A systematic evaluation of normalization methods and probe replicability using infinium EPIC methylation data

BACKGROUND

The Infinium EPIC array measures the methylation status of > 850,000 CpG sites. The EPIC BeadChip uses a two-array design: Infinium Type I and Type II probes. These probe types exhibit different technical characteristics which may confound analyses. Numerous normalization and pre-processing methods have been developed to reduce probe type bias as well as other issues such as background and dye bias.

METHODS

This study evaluates the performance of various normalization methods using 16 replicated samples and three metrics: absolute beta-value difference, overlap of non-replicated CpGs between replicate pairs, and effect on beta-value distributions. Additionally, we carried out Pearson's correlation and intraclass correlation coefficient (ICC) analyses using both raw and SeSAMe 2 normalized data.

RESULTS

The method we define as SeSAMe 2, which consists of the application of the regular SeSAMe pipeline with an additional round of QC, pOOBAH masking, was found to be the best performing normalization method, while quantile-based methods were found to be the worst performing methods. Whole-array Pearson's correlations were found to be high. However, in agreement with previous studies, a substantial proportion of the probes on the EPIC array showed poor reproducibility (ICC < 0.50). The majority of poor performing probes have beta values close to either 0 or 1, and relatively low standard deviations. These results suggest that probe reliability is largely the result of limited biological variation rather than technical measurement variation. Importantly, normalizing the data with SeSAMe 2 dramatically improved ICC estimates, with the proportion of probes with ICC values > 0.50 increasing from 45.18% (raw data) to 61.35% (SeSAMe 2).

Increased H2O2 levels and p53 stabilization lead to mitochondrial dysfunction in XPC-deficient cells

XPC deficiency is associated with mitochondrial dysfunction, increased mitochondrial H2O2 production and sensitivity to the Complex III inhibitor antimycin A (AA), through a yet unclear mechanism. We found an imbalanced expression of several proteins that participate in important mitochondrial function and increased expression and phosphorylation of the tumor suppressor p53 in Xeroderma pigmentosum complementation group C (XP-C) (XPC-null) cells compared with an isogenic line corrected in locus with wild-type XPC (XPC-wt). Interestingly, inhibition of p53 nuclear import reversed the overexpression of mitochondrial proteins, whereas AA treatment increased p53 expression more strongly in the XP-C cells. However, inhibition of p53 substantially increased XP-C cellular sensitivity to AA treatment, suggesting that p53 is a critical factor mediating the cellular response to mitochondrial stress. On the other hand, treatment with the antioxidant N-acetylcysteine increased glutathione concentration and decreased basal H2O2 production, p53 levels and sensitivity to AA treatment in the XPC-null back to the levels found in XPC-wt cells. Thus, the results suggest a critical role for mitochondrially generated H2O2 in the regulation of p53 expression, which in turn modulates XP-C sensitivity to agents that cause mitochondrial stress.

Where do we aspire to publish? A position paper on scientific communication in biochemistry and molecular biology

The scientific publication landscape is changing quickly, with an enormous increase in options and models. Articles can be published in a complex variety of journals that differ in their presentation format (online-only or in-print), editorial organizations that maintain them (commercial and/or society-based), editorial handling (academic or professional editors), editorial board composition (academic or professional), payment options to cover editorial costs (open access or pay-to-read), indexation, visibility, branding, and other aspects. Additionally, online submissions of non-revised versions of manuscripts prior to seeking publication in a peer-reviewed journal (a practice known as pre-printing) are a growing trend in biological sciences. In this changing landscape, researchers in biochemistry and molecular biology must re-think their priorities in terms of scientific output dissemination. The evaluation processes and institutional funding for scientific publications should also be revised accordingly. This article presents the results of discussions within the Department of Biochemistry, University of São Paulo, on this subject.

