Brain Uptake of a Fluorescent Vector Targeting the Transferrin Receptor: A Novel Application of in...

Brain Uptake of a Fluorescent Vector Targeting the TransferrinReceptor: A Novel Application of in Situ Brain PerfusionWael Alata,†,‡ Sarah Paris-Robidas,†,‡ Vincent Emond,‡ Fanchon Bourasset,§ and Frederic Calon*,†,‡

†Faculty of Pharmacy, Universite Laval, Quebec, QC G1V 0A6, Canada‡Neurosciences Axis, Centre de recherche du CHU de Quebec, Quebec (QC), QC G1V 4G2, Canada§Department of Pharmacokinetics, INSERM U705, CNRS UMR 8206, Faculty of Pharmacy, Universite Paris Descartes, Universite Paris Diderot, Sorbonne Paris Cite, Paris, France

ABSTRACT: Monoclonal antibodies (mAbs) targeting blood−brain barrier (BBB) transporters are being developed for braindrug targeting. However, brain uptake quantification remains achallenge, particularly for large compounds, and often requires theuse of radioactivity. In this work, we adapted an in situ brainperfusion technique for a fluorescent mAb raised against the mousetransferrin receptor (TfR) (clone Ri7). We first confirmed in vitrothat the internalization of fluorolabeled Ri7 mAbs is saturable anddependent on the TfR in N2A and bEnd5 cells. We next showed thatthe brain uptake coefficient (Clup) of 100 μg (∼220 nM) of Ri7mAbs fluorolabeled with Alexa Fluor 750 (AF750) was 0.27 ± 0.05μL g−1 s−1 after subtraction of values obtained with a control IgG. Alinear relationship was observed between the distribution volume VD(μL g−1) and the perfusion time (s) over 30−120 s (r2 = 0.997), confirming the metabolic stability of the AF750-Ri7 mAbsduring perfusion. Co-perfusion of increasing quantities of unlabeled Ri7 decreased the AF750-Ri7 Clup down to control IgGlevels over 500 nM, consistent with a saturable mechanism. Fluorescence microscopy analysis showed a vascular distribution ofperfused AF750-Ri7 in the brain and colocalization with a marker of basal lamina. To our knowledge, this is the first reported useof the in situ brain perfusion technique combined with quantification of compounds labeled with near-infrared fluorophores.Furthermore, this study confirms the accumulation of the antitransferrin receptor Ri7 mAb in the brain of mice through asaturable uptake mechanism.

KEYWORDS: blood−brain barrier, in situ brain perfusion, transferrin receptor, fluorescence

■ INTRODUCTION

Poor brain bioavailability has been recognized as one of theleading causes of failure in drug development.1−3 Such lowbrain access to drugs is mainly attributed to the presence of theblood−brain barrier (BBB), which protects the brain from mostbloodborne hydrophilic molecules. The BBB consists of braincapillary endothelial cells (BCECs), which are distinct fromother endothelial cells by virtue of the presence of tightjunctions and the expression of multiple influx and effluxtransport systems.4 The basal lamina, pericytes, and astrocyticend-feet located near the BCECs also contribute to thecohesion of the BBB structure.5 Even as of today, manycompounds enter phase II and phase III clinical trials with brainbioavailability issues unresolved. Thus, it is essential to developvalid methods to determine the capacity of drug candidates tocross the BBB as early as possible in the drug developmentprocess.2,6

Among the few techniques developed to assess the brainuptake of molecules, in vivo methods involving an injection intothe carotid artery, such as the in situ brain perfusion technique(ISBP), present key quantitative advantages.7−9 ISBP came tolight in 1984 when it was applied to rats by Takasato et al.10 and

was later applied to mice by Dagenais et al. in 200011 (Figure1). Besides high sensitivity,12 ISBP displays several otheradvantages, such as complete control over the perfusate contentand flow rate while avoiding peripheral metabolism and thepossibility of directly assessing the transport mechanisms.13,14

However, the use of ISBP is generally limited to compounds forwhich a radiolabeled version is available. Because of cost,growing concerns regarding radioprotection, and the relativesimplicity and versatility of recently developed fluorolabelingprocedures, the need to develop a fluorescence-based ISBPtechnique is becoming more evident. The recent developmentof far-red and near-infrared (NIR) fluorescent dyes withspectroscopic characteristics that allow a differentiationbetween the true signal and autofluorescence has renderedthe quantification of total fluorescence in brain homogenatespossible.15

Received: July 19, 2013Revised: October 29, 2013Accepted: November 11, 2013Published: November 11, 2013

Article

pubs.acs.org/molecularpharmaceutics

© 2013 American Chemical Society 243 dx.doi.org/10.1021/mp400421a | Mol. Pharmaceutics 2014, 11, 243−253

pubs.acs.org/molecularpharmaceutics

As the brain requires bloodborne molecules for its normalfunction, the BBB expresses transporters, such as the transferrinreceptor (TfR) and insulin receptor. Vectors targeting thesetransporters have thus been developed using monoclonalantibody (mAb) technology.16,17 Because of the highexpression of the TfR on BCECs,18 reports suggest thatmAbs recognizing rat (clone OX-26) and mouse (clones 8D3and Ri7) TfR are uptaken by the brain after systemicinjections.19−25 However, the quantitative evidence reportedwas obtained using 8D3 and Ri7 mAbs labeled with the γ-ray-emitting radioisotope 125I.21 We recently discovered that NIR-fluorescent dyes such as Alexa Fluor 750 allow quantification ofvector penetration in brain homogenates after IV injection as aresult of the relatively low brain autofluorescence levels with theabsorption/emission parameters selected.22 A window ofopportunity was thus opened for using ISBP with fluorescentcompounds. Therefore, the goal of this study was to develop anonradioactive, fluorescence-based ISBP method to quantifyand characterize the brain uptake of a fluorolabeled protein invivo. Because of the importance of TfR in brain drug delivery,the Ri7 mAb vector targeting the TfR was selected as the maintest compound.

■ EXPERIMENTAL SECTIONAnimals. Adult (4 week old) male BALB/c mice (Charles

River Laboratories Inc., Wilmington, MA, USA) weighing 20 to30 g were used. They had free access to food and water. Allprocedures were performed in accordance with the CanadianCouncil on Animal Care standards and were approved by the

Animal Ethics Committee of the Centre Hospitalier del’Universite Laval.

Reagents. [14C]sucrose (0.6 Ci mmol−1) was purchasedfrom Moravek Biochemicals (Brea, CA, USA). SOLVABLEsolubilizer and Ultima Gold scintillation cocktail werepurchased from PerkinElmer (Boston, MA, USA). AlexaFluor (AF) dyes, 4′,6-diamidino-2-phenylindole dilactate(DAPI), and culture reagents were purchased from LifeTechnologies (Burlington, ON, Canada).

