Next Article in Journal
Organizational Factors Moderating Changes in Tobacco Use Dependence Care Delivery Following a Comprehensive Tobacco-Free Workplace Intervention in Non-Profit Substance Use Treatment Centers
Next Article in Special Issue
Seasonal and Spatial Variation in Dissolved Heavy Metals in Liaodong Bay, China
Previous Article in Journal
Impact of COVID-19 on Life of Students: Case Study in Hong Kong
Previous Article in Special Issue
The Association of Renal Function and Plasma Metals Modified by EGFR and TNF-α Gene Polymorphisms in Metal Industrial Workers and General Population
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
IJERPH
Volume 18
Issue 19
10.3390/ijerph181910484
ijerph-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The Mediation Effects of Aluminum in Plasma and Dipeptidyl Peptidase Like Protein 6 (DPP6) Polymorphism on Renal Function via Genome-Wide Typing Association

by
Ting-Hao Chen
1,
Chen-Cheng Yang
2,
Kuei-Hau Luo
1,
Chia-Yen Dai
2,
Yao-Chung Chuang
3 and
Hung-Yi Chuang
1,2,*
1
Department of Public Health and Environmental Medicine, Research Center for Environmental Medicine, Kaohsiung Medical University, Kaohsiung 807, Taiwan
2
Department of Occupational and Environmental Medicine, Kaohsiung Medical University Hospital, Kaohsiung 807, Taiwan
3
Department of Neurology, Kaohsiung Chang Gung Memorial Hospital, Kaohsiung 833, Taiwan
*
Author to whom correspondence should be addressed.
Int. J. Environ. Res. Public Health 2021, 18(19), 10484; https://doi.org/10.3390/ijerph181910484
Submission received: 23 September 2021 / Accepted: 3 October 2021 / Published: 6 October 2021
(This article belongs to the Special Issue Evaluation and Health Impact of Toxic Metals Pollution)
Browse Figures
Versions Notes

Abstract

:
Aluminum (Al) toxicity is related to renal failure and the failure of other systems. Although there were some genome-wide association studies (GWAS) in Australia and England, there were no GWAS about Han Chinese to our knowledge. Thus, this research focused on using whole genomic genotypes from the Taiwan Biobank for exploring the association between Al concentrations in plasma and renal function. Participants, who underwent questionnaire interviews, biomarkers, and genotyping, were from the Taiwan Biobank database. Then, we measured their plasma Al concentrations with ICP-MS in the laboratory at Kaohsiung Medical University. We used this data to link genome-wide association (GWA) tests while looking for candidate genes and associated plasma Al concentration to renal function. Furthermore, we examined the path relationship between Single Nucleotide Polymorphisms (SNPs), Al concentrations, and estimated glomerular filtration rates (eGFR) through the mediation analysis with 3000 replication bootstraps. Following the principles of GWAS, we focused on three SNPs within the dipeptidyl peptidase-like protein 6 (DPP6) gene in chromosome 7, rs10224371, rs2316242, and rs10268004, respectively. The results of the mediation analysis showed that all of the selected SNPs have indirectly affected eGFR through a mediation of Al concentrations. Our analysis revealed the association between DPP6 SNPs, plasma Al concentrations, and eGFR. However, further longitudinal studies and research on mechanism are in need. Our analysis was still be the first study that explored the association between the DPP6, SNPs, and Al in plasma affecting eGFR.

1. Introduction

Aluminum (Al), which is the third most common element on earth, is widely used in many daily necessities and energy development [1]. Even daily drinking water may contain a little bit of Al. However, as industries advance, humans demand more Al, and Al-related occupational diseases have increased gradually [2,3,4]. Al is an unnecessary element for the human body. There are many studies that indicate that Al toxicity would damage renal function [5,6] and other systems in humans [7,8,9,10]. Al is excluded through 95% kidney and 2% bile. However, Al toxicity and decline of renal function were associated with each other. Patients with renal failure were damaged by Al. On the other hand, chronic kidney disease decreased Al excretion and accumulated it in the human body. The metabolism of Al in the human body is not clear; however, some enzymes and genotypes have been studied [11].
A genome-wide association study showed that the most significant SNP for Al in serum was at chr8:9095620, which is near PPP1R3B, but the biological information of this SNP was unknown [11]. Moreover, an occupational epidemiology study indicated that exposure to metal particles could cause DNA methylation; in an unadjusted model, exposure to Al particles was associated with DNA methylation in tumor suppressor genes; however, after adjustment for smoking, age, and body mass index (BMI), Al was not associated with DNA methylation [12]. Some epidemiological studies indicated that there was no association between Al exposure and Long Interspersed Element-1 (LINE-1) methylation [13,14]. In addition, a Chinese study group investigated whether those with occupational exposure to Al would have mild cognitive dysfunction and global DNA methylation [15], which causes a significant decrease in global DNA methylation. Another occupational epidemiological study from China indicated that occupational exposure to Al would have lowered the Amyloid Precursor Protein (APP) gene-methylation [15]. On the other hand, gene expression had an important role in influencing the maximum tolerance of Al toxicity [16], and different single nucleotide polymorphisms (SNPs) also affected different gene expressions in an animal model [11,17]. To our knowledge, there is no relevant study about the Han Chinese population.
Using the genome-wide data of Taiwan’s Han population, we could provide some evidence to look for potentially susceptible SNPs of a southern Chinese population to Al toxicity and understand the relationship between Al levels and genetic variants. Therefore, the goal of this study was to explore potential SNPs that involve plasma Al-affected renal function.

