Antiviral Therapy for Human Immunodeficiency Virus Infections

40 Pages·1995·0.85 MB·English

CLINICALMICROBIOLOGYREVIEWS,Apr.1995,p.200–239 Vol.8,No.2
0893-8512/95/$04.00(cid:49)0
Copyright(cid:113)1995,AmericanSocietyforMicrobiology
Antiviral Therapy for Human Immunodeficiency
Virus Infections
ERIKDECLERCQ*
RegaInstituteforMedicalResearch,KatholiekeUniversiteitLeuven,B-3000Leuven,Belgium
INTRODUCTION.......................................................................................................................................................200
ANTI-HIVAGENTS...................................................................................................................................................201
VirusAdsorptionInhibitors..................................................................................................................................201
Virus-CellFusionInhibitors.................................................................................................................................207
VirusUncoatingInhibitors....................................................................................................................................207
ReverseTranscriptionInhibitors.........................................................................................................................209
Substrateanalogs................................................................................................................................................209
Nonsubstrateanalogs.........................................................................................................................................210
MiscellaneousRTinhibitors.............................................................................................................................215
IntegrationInhibitors.............................................................................................................................................216
DNAReplicationInhibitors...................................................................................................................................216
TranscriptionInhibitors........................................................................................................................................216
TranslationInhibitors............................................................................................................................................217
MaturationInhibitors............................................................................................................................................218
Proteaseinhibitors..............................................................................................................................................218
Myristoylationinhibitors...................................................................................................................................220
Glycosylationinhibitors.....................................................................................................................................220
Budding(Assembly/Release)Inhibitors...............................................................................................................222
COMBINATIONTHERAPY.....................................................................................................................................223
VIRUS-DRUGRESISTANCE...................................................................................................................................224
CONCLUSION............................................................................................................................................................225
ACKNOWLEDGMENTS...........................................................................................................................................226
REFERENCES............................................................................................................................................................226
INTRODUCTION thestageatwhichtheyinterferewiththeHIVreplicativecycle
(Table1).
Numerous compounds have been reported to inhibit the However, not all substances to which anti-HIV activity has
replication of human immunodeficiency virus (HIV) in vitro been attributed easily fit within the proposed scheme (Fig. 1;
(118,410),yetonlyfouragentshaveatthistimebeenformally Table1).Forexample,somerecombinant(chimeric)proteins
licensed(intheUnitedStates)forclinicaluseinthetreatment inwhichatoxin,Pseudomonasaeruginosatoxin(7,16,17,65)
of AIDS. These are zidovudine (3(cid:57)-azido-2(cid:57),3(cid:57)-dideoxythymi- ordiphtheriatoxin(18),hasbeenlinkedtotheHIVenvelope
dine or azidothymidine [AZT]; Retrovir) (269), didanosine glycoprotein (gp120)-binding domain of human CD4 have
(2(cid:57),3(cid:57)-dideoxyinosine [ddI]; Videx) (156), zalcitabine (2(cid:57),3(cid:57)- been described: by virtue of their affinity for gp120, these
dideoxycytidine[ddC];Hivid)(479),andstavudine(2(cid:57),3(cid:57)-dide- hybrid toxins selectively bind to and kill HIV-infected cells.
hydro-2(cid:57),3(cid:57)-dideoxythymidine [D4T]; Zerit). The basic strate- AlthoughbothacutelyandchronicallyHIV-infectedcellscan
gies and molecular targets for anti-HIV therapy have been be selectively killed by this gp120-targeted cytotoxicity ap-
repeatedlyreviewedstartingfrom1985,thusshortlyafterHIV proach, it does not prevent the emergence of HIV-infected
hadbeenidentifiedasthecausativeagentofAIDS(116–121, cells that are resistant to the chimeric toxins (18). Also, gene
127, 324, 327). More recent reviews have addressed the chal- therapy approaches have been advocated to introduce the
lenges and prospects for the therapy of HIV infection (236, diphtheria toxin gene directly to HIV-infected cells (198),
490). which should ultimately result in the eradication of the cells
The replicative cycle of HIV comprises a number of steps whenthediphtheriatoxingeneisexpressed.
thatcouldbeconsideredadequatetargetsforchemotherapeu- Anotherapproachthatcouldnotbereadilyaccommodated
tic intervention (Fig. 1). In fact, HIV follows a replicative bytheproposedscheme(Fig.1;Table1)isthatbasedonthe
pathway that is similar to that of the classical cytolytic RNA targeting of antiviral agents (i.e., pokeweed antiviral protein)
viruses, except for reverse transcription (step 4) and integra- to CD4(cid:49) cells (whether infected or not) by conjugation of
tion (step 5), which lead to the formation and integration of theseantiviralagentswithmonoclonalantibodiesreactivewith
theproviralDNAintothecellularDNAgenome.Mostofthe normal antigens on CD4(cid:49) cells. Such conjugates have been
substancesthathavebeenidentifiedasanti-HIVagentscanbe showntoinhibitHIVtype1(HIV-1)replicationinCD4(cid:49)cells
assignedtooneofthe10classesofHIVinhibitorsaccordingto andweresurmisedtoinhibitthereplicationofothervirusesas
well(152,497).
