Synthesis, Catalytic GPx-like Activity, and SET Reactions of Conformationally Constrained 2,7-Dialkoxy-Substituted Naphthalene-1,8-peri-diselenides

Abstract

Several 2,7-dialkoxy-substituted naphthalene-1,8-peri-diselenides were prepared and tested for catalytic antioxidant activity in an NMR-based assay employing the reduction of hydrogen peroxide with stoichiometric amounts of benzyl thiol. Acidic conditions enhanced their catalytic activity, whereas basic conditions suppressed it. The highest activity was observed with a 2,7-bis(triethyleneglycol) derivative. These compounds serve as mimetics of the antioxidant selenoenzyme glutathione peroxidase. Studies based on NMR peak-broadening effects and EPR spectroscopy indicated that a thiol-dependent SET reaction occurs under the conditions of the assay, which can be reversed by the addition of triethylamine. In contrast, peak broadening induced by proton-catalyzed electron transfer during the treatment of naphthalene-1,8-peri-diselenides with trifluoroacetic acid can be suppressed by the addition of excess thiol. These observations provide new insights into the redox mechanisms of these processes.

Introduction

Naphthalene peri-diselenide 1a was first reported by Meinwald and Wudl et al.(1) by the lithiation and selenation of 1,8-dibromonaphthalene. This and related compounds have proven to be of interest as electron donors in charge transfer complexes and organic conductors.² Woollins et al. reported the solid-state ⁷⁷Se NMR spectrum³ and X-ray crystal structure⁴ of 1a, as well as complexes of 1a and its derivatives with various metals and metalloids.⁵ Diselenide 1a and its diselenol analogue also effect the deiodination of thyroxine, as demonstrated by Manna, Mugesh and Mondal,⁶ while Grainger et al. employed 1a and its congeners as mimetics of [FeFe]-hydrogenase.⁷

As part of our ongoing studies⁸ of organoselenium compounds that serve as mimetics of the selenoenzyme glutathione peroxidase (GPx),⁹ we considered that 1a and its analogues might be effective for this purpose. GPx protects cells from oxidative stress caused by hydrogen peroxide and other reactive oxygen species (ROS) that are formed during aerobic metabolism by catalyzing the reduction of ROS with the tripeptide thiol glutathione.¹⁰ Oxidative stress is in turn implicated in numerous diseases and degenerative conditions,¹¹ examples of which include reperfusion injury in heart attack and stroke patients,¹² as well as hearing loss,¹³ for which the selenium compound ebselen has been tested in clinical trials. Ebselen is also in clinical trials as a lithium surrogate in the treatment of bipolar disorder.¹⁴ In acute cases, where ROS levels are particularly elevated, GPx is overwhelmed, and administration of a supplementary antioxidant is desirable to suppress excessive oxidative stress.

Typically, acyclic diaryl diselenides have C–Se–Se–C dihedral angles that are close to orthogonal, presumably to minimize destabilizing selenium lone pair interactions. For example, the dihedral angle in diphenyl diselenide was determined to be 82.0° by X-ray crystallography.¹⁵ However, several years ago we found that the highly sterically hindered diselenide 2 (Figure 1) displayed a dihedral angle of 112.1°, along with a bathochromic shift in its UV–visible spectrum compared to less hindered aliphatic diselenides.¹⁶ This indicated that the conformational constraint imposed by the bulky t-butyl substituents forced an increase in the dihedral angle and simultaneously raised the energy level of the HOMO while decreasing the HOMO–LUMO gap in 2. Since selenium oxidation is generally the rate-determining step in the catalytic cycle of various GPx mimetics,^8a,17 this observation suggested that the catalytic activity of other diselenides might be further enhanced by constraining their dihedral angles from the usual nearly orthogonal geometry to a coplanar one, thus again raising their HOMO energies and lowering their oxidation potentials. The rigid naphthalene diselenide 1a, where the C–Se–Se–C dihedral angle is essentially 0° (Figure 1), was chosen to test this possibility, as reported in our preliminary communication.¹⁷ Thus, 1a displayed catalytic activity ca. 13 times greater than that of diphenyl diselenide, while the 2,7-dimethoxy derivative 1b revealed a further increase in activity to 17.4 times that of the diphenyl derivative, consistent with mesomeric stabilization of positive charge during the rate-determining selenium oxidation step. As a result, it was desirable to retain oxygen substituents at the 2- and 7-positions in the investigation of new compounds.

Figure 1.

C–Se–Se–C dihedral angles of diselenides.

It is also recognized that intramolecular coordination can have profound effects on the reactivity of selenium compounds.^18,19 Since the angle strain associated with the four-centered O–Se interaction in 1b impedes intramolecular coordination, it was of interest to investigate less studied structures where six- or seven-centered coordination might be possible. Moreover, the installation of longer alkoxy chains on the naphthalene moiety, along with more highly oxygenated analogues, would also provide a range of compounds with varying lipophilic and hydrophilic properties, for eventual optimization of bioavailability.

We had previously discovered that naphthalene peri-diselenides 1a–1c, as well as their sulfur and tellurium analogues, undergo a facile proton-catalyzed electron transfer (PCET) reaction that could be suppressed under basic conditions, as evidenced by NMR and EPR experiments.²⁰ However, the effect of PCET or related processes on the catalytic GPx-like activity of naphthalene peri-diselenides was unknown. We now report the preparation of several additional 2,7-disubstituted analogues that contain either long-chain alkoxy substituents, or more highly oxygenated ones, and the in vitro assessment of their catalytic activities, as well as new insights into the mechanism of such processes.

