Synthesis and anti-human immunodeficiency virus type 1 integrase ...

2 downloads 0 Views 2MB Size Report
1968)]; IR (KBr) 3280, 1600; 1H NMR (DMSO d6) δ 9.88. Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters. 499. Antiviral Chemistry & Chemotherapy 9: ...
Antiviral Chemistry & Chemotherapy 9: 497-509

Synthesis and anti-human immunodeficiency virus type 1 integrase activity of hydroxybenzoic and hydroxycinnamic acid flavon-3-yl esters N Desideri1*, I Sestili1, ML Stein1, E Tramontano2, S Corrias2 and P La Colla2* 1 Dipartimento di Studi Farmaceutici, Università ‘La Sapienza’ di Roma, P le Aldo Moro 5, Box 36 Roma 62, 00185 Roma, Italy 2 Dipartimento di Biologia Sperimentale, Università di Cagliari, V le Regina Margherita 45, 09124 Cagliari, Italy

*Corresponding authors: ND; Tel/Fax: +39 6 49 14 91; E-mail: [email protected]; PLC; Tel: +39 70 67 09 89; Fax: +39 70 67 08 02

A series of new hydroxybenzoic and hydroxycinnamic acid flavon-3-yl esters were synthesized in order to obtain compounds targeting the human immunodeficiency virus (HIV) type 1 integrase (IN). The esters were tested for anti-IN and anti-reverse transcriptase (RT) activity in enzyme assays and for anti-HIV-1, anti-proliferative and anti-topoisomerase activity in cell-based assays. In enzyme assays, the two gallic acid flavon-3-yl esters showed a notable IN inhibition (IC50 values were 8.3 and 9.1 µM, respectively), while the two caffeic acid

flavon-3-yl esters exhibited a modest activity (IC50 75 and 60 µM, respectively). Replacement of hydroxyl groups resulted in loss of potency. Caffeic acid 3′,4′-dichloroflavon-3-yl ester also inhibited the RT activity whereas it was not active on human topoisomerases. It therefore represents an interesting example of a compound specifically targeting more than one step of the virus replication cycle. Keywords: HIV-1; integrase inhibitors; flavonoids; caffeic acid esters; gallic acid esters.

Introduction Single-agent therapy with the currently available drugs has proved ineffective in the long-term treatment of human immunodeficiency virus (HIV) type 1 infection (Johnson, 1996). This failure is due to the development of drug-resistant virus, which arises within months or even weeks according to the drug used and the subsequent rebound of virus replication (De Clercq, 1994; Johnson, 1996). Combination therapy employing multiple agents has effectively reduced the occurrence of drug-resistant strains and provides long-term suppression of virus replication (Johnson, 1996). To date, most of the antiretroviral drugs target two viral enzymes; reverse transcriptase (RT) and protease (De Clercq, 1995). Therefore, a possible approach to improve further the efficacy of combination therapy is the development of new compounds that target additional and different steps of the virus replication cycle. The HIV-1 integrase (IN) is an attractive viral target for drug development as it is absolutely required for virus replication and it is absent in human cells (Kulkosky & Skalka, 1994; Andrake & Skalka, 1996). IN catalyses the integration of the viral dsDNA into the host chromosome in two ©1998 International Medical Press 0956-3202/98/$17.00

distinct steps. The first, a processing reaction, cuts the 3′ends of both strands, removing the dinucleotide GT; the second, a joining reaction, includes a staggered cleavage of the host DNA coupled to ligation with the 3′-processed ends of the viral DNA (Kulkosky & Skalka, 1994; Andrake & Skalka, 1996). Recently, several compounds have been shown to inhibit IN activity in enzyme assays (Pommier et al., 1997). Among them are flavones, such as quercetin and quercetagenin (Fesen et al., 1994; Raghavan et al., 1995), caffeic acid phenethyl ester (CAPE) and related amides (Fesen et al., 1994; Burke et al., 1995), curcumin (Mazunder et al., 1995a), tyrphostins (Mazunder et al., 1995b), bis-catechols such as β-conidedrol and 1,4-dicaffeoylquinic acid (La Femina et al., 1995; Robinson et al., 1996), diarylsulphones (Neamati et al., 1997a), 2-mercaptobenzenesulphonamides (Neamati et al., 1997b) and others (Pommier et al., 1997). The structures of a large number of active compounds share the presence of multiple aromatic rings with at least two vicinal hydroxyl groups (Pommier et al., 1997) (Figure 1). Unfortunately, 497

N Desideri et al.

Figure 1. Chemical structures of some HIV-1 IN inhibitors with a catechol moiety OH OH O HO

O

HO O

CAPE R

OH OH

HO

O OH

R= H: quercetin R= OH: quercetagenin

O HO N

OH

H HO

An amide related to CAPE O

O

A tyrphostin

HO CN

OH

N

N

H

H

CN OH

HO

HO

O

OH HO

COOH O O OH

HO

HO

O

1,4-Dicaffeoylquinic acid

O O H

H

OH

HO

β-Conidedrol

OH HO

the great majority of these compounds do not show specific IN activity in enzyme assays and selective HIV-1 inhibition in cell-based assays. Quercetin, for example, inhibits also the activities of RT from HIV-1 and Rauscher murine leukaemia virus (R-MuLV), human DNA polymerase β, terminal deoxynucleotidyltransferase (Ono et al., 1990), topoisomerase II (Austin et al., 1992) and tyrosine kinase (Levitzki & Gazit, 1995). In the last few years, our group has searched for new antiviral compounds among natural and synthetic flavonoids (Burali et al., 1987; Conti et al., 1988, 1990a,b, 1992; Superti et al., 1989; Desideri et al., 1990, 1992, 1995, 1997; Quaglia et al., 1991, 1992, 1993; Genovese et al., 1995). Various esters of gallic and caffeic acid have been obtained from plant sources; they are formed by

498

condensation with sugars, polyols, glycosides and other phenols such as flavonoids. Among them are epicatechin3-gallate and epigallocatechin-3-gallate, two components of Camellia sinensis (tea plant) which differentially inhibit the activity of viral RTs, eukaryotic DNA polymerases, Escherichia coli RNA polymerase (Nakane & Ono, 1990; Moore & Pizza, 1992) and topoisomerase II (Austin et al., 1992). The biological activity has been associated with the presence of the galloyl group (Nakane & Ono, 1990; Moore & Pizza, 1992), which has also been linked to the activity of tetragalloylquinic acid, a compound that inhibits the HIV-1 RT in vitro and HIV-1 multiplication in cell-based assays (Nishizawa et al., 1989) (Figure 2). In view of these observations and with the aim of obtaining synthetic compounds that target the HIV-1 IN, we syn-

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

Figure 2. Chemical structures of epicatechin-3gallate, epigallocatechin-3-gallate and 1,3,4,5-tetragalloylquinic acid (a)

Material and Methods: Chemistry Melting points were determined on a Büchi SMP-20 apparatus and are uncorrected. 1H NMR spectra were detected with a Bruker AM-200 instrument (using Me4Si as the internal standard). IR spectra were recorded on a Perkin-Elmer 1310 spectrophotometer. All compounds were routinely checked by thin-layer chromatography (TLC) and 1H NMR. TLC was performed using 0.25 mm silica gel or aluminium oxide fluorescent coated plates (Merck, Kieselgel or Aluminium oxide 60 F254). Components were visualized by UV light. Column chromatography was performed using silica gel (Carlo Erba; 0.05–0.20 mm) or aluminium oxide (Merck; 70–230 mesh). Elemental analyses were determined by the Microanalytical Laboratory of the University of Padova, Italy, and were within ±0.4% of theoretical values.

OH OH

HO

O R

O OH

OH O

OH

(b) OH

OH

HO

2′-Hydroxychalcones (1a–b) 2′-Hydroxyacetophenone (10 mmol) and substituted benzaldehyde (10 mmol) were dissolved in EtOH (50 ml) under stirring, and aqueous NaOH (60%, 30 ml) was added dropwise, in an ice bath. The reaction mixture was stirred at room temperature for 5 h, and acidified with 2 M HCl. The precipitate was filtered, washed with water, and crystallized from EtOH.

