17 May 2012. New evidence appearing in today’s Nature implicates the ApoE4 allele, the primary genetic risk factor for late-onset Alzheimer’s disease, as a prime culprit in damaging brain blood vessels. Although scientists suspected that ApoE4 worked some mischief at the blood-brain barrier (BBB), the mechanism was unknown. Now, researchers led by Berislav Zlokovic, previously at the University of Rochester, New York, and now at the University of Southern California, Los Angeles, describe a detailed inflammatory pathway through which human ApoE4 triggers BBB breakdown in transgenic mice. This breakdown causes toxic serum proteins to accumulate in the brain and provokes neuronal degeneration, the authors report. Notably, this occurs in the absence of any Aβ. Zlokovic and colleagues were able to restore the BBB and improve the neuronal health of the mice through genetic and pharmacological manipulations of the pathway, suggesting this mechanism could be a therapeutic target in ApoE4 carriers, who make up the majority of sporadic AD cases. However, it remains to be seen whether the results will translate to humans, and whether the pathway will be amenable to drug development. First author Robert Bell at the University of Rochester previously presented some of this research at the 2010 Society for Neuroscience annual conference in San Diego, California (see ARF related news story).
Other scientists contacted by Alzforum expressed enthusiasm for the findings. “The paper is a tour de force in pinning down a role for ApoE in vascular protection,” said Cheryl Wellington at the University of British Columbia, Vancouver, Canada. “This is an important paper, because it describes a very early event in AD pathogenesis that could precede a lot of other downstream things.”
Researchers have struggled to nail down precisely how ApoE4 inflates AD risk because the protein acts on so many cellular processes (see ARF related news story). In particular, ApoE4 promotes amyloid-β deposition, giving it a direct role in AD pathology (see ARF related news story). How does Aβ, which is present in human AD, relate to ApoE’s actions at the BBB? Zlokovic told Alzforum he favors a two-hit hypothesis. He believes ApoE4 first damages the cerebrovasculature, kicking off a cascade of brain damage, then, as a second hit, amplifies Aβ deposition (see Zlokovic, 2011). Other scientists agreed that the BBB mechanism is probably an important contributor to brain damage, but is unlikely to explain all of the AD risk conferred by ApoE4, and may act in tandem with Aβ-dependent pathways. “Cerebrovascular degeneration in concert with Aβ [deposition] could have a synergistic effect on cognition,” suggested Donna Wilcock at the University of Kentucky, Lexington.
To study the vascular role of ApoE, Bell and colleagues used mice in which the endogenous mouse protein was replaced with human ApoE2, 3, or 4, as well as ApoE knockout animals. In two-week-old mice with ApoE4 or no ApoE, cerebral blood vessels leaked profusely, capillary length declined, and cerebral blood flow dropped. These changes grew worse with age. Intriguingly, levels of the proinflammatory cytokine, cyclophilin A (CypA), which has been shown to damage blood vessels (see Satoh et al., 2009; Jin et al., 2004), jumped fivefold in these animals compared to ApoE2 and ApoE3 mice. The CypA explosion occurred in pericytes, a type of cell that wraps around small blood vessels. When the authors crossed ApoE4 and knockout animals with mice lacking CypA, the BBB remained intact. Feeding the mice cyclosporine A, a CypA inhibitor, also tightened up the BBB.
The authors then dissected how ApoE4 induces CypA. They found that ApoE, which is predominantly made by astrocytes in the brain, binds to low-density lipoprotein receptor-related protein 1 (LRP1) on pericytes. The ApoE4 allele, however, fails to bind the receptor, as shown by proximity ligation assay, a highly sensitive type of immunoassay (see Fredriksson et al., 2002; Söderberg et al., 2006). In ApoE4 or ApoE knockout mice, CypA synthesis goes wild. The cytokine then activates pericyte nuclear factor κB (NF-κB), which translocates to the nucleus and pumps up production of matrix metalloproteinase 9 (MMP9). This proteinase chews up capillary basement membrane and tight junction proteins, effectively punching holes in the blood-brain barrier. Interfering with any step in this pathway, by pharmacological inhibitors, short interfering RNA, or genetic deletion, restored BBB function, the authors report.