2. Materials and Methods

2.1. Study Population

The subjects of this cross-sectional study consisted of 500 Taiwan Han Chinese subjects from Taiwan Biobanks [18,19], included 30–70-year-old healthy people and excluded those diagnosed with cancer, non-Taiwanese people, and patients who were hospitalized from inception to date. The purpose of the Taiwan Biobank data is to establish the baseline characteristics of the Taiwan population based on volunteers who genuinely cared about their own health status. The Taiwan Biobank consisted of a population-based databank and storage-plasma bank. The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMU-HIRB-E(I)-20150259, the initial date of approval: 6 January 2016). Approval was waived for individual consent forms due to de-identification in Taiwan Biobank data and specimens.

2.2. Questionnaires

Each subject was administered questionnaires about demographic information, personal health behavior, present medication, drug use, diet, female-related questions, family history, lifestyle, economic status, and personalized information about the living environment.

2.3. Laboratory Analysis

Biomarkers of the Taiwan Biobank were all analyzed in the Linkou Chang Gung Memorial Hospital. However, the plasma Al concentrations were measured in the laboratory at Kaohsiung Medical University by an inductively coupled plasma mass spectrometry (ICP-MS, Thermo Scientific XSERIES 2®, Waltham, MA USA). The resolution was set to 0.8 and 0.4 amu at 10% peak height, which complied with the typical Al analysis. Radio Corporation of America (RCA) cleaning standard was used on all equipment used in the laboratory. For sample preparation, 1% HNO3 was added to the plasma samples to make a 1:10 dilution, and then we waited for 10 min. For checking the high linearity, ICP-MS calibration standard solution (from Accu Standard, MES-04-1) was diluted to 0.1, 0.2, 0.5, 1, 2, 5, 10, 20, 50, 100, 200, 500, 1000, 2000, and 3000 ug/L to estimate the calibration curve, which showed a high correlation coefficient (r > 0.995). Before analyzing the unknown concentrations, we conducted quality assurance (QA) and quality control (QC) to ensure precision and accuracy. QA was used to analyze standard reference materials (SRMs). To ensure consistency of the laboratory test, we took random SRMs to conduct repeated analyses. Each result had to fit the reference between 90% and 110%. QC was to ensure the stability of the system by triple testing the SRM sample, of which the coefficient of variance (CV) should be less than 3%.

2.4. Genotyping

DNA samples were genotyped by a customized Axiom Taiwan Biobank Array Plate (Taiwan Biobank chip, TWB chip) to obtain 645,918 single-nucleotide polymorphisms (SNP). The TWB chip was based on Affymetrix databases, the HapMap Project [20], the 1000-Genome Project [21,22], and published GWAS research about the Han Chinese population in Beijing and Taiwan to pick up SNPs related to Taiwan’s Han population.
To increase quality control in the genotype data, we looked for the subjects with high missing genotyping rates and extreme heterozygosity rates. The sample was excluded if the missing genotype rate was >5%, and SNP were excluded if the missing rate was >5%. We used (N-O)/N to calculate the mean heterozygosity rate, where N was the number of non-missing genotypes and O was the observed number of homozygous genotypes. Individuals with more than 3 standard deviations from the mean heterozygosity rate indicated that the DNA sample was contaminated or inbred; therefore, we excluded the subjects with extreme heterozygosity rates. We estimated the exact p-value of the Hardy–Weinberg equilibrium (HWE), and SNPs were excluded if the p-value of HWE was < 10−6. Furthermore, we calculated identity by descent (IBD). To understand whether our sample was independent, we excluded individuals if IBD was > 0.1875. All quality controls were conducted using PLINK software (version 1.9) [23] and R language (version 3.4.1) [24].

2.5. Statistical Analysis

2.5.1. Descriptive Analysis

We used the mean ± SD and quartiles to represent continuous data and n (%) to represent categorical data. Blood measurements were log-transformed to generate normal distribution such as Al concentration. The MDRD equation was used to evaluate renal function [25]:
eGFR = 175• (Serum creatinine)−1.154 • (Age)−0.203 • 0.742 (if female)• 1.21 (if African American)
We used eGFR as the x-axis and Al in plasma as the y-axis to make the scatter plot with a regression line to understand the distribution of eGFR and Al levels (log-transformed) in plasma. All descriptive analyses were conducted with SAS software (version 9.4).

2.5.2. Genome-Wide Association (GWA)

Association testing for log-transformed Al was conducted in the linear regressions through an additive model: SNP types were coded by assigning 0, 1, and 2 for homozygous genotypes with 0 minor alleles, heterozygous genotypes with 1 minor allele, and homozygous genotypes with 2 minor alleles, respectively.
We performed the crude model for log-transformed Al to avoid Type I errors because of multiple comparisons. The Bonferroni Correction [26] was used to correct the p-value of multiple testing. The corrected p-value = 0.05/numbers of SNPs, therefore the corrected p-value was (0.05/645918) = 8 × 10−8 and the suggested p-value was 1 ×∙10−6. In addition, a Manhattan plot was used to visualize the results of association testing. The vertical axis was −log10 (p-value), where a higher location meant a smaller p-value, and the horizontal axis represented the numbers of the chromosomes, and each dot represented each SNP. Meanwhile, we generated a QQ-plot to check the quality of the GWA results, where the Y-axis represented the observed −log10 (p-value), X-axis meant the expected −log10 (p-value), and each dot represented each SNP. We used g-profiler [27] to finish gene annotation.
All association tests were performed with PLINK 1.9 [23], and R statistical software (version 3.4.1) [24]. The Manhattan plot was conducted by Haploview software [28,29].

