ATI-2173, a Novel Liver-Targeted Non-Chain-Terminating Nucleotide for Hepatitis B Virus Cure Regimens

ATI-2173 is a novel liver-targeted molecule designed to deliver the 5′-monophosphate of clevudine for the treatment of chronic hepatitis B infection. Unlike other nucleos(t)ides, the active clevudine-5′-triphosphate is a noncompetitive, non-chain-terminating inhibitor of hepatitis B virus (HBV) polymerase that delivers prolonged reduction of viremia in both a woodchuck HBV model and in humans for up to 6 months after cessation of treatment. However, long-term clevudine treatment was found to exhibit reversible skeletal myopathy in a small subset of patients and was subsequently discontinued from development.


H epatitis B virus (HBV) and subsequent chronic hepatitis B infection has a global
impact on over 350 million individuals, predominantly impacting sub-Saharan Africa and East Asia (1). Left untreated, chronic HBV can lead to impaired liver function, cirrhosis, and eventually hepatocellular carcinoma (1). Accordingly, great effort has been invested into the development of treatments and, ultimately, a cure for HBV infection. Current options comprise interferon-␣, which comes with myriad side effects, as well as chain-terminating nucleoside analogues, including entecavir, tenofovir, lamivudine, adefovir, and telbivudine (2). Although chain-terminating nucleoside analogues have a much-improved safety profile over interferon, they typically necessitate a longer duration of treatment that can result in viral breakthrough with drug resistance (3,4) and have a low cure rate (5). Altogether, the landscape of HBV therapy is complex for patients and primed for the introduction of a curative treatment regimen (6).
Clevudine (L-FMAU; [1-(2-fluoro-5-methyl-␤,L-arabinofuranosyl) uracil]), a small molecule inhibitor of HBV polymerase, was found to have antiviral activity in a woodchuck model of HBV infection nearly 20 years ago (7). Curiously, there was a dose-dependent delay in reemergence of viremia following withdrawal of treatment, with a select few animals remaining suppressed for months during the follow-up period, differentiating clevudine from the chain-terminating nucleoside analogues. In a subsequent phase IIA 28-day clinical trial, clevudine was found to be safe and effective (8). At the end of treatment, viral load (HBV DNA) was reduced by 2.5 to 3.0 log 10 , and viral suppression was sustained around a 2 log 10 reduction from baseline for 6 months following treatment. Further, alanine aminotransferase (ALT) levels were reduced in all cohorts, with a subset of patients showing sustained normalized ALT at a 1-year follow-up.
The potent and prolonged HBV suppression (and associated biochemical response) was reproducible in multiple ensuing clinical studies, including 12-week treatment regimens (9, 10) and 24-week regimens in both e-antigen-positive (HBeAg ϩ ) (11) and -negative (HBeAg Ϫ ) (12) patients. Despite the demonstrated safety of clevudine up to 24 weeks in the above-mentioned studies, reversible proximal myopathy symptoms emerged in South Korean patients after at least 8 months of clevudine exposure (13)(14)(15), prompting further investigation. Mitochondrial damage was not readily observed in the first report of clevudine-associated myopathy (14), and mitochondrial DNA polymerase ␥ was not found to be a target of clevudine in preclinical studies (16). However, a later report demonstrated long-term clevudine-associated myopathy was linked to mitochondrial injury by electron microscopy and mitochondrial DNA depletion (15), possibly due to exhaustion of available thymidine kinase 2 (TK2) (17), a key enzyme needed for mitochondrial DNA synthesis and utilized in the conversion of clevudine to the 5=-monophosphate. Clevudine (Fig. 1A), an unnatural L-nucleoside thymidine analogue (16), is unique as a noncompetitive, non-chain-terminating inhibitor of the HBV polymerase (18). Clevudine does so via a series of intracellular phosphorylations to generate the active 5=-triphosphate molecule (clevudine-5=-triphosphate) (16,19) (Fig. 1B). ATI-2173 is a new chemical entity (patent no. 10,683,319, formerly EIDD-2173) derived from clevudine with a 5=-phosphoramidate modification (Fig. 1A) and the potential of reducing peripheral clevudine exposure by bypassing the first clevudine phosphorylation step, leading to ion trapping in the liver (20). Of note, any ATI-2173 that does enter peripheral circulation would not be a target for intracellular TK2, potentially mitigating the risk of enzymatic depletion and subsequent skeletal mitochondrial damage. We sought to define the anti-HBV biological and pharmacological properties of ATI-2173. Our results show that ATI-2173 is a highly selective and potent inhibitor of HBV replication in vitro and demonstrated a robust preclinical safety profile. Oral administration of ATI-2173 leads to a marked reduction in peripheral plasma clevudine while delivering the 5=-monophosphate to the liver. Altogether, ATI-2173, via the active 5=-triphosphate, is a potent non-chain-terminating anti-HBV nucleotide analogue that provides an improved pharmacokinetic profile over clevudine.

