Glutathione (GSH), a cysteine-containing tripeptide (γ-glutamyl-cysteinyl-glycine) found in eukaryotic cells at mil-limolar concentrations, has a number of important functionsin cell physiology (Meister & Anderson, 1983; Nakamura etal., 1997). In its oxidized disulphide form, GSH protectscells from oxidative stress damage by maintaining the intra-cellular redox state and modulating various intracellularfunctions via metabolic interconversion (Reed, 1986).

Several studies show an imbalance in the intracellularredox state during viral infections, accompanied by a pro-gressive depletion of GSH and, consequently, disruption tothe intracellular redox state. Recent data reported by ourgroup have demonstrated a progressive decrease in intracel-lular GSH levels during parainfluenza-1 Sendai virus (SV),herpes simplex-1 virus (HSV-1) and human immuno- deficiency virus (HIV) infections in vitro (Ciriolo et al.,

1997; Palamara et al., 1995; Garaci et al., 1997). Several invivo studies have also reported an imbalance in the redoxstate in the cells and bodily fluids of HIV- and hepatitis C-infected patients, as well as in the lungs and livers ofinfluenza virus-infected mice (Boya et al., 1999; Elbim etal., 1999; Mileva et al., 2000). Several pieces of data suggestthat the imbalance in the intracellular redox state is a keyevent in the development of viral replication, either byincreasing viral replication (Palamara et al., 1996a) or acti-vating nuclear transcription factors (Schreck et al., 1991).Indeed, many antioxidant substances are known to inhibitthe life cycle of viruses through different mechanisms ofaction. Our previously published data show that theadministration of reduced GSH to infected cells preventsthis decrease in intracellular GSH and inhibits viral repli-cation in SV, HSV-1 and HIV infections (Palamara et al.,

Antiviral Chemistry & Chemotherapy 15:77–85

New synthetic glutathione derivatives with increasedantiviral activitiesAnna Teresa Palamara1, Giorgio Brandi2, Luigia Rossi3, Enrico Millo4, Umberto Benatti5, LuciaNencioni1, Alessandra Iuvara6, Enrico Garaci6 and Mauro Magnani3*

1Institute of Microbiology, University of Rome ‘La Sapienza’, Rome, Italy 2Institute of Hygiene, University of Urbino ‘Carlo Bo’, Urbino (PU), Italy3Institute of Biochemistry ‘G Fornaini’, University of Urbino ‘Carlo Bo’, Urbino (PU), Italy4Insitute ‘Giannina Gaslini’, Genoa, Italy5Department of Experimental Medicine Biochemistry Section, University of Genoa, Genoa, Italy6Department of Experimental Medicine & Biochemical Sciences, University of Rome ‘Tor Vergata’, Rome, Italy

*Corresponding author: Tel: +39 722 305 211; Fax: +39 722 320 188; E-mail: magnani@uniurb.it

A series of glutathione (GSH) derivatives withaliphatic chains of different lengths, coupled bypeptides bound to the α-NH2 group of Glu, weresynthesized. When added to several cell lines, theC6 (n-hexanoyl), C8 (n-octanoyl) and C12 (n-dode-canoyl) derivatives were toxic while the C2 (n-ethanoyl) and C4 (n-butanoyl) derivatives werenot. Preliminary experiments were performed toinvestigate the potential antiviral activity of theC2 and C4 derivatives compared to GSH. The C4derivative was the most potent and fully charac-terized. GSH-C4 is a poor substrate of GSH metab-olizing enzymes; once oxidized by disulphide-bound formation, C4 is slowly reduced by GSH-reductase. GSH-C4 completely abrogated Sendai

virus replication at 7.5 mM with an EC50 of 3.6 mM,compared to 7.5 mM for GSH. GSH-C4 completelyinhibited herpes simplex virus (HSV-1) virus pro-duction in Vero cells at 10 mM, while the samedose of GSH caused only a 2.5 log10 reduction.Furthermore, the GSH-C4 treatment (7.5 mM) wasable to markedly reduce the cytopathic effect ofHSV-1 in Vero cells. Thus, GSH derivatives withincreased hydrophobic properties are more effec-tive antiviral agents against Sendai and HSV-1viruses than GSH, suggesting their usefulness inantiviral therapy.

Key words: glutathione derivatives, antiviral effi-cacy, Sendai virus, herpes simplex-1

Introduction

77©2004 International Medical Press 0956-3202/02/$17.00

AT Palamara et al.

78 ©2004 International Medical Press

1995; Palamara et al., 1996b; Garaci et al., 1992). In theseexperimental models, GSH antiviral activity appears tobe related to an inhibition of the post-transcriptionalstages of viral replication, probably by impeding the cor-rect folding and maturation of specific virus envelopeproteins. Other authors report that the administration ofN-acetyl-cysteine, a precursor of GSH, inhibits HIVreplication (Gu et al., 2001; Roederer et al., 1992) andhepatitis C replication (Gong et al., 2001) by preventingthe activation of NF-kB.

