English

• Chemotherapy is deemed to be an efficacious modality for clinical cancer therapy [1-5]. Paclitaxel (PTX), one of the most potent natural anticancer drugs, has become the mainstay of cancer chemotherapy regimens. The commercial injection, Taxol, has to incorporate the allergenic excipients Cremophor EL and ethanol to solve the water-solubility of PTX. However, poor tolerance, severe side effects and unsatisfactory pharmacokinetics greatly compromise the therapeutic efficacy of Taxol. Nanomedicines have been shown to improve solubility, extend circulation of blood and enhance tumor accumulation of chemotherapeutics. The most representative PTX nano-formulation, Abraxane, not only solves the water solubility problem of PTX but also increases the tolerated dose of Taxol by about 0.5 times [6, 7]. However, lower drug loading, immunogenicity of exogenous albumin, and similar pharmacokinetics to Taxol are still drawbacks of Abraxane. Engineering the innovative PTX nano delivery system to address these dilemmas remains challenging [8, 9].

  The emergence of self-assembling prodrug nano delivery systems provides an attractive strategy to overcome the above obstacles [10-14]. The prodrugs are capable of self-assembling to form nanostructure after rational design [15-19]. Moreover, prodrug nanoassemblies have ultra-high drug loading owing to the prodrugs acting as both cargoes and carriers [20, 21]. In addition, the preparation process of prodrug nanoassemblies is simple and reproducible, which facilitates industrialized scale-up production [22]. However, the rational design of prodrugs is still a concern. For example, paclitaxel-docosahexaenoic acid (PTX-DHA) prodrug significantly reduces the toxic side effects of PTX, increases the maximum tolerated dose by 4 times, and has successfully entered clinical phase III studies [23, 24]. However, since PTX and DHA are linked by ester bond, PTX may not be released easily, resulting in inferior efficacy. Therefore, the modular design is utilized to divide the prodrug into drug module, response module and modification module [25, 26]. Due to the abnormal proliferation and growth of tumors, tumor cells possess higher redox levels than normal cells, which provides favorable opportunities to design redox-responsive modules [27-29]. Disulfide bond has been extensively used as a classical redox response module, and there are a variety of marketed drugs containing disulfide bonds, which have promising application prospects [30-32].

  Modification modules usually consist of aliphatic or aromatic structure that facilitate the assembly of the prodrug [33, 34]. Paclitaxel, a hydrophobic drug with structural rigidity, is prone to aggregate and precipitate in water. Aliphatic chains can provide steric hindrance to disrupt the ordered arrangement of drug molecules and promote the assembly of prodrugs into stable nanoparticles [35-37]. In general, single-tailed fatty alcohols are widely used as modification modules. Based on the steric-hindrance effect [38, 39], we hypothesized that two-tailed fatty alcohols may provide stronger steric-hindrance, which affects the assembly mechanism, drug release characteristics and therapeutic efficacy of the prodrug nanoassemblies. Therefore, exploring the action mechanism of different types of modification modules is conducive to resolving the structure-activity relationship of the prodrug nanoassemblies.

  Herein, single- and two-tailed modification chains were selected for the construction of PTX prodrug nanoassemblies. The two-tailed modified prodrugs exhibited better size stability as the two-tailed modification module provided higher steric-hindrance. However, the steric-hindrance effect also impeded the attack of redox agents, resulting in a lower redox sensitivity of the two-tailed modified prodrug nanoassemblies. The single-tailed modified prodrug nanoassemblies exhibited higher redox sensitivity, which released more PTX at the tumor site with better antitumor efficacy. The two-tailed modified prodrug nanoassemblies displayed more advantages in terms of pharmacokinetics and tumor accumulation due to better colloidal stability. In addition, the antitumor effectiveness and safety of the two-tailed modified prodrug nanoassemblies were optimized due to the balance of multiple influencing factors (Scheme 1). Therefore, by investigating the steric-hindrance effect of modification modules, this study proposed new perspectives into the rational design of prodrug nanoassemblies.

  Scheme 1

  

  Scheme 1.  Two-tailed modification module tuned steric-hindrance effect enabling high therapeutic efficacy of paclitaxel prodrug nanoassemblies.

  DownLoad: Full-Size Img PowerPoint

  To investigate the potential mechanism of the steric-hindrance effect on the behavior of prodrug nanoassemblies, prodrugs were constructed by linking PTX and modification modules with disulfide bonds. Single-tailed 1-octadecanol and two-tailed 2-heptylundecanol were selected as modification modules, and the prodrugs were named P-LA_C18 and P-BA_C18. The synthetic pathways were illustrated in Fig. S1 (Supporting information). The structures of P-LA_C18 and P-BA_C18 were verified using nuclear magnetic resonance hydrogen spectroscopy and mass spectrum (Figs. S2 and S3 in Supporting information). The purities of P-LA_C18 and P-BA_C18 were all above 99%, which were determined by high performance liquid chromatography (HPLC) (Figs. S2 and S3).

