Alkoxy Substituted Brominated closo-Dodecaborates with Functionalized Aliphatic Spacers

Abstract

The utilization of dodecaborate boron clusters, [B₁₂X₁₂]²⁻ (X = Cl, Br, or I), as membrane carriers has been demonstrated recently, and their activity is known to be due to their superchaotropic nature. In this work, the mono-alkylation of [B₁₂Br₁₁OH]²⁻ to functionalize it with an aliphatic spacer was developed with a view to expanding the known chemical space of membrane carriers based on [B₁₂Br₁₂]²⁻. A new and improved facile route for the preparation of [B₁₂Br₁₁OH]²⁻, which is an important precursor to other [B₁₂Br₁₁OR]²⁻ species, is reported. Various alkoxylated [B₁₂Br₁₁O(CH₂)₅Z]²⁻ (Z = OH, N(CO)₂C₆H₄, CN and N₃) derivatives were prepared via a divergent synthesis based on [B₁₂Br₁₁O(CH₂)₅Br]²⁻. One of the newly synthesized compounds was utilized as a membrane carrier, and its impact on cell viability was examined.

1. Introduction

Halogenated dodecaborates [B₁₂X₁₂]²⁻ (X = F, Cl, Br, or I) have received much attention in the fields of materials science and medicinal chemistry due to their rigid globular icosahedral structure, delocalized charge and high chemical stability [1,2,3,4]. Recently, the discovery of the superchaotropic ionic character of halogenated dodecaborates has led to the development of new chaotropic membrane carriers that utilize these boron clusters for transmembrane transport [5,6,7,8,9,10,11,12,13]. The term “superchaotropic” refers to ions that strongly disrupt the structure of water, facilitating the transport of hydrophilic cargos across lipid membranes. Among the perhalogenated dodecaborates, [B₁₂Br₁₂]²⁻ has emerged as a particularly promising chaotropic membrane carrier due to its exceptional properties. It exhibits broad delivery activity with low levels of membrane lysis, does not tend to aggregate—a common issue with classic amphiphilic carriers—and the transported molecular cargo is not trapped in endosomes but is evenly distributed in the cytoplasm. These characteristics make [B₁₂Br₁₂]²⁻ an effective chaotropic membrane carrier for delivering various bioactive molecules into cells. However, despite these advantages, there are still limitations in the range of transmembrane cargo that can be transported by chaotropic membrane carriers. Specifically, effective transport has been demonstrated primarily for positively charged cargos up to c.a. 12 kDa, and there have been no reports of the successful transport of negatively charged cargos. This limitation hinders the broader application of [B₁₂Br₁₂]²⁻ as a universal membrane carrier.

To overcome these limitations and enhance the utility of chaotropic membrane transporters based on anionic boron clusters, it is crucial to develop versatile methods for the functionalization of [B₁₂Br₁₂]²⁻. One promising strategy involves attaching aliphatic spacers or linkers to the boron cluster. Although introducing an aliphatic spacer might reduce the chaotropic properties due to decreased symmetry and charge distribution, incorporating such spacers can contribute to the development of more effective chaotropic membrane transporters. Specifically, the aliphatic spacer may beneficially influence the interactions between the boron cluster and the lipid bilayer during membrane translocation [11]. Moreover, designing molecules with spacers allows for the linkage and incorporation of multiple boron clusters into a single molecule, potentially achieving a cooperative effect that enhances the overall chaotropic activity. Functionalizing the spacer also offers a versatile platform to fine-tune interactions with various cargo molecules, including the possibility of covalent attachment. Despite the potential benefits of functionalizing [B₁₂X₁₂]²⁻ with aliphatic spacers, structural modifications of these boron clusters have generally been limited to simple alkoxylated ([B₁₂X₁₁OR]²⁻/aminated ([B₁₂X₁₁NHR]²⁻) derivatives [14,15,16]. Therefore, developing versatile methods for introducing a broader range of functional groups is essential to expand the range of cargos that can be transported, including negatively charged species.