Also, various other compounds that have been reported to
*Mailingaddress:RegaInstituteforMedicalResearch,K.U.Leu- inhibit HIV replication cannot be unequivocally allocated to
ven,Minderbroedersstraat10,B-3000Leuven,Belgium.Phone:32-16- one of the 10 classes of HIV inhibitors (Table 1; Fig. 1),
33.73.41.Fax:[phone]. primarilybecausetheirtargetofactionhasnotbeenelucidated
200
VOL.8,1995 ANTIVIRAL THERAPY FOR HIV INFECTIONS 201
FIG. 1. EssentialstepsintheHIVreplicativecycle:1,adsorption;2,fusion;3,uncoating;4,reversetranscription;5,integration;6,DNAreplication;7,transcription;
8,translation;9,maturation;10,budding(assembly/release).
or does not fall within the proposed scheme. To the more in HIV-infected pregnant women for the prevention of HIV
recent group of HIV inhibitors, for which the mechanism of infection of the fetus, since CD4-immunoglobulin G, like the
anti-HIVactionneedstobeelucidated,belongdiphenylhydan- parentimmunoglobulinGmolecule,efficientlycrossesthepla-
toin(97),ascorbicacid(194),pradimycin(444),oxophenarsine centa. Yet, there are several problems linked to the use of
(188), fluoroquinolones (354), prostaglandins (10), glutathi- CD4-based therapeutics, in particular, the fact that much
one, glutathione ester, N-acetylcysteine (241), ((cid:50))-gossypol higher concentrations of CD4 are needed to inhibit primary
(282) and various analogs of gossypol (284), and the HIV-1 HIV-1 isolates than laboratory strains of HIV-1 (109), for
inhibitorsproducedbymyxobacteria(237)orinducedbyPinus reasonsthatstillhavetobeclarified(15).Also,cell-associated
parvifloraextracts(440).OtherHIVinhibitorssuchastheC60 virus may be less easily inhibited by CD4 derivatives than
fullerene derivatives seem to interact at multiple steps of the cell-freevirus.
virallifecycle,i.e.,directvirusinactivationaswellasinhibition AsCD4isnotonlythereceptorforHIVbutalsotherecep-
oftheHIVreversetranscriptase(RT)andHIVprotease(166, torforclassIImajorhistocompatibilitycomplexantigens,sol-
411, 423). Until the modes (targets) of action of these com- ubleformsofCD4mayalsointerferewithimmuneprocesses
poundsarebetterdelineated,itwouldseemdifficulttoassess
involving the class II major histocompatibility complex pro-
theirpositionorpotentialforthetreatmentofHIVinfections.
teins,andinaddition,theCD4derivativesmayhavedelivery,
stability,andexpenseproblems.Thesmallerthepeptides,the
ANTI-HIVAGENTS smaller these problems may turn out to be, and in this per-
spectivetheN-carbomethoxycarbonyl-prolyl-phenylalanylben-
VirusAdsorptionInhibitors zyl esters (CPFs) were conceived (160). These compounds
Since the CD4 molecule on helper T4 lymphcoytes and interactdirectlywiththeviralglycoproteingp120,blockbind-
monocytes/macrophages is the principal receptor for the ingoftheHIVtotheCD4receptor,donotinterferewiththe
HIV-1 envelope glycoprotein gp120, various forms of recom- binding of CD4 to class II major histocompatibility complex
binant soluble CD4 (rsCD4), including truncated CD4 mole- proteins,andpreventthespreadofHIVfromasmallnumber
cules(i.e.,CD4[segment74-95]orCD4[segment81-92]pep- ofafflictedcellstoalargerpopulationofuninfectedcells(160).
tides [384, 418] and benzylated or phenylalanine-substituted The questions of how the CPFs perform in vivo and whether
derivativesthereof[275])aswellasCD4-immunoglobulincon- they indeed block dissemination of HIV-1 in vivo have so far
jugates (i.e., CD4 immunoadhesins [86, 268]) and CD4-albu- remained unanswered. Given their poor aqueous solubility,
minconstructs(491),havebeencreatedwiththeaimofblock- thesecompoundsmightalsohavebioavailabilityproblems.
ing HIV-1 binding (adsorption) to the cells. The chimeric In addition to the CPFs, several other, miscellaneous com-
forms (CD4 immunoadhesins and CD4-albumin constructs) poundshavebeenpostulatedtoinhibitHIVinfectionthrough
were obviously made to increase the plasma half-life of the aninteractionwiththeviralglycoproteingp120,thusblocking
otherwise short-lived CD4. The CD4 immunoadhesin (CD4- the binding of gp120 to the CD4 receptor: pyridoxal 5(cid:57)-phos-
immunoglobulinG)didnotoffermuchprotectionagainstsim- phate (187a), Prunella vulgaris extract (488), tannins (474),
ian immunodeficiency virus infection in macaques (268) but caffeoylquinic acid derivatives (294), flavans (i.e., daphnodor-
provedcapableofpreventingHIV-1infectioninchimpanzees ins [495a]), and flavanoids (i.e., ((cid:50))epicatechin 3-O-gallate]
(471),andthisoffershopefortheuseofCD4-immunoglobulin (295). In contrast with the sulfated polysaccharides (i.e., dex-