  To explore the self-assembly ability, non-PEGylated P-LA_C18 and P-BA_C18 nanoparticles (NPs) were prepared. The prodrugs could self-assemble to form nanoassemblies at the concentrations of 0.1, 0.2 and 0.4 mg/mL. Among them, the particle size of non-PEGylated P-BA_C18 NPs was smaller (Table S1 in Supporting information). When the concentration increased to 0.6 mg/mL, P-LA_C18 precipitated in deionized water (Fig. 1A, Fig. S4 and Table S1 in Supporting information), while the P-BA_C18 NPs exhibited uniform particle size distribution. In addition, the stability of non-PEGylated P-LA_C18 and P-BA_C18 NPs was investigated under room temperature. From day 8, the particle size of non-PEGylated P-LA_C18 NPs gradually grew, while non-PEGylated P-BA_C18 NPs remained unchanged (Fig. S5 in Supporting information). Therefore, the two-tailed modified prodrug nanoassemblies (P-BA_C18 NPs) displayed superior size stability.

  Figure 1

  

  Figure 1.  Self-assembly capacity and characterization of prodrug nanoassemblies. (A) The appearance of non-PEGylated P-LA_C18 NPs and P-BA_C18 NPs at concentrations of 0.1, 0.2, 0.4 and 0.6 mg/mL. (B) The variation of the particle size of non-PEGylated prodrug nanoassemblies after co-incubated with NaCl, SDS or Urea. (C) Steric-hindrance effects of P-LA_C18 and P-BA_C18. (D) Intermolecular interactions of P-LA_C18 NPs and P-BA_C18 NPs during the self-assembly process. (E) LogP values of two prodrugs. (F) The particle size and (G) morphology of P-LA_C18 NPs and P-BA_C18 NPs. Scale bar: 200 nm. Storage stability of P-LA_C18 NPs and P-BA_C18 NPs (H) at 4 ℃ and (I) at 25 ℃. (J) The changes in particle size of P-LA_C18 NPs and P-BA_C18 NPs after coincubation with PBS containing FBS for 12 h. Data are presented as mean ± SD (n = 3). P < 0.05 by two-tailed Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  The self-assembly mechanism of the prodrugs was studied to explore the factors contributing to the enhanced assembly capacity of the two-tailed P-BA_C18 NPs. The LogP values of P-LA_C18 and P-BA_C18 were 12.05 and 12.35, suggesting that the hydrophobic interactions of P-BA_C18 might be stronger than P-LA_C18 (Fig. 1E). In addition, non-PEGylated prodrug nanoassemblies were co-incubated with NaCl (shielding of electrostatic interactions), sodium dodecyl sulfate (SDS, shielding of hydrophobic interaction) and urea (shielding of hydrogen bonds) to validate the intermolecular forces. As shown in Fig. 1B, the particle size of the non-PEGylated prodrug nanoassemblies was significantly increased, indicating that three forces were involved in the assembly of the prodrugs. Notably, the particle size of non-PEGylated P-BA_C18 NPs was changed remarkably, suggesting that the assembly process might involve stronger intermolecular forces.

  However, too strong hydrophobic forces could lead to aggregation between molecules. Aliphatic chains provided steric-hindrance to disrupt the ordered arrangement of drug molecules. Thus, the steric-hindrance of the prodrugs was simulated. In addition, volume calculations were performed using Yasara, with P-BA_C18 (3092.70 ų) > P-LA_C18 (3069.50 ų), suggesting that P-BA_C18 might possess stronger steric-hindrance (Fig. 1C). Then, the binding energies and intermolecular forces were calculated using molecular simulations. The binding free energy was as follows: P-BA_C18 NPs (−2.8 kcal/mol) and P-LA_C18 NPs (−1.9 kcal/mol). The relatively lower binding energy value of P-BA_C18 NPs signified heightened size stability (Fig. 1D). The above results revealed that the composition of the modification modules had a substantial impact on the assembly mechanism and size stability of the prodrugs. Our previous study found that modification modules could provide structural defects to facilitate the assembly of prodrugs. Branched-chain fatty alcohols could provide greater steric hindrance for prodrugs to balance intermolecular forces, which was expected to improve assembly capacity. The smaller particle size of P-BA_C18 NPs suggested the possibility of forming more compact nanostructures and the non-PEGylated P-BA_C18 NPs also displayed better assembly capacity. In addition, transmission electron microscope (TEM) showed that prodrug nanoassemblies were spherical with no significant difference in morphology (Fig. S6 in Supporting information).

