VOLUME 8
`
`553-1061
`
`ISSN. 0929-8673
`
`urrent
`
`Medicinal
`
`Chemistry
`
`Th e
`
`International
`
`Journalfbr
`
`Timely
`
`In-Depth
`
`Reviews in
`
`Medicinal
`
`Chemistry
`
`Case No. 2:10-ev-D5954
`Janssan Products, LP. 51 al.
`v‘ Lupin Limited‘ et al.
`
`PTX78
`
`(Page 1 of 31)
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`

`
`Currcnr Medicirrn! Cl'remv'5Iry 2001, 8. I543-I572
`
`[543
`
`New Developments in Anti-HIV Chemotherapy
`
`Erik De Clercq*
`
`Rega Inslimrejbi‘ Medical Research, Karhoiielre Uiriversfren Leuven. Minderbroederssrraar I0,
`B-3000 Lerrverz, Belgium
`
`
`
`Abstract: Virtually all the compounds that are currently used, or under advanced clinical
`trial, for the treatment of HIV infections, belong to one of the following classes:
`(1)
`nucleosidc/nucleotide reverse transcriptasc inhibitors (NRTls):
`i.e., zidovudine (AZT),
`didanosine (ddl), zalcitabine (_ddC), stavudino (cl4T),
`lamivudine (STC), abacavir (ABC),
`emtricitabine [(-)FTC], tenofovir (PMPA) disoproxil fumaratc; (ii) non—nuc|eoside reverse
`transcriptase inhibitors (NNRTls): i.e., nevirapine, delavirdinc, efavircnz, emivirine (_MKC-
`442); and (m) protease inhibitors (Pls):
`i.e., saquinavir. ritonavir,
`indinavir, nelfinavir, atnprenavir, and
`lopinavir. in addition to the reverse transcriptase and protease step, various other events in the HIV replicative
`cycle are potential
`targets for chemotherapeutic intervention: (1) viral adsorption, through binding to the viral
`envelope glycoprotein gpl20 (polysulfalcs, polysulfonates, polyoxometalates, zintevir, negatively charged
`albumins, cosalane analogues); (Q) viral entry, through blockade of the viral coreceptors CXCR4 and CCRS
`lbicyclams (i.e. AMD3l00), polyphemusins (T22), 'I‘AK—779, Mil’-lot LD78B isofonn]; (iii) virus-cell fusion.
`through binding to the viral glycoprotein gp4l
`[T-20 (DP-I78). T-1249 (DP-I07), siarnycins, betulinic acid
`derivatives];
`(J1) viral assembly and disassembly,
`through NCp7 zinc finger-targeted agents [2,2'—
`dithiobisbenzamides (D1BAs), azadicarbonamide (ADA) and NCp7 peptide mimics];
`(y_) proviral DNA
`integration,
`through integrase inhibitors such as L—chicoric acid and diketo acids (Le. L-731.988); (11) Viral
`mRNA transcription, through inhibitors of the transcription (transactivation) process (fluoroquinolonc K-12,
`Str-epromyces product EM2487. temacrazine, CGP64222). Also,
`in recent years new NRTls, NNRTIS and Pls
`have been developed that possess respectively improved metabolic characteristics (i.e. phosphoramidatc and
`cyclosaligenyl pronttcleotides of MT), or increased activity against NNRTI-resistant
`ll1V strains [second
`generation NNR'l‘ls,
`such
`as
`capravirine
`and
`the
`novel
`quinoxaline,
`quinazolinone,
`phenylcthyIthiazolylthiourea (l’B’["I‘) and cmivirinc (MKC—442) analogues], or, as in the case of Pls, a dif'l'erent,
`non-pcptidic scaffold [i.o. cyclic urea (DMP 450), 4-ltydroxy—2—pyrone (tipranavir]]. Given the multitude of
`molecular targets with which anti-HIV agents can interact, one should be cautious in extrapolating from cell-
`frcc enzymatic assays to the mode of action of these agents in intact cells. A number of compounds (Le. zintcvir
`and L-chicoric acid, on the one hand; and CGP64222 on the other hand) have recently been found to interact
`with virus-cell binding and viral entry in contrast to their proposed modes of action targeted at the integrasc
`and transactivation process, respectively.
`
`INTRODUCTION
`
`Combination therapy, comprising at least three anti-HIV
`drugs, has become the standard treatment of AIDS or HIV-
`infcctcd patients. Virtually all drugs that have been licensed
`for clinical use (or made available through expanded access
`programmes) for the treatment of HIV infections fall into one
`of‘ the following three categories: (1), nuclcosidc/nucleotide
`reverse transcriptase inhibitors (NRTlS), that. following two
`phosphorylation steps (tenofovir) or three phosphorylation
`steps (zidovudine, didanosine, zalcitabine, stavudine,
`lamivudine, ab-acavir), act, as chain terminalors, at
`the
`substrate binding site of the reverse transcriptase; (Q), non-
`nucleosidc reverse transcriptasc inhibitors (NNRTIS), that
`interact with the reverse transcriptase at an allostcric, non-
`substrate binding site (ncvirapine, dclavirdine, efavircnz);
`
`
`
`‘Address correspondence to this author at the Raga Institute for Medical
`Research, Katholieke Univcrsiteit Leuven. Minderbrocdcrsstraat 10, B-
`3000 Lcuvcn, Belgium; c-mail; erik.t.tcc|ercq@rcga.ltulcuven.ac.bc
`
`and (Q1), protease inhibitors (Pls), that specifically inhibit.