The efficacy of antioxidant substances against viralinfection has also been demonstrated in in vivo studies. Inparticular, our data report that the administration of highdoses of GSH reduces viral infection and inhibits diseaseprogression in a murine model of AIDS (Palamara et al.,1996c). Furthermore, topical GSH treatment is suggestedas a therapy in HSV-induced keratitis (Nucci et al., 2000).Other authors report that administering vitamin E sup-plements significantly decreases pulmonary viral titre ininfluenza-infected old mice (Han et al., 2000).

Even though the antiviral activity of antioxidant sub-stances has been clearly demonstrated, it has been proventhat some molecules, such as GSH, are not freely trans-ported to most cells or tissues. For this reason, we haveevaluated the antiviral activity of some GSH derivativesthat can more easily cross the membranes of many celltypes. We investigated the effect of GSH derivatives on thereplication of RNA virus (SV) and DNA virus (HSV-1) infreely permissive cells [Madin Darby canine kidney(MDCK), African green monkey kidney (Vero)]. Theresults reported in this paper show that GSH-C4 exertsremarkable antiviral activity in vitro against both viruseswithout having toxic effects on uninfected cells, suggestingits potential use in antiviral therapy.

Material and methods

Synthesis of GSH derivatives (GSH-C2, GSH-C4,GSH-C6, GSH-C8, GSH-C12)The compounds were manually synthesized using the stan-dard method of solid-phase peptide synthesis which fol-lows the 9-fluorenylmethoxycarbonyl (Fmoc) strategy(Wellings & Atherton, 1997), with minor modifications.

All manual syntheses were carried out on a 0.1 mmolscale using a rotary reaction vessel with the Fmoc protect-ed N-terminal glycine, preloaded onto Wang resin(Novabiochem AG, Laufelfingen, Switzerland).

The standard synthesis cycle involved swelling the resinin dichloromethane (Biosolve LTD, the Netherlands) forone night. Deprotection of the Fmoc group was thenobtained with 20% piperidine (Fluka Chemie AG, Buchs,Switzerland) in N,N-dimethyl formamide (DMF)(Biosolve LTD, the Netherlands) for 20 min.

Five equivalents (eq.) of the appropriate Fmoc aminoacid (Advanced Biotech Italia, Italy) were preactivated with4.5 eq. of O-(benzotriazol-1-yl)-1,1,3,3 tetramethyluroni-umhexafluorophosphate (Advanced Biotech Italia) and fiveeq. of N,N diisopropilethylamine at a final concentration of0.2 M amino acid in anhydrous N-methylpyrrolidone(Biosolve LTD, the Netherlands). This solution was cou-pled with in situ neutralized resin for 1 h at 40°C. As a cap-ping step, a mixture of 5% acetic anhydride (Fluka ChemieAG, Buchs, Switzerland) in DMF after the coupling ofeach amino acid was included.

GSH-C4 derivative was prepared by treating peptideresin with n-butanoic acid (Fluka Chemie AG, Buchs,Switzerland), activated as previously described for aminoacids. Ethanoyl, hexanoyl, octanoyl and dodecanoyl deriva-tives (GSH-C2, GSH-C6, GSH-C8 and GSH-C12) weresimilarly prepared using acetic anhydride, hexanoic,octanoic and dodecanoic acids (Fluka Chemie AG, Buchs,Switzerland) in place of butanoic acid, respectively.

Cleavage of oligomer from Wang resin and removal ofthe remaining side chain protecting group was performedin a single step, using trifluoroacetic acid (TFA) (BiosolveLTD, the Netherlands): ethanedithiol (Fluka Chemie AG,Buchs, Switzerland): water: triisopropylsilane (FlukaChemie AG, Buchs, Switzerland) (92.5:2.5:2.5:1 v/v) for 2 h at room temperature. The resulting solution was con-centrated to about 1 ml under vacuum and finally precipi-tated and washed in ice-cold diethyl ether.

All the molecules were purified by reverse phase highpressure liquid chromatography (RP-HPLC) with aWaters C18 µBondapack column. Solvent A was 0.1%TFA in water and solvent B was the same in acetonitrile.The solvent program was as follows: a gradient started with100% solvent A for 5 min and linearly increased to 60%solvent B in 30 min and up to 100% B in 5 min.

The fractions containing the relevant molecules werecollected, vacuum dried and, finally, lyophilized. The mol-ecular weights of GSH derivatives confirmed by an electro-spray mass spectrometer were: 349.3, 377.3, 405.3, 433.3and 489.3 daltons for GSH-C2, GSH-C4, GSH-C6,GSH-C8 and GSH-C12, respectively. Mass spectra ofeach compound were acquired in the negative ion modeusing a single quadrupole HP Engine 5989-A equippedwith an electrospray ion source. All products were obtainedwith a final yield ranging from 75–78% and a purity greaterthan 95% after HPLC analysis. An example of mass spec-trum for GSH-C4 is shown in Figure 1.

Cells and virusesMDCK, Vero and human histiocytic lymphoma cells(U937) were grown in RPMI 1640 medium supplement-ed with 10% heat inactivated fetal calf serum (FlowLaboratories, Milan, Italy).

