Facile Synthesis of a Cholesterol–Doxorubicin Conjugate Using Cholesteryl-4-nitrophenolate as an Activated Ester and Biological Property Analysis

Abstract

Developing new biomolecule–drug conjugates as prodrugs is a promising area for natural products and pharmaceutical chemistry. Herein, a cholesterol–doxorubicin (Chol-DOX) conjugate was synthesized using cholesteryl-4-nitrophenolate as a facile, stable, and controllable activated ester. This approach offers an alternative to the conventional HCl-emitting cholesteryl chloroformate method. Semi-empirical theoretical calculations showed that cholesteryl-4-nitrophenolate exhibits moderate reactivity, greater thermodynamic stability, a higher dipole moment, and a lower HOMO-LUMO energy gap compared to cholesteryl chloroformate, suggesting that cholesteryl-4-nitrophenolate could be used as a more controllable acylating agent. The structure of the synthesized Chol-DOX conjugate was characterized using NMR, MS, and FT-IR techniques. Biological properties of the Chol-DOX conjugate were analyzed with a comparison of theoretical and experimental data. This work provides a facile and controllable method to synthesize natural lipid–DOX prodrugs and offers an in-depth data analysis of the related biological properties.

1. Introduction

Doxorubicin (DOX, also known as Adriamycin) is an anthracycline with an aglycone linked to an amino-sugar moiety (daunosamine), which has been employed as a chemotherapeutic drug for the treatment of solid tumors and malignancies [1]. DOX inhibits topoisomerase II (Topo II) by DNA intercalation and by reactive oxidative species (ROS)-associated apoptosis. Free DOX is highly toxic, and its pharmacodynamics are challenging to control. To mitigate toxic side effects of DOX, conjugating natural products to the daunosamine moiety has proven to be an effective strategy, improving the selectivity and enhancing cellular uptake of drugs [2]. Particularly, the conjugation of DOX with biocompatible natural lipids could bring about enhanced cellular uptake and controllable drug release properties.

As a natural steroid lipid, cholesterol plays essential roles as a precursor to all steroid hormones and bile acids with outstanding bioavailability and biocompatibility, which allow it to serve as a potential sustainable building block for biomaterials [3,4,5,6,7,8,9]. The conjugation of anti-cancer drugs with cholesterol endowed them with controllable, quantitative, and enhanced drug delivery features for constructing nano-chemotherapeutics. In our previous work, a cholesterol–doxorubicin conjugate-based nanoprodrug with enhanced breast cancer inhibition was synthesized [10] using cholesteryl chloroformate as an acylating agent. Nevertheless, cholesteryl chloroformate is a highly reactive agent that could result in drawbacks, such as low storage stability, lack of selectivity, and high toxicity. Particularly, it could emit very toxic HCl gas in a moist environment through hydrolysis. The drawbacks prompted us to develop another activated ester, cholesteryl-4-nitrophenolate, as an alternative acylating reagent with better stability and controllable reactivity. Moreover, cholesteryl-4-nitrophenolate, as a chromophore, could react with primary amino group-containing molecules via amidation (or aminolysis) reaction and release another chromophore, 4-nitrophenol, with a stronger intramolecular charge transfer (ICT) effect, thus resulting in a red-shift of the absorption wavelength. This offers a possible way to monitor the amidation/aminolysis reaction of cholesteryl-4-nitrophenolate by thin-layer chromatography (TLC) or UV-vis spectrometry.

In this work, the molecular parameters, electrostatic potential, and local electron density of carbonyl groups of cholesteryl chloroformate and cholesteryl-4-nitrophenolate were first theoretically calculated by a semi-empirical method. In our experimental approach, the unstable, HCl-emitting cholesteryl chloroformate was converted to the more stable cholesteryl-4-nitrophenolate, which was then coupled with free DOX in the presence of an organic base, triethylamine (TEA), to facilely synthesize the Chol-DOX conjugate. The molecular structures of both the synthesized cholesteryl-4-nitrophenolate intermediate and the final product, Chol-DOX, were characterized. In addition, the theoretical and experimental biological properties of the Chol-DOX conjugate were analyzed and discussed.

