Antiviral Activity of Clevudine [L-FMAU, (1-(2-fluoro-5-methyl-b,L-arabinofuranosyl) uracil)] Against Woodchuck Hepatitis Virus

Replication and Gene Expression in Chronically InfectedWoodchucks (Marmota monax)

SIMON F. PEEK,1 PAUL J. COTE,2 JAMES R. JACOB,1 ILIA A. TOSHKOV,1 WILLIAM E. HORNBUCKLE,1 BETTY H. BALDWIN,1

FRANCES V. WELLS,2 C. K. CHU,3 JOHN L. GERIN,2 BUD C. TENNANT,1 AND BRENT E. KORBA2

L-FMAU [1-(2-fluoro-5-methyl-b,L-arabinofuranosyl) ura-cil] has been shown to be an effective inhibitor of hepatitis Bvirus (HBV) and duck hepatitis B virus replication in cellculture and duck hepatitis B virus replication in acutelyinfected Peking ducks. The woodchuck hepatitis virus(WHV) and its natural host, the Eastern woodchuck (Mar-mota monax), have been established as a predictive modelfor the evaluation of antiviral therapies against chronic HBVinfection. In this report, the antiviral activity of L-FMAUagainst WHV replication in chronically infected wood-chucks is described. Four weeks of once-daily oral adminis-tration of L-FMAU significantly reduced viremia, antigenemia,intrahepatic WHV replication, and intrahepatic expression ofwoodchuck hepatitis virus core antigen (WHcAg) in a dose-dependent manner. At the highest dose administered (10mg/kg/d), significant reductions of intrahepatic WHV RNAand covalently closed circular (ccc)WHV-DNA levels alsowere observed. The reduction in viremia was remarkablyrapid at the higher doses of L-FMAU, with greater than1,000-fold reductions in WHV-DNA serum levels observedafter as little as 2 to 3 days of therapy. Following the with-drawal of therapy, a dose-related delay in viremia reboundwas observed. At the highest doses used, viremia remainedsignificantly suppressed in at least one half of the treatedanimals for 10 to 12 weeks’ posttreatment. No evidence ofdrug-related toxicity was observed in the treated animals.L-FMAU is an exceptionally potent antihepadnaviral agentin vitro and in vivo, and is a suitable candidate for antiviral

therapy of chronic HBV infection. (HEPATOLOGY 2001;33:254-266.)

The nucleoside analogue, L-FMAU [1-(2-fluoro-5-methyl-b,L-arabinofuranosyl) uracil] has been shown to have signifi-cant antiviral activity against Epstein-Barr virus (EBV), hepa-titis B virus (HBV), and duck hepatitis B virus replication incell culture, and duck hepatitis B virus replication in acutelyinfected Peking ducks.1-10 Recent cell-culture studies havedemonstrated that HBV variants that have been associated withresistance to famciclovir and lamivudine therapy in chronicallyinfected patients are sensitive to L-FMAU to varying degrees.4

L-FMAU has markedly less toxicity than its D-enantiomer,D-FMAU, and L-FMAU does not induce an increase in lacticacid production in the human hepatoblastoma cell line,HepG2, as observed following exposure to D-FMAU and therelated nucleoside analogue, D-FIAU (fialuridine).1-3,5-8,11-15

Fialuridine demonstrated potent anti-HBV activity in clinicaltrials for chronic HBV infection, but also induced severe de-layed toxicity associated with lactic acidosis and hepatic fail-ure that resulted in the death of several patients.13,14 D-FMAU,D-FIAU, and D-FEAU have been shown to be potent inhibitorsof woodchuck hepatitis virus (WHV) replication in chroni-cally infected woodchucks, but all of these nucleoside ana-logues also demonstrated severe toxicity at daily doses greaterthan 2 mg/kg body weight.11,12,15

WHV and its natural host, the Eastern woodchuck (M.monax), constitute a useful model of HBV-induced disease,including hepatocellular carcinoma (HCC).16,17 A variety ofpublished studies from several laboratories have reported theuse of chronic WHV infection in woodchucks to investigatepotential antiviral therapies for chronic HBV infec-tion.11,15,18-36 Most of the antiviral agents that have been usedto treat chronic HBV infection (acyclovir,24 interferon alfa,25

emtricitabine [FTC],28,33 famciclovir [oral prodrug for penci-clovir],32,34 fialuridine [D-FIAU],11,15 ganciclovir,27 lamivu-dine [3TC],26,31,32,34,35 ribavirin,34 thymosin a1,20 vidarabine(ara-A),21,34 zidovudine (AZT),34 combination therapy witheither lamivudine/interferon alfa35 or lamivudine/famciclo-vir32) have shown similar relative antiviral activities againstWHV and similar toxicity profiles in chronically infectedwoodchucks. These studies demonstrate that the WHV/wood-chuck model of chronic HBV infection can be considered to bea predictive model for new, potential therapeutic applicationsagainst HBV infection in humans, especially nucleosides.

Abbreviations: L-FMAU, [1-(2-fluoro-5-methyl-b,L-arabinofuranosyl) uracil]; EBV,Epstein-Barr virus; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; WHV, wood-chuck hepatitis virus; PCR, polymerase chain reaction; WHsAg, woodchuck hepatitissurface antigen; WHcAg, woodchuck hepatitis core antigen; ccc, covalently closed cir-cular; SDS, sodium dodecyl sulfate; RI, replication intermediate; L-FMAU-TP, L-FMAU–triphosphate.

From the 1College of Veterinary Medicine, Cornell University, Ithaca, NY; 2Division ofMolecular Virology and Immunology, Georgetown University Medical Center, Rock-ville, MD; and 3Department of Chemistry, University of Georgia, Athens, GA.

Received August 3, 2000; accepted October 16, 2000.Supported by contract N01-AI-45179 between the National Institute of Allergy and

Infectious Diseases (NIAID) and Georgetown University, contract N01-AI-35164 be-tween the NIAID and the College of Veterinary Medicine of Cornell University.

Address reprint requests to: Brent E. Korba, Ph.D., Division of Molecular Virology andImmunology, Georgetown University Medical Center, 13 Taft Court, Suite 101, Rock-ville, MD 20850. E-mail: Korbabe@georgetown.edu.

Copyright © 2001 by the American Association for the Study of Liver Diseases.0270-9139/01/3301-0033$3.00/0doi:10.1053/jhep.2001.20899

254

L-FMAU has been shown to have a favorable pharmacokineticprofile and sufficient oral bioavailability in rats and woodchucksthat makes it suitable for once-daily administration.12,37,38 Studiesin cultured human erythrocytes demonstrate that L-FMAU istransported into cells by both nucleoside transporters and by non-facilitated membrane diffusion.9,39 L-FMAU has also been shownto be a substrate for both human thymidine kinase and humandeoxycytidine kinase, and is phosphorylated efficiently in a step-wise manner to mono-, di-, and triphosphate forms in culturedhuman cells.39

In this report, we examined the antiviral activity of L-FMAUagainst WHV replication in chronically infected woodchucksin a placebo-controlled, dose-ranging study. L-FMAU wasfound to be a remarkably potent and fast-acting antiviral agentagainst WHV replication in vivo in the absence of any obvioussigns of toxicity.

MATERIALS AND METHODS

Woodchucks. The woodchucks used in these studies were born toWHV-negative females in a breeding colony maintained at CornellUniversity. Animals were inoculated at 3 days of age with 5 millionwoodchuck infectious doses of a standardized WHV inoculum pool(WHV7p1) that characteristically produces a 65% to 70% chronicWHV carrier rate.17,40 The diet consisted of laboratory animal chowformulated for rabbits (Agway Red Rabbit Food, Syracuse, NY), spe-cially pelleted in blocks for woodchucks. Both food and water wereprovided ad libitum. Experiments were conducted in accordancewith the Guide for the Care and Use of Laboratory Animals (NationalAcademy Press, revised 1996) and were reviewed and approved bythe Cornell University Institutional Animal Care and Use Com-mittee.

General Experimental Design. Six groups of 4 adult chronic carrierwoodchucks (approximately 2 years old) were treated with L-FMAUat 1 of 6 doses: 0.03, 0.1, 0.3, 1.0, 3.0, or 10 mg/kg body weight. Anadditional group of 8 age-matched chronic WHV carriers served as aplacebo control. L-FMAU was solubilized in isotonic saline solutionand administered orally in a liquid diet once daily for 28 days.11 Theliquid diet was also administered daily to the control animals.

