DNA damage has the potential to alter many essential cellular processes through blocked or error-prone replication and transcription, and if left unrepaired could adversely impact the overall integrity of cells and tissues. At an organismal level, unrepaired or mis-repaired DNA damage that arise from environmental exposures or endogenous metabolites can generate base substitutions, insertions, deletions, and chromosomal loss or rearrangements, with the potential to produce disease outcomes, including neurodegeneration, metabolic syndromes, and cancer. Rescue of replication of damaged genomic DNA can occur through the activities of translesion DNA polymerases, though the fidelity of these bypass reactions is generally low [1], [2]. Alternatively, to maximize genomic integrity and rescue stalled replication and transcription, cells possess multiple strategies including i) repair of DNA damage via high fidelity excision followed by gap-filling DNA synthesis [3], [4], [5], [6], ii) direct enzymatic DNA damage reversal [7], iii) mitigation of DNA damage impact via recombinational processes [8], and iv) minimal loss of genetic material via non-homologous or microhomology-mediated end-joining pathways [9], [10].

Except for direct damage reversal pathways, excision repair strategies involving damage-specific recognition, removal and repair offer the highest fidelity outcomes and include base excision repair (BER) [4], nucleotide excision repair (NER) [5], and nucleotide incision repair (NIR) [6]. Although these pathways differ in the detailed parameters that guide recognition, incision, and resynthesis, the overall objective is to create damage-free templates on which high-fidelity synthesis can occur. In general, substrates that are recognized and corrected by BER and NIR are classified as simple base modifications that minimally distort the structure of B-DNA, while typical NER substrates are categorized as bulky, helix-distorting adducts [3], [4], [5], [6]. Although most DNA damages are repaired via one predominant pathway, there is a limited number of lesions that appear to be substrates for both BER and NER [11], [12], [13], [14], with aflatoxin-induced guanine adducts being a well characterized example [12], [13], [14].

Aflatoxin B1 (AFB1) is a recognized liver carcinogen. Human exposures occur from ingestion of foods contaminated with various fungal species that belong to the genus, Aspergillus [15]. Except for rare cases of acute aflatoxicosis with near-term mortalities [16], [17], most human exposures to aflatoxin are anticipated to be at low concentrations spread over an individual’s lifetime [18]. Epidemiological evidence for these chronic exposures has been revealed by analyses of urine samples for the presence of aflatoxin metabolites [19], [20], [21], [22]. Following ingestion and transport to the liver, the toxin is metabolically activated by cytochrome P450s to the reactive AFB1 exo-8,9-epoxide [23], [24], which primarily reacts with guanines in DNA [25]. The resulting short-lived cationic AFB1-N7-Gua species either depurinates, generating an abasic (apurinic/apyrimidinic, AP) site, or undergoes imidazole ring opening to form the stable and persistent 8,9-dihydro-8-(2,6-diamino-4-oxo-3,4-dihydropyrimid-5-yl-formamido)-9-hydroxyaflatoxin B1 (AFB1-FapyGua) [14], [23], [26], [27], [28]. Thus, it is anticipated that under chronic exposure, aflatoxin DNA adducts are primarily represented by AFB1-FapyGua.

Analyses of aflatoxin-induced carcinogenesis in mice require an acute, early-life exposure in which the window of cancer susceptibility is within the first two weeks of life [29], [30], [31]. In these studies, aflatoxin is delivered via intraperitoneal injection. Although the route and conditions of exposure differ from chronic dietary exposure in humans, the parameters which define the kinetics of formation and disappearance of aflatoxin adducts in mouse DNA are applicable to human exposures. AFB1 DNA adducts have been measured in DNA repair-proficient mice with both AFB1-N7-Gua and AFB1-FapyGua readily detectable 6 h post-exposure. By 48 h, the level of AFB1-N7-Gua was reduced by over 10-fold, while no statistically significant difference was observed between the two time points for AFB1-FapyGua [14]. In agreement with this trend, levels of AFB1 adducts have also been analyzed using a rat model [27]. After a single aflatoxin dose, AFB1-N7-Gua was initially the most abundant form but depleted steadily over a period of a few days. Approximately 20% of AFB1-N7-Gua had converted to the imidazole ring-opened form and the rest had been removed from the genome, such that 98% of the aflatoxin adducts at 72 h were AFB1-FapyGua [27].

The biological importance of AFB1-induced adducts is evident from data showing that replication past either of these adducts is highly mutagenic (80-97% for AFB1-FapyGua [32], [33] and 32% for AFB1-N7-Gua [34] in primate cells), with mutational spectra corresponding to the spectra of AFB1-induced mutations in exposed cells or animals (predominantly G to T transversions) [35], [36], [37]. BER-initiated repair of the highly mutagenic AFB1-FapyGua adduct is catalyzed by endonuclease VIII-like 1 (NEIL1) [14], [38], [39], [40], [41], a bifunctional glycosylase associated with pre-replicative repair [42]. It first hydrolyzes the N-glycosidic bond between the base and sugar moieties, and cleaves the resulting AP site in sequential β,δ elimination reactions [43]. The biological role of NEIL1 in removal of AFB1-FapyGua has been validated by a study showing that DNA in Neil1-deficient mice accumulated >2-fold higher amounts of AFB1-FapyGua compared to wild-type (wt) mice [14], with levels ranging from ≈20 to ≈50 lesions per million nucleotides [44], [45].

In addition to recognition and excision of AFB1-FapyGua, NEIL1 removes a variety of base lesions including cis- and trans-thymine glycols (ThyGly), 5-hydroxy-5-methylhydantoin (5-OH-5-MeHyd), 8-hydroxyadenine (8-OH-Ade), 2,6-diamino-4-hydroxy-5-formamidopyrimidine (FapyGua), 4,6-diamino-5-formamidopyrimidine (FapyAde), urea, methyl- and nitrogen mustard-FapyGua, and psoralen crosslinks [38], [39], [41], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57]. Given the protective role of NEIL1 in cleansing the genome of both oxidatively-induced and AFB1-FapyGua lesions, deficiencies in this glycosylase are believed to contribute to the initiation and progression of aflatoxin-induced hepatocellular carcinoma (HCC). This has been explored in a murine model, where Neil1 knockout mice had a 3.15-fold higher risk of developing a tumor after a single aflatoxin dose compared to wt-mice. NER deficient mice had only a 1.2-fold increased risk relative to wt-mice [14].

Although only a small fraction of the global human population suffers from NEIL1 deficiency, naturally occurring polymorphisms have the potential to pathogenically alter the enzyme’s conventional structure and function. It is hypothesized that such single nucleotide polymorphic (SNP) variants of NEIL1 could increase human risk for HCC, especially in regions with high occurrences of aflatoxin exposures and hepatitis B viral (HBV) infections [58]. This hypothesis is partially supported by DNA sequencing analyses conducted in Qidong China. The data revealed that out of 49 HCC patients, 3 of them contained SNP NEIL1 variant alleles [39], [59]. Given that these variants present only in ≈0.1% to 2% of the general population in East Asia, their presence in such a small cohort suggests an enrichment in HCC patients. Since the aflatoxin exposures and HBV infections represent major problems in the developing countries of South Asia [60], [61], the focus of this study was to characterize the most common SNP NEIL1 variants in this region, Q67K, T103A, P206L, and T278I.