Research ArticleCELL BIOLOGY

On the shuttling across the blood-brain barrier via tubule formation: Mechanism and cargo avidity bias

See allHide authors and affiliations

Science Advances  27 Nov 2020:
Vol. 6, no. 48, eabc4397
DOI: 10.1126/sciadv.abc4397
  • Fig. 1 LRP1 intracellular mapping.

    Expression levels of LRP1 in BECs assessed by WB (A) and immunofluorescence (green) with cell nuclei counterstained with DAPI (blue) (B). 3D renderings of BECs with both DNA (DAPI in blue) and LRP1 [anti–immunoglobulin G (IgG) in green] labeled shown as top view (C) and projection (D). (E) PLA between LRP1 and several intracellular proteins associated with endocytosis and trafficking reported as number of PLA events per cell and the total PLA signal per cell. 3D rendering of BECs showing PLA events between LRP1 and clathrin (F), Rab5 (G), actin (H), and syndapin-2 (I).

  • Fig. 2 Ligand avidity versus BBB crossing.

    (A) Heatmap showing the experimental measurement of % of AL-P crossing as a function of incubation time and ligand number per particle (L). (B) Ex vivo fluorescent photographs of whole murine brains imaged 2 hours after intravenous injection of PBS, pristine POs (L = 0), free angiopep-2 peptide (L = 1), A22-P, or A110-P. Violin plots showing the quantification in the brain parenchyma of the various preparations tested. **P < 0.01, ***P < 0.001, and ****P < 0.0001, one-way ANOVA (n = 6). (C) Concentration of angiopep-2 functionalized cargo expressed as percentage of injected dose (% ID) per gram of tissue as a function of the number of ligands. (D) Heatmap of the apparent permeability, P, obtained from agent-based simulations as a function of the ligand number per particle and the single ligand dissociation constant, Kd, with the LRP1 receptor. (E) Comparison between apparent permeability, P, across BBB experimental data (red markers and solid line) and simulation (blue markers and dashed lines) calculated for two different receptor densities and single ligand dissociation constant, Kd = 300 nM. Note that the control pristine PO apparent permeability was subtracted to the other formulations to remove passive diffusion. (F) Phase diagram showing different regimes of nanoparticle aggregation across the receptor densities and nanoparticle-receptor affinities expressed in kBT (with kB being the Boltzmann constant and T the temperature) as observed in MD simulations. Nanoparticle distributions are illustrated MD simulations using a coarse-grained membrane surface patch.

  • Fig. 3 LRP1 subcellular localization and expression as a function of avidity.

    (A) Deviation of the number of proximity events measured by a PLA between untreated endothelial cells and treated for 0.25, 1, and 2 hours of incubation with free angiopep-2 peptide, L = 1, and AL-P, with L = 22 and L = 110. Note that zero corresponds to no variation, while positive and negative values indicate up- and down-regulation, respectively. (B) Ratio between LRP1/Rab5 and LRP1/syndapin-2 number of proximity events for the different treatments with free angiopep-2 peptide, L = 1, and AL-P, where L = 22 and L = 110, with Rab5/syndapin-2 being 10 for the untreated cells. (C) WB measuring the LRP1 expression relative to the untreated cells for free angiopep-2 peptide, L = 1, and AL-P, with L = 22 and L = 110 measured at different incubation times with 0.25, 1, and 2 hours. *P < 0.05, **P < 0.01, and ***P < 0.001, one-way ANOVA (n = 6). Note that LRP1 expression is normalized to the loading control. (D) Diagram showing the syndapin-2–mediated transcellular route and the intracellular degradation of LRP1.

  • Fig. 4 Syndapin-2–mediated transport.

    3D rendering of confocal laser scanning micrographs of polarized BECs incubated with A22-P (red). (A) Cell nuclei were stained with DNA binding DAPI (blue), and syndapin-2 is shown in green (anti-IgG). (B) Fluorescence photograph of ex vivo whole mouse brains imaged 30 min after intravenous injection of A22-P and pristine POs loaded with PtA2. Pt tissue concentration in brain, kidney, lung, spleen, liver, and heart expressed as microgram per gram and measured by ICP-MS. (C) Tissues were collected 30 min after intravenous injection of A22-P and pristine POs both loaded with PtA2. STED micrographs of coronal brain sections showing the distribution of PtA2-loaded A22-P (red) 30 min after injection with capillary stained by lectin (green). (D) Two different regions of interests (ROIs) show the detail of the tubulation across the BECs. (E) 3D renderings as projections of STED micrographs of brain capillary (lectin in green) showing the detail of PtA2-loaded A22-P (red) and syndapin-2 (anti-IgG stained in blue). (F) Details of the tubule formed by the PtA2-loaded A22-P (red) surrounded by syndapin-2 (anti-IgG stained in blue).

  • Fig. 5 4D microscopy of transcytosis.

    (A) 3D rendering at three different times extracted from 4D (xyzt) live imaging of BECs stained for cellular membrane CellMask (green) and nuclei (blue) and incubated with Cy5-labeled A22-P (red). (B) 3D renderings of the same cell, with each event color-coded by its occurrence within periods of 144 s. (C) Graphs showing the red channel fluorescence intensity and the number of events (threshold in the red channel) as a function of time and zeta-averaged across the full cell thickness. a.u., arbitrary units. (D) Each event radius and length is monitored over time, and the average values across 20 events are plotted as a function of time. (E) The corresponding 3D renderings of the single events show an evolution from few puncta to large clusters, to membrane-bound tubulations, to tubular carriers. (F) Two sequences of 3D renderings extracted by fast 4D videos of the tubular carriers filled up with Cy5-labeled A22-P (red) crossing from one side to other BECs; note the cell membrane is stained by CellMask (green). (G) Normalized MSD as a function of normalized time τ*=tτcrossing, where τcrossing is the time each event takes to fully cross from apical to basal and vice versa. Distribution of τclustering (H), τtubulation (I), and τfission (J) measured from the graph in (D) and τcrossing (K) measured from the graph in (G).

  • Fig. 6 Super-resolution imaging of the tubular carriers.

    (A) 3D renderings shown as top and side views color-coded as a function of the depth (i.e., the distance from apical to basal) of optical sections of BECs incubated with PtA2-loaded A22-P. The 3D rendering was captured at a different time, and the normalized fluorescence measured across each section is plotted as a function of time. (B) The 3D rendering at 120 s is shown enlarged as top and side views and the arrows point at single A22-P particles, while the structure emerged as a network of tubules. (C) Close-up detail of top and side view of the same 3D rendering shows the evolution of the single tubule from apical to basal membrane showing the different stages of tubulation, fission from the apical membrane, and transport and fusion to the basal membrane. The same tubulations were observed in MD simulations. Here, the anisotropic growth of the membrane emerges from the collapse of a tubular aggregate of a particle on the surface. The membrane buckling can occur in different ways depending on the cluster size of assembled nanoparticles, leading to the formation of tubules (D), short tubes (E), or multiple assembly tubules (F). (G) The final tubule can thus be one, two, or three nanoparticles (NP) thick.

Supplementary Materials

Stay Connected to Science Advances

Navigate This Article