Inhibition of leptin-ObR interaction does not prevent leptin translocation across a human blood-brain barrier model

Daniel Gonzalez-Carter1,2*, Angela Goode2, Roberto Fiammengo3, Iain E. Dunlop2, David T. Dexter1, Alexandra Porter2*

1. Centre for Neuroinflammation and Neurodegeneration, Department of Medicine, Division of Brain Sciences, Imperial College London, London, UK, W12 0NN

2. Department of Materials, Imperial College London, Exhibition Road, London, UK, SW7 2AZ

3 Centre for Biomolecular Nanotechnologies, UniLe, Istituto Italiano di Tecnologia, (ITT), Via Barsanti, 73010 Arnesano, Lecce, Italy

*Corresponding authors: Daniel Gonzalez-Carter (daniel.gonzalez-carter08@imperial.ac.uk); Alexandra Porter (a.porter@imperial.ac.uk)

Key words: Blood-brain barrier, leptin, leptin receptor, transcytosis, LRP-2

Abstract

The adipocyte-derived hormone leptin regulates appetite and energy homeostasis through activation of leptin receptors (ObR) on hypothalamic neurons, hence leptin must be transported through the blood-brain barrier (BBB) to reach its CNS target sites. During obesity however, leptin BBB transport is decreased, in part precluding leptin as a viable clinical therapy against obesity. Though the short isoform of the ObR (ObRa) has been implicated in the transport of leptin across the BBB due to its elevated expression in cerebral microvessels, accumulating evidence indicates leptin BBB transport is independent of ObRa. Here, we employed an ObR-neutralizing antibody (9F8) to directly examine the involvement of endothelial ObR in leptin transport across an in vitro human BBB model composed of the human endothelial cell line hCMEC/D3. Our results indicate that, while leptin transport across the endothelial monolayer was non-paracellular, and energy- and endocytosis-dependent, it was not inhibited by pre-treatment with 9F8, despite the latter’s ability to recognize hCMEC/D3-expressed ObR, prevent leptin-ObR binding and inhibit leptin-induced STAT-3 phosphorylation in hCMEC/D3 cells. Furthermore, hCMEC/D3 cells expressed the transporter protein LRP-2, capable of binding and endocytosing leptin. In conclusion, our results demonstrate leptin binding to and signalling through ObR is not required for efficient transport across human endothelial monolayers, indicating ObR is not the primary leptin transporter at the human BBB, a role which may fall upon LRP-2. A deeper understanding of leptin BBB transport will help elucidate the exact causes for leptin resistance seen in obesity and aid in the development of more efficient BBB-penetrating leptin analogues.

Introduction

Leptin, a 16 kDa adipocyte-derived hormone, is intimately involved in appetite regulation and energy homoeostasis through modulation of hypothalamic neuronal activity [1, 2]. In order to exert its central actions, circulating leptin released by adipose tissue must cross the blood-brain barrier (BBB) to bind to leptin receptors (ObR) in the cell surface and vesicles of hypothalamic neurons [3, 4]. Lack of leptin signalling, either due to absence of circulating leptin or functional ObR, leads to obesity [5]. In addition however, diet-induced obesity in rodents leads to leptin resistance by inhibiting leptin transport through the BBB into the brain [6, 7]. Similarly, obese humans with functional ObR have

elevated circulating leptin levels [8], indicating impaired leptin entry into the brain. Hence, a better understanding of leptin translocation across the BBB would shed light on peripheral leptin resistance, presenting new strategies to combat obesity.

The BBB is primarily composed of the endothelial cells lining the brain microvasculature. Unlike peripheral endothelial cells, cerebral endothelial cells have a low permeability due to the presence of intercellular tight-junction structures and active efflux systems, preventing the passive entry of molecules greater than ~500 daltons [9]. Given its high molecular weight, leptin is unable to passively permeate across the BBB. Therefore, specific transport mechanisms must exist to ensure rapid and selective leptin transport into the brain parenchyma. Indeed, a large body of evidence supports the existence of a selective leptin transporter at the BBB. Studies have demonstrated leptin transport into the brain from the circulation is unidirectional and saturable [10, 11]. In addition, entry of leptin into the brain may be enhanced or inhibited pharmacologically [12, 13], indicating leptin transport depends on a specific transporter at the apical surface of the BBB which may be occupied competitively as well as stimulated to specifically increase leptin entry into the brain.

The identity of the BBB leptin transporter has been suggested to be the short splice variant of the leptin receptor (ObRa). This suggestion is based on the fact that the long form of ObR (ObRb) is the major signalling receptor, with ObRa inducing only weak signalling but efficient leptin internalization [14, 15]. In addition, ObRa expression is particularly enhanced in the brain microvessels [16], positioned optimally to aid BBB transport. Furthermore, knock-out animal models have indicated ablation of ObRa expression leads to a reduction in leptin transport into the brain under certain conditions [17, 18].

However, there is a growing body of evidence indicating ObRa is not the primary leptin transporter at the BBB. Firstly, age-dependent ObRa expression changes do not correlate with age-dependent changes in leptin brain transport rates. Through real-time PCR, Pan et al. [19] have demonstrated ObRa mRNA levels within the cerebral microvessels are greatly enhanced in 7 day-old mice compared to 2 month-old mice. However, leptin brain transport is comparable between the two ages. Importantly, leptin brain transport was measured through in situ brain perfusion, therefore confounding factors, such as increased levels of circulating leptin, were eliminated. In addition, Hileman et al. [11] have found that, while leptin brain transport is decreased in the New Zealand obese mouse model compared to control mice, the level of ObRa mRNA expression at the cerebral microvessels is comparable between both. Similarly, levels of ObRa mRNA have been shown to be up-regulated at the BBB following a high-fat diet [20], which is associated with a decrease in leptin brain penetration.

