Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Tetrahedron Lett. Author manuscript; available in PMC 2010 June 24.
Published in final edited form as:
Tetrahedron Lett. 2009 June 24; 50(25): 3023–3026.
doi:  10.1016/j.tetlet.2009.03.209
PMCID: PMC2697396

Synthesis and redox-enzyme modulation by amino-1,4-dihydro-benzo[d][1,2]dithiine derivatives


A convenient method to prepare a series of benzodithiine derivatives was developed, via the synthesis of cyclic disulfide building blocks containing an amino-group linker. Some of the novel cyclic disulfide compounds are shown to modulate the activity of the redox-enzyme glutathione reductase.

The redox-activities of cyclic disulfides and their reactivity with thiols and relative redox-potentials has been the subject of detailed physico-chemical studies1. More recently, the utility of the cyclic disulfide group has found interest in a variety of applications. For example, cyclic disulfides have been used to serve as a bridging molecule between a gold layer and a single-walled carbon nanotube through a thioalkylthiol linkage2, to attach an oligonucleotide to a gold nanoparticle providing increased stability3, or as a linker to connect folic acid to gold nanoparticles for potential use as drug delivery vehicles4. Although the role of the dithiolane α-lipoic acid in biological processes is well-established, the study synthetic cyclic disulfides remains largely unexplored. Recent examples include the unsubstituted benzodithiine 4 (NO2 replaced by H), which inhibits Respiratory Syncytial Virus replication5 and oxidized dithiothreitol, which interferes with HIV-replication by ejecting zinc from Zinc-finger protein via interaction with active-site sulfhydryl groups6.

In view of the above, a systematic exploration of a broader range of compounds containing cyclic disulfide building blocks could prove interesting. Although various non-cyclic disulfide compounds have been demonstrated to interact with active-site cysteine residues of a number of proteins7, they are prone to intracellular reductive inactivation. The 1,4-dihydro-benzo[d][1,2]dithiines have a much more negative reduction potential than linear disulfides1, which provides increased stability. Therefore, we selected the benzodithiines for further exploration. The most convenient way to potentially synthesize a wide variety of cyclic disulfide derivatives would be via a benzodithiine core that has a handle for further derivatization. We now present the synthesis of substituted 1,4-dihydro-benzo[d][1,2]dithiines that contain an amino-group as a linking unit. These novel building blocks have been reacted with various electrophilic compounds and preliminary studies of their potential biological activity has been carried out via investigation of their potential for modulation of the redox enzyme glutathione reductase.

A general synthetic scheme is provided in figure 1. Starting with either 3- or 4-nitro substituted ortho-xylene, both 5-amino- and 6-amino subsituted benzodithiine derivatives 6 were synthesized respectively. In the first step, the nitroxylene 1 is brominated with bromine in a biphasic methylene chloride/water mixture, from which the α,α′-dibromide 2 can be isolated and purified via a single crystallization. Nucleophilic substitution with potassium thioacetate in methanol provides the bis-thioacetate 3. Direct hydrolysis of 3 with sodium hydroxide in different solvents led to insoluble products, most likely polymeric polysulfides. However, when the hydrolysis was carried out with ammonium hydroxide in high dilution (0.01M) in methanol, concomitant air oxidation provided the cyclic disulfide 4 in good yields8,9. A last challenge was to reduce the nitro-group of compound 4 to an amino-group, while maintaining the reduction-prone cyclic disulfide group intact. Reducing agents such as Fe/HCl, Zn/H2NNH2, Ni/HCOOH or Pd/C/H2, either did not react, or provided a mixture of unidentified or polymeric products. However, reduction with sodium dithionite successfully provided the desired amino-substituted dihydrobenzodithiins 512. Subsequently, the amino group could be reacted with a number of electrophilic compounds to provide the derivatives 6. The products synthesized are summarized in table 1.

