This is Part 1 of a three-part series. See also Part 2 and Part 3.
29 June 2010. The apolipoprotein E gene is the strongest genetic risk factor for late-onset Alzheimer disease. One ApoE4 allele almost quadruples a person’s chances of getting the disease, while having two drives up the risk by 12- to 15-fold. Not surprisingly, some researchers believe that understanding ApoE could be the key to finding much-needed treatments and preventions for AD. But does ApoE get the attention it deserves? The apolipoprotein has been—and unfortunately still is—a Cinderella compared to amyloid-β precursor protein (APP) and the presenilins, which attract the lion’s share of the research dollars and most of the attention at scientific meetings. “ApoE, ApoE Receptors and Neurodegeneration,” a one-day symposium held 7 June 2010 at Washington University, St. Louis, Missouri, bucked the trend by putting ApoE center stage. Though the meeting addressed the relationship between the apolipoprotein and the amyloid-β cascade, it also stressed ApoE’s own intricate biology. The take-home message from the meeting, which was organized by members of the NIA/NIH program project grant (PPG) on ApoE Receptor Biology and Neurodegeneration, was that there may be a lot more to ApoE’s involvement in AD than its dalliances with Aβ.
A case in point is signaling through the ApoE receptor 2 (ApoER2), one of several members of the low-density lipoprotein (LDL) receptor family. Researchers led by Joachim Herz, University of Texas Southwestern Medical Center, Dallas, had previously reported that this receptor plays a key role in neuronal survival, and at the conference, Herz presented the latest twist implicating ApoE in the plot, as well. ApoER2 knockout mice lose almost half their corticospinal neurons by four months of age (see ARF related news story on Beffert et al., 2005). The reasons for this are not entirely clear, but upon binding to its normal ligand reelin, ApoER2 activates downstream cascades such as those triggered by Jun terminal kinases. Reelin also enhances long-term potentiation (LTP) and reverses Aβ-induced reduction of LTP in the hippocampus (see ARF related news story and ARF news story). The ApoER2 ligand even neutralizes LTP suppression by toxic Aβ dimers isolated from human AD patients (see ARF related news story). All told, the data suggest a complex interplay among Aβ, ApoE, and ApoE receptors, Herz said in his keynote talk. One of the unanswered questions—one that is particularly relevant to AD risk—is whether ApoE genotype influences this interplay. In St. Louis, Herz presented newly published data to suggest that it does.
Herz showed that, in keeping with its role in beefing up LTP, reelin enhances NMDA receptor-driven calcium spikes when added to neurons in culture. The spikes still occur if ApoE2 or ApoE3 are added to cells prior to reelin (though in the presence of ApoE3 the spikes do not last long); however, pre-incubating the neurons with ApoE4 completely suppresses the calcium influx. ApoE4 also suppressed LTP in a more physiological setting. For this, Herz and colleagues turned to ApoE targeted replacement (TR) mice developed by Patrick Sullivan at Duke University. These animals have their mouse ApoE gene replaced with one of the three human ApoE isoforms and have become a model of choice in the field. Reelin enhanced LTP in hippocampal tissue slices from ApoE2 or ApoE3 TR mice, just like it does in wild-type tissue, but failed to do so in ApoE4 slices. How might this relate to Aβ, which puts a damper on LTP? Herz said that Aβ-containing extracts from Alzheimer’s Disease Research Center brain tissue samples had little effect on LTP in ApoE2 slices and no effect on reelin-driven LTP. In ApoE3 slices, AD extracts suppressed LTP slightly, and this was overcome with reelin. But in ApoE4 slices, the Aβ preparation suppressed LTP, and reelin was unable to restore it. The experiments, by Herz lab members Ying Chen, Murat Durakoglugli, and Xunde Xian, were described in the June 14 PNAS online, and the paper is freely available for download.
