ApoE Isoform Expression as a Risk Factor for Alzheimer's Disease
Alzheimer's disease (AD) is the most prevalent neurodegenerative disorder, affecting approximately 10% of the population over the age of sixty-five.1 The brains of AD patients are biochemically characterized by the accumulation of beta-amyloid peptide (A-beta), the proposed causative agent of AD-related neuronal dysfunction and neurodegeneration. However, AD patients also experience neurovascular defects including cerebral amyloid angiopathy, the deposition of A-beta in blood vessels within and surrounding the brain. Currently, it is unknown if the vascular deposition of A-beta is etiologically relevant to the development of AD or is merely a consequence of the disease.
The risk of an individual developing AD is strongly determined by the Apolipoprotein E (ApoE) allele that they possess (APOE2, 3, or 4).2,3 Inheritance of APOE3 lowers the genetic risk of developing this devastating neurological disorder, in a dose-dependent manner. In contrast, individuals that possess one or two APOE4 alleles have a significantly greater probability of developing AD. In addition, the expression of the APOE4 isoform has been linked to neurovasculature dysfunction, exacerbated Down’s syndrome-associated dementia, and inferior neurological recovery following traumatic brain injury or hemorrhage.4 Although ApoE is known to regulate microvasculature permeability and promote A-beta clearance, the biological mechanisms underlying the relationship between specific ApoE isoforms and the development of AD remain to be clearly demonstrated.
ApoE Maintains Blood-brain Barrier Function and Cerebrovascular Integrity
To test the hypothesis that the ApoE genotype influences the probability of developing AD by affecting microcirculation in the brain, Bell et al. studied blood-brain barrier (BBB) function in mice that lacked murine ApoE (Apoe–/–) and mice that expressed only one human ApoE isoform (TR-APOE) as a result of targeted replacement.5 Multiphoton microscopy experiments following tetramethylrhodamine-conjugated dextran administration revealed an intact BBB in TR-APOE2 and TR-APOE3 mice, but a leaky BBB in Apoe–/– and TR-APOE4 models. These data support a functional role for APOE2 and APOE3 but not APOE4 in maintaining BBB integrity.
ApoE4-induced Basement Membrane Degradation & Blood-brain Barrier Dysfunction. Under normal healthy conditions, astrocyte-derived ApoE2 and ApoE3 bind LRP-1 to suppress pericyte CypA concentrations and maintain blood-brain barrier (BBB) function. However, the ApoE4 isoform does not bind LRP-1, leading to elevated intracellular CypA concentrations, nuclear translocation of NFkB, and increased expression of MMP-9.5 MMP-9 degrades endothelial cell tight junctions and the basement membrane, resulting in BBB dysfunction and the extravascular accumulation of neurotoxic proteins.
To determine the signaling cascade responsible for ApoE4-induced BBB impairment, the authors investigated the involvement of Cyclophilin A (CypA), a proinflammatory cytokine that is known to induce vascular dysfunction.6 Immunohistochemical studies showed that CypA expression was specifically increased in pericytes, cells previously shown to modulate microcirculation in response to astrocyte-secreted ApoE.7 This effect was only observed in Apoe–/– and TR-APOE4 mice, suggesting that APOE2 and APOE3 are required to maintain a physiologically low concentration of CypA. In the Apoe–/– and TR-APOE4 groups, pharmacological inhibition of CypA using Cyclosporine A reversed the leaky BBB phenotype, supporting a causal role for elevated CypA in the disruption of BBB function.
To further elucidate the underlying molecular mechanism, Bell et al. investigated the importance of Matrix Metalloproteinase 9 (MMP-9). MMP-9 is a gelatinase that is known to be activated by CypA and capable of degrading proteins that are essential to the composition of capillary basement membranes (Collagen IV) and tight junctions (ZO-1, Occludin, and Claudin-5).8 Gelatin zymography of Apoe–/– and TR-APOE4 brain tissues indicated a significant increase in pro-MMP-9 and activated MMP-9 that correlated with the immunohistochemical detection of MMP-9+ pericytes. In the hippocampus, inhibition of MMP-9 activity via SB-3CT pharmacological blockade or short interfering (si) RNA, eliminated BBB dysfunction in Apoe–/– and TR-APOE4 mice. Further pharmacological experiments, using pyrrolidine dithiocarbonate (PDTC), showed that disruption of Nuclear Factor kappa B (NFkB) nuclear translocation prevented MMP-9 activation and corrected BBB leakage in these mice.