In this work, we report a convenient method for the preparation of a brominated and hydroxylated dodecaborate, [B₁₂Br₁₁OH]²⁻, starting from cyclic oxonium derivatives of [B₁₂H₁₂]²⁻. Likewise, we show the divergent synthesis of various alkoxylated [B₁₂Br₁₁O(CH₂)₅Z]²⁻ (Z = OH, N(CO)₂C₆H₄, CN and N₃) derivatives that have functionalized aliphatic spacers. Furthermore, membrane transport experiments in living cells and a cell viability assay were conducted on one of the newly synthesized compounds [17].

2. Materials and Methods

2.1. Instrumentation

NMR spectra were obtained on a BRUKER AVANCE NEC400 OneBay (Bruker, Billerica, MA, USA). Chemical shifts are expressed in δ (ppm) values. ¹H NMR spectra were referenced to the residual solvent as an internal standard. ¹¹B NMR spectra were unreferenced. ¹³C NMR spectra were referenced to the residual solvent as an internal standard. The following abbreviations are used: s = singlet, d = doublet, t = triplet, m = multiplet, and bs = broad singlet. IR spectra were obtained on a Shimadzu Corporation IR Prestige-21 spectrometer (Shimadzu, Kyoto, Japan). ESI mass spectra were measured on a Bruker micrOTOF-II-SF spectrometer (Bruker, Billerica, MA, USA). Melting points were determined with a Yanaco micro melting point (Yanaco Analytical Systems Inc., Tokyo, Japan). Normal-phase column chromatography was performed with Biotage Isolera One 1SW (Biotage AB, Uppsala, Sweden).

2.2. Materials

Unless otherwise noted, materials were purchased from Aldrich Inc. (St. Louis, MO, USA), Wako Pure Chemical Industries Ltd. (Osaka, Japan), Tokyo Kasei Co. (Tokyo, Japan) and other commercial suppliers and were used after appropriate purification. [Cs⁺]₂[B₁₂H₁₂²⁻] (Cs•1) was purchased from KATCHEM spol (Prague, Czech Republic). Air- and moisture-sensitive manipulations were performed with standard Schlenk techniques. Normal-phase column chromatography was performed with silica gel 60 (230–400 mesh) from Merck (Merck KGaA, Darmstadt, Germany).

2.3. Experimental Procedures

2.3.1. Procedure for One-Pot Synthesis of [NBu₄⁺]₂[B₁₂Br₁₁OH²⁻]

(NBu₄)₂•1 (712 mg, 1.14 mmol), NaBF₄ (616 mg, 5.61 mmol), THF (78 mL), and HCl (prepared by mixing 1 vol of 37% HCl in water with 3 vol of THF) (2.2 mL) were placed in a Schlenk tube and stirred at 90 °C for 3 h. The generation of the intermediate (2) was checked by ¹H NMR and ¹¹B NMR analysis (Figures S1 and S2). The solvent was removed under vacuum. Bromine (5.1 mL, 0.099 mol) and 25 mL of 50% aqueous methanol solution were added, and the solution was stirred at 70 °C for 2 days. The solvent was removed under vacuum, the resulting material was again dissolved in water, and the precipitate was filtered off, washed with water, dried in air to give pure (NBu₄)₂•3 (powder A). The filtrate was added to the tetrabutylammonium bromide (1.61 g, 4.99 mol), and the precipitate was filtered off, washed with water, dried in air to give pure (NBu₄)₂•3 (powder B). The powders A and B were combined (1.41 g, 82% yield based on Cs•1).

Compound: [NBu₄⁺]₂[B₁₂Br₁₁OH²⁻] ((NBu₄)₂•3)

White powder; mp > 300 °C; ¹H-NMR (400 MHz, methanol-d₄) δ 4.60 (s, 1H), 3.26 (t, J = 8.6 Hz, 16H), 1.65–1.73 (m, 16H), 1.45 (m, 16H), 1.05 (t, J = 7.4 Hz, 24H); ¹¹B-NMR (128 MHz, methanol-d₄) δ −4.01 (s, 1B), −13.92 (s, 10B), −16.44 (s, 1B); ¹³C-NMR (101 MHz, acetone-d₆) δ 58.56, 23.66, 19.56, 13.12; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁OH 513.1070, found 513.1073; Anal. Calcd for C₃₂H₇₃B₁₂Br₁₁N₂O: C, 25.44; H, 4.87; N, 1.85. Found: C, 25.11; H, 4.31; N, 1.99.

