|Home | About | Journals | Submit | Contact Us | Français|
Apurinic/apyrimidinic endonuclease 1 (APE1), an essential protein in mammals, is known to be involved in base excision DNA repair, acting as the major abasic endonuclease; the protein also functions as a redox coactivator of several transcription factors that regulate gene expression. Recent findings highlight a novel role for APE1 in RNA metabolism. The new findings are as follows: (i) APE1 interacts with rRNA and ribosome processing protein NPM1 within the nucleolus; (ii) APE1 interacts with proteins involved in ribosome assembly (i.e., RLA0, RSSA) and RNA maturation (i.e., PRP19, MEP50) within the cytoplasm; (iii) APE1 cleaves abasic RNA; and (iv) APE1 cleaves a specific coding region of c-myc mRNA in vitro and influences c-myc mRNA level and half-life in cells. Such findings on the role of APE1 in the posttranscriptional control of gene expression could explain its ability to influence diverse biological processes and its relocalization to cytoplasmic compartments in some tissues and tumors. In addition, we propose that APE1 serves as a “cleansing” factor for oxidatively damaged abasic RNA, establishing a novel connection between DNA and RNA surveillance mechanisms. In this review, we introduce questions and speculations concerning the role of APE1 in RNA metabolism and discuss the implications of these findings in a broader evolutionary context.
There is an emerging body of evidence that links DNA repair proteins to specific aspects of RNA metabolism. Currently, little is known about how cells cope with damaged RNA, either modified or oxidized, but it is clear that such RNA can impair protein synthesis, thus affecting cell function and viability. Specific surveillance mechanisms are therefore needed to remove damaged molecules from the RNA pool to guarantee the biological integrity of cells.
The idea that quality control mechanisms might exist to repair RNA was brought to the forefront with the identification of the biochemical activities of the mammalian AlkB homologs. In particular, it was discovered that AlkB (from Escherichia coli) and the human homolog hABH3, besides being able to directly reverse alkylation damage on DNA bases, are able to demethylate damaged bases on RNA, thus playing a key role in the repair of specific RNA lesions (1, 30). While repair mechanisms have been demonstrated for alkylated RNA, such activity is not yet known for oxidatively damaged RNA. However, their existence seems improbable, in the absence of direct reversal strategies, due to the lack of a template for accurate repair, as in the case of double-stranded DNA. Since, under oxidative stress conditions, oxidation of RNA can occur up to a 10- to 20-fold-higher extent than oxidation of DNA (24), the question is: how is oxidized RNA specifically removed or repaired?
The recent findings of Berquist et al. (7), Barnes et al. (4), and Vascotto et al. (39) highlight a novel “moonlighting” role for the repair apurinic/apyrimidinic (AP) endonuclease 1 (APE1) in RNA metabolism, both as a possible “cleansing” factor for damaged abasic RNA and as a regulator of c-myc gene expression through mRNA decay.
After cloning by independent groups, first as a DNA repair enzyme (13, 32) and then as a redox protein (43), APE1 has been described, in about 400 papers, to operate in several biological contexts. APE1 is a protein of 318 amino acids with a well-established role in the base excision repair (BER) pathway, possessing a strong AP DNA endonuclease activity in vitro. Indeed, APE1 is believed to be the major AP endonuclease in mammals. In the BER process, APE1 hydrolyzes the phosphodiester backbone immediately 5′ to an AP site to generate a 3′-hydroxyl group and a 5′-abasic deoxyribose-5-phosphate. In addition to AP DNA endonuclease activity, APE1 has been shown to possess other nuclease activities, which presumably assist in DNA damage processing. These include 3′ DNA phosphodiesterase or phosphatase activities (12), as well as 3′-5′ DNA exonuclease activity (11, 34, 42). APE1 has also been demonstrated to have RNase H-like activity in vitro in degrading the RNA strand of a DNA-RNA duplex (5), yet the biological significance of this function remains unknown. A feature shared among these DNA nuclease activities is that they are performed by the same catalytic center, in which amino acids His309, Glu96, and Asp210 of APE1 play an essential role (28, 41).
As a redox factor (Ref-1), a function separable from its nuclease activities, APE1 serves to maintain critical cysteine residues of several transcription factors in their active reduced state (reviewed in reference 37). In this process, APE1 controls transcriptional regulation of gene expression by enhancing the DNA binding activity of AP-1, p53, HIF-1α, NF-κB, CREB, and Egr-1, to name a few (37).
