Toward redesigning the PEG surface of nanocarriers for tumor targeting: impact of inner functionalities on size, charge, multivalent binding, and biodistribution

A simple strategy to enhance the tumor-targeting efficiency of PEGylated nanocarriers is demonstrated.

. Structural analysis of synthesized dendrimer conjugates by NMR and MALDI-TOF MS.
p. 28 Table S2. Structural parameters obtained from SAXS data of targeted and untargeted agents in aqueous solution. p. 29 Table S3. Surface charge of targeted and untargeted agents. p. 30 Table S4. IC 50 values of multivalent dendritic ligands and c(RGDfK) against binding of [ 125 I]echistatin to the α v β 3 integrin receptors on U87MG cells.
p. 38 Fig. S7. Synthetic scheme to prepare precursors of four high-avidity ligands.
pp. 45-48 & 50-51 Fig. S17. COSY and NOESY spectra of L SA in DMSO-d 6 . p. 49 Fig. S20-S23. 1 H NMR spectra of PH X and H X in DMSO-d 6 . pp. 52-55  dendrimer had a significant amount of structural defects, 1,2 and thus the average molecular weights (MWs) of its conjugates were generally overestimated by NMR integration (assuming theoretical 32 peripheral groups). Detailed methods for the structural analysis of PAMAM dendrimer conjugates by NMR were reported previously. 1 The stoichiometry and the average MWs of the dendrimer conjugates determined by NMR integration are summarized in Table S1.
Matrix-assisted laser desorption ionization time-of-flight mass spectrometry pattern was radially averaged from the beam center and was normalized to the transmitted X-ray beam intensity, which was monitored with a scintillation counter placed behind the sample. The scattering of a blank solution (i.e., 2.5 mM NaCl, pH 7.4) was used as the experimental background. The R g,G (radius of gyration) values were estimated from the slope of the linear scattering data in the q 2 -region using Guinier analysis 3 (Fig. 2a,b and Table S2). The SCATTER program was used for the analysis of one-dimensional X-ray scattering data, providing more detailed information on particle shape, size, and size distribution 4,5 (Fig. 2c,d). Additionally, the serum stability of our  Table S3, respectively.

Synthesis of PL 12Ac
To a portion of the crude reaction mixture of PL NH2 (1.70

Synthesis of PL 19Ac
To a portion of the crude reaction mixture of PL NH2 (1.70
The reaction was protected from light and stirred at room temperature for 60 h under a dry Ar atmosphere. In a darkroom, the crude mixture was first filtered through a short

Synthesis of PL PEG
To a portion of the crude reaction mixture of PL NH2 (1.70 mL, ca.
The reaction was protected from light and stirred at room temperature for 48 h under a dry Ar atmosphere. In a darkroom, the crude mixture was first filtered through a short
The reaction was protected from light and stirred at room temperature for 48 h under a dry Ar atmosphere. In a darkroom, the crude mixture was first filtered through a short

Synthesis of L 12Ac
To a solution of PL 12Ac (31.

Synthesis of L 19Ac
To a solution of PL 19Ac (30.6 mg, 1.79 μmol) in DMSO-d 6

Synthesis of H NH2
To a solution of PH NH2 (53.7 mg, 2.12 μmol) in DMSO-d 6  In a darkroom, the crude mixture was purified by the preparative SEC first using Bio-

Synthesis of G3-32PEG
The methanolic solution of G3 PAMAM dendrimer was dried in vacuo overnight and the resulting solid was weighed (1.14 mg, 0.16 μmol).

Cytotoxicity Assays
Stock solutions of 10 multivalent ligands (L X and H X ) at 600 μM in deionized water were diluted serially with culture media to prepare samples of the following concentrations: 10 -13 , 10 -12 , 10 -11 , 10 -10 , 10 -9 , 10 -8 , 10 -7 , 10 -6 , and 10 -5 M. U87MG cells were seeded in three flat-bottomed 96-well microculture plates each at a density of 5 × proportional to the number of live cells, was measured and normalized against that of the control (i.e., 100%) prepared under the same conditions. All experiments were repeated twice in triplicate (i.e., n = 6), and the results are shown as the mean ± SD in Fig. S36.

In Vitro Competitive Binding Assays
To determine the relative binding affinity of our multivalent ligands (i.e., All experiments were repeated twice in triplicate (i.e., n = 6), and the results are shown as the mean ± SD in Fig. 3a,b and Table S4.

