MPS blockade with liposomes controls pharmacokinetics of nanoparticles in a size-dependent manner

We prepared liposomes composed of L-α-PC and Chol, common components of clinically approved lipid-based therapeutics [23]. While non-PEGylated PC/Chol liposomes are used for drug delivery and tumor treatment [24], liver and spleen remain main organs of their accumulation in the body [25, 26]. Pre-dosing with liposomes of PC/Chol composition has been reported to deplete the MPS activity by occupation of the hepatic and splenic phagocytes [27]. Moreover, the high biocompatibility and cost-effectiveness of these lipids make them attractive for preparing MPS-blocking liposomes, especially considering the typically high doses required for an effective blocking effect in vivo (∼100 mg kg⁻¹ of body weight) [12, 14, 28].

Nanosized PC/Chol (1/1 molar ratio) liposomes were synthesized via a thin-film hydration method, followed by membrane extrusion-based size fractionation. We produced liposomes larger 100 nm to avoid renal excretion, while liver and spleen uptake dominates for liposomes of this size [25, 29].

Three size fractions of liposomes were employed to induce MPS-blockade, with mean hydrodynamic diameters of 160 nm (small, LP_S), 210 nm (medium, LP_M) and 370 nm (large, LP_L), each having low polydispersity (PDI < 0.25). The size distribution and corresponding dispersity parameters are presented in figure 1(a) and supplementary table 1. All formulations demonstrated stability of the mean hydrodynamic size in PBS over a period of 8 h (figure 1(b)). The size of all liposomal fractions was higher than endothelial fenestrae in liver, preventing these nanoparticles from entering the space of Disse and increasing the probability of interaction with Kupffer cells [30]. Nanosized PC/Chol liposomes possessed a strong negative ζ-potential in 10 mM NaCl (LP_S: −43 ± 9 mV, LP_M: −30 ± 10 mV, LP_L: −34 ± 7 mV), indicating similar surface properties and high colloidal stability of nanoparticles (figure 1(c)). The negative surface charge of liposomes composed of zwitterionic PC lipid and Chol is consistent with other studies [31] and can be explained by high affinity of the PC chains to anions in salt solution.

Zoom In Zoom Out Reset image size

Macrophages represent a target cell population in MPS-blockade approaches, as they are responsible for elimination of major fraction of exogeneous nanomaterials in vivo [32, 33]. Therefore, it is important to access cytotoxicity of liposomal blockers towards these cells. For this aim, we utilized the commonly used murine macrophage RAW264.7 line and measured their viability by MTT test. The experiment was conducted under high concentrations of liposomes (up to 500 μg ml⁻¹), which are expected to saturate macrophage uptake function in vitro [21, 28].

For all three types of prepared liposomes, a gradual decrease in viability was observed with an increase in particle concentration (figure 1(d)). No significant difference in cytotoxicity was found between liposomes of the studied size ranges. At a concentration of 50 μg ml⁻¹, the viability was approximately 85%, and after a 10-fold dose increase it plateaued at the 70%–73% level. The mild toxicity of liposomes may be explained by the presence of remnants of organic solvents in the particles structure [34]. Nevertheless, nanosized liposomes demonstrate a low level of cytotoxicity to macrophages, compared to many other nanoparticles, used for the MPS-blockade. For example, silica (IC₅₀ < 600 μg ml⁻¹ [35]) and polystyrene beads (IC₅₀ < 500 μg ml⁻¹ [36]) are found to be toxic at high doses. For the designed liposomes, the IC₅₀ was higher than 500 μg ml⁻¹ and could not be determined from the graphs. Thus, the PC/Chol liposomes of the 100–500 nm size range have low toxicity to macrophages even at high doses.

