Enhanced Tumor Imaging Using Glucosamine-Conjugated Polyacrylic Acid-Coated Ultrasmall Gadolinium Oxide Nanoparticles in Magnetic Resonance Imaging

Owing to a higher demand for glucosamine (GlcN) in metabolic processes in tumor cells than in normal cells (i.e., GlcN effects), tumor imaging in magnetic resonance imaging (MRI) can be highly improved using GlcN-conjugated MRI contrast agents. Here, GlcN was conjugated with polyacrylic acid (PAA)-coated ultrasmall gadolinium oxide nanoparticles (UGONs) (davg = 1.76 nm). Higher positive (brighter or T1) contrast enhancements at various organs including tumor site were observed in human brain glioma (U87MG) tumor-bearing mice after the intravenous injection of GlcN-PAA-UGONs into their tail veins, compared with those obtained with PAA-UGONs as control, which were rapidly excreted through the bladder. Importantly, the contrast enhancements of the GlcN-PAA-UGONs with respect to those of the PAA-UGONs were the highest in the tumor site owing to GlcN effects. These results demonstrated that GlcN-PAA-UGONs can serve as excellent T1 MRI contrast agents in tumor imaging via GlcN effects.


Introduction
Tumor diagnosis is a major challenge in the medical field [1,2]. Among various imaging techniques, magnetic resonance imaging (MRI), a noninvasive imaging technique in which radiofrequency proton spin signals are processed, has been widely applied in tumor diagnosis because of its high anatomical resolution and sensitivity due to the ample existence of protons in the body [3,4].
Tumor imaging using MRI can be improved via active and/or passive tumor-targeting methods, such as drug delivery [5,6]. For active targeting [5][6][7], tumor-targeting ligands are conjugated with imaging agents, while for passive targeting [5,6,8], the enhanced permeability and retention (EPR) effects of nanoparticle imaging agents are utilized, implying that nanoparticle agents can allow active and passive targeting. In addition, tumor imaging can be improved using conjugation ligands, which are highly consumed in the metabolic process of tumor cells. For example, glucose as the food of cells is more highly consumed by tumor cells than by normal cells because of the higher metabolic activity of the tumor from 1.0 to 3.0 nm. The insets show the magnified HRTEM images on a 2 nm scale. The average particle diameter (d avg ) was estimated to be 1.76 nm for the PAA-UGONs and GlcN-PAA-UGONs from log-normal function fits to the observed particle diameter distributions (Figure 1c,d and Table 1). Energy-dispersive X-ray spectroscopy (EDS) confirmed the presence of C, O, and Gd in the PAA-UGONs ( Figure 1e)    Hydrodynamic diameters were determined using dynamic light scattering (DLS) patterns ( Figure 2a). The average hydrodynamic diameters (a avg ) were estimated to be 9.2 and 10.6 nm from the log-normal function fits to the observed hydrodynamic diameter distributions of the PAA-UGONs and GlcN-PAA-UGONs, respectively ( Figure 2a and Table 1). The a avg values are larger than the d avg values because of hydrophilic surface coating and accompanying hydration by numerous water molecules around nanoparticles. The a avg of the GlcN-PAA-UGONs was slightly larger than that of the PAA-UGONs because of additional GlcN coating in the GlcN-PAA-UGONs. These large a avg values in both nanoparticles suggested that the nanoparticles attracted a large number of water molecules around them via the hydrophilic -COOH groups of PAA in the PAA-UGONs and via the hydrophilic -OH groups of GlcN and -COOH groups of PAA in the GlcN-PAA-UGONs. This explains the observed excellent colloidal stability in aqueous media; the nanoparticle colloids never settled down after the synthesis (>1 year). High zeta potentials (ξ) were observed for the PAA-UGONs and GlcN-PAA-UGONs in aqueous media, i.e., −36.0 and −30.7 mV, respectively ( Figure 2b and Table 1), confirming the observed excellent colloidal stability of the nanoparticles, as observed in similar nanoparticles grafted with hydrophilic polymers [31,39]. The zeta potential of the GlcN-PAA-UGONs was slightly lower than that of the PAA-UGONs. This was probably because the -OH groups of GlcN in the GlcN-PAA-UGONs were less negative than the -COOH group of PAA in the PAA-UGONs. The aqueous nanoparticle solution samples of the PAA-UGONs and GlcN-PAA-UGONs are shown in Figure 2c, exhibiting transparency, owing to the excellent colloidal stability of the samples. The Tyndall effect (light scattering by colloids) was observed only for the solution samples (middle vial for the PAA-UGONs and right vial for the GlcN-PAA-UGONs, as shown in Figure 2d), but not for the triple-distilled water (left vial, as shown in Figure 2d), confirming the colloidal dispersions in aqueous media.

