In Search of a Selective Antiviral Chemotherapy

20 Pages·1997·3 MB·English

CLINICALMICROBIOLOGYREVIEWS,Oct.1997,p.674–693 Vol.10,No.4
0893-8512/97/$04.00(cid:49)0
Copyright©1997,AmericanSocietyforMicrobiology
In Search of a Selective Antiviral Chemotherapy
ERIKDECLERCQ*
RegaInstituteforMedicalResearch,KatholiekeUniversiteitLeuven,B-3000Leuven,Belgium
INTRODUCTION.......................................................................................................................................................674
SELECTIVEANTIHERPESVIRUSAGENTS........................................................................................................674
BROAD-SPECTRUMANTI-DNAVIRUSAGENTS..............................................................................................677
SELECTIVEINHIBITORSOFRETROVIRUSESANDHEPADNAVIRUSES................................................681
SELECTIVEINHIBITORSOFHIV-1....................................................................................................................683
KNOCKINGOUTTHEVIRUS...............................................................................................................................686
OUTLOOKFORANTI-HIVDRUGS......................................................................................................................688
D
ACKNOWLEDGMENTS...........................................................................................................................................688
o
REFERENCES............................................................................................................................................................689 w
n
lo
INTRODUCTION primarily in the treatment of some hemorrhagic fever virus a
d
infections (e.g., Lassa fever) and respiratory syncytial virus e
Viral diseases have for a long time been considered intrac- infections. Although ribavirin (in aerosolized form) has been d

table by chemotherapeutic means because of the innate asso- approved for the latter indication (in 1985), its effectiveness fr
o
ciation of viruses with the normal cell machinery. The first, hasremainedcontroversial(70). m
albeittimid,attemptstocurtailvirusinfectionsdatefromthe In retrospect, it can be stated that antiviral chemotherapy
h
early 1950s when the thiosemicarbazones, introduced by came of age with the advent in 1977 of acyclovir [9-(2-hy- t
t
Domagk et al. (83) for the treatment of tuberculosis, were droxyethoxymethyl)guanine]asthefirsttrulyspecificantiviral p
:
foundtobeactiveagainstvacciniavirusaswell(93).Thiswork agent(86,156).Thecompoundwasfoundtoeffectaselective //
c
was continued by Bauer (36) and culminated in the demon- inhibitionofthereplicationofHSV,essentiallybecauseitwas m
strationbyBaueretal.(37)in1963thatthethiosemicarbazone phosphorylated by the virus-induced thymidine kinase (TK) r
.
derivativeMarboranwaseffectiveintheprophylaxisofsmall- (91). a
s
poxinfection. From its chemical structure, it could be hardly anticipated m
Thefirstantiherpescompound,IDU(idoxuridine),wassyn- that acyclovir would be recognized as substrate by a TK, .
o
thesizedbyPrusoff(145)in1959;itwasthenshownbyKauf- whether virus induced or not. At best, the guanine part of r
g
man to “cure” herpetic keratitis in rabbits (102) and humans acyclovir could be viewed as bearing some resemblance to /

(104) and in the same year (1962), it was approved in the cytosine, and this would explain why it was recognized by the o
n
United States for the (topical) therapy of herpetic keratitis. viralTK,whichactsnotonlyasaTKbutalsoasadeoxycyti-
M
Two years later (1964), TFT (trifluridine) was introduced for dinekinase(54).
a
the treatment of herpetic keratitis (103), and both IDU and Empirical search has been the cornerstone of both initial r
c
TFT, now more than 30 years later, are still used for this discovery and subsequent improvement of nearly all useful h
purpose(aseyedropsand,occasionally,aseyeointment). chemotherapeutic agents (155). The first synthetic centrally 2
The first description of the antiviral activity of amantadine active drugs, i.e., diuretics, sulfonamides, cardiac glucosides, 9
,
(52) and vidarabine (144, 154) also dates from the “avant andantibioticssuchaspenicillin,wereinnowayinfluencedby 2
garde”year1964.Later,thesedrugswouldbeapprovedforthe any fundamental knowledge of the processes with which they 0
1
prophylaxis and treatment of influenza A virus infections (in interfere (116). This knowledge was acquired later through 9
1966) and for the therapy of herpes simplex virus (HSV) in- logicandorderlythinking.Itwouldappearthatantiviralche- b
fections(e.g.,herpeticencephalitis)(in1972). motherapyhasmeanderedalongsucharoute,withanempir- y

Inthemeantime,interferonandinterferoninducers[partic- ical start and a logical and rational, yet often winding course g
u
ularly the double-stranded RNA types of interferon inducers beforearriving(successfully)atitsfinaldestination. e
suchaspoly(I-C)]attractedconsiderableattentionasantiviral (This review is an extended version of the Hoechst Marion s
t
agentsbecauseoftheirbroad-spectrumantiviralactivity(55). Roussel Award Lecture presented at the 36th Interscience
Althoughinterferonhasshownabeneficialeffectinthetreat- ConferenceonAntimicrobialAgentsandChemotherapy,New
mentofanumberofvirusinfections,includinghepatitisBand Orleans,La.,15to18September1996.)
hepatitisC,itwouldfinditsmajortherapeutic“niche”outside
thefieldofantiviralagents,namely,forthetreatmentofsome
SELECTIVEANTIHERPESVIRUSAGENTS
formsofcancerand,nowadays,alsoforthetreatmentofmul-
tiplesclerosis. Shortlyafterithadbeendescribedasapotentandselective
Thefirstsyntheticchemicaltobecreditedwithbroad-spec- inhibitor of the replication of HSV and, to a lesser extent, of
trumantiviralactivitywasribavirin(Virazole).Whilefirstde- varicella-zoster virus (VZV), acyclovir became the drug of
scribed in 1972 as a broad-spectrum antiviral agent active choice for the treatment of HSV and VZV infections, partic-
againstbothDNAandRNAviruses(159),ithasprovenuseful ularly primary and recurrent genital herpes and mucocutane-
ous HSV and VZV infections in immunosuppressed patients.
Also, acyclovir proved to be superior to vidarabine in the
*Mailingaddress:RegaInstituteforMedicalResearch,Katholieke treatmentofherpeticencephalitisandVZVinfectionsandhas
Universiteit Leuven, Minderbroedersstraat 10, B-3000 Leuven, Bel- replaced vidarabine for these indications (63). Because of its
gium.Phone:[phone].Fax:[phone]. limited oral bioavailability (only 20%), acyclovir has, in turn,
674
VOL.10,1997 SELECTIVE ANTIVIRAL CHEMOTHERAPY 675
D
o
w
n
lo
a
d
e
d