RESULTS
ATI-2173 is converted to the active 5=-triphosphate in hepatocytes. ATI-2173 was incubated with cryopreserved mouse, rat, dog, monkey, and human hepatocytes for 4 h. Although ATI-2173 was not completely consumed, hepatocyte preparations from mice, rats, dogs, monkeys, and humans revealed five distinct metabolites, including clevudine and the 5=-monophosphate (MP) (Fig. 2A). The most abundant metabolite (other than the prodrug, clevudine, or MP) was M411 (M1), a product of rapid, nonenzymatic hydrolysis. ATI-2173 likely undergoes a preliminary hydrolysis step at the carboxyl ester catalyzed by either human cathepsin A (Cath A) and/or carboxylesterase 1 (CES-1), as is the case for other nucleos(t)ide phosphoramidates (20). This intermediate metabolite then undergoes rapid nonenzymatic nucleophilic attack on the phos-phorous by the carboxyl group, leading to a short-lived intermediate that rapidly and noncatalytically hydrolyzes into the stable M1 metabolite. The relative abundance of each metabolite identified (Fig. 2B) indicates that ATI-2173 and M1 are the two most abundant species in human hepatocytes, followed by clevudine and MP, and finally two other minor metabolites (M415 and M453) that together account for less than 0.5% of the identified compounds. Following the formation of M1, the amino acid is cleaved to generate MP, putatively by the histidine triad nucleotide-binding protein 1 (HINT-1), a known bio-activator of pronucleotides (20)(21)(22). From here, the MP follows successive metabolism to generate the 5=-di-and 5=-triphosphates (23), which are not identified in the present study due to the relatively short incubation period. Figure 2C describes the proposed metabolites and associated metabolic enzymes. Sofosbuvir, an FDAapproved drug used in the treatment of HCV also possesses a 5=-phosphoramidate, undergoes a series of similar intracellular transformations, yielding an intermediate metabolite, before being converted to the 5=-monophosphate, 5=-diphosphate, and finally the active 5=-triphosphate (20). Though the backbone of ATI-2173 is an L-nucleotide, the metabolic pathway for releasing the 5=-monophosphate from the 5=-phosphoramidate in hepatocytes appears to follow a similar pathway to that of Sofosbuvir, a D-nucleotide.