Crystal Structures
X-ray diffraction (XRD) patterns of the synthesized nanoparticle powder samples were recorded prior to and after thermogravimetric analysis (TGA) (Figure 3). The XRD patterns prior to the TGA did not exhibit sharp peaks because the nanoparticles were not fully crystallized, owing to their ultrasmall particle sizes. However, the XRD patterns obtained after TGA exhibited sharp peaks of body-centered cubic (bcc) Gd 2 O 3 [40]. This was attributed to both particle size and crystal growth during the TGA up to 900 • C. The lattice constant of the TGA-treated powder samples was estimated to be 10.814 Å, which were consistent with a reported value (10.813 Å) [40].

Surface-Coating Results
The surface coating of the UGONs with PAA and the successful conjugation of GlcN with PAA in the PAA-UGONs via the amide bond were confirmed using the Fourier transform infrared (FT-IR) absorption spectra of PAA, GlcN, PAA-UGONs, and GlcN-PAA-UGONs ( Figure 4a). The C=O stretching vibration at 1700 cm −1 in PAA was red-shifted and split into two peaks in the PAA-UGONs and GlcN-PAA-UGONs (COO − antisymmetric stretching vibration at 1538 cm −1 and COO − symmetric stretching at 1401 cm −1 ), confirming the successful coating of PAA on the UGON surface. This red-shift was due to the hard acid-hard base type of bonding between the COO − group of PAA (hard base) and Gd 3+ of the UGONs (hard acid) [41][42][43], and the splitting was due to the bridge-bonding between the COO − group and Gd 3+ [44]. The C-H stretching vibration at 2957 cm −1 in PAA appeared at 2945 cm −1 in the PAA-UGONs and GlcN-PAA-UGONs, confirming the surface coating of UGON with PAA again. The N-H bending and C-N stretching vibrations at 1540 and 1415 cm −1 , respectively, in GlcN overlapped with the COO − antisymmetric and symmetric stretching vibrations in the GlcN-PAA-UGONs. The C-O stretching vibration at 1026 cm −1 in GlcN appeared at 1060 cm −1 in the GlcN-PAA-UGONs. The observed FT-IR absorption frequencies were consistent with those in previous studies [30,[45][46][47], and are summarized in Table 2.

Crystal Structures
X-ray diffraction (XRD) patterns of the synthesized nanoparticle powder samples were recorded prior to and after thermogravimetric analysis (TGA) (Figure 3). The XRD patterns prior to the TGA did not exhibit sharp peaks because the nanoparticles were not fully crystallized, owing to their ultrasmall particle sizes. However, the XRD patterns obtained after TGA exhibited sharp peaks of body-centered cubic (bcc) Gd2O3 [40]. This was attributed to both particle size and crystal growth during the TGA up to 900 °C. The lattice constant of the TGA-treated powder samples was estimated to be 10.814 Å, which were consistent with a reported value (10.813 Å) [40].