Results and Discussion

Compounds 1b(17) and 1c(20) were obtained by variations of the original method of Meinwald and Wudl,¹ along with new analogues 1d–1g. In general, 2,7-dihydroxynaphthalene (3) was O-alkylated to afford 4 and brominated selectively at the 1,8-positions with N-bromosuccinimide (NBS), providing 5. Metalation and selenation, followed by aerial oxidation, then afforded the desired products 1b–1g. Alternatively, 3 was first brominated and then alkylated to afford 5via6. These processes are summarized in Scheme 1.

Scheme 1. Synthesis of Naphthalene peri-Diselenides.

With diselenides 1b–1g in hand, we proceeded to measure their catalytic properties in promoting the reduction of hydrogen peroxide with benzyl thiol as the stoichiometric reductant. In our previous investigations of compounds 1a and 1b, an HPLC-based assay was successfully employed. However, the widely differing solubilities of the present range of diselenides in various available mobile phases precluded their direct comparison by HPLC analysis. Consequently, an NMR-based assay was developed, where each of 1b–1g could be run under the same conditions. Thus, a mixture of benzyl thiol, dimethyl terephthalate (DMT, internal standard), and 10 mol % of the catalyst (relative to benzyl thiol) was stirred in CDCl₃-CD₃OD (95:5) at 18 °C. A small excess of 50% H₂O₂ was added, and the mixture was periodically analyzed by comparison of the NMR integration of the methylene singlet of dibenzyl disulfide at 3.57 ppm with the aryl signal of the internal standard at 8.06 ppm.²¹ A typical NMR assay with diselenide 1b as the catalyst is shown in Figure 2, indicating the increase in dibenzyl disulfide concentration with time. Each diselenide was run at least three times, and the averaged plots of % yield of dibenzyl disulfide versus time are provided in Figure 3. The linear nature of these plots is consistent with zero-order kinetics of a catalytic system. The time required for 50% completion of the oxidation of the thiol to its corresponding disulfide (t_1/2) also provides a convenient parameter for comparing the catalytic activities of various types of GPx mimetics.^8a,22 These values are provided for 1b–1g in Table 1. Diphenyl diselenide and the unsubstituted naphthalene peri-diselenide 1a are included for comparison.

Figure 2.

¹H NMR assay of diselenide catalyst 1b in the oxidation of benzyl thiol with hydrogen peroxide. The ¹H NMR spectra were recorded at 400 MHz in CDCl₃-CD₃OD (95:5).

Figure 3.

(a–f) Kinetic plots for assays of diselenide catalysts: (a) 1b, (b) 1c, (c) 1d, (d) 1e, (e) 1f, (f) 1g. Reactions were performed at 18 °C in CDCl₃–CD₃OD (95:5). Initial concentrations were as follows: benzyl thiol, 0.031 M; H₂O₂, 0.035 M; catalyst, 0.0031 M; and DMT, 0.0155 M. The formation of dibenzyl disulfide was monitored by ¹H NMR spectroscopy via integration of the disulfide methylene signal at 3.57 ppm vs the aromatic signal of DMT at 8.06 ppm. The plots are averages of either three or four runs.

Table 1. Catalytic Activities of Diselenides 1a–1g^a.

entry	diselenide	additiveb	t1/2 (h)c
1	PhSeSePh	nil	(129)
2	1a	nil	(9.7)
3	1b	nil	7.1 (7.4)
4	1b	1 mol % TFA	4.2
5	1b	10 mol % TFA	3.0
6	1b	100 mol % Py-d5	13.4
7	1c	nil	6.7
8	1d	nil	6.8
9	1e	nil	5.1
10	1f	nil	6.5
11	1g	nil	2.3

^aReactions were performed at 18 °C in CDCl₃–CD₃OD (95:5). Initial concentrations were as follows: benzyl thiol, 0.031 M; H₂O₂, 0.035 M; catalyst, 0.0031 M; and DMT, 0.0155 M.

^bTFA is trifluoroacetic acid, and Py-d₅ is deuterated pyridine.

^cThe values of t_1/2 (time taken for 50% completion of the oxidation of the thiol to its disulfide) are averages of either three or four runs. The values in parentheses were taken from ref (17) and were obtained under slightly different conditions via an HPLC-based assay.

These results indicate that the n-pentyloxy and n-dodecyloxy substituents of 1c and 1d, respectively, produced similar catalytic effects to that of the methoxy derivative 1b (compare entries 7 and 8 with entry 3 in Table 1). This was expected because all three diselenides are subject to comparable mesomeric effects from electron-donating alkoxy groups but lack an additional nucleophilic center for coordination with selenium during the oxidation step. On the other hand, the methoxymethyl analogue 1e, where a second oxygen is able to coordinate with selenium via a six-membered cyclic interaction, showed a modest improvement in activity [t_1/2 = 5.1 h for 1evs 7.1 h for 1b (entries 9 and 3, respectively, in Table 1)], while such an enhancement was less evident in the (methoxyethoxy)methoxy analogue 1f (t_1/2 = 6.5 h; entry 10, Table 1).²³ In contrast, the triethylene glycol derivative 1g revealed a more pronounced effect, affording a significantly faster rate (t_1/2 = 2.3 h; entry 11, Table 1) than any of the other diselenides shown in Table 1.