2.5.3. Mediation Analysis

We performed a mediation analysis to examine if there was an indirect effect between SNPs and renal function (Figure 1).
According to the assumption of Baron et al. [30], there were three criteria that had to be met: first of all was the exposure to outcome association; the second was exposure to mediator association; the third was the mediator to outcome association. However, MacKinnon et al. indicated that even it did not follow the first assumption of Baron, there was still mediation [31]. Hence, we built three linear models to test whether to follow the assumption of Baron.
Y = b + b1 X1 + b2 X2 + ….+biXi
M = a + a1 X1
Y = r + rxXi + rmMi
To generate the normalized residual, we used log-transformed Al concentration and assigned 0, 1, and 2 for homozygous genotypes, heterozygous genotypes, and homozygous genotypes with 2 minor alleles to fit an additive model. The causal mediated effect was estimated using the product-of-coefficient method to generate the mediation coefficient [31,32,33,34].
We used the bootstrap method to estimate the indirect effect because bootstrap CIs did not require the assumption of normal distribution and did not require the sample size to be too large [35,36]. We used percentile bootstrap CIs to estimate the confidence interval for avoiding the type-I error. All mediation analyses were conducted with STATA 13.0 software, and a p-value <0.05 was considered statistically significant.

3. Results

3.1. Study Population and SNPs Typing

After collecting the questionnaires and blood samples, 500 subjects were involved in this study; however, there was a subject who rejected to participate in this study. In the beginning, there were 645,918 SNPs in the Taiwan Biobank chip (TWB chip) and 500 subjects (227 males and 223 females). As illustrated in Figure 1, there were 6 people whose heterozygosity rate was over ± 3 standard deviations and 29,619 variants whose call rate was <95%. There were 618 variants that did not follow the Hardy–Weinberg equilibrium. Finally, we conducted a further analysis with 494 subjects and 615,681 variants in which the total genotyping rate was 99.7649% (Figure 2).

3.2. Characteristics of Study Population

The demographic characteristics of the study population are shown in Table 1. The average age was 48.28 years old, the rate of people who had drunk for more than six months was about 7.9%, and the percentage of people who had smoking experience was 34.21%. Regarding the indication of renal function, blood urine nitrogen (BUN), creatinine, and uric acid (UA) were measured. Among 75% of participants who were measured, BUN was under 15.20 mg/dL. A total of 75% of subjects were under 0.91 mg/dL, and 75% of subjects were under 6.9 mg/dL, respectively. The mean (SD) values for BUN, creatinine, and UA were 13.06 (3.48), 0.78 (0.19), and 5.73 (1.56), respectively. Using creatinine levels to estimate the eGFR, the mean (SD) was 99.52 (22.00), and almost 75% of subjects were >85.03. As for the blood measurement, the mean (SD) values for HbA1c (%) were 5.66 (0.55). Almost 75% of the participants were under the normal range in blood measurement. In the indicator of total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and liver function (AST and ALT), the mean (SD) values were 197.72 (35.75), 53.79 (13.19), 125.52 (32.42), 23.38 (11.44), and 25.56 (23.36), respectively, which showed this was a relatively healthy population base. Regarding the Al concentration in plasma (μg/L), the mean value was 1.13 ± 0.80 (Table 1).

3.3. Association between eGFR and Al

Figure 3 showed the association between eGFR and Al (log-transformed). We used a smoothing curve (smooth function: loess, 65% of point to fit, df = 4, kernel: Epanechkinov) to reveal that eGFRs ranged normal (between 85 and 120) and had lower plasma Al levels.

3.4. Genome-Wide Association Test

We did not use the SNPs of the Y chromosome. In Figure 4, there were two loci beyond the significant level (p-value < 8 × 10−8) and three variants between the significant level and suggestive level (p-value < 1 × 10−6) in the genome-wide association test. The most significant SNP was rs9857275, which was an intron variant located in ZBTB38 (zinc finger and BTB domain containing 38), a protein encoded by this gene was associated with DNA methylation [37]. The second was rs764111946, which was a missense variant and found in chromosome 2, showed close association to the BAZ2B (bromodomain adjacent to zinc finger domain 2B) gene [38]. Within (or close to) the dipeptidyl peptidase-like protein 6 (DPP6) gene were three intron variants that were at the suggestive level (p < 10−6). However, the minor allele frequencies of SNPs (rs9857275, rs764111946) were too small (both MAF = 0.001), and the first 2 SNPs beyond the significant level were so rare that it was difficult to estimate the true effect of the rare variants based on the current sample size (minor allele frequency = 0.001). Hence, we focused on the variants with bigger MAF, and we decided to focus on DPP6 variants (rs10224371, rs2316242, rs10268004) using a medication analysis.

3.5. Mediation Analysis

3.5.1. rs10224371→ln(Al)→eGFR

As shown in Figure 5a, rs10223471, which was located in DPP6, was significantly correlated with ln(Al)(coefficient = 0.287, 95%CI = 0.186, 0.387) after controlling for potential confounders, and ln(Al) was correlated with eGFR (coefficient = 5.124, 95%CI = 0.971, 9.277). The evidence indicated that rs10223471 within DPP6 would lean to increased eGFR through a mediated effect of ln(Al) (coefficient = 1469, 95%CI = 0.349, 2.751); however, the direct effect was non-significant (coefficient = −1.814, 95%CI = −6.332, 3.424). This polymorphism only had an indirect effect. what it meant was that there was to for the exposure to Al which thus resulted in increased eGFR.

3.5.2. rs2316242→ln(Al)→eGFR

As shown in Figure 5b, the first path (rs2316242→ln(Al)) showed that ln(Al) would increase by 0.311 units when a minor T allele of rs2316242 existed, and ln(Al) would increase eGFR by 4.639 units (95%CI = 0.509, 8.768). A bootstrap with 3000 replications showed that every T allele of rs2316242 would cause an increase in eGFR by 1.443 units (95%CI = 0.320, 2.936) through a mediated effect of ln(Al). However, rs2313242 did not result in an increased eGFR directly (coefficient = 0.6848, 95%CI = −4.995, 7.446).