Surface-Coating Results
The surface coating of the UGONs with PAA and the successful conjugation of GlcN with PAA in the PAA-UGONs via the amide bond were confirmed using the Fourier transform infrared (FT-IR) absorption spectra of PAA, GlcN, PAA-UGONs, and GlcN-PAA-UGONs (Figure 4a). The C=O stretching vibration at 1700 cm −1 in PAA was   The surface-coating amounts (P) of PAA in the PAA-UGONs and GlcN-PAA in the GlcN-PAA-UGONs were measured via TGA and elemental analysis (EA). As shown in the TGA curves (Figure 4b), the surface-coating amounts in wt.% were estimated to be 41.7 and 50.3% for the PAA-UGONs and GlcN-PAA-UGONs, respectively, from the mass drops after considering the initial mass drops between room temperature and~105 • C, owing to water and air desorption ( Table 1). The remaining masses were due to the UGONs ( Figure 4b and Table 1). The difference in the coating amount (8.6%) between the PAA-UGONs and GlcN-PAA-UGONs was due to the GlcN in the GlcN-PAA-UGONs. For EA, the surfacecoating amounts in wt.% were estimated to be 48. The difference (6.2%) was due to the GlcN in the GlcN-PAA-UGONs, as previously mentioned. The higher p values of EA, compared with those of TGA, were due to the water and air contributions to the P in the EA because all elements in the sample except for UGONs were measured during the EA. Using the TGA data, the grafting density (σ) of the PAA-UGONs, corresponding to the average number of PAA polymers coating a UGON unit surface area [48], was estimated to be 0.64 nm −2 using the bulk density of Gd 2 O 3 (7.407 g/cm 3 ) [49], p value previously estimated, and d avg determined from the HRTEM imaging. By multiplying σ by the UGON surface area (πd avg 2 ), the average number (N NP ) of PAA polymers per UGON was estimated to be~6. To estimate the number of GlcN molecules per UGON in the GlcN-PAA-UGONs, the molecular weight of GlcN-PAA increased in the GlcN unit until the N NP was 6 or σ was 0.64 because the N NP and σ of PAA in the GlcN-PAA-UGONs should be the same as those in the PAA-UGONs. Approximately, five GlcN molecules per PAA were conjugated. Thus,~30 GlcN molecules per UGON were conjugated. The surface-coating results are summarized in Table 1.

Magnetic Properties
The magnetic properties of the UGONs were investigated by measuring the magnetization (M) versus applied field (H) (or M-H) curves (−2.0 T ≤ H ≤ 2.0 T) at 300 K using a vibrating sample magnetometer (VSM) ( Figure 5). The measured M values were mass-corrected using net masses of UGONs without ligands, which were obtained from the net masses of UGONs in the TGA curves. As shown in Figure 5, all nanoparticle samples were paramagnetic, exhibiting no hysteresis, small unsaturated M values, zero remanence, and zero coercivity, similar to bulk Gd 2 O 3 [50,51]. From the mass-corrected M-H curves, the unsaturated net M values of the UGONs at H = 2.0 T were estimated to be 1.89 and 1.98 emu/g for the PAA-UGONs and GlcN-PAA-UGONs, respectively (Table 3). Therefore, the average M value for the UGONs was 1.94 emu/g. This appreciable M value at room temperature was due to a high spin magnetic moment (s = 7/2) of Gd 3+ [29] and was attributable to high r 1 values of Gd 3+ -based MRI contrast agents [29][30][31][32][33][34].

r 1 and r 2 Values and R 1 and R 2 Map Images
The r 1 and r 2 values and longitudinal (R 1 ) and transverse (R 2 ) map images were measured at H = 1.5 and 3.0 T MR fields; the r 1 and r 2 values were estimated from the plots of inverse longitudinal (T 1 ) and transverse (T 2 ) water proton spin relaxation times, 1/T 1 and 1/T 2 as a function of the Gd concentration, respectively (Figure 6a,b and Table 3). The r 1 and r 2 values increased with an increase in H from 1.5 to 3.0 T (Table 3). This was because they were proportional to M 2 [52,53], and M increased with an increase in H because the UGONs were paramagnetic (refer to M-H curves in Figure 5). As shown in Figure 6c, dose-dependent contrast enhancements were observed in the R 1 and R 2 map images for both samples in aqueous media. This demonstrated in vitro that both samples induced T 1 and T 2 water proton spin relaxations. However, both nanoparticles were more suitable as T 1 MRI contrast agents rather than as T 2 MRI contrast agents because their r 2 /r 1 ratios were close to one, and their r 1 values were extremely high (3 to 4 times higher than those [29] of commercial Gd-chelates).