Figure 1
Synthesis of cyclic disulfides depicted in Table 1. In brackets: yields for 6-substituted derivatives
Table 1
Preparation of benzodithiine derivatives and their effects on glutathione reductase activity

In order to provide a preliminary investigation of their biological potential, the interaction of the new cyclic disulfide compounds 7 to 13 with commercially available yeast glutathione reductase was investigated. The redox-enzyme glutathione reductase is responsible for the reduction of oxidized glutathione (GSSG) to two molecules of reduced glutathione (GSH), with NADPH as the co-reductant (figure 2). The enzyme maintains GSH concentrations, and possible inhibitors of this enzyme have been indicated as potential anti-malarial13 or anti-cancer agents14. A reversible dithiol-disulfide couple, formed by two active-site cysteine residues provides the key-functionality in the oxidation-reduction process. Based on the reactivity of thiols with disulfides, our novel benzodithiine derivatives seem especially suitable for enzyme modulation, and are intended to be able to interact with the active-site sulfhydryl groups of the enzyme (figure 3). Depending on enzyme kinetics, the benzodithiines could prove to be reversible covalent inhibitors, or alternatively function as competitive substrates, leading to dithiol products. Although an extensive pharmacological study is beyond the scope of the present article, a proof-of-principle for the structure-dependent activity of benzodithiine derivatives to serve as modulators of cysteine-containing redox enzymes was evidenced by their effect on yeast glutathione reductase. The interference with enzymatic activity of the novel benzodithiine products is summarized in table 1.

Figure 2
Reaction catalyzed by glutathione reductase.
Figure 3
Possible interaction of benzodithiine derivatives with the active-site cysteine residues of glutathione reductase

As can be seen from table 1, at a concentration of 50 µM, three of the compounds (8, 9 and 10) reduce the activity of yeast glutathione reductase to 19, 59 and 66% respectively compared with the baseline enzymatic activity. Thus, at this concentration, compound 8 inhibits more than 80% of enzyme activity, as measured by NADPH consumption. At a concentration of 25µM, as can be expected, the effect of all three compounds on enzyme activity is much reduced. Nevertheless, even at this reduced concentration, compound 8 shows an enzyme inhibitory activity of about 50%. For comparison, a bis-dithiocarbamate18 was recently revelealed as a covalent irreversible inhibitor of yeast glutathione reductase with Ki=56µM and kinact=0.1min-1.

In contrast to the above compounds, at both 25 and 50 μM concentrations, the modulatory effect of compounds 11 and 12 on enzyme activity is only moderate at most. These data indicate that the observed effects are not only the result of interaction with the cyclic disulfide group, but are indeed dependent on the complete molecular architecture of the modulating compounds. Interestingly, at a concentration of 50μM, compound 7 seems to exhibit an increase of activity of glutathione reductase, as measured by an increased rate of NADPH consumption19. A similar increase of NADPH consumption was observed in the interaction of human glutathione reductase with either ajoene (50% inhibition within 15 min. at 200μM)21 or with fluoro-M5 (IC50=4.1μM)22. These compounds respectively thioalkylate or alkylate an active-site cysteine residue of the enzyme, thereby inhibiting the reduction of GSSG to GSH. Nevertheless, the covalently inhibited enzyme shows an increased NADPH-oxidase activity, with a faster turnover of NADPH than in the non-inhibited enzyme. In this case the substrate is not GSSG, but rather, either molecular oxygen or a naphtoquinone derivative, that presumably binds to a second, unidentified binding site. We suggest that the observed increased activity of compound 7, when compared to enzyme activity in the absence of a 7, is due to a similar increase in oxidase activity, leading to the observed NADPH consumption23.

In summary, we have developed a procedure for the easy generation of a library of cyclic disulfides, via the preparation of amino-benzodithiine building blocks. These can easily be connected to a variety of electrophiles, some examples of which have been presented. Potential applications of the presented, and of other benzodithiine derivatives, could be in the fields of nanotechnology, as well as in the modulation of redox enzymes. Some of the examples prepared have been shown to interact with glutathione reductase. Possible interactions of other compounds with this novel pharmacophoric group, targeted to other enzymes with active-site cysteine residues will be investigated.