3.5.3. rs10268004→ln(Al)→eGFR

As shown in Figure 5c, rs10268004, within DPP6, would cause a significant increase in ln(Al) by 0.311 units (95%CI = 0.189, 0.433), and a unit of increase in ln(Al) would cause eGFR to increase by 4.64 units (95%CI = 0.509, 8.768). A bootstrap with 3000 replications suggested that rs10268004 would result in increased eGFR by 1.443 units (95%CI = 0.379, 2.964) through a mediated effect of ln(Al). However, no direct evidence indicated that rs10268004 could result in increased eGFR directly.

4. Discussion

Some studies have been based on subjects who had occupational exposure to Al [2,3,4], while in our study, subjects were from the Taiwanese population. Al is an unnecessary element for the human body. There were probably some differences that affected the element’s levels in blood, such as dietary habits, local customs, or ethnic differences. In addition, the element’s levels in whole blood, serum, or plasma were different, while too many impurities in the whole blood would cause the test results to be unstable; moreover, the differences between serum and plasma we34re the proteins that would bind with some special metal such as lead, mercury, arsenic, and nickel [39]. Al levels in plasma could be consistent with the reference value from France [40], which shows that there was no occupational exposure to Al. Thus, this research could be a representation of plasma Al levels associated with renal function in the Taiwanese population.
A GWAS technique was used to find out the potential SNPs of the interested phenotype; therefore, we used plasma Al levels as a phenotype to perform the association analysis. There were three SNPs found in the same linkage disequilibrium (LD) blocks, rs10224371, rs2316242, and rs10268004, within DPP6. This gene, DPP6, has been proven as a candidate gene for amyotrophic lateral sclerosis (ALS) in the patients with ALS against the non-ALS population [41,42,43] and could also be a potential factor that led to mental diseases, such as autism [44] and posttraumatic stress disorder (PTSD) [45]. Therefore, the protein encoded by DPP6 is associated with the regulation of the potassium ion transport channel and further causes neurodevelopmental impairment [46,47] and mental diseases [44,45]. In our findings, there were three SNPs within DPP6 with suggestively significant association levels of Al concentration in plasma. Based on the biological evidence, Al may replace iron ions and calcium ions in the brain and interfere with other trace elements [48], and the protein encoded by DPP6 regulates the potassium ion transport channel and causes abnormal synapse activity [46,47]. Esther Ng et al. found a potential variant within chr8:9095620, which was reported at PPP1R3B, and three SNPs were found in chr7:89168136, chr7:73882011, and chr7:136945107, which report genes located at ZNF804B, GTF2IRD1, and PTN. The differences in our findings were probably due to race and type of experimental sample; Esther Ng et al. used the serum to measure the Al levels, and the study population was from Uppsala, Sweden [11]. There are more than one-third of genome differences between the Han Chinese population and Caucasians. This may be the main reason that leads to inconsistencies between Esther Ng et al. and our study [11].
The previous studies indicated that metal toxicity caused renal tubular damage [39,49] and makes the body unable to excrete the metal. However, according to some studies, the ability of Al excretion for patients with renal failure was worse than the ability of the normal population [50]. In our research, the majority of the study population were volunteers who greatly care about their own health status. As illustrated in Figure 3, a U-shaped curve existed. When eGFR was between 90 and 110, Al levels in plasma were the lowest; however, this U-shaped curve supports the evidence that previous studies indicated [50].
In addition, the indirect effect of DPP6 was observed. This represented that DPP6 would have no effect on eGFR if there was no Al in human plasma. This proved that DPP6 was associated with some neurological diseases, including cognitive diseases and memory impairment [41,42,43,44,45,47,51]. Some studies also mentioned that exposure to Al would be a potential factor in the development of Alzheimer’s disease, but the causal relationship was still controversial [52,53,54,55]; however, our finding showed that there was an association between Al in plasma, DPP6, and eGFR. To our knowledge, there were few studies that explained the indirect effect of DPP6 on eGFR. Moreover, in our findings, three variants in DPP6 were within the same LD-blocks. Therefore, it may be interesting to further investigate the probable causal effect through deep sequencing and the gene expression of DPP6 on renal function through molecular biology experiments.
The first limitation of this study was its relatively small sample size. To detect the rare variant with enough statistical power, increasing the sample size is needed. Meanwhile, statistical power would be limited because of multiple-testing correction; therefore, only large-scale gene-versus-phenotype effects can be discovered. The second limitation was the study design. Since a cross-sectional study could not ensure a causal relationship, longer follow-up data is in need to evaluate the long-term impact of Al levels on renal function. Eventually, due to the low exposure to Al in the normal population, the data from occupationally exposed workers will provide more information for this kind of study.

5. Conclusions

To the best of our knowledge, this is the first GWA study to explore the association between candidate variants and Al concentration in plasma on the Han Chinese population. Furthermore, we found that DPP6 variants played a mediator between Al in plasma and renal function. Though, longitudinal data or data of occupational exposure are in need. Nevertheless, these findings can provide additional information and medical care for the decision making of occupational medicine.

Author Contributions

Conceptualization, T.-H.C. and H.-Y.C.; methodology, T.-H.C. and H.-Y.C.; validation, C.-C.Y., K.-H.L., and C.-Y.D.; formal analysis, T.-H.C.; investigation, Y.-C.C.; resources, H.-Y.C.; data curation, T.-H.C., C.-C.Y., and H.-Y.C.; writing—original draft preparation, T.-H.C.; writing—review and editing, H.-Y.C.; supervision, H.-Y.C.; funding acquisition, H.-Y.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Ministry of Science and Technology, grant number MOST108-2314-B-037-062, and Kaohsiung Medical University Hospital (KMUH109-9T05).