The mechanism for the catalytic activity of 1a and 1b was reported in our preliminary communication¹⁷ and is shown in more detail in Scheme 2. The diselenides were slowly oxidized by hydrogen peroxide to the isolable selenolseleninates 8, presumably formed from the hydroxyselenonium intermediates 7, followed by rapid reduction back to the parent diselenides by benzyl thiol via9 and/or 10. It is possible that the modest to significant rate enhancements observed in 1e–1g are due to coordination effects, as shown by structure 11 (Figure 4), resulting in a further increase in the reactivity of the selenium atoms toward rate-determining oxidation. However, it is not entirely clear why 1g, where such coordination would require a seven-membered or larger ring, produced a faster reaction than 1e or 1f, where enhanced coordination via a six-centered structure would be expected. A possible explanation is that the anomeric effect in acetals 1e and 1f (i.e., structure 11 where n = 1 in Figure 4) decreases the magnitude of mesomeric electron donation from the alkoxy oxygen atom to the naphthalene diselenide moiety. Since electron donation increases the rate of the rate-determining oxidation of the diselenide moiety, the anomeric effect in 1e and 1f is expected to decrease their reaction rates relative to that of 1g (i.e., structure 11 where n = 2), where such anomeric deactivation is not possible.²⁴ However, the alkoxy derivatives 1b–1d, where the anomeric effect is similarly precluded, also show lower reactivity than 1g. In this case, the decreased reactivity may be attributed to the lack of a second oxygen atom capable of coordinating with selenium. Thus, the relative reaction rates of the diselenides, as shown in Table 1, are dependent upon a balance of mesomeric, coordination, and, in some examples, anomeric effects.

Scheme 2. Catalytic Cycle for the Reduction of Hydrogen Peroxide with Naphthalene peri-Diselenides and Benzyl Thiol.

Figure 4.

Possible O···Se coordination in naphthalene peri-diselenides 1e–1g.

It was also of interest to determine whether the previously discovered PCET reaction of naphthalene peri-chalcogenides²⁰ (e.g., as shown for 1b in Scheme 3) played a role in the catalytic activity of compounds 1b–1g. This phenomenon was first identified when various naphthalene peri-dichalcogenides were exposed to increasing concentrations of acids, which resulted in extreme broadening, coalescence, and eventually complete disappearance of their ¹H and ¹³C NMR signals, attributed to the formation of paramagnetic species. The original spectra were restored upon addition of excess pyridine-d₅.

Scheme 3. PCET Reaction of Naphthalene peri-Diselenide 1b.

During the assays of diselenides 1b–1g for GPx-like activity, where both hydrogen peroxide and benzyl thiol were present, such peak broadening was also observed but was less evident because of the low concentration of the catalyst (10 mol %).²⁵ Moreover, the presence of trifluoroacetic acid (TFA) resulted in a significantly faster rate of reaction with 1b than that in its absence (entries 4 and 5 in Table 1). When pyridine-d₅ was included instead of TFA, the reaction rate was considerably suppressed (entry 6, Table 1). These results are parallel to those for the PCET reaction of naphthalene peri-diselenides, where TFA promoted NMR peak broadening and coalescence, while pyridine-d₅ reversed the effect. While it is therefore tempting to assume that the formation of the radical and radical cation species from PCET enhances the catalytic activity of 1b in the presence of TFA, this effect could instead reflect the increased reactivity of hydrogen peroxide through its protonation by the acid or via the in situ formation of the corresponding peroxytrifluoroacetic acid. In the absence of TFA, where PCET is precluded, we first considered that peak broadening under the normal assay conditions was the result of single-electron transfer (SET instead of PCET) from diselenides 1 to the more electrophilic selenolseleninates 8 or their precursors 7. However, when diselenide 1b was treated with 50 mol % of hydrogen peroxide in order to generate an equimolar mixture of 1b and 8b, the ¹H NMR spectrum of the resulting solution indicated no peak broadening, thereby ruling out this possibility (Figure 5). This experiment suggests that, in the absence of an acid (and associated PCET), the presence of the thiol is required to generate paramagnetic species. We therefore postulate that intermediate 9b, formed by thiolysis of 8b(26) (as shown in Scheme 2), serves as the SET acceptor under the conditions of catalytic assay. Further reaction of the selenenyl sulfide moiety of radical 15b with benzyl thiol then produces radical anion 16b, followed by charge annihilation and regeneration of 1b. This tentative mechanism is shown in Scheme 4.²⁷

Figure 5.

¹H NMR spectrum of 1:1 mixture of diselenide 1b and selenolseleninate 8b. The ¹H NMR spectrum was obtained in CDCl₃-CD₃OD (95:5) at 400 MHz by treating 1b with 50 mol % of H₂O₂ for 4 h.

Scheme 4. SET Reaction of Diselenide 1b after Treatment with Benzyl Thiol.