`as peptidomimctics,
`the virus-associated protease
`(saquinavir, ritonavir,
`indinavir, nclfinavir, amprenavir,
`lopinavir). Guidelines to the major clinical trials with these
`compounds have been recently published [I].
`
`in numerous studies, combinations of NRTIS, I\_lNRTis
`and Pls have been found to decrease HIV viral load, Increase
`CD4 count, decrease mortality and delay disease
`progression, particularly in AIDS patients with advanced
`immune suppression [2]. When initiated during early
`asymptomatic HIV infection, highly active antiretroviral
`(combination) therapy (HAART] initiates rapid reversal of
`disease-induced T—cell activation, while preserving
`prctherapy levels of immune function, suggesting that
`therapeutic benefit may be gained from early aggressive anti-
`HlV chemotherapy [3]. Combination therapy that produces
`sustained suppression ofplasma HIV RNA may also be able
`to reduce the virus burden in the lymphoid tissues [4],
`although clearance of plasma viremia is not
`invariably
`associated with immune restoration [5] and not at all
`
`0929-8673/0l $28.l)0+.0l)
`
`in 2001 Bentham Science Pubtisltcrs Ltd.
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`(Page 2 of 31)
`
`

`
`I544 Current Medicinal Cliernisrnt, 200!-. Vol. 8, Na. 13
`
`Erik Dc Clef""l'
`
`paralleled by 21 reduction in viral DNA burden in the
`peripheral blood mononuclcar cells [6].
`
`The introduction of HAART or combined anti-HIV drug
`regimens has had profound repercussions on various AIDS-
`assouiatcd diseases. While partial
`immune:
`restitution
`
`induced by HAART in patients with advanced HlVlinI_'c_cfi0“
`can exacerbate clinically apparent cryptococcal meningitis l7l
`or CMV vitritis, HAART may reduce the prevalence 01"
`cervical
`squamous
`intra-epithelial
`lcsions
`in I-ll'V~
`scropositive women caused by HPV [8]. Scvcral studlex
`have indicated that HAART significantly improves ilk‘
`
`HIV-1 SU (gp120)
`
`
`
`Fig. (1). Model for the stirlhce ISU) glycoprotein gplll). Figure taken frurn rel‘, 2'2.
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`|PR2015-01030
`
`(Page 3 of 31)
`
`

`
`Developments in Ami-HI V CIlematlierapy
`
`Currerirfiledic-incl C'Imm'.s'tI;v, 2001. Val. 8,/Va. .13
`
`I545
`
`prognosis of AIDS patients with progressive fnuhifgcal
`leukoenceplialopathy [9-1 1]. Likewise, HAART has been
`shown to favorably alter the prognosis of CMV retinitis in
`HIV-infected individuals, as attested by long-lasting
`remission of CMV retinitis without CMV niaingenance
`therapy [[2], lack of reactivation of CMV retinitis after
`stopping CMV maintenance therapy [13], decrease of CMV
`replication (viremia) [14,15] and increased survival [l6,l 7].
`In the era of combination antiretroviral
`therapy the
`management of CMV diseases in patients with AIDS has
`undergone dramatic modifications [18,19].
`
`Although the long-term goal ofcradicating the virus from
`latently and chronically infected cells remains forbidding [20]
`the advent of so many new compounds, other than those that
`have been lbrinally approved. for the treatment of HIV
`infcctioiis, will undoubtedly improve the prognosis of
`patients with AIDS and AIDS-associated diseases. Here I
`will primarily address those new anti-HIV compounds that
`(i) have emerged as promising anti—HlV drug candidates
`during the last few years, that (i_i) are in preclinical or early-
`clinical development, and that (fl) are targeted at we”.
`defined steps in the HIV rcplicative cycle.
`
`VIRUS ADSORPTION (gpl20) INHIBITORS
`
`A great variety of polyanionic compounds have been
`described to block HIV replication through interference with
`virus adsorption (or binding) to the cell surface;
`i_e_
`polysulfates, polysulfonates, polycm-b()xy1a[e5_ pmyphos:
`phates, polypliosphonates, polyoxoinetalates, etc, This ctass
`of compounds also comprises the sulfated polysaccharides
`extracted from sea algae [21]. All these compounds \'v'he[h3r
`synthetic or ofnatural origin. are assumed to exert their anti-
`HIV activity by shielding off the positively charged sites in
`the V3 loop ofthc viral envelope glycoprotein (gpl20) [22]
`which is necessary for virus attachment to the cell surface
`hcparan sulfate. a primary binding site, before more specific
`binding occurs to the CD4 receptor of the CD4* cells (Fig.
