Evidence for genetic causality between iron homeostasis and Parkinson's disease a two-sample Mendelian randomization study

doi:10.21203/rs.3.rs-3214069/v1

Download PDF

Research Article

Evidence for genetic causality between iron homeostasis and Parkinson's disease a two-sample Mendelian randomization study

https://doi.org/10.21203/rs.3.rs-3214069/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Jul, 2024

Read the published version in Journal of Trace Elements in Medicine and Biology →

Version 1

posted

You are reading this latest preprint version

Background

The specific etiology of Parkinson's disease (PD), a degenerative disease of the central nervous system, is still unclear, and it is currently believed that its main pathological basis is a decrease in dopamine concentration in the striatum of the brain.Although many previous studies have suggested that iron, as an important nutrient, is involved in the development of PD, there has been a lack of validated genetic evidence on whether there is a causal association between iron homeostasis indices (ferritin, serum iron, TIBC, and TSAT) and PD.

Methods

We used Mendelian randomization (MR) as an analytical method to efficiently assess the genetic association of exposure and outcome, based on the largest genome-wide association study (GWAS) data so far, for the causal association between iron homeostasis indicators and Parkinson's disease, controlling for confounders, by using genetic instrumental variables, that is, single-nucleotide polymorphisms (SNPs), which are randomly assigned and are not subject to any causative effect.

Results

By coordinated analysis of 86 SNPs associated with iron homeostasis markers and 12,858,066 SNPs associated with PD, a total of 56 SNPs were finally screened for genome-wide significance of iron homeostasis associated with Parkinson's disease.The results of the IVW analysis suggested that total iron binding capacity ( β= -0.142; 95%Cl = -0.197 to 0.481; P = 0.4138421 ), transferrin saturation ( β= -0.316 ; 95%Cl = -0.861 to 0.229; P = 0.2553290 ), ferritin ( β= -0.387 ; 95%Cl = -1.179 to 0.405; P = 0.33783807 ) were not genetically causally associated with PD, iron ( β= -0.524; 95%Cl = -0.046 to -0.002; P = 0.03191512 ) was considered to have genetic causality with PD. Cochran's Q test for MR-IVW suggested that TIBC ( P = 0.1618872), TSAT ( P = 0.7099448), ferritin ( P = 0.09768154), iron ( P = 0.8557510) were not heterogeneous with the results of Mr analysis of PD. MR-PRESSO global test showed that TIBC ( P = 0.404), TSAT ( P = 0.759), ferritin ( P = 0.113), iron ( P = 0.87 ) were not detected the presence of horizontal pleiotropy.

Conclusion

Our study found that of the four iron homeostasis markers, TIBC, TSAT, and ferritin were not genetically causally associated with PD, whereas there was a genetically causal association between serum iron and PD, and the increase of serum iron level may reduce the risk of PD.

Iron homeostasis

Iron

Ferritin

Parkinson's disease

Mendelian randomization

Parkinson's disease (PD) is a movement disorder and the second most common degenerative disease of the central nervous system worldwide after Alzheimer's disease. It was first described and documented by James Parkinson in 1817, and its main symptomatic manifestations include resting tremor, bradykinesia, myotonia, and postural balance disorders, as well as a number of non-motor symptoms including olfactory dysfunction, autonomic dysfunction, psychiatric disorders, sleep disorders, and cognitive disorders (1–2). The exact etiology of Parkinson's disease is unknown, and studies have reported that smoking, alcohol consumption, vitamin D exposure, insecticide use, and uric acid levels may be potential risk factors [3]. Its main pathological basis is the reduction of dopamine concentration in the striatum and the presence of discovered neuroleptic vesicles (4–6). In recent years, many scholars have conducted various and in-depth studies on its pathological mechanisms, which include gene mutations, dysfunctions of mitochondria, damage to the ubiquitin-proteasome system, activation of the oxidative stress enzyme signaling pathway, and accumulation of misfolded proteins in the autophagy-lysosomal system, endoplasmic reticulum stress response, inflammatory response, etc. (7–16), and these pathological mechanisms are believed to be important factors leading to the death of dopamine neurons in the brain.

As one of the essential elements, iron participates in many important biological processes in the human body and plays an important role in mitochondrial respiration, redox homeostasis, ATP metabolism, immune regulation, DNA and hormone synthesis (17–18). In order to maintain sufficient iron content in the organism, the cell maintains the stability of iron content by coordinating multiple genes so that they work together to restrict intracellular iron metabolism in the organism (19–21), so it is called iron homeostasis. In recent years, with the introduction of the concept of iron homeostasis, the number of studies in this field has increased year by year. It is believed that iron homeostasis is involved in the development of many diseases, and similar to the role of iron, it is associated with inflammatory response and immune regulation (22–23). In particular, changes in iron levels are considered to be associated with a variety of central nervous system disease pathologies, including Parkinson's disease, Alzheimer's disease, multiple sclerosis, stroke, etc. (24–27). Studies have shown that Parkinson's disease patients have higher iron concentrations in the substantia nigra (28–29), which may be one of the main reasons for inducing oxidative stress in cells of PD patients, affecting mitochondrial function, and causing apoptosis of neuronal cells (30–31).

Mendelian randomization, as an analytical method to assess the causal relationship between exposure factors and outcome factors, by using genetic instrumental variables, namely single nucleotide polymorphisms (SNPs), that are randomly assigned and not subject to any causal relationship, while controlling for confounders [32–33]. The method was first proposed by econometricians and then gradually applied in medical statistics [34–35]. Through reviewing the literature, we found that although many experts and scholars have studied the pathogenesis of iron and Parkinson's disease, there are very few genetic Mendelian randomization studies, especially in combination with the newly proposed theory of "iron homeostasis", so is there a genetic causal association between iron homeostasis and Parkinson's disease? In this study, we tried to explain the association between iron homeostasis and Parkinson's disease from a genetic point of view by using two-sample Mendelian randomization analysis with the genome-wide association summary data.

