2017 Volume 40 Issue 7 Pages 1002-1009
In the active targeting of a drug delivery system (DDS), the density of the ligand on the functionalized liposome determines its affinity for binding to the target. To evaluate these densities on the surface of different sized liposomes, 4 liposomes with various diameters (188, 137, 70, 40 nm) were prepared and their surfaces were modified with fluorescently labeled ligand–lipid conjugates by the post-insertion method. Each liposomal mixture was fractionated into a series of fractions using size exclusion chromatography (SEC), and the resulting liposome fractions were precisely analyzed and the surface ligand densities calculated. The data collected using this methodology indicate that the density of the ligand on a particle is greatly dependent on the size of the liposome. This, in turn, indicates that smaller liposomes (75–40 nm) tend to possess higher densities. For developing active targeting systems, size and the density of the ligands are two important and independent factors that can affect the efficiency of a system as it relates to medical use.
Active targeting is at the forefront of liposomal design for the second generation of drug delivery system (DDS).1,2) It permits liposomes to be actively and specifically recognized at locations. As a result, developing suitable ligands to graft on the surfaces of liposomes has become a prime goal. That is, compared to the first generation, these ligand-modified liposomes are not restricted by being dependent on the enhanced permeability and retention (EPR) effect for delivering bioactive substances to tissues that have porous characteristics such as the liver, spleen or tumors.3,4) In contrast, they extend the possibility of targeting to other kinds of tissue with blood vessels that have non-fenestrated features, i.e., the blood brain barrier or normal tissues and cells.5–7) These ligands have the ability to permit liposomes, carrying bioactive materials to specific organelles, tissues, organs or pathological locations in the human body. Thus, a ligand that efficiently recognizes the target tissue is critical, and the density of the ligand on the surface of the liposome serves as an indicative index for assessing its binding affinity.8–12)
To define and assess surface ligand density is an important issue that needs to be addressed before liposomal applications can be undertaken.13–16) However, most studies have reported various liposomal bioactivities without providing a clear value that reflects surface ligand density. Instead, bioactivities are reported as the difference in the original lipid constituents of the liposome.17–32) To address this issue, a method for estimating and calculating these values is needed and the liposomal preparation should be standardized, since these factors affect the density of the ligand on the surface of a particle.33) Considering the clinical aspects of regulatory science (medicine), the characterization of liposomal surfaces guarantees the quality of a medical application. An accurate value for the surface ligand density provides an accurate assessment of the degree of binding affinity for a modified liposome.
Recent literature reports have placed great attention on the fact that liposomes with smaller sizes (diameters of around 40 nm) have a higher distribution and penetration capability, especially, in a tumor microenvironment, in terms of enhancing therapeutic efficiency.34,35) For developing the active targeting of DDS, small liposomes with attached-ligand assistance are being prepared. In order to easily and precisely monitor and detect the ligands, a fluorescent ligand was introduced on the particles. This permitted us to analyze the surface ligand densities on different sized liposomes and to determine how liposome size affects ligand density. In this study, liposomes of different sizes were prepared and purified and their surface ligand densities were then determined.36) The parameter, ‘incorporation ratio,’ is introduced to define this density.33) In an analysis of these ratios, our new discovery shows that when liposome size decreases, the incorporation ratio is not constant but rapidly increases. Since liposomes with smaller sizes promise to become more important in the near future, a significant underestimation of the incorporation ratio may be misleading and lead to erroneous conclusions regarding their bioactivities. Our methodology and this new discovery provide important information regarding the development of active targeting systems.
