Application of 19F NMR Spectroscopy Using a Novel a-Tocopherol Derivative as a 19F NMR Probe for a Pharmacokinetic Study of Lipid Nano-Emulsions in Mice

Methods: An α−tocopherol derivative, 19F-TP, in which a 4-(trifluoromethyl) benzoyl group was introduced to the hydroxyl group of α-tocopherol was newly synthesized as a 19F NMR probe. Three different LNEs containing 19F-TP, denoted 19F-TP-LNEs (Small-LNE, Large-LNE, and polyethylene glycol-modified LNE (PEG-LNE)) were prepared by the sonication method and characterized using a dynamic light-scattering method and zeta potential analysis. The concentrations of the three 19F-TP-LNEs in the blood, liver and kidney of mice were periodically evaluated based on the 19F NMR signal intensity ratio of 19F-TP using 0.1 mM of trifluoromethane sulfonic acid sodium salt as an internal reference.


Introduction
Investigation of the pharmacokinetics of drug carriers in the body provides valuable information for drug delivery system (DDS) research. Fluorescent probes [1] and radioisotope-labeled molecular probes [2] are generally used for this purpose. Evaluation of drug carriers distributed in the blood and organs has been performed by quantitative determination of molecular probes loaded onto drug carriers in the blood and in each organ. However, the complicated tasks of deproteination of the matrix and extraction of the target compound must be completed before analysis using high-performance liquid chromatography (HPLC) can be carried out. Moreover, the radioactive nature of the radioisotopes makes human exposure a risk and necessitates the use of strict control measures in dedicated facilities. These complicated restrictions and pretreatment requirements interfere with the development of fast-acting, effective drug carriers. 19 F nuclear magnetic resonance (NMR) spectroscopy has the potential to be a powerful tool for pharmacokinetic studies of drug carriers. The usefulness of 19 F NMR can be attributed to the fact that the natural abundance of the 19 F nucleus is 100% and its sensitivity relative to protons is approximately 83%. In addition, the 19 F NMR chemical shift has a range of approximately 250 ppm, which is much greater than that of the 1 H NMR chemical shift; that is, 19 F NMR signals are more sensitive to changes in the chemical environment than 1 H NMR signals. Because the 19 F nucleus is not present in natural biological substances, it is easily detectable without interfering signals even in the presence of low concentrations of 19 F-containing compounds [3].
Lipid nanoparticles such as lipid emulsions (LEs), liposomes, solid lipid nanoparticles and micelles have been a focus of DDS research as they are physiologically compatible, targetable, generally non-toxic and amenable to large-scale production. Compared with other carriers, LEs have many advantages including that they exhibit a higher drug solubilization capacity that are easier to process and manufacture, and are more cost effective [4,5]. LEs are frequently used for safe administration of parenteral nutrition in clinical settings. Because LEs are expected to act as good drug carriers because of their high lipophilicity and apolarity, which allows them to cross cell membranes, they have also been used as parenteral DDS carriers [6] for sites of inflammation [7], as well as the heart [8] and lymphatic system [9], because of their tendency to accumulate in these areas. Moreover, recently, LEs have been employed as carriers of anticancer agents to improve their therapeutic indices and minimize drug cytotoxicity in normal cells [10][11][12][13][14][15][16]. It has been recognized that only drug carriers less than 100 nm in diameter can pass through the discontinuous capillary endothelium of tumors [17]. LEs with droplet sizes on the nanometer scale are characterized as lipid nano-emulsions (LNEs). LNEs with droplet sizes of less than 100 nm show high selectivity towards tumor tissues [18,19] because they accumulate passively because of leaky tumor vasculature. This is known as the enhanced permeation and retention (EPR) effect [20,21]. In a previous study, we developed an LNE that was prepared from a lipid mixture of soybean oil (SO), phosphatidylcholine (PC), sodium palmitate (PANa) and sucrose fatty acid ester [22]. As the mean droplet size of this LNE was approximately 50 nm, it was investigated as a DDS carrier for cancer therapy [23,24].
The aim of our study was to use 19 F NMR spectroscopy as an analytical technique to investigate the pharmacokinetics of LNEs. We focused on α-tocopherol (α-TP), a lipid-soluble vitamin with no reported adverse reactions, as the 19 F NMR probe compound. In this study, we synthesized a novel 19 F derivative of α-TP ( 19 F-TP) for use as a 19 F NMR probe by introducing a 4-(trifluoromethyl)benzoyl group to the hydroxyl group of α-TP, and used 19 F NMR to establish a procedure for convenient evaluation of LNE pharmacokinetics without the need for complicated pretreatment procedures.

