Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells.

The blood-brain barrier (BBB) is one of the main obstacles for therapies targeting brain diseases. Most macromolecules fail to pass the tight BBB, formed by brain endothelial cells interlinked by tight junctions. A wide range of small, lipid-soluble molecules can enter the brain parenchyma via diffusion, whereas macromolecules have to transcytose via vesicular transport. Vesicular transport can thus be utilized as a strategy to deliver brain therapies. By conjugating BBB targeting antibodies and peptides to therapeutic molecules or nanoparticles, it is possible to increase uptake into the brain. Previously, the synthetic peptide GYR and a peptide derived from melanotransferrin (MTfp) have been suggested as candidates for mediating transcytosis in brain endothelial cells (BECs). Here we study uptake, intracellular trafficking, and translocation of these two peptides in BECs. The peptides were synthesized, and binding studies to purified endocytic receptors were performed using surface plasmon resonance. Furthermore, the peptides were conjugated to a fluorophore allowing for live-cell imaging studies of their uptake into murine brain endothelial cells. Both peptides bound to low-density lipoprotein receptor-related protein 1 (LRP-1) and the human transferrin receptor, while lower affinity was observed against the murine transferrin receptor. The MTfp showed a higher binding affinity to all receptors when compared to the GYR peptide. The peptides were internalized by the bEnd.3 mouse endothelial cells within 30 min of incubation and frequently co-localized with endo-lysosomal vesicles. Moreover, our in vitro Transwell translocation experiments confirmed that GYR was able to cross the murine barrier and indicated the successful translocation of MTfp. Thus, despite binding to endocytic receptors with different affinities, both peptides are able to transcytose across the murine BECs.

Introduction

The brain capillary endothelial cells (BECs) are a key component of the tight blood-brain barrier (BBB), which protects the brain from potentially harmful substances [1-3]. The transport across the brain endothelium is highly selective. As a result, most small and large molecule brain therapeutics fail to cross the BBB and do not reach their target within the brain parenchyma [1-3]. In order to overcome this hurdle, an endogenous transportation route is often used as a delivery strategy. These routes include carrier-mediated transport, adsorptive- and receptor-mediated transcytosis. Among these strategies, the receptor-mediated transcytosis pathway has been used to transcytose macromolecules, such as monoclonal antibodies, antibody fragments, and peptides, to the brain [4-8]. Peptides are generally easier and less costly to produce and modify than antibodies, making them excellent, specific targeting moieties [9-13]. Therefore, an increasing number of peptides have been investigated for their ability to transcytose from the circulation to the brain [11, 14, 15]. Although a few transcytosing peptides have been identified, details of their target receptor and the transcytotic mechanism are often limited [11, 15–17].

Human melanotransferrin (MTf) is expressed in melanomas as well as in multiple other tissues in lower amounts, including the brain endothelium [18-20]. Interestingly, MTf was shown to cross the BBB to a much greater extent than bovine serum albumin (BSA) and transferrin (Tf) after intravenous injection [21]. Despite the structural similarity between Tf and MTf, competition assays with holo-transferrin indicated that the transferrin receptor (TfR) was not involved in the uptake. Instead, a 50% inhibition of MTf uptake in BECs was obtained using a high concentration of receptor-associated protein, suggesting that the low-density lipoprotein receptor-related protein 1 (LRP-1) is involved in the uptake [21]. Furthermore, when MTf was conjugated to doxorubicin, a chemotherapeutic that does not normally cross the BBB, increased drug accumulation and reduced tumor growth were observed in mice [22]. Interestingly, a peptide derived from MTf showed even higher concentrations in the brain parenchyma than the full-length MTf [23]. Other MTf-derived sequences have also demonstrated transcytotic capabilities in vitro [17]. Thus, MTf and MTf-derived peptides are particularly promising as brain delivery agents.

Besides sequences derived from transcytosing proteins, phage display has guided the discovery of novel peptide sequences that can internalize and transcytose BECs. In 2010, van Rooy et al. [11] identified seventeen 15-mer peptides that bound to the murine BEC after the infusion of a random peptide library. Two of the sequences had a high affinity to BEC [11, 12]. One of these 15-amino acid peptides were the so-called GYR peptide (GYRPVHNIRGHWAPG). Recently, it has been demonstrated that the self-assembled GYR core-shell nanoparticles and nanofibers are capable of crossing BBB in vivo and that TfR and RAGE act as predominant receptors of this process [24].

This study points to future usages of GYR and MTf-derived peptides for drug delivery to the brain. Therefore, the endocytosis, subcellular trafficking, and translocation of MTf-derived peptide (Mtfp) and GYR were characterized in BEC. We evaluated live uptake into in vitro-cultured BECs using spinning disk confocal microscopy. Affinity for transcytosis-relevant receptors on BECs was measured using surface plasmon resonance (SPR).

Materials and methods

Peptide synthesis

The peptides were synthesized and labeled with fluorophores on either the N- or C-terminus. One peptide, MTfp (peptide 1, Table 1), was derived from MTf [17]. The second peptide, GYR (peptide 2, Table 1), was discovered by the aforementioned phage display [11].

Table 1

The peptide sequences MTfp and GYR.

Abbreviation	MTfp	GYR
Peptide no.	1	2
Full sequence	H–FRCLVENRGDVPFVTRIR–NH2	H–GYRPVHNIRGHWAPGK–NH2
Derived from	Portions corresponding to residues 210–229 and 556–575 of the full sequence of melanotransferrin [17]	Discovered by phage display from a library of 15-mer peptides [11, 12]

Materials and instrumentation

Fmoc-amino acids, HATU, and Fmoc-Sieber-polystyrene resin were purchased from Iris Biotech GMBH (Marktredwitz, Germany), Boc-Gly-OH from Bachem (Bubendorf, Switzerland), TAMRA-NHS from ThermoFisher Invitrogen™ (Oregon, USA), and Fmoc-Rink Amide-Tentagel resin (TGRA) from Rapp Polymere (Tübingen, Germany). All other reagents were purchased from Sigma-Aldrich Co. (Merck KGaA, Darmstadt, Germany). Peptides were synthesized by automated solid-phase peptide synthesis (SPPS) using a Biotage® Initiator+ Alstra™ Microwave Peptide Synthesizer. MALDI-TOF spectra were recorded on a Bruker AutoFlex™ MALDI-ToF MS spectrometer, in a positive-ion mode, using 2,4-dihydrobenzoic acid (DHB, 60 mg/mL) spiked with sodium trifluoroacetate in MeCN as a matrix, analyzed using Bruker autoFlex software and reported as m/z in atom mass units. UPLC-MS spectra were recorded on a Waters Acquity UPLC-MS instrument with a Waters Acquity sample manager, Waters Acquity binary solvent manager, equipped with a Waters UPLC XTerra BEH C8 or C18 column (130 Å, 1.7 μm, 2.1x50 mm), using a linear gradient of 5–100% MeCN in water over 6 min, both containing 0.1% formic acid, with a flow rate of 0.4 mL/min, detection at 220–280 nm on a TUV detector and 200–1200 m/z with ES+ and ES- ionization on a QDa single quadrupole detector. The purity of the final products was measured by analytical HPLC on a Shimadzu Nexera-x2 UHPLC instrument, equipped with a Waters XTerra RP C8 or C18 (125 Å, 5 μm, 4.6 x 150 mm) column, and using a linear gradient of MeCN in water, both containing 0.1% TFA, a flow rate of 1 mL/mL, and detection at 220 nm.

General SPPS procedure

The resin was swelled in DMF for 30 min prior to the synthesis. Each coupling reaction was carried out using HATU (3.92 equiv.), collidine (8 equiv.), and the standard Fmoc-protected L-amino acid derivative of residues unless stated otherwise (4 equiv.) in DMF solvent. For all Gly, Tyr, Pro, Val, Asn, Ile, Trp, Ala, and Lys residues, the coupling reactions were heated to 75°C with microwave irradiation for 5 min, while for all Arg, Cys, and His residues, the reactions were instead stirred at room temperature for 30 min. All Leu, Asn, and Val residues and residues after the first ten were coupled twice to ensure full conversion. Deprotection of the Nα-Fmoc group was facilitated by two rounds of 20% piperidine in DMF and stirring at room temperature for 3 and 10 min. The resin was washed twice with each DMF and DCM after the synthesis. The peptide identity of the on-resin products was confirmed by suspending a small amount of resin in 100 μL TFA:TIPS:H₂O 95:2,5:2,5 cleavage mixture for 30 min, filtering off the resin, removing the TFA with nitrogen flow and dissolving in water/MeCN (1:1) before confirming the mass by MALDI-TOF and UPLC-MS. Cleavage and full deprotection were achieved by suspending the resin in TFA/TIPS/H2O/EDT (95:1:2:2) for MTfp or TFA/TIPS/H2O (95:2.5:2.5) for GYR for 3–4 h with agitation, filtering off the resin, removing the TFA in vacuo and precipitation from cold ether.

Synthesis MTfp (1)

The resin-bound linear L-octadecapeptide H-Phe-Arg(Pbf)-Cys(Trt)-Leu-Val-Glu(tBu)-Asn(Trt)-Arg(Pbf)-Gly-Asp(tBu)-Val-Pro-Phe-Val-Thr(tBu)-Arg(Pbf)-Ile-Arg(Pbf)-TGRA was synthesized by automated SPPS. It was carried out on a 0.500 mmol scale, starting with 2.51 g of Tentagel Rink Amide resin (TGRA) with 0.25 mmol/g loading. Cleavage of a part of the resin (0.05 mmol) and purification by preparative HPLC (C18 column, 20–50% linear gradient of MeCN in water with 0.1% TFA over 15 min, flow rate 17 mL/min) afforded peptide 1 (37.3 mg, 36%) as a white powder. Purity = 92.3% and tr = 4.3 min (anal. HPLC, C8 column, 20–80% MeCN in water over 15 min). Calc. mass [M+H]+ 2176.2; found mass (MALDI-TOF) [M+H]+ 2175.7.

Synthesis of GYR (2)

The resin-bound linear L-pentadecapeptide Boc-Gly-Tyr(tBu)-Arg(Pbf)-Pro-Val-His(Trt)-Asn(Trt)-Ile-Arg(Pbf)-Gly-His(Trt)-Trp(Boc)-Ala-Pro-Gly-Lys(Mtt)-SR was synthesized by SPPS. It was carried out on a 0.100 mmol scale, starting with 165.7 mg of Sieber resin (SR) with 0.61 mmol/g loading. Fmoc-Lys(Mtt)-OH and Boc-Gly-OH were used instead of standard Fmoc-amino acids for the first and last coupling, respectively. Cleavage and purification by preparative HPLC (C18 column, 20–80% linear gradient of MeCN in water with 0.1% TFA over 30 min, flow rate 17 mL/min) afforded peptide 2 (38.2 mg, 20%) as a white powder. Purity = 95.8% and tr = 5.0 min. (UPLC, C8 column, 5–60% MeCN in water over 6 min). Calc. mass [M+H]+ 1844.0; found mass (MALDI-TOF) [M+H]+ 1844.9.

Synthesis of TAMRA-labeled MTfp (3)

To the pre-swelled resin-bound side-chain protected MTfp (1) (0.01 mmol) was added TAMRA-NHS (38.9 mg, 0.100 mmol, 2.0 equiv.) in dry DMF, followed by triethylamine (30 μL, 0.409 mmol, 8.2 equiv.). The mixture was agitated overnight at room temperature under nitrogen atmosphere and protected from light. The resin was washed three times with each DMF and DCM. Cleavage and purification by preparative HPLC (C18, 5–50% linear gradient of MeCN in water with 0.1% TFA over 30 min, flow rate 20 mL/min) afforded TAMRA-labeled peptide 3 (4.7 mg, 18%) as a pink powder. Purity = 99.7% and tr = 8.8 min (anal. HPLC, C8 column, 5–100% MeCN in water over 15 min). Calc. mass [M+H]+ 2588.3; found mass (MALDI-TOF) [M+H]+ 2587.8.