202 DE CLERCQ CLIN.MICROBIOL.REV.
TABLE 1. ReviewofHIVinhibitorsaccordingtostageofinterventionwiththeHIVreplicativecycle
StageofHIVintervention HIVinhibitor
Adsorption rsCD4constructs(CD4fragments,CD4immunoadhesins,andCD4-albuminconstructs)
CPFs(N-carbomethoxycarbonyl-prolyl-phenylalanylbenzylesters),tannins,andflavanoids[((cid:50))epicatechin-3-O-
gallate]
Polysulfates(heparin,dextransulfate,dextrinsulfate,curdlansulfate,pentosanpolysulfate,mannansulfate,
sulfoevernan,fucoidan,polyvinylalcoholsulfate,polyacetalpolysulfate,O-acylatedheparin,cyclodextrin
sulfate,andmodifiedcyclodextrinsulfates)
Polysulfonates[suramin,Evansblue,bis(naphthalenedisulfonate)derivatives,polyvinylsulfonate,polystyrene
sulfonate]
Polycarboxylates(ATA),polyhydroxycarboxylates(phenyl-derivedpolyhydroxycarboxylates),and
polyfluoroalkylcarboxylates(MAA-HFPO5)
Polyoxometalates{H SiW O (JM1493),K[PTiW O ](cid:122)6HO[PM-19],K [Ce(SiW O )](cid:122)26HO
4 12 40 7 2 10 40 2 13 11 39 2 2
[JM1590],and[MeNH][SiNbW O ](JM2820)}
3 8 2 6 18 77
Fusion Plantlectins(fromListeraovata,Hippeastrumhybrid,Cymbidiumhybrid,Epipactishelleborine,andUrticadioica)
PeptideT22[(Tyr-5,12,Lys-7)polyphemusinII]
SuccinylatedandaconitylatedHSA
BetulinicacidRPR103611
Uncoating Bicyclams(JM2763andJM3100)
Reversetranscription Substrateanalogs
2(cid:57),3(cid:57)-Dideoxynucleosideanalogs(zidovudine[AZT],didanosine[ddI],zalcitabine[ddC],stavudine[D4T],
lamivudine[3TC],FTC,andFddClUrd)
Acyclicnucleosidephosphonates(PMEA,FPMPA,PMPA,andPMPDAP)
Nonsubstrateanalogs(NNRTIs:TIBO[R82150,R82913,andR86183],HEPT[E-EPU,E-EBU-dM,andI-
EBU],nevirapine[BI-RG-587],pyridinone[L-696,229andL-697,661],BHAP[U-88204andU90152],TSAO,
(cid:97)-APA,andPETT)
MiscellaneousRTinhibitors,includingantisenseoligonucleotides
Integration Antisenseconstructs
DNAreplication Antisenseconstructs
Transcription AntisenseODNs
Tatantagonists(benzodiazepines[Ro5-3335andRo24-7429]and3-keto/enol-4,5-epoxysteroids)
LTR-directedgeneexpressioninhibitors(topotecan)andPKCinhibitors(indolocarbazoles)
Translation AntisenseODNs(phosphorothioates,phosphorodithioates,andmethylphosphonates)
Ribozymes(hammerheadandhairpinribozymes)thatcanbedeliveredexogenouslyorendogenouslyvia
retroviralvectors)
Trichosanthin(?)
Maturation Proteaseinhibitors:transition-statepeptidomimetics(Ro31-8959,U-81749,A-77003,andKNI-227),and
nonpeptidecyclicureas(XM323)
Myristoylationinhibitors(12-azidododecanoicacid)
Glycosylationinhibitors(NBuDNJanditsprodrug[N-butyldeoxynojirimycin-6-phosphate])
Budding(assembly/release) IFN(alsointerfereswithotherstages)
Hypericin(?)
Cyclosporineanalogs(SDZNIM811)(alsointerferewithtransportofviralDNAintothenucleus)
transulfate),whoseactionisreversible,theflavanoidsirrevers- polysulfate,polyvinylalcoholsulfate,andmodifiedcyclodextrin
ibly inactivate virus infectivity (295). Also, some of the fla- sulfates)(Fig.2)havebeenfoundtoinhibitHIVreplicationin
vanoids have been shown to inhibit the RT of certain vitroat45concentrationsthatareupto10,000-foldlowerthan
retroviruses(includingHIV),butthiseffectwouldnotcontrib- the cytotoxic concentration (124). These compounds are tar-
ute to their anti-HIV action observed in cell culture. Other geted at the interaction between the viral envelope glycopro-
compoundsthathavebeenpostulatedtointerferewithseveral teingp120andtheCD4receptor,andasaconsequence,they
stepsoftheHIVreplicativecycle,i.e.,cosalane(disalicylmeth- inhibit not only virus adsorption to the cells but also virus-
ane linked to cholestane [106a]) and GTOs (oligonucleotides induced syncytium (giant cell) formation (29). The inhibitory
composed entirely of guanosine and thymidine [356a]), may effects of dextran sulfate and its congeners on viral binding,
owe their anti-HIV activity primarily to inhibition of gp120- viralreplication,andsyncytiumformationappeartobemedi-
CD4binding. atedbyaspecificinteractionwiththeV3regionofgp120(64,
Varioussulfatedpolysaccharides(e.g.,heparin,dextransul- 82). In addition, sulfated polysaccharides may also directly
fate,dextrinsulfate,cyclodextrinsulfate,curdlansulfate,pen- interfere with the binding of HIV particles to the heparan
tosanpolysulfate,mannansulfate,sulfoevernan,andfucoidan) sulfate proteoglycan at the cell surface, whether or not this
and derivatives thereof (e.g., O-acylated heparin, polyacetal process occurs independently of, or cooperatively with, the
VOL.8,1995 ANTIVIRAL THERAPY FOR HIV INFECTIONS 203
FIG. 2. Structuresofpolysulfates.(A)Dextransulfate[sulfated(136)-(cid:97)-
D-glucan],dextrinsulfate[sulfated(134)-(cid:97)-D-glucan],curdlansulfate[sulfated
(133)-(cid:98)-D-glucan],pentosanpolysulfate[sulfated(134)-(cid:98)-D-xylan],mannan
sulfate[sulfated(134)-(cid:97)-D-mannan],sulfoevernan{sulfated(133)[80%],(1
3 4) [20%]-(cid:97)-D-glucan}, fucoidan [composed of sulfated (1 3 2)-linked L-
fucoseunits],PAPS(polyacetalpolysulfatepreparedfromdextran),andPVAS
(polyvinyl alcohol sulfate). (B) Heparin [composed of L-iduronic acid or D-
glucuronicacid(134)linkedtoD-glucosamine],O-acylated(butyrylatedor
hexanoylated)heparin,supersulfateddermatansulfate[chondroitinsulfateB;
consistsofL-iduronicacid(133)linkedtoD-(N-acetyl)galactosamine],PAVAS
[poly(acrylicacidvinylalcoholsulfate)copolymer],andsulfated(cid:98)-cyclodextrin
[cyclicdextrinconsistingofseven(134)-linked(cid:97)-D-glucans]andderivatives
thereof(mCDS11andmCDS71[containing6-benzylthio-6-deoxyor2-O-benzyl
substituents,respectively]).