ATI-2173 is a potent and specific suppressor of HBV replication. Clevudine was found to have an anti-HBV EC 50 of around 0.1 M in cell-based in vitro systems (16,24,25). In humans, a dose found to be effective in the clinic (30 mg) corresponds with ϳ1 M plasma concentration and an intracellular 5=-triphosphate concentration of ϳ10 M (26) in primary human hepatocytes, although an EC 50 value for clevudine in primary human hepatocytes has not been reported. As ATI-2173 is a liver-targeting 5= phosphoramidate of clevudine, its active 5=-triphosphate would have similar in vitro activity to clevudine. ATI-2173 was incubated with HepG2.2.15 cells and primary human hepatocytes, and HBV DNA was quantified by quantitative PCR (qPCR). ATI-2173 was found to reduce extracellular HBV DNA in both cell types, with an EC 50 of 0.26 M in HepG2.2.15 cells ( Fig. 3A; 50% cytotoxic concentration [CC 50 ] Ͼ10 M, Fig. S1) and an EC 50 of 1.31 nM in primary human hepatocytes (PHHs) ( Fig. 3B; CC 50 Ͼ20 nM, Fig. S1). In these same assays, control compounds entecavir (EC 50 of 0.6 nM, CC 50 of Ͼ50 nM), clevudine (EC 50 of 0.1 m, CC 50 of Ͼ10 M), and peg-IFN-␣2a (EC 50 of 2 IU/ml, CC 50 of Ͼ100 IU/ml) demonstrated potent and comparable anti-HBV activity (Fig. S2). In the same HepG2 assay, ATI-2173 showed similar activity across HBV genotypes A, B, C, D, E, F, G, and H (EC 50 of 212 nM to 718 nM) ( Fig. 3C and D). The EC 50 variability is consistent with strain-to-strain differences, as seen with our positive-control compound, entecavir (Fig. S3)  (data not shown). Finally, when combined with other anti-HBV compounds in primary human hepatocytes, ATI-2173 had an anti-HBV activity that was additive with GLS4, lamivudine, and tenofovir (Table 1). Entecavir and adefovir had synergistic anti-HBV activity when combined with ATI-2173 (Table 1). Overall, these data indicate that ATI-2173 is an HBV-specific inhibitor of viral replication with nanomolar potency in primary hepatocytes and the potential for additive/synergistic activity when combined with other anti-HBV compounds. ATI-2173 is cross-resistant with other nucleosides. Viral break-through, a phenomenon linked to resistance mutations in the HBV polymerase (27), has been demonstrated for chain-terminating nucleos(t)ides adefovir, lamivudine, telbivudine, and entecavir, among others (28)(29)(30). Although clevudine resistance and viral breakthrough was not found in 24-week, phase III clinical trials (11,12), several patients did experience viral break-through associated with multiple mutations, predominantly M204I (4), after nine or more months of treatment. Notably, tenofovir and adefovir were active against these clevudine-resistant mutants, but not lamivudine nor entecavir (4). Since ATI-2173 and clevudine generate the same active 5=-triphosphate, it was hypothesized that ATI-2173 has the same resistance profile as that of clevudine. In HepG2 cells,  ATI-2173 was incubated against ten mutant strains of HBV, eight of which confer nucleos(t)ide resistance (30)(31)(32), and two CpAM-resistant strains as non-nucleos(t)ide controls.
While ATI-2173 effectively inhibited replication of wild-type HBV, ATI-2173 resistance generally overlapped with lamivudine resistance, with the exception of M204V. The double and triple mutants associated with entecavir (and multidrug) resistance, rtS202GϩM204I and rtS202GϩM204IϩM250V, also demonstrated cross-resistance with ATI-2173 ( Table 2). As has been seen with other nucleosides (33,34), the A181V mutation associated with adefovir resistance also generated resistance to ATI-2173. Based on the mechanism of action, the CpAM-resistant mutants, P25A and T33N, showed no cross-resistance with ATI-2173. Table 2 summarizes the cross-resistance found with ATI-2173 and entecavir, lamivudine, GLS4, or AT-130 resistance mutants.
ATI-2173 has no apparent toxic effect on cells. ATI-2173 was evaluated for potential acute cellular and mitochondrial toxicity. Proliferating cells (HK-2 human kidney cells and SkMC human skeletal muscle cells) and nonproliferating cells (cryopreserved primary human hepatocytes, cryopreserved iCell cardiomyocytes, and cryopreserved human bone marrow mononuclear cells) were incubated with increasing concentrations of ATI-2173 or the control compound doxorubicin. ATI-2173 up to 100 M did not demonstrate any cytotoxicity in primary hepatocytes, kidney cells, or cardiomyocytes ( Fig. 4A to C). Minimal cytotoxicity was observed for bone marrow cells and skeletal muscle cells at concentrations of 100 M or greater, (Fig. 4D and E), which is Ͼ100-fold higher than relevant clinical doses. Similar to ATI-2173 and consistent with previous literature (35), clevudine did not demonstrate cytotoxicity, except minimally in bone marrow cells (CC 50 of Ͼ100 M) (Fig. S4).
