Dopasomes: dopamine-mediated cross-linked lipid vesicles

Abstract

Liposomes offer significant advantages for drug delivery but are limited by poor stability, which hinders product development and therapeutic efficacy. In biological environments containing surfactant-like components such as serum proteins and bile salts, liposomes are susceptible to destabilization. In this study, we designed and synthesized dopamine-containing, self-polymerizing lipids that self-assemble into liposomal structures, termed as “dopasomes.” Dopasomes retained their structural integrity even in the presence of Triton X-100 and enabled efficient intracellular delivery of doxorubicin hydrochloride.

---

Lipid-based nanoparticles are highly versatile platforms that have been extensively studied for diverse applications, including drug delivery systems,^1,2 biosensors,^3,4 artificial cell membranes,^5,6 and chemical reactors.^7,8 They have recently garnered considerable attention in the medical field, particularly for their pivotal role during the COVID-19 pandemic.⁹ Among these, liposomes are among the most thoroughly investigated lipid-based nanoparticles, especially for drug delivery. Several liposome-based drugs have been approved by the Food and Drug Administration,^10,11 with many others currently in clinical trials.^12,13 Their broad appeal stems from desirable delivery features such as biocompatibility and tunable physicochemical properties, including morphology, surface charge, surface modifiability, and the ability to encapsulate both hydrophilic and hydrophobic agents.¹⁴

Despite their advantages, liposomes face considerable challenges, particularly regarding stability, which hampers both product development and therapeutic performance.^15,16 Liposomes are artificially prepared vesicles composed of one or more phospholipid bilayers and are subject to both chemical and structural instability. Chemically, phospholipids are prone to ester bond hydrolysis and lipid peroxidation under physiological conditions, leading to membrane degradation.¹⁷ Physically, liposomes tend to aggregate and fuse, causing drug leakage and loss of functionality.¹⁷ Additionally, interactions with surfactants such as amphiphilic plasma proteins can further compromise liposomal structure and integrity.¹⁸ Consequently, conventional liposomes often exhibit poor shelf life and in vivo stability.

To address this challenge, various strategies have been investigated, including the incorporation of membrane stabilizers,¹⁹ lipid polymerization,²⁰ and surface modifications with polymers²¹ or inorganic materials.²² One promising approach to stabilize liposomes involves cross-linking the lipid membrane using polymerizable lipids incorporated into the bilayer. However, synthesizing polymerizable amphiphilic molecules remains challenging due to the complexity of multi-step reactions and the need for precise control over the polymerization conditions.

In this study, we designed and synthesized a self-polymerizable lipid containing a dopamine moiety as a cross-linker, which self-assembled into a liposomal structure termed a “dopasome” and exhibited enhanced colloidal stability due to its polydopamine framework (Fig. 1). Dopamine derivatives are prone to oxidation under alkaline conditions in the presence of molecular oxygen, leading to polydopamine formation.²³ In the dopasome design, the hydrophilic head group is replaced with dopamine derivatives (L-DOPA) to enable in situ coupling. The L-DOPA moiety undergoes oxidation in alkaline aqueous environments and cross-links within the lipid layer to form oligomers, thereby enhancing both physical and chemical stability. Previous reports have examined other types of lipid-crosslinked liposomes, including partially silica- or ceramic-coated vesicles known as cerasomes.²⁰ In general, cerasomes exhibit excellent stability owing to their surface siloxane network. In contrast, dopasomes are composed of fully organic components, including lipids and dopamine derivatives, and therefore offer excellent biocompatibility and biodegradability. Moreover, with dopasomes, polymerization is triggered on demand by the addition of a base or oxidant, and cross-linking can occur under milder conditions. The resulting dopasomes demonstrated strong resistance to surfactants and effectively minimized the leakage of encapsulated agents. These results suggest that dopasomes are a promising platform for developing next-generation lipid-based nanocarriers.

Fig. 1 Schematic illustration of the formation process of the novel cross-linked liposome “dopasome”.