`1). llcparan sulfate is widely expressed on animal cells and,
`as it
`is involved in the virus-cell binding of a broad
`spectrum of enveloped viruses, including HSV [23], it also
`explains why polysulfates have a broad~spcctriim antiviral
`activity against HIV, HSV and various other enveloped
`viruses.
`
`The major role of polysulfates or polyariioiiic substances
`in general in the management of HIV infections may reside
`in the prevention ofscxual transmission oft-{IV inlcctioii. as
`these compounds, if applied as a vaginal forniulatioii, may
`succcssfiilly block HIV infection through both virus-to-cell
`and cell-to-cell contact. These compounds therefore merit
`being pursued as vaginal inicrobicidcs. The fact that in
`addition to their anti-HIV activity,
`these polyanionic
`substances also inhibit other sexually transmitted disease
`(STD) pathogens further adds to their potential tlierapcutic
`and preventive value. One candidate compound is
`polytsodiuin (4-styrene sullbnatc) which is highly effective
`against several STD patliogcns including not only IISV but
`also Nei'sseri'u goriorr/ioene and CI'I[(m‘l_1’t[.l'(I n-ac/ioniaris
`124].
`
`Foremost among the polyanionic substances that have
`been described as virus—ce|l binding inhibitors are dexlran
`sulfate. dexti-in sulfate, polyvinylalcohol sulfate (PVAS)
`(Fig. 2), polyacrylic acid/polyvinylalcohol sulfate copolymer
`(PA)/AS} (Fig. 2)
`(and many other polysulfates),
`naphthalene sulfonate (PROZZOOO) (Fig. 2) [25], polyoxo—
`metalates such as JMl59D or K13[Cct'SiW] ;O39)2].26 l-I20
`(Fig. 2) and JM2766 or K6[BGa(H3O)W] |O3q].lS H20
`[26], and the negatively charged (i.e., succinylatcd) human
`serum albuniins [28]. Polyoxometalates can assume various
`structures, such as Keggin, Dawson, “Doublc" Kcggin,
`“Double" Dawson, “Triple” Keggin and ‘‘Large‘’ Ring
`structures:
`they all
`inhibit virus adsorptioii
`through
`iiitcrlcrcncc with the binding of the viral envelope
`glycoprotein gpl20 to CD4“ cells [27].
`
`Zintevir (ARJ77, T301 77), which is capable of" forming a
`double guanosine quartet (G-quartet) [293 I], was originally
`described as an inhibitor of HIV-1 integrase (sec infra). Yet,
`its primary target, accounting for its anti-HIV activity in cell
`culture, does not appear to be the intcgrase, but the viral
`adsorption step [32,33] As first described for dextran sulfate
`[34], resistance to polyanionic substances can arise through
`repeated passage of HlV—l
`in the presence of the compound,
`and this resistance is invariably mediated by mutations in
`the gpl20 molecule,
`i.e.. Sll4N (VI loop), S|34N (V2
`loop), K269E, Q278H, N293D (V3 loop). N3?.3S (C3
`region), deletion of FNSTW at positions 364-368 (V4 loop)
`and R3871 (CD4 binding doinain). It was later shown that
`HIV-l resistance to other polyanionic substances, that are
`targeted at the virus adsorption stop, such as zintcvir [33],
`cyclodextrin sulfate [35] and negatively charged _alburnIns
`[36], emanates through a similar set of niutatioris in gpl20.
`The viral gpl20 glycoprotcin must thus be considered as tht'
`main target for the anti-HIV-1 action ofzintevir and the Olllfir
`polyanionic substances.
`
`Cosalane analogues represent another class 0i:t't'l0i6CUl'33
`that owe their anti-l—llV activity to the polyatiiomc 6033111111‘:
`pharmacophore [37—39]. Cosalaiie analogues such as that
`shown in Fig. 2 are able to bind to either gpl20 0!‘ CD4 or
`both, thus affectiiig the interaction ol'gpl2U with CD4. Find
`the ensuing virus-cell attachment and fusioii processes [37-
`39]. A hypothetical model has been proposed for the binding
`of the cosalane rnotil' with CD4, which should assist in the
`design of more cllicient congencrs [38].
`
`VIRAL CORECEPTOR ANTAGONISTS
`
`To enter cells, following binding with the CD4 receptor,
`the HIV-1 particles must interact, again through the virai
`envelope glycoprotein gpl20, with the CXCR4 coreceptor
`[40] or CCR5 coreccptor [4]] (Fig. 3). CXCR4 is the
`corcceptor for HIV-l strains that iiilizct T—cc|ls (‘T-tropic or
`X4 strains), and CCR5 is the coreccptor for HIV-1 strains
`that infect inacrophagcs (M-tropic or R5 strains). CXCR4
`and CCR5 have not evolved simply to act as corcceptors for
`l-l[V entry; they iioriiially act as receptors for clicmokines
`(clicinoattractant cytokiiies). The nornial ligands for C CR5
`are RAN'Tl£S (“regulated upon activiitioii, normal T—cc||
`expressed and secreted") and Mll’-lot and - I [5 (“inucrophage
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`(Page 4 of 31)
`
`

`
`1546
`
`cm-mu Medicinal Clmmislry. 200:. Va. 3, Na. 13
`
`Iirik 0°’ ‘3"“""?