Blocking nanoparticles are designed to inhibit endocytosis of macrophages via temporal occupation and saturation of their surface receptors [12, 28]. Therefore, we compared uptake of MNPs in macrophages after their blockade with LP_S, LP_M and LP_L formulations. MNPs are commonly applied in nanomedicine for therapeutic and diagnostic purposes, but their rapid elimination from circulation by professional phagocytes remains a challenging problem [37]. In this study, we used multicore magnetic iron oxide nanoparticles stabilized by glucuronic acid (FluidMag-ARA) as model of therapeutic nanoparticles. They have hydrodynamic diameter of (160 ± 40) nm and ζ-potential of (−34 ± 7) mV (figures S1(a) and (b)). Nanoparticles of similar size are widely employed for drug delivery due to their sufficient payload loading, avoidance of renal filtration, and potential for their retention in tumors via EPR effect [38]. Nevertheless, rapid macrophage uptake hampers their performance. Superparamagnetic properties of FluidMag-ARA nanoparticles allow to quantify them in samples and tissues by the magnetic particle quantification (MPQ) technique [39].

The liposomal blockade in vitro was performed using RAW264.7 macrophages. Liposomes were added to the cells at a concentration of 500 μg ml⁻¹ and incubated for 24 h to saturate endocytic pathways. The cells were then washed with PBS to remove the unbound liposomes, and MNPs labeled with a lipophilic fluorescent dye (FluidMAG-ARA/R) were added to the cells at 25 μg ml⁻¹ dose and co-incubated for 24 h. The fluorescent MNPs had similar size and ζ-potential distributions as non-labeled ones (figures S2(a) and (b)). Cellular uptake of MNPs was accessed by flow cytometry (figure 2(a)) and compared by the distribution of cell fluorescence intensity (figure 2(b)). The blockade of macrophages was manifested when cells were incubated with LP_M and LP_L formulations. Flow cytometry histograms shifted to the smaller fluorescence values 24 h after the blockade (figure 2(a)), and the calculated uptake of MNPs decreased by 45% and 25% after the saturation with LP_M and LP_L, respectively (figure 2(b)). However, after the treatment of macrophages with small-sized LP_S liposomes, no significant change in the uptake of MNPs was observed.

Zoom In Zoom Out Reset image size

Observed difference may be explained by the kinetics of liposomal uptake by cells. To test this assumption, we synthesized liposomal formulations of similar size distributions labeled with a fluorescent 1,1'-dioctadecyl-3,3,3',3'-tetramethylindodicarbocyanine (DiD) dye (see characterization at table S1 and figure S3). Labeled liposomes were added to RAW264.7 macrophages at an equal lipid dose (500 μg ml⁻¹), incubated for 1 h or 24 h, and analyzed by flow cytometry for DiD fluorescence. Macrophages showed similar uptake of all types of liposomes at 1 h time-point (figure 2(c)). However, after incubation for 24 h, DiD-LP_M and DiD-LP_L demonstrated higher cellular uptake with a 20% and 110% increase in median fluorescence intensity compared to DiD-LP_S, respectively (figures 2(c) and (d)). Low accumulation of LP_S in cells was additionally confirmed by fluorescence microscopy (figure S4). Significant difference in the blockade efficiency by LP_S and LP_M formulations cannot be explained by the liposomal uptake alone, and other factors, such as the speed of receptor recycling and rate of endocytosis, are likely involved.

Pharmacokinetics of medium-sized liposomes was determined to understand time delay between blocking and magnetic particle administration. For this aim, fluorescent DiD-labeled liposomes of intermediate size (DiD-LP_M) were administered in BALB/c mice at a dose of 3 mg via retro-orbital injection. Their blood availability was evaluated by measuring the fluorescence of plasma samples (figure 3(a)). The pharmacokinetics exhibited a biexponential profile with half-lives of (13.2 ± 2.5) min and (3.7 ± 2.3) h for the rapid and slow phases, respectively. Within the first hour of circulation, the signal of fluorescence in blood dropped 2.2-fold. After 1 h, spleen and liver demonstrated the strongest fluorescence among organs (figure 3(b)), confirming effective capture of liposomes by resident macrophages. For the next 6 h, the fluorescence of spleen decreased by 55%, while the liver retained the same signal. From this study, we conclude that 1 h is a sufficient interval for MPS blockage in vivo with PC/Chol liposomes, considering that the liver has substantial uptake of liposomes, and part of the blocking materials remains in the bloodstream to compete with exogenous nanoparticles for recognition [40].