Synthesis of TAMRA-labeled GYR (4)

The resin-bound linear L-pentadecapeptide Boc-Gly-Tyr(tBu)-Arg(Pbf)-Pro-Val-His(Trt)-Asn(Trt)-Ile-Arg(Pbf)-Gly-His(Trt)-Trp(Boc)-Ala-Pro-Gly-Lys(Aloc)-TGRA was synthesized by automated SPPS. It was carried out on a 0.250 mmol scale, starting with 1.14 g of Tentagel Rink Amide resin (TGRA) with 0.25 mmol/g loading. Fmoc-Lys(Alloc)-OH and Boc-Gly-OH were used instead of standard Fmoc-amino acids for the first and last coupling, respectively. The Alloc protection was removed by adding 5 mL dry DCM, phenylsilane (406 μL, 13 equiv.) and tetrakis(triphenylphosphine)palladium(0) (30 mg, 0.1 equiv.) in 1 mL DCM to the pre-swelled resin under nitrogen atmosphere and agitating for 15 min. The procedure was repeated four times to ensure full removal of the Alloc protection group before washing the resin five times with DCM. TAMRA-NHS (10.1 mg, 1 equiv.) was added to part of the pre-swelled resin (0.029 mmol) in dry DMF/DCM (1:1, 1 mL) followed by triethylamine (4 μL, 3 equiv.). The mixture was agitated overnight at room temperature under nitrogen atmosphere and protected from light. The resin was washed three times with each DMF and DCM. Cleavage and purification by preparative HPLC (C18 column, 5–50% linear gradient of MeCN in water with 0.1% TFA over 30 min, flow rate 20 mL/min) afforded TAMRA-labeled peptide 4 as a pink powder (28.1 mg, 43%). Purity = 100% and tr = 7.3 min. (anal. HPLC, C8 column, 5–100% MeCN in water over 15 min). Calc. mass [M+H]+ 2256.1; found mass (MALDI-TOF) [M+H]+ 2255.8.

Surface Plasmon Resonance (SPR)

The binding affinity analysis of the MTf and GYR peptides was performed using the Biacore 3000 system (Cytiva, UK) equipped with a CM5 sensor chip. The sensor chip was initially activated by injection of 0.2 M 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride and 0.05 M N-hydroxysuccinimide in water. Recombinant human and mouse TfR protein (SinoBiological, Inc., Beijing, PRC) was purchased with a His-tag and immobilized on the sensor chip at 58 fmol/mm2. LRP-1 was purified from human placenta according to a previously described protocol [25] and immobilized to densities of 0.025 fmol/mm2. The remaining carboxylate groups were blocked with 1 M ethanolamine. The MTf and GYR peptides were injected for binding to the immobilized TfR or LRP-1 protein at 5 μL/min at 25°C in 10 mM HEPES, 150 mM NaCl, 1.5 mM CaCl2, 1 mM EGTA, and 0.005% Tween 20 (pH 7.4) at five different concentrations (0.25 μM, 0.5 μM, 1 μM, 2 μM and 4 μM). The reference flow channel (FC1) was activated and blocked but without immobilized TfR or LRP-1. The dissociation constant (KD) values were calculated using the BIAevaluation 4.1 software (Cytiva, UK) using the predefined Langmuir 1:1 interaction model and fitted using a drifting baseline with global fitting to the curves of the considered concentration range.

Routine cell culturing

Immortalized mouse BEC, bEnd.3 (ATCC, Cat no. CRL-2299) between passage 30 and 38 were used. For routine cell culture, the cells were seeded in rat-tail type I collagen (100 μg/ml, Sigma, Cat no. C3867-1VL) pre-coated flasks (Thermo Scientific™ Nunc™ Cell Culture Treated EasYFlasks™ or Greiner Bio-One GmbH) and were cultured in Dulbecco’s Modified Eagle Medium (DMEM) (Sigma, Cat no. D0819) supplemented with 10% fetal bovine serum (FBS) (Sigma, Cat no. F9665), 1% penicillin-streptomycin (Thermo Fisher Scientific, Cat no. 15140–122), and 1 mM sodium pyruvate (Sigma, Cat no. S8636); referred as cDMEM. Please note that all cell cultureware was coated with 100 μg/mL rat-tail type I collagen (Sigma, Cat no. C3867-1VL) diluted in ddH20 for 1 hour @ 37°C prior to cell seeding. Cells were cultured in an incubator at 37°C with 5% CO2/95% air and saturated humidity. The cell culture medium was changed every three days. Cells were sub-cultured once they reached 85–90% confluency.

Immunocytochemistry

The bEnd.3 cells were seeded on collagen-coated 8-well imaging chamber (10,000 cells/well) (Thermo Fisher Scientific, Lab-Tek® II Chamber Slide™ Cat no. 155409) or polyester Transwell inserts (40,000 cells/insert) (Corning Cat no. 3460) and grown for 5 days, until they formed a confluent monolayer. On the day of the experiment, the cells were washed with pre-heated cDMEM before fixation with 4% paraformaldehyde (Sigma, Cat no. 441244) in phosphate buffer saline (PBS) for 10 min at room temperature, followed by three PBS washes. The cells were permeabilized with 0.2% Triton X-100 (Sigma, Cat no. X100) (PBS-TX) in PBS for 10 min and blocked in 2% BSA (VWR, Cat no. 0332) (in 0.05% PBS-TX) for 20 min. For the LRP-1 and TfR surface staining, the cells were not permeabilized, and the BSA was diluted in PBS only. Cells were then incubated with primary antibody diluted in 2% BSA for 1 hour at room temperature (see list of antibodies in Table 2).

Table 2

List of antibodies.

Antigen	Antibody type	Manufacturer, catalog number	Concentration
Claudin-5	Mouse monoclonal, Clone 4C3C2	Thermo Fisher Scientific, 35–2500	2.5 μg/mL
CD31 (PECAM-1)	Mouse monoclonal, Clone 2H8	DSHB, 2H8	1.0 μg/mL
ZO-1	Mouse monoclonal	Thermo Fisher Scientific, 61–7300	2.5 μg/mL
LRP-1	Rabbit polyclonal	N/A	1.5 μg/mL
TfR	Mouse monoclonal with human FC, Clone 8D3	Provided by H. Lundbeck A/S	6 μg/mL
Mouse IgG (H+L)	Goat polyclonal, Alexa Fluor 488-coupled	Thermo Fisher Scientific, A-11001	2 μg/mL
Rabbit IgG (H+L)	Goat polyclonal, Alexa Fluor 488-coupled	Thermo Fisher Scientific, A-11008	2 μg/mL
Rabbit IgG (H+L)	Goat polyclonal, Alexa Fluor 647-coupled	Thermo Fisher Scientific, A-21244	2 μg/mL
Human IgG (H+L)	Goat polyclonal, Alexa Fluor 488-coupled	Thermo Fisher Scientific, A-11013	2 μg/mL

After incubation, cells were washed three times for 5 min with PBS and incubated with fluorescently labeled secondary antibodies (Table 2) for 1 hour in dark at room temperature, followed by rinsing three times 5 min with PBS. Then, the cells were quickly washed twice with ddH2O and the nuclei of bEnd.3 cells were stained with Hoechst 33342 (1 μg/ml concentration in ddH2O) (Sigma, Cat no. B2261) for 15 min in dark at room temperature. Finally, the cells were washed twice for 2 min with ddH2O and were kept in PBS until imaging. The cells were observed and photographed using a spinning disk confocal microscopy system consisting of a CSU-X1 spinning disk unit (Yokogawa Electric Corporation, Japan) and an Andor iXon-Ultra 897 EMCCD camera (Andor, UK), mounted on an inverted fully motorized Olympus IX83 microscope body and a UPlanSApo 60x/NA1.20 (WD = 0.28 mm) water immersion objective (Olympus Corporation, Japan). The following excitation laser lines and emission filters were used: tight junction proteins (Alexa Fluor 488-coupled secondary antibody): λex = 488 nm, λem = 525/50 nm bandpass filter and nuclei (Hoechst 33342): λex = 405 nm, λem = 440/521/607/700 nm quad-band bandpass filter. The raw images were processed using Imaris (version 8.2.1, Bitplane AG, Switzerland).

Subcellular localization studies of MTf and GYR peptides

One hundred twenty-five thousand bEnd.3 cells were seeded on collagen-coated 35 mm glass-bottom dishes (MatTek Cat no. P35G-1.5-14-C) and were grown for 4–5 days in cDMEM, until they formed a confluent monolayer. On the day of the experiment, bEnd.3 cells were incubated with 1 mL of 10 μM TAMRA-labeled MTf and GYR peptides in phenol red-free cDMEM for 30 min. After the 30 min exposure time, the medium was removed, and the samples were washed three times with peptide-free cDMEM, and finally, fresh medium was added to the cells. The imaging chambers were kept at 37°C with 5% CO2/95% air and saturated humidity during imaging. Following peptide incubation and prior to image acquisition, the acidic organelles were stained with 200 nM LysoTracker™ Green DND-26 (Thermo Fisher Scientific, Cat no. L7526) for 3 min, and the cell membrane was labeled with 5 μg/ml Wheat Germ Agglutinin, Alexa Fluor™ 488 Conjugate (WGA) (Thermo Fisher Scientific, Cat no. W11261) for 5 min and washed once with cell culture medium. Three-dimensional images (’z-stacks’) were obtained using the spinning disk confocal microscopy system described above with a UPlanSApo 60x/NA1.20 (WD = 0.28 mm) water immersion or a UPlanSApo 100xS/NA1.35 (WD = 0.20 mm) silicone immersion objectives (Olympus Corporation, Japan). The following excitation laser lines and emission filters were used: WGA, LysoTracker™ Green: λex = 488 nm, λem = 525/50 nm bandpass filter, TAMRA-labeled MTf and GYR peptides: λex = 561 nm, λem = 625/90 nm bandpass filter. Images were acquired using Olympus cellSens software (version 1.18) and processed using Imaris imaging software (version 8.2.1).

Energy-dependent uptake of MTf and GYR peptides

One hundred twenty-five thousand bEnd.3 cells were seeded on collagen-coated 35 mm glass-bottom dishes (MatTek Cat no. P35G-1.5-14-C) and were grown for 5 days in cDMEM, until they formed a confluent monolayer. On the day of the experiment, bEnd.3 cells were labeled with 5 μg/ml Alexa Fluor™ 488 conjugated WGA (Thermo Fisher Scientific, Cat no. W11261) for 5 min and washed once with cell culture medium. Then ice-cold phenol red-free cDMEM was added to the cells, and the cells were pre-incubated at 4°C for 15 min prior to peptide exposure. The bEnd.3 cells were incubated with 1 mL of 10 μM TAMRA-labeled MTf and GYR peptides in phenol red-free cDMEM at 4°C for 1 hour. Then the peptide solution was removed, and the cells were washed once with ice-cold PBS containing heparin, and warm (37°C) medium was added to the cells. Immediately after that, three-dimensional images were obtained using the spinning disk confocal microscopy system described above with a UPlanSApo 100xS/NA1.35 (WD = 0.20 mm) silicone immersion objectives (Olympus Corporation, Japan) and climate control chamber (37°C, controlled CO2 and humidity), i.e., 0 min chase. The three-dimensional images were obtained from several regions every 30 min for 2 hours. The following excitation laser lines and emission filters were used: WGA, LysoTracker™ Green: λex = 488 nm, λem = 525/50 nm bandpass filter, TAMRA-labeled MTf and GYR peptides: λex = 561 nm, λem = 625/90 nm bandpass filter. Images were acquired using Olympus cellSens software (version 1.18) and processed using Arivis Vision 4D (version 3.3.0).