An EPR spectrum of the reaction mixture from 1b under the usual assay conditions (Figure 6) revealed the presence of paramagnetic species, consistent with the SET process in Scheme 4.²⁸ An additional experiment was performed to confirm the role of the thiol in initiating radical formation from the otherwise inert 1:1 mixture of diselenide 1b and selenolseleninate 8b, as generated in Figure 5. Again, the 1:1 mixture revealed no line broadening in the NMR spectrum (Figure 7a), but when treated with a substoichiometric amount (20 mol %) of benzyl thiol in the absence of hydrogen peroxide, the signals from the diselenide completely disappeared, leaving only those from the remaining unreacted selenolseleninate (Figure 7b). This is consistent with the required formation of 9b from the reaction of the thiol with 8b as a precondition to the electron transfer shown in Scheme 4. Finally, the addition of triethylamine to the latter mixture effected an immediate restoration of the NMR signals of diselenide 1b (Figure 7c). We considered that this could be attributed to the reversal of the formation of 9b from 8b under basic conditions, as shown in path A of Scheme 5, or from triethylamine attack at the sulfur atom of 9b (path B of Scheme 5). However, NMR integration of the methyl signals of 1b and 8b at the start and end of the experiment revealed that the proportion of the diselenide relative to the selenolseleninate increased from the initial 1:1 ratio to ca. 1.5:1, as expected from the consumption of 20 mol % of 8b through its reaction with the thiol to afford an additional 20 mol % of 1bvia path B.

Figure 6.

EPR spectrum of diselenide 1b in the presence of benzyl thiol and hydrogen peroxide under the usual conditions of the antioxidant assay.

Figure 7.

¹H NMR peak broadening during the reaction of diselenide 1b and selenolseleninate 8b with benzyl thiol and triethylamine. (a) Spectrum of diselenide 1b (0.05 mmol) and selenolseleninate 8b (0.05 mmol); (b) after addition of 0.01 mmol of benzyl thiol; and (c) after addition of 0.036 mmol of triethylamine. ¹H NMR spectra were obtained at 400 MHz in CDCl₃-CD₃OD (95:5). The complete spectra showing integration and chemical shifts are provided in the Supporting Information.

Scheme 5. Reaction of Postulated Intermediate 9b with Triethylamine.

A separate experiment related to our previously reported PCET studies was also performed, where 1c was first treated with 50 mol % of TFA in the absence of hydrogen peroxide and thiol, resulting in PCET and the usual coalescence and disappearance of ¹H NMR peaks. Interestingly, and in contrast to the SET process in the absence of acid, where the presence of thiol is required, the normal NMR spectrum was restored when a large excess of benzyl thiol was added subsequently to that of the acid (Figure 8). While the precise mechanism for this quenching effect is not presently known, it appears that the thiol interrupts the electron transfer step when present in high concentrations.

Figure 8.

Effect of benzyl thiol and TFA on peak broadening of catalyst 1c. (a) 1c: 15.5 mg, 0.034 mmol. (b) TFA: 1.3 μL, 1.9 mg, 0.017 mmol. (c) Benzyl thiol (1st portion): 2.0 μL, 2.1 mg, 0.017 mmol. (d) Benzyl thiol (2nd portion): 2.0 μL, 2.1 mg, 0.017 mmol. (e) Benzyl thiol (3rd portion): 36 μL, 38 mg, 0.306 mmol. The ¹H NMR spectra were taken at 400 MHz. The spectra were recorded in 1.0 mL of CDCl₃.

Conclusions

In conclusion, the GPx-like catalytic activities of a series of 2,7-dialkoxynaphthalene peri-diselenides 1b–1g, including the novel compounds 1e–1g, were measured by means of an NMR-based assay employing hydrogen peroxide as the oxidant and benzyl thiol as the stoichiometric reductant. In particular, the substituents in compounds 1e–1g contain both an oxygen that provides mesomeric electron donation to the diselenide moiety, which is further activated by its constrained conformation, as well as by a second oxygen capable of coordinating with the proximal selenium atom via a six- or seven-centered interaction. The anomeric effect in acetals 1e and 1f may account for their reduced activities compared to that of the triethylene glycol derivative 1g, where no such effect is possible. The rate of disulfide formation is also enhanced by TFA and is suppressed by pyridine-d₅.

In the absence of TFA and under the usual assay conditions, peak broadening was again observed and is attributed to a SET reaction where diselenide 1 acts as the electron donor and selenenyl sulfide 9 serves as the acceptor. The possibility that selenolseleninate 8 functions as the electron acceptor instead of 9 was ruled out by the failure to observe NMR peak broadening in an equimolar mixture of 1b and 8b, while the dramatic broadening and disappearance of the signals of 1b upon addition of 20 mol % of benzyl thiol provide further support of 9b as the acceptor. The regeneration of the diselenide signals at the expense of those of the selenolseleninate when triethylamine was introduced is consistent with interception of 9b by the amine, resulting in the suppression of SET by removal of the postulated electron acceptor 9b, while the increase in the ratio of 1b to 8b during this process indicates that path B is favored over path A in Scheme 5.

In any event, NMR spectroscopy is highly sensitive to the presence of paramagnetic species and their formation likely comprises a relatively minor side reaction of the processes involved in the assay of GPx-like activity of diselenides 1. Overall, 1g affords a ca. three-fold improvement in catalytic activity compared to the previously reported 2,7-dimethoxy derivative 1b, and the substituents in 1b–1g provide a broad range of hydrophilic and lipophilic properties that may be useful in optimizing the biological activity.

