pubs.acs.org/Biomac Article Metal-Free Radical Dendrimers as MRI Contrast Agents for Glioblastoma Diagnosis: Ex Vivo and In Vivo Approaches Songbai Zhang, Vega Lloveras,* Silvia Lope-Piedrafita, Pilar Calero-Pérez, Shuang Wu, Ana Paula Candiota,* and José Vidal-Gancedo* Cite This: Biomacromolecules 2022, 23, 2767−2777 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: Simultaneously being a nonradiative and noninvasive technique makes magnetic resonance imaging (MRI) one of the highly required imaging approaches for the early diagnosis and follow- up of tumors, specifically for brain cancer. Paramagnetic gadolinium (Gd)-based contrast agents (CAs) are the most widely used ones in brain MRI acquisitions with special interest when assessing blood− brain barrier (BBB) integrity, a characteristic of high-grade tumors. However, alternatives to Gd-based contrast agents (CAs) are highly required to overcome their established toxicity. Organic radicals anchored on a dendrimer macromolecule surface (radical den- drimers) are promising alternatives since they also exhibit para- magnetic properties and can act as T1 CAs like Gd-based CAs while being organic species (mitigating concerns about toxic metal accumulation). Here, we studied the third generation of a water-soluble family of poly(phosphorhydrazone) radical dendrimers, with 48 PROXYL radical units anchored on their branches, exploring their potential of ex vivo and in vivo contrast enhancement in brain tumors (in particular, of immunocompetent, orthotopic GL261 murine glioblastoma (GB)). Remarkably, this radical species provides suitable contrast enhancement on murine GL261 GB tumors, which was comparable to that of commercial Gd-based CAs (at standard dose 0.1 mmol/kg), even at its 4 times lower administered dose (0.025 mmol/kg). Importantly, no signs of toxicity were detected in vivo. In addition, it showed a selective accumulation in brain tumor tissues, exhibiting longer retention within the tumor, which allows performing imaging acquisition over longer time frames (≥2.5 h) as opposed to Gd chelates. Finally, we observed high stability of the radicals in biological media, on the order of hours instead of minutes, characteristic of the isolated radicals. All of these features allow us to suggest that the G3-Tyr-PROXYL-ONa radical dendrimer could be a viable alternative to metal-based MRI contrast agents, particularly on MRI analysis of GB, representing, to the best of our knowledge, the first case of organic radical species used for this purpose and one of the very few examples of these types of radical species working as MRI CAs in vivo. ■ INTRODUCTION Magnetic resonance imaging (MRI) is one of the most barrier (BBB) integrity, which is a relevant feature in aggressive brain tumors such as glioblastomas (GBs), a high-grade glial versatile and widely used clinical diagnostic tools nowadays. tumor with overall survival below 18 months. MRI provides images of soft tissue anatomy in excellent detail, Gd-based CAs are categorized as mainly T1 CAs, also with high spatial resolution, unlimited penetration depth, long referred to as “positive” contrast agents, leading to an increased effective imaging window, rapid in vivo imaging acquisition, (brighter) signal in T1-weighted images. They present high and the absence of ionizing radiation. Accordingly, MRI is relaxivities due to the high spin of the paramagnetic Gd(III) largely used for noninvasive diagnosis of tumors, as well as ion, which possesses seven unpaired electrons (spin 7/2).1 following-up response to therapy or relapse in several organs, These CAs have historically been considered safe, but for more but it is especially useful in brain tumors. Brain cancer is one of than a decade, they have been associated with potentially lethal the most lethal and difficult-to-treat cancers. Surgical resection is the main approach for tumor mass reduction in brain cancer Received: January 19, 2022 treatment; thus, it is of paramount importance in precise tumor Revised: June 2, 2022 localization and delineation. Published: June 24, 2022 Gadolinium (Gd)-based contrast agents (CAs) are the most widely used ones in brain MRI acquisitions for improving the intrinsic contrast enhancement and assessing blood−brain © 2022 The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.biomac.2c00088 2767 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article nephrogenic systemic fibrosis.2−4 Moreover, recent reports following systemic administration.17 Linear and cross-linked have emerged regarding the accumulation of residual toxic poly(carboxylate ester) PEG-modified PROXYL systems were Gd(III) ions in the brain and other organs from patients with used to provide MR imaging contrast enhancement to breast an intact blood−brain barrier after the administration of such cancer tumors.18 In addition, amphiphilic poly(ethylene Gd-based CAs.5,6 Since the use of CAs as cancer diagnosis glycol)-b-polycarbonate-based diblock copolymers containing agents in MRI is essential for cancer treatment, in particular for pendant persistent PROXYL radicals were locally administered glioblastoma, it is critical to find alternative imaging probes in the hindlimb muscle of a female C57BL/6J mouse.18 Finally, that provide the same or even better paramagnetic properties when adult female BALB/c nude mice bearing subcutaneous of current Gd-based CAs. tumors of HeLa cells in their back were intravenously injected Nitroxides (or nitroxyl radicals) such as PROXYL or with polyacetylenes containing TEMPO and PEG, the MRI (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) radicals are signal intensity significantly increased in the tumor paren- stable organic paramagnetic species possessing unpaired chyma.20 Nevertheless, none of these studies describe MRI electrons having the ability to provide imaging contrast by performance in orthotopic glioblastoma. shortening the T1 relaxation of water in a manner analogous to While dendrimer-based magnetic resonance imaging agents paramagnetic Gd3+. In fact, they have been investigated as T1 decorated with Gd have been reported for brain cancer,25 CAs for MRI7−10 and have been shown to be nontoxic in nothing has been described with metal-free organic water- vivo.11,12 However, they have two major limitations. On the soluble radical dendrimers. one hand, nitroxides present inherent low water 1H relaxivity In some recent works, we have proposed the use of radical since they possess only one unpaired electron (spin 1/2), and, dendrimers26,27 as a suitable alternative to Gd-based CAs.24,28 on the other hand, they are rapidly reduced in vivo (half-lives In general, one of the main drawbacks of organic macro- on the order of minutes) to diamagnetic hydroxylamines, molecules such as dendrimers is their low water solubility, hence losing their contrast ability and making them ineffective especially for large dendrimer generations. One of our as contrast agents shortly after injection.13−16 proposals was an innovative strategy to increase water One strategy to achieve higher molecular