O HO O

O

HO

COOH O

HO

O

O

4-Chloro-2′-hydroxychalcone (1a)

OH HO O

HO

HO

OH OH

OH

Yield 61%; m.p. 147–148°C [lit. m.p. 150°C (Cheng et al., 1963)]; IR (KBr) 3200–2400, 1630 cm–1; 1H NMR (CDCl3) δ 12.70 (s, 1 H, OH), 7.95 (dd, 1 H, H6′, J5′–6′ 8.0 Hz, J4′–6′ 2.0 Hz), 7.90–7.25 (m, 7 H, Hα, Hβ, H2, H3, H5, H6, H4′), 7.20–6.80 (m, 2H, H3′, H5′).

(a) R= H; (–)-epicatechin-3-gallate; R= OH; (–)-epigallocatechin-3-gallate (b) 1,3,4,5-tetragalloylquinic acid

thesized a series of new derivatives by esterification of the hydroxyl group at position 3 of synthetic flavonoles with gallic acid and other naturally occurring hydroxybenzoic and hydroxycinnamic acids (such as syringic, caffeic, vanillic and ferulic acids). Owing to the poor selectivity of natural polyhydroxylated flavones such as quercetin (Ono et al., 1990; Austin et al., 1992; Levitzki & Gazit, 1995), we designed compounds without free hydroxyl groups in the flavone portion of the molecule. Chlorine atoms were introduced in positions 4′ and 3′ in order to study the impact on activity of the change in physicochemical parameters of the molecule. The new esters (3a–h, 4a–f, 4i–l, 6a–b, 7a–b, 8a–b) and the flavon-3-ols (2a–b) were tested for anti-IN and anti-RT activity in enzyme assays, and for anti-proliferative, anti-HIV-1 and anti-topoisomerase activity in cell-based assays.

Antiviral Chemistry & Chemotherapy 9:6

3,4-Dichloro-2′-hydroxychalcone (1b) Yield 62%; m.p. 151–152°C (Kallay et al., 1974); IR (KBr) 3100–2300, 1630 cm–1; 1H NMR (CDCl3) δ 12.67 (s, 1 H, OH), 7.90 (dd, 1 H, H6’, J5′–6′ 7.9 Hz, J4′–6′ 2.0 Hz), 7.85–7.41 (m, 6 H, Hα, Hβ, H2, H5, H6, H4’), 7.12–6.89 (m, 2H, H3′, H5′).

3-Hydroxyflavones (2a–b) 2′-Hydroxychalcone 1a–b (10 mmol) was dissolved in a solution of MeOH (100 ml) and NaOH (5%, 35 ml). H2O2 solution (35%, 4 ml) was added to the stirred solution kept at 0–5°C. The solution was stirred for 2 h at 0–5°C and then for 16 h at room temperature. The reaction mixture was diluted with water and acidified with 2 M HCl. The precipitate was filtered, washed with water and crystallized from EtOH.

4′-Chloro-3-hydroxyflavone (2a) Yield 66%; m.p. 200–201°C [lit. 203–205°C (Smith et al., 1968)]; IR (KBr) 3280, 1600; 1H NMR (DMSO d6) δ 9.88

499

N Desideri et al.

(bs, 1 H, OH), 8.26 (d, 2 H, H2′, H6′, J2′–3′ 8.7 Hz), 8.12 (d, 1 H, H5, J5–6 7.7 Hz), 7.89–7.71 (m, 2H, H7, H8), 7.65 (d, 2 H, H3′, H5′), 7.48 (t, 1 H, H6, J5–6 J6–7 7.7 Hz).

Hz), 7.57 (d, 1 H, H8, J7–8 8.5 Hz), 7.52–7.39 (m, 5 H, H6, H3′, H5′, 2Hsir), 3.89 (s, 6 H, 2 OCH3), 2.36 (s, 3 H, COCH3); Anal. (C26H19ClO8) C, H, Cl.

3′,4′-Dichloro-3-hydroxyflavone (2b)

Acetylsyringic acid 3′,4′-dichloroflavon-3-yl ester (3d)

Yield 53%; m.p. 171–176°C; IR (KBr) 3270, 1600 cm–1; 1 H NMR (DMSO d6) δ 10.14 (s, 1 H, OH), 8.42 (s, 1 H, H2′), 8.18 (d, 1 H, H6′, J5′–6′ 8.2 Hz), 8.09 (d, 1 H, H5, J5–6 7.7 Hz), 7.96–7.70 (m, 3 H, H7, H8, H5′), 7.40 (t, 1 H, H6, J5–6 J6–7 7.7 Hz); Anal. (C15H8Cl2O3) C, H, Cl.

Preparation of esters (3a–h, 7a–b): general procedure Dry pyridine (0.8 ml) was added to a suspension of 3hydroxyflavone 2a or 2b (10 mmol) and the appropriate acid chloride (10 mmol) in dry CH2Cl2 (200 ml). The reaction mixture was stirred at 50°C for 5 h. After cooling, the precipitate was filtered off, and the filtrate was washed with diluted HCl and water. The organic layer was dried over Na2SO4, and the solvent was removed under reduced pressure. The residue was chromatographed on aluminium oxide eluting with AcOEt and crystallized.

Acetylvanillic acid 4’-chloroflavon-3-yl ester (3a) Prepared from 4′-chloro-3-hydroxyflavone (2a) and Oacetylvanilloyl chloride (Deulofeu & Schopflocher, 1953). Yield 43%; m.p. 164–165°C from AcOEt; IR (KBr) 1755, 1730, 1640 cm–1; 1H NMR (CDCl3) δ 8.25 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.93–7.66 (m, 5 H, H2′, H6′, H7, H2van, H6van), 7.66–7.40 (m, 4 H, H3′, H5′, H6, H8), 7.17 (d, 1 H, H5van, J5–6van 8.3 Hz), 3.90 (s, 3 H, OCH3), 2.30 (s, 3 H, COCH3); Anal. (C25H17ClO7) C, H, Cl.

Acetylvanillic acid 3′,4′-dichloroflavon-3-yl ester (3b) Prepared from 3′,4′-dichloro-3-hydroxyflavone (2b) and Oacetylvanilloyl chloride (Deulofeu & Schopflocher, 1953). Yield 34%; m.p. 149–151°C from AcOEt/petroleum ether; IR (KBr) 1750, 1725, 1630 cm–1; 1H NMR (CDCl3) δ 8.27 (dd, 1 H, H5, J5–6 8.0 Hz, J5′7 1.5 Hz), 8.07 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.90–7.70 (m, 4 H, H2van, H6van, H6′, H7), 7.61 (d, 1 H, H8, J7–8 8.5 Hz), 7.55 (d, 1 H, H5′, J5′–6′ 8.5 Hz), 7.48 (m, 1 H, H6, J5–6 8.0 Hz, J6–7 7.0 Hz, J6–8 0.9 Hz), 7.2 (d, 1 H, H5van, J5–6van 8.2 Hz), 3.9 (s, 3 H, OCH3), 2.36 (s, 3 H, COCH3); Anal. (C25H16Cl2O7) C, H, Cl.

Acetylsyringic acid 4′-chloroflavon-3-yl ester (3c) Prepared from 4′-chloro-3-hydroxyflavone (2a) and Oacetylsyringoyl chloride (Freudenberg & Hüber, 1952). Yield 35%; m.p. 198–200°C from CH3CN; IR (KBr) 1750, 1720, 1640 cm–1; 1H NMR (CDCl3) δ 8.26 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.86 (m, 2H, H2′, H6′, J2′–3′ 8.7 Hz), 7.73 (m, 1 H, H7, J5′7 1.5 Hz, J6–7 7.0 Hz, J7–8 8.5

500

Prepared from 3′,4′-dichloro-3-hydroxyflavone (2b) and O-acetylsyringoyl chloride (Freudenberg & Hüber, 1952). Yield 42%; m.p. 224–225°C from AcOEt; IR (KBr) 1750, 1720, 1640 cm–1; 1H NMR (CDCl3) δ 8.28 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.07 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.80–7.68 (m, 2H, H7, H6’), 7.65–7.30 (m, 5 H, H6, H8, H5′, 2 Hsir), 3.90 (s, 6 H, 2 OCH3), 2.35 (s, 3 H, COCH3); Anal. (C26H18Cl2O8) C, H, Cl.