Zoom In Zoom Out Reset image size

To evaluate the circulation kinetics of MNPs in the bloodstream and their biodistribution in mice, we utilized the MPQ technique, described in detail by Zelepukin et al [39]. The MPQ method quantitatively and non-invasively measures concentration of non-linear magnetics, such as superparamagnetic nanoparticles, in the blood vessels of mouse tail. Linear paramagnetic and diamagnetic materials, such as animal tissue and iron in the blood, do not contribute to the measured signal, setting the limit of detection of nanoparticles at the nanogram level. This method enables real-time tracking of pharmacokinetics of magnetic particles with high sensitivity.

Macrophage blockade was induced by injection of 3 mg of liposomes into the orbital venous sinus followed by administration of MNPs after 1 h via the same route. MNPs blood circulation kinetics and biodistribution at 3 h were analyzed (figure 4(a)). In the absence of blockade, the elimination kinetics of MNPs exponentially decreased and 90% of the ID was cleared within 7 min (figure 4(b)). MPS blockade with liposomes significantly prolonged the circulation of nanoparticles. The half-life increased from (1.2 ± 0.1) min to (6.2 ± 2.1) min, (7.0 ± 2.7) min and (7.5 ± 2.8) min using LP_S, LP_M and LP_L blockers, respectively. Other common parameters of pharmacokinetics analysis, such as area under curve and Cl demonstrated the beneficial effect of blockade on the prolongation of MNP circulation (table S2). Furthermore, the highest values were observed for LP_L blockers, as consistent with the in vitro study. Nevertheless, LP_S nanoparticles, which were completely inefficient in vitro, showed in vivo blockade effect similar in magnitude to LP_M nanoparticles.

Zoom In Zoom Out Reset image size

This difference can be explained by the involvement of other saturation mechanism in complex in vivo environment. For instance, depletion of opsonins in serum by liposomes can additionally decrease the subsequent uptake of MNPs [12, 41]. In addition, small liposomes generally demonstrate longer blood circulation time and can better compete for binding to macrophage surface during MNPs circulation [40].

The biodistribution of MNPs after liposomal blockade is presented in figures 4(c)–(h) and S5, raw data are summarized in table S3. Without the blockade induction, 99% of the ID accumulated in liver and spleen, two major organs of MPS. Pre-treatment with liposomes significantly decreased total accumulation in these organs by 8.5%, 10%, and 6.6% ID for LP_L, LP_M, and LP_S liposomes, respectively. Low effect of LP_S liposomes on depleting activity of the liver and spleen phagocytes corroborates with the observation on the lower uptake by macrophages in vitro. The delivery of MNPs to the lungs increased in order of magnitude after the blockade with all three blocker types, but the effect was most pronounced for the largest LP_L liposomes. In addition, the blockade with LP_M and LP_L liposomes induced a 3-fold increase in delivery to the kidneys and 4.3- and 3.7-fold higher uptake in the heart, respectively. At the same time, LP_S liposomes did not change significantly the uptake of MNPs in these organs compared to the non-blocking condition (table S3).

Importantly, even after the liposomal blockade, liver remains the main organ eliminating nanoparticles from the bloodstream. Kupffer cells in the liver constitute 80%–90% of the total macrophage population in the body [32, 42], and they reside in sinusoids with slow blood flow, which increases the probability of nanoparticle capture [1]. Tavares et al reported that even after elimination of 75% of liver macrophages by clodronate liposomes, the liver continues to be the second organ of nanoparticle uptake after the spleen [43]. MPS blockade with liposomes is a mild method and only partially and temporally reduces the activity of the liver macrophages, so magnetic particles still accumulate in this organ in high quantities.