204 DE CLERCQ CLIN.MICROBIOL.REV.
viral envelope-CD4 receptor interaction (364). Yet, sulfated macokinetic properties in HIV-infected individuals (372). It
polysaccharides would be unable to block the viral gp120 in- has also been investigated, but found inactive, against HIV-
teractionwiththeCD4ofmonocytes(292a). associated Kaposi’s sarcoma (375). Since Kaposi’s sarcoma is
Among the more promising congeners of dextran sulfate characterizedbymicrovascularproliferation(angiogenesis)in
rank polyacetal polysulfate (484) and polyvinylalcohol sulfate theinitialstageoflesiondevelopment,itwouldseemjustified
(27), which show potent activity against HIV-1, HIV-2, and to study sulfated polysaccharides because of their angiostatic
several other enveloped viruses, including simian immunode- potentialagainstKaposi’ssarcoma.Perhapspentosanpolysul-
ficiency virus, herpes simplex virus (HSV), cytomegalovirus fatewasnotthebestchoice,andothersulfatedpolysaccharides
(CMV), influenza A virus, and respiratory syncytial virus, as suchasthesulfatedpolysaccharide-peptidoglycanproducedby
wellastoga-,flavi-,arena-,bunya-,andrhabdoviruses(8,124, Arthrobactersp.(343)mightbemoreefficaciousagainstKapo-
215, 414). Thus, the spectrum of activity of the polysulfates si’ssarcoma.
extends to various viruses other than HIV that may occur as Inthewakeofanysolidevidencefortheinvivoefficacyof
opportunisticpathogensinimmunosuppressed(i.e.,AIDS)pa- the polysulfates against HIV or other viral infection, one
tients. should consider their potential application in the (systemic)
Ofadditionalimportanceisthefactthatthepolysulfatescan prophylaxis of HIV infection following an accidental needle
be obtained from natural sources (i.e., marine invertebrates) stick injury or stab wound, i.e., conditions in which AZT has
(66).Theycanbepreparedandmadeavailableinlargequan- provedinefficacious,and/ortopicalprophylaxisofHSVorHIV
titiesatreasonablecost.Theycanactsynergisticallywithother infectioncontractedthroughsexualintercourse.
anti-HIVdrugs(i.e.,AZT,ddI,andddC)(415).Theyarenot The principles guiding the anti-HIV activity of polysulfates
knowntoleadtothedevelopmentofvirus-drugresistance,and are also applicable to the polysulfonates. Several polysulfon-
theyshouldbeeffectiveagainstHIVmutantsthatareresistant ates of varying molecular weights and degrees of sulfonation
toAZTorotherRTinhibitors(461). have been described as potent anti-HIV agents (Fig. 3): e.g.,
However,polysulfates(suchasdextransulfate)sufferfroma naphthalene sulfonates (330, 330a, 332–334) {i.e., 4,4(cid:57)-[1,6-
number of drawbacks which seem to argue against their po- hexanediylbis(carbonylamino)]bis(5-hydroxy-2,7-naphthalene-
tentialusefulnessinvivo.Theyarepoorlyabsorbedafteroral disulfonic acid) (335)}, stilbene sulfonates (87), Evans blue
administration, as noted in humans (2, 288), rats (200), and andvariousothersulfonateddyes(47,96,263,363,473),poly-
mice(256).However,highoralbioavailabilitycanbeobtained styrenesulfonate,polyanetholesulfonate,andpolyvinylsulfon-
by the appropriate chemical modifications, as shown for the ate(331).Thesecompoundswouldbindprimarilytotheviral
modified (cid:98)-cyclodextrin sulfates (mCDS11 and mCDS71) envelope gp120 glycoprotein (28) and thus interfere with the
(338,339,359).Dextransulfate,uponintravenousadministra- interaction between the viral gp120 glycoprotein and the cel-
tion,producesthrombocytopenia(164).Sulfatedpolymersare lular CD4 receptor and block virus adsorption and virus-in-
alsonotoriousfortheiranticoagulantactivity,butashasbeen ducedsyncytiumformation.Likethepolysulfates,thepolysul-
demonstratedwithperiodate-treatedheparin(19)andO-acy- fonatesinhibitednotonlythereplicationofHIVbutalsothat
latedheparin(63),thisproblemcanbeovercomebyappropri- ofotherenvelopedviruses,i.e.,CMV(24).
atechemicalmodifications. In fact, the prototype of the polysulfonates, suramin (117),
The sulfated polymers owe their anti-HIV activity to the wasthefirstcompoundtoberecognizedasananti-HIVagent
presence of the sulfate groups, which in turn are responsible (325)andalsothefirsttobeusedintheclinicforthetreatment
fortheinhibitionofvirus-cellbinding.Inthissense,anycom- ofAIDS(74).Itwasoriginallyassumedthatsuramin,aswellas
pound could be turned into an anti-HIV agent targeted at Evans blue (47), inhibits the replication of HIV through an
virus-cell binding provided it contains the necessary hydroxyl inhibitoryeffectontheviralRT.Hence,initialstructure-func-
groupsforattachmentofthesulfategroups,andthusvarious tionrelationshipstudieswerebasedontheinhibitoryeffectof
compounds, i.e., glycyrrhizin, lentinan, amphotericin B, and suramin and its congeners on the viral RT (230). It has now
gangliosides (191, 204, 347, 358, 453), were found to gain become evident, however, that the polysulfonates also inter-
anti-HIVactivityfollowingsulfation. ferewiththeviraladsorptionprocess(412).Inhibitionofvirus-
Giventheirwidelyvaryingmolecularweightsanddegreesof cell binding may well be their principal mode of anti-HIV
sulfation,itisverydifficulttoobtainstandardizedpreparations action,sinceasarule,inhibitionoftheRTdoesnotcorrelate
of dextran sulfate or other sulfated polymers. This lack of withinhibitionofHIVreplicationinthevirus-cellassay,prob-
homogeneity,togetherwiththeinherentvariabilityofthemo- ably due to the lack of cellular entry of the polysulfonates
leculartarget(V3loopofgp120)withwhichthesulfatedpoly- (330a).
