On Further Development of Barrier Modulation as a Technique for Systemic Ocular Drug Delivery
Many systemically deliverable low-molecular-weight drugs with proven efficacy as neuroprotective, anti-inflammatory, anti-neovascular, cytotoxic, or anti-microbial agents are blocked from entering the eye by the inner blood-retina barrier. Recent reports from this laboratory have shown that it is possible to modulate the iBRB reversibly in mice by systemic administration of siRNA targeting transcripts derived from claudin 5, a tight junction protein of the endothelial cells lining the inner retinal microvasculature, rendering the iBRB transiently permeable to compounds up to approximately 1,000 Da in molecular weight. The system has been validated by demonstrating improved visual function in a model of autosomal recessive retinitis pigmentosa (IMPDH1−/− mice) and by suppression of light-induced damage to the retina, in each case by systemic drug delivery following barrier modulation. In this review, we explore how this technique could be improved by the down-regulation of transcripts encoding other tight junction endothelial proteins, particularly ones which would enable outer blood-retina barrier modulation.
KeywordsBlood-retinal barrier Drug delivery Tight junction Claudin-5 Barrier modulation Retinal pigmented epithelium IMPDH1
The use of the monoclonal antibody Lucentis® to treat neovascularization in wet AMD has been a huge milestone in the treatment of retinal degeneration. However, the blood-retinal barrier (BRB) remains the largest obstacle to the treatment of retinal conditions, with up to 98% of low-molecular-weight drugs not crossing the BRB (Pardridge 2007). The majority of therapeutics used to treat the retina currently, such as Lucentis®, are delivered via intraocular injection. This requires repeated visits to clinics, great discomfort for patients, as well as the risk of endophthalmitis, cataract, retinal detachment and toxic vitreitis (Ness et al. 2010).
The BRB is made up of two distinct parts – the inner and outer BRB. The iBRB is formed by cells of the retinal vascular endothelium, while the oBRB is formed by the cells of the retinal pigmented epithelium (RPE). In both cases, the crucial part of the barrier is the tight junction, a protein complex that spans the plasma membrane at the apical end of contacting endothelial or epithelial cells. These tight junction complexes limit paracellular transport between the cells of the iBRB and oBRB.
Tight junctions are made up of at least three different types of membrane-spanning proteins – occludin, members of the junctional adhesion molecule family of proteins and claudin proteins (Ben-Yosef et al. 2003), as well as many structural and transport proteins inside the cell, such as zona occludens proteins (Anderson and Van Itallie 2008).
21.2 Therapeutic Delivery Across the iBRB
This lab has recently reported that using systemically injected small interfering RNA (siRNA) targeting claudin-5, a tight junction component of the iBRB and not of the oBRB, the iBRB was modulated to allow for the selective passive diffusion of low-molecular-weight compounds from the blood to the retina, 24 and 48 h post injection of siRNA (Campbell et al. 2009). This technique was used to treat a number of animal models of retinopathies.
This technique was demonstrated to be size-selective as well as transient, with microperoxidase – molecular weight 1,881 Da – being maintained within retinal microvessels even while claudin-5 was suppressed, and levels of claudin-5 expression and full barrier strength were shown to return to normal 72 h post-siRNA injection. Concomitantly, no cell death was observed, and neuronal transcription profiles remained largely unchanged as a result of barrier modulation (Campbell et al. 2009).
Claudin-5 is expressed in the endothelial cells of the iBRB and not in RPE cells of the oBRB (Bai et al. 2008; Tachikawa et al. 2008), and as such, the delivery of therapeutics following claudin-5 suppression results in the delivery of the therapeutic across the iBRB alone. What remains to be investigated is whether there would be any advantage in the delivery of low-molecular-weight therapeutics across the oBRB alone, or both the iBRB and oBRB.
21.3 The oBRB and Its Potential in Barrier Modulation
The RPE that constitutes the oBRB is located at the interface between the photoreceptor outer segments on its apical side and Bruch membrane on its basolateral side, with the choriocapillaris beyond (Simo et al. 2010). One of the main functions of the RPE is the maintenance of the oBRB – important for the homeostatic microenvironment of the retina – as well as the transport of nutrients, ions, and water. However, unlike the retinal endothelial vasculature, the RPE also protects against photo-oxidation, phagocytoses shed photoreceptor outer segments, re-isomerises all-trans-retinal into 11-cis-retinal – a molecule necessary for photo-transduction, and secretes factors essential to the integrity of the retina, including immunosuppressive factors and growth factors (Simo et al. 2010). Indeed, transplanted RPE cells have been observed to slow the progression of retinal degeneration in animal models without forming a functional BRB (Litchfield et al. 1997).
Despite these differences in characteristics and anatomy, the oBRB has been investigated to a far lesser extent. The barrier selectivity of the RPE appears to be very similar to its endothelial counterpart (Steuer et al. 2005), despite an apparent difference in transepithelial resistance (Rizzolo 2007), and changes in RPE tight junction size selectivity appear to follow alterations in the levels of tight junction proteins such as claudins and occludin (Abe et al. 2003). One example where delivery across the oBRB could potentially be beneficial would be for the delivery of anti-angiogenic substances, such as sunitinib (molecular weight 400 Da), to conditions such as wet AMD where choroidal neovascularization is the central pathology. This is because it is the RPE which secretes VEGF across both its apical and basolateral membranes (Peng et al. 2010), and so hypothetically delivery of anti-angiogenic therapies here could maximise their effect.
Another potential benefit of oBRB modulation is improvement of the therapeutic window of a drug. 9-cis-retinyl acetate, for instance, has been demonstrated to partially rescue cone photoreceptor loss in an animal model of Leber’s congenital amaurosis, even when delivered by oral gavage or intraperitoneally (Maeda et al. 2009). To generate this effect, however, a very high dose of the artificial chromophore was required, and side effects of this treatment could be quite negative (Palczewski 2010). In this case, and in the case of many other drugs which do manage to cross the oBRB, the therapeutic window of these therapeutics, as well as their efficacy, could be greatly improved by their delivery across a modulated oBRB. In the case of 11-cis-retinal replacements such as 9-cis-retinyl acetate, delivery across the oBRB might be preferable due to 11-cis-retinal’s production being in the RPE, therefore delivery across the oBRB would be more akin to its natural delivery pattern, as well as improving the therapeutic window. Delivery across the oBRB as well as the iBRB would maximise this improvement, and this could hypothetically be done by suppressing the tight junction protein occludin, which is expressed in both the iBRB (Leal et al. 2010) and the oBRB (Miura et al. 2010).
One final strategy behind targeting the oBRB rather than the iBRB for delivery of therapeutics is that, by modulating the iBRB by systemic siRNA delivery, the blood–brain barrier is also modulated due to the apparent equivalence in tight junction protein composition (Campbell et al. 2008). For some therapeutics, this would be undesirable if they have psychoactive properties or are neuronally damaging. One way to circumvent this problem would be to subretinally inject an AAV carrying a claudin-5 shRNA under the control of an inducible promoter. Another option, however, would be to target the oBRB and not the iBRB, for instance by suppression of claudin-1 which does not appear to be prominently expressed in the endothelial barrier (Xu et al. 2005).
21.4 Closing Remarks
As has been previously demonstrated (Campbell et al. 2009), modulation of the iBRB by the suppression of tight junction proteins holds much promise for the delivery of therapeutics that could aid in combatting many conditions of the retina. Modulation of the oBRB, in conjunction with or in the absence of iBRB modulation, could aid in the efficient and safe delivery of therapeutics to the retina, as well as potentially opening up new therapeutic opportunities unique to delivery across the oBRB.