Residue cytotoxicity of a hydrazone-linked polymer–drug conjugate: implication for acid-responsive micellar drug delivery

Abstract

We report that the polymer residue of hydrazone-containing pH-responsive polymeric conjugate micelles could induce considerable cytotoxicity in a model cell line (HeLa). However, there was no significant difference between the cytotoxicity of the residue of the model drug (curcumin) and its parent form post hydrolysis. The results demonstrated that both the polymer residue and active drug could be beneficial for cancer treatment, whereas such a synergistic role of an amine-containing polymer residue was often neglected.

---

Polymer–drug conjugate micelles exhibit many favorable characteristics for cancer chemotherapy to achieve enhanced therapeutic efficacy accompanied with reduction in adverse effects.¹ The problem of poor aqueous solubility of active agents can be readily solved without the issue of premature drug release.² The tunable micelle size can utilize the enhanced permeability and retention (EPR) effect to achieve passive tumor tissue targeting.³ The co-assembly of different types of conjugates makes the ratiometrically combinational delivery possible.⁴ Micelle surface decoration with targeting molecules enables the active targeting to the tumor cells.⁵ The reversal of cancer multidrug resistance has also been reported via the introduction of chemosensation molecules in the nanoparticles.⁶ Hence such a delivery system exhibits high translational potential and several micelle formulations are currently at different stages of clinical evaluation.⁷

The polymer–drug conjugates are often created via the ester bond that can be hydrolyzed to release the parent drug molecule.² However, the drug release rate from ester bond-linked conjugate micelles is usually slow, which would delay the rapid onset of drug action. Hence, stimuli-responsive linkers have been designed to expedite the breaking-down of the conjugate, and a number of stimuli have been previously utilized, including temperature, ultrasound, pH, redox potential, light, and enzymes.^8,9 Among these, pH-responsive delivery is unique since such stimulus is naturally present in vivo; there are marked pH gradients between the neutral blood and certain tissues (e.g. tumor), and the endosome/lysosome also shows an acidic microenvironment (pH 4.5–6.5) compared to other cellular compartments and the cytoplasm.¹⁰ In addition, the rich collection of pH-labile linkers (e.g. acetal, dimethyl maleate, hydrazone, hydrazide, imine, ketal, and oxime) facilitates the tailored fabrication of smart polymer–drug conjugate nanostructures.¹¹

Hydrazone has been a widely used pH-liable bond for connecting the polymer backbone and various active therapeutics, possibly due to the ease of synthesis.¹² The conjugation process necessitates the presence of an amine-ended polymer and a carbonyl group in the drug molecule. Hydrazine was usually employed to achieve the former under ambient conditions. It is an aircraft fuel and propellant, showing potent reducing capability and toxicity. Except for a few active compounds such as doxorubicin, most therapeutic drugs need modification by another keto acid (e.g. levulinic acid) to bring in the carbonyl group before conjugation.

In the current study, we used the biodegradable hydrazine-modified poly(ethylene glycol)–poly(lactic acid) (mPEG–PLA–NHNH₂) as the amphiphilic block copolymer and levulinic acid-modified curcumin (Cur-L) as the model drug. The detailed synthesis and characterization procedure of the curcumin derivative and amine-ended mPEG–PLA was reported recently.^2,12 Such a conjugate maintained the amphiphilic nature of mPEG–PLA and could self-assemble into micellar nanostructures. Upon hydrolysis, the conjugate would degrade back to mPEG–PLA–NH–NH₂ and Cur-L (Fig. 1). It is well known that an amine-ended polymer such as polyethylenimine and poly(amido amine) would exhibit cytotoxicity.^13,14 Hence it was postulated that the cationic polymer residue would show cytotoxicity; the aim of this study was to investigate the extent to which the hydrazone-linked polymer–drug conjugate affected the cell viability and to examine whether the cytotoxicity of the keto acid-modified drug derivative is different to that of the parent drug.

Fig. 1 The hydrazone-linked mPEG–PLA–Cur conjugate and its hydrolysis residue. mPEG, PLA and Cur are short for methoxy poly(ethylene glycol), poly(lactic acid) and curcumin, respectively.

HeLa cells were used as the model system for cell viability assessment and the CCK-8 (cell counting kit-8) method was employed as described previously.¹² In contrast to the hydroxyl-ended mPEG–PLA–OH control, amine-terminated mPEG–PLA–NH–NH₂ showed substantial cytotoxicity leading to a ca. 50% cell viability at the maximum polymer concentration (100 μg mL⁻¹ or 28.6 mM) (Fig. 2).

Fig. 2 The dose-dependent viability of HeLa cells in response to amine/hydroxyl-terminated polymers: mPEG–PLA–NH–NH₂ (white), and mPEG–PLA–OH (hatched) micelle (n ≥ 4) (p < 0.05 at the polymer concentration of 20–100 μg mL⁻¹).

The half maximal inhibitory concentration (IC₅₀) of another control (hydrazine) was 2.8 ± 0.2 mM (Fig. 3). Previous work showed that the cytotoxicity of hydrazine was accompanied with increased lactate dehydrogenase (LDH) leakage, decreased mitochondrial activity, depletion of glutathione, and elevated reactive oxygen species generation as well as lipid peroxidation using rat hepatocytes.¹⁵ Other previous studies have revealed that the amine-containing polymers such as polyamidoamine and polyethylenimine displayed cytotoxicity in several clinically relevant human cell lines via inducing membrane damage and initiating apoptosis.^16,17 The cytotoxicity of cationic mPEG–PLA–NH–NH₂ in the current study might experience similar molecular processes.

Fig. 3 The dose-dependent viability of HeLa cells in response to hydrazine (n ≥ 4).