2.1 Study design

In this study, we chose SNPs from the Genome-Wide Association Study (GWAS) dataset as instrumental variables to explore the genetic causal relationship between exposure and outcome. This study satisfies the three prerequisite assumptions for Mendelian randomization studies: (1) all instrumental variables have a clear correlation with the exposure factor (p < 5 × 10 − 8, F-statistic > 10); (2) the instrumental variables are not associated with any confounders and are independent between the exposure and outcome factor; (3) all instrumental variables must be, and are only, associated with the outcome event through the exposure factor and not influenced by other pathways. All datasets used in this study were publicly available and accessible in the public databases. A diagram of the study design is shown in Fig. 1.

2.2 Data sources

Iron homeostasis indexes data (36) were derived from genome-wide association summary data obtained from the Decode genetic database ( https://www.decode.com/ ), all study participants were of European descent and had provided informed consent upfront.The iron homeostasis data consisted of a total of four indexes, namely ferritin (N = 246,139), serum iron (N = 163,511), total iron binding capacity (N = 135,430) and transferrin saturation (N = 131,471). Serum iron was measured by colorimetric assay (elemental iron and zinc). Ferritin was measured by immunoagglutination principle using latex-enhanced reaction. TIBC is calculated from the results of serum iron and unsaturated iron binding capacity. TSAT is calculated by dividing serum iron by TIBC concentration. Quality control was performed by removing gender mismatch, duplicate samples, low call rates, extreme heterozygosity and non-European ancestry. SNPs associated with all metrics were estimated after adjusting for covariates including age, principal component scores, and other study-specific covariates.

Genome-wide association summary data for Parkinson's disease (37) were obtained from the IEU database ( http://gwas.mrcieu.ac.uk/ ), and all participants were of European ancestry and had provided informed consent upfront. The meta-analysis involved a sample size of 2,829, including 1,570 patients diagnosed with Parkinson's disease and 1,259 controls of European origin, and all cases included in the analysis were diagnosed with PD according to the standard diagnosis (38).Meanwhile, imputation and quality control were carried out in the meta-analysis of the GWAS for PD, which yielded a total of 12,858,066 genotyping and imputation variants. SNPs associated with iron homeostasis were extracted from this data.Detailed information on participants, genotyping, imputation, and quality control can be found at the IEU database website.

2.3 Selection of instrumental variables

We performed a series of strict quality control on the instrumental variables so as to select valid instrumental variables. By obtaining snps (p < 5×10^− 8) related to iron homeostasis indicators (ferritin, serum iron, TIBC, and TSAT), a clustering process was performed on this basis (r2 < 0.001, cluster distance = 10000kb), thus excluding SNPs with similar effects that might have an influence on each other.At the same time, we excluded SNPs (PD; p < 5×10^− 8) associated with the outcome (PD; p < 5×10^− 8). Then, we further excluded SNPs associated with confounding factors. By reviewing the literature, we found that the risk factors associated with PD include smoking, alcohol consumption, pesticides, and uric acid levels (39–44). To ensure that the instrumental variables were strongly correlated with the exposure factors, we selected the instrumental variables with F > 10 as the valid instrumental variables by calculating the statistic F. The F statistic was calculated by the formula: F = R² (N-K-1)/K (1-R²), and the R² was calculated by the formula: R² = 2 × MAF × (1-MAF) Beta². To ensure that the SNPs had an effect on the results only through the exposure factors, we excluded SNPs with moderate allele frequencies (45). Finally, when SNPs were not available from the GWAS results, we identified surrogate SNPs for MR analysis through the LD Link online platform ( https://ldlink.nci.nih.gov/ ).

2.4 Mendelian randomization analysis

The two-sample Mendelian randomization analysis of the relationship between iron homeostasis and Parkinson's disease was performed using the "TwoSampleMR" package in the R language software (version 4.2.1). Inverse variance weighting (IVW) with random effects was used as the main method, while MR Egger, weighted median (WM), simple modal and weighted modal were used as supplementary methods. Among them, the IVW method can provide a more stable, accurate and reliable estimation of the causal relationship between exposure factors and outcome factors by combining the Wald value of each instrumental variable through meta-analytic methods (46–47), so the results of IVW analysis are considered as the main results of MR. In addition, Egger's regression method, as a complementary method, can be validated and assessed for multiple validity through its intercepts. The weighted median analysis method allows the median of the distribution function obtained by ranking all SNP effect values according to their weights, and WM can obtain relatively robust estimates when at least 50% of the information comes from valid instrumental variables. Simple modal and weighted modal are usually used as auxiliary analysis methods due to their inferior evaluation ability compared to IVW (48).

2.5 Sensitivity analysis

We performed sensitivity analysis on the results of causality between iron homeostasis-related indicators and Parkinson's disease, which mainly included horizontal pleiotropy, heterogeneity test and Leave-one-out method (49–52). Horizontal pleiotropy was assessed using the MR- Egger method, which indicated the absence of genetic pleiotropy when the regression intercept was not zero and P > 0.05. The Cochran Q-test for P-value was used to assess heterogeneity among individual SNPs; if P > 0.05, indicating no heterogeneity, the fixed IVW method was considered as the primary method; otherwise, the random-effects IVW method was used. The leave-one-out method was used to observe the changes in the results of the remaining SNPs by progressively removing each SNP; if the overall error line did not change much after the removal of a SNP, the results were more reliable, and conversely, the results were unreliable.

3.1 Association estimation of SNPs associated with iron homeostasis and Parkinson's disease in a genome-wide sense

According to the screening criteria of instrumental variables in this study, a total of 56 SNPs associated with iron homeostasis and Parkinson's disease in a genome-wide sense were finally screened out through the coordinated analysis of 86 SNPs associated with iron homeostasis markers and 12,858,066 SNPs associated with PD, among which 12 SNPs were associated with Parkinson's disease in total iron binding capacity, 7 SNPs were associated with transferrin saturation, 25 SNPs were associated with ferritin, 12 were associated with serum iron, and 7 were associated with parkinson's disease. SNPs, 25 SNPs associated with ferritin and 12 SNPs associated with serum iron (Table 1) .