Cellular Uptake Studies
Stock solutions (600 μM) of targeted (L X and H X ) and untargeted (PL X and PH X ) agents in deionized water were diluted with culture media (MEM containing 10% FBS and antibiotic-antimycotic) to prepare solutions at 1.8 μM. hpi. The obtained SPECT images were analyzed using InVivoScope software (Bioscan), and the results (range: 0-3 kBq) are shown in Fig. 4 and S38, and Movies S1 and S2. 100 μCi)) was injected into the mouse tail vein. The mice were sacrificed at either 2 hpi or 24 hpi, and their target organs (brain, blood, heart, lung, liver, spleen, kidney, stomach, intestine, femur, muscle, and tumor; Fig. S39) were collected, harvested, and weighed.

Biodistribution Studies
The radioactivity of each organ was measured using a PerkinElmer 1480 WIZARD 3″ automatic gamma counter. Finally, the percent of injected dose per gram of tissue (% ID/g) for each organ was calculated from the radioactivity data. All experiments were performed in triplicate, and the results are shown as the mean ± SD in Fig. 5a,b and Table   S5. Statistical analysis was performed by the unpaired t-test using the GraphPad Prism 5.0 software. P values less than 0.05 were considered significant. Additionally, from the biodistribution results, tumor-targeting efficiency for up to 24 hpi was estimated quantitatively as the area-under-the-curve (AUC; Fig. 5c) based on the noncompartmental linear trapezoidal analysis model. 8 For all compounds, one additional time point, 0 hpi (i.e., before injection, 0% ID/g), was considered for the estimation of AUC.
The relative tumor-targeting efficiency was also calculated as the tumor-to-organ ratios (mean ± SD, n = 3) as shown in Fig. 5d and Table S6.  a Radius of gyration obtained from the scattering data by the Guinier analysis. b Average sphere radius estimated from the particle size distribution. c Relative standard deviation (RSD), σ R = standard deviation/mean value. d Radius obtained as R = R g,G / (3/5) ½ . e Data were not obtained due to the limited supply of samples.   Fig. 3a,b in the main text for the binding curves. b Results, determined by the nonlinear regression analysis (sigmoidal dose response equation) using the GraphPad Prism 6 software, are shown as the mean ± SD (n = 6). c Calculated from the mean value of the corresponding log(IC 50 ).               , and H 9′(C) ) of the DBCO moiety was done similarly as described in Fig. S17.       S31. Characterization of synthetic intermediates, PPL and PPH, and a fully PEGylated G3 PAMAM dendrimer, G3-32PEG. G3-32PEG was prepared as a control using mPEG-NHS ester (11, see Fig. S5) in order to roughly estimate the hydrodynamic diameter of fully PEGylated compounds, such as PL PEG and PH PEG , by dynamic light scattering (DLS) (see Experimental Section in the main text for details). (a) Schematic illustrations; (b) size distribution plots obtained by DLS at a sample concentration of 2 mg/mL in 10 mM NaCl solution (pH 7.4) at 25 °C; (c) the proton labeling method of G3-32PEG for the peak assignment of its 1 H NMR spectrum; (d) MALDI-TOF mass spectra using DHB as a matrix. The peak-average molar mass (M p ) values were assigned (arrow) approximately at the center of the bandwidth at the half-height of each peak.   (Fig. 2b), suggesting that these untargeted agents remained stable by retaining their original sizes in the media containing 10% FBS. The average R g,G (radius of gyration) of the particles (e.g., bovine serum albumin (BSA), R g,G = 3.17 ± 0.05 nm) in FBS, 3.18 ± 0.04 nm (mean ± standard deviation (SD)), was estimated from the slope of the linear scattering data in the q 2 -region using Guinier analysis. Data were not obtained for other untargeted agents due to the limited supply of samples for this study.  (Fig. 2a), suggesting that these compounds remained stable by retaining their original sizes in the media containing 10% FBS. The average R g,G (radius of gyration) of the particles (e.g., BSA) in FBS, 3.18 ± 0.04 nm (mean ± SD), was estimated from the slope of the linear scattering data in the q 2 -region using Guinier analysis. These results are a part of the time-course SPECT imaging series to estimate the tumortargeting efficiency of our nano-sized agents, which are also partially shown (2 hpi, 7 hpi, and 24 hpi) in Fig. 4a in the main text. Here, the targeted agents were injected as radiolabeled with iodine-125. See Experimental Section for details.