Antiviral activities of glutathione derivatives

Antiviral Chemistry & Chemotherapy 15:2 79

SV is a nonsegmented single-stranded RNA virus ofnegative polarity. Like other paramyxoviruses, SV was pre-pared by allantoic inoculation of embryonated eggs as pre-viously described (Garaci et al., 1992). Confluent monolay-ers of MDCK cells were infected with SV [three haemag-glutinating units (HAU)×105 cells]. After incubation for 1h at 37°C (adsorption period), unadsorbed virus wasremoved. The monolayers were washed and then incubatedwith complete medium containing 2% fetal calf serum.Virus production was determined in the supernatants ofinfected cells at different times post-infection (p.i.), bymeasuring the HAU according to standard procedures,using human type 0 Rh+ erythrocytes. For the evaluation ofantiviral activity, the compounds GSH-C2, GSH-C4 andGSH-C6 were diluted in RPMI pH 7.3 and added at thedesired concentration, just after the viral adsorption period.

Clinically isolated HSV-1 TV1 was grown and titratedin Vero cells as previously described (Palamara et al., 1995).For infection, confluent monolayers of Vero cells wereinfected with HSV-1 at m.o.i. (multiplicity of infection) of0.03 PFU/cell. After incubation for 1 h at 37°C (adsorp-tion period), unadsorbed virus was removed. The monolay-ers were washed and then incubated with complete

medium containing 2% fetal calf serum. GSH-C4 wasadded, at the indicated doses, just after the viral adsorptionperiod and maintained in the culture medium until theexperiments were finished. Supernatants from infected cellswere harvested at different time points after virus challengeand tested for their ability to form plaques in Vero cells,using a standard titration method. Similar experimentswere performed also on a thymidine-kinases-negativestrain (HSV-1 TK- ∆305), following the same procedures.

Cell toxicity studiesAll of the compounds (GSH-C2, GSH-C4, GSH-C6,GSH-C8, GSH-C12) were studied for cytotoxicity onconfluent monolayers of uninfected MDCK cells.Cytotoxicity was evaluated on the basis of microscopicexamination of cell morphology, evaluation of cell viabili-ty after trypan blue staining and cell count. All the com-pounds were diluted in RPMI; the final pH of the solutionwas about 7–7.3. GSH-C2 and GSH-C4 were also stud-ied for their eventual cytostatic effect on exponentiallygrowing U937 cell line. To this aim, cells were stained withtrypan blue and counted 24, 48 and 72 h after addition ofdifferent doses of the substances.

0

300

250

200

150

100

50

350

650

600

550

500

450

400

700

800

750

300250200150100 350 650600550500450400 700 800750m/z

Abundance 376.7

OH

OHNH

NH

NH

SH

CH3

O

OO

OO

Figure 1. Molecular-chemical structure of n-butanoyl α-glutamyl-cysteinyl-glycine (GSH-C4) and its MS analy-sis after purification

Performed in negative ion mode, the acquisition revealed a unique peak at 376.7 corresponding to [M-H]–.

Enzymatic determinations

GSH peroxidase (EC 1.11.1.9.) activity in humanhaemolysate in the presence of GSH or its derivative GSH-C4 was measured spectrophotometrically in a system cou-pled with GSH reductase (1.6.4.2.) (prepared as describedby Beutler, 1984). Briefly, 0.1 M Tris-HCl/ 0.5 mM EDTApH 8.0, 2.0 mM GSH or GSH-C4, 1.0 U GSH reductase,0.2 mM NADPH and 10 µl haemolysate 1:20 were addedto cuvettes with a critical volume of less than 1.0 ml.Reagents were incubated at 37°C for 10 min; 0.07 mM t-butylhydroperoxide was then added and the system’s opticaldensity (OD) decrease was measured at 340 nm.

GSH S-transferase (EC 2.5.1.18.) activity in humanhaemolysate, in the presence of GSH or the derivativeGSH-C4 was measured spectrophotometrically asdescribed (Beutler, 1984). Briefly, 0.1 M K2HPO4/KH2PO4 pH 6.5 and 0.5 mM 1-chloro-2,4-dinitrobenzene(CDNB) were incubated at 37°C for 10 min, then 1.0 mMGSH or GSH-C4 and 50 µl haemolysate 1:20 were addedto the mixture and the increase of OD, due to the productionof CDNB-GSH conjugate, at 340 nm was measured.

GSH and GSH-C4 were then oxidized by overnight incu-bation at room temperature in the presence of 35% (v/v) H2O2

to obtain GSSG and C4-GSSG-C4, respectively. GSHreductase (EC 1.6.4.2.) activity was then evaluated spec-trophotometrically on human haemolysate, as previouslydescribed (Beutler, 1984). Briefly, 0.05 mM Tris-HCl/ 0.25mM EDTA pH 8.0, 10 µl haemolysate 1:20 and 0.1 µMFAD were incubated 10 min at 37°C, then 1.5 mM GSSG orC4-GSSG-C4 were added and the mixture was incubatedagain at 37°C for 10 min. Finally, 0.1 mM NADPH wasadded and the decrease of OD at 340 nm was evaluated.