  DSPE-mPEG_2K modification could prolong the systemic circulation of the prodrug nanoassemblies by reducing the uptake of the mononuclear phagocytic system (MPS). Therefore, PEGylated P-LA_C18 NPs and P-BA_C18 NPs were prepared, named P-LA_C18 NPs and P-BA_C18 NPs (Fig. S7A in Supporting information). A significant increase in the particle size of the prodrug nanoassemblies was observed only in the presence of SDS, and the change in the particle size of the P-BA_C18 NPs was more dramatic (Fig. S7B in Supporting information). Thus, hydrophobic force might be the primary force driving the assembly of prodrugs. The particle size of P-BA_C18 NPs and P-LA_C18 NPs were 86.5 and 105.6 nm, respectively. The particle size distribution was uniform, with the poly dispersity index (PDI) less than 0.2 (Fig. 1F and Table S2 in Supporting information). In addition, the P-BA_C18 NPs possessed a lower zeta potential of approximately −28 mV (Table S2), which could prevent the aggregation of the nanoassemblies. Moreover, the drug loading of the prodrug nanoassemblies was greater than 50%. Both prodrug nanoassemblies showed uniform spherical structures as observed by transmission electron microscopy (Fig. 1G).

  As shown in Figs. 1H and I, P-BA_C18 NPs and P-LA_C18 NPs exhibited good stability at 4 ℃ (30 days) and room temperature (22 days). In addition, the prodrug nanoassemblies exhibited negligible changes in particle size following co-incubation with phosphate buffered solution (PBS, pH 7.4) supplemented with 10% fetal bovine serum (FBS) for 12 h, demonstrating the excellent colloidal stability (Fig. 1J). As shown in Fig. S8 (Supporting information), the particle size of prodrug nanoassemblies gradually increased after co-incubation with plasma, and the particle size of P-LA_C18 NPs increased significantly at 1 h. In contrast, P-BA_C18 NPs showed good colloidal stability.

  The impact of steric-hindrance effect on the redox reactivity of P-LA_C18 NPs and P-BA_C18 NPs was investigated using hydrogen peroxide (H₂O₂) and glutathione (GSH) as triggering agents. As shown in Fig. S9 (Supporting information), little PTX was released from P-LA_C18 NPs and P-BA_C18 NPs in the blank medium without the addition of H₂O₂ and GSH. In the presence of H₂O₂, the release of the prodrug nanoassemblies showed concentration- and time-dependence. The release of P-LA_C18 NPs was greater than that of P-BA_C18 NPs regardless of the concentration of H₂O₂ (Figs. 2A and B). In addition, the prodrug solution was added to the release medium containing 10 mmol/L H₂O₂ and 0.01 mmol/L GSH. In this case, the release difference between P-LA_C18 and P-BA_C18 was more significant (Figs. 2E and F). Therefore, the redox sensitivity was mainly affected by the molecular conformation of the prodrugs, but the release trend was consistent with prodrug nanoassemblies. To investigate the underlying mechanism, the oxidative release mechanism of P-LA_C18 NPs and P-BA_C18 NPs was examined. As shown in Fig. 2G, the disulfide bonds could be oxidized to sulfoxide or sulfone, which increased the hydrophilicity of the system and facilitated the hydrolysis of adjacent ester bonds and the release of PTX. Compared with the single-tailed modification module, the increased steric-hindrance effect of the two-tailed modification module hindered the attack of H₂O₂ on the sulfur atoms and ester bonds, and the oxidative sensitivity of the P-BA_C18 NPs was weaker than P-LA_C18 NPs. The oxidation intermediates were captured using ultra performance liquid chromatography tandem mass spectrometry (UPLC-MS-MS) (Figs. S10 and S12 in Supporting information).

  Figure 2

  

  Figure 2.  Redox dual-sensitive drug release. The drug release profiles of P-LA_C18 NPs and P-BA_C18 NPs were measured under the following conditions: (A) 10 mmol/L H₂O₂; (B) 50 mmol/L H₂O₂; (C) 0.01 mmol/L GSH; (D) 0.1 mmol/L GSH. Redox dual-sensitive drug release of prodrugs in (E) 10 mmol/L H₂O₂ and (F) 0.01 mmol/L GSH. (G) The oxidization mechanism and (H) the reduction mechanism of P-LA_C18 NPs and P-BA_C18 NPs. Data are presented as mean ± SD (n = 3).