f
r
o
m

h
t
FIG. 1. Arabinofuranosyladenineandacyclicnucleosideanalogs.Ara-A,vidarabine,9-(cid:98)-D-arabinofuranosyladenine;ACV,acyclovir,9-(2-hydroxyethoxymethyl)guanine; tp
GCV, ganciclovir, 9-(1,3-dihydroxypropoxymethyl)guanine; PCV, penciclovir, 9-(4-hydroxy-3-hydroxymethyl-but-1-yl)guanine; VACV, valaciclovir, L-valyl ester of :/
acyclovir;FCV,famciclovir,diacetylesterof6-deoxypenciclovir. /c
m
r
.
a
beenreplacedbyitsprodrug,valaciclovir,intheoraltreatment s
m
ofHSVandVZVinfections(47).
.
o
TheremarkablepotencyofacycloviragainstHSV-1,HSV-2,
r
andtoalesserextentVZVhaspromptedthedevelopmentof g
/
several structural analogs of acyclovir (Fig. 1) (65). Foremost o
amongtheseacyclovircongenersareganciclovirandpenciclo- n

vir. Ganciclovir has a more pronounced activity against cyto- M
megalovirus (CMV) than acyclovir and became the drug of a
r
choice(untiltheadventofHPMPC[seebelow])forthetreat- c
h
mentofCMVinfections,particularlyCMVretinitis,inimmu-
2
nosuppressed patients (121). Penciclovir has a similar activity 9
spectrum and mechanism of action to acyclovir and has been ,
2
used under its oral prodrug form, famciclovir, for the same 0
indications as valaciclovir, i.e., for the treatment of HSV and 1
9
VZVinfections(88,182).Inaddition,penciclovirshowsactiv-
b
ityagainsthumanhepatitisBvirus. y

Whereas acyclovir emanated from a program centered on g
u
examining the substrate specificity of adenosine deaminase
e
(which is known to rapidly convert the antiviral compound s
vidarabine [ara-A, adenine arabinoside, 9-(cid:98)-D-arabinofurano- t
syladenine] to its inactive metabolite ara-Hx [hypoxanthine
arabinoside]),brivudin[(E)-5-(2-bromovinyl)-2(cid:57)-deoxyuridine,
BVDU] was originally developed at the Chemistry Depart-
ment of the University of Birmingham by P. J. Barr, A. S.
Jones, and R. T. Walker as a potential radiation-sensitizing
agent (assuming that it would be incorporated into DNA).
Likeacyclovir,BVDUturnedouttobeapotentandselective
anti-HSV agent (74). In its potency and selectivity against
HSV-1,BVDUexceededallotheranti-HSVagents,including
acyclovir and the older 5-substituted 2(cid:57)-deoxyuridines such as
IDU,TFT,and5-ethyl-2(cid:57)-deoxyuridine(EDU)(58)(Fig.2).
The 2(E)-bromovinyl group, with the bromine in the trans FIG. 2. 5-Substituted 2(cid:57)-deoxyuridines. IDU, idoxuridine, 5-iodo-2(cid:57)-deoxy-
(orEntgegen)configuration,iscrucialfortheantiviralselectiv- uridine; TFT, trifluridine, 5-trifluoromethyl-2(cid:57)-deoxyuridine; EDU, 5-ethyl-2(cid:57)-
deoxyuridine;BVDU,brivudin,(E)-5-(2-bromovinyl)-2(cid:57)-deoxyuridine;BVaraU,
ity of BVDU, and hence various other compounds that have sorivudine, 1-(cid:98)-D-arabinofuranosyl-(E)-5-(2-bromovinyl)uracil; C-BVDU, car-
beensynthesizedsinceBVDUandthatsharewithBVDUthe bocyclicBVDU;BTDU,5-(5-bromothien-2-yl)-2(cid:57)-deoxyuridine.
676 DE CLERCQ CLIN.MICROBIOL.REV.
D
o
w
n
lo
a
d
e
d

f
r
o
m

h
t
t
p
:
/
FIG. 3. Mechanism of action of BVDU. Following uptake by the cells and intracellular phosphorylation by the virus-encoded TK to the 5(cid:57)-monophosphate /c
(BVDU-MP) and the 5(cid:57)-diphosphate (BVDU-DP) and further phosphorylation (presumably by the NDP kinase) to the 5(cid:57)-triphosphate (BVDU-TP), the last m
compoundactsasacompetitiveinhibitor/alternativesubstrateoftheviralDNApolymeraseandcanbeincorporatedinternally(viainternucleotidelinkage)intothe r
DNAchain. .a
s
m
.
o
same5-2(E)-bromovinylsubstituentdemonstratesimilarselec- described. At nanomolar concentrations, they inhibit the rep-
r
tivity,potency,andactivityspectrumtothoseofBVDU;these lication of VZV (4, 158). This is in marked contrast to the g
/
include BVaraU (sorivudine) (119), C-BVDU (carbocyclic established anti-VZV drugs, acyclovir and penciclovir, whose o
BVDU) (97), and S-BVDU (4(cid:57)-thio-BVDU) (Fig. 2) (185). inhibition of VZV replication takes place only at micromolar n