Serum samples were taken for analysis on the first day of treatmentbefore administration of the initial dose of drug (“day 0” or “week0”), after 0.5, 1, 2, 3, 5, 7, 14, 21, and 28 days of treatment, and at 1,2, 3, 4, 6, 8, and 12 weeks following the end of therapy. Needlebiopsies of the liver were obtained at the time of the first serumspecimen (“week 0”), at the end of the treatment period (week 4),and at 12 weeks’ posttreatment (week 16).11 Serum and liver biopsyspecimens were obtained while the animals were under general an-esthesia (ketamine/xylazine) using 14-gauge disposable biopsy nee-dles directed by ultrasound imaging as previously described.11 Theneedle was inserted at a site near the ventral midline just caudal tothe xiphoid cartilage and directed dorsolaterally and somewhat cra-nially into the margin of the left lateral lobe of the liver.11 Onespecimen was placed immediately in liquid nitrogen and stored at270°C for nucleic acid analysis.11 A second specimen was fixed inphosphate-buffered formalin, embedded in paraffin, sectioned, andstained with hematoxylin-eosin for histopatholologic analysis.11

Liver specimens were scored, under code, for severity of hepatitis(portal and parenchymal), bile duct proliferation, and fatty acidchange using a scale of 0 to 4 as previously described.11 Steatosis wasalso qualitatively classified, and oil Red O–stained sections were usedto confirm presence and severity.11

The general health of the woodchucks was assessed by daily ob-servations at the time they received food and water, at the time ofdrug (or placebo) administration, and at the times they were anes-thetized. Any abnormalities in behavior, appearance, or intake offood or water were recorded. Body weights were determined whenserum samples were taken. Hematologic (hematocrit, white bloodcells, red blood cells, segmented neutrophils, platelet count, etc.)

and routine blood chemistry (alanine transaminase, aspartatetransaminase, sorbitol dehydrogenase, g-glutamyl transpeptidase,lactate, bicarbonate, serum electrolytes, acid-base status, blood ureanitrogen, total bilirubin, total protein, albumin, glucose, serumlipase activity, etc.) analyses were performed at the beginning andend of the treatment periods, and at 4 and 12 weeks following termi-nation of treatment.11

Markers of WHV Replication/Infection. Viremia in serum samples wasassessed by different quantitative methods depending on the concen-tration of WHV DNA. Serum containing concentrations of WHVvirion DNA above 1 3 107 WHV genome equivalents per milliliter ofserum (WHVge/mL) were analyzed by dot-blot hybridization (four10-mL replicates per sample).41,42 Samples containing WHV DNAbelow the dot-blot sensitivity cutoff and above 1,000 WHVge/mLwere analyzed using a quantitative polymerase chain reaction (PCR)-based method.11,34

Serum that contained WHV-DNA levels below 1,000 WHVge/mLwas processed in duplicate 100-mL aliquots. Nucleic acids from theseserum samples were extracted by a proteinase K–sodium dodecylsulfate (SDS)/phenol/chloroform procedure, followed by ethanolprecipitation.41,42 WHV-DNA samples then were resuspended in 10mL of nuclease-free water and subjected to PCR amplification andblot hybridization using standards representing 30 to 1,000 WHVge/mL.11,34 The sensitivity of this detection procedure was approxi-mately 30 WHVge/mL serum.

Assessments of the level of WHV genome equivalents for the PCRanalyses were determined by direct comparisons with parallel PCRamplifications of a dilution series (30 to 1,000,000 WHVge/mL) ofthe standardized serum pool (WHV7p1). This pool was used to in-fect the experimental animals and has a known WHV genome con-tent.11,34,40 The dilution standards were included, in duplicate, witheach run of the thermocycler and were blotted, in duplicate, on eachhybridization membrane containing the test samples amplified in thesame thermocycler run. Two negative controls also were included, induplicate, in the standards: uninfected woodchuck serum used forthe dilution of the standards and the water/buffers used as PCRcomponents.11,34

Levels of intrahepatic WHV nucleic acids were quantitatively de-termined by Southern or Northern blot hybridization.41,42 Levels ofWHV surface antigen (WHsAg) and the presence of antibodies toWHsAg and WHV core antigen (WHcAg) in serum samples weredetermined using WHV-specific enzyme immunoassays.43,44 Assess-ments of intrahepatic expression of WHcAg were performed by im-munohistochemical methods using a WHcAg-specific antibody.45,46

Calculations of the apparent half-lives of serum virions and he-patic WHV covalently closed circular (ccc)DNA (see below) andminimal effective doses were based on previously described meth-ods.31 Statistical comparisons of the levels of WHV nucleic acids inserum and liver tissues was performed using Student t test, withcorrections for small sample sizes.

Analysis of Hepatic WHV cccDNA. To directly correlate levels ofWHV cccDNA with the levels of hepatic WHV-DNA replication andWHV RNA in individual liver biopsy samples, the following proce-dure was used. Approximately 1 to 10 mg of tissue obtained fromindividual needle-punch biopsy material was processed. Hirt extrac-tion was performed as previously described.47 Briefly, half the tissuehomogenate was mixed with an equal volume of 4% SDS (final con-centration 2%). Afterward, 2.5 mol/L potassium chloride was addedto a final concentration of 0.5 mol/L. Samples were next subjected tocentrifugation at 16,000g for 30 minutes at 4°C. Supernatant fluidswere transferred to fresh tubes and extracted sequentially one timeeach with phenol, phenol/chloroform, and chloroform. DNA wasprecipitated with 0.3 mol/L sodium acetate (pH 5.2) and 2 volumesethanol, followed by centrifugation. Pellets were resuspended insterile distilled water, and DNA concentrations were determined byspectrophotometry (A260). Total cellular DNA and replicative DNAintermediates of WHV were isolated from the remaining half of thetissue homogenate by adding SDS and proteinase K to final concen-trations of 0.2% and 100 mg/mL, respectively, followed by incubation

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 255

at 60°C overnight. Samples were extracted sequentially one timeeach with phenol, phenol/chloroform, and chloroform. DNA wasprecipitated with 0.3 mol/L sodium acetate (pH 5.2) and 2 volumesethanol, followed by centrifugation. Pellets were resuspended insterile distilled water and DNA concentrations determined by spec-trophotometry (A260). DNA samples were not digested with restric-tion enzymes. Before electrophoresis, DNA samples (5 mg) weremixed with an equal volume of liquified 0.4% agarose in 23 TAE(TAE: 0.04 mol/L Tris-acetate, 1 mmol/L ethylenediaminetetraaceticacid [pH 8.0]) and loaded into the wells of a 0.9% agarose DNA gel(25 3 20 cm). Samples were allowed to solidify in the wells beforesubmersion of the gel into 13 TAE running buffer without ethidiumbromide.

Total RNA was isolated from the remainder of the biopsy samplein a solution of guanidine isothiocyanate and phenol, and sampleswere processed as recommended by the manufacturer (Trizol Re-agent; Life Techonologies).48 After precipitation of RNA, sampleswere dissolved in RNAse-free water (Life Technologies, Inc., Gaith-ersburg, MD), and RNA concentrations were determined by spec-trophotometry (A260). RNA samples were dried under vacuum,dissolved in RNA sample buffer (2.2 mol/L formaldehyde, 50% for-mamide, and glycerol) and heated at 65°C for 5 minutes, followed byquick chill on ice, before loading a 1.4% agarose gel with formalde-hye as denaturant (Gerard and Miller, 1986, BRL Focus, 8:3, LifeTechnologies).

After gel electrophoresis (3 hours at 50 V), the nucleic acids weretransferred to nylon membranes as recommended by the manufac-turer (GeneScreenPlus, NEN) by capillary blot with 103 SSC (SSC:0.3 mol/L sodium chloride, 30 mmol/L sodium citrate [pH 7.0])overnight. Membranes containing DNA were soaked in 0.25 N so-dium hydroxide, neutralized in 0.2 mol/L Tris (pH 8.0) and 13 SSC,baked at 80°C for 2 hours, prehybridized at 60°C for 2 to 3 hours, andhybridized at 60°C overnight in 53 SSC containing 1% SDS and 100mg/mL sheared salmon DNA. Immediately following the overnighttransfer procedure, membranes containing RNA were baked andthen hybridized in 0.5 mol/L sodium phosphate (pH 7.0) containing7% SDS and 100 mg/mL sheared salmon DNA. A WHV-specific ran-dom-primed, [a32P]dCTP-labeled, DNA probe made from the EcoRIfragment of WHV840 was heat-denatured at 100°C for 10 minutesbefore addition to the prehybridization solution. Following over-night incubation, membranes were washed several times in 13 SSCand 1% SDS, air-dried, and exposed to phosphor-screens for imageanalysis (Storm 840/ImageQuant; Molecular Dynamics, Sunnyvale,CA).

RESULTS

L-FMAU Is a Potent and Fast-Acting Inhibitor of WHV Viremia.L-FMAU induced a dose-dependent decline in viremia inchronically infected woodchucks (Figs. 1 and 6, Table 1). Thelowest dose of L-FMAU used in these studies (0.03 mg/kg) wasthe only one that did not reduce the average levels of viremiasignificantly by the end of the treatment period. The averagelevel of viremia in woodchucks treated with 0.1 mg/kgL-FMAU was reduced approximately 12-fold by the end of thetreatment period. There was, however, a high degree of vari-ation in the antiviral response in this group of animals. Nosignificant change in viremia was observed in 2 of the 4 wood-chucks treated with 0.1 mg/kg (F3712, M5375), whilemarked reductions in viremia were observed in the other 2animals in this group (M3703, F5259). The average level ofviremia in woodchucks treated with the highest dose ofL-FMAU used in this study (10 mg/kg) was reduced approx-imately 100,000-fold after 7 days of therapy and more than100 million-fold by the end of the treatment period.