2. Materials and Methods

2.1. Theoretical Molecular Parameter and Molecular Electrostatic Potential Analysis

The molecular parameters and molecular electrostatic potential (MEP) of cholesteryl chloroformate and cholesteryl-4-nitrophenolate were calculated by WebMO online software (version 24.0.018e) [11], using the semi-empirical Molecular Orbital PACkage (MOPAC, v22.1.1 Linux) and the Parameterization Method (PM) [12].

2.2. Chemical Synthesis Experimental Method

2.2.1. Materials and Methods for the Synthesis

Cholesteryl chloroformate (98.0%), 4-nitrophenol (98.0%), doxorubicin hydrochloride (DOX·HCl; 99.0%), and triethylamine (TEA; 99.0%) were purchased from Sigma-Aldrich and used as received. Other reagents and solvents were of analytical grade and used without further purification. Aluminum TLC Silica Gel 60 F254 Plates were purchased from Merck (Germany). ¹H NMR and ¹³C NMR values were measured on a Bruker Avance II+ UltraShield^TM 400 Plus Ultra Long Hold NMR spectrometer (Wissembourg, France, working at room temperature, ¹H nuclei at 400 MHz and ¹³C at 101 MHz). The mass spectrum with electrospray ionization technique (ESI-MS) was conducted on a Bruker Autoflex maX MALDI-TOF-Mass Spectrometer (Bremen, Germany, working in negative ion mode, with the 2,5-dihydroxybenzoic acid as the matrix) at the Centro de Quimica da Madeira. Fourier transform infrared spectroscopy (FT-IR) was carried out with a PerkinElmer Spectrum Two spectrometer apparatus (UK).

2.2.2. Synthesis of Cholesteryl-4-nitrophenolate (IUPAC Name: [(3S,10R,13R,14S,17R)-10,13-Dimethyl-17-(6-methylheptan-2-yl)-2,3,4,7,8,9,11,12,14,15,16,17-dodecahydro-1H-cyclopenta[a]-phenanthren-3-yl] (4-Nitrophenyl) Carbonate)

Cholesteryl chloroformate (45 mg, 0.1 mmol) was dissolved in 5 mL tetrahydrofuran (THF) with TEA (15 mg, 0.15 mmol) and put inside a 50 mL flask; then, 4-nitrophenol (17 mg, 0.12 mmol) was dissolved in 1 mL THF and the solution was added dropwise into the cholesteryl chloroformate solution, after which the mixture was stirred at room temperature (r.t.) for 6 h. The solvent, THF, was removed by rotavaporation under reduced pressure, after which 5 mL of saturated citric acid solution was added to neutralize the TEA and then extracted with 20 mL CHCl₃. The crude product was further purified by silica gel flash column chromatography (ethyl acetate: hexane = 1:5, v:v) to obtain the cholesteryl-4-nitrophenolate (39.3 mg, yield: 71.2%).

• ¹H NMR (CDCl₃, 400 MHz, δ in ppm) δ 8.26 (2H, Ar-H, ²J_ortho = 12.0 Hz, nitrophenolate), 7.40 (2H, Ar-H, ²J_ortho = 12.0 Hz, nitrophenolate), 5.42 (1H, -CH=, Chol), 4.60 (1H, COCH), 2.50 (2H, -CH₂-C=, Chol) 2.30–0.79 (38 H, cholesterol skeleton), 0.68 (3H, -CH₃, Chol).
• ¹³C NMR (CDCl₃, 101 MHz, δ in ppm) 155.79, 151.91, 145.42, 138.98, 125.41, 123.67, 121.93, 79.90, 56.81, 56.27, 50.11, 42.46, 39.84, 39.65, 38.00, 36.93, 36.69, 36.32, 35.93, 32.05, 31.97, 28.36, 28.16, 27.73, 24.42, 23.97, 22.96, 22.71, 21.19, 19.41, 18.86, 12.01.
• FT-IR (ν cm⁻¹) 2936.32 (stretching of aromatic C-H), 1756.13 (stretching of ester C=O), 1519.33 (stretching of aromatic C=C), 1491.74 (bending of aromatic C-H), 1350.03 (bending of C-H), 1257.45, 1230.17, 1165.52, 1110.07 (C-O-C stretching), 972.35, 849.06, 745.34, 681.91 (C-H out-of-plane bending).