Secondly, animal selective gene knock-out studies have shown leptin transport into the brain and its central actions following peripheral administration are retained even under conditions of ObRa ablation. For instance, in the Koletsky rat obesity model, which lacks global functional leptin receptors, restoration of ObRb expression in hypothalamic neurons through viral gene therapy results in reduced food intake [21], indirectly demonstrating leptin brain transport is still functional even in the absence of ObRa. Similarly, while specific ablation of ObRa gene expression in mice leads to a small increase in body weight compared to wild-type controls under a high-fat diet (though not under a normal diet), peripheral leptin administration is still able to induce a marked reduction in body weight [22]. More importantly, peripheral leptin administration in these mice

leads to activation of central (hypothalamic) neuronal leptin receptors, as evidenced by phosphorylation of STAT-3, again providing evidence of leptin brain entry even in the absence of ObRa. Direct measurement of leptin brain transport has also been carried out in ObRa knock-out animal models, demonstrating sustained leptin entry into the brain despite lack of ObRa. For example, Banks et al. [18] measured the entry of radioactively labelled leptin into the brain in both ObR-deficient Koletsky rats and wild-type controls. The authors found that, while radioactive leptin entry is decreased in the Koletsky rats following intravenous injection ( i.e. when blood components are able to compete with exogenous leptin), entry is comparable between the two groups if leptin administration is carried out through brain perfusion (i.e. in the absence of blood components). Importantly, the entry of radioactive leptin following perfusion could be inhibited if co-administered with non-radioactive leptin, indicating entry was mediated by a saturable transporter, as opposed to passive diffusion due to a leaky BBB.

Fully elucidating the mechanism of leptin transport across the BBB is of great importance as it will allow for a better understanding of leptin resistance in obese individuals, as well as synthesis of leptin peptides with enhanced BBB penetration to improve leptin’s central effects. Though animal models have yielded valuable evidence to support non-ObR mediated leptin transport across the BBB, it is difficult to accurately gauge the extent of ObR involvement from these studies due to the many variables and confounding factors inherent in the models (e.g. elevated circulating leptin levels, capillary leptin retention). Furthermore, no study has examined the role of endothelial ObR in leptin transport across the human blood brain barrier. With this in mind, the present study aimed to elucidate the role of endothelial ObR in leptin transcytosis across a model of the human BBB by utilizing the human brain endothelial cell line hCMEC/D3 and a leptin receptor neutralizing antibody to block ObR-leptin interaction, allowing for specific examination of ObR involvement during leptin transcytosis.

Methods

Cell cultureThe immortalized human brain microvascular endothelial cell line hCMEC/D3 was obtained from Dr. Ignacio Romero, Open University, U.K. The cell line was established by lentiviral vector transduction of endothelial cells derived from human temporal lobe microvessels with the catalytic subunit of human telomerase (hTERT) and SV40 large T antigen [23]. hCMEC/D3 cells were cultured in tissue culture flasks coated with Cell-bind technology (Corning, UK) in EBM-2 medium (Lonza, Switzerland) supplemented with VEGF (unless otherwise stated), IGF1, EGF, basic FGF, hydrocortisone, ascorbic acid, gentamycin, and 2.5% (v/v) FBS as recommended by the manufacturer (hereafter termed full EBM). Cells were incubated at 37oC in 5% CO2, replenished with fresh medium every 2-3 days and passaged when they reached 80% confluency.

For transwell experiments, hCMEC/D3 cells were grown on Transwell polyester micropore membranes (1 m pore diameter, 12 mm membrane diameter) (Millipore, UK) coated with calf skin collagen type I (Sigma-Aldrich, UK) and bovine plasma fibronectin (Sigma-Aldrich, UK) in 12-well plates, with full EBM in the upper (0.5 mL) and lower (1.5 mL) chamber. Once cells reached ~90% confluency (after 3-4 days), medium was changed to full EBM without VEGF (hereafter termed NV-

EMB), and cells cultured for a further 2 days before experimentation began to allow the monolayer formation to acquire a BBB phenotype.

For experiments not requiring transwells, cells were plated directly onto collagen/fibronectin-coated wells/glass cover slips and grown as above before experimentation began.

TEER measurementTrans endothelial electrical resistance (TEER) measurements were carried with an EVOM ohm meter equipped with an STX2 electrode set (World Precision Instruments Inc., UK). TEER readings of blank transwells were subtracted from primary TEER readings of transwells coated with cells to obtain TEER measurement of the cell monolayer per se. TEER measurement was multiplied by the transwell’s surface area (1.12 cm2) to obtain the final measurement in Ω/cm2.

Permeability coefficient measurementMonolayer permeability to 4 and 70 kDa FITC-conjugated dextran (paracellular transport markers) (Sigma-Aldrich, UK) and full-size human leptin (Life Technologies, UK) (human leptin was employed throughout the study) was quantified as previously described [24]. Transwells were washed with phenol-free DMEM with 2% FCS (hereafter termed transport medium). Transport medium containing study compound (1 mg/mL FITC-Dextran or 4.65 nM leptin) was administered into the upper chamber (0.5 mL) and the flux of compound into the lower chamber (containing 1.5 mL transport medium) measured at fixed intervals (10 mins FITC-dextran, 30 mins leptin) by fluorescence quantification of the basolateral medium at exc 490 nm/em 510 nm in the case of FITC-dextran or an enzyme-linked immunosorbent assay (ELISA) (see below) in the case of leptin. The basolateral compound concentration was subsequently used to calculate the cleared volume (vol cl) as follows:

volcl = (Cl*Vl)/Cu

where Cl and Cu represent the concentration in the lower and upper chamber, respectively, and V l

represents the volume in the lower chamber.

The cumulative cleared volume was plotted against time and the rate of transport (Rt) was calculated from the slope of the curves. The Rt across the cells per se (Rtc) was calculated by subtracting the Rt across empty transwell membranes (Rt tw) from the Rt of transwell membranes containing cells (Rtt) following the below equation:

1/Rtc = 1/Rtt- 1/Rttw

The permeability coefficient (Pe) (measured in cm.min -1) across cells was calculated by dividing the Rtc by the surface area (A) of the transwell membrane:

Pe = Rtc/A

Leptin-induced STAT-3 phosphorylationTo examine leptin signalling, quantification of STAT-3 phosphorylation was carried through immunostaining of Western blots. Cells plated in 6 well plates (see above) were treated with leptin (in NV-EMB) for 5 mins at room temperature (RT). In experiments examining 9F8-inhibition of leptin

signalling, the cells were pre-treated with 9F8 (in NV-EBM) for 30 mins (at 37 oC) prior to treatment with leptin (with 9F8 present) for 5 mins at RT. Following treatment, cells were washed with PBS and lysed to quantify phosphorylated STAT-3 through Western blotting (see below).

Western blotting

Cells treated on 6 well plates (see above) were collected by trypsinization into Eppendorf tubes and centrifuged. The resulting pellet was resuspended in RIPA buffer (200 L)(with 10% protease inhibitors, Sigma-Aldrich UK), incubated over ice for 10 mins and cleared by centrifugation. Following protein quantification by a Bradford assay, 25 g of protein were separated through SDS-PAGE (10% gel) and transferred onto a PVDF membrane. After blocking in TBS-0.1% tween with 5% BSA (2hrs at RT), the membrane was incubated with primary antibodies (see table 1) in TBS-0.1% tween with 1% BSA overnight (4oC). The membrane was then incubated in HRP-conjugated secondary antibodies (see table 1) and the signal developed through enhanced chemiluminescence.