Experimental Section

General Experiment

All synthetic reactions were performed using oven-dried glassware under a nitrogen atmosphere, unless otherwise indicated. THF was dried over LiAlH₄ and was freshly distilled before use. Hydrogen peroxide was titrated before use and had a concentration of 50 ± 1%. The reported yields are based on isolated products. Reaction temperatures are reported as the temperature of the bath. ¹H, ¹³C, and ⁷⁷Se NMR spectra were recorded in CDCl₃ unless otherwise noted, at 400, 101, and 76 MHz, respectively. Diphenyl diselenide was employed as an external reference (δ 463 ppm relative to Me₂Se) for ⁷⁷Se NMR spectra.²⁹¹³C NMR spectra were recorded with broadband proton decoupling. Traces of acid were removed from CDCl₃ by treatment with anhydrous K₂CO₃, followed by filtration. NMR multiplets are reported as: s = singlet, d = doublet, t = triplet, q = quartet, and m = multiplet. HRMS were obtained using either electrospray ionization (ESI) or electron impact (EI), as indicated.2,7-Dihydroxynaphthalene (3) was obtained from commercial sources, and compounds 1b,¹⁷1c,²⁰ and 6(30) were prepared, as reported previously.

2,7-Di(n-dodecyloxy)naphthalene (4d)

A mixture of 2,7-dihydroxynaphthalene (3) (1.62 g, 10.1 mmol), K₂CO₃ (6.9 g, 50 mmol), and 1-bromododecane (7.5 mL, 7.8 g, 31 mmol) in 30 mL of DMF was heated at 95 °C for 16 h. It was cooled to room temperature and poured into 100 mL of water. The precipitate was filtered, washed with water and methanol, and dried in vacuo. The light-brown powder (4.82 g, 96%) was of sufficient purity for use in the next step. A sample was purified by flash chromatography (99:1 hexanes/ethyl acetate) followed by recrystallization from hexanes, to afford 2,7-bis(dodecyloxy)naphthalene as a white solid: mp 60–63 °C; ¹H NMR (400 MHz, CDCl₃): δ 7.64 (d, J = 8.9 Hz, 2H), 7.04 (d, J = 2.3 Hz, 2H), 6.99 (dd, J = 8.8, 2.4 Hz, 2H), 4.06 (t, J = 6.6 Hz, 4H), 1.85 (crude pentet, J = 7.05 Hz, 4H), 1.56–1.45 (m, 4H), 1.28 (m, 32H), 0.90 (t, J = 6.8 Hz, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 157.8, 136.2, 129.2, 124.3, 116.4, 106.2, 68.2, 32.1, 29.83, 29.80, 29.78, 29.76, 29.6, 29.51, 29.45, 26.3, 22.9, 14.3; MS (EI-TOF) (m/z, %) 496 (M⁺, 100), 328 (5), 160 (10); HRMS (EI-TOF) calcd for C₃₄H₅₆O₂ (M⁺), 496.4280; found, 496.4280.

1,8-Dibromo-2,7-di(n-dodecyloxy)naphthalene (5d)

N-Bromosuccinimide (3.80 g, 21.4 mmol) and pyridine (1.72 mL, 1.68 g, 21.3 mmol) were dissolved in 100 mL of dichloromethane, and 2,7-di(n-dodecyloxy)naphthalene (4d) (4.82 g, 9.70 mmol) was added. The solution was refluxed for 3 h under a nitrogen atmosphere. The solution was cooled to room temperature and concentrated in vacuo to ca. 40 mL. It was then poured into methanol, and the resulting precipitate was filtered, washed with methanol, and dried in vacuo. Product 5d was recrystallized from hexanes as white crystals (4.38 g, 69%): mp 81–82 °C; ¹H NMR (400 MHz, CDCl₃): δ 7.68 (d, J = 9.0 Hz, 2H), 7.11 (d, J = 8.9 Hz, 2H), 4.14 (t, J = 6.5 Hz, 4H), 1.88 (crude pentet, J = 7.0 Hz, 4H), 1.59–1.50 (m 4H), 1.27 (m, 32H), 0.89 (t, J = 6.8 Hz, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 156.6, 132.4, 130.3, 127.9, 113.5, 107.3, 70.9, 32.4, 30.14, 30.12, 30.06, 30.04, 29.90, 29.83, 29.81, 26.5, 23.2, 14.6; MS (EI-TOF) (m/z, %) 654 (M⁺, 25), 576 (100), 408 (15), 240 (60), 55 (15); HRMS (EI-TOF) calcd for C₃₄H₅₄⁷⁹Br⁸¹BrO₂ (M⁺), 654.2470; found, 654.2461.