Chemical synthesis of 19 F-TP
The chemical reaction used in this study, which involves the benzoylation of a phenolic hydroxyl group, is shown in Figure 1. α-TP (2.00 g, 4.64 mmol), 4-(dimethylamino)pyridine (113 mg, 0.93 mmol) and 4-(trifluoromethyl)benzoyl chloride (1.52 mL, 10.21 mmol) were added to an empty 100-mL round-bottomed flask along with 50 mL of pyridine and the resulting suspension was stirred for 18 h at room temperature. Progress was monitored by silica gel thin-layer chromatography (TLC). The pyridine was removed using a rotary evaporator and the product was extracted with ethyl acetate. The resulting clear solution was washed with 3% HCl solution followed by saturated NaCl solution, and then dried over anhydrous Na 2 SO 4 , which was subsequently removed by filtration. Ethyl acetate was removed using a rotary evaporator, and the residue was purified by silica gel chromatography using hexane/ethyl acetate (30:1 v/v) to give white oil. The final product, 19 F-TP, was obtained in 96% yield. The purity of 19 F-TP was confirmed by 1 H NMR and TLC using hexane:ethyl acetate=30:1 (v/v). The retardation factor (R f ) of 19 F-TP was 0.3 by TLC.

Preparation of 19 F-TP-LNEs
The formulations of the three LNEs containing 19 F-TP (Small-LNE, Large-LNE, and PEG-LNE), generalized as 19 F-TP-LNEs, are shown in Table 1. The preparation of spherical LNE particles was carried out using a sonication method that is described in detail elsewhere [22]. The mixture was emulsified by sonication using a VC-501 instrument (Tokyo Rikakikai Co. Ltd., Tokyo, Japan) for 1 h at 55°C in a thermostatic water bath. Sonication for 3 min was repeated at 3-min intervals. The 19 F-TP-LNE suspensions were centrifuged at 2000 × g to eliminate sediment from the sonication tip and then stored in tightly closed, light-resistant, glass containers at room temperature under a nitrogen atmosphere. The exact 19 F-TP concentrations in three 19 F-TP-LNE suspensions were measured using HPLC as indicated below. Analytical samples were dissolved in methanol before injection. All measurements were carried out in triplicate. The mean and standard deviation (S.D.) of 19 F-TP concentrations in three 19 F-TP-LNE suspensions were 27.3 ± 1.2, 27.5 ± 0.9, and 27.1 ± 0.8 mM for Small-LNE, Large-LNE and PEG-LNE, respectively.

Characterization of 19 F-TP-LNE preparations
The 19 F-TP-LNE preparations were further diluted with deionizeddistilled water to 1:1000 for droplet size measurement and to 1:10,000 for zeta potential measurement. The mean diameters and droplet size distributions of the 19 F-TP-LNE particles were determined by dynamic light-scattering (DLS) using a Nicomp 380 analyzer (Particle Sizing Systems, Santa Barbara, CA, USA) and the 19 F-TP-LNE droplet size was reported as a volume-weighted distribution. Zeta potential values were measured using a Zeecom ZC-3000 analyzer (Microtec Co., Ltd., Chiba, Japan), based on the principle of electrophoresis.