relaxivity and solubility and, at the same time, to obtain a full radical protection against reduction is through the anchoring of many functionalization of branches. By using amino acids as linkers nitroxide units to a conventional linear polymer or a between the dendrimer branches and the radicals we obtained hyperbranched polymer,17−21 where the relatively low four generations of poly(phosphorhydrazone) (PPH)-based relaxivity per nitroxide is multiplied by the number of bounded Gn-Tyr-PROXYL (n = 0, 1, 2, 3) radical dendrimers fully nitroxides and a protective shield effect can be provided to the soluble in water and completely functionalized with 6, 12, 24, radicals. Dendrimers could be excellent scaffold candidates for and 48 PROXYL radicals, respectively.28 We demonstrated the anchoring of radicals since they are a specific type of that such radical dendrimers offered the pendant radicals polymers characterized by strict control over their structure, higher stability (in vitro) against reduction with ascorbate ions making them nearly perfect monodisperse macromolecules. and showed negligible in vitro cytotoxicity, demonstrating that Besides, they present multifunctionality, globular structure and they could be excellent candidates to be used as MRI contrast tunable size (through different generations). It is important to agents suitable for biomedical applications. However, the highlight that the control over the size of the dendrimers opens assessment of their in vivo properties such as stability and the opportunity to modulate their distribution profile in the toxicity along with biodistribution and MRI studies was still body, which is not feasible in the case of Gd chelates. Only few lacking on tumor-bearing mice, and this is essential to consider reports describe the functionalization of dendrimers with them as plausible alternatives to Gd-based CA. organic radicals (radical dendrimers), with most of them being In the present work, we have explored the ex vivo and in vivo devoted to studying their electronic, magnetic, or structural MRI potential characteristics of the highest generation of that properties and only a few of them devoted to MRI CA family of radical dendrimers, G3-Tyr-PROXYL, as MRI applications.22−24 Still, very little has been reported about their contrast agents in immunocompetent, orthotopic GL261 real behavior in vivo, remaining a crucial challenging goal. To murine glioblastoma. The in vivo toxicity and stability were the best of our knowledge, there are only two examples of in also assessed. The G3 generation was the compound of choice vivo studies using radical dendrimers and very few reports for these studies since it showed the highest relaxivity and a using other types of macromolecular polynitroxides. Despite larger molecular size. We have synthesized the corresponding the low solubility in water of the third-generation poly- sodium salt derivative instead of the previously reported (propylenimine) (PPI) dendrimers conjugated with nitroxides, lithium salt derivative to improve biocompatibility and safety. their intra-articular administration to rabbit stifle joints produced significant enhancement of the articular cartilage in T1-weighted images.22 On the other hand, biodistribution ■ EXPERIMENTAL SECTION Synthesis. The synthesis of G3-Tyr-PROXYL-ONa was carried studies were performed with PPI dendrimers conjugated with out following the procedure previously described by us, with little spirocyclohexyl nitroxides and poly(ethylene glycol) (PEG) modifications.28 chains, providing selectively enhanced magnetic resonance G3-Tyr-PROXYL-ONa. Under dark conditions, G3-Tyr-PROXYL- imaging in mice for over 1 h.23 Only more recently, another OMe28 (110 mg, 4.17 μmol, 1 equiv) was added into a round- kind of macromolecular polynitroxide, not based on bottomed flask equipped with a stir bar and dissolved in 2 mL of dendrimers but on polymers or polymeric nanoparticles, has tetrahydrofuran (THF). NaOH (129 mg, 3.0 mmol, 720 equiv) was dissolved in 2 mL of Milli-Q water and transferred to the THF shown interesting properties as MRI CA. For example, solution of G3-Try-PROXYL-OMe. The reaction mixture was allowed nitroxide-functionalized brush-arm star polymer organic radical to stir at room temperature overnight. Afterward, THF was removed contrast agents (BASP-ORCAs) have shown extremely high r2 under a vacuum, and the aqueous solution was purified by dialysis relaxivity and accumulation in murine subcutaneous tumors (MWCO, 0.1−0.5 kDa) to remove the excess NaOH. The external (A459 tumor-bearing NCR-NU mice) for a long time water was changed after 2, 4, 21, and 46 h. Then, the aqueous solution 2768 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article of the dialysis bag was collected and the water was eliminated by NANBIOSIS, more specifically by the U20/FVPR at the Vall freeze-drying to afford the final product G3-Try-PROXYL-ONa as a d’Hebron Institute of Research (VHIR), using the limulus amebocyte pale-yellow solid in 67% yield. The full radical substitution was lysate (LAL) chromogenic method (A39553 de Pierce). quantitatively characterized by electron paramagnetic resonance Magnetic Resonance Imaging (MRI) Studies. Magnetic (EPR) and its purity by size exclusion chromatography (SEC) (see resonance imaging (MRI) studies were carried out at the joint the Supporting information). NMR facility of the Universitat Autònoma de Barcelona and CIBER- Materials and Methods. Chemicals. N-(tert-Butoxycarbonyl)-L- BBN (Cerdanyola del Vallès, Spain), Unit 25 of NANBIOSIS ICTS tyrosine methyl ester (Boc-Tyr-OMe), Cs2CO3, 3-carboxy-PROXYL, (https://www.nanbiosis.es/portfolio/u25-nmr-biomedical- 1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]- application-i/). MRI studies were performed in a 7.0 T horizontal- pyridinium 3-oxid hexafluorophosphate (HATU), N,N-diisopropyle- bore superconducting magnet (BioSpec 70/30; Bruker BioSpin, thylamine (DIPEA), and NaOH were obtained from Sigma-Aldrich. Ettlingen, Germany) equipped with actively shielded gradients (B- Trifluoracetic acid (TFA) was obtained from EMD Millipore. CH2Cl2 GA12 gradient coil inserted into a B-GA20S gradient system). For and CHCl3 were distillated from CaH2. Tetrahydrofuran (THF) was mouse brain MRI, a 72 mm inner-diameter linear volume coil was distillated from Na/benzophenone. Ultrafiltration was performed on used as the transmitter, and a dedicated mouse brain quadrature solvent-resistant stirred cells from EMD Millipore with regenerated surface coil was used as the receiver. For whole-body MRI, a 72 mm cellulose membranes (3, 5, and 10 kDa) and dialysis with a dialysis kit inner-diameter quadrature volume coil was used as the transceiver. (MWCO 100−500 Da, Thermo Fisher Scientific Inc.). Ultrapure MR data were acquired and processed on a Linux computer using water (Milli-Q, EMD Millipore) was used for ultrafiltration (together Paravision 5.1 software (Bruker BioSpin GmbH, Ettlingen, Germany). with high-performance liquid chromatography (HPLC)-grade ace- Solutions of gadopentetate