2.2.3. Synthesis of Chol-DOX Conjugate (IUPAC Name: [(3S,10R,13R,17R)-10,13-Dimethyl-17-(6-methylheptan-2-yl)-2,3,4,7,8,9,11,12,14,15,16,17-dodecahydro-1H-cyclopenta[a]phenanthren-3-yl] N-[(2S,3S,4S,6R)-3-Hydroxy-2-methyl-6-[[(1S,3S)-3,5,12-trihydroxy-3-(2-hydroxyacetyl)-10-methoxy-6,11-dioxo-2,4-dihydro-1H-tetracen-1-yl]oxy]oxan-4-yl]carbamate)

A quantity of 24.4 mg of DOX·HCl (0.04 mol) was weighed and mixed with 0.5 mL of dimethyl sulfoxide (DMSO), 0.25 mL of methanol (MeOH), and 0.25 mL of TEA. Cholesteryl-4-nitrophenolate (38.8 mg, 0.07 mol) was dissolved in 0.5 mL of DMSO and 0.5 mL of THF and added dropwise into the DOX solution. Then, the reaction mixture was stirred at 40 °C for 12~16 h. The reaction was monitored using thin-layer chromatography (TLC) silica plates, and when the reaction was completed, the solvent, THF, was removed by rotavaporation at 40 °C under reduced pressure, after which 5 mL of saturated citric acid solution was added to neutralize the TEA and then extracted with 20 mL CHCl₃; the crude product was further purified by silica gel flash column chromatography (MeOH: CHCl₃=1:10, v:v). The eluent was removed under reduced pressure to obtain the conjugate Chol-DOX as a dark-red powder (26.1 mg, yield: 68.3%).

• ¹H NMR (CDCl₃, 400 MHz, δ in ppm): δ 13.97 (1H, Ar-OH, DOX), 13.24 (1H, Ar-OH, DOX), 8.02 (1H, Ar-H, DOX), 7.78 (1H, Ar-H, DOX), 7.28 (1H, Ar-H, DOX), 5.50 (1H, CH₂-O(H), DOX), 5.29 (1H, -CHO(O)-, DOX) and 1H, -CH=, DOX), 5.11 (1H, Ar-C(O)HCH), 4.78 (2H, -CH₂CO, DOX), 4.58 (1H, OCOO-CH, Chol), 4.10 (3H, -OCH₃, DOX), 3.81 (1H, -CH-NH, DOX), 3.68 (1H, -CH-O-, DOX), 3.18 and 3.06 (2H, Ar-CH₂ (H_a and H_b), Dox), 2.23–0.74 (51 H, Chol and Dox skeletons), 0.63 (3H, -CH₃, Chol).
• ¹³C NMR (CDCl₃, 101 MHz, δ in ppm): δ 214.6, 187.3, 186.76, 161.22, 156.14, 155.72, 155.46, 136.09, 135.61, 135.17, 133.74, 133.56, 122.63, 120.81, 120.03, 118.62, 111.78, 111.42, 69.86, 65.73, 62.29, 56.83, 56.27, 50.14, 42.45, 39.87, 39.65, 38.74, 38.57, 37.11, 37.06, 36.69, 36.64, 36.32, 35.93, 35.76, 34.01, 28.37, 28.15, 24.42, 23.97, 22.96, 22.70, 21.18, 19.46, 18.85, 17.03, 12.00.
• MALDI-TOF-MS (negative ion mode): [C₅₅H₇₃NO₁₃-H⁺]⁻ calculated: 954.50, found: 954.649.
• FT-IR (ν cm⁻¹) 3620~3230 (broad peak, stretching of -OH and -NH), 2935.36 (stretching of aromatic C-H), 1759.12 (stretching of ester C=O), 1614.52, 1578.20, 1518.82 (stretching of aromatic C=C), 1410.78 (bending of aromatic C-H), 1350. 29 (bending of C-H), 1258.31, 1207.10, 1111.54, 1069.62 (C-O-C stretching), 1009.80, 983.15, 765.01 (C-H out-of-plane bending).