In Vitro Cellular Cytotoxicity Results
The biocompatibility of the PAA-UGONs and GlcN-PAA-UGONs was investigated by measuring the in vitro cell viabilities in normal mouse hepatocyte (NCTC1469) and human prostate tumor (DU145) cell lines (Figure 7). The cytotoxicity results up to 0.5 mM Gd are consistent with those measured in PAA-coated lanthanide oxide nanoparticles [39,54], indicating the suitability of PAA as an excellent biocompatible surface-coating ligand. The IC50 (the inhibitory concentration of chemicals that cause 50% of the maximum inhibition) of the PAA-UGONs and GlcN-PAA-UGONs were estimated to be 2.75 and 0.78 mM Gd in DU145 cell lines, respectively, and 1.53 and 1.30 mM Gd in NCTC1469 cell lines, respectively (Figure 7c). These values are more or less consistent with 1.93 mM (or 304 µg/mL) of gadolinium oxide nanoparticles prepared via thermal decomposition method in human umbilical vein endothelial (HUVEC) cells [55].

In Vivo Tumor Imaging: T 1 MR Images in U87MG Tumor-Bearing Mice
In vivo T 1 MR images of the U87MG tumor-bearing mice were obtained prior to (labeled as "0") and after the intravenous injection of aqueous solution samples of the GlcN-PAA-UGONs and PAA-UGONs as control into the tail veins ( Figure 8). As shown in Figure 8, positive contrast enhancements were observed after the injection in various organs (liver, kidney, and bladder), including the tumor site, for all samples. However, the contrast enhancements of the GlcN-PAA-UGONs were higher than those of the PAA-UGONs, except for the bladder because the PAA-UGONs were rapidly excreted through the renal system within~4 h after injection. In addition, the contrast enhancements of the GlcN-PAA-UGONs were retained longer in all organs, including the tumor site, compared with those of the PAA-UGONs. These results were attributable to the GlcN effects in the GlcN-PAA-UGONs and are quantitatively discussed in the following section, by measuring the signal-to-noise ratios (SNRs) of regions of interest (ROIs) in T 1 MR images.