Acetylferulic acid 4′-chloroflavon-3-yl ester (3e) Prepared from 4′-chloro-3-hydroxyflavone (2a) and Oacetylferuloyl chloride (Allen & Byers, 1949). Yield 66%; m.p. 200–201°C from AcOEt/EtOH; IR (KBr) 1750, 1715, 1630 cm–1; 1H NMR (CDCl3) δ 8.30 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.00–7.78 (m, 3H, H2′, H6′, Hβ), 7.80–7.35 (m, 5 H, H6–8, H3′, H5′), 7.25–7.00 (m, 3 H, H2fer, H5fer, H6fer), 6.63 (d, 1 H, Hα, Jα–β 15.8 Hz), 3.90 (s, 3 H, OCH3), 2.30 (s, 3 H, COCH3); Anal. (C27H19ClO7) C, H, Cl.

Acetylferulic acid 3′,4′-dichloroflavon-3-yl ester (3f) Prepared from 3′,4′-dichloro-3-hydroxyflavone (2b) and O-acetylferuloyl chloride (Allen & Byers, 1949). Yield 43%; m.p. 174–176°C from AcOEt; IR (KBr) 1750, 1715, 1645 cm–1; 1H NMR (CDCl3) δ 8.30 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.10 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.90 (d, 1 H Hβ, Jα–β 15.8 Hz), 7.80–7.30 (m, 5 H, H6–8, H5′, H6′), 7.25–7.00 (m, 3 H, H2fer, H5fer, H6fer), 6.65 (d, 1 H, Hα, Jα–β 15.8 Hz), 3.90 (s, 3 H, OCH3), 2.30 (s, 3 H, COCH3); Anal. (C27H18Cl2O7) C, H, Cl.

Diacetylcaffeic acid 4′-chloroflavon-3-yl ester (3g) Prepared from 4′-chloro-3-hydroxyflavone (2a) and Odiacetylcaffeoyl chloride (Freudenberg & Heel, 1953). Yield 55%; m.p. 205–208°C from AcOEt/petroleum ether; IR (KBr) 1755, 1710, 1630 cm–1; 1H NMR (CDCl3) δ 8.26 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.90–7.65 (m, 4 H, H2′, H6′, Hβ, H7), 7.60–7.40 (m, 6 H, H3′, H5′, H6, H8, H2caff, H6caff), 7.25 (d, 1 H, H5caff, J5–6caff 8.2 Hz), 6.60 (d, 1 H, Hα, Jα–β 15.9 Hz), 2.29 (s, 6 H, 2COCH3); Anal. (C28H19ClO8) C, H, Cl.

Diacetylcaffeic acid 3′,4′-dichloroflavon-3-yl ester (3h) Prepared from 3′,4′-dichloro-3-hydroxyflavone (2b) and O-diacetylcaffeoyl chloride (Freudenberg & Heel, 1953).

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

Yield 59%; m.p. 193–195°C from AcOEt; IR (KBr) 1760, 1715, 1630 cm–1; 1H NMR (CDCl3) δ 8.26 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.01 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.83 (d, 1 H, Hβ, Jα–β 16 Hz), 7.77–7.68 (m, 2 H, H6′, H7), 7.62–7.38 (m, 5 H, H5′, H2caff, H6caff, H6, H8), 7.25 (d, 1 H, H5caff, J5–6caff 8.2Hz), 6.60 (d, 1 H, Hα, Jα–β 16 Hz), 2.30 (s, 6 H, 2 COCH3); Anal. (C28H18Cl2O8) C, H, Cl.

3-Acetoxy-4′-chloroflavone (6a) Obtained from the reaction of 4′-chloro-3-hydroxyflavone (2a) with triacetylgalloyl chloride (Freudenberg & Hüber, 1952). Yield 56%; m.p. 136–137°C from AcOEt/petroleum ether; IR (KBr) 1755, 1630 cm–1; 1H NMR (CDCl3) δ 8.32 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.85 (d, 2 H, H2′, H6′, J2′–3′ 8.7 Hz), 7.80–7.30 (m, 5 H, H5′, H6′, H6–8), 2.30 (s, 3 H, COCH3); Anal. (C17H11ClO4) C, H, Cl.

3-Acetoxy-3′,4′-dichloroflavone (6b) Obtained from the reaction of 3′,4′-dichloro-3-hydroxyflavone (2b) with triacetylgalloyl chloride (Freudenberg & Hüber, 1952). Yield 53%; m.p. 126–127°C from AcOEt/petroleum ether; IR (KBr) 1760, 1635 cm–1; 1H NMR (CDCl3) δ 8.30 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.00 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.90–7.30 (m, 5 H, H5′, H6′, H6–8), 2.30 (s, 3 H, COCH3); Anal. (C17H10Cl2O4) C, H, Cl.

Tribenzylgallic acid 4′-chloroflavon-3-yl ester (7a) Prepared from 4′-chloro-3-hydroxyflavone (2a) and tribenzylgalloyl chloride (Schmidt & Schach, 1950). Yield 47%; m.p. 174–176°C from CH3CN; IR (KBr) 1720, 1645 cm–1; 1H NMR (CDCl3) δ 8.27 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1 .5 Hz), 7.83–7.68 (m, 3 H, H2′, H6′, H7), 7.57 (dd, 1 H, H8, J6–8 0.8 Hz, J7–8 8.5 Hz), 7.51 (s, 2 H, 2 Hgall), 7.50–7.23 (m, 18 H, H3′, H5′, H6, 15 Hbz), 5.15 (s, 6 H, 3 OCH2); Anal. (C43H31ClO7) C, H, Cl.

Tribenzylgallic acid 3′,4′-dichloroflavon-3-yl ester (7b)

in an ice bath. The precipitate was filtered, washed with water and crystallized.

Vanillic acid 4′-chloroflavon-3-yl ester (4a) Yield 64%; m.p. 175–176°C from benzene; IR (KBr) 3600–3100, 1720, 1630 cm–1; 1H NMR (CDCl3) δ 8.25 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.89 (d, 2 H, H2′, H6′, J2′–3′ 8.7 Hz), 7.81–7.65 (m, 2 H, H6van, H7), 7.60 (d, 1 H, H2van, J2–6van 1.9 Hz), 7.55 (d, 1 H, H8, J7–8 8.6 Hz), 7.50–7.38 (m, 3 H, H3′, H5′, H6), 6.96 (d, 1 H, H5van, J5–6van 8.3 Hz), 6.41 (s, 1 H, OH), 3.90 (s, 3 H, OCH3); Anal. (C23H15ClO6) C, H, Cl.

Vanillic acid 3′,4′-dichloroflavon-3-yl ester (4b) Yield 60%; m.p. 187–188°C from benzene; IR (KBr) 3500–3100, 1720, 1630 cm–1; 1H NMR (CDCl3) δ 8.28 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.08 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.83–7.70 (m, 3 H, H6van, H7, H6′), 7.65 (d, 1 H, H2van, J2–6van 1.9 Hz), 7.60 (d, 1 H, H8, J7–8 8.6 Hz), 7.55–738 (m, 2 H, H5′, H6), 7.00 (d, 1 H, H5van, J5–6van 8.3 Hz), 6.20 (s, 1 H, OH), 3.96 (s, 3 H, OCH3); Anal. (C23H14Cl2O6) C, H, Cl.

Syringic acid 4′-chloroflavon-3-yl ester (4c) Yield 67%; m.p. 215–218°C from benzene; IR (KBr) 3600–3200, 1720, 1630 cm–1; 1H NMR (CDCl3) δ 8.26 (dd, 1 H H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 7.87 (d, 2 H, H2′, H6′, J2′–3′ 8.7 Hz), 7.73 (m, 1 H, H7, J5–7 1.5 Hz, J6–7 7.0 Hz, J7–8 8.6 Hz), 7.58 (d, 1 H, H8, J7–8 8.6 Hz), 7.50–7.35 (m, 5 H, H6, H3′, H5′, 2 Hsir), 6.12 (s, 1 H, OH), 3.94 (s, 6 H, 2 OCH3); Anal. (C24H17ClO7) C, H, Cl.

Syringic acid 3′,4′-dichloroflavon-3-yl ester (4d) Yield 43%; m.p. 203–205°C from benzene; IR (KBr) 3520–3300, 1720, 1640 cm–1; 1H NMR (CDCl3) d 8.28 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.10 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.82–7.70 (m, 2 H, H7, H6′), 7.65–7.43 (m, 5 H, H6, H8, H5′, 2 Hsir), 6.10 (s, 1 H, OH), 3.97 (s, 6 H, 2 OCH3); Anal (C24H16Cl2O7) C, H, Cl.