Non-hydrogenated egg phosphatidylcholine (EPC), N-[(3-maleimide-1-oxopropyl)aminopropyl polyethyleneglycol-carbamyl]distearoylphosphatidyl-ethanolamine (maleimide-PEG2000–DSPE, DSPE–020MA), and N-(carbonyl-methoxy polyethyleneglycol)-1,2-distearoyl-sn-glycero-3-phosphoethanolamine, (mPEG2000–DSPE) were purchased from NOF Corp. (Tokyo, Japan). Cholesterol (chol) and Sepharose® CL-4B were purchased from Sigma (St. Louis, MO, U.S.A.). 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt) (rhodamine-DOPE, rho lipid) was purchased from Avanti Polar Lipids Inc. (Alabaster, AL, U.S.A.). 5-Carboxyfluorescein (5-FAM) was purchased from TCI Corp. (Tokyo, Japan). 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Sephadex® G-25 was purchased from GE healthcare Corp. (Sweden). All reagents were without further purification before use.
Preparation of Inserted LiposomesThe plain liposomes were composed of EPC : chol : rho lipid (48 : 50 : 2). The lipids were dissolved in ethanol, chloroform was added and the solution was vigorously mixed. The solution was evaporated to form a lipid film, and hydrated in HEPES–NaOH (pH=7.4) for 10 min at room temperature. The hydration method was followed by a probe sonication technique. While the 300 µL, 15.05 mM (ensured by a 2% rho lipid UV absorbance, total lipids=4515 nmol) plain liposomes with 70 nm diameters were produced, the 5-FAM–PEG2000–DSPE conjugate lipid (5-FAM conj.) (see synthetic details in the supplementary data, Charts S1, S2) that was synthesized and purified in advance was then added (molar ratio 5-FAM conj. : plain liposomes lipids=1 : 16). The mixture was equally divided into 5 eppendorfs before undergoing the post-insertion procedure at the same incubation temperature 37°C but for 5 different insertion time intervals, 10, 30, 60, 120, and 180 min. This resulted in 5 liposomal mixtures with conjugate micelles prepared using different incubation times.
The inserted liposomal mixtures of other diameters 188, 137, and 40 nm around 15 mM were also prepared by the same process as above. The probe sonicator was adjusted and optimized by pulse power and time to produce liposomes with these different diameters (188, 137, 70, 40 nm). These 4 inserted liposomes with different diameters (188, 137, 70, 40 nm) were incubated at the same ratio of 5-FAM conj. : plain liposomes lipids 1 : 16, at a temperature of 37°C but the divided 5 eppendorfs of each diameter, which were subjected to different insertion time intervals resulted in 20 different liposomal mixtures that were ready for further separation.
Optimization and Calibration Curve of a Sepharose® CL-4B ColumnTo ensure that contaminating micelles were completely separated from the inserted liposomal mixtures, the separation conditions were optimized as follows: a 45 mL bed volume of Sepharose® CL-4B in a ϕ15×350 mm column was prepared. The sample was suspended in 250 µL and eluted by a HEPES–NaOH (pH=7) buffer. Before each separation, the column was washed with 200 mL of buffer. While the samples were mounted, the column was being eluted with a HEPES–NaOH (pH=7) buffer, and 1 mL fractions were collected by means of an autocollector for 40 s tubes after void volumes were eluted (the void volume was determined by the elution of Blue Dextran 2000).
To prepare a calibration curve, we used plain liposomes composed of EPC : chol : mPEG2000–DSPE : rho lipid (60 : 34 : 5 : 1). The hydration method followed by bath sonication was the applied. While 80 µL, 5.0 mM (ensured by 1% rho lipid based on UV absorbance) PEG liposomes were produced, they were applied to the same Sepharose® CL-4B column to obtain 40 s tubes that were used to prepare the curve. Particle size, size distribution, and zeta potentials were determined by dynamic laser light scattering (Malvern Zetasizer Spectrometer). Finally, a plot of liposomal diameter versus elution volume was prepared, as shown in Fig. S1 (see the supplementary materials for the SEC calibration details).