Co-localization analysis

Lysosomal co-localization was determined using Imaris XTension. First, all peptides and lysosomes were identified as spots using the built-in ’spot detection’ algorithm. The ’quality intensity threshold’ parameters were adjusted manually for each image to account for differences in background intensity. The threshold was adjusted until the majority of identifiable peptides and lysosomes were labeled. Spots located within 1 μm distance were identified as co-localized spots. The Mander’s overlap coefficient (MOC) was calculated based on the following formula: The final MOC represents the mean and standard deviation (SD) of three independent experiments; 10–14 regions were imaged in each experiment.

Translocation studies of MTf and GYR peptides

For the translocation studies, an in vitro contact and non-contact co-culture model was used. Forty thousand bEnd.3 cells were seeded on the upper side of the collagen-coated Transwell insert and maintained in cDMEM for 24 hours (contact co-culture) or immediately transferred into a 12-well plate in which rat astrocytes were grown for at least three weeks (non-contact co-culture). In the case of contact co-culture, 24 hours after plating the bEnd.3 approx. 150,000 rat astrocytes were seeded on the bottom side of the flipped Transwell inserts, which was coated with 5 μg/ml poly-L-lysine (P1524, Sigma-Aldrich), and incubated for 2 hours at 37°C. The inserts were then flipped back to the original side and placed into a 12-well plate with rat astrocytes. The cells were cultured in cDMEM at 37°C with 5% CO2/95% air and saturated humidity, and the experiments were performed 4.5–5 days post-seeding of bEnd.3 cells. The medium was switched to phenol red-free cDMEM the day before the experiment.

Translocation study using live-cell imaging

These experiments were performed using the contact co-culture model. On the day of the experiment, the inserts were placed into an empty 12-well plate, and the cells were gently washed with phenol red-free cDMEM, followed by a medium exchange on both the apical and basolateral sides. The apical side of the membrane was exposed to 10 μM MTfp or GYR for 30 min, then washed once with phenol red-free cDMEM, and both sides of the membrane were stained with 5 μg/ml Alexa Fluor™ 488 conjugated WGA (Thermo Fisher Scientific, Cat no. W11261) for 3 min. Both sides of the membrane were gently washed with phenol red-free cDMEM, and the inserts were placed into a glass-bottom imaging dish filled with experimental medium. The cells were imaged with the same spinning disk microscopy system and settings as described previously. Images were processed using Arivis Vision 4D (version 3.3.0).

Translocation study based on fluorescence intensity measurement

These experiments were performed using the bEnd.3 cells that were cultured with rat astrocytes in a non-contact co-culture model and with cell-free filters as negative controls. On the day of the experiment, the inserts were transferred to a new astrocyte-free 12-well plate, and the medium was changed on both the apical and basolateral sides. The apical side of the membrane was exposed to 10 μM MTfp or GYR for 2 hours while the plates were gently shaken to overcome unstirred water-layer effects. After the 2 hours exposure time, 150 μL sample was taken from both apical and basolateral compartments, and the fluorescence intensity was read by a CLARIOstar® Plus microplate reader (BMG Labtech) (λExc = 535 nm, λEm = 585 nm, Bandwidth = 20 and 30 nm, respectively). The translocation of the peptides is presented as permeability coefficient (a concentration independent transport parameter) and was determine by using the clearance principle [26].

The following equations [27] were used to calculate the permeability coefficient: Where Papp is the apparent permeability, B is the relative fluorescence unit (RFU) at time t, T is the top chamber RFU at time 0 (constant), Vb is the volume of the bottom channel [ml], A is the cross-section area of the membrane [cm2], and t is the time [min].

The final endothelial permeability (Pe) was calculated based Eq (2): Where Papp, bEnd.3+filter is the apparent permeability of the bEnd.3 model (cells + filter), and Papp, filter is the apparent permeability of the collagen-coated blank porous membrane.

The translocation experiments were performed three times in technical triplicates, and the data are represented as mean (SD).

Paracellular permeability of 4 kDa dextran

The paracellular permeability of fluorescein isothiocyanate (FITC) labeled 4 kDa dextran (FD4, Sigma Aldrich) to determine the tightness of our bEnd.3 model.

The bEnd.3 cells were cultured with rat astrocytes as described above. On the day of the experiment, the inserts were transferred to a new astrocyte-free plate, and the medium was changed on both the apical and basolateral sides. To initiate the permeability experiment and avoid the temporary disruption of the barrier, 50 μL medium was removed from the apical side and 50 μL 200 μM working solution of 4kDa FITC-dextran was added to the upper well, i.e. the apical well had 500 μl 10 μM 4kDa FITC-dextran. In every 15 min for a period of 60 min, starting from 0 min, 100 μl medium was removed from the basolateral side. At the final time point, 100 μl medium was also removed from the apical compartment. The fluorescence intensity of the sample was read by a CLARIOstar® Plus microplate reader (BMG Labtech) (λExc = 485 nm, λEm = 515 nm, Bandwidth = 15 and 20 nm, respectively), and the paracellular permeability was calculated according to Eqs (1) and (2). The permeability experiments were repeated three times in technical triplicates, and the data are represented as mean (SD).

Statistical analysis

The normality of the co-localisation, permeability, and translocation data was tested and confirmed with GraphPad Prism 9.0.2 (GraphPad Software, Inc, CA, USA) using its built-in normality tests and QQ plots (not shown). Since the experimental data exhibited a normal/Gaussian distribution, they are presented as mean (SD).

Results

Peptide synthesis and fluorophore labeling

We synthesized two peptides (MTfp and GYR) by standard Fmoc SPPS, followed by cleavage and purification by preparative HPLC. They were used for SPR (peptide 1 and 2, respectively) to evaluate the binding affinities to key receptors. To study their uptake in vitro, the peptides were labeled with the tetramethylrhodamine (TAMRA) fluorophore. The MTfp was labeled on-resin at the free N-terminal amine with an amine-reactive NHS ester of the TAMRA carboxylic acid derivative (TAMRA-NHS) to form an amide bond. The peptide with the N-terminal TAMRA (3) was subsequently cleaved off the resin and purified by preparative HPLC. The synthesis and TAMRA labeling of MTfp is outlined in Fig 1. In order to conjugate a fluorophore to the C-terminus of GYR on-resin, a lysine residue with an orthogonal Nε-protection group was added to the C-terminus of the peptide sequence. The Alloc-protection group was removed by the palladium-catalyzed transfer to a scavenger such as phenylsilane [28, 29] and was, therefore, fully orthogonal to the acid- and base-labile protection groups used in the standard Fmoc peptide synthesis. The GYR peptide was synthesized by SPPS, the Alloc protection removed, and the free amine on the C-terminal lysine reacted with TAMRA-NHS. Full cleavage and purification by preparative HPLC afforded the TAMRA-labeled GYR (4). The synthesis and TAMRA labeling of GYR is outlined in Fig 2. The peptides were afforded in moderate yields and measured in purities above 90% by HPLC.

Fig 1

The synthesis of unlabeled (1) and TAMRA-labeled MTfp (3).

Peptide sequences are shown using the one-letter code.

Fig 2

The synthesis of unlabeled (2) and TAMRA-labeled GYR (4).

Peptide sequences are shown using the one-letter code.

Binding affinity to TfR and LRP-1

TfR and LRP-1 expression in bEnd.3 cells were confirmed by Western blotting (not shown). By using SPR, we measured the binding affinities and kinetics of the MTfp and GYR (peptide 1 and 2, respectively) to endothelial cell surface receptors: human and mouse TfRs, and human LRP-1. The equilibrium dissociation constants (KD) were determined based on the SPR sensorgrams (Fig 3). For all three immobilized receptors, the KD values of MTfp were several magnitudes lower than those of GYR. The binding affinities for both MTfp and GYR decreased in the following order: LRP-1 > human TfR > mouse TfR.

Fig 3

Receptor binding.

Binding affinities of MTfp 1 (a) and GYR 2 (b) to LRP-1 (i), human TfR (hTfR) (ii), and mouse TfR (mTfR) (iii). KD is denoted for each equilibrium.

Cellular uptake of MTfp and GYR

Next, we investigated the cellular uptake and localization of TAMRA-labeled MTfp and GYR (peptides 3 and 4, respectively) in confluent bEnd.3 cells. Since the cells were grown on glass-bottom imaging dishes, the endothelial cell barrier’s integrity and tightness were validated by the expression of endothelial junctional proteins using immunocytochemistry. S1 Fig shows that bEnd.3 cells expressed claudin-5, CD31, and ZO-1 junctional proteins. Live-cell imaging showed that both MTfp and GYR (red) were taken up after a 30-minute incubation (representative images in Fig 4). Both peptides were detected in all horizontal planes of the BECS, also on the lower (basal) cell membrane indicating translocation (cross-sectional view, Fig 4A(iv) and 4B(iv)). This uptake was energy-dependent and most likely receptor-mediated, as we have not seen any significant peptide uptake even after 2 hours (37°C) following a 1-hour peptide exposure at 4°C (S2 and S3 Figs). Furthermore, surface staining of LRP-1 and TfR demonstrated that LRP-1 is primarily located on the basolateral membrane, whereas TfR seems to be located on the apical side as well (S4 Fig). This indicates that TfR is a more important receptor for the observed uptake.

Fig 4

Internalization of TAMRA-labeled MTfp and GYR.

Uptake of 10 μM TAMRA-labeled MTfp 3 (a) and GYR 4 (b) in confluent bEnd.3 cells 15 min after a 30-min peptide exposure. (i-iii) 2D micrographs of the middle section of the WGA labeled bEnd.3 cells (green). (iv) Cross-sectional view of the randomly selected 15 μm thick region (indicated with yellow in subfigure a(i) and b(i)). Scale bars: 10 μm, image acquisition: 100x silicone immersion objective.

Moreover, live-cell imaging suggests that both MTfp and GYR accumulate in intracellular vesicles (Fig 5A(i–iii) and 5B(i–iii)). To further investigate the peptides’ subcellular localization and determine whether they accumulated in acidic organelles (mainly lysosomes), bEND.3 cells were labeled with LysoTracker™ Green. The raw data suggested high lysosomal accumulation of both MTfp and GYR (representative images in Fig 5A(i–iii) and 5B(i–iii)). Lysosomal co-localization was determined by first identifying the peptides and lysosomes as individual spots using the built-in ’spot detection’ algorithm (Fig 5A(iv–v) and 5B(iv–v)). Spots located within 1 μm distance (default spot co-localization setting of Imaris XTension) were identified as co-localized spots (Fig 5A(vi) and 5B(vi)). Mander’s coefficients of MTfp and GYR co-localization with lysosomes were 0.57 (0.07) and 0.55 (0.06), respectively, suggesting that a substantial fraction of the endocytosed peptides followed the endo-lysosomal pathway.

Fig 5

Cellular endocytosis and localization.

Subcellular localization of 10 μM TAMRA-labeled MTfp 3 (a) and GYR 4 (b) in confluent bEnd.3 cells 15 min after a 30-minute peptide exposure. (i–iii) 2D micrographs of the middle section of the LysoTrackerTM Green labeled bEnd.3 cells (lysosomal stain, green). (iv–v) 3D rendering showing all identified peptides and lysosomes as spots and in (vi) the co-localized spots. Green; lysosomes, red: MTfp and GYR. Scale bars: 10 μm, image acquisition: 100x silicone immersion objective.