Animals
Male specific-pathogen-free ddY mice (aged 5-6 weeks, 28-30 g) were purchased from Japan SLC Inc. (Shizuoka, Japan) and maintained under conventional housing conditions. All animal experiments were conducted in accordance with the principles and procedures outlined in the National Institutes of Health Guide for the Care and Use of Laboratory Animals. The protocols for animal experiments were approved by the Animal Experimentation Committee of Kyoto Pharmaceutical University.
A dose of 100 μL of each of the prepared 19 F-TP-LNE suspensions was injected into the mice via the tail vein. At selected intervals thereafter, the mice were lightly anesthetized, dissected and bled via the vena cava using a hypodermic needle treated with heparin, after which both the liver and kidney of the mice were excised and washed with saline.   19 F-TP concentrations under the conditions described above, and a 19 F-TP calibration curve was prepared in a manner similar to that described above.
Determination of 19 F-TP concentration in mouse blood using 19

F-NMR
At suitable time intervals, an analytical sample was prepared by adding a 300-μL blood sample from a mouse to 240 μL of D 2 O and 60 μL of a 1 mM TFMS-D 2 O stock solution to achieve a concentration of 0.1 mM, and 19 F NMR measurements were carried out using the conditions described above. 19 F-TP concentrations were calculated using the calibration curve described above.

Determination of 19 F-TP concentration in mouse plasma using HPLC
The blood samples from mice were centrifuged at 10,000 rpm (× 5500 g) for 5 min to give plasma samples, after which 100 μL of plasma was added to 900 μL of ultrapure water and 6 mL of ethyl acetate. After deproteination, the mixtures were centrifuged at 3500 rpm (× 2000 g) for 10 min. Thereafter, 5 mL of the upper layer was withdrawn from the mixture and dried at 60°C. The analytical samples were prepared by adding 100 μL of methanol to the residue. The 19 F-TP concentrations were measured using the HPLC experimental conditions described above.

Determination of 19 F-TP concentrations in mouse liver and kidneys using 19 F-NMR
At suitable time intervals, 1 mL of lysis buffer (0.1 M Tris/HCl, 0.05% Triton X100, 2 mM EDTA and pH 7.8) was added to 1-g samples of each organ, and the organ suspensions were homogenized at 30,000 rpm for 1 min. To prepare analytical samples, 300 μL of the prepared organ suspension was added to 240 μL of D 2 O and 60 μL of a 1 mM TFMS-D 2 O stock solution to achieve a final concentration of 0.1 mM. 19 F NMR measurements were carried out using the conditions described above. 19 F-TP concentrations for each organ were calculated using the calibration curve described above.

Characterization of 19 F-TP-LNE preparations
The mean droplet sizes and zeta potential values of the three 19 F-TP-LNE preparations are shown in Table 2. The mean droplet sizes of the Small-LNE, Large-LNE and PEG-LNE were 58, 157 and 174 nm, respectively. The latter two values were comparable and approximately three times larger than that of the Small-LNE. The zeta potential of the Small-LNE was -34 mV, while that of the Large-LNE was lower at -53

HPLC assay
A reverse-phase HPLC method was used for analysis of 19 F-TP. Quantitative determination of 19 F-TP in the analytical samples prepared from the 19 F-TP-LNE preparations and mouse plasma was performed by the absolute calibration method using a COSMOSIL 5C 18 -MS-II column (4.6×150 mm, 5 µm, Nacalai Tesque Co., Kyoto, Japan) using methanol as a mobile phase at a flow rate of 1.0 mL/min. The injection volume of the analytical samples was 20 μL and 19 F-TP detection was performed using a ultraviolet detector at 280 nm. The t R value of 19 F-TP was 16.7 min. This HPLC method was linear (R = 0.999) over a 19 F-TP concentration range of 5-100 μM. The lower limit of quantification (LLOQ) was set at 5 μM.