The work suggests that the ApoE4 isoform might increase susceptibility to AD by undermining the normal processes of LTP and by exacerbating Aβ’s suppression of it. As for a mechanism, Herz believes that ApoE4 blocks beneficial reelin signaling because it sequesters ApoER2 receptors inside cells. When Chen and colleagues looked at steady-state levels of ApoER2 receptors in primary cortical neurons, they found that the receptors were internalized and became trapped in cells treated with ApoE4. However, receptors traffic normally and return to the cell surface in the presence of ApoE2/3. The upshot is that ApoE4 dampens ApoER2 signaling, prevents NMDA receptor activation, and blocks LTP. But some big questions remain, said Herz. These include how ApoE4 alters ApoER2 trafficking and whether the mechanism is amenable to drug treatment. He also thought it will be important to find out how Aβ and ApoE dovetail with respect to synaptic regulation.
Given these findings, one potential therapy for AD might be to activate ApoER2 receptors. Edwin Weeber, University of South Florida, Tampa, has examined this idea using reelin heterozygote mice, whose reelin level is reduced by half. These mice show defects in LTP and perform poorly in a contextual fear learning and memory paradigm. Injecting reelin into their brain ventricles restores reelin and rescues learning and memory.
But can reelin have any effect in wild-type mice, and how might it work? To test this idea, Weeber turned to an in-vivo model of receptor blockade. Injecting into the brain of mice a daily shot of RAP, or receptor associated protein—which binds to all the apolipoprotein receptors—cripples the animal’s performance in water maze tests of spatial memory. But a single shot of reelin into the brain ventricles counteracted RAP; it enhanced associative learning and improved spatial learning performance for days, significantly reducing escape latency in the water maze, Weeber said.
Weeber also reported improved LTP when adding reelin to hippocampal slices from Tg2576 AD mice. He has not tried to give reelin to AD mice in vivo yet. Nonetheless, Weeber thinks that the ApoER2 receptor system could be of interest as a therapeutic target for AD and other neurodegenerative diseases.
Hyang-Sook Hoe, Georgetown University, Washington, DC, also studies ApoER2 receptors—in synapse and spine formation. Using a cell co-culture system, she found increased clustering between COS7 cells and primary hippocampal neurons—as the two form synapses with each other—if the former overexpress ApoER2. The neurons also expressed a third more synaptophysin, a marker of synaptic vesicles, suggesting that ApoER2 stimulates synaptic differentiation. In addition, she tested if the apolipoprotein receptor has post-synaptic effects by looking at dendritic spine formation. Overexpressing ApoER2 in the COS7 cells boosted neuronal spine density in the primary neurons, whereas knocking down the receptor had the opposite effect. These results suggest that ApoER2 can profoundly affect synaptic development. Fitting this idea, Hoe found that ApoER2 knockout mice have about a third fewer dendrites than wild-type animals.
Delving into the mechanism, Hoe focused on PSD-95 and X-11, post-synaptic proteins that both bind to ApoER2. In St. Louis, she reported that in the same heterologous co-culture system, PSD95 and ApoER2 synergistically enhance presynaptic differentiation and also have an additive effect on dendritic spine formation in hippocampal neurons. X11 had the opposite effect. The results point to a model where PSD-95 would bind to ApoER2 to enhance spine formation and X11 binds to ApoER2 to suppress it.
What do these findings have to do with ApoE—and perhaps AD? For this question, Hoe also availed herself of ApoE TR mice, counting the spine density in cortical layers II/III of their brains. She found a dearth of dendritic spines in ApoE4 TR mice compared to wild-type. ApoE4 TR mice as young as four weeks old lose 25 percent of their shaft spines; by one year of age the loss amounts to 55 percent. Hippocampal spines appeared normal, however. It is not clear how this loss relates to the ApoE4-dependent LTP suppression and aberrant ApoER2 trafficking described by Herz, or to Aβ in any way. Going forward, Hoe said she plans to determine whether ApoE isoforms differentially influence the interaction between ApoER2 and PSD-95 or X-11.
In summary, several speakers showed that ApoER2 can profoundly affect synapse formation and function, suggesting one rationale for how different ApoE isoforms might render a person susceptible to cognitive decline.—Tom Fagan.
This is Part 1 of a three-part series. See also Part 2 and Part 3.
Chen Y, Durakoglugil MS, Xian X, Herz J. ApoE4 reduces glutamate receptor function and synaptic plasticity by selectively impairing ApoE receptor cycling. PNAS 2010 June 14 online. Abstract