Translocation of MTfp and GYR peptides

To further investigate the peptides’ BEC crossing ability, we set-up an in vitro contact and non-contact co-culture model, in which the bEnd.3 cells were co-cultured with rat astrocytes (rAstro). The integrity of the barrier was confirmed by immunostaining for tight junction markers (not included) and permeability measurement of 4 kDa FITC-Dextran (Table 3). The translocation was assessed by live-cell imaging and fluorescence intensity-based studies. As Figs 6 and 7 show, both MTfp and GYR peptides crossed the bEnd.3 monolayer after 30-min peptide exposure) however, most of them were trapped on the basal side of the membrane pores, and MTfp and GYR peptides were visually not found inside the rAstro. Fig 6B and 7B (cross-sectional view of the contact co-culture model) clearly show that most of the peptides were located on the basolateral side of the membrane (the same was also observed with cell-free controls, see S5 Fig). The translocation was quantified by measuring the fluorescence intensity of the apical and basolateral compartment following a 2-hour peptide uptake and calculating the permeability coefficient. The apparent (Papp) and “real endothelial” permeability (Pe) values are presented in Table 3. GYR had approx. 3 times higher permeability coefficient than MTfp, indicating a better translocation ability through the BBB. However, our Papp results suggest MTfp is bound to the collagen-coated polyester membrane or trapped inside the pores of the membrane as the Papp of 4kDa Dextran across empty filter was 1.45 times higher than the Papp of MTfp (approx. 1.5 less molecular weight; i.e. smaller size and consequently higher expected permeability).

Table 3

Paracellular permeability and peptide translocation studies.

Permeability coefficients of 4 kDa FITC-Dextran, TAMRA-labeled MTfp and GYR. Permeability (P) is presented as mean (SD) of three independent experiments with triplicates.

Name	MW (Da)	Papp,F (x 10−6 cm/s)	Papp, bEnd.3+F (x 10−6 cm/s)	Pe (x 10−6 cm/s)
FITC-Dextran	~4000	8.49 (1.90)	3.32 (0.48)	5.79 (0.44)
MTfp-TAMRA	2589.0	5.86 (1.27)	2.55 (0.45)	4.57 (0.64)
GYR-TAMRA	2256.6	12.82 (3.11)	6.64 (1.16)	15.17 (1.83)

MW- molecular weight, Papp,F−apparent permeability through filter, Papp,bEnd.3+F−apparent permeability through bEnd.3 grown on filter, Pe−“real” endothelial permeability. Membrane specifications: Polyester (PET) membrane with 10 μm thickness, 0.4 μm pore size, with 4x106 pores/cm2, i.e. 0.50% porosity (Corning #3460)

Fig 6

Translocation of MTfp (3).

Translocation of 10 μM TAMRA-labeled MTfp across confluent bEnd.3 cells cultured as contact co-culture with rAsto (a-b) or monoculture (c). (a) Representative 2D micrographs of different regions of the co-culture show that most of the MTfp is trapped inside the membrane pores (red: MTfp, green: WGA labeled cells). Scale bars: 20 μm. (b) Cross-sectional view of the entire field-of-view. The white dashed line marks the top part of the 10 μm thick membrane. Image acquisition: 60x water immersion objective.

Fig 7

Translocation of GYRp (4).

Translocation of 10 μM TAMRA-labeled GYRp across confluent bEnd.3 cells cultured as contact co-culture with rAsto (a-b) or monoculture (c). (a) Representative 2D micrographs of different regions of the co-culture show that most of the GYRp is trapped inside the membrane pores (red: GYRp, green: WGA labeled cells). Scale bars: 20 μm. (b) Cross-sectional view of the entire field-of-view. The white dashed line marks the top part of the 10 μm thick membrane. Image acquisition: 60x water immersion objective.

Discussion

Transport across the BBB and targeting brain diseases is difficult, and thus mechanistic insight into how transport is governed could lead to more efficient therapies. Peptides have shown promise as targeting ligands [14, 23, 30, 31], and here we show that two BBB-targeting peptides binding to relevant receptors are taken up and transported in cultured BECs. MTfp and GYR both show strong binding to LRP-1 and a less favorable binding to TfR.

Using solid-phase synthesis, we conjugated the photostable TAMRA fluorophore at either the N- or C-terminus of MTfp and GYR, respectively. By doing so, we could track the live uptake of these relevant brain-targeting peptides in vitro. Previously only quantitative assays have documented their improved BBB translocation [11, 12], but fluorescence microscopy enables studies on uptake dynamics as well as intracellular localization and sorting [7, 32]. One observation made was that a significant fraction of the peptide was co-localized with the endo-lysosomal system. Ideally, the peptide shuttle should stimulate the fusion of endosomes to tubules facilitating translocation [33], yet in a monocellular system, translocation is not a relevant output. The 4 kDa FITC-Dextran was used to measure the paracellular permeability of the model and to determine whether our observed peptide permeability was due to the leakiness of the modell or not. The literature is lacking of 4 kDa Dextran permeability data on bEnd.3 cells and only one paper was found, the measured permeability coefficient was ranging from 2.48 to 4.01 x 10−6 cm/s with a mean Pe of 2.91 (0.43) x 10−6 cm/s [34]. Our model was a bit leakier than this, as we measured a Pe of 5.79 (0.44) x 10−6 cm/s; this permeability value is similar to the hCMEC/D3 cell line, where the permeability of 4 kDa Dextran is ranging between ~5–14 x 10−6 cm/s when cultured on Transwell membrane [35-37]. Our translocation studies suggest that GYR can transcytose across the BBB. The comparison of Papp values of cell-free filters and the imaging studies (Table 3 and S5 Fig) revealed MTfp binds to the membrane and/or is trapped inside the pores of the membrane; consequently, these methods were not suitable to quantitatively determine the translocation ability of MTfp. However, as Fig 6 indicates, MTfp is able to cross the endothelial cell layer, but it is trapped inside the membrane afterwards. The use of different type of membranes could resolve this issue as the material of the membrane, pore size, and porosity could affect the peptide penetration. Furthermore, binding of MTfp to the endocytic receptors, LRP-1 and TfR, indicates that at a least a fraction of the observed translocated MTfp is conveyed by transcellular transport. Since bEnd.3 cells form a rather leaky barrier, the translocation studies should be repeated with a tighter in vitro BBB model, such as induced pluripotent stem cell-based models, or in vivo experiments in mice.

One aspect that has been discussed in the literature in terms of uptake and delivery has been the affinity to the targeting receptor. A high affinity to the receptor leads to efficient binding but may also lead to impaired release from the receptor and decreased transcellular transport [5, 38–40]. Thus, an intermediate binding affinity could be preferred. Interestingly, despite being developed for human targets [11, 12, 17], both MTfp and GYR bind to murine TfR, as shown by SPR. For the GYR peptide, the applied concentration suggests that other uptake mechanisms may be in place (10 μM GYR vs. a KD of 1 mM), whereas at this concentration, strong binding to the murine TfR by the MTf peptide is to be expected. However, LRP-1 and TfR are only two out of a multitude of possible transporters available, but our energy-dependent uptake experiments indicate that peptides enter the bEnd.3 cells by receptor-mediated endocytosis and not by a passive mechanism. The binding partners for GYR have not yet been identified in the literature [11, 12]. However, the submicromolar affinity to human LRP-1 indicates that this receptor might be one of them. In the literature, it has been indirectly shown that MTf is a ligand of LRP-1 [21], and GYR is a ligand of TfR and RAGE [24]. To our knowledge, the direct binding kinetics of the peptide sequences used in this study have not been investigated. From our data, it seems that a 30-minute incubation followed by a 15-minute chase stimulates a high level of uptake and transport in the endo-lysosomal pathway.

Although it has been suggested that MTf is not a ligand of TfR [21], these studies were competition experiments using Tf or the OX26 TfR antibody. Since the binding affinities of Tf or the OX26 antibody are much higher to the TfR [41] than our measured affinity of MTfp, it is plausible that in the presence of Tf and OX26, the uptake of MTf is diminished. It should be noted that the determined KD values were based on SPR binding kinetics of unlabeled peptides, and thus the affinity of the TAMRA-labeled peptides may be different from the one reported in the present SPR sensorgrams. Furthermore, the observed response in our SPR studies is much higher than what one would expect from a small peptide, indicating that both GYR and MTfp could bind to LRP-1 and TfR at multiple binding sites. Multiple binding sites combined with the fact that MTf has at least three receptor binding domains [17, 23, 42] could explain why no competition was observed with holo-transferrin in the study by Demeule et al. [21]. In addition, it is notable that LRP-1 in BECs has been reported to be localized at the abluminal surface [43] and that many RNA sequence databases only report very low amounts of expressed LRP-1 in these cells [44-46]. In our study, the abluminal localization was confirmed in bEnd.3 cells, and the data suggests that LRP-1 is not the main receptor facilitating the internalization of MTf to the brain in vivo. However, LRP-1 might have a function in the delivery of the GYR and MTfp to the abluminal surface by binding the peptides in endosomes after they have been endocytosed by TfR or other receptors. Such a handover mechanism needs to be followed up in future studies.

One factor, which challenged further studies of these peptides, was lack of fixability. Without a number of lysine residues in their amino acid sequences, aldehyde-based fixatives cannot keep the peptides in place. Nevertheless, by performing live-cell imaging, we could address whether the peptides passed the luminal glycocalyx (WGA). The lysotracker enabled us to assess the level of co-localization with lysosomes. Multiple drug delivery vehicles pass by the endo-lysosomal pathway [33, 47] and while some stay in this pathway and are degraded, others are transcytosed. A natural next step for a development platform, like the one described in this paper, is to address the level of translocation. This is outside the scope of the current work but will be of interest in future studies.

Conclusions

In conclusion, our data quantitatively confirmed the binding of MTfp and GYR to LRP-1 and TfR, providing important knowledge for future experiments and clinical approaches using MTfp and GYR for brain optimized drug delivery. Furthermore, we studied the in vitro uptake and distribution of fluorescently labeled peptides in brain endothelial cells. Information on receptor binding affinity and intra-cellular transport is often lacking but is needed if one wishes to improve brain targeting of biotherapeutics.

Expression of endothelial and BBB markers by bEnd.3 cells.

Confluent monolayers of bEnd.3 cells were grown on glass-bottom imaging dish and were stained for claudin-5, CD31 (PECAM-1), and ZO-1 tight and adherens junction proteins (green), and the nuclei were stained by Hoechst 33342 (blue). Scale bars: 10 μm.

(PDF)

Click here for additional data file.

Energy dependent uptake of TAMRA-labeled MTfp in confluent bEnd.3 cells.

The confluent bEnd.3 cells were exposed 10 μM TAMRA-labeled MTfp (red) for 1 hour at 4°C, then the internalization of the MTfp was followed for 120 minutes (chase). Representative 2D micrographs and cross-sectional views of the regions with MTfp (yellow dashed rectangles) show that no peptide internalization happened even after 2 hours chase. MTfp was only found on the cell surface. Cells were labeled by WGA (green). Scale bars: 10 μm, image acquisition: 100x silicone immersion objective.

(PDF)

Click here for additional data file.

Energy dependent uptake of TAMRA-labeled GYR peptide in confluent bEnd.3 cells.

The confluent bEnd.3 cells were exposed 10 μM TAMRA-labeled GYR peptide (red, circled with magenta) for 1 hour at 4°C, then the internalization of the peptide was followed for 120 minutes (chase). Representative 2D micrographs and cross-sectional views of the few regions with GYR peptide (yellow dashed rectangles) show that almost no peptide internalization happened even after 2 hours chase. Cells were labeled by WGA (green). Scale bars: 10 μm, image acquisition: 100x silicone immersion objective.