Immunocytochemistry

Cells grown on glass cover slips (GCS) (see above) were fixed with ice-cold 4% paraformaldehyde (PFa) in PBS at 4oC and permeabilized with PBS/0.2% Triton-x 100 (Tx). Following blocking with 5% normal serum/0.1% Tx in PBS, cells were incubated with primary antibodies (see table 1) (in 5% normal serum/0.1% Tx in PBS) at 4oC overnight. Cells were then incubated with secondary Alexa-488/546 antibodies (Invitrogen, UK) (in 5% normal serum/0.1% Tx in PBS) (see table 1) at RT, followed by incubation with 4',6-diamidino-2-phenylindole (DAPI) to stain cell nuclei. Glass coverslips were then mounted onto glass slides with Vectashield anti-fading medium, and imaged through fluorescence microscopy utilizing a Nikon E1000M eclipse fluorescence microscope equipped with a Q-imaging Q-CAM camera.

Translocation assaysTranslocation of leptin across hCMEC/D3 was carried out by administering leptin (4.65nM) into the apical chamber in transport medium (0.5 mL). Following a 90 mins incubation, leptin concentration in the basolateral medium was quantified by an ELISA (Peprotech, UK) according to the manufacturer’s insctructons. In experiments examining leptin translocation inhibition by sodium azide (NaN3) or the ObR neutralizing 9F8 antibody [25] (a kind gift from Prof. Richard Ross, Sheffield University), hCMEC/D3 monolayers were pre-incubated with NaN3 or 9F8 (30 mins, 37oC) before the addition of leptin (with NaN3 or 9F8 present), and translocation measured as above. In experiments studying leptin translocation inhibition by cold temperature, cells were pre-incubated at either 37oC or at 4oC for 30 mins before the addition of leptin. Translocation incubation (90 mins) was then carried out at 37oC or 4oC and basolateral leptin measured as above. In order to ensure neither the cold temperature nor administration of 9F8 affected the monolayer permeability, the translocation of 4 kDa FITC-dextran across the monolayer was monitored following leptin translocation assay as described above. Competition plate binding assay

In order to ensure the neutralizing antibody 9F8 inhibited human leptin-ObR binding, a competition plate-binding assay was carried out. Briefly, a high-binding 96-well plate (Corning, USA) was coated with human ObR protein (Thr20-Asp830, corresponding to the extracellular receptor domain) (1 g/mL in PBS, pH 7.4) (R and D systems, UK) overnight at RT. Following blocking with 3% BSA (in PBS) for 2.5hr at RT, the plate was washed with wash buffer (PBS with 0.05% Tween-20) and incubated with human leptin (4.65 nM, in diluent buffer (PBS/0.05% Tween-20 with 0.1% BSA)) with or without increasing concentrations of 9F8 neutralizing antibody for 2hr at RT (in duplicate wells) (wells were pre-incubated with 9F8 for 30 mins prior to addition of leptin). Following further washings, the plate was incubated with biotinylated rabbit anti-leptin (0.25 g/mL in wash buffer with 0.1% BSA) (Peprotech, UK) for 2 hrs at RT. Following further washings, the plate was incubated with avidin-horse radish peroxidase (HRP) conjugate (Peprotech, UK) for 30 mins at RT. The washing step was repeated and leptin binding quantified by addition of 2,2 -azino-bis(3-ethylbenzothiazoline-6-sulfonic′ acid) (ABTS, Sigma-Aldrich, UK) at RT and quantification of optical density (OD) with a plate reader with wavelength set at 405 nm. In order to ensure 9F8 inhibition of leptin-ObR binding was specific for ObR, the competition plate binding assay was repeated replacing ObR coating with a rabbit anti-leptin capture antibody (1 g/mL in PBS) (Peprotech, UK). Because of the high affinity of the capture antibody, the plate was incubated with 62.5 pM leptin instead of 4.65 nM.

Statistical analysis

Results are expressed as mean ± SEM of n = 3, unless stated otherwise. Lines of best fit were calculated through linear regression analysis using GraphPad Prism 4.0 (GraphPad, USA). Statistics were performed using GraphPad Prism 4.0 by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test to identify statistically significant differences between three or more groups; or students t-test to identify statistically significant differences between two groups. Statistical significance was set at p < 0.05.

Results

Establishment of an in vitro human BBB model to examine endothelial ObR-mediated leptin translocation

In order to employ the hCMEC/D3 cell line as a BBB model to examine leptin translocation, their ability to form a tight monolayer barrier with low paracellular flux first had to be confirmed. To this end, hCMEC/D3 cells grown on collagen/fibronectin-coated transwell membranes were examined for acquirement of a BBB phenotype, formation of tight junction complexes and low paracellular permeability. Immunocytochemical staining of fixed endothelial monolayers revealed hCMEC/D3 successfully acquired a BBB phenotype, as evidenced by expression of von Willebrand factor (vWF) (fig. 1a), an endothelial-marker protein abundantly expressed in cerebral endothelial cells [26, 27]; and expression and correct localization (i.e. at cellular interfaces) of the tight junction-marker protein zona occludens (ZO)-1 (fig. 1b). Furthermore, immunoblotting of hCMEC/D3 Western blots with an antibody directed against occludin, a protein responsible for bridging adjacent cell membranes to construct tight junction complexes [28], identified a specific protein band at the predicted occludin molecular weight (~65 kDa) [29] (fig. 1c), confirming expression of occludin by

hCMEC/D3 monolayers. Bright field TEM imaging of the junctions between cells revealed closely-associated plasma membranes between monolayer cells, indicating appropriate tight-junction complex architecture (fig. 1d). In addition, hCMEC/D3 monolayers exhibited low paracellular permeability, as evidenced by low permeability coefficient to the paracellular flux markers 70 and 4 kDa FITC-dextran (fig. 1e, f), and TEER values comparable to published BBB in vitro models (fig. 1g) [23, 30, 31].

In order to employ the hCMEC/D3 cell line to examine endothelial ObR-mediated leptin transport, the expression of ObR had to be established. Positive ObR expression was detected through immunocytochemical labelling of fixed hCMEC/D3 monolayers (fig. 1h, upper panels). The specificity of immunostaining was confirmed through negative control staining (omission of primary antibody) (fig. 1h, lower panels). Furthermore, immunoblotting of hCMEC/D3 Western blots with an anti-ObR antibody identified a specific protein band at the appropriate size (150-170 kDa) [32-35](fig. 1i).