F NMR measurements
All 19 F NMR spectra were measured using a UNITY INOVA spectrometer (Agilent Technologies, Inc., Santa Clara, CA, USA) operating at 376.21 MHz without proton decoupling. The set parameters were a 3.0-µs pulse width (30° for the flip angle), a relaxation delay of 0.5 s, and an acquisition time of 0.5 s. The probe temperature was 25°C. The number of free induction decay (FID) accumulations to improve the signal-to-noise (S/N) ratio was from 1000 to 80,000, which corresponded to an accumulation time of approximately 16 min to 22 h.

Calibration curve of 19 F-TP from 19 F-TP-LNE in blood
Fresh blood taken from the vena cava of male ddY mice anesthetized with ether was used without removal of the blood cells. The calibration samples were prepared by adding 300 μL of blood suspension containing various amounts of 19 F-TP-LNE suspension to 240 μL of D 2 O and 60 μL of a 1 mM TFMS-D 2 O stock solution so as to achieve a concentration of ca. 0.1 mM TFMS. The samples were stirred and transferred into 5-mm-diameter NMR sample tubes, and 19 F NMR was carried out in triplicate at five suitable 19 F-TP concentrations under the conditions described above. A 19 F-TP calibration curve was prepared by plotting the 19 F-TP concentration on the horizontal axis and the ratio of the signal intensity of 19 F-TP to the signal intensity of the trifluoromethyl group of 0.1 mM TFMS, the internal standard, on the vertical axis.

Calibration curves of 19 F-TP from 19 F-TP-LNE in liver and kidneys
First, 1 mL of lysis buffer (0.1 M Tris/HCl, 0.05% Triton X100, and 2 mM EDTA, pH 7.8) was added to 1-g samples of liver or kidney. The organ suspensions were homogenized using a Physcotron NS-360 instrument (Microtec Co. Ltd., Chiba, Japan) at 30,000 rpm for 1 min, and were used without further separation such as centrifugation or filtration. To prepare calibration samples, 300 μL of the prepared organ suspension containing various amounts of 19  mV. Using the mean droplet sizes of both LNEs, the total surface area of the Small-LNE droplets was calculated to be approximately three times as large as that of the Large-LNE droplets. The amount of PANa used for the preparation of the Small-LNE was twice that used for the Large-LNE as shown in Table 1. Thus, this difference was attributable to the larger total surface area of the Small-LNE droplet compared with the Large-LNE droplet, which resulted in a smaller amount of palmitate, which contains a COOgroup, per unit surface area. The zeta potential of the PEG-LNE, which had a similar droplet size to the Large-LNE, was -32 mV, which was comparable to the zeta potential of the Small-LNE. This may be the result of the fact that the surface of the PEG-LNE droplets, which was covered with a hydrophilic PEG layer, was not significantly influenced by the negative charge of the palmitate COOgroup [25].