The glucose/galactose assay was conducted in HepG2 cells with increasing concentrations of ATI-2173 and the control compound chloramphenicol to assess potential mitochondrial toxicity. Although cell toxicity increased with chloramphenicol (due to mitochondrial toxicity), ATI-2173 demonstrated no apparent toxicity, with no differences in cell viability in glucose-supplemented medium versus galactose-supplemented medium (Fig. 4F), indicating no apparent mitochondrial toxicity. Similarly, clevudine demonstrated no differences between glucose and galactose cell viability, indicating no apparent mitochondrial toxicity (Fig. S4). Altogether, these data suggest ATI-2173 has a low risk of acute cytotoxicity and mitochondrial toxicity at doses relevant to the clinic.
ATI-2173 is orally bioavailable and liver targeted. Clevudine is orally bioavailable and yielded plasma concentrations up to 1 M at doses found to be both effective (8,10) but also linked with skeletal myopathy (15). Ideally, an improved pharmacokinetic profile would demonstrate a significantly reduced plasma exposure of clevudine following ATI-2173 dosing, while maintaining similar liver exposure of the 5=-triphosphate. Sprague-Dawley rats were given equimolar, single oral doses of either ATI-2173 (50 mg/ a ATI-2173 was tested against eight nucleoside-resistant HBV genotypes and two cpAM-resistant HBV genotypes. Symbol -, not applicable. kg) or clevudine (25 mg/kg), and plasma samples were collected at 1, 2, 4, 6, 8, and 24 h postdose. Figure 5A shows a nearly 5-fold reduction in peak plasma concentration of clevudine following ATI-2173 administration compared to clevudine administration at an equimolar dose. Clevudine and 5=-triphosphate levels in liver and skeletal muscle tissue samples from rats dosed as described above were measured at 1, 2, 4, 6, 8, and 24 h postdose. Figure 5B demonstrates nearly identical liver exposure to the 5=triphosphate following equimolar dosing of ATI-2173 and clevudine, despite considerable reduction in plasma exposure to clevudine with ATI-2173 dosing (Fig. 5A). Accordingly, skeletal muscle concentrations of clevudine (Fig. 5C) and the 5=-triphosphate (Fig. 5D) showed a nearly 10-fold and 4-fold reduction in peak concentrations, respectively, following administration of ATI-2173 versus clevudine. These results are consistent with the marked reduction in plasma exposure to clevudine. Rats administered a single oral 50 mg/kg ATI-2173 dose had no discernible differences in plasma exposure (maximum concentration of drug in serum [C max ] or area under the concentration curve from 0 to last [AUC 0-last ]) to ATI-2173 or clevudine (data not shown) when fed or fasted. ATI-2173 underwent high hepatic extraction (82%) in cynomolgus monkeys with portal vein cannulation following single oral (20 mg/kg) and intravenous (2 mg/kg) administration (data not shown). These results indicate ATI-2173 is both orally bioavailable and liver targeted, which leads to reduced systemic plasma and skeletal muscle clevudine exposure while still maintaining effective concentrations of active 5=-triphosphate in the liver.

DISCUSSION
The active 5=-triphosphate of clevudine has a unique mechanism of action as a noncompetitive, non-chain-terminating inhibitor of the HBV reverse transcriptase (18). It binds to and distorts the active site of the polymerase such that it blocks DNA replication at all stages (priming initiation, priming polymerization, and DNA synthesis), but is not incorporated into the elongating chain, unlike traditional chain-terminating nucleos(t)ide analogues. The 5=-triphosphate is also unique from other nucleos(t)ide analogues in that as few as 4 weeks of daily administration can maintain suppressed viremia for months following treatment cessation in the woodchuck model of HBV (7) and in chronically infected HBV patients (8). ATI-2173 is a novel 5=-phosphoramidate of clevudine, which inhibits HBV with nanomolar potency in primary human hepatocytes. No significant acute, off-target mitochondrial or cellular toxicities were detected in vitro at clinically relevant concentrations. Sequence divergence among HBV populations throughout the world has led to different genotype classifications. In general, genotypes A and D predominate in central and southern Europe (36), genotype D prevails in the middle east (37)(38)(39), genotypes B and C dominate Asia (40,41), and genotypes E to H are less common or more globally restricted (42). ATI-2173 demonstrates similar activity across all genotypes.