Separation of the Liposomes by Sepharose® CL-4B ChromatographyTo purify and separate each inserted liposomal mixture (as described in method “Preparation of Inserted Liposomes”) into different tubes, each mixture was passed through the Sepharose® CL-4B column (ϕ15×350 mm, as mentioned in method “Optimization and Calibration Curve of a Sepharose® CL-4B Column”) and eluted with HEPES–NaOH (pH=7) to produce pure inserted liposomes. Each sample in 125 µL was applied, and manipulated as described in method “Optimization and Calibration Curve of a Sepharose® CL-4B Column” (an example of a chromatogram is shown in Fig. S2). Particle size, size distribution, and zeta potential of the plain liposomes, the inserted liposomes, and the inserted liposomes after purification were determined using the same instrument as described above (see the physical data in Tables S1–S4 of the supplementary materials).
Calculation of Surface Ligand DensityFor preparing a HPLC standard analytical curve (area under the curve (AUC) vs. UV absorbance), covering the concentration range of all samples, a series of concentrations (2–50 µM, 100 µL) of 5-FAM conj. standard solutions (HEPES–NaOH, pH=7.4) were prepared. The analytical curve of rhodamine-DOPE also prepared in the same manner. Also, for preparing the UV-Vis standard analytical curves (nmol vs. UV absorbance) of 5-FAM conj. and rhodamine-DOPE the same procedure was used as mentioned above. The following values: λex=495 nm (in HEPES–NaOH pH=7.4), and ε=31000 were utilized for 5-FAM conj.; λex=570 nm (in HEPES–NaOH pH=7.4), and ε=142000 were utilized for rhodamine-DOPE. Each sample (70–99 µL), which was separated by SEC mentioned in method “Preparation of Inserted Liposomes” was then analyzed by injecting it into the HPLC to calculate molar amount of 5-FAM conj. and rhodamine-DOPE using the following conditions: C4 column, eluting by triethylammonium acetate (TEAA) buffer (100 mM, pH=7.4)/acetonitrile and gradient elution of 0–90% (acetonitrile) in 25 min at 40°C, λex=495 nm.
To calculate the 5-FAM conj. molar amount, the AUC of each sample was analyzed by HPLC. Calculating the amount of total lipids amount was based the 2% rho lipid molar amount contained in the liposomes, and the rho lipid AUC of each sample was analyzed by HPLC as well. Further applying the Eq. 2 in Chart 1, all incorporation ratios (surface ligand density) were obtained. The calculation of incorporation ratios for pre-inserted liposomes was based on that the conjugates are exposed outward/inward of the membrane with a theoretical ratio 55/45 (the calculation formula is mentioned in Results and Discussion “Incorporation Ratio of Pre-inserted Liposomes”).
Preparation of Pre-inserted LiposomesThe plain liposomes were composed of EPC : chol : rho lipid : 5-FAM conj. (48 : 50 : 2 : 10). The hydration method was followed by a bath sonicator and the same procedure mentioned in method “Preparation of Inserted Liposomes.” The plain liposomes with 102 nm diameters were produced, and further separated by SEC, whose method was mentioned in methods “Separation of the Liposomes by Sepharose® CL-4B Chromatography,” and whose calculation was mentioned in methods “Calculation of Surface Ligand Density” to obtain the Fig. 5.