Institutional Review Board Statement

The study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the Institutional Review Board at Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20150259). Approval was waived for individual consent forms due to de-identification in the Taiwan Biobank data and specimens.

Informed Consent Statement

Patient consent was waived due to de-identification in Taiwan Bio-Bank data and specimen.

Data Availability Statement

The data from Taiwan Biobank in this study can be applied from the Taiwan Biobank at https://www.twbiobank.org.tw/new_web_en/about-export.php (accessed on 20 August 2021).

Acknowledgments

This work was supported partially by the Research Center for Environmental Medicine at Kaohsiung Medical University in Kaohsiung City, Taiwan from The Featured Areas Research Center Program within the framework of the Higher Education Sprout Project by the Ministry of Education (MOE) in Taiwan and by Kaohsiung Medical University Research Center Grant (KMU-TC109A01-1).

Conflicts of Interest

The authors declare no conflict of interest.

Ethics Information

The protocol of this study was approved by the Institutional Review Board of Kaohsiung Medical University Hospital (KMUHIRB-E(I)-20150259, initial date of approval: 06 January 2016).

References

  1. Aluminum. Available online: https://en.wikipedia.org/wiki/Aluminium (accessed on 20 August 2021).
  2. Letzel, S.; Lang, C.; Schaller, K.; Angerer, J.; Fuchs, S.; Neundörfer, B.; Lehnert, G. Longitudinal study of neurotoxicity with occupational exposure to aluminum dust. Neurology 2000, 54, 997–1000. [Google Scholar] [CrossRef]
  3. Sinczuk-Walczak, H.; Szymczak, M.; Razniewska, G.; Matczak, W.; Szymczak, W. Effects of occupational exposure to aluminum on nervous system: Clinical and electroencephalographic findings. Int J Occup Med Environ Health 2003, 16, 301–310. [Google Scholar] [PubMed]
  4. Rønneberg, A.; Haldorsen, T.; Romundstad, P.; Andersen, A. Occupational exposure and cancer incidence among workers from an aluminum smelter in western Norway. J Work Environ Health 1999, 25, 207–214. [Google Scholar] [CrossRef] [Green Version]
  5. Alfrey, A.; Hegg, A.; Craswell, P. Metabolism and toxicity of aluminum in renal failure. Am. J. Clin. Nutr. 1980, 33, 1509–1516. [Google Scholar] [CrossRef] [PubMed]
  6. Recker, R.; Blotcky, A.; Leffler, J.; Rack, E. Evidence of aluminum absorption from the gastrointestinal tract and bone deposition by aluminum carbonate ingestion with normal renal function. J Lab Clin Med 1977, 90, 810–815. [Google Scholar]
  7. Jeffery, E.; Abreo, K.; Burgess, E.; Cannata, J.; Greger, J. Systemic aluminum toxicity: Effects on bone, hematopoietic tissue, and kidney. J. Toxicol. Environ. Health Part A 1996, 48, 649–666. [Google Scholar]
  8. Hewitt, C.; Savory, J.; Wills, M. Aspects of aluminum toxicity. Clin. Lab. Med. 1990, 10, 403–422. [Google Scholar] [CrossRef]
  9. Nayak, P. Aluminum: Impacts and disease. Environ. Res. 2002, 89, 101–115. [Google Scholar] [CrossRef] [PubMed]
  10. Darbre, P. Aluminium, antiperspirants and breast cancer. J. Inorg. Biochem. 2005, 99, 1912–1919. [Google Scholar] [CrossRef]
  11. Ng, E.; Lind, P.; Lindgren, C.; Ingelsson, E.; Mahajan, A.; Morris, A.; Lind, L. Genome-wide association study of toxic metals and trace elements reveals novel associations. Hum Mol Genet 2015, 24, 4739–4745. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  12. Hou, L.; Zhang, X.; Tarantini, L.; Nordio, F.; Bonzini, M.; Angelici, L.; Marinelli, B.; Rizzo, G.; Cantone, L.; Apostoli, P. Ambient PM exposure and DNA methylation in tumor suppressor genes: A cross-sectional study. Part. Fibre Toxicol. 2011, 8, 25. [Google Scholar] [CrossRef] [Green Version]
  13. Tajuddin, S.M.; Amaral, A.F.S.; Fernández, A.F.; Rodríguez-Rodero, S.; Rodríguez, R.M.; Moore, L.E.; Tardón, A.; Carrato, A.; García-Closas, M.; Silverman, D.T.; et al. Genetic and Non-genetic Predictors of LINE-1 Methylation in Leukocyte DNA. Environ. Health Perspect. 2013, 121, 650–656. [Google Scholar] [CrossRef] [Green Version]
  14. Kile, M.L.; Fang, S.; Baccarelli, A.A.; Tarantini, L.; Cavallari, J.; Christiani, D.C. A panel study of occupational exposure to fine particulate matter and changes in DNA methylation over a single workday and years worked in boilermaker welders. J. Environ. Health 2013, 12, 47. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  15. Yang, X.; Yuan, Y.; Lu, X.; Yang, J.; Wang, L.; Song, J.; Nie, J.; Zhang, Q.; Niu, Q. The Relationship Between Cognitive Impairment and Global DNA Methylation Decrease Among Aluminum Potroom Workers. J. Occup. Environ. 2015, 57, 713–717. [Google Scholar] [CrossRef]
