4-Sulfobutyl

Article Cite This: Bioconjugate Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/bc

Gd3+:DOTA-Modified 2‑Hydroxypropyl-β-Cyclodextrin/4-Sulfobutyl Ether-β-Cyclodextrin-Based Polyrotaxanes as Long Circulating High Relaxivity MRI Contrast Agents Yawo A. Mondjinou,† Bradley P. Loren,† Christopher J. Collins,† Seok-Hee Hyun,† Asher Demoret,† Joseph Skulsky,† Cheyenne Chaplain,† Vivek Badwaik,§ and David H. Thompson*,†,‡,§ Department of Chemistry, ‡Weldon School of Biomedical Engineering, and §Center for Cancer Research, Multi-disciplinary Cancer Research Facility, Bindley Bioscience Center, Purdue University, 1203 West State Street, West Lafayette, Indiana 47907, United States

Downloaded via UNIV OF SUNDERLAND on November 8, 2018 at 09:43:59 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

S Supporting Information *

ABSTRACT: A family of five water-soluble Gd3+:1,4,7,10-tetraazacyclododecane-1,4,7-tetraacetic acid-modified polyrotaxane (PR) magnetic resonance contrast agents bearing mixtures of 2-hydroxypropyl-β-cyclodextrin and 4-sulfobutylether-β-cyclodextrin macrocycles threaded onto Pluronic cores were developed as long circulating magnetic resonance contrast agents. Short diethylene glycol diamine spacers were utilized for linking the macrocyclic chelator to the PR scaffold prior to Gd3+ chelation. The PR products were characterized by 1H NMR, gel permeation chromatography/ multiangle light scattering, dynamic light scattering, and analytical ultracentrifugation. Nuclear magnetic relaxation dispersion and molar relaxivity measurements at 23 °C revealed that all the PR contrast agents displayed high spin−spin T1 relaxation and spin−lattice T2 relaxation rates relative to a DOTAREM control. When injected at 0.05 mmol Gd/kg body weight in BALB/c mice, the PR contrast agents increased the T1-weighted MR image intensities with longer circulation times in the blood pool than DOTAREM. Excretion of the agents occurred predominantly via the renal or biliary routes depending on the polyrotaxane structure, with the longest circulating L81 Pluronic-based agent showing the highest liver uptake. Proteomic analysis of PR bearing different β-cyclodextrin moieties indicated that lipoproteins were the predominant component associated with these materials after serum exposure, comprising as much as 40% of the total protein corona. We infer from these findings that Gd(III)-modified PR contrast agents are promising long-circulating candidates for blood pool analysis by MRI.

■

INTRODUCTION Interest in magnetic resonance imaging (MRI) has grown rapidly over the past four decades. As a noninvasive and nonradiative modality that provides deep tissue penetration and high temporal and spatial resolution, it is widely used for revealing dynamic biological processes and facilitating clinical diagnoses.1,2 Contrast agents are often used to enhance relaxation by decreasing the spin−spin or longitudinal (T1) and spin−lattice or transverse (T2) relaxation times.3 Contrast agents can consist of paramagnetic or superparamagnetic metal ions; however, Gd3+ complexes are efficient and commonly used T1-weighted agents that produce positive images. A manganese-based contrast agent has also been recently reported.4 Among the intrinsic parameters that control the relaxation rate of contrast agents, the molecular rotation time and the number of water molecules coordinated in the inner coordination sphere are the most important.5−8 Most commercial T1 weighted contrast agents are based on either acyclic diethylenetriaminepentaacetic acids (e.g., Gd-DTPA, Magnevist; Gd-BOPTA, Multihance; Gd-DTPA-BMA, Om© XXXX American Chemical Society

niscan; Gd-DTPA-BMEA, OptiMARK; and Gd-EOB-DTPA, Eovist) or macrocyclic ligands such as 1,4,7,10-tetraazacyclododecane (i.e., Gd-DOTA, DOTAREM; Gd-HP-DO3A, ProHance; and Gd-DO3A-butrol, Gadovist).9 These lowmolecular-weight MR agents tumble rapidly in solution, leading to low relaxivities in the range of 4−5 mM−1 s−1 in magnetic fields ranging from 0.3 to 7 T.6,10 Unfortunately, these clinical contrast agents also suffer from rapid renal clearance that seriously limit the MRI scanning time window. Low-molecular-weight agents can also extravasate from the vascular compartment, thereby reducing contrast with surrounding tissues. Consequently, various efforts have been devoted to the development of macromolecular and supramolecular Gd(III)-based contrast agents including dendrimers, 11−13 micellar/liposomal carriers and CEST agents,14−18 peptides,19 poly(L-lysines),20 polyethylene glycol Received: July 24, 2018 Revised: September 27, 2018

A

DOI: 10.1021/acs.bioconjchem.8b00525 Bioconjugate Chem. XXXX, XXX, XXX−XXX

Article

Bioconjugate Chemistry

Figure 1. Structure of Gd:DOTA-HPβCD/SBEβCD Pluronic-based polyrotaxane contrast agents with cholesteryl end-caps. Blue conical structure = HPβCD. Yellow conical structure = SBEβCD. Pluronic cores = F127 (m = 200, n = 65); F68 (m = 151, n = 29); L35 (m = 21, n = 16); L64 (m = 26, n = 30); L81 (m = 6, n = 43).

(PEG),21 albumin,22,23 antibodies,24,25 and carbohydrates (e.g., dextran or starch).26−28 These macromolecular agents were found to limit extravasation across a healthy vascular endothelium and have comparatively enhanced circulation times relative to low-molecular-weight agents. Ideal macromolecular contrast agents should have high relaxivities, persist in the blood pool, and display high kinetic stability to avoid toxicity such as nephrogenic systemic fibrobrosis (NSF) in patients with renal impairment.29 Polyrotaxanes (PR) are self-assemblies obtained by threading macrocyclic molecules such as cyclodextrins onto polymer axles that noncovalently retain the macrocycles on the polymer core through bulky stoppers that prevent dethreading. These noncovalent polymer constructs have attracted wide interest due to their potential utility in sliding ring gels,30,31 molecular machines,32 and, as degradable, low toxicity carriers for drugs and gene delivery.33−35 They also provide flexibility for functionalization by virtue of the multiple hydroxyl groups present on the threaded cyclodextrin macrocycles. Since the first α-cyclodextrin (αCD):PEG-based PR was reported by Harada et al.,36,37 many subsequent efforts have been dedicated to the design of various Pluronic-based PR using βCD and its derivatives.38,39 In spite of these efforts, the biological imaging applications of these supramolecular constructs have been comparatively limited. In the work reported by Hosseini et al., superparamagnetic iron oxide nanoparticles (SPION) nanoparticles were coated with αCD:PEG-based PR for particle stabilization and postfunctionalization of superparamagnetic T2

weighted MR agents.40 Fredy and co-workers have developed polyrotaxanes as MRI contrast agents,41,42 including a bimodal MR and fluorescent PR with functionalized α-CD and poly(N,N-dimethylamonio)undecamethylene polyamine polymer that displayed a 5-fold improvement in relaxivity relative to Gd-DOTA (DOTAREM).41 A long circulating, degradable Gd3+-DO3A-HPβCD:Pluronic 127 PR that has the intravascular imaging capabilities of a macromolecular contrast agent, while potentially retaining the renal elimination properties of a small molecule agent after end-cap cleavage has also been reported.43 This construct was shown to circulate for more than 30 min and provide >100-fold vascular enhancement relative to the monomeric Gd 3+ -DO3AHPβCD control that was rapidly cleared via the kidney. In the present study, a family of five water-soluble 2hydroxypropyl-β-cyclodextrin (HPβCD) and 4-sulfobutylether-β-cyclodextrin (SBEβCD) PR have been developed using five different Pluronic polymer cores with varying molecular weights and ratios of their PEG and polypropylene (PPG) blocks. We envisioned that the threading efficiency, polymer molecular weight, and PEG block sizes would play an important role in the observed contrast agent relaxivity and tissue contrast enhancement, in part due to their varying rodlike conformational propensities.44−46 Because of the low water solubility of our previously reported HPβCD:Pluronic PR,39 SBEβCD was blended with HPβCD in the cyclodextrin threading mixture to help circumvent the low aqueous solubility of those PR derivatives. Dramatically improved B


