001;409:198–201. doi: 10.1038/35051594. [DOI] [PubMed] [Google Scholar]
- Kurata C. Medical check-up findings characteristic of smokers: aimed at improving smoking cessation interventions by physicians. Intern Med. 2006;45:1027–1032. doi: 10.2169/internalmedicine.45.1537. [DOI] [PubMed] [Google Scholar]
- Langlois MR, Delanghe JR. Biological and clinical significance of haptoglobin polymorphism in humans. Clin Chem. 1996;42:1589–1600. [PubMed] [Google Scholar]
- McCormick DJ, Atassi MZ. Hemoglobin binding with haptoglobin: delineation of the haptoglobin binding site on the alpha-chain of human hemoglobin. J Protein Chem. 1990;9:735–742. doi: 10.1007/BF01024768. [DOI] [PubMed] [Google Scholar]
- Melamed-Frank M, Lache O, Enav BI, Szafrenek T, Levy NS, Ricklis RM, Levy AP. Structure-function analysis of the antioxidant properties of haptoglobin. Blood. 2001;98:3693–3698. doi: 10.1182/blood.v98.13.3693. [DOI] [PubMed] [Google Scholar]
- Morens DM, Grandinetti A, Davis JW, Ross GW, White LR, Reed D. Evidence against the operation of selective mortality in explaining the association between cigarette smoking and reduced occurrence of idiopathic Parkinson’s disease. Am J Epidemiol. 1996;144(4):400–404. doi: 10.1093/oxfordjournals.aje.a008941. [DOI] [PubMed] [Google Scholar]
- Northrop-Clewes CA, Thurnham DI. Monitoring micronutrients in cigarette smokers. Clin Chim Acta. 2007;377:14–38. doi: 10.1016/j.cca.2006.08.028. [DOI] [PubMed] [Google Scholar]
- Ritz B, Ascherio A, Checkoway H, Marder KS, Nelson LM, Rocca WA, Ross GW, Strickland D, Van Den Eeden SK, Gorell J. Pooled analysis of tobacco use and risk of Parkinson disease. Arch Neurol. 2007;64:990–997. doi: 10.1001/archneur.64.7.990. [DOI] [PubMed] [Google Scholar]
- Searles Nielsen S, Gallagher LG, Lundin JI, Longstreth WT, Jr, Smith-Weller T, Franklin GM, Swanson PD, Checkoway H. Environmental tobacco smoke and Parkinson’s disease. Mov Disord. 2012;27:293–296. doi: 10.1002/mds.24012. doi:10.1002/mds.24012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Torti SV, Torti FM. Iron and ferritin in inflammation and cancer. Adv Inorg Biochem. 1994;10:119–137. [PubMed] [Google Scholar]
- Tsuang D, Larson EB, Li G, Shofer JB, Montine KS, Thompson ML, Sonnen JA, Crane PK, Leverenz JB, Montine TJ. Association between lifetime cigarette smoking and Lewy body accumulation. Brain Pathol. 2010;20:412–418. doi: 10.1111/j.1750-3639.2009.00296.x. doi: 10.1111/j.1750-3639.2009.0029.00296.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Walter U, Witt R, Wolters A, Wittstock M, Benecke R. Substantia nigra echogenicity in Parkinson’s disease: relation to serum iron and C-reactive protein. J Neural Transm. 2012;119:53–57. doi: 10.1007/s00702-011-0664-z. doi: 10.1007/s00702-011-0664-z. [DOI] [PubMed] [Google Scholar]
- Wypijewska A, Galazka-Friedman J, Bauminger ER, Wszolek ZK, Schweitzer KJ, Dickson DW, Jaklewicz A, Elbaum D, Friedman A. Iron and reactive oxygen species activity in parkinsonian substantia nigra. Parkinsonism Relat Disord. 2010;16:329–333. doi: 10.1016/j.parkreldis.2010.02.007. doi:10.1016/j.parkreldis.2010.02.007. [DOI] [PubMed] [Google Scholar]
- Youdim MBH, Stephenson G, Ben Shachar D. Ironing out iron in Parkinson’s disease and other neurodegenerative diseases with iron chelators. A lesson from 6-hydroxydopamine and iron chelators desferal and VK-28. Ann NY Acad Sci. 2004;1012:306–326. doi: 10.1196/annals.1306.025. [DOI] [PubMed] [Google Scholar]
- Young CJ, Moss J. Smoke inhalation: diagnosis and treatment. J Clin Anesth. 1989;1:377–386. doi: 10.1016/0952-8180(89)90079-2. [DOI] [PubMed] [Google Scholar]
- Zabetian CP, Samii A, Mosley AD, Roberts JW, Leis BC, Yearout D, Raskind WH, Griffith A. A clinic-based study of the LRRK2 gene in Parkinson disease yields new mutations. Neurology. 2005;65:741–744. doi: 10.1212/01.wnl.0000172630.22804.73. [DOI] [PubMed] [Google Scholar]
- Zhao X, Song S, Sun G, Strong R, Zhang J, Grotta JC, Aronowski J. Neuroprotective role of haptoglobin after intracerebral hemorrhage. J Neurosci. 2009;29:15819–15827. doi: 10.1523/JNEUROSCI.3776-09.2009. doi:10.1523/JNEUROSCI.3776-09.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]