  16. Li, J.; Liu, J.; Dong, D.; Jia, X.; McCouch, S.; Kochian, L. Natural variation underlies alterations in Nramp aluminum transporter (NRAT1) expression and function that play a key role in rice aluminum tolerance. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 6503–6508. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  17. Famoso, A.; Zhao, K.; Clark, R.; Tung, C.; Wright, M.; Bustamante, C.; Kochian, L.; McCouch, S. Genetic architecture of aluminum tolerance in rice (Oryza sativa) determined through genome-wide association analysis and QTL mapping. PLoS Genet. 2011, 7, e1002221. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  18. Fan, C.; Lin, J.; Lee, C. Taiwan Biobank: A project aiming to aid Taiwan’s transition into a biomedical island. Pharmacogenomics 2008, 9, 235–246. [Google Scholar] [CrossRef] [PubMed]
  19. Chen, C.; Yang, J.; Chiang, C.; Hsiung, C.; Wu, P.; Chang, L.; Chu, H.; Chang, J.; Song, I.; Yang, S. Population structure of Han Chinese in the modern Taiwanese population based on 10,000 participants in the Taiwan Biobank project. Hum Mol Genet 2016, 25, 5321–5331. [Google Scholar] [CrossRef] [Green Version]
  20. Gibbs, R.; Belmont, J.; Hardenbol, P.; Willis, T.; Yu, F.; Yang, H.; Ch’’ang, L.; Huang, W.; Liu, B.; Shen, Y. The international HapMap project. Nature 2003, 426, 789–796. [Google Scholar]
  21. Siva, N. 1000 Genomes project. Nat. Biotechnol. 2008, 26, 256. [Google Scholar] [CrossRef]
  22. Genomes Project Consortium. An integrated map of genetic variation from 1,092 human genomes. Nature 2012, 491, 56–65. [Google Scholar] [CrossRef] [Green Version]
  23. Purcell, S.; Neale, B.; Todd-Brown, K.; Thomas, L.; Ferreira, M.; Bender, D.; Maller, J.; Sklar, P.; De Bakker, P.; Daly, M. PLINK: A tool set for whole-genome association and population-based linkage analyses. Am J Hum Genet 2007, 81, 559–575. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  24. Ihaka, R.; Gentleman, R. R: A language for data analysis and graphics. J Comput Graph Stat 1996, 5, 299–314. [Google Scholar]
  25. Ma, Y.C.; Zuo, L.; Chen, J.H.; Luo, Q.; Yu, X.Q.; Li, Y.; Xu, J.S.; Huang, S.M.; Wang, L.N.; Huang, W.; et al. Modified glomerular filtration rate estimating equation for Chinese patients with chronic kidney disease. J. Am. Soc. Nephrol. 2006, 17, 2937–2944. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  26. Bland, J.; Altman, D. Multiple significance tests: The Bonferroni method. Br. Med. J. 1995, 310, 170. [Google Scholar] [CrossRef] [Green Version]
  27. Reimand, J.; Arak, T.; Adler, P.; Kolberg, L.; Reisberg, S.; Peterson, H.; Vilo, J. g: Profiler—a web server for functional interpretation of gene lists (2016 update). Nucleic Acids Res. 2016, 44, W83–W89. [Google Scholar] [CrossRef]
  28. Barrett, J.; Fry, B.; Maller, J.; Daly, M. Haploview: Analysis and visualization of LD and haplotype maps. Bioinformatics 2004, 21, 263–265. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  29. Turner, S. qqman: An R package for visualizing GWAS results using QQ and manhattan plots. J. Open Source Softw. 2018, 3, 0731. [Google Scholar] [CrossRef] [Green Version]
  30. Baron, R.; Kenny, D. The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986, 51, 1173. [Google Scholar] [CrossRef]
  31. MacKinnon, D.; Fairchild, A.; Fritz, M. Mediation analysis. Annu. Rev. Psychol. 2007, 58, 593–614. [Google Scholar] [CrossRef] [PubMed]
  32. MacKinnon, D.P.; Lockwood, C.M.; Williams, J. Confidence Limits for the Indirect Effect: Distribution of the Product and Resampling Methods. Multivariate Behav Res 2004, 39, 99. [Google Scholar] [CrossRef] [Green Version]
  33. Preacher, K.; Hayes, A. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav. Res. Methods 2008, 40, 879–891. [Google Scholar] [CrossRef]
  34. Thakkinstian, A.; Anothaisintawee, T.; Chailurkit, L.; Ratanachaiwong, W.; Yamwong, S.; Sritara, P.; Ongphiphadhanakul, B. Potential causal associations between vitamin D and uric acid: Bidirectional mediation analysis. Sci. Rep. 2015, 5, 14528. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Preacher, K.; Selig, J. Advantages of Monte Carlo confidence intervals for indirect effects. Commun Methods Meas 2012, 6, 77–98. [Google Scholar] [CrossRef] [Green Version]
  36. Hayes, A.; Scharkow, M. The relative trustworthiness of inferential tests of the indirect effect in statistical mediation analysis: Does method really matter? Psychol. Sci. 2013, 24, 1918–1927. [Google Scholar] [CrossRef] [PubMed]
  37. Miotto, B.; Chibi, M.; Xie, P.; Koundrioukoff, S.; Moolman-Smook, H.; Pugh, D.; Debatisse, M.; He, F.; Zhang, L.; Defossez, P.A. The RBBP6/ZBTB38/MCM10 axis regulates DNA replication and common fragile site stability. Cell Rep. 2014, 7, 575–587. [Google Scholar] [CrossRef]