Kinetics of GSH and GSH-C4 influx into erythrocytesAssays of GSH and GSH-C4 influx into erythrocytes(RBCs) were performed at 37°C in 1.5 ml polypropylenemicro centrifuge tubes, using ‘oil-stop’ methodology.Briefly, red blood cell suspensions at 10% haematocrit con-taining 5 mM GSH or GSH-C4 and 10 mM GSH wereincubated 2 h at 37°C. At different times (0, 5, 15, 30, 60and 120 min), 600 µl aliquots were stratified on 600 µl bro-mododecane, centrifuged 5 min at 10000 g and spec-trophotometrically evaluated for GSH and GSH-C4 con-tent in RBCs, as described (Beutler, 1984). Briefly, 30 µlRBC pellets were resuspended in 70 µl of phosphate salinesolution (PBS) pH 7.4, diluted with H2O (1:10) and 1.5 mlof precipitating solution (for composition see Beutler,1984) was added. GSH and GSH-C4 were evaluated onclear supernatant at 412 nm after the addition of DTNB[5′,5′-dithiobis (2-nitrobenzoic acid)] reagent, a disulphidecompound that is readily reduced by sulphydryl compoundsforming a highly coloured yellow anion.

GSH-C4 stability in plasma

GSH-C4 (1 mM) was incubated in human plasma at 37°C.And at different time points of incubation (0, 15, 30, 60and 120 min), 200 µl aliquots were ultra filtered, using theAmicon Centricon 30 microconcentrators, by centrifuga-tion at 2000 g for 30 min. The filtered solutions weredirectly analysed by high-performance capillary elec-trophoresis (HPCE) to evaluate GSH-C4 and C4-GSSG-C4 content as described (Piccoli et al., 1994).

Results

GSH derivatives toxicity and antiviral activityThe addition of C8 (n-octanoyl) and C12 (n-dodecanoyl)derivatives caused marked toxic effects and damage to unin-fected MDCK monolayer cells. These effects increaseddose-dependently, and were observed starting from the 0.1mM concentration. For this reason, their activity on viralreplication was not considered. The treatment of MDCKcells with GSH-C6 induced monolayer cell damage exclu-sively at the concentration of 10 mM. Some modification ofcell morphology was detected 24 h after addition of 2.5 and5 mM. This substance caused an inhibition of SV replica-tion, measured as haemagglutinating activity in the super-natant, ranging between 50–100% (2.5–10 mM, respective-ly). Because of the limited difference between cytotoxic andantiviral doses, we did not continue the study of this deriv-ative. GSH-C4 and GSH-C2 did not induce any toxiceffect or change in morphology of cell monolayer at theconcentrations used for the study (0.1–10 mM). To betterinvestigate the eventual cytostatic effect of these compoundson exponentially growing cells, U937 cells were plated at theconcentration of 2×105/ml and increasing doses of GSH-C2 and GSH-C4 were added. Cell counts were performed24–48 and 72 h after drug addition. Both compoundsinduced a slight, non-significant inhibition of cell prolifera-tion compared to untreated cells when used at 10 mM con-centration (10–15% decrease for GSH-C2 and GSH-C4,respectively). No differences were detected after addition ofboth compounds at 1, 2, 2.5, 5 and 7.5 mM concentrations.

Regarding the antiviral activity of GSH-C2, its additionto SV infected MDCK cells caused a slight reduction ofviral replication. The viral titre detected in the supernatantof cells treated with 5.0 and 7.5 mM GSH-C2 wasdecreased by 30 and 40%, respectively, compared to untreat-ed infected cells. Furthermore, GSH-C2 had no effect onHSV-1 replication. For these reasons, the study was thenfinalized in order to better characterize GSH-C4 activities.

GSH-C4 metabolismThe enzymatic activities of GSH peroxidase and GSH S-transferase from human haemolysate were evaluated in thepresence of either 2 mM GSH-C4 or GSH as substrates.

AT Palamara et al.

80 ©2004 International Medical Press

Antiviral activities of glutathione derivatives

GSH peroxidase and GSH S-transferase activities in thepresence of GSH-C4 were 1.8% and 3.5% compared toGSH, respectively. Moreover, GSH and GSH-C4 in thecorresponding disulphide forms were chemically oxidized.GSH reductase activity on both substrates was then evalu-ated. The results obtained show that C4-GSSG-C4 isslowly reduced by the enzyme, having an activity of 0.63IU/g haemoglobin (Hb), a Km of 50 mM and a Vmax of12.6 µmol/min/mg Hb compared to an activity of 3.0 IU/gHb, a Km of 1.3 mM and a Vmax of 6.3 µmol/min/mg Hbobtained with GSSG.

GSH-C4 influx into erythrocytes The influx of GSH-C4 into erythrocytes was evaluated andcompared to GSH. The results obtained show that GSH-C4 permeates the erythrocyte membrane more quicklythan GSH; in fact, the rate of influx (calculated in the first30 min) was 4.33 nmol/min/ml RBCs for GSH-C4 and2.33 nmol/min/ml RBCs for GSH.

GSH-C4 stability in plasmaThe ability of human plasma enzymes to convert GSH-C4in C4-GSSG-C4 was also evaluated. As shown in Figure2, plasma enzymes are able to oxidize GSH-C4 with thestoichiometric production of C4-GSSG-C4 (t1/2 50 min).