Furthermore, the brains of ApoE4 and knockout mice accumulated serum proteins such as fibrin, thrombin, and hemosiderin, which can poison neurons (see Grammas, 2011; Paul et al., 2007). Bell and colleagues showed that, by four months of age, ApoE4 mice had less neuronal activity and were losing neurites and synaptic proteins. Inhibiting the CypA-MMP9 pathway partially reversed this neurodegeneration, improving neuron structure and function. In future work, Zlokovic said he will test behavior and information processing in these mice to see if the neuronal losses correlate with cognitive problems.
One big question is whether these findings relate to humans. Zlokovic plans to examine cerebrospinal fluid from AD patients to see if the main markers of this inflammatory pathway, CypA and MMP9, are elevated in people with the ApoE4 allele. In collaboration with colleagues at the University of Southern California, he is also developing methods using MRI to look at BBB health in AD patients. The task is challenging, because small flaws in capillaries typically do not show up on MRIs, he noted.
Neurodegenerative conditions such as AD frequently go hand-in-hand with BBB disruption (see, e.g., Farrall and Wardlaw, 2009; Dickstein et al., 2010), and vascular flaws appear to be more pronounced in people carrying the ApoE4 allele (see Salloway et al., 2002). People with the ApoE4 allele are known to have higher levels of cerebral amyloid angiopathy (CAA) and be more susceptible to microhemorrhages compared to non-carriers, which has caused problems for this group in immunotherapy trials, Wilcock pointed out (see, e.g., ARF related news story). Wilcock recently showed that giving immunotherapy to mice activates the MMP9 pathway, which may help explain some of the vascular damage seen in trials (see Wilcock et al., 2011). Zlokovic’s study now highlights why ApoE4 carriers may be particularly susceptible to this mechanism. It also holds out hope that “Maybe we now have a targetable pathway in ApoE4s,” Wilcock said. “I think all of this is going to start pointing us toward a more personalized therapeutic approach, as opposed to a one-size-fits-all.”
The ApoE4 allele is also a risk factor for other neurodegenerative conditions, such as Parkinson’s and multiple sclerosis (see ARF related news story). Having an E4 allele worsens a person’s outcome after ischemic or traumatic brain injury, noted Yadong Huang at the Gladstone Institute for Neurological Disease, San Francisco, California (see Mayeux et al., 1995 and ARF related news story). Huang believes the new findings have relevance for these conditions, too. “Whenever you have trauma or brain injury, cerebrovascular integrity is critical,” he said.
While the new data suggest that inhibiting the CypA-MMP9 pathway could reverse vascular problems and perhaps prevent brain damage in ApoE4 carriers, the route from an academic result to a usable drug is a long and arduous one, cautioned Ryan Watts at Genentech, South San Francisco, California. Because most chemical inhibitors are “dirty,” hitting many targets, researchers need to perform careful pharmacodynamic and pharmacokinetic studies over a range of doses to be sure a drug is selectively inhibiting the desired target, he said. Also, molecules such as MMP9 and CypA act on many pathways and have beneficial effects as well, which means that inhibiting them long term could have undesirable side effects. “Although promising, substantially more work will be necessary to determine if these pathways/targets proposed in this manuscript are suitable for the treatment of Alzheimer's and other diseases associated with ApoE4,” Watts wrote (see full comment below).—Madolyn Bowman Rogers.
Bell RD, Winkler EA, Singh I, Sagare AP, Deane R, Wu Z, Holtzman DM, Betsholtz C, Armulik A, Sallstrom J, Berk BC, Zlokovic BV. Apolipoprotein E controls cerebrovascular integrity via cyclophilin A. Nature. 2012 May 16. Abstract