In recent years, there have been several reports on the ability of APE1 to influence other biological processes. For instance, APE1 has been shown to inhibit the activation of PARP-1 during the response to DNA damage generated by oxidative stress, presumably by competing for binding to single-strand breaks, and as a result APE1 prevents apoptosis (31). APE1 is also involved in a caspase-independent cell death pathway potentially by serving as a substrate for granzyme A proteolytical inactivation (15), as well as in negative regulation of the parathyroid hormone gene by acting as a transcriptional repressor (8). In addition, it has been suggested that APE1 controls vascular tone and endothelial nitric oxide production by negatively regulating Rac1/GTPase activity (22). The new role for APE1 in regulating aspects of RNA metabolism, discussed in greater detail below, must now be considered as possibly contributing to any number of the previous biological outcomes.
Besides its understandable role in mitochondria as part of the genome surveillance system of this organelle, the discovery that APE1 has cytoplasmic localization, which was influenced by cancer type and grade at least in part, was “confusing” in terms of its known biological functions (reviewed in reference 38). To date, a causal involvement of APE1 dysregulation in disease pathology has remained largely unsubstantiated. The recent discoveries of the novel functions of APE1 in controlling RNA metabolism (4, 39) open up new paradigms that perhaps explain its cytoplasmic distribution, new areas of investigation to define the biological contributions of its multiple functions, and future avenues for translational research.
APE1 is able to cleave abasic RNA and may participate in an RNA quality control process (7, 39). Within nucleoli, the process of ribosome biogenesis, which includes rRNA metabolism, is regulated in part through an interaction of nucleophosmin (NPM1) and the N-terminal portion of APE1. Interestingly, this disordered protein domain of APE1 is a recent evolutionary acquisition and, thus, may represent a phylogenetic “gain of function.” The experimental data in support of a role for APE1 in RNA quality control are (i) its endonuclease activity on abasic RNA; (ii) the accumulation of oxidized rRNA, such as 8-hydroxyguanine (8-OHG) rRNA, in APE1-depleted cells; and (iii) the disruption of the APE1-NPM1 interaction in H2O2-treated cells, resulting in increased APE1 incision potential.
Why do APE1-depleted cells have increased 8-OHG rRNA content upon oxidative stress? This result could be explained assuming cells maintain the ability to excise oxidized bases from damaged RNA (27) and if APE1, as seen with DNA substrates (18), is required to stimulate this hypothetical glycosylase function. The existence of specific N-ribohydrolases (33), including the toxin ricin, has been documented. Additionally, a role for the YB-1 protein, a known APE1 interacting partner (10), in recognizing 8-OHG sites in RNA has been reported, although no specific enzymatic activity to remove oxidized bases has been found to date (17). Furthermore, it is reasonable to postulate that, in APE1-depleted cells, abasic RNA damage would also increase, as seen with abasic DNA damage (16). Unfortunately, at present, due to the lack of reliable methods, the amount of abasic RNA damage has not been quantified.
Due to its intrinsic chemical properties (i.e., mostly single stranded and with bases not protected by hydrogen bonding or by binding to specific structural proteins such as histones), its comparatively larger amount, and its intracellular location, RNA may be more susceptible to oxidative insults than DNA (29). Not only 8-OHG but also 5-hydroxycytidine, 5-hydroxyuridine, and 8-hydroxyadenosine have been identified in oxidized RNA (44). If not repaired, these damages can cause altered pairing in rRNA leading to ribosomal dysfunction and erroneous translation, thus significantly affecting overall protein synthesis rate and cellular integrity (14, 36). In agreement with these findings, there was a significant reduction in protein synthesis upon APE1 silencing coupled to cell growth arrest and apoptosis, without compensation by an APE1 mutant lacking its N-terminal domain, which is critical for the RNA binding function of the protein, suggesting that the APE1 RNA cleansing activity has a significant impact on cell viability (39).
Oxidative damage to RNA molecules, both coding for proteins (mRNA) and participating in translation (rRNA and tRNA), has been recently associated with the occurrence of neurodegenerative diseases, such as Alzheimer's, and cannot be excluded from playing a role in cancer development (6, 29). Thus, all hypotheses made about the involvement of APE1 in the development of disease pathology should be reconsidered in light of the recent findings (4, 39). Although the role that APE1 has in RNA-metabolic processes needs further study, the ability of the protein to recognize and endonucleolytically cleave abasic RNA suggests that APE1 participates in some aspect of RNA quality control, presumably as a novel “cleansing” enzyme to maintain a functional RNome.