Since the free amino group of L-DOPA is known to be essential for initiating self-polymerization,²⁴ we designed the lipid structure to retain this functional group. The L-DOPA-based lipid (DO-DOPA) was synthesized by covalently conjugating L-DOPA with suitable aspartate diesters bearing unsaturated alkyl chains through conventional peptide coupling, as schematically illustrated in Scheme S1. As shown in Fig. S1, the success of the condensation reaction was confirmed by the integration ratio of the terminal methyl protons of the lipid (δ = 0.86 ppm) to the aromatic protons (δ = 6.5–6.8 ppm), which was approximately 2 : 1. The structure of the final product was further verified using electrospray ionization mass spectrometry (ESI-MS) (Fig. S2).

We investigated the liposome-forming capability of DO-DOPA using ethanol injection²⁵ and thin-film hydration²⁶ methods to optimize the liposome preparation conditions. Initially, liposomes were prepared via the ethanol injection method under varying alkaline conditions—commonly used in dopamine polymerization protocols.²⁷ To assess the effect of base concentration on the reaction, liposomes were prepared with increasing NaOH concentrations by varying the [lipid]/[NaOH] ratios from 1 : 0 to 1 : 8. In the absence of NaOH, nanoparticles with an average diameter of 170 nm were detected via dynamic light scattering (DLS), although the dispersion exhibited poorly defined particles and broad size distributions (Fig. S3). At a 1 : 1 ratio, the particle size increased after 30 min, likely due to incomplete deprotonation of the catechol groups. In contrast, at ratios of 1 : 4 and higher, the particle size remained stable for at least 24 h, suggesting that adequate deprotonation of catechol moieties improved colloidal stability (Table 1). DLS correlogram fittings are provided in Fig. S4, SI, showing good agreement with the measured curves. Consequently, a ratio of 1 : 4 was identified to be optimal for liposome formation, yielding the lowest polydispersity index (PDI).

Table 1 Solution properties of the dopasome

Label	Preparation method	NaOH [equiv.]	T/°C	Dhy/nm	PDI	ζ-potential /mV
Dynamic light scattering (DLS) measurements were performed at 25 °C in MilliQ water. PDI values were calculated using the cumulant method. ζ-potential was measured using a capillary cell.
a	Ethanol injection	0	25	169.6 ± 3.2	0.314	+52.9 ± 4.3
b	Ethanol injection	1	25	831.2 ± 154.4	0.788	−17.8 ± 0.7
c	Ethanol injection	4	25	130.1 ± 1.3	0.100	−35.6 ± 0.2
d	Ethanol injection	8	25	121.6 ± 2.2	0.141	−76.8 ± 6.9
e	Ethanol injection	4	60	74.2 ± 5.1	0.197	−33.4 ± 1.3
f	Thin-film	0	25	78.4 ± 0.1	0.238	+64.7 ± 0.5
g	Thin-film	4	60	92.8 ± 1.7	0.241	−31.2 ± 1.7

---

Next, we investigated how the cross-linking behavior of DO-DOPA varied with alkaline concentrations and temperature. During oxidative polymerization of dopamine, the catechol moiety is first oxidized to o-quinone, which undergoes intramolecular cyclization via a Michael-type addition.²⁷ Further oxidation produces dopaminechrome, which can rearrange into 5,6-dihydroxyindole. This 5,6-dihydroxyindole moiety plays a key role in polymerization through both covalent and non-covalent interactions (Scheme S2).²⁷ These transformations can be monitored using ultraviolet-visible (UV-vis) absorption spectroscopy. Polydopamine exhibits a broad absorption band in the visible region and causes a color change in solution from colorless to brown or dark brown. As shown by the black line in Fig. 2a, the UV-vis spectrum of the lipid in ethanol shows a characteristic peak at around 280 nm, attributed to the catechol group.²⁸ Upon liposome formation under alkaline conditions, a new peak emerged at around 320 nm after 30 min of incubation at 25 °C, dictating quinone formation through oxidation by dissolved oxygen (Fig. 2a).²⁹ After 24 h, the solution exhibited broad absorption and turned brown, indicating polydopamine formation. We next examined the effect of high temperatures on the cross-linking process. Heating the sample at 60 °C for 24 h resulted in a notable increase in UV-vis absorption, particularly within the first 3 h (Fig. 2a). These findings suggest that elevated temperatures accelerate the cross-linking reaction. However, precipitation occurred after 6 h of preparation, indicating interparticle cross-linking. To prevent this, the heating duration was reduced to 30 min, which effectively maintained the particle size at approximately 70 nm and suppressed precipitation.

