Modern nanoparticles useful for applications in nanomedicine are often designed to respond to a multiplicity of tasks (Lammers et al., 2011; Biffi et al., 2015). By always keeping in mind the need for the lowest possible cytotoxicity, the versatility of these nanocarriers toward different administration routes and the possibility of modifying their surface via functionalization/decoration with targeting agents and/or fluorescent dyes are additional examples of desired qualities. In this arena, amphiphilic lipids are considered suitable building blocks (Aleandri and Mezzenga, 2020; Murgia et al., 2020), as they fulfill these prerogatives, simultaneously offering high biodegradability and showing rich polymorphism in water (Sahdev et al., 2014).

Among the various lipid-based self-assembled nanoparticles, lamellar phase dispersions, composed of a lipid bilayer enclosing a water compartment and being known as liposomes or vesicles, are the first and most widely studied, as they resemble the natural structure of cell membranes. Liposomes are largely proposed as nanocarriers for nanomedicine applications, and the first FDA-approved nanodrug was indeed a liposomal formulation (Doxil) (Barenholz, 2012), while scale-up, stability, cost, and reproducibility issues have now been addressed (Crommelin et al., 2020).

The non-lamellar counterparts of lamellar dispersions, namely, cubosomes (Gustafsson et al., 1997) and hexosomes (Yaghmur and Glatter, 2009), are now under investigation and often proposed as a valid alternative. Moreover, in recent years, the “melted” cubic phase known as the sponge phase (Angelov et al., 2011; Zhai et al., 2020a) has gained much interest since it can accommodate quite large macromolecules (e.g., proteins and enzymes) and can be dispersed in water similarly to cubosomes and hexosomes to form sponge nanoparticles, known as spongosomes (Chen et al., 2015; Valldeperas et al., 2019). Figure 1 reports the representation and morphology by an electron microscopy connection for the lipid nanoparticles under discussion.

This review focuses on different aspects of these lipid nanoparticles, cubosomes, hexosomes, and spongosomes, highlighting the advances in physicochemical characterization and biological performance. The most significant findings and studies for the biennium 2020–2022 are presented.

In comparison to the original liquid crystalline phases, cubosome, hexosome, and spongosome dispersions are diluted, non-viscous aqueous materials with an internal nanostructure suitable for nanomedicine applications.

They can usually be obtained via a high-energy emulsification procedure (Murgia et al., 2020), such as ultrasonication, high-pressure homogenization, or high-speed shearing, from their bulk phase or by adding a hydrotrope, an amphiphile incapable of displaying phase behavior by itself, thus providing a salting-in effect on the building blocks otherwise insoluble in water (Barriga et al., 2019; Murgia et al., 2020).

Larsson et al. reported the first case of bicontinuous inverse cubic dispersion in the late 1990s (Longley and McIntosh, 1983; Gustafsson et al., 1997), opening the path to the history and application of cubosomes. As for the lamellar phase, the reconstruction on the nanoscale as dispersion does not affect the inner structure, and cubosomes can present bicontinuous inverse cubic phases of symmetry, where Im3m, Pn3m, and Ia3d are their original bulk phases (Aleandri and Mezzenga, 2020). Generally, two sets of non-connected water channels are formed by a tridimensional arrangement of a lipid bilayer with an average zero curvature, yielding a porous material suitable for the confinement of biomolecules. The L3 or sponge phase is considered an intermediate mesophase between the lamellar and cubic phase, a “melted” cubic phase, in which it has a similar short-range structure of interconnected water channels, but with no long-range order. Spongosomes are obtained by the dispersion of the L3 phase and have been studied extensively by the group of Nylander, especially with regard to how to produce them from bulk phases with low-energy methods (Gilbert et al., 2019; Valldeperas et al., 2019).

Similar to the bicontinuous inverse cubic and sponge phases, the inverse hexagonal bulk phase can be fragmented into nanoparticles called hexosomes. These are composed of cylindrical inverse micelles closely packed in a hexagonal array, providing a network of water channels. Usually, they are obtained using the very same procedure and building blocks that should lead to cubosomes but with the addition of surfactants characterized by small head groups and long tails that affect the curvature of the system making it more negative (Salentinig et al., 2010).