These data support the hypothesis that APOE4 induces cerebrovascular dysfunction in a CypA-NFkB-MMP-9-dependent manner but do not address the mechanism by which astrocyte-secreted APOE4 affects pericytes. Since ApoE is a known ligand for the ApoE receptor and Low Density Lipoprotein receptors (LDL R), the authors designed experiments to determine if these receptors were necessary for ApoE activity. Specific receptor blockade via antibody neutralization or siRNA revealed that LDL R-related protein 1 (LRP-1) was required by murine ApoE and human APOE3 to maintain intracellular pericyte CypA concentrations at a physiological range. Proximity ligation imaging experiments showed that although murine ApoE and human APOE3 bind to LRP-1 with high affinity, an interaction between human APOE4 and LRP-1 was almost undetectable. This finding further supports the conclusion that loss of ApoE:LRP-1 binding underlies the increase in pericyte CypA expression observed following deletion of ApoE or expression of ApoE4.
Recovery of Blood-brain Barrier Function Reverses ApoE4-induced Neuronal and Synaptic Dysfunction
To determine how changes in BBB permeability induced by deletion of ApoE might affect neuronal function, Bell et al. examined levels of endogenous blood-derived proteins in the brain. Consistent with impaired BBB integrity, extravascular accumulation of serum IgG was detected only in the hippocampi of mice lacking murine ApoE or expressing human ApoE4. These mice also displayed neuronal accumulation of blood-derived Fibrin and Thrombin, and perivascular deposition of Hemosiderin. Because Fibrin is known to exacerbate neurovascular dysfunction, Thrombin can be neurotoxic, and Hemosiderin induces the generation of reactive oxygen species, these data present multiple mechanisms by which BBB dysfunction could result in neuronal injury and increased vulnerability to neurodegenerative disease.9,10,11 Importantly, inhibition of the CypA-NFkappaB-MMP-9 pathway by Cyclosporine A restored BBB function as measured by the normalization of the extravascular levels of these markers in the hippocampus.
Vascular defects were detectable at 2 weeks of age in Apoe–/– and TR-APOE4 mice. To test the hypothesis that disruption of BBB integrity and extravascular accumulation of cytotoxic proteins have deleterious functional consequences in the brain, Bell and colleagues conducted a series of biochemical and electrophysiological analyses. Although no changes were detected at 2 weeks of age, synaptic protein expression, neurite density, and cortical activity were reduced in Apoe–/– and TR-APOE4 mice at 4 months of age, suggesting that the observed vascular defects preceded neuronal and synaptic dysfunction. Furthermore, blocking the CypA-NFkB-MMP-9 pathway with Cyclosporine A, PDTC, or SB-3CT reversed the pathological neuronal and synaptic changes that occurred following BBB impairment. Collectively, these data support the pursuit of CypA as a pharmacological target for the prevention and treatment of ApoE4-induced neurovascular dysfunction and the associated increased risk of neurodegenerative disease.
The Neurovascular Unit. Astrocytes, pericytes, neurons, endothelial cells, and the basement membrane function together to maintain the blood-brain barrier. In the first study described here, Bell et al. investigated the hypothesis that changes incerebrovascular integrity underlie the relationship between different Apolipoprotein E isoforms and the risk of developing Alzheimer’s disease.5
Inducing the Beneficial Actions of ApoE: Clearance of A-beta
In addition to blocking the detrimental effects of ApoE4, recent studies suggest that exogenously enhancing other beneficial neuronal actions of APOE2/APOE3 signaling may represent a complementary therapeutic approach. For example, Cramer et al. recently examined the therapeutic potential of pharmacologically augmenting ApoE-dependent A-beta degradation.12 ApoE expression is transcriptionally regulated by obligate heterodimers composed of a retinoid X receptor (RXR) and either a peroxisome proliferator-activated receptor gamma (PPAR gamma) or a liver X receptor (LXR). Preliminary experiments using Bexarotene, a highly selective RXR agonist, increased ApoE expression and A-beta phagocytosis in primary cultures of mouse astrocytes and microglia. Consistent with the established feed-forward mechanism of activation, Bexarotene also induced expression of the ApoE lipid transporters ABCA1 and ABCG1, and the three nuclear receptors (RXR, PPAR gamma, and LXR).