2.3.2. Procedure for the Alkoxylation of [B₁₂Br₁₁OH²⁻]

(NBu₄)₂•3 (98.8 mg, 0.065 mmol), KOH (26.0 mg, 0.46 mol) 1,5-dibromopentane (0.2 mL, 1.53 mol) and DMSO (1.5 mL) were placed in a Schlenk tube and stirred at room temperature for 16 h. The solvent was removed under vacuum, and the solid was washed with water and hexane, dried in air to give pure (NBu₄)₂•4 (78 mg, 72% yield).

Compound: [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₄Br²⁻] ((NBu₄)₂•4)

White powder; mp > 300 °C; ¹H-NMR (400 MHz, acetone-d₆) δ 4.17 (t, J = 5.9 Hz, 2H), 3.53 (t, J = 7.0 Hz, 2H), 3.46 (t, J = 8.5 Hz, 16H), 1.92–1.99 (m, 2H), 1.81–1.89 (m, 16H), 1.51–1.61 (m, 2H), 1.47 (m, 16H), 1.00 (t, J = 7.4 Hz, 24H); ¹¹B-NMR (128 MHz, acetone-d₆) δ −4.51 (s, 1B), −14.03 (s, 10B), −16.55 (s, 1B); ¹³C-NMR (101 MHz, acetone-d₆) δ 64.66, 58.57, 34.78, 32.59, 31.03, 24.52, 23.62, 19.54, 13.06; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅Br 587.6010, found 587.5917; Anal. Calcd for C₃₇H₈₂B₁₂Br₁₂N₂O: C, 26.78; H, 4.98; N, 1.69. Found: C, 26.43; H, 5.12; N, 1.88.

2.3.3. General Procedure for the Synthesis of [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₅X²⁻] (X = CN, N₃, NC₆H₄(CO)₂)

(NBu₄)₂•4 (23.4 mg, 0.014 mmol), KCN (10.8 mg, 0.17 mol) and DMF (0.5 mL) were placed in a Schlenk tube and stirred at 55 °C overnight. The solvent was removed under vacuum, and the solid was washed with water, dried in air to give pure (NBu₄)₂•5 (10.8 mg, 60% yield).

Compound: [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₄CN²⁻] ((NBu₄)₂•5)

White powder; mp > 300 °C; ATR-FTIR (neat) ν 416, 448, 482, 500, 532, 555, 579, 603, 665, 737, 799, 888, 921, 981, 997, 1035, 1065, 1184, 1218, 1306, 1349, 1380, 1420, 1468, 1646, 1713, 1930, 1940, 1959, 2040, 2074, 2105, 2146, 2169, 2188, 2215, 2245, 2295, 2342, 2369, 2423, 2464, 2566, 2873, 2931, 2961, 3123, 3280, 3352, 3447, 3513, 3576, 3618, 3688, 3798, 3819, 3849, 3966 cm⁻¹; ¹H-NMR (400 MHz, acetone-d₆) δ 4.18 (t, J = 6.0 Hz, 2H), 3.46 (t, J = 8.6 Hz, 16H), 2.47 (t, J = 7.3 Hz, 2H), 1.81–1.89 (m, 16H), 1.75 (m, 4H) 1.60 (m, 2H), 1.47 (m, 16H), 1.00 (t, J = 7.4 Hz, 24H); ¹¹B NMR (128 MHz, acetone-d₆) δ −4.53 (s, 1B), −14.01 (s, 10B), −16.37 (s, 1B); ¹³C-NMR (101 MHz, acetonitrile-d₃) δ 65.06, 58.39, 31.22, 25.06, 23.34, 19.36, 16.53, 12.84; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅CN 560.6439, found 560.6531; Anal. Calcd for C₃₇H₈₂B₁₂Br₁₂N₂O: C, 26.78; H, 4.98; N, 1.69. Found: C, 26.43; H, 5.12; N, 1.88.