3,8-Di(n-dodecyloxy)naphtho[1,8-cd]-1,2-diselenole (1d)

1,8-Dibromo-2,7-di(n-dodecyloxy)naphthalene (5d) (1.99 g, 3.04 mmol) was dissolved in 90 mL of dry THF, and the solution was cooled to −78 °C. n-Butyllithium (5.13 mL, 1.6 M in hexanes, 8.21 mmol) was then added dropwise, and the reaction mixture was stirred at −78 °C for 30 min, then stirred at 0 °C for 30 min, and for an additional 30 min at room temperature. The solution was then cooled to 0 °C, and elemental selenium (0.676 g, 8.56 mmol) was added. The mixture was warmed to room temperature and stirred for 1 h. The reaction was quenched with saturated NH₄Cl, and air was bubbled through the mixture for 30 min. The mixture was filtered through celite, and the latter was washed with ethyl acetate and dichloromethane. The combined organic layers were washed with brine, dried with anhydrous Na₂SO₄, and evaporated under reduced pressure. The resulting solid was triturated with methanol and diethyl ether until the washings were no longer yellow. Product 1d was recrystallized from hexanes as purple needles (0.833 g, 42%): mp 84–85 °C; ¹H NMR (400 MHz, CDCl₃): δ 7.52 (d, J = 8.8 Hz, 2H), 6.95 (d, J = 8.8 Hz, 2H), 4.14 (t, J = 6.4 Hz, 4H), 1.82 (crude pentet, J = 7.0 Hz, 4H), 1.50 (crude pentet, J = 7.3 Hz, 4H), 1.28 (s, 32H), 0.89 (t, J = 6.8 Hz, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 152.6, 139.9, 127.8, 125.4, 123.6, 113.0, 69.5, 31.9, 29.66, 29.63, 29.58, 29.56, 29.44, 29.35, 26.0, 22.7, 14.1; ⁷⁷Se{¹H} NMR (CDCl₃, 76 MHz) δ 405.2; MS (ESI-QTOF) (m/z, %), 655 ([M + H]⁺, 100), 577 (20), 497 (10), 371 (30), 220 (25), 101 (15); HRMS (EI-TOF) calcd for C₃₄H₅₄O₂⁸⁰Se₂ (M⁺), 654.2454; found, 654.2475. Anal. Calcd for C₃₄H₅₄O₂Se₂: C, 62.56; H, 8.34. Found: C, 62.30; H, 8.09.

1,8-Dibromo-2,7-di(methoxymethoxy)naphthalene (5e)³¹

To a solution of 1,8-dibromonaphthalene-2,7-diol (6) (4.74 g, 14.9 mmol) in THF was added NaH (787 mg, 32.8 mmol) at 0 °C, followed by chloromethyl methyl ether (2.48 mL, 2.63 g, 32.7 mmol). The solution was stirred at room temperature for 4 h and then poured into ice-cold water and extracted with ethyl acetate. The organic extracts were combined, washed with water and brine, dried over anhydrous Na₂SO₄, and evaporated in vacuo. The crude mixture was purified by flash chromatography (hexanes-ethyl acetate, 2:1, increasing to 3:2) to yield product 5e as a light brown solid (5.7 g, 94%). This compound decomposed to an intractable black solid after a few days at room temperature. A fresh sample gave: mp 92–94 °C; ¹H NMR (400 MHz, DMSO-d₆): δ 7.96 (d, J = 9.0 Hz, 2H), 7.45 (d, J = 9.0 Hz, 2H), 5.41 (s, 4H), 3.45 (s, 6H); ¹³C{¹H} NMR (101 MHz, DMSO-d₆): δ 154.5, 131.2, 130.8, 128.4, 115.6, 106.6, 95.4, 56.6; HRMS (ESI): calcd for C₁₄H₁₄⁷⁹Br₂O₄, 404.9332 (M + H); found, 404.9334 (M + H). Anal. Calcd for C₁₄H₁₄O₄Br₂: C, 41.41; H, 3.48. Found: C, 41.15; H, 3.48.

3,8-Di(methoxymethoxy)naphtho[1,8-cd]-1,2-diselenole (1e)

n-Butyllithium (10.1 mL, 2.5 M in hexanes, 25 mmol) was added to a solution of 1,8-dibromo-2,7-di(methoxymethoxy)naphthalene (5e) (3.7 g, 9.1 mmol) in THF at −78 °C. The solution was slowly warmed to room temperature and then stirred for 1.5 h. It was cooled to 0 °C, elemental selenium (1.98 g, 25.1 mmol) was added, and the solution was stirred for 21 h. It was quenched with aqueous NH₄Cl, and air was bubbled through the solution for 30 min. The solution was extracted with ethyl acetate, and the organic extracts were combined and washed with water and brine, dried over anhydrous Na₂SO₄, and concentrated in vacuo. The crude mixture was purified by flash chromatography (dichloromethane/hexanes, 3:1 increasing to 5:1) to yield the product as a purple solid (742 mg, 20%) as well as a second fraction of slightly lower purity (421 mg, 11%). The first fraction gave mp 110–111 °C; ¹H NMR (400 MHz, CDCl₃): δ 7.51 (d, J = 8.9 Hz, 2H), 7.15 (d, J = 8.9 Hz, 2H), 5.29 (s, 4H), 3.53 (s, 6H); ¹³C{¹H} NMR (101 MHz, CDCl3): δ 150.8, 139.8, 129.1, 125.7, 125.1, 115.2, 95.0, 56.4; ⁷⁷Se{¹H} NMR (76 MHz, CDCl₃): δ 408.4. HRMS (EI -TOF): calcd for C₁₄H₁₄O₄⁸⁰Se₂, 405.9220; found, 405.9223. Anal. Calcd for C₁₄H₁₄O₄Se₂: C, 41.60; H, 3.49. Found: C, 41.76; H, 3.42.