F NMR spectroscopic behavior of F-TP and F-TP-LNE in the biological samples
The LNE particles interact with various biological substances in the body after administration. If 19 F-TP molecules localized at the PC/ water interface of the LNE particles are pulled from the LNE particles by biological substances such as serum albumins and blood cells, this could prevent proper evaluation of the pharmacokinetics of LNE. For this reason, 19 F-NMR was used to examine the behavior of 19 F-TP in the LNE suspension and biological samples. Figure 2a shows the 19 F NMR spectrum of the Small-LNE containing 2 mM 19 F-TP in 100 mM phosphate buffer solution (pH 7.4). As shown in Figure 2a, a single sharp signal attributable to the trifluoromethyl group of 19 F-TP was observed at 15.4 ppm. In contrast, the signal of 2 mM 19 F-TP spiked with buffer solution containing bovine serum albumin (BSA) at a physiological concentration of 0.6 mM was shifted upfield to 14.9 ppm and considerably broadened as seen in Figure 2b. Because 19 F-TP is a highly lipophilic compound, the broadened signal is considered to be derived from 19 F-TP binding to BSA molecules. Meanwhile, the addition of 0.6 mM BSA to the Small-LNE buffer solution did not induce any significant changes in the chemical shift value of the 19 F-TP signal, i.e., 15.4 ppm, as depicted in Figure 2c. As shown in Figures 2a,  2b and 2c, the 19 F NMR signals of 19 F-TP in the Small-LNE suspension were not significantly different in the absence and presence of BSA, whereas 19 F-TP bound to BSA resulted in a clearly broadened signal that was shifted upfield. Thus, these results suggest that 19 F-TP may be localized in the inner SO phase of the Small-LNE particles and is not present in the water phase or on the lipid monolayer/water interface of the LNE particles where 19 F-TP can interact with BSA. Figure 2d shows the 19 F NMR spectrum of the Small-LNE containing 2 mM 19 F-TP in a mouse blood suspension. A single sharp signal attributable to the trifluoromethyl group of 19 F-TP was observed at 15.4 ppm as shown in Figure 2d. The chemical shift value of this signal was similar to the corresponding signal in Figure 2a. In contrast, as seen in Figure 2e, the signal of free 2 mM 19 F-TP spiked in a blood suspension was considerably shifted downfield and slightly broadened at 16.50 ppm. The same results were also obtained for a liver suspension. This may have resulted from the single signal attributable to 19 F-TP interacting with the lipid membrane of blood cells and liver tissues. As indicated by the results in Figure 2d and 2e, the magnetic environment of 19 F-TP is different in the LNE particles and in the lipid membrane of biological cells, i.e., the presence of 19 F-TP can be detected by examining the 19 F NMR signal. To demonstrate the 19 F NMR spectroscopic behavior of 19 F-TP, further experiments were carried out. Figure 2f shows the 19 F NMR spectrum of both free 2 mM 19 F-TP and the Small-LNE containing 2 mM 19 F-TP spiked in a blood suspension. As seen in Figure 2f, two single signals were separately observed at 15.4 and 16.5 ppm, and are attributable to 19 F-TP in the Small-LNE and 19 F-TP interacting with the lipid membrane of blood cells, respectively. The same result was also obtained for the liver suspension. The spectral results indicate that the exchange rate of 19 F-TP between the two states in the LNE particles and in the blood cell membranes is slow on the 19 F NMR time scale. Therefore, if the 19 F-TP molecules are released from the LNE particles during blood circulation and organ accumulation, the signal at 16.5 ppm will be observed in the 19 F-NMR spectra for the biological samples. Figure 3 shows the 19 F NMR spectra of 19 F-TP in blood taken from the mice at 30, 60 and 360 min after the administration of the Small-LNE. A single sharp signal attributable to 19 F-TP was observed at 15.4 ppm in each spectrum, but was observed to decrease over time. As sedimentation of blood cells and LNE particles was not observed in all analytical samples after the 19 F NMR measurement, it was concluded that the condition of the blood samples could be maintained over long accumulation times, such as the 22 h required to improve the S/N ratio of the 19 F NMR spectra in the analytical samples at 360 min after administration.

F NMR spectra of F-TP-LNE in blood
The chemical shift values obtained from the spectral data at each time are shown in Table 3. The values obtained from the 19 F NMR spectra of the calibration samples used for preparation of the calibration curve are also shown in Table 3. These chemical shift values did not 19 F-TP-LNE Diameter (nm) Zeta potential (mV)

Small-LNE 58 ± 3 -34 ± 3
Large-LNE 157 ± 4 -53 ± 9 PEG-LNE 174 ± 4 -32 ± 2   change over time and were the same compared with the corresponding signals in Figure 2a and 2d. In addition, the signal attributable to the released 19 F-TP interacting with the blood cells observed at 16.50 ppm in Figure 2e was not observed in the 19 F NMR spectra for the analytical and calibration samples at any time. These results indicate that 19 F-TP was present in the same magnetic environment over time; that is, it did not leak from the Small-LNE particles and instead, remained encapsulated in them during blood circulation and the long 19 F NMR measurements. The signal intensity ratios of 19 F-TP to 0.1 mM TFMS were calculated and the calibration curve was used to quantitatively determine 19 F-TP. The curve showed good linearity (R = 0.998) over a 19 F-TP concentration range of 6-2800 μM. The LLOQ of 19 F-TP in blood was set at 6 μM.