Ferulic acid 4′-chloroflavon-3-yl ester (4e)

Prepared from 3′,4′-dichloro-3-hydroxyflavone (2b) and tribenzylgalloyl chloride (Schmidt & Schach, 1950). Yield 48%; m.p. 197–200°C from CH3CN; IR (KBr) 1720, 1640 cm–1; 1H NMR (CDCl3) δ 8.26 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.04 (d, 1 H, H2′, J2′–6′ 2.0 Hz), 7.80–7.58 (m, 3 H, H7, H6′, H8), 7.53 (s, 2 H, 2 Hgall), 7.50–7.20 (m, 17 H, H5′, H6, 15Hbz), 5.15 (s, 6 H, 3 OCH2); Anal. (C43H30Cl2O7) C, H, Cl.

Yield 60%; m.p. 163–165°C from EtOH; IR (KBr) 3600–3000, 1715, 1620 cm–1; 1H NMR (acetone d6) δ 9.90–9.50 (bs, 1 H, OH), 8.30–7.50 (m, 9 H, H5–8, H2′, H3′, H5′, H6′, Hβ), 7.48 (d, 1 H, H2fer, J2fer–6fer 1.8 Hz), 7.27 (dd, 1 H, H6fer, J2fer–6fer 1.8 Hz, J5fer–6fer 8.2 Hz), 6.90 (d, 1 H, H5fer, J5fer–6fer 8.2 Hz), 6.65 (d, 1 H, Hα, Jα–β 15.8 Hz), 3.95 (s, 3 H, OCH3); Anal. (C25H17ClO6) C, H, Cl.

Hydrolysis to the esters 4a–f,i,l

Ferulic acid 3′,4′-dichloroflavon-3-yl ester (4f)

1M NaOH (50 ml) was added to a solution of ester (3a–h, 1 mmol) in acetone (50 ml). The mixture was stirred at room temperature for 20 min, and acidified with 2 M HCl

Yield 46%; m.p. 162–167°C from benzene; IR (KBr) 3500–3100, 1720, 1640 cm–1; 1H NMR (DMSO d6) δ 10.0–9.60 (bs, 1 H, OH), 8.25–8.05 (m, 2 H, H5, H2′),

Antiviral Chemistry & Chemotherapy 9:6

501

N Desideri et al.

7.95–7.70 (m, 5 H, H7, H8, H5′, H6′, Hβ), 7.60–7.45 (dd, 1 H, H6, J5–6 8.0 Hz, J6–7 7.0 Hz), 7.35 (d, 1 H, H2fer, J2fer–6fer 1.8 Hz), 7.15 (dd, 1 H, H6fer, J2fer–6fer 1.8 Hz, J5fer–6fer 8.2 Hz), 6.82 (d, 1 H, H5fer, J5fer–6fer 8.2 Hz), 6.65 (d, 1 H, Hα, Jα–β 15.9 Hz), 3.90 (s, 3 H, OCH3); Anal. (C25H16Cl2O6) C, H, Cl.

Material and Methods: Virology Compounds Stock solutions were prepared in DMSO at 10 mM concentration and kept at –20°C.

IN and DNA substrates Caffeic acid 4′-chloroflavon-3-yl ester (4i) Yield 55%; m.p. 223–225°C from benzene; IR (KBr) 3600–3000, 1710, 1620 cm–1; 1H NMR (DMSO d6) δ 9.90–9.10 (bs, 2 H, 2 OH), 8.12 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.00–7.80 (m, 4 H, H2′, H6′, H7, H8), 7.80–7.62 (m, 3 H, Hβ, H3′, H5′), 7.57 (m, 1 H, H6, J5–6 8.0 Hz, J6–7 7.0 Hz, J6–8 1.0 Hz), 7.36 (d, 1 H, H2caff, J2caff–6caff 1.4 Hz), 7.15 (dd, 1 H, H6caff, J5–6caff 8.2 Hz, J2–6caff 1.4 Hz), 6.83 (d, 1 H, H5caff, J5–6caff 8.2 Hz), 6.55 (d, 1 H, Hα, Jα–β 15.9 Hz); Anal. (C24H15ClO6) C, H, Cl.

Caffeic acid 3′,4′-dichloroflavon-3-yl ester (4l) Yield 58%; m.p. 203–205°C from benzene; IR (KBr) 3500–3000, 1705, 1625 cm–1; 1H NMR (DMSO d6) δ 10.10–9.10 (bs, 2 H, 2 OH), 8.30–7.50 (m, 8 H, H5–8, H2′, H5′, H6′, Hβ), 7.30–7.05 (m, 2 H, H2caff, H6caff), 6.80 (d, 1 H, H5caff, J5–6caff 8.2 Hz), 6.55 (d, 1 H ,Hα, Jα–β 15.9 Hz); Anal. (C24H14Cl2O6) C, H, Cl.

Expression of the IN protein with an N-terminal polyhistidine tag was obtained by isopropyl β-D-thiogalactopyranoside (IPTG) induction of the E. coli strain BL21(DE3) (Novagen) containing the pINSD.His vector. Protein purification was carried out following the Novagen procedure, except for the presence of 5 mM CHAPS in binding, washing and elution buffer. The following oligonucleosides representing the terminal 21 nucleotides of the HIV-1 U5 LTR were used in this study: B: 5′-ACTGCTAGAGATTTTCCACAC-3′ (minus strand); C: 5′ GTGTGGAAAATCTCTAGCA-3′. For standard 3′-processing assays, C was annealed with B in 0.1 M NaCl by heating at 80°C and slowly cooling to room temperature overnight. This double-stranded substrate was labelled by introducing at the 3′ end of C the two missing nucleotides using [a-32P]dGTP, unradiolabelled dTTP and Klenow polymerase. Unincorporated [α-32P]dGTP was separated from the duplex substrate by two consecutive runs through G-25 Sephadex quick spin columns.

Gallic acid flavon-3-yl esters (8a–b) A solution of ester (7a–b, 1 mmol) in THF (30 ml) was hydrogenated at room temperature over 10% Pd/C (100 mg) until absorbtion of the theoretical hydrogen amount (4.5 h). After filtration of the catalyst and evaporation of the solvent, the residue oil was mixed with CH2Cl2; the resulting solid was filtered and crystallized.

Gallic acid 4′-chloroflavon-3-yl ester (8a) Yield 54%; m.p. 217–219°C from AcOEt; IR (KBr) 3600–3000, 1720, 1610 cm–1; 1H NMR (acetone d6) δ 8.40–8.28 (bs, 3 H, 3 OH), 8.18 (dd, 1 H, H5, J5–6 8.0 Hz, J5–7 1.5 Hz), 8.05 (d, 2 H, H2′, H6′, J2′–3′ 8.8 Hz), 7.89 (m, 1 H, H7, J 5–7 1.5 Hz, J6–7 7.0 Hz, J7–8 8.6 Hz), 7.78 (dd, 1 H, H8, J6–8 0.8 Hz, J7–8 8.6 Hz), 7.67–7.50 (m, 3 H, H3–, H5′, H6,), 7.27 (s, 2 H, 2 Hgall); Anal. (C22H13ClO7) C, H, Cl.

3′ Processing assay Standard reaction conditions were: 40 mM NaCl, 10 mM MnCl2, 25 mM Tris–HCl pH 7.5, 1 mM DTT, 2% glycerol, 1 nM duplex B:C labelled at the 3′ end and 5 nM IN (considered as monomer). Incubation was carried out at 37°C for 30 min in a volume of 50 µl. The reaction was stopped by adding an equal volume of 0.5 M Na2HPO4. Each sample (90 µl) was added to a well of a Multiscreen 96-well filtration plate pre-loaded with 100 µl DEAE Sephacel slurry which had been vacuum-dried. Vacuum was applied after incubation at room temperature for 10 min. The uncleaved, labelled substrate was retained in the resin, whereas labelled dinucleotide (GT) products were collected at the bottom of 96-well filtration plates and counted for radioactivity.

RT assays Gallic acid 3′,4′-dichloroflavon-3-yl ester (8b) Yield 59%; m.p. 219–221°C from CH3CN; IR (KBr) 3600–3000, 1715, 1615 cm–1; 1H NMR (acetone d6) d 8.59–8.29 (bs, 3 H, 3 OH), 8.24 (d, 1 H, H2′, J2′–6′ 1.9 Hz), 8.17 (dd, 1 H, H5, J5–6 7.9 Hz, J5–7 1 .5 Hz), 8.00 (dd, 1 H, H6′, J2′–6′ 1.9 Hz, J5′–6′ 8.5 Hz), 7.96–7.72 (m, 3 H, H7, H8, H5′), 7.56 (m, 1 H, H6, J5–6 7.9 Hz, J6–7 7.0 Hz, J6–8 0.9 Hz), 7.30 (s, 2 H, 2 H gall); Anal (C22H12Cl2O7) C, H, Cl.