Energy-dependent translocation of leptin across hCMEC/D3 endothelial monolayers

As leptin-induced ObR signalling involves STAT-3 phosphorylation [36, 37], the ability of hCMEC/D3 cells to recognize leptin through ObR binding was examined through quantification of STAT-3 phosphorylation. Cell monolayers treated (5 mins) with increasing concentrations of leptin showed a dose-dependent increase in phosphorylated STAT-3 compared to control cells (fig. 2a, b), indicating functional leptin-ObR interaction in hCMEC/D3 cells. Though there was an increase in STAT-3 phosphorylation with all leptin concentrations examined, only the highest dose (46.5nM) resulted in a statistically significant increase, probably due to the low sensitivity of Western blot detection.

Once interaction of leptin with hCMEC/D3 cells was established, the ability of leptin to be actively transcytosed through the cell monolayer was examined. To this end, monolayers plated on transwells were treated with leptin in the apical chamber, and leptin concentration in the basolateral medium quantified through an ELISA at 30 min intervals for a total of 2h. A dose of 4.65 nM leptin was chosen as it closely recreates the concentration of circulating leptin seen in vivo [38]. Leptin translocated into the basolateral chamber with a constant rate of 0.16 + 0.005 % of apical quantity/minute (fig. 2c). In contrast, translocation of 4 kDa dextran (1 mg/mL, measured at 10 min intervals for 40 mins) (employed as an internal control) had a constant rate of 0.06 + 0.004 % of apical quantity/minute (fig. 2c). The translocation rates across empty transwells were subsequently used to calculate the permeability coefficient (Pe) of leptin and 4 kDa dextran across hCMEC/D3 monolayers per se. As can be seen in fig. 2d, leptin had a significantly higher P e (7.96 + 0.44 x 10-4

cm/min) than 4 kDa dextran (3.79 + 0.32 x 10-4 cm/min). The energy-dependence of leptin translocation across hCMEC/D3 monolayers was examined by quantifying leptin translocation at 37oC and 4oC over a 90 min incubation. Incubation at 4oC has been widely used in the past to inhibit energy-dependent cell internalization as the rate of active physiological processes is significantly inhibited at this temperature [39-43]. In agreement with an active transcytosis mechanism, translocation of leptin was significantly inhibited by incubation at 4oC compared to translocation at 37oC (fig. 2e). Importantly, the decreased leptin translocation at lower temperature was not due to a decrease in permeability, as evidenced by comparable 4kDa dextran flux across monolayers following incubation at 37oC and 4oC for 90 mins (fig. 2f). Similarly, transport of leptin across empty transwells was not affected by incubation (90 mins) at 4oC or 37oC (fig. 2g), indicating decreased

leptin detection efficiency or retention by the transwell membrane were not responsible for lower basolateral leptin concentrations when incubated at 4oC. In order to examine whether the evidenced leptin translocation relied on ATP-dependent endocytosis, the translocation experiments were repeated in the presence or absence of NaN3 (5 mM), a metabolic inhibitor which prevents ATP-dependent membrane internalization and endocytosis [43-46]. Total leptin translocation in the presence of NaN3 decreased compared to control conditions, though this did not reach statistical significance (fig. 2h). However, treatment with NaN3 increased monolayer permeability (data not shown). Therefore, in order to compensate for the increase in paracellular leptin leakage, leptin translocation was normalized against 4 kDa dextran translocation, which resulted in a marked and statistically significant reduction in leptin translocation under NaN3 treatment conditions (fig. 2i).

Leptin translocation is not inhibited by an ObR-neutralizing antibody

The role of ObR in the evidenced leptin translocation across hCMEC/D3 monolayers was examined by inhibiting ObR-leptin conjugation utilizing the ObR-neutralizing antibody 9F8 [25, 47]. Firstly, the ability of 9F8 to inhibit human ObR-human leptin interaction was examined through a plate-binding assay. As can be seen in fig. 3a, coating of a 96 well-plate with human ObR led to a significant increase in leptin binding following a 2 hr leptin (4.65 nM) incubation. Leptin binding was specific for ObR, as leptin failed to bind to wells lacking ObR coating (fig. 3a). The ability of 9F8 to inhibit human ObR-human leptin interaction was confirmed by the significant and dose-dependent reduction in leptin binding when incubated with increasing concentrations of 9F8 (fig. 3a). The specificity of 9F8 for ObR-leptin interaction was confirmed by repeating the plate-binding assay replacing ObR coating for coating with an anti-leptin antibody. In this case, leptin binding was not impaired by co-incubation with increasing concentrations of 9F8 (fig. 3b). The ability of 9F8 to bind ObR expressed in hCMEC/D3 cells was examined by incubating Western blots of hCMEC/D3 lysates with 9F8. As can be seen in figure 3c, immunolabelling with 9F8 detected a protein band of ~150 kDa, corresponding to the expected ObR molecular weight, indicating 9F8 was able to bind to human endothelial-derived ObR. The ability of 9F8 to prevent ObR-leptin conjugation in hCMEC/D3 was assessed by its ability to inhibit leptin-induced STAT-3 phosphorylation. Co-incubation with 9F8 (100 g/mL) ablated the increase in leptin-induced STAT-3 phosphorylation (fig. 3d, optical density quantified in 3e), indicating 9F8 prevented ObR-leptin interaction in hCMEC/D3 cells. Once the interaction of 9F8 with ObR in hCMEC/D3 cells was established, its ability to inhibit leptin translocation across hCMEC/D3 monolayers was examined. Accumulation of leptin in the basolateral chamber following a 90 min incubation of hCMEC/D3 monolayers with leptin (4.65 nM) was not decreased by co-incubation with 9F8 even at the highest concentration (100 g/mL) (fig. 3f), indicating leptin transcytosis was not inhibited by prevention of ObR-leptin binding. Importantly, barrier permeability to 4 kDa dextran was not affected by 9F8 treatment (fig. 3g), discarding the possibility that leptin translocation was unaffected by 9F8 due to an increase in barrier permeability counteracting the inhibition of transcytosis. Several studies have indicated leptin is able to bind to and be endocytosed by the transporter protein low density lipoprotein-related protein (LRP)-2 [48, 49]. Hence, LRP-2 expression by hCMEC/D3 cells was examined through Western blotting. Immunostaining of hCMEC/D3 lysate-derived Western blots with an anti-LRP2 antibody revealed a specific protein greater than 250 kDa, corresponding to the predicted molecular weight of LRP-2 (300-500 kDa) (fig. 3h, left panel). The band was shown to be specific for LRP-2, as omission of primary antibody resulted in disappearance of the band (fig. 3h, right panel).

