Pulmonary Arterial Hypertension Is Associated with Oxidative Stress–induced Genome Instability

DNA damage, defined as chemical changes to the sugar–phosphate backbone or one of the four bases, can occur through exogenous chemical and physical agents such as sunlight, ionizing radiation, or chemical carcinogens or through endogenous agents such as reactive oxygen species (ROS) from mitochondrial respiration or proinflammatory processes (1) (Figure 1). This DNA damage, if not properly repaired, can cause DNA replication abnormalities and manifest itself as genome instability in the form of simple base changes (point mutations) or deletions and chromosome aberrations. Many human pathologies, including cancers and several neurodegenerative diseases, have been associated with either increased DNA damage or loss of specific DNA repair pathways (2). However, direct measurements of DNA damage in human tissues is technologically challenging, and surrogate biomarkers of genome instability have been developed by the field of genetic toxicology. One commonly used biomarker is the micronucleus assay (3). Micronuclei, as the name implies, are fragments of nuclear DNA resulting from chromosome breakage or loss of up to a third the size of the bifurcated nucleus, which can be easily observed in cells arrested during cytokinesis under a light microscope (Figure 1). Micronucleus formation can be caused by chromosomal breakage or through the dysfunction of the mitotic spindle apparatus. Thus, micronucleus formation can be, in part, caused by replication errors as a result of persistent DNA damage at the time of S-phase.

Figure 1.

Potential role of mitochondria-generated reactive oxygen species (ROS) in the formation of micronuclei (bottom, white boxes) in pulmonary arterial hypertension. Another important source of ROS is NOX (NADPH [nicotinamide adenine dinucleotide phosphate] oxidase) (not shown). Hydrogen peroxide released by mitochondria could cause oxidative DNA damage in the nucleus, which, if unrepaired, could cause genome instability. ETC = electron transport chain; GPx = glutathione peroxidase; mtDNA = mitochondrial DNA; Prx = peroxiredoxin; SOD = superoxide dismutase; Ub = ubiquinone, also known as Coenzyme Q10.

The molecular events underlying the etiology of pulmonary arterial hypertension (PAH) are not well understood: This disease has been associated with DNA microsatellite instability and chromosomal abnormalities in cultured pulmonary artery endothelial cells (PAECs) from patients with PAH (4). In an important new study published in this issue of the Journal, Federici and colleagues (pp. 219–228) at the Lerner Research Institute, Cleveland Clinic, have found evidence for increased genome instability in patients with heritable, idiopathic, or associated PAH (5). The authors of this study first isolated PAECs and used molecular karyotyping on single-nucleotide polymorphism arrays to assess copy number variation and found that 13 of 40 patients with PAH (compared with 1 of 19 controls) displayed copy number changes. Further analysis of DNA from other organ sites within 11 of the patients with PAH confirmed that the changes induced in PAECs were indeed newly arising somatic mutations. The authors next performed a micronucleus assay on four PAEC samples with at least one from each of the three groups of patients with PAH and found a statistical increase in micronuclei from patients who showed chromosomal abnormalities. Staining cells for γ-H2AX, which is sometimes referred to as a marker of DNA double-strand breaks (DSBs), but is more appropriately called a marker of replicative stress (6), also showed increases in patients with PAH and correlated nicely with the number of micronuclei. Staining for 53BP is a more suitable marker for DSB (7, 8), and future studies with this marker would be appropriate. Surprisingly, these researchers documented increased micronuclei in immortalized Epstein-Barr virus–transformed lymphoblastoid cells and peripheral blood mononuclear cells from both patients with PAH, as well as from their relatives, but not in those from unrelated controls. This fact, coupled with the observation that this phenotype persists even in immortalized B cells grown in culture, strongly suggests a genetically inheritable trait.

Because endogenous agents such as ROS can drive genome instability, the authors next examined ROS production in patient-derived lymphoblastoid cells and PAECs, using a dye that becomes fluorescent when altered by ROS. An increase in fluorescence was noted for patients with PAH and correlated well with the number of micronuclei in PAECs. This increased ROS was also verified by measuring hydrogen peroxide released into the media and the use of antioxidants to reduce the fluorescence signal. Finally, as ROS can arise from several different sources, the authors examined a mitochondrially targeted superoxide probe that also showed elevation in PAECs from patients with PAH.

Taken together, these studies suggest that an inheritable set of genes may be responsible for a ROS-associated increase in genome instability not only in the target sites but also in peripheral blood cells, suggesting a potential prognostic indicator. The important question arises: What types of genetic differences may predispose patients to PAH? Clearly an increase in ROS could be a result of either increased ROS production and/or decreased antioxidant defenses. NOX (NADPH [nicotinamide adenine dinucleotide phosphate] oxidase) and mitochondria are the two largest sources of cellular ROS (Figure 1). Importantly, damage to mitochondria can cause an increase in ROS production, leading to a vicious cycle of ROS production by the mitochondria (9). Furthermore, deficiencies in iron could affect mitochondrial function and have been associated with PAH (10). The observed increased superoxide production in the mitochondria of PAECs in patients with PAH supports a potential defect in mitochondrial respiration. It is interesting to note that abnormal mitochondria have been noted in PAH (11). Some pesticides, such as rotenone, have been associated with mitochondrial dysfunction; however, the authors indicated that environmental effects seemed unlikely, as the effects were seen among parents, their siblings, and their offspring, but not in spouses of patients with PAH. Superoxide is rapidly converted to hydrogen peroxide by superoxide dismutases in either the mitochondria or the cytoplasm. Hydrogen peroxide can be converted to water through the action of glutathione peroxidase, an abundant enzyme that uses glutathione as a cofactor (Figure 1). Changes in these components could make the cells more prone to DNA injury.

If left unrepaired, DNA damage could result in elevated micronuclei, and another contributor to genome instability could be changes in DNA repair enzymes. ROS produces a spectrum of damage ranging from base damage to single-strand breaks and DSB, each with dedicated repair pathways. It is well known that DNA repair capacity varies among people in the population (12), so it is plausible that defects in repair could make people more susceptible to PAH. To test this hypothesis, the authors treated PAECs with DNA-damaging agents that produce DNA DSBs and found that patients representing groups of one of the three forms of PAH all had elevated micronuclei after DNA DSB-inducing damaging agents. Future studies with larger cohorts and exome DNA sequencing may help uncover consistent changes in either antioxidant defense enzymes and/or DNA repair enzymes that help predispose certain people to PAH. A more mechanistic understanding of genome instability in PAH will direct measurement of either mitochondrial or nuclear DNA damage, using antibodies to certain oxidative base lesions (11), quantitative PCR (13), or mass spectrometry (14). Finally, direct measures of DNA repair enzyme levels or overall capacity in blood using new techniques such as high-throughput flow cytometry (15) would also be helpful to answer these questions.