Research ReportVEGF increases blood–brain barrier permeability to Evans blue dye and tetanus toxin fragment C but not adeno-associated virus in ALS mice
Introduction
Most drugs fail to enter the CNS because of the blood–brain barrier (BBB). This restriction particularly affects drugs that are not substrates for active transport into the CNS, hydrophilic molecules larger than 500 Da, and high molecular weight therapeutic modalities such as monoclonal antibodies, antisense oligonucleotides, viral vectors, stem cells, and nanoparticles (de Boer and Gaillard, 2007). The BBB is a physiological entity that is a consequence of the relatively poor permeability of endothelial tight junctions, the absence of endothelial fenestrae, a low rate of CNS endothelial cell endocytosis, and the presence of drug efflux pumps. Several strategies have been devised to bypass or permeate the BBB including administration of drugs directly into the CSF, osmotic disruption of the BBB by agents such as mannitol, enhancement of endocytosis/transcytosis through CNS endothelium (e.g. coupling of proteins to anti-transferrin receptor antibodies), or the use of agents that augment vascular permeability directly, such as papaverine (Bhattacharjee et al., 2001, Muldoon et al., 2007, Pardridge, 2007). To date, with important exceptions, these strategies have not been sufficiently effective to merit significant clinical application (Doolittle et al., 2000, Kroll and Neuwelt, 1998).
Recent studies have indicated that vascular endothelial growth factor (VEGF) may provide new opportunities for manipulating the permeability of the BBB. VEGF is a 34 to 46 kDa angiogenic and vasculogenic factor with five isoforms having 121, 145, 165, 189, and 206 amino acids; the 165 amino acid form is the most abundant in vivo (Ferrara et al., 2003). VEGF increases the permeability of peripheral and CNS vasculature by permeating endothelial cells of venules and capillaries (Mayhan, 1999, Roberts and Palade, 1995). Topical or intradermal application of this growth factor causes fenestrae formation in the skin and muscle capillary endothelium, resulting in leakage of solutes like Evans blue dye within 10 min of treatment (Roberts and Palade, 1995). Likewise, intracerebral injection of VEGF induces characteristic morphological changes in CNS endothelium including interendothelial gaps, fragmentation of endothelium with formation of segmental, fenestrae-like narrowings, degenerative changes of the vascular basement membrane, and the appearance of fibrin gel in the vessel lumen (Dobrogowska et al., 1998). These structural alterations were associated with transient impairment of BBB function as detected by leakage of plasma albumin into the CNS. This begins 10 min after VEGF injection and continues for up to 24 h (Dobrogowska et al., 1998).
Endothelial cells within the BBB adapt to different maturational stages and pathological conditions by altering levels of VEGF expression. Although VEGF is expressed at high levels in endothelial cells during early embryogenesis, inhibition of VEGF is a crucial step in the formation of the BBB. This inhibition continues throughout the adult life and is reflected by low CNS endothelial cell VEGF levels (Darland and D'Amore, 1999, Lee et al., 2003). However, in some pathological conditions associated with increased leakiness of the BBB (e.g., cerebral ischemia, brain injury, and brain tumors), VEGF expression is elevated to embryonic levels (Marti et al., 2000, Nag et al., 1997, Zhang et al., 2002). It is likely that the elevated VEGF levels contribute directly to the increased BBB permeability in these disorders, as inhibition of VEGF signaling reverses the phenomenon (van Bruggen et al., 1999). A broader implication is that VEGF might be a useful reagent for achieving controlled, reversible pharmacological manipulation of BBB permeability. To test this hypothesis, we evaluated the effect of VEGF on BBB permeability in a mouse model of amyotrophic lateral sclerosis (ALS), a neurodegenerative motor neuron disease that leads to rapid paralysis and death. It is important to note in here that there is some degree of morphological damage to the BBB in ALS mice (Garbuzova-Davis et al., 2007). Therefore, interpretation of our results should be limited to ALS mice. Compounds tested in this study were selected based on the potential therapeutic modalities in ALS patients and included Evans blue dye (a protein-bound small molecule to represent small molecules that cannot pass through the BBB due to their high affinity plasma protein binding property), a non-toxic fragment of tetanus toxin (as a representative peptide), and adeno-associated viral particles. We found that pretreatment with VEGF increased the permeability of BBB to Evans blue dye and the peptide but not to the viral vector. These results suggest that the effectiveness of VEGF for pharmacological opening of BBB depends on the size of the delivered substance.