In an active targeting system, an appropriate ligand is required for binding to the target. Its density on the liposomal surface also needs to be analyzed to permit the relationship between ligand–receptor binding affinity versus ligand density to be evaluated. To provide an indicative parameter for the degree of liposomal binding affinity, we defined the liposomal surface ligand density and developed a standardized liposomal preparation method. In our methodology, the term ‘incorporation ratio’ is introduced as a parameter that reflects the accurate surface ligand density.33) In Chart 1, to analyze the incorporation ratio of the inserted (functionalized) liposomes a post-insertion procedure was employed using plain liposomes with a 5-FAM–PEG2000–DSPE conjugate lipid (5-FAM conj.), in which the liposomal surface was modified with a fluorescently labeled functionalized conjugate. After removing contamination, e.g., 5-FAM conj. micelles, by SEC (Sepharose® CL-4B), inserted liposomes were prepared for use in estimating the incorporation ratio37–39) (see an example of a chromatogram in Fig. S2). In conventional considerations, the surface ligand density is assumed to be all of the added ligand–lipid used in the post-insertion procedure ((1) in Chart 1). That is, all the 5-FAM conj. was assumed to be incorporated into the inserted liposomes and the SEC separation is omitted. However, our methodology denies this assumption and the value of this parameter, the incorporation ratio is defined as:
 | (2) |
The above flowchart was designed to calculate surface ligand density and analyze the post-insertion characteristics of various sized liposomes. To correct for the conventional surface ligand density (Eq. 1), we defined the ‘incorporation ratio’ as a parameter that reflects the surface ligand density as Eq. 2. The insertion efficiency is also defined as the percentage of conjugates that are attached to liposomes of the total amount of conjugates that were added (Eq. 3). Liposomes of four sizes (diameters 188, 137, 70, 40 nm) were prepared to determine the effect of liposome size on the incorporation ratio. The 5-FAM conj. was introduced as a fluorescently labeled ligand–lipid conjugate in the post-insertion procedure. It is important that the molar ratio of liposomal lipids–conjugates is maintained constant (16 : 1) in the insertion process. Each aliquot of a liposomal solution at different time intervals (10, 30 min, 1, 2, 3 h) were collected and the incorporation ratio precisely were analyzed after removing the micelles by SEC.
Smaller liposomes have unique physical properties that allow them to penetrate and become distributed in pathological tissues such as in a tumor microenvironment. If the effect of particle size is considered to play a role in therapeutic efficacy, the surface ligand densities on liposomal surfaces with different sizes remains an interesting issue to examine. Plain liposomes with 4 different diameters (188, 137, 70, 40 nm) were prepared by processing them through the same post-insertion conditions, i.e., molar ratio of conjugates/liposomal lipids and temperature, since these factors are known to govern the post-insertion efficiency. These liposomes of each diameter were then continuously sampled to evaluate the incorporation ratios for different incubation times to monitor the insertion process.
An example calculation for the 70 nm diameter liposomes incubated by 10 min post-insertion is shown in Fig. 1. The Sepharose® CL-4B chromatogram gave two overlapped peaks in the dotted line rectangle, which represents the area of elution collection for 5 tubes. In an HPLC analysis of the 5 tubes, a plot of the AUC of HPLC versus tube number indicated that the gray bars represent the molar amount of conjugate (5-FAM conj.) and the black bars represent the molar amount of liposomal lipids. Applying Eq. 2, those molar amounts were then converted to incorporation ratios for different tubes (sizes) respectively. Here, it was observed that different fractions (sizes) derived from 70 nm diameter liposomes have various incorporation ratios. These ratios demonstrate their size-dependency and provide an ‘overall’ incorporation ratio for the 70 nm liposomes as well.
To analyze the incorporation ratios of inserted liposomes, liposomes with most reliable qualities in the 5 tubes were collected after Sepharose® CL-4B separation. In the upper right, the AUC of each tube was analyzed by HPLC to determine the molar amount of conjugate (5-FAM conj.) and lipid in the liposomes. By applying Eq. 2 the incorporation ratio for each tube was then determined, as shown in the lower right, and the process repeated for samples prepared using different post-insertion time intervals.
In addition to nanoparticle size, the incubation time used for post-insertion is the other main factor that affects the value of the incorporation ratio. Thus, repeating the same process shown in Fig. 1 for various incubation times and further for liposomes with 4 different diameters (188, 137, 70, 40 nm), the incubation time axis could be used to prepare 4 3D charts of incorporation ratios versus size (fraction) and time, as shown in Figs. 2(a) to (d) (see the physical data in Tables S1 to S4 of the supplementary section). Analyzing these 3D charts, the distribution of the incorporation ratio values in the sizes of fractional liposomes and time used for their preparation showed some trends. First, the incorporation ratios in (b) and (c) (137, 70 nm liposomes) increased along the time axes, showing an obvious time-dependency. In the case of such ratios in (a) and (d) (188, 40 nm liposomes), however, no dependency on incubation time was found. Secondly, in observing these fractions in the 3D charts, the size of the nanoparticle was clearly an important factor that affects the incorporation ratio. Although the incorporation ratios of fractions in (a), (b), and (d) (188, 137, 70 nm liposomes) did not show an obvious size-dependency, the ratios of fractions in (c) (70 nm liposomes), however, presented a remarkable size-dependency; that is, smaller liposomes (fractions of smaller sizes) tended to show a higher incorporation ratio. Surprisingly, it was also found that all of the fractions of the smaller liposomes (d) (40 nm liposomes) approached the maximum value,* 5.88%, of the incorporation ratio but, in contrast, the values for the incorporation ratio of all fractions of the larger liposomes (a) (188 nm liposomes) were around 3.0%. This suggests that smaller liposomes have a greater capability to incorporate conjugates, especially, in a short time interval and that larger liposomes have only limited capability.