Production and Purification of mAbs in Vitro. Hybrid-oma cell lines were cultured in CELLine bioreactors in mAbserum-free medium (BD Biosciences, Mississauga, ON,Canada). Supernatants were harvested weekly. mAbs werepurified using HiTrap protein G columns and the Akta PrimePlus system (GE Healthcare, Baie d’Urfe, QC, Canada)according to the manufacturer’s recommendations. Purifiedantibodies were concentrated with Amicon (Millipore Corpo-ration, Billerica, MA, USA) ultracentrifugal devices [molecularmass cutoff (MMCO) 30 kDa] and subsequently dialyzedagainst 0.01 M phosphate-buffered saline (PBS), pH 7.4, using10 kDa MMCO Slide-A-Lyzer dialysis cassettes (PierceChemical, Rockford, IL, USA). Protein concentrations weredetermined using bicinchoninic acid assays (Pierce Chemical).The weekly yield of purified mAbs averaged 3 mg. Hybridomacell line Ri7.217.1.4 (Ri7) secreting the rat mAbs specific forthe mouse TfR was obtained from Dr. Jayne Lesley (SalkInstitute, La Jolla, CA, USA) via Dr. Pauline Johnson(University of British Columbia, Vancouver, BC, Canada).Purified rat IgG2a isotype control (clone 2A3) was acquiredfrom Bio X Cell (West Lebanon, NH, USA).

mAb Conjugation to Alexa Fluor Dyes. mAbs (Ri7,control IgG) were thiolated with a 40:1 M excess of freshlyprepared 2-iminothiolane (Traut’s reagent, Sigma-Aldrich,Oakville, ON, Canada) after a 1 h incubation in 0.15 Msodium borate/0.1 mM EDTA, pH 8.5. Thiolated mAbs werediluted in 0.05 M HEPES/0.1 mM EDTA, pH 7.0, and thenconcentrated using a 30 kDa MMCO Vivaspin filter device(Sartorius Stedim Biotech, Aubagne, France). To conjugatethiolated mAbs, AF750 or AF488 C5-maleimide (125 nM) wasadded and incubated overnight in 2 mL glass bottles under aninert nitrogen atmosphere. Vivaspin devices were then used todiscard unbound AF maleimide and concentrate the AF-conjugated mAbs. Volumes were completed to have the desiredconcentration (5 μg/μL) with HEPES/EDTA buffer before theperfusion.

Cell Culture Studies. Murine Neuro-2a (N2A) cells (CCL-131, ATCC, Manassas, VA, USA) and murine bEnd5 cells(96091930, ECACC, Public Health England, Salisbury, UK)were used for in vitro internalization studies. The N2A cell linehas been established from a spontaneous neuroblastoma tumorof a strain A albino mouse. These cells have a neuronal andamoeboid stem cell morphology and express acetylcholinester-ase and tubulin.26 The bEnd5 cell line has been establishedfrom brain endothelial cells of BALB/c mice. These cells havean endothelial-like morphology and are positive for PECAM-1,endoglin, -32, and Flk-1.27−29 Cells were seeded in 96-wellplates (40 000 and 30 000 cells per well for N2A and bEnd5,respectively) containing 100 μL of Dulbecco’s Modified EagleMedium (DMEM) with 10% fetal bovine serum (FBS). Thenext day, cells were incubated with the specified concentration(0.4, 2, or 10 nM) of either AF488-Ri7 or AF488-IgG for 5, 15,or 45 min at 37 °C, unless specified otherwise. Alternatively, forcompetition studies, N2A and bEnd5 cells were first pretreated

Figure 1. Illustrative schema of the in situ brain perfusion technique.The right common carotid artery and the external carotid artery areligated at the heart side and the level of bifurcation, respectively. Theright common carotid artery is then catheterized with a polyethylenecatheter filled with heparin. Before perfusion, the thoracic cavity isopened and the heart is cut, and the perfusion is started immediatelyvia the internal carotid artery. (Adapted from ref 10.)

Molecular Pharmaceutics Article

dx.doi.org/10.1021/mp400421a | Mol. Pharmaceutics 2014, 11, 243−253244

for 15 min with unlabeled Ri7 or IgG (0, 10, 50, or 250 nM)and then exposed for a further 15 min to AF488-Ri7 (5 nM).Cells were then washed once with cold 0.1 M PBS and thenwith 0.025% trypan blue to quench noninternalized fluo-rescence. Measures were taken with a Synergy HT plate reader(filters: excitation, 485/20 nm; emission, 528/20 nm) fromBiotek (Winooski, VT, USA). To visualize AF488-Ri7accumulation, N2A and bEnd5 cells were seeded as mentioned,and 10 nM AF488-Ri7 was added to the medium at 4 °C.Incubation was resumed for 30 min at either 4 or 37 °C. Cellswere then washed with PBS, fixed in 4% paraformaldehyde,stained with DAPI, washed, and imaged using an EVOS FLAuto Cell Imaging System (Life Technologies) with DAPI andGFP filter cubes. Z-stack images (12 μm depth, 60 photos,0.366 μm step) were acquired, and deconvolution wasprocessed with the DeconvolutionLab plugin (BiomedicalImaging Group, EPFL, Lausanne, Switzerland) for ImageJ(National Institutes of Health, Bethesda, MD, USA).Surgical Procedure and Perfusion Technique. The

surgery and conditions of perfusion were done as previouslydescribed.11,30 Mice were anesthetized by intraperitonealinjection of xylazine/ketamine (8/140 mg kg−1). Next, theright common carotid artery was catheterized following ligationof the external branch (Figure 1). Before perfusion, the thoraxof the animal was opened and the heart was cut, and perfusionwas immediately started with a flow rate of 2.5 mL min−1. Theperfusion fluid consisted of bicarbonate-buffered physiologicalsaline (mM solute: 128 NaCl, 24 NaHCO3, 4.2 KCl, 2.4NaH2PO4, 1.5 CaCl2, 0.9 MgCl2, 9 D-glucose). The solutionwas gassed with 95% O2/5% CO2 to obtain a pH of 7.4 andheated to 37 °C. Unless specified otherwise, a fixed quantity(100 or 200 μg) of either AF750-Ri7 or AF750-IgG wereperfused during 1 min, followed by a 4 min washout with 10mL of the same buffer. The washout time was included (1) toremove unbound AF750-labeled mAbs from brain vessels andcapillaries and (2) to allow enough time for receptor-mediateduptake, which was expected to require a few minutes.12 Such along washout carries the risk of release of AF750-Ri7 fromBCECs back to the capillary lumen in the last minutes ofperfusion. However, this is unlikely given the high affinity ofRi7 to the TfR31 and the fact that the linear relationshipbetween Clup and time (see Figure 5 in the Results) wasconsistent with unidirectional uptake of AF750-Ri7.BBB Integrity during the Perfusion. In a subset of

animals, co-perfusion with 0.3 μCi/μL [14C]sucrose (a vascularspace marker that does not cross the BBB) during the lastminute of the washout provided a measure of the vascularvolume and an evaluation of the physical integrity of the BBB.Quantification of [14C]sucrose was performed by digesting theperfused hemisphere with 1 mL of SOLVABLE at 50 °Covernight and then adding 9 mL of Ultima Gold scintillationcocktail to the mix. Total [14C] was determined in a PackardTri-Carb model 1900TR liquid scintillation analyzer. Thecalculations of vascular volumes were done as describedpreviously.11,30

Brain Homogenates. The hemispheres perfused withAF750-mAbs were homogenized in 4 volumes of lysis buffer(150 mM NaCl, 10 mM NaH2PO4, 1% Triton X-100, 0.5%sodium dodecyl sulfate, and 0.5% sodium deoxycholate)containing Complete protease inhibitor cocktail (RocheDiagnostics, Indianapolis, IN, USA) with phosphatase inhib-itors (50 mM sodium fluoride and 1 mM sodiumpyrophosphate). Samples were sonicated briefly (3 × 10 s)

and centrifuged at 10000g for 20 min at 4 °C to generate afraction containing all of the detergent-soluble proteins,including cytosolic, extracellular, nuclear, and membrane-bound proteins. Samples of the perfusate and brainhomogenates were added to black 96-well plates, andfluorescence was measured with a Kodak 4000MM imagestation with filters appropriate for AF750 (excitation, 720 nm;emission, 790 nm). Images were analyzed and pseudocoloredwith the Kodak MI software (Carestream Health, Woodbridge,CT, USA). The final fluorescence values were normalizedaccording to exposure time, total fluorescence dose adminis-tered, and brain weight.