Effect of GSH-C4 on SV replicationGSH-C4 inhibits replication of SV in MDCK cells in adose dependent manner (Figure 3). An inhibition ranging

between 88% (24 h p.i.) and 93% (48 h p.i.) was found inthe presence of 5 mM GSH-C4. No virus was detected inthe supernatant of cells treated with 7.5 mM of the drug.The optimal antiviral dose of GSH-C4 (7.5 mM) was nottoxic for the cells; this was confirmed by microscopicexamination of the monolayers and by vital dye exclusion.The 50% inhibition of viral production (EC50) at 48 h p.i.was obtained at the concentration of 3.6 mM; 7.5 mM ofGSH was necessary to obtain the same results. It is note-worthy that in different experimental models (Garaci etal., 1992; Palamara et al., 1995; Kalebic et al., 1991; Vogelet al., 2003) the GSH antiviral activity has been describedin doses ranging between 1–15 mM.

At the concentrations used in this study, GSH-C4 didnot induce toxic effects on uninfected cells. This wasconfirmed by microscopic examination of the monolayersand by vital dye exclusion (not shown).

Effect of GSH-C4 on HSV-1 replication The effect of different doses of GSH-C4 on HSV-1 pro-duction in Vero cells is shown in Figure 4; data representthe viral titre in the supernatant of infected cells 48 h afterinfection. The results obtained show that GSH-C4 inhib-ited virus replication in a dose-dependent manner. Amarked decrease (approx. 3 log10) in viral replication wasachieved through the addition of 7.5 mM GSH-C4. Novirus particles were detected at 10 mM. The same inhibi-tion was found 72 h after infection (not shown).

Antiviral Chemistry & Chemotherapy 15:2 81

0

0.2

0.4

0.6

0.8

1.2

0 20 40 60 80 100 120 140

Time (min)

mM

(µ

mo

l/ml p

lasm

a)

GSH-C4C4-GSSG-C4

1.0

Figure 2. GSH-C4 stability in plasma

GSH-C4 (1 µmol/ml plasma) was incubated for 2 h at 37°C. Ultrafiltered samples were prepared at different incubation times and analysed byhigh-performance capillary electrophoresis (HPCE). Values are the mean ± SD of three different experiments.

The addition of GSH at the concentration of 10 mMdid not induce a complete inhibition of the virus, and pro-duced only a 2.5 log10 reduction of virus replication (Figure4). Furthermore, unlike GSH, GSH-C4 (7.5 mM)markedly reduced the typical cytopathic effect induced byHSV-1 (data not shown).

A percentage of HSV strains (approx. 5% in immuno-compromised and <1% in immunocompetent patients) hasbecome resistant to conventional therapy (Field, 2001), suchas acyclovir, as a consequence of a mutation in the thymi-dine-kinase gene (TK–). For this reason, we also tested theeffect of GSH-C4 on a TK– strain. The results obtained(Table 1) show that GSH-C4 is less active against the TK–

strain than the wild type strain. In fact, a significant inhibi-tion of virus titre (about 1 log10) was observed only at a con-

centration of 7.5 mM 72 h p.i. However, the HSV-1 TK–

strain used in this study was a deletion clone (∆305), witha lower replicative rate (about 3 log10) than the parentalstrain. Thus, the lower sensitivity of this HSV-1 TK– strainto inhibition by GSH-C4 could not be considered repre-sentative of all TK– strains isolated from clinical samplesthat usually replicate normally in tissue cultures.

Discussion

GSH probably exerts its antimicrobial activity in a numberof different ways; for example, by controlling viral or micro-bial growth by affecting the correct folding of essentialpathogen proteins [see Palamara et al., 1996b (HIV-1 Gp120); Nencioni et al., 2003, Cai et al., 2003 (influenza A

AT Palamara et al.

82 ©2004 International Medical Press

0

10

20

30

40

50

60

70

0 2.5 5 7.5 7.5

HA

U/m

l

24 h48 h

GSH-C4(mM)

GSH(mM)

******

******

**

*

1.0

Figure 3. Effect of GSH-C4 on Sendai virus replication

MDCK cells were infected for 1 h at 3 haemogglutinating units (HAU)x105 cells.The cells were extensively washed and then cultured in thepresence of different GSH-C4 concentrations (0–7.5 mM range) or in the presence of 7.5 mM GSH for 2 days. At 24 h and 48 h virus produc-tion was assayed by measuring the HAU using human type 0 Rh+ erythrocytes. The results shown are from one representative experiment ofthree performed, each of one in triplicate. t test was calculated in the three different experiments.*P<0.05; **P<0.001; ***P<0.0001.

Table 1. Effect of GSH-C4 on the herpes simplex virus (HSV)-1 thymidine-kinase gene (TK–)

HSV-1 TK– (pfu/ml)

24h 48h 72hCtr. 1.38x105 1.96x105 2.56x106

5 mM 2.59x105 2.42x106 2.13x106

7.5 mM 1.59x105 2.56x105 4.33x105

10 mM 1.53x105 2.60x105 3.44x105

Vero cells were infected for 1 h at 37°C with the TK– negative strain ∆305 and then cultured in the presence of different GSH-C4 concentra-tions. Virus production was evaluated 1, 2 and 3 days post infection by plaque assay. The results shown are from one representative experi-ment of three that agree within 10% of the reported values.