Discussion

The discovery of leptin as a powerful anorexic adipokine capable of inducing marked weight loss in animal models of obesity and certain rare human genetic cases of leptin deficiency led to significant expectation of its use as a general anti-obesity therapy in humans [50-52]. However, its efficacy in the clinic has not met this promise [53], mainly due to leptin resistance in obese patients resulting, at least in part, from inadequate leptin brain entry. Therefore, fully elucidating the mechanisms of leptin blood-brain barrier translocation will allow for the changes leading to transport inefficiency to be understood and overcome, allowing leptin, or a derived functional analogue, to be successfully implemented as an anti-obesity therapy.

With that goal in mind, the main aim of the current study was to examine the role of endothelial ObR in leptin translocation across the human blood-brain barrier. To that end, the human endothelial cell line hCMEC/D3 was employed to model leptin transcytosis across the human blood brain barrier in vitro. The use of an in vitro BBB model to examine leptin transcytosis was chosen as such models have several advantages over in vivo studies. Firstly, entry of proteins into the brain parenchyma in vivo may be confounded by capillary retention, whereby proteins are retained in the capillaries without actual penetration into the brain [54]. Secondly, obesity leads to elevated circulating leptin levels. Given that the leptin transcytosis machinery is saturated at low levels of circulating leptin [55, 56], quantifying exogenous leptin brain entry in obesity situations, such as occurring in ObR-deficient animals, may result in reduced levels due to transporter saturation as opposed to decreased transport efficiency. Indeed, while measuring exogenous radioactive leptin brain entry following intravenous injections to ObR-deficient rats leads to reduced levels of brain radioactivity compared to wild-type controls, this reduction is ablated if leptin brain entry is measured through brain perfusion, where the excess circulating leptin levels are removed [18]. Hence, employing an in vitro assay allows control over the leptin concentration which the endothelial cells are exposed to, thereby avoiding the confounding effect of variable circulating leptin. Similarly, in vitro models allow for acute and direct inhibition of endothelial leptin-ObR interaction, thereby avoiding confounding secondary effects of ObR ablation in vivo. Thirdly, to the best of our knowledge, all studies examining the role of endothelial ObR in leptin transcytosis across the BBB have employed non-human models, therefore examining leptin transcytosis across a human immortalized endothelial cell monolayer gave our study higher clinical relevance. A single previous study has examined in vitro leptin transport across a human brain endothelial monolayer [57], however this study focused on the effects of astrocytic leptin receptors on general endothelial monolayer permeability, without examining the direct effects of endothelial leptin receptors on leptin translocation. Furthermore, the mechanisms of leptin transcytosis across the monolayer were not studied. However, despite these shortcomings, the Hsuchou study does highlight the role of cell-type specific leptin receptors on BBB permeability, and specifically astrocytic receptors.

In order to successfully employ the hCMEC/D3 cell line as a human BBB model to examine the role of endothelial ObR in leptin translocation, we first had to confirm the expression of ObR by hCMEC/D3 cells, as well as confirming leptin transport across the monolayer occurred through a non-paracellular, energy-dependent, endocytic process, as previous studies employing hCMEC/D3 cells have not addressed these issues [57]. Leptin receptor expression was established through

immunocytochemistry and Western blotting. Furthermore, the rate of leptin transport was shown to be significantly higher than the paracellular marker dextran, and the level of leptin translocation was significantly inhibited by both cold temperatures and pre-incubation with the endocytosis inhibitor sodium azide, demonstrating leptin transcytosis across the hCMEC/D3 monolayer required an energy-dependent transporter relying on endocytosis.

Subsequently, the role of ObR in the evidenced active transcytosis of leptin through hCMEC/D3 monolayers was examined by utilizing an anti-ObR neutralizing antibody (9F8). 9F8 successfully blocked human ObR-human leptin interaction, as evidenced by almost complete inhibition of leptin binding in an ObR-coated plated binding assay with the highest 9F8 concentration tested (100 g/mL). Importantly, a similar 9F8 concentration did not affect leptin binding to an anti-leptin capture antibody, indicating 9F8 interaction was specific for ObR protein. Similarly, 9F8 was shown to bind ObR when expressed in hCMEC/D3 cells, indicating its ability to inhibit leptin-ObR interaction was maintained in our BBB model. This was further confirmed by prevention of leptin-stimulated increase in endothelial STAT-3 phosphorylation by 9F8 pre-incubation.

Despite the ability of the neutralizing antibody to prevent leptin-ObR interaction, it failed to modulate the active and specific leptin translocation across the hCMEC/D3 monolayer, indicating ObR is not the primary leptin transporter across our human BBB model. In addition, given that 9F8 binds to the extracellular domain of ObR [47], which is conserved throughout all ObR splice variants [58] and hence prevents interaction between leptin and all leptin receptor isoforms expressed in hCMEC/D3 endothelial cells, these results indicate none of the isoforms are involved in leptin transport. Speculatively, the high expression of ObRa in brain microvessels, instead of aiding BBB transcytosis, could function to sequester leptin intracellularly leading to degradation, thereby modulating the brain entry of leptin in a localized fashion independent of circulating leptin levels. Indeed, leptin receptors have been shown to be able to internalize and degrade leptin in a lysosomal fashion, with ObRa being more efficient at leptin degradation [59]. Furthermore, ObRa is recycled to the cell surface following leptin-induced internalization more efficiently than ObRb [60]. Hence, ObRa is optimally localized and has the adequate leptin processing mechanisms to negatively regulate leptin brain entry. A mechanism by which ObRa expression/activity aids in regulating regional leptin brain entry would be advantageous as circulating leptin levels would not be able to be tailored to the needs of specific regions by modulating adipose-tissue secretion.

An alternative candidate BBB transporter for leptin would be the multi-ligand receptor LRP-2. Indeed, LRP-2 has been implicated in the transport of leptin across the blood-cerebrospinal fluid barrier [48] and is known to bind and internalize leptin in the kidney over ObR [49]. Therefore, a similar mechanism may be present in the BBB. In order to explore this possibility, we examined the expression of LRP-2 by hCMEC/D3 cells. Staining of Western blots with an anti-LRP2 antibody identified a specific protein band with molecular weight greater than 250kDa, in accordance with the predicted molecular weight of LRP-2 (300-500 kDa). Hence, hCMEC/D3 cells express the appropriate transporter to support an ObR-independent, LRP-2-dependent leptin transcytosis mechanism.