Calculations of Transport Parameters. In AF750-mAbperfusion experiments, the calculations were similar to those forpreviously published data,11,30 but there were some mod-ifications. The distribution volume (VD, μL g−1) of afluorolabeled vector was calculated using the followingequation:

=V X C/D brain perf

where Xbrain (fluo g−1) is the amount of fluorescence in the right

brain hemisphere and Cperf (fluo μL−1) is the concentration offluorescence in the perfusate.The apparent brain uptake coefficient (Clup, μL g−1 s−1) was

calculated using the following equation:

= V TClup /D perf

where Tperf is the perfusion time (s).The quantity of fluorolabeled vector per gram of brain (Q, μg

g−1) was calculated using the following equation:

=Q X F/brain perf

where Fperf (fluo μg−1) is the fluorescence per microgram offluorolabeled vector, which was calculated by dividing Cperf(fluo μL−1) by the quantity of fluorolabeled vector permicroliter (μg μL−1).

Immunohistofluorescence. Tissue preparation for brainlocalization of Ri7 was performed on hemispheres perfusedwith 100 μg of AF750-Ri7, postfixed in 4% paraformaldehydefor 2 days, and transferred into a 20% sucrose/0.5% sodiumazide solution for cryoprotection. Coronal brain sections with athickness of 25 μm were cut with a freezing microtome (LeicaMicrosystems, Richmond Hill, ON, Canada). Experiments wereperformed on three animals and on at least five sections peranimal.Washes in 0.1 M PBS, pH 7.4, were performed between each

step. Free-floating brain sections from mice perfused withAF750-Ri7 were blocked for 1 h in a PBS solution containing5% horse serum (Invitrogen) and 0.2% Triton X-100. Sectionswere then incubated overnight at 4 °C with primary antibodiesin the blocking solution: goat anti-type IV collagen (Coll IV,1:500; Millipore Bioscience Research Reagents, Temecula, CA)and mouse antineuronal nuclei (NeuN, 1:1000; MilliporeBioscience Research Reagents). After incubation with primaryantibodies, slices were exposed to AF488 donkey anti-goat andAF568 anti-mouse secondary antibodies (1:1000; Invitrogen).Finally, slices were mounted onto SuperFrost Plus slides(Thermo Fisher Scientific, Waltham, MA, USA) and placedunder coverslips with Mowiol mounting medium. Differentappropriate filters were used for Ri7 (excitation, 710/75 nm;emission, 810/90 nm), Coll IV (excitation, 480/30 nm;emission, 535/40 nm), and NeuN (excitation, 540/40 nm;



Figure 2. Accumulation of AF488-Ri7 in murine neuroblastoma (N2A) and brain endothelial (bEnd5) cells. (A) N2A and (B) bEnd5 cells wereseeded in 96-well plates containing DMEM with 10% FBS. One day later, they were exposed to AF488-Ri7 or AF488-IgG (0.4, 2, or 10 nM) for 5,15, or 45 min. Cells were then washed once with cold PBS and 0.025% trypan blue to quench noninternalized fluorescence (n = 3 experiments intriplicate). Fluorescence was measured with a Synergy HT plate reader (485/528 filters). (C, E, G) N2A and (D, F, H) bEnd5 cells were cultured asmentioned above and exposed for 30 min to 10 nM AF488-Ri7 at either 4 °C (C, D) or 37 °C (E, F) or to 10 nM AF488-IgG at 37 °C (G, H). Cellswere then washed with cold PBS, fixed with 4% paraformaldehyde, and stained with DAPI. AF488-Ri7 can be seen in green and DAPI-stained nucleiin blue. Vesicular-like staining is visible in (E) and (F). The images represent compressed, deconvolved Z-stacks. Data are shown as means ± SEM.*, p < 0.05; **, p < 0.01; ***, p < 0.001. Abbreviations: AF488, Alexa Fluor 488; Ri7, Ri7.217.1.4; IgG, control rat IgG2a; mAb, monoclonalantibody; RFU, relative fluorescence units.



emission, 600/50 nm). Photomicrographs were acquired usingSimple PCI version 5.0 software (Hamamatsu, Sewickley, PA,USA) linked to a Nikon Eclipse 90i microscope (NikonInstruments, Toronto, ON, Canada).Capillary Depletion. A capillary depletion technique was

used to isolate brain microvessels by density gradientcentrifugation as adapted from a previous study.32 Four maleBALB/c mice were anesthetized deeply and then perfusedtranscardially with 50 mL of ice-cold 0.1 M PBS. Next, brainswere collected and transferred into ice-cold 0.1 M PBS, wherecerebellum, meninges, brain stem, and large superficial bloodvessels were removed. The brain was gently homogenized inice-cold DMEM containing 10% FBS using Potter homoge-nizer, and a fraction of this homogenate was kept and identifiedas the total brain homogenate fraction (H). The remaininghomogenate was centrifuged at 500g for 10 min at 4 °C. Thesupernatant was excluded, and the pellet was homogenized in 5mL of ice-cold DMEM containing 25% bovine serum albumin(BSA) and centrifuged at 1500g for 20 min at 4 °C. Thesupernatant was excluded, and the remaining pellet was gentlyhomogenized in 1 mL of ice-cold DMEM containing 10% FBS.The homogenate was filtered through a 60 μm filter toeliminate larger vessels, and the filtrate was centrifuged at12000g for 45 min at 4 °C. The pellet containing themicrovessels was washed in ice-cold 0.1 M PBS and centrifugedagain at 12000g for 20 min at 4 °C. The supernatant wasdiscarded, and the pellet was identified as the capillary fraction(C). The two fractions (H and C) were stored at −80 °C untilprocessed for Western blot analysis.Tissue Processing. N2A and bEnd5 cells and the H and C

brain fractions were homogenized in 8 volumes of lysis buffer(150 mM NaCl, 10 mM NaH2PO4, 1% Triton X-100, 0.5%SDS, and 0.5% deoxycholate) containing Complete proteaseinhibitors, 10 mg/mL pepstatin A, and phosphatase inhibitors(1 mM sodium pyrophosphate, 50 mM sodium fluoride). Theobtained suspension was sonicated briefly (3 × 10 s) andcentrifuged at 100000g for 20 min at 4 °C. The proteinconcentration was determined with the BCA assay in thesupernatant, which was then stored at −80 °C until Westernblotting.Western Blot Analysis. Equal amounts of proteins were

added to Laemmli’s loading buffer, heated to 95 °C for 5 minbefore loading, and subjected to sodium dodecyl sulfate (SDS)polyacrylamide gel electrophoresis. Proteins were electro-blotted onto PVDF membranes (Immobilon, Millipore, MA,USA) before blocking in 5% nonfat dry milk, 0.5% BSA, and0.1% Tween 20 in 0.1 M PBS for 1 h at room temperature. Themembrane was washed three times for 10 min in 0.1% PBScontaining 0.1% Tween 20. The membrane was then incubatedovernight at 4 °C with primary antibodies diluted in 0.1 M PBScontaining 0.1% Tween 20, 5% nonfat dry milk, and 0.5% BSA.The dilutions of primary antibodies were 1:5000 for mouseanti-TfR (Abmart) and 1:10000 for mouse anti-GAPDH(ABM, Richmond, BC, Canada). The next day the membranewas washed three times for 10 min in 0.1% PBS containing0.1% Tween 20 and then incubated for 1 h at roomtemperature with horseradish peroxidase-labeled goat anti-mouse (Jackson, West Grove, PA, USA) diluted 1:100000 in0.1 M PBS containing 0.1% Tween 20 and 1% BSA. Themembrane was again washed three times for 10 min in 0.1%PBS containing 0.1% Tween 20 and then probed withchemiluminescence reagents (Lumiglo Reserve, KPL, Gaithers-burg, MD, USA). Immunoblots were analyzed with a KODAK

Imaging Station 4000 MM Digital Imaging System (MolecularImaging Software version 4.0.5f7, Carestream Health, Ro-chester, NY, USA).

