Figure 1.
The choroid plexus and the blood–CSF barrier.
A. The choroid plexuses are highly vascularised epithelial tissue that originates from the neuroependyma in the developing brain (ependyma in adult) and float freely in each of the brain ventricles (the two lateral, third and fourth). These specialised organs have two major functions: (i) to act as a diffusion barrier between the blood and the CSF; and (ii) they produce and secrete the CSF that fills the ventricles and subarachnoid spaces. B. The choroid plexuses are comprised of a central stroma with many blood vessels covered by a single layer of specialised epithelial cells, which rest on a thick basement membrane. C. The choroid plexuses are one of the circumventricular organs (i.e. they are positioned at sites around the margin of the ventricular system of the brain) and, like blood vessels in this region of the brain, the vessels of the choroid plexus stroma are fenestrated and the junctional strands linking adjacent endothelial cells are discontinuous. Accordingly, molecules are able to leave the blood vessels and enter the basement membrane of the plexus (arrows). However, a protective barrier is present in the choroid plexus, itself provided by tight, adherens, gap and other junctions between intimately apposed plexus epithelial cells – forming the blood–CSF barrier. The presence of many junctions between adjacent plexus epithelial cells from its first appearance in development means that the paracellular route for small molecules between blood and CSF is closed off in the embryo as well as in the adult. Abbreviations: BV, blood vessel; CP, choroid plexus; CPEC, choroid plexus epithelial cell; CSF, cerebrospinal fluid; LV, lateral ventricle; St, stroma; TJ, tight junction; Scale bar: 500 µm in A, 25 µm in B.
Table 1.
Primers (5′→3′) used for qPCR validation of targets from array dataset.
Table 2.
List and characterisation of antibodies.
Figure 2.
Gene expression in lateral ventricular choroid plexus of E15 and adult mice.
A. The data are presented in a dot plot on a logarithmic scale, where each point represents a separate probe set on the GeneChip. Only probe sets that hybridized are represented, with no absent probe sets present in the plot. Data presented in plot are representative of a single GeneChip experiment, however three biological replicates were completed. Probe sets lying on the diagonal axis indicate similar levels of expression at E15 and adult. Validation using qPCR was completed on a select number of targets (see Fig. 3 and 4). B. Pie chart of the percentage of gene ontology of 1803 genes with altered expression in mouse lateral ventricular choroid plexus between the two ages. Protein binding, receptors and ATP binding genes accounted for 50% of all genes altered during development.
Table 3.
Top 50 genes enriched in the embryonic (A) or adult (B) lateral ventricular choroid plexus of the mouse.
Table 4.
Tight junction and associated proteins enriched in mouse lateral ventricular choroid plexus.
Table 5.
Influx transporters and ion channels enriched in mouse lateral ventricular choroid plexus.
Table 6.
Expression and function of transporters in developing choroid plexus.
Figure 3.
qPCR validation of microarray gene targets.
Several members of the ABC family of efflux transporters were up-regulated in the embryo (A) and in the adult (B). The direction of these changes was in line with what was found using the Affymetrix GeneChip platform, though the magnitude of the expression was different. C. Three members of the vesicle-associated membrane protein (Vamp) family of proteins were down-regulated in the embryo (up-regulated in the adult).
Table 7.
ABC transport/drug efflux gene expression in mouse lateral ventricular choroid plexus.
Table 8.
Protein binding targets enriched in mouse lateral ventricular choroid plexus.
Figure 4.
Co-localisation of albumin and total plasma protein with SPARC, Glycophorin A or C.
A. Immunoreactivity in the lateral ventricular choroid plexus of E15 mouse embryo immunostained to detect endogenous albumin. Several epithelial cells of the plexus were deemed albumin-immunopositive (filled arrows), while all other cells were albumin-immunonegative (unfilled arrowheads). To be included in cell counts, the reaction product had to be present from basolateral to apical membranes and a cell nucleus had to be visible. B. Summary of cell count data from E15 and adult mouse lateral ventricular choroid plexus. The number of both total protein- and albumin-immunopositive plexus cells increased with age. As a percentage (numbers in bars), there was a decrease between E15 and adult, with E15 having the higher percentage of protein-positive cells of any age of the developing mouse. C. Quantitative PCR validation of microarray experiments showed that Sparc, along with Gypa and Gypc were up-regulated from about 20- to 30-fold in the embryo compared to the adult. D. Western blots of SPARC and GYPA in lateral ventricular choroid plexus and kidney (used as control) of E15 and adult mice. The level of SPARC did not change in kidney at the two ages: there was a substantial decrease in adult lateral ventricular choroid plexus compared to E15. A single dark staining band at the correct molecular weight (∼35 kDa, mouse SPARC MW 32 kDa) confirms specificity of the antibody. Levels of GYPA appeared to be higher in the adult kidney compared to E15. The highest level in the choroid plexus was in the embryo. A dark staining band at ∼40 kDa (mouse GYPA MW 43 kDa) confirms cross-reactivity of antibody. E. Single-cell PCR was performed on cells immunostained for either total protein or albumin. The percentage of immunopositive plexus cells that also displayed positive gene expression increased with age for both glycophorins, however it decreased for Sparc. Abbreviations: E, embryonic day; Gypa/GYPA, Glycophorin A; GYPC, Glycophorin C; LV CP, lateral ventricular choroid plexus; MW molecular weight (kDa).