dimeglumine (Gd-DTPA, Magnevist) tone) and dialysis. and the G3-Tyr-PROXYL-ONa radical dendrimer were prepared in a Animals. GL261 mouse glioma cells were obtained from the saline solution (NaCl 0.9%, B. Braun), and the injection volume was Tumour Bank Repository at the National Cancer Institute (Frederick, adjusted according to mice weight. Maryland). Cells were checked for the mouse short tandem repeat Animal Experimental Design. Tolerability and biodistribution (STR) profile as well as interspecies contamination. In addition, studies were performed in healthy (nontumor-bearing) C57BL/6 polymerase chain reaction (PCR) studies were performed to discard female mice. Since contrast-enhancement explorations are not mycoplasma and virus presence. All studies involving animals were expected to be repeated and cumulative, these studies were performed approved by the local ethics committee (Comissió d’È tica en with single-dose administrations. For biodistribution studies, a Experimentació Animal i Humana, CEEAH), according to regional 0.00625 mmol/kg dosage was used, with the objective of assessing and state legislations (protocol references CEA-OH-9685/CEEAH- the main organs related to G3-Tyr-PROXYL-ONa radical dendrimer 3665). Mice were purchased from Charles River Laboratories metabolization through MRI studies, which is not expected to vary for (l’Abresle, France) and housed at the animal facility of the Universitat different doses. However, tolerability studies were performed with the Autònoma de Barcelona. GL261 tumors were induced in a total of same dosage foreseen to be used in the dynamic contrast-enhanced 10C57BL/6 female wild-type (wt) mice by intracranial stereotactic magnetic resonance imaging (DCE-MRI) studies with the objective injection of 105 GL261 cells as already described by us,30 and the n = to ensure that no harmful effect was observed, at least with C57BL/6 7 mice with most homogeneous tumor volumes were chosen for healthy female mice. further studies. Mice were weighed twice a week, and tumor volumes MRI studies for assessing brain tumor contrast enhancement were were followed up using T2-weighted image (T2w) MRI acquisitions. performed with GL261 GB-bearing mice. Contrast agent admin- In addition, wt mice were used for studies such as ex vivo, istration (both the G3-Tyr-PROXYL-ONa radical dendrimer and the biodistribution and tolerability (described in the corresponding Gd-based commercial CA) was done intravenously under anesthesia. sections). Overall, a total of n = 27 C57BL/6 mice (weighing 19.43 The MRI exploration was performed, and animals were allowed to ± 1.46 g, aged 12 weeks) were used in this study. recover in a warm environment. Mice were euthanized after the whole Size Exclusion Chromatography (SEC). Size exclusion procedure was finished. chromatography (SEC) analysis was carried out using an Agilent MRI Studies. Ex Vivo Brain MRI Studies. The CAs for ex vivo 1260 infinity II liquid chromatography system apparatus equipped studies were dissolved in a saline solution (0.9% NaCl, B. Braun). The with a diode array detector. For the G3-Tyr-PROXYL-ONa amount finally used for each mouse was 5 nmol of Gd dissolved in 4 dendrimer, a PSS Suprema precolumn (10 μm, 8 × 50 mm2) and a μL of saline solution and 1.25 nmol of G3-Tyr-PROXYL-ONa radical PSS Suprema analytical column (10 μm, 100 Å, 8 × 300 mm2) were dendrimer. Mice were euthanized by cervical dislocation and used. LiCl 0.25 mM in water was used as the eluent at a flow rate of immobilized on a stereotactic holder (Kopf Instruments, Tujunga, 0.35 mL/min at 35 °C. The dendrimer was dissolved in the eluent to California). The contrast administration was carried out as described reach a final concentration of 1 mg/mL dendrimer and filtered for tumor generation, with three injection points as previously through a 0.2 μm nylon filter before injection. described.31 The whole process of CA injection ex vivo took 30 min. Electron Paramagnetic Resonance Spectroscopy (EPR). All imaging studies started with T2-weighted MRI screening with Electronic paramagnetic resonance spectroscopy (EPR) spectra high-resolution coronal T2w images using a rapid acquisition with were obtained with an X-Band (9.7 GHz) Bruker ELEXSYS 500 relaxation enhancement (RARE) sequence to evaluate brain tumor spectrometer equipped with a ST8911 microwave cavity, a Bruker presence and to monitor its evolution stage. The acquisition variable-temperature unit, a field frequency lock system Bruker ER parameters for MRI studies were as follows: repetition time (TR)/ 033 M and equipped with an NMR Gaussmeter Bruker ER 035 M. effective echo time (TEeff) = 4200:36 ms; echo train length (ETL) = The modulation amplitude was kept well below the line width, and 8; field of view (FOV) = 19.2 × 19.2 mm2; matrix size (MTX) = 256 the microwave power was well below saturation. All liquid samples × 256 (75 μm/pixel × 75 μm/pixel); slice thickness (ST) = 0.5 mm; were previously degassed with Ar. A quantitative EPR study was inter-ST = 0.1 mm; number of slices (NS) = 10; number of averages performed for G3-Tyr-PROXYL-ONa under the same conditions and (NA) = 4; and total acquisition time (TAT) = 6 min and 43 s. at the same concentration as for G0- to G3-Tyr-PROXYL-OLi,28 T1-weighted MRI: A multispin−multiecho (MSME) sequence was comparing the corresponding double integration value of the EPR used with FOV = 17.6 × 17.6 mm2; MTX = 128 × 128 matrix (138 spectrum with those of the former ones, resulting in an area matching μm/pixel × 138 μm/pixel); TR/TEeff = 200:8.5 ms; number of the full radical substitution. EPR spectra of urine were carried out in a repetitions (NR) = 1; and TAT = 4 min 16 s. quartz flat cell, and the different organ tissues were analyzed using a In Vivo MRI Studies. Mice anesthesia was performed with quartz tissue cell. Previously, tissue organs were weighed in an isoflurane (B. Braun, Melsungen, Germany) at 0.5−1.5% in O2, and analytical balance. the respiratory frequency was maintained between 40 and 60 breaths/ Endotoxin Determination. The endotoxin determination of the min. Body temperature was maintained with a recirculating water G3-Tyr-PROXYL-ONa sample was performed by the ICTS− system incorporated in the animal bed and measured with a rectal 2769 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Figure 1. Structure of the G3-Tyr-PROXYL-ONa radical dendrimer. probe. Respiration rate and temperature were constantly monitored T1 MRI studies. Three glioma-bearing mice were injected with (SA Instruments, Inc., New York). Before immobilization in the gadoterate meglumine, and another three were injected with G3-Tyr- animal holder, each mouse was cannulated in the tail vein using a PROXYL-ONa. A DCE T1 study was then performed using three home-built multidelivery polyethylene tubing system. In this case, a coronal sections. For this, an MSME sequence was used with FOV = 30G two-way catheter was connected through polyethylene tubing, to 17.6 × 17.6 mm2; MTX = 128 × 128 matrix (138 μm/pixel × 138 two independent 1 mL syringes (Becton−Dickinson