mersinteract,mayaccountforthedifferencesinsusceptibility Suramin may interfere with a number of processes, i.e.,
ofdifferentHIVstrainstodifferentpolysulfates(79,416).This proteinkinaseC(PKC)-mediatedprocesses(296),involvedin
differential virus-drug susceptibility obviously raises questions virus infectivity. Furthermore, suramin and other polysulfon-
astotheinvivoefficacythatmaybeexpectedforthepolysul- ates(i.e.,sulfonateddistamycinAderivatives)(95)areknown
fatesineachparticularHIVinfection. to block basic fibroblastic growth factor and other factors in-
There is little, if any, evidence for the in vivo efficacy of volvedintumorangiogenesisandshouldthereforebepursued
sulfated polysaccharides against HIV infection or any other for their antitumor potential, i.e., against Kaposi’s sarcoma.
viralinfection.Dextransulfatedidnotproveefficaciousagainst Also,suraminisnotoriousforitsstickinesstoplasmaproteins,
felineleukemiavirusinfectionincats(299)orduckhepatitisB i.e.,albumin(70);thus,albuminreversestheabilityofsuramin
virus (HBV) infection in ducklings (356). On the other hand, to block the CD4-gp120 interaction, thereby attenuating its
sulfoevernan was reported to completely suppress Rauscher anti-HIV activity (489). Although the high affinity of suramin
leukemiavirusinfectioninmiceifadministeredatadoseof20 for plasma proteins, a propensity it undoubtedly shares with
mg/kg/day for 8 days, starting 1 day after infection (477). other polysulfonates, is likely to affect the in vivo efficacy of
Equallyimpressivehavebeentheprotectiveeffectsofdextran these compounds, suramin has proved to be effective in sup-
sulfate and, recently, pentosan polysulfate (139) in mice in- pressing retrovirus (Rauscher leukemia virus) replication in
fectedwiththeunconventionalscrapieagent. mice(398).Nowthatsomanymorepolysulfonateshavebeen
Pentosanpolysulfatehasbeenfurtherpursuedforitsphar- shown to be antivirally active, not only against HIV but also
VOL.8,1995 ANTIVIRAL THERAPY FOR HIV INFECTIONS 205
FIG. 4. Structuresofpolycarboxylates:aurintricarboxylicacid(ATA),phe-
nol-derivedpolyhydroxycarboxylates[KOP(fromcaffeicacid),HYKOP(from
hydrocaffeicacid),andGENOP(fromgentisinicacid)],andpolyfluoroalkylcar-
boxylates [bis(perfluoro-1,4,7,10-tetramethyl-2,5,8,11-tetraoxatetradecylated)
methacrylicacidoligomer(MAA-HFPO5)].PolymericformforATA,aspro-
posedbyCushmanetal.(108).
thehigherthemolecularweight,thehigherthecapacityofthe
ATA fractions to block HIV binding to the cells, HIV repli-
cation,andHIV-inducedsyncytiumformation(107,108).
Also,polyhydroxycarboxylatesderivedfromphenolic(PDP)
compoundshavebeenfoundtoblockHIVbindingtothecells,
HIVreplication,andHIV-inducedsyncytiumformation(417).
The anti-HIV activity of the polyhydroxycarboxylates can be
ascribed to inhibition of the gp120-CD4 interaction, and this
inhibitory effect would depend essentially on the presence of
thecarboxylategroups(417).Asimilarmodeofactionmaybe
postulated for the polyfluoroalkylcarboxylates (i.e., MAA-
FIG. 3. Structures of polysulfonates: suramin, Evans blue, bis(naphtha-
HFP05), which have been recently shown to inhibit HIV-1
lenedisulfonate)derivatives,polystyrenesulfonate,andpolyvinylsulfonate.
replication, HIV-1 binding to the cells, and HIV-1-induced
syncytiumformation(23).
Asnotedaboveforthepolysulfonates,thepoly(hydroxy)car-
againstotherviruses(e.g.,CMV),itwouldseemimperativeto boxylates (i.e., ATA and PDP) were also found to inhibit the
exploretheirinvivoantiviralactivityintheappropriateanimal replicationofherpesviruses(i.e.,HSVandCMV)(353),which
models. again could be ascribed to inhibition of the viral adsorption
Akin to the polysulfonates (i.e., Evans blue), the polycar- process(353).Asforthepolysulfonates,thepoly(hydroxy)car-
boxylates (i.e., aurintricarboxylic acid [ATA]) (Fig. 4) were boxylatesneedtobefurtherexploredfortheirinvivoefficacy
originally assumed to inhibit HIV replication through inhibi- intheappropriateanimalvirusinfectionmodels.