Fig. 2 Characterization of the cross-linking behavior of the L-DOPA-based lipid (DO-DOPA). (a) UV-vis spectra of DO-DOPA in ethanol (black) and in the presence of NaOH at room temperature (blue) and at 60 °C (red). Inset images: (1) unpolymerized ethanol solution and (2) polymerized product after heating at 60 °C. (b) Time-dependent UV-vis spectral changes of liposomes from 30 min to 24 h. Line darkness increases with time; darker lines represent later time points. Inset: change in absorbance at 360 nm. (c) ESI-MS spectra of DO-DOPA treated with NaOH.

Additionally, the formation of dopamine cross-linking within the bilayer was analyzed using ESI-MS spectra (Fig. 2c). Only dimer peaks were detected, likely due to the poor solubility of higher-order oligo-dopamine species in common organic solvents, possibly resulting from strong intramolecular interactions. Liposomal stability is known to improve through enhanced lipid–lipid interactions, particularly those involving cholesterol²⁹ and bipolar lipids.³⁰ This suggests that synthetic dimeric lipids may similarly contribute to membrane stabilization. Furthermore, several oxidizing agents commonly used for dopamine oxidation, including aqueous ammonia and sodium periodate (NaIO₄), were tested. Both agents induced cross-linking (Fig. S5); however, NaIO₄ exhibited strong oxidative activity, leading to precipitation due to interparticle cross-linking.

Next, dopasomes were also prepared using the thin-film hydration method followed by sonication, commonly known as the Bangham method.²⁶ The resulting dopasomes exhibited particle sizes comparable to those obtained via the ethanol injection method (Table 1). Unlike the ethanol injection method, this approach enabled dopasome preparation under neutral pH conditions, likely due to the favorable alignment of lipid molecules during thin-film formation.

To gain deeper insights into the lipid structures in these formulations, we employed transmission electron microscopy (TEM), cryo-TEM, and small-angle X-ray scattering (SAXS) measurements. TEM images revealed spherical particles with diameters ranging from 20 to 56 nm (Fig. S6). Cryo-TEM observations showed vesicle-like morphology in both cross-linked and uncross-linked dopasomes (Fig. 3a and b), indicating that DO-DOPA lipids self-assemble into liposome-like vesicles in aqueous environments, regardless of cross-linking. We then performed SAXS measurements to further investigate the structural features inferred from TEM. As shown in the SAXS profile (Fig. 3c), the scattering intensity decayed gradually at low q, suggesting the presence of large or soft nanostructures.³¹

Fig. 3 Cryo-TEM images of dopasomes: (a) non-cross-linked and (b) cross-linked. (c) SAXS profile of the cross-linked dopasome (2 mM).

A broad peak near q ≈ 1.7 nm⁻¹ indicated the presence of a periodic structure. Although the profile could not be fitted with conventional form factor models, the results support the formation of partially ordered, membrane-like nanostructures.

One of the objectives of this research was to fabricate vesicles with high colloidal stability against surfactants. Fig. 4a shows the resistance of dopasomes to Triton X-100, a commonly used detergent, as evaluated by monitoring the light scattering intensity of the vesicles.²⁰ For comparison, we used conventional liposomes composed of dioleoyl phosphatidylcholine (DOPC), which has alkyl chains similar to those in DO-DOPA. Upon the addition of 5 equivalents of Triton X-100 relative to the lipid, the light scattering intensity of the DOPC liposomes dropped significantly, indicating disruption by Triton X-100 micelles. In contrast, the scattering intensity of dopasome remained unchanged even in the presence of 100 equivalents of Triton X-100. These results demonstrate that dopasomes possess substantially greater colloidal stability than conventional phospholipid-based liposomes. This enhanced stability is attributed to a combination of covalent cross-linking, π–π stacking interactions, and hydrogen bonding via catechol moieties. Furthermore, the ability of liposomes to retain their internal contents in the presence of surfactants was evaluated using a fluorescence recovery assay with calcein (Fig. 4b).³² Calcein, encapsulated in the hydrophilic core of the liposomes, is self-quenched at high concentrations; its fluorescence increases upon release following vesicle rupture. Remarkably, dopasomes exhibited minimal calcein leakage (∼5%) even in the presence of 100 equivalents of Triton X-100, whereas DOPC liposomes showed approximately 85% leakage with just 5 equivalents. This inhibition of drug leakage is attributed to the prevention of vesicle rupture, as demonstrated by the preceding experimental data.