Comparison of 19 F NMR and HPLC methods for examining the circulation of the 19 F-TP-LNE preparations
To confirm the usefulness of 19 F NMR as a convenient technique for assessing the pharmacokinetics of LNE, the 19 F-TP concentrations of the Small-LNE in blood and plasma were determined using 19 F NMR and traditional HPLC methods, respectively. The results are illustrated in Figure 4. There was a significant difference between the 19 F-TP concentration profiles determined using 19 F NMR and those determined using HPLC, and the concentrations obtained from the 19 F NMR method were similar to but higher than those obtained using the HPLC method. This is probably because during the extraction of 19 F-TP from the blood samples for HPLC analysis, no correction was made for any loss of 19 F-TP resulting from the extraction process. While the 19 F-TP concentration can be measured within approximately 20 min using the HPLC method (the t R value of 19 F-TP is 16.7), complicated pretreatments such as deproteination, extraction and separation procedures are required before analysis. In contrast, using the 19 F NMR method, the 19 F-TP concentration can be easily obtained from intact blood samples without such complex procedures. However, there is still the need for time-consuming FID accumulations to improve the S/N ratio at lower 19 F-TP concentrations; e.g., for measurement of an analytical sample at 360 min, it takes more than 22 h to acquire a 19 F NMR spectrum with a S/N ratio sufficient to determine the quantity of 19 F-TP. In terms of applying this procedure to the measurement of biological samples with simple pretreatments, this result demonstrates that the 19 F NMR method is useful for determining the blood circulation of 19 F-TP-LNE.

Circulation of 19 F-TP-LNE preparations in blood
To examine whether 19 F NMR can detect different pharmacokinetics that may arise from the droplet size and surface physical properties of the LNE particles, the 19 F NMR spectra of three different 19 F-TP-LNEs (Small-LNE, Large-LNE, and PEG-LNE) in blood samples periodically bled from mice were determined and the corresponding 19 F-TP blood profiles were calculated. As shown in Figure 5, elimination of Large-LNE from the blood is almost complete 60 min after administration. More Small-LNE than Large-LNE is present in blood at 60 min, but the amount of Small-LNE decreases sharply from 42% to 12% from 30 to 60 min. We previously reported that Small-LNE flocculated in the presence of the counter ion Na + in saline and that the droplet size increased from 50 nm to 150 nm at more than 30 min after addition [22]. A rapid drop in the level of Small-LNE occurred when the droplet sizes increased in the presence of high concentrations of Na + in the blood. Approximately 8% of the amount of PEG-LNE administered remained in the blood 360 min later; the blood circulation of PEG-LNE was clearly better than those of the other two 19 F-TP-LNEs.
To quantitatively evaluate the differences in blood circulation among the three 19 F-TP LNEs, circulation data were calculated using MULTI [26], a nonlinear least-squares program based on a onecompartment model. The program was used to determine the areas under the blood concentration-time curves (AUCs) of the 19 F-TP-LNEs in Figure 5. It was found that the AUC of Small-LNE was 45.69   h⋅% of dose/mL, approximately 2.7-fold greater than that of Large-LNE at 16.64 h⋅% of dose/mL. It has been reported that physical and chemical properties such as particle size, surface charge and surface hydrophilicity are important in evaluating the biological fate of nanoparticles after intravenous administration [27,28]. LNEs disappear from the blood following uptake into the Kupffer cells of the liver, spleen macrophages and through other endocytotic processes of the reticuloendothelial system (RES). Larger mean droplet sizes are more easily captured by the RES [2,29]. Moreover, the degree of phagocytosis increases in proportion to the absolute zeta potential value [30,31]. On this basis, Large-LNE, which has a large droplet size and the smallest zeta potential value, was expected to have lower blood circulation than the Small-LNE. Thus, the LNEs are not captured by the RES and thus, they have increased blood circulation when the droplet size is sufficiently small. The AUC of PEG-LNE was 117.04 h⋅% of dose/mL, which was approximately 2.6 times greater than that of the Small-LNE, and the blood circulation time of PEG-LNE was greater than those of Small-LNE and Large-LNE despite its larger droplet size. The uptake of PEG into the RES is low because the substance has limited interactions with plasma proteins and cells [32][33][34]. The different blood circulation profiles of the three 19 F-TP-LNEs show 19 F-TP to be a useful 19 F NMR probe for evaluating LNE blood circulation. 19 F NMR was also used to determine the amount of 19 F-TP present in the livers and kidneys of mice. The calibration curve for each organ showed good linearity (R = 0.999) over a 19 F-TP concentration range of 6-2400 μM. The LLOQ of 19 F-TP in both organs was set at 6 μM. In addition, all of the chemical shift values of the analytical samples from both organs were unchanged compared with those in blood (Table 3). For example, the chemical shift was 15.44 ± 0.01 (ppm) in liver (N = 15), and 15.44 ± 0.01 (ppm) in kidney (N = 12), respectively. Furthermore, the signal at 16.4 ppm attributed to the released 19 F-TP interacting with the lipid membrane of organ cells as shown in Figure  2e was not observed in the 19 F NMR spectra for the analytical and calibration samples of both organs at any time.