Article

Bioconjugate Chemistry Table 1. Summary of Gd:DOTA-HPβCD/SBEβCD PR Contrast Agent Properties Polyrotaxane Gd3+-HPβCD/SBEβCD F127 Gd3+-HPβCD/SBEβCD F68 Gd3+-HPβCD/SBEβCD L35 Gd3+-HPβCD/SBEβCD L64 Gd3+-HPβCD/SBEβCD L81

# of CDa

# of HPβCDa

# of SBEβCDa

% CD Coverage

# of DOTAb

% Gd Contentc

MWd (NMR)

MWd (GPC)

MWd (AUC)

10

6

4

31

12

4.5

39.2

34.1

33.3

5

3

2

38

15

29.5

24.8

11.8

16

10

6

100

14

6.2

40.2

52.6

29.1

11

6

5

73

22

8.0

39.1

30.6

11.8

20

10

10

91

22

5.2

54.7

49.4

nd

11

a

Determined by 1H NMR before Gd3+ complexation. bEstimated by NMR integration of the DOTA phenyl protons vs the PPG methyl protons. ICP-MS analysis. dValues reported in kD.

c

water solubilities of Pluronic L81 PR have been observed when SBEβCD was included in threading mixtures with various βCD derivatives (including βCD, HPβCD, methyl-βCD, and azidoβCD).44 This effect is attributed to the presence of SBEβCD negative charges that limit intermolecular aggregation of the corresponding PR. These constructs can then be postfunctionalized with a macrocyclic chelator and loaded with Gd3+ to produce MR contrast agents that can potentially degrade into excretable, low toxicity βCD and Pluronic components. The relaxivity, vascular enhancement, and protein corona properties of this family of Gd3+:DOTA-HPβCD/SBEβCD Pluronic polyrotaxanes (Gd:DOTA-HPβCD/SBEβCD PR) are reported. Our data show that the L81 variant of this library is the most efficient MR agent relative to DOTAREM. We infer from these findings that Gd:DOTA-HPβCD/SBEβCD L81 PR has significant utility as a blood pool contrast agent for vascular and potential tumor imaging.

■

be seen most clearly for L-35, L-64, and L-81 (PPG:PEG ratios of 0.72, 1.2, and 7.2, respectively) that displayed threading efficiencies of 100%, 73%, and 91%, respectively. The SBEβCD proportions in the PR ranged between 40 and 50 mol %, indicating that the ionic βCD threads more efficiently for reasons that that are unclear at this time. The HPβCD/ SBEβCD:Pluronic PR intermediates were then activated with CDI before modification with an excess of 1,10-diamino-4,7dioxadecane to increase their water solubility and enable subsequent conjugation with DOTA-Bn-SCN, such that loading with GdCl3 provides the final Gd:DOTA-HPβCD/ SBEβCD PR (Figure 1, Table 1). The PR molecular weights were estimated by 1H NMR, gel permeation chromatography-multiangle light scattering/refractive index (GPC-MALS/RI), and analytical ultracentrifugation analysis (AUC) in DMSO (Table 1). The values determined by GPC analysis were in general agreement with the molecular weights calculated by 1H NMR analysis; however, the masses determined by AUC were generally lower, presumably due to the presence of low-molecular-weight PR in the polydisperse mixture and/or the occurrence of dethreading during centrifugal analysis. The weight percentage of Gd in the Gd:DOTA-HPβCD/SBEβCD PR samples, evaluated by ICPMS, are indicated in Table 1. The hydrodynamic diameters of the compounds in water at pH 7 ranged between 116 and 230 nm, sizes that suggest aggregation under the conditions of the DLS measurements. We anticipate that these are weakly associated assemblies that dissociate upon introduction into the bloodstream due to apolipoprotein binding of the end-caps and adsorption of other proteins that contribute to corona formation (see below). The zeta potential (ζ) measurements for the PR contrast agents indicated the presence of a slight negative charge due to the presence of SBEβCD sulfonate groups on the PR scaffold (Table S1).44 Molar Relaxivity of Gd:DOTA-HPβCD/SBEβCD PR. The proton nuclear magnetic relaxation dispersion (1H NMRD) was measured to evaluate the T1 relaxivities of the Gd:DOTAHPβCD/SBEβCD PR contrast agents as a function of field strength. The 1H NMRD profiles obtained at 30 °C in H2O over 30 values of magnetic field between 0.24 mT to 0.97 T are shown in Figure 2 for F127, F68, L35, and L81 Pluronic PR contrast agents. A slight increase in T1 relaxivity is observed between 20 and 40 MHz for all of the compounds, with a leveling off of r1 around 30 MHz that is characteristic of macromolecular contrast agents47−49 compared to lowmolecular-weight Gd(III) chelates such as DOTAREM and Prohance that have 1H NMRD profiles lacking incremental r1 increases in the same region. The highest and lowest

RESULTS AND DISCUSSION

Synthesis of HPβCD/SBEβCD:Pluronic PR Contrast Agents. The PR contrast agents shown in Figure 1 were synthesized using an adaptation of the methods described previously.43,44 In brief, a library of HPβCD/SBEβCD:Pluronic PR were initially prepared under heterogeneous conditions39 using TREN-modified F-127, F-68, L-35, L-64, and L-81 Pluronics in the presence of 30:70 mol % SBEβCD:HPβCD that had been thoroughly mixed in the solid state by extensive grinding of the two cyclodextrins into a finely powdered mixture before initiating the polymer threading reaction using hexane as solvent. The pseudopolyrotaxane intermediates generated by this procedure were then end-capped with cholesterol chloroformate to generate the corresponding PR precursors. 1H NMR analysis was used to quantify the number of cyclodextrins threaded onto the Pluronic axles by comparing the integral intensities of the cyclodextrins (HPβCD C1−H at 4.5−5.0 ppm and SBEβCD CH2 at 1.6 ppm) relative to the Pluronic central block signal (PPG CH3 at 1.0 ppm). The coverage ratio was calculated based on the assumption that one βCD molecule was capable of including two PPG units within its cavity. The broadening of peaks in the 4.5−5.0 ppm region confirms that the Pluronic copolymers were successfully threaded with β-CD derivatives (Figures S1−S10). Table 1 summarizes the number of βCD units held by each Pluronic axle, the percent coverage of the PPG block, as well as the percentage of SBEβCD contained in the PR. These data show that the threading efficiency tends to improve for Pluronics having higher PPG:PEG ratios. This can C


Article


Table 2. Molar Relaxivities of Gd:DOTA-HPβCD/SBEβCD PR and DOTAREM Contrast Agent DOTAREM Gd:DOTA-HPβCD/ SBEβCD F127 Gd:DOTA-HPβCD/ SBEβCD F68 Gd:DOTA-HPβCD/ SBEβCD L35 Gd:DOTA-HPβCD/ SBEβCD L64 Gd:DOTA-HPβCD/ SBEβCD L81

Figure 2. 1H NMRD spectra of Gd:DOTA-HPβCD/SBEβCD Pluronic F127, F68, L35, and L81 polyrotaxanes.

r1 (mM−1 s−1) (1H NMRD)

r1 (mM−1 s−1) (1/T1)

r2 (mM−1 s−1) (1/T2)