Discussion
The r 1 and r 2 values of the GlcN-PAA-UGONs were slightly lower than those of the PAA-UGONs (Table 3). This was probably because the PAA with many -COOH groups attracted a larger number of water molecules around the UGONs than GlcN with -OH groups. In addition, the hydrodynamic diameter of the GlcN-PAA-UGONs was slightly larger than that of the PAA-UGONs (Table 1), owing to the extra GlcN coating in the GlcN-PAA-UGONs as previously mentioned, rendering the water molecules in the GlcN-PAA-UGONs slightly farther apart from the UGONs than those in the PAA-UGONs. Based on these, hypothesized distributions of water molecules around the nanoparticle are drawn in Figure 9. The first case allows the GlcN-PAA-UGONs to interact with a fewer number of water proton spins than the PAA-UGONs, and the second case allows the GlcN-PAA-UGONs to less strongly interact with water proton spins than the PAA-UGONs, resulting in lower r 1 and r 2 values in the GlcN-PAA-UGONs. To quantitatively determine the enhanced tumor imaging of GlcN-PAA-UGONs via GlcN effects, the SNRs of ROIs (labeled as small circles in the T 1 MR images prior to injection, as shown in Figure 8) at the tumor site, liver, kidney, and bladder were estimated and subtracted from those prior to injection to estimate the contrast enhancements [=SNR-ROI (time) − SNR-ROI (0)]. The contrast enhancements were plotted as a function of time in Figure 10a,b for the PAA-UGONs and GlcN-PAA-UGONs, respectively. As shown in Figure 10a, most of the PAA-UGONs were excreted via the renal system within 4 h after injection because of their ultrasmall particle sizes, which can be noticed from considerably higher contrast enhancements in the bladder, compared with those in other organs, including the tumor site. However, the GlcN-PAA-UGONs exhibited no such drastic contrast enhancements in the bladder (Figure 10b), indicating their slower excretion through the renal system, compared with PAA-UGONs, owing to their uptakes by GLUT and SGLT transporters expressed on the normal and tumor cell membranes [13][14][15]. Consequently, apart from the bladder (Figure 10c Figure 10g. Higher areas below the curves for the GlcN-PAA-UGONs, compared with those of the PAA-UGONs in each organ and the tumor site except for the bladder were observed, implying higher accumulations of GlcN-PAA-UGONs than PAA-UGONs in these organs and tumor site, owing to the aforementioned reason. Larger areas of organs (liver and kidney) below the curves than those of the tumor site for the PAA-UGONs and GlcN-PAA-UGONs were probably because the nanoparticles were excreted through these organs. Importantly, the area ratios of the GlcN-PAA-UGONs to the PAA-UGONs (Figure 10h) was the highest at the tumor site, implying the highest accumulations of the GlcN-PAA-UGONs with respect to those of PAA-UGONs at the tumor site, which is due to GlcN effects. It is well-known that tumor cells grow via angiogenesis after tumor cell inoculation into mice [56,57] and that nanoparticles accumulate at tumor site via EPR effects [5,6,8], thus enhancing T 1 contrasts in MR images. The EPR effects are the same for the GlcN-PAA-UGONs and PAA-UGONs, but the GlcN-PAA-UGONs have the additional GlcN effects. Using these two effects, the observed contrast enhancements are schematically illustrated in Figure 11 using the number (N) for the UGONs. As shown in Figure 11, the N tumor cell (GlcN-PAA-UGONs) > N normal cell (GlcN-PAA-UGONs) is due to GlcN and EPR effects and the N tumor cell (PAA-UGONs) > N normal cell (PAA-UGONs) is due to EPR effects. As the contrast enhancement is proportional to N, the observed highest contrast enhancements in the tumor site (Figure 10h) can be explained as due to GlcN effects. Therefore, the GlcN-PAA-UGONs can highly enhance contrasts in tumor site via the additional GlcN effects. GlcN is not tumor-specific; therefore, GlcN-PAA-UGONs can be ap-plied to any tumor type imaging. As shown in Figure 7c, the GlcN-PAA-UGONs exhibited lower IC50 values than the PAA-UGONs for the same cell lines, indicating higher toxicities than PAA-UGONs for the same cell lines, likely due to higher accumulation of GlcN-PAA-UGONs in tumor DU145 and normal NCTC1469 cells than PAA-UGONs because of GlcN. Notably, IC50 value of the GlcN-PAA-UGONs in DU145 cells was the lowest among four IC50 values. Assuming that the cellular toxicity is proportional to the number (N) of UGONs accumulated into cells (i.e., N∝1/IC50), N DU145 (GlcN-PAA-UGONs)/N DU145 (PAA-UGONs) = 3.53 > N NCTC1469 (GlcN-PAA-UGONs)/N NCTC1469 (PAA-UGONs) = 1.18, also explaining the GlcN effects.

One-Pot Polyol Synthesis of the PAA-UGONs
The PAA-UGONs were synthesized using a simple one-pot polyol method (Figure 12a). First, 2 mmol of GdCl 3 ·6H 2 O was added to 15 mL of TEG in a three-necked round-bottom flask, and the mixture solution was magnetically stirred at 60 • C under atmospheric conditions until the precursor dissolved in TEG (it lasted for~2 h). In a separate beaker, 10 mmol of NaOH in 10 mL of TEG was prepared. The NaOH solution was added to the aforementioned precursor solution until the pH of the solution was within a range of 9-11. The reaction temperature increased to 120 • C, and the mixture solution was magnetically stirred for 4 h. Thereafter, 0.25 mmol of PAA was added to the reaction solution with magnetic stirring for 14 h. The product solution containing the PAA-UGONs was cooled to room temperature and transferred into a 500 mL beaker. Subsequently, 400 mL of ethanol was added to the product solution with magnetic stirring for 10 min. The product solution was preserved in a refrigerator until the PAA-UGONs settled to the beaker bottom. The top transparent solution was removed, and the remaining product solution was washed again with ethanol using the same process thrice. The product solution was dialyzed (MWCO =~2000 Da) against 1.5 L of triple-distilled water for one day to remove the remaining impurities, such as Gd 3+ , Cl − , Na + , TEG, ethanol, and PAA from the product solution.