Data and Statistical Analysis. Data are shown as mean ±standard error of the mean (SEM). Student’s unpaired t testwas used to identify significant differences between two groupswhen appropriate. Statistical differences between three groupswere determined using the appropriate one-way analysis ofvariance (ANOVA) and posthoc tests for comparison betweengroups. All of the tests were two-tailed, and statisticalsignificance was set as follows: *, P < 0.05; **, P < 0.01;***, P < 0.001.

■ RESULTSCellular Uptake of AF488-Ri7 in N2A and bEnd5 Cells.

To first confirm the specific targeting and uptake of the Ri7mAb by mouse TfR, AF488-Ri7 accumulation was evaluated invitro using two mouse-TfR-expressing cell lines, N2A andbEnd5. As shown in Figure 2A,B, a time- and concentration-dependent (0.4 to 10 nM) accumulation of AF488-Ri7 wasobserved in both cell lines. The observed curves wereconsistent with a TfR-dependent saturable mechanism. Toverify whether the cellular accumulation of AF488-Ri7 was dueto endocytosis rather than receptor binding, we performeduptake experiments at 4 and 37 °C in both cell lines. At 4 °C,the signal was limited to the cell surface (Figure 2C,D),suggesting binding of Ri7 to TfR without endocytosis. On theother hand, vesicular-like staining indicated that internalizationof AF488-Ri7 occurred at 37 °C (Figure 2E,F), suggesting thatendocytic mechanisms were activated after Ri7 binding to TfRin both N2A and bEnd5 cells. In contrast, uptake of thenegative control AF488-IgG remained negligible comparedwith AF488-Ri7 (10 nM) in the two cell lines after a 30 minincubation at 37 °C (Figure 2G,H).

Competition of Unlabeled Ri7 with Cellular Uptake ofAF488-Ri7 by N2A and bEnd5 Cells. To establish whetherRi7 uptake into N2A and bEnd5 cells was a saturable processinvolving the mouse TfR, we carried out a competition studyusing increasing concentrations of unlabeled Ri7. In theseexperiments, a 15 min pre-exposition with three escalatingconcentrations of unlabeled Ri7 (10, 50, and 250 nM) wasperformed to compete with AF488-Ri7 (5 nM) for bindingsites on TfR. As demonstrated in Figure 3A,B, unlabeled Ri7was highly potent to prevent binding of AF488-Ri7 in the twocell lines, decreasing its cellular accumulation by 75% at 250nM. As expected, addition of unlabeled IgG (10 to 250 nM) didnot compete with AF488-Ri7 internalization. To document thelevel of TfR expression in the two cell lines, a Western blotanalysis was performed. Relative to GAPDH, a higherconcentration of mouse TfR was found in N2A cells than inbEnd5 cells (Figure 3 C).

Accumulation of Fluorescent Anti-TfR mAb (AF750-Ri7) in the Brain after in Situ Brain Perfusion. Because ofautofluorescence of brain tissue in the visible spectrum, theproposed experimental paradigm was developed using an NIR-fluorescent dye. To quantify Ri7 brain uptake, we thus coupledRi7 with AF750.22 The fluorescence in brain homogenates wasfirst measured after the perfusion of 100 or 200 μg of AF750-mAb (Ri7 or control IgG), equivalent to concentrations of∼220 and ∼440 nM, respectively, in the perfusate. The degreesof fluorolabeling per microgram of proteins for Ri7 and IgGwere comparable. The fluorescence retrieved in brainhomogenates from animals perfused with either 100 or 200



μg of AF750-Ri7 was significantly higher than that in animalsperfused with control AF750-IgG (Figure 4A,B) indicating asignificant accumulation of Ri7 vectors in the brain. On thebasis of our in vitro data indicating that the signal associatedwith the IgG control was nonspecific, distribution volume (VD)values obtained in mice perfused with control AF750-IgG weresubtracted from the data obtained with AF750-Ri7 todetermine more accurately the VD of Ri7. From these values,the brain uptake coefficient (Clup) values of AF750-Ri7 were

estimated to be 0.27 and 0.26 μL g−1 s−1, for perfused doses of100 and 200 μg, respectively.We verified the physical integrity of the BBB to ensure that

Ri7 accumulation in the brain was not due to leakage of theBBB during perfusion. To that aim, we measured the brainvascular volume calculated by co-perfusing [14C]sucrose (Vvasc,μL g−1), a compound that does not cross the BBB and remainsin the vascular space. All groups displayed Vvasc values of

Figure 3. Decrease of accumulation of AF488-Ri7 in N2A and bEnd5cells and higher expression of murine transferrin receptor (TfR) inN2A cells compared with bEnd5 cells. (A, B) N2A and bEnd5 cellswere seeded in 96-well plates containing DMEM with 10% FBS. Oneday later, pretreatment for 15 min with unlabeled Ri7 or unlabeled IgG(0, 10, 50, or 250 nM) competed against the binding of AF488-Ri7.After further exposure to 5 nM AF488-Ri7 for 15 min, the cells werewashed once with cold PBS and 0.025% trypan blue to quenchnoninternalized fluorescence (n = 2 experiments in triplicate).Fluorescence was measured with a Synergy HT plate reader (485/528 filters). (C) TfR content was measured by Western blotting incultured N2A and bEnd5 cells and normalized with GAPDH. n = 4dishes per cell line, 48 μg of proteins per lane. Data are shown asmeans ± SEM. Control values (without unlabeled mAb) were adjustedto 100%. *, p < 0.05; ***, p < 0.001. Abbreviations: AF488, AlexaFluor 488; Ri7, Ri7.217.1.4; IgG, control rat IgG2a (2A3); RFU,relative fluorescence units; mAb, monoclonal antibody.

Figure 4. Brain uptake of mAb targeting the TfR (AF750-Ri7) whenthe physical integrity of the BBB is maintained. (A, B) Shown areexamples of pseudocolored brain homogenates in a 96-well plate asdetected with 720/790 filters along with brain distribution volumes VD(μL g−1) after in situ brain perfusion of (A) 100 or (B) 200 μg ofAF750-mAb (n = 9−10 per group). The brain uptake coefficientClupRi7 (μL g−1 s−1) was calculated as (VD,Ri7 − VD,IgG)/perfusion time.Black column, group perfused with AF750-IgG; white column, groupperfused with AF750-Ri7. Data were normalized by subtraction ofbackground values (Bckd) obtained from mice perfused with perfusionbuffer alone. (C) The BBB is intact after the perfusion of Ri7.Distribution volumes of the vascular space marker [14C]sucrose weremeasured by in situ brain perfusion. The vascular volume wasunchanged for the group perfused with Ri7 relative to the groupperfused with buffer alone (n = 6). Black column, group perfused withbuffer; white column, group perfused with Ri7. Data are shown asmeans ± SEM. **, p < 0.01; ***, p < 0.001. Abbreviations: AF750,Alexa Fluor 750; Ri7, Ri7.217.1.4; IgG, control rat IgG2a (2A3); mAb,monoclonal antibody; TfR, transferrin receptor.



approximately 15 μL g−1 (Figure 4C), consistent with an intactBBB. These data further confirm that the ISBP technique usedhere with AF750-Ri7 does not impair the integrity of the BBB.Linearity of the Brain Uptake of AF750-Ri7 over Time.