tion of the viral RT (47). Later it was ascertained that ATA Beginning with HPA-23 ([NH ] Na[NaSb W O ](cid:122)
417 9 21 86
inhibits HIV replication primarily through a specific interac- 14H O) as the prototype (142), numerous polyoxometalates
2
tionwiththeCD4receptor(413),thuspreventingthebinding have been synthesized and found to be effective as anti-HIV
of the viral gp120 with its receptor (413, 472). In addition to agents(210,223,439,475,487).Representativeexamples(Fig.
the cellular CD4 receptor, the viral gp120 glycoprotein (V3 5) of these inorganic complexes are H SiW O (JM1493)
4 12 40
loop) may also serve as a target for the interaction of ATA (210), K [PTi W O ] (cid:122) 6H O (PM-19) (439), [NH ]
7 2 10 40 2 4 2
(351,413).DifferentfractionsofATA,withvaryingmolecular H [EU (MoO )(H O) (Mo O ) ](cid:122)13H O (PM-104) (223),
2 4 4 2 16 7 24 4 2
weights, have been prepared, and a direct correlation was K [Ce(SiW O ) ] (cid:122) 26H O (JM1590) (487), K [BGa
13 11 39 2 2 6
found between antiviral potency and molecular weight; thus, (H O)W O ](cid:122)15H O (JM2766) (487), and [Me NH]
2 11 39 2 3 8
FIG. 5. Structures of polyoxometalates: HSiW O (JM1493), [MeNH]-8[SiNbW O ] (JM2820), and K [Ce(SiW O )](cid:122)26HO (JM1590). JM1493
4 12 40 3 2 6 18 77 13 11 392 2
representsa‘‘keggin’’structure;JM2820,a‘‘doublekeggin’’structure;andJM1590,a‘‘kegginsandwich’’structure.
206
VOL.8,1995 ANTIVIRAL THERAPY FOR HIV INFECTIONS 207
[Si Nb W O ](JM2820)(487).Likealloftheotherpolyan- pursued for their therapeutic potential in the treatment of
2 6 18 77
ionic substances, polyoxometalates inhibit HIV replication, retro-,herpes-,andmyxovirusinfectionsinvivo.
HIV binding to the cells, and HIV-induced syncytium forma- ThepeptideT22([Tyr-5,12,Lys-7]polyphemusinII),aderiv-
tion. ative of polyphemusin that is highly abundant in hemocyte
Although the polyoxometalates also inhibit the viral RT, debrisofthehorseshoecrabLimuluspolyphemus,alsoqualifies
theirmechanismofanti-HIVactioncanbeattributedprimar- as an HIV-cell fusion inhibitor: it is only weakly inhibitory to
ilytoinhibitionofvirus-cellbinding.Thismodeofactionwas virus-cell binding, yet it is strongly inhibitory to syncytium
suggested by ‘‘time of addition’’ experiments, in which the formation,andfromtimeofadditionexperimentsitappearsto
polyoxometalateswereaddedatdifferenttimesaftervirusin- interactwithastageofthevirusreplicativecyclethatmaywell
fection(487).Inhibitionofvirus-cellbindingapparentlyresults correspondwithvirus-cellfusion(346).Itwouldseemmanda-
from the interaction of the polyoxometalates with the viral torytoexaminewhethertheantiviralactivityspectrumofT22
glycoproteingp120. extendstovirusesotherthanHIV(i.e.,HSV,CMV,orrespi-
In keeping with the other polyanionic substances, polyoxo- ratorysyncytialvirus)and/orwhetheritisasefficaciousinvivo,
metalatesarealsoinhibitorytovariousenvelopedviruses(oth- asitsinvitropotencytendstosuggest.
er than HIV), including herpesviruses (i.e., HSV and CMV) Anotherclassofmoleculesthatisapparentlytargetedatthe
and ortho- and paramyxoviruses (influenza A and respiratory fusionprocessisthesuccinylatedlectins(i.e.,succinylatedcon-
syncytial virus) (167, 221, 487). This broad-spectrum antiviral canavalinA[300])andsuccinylatedalbumins(whetherornot
activity adds to the therapeutic potential of the polyoxometa- thesealbuminsarealsoglycosylated[228]).Theanti-HIVac-
latesandalsojustifiestheirfurtherfollow-upintheappropri- tivityofthesuccinylatedalbuminsincreaseswiththeirnegative
ateanimalvirusinfectionmodels.Infact,thepolyoxotungstate charge;theyinhibitsyncytiumformationatconcentrationsthat
PM-19 has proved effective against HSV-2 infection in mice correspondto(orareslightlyhigherthan)theconcentrations
whengivenintraperitoneallyoveradosagerangeof0.1to50 requiredtoinhibitHIVreplication,whilevirus-cellbindingis
mg/kg/day under conditions in which acyclovir was ineffective inhibited only partially at much (100-fold) higher concentra-
atdosesofupto100mg/kg/day(222). tions (228). In addition to the succinylated human serum al-
bumins (HSA), aconitylated HSA (Fig. 6) have also been
foundtoinhibitHIVreplication(229).Aconitylatedalbumins
Virus-CellFusionInhibitors inhibit HIV-1 replication at lower concentrations than succi-
nylatedalbumins,probablybecauseinadditiontotheirinhib-
To qualify as a specific virus-cell fusion inhibitor, a given itoryeffectonvirus-cellfusion,aconitylatedalbuminsalsoin-
compound, while not inhibitory to virus-cell binding, should hibit virus-cell binding by shielding off viral gp120. Both
inhibitsyncytiumformationinthedirectsyncytiumformation succinylated and aconitylated HSA are less active against
assay. The latter test is based on the formation of giant cells HIV-2thanHIV-1,andincontrasttothesulfatedpolysaccha-
following cocultivation of uninfected CD4(cid:49) cells with HIV- rides (dextran sulfate), they are inactive against viruses other
infected cells expressing the viral glycoproteins gp120 and thanHIV.Alsoincontrasttodextransulfate,succinylatedand
gp41. This giant cell (or syncytium) formation requires the aconitylatedHSAlackanticoagulantactivity.Succinylatedand
interaction of the CD4 receptor with the viral glycoproteins. aconitylated albumins offer the potential to block HIV infec-
Direct syncytium formation should be distinguished from in- tivity in blood, plasma, and plasma products and should be
directsyncytiumformation,inwhichgiantcellsareinducedby furtherexaminedforthispurpose.