Regardless of the identity of the leptin transporter, conclusively discarding the involvement of ObR in leptin BBB transcytosis will allow for the leptin brain entry impairment seen during obesity to be more deeply understood, thereby opening up further possibilities for disease treatment. Similarly, the design of brain-penetrating leptin analogues and peptides would greatly benefit from a further

clarification of leptin brain entry mechanisms, as in some instances the design rests on encompassing the ObR binding domains within leptin. For instance, Barrett et al. [61, 62] examined the brain penetrating capacity of a leptin peptide containing the 61-90 amino acid segment of the full length leptin protein spanning a receptor binding sequence. Interestingly, though efficient brain penetration was evidenced, the peptide acted to inhibit leptin function, inducing marked weight gain in wild-type female rats presumably through antagonism of hypothalamic ObR, indicating ObR activation is not a requirement for BBB transport of leptin peptides. Therefore, a deeper understanding of leptin BBB transport will allow for a more informed peptide synthesis strategy resulting in both efficient brain entry and ObR-activation capacity. Similarly, identification of the leptin transporter would provide a clearer target to study the effects of astrocytic leptin signalling on BBB permeability and leptin transcytosis [57].

In conclusion, the results presented in this study indicate that, while leptin is able to signal through ObR in human endothelial cells and its transport across the BBB is an energy-dependent endocytic process, leptin transcytosis across the human BBB is independent of leptin-ObR interaction.

Acknowledgments

The authors gratefully acknowledge funding from an ERC starting investigator grant # 257182. In addition, the authors thank Prof. Richard Ross (Sheffield University) for the generous supply of the 9F8 antibody.

Conflict of interest

The authors of the manuscript have no conflicts of interest to declare.

References

[1] Schwartz M, Woods S, Porte D, Seeley R, Baskin D. Central nervous system control of food intake. Nature. 2000;404:661-71.

[2] Perry B, Wang Y. Appetite regulation and weight control: the role of gut hormones. Nutrition and diabetes. 2012;16:e26.[3] Sahu A. Leptin signaling in the hypothalamus: emphasis on energy homeostasis and leptin resistance. Frontiers in Neuroendocrinology. 2003;24:225-53.[4] Ha S, Baver S, Huo L, Gata A, Hairston J, Huntoon N, et al. Somato-dendritic localization and signaling by leptin receptors in hypothalamic POMC and AgRP neurons. PLoS One. 2013;8:e77622.[5] Ahima R. Revisiting leptin's role in obesity and weight loss. Journal Clinical Investigation. 2008;118:2380-3.[6] Van Heek M, Compton d, France C, Tedesco R, Fawzi A, Graziano M, et al. Diet-induced obese mice develop peripheral, but not central, resistance to leptin. Journal Clinical Investigation. 1997;99:385-90.[7] Halaas J, Boozer C, Blair-West J, Fidahusein N, Denton D, Friedman J. Physiological response to long-term peripheral and central leptin infusion in lean and obese mice. PNAS. 1997;94:8878-83.[8] Maffei M, Halaas J, Ravussin E, Pratley R, Lee G, Zhang Y, et al. Leptin levels in human and rodent: measurement of plasma leptin and ob RNA in obese and weight-reduced subjects. Nature Medicine. 1995;1:1155-61.[9] Pardridge W. The blood-brain barrier: bottleneck in brain drug development. NeuroRx. 2005;2:3-14.[10] Banks W, Kastin A, Huang W, Jaspan J, Maness L. Leptin enters the brain by a saturable system independent of insulin. Peptides. 1996;17:305-11.[11] Hileman S, Pierroz D, Masuzaki H, Bjorbaek C, El-Haschimi K, Banks W, et al. Characterization of short isoforms of the leptin receptor in rat cerebral microvessels and of brain uptake of leptin in mouse models of obesity. Endocrinology. 2002;143:775-83.[12] Banks W. Enhanced leptin transport across the blood-brain barrier by a1-adrenergic agents. Brain Research. 2001;899:209-17.[13] Banks W, Coon A, Robinson S, Moinuddin A, Shultz J, Nakaoke R, et al. Triglycerides induce leptin resistance at the blood-brain barrier. Diabetes. 2004;53:1253-60.[14] Belouzard S, Rouille Y. Ubiquitylation of leptin receptor ObRa regulates its clathrin-mediated endocytosis. EMBO journal. 2006;25:932-42.[15] Bjorbaek C, Uotani S, da Silva B, Flier J. Divergent signaling capacities of the long and short isoforms of the leptin receptor. Journal of Biological Chemistry. 1997;272:32686-95.[16] Bjorbaek C, Elmquist J, Michi P, Ahima R, van Bueren A, McCall A, et al. Expression of leptin receptor isoforms in rat brain microvessels. Endocrinology. 1998;139:3485-91.[17] Kastin A, Pan W, Maness L, Koletsky R, Ernsberger P. Decreased transport of leptin across the blood-brain barrier in rats lacking the short form of the leptin receptor. Peptides. 1999;20:1449-53.[18] Banks W, Niehoff M, Martin D, Farrell C. Leptin transport across the blood-brain barrier of the Koletsky rat is not mediated by a product of the leptin receptor gene. Brain Research. 2002;950:130-6.[19] Pan W, Hsuchou H, Tu H, Kastin A. Developmental changes of leptin receptors in cerebral microvessels: unexpected relation to leptin transport. Endocrinology. 2008;149:877-85.[20] Boado R, Golden P, Levin N, Pardridge W. Up-regulation of blood-brain barrier short-form leptin receptor gene products in rats fed a high fat diet. Journal of Neurochemistry. 1998;71:1761-4.[21] Morton G, Blevins J, Williams D, Niswender K, Gelling R, Rhodes C, et al. Leptin action in the forebrain regulates the hindbrain response to satiety signals. Journal Clinical Investigation. 2005;115:703-10.[22] Li Z, Ceccarini G, Eisenstein M, Tan K, Friedman J. Phenotypic effects of an induced mutation of the ObRa isoform of the leptin receptor. Molecular Metabolism. 2013;2:364-75.