`
`CIIa——(l.H
`OH
`
`CH2-—i.‘H
`080;
`
`|
`
`PVAS
`
`m
`
`CHg—(|.‘!l
`C0;
`
`Cl lgifll
`OH
`
`Cl—l2—-CITH
`0503'
`
`1;
`
`1
`
`n1
`
`PM/AS
`
`f :1‘
`
`Naphthalene sulfonate
`PRO2000
`
`
`
`K,,|Cc(SiW,,(J,,),|.2IS H10
`JM 1590
`
`I IOUC
`
`
`
`IIOOC
`
`Cosalane analogue
`
`the viral envelope glycoprotein gp|20: PVAS, PAVAS, PRO2000 and JMI590; also
`Fig. (2). Polyanionic substances targeted at
`targeted at the cellular CD4 receptor: cosalane analogues.
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`IPR2015-01030
`
`(Page 5 of 31)
`
`

`
`9*-""’°P”""“’ "' """'”""~”"e"w"'¢""PJ’
`
`Current Medicinal Cllelnisrqt 2001 Val. s No I3
`
`1547
`
`:3
`‘Me:
`
`.309 VI‘? 09
`$9199
`96:
`09909
`:2-'
`'3
`0,39as 0
`
`Cytoplasmic
`
`
`
`_
`
`SIIITHI
`lifll
`
`9
`0099 9099
`HA Antigenic Tag J
`
`CXCR4: coreceptor for X4 HIV-1 strains
`
`91, cf) 3' - NH;
`8.39 N 0 Y -r 9
`
`5 0
`
` Q
`
`an-egg"
`
`Fig. (3). CXCR4: receptor for the CXC chemokine SDF-l, and coreceptor for T-tropic (X4) HIV-I strains. CCR5: receptor for the CL
`chemokines Mil’-lot. MIP-IB. RANTES, and corcceptor for M-lropic (RS) HIV-I strains. Figure taken from rclé. 40 and 4].
`
`CCR5: coreceptnr for R5 HIV-1 strains
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`(Page 6 of 31)
`
`

`
`I548 Curr£ruMetlt'cina.l Chemistry. 2001. Vol. 8. No. I3
`
`Erik De Cle-r.-.-q
`
`inflammatory proteins”), whereas for CXCR4 only one
`natural ligand, namely Sf)F—l {“stromal-cell derived factor”)
`has been identified. RANTES, MIP—l0t and MIP~l{3 block
`the entry ofM-tropic, whereas SDF—l blocks the entry of T-
`tropic HIV strains [42]. Of these ehemokines. the LD78B
`isoform of MIP-lot has emerged as the most potent
`
`chemokine for inhibiting HIV-1 infection [43,44]. LD785
`can be considered as a potentially important drug candidate
`for the treatment ofinfeetions with R5 HIV-1 strains.
`
`TAK-779. £1 quatemary ammonium derivative (Fig. 4) is
`the first non—peptidie molecule that has been described to
`
`N /*< >:\ m‘
`j N“
`[NH HN:|
`|:Nll
`lINj
`AMDJIDO
`K/I
`
`/—\ I S \ /j\
`NH N]
`[N HN
`rm
`mi
`N‘ \
`I /
`
`AMD3329
`
`/\
`
`R
`
`\\ Rmnlt
`
`T222
`
`K
`
`R
`
`Rtcowtm
`
`\\ RtNI 1,)
`
`T134
`
`0
`
`I-IzN %Nt-I2
`JNH
`=
`D
`E
`I
`It W H
`0
`
`I
`
`.
`
`T
`0
`
`um? Ni-:2
`Nll
`U
`
`,
`
`:
`
`|l2N §",»Nn2
`mu % um
`Nil
`Nil
`-
`0 J 0
`
`'-I
`
`.
`
`5
`
`0
`
`0
`
`0
`
`N”:
`
`“N
`ll2N’T"Nl-lg
`
`Iii]:
`H2N’+"‘NH;
`
`IIN
`li2N ’?~Nt|g
`
`ALX-ll)-4C
`
`HN
`llgN’/”-‘+\"Nl-in
`
`“N
`ll2N ’I‘N]lo
`
`M Qp
`0
`cu;
`0
`N _<:/\,
`
`1+
`I
`
`CH3
`
`‘
`
`.
`
`ll3(:
`
`TAK-779
`
`Fig-(41 CXCR4 antagonists: AMDNUO. AMD3329. T22. ‘H34 and ALX40-4C. CCR5 antagonist: “rm<-779. TAK-779 eorresputlds
`1” N«N‘dl_'"°lhY"N'l4'[l[344‘mfilhltlphenyTl-6.7-diltydro-SI!-henzoeyclohepten-8-yljcarbonylIantinojbcnzyll
`tetraItydro-2H—pyr:m—
`4-ammonium cltloride.
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`(Page 7 of 31)
`
`

`
`Develapmems in Aim‘-HI V Clmuarlitmpy
`
`Cnrreiir Medicliial CI:ei:n‘i'!1;p, 2001. Vol. 8. No. 13
`
`I549
`
`block the replication of M—tropic R5 HIV-1 strains at the
`CCR5 level [45]. A binding site for this molectile has been
`identified within the transmembrane helices of CCR5 [46].