7.9

3.8 22

3.7 19

12

10

17

6.9

8.2

14

-

7.0

10

11

14

14

The higher relaxivities observed for the Gd:DOTAHPβCD/SBEβCD PR at increased field strength can be attributed to an increase in their τR due to a combined effect of their macromolecular motion and their side chain motions that also influence relaxation rates as described by Lipari and coworkers.52 The r1 and r2 values of the F127, F68, and L81 Gd:DOTA-HPβCD/SBEβCD PR were greater than those of the L35 PR and L64 variants. This can be explained in part by the size of the terminal PEG blocks of the Pluronic cores within the compounds. The large PEG blocks of the F127 and F68 PR derivatives (e.g., 200 and 151 ethylene oxide units, respectively) likely facilitate access of water molecules to the Gd chelate that is appended to the PR βCD units. Kojima et al. observed the same effect of PEG block size on the relaxivity of PEGylated dendrimers.53 In the case of Gd:DOTA-HPβCD/ SBEβCD L81 PR, its high threading coverage (90%) confers a rod-like shape to the PR molecule, thus limiting lateral diffusion of the Gd3+:DOTA-βCD units along the polymer axle and lowering the molecular tumbling rate to shorten its relaxation time as a consequence. In Vitro MR Imaging of Gd:DOTA-HPβCD/SBEβCD PR. To evaluate the signal enhancement properties of these PR constructs, T1-weighted spin−echo MR images were recorded for aqueous solutions of the samples at increasing Gd concentration. DOTAREM and pure water images were also obtained under the same conditions as controls. High positive contrast enhancement was observed for all the Gd:DOTAHPβCD/SBEβCD PR, producing stronger contrast with increasing Gd concentration (Figure S23). It should be noted that the lowest concentrations of the F127 and L35

relaxivities were observed for Gd:DOTA-HPβCD/SBEβCD F68 PR (12.2 mM−1 s−1 at 0.24 mT) and Gd:DOTA-HPβCD/ SBEβCD L35 PR (6.7 mM−1 s−1 at 0.24 mT). These values are relatively low compared to dendrimeric contrast agents due to the comparatively high flexibility of the PR constructs (i.e., free rotation and translation of the Gd3+:DOTA-βCD units around the polymer axle), that induces a long residence time (τM), of water molecules bound to the metal. In general, the residence time should be short to enhance the relaxivity of macromolecular contrast agents.6,50 In addition, the rotational time (τR) of the PR structures are expected to be short, leading to lower relaxivities compared to dendrimers having compact structure with restricted isotropic rotational dynamics that increase τR.51 Nonetheless, evaluation of the PR relaxivities at higher discrete fields (1.5, 3, or 7 T) was necessary to more fully assess their utility as MR contrast agents as these are the field strengths utilized in clinical and animal imaging applications. Fixed field relaxivity measurements were performed at 25 °C on aqueous solutions of the contrast agents at concentrations between 0.05 and 1.0 mM using a 7 T Bruker BioSpec scanner. The r1 and r2 relaxivity values of the PR constructs were determined by plotting the inverse longitudinal (1/T1) and transversal (1/T2) relaxation times, respectively, as a function of Gd concentration. As shown in Figure 3 and Table 2, the r1 and r2 of the polyrotaxane contrast agents are higher than that of DOTAREM, suggesting that these compounds may have useful MR contrast properties at this field strength.

Figure 3. Relaxivity determinations (A, r1; B, r2) of Gd:DOTA-HPβCD/SBEβCD Pluronic polyrotaxanes and DOTAREM at 7 T, 25 °C. Blue: F127; Red, L81; Purple: F68; Yellow: L35; Aqua: L64; Green: DOTAREM. D


Article

Bioconjugate Chemistry PR contrast agents (0.050 mM) produced signals whose contrast enhancement was similar or better than DOTAREM at the highest concentration (1 mM). In Vivo MRI Contrast Enhancement of Gd:DOTAHPβCD/SBEβCD PR in Balb/c Mice. T1-weighted 3D MR images of Balb/c mice after intravenous injection of the PR contrast agents were acquired using a 7 T Bruker BioSpec small animal scanner before injection, and 10, 20, 35, 50, and 60 min post-injection, to investigate the MR contrast enhancement and circulation fate of the PR in vivo. The tissue contrast enhancement of the compounds was found to have different outcomes depending on the threading efficiency, molecular weight, and surface charge of the contrast agents. Maximum intensity projection images (Figure 4) obtained for DOTAREM and all the PR contrast agents (except Gd:DOTAHPβCD/SBEβCD L64 PR due to material limitations) reveals excellent contrast enhancements after 5−10 min for most of the PR compounds in heart, liver, and kidney. In general, the contrast intensities progressively decreased over time, with renal clearance appearing to be the main route of elimination for the lower-molecular-weight PR and biliary clearance predominating for the longer circulating, higher-molecularweight PR. As expected, DOTAREM was completely eliminated within 20 min, as is commonly observed for small molecule contrast agents. Surprisingly, Gd:DOTA-HPβCD/SBEβCD F127 PR displayed lower contrast enhancement compared to the other PR agents. This is likely due to the low theading efficiency (31%) that exposes the large Pluronic F127 PEG blocks and nearly half of the PPG block to blood. Since Kojima et al. reported strong influences of large PEG blocks on relaxivity reduction compared to short PEG blocks (e.g., 5k vs 2k),53 we attribute the loss in relaxivity to hydrophobic collapse of the PPG blocks such that the Gd3+-DOTA-βCD units become encased within a corona of the PEG blocks, thus limiting their access to water. Conversely, the highly threaded PR contrast agents such as Gd:DOTA-HPβCD/SBEβCD L35 PR and Gd:DOTAHPβCD/SBEβCD L81 PR show the best contrast enhancements, likely due to their high threading efficiency that confers a rod-like morphology in circulation as previously reported wherein the longest circulating species were those with the highest threading efficiencies.45 Contrast signal-to-noise ratio (CNR) enhancements were calculated from regions of interest (ROI) that were carefully drawn in cross-sectional images of heart, liver, and kidney obtained from the 2D T1-weighted scans (Figures S13−S22). In heart, meaningful signal enhancements were observed for all PR contrast agents, with the most notable increases obtained with F68, L35, and L81 constructs (Figure 5). It appears that these compounds persist in circulation longer than DOTAREM and Gd:DOTA-HPβCD/SBEβCD F127 PR. We attribute these observations to the higher threading efficiencies and higher molecular weights of these PR constructs. Vascular contrast persists for as long as 60 min for all PR agents except the F127 PR, while the L81 PR shows the highest level of accumulation in liver relative to the other PR. The L35 and F68 PR also displayed some liver accumulation after 70 min (i.e., ranging between 10−25% of the initial contrast intensity). Protein Corona Analysis. To better understand the serum−material interactions that dictate the PR fate in vivo, we evaluated the effects of surface chemistry and charge on the PR protein corona using a common Pluronic L81 core and a variety of βCD as the threaded macrocycle. Four different

Figure 4. 3D Maximum intensity projections of Balb/c mice recorded at 7 T before and after tail vein injection of 0.05 mmol Gd/kg body weight of DOTAREM and Gd:DOTA-HPβCD/SBEβCD PR MR contrast agents. Images were recorded using a T1-weighted 3D gradient echo sequence.

cholesteryl end-capped PR containing βCD, HPβCD, HPβCD/SBEβCD, and MeβCD were prepared using the synthesis method described above (Table S3) and compared with the Gd:DOTA-HPβCD/SBEβCD PR library. The most striking difference observed was the total amount of protein present within the coronas of the Gd:DOTA HPβCD/ E


Article


Figure 5. Normalized intensities of Gd:DOTA-HPβCD/SBEβCD Pluronic PR contrast agents and DOTAREM control. Aqua diamonds, DOTAREM; Red squares, F127; Green triangles, L35; Purple crosses, L81; Blue crosses, F68.

Figure 6. Total protein adsorption onto Gd:DOTA-HPβCD/SBEβCD PR (left) and L81-based PR containing different βCD derivatives (right) as determined by total MS ion current. Inset: Ordinate expansion to emphasize the differences in protein deposition onto L81-based PR containing different βCD derivatives.