  DownLoad: Full-Size Img PowerPoint

  As shown in Figs. 2C and D, the reduction sensitivity of P-BA_C18 NPs remained weaker than that of P-LA_C18 NPs in the presence of GSH. The mechanism was that the disulfide bond was cleaved to generate hydrophilic thiols, which further enhanced the hydrophilicity of the system and facilitated the hydrolysis of neighboring ester bonds (Fig. 2H). Despite having identical intermediates (Figs. S11 and S13 in Supporting information), the steric-hindrance effect of the two-tailed modification module was greater than the single-tailed modification module, which led to differences in the attack of GSH on the disulfide bonds.

  The effective release of the parent drug was essential for ensuring the potent antitumor efficacy of the prodrug nanoassemblies. Thus, the intracellular drug release was examined. The release of PTX from the prodrug nanoassemblies was delayed compared to that of Taxol and Abraxane. Additionally, P-LA_C18 NPs demonstrated a higher PTX release compared to P-BA_C18 NPs, which was consistent with the in vitro release results (Figs. 3A–C).

  Figure 3

  

  Figure 3.  Cell assays and pharmacokinetics of P-LA_C18 NPs and P-BA_C18 NPs. PTX released from different concentrations of P-LA_C18 NPs and P-BA_C18 NPs, including 100 ng/mL (A), 200 ng/mL (B) and 500 ng/mL (C). The IC₅₀ values of Taxol, Abraxane, P-LA_C18 NPs and P-BA_C18 NPs in 4T1 cells (D), A549 cells (E) and 3T3 cells (F) were determined. Data are presented as mean ± SD (n = 3). (G) Schematic representation of the tumor-selective cytotoxicity of P-LA_C18 NPs and P-BA_C18 NPs. (H) P-LA_C18 NPs and P-BA_C18 NPs were present in the systemic circulation. The AUC_{0–24 h} of the (I) prodrugs, (J) the released PTX, and (K) the sum (prodrug and PTX, PTX equivalent). (L) The AUC_{0–24 h} of the sum (prodrugs and PTX, PTX equivalent). Data are presented as mean ± SD (n = 5). n.s., no significance. **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-tailed Student's t-test.

  DownLoad: Full-Size Img PowerPoint

  Subsequently, the cytotoxicity of P-LA_C18 NPs and P-BA_C18 NPs was evaluated against two tumor cell lines (4T1 and A549) and one normal cell line (3T3). As shown in Fig. 3D and E, Figs. S14A and B and Table S3 (Supporting information), the delayed release of PTX resulted in lower cytotoxicity of P-LA_C18 NPs and P-BA_C18 NPs compared to Taxol and Abraxane. Moreover, the cytotoxicity of the P-LA_C18 NPs was stronger than P-BA_C18 NPs, which may be attributed to more PTX release from P-LA_C18 NPs within the tumor redox microenvironment (Figs. 3D and E).

  The cytotoxicity of the formulations was decreased for normal cells, especially for the prodrug nanoassemblies (Fig. 3F and Fig. S14C in Supporting information). To accurately demonstrate the tumor selectivity of P-LA_C18 NPs and P-BA_C18 NPs, the tumor selectivity index (SI) was calculated as shown in Table S4 (Supporting information). P-LA_C18 NPs and P-BA_C18 NPs exhibited higher tumor selectivity compared with Taxol and Abraxane. P-LA_C18 NPs and P-BA_C18 NPs released more PTX in the high redox microenvironment of tumor cells and reduced toxicity to normal cells (Fig. 3G). Moreover, P-BA_C18 NPs exhibited the highest tumor selectivity, which might show superior safety.

  P-LA_C18 NPs and P-BA_C18 NPs were co-incubated with fresh rat plasma to investigate the plasma stability. As shown in Fig. S15 (Supporting information), P-BA_C18 NPs had more undegraded prodrugs compared to P-LA_C18 NPs. The better plasma stability of the two-tailed prodrug nanoassemblies (P-BA_C18 NPs) was due to the improved size stability. Moreover, the biosafety of the prodrug nanoassemblies was further demonstrated by hemolysis assay before the in vivo studies. Neither of the prodrug nanoassemblies caused hemolysis in the range of 0.5–2 mg/mL, and the hemolysis percentage (HP%) was less than 5%, which demonstrated the safety of intravenous administration (Figs. S16A–D in Supporting information).