Also,BTDU(Fig.2),containinga5-(5-bromothien-2-yl)sub- concentrations. M
stituentwithabrominegroupinroughlythesamepositionas BVDU is also a potent inhibitor of the replication of Ep- a
r
in BVDU, exhibits similar selectivity and potency to BVDU stein-Barrvirus(EBV)invitro(118),butnotCMVorhuman c
h
(188). herpesvirus 6 (HHV-6) and 7 (HHV-7) (148). In addition,
2
The mechanism of action of BVDU (Fig. 3) depends on a BVDU has been found effective against a number of herpes- 9
specificphosphorylationbythevirus-encodedTK,i.e.,theTKs virusesofveterinaryimportancesuchasswineherpesvirustype ,
2
inducedbyHSV-1andVZV,whichconvertthecompoundto 1 (SHV-1), bovine herpesvirus type 1 (BHV-1), simian vari- 0
its 5(cid:57)-monophosphate (BVDU-MP) and 5(cid:57)-diphosphate cellavirus(SVV),andherpesvirusplatyrrhinae(HVP)butnot 1
9
(BVDU-DP)(5,79).Uponfurtherphosphorylationbycellular equine herpesvirus type 1 (EHV-1) (57). As for HSV-2, the
b
kinase(s), i.e., nucleoside 5(cid:57)-diphosphate (NDP) kinase, insensitivity of EHV-1 to BVDU could be attributed to an y
BVDU5(cid:57)-triphosphate(BVDU-TP)canactinadualfashion inefficientphosphorylationofBVDU-MPtoBVDU-DPinthe g
with the viral DNA polymerase: (i) as a competitive inhibitor virus-infectedcells(107). u
e
with respect to the natural substrate (dTTP) (3) or (ii) as an Due to its unique propensity for some virus-encoded TKs, s
t
alternativesubstrate,whichwouldthenallowBVDU-TPtobe BVDUcouldalsobeusedasacytostaticagent,providedthat
incorporated(asBVDU-MP)intotheDNAchain.Thisincor- thetumorcellshavebeentransfectedbythevirus-specificTK.
poration may, in turn, affect both the stability and the func- Thus, murine mammary FM3A carcinoma cells transformed
tioningoftheDNA.Infact,aclosecorrelationhasbeenfound with the HSV-1 TK gene are inhibited in their growth by
between the incorporation of BVDU into HSV-1 DNA, viral BVDUat1ng/ml,a10,000-fold-lowerconcentrationthanthat
infectivity,andDNAintegrity(16,120). requiredtoinhibittheuntransformedcancercells(12,13).The
BVDU displays a remarkable specificity in its antiviral ac- cytostaticactivityofBVDUagainstcancercellsthathavebeen
tivityspectrum:itisahighlypotentinhibitorofHSV-1butnot transfected by the HSV-1 (or HSV-2) TK gene is due to the
HSV-2, so that it can be used as a marker for differentiating inhibitory effect of BVDU-MP (formed intracellularly by the
HSV-2 from HSV-1 strains (57). The reason for the relative viral TK) on the cellular thymidylate synthase (14, 21). In
inactivityofBVDUagainstHSV-2isthattheHSV-2-encodedTK addition to BVDU, S-BVDU and BTDU (but not BVaraU)
is unable to phosphorylate BVDU-MP to BVDU-DP. This re- can serve as HSV TK-activated cytostatic agents targeted at
sultsinasubstantialreductioninthesupplyoftheactiveBVDU the thymidylate synthase (24, 41). BVDU and its congeners
metabolite,BVDU-TP,intheHSV-2-infectedcells(5,90). S-BVDUandBTDUshouldthereforebeconsideredpotential
BVDU and its arabinofuranosyl counterpart BVaraU be- candidatedrugsforthetreatmentofHSVTKgene-transfected
longtothemostpotentinhibitorsofVZVthathaveeverbeen tumors.
VOL.10,1997 SELECTIVE ANTIVIRAL CHEMOTHERAPY 677
70a). As a 5% cream, BVDU could be advocated for the
topicaltreatmentofherpeslabialis.Ithasbeenusedoccasion-
allyforthispurpose.
BROAD-SPECTRUMANTI-DNAVIRUSAGENTS
Emerging from a collaborative program (started in 1976)
with A. Hol´y at the Institute of Organic Chemistry and Bio-
chemistryoftheCzechoslovakAcademyofSciencesinPrague,
(S)-9-(2,3-dihydroxypropyl)adenine (DHPA) (Fig. 4) was re-
portedin1978asthefirstacyclicnucleosideanalogpossessing
broad-spectrumantiviralactivity(73).DHPAwaslatershown
tobeaninhibitorofS-adenosylhomocysteinehydrolase(183)
and thus paved the way for the development of various other
adenosine analogs, including neplanocin A and 3-deazanepl-
anocinA,asbroad-spectrumantiviralagentsowingtheiranti- D
viral activity to inhibition of S-adenosylhomocysteine hydro- o
lase (59). These compounds are particularly inhibitory to w
virusesthatrelyfortheirreplicationonS-adenosylmethionine- n
lo
dependent methylations (of the viral mRNA): i.e., negative- a
strand RNA viruses (rhabdoviruses, arenaviruses, and d
e
paramyxoviruses), double-strand RNA viruses (reovirus), and d
someDNAviruses(vacciniavirusandCMV)(77,162).Also, f
r
the replicative cycle of human immunodeficiency virus (HIV) o
m

h
t
TABLE 1. AntiviralactivityspectrumofHPMPC tp
:
Virusa Activity //
c
m
Polyomaviridae
Murinepolyomavirus...............................................................Active r.
a
Humanpolyomavirus...............................................................Active s
m
Papillomaviridae .
o
Rabbitpapillomavirus..............................................................Active r
FIG. 4. Acyclicnucleosides,phosphonates,andacyclicnucleosidephospho- HPV(severaltypes).................................................................Active g
nates.DHPA,(S)-9-(2,3-dihydroxypropyl)adenine;PAA,phosphonoaceticacid; /
o
PFA,phosphonoformicacid,foscarnet;HPMPA,(S)-9-(3-hydroxy-2-phospho- Adenoviridae n
nylmethoxypropyl)adenine; HPMPC, (S)-1-(3-hydroxy-2-phosphonylme-
Humanadenoviruses(severaltypes).....................................Active M
thoxypropyl)cytosine;cHPMPC,cyclicHPMPC.
a
Herpesviridae rc
HSV-1........................................................................................Active h
HSV-2........................................................................................Active 2
From a clinical viewpoint, BVDU has been used for many VZV...........................................................................................Active 9
years in the topical treatment (as 0.1% eyedrops) of herpetic EBV...........................................................................................Active , 2
keratitis,sinceitisefficaciousagainstvariousformsofherpetic HCMV.......................................................................................Active 0
keratitis(dendriticandgeographiccornealulcers,andstromal HHV-6.......................................................................................Active 1
9
keratitis) that were clinically resistant to other antiviral drugs HHV-7.......................................................................................Active
(i.e.idoxuridine,fluridine,vidarabine,oracyclovir)(122). TK(cid:50)HSV..................................................................................Active by
TK(cid:50)VZV.................................................................................Active
Initial (uncontrolled) clinical trials have pointed to the ef- g
PK(cid:50)HCMV..............................................................................Active u
fectiveness of oral BVDU (at a dose of 7.5 mg/kg/day [four
SVV............................................................................................Active e
125-mg tablets] in adults or 15 mg/kg/day in children for 5 s
EHV-1.......................................................................................Active t
days)inthetreatmentofHSV-1andVZVinfections(38,175, BHV-1.......................................................................................Active
189).TheefficacyofBVDUinthetreatmentofherpeszoster BHV-2.......................................................................................Active
was then proven in a randomized double-blind trial, in which MCMV......................................................................................Active
oralBVDU(at7.5mg/kg/day[four125-mgtablets]for5days) RCMV.......................................................................................Active
wasatleastaseffectiveasintravenousacyclovir(at30mg/kg/ GPCMV.....................................................................................Active
day[10mg/kgevery8h],for5days)inarrestingtheprogres-
Iridoviridae
sionofthedisease(190).Thesestudiesarenowbeingextended
Africanswinefevervirus.........................................................Active
in several European clinical centers to various other BVDU
dosageschedules(i.e.,125or50mgtwicedaily;and125,62.5,
Hepadnaviridae
or31.25mgoncedaily)incomparisonwithfive800-mgdoses HBV...........................................................................................Inactiveb
of acyclovir daily for 7 days in the management of herpes
zoster in both immunocompromised and immunocompetent Poxviridae
patients. Vacciniavirus...........................................................................Active
BVDUhasrepeatedlyshownmarkedefficacyinthetopical
aTK(cid:50), thymidine kinase deficient; PK(cid:50), protein kinase deficient; MCMV,
treatment of cutaneous HSV-1 infection in hairless (hr/hr)
murineCMV;RCMV,ratCMV;GPMCV,guineapigCMV.
micewhenappliedoveraconcentrationrangeof1to10%(56, bInactiveoronlyslightlyactive.
678 DE CLERCQ CLIN.MICROBIOL.REV.
D
o
w
n
lo
a
d
e
d