Figure 7. Average percentage of PR protein corona as a function of L81 PR type.

500-fold increase in protein deposition as a result of gadolinium chelate modification and the presence of unreacted PEGamines on the PR scaffold. Furthermore, the more highly threaded PR contrast agents, Gd:DOTA-HPβCD/SBEβCD

SBEβCD PR relative to the PR lacking the gadolinium chelate modification (Figure 6). Comparison of the total protein counts for the HPβCD/SBEβCD L81 PR containing and lacking the Gd3+:DOTA chelate reveals that there is nearly a F


Article


incubation is similar, despite differences in βCD type. This suggests that the similarities between the PR structures (e.g., polymer core, end-cap chemistry, rod-like shape, presence of Gd:DOTA chelate) play a larger role in protein recruitment than βCD character, even though the latter appears to contribute strongly to the magnitude of protein adsorption onto the PR scaffold.

L35 and Gd:DOTA-HPβCD/SBEβCD L81, have the lowest associated protein relative to the other members of the Gd:DOTA PR library. Notably, changes in the βCD derivatives borne by the PR scaffolds yielded significant differences in the amount and identity of protein species found in the protein corona. Total amounts of adsorbed protein were found to vary significantly with βCD composition (Figure 6, inset). PR bearing unmodified βCD accumulated the largest quantity of serum proteins, possibly due to the rigidity of the hydrogen bonded network between neighboring βCD units and the hydrophobicity of the βCD surface due to the axial C−H bonds of the βCD glucose units. HPβCD had the second highest average protein deposition. The observed decrease in protein adsorption onto HPβCD versus unmodified βCD PR may be due to a brush-like shielding of the PR surface that originates from the multiple 2-hydroxypropyl modifications along the HPβCD PR scaffold. MeβCD and HPβCD/ SBEβCD PR displayed similar levels of protein deposition despite having significantly different surface chemistries. Decreased adsorption to the MeβCD PR may originate from reduced hydrogen bonding potential from the capped hydroxyl groups of MeβCD, whereas incorporation of negative charge along the backbone of the SBEβCD PR may limit protein−PR interactions due to electrostatic repulsion and/or reduced hydrophobic interactions. Additional studies are needed to more fully characterize the structural factors contributing to protein deposition in these materials. We then evaluated the proteins present within the corona of each PR type and sorted them into their respective families (Figure 7). Variation in βCD type within the PR scaffold led to changes in the classes of proteins that were preferentially deposited onto the polyrotaxane. Lipoproteins were the predominant protein class observed in the corona of all the PR (∼40% of the total), except for the unmodified βCD species that accumulated ∼27% lipoprotein content. We infer from the similarity in lipoprotein adsorption with each type of PR that the lipoproteins are binding via association with the cholesteryl end-caps present in each of these PR constructs rather than through interactions with the different βCD units. Notably, βCD PR and HPβCD PR displayed greater immunoglobulin deposition than their HPβCD/SBEβCD PR and MeβCD PR counterparts. This may be due to antibody recognition of the 1,4-glucose motifs within the more rigid intermolecularly hydrogen-bonded βCD units that present glycals in a manner that resembles the surface of invading bacteria. Increased antibody binding would be expected to contribute to accelerated clearance during circulation. Since immunoglobulin binding to the HPβCD/SBEβCD PR was low, the long circulation behavior of the Gd:DOTA-HPβCD/ SBEβCD L81 PR (Figure 4E) is consistent with this interpretation. Although each PR contrast agent bound albumin similarly, likely via the cholesteryl end-cap moieties, the MeβCD and HPβCD/SBEβCD PR displayed elevated complement protein binding relative to the βCD and HPβCD PR. Since the latter constructs have lower degrees of βCD threading, the βCD units in these PR likely have greater lateral and rotational dynamics that contribute to reduced complement protein binding. A table showing the 20 most abundant proteins found in the corona of each of these PR appears in Table S3. Collectively, all PR share 10 of their top 20 most abundant proteins, while pairwise evaluation of any two reveal that those PR share between 13 and 15 identified proteins. We infer from these findings that the serum protein response to PR

■

CONCLUSION In summary, we have synthesized and characterized five Gd:DOTA-based HPβCD/SBEβCD Pluronic polyrotaxane MR contrast agents that were threaded with 40−50 mol % SBEβCD. DLS results suggested that the polyrotaxanes are polydisperse with sizes ranging between 110 and 230 nm. Zeta potential measurements indicated that the compounds bear a slight negative charge due to the presence of SBEβCD sulfonate groups that substantially improves their water solubility. 1H NMRD and molar relaxivity measurements at 25 °C revealed that all of the PR constructs had higher spin− spin relaxation (T1) and spin−lattice relaxation (T2) rates than DOTAREM as control. MR images of Balb/c mice receiving 0.05 mmol Gd/kg body weight via tail vein injection revealed that Gd:DOTA-HPβCD/SBEβCD F68 PR, Gd:DOTAHPβCD/SBEβCD L35 PR, and Gd:DOTA-HPβCD/SBEβCD L81 PR gave the best vascular contrast enhancement, for periods as long as 60 min. All PR agents, except Gd:DOTAHPβCD/SBEβCD L81 PR, appeared to clear predominantly via the renal route, whereas the highly threaded L81-based construct appeared to clear by both the renal and biliary pathways. Proteomics evaluation of serum treated PR contrast agents bearing different β-CD types on a common Pluronic L81 scaffold revealed that lipoproteins were the most abundant blood protein species associated with the PR, likely due to binding with the biantennary cholesterol end-caps. Immunoglobulins and complement proteins were the next most abundant corona proteins associated with the PR. Modification of the PR scaffolds with Gd3+:DOTA led to enhanced protein deposition into the corona of the contrast agents. β-CD-based PR were found to bind the highest levels of blood proteins, whereas the HPβCD/SBEβCD-based PR displayed the lowest protein corona levels. Structure−property considerations suggest that the best performing PR agents are those that possess a rod-like structure that presumably promotes flow alignment during blood circulation to enhance contrast persistence. We infer from these findings that rod-like L81based PR contrast agents with ≥40 mol % SBEβCD content are promising candidates for vascular imaging applications.

■

MATERIALS AND METHODS

Materials. The Pluronic triblock copolymers F127 (EO 200, PO 65, MW = 12600), F68 (EO 153, PO 29, MW = 8350), L35 (EO 22, PO 16, MW = 1900), L64 (EO 26, PO 30, MW = 2900), and L81 (EO 6, PO 43, MW = 2800) were purchased from Sigma-Aldrich and dried by azeotropic distillation from benzene under vacuum before use. 2Hydroxypropyl-β-cyclodextrin, with an average degree of hydroxypropyl substitution of 6.8, carbonyldiimidazole (CDI), triethylamine (TEA), tris(2-aminoethyl)amine (TAEA), and cholesteryl chloroformate were also purchased from Sigma-Aldrich and used received. 4-Sulfobutylether-βcyclodextrin (SBEβCD, a.k.a. Captisol) with an average degree of SBE substitution of 7 was generously supplied by Cydex G