3. Results and Discussion

3.1. Theoretical Analysis of Molecular Parameters

For nucleophilic reactions, the acyl group’s activity was determined by the electron density of the sp²-carbon atom of the active carbonyl; due to a higher positive charge (electropositivity), it tended to readily react with electron-rich (nucleophilic) groups (such as NH₂ and OH). Herein, the molecular parameters and molecular electrostatic potential (MEP) were calculated by the MOPAC (v22.1.1 Linux) with PM6 semi-empirical method using WebMO online software [11]. The electron density and charge distribution could be visually presented on the MEP surfaces of the molecules. Various colors represented the different electrostatic potentials of atoms (Figure 1); cholesteryl-4-nitrophenolate had a much lower local positive charge on the carbonyl group (ρ_COPh = +0.7277 − 0.3332 = 0.3945) than that of cholesteryl chloroformate (ρ_COCl = +0.2493 + 0.5256 = +0.8049), suggesting its moderate and controllable reactivity [13]. Moreover, the positive charge of cholesteryl chloroformate localized on both sp²-carbon (+0.2493) and chloride (+0.5256) atoms, while the cholesteryl-4-nitrophenolate’s phenol oxygen atom had a negatively charged oxygen atom (−0.3332) and the positive charge only localized on the sp²-carbon (+0.7277), suggesting its less nucleophilic reactivity compared to that of cholesteryl chloroformate. The controllable reactivity of cholesteryl-4-nitrophenolate could bring about higher storage stability under moisture, easier handling, and fewer side reactions, which might improve its applicability in cholesteryl lipid conjugation and biomaterial modification.

Theoretical molecular parameters of the cholesteryl chloroformate and cholesteryl-4-nitrophenolate (under non-solvent conditions) were also calculated using WebMO online software for the optimized structures; the data are shown in

Table 1. Comparison of the theoretical molecular properties of cholesteryl chloroformate and cholesteryl-4-nitrophenolate (hybridized N atom) under non-solvent conditions calculated by WebMO software (* Calculated by Hyperchem 8.0 software).

Molecular Parameters	Cholesteryl Chloroformate	Cholesteryl-4-nitrophenolate
Molecular formula	C28H45O2Cl	C34H49NO5
Molecular weight (MW)	449.115	551.765
Heat of formation (Kcal/mol)	98.164	469.093
Molecular area (Å2) *	465.440	576.180
Molecular volume (Å2) *	595.540	713.880
LUMO energy (ELUMO, eV)	+0.088	−12.812
HOMO energy (EHOMO, eV)	−9.000	−13.946
HOMO-LUMO energy gap (ΔE= ELUMO-EHOMO, eV)	9.088	0.134
Dipole moment (Debye)	2.855	10.721

. Cholesteryl-4-nitrophenolate had a significantly higher heat of formation (469.093 Kcal/mol) compared to cholesteryl chloroformate (98.164 Kcal/mol), suggesting its greater thermodynamic stability. The large HOMO-LUMO gap for cholesteryl chloroformate (9.000 eV) indicated its low light absorption capability; in contrast, the smaller HOMO-LUMO gap of cholesteryl-4-nitrophenolate (0.134 eV) suggested its light absorption property, which is attributed to its aromatic 4-nitrophenolate structure. Moreover, cholesteryl-4-nitrophenolate had a much higher dipole moment (10.721 Debye), which indicated its stronger polarity, and could easily interact with amphiphilic molecules, particularly under polar organic solvent conditions (Table S1) [14].