The research was funded by NIH Grant 3-S06-GM08224-21 via the MBRS-SCORE program. The authors thank Mr. Melvin de Jesús at the University of Puerto Rico, Humacao Campus for his assistance with the NMR-measurements.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Lees WJ, Whitesides GM. J Org Chem. 1993;58:642–647.
2. Nakamura T, Isihara TO, Hasegawa YK. Diamond & Related Materials. 2007;16:1091–1094.
3. (a) Letsinger RL, Elghenian R, Viswanadham G, Mirkin CA. Bioconjugate Chem. 2000;11:289–291. [PubMed] (b) Li Z, Jin R, Mirkin CA, Letsinger RL. Nucleic Acids Research. 2002;30:1558–1562. [PubMed]
4. Dixit V, Van den Bossche J, Sherman DM, Thompson DH, Andres RP. Bioconjugate Chem. 2006;17:603–609. [PubMed]
5. Sudo K, Watanabe W, Konno K, Sato R, Kajiyashiki RT, Shigeta S, Yokota T. Antimicrob Agents Chemother. 1999;43:752–757. [PMC free article] [PubMed]
6. Rice WG, Baker DC, Schaeffer CA, Graham L, Bu M, Terpening S, Clanton D, Schultz R, Bader JP, Buckheit RW, Jr, Field L, Singh PK, Turpin JA. Antimicrob Agents Chemother. 1997;41:419–426. [PMC free article] [PubMed]
7. (a) Baum E, Siegel MM, Bebernitz GA, Hulems JD, Sridharan L, Sun L, Tabei K, Johnston SH, Wildey MJ, Nygaard J, Jones TR, Gluzman Y. Biochemistry. 1996;35:5838–5846. [PubMed] (b) Boerzel H, Koeckert M, Bu W, Spingler B, Lippard SJ. Inorg Chem. 2003;42:1604–1615. [PubMed]
8. Synthesis of 6-nitro-1,4-dihydro-benzo[d][1,2]dithiine 4 (m-nitrobenzodithiine): In a 2L erlenmeyer, 2.99g (10.0mmol) 3, was dissolved in 1000mL methanol. To this solution was added 5.8 mL NH4OH and, while maintaining the vessel open to air, the solution was stirred vigorously for 16 hours at room temperature, to obtain a dark-brown solution. The methanol was removed in vacuo on the rotavap (can be recycled for a subsequent batch of the reaction) and the remaining solid was extracted with water and ethyl acetate (2×) (THF may be added to aid in dissolution of the solids). Extraction of the combined organic phases with brine, drying on Na2SO4 and removal of the solvents gave a dark-brown solid that was purified by silica gel column chromatography (hexanes/ethyl acetate, 6:1) to provide 1.79g (8.4 mmol) of a yellowish solid which by TLC (hexanes/ethyl acetate=6:1: Rf = 0.78) contained a minor impurity (Rf 0.72) (probably a dimeric tetrasulfide product, as more of it is observed when the reaction is carried out at higher concentrations). Although further purification by column chromatography is possible, the product can be utilized as obtained for the next step. 1H NMR (DMSO-d6) δ 4.28 (2H, s), 4.30 (2H, s), 7.47 (1H, d, J=8.4 Hz), 8.06 (1H, dd, J=8.6Hz, J=2.4Hz), 8.11 (1H, d, J=2.4Hz); 13C NMR δ 34.1, 34.7, 114.0, 116.1, 122.8, 131.0, 133.5, 144.9; HRMS m/z calcd. for [C8H7O2N32S2 + H]+ 213.9991, found 213.9992.
9. A possible alternative method could be to convert the α,α′-dibromide 2 directly into cyclic disulfide 4 via reaction with Na2S210, if needed in the presence of a phase transfer catalyst10b. However, as aqueous polysulfide Sn2- consists of various homologous anions in equilibrium (n=1,2,3…), the resulting product mixture might contain products with different amounts of Sn that are difficult to separate at a preparative scale11. As our further investigations required cyclic disulfides without the possibility of contamination with mono-, tri- or polysulfides, we preferred the above-described multi-step procedure (figure 1).
10. (a) Harpp DN, Gleason JG. J Org Chem. 1970;35:3258–3263. (b) Sonavane SU, Chidambaram M, Khalil S, Almog J, Sasson Y. Tetrahedron Lett. 2008;49:520–522. and references cited.
11. (a) Rys AZ, Abu-Yousef IA, Harpp DN. Tetrahedron Lett. 2008;49:6670–6673. (b) Steudel R. Chem Rev. 2002;102:3905–3946. [PubMed] (c) Kamyshny A, Ekeltchlk I, Gun J, Lev O. Anal Chem. 2006;78:2631–2639. [PubMed]