[23] Weksler B, Subileau E, Perriere N, Charneau P, Holloway K, Leveque M, et al. Blood-brain barrier-specific properties of a human adult brain endothelial cell line. FASEB journal. 2005;19:1872-4.[24] Lopez-Ramirez A, Fischer R, Torres-Badillo C, Davies H, Logan K, Pfizenmaier K, et al. Role of caspases in cytokine-induced barrier breakdown in human brain endothelial cells. Journal of Immunology. 2012;189:3130-9.[25] Fazeli M, Zarkesh E, H, Wu Z, Maamra M, Bidlingmaier M, Pockley A, et al. Identification of a monoclonal antibody against the leptin receptor that acts as an antagonist and blocks human monocyte and T cell activation. Journal of Immunological methods. 2006;312:190-200.[26] Pusztaszeri M, Seelentag W, Bosman F. Immunohistochemical expression of endothelial markers CD31, CD34, von Willebrand factor, and Fli-1 in normal human tissues. Journal of histochemistry and cytochemistry. 2006;54:385-95.[27] Yamamoto K, de Waard V, Fearns C, Loskutoff D. Tissue distribution and regulation of murine von willerbrand factor gene expression in vivo. Blood. 1998;92:2791-801.[28] Furuse M, Itoh M, Hirase T, Nagafuchi A, Yonemura S, Tsukita S. Direct association of occludin with ZO-1 and its possible involvement in the localization of occludin at tight junctions. J Cell Biol. 1994;127:1617-26.[29] Feldman G, Mullin J, Ryan M. Occludin: structure, function and regulation. Advance Drug Delivery Review. 2005;57:883-917.[30] Weksler B, Romero I, Couraud P-O. The hCMEC/D3 cell line as a model of the human blood brain barrier. Fluids and Barriers of the CNS. 2013;10:1-10.[31] Forster C, Burek M, Romero I, Weksler B, Couraud P, Drenckhahn D. Differential effects of hydrocortisone and TNFalpha on tight junction proteins in an in vitro model of the human blood-brain barrier. Journal of physiology. 2008;586:1937-49.[32] Guerra B, Santana A, Fuentes T, Delgado-Guerra S, Cabrera-Socorro A, Dorado C, et al. Leptin receptors in human skeletal muscle. Journal of applied physiology. 2007;102:1786-92.[33] Haniu M, Arakawa T, Bures E, Young Y, hui J, Rohde M, et al. Human leptin receptor. Determination of disulfide structure and N-glycosylation sites of the extracellular domain. Journal Biological Chemistry. 1998;273:28691-9.[34] Liu C, Liu X, Barry G, Ling N, Maki R, De Souza E. Expression and characterization of a putative high affinity human soluble leptin receptor. Endocrinology. 1997;138:3548-54.[35] Baskin D, Schwartz M, Seeley R, Woods S, Porte D, Breininger J, et al. Leptin receptor long-form splice-variant protein expression in neuron cell bodies of the brain and co-localization with neuropeptide Y mRNA in the arcuate nucleus. Journal of histochemistry and cytochemistry. 1999;47:353-62.[36] Ghilardi N, Ziegler S, Wiestner A, Stoffel R, Heim M, Skoda R. Defective STAT signaling by the leptin receptor in diabetic mice. PNAS. 1996;93:6231-5.[37] Vaisse C, Halaas J, Horvath C, Darnell J, Stoffel R, Friedman J. Leptin activation of stat3 in the hypothalamus of wild-type and ob/ob mice but not db/db mice. Nature Genetics. 1996;14:95-7.[38] Chow V, M P. Measurement of serum leptin concentrations in university undergraduates by competitive ELISA reveals correlations with body mass index and sex. Advances in physiology education. 2003;27:8.[39] Jayanthi L, Samuvel D, Ramamoorthy S. Regulated internalization and phosphorylation of the native norepinephrine transporter in response to phorbol esters. Evidence for licalization in lipid rafts and lipid raft-mediated internalization. Journal Biological Chemistry. 2004;279:12.[40] Jiao C, Delaroche D, Burlina F, Alves I, Chassaing G, Sagan G. Translocation and endocytosis for cell-penetrating peptide internalization. Journal Biological Chemistry. 2009;284:9.[41] Yaron P, Holt B, Short P, Losche M, Islam M, Dahl K. Single wall carbon nanotubes enter cells by endocytosis and not membrane penetration. Journal Nanobiotechnology. 2011;9:15.[42] Blitzer J, Nusse R. A critical role for endocytosis in Wnt signaling. BMC cell biology. 2006;7.

[43] Drin G, Cottin S, Blanc E, Rees A, Temsamani J. Studies on the internalization mechanism of cationic cell-penetrating peptides. Journal Biological Chemistry. 2003;278:10.[44] Gao H, Yang Z, Zhang S, Cao S, Shen S, Pang Z, et al. Ligand modified nanoparticles increases cell uptake, alters endocytosis and elevates glioma distribution and internalization. Scientific Reports. 2013;3.[45] Scarmato P, Durand G, Agneray J, Feger J. Inhibitory effect of sodium arsenite and azide on asialglycoprotein receptor mediated endocytosis in suspended rat hepatocytes. Biol Cell. 1986;56:4.[46] Sato K, Nagai J, Mitsui N, Ryoko Y, Takano M. Effects of endocytosis inhibitors on internalization of human IgG by Caco-2 human intestinal epithelial cells. Life Sciences. 2009;85:8.[47] Carpenter B, Hemsworth G, Wu Z, Maamra M, Strasburger C, Ross R, et al. Structure of the human obesity receptor leptin-binding domain reveals the mechanism of leptin antagonism by a monoclonal antibody. Structure. 2012;20:487-97.[48] Dietrich M, Spuch C, Anteguera D, Rodal I, de Yebenes J, Molina J, et al. Megalin mediates the transport of leptin across the blood-CSF barrier. Neurobiology of Aging. 2008;29:902-12.[49] Hama H, Saito A, Takeda T, Tanuma A, Xie Y, Sato K, et al. Evidence indicating that renal tubular metabolism of leptin is mediated by megalin but not by the leptin receptors. Endocrinology. 2004;145:3935-40.[50] Halaas J, Gajiwala K, Maffei M, Cohen S, Chait B, Rabinowitz D, et al. Weight-reducing effects of the plasma protein encoded by the obese gene. Science. 1995;269:543-6.[51] Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman J. Positional cloning of the mouse obese gene and its human homologue. Nature. 1994;372:425-32.[52] Paz-Filho G, Wong M, Licinio J. Ten years of leptin replacement therapy. Obesity Reviews. 2011;12:e315-23.[53] Heymsfield S, Greenberg A, Fujioka K, Dixon R, Kushner R, Hunt T, et al. Recombinant leptin for weight loss in obese and lean adults: a randomized, controlled, dose-escalation trial. JAMA. 1999;282:1568-75.[54] Triguero D, Buciak J, Pardridge W. Capillary depletion method for quantification of blood-brain barrier transport of circulating peptides and plasma proteins. Journal of Neurochemistry. 1990;54:1882-8.[55] Banks W, Clever C, Farrell C. Partial saturation and regional variation in the blood-to-brain transport of leptin in normal weight mice. American Journal of Physiology Endocrinology and Metabolism. 2000;278:e1158-65.[56] Pan W, Kastin A. Dirunal variation of leptin entry from blood to brain involving partial saturation of the transport system. Life Sciences. 2001;68:2705-14.[57] Hsuchou H, Kastin A, Tu H, Abbott J, Couraud P, Pan W. Role of astrocytic leptin receptor subtypes on leptin permeation across hCMEC/D3 human brain endothelial cells. Journal of Neurochemistry. 2010;115:11.[58] Friedman J, Halaas J. Leptin and the regulation of body weight in mammals. Nature. 1998;395:763-70.[59] Uotani S, Bjorbaek C, Tornoe J, Flier J. Functional properties of leptin receptor isoforms: internalization and degradation of leptin and ligand-induced receptor downregulation. Diabetes. 1999;48:279-86.[60] Lundin A, Rondahl H, Walum E, Wilcke M. Expression and intracellular localization of leptin receptor long isoform-GFP chimera. Biochimica et biophysica acta. 2000;499:130-8.[61] Barrett G NT, Trieu J. Leptin-derived peptides that stimulate food intake and increase body weight following peripheral administration. Regulatory peptides. 2011;170:24-30.[62] Barrett G TJ, Naim T. The identification of leptin-derived peptides that are taken up by the brain. Regulatory peptides. 2009;155:55-61.