Article


DMSO and deionized water, respectively, for 3 d, then lyophilized to generate white powders of the modified PR product. 1H NMR (500 MHz, Cryo probe, DMSO-d6): δ = 6.92 ppm (S, H-NCO carbamate), 5.25 ppm (t, 1H, Chlethylene H), 4.5−5.0 ppm (b, C1−H of CDs), 4.5 ppm (b, OH propyl of HPβCD), 3.5−3.8 ppm (m, C3,5,6-H of CDs), 3.5 ppm (m, PEG-CH2), 2.6−2.8 ppm (m, 16H, CH2 of TAEA), 1.8 ppm (b, NH2 of DADO), 1.6 ppm (b, (CH2)-SO3−), 1.2 ppm (d, CH3−HPβCD), 1.0 ppm (d, CH3 of PPG), 0.8−0.6 ppm (m, Chl-CH3). Synthesis of p-SCN-Bn-DOTA-Modified Polyrotaxanes (DOTA-HPβCD/SBE-βCD PR). The PEGylated PR (150 mg) were dissolved in 5 mL DMSO and stirred under Ar. To this mixture, p-SCN-Bn-DOTA (50 mg, 0.073 mmol) was added and allowed to stir for 24 h at 20 °C. The reaction was stopped and the products purified by dialysis to remove any unreacted reagents using 12000−14000 and 6000−8000 MWCO regenerated cellulose membranes in DMSO, followed by water, before lyophilization to generate powdered samples of p-SCN-Bn-DOTA modified polyrotaxane (DOTAHPβCD/SBE-βCD-PR). 1H NMR (500 MHz, Cryo probe, DMSO-d6): δ = 13 ppm (Ph-NH−CO), 9.5 ppm (CO-NH−) 7.5−7.0 ppm (d, 4H−Ar-DOTA), 6.92 ppm (S, H-NCO), 5.25 ppm (t, 1H, Chl-ethylene H), 4.5−5.0 ppm (b, C1−H of CDs), 4.5 ppm (b, propyl OH of HPβCD), 3.5−3.8 ppm (m, C3,5,6-H of CDs), 3.5 ppm (m, PEG-CH2), 2.6−2.8 ppm (m, 16H, CH2 of TAEA), 1.8 ppm (b, NH2 of DADO), 1.6 ppm (b, (CH2)-SO3−), 1.2 ppm (d, CH3−HPβCD), 1.0 ppm (d, CH3 of PPG), 0.8−0.6 ppm (m, Chl-CH3). Gadolinium(III) Loading of DOTA-Modified Polyrotaxanes (Gd:DOTA-HPβCD/SBEβCD PR). Gadolinium(III) complexation was performed as follows: GdCl3·6H2O (2 equiv per DOTA) was dissolved in H2O (5 mL) and mixed with the DOTA-HPβCD/SBE-βCD PR samples dissolved in H2O (10 mL) with adjustment of the mixture pH to 5.5−6.5 using NaOH or HCl solutions. The reaction mixtures were stirred at 50 °C for 48 h and the products were dialyzed against water (pH 7) for 3 d using a 6.0−8.0 kDa MWCO membrane. To remove nonchelated Gd3+ ions, the samples were eluted through a Sephadex G 25 column and the solution eluting after the void volume assayed with xylenol orange to detect traces of free Gd(III) ions. No color change at 435 and 578 nm was observed after xylenol orange addition to the fractions isolated by gel filtration. The fractions collected were lyophilized and stored at −80 °C until use. Nuclear Magnetic Resonance, NMR. 1H NMR spectra were collected using a Bruker AV-III-500-HD spectrometer equipped with a CryoProbe. Spectra were recorded at 25 °C in DMSO-d6 unless otherwise indicated using approximately 15 mg of each PR dissolved in 1 mL of solvent. GPC-MALS/RI. Absolute masses of the PR contrast agents were obtained using an Agilent Technologies 1200 series chromatograph equipped with a Shodex SB-803-HQ column with DMSO as eluant at a flow rate of 0.1 mL/min using RI and multiangle light scattering detections (Wyatt Optilab REx and DAWN HELEOS-II, respectively). Pullulan (MW 12000 kDa), and three dextrans (MW 11600; 48600; and 667,800 kDa) were used as calibration standards. The samples were dissolved in DMSO (2 mg/mL) and eluted for 150 min. AUC. AUC (Beckman-Coulter Optima XL1) was used to independently determine the molecular weight of the PR constructs. The sedimentation velocity method was employed at a speed of 50000 rpm for 20 h. The samples were dissolved

Pharmaceuticals (Lawrence, KS) and was used without further purification. S-2-(4-Isothiocyanatobenzyl)-1,4,7,10-tetraazacyclododecane tetraacetic acid (p-SCN-Bn-DOTA) was obtained from Macrocyclics (Dallas, TX). All solvents were reagent grade, purchased from commercial sources, and were dried over CaH2, filtered, and distilled at reduced pressure before storing under Ar prior to use. Cellulose dialysis membranes were obtained from Spectrum Laboratories and immersed in deionized water for at least 30 min prior to use. A Barnstead MicroPure water purification system was used to produce ultrapure water (resistivity ≈ 18.2 MΩ/cm−1). Synthesis of Bis-Cholesteryl End-Capped HPβCD/ SBEβCD Pluronic Polyrotaxanes. Before the synthesis of the polyrotaxanes, all the Pluronic copolymers were modified to obtain α,ω-bis-tris(2-aminoethyl)amine Pluronic as described previously.39 Polyrotaxane syntheses were then conducted using a slight modification of the previously described heterogeneous solvent-assisted threading procedure.39,43,44 Typically, 2 g of each dried α,ω-bis-tris(2aminoethyl)amine Pluronic triblock copolymer was suspended in 70 mL hexane and dispersed by bath sonication for about 5 min before stirring until the solution appeared homogeneous. HPβCD and SBEβCD were physically mixed at a molar ratio of 30% SBEβCD using a cyclodextrin loading of one β-CD per two PO units within the Pluronic to be used and the β-CD mixture finely ground for 30 min using an agate mortar and pestle. The powdered cyclodextrin mixtures were added to the polymer suspension and then vigorously stirred for 2 h. The mixtures were then bath sonicated for 1 h at 20 °C, followed by 10 min probe sonication (Heat Systems Model W-350, 50 W, 1/2 in. probe) to improve the dispersion of the Pluronic copolymers. The mixtures were allowed to stir for 72 h at 20 °C before the solvent was removed under reduced pressure and the resulting materials were redissolved in 40 to 60 mL of dried CH2Cl2 before cholesteryl chloroformate (12 equiv) was added. The reaction mixtures were stirred at 20 °C for 24 h, concentrated, and then precipitated in Et2O (700 mL). To remove unreacted reagents and unthreaded cyclodextrins, the crude products were dissolved in CH3OH (300 mL), precipitated in 500 mL Et2O, and gathered by filtration. Finally, the products were purified by sequential dialysis using 12000−14000 and 6000−8000 MWCO regenerated cellulose membranes in DMSO first and progressively exchanged with deionized water over 5 d before lyophilization to generate white HPβCD/SBEβCD PR powders. 1H NMR (500 MHz, Cryo probe, DMSO-d6): δ = 6.92 ppm (S, H-NCO carbamate), 5.25 ppm (t, 1H, Chl-ethylene H), 4.5−5.0 ppm (b, C1−H of CDs), 4.5 ppm (b, OH propyl of HPβCD), 3.5− 3.8 ppm (m, C3,5,6-H of CDs), 3.5 ppm (m, PEG-CH2), 2.6− 2.8 ppm (m, 16H, CH2 of TAEA), 1.6 ppm (b, (CH2)-SO3−), 1.2 ppm (d, CH3−HPβCD), 1.0 ppm (d, CH3 of PPG), 0.8− 0.6 ppm (m, Chl-CH3). Synthesis of 1,10-Diamino-4,7-dioxadecane Modified Polyrotaxanes. Dried polyrotaxane (200 mg) from the previous step was dissolved in 20 mL DMSO and stirred under an Ar atmosphere. To this mixture, TEA (1.5 equiv. per OH group) was added and the reactions were stirred for 30 min before addition of excess 1,1′-carbonyldiimidazole (700 equiv) and stirring for 24 h. Next, an excess of 1,8-diamino-3,6dioxaoctane (700 equiv) was slowly added to the solutions and the mixtures were allowed to stir for 24 h at 20 °C. The products were purified by dialysis using 12000−14000 and 6000−8000 MWCO regenerated cellulose membranes in H