Section snippets
Effect of VEGF pretreatment on CNS distribution of Evans blue dye
Two hours after intraperitoneal injection of Evans blue dye (2%, 4 ml/kg), the dye was not detected in brain or spinal cord tissues of control animals (Figs. 1 and 2a). By contrast, 2 h after pretreatment with VEGF (20 μg/200 μl; i.v.), blue staining of the spinal cord sections was clearly evident (Fig. 2b). Quantification of fluorescence showed an ample penetration of Evans blue dye into the cerebrum (341.66 ± 410.56 ng/mg tissue), brainstem (840.54 ± 831.94 ng/mg tissue), and spinal cord (413.49 ±
Discussion
Using transgenic ALS mice, we have examined the influence of VEGF-induced alterations in BBB on the CNS permeability of three agents of distinctly different sizes. Each of these agents represents a potential therapeutic modality for the disorders of the CNS. In each instance, we have measured the CNS tissue concentration or enzyme activity level of the systemically administered agents in control and VEGF-pretreated animals. Our findings suggest that pretreatment with VEGF increases the
Animals
Male and female mice (70–90 days old; 20–23 g) expressing transgenic, mutant human superoxide dismutase 1 (SOD1) (G93A or SOD1G93A) were obtained from Jackson Laboratories (Bar Harbor, ME, USA). A total of 60 mice were included in this study; 31 received vehicle injection and 29 were injected with VEGF. A breakdown of the number of animals is as follows: Evans blue dye quantification: control n = 10 (two sets of experiments using 5 mice in each), VEGF n = 8 (two sets of experiments using 4 mice in
Acknowledgments
We thank Drs. Michele M. Maxwell and Zhanyun Fan for their assistance with these experiments. Robert H. Brown, Jr receives support from the NIA and NINDS, Project ALS, the ALS Association, the Angel Fund, the Al-Athel ALS Research Foundation, and the Pierre L. deBourghknect ALS Research Foundation. Jonathan W. Francis is supported by NINDS and the Muscular Dystrophy Association.
References (32)
- et al.
Quantitative analysis of papaverine-mediated blood–brain barrier disruption in rats
Biochem. Biophys. Res. Commun.
(2001) - et al.
Recombinant AAV viral vectors pseudotyped with viral capsids from serotypes 1, 2, and 5 display differential efficiency and cell tropism after delivery to different regions of the central nervous system
Mol. Ther.
(2004) - et al.
Enhanced CNS uptake of systemically administered proteins through conjugation with tetanus C-fragment
J. Neurol. Sci.
(1990) - et al.
Ultrastructure of blood–brain barrier and blood–spinal cord barrier in SOD1 mice modeling ALS
Brain Res.
(2007) - et al.
Hypoxia-induced vascular endothelial growth factor expression precedes neovascularization after cerebral ischemia
Am. J. Pathol.
(2000) - et al.
VEGF disrupts the neonatal blood–brain barrier and increases life span after non-ablative BMT in a murine model of congenital neurodegeneration caused by a lysosomal enzyme deficiency
Exp. Neurol.
(2004) - et al.
Tetanus toxin fragment C fusion facilitates protein delivery to CNS neurons from cerebrospinal fluid in mice
J. Neurochem.
(2005) - et al.
An antigenic polypeptide fragment isolated from tetanus toxin: chemical characterization, binding to gangliosides and retrograde axonal transport in various neuron systems
J. Neurochem.
(1977) - et al.
Effects of VEGF and nitric oxide synthase inhibition on blood–brain barrier disruption in the ischemic and non-ischemic cerebral cortex
Neurol. Res.
(2005) - et al.
Blood vessel maturation: vascular development comes of age
J. Clin. Invest.
(1999)