Figure legends

Table 1. Antibody details

Figure 1. Establishment of a suitable in vitro human BBB model to examine endothelial ObR-mediated leptin transcytosis. The formation of an endothelial monolayer with BBB characteristics by hCMEC/D3 cells was examined by detection of the cerebral endothelial marker protein vWF (a, red stain = vWF; blue stain = DAPI), and the tight-junction proteins ZO-1 (b, green stain) and occludin (c). In addition, appropriate tight junction architecture between monolayer cells was confirmed through bright field transmission electron microscopy (d, tight junction is depicted between the black arrows). Furthermore, paracellular permeability was examined through quantification of the permeability coefficient to 70 kDa (e) and 4 kDa (f) FITC-dextran, as well as quantification of transendothelial electrical resistance (TEER) (g). The expression of ObR protein was detected through immunocytochemical staining of hCMEC/D3 cells fixed on glass coverslips (h, upper panels; red stain = ObR, blue stain = DAPI). Omission of primary antibody served as negative control to confirm specificity of ObR immunolabelling (h, lower panels). ObR expression was further confirmed through immunolabelling of Western blots from hCMEC/D3 lysates (i). Results are displayed as mean + SEM of three independent experiments. Images are representative of three independent experiments, with the exception of (c) which represents two independent experiments.

Figure 2. Leptin signalling and transcytosis across hCMEC/D3 endothelial monolayers. The interaction between leptin and hCMEC/D3 cells was examined by quantifying STAT-3 phosphorylation following leptin treatment (5 minutes) (a, optical density quantified in ‘b’). The translocation rate of leptin (4.65 nM apical concentration) or 4 kDa FITC-dextran (1 mg/mL apical concentration) across hCMEC/D3 monolayers was measured by quantifying leptin or dextran basolateral concentration at 30 or 10 min intervals for 120 or 40 minutes, respectively (c). The translocation rates across empty transwells were subsequently used to calculate permeability coefficients for leptin and 4 kDa across hCMEC/D3 monolayers per se (d). The energy-dependence of leptin translocation was examined by quantifying basolateral leptin concentration following a 90 min incubation at 37oC or 4oC (e). To ensure lower temperatures did not affect monolayer permeability, the rate of transport of 4kDa dextran was quantified following the 90 min incubation at 37oC or 4oC (f). The translocation of leptin across empty transwells was similarly quantified at 37 oC or 4oC (g). The role of endocytosis in leptin translocation was examined by quantifying basolateral leptin concentration following a 90 min incubation with or without sodium azide (NaN3, 5 mM) (h). Basolateral leptin concentration was normalized against the translocation of 4kDa FITC-dexran to compensate for barrier-opening effects of NaN3 (i). Image in (a) is representative of three independent experiments. Results are displayed as mean + SEM of three independent experiments. ** , *** indicate p < 0.01, 0.005, respectively, as determined by an unpaired student t-test.

Figure 3. Effect of an ObR neutralizing antibody on ObR-leptin conjugation and leptin transcytosis across hCMEC/D3 monolayers. The ability of the ObR neutralizing antibody 9F8 to prevent human ObR-human leptin binding was examined through an ObR-coated plate binding assay incubating leptin (4.65 nM) with increasing concentrations of 9F8 (a). To ensure prevention of leptin binding by 9F8 was specific for ObR-leptin interaction, the plate binding assay was repeated coating the plate with an anti-leptin capture antibody (b). In order to examine 9F8 interaction with hCMEC/D3-expressed ObR, Western blots from hCMEC/D3 lysates were immunoblotted with 9F8 (c). Similarly, prevention of leptin-induced ObR signalling on hCMEC/D3 by 9F8 was examined by quantifying STAT-3 phosphorylation following leptin treatment (46.5 nM, 5 mins) with or without 9F8 (100 g/mL) pre-incubation by Western blotting (d, optical density quantified in ‘e’). The effect of 9F8 treatment on leptin translocation across hCMEC/D3 monolayers was examined by quantifying basolateral leptin following a 90 min incubation leptin (4.65 nM apical concentration) with or without increasing concentrations of 9F8 (f). The effect of 9F8 treatment on barrier permeability was examined by measuring the translocation rate of 4kDa FITC-dextran following the 90 min incubation (g). The expression of LRP-2 by hCMEC/D3 cells was examined by immunostaining Western blots derived from hCMEC/D3 cell lysates with an anti-LRP-2 antibody (h). Results are displayed as mean + SEM of three independent experiments. Images in (c) and (d) are representative of three independent experiments. Images in (h) are representative of two independent experiments. *, *** denote p < 0.01, 0.005, respectively vs. 9F8 control; #, ### denote p < 0.05, 0.005, respectively, vs. leptin control, as determined by a one-way ANOVA with Tukey’s post-hoc test or a student t-test (e).

Figure 1)

a) b) c)

d)

e) f) g)

h) i)

Figure 2)

a) b) c)

d) e) f)

g) h) i)

STAT-3~80 kDa

Figure 3)

a) b)

c) d) e)

f) g) h)