Similar to chain-terminating nucleos(t)ide analogues, the active 5=-tripohosphate of ATI-2173 is susceptible to a number of resistance mutations and long-term monotherapy may be challenging with viral breakthrough and drug resistance (4). Mutations associated with clevudine resistance in the clinic, such as M204I, were also linked with ATI-2173 resistance and overlap lamivudine and entecavir mutations. Overall, ATI-2173 resistance overlaps with lamivudine and entecavir. A181V, an adefovir-associated resistance mutation, was also found to reduce ATI-2173 anti-HBV EC 50 . Combination therapy in the clinic of 20 mg clevudine and adefovir effectively maintained viral suppression at 96 weeks without generating resistance (43). In primary human hepa- tocytes, ATI-2173 demonstrated additive-to-synergistic anti-HBV activity in combination with adefovir, entecavir, and tenofovir, indicating combination therapy may lead to improved treatment or cure regimens. Though cross-resistance and viral reemergence could be of clinical concern for ATI-2173, chain-terminating nucleos(t)ide analogues with a higher barrier to resistance, such as tenofovir (44), could be advantageously paired with ATI-2173 to target multiple mechanisms of DNA replication, preventing viral breakthrough and emergence of HBV drug resistance, and result in a potent combination to potentially shut down the HBV polymerase activity.
ATI-2173 targets the 5=-monophosphate to the liver and leads to an improved pharmacokinetic safety profile. Monkeys dosed with ATI-2173 showed an 82% hepatic extraction ratio, while rats dosed with ATI-2173 had significant reductions in plasma clevudine, skeletal muscle clevudine, and skeletal muscle 5=-triphosphate exposures compared to rats dosed with equimolar clevudine. Despite significant reductions in peripheral clevudine, liver concentrations of the 5=-triphosphate remained nearly equivalent between ATI-2173-and clevudine-treated rats, indicating ATI-2173 may maintain the same benefits of the active 5=-triphosphate in the liver while minimizing the risks associated with peripheral clevudine exposure. The 5=-phosphoramidate of ATI-2173 generates the 5=-monophosphate bypassing the first phosphorylation step of clevudine, which uses the TK2 enzyme and may be responsible for the reversible skeletal myopathy seen in chronically dosed patients (15,17,45). Thus, ATI-2173 that does leak into the systemic circulation may be less likely than clevudine to lead to myopathy. Altogether, ATI-2173 is a novel 5=-phosphoramidate of clevudine that delivers the 5=-monophosphate to the liver, reducing peripheral exposure to clevudine and potentially reducing the risk of skeletal myopathy. A drug candidate such as ATI-2173 that maintains 5=-triphosphate (a nucleotide analogue with a unique mechanism of action) levels in the liver but decreases plasma nucleoside exposure may have an improved clinical profile, particularly when paired with a chain-terminating nucleos(t)ide analogue that has a different mechanism of action and different bioactivation pathway (e.g., tenofovir) for chronically infected HBV patients (18). To this end, a phase I clinical trial is ongoing to investigate the safety, tolerability, pharmacokinetics, and anti-HBV activity of ATI-2173 in healthy subjects and subjects with chronic HBV (46).