Compound: [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₄N₃²⁻] ((NBu₄)₂•6)

According to the general procedure using NaN₃ as a nucleophile, (NBu₄)₂•6 was obtained in 49% isolated yield (11.4 mg). White powder; mp > 300 °C; ATR-FTIR (neat) ν 426, 449, 501, 531, 556, 606, 674, 737, 800, 882, 924, 981, 997, 1036, 1067, 1173, 1202, 1219, 1263, 1297, 1349, 1381, 1469, 1539, 1647, 1712, 2021, 2037, 2093, 2144, 2169, 2185, 2342, 2874, 2931, 2962, 3175, 3351, 3513, 3591, 3619, 3674, 3800 cm⁻¹; ¹H-NMR (400 MHz, acetone-d₆) δ 4.17 (t, J = 5.8 Hz, 2H), 3.44–3.49 (t, J = 5.6 Hz,16H), 3.33 (t, J = 7.1 Hz, 2H), 1.81–1.89 (m, 16H), 1.68 (m, 4H), 1.53 (m, 2H), 1.42–1.50 (m, 16H), 1.00 (t, J = 7.4 Hz, 24H); ¹¹B NMR (128 MHz, acetone-d₆) δ −4.57 (s, 1B), −14.02 (s, 10B), −16.37 (s, 1B); ¹³C-NMR (101 MHz, acetonitrile-d₃) δ 65.16, 58.33, 51.35, 31.76, 28.31, 23.34, 22.82, 19.36, 12.83; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅N₃ 568.6469, found 568.6490; Anal. Calcd for C₃₆H₈₀B₁₂Br₁₁N₅O: C, 26.89; H, 5.02; N, 4.36. Found: C, 26.59; H, 5.33; N, 3.32.

Compound: [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₄NC₆H₄(CO)₂²⁻] ((NBu₄)₂•7)

According to the general procedure using KNC₆H₄(CO₂) as a nucleophile, (NBu₄)₂•7 was obtained in 72% isolated yield (32.2 mg). White powder; mp 268 °C; ¹H-NMR (400 MHz, acetone-d₆) δ 7.84 (m, 4H), 4.16 (t, J = 6.1 Hz, 2H), 3.65–3.69 (t, J = 8.4 Hz, 2H), 3.45 (t, J = 8.4 Hz, 16H), 1.80–1.88 (m, 16H), 1.69–1.77 (m, 4H), 1.57 (m, 2H), 1.47 (m, 16H), 1.00 (t, J = 7.3 Hz, 24H); ¹¹B-NMR (128 MHz, acetone-d₆) δ −4.47 (s, 1B), −13.99 (s, 10B), −16.45 (s, 1B); ¹³C NMR (101 MHz, acetone-d₆) δ 167.86, 133.99, 132.32, 122.72, 64.86, 58.58, 37.86, 32.12, 23.64, 23.23, 19.55, 13.11; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅NC₆H₄(CO)₂ 620.6546, found 620.6550; Anal. Calcd for C₄₅H₈₆B₁₂Br₁₁N₃O₃: C, 31.32; H, 7.52; N, 2.43. Found: C, 30.83; H, 6.75; N, 2.52.

2.3.4. Procedure for the Hydroxylation of Terminal Bromine of (NBu₄)₂•4

(NBu₄)₂•4 (24.2 mg, 0.015 mmol) and H₂O/DMSO (7:3, 4 mL) was placed in a Schlenk tube (Yazawa Scientific Co., Tokyo, Japan) and stirred at 120 °C overnight. The solvent was removed under vacuum, and the solid was washed with water and hexane, dried in air to give pure (NBu₄)₂•8 (14.6 mg, 63% yield).