(PDF)

Click here for additional data file.

Localization of TfR and LRP-1 receptors on bEnd.3 cell surface.

Representative maximum intensity projection images and cross sectional views of the highlighted sections (yellow rectangle) show the distribution of TfR (a) and LRP-1 (b) receptors (green) on bEnd.3 cell surface. Scale bars: 10 μm, image acquisition: 100x silicone immersion objective.

(PDF)

Click here for additional data file.

Transcytosis control of MTfp (3) and GYRp (4).

Transcytosis of 10 μM TAMRA-labeled MTfp (a) and GYR (b) across a coated cell-free polyester Transwell membrane. (a-b) Representative 2D micrographs of the top and bottom side of the membrane and crosssectional views show that both MTfp and GYR are trapped inside the pores (red: peptides). Scale bars: 20 μm. Image acquisition: 60x water immersion objective. The white dashed line marks the top part of the 10 μm thick membrane.

(PDF)

Click here for additional data file.

4 Nov 2020

PONE-D-20-31660

Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells

PLOS ONE

Dear Dr. Nielsen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

Three experts have evaluated the manuscript. They agreed, that the research topic is very relevant, but also that amendment is needed including some additional experiments  (regarding the specificity of the peptides eg. uptake inhibitors, 4C). The authors need to justify the selection of the two peptides and the use of the b.End3 model.

==============================

Please submit your revised manuscript by Dec 19 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Mária A. Deli, M.D., Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please ensure that you refer to Figures 1, 2 and 3 in your text as, if accepted, production will need this reference to link the reader to each figure.

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Partly

Reviewer #3: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: N/A

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this manuscript, Sigurdardóttir and colleagues investigate the ability of 2 synthetic peptides to bind and to be uptaken by murine brain endothelial cells. This topic is very interesting because there is a need to identify new strategies to overcome the blood-brain barrier in order to treat neurodegenerative diseases. From my point of view, the manuscript is well written but conclusions need to be supported by more experiments and addition of controls. In particular, I would suggest to perform inhibition experiments or transcytosis experiments, and to use scramble peptides as controls.

Please find below my comments that need to be addressed:

- References need to be carefully checked because it appears in several places that the reference sources were not found (lines 240, 247, 258, 280 to 290, etc).

- As written in the conclusion, expression and localization of LRP1 and other members of the LDLR family are very confusing in the literature. To support their results, authors need to quantify these expressions and to investigate the luminal/abluminal expression of them.

- Lines 52-53, authors claim that RAP inhibits LRP1, but in fact, RAP inhibits all members of the LDLR family. Therefore, to specifically inhibit LRP1, authors should use Sh/SiRNA or mutagenesis approaches.

- Picture quality needs to be improved. To really demonstrate that the 2 peptides are specifically endocytosed by LRP/TfR, authors might perform several additional experiments including competitive experiments (using RAP or LRP antibodies), and should compare results obtained together in 4°C versus 37°C experiments. Inhibition of transyctosis pathways can be also performed using inhibitors or cyclodextrins, etc. Furthermore, scramble peptides should be designed to demonstrate that internalization process is specific, etc.

- Authors wrote in numerous places of the manuscript that they study transcytosis of these peptides but there is no experiment showing transcytosis results. Authors should re-phrase or might perform these experiments by seeding b.end3 cells on transwell inserts.

- In addition, authors wrote (line 270) that “Both peptides were detected in all horizontal planes of the BECS, also on the lower (basal) cell membrane indicating transcytosis”, but if the cells are leaky it is likely that these peptides cross the cells through the opened tight junctions and accumulate directly on the slides. In addition, if transcytosis process occurs, it could be mediated by adsorptive pathway and not necessary by receptor-mediated transcytosis. Again, adding adequate controls (scramble peptides) and performing competitive experiments might allow to generate more relevant data.

- Because these peptides seem to preferentially interact with the human LRP1 and TfR, why authors decided to use a murine in vitro BBB model instead to use human iPSC or stem cells ? This point should be discussed in the introduction/discussion parts.

Reviewer #2: In “Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells”, Sigurdardóttir et al. describes the interaction of two transBBB peptides with endothelial receptors, as well as the internalization pathway followed by both peptides in brain endothelial cells. The work is well organized and clear. However, the manuscript needs extra experiments and clarifications:

Major general comments:

1. I believe the two peptides employed in this study are not clearly described, should be clear their relevance and what distinguishes the peptides from other in the literature that target the same receptors. Therefore, I would like to know more about the rationale behind the design of both peptides. I think that the authors should include information on previous results that support the inclusion of both peptides in the study, and current applications (if possible).

2. In the literature, there are numerous examples of the use of bEnd.3 cells in in vitro model of BBB to evaluate peptides translocation ability. The manuscript would benefit from those assays to prove the translocation of MTfp and GYRp? It would also be interesting to should the capability of these peptides to deliver a cargo.

3. Another concern with the experimental design is related to the use of a mouse cell line (bEnd.3) instead of a human endothelial cell line. In the manuscript the authors show that both peptides have a good binding affinity towards human receptors (LRP-1 and hTfR) and poor binding affinity towards mouse receptors (mTfR). So, based on these results and the increased interest of human models, why did you select the bEnd.3 as your cell line model?

4. Figure resolution should be improved. For instance, in figure 3 the text and sensorgrams are out of focus.

5. All references from results section are gone. “Error! Reference source not found”.

My major concern is related to the innovative aspects of the manuscript. At first, it seems that peptides are original or at least not tested for BBB interaction. However, GYRp was developed by phage isolation from the brain and MTfp was used to deliver antibodies in vivo, where biodistribution demonstrates delivery of MTfp-antibody into brain parenchyma. Both peptides were tested in mice, thus it would be pertinent to test their ability in BBB models of human brain endothelial cells. The SPR results do not add additional information, since authors state “This suggests that LRP-1 is not the main receptor facilitating the transcytosis of MTf to the brain in vivo”.

Other particular comments:

Abstract (Page 2)

Line 26 – Some issue raise above concerning the use of a murine model if both peptides have more affinity towards human receptors.

Line 30 – “frequently” is a subjective concept.

Introduction (Page 3)

The introduction presents the problem of crossing the BBB in a very clear way. It shows the need for innovative strategies and the authors try to address this problem by using two peptides.

Line 47 – For MTf a fair description is present. Nevertheless, I would like to know a little more about the physiological role of MTf.

Line 50 – The sentences are confusing.

Line 54 – Do you think that the interaction with BBB receptors might be compromised by the conjugation to small molecule drugs? Did you consider the use of conjugated peptides in the study?

Line 57 – “(…) MTf and MTf-derived peptides”, Why did you select MTfp among the other “promising as brain delivery agents?”

Line 59 – Poor description. The authors need to present GYR peptide in a more clear way.

Line 63 – “(…) of GYR and MTf-derived peptides (…)” is not a completely true sentence. The authors use one MTf-derived peptide, namely the MTfp.

Materials and Methods (Page 4)

Line 70 – It is not fluorophores. The authors only conjugated both peptide to one fluorophore (TAMRA).

Line 70 – Why did you conjugated the fluorophore in the N-terminus for one peptide and in the C-terminus for the other one?

Line 176 – “relevant secondary antibodies” …

Line 193 – Why did the authors incubated both peptides for 30 min?

Line 228 – The concentrations should be presented in the same way. There are differences between the text and the figure, for instance, (250 mM, 500 mM) and (0.25 uM, 0.5 uM).

Results (Page 13)

All references are absent.

Line 234-248 – are not results but methods.

Line 260 – Please comment the SPR sensorgrams for GYRp. Sensorgrams are not a typical binding curve with association, steady state and dissociation. The fast dissociation is typical of non-specific binding.

Line 255 – The authors mention the use of WB to confirm the TfR and LRP-1 expression in bEnd.3; however, they do not present the results. Since this section is basically based on the SPR results, and in these experiments the authors do not use cells, is important to present those results.

Line 271 – “on the lower (basal) cell membrane indicating transcytosis”- is not correct to say “transcytosis”. The term transcytosis means that the molecule to be transported is captured in vesicles on one side of the cell, drawn across the cell, and release on the other side. In the results presented there is no confirmation of release on the other side.

The authors should use other controls to demonstrate internalization, for instance, perform the experiment at 4 C.

Discussion (Page 16)

Line 300 – “(…) targeting brain diseases (…)” instead of “(…) targeting diseases (…)”.

Line 305 – A comment on the conjugation at the N- or C-terminus would be helpful to understand the difference if the authors considered it interesting.

Line 319 – With substantially less affinity. Comment on that in the discussion.

Line 321 – Why is this to be expected?

Line 331 – The authors introduce OX26 as a “control”; however, they do not explain the relevance of using it. An explanation would improve its impact.

Line 334 – If you consider that the affinity would be different, why did you not use fluorescent-labeled peptides instead of unlabeled ones?

Line 343 – “fixability”

Conclusions:

The authors claim they have developed a platform for synthesis and testing of brain-targeting peptides. Do the authors consider LRP1 and TfR brain specific receptors? Do results show strong evidence of targeting to these receptors?

“Information on receptor binding affinity and intra-cellular transport is often lacking” the authors should consider other methods to determine transport pathways. Labeling of a single compartment, such as lysosomes is not sufficient to determine “intra-cellular transport”.

Reviewer #3: In the manuscript “Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells”, the authors investigate two synthetic peptides, MTfp and GYRp as BBB transcytosing peptides. Sigurdardóttir et al. demonstrate the binding ability to LRP1 and TfR receptors and the uptake and intracellular trafficking of these peptides in mouse brain endothelial cells, bEnd3. The topic of the present work is very important because transBBB peptides are promising new tools for drug delivery to the brain. The manuscript contains interesting results but in my opinion, authors need to perform extra experiments to support the conclusions.

Major comments:

1. The authors used mouse bEnd3 cell line as BBB model, however it is well known from the literature, that this model is leaky and very week as in vitro BBB model. Furthermore, the authors demonstrate that both peptides have better binding affinity towards human receptors compare to mouse ones, so authors should clarify why they selected bEnd3 cells for these experiments.

2. From the introduction part, I miss more information about the two peptides and the background results, mostly about the GYR peptides. Line 57 and 62, authors mentioned other promising transBBB peptides, why did the authors choose these ones for further studies.

3. The quality of the figures of immunocytochemistry should be improved, the text is unreadable.

4. To prove the uptake of the tested peptides in bEnd3 cells, the authors should perform extra experiments with scramble peptides as controls, investigate the temperature dependency of the cellular uptake (37°C - 4 °C) or use inhibitors.

5. The authors mentioned BBB transcytosis experiment in the abstract and later in several parts of the manuscript but there is no data demonstrating these results. Authors should show these results.

Minor comments:

1. The first part of the Results ”Peptide synthesis and fluorophore labeling” belongs to the Methods, please correct it.

2. All of the references are missing from the Results part (Error! Reference source not found.), please correct them.

In conclusion, the manuscript needs more data to underlay the authors claims.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Pr. Fabien Gosselet

Reviewer #2: Yes: Vera Neves

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-20-31660_ Reviewer Comments to authors.docx

Click here for additional data file.