f
r
o
m

h
t
t
p
:
/
/
c
m
r
.
a
s
m
.
o
r
g
/

o
n

M
a
r
c
h

2
FIG. 5. MechanismofactionofHPMPC.Followinguptakebythecellsandintracellularphosphorylation(presumablybyapyrimidinenucleosidemonophosphate 9
[PNM]kinasefollowedbyNDPkinase)tothediphosphorylderivativeHPMPCpp,thelatterwouldblockDNAchainelongationastheresultoftwoconsecutive ,
incorporationsofHPMPCatthe3(cid:57)end(asdemonstratedwithCMV). 20
1
9

b
appears to be susceptible (presumably at the transcription and cytosine derivatives (termed HPMPDAP, HPMPG, and y
level) to the inhibitory effect of S-adenosylhomocysteine hy- HPMPC, respectively) exhibited marked activity against vari- g
u
drolase inhibitors (123). The therapeutic potential of these ousDNAviruses(76).
e
compoundshasyettomaterialize. The activity spectrum of HPMPC (Cidofovir) (Fig. 4) en- s
t
Starting with DHPA and phosphonoacetic acid, which had compasses papovaviruses (polyomaviruses and papillomavi-
long been recognized as an anti-DNA virus agent (and gave ruses), adenoviruses, herpesviruses, iridoviruses, and poxvi-
rise to phosphonoformic acid [foscarnet] [136], which is now ruses (Table 1), and virtually all members of the herpesvirus
used in the treatment of CMV disease), (S)-9-(3-hydroxy-2- family, including the TK-deficient HSV and VZV strains and
phosphonylmethoxypropyl)adenine (HPMPA) (Fig. 4) was the protein kinase-deficient CMV strains that are resistant to
conceived as a hybrid molecule combining the features of an acyclovir and ganciclovir, respectively, have been found to be
acyclic nucleoside analog and a phosphonate analog (60). susceptibletoHPMPC(62).TheantiviralefficacyofHPMPC
HPMPA was found to be active against a broad spectrum of has been demonstrated in a large variety of animal models
DNA viruses, including papovaviruses, adenoviruses, herpes- (62).WhenHPMPCwascomparedwiththe“classical”antivi-
viruses,iridoviruses,poxviruses,andhepadnaviruses(75),and ral drugs, it proved clearly more efficacious than acyclovir
its in vivo efficacy was demonstrated in a variety of animal againstHSVinfectioninmice(71)andmoreefficaciousthan
modelsforHSVandvacciniavirusinfection(78). ganciclovir against CMV infection in mice (133). Also, in a
Starting from HPMPA, several other HPMP analogs were rabbitmodelforviralretinitis,HPMPC,followingintravitreal
synthesized, and while the hypoxanthine, uracil, and thymine injection,affordedamorepotentandprolongedantiviraleffect
derivatives(termedHPMPHx,HPMPU,andHPMPT,respec- thanganciclovirdid(89).
tively) were virtually inactive, the 2,6-diaminopurine guanine HPMPC has been used primarily as an anti-CMV agent: it
VOL.10,1997 SELECTIVE ANTIVIRAL CHEMOTHERAPY 679
TABLE 2. MajorclinicalindicationsofHPMPC TABLE 3. AntiviralactivityspectrumofPMEAandPMPA
Topical Activityof:
MucocutaneousHSVlesions(HPMPCgel),particularlywhen Virus
PMEA PMPA
resistanttoacyclovir
Recurrentgenitalherpes(HPMPCgel) Herpesviridae
MucocutaneousHPVlesions(HPMPCgelorintratumoral HSV-1 Active Inactivea
injections) HSV-2 Active Inactive
HSVoradenoviruskeratoconjunctivitis(HPMPCeyedrops) VZV Active Inactive
CMVretinitis(intravitrealinjection) EBV Active
HCMV Active Inactive
Systemica TK(cid:50)HSV Active Inactive
CMVretinitis(inimmunocompromisedpatients) TK(cid:50)VZV Active Inactive
SystemicCMVinfections(inimmunocompromisedpatients)
HSVandVZVinfections(inimmunocompromisedpatients), Hepadnaviridae
particularlywhenresistanttoacyclovir HBV Active Active
EBV-associateddiseasessuchashairyleukoplakiaandothers(to DHBV Active Active
befurtherexplored) D
o
Retroviridae
aIntravenous. w
HIV-1 Active Active n
HIV-2 Active Active lo
SIV Active Active a
inhibitshumanCMV(HCMV)replicationataconcentration FIV Active Active de
about 1,000-fold lower than the concentration that is toxic to Visna/maedivirus Active Active d
Felineleukemiavirus Active
thehostcells(160),andthisselectivityisalsoreflectedinthe LP-BM5(murineAIDS)virus Active Active fro
1,000-fold-lowerconcentrationrequiredforHPMPCtoinhibit
MSV Active Active m
HCMV DNA synthesis than for it to inhibit host cell DNA

