The G-protein coupled receptor Gpr3 helps γ-secretase pump out Aβ, and targeting the receptor may present a therapeutic strategy, according to a study published in Science Translational Medicine on October 14. Reducing or eliminating the expression of Gpr3 dampened Aβ production in four different mouse models of Alzheimer’s disease, including APP knock-in mice. The researchers, led by Amantha Thathiah and Bart De Strooper at KU Leuven in Belgium, found that AD mice lacking the receptor fared better on memory tests. They also reported that in people, elevated levels of the protein correlated with progression of the disease. The researchers plan to test a suite of potential Gpr3 inhibitors in mice as potential therapeutics.
The specific biological roles played by G-protein coupled receptors makes them attractive targets for small-molecule therapeutics. Roughly half of all marketed drugs bind G-protein coupled receptors (GPCRs) (see Ma and Zemmel, 2002). When triggered, these 7-transmembrane receptors touch base with various G proteins in the cytoplasm, which then modulate myriad signaling pathways. Many GPCRs have no known endogenous ligand, yet still have important known functions. Gpr3 is one of these “orphan” receptors. In 2009, researchers in De Strooper’s lab fingered Gpr3 for triggering Aβ production by γ-secretase, and reported that APP/PS1 mice deficient in the receptor had a reduced amyloid burden (see Feb 2009 news). A subsequent study from the group reported revealed a twist—the receptor promoted APP processing through recruitment and interaction with β-arrestin 2, an intracellular scaffolding protein, rather than through coupling with G-proteins (see Dec 2012 news). The researchers found that following its recruitment to the membrane by Gpr3, β-arrestin 2 also interacted directly with the Aph-1α subunit of γ-secretase, and promoted localization of the complex to lipid rafts in the plasma membrane, Thathiah told Alzforum. The researchers speculated that this may somehow ramp up the APP processing activity of the secretase.
First author Yunhong Huang and colleagues wanted to make certain that Gpr3’s effects on Aβ production would hold up in multiple models of AD. The recent advent of APP-KI mice, which express human APP with familial AD mutations at physiological levels, motivated the researchers to conduct such a comparison study, Thathiah said.
The researchers started off by analyzing Aβ production in their tried-and-true model, the APP/PS1 mouse. At 9 to 12 months of age, APP/PS1 mice lacking one or both copies of Gpr3 had less than half as much soluble and insoluble Aβ40 and Aβ42 in the hippocampus and cortex as control animals expressing two copies of the gene. Amyloid plaque burden, as measured by confocal imaging of brain slices, also dropped by half in mice lacking Gpr3.
Roughly the same pattern emerged when the researchers created APPDutch mice lacking Gpr3. This mouse strain overexpresses human APP harboring the E693Q mutation, which causes a form of familial cerebral amyloid angiopathy and does not develop parenchymal amyloid plaques. APPDutch mice lacking Gpr3 had less Aβ40 and Aβ42 in the hippocampus and cortex than those expressing Gpr3. Wild-type mice, which express only endogenous mouse APP, also had a moderate reduction in mouse Aβ40 and Aβ42 when crossed onto the Gpr3 knockout background.
Finally, the researchers turned to two strains of APP-KI mice: the APPNL strain, which expresses physiological levels of APP with the Swedish mutation, and APPNL-F mice, which also harbors the Beyreuther/Iberian APP mutation. The Swedish mutation increases the total amount of Aβ40 and Aβ42, while the Beyreuther/Iberian mutation raises the A42/40 ratio. When crossed to a Gpr3 KO background, both strains of APP-KI mice had reduced levels of Aβ40 and Aβ42. Interestingly, the Aβ42/40 ratio fell in the APPNL-F mice compared to their counterparts expressing Gpr3.