Nevertheless, since liver serves as the first barrier encountered by nanoparticles after the injection, MPS blockade induces a 'spillover effect' of nanoparticle redistribution to other macrophage-reach tissues like spleen and bone marrow [44]. The improvement of the splenic uptake of nanomedicines is currently desirable for gene delivery and immunization applications, as well as for direct treatment of spleen pathologies [45, 46]. Therefore, we further examined difference in distribution between liver and spleen dependent on blocking liposome size (figures 4(c), (d) and (h)). The average liver accumulation of MNPs was observed to decline gradually with the increase in liposome size, from 78.6% ID (LP_S) to 71.6% ID (LP_L). On the other hand, the average quantity of MNPs reaching spleen was highest for LP_L formulation (18.5% ID), with no significant difference observed between blockade by LP_M and LPs formulations (∼14% ID). The ratio of liver to spleen accumulation of magnetic particles decreased from 7.9 in animals without blockade to 3.9 in mice with blockade by LP_L liposomes. Decreased liver-to-spleen delivery ratios were established in LP_S and LP_M groups, with corresponding values of 5.8 and 5.3. Our observations suggest that the increase of mean liposome size towards 370 nm provides higher splenic uptake of post-injected therapeutic nanoparticles by hindering their capture by liver. It corroborates with recent meta-analysis of MPS blockade studies, where the spillover effect was observed to increase for nanomaterials larger 200 nm [6]. However, further size increase of blocking liposomes does not guarantee an elevation of liver-evasion capacity, since the hepatic uptake dependence for liposomes typically demonstrates saturation at 250–400 nm mean liposome sizes [36].

L-α-PC, Chol and methanol were obtained from Sigma-Aldrich, USA. Chloroform and dimethyl sulfoxide (DMSO) were obtained from Chimmed, Russia. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) and 1,1-dioctadecyl-3,3,3,3-tetramethylindodicarbocyanine (DiD) were obtained from Thermo Fisher Scientific, USA. FluidMAG-ARA100 and FluidMAG-ARA100/R nanoparticles were purchased from Chemicell GmbH, Germany. L-glutamine, penicillin, streptomycin, and amphotericin B were purchased from Gibco, USA.

56 mg of L-α-PC and 28 mg of Chol were dissolved in 5 ml of 6:1 chloroform/methanol solution and then dried under reduced pressure using a rotary evaporator at 40 °C and 90 rpm rotation speed. The obtained lipid film was hydrated with PBS (0.1 M, pH 7.4) in an ultrasonic bath. Microscale liposomes were collected by centrifugation at 100 g for 5 min. Then, liposomes were extruded through 400 nm, 200 nm and 100 nm polycarbonate membranes to obtain liposomes of the desired sizes.

To prepare fluorescent liposomes, DiD dye was added to chloroform/methanol mixture in a weight ratio of 0.01% of a total lipid mass. After extrusion, liposomes were washed from the unbound dye by centrifugation in spin columns with 7 kDa weight cut-off.

The size distribution and ζ-potential of liposomes and MNPs were characterised using Zetasizer Nano ZS device (Malvern, UK). The hydrodynamic size (diameter) of liposomes was determined from intensity size distribution. ζ-potential of liposomes and MNPs was measured in 10 mM NaCl. The Smoluchowski approximation was used for analysis.

RAW264.7 (ATCC TIB-71™) macrophage line were cultured in DMEM medium supplemented with 10% FBS, 2 mM L-glutamine and Gibco antibiotic–antimycotic solution. Cells were seeded in 96-well plates at density of 10⁴ cells per well and incubated at 37 °C under humidified atmosphere containing 5% CO₂ for 24 h before the experiments.