502

Purified recombinant RT was assayed as previously described (Tramontano & Cheng, 1992) in a 50 µl volume containing 50 mM Tris–HCl pH 7.8, 80 mM KCl, 6 mM MgCl2, 1 mM DTT, 0.1 mg/ml BSA, 0.3 OD/ml poly(rC).oligo(dG)12–18, 10 µM [3H]dGTP (1 Ci/mmol).

Biological assays MT4 and KB cells were grown at 37°C in a 5% CO2 atmosphere in RPMI 1640 medium supplemented with

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

Figure 3. Synthesis of flavon-3-yl esters (3a–h, 4a–f i, l) 1 a–b

R OH

NaOH OH

Cl

Cl

+ CH 3

H R

r.t., EtOH O

H2O2, NaOH, MeOH

O

O

R' CH3COO R

Cl Cl

(

R''

)n

COCl O R

O

CH2Cl2, Py, 50°C

OH

O O

C

)n

R'

3 a–h NaOH, CH3COCH3, r.t., 20 min

3,4 a 3,4 b 3,4 c 3,4 d 3,4 e 3,4 f 3g 3h 4i 4l

CH3COO

R''

R Cl

O

O O

2 a–b

O

( O

C

( O

n 0 0 0 0 1 1 1 1 1 1

R H Cl H Cl H Cl H Cl H Cl

R′ OCH3 OCH3 OCH3 OCH3 OCH3 OCH3 OCOCH3 OCOCH3 OH OH

R′′ H H OCH3 OCH3 H H H H H H

)n

R'

4 a–f, i, l

R''

OH

5% fetal calf serum (FCS), 100 IU/ml penicillin G and 100 µg/ml streptomycin. Cell cultures were checked periodically for the absence of mycoplasma contamination with a MycoTect Kit (Gibco). HIV-1 stock solutions had titres of 4.5×106 50% cell culture infectious dose (CCID50)/ml. Cytotoxicity evaluation was based on the viability of mockinfected cells, as monitored by the 3-(4,5-dimethylthiazol1-yl)-2,5-diphenyltetrazolium bromide (MTT) method (Pauwels et al., 1988). Activity of the compounds against HIV-1 multiplication in acutely infected cells was based on inhibition of virus-induced cytopathicity in MT4 cells and was determined by the MTT method.

Antiviral Chemistry & Chemotherapy 9:6

Quantification of protein-linked DNA breaks A modified in vivo K-SDS coprecipitation assay (Lee et al., 1989) was used for quantification of protein-linked DNA breaks (PLDB) levels. Briefly, KB cells were seeded at a density of 2.5×105/ml and labelled with [14C]thymidine (0.75 µCi/ml) for 48 h. Monolayers were washed twice and, after trypsinization, cells were resuspended in aliquots of 3×105/ml and were incubated for 1 h at 37°C. Then, duplicate samples were treated with test drugs at the indicated concentrations and further incubated for 1 h. Cells were collected by centrifugation at 2000 g for 10 min, resuspended in 1 ml of warm lysis buffer

503

N Desideri et al.

Figure 4. Synthesis of gallic acid flavon-3-yl esters (8 a–b)

OCOCH3 Cl

2 a–b

R

OCOCH3

C Cl

O

R OCOCH3

Cl

O

5 a–b

O OH O

COCl O

O

C

OAc

O

BzO

OAc

OBz OAc

+

OBz

R

R

Cl

Cl

6 a–b O

O

7 a–b O O

C

OCOCH3

OBz O

O

H2 Pd/C a.p.; r.t.

OBz OBz

2, 5, 6, 7, 8 a R= H 2, 5, 6, 7, 8 b R= Cl

R Cl

O

O O

C

OH

8 a–b

O

OH OH

(1.25% SDS, 5 mM EDTA, 0.4 mg/ml salmon sperm DNA) and the viscous cell lysates were sheared by passage through a 22 gauge needle five times. After 10 min incubation at 65°C, KCl was added to a final concentration of 100 mM, samples were chilled in an ice bath for 10 min and centrifuged at 3500 g for 10 min at 4°C. Pellets were resuspended in 1 ml of warm washing buffer (10 mM Tris–HCl pH 8.0, 100 mM KCl, 1 mM EDTA and 0.1 mg/ml salmon sperm DNA) and again incubated for 10

504

min at 65°C, chilled in ice bath and centrifuged as above. After a second washing step, pellets were solubilized at 65°C in 400 µl of water and counted for radioactivity.

Results Chemistry 3-Hydroxyflavones (2a–b) were prepared by commonly used synthetic methods. As shown in Figure 3, 2′-hydrox-

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

Table 1. Cytotoxicity and inhibition of HIV-1 IN, RT and topoisomerase of 3-hydroxyflavones (2a–b) and esters (3a–h; 4a–f, i, l; 6a–b; 7a–b; 8a–b) Compound

n

R

R’

R”

CC50 (µM)*

IC50 IN (µM)†

IC50 rRT (µM)‡

PBLD (fold)§

2a H 9.4±2 >100 >100 1.0 2b Cl 4.2 ±1 >100 >100 1.3 3a 0 H OCH3 H 10.4±2 >100 >100 1.2 3b 0 Cl OCH3 H >100 >100 >100 1.2 3c 0 H OCH3 OCH3 6.5±1 >100 >100 1.3 3d 0 Cl OCH3 OCH3 64.0±9 >100 >100 1.1 3e 1 H OCH3 H 19.0±3 >100 >100 1.5 3f 1 Cl OCH3 H >100 ND§ ND ND 3g 1 H OCOCH3 H 57.0±3 >100 >100 1.4 3h 1 Cl OCOCH3 H 20.0±2 >100 >100 1.4 4a 0 H OCH3 H 20.0±3 >100 >100 1.4 4b 0 Cl OCH3 H 4.7 ±1 >100 >100 1.1 H OCH3 >100 >100 1.5 4c 0 OCH3 26.0±4 4d 0 Cl OCH3 OCH3 7.1±1 >100 >100 1.3 4e 1 H OCH3 H 4.4±1 >100 >100 0.9 4f 1 Cl OCH3 H 2.1±1 >100 >100 1.3 H OH H 3.7±1 75±8 >100 1.1 4i 1 4l 1 Cl OH H 6.0±2 60±10 50±10 0.9 6a H 6.5±1 >100 >100 1.1 6b Cl 6.4±1 >100 >100 1.0 7a H >100 >100 100±10 ND 7b Cl >100 >100 >100 ND 8a H 6.7±1 8.3±2.0 100±5 1.2 8b Cl 4.5±1 9.1±1.3 >100 0.8 Quercetin 25.0±1 1.6±0.9 >100 4.0 Etoposide 0.4 >100 >100 4.1 Camptothecin 0.01 >100 >100 3.0 Nevirapine ND >100 0.19±0.02 ND * Compound concentration required to reduce the viability of KB cells by 50% as determined by the MTT method. † Inhibitor concentration required to reduce by 50% the 3’-processing activity of HIV-1 IN activity. Data represent the average ± SD of three independent determinations. ‡ Inhibitor concentration required to reduce HIV-1 RT activity by 50%. Data represent the average ± SD of three independent determinations. § Fold increase in PLDB as compared with untreated control, compound concentration was 200 µM with the exception of etoposide (10 µM) and camptothecin (1 µM). Data represent average of three independent determinations. ¶ Not done

ychalcones (1a–b) were obtained from 2-hydroxyacetophenone and the appropriate substituted benzaldehyde in alkaline medium. These chalcones were cyclized with alkaline hydrogen peroxide to 3-hydroxyflavones (2a–b). The acylation of 3-hydroxyl group with acetoxybenzoylchlorides or acetoxycinnamoylchlorides was carried out by mild heating in dichloromethane in the presence of a catalytic amount of pyridine. In these conditions, the esterification was partial but no side products were formed, so that esters 3a–b were easily purified by column chromatography. The desacetylation was performed under mild conditions to avoid hydrolysis of the other ester function (Figure 3). The esterification of 2a–b with triacetylgalloylchloride gave a mixture of 3-galloyloxyflavones (5a–b) and 3-acetoxyflavones (6a–b). Our attempts to separate the mixture allowed only the purification of 3-acetoxyflavones, whereas 3-galloyloxyflavones were hydrolysed or not obtained as pure substances. However, the galloylesters (8a–b) were satisfactorily prepared using benzylether protection of the

Antiviral Chemistry & Chemotherapy 9:6

hydroxyls; debenzylation was performed by catalytic hydrogenation (Figure 4).