The rate of the L-FMAU–induced reduction of viremia wasremarkably rapid (Figs. 1 and 6, Table 1). There was no effecton viremia during the first 12 hours following the initial ad-

ministration of drug at any of the doses used. After 72 hours oftherapy at a dose of 10 mg/kg, viremia was reduced an averageof 10,000-fold. Although the initial rate of clearance of vire-mia in the group treated with 3 mg/kg L-FMAU was slowerthan that observed for the group treated with 10 mg/kgL-FMAU, the levels of viremia in animals treated with either 3or 10 mg/kg were virtually the same from 5 to 28 days oftreatment.

Viremia began to rebound following the end of treatment atintervals that were generally dose-related (Figs. 1 and 6, Table1). Viremia in all animals treated with 0.3 mg/kg L-FMAU rosesignificantly within 1 week following the termination of ther-apy and had returned to pretreatment levels in all animals by2 weeks’ posttreatment. Viremia in the animals treated withhigher doses of L-FMAU remained suppressed for variouslengths of time following the withdrawal of treatment. In gen-eral, viremia in animals treated with either 1.0 or 3.0 mg/kgL-FMAU did not return to pretreatment levels until 8 to 10weeks after the end of therapy. Viremia in 2 animals treatedwith 1.0 mg/kg (M5202, F5280) remained suppressed for atleast 8 weeks following the end of therapy. Viremia in theselatter 2 woodchucks was transiently reduced to undetectablelevels (,30 WHV virions/mL serum) during the treatmentperiod. Viremia in 3 of the 4 animals treated with 3.0 mg/kgrose slowly during the period of 2 to 8 weeks following theend of therapy and was still suppressed in 2 animals (M5160,M5191) at 8 weeks’ posttreatment. Viremia in 1 animal treatedwith 3.0 mg/kg (M5191) was reduced to undetectable levelsafter 2 weeks of treatment and remained undetectable until 6weeks’ posttreatment. Viremia in all of the animals treatedwith 10 mg/kg remained suppressed at the same levels (orlower) than that observed at the end of the treatment periodfor 6 weeks following the end of therapy. Viremia in 2 of theanimals treated with 10 mg/kg (F5242, F5279) began to re-bound to pretreatment levels over the next 6 weeks. Viremiain 2 of the 4 animals treated with 10 mg/kg L-FMAU (M3702,F5273) did not return to pretreatment levels by the end of thestudy period. Viremia in M3702 was reduced to undetectablelevels after 2 weeks of treatment, remained at undetectablelevels until 8 weeks’ posttreatment, then fluctuated betweenapproximately 1 3 106 and 1 3 108 virions per milliliter ofserum until the death of this animal as a result of complica-tions related to HCC 29 weeks’ posttreatment. F5273 alsodied of complications related to HCC 13 weeks’ posttreat-ment. Both deaths were considered to be unrelated to treat-ment. As observed for HBV chronic carriers, the developmentof HCC is a natural consequence of chronic WHV infection inwoodchucks.16,17 HCC develops in 100% of WHV carrier an-imals within 4 years of infection, with a median developmentinterval of approximately 29 months following neonatal in-fection.16,17

No significant changes in the average levels of viremia wereobserved in the placebo-treated (control) groups of animalsduring the study period (Figs. 1 and 6, Table 1).

In these studies, the decline in viremia began between 12and 24 hours following the initiation of therapy. In animalstreated with 10 mg/kg, viremia was reduced approximately10-fold during this period. Based on the kinetics of the reduc-tion in viremia in animals treated with 10 mg/kg L-FMAU 12to 72 hours after the initiation of treatment, we estimate theminimal apparent half-life of WHV in serum to be 4.0 hours(Table 2). The rate of the reduction in viremia in animals

256 PEEK ET AL. HEPATOLOGY January 2001

treated with 10 mg/kg slowed dramatically after 72 hours fol-lowing the start of treatment (Figs. 1 and 6). The reasons forthis decline are unclear at the present time. Based on thekinetics of the decline in viremia during the initial 3, 5, or 7days of therapy, the minimal effective dose of L-FMAU inchronically infected woodchucks was calculated to be 0.04 to0.05 mg/kg (Fig. 2). Different time intervals were used for thiscalculation to determine if the dose- and time-dependentchanges in the rates of the reduction in viremia over the first 7days of therapy would appreciably affect estimations of a min-imal effective dose. Although the slopes of the linear regres-sions for the 3 different time intervals used appear to differ,the regression coefficients and estimations of the minimal ef-fective dose were virtually identical for all 3 time intervals(Fig. 2).

L-FMAU Therapy Reduces Intrahepatic WHV-DNA Replicationand WHV-RNA Levels. Hepadnaviral DNA replication interme-diates in liver tissue are comprised of a heterogeneous popu-lation of single-stranded and partially double-stranded viralDNA molecules that migrate as a distinctive smear with anapparent molecular size of 0.2 kb to 3.0 kb in Southern blothybridization analyses.49,50 Analysis of liver biopsy samplesdemonstrated changes in the levels of intrahepatic WHV rep-lication intermediates (RI) that were generally consistent withthe viremia patterns (Figs. 3 and 6, Table 1). Significant re-ductions in the average levels of WHV RI were observed by theend of the treatment period in animals treated with 0.3 mg/kgor higher amounts of L-FMAU. The average levels of WHV RIin animals treated with 0.3 mg/kg, 1.0 mg/kg, and 3.0 mg/kgwere reduced approximately 3-fold, 10-fold, and 28-fold, re-

FIG. 1. Effect of 7 and 28 days ofL-FMAU therapy on WHV viremia inchronic WHV carrier woodchucks.Values for individual animals in eachtreatment group are displayed. Seetext and Materials and Methods sec-tion for experimental details. Hori-zontal bars denote treatment period.“WHVge,” WHV genomic equiva-lents (virion or WHV-DNA–con-taining virus particles); ud, unde-tectable level of serum WHV DNA(,30 WHVge/mL serum).

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 257

TA

BL

E1.

Sum

mar

yof

An

tivi

ral

Act

ivit

y

Gro

up

Tim

eP

oin

t

Ser

um

WH

VV

irio

nD

NA

(log

10

WH

Vge/

mL

seru

m)

Hep

atic

WH

VR

I(p

g/m

gce

llD

NA

)H

epat

icW

HV

RN

A(p

g/m

gto

tal

cell

RN

A)

Ser

um

WH

sAg

(%p

retr

eatm

ent

val

ue)

Hep

atic

WH

cAg

(%W

HcA

g1

cell

s)

Mea

n6

SD

(Ran

ge)

Mea

n6

SD

(Ran

ge)

Mea

n6

SD

(Ran

ge)

Mea

n6

SD

(Ran

ge)

Mea

n6

SD

(Ran

ge)

Pla

cebo

Con

trol

s(N

58)

pret

reat

men

t(w

k0)

10.6

60.

9(9

.1-1

1.7)

1,25

76

415

(950

-2,1

00)

516

11(3

9-67

)10

061

610

(50-

75)

end

oftr

eatm

ent

(wk

4)10

.76

0.8

(9.1

-11.

7)1,

381

641

8(9

70-2

,200

)53

610

(41-

66)

112

651

(77-

217)

616

14(5

0-75

)4

wks

post

trea

t.(w

k8)

10.7

60.

8(9

.0-1

1.8)

ND

ND

896

54(5

3-21

6)N

D12

wks

post

trea

t.(w

k16

)10

.66

1.0

(8.5

-11.

4)1,

268

635

9(9

70-1

,900

)60

612

(43-

75)

966

58(4

2-21

3)71

68

(60-

75)

0.03

mg/

kgL-F

MA

Upr

etre

atm

ent

(wk

0)10

.36

0.4

(9.1

-10.

5)1,

432

643

5(9

50-1

,900

)56

613

(40-

69)

100

end

oftr

eatm

ent

(wk

4)9.

86

0.7

(9.0

-10.

5)N

DN

D10

46

25(6

7-18

0)N

D4

wks

post

trea

t.(w

k8)

10.1

60.

6(9

.5-1

0.6)

1,56

46

469

(990

-2,1

00)

506

5(4

3-52

)61

633

(58-

91)

ND

12w

kspo

sttr

eat.

(wk

16)

10.4

60.

4(9

.8-1

0.8)

1,77

36

116

(1,7

00-1

,900

)58

67

(50-

66)

606

28(1

2-12

5)N

D0.

1m

g/kg

L-F

MA

Upr

etre

atm

ent

(wk

0)10

.66

0.3

(10.

3-10

.9)

1,11

86

212

(930

-1,4

00)

576

9(5

0-71

)10

0N

Den

dof

trea

tmen

t(w

k4)

9.5

60.

6*†

(8.9

-10.

0)1,

489

649

3(1

,000

-2,1

00)

676

7(5

9-72

)90

623

(67-

121)

ND

4w

kspo

sttr

eat.