Localization of mitochondrial DNA base excision repair to an inner membrane-associated particulate fraction

Mitochondrial DNA (mtDNA) contains high levels of oxidative damage relative to nuclear DNA. A full, functional DNA base excision repair (BER) pathway is present in mitochondria, to repair oxidative DNA lesions. However, little is known about the organization of this pathway within mitochondria. Here, we provide evidence that the mitochondrial BER proteins are not freely soluble, but strongly associated with an inner membrane-containing particulate fraction. Uracil DNA glycosylase, oxoguanine DNA glycosylase and DNA polymerase γ activities all co-sedimented with this particulate fraction and were not dissociated from it by detergent (0.1% or 1.0% NP40) treatment. The particulate associations of these activities were not due to their binding mtDNA, which is itself associated with the inner membrane, as they also localized to the particulate fraction of mitochondria from 143B (TK⁻) ρ⁰ cells, which lack mtDNA. However, all of the BER activities were at least partially solubilized from the particulate fraction by treatment with 150–300 mM NaCl, suggesting that electrostatic interactions are involved in the association. The biological implications of the apparent immobilization of BER proteins are discussed.

DNA base excision repair activities and pathway function in mitochondrial and cellular lysates from cells lacking mitochondrial DNA

Mitochondrial DNA (mtDNA) contains higher steady-state levels of oxidative damage and mutates at rates significantly greater than nuclear DNA. Oxidative lesions in mtDNA are removed by a base excision repair (BER) pathway. All mtDNA repair proteins are nuclear encoded and imported. Most mtDNA repair proteins so far discovered are either identical to nuclear DNA repair proteins or isoforms of nuclear proteins arising from differential splicing. Regulation of mitochondrial BER is therefore not expected to be independent of nuclear BER, though the extent to which mitochondrial BER is regulated with respect to mtDNA amount or damage is largely unknown. Here we have measured DNA BER activities in lysates of mitochondria isolated from human 143B TK(-) osteosarcoma cells that had been depleted of mtDNA (rho(0)) or not (wt). Despite the total absence of mtDNA in the rho(0) cells, a complete mitochondrial BER pathway was present, as demonstrated using an in vitro assay with synthetic oligonucleotides. Measurement of individual BER protein activities in mitochondrial lysates indicated that some BER activities are insensitive to the lack of mtDNA. Uracil and 8-oxoguanine DNA glycosylase activities were relatively insensitive to the absence of mtDNA, only about 25% reduced in rho(0) relative to wt cells. Apurinic/apyrimidinic (AP) endonuclease and polymerase gamma activities were more affected, 65 and 45% lower, respectively, in rho(0) mitochondria. Overall BER activity in lysates was also about 65% reduced in rho(0) mitochondria. To identify the limiting deficiencies in BER of rho(0) mitochondria we supplemented the BER assay of mitochondrial lysates with pure uracil DNA glycosylase, AP endonuclease and/or the catalytic subunit of polymerase gamma. BER activity was stimulated by addition of uracil DNA glycosylase and polymerase gamma. However, no addition or combination of additions stimulated BER activity to wt levels. This suggests that an unknown activity, factor or interaction important in BER is deficient in rho(0) mitochondria. While nuclear BER protein levels and activities were generally not altered in rho(0) cells, AP endonuclease activity was substantially reduced in nuclear and in whole cell extracts. This appeared to be due to reduced endogenous reactive oxygen species (ROS) production in rho(0) cells, and not a general dysfunction of rho(0) cells, as exposure of cells to ROS rapidly stimulated increases in AP endonuclease activities and APE1 protein levels.

Mitochondrial and nuclear DNA base excision repair are affected differently by caloric restriction

Aging is strongly correlated with the accumulation of oxidative damage in DNA, particularly in mitochondria. Oxidative damage to both mitochondrial and nuclear DNA is repaired by the base excision repair (BER) pathway. The "mitochondrial theory of aging" suggests that aging results from declining mitochondrial function, due to high loads of damage and mutation in mitochondrial DNA (mtDNA). Restriction of caloric intake is the only intervention so far proven to slow the aging rate. However, the molecular mechanisms underlying such effects are still unclear. We used caloric-restricted (CR) mice to investigate whether lifespan extension is associated with changes in mitochondrial BER activities. Mice were divided into two groups, receiving 100% (PF) or 60% (CR) of normal caloric intake, a regime that extends mean lifespan by approximately 40% in CR mice. Mitochondria isolated from CR mice had slightly higher uracil (UDG) and oxoguanine DNA glycosylase (OGG1) activities but marginally lower abasic endonuclease and polymerase gamma gap-filling activities, although these differences were tissue-specific. Uracil-initiated BER synthesis incorporation activities were significantly lower in brain and kidney from CR mice but marginally enhanced in liver. However, nuclear repair synthesis activities were increased by CR, indicating differential regulation of BER in the two compartments. The results indicate that a general up-regulation of mitochondrial BER does not occur in CR.