To carefully scrutinize the effects of Gpr3 loss on amyloid plaque deposition in APPNL-F mice, the researchers rendered the animals’ brains transparent using a technique similar to 3DISCO. A flurry of optical clearing methods have come on the scene in recent years (for examples, see Jul 2014 news and Sep 2015 news), and the researchers chose this one, which uses an organic solvent to dissolve lipids. They injected the mice intravenously with fluorescently labeled lectin to label the brain vasculature, then clarified the brains and stained them with Thioflavin S to label amyloid plaques. Fluorescence microscopy of the clear brain allowed the researchers to accurately calculate the total volume of amyloid plaques in the animals, a feat that would not have been possible with two-dimensional imaging. They found that APPNL-F mice lacking Gpr3 had roughly half the number of plaques and only a third the total amyloid volume as those expressing Gpr3.
Could ablating Gpr3 also reduce cognitive deficits? To find out, the researchers looked to APP/PS1 mice, which develop memory problems at a young age. Though not as smart as wild-type mice, APP/PS1 mice more quickly learned the location of a submerged platform in the Morris water maze when they lacked the Gpr3 gene. They were also less anxious and willing to spend more time in an exposed area than APP/PS1 mice. Thathiah said they plan to assess cognition in the knock-in mice as well, but those animals take at least a year to develop cognitive symptoms.
Given the role of Gpr3 in Aβ production and memory in mice, the researchers wondered if it tied into AD pathology in people. They measured Grp3 in brain samples from 131 people of various ages, and found no correlation between protein level and age. However, Gpr3 was elevated in brains from 18 AD patients. In two other sample sets totaling 40 brains, Grp3 levels in the brain correlated with the disease severity measured through Braak staging.
“This study provides an important validation of Gpr3 as a therapeutic target for AD,” commented Michael Wolfe of Brigham and Women’s Hospital in Boston, who called the results impressive. “It will be critical to show that these effects occur upon pharmacological inhibition in adult mice, rather than on genetic deletion from conception onward,” he said.
A handful of commercially available Gpr3 inhibitors exist, Thathiah said, but they have not been tested for their ability to block the recruitment of β-arrestin 2, a requisite for Gpr3’s effect on γ-secretase activity. Thathiah told Alzforum that her lab, which will soon move to the University of Pittsburg in Pennsylvania, plans to screen for new inhibitors and test commercially available ones.
Klaus Heese of Hanyang University in Seoul noted that the relationship between GPCRs and APP processing has a long history, as previous studies pinpointed interactions between other GPCRs and γ-secretase function. “Huang and colleagues add further evidence about the significance of GPCRs and APP processing,” he wrote. Targeting Gpr3 might offer a promising therapeutic approach, Heese added.—Jessica Shugart
- Big Haul? A G Protein-coupled Receptor Regulates Aβ Production
- Could β-Arrestin Provide New Way to Halt Aβ Accumulation?
- Transparent Bodies Allow Neural Networks to ‘Apparate’
- With ScaleS, You Can See Through the Brain to Behold Synapses
Research Models Citations
- Huang Y, Thathiah A. Regulation of neuronal communication by G protein-coupled receptors. FEBS Lett. 2015 Jun 22;589(14):1607-19. Epub 2015 May 14 PubMed.
- Heese K. G Proteins, p60TRP, and Neurodegenerative Diseases. Mol Neurobiol. 2013 Jan 24; PubMed.
- Thathiah A, De Strooper B. The role of G protein-coupled receptors in the pathology of Alzheimer's disease. Nat Rev Neurosci. 2011 Feb;12(2):73-87. PubMed.
- Huang Y, Skwarek-Maruszewska A, Horré K, Vandewyer E, Wolfs L, Snellinx A, Saito T, Radaelli E, Corthout N, Colombelli J, Lo AC, Van Aerschot L, Callaerts-Vegh Z, Trabzuni D, Bossers K, Verhaagen J, Ryten M, Munck S, D'Hooge R, Swaab DF, Hardy J, Saido TC, De Strooper B, Thathiah A. Loss of GPR3 reduces the amyloid plaque burden and improves memory in Alzheimer's disease mouse models. Sci Transl Med. 2015 Oct 14;7(309):309ra164. PubMed.