Liposomes with 4 different diameters (188, 137, 70, 40 nm) were incubated with conjugates for different post-insertion times and then further separated by SEC to produce fractions of different sizes. The incorporation ratios of each liposome size (fraction) were analyzed and the results were used to prepare a 3D chart, incorporation ratio versus time and size of fractions for the 4 different liposomes. The incorporation ratio in the 3D chart (b) (137 nm liposomes) and (c) (70 nm liposomes) showed a time-dependency but not the (a) (188 nm liposomes) and (d) (40 nm liposomes). Surprisingly, such ratios along with the size of fractions axis in (c) (70 nm liposomes) showed a remarkable size-dependency, in that the ratio increased with decreasing size. Also, all fractions in (d) (40 nm liposomes) demonstrated a higher capability to incorporate the conjugate, resulting in higher incorporation ratios compared with other liposomes with larger diameters.
The fractional incorporation ratios of 4 liposomes of different diameters showed different degrees of size-dependency and conjugate-incorporated capability. To carefully analyze these dependencies and capabilities, the incubation time (3 h) was fixed and each of the 4 diameter liposomes was analyzed and the data were used to prepare a chart of incorporation ratio versus size of fractional liposomes as shown in Figs. 3(a) to (d). In (a) (188 nm liposomes) and (b) (137 nm liposomes) the ratios and sizes (fractions) showed a relatively low linearity and the values for incorporation ratio were relatively low, around half of the maximum value* of the incorporation ratio, 3.0% as well. Nonetheless, those ratios and sizes (fractions) in (c) (70 nm liposomes) and (d) (40 nm liposomes) showed a nearly perfect linearity. It is particularly noteworthy that all fractions (sizes) in the 40 nm liposomes (d) possessed a nearly maximum value* of incorporation ratio, indicating that nearly all of the conjugates had been incorporated. Moreover, while considering each slope of the linearity in (a)–(d), the steeper slope of (c) (70 nm liposomes) showed that the ratios dramatically increased in parallel with the size of fractional liposomes from 95 to 70 nm. This suggests that liposomal sizes in this range were more able to incorporate the conjugate and that the process was dependent on the size of the particle. The reason for the extraordinary conjugate-incorporated capability in smaller sized liposomes, e.g., from 75–40 nm, may be the larger curvature on their surfaces that would permit the surface to provide more steric space on the liposomal surfaces to permit the conjugates to be incorporated and stabilize them when the post-insertion manner was performed.
 |
The incorporation ratios versus sizes (fractions) for 4 liposomes with different diameters (188, 137, 70, 40 nm) were analyzed to assess the degree of size-dependency and the conjugate-incorporating capability. The incubation time for post-insertion was fixed at 3 h. The linearity of 188 nm liposomes (a) (y=−0.0133x+5.1857, R2=0.22239), and 137 nm liposomes (y=0.0622x−5.3674, R2=0.34024) gave a low linearity of the fractional incorporation ratio to the size of the fractional liposomes. However, the 70 nm liposomes (c) (y=−0.0543x+8.5732, R2=0.99765), and the 40 nm liposomes (y=−0.0765x+9.1685, R2=0.96875) gave a perfect linearity of those ratios and sizes. The steeper slope of the linearity for the 70 nm liposomes (c) clearly showed that in the range of these sizes of fractional liposomes, their incorporation ratios dramatically increased with decreasing size. Each value of fractional incorporation ratio was calculated in 3 separate experiments (n=3).