Anti-HBV activity in HepG2.2.15 cells. ATI-2173 was obtained from WuXi SynTheAll. Entecavir (positive control) was obtained from Shanghai Titan Scientific Co., Ltd. HepG2.2.15 cells (derived from human hepatoblastoma cell-like HepG2) contain four 5= to 3= tandem copies of the HBV genotype D and were provided by WuXi. HepG2.2.15 cells were cultured in Dulbecco's modified Eagle medium (DMEM)/ F12 (Invitrogen, 11330032) supplemented with 2% fetal bovine serum (FBS) (HyClone, SV30087.03), 1% L-glutamine, 1% minimum essential medium (MEM) nonessential amino acid solution (NEAA), and 1% penicillin-streptomycin solution. QIAamp 96 DNA blood kit was purchased from Qiagen (51162). On day 1, HepG2.2.15 cells were seeded in a 96-well plate at a density of 6.0 ϫ 10 4 cells/well and cultured at 37°C and 5% CO 2 . Culture medium containing ATI-2173 or entecavir (serially diluted) was added to cells on day 2 and refreshed on day 5. ATI-2173 dilutions were: 0.0006, 0.0024, 0.0098, 0.0391, 0.1563, 0.6250, 2.5000, and 10.0000 M. On day 8, cell culture supernatants were collected and DNA was extracted by the QIAamp 96 DNA blood kit. qPCR was run at 95°C for 10 min, then cycled at 95°C for 15 s, 60°C for 1 min, for a total of 40 cycles. HBV DNA content in the samples was calculated according to the standard curve and C T (cycle threshold of qPCR) values of a sample. Data were normalized for maximum and minimum copy numbers.
For HBV combination studies, primary human hepatocytes were treated as described above and single-compound assays were conducted to determine the appropriate test concentrations. ATI-2173 was tested at concentrations between 0.5 nM and 8 nM. Entecavir was obtained from Shanghai Titan Scientific Co., Ltd. and tested at concentrations between 0.0025 nM and 0.04 nM. GLS4 was obtained from WuXi AppTec and tested at concentrations between 50 nM and 800 nM. Lamivudine was obtained from Dalian Meilun Biotech Co., Ltd. and tested at concentrations between 2.5 nM and 40 nM. Tenofovir DF was obtained from Shanghai Titan Scientific Co., Ltd. and tested at concentrations between 0.125 nM and 2 nM. Adefovir dipivoxil was obtained from Sun Chemical Technology (Shanghai) Co., Ltd. and tested at concentrations between 0.25 nM and 4 nM. Combination indexes were determined using MacSynergy and the 3-D model described by Prichard and Shipman (47). Combination indexes less than 25 were considered additive, between 25 and 50 were considered minor synergism, between 50 and 100 were considered moderate synergism, and greater than 100 were considered strong synergism.
Cytotoxicity. HK-2 human kidney cells were obtained from ATCC (Manassas, VA) and cultured in Gibco Keratinocyte serum-free medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% FBS (Peak Serum, Wellington, CO). Human skeletal muscle cells were obtained from Lonza (Walkersville, MD) and cultured in SkGM-2 skeletal muscle cell growth medium-2 (Lonza). Both HK-2 and skeletal muscle cells were cultured humidified at 37°C and 5% CO 2 . Cryopreserved human primary hepatocytes and medium were received from Lonza. Cryopreserved iCell cardiomyocytes and cardiomyocyte medium were purchased from Stem Cell Technologies (Vancouver, BC, Canada). Cryopreserved human bone marrow mononuclear cells were purchased from Lonza with HemaTox Myeloid medium purchased from STEMCELL Technologies (Vancouver, BC, Canada). All cryopreserved cells (human primary hepatocytes, iCell cardiomyocytes, and human bone marrow mononuclear cells) were stored in vapor phase liquid nitrogen until the day of plating. ATI-2173 was obtained from WuXi and doxorubicin was purchased from Sigma Millipore (Burlington, MA). Serial dilutions of compounds in DMSO were as follows: 100, 10, 1, 0.1, 0.01, 0.001, 0.0001, and 0.00001 M (final concentration of DMSO in medium was 0.5%).
