Altered blood-brain barrier transport of nanotherapeutics in lysosomal storage diseases

Abstract

Treatment of neurological lysosomal storage disorders (LSDs) are limited because of impermeability of the blood-brain barrier (BBB) to macromolecules. Nanoformulations targeting BBB transcytosis are being explored, but the status of these routes in LSDs is unknown. We studied nanocarriers (NCs) targeted to the transferrin receptor (TfR), ganglioside GM1 or ICAM1, associated to the clathrin, caveolar or cell adhesion molecule (CAM) routes, respectively. We used brain endothelial cells and mouse models of acid sphingomyelinase-deficient Niemann Pick disease (NPD), and postmortem LSD patients' brains, all compared to respective controls. NC transcytosis across brain endothelial cells and brain distribution in mice were affected, yet through different mechanisms. Reduced TfR and clathrin expression were found, along with decreased transcytosis in cells and mouse brain distribution. Caveolin-1 expression and GM1 transcytosis were also reduced, yet increased GM1 levels seemed to compensate, providing similar NC brain distribution in NPD vs. control mice. A tendency to lower NHE-1 levels was seen, but highly increased ICAM1 expression in cells and human brains correlated with increased transcytosis and brain distribution in mice. Thus, transcytosis-related alterations in NPD and likely other LSDs may impact therapeutic access to the brain, illustrating the need for these mechanistic studies.

Keywords: Blood-brain barrier; Lysosomal storage disorders; Neurological diseases; Targeted nanocarriers; Transcytosis pathways.

Figure 1:. Biodistribution of targeted NCs in control and ASM KO mice.

(A) Circulation at the indicated times and (B, C) brain biodistribution 30 min after i.v. injection in wildtype (Control) or ASM KO mice of model polystyrene NCs coated with tracer amounts of ¹²⁵I-IgG along with anti-TfR, anti-GM1 or anti-ICAM1. Weight and ¹²⁵I content were determined after sacrifice to calculate: (A) the percentage of the injected dose (%ID) in the circulation of control mice, (B) the localization ratio (LR) calculated as %ID/g in brain : %ID/g in blood, and (C) the specificity index (SI) of the brain (brain LR for targeted NCs : brain LR for non-targeted IgG NCs) normalized to the SI of the liver (liver LR for targeted NCs : liver LR for non-targeted IgG NCs). Data are Mean ± SEM. *Compared to anti-TfR NCs, #compared to anti-GM1 NCs, ^$compared to control mice (p < 0.05 by Student’s t test).

Figure 2:. Targeting of NCs to control and NPD brain endothelial cell models.

Immunofluorescence micrographs (top panels) and quantification (bottom panels), showing interaction of PLGA NCs coated with anti-TfR, anti-GM1 or anti-ICAM1 with HBMECs grown on (A) coverslips or (B) Transwells, either untreated (Ctrl) or treated with 20 μM imipramine + 10 ng/ml TNμ (TNF+Imi) to mimic NPD. NC incubations were conducted for 30 min at 37 °C, after which cells were fixed, permeabilized, and incubated with respective FITC-labeled secondary antibodies (green) to detect antibody-coated NCs and DAPI (blue) to stain cell nuclei. Scale bar = 10 μm. Data expressed as mean ± SEM. *Comparison with control cells (p<0.05 by Student’s t-test).

Figure 3:. Brain endothelial cell barrier model.

(A) TEER measurements of HBMECs grown on transwell filters over a period of 7 days, either as untreated (Control) or treated with 20 μM imipramine on day 5 (Disease). (B) Confluent HBMECs stained for VE-cadherin (arrows). Scale bar = 10 μm. (C) HBMECs were incubated for 30 min at 37 °C with Texas Red dextran, ¹²⁵I-labeled anti-ICAM1 antibody alone or ¹²⁵I-labeled anti-ICAM1 PLGA NCs added to the apical chamber, after which both the apical and basolateral cell medium were collected. For dextran, fluorescence measurements were determined using a plate reader while for the antibody and antibody-coated NC, ¹²⁵I content was determined to calculate respective NC concentrations in both chambers to assess barrier function. Data expressed as mean ± SEM. *Comparison with the apical chamber (p<0.05 by Student’s t-test).

Figure 4.. NCs maintain their antibody coat during binding, uptake, and transcytosis across brain endothelial cells.