22 Dec 2020

Dear Reviewers,

Thanks for the thorough review of our manuscript and the many good suggestions. We have added 6 addition figures to the manuscript (two main and 4 supplementary) with several new experiments. Below you will find a point by point response to all your comments. We hope you will find our revised version acceptable for publications in PLOS One

Sincerely

Morten Nielsen

Reviewer #1: In this manuscript, Sigurdardóttir and colleagues investigate the ability of 2 synthetic peptides to bind and to be uptaken by murine brain endothelial cells. This topic is very interesting because there is a need to identify new strategies to overcome the blood-brain barrier in order to treat neurodegenerative diseases. From my point of view, the manuscript is well written but conclusions need to be supported by more experiments and addition of controls. In particular, I would suggest to perform inhibition experiments or transcytosis experiments and to use scramble peptides as controls.

Please find below my comments that need to be addressed:

- References need to be carefully checked because it appears in several places that the reference sources were not found (lines 240, 247, 258, 280 to 290, etc).

This has been corrected. Last formatting steps with Endnote resulted in some errors we overlooked.

- As written in the conclusion, expression and localization of LRP1 and other members of the LDLR family are very confusing in the literature. To support their results, authors need to quantify these expressions and to investigate the luminal/abluminal expression of them.

We have now added new data to the manuscript. The localization and particularly the quantification at the luminal and abluminal membrane are difficult in BECs due to the thickness of the cell. As demonstrated in the new S4 Fig., with z-sections over the nucleus, we cannot observe any significant staining of LRP-1 at the luminal membrane (as observed for the TfR). These data support the published data from Zhao Z et al. (Ref 36). It has not been possible to quantify the luminal vs abluminal expression due to the limitation in the resolution of our confocal systems in the Z-axis. This important issue and consequences for the RMT of the peptides are now discussed in more details in the manuscript as well.

- Lines 52-53, authors claim that RAP inhibits LRP1, but in fact, RAP inhibits all members of the LDLR family. Therefore, to specifically inhibit LRP1, authors should use Sh/SiRNA or mutagenesis approaches.

This is correct and a good suggestion. LRP-1 ligands (inclusive RAP) do cross-react between the LDLR receptors, and the lines in 52-54 are referring to data published in Demeule M et al. (Ref 21) – it is not our claim. We did try to inhibit the cellular uptake of both peptides with recombinant RAP but did not observe any visual differences in comparison to the uninhibited uptake (difficult to quantify). However, we did not follow up on this, as the localization of LRP-1 is mainly abluminal, and therefore we do not expect any major contribution from LRP-1 regarding luminal uptake. Speculations of a plausible function of LRP-1 in BECs is now discussed in the second last paragraph in the discussion. Due to the low amount of transcytosed peptide and the fact that transfection of BEC is difficult (and therefore difficult to knock down LRP), we did not make the siRNA knock down.

- Picture quality needs to be improved. To really demonstrate that the 2 peptides are specifically endocytosed by LRP/TfR, authors might perform several additional experiments including competitive experiments (using RAP or LRP antibodies), and should compare results obtained together in 4°C versus 37°C experiments. Inhibition of transyctosis pathways can be also performed using inhibitors or cyclodextrins, etc. Furthermore, scramble peptides should be designed to demonstrate that internalization process is specific, etc.

We apologize for the picture quality. They are default included in the manuscript by the journal at low quality. They are made and submitted as high-quality images, and it should be possible to download and expect them by clicking at the link in Manuscript/PDF document. We have now added new data (S2 Fig. and S3 Fig.) demonstrating the energy depending uptake of the peptides and included corresponding text in the manuscript. The use of scrambled peptides is not straight forward due to the labeling approach and methods we have used.

- Authors wrote in numerous places of the manuscript that they study transcytosis of these peptides but there is no experiment showing transcytosis results. Authors should re-phrase or might perform these experiments by seeding b.end3 cells on transwell inserts.

We have added transcytosis experiments to the manuscript, see new Fig. 6 and Fig. 7 and corresponding text. We do underline in the discussion that these data should be further demonstrated in vivo (in mice) or by iPSC, as the bEnd3 model is not completely tight.

- In addition, authors wrote (line 270) that “Both peptides were detected in all horizontal planes of the BECS, also on the lower (basal) cell membrane indicating transcytosis”, but if the cells are leaky it is likely that these peptides cross the cells through the opened tight junctions and accumulate directly on the slides. In addition, if transcytosis process occurs, it could be mediated by adsorptive pathway and not necessary by receptor-mediated transcytosis. Again, adding adequate controls (scramble peptides) and performing competitive experiments might allow to generate more relevant data.

We have now demonstrated using 4-37 °C uptake (S2 Fig. and S3 Fig.) that the uptake most likely are energy-dependent receptor-mediated uptake, and not a result of passive diffusion. As stated above, we agree that the model is leaky and it should be verified by other models as well and particularly by in vivo experiments. We agree that scramble peptides might be a good control, but we are also aware that there are numerous peptides that binds to LRP and TfR, and we might very well get an inhibition using this approach. We think that the added 4-37 °C experiments in combination with SPR would be enough for the conclusions we have in the article.

- Because these peptides seem to preferentially interact with the human LRP1 and TfR, why authors decided to use a murine in vitro BBB model instead to use human iPSC or stem cells? This point should be discussed in the introduction/discussion parts.

Thanks for this comment. It is a very relevant question. We do have binding to mice TfR in the manuscript (Fig. 3) as the experiments were performed at bEnd3 cells. It would have been nice to have binding to mice LRP-1, but it is not possible for us to purify this receptor from mice-placenta. Moreover, as we would like to proceed with mice in vivo experiment we found that the b.End3 cells was the best choice. We have now included these considerations in the discussion.

Reviewer #2: In “Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells”, Sigurdardóttir et al. describes the interaction of two transBBB peptides with endothelial receptors, as well as the internalization pathway followed by both peptides in brain endothelial cells. The work is well organized and clear. However, the manuscript needs extra experiments and clarifications:

Major general comments:

1. I believe the two peptides employed in this study are not clearly described, should be clear their relevance and what distinguishes the peptides from other in the literature that target the same receptors. Therefore, I would like to know more about the rationale behind the design of both peptides. I think that the authors should include information on previous results that support the inclusion of both peptides in the study, and current applications (if possible).

Indeed, a good point. We have added more text in the introduction

2. In the literature, there are numerous examples of the use of bEnd.3 cells in in vitro model of BBB to evaluate peptides translocation ability. The manuscript would benefit from those assays to prove the translocation of MTfp and GYRp? It would also be interesting to should the capability of these peptides to deliver a cargo.

Correct. We have added some transcytosis data to the manuscript, see new Fig. 6 and Fig. 7

3. Another concern with the experimental design is related to the use of a mouse cell line (bEnd.3) instead of a human endothelial cell line. In the manuscript the authors show that both peptides have a good binding affinity towards human receptors (LRP-1 and hTfR) and poor binding affinity towards mouse receptors (mTfR). So, based on these results and the increased interest of human models, why did you select the bEnd.3 as your cell line model?

This relevant point was also raised by Reviewer 1. Good human models are just as difficult to find as mouse models. However, we would like to proceed with in vivo experiments, and therefore we decided to go with the mouse cell line in the end.

4. Figure resolution should be improved. For instance, in figure 3 the text and sensorgrams are out of focus.

Yes, the figures in the downloaded PDF are terrible. However, they were uploaded as high-quality images and the images in the PDF is a result of the conversion made by the journal. There should be a link in the PDF file, which should give you access to the original version.

5. All references from results section are gone. “Error! Reference source not found”.

We apologize for this. This error from EndNote should now have been corrected.

My major concern is related to the innovative aspects of the manuscript. At first, it seems that peptides are original or at least not tested for BBB interaction. However, GYRp was developed by phage isolation from the brain and MTfp was used to deliver antibodies in vivo, where biodistribution demonstrates delivery of MTfp-antibody into brain parenchyma. Both peptides were tested in mice, thus it would be pertinent to test their ability in BBB models of human brain endothelial cells. The SPR results do not add additional information, since authors state “This suggests that LRP-1 is not the main receptor facilitating the transcytosis of MTf to the brain in vivo”.

We do of course agree with these points. These studies were based on a long-term strategy to make better constructs for drug delivery to the brain. We found the MTfp in a parent from BIOasis, but previously published work claiming that TfR is not involved in the uptake of MTf (and therefore presumably also the MTfp) but is mediated by LRP-1, was not fitting with our perception of LRP-1 expression and localization in BEC. Therefore, if we should use MTfp for our future strategies to make complex dual targeting constructs, we would like to test which receptors might capable of luminal endocytosis in BEC. We think that our main conclusion, that TfR does bind MTfp and that the peptide undergoes the receptor-mediated uptake, are important for the society. Although debated, it has also demonstrated in the literature that low affinity to TfR might be important for efficient transcytosis due to the possibility that TfR might not be transported all the way to the abluminal membrane. Therefore, LRP-1 binding might still be important, since it could be involved in the exocytosis at the abluminal membrane. A final comment to this relevant discussion is that only entire MTf protein has been tested for binding to receptors before. To understand how the peptide works, binding studies (with e.g. SRP) are relevant. The GYR peptide was initially included as a positive control to our work.

In conclusion, we agree that our results are only small bricks in a large puzzle, but we find them relevant and suitable for publication in Plos One. We hope you agree.

Other particular comments:

Abstract (Page 2)

Line 26 – Some issue raise above concerning the use of a murine model if both peptides have more affinity towards human receptors.

Discussed above

Line 30 – “frequently” is a subjective concept.

Corrected

Introduction (Page 3)

The introduction presents the problem of crossing the BBB in a very clear way. It shows the need for innovative strategies and the authors try to address this problem by using two peptides.

Line 47 – For MTf a fair description is present. Nevertheless, I would like to know a little more about the physiological role of MTf.

Not much is known about MTf in the BBB, but we have added some text.

Line 50 – The sentences are confusing.

Corrected.

Line 54 – Do you think that the interaction with BBB receptors might be compromised by the conjugation to small molecule drugs? Did you consider the use of conjugated peptides in the study?

It is a highly relevant question. But as we do not have antibodies against the MTfp, this is the best alternative. For transcytosis (but not for imaging), we could have used radiolabeled peptides, but new regulations in Denmark and many other countries, unfortunately, complicates this approach.

Line 57 – “(…) MTf and MTf-derived peptides”, Why did you select MTfp among the other “promising as brain delivery agents?”

We (as many others) aim to make new dual-targeting constructs, and are therefore trying to collect several peptides with the capacity to cross the BBB. We find the MTf and MTfp are highly “underinvestigated” and might have some potential. We recently also published a paper about self-penetrating peptides (PMID: 32674358).

Line 59 – Poor description. The authors need to present GYR peptide in a more clear way.

More explanation to GYR in the second last paragraph of the introduction has been added.

Line 63 – “(…) of GYR and MTf-derived peptides (…)” is not a completely true sentence. The authors use one MTf-derived peptide, namely the MTfp.

Corrected to singular.

Materials and Methods (Page 4)

Line 70 – It is not fluorophores. The authors only conjugated both peptide to one fluorophore (TAMRA).

Corrected.

Line 70 – Why did you conjugated the fluorophore in the N-terminus for one peptide and in the C-terminus for the other one?

We wanted to use the same type of chemistry to label the peptides. Thus, the NHS ester is reacted with a primary amine in both peptides forming an amide binding. Since the amino acid sequence is different in the two peptides, labelling end was also not similar.

Line 176 – “relevant secondary antibodies” …

Secondary antibodies added to the table

Line 193 – Why did the authors incubated both peptides for 30 min?

We tried to visualize the uptake of the peptides in a continuous manner, i.e. visualization during continuous exposure starting from 0 min up to 60 min; however, the background from the peptide solution made the image acquisition difficult. Five to 15 min peptide incubation did not result in significant peptide internalization, i.e. they were visually not present; however, after a 30 min exposure time a significant receptor-mediated endocytosis occurred and it was possible to visually detect them. In other studies in the literature, peptide exposure varies between a few minutes and up to 60 min or even up to 24 hours.