`'l‘AK-77‘) has been Found to inhibit R5 I-{IV-l strains in the
`
`nanomolar concentration range, while not affecting X4 l~llV—l
`strains at
`l0,0()0-Fold higher concentrations [45]. TAK-779
`is not ti “pure” CCR5 antagonist, as it also demonstrates
`some antagonism towards CCR2b. Unlike RANTES, TAK-
`779 does not induce internalization ot‘Ct.‘l{5. its therapeutic
`potential for HIV-1 infections, remains to be further explored.
`
`i.e.
`Almost simultaneously [47-49], three compounds.
`the bicyclam AMD3 100 (Fig. 4} [47].
`iTyr-5,l2.Lys-
`7]polyphcniusin or T22 (Fig. 4) [48] and the nonapeptide
`[49] (D-/\rg)g or ALX40—4C were announced as CXCR4
`antagonists, blocking the replication ot‘T-tropic X4, but not
`M-tropic R5, HIV-1 strains through selective zintaigonisiii of
`CXCR4. Yet, the third compound (ALX40-4C) cannot be
`regarded as truly specific for CXCR4, since through its
`highly cationic character, it also interferes with the entry of‘
`viruses other than HIV. The second compound (T22), an l8-
`arniiio-acid peptide, can be shortened to 14 amino acids, as
`in T134 (Fig. 4), without loss of activity [50]. It has also
`been claimed that T134 would be active against AMD3lUO-
`resistant HIV-1 strains [51], but this claim has not (yet) been
`substantiated. Whether T22 and/or T134 offer potential in
`the prevention and/or therapy of HIV infections is another
`issue that needs to be further addressed.
`
`The bicyclams are the most specific and most potent
`CXCR4 antagonists that have been described to date
`[52,53]. The p-phenylenebis(methylene)-linked dimer of the
`py[iso-14]ane N4 (AMD3329) (Fig. 4) displayed the highest
`antiviral activity of the bis-azaniacroeyciic analogues reported
`[53], exhibiting a 50% eltective concentration (EC5g) against
`HIV-1 and HIV-2 of 0.8 and L6 nM, respectively, that is
`about 3- to 5-fold lower than the EC50 of AMD3 l 00.
`
`The bicyclams had been known as potent and selective
`HIV inhibitors for a number ofycars [S4,55], before their
`target of action was identified as the CXCR4 coreccptor
`[47.56,57]. The bicyclatii AMD3 I00 inhibits the replication
`of X4 HIV-1 strains within the nanoniolar concentration
`
`toxic to the host cells at
`is not
`it
`range [55]. As
`concentrations up to 500 ttM., its selectivity index, or ratio
`of 50% cytotoxic concentration (CC50) to 50% antivirally
`eftiective concentration (EC5g) can be estimated at > 100,000.
`
`It took more than 60 passages (300 days) in cell culture
`for the HIV-1 clone NL4-3 to become 300- to 400-fold
`
`the resistant virus had
`to AMD3100 [58,59];
`resistant
`several mutations
`scattered over
`the whole gpl2O
`glycoprotein, but primarily clustered in the V3 loop (i.e.,
`R.272T, S27-4R, Q2781-I,
`l288V, N292!!! and A297T).
`Most, if not all, of these mutations may have contributed to
`the resistant phenotype, as indicated by recombination
`experiments with overlapping parts of the envelope gene
`[60].
`It was postulated that the overall, three-diinensional.
`conlorinatioii of gp120, rather than individual amino acid
`substitutions,
`is
`the
`prime determinant of
`the
`resistance/sensitivity prolilc of HIV strains to bicyelams
`
`[61]. Resistance to AMD3 100 (or SDF— I) does not lead to a
`switch in coreccplor use [62].
`
`A close correlation has been found, over a concentration
`range of 0.|—l0O0 ng/ml, between the AMl)3l00
`concentrations required to inhibit
`{i_) HIV-1 NL4-3
`replication, (fl) monoclonal antibody (_inAb IIZGS) binding
`to the CXCR4 coreceptor, and (ji_i) SDI’-l-induced signal
`transduction (Ca-" tlux), Suggesting an iiitiinate rclationsliip
`between these three parameters [56,57]. The inhibitory effects
`OFAMD3 l 00 on the T-tropic HIV-I NL4—3 strain have been
`demonstrated in a wide variety ol’ cells expressing C XCR4_
`includiiig PBMC, and. vice W.’l‘.‘o‘(I. various T-tropic and
`dual-tropic. but not M—tropic,
`lvlIV—l strains have proven
`sensitive to AMD3 100 in PBMC.
`
`tispartic acid)
`Negatively charged amino acid (i.c.