(wk

8)10

.16

1.0

(8.7

-10.

8)N

DN

D35

628

(4-6

8)N

D12

wks

post

trea

t.(w

k16

)10

.16

1.1

(8.5

-11.

0)1,

354

627

5(1

,100

-1,7

00)

626

10(5

3-76

)51

630

(1-9

4)N

D0.

3m

g/kg

L-F

MA

Upr

etre

atm

ent

(wk

0)10

.86

0.3

(10.

6-11

.1)

1,48

96

574

(900

-2,1

00)

476

20(3

8-79

)10

061

614

(50-

75)

end

oftr

eatm

ent

(wk

4)6.

76

1.5‡

¶(5

.0-8

.1)

525

661

*§(4

60-6

00)

486

16(3

1-55

)42

629

(15-

133)

436

29(2

5-75

)4

wks

post

trea

t.(w

k8)

10.8

60.

3(1

0.6-

11.2

)N

DN

D31

628

(8-1

14)

ND

12w

kspo

sttr

eat.

(wk

16)

10.8

60.

5(1

0.3-

11.4

)1,

351

617

6(1

,100

-1,8

00)

506

12(3

5-55

)69

634

(35-

183)

686

13(5

0-75

)1.

0m

g/kg

L-F

MA

Upr

etre

atm

ent

(wk

0)11

.06

0.2

(10.

7-11

.2)

1,27

16

432

(980

-1,8

00)

516

20(3

6-77

)10

064

612

(50-

75)

end

oftr

eatm

ent

(wk

4)2.

86

1.7‡

¶(u

d-4.

9)12

16

22†\

(110

-150

)46

65

(40-

50)

416

28(7

-143

)44

621

(25-

75)

4w

kspo

sttr

eat.

(wk

8)4.

36

4.0‡

¶(u

d-9.

5)N

DN

D9

622

*(1

-98)

ND

12w

kspo

sttr

eat.

(wk

16)

11.1

60.

2(1

0.7-

11.2

)1,

350

631

0(1

,100

-1,8

00)

446

11(3

1-52

)70

635

(47-

190)

636

13(5

0-75

)3.

0m

g/kg

L-F

MA

Upr

etre

atm

ent

(wk

0)11

.26

0.3

(10.

9-11

.6)

1,45

36

601

(970

-2,2

00)

576

20(4

0-80

)10

055

613

(50-

75)

end

oftr

eatm

ent

(wk

4)2.

06

1.8‡

¶(u

d-4.

6)52

613

‡¶(4

2-69

)49

68

(35-

69)

516

14(2

5-92

)3

629

(0-7

5)4

wks

post

trea

t.(w

k8)

3.9

63.

3‡¶

(ud-

8.8)

ND

ND

56

12*

(0.1

-53)

ND

12w

kspo

sttr

eat.

(wk

16)

10.9

60.

3(1

0.7-

11.3

)1,

258

669

2(7

80-2

,300

)38

65

(33-

42)

556

18(2

7-10

8)73

620

(50-

100)

10m

g/kg

L-F

MA

Upr

etre

atm

ent

(wk

0)11

.16

0.5

(10.

5-11

.4)

1,43

16

689

(970

-2,4

00)

646

10(5

1-75

)10

052

613

(25-

60)

end

oftr

eatm

ent

(wk

4)2.

96

1.5‡

¶(u

d-4.

2)21

68‡

¶(1

8-30

)24

615

†¶(0

-30)

256

13*

(7-6

0)8

628

*§(0

-70)

4w

kspo

sttr

eat.

(wk

8)2.

06

1.7‡

¶(u

d-4.

2)N

DN

D5

62†

(4-9

)N

D12

wks

post

trea

t.(w

k16

)7.

96

3.0‡

¶(4

.2-1

1.1)

516

669

2*§

(110

-1,5

00)

506

28(3

0-80

)17

634

*(3

-150

)9

631

*§(0

-75)

NO

TE

.A

vera

ge(6

SD)

and

over

all

ran

geof

valu

esfo

rW

HV

nu

clei

cac

ids

are

pres

ente

d.L

evel

sof

intr

ahep

atic

intr

acel

lula

rW

HV

-DN

Are

plic

atio

nin

term

edia

tes

(WH

VR

.I.)

and

tota

lW

HV

RN

Aw

ere

nor

mal

ized

for

tota

lce

llu

lar

DN

Aor

RN

Au

sin

gth

ere

lati

vele

vels

ofb

-act

inD

NA

orR

NA

(b-a

ctin

hyb

ridi

zati

onpr

obe

purc

has

edfr

omO

nco

r,In

c.,

Gai

ther

sbu

rg,

MD

).B

ecau

seof

the

subs

tan

tial

degr

eeof

vari

atio

nin

WH

sAg

leve

lsin

the

plac

ebo

con

trol

grou

psdu

rin

gth

est

udy

peri

od,s

tati

stic

alco

mpa

riso

ns

for

WH

sAg

wer

em

ade

only

toth

epl

aceb

oco

ntr

olgr

oups

.A

bbre

viat

ion

s:W

HV

ge,W

HV

gen

ome

equ

ival

ents

;ND

,no

live

rbi

opsi

esw

ere

obta

ined

4w

eeks

’pos

ttre

atm

ent;

ud,

un

dete

ctab

lele

velo

fW

HV

vire

mia

(,1.

48lo

g 10

WH

Vge

/mL

seru

m).

Stat

isti

call

ydi

ffer

ent

from

plac

ebo

con

trol

(wee

k4)

:*P

,.0

5;†

P,

.01;

‡P

,.0

01.S

tati

stic

ally

diff

eren

tfr

ompr

etre

atm

ent

valu

es:§

P,

.05;

\P

,.0

1;¶

P,

.001

.

258 PEEK ET AL. HEPATOLOGY January 2001

spectively, by the end of treatment. By the end of the 12-weekposttreatment period, the average levels of WHV RI had re-turned to pretreatment values in animals treated with 3.0mg/kg or lower doses of L-FMAU. The level of WHV RI in 1animal treated with 3.0 mg/kg (F5196) was still approxi-mately 3-fold below the pretreatment value for that animal 12weeks’ posttreatment.

The average level of WHV RI in animals treated with 10mg/kg L-FMAU was reduced nearly 70-fold by the end of thetreatment period and were still reduced approximately 3-foldbelow average pretreatment levels 12 weeks’ posttreatment(Figs. 3 and 6, Table 1). However, at 12 weeks’ posttreatment,there was substantial heterogeneity in the levels of WHV RI inthe liver tissues of these 4 animals that correlated with theindividual patterns of viremia. WHV RI levels were at pre-treatment values in the 2 animals in which viremia had re-turned to pretreatment levels by 12 weeks’ posttreatment(F5242, F5279). WHV RI remained at reduced levels in the 2animals in which viremia remained suppressed at 12 weeks’posttreatment (M3702, F5273).

The average levels of intrahepatic WHV RNA in chroniccarrier woodchucks treated with 10 mg/kg L-FMAU were re-duced approximately 3-fold from pretreatment values at theend of the treatment period (Table 1). The average levels ofWHV RNA in these animals had returned to pretreatmentlevels by 12 weeks’ posttreatment. The average levels of WHV-RNA levels were still reduced approximately 2- to 3-fold in the2 animals in which viremia and WHV RI remained reduced at12 weeks’ posttreatment (M3702, F5273). WHV-RNA levelsin the liver samples obtained from the groups of animalstreated with 3.0 mg/kg or lower doses of L-FMAU did notchange significantly during the study period (Table 1).

No significant changes in the average levels of intrahepaticWHV RI or WHV RNA were observed in the placebo-treated(control) groups of animals during the study period (Figs. 3and 6, Table 1).

L-FMAU Therapy Reduces the Levels of WHsAg and WHcAg.L-FMAU therapy lowered levels of WHsAg in the serum of thetreated animals in a dose-dependent manner (Figs. 4 and 6,Tables 1 and 2). There was a wide range of WHsAg levels inthe pretreatment serum of individual animals in this study(approximately 70-2,000 mg/mL serum). Because of the vari-

ation in the initial amounts of WHsAg in the serum of indi-vidual animals, the relative levels of WHsAg present at thedifferent time points during this study are presented as a per-centage of the pretreatment value for each animal.

No significant reduction in the average levels of WHsAg inthe serum of the groups of animals treated with 0.03 mg/kg,0.1 mg/kg, or 0.3 mg/kg L-FMAU was observed (Figs. 4 and 6,Tables 1 and 2). However, some changes in WHsAg wereobserved in individual animals treated with these doses ofL-FMAU. WHsAg levels in 1 animal treated with 0.03 mg/kg(M3707) were transiently reduced approximately 4-fold at 4weeks’ posttreatment, WHsAg in 1 animal treated with 0.1mg/kg (M3703) was reduced over 20-fold from 4 to 12 weeks’posttreatment, and WHsAg levels were reduced approxi-mately 5- to 10-fold by the end of the treatment period in 2animals treated with 0.3 mg/kg L-FMAU (M5175, F5247) thatpersisted for at least 4 weeks’ posttreatment in M5175 and 8weeks’ posttreatment in F5247 (Fig. 4).