Synthesis of the GlcN-PAA-UGONs
The -NH 2 group of GlcN was conjugated with the -COOH group of PAA in the PAA-UGONs via the EDC/NHS coupling method (Figure 12b). The prepared PAA-UGONs were dispersed in 40 mL of triple-distilled water in a 250 mL beaker. Thereafter, 1 mmol of EDC and 2 mmol of NHS were added to the aforementioned solution at room temperature under atmospheric conditions with magnetic stirring for 30 min (the solution pH was~6). Afterward, 10 mmol of GlcN prepared in 4 mL of triple-distilled water was added to the solution with magnetic stirring for 20 min. Subsequently, a 0.1 M NaOH solution was slowly added to the solution to obtain a solution pH of~7. The solution was magnetically stirred for 24 h. The product solution containing the GlcN-PAA-UGONs was dialyzed (MWCO =~2000 Da) against 1.5 L of triple-distilled water for one day to remove the remaining impurities, such as Na + , Cl − , EDC, NHS, and GlcN from the product solution.

Physicochemical Property Characterization
The nanoparticle diameter was measured using an HRTEM (Titan G2 ChemiSTEM CS Probe; FEI, Hillsboro, OR, USA) at 200 kV acceleration voltage. For the measurements, a drop of the diluted colloidal nanoparticle sample dispersed in ethanol was dropped onto a carbon film supported by a 200-mesh copper grid (Pelco No. 160, Ted Pella Inc., Redding, CA, USA) using a micropipette (2-20 µL, Eppendorf, Hamburg, Germany) and allowed to dry in air at room temperature. The copper grid with the nanoparticle sample was placed inside the HRTEM for measurements. An EDS instrument (Quantax Nano, Bruker, Berlin, Germany) installed in the HRTEM was used to analyze elements (C, N, O, and Gd) in the nanoparticle samples. The Gd concentration of the nanoparticle samples in aqueous media was determined via inductively coupled plasma-atomic emission spectroscopy (ICP-AES) (IRIS/AP, Thermo Jarrell Ash Co., Waltham, MA, USA). A DLS particle size analyzer (Zetasizer Nano ZS, Malvern, Malvern, UK) was used to measure the hydrodynamic diameter

In Vitro Cellular Cytotoxicity Measurements
The in vitro cellular cytotoxicity of the aqueous nanoparticle suspension samples was measured using a CellTiter-Glo Luminescent Cell Viability Assay (Promega, Madison, WI, USA). The intracellular adenosine triphosphate was quantified using a Victor 3 luminometer (Perkin Elmer, Waltham, MA, USA). Two cell lines, NCTC1469 and DU145, were used. Each cell line was seeded onto a separate 24-well cell culture plate and incubated for 24 h (5 × 10 4 cell density, 500 µL cells/well, 5% CO 2 , and 37 • C). Nine test solutions (0.01, 0.05, 0.1, 0.2, 0.5, 1, 2, 3, 5 mM Gd) were prepared by diluting the original concentrated nanoparticle suspension sample dispersed in triple-distilled water with a sterile phosphatebuffered saline (PBS) solution. Afterward, 2 µL aliquots were used to treat the cells, which were subsequently incubated for 48 h. Cell viabilities were measured thrice to obtain average cell viabilities, which were normalized with respect to that of the untreated control cells (0.0 mM Gd).