Glycerol monooleate (GMO; monoolein) and phytantriol (PHYT) are the most common lipids employed for the preparation of cubosomes and hexosomes (Zhai et al., 2019; Murgia et al., 2020), and their phase behavior was thoroughly studied. The phase sequence with an increasing water content includes an inverse micellar, a lamellar, and two inverse bicontinuous cubic phases, the Schoen double gyroid (Ia3d space group) and the Schwarz double diamond (Pn3m space group) (Mezzenga et al., 2019). In addition, an inverse hexagonal phase appears at high temperatures.

Scientific interest relies on the fact that above a certain percentage of solvent and water, Pn3m or inverse hexagonal phases can coexist separately (liquid crystal + water excess), indicating the maximum solubilization capacity of that phase for the solvent (Kulkarni et al., 2011).

The sponge phase can be obtained with monoglycerides by adding organic solvents as propylene glycol, polyethylene glycol, or ethanol that disrupt the short-range cubic network, yielding the peculiar melted phase (Angelov et al., 2011). Recently, diglycerol monooleate (DGMO) and a mixture of mono-, di-, and tri-glycerides (Capmul GMO-50) were employed as the main components of the sponge nanoparticle formulation, and its phase behavior was reported (Valldeperas et al., 2016). Figure 2 summarizes the most common components for cubosome, hexosome, and sponge nanoparticles, also referring to the conventional stabilizers employed in their formulation.

Several studies also appointed polymers as suitable materials for cubosome formulations (Blanazs et al., 2009; Chen and Li, 2021). Ha et al showed how dendritic block copolymers can yield cubosomes by tuning the molecular geometry of the amphiphile (Ha et al., 2020). Chen et al. employed fluorescently labeled PEG-b-PTPEMA as the material for preparing cubosomes and hexosomes with aggregation-induced emissions (Ptissam, 2021). Mesoporous cubosomes (average water channel diameter of ca. 45 nm) formulated using a soft-template strategy were also tested for the controllable encapsulation of albumin (Wu et al., 2022); the tunable surface charge of this carrier allowed the selective adsorption of only albumin, and its release was pH dependent.

Dispersing a lamellar phase in water to obtain vesicles is an easy task. Indeed, the ability of the lamellae to bend into a closed structure that avoids contact between the hydrophobic domain and bulk water permits the breaking of the bulk phase easily. On the contrary, the dispersions of inverse hexagonal and bicontinuous cubic phases are more difficult given the boundary conditions imposed on fragmented crystals (Kulkarni et al., 2011; Murgia et al., 2013; Barriga et al., 2019; Aleandri and Mezzenga, 2020). These aggregates possess limited colloidal stability in aqueous dispersions, and they are prone to flocculation. Therefore, an additional surfactant is required to facilitate the dispersion of these phases, and it acts as a stabilizer.

The non-ionic tri-block copolymers that present poly(ethylene oxide)–poly(propylene oxide)–poly(ethylene oxide) moieties, known as Pluronics, are historically chosen as stabilizers for cubosomes and hexosomes (Meli et al., 2015; Akhlaghi et al., 2016). The hydrophilic corona surrounding the nanoparticles provides steric stabilization, whereas the polymer hydrophobic portion (the poly(propylene oxide) moiety) interacts with the lipid bilayer, partially anchoring to it. To ensure the functionalization of the stabilizer to impart targeting, therapeutical, or imaging properties, poly(propylene oxide) ends can be modified. Our group has been quite active in formulating cubosomes with modified Pluronics and new stabilizers. We investigated the functionalization of both Pluronic F108 and F127 by coupling folic acid and fluorophores, formulating cubosomes and hexosomes suitable for theranostic applications (Caltagirone et al., 2015; Meli et al., 2015; Meli et al., 2017). Moreover, we conjugated a photosensitizer to Pluronic F108 (Jenni et al., 2020), in order to formulate cubosomes suitable for photodynamic therapy.