To determine the appropriate perfusion time and to ensure thatno metabolic changes occurred during the perfusion, it wasimportant to investigate the linearity of AF750-Ri7 brain uptakeover time. The time course of AF750-Ri7 brain uptake ispresented in Figure 5. The distribution volume of AF750-Ri7

was linear over all of the perfusion time points (30, 60, and 120s), confirming the stability of AF750-Ri7 and indicating that thetransport rate remained unchanged over time and that the brainuptake was unidirectional33 from blood to brain over 120 s ofbrain perfusion. A perfusion time of 60 s was thus chosen forsubsequent single-time-point experiments.Saturation of AF750-Ri7 Brain Uptake. To determine

whether the brain uptake of AF750-Ri7 is saturable, increasingquantities of unlabeled Ri7 were co-perfused with a constantdose of AF750-Ri7. The brain uptake coefficient (Clup) ofAF750-Ri7 was reduced significantly with co-perfusion of 400μg or 900 μg of unlabeled Ri7 (Figure 6), down to the residualcontrol values of AF750-IgG. To further confirm this result andto estimate the saturation point, increasing doses (50−1000μg) of AF750-Ri7 were used next. The quantity of AF750-Ri7per gram of brain tissue increased until a plateau was reached atapproximately 300 μg (Figure 7A). The saturating perfuseddose associated with 50% TfR occupancy (SD50) correspondsto half of the graphically determined maximum receptorpopulation (Bmax). After subtraction of control values (non-specific binding of AF750-IgG), curve fitting analysis resulted inan SD50 of approximately 226 μg (∼500 nM). Accordingly, theClup was reduced progressively from 1.67 ± 0.15 μL g−1 s−1 forperfusion of 50 μg of AF750-Ri7 down to a plateau at ∼0.20 μLg−1 s−1 with a perfused dose above 400−500 μg (Figure 7B).Confinement of AF750-Ri7 in BCECs. Fluorescence

quantification from brain homogenates does not discriminateparenchymal penetration from retention in BCECs. To verifywhether the AF750-Ri7 signal originated from BCECs and/or

the cerebral parenchyma, a fluorescence microscopy analysiswas performed on sections of brain perfused with 100 μg ofAF750-Ri7 in situ. As shown in Figure 8A, fluorescence fromAF750-Ri7 was restricted to cerebral microvessels. Immunolab-eling of anti-collagen IV, a basal lamina marker, confirmed thecolocalization and accumulation of AF750-Ri7 in BCECs.Colocalization analyses with the neuronal marker NeuN werealso performed, and no colocalization of AF750-Ri7 andneurons was observed. Finally, using the capillary depletiontechnique followed by Western blotting, we found that the TfRwas concentrated in capillaries, compared with total brainhomogenates from four mice (Figure 8B).

■ DISCUSSIONThe results of this series of studies are consistent with thefollowing conclusions. First, in vitro determination of AF488-Ri7 uptake in mouse N2A and bEnd5 cells showed that theuptake mechanism involves a site-specific and saturable processthrough the TfR. An IgG control did not compete with thecellular accumulation of AF488-Ri7. Second, ISBP experimentsdemonstrated the accumulation of AF750-Ri7 into the brain,confirming the feasibility of combining NIR-fluorescentdetection with ISBP. Third, we showed that the brain uptakeof AF750-Ri7 is a saturable process, as the Clup valuesplummeted after the total perfused dose was increased. Thesaturating dose (SD50) of AF750-Ri7 was estimated to be 226μg, equivalent to a concentration of ∼500 nM in the perfusedbuffer. Fourth, fluorescence microscopy analysis demonstratedthat BCECs retained most of the AF750-Ri7 after ISBP.Quantitative assessment of the brain uptake of large moleculesis an essential prerequisite for their development into clinicalapplications for central nervous system (CNS) diseases. Sincefluorolabeling can be readily performed in large proteins, thepresent fluorescence-based ISBP technique stands as a methodof choice for routine assessment of brain uptake of large

Figure 5. Time course of 100 μg of mAb targeting the TfR (AF750-Ri7) uptaken by the right cerebral hemisphere after in situ brainperfusion, expressed as distribution volume VD (μL g−1). Regressionanalysis of individual data gave r2 = 0.997. Data were normalized bysubtraction of background values (Bckd) obtained from mice perfusedfor 60 s with perfusion buffer alone. Data are shown as means ± SEMof nine mice per data point. Also shown is an example ofpseudocolored brain homogenates in a 96-well plate as detectedwith 720/790 filters. Abbreviations: AF750, Alexa Fluor 750; Ri7,Ri7.217.1.4; mAb, monoclonal antibody; TfR, transferrin receptor.

Figure 6. Decrease in brain uptake of mAb targeting the TfR (AF750-Ri7). The brain uptake coefficient, Clup (μL g−1 s−1), of 100 μg offluorolabeled Ri7 (AF750-Ri7) was determined after in situ brainperfusion. Co-perfusion with 400 or 900 μg of unlabeled Ri7 during 60s impeded AF750-Ri7 uptake. Black column, AF750-Ri7; gray column,AF750-Ri7 + 400 μg of unlabeled Ri7; white column, AF750-Ri7 +900 μg of unlabeled Ri7. Data were normalized by subtraction ofbackground values (Bckd) obtained from mice perfused with perfusionbuffer alone. Data are shown as means ± SEM of eight or nine miceper group. ***, p < 0.001. Also shown is an example of pseudocoloredbrain homogenates in a 96-well plate as detected with 720/790 filters.Abbreviations: AF750, Alexa Fluor 750; Ri7, Ri7.217.1.4; mAb,monoclonal antibody; TfR, transferrin receptor.



biotherapeutics, provided that the fluorolabeling does notchange the properties of the targeted biotherapeutic.Because of its large expression on microvessels arborizing

throughout the brain and its capacity to ferry iron-bound Tfinto the brain, the TfR has attracted a lot of interest for braindrug delivery.17,23,34 Our data confirmed that TfR expression isrelatively higher in brain capillaries compared with other braincompartments. Vectors targeting the TfR include the murineOX-26 mAb for brain transport through the rat BBB20,35 andrat mAbs targeting the murine TfR, such as 8D3 and Ri7.21,36

These vectors have been conjugated to therapeutic peptide/protein or liposomes for preclinical assays in various models ofhuman diseases.16,25,37−40 Strong evidence of brain drugdelivery has been gathered with OX-26 in the rat using varioustechniques,25,41−43 with the exception of reports usinghistochemistry and brain capillary depletion techniques.44,45

In the mouse, evidence of transport of Ri7 or 8D3 anti-TfRantibodies across the BBB21,36 has not been confirmed by aconfocal microscopy report in which intravenously adminis-tered AF647-Ri7 remained confined to BCECs in the mouse.22

The disparity between these results may reflect differences in

experimental approaches, such as the method of terminalperfusion or the relative dilution of the anti-TfR mAb in a largepostendothelial volume, hindering its detection in brainparenchyma cells. It should be kept in mind that the presentISBP method alone is not sufficient to conclude that fulltransport across the BBB occurred. Complementary micro-scopic data are necessary to assess the cellular brain distributionof a perfused fluorolabeled compound. Although theimmunohistofluorescence data provided here corroborate theview that the total increase of Ri7 mAbs in the brain is mostlymediated by binding and endocytosis into BCECs, we cannotrule out the possibility that undetectable amounts of the mAbcrossed BCECs and reached the brain parenchyma.

Figure 7. Saturation of brain uptake of mAb targeting the TfR(AF750-Ri7) after perfusion of increasing doses. (A) Cerebralconcentrations (μg g−1) of AF750-Ri7 (n = 3−4) and (B) brainuptake coefficients Clup (μL g−1 s−1) (n = 3−4) were determinedusing in situ brain perfusion after normalization by subtracting theautofluorescence background obtained from mice perfused withperfusion buffer alone. Data are shown as means ± SEM. Alsoshown is an example of pseudocolored brain homogenates in a 96-wellplate as detected with 720/790 filters. Abbreviations: SD50, perfuseddose associated with 50% TfR occupancy; Bmax, maximum transferrinreceptor population; AF750, Alexa Fluor 750; Ri7, Ri7.217.1.4; mAb,monoclonal antibody; TfR, transferrin receptor.