Figure 5.
Cellular and subcellular distribution of GYPA immunoreactivity in E15 and adult choroid plexus.
A. Embryonic day 15 (E15) lateral ventricular choroid plexus. The more medial younger segment of the plexus emerging from the hem and hippocampal anlage is shown in the lower and right side of the figure. This part of the plexus shows only weak immunopositive staining for GYPA (filled arrowheads). At the tip of the plexus most epithelial cells exhibit a strong granular immunoreactivity in the apical cytoplasm (arrows) and an apical membrane-associated surface immunostaining (open arrowheads). Basolateral cell membranes (BM) are also immunopositive, as are vascular endothelial cells, which show a marked immunoreactivity (open arrows), in particular in areas with basolateral and apical epithelial membrane-associated immunostaining. Strong membrane immunostaining of erythrocytes (e) is also evident. B. Adult lateral ventricular choroid plexus. A very weak fine granular cytoplasmic immunoreactivity is seen in almost all choroid plexus epithelial cells in contrast to a stronger immunoreactivity found in few small segments of the plexus (framed area). Note the apparent lack of cytoplasmic immunostaining in the apical-most part of most epithelial cells (filled arrowheads), but a distinct immunostaining of small vesicles within the cytoplasm (arrows). Very few cells exhibit an apical membrane-associated immunoreactivity (open arrowheads). There was no immunoreactivity for GYPA in vascular endothelial cell membranes (open arrows). Scale bar: 10 µm in all.
Figure 6.
Cellular and subcellular distribution of SPARC immunoreactivity in E14, E16 and adult choroid plexus.
A. Embryonic day 14 (E14) lateral ventricular choroid plexus. The most lateral (older) segment of the plexus close to the tip is shown in the upper part of the figure. In this area small subsets of SPARC-immunopositive epithelial cells exhibit a fine granular immunoreactivity (open arrowheads) and a few strongly immunostained large ‘vacuoles’ (arrows) in contrast to neighboring epithelial cells without immunoreactivity (closed arrowheads). Note the absence of SPARC-immunostaining in perivascular spaces and in endothelial cell membranes (open arrow). B. E16 lateral ventricular choroid plexus. A segment of the plexus similar to that shown in A exhibits many strongly immunoreacting choroid plexus epithelial cells (open arrowheads). The patchy reaction product in these cells is associated with a tubulocisternal endoplasmic reticulum (TER). Note the immunostaining along the basolateral cell membranes (BM) in the strongly SPARC-immunopositive cells. A few large granules (arrows) are present in the majority of the epithelial cells including those showing low SPARC-immunoreactivity (filled arrowheads). SPARC-immunostaining is absent from perivascular spaces and endothelial cell membranes (open arrow). C. Adult lateral ventricular choroid plexus. Note the apparent lack of cytoplasmic immunoreactivity in the apical-most part of the epithelial cells, but the pronounced immunostaining of large ‘vacuoles’ (arrows) which represent a combination of multivescicular bodies and lysosomes. There was no immunoreactivity for SPARC either in perivascular spaces or in individual endothelial cell membranes. Scale bar: 10 µm in all.
Figure 7.
Proposed transepithelial pathway for albumin through choroid plexus epithelial cells.
A. Whole choroid plexus showing single layer of epithelial cells sitting on thick basement membrane (see also Fig. 1C). B. Illustration depicting the suggested routes of albumin from plasma into CSF across the choroid plexus epithelium. GYPA/C in the endothelial cells may deliver albumin to the basement membrane (1) from where it can be taken up into the plexus epithelium by GYPA/C or SPARC (2). From here albumin may travel along a SPARC-specific pathway through the tubulocisternal endoplasmic reticulum (3, and see C) and Golgi (4a), or via a VAMP-mediated pathway in vacuoles, lysosomes or multivesicular bodies (4b, and see panel C). On the apical surface of the plexus epithelium, GYPA/C may be involved in efflux of protein from the cell into the CSF of the ventricles (5), as validated by extensive GYPA immunoreactivity in embryonic plexus (Fig. 5A). In the adult, the lack of immunoreactivity in the endoplasmic reticulum and Golgi (see Fig. 5 and Fig. 6) along with increased expression of gene products for VAMP molecules (see Fig. 3C) suggest that the majority of transport possibly occurs via VAMP-mediated vesicular/lysosomal transport such as shown in (4b). C. Transmission electron micrograph of ultracryosection from E60 fetal sheep choroid plexus [87]. Immunolabelled human albumin 6 nm particles and sheep albumin 12 nm gold particles are shown to co-localise within the tubulocisternal endoplasmic reticulum. Abbreviations: CPEC, choroid plexus epithelial cell; CSF, cerebrospinal fluid; GYPA, glycophorin A; GYPC, glycophorin C; MVB, multivescicular body; TER, tubulocisternal endoplasmic reticulum. Scale bar: 0.2 µm in C.