  38. Jones, M.H.; Hamana, N.; Nezu, J.; Shimane, M. A novel family of bromodomain genes. Genomics 2000, 63, 40–45. [Google Scholar] [CrossRef]
  39. Baldwin, D.; Marshall, W. Heavy metal poisoning and its laboratory investigation. Ann. Clin. Biochem 1999, 36, 267–300. [Google Scholar] [CrossRef] [Green Version]
  40. Goullé, J.; Mahieu, L.; Castermant, J.; Neveu, N.; Bonneau, L.; Lainé, G.; Bouige, D.; Lacroix, C. Metal and metalloid multi-elementary ICP-MS validation in whole blood, plasma, urine and hair: Reference values. Sci. Int. 2005, 153, 39–44. [Google Scholar] [CrossRef] [PubMed]
  41. Cronin, S.; Tomik, B.; Bradley, D.G.; Slowik, A.; Hardiman, O. Screening for replication of genome-wide SNP associations in sporadic ALS. Eur. J. Hum. Genet. 2009, 17, 213–218. [Google Scholar] [CrossRef] [Green Version]
  42. Chio, A.; Schymick, J.C.; Restagno, G.; Scholz, S.W.; Lombardo, F.; Lai, S.L.; Mora, G.; Fung, H.C.; Britton, A.; Arepalli, S.; et al. A two-stage genome-wide association study of sporadic amyotrophic lateral sclerosis. Hum. Mol. Genet. 2009, 18, 1524–1532. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  43. Kwee, L.C.; Liu, Y.; Haynes, C.; Gibson, J.R.; Stone, A.; Schichman, S.A.; Kamel, F.; Nelson, L.M.; Topol, B.; Van den Eeden, S.K.; et al. A high-density genome-wide association screen of sporadic ALS in US veterans. PloS ONE 2012, 7, e32768. [Google Scholar] [CrossRef] [Green Version]
  44. Maussion, G.; Cruceanu, C.; Rosenfeld, J.A.; Bell, S.C.; Jollant, F.; Szatkiewicz, J.; Collins, R.L.; Hanscom, C.; Kolobova, I.; de Champfleur, N.M.; et al. Implication of LRRC4C and DPP6 in neurodevelopmental disorders. Am. J. Med. Genet. Part A. 2017, 173, 395–406. [Google Scholar] [CrossRef] [PubMed]
  45. Wolf, E.J.; Rasmusson, A.M.; Mitchell, K.S.; Logue, M.W.; Baldwin, C.T.; Miller, M.W. A genome-wide association study of clinical symptoms of dissociation in a trauma-exposed sample. Depress Anxiety. 2014, 31, 352–360. [Google Scholar] [CrossRef] [Green Version]
  46. Lin, L.; Long, L.K.; Hatch, M.M.; Hoffman, D.A. DPP6 Domains Responsible for Its Localization and Function. J. Biol. Chem. 2014, 289, 32153–32165. [Google Scholar] [CrossRef] [Green Version]
  47. Lin, L.; Murphy, J.G.; Karlsson, R.M.; Petralia, R.S.; Gutzmann, J.J.; Abebe, D.; Wang, Y.X.; Cameron, H.A.; Hoffman, D.A. DPP6 Loss Impacts Hippocampal Synaptic Development and Induces Behavioral Impairments in Recognition, Learning and Memory. Front. Cell. Neurosci. 2018, 12, 84. [Google Scholar] [CrossRef] [Green Version]
  48. Keith, S.; Jones, D.; Rosemond, Z.; Ingerman, L.; Chappell, L. Toxicological profile for aluminum. DIANE Publishing: Collingdale, PA, USA, 2008. [Google Scholar]
  49. Tokar, E.; Boyd, W.; Freedman, J.; Waalkes, M. Toxic effects of metals. In Casarett & Doull’s Toxicology: The Basic Science of Poisons, 8th ed.; Klaassen, C., Ed.; McGraw-Hill Health Professions Division: New York, NY, USA, 2013. [Google Scholar]
  50. Mayor, G.H.; Burnatowska-Hledin, M.A. Impaired renal function and aluminum metabolism. FASEB J. 1983, 42, 2979–2983. [Google Scholar]
  51. Brambilla, P.; Esposito, F.; Lindstrom, E.; Sorosina, M.; Giacalone, G.; Clarelli, F.; Rodegher, M.; Colombo, B.; Moiola, L.; Ghezzi, A.; et al. Association between DPP6 polymorphism and the risk of progressive multiple sclerosis in Northern and Southern Europeans. Neurosci. Lett. 2012, 530, 155–160. [Google Scholar] [CrossRef] [PubMed]
  52. Martyn, C.; Coggon, D.; Inskip, H.; Lacey, R.; Young, W. Aluminum concentrations in drinking water and risk of Alzheimer’’s disease. Epidemiology. 1997, 8, 281–286. [Google Scholar] [CrossRef]
  53. McLachlan, D. Aluminium and the risk for Alzheimer’’s disease. Environmetrics. 1995, 6, 233–275. [Google Scholar] [CrossRef]
  54. Virk, S.A.; Eslick, G.D. Occupational Exposure to Aluminum and Alzheimer Disease: A Meta-Analysis. J. Occup. Environ. Med. 2015, 57, 893–896. [Google Scholar] [CrossRef] [PubMed]
  55. Krewski, D.; Yokel, R.; Nieboer, E.; Borchelt, D.; Cohen, J.; Harry, J.; Kacew, S.; Lindsay, J.; Mahfouz, A.; Rondeau, V. Human health risk assessment for aluminium, aluminium oxide, and aluminium hydroxide. J. Toxicol. Environ. Health Part B. 2007, 10, 1–269. [Google Scholar] [CrossRef] [PubMed]
Ijerph 18 10484 g001 550
Figure 1. The relationship between X (e.g., SNPs), M (e.g., Al concentrations), and Y (e.g., eGFR) using a mediation analysis.
Figure 1. The relationship between X (e.g., SNPs), M (e.g., Al concentrations), and Y (e.g., eGFR) using a mediation analysis.
Ijerph 18 10484 g001
Ijerph 18 10484 g002 550
Figure 2. Flow chart for quality control of SNPs and selection of study population.
Figure 2. Flow chart for quality control of SNPs and selection of study population.