Figure 8. (A) Brain distribution of AF750-Ri7 after in situ brainperfusion of mAb targeting the TfR (AF750-Ri7). Mice were perfusedwith 100 μg of AF750-Ri7 targeting murine TfR during 60 s. Theexamples shown are representative of results obtained from threeanimals. Microscopy analysis showed that AF750-Ri7 (white)colocalized with the basal lamina marker collagen IV (green) oncerebral microvessels, whereas no signal was found in neuronsimmunostained with NeuN (red). Scale bars represent 50 μm.Abbreviations: AF750, Alexa Fluor 750; Ri7, Ri7.217.1.4; mAb,monoclonal antibody; TfR, transferrin receptor. (B) Western blotswere performed to measure TfR expression in brain homogenates (H)and isolated capillaries (C) fractions from four mice. Data are shownas means ± SEM. **, p < 0.01; ***, p < 0.001.



The specificity of the Ri7 mAb for the TfR has been furtherconfirmed in the present study on the basis of our in vitro andin vivo data. The Ri7 mAb was initially chosen in this studybecause of its specific uptake by the brain compared with 8D3,which is also significantly taken up by the liver and kidney.21 Invitro experiments confirmed that accumulation of AF488-Ri7 inthe two cell lines N2A and bEnd5 was inhibited by increasingconcentrations of unlabeled Ri7, with no interference from theisotype control (IgG), indicating that the cellular uptake of Ri7requires specific binding with its receptor. This result wasconsistent with in vitro work using micelles complexed to Tf46

or transfection experiments with liposomes conjugated to 8D3or Ri7.47 Endocytosis is an active process that is slowed atreduced temperatures.48−50 Our findings that the vesicular-likesignal associated with AF488-Ri7 accumulates in cells at 37 °Cbut is confined to the cell surface at 4 °C argues in favor of anendocytotic process. In vivo competition, as assessed with ISBP,further supports specific binding of Ri7 to the luminallyexposed TfR in brain microvessels.From a quantitative perspective, the brain uptake rate of Ri7

reported here remained relatively low, at least at highconcentrations. Perfusion of 100 μg of AF750-Ri7 led to aClup of 0.27 μL g−1 s−1 after correction for nonspecific signal.The relevance of such a correction is supported by in vitro andin vivo data showing that IgG binding was nonspecific. Incontrast, the uptake of AF750-Ri7 reached 1.47 ± 0.15 μL g−1

s−1 when 50 μg was perfused (∼110 nM), consistent with asaturable mechanism. Nevertheless, our data stand in relativeagreement with the results of previous studies using comparableperfusion techniques after conversion into similar units. In therat brain, the Clup of the radiolabeled anti-TfR vector OX-26was approximately 0.07 μL g−1 s−1 using transcardialperfusion,44 whereas the Clup of OX-26 and Tf itself rangedbetween 0.08 and 0.18 μL g−1 s−1 using ISBP.41 Still, thesevalues remain relatively low compared with those for fullydiffusible compounds such as diazepam and fatty acids, whichdisplay Clup values of up to 40 μL g−1 s−1.11,51 In contrast, aroutine-use CNS drug such as morphine, which is partiallypumped out of the brain by the ABC-transporter P-glycoprotein (Pgp), displays a Clup of 0.3 μL g−1 s−1.52

Thus, the measured brain uptake of Ri7 is consistent withprevious reports and stays within an acceptable range for drugdelivery purposes.Our present in vitro and in vivo data argue for a saturable

mechanism underlying AF750-Ri7 accumulation in brain cells.The transport rate of AF750-Ri7 became negligible atconcentrations over 500 nM, evidencing complete saturationof TfR-mediated transport in vivo. Such an interpretation is inline with previous data on saturable transport of perfused Tf orOX-26 in the rat brain.41,53 Similarly, uptake of radiolabeled8D3 or AF750-Ri7 was fully inhibited by intravenouscoinjection of a saturating dose of unlabeled mAb.21,22 OurISBP approach allowed the calculation of a saturating perfuseddose (SD50), which was estimated to be 226 μg, correspondingto a concentration of ∼500 nM; this value is not very far fromthe dissociation constant of human Tf to the TfR, which hasbeen estimated to be slightly above 700 nM.54 Still, thesenumbers remain below the physiological concentration of Tf(25 μM or 2000 μg/mL),55 suggesting that TfR at the BBB isfully saturated under physiological conditions.34 Nevertheless,systemically administered fluorolabeled/radiolabeled Ri7 orRi7-loaded nanocarriers in rodent accumulate in thebrain,21,22,56 supporting the hypothesis that the binding site

of Tf is spatially different from that of TfR-targeting mAbs.57,58

Another intriguing possibility raised recently is that the BCECconfinement of AF750-Ri7 observed here results from a strongaffinity with the TfR. According to this hypothesis, only TfRmAbs with low affinity are ferried through BCECs andsubsequently released across the BBB.59 Here, an SD50 of 500nM is consistent with a relatively high binding affinity of Ri7 forthe TfR, possibly explaining why AF750-Ri7 remains confinedin BCECs after perfusion.22,59,60

The use of fluorescence-coupled ISBP as shown herecircumvents the safety issues associated with radioactivity.Exposure to external radiation, even for medical purposes,increases the risk for cancer, and the American NationalResearch Council concluded that no exposure level can beassumed to be truly safe.61 Despite drawbacks such asunavoidable background autofluorescence in tissues, limitedstability, and light sensitivity, fluorescence detection is ananalytical technique that is relatively simple to implement.Conjugation between fluorophores and amine or thiol groupson a protein can be done in the laboratory by application ofroutine-use protocols. Operating costs are also modestcompared with those for radioactivity, allowing the use oflarger quantities of fluorolabeled molecules without particularsafety provisions. Finally, coupling of vectors to differentfluorophores offers the possibility to detect several wavelengthssimultaneously and combine homogenate quantification withmicroscopy.

■ CONCLUSION

In summary, this study confirms the accumulation of the anti-TfR mAb AF750-Ri7 into BCECs of mice within minutes afterintracarotid perfusion. Taken together, the results areconsistent with TfR-specific unidirectional brain uptake ofAF750-Ri7 through a fully saturable mechanism. The presentwork thus offers a novel application of ISBP to providequantitative data on the brain transport of large fluorolabeledmolecules, avoiding the use of radioactivity.

■ AUTHOR INFORMATION

Corresponding Author*Address: Neurosciences Axis, Centre de recherche du CHUde Quebec, 2705 Laurier Blvd., Room T2-05, Quebec, QC G1V4G2, Canada. Tel: (418) 656-4141 ext. 48697. Fax: (418) 654-2761. E-mail: [email protected].

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTS

We thank Melissa Ouellet for the illustrative schema of the insitu brain perfusion technique and Eric Beliveau for the Z-stackdeconvolution. Financial support was provided by the CanadianInstitute of Health Research (CIHR, MOP84251) and theCanada Foundation for Innovation.