`residues in both the ainino~terminus and second extracellular
`
`loop (ECL2) ofCXCR-4 are thought to be involved in the
`recognition ofCXCR4 by X4 HIV-1 strains [63]. Dil‘l'ci'cnt
`single amino acid substitutions of a neutral amino acid
`residue for aspttrtie acid and of a non-aromatic residue for
`phenylalanine, in the second extracellular loop (ECL2) or in
`the adjacent
`inembrane-spanning domain (TM-4), were
`associated with resistance to AMD3lO0 [04]. From these
`studies a model emerged for the interaction cl‘ AMD3l00
`with CXCR4, whereby the positive charges ofthe bieyclanis
`interact with the aspartic acid residues of ECL2 and TM4,
`whereas the aromatic linker [i.e., phenylenebis(mcthylene) in
`AMD3l00] might engage in hydrophobic interactions with
`the Plie-X-Plie motifs in ECL2 and TM4.
`
`Feline iininunodelicicney virus (FIV), that predominantly
`uses CXCR4 for entering its target cells, is, just like I-IIV,
`highly sensitive to inhibition by the bieyclams [65]. The
`high anti-FIV potency and selectivity oi‘ the bieyelams may
`serve as the starting point for establishing an appropriate
`therapy for the treatment of FIV infections in cats, but,
`furthermore,
`it may prove most valuable as a model
`to
`delineate novel strategies to prevent AIDS progression in
`humans. When the bicyclam AMD3l00 was added I0
`PBMC infected with clinical HIV isolates displaying the
`syneytium-inducing (SI) phenotype, these strains reverted to
`the non—syneytium-inducing (NSI) phenotype, and,
`concomitantly,
`these strains switched from CXCR4 to
`CCR5 coreceptor use [66]. These findings indicate that
`selective blockade ofCXCR4 by AMD3 I00 may prevent the
`switch from the less pathogenic M-tropic R5 to the more
`pathogenic T-tropic X4 strains of HIV, that in vivo heralds
`the progression to AIDS.
`
`AMD3 I00 has proved eflieacious, alone and in
`combination with other anti-l-llV drugs,
`in achieving a
`marked reduction in viral
`load in the SClD—hti Thy/Liv
`mouse model [67]. Its in viva activity against T-tropic HIV
`infections has thus been demonstrated. Following a phase I
`clinical trial for safety in normal healthy volunteers [68],
`AMD3l00 has recently entered phase II clinical trials in
`HIV-iiiI"ectcd iiidividtials. AMI)3t00 can be considered as a
`
`highly specilic CXCR4 antagonist, which consistently
`blocks the outgrowth of‘ all I-IIV variants (X4 and Lltlill-ll'0plC
`X4/R5) that use the CXCR4 receptor for entering the cells
`(lymphocytes or riiacropliages/inoiioeytes) [69].
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`lPR2015-01030
`
`(Page 8 of 31)
`
`

`
`1550
`
`("nrrt'u!Medicinal (‘lit-rr:i.rrr'_i'. 2t-WI. VUL 3- r‘V"- 13
`
`!:'ri:'.' De {'t't:.t't'tf
`
`in‘-\sk\q‘‘55o\.s.(‘~"'n
`
`15CI
`
`I4
`
`‘a
`
`
`
`Fig, (5), ‘rite t]IV.| entry process: sequential binding of the HIV-I gpl2tJ glycoprotein to CD4 and then to the coreeeptor [Lt R: ul
`CX(‘R~’l) is believed to irtdttce conformational changes in the envelope _t_Llyeoprott.:iti contple.\' (middle panel as C0lIlt|‘lt1t‘(.'t.l[tt).lt.‘ll
`l’5""‘:’l)'
`in the hypothetical intermediate tniotlel depicted in the middle ptinel), the N36 coiled coil tbluel in the gpcll eetodoinain is appatLnl~
`and the gpctl amino termini t"l'usion peptides“) are interacting, with the target cell mt.nilirane. The ltytlropliobie grooves on the outclt
`face oi’ the N30 coiled coil are ttnoeetipied and available tor binding with either the gp-’ll
`('34 helices ttiiageitttil or Cl’(ll“¢illC()l.l5:
`inhibitors, one ol‘ which tmagenta) is depicted on the lelt side in the middle panel. The panel on the right depicts the association oi: tltt.
`C34 helices tniagenta) and the N30 coil (bloc; resulting in the approximation of the viral and target cell membranes. I-'igttre Itiktll lftlm
`rel‘. 70.
`
`VIRAL FUSION (gp4l) INHIBITORS
`
`The interaction of the X4 or R5 |llV—l enveiope
`glycoptmcitt gpl2t) with the coreceptor CXCR4 or CCRS.
`respectively,
`is ioiloxved by a spring-ioaded action oi‘ the
`viral glycoprotein gpal
`tnormally covered by the bulkier
`gpl'.ZU). that then anchors through its amino termini (the
`“fusion peptides") into the target cell
`tnembrane. This
`initiates the fusion ofthe two lipid bilayers, that ol‘ the viral
`envelope with that oi" the cellular plasma membrane (Fig. 5)
`[70]. At the onset of the l‘usion process, the hydrophobic
`grooves on the surface ofthe N30 coiled coil and the surlacc
`ofthe N36 pocket, in the gpéll ectodomain bcconic available
`for binding with either the gptll C34 hclii-;, or extraneous
`inhibitors, such as D—amino acid t'lt3— to I8-residue) peptides.