S.A., Madrid, μm/pixel); TR/TE = 200:8.5 ms; ST = 1 mm; NA = 2; NR = 70; Spain) loaded with heparinized-saline (40 U/mL) (0.9% NaCl, B. TAT = 59 min 44 s. The contrast bolus was administered after the Braun and heparin, Mayne Pharma España, Madrid, Spain). third repetition of the complete T1-weighted sequence (about 2.5 min Brain. The T2- and T1-weighted MRI were performed with the after the start of the image acquisition protocol). DCE-MRI data were same parameters as described for ex vivo MRI. analyzed with DCE-@urLAB (http://oa.upm.es/28901/). T1 maps were performed with the RARE-VTR sequence with FOV Body. T1-weighted images were acquired using a respiratory gated = 17.6 × 19.28 mm2; MTX = 128 × 128 matrix (138 μm/pixel × 150 spin echo sequence (repetition time (TR)/echo time (TE) = μm/pixel); Teff = 7.5 ms and TR list: 100, 400, 700, 1000, 1300, 600:10.5 ms) acquiring 23 coronal sections with a field of view = 1700, 2000, 2600, 3500, and 5000 ms. NR = 1, TAT = 19 min 31s. 10 × 5 cm2, matrix size = 512 × 256, and slice thickness of 1 mm with DCE-MRI Studies. The contrast agent was injected into the mice as a 0.1 mm gap between slices a bolus (72−84 μL, doses varied according to the CA ranging from T2- and T1-weighted scout images were initially performed to be 0.00625 to 0.1 mmol/kg) during dynamic contrast-enhanced (DCE)- used as reference images for prescribing the final coronal sections 2770 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Figure 2. (a) Schematic representation of the region of interest (ROI) selection (blue circles) for analysis in T1-weighted MRI, indicating ipsilateral ROIs where CA are injected (right) and contralateral ROIs (left). T1-weighted MRI after ex vivo administration of (b) 5 nmol of gadopentetate dimeglumine to each injection point and (c) 1.25 nmol of the G3 radical dendrimer at each injection point. Figure 3. RCE% time-course curve (left) and data (right) obtained from the quantification of T1w MRI images. The curve displays the average RCE obtained along the time after ex vivo administration of 1.25 nmol G3 radical dendrimer. Bars show ± SD. through the main organs of interest (kidneys, liver, bladder, muscle, investigate whether some aspects of the tumor environment spleen). Afterward, DCE-T1w images and T1 map respiration gated (not possible to be evaluated in vitro) could modify/modulate acquisitions were performed before and after contrast agent injection. the ability of CA to produce contrast enhancement. A setup For whole-body DCE-T1w, an MSME sequence was used with FOV = experiment with gadopentetate dimeglumine was performed 9 × 3 cm2; MTX = 256 × 128 matrix; TR/TE = 400:10 ms; NA = 2; ST = 1 mm with 0.2 mm gap between slices; TAT = 1 min 42 s. for comparison purposes, using 5 nmol of gadopentetate T1 maps were performed on the same sections as DCE-T1w images dimeglumine to each injection point in the brain parenchyma. using a RARE-VTR sequence with FOV = 9 × 3 cm2; MTX = 128 × The relative contrast enhancement (RCE) was calculated, 128 matrix, TE = 7.5 ms; and TR list: 250, 400, 800, 1300, 1700, achieving a value of 232% ± 29, n = 3, and in agreement with 2400, and 3500 ms. NR = 1; TAT = 11 min 2 s. T1 map acquisitions previous ex vivo experiments with similar Gd-based contrast were used to calculate T1 values, drawing ROIs in defined zones and agents. The resulting T1-weighted MRI is shown in Figure 2b. measuring the estimated T1 before and after CA administration, After a first trial using 0.1 nmol of the G3 radical dendrimer, further expressed as a percentage change. which only produced a faint relative contrast enhancement, the Data Availability. The raw/processed data required to reproduce final administered amount that produced a noticeable relative these findings will be available under reasonable request to corresponding author(s). contrast enhancement in T1w MRI was found to be 1.25 nmol ■ of G3. With this administered amount, a clear enhancement was observed (Figure 2c). The corresponding calculated RESULTS AND DISCUSSION relative contrast enhancement was 237% ± 40, n = 3, which Synthesis. First, we synthesized the G3-Tyr-PROXYL- meant no significant differences compared with RCE OMe radical dendrimer derivative following the procedure calculated for 5 nmol Gd, suggesting that this concentration already described in the literature by us.28 was enough to get comparable RCE, at least in the ex vivo Then, the methyl ester was hydrolyzed with NaOH in THF/ environment. H2O (1:1), resulting in the water-soluble G3-Tyr-PROXYL- We also investigated whether RCE presented significant ONa dendrimer. The excess NaOH was removed by dialysis to changes along time due to possible radical inactivation. This purify the G3-Tyr-PROXYL-ONa dendrimer, henceforth was a first attempt to assess the stability of the radical referred to as the G3 radical dendrimer (Figure 1). The full dendrimer in a real biological medium, with a view to its use in functionalization of the G3 dendrimer with radicals was vivo. The first T1w MRI was acquired 31 min after injection confirmed by EPR, and its purity was monitored by SEC (see (the shorter timeframe feasible considering the experimental the Supporting Information). setup). In this first acquisition, the measured RCE was 237% ± Ex Vivo MRI Analysis. We followed an ex vivo method 40. After this first T1w acquisition, quantitative T1 maps were developed to select CAs with a suitable potential for in vivo acquired, followed by sequential T1w MRI until 111 min post efficiency using small amounts of compounds and minimum the first injection (Figure S2), and RCE values were animal use.31 Note that these analyses are not intended to set accordingly measured (Figure 3). Although the values tended the amount of CA to be administered in vivo but rather to to be slightly lower in the last time points analyzed, the overall 2771 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article RCE variation did not present statistical significance when the T1w MRI (Figure 4), confirmed by a decrease in T1 values different time points were compared (p always > 0.05). Thus, calculated in T1 maps. The corresponding T1w enhancement the stability of the radicals anchored to the dendrimer scaffold seemed quite prolonged in time in such a biological medium, presenting a similar RCE well beyond 1 h after injection. These data were encouraging for its subsequent use in vivo, since, as explained above, one of the main limitations of nitroxides is their rapid bioreduction. T1 values after the G3 radical dendrimer local injection were also assessed, and significant changes were detected. Ipsilateral (injection) ROIs presented an estimated T1 of 1076 ± 320 ms, whereas contralateral ROIs (control) presented values of 1914 ± 23 ms, with a significant average decrease of 46%, in agreement with satisfactory results obtained for RCE measure- ments in T1w MRI. Examples of curve adjustments for T1 maps are shown in Figure S3. Endotoxin Analysis of the G3 Radical Dendrimer. Before in vivo studies, an endotoxin analysis of the G3 