Compound: [NBu₄⁺]₂[B₁₂Br₁₁O(CH₂)₄OH²⁻] ((NBu₄)₂•8)

White powder; mp > 300 °C; ATR-FTIR (neat) ν 408, 449, 489, 504, 596, 739, 798, 838, 881, 984, 1061, 1173, 1208, 1381, 1469, 1719, 1976, 2140, 2202, 2267, 2873, 2932, 2960, 3037, 3210, 3263, 3362, 3458, 3524, 3763, 3928 cm⁻¹; ¹H NMR (400 MHz, acetone-d₆) δ 4.15 (t, J = 6.3 Hz, 2H), 3.53 (t, J = 6.6 Hz, 2H), 3.46 (t, J = 8.6 Hz, 16H), 1.84 (m, 16H), 1.53–1.42 (m, 22H), 1.00 (t, J = 7.4 Hz, 24H); ¹¹B-NMR (128 MHz, acetone-d₆) δ −4.46 (s, 1B), −13.99 (s, 10B), −16.40 (s, 1B); ¹³C NMR (101 MHz, acetone-d₆) δ 65.12, 61.87, 58.55, 32.94, 32.38, 23.59, 22.07, 19.53, 13.03; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅OH 556.1529, found 556.1486; Anal. Calcd for C₃₇H₈₃B₁₂Br₁₁N₂O₂: C, 27.83; H, 5.24; N, 1.75. Found: C, 28.42; H, 5.63; N, 2.05.

2.3.5. Procedure for the Cation Exchange from NBu₄⁺ to K⁺

(NBu₄)₂•8 was charged in a Schlenk flask and dried under reduced pressure at 120 °C for 1 h. The flask was charged with argon, and 2.0 mL of anhydrous THF was added at room temperature. The solution was cooled to 0 °C, and n-BuLi (2.69 M, 0.525 mmol) was added. The mixture was stirred at 0 °C for 30 min, warmed to room temperature and stirred for 1 h. The reaction mixture was added to distilled water.

The counter cation exchange to potassium cation (K⁺) was performed as follows: the AcOEt solution (ca. 10 mL) of 8 was washed with 20% aqueous KCl solution (3 × 5 mL), and, then, the combined water layers were extracted with AcOEt (3 × 10 mL). The combined organic phases were evaporated, and the solvent was removed under vacuum. The solid was added to acetone, filtered off to remove KCl, washed with acetone, and the filtered solution was evaporated under vacuum and dried in air to give pure K₂•8 (19.4 mg, 12% yield based on the starting material).

Compound: [K⁺]₂[B₁₂Br₁₁O(CH₂)₄OH²⁻] ((K)₂•8)

White powder; mp > 300 °C; ATR-FTIR (neat) ν 449, 535, 585, 728, 982, 1092, 1175, 1204, 1228, 1353, 1364, 1414, 1561, 1708, 1884, 2020, 2120, 2147, 2164, 2272, 2410, 2478, 2566, 2682, 2763, 2800, 2853, 2886, 2921, 2941, 3252, 3366, 3452, 3502, 3570, 3618, 3651, 3855, 3936, 3971 cm^{−1 1}H NMR (400 MHz, acetone-d₆) δ 4.15 (t, J = 6.5 Hz, 2H), 3.54 (t, J = 6.6 Hz, 2H), 1.50–1.57 (m, 4H), 1.41–1.47 (m, 2H); ¹¹B NMR (128 MHz, acetone-d₆) δ −4.40 (s, 1B), −14.00 (s, 10B), −16.31 (s, 1B); ¹³C NMR (101 MHz, acetone-d₆) δ 65.24, 61.92, 32.89, 32.35, 22.02; HR-ESI-TOF-MS m/z calcd for B₁₂Br₁₁O(CH₂)₅OH 556.1529, found 556.1492; Anal. Calcd for C₅H₁₁B₁₂Br₁₁KO₂: C, 5.22; H, 0.96; N, 3.40. Found: C, 5.34; H, 1.14; N, 3.72.

2.4. Materials and Methods for Cell Viability Assay

2.4.1. Cell Culture and Imaging

HeLa cells were seeded on 48-well plates (Hounisen, Nagoya, Japan, 83.3923) and cultured in DMEM (Thermo Fisher Scientific, Waltham, MA, USA, 10566016) with 15% Fetal Bovine Serum (FBS) and 1000 U/mL of Penicillin–Streptomycin (P/S) (Thermo Fisher Scientific, 15140122) until they became approximately 60% confluent. The boron clusters and 3.6 μM of DAPI (Thermofisher, 62248) were diluted in an HKR buffer (5 mM of HEPES pH 7.5, 137 mM of NaCl, 2.68 mM of KCl, 2.05 mM of MgCl₂, 1.8 mM of CaCl₂, sterilized by filter) and incubated for 15 min at room temperature. After HeLa cells were rinsed with the HKR buffer, the boron cluster and DAPI mixture were added, and cells were incubated for 1.5 h at 37 °C, 5% CO₂. We carried out imaging using Zeiss Axio Observer Z1 Apotome 2 (Carl Zeiss AG, Oberkochen, Germany, inverted setup), and images were analyzed by ImageJ [18].

2.4.2. Cell Viability Assay

HeLa cells were seeded on 48-well plates and incubated in DMEM with 15% FBS and P/S until they became approximately 60% confluent. The boron clusters were diluted in the HKR buffer and added to the HeLa cells. After 1.5 h of incubation, HeLa cells were cultured in DMEM with 15% FBS and P/S for 24 h. To dissociate HeLa cells, cells were rinsed with PBS and incubated with 0.05% Trypsin EDTA (Thermo Fisher Scientific, 25300054) at 37 °C for 5 min. Subsequently, trypsin was neutralized by adding an equal volume of DMEM with 15% FBS, and cells were completely dissociated by pipetting. After adding an equal volume of trypan blue (Thermo Fisher Scientific, 15250-061), live and dead cells were counted by a hemocytometer. We confirmed that the total cell number did not show a statistically significant difference among groups and quantified the percentage of live cells over the total cell number.

3. Results

3.1. Preparation of [B₁₂Br₁₁OH]²⁻

Our study commenced with an improved synthesis of [B₁₂Br₁₁OH]²⁻ (3). A previously reported preparation of [B₁₂Br₁₁OH]²⁻ utilizes two steps: (1) the mono-hydroxylation of [B₁₂H₁₂]²⁻ (1) to give [B₁₂H₁₁OH]²⁻ and (2) the subsequent bromination of this species. Hawthorne reported that the elegant sequential acid-catalyzed hydroxylation of 1 in aqueous sulfuric acid results in the formation of 3 [19]. However, we found that the preparation of 3 was only reproducible on a small scale and that a prolonged reaction time was needed at larger scales. The formation of an over-hydroxylated product, [B₁₂H₁₀(OH)₂]²⁻, was also inevitable, and the separation of the mono- and di- hydroxylated dodecaborate was not practical, either before or after the bromination. To produce highly pure [B₁₂Br₁₁OH]²⁻ without any over-hydroxylated products, we supposed that an oxonium derivative of 1 could serve as an alternative precursor for 3. A dodecaborate containing a cyclic oxonium group, [B₁₂H₁₁O(CH₂)₄]²⁻ 2, can be obtained on a gram scale in a single step in a moderate yield without the formation of over-hydroxylated byproducts [20]. We supposed that the acid-mediated rupture of the C–O bond of 2 during the bromination step would generate highly pure 3. Thus, 2 was prepared as tetrabutylammonium salts, according to a reported method (Figure 1a, step 1), as confirmed using NMR spectroscopy (Figures S1 and S2) and was directly brominated in aqueous methanol without purification (Figure 1b, step 2) [21]. The preparation of (Cs)₂•3 starting from (Cs)₂•1 failed probably due to the poor solubility of (Cs)₂•1. Previous reports adopted tetrabutylammonium salts. As for the mechanism of the transformation from 2 to 3, we propose two possibilities: (1) cleavage of the C–O bond in the oxonium ring, and (2) cleavage of the B–O bond, generating a borenium ylide with a naked boron vertex that reacts with water to yield the hydroxylated dodecaborate cluster. Although B–O bond cleavages are rare, similar examples have been reported [22]. We conducted an ESI-TOF-MS analysis of the reaction mixture (step 2) after 5 min at room temperature (Figure S3). The spectra showed ring-opened perbrominated products in addition to [B₁₂Br₁₁OH]²⁻ and [B₁₂HBr₁₀OH]²⁻, indicating that at least the C–O bond cleavage had occurred. Since the reaction proceeds rapidly at room temperature, we were unable to determine whether bromination or C–O bond cleavage occurs first by ESI-TOF-MS analysis. This two-stage one-pot reaction could be conducted on a gram scale, and simple purification (removal of volatiles/filtration/precipitation in H₂O as the ammonium salt) was sufficient to obtain (NBu₄)₂•3 in high purity (82%). The obtained product was characterized using ¹¹B NMR and ¹H NMR spectroscopy and ESI-TOF mass spectrometry. The obtained spectra completely matched reported results without any traces of [B₁₂Br₁₀(OH)₂]²⁻ detected (Figure 1b,c). Although ESI-TOF mass spectrometry showed peaks corresponding to [B₁₂Br₁₂]²⁻, no signals for [B₁₂Br₁₂]²⁻ were observed in ¹¹B NMR spectroscopy, indicating that contamination by [B₁₂Br₁₂]²⁻ is negligible. We expect this method to be applicable for the synthesis of other hydroxylated halogenated dodecaborates [B12X11OH]²⁻ (X = F, Cl, I) [23].