Inhibition of HIV-1 integrase It is already known that (i) many compounds active against IN present multiple aromatic rings with polyhydroxylation, frequently in the 1,2-catechol arrangement (Pommier et al., 1997); (ii) the flavonoid structure, with various substitutions, has led to compounds that inhibit many biological activities (Nakane & Ono, 1990; Ono et al., 1990; Austin et al., 1992; Moore & Pizza, 1992; Levitzki & Gazit, 1995), often simultaneously; (iii) some galloyl derivatives showed anti-HIV-1 activity in a cell-based assay. In view of these observations we synthesized 4′-chloro flavon-3-oles esterified with hydroxybenzoic and hydroxycinnamic acids and addressed (i) the requirement of the catechol unit for the anti-IN activity of these derivatives; (ii) the impact of the increase of the molecule lipophilicity with the introduction of another chlorine atom in position

505

N Desideri et al.

Figure 5. Concentration–response curve for inhibition of the HIV-1 IN 3′processing of activity by Quercetin and flavonoles 4i, 4l, 8a, 8b

3’-processing activity (% of control)

120

100 80

60

Inhibition of HIV-1 RT

40

20 0 0.1

10 1 Compound concentration (µM)

100

Experiments were performed as described in Methods. Values represent the average ± SD of three independent determinations: Quercetin (q); 4i (q); 4l (◊); 8a (O) and 8b ( ).

3′; (iii) the correlation of the anti-IN activity with the antiRT and -topoisomerases activities, i.e. the specificity of the IN inhibition. For this purpose, the effects of the test compounds on the 3′ processing reaction was first measured using a recombinant HIV-1 IN expressed in E. coli and as substrate, a 21mer double-stranded oligonucleotide corresponding to the U5 end of the HIV-1 LTR, labelled at the 3′ end of the plus strand. This allowed the rapid and quantitative recovery of the acid soluble dinucleotides released by IN in the 3′-processing reaction. In agreement with other studies (Fesen et al., 1994), quercetin, used as a reference compound, inhibited the 3′processing activity of the HIV-1 IN with an IC50 value of 1.6 µM (Table 1). The inhibition of IN activity by flavonoles (2a–b) and esters (3a–h, 4a–f, i, l, 6a–b, 7a–b, 8a–b) was measured at 100 µM and the IC50 values were determined only for compounds that exibited more than 50% inhibition at 100 µM. As shown in Table 1, while the flavonoles 2a and 2b, and their acetyl derivatives 6a and 6b were inactive against IN, the two esters of caffeic acid 4i and 4l, with ortho-bis hydroxyl substitution, showed a weak inhibition of the IN activity (IC50 values were 75 and 60 µM, respectively) (Figure 5). A notable anti-IN activity was observed for the two esters of gallic acid 8a and 8b, which contain three vicinal hydroxyl groups (IC50 values were 8.3 and 9.1 µM, respectively). Introduction of another chlorine atom (derivatives 4l and 8b) did not increase the anti-IN activity shown by the parent monochlorine compounds 4i and 8a, respectively. Replacement of hydroxyl groups in positions 3 and 5 of gallic acid esters 8a and 8b with either two methoxyl groups

506

(compounds 4c and 4d) or one hydrogen and one methoxyl group (compounds 4a and 4b) resulted in loss of activity, as did the replacement of 3-hydroxyl group (compounds 4e and 4f) of caffeic acid esters 4i and 4l. These data clearly demonstrate that the presence at least of two hydroxyl groups on an aromatic ring was necessary for IN inhibition. Therefore, owing to the protection of the hydroxyls, all the intermediates in the synthesis (3a–h; 7a–b) were inactive.

Since both flavone and galloyl acid derivatives have been shown to inhibit the HIV-1 RT (Nakane & Ono, 1990; Ono et al., 1990; Moore & Pizza, 1992; Nishizawa et al., 1989), we evaluated the effect of the test compounds on this enzyme, using poly(rC).oligo(dG)12–18 and dGTP as reaction substrates and nevirapine as the reference compound. Among the test compounds only derivative 4l inhibited the RT activity with an IC50 value (50 µM) that is comparable to the IC50 value observed for IN (60 µM) inhibition (Table 1). Under these experimental conditions, quercetin was not able to inhibit HIV-1 RT. However, since in other studies (Ono et al., 1990) the reaction substrates used were poly(rA).oligo(dT)10 and dTTP, we assayed the HIV-1 RT activity in these conditions. Using poly(rA).oligo(dT)10 and dTTP as reaction substrates, quercetin inhibited the HIV-1 RT; however, the IC50 value (25 µM) was higher than the one previously reported (0.5 µM) (Ono et al., 1990). Out of the test compounds 4i, 4l, 8a and 8b, only 4l was active (its IC50 value was 80 µM). It is also worth noting that the anti-RT activity of compound 4l did not correlate with the presence of two vicinal hydroxyl groups on an aromatic ring since the strictly related compound 4i, lacking a chlorine atom in position 3′, as well as the two galloyl esters (8a–b), did not possess any anti-RT activity.

Antiretroviral effect Owing to the fact that some galloyl acid derivatives have been reported to be active against HIV-1 multiplication in cell-based assays, we measured the ability of the test compounds to selectively inhibit the HIV-1-induced cytopathogenicity in de novo-infected MT4 cells. However, in conditions where nevirapine and AZT (reference compounds), showed EC50 values of 0.01 and 0.25 µM, respectively, none of them were able to prevent the virus-induced cytopathogenic effect at concentrations below those cytotoxic for MT4 cells (data not shown).

Anti-proliferative and anti-topoisomerase activity Several flavonoids have been shown to inhibit cell proliferation (Ono et al., 1990; Austin et al., 1992; Levitzki & Gazit, 1995), therefore we evaluated the ability of the test

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

compounds to inhibit proliferation of the epithelial cell line KB. None of the substituted esters significantly increased the anti-proliferative activity of the flavonoles 2a–b and no clear correlation emerged between the cell growth inhibition and the presence of the various substituents. However, since quercetin, catechin-3-gallate and epigallocatechin-3-gallate have been shown to inhibit topoisomerase II activity (Austin et al., 1992) we investigated the ability of our derivatives to induce PLDB in cell-based assays. The assay allowed us to assess simultaneously the possible inhibition of both topoisomerase I and II by measuring the amount of pre-labelled-DNA trapped as cleavage complex and precipitated with the proteins. Quercetin, etoposide and camptothecin, used as reference compounds at the indicated concentrations, induced the formation of PLDB by 4.0-, 4.1- and 3.0- fold, respectively (Table 1). In contrast, none of the test compounds induced PLDB, indicating that the anti-IN and -RT activities of compounds 8a–b and 4i–l do not correlate with topoisomerase inhibition.