  Pharmacokinetic studies were conducted to investigate the influence of steric-hindrance on the in vivo behavior of P-LA_C18 NPs and P-BA_C18 NPs. All experimental procedures were carried out in strict compliance with the protocols for animal testing and received approval from the Institutional Animal Ethics Committee (IAEC) of Shenyang Pharmaceutical School. The concentration-time profiles of the prodrugs, released PTX and the sum of both them were depicted in Figs. 3I–K. Taxol was cleared quickly from the circulatory system, with a half-life of only 2.2 h (Table S5 in Supporting information). In contrast, the prodrug nanoassemblies extended the duration of blood circulation. (Fig. 3H and Table S5). In addition, the area under the curve (AUC) of P-BA_C18 NPs and P-LA_C18 NPs were approximately 20 and 19 times higher than Taxol (Fig. 3L), respectively. For the prodrug nanoassemblies, a minimal amount of free PTX was released from P-BA_C18 NPs compared to P-LA_C18 NPs. This might be due to the stronger redox sensitivity of P-LA_C18 NPs and the presence of some oxidizing substances within the circulatory system, such as oxygen, leading to the release of PTX from P-LA_C18 NPs. In comparison, P-BA_C18 NPs exhibited superior pharmacokinetic behavior due to appropriate redox sensitivity and better size stability.

  The positive antitumor efficacy of the prodrug nanoassemblies relied on elevated tumor accumulation and effective drug release. Therefore, an examination of the biodistribution and tumor accumulation of the prodrug nanoassemblies were conducted (Fig. S17 in Supporting information). Notably, P-BA_C18 NPs displayed the highest tumor accumulation (Fig. S17F). In addition, the prodrug of single-tailed modified P-LA_C18 NPs was completely degraded at 4 h, and PTX was gradually eliminated at 8 h. However, the prodrug of two-tailed modified P-BA_C18 NPs could still be detected at 8 h, and the release of PTX increased with time. In addition, similar results were found in the in vivo imaging in mice (Fig. S17H). The higher tumor accumulation and sustained PTX release contributed to the excellent pharmacodynamics and safety of the two-tailed modified prodrug nanoassemblies.

  Next, orthotopic breast cancer model (4T1) was developed (Fig. 4A). As shown in Fig. 4B and Fig. S18A (Supporting information), P-LA_C18 NPs and Taxol showed excellent tumor growth inhibition. In addition, the tumor burden of P-LA_C18 NPs group was the lowest (Fig. 4D), which could be attributed to the higher redox release capacity and good tumor selectivity. P-BA_C18 NPs showed weaker tumor growth inhibition than P-LA_C18 NPs and Taxol, but similar to Abraxane. Moreover, lung metastasis was significantly inhibited in the prodrug nanoassemblies groups (Fig. 4E and Fig. S18B in Supporting information). The tumor tissues of the administered groups showed obvious cell necrosis (Fig. 4F). The terminal dUTP nick end labeling (TUNEL) (Figs. S19B and C in Supporting information) and Ki-67 staining (Figs. S19D and E in Supporting information) also showed that P-LA_C18 NPs induced higher levels of tumor apoptosis and inhibited tumor proliferation.

  Figure 4

  

  Figure 4.  Antitumor efficacy of P-LA_C18 NPs and P-BA_C18 NPs. (A) Antitumor effects against orthotopic 4T1 tumors. (B–D) Tumor growth kinetics, body weight changes, and tumor burden were evaluated in BALB/c mice at a dose of 10 mg/kg. (E) The number of metastatic nodules. (F) H & E staining of tumor sections of 4T1 tumor-bearing mice. Scale bar: 100 µm. (G) Schematic of the administration schedule in A549 xenograft tumors. (H, I) Tumor growth profiles and body weight at 30 mg/kg. (J, K) Tumor growth profiles and body weight at 45 mg/kg of Abraxane, and 60 mg/kg of prodrug nanoassemblies. Data are presented as mean ± SD (n = 3). P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way analysis of variance (ANOVA).

  DownLoad: Full-Size Img PowerPoint

  To further validate the antitumor efficacy of P-LA_C18 NPs and P-BA_C18 NPs in different tumor models and at different administered doses, A549 tumor-bearing nude mice were constructed (Fig. 4G). As shown in Figs. 4H and J, at low doses, the P-LA_C18 NPs exhibited the smallest tumor volume due to the fastest release of PTX. The antitumor effect of P-BA_C18 NPs was slightly weaker than P-LA_C18 NPs. At high doses, the tumor growth inhibition of P-BA_C18 NPs was comparable to P-LA_C18 NPs. The enhanced antitumor effect of P-BA_C18 NPs was ascribed to good colloidal stability, improved pharmacokinetics and high tumor accumulation.