Antiviral Chemistry & Chemotherapy 15:2 83

Antiviral activities of glutathione derivatives

haemoagglutinin)], interfering with the activation ofselected transcription factors [Roederer et al., 1990; Mihmet al., 1995 (NF-kB)] and affecting viral receptor expres-sion (Papi et al., 2002) or function (Matthias et al., 2002).Thus, GSH can be considered an antimicrobial agent thatexerts its activity through several different mechanisms,depending on the host-pathogen systems considered andon the GSH concentration applied. Unfortunately, GSH invivo is rapidly oxidized, particularly in the presence of viralinfections when the redox state of the cells is unbalanced bya massive production of reactive oxygen species. OxidizedGSH is in turn reduced by cellular GSH reductases orpumped out of the cell by an ATP-dependent mechanism.

GSH reduction by cellular enzymes depends on the avail-ability of cell-reducing equivalents (NADH, NADPH); thisis not favourable in cases of pathogenic infection. Instead,the expulsion of oxidized GSH by the cell can only be coun-terbalanced by new intracellular synthesis of the tripeptide.Taking into account these considerations, we deduced thepossibility of using GSH derivatives capable of crossing bio-logical membranes in the treatment of several viral infec-tions. The molecules reported in this paper represent a firstclass of such GSH derivatives. The addition of aliphaticchains at the α-NH2 of glutamic acid represents, in effect, auseful approach in the production of diffusible drugs becauseof the hydrophobic properties of the chain and the neutral-ization of a charged NH2 group. Unfortunately, aliphatic

chains longer than four C atoms are toxic to the cells, whilea single ethanoyl group show no or little effect. Thus, thebutanoyl derivative of GSH was extensively characterized. Itis important to note that, like GSH, this GSH derivativecould act via different mechanisms. It probably interfereswith different phases of the viral life-cycle, since it is able toprevent the cytopathic effect of HSV-1 in Vero cells but alsoinhibits SV replication in MDCK- infected cells.Understanding butanoyl GSH’s mechanism of action isbeyond the scope of this paper but certainly will be furtherinvestigated in our laboratories. The fact that butanoyl GSHcan probably be used not only as a water-soluble drug butalso in the form of a cream or lotion, particularly in the inhi-bition of HSV-1 cytopathic effect by local applications, hasnot escaped our attention.

The GSH derivatives we have designed, synthesized andcharacterized are new but other derivatives have been pre-viously reported. Meister and co-workers reported almost20 years ago on the preparation and characterization ofGSH monoethyl ester (Puri & Meister, 1983; Anderson etal., 1985; Wellner et al., 1984; Anderson et al., 1989;Naganuma et al., 1990), a derivative that is intracellularlytransported and split into GSH. Burg et al. (2002a) havereported on the synthesis of peptidomimetic GSH-conju-gate analogues that are effective inhibitors of multidrugresistance protein 1 (MRP1) and thus can have a role inanticancer drug resistance (Burg & Mulder, 2002b).

0 2.5 7.5 10.0 10.0

Vir

us

yiel

d (

PFU

/ml)

102

103

104

105

106

107

108

109

GSH-C4(mM)

GSH(mM)

***

***

*

1.0 5.0

Figure 4. Effect of GSH-C4 on HSV-1 replication

Vero cells were infected for 1 h with HSV-1 at m.o.i. of 0.03 PFU/cell. After extensive washings, GSH-C4 at different concentrations (0–10 mMrange) or 10 mM GSH were added to cell cultures. Virus replication was assayed in Vero cells 48 h post infection by using the plaque assay.Each point represents the average ± SD of quadruplicate samples. Data are from a single experiment representative of 4.*P<0.05; ***P<0.0001.

Finally, nitric oxide derivatives (S-nitrosoglutathione) havebeen extensively investigated as antioxidants (Rigobello etal., 2002) and nitric oxide donors (Kuo et al., 2003). Theselast derivatives, as well as a number of GSH conjugates, canalso occur in vivo and represent a useful system for thetransport of chemically reactive metabolites of both autol-ogous and exogenous origins (Baillie & Slatter, 1991).Many GSH derivatives are formed through the SH reactivegroup of this peptide; however, usually these compoundsare unstable because of the action of endogenous GSHmetabolizing enzymes. Butanoyl GSH is stable but can beoxidized as free GSH. Once oxidized in dimeric form, thecellular GSH reductase is not able to efficiently reduce itback to butanoyl GSH. Thus, butanoyl GSH is not a goodsubstrate for GSH metabolizing enzymes and exerts its pri-mary effect as a reducing agent, with a plasma half-life ofabout 50 min (see Figure 2), while free GSH has a half-lifeof 30 min (Magnani et al., 1984). The results reported inthis paper and in the available literature suggest thatbutanoyl GSH can be considered an interesting antimicro-bial agent with proven efficacy in vitro against differentpathogens, with a potential place among compounds thatcan reduce both viral infectivity in early disease and viralproduction in later disease. This property is nearly uniqueand worth further exploitation.

Acknowledgements

We thank Dr Giovanni Piccoli (Istituto di Ricercasull’Attività Motoria, Università degli Studi di Urbino‘Carlo Bo’, Urbino, Italy) for HPCE determinations.

References

Anderson M-E & Meister A (1989) Glutathione monoesters.Analytical Biochemistry 183:16–20.