virus that has gone through its replicative cycle. The indirect A novel class of triterpene (i.e., betulinic acid) derivatives
syncytiumformationassaycannotbeusedforidentifyingcom- has been recently identified as a potent and selective HIV-1
pounds that specifically interfere with virus-cell fusion, since inhibitor (303). These betulinic acid derivatives (Fig. 7) are
inhibitionofindirectsyncytiumformationmayreflectinterfer- inactiveagainstHIV-2andapparentlytargetedatapostbind-
encewithanystepofthevirusreplicativepathway. ing, virus-cell fusion step. As some HIV-1 strains (i.e., NDK)
The mannose-specific plant lectins (i.e., from Listera ovata, are not susceptible to betulinic acid RPR 103611, the com-
Hippeastrumhybrid,Cymbidiumhybrid,andEpipactishellebo- poundmaybesurmisedtointeractwithaveryspecificmolec-
rine) and N-acetylglucosamine-specific plant lectin (i.e., from ularsite.TheprecisemodeofactionofRPR103611,aswellas
Urtica dioica) qualify as specific inhibitors of the virus-cell itspotentialtherapeuticusefulness,remainsasubjectforfur-
fusionprocess:theydonotinhibitvirusattachmenttothecells, therstudy.
yettheyblocksyncytiumformationbetweenHIV-infectedcells
and uninfected cells (52, 58). Those plant lectins that inhibit VirusUncoatingInhibitors
syncytium formation also inhibit HIV replication, and it is
likelythattheyintervenewiththevirusreplicativecycleatthe Of all the retrovirus inhibitors that have been described to
fusion step. This may also be the case for mannose-specific date, the bicyclams, consisting of 2 cyclam (1,4,8,11-tetraaza-
lectinsfromGerardiasavaglia(340)(althoughthelatterlectin cyclotetradecane) units tethered by various bridges (Fig. 8),
was mentioned, but not shown, to block virus binding to the are the only ones that have been postulated to interfere with
cells)andothersources(MachaeriumbiovulatumandMachae- theuncoatingprocess.Thisassumptionhasbeenbasedonthe
riumlunatus)(9). fact that the prototype (JM2763) of this class of compounds
Mannose- and N-acetylglucosamine-specific plant lectins inhibitsthereplicativecycleofHIVatastagethatfollowsthe
may be assumed to interact with specific glycosylation sites virus adsorption step but precedes the reverse transcription
within the viral envelope glycoproteins gp120 and/or gp41, step,andasthecompoundhadapparentlynoeffectonsyncy-
particularly those sites that are rich in mannose (or N-acetyl- tium formation (in a direct syncytium formation assay), its
glucosamine).Theseplantlectinswerealsofoundtoinhibita modeofactioncouldbeattributedtoaninhibitionoftheviral
numberofvirusesotherthanHIV,i.e.,CMV,respiratorysyn- uncoating event (129). This hypothesis was corroborated by
cytialvirus,andinfluenzaAvirus(52).Astheseantiviraleffects ‘‘uncoating’’experimentsinwhichsensitivitytoRNaseAwas
were achieved at concentrations well below the cytotoxicity monitored for the viral RNA that was recovered from HIV-
threshold, the most promising plant lectins should be further infectedcellsthathadbeenexposedtoJM2763:thecompound
208 DE CLERCQ CLIN.MICROBIOL.REV.
FIG. 6. Succinylated(Suc)andaconitylated(Aco)HSA,followingtreatmentofHSAwithsuccinicanhydrideorcis-aconiticanhydride.Perlysineresidue,suc-HSA
andaco-HSAacquireoneortwonegativecharges,respectively,whichmeansagain((cid:68))oftwoorthreenegativechargesoverall.
effectedaconcentration-dependentinhibitionofthedegrada-
tion of viral RNA by RNase A, as could be anticipated if the
uncoating (i.e., dissociation) of the viral RNA from the sur-
roundingviralproteinshadbeenblocked(123).
BicyclamsrepresentanentirelynewclassofHIVinhibitors
and new approach toward the treatment of HIV infections.
Some of the recently synthesized bicyclams (e.g., JM3100), in
which the cyclam moieties are tethered via an aromatic phe-
nylenebis(methylene)bridge(Fig.8),inhibitthereplicationof
HIV-1 and HIV-2 at concentrations which are more than
100,000-fold lower than the cytotoxic concentration (130). In
primaryT4lymphocytesormonocytes,JM3100inhibitsHIV-1
replication at concentrations lower than 1 nM. From time of
additionexperiments,JM3100appearedtointerferewithviral
uncoating,andthiswasfurthercorroboratedbyuncoatingex-
periments in which the RNase A sensitivity of the viral RNA
was monitored (130). JM3100 was also found to interfere di-
rectlywithvirus-inducedsyncytiumformationformation,albeit
at a higher concentration ((cid:59)1 (cid:109)M) than that required for
inhibitionofviralreplication.
FIG. 8. Bicyclams, consisting of two cyclam (1,4,8-11-tetraazacyclotetrade-
FIG. 7. Betulinic acid, RPR 103611: N(cid:57)-{N-[3(cid:98)-hydroxy-1up-20(29-ene-28- cane)moieties,tetheredviaanaliphaticbridge(i.e.,propylene,asinJM2763)
oyl]-8-aminooctanoyl}-L-statine. oranaromaticbridge[i.e.,phenylenebis(methylene),asinJM3100].