aInactiveoronlyslightlyactive. h
t
t
p
:
/
/
synthesis(132).ToachievethisinhibitoryeffectonviralDNA c
m
synthesis,HPMPCmustbeconvertedintracellularlytoitsac-
r
tive metabolite, the diphosphorylated form of HPMPC, .a
HPMPCpp (Fig. 5). HPMPC enters the cells by endocytosis s
m
(46) and is then metabolized, presumably through successive
.
o
phosphorylation by a pyrimidine nucleoside monophosphate
r
(PNM) kinase and NDP kinase, to HPMPCpp (45, 101). g
/
HPMPCpp acts as both a competitive inhibitor and an alter- o
native substrate of dCTP in the viral DNA polymerase reac- n

tion, thereby exhibiting a much greater affinity (lower K) for M
theDNApolymerasesofHSV-1,HSV-2,andHCMVthainfor a
r
thecellularDNApolymerases(cid:97),(cid:98),and(cid:103)(43,98).Theincor- c
h
poration of a single HPMPC molecule into DNA by HCMV
2
DNApolymerasedoesnotleadtochaintermination(192).To 9
effectivelyterminatetheDNAchain(Fig.5),twoconsecutive ,
2
HPMPCmoleculesmustbeincorporatedatthe3(cid:57)end(191). 0
HPMPC adds a new dimension to the “art” of antiviral 1
9
chemotherapy in that it confers a long-lasting antiviral re-
b
sponsethatlastsforseveraldays(orevenweeks)afterasingle y

administration, which allows infrequent dosing of the com- g
u
pound(167).Thisprolongedantiviralresponsemaybeattrib-
e
uted to the accumulation inside the cells of the HPMPC me- s
t
tabolites HPMPCp (half-life, 24 h), HPMPCpp (half-life,
65h),andHPMPCpcholine(half-life,87h)(1,98).Inpartic-
ular,theHPMPCpcholineadductwouldactasareservoirfor
the generation of HPMPCpp, thus explaining the prolonged
antiviralactivityofHPMPC.Inkeepingwiththelongintracel-
lular half-life of HPMPCpp and HPMPCp choline, a pro-
longed elimination phase (up to 36 h) has been observed for
HPMPC in monkeys (49), which again supports infrequent
dosesofthedrugforantiviraltherapy.
Fromaclinicalviewpoint(Table2),HPMPChasbeenpur-
sued primarily for the treatment of CMV, HSV, and human
papillomavirus (HPV) infections. The dose-limiting toxicity fol-
FIG. 6. Acyclicnucleosidephosphonates.PMEA,9-(2-phosphonylmethoxy- lowing systemic (i.e., intravenous) administration of HPMPC
ethyl)adenine; PMEDAP, 9-(2-phosphonylmethoxyethyl)-2,6-diaminopurine; hasproventobenephrotoxicity,andthisnecessitatesthecon-
PMPA,(R)-9-(2-phosphonylmethoxypropyl)adenine;PMPDAP,(R)-9-(2-phos-
comitant administration of probenecid. The cyclic form of
phonylmethoxypropyl)-2,6-diaminopurine;bis(POM)-PMEA,bis(pivaloyloxym-
ethyl)esterofPMEA. HPMPC, cHPMPC (Fig. 4), is significantly less nephrotoxic
680 DE CLERCQ CLIN.MICROBIOL.REV.
D
o
w
n
lo
a
d
e
d

f
r
o
m

h
t
t
p
:
/
/
c
m
r
.
a
s
m
.
o
r
g
/

o
n

M
a
r
FIG. 7. MechanismofactionofPMEA.Followinguptakebythecellsandintracellularphosphorylation(byeitherAMPkinaseor5-phosphoribosyl-1-pyrophos- c
phate[PRPP]synthetase)tothediphosphorylderivativePMEApp,thelatteractsasaDNAchainterminatorintheRTreaction. h

2
9
,

2
thanHPMPC(39)becauseitismoreefficientlysecretedbythe resistant HSV infections (161). In two patients, an AIDS pa- 0
kidneys (48), and, since this is reflected by an improved ther- tientwithaperinealHSV-2infectionresistanttoacyclovir,and 1
9
apeuticindex,cHPMPChasalsobeenselectedasadrugcan- a bone marrow transplant recipient with an orofacial HSV-1
b
didateforfurtherdevelopment. infection resistant to both ganciclovir and foscarnet, topical y

