The transporter and permeability interactions of asymmetric dimethylarginine (ADMA) and L-arginine with the human blood-brain barrier in vitro

Abstract

The blood-brain barrier (BBB) is a biological firewall that carefully regulates the cerebral microenvironment by acting as a physical, metabolic and transport barrier. This selectively permeable interface was modelled using the immortalised human cerebral microvascular endothelial cell line (hCMEC/D3) to investigate interactions with the cationic amino acid (CAA) L-arginine, the precursor for nitric oxide (NO), and with asymmetric dimethylarginine (ADMA), an endogenously derived analogue of L-arginine that potently inhibits NO production. The transport mechanisms utilised by L-arginine are known but they are not fully understood for ADMA, particularly at the BBB. This is of clinical significance giving the emerging role of ADMA in many brain and cerebrovascular diseases and its potential as a therapeutic target. We discovered that high concentrations of ADMA could induce endothelial dysfunction in the hCMEC/D3s BBB permeability model, leading to an increase in paracellular permeability to the paracellular marker FITC-dextran (40kDa). We also investigated interactions of ADMA with a variety of transport mechanisms, comparing the data with L-arginine interactions. Both molecules are able to utilise the CAA transport system y(+). Furthermore, the expression of CAT-1, the best known protein from this group, was confirmed in the hCMEC/D3s. It is likely that influx systems, such as y(+)L and b(0,+), have an important physiological role in ADMA transport at the BBB. These data are not only important with regards to the brain, but apply to other microvascular endothelia where ADMA is a major area of investigation.

Keywords: Cationic amino acid transporter 1; Cerebral microvasculature; Guanosine triphosphatases (GTPases); Nitric oxide.

Fig. 1

Accumulation of [³H]L-arginine in hCMEC/D3 cells. (A) To understand the roles CAA transporters played in the transport and subsequent accumulation of [³H]L-arginine, self-inhibition and the addition of known AA transporter interacting molecules was performed in an accumulation model using confluent hCMEC/D3s. Control L-arginine values were compared to treatments. ***p<0.001. (B) To investigate the competitive effects of unlabelled ADMA on [³H]L-arginine accumulation, 0.5 µM, 3 µM and 500 µM ADMA were added individually to accumulation buffer. ***p<0.001. All [³H]L-arginine data are corrected for [¹⁴C]sucrose and protein content (with [¹⁴C]sucrose corrected for protein only) and expressed as means±SEM, n=6 plates with 6 replicates per plate.

Fig. 2

Accumulation of [³H]ADMA in hCMEC/D3 cells. (A) To investigate self and competitive inhibition on [³H]ADMA transport and accumulation, 0.5 µM, 3 µM and 500 µM unlabelled ADMA and 100 µM L-arginine were added individually to the accumulation buffer. *p<0.05,***p<0.001. (B) To understand the roles CAA transporters played in the transport and subsequent accumulation of [³H]ADMA, the addition of known AA transporter interacting molecules was performed in an accumulation model using confluent hCMEC/D3s. *p<0.05,**p<0.01. All [³H]ADMA data are corrected for protein content and [¹⁴C]sucrose (with [¹⁴C]sucrose corrected for protein only) and expressed as means±SEM, n=5 (A) or 4–6 (B) plates with 6 replicates per plate.

Fig. 3

Expression of CAT-1 in hCMEC/D3 cells. (A) SDS-PAGE and WB analysis revealed CAT-1 expression in hCMEC/D3 (P28) and wild type MCF7 whole cell lysate lysed in TGN lysis buffer, as described in Section 4.9. (B) CAT-1 expression was also demonstrated by IF performed with hCMEC/D3 cells (P28) grown on rat tail collagen type 1-coated coverslips, fixed with 4% formaldehyde and stained with primary and secondary antibody, and viewed at 63× with oil emersion using a Zeiss LSM710 confocal microscope and image analysis software Zen 2009 as described in Section 4.9. This localization appears to be on the membrane, similar to the results seen in other studies (Closs et al., 2004). Scale bar, 10 µm. Cell nuclei were counterstained with 1 µg/ml DAPI. For negative staining, cells were stained with secondary antibody only along with DAPI (inserts).

Fig. 4

Impact of ADMA, L-NIO, TNF-α and IFN-γ on hCMEC/D3 permeability to 40 kDa FITC-Dex. (A) Confluent monolayers of hCMEC/D3 cells, grown on rat tail collagen type 1-coated transwell filter inserts for 8–10 days, were incubated for 24 hr with a range of ADMA concentrations and their permeability to 40 kDa FITC-Dex was investigated and compared to control (untreated) cells. (B) Confluent monolayers of hCMEC/D3 cells, grown on rat tail collagen type 1-coated transwell filter inserts for 8–10 days, were incubated for 24 h with either 5 µM L-NIO or 5 ng/ml TNF-α & 5 ng/ml IFN-γ, and their permeability to 40 kDa FITC-Dex was investigated and compared to control (untreated) cells. Data represent means±SEM from 12 to 15 transwell filter inserts per treatment. Data analysed with one-way ANOVA and Bonferroni’s post hoc test.

Fig. 5

Cytotoxicity of permeability treatments on hCMEC/D3s. The pharmacological treatments used were assessed for any cytotoxic potential using an MTT assay with confluent monolayers of hCMEC/D3 cells in 96 well plates, grown in permeability medium as described in Section 4.3. The results are expressed as percentage viability±SEM and compared to control untreated cells, which were incubated in permeability medium alone. TNF-α and IFN-γ were used together at 5 ng/ml each. N=3 plates, with 6 replicates per plate.

Fig. 6

Formation of ROS in hCMEC/D3 cells. hCMEC/D3 cells were treated with increasing doses of ADMA for 24 h before assessment of ROS levels with DHE. As a positive control, some wells were treated with 300 µM H₂O₂ alone for 30 min before measuring ROS levels. ADMA treatment had no significant effect on ROS levels in hCMEC/D3 cells, while exposure to H₂O₂ significantly increased ROS in these cells (one way ANOVA,*p<0.05).