Line 228 – The concentrations should be presented in the same way. There are differences between the text and the figure, for instance, (250 mM, 500 mM) and (0.25 uM, 0.5 uM).

Corrected.

Results (Page 13)

All references are absent.

Line 234-248 – are not results but methods.

Since the peptides were synthesized and labeled in-house (which required a long optimization process and have not been published previously), we wanted to include the synthesis and labeling as part of the result section. We understand that in some papers it is not common, but some readers with medicinal chemistry background might may find it relevant

Line 260 – Please comment the SPR sensorgrams for GYRp. Sensorgrams are not a typical binding curve with association, steady state and dissociation. The fast dissociation is typical of non-specific binding.

Text added

Line 255 – The authors mention the use of WB to confirm the TfR and LRP-1 expression in bEnd.3; however, they do not present the results. Since this section is basically based on the SPR results, and in these experiments the authors do not use cells, is important to present those results.

The WB displays the presence of the receptors in the cells. We do not find it important for the discussion of the binding. The expression is also described in the literature. Below are our blots (Figure is only in uploaded responce and can be found in end of this PDF)

Line 271 – “on the lower (basal) cell membrane indicating transcytosis”- is not correct to say “transcytosis”. The term transcytosis means that the molecule to be transported is captured in vesicles on one side of the cell, drawn across the cell, and release on the other side. In the results presented there is no confirmation of release on the other side.

The authors should use other controls to demonstrate internalization, for instance, perform the experiment at 4 C.

Several new experiments have been added

Discussion (Page 16)

Line 300 – “(…) targeting brain diseases (…)” instead of “(…) targeting diseases (…)”.

Done

Line 305 – A comment on the conjugation at the N- or C-terminus would be helpful to understand the difference if the authors considered it interesting.

This was in part chosen since we wanted to use the same fluorophore and thus not have this difference as a confounding factor. Thus, we have not included a discussion of N vs C terminal conjugation.

Line 319 – With substantially less affinity. Comment on that in the discussion.

Very low affinity is not desirable as the target (peptide, antibody etc.) needs to bind the cellular receptor for efficient uptake. In general, the binding affinity found in the literature varies between 0.4 nM to 1000 nM.

Line 321 – Why is this to be expected?

The binding affinity of GYR to mouse TfR was only 1 mM, which is considered to be too low for successful RMT delivery. Such a low binding affinity (KD of 810 nM) could lead to successful brain delivery; however, it was reported for bispecific antibodies by Genentech (ref 5 - Yu et al. 2014, doi:10.1126/scitranslmed.3009835); they used an antibody construct with TfR and BACE1 binding. In the same year, Roche claimed that low binding affinity was needed for successful delivery (ref 7 – Niewoehner et al. 2014, doi:10.1016/j.neuron.2013.10.061)- they used monovalent antibody fragment (sFab).

Hence, we expect that GYR has a binding affinity to other receptors as well. On the other hand, we expect a stronger binding between MTfp and mouse TfR as the affinity was much lower (KD of 1.3 µM). We slightly changed the wording in the text and use “stronger” instead of “strong”.

Line 331 – The authors introduce OX26 as a “control”; however, they do not explain the relevance of using it. An explanation would improve its impact.

Text added

Line 334 – If you consider that the affinity would be different, why did you not use fluorescent-labeled peptides instead of unlabeled ones?

Yes, it could have been done. We decided to use only one constructs to get comparable results.

Line 343 – “fixability”

It is in the English dictionary. But we agree, it is not used often so we changed the text.

Conclusions:

The authors claim they have developed a platform for synthesis and testing of brain-targeting peptides. Do the authors consider LRP1 and TfR brain specific receptors? Do results show strong evidence of targeting to these receptors?

“Information on receptor binding affinity and intra-cellular transport is often lacking” the authors should consider other methods to determine transport pathways. Labeling of a single compartment, such as lysosomes is not sufficient to determine “intra-cellular transport”.

Yes, we agree. The main problem is that we cannot fixate the peptides, and therefore the use of vesicular specific antibodies and ordinary immunofluorescence staining is not possible. We are only aware of live tracing markers for lysosomes and mitochondria, and the latter is not relevant. We are also aware that we could have tried the CellLight™ BacMam 2.0 family for staining further intracellular compartments; however, this “staining” is based on a 24 hours transfection and since we had difficulties with efficient transfection of BECs, we decided not to proceed with this technology.

Reviewer #3: In the manuscript “Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells”, the authors investigate two synthetic peptides, MTfp and GYRp as BBB transcytosing peptides. Sigurdardóttir et al. demonstrate the binding ability to LRP1 and TfR receptors and the uptake and intracellular trafficking of these peptides in mouse brain endothelial cells, bEnd3. The topic of the present work is very important because transBBB peptides are promising new tools for drug delivery to the brain. The manuscript contains interesting results but in my opinion, authors need to perform extra experiments to support the conclusions.

Major comments:

1. The authors used mouse bEnd3 cell line as BBB model, however it is well known from the literature, that this model is leaky and very week as in vitro BBB model. Furthermore, the authors demonstrate that both peptides have better binding affinity towards human receptors compare to mouse ones, so authors should clarify why they selected bEnd3 cells for these experiments.

This is relevant and raised by the other referees as well. We aim to follow up this in vitro study with in vivo experiments in mice, and therefore we decided to perform our in vitro experiments with the same species. Second, the available human models (particularly the hCMEC/D3) are just as leaky as b.End3 cells, and the alternative iPSC models have several other concerns of being more epithelial-like than endothelial.

2. From the introduction part, I miss more information about the two peptides and the background results, mostly about the GYR peptides. Line 57 and 62, authors mentioned other promising transBBB peptides, why did the authors choose these ones for further studies.

More information on the peptide has been added. There are many interesting and promising peptides that can and should be investigated. We think the MTf peptides could be a peptide that can complement other BBB crossing peptides, particularly in a dual-targeting strategy. Therefore, we think it is relevant to study the MTfp as well.

3. The quality of the figures of immunocytochemistry should be improved, the text is unreadable.

The images are all high quality, but the formatting done by the journal resulted in poor quality images in the downloaded PDF. To get the original images, they should be downloaded separately. We apologize, but we can not do this differently.

4. To prove the uptake of the tested peptides in bEnd3 cells, the authors should perform extra experiments with scramble peptides as controls, investigate the temperature dependency of the cellular uptake (37°C - 4 °C) or use inhibitors.

Done

5. The authors mentioned BBB transcytosis experiment in the abstract and later in several parts of the manuscript but there is no data demonstrating these results. Authors should show these results.

Done

Minor comments:

1. The first part of the Results ”Peptide synthesis and fluorophore labeling” belongs to the Methods, please correct it.

Since the peptides were synthesized and labeled in-house (which required a long optimization process and have not been published previously), we wanted to include the synthesis and labeling as part of the result section. We understand that in some papers it is not common, but some reader with medicinal chemistry background might find it relevant

2. All of the references are missing from the Results part (Error! Reference source not found.), please correct them.

Sorry, formatting issue. Should be corrected.

In conclusion, the manuscript needs more data to underlay the authors claims.

Several new data have been included to improve the manuscript conclusions.

Submitted filename: Responce to Reviewers.docx

Click here for additional data file.

25 Jan 2021

PONE-D-20-31660R1

Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells

PLOS ONE

Dear Dr. Nielsen,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses all the points raised during the review process.

==============================

The manuscript has been greatly improved and many of the original concerns were addressed. A couple of requests remained to be answered, especially the calculation of the transcytosed peptide as a permeability coefficient, either as Papp or Pe. Another important point is the characterization of the model in terms of paracellular permeability using a marker molecule with the same MW as the peptide, eg. fluorsecently labele dextran. All the other points are minor that can be easily amended.

==============================

Please submit your revised manuscript by Mar 11 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Mária A. Deli, M.D., Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: N/A

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Authors partially replied to my concerns, that has, from my opinion, improved the quality of the results and the manuscript. However, some parts can still be improved, in particular the transcytosis experiments :

- Authors claim for the MTFp that 103 % of the peptide are still in the apical compartment at the end of the experiment, whereas 3 % have transcytosed and that the rest is trapped in the coated insert. This kind of data is no really relevant and raise doubt about the methods and the quantification methods. I would suggest to the authors to use the clearance principle to calculate the MTFp permeability (Pe) across the cells. For this, I would suggest to refer to the method to generate a concentration-independent parameter as described by Siflinger-Birnboim et al. (A. Siflinger-Birnboim, P.J. Del Vecchio, J.A. Cooper, F.A. Blumenstock, J.M. Shepard, A.B. Malik, Molecular sieving characteristics of the cultured endothelial monolayer, J Cell Physiol, 132 (1987). 111-117. 10.1002/jcp.1041320115). At the end, with this method, the authors will be also able to calculate the % of recovery. Difference between the starting quantity of peptide and the remaining quantity will correspond to the part trapped in inserts and/or degraded by BEC. Because these peptides are also observed in lysosomes, this latter point might be then discussed in the manuscript. Authors are also encouraged to read this review on transport assessment when using in vitro BBB models : ”Santa-Maria AR, Heymans M, Walter FR, Culot M, Gosselet F, Deli MA, Neuhaus W. Transport Studies Using Blood-Brain Barrier In Vitro Models: A Critical Review and Guidelines. Handb Exp Pharmacol. 2020 Oct 11.”

- Then, to justify that this model, in this condition, is suitable for permeability studies, authors should include a paracellular marker such as a 3kDa-Dextran control, with an almost similar molecular weight than the transcytosed peptides of interest. Then permeabilities of the peptides and dextran need to be compared together.

- Authors wrote in the conclusion : “In conclusion, we have designed a platform allowing for the synthesis and testing of BBB-targeting peptides.”. I am not convinced that this is the main message of the manuscript and that this is really true. In addition, this kind of BBB-targeting system already exist because several labs around the world make this transcytosis or paracellular experiments, routinely, even with human BBB models that are often more tight and easy to handle than b.end3 cells.

Reviewer #2: The authors have addressed the main concerns pointed out in the review report. Thus, the quality of the manuscript increased substantiality. However, I still have some concerns related to the translocation assays and the respective statements.

The authors followed the reviewers’ suggestion and introduced a translocation assay. This is a big and important part of the work, since these peptides were designed to penetrate the brain. The model used is a co-culture, which is usually referred as an improvement to single cell models.

Major comments I would like to be addressed:

1. Was the integrity of barrier measured, using a fluorescent probe?

2. Why did you use microscopy to detect translocation? You could have detect fluorescence in the basolateral compartment by measuring the fluorescence intensity of labeled-peptides.

3. You should check stability of the peptides. Are peptides degraded in acidic conditions?

4. Can 2.9 and 3.7% translocation be considered good translocation? Have you determine a threshold for translocation in your BBB models? In my view, the authors cannot state that MTfp and GYR “transcytose across the BBB”. The authors should rephrase the following sentences:

Line 31 – “Moreover, our in vitro Transwell transcytosis experiments confirmed that both MTfp and GYR peptides were able to transcytose across a murine barrier. Thus, despite binding to endocytic receptors with different affinities, both peptides are able to transcytose across the murine BECs.”

Line 398 – “Our transcytosis studies suggest that both peptides can transcytose across the BBB.

5. What do the authors mean with “The rest were most likely trapped inside the membranes.”? Since the results are “2.9±0.8% MTfp and 3.7±0.2% GYR were able to transcytose into the basolateral compartment, whereas 103.1±2.6% and 94.6±0.5% remained inside the apical compartment” is there something left?