The average levels of WHsAg in animals treated with 1.0mg/kg were reduced approximately 10-fold from pretreat-ment levels at 4 weeks’ posttreatment, and had returned topretreatment levels by 12 weeks’ posttreatment (Figs. 4 and 6,Table 2). Only 1 animal treated with 1.0 mg/kg (M5202)showed a marked reduction in WHsAg at the end of the treat-ment period. While substantial reductions in WHsAg levelswere observed in 3 of the 4 animals treated with 1.0 mg/kg,WHsAg levels in 1 animal (M5140) were essentially unaf-fected by L-FMAU therapy.

The average level of WHsAg in the group of animals treatedwith 3.0 mg/kg was reduced approximately 20-fold by 4weeks’ posttreatment and had returned to pretreatment levelsby 12 weeks’ posttreatment (Figs. 4 and 6, Table 2). WHsAglevels in 1 animal (F5196) displayed little change during thestudy period. WHsAg levels in another woodchuck (F5238)were only transiently reduced at 4 weeks’ posttreatment. Lev-els of WHsAg in the other 2 animals (M5160, M5191) showedprogressive reductions in WHsAg that were sustained for atleast 8 weeks following the end of therapy.

The average levels of WHsAg in the group of animalstreated with 10 mg/kg L-FMAU were reduced approximately

FIG. 2. Effect of the dose of L-FMAU on the apparent half-life of serumvirions in treated chronic carrier woodchucks. The apparent half-lives ofserum virions were calculated based on previously reported methods.31,53,54

The reciprocal of the half-lives for doses of 0.1 to 10 mg/kg for the initial 3, 5,or 7 days of therapy are displayed. Lines represent regression analyses for eachtime interval. The regression coefficient (r) for each data set is listed in thefigure.

TABLE 2. Effect of L-FMAU Dosage and Duration of Therapyon the Apparent Half-Life of WHV in the Serum of

Chronically Infected Woodchucks

Dose (mg/kg)

Apparent Half-Life of Serum Virions (hours)

3 Days 5 Days 7 Days

0.03 71 82 990.1 44 47 730.3 9.0 13 181.0 7.0 10 123.0 6.0 6.8 8.8

10.0 4.0 6.4 8.1

NOTE. The apparent half-life (in hours) of WHV in the serum of L-FMAU–treated, chronically infected woodchucks following the initial 3, 5, or 7 daysof L-FMAU therapy are displayed. The apparent half-life of serum virionswere calculated31,53,54 based on the mean reduction in viremia during theinterval between 12 hours after the initiation of therapy and either 3, 5, or 7days of treatment in these studies. There was no reduction in viremia duringthe first 12 hours following the initiation of therapy (Fig. 1).

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 259

4-fold at the end of treatment, 20-fold at 4 weeks’ posttherapy,and were still reduced more than 5-fold at 12 weeks followingthe end of treatment (Figs. 4 and 6, Table 2). In the 2 animalsthat showed the most rapid rebound of viremia following thetermination of therapy (F5242, F5279), WHsAg levels wereonly modestly reduced by the end of the treatment period,declined further by 4 weeks’ posttreatment, but began to re-bound by 8 weeks’ posttreatment. WHsAg levels in the 2 an-imals in which viremia remained reduced following the end oftreatment (M3702, F5273) were markedly reduced at the endof the treatment period and remained suppressed throughoutthe posttreatment study period.

Reductions in the intrahepatic expression of WHcAg wereobserved following 4 weeks of treatment with the 2 highestdoses of L-FMAU (3.0 and 10 mg/kg)(Figs. 5 and 6, Table 2).Treatment with these 2 doses of L-FMAU reduced the numberof WHcAg-positive cells to undetectable levels in at least onehalf of each group of treated animals by the end of the treat-ment period. WHcAg remained at undetectable levels at 12weeks’ posttreatment in 2 of 4 animals (M3702, F5273)treated with 10 mg/kg. WHcAg expression in the livers of all 4animals treated with 3.0 mg/kg had returned to pretreatmentlevels by 12 weeks following the end of therapy. Doses of 0.3and 1.0 mg/kg L-FMAU had less of an effect. Treatment with

these doses reduced the number of WHcAg-positive hepato-cytes in the livers of 1 or 2 animals in each group, but gener-ally had no effect on the relative intensity of WHcAg stainingin the positive cells (Figs. 5 and 6, Table 2).

No significant changes in the average levels of antigenemiaor intrahepatic WHcAg expression were observed in the pla-cebo-treated (control) groups of animals during the studyperiod (Figs. 1, 4-6, Table 1). WHsAg serum levels in 2 wood-chucks in the control group (F5197, F5223) were transientlyreduced approximately 3- to 4-fold at the 8-week posttreat-ment time point. No correlation between the relative changesin WHsAg levels in response to L-FMAU therapy and the ini-tial (pretreatment) levels of WHsAg in individual animalswere observed. None of the control animals or any of thewoodchucks treated with L-FMAU in these studies had evi-dence of antibodies to WHsAg (data not shown).

L-FMAU Treatment Leads to Reductions of Hepatic WHV cccDNA.Hepadnaviral cccDNA (superhelical cccDNA) resides in thenucleus and is the template for RNA transcription, includingviral pregenomic RNA that is used as the template for thesynthesis of the first stand of viral DNA.49,50 The reductions inthe levels of serum WHsAg, hepatic WHcAg, and hepaticWHV RNA in the animals treated with 10 mg/kg L-FMAU(Figs. 1, 3-6, Table 2) was consistent with a reduction in the

FIG. 3. Effect of L-FMAU ther-apy on hepatic WHV replication inchronic carrier woodchucks. Seetext and Materials and Methods forexperimental details. Levels of he-patic cellular DNA were quantifiedby hybridization to a commercialb-actin gene probe (Oncor, Inc.,Gaithersburg, MD) using Southernor dot-blot hybridization techniquesas described in Materials and Meth-ods.

260 PEEK ET AL. HEPATOLOGY January 2001

levels of this hepatic WHV genomic form. Sufficient liver bi-opsy material remained from the week-0 and week-4 liverbiopsies obtained from the 4 woodchucks treated with 10mg/kg to permit a preliminary quantitative analysis to deter-mine if L-FMAU therapy induced a reduction in the levels ofcccDNA in the livers of these treated animals.

WHV nucleic acid analysis on liver samples obtained beforedrug treatment (week 0) showed typical profiles for WHV-infected hepatocytes: partially double-stranded DNA/RNAreplicative intermediates (1-3 kb), cccDNA (2.2 kb), and pre-genomic (3.6 kb) and WHsAg (2.7 kb) mRNA transcripts(Fig. 7). Pretreatment levels of WHV cccDNA were calculatedat an average of 20 copies per cell (range, 18-35 copies/cell) inthese 4 woodchucks. Identical nucleic acid analysis on liversamples taken at the end of L-FMAU therapy with 10 mg/kg(week 4) yielded an entirely different WHV profile (Fig. 7).The level of replicative intermediates was dramatically re-duced in all 4 animals, and WHV RNA was reduced to nearlyundetectable levels in 2 animals (M3702, F5273). WHVcccDNA was observed in only 2 of these 4 animals (3.0 copies/cell in F5242, 11 copies/cell in F5279), representing 6-foldand 2-fold reductions from the pretreatment levels in these 2animals, respectively. In the other 2 animals (M3702, F5273),cccDNA was reduced to undetectable levels (cutoff of approx-imately 1.0 copies/cell), representing more than a 20-fold re-duction from pretreatment values. Following 4 weeks of ther-apy, the average WHV cccDNA levels for the entire group ofanimals were calculated to be 2.0 copies per cell (range, ,1.0to 11 copies/cell). The average 10-fold decrease in cccDNA

levels over a 4-week period in these woodchucks was consis-tent with an average apparent half-life of approximately 8.5days (range, ,6.0-14 days).

Four Weeks of Therapy With L-FMAU Was Well Tolerated. Noobvious treatment-related clinical, hematologic, or clinicalbiochemical abnormalities indicative of toxicity, includinghepatic toxicity, were observed in any of the placebo-treatedor L-FMAU–treated animals in this study (data not shown).No evidence of lactic acidosis was observed in any of thetreated animals (data not shown). No fatty liver or other sig-nificant changes in liver histology that suggested drug-relatedtoxicity were observed in any of the L-FMAU–treated animalsduring the 16-week study period (data not shown).