Using rat liver polysomes as the starting material, biochemical purification was initially performed in an effort to identify an endoribonuclease(s) that is capable of cleaving a specific coding region of c-myc RNA. APE1 was unexpectedly discovered as one of the endoribonucleases (4). Immunodepletion experiments on the purified native enzyme and biochemical characterization of both the recombinant wild-type and mutant APE1 have further confirmed this finding. When HeLa cells were transiently and specifically knocked down with small interfering RNA against APE1, the steady-state c-myc mRNA levels increased 2- to 5-fold and c-myc mRNA half-life was stabilized. These results clearly demonstrate the ability of APE1 to influence the stability of c-myc mRNA and provide further support for its role in RNA metabolism. The effect on c-myc mRNA stability is likely a direct action of APE1, presumably through its endoribonuclease activity, and unlikely due to the other known functions of APE1, including its redox regulation of transcription factors. In the latter case, a decrease in active transcription factors upon knockdown of APE1 would have resulted in decreased c-myc mRNA expression.
The evidence that APE1 affects c-myc mRNA half-life inevitably leads one to ask the following questions concerning its more general role in mRNA decay. Is APE1 localized to cytoplasmic structures commonly associated with mRNA metabolism, such as P bodies or stress granules (2)? Is APE1 involved in mRNA surveillance mechanisms, such as nonsense-mediated decay (21)? Does APE1 cleave other mRNAs and influence their stability, gene expression, and therefore biological functions? Does APE1 cleave other types of RNAs, such as tRNA, rRNA, and microRNAs (miRNAs), and hence affect a broad range of cellular functions? Is the endoribonuclease activity of APE1 dependent upon cellular stress? The purified native APE1, which copurified with a few proteins, was about 40 times more active than the single polypeptide recombinant APE1 (W. C. Kim et al. and C. Vascotto et al., unpublished observations). So, does this mean there exist other cofactors that act to enhance the RNA-cleaving function of APE1? This conclusion seems possible, given that APE1 copurified with additional proteins (4) and interacts with proteins known to associate with RNA processing (39).
Are there any other DNA repair proteins with nuclease activity capable of cleaving RNA in the cytoplasm? In fact, FEN1 marginally satisfies such criteria. Like APE1, FEN1 is a multifunctional nuclear protein with a central role in DNA metabolism (26). Multiple functions of FEN1 include RNA primer removal during DNA replication, double-stranded DNA 5′-3′ exonuclease, gap endonuclease, and RNase H activity (26). Human FEN1 has been demonstrated to possess endoribonuclease activity against 5′ stem structures of synthetic and native RNAs in vitro (35). Interestingly, FEN1 is physically associated with heterogeneous nuclear ribonucleoprotein A1 (hnRNP A1), a protein that participates in RNA metabolism, and this association enhanced FEN1's DNA endonuclease activity (9). It is currently unknown if FEN1 has any endoribonuclease activity in the cytoplasm and therefore if it can influence mRNA expression and hence biological functions through this activity. Surprisingly, FEN1 has been recently discovered to participate in long-patch BER in mitochondria (25), sharing an additional feature with APE1.
Two independent studies provided in vitro and in vivo evidence that APE1 acts on RNA molecules and is involved in RNA metabolism (Fig. (Fig.1).1). First, upon interaction with NPM1, a protein involved in rRNA maturation and ribosome biogenesis, APE1 is localized to the nucleolus (39). Upon knockdown of APE1, impairment in translation and cell growth were observed. APE1 was also demonstrated to cleave abasic RNA (7, 39). Second, both the purified native and recombinant APE1 proteins were shown to cleave c-myc mRNA at identical sites (4). Transient knockdown of APE1 in this study led to increased steady-state c-myc mRNA levels, as well as an extended c-myc mRNA half-life. It is currently unclear if the two separate investigations are directly related, but what is clear is that APE1 can catalyze RNA cleavage in test tubes and can influence RNA processing in cells. One of the major challenges in the future is to provide evidence that APE1 can act on RNA molecules in cells and in animals to affect biological outcomes.
Although still in the early phases, the recent findings have raised many questions and speculations concerning the role of APE1 in RNA metabolism.
(i) Why should a protein involved in DNA repair play a role in RNA metabolism? One advantage would be that APE1 could protect genetic stability not only through its DNA repair activity but through its capacity to cleanse damaged RNA that may otherwise be inaccurately translated or to degrade unwanted foreign RNA, such as viral RNA.
(ii) Is APE1 an ancient protein with a previously unknown ancient role, or is it an ancient protein with a newly identified and yet unappreciated role? The current information regarding the primary amino acid sequence of APE1 across species seems to suggest a phylogenetic gain of function and hence supports the latter hypothesis. However, further experimental and bioinformatics studies that delineate the biochemical and higher-order structures of APE1 orthologs may reveal additional insights into this question.