Minor comments:

Line 20 – conjugating instead of attaching

Line 56 – I believe reference 21 is not correct, as Demeule et al 2002 does not use doxorubicin. Please check all references.

Line 182 – lacks a for five minutes

Line 224 – Lacks a “(“

Reviewer #3: The authors have adequately addressed my comments and added several new data to the manuscript. They tested both the temperature dependency of the cellular uptake and the transcytosis of the peptides upon my request.

I have only one minor finding: the transcytosis of MTfp and GYR peptides would be better to show with the calculation of the apparent permeability coefficient and not with the form of % of transcytosed peptide.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Fabien Gosselet

Reviewer #2: Yes: Vera Neves

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: PONE-D-20-31660R1_ Reviewer Comments to authors.docx

Click here for additional data file.

5 Mar 2021

The manuscript has been greatly improved and many of the original concerns were addressed. A couple of requests remained to be answered, especially the calculation of the transcytosed peptide as a permeability coefficient, either as Papp or Pe. Another important point is the characterization of the model in terms of paracellular permeability using a marker molecule with the same MW as the peptide, eg. fluorsecently labele dextran. All the other points are minor that can be easily amended.

Dear Editor and reviewers,

Thanks for the many nice comments and suggestions. We have now studied the permeability in more detail and evaluated the translocation using Papp and Pe. As discussed by you and our comments in the last rebuttal, the bEnd.3 cells are not the best model to study transcytosis. Our major message in this publication is the binding and endocytosis of the peptides to endocytic receptors, which is the first step in a successful transcytosis from blood to brain. The transcytosis/translocation studies we have provided is indicative and should be further validated in vivo. This should hopefully be clear from the discussion and conclusion. Below we have addressed all concerns and comments point-by-point. We hope that the manuscript is acceptable now for publication in PlosOne.

Sincerely

Morten Nielsen

Reviewer #1: Authors partially replied to my concerns, that has, from my opinion, improved the quality of the results and the manuscript. However, some parts can still be improved, in particular the transcytosis experiments :

- Authors claim for the MTFp that 103 % of the peptide are still in the apical compartment at the end of the experiment, whereas 3 % have transcytosed and that the rest is trapped in the coated insert. This kind of data is no really relevant and raise doubt about the methods and the quantification methods. I would suggest to the authors to use the clearance principle to calculate the MTFp permeability (Pe) across the cells. For this, I would suggest to refer to the method to generate a concentration-independent parameter as described by Siflinger-Birnboim et al. (A. Siflinger-Birnboim, P.J. Del Vecchio, J.A. Cooper, F.A. Blumenstock, J.M. Shepard, A.B. Malik, Molecular sieving characteristics of the cultured endothelial monolayer, J Cell Physiol, 132 (1987). 111-117. 10.1002/jcp.1041320115). At the end, with this method, the authors will be also able to calculate the % of recovery. Difference between the starting quantity of peptide and the remaining quantity will correspond to the part trapped in inserts and/or degraded by BEC. Because these peptides are also observed in lysosomes, this latter point might be then discussed in the manuscript. Authors are also encouraged to read this review on transport assessment when using in vitro BBB models : ”Santa-Maria AR, Heymans M, Walter FR, Culot M, Gosselet F, Deli MA, Neuhaus W. Transport Studies Using Blood-Brain Barrier In Vitro Models: A Critical Review and Guidelines. Handb Exp Pharmacol. 2020 Oct 11.”

As suggested, the translocation was recalculated using the clearance principle and is presented as apparent and effective permeability (Papp and Pe). To calculate the apparent permeability, we used the following equation based on Czupalla et al. (doi: 10.1007/978-1-4939-0320-7_34):

P_app [cm/s]=B/T∙V_b/(A∙t∙60)

Where Papp is the apparent permeability, B is the relative fluorescence unit (RFU) at time t (120 min), T is the top chamber RFU at time 0 (assumed to be constant; hence, we used the top chamber RFU at 120 min), Vb is the volume of the bottom channel [ml], A is the cross-section area of the membrane [cm2], and t is the time [min]. The reason for using this equation instead of the suggested V=(〖[A]〗_b∙V_b)/〖[A]〗_t and P_app=dV/(dt∙A) (where [A]b is the concentration of the tracer in the basolateral compartment and [A]t is the initial concentration of the tracer in the basolateral/top compartment) from Siflinger-Birnboim et al. (doi: 10.1002/jcp.1041320115) is that the former equation is entirely based on the measured fluorescence intensity and not on the calculated concentrations based on a standard curve, i.e. in our opinion, a slightly bit more accurate. Furthermore, we calculated the permeability coefficients using both equations and the results were very similar. In the case of GYR-TAMRA, the Papp were 6.64 x10-6 cm/s vs. 6.71 x10-6 cm/s vs. 6.31 x10-6 cm/s using RFU values, the calculated concentrations based on a standard curve or the calculated final basal concentration and the 10 µM initial apical concentration values, respectively. The new data has been included in the manuscript as Table 3.

- Then, to justify that this model, in this condition, is suitable for permeability studies, authors should include a paracellular marker such as a 3kDa-Dextran control, with an almost similar molecular weight than the transcytosed peptides of interest. Then permeabilities of the peptides and dextran need to be compared together.

We used 4 kDa Dextran (FD4 from Sigma, molecular weight (MW) is ranging between 3 and 5 kDa). Both Papp and Pe was calculated and included in the manuscript and the values were compared with each other. The Papp through the cell-free filter revealed that most of the MTfp are trapped inside the membrane (as it was confirmed by microscopy) as the Papp, filter, MTfp was smaller than Papp, filter, Dextran. Since the MW of the labelled MTfp is smaller than the dextran (~2.6 kDa vs. ~4 kDa), the expected Papp, filter, MTfp is higher than Papp, filter, Dextran as it was seen in the case of GYR. Hence, Transwell membrane (or at least polyester membrane with 0.4 µm pore size) and fluorescence intensity-based analysis of the basolateral compartment are not suitable to measure the translocation of MTfp across the BBB. This part is included and explained in the manuscript as well.

All permeability values are presented in Table 3.

- Authors wrote in the conclusion : “In conclusion, we have designed a platform allowing for the synthesis and testing of BBB-targeting peptides.”. I am not convinced that this is the main message of the manuscript and that this is really true. In addition, this kind of BBB-targeting system already exist because several labs around the world make this transcytosis or paracellular experiments, routinely, even with human BBB models that are often more tight and easy to handle than b.end3 cells.

Thank you, we agree with you and it has been modified in the manuscript.

Reviewer #2: The authors have addressed the main concerns pointed out in the review report. Thus, the quality of the manuscript increased substantiality. However, I still have some concerns related to the translocation assays and the respective statements.

The authors followed the reviewers’ suggestion and introduced a translocation assay. This is a big and important part of the work, since these peptides were designed to penetrate the brain. The model used is a co-culture, which is usually referred as an improvement to single cell models.

Major comments I would like to be addressed:

1. Was the integrity of barrier measured, using a fluorescent probe?

The barrier integrity was determined by measuring the paracellular permeability of fluorescein isothiocyanate (FITC) labeled 4 kDa dextran and sodium fluorescein (376 Da) and these data are now included in the manuscript.

2. Why did you use microscopy to detect translocation? You could have detect fluorescence in the basolateral compartment by measuring the fluorescence intensity of labeled-peptides.

We used microscopy and traditional fluorescence intensity-based techniques to detect translocation. We used microscopy as we wanted to see if the peptides were able to be endocytosed and translocate through the membrane and taken up by the astrocytes grown in juxtaposition on the opposite side of the Transwell membrane. Binding to endocytic receptors is a critical point for further success in in vivo translocation and successful drug delivery. Although we did not observe any translocated peptides inside the astrocytes using microscopy (most of them trapped inside the pores of the polyester membrane), a degree of translocation was observed by fluorescence intensity measurement of the basolateral compartment.

3. You should check stability of the peptides. Are peptides degraded in acidic conditions?

This is a valid request. We initially labeled a large fraction of peptides with TAMRA, which subsequently was purified and validated using HPLC in smaller portions when needed. The purification over time did not indicate any degradation but we can unfortunately not provide any quantitative data.

4. Can 2.9 and 3.7% translocation be considered good translocation? Have you determine a threshold for translocation in your BBB models? In my view, the authors cannot state that MTfp and GYR “transcytose across the BBB”. The authors should rephrase the following sentences:

Line 31 – “Moreover, our in vitro Transwell transcytosis experiments confirmed that both MTfp and GYR peptides were able to transcytose across a murine barrier. Thus, despite binding to endocytic receptors with different affinities, both peptides are able to transcytose across the murine BECs.”

Line 398 – “Our transcytosis studies suggest that both peptides can transcytose across the BBB.

The translocation was recalculated and is now presented as apparent and effective permeability. These permeability values were compared with the permeability of 4 kDa dextran, and they are discussed in the manuscript.

We agree that we have made conclusions about transcytosis mechanisms that are not proved by our data. Therefore, the term “trancytosis” is now corrected to “translocation” in all relevant places in the manuscript text.

5. What do the authors mean with “The rest were most likely trapped inside the membranes.”? Since the results are “2.9±0.8% MTfp and 3.7±0.2% GYR were able to transcytose into the basolateral compartment, whereas 103.1±2.6% and 94.6±0.5% remained inside the apical compartment” is there something left?

As our translocation data is now presented as permeability (see Table 3), and the referred texts are now removed from the text.

However, we meant that in the case of the GYR, we collected 3.7 + 94.6 = 98.3% < 100%, and the remaining 1.7% is most likely trapped inside the filter membrane or in the endothelial cell layer. Again, we would like to stress that the receptor identification and binding is the most important message of the manuscript. Therefore the conclusion on translocation is less strong at the moment.

Minor comments:

Line 20 – conjugating instead of attaching

Line 56 – I believe reference 21 is not correct, as Demeule et al 2002 does not use doxorubicin. Please check all references.

Line 182 – lacks a for five minutes

Line 224 – Lacks a “(“

Many thanks for these observations, they are now corrected in the text.

Reviewer #3: The authors have adequately addressed my comments and added several new data to the manuscript. They tested both the temperature dependency of the cellular uptake and the transcytosis of the peptides upon my request.

I have only one minor finding: the transcytosis of MTfp and GYR peptides would be better to show with the calculation of the apparent permeability coefficient and not with the form of % of transcytosed peptide.

Thank you very much. The translocation of the peptides is now presented as apparent permeability coefficient. Furthermore, the calculated effective permeability coefficient values were also calculated (see Table 3).

Furthermore, following the statistical analysis guidelines of PLOS One, our data is presented as mean (SD) instead of mean ± SD. Furthermore, the normality of data distribution was assessed with GraphPad Prism 9.0; the QQ plots can be found below. These plots are not included in the manuscript.

Submitted filename: Responses to Reviewers Comments.docx

Click here for additional data file.

23 Mar 2021

Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells

PONE-D-20-31660R2

Dear Dr. Nielsen,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Mária A. Deli, M.D., Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

Reviewer #2: In “Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells”, Hudecz et al. describe the interaction of two BBB translocation peptides with endothelial receptors, as well as the internalization pathway and translocation efficiency of both peptides in brain endothelial cells. The authors have addressed the main concerns pointed out in previous review report and I endorse its publication.

Reviewer #3: The authors have adequately addressed my comments raised in the previous round of review and I feel that this manuscript is now acceptable for publication.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Pr. Fabien Gosselet

Reviewer #2: No

Reviewer #3: No

25 Mar 2021

PONE-D-20-31660R2

Two peptides targeting endothelial receptors are internalized into murine brain endothelial cells

Dear Dr. Nielsen:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Mária A. Deli

Academic Editor

PLOS ONE