Ijerph 18 10484 g002
Ijerph 18 10484 g003 550
Figure 3. Scatter plot for eGFR and log-transformed Al concentration.
Figure 3. Scatter plot for eGFR and log-transformed Al concentration.
Ijerph 18 10484 g003
Ijerph 18 10484 g004 550
Figure 4. Manhattan plot for plasma aluminum after controlling for gender, age, smoking, and alcohol consumption. The vertical axis was −log10 (p-value) according to the strength of association. The horizontal axis represents chromosome location according to the genomic location. The red line represents significant levels (p-value < 8 × 10−8). The blue line represents suggestive levels (p-value < 1 × 10−6).
Figure 4. Manhattan plot for plasma aluminum after controlling for gender, age, smoking, and alcohol consumption. The vertical axis was −log10 (p-value) according to the strength of association. The horizontal axis represents chromosome location according to the genomic location. The red line represents significant levels (p-value < 8 × 10−8). The blue line represents suggestive levels (p-value < 1 × 10−6).
Ijerph 18 10484 g004
Ijerph 18 10484 g005 550
Figure 5. The mediation analysis of 3 SNPs within DPP6, ln (aluminum) levels in plasma, and eGFR. (a): rs10224371 was fitted as an additive model effect with 0, 1, and 2 for CC, CT, and TT. (b): rs2316242 was fitted as an additive model effect with 0, 1, and 2 for CC, CT, and TT. (c): rs10268004 was fitted as an additive model effect with 0, 1, and 2 for TT, GT, and GG. (a): each effect with adjusted gender, age, smoke experience, and drinking habits. (b): coefficient; SE, standard error; 95%CI, 95% confidence interval estimated by percentile bootstrap methods. A summary is shown in the right lower quadrant. * p < 0.05.
Figure 5. The mediation analysis of 3 SNPs within DPP6, ln (aluminum) levels in plasma, and eGFR. (a): rs10224371 was fitted as an additive model effect with 0, 1, and 2 for CC, CT, and TT. (b): rs2316242 was fitted as an additive model effect with 0, 1, and 2 for CC, CT, and TT. (c): rs10268004 was fitted as an additive model effect with 0, 1, and 2 for TT, GT, and GG. (a): each effect with adjusted gender, age, smoke experience, and drinking habits. (b): coefficient; SE, standard error; 95%CI, 95% confidence interval estimated by percentile bootstrap methods. A summary is shown in the right lower quadrant. * p < 0.05.
Ijerph 18 10484 g005
Table
Table 1. Characteristics of study population.
Table 1. Characteristics of study population.
Term (n = 494)Mean ± SD, N(%)MinQ1Q2Q3Max
Gender (Male)275(55.7%)
Drink (Current drinking) a39(7.9%)
Smoke experience (Yes)169(34.2%)
Age (year)48.3 ± 10.93040475470
Height (cm)165.18 ± 8.63142.0188.5158.5165.5171.5
Weight (kg)66.95 ± 12.3239.4120.858.266.8074.8
γ-GT (U/L)26.25 ± 27.096.013.018.030.0293.0
BUN (mg/dL)13.06 ± 3.485.010.612.615.228.2
Creatinine (mg/dL)0.78 ± 0.190.330.620.790.911.50
eGFR b (mL/min/1.73 m2)99.52 ± 22.0049.1585.3096.99110.37216.66
Uric acid (mg/dL)5.73 ± 1.562.64.55.66.912.4
HbA1C (%)5.66 ± 0.554.45.45.65.811.9
Total cholesterol (mg/dL)197.7 ± 35.8117358174195217
TG (mg/dL)122.9 ± 85.73079868101145
HDL (mg/dL)53.8 ± 13.226108445262
LDL (mg/dL)125.5 ± 32.447258105124146
AST (U/L)23.38 ± 11.4410182126167
ALT (U/L)25.6 ± 23.47152028344
Aluminum (μg/L)1.13 ± 0.800.290.780.991.2713.94
a To continue alcohol consumption for more than six months, 155 cc per week. b Estimated glomerular filtration rate (MDRD equation); Q1: The first quartile; Q2: The second quartile; Q3: The third quartile.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Chen, T.-H.; Yang, C.-C.; Luo, K.-H.; Dai, C.-Y.; Chuang, Y.-C.; Chuang, H.-Y. The Mediation Effects of Aluminum in Plasma and Dipeptidyl Peptidase Like Protein 6 (DPP6) Polymorphism on Renal Function via Genome-Wide Typing Association. Int. J. Environ. Res. Public Health 2021, 18, 10484. https://doi.org/10.3390/ijerph181910484

AMA Style

Chen T-H, Yang C-C, Luo K-H, Dai C-Y, Chuang Y-C, Chuang H-Y. The Mediation Effects of Aluminum in Plasma and Dipeptidyl Peptidase Like Protein 6 (DPP6) Polymorphism on Renal Function via Genome-Wide Typing Association. International Journal of Environmental Research and Public Health. 2021; 18(19):10484. https://doi.org/10.3390/ijerph181910484

Chicago/Turabian Style

Chen, Ting-Hao, Chen-Cheng Yang, Kuei-Hau Luo, Chia-Yen Dai, Yao-Chung Chuang, and Hung-Yi Chuang. 2021. "The Mediation Effects of Aluminum in Plasma and Dipeptidyl Peptidase Like Protein 6 (DPP6) Polymorphism on Renal Function via Genome-Wide Typing Association" International Journal of Environmental Research and Public Health 18, no. 19: 10484. https://doi.org/10.3390/ijerph181910484

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop