Structure–activity relationships study of N-ethylene glycol-comprising alkyl heterocyclic carboxamides against A549 lung cancer cells

Abstract

Aim: Certain cancer cells depend on oxidative phosphorylation for survival; thus, inhibiting this process may be a promising treatment strategy. This study explored the structure–activity relationships of the mitochondrial inhibitor N-ethylene glycol-comprising alkyl thiophene-3-carboxamide 3.

Methods & results: We synthesized and evaluated 13 analogs (5a–m) with different ethylene glycol units, heterocycles and connecting groups for their growth-inhibitory effects on A549 non-small cell lung cancer cells. We found that increasing the number of ethylene glycol units significantly enhanced inhibitory activity. Some analogs activated adenosine monophosphate-activated protein kinase, similar to 3. Notably, analog 5e, which contains tetraethylene glycol units, significantly inhibited tumor growth in vivo.

Conclusion: Analog 5 may be a potential therapeutic agent for non-small cell lung cancer treatment.

Graphical Abstract

Plain language summary

Article highlights.

• This study explored the structure–activity relationships of the mitochondrial inhibitor N-ethylene glycol-comprising alkyl thiophene-3-carboxamide.

• We synthesized and evaluated 13 analogs with varying ethylene glycol units, heterocycles and connecting groups for their growth–inhibitory activity against A549 non-small cell lung cancer cells.

• Our findings indicate that increasing the number of ethylene glycol units significantly enhanced inhibitory activity.

• Some analogs induced phosphorylated AMP-activated protein kinase.

• Analog 5e, with tetraethylene glycol units, significantly inhibited tumor growth in vivo.

• These results suggest that analog 5 is a potential agent for non-small cell lung cancer treatment.

1. Introduction

Warburg’s 1927 findings uncovered that cancer cells augment energy production through glycolysis, even in aerobic conditions [1]. Recent studies suggest that some cancer cells depend on oxidative phosphorylation (OXPHOS) for energy production [2], and increased OXPHOS activity contributes to cancer cell resistance to radiotherapy [3] or chemotherapy [4]. Consequently, inhibiting OXPHOS has emerged as a promising strategy for cancer treatment [5–7]. Notably, metformin, an OXPHOS inhibitor that blocks mitochondrial complex I, has exhibited antitumor activity [8]. Clinical trials are currently underway to investigate the combination of metformin and temozolomide, a standard chemotherapeutic agent, for treating glioblastoma [9].

Lung cancer was ranked as the most frequently diagnosed cancer worldwide in 2022, with 2.5 million new cases, comprising 12.4% of the total [10]. The arsenal of treatment options include surgery, chemotherapy, radiation, immunotherapy or their combinations. Despite these interventions, lung cancer persists as the foremost cause of cancer-related mortality, accounting for 1.8 million deaths, or 18.7% of the total. The disease is classified into two types: small cell lung cancer and non-small cell lung cancer (NSCLC), with NSCLC accounting for the majority of cases [11]. A predominant causative factor in NSCLC is mutations in the EGFR gene, with a higher prevalence in Asian patients (30–50%) compared with other ethnicities [12]. Tyrosine kinase inhibitors serve as first-line drugs for treating EGFR-mutated NSCLC [13]. However, the emergence of tyrosine kinase inhibitor-resistant strains, exhibiting enhanced OXPHOS function, poses a challenge. Consequently, a strategy to overcome this resistance involves the combination of metformin and afatinib, a tyrosine kinase inhibitor, for NSCLC treatment [14]. Hence, the inhibition of mitochondrial function has emerged as a potential therapeutic strategy for effectively targeting lung cancer cells.

We conducted a study on the structure–activity relationships (SARs) of analogs derived from annonaceous acetogenin, aiming to develop novel antitumor agents with a unique mode of action [15–17]. Annonaceous acetogenins are polyketides featuring one to three tetrahydrofuran(s) (THFs) at the molecule’s center and an α,β-unsaturated-γ-lactone ring at the molecule's end [18]. In a previous report, we highlighted that the 1-methylpyrazole-5-carboxamide analog 1, based on solamin [19–21], a natural mono-THF acetogenin, exhibited a remarkable 1460-fold increase in growth inhibitory activity against human lung cancer NCI-H23 cells compared with solamin (Figure 1) [22]. However, 1 displayed in vivo acute toxicity [23]. Continuing the SAR investigation of heterocycles at the end of compound 1, it was observed that the thiophene carboxamide analog 2 exhibited antitumor activity against NCI-H23 [24] and glioblastoma stem cells (GSCs) [25,26] without causing acute toxicity. Subsequently, we investigated the activity of an analog containing a bis-THF moiety instead of a mono-THF moiety. This interest arose from the observation that natural acetogenins with bis-THF rings typically demonstrate more potent growth inhibitory activity than those with mono-THF rings [18]. However, the synthesis of bis-THF analogs poses challenges owing to the involvement of more synthetic steps, given the increased number of chiral centers on the THF moiety. Consequently, analog 3 was designed, synthesized, and evaluated for biological activity. The results revealed that analog 3 exhibited robust in vivo antitumor activity against GSCs by activating adenosine monophosphate-activated protein kinase (AMPK) [27]. We also documented that analog 4, wherein the amide bond between 1-methylpyrazole and the linker moiety of 1 was substituted with a sulfonamide bond, demonstrated antitumor activity without causing acute toxicity in vivo [23]. Furthermore, earlier SAR investigations focusing on the heterocycle at the end of the molecule—for instance, 2-thiophene [23] or the pyrimidine ring [15,28] — revealed that certain analogs with substituted heterocycles exhibited promising growth inhibitory activity.

Figure 1.

Previous structure–activity relationships studies and this work for novel antitumor agents.

This study details the synthesis of analogs with varying numbers of ethylene glycol units, connecting groups and heterocycle moieties; the assessment of growth inhibitory activities against A549 cells; the identification of the mode of action; and an in vivo study using xenograft mice.

2. Materials & methods

2.1. Chemistry

2.1.1. General information

Melting points were measured using a Yanaco MP micro-melting point apparatus or a SANSYO melting point apparatus SMP-500, and were uncorrected. NMR spectra were recorded in the specified solvents using Bruker Ascend™ 500 (¹H: 500 MHz; ¹³C: 125 MHz), JEOL ECS-400 (¹H: 400 MHz; ¹³C: 100 MHz), or Bruker Ultrashield™ 300 (¹H: 300 MHz; ¹³C: 75 MHz) spectrometers. Chemical shifts were recorded in ppm relative to that of the internal solvent signal [CDCl₃: 7.26 ppm (¹H NMR), 77.0 ppm (¹³C NMR)] or tetramethylsilane [0 ppm] used as the internal standard. The following abbreviations were used: broad singlet = br s, singlet = s, doublet = d, broad triplet = br t, triplet = t, quartet = q, quintet = qn, sextet = sext, septet = sep and multiplet = m. IR absorption spectra (FT = diffuse reflectance spectroscopy) were recorded with attenuated total reflectance (ATR) using a JASCO FT/IR-4600 spectrophotometer and listed only noteworthy absorptions (in cm⁻¹). Mass spectra were obtained using a JEOL JMS-600H, JEOL JMS-700, JEOL GC-mate II, JEOL SX-102A, or JEOL LCMS–IT–TOF mass spectrometer. Column chromatography was performed using a Kanto Chemical Silica Gel 60 N (spherical, neutral, 63–210 μm) column, and flash column chromatography was performed using a Merck Silica Gel 60 (40–63 μm) column. All air- and moisture-sensitive reactions were carried out in flame-dried glassware in an atmosphere consisting of Ar or N₂. All solvents were dried and distilled according to standard procedure, if necessary, while organic extracts were dried, filtered, and concentrated under reduced pressure using a rotary evaporator. The synthetic procedure for 2 [24], 3 [27], 1-azido-9-bromononane [29] and 1-methyl-1H-pyrazole-3-sulfonyl chloride [23] is presented in the previous report. The purity of all tested compounds was >95% as determined on a high-performance liquid chromatography using the following method: mobile phase, CH₂Cl₂; flow rate, 0.2 ml/min; column, COSMOSIL 5C₁₈-Ar-II 5.0 μm (4.6 mm × 250 mm) for 3 and 5a–m, or mobile phase, CH₂Cl₂; flow rate, 0.2 ml/min; column, CHIRALPAK IB 5 μm (4.6 mm × 250 mm) for 2.

2.1.2. 1-Azido-13,16,19-trioxanonacosane (7a)

NaH (60% in oil, 283 mg, 7.08 mmol) and 15-crown-5 ether (1.41 ml, 7.08 mmol) were added to a solution of 6a (873 mg, 3.54 mmol) in N,N-dimethylformamide (DMF; 10.0 ml) at 0°C. After stirring for 10 min at the same temperature, 1-azido-12-bromododecane (1.54 g, 5.31 mmol) in DMF (1.8 ml) was added to the mixture at the same temperature. After stirring for 23 h at rt, water was added to the reaction mixture at the same temperature and the mixture was extracted with EtOAc. The combined organic layers were washed with brine and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane to n-hexane/EtOAc (30:1 to 10:1) as eluent yielded 7a (1.38 g, 86%) as colorless needles. M.p. 40°C or lower; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 7.0 Hz), 1.26–1.37 (m, 30H), 1.55–1.62 (m, 6H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 4H, J = 6.8 Hz), 3.58–3.60 (m, 4H), 3.64–3.66 (m, 4H); ¹³C NMR (75 Hz, CDCl₃) δ: 14.1, 22.7, 26.1 (2C), 26.7, 28.8, 29.1, 29.3, 29.45, 29.48, 29.51, 29.53, 29.54, 29.56 (2C), 29.60, 29.64 (2C), 31.9, 51.5, 70.1 (2C), 70.6 (2C), 71.5 (2C); IR (ATR) cm^–1: 2094; MS (FAB) m/z (%): 456 [M+H]⁺, HRMS (FAB) m/z: calculated (Calcd) for C₂₆H₅₄N₃O₃: 456.4165; found: 456.4161 [M+H]⁺.

2.1.3. 1-Azido-13,16,19,22,25-pentaoxapentatriacontane (7b)

The procedure was the same as that used for the preparation of 7a using 6b instead of 6a, giving 7b (yield: 73%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.86 (t, 3H, J = 6.7 Hz), 1.24–1.37 (m, 30H), 1.51–1.63 (m, 6H), 3.24 (t, 2H, J = 7.0 Hz), 3.43 (t, 4H, J = 6.8 Hz), 3.54–3.58 (m, 4H), 3.61–3.64 (m, 12H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 26.0, 26.7, 28.8, 29.1, 29.3, 29.41, 29.43, 29.45, 29.46, 29.49, 29.50, 29.52, 29.57, 29.59 (2C), 29.62, 31.9, 51.4, 70.0 (2C), 70.55 (4C), 70.58 (2C), 71.5 (2C); IR (ATR) cm^–1: 2092; MS (FAB) m/z: 544 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₃₀H₆₂N₃O₅: 544.4689; found: 544.4700 [M+H]⁺.

2.1.4. 1-Azido-13,16,19,22,25,28-hexaoxaoctatriacontane (7c)

The procedure was the same as that used for the preparation of 7a using 6c instead of 6a, giving 7c (yield: 63%) as a white waxy solid. M.p. 40°C or lower; ¹H NMR (500 MHz, CDCl₃) δ: 0.85 (t, 3H, J = 6.8 Hz), 1.24–1.36 (m, 30H), 1.50–1.62 (m, 6H), 3.23 (t, 2H, J = 7.0 Hz), 3.42 (t, 4H, J = 6.8 Hz), 3.53–3.57 (m, 4H), 3.60–3.63 (m, 16H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.6, 26.0 (2C), 26.6, 28.8, 29.1, 29.2, 29.39, 29.40 (2C), 29.43, 29.47, 29.49 (2C), 29.52, 29.6 (2C), 31.8, 51.4, 70.0 (2C), 70.51 (5C), 70.54 (3C), 71.5 (2C); IR (ATR) cm^–1: 2094; MS (FAB) m/z: 610 [M + Na]⁺, HRMS (FAB) m/z: Calcd for C₃₂H₆₅N₃NaO₆: 610.4771; found: 610.4762 [M + Na]⁺.

2.1.5. N-(13,16,19-Trioxanonacosan-1-yl)-thiophene-3-carboxamide (5a)

PPh₃ (231 mg, 0.881 mmol) was added to a solution of 7a (200 mg, 0.434 mmol) in Et₂O/H₂O (1:1, 4.3 ml) at rt. After stirring 44 h at the same temperature, 3M NaOH aq. was added to the reaction mixture at the same temperature, and the mixture was extracted with Et₂O. The combined organic layers were dried over Na₂SO₄ prior to solvent evaporation, giving a crude 8a. Thiphene-3-carboxylic acid (185 mg, 1.44 mmol), 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride (250 mg, 1.30 mmol) and N,N-dimethylaminopyridine (159 mg, 1.30 mmol) were added to a solution of the crude 8a in THF (8.7 ml) at 0°C. After stirring for 18 h at rt, water was added to the reaction mixture at the same temperature and the mixture was extracted with Et₂O. The combined organic layers were washed with brine and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (2:1) as eluent yielded 5a (201 mg, 85%) as a white powder. M.p. 74.8–75.2°C; ¹H NMR (500 MHz, CDCl₃) δ: 0.86 (t, 3H, J = 6.7 Hz), 1.24–1.36 (m, 30H), 1.51–1.62 (m, 6H), 3.35–3.45 (m, 6H), 3.55–3.65 (m, 8H), 6.18 (br s, 1H), 7.30 (dd, 1H, J = 5.0, 3.0 Hz), 7.37 (dd, 1H, J = 5.0, 1.3 Hz), 7.84 (dd, 1H, J = 3.0, 1.3 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 26.0 (2C), 26.9, 29.3 (2C), 29.41, 29.43, 29.47, 29.50 (3C), 29.51, 29.55, 29.57 (2C), 29.7, 31.8, 39.8, 70.0 (2C), 70.6 (2C), 71.5 (2C), 126.0, 126.3, 127.8, 137.8, 163.0; IR (ATR) cm^–1: 3330, 1623; MS (FAB) m/z: 540 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₃₁H₅₈NO₄S: 540.4087; found: 540.4083 [M+H]⁺.

2.1.6. N-(13,16,19,22,25-Pentaoxapentatriacontan-1-yl)-thiophene-3-carboxamide (5b)

The procedure was the same as that used for the preparation of 5a using 7b instead of 7a, giving 5b (yield: 98%) as a white waxy solid. M.p. 60.2–61.1°C; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.26–1.39 (m, 30H), 1.57 (qn, 6H, J = 6.7 Hz), 3.38–3.47 (m, 6H), 3.56–3.59 (m, 4H), 3.62–3.66 (m, 12H), 5.93 (br s, 1H) 7.34 (dd, 1H, J = 5.1, 3.0 Hz), 7.37 (dd, 1H, J = 5.1, 1.5 Hz), 7.84 (dd, 1H, J = 3.0, 1.5 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 26.0 (2C), 26.9, 29.3 (2C), 29.40, 29.43, 29.45, 29.48 (2C), 29.50, 29.55, 29.56, 29.57 (2C), 29.7, 31.8, 39.8, 70.0 (2C), 70.5 (4C), 70.6 (2C), 71.5 (2C), 126.0, 126.3, 127.8, 137.8, 163.0; IR (ATR) cm^–1: 3332, 1623; MS (FAB) m/z: 628 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₃₅H₆₆NO₆S: 628.4611; found: 628.4619 [M+H]⁺.

2.1.7. N-(13,16,19,22,25,28-Hexaoxaoctatriacontan-1-yl)-thiophene-3-carboxamide (5c)

The procedure was the same as that used for the preparation of 5a using 7c instead of 7a, giving 5c (yield: 69%) as a white waxy solid. M.p. 57.4–57.8°C; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.26–1.33 (m, 30H), 1.57 (qn, 6H, J = 7.0 Hz), 3.38–3.46 (m, 6H), 3.56–3.59 (m, 4H), 3.62–3.65 (m, 16H), 5.97 (br s, 1H), 7.35 (dd, 1H, J = 5.1, 2.9 Hz), 7.37 (dd, 1H, J = 5.1, 1.4 Hz), 7.84 (dd, 1H, J = 2.9, 1.4 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 26.01, 26.03, 26.9, 29.3 (2C), 29.40, 29.43, 29.45, 29.48 (2C), 29.51, 29.55, 29.56 (2C), 29.57, 29.7, 31.8, 39.8, 70.0 (2C), 70.5 (5C), 70.6 (3C), 71.48, 71.49, 126.0, 126.3, 127.8, 137.8, 163.0; IR (ATR) cm^–1: 3334, 1623; MS (FAB) m/z: 672 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₃₇H₇₀NO₇S: 672.4873; found: 672.4879 [M+H]⁺.

2.1.8. 1-Azido-24-(t-butyldimethylsilyloxy)-13,16,19,22-tetraoxatetracosane (10a)

NaH (60% in oil, 31.1 mg, 0.778 mmol) was added to a solution of 9 (200 mg, 0.648 mmol) in DMF (0.65 ml) at 0°C. After stirring for 10 min at the same temperature, 1-azido-12-bromododecane (376 mg, 1.30 mmol) was added to the mixture at the same temperature. After stirring for 18 h at rt, water was added to the reaction mixture at same temperature and the mixture was extracted with EtOAc. The combined organic layers were washed with water (three-times) and brine, and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (20:1 to 5:1) as eluent yielded 10b (181 mg, 54%) as a colorless oil. ¹H NMR (400 MHz, CDCl₃) δ: 0.07 (s, 6H), 0.89 (s, 9H), 1.24–1.40 (m, 16H), 1.54–1.66 (m, 4H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 2H, J = 6.8 Hz), 3.54–3.59 (m, 4H), 3.63–3.68 (m, 10H), 3.77 (t, 2H, J = 5.4 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: –5.3 (2C), 18.3, 25.9 (3C), 26.0, 26.7, 28.8, 29.1, 29.42, 29.43, 29.47, 29.51, 29.52, 29.6, 51.4, 62.7, 70.0, 70.55, 70.58, 70.65, 70.66, 71.49, 71.50, 72.6; IR (ATR) cm^–1: 2093; MS (ESI) m/z: 540 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₂₆H₅₅N₃NaO₅Si: 540.3803; Found: 540.3794 [M + Na]⁺.

2.1.9. 1-Azido-21-(t-butyldimethylsilyloxy)-10,13,16,19-tetraoxahenicosane (10b)

The procedure was the same as that used for the preparation of 10a using 1-azido-9-bromononane instead of 1-azido-12-bromododecane, giving 10b (yield: 56%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.06 (s, 6H), 0.89 (s, 9H), 1.25–1.37 (m, 10H), 1.55–1.65 (m, 4H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 2H, J = 6.8 Hz), 3.55–3.59 (m, 4H), 3.63–3.66 (m, 10H), 3.77 (t, 2H, J = 5.5 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: –5.3 (2C), 18.3, 25.9 (3C), 26.0, 26.7, 28.8, 29.0, 29.3, 29.4, 29.6, 51.4, 62.7, 70.0, 70.57 (2C), 70.60, 70.7, 71.5 (2C), 72.6; IR (ATR) cm^–1: 2093; MS (ESI) m/z: 498 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₂₃H₄₉N₃NaO₅Si: 498.3334; Found: 498.3339 [M + Na]⁺.

2.1.10. 24-Azido-3,6,9,12-tetraoxatetracosan-1-ol (11a)

Tetra-n-butylammonium fluoride (ca. 1M in THF, 6.29 ml, 6.29 mmol) was added to a solution of 10a (326 mg, 0.629 mmol) in THF (6.3 ml) at 0°C. After stirring for 1 h at rt, sat. NH₄Cl aq. was added to the reaction mixture at the same temperature and the mixture was extracted with EtOAc. The combined organic layers were washed with brine and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (1:1 to 1:2) as eluent yielded 11a (204 mg, 80%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 1.26–1.39 (m, 16H), 1.55–1.74 (m, 4H), 2.69 (br, 1H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 2H, J = 6.8 Hz), 3.58–3.74 (m, 16H); ¹³C NMR (75 MHz, CDCl₃) δ: 26.0, 26.6, 28.7, 29.1 (2C), 29.37 (2C), 29.42, 29.45, 29.47, 51.4, 61.6, 69.9, 70.2, 70.4, 70.47, 70.50, 70.51, 71.5, 72.5; IR (ATR) cm^–1: 3458, 2093; MS (ESI) m/z: 426 [M + Na]⁺; HRMS (ESI) m/z: Calcd for C₂₀H₄₁N₃NaO₅: 426.2938; Found: 426.2939 [M + Na]⁺.

2.1.11. 21-Azido-3,6,9,12-tetraoxahenicosan-1-ol (11b)

The procedure was the same as that used for that used the preparation of 11a using 10b instead of 10a, giving 11b (yield: 93%) as a pale yellow oil. ¹H NMR (500 MHz, CDCl₃) δ: 1.25–1.39 (m, 10H), 1.55–1.62 (m, 4H), 2.84 (br, 1H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 2H, J = 6.8 Hz), 3.58–3.74 (m, 16H); ¹³C NMR (75 MHz, CDCl₃) δ: 25.9, 26.6, 28.7, 29.0, 29.2, 29.3, 29.4, 51.3, 61.6, 69.9, 70.2, 70.42, 70.44, 70.47, 70.48, 71.4, 72.5; IR (ATR) cm^–1: 3456, 2092; MS (ESI) m/z: 384 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₁₇H₃₅N₃NaO₅: 384.2469; Found: 384.2471 [M + Na]⁺.

2.1.12. 1-Azido-13,16,19,22,25-pentaoxadotriacontane (12a)

The procedure was the same as that used for the preparation of 10b using 11a and 1-iodoheptane instead of 9 and 1-azido-9-bromononane, giving 12a (yield: 74%) as a colorless oil. ¹H NMR (400 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.9 Hz), 1.23–1.40 (m, 24H), 1.54–1.68 (m, 6H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 4H, J = 6.8 Hz), 3.57–3.59 (m, 4H), 3.63–3.66 (m, 12H); ¹³C NMR (100 MHz, CDCl₃) δ: 14.1, 22.6, 25.98, 26.01 (2C), 26.7, 28.8, 29.1, 29.42, 29.43, 29.47, 29.49, 29.50, 29.52, 29.6, 31.8, 51.4, 70.0, 70.49 (2C), 70.50 (2C), 70.53 (2C), 70.54, 71.5 (2C); IR (ATR) cm^–1: 2093; MS (ESI) m/z: 524 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₂₇H₅₅N₃NaO₅: 524.4034; Found: 524.4039 [M + Na]⁺.

2.1.13. 1-Azido-10,13,16,19,22-pentaoxadotriacontane (12b)

The procedure was the same as that used for that used the preparation of 10b using 11b and 1-iododecane instead of 9 and 1-azido-9-bromononane, giving 12b (yield: 80%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 7.0 Hz), 1.26–1.38 (m, 24H), 1.55–1.69 (m, 6H), 3.26 (t, 2H, J = 7.0 Hz), 3.45 (t, 4H, J = 6.8 Hz), 3.57–3.60 (m, 4H), 3.62–3.66 (m, 12H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 25.97, 26.02, 26.6 (2C), 28.8, 29.0, 29.25, 29.28, 29.33, 29.4, 29.50, 29.54, 29.6, 31.8, 51.4, 70.0 (2C), 70.52 (2C), 70.54 (2C), 71.4 (2C), 71.5 (2C); IR (ATR) cm^–1: 2093; MS (ESI) m/z: 524 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₂₇H₅₅N₃NaO₅: 524.4034; Found: 524.4035 [M + Na]⁺.

2.1.14. N-(13,16,19,22,25-Pentaoxadotriacontan-1-yl)-thiophene-3-carboxamide (5d)

The procedure was the same as that used for that used the preparation of 5a using 12a instead of 7a, giving 5d (yield: 58%) as a white waxy solid. M.p. 51.6–52.4°C; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 7.0 Hz), 1.22–1.39 (m, 24H), 1.55–1.63 (m, 6H), 3.40–3.46 (m, 6H), 3.57–3.60 (m, 4H), 3.62–3.66 (m, 12H), 5.96 (br, 1H), 7.34 (dd, 1H, J = 5.0, 3.0 Hz), 7.37 (dd, 1H, J = 5.0, 1.3 Hz), 7.84 (dd, 1H, J = 3.0, 1.3 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.5, 25.95, 25.98, 26.9, 29.1, 29.2, 29.37 (2C), 29.43, 29.45 (2C), 29.52, 29.53, 29.6, 31.7, 39.8, 69.9 (3C), 70.48 (2C), 70.52 (3C), 71.4 (2C), 126.1, 126.2, 127.8, 137.7, 163.0; IR (ATR) cm^–1: 3333, 1623; MS (EI) m/z (%): 585 (11.3) [M]⁺, 310 (44.3), 294 (42.4), 111 (100.0); HRMS (EI) m/z: Calcd for C₃₂H₅₉NO₆S: 585.4063; Found: 585.4066 [M]⁺.

2.1.15. N-(10,13,16,19,22-Pentaoxadotriacontan-1-yl)-thiophene-3-carboxamide (5e).

The procedure was the same as that used for that used the preparation of 5a using 12b instead of 7a, giving 5e (yield: 84%) as a white waxy solid. M.p. 50.5–52.3°C; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 7.0 Hz), 1.23–1.40 (m, 24H), 1.55–1.63 (m, 6H), 3.43 (td, 2H, J = 7.2, 6.0 Hz), 3.45 (t, 4H, J = 6.8 Hz), 3.57–3.60 (m, 4H), 3.62–3.66 (m, 12H), 5.96 (br, 1H), 7.34 (dd, 1H, J = 5.1, 3.1 Hz), 7.38 (dd, 1H, J = 5.1, 1.3 Hz), 7.85 (dd, 1H, J = 3.1, 1.3 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.6, 25.9, 26.0, 26.8, 29.1, 29.21, 29.24, 29.3, 29.4 (2C), 29.45 (2C), 29.49, 29.6, 31.8, 39.7, 69.90, 69.91, 70.4 (3C), 70.5 (2C), 71.36, 71.43 (2C), 126.09, 126.10, 127.8, 137.7, 163.0; IR (ATR) cm^–1: 3332, 1622; MS (EI) m/z (%): 585 (17.6) [M]⁺, 268 (69.6), 252 (71.2), 111 (100.0); HRMS (EI) m/z: Calcd for C₃₂H₅₉NO₆S: 585.4063; Found: 585.4064 [M]⁺.

2.1.16. 1-Benzyloxy-7,10,13,16,19,22-hexaoxadotriacontane (14)

The procedure was the same as that used for the preparation of 7a using 6c and 1-benzyloxy-6-bromohexane instead of 6a and 1-azido-12-bromododecane, giving 14 (yield: 60%) as a pale yellow oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.87 (3H, t, J = 6.7 Hz), 1.25–1.43 (m, 18H), 1.52–1.66 (m, 6H), 3.44 (t, 4H, J = 6.7 Hz), 3.45 (t, 2H, J = 6.7 Hz), 3.55–3.58 (m, 4H), 3.61–3.65 (m, 16H), 4.49 (s, 2H), 7.23–7.37 (m, 5H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.6, 25.9, 25.98, 26.00, 29.2, 29.4, 29.49 (2C), 29.52, 29.54, 29.6, 31.8, 70.0 (2C), 70.3, 70.50 (6C), 70.53 (2C), 71.3, 71.5, 72.8, 127.4, 127.5 (2C), 128.2 (2C), 138.6; IR (ATR) cm^–1: No characteristic absorptions were observed.; MS (EI) m/z (%): 568 (3.8) [M]⁺, 477 (2.0), 91 (100); HRMS (EI) m/z: Calcd for C₃₃H₆₀O₇: 568.4339; found: 568.4338 [M]⁺.

2.1.17. 7,10,13,16,19,22-Hexaoxadotriacontan-1-ol (15)

A solution of 14 (614 mg, 1.08 mmol) in EtOAc (10.8 ml) was hydrogenated on 20% Pd(OH)₂–C (wetted with ca. 50% water, 61.4 mg) with stirring at rt for 3 h under 3 atm pressure of hydrogen. The catalyst was filtered off through a pad of Celite^® and the filtrate was concentrated under reduced pressure. Purification using column chromatography over silica gel with EtOAc as eluent yielded 15 (475 mg, 92%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.84 (t, 3H, J = 6.8 Hz), 1.22–1.36 (m, 18H), 1.49–1.58 (m, 6H), 1.80 (br s, 1H), 3.41 (t, 2H, J = 6.6 Hz), 3.42 (t, 2H, J = 6.6 Hz), 3.52–3.56 (m, 4H), 3.59–3.62 (m, 18H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.6, 25.5, 25.8, 26.0, 29.2, 29.38, 29.44, 29.46, 29.50, 29.51, 31.8, 32.6, 62.6, 69.9, 70.0, 70.48 (6C), 70.50 (2C), 71.2, 71.4; IR (ATR) cm^–1: 3472; MS (FAB) m/z: 479 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₂₆H₅₅O₇: 479.3948; found: 479.3949 [M+H]⁺.

2.1.18. 7,10,13,16,19,22-Hexaoxa-1-iododotriacontane (16)

I₂ (269 mg, 1.06 mmol) and imidazole (72.1 mg, 1.06 mmol) were added to a solution of PPh₃ (278 mg, 1.06 mmol) in CH₂Cl₂ (1.1 ml) at 0°C. After stirring for 1 h at the same temperature, 15 (392 mg, 0.818 mmol) in CH₂Cl₂ (3.0 ml) was added to the mixture at the same temperature. After stirring for 45 min at rt, sat. Na₂S₂O₃ aq. was added to the reaction mixture at the same temperature and the mixture was extracted with EtOAc. The combined organic layers were washed with brine and dried over MgSO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (10:1) as eluent yielded 16 (441 mg, 92%) as a pale yellow oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.85 (t, 3H, J = 6.7 Hz), 1.23–1.44 (m, 18H), 1.50–1.61 (m, 4H), 1.80 (qn, 2H, J = 7.0 Hz), 3.16 (t, 2H, J = 7.0 Hz), 3.42 (t, 2H, J = 7.0 Hz), 3.43 (t, 2H, J = 6.6 Hz), 3.53–3.57 (m, 4H), 3.60–3.63 (m, 16H); ¹³C NMR (75 Hz, CDCl₃) δ: 7.0, 14.1, 22.6, 25.0, 26.0, 29.3, 29.4, 29.49, 29.53 (2C), 29.6, 30.2, 31.8, 33.4, 69.97, 70.03, 70.5 (6C), 70.6 (2C), 71.1, 71.5; IR (ATR) cm^–1: No characteristic absorptions were observed.; MS (FAB) m/z: 589 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₂₆H₅₄IO₆: 589.2965; found: 589.2980 [M+H]⁺.

2.1.19. 1-Azido-7,10,13,16,19,22-hexaoxadotriacontane (17)

NaN₃ (85.7 mg, 1.32 mmol) was added to a solution of 16 (382 mg, 0.649 mmol) in DMSO (2.2 ml) at rt. After stirring for 2 h at the same temperature, water was added to the reaction mixture at the same temperature and the mixture was extracted with EtOAc. The combined organic layers were washed with brine and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (20:1 to 1:1) as eluent yielded 17 (308 mg, 94%) as a colorless oil. ¹H NMR (500 MHz, CDCl₃) δ: 0.85 (t, 3H, J = 6.7 Hz), 1.23–1.38 (m, 18H), 1.50–1.62 (m, 6H), 3.23 (t, 2H, J = 6.8 Hz), 3.42 (t, 2H, J = 6.8 Hz), 3.43 (t, 2H, J = 6.8 Hz), 3.53–3.56 (m, 4H), 3.60–3.63 (m, 16H); ¹³C NMR (75 MHz, CDCl₃) δ: 14.0, 22.6, 25.6, 26.0, 26.5, 28.7, 29.2, 29.39, 29.41, 29.48, 29.52, 29.6, 31.8, 51.3, 69.96, 70.02, 70.5 (8C), 71.1, 71.5; IR (ATR) cm^–1: 2092; MS (FAB) m/z: 504 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₂₆H₅₄N₃O₆: 504.4013; found: 504.4034 [M+H]⁺.

2.1.20. N-(7,10,13,16,19,22-Hexaoxadotriacontan-1-yl)-thiophene-3-carboxamide (5f)

The procedure was the same as that used for that used the preparation of 5a using 17 instead of 7a, giving 5f (yield: 90%) as a pale brown waxy solid. M.p. 40°C or lower; ¹H NMR (500 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.26–1.42 (m, 18H), 1.52–1.65 (m, 6H), 3.38–3.48 (m, 6H), 3.55–3.59 (m, 4H), 3.62–3.65 (m, 16H), 6.07 (br s, 1H), 7.33 (dd, 1H, J = 5.1, 3.1 Hz), 7.39 (dd, 1H, J = 5.1, 1.4 Hz), 7.86 (dd, 1H, J = 3.1, 1.4 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 25.7, 26.0, 26.6, 29.3, 29.37, 29.42, 29.50, 29.54 (2C), 29.6, 31.8, 39.6, 69.97, 69.99, 70.5 (6C), 70.6 (2C), 71.1, 71.5, 126.1, 126.2, 127.9, 137.7, 163.1; IR (ATR) cm^–1: 3329, 1623; MS (FAB) m/z: 588 [M+H]⁺, HRMS (FAB) m/z: Calcd for C₃₁H₅₈NO₇S: 588.3934; found: 588.3925 [M+H]⁺.

2.1.21. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-thiophene-2-carboxamide (5g)

The procedure was the same as that used for the preparation of 5a using 19 and thiophene-2-carboxylic acid instead of 7a and thiophene-3-carboxylic acid, giving 5g (yield: 86%) as a white waxy solid. M.p. 42.1–43.3°C; ¹H NMR (300 MHz, CDCl₃) δ: 0.87 (t, 3H, J = 6.9 Hz), 1.25–1.38 (m, 30H), 1.57 (qn, 4H, J = 7.0 Hz), 1.60 (qn, 2H, J = 7.0 Hz), 3.41 (t, 2H, J = 6.9 Hz), 3.44 (t, 4H, J = 6.9 Hz), 3.56–3.58 (m, 4H), 3.63–3.66 (m, 8H), 5.98 (br s, 1H), 7.07 (dd, 1H, J = 5.0, 3.8 Hz), 7.46 (d, 1H, J = 5.0 Hz), 7.48 (d, 1H, J = 3.8 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.7, 26.0, 26.1, 26.9, 29.26, 29.29, 29.4, 29.46, 29.47, 29.50, 29.51, 29.52, 29.53, 29.58, 29.59, 29.65, 29.66, 31.9, 40.0, 70.0 (2C), 70.57 (2C), 70.59 (2C), 71.5 (2C), 127.5, 127.7, 129.5, 139.2, 161.8; IR (ATR) cm^–1: 3321, 1622; MS (EI) m/z (%): 583 (13.6) [M]⁺, 472 (7.5), 310 (83.9), 294 (66.1), 111 (100), 83 (13.7); HRMS (EI) m/z: Calcd for C₃₃H₆₁NO₅S: 583.4270; found: 583.4268 [M]⁺.

2.1.22. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-1-methyl-1H-pyrazole-5-carboxamide (5h)

The procedure was the same as that used for the preparation of 5a using 19 and 1-methyl-1H-pyrazole-5-carboxylic acid instead of 7a and thiophene-3-carboxylic acid, giving 5 h (yield: 58%) as a white waxy solid. M.p. 40°C or lower; ¹H NMR (400 MHz, CDCl₃) δ: 0.86 (t, 3H, J = 6.7 Hz), 1.18–1.32 (m, 30H), 1.52–1.61 (m, 6H), 3.35–3.45 (m, 6H), 3.55–3.65 (m, 12H), 4.16 (s, 3H) 5.95 (br, 1H), 6.45 (d, 1H, J = 2.0 Hz), 7.42 (d, 1H, J = 2.0 Hz); ¹³C NMR (100 MHz, CDCl₃) δ: 14.1, 22.6, 26.0, 26.1, 26.9, 29.2, 29.3 (2C), 29.4 (4C), 29.5 (4C), 29.7 (2C), 31.8, 39.2, 39.5, 70.0, 70.5 (4C), 70.9 (2C), 71.5, 105.9, 135.4, 137.4, 160.0; IR (ATR) cm^–1: 3271, 1641; MS (EI) m/z (%): 581 (8.2) [M]⁺, 308 (78.4), 270 (69.0), 109 (100.0); HRMS (EI) m/z: Calcd for C₃₃H₆₃N₃O₅: 581.4768; Found: 581.4768 [M]⁺.

2.1.23. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-pyrimidine-4-carboxamide (5i)

The procedure was the same as that used for the preparation of 5a using 19 and pyrimidine-4-carboxylic acid instead of 7a and thiophene-3-carboxylic acid, giving 5i (yield: 98%) as a white waxy solid. M.p. 46.9–47.5°C; ¹H NMR (300 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.26–1.41 (m, 30H), 1.53–1.69 (m, 6H), 3.42–3.51 (m, 6H), 3.56–3.60 (m, 4H), 3.63–3.66 (m, 8H), 8.00 (br s, 1H), 8.13 (dd, 1H, J = 5.2, 1.2 Hz), 8.97 (d, 1H, J = 5.2 Hz), 9.23 (d, 1H, J = 1.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.7, 26.1, 26.9, 29.27, 29.31, 29.45, 29.48, 29.50, 29.55, 29.56, 29.60, 29.62 (4C), 29.7 (2C), 31.9, 39.6, 70.0 (2C), 70.60 (2C), 70.62 (2C), 71.5 (2C), 118.5, 156.4, 157.7, 159.2, 162.5; IR (ATR) cm^–1: 3364, 1656; MS (ESI) m/z: 602 [M + Na]⁺, HRMS (FAB) m/z: Calcd for C₃₃H₆₁N₃NaO₅: 602.4509; found: 602.4504 [M + Na]⁺.

2.1.24. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-thiophene-3-sulfonamide (5j)

PPh₃ (84.7 mg, 0.323 mmol) was added to a solution of 19 (100 mg, 0.200 mmol) in Et₂O/H₂O (1:1, 2.0 ml) at rt. After stirring 14 h at the same temperature, 1M NaOH aq. was added to the reaction mixture at the same temperature, and the mixture was extracted with Et₂O. The combined organic layers were dried over Na₂SO₄ prior to solvent evaporation, giving a crude 20. Thiophene-3-sulfonyl chloride (110 mg, 0.600 mmol) and triethylamine (0.139 ml, 1.00 mmol) were added to a solution of the crude 20 in CH₂Cl₂ (2.0 ml) at rt. After stirring for 19 h at rt, water was added to the reaction mixture at the same temperature and the mixture was extracted with CH₂Cl₂. The combined organic layers were washed with brine and dried over Na₂SO₄ prior to solvent evaporation. Purification using column chromatography over silica gel with n-hexane/EtOAc (5:1) as eluent yielded 5j (95.0 mg, 77%) as a white waxy solid. M.p. 40.6–41.7°C; ¹H NMR (300 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.22–1.33 (m, 30H), 1.42–1.58 (m, 6H), 3.00 (q, 2H, J = 6.8 Hz), 3.45 (t, 4H, J = 6.9 Hz), 3.56–3.60 (m, 4H), 3.63–3.66 (m, 8H), 4.33 (br t, 1H, J = 6.0 Hz), 7.33 (dd, 1H, J = 5.1, 1.3 Hz), 7.43 (dd, 1H, J = 5.1, 3.0 Hz), 7.96 (dd, 1H, J = 3.0, 1.3 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.7, 26.07, 26.08, 26.5, 29.0, 29.3, 29.4, 29.46, 29.49 (2C), 29.53, 29.57 (2C), 29.60, 29.62 (2C), 29.7, 31.9, 43.3, 70.0 (2C), 70.61 (2C), 70.62 (2C), 71.5, 71.6, 125.4, 127.9, 130.3, 140.1; IR (ATR) cm^–1: 3277, 1320; MS (ESI) m/z: 642 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₃₂H₆₁NNaO₆S₂: 642.3838; found: 642.3837 [M + Na]⁺.

2.1.25. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-thiophene-2-sulfonamide (5k)

The procedure was the same as that used for the preparation of 5j using thiophene-2-sulfonyl chloride instead of thiophene-3-sulfonyl chloride, giving 5k (yield: 71%) as a white waxy solid. M.p. 40°C or lower; ¹H NMR (300 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.23–1.33 (m, 30H), 1.44–1.60 (m, 6H), 3.04 (q, 2H, J = 6.9 Hz), 3.45 (t, 4H, J = 6.9 Hz), 3.56–3.60 (m, 4H), 3.63–3.66 (m, 8H), 4,43 (br t, 1H, J = 6.0 Hz), 7.10 (dd, 1H, J = 3.7, 1.4 Hz), 7.59 (dd, 1H, J = 5.0, 1.4 Hz), 7.61 (dd, 1H, J = 5.0, 3.7 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.7, 26.06, 26.08, 26.5, 29.0, 29.3, 29.4, 29.45 (3C), 29.49 (2C), 29.53, 29.57, 29.61, 29.62, 29.7, 31.9, 43.5, 70.0 (2C), 70.6 (2C), 71.5 (2C), 71.6 (2C), 127.3, 131.7, 132.0, 141.0; IR (ATR) cm^–1: 3269, 1336; MS (ESI) m/z: 642 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₃₂H₆₁NNaO₆S₂: 642.3838; found: 642.3827 [M + Na]⁺.

2.1.26. N-(13,16,19,22-Tetraoxadotriacontan-1-yl)-1-methyl-1H-pyrazole-5-sulfonamide (5l)

The procedure was the same as that used for the preparation of 5j using 1-methyl-1H-pyrazole-3-sulfonyl chloride instead of thiophene-3-sulfonyl chloride, giving 5l (yield: 94%) as a white waxy solid. M.p. 40°C or lower; ¹H NMR (300 MHz, CDCl₃) δ: 0.86 (t, 3H, J = 6.7 Hz), 1.22–1.32 (m, 30H), 1.48 (qn, 2H, J = 6.7 Hz), 1.56 (qn, 4H, J = 6.7 Hz), 3.04 (q, 2H, J = 6.7 Hz), 3.45 (t, 4H, J = 6.7 Hz), 3.55–3.58 (m, 4H), 3.61–3.65 (m, 8H), 4.08 (s, 3H), 4.89 (br s, 1H), 6.74 (d, 1H, J = 2.0 Hz), 7.45 (d, 1H, J = 2.0 Hz); ¹³C NMR (75 MHz, CDCl₃) δ: 14.1, 22.6, 26.0, 26.1, 26.4, 29.0, 29.3, 29.35, 29.40 (2C), 29.44, 29.45 (2C), 29.48, 29.54, 29.57 (2C), 29.63, 31.9, 38.5, 43.3, 70.0, 70.56, 70.57 (2C), 71.50 (2C), 71.52 (2C), 111.0, 137.6, 139.0; IR (ATR) cm^–1: 3291, 1325; MS (FAB) m/z: 618 [M+H]⁺, HRMS (FAB) m/z: calcd for C₃₂H₆₄N₃O₆S: 618.4516; found: 618.4511 [M+H]⁺.

2.1.27. 21-Azido-3,6,9-trioxahenicosan-1-ol (22)

The procedure was the same as that used for the preparation of 11a using 21 instead of 10a, giving 22 (yield: 98%) as a colorless oil. ¹H NMR (400 MHz, CDCl₃) δ: 1.23–1.40 (m, 16H), 1.55–1.67 (m, 4H), 2.54 (t, 1H, J = 6.4 Hz), 3.26 (t, 2H, J = 7.1 Hz), 3.45 (t, 2H, J = 6.8 Hz), 3.58–3.75 (m, 12H); ¹³C NMR (100 MHz, CDCl₃) δ: 25.9 (2C), 26.6 (2C), 28.7 (2C), 29.0 (2C), 29.3, 29.4, 51.3, 61.5, 69.8, 70.2, 70.4 (2C), 71.4, 72.4; IR (ATR) cm^–1: 3445, 2091; MS (ESI) m/z: 382 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₁₈H₃₇N₃O₄: 382.2676; Found: 382.2680 [M + Na]⁺.

2.1.28. 1-Azido-13,16,19,22-tetraoxaoctacosane (23)

The procedure was the same as that used for the preparation of 10b using 22 and 1-iodohexane instead of 9 and 1-azido-9-bromononane, giving 23 (yield: 43%) as a colorless oil. ¹H NMR (300 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.8 Hz), 1.27–1.35 (m, 22H), 1.53–1.64 (m, 6H), 3.25 (t, 2H, J = 7.0 Hz), 3.45 (t, 4H, J = 6.8 Hz), 3.56–3.66 (m, 12H); ¹³C NMR (100 MHz, CDCl₃) δ: 14.0, 22.5, 25.7, 26.0, 26.6, 28.7, 29.1, 29.39 (2C), 29.43, 29.46, 29.48, 29.52, 29.6, 31.6, 51.4, 70.0 (2C), 70.5 (4C), 71.4 (2C); IR (ATR) cm^–1: 2093; MS (ESI) m/z: 466 [M + Na]⁺, HRMS (ESI) m/z: Calcd for C₂₄H₄₉N₃O₄: 466.3615; Found: 466.3614 [M + Na]⁺.

2.1.29. N-(13,16,19,22-Tetraoxaoctaconsan-1-yl)-1-methyl-1H-pyrazole-5-carboxamide (5m)

The procedure was the same as that used for the preparation of 5a using 23 and 1-methyl-1H-pyrazole-5-carboxylic acid instead of 7a and thiophene-3-carboxylic acid, giving 5m (yield: 69%) as a pale yellow oil. ¹H NMR (400 MHz, CDCl₃) δ: 0.88 (t, 3H, J = 6.7 Hz), 1.26–1.33 (m, 22H), 1.54–1.61 (m, 6H), 3.37–3.46 (m, 6H), 3.57–3.66 (m, 12H), 4.18 (s, 3H), 5.98 (br, 1H), 6.47 (d, 1H, J = 2.0 Hz), 7.44 (d, 1H, J = 2.0 Hz); ¹³C NMR (100 MHz, CDCl₃) δ: 13.9, 22.5, 25.7, 26.0, 26.9, 29.2, 29.4 (4C), 29.5 (3C), 31.6 (2C), 39.1, 39.5, 70.0 (2C), 70.5 (4C), 71.4 (2C), 105.9, 135.4, 137.4, 159.9; IR (ATR) cm^–1: 3352, 1666; MS (EI) m/z (%): 525 (15.3) [M]⁺, 416 (25.1), 308 (100.0), 292 (82.2), 138 (38.8), 109 (100.0), 85 (51.2), 43 (75.9); HRMS (EI) m/z: Calcd for C₂₉H₅₅N₃O₅: 525.4142; Found: 525.4141 [M]⁺.

2.2. Biology

2.2.1. Reagents

The tested compounds were dissolved in dimethyl sulfoxide (DMSO; Nacalai Tesque, Kyoto, Japan).

2.2.2. Cell culture

A549 human NSCLC cells (ATCC, Manassas, VA, USA) and IMR90 normal human lung fibroblast cells (JCRB Cell Bank, Osaka, Japan) were maintained in DMEM (Wako Pure Chemical Industries, Osaka, Japan), supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin solution (Wako) at 37°C in humidified air containing 5% CO₂.

2.2.3. Cell growth assay

The WST-8 assay was performed using an SF kit (Nacalai Tesque). Absorbance at 450 nm was measured and normalized with DMSO-treated control cells.

2.2.4. Calculation of GI₅₀

Cell proliferation was assessed with counting numbers with trypan blue staining of living A549 cells and IMR90 treated with DMSO or the indicated concentrations of the compounds for 3 days. GI₅₀ value was calculated with CalcuSyn 2.11 software (Biosoft, Cambridge, UK).

2.2.5. Western blot analysis

SDS buffer (1%) supplemented with a protease inhibitor cocktail (Nacalai Tesque) and PhosSTOP EASYpack (Roche Diagnostics, Indianapolis, IN, USA) was used for preparation of the protein samples, which were separated by SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, MA, USA). Fat-free dried milk (5%) in Tris-buffered saline with 0.05% Tween20 (TBST) or Blocking One-P (Nacalai Tesque) were used for blocking. Membranes were then incubated with primary and secondary antibodies. The chemiluminescent signals were visualized using a ChemiDoc XRS Plus system (Bio-Rad) with Clarity Western ECL substrate (Bio-Rad Laboratories, Hercules, CA, USA). The following antibodies were used; p-AMPKα (1:1000; #2535, Cell Signaling Technology, Danvers, MA, USA), glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1:1000; 016–25523, Wako Pure Chemical Industries, Osaka, Japan), horseradish peroxidase-conjugated horse anti-mouse IgG (1:2000; PI-2000, Vector Laboratories, Burlingame, CA, USA) and HRP-conjugated goat anti-rabbit IgG (1:2000; #7074, Cell Signaling Technology).

2.2.6. Xenograft study

All experiments were carried out under the approval of the Institutional Ethical Committee for Animal Experiments of Kyoto Pharmaceutical University (approval number: CLON19-001). SCID mice (6–8 weeks old) were subcutaneously inoculated with 1 × 10⁶ A549 cells on the lower sides of trunk. Analog 5e was diluted at 15 mg/ml in DMSO with Kolliphor EL (Sigma-Aldrich) (final 7.5%) and then diluted in saline (total 200 μl/20 g body weight). Mice were treated with DMSO or 5e intraperitoneally at a dose of 15 mg/kg, three-times a week for 4 weeks from the next day of the inoculation. Tumor size was assessed with vernier calipers once a week for 4 weeks, and tumor volume was calculated as 0.5 × (length × width²).

2.2.7. Statistical analysis

Data are displayed as the mean ± SD unless otherwise indicated. All analysed data were obtained from at least three independent experiments. Statistical analysis was performed with Bell Curve for Excel (Social Survey Research Information Co., Ltd. Tokyo, Japan). Dunnett’s multiple comparison test or a two-tailed Student’s t-test were used for the in vitro or in vivo studies, and a value of p < 0.05 was considered statistically significant.

3. Results

3.1. Synthesis

Figure 2A illustrates the synthesis of analogs 5a–c, each containing a distinct number of ethylene glycol units. Diethylene glycol monodecyl ether 6a [30] was etherified using 1-azido-12-bromododecane [31] to yield azide 7a. The reduction of the azido group in 7a produced the primary amine 8a, which was then condensed with 3-thiophene carboxylic acid in the presence of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide (EDC), resulting in a high yield of the diethylene analog 5a. Analogs 5b–c, featuring varying numbers of ethylene glycol units, were synthesized using the same method employed for the corresponding starting materials 6b–c [32,33].

Figure 2.

(A-C) Synthesis of analogs 5a–f. Reagents and conditions: (a) 1-azido-12-bromododecane, NaH, 15-crown-5 ether, DMF, 0°C to rt, 7a (86%); 7b (73%); 7c (63%); (b) PPh₃, H₂O, Et₂O, 0°C to rt; (c) thiophene-3-carboxylic acid, EDC·HCl, DMAP, THF, 0°C to rt, 5a (85% from 7a); 5b (98% from 7b); 5c (69% from 7c); 5d (58% from 12a); 5e (84% from 12b); 5f (90% from 17); (d) 1-azido-bromoalkane, NaH, DMF, 0°C to rt, 10a (54%); 10b (56%); (e) n-Bu₄NF, THF, rt, 11a (80%); 11b (93%); (f) alkyl iodide, NaH, DMF, 0°C to rt, 12a (74%); 12b (80%); (g) 1-benzyloxy-6-bromohexane, NaH, 15-crown-5 ether, DMF, 0°C to rt, 60%; (h) H₂ (3 atm), 10% Pd(OH)₂-C, EtOAc, rt, 92%; (i) I₂, PPh₃, imidazole, CH₂Cl₂, 0°C to rt, 92%; (j) NaN₃, DMSO, rt, 94%.

Bn: Benzyl; DMF: N,N-Dimethylformamide; DMAP: N,N-Dimethylaminopyridine; DMSO: Dimethyl sulfoxide; EDC: 1-(3-Dimetylaminopropyl)-3-ethylcarbodiimide; TBS: t-Butyldimethylsilyl; THF: Tetrahydrofuran.

The synthesis of tetraethylene analogs 5d–e, with alkyl chains of varying lengths compared with 5b, is depicted in Figure 2B. The process commenced with the etherification of mono-t-butyldimethylsilyl (TBS)-protected tetraethylene glycol 9 [34], producing 10a in moderate yields. Deprotection of 10a led to the formation of primary alcohol 11a. Subsequent etherification of alcohol 11a, utilizing NaH in the presence of 15-crown-5 ether, yielded 12a. The reduction of the azido group in 12a generated primary amine 13a, followed by condensation with 3-thiophene carboxylic acid, resulting in the synthesis of analog 5d in moderate yields. Analog 5e, featuring a linker chain shorter than that of 5b, was synthesized using the same protocol employed for the corresponding alkylating reagent.

The pentaethylene glycol analog 5f, incorporating an n-hexyl alkyl linker chain, could have been synthesized by etherifying 6c with 1-azido-6-bromohexane. However, this reaction was abandoned because of concerns regarding the explosive nature of 1-azido-6-bromohexane. Consequently, an azido group was introduced at a later stage of the synthesis (Figure 2C). Alkylation of 6c with 1-benzyloxy-6-bromohexane [35] yielded benzyl ether 14. The benzyl group of 14 was deprotected by hydrogenolysis to produce primary alcohol 15. The hydroxyl group of 15 was converted to an azido group via iodination to yield azide 17. Subsequently, the desired analog 5f was synthesized under the same reaction conditions as those of 5a.

The synthesis of analogs 5g–l, which included various heterocycles and connecting groups, is shown in Figure 3A. Amine 20, which was prepared via 19, as previously reported [27] was condensed to 2-thiophene carboxylic acid to give the 2-thiophene carboxamide analog 5g in high yields. Other carboxamide analogs 5h–i were also synthesized under the same conditions, and sulfonamides 5j–l were obtained with the corresponding sulphonyl chlorides from 20.

Figure 3.

(A & B) Synthesis of analogs 5g–m. Reagents and conditions: (a) PPh₃, H₂O, Et₂O, 0°C to rt; (b) Ar-COOH, EDC·HCl, DMAP, THF, 0°C to rt, 5g (86% from 19); 5h (53% from 19); 5i (98% from 19); (c) Ar-SO₂Cl, Et₃N, CH₂Cl₂, rt, 5j (77% from 19); 5k (71% from 19); 5l (94% from 19); (d) n-Bu₄NF, THF, rt, 98%; (e) 1-iodohexane, NaH, DMF, 0°C to rt, 43%; (f) 1-methyl-1H-pyrazole-5-carboxylic acid, EDC·HCl, DMAP, THF, 0°C to rt, 69% from 23.

Finally, we sought to synthesize the 1-methylpyrazole carboxamide analog 5m with an n-hexyl alkyl tail chain by applying the same synthetic route as that of 3. However, an undesired N-alkylation reaction occurred upon the introduction of the n-hexyl group. Therefore, the alkyl tail chain of 5m was introduced before the condensation reaction. Figure 3B outlines the synthesis of 5m obtained from TBS-protected azide 21 [27] through the following steps: deprotection of the TBS group, etherification using 1-iodohexane, hydrogenation and condensation with 1-methyl-1H-pyrazole-5-carboxylic acid.

3.2. In vitro A549 growth inhibitory activity of synthesized analogs

The growth inhibitory activity of the synthesized analogs 5a–l and lead compounds (2 and 3) against A549 cells was evaluated (Figure 4A & B). A549 cells have been shown to respond to hypoxia and glucose depletion stimuli, with an effective switching mechanism between mitochondrial oxidative phosphorylation and glycolytic metabolism [36]. Therefore, A549 is a representative NSCLC cell line with confirmed active energy metabolism through mitochondrial oxidative phosphorylation, making it suitable for this study. The A549 cells were treated with 500 nM of each compound. Both lead compounds (2 and 3) inhibited cell growth compared with the controls; however, 2 showed significantly greater potency against A549 cells than 3 (2: 35 ± 7%; 3: 49 ± 9%), unlike their similar activity against GSCs.

Figure 4.

(A) Structures of lead compounds (2 and 3) and analogs 5a–l. (B) Cell viability profile in A549 cells treated with the compounds. A549 cells were treated with DMSO or the compounds (500 nM) for 3 days, and a WST-8 assay was performed (n = 9). (C) Cell proliferation assessed by counting absolute number of living A549 cells treated with DMSO or the indicated concentrations of the compounds for 3 days (n = 6).

*p < 0.05 and ***p < 0.001 by Dunnett’s multiple comparison test.

Initially, the effect of the number of ethylene glycol units was investigated. The diethylene glycol analog 5a showed no activity, whereas each of the three tetraethylene glycol analogs (5b, 5d and 5e) exhibited greater potency than 3, suggesting that the increased number of ethylene glycol units contributed to the enhanced activity. There was no significant difference in activity between 5b and 5d, or 5b and 5e, whereas 5e displayed significantly more potent activity than 5d (5b: 31 ± 5%; 5d: 38 ± 9%; 5e: 27 ± 6%), indicating that the length of the tail alkyl chain might be more crucial than that of the linker alkyl chain for activity. Analog 5c significantly inhibited cell proliferation with potent activity, whereas the activity of 5f, featuring a shorter linker-alkyl chain than 5c to reduce molecular weight, was weaker than that of 5c (5c: 26 ± 6%; 5f: 51 ± 8%). Yao and Zhou, along with their colleagues, synthesized AA101, which comprises an ethylene glycol unit at the molecule’s center and lactone at both ends. They reported that the incorporation of additional ethylene glycol units into AA101 slightly enhanced its anti-proliferative activity against human cancer cell lines [37]. On the other hand, comparison of our analogs with equivalent tail and linker alkyl chain lengths indicates a discernible positive correlation between activity and the number of ethylene glycol units (5a: 86 ± 6%; 3: 49 ± 9%; 5b: 31 ± 5%; 5c: 26 ± 6%). Based on these findings, we conclude that the optimal number of ethylene glycol units for our analogs is four. This conclusion is rooted in the observation that while the activity of 5c surpassed that of 5b, the disparity was minimal, prompting a preference for a smaller molecular weight.

Second, we focused on the effects of heterocycles. Analogs with the THF moiety, including 2, have been reported to exhibit activity regardless of the regioisomers of the heterocycle or different heterocycles other than the 1-methylpyrazole or 3-thiophene rings [15,24,28]. Thiophene-2-carboxamide analog 5g, a regioisomer of 3, showed similar activity to 3 (3: 49 ± 9%; 5g: 48 ± 16%), as observed in previous results [24]. Although the pyrimidine analog 5i exhibited weaker activity than 3, unlike previous reports (3: 49 ± 9%; 5i: 79 ± 8%) [15,28], both analogs 5h and 5m with the 1-methylpyrazole moiety reduced cell growth with an effect similar to that of 3 (5h: 37 ± 9%; 5m: 46 ± 14%). These SAR results indicate that the substitution of heterocycles generated new potent analogs. There was almost no change in activity due to differences in the tail length of 1-methylpyrazoles (5h and 5m); however, when considering the results for the tetraethylene glycol analogs (5b and 5d) or the lead compounds (2 and 3) [16,27], it was suggested that a certain length of the tail alkyl chain is required for activity.

Third, we focused on the effect of the connecting group between the heterocycle and the linker moiety. Although sulfonamide analog 4 with the THF moiety exhibited potent growth inhibitory activity similar to 1 [23], the sulfonamide analogs 5j–l showed weaker inhibitory activities than the corresponding carboxamide analogs (3 vs 5j, 5g vs 5k, 5h vs 5l). Unexpectedly, the thiophene sulfonamide analogs (5j, 5k and 5l) did not show strong activity, the reason for which remains to be determined. Based on these in vitro studies, modifying the number of ethylene glycol units was more effective than altering the heterocycle and connecting groups in enhancing the activity of analog 3.

Fourth, comprehensive tests were conducted to measure the cell proliferation inhibitory effects of two analogs (5c and 5e) showing potent activity, alongside analog 3, which was used as a benchmark (Figure 4C). The findings indicated that each analog reduced the growth of A549 cells in a manner proportional to its concentration. The GI₅₀ value of 5e was lower than that of 3 (5e, 0.234 μM; 3, 0.516 μM), suggesting its superior efficacy. Although 5c showed a lower GI₅₀ than 5e (5c, 0.122 μM), analog 5e was concluded to be a more effective option considering its molecular size.

Finally, we used human IMR90 lung fibroblasts to assess the tumor cell-specific effect of 5e. Treatment with 5e for 3 days, similar to that in A549 cells, demonstrated relatively limited efficacy, yielding a GI₅₀ value of 3.13 μM, which was higher than that observed in A549 cells (Figure 5A). These findings indicate the preferential activity of 5e against malignant cells.

Figure 5.

(A) Cell proliferation assessed by determining the absolute number of living IMR90 normal cells treated with DMSO or the indicated concentrations of the compounds for 3 days (n = 3). ***p < 0.001 by Dunnett's multiple comparison test. (B & C) Analogs induce phosphorylation of AMPKα proteins. (B) Expression levels of p-AMPKα proteins in A549 cells treated with 500 nM 2, 3, 5a, 5e, or 5h for 72 h. (C) Expression levels of p-AMPKα proteins in A549 cells treated with 0.1, 0.25, 0.5, or 1 μM 5e for 72 h. Protein expression was evaluated by western blotting. (D & E) Tumor growth and body weight changes in nude mice bearing a human lung cancer A549 xenograft. A549 cells were implanted into CB17 SCID mice, and DMSO or 5e was intraperitoneally administered three-times a week (n = 14 for DMSO, n = 12 for 5e). (D) Tumor size was measured once a week. The graph represents tumor volume expressed as the mean ± SE. *p < 0.05 and ***p < 0.001 by a two-tailed Student’s t-test. (E) Body weight results after four weeks of treatment are shown as a boxplot.

3.3. Effect of synthesized analogs on AMPK activity in A549 cells

We evaluated the phospho-AMPK (p-AMPK) protein levels in A549 cells treated with the compounds. In our previous studies, lead compounds (2 and 3) demonstrated growth inhibitory activities against GSCs by inhibiting mitochondrial function and subsequently increasing the phosphorylation of AMPK [25,27]. First, the activation of AMPK protein by the control analogs (2 and 3) was observed in A549 cells, similar to that in GSCs (Figure 5B). Keçeci and İncesu reported that A549 cells depend on OXPHOS for energy production under aglycemic hypoxic conditions [36]. This is consistent with the results of our lead compounds, since AMPK is activated when energy production is reduced. Analogs with potent growth inhibitory activities (5e and 5h) increased the protein levels of p-AMPK, whereas analog 5a, which lacks growth inhibitory activity, did not activate AMPK. Analog 5e increased p-AMPK protein levels in a concentration-dependent manner (Figure 5C). These results suggest that growth inhibitory activity is correlated with the activation of AMPK in A549 cells. Therefore, our analogs, including analog 3, showed growth-inhibitory activity via AMPK activation in A549 cells, similar to that in GSCs, regardless of structural changes.

3.4. In vivo study of the effect of 5e on A549 cells

The in vivo efficacy of the analogs was assessed in xenograft mice implanted with A549 cells. We chose 3-thiophene carboxamide 5e rather than 5h for this test. This selection was based on concerns that 5h might exhibit acute toxicity similar to that of compound 1, owing to its 1-methylpyrazole carboxamide component. After four weeks of treatment with 5e (15 mg/kg) three-times a week, the average tumor size was reduced to 45% compared with that in the control mice (5e, 94 ± 9 mm³, control, 208 ± 26 mm³), demonstrating a significant inhibition of tumor growth (Figure 5D). Furthermore, the administration of 5e did not induce changes in body weight, suggesting that treatment with 5e at this dose did not cause acute toxicity (Figure 5E). These results indicate the potential of 5e as a novel therapeutic agent for NSCLC.

4. Discussion

According to the present study, increasing the number of ethylene glycol units resulted in more potent activity than substituting the heterocycle and connecting groups. Considering the calculated lipophilicity values (XlogP) [38] (3: 9.34; 5e: 7.57), analog 5e emerged as a more promising antitumor agent than 3 for A549 cells. This was attributed to the lower lipophilicity of 5e, suggesting a potential advantage in avoiding the risk of toxicity. Moreover, analog 5e is anticipated to demonstrate antitumor activity against GSCs and SW48 cells, similar to analog 3, because both analogs have been suggested to inhibit tumor growth by activating AMPK. In addition, our analogs may exhibit antitumor activity against other cancer cells that rely on OXPHOS for energy production.

Recently, numerous researchers have focused on the development of innovative antitumor agents that target mitochondrial function, such as metformin, as discussed in the Introduction. Notably, darinaparsin was approved as a chemotherapeutic agent for relapsed or refractory peripheral T-cell lymphoma in 2022 [39]. Additionally, ME-344 [40], a mitochondrial complex I inhibitor, is currently undergoing phase Ib clinical trials for previously treated metastatic colorectal cancer in combination with bevacizumab (NCT05824559). This underscores the potential of OXPHOS-targeting compounds as chemotherapeutic agents against not only blood cancers but also potentially solid cancers, suggesting that our analogs may represent novel therapeutic agents for solid cancers, including NSCLC.

5. Conclusion

We synthesized 13 analogs of lead compound 3 by varying the number of ethylene glycol units, heterocycles and connecting groups. SAR studies indicated that the ethylene glycol unit moiety contributed significantly to the enhancement of anti-proliferative activity compared with the other two moieties, implying a preference for four ethylene glycol units. The current analogs activated AMPK, similar to 3, signifying a shared mode of action for inhibiting tumor growth. Notably, analog 5e, with four ethylene glycol units, demonstrated antitumor activity against A549 cells without acute toxicity in vivo, suggesting its potential as a novel therapeutic agent for NSCLC.

Funding Statement

This study was supported in part by JSPS KAKENHI (grant numbers 16K08330 and 24K09742 [N Kojima]), MEXT-Supported Program for the Strategic Research Foundation at Private Universities, 2015–2019 (grant number S1511024L [N Kojima and S Nakata]) and the Kyoto Pharmaceutical University Fund for the Promotion of Scientific Research (N Kojima).

Supplemental material

Supplemental data for this article can be accessed at https://doi.org/10.1080/17568919.2024.2394016

Author contributions

K Ohta, S Nakata and N Kojima conceived the idea and designed the experiments. K Ohta and M Takahashi conducted the chemical experiments (under the guidance of H Nambu and N Kojima). H Ii, C Moyama, S Ando, M Masuda and M Mori conducted the biological experiments (under the guidance of S Nakata). K Ohta, H Ii, S Nakata and N Kojima analyzed the data. K Ohta, H Ii, S Nakata and N Kojima wrote the original draft. K Ohta, H Ii, H Nambu, S Nakata and N Kojima reviewed and edited the paper. S Nakata and N Kojima administrated and supervised this study. All the authors approved the final version of the manuscript.

Financial disclosure

This study was supported in part by JSPS KAKENHI (grant numbers 16K08330 and 24K09742 [N Kojima]), MEXT-Supported Program for the Strategic Research Foundation at Private Universities, 2015–2019 (grant number S1511024L [N Kojima and S Nakata]) and the Kyoto Pharmaceutical University Fund for the Promotion of Scientific Research (N Kojima). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.