3.2. Preparation of [B₁₂Br₁₁OH]²⁻

Next, we examined the synthesis of [B₁₂Br₁₁O(CH₂)₅Br]²⁻ 4, a precursor of various [B₁₂Br₁₁OR]²⁻ species with functionalized aliphatic spacers (Scheme 1). The reaction of 3 with 1,5-dibromopentane in DMSO was found to proceed smoothly and give 4 in a moderate yield [14]. A twenty-fourfold excess of 1,5-dibromopentane was used to suppress the undesirable formation of [B₁₂Br₁₁O]²⁻–(CH₂)₂–[B₁₂Br₁₁O]²⁻.

Subsequently, various alkoxylated [B₁₂Br₁₁O(CH₂)₅Z]²⁻ (Z = OH, N(CO)₂C₆H₄, CN and N₃) derivatives were prepared from 4 (Scheme 2). Precursor 4 reacted smoothly with either potassium cyanide or sodium azide in DMF to give the corresponding products [B₁₂Br₁₁O(CH₂)₅CN]²⁻ 5 and [B₁₂Br₁₁O(CH₂)₅N₃]²⁻ 6 in 60% and 49% yields, respectively. The IR spectra of 5 and 6 demonstrate the presence of CN (5: 2245 cm⁻¹) and N₃ (6: 2093 cm⁻¹) absorption bands, respectively. One of the most sought-after molecular designs for superchaotropic membrane transport carriers is a structure possessing multiple chaotropic moieties. A useful building block would be 5 for constructing a molecule that is rich in [B₁₂Br₁₁O–]²⁻ species by utilizing click chemistry [24].

The treatment of 4 with potassium phthalimide in DMF results in the formation of 7 in a 72% yield. The ¹H NMR spectra of 4 and 7 show that the low field Br–CH₂– signals disappeared and that new peaks corresponding to a –CH₂– group adjacent to N appeared. It is noteworthy that no reaction between 3 and the N-(2-bromopentyl)phthalimide moiety took place under the same or representative conditions (dimethoxyethane/NaH/reflux, CH₃CN/K₂CO₃/reflux), justifying our approach of utilizing 4 as a precursor [25].