Age-associated increase in 8-oxo-deoxyguanosine glycosylase/AP lyase activity in rat mitochondria

The mitochondrial theory of aging postulates that organisms age due to the accumulation of DNA damage and mutations in the multiple mitochondrial genomes, leading to mitochondrial dysfunction. Among the wide variety of DNA damage, 8-oxo-deoxyguanosine (8-oxo-dG) has received the most attention due to its mutagenicity and because of the possible correlation between its accumulation and pathological processes like cancer, degenerative diseases and aging. Although still controversial, many studies show that 8-oxo-dG accumulates with age in the mitochondrial (mt) DNA. However, little is known about the processing of this lesion and no study has yet examined whether mtDNA repair changes with age. Here, we report the first study on age-related changes in mtDNA repair, accomplished by assessing the cleavage activity of mitochondrial extracts towards an 8-oxo-dG-containing substrate. In this study, mitochondria obtained from rat heart and liver were used. We find that this enzymatic activity is higher in 12 and 23 month-old rats than in 6 month-old rats, in both liver and heart extracts. These mitochondrial extracts also cleave oligonucleotides containing a U:A mismatch, at the uracil position, reflecting the combined action of mitochondrial uracil DNA glycosylase (mtUDG) and mitochondrial apurinic/apyrimidinic (AP) endonucleases. The mtUDG activity did not change with age in liver mitochondria, but there was a small increase in activity from 6 to 23 months in rat heart extracts, after normalization to citrate synthase activity. Endonuclease G activity, measured by a plasmid relaxation assay, did not show any age-associated change in liver, but there was a significant decrease from 6 to 23 months in heart mitochondria. Our results suggest that the mitochondrial capacity to repair 8-oxo-dG, the main oxidative base damage suggested to accumulate with age in mtDNA, does not decrease, but rather increases with age. The specific increase in 8-oxo-dG endonuclease activity, rather than a general up-regulation of DNA repair in mitochondria, suggests an induction of the 8-oxo-dG-specific repair pathway with age.

Age-associated change in mitochondrial DNA damage

There is an age-associated decline in the mitochondrial function of the Wistar rat heart. Previous reports from this lab have shown a decrease in mitochondrial cytochrome c oxidase (COX) activity associated with a reduction in COX gene and protein expression and a similar decrease in the rate of mitochondrial protein synthesis. Damage to mitochondrial DNA may contribute to this decline. Using the HPLC-Coularray system (ESA, USA), we measured levels of nuclear and mitochondrial 8-oxo-2'-deoxyguanosine (8-oxodG) from 6-month (young) and 23-month-old (senescent) rat liver DNA. We measured the sensitivity of the technique by damaging calf thymus DNA with photoactivated methylene blue for 30s up to 2h. The levels of damage were linear over the entire time course including the shorter times which showed levels comparable to those expected in liver. For the liver data, 8-oxodG was reported as a fraction of 2-deoxyguanosine (2-dG). There was no change in the levels of 8-oxodG levels in the nuclear DNA from 6 to 23-months of age. However, the levels of 8-oxodG increased 2.5-fold in the mitochondrial DNA with age. At 6 months, the level of 8-oxodG in mtDNA was 5-fold higher than nuclear and increased to approximately 12-fold higher by 23 months of age. These findings agree with other reports showing an age-associated increase in levels of mtDNA damage; however, the degree to which it increases is smaller. Such damage to the mitochondrial DNA may contribute to the age-associated decline in mitochondrial function.