VOL.8,1995 ANTIVIRAL THERAPY FOR HIV INFECTIONS 209
FIG. 9. 2(cid:57),3(cid:57)-Dideoxynucleosideanalogs(clockwise):a,carboxylicoxetanocinanalogs;b,oxetanocinanalogs;c,carbocyclic2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxynucleo-
sides; d, 2(cid:57),3(cid:57)-dideoxynucleoside isomers; e, 1,3-dioxolane nucleosides; f, 1-oxo-3-thiolane nucleosides (3TC and FTC); g, 2(cid:57),3(cid:57)-dideoxy-L-nucleosides; h, 2(cid:57),3(cid:57)-
dideoxynucleosides (ddI and ddC); i, 3(cid:57)-azido-2(cid:57),3(cid:57)-dideoxynucleosides (AZT); j, 3(cid:57)-fluoro-2(cid:57),3(cid:57)-dideoxynucleosides (FddClUrd); k, 2(cid:57)-fluoro(‘‘up’’)-2(cid:57),3(cid:57)-
dideoxynucleosides; 1, 2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxynucleosides (D4C and D4T); and m, phosphonate isosteres of 2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxynucleoside 5(cid:57)-
monophosphates.
ReverseTranscriptionInhibitors viral RT), the anti-HIV activity (or inactivity) of 2(cid:57),3(cid:57)-
dideoxynucleosides may be more critically dependent on the
Substrate analogs. All four anti-HIV drugs that have been initialintracellularphosphorylationthanontheireventualin-
formallyapprovedforthetreatmentofHIVinfection,namely, teractionwiththeirtargetenzyme(192,193).
AZT, ddI, ddC, and D4T, belong to the class of the 2(cid:57),3(cid:57)- Thebottleneckintheintracellularmetabolismofthe2(cid:57),3(cid:57)-
dideoxynucleosideanalogs(Fig.9).Theiranti-HIVactivitywas
dideoxynucleosides is the first phosphorylation step by nucle-
disclosed(323,326)shortlyaftersuraminhadbeendescribed
osidekinases.Manydideoxynucleosides(suchasddU)havea
as an anti-HIV agent (235). Following the saturated 2(cid:57),3(cid:57)-
dideoxynucleosides (323), their 2(cid:57),3(cid:57)-unsaturated derivatives lowaffinityfornucleosidekinases(suchasthymidinekinase),
(i.e.,2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)-dideoxycytidineor2(cid:57),3(cid:57)-dideoxycyt- and moreover, the nucleoside kinase activity of some cells
idinene [also referred to as D4C] and 2(cid:57),3(cid:57)-didehydro-2(cid:57),3(cid:57)- (suchasmonocytes/macrophages)atrestmaybeinsufficientto
dideoxythymidine or 2(cid:57),3(cid:57)-dideoxythymidinene [also referred satisfactorily phosphorylate even those dideoxynucleoside an-
to as D4T]) (26, 53, 190, 281, 283) and various other 2(cid:57),3(cid:57)- alogs (i.e., AZT) that have high affinity for the enzyme. In
attemptstoovercomethisproblem,specialprodrugs,i.e.,aryl
dideoxynucleoside analogs were reported to inhibit HIV rep-
methoxyglycinyl derivatives (308) and bis[S-(2-hydroxyethyl-
lication, with selectivity indexes that in some instances (i.e.,
5-chloro-3(cid:57)-fluoro-2(cid:57),3(cid:57)-dideoxyuridine [FddClUrd]) (60, 128, sulfidyl)-2-thioethyl]esters(379),havebeendesignedthatde-
463)approachedtheselectivityindexofAZT(118,122,349). liverthemonophosphateformsintracellularlyandthusbypass
WhileitsselectivityindexiscomparabletothatofAZT,Fdd- thefirstphosphorylationstep.
ClUrd is much less toxic for the host cells than are AZT and Amongthemostpromising2(cid:57),3(cid:57)-dideoxynucleosideanalogs
various other 2(cid:57),3(cid:57)-dideoxynucleoside analogs (60, 128, 463). that have recently been described are 3TC, the ((cid:50))-(cid:98)-enanti-
This compound (BW 935U83) has been selected for further omerof2(cid:57),3(cid:57)-dideoxy-3(cid:57)-thiacytidine(BCH-189),andthe((cid:50))-
development(109a). (cid:98)-enantiomerof2(cid:57),3(cid:57)-deoxy-5-fluoro-3(cid:57)-thiacytidine[((cid:50))FTC]
All 2(cid:57),3(cid:57)-dideoxynucleoside analogs may be assumed to act (83, 405, 406, 409, 429). In both cases the ((cid:50))-(cid:98)-enantiomer
inasimilarfashion;thatis,followingintracellularphosphory- was found to be less toxic and/or more potent than the ((cid:49))-
lation to the 5(cid:57)-triphosphate form, they serve as chain termi- (cid:98)-enantiomer.Theabsoluteconfigurationof((cid:50))FTChasbeen
nators of the RT reaction (as has been clearly demonstrated determinedbyX-raycrystallography,andtheresultsconfirmed
with AZT) (169, 218, 434). As attested to by the inactivity of that the L-isomer [or ((cid:50))-(cid:98)-enantiomer] is indeed the most
2(cid:57),3(cid:57)-dideoxyuridine (ddU) against HIV replication (despite activeenantiomer(465).Akintoallother2(cid:57),3(cid:57)-dideoxynucleo-
the potent inhibitory effect of its 5(cid:57)-triphosphate form on the sideanalogs,3TCand((cid:50))FTCfunction,followingtheirintra-