radical dendrimer was carried out to confirm the absence of such toxin Figure 4. T1w MRI at different times after G3 radical dendrimer in samples administered to mice. We performed the analysis at administration (0.00625 mmol/kg). T1w representative images two different concentrations: 0.4 and 4 mg/mL. In the sample showing a slice through the kidneys (upper row) and through the with a lower concentration, the value obtained (0.02 EU/mL) bladder (lower row). The main enhancement was seen in the kidney was below the detection range, and in the sample with a higher cortex/pelvis and bladder. concentration, very low endotoxin levels were detected: 0.034 ± 0.003 EU/mL. These levels were considered acceptable was mostly observed in the kidney cortex and pelvis. Kidney T1 since the limit for administration in mice would be below 1 decreased by 37.7% at 20 min and by 23.8% at 60 min EU/mL (5 EU/kg for the usual routes of administration). postadministration, which corresponded to 41 and 16% overall In Vivo Studies. In an attempt to use the minimum increases in the T1w signal intensity, suggesting the relevance amount of the G3 radical dendrimer in the in vivo studies and of renal excretion for these compounds. This was further produce as similar as possible a Gd-like enhancement, we used confirmed by data obtained from the bladder: T1 decrease of two types of concentrations: first, a lower one of 0.00625 ca. 36% and signal intensity increase of 84%. mmol/kg; and later, a higher one (0.025 mmol/kg). These The biodistribution data for the higher dose finally chosen concentrations were chosen taking into account the usual Gd- for MRI studies was expected to be fully comparable to the based CA dose used in clinical and preclinical approaches (0.1 lower dose. For confirmation purposes, one GL261 GB- mmol/kg) and the in vitro relaxivity of the G3-Tyr-PROXYL- bearing mouse was explored with whole-body MRI 3 h after ONa radical dendrimer: r1 relaxivity measured at 7 T (13 s−1· G3 radical dendrimer administration at 0.025 mmol/kg once mM−1) was on the order of 4 times higher than Gd-DTPA DCE-MRI studies were finished. Although the timing did not relaxivity (3.2 s−1·mM−1),29 similar to the lithium salt completely match, the main contrast enhancement and T1 derivative.28 value decrease were also seen in the kidney cortex and pelvis The lower dose (0.00625 mmol/kg) of the G3 dendrimer is (41−59% in comparison with the basal measurements in wt 16 times lower than the usual Gd-based CA dose (0.1 mmol/ mice) and bladder (95% decrease), which is even higher than kg), but taking into account the 4 times higher relaxivity of G3, the values found in the biodistribution study with the lower we could expect only “4 times lower enhancement” than with dose. Since the main biodistribution parameters were Gd-based CA. Thus, it could be a good starting point. In fact, confirmed, we considered that the previous biodistribution this dose proved suitable to determine properly the experiment produced enough signal to be analyzed and was biodistribution due to reasonable enhancement detection. not repeated in a whole cohort of animals with the increased However, it proved inappropriate for tumor detection since it dose. Moreover, biodistribution obtained from EPR data at produced only a slight enhancement in the tumor periphery. lower and higher doses (see the next section) was in agreement On the other hand, the higher dose chosen later (0.025 with the data obtained from whole-body MRI acquisitions. mmol/kg) is 4 times lower than the standard dose of Gd-based A tolerability study was performed on three healthy C57BL/ CA (0.1 mmol/kg), but “in terms of enhancement”, we could 6 female mice administered with the higher dose 0.025 mmol/ expect similar enhancement as Gd-based CA taking into kg G3 radical dendrimer and closely followed up 10 days after account the aforementioned relaxivity values. In fact, this dose injection since acute toxicity would show up in the first few proved suitable for tumor detection. hours/days. In addition, they were also inspected for health In Vivo Biodistribution and Tolerability. The prelimi- status and welfare for a month. nary biodistribution studies (MRI-based) were performed with Administered mice did not show toxicity symptoms and did the G3 radical dendrimer at 0.00625 mmol/kg (lower dose) not experience body-weight loss either during the first 10 days administered through the tail vein, with n = 3 wt female or in the whole month of the follow-up. Figure 5 shows the C57BL/6 mice. With this experiment, we aimed to assess any variation of weight along time. Their weight evolution was as T1 changes in different organs to check for the biodistribution expected according to the Charles River growth chart for of G3 after intravenous administration. healthy C57BL/6 females of this age. Their overall states such The G3 radical dendrimer administration produced a slight as fur aspect, hydration, and behavior/activity were satisfac- enhancement in different organs of wt mice on whole-body tory. These results demonstrate the nontoxicity of the G3 2772 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Figure 5. Percentage of body-weight variation of mice injected with the G3 radical dendrimer at dose 0.025 mmol/kg in the G3 dendrimer. The red line indicates the 20% weight reduction point. Average starting values and body-weight evolution are in agreement with values described from the Jackson Laboratory (https://www.jax. org/jax-mice-and-services/strain-data-sheet-pages/body-weight-chart- 000664). radical dendrimer in vivo after systemic intravenous tail-vein injection at this dose. EPR-Based Biodistribution and Radical Stability In Vivo. Mice administered with the G3 radical dendrimer at lower and higher doses were euthanized to check the amount Figure 6. EPR spectra of urine, kidneys and liver from a C57BL/6 of G3 in some organs by electron paramagnetic resonance female wt mouse administered with 0.00625 mmol/Kg of G3-Tyr- (EPR) spectroscopy. A similar mass of organ tissues was PROXYL-ONa radical dendrimer, euthanized 1.5 h after of analyzed to obtain comparable results, and no significant intravenous administration. The urine spectrum was obtained from differences were found between the lower and higher doses’ the collected urine diluted to 1:2 with miliQ water. results. We analyzed the urine, kidneys and liver of a C57BL/6 surrounding brain. This is a relevant and desirable character- female wt mouse administered with 0.00625 mmol/kg (see istic in a contrast agent intended to be used for brain tumors. Figure 6 and the Supporting Information for additional data). In addition, a GL261 GB-bearing mouse with 0.025 mmol/ The highest amount of the G3 radical dendrimer (highest EPR kg dose was euthanized 15 h after administration, and EPR was signal intensity) was found in urine (even taking into account performed on bladder, kidneys, liver, heart, brain tumor, that it was diluted 1:2 with miliQ water to get the optimum healthy brain, and muscle. Similar organ tissue mass was volume to be