Article


100% oxygen using a SomnoSuite device. The induction conditions were fixed at 250 mL/min O2 with 2.5% isofluorane, followed by immobilization in the scanner at 250 mL/min O2 with 1.8% isofluorane. The temperature of the mice was maintained by a water heating bath bed system, whereas the respiration rate was monitored using a SA system equipped with an air pad inserted under the animals. Methods. A high resolution T1-weighted 3D gradient echo sequence was run to scan the mice in a 7 T Bruker BioSpec 47/30 scanner equipped with a Bruker Biospin MRI GmbH volume coil, 86 mm ID, operating at 300 MHz, and interfaced to ParaVision software to collect coronal images. The mice were then injected via tail vein with 200 μL of saline solution containing PR construct (0.05 mmol Gd/kg body weight) before acquisition of post-injection images at five different time points using a high resolution T1-weighted 3D sequence: TR/ TE = 15/3.0 ms, FOV = 80 × 30 × 30 mm3, matrix = 384 × 192 × 32, slice thickness = 30 mm with no slice gap, FA = 20°, acquisition number = 3, resolution = 0.20 × 0.15 × 0.90 mm3. Similarly, 2D T1-weighted transversal images were acquired at different time points to reveal different vascularized regions under the following parametrs: TR/TE = 500/1.5 ms, FOV = 50 × 50 mm2, matrix = 256 × 256, slice thickness = 2 mm, FA = 30°, acquisition number = 2, resolution = 0.20 × 0.20 mm2, slice gap = 1.0 mm. Image Analysis. To quantify the changes in tissues signal intensity, ROI were drawn in images of different organs obtained from the 2D T1-weighted scans and ParaVision software was used to evaluate the mean intensities. The contrast-to-noise ratio (CNR) was calculated for the tissues using the following equation

in DMSO at a concentration of 1 mg/mL and introduced into 7.15 cm cells in the instrument. The reference cell was filled with pure DMSO. The data were analyzed by SEDFIT using a c(s) model. Inductively Coupled Plasma Mass Spectrometry. A quadrapole ICP-MS, Agilent Technologies 7500am, was used to analyze the gadolinium content in the polyrotaxanes. Samples were digested with 2% nitric acid (TraceMetal grade, Fisher Scientific) and diluted from a gadolinium/2% HNO3 Certiprep ME 1 standard solution of 1 mg/mL (SPX CertiPrep, Metuchen, NJ) using a standard additive method and introduced into a temperature controlled spray chamber with a MicroMist Nebulizer (Pocasset, MA). Measurement of the Hydrodynamic Diameter and ζ Potential of the Gd3+-HPβCD/SBEβCD PR. The diameters, size distributions, and ζ potentials of the PR constructs were evaluated by dynamic light scattering using a particle size analyzer (Zetasizer Nano S, Malvern Instruments Ltd.) at 20 °C with a scattering angle of 90°. At least 3 measurements were made and averaged for each sample. 1 H NMRD. The longitudinal relaxation rates of all the samples were acquired using 0.5 mM aqueous solutions of PR contrast agent dissolved in H2O using a Stelar SPINMASTER 1T fast field cycling (FFC) relaxometer (Stelar Mede, Pavia, Italy). The conventional FFC method was used, evaluating 30 values of magnetic fields from 0.24 mT to 0.97 T, corresponding to a proton Larmor frequency range of 0.010−40 MHz. Standard Pre-Polarized (PP/S) and NonPolarized (NP/S) acquisition sequences were used at 30 °C. The samples were polarized in a high magnetic field Bpol (25 MHz) until the nuclear magnetization of the 1H nucleus reached saturation. Then, the magnetic field was switched to the detection field Bacq (16 MHz) and the magnetization was measured by a 90° pulse, followed by acquisition of the time dependent decay curves. Relaxivity Measurements. The longitudinal and transversal relaxation rates of the contrast agents were measured using a 7 T Bruker BioSpec scanner 47/30 equipped with a Bruker Biospin MRI GmbH volume coil, 86 mm ID, operating at 300 MHz at 23 °C (RES 300 1 H 112/086 QNS TO AD). The PR contrast agents were dissolved in deionized water and 300 μL solutions at 0.05, 0.10, 0.30, 0.50, 0.80, and 1.0 mM Gd content were prepared. T1 and T2 values were measured using inversion recovery sequence and multislice-multiecho (MSME) sequence, respectively. For each image obtained, nonlinear magnetization regression equations, Mz = Mo (1 − e−t/T1) for T1 and Mxy = (e−t/T2) for T2 were fit by a least-squares method. The parameters used were the following. r1: Repetition times, TE = 50.72, 100, 350, 750, 1250, 2500, 3500, and 5000 ms, echo time, TE = 22.22 ms, FOV = 10 × 10 mm2, matrix = 128 × 128, slice thickness = 1.0 mm, acquisition number = 1. r2: TR = 2000 ms, TE = 15, 30, 45, 60, 75, 90, and 105 ms, FOV = 20 × 20 mm2, matrix = 128 × 128, slice thickness = 1.0 mm, acquisition number = 1. r1 and r2 values were determined as the slopes of the lines in the 1/T1 and 1/ T2 vs Gd concentration plots, respectively, using the same ROI. In Vivo MR Imaging. In vivo evaluation of the PR contrast agents were performed using 7−9-weeks-old female Balb/c mice (20 g each) following a protocol (1112000342) approved by the Purdue Animal Care and Use Committee (PACUC). The mice (n = 3) were anesthetized with isoflurane mixed with

CNR =

MSI(tissue) − MSI(muscle) StD(background)

where MSI stands for mean signal intensity and StD denotes standard deviation. Proteomics Analysis Methods.45 Normal human serum was purchased from Complement Technology (Tyler, TX) and thawed immediately before use. PR (100 μg) were taken up in PBS and bath sonicated for 10 min at 20 °C before incubation with undiluted human serum (1:1 v:v) for 1 h at 37 °C. After incubation, samples were centrifuged for 10 min at 4000 × g and the PR pellets washed four times with 150 μL cold PBS and recentrifuged/resuspended as before. Three separate incubations were performed for each PR type and the data averaged for each compound. Tryptic peptides were separated on a nanoLC system (Agilent Technologies, Series 1100 LC, Santa Clara, CA). The peptides were loaded onto an Agilent 300SB-C18 enrichment column (5 × 0.3 mm, 5 μm particles) for concentration before switching the enrichment column into the nanoflow path after 5 min. Peptides were separated with an Agilent C18 reversed phase ZORBAX 300SB-C18 analytical column (0.75 × 150 mm, 3.5 μm particles). The column was connected to an emission tip from New Objective and coupled to the nanoelectrospray ionization (nESI) source of a high-resolution hybrid ion trap mass spectrometer (LTQ-Orbitrap XL, Thermo Scientific). The peptides were eluted from the column using an ACN/0.1% formic acid (Mobile Phase B) linear gradient. For the first 5 min, the column was equilibrated with 95% H2O/0.1% formic acid (Mobile Phase A), followed by a linear gradient of 5% → 40% B over 65 min at 0.3 μL/min, I


Article

Bioconjugate Chemistry then from 40% → 100% B over an additional 10 min. The column was washed with 100% ACN/0.1% formic acid and equilibrated with 95% H2O/0.1% formic acid before the next sample was injected (total method time = 95 min). A blank injection was run between samples to avoid carryover. The LTQ-Orbitrap mass spectrometer was operated in the datadependent positive acquisition mode wherein each full MS scan (30,000× resolving power) was followed by eight MS/MS scans where the eight most abundant molecular ions were selected and fragmented by collision-induced dissociation using a normalized collision energy of 35%. Database searching was conducted using MaxQuant for LFQ (label free quantitation). The Human (SwissProt) annotated database was used for digest searches. Initial Spectral Counting was performed using the Mascot Database search results.