  To evaluate the safety of P-LA_C18 NPs and P-BA_C18 NPs, changes in body weight of the mice were monitored during the pharmacodynamic experiments. In the breast cancer model, mice in the Taxol group experienced greater weight loss compared with the saline group (Fig. 4C). No significant body weight change was observed in other preparation groups. At the end of the pharmacodynamic experiment, the tissues from each group were stained with hematoxylin-eosin (H & E) staining. Hepatic edema and hemorrhage were seen in the Taxol group (Fig. S19A in Supporting information). Blood cell analysis revealed lower Gran% values in the Taxol group, indicating higher toxic side effects (Figs. S20A–D in Supporting information). In the lung cancer model, mice in the Abraxane group showed slight weight loss at different administered doses. In addition, mice in the P-LA_C18 NPs group showed significant weight loss and death when administered at doses of 30 mg/kg and 60 mg/kg (Figs. 4I and K). The body weight of mice in the P-BA_C18 NPs group showed no significant change at both low and high doses, indicating a favorable safety.

  Subsequently, the tolerance of P-LA_C18 NPs and P-BA_C18 NPs was assessed in BALB/c-nude mice, using body weight and survival rate as indicators. As shown in Figs. S20E and F and Table S6 (Supporting information), the tolerance of P-LA_C18 NPs was poor, and all mice died on the 6th day with continuous weight loss. Similarly, mice in the Abraxane group showed poor tolerance with the same weight loss and death after continuous administration. In contrast, mice in P-BA_C18 NPs group demonstrated no significant abnormalities in body weight, indicating an outstanding tolerance.

  In this study, the disulfide bond-bridged single-tailed and two-tailed modified prodrug nanoassemblies (P-LA_C18 NPs and P-BA_C18 NPs) were constructed, and the steric-hindrance effect of the two-tailed modification module was found to significant impact the formulation properties and therapeutic index of the prodrug nanoassemblies. Firstly, the two-tailed modified prodrug nanoassemblies exhibited lower binding energy and stronger intermolecular forces, which promoted the stable assembly P-BA_C18 NPs. Secondly, in comparison to the single-tailed prodrug nanoassemblies, the redox sensitivity of the two-tailed prodrug nanoassemblies was reduced due to the steric-hindrance effect. Finally, due to the favorable size stability and appropriate redox sensitivity, the two-tailed modified prodrug nanoassemblies showed advantages in pharmacokinetics, biodistribution, pharmacodynamics and safety. The single-tailed modified prodrug nanoassemblies showed impressive antitumor effects due to rapid PTX release, but the safety remained to be a concern. Our findings shed light on the relationship between steric-hindrance effects and the activities of the prodrug nanoassemblies, and offered new viewpoints into the rational design of self-assembled prodrugs for high performance cancer therapeutics.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Wenfeng Zang: Writing – original draft, Methodology, Investigation. Yixin Sun: Writing – original draft, Methodology, Investigation. Jingyi Zhang: Data curation. Yanzhong Hao: Methodology, Formal analysis. Qianhui Jin: Data curation. Hongying Xiao: Validation, Supervision. Zuo Zhang: Validation, Supervision. Xianbao Shi: Validation, Supervision. Jin Sun: Writing – review & editing, Funding acquisition. Zhonggui He: Writing – review & editing, Funding acquisition. Cong Luo: Writing – review & editing. Bingjun Sun: Writing – review & editing, Funding acquisition, Data curation.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China, (Nos. 82272151, 82204318), Liaoning Revitalization Talents Program (No. XLYC2203083), and Shenyang Young and Middle-aged Science and Technology Innovation Talent Support Program (No. RC220389), Postdoctoral Fellowship Program of CPSF (No. GZC20231732), China Postdoctoral Science Foundation (Nos. 2023TQ0222, 2023MD744229).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110230.

  
  
• 1. [1]

     C. Swanton, E. Bernard, C. Abbosh, et al., Cell 187 (2024) 1589–1616. doi: 10.1016/j.cell.2024.02.009

  2. [2]

     J. Shi, P.W. Kantoff, R. Wooster, et al., Nat. Rev. Cancer 17 (2016) 20–37.

  3. [3]

     R. Walther, J. Rautio, A.N. Zelikin, Adv. Drug Deliv. Rev. 118 (2017) 65–77. doi: 10.1016/j.addr.2017.06.013

  4. [4]

     B. Sun, C. Luo, W. Cui, et al., J. Control. Release 264 (2017) 145–159. doi: 10.1016/j.jconrel.2017.08.034

  5. [5]

     Q. Song, Y. Yin, L. Shang, et al., Nano Lett. 17 (2017) 6366–6375. doi: 10.1021/acs.nanolett.7b03186

  6. [6]

     Q. Chen, Z. Liu, Adv. Mater. 28 (2016) 10557–10566. doi: 10.1002/adma.201600038

  7. [7]

     M. Hama, Y. Ishima, V.T.G. Chuang, et al., ACS Appl. Mater. Interfaces 13 (2021) 19736–19744. doi: 10.1021/acsami.1c03065

  8. [8]

     F. Kratz, J. Control. Release 190 (2014) 331–336. doi: 10.1016/j.jconrel.2014.03.013

  9. [9]

     S. Ezrahi, A. Aserin, N. Garti, Adv. Colloid Interface Sci. 263 (2019) 95–130. doi: 10.1016/j.cis.2018.11.004

  10. [10]

      B. Sun, C. Luo, H. Yu, et al., Nano Lett. 18 (2018) 3643–3650. doi: 10.1021/acs.nanolett.8b00737

  11. [11]

      C.M. Hartshorn, M.S. Bradbury, G.M. Lanza, et al. ACS Nano 12 (2017) 24–43.