12. Synthesis of 6-amino-1,4-dihydro-benzo[d][1,2]dithiine 5. A solution of 1.65g (7.7 mmol) 4 in methanol/THF/water (60mL/20mL/20mL) was heated in an oil-bath of 60°C. To this solution was added in one portion 5.53 g (27.0 mmol) 85% Na2S2O4 (should be fresh). After 30 minutes, the reaction was shown to be finished by TLC, and after cooling to room temperature, CH2Cl2 and 10% K2CO3 were added. The precipitated solids were removed by vacuum filtration, and the aqueous phase was extracted 2× with CH2Cl2. The combined organic phases were extracted with water, dried on Na2SO4 and after removing the solvent, the obtained solid was purified by silica gel column chromatography (hexanes ethyl acetate 3:1) to obtain 0.60 g (3.3 mmol) of an off-white solid. 1H NMR (CDCl3) δ 3.59 (2H, br s), 3.95 (2H, s), 3.96 (2H, s), 6.40 (d, J=2.4Hz), 6.52 (1H, dd, J=8.4Hz, J=2.4Hz), 6.86 (1H, d, J=8.4Hz); 13C NMR δ 34.1, 34.7, 114.0, 116.1, 122.8, 131.0, 133.5, 144.9; HRMS m/z calcd. for [C8H9N32S2 + H]+ 184.0249, found 184.0250.
13. (a) Sarma GN, Savvides SN, Becker K, Schirmer M, Karplus PA. J Mol Biol. 2003;328(4):893–907. [PubMed] (b) Friebolin W, Jannack B, Wenzel N, Furrer J, Oeser T, Sanchez CP, Lanzer M, Yardley V, Becker K, Davioud-Charvet E. J Med Chem. 2008;51:1260–1277. [PubMed]
14. (a) Seefeldt T, Dwivedi C, Peitz G, Herman J, Carlson L, Zhang Z, Guan X. J Med Chem. 2005;48:5224–5231. [PubMed] (b) Rice KP, Penketh PG, Shyam K, Sartorelli AC. Biochem Pharmacol. 2005;69:1463–1472. [PubMed]
15. Zhang H, Cai Q, Ma D. J Org Chem. 2005;70:5164–5173. [PubMed]
16. The reduction of GSSG by yeast glutathione reductase was determined similar as described inBoese M, Keese MA, Becker K, Busse R, Mülsch A. J Biol Chem. 1997;272:21767–21773. [PubMed]In short: The enzyme (Sigma-Aldrich) was diluted (1 in 100) in assay buffer (200mM KCl, 1mM EDTA, 50mM potassium phosphate, pH 6.9). Seperately, 1.85 mg NADPH was dissolved in 1 mL assay buffer and 14.2 mg GSSG in 2mL assay buffer. Compounds 7 to 13 were dissolved in DMSO to obtain 5mM and 2.5mM solutions. At room temperature, 790μL assay buffer was transferred to a quartz cuvette, followed by 50μL enzyme solution, 50μL NADPH solution and 10μL compound solution (or 10μL DMSO for baseline enzyme activity). After 2 min., 100μL GSSG was added and the rate of consumption of NADPH was measured for 3 minutes (scan every 5 seconds) via its absorption at 340nm. Each compound concentration was measured in triplicate, and the rate of NADPH consumption was compared with baseline enzyme activity.
17. For the synthesis of related N-phenylphtalanilic acidsPerry CJ, Parveen ZJ. Chem Soc, Perkin Trans. 2(2001):512–521.
18. Bisdithiocarbarmate inhibitor: 2-acetylamino-3-[4-(2-acetylamino-2-carboxyethylsulfanylthiocarbonylamino)-phenylthiocarbamoylsulfanyl]propionic acidSeefeldt T, Zhao Y, Chen W, Raza AS, Carlson L, Herman J, Stoebner A, Hanson S, Foll R, Guan X. J Biol Chem. 2009;284:2729–2737. [PMC free article] [PubMed]
19. The dithiol derivative of compound 7 was prepared independently via reduction with tributylphosphine/H2O20. Separate measurements indicated very low absorption of both 7 and its dithiol reduction product at 340nm, demonstrating that the observed increase in absorption is not due to susbtrate interference in the assay procedure, but confirms an increase of NADPH consumption.
20. Ayers JT, Anderson SR. A preparative scale reduction of alkyl disulfides with tributyl phosphine and water. Synthetic Commun. 1999;29(3):351–358.
21. Ajoene: (E,Z)-4,5,9-trithiadodeca-1,6,11-triene 9-oxide)Gallwitz H, Bonse S, Martinez-Cruz A, Schlichting I, Schumacher K, Krauth-Siegel RL. J Med Chem. 1999;42:364–372. [PubMed]
22. Fluoro-M5: 6-[2′-(3′-fluoromethyl)-1′,4′-naphtoquinolyl]-hexanoic acidBauer H, Fritz-Wolf K, Winzer A, Khner S, Little S, Yardley V, Vezin H, Palfey B, Schirmer RH, Davioud-Charvet E. J Am Chem Soc. 2006;128:10784–10794. [PubMed]
23. The observed effect of increased NADPH consumption observed at the lower concentration of compound 13 could be suggested to occur via a similar mechanism.