Water Proton Spin Relaxivity and Map Image Measurements
The T 1 and T 2 water proton spin relaxation times and the R 1 and R 2 map images were measured using a 1.5 T MRI scanner (GE 1.5 T Signa Advantage, GE Medical Systems, Chicago, IL, USA) equipped with a knee coil (EXTREM) and 3.0 T MRI scanner (Magnetom Trio Tim, Siemens, Munich, Bayern, Germany). Aqueous dilute solutions (1, 0.5, 0.25, 0.125, and 0.0625 mM Gd) were prepared by diluting the original concentrated solutions with triple-distilled water. These dilute solutions were used to measure the T 1 and T 2 relaxation times and R 1 and R 2 map images. The r 1 and r 2 water proton spin relaxivities were estimated from the slopes of the plots of 1/T 1 and 1/T 2 versus Gd concentration, respectively. T 1 relaxation time measurements were conducted using an inversion recovery method. In this method, the inversion time (TI) was varied, and the MR images were acquired at 35 different TI values in the range of 50-1750 ms. The T 1 relaxation times were obtained from the nonlinear least-square fits to the measured signal intensities at various TI values. For the measurements of T 2 relaxation times, the Carr-Purcell-Meiboom-Gill pulse sequence was used for multiple spin-echo measurements, and 34 images were acquired at 34 different echo time (TE) values in the range of 10-1900 ms. The T 2 relaxation times were obtained from the nonlinear least-square fits to the mean pixel values of the multiple spin-echo measurements at various TE values.

Animal Experiments
All in vivo experiments on mice were performed following the rules, regulations, and permission of the animal research committee of the Korea Institute of Radiological and Medical Sciences (approval number: Kirams2018-0072 and approval date: 9 January 2019).

Tumor Model Nude Mice Preparation
The U87MG tumor cells were incubated for 24 h at 37 • C in air containing 5% CO 2 . Roswell Park Memorial Institute (RPMI-1640) containing 10% (v/v) fetal bovine serum and 1% (v/v) penicillin streptomycin was used as the culture medium of the cells. BALB/c nude mice (male, 5-week old, 20 g) were administered inoculation into their subcutaneous tissue in one of their hind legs (thighs) with 5 × 10 6 U87MG tumor cells suspended in 100 µL of PBS solution. In vivo MRI experiments were conducted three weeks after the tumor cell inoculation.

In Vivo T 1 MR Image Measurements
In vivo T 1 MR images were acquired using a 3.0 T MRI scanner (Magnetom Trio Tim, Siemens, Munich, Bayern, Germany). For imaging, U87MG tumor-bearing mice were anesthetized with 1.5% isoflurane in oxygen. The aqueous nanoparticle suspension samples (PAA-UGONs and GlcN-PAA-UGONs) were injected into the tail veins of the mice (0.1 mmol Gd/kg) as a bolus: two mice were used for each nanoparticle sample. All mice recovered after injection. Measurements were performed prior to and after the injection. During the measurements, the body temperature of the mice was maintained at 37 • C using a warm water blanket. The spin-echo sequence was used to obtain T 1 MR images. The typical measurement parameters used for coronal (or axial) image measurements are as follows: H = 3.0 T, TE = 10 (9.3) ms, repetition time = 385 (455) ms, echo train length = 3 (3 mm, pixel bandwidth = 299 (299) Hz, flip angel = 120 (120) degree, width = 41.875 (60) mm, height = 60 (45) mm, number of acquisitions = 8 (4), field of view = 70 (70) mm, slice thickness = 1.0 (1.5) mm, and spacing = 1.1 (3.75) mm. The numbers in parentheses are those used for the axial image measurements.

Conclusions
Using GlcN effects, the enhanced tumor imaging of GlcN-PAA-UGONs was investigated using tumor model nude mice. The PAA-UGONs were used as control. The results are summarized as follows.
(2) The PAA-UGONs and GlcN-PAA-UGONs exhibited excellent colloidal stability in aqueous media (no precipitation after synthesis, >1 year) and low cellular toxicities up to 0.5 mM Gd. (3) The PAA-UGONs and GlcN-PAA-UGONs exhibited three to four times higher r 1 values than those of commercial Gd-chelates, and their r 2 /r 1 ratios were close to one, indicating that they are potential high-performance T 1 MRI contrast agents. (4) The GlcN-PAA-UGONs exhibited higher contrast enhancements at various organs, including the tumor site, compared with the PAA-UGONs, and such contrast enhancements were the highest at the tumor site, owing to the GlcN effects. Consequently, the GlcN-PAA-UGONs can be applied to highly enhancing contrasts in tumor. Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.