Interestingly, there was no significant difference between the IC₅₀ of the model drug, curcumin (35.9 ± 4.6 μM), and its derivative Cur-L (37.1 ± 4.5 μM) (p > 0.05) (Fig. 4). Many previous works on the structure–activity relationship of curcumin and its derivatives revealed that the phenolic group is vital for the free-radical scavenging activity and the diketone moiety is essential for cell proliferation inhibition.¹⁸ Hence the modification of the phenol group in the current study did not significantly influence the cytotoxicity of curcumin. The uncompromised drug cytotoxicity is beneficial to maintain the efficacy of active agents for the pH-responsive delivery systems. Our recent work showed that the conjugates exhibited a pH-dependent drug release profile.¹² Compared to the IC₅₀ (170.4 ± 10.3 μM) of the control (polymer–drug conjugate linked with an ester bond),¹⁹ the polymer–curcumin conjugate exhibited an IC₅₀ of 64.0 ± 9.3 μM (Fig. 5), which was the action of both the active agent and polymer carrier residue. This phenomenon would be beneficial for killing tumor cells via synergism, but this might also enhance the adverse effects to healthy tissues since an absolute tumor targeting system does not exist.²⁰ However, the synthesis of a new active curcumin derivative for ease of conjugate generation is typically not advantageous in the process of commercial pharmaceutical product development and regulatory approval.

Fig. 4 The dose-dependent viability of HeLa cells in response to curcumin (white) and its derivative Cur-L (hatched) (n ≥ 4) (p > 0.05 in terms of IC₅₀ of curcumin and its derivative).

Fig. 5 The dose-dependent viability of HeLa cells in response to hydrazone-linked mPEG–PLA–Cur conjugate (hatched) and ester-linked mPEG–PLA–Cur conjugate (white) (n ≥ 4) (p < 0.05 at the conjugate concentration of 10–100 μg mL⁻¹).

To sum up, we report the residue cytotoxicity of the hydrazone-linked polymer–curcumin conjugate. Although only one type of cell line was used, the current study highlights the importance of linker selection for designing acid-responsive delivery systems. Further work will test whether the cytotoxicity of these residues is cell type-dependent. In addition, the potential alteration of drug efficacy should be considered when modifying its parent form, which is necessary for developing novel polymer–drug conjugate micellar delivery systems.

Acknowledgements

The work was supported by National Basic Research Program of China (2015CB856500), National Natural Science Foundation of China (2134068), and Tianjin Research Program of Application Foundation and Advanced Technology (14JCZDJC38400).

Notes and references

1. X. Hu and X. Jing, Expert Opin. Drug Delivery, 2009, 6, 1079.
2. R. Yang, S. Zhang, D. Kong, X. Gao, Y. Zhao and Z. Wang, Pharm. Res., 2012, 29, 3512.
3. H. Maeda, J. Controlled Release, 2012, 164, 138.
4. S. Aryal, C. M. J. Hu and L. Zhang, Mol. Pharm., 2011, 8, 1401.
5. R. Liu, M. Sun, X. Liu, A. Fan, Z. Wang and Y. Zhao, Int. J. Pharm., 2014, 462, 103.
6. V. Drinberg, R. Bitcover, W. Rajchenbach and D. Peer, Cancer Lett., 2014, 354, 290.
7. H. Cabral and K. Kataoka, J. Controlled Release, 2014, 190, 465.
8. S. Ganta, H. Devalapally, A. Shahiwala and M. Amiji, J. Controlled Release, 2008, 126, 187.
9. S. Mura, J. Nicolas and P. Couvreur, Nat. Mater., 2013, 12, 991.
10. Z. Wang, C. Chen, R. Liu, A. Fan, D. Kong and Y. Zhao, Chem. Commun., 2014, 50, 14025.
11. E. Fleige, M. A. Quadir and R. Haag, Adv. Drug Delivery Rev., 2012, 64, 866.
12. Z. Wang, C. Chen, Q. Zhang, M. Gao, J. Zhang, D. Kong and Y. Zhao, Eur. J. Pharm. Biopharm., 2015, 90, 53.
13. Y. Zhao, R. Yang, D. Liu, M. Sun, L. Zhou, Z. Wang and Y. Wan, J. Biomed. Mater. Res., Part A, 2012, 100, 134.
14. Y. Fu, X. Gao, Y. Wan, J. Zhang, D. Kong, Z. Wang and Y. Zhao, RSC Adv., 2014, 4, 46020.
15. S. M. Hussain and J. M. Frazier, Toxicol. Sci., 2002, 69, 424–432.
16. R. Jevprasesphant, J. Penny, R. Jalal, D. Attwood, N. B. McKeown and A. D’Emanuele, Int. J. Pharm., 2003, 252, 263.
17. S. M. Moghimi, P. Symonds, J. C. Murray, A. C. Hunter, G. Debska and A. Szewczyk, Mol. Ther., 2005, 11, 990.
18. B. B. Aggarwal, I. D. Bhatt, H. Ichikawa, K. S. Ahn, G. Sethi, S. K. Sandur, C. Natarajan, N. Seeram and S. Shishodia, in Turmeric: The genus Curcuma, ed. P. N. Ravindran, K. N. Babu and K. Sivaraman, CRC Press, London, 2007, pp. 297–368.
19. Y. Cao, M. Gao, C. Chen, A. Fan, J. Zhang, D. Kong, Z. Wang, D. Peer and Y. Zhao, Nanotechnology, 2015, 26, 115101.
20. I. K. Kwon, S. C. Lee, B. Han and K. Park, J. Controlled Release, 2012, 164, 108.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra02097b

---