`ti.-.'., “l)ltl—l’X-2l(") that bind to the N36 pocket [71] or
`DI’-l7l'.l ‘(T-20), a 36—residLte peptide that binds to the
`ltydtnphoiiie groove oI'N36 [70].
`
`'1‘-'20 tpentatltside) is a synthetic, 36-amino acid peptide
`correspontling to resitlues l27'—l62 ol"tlte ectodomain ofgpnll
`tor residues 643-678 in the gplfill precursor‘) (l"i_t;. 6). T—2t.l.
`previously called [JP-178, was modeled alter a specilic
`domain (within gp4l) predictive oi’ U.-Il(.‘llL‘{ll seeontlary
`structttre: I)l’—l'/'8 consistently al‘t'ordetl
`|t)tl"u blockade oi‘
`virus-ntetliatetl cell-cei]
`litsion (syncytitun formation) at
`concentrations t'aiti_;in_t; from 1
`to ill ng,/nil, i.e., 104- to it)”-
`l‘oitl
`lower titan the cytotoxic concentration [72,73]. In an
`cliort to ttiid-:rstant.l the mechanism ol' actiott 0|‘ DI’-l7i%,
`
`resistant variants of HIV-1 were generated by serial passage
`otthe virus in the presence olinereasing. doses ofthe peptide.
`Tltese resistant variants revealed mttttttions in a contiguous
`3—atnino aeid stretch t(.ilV —-> SIM. DIM, IJTV) at positions
`°u(:~3H within the antino terminal heplad motiI‘ol‘gp4l [74].
`
`'l'—2t)
`that
`is
`liypotltesis [75]
`the current
`Altlttiuglt
`inhibits virus-cell fusion by preventing the fortitzttitiit oi‘ the
`
`tletcrminants 0t’
`interinediatt: (Fig. 5),
`stable coiled coil
`coreceptor specificity within the gpl20 V3 loop ‘can
`modulate T-20 cl'l'tcaey. In particular, the interaction oi the
`gp|Zt) V3 loop with CXCIUI may contribute to the
`sensitivity to "I-20, as X-'1 HIV-1 strains were iinind to be, at
`an average 0.8 log,..,+l'oltl more sensitive to T-20 than RS
`viruses [76].
`In addition.
`it has been suggestetl that ‘T-_3l')
`may interfere with membrane ltision at a post lipid-tntxntg
`stage [75].
`
`An initial clinical trial has been carried out with 'l‘-Zt) in
`four doses (3, 10, 30 and tilt) mg twice daily. intrave1iotIsl‘_v,
`ibr I4 days) in sixteen lllV-inlected adults: at the liigliesi
`dose t ltltl mg. twice daily), '1‘-20 achieved by the 15"‘ day a
`1.5- to 2.0-fold reduction in plasma HIV RNA [77]. Tltcfic
`data provide proo|'-oi’-concept that lllV fusion inliibitors are
`ablc to reduce virus replication in vivo. in trials ol’28 tlztys it,-
`less T-20 has been given as twice-daily sttbctttancolls
`ittjeclintts or continuous intitsion, and in one trial, with (1
`single patient, the compound has been given for more than
`28 days in combination with other anti-lllV ttgettts ]78], In
`the latter study,
`l(lt)tl—l'old suppression ol‘|l|V—l RNA \\’i\5
`tttaintaittctl liar 20 weeks. no evidence ofgenotypic resistance
`to 'l'-El) was observed. and no anti-'l'-Ztl antibodies wet-t;
`detected alter 28 weeks of administration ol‘ T-20 [Tl->']_
`Mettnwhile, T-20 has proeeetletl to pltase ll clinical trialg]
`and phase I clinical trials have been initiated with ‘I’-IEJA),
`39 mer peptide derived l'rotn Dl’-I07, a .38 amino acid
`peptide correspotitling to residues 558-595 ol'g,pl()U; 'l‘-I24‘)
`would be I0-fold more potent than T-20 when evaluated tin
`i'i'i'm) tinder the same eircttntstatttzes [79].
`
`Anotlter polypeptide that may interact directly with gp4l
`is siantyein (Fig. 6). Siatttycitt is a tricyclic 21-amino acid
`peptide isolated li'om Si"i'L‘pIr)NI_)-‘€.'L'.5'I
`three vat'ietie.~i ml‘
`siamycin have been described ]siamycin ll
`(RP 7l‘J55.
`l5l'vtY-293t)3)
`]i%tl—R'.Z], Ni’-06 (FR ‘)t)l724l [83,84] anti
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`|PR2015-01030
`
`(Page 9 of 31)
`
`

`
`Developments in Am‘!-HIV Chemotherapy
`
`CurrentM¢dicInalCI1emist:;y, 2001'. Val 3. Na 13
`
`I551
`
`_
`Fusion
`Leucine‘ zipper
`Peptide
`region
` NH;
`558
`517
`532
`
`595
`
`DP- I0
`
`7
`
`Membrane
`spanning
`mgion
`
`I52 am inc acids
`
`/~—cooH
`
` ..