■ REFERENCES(1) Pardridge, W. M. Alzheimer’s disease drug development and theproblem of the blood−brain barrier. Alzheimer’s Dementia 2009, 5,427−432.(2) Vlieghe, P.; Khrestchatisky, M. Medicinal chemistry basedapproaches and nanotechnology-based systems to improve CNS drugtargeting and delivery. Med. Res. Rev. 2013, 33, 457−516.



mailto:[email protected]

(3) Pangalos, M. N.; Schechter, L. E.; Hurko, O. Drug developmentfor CNS disorders: Strategies for balancing risk and reducing attrition.Nat. Rev. Drug Discovery 2007, 6, 521−532.(4) Weiss, N.; Miller, F.; Cazaubon, S.; Couraud, P. O. The blood−brain barrier in brain homeostasis and neurological diseases. Biochim.Biophys. Acta 2009, 1788, 842−857.(5) Sa-Pereira, I.; Brites, D.; Brito, M. A. Neurovascular unit: A focuson pericytes. Mol. Neurobiol. 2012, 45, 327−347.(6) Jeffrey, P.; Summerfield, S. Assessment of the blood−brainbarrier in CNS drug discovery. Neurobiol. Dis. 2010, 37, 33−37.(7) Mensch, J.; Oyarzabal, J.; Mackie, C.; Augustijns, P. In vivo, invitro and in silico methods for small molecule transfer across the BBB. J.Pharm. Sci. 2009, 98, 4429−4468.(8) Dash, A. K.; Elmquist, W. F. Separation methods that are capableof revealing blood−brain barrier permeability. J. Chromatogr., B 2003,797, 241−254.(9) Hawkins, B. T.; Egleton, R. D. Pathophysiology of the blood−brain barrier: Animal models and methods. Curr. Top. Dev. Biol. 2008,80, 277−309.(10) Takasato, Y.; Rapoport, S. I.; Smith, Q. R. An in situ brainperfusion technique to study cerebrovascular transport in the rat. Am.J. Physiol. 1984, 247, H484−H493.(11) Dagenais, C.; Rousselle, C.; Pollack, G. M.; Scherrmann, J. M.Development of an in situ mouse brain perfusion model and itsapplication to mdr1a P-glycoprotein-deficient mice. J. Cereb. BloodFlow Metab. 2000, 381−386.(12) Bickel, U. How to measure drug transport across the blood−brain barrier. NeuroRx 2005, 2, 15−26.(13) Smith, Q. R. A review of blood−brain barrier transporttechniques. Methods Mol. Med. 2003, 89, 193−208.(14) Nicolazzo, J. A.; Charman, S. A.; Charman, W. N. Methods toassess drug permeability across the blood−brain barrier. J. Pharm.Pharmacol. 2006, 58, 281−293.(15) Egawa, T.; Hanaoka, K.; Koide, Y.; Ujita, S.; Takahashi, N.;Ikegaya, Y.; Matsuki, N.; Terai, T.; Ueno, T.; Komatsu, T.; Nagano, T.Development of a far-red to near-infrared fluorescence probe forcalcium ion and its application to multicolor neuronal imaging. J. Am.Chem. Soc. 2011, 133, 14157−14159.(16) Pardridge, W. M.; Boado, R. J. Reengineering biopharmaceut-icals for targeted delivery across the blood−brain barrier. MethodsEnzymol. 2012, 503, 269−292.(17) Dennis, M. S.; Watts, R. J. Transferrin antibodies into the brain.Neuropsychopharmacology 2012, 37, 302−303.(18) Jefferies, W. A.; Brandon, M. R.; Hunt, S. V.; Williams, A. F.;Gatter, K. C.; Mason, D. Y. Transferrin receptor on endothelium ofbrain capillaries. Nature 1984, 312, 162−163.(19) Boado, R. J.; Zhang, Y.; Wang, Y.; Pardridge, W. M. Engineeringand expression of a chimeric transferrin receptor monoclonal antibodyfor blood−brain barrier delivery in the mouse. Biotechnol. Bioeng. 2009,102, 1251−1258.(20) Friden, P. M.; Walus, L. R.; Musso, G. F.; Taylor, M. A.;Malfroy, B.; Starzyk, R. M. Anti-transferrin receptor antibody andantibody−drug conjugates cross the blood−brain barrier. Proc. Natl.Acad. Sci. U.S.A. 1991, 88, 4771−4775.(21) Lee, H. J.; Engelhardt, B.; Lesley, J.; Bickel, U.; Pardridge, W. M.Targeting rat anti-mouse transferrin receptor monoclonal antibodiesthrough blood−brain barrier in mouse. J. Pharmacol. Exp. Ther. 2000,292, 1048−1052.(22) Paris-Robidas, S.; Emond, V.; Tremblay, C.; Soulet, D.; Calon,F. In vivo labeling of brain capillary endothelial cells after intravenousinjection of monoclonal antibodies targeting the transferrin receptor.Mol. Pharmacol. 2011, 80, 32−39.(23) Qian, Z. M.; Li, H.; Sun, H.; Ho, K. Targeted drug delivery viathe transferrin receptor-mediated endocytosis pathway. Pharmacol.Rev. 2002, 54, 561−587.(24) Watts, R. J.; Dennis, M. S. Bispecific antibodies for delivery intothe brain. Curr. Opin. Chem. Biol. 2013, 17, 393−399.(25) Zhang, Y.; Calon, F.; Zhu, C.; Boado, R. J.; Pardridge, W. M.Intravenous nonviral gene therapy causes normalization of striatal

tyrosine hydroxylase and reversal of motor impairment in experimentalParkinsonism. Hum. Gene Ther. 2003, 14, 1−12.(26) Olmsted, J. B.; Carlson, K.; Klebe, R.; Ruddle, F.; Rosenbaum, J.Isolation of microtubule protein from cultured mouse neuroblastomacells. Proc. Natl. Acad. Sci. U.S.A. 1970, 65, 129−136.(27) Williams, R. L.; Risau, W.; Zerwes, H. G.; Drexler, H.; Aguzzi,A.; Wagner, E. F. Endothelioma cells expressing the polyoma middle Toncogene induce hemangiomas by host cell recruitment. Cell 1989, 57,1053−1063.(28) Wagner, E. F.; Risau, W. Oncogenes in the study of endothelialcell growth and differentiation. Semin. Cancer Biol. 1994, 5, 137−145.(29) Rohnelt, R. K.; Hoch, G.; Reiss, Y.; Engelhardt, B.Immunosurveillance modelled in vitro: Naive and memory T cellsspontaneously migrate across unstimulated microvascular endothe-lium. Int. Immunol. 1997, 9, 435−450.(30) Ouellet, M.; Emond, V.; Chen, C. T.; Julien, C.; Bourasset, F.;Oddo, S.; LaFerla, F.; Bazinet, R. P.; Calon, F. Diffusion ofdocosahexaenoic and eicosapentaenoic acids through the blood−brain barrier: An in situ cerebral perfusion study. Neurochem. Int. 2009,55, 476−482.(31) Lesley, J. F.; Schulte, R. J. Selection of cell lines resistant to anti-transferrin receptor antibody: Evidence for a mutation in transferrinreceptor. Mol. Cell. Biol. 1984, 4, 1675−1681.(32) Do, T. M.; Noel-Hudson, M. S.; Ribes, S.; Besengez, C.;Smirnova, M.; Cisternino, S.; Buyse, M.; Calon, F.; Chimini, G.;Chacun, H.; Scherrmann, J. M.; Farinotti, R.; Bourasset, F. ABCG2-and ABCG4-mediated efflux of amyloid-β peptide 1−40 at the mouseblood−brain barrier. J. Alzheimer’s Dis. 2012, 30, 155−166.(33) Cattelotte, J.; Andre, P.; Ouellet, M.; Bourasset, F.; Scherrmann,J. M.; Cisternino, S. In situ mouse carotid perfusion model: Glucoseand cholesterol transport in the eye and brain. J. Cereb. Blood FlowMetab. 2008, 1449−1459.(34) Pardridge, W. M. Drug transport across the blood−brain barrier.J. Cereb. Blood Flow Metab. 2012, 1959−1972.(35) Broadwell, R. D.; Baker-Cairns, B. J.; Friden, P. M.; Oliver, C.;Villegas, J. C. Transcytosis of protein through the mammalian cerebralepithelium and endothelium. III. Receptor-mediated transcytosisthrough the blood−brain barrier of bloodborne transferrin andantibody against the transferrin receptor. Exp. Neurol. 1996, 142,47−65.(36) Shi, N.; Zhang, Y.; Zhu, C.; Boado, R. J.; Pardridge, W. M.Brain-specific expression of an exogenous gene after i.v. administration.Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 12754−12759.(37) Fu, A.; Hui, E. K.; Lu, J. Z.; Boado, R. J.; Pardridge, W. M.Neuroprotection in stroke in the mouse with intravenouserythropoietin−Trojan horse fusion protein. Brain Res. 2011, 1369,203−207.(38) Zhou, Q. H.; Hui, E. K.; Lu, J. Z.; Boado, R. J.; Pardridge, W. M.Brain penetrating IgG−erythropoietin fusion protein is neuro-protective following intravenous treatment in Parkinson’s disease inthe mouse. Brain Res. 2011, 1382, 315−320.(39) Zhou, Q. H.; Fu, A.; Boado, R. J.; Hui, E. K.; Lu, J. Z.; Pardridge,W. M. Receptor-mediated abeta amyloid antibody targeting toAlzheimer’s disease mouse brain. Mol. Pharmaceutics 2011, 8, 280−285.(40) Lee, H. J.; Zhang, Y.; Zhu, C.; Duff, K.; Pardridge, W. M.Imaging brain amyloid of Alzheimer disease in vivo in transgenic micewith an Abeta peptide radiopharmaceutical. J. Cereb. Blood Flow Metab.2002, 223−231.(41) Skarlatos, S.; Yoshikawa, T.; Pardridge, W. M. Transport of[125I]transferrin through the rat blood−brain barrier. Brain Res. 1995,683, 164−171.(42) Cerletti, A.; Drewe, J.; Fricker, G.; Eberle, A. N.; Huwyler, J.Endocytosis and transcytosis of an immunoliposome-based brain drugdelivery system. J. Drug Targeting 2000, 8, 435−446.(43) Bickel, U.; Kang, Y. S.; Yoshikawa, T.; Pardridge, W. M. In vivodemonstration of subcellular localization of anti-transferrin receptormonoclonal antibody−colloidal gold conjugate in brain capillaryendothelium. J. Histochem. Cytochem. 1994, 42, 1493−1497.