properly measured), followed by kidneys and analyzed by EPR (around 52 mg each), and the corresponding liver. This result was in agreement with the whole-body T1w spectra are plotted in Figure 7. Similar to what was obtained MRI, also suggesting excretion of the radical dendrimer with the lower dose administered, the largest EPR signal through the kidneys. Interestingly, the EPR spectrum shape of intensity was found in the kidneys, followed by the bladder, urine sample was almost identical to the G3 spectrum before liver, brain tumor, and heart. It is also important to remark that injection (see Figure S4). This means that the radical character the selective accumulation behavior in the brain tumor was of PROXYL radicals in the G3 radical dendrimer was not reproduced at a higher dose (see the inset in Figure 7). quenched by circulating in the bloodstream, passing from the The observation of an intense EPR signal after 15 h of blood to the kidneys and the bladder, 1.5 h postadministration. intravenous administration confirmed the stability of PROXYL This is a relevant result that demonstrates the high stability of radicals in vivo, farther beyond that of isolated nitroxides. the radicals when anchored to the dendrimer, in contrast to the Related to this, the EPR signal observed in the heart means a fast reduction in vivo experienced by isolated nitroxides, long circulation half-life of the radical dendrimer. especially in the bloodstream and tissues, losing their contrast In Vivo MRI Studies with GL261 GB-Bearing Mice. As ability shortly after injection.13−16 previously mentioned, the lower dose of the G3 radical We also analyzed the kidneys, liver, brain tumor, healthy dendrimer (0.00625 mmol/kg) administered to GL261 GB- brain, and muscle of a GL261-tumor-bearing mouse bearing mice only produced a slight enhancement in the tumor administered with the same dose 0.00625 mmol/kg 1.5 h periphery and some hot spots within the tumor (Figure S6), postinjection (see Figure S5), showing similar results. The which may correspond to more highly perfused regions. A largest amount of the G3 radical dendrimer was found in the whole set of mouse brain MRI studies, including DCE-MRI, kidneys, followed by the liver and the brain tumor, while an was performed, as described in the Experimental Section. almost imperceptible EPR signal was detected in the healthy However, importantly, the higher dose (0.025 mmol/kg) brain and muscle. It is worth mentioning that an important produced a noticeable RCE in the tumor (Figures 8 and S7). amount of the G3 radical dendrimer was detected in the brain The aforementioned G3 radical dendrimer dose was intra- tumor, while the healthy surrounding brain barely showed any venously injected to n = 3 GL261 GB-bearing mice, and G3 signal, suggesting that the G3 radical dendrimer can experiments were conducted as for the lower dose previously selectively accumulate in tumors with few spreading to the described. 2773 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Figure 7. EPR spectra of kidneys, bladder, liver, heart, muscle, brain tumor, and contralateral healthy brain from GL261-tumor-bearing mice after 15 h of 0.025 mmol/kg administration of the radical dendrimer. On the other hand, enhancement achieved with G3 radical dendrimer administration proved mostly sustained along the time measured. The enhancement of the tumor zone reached a maximum of 126% after 6 min and remained essentially unchanged during the whole time course (RCE of 121% after 60 min). RCE data suggests that T1 contrast enhancement was sustained as long as 2.5 h after CA administration (not shown). Figure 8. Top: Axial T1w MRI for the follow-up of tumor contrast Therefore, the measured data suggest that tissue enhancement enhancement before and after 5 min (ca. maximum enhancement) and 60 min (final of first DCE-T1w-MRI follow-up, refer to Figure 9) triggered by the G3 radical dendrimer administration may of G3 radical dendrimer administration (0.025 mmol/kg) to GL261 persist well beyond the washout time of Gd. GB-bearing mice. This is also a relevant point to highlight, suggesting that an improvement in the imaging time frames could be achieved with G3 radical dendrimer administration when compared to Changes observed in T1 values were mostly seen in tumors Gd chelates, which show a very fast clearance. rather than in the contralateral brain. T1 values were calculated In summary, the selective accumulation behavior of the G3 before and after G3 administration with T1 map sequences ca. radical dendrimer in the brain tumor, the high stability of the 1 h after G3 administration. T1 decrease in the tumor was radicals in vivo, and the long circulation half-life of the radical 19.2% (29.8% increase in the measured signal), while no dendrimer make G3 radical dendrimers capable of imaging noticeable changes were found in the corresponding brain tumors over clinically meaningful time scales following contralateral part. systemic administration, at longer time periods than Gd The RCE and the kinetics of uptake and washout of the G3 chelates, without concerns over long-term tissue accumulation radical dendrimer at 0.025 mmol/kg dose were compared with of metals. ■ Gd-based CA administration at the standard dose of 0.1 mmol/kg, as well as for 0.04 mmol/kg. Only the first 60 min in the first DCE-MRI experiment of G3 were used for CONCLUSIONS comparison with Gd. The potential for in vivo contrast enhancement capabilities of a Remarkably, the RCE measured for the G3 radical radical dendrimer specially designed to act as a T1 contrast dendrimer at 0.025 mmol/kg dose was similar to, although agent for MRI was described in GL261 orthotopic GB-bearing slightly lower than, the RCE obtained with Gd-DTPA at the mice. In particular, the third generation of a water-soluble standard dose of 0.1 mmol/kg and proved higher than the radical dendrimer family based on poly(phosphorhydrazone) RCE obtained with Gd at 0.04 mmol/kg (Figure 9). However, dendrimers fully functionalized with PROXYL radicals on the interestingly, the kinetics of washout was completely different periphery (G3-Tyr-PROXYL-ONa radical dendrimer), pre- between both contrast agents. The maximum contrast senting high r1 relaxivity (13 mM−1·s−1). MR-based biodis- enhancement with Gd-DTPA varied with dose, ranging from tribution studies showed contrast enhancement mostly in the 113 to 158%, and started to decrease sharply after the first 5−6 kidney cortex and pelvis, suggesting the relevance of renal min, suggesting a fast washout. The RCE decreased to 125% excretion for this compound, confirmed by EPR analyses of the after 30 min in the case of the 0.1 mmol/kg dose and selected organs. Remarkably, it provides suitable contrast recovering basal values in the case of the 0.04 mmol/kg dose. enhancement on murine GL261 glioblastoma tumors com- 2774 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Figure 9. Top: Color-code scale for RCE (Gd 0.1 and 0.04 mmol/kg, 30 min; and G3 radical dendrimer 0.025 mmol/kg, 60 min). Bottom: ROI kinetics for these