DISCUSSION

In this report, we examined the relative antiviral activity ofthe novel nucleoside, L-FMAU, against WHV in chronicallyinfected woodchucks. The data establish that there was anabsence of obvious treatment-related toxicity over a 4-weekperiod of daily oral administration, and that L-FMAU was anexceptionally effective inhibitor of WHV replication in vivo.Dramatic reductions in all measured serologic and intracellu-lar markers of viral replication and viral gene expression wereobserved at the highest doses of L-FMAU used in this study.The relative antiviral activity of L-FMAU in this study wasgreater than that previously reported for other antiviral drugsadministered to WHV chronic carriers.11,15,18-36 The sustainedantiviral effects on WHV replication observed following with-

FIG. 4. Effect of L-FMAU therapyon WHV antigenemia in chronicWHV carrier woodchucks. Valuesfor individual animals in each treat-ment group are displayed. See textand Materials and Methods for ex-perimental details. Horizontal barsdenote treatment period. The levelsof WHsAg for each animal are ex-pressed as a percentage of the level ofWHsAg present in its pretreatmentserum.

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 261

drawal of L-FMAU, along with similar observations in WHVcarrier woodchucks with another nucleoside (BMS200475)30

and in some patients treated with lamivudine,51 raises thepossibility that life-long therapy may not be necessary to con-trol chronic HBV infection.

Four weeks of L-FMAU therapy was well tolerated bychronically infected woodchucks. No evidence of lactic aci-dosis, characteristic of the patterns of toxicity induced by theD-isomers of this family of nucleosides (filauridine, D-FEAU,D-FMAU),11-15 was observed in any of the treated animals inthis study. Periods of treatment longer than the 4-week dura-tion of therapy used in this study were typically necessary todemonstrate the toxicologic profiles of these other nucleo-sides.11-15 However, it should be noted that in previous stud-ies, no evidence of toxicity was observed in non–WHV-in-fected woodchucks following 12 weeks of therapy with 2

mg/kg L-FMAU, while severe toxicity that led to death wasobserved in all woodchucks treated for 12 weeks with 2mg/kgD-FMAU.12 Longer durations of therapy with L-FMAU will beneeded to confirm the safety profile of this potential antiviralagent.

L-FMAU inhibited WHV replication in WHV carrier wood-chucks in a dose-dependent manner. Based on the kinetics ofreduction in viremia and intrahepatic WHV replication, thehighest dose used in these studies, 10 mg/kg body weight,appeared to be near the maximally effective dose for treatmentof WHV chronic carriers. Doses of 3.0 mg/kg and 10 mg/kgappeared to be similar in effectiveness with respect to maxi-mal suppressions of viremia and intrahepatic WHV replica-tion by the end of the treatment period. However, treatmentwith 10 mg/kg conferred additional antiviral benefit in theform of a more sustained reduction in WHV replication, se-

FIG. 5. Effect of L-FMAU therapyon intrahepatic WHcAg expressionin chronic WHV carrier wood-chucks. Values for individual ani-mals in each treatment group are dis-played. See text and Materials andMethods for experimental details.Left-hand panels display the percent-age of hepatocytes in individual ani-mals staining positive for WHcAg;right-hand panels display the averagerelative intensity of staining in WH-cAg-positive hepatocytes (nonstain-ing cells are not included in the de-termination of staining intensity).

262 PEEK ET AL. HEPATOLOGY January 2001

rum WHsAg, and intrahepatic WHcAg expression followingdrug withdrawal. These results suggest the possible benefit ofreducing hepadnaviral replication as rapidly as possible.Achieving similar reductions in viral load over a longer treat-ment period may not induce the same prolonged antiviraleffect observed when viral load is reduced more rapidly.

The timing and the dose-related nature of the decline inWHsAg levels, intrahepatic gene expression, and the delay inthe rebound of markers of viral replication at the higher dosesof L-FMAU are consistent with a mechanism of an L-FMAU–induced reduction of WHV polymerase activity below thelevel necessary to maintain steady-state levels of WHVcccDNA in the livers of the treated animals. A preliminaryanalysis of cccDNA in animals treated with 10 mg/kg L-FMAUin this study demonstrated an average reduction of cccDNAlevels of at least 10-fold after 4 weeks of therapy with 10 mg/kg

L-FMAU. Importantly, the relative levels of the markers ofviral replication and viral gene expression were directly cor-related with the levels of WHV cccDNA in the individualanimals treated with 10 mg/kg L-FMAU, providing support forthis mechanism of action. WHV-RNA transcripts were onlydetectable in the 2 animals (F5242, F5279) in which cccDNAcould be seen (Fig. 7). WHV RI levels (Fig. 3), as well as thelevels of serum WHsAg (Fig. 4) and intrahepatic WHcAg (Fig.5), in these 2 animals were also the highest among the group.Finally, among the 4 woodchucks treated with 10 mg/kgL-FMAU, viremia rebounded most rapidly following the termina-tion of therapy in these 2 animals (Fig. 1). Because L-FMAU isa known inhibitor of DNA polymerase,2,3,6 it is reasonable toassume that the drug would not directly interfere with theproduction of viral proteins, although this has not been for-mally tested.

FIG. 6. Summary of the effect ofL-FMAU therapy on WHV replica-tion in chronic WHV carrier wood-chucks. Geometric mean values foreach experimental group are dis-played. See text and Materials andMethods for experimental details.Horizontal bars denote treatment pe-riod. Vertical lines denote standarddeviations. “ WHVge, ” WHV ge-nomic equivalents (virion or WHV-DNA–containing virus particles).The levels of WHsAg for each animalare expressed as a percentage of thelevel of WHsAg present in its pre-treatment serum. ud, undetectablelevel of serum WHV DNA (30WHVge/mL serum).

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 263

The overall loss of cccDNA could be a consequence of re-ductions of cccDNA in individual infected cells and/or resultfrom the replacement of infected cells at a rate that is fasterthan the rate of infection of these new cells by the low, resid-ual level of virions in the serum. While it is not possible todetermine the primary mechanism by which cccDNA levelswere reduced in this study, it is of interest to note that theestimated apparent half-lives of cccDNA in the 2 animals inwhich cccDNA declined most rapidly in this study are in the

same range as previously reported estimates for the half-life ofWHV-infected hepatocytes (approximately 4.5 days).52

In this study, the relative levels of cccDNA in individualanimals at the end of 4 weeks of therapy with 10 mg/kgL-FMAU are inversely related to the duration of the suppres-sion of WHV replication following drug withdrawal. The de-lay in the rebound of viral replication markers following thetermination of therapy in these studies may be a reflection ofthe time required to re-establish cccDNA pools through rein-fection of new hepatocytes and/or reamplification of existingintracellular pools of cccDNA in infected cells. The prelimi-nary assessment of cccDNA in this study illustrates that rela-tively modest reductions (e.g., 50%-85%) in hepadnaviralcccDNA levels can potentially induce substantial retention(e.g., 4-6 weeks) of antiviral effects following the terminationof therapy. It is equally important to note that a reduction ofover 95% in hepatic cccDNA is insufficient to induce a per-manent suppression of viral replication following the with-drawal of therapy. This indicates the necessity of reducing thelevels of this critical viral template to even lower levels and forperhaps longer periods of time.

The minimally effective dose of L-FMAU in this animalmodel was mathematically calculated at approximately 0.04mg/kg based on a measurable effect on viremia (Fig. 2). Thisvalue is consistent with the reductions in viremia observed atthe end of the treatment period, which demonstrated a slight,but statistically insignificant, reduction in viremia followingtreatment with 0.03 mg/kg, and a significant antiviral effectwith therapy at 0.1 mg/kg L-FMAU (Table 1).

The rate of reduction and the minimal apparent half-life ofserum virions following treatment with 10 mg/kg/d L-FMAU(4 hours) is considerably less than that previously reported forlamivudine treatment of woodchucks with 15 mg/kg/d (48hours)31 or humans at 100 mg/d (22 hours),53 or for the treat-ment of humans with 30 mg/d adefovir dipovoxil (26hours).54 The changes in the apparent half-life of serum viri-ons in woodchucks at the various doses of L-FMAU and fordifferent time intervals indicate multiple-order kinetics forthe clearance of viremia, which can most likely be groupedinto 2 general categories: virus production and the removal ofvirus. Assuming that L-FMAU does not, at least initially, affectthe rate of the removal of WHV from serum, then the differ-ences in the apparent half-life of serum virions observed at thevarious doses of L-FMAU are most likely a reflection of therelative degree of the inhibition of WHV replication at eachdose.

The biphasic rates of the reduction of viremia induced byL-FMAU therapy are similar to patterns of the loss of HBVfrom the serum of HBV-infected patients treated with lamivu-dine or adefovir dipovoxil,53,54 although the overall loss ofviremia in L-FMAU–treated woodchucks is considerably morerapid. It was proposed in these studies that the initial (faster)rate of viral loss reflects direct interference with virus replica-tion/production, and the subsequent (slower) rate of declinereflects a loss of virus-infected cells from the liver. The obser-vations compiled in L-FMAU–treated woodchucks in the cur-rent study, including reductions in cccDNA and in hepato-cytes expressing viral antigens, are consistent overall with thishypothesis. However, further studies will be necessary tomore precisely define the mechanisms related to the clearancepatterns of virus as a result of drug therapy.

FIG. 7. WHV cccDNA and nucleic acid analysis for woodchucks treatedwith 10 mg/kg L-FMAU. See text and Materials and Methods for experimentaldetails. (Top) WHV RI. (Middle) WHV cccDNA. (Bottom) WHV RNA (lineson the left side of the panel denote the major WHV transcripts49,50).