HPMPC has been recently approved by the U.S. Food and treatmentwith1%HPMPC(inBeelerbaseorOrabase)once g
u
DrugAdministrationforthesystemic(intravenous)treatment daily for three consecutive days resulted in complete regres-
e
of CMV retinitis in patients with AIDS. This approval came sionofthelesions(163).WhentheHSVlesionsrecurred,the s
t
after it was found that HPMPC, when administered intrave- recurringvirusappearedtohaveregainedsusceptibilitytoacy-
nouslyat5mg/kgonceweeklyfor2weeks(inductiontherapy) clovir, but it then redeveloped resistance to it, so that the
and then once every other week (maintenance therapy), ex- HPMPCtreatmenthadtobereinstituted.Atnopointduring
tendedthemediantimetoprogressionofretinitisfrom22days therecurrentherpesvirusepisodesdidthevirusdevelopresis-
(deferred group) to 120 days (treated group) (P (cid:44) 0.0001) tance to HPMPC (163). These observations point to the use-
(114). Not only did HPMPC prove efficacious in delaying the fulness of HPMPC in the topical treatment of acyclovir-resis-
progression of previously untreated CMV retinitis, but also it tant HSV infections. The fact that TK(cid:50) VZV, while resistant
causedasignificantdelayintheprogressionofrelapsingCMV toacyclovir,remainssusceptibletoHPMPC(164)alsopoints
retinitis (following treatment with ganciclovir). Prolonged ar- to the potential usefulness of HPMPC for the treatment of
restoftheprogressionofCMVretinitishasalsobeenobtained acyclovir-resistantVZVinfections.
following intravitreal injection of HPMPC in patients with When intravenously administered HPMPC was being eval-
AIDS(105,106).Followingasingleintravitrealinjectionof20 uated for its anti-CMV activity based on inhibition of CMV
(cid:109)g of HPMPC, progression of the disease was stopped for a excretionintheurine(112),itwasnotedthatanintercurrent
mediantimeof55days;ifthisinjectionwasrepeated,ittook HSV-2 infection that was refractory to acyclovir responded
amediantimeof63daysforthediseasetoprogress(105). remarkably well to HPMPC treatment (111): a patient with
HPMPC has been hailed as a new topical treatment for AIDS who had had a 6-month history of acyclovir-resistant
VOL.10,1997 SELECTIVE ANTIVIRAL CHEMOTHERAPY 681
TABLE 4. MajorclinicalindicationsforPMEAandPMPA ble 3). In particular, PMEA has proved to be inhibitory to a
widerangeofretroviruses(leukemia/sarcomavirusesandim-
PMEAinitsoralprodrugform[bis(POM)-PMEA(Adefovir
munodeficiency viruses) both in vitro and in vivo (130). The
dipivoxil)]:
antiretrovirusactivityofPMEAhasbeendocumentedinvivo
HIVinfection(AIDS)
inavarietyofretrovirusmodels,i.e.,againstsimianimmuno-
HBVinfection(hepatitisB)
IntercurrentherpesvirusinfectionsinHIV-infectedindividuals deficiencyvirus(SIV)infectioninmonkeys(19),felineimmu-
nodeficiencyvirus(FIV)infectionincats(85),murine(Molo-
PMPA(fromaprospectiveviewpoint): ney) sarcoma virus (MSV) infection in mice (15), and visna
HIVinfection(AIDS) virus infection in lambs (174). As PMEA has also demon-
HBVinfection(hepatitisB) stratedactivityagainstherpesvirus(e.g.,HSV)infections(78)
PreventionofHIVinfection and hepadnavirus (e.g. hepatitis B virus [HBV]) infections
Bytheparenteralroute(needlestick):systemicPMPA
(95), both in vitro and in vivo, it seems to be a particularly
Perinatally(mother-to-child):systemicPMPA
promising drug candidate for the treatment of mixed HIV,
Throughsexualintercourse:topicalPMPA(gel)
HBV,and/orherpesvirusinfections.
WhereastheantiviralactivityspectrumofPMEAand9-(2-
phosphonylmethoxyethyl)-2,6-diaminopurine (PMEDAP) en- D
perinealHSV-2infectionshowed,followingfourweeklydoses compasses retroviruses, hepadnaviruses, and herpesviruses, o
w
of intravenous HPMPC at 5 mg/kg/week, an almost complete theactivityspectrumofPMPA,(R)-9-(2-phosphonylmethoxy-
n
healingofthelesions.Anotherpatientinthisstudygrouphad propyl)-2,6-diaminopurine (PMPDAP), and FPMPA is re- lo
bilateral oral hairy leukoplakia (presumably due to Epstein- stricted to retroviruses and hepadnaviruses (Table 3). In a
d
Barrvirusinfection):thelesionsresolvedwithin1weekaftera particular,PMPAandPMPDAPwerefoundtobehighlypo-
e
singleintravenousadministrationofHPMPC(5mg/kg)(112). tent inhibitors of HIV replication in cell culture and MSV- d

Two randomized, double-blind, placebo-controlled trials induced tumor formation in mice (22). Subsequently, PMPA f
r
haverecentlydemonstratedtheefficacyofHPMPCtopicalgel andPMPDAPwerealsofoundtobehighlypotentandselec- o
m
(at 0.3, 1, or 3%) in the treatment of acyclovir-resistant HSV tiveinhibitorsofvisnavirusreplication(173),humanHBVand

infections in AIDS patients (113) and in the (single-dose) duckHBVreplication(96),andFIVinfectionbothinvitroand h
t
therapy of recurrent genital herpes in immunocompetent pa- in vivo in cats (179). From both the mouse and cat experi- tp
tients(153),respectively. :/
/
OfmajorclinicalrelevanceistheactivityofHPMPCagainst c
m
HPV-associated lesions, which was first demonstrated in a
r
patient with extensive squamous papillomatous lesions of the .a
hypopharynxandesophagus(180):thelesionsshowedacom- s
m
plete and permanent regression following a few weekly injec-
.
o
tionsof1.25mgofHPMPCintothetumor.Asimilarcomplete
r
and permanent regression has been noted following intratu- g
/
moral injection of HPMPC in a number of patients with re- o
current laryngeal papillomatosis (166). Finally, HPMPC topi- n

cal gel or ointment (at 1%) has been found to induce a M
complete regression of relapsing anogenital HPV lesions in a
r
three patients with AIDS (165), and a double-blind, placebo- c
h
controlledtrial(inimmunocompetentpatients)hasnowbeen
2
initiated to corroborate the initial clinical findings with 9
HPMPCinthetopicaltreatmentofgenitalwarts. ,
2
0
1
SELECTIVEINHIBITORSOFRETROVIRUSES 9
ANDHEPADNAVIRUSES b
y

Simultaneously with HPMPA, [9-(2-phosphonylmethoxy- g
u
ethyl)adenine (PMEA) was mentioned as an antiviral agent
e
(75).PMEA(Adefovir)(Fig.6)canbeconsideredatruncated s
t
formofHPMPAinwhichthe3-hydroxymethylgroupiselim-
inated. While, due to the presence of the hydroxyl group,
HPMPA may, at least theoretically, be incorporated into the
DNA chain via an internucleotide linkage, PMEA must act
obligatorilyasachainterminator(60).Thisnotionalsoholds
foranumberofcloselyrelatedanalogssuchas(S)-9-(3-fluoro-
2-phosphonylmethoxypropyl)adenine (FPMPA), (R)-9-(2-phos-
phonylmethoxypropyl)adenine (PMPA), and the 2,6-diamin-
opurine counterparts of PMEA and PMPA (Fig. 6). In
attemptstoimprovetheoralbioavailabilityofPMEA,several
lipophilic ester prodrugs were developed (168), and of these FIG. 8. Dideoxynucleosideanalogs.AZT,zidovudine,3(cid:57)-azido-2(cid:57),3(cid:57)-dideoxy-
prodrugs, bis(pivaloyloxymethyl)-PMEA [bis(POM)-PMEA] thymidine; ddI, didanosine, 2(cid:57),3(cid:57)-dideoxyinosine; ddC, zalcitabine, 2(cid:57),3(cid:57)-dide-
(Fig.6)hasprovedtobeanacceptableoraldrugformforthe oxycytidine; d4T, stavudine, 2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxythymidine; 3TC,
treatmentofretrovirusinfections(131). lamivudine,((cid:50))3(cid:57)-thia-2(cid:57),3(cid:57)-dideoxycytidine;((cid:50))FTC,((cid:50))3(cid:57)-thia-2(cid:57),3(cid:57)-dideoxy-
5-fluorocytidine;FddaraA,2(cid:57)-fluoro-ara-2(cid:57),3(cid:57)-dideoxyadenosine;CBV,carbovir,
TheantiviralactivityspectrumofPMEAencompassesboth carbocyclic 2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxyguanosine; 1592U89, 6-N-cyclopropyl
retrovirusesandhepadnaviruses,aswellasherpesviruses(Ta- derivativeofcarbovir.
682 DE CLERCQ CLIN.MICROBIOL.REV.
D
o
w
n
lo
a
d
e
d