1,8-Dibromo-2,7-di[(2-methoxyethoxy)methoxy]naphthalene (5f)

The same procedure as in the preparation of 5e was employed. Product 5f was obtained from compound 6 (302 mg, 0.950 mmol), sodium hydride (69 mg, 2.9 mmol), and 1-(chloromethoxy)-2-methoxyethane (217 μL, 237 mg, 1.90 mmol). The resulting oil was purified by flash chromatography to afford the product as a red oil (203 mg, 43%). This compound decomposed to an intractable black solid after a few days at room temperature. A fresh sample gave: ¹H NMR (400 MHz, CDCl₃): δ 7.72 (d, J = 9.0 Hz, 2H), 7.39 (d, J = 9.0 Hz, 2H), 5.45 (s, 4H), 3.95–3.93 (m, 4H), 3.59–3.57 (m, 4H), 3.37 (s, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 154.4, 131.6, 129.9, 128.7, 115.4, 107.9, 94.5, 71.5, 68.2, 59.0; HRMS (ESI) calcd for C₁₈H₂₂⁷⁹Br⁸¹BrO₆, 512.011 (M + NH₄); found, 512.0108 (M + NH₄).

3,8-Di[2-(methoxyethoxy)methoxy)]naphtho[1,8-cd]-1,2-diselenole (1f)

The same procedure as in the preparation of 1e was employed. Product 1f was obtained from compound 5f (1.07 g, 2.17 mmol), n-butyllithium (2.4 mL, 2.5 M in hexanes, 6.0 mmol), and elemental selenium (485 mg, 6.14 mmol). The crude product was purified by flash chromatography (hexanes/ethyl acetate, 1:1) to afford a purple oil which solidified upon standing to provide 1f (447 mg, 42%); mp 64–66 °C; ¹H NMR (400 MHz, CDCl₃): δ 7.53 (d, J = 8.8 Hz, 2H), 7.21 (d, J = 8.8 Hz, 2H), 5.40 (s, 4H), 3.90–3.87 (m, 4H), 3.58–3.56 (m, 4H), 3.38 (s, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 150.8, 139.8, 129.1, 125.7, 124.9, 115.2, 94.0, 71.6, 68.0, 59.0; ⁷⁷Se{¹H} NMR (76 MHz, CDCl₃): δ 407.9; HRMS (ESI): calcd for C₁₈H₂₂ O₆⁸⁰Se₂, 512.0085 (M + NH₄); found, 512.0068. Anal. Calcd for C₁₈H₂₂O₆Se₂: C, 43.92; H, 4.50. Found: C, 43.64; H, 4.37.

2,7-Di{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}naphthalene (4g)

2,7-Dihydroxynaphthalene (3) (247 mg, 1.54 mmol) and anhydrous K₂CO₃ (1.04 g, 7.54 mmol) were stirred in 30 mL of acetonitrile under nitrogen. 2-[2-(2-methoxyethoxy)ethoxy]ethyl p-toluenesulfonate (1.00 g, 3.14 mmol) was then added, and the reaction was refluxed under nitrogen. After 24 h, the opaque, green reaction mixture was cooled to room temperature, washed with saturated NaHCO₃, and extracted with ethyl acetate. The combined extracts were dried over anhydrous MgSO₄ and concentrated in vacuo. The crude product was purified by flash chromatography (ethyl acetate/hexanes, 6:1– to 9:1) to afford 573 mg (82%) of product 4g as a yellow oil; ¹H NMR (400 MHz, CDCl₃): δ 7.64 (d, J = 8.7 Hz, 2H), 7.03–7.02 (m, 2H), 7.00 (d, J = 2.4 Hz, 2H), 4.23 (t, J = 4.9 Hz, 4H), 3.92 (t, J = 4.9 Hz, 4H), 3.79–3.74 (m, 4H), 3.72–3.64 (m, 8H), 3.57–3.53 (dd, m, 4H), 3.37 (s, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 157.6, 136.0, 129.3, 124.7, 116.6, 106.6, 72.2, 71.1, 70.9, 70.8, 70.0, 67.7, 59.2; HRMS (ESI-TOF) [M + Na] calcd for C₂₄H₃₆O₈, 475.2302; found, 475.2280.

1,8-Dibromo-2,7-di{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}naphthalene (5g)

Compound 4g (2.27 g, 5.00 mmol) and pyridine (1.31 mL, 1.28 g, 16.2 mmol) were dissolved in ethyl acetate, followed by NBS (3.74 g, 21.0 mmol), and the solution was stirred at room temperature for 39 h. The mixture was washed with aqueous NaHCO₃ and brine, dried over anhydrous Na₂SO₄, and concentrated in vacuo. The crude product was purified by flash chromatography (ethyl acetate/methanol, 49:1) to afford 1.10 g (36%) of product 5g as an orange oil. It was rigorously dried under high vacuum and stored under argon. ¹H NMR (400 MHz, CDCl₃): δ 7.70 (d, J = 9.0 Hz, 2H), 7.17 (d, J = 9.0 Hz, 2H), 4.31 (t, J = 5.0 Hz, 4H), 3.95 (t, J = 4.9 Hz, 4H), 3.83–3.80 (m, 4H), 3.70–3.61 (m, 8H), 3.55–3.52 (m, 4H), 3.37 (s, 6H); ¹³C{¹H} NMR (100 MHz, CDCl₃): δ 156.3, 132.1, 130.1, 128.1, 114.1, 107.7, 72.2, 71.4, 71.0, 70.8, 70.5, 70.0, 59.2; HRMS (ESI-TOF) [M + Na]⁺ calcd for C₂₄H₃₄O₈⁷⁹Br₂, 631.0513; found, 631.0521.