Discussion Among natural products, several flavonoids have attracted recent attention as novel inhibitors of HIV-1 IN (Fesen et al., 1994; Pommier et al, 1997). Similar to most HIV-1 IN inhibitors identified to date, the active flavonoids are characterized by the presence of a catechol moiety together with at least one or two additional hydroxyl groups (Fesen et al., 1994). However, the presence of several hydroxyl substituents have been linked to the lack of selectivity of these compounds (Ono et al., 1990; Austin et al., 1992; Levitzki & Gazit, 1995). Similarly to flavonoids, CAPE, another IN inhibitor, is characterized by the presence of a catechol moiety together with a carbonyl group (Fesen et al., 1994). Unfortunately, the great majority of compounds, active against IN in enzyme assays, fail to inhibit virus replication in cell-based assays. Some galloyl derivatives have, however, proved effective in preventing virus replication (Nishizawa et al., 1989). In the present work, we synthesized a series of new derivatives starting from 4′-chloroflavonols by acylation of 3-hydroxyl group with galloyl and naturally occurring acids and tested them for antiviral activity. Results show that also in this series of inhibitors the presence of an ortho-bis hydroxylation on an aryl ring is required for anti-IN activity. In fact, while the two caffeoyl esters 4i–l and the two galloyl esters 8a–b displayed a significant inhibitory effect, all modifications in o-hydroxyl groups resulted in loss of activity. The remarkable inhibitory effect of the galloyl esters with respect to caffeoyl esters is probably due to the introduction of a third adjacent hydroxyl rather than the elimination of the 1,2-ethenyl spacer between the hydroxylated ring and the carbonyl

Antiviral Chemistry & Chemotherapy 9:6

group, since an enhancing effect was also observed when a third hydroxyl group was added to other classes of IN inhibitors such as flavones and CAPE analogues (Fesen et al., 1994; Burke et al., 1995). The increase in lipophilicity of the molecules, owing to the introduction of another chlorine atom in position 3′ (compounds 4l and 8b), did not have any impact on the anti-IN activity, resulting in compounds of similar activity to the parent monochloroflavones (compounds 4i and 8a). It has been reported that CAPE, but not flavones, need to be preincubated with the enzyme for 20 min before the addition of the DNA substrate in order to inhibit the 3′processing activity (Fesen et al., 1994). In our experimental conditions we did not preincubate test compounds with the enzyme. Therefore, the fact that our flavonyl esters were inhibitory without this requirement suggests that even though the catechol moiety seems to be a requisite for the anti-IN activity, the flavone portion of the molecule is important for the interaction between compounds and IN. In addition to IN, several natural polyhydroxylated flavonoids have been shown to inhibit also the HIV-1 RT (Ono et al., 1990). Requisites for this activity are the presence of an unsaturated bond between positions 2 and 3 of the pyrone ring (flavone structure) and the presence of three hydroxyl groups in positions 5, 6 and 7 of the molecule. Removal of one of these hydroxyl groups requires the introduction of three additional hydroxyl groups to restore anti-RT activity (Ono et al., 1990). Among flavans, only epicatechin-3-gallate and epigallocatechin-3-gallate, two polyhydroxylated flavan-3-yl esters, were found to be potent inhibitors of HIV-1 RT (Nakane & Ono, 1990; Moore & Pizza, 1992). To our knowledge, no data are available on RT inhibition by flavon-3-yl esters and flavones with substituents other than hydroxyl and methoxyl groups. Structure–activity studies on our series of chloro-substituted flavon-3-yl esters reveal that the presence of the caffeoyl moiety and two chlorine atoms on the flavone portion of the molecule are both required for RT inhibition. In fact, only caffeic acid 3′,4′-dichloroflavon-3yl ester (4l) showed moderate activity and all modifications of its structure led to inactive derivatives. In contrast to tetragalloylquinic acid (Nishizawa et al., 1989), the galloyl derivatives 8a–b were not able to inhibit HIV-1 RT. Also unlike other galloyl derivatives (Nishizawa et al., 1989), our flavonyl esters were not able to prevent virus cytopathogenicity in de novo-infected MT4 cells. This failure is probably due to their high cytotoxicity. However, differently from both flavone and catechin derivatives, which promote enzymatic cleavage of the DNA by topoisomerase II (Austin et al., 1992), our flavonoids did not induce the formation of PLDB. In addition, their structures significantly differ from the minimum flavone structure that has been proposed to be required for flavone inhibition of

507

N Desideri et al.

ATP-dependent enzymes such as protein kinase C, tyrosine kinase and others (Austin et al., 1992; Ferriola et al, 1989). Therefore, even though the importance of such mechanisms is unclear, since other reports have shown that flavonoids are toxic in T lymphocytes assayed in cell culture but are virtually non-toxic in vivo to animals and humans (Havsteen, 1983), different mechanisms of cellular toxicity may be involved for our flavonyl esters. Finally, it is interesting to underline that even though the presence of at least two vicinal hydroxyl groups has been described to be important for the interaction between IN and several hydroxylated aromatic inhibitors (Pommier et al., 1997), the mechanism of interaction is not yet clear. Recent reports suggest that polyhydroxylated compounds are probably active against the assembly of the preintegration complex (Hazuda et al., 1997) and the possibility that they may bind to the IN active site has been proposed (Pommier et al., 1997). Recently, it has also been shown that in case of quinoline derivatives the presence of the catechol structure with free hydroxyl groups is not essential for activity but it may be replaced by a carboxyl or a cyano group (Mekouar et al., 1998). Furthermore, the presence of a carboxyl group has been shown to decrease the cytotoxicity of some derivatives (Mekouar et al., 1998). In view of these observations, the synthesis of new derivatives with this substituent is currently under evaluation. In conclusion, it is worth to note that the failure of test compounds to block HIV-1 replication in cell-based assays does not necessarily indicate that they are without prospects as antiviral agents. In fact, it has already been proposed that structure modifications may reduce their cytotoxicity and enhance their antiviral effect (Ono et al., 1990; Austin et al., 1992). Particularly, structure modifications may be performed on compound 4l which, as far as we know, is the first synthetic derivative able to selectively inhibit both HIV-1 RT and IN and represents an example of the possibility of multiple targeted compounds.

Acknowledgements This research has been supported by a contribution of the ‘Istituto Pasteur-Fondazione Cenci Bolognetti’, Università degli Studi di Roma ‘La Sapienza’, and by a contribution from the Italian Ministero della Sanità - Istituto Superiore di Sanità - IX Progetto AIDS 1996 grant number 9403–59 and from Assessorato Igiene e Sanità, Regione Sardegna, Università di Cagliari.

References Allen CFH & Byers JR Jr (1949) Synthesis of coniferyl alcohol and coniferyl benzoate. Journal of the American Chemistry Society 71:2683–2684.

508

Andrake MD & Skalka AM. (1996) Retroviral integrase, putting the pieces together. Journal of Biological Chemistry 271:19633–19636. Austin CA, Patel S, Ono K, Nakane H & Fisher LM (1992) Sitespecific DNA cleavage by mammalian DNA topoisomerase II induced by novel flavone and catechin derivatives. Biochemical Journal 282:883–889. Burali C, Desideri N, Stein ML, Conti C & Orsi N (1987) Synthesis and anti-rhinovirus activity of halogen substituted isoflavenes and isoflavans. European Journal of Medicinal Chemistry 22:119–123. Burke TR, Fesen MR, Mazumder A, Wang Y, Carothers AM, Grunberger D, Driscoll J, Khon KW & Pommier Y (1995) Hydroxylated aromatic inhibitors of HIV-1 integrase. Journal of Medicinal Chemistry 38:4171–4178. Cheng PL, Fournari P & Tirouflet J (1963) Reserches en série hétérocyclique. VII. Syntèse de chromannones et o-hydroxychalcones substituées sur le differents cycles. Bulletin de la Societe Chimique de France 2248–2251. Conti C, Orsi N & Stein ML (1988) Effect of isoflavans and isoflavenes on rhinovirus 1B and its replication in HeLa cells. Antiviral Research 10:117–127. Conti C, Desideri N, Orsi N, Sestili I & Stein ML (1990a) Synthesis and anti-rhinovirus activity of cyano and amidino substituted flavanoids European Journal of Medicinal Chemistry 25:725–730. Conti C, Genovese D, Santoro R, Stein ML, Orsi N & Fiore L (1990b) Activities and mechanisms of action of halogen-substituted flavanoids against poliovirus type 2 infection in vitro. Antimicrobial Agents and Chemotherapy 34:460–466. Conti C, Tomao P, Genovese D, Desideri N, Stein ML & Orsi N (1992) Mechanism of action of antirhinovirus flavanoid 4′,6dicyanoflavan. Antimicrobial Agents and Chemotherapy 36:95–99. De Clercq E (1994) HIV resistance to reverse transcriptase inhibitors. Biochemical Pharmacology 47:155–169. De Clercq E (1995) Toward improved anti-HIV chemotherapy: therapeutic strategies for intervention with HIV infections Journal of Medicinal Chemistry 38:2491–2517. Deulofeu V & Schopflocher N (1953) The flavonic glucosides of Ombu (Phytolacca dioica). III. The synthesis of ombuin and of other flavonic derivatives of isovanillic series. Gazzetta Chimica Italiana 83:449–453. Desideri N, Sestili I, Stein ML, Conti C, Tomao P & Orsi N (1990) Synthesis and anti-rhinovirus 1B activity of oxazolinylflavans. Antiviral Chemistry and Chemotherapy 1:307–312. Desideri N, Conti C, Sestili I, Tomao P, Stein ML & Orsi N (1992) Synthesis and evaluation of anti-rhinovirus 1B activity of oxazolinyl-isoflavans and -3(2 H)-isoflavenes. Antiviral Chemistry and Chemotherapy 3:195–202. Desideri N, Conti C, Sestili I, Tomao P, Stein ML & Orsi N (1995) In vitro evaluation of the anti-picornavirus activities of new synthetic flavonoids. Antiviral Chemistry and Chemotherapy 6:298–306. Desideri N, Olivieri S, Stein ML, Sgro R, Orsi N & Conti C (1997) Synthesis and anti-picornavirus activity of homoisoflavonoids. Antiviral Chemistry and Chemotherapy 8:545–555. Ferriola PC, Cody V & Middetlon E (1989) Protein kinase C inhibition by plant flavanoids. Kinetic mechanisms and structure–activity relationships. Biochemical Pharmacology 38:1617–1622. Fesen MR, Pommier Y, Leteurtre F, Hiroguchi S, Yung J & Khon KW (1994) Inhibition of HIV-1 integrase by flavones, caffeic acid phenethyl ester (CAPE) and related compounds. Biochemical Pharmacology 48:595–608.