To more directly test the relevance of Bexarotene-induced A-beta clearance, the authors extended their studies to the APPswe/PS1 delta e9 transgenic mouse model of amyloidosis (APP/PS1). In the brains of AD patients, neurotoxic, insoluble A-beta is deposited as extracellular lesions called senile plaques. When Bexarotene was administered to APPswe/PS1 delta e9 mice at an age that precedes the appearance of senile plaques (2 months), in vivo microdialysis indicated a rapid reduction in the levels of A-beta in the hippocampal interstitial fluid (ISF). Bexarotene also lowered ISF A-beta concentrations in wild-type C57BL/6 controls but had no effect in ApoE-null mice, confirming the dependence of Bexarotene-induced A-beta clearance on ApoE expression. In contrast to the study by Bell et al., the authors did not investigate the isoform-specificity of this ApoE-dependent effect.
Bexarotene treatment also decreased A-beta ISF levels in APP/PS1 mice in the presence of senile plaques. Acute Bexarotene administration reduced ISF A-beta concentrations in 6 month old APP/PS1 mice, and significantly reduced the number of fibrillar, Thioflavin-S+ plaques in the hippocampus and cortex. These results were supported by similar findings in 11 month old APP/PS1 mice and in a more aggressive model of amyloidosis (APPPS1-21 mice). To determine if ApoE-dependent Bexarotene-induced A-beta clearance was associated with improved cognitive function, Cramer and colleagues conducted a battery of behavioral and electrophysiological tests in these transgenic mouse models of amyloidosis. Bexarotene-induced improvements in contextual fear conditioning, spatial reference memory, nest construction behavior, and odor habituation further supported the theory that increased A-beta levels lead to pathological changes in memory and cognition.
Manipulation of ApoE Signaling: A Pharmacological Target for Alzheimer’s Disease
The two recent studies described here support the pursuit of ApoE as a pharmacological target for the prevention and treatment of neurodegenerative disease. Bell et al. demonstrated that the physiological functions of APOE2 and APOE3 include pericyte-dependent control of microvessel integrity and BBB permeability. Although this action was lost following murine ApoE knockout or targeted human APOE4 expression, the deleterious consequences could be antagonized by the inhibition of proinflammatory cytokine CypA. Interestingly, BBB dysfunction was shown to precede pathological changes in neuronal and synaptic structure, suggesting the isoform-specific cerebrovascular actions of APOE4 may contribute to its association with increased risk for developing AD.
In contrast, data from Cramer et al. showed that stimulation of ApoE expression using the RXR agonist Bexarotene enhanced the clearance of A-beta from the brain. Importantly, it was demonstrated that the neuronal and behavioral impairments associated with BBB dysfunction and A-beta deposition could be reversed following the administration of Cyclosporine A and Bexarotene, respectively. These data present ApoE signaling as a central pharmacological target to combat the cerebrovascular and neurodegenerative changes that characterize AD pathology.
- Zhang, Y. et al. (2011) Mol. Brain 4:3.
- Deane, R. et al. (2008) J. Clin. Invest. 118:4002.
- Holtzman, D.M. et al. (2012) Cold Spring Harb. Perspect. Med. 2:a006312.
- Verghese, P.B. et al. (2011) Lancet Neurol. 10:241.
- Bell, R.D. et al. (2012) Nature 484:512.
- Bahmed, K. et al. (2012) Cancer Cell Int. 12:19.
- Bell, R.D. et al. (2010) Neuron 68:407.
- Zhang, M. et al. (2011) Mol. Cell. Biochem. 357:387.
- Yue, M. et al. (2008) Brain 132:26.
- Paul, J. et al. (2007) J. Exp. Med. 204:1999.
- Fisher, M. et al. (2010) Stroke 41:278.
- Cramer, P.E. et al. (2012) Science 335:1503.
This symbol denotes references that cite the use of R&D Systems products.