Table 1

Iron homeostasis and Parkinson's disease-related SNPs
NO.	Expose	SNP	β	SE	P	NO.	Expose	SNP	β	SE	P
1	TIBC	rs112727702	-2.64883721	1.6511628	0.10866439	10	Ferritin	rs1800562	1.5984615	0.9023077	0.076473273
2	TIBC	rs1132274	-0.35277778	2.3527778	0.88081119	11	Ferritin	rs2529440	0.1742857	1.82	0.923710089
3	TIBC	rs12693541	0.85208333	2.1333333	0.68958813	12	Ferritin	rs34523089	2.4681159	1.1971014	0.039232178
4	TIBC	rs174546	2.80217391	1.3652174	0.04011646	13	Ferritin	rs36184164	-5.3722222	2.9722222	0.070687994
5	TIBC	rs1800562	-0.46177778	0.2606667	0.07647327	14	Ferritin	rs4841429	2.67	2.0066667	0.183332256
6	TIBC	rs3817672	-0.0483871	1.9451613	0.98015417	15	Ferritin	rs4938939	1.2681818	3.0136364	0.673890558
7	TIBC	rs4854760	0.06882353	0.185	0.70987865	16	Ferritin	rs57659670	0.3764286	0.8428571	0.655156552
8	TIBC	rs57659670	-0.68441558	1.5324675	0.65515655	17	Ferritin	rs601338	-1.7928571	2.1357143	0.401208487
9	TIBC	rs59950280	-1.80606061	1.9787879	0.36139479	18	Ferritin	rs6025	1.3026667	1.422	0.359624602
10	TIBC	rs6025	-2.10107527	2.2935484	0.3596246	19	Ferritin	rs6029148	1.25	2.8695652	0.663122573
11	TIBC	rs855791	-2.71923077	2.3230769	0.24178783	20	Ferritin	rs6029148	1.39375	2.06875	0.50049188
12	TIBC	rs9399136	-3.51818182	2.0575758	0.08729039	21	Ferritin	rs708686	-5.6935484	2.1774194	0.008927584
1	TSAT	rs13007705	-1.16060606	1.9545455	0.55264693	22	Ferritin	rs7865362	-3.88	2.736	0.156153207
2	TSAT	rs3817672	0.05769231	2.3192308	0.98015417	23	Ferritin	rs855791	1.6068182	1.3727273	0.241787829
3	TSAT	rs4854760	-0.24375	0.6552083	0.70987865	24	Ferritin	rs9921222	-0.244	2.452	0.920732869
4	TSAT	rs57659670	0.90862069	2.0344828	0.65515655	25	Ferritin	rs996347	1.4061224	1.4040816	0.316607616
5	TSAT	rs7385804	-0.25	0.9967742	0.80196174	1	IRON	rs12718598	2.1185185	2.1518519	0.32486508
6	TSAT	rs855791	0.41588235	0.3552941	0.24178783	2	IRON	rs13007705	-1.3206897	2.2241379	0.55264693
7	TSAT	rs9399136	1.73283582	1.0134328	0.08729039	3	IRON	rs1800562	0.7696296	0.4344444	0.07647327
1	Ferritin	rs10801913	2.3333333	2.8916667	0.419714634	4	IRON	rs2228145	-0.7846154	2.3307692	0.73639308
2	Ferritin	rs1131262	-4.584375	3.115625	0.141179124	5	IRON	rs35945185	-1.2032258	1.9548387	0.53821694
3	Ferritin	rs12419620	1.1	2.6967742	0.683350957	6	IRON	rs4854760	0.4415094	1.1867925	0.70987865
4	Ferritin	rs1260326	1.988	2.44	0.41521308	7	IRON	rs57659670	1.2547619	2.8095238	0.65515655
5	Ferritin	rs12693541	-0.5177215	1.2962025	0.689588134	8	IRON	rs7385804	-0.2719298	1.0842105	0.80196174
6	Ferritin	rs12807014	-1.9103448	2.2827586	0.402672388	9	IRON	rs7630745	2.004	2.544	0.43085123
7	Ferritin	rs13253974	-3.3583333	2.7666667	0.224802938	10	IRON	rs77262773	-1.9703704	2.5617284	0.44180034
8	Ferritin	rs1542752	-3.1705882	2.4176471	0.189710664	11	IRON	rs855791	0.4158824	0.3552941	0.24178783
9	Ferritin	rs17476364	-1.9372093	2.3930233	0.418213913	12	IRON	rs9399136	2.0368421	1.1912281	0.08729039

3.2 Results of Mendelian randomization analysis

The results of IVW analysis suggested that TIBC ( β= -0.142; 95%Cl = -0.197 to 0.481; P = 0.4138421 ), TSAT ( β= -0.142; 95%Cl = -0.197 to 0.481; P = 0.4138421 ), ferritin ( β= -0.387 ; 95%Cl = -1.179 to 0.405; P = 0.33783807 ) were not genetically causally associated with PD, and iron ( β= -0.524; 95%Cl = -0.046 to -0.002; P = 0.03191512 ) was considered to be genetically causally associated with PD. The results of the MR Egger's analysis method suggests that ferritin may be genetically causally associated with PD (p < 0.05), the results of weighted median method suggested that ferritin might be genetically causally associated with PD (p < 0.05), the results of simple model suggested that the four indicators were not causally associated with PD (p > 0.05), and the results of weighted model suggested that ferritin might be genetically causally associated with PD (p < 0.05). The results are shown in Table 2.Scatter plots were used to visualize the results, in which the regression lines for inverse variance weighted (IVW), weighted median, MR-Egger, simple mode and weighted mode are shown in Fig. 2(a)(b)(c)(d). The forest plot suggested that of the iron homeostasis metrics, only the red line of the iron index was located on the "0" side, indicating that iron had a significant effect on PD (Fig. 3a, b, c, d ).