The design of thermoresponsive nanoparticles can be achieved by the inclusion of thermoresponsive polymers as shown by Balestri et al. They used NIPAM-based block copolymers as stabilizers for GMO cubosomes, indicating that the polymer phase transition is preserved even after inclusion into the formulation (Balestri et al., 2022). The inclusion of the polymer in the formulation yielded Im3m at a low polymer content with a typical morphology of cubosomes, as shown by cryo-TEM investigations. When approaching the polymer critical temperature, the water channel radius of the Im3m phase decreases, stabilizing cubosomes by NIPAM block copolymers suitable for thermoresponsive drug delivery. Notably, increasing the concentration of the polymer in cubosomes induced a phase transition toward vesicles, as already seen in formulations of cubosomes stabilized by Pluronic F127 (Gustafsson et al., 1997).

Although poloxamers are largely employed as stabilizers for lipid nanoparticles, they are not biodegradable in vivo and can exert a significant cytotoxic effect, which is concentration dependent (Azmi et al., 2016). For this reason, investigations on new stabilizers with a biodegradable moiety have been conducted in the past years. A Pluronic F127 analog was synthesized based on polyphosphoesters (PPEs) and used to stabilize GMO cubosomes (Fornasier et al., 2020). The study highlighted that even at high concentrations, PPEs did not affect the inner structure of nanoparticles, and it improved the stability of the formulation.

The biodegradability of natural materials can also be exploited in this regard. Hemicellulose, an abundant material discarded by the wood industry, was still applied as a stabilizer for GMO cubosomes (Naidjonoka et al., 2021). Two purified hemicellulosic extracts were tested, which showed that the content of lignin plays a crucial role in the stabilization of the water–oil interface.

Indeed, non-polymeric stabilizers have been suggested in the past years. A mixture of propylene glycol and phospholipids can ensure the stabilization of GMO cubosomes without any polymeric material being added to the formulation (Bazylińska et al., 2018). Even biosurfactants such as bile salts are able to stabilize the interface of hexosomes suitable for the topical administration of antioxidants (Fornasier et al., 2021).

Tween 80 has been employed mostly in cubosomes suitable for crossing the blood–brain barrier. A recent study by Boyd and coauthors showed that PHYT cubosomes coated by Tween 80 managed to significantly increase the uptake of the model drug rhodamine B in zebrafish as compared to a control suspension of the drug or cubosomes stabilized with Pluronics (Azhari et al., 2021).

Linoleylethanolamide-based cubosomes stabilized with either Pluronic F68, F127, and Tween 80 greatly influenced the size, inner structure, and stability, as the interaction between the stabilizer and the lipid rules the physicochemical properties of nanoparticles. Interestingly, the authors showed that such stabilizers did not impact their cytotoxic effect on cells (Mohammad et al., 2020), even though this effect was highlighted in GMO-based cubosomes stabilized by the same polymers. This finding strongly suggests that the stabilizer–lipid interaction of the material composing nanoparticles plays a significant role in their biological impact.

Scattering and microscopy methods have always been the conventional approaches to characterizing lipid nanoparticles. Given the increased interest in the field and the variety of functionalities that were added to the systems, new different approaches can be employed to investigate the nanoparticle features at the physicochemical and biological level, depending on the aim and application of these formulations.

Mainly, such techniques can be divided into three categories (a schematic flow scheme is reported in Figure 3):

• Bulk techniques, which aim to assess the bulk properties of aggregates such as the size, morphology, structure, and colloidal stability

Spectroscopies such as nuclear magnetic resonance (NMR), UV–Vis, fluorescence, Raman, and Fourier transform infrared (FT-IR) can improve the knowledge of the local structure or the orientation of specific molecules in the lipid bilayer.

The figures obtained by bulk and surface characterization can give a hint about the possible biological impact of aggregates in vitro or in vivo. Indeed, the correlation between all these complementary techniques is crucial to developing the most efficient formulation for drug delivery or imaging purposes.

The biocompatibility and biodegradability of lipids have attracted great attention in the pharmaceutical field since they can be employed as emulsifiers, solubilizers, and, especially, building blocks for drug delivery platforms (Tan et al., 2022). Indeed, the use of lipid nanoparticles as drug carriers is less invasive in comparison to other methods, and it allows an increase in the selectivity and effectiveness of the therapy (Barriga et al., 2019; Lai et al., 2020; Murgia et al., 2020).