Fig. 4 Structural stability of dopasomes against Triton X-100. (a) Light scattering intensities of dopasomes and DOPC liposomes as a function of added equivalents of Triton X-100 at 25 °C. (b) Cumulative release of calcein from dopasomes in the presence of Triton X-100: dopasome (blue circles); DOPC liposome (red circles).

Finally, to evaluate the suitability of the system for cancer therapy, we assessed its potential as a drug delivery carrier. First, the cytotoxicity of dopasomes was tested on two cell lines: the murine colon carcinoma cell line (Colon26) and the mouse fibroblast-like cell line (L929). As shown in Fig. S7, no apparent cytotoxicity was observed for either cell line at concentrations of 0–100 μM after 24 and 48 h of exposure. Additionally, cellular uptake and subcellular localization of the liposomes were examined using confocal laser scanning microscopy (CLSM). Typically, internalized liposomes are predominantly localized within lysosomes.³³ To assess this, rhodamine B-labeled liposomes were incubated with Colon26 cells for 4 and 24 h. As shown in Fig. S8, red fluorescence from the liposomes was clearly detected inside the cells. Merged images with LysoTracker-stained lysosomes revealed significant co-localization, indicating that the liposomes were primarily accumulated in lysosomal compartments following endocytosis.

We evaluated the potential of dopasomes as anticancer carriers using doxorubicin hydrochloride (DOX), a widely used chemotherapeutic agent. DOX has been extensively studied as a drug for liposomal formulation. Moreover, the first chemotherapeutic liposomal formulation has been approved under the product name Doxil.³⁴ As a reference treatment for this study, we employed liposomes composed of egg phosphatidylcholine (EggPC), a phospholipid used in DOX-loaded liposomal formulations.³⁵ EggPC has an unsaturated acyl chain that is similar to the dopamine-functionalized lipids found in dopasomes, thereby making it a suitable control for comparison. Since DOX can partition into both the lipid bilayer and the aqueous core, DOX-loaded dopasomes were prepared using the thin-film hydration method. To enhance dispersibility, a small amount of PEGylated lipid was included in the lipid composition. Given the sensitivity of DOX to heat and strongly basic conditions,³⁶ the pH of the resulting dispersion was adjusted to 8.5 using a mild base, sodium bicarbonate (NaHCO₃), to enable the cross-linking reaction. The dispersion was then incubated at room temperature for 48 h. The encapsulation efficiency of DOX in the dopasomes was calculated to be 96.3 ± 0.7%, whereas the DOX encapsulation efficiency in EggPC liposomes was 43 ± 11%. The higher dopasome loading efficiency is likely attributable to a more hydrophobic membrane environment that can accommodate DOX more effectively. The antitumor activity of DOX-loaded dopasomes was compared with that of free DOX using the Colon26 cell line. Both formulations were tested at two different incubation periods: 24 and 48 h. After 24 h, DOX-loaded dopasomes exhibited lower cytotoxicity than free DOX, likely due to a slower drug release rate (Fig. 5a). However, after 48 h, the cytotoxicity of the DOX-loaded dopasomes increased and was comparable to that of free DOX (Fig. 5b). In contrast, EggPC liposomes showed rapid onset of cytotoxicity, similar to that of free DOX. Moreover, this was true even at the 24 h time point, suggesting faster drug release from EggPC bilayers. To further assess potential off-target effects, we also evaluated cytotoxicity against the normal fibroblast cell line L929. As shown in Fig. S9, EggPC liposomes exhibited higher toxicity than both free DOX and dopasomes, whereas dopasomes showed relatively lower toxicity. These results suggest that the lower cytotoxicity of dopasomes in normal cells may be partly attributed to the higher overall toxicity of EggPC liposomes, in combination with the inherently slower drug release from dopasome membranes. As DOX exerts its antitumor effects primarily through DNA intercalation,³⁷ we investigated the intracellular release and nuclear localization of DOX using CLSM. The fluorescence of encapsulated DOX, typically quenched due to crystallization and lipid bilayer interactions,³⁸ was also significantly suppressed in our system (Fig. S10), but recovered upon drug release. As shown in Fig. 5c, the released DOX was predominantly localized in the cell nuclei after 24 h of incubation, suggesting successful nuclear delivery, which is critical for its antitumor activity.