©1998 International Medical Press

Anti-HIV-1 integrase activity of synthetic flavon-3-yl esters

Freudenberg K & Hüber HH (1952) Oxyzimtalkohole und ihre Dehydrierungs-polymerisate. Chemische Berichte 85:1181–1191. Freudenberg K & Heel W (1953) Dioxy- und Trioxy-zimtalkohol. Chemische Berichte 86:190–196. Genovese D, Conti C, Tomao P, Desideri N, Stein ML, Catone S & Fiore L (1995) Effect of chloro, cyano, and amidino-substituted flavanoids on enterovirus infection in vitro. Antiviral Research 27:123–136. Havsteen B (1983) Flavonoids, a class of natural products of high pharmacological potency. Biochemical Pharmacology 32:1141–1148. Hazuda D, Felock P, Hastings JC, Pramanik B & Wolfe AL (1997) Differential divalent cation requirements uncouple the assembly and catalytic reactions of human immunodeficiency virus type 1 integrase. Journal of Virology 71:7005–7011. Johnson VA (1996) Combination therapy for HIV-1 infection overview: preclinical and clinical analysis of antiretroviral combinations. Antiviral Research 29:35–39. Kallay F & Janzs G (1974) Flavonoid chemistry research. Kemiai Kazlemenyek 42:213–223. Kulkosky J & Skalka AM (1994) Molecular mechanism of retroviral integration. Pharmacology and Therapeutics 61:185–203. La Femina RL, Graham PL, Le Grow K, Hastings JC, Wolfe A, Young SD, Emini EA & Hazuda DJ (1995) Inhibition of human immunodeficiency virus integrase by bis-catechols. Antimicrobial Agents and Chemotherapy 39:320–324. Lee KH, Imakura Y, Beers SA, Thurston LS, Dai HJ & Chen CH (1989) Antitumor agent 107. New cytotoxic 4-alkylamino analogues of 4′-demethyl-epipodophyllotoxin as inhibitors of human DNA topoisomerase II. Journal of Natural Products 52:606–613. Levitzki A & Gazit A (1995) Tyrosine kinase inhibition: an approach to drug development. Science 267:1782–1788. Mazumder A, Raghavan K, Weistein J, Kohn KW & Pommier Y (1995a) Inhibition of human immunodeficiency virus type-1 integrase by curcumin. Biochemical Pharmacology 49:1165–1170. Mazumder A, Gazit A, Levitzki A, Nicklaus M, Yung J, Kohlhagen G & Pommier Y (1995b) Effects of tyrphostins, protein kinase inhibitors, on human immunodeficiency virus type 1 integrase. Biochemistry 34:15111–15122. Mekouar K, Mouscadet JF, Desmaele D, Subra F, Leh H, Savouré D, Auclair C & d’Angelo J (1998) Styrylquinoline derivatives: a new class of potent HIV-1 integrase inhibitors that block HIV-1 replication in CEM cells. Journal of Medicinal Chemistry 41:2846–2857. Moore PS & Pizza C (1992) Observations on the inhibition of HIV-1 reverse transcriptase by catechins. Biochemical Journal 288:717–719. Nakane H & Ono K (1990) Differential inhibitory effects of some catechin derivatives on the activities of human immunodeficiency virus reverse transcriptase and cellular deoxyribonucleic and ribonucleic acid polymerases. Biochemistry 29:2841–2845. Neamati N, Mazumder A, Zhao H, Sunder S, Burke TR, Schultz RJ & Pommier Y (1997a) Diarylsulfones, a novel class of human immunodeficiency virus type 1 integrase inhibitors. Antimicrobial Agents and Chemotherapy 41:385–393.

Neamati N, Mazumder H, Sunder S, Owen JM, Schultz RJ & Pommier Y (1997b) 2-Mercaptobenzenesulfonamides as novel inhibitors of human immunodeficiency virus type 1 integrase and replication. Antimicrobial Agents and Chemotherapy 41:485–495. Nishizawa M, Yamagishi T, Dutschman GE, Parker WB, Bodner AJ, Kilkuskie RE, Cheng YC & Lee KH (1989) Anti-AIDS agents, 1. Isolation and characterization of four new tetragalloylquinic acids as a new class of HIV reverse transcriptase inhibitors from tannic acid. Journal of Natural Products 52:762–768. Ono K, Nakane H, Fukushima M, Chermann JC & Barré Sinoussi F (1990) Differential inhibitory effects of various flavonoids on the activities of reverse transcriptase and cellular DNA and RNA polymerases. European Journal of Biochemistry 190:469–476. Pauwels R, Balzarini J, Baba M, Snoeck R, Schols D, Herdewijn P, Desmyster J & DeClercq E (1988) Rapid and automated tetrazolium-based assay for the detection of anti-HIV compounds. Journal of Virological Methods 20:309–321. Pommier Y, Pilon AA, Bajaj K, Mazumder A & Neamati N (1997) HIV-1 integrase as a target for antiviral drugs. Antiviral Chemistry and Chemotherapy 8:463–483. Quaglia MG, Desideri N, Bossù E, Melchiorre P & Conti C (1991) Enantiomeric separation by HPLC/DAD of substituted flavanoids. Journal of Pharmaceutical and Biomedical Analysis 9:811–815. Quaglia MG, Desideri N, Bossù E, Sestili I & Conti C (1992) 4′,6Dichloroflavan (BW863C): enantiomeric separation by HPLC/DAD and antirhinovirus activity. Chirality 4:65–67. Quaglia MG, Desideri N, Bossù E, Sestili I, Tomao P, Conti C & Orsi N (1993) Chiral discrimination and antipicornavirus activity of 6-oxazolinylisoflavan. Chirality 5:356–358. Raghavan K, Buolamwini JK, Fesen M, Pommier Y, Kohn KW & Winstein JN (1995) Three-dimensional quantitative structure–activity relationship (QSAR) of HIV integrase inhibitors: a comparative molecular field analysis (CoMFA) study. Journal of Medicinal Chemistry 38:890–897. Robinson WE, Cordeiro M, Abdel-Malek S, Jia Q, Chow SA, Reinecke MG & Mitchell WM (1996) Dicaffeoylquinic acid inhibitors of human immunodeficiency virus integrase: inhibition of the core catalytic domain of human immunodeficiency virus integrase. Molecular Pharmacology 50:846–855. Schmidt OT & Schach A (1950) Synthese der 3-Galloyl-glucose, 6Galloyl-glucose und 3,6-Digalloyl-glucose. VIII. Mitteilung über natürliche Gerbstoffe. Liebigs Annalen der Chemie 571:29–41. Smith MA, Neumann RM & Webb RA (1968) A modification of Algar-Flynn-Oymada preparation of flavonols. Journal of Heterocyclic Chemistry 5:425–426. Superti F, Seganti L, Orsi N, Divizia M, Gabrieli R, Panà A & Stein ML (1989) Effect of isoflavans and isoflavenes on the infection of Frp/3 cells by hepatitis A virus. Antiviral Research 11:247–254. Tramontano E & Cheng YC (1992) HIV-1 reverse transcriptase inhibition by a dipyridodiazepinone derivative: BI-RG-587. Biochemical Pharmacology 43:1371–1376.

Received 27 July 1998; accepted 1 October 1998

Antiviral Chemistry & Chemotherapy 9:6

509