(44) Gosk, S.; Vermehren, C.; Storm, G.; Moos, T. Targeting anti-transferrin receptor antibody (OX26) and OX26-conjugated lip-osomes to brain capillary endothelial cells using in situ perfusion. J.Cereb. Blood Flow Metab. 2004, 1193−1204.(45) Moos, T.; Morgan, E. H. Restricted transport of anti-transferrinreceptor antibody (OX26) through the blood−brain barrier in the rat.J. Neurochem. 2001, 79, 119−129.(46) Zhang, P.; Hu, L.; Yin, Q.; Zhang, Z.; Feng, L.; Li, Y.Transferrin-conjugated polyphosphoester hybrid micelle loadingpaclitaxel for brain-targeting delivery: Synthesis, preparation and invivo evaluation. J. Controlled Release 2012, 159, 429−434.(47) Rivest, V.; Phivilay, A.; Julien, C.; Belanger, S.; Tremblay, C.;Emond, V.; Calon, F. Novel liposomal formulation for targeted genedelivery. Pharm. Res. 2007, 24, 981−990.(48) Thomas, F. C.; Taskar, K.; Rudraraju, V.; Goda, S.; Thorsheim,H. R.; Gaasch, J. A.; Mittapalli, R. K.; Palmieri, D.; Steeg, P. S.;Lockman, P. R.; Smith, Q. R. Uptake of ANG1005, a novel paclitaxelderivative, through the blood−brain barrier into brain andexperimental brain metastases of breast cancer. Pharm. Res. 2009,26, 2486−2494.(49) Visser, C. C.; Stevanovic, S.; Heleen Voorwinden, L.; Gaillard, P.J.; Crommelin, D. J.; Danhof, M.; De Boer, A. G. Validation of thetransferrin receptor for drug targeting to brain capillary endothelialcells in vitro. J. Drug Targeting 2004, 12, 145−150.(50) Kohlhaas, S. L.; Craxton, A.; Sun, X. M.; Pinkoski, M. J.; Cohen,G. M. Receptor-mediated endocytosis is not required for tumornecrosis factor-related apoptosis-inducing ligand (TRAIL)-inducedapoptosis. J. Biol. Chem. 2007, 282, 12831−12841.(51) Bourasset, F.; Ouellet, M.; Tremblay, C.; Julien, C.; Do, T. M.;Oddo, S.; LaFerla, F.; Calon, F. Reduction of the cerebrovascularvolume in a transgenic mouse model of Alzheimer’s disease.Neuropharmacology 2009, 56, 808−813.(52) Bourasset, F.; Cisternino, S.; Temsamani, J.; Scherrmann, J. M.Evidence for an active transport of morphine-6-β-D-glucuronide butnot P-glycoprotein-mediated at the blood−brain barrier. J. Neurochem.2003, 86, 1564−1567.(53) Kang, Y. S.; Bickel, U.; Pardridge, W. M. Pharmacokinetics andsaturable blood−brain barrier transport of biotin bound to a conjugateof avidin and a monoclonal antibody to the transferrin receptor. DrugMetab. Dispos. 1994, 22, 99−105.(54) Dautry-Varsat, A.; Ciechanover, A.; Lodish, H. F. pH and therecycling of transferrin during receptor-mediated endocytosis. Proc.Natl. Acad. Sci. U.S.A. 1983, 80, 2258−2262.(55) Jones, A. R.; Shusta, E. V. Blood−brain barrier transport oftherapeutics via receptor-mediation. Pharm. Res. 2007, 24, 1759−1771.(56) Papademetriou, I. T.; Garnacho, C.; Schuchman, E. H.; Muro, S.In vivo performance of polymer nanocarriers dually-targeted toepitopes of the same or different receptors. Biomaterials 2013, 34,3459−3466.(57) Sumbria, R. K.; Zhou, Q. H.; Hui, E. K.; Lu, J. Z.; Boado, R. J.;Pardridge, W. M. Pharmacokinetics and brain uptake of an IgG−TNFdecoy receptor fusion protein following intravenous, intraperitoneal,and subcutaneous administration in mice.Mol. Pharmaceutics 2013, 10,1425−1431.(58) Zhou, Q. H.; Boado, R. J.; Hui, E. K.; Lu, J. Z.; Pardridge, W. M.Chronic dosing of mice with a transferrin receptor monoclonalantibody−glial-derived neurotrophic factor fusion protein. Drug Metab.Dispos. 2011, 39, 1149−1154.(59) Yu, Y. J.; Zhang, Y.; Kenrick, M.; Hoyte, K.; Luk, W.; Lu, Y.;Atwal, J.; Elliott, J. M.; Prabhu, S.; Watts, R. J.; Dennis, M. S. Boostingbrain uptake of a therapeutic antibody by reducing its affinity for atranscytosis target. Sci. Transl. Med. 2011, 3, No. 84ra44.(60) Couch, J. A.; Yu, Y. J.; Zhang, Y.; Tarrant, J. M.; Fuji, R. N.;Meilandt, W. J.; Solanoy, H.; Tong, R. K.; Hoyte, K.; Luk, W.; Lu, Y.;Gadkar, K.; Prabhu, S.; Ordonia, B. A.; Nguyen, Q.; Lin, Y.; Lin, Z.;Balazs, M.; Scearce-Levie, K.; Ernst, J. A.; Dennis, M. S.; Watts, R. J.Addressing Safety Liabilities of TfR Bispecific Antibodies That Crossthe Blood-Brain Barrier. Sci Transl Med. 2013, 5, No. 183ra157.

(61) Barrett, B.; Stiles, M.; Patterson, J. Radiation risks: Criticalanalysis and commentary. Prev. Med. 2012, 54, 280−282.



Brain Uptake of a Fluorescent Vector Targeting the Transferrin Receptor: A Novel Application of in...

Documents

Transcript of Brain Uptake of a Fluorescent Vector Targeting the Transferrin Receptor: A Novel Application of in...