Anderson M-E, Powrie F, Puri R-N & Meister A (1985)Glutathione monoethyl ester: preparation, uptake by tissue andconversion to glutathione. Archives of Biochemistry & Biophysics239:538–548.

Baillie T-H & Slatter J-G (1991) Glutathione: a vehicle for thetransport of chemically reactive metabolites in vivo. AmericanChemical Society 24:264–270.

Beutler E (1984) Red Cell Metabolism. A Manual of BiochemicalMethods. 3rd edition. New York: Grune & Stratton.

Boya P, De La Pena A, Beloqui O, Larrea E, Conchillo M,Castelruiz Y, Civeira M-P & Prieto J (1999). Antioxidant statusand glutathione metabolism in peripheral blood mononuclear cellsfrom patients with chronic hepatitis C. Journal of Hepatology31:808–814.

Burg D, Wielinga P, Zelcer N, Saeki T, Mulder G-J & Borst P(2002a) Inhibition of the multidrug resistance protein 1 (MRP1)by peptidomimetic glutathione-conjugate analogs. MolecularPharmacology 62:1160–1166.

Burg D & Mulder G J (2002b) Glutathione conjugates and theirsynthetic derivatives as inhibitors of glutathione-dependentenzymes involved in cancer and drug resistance. Drug MetabolismReviews 34:821–863.

Cai J, Chen Y, Seth S, Furukawa S, Compans R-W & Jones D-P(2003) Inhibition of influenza infection by glutathione. FreeRadical Biology & Medicine 34:928–936.

Ciriolo M-R, Palamara A-T, Incerpi S, Lafavia E, Bue M-C, DeVito P, Garaci E & Rotilio G (1997) Loss of GSH, oxidativestress and decrease of intracellular pH as sequential steps in viralinfection. Journal of Biological Chemistry 272:2700–2708.

Elbim C, Pillet S, Prevost M-H, Preira A, Girard PM, Rogine N,Matusani H, Hakim J, Israel N & Gougerot-Pocidalo M-A (1999)Redox and activation status of monocytes from human immunode-ficiency virus-infected patients: relationship with viral load. Journalof Virology 73:4561–4566.

Field HJ (2001) Herpes simplex virus antiviral drug resistance-cur-rent trends and future prospects. Journal of Clinical Virology21:261–269.

Garaci E, Palamara A-T, Ciriolo M-R, D'Agostini C, Abdel-LatifM-S, Aquaro S, Lafavia E, Rotilio G & del-Latif M-S (1997)Intracellular GSH content and HIV replication in Humanmacrophages. Journal of Leukocyte Biology 62:54–59.

Garaci E, Palamara A-T, Di Francesco P, Favalli C, Ciriolo M-R &Rotilio G (1992). Glutathione inhibits replication and expressionof viral proteins in cultured cells infected with Senday virus.Biochemical & Biophysical Research Communications 188:1090–1096.

Gong G, Waris G, Tanveer R & Siddiqui A (2001) Human hepatitisC virus NS54 protein alters intracellular calcium levels, inducesoxidative stress, and activates STAT-3 and NF-kappa B.Proceedings of the National Academy of Sciences, USA 98:9599–604.

Gu Y, Wu R-F, Xu Y-C, Flores S-C & Terada L-S (2001). HIV Tatactivates c-Jun amino-terminal kinase throgh an oxidant-depen-dent mechanism. Virology 286:62–71.

Han S-N, Meydani M, Wu D, Bender B-S, Smith D-E, Vina J, CaoG, Prior R-L & Meydani S-N (2000). Effect of long-term dietaryantioxidant supplementation on influenza virus infection. Journalsof Gerontology Series A: Biological Sciences & Medical Sciences55:B496–503.

Kalebic T, Kinter A, Poli G, Anderson ME, Meister A & Fauci AS(1991) Suppression of human immunodeficiency virus expressionin chronically infected monocytic cells by glutathione, glutathioneester, and N-acetylcysteine. Proceedings of the National Academy ofSciences, USA 88:986–990.

Kuo W-N, Kocis J-M & Nibbs J (2003) Nitrosation of cysteine andreduced glutathione by nitrite at physiological pH. Frontiers inBioscience 8:A62–A69.

Magnani M, Novelli G & Palloni R (1984) Human plasma glu-tathione oxidation in normal and pathological conditions. ClinicalPhysiology & Biochemistry 2:287–290.

Matthias L-J, Yam P-T, Jiang X-M, Vandegraaff N, Li P,Poumbourios P, Donoghue N & Hogg P-J (2002) Disulfideexchange in domain 2 of CD4 is required for entry of HIV-1.Nature Immunology 3:727–732.

Meister A. & Anderson M E (1983). Glutathione. Annual Review ofBiochemistry 52:711–760.

Mihm S, Galter D & Droge W (1995) Modulation of transcriptionfactor NF kappa B activity by intracellular glutathione levels andby variations of the extracellular cysteine supply. Federation ofAmerican Societies for Experimental Biology Journal 9:246–252.

Mileva M, Tancheva L, Bakalova R, Galabov A, Savov V & RibarovS (2000) Effect of vitamin E on lipid peroxidation and livermonooxigenase activity in experimental influenza virus infection.Toxicology Letters 114:39–45.