  12. [12]

      J. Wang, B. Li, U.B. Kompella, et al., MedComm Biomater. Appl. 2 (2023) e30.

  13. [13]

      T. Tong, L. Chen, S. Wu, et al., Chin. Chem. Lett. (2024), https://doi.org/10.1016/j.cclet.2024.109689. doi: 10.1016/j.cclet.2024.109689

  14. [14]

      S. Li, Y. Yang, S. Wang, et al., Exploration 2 (2022) 20210223. doi: 10.1002/EXP.20210223

  15. [15]

      Y. Huang, T. Wang, Q. Tan, et al., Int. J. Nanomed. Volume 16 (2021) 4117–4146. doi: 10.2147/ijn.s315368

  16. [16]

      J. Li, Y. Li, Y. Wang, et al., Nano Lett. 17 (2017) 6983–6990. doi: 10.1021/acs.nanolett.7b03531

  17. [17]

      X. Wu, X. Chen, X. Wang, et al., Chin. Chem. Lett. 35 (2024) 108756. doi: 10.1016/j.cclet.2023.108756

  18. [18]

      R. Zhang, X. You, M. Luo, et al., Carbohydr Polym. 292 (2022) 119695. doi: 10.1016/j.carbpol.2022.119695

  19. [19]

      W. Zhong, X. Zhang, X. Duan, et al., Acta Biomater. 144 (2022) 67–80. doi: 10.1016/j.actbio.2022.03.035

  20. [20]

      L. Li, B. Sun, J. Sun, et al., Chin. Chem. Lett. (2024), https://doi.org/10.1016/j.cclet.2024.109538. doi: 10.1016/j.cclet.2024.109538

  21. [21]

      H. Liu, T. Nie, X. Duan, et al., J. Control. Release 359 (2023) 132–146. doi: 10.1016/j.jconrel.2023.05.040

  22. [22]

      J. Zhang, H. Guo, M. Liu, et al., Exploration 4 (2024) 20230087.

  23. [23]

      C. Zhang, Y. Zuo, T. Zhang, et al., Acta Mater. Med. 2 (2023) 430–448.

  24. [24]

      R. Wang, X. Zhang, K. Feng, et al., Asian J. Pharm. Sci. 18 (2023) 100857.

  25. [25]

      C. Luo, J. Sun, B. Sun, et al., Small 12 (2016) 6353–6362. doi: 10.1002/smll.201601597

  26. [26]

      B. Li, T. Tan, W. Chu, et al., Drug Deliv. 29 (2021) 75–88. doi: 10.1360/ssi-2019-0226

  27. [27]

      X. Wang, T. Liu, Y. Huang, et al., Nanoscale Horiz. 8 (2023) 235–244. doi: 10.1039/d2nh00425a

  28. [28]

      Q. Pei, B. Jiang, D. Hao, et al., Acta Pharmacol. Sin. B 13 (2023) 3252–3276. doi: 10.1016/j.apsb.2023.02.021

  29. [29]

      Y. Li, Q. Bao, S. Yang, et al., View 3 (2022) 20200027. doi: 10.1002/VIW.20200027

  30. [30]

      J. Zhang, Y. Zhang, Y. Huang, et al., Chem. Eng. J. 458 (2023) 141510. doi: 10.1016/j.cej.2023.141510

  31. [31]

      L. Li, S. Zuo, F. Dong, et al., Asian J. Pharm. Sci. 16 (2021) 337–349.

  32. [32]

      C. Luo, B. Sun, C. Wang, et al., J. Control. Release 302 (2019) 79–89. doi: 10.1016/j.jconrel.2019.04.001

  33. [33]

      Y. Sun, S. Wang, Y. Li, et al., Acta Biomater. 157 (2023) 417–427. doi: 10.1016/j.actbio.2022.12.005

  34. [34]

      Y. Gao, S. Zuo, L. Li, et al., Nanoscale 13 (2021) 10536–10543. doi: 10.1039/d1nr01359a

  35. [35]