Fig. 5 Antitumor activity of DOX-loaded dopasomes against Colon26 cells. Cytotoxicity of DOX-loaded dopasomes (blue), free DOX (red) and DOX-loaded EggPC liposomes (grey) after incubation with Colon26 cells for (a) 24 h and (b) 48 h. Data are presented as mean ± SD (n = 6); error bars represent standard deviation. (c) Subcellular distribution of DOX after incubation of Colon26 cells with DOX-loaded dopasomes for 24 h. Nuclei were stained with Hoechst 33 342 and samples were visualized by CLSM. Scale bar = 20 μm.

In conclusion, we developed a novel liposomal-based nanoparticle, termed a “dopasome”, using dopamine-derived lipids. Structural stability was achieved through cross-linking driven by the oxidative polymerization of the dopamine moieties. Notably, the dopasomes exhibited excellent stability and effectively retained cargo, even in the presence of excess surfactants. Their ability to preserve cargo integrity and enable efficient intracellular delivery underscores their potential as a promising platform for diverse biomedical applications.

Author contributions

All authors discussed the results and contributed to the final manuscripts. K. Y. and K. F. mainly conducted all the examinations. K. Y., R. K. and A. I. designed the experiments.

Conflicts of interest

There are no conflicts to declare.

Data availability

All relevant data are within the paper.

The supporting file includes the mechanism of polymerization, NMR and MS spectra, TEM images, DLS fitting data, and additional cytotoxicity assays. See DOI: https://doi.org/10.1039/d5tb01196e.

Acknowledgements

This research was financially supported by JSPS KAKENHI, Grant-in-Aid for Young Scientists (R. K., JP22K18196). JST ACT-X (R. K., JPMJAX2225). TEM and CLSM experiments were conducted in N-BARD in Hiroshima University. SAXS experiments were conducted at the BL40B2 beamline of SPring-8 under proposal number 2025A1078.