264 PEEK ET AL. HEPATOLOGY January 2001

The high degree of in vivo antiviral potency of L-FMAU maybe the result of multiple pharmacologic and metabolic factors,as well as the proposed mechanism of action of L-FMAU onviral DNA polymerases. Studies with EBV show that L-FMAUinhibits viral DNA polymerase in a noncompetitive manner,inhibiting viral polymerase elongation reactions and 39-to-59exonuclease activity without incorporation of L-FMAU–triphosphate (L-FMAU-TP) into either EBV or cellular DNA.6However, L-FMAU does appear to compete preferentially withdeoxythymidine for the EBV polymerase.6 L-FMAU is effi-ciently and rapidly transported into cells by equilibrative-sen-sitive and equilibrative-insensitive nucleoside transport, aswell as through nonfacilitated, passive diffusion.9,39 L-FMAUis a substrate for both thymidine and deoxycytidine kinases,and it is phosporylated in a stepwise manner from mono-, todi-, to triphosphate forms rapidly and efficiently.39 L-FMAU-TP isthe overwhelmingly predominant metabolite following cellu-lar uptake.39 Previous pharmacokinetic studies demonstratedthat plasma levels of L-FMAU above the 50% effective doseobserved in cell culture are maintained for more than 24hours following a single oral dose of 25 mg/kg in wood-chucks.12 These findings indicate that L-FMAU-TP most likelyrapidly accumulates and is maintained in hepatocytes at levelsthat are above the inhibitory threshold concentration forWHV polymerase. Unfortunately, no information is currentlyavailable on estimates of intracellular concentrations, thehalf-life of L-FMAU-TP in woodchuck tissues, or the inhibi-tory equilibrium constant values of L-FMAU for hepadnaviralpolymerases. Information on these parameters will enablebetter modeling of the interference with hepadnaviral replica-tion by L-FMAU.

L-FMAU is the most potent single antihepadnaviral agentstudied thus far in the woodchuck model of chronic HBVinfection. The longevity of antiviral effects following rela-tively short periods of once-daily oral administration makeL-FMAU a highly attractive and unique antihepadnaviralagent and one that should be considered as a candidate fortherapy against chronic HBV infection in humans.

Acknowledgment: The technical assistance of Kristine Far-rar, Tracie Franklin, and Karen Gay (Georgetown University),Mary Ascenzi, Virginia Bayer, Christine Bellezza, CurtisFullmer, and Lou Ann Graham (Cornell University) is grate-fully acknowledged.

REFERENCES

1. Aguesse-Germon SS, Liu H, Chevallier M, Pichoud C, Jamard C, Borel C,Chu CK, et al. Inhibitory effect of 29-fluoro-5-methyl-beta-L-arabino-furanosyl-uracil on duck hepatitis B virus replication. Antimicrob AgentsChemother 1998;42:369-376.

2. Balakrishna Pai SS, Liu H, Zhu YL, Chu CK, Cheng YC. Inhibition of hepatitisB virus by a novel L-nucleoside, 29-fluoro-5- methyl-beta-L-arabinofuranosyluracil. Antimicrob Agents Chemother 1996;40:380-386.

3. Chu CK, Ma T, Shanmuganathan K, Wang C, Xiang Y, Pai SB, Yao GQ, etal. Use of 29-fluoro-5-methyl-beta-L-arabinofuranosyluracil as a novelantiviral agent for hepatitis B virus and Epstein-Barr virus. AntimicrobAgents Chemother 1995;39:979-981.

4. Fu L, Liu SH, Cheng YC. Sensitivity of L-(2)2,3-dideoxythiacytidineresistant hepatitis B virus to other antiviral nucleoside analogues. Bio-chem Pharmacol 1999;57:1351-1359.

5. Kotra LP, Xiang Y, Newton MG, Schinazi RF, Cheng YC, Chu CK. Struc-ture-activity relationships of 29-deoxy-29,29-difluoro-L-erythro-pento-furanosyl nucleosides. J Med Chem 1997;40:3635-3644.

6. Kukhanova M, Lin ZY, Yas’co M, Cheng YC. Unique inhibitory effect of1-(29-deoxy-29-fluoro-b,L-arabinofuranosyl)-5-methyluracil 59-triphos-phate on Epstein-Barr virus and human DNA polymerases. BiochemPharmacol 1998;55:1181-1187.

7. Ma T, Lin JS, Newton MG, Cheng YC, Chu CK. Synthesis and anti-hepatitis B virus activity of 9-(2-deoxy-2-fluoro-b,L-arabinofuranosyl)purine nucleosides. J Med Chem 1997;40:2750-2754.

8. Ma T, Pai SB, Zhu YL, Lin JS, Shanmuganathan K, Du J, Wang C, et al.Structure-activity relationships of 1-(2-deoxy-2-fluoro-b,L-arabinofurano-syl)pyrimidine nucleosides as anti-hepatitis B virus agents. J Med Chem1996;39:2835-2843.

9. Xu AS, Chu CK, London RE. 19F NMR study of the uptake of 29-fluoro-5-methyl-beta-L-arabinofuranosyluracil in erythrocytes: evidence oftransport by facilitated and nonfacilitated pathways. Biochem Pharmacol1998;55:1611-1619.

10. Yao GQ, Liu SH, Chou E, Kukhanova M, Chu CK, Cheng YC. Inhibition ofEpstein-Barr virus replication by a novel L-nucleoside, 29-fluoro-5-methyl-beta-L-arabinofuranosyluracil. Biochem Pharmacol 1996;51:941-947.

11. Tennant BC, Baldwin BH, Graham LA, Ascenzi MA, Hornbuckle WE,Rowland PH, Tochkov IA, et al. Antiviral activity and toxicity of fialuri-dine in the woodchuck model of hepatitis B virus infection. HEPATOLOGY

1998;28:179-191.12. Witcher JW, Boudinot FD, Baldwin BH, Ascenzi MA, Tennant BC, Du JF,

Chu CK. Pharmacokinetics of 1-(2-fluoro-5-methyl-b,L-arabinofurano-syl)uracil in woodchucks. Antimicrob Agents Chemother 1997;41:2184-2187.

13. Kleiner DE, Gaffey MJ, Sallie R, Tsokos M, Nichols L, McKenzie R, StrausSE, et al. Histopathologic changes associated with fialuridine hepatotox-icity. Mod Pathol 1997;10:192-199.

14. McKenzie R, Fried MW, Sallie R, Conjeevaram H, Di Bisceglie AM, ParkY, Savarese B, et al. Hepatic failure and lactic acidosis due to fialuridine(FIAU), an investigational nucleoside analogue for chronic hepatitis B.N Engl J Med 1995;333:1099-1105.

15. Fourel I, Hantz O, Watanabe KA, Jacquet C, Chomel B, Fox JJ, Trepo C.Inhibitory effects of 29-fluorinated arabinosyl-pyrimidine nucleosides towoodchuck hepatitis virus replication in chronically infected wood-chucks. Antimicrob Agents Chemother 1990;34:473-475.

16. Gerin JL. Experimental WHV infection of woodchucks: an animal modelof hepadnavirus-induced liver cancer. Gastroenterol Jpn 1990;25:38-42.

17. Tennant BC, Hornbuckle WE, Baldwin BH, King JM, Cote P, Popper H,Purcell RH, et al. Influence of age on the response to experimental wood-chuck hepatitis virus infection. In Zuckerman AJ, ed. Viral Hepatitis andLiver Disease. New York: Alan Liss, 1988:462-465.

18. Venkateswaran PS, Millman I, Blumberg BS. Effects of an extract fromPhyllanthus niruri on hepatitis B and woodchuck hepatitis viruses: in vitroand in vivo studies. Proc Natl Acad Sci U S A 1987;84:274-278.

19. Nordenfelt E, Widell A, Hansson BG, Lofgren B, Moller-Nielsen C, ObergB. No in vivo effect of trisodium phosphonoformate on woodchuck hep-atitis virus production. Acta Pathol Microbiol Immunol Scand [B] 1992;90:449-451.

20. Gerin J, Korba B, Cote P, Tennant B. A preliminary report of a controlledstudy of thymosin alpha-1 in the woodchuck model of hepadnavirusinfection. In: Block T, Jungkind D, Crowell R, Denison M, eds. Innova-tions in Antiviral Development and the Detection of Virus Infection. NewYork: Plenum, 1992:121-123.

21. Fiume L, DiStefano G, Busi C, Mattioli A, Rapicetta M, Giuseppetti R,Ciccaglione AR, et al. Inhibition of woodchuck hepatitis virus replicationby adenine arabinoside monophosphate coupled to lactosaminated poly-L-lysine and administered by intramuscular route. HEPATOLOGY 1995;22:1072-1077.

22. Bartholomew RM, Carmichael EP, Findeis MA, Wu CH, Wu GY. Tar-geted delivery of antisense DNA in woodchuck hepatitis virus–infectedwoodchucks. J Viral Hepat 1995;2:273-278.