`
`43
`
`678 689
`
`710
`
`DP-I78
`
`T-20
`
`YTSLIHSLIEESQNQQEKNEQELLELDKWASLWNWF
`ss~_.q', L...G‘__ 4
`
`“
`G
`
`: Siamycin II = RP ‘H955 = B-MY-29303
`: NP-06 = FR90I72-I
`
`4V-t I
`l'lV—> I
`
` Siamycin I = BMY-19304
`
`RPR I036ll
`
`Fig. (6). Compounds targeted at the transmembrane (TM) glycoprolein gp4l: T-20. Siamycins and RPR 103611.
`
`siamycin I (BMY-29304) [82,85]], which differ from one
`another only at position 4 (V or I) or position 17 (again, V
`or I). The siamycins have been found to inhibit HIV
`infection in vilro and qualify as fusion inhibitors also,
`because they exert a strong inhibitory effect on syncytium
`formation while only weakly interfering with virus-cell
`binding [83,84]. Siamycimresistance mutations have been
`detected in gp41, thus pointing to gp4l as a likely target for
`the action of siamycin II [85]. There is,
`in fact, some
`homology between siamycin II and residues 608-628 of gp4l
`(amino acid numbering for the gpl60 precursor) [81].
`Siamycins may hamper the fusogenic activity of gp4l by
`different mechanisms: for example. through direct binding to
`the ectodomain of gp4l and/or induction of conformation
`changes in this domain. The exact mechanism of action of
`siamycin remains to be resolved, as do its therapeutic
`potential and pharmaookinetic profile.
`
`The betulinic acid derivative RPR 103611 (Fig. 6)
`represents the only non-peptidic low-molecular-weight
`compound known to block HIV-1 infection through
`interaction with gp4l: this triterpene derivative has been
`found to inhibit the infeclivity of a number of HIV-1 strains
`in the 10 nM concentration range [86], apparently through
`interference with a post-binding, envelope-dependent step
`involved in the fusion of the virus with the cell plasma
`membrane. The exact mode of action of RPR 1036ll
`
`remains to be elucidated. Resistance to this compound
`appears to be associated with the emergence of two amino
`acid substitutions within gp4i (R —> A at position 22, i —>
`S at position 84) [87]. In the context of the HW-I strain LA1
`(subtype B), the I843 mutation would be sufficient for drug
`resistance [87]. Also, the L9IH mutation would impart
`resistance to RPR |036ll
`[88]. Both [84 and L9! are
`located in the “loop region” of gp-41 separating the proximal
`
`Janssen Ex. 2017
`
`Lupin Ltd. v. Janssen Sciences Ireland UC
`IPR2015-01030
`
`(Page 10 of 31)
`
`

`
`1552 Current Medicinal Chemistry. 2001. Vol. 8. Na. I3
`
`and distal helix domains. The antiviral efficacy of RPR
`103611 therefore depends on the sequence, and accessibility,
`ofthe gp4l loop region [88].
`
`RPR l036ll may be regarded as an interesting lead
`compound in the pursuit of non-peptidic low-molecular
`weight HIV fusion inhibitors targeted at gp4l. Recently, a
`stereoisomer of RPR [0361 l, namely [C 9564 (43-[8—(28~
`betuliniyl)amino octanoylamino]-3R-hydroxy-6-methylhepta-
`noic acid), has been found to inhibit HIV replication through
`interference with the viral envelope-induced membrane fusion
`[89]: in this case, the viral gpl20 was thought to play a key
`role in anti-HIV activity, since two mutations (G237l{ and
`RZSZK)
`in gpl20 affected the viral sensitivity to the
`compound [89].
`
`the search has
`Using molecular docking techniques.
`begun for small molecules that are targeted at hydrophobic
`cavity within the gp4l core [90]. This Search has already
`yielded two compounds (ADS-Jl and ADS-J2) having
`inhibitory activity at ].tM concentrations on the fonnation of
`the gp4I core structure and on HIV-I
`infection [90].
`However, ADS-J I and ADS-J2 can be considered as tetra-
`
`and trisulfonatcs. respectively, and are thus reminiscent of
`other polysulfonatcs that have been shown to inhibit virus-
`cell binding through interference with gpl 20.
`
`NUCLEOCAPSID PROTEIN (NCp7) Zn FINGER-
`TARGETEI) AGENTS
`
`The two zinc fingers [Cys—X3—Cys-X4-His-X4-Cys
`tCCl|C), whereby X = any amino acid] in the nucleocapsid
`
`Erik De Clercq
`
`[91] comprise the proposed
`(NCp7) protein (Fig. 7)
`molecular target for zinc-ejecting compounds (Fig. 8) such as
`NOBA (3-nitrosobenzamide)
`[92,93]. DIBA (2,2'-
`dithiobisbenzamide)
`[94,95]. SRR-S133 (cyclic 2,2‘-
`dithiobisbenzamidc)
`[96], dithiane (I.2-dithiane-4,S-
`diol,l,l-dioxide,eis) [97] and ADA (azodicarbonamide)
`[93,99]. These compounds should be able to interfere with
`both early (uncoating, disasscmbly) and late phases
`(packaging, assembly) of retrovirus replication. Their effect at
`the late p