After obtaining those values and trends for incorporation ratios in those fractions (sizes), we eventually calculated their overall incorporation ratios for the 4 liposomes of different diameters (188, 137, 70, 40 nm) and an overview graph is shown in Fig. 4. Each overall incorporation ratio was obtained using the concentration normalizing calculation** from the liposomal fractions, and the overall trend clearly showed that the incorporation ratios were size-dependent as well. Those trends and the linearity mentioned in “The Distribution of Incorporation Ratios in Different Diameters” verified this dependency, and liposomal diameters in the area of 75 nm began to dramatically increase the potential for conjugate-incorporation to a diameter of 40 nm where the maximum incorporation ratio was found.* Although we fixed the post-insertion conditions and applied a fluorescent lable in place of a real functioal ligand to discover the size and curvature effect, in an overview of the post-tinsertion methods used for incorporating lipids into liposomes, factors that have been reported in the literature, such as temperature, the initial ratio of PEG-lipid in the liposomes, the ratio of additional PEG-lipid to the liposomal lipids, the type of compositional lipids in the liposomes, lipid concentration and the incubation times etc. were taken into consideration.24,37,38) Nonetheless, this result impacts the conventional assumption that the surface ligand density, remains constant, irrespective of liposome size.
 |
An overview graph showing the overall incorporation ratios for the 4 liposomes of different diameters (188, 137, 70, 40 nm) demonstrating the size-dependency. Liposomal diameters in the range of 75 to 40 nm had a high potential for conjugate-incorporation on the liposomal surfaces. Each value for the overall incorporation ratio was calculated using 3 repeated experiments (n=3).
An interesting question rose as to whether the size-dependency occurs during the post-insertion process or not. To examine this, pre-inserted liposomes (incorporated with 5-FAM conj.) were prepared. SEC further gave liposomal fractions of different sizes and the incorporation ratio was calculated for each fraction in Fig. 5 (see the physical data on Table S5 in the supplementary). The calculation was based on the assumption that the conjugates are exposed outward/inward from the membrane with a theoretical ratio 55/45 on the pre-inserted liposomes, i.e., incorporation ratio=(incorporated 5-FAM conj.×(55/100))/(lipids mixture+incorporated 5-FAM conj.).33) Surprisingly, small differences in the incorporation ratios in eight different sizes of fractional liposomes suggested the absence of any size-dependency trend during the formation of pre-inserted liposomes. This result may be caused by homogeneous lipid constituents in the initial hydrated liposomes that form the liposomes during the sonication process. Hence, each unit of surface area of liposomes contained similar constituents of lipids, irrespective of the sizes of liposomes so that its incorporation ratio for every fraction was similar. We finally conclude that pre-inserted liposomes offer functionalized liposomes with a homogeneous surface ligand density (incorporation ratio).
SEC gave fractions with sizes of 60–90 nm from the pre-inserted liposomes, and the incorporation ratio of each fraction was calculated. No size-dependency was observed on these liposomal fractions but these fractional liposomes showed homogeneous incorporation ratios.
The unexpected results obtained in this study demonstrated the non-linearity of the incorporation ratios for nanoparticles. This ratio rapidly increases with decreasing liposomal diameter. This non-linearity may lead to an underestimation of the surface ligand density, thus leading to erroneous interpretations among the size, density of surface ligands, and biological activities. For the significance of smaller liposomes in the near future, this finding provides important information regarding the DDS development from the view point of regulatory science (medicine).
This work was supported in parts by Grants from the Special Education and Research Expenses of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan, and a Grant-in-Aid for Research Activity Start-up from the Japan Society for Promotion of Science (JSPS) (Research Project Number: 26893001). The authors also wish to thank Dr. Milton S. Feather for his helpful advice in writing the English manuscript.
The authors declare no conflict of interest.
The online version of this article contains supplementary materials.