23. Korba BE, Xie H, Wright KN, Hornbuckle WE, Gerin JL. Liver-targetedantiviral nucleosides: enhanced antiviral activity of phosphatidyl-dideoxyguanosine versus dideoxyguanosine in woodchuck hepatitis vi-rus infection in vivo. HEPATOLOGY 1996;23:958-963.

24. Tencza, MG, Newbold JE. Heterogeneous response for a mammalianhepadnavirus infection to acyclovir: drug-arrested intermediates of mi-nus-strand viral DNA synthesis are enveloped and secreted from infectedcells as virion-like particles. J Med Virol 1997;51:6-16.

25. Gangemi D, Korba B, Tennant B, Gerin J, Cote P, Ueda H, Jay G. Antiviraland anti-proliferative activities of a-interferon in experimental hepatitisB virus infections. Antiviral Ther 1997;1:64-70.

26. Mason WS, Cullen J, Moraleda G, Saputelli J, Aldrich CE, Miller DS,Tennant B, et al. Lamivudine therapy of WHV-infected woodchucks.Virology 1998;245:18-32.

27. Zahm FE, Bonino F, Giuseppetti R, Rapicetta M. Antiviral activity ofganciclovir, 9-(1,3-dihydroxy-2-propoxymethyl) guanine against wood-chuck hepatitis virus: quantitative measurement of woodchuck hepatitis

HEPATOLOGY Vol. 33, No. 1, 2001 PEEK ET AL. 265

virus DNA using storage phosphor technology. Ital J Gastroenterol Hepa-tol 1998;30:510-516.

28. Cullen JM, Smith SL, Davis MG, Dunn SE, Botteron C, Cecchi A, LinseyD, et al. In vivo antiviral activity and pharmacokinetics of (2)-cis-5-fluoro-1- [2-(hydroxymethyl)-1,3-oxathiolan-5-yl]cytosine in wood-chuck hepatitis virus-infected woodchucks. Antimicrob Agents Che-mother 1997;41:2076-2082.

29. Block TM, Lu X, Mehta AS, Blumberg BS, Tennant B, Ebling M, Korba B,et al. Treatment of chronic hepadnavirus infection in a woodchuck ani-mal model with an inhibitor of protein folding and trafficking. Nat Med1998;4:610-614.

30. Genovesi EV, Lamb L, Medina I, Taylor D, Seifer M, Innaimo S, ColonnoRJ, et al. Efficacy of the carbocyclic 29-deoxyguanosine nucleoside BMS-200475 in the woodchuck model of hepatitis B virus infection [publishederratum appears in Antimicrob Agents Chemother 1999;43:726]. Anti-microb Agents Chemother 1998;42:3209-3217.

31. Hurwitz S, Tennant B, Korba B, Gerin J, Schinazi R. Viral pharmacody-namics of (2)-B-29,39-dideoxy-39-thiacytidine in chronically infectedwoodchucks compared to humans. Antimicrob Agents Chemother 1998;42:2804-2809.

32. Korba BE, Cote PJ, Hornbuckle WE, Schinazi RF, Gerin JL, Tennant BC.Enhanced antiviral benefit of combination therapy with lamivudine andfamciclovir against WHV replication in chronic WHV carrier wood-chucks. [Corrigendum published in Antivir Res 2000;45;155-156]. An-tiviral Res 2000;45:19-22.

33. Korba BE, Schinazi RF, Cote PJ, Tennant BC, Gerin JL. Effect of oraladministration of emtricitabine on woodchuck hepatitis virus replicationin chronically-infected woodchucks. Antimicrobl Agents Chemother2000;44:1757-1760.

34. Korba BE, Cote PJ, Hornbuckle WE, Tennant BC, Gerin JL. Treatment ofchronic WHV infection in the Eastern woodchuck (M. monax) with nu-cleoside analogues is predictive of therapy for chronic hepatitis B virusinfection in man. HEPATOLOGY 2000;31:1165-1175.

35. Korba BE, Cote PJ, Hornbuckle WE, Schinazi RF, Gangemi DL, TennantBC, Gerin JL. Enhanced antiviral benefit of combination therapy withlamivudine and alpha interferon against WHV replication in chronicWHV carrier woodchucks. Antiviral Therapy 2000;5:95-104.

36. Hostetler KY, Beadle JR, Korba BE, Hornbuckle WE, Bellezza CA, TochovIA, Cote PJ, et al. Antiviral activity of oral 1-O-hexadecyl-propanediol-3-P-acyclovir in woodchucks with chronic woodchuck hepatitis virusinfection. Antimicrob Agents Chemother 2000;44:1964-1969.

37. Wright JD, Ma T, Chu CK, Boudinot FD. Discontinuous oral absorptionpharmacokinetic model and bioavailability of 1-(2-fluoro-5-methyl-b,L-arabinofuranosyl)uracil (L-FMAU) in rats. Biopharm Drug Dispos1996;17:197-207.

38. Wright JD, Ma T, Chu CK, Boudinot FD. Pharmacokinetics of 1-(2-deoxy-2-fluoro-beta-L-arabinofuranosyl)-5-methyluracil (L-FMAU) inrats. Pharmacol Res 1995;12:1350-1353.

39. Liu SH, Grove KL, Cheng YC. Unique metabolism of a novel antiviralL-nucleoside analog, 29-fluoro-5- methyl-b,L-arabinofuranosyluracil: asubstrate for both thymidine kinase and deoxycytidine kinase. Antimi-crob Agents Chemother 1998;42:833-839.

40. Cote PJ, Korba BE, Miller RH, Jacob JR, Baldwin BH, Hornbuckle WE,Purcell RH, et al. Effects of age and viral determinants on chronicity as anoutcome of experimental woodchuck hepatitis virus infection. HEPATOL-OGY 2000;31:190-200.

41. Korba BE, Wells F, Tennant BC, Yoakum GH, Purcell RH, Gerin JL.Hepadnavirus infection of peripheral blood lymphocytes in vivo: wood-chuck and chimpanzee models of viral hepatitis. J Virol 1986;58:1-8.

42. Korba BE, Wells FV, Baldwin B, Cote PJ, Tennant BC, Popper H, Gerin JL.Hepatocellular carcinoma in woodchuck hepatitis virus–infected wood-chucks: presence of viral DNA in tumor tissue from chronic carriers andanimals serologically recovered from acute infections. HEPATOLOGY 1989;9:461-470.

43. Cote PJ, Roneker C, Cass K, Schodel F, Peterson D, Tennant B, DeNo-ronha F, Gerin JL. New enzyme immunoassays for the serologic detec-tion of woodchuck hepatitis virus infection. Viral Immunol 1993;6:161-169.

44. Ponzetto A, Cote PJ, Ford EC, Purcell RH, Gerin JL. Core antigen andantibody in woodchucks after infection with woodchuck hepatitis virus.J Virol 1984;52:70-76.

45. Jacob JR, Ascenzi MA, Roneker CA, Toshkov IA, Cote PJ, Gerin JL,Tennant BC. Hepatic expression of the woodchuck hepatitis X-antigenduring acute and chronic infection and detection of a woodchuck hepa-titis virus X-antigen antibody response. HEPATOLOGY 1997;26:1607-1615.

46. Cote PJ, Toshkov I, Bellezza C, Ascenzi M, Roneker C, Graham LA,Baldwin BH, et al. Temporal pathogenesis of experimental neonatalwoodchuck hepatitis virus infection: Increased initial viral load and de-creased severity of acute hepatitis during the development of chronicviral infection. HEPATOLOGY 2000;32:807-817.

47. Wu TT, Coates L, Aldrich CE, Summers J, Mason WS. In hepatocytesinfected with duck hepatitis B virus, the template for viral RNA synthesisis amplified by an intracellular pathway. Virology 1990;175:255-261.

48. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acidguanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem1987;162-156.

49. Summers J, Mason WS. Replication of the genome of a hepatitis-B–likevirus by reverse transcription of an RNA intermediate. Cell 1982;29:403-415.

50. Ganem D. Hepadnaviridae: the viruses and their replication. In: FieldsBN, Knipe DM, Howley PM, eds. Fields Virology. Philadelphia: Lippin-cott-Raven, 1997:2703-2737.

51. Jarvis B, Faulds D. Lamivudine. A review of its therapeutic potential inchronic hepatitis B. Drugs 1999;58:101-141.

52. Seeger C, Mason WS. Hepatitis B virus biology. Microbiol Mol Biol Rev2000;64:51-68.

53. Nowak MA, Bonhoeffer S, Hill AM, Boehme R, Thomas HC, McDade H.Viral dynamics in hepatitis B virus infection. Proc Natl Acad Sci U S A1996;93:4398-4402.

54. Tsiang M, Rooney JF, Toole JJ, Gibbs CS. Biphasic clearance kinetics ofhepatitis B virus from patients during adefovir dipivoxil therapy. HEPA-TOLOGY 1999;29:1863-1869.

266 PEEK ET AL. HEPATOLOGY January 2001