(A) Fluorescence microscopy of TNF-treated HBMECs incubated for 30 min with anti-ICAM1 followed by red-AlexaFluo555 secondary antibody. (B) Fluorescence microscopy of TNFα-activated HBMECs incubated for 30 min with green-fluorescent polystyrene NCs coated with anti-ICAM1 followed by red-AlexaFluor555 secondary antibody. (A,B) Fixed cells to allow cell-surface binding without uptake. (C) Same as (B) but in live cells to allow uptake. In this case, after staining cell-surface bound anti-ICAM1 antibody counterparts with blue AlexaFluor350 secondary antibody, cells were permeabilized and stained with red AlexaFluor555 to access both surface and internalized anti-ICAM1. Thus, in (C) anti-ICAM1 colocalizing with cell-surface NCs appears white (blue + red + green; open arrows) and anti-ICAM1 colocalizing with internalized NCs appears yellow (green + red = yellow, white arrowheads). (A-C) Scale bar = 10 μm. Dashed lines = cell borders viewed by bright field. (D) DLS measurements of the hydrodynamic diameter and PDI of non-coated polystyrene NCs, NCs coated with anti-ICAM1 added to the apical chamber above TNFα-treated HBMECs (AP input), the same NCs harvested from the basolateral chamber after 5 h incubation (BL output), or control cell medium (CM). Data expressed as mean ± SEM. *Coated NCs vs. non-coated NCs (p<0.05 by Student’s t-test).

Figure 5:. Transport of antibody coated NCs across control and NPD brain endothelial cell models.

HBMECs were grown on transwells either untreated (Control) or treated with 20 μM imipramine + 10 ng/mL TNμ (Disease), following which they were incubated for a 30 min binding pulse with PLGA NCs coated with ¹²⁵I-anti-TfR, ¹²⁵I-anti-GM1 or ¹²⁵I-anti-ICAM1. Non-bound NCs were then removed from the apical and basolateral chambers and incubations continued in NC-free medium. (A) Basolateral fractions were collected at 3, 8 and 24 h to determine the number of NCs using a gamma counter and corrected for free iodine arising by degradation and was expressed for diseased cells as % of the control cells. (B) Apparent permeability coefficient (Papp) calculated from the same data using the formula described in Materials and Methods. (C) The apical fraction of NCs targeted to each of the pathways was also collected at 2 h and quantified. Data expressed as mean ± SEM. *Comparison with control cells (p<0.05 by Student’s t-test).

Figure 6:. Levels of cell-surface markers to which NCs were targeted.

Flow cytometry analysis of cell-surface markers to which NCs had been targeted i.e. TfR, ganglioside GM1, and ICAM1 in HBMECs, either untreated (Control) or treated with 20 μM imipramine + 10 ng/ml TNFμ (disease) to mimic NPD. Solid black line = non-specific IgG. Solid colored lines = specific antibodies under control condition. Dashed colored lines = specific antibodies under disease condition. Quantification was done using FlowJo^® software and expressed taking into consideration both intensity levels and positive fraction of the cell population, as described in Materials and Methods. Data expressed as mean ± SEM. *Compared to control (p<0.05 by Student’s t-test).

Figure 7:. Expression of regulatory markers of endocytosis in NPD cell models.

(A) Representative Western blots of NHE-1 involved in the ICAM-1-mediated transport, clathrin heavy chain (CHC) involved in TfR-mediated transport, caveolin-1 (Cav-1) associated with caveolae, and housekeeping GAPDH in untreated HBMEC (control, Ctrl) or HBMEC treated with imipramine to mimic NPD (Disease). (B) Densitometric quantification of bands shown in (A) using ImageJ and normalized to respective housekeeping bands. The solid line indicates control condition. Data expressed as mean ± SEM. *Compared to control (p<0.05 by Student’s t-test).

Figure 8:. Expression of endocytic markers in human brains from LSD patients and non-LSD controls.

(A) Representative immunohistochemistry images (top) and fluorescence intensity quantification (bottom) of brain samples from LSD patients (disease) vs. age-matched non-LSD controls, probed for clathrin heavy chain (CHC; green channel), caveolin-1 (Cav-1; red channel), and ICAM-1 (red channel). The signal of cell nuclei, visible in the blue channel, was used to normalize the fluorescence observed in the green and red channels, and the resulting expression levels in disease samples were expressed as % of expression in control samples. Scale bar = 40 μm. (B) Representative protein bands from Western blotting and densitometric quantification of human brain lysates probed for ICAM-1 and housekeeping protein GAPDH. (A, B) Samples are described in Supplementary Table S1 C= non-LSD control; D= LSD. Data expressed as mean ± SEM (n=6). *Compared to control (p<0.05 by one-way ANOVA, Holm-Sidak post hoc test).