agents, showing the slower washout of G3. The maximum RCE calculated was 158% for Gd 0.1 mmol/kg, 113% for Gd 0.04 mmol/kg, and 126% for G3 0.025 mmol/kg. Contralateral enhancement was reproducible in all cases (102 ± 3%). Please note that 0.04 mmol/kg administration data comes from retrospective cases administered with Gd-DOTA instead of Gd-DTPA. Analyses were performed with the DCE@ urLAB software package.32 parable to commercial Gd-based contrast agents (0.1 mmol/ their established toxicity. These important features, together kg), at a 4 times lower concentration (0.025 mmol/kg dose), with the selective accumulation in brain tumor tissues and the mitigating concerns about toxic metal accumulation. In fact, no longer imaging time frames in comparison to Gd-based CA, signs of toxicity or weight loss were detected in mice after make the G3-Tyr-PROXYL-ONa radical dendrimer a viable systemic intravenous tail-vein injection of the radical alternative to metal-based MRI contrast agents, particularly on dendrimer. In addition, the selective accumulation of the G3 MRI analysis of glioblastomas. radical dendrimer in brain tumor tissue is a relevant and desirable characteristic in a contrast agent to be used for brain tumors. The G3 radical dendrimer also exhibited longer ■ ASSOCIATED CONTENT * Supporting Information sı retention within the tumor, which allows tumor imaging over The Supporting Information is available free of charge at longer time frames (≥2.5 h) than Gd chelates which present https://pubs.acs.org/doi/10.1021/acs.biomac.2c00088. faster clearance profiles. Moreover, high stability of the radicals Characterization of G3-Tyr-PROXYL-ONa, sequential anchored on the dendrimer surface when subjected to axial T1w MRI acquired along time after ex vivo biological media has been demonstrated both ex vivo and in stereotactic injection of the G3 dendrimer, T1 estimation vivo. By EPR, it proved to be much higher (≥15 h) than that of from T1 map sequences from ex vivo MRI acquisitions, isolated nitroxides, which are rapidly reduced (half-lives on the EPR spectra of the G3 radical dendrimer prior to be order of minutes). injected and after injection (from the collected urine), Therefore, the high contrast enhancement (high relaxivity) additional EPR-based biodistribution data, DCE-MRI of and high stability of the G3 radical dendrimer species tumor-bearing mouse administered with 0.00625 mmol/ demonstrate that the two major limitations of nitroxyl radicals kg, and T1w MRI for follow-up of tumor contrast have been overcome, even in in vivo conditions. At the same enhancement after G3 administration to tumor-bearing time, it may also solve the major concern of Gd-based CAs, mouse at 0.025 mmol/kg (PDF) 2775 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules ■ pubs.acs.org/Biomac Article AUTHOR INFORMATION TRIBUTE (PID2020-113058GB-I00/AEI/10.13039/ Corresponding Authors 501100011033). Vega Lloveras − Institut de Ciència de Materials de Barcelona, Notes ICMAB−CSIC, 08193 Bellaterra, Spain; CIBER de The authors declare no competing financial interest. Bioingeniería, Biomateriales y Nanomedicina, Instituto de Salud Carlos III, 08913 Bellaterra, Spain; orcid.org/ 0000-0002-9653-3340; Email: vega@icmab.es ■ ACKNOWLEDGMENTS S.Z. and W.S. acknowledge their China Scholarship Council Ana Paula Candiota − CIBER de Bioingeniería, Biomateriales (CSC) grants. We acknowledge the UAB Predoctoral training y Nanomedicina, Instituto de Salud Carlos III, 08913 program (14a̲ Convocatoria PIF-19612, predoctoral fellow- Bellaterra, Spain; Departament de Bioquímica i Biologia ships for P.C.-P.). A.P.C. acknowledges 2018 XARDI 00016/ Molecular, Unitat de Bioquímica de Biociències, Edifici Cs, IU68-013944 (XarTEC SALUT) for support. The authors Universitat Autònoma de Barcelona, 08193 Bellaterra, thank Yufei Wu for his help in the radical dendrimer Spain; Institut de Biotecnologia i de Biomedicina (IBB), characterization. ■ Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain; orcid.org/0000-0002-1523-6505; REFERENCES Email: AnaPaula.Candiota@uab.cat (1) Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B. José Vidal-Gancedo − Institut de Ciència de Materials de Gadolinium(III) Chelates as MRI Contrast Agents: Structure, Barcelona, ICMAB−CSIC, 08193 Bellaterra, Spain; CIBER Dynamics, and Applications. Chem. Rev. 1999, 99, 2293−2352. de Bioingeniería, Biomateriales y Nanomedicina, Instituto de (2) Braverman, I. M.; Cowper, S. Nephrogenic Systemic Fibrosis. Salud Carlos III, 08913 Bellaterra, Spain; Email: j.vidal@ F1000 Med. Rep. 2010, 2, No. 84. icmab.es (3) Kanda, T.; Fukusato, T.; Matsuda, M.; Toyoda, K.; Oba, H.; Kotoku, J.; Haruyama, T.; Kitajima, K.; Furui, S. Gadolinium-based Authors Contrast Agent Accumulates in the Brain Even in Subjects without Songbai Zhang − Institut de Ciència de Materials de Severe Renal Dysfunction: Evaluation of Autopsy Brain Specimens Barcelona, ICMAB−CSIC, 08193 Bellaterra, Spain; with Inductively Coupled Plasma Mass Spectroscopy. Radiology 2015, orcid.org/0000-0003-1084-6483 276, 228−232. (4) Rogosnitzky, M.; Branch, S. Gadolinium-based Contrast Agent Silvia Lope-Piedrafita − Servei de Ressonància Magnètica Toxicity: a Review of Known and Proposed Mechanisms. Biometals Nuclear, Universitat Autònoma de Barcelona, 08193 2016, 29, 365−376. Bellaterra, Spain; CIBER de Bioingeniería, Biomateriales y (5) Ranga, A.; Agarwal, Y.; Garg, K. J. Gadolinium based Contrast Nanomedicina, Instituto de Salud Carlos III, 08913 Agents in Current Practice: Risks of Accumulation and Toxicity in Bellaterra, Spain; orcid.org/0000-0002-8127-6425 Patients with Normal Renal Function. Indian J. Radiol. Imaging 2017, Pilar Calero-Pérez − Departament de Bioquímica i Biologia 27, 141−147. Molecular, Unitat de Bioquímica de Biociències, Edifici Cs, (6) Olchowy, C.; Cebulski, K.; Łasecki, M.; Chaber, R.; Olchowy, Universitat Autònoma de Barcelona, 08193 Bellaterra, A.; Kałwak, K.; Zaleska-Dorobisz, U. The Presence of the Spain; CIBER de Bioingeniería, Biomateriales y Gadolinium-based Contrast Agent Depositions in the Brain and Nanomedicina, Instituto de Salud Carlos III, 08913 Symptoms of Gadolinium Neurotoxicity - A Systematic Review. PLoS One 2017, 12, No. e0171704. Bellaterra, Spain (7) Brasch, R. C.; London, D. A.; Wesbey, G. E.; Tozer, T. N.; Shuang Wu − Departament de Bioquímica i Biologia Nitecki, D. E.; Williams, R. D.; Doemeny, J.; Tuck, L. D.; Lallemand, Molecular, Unitat de Bioquímica de Biociències, Edifici Cs, D. P. Work in Progress: Nuclear Magnetic Resonance Study of a Universitat Autònoma de Barcelona, 08193 Bellaterra, Spain Paramagnetic Nitroxide Contrast Agent for Enhancement of Renal Complete contact information is available at: Structures in Experimental Animals. Radiology 1983, 147, 773−779. https://pubs.acs.org/10.1021/acs.biomac.2c00088 (8) Brasch, R. C.; Nitecki, D. E.; Brant-Zawadzki, M.; Enzmann, D. R.; Wesbey, G. E.; Tozer, T. N.; Tuck, L. D.; Cann, C. E.; Fike; Sheldon, P. Brain Nuclear Magnetic Resonance Imaging Enhanced by Author Contributions a Paramagnetic Nitroxide Contrast Agent: Preliminary Report. Am. J. The manuscript was written through contributions of all Roentgenol. 