3.2. Synthesis and Characterization of the Chol-DOX Conjugate

First, cholesteryl chloroformate was reacted with 4-nitrophenol to prepare cholesteryl-4-nitrophenolate using TEA as a catalyst [15,16], which was also a proton scavenger used to absorb HCl released from the chloroformate. The reaction proceeded at room temperature for 6 h (Figure 2). Then, the solvent was removed, and the TEA residue was neutralized with citric acid; the product was purified via silica gel flash column chromatography (ethyl acetate: hexane= 1:5, v:v) with a yield of 71.2%. The as-synthesized cholesteryl-4-nitrophenolate further reacted with the primary amino group of DOX via amidation (or aminolysis) reaction to prepare the Chol-DOX conjugate. The reaction can be easily monitored by a silica gel 60 F₂₅₄ aluminum TLC plate or a UV-vis spectrometer. For the facile and efficient synthesis of Chol-DOX conjugates, it is essential to use an organic base to catalyze the amide bond formation (amidation) under mild conditions. TEA was employed as a catalyst and proton scavenger in the reaction, as it can convert the hydrochloride form of DOX into neutralized free DOX. Pyridine (PY) is widely used as a catalyst and/or solvent [17], while the PY-catalyzed amidation proceeded at a slower rate and required complex purification steps, which might have been due to the hydrogen-bonding or π-π stacking interactions between pyridine and DOX and resulted in the formation of some complexes. Compared to PY, TEA is a more efficient basic catalyst. Moreover, excess TEA catalyst in the reaction mixture was much easier to remove under vacuum distillation (boiling point: ~40 °C at 0.15 atm or 151.98 mbar) [18,19]. After purification by silica gel flash column chromatography (MeOH: CHCl₃ = 1:10, v:v), the synthesized Chol-DOX conjugate was obtained as a dark-red solid powder with a yield of 68.3%.

The ¹H and ¹³C NMR spectra of the cholesteryl-4-nitrophenolate activated ester are shown in Figure S1 (Supplementary Materials). The aromatic proton signals of 4-nitrophenolate were found at δ 8.26 ppm and δ 7.40 ppm, with a strong ortho H-H coupling constant (²J_ortho = 12.0 Hz). The signals at δ 5.42 ppm and δ 4.60 ppm were identified as the double-bond (alkenyl) proton (-C=CH-) of the cholesteryl moiety and the proton attached to the carbonate (-OCOO-) group, respectively. The signals at 2.50 ppm and 2.01 ppm were identified as the methylene (-C=C-CH₂) protons adjacent to the cholesterol’s alkenyl group (Figure S1a). In the ¹³C NMR analysis, the chemical shift of the carbon atom on the carbonyl group (-C=O) was observed at δ 155.79 ppm. The carbon signals of the 4-nitrophenolate ring were observed at δ 151.9, 145.42, 125.41, and 121.93 ppm. The alkenyl carbon atoms (-C=CH-) of the cholesteryl moiety were found at δ 138.98 and 123.67 ppm. The signal at δ 79.90 ppm was identified as the carbon atom on the cholesteryl moiety attached to the oxygen of the carbonate (-OCOO-CH) group (Figure S1b). The structure was confirmed by FT-IR spectrometry (Figure S2): the peak at 2936.32 cm⁻¹ was identified as the stretching of aromatic C-H; that at 1756.13 cm⁻¹ as the stretching of ester C=O; that at 1519.33 cm⁻¹ as the stretching of aromatic C=C; that at 1491.74 cm⁻¹ as the bending of aromatic C-H; that at 1350.03 cm⁻¹ as the bending of C-H; and those at 1257.45, 1230.17, 1165.52, and 1110.07 cm⁻¹ as C-O-C stretching.