■

(6) Caravan, P., Ellison, J. J., McMurry, T. J., and Lauffer, R. B. (1999) Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem. Rev. 99, 2293−2352. (7) Luz, Z., and Meiboom, S. (1964) Proton relaxation in dilute solutions of cobalt(II) and nickel(II) ions in methanol and the rate of methanol exchange of the solvation sphere. J. Chem. Phys. 40, 2686. (8) Swift, T. J., and Connick, R. E. (1962) NMR-relaxation mechanisms of O17 in aqueous solutions of paramagnetic cations and the lifetime of water molecules in the first coordination sphere. J. Chem. Phys. 37, 307−20. (9) Davies, G.-L., Kramberger, I., and Davis, J. J. (2013) Environmentally responsive MRI contrast agents. Chem. Commun. 49, 9704−9721. (10) Aime, S., Botta, M., Fasano, M., and Terreno, E. (1999) Prototropic and water-exchange processes in aqueous solutions of Gd(III) chelates. Acc. Chem. Res. 32, 941−949. (11) Wiener, E. C., Brechbiel, M. W., Brothers, H., Magin, R. L., Gansow, O. A., Tomalia, D. A., and Lauterbur, P. C. (1994) Dendrimer-based metal chelates: a new class of magnetic resonance imaging contrast agents. Magn. Reson. Med. 31, 1−8. (12) Langereis, S., de Lussanet, Q. G., van Genderen, M. H. P., Backes, W. H., and Meijer, E. W. (2004) Multivalent contrast agents based on gadolinium−diethylenetriaminepentaacetic acid-terminated poly(propylene imine) dendrimers for magnetic resonance imaging. Macromolecules 37, 3084−3091. (13) Villaraza, A. J. L., Bumb, A., and Brechbiel, M. W. (2010) Macromolecules, dendrimers, and nanomaterials in magnetic resonance imaging: the interplay between size, function, and pharmacokinetics. Chem. Rev. 110, 2921−2959. (14) Kabalka, G. W., Davis, M. A., Moss, T. H., Buonocore, E., Hubner, K., Holmberg, E., Maruyama, K., and Huang, L. (1991) Gadolinium-labeled liposomes containing various amphiphilic GdDTPA derivatives: targeted MRI contrast enhancement agents for the liver. Magn. Reson. Med. 19, 406−15. (15) Tilcock, C., Ahkong, Q. F., Koenig, S. H., Brown, R. D., Davis, M., and Kabalka, G. (1992) The design of liposomal paramagnetic MR agents: effect of vesicle size upon the relaxivity of surfaceincorporated lipophilic chelates. Magn. Reson. Med. 27, 44−51. (16) Gløgård, C., Stensrud, G., and Aime, S. (2003) Novel radicalresponsive MRI contrast agent based on paramagnetic liposomes. Magn. Reson. Chem. 41, 585−588. (17) Ward, K. M., Aletras, A. H., and Balaban, R. S. (2000) A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J. Magn. Reson. 143, 79−87. (18) Zhao, J. M., Har-El, Y.-E., McMahon, M. T., Zhou, J., Sherry, A. D., Sgouros, G., Bulte, J. W. M., and van Zijl, P. C. M. (2008) Sizeinduced enhancement of chemical exchange saturation transfer (CEST) contrast in liposomes. J. Am. Chem. Soc. 130, 5178−5184. (19) Li, Y., Han, Z., Roelle, S., DeSanto, A., Sabatelle, R., Schur, R., and Lu, Z.-R. (2017) Synthesis and assessment of peptide Gd− DOTA conjugates targeting extradomain B fibronectin for magnetic resonance molecular imaging of prostate cancer. Mol. Pharmaceutics 14, 3906−3915. (20) Spanoghe, M., Lanens, D., Dommisse, R., Van der Linden, A., and Alderweireldt, F. (1992) Proton relaxation enhancement by means of serum albumin and poly-L-lysine labeled with DTPA-Gd3+: relaxivities as a function of molecular weight and conjugation efficiency. Magn. Reson. Imaging 10, 913−7. (21) André, J. P., Tóth, É ., Fischer, H., Seelig, A., Mäcke, H. R., and Merbach, A. E. (1999) High relaxivity for monomeric Gd(DOTA)based MRI contrast agents, thanks to micellar self-organization. Chem. - Eur. J. 5, 2977−2983. (22) Schmiedl, U., Ogan, M. D., Moseley, M. E., and Brasch, R. C. (1986) Comparison of the contrast-enhancing properties of albumin(Gd-DTPA) and Gd-DTPA at 2.0 T: an experimental study in rats. AJR, Am. J. Roentgenol. 147, 1263−70. (23) Wikstrom, M. G., Moseley, M. E., White, D. L., Dupon, J. W., Winkelhake, J. L., Kopplin, J., and Brasch, R. C. (1989) Contrast-

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.bioconjchem.8b00525. NMR, GPC, AUC, MR images, and proteomics table (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge the assistance of Dr. Gregory Tamer (Purdue University MRI Facility) for his assistance in the collection of MR images. We would like to express our special thanks to Dr. Ben D. Elzey, Ms. Sandra TerregrosaAllen, and Mr. Benjamin D. Ramsey from the Center for Cancer Research − Biological Evaluation Shared Resource at Purdue University for their extensive assistance in the animal preparation and handling. We are grateful to Lake Paul from Bindley Biophysical Analysis Laboratory at Purdue University for his help with the AUC analyses. The authors want to thank Gianni Ferrante and Rebecca Steele from Stelar s.r.l., Italy for collecting the relaxivity of the PR contrast agents using the Fast Field Cycling Relaxometry (FFCR) proton NMR dispersion technique. We are especially grateful for the financial support of NIH grants EB017921 and CCSG CA23168 and special project and shared resource funds provided by the Purdue University Center for Cancer Research.

■

REFERENCES

(1) Weissleder, R., and Pittet, M. J. (2008) Imaging in the era of molecular oncology. Nature 452, 580−9. (2) Cassidy, P. J., and Radda, G. K. (2005) Molecular imaging perspectives. J. R. Soc., Interface 2, 133−44. (3) Frullano, L., and Meade, T. J. (2007) Multimodal MRI contrast agents. JBIC, J. Biol. Inorg. Chem. 12, 939−49. (4) Gale, E. M., Wey, H.-Y., Ramsay, I., Yen, Y.-F., Sosnovik, D. E., and Caravan, P. (2018) A Manganese-based alternative to gadolinium: contrast-enhanced MR angiography, excretion, pharmacokinetics, and metabolism. Radiology 286, 865−872. (5) Lauffer, R. B. (1987) Paramagnetic metal complexes as water proton relaxation agents for NMR imaging: theory and design. Chem. Rev. 87, 901−927. J