Organ distribution of 19 F-TP-LNE preparations
The concentration profiles of the three 19 F-TP-LNEs in the liver and kidney are shown in Figure 6. The profiles of the 19 F-TP-LNEs in the liver correlated well with the obtained blood circulation profiles Figure 6a. As mentioned above, the Large-LNEs were eliminated from the blood just 30 min after administration. Its distribution in the liver at 30 min was approximately 14% and the concentration did not change subsequently. The uptake of both Small-LNE and PEG-LNE 30 min after administration was approximately 10%. At 60 min after administration, the Small-LNE distribution increased to 22%, compared with a distribution of approximately 15% for the PEG-LNE. This is explained by the finding that nanoparticles with diameters below approximately 70 nm will accumulate in the liver because of their penetration through the fenestrated endothelial lining [35]. The rapid increase in the Small-LNE distribution in the liver may also be attributed to the increase in LNE droplet size between 30 and 60 min after administration, which resulted in enhanced RES uptake in the Kupffer cells. The presence of PEG in the PEG-LNE resulted in a RES uptake that was lower than the distribution of the Small-LNE in the liver through a mechanism similar to that associated with the improved blood circulation of the PEG-LNE.
The distribution of the 19 F-TP-LNEs in the kidneys was relatively low Figure 6b, with the Small-LNE having the highest distribution. It is likely that the Small-LNE, with a mean droplet size of approximately 60 nm, had a greater renal uptake than the other LNEs because droplets in the 50-60 nm range are susceptible to interaction with the RES in the kidneys [36]. The renal distribution of the Large-LNE was low because a large proportion had already been taken up by the liver, while the renal distribution of the PEG-LNE was low because the presence of PEG excluded the LNE from the RES. The different concentration profiles of the three 19 F-TP-LNEs in the liver and kidneys show that 19 F-TP is a useful 19 F NMR probe for evaluating LNE distribution in various organs.

Conclusions
Differences in the droplet sizes and surface physical characteristics of the three 19 F-TP-LNEs resulted in differences in their blood circulation and organ distribution characteristics. This demonstrates the validity and usefulness of 19 F NMR as a convenient technique for assessing LNE pharmacokinetics. The use of 19 F-TP and 19 F NMR allows for convenient evaluation of LNEs and other drug carriers, and the results of this research should be useful in the development of fastacting, effective drug carriers.

Acknowledgments
The authors thank Microtec Co. Ltd., for allowing us to use the homogenizer (Physcotron NS-360).