Table 2

Results of MR analysis of iron homeostasis and Parkinson's disease
Exposure	Method	nSNP	β	95%Cl	P-val
TIBC	MR Egger	12	0.032	-0.397,0.461	0.8872405
	Weighted median	12	-0.005	-0.354,0.344	0.9766492
	Inverse variance weighted	12	0.142	-0.197,0.481	0.4138421
	Simple mode	12	0.465	-0.854,1.784	0.5041971
	Weighted mode	12	0.109	-0.199,0.417	0.5027566
TSAT	MR Egger	7	-0.492	-1.505,0.521	0.3845577
	Weighted median	7	-0.302	-0.927,0.323	0.3440717
	Inverse variance weighted	7	-0.316	-0.861,0.229	0.2553290
	Simple mode	7	0.088	-0.010,1.186	0.8806408
	Weighted mode	7	-0.367	-0.996,0.262	0.2973331
FERRITIN	MR Egger	25	-1.735	-3.125,-0.345	0.02251811
	Weighted median	25	-1.272	-2.252,-0.272	0.01101036
	Inverse variance weighted	25	-0.387	-1.179,0.405	0.33783807
	Simple mode	25	-1.348	-2.926,0.230	0.10717302
	Weighted mode	25	-1.229	-2.366,-0.092	0.04455122
IRON	MR Egger	12	-0.612	-1.306,0.082	0.11514440
	Weighted median	12	-0.432	-1.014,0.150	0.14488879
	Inverse variance weighted	12	-0.524	-0.046,-0.002	0.03191512
	Simple mode	12	-0.749	-1.766,0.268	0.17726917
	Weighted mode	12	-0.527	-1.070,0.016	0.08400243

3.3 Sensitivity analysis of Mr results

The Cochran's Q test of MR-IVW suggested that TIBC ( P = 0.1618872), TSAT ( P = 0.7099448), ferritin ( P = 0.09768154), and iron ( P = 0.8557510) did not have any results of the Mr analysis with PD heterogeneity, and MR Egger's test showed that the Mr analysis results of TIBC ( P = 0.1540043), TSAT ( P = 0.6096557), ferritin ( P = 0.24083554), and iron ( P = 0.8033798) suggested no significant heterogeneity at the same time. The funnel plots (Fig. 4a, b, c, d ) showed that the SNPs of the four indicators, TIBC, TSAT, ferritin, and iron, were relatively equally distributed on both sides of the straight line.The results of the MR Egger's level of pleiotropy test showed that, except for ferritin ( P = 0.0353461 ), TIBC ( P = 0.4167759 ), TSAT ( P = 0.7032719 ), and iron ( P = 0.7403369 ) were without horizontal pleiotropy.MR-PRESSO global test showed that TIBC ( P = 0.404 ), TSAT ( P = 0.759 ), ferritin ( P = 0.113 ), and iron ( P = 0.87 ) did not detect the presence of horizontal pleiotropy (Table 3).

The Leave-one-out method was used to visualize the results of the IVW method, and by sequentially eliminating individual SNPs, we can see that the distributions of the remaining SNPs of the four iron homeostasis indexes, TIBC, TSAT, ferratin, and iron, are still close to the positions marked in red in the figure, and the overall results have not been significantly affected, which suggested that the results of the IVW analysis were stable and reliable ( Fig. 5a, b, c, d ).

Table 3

Sensitivity analysis results of MR in iron homeostasis and Parkinson's disease
Expose	Outcome	Heterogeneity test		Pleiotropy test	Mr presso test
		Cochran’s Q test(P value)	Rucker’s Q test(P value)	Egger intercept(P value)	Main MR	Mr presso test
		Inverse variance weighted	MR Egger	MR Egger	P value	P value
TIBC	PD	0.1618872	0.1540043	0.4167759	0.4311849	0.404
TSAT	PD	0.7099448	0.6096557	0.7032719	0.2004497	0.759
FERRITIN	PD	0.09768154	0.24083554	0.0353461	0.3473946	0.113
IRON	PD	0.8557510	0.8033798	0.7403369	01594058	0.87

This study is the first to explore the causal relationship between iron homeostasis and Parkinson's disease. All the data in this study were obtained from genome-wide association summary data. SNPs related to iron homeostasis indicators were selected after strict quality control, while imputation and quality control were performed in a meta-analysis of GWAS sources for PD. We obtained a total of 12,858,066 PD-associated genotyping and imputation variants, on the basis of which we extracted SNPs associated with iron homeostasis. Finally, we performed Mendelian randomization and sensitivity analyses, and the results suggested that there was no significant causal association between TIBC, TSAT, and ferritin and PD ( IVW ; P ༞ 0.05), while there was a genetic causal association between iron and PD ( β = -0.524; 95%Cl = -0.046,-0.002; P value = 0.03191512 ༜0.05), and increased iron levels may reduce the risk of PD. The results of the sensitivity analysis showed that there was no heterogeneity and pleiotropy ( P < 0.05) in the results of the Mr analysis of the four indicators with PD, suggesting that the results were relatively stable and reliable. Therefore, based on the study of genome-wide significance of iron homeostasis and Mendelian randomization in Parkinson's disease, we concluded that among the iron homeostasis indices, there was no causal association between TIBC, TSAT, ferritin and PD, and there was a genetically causal association between serum iron and PD.

Iron is an abundant element in the human body, which is involved in many physiological processes in the body including neurotransmitter delivery, O2 transport and storage, DNA synthesis, and cellular metabolism (53–54). And iron homeostasis, as an important organismal regulatory mechanism, plays an important role in the physiological and pathological processes of the nervous system, especially in certain degenerative neurological diseases such as PD and AD. Recent evidence suggests that iron metabolism is involved in key proteins in the pathogenesis of PD, suggesting that imbalance of iron homeostasis is an important mechanism in the development of PD (55). However, the role of iron in the pathogenesis and progression of PD has been controversial. Iron has long been recognized as one of the risk factors for PD.Thomas(56) et al. found that localized magnetic susceptibility of the brain was significantly increased in patients with Parkinson's disease, possibly due to the accumulation of cerebral iron in the frontal, cingulate, and insular cortices, while higher levels of gene expression related to heavy metal detoxification and synaptic function were found in the corresponding sites, suggesting that cerebral iron involvement may be one of the mechanisms that lead to regression in Parkinson's disease mechanism. Jiang(57) et al. concluded that specific iron accumulation exists in the substantia nigra of the brains of Parkinson's disease patients and induces apoptosis of dopamine neurons, and presented evidence for the use of iron chelators as a therapeutic option for the treatment of PD. It has been found that iron can generate toxic reactive oxygen species (ROS) by reacting with hydrogen peroxide, and in patients with Parkinson's disease, the latter can further induce oxidative stress by depleting endogenous antioxidants, which is one of the important mechanisms involved in the development of PD (58–60). Prasuhn(61) et al. have found that cerebral iron deposition can be highly predictive of mitochondrial damage in vivo, suggesting an association between brain iron deposition and mitochondrial damage in vivo. Zeng et al.(62) found that iron overload could lead to neuronal cell death in a PD model by activating apoptosis and iron death pathways, whereas the use of iron chelators could play a role in protecting neurons by inhibiting iron death.