f
r
o
m

h
t
t
p
:
/
/
c
m
r
.
a
s
m
.
FIG. 9. NNRTIs.4,5,6,7-Tetrahydroimidazo[4,5,1-jk][1,4]-benzodiazepin-2(1H)-one(TIBO)derivativesR82150,R82913,andR86183;1-(2-hydroxyethoxymethyl)- o
r
6-(phenylthio)thymine (HEPT) and its derivative I-EBU [5-isopropyl-1-(ethoxymethyl)-6-benzyluracil]; nevirapine (BI-RG-587); pyridinone L-697,661; bis(het- g
eroaryl)piperazine(BHAP)U-88204andU-90152(Delavirdine);(cid:97)-Anilinophenylacetamide((cid:97)-APA)R18893andR89439(Loviride);quinoxalinesS-2720andHBY /
097;OxathiincarboxanilideUC-84andthethiocarboxanilidesUC-10,UC-781andUC-82;2(cid:57),5(cid:57)-bis-O-(tert)-butyldimethylsilyl-3(cid:57)-spiro-5(cid:48)-(4(cid:48)-amino-1(cid:48),2(cid:48)-oxathiole- o
2(cid:48),2(cid:48)-dioxide)(TSAO)derivativeof3-methylthymine(TSAO-m3T). n

M
a
r
c
h
ments, PMPA and PMPDAP emerged as more potent and eitheronestep,catalyzedbyPRPP(5-phosphoribosyl1-pyro-
2
moreselectiveantiretrovirusagentsthanPMEAorPMEDAP phosphate)synthetase(11),ortwoconsecutivesteps,catalyzed 9
(22,179). byAMPkinase(126).Inhumanlymphoidcells,PMEAwould ,
2
PMEA has proved efficacious in the preexposure prophy- be anabolized to PMEApp mainly through (mitochondrial) 0
laxis of SIV infection in macaques. When PMEA was admin- AMP kinase (150). In their diphosphorylated form, PMEA, 1
9
isteredsubcutaneouslyat20mg/kg/dayfor28daysstarting48h PMPA, and their congeners act as alternative substrates to
b
before SIV inoculation, it protected 83% of the macaques dATP/chainterminatorsintheHIVreversetranscriptase(RT) y

from acute SIV infection (177); this compared favorably with reaction(17,18).Also,theinhibitoryeffectsofPMEA,PMPA, g
u
theminimaleffectseenwithzidovudine(AZT)(atadailydose and their congeners on HBV replication may be attributed
e
of 100 mg/kg for 4 weeks), where only 6% of the macaques to chain termination in the HBV-associated RT reaction. s
t
wereprotectedfrominfection(176).WithPMEA,sometoxic PMEApp and PMPApp apparently possess a greater affinity
sideeffectsintheformofmildskinlesionswereobserved,and for HIV (and HBV) RT than for the cellular DNA poly-
with AZT, there were signs of hematologic toxicity. If, how- merases(cid:97),(cid:98),(cid:103),(cid:100),ande (44,109),andthismayexplaintheir
ever,PMPAwasinvestigatedundersimilarexperimentalcon- selectivityforretrovirusesandhepadnaviruses.
ditions(subcutaneousinjectionofadailydoseof30mg/kgfor Fromaclinicalviewpoint(Table4),PMEA,initsoralpro-
4weeks,startingeither48hbefore,4hafter,or24haftervirus drugfrombis(POM)-PMEA,hasprovedeffectiveinreducing
inoculation),itcompletelypreventedtheestablishmentofSIV the viral load in HIV-infected individuals (35). Bis(POM)-
infectionin100%ofthemacaqueswithoutanysignoftoxicity PMEAhasalsobeenthesubjectofclinicalstudiesinpatients
(178). More recent data indicate that PMPA (given at 30 withHBVinfectionandhasrecentlyenteredamulticentered
mg/kg/day for 4 weeks) is also effective in the treatment of clinical phase II/III trial for the treatment of HIV infection.
established SIV infection (2 to 3 log unit reduction in the Given the potential of PMEA to inhibit herpesviruses (e.g.,
10
viral load in plasma) (40), and in preventing intravaginal SIV CMV), HIV-infected individuals treated with bis(POM)-
transmission(ifappliedtopicallyasa10%PMPAgel)(127). PMEAwillbemonitoredforHIVaswellasCMVloadreduc-
To exert their antiviral action, PMEA, PMPA, and their tions.
congeners must be phosphorylated to their diphosphorylated PMPAhasonlyrecentlybecomethesubjectofclinicaltrials,
form (PMEApp, PMPApp, etc.) (Fig. 7). This could occur in but in view of its pronounced therapeutic and prophylactic
VOL.10,1997 SELECTIVE ANTIVIRAL CHEMOTHERAPY 683
TABLE 5. AntiviralactivityspectrumofNNRTIsa
Activeagainstb:
Mutant
NNRTI
Wildtype
(HIV-1III ) HIV-1 HIV-1 HIV-1 HIV-1 HIV-1 HIV-1 HIV-1
B
L100I K103N V106A E138K Y181C Y181I Y188H
TIBOR82913 Yes No No No Yes No No No
I-EBU(MKC-442) Yes Yes No Yes Yes Yes No No
Nevirapine Yes Yes No No Yes No No No
Pyridinone(L-697,661) Yes Yes No Yes Yes No No No
BHAPU-88204 Yes No No Yes Yes Yes No Yes
BHAPU-90152(Delavirdine) Yes No Yes Yes Yes Yes No Yes
TSAO-m3T Yes Yes Yes No No No No No
QuinoxalineS-2720 Yes Yes Yes Yes Yes Yes No Yes
ThiocarboxanilideUC-10 Yes Yes No Yes Yes Yes No Yes
QuinoxalineHBY097 Yes Yes Yes Yes Yes Yes Yes Yes D
ThiocarboxanilideUC-781 Yes Yes Yes Yes Yes Yes Yes Yes o
w
aDatafromreferences27,30,32,34a,and72. n
bYes,EC50(cid:35)1(cid:109)M;No,EC50(cid:46)1(cid:109)M. lo
a
d
e
efficacyagainstSIVinfectioninmonkeys,itshouldbefurther vicinity of the substrate-binding site. NNRTIs can be readily d