HK-2 and skeletal muscle cells were seeded into 96-well plates at 5,000 cells/well and incubated overnight at 37°C and 5% CO 2 . Human hepatocytes were thawed, diluted in hepatocyte plating medium (Lonza) and seeded into a 96-well, collagen-coated plates at 2.5 ϫ 10 4 cells/well. Hepatocytes were incubated for 6 h at 37°C in 5% CO 2 , then the medium was aspirated and replaced with hepatocyte culture medium at 50 l/well, and finally incubated overnight. The iCell cardiomyocytes were thawed, suspended in iCell thawing/plating medium, and plated at 4 ϫ 10 4 cells/well in a gelatin-coated 96-well plate. iCells were incubated for 2 days at 37°C in 7% CO 2 . After 2 days of incubation, iCells were visually checked to ensure a uniform layer of attached cells. Human bone marrow cells were thawed, diluted in HemaTox medium, and seeded into a 96-well tissue culture plate at 5 ϫ 10 4 cells/well. Bone marrow cells were incubated for 1 h at 37°C in 5% CO 2 .
Following incubation, cells were dosed and incubated for 72 h at 37°C and 5% CO 2 . Following incubation, the endpoint cell viability CellTiter-Glo 2.0 assay (Promega, San Luis Obispo, CA) was performed and luminescence data collected on a Synergy 4 plate reader (Biotek, Winooski, VT). Where possible, the 50% inhibitory concentration (IC 50 ) was calculated using the Biotek GEN5 software.
Mitochondrial toxicity. Proliferating human hepatocytes (HepG2 cells) were acquired from Southern Research Institute. ATI-2173 was provided by WuXi and chloramphenicol was purchased from Millipore Sigma. Cells were cultured in 225 cm 2 tissue culture flasks (Corning) containing 50 ml of DMEM (Life Technologies), 20 mM glucose (Millipore Sigma, St. Louis, MO), and 10% FBS (termed "glucose medium"). At least 1 week prior to seeding in assay plates, the HepG2 cultures were split into parallel cultures, where one culture continued in glucose medium while the second culture was grown in DMEM containing 10 mM galactose (Sigma Millipore) and 10% FBS ("galactose medium"). Cells were incubated at 37°C with 5% CO 2 . On day 1, HepG2 cells were harvested from the glucose and galactose media parallel cultures seeded on 6-well plates with 1.5 ϫ 10 5 cells/well in the appropriate glucose or galactose medium. Following seeding, cells were dosed with controls and compounds to give in-well concentrations of 100, 10, 1, and 0.1 M (final concentration of DMSO was 0.5%). Cells were subcultured at a 1:3 ratio, and redosed on days 4 and 8. On day 11, cells were harvested and viable cell counts were determined by hemocytometer with trypan blue as the viability indicator.
Ethics statement. All animal protocols using rats were reviewed and approved by the Institutional Animal Care and Use Committees (IACUC) at the facilities where the studies were performed. Animal experiments were conducted in the Association for Assessment and Accreditation of Laboratory Care (AAALAC)-accredited and Public Health Services Animal Welfare Assurance-approved animal care suites at the Yerkes National Non-Human Primate research facility.
ATI-2173 pharmacokinetics and tissue concentration. Female Sprague Dawley (SD) rats (from Envigo, NJ), 225 to 249 g, were used in the studies. Clevudine and ATI-2173 were administered by oral gavage (p.o.) in carboxyl methyl cellulose/PEG 400/Tween 20 (0.5%/20%/5%). The oral doses were administered at 25 mg/kg and 50 mg/kg, respectively. Blood samples were collected at 1, 2, 4, 6, 8, and 24 h into lithium-heparin microtainer tubes (BD). Plasma samples were prepared within 30 min after collection by centrifugation at 2,000ϫ rcf for 10 min at 4°C and stored at -80°C before processing for analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Rat organs (skeletal muscle, liver, and heart) were collected from all rats immediately following blood collection at all time points. The tissues were immediately snap-frozen in liquid nitrogen and stored at -80°C before processing for analysis by LC-MS/MS.