However, studies in recent years have suggested that serum iron levels may be protective against the development of PD, which is also consistent with the results of the present study (63–64). Nigrostriatal echoes in the brain of PD patients were found to be negatively related to markers of elevated serum iron levels (65).Logroscino et al.(67) found that male patients experiencing multiple blood donations may be at an increased risk of PD due to depletion of iron stores in the body.Savica et al.(67) found that anemia may increase the risk of PD at an early stage of life, by using anemia as a surrogate marker of iron deficiency (67). Xu et al.(68) found that patients with non-motor symptoms of PD had significantly lower serum iron levels and significantly higher serum transferrin levels.(69) Xia et al.(69) found that high serum iron associated with the C282Y mutation in the hemochromatosis gene, HFE, reduced the risk of developing PD. In fact, there is uncertainty regarding the protective mechanism of reduced serum iron and PD, and it has been suggested that reduced peripheral serum iron levels may have an effect on dopamine neuron development by affecting tyrosine hydroxylase (TH) synthesis (70–71). In addition, a meta-analysis study found (72) that no significant differences in iron and copper levels were observed in serum, plasma, and cerebrospinal fluid of patients with Parkinson's disease, but increased transferrin saturation suggests the importance of the simultaneous incorporation of iron homeostasis-related indices into the study.

Whereas ferritin, as an important iron storage protein, plays an important role in iron transport, storage, and metabolism, the present study did not find an association between serum ferritin and PD, which is in agreement with the results of some studies (73–75), and previous studies have also found increased serum ferritin levels in PD patients (76–77). There is also no unified conclusion on this aspect. In recent years, studies have shown that there is no significant difference in TIBC and TSAT levels in PD patients compared to the controls (78), and it has also been suggested that TIBC is negatively correlated with the severity of PD (79). The existence of a significant association between TIBC, TSAT and PD was not found in our study, and these inconsistent results suggest the complexity of the mechanisms involved, and more and more in-depth studies are needed in the future.

Mendelian randomization research starts from the gene aspect by selecting candidate genes and genome-wide association study (GWAS) databases, and introduces instrumental variables that have strong associations with exposure factors, namely, SNPs. Since the alleles of SNPs are assigned to an individual before affected by the exposure or the outcome, which is immutable, and can exclude the influence of confounding factors brought about by the transcription process of the cell. Therefore, the association between exposure and outcome can be assessed more accurately and even use it to validate the results of RCTs.As a prospective study, RCT is an important research method for assessing causality, but it is not applicable to all research situations due to the difficulty of its implementation and the moral and ethical issues it sometimes involves (32–33). In contrast, in observational studies, due to the uncontrollable nature of confounders, it is often not possible to provide authentic causal association results between exposure and outcome by regression methods (80). Therefore, the use of Mendelian randomization analysis not only can control the effects of confounders and reverse causality, but also has better feasibility and high assessment value. Thus, exploring the association between iron homeostasis and PD based on this approach is expected to obtain relatively stable results from a genetic perspective. However, previous genetic studies focusing on this aspect are rare. A decade-old Mendelian randomization study showed an association between an increase in serum iron and a reduced risk of developing Parkinson's disease (81). The genetic association of other indicators of iron homeostasis, including TIBC, TSAT, and ferritin, with the disease remains to be further investigated.

This study used Mendelian randomization analysis to investigate the association between iron homeostasis and Parkinson's disease in order to explore the genetic evidence for the association. However, this study still has limitations, firstly, the database in this study only involves patients of European origin, and does not comprehensively involve patients from various countries and regions around the world, and although the most recent indexes related to iron homeostasis are currently used, the relevant indexes are still relatively small, and the instrumental variables used for the study are still far from being sufficient, and in addition, the database on Parkinson's disease does not involve a large enough sample size of cases, all of which might have an impact on the results of the study. This requires the incorporation of a larger and more comprehensive sample size and the use of more indicators for further validation and measurement in the future to make the study results more reliable and credible.

In conclusion, by studying the causal relationship between iron homeostasis indicators and Parkinson's disease at the genetic level using Mendelian randomization analysis, it was found that among the four iron homeostasis indicators, TIBC, TSAT, and ferritin had no causal relationship with Parkinson's disease. Moreover, serum iron is genetically associated with Parkinson's disease, and increased serum iron levels may reduce the risk of the development of PD, suggesting that impaired iron metabolism plays an important role in the pathogenesis of Parkinson's disease. In the future, it is expected that further studies on Parkinson's disease will be conducted for other iron homeostasis indexes, with a view to providing some ideas and basis for the pathomechanism and clinical diagnosis and treatment of Parkinson's disease.

Acknowledgement

We appreciate the generosity and contribution of all participants and relevant GWAS databases.

Funding

This study was funded by the National Natural Science Foundation of China (Grant No. 82274493).

Data availability

The original contributions proposed in this study are included in the article/supplementary materials, and further inquiries can be directly addressed to the corresponding authors.

Author contributions

HC designed the study. XW and ZC wrote the analysis plan, performed the data analysis, and interpreted the results.HC wrote the first draft of the manuscript with critical feedback and revisions from JZ and DX, and all authors have read and approved the final manuscript.