investigatedforitspotential(i)topreventHIVinfection(fol- differentiated from the nucleoside type of RT inhibitors in fr
o
lowing needlestick contamination, sexual transmission, or that they inhibit HIV-1 but not HIV-2 replication. The first
m
mother-to-child transmission) and (ii) to suppress HIV infec- compounds that were found to behave in this way were the
h
tion in chronically infected patients. In view of its marked 1-[(2-hydroxyethoxy)methyl]-6-(phenylthio)thymine (HEPT) t
t
activity against HBV (96), PMPA should also be investigated derivatives (6, 128) and the tetrahydroimidazo[4,5,1-jk][1,4]- p
:
foritspotentialintheprophylaxisortherapyofHBVinfection. benzodiazepin-2(1H)-oneand-thione(TIBO)derivatives(53, //
c
PMEAandPMPAshouldbeconsideredacyclicnucleotide 140).Thesecompoundswerethenshowntoexertthisunique m
analogs,i.e.,acyclicnucleosidestowhichaphosphonategroup selectivity for HIV-1 through a specific interaction with the r
.
hasbeenattached(viatheunusualP-Clinkage).Inthissense, HIV-1RTthat,unliketheinteractionofthenucleosidetypeof a
s
PMEA and PMPA circumvent the first phosphorylation step RT inhibitors, appeared to be noncompetitive with regard to m
that is needed for and often represents the bottleneck in, the thenormalsubstrates(7,8,53). .
o
intracellular metabolism of the nucleoside analogs to their SincethedescriptionsofHEPTandTIBO,literallydozens r
5(cid:57)-triphosphateform. of different classes of NNRTIs have been described, and all g/

In the search for an effective chemotherapy of HIV infec- thesecompoundswerefoundtoinhibitHIV-1butnotHIV-2 o
n
tions, initial attempts were focused on the development of
replicationinvitroatconcentrationswellbelowthecytotoxic-
dideoxynucleosides that in their 5(cid:57)-triphosphate form would M
ity threshold: dipyridodiazepinones (e.g., nevirapine) (125),
a
inhibit the virus-associated RT reaction by terminating DNA
pyridinones (e.g., L-697,661) (92), bis(heteroaryl)piperazines r
chain elongation (61). These efforts yielded a number of se- (e.g., BHAP U-88204) (151), 2(cid:57),5(cid:57)-bis-O-(tert-butyldimethylsi- ch
lective HIV inhibitors (65, 66) that have now been approved lyl)-3(cid:57)-spiro-5(cid:48)-(4(cid:48)-amino-1(cid:48),2(cid:48)-oxathiole-2(cid:48),2(cid:48)-dioxide)pyrimi- 2
for the treatment of HIV infections (AIDS), i.e., zidovudine dine(TSAO)derivatives(e.g.,TSAO-m3T)(20),(cid:97)-anilinophe- 9,
(AZT) (115), didanosine (ddI) (87), zalcitabine (ddC) (187), nylacetamides ((cid:97)-APA) (141), quinoxalines (e.g., S-2720) 2
stavudine(d4T)(117),andlamivudine(3TC)(Fig.8).Someof 0
(108), and PETT derivatives (2). Meanwhile, the original 1
thesedideoxynucleosideanalogues,particularly3TC,are,like
HEPTandTIBOparentcompoundshadgivenrisetonewand 9
PMEAandPMPA,alsoactiveagainstHBV,and3TCisbeing
more active congeners, i.e., I-EBU (MKC-442) (9) and 8- b
furthertestedforthetreatmentofHBVinfections. y
chloro-TIBO R86183 (Tivirapine) (142). The different variet-
Other dideoxynucleoside analogs such as FddaraA, g
1592U89,and((cid:50))FTC(Fig.8)arestillunderdevelopmentas i9esdeopfiNctNstRhTosIesNhaNveRTbeIesnthraetvhieawveedbreeecnenmtolyst(6in8t,e6n9s)i,vealnydstFuidg-. ue
candidateanti-HIVdrugs.Thegoaloftheseefforts(66,67)has s
ied. t
been to increase the safety (activity/toxicity) profile of the
One of the compounds shown in Fig. 9, namely, oxathiin
anti-HIVagents,therebyprovidingagreaterchoiceofpoten-
carboxanilide UC-84 (NSC 615985), was initially not recog-
tial candidates for drug combinations. Indeed, it has become
nizedasanNNRTI:itwas,infact,saidnottoinhibittheRT
increasinglyclearthattoachievethegreatestbenefitinterms
underconditionswhereTIBOeffectivelydidso(10).Fromthe
ofreductionofviralload,increaseinCD4cellcounts,delayof
resistancedevelopment,andclinicaloutcome,thecompounds original UC-84, several carboxanilide and thiocarboxanilide
shouldbecombinedinmultiple-drugregimens. derivatives(UC-10,UC-781,andUC-82)havebeenprepared
thatarenotonlymuchmoreactivethantheparentcompound
(and active against the RT) but also effective against mutant
SELECTIVEINHIBITORSOFHIV-1
HIV-1strainsresistanttootherNNRTIs(28,29,31).
While the dideoxynucleosides in their 5(cid:57)-triphosphate form NNRTIshavebecomenotoriousfortheirpropensitytorap-
and the acyclic nucleoside phosphonates, PMEA and PMPA, idly elicit resistance due to mutations in the HIV-1 RT (64).
in their diphosphorylated form interact as competitive inhibi- Thepositionsthataremostpronetomutationsare100(Leu3
tors or alternative substrates for the substrate-binding site of Ile),103(Lys3Asn),106(Val3Ala),138(Glu3Lys),181
theHIVRT,nonnucleosideRTinhibitors(NNRTIs)interact (Tyr 3 Cys 3 Ile), 188 (Tyr 3 Cys/His), 190 (Gly 3 Glu/
with a non-substrate-binding site that is located in the close Ala/Gln), and 236 (Pro 3 Leu) (68). While these mutations