The ¹H and ¹³C NMR spectra of the Chol-DOX conjugate are shown in Figure 3. For the ¹H NMR spectrum (Figure 3a), the proton signals in the range of δ 7.28–7.41 ppm were identified as the aromatic protons (-ArH). The signals at δ 13.97 and 13.21 ppm were attributed to hydroxyl groups (-OH), and that at δ 4.10 ppm belonged to the methoxy group (-OCH₃) on the DOX. The alkenyl proton (-C=CH-) appeared at δ 5.28 ppm. Moreover, aliphatic proton signals (δ 0.63 to 2.41 ppm) on the cholesterol skeleton could be clearly observed. For the ¹³C NMR spectrum (Figure 3b), δ 214.6, 187.3, and 186.7 ppm were assigned as signals of carbonyl carbon atoms. The carbon signals of δ 136.09–111.42 ppm were identified as the anthracycline aromatic carbons of the DOX, the glycoside ring carbon signals were located at about δ 69.86–56.27 ppm, and the cholesterol skeleton exhibited various carbon signals in the range of δ 69.86 ppm to 12.00 ppm. Moreover, the MALDI-TOF-MS spectrum (Figure 3c) showed a negative ion peak at m/z 954.649, which was very close to the calculated m/z 954.50 for the deprotonated Chol-DOX conjugate ([C₅₅H₇₃NO₁₃-H⁺]⁻ or [Chol-DOX-H⁺]⁻). For the FT-IR spectrum of Chol-DOX (Figure S3), the broad peak at 3620~3230 cm⁻¹ was identified as the stretching of -OH and -NH; that at 2935.36 cm⁻¹ as the stretching of aromatic C-H; that at 1759.12 cm⁻¹ as the stretching of ester C=O; those at 1614.52, 1578.20, and 1518.82 as the stretching of aromatic C=C; and that at 1410.78 as the bending of aromatic C-H. The ¹H NMR, ¹³C NMR, MALDI-TOF-MS, and FT-IR results confirmed the successful preparation of the Chol-DOX conjugate.

The synthesized Chol-DOX conjugate appeared as a dark-red solid powder (Figure S4a, Supplementary Materials); the solid state indicated that it could be conveniently handled and stored. When the Chol-DOX conjugate solid powder was added into PBS (0.1 M) solution, precipitates of Chol-DOX could be clearly observed, indicating that Chol-DOX has poor solubility (is almost insoluble) in water or an aqueous environment (Figure S4b, Supplementary Materials). The poor solubility was due to the conjugation of a strong hydrophobic cholesteryl moiety to the amphiphilic DOX. Thin-layer chromatography (TLC) plate analysis of the Chol-DOX conjugate showed a high R_f value (0.85) and low polarity (Figure S2c), which further suggested that the introduction of a cholesteryl moiety could significantly reduce the polarity of DOX. The low polarity of Chol-DOX (calculated LogP value: 8.57) could benefit higher cell membrane permeability and enhance cellular uptake efficiency, which might change its pharmacokinetics and intracellular behaviors.

3.3. Theoretical and Experimental Biological Properties of the Chol-DOX Conjugate

In the above study, we successfully synthesized a Cholesteryl–Doxorubicin conjugate using cholesteryl-4-nitrophenolate as a controlled and stable activating ester. To access the potential pharmaceutical and biomedical applications of the conjugate, herein, theoretical (predicted on the SwissADME software (version: SwissDrugDesign) [20,21]) and experimental data analyses were conducted to explore the biological properties of the Chol-DOX conjugate using free DOX as the control. As shown in

Table 2. Theoretical and experimental biological properties (pharmacokinetics, druglikeness, cell inhibition efficacy, and localization) of free DOX and the Chol-DOX conjugate.