Ethics approval

Only publicly available GWAS summary statistics were used in this study and therefore ethical approval was not required. The relevant GWAS researchers have obtained their respective ethical approvals.

Conflict of interest

They do not have known competitive economic interests or personal relationships, which may affect the work reported in this article.

Parkinson J. An essay on the shaking palsy. J Neuropsychiatry Clin Neurosci 2002;14:223–36.
Kempster PA, Hurwitz B, Lees AJ. A new look at James Parkinson’s essay on the shaking palsy. Neurology 2007;69:482–5
Kalia LV, Lang AE (2015) Parkinson’s disease. Lancet, 386(9996):896–912. doi:10.1016/S0140-6736(14)61393-3
Bjorklund A, Dunnett SB. Dopamine neuron systems in the brain: an update. Trends Neurosci 2007;30:194–202.
Hornykiewicz O. The discovery of dopamine deficiency in the parkinsonian brain. J Neural Transm 2006;70:9–15
Braak H, Del Tredici K, Rub U, de Vos RA, Jansen Steur EN, and Braak E. Staging of brain pathology related to sporadic Parkinson’s disease. Neurobiol Aging 24: 197–211, 2003.
Lücking C, Dürr A, Bonifati V, Vaughan J, De Michele G, Gasser T, et al. Association between early-onset Parkinson’s disease and mutations in the parkin gene. N Engl J Med. (2000) 342:1560–7. doi: 10.1056/nejm200005253422103
Lill CM (2016) Genetics of Parkinson’s disease. Mol Cell Probes 30(6):386–396. doi:10.1016/j.mcp.2016.11.001
McNaught KSP, Jenner P, Olanow CW. Protein mishandling: Role of the ubiquitin proteasome system in the pathogenesis of Parkinson’s disease. In: Jankovic J, Tolosa E, eds. Parkinson’s disease and movement disorders. Philadelphia: Lippincott Williams and Wilkins, 2007:33–49.
Pan T, Kondo S, Le W, et al. The role of autophagy-lysosome pathway in neurodegeneration associated with Parkinson’s disease. Brain 2008 (Epub ahead of print).
Deng H, Jankovic J, Guo Y, Xie W, and Le W. Small interfering RNA targeting the PINK1 induces apoptosis in dopaminergic cells SH-SY5Y. Biochem Biophys Res Commun 337: 1133–1138, 2005.
Schroder M and Kaufman RJ. ER stress and the unfolded protein response. Mutat Res 569: 29–63, 2005.
Schroder M and Kaufman RJ. The mammalian unfolded protein response. Annu Rev Biochem 74: 739–789, 2005.
Selkoe DJ. Folding proteins in fatal ways. Nature 426: 900–904, 2003.
Hernández-Romero M, Delgado-Cortés M, Sarmiento M, de Pablos R, Espinosa Oliva A, Argüelles S, et al. Peripheral inflflammation increases the deleterious effffect of CNS inflammation on the nigrostriatal dopaminergic system. Neurotoxicology. (2012) 33:347–60. doi: 10.1016/j.neuro.2012.01.018
Villarán R, Espinosa-Oliva A, Sarmiento M, De Pablos R, Argüelles S, DelgadoCortés M, et al. Ulcerative colitis exacerbates lipopolysaccharide-induced damage to the nigral dopaminergic system: potential risk factor in Parkinson‘s disease. J Neurochem. (2010) 114:1687–700. doi: 10.1111/j.1471-4159.2010.06879.x
Muckenthaler MU, Rivella S, Hentze MW, Galy B. A red carpet for iron metabolism. Cell. (2017) 168:344–61. doi: 10.1016/j.cell.2016.12.034
Hentze MW, Muckenthaler MU, Andrews NC. Balancing acts: molecular control of mammalian iron metabolism. Cells. (2004) 117:285–97. doi: 10.1016/ s0092-8674(04)00343-5
MacKenzie, E.L. et al. (2008) Intracellular iron transport and storage: from molecular mechanisms to health implications. Antioxid. Redox Signal. 10, 997–1030
Hentze, M.W. et al. (2010) Two to tango: regulation of mammalian iron metabolism. Cell 142, 24–38
Ganz, T. and Nemeth, E. (2012) Hepcidin and iron homeostasis. Biochim. Biophys. Acta 1823, 1434–1443
Ganz T, Nemeth E. Hepcidin and iron homeostasis. Biochim Biophys Acta. (2012) 1823:1434–43. doi: 10.1016/j.bbamcr.2012.01.014
Weiss G, Ganz T, Goodnough LT. Anemia of inflammation. Blood. (2019) 133:40–50. doi: 10.1182/blood-2018-06-856500
Mochizuki H, Choong CJ, Baba K. Parkinson's disease and iron. J Neural Transm (Vienna). 2020;127(2):181-187. doi:10.1007/s00702-020-02149-3
Ma H, Dong Y, Chu Y, Guo Y, Li L. The mechanisms of ferroptosis and its role in alzheimer's disease. Front Mol Biosci. 2022;9:965064. Published 2022 Aug 26. doi:10.3389/fmolb.2022.965064
Stankiewicz JM, Neema M, Ceccarelli A. Iron and multiple sclerosis. Neurobiol Aging. 2014;35 Suppl 2:S51-S58. doi:10.1016/j.neurobiolaging.2014.03.039
Guo J, Tuo QZ, Lei P. Iron, ferroptosis, and ischemic stroke. J Neurochem. 2023;165(4):487-520. doi:10.1111/jnc.15807
Dexter, D. T. et al. Increased nigral iron content and alterations in other metal ions occurring in brain in Parkinson’s disease. J. Neurochem. 52, 1830–1836 (1989).
Sofific, E., Paulus, W., Jellinger, K., Riederer, P. & Youdim, M. B. Selective increase of iron in substantia nigra zona compacta of Parkinsonian brains. J. Neurochem. 56, 978–982 (1991).
Duran, R. et al. Oxidative stress and aminopeptidases in Parkinson’s disease patients with and without treatment. Neurodegener. Dis. 8, 109–116 (2011).
Agil, A. et al. Plasma lipid peroxidation in sporadic Parkinson’s disease. Role of the L-dopa. J. Neurol. Sci. 240, 31–36 (2006).