Notes and references

1. A. L. Nelson, E. Dhimolea and J. M. Reichert, Nat. Rev. Drug Discovery, 2010, 9, 767–774.
2. R. Tenchov, R. Bird, A. E. Curtze and Q. Zhou, ACS Nano, 2021, 15, 16982–17015.
3. Z. Yang, A. S. Malinick, T. Yang, W. Cheng and Q. Cheng, Sens. Actuators Rep., 2020, 2, 100023.
4. F. Mazur, M. Bally, B. Städler and R. Chandrawati, Adv. Colloid Interface Sci., 2017, 249, 88–99.
5. M. F. Mougenot, V. S. Pereira, A. L. R. Costa, M. Lancellotti, M. A. Porcionatto, J. C. da Silveira and L. G. de la Torre, Pharmaceutics, 2022, 14, 2008.
6. S. Pramanik, H. Tateishi-Karimata and N. Sugimoto, Nucleic Acids Res., 2014, 42, 12949–12959.
7. S. E. H. Piper, C. Casadevall, E. Reisner, T. A. Clarke, L. J. C. Jeuken, A. J. Gates and J. N. Butt, Angew. Chem., Int. Ed., 2022, 61, e202210572.
8. M. Godoy-Gallardo, C. Labay, M. M. T. Jansman, P. K. Ek and L. Hosta-Rigau, Adv. Healthcare Mater., 2017, 6, 1601190.
9. X. Hou, T. Zaks, R. Langer and Y. Dong, Nat. Rev. Mater., 2021, 6, 1078–1094.
10. Y. Chezy Barenholz, J. Control. Release, 2012, 160, 117–134.
11. R. Nisini, N. Poerio, S. Mariotti, F. De Santis and M. Fraziano, Front. Immunol., 2018, 9, 155.
12. D. Guimarães, A. Cavaco-Paulo and E. Nogueira, Int. J. Pharm., 2021, 601, 120571.
13. J. Nel, K. Elkhoury, É. Velot, A. Bianchi, S. Acherar, G. Francius, A. Tamayol, S. Grandemange and E. Arab-Tehrany, Bioact. Mater., 2023, 24, 401–437.
14. V. P. Torchilin, Nat. Rev. Drug Discovery, 2005, 4, 145–160.
15. J. Y. Yu, P. Chuesiang, G. H. Shin and H. J. Park, Pharmaceutics, 2021, 13, 1023.
16. M. Grit and D. J. A. Crommelin, Chem. Phys. Lipids, 1993, 64, 3–18.
17. V. G. S. Sainaga Jyothi, R. Bulusu, B. Venkata Krishna Rao, M. Pranothi, S. Banda, P. K. Bolla and N. Kommineni, Int. J. Pharm., 2022, 624, 122022.
18. R. T. Poon and N. Borys, Expert Opin. Pharmacother., 2009, 10, 333–343.
19. P. Nakhaei, R. Margiana, D. O. Bokov, W. K. Abdelbasset, M. A. Jadidi Kouhbanani, R. S. Varma, F. Marofi, M. Jarahian and N. Beheshtkhoo, Front. Bioeng. Biotechnol., 2021, 9, 705886.
20. K. Katagiri, M. Hashizume, K. Ariga, T. Terashima and J. Kikuchi, Chem. – Eur. J., 2007, 13, 5272–5281.
21. C. E. Smith and H. Kong, Langmuir, 2014, 30, 3697–3704.
22. Y. Fukui and K. Fujimoto, Chem. Mater., 2011, 23, 4701–4708.
23. Y. Liu, K. Ai and L. Lu, Chem. Rev., 2014, 114, 5057–5115.
24. R. Mrówczyński, R. Markiewicz and J. Liebscher, Polym. Int., 2016, 65, 1288–1299.
25. A. Gouda, O. S. Sakr, M. Nasr and O. Sammour, J. Drug Delivery Sci. Technol., 2021, 61, 102174.
26. A. D. Bangham and R. W. Horne, J. Mol. Biol., 1964, 8, 660–668.
27. M. L. Alfieri, T. Weil, D. Y. W. Ng and V. Ball, Adv. Colloid Interface Sci., 2022, 305, 102689.
28. D. K. Yeon, S. Ko, S. Jeong, S. P. Hong, S. M. Kang and W. K. Cho, Langmuir, 2018, 35, 1227–1234.
29. Q. Wei, F. Zhang, J. Li, B. Li and C. Zhao, Polym. Chem., 2010, 1, 1430–1433.
30. P. Nakhaei, R. Margiana, D. O. Bokov, W. K. Abdelbasset, M. A. Jadidi Kouhbanani, R. S. Varma, F. Marofi, M. Jarahian and N. Beheshtkhoo, Front. Bioeng. Biotechnol., 2021, 9, 705886.
31. G. Mahmoud, J. Jedelská, B. Strehlow and U. Bakowsky, Eur. J. Pharm. Biopharm., 2015, 95, 88–98.
32. Y. Yang, X. Liu, W. Ma, Q. Xu, G. Chen, Y. Wang, H. Xiao, N. Li, X. Li. Liang, M. Yu and Z. Yu, Biomaterials, 2021, 265, 120456.
33. G. Lin, H. Zhang and L. Huang, Mol. Pharm., 2015, 12, 314–321.
34. P. K. Working and A. D. Dayan, Hum. Exp. Toxicol., 1996, 15, 751–7856.
35. H. Farzaneh, M. E. Nik, M. Mashreghi, Z. Saberi, M. R. Jaafari and M. Teymouri, Int. J. Pharm., 2018, 551, 300–308.
36. J. H. Beijnen, O. A. G. J. van der Houwen and W. J. M. Underberg, Int. J. Pharm., 1986, 32, 123–131.
37. J. H. Lee, T. Tachibana, K. Yamana, R. Kawasaki and A. Yabuki, Langmuir, 2021, 37, 11269–11275.
38. Y. Tsukioka, Y. Matsumura, T. Hamaguchi, H. Koike, F. Moriyasu and T. Kakizoe, Cancer Sci., 2002, 93, 1145.

---

Footnote

† These authors contributed on this work equally.

---