Article

Bioconjugate Chemistry enhanced MRI of tumors. Comparison of Gd-DTPA and a macromolecular agent. Invest. Radiol. 24, 609−15. (24) Curtet, C., Bourgoin, C., Bohy, J., Saccavini, J. C., Thedrez, P., Akoka, S., Tellier, C., and Chatal, J. F. (1988) Gd-25 DTPA-MAb, a potential NMR contrast agent for MRI in the xenografted nude mouse: preliminary studies. Int. J. Cancer 41 (2), 126−32. (25) Curtet, C., Tellier, C., Bohy, J., Conti, M. L., Saccavini, J. C., Thedrez, P., Douillard, J. Y., Chatal, J. F., and Koprowski, H. (1986) Selective modification of NMR relaxation time in human colorectal carcinoma by using gadolinium-diethylenetriaminepentaacetic acid conjugated with monoclonal antibody 19−9. Proc. Natl. Acad. Sci. U. S. A. 83, 4277−81. (26) Rongved, P., and Klaveness, J. (1991) Water-soluble polysaccharides as carriers of paramagnetic contrast agents for magnetic resonance imaging: synthesis and relaxation properties. Carbohydr. Res. 214, 315−23. (27) Chu, W.-J., and Elgavish, G. A. (1995) Gadolinium and dysprosium chelates of DTPA-amide-dextran: synthesis, 1H NMR relaxivity, and induced 23Na NMR shift. NMR Biomed. 8, 159−163. (28) Lai, W.-F., Rogach, A. L., and Wong, W.-T. (2017) Chemistry and engineering of cyclodextrins for molecular imaging. Chem. Soc. Rev. 46, 6379−6419. (29) Rofsky, N. M., Sherry, A. D., and Lenkinski, R. E. (2008) Nephrogenic systemic fibrosis: a chemical perspective. Radiology 247, 608−12. (30) Fleury, G., Schlatter, G., Brochon, C., Travelet, C., Lapp, A., Lindner, P., and Hadziioannou, G. (2007) Topological polymer networks with sliding cross-link points: the “sliding gels”. Relationship between their molecular structure and the viscoelastic as well as the swelling properties. Macromolecules 40, 535−543. (31) Okumura, Y., and Ito, K. (2001) The polyrotaxane gel: a topological gel by figure-of-eight cross-links. Adv. Mater. 13, 485−487. (32) Neal, E. A., and Goldup, S. M. (2014) Chemical consequences of mechanical bonding in catenanes and rotaxanes: isomerism, modification, catalysis and molecular machines for synthesis. Chem. Commun. 50, 5128−5142. (33) Ooya, T., Inoue, D., Choi, H. S., Kobayashi, Y., Loethen, S., Thompson, D. H., Ko, Y. H., Kim, K., and Yui, N. (2006) pHResponsive movement of cucurbit[7]uril in a diblock polypseudorotaxane containing dimethyl β-cyclodextrin and cucurbit[7]uril. Org. Lett. 8, 3159−3162. (34) Kulkarni, A., DeFrees, K., Schuldt, R. A., Hyun, S.-H., Wright, K. J., Yerneni, C. K., VerHeul, R., and Thompson, D. H. (2013) Cationic α-cyclodextrin:poly(ethylene glycol) polyrotaxanes for siRNA delivery. Mol. Pharmaceutics 10, 1299−1305. (35) Loethen, S., Ooya, T., Choi, H. S., Yui, N., and Thompson, D. H. (2006) Synthesis, characterization, and pH-triggered dethreading of α-cyclodextrin-poly(ethylene glycol) polyrotaxanes bearing cleavable endcaps. Biomacromolecules 7, 2501−6. (36) Harada, A., Li, J., and Kamachi, M. (1992) The molecular necklace: a rotaxane containing many threaded α-cyclodextrins. Nature 356, 325. (37) Harada, A., Li, J., and Kamachi, M. (1994) Double-stranded inclusion complexes of cyclodextrin threaded on poly(ethylene glycol). Nature 370, 126. (38) Fujita, H., Ooya, T., and Yui, N. (1999) Thermally induced localization of cyclodextrins in a polyrotaxane consisting of βcyclodextrins and poly(ethylene glycol)−poly(propylene glycol) triblock copolymer. Macromolecules 32, 2534−2541. (39) Mondjinou, Y. A., McCauliff, L. A., Kulkarni, A., Paul, L., Hyun, S.-H., Zhang, Z., Wu, Z., Wirth, M., Storch, J., and Thompson, D. H. (2013) Synthesis of 2-hydroxypropyl-β-cyclodextrin/Pluronic-based polyrotaxanes via heterogeneous reaction as potential Niemann-Pick type C therapeutics. Biomacromolecules 14, 4189−4197. (40) Casula, M. F., Floris, P., Innocenti, C., Lascialfari, A., Marinone, M., Corti, M., Sperling, R. A., Parak, W. J., and Sangregorio, C. (2010) Magnetic resonance imaging contrast agents based on iron oxide superparamagnetic ferrofluids. Chem. Mater. 22, 1739−1748.

(41) Fredy, J. W., Scelle, J., Guenet, A., Morel, E., Adam de Beaumais, S., Menand, M., Marvaud, V., Bonnet, C. S., Toth, E., Sollogoub, M., Vives, G., and Hasenknopf, B. (2014) Cyclodextrin polyrotaxanes as a highly modular platform for the development of imaging agents. Chem. - Eur. J. 20, 10915−10920. (42) Fredy, J. W., Scelle, J., Ramniceanu, G., Doan, B.-T., Bonnet, C. S., Tóth, É ., Ménand, M., Sollogoub, M., Vives, G., and Hasenknopf, B. (2017) Mechanostereoselective one-pot synthesis of functionalized head-to-head cyclodextrin [3]rotaxanes and their application as magnetic resonance imaging contrast agents. Org. Lett. 19, 1136− 1139. (43) Zhou, Z., Mondjinou, Y., Hyun, S.-H., Kulkarni, A., Lu, Z.-R., and Thompson, D. H. (2015) Gd3+-1,4,7,10-tetraazacyclododecane1,4,7-triacetic-2-hydroxypropyl-β-cyclodextrin/Pluronic polyrotaxane as a long circulating high relaxivity MRI contrast agent. ACS Appl. Mater. Interfaces 7, 22272−22276. (44) Mondjinou, Y. A., Hyun, S.-H., Xiong, M., Collins, C. J., Thong, P. L., and Thompson, D. H. (2015) Impact of mixed β-cyclodextrin ratios on Pluronic rotaxanation efficiency and product solubility. ACS Appl. Mater. Interfaces 7, 23831−23836. (45) Collins, C. J., Mondjinou, Y., Loren, B., Torregrosa-Allen, S., Simmons, C. J., Elzey, B. D., Ayat, N., Lu, Z.-R., and Thompson, D. (2016) Influence of molecular structure on the in vivo performance of flexible rod polyrotaxanes. Biomacromolecules 17, 2777−2786. (46) Collins, C. J., Loren, B. P., Alam, M. S., Mondjinou, Y., Skulsky, J. L., Chaplain, C. R., Haldar, K., and Thompson, D. H. (2017) Pluronic based β-cyclodextrin polyrotaxanes for treatment of Niemann-Pick type C disease. Sci. Rep. 7, 46737. (47) Merbach, A., Helm, L., and Toth, E., Eds. (2013) The Chemistry of Contrast Agents in Medical Magnetic Resonance Imaging, p 496, John Wiley & Sons Ltd. (48) Kobayashi, H., Kawamoto, S., Saga, T., Sato, N., Hiraga, A., Ishimori, T., Konishi, J., Togashi, K., and Brechbiel, M. W. (2001) Positive effects of polyethylene glycol conjugation to generation-4 polyamidoamine dendrimers as macromolecular MR contrast agents. Magn. Reson. Med. 46, 781−788. (49) Ogawa, M., Regino, C. A. S., Marcelino, B., Williams, M., Kosaka, N., Bryant, L. H., Choyke, P. L., and Kobayashi, H. (2010) New nano-sized biocompatible MR contrast agents based on lysinedendri-graft macromolecules. Bioconjugate Chem. 21, 955−960. (50) Nwe, K., Bryant, L. H., and Brechbiel, M. W. (2010) Poly(amidoamine) dendrimer based MRI contrast agents exhibiting enhanced relaxivities derived via metal pre-ligation techniques. Bioconjugate Chem. 21, 1014−1017. (51) Kotkova, Z., Helm, L., Kotek, J., Hermann, P., and Lukes, I. (2012) Gadolinium complexes of monophosphinic acid DOTA derivatives conjugated to cyclodextrin scaffolds: efficient MRI contrast agents for higher magnetic fields. Dalton Trans. 41, 13509−13519. (52) Lipari, G., and Szabo, A. (1982) Model-free approach to the interpretation of nuclear magnetic resonance relaxation in macromolecules. 1. Theory and range of validity. J. Am. Chem. Soc. 104, 4546−4559. (53) Kojima, C., Turkbey, B., Ogawa, M., Bernardo, M., Regino, C. A., Bryant, L. H., Choyke, P. L., Kono, K., and Kobayashi, H. (2011) Dendrimer-based MRI contrast agents: the effects of PEGylation on relaxivity and pharmacokinetics. Nanomedicine 7, 1001−8.

K


4-Sulfobutyl

Recommend Documents