3,8-Di{2-[2-(2-methoxyethoxy)ethoxy]ethoxy}naphtho[1,8-cd]-1,2diselenole (1g)

The same procedure as in the preparation of 1e was employed. Product 1g was obtained from compound 5g (205 mg, 0.336 mmol), n-butyllithium (0.38 mL, 2.5 M, 0.95 mmol), and elemental selenium (122 mg, 1.55 mmol). The crude product was purified by flash chromatography (ethyl acetate) to afford 96 mg (47%) of diselenide 1g as a purple oil; ¹H NMR (400 MHz, CDCl₃): δ 7.50 (d, J = 8.9 Hz, 2H), 6.97 (d, J = 8.9 Hz, 2H), 4.29 (crude t, J = 4.5 Hz 4H), 3.87 (crude t, J = 5.0 Hz,, 4H), 3.77–3.75 (m, 4H), 3.70–3.62 (m, 8H), 3.55–3.52 (m, 4H), 3.36 (s, 6H); ¹³C{¹H} NMR (101 MHz, CDCl₃): δ 152.5, 140.0, 128.4, 125.7, 124.3, 113.7, 72.1, 71.2, 70.9, 70.7, 70.0, 69.3, 59.2; ⁷⁷Se{¹H} NMR (CDCl₃, 76 MHz): δ 408.4; HRMS (ESI-TOF) [M + NH₄]⁺ calcd for C₂₄H₃₄O₈⁸⁰Se₂, 628.0922; found, 629.0900.

General Procedure for the Kinetic Assay of GPx Mimetics

Deuteriochloroform was treated with anhydrous K₂CO₃ to remove acidic impurities before use. DMT (30.1 mg, 0.155 mmol), benzyl thiol (36.3 μL, 38.4 mg, 0.310 mmol), and the catalyst (0.0310 mmol) were dissolved in 10 mL of CDCl₃-CD₃OD (95:5). Aqueous hydrogen peroxide (50%, 20.0 μL, 0.350 mmol) was added, and the temperature was maintained at 18 °C with vigorous stirring. The reaction was monitored by ¹H NMR spectroscopy, and the amount of dibenzyl disulfide was determined from the integrated ratio of its methylene signal with that of the aromatic signal of DMT. The t_1/2 for each catalyst was defined as the time required for the formation of 50% of the expected amount of the disulfide.

EPR Spectrum from the Assay of Diselenide 1b

Deuteriochloroform was treated with anhydrous K₂CO₃ to remove acidic impurities before use. DMT (15.0 mg, 0.0772 mmol, internal standard for ¹H NMR spectra), benzyl thiol (18.1 μL, 0.154 mmol), and diselenide 1b (5.3 mg, 0.015 mmol) were dissolved in 5 mL of CDCl₃-CD₃OD (95:5). Aqueous hydrogen peroxide (50%, 10.0 μL, 0.18 mmol) was added, and the temperature was maintained at 18 °C. ¹H NMR spectra were recorded to ensure that the assay was proceeding in the usual manner with evidence of peak broadening. An aliquot was removed after 2 h, and the EPR spectrum in Figure 6 was obtained at 160 K using an X-band EPR spectrometer operating at 9.6 GHz.³²

Reaction of Equimolar Amounts of 1b and 8b with Benzyl Thiol and Triethylamine

Diselenide 1b (35.5 mg, 0.103 mmol) was dissolved in 1 mL of CDCl₃-CD₃OD (95:5), and aqueous hydrogen peroxide (50%, 3.1 μL, 0.054 mmol) was added. The reaction was stirred at room temperature for 2 h, at which time a ¹H NMR spectrum indicated the presence of equimolar amounts of 1b and selenolseleninate 8b (see Figure 7a and Supporting Information 15). Benzyl thiol (1.0 μL, 0.01 mmol) was added, the NMR tube was shaken vigorously, and the NMR spectrum revealed extreme peak broadening of signals from the diselenide (see Figure 7b and Supporting Information 15). While these signals seemed to disappear into the baseline, their presence was detected by integration of the spectrum. Triethylamine (5.0 μL, 0.036 mmol) was then added, the NMR tube was shaken vigorously, and the resulting NMR spectrum showed restoration of the diselenide peaks (see Figure 7c and Supporting Information 16).

PCET Reaction of Diselenide 1c with TFA and Excess Benzyl Thiol

3,8-Di(n-pentyloxy)naphtho[1,8-cd]-1,2-diselenole (1c) (15.5 mg, 0.034 mmol) was dissolved in 1.00 mL of CDCl₃, and the ¹H NMR spectrum was recorded. TFA (1.3 μL, 1.9 mg, 0.017 mmol) was added, and the spectrum was again recorded. Benzyl thiol was added in portions (1st portion: 2.0 μL, 2.1 mg, 0.017 mmol; 2nd portion: 2.0 μL, 2.1 mg, 0.017 mmol; and 3rd portion: 36 μL, 38 mg, 0.306 mmol), and the spectrum was recorded after each addition (see Figure 8).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.2c02286.

• ¹H, ¹³C, and ⁷⁷Se NMR spectra of compounds 1, 4, 5, and 6, as well as the full spectra from Figure 7 (PDF)

The authors declare no competing financial interest.