The transformation from a terminal bromine group to a hydroxyl group was also examined. Based on a previously reported method, stirring (NBu₄)₂•4 in DMSO/H₂O at 80 °C resulted in the formation of (NBu₄)₂•8 in a good yield [26]. In the IR spectrum of 8, absorption bands for OH groups were observed between 3200 and 3100 cm⁻¹, indicating that the introduction of the hydroxyl group had been successful. The tetrabutylammonium salt, (NBu₄)₂•8, was converted to the corresponding potassium salt, K₂[B₁₂Br₁₁O(CH₂)₅OH]²⁻, by treating (NBu₄)₂•8 with an excess amount of n-BuLi in THF followed by liquid–liquid extraction using aqueous KCl. Attempts to decompose the tetrabutylammonium cation of (NBu₄)₂•8 using other bases (NaH/THF, KOH/DMSO, NaOH/water) were unsuccessful. The primary reason of the low yield of (K)₂•8 may be the difficulty in completely decomposing the tetrabutylammonium cation. The potassium salt, K₂•8, was found to be highly soluble in water (>2 mM), despite possessing a hydrophobic aliphatic chain. In general, all the tetrabutylammonium salts (4–8) can be purified using a simple procedure of washing with H₂O and hexane.

All the [B₁₂Br₁₁OR]²⁻ species with functionalized aliphatic spacers show a 1:10:1 pattern in their ¹¹B NMR spectra, and the chemical shift is independent of the terminal substituent (ESI Figure S11). Likewise, in the ¹H NMR spectra, the terminal substituent does not change the chemical shift in the signal for the –CH₂–O–B₁₂ peak. These results indicate that the inducive effect of the terminal functional groups of the [B₁₂Br₁₁O(CH₂)₅Z]²⁻ species (5–8) is not passed to the B₁₂ core and that the superchaotropic properties of B₁₂Br₁₁O–R would be not affected.

3.3. Cell Viability Assay and Membrane Translocation in Living Cells

Finally, the high water solubility of K₂•8 encouraged us to investigate its cytotoxicity and membrane transport activity in HeLa cells, which have been used in the proceeding studies [6,7,8,9,10,12]. To validate the delivery of a molecular cargo in the HeLa cells, we followed a previous report and utilized a membrane-impermeable dicationic fluorescent DNA-staining dye, 4′,6-diamidino-2-phenylindole (DAPI) [9]. Similar to other reported results, [B₁₂Br₁₂]²⁻ showed no cytotoxicity up to 100 μM and facilitated the intracell entry of DAPI and nuclear staining (Figure 2). Gratifyingly, 8 caused no significant cytotoxicity up to 100 μM and facilitated the uptake of DAPI, albeit with a decreased uptake efficiency, into the HeLa cells. One possible explanation for the lower uptake efficiency of 8 compared with [B₁₂Br₁₂]²⁻ is that the interaction between 8 and DAPI might be weaker than that of DAPI and [B₁₂Br₁₂]²⁻. A previous report suggests that the interaction between the cargo and the boron cluster drives desolvation of the cargo and facilitates direct membrane and cargo translocation. However, the hydrophobic spacer could physically and chemically hinder access from the boron cluster core of 8 to the hydrophobic cargo, DAPI. For comparison, [B₁₂Br₁₁OH]²⁻ (3) was also tested and exhibited no significant cytotoxicity up to 100 μM, with an uptake efficiency comparable to that of 8 (Figure S13). This suggests that the substitution of Br with the more hydrophilic hydroxyl group reduces the chaotropic properties compared to [B₁₂Br₁₂]²⁻. Considering that 8 is a prototype of a [B₁₂Br₁₁OR]²⁻-type membrane carrier, we expect that rational design of the spacer would afford a more effective superchaotropic carrier.

4. Conclusions

A two-stage one-pot facile preparation of high purity [B₁₂Br₁₁OH]²⁻ that avoids the generation of over hydroxylated byproducts was achieved by applying [B₁₂Br₁₁O(CH₂)₄]²⁻ as an intermediate. Various monoalkoxylated [B₁₂Br₁₁O(CH₂)₅Z]²⁻ species were prepared using [B₁₂Br₁₁O(CH₂)₅Br]²⁻ as a precursor. One of the newly synthesized compounds acts as a low toxicity membrane transporter in living cells, demonstrating the potential value of surface-modified halogenated dodecaborates for generating superchaotropic membrane carriers. We believe that this work is beneficial for expanding the known chemical space of membrane carriers, specifically carriers based on zwitterionic species and dendrimers of high boron content clusters.