The route chosen directly affects the drug's bioavailability (Suk et al., 2016), thereby determining the onset and duration of pharmacological effects. Such routes are influenced by various factors, e.g., state and age of the patient, nature of the drug, first-pass metabolism, and convenience of application.

Depending on these parameters, administration routes can be divided into local (or topical) and systemic. The former is the simplest mode of administration; the formulation containing the drug is applied where the desired action is required. On the other hand, the systemic administration of a drug or a formulation is a more complex route, since the enteral (via the gastrointestinal system) or the parenteral (bypassing the gastrointestinal system) mode can be adopted.

The enteral route involves drug adsorption via the gastrointestinal tract and includes oral, sublingual, and rectal administration, whereas the parenteral route bypasses the gastrointestinal system, and it exploits injection (e.g., intravenous, intramuscular, and subcutaneous), inhalation, or transdermal delivery. Indeed, each route presents advantages and drawbacks depending on the nature of the drug and the biodistribution necessary to achieve a therapeutic effect. Figure 4 summarizes the administration routes of lipid nanoparticles discussed in the following sections.

Up to date, lipid nanoparticle formulations exploit both local (Lai et al., 2020) and systemic administration routes (Mulet et al., 2013).

Since cubosomes enable the encapsulation of poorly water-soluble drugs and can be engineered with specific functionalities, they are proposed as suitable candidates as nanomedicines. The variety of applications of such formulations shows their potency and versatility. Moreover, given their bigger water channel diameters, hexosomes and sponge phase nanoparticles can ensure the immobilization of large macromolecules (Meli et al., 2015; Meli et al., 2017; Barriga et al., 2019; Gilbert et al., 2019; Murgia et al., 2020). Even though hexosomes and spongosomes have been studied for more than 20 years, few papers have been published in comparison with cubosomes. For instance, most of the research papers devoted to spongosomes are related to their physicochemical aspects, while their biological impact remains poorly understood, and studies on their possible use as nanomedicines are missing. Furthermore, their surface/interfacial properties lack proper investigation. While cubosomes were investigated early under this aspect (Larsson, 1983; Hyde et al., 1984; Neto et al., 1999), hexosomes and sponge nanoparticles surface are yet to be probed in a systematic manner on bare surfaces (hydrophilic and hydrophobic) and on model membranes. This appears as one of the most important missing links between the physicochemical features and the biological/medical applications of these nanoparticles. In general, as underlined by many reviews (Biffi et al., 2017; Tran et al., 2017; Zhai et al., 2020b), only a limited number of research papers have appeared presenting in vitro and in vivo investigations on non-lamellar nanoparticles, and this is, especially, true for hexosomes and spongosomes. As a consequence, important aspects, such as the nanostructure relationship with nanoparticles’ cellular uptake, remain rather obscure and yet to be fully clarified (Bor et al., 2022b), not allowing the drawing of fundamental conclusions on the fate of these nanoparticles once administered and on their effect on our bodies. In addition, non-lamellar nanoparticles are often proposed as possible alternatives to liposomes, so the question arises in which cases they may be superior. Due to the higher lipid volume convoluted in the space in non-lamellar nanoparticles, they certainly allow greater encapsulation of drugs per unit of volume, but a control sample of vesicles is usually missing as a comparison in most of the studies, making it harder to assess the best option of treatment between lamellar or non-lamellar formulations.

All the nanoparticles discussed in this review, either lamellar, cubic, sponge, or hexagonal, exhibit peculiar aspects that place them, in terms of use, in the pharmaceutical area of controlled drug release for various pathologies. Definitely, each of them has intrinsic advantages and disadvantages, ultimately related to the physics and the chemistry of the molecules used for their preparation. A clarifying example could be the better performance shown by the non-hydrolyzable PHYT compared to GMO in oral drug administration (see Section 6.3), although it is commonly found more cytotoxic than GMO. However, rather than simple limitations, such disadvantages should be considered florid fields of research where investigations are needed to clarify the biological role of the components of these nanoparticles, which is well beyond the physicochemical properties of the nanostructure they originate when dispersed in water.