      H. Xu, S. Zuo, D. Wang, et al., J. Control. Release 360 (2023) 784–795. doi: 10.1016/j.jconrel.2023.07.012

  36. [36]

      Y. Yang, X. Li, J. Song, et al., Nano Lett. 23 (2023) 1530–1538. doi: 10.1021/acs.nanolett.2c04976

  37. [37]

      G. Li, F. Xia, H. Xiao, et al., Nano. Res. 17 (2023) 2908–2918. doi: 10.1177/0020720921996606

  38. [38]

      G. Wang, Q. Bai, H. Liu, et al., Inorg. Chem. 62 (2023) 6537–6543. doi: 10.1021/acs.inorgchem.3c00026

  39. [39]

      J. Liao, S. Liu, Y. Yuan, et al., New J. Chem. 42 (2018) 5698–5708. doi: 10.1039/c8nj00136g

• 

  Scheme 1  Two-tailed modification module tuned steric-hindrance effect enabling high therapeutic efficacy of paclitaxel prodrug nanoassemblies.

  下载: 全尺寸图片 幻灯片
  

  Figure 1  Self-assembly capacity and characterization of prodrug nanoassemblies. (A) The appearance of non-PEGylated P-LA_C18 NPs and P-BA_C18 NPs at concentrations of 0.1, 0.2, 0.4 and 0.6 mg/mL. (B) The variation of the particle size of non-PEGylated prodrug nanoassemblies after co-incubated with NaCl, SDS or Urea. (C) Steric-hindrance effects of P-LA_C18 and P-BA_C18. (D) Intermolecular interactions of P-LA_C18 NPs and P-BA_C18 NPs during the self-assembly process. (E) LogP values of two prodrugs. (F) The particle size and (G) morphology of P-LA_C18 NPs and P-BA_C18 NPs. Scale bar: 200 nm. Storage stability of P-LA_C18 NPs and P-BA_C18 NPs (H) at 4 ℃ and (I) at 25 ℃. (J) The changes in particle size of P-LA_C18 NPs and P-BA_C18 NPs after coincubation with PBS containing FBS for 12 h. Data are presented as mean ± SD (n = 3). P < 0.05 by two-tailed Student's t-test.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Redox dual-sensitive drug release. The drug release profiles of P-LA_C18 NPs and P-BA_C18 NPs were measured under the following conditions: (A) 10 mmol/L H₂O₂; (B) 50 mmol/L H₂O₂; (C) 0.01 mmol/L GSH; (D) 0.1 mmol/L GSH. Redox dual-sensitive drug release of prodrugs in (E) 10 mmol/L H₂O₂ and (F) 0.01 mmol/L GSH. (G) The oxidization mechanism and (H) the reduction mechanism of P-LA_C18 NPs and P-BA_C18 NPs. Data are presented as mean ± SD (n = 3).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Cell assays and pharmacokinetics of P-LA_C18 NPs and P-BA_C18 NPs. PTX released from different concentrations of P-LA_C18 NPs and P-BA_C18 NPs, including 100 ng/mL (A), 200 ng/mL (B) and 500 ng/mL (C). The IC₅₀ values of Taxol, Abraxane, P-LA_C18 NPs and P-BA_C18 NPs in 4T1 cells (D), A549 cells (E) and 3T3 cells (F) were determined. Data are presented as mean ± SD (n = 3). (G) Schematic representation of the tumor-selective cytotoxicity of P-LA_C18 NPs and P-BA_C18 NPs. (H) P-LA_C18 NPs and P-BA_C18 NPs were present in the systemic circulation. The AUC_{0–24 h} of the (I) prodrugs, (J) the released PTX, and (K) the sum (prodrug and PTX, PTX equivalent). (L) The AUC_{0–24 h} of the sum (prodrugs and PTX, PTX equivalent). Data are presented as mean ± SD (n = 5). n.s., no significance. **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-tailed Student's t-test.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Antitumor efficacy of P-LA_C18 NPs and P-BA_C18 NPs. (A) Antitumor effects against orthotopic 4T1 tumors. (B–D) Tumor growth kinetics, body weight changes, and tumor burden were evaluated in BALB/c mice at a dose of 10 mg/kg. (E) The number of metastatic nodules. (F) H & E staining of tumor sections of 4T1 tumor-bearing mice. Scale bar: 100 µm. (G) Schematic of the administration schedule in A549 xenograft tumors. (H, I) Tumor growth profiles and body weight at 30 mg/kg. (J, K) Tumor growth profiles and body weight at 45 mg/kg of Abraxane, and 60 mg/kg of prodrug nanoassemblies. Data are presented as mean ± SD (n = 3). P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by one-way analysis of variance (ANOVA).

  下载: 全尺寸图片 幻灯片
•