doi: 10.3389/fnagi.2021.713084.
eCollection 2021.
Affiliations
Item in Clipboard
Front Aging Neurosci.
.
Abstract
Objective: To establish a workflow for mitochondrial DNA (mtDNA) CpG methylation using Nanopore whole-genome sequencing and perform first pilot experiments on affected Parkin biallelic mutation carriers (Parkin-PD) and healthy controls. Background: Mitochondria, including mtDNA, are established key players in Parkinson’s disease (PD) pathogenesis. Mutations in Parkin, essential for degradation of damaged mitochondria, cause early-onset PD. However, mtDNA methylation and its implication in PD is understudied. Herein, we establish a workflow using Nanopore sequencing to directly detect mtDNA CpG methylation and compare mtDNA methylation between Parkin-related PD and healthy individuals. Methods: To obtain mtDNA, whole-genome Nanopore sequencing was performed on blood-derived from five Parkin-PD and three control subjects. In addition, induced pluripotent stem cell (iPSC)-derived midbrain neurons from four of these patients with PD and the three control subjects were investigated. The workflow was validated, using methylated and unmethylated 897 bp synthetic DNA samples at different dilution ratios (0, 50, 100% methylation) and mtDNA without methylation. MtDNA CpG methylation frequency (MF) was detected using Nanopolish and Megalodon. Results: Across all blood-derived samples, we obtained a mean coverage of 250.3X (SD ± 80.5X) and across all neuron-derived samples 830X (SD ± 465X) of the mitochondrial genome. We detected overall low-level CpG methylation from the blood-derived DNA (mean MF ± SD = 0.029 ± 0.041) and neuron-derived DNA (mean MF ± SD = 0.019 ± 0.035). Validation of the workflow, using synthetic DNA samples showed that highly methylated DNA molecules were prone to lower Guppy Phred quality scores and thereby more likely to fail Guppy base-calling. CpG methylation in blood- and neuron-derived DNA was significantly lower in Parkin-PD compared to controls (Mann-Whitney U-test p < 0.05). Conclusion: Nanopore sequencing is a useful method to investigate mtDNA methylation architecture, including Guppy-failed reads is of importance when investigating highly methylated sites. We present a mtDNA methylation workflow and suggest methylation variability across different tissues and between Parkin-PD patients and controls as an initial model to investigate.
Keywords:
Nanopore; Parkin; Parkinson’s disease; methylation; mitochondrial DNA; third-generation sequencing.
Copyright © 2021 Lüth, Wasner, Klein, Schaake, Tse, Pereira, Laß, Sinkkonen, Grünewald and Trinh.
Conflict of interest statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Figures
Figure 1
Mitochondrial CpG methylation frequency detected by Nanopore sequencing in eight individuals. Radar chart showing mitochondrial DNA methylation from eight blood-derived DNA samples detected by Nanopolish. Methylation frequency is indicated by blue spikes. 12S, small subunit rRNA; 16S, large subunit rRNA; ND1, NADH dehydrogenase, subunit 1; ND2, NADH dehydrogenase, subunit 2; COI, cytochrome c oxidase, subunit 1; COII, cytochrome c oxidase, subunit 2; ATPase8, ATP synthase, subunit 8; ATPase6, ATP synthase, subunit 6; COIII, cytochrome c oxidase, subunit 3; ND3, NADH dehydrogenase, subunit 3; ND4L, NADH dehydrogenase, subunit 4L; ND4, NADH dehydrogenase, subunit 4; ND5, NADH dehydrogenase, subunit 5; ND6, NADH dehydrogenase, subunit 6; Cyt b, cytochrome b; D-Loop, displacement loop.
Figure 2
Relationship between coverage and methylation frequency. (A) Overall relationship between the mean coverage and mean methylation frequency of different subsections of the obtained sequencing data. (B) Overall relationship between the coverage of each individual CpG site and the corresponding methylation frequency. r, Spearman’s rank correlation coefficient; p, Spearman’s exploratory p-value.
Figure 3
Mitochondrial and 45S rRNA CpG methylation comparison by strand. (A) Correlation between the methylation frequency of the mitochondrial plus- and minus-strand (plus-strand = light-strand, minus-strand = heavy-strand). Data of the six blood-derived samples combined. (B) Correlation between the methylation frequency detected from the 45S rRNA gene plus- and minus-strand. Data of the six blood-derived samples combined. r, Spearman’s rank correlation coefficient; p, Spearman’s exploratory p-value.
Figure 4
Analysis of synthetic DNA for the validation of the data analysis pipeline. (A) Methylation frequency detected by Nanopolish from three samples of synthetic DNA with different proportions of methylated DNA (0, 50 and 100%), only reads that passed Guppy quality threshold were included in the analysis. (B) Methylation frequency detected by Nanopolish from three samples of synthetic DNA with different proportions of methylated DNA (0, 50 and 100%), only reads that failed Guppy quality threshold were included in the analysis. (C) Stacked bar plot of the fraction of reads that passed (blue) or failed (red) the Guppy quality threshold, stratified by the ratio of methylated reads in the sample. (D) Relationship between the coverage and methylation frequency in the sample with 100% methylated reads, only reads that passed Guppy quality threshold were included in the analysis. (E) Relationship between the coverage and methylation frequency in the sample with 100% methylated reads, only reads that failed Guppy quality threshold were included in the analysis. r, Spearman’s rank correlation coefficient; p, Spearman’s exploratory p-value.
Figure 5
Comparison of mtDNA CpG methylation between blood- and neuron-derived DNA. (A,B) The bar plot shows the mtDNA CpG methylation from blood- or neuron-derived DNA (iPSC-derived midbrain neurons) which was either detected with Nanopolish (A) or Megalodon (B). Bars indicate means and 95% confidence interval. The asterisks represent the level of significance (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001), p-value = Mann Whitney U-test performed for pairwise comparisons, blood-derived DNA is indicated in red and neuron-derived DNA in blue.
Figure 6
Comparison of mtDNA CpG methylation between patients with Parkin-PD and healthy controls. (A,B) The bar plot shows the mtDNA CpG methylation from blood-derived DNA in patients with Parkin-PD and healthy control subjects, which was either detected with Nanopolish (A) or Megalodon (B). (C,D) The bar plot shows the mtDNA CpG methylation from neuron-derived DNA (iPSC-derived midbrain neurons) in patients with Parkin-PD and healthy control subjects, which was either detected with Nanopolish (A) or Megalodon (B). Bars indicate means and 95% confidence interval. The asterisks represent the level of significance (*p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001), p-value = Mann Whitney U-test performed for pairwise comparisons, Parkin++ = Parkin biallelic; PD+ = PD patient, patients are indicated in green and control subjects in gray.
References
-
-
Berk A. J., Clayton D. A. (1974). Mechanism of mitochondrial DNA replication in mouse L-cells: asynchronous replication of strands, segregation of circular daughter molecules, aspects of topology and turnover of an initiation sequence. J. Mol. Biol. 86, 801–824. 10.1016/0022-2836(74)90355-6
–
DOI
–
PubMed
-
-
-
Bi C., Wang L., Fan Y., Ramos-Mandujano G., Yuan B., Zhou X., et al. . (2020). Single-cell individual complete mtDNA sequencing uncovers hidden mitochondrial heterogeneity in human and mouse oocytes. bioRxiv [preprint]. 10.1101/2020.12.28.424537
–
DOI
-
Read more here: Source link