Lawlor DA, Harbord RM, Sterne JAC, Timpson N, Davey Smith G: Mendelian randomization: Using genes as instruments for making causal inferences in epidemiology. Stat Med 27: 1133–1163, 2008
Evans DM, Davey Smith G: Mendelian randomization: New applications in the coming age of hypothesis-free causality. Annu Rev Genomics Hum Genet 16: 327–350, 2015
Greenland S: An introduction to instrumental variables for epidemiologists. Int J Epidemiol 29: 722–729, 2000
Thomas DC, Conti DV: Commentary: The concept of ‘Mendelian randomization’. Int J Epidemiol 33: 21–25, 2004
Bell S, Rigas AS, Magnusson MK, Ferkingstad E, Allara E, Bjornsdottir G, et al. A genome-wide meta-analysis yields 46 new loci associating with biomarkers of iron homeostasis. Commun Biol. (2021) 4:156. doi: 10.1038/s42003-020-01575-z
Alfradique-Dunham I, Al-Ouran R, von Coelln R, et al. Genome-Wide Association Study Meta-Analysis for Parkinson Disease Motor Subtypes. Neurol Genet. 2021;7(2):e557. Published 2021 Jan 28. doi:10.1212/NXG.0000000000000557
Tolosa E, Wenning G, Poewe W. The diagnosis of Parkinson's disease. Lancet Neurol. 2006;5(1):75-86. doi:10.1016/S1474-4422(05)70285-4
Ascherio A, Schwarzschild MA. The epidemiology of Parkinson's disease: risk factors and prevention. Lancet Neurol. 2016;15(12):1257-1272. doi:10.1016/S1474-4422(16)30230-7
Tysnes OB, Storstein A. Epidemiology of Parkinson's disease. J Neural Transm (Vienna). 2017;124(8):901-905. doi:10.1007/s00702-017-1686-y
Hernan MA, Zhang SM, Rueda-deCastro AM, Colditz GA, Speizer FE, Ascherio A. Cigarette smoking and the incidence of Parkinson’s disease in two prospective studies. Ann Neurol 2001; 50: 780–86.
Hernan MA, Chen H, Schwarzschild MA, Ascherio A. Alcohol consumption and the incidence of Parkinson’s disease. Ann Neurol 2003; 54: 170–75.
Baldi I, Cantagrel A, Lebailly P, et al. Association between Parkinson’s disease and exposure to pesticides in southwestern France. Neuroepidemiology 2003; 22: 305–10.
Parkinson Study Group SURE-PD3 Investigators, Schwarzschild MA, Ascherio A, et al. Effect of Urate-Elevating Inosine on Early Parkinson Disease Progression: The SURE-PD3 Randomized Clinical Trial. JAMA. 2021;326(10):926-939. doi:10.1001/jama.2021.10207
Sekula P, Del Greco M F, Pattaro C, Köttgen A. Mendelian Randomization as an Approach to Assess Causality Using Observational Data. J Am Soc Nephrol. 2016;27(11):3253-3265. doi:10.1681/ASN.2016010098
Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from mendelian randomization between complex traits and diseases. Nat Genet. 2018;50:693-698.
Bowden J, del Greco MF, Minelli C, Davey Smith G, Sheehan NA, Thompson JR. Assessing the suitability of summary data for twosample mendelian randomization analyses using MR-Egger regression: the role of the I2 statistic. Int J Epidemiol. 2016;45:1961-1974.
Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method [published correction appears in Eur J Epidemiol. 2017 Jun 29;:]. Eur J Epidemiol. 2017;32(5):377-389. doi:10.1007/s10654-017-0255-x
Birney E. Mendelian Randomization. Cold Spring Harb Perspect Med. 2022;12(4):a041302. Published 2022 May 17. doi:10.1101/cshperspect.a041302
Sekula P, Del Greco M F, Pattaro C, Köttgen A. Mendelian Randomization as an Approach to Assess Causality Using Observational Data. J Am Soc Nephrol. 2016;27(11):3253-3265. doi:10.1681/ASN.2016010098
Grover S, Del Greco M F, Stein CM, Ziegler A. Mendelian Randomization. Methods Mol Biol. 2017;1666:581-628. doi:10.1007/978-1-4939-7274-6_29
Rees JMB, Foley CN, Burgess S. Factorial Mendelian randomization: using genetic variants to assess interactions. Int J Epidemiol. 2020;49(4):1147-1158. doi:10.1093/ije/dyz161
Benarroch, E. E. Brain iron homeostasis and neurodegenerative disease. Neurology 72:1436–1440; 2009.
Crichton, R. R.; Dexter, D. T.; Ward, R. J. Brain iron metabolism and its perturbation in neurological diseases. J. Neural Transm. 118:301–314; 2011.
Singh N, Haldar S, Tripathi AK, et al. Brain iron homeostasis: from molecular mechanisms to clinical significance and therapeutic opportunities. Antioxid Redox Signal. 2014;20(8):1324-1363. doi:10.1089/ars.2012.4931
Thomas GEC, Zarkali A, Ryten M, et al. Regional brain iron and gene expression provide insights into neurodegeneration in Parkinson's disease. Brain. 2021;144(6):1787-1798. doi:10.1093/brain/awab084
Jiang H, Wang J, Rogers J, Xie J. Brain Iron Metabolism Dysfunction in Parkinson's Disease. Mol Neurobiol. 2017;54(4):3078-3101. doi:10.1007/s12035-016-9879-1
Lan, J.; Jiang, D. H. Excessive iron accumulation in the brain: a possible potential risk of neurodegeneration in Parkinson’s disease. J. Neural Transm 104:649–660; 1997.
Han, J.; Cheng, F. C.; Yang, Z.; Dryhurst, G. Inhibitors of mitochondrial respiration, iron(II), and hydroxyl radical evoke release and extracellular hydrolysis of glutathione in rat striatum and substantia nigra: potential implications to Parkinson’s disease. J. Neurochem 73:1683–1695; 1999.