Aliquots of rat plasma were precipitated with acetonitrile and centrifuged at 1,500 rcf for 10 min at 4°C. The supernatants were transferred to an HPLC vial for analysis by LC-MS/MS. Frozen rat tissue samples (ϳ50 mg) were extracted with 70% acetonitrile in water at 4°C and homogenized in a Bead Ruptor 24 (Omni International, Kennesaw, GA, USA). To remove large solids, homogenates were centrifuged at 1,500ϫ rcf for 10 min at 4°C. The clarified supernatants were transferred to microcentrifuge tubes and centrifuged again at 15,000ϫ rcf for 10 min at 4°C. The remaining supernatant was transferred to an HPLC vial for analysis by LC-MS/MS.
Samples were maintained at 4°C in a Leap Pal Autosampler (CTC Analytics AG, Zwingen, Switzerland). Plasma concentrations of ATI-2173 and clevudine were determined using qualified methods with an assay range of 10 to 1,000 ng/ml. Calibrators and quality control samples were prepared in blank rat plasma (Pel Freez Biologicals, Rodgers, AR, USA) and showed linearity with R 2 values of Ͼ0.99. QCs were prepared at 30, 500, and 900 ng/ml and were measured within Ϯ 30% of actual values. HPLC analysis of ATI-2173 was performed with an Agilent 1200 HPLC (Agilent Technologies, Santa, Clara, CA, USA) equipped with a column oven, UV lamp, and binary pump. For ATI-2173, a 5-min isocratic HPLC method with 0.1% formic acid in HPLC-grade water and methanol (45:55) was employed with an ACE 5 C 18 column (50 ϫ 2.1 mm, 5 m particle size) (Advanced Chromatography Technologies LTD., Aberdeen, Scotland). Mass spectrometry was performed on an API 4000 mass spectrometer (Sciex, Framingham, MA, USA) using positive electrospray ionization (ESI) in multiple reaction monitoring mode (MRM). For clevudine and its internal standard, a 6-min gradient HPLC method was used from 15% mobile phase B to 90% mobile phase B with a Hypercarb PGC column (100 ϫ 4.6 mm, 5 m particle size) (Thermo Fisher Scientific, Waltham, MA, USA). Mobile phase A consisted of 25 mM ammonium bicarbonate in HPLC grade water at pH 9.8 and acetonitrile was used as mobile phase B. Mass spectrometry was performed on a QTRAP 5500 mass spectrometer (Sciex, Framingham, MA, USA) using positive electrospray ionization (ESI) in multiple reaction monitoring mode (MRM). Data analysis was performed using Analyst Software (AB Sciex, Framingham, MA, USA).
Tissue levels were measured according to qualified LC-MS/MS methods with an assay range of 18 to 18,000 ng/g of tissue. Calibrators were prepared in blank rat tissue lysate and showed linearity with R 2 values of Ͼ0.99. HPLC separation was performed on an Agilent 1200 system (Agilent Technologies, Santa Clara, CA, USA) equipped with a column oven, UV lamp, and binary pump. For the separation of ATI-2173 and its internal standard, an Acclaim Polar Advantage II column (150 ϫ 4.6 mm, 5 m particle size) (Thermo Fisher Scientific, Waltham, MA, USA) was used. A 4-min gradient from 5% to 60% mobile phase B was used with 50 mM ammonium formate in HPLC-grade water as mobile phase A and acetonitrile as mobile phase B. Mass spectrometry was performed on a QTRAP 5500 mass spectrometer (Sciex, Framingham, MA, USA) using positive electrospray ionization (ESI) in multiple reaction monitoring mode (MRM). For the separation of clevudine, clevudine-5=-monophosphate, and clevudine-5=-triphosphate and internal standards, a Hypercarb PGC column was used (100 ϫ 4.6 mm, 5 m particle size) (Thermo Fisher Scientific, Waltham, MA, USA). A 7-min gradient from 5% to 90% mobile phase B was used with 25 mM ammonium bicarbonate in HPLC-grade water at pH 9.8 as mobile phase A and acetonitrile as mobile phase B. Mass spectrometry was performed on a QTRAP 5500 mass spectrometer (Sciex, Framingham, MA, USA) using negative electrospray ionization (ESI) in multiple reaction monitoring mode (MRM).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 1 MB.