1983, 141, 1019−1023. authors. All authors have given approval to the final version of (9) Bennett, H. F.; Brown, R. D.; Koenig, S. H.; Swartz, H. M. the manuscript. Effects of Nitroxides on the Magnetic Field and Temperature Dependence of 1/T1 of Solvent Water Protons. Magn. Reson. Med. Funding 1987, 4, 93−111. This work was supported by Ministerio de Ciencia e (10) Rosen, G. M.; Griffeth, L. K.; Brown, M. A.; Drayer, B. P. Innovación MICINN (PID2019-105622RB-I00, CTQ2017- Intrathecal Administration of Nitroxides as Potential Contrast Agents 90596-REDT, Severo Ochoa FUNMAT-FIP-2018, and for MR Imaging. Radiology 1987, 163, 239−243. FUNFUTURE-FIP-2020 projects), CSIC (Intramural CSIC (11) Afzal, V.; Brasch, R. C.; Nitecki, D. E.; Wolff, S. Nitroxyl Spin project 201760E080), and Generalitat de Catalunya (2017 Label Contrast Enhancers for Magnetic Resonance Imaging. Studies SGR918). ICMAB acknowledges Spanish MICINN through of Acute Toxicity and Mutagenesis. Invest. Radiol. 1984, 19, 549−552. the Severo Ochoa Centres of Excellence Programme Grants (12) Sosnovsky, G. A Critical Evaluation of the Present Status of SEV-2015-0496 and CEX2019-000917-S. The work was also Toxicity of Aminoxyl Radicals. J. Pharm. Sci. 1992, 81, 496−499. (13) Hyodo, F.; Soule, B. P.; Matsumoto, K.-I.; Matusmoto, S.; funded by Centro de Investigación Biomédica en Red − Bioingenieria,́ Biomateriales y Nanomedicina, an initiative of Cook, J. A.; Hyodo, E.; Sowers, A. L.; Krishna, M. C.; Mitchell, J. B. Assessment of Tissue Redox Status Using Metabolic Responsive the Instituto de Salud Carlos III (Spain) cofunded by EU Contrast Agents and Magnetic Resonance Imaging. J. Pharm. Fondo Europeo de Desarrollo Regional (FEDER). A.P.C. Pharmacol. 2010, 60, 1049−1060. received funding from the ATTRACT project funded by the (14) Hyodo, F.; Chuang, K.-H.; Goloshevsky, A. G.; Sulima, A.; EC under Grant Agreement 777222 and the MICINN grant Griffiths, G. L.; Mitchell, J. B.; Koretsky, A. P.; Krishna, M. C. Brain 2776 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777 Biomacromolecules pubs.acs.org/Biomac Article Redox Imaging Using Blood-brain Barrier-permeable Nitroxide MRI (30) Simões, R. V.; García-Martín, M. L.; Cerdán, S.; Arús, C. Contrast Agent. J. Cereb. Blood Flow Metab. 2008, 28, 1165−1174. Perturbation of Mouse Glioma MRS Pattern by Induced Acute (15) Hyodo, F.; Matsumoto, K.; Matsumoto, A.; Mitchell, J. B.; Hyperglycemia. NMR Biomed. 2008, 21, 251−264. Krishna, M. C. Probing the Intracellular Redox Status of Tumors with (31) Candiota, A. P.; Acosta, M.; Simões, R. V.; Delgado-Goñi, T.; Magnetic Resonance Imaging and Redox-Sensitive Contrast Agents. Lope-Piedrafita, S.; Irure, A.; Marradi, M.; Bomati-Miguel, O.; Cancer Res. 2006, 66, 9921−9928. Miguel-Sancho, N.; Abasolo, I.; Schwartz, S., Jr.; Santamaria, J.; (16) Davis, R. M.; Matsumoto, S.; Bernardo, M.; Sowers, A.; Penadés, S.; Arús, C. A New Ex Vivo Method to Evaluate the Matsumoto, K.; Krishna, M. C.; Mitchell, J. B. Magnetic Resonance Performance of Candidate MRI Contrast Agents: A Proof-of-Concept Imaging of Organic Contrast Agents in Mice: Capturing the Whole- Study. J. Nanobiotechnol. 2014, 12, 1−14. Body Redox Landscape. Free Radical Biol. Med. 2011, 50, 459−468. (32) Ortuño, J. E.; Ledesma-Carbayo, M. J.; Simões, R. V.; Candiota, (17) Nguyen, H.V.-T.; Chen, Q.; Paletta, J. T.; Harvey, P.; Jiang, Y.; A. P.; Arús, C.; Santos, A. DCE@urLAB: A Dynamic Contrast- Zhang, H.; Boska, M. D.; Ottaviani, M. F.; Jasanoff, A.; Rajca, A.; Enhanced MRI Pharmacokinetic Analysis Tool for Preclinical Data. Johnson, J. A. Nitroxide-Based Macromolecular Contrast Agents with BMC Bioinf. 2013, 14, 316. Unprecedented Transverse Relaxivity and Stability for Magnetic Resonance Imaging of Tumors. ACS Cent. Sci. 2017, 3, 800−811. (18) Guo, S.; Wang, X.; Dai, Y.; Dai, X.; Li, Z.; Luo, Q.; Zheng, X.; Gu, Z.; Zhang, H.; Gong, Q.; Luo, K. Enhancing the Efficacy of Metal-Free MRI Contrast Agents via Conjugating Nitroxides onto PEGylated Cross-Linked Poly(Carboxylate Ester). Adv. Sci. 2020, 7, No. 2000467. (19) Chan, J. M. W.; Wojtecki, R. J.; Sardon, H.; Lee, A. L.; Smith, C. E.; Shkumatov, A.; Gao, S.; Kong, H.; Yang, Y. Y.; Hedrick, J. L. Self-Assembled, Biodegradable Magnetic Resonance Imaging Agents: Organic Radical-Functionalized Diblock Copolymers. ACS Macro Lett. 2017, 6, 176−180. (20) Huang, L.; Yan, C.; Cui, D.; Yan, Y.; Liu, X.; Lu, X.; Tan, X.; Lu, X.; Xu, J.; Xu, Y.; Liu, R. Organic Radical Contrast Agents Based on Polyacetylenes Containing 2,2,6,6-Tetramethylpiperidine 1-Oxyl (TEMPO): Targeted Magnetic Resonance (MR)/Optical Bimodal Imaging of Folate Receptor Expressing HeLa Tumors In Vitro and In Vivo(a). Macromol. Biosci. 2015, 15, 788−798. (21) Fu, C.; Yu, Y.; Xu, X.; Wang, Q.; Chang, Y.; Zhang, C.; Zhao, J.; Peng, H.; Whittaker, A. K. Functional Polymers as Metal-Free Magnetic Resonance Imaging Contrast Agents. Prog. Polym. Sci. 2020, 108, No. 101286. (22) Winalski, C. S.; Shortkroff, S.; Mulkern, R. V.; Schneider, E.; Rosen, G. M. Magnetic Resonance Relaxivity of Dendrimer-Linked Nitroxides. Magn. Reson. Med. 2002, 48, 965. (23) Rajca, A.; Wang, Y.; Boska, M.; Paletta, J. T.; Olankitwanit, A.; Swanson, M. A.; Mitchell, D. G.; Eaton, S. S.; Eaton, G. R.; Rajca, S. Organic Radical Contrast Agents for Magnetic Resonance Imaging. J. Am. Chem. Soc. 2012, 134, 15724−15727. (24) Zhang, S.; Lloveras, V.; Pulido, D.; Liko, F.; Pinto, L. F.; Albericio, F.; Royo, M.; Vidal-Gancedo, J. Radical Dendrimers Based on Biocompatible Oligoethylene Glycol Dendrimers as Contrast Agents for MRI. Pharmaceutics 2020, 12, 772. (25) Ding, L.; Lyu, Z.; Dhumal, D.; Kao, C.-L.; Bernard, M.; Peng, L. Dendrimer-Based Magnetic Resonance Imaging Agents for Brain Cancer. Sci. China Mater. 2018, 61, 1420−1443. (26) Badetti, E.; Lloveras, V.; Wurst, K.; Sebastián, R. M.; Caminade, A.-M.; Majoral, J.-P.; Veciana, J.; Vidal-Gancedo, J. Synthesis and Structural Characterization of a Dendrimer Model Compound based on a Cyclotriphosphazene Core with TEMPO Radicals as Substituents. Org. Lett. 2013, 15, 3490−3493. (27) Badetti, E.; Lloveras, V.; Muñoz-Gómez, J. L.; Sebastián, R. M.; Caminade, A. M.; Majoral, J. P.; Veciana, J.; Vidal-Gancedo, J. Radical Dendrimers: A Family of Five Generations of Phosphorus Dendrimers Functionalized with TEMPO Radicals. Macromolecules 2014, 47, 7717−7724. (28) Pinto, L. F.; Lloveras, V.; Zhang, S.; Liko, F.; Veciana, J.; Muñ oz-Gómez, J. L.; Vidal-Gancedo, J. Fully Water-Soluble Polyphosphorhydrazone-Based Radical Dendrimers Functionalized with Tyr-PROXYL Radicals as Metal-Free MRI T1 Contrast Agents. ACS Appl. Bio Mater. 2020, 3, 369−376. (29) Shen, Y.; Goerner, F. L.; Snyder, C.; Morelli, J. N.; Hao, D.; Hu, D.; Li, X.; Runge, V. M. T1 Relaxivities of Gadolinium-Based Magnetic Resonance Contrast Agents in Human Whole Blood at 1.5, 3, and 7 T. Invest. Radiol. 2015, 50, 330−338. 2777 https://doi.org/10.1021/acs.biomac.2c00088 Biomacromolecules 2022, 23, 2767−2777