Bharath, S.; Hsu, M.; Kaur, D.; Rajagopalan, S.; Andersen, J. K. Glutathione, iron and Parkinson’s disease. Biochem. Pharmacol. 64:1037–1048; 2002.
Prasuhn J, Göttlich M, Gerkan F, et al. Relationship between brain iron deposition and mitochondrial dysfunction in idiopathic Parkinson's disease. Mol Med. 2022;28(1):28. Published 2022 Mar 4. doi:10.1186/s10020-021-00426-9
Zeng X, An H, Yu F, et al. Benefits of Iron Chelators in the Treatment of Parkinson's Disease. Neurochem Res. 2021;46(5):1239-1251. doi:10.1007/s11064-021-03262-9
No authors listed. Higher iron concentrations may protect against Parkinson's disease. BMJ. 2013;346:f3691. Published 2013 Jun 12. doi:10.1136/bmj.f3691
Berg D, Roggendorf W, Schroder U, Klein R, Tatschner T, et al. (2002) Echogenicity of the substantia nigra: Association with increased iron content and marker for susceptibility to nigrostriatal injury. Arch Neurol 59: 999–1005.
Walter U, Witt R, Wolters A, Wittstock M, Benecke R (2012) Substantia nigra echogenicity in Parkinson’s disease: Relation to serum iron and C-reactive protein. J Neural Transm 119: 53–57.
Logroscino G, Chen H, Wing A, Ascherio A (2006) Blood donations, iron stores, and risk of Parkinson’s disease. Mov Disord 21: 835–838.
Savica R, Grossardt BR, Carlin JM, Icen M, Bower JH, et al. (2009) Anemia or low hemoglobin levels preceding Parkinson disease: a case-control study. Neurology 73: 1381–1387.
Xu W, Zhi Y, Yuan Y, et al. Correlations between abnormal iron metabolism and non-motor symptoms in Parkinson's disease. J Neural Transm (Vienna). 2018;125(7):1027-1032. doi:10.1007/s00702-018-1889-x
Xia J, Xu H, Jiang H, Xie J. The association between the C282Y and H63D polymorphisms of HFE gene and the risk of Parkinson's disease: A meta-analysis. Neurosci Lett. 2015;595:99-103. doi:10.1016/j.neulet.2015.04.010
Ramsey AJ, Hillas PJ, Fitzpatrick PF (1996) Characterization of the active site iron in tyrosine hydroxylase. Redox states of the iron. J Biol Chem 271: 24395– 24400.
Beard J, Erikson KM, Jones BC (2003) Neonatal iron deficiency results in irreversible changes in dopamine function in rats. J Nutr 133: 1174–1179
Mariani S, Ventriglia M, Simonelli I, Donno S, Bucossi S, et al. (2013) Fe and Cu do not differ in Parkinson’s disease: a replication study plus meta-analysis. Neurobiol Aging 34: 632–633
Jiménez-Jiménez FJ, Alonso-Navarro H, García-Martín E, Agúndez JAG. Biological fluid levels of iron and iron-related proteins in Parkinson's disease: Review and meta-analysis. Eur J Neurol. 2021;28(3):1041-1055. doi:10.1111/ene.14607
Pretorius E, Swanepoel AC, Buys AV, Vermeulen N, Duim W, Kell DB. Eryptosis as a marker of Parkinson's disease. Aging (Albany NY). 2014;6(10):788-819. doi:10.18632/aging.100695
Farhoudi M, Taheraghdam A, Farid GA, Talebi M, Pashapou A, Majidi J and Goldust M. Serum iron and ferritin level in idiopathic Parkinson. Pakistan journal of biological sciences: PJBS. 2012; 15:1094‐1097.
Madenci G, Bilen S, Arli B, Saka M and Ak F. Serum iron, vitamin B12 and folic acid levels in Parkinson's disease. Neurochemical research. 2012; 37:1436‐1441.
Wei Z, Li X, Li X, Liu Q, Cheng Y. Oxidative stress in Parkinson's disease: a systematic review and meta-analysis. Front Mol Neurosci. 2018;11:236.
Shen X, Yang H, Zhang D, Jiang H. Iron Concentration Does Not Differ in Blood but Tends to Decrease in Cerebrospinal Fluid in Parkinson's Disease. Front Neurosci. 2019;13:939. Published 2019 Sep 26. doi:10.3389/fnins.2019.00939
Deng, Q., Zhou, X., Chen, J., Pan, M., Gao, H., Zhou, J., et al. (2017). Lower hemoglobin levels in patients with parkinson’s disease are associated with disease severity and iron metabolism. Brain Res. 1655, 145–151. doi: 10.1016/ j.brainres.2016.11.007
Klungel OH, Martens EP, Psaty BM, Grobbee DE, Sullivan SD, Stricker BH, Leufkens HG, de Boer A: Methods to assess intended effects of drug treatment in observational studies are reviewed. J Clin Epidemiol 57: 1223– 1231, 2004
Pichler I, Del Greco M F, Gögele M, et al. Serum iron levels and the risk of Parkinson disease: a Mendelian randomization study [published correction appears in PLoS Med. 2013;10(6):e1001462. doi:10.1371/journal.pmed.1001462

Download PDF

Journal Publication

published 01 Jul, 2024

Read the published version in Journal of Trace Elements in Medicine and Biology →

Version 1

posted

You are reading this latest preprint version

Evidence for genetic causality between iron homeostasis and Parkinson's disease a two-sample Mendelian randomization study

Status:

Journal Publication

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials and Methods

2.1 Study design

2.2 Data sources

2.3 Selection of instrumental variables

2.4 Mendelian randomization analysis

2.5 Sensitivity analysis

3 Results

3.1 Association estimation of SNPs associated with iron homeostasis and Parkinson's disease in a genome-wide sense

3.2 Results of Mendelian randomization analysis

3.3 Sensitivity analysis of Mr results

4. Discussion

5. Summary

Declarations

References

Status:

Journal Publication

Version 1