Macrophage cells were co-incubated with liposomes at final concentrations of 0.5, 0.1 and 0.05 mg ml⁻¹ in 100 μl of medium. After 24 h of incubation, the medium was removed and 100 μl of the MTT solution (0.5 g l⁻¹ in DMEM) was added to each well. The cells were incubated for 1 h at 37 °C, MTT solution was removed and 100 μl of DMSO was added to the wells. Then the plate was shaken in the dark until complete dissolution of formazan crystals. Cell viability was measured by optical density at 570 nm wavelength, using Infinite 1000 PRO (Tecan, Austria) microplate reader, and was normalized to the viability of untreated cells.

Liposomes at concentration of 0.5 mg ml⁻¹ were incubated with RAW264.7 macrophages for 24 h. Then, cells were washed with PBS and treated with 25 μg ml⁻¹ of FluidMAG-ARA100/R fluorescent nanoparticles in DMEM medium (excitation: 578 nm, emission: 613 nm). In 3 h, cells were washed 3 times with 1% BSA in PBS and analyzed by a Novocyte 3000 cytometer in the YL2 channel (excitation laser at 561 nm, emission filter at 615/20 nm). The uptake of fluorescent nanoparticles after liposomal blockade was determined by median intensity of cell fluorescence and was normalized to the uptake values in cells without blockade.

To visualize liposomal blockade of macrophages, DiD-labelled liposomes were added to the cells to a final concentration of 0.5 mg ml⁻¹. In 1 h or 24 h, the cells were washed with 1% BSA in PBS and analyzed by flow cytometry in the YL3 channel (excitation laser at 561 nm, emission filter at 660/20 nm), and by fluorescence microscopy on a DMI 6000B microscope (Leica, Germany) equipped with a DFC295 camera.

All experiments with animals were approved by the institutional animal care and use committee of Shemyakin–Ovchinnikov Institute of Bioorganic Chemistry Russian Academy of Sciences (IBCh RAS), protocol 367/2022 from 15.12.2022. The procedures were performed according to Institutional guidelines of IBCh RAS, based on European convention on protection of vertebrate animals (ETS 123 from 18.03.1986) and Russian animal legislation. Female BALB/c mice of 18–22 g (10–14 weeks old) were obtained from Pushchino animal facility (Pushchino, Russia) and maintained at the vivarium of IBCh RAS. Animals were maintained under 12 h dark-light cycle with a constant supply of the food and water. Before nanoparticles administration, mice were anesthetized with intraperitoneal injection of Zoletil/Rometar mixture at 53 and 2 mg kg⁻¹ doses, respectively.

To study the blood circulation of liposomes, 3 mg of DiD-tagged liposomes in 300 μl of PBS (pH 7.4) were injected into retro-orbital sinus of BALB/c mice. Blood samples (30 μl) were collected with heparin within the following 7 h via puncture of the opposite retro-orbital sinus and diluted with 120 μl of PBS. Blood plasma was separated by centrifugation of samples at 1000 g for 3 min. Liposome concentration in samples was measured by fluorescence intensity at 666 nm with excitation at 647 nm on an Infinite M1000 microplate reader.

To study the biodistribution of liposomes, mice were sacrificed by cervical dislocation and major organs were analyzed by averaged radiance efficiency of epifluorescence using IVIS Spectrum CT device (PerkinElmer, USA) with excitation at 647 nm.

3 mg of liposomes were injected into retro-orbital venous sinus of mice to induce macrophage blockade. Then, in 1 h, 300 μg of FluidMAG-ARA MNPs were injected into opposite retroorbital sinus. Magnetic particle pharmacokinetics was measured quantitatively by MPQ technique [39]. Briefly, mice tail was placed into a coil of the MPQ detector and gently fixed with a tape lent. Nanoparticle concentration in tail veins and arteries was measured with a frequency of 15.2 Hz.

Biodistribution of MNPs was measured ex vivo with MPQ technique 3 h after their administration. Mice were euthanized by cervical dislocation and major organs were extracted and placed into the coil of the MPQ detector. The measured magnetic signals were normalized to the total detected dose of nanoparticles in all extracted organs and presented in % of ID.