Naganuma A, Anderson M E & Meister A (1990) Cellular glu-tathione is a determinant of sensivity to mercuric chloride toxicity:prevention of toxicity by giving glutathione monoester. BiochemicalPharmacology 40:693–697.

84 ©2004 International Medical Press

AT Palamara et al.

Nakamura H, Nakamura K & Yodoi J (1997). Redox regulation ofcellular activation. Annual Review of Immunology 15:351–369.

Nencioni L, Iuvara A, Aquilano K, Ciriolo M-R, Cozzolino F,Rotilio G, Garaci E & Palamara A-T (2003) Influenza A virusreplication is dependent on an antioxidant pathway that involvesGSH and Bcl-2. Federation of American Societies for ExperimentalBiology Journal 17:758–760.

Nucci C, Palamara A-T, Ciriolo M-R, Nencioni L, Savini P,D'Agostini C, Rotilio G, Cerulli L & Garaci E (2000). Imbalancein corneal redox state during Herpes Simplex virus 1- induced ker-atitis in rabbits. Effectiveness of exogenous glutathione supply.Experimental Eye Research 70:215–220.

Palamara A-T, Perno C-F, Ciriolo M-R, Dini L, Balestra E,D'Agostini C, Di Francesco P, Favalli C, Rotilio G & Garaci E(1995) Evidence for antiviral activity of glutathione: in vitro inhi-bition of herpes simplex virus type 1 replication. Antiviral Research27:237–253.

Palamara A-T, Di Francesco P, Ciriolo M, Bue C, Lafavia E, RotilioG & Garaci E (1996a) Cocaine increases Sendai virus replicationin cultured epithelial cells: critical role of the intracellular redoxstatus. Biochemical & Biophysical Research Communications228:579–585.

Palamara A-T, Perno C-F, Aquaro S, Bue M-C, Dini L & Garaci E(1996b) Glutathione inhibits HIV replication by acting at latestages of the virus life cycle. AIDS Research & Human Retroviruses12:1537–1541.

Palamara A-T, Garaci E, Rotilio G, Ciriolo M-R, Casabianca A,Fraternale A, Rossi L, Schiavano G-F, Chiarantini L, Magnani M(1996c) Inhibition of murine AIDS by reduced glutathione. AIDSResearch & Human Retroviruses 12:1373–1381.

Papi A, Papadopoulos N-G, Stanciu L-A, Bellettato C-M,Pinamonti S, Degitz K, Holgate S-T & Johnston S-L (2002)Reducing agents inhibit rhinovirus-induced up-regulation of therhinovirus receptor intercellular adhesion molecule-1 (ICAM-1) inrespiratory ephithelial cells. FASEB Journal 16:1934–1936.

Piccoli G, Fiorani M, Biagiarelli B, Palma F, Potenza L, Amicucci A& Stocchi V (1994) Simultaneous high-performance capillaryelectrophoretic determination of reduced and oxidized glutathionein red blood cells in the femtomole range. Journal ofChromatography A 676:239–246.

Puri R N. & Meister A (1983) Transport of glutathione, as _-glu-tamylcysteinylglycyl ester, into liver and kidney. Proceedings ofNational Academy of the Sciences, USA 80:5258–5260.

Reed D.J (1986) Regulation of reductive processes by glutathione.Biochemical Pharmacology 35:7–13.

Rigobello M-P, Scutari G, Boscolo, R & Bindoli A (2002) Inhibitionof lipid peroxidation by S-nitrosoglutathione and copper. FreeRadical Research 36:1071–1077.

Roederer M, Ela S-W, Staal F-J-T, Herzenberg LA & HerzenbergLA (1992) N-acetylcysteine: a new approch to anti-HIV therapy.AIDS Research & Human Retroviruses 8:209–217.

Roederer M, Staal F-J, Raju P-A, Ela S-W, Herzenberg L-A &Herzenberg L-A (1990) Cytokine-stimulated human immunodefi-ciency virus replication is inhibited by N-acetyl-L-cysteine.Proceedings of the National Academy of Sciences, USA 87:4884–4888.

Schreck R, Rieber P & Baeuerle P-A (1991). Reactive oxygen inter-mediates as apparently widely used messengers in the activation ofNF-kB transcription factor and HIV-1. EMBO Journal10:2247–2258.

Vogel JU, Cinatl J, Dauletbaev N, Buxbaum S, Treusch G, Cinatl JJr, Gerein V, Doerr HW (2003) Effects of S-acetylglutathione incell and animal model of herpes simplex virus type 1 infection.Medical Microbiology & Immunology (Epub ahead of print)

Wellings D-A & Atherton E (1997) Standard Fmoc Protocols. InMethods Enzimology. Edited by GB Fields. New York: AcademicPress, pp. 44–66.

Wellner V-P, Anderson M-E, Puri R-N, Jensen G-L & Meister A(1984) Radioprotection by glutathione ester; transport of glu-tathione ester into human lymphoid cells and fibroblast.Proceedings of National Academy of the Sciences, USA 81:4732–4735.

85

Antiviral activities of glutathione derivatives

Antiviral Chemistry & Chemotherapy 15:2

Received 3 December 2003; accepted 18 February 2004