Biological Properties	Free DOX	Chol-DOX
GI absorption a	Low	Low
BBB permeant a	No	No
P-gp substrate a	Yes	Yes
CYP1A2 inhibitor a	No	No
CYP2C19 inhibitor a	No	No
CYP2C9 inhibitor a	No	No
CYP2D6 inhibitor a	No	No
CYP3A4 inhibitor a	Yes	Yes
Log Kp (skin permeation, cm/s) a	−8.71	−4.37
Log Po/w (iLOGP) a	2.58	5.80
Lipinski druglikeness a	No (3 violations): MW > 500; number of N or O atoms >10; NH or OH groups >5)	No (3 violations): MW > 500; number of N or O atoms >10; NH or OH groups >5
Bioavailability score a	0.17	0.17
MDA-MB-231 cell viability (%) with 2, 6, and 10ug/mL DOX b	87.1, 74.3, 70.2	98.9, 94.3, 82.1
MCF-7 cell viability (%) with 2, 6, and 10ug/mL DOX b	67.3, 64.1, 60.7	95.9, 79.4, 62.4
Intracellular localization b	Cell nuclei	Lysosomes

^a The theoretical data were calculated in the SwissADME online software. ^b The experimental/tested data are cited from our previous research work [10].

, both free DOX and Chol-DOX exhibited low gastrointestinal (GI) absorption and blood–brain barrier (BBB) penetration capability due to their potential to serve as substrates for P-glycoprotein (P-gp). Both Chol-DOX and free DOX have no inhibitory effects on the cytochrome P450 (CYP) enzymes CYP1A2, CYP2C19, and CYP2D6, which could reduce their CYP-catalyzed drug degradation. The inhibition of CYP3A4 might increase their plasma levels. Notably, Chol-DOX demonstrated a significantly improved octane–water partition coefficient (Log P_o/w = 5.80) compared to free DOX (Log P_o/w = 2.58), implying its higher hydrophobicity (or lipophilicity) [22]. This property was beneficial to improve the cell membrane permeability and to increase the cellular uptake efficiency of Chol-DOX.

The experimental data revealed that the Chol-DOX conjugate had obviously lower breast cancer cytotoxicity in MDA-MB-231 (98.9%, 94.3%, and 82.1% cell viability for 2, 6, and 10 µg/mL, respectively) and MCF-7 cell lines (95.9%, 79.4%, and 62.4% for the same concentrations) compared to that of free DOX (87.1%, 74.3%, and 70.2% cell viability for MDA-MB-231 and 67.3%, 64.1%, and 60.7% for MCF-7 cells), which might be due to the ‘endosome–lysosome–autophagosome encapsulation process’ of Chol-DOX and the slow acidic hydrolysis of its carbonate linkage. Interestingly, the intracellular localization of Chol-DOX in lysosomes instead of cell nuclei (free DOX mainly localized in cell nuclei after 12 h incubation) implied a more controllable drug delivery mechanism of Chol-DOX [10]. The bioavailability radar plots (Figure 4) of free DOX and Chol-DOX (predicted by SwissADME software [23]) also demonstrated their significant difference. This profile suggested that Chol-DOX has more limitations in terms of absorption, distribution, metabolism, and excretion properties compared to free DOX; it is suitable to serve as a prodrug rather than a ‘direct’ drug. In addition, the biological properties indicated that the Chol-DOX conjugate might be employed as a more controllable alternative to free DOX. Further study on optimizing Chol-DOX’s bioavailability, anti-cancer efficacy, cellular uptake, and drug metabolism, as well as its intracellular localization and transportation, is ongoing in our lab.

4. Conclusions

In summary, we synthesized a cholesterol–doxorubicin conjugate using a stable cholesteryl-4-nitrophenolate ester, which could be employed as a more controllable and safer alternative to the unstable, HCl-emitting cholesteryl chloroformate. The theoretical calculations indicated that cholesteryl-4-nitrophenolate exhibited moderate reactivity, greater thermodynamic stability, a higher dipole moment, and a smaller HOMO-LUMO gap compared to cholesteryl chloroformate. The synthesized Chol-DOX conjugate was fully characterized using NMR, MS, and FT-IR techniques. The synthesized Chol-DOX conjugate demonstrated high lipophilicity and low water solubility, which may improve its cell membrane permeability and enhance therapeutic efficiency. This work provided a controllable method to synthesize Chol-DOX and offered an in-depth analysis of its biological properties.