(iii) Is there any role of APE1 in controlling gene expression through, for example, miRNA metabolism? APE1 is capable of cleaving miRNAs and reducing the ability of the Dicer enzyme to process pre-miRNAs in vitro (Kim et al., unpublished observation). Hence, it remains a possibility that APE1 can influence gene expression by operating during miRNA processing, potentially explaining APE1's ability to influence diverse biological functions.
(iv) Suppressing the amount of APE1 has proven to be effective in sensitizing cancer cells to chemotherapeutic agents. This finding has led to the proposal that selective inhibition of APE1's DNA repair activity is an attractive avenue for developing novel anticancer therapies (3). One could envision targeting the non-DNA repair functions of APE1 or using APE1's RNA repair and/or RNA-cleaving activities as an approach for the treatment of cancer or neurological disorders. However, such therapeutic goals await further understanding of the role of APE1 in RNA metabolism.
In closing, a need for productive biological cross talk between DNA repair enzymes and proteins involved in RNA metabolism seems reasonable. Perhaps an early example of this phenomenon was described for the ribosomal protein hS3. Specifically, hS3 has been found to protect human cells from genotoxic stress (19, 20) through (i) its DNase activity at abasic sites (40), (ii) its ability to stimulate the glycosylase activity of the uracil-DNA glycosylase hUNG (23), (iii) its ability to bind with high affinity to the oxidative lesion 7,8-dihydro-8-oxoguanine (8-oxoG) (19), and (iv) its functional interaction with other well-known DNA repair proteins such as hOGG1 and hAPE1 (20). Such findings would seem to justify further analysis of other proteins, such as FEN1, and may imply that nucleic acid processing enzymes are more strategically promiscuous than originally thought or may have evolved DNA-targeted functions after a prior life in the early RNA world. The observation of cytoplasmic localization for a “classic” DNA repair protein, beyond simply mitochondrial targeting, may therefore suggest much more than just an “abnormal” distribution pattern.
We regret the omission of many important references due to space limitations.
This work was supported by grants from MIUR (FIRB RBRN07BMCT_008) to G.T., NSERC Discovery Grant (grant no. 227158) to C.H.L., and the Intramural Research Program of NIH, National Institute on Aging to D.M.W.
Prof. Gianluca Tell is Associate Professor of Molecular Biology at the School of Medicine—University of Udine (UNIUD), Italy. He obtained his Ph.D. from the University of Trieste in Italy and was a visiting scientist in 1996 at the NCI, NIH, Bethesda, MD. His prevailing scientific interest is the study of molecular mechanisms of gene expression in the field of redox signaling and cell oxidative stress. In the last 10 years he devoted himself to the study of the role of the protein APE1 in transcriptional regulation in different cell models and human cancers, discovering a new unsuspected role for this protein in RNA metabolism and coordinating several research projects on these topics. He authored more than 100 publications in international peer-reviewed journals. Before joining UNIUD, he was an Assistant Professor at the University of Trieste and visiting professor in 2006 at the UTMB, Galveston, TX.
Dr. David M. Wilson III received a bachelor of arts degree in both biology and political science from Bucknell University in 1989. He completed his Ph.D. work in 1993 at Loyola University of Chicago, Stritch School of Medicine, as part of the Molecular Biology Program. Dr. Wilson performed his postdoctoral training at the Harvard School of Public Health until 1997, when he became a Senior Biomedical Scientist at Lawrence Livermore National Laboratory in the Biology and Biotechnology Research Program. While at Livermore, he was briefly an adjunct faculty member in the Radiation Oncology Department at the University of California Cancer Center—Sacramento. Dr. Wilson started at the National Institute on Aging Intramural Research Program in the Laboratory of Molecular Gerontology in 2002 and became a tenured Senior Investigator in 2008. He has been working on specific aspects of base excision DNA repair since a graduate student.
Dr. Chow H. Lee is an Associate Professor in the Chemistry Program, College of Science and Management, at the University of Northern British Columbia, Prince George, BC, Canada. He has been working on endoribonucleases and their role in the control of mRNA turnover and gene expression as it relates to cancer since 1996. Before joining the University of Northern British Columbia in 1998, Dr. Lee spent his postdoctoral research work at the Ontario Cancer Institute, BC Cancer Agency, and McArdle Laboratory for Cancer Research, University of Wisconsin—Madison. Dr. Lee obtained his Ph.D. from Flinders University in 1992 and his B.Sc. with first class honors from the University of New South Wales in 1988, both in Australia. Using biochemical purification techniques, he fortuitously discovered APE1 as an endoribonuclease that cleaves c-myc mRNA in vitro and regulates c-myc mRNA level and stability in cells.
Published ahead of print on 9 November 2009.