20 Aug 2014

A mouse strain expressing low levels of human β-secretase 1 (hBACE1) may model the early stages of sporadic AD. Called PLB4, the mouse made its debut on August 6 in The Journal of Neuroscience. There, researchers reported that the animals accumulate an array of amyloid oligomers along with a smattering of plaques in the brain. The animals also perform poorly in cognitive tests. Led by Bettina Platt at the University of Aberdeen in Scotland, the researchers expressed hBACE1 at just above physiological levels using a targeted insertion strategy. They detected amyloid abnormalities without overexpressing human amyloid precursor protein (APP) in the mice, which other Aβ models typically do.

“This represents the next wave of AD mouse models, which are more physiological and less likely to lead us astray,” said Scott Small of Columbia University in New York, who was not involved in the study.

β-secretase performs the rate-limiting step in the generation of Aβ peptides from APP. Researchers created mouse strains expressing human BACE1 years ago (see Bodendorf et al., 2002; Lee et al., 2005; and research models database on hBACE54 and hBACE). However, these mice express high levels of the protein and harbor no amyloid plaques, which at that time were considered the most important marker of AD pathology, Platt said. Researchers therefore crossed hBACE mice to those overexpressing human APP, and observed that BACE1 overexpression boosted plaque production. However, how human BACE1 expression affected the production of other potentially toxic Aβ species—such as oligomers—was never explored when APP was only expressed endogenously, Platt said.

Platt initially planned to cross her hBACE1 mouse directly to APP transgenics, as well. "However, when we started working with this mouse, we found that it had a lot of striking phenotypes,” she said.

First author Kaja Plucinska and colleagues generated the PLB4 mice by targeting the hBACE1 gene to the hypoxanthine phosphoribosyltransferase (HPRT) locus on the X chromosome, a region known as a safe place to insert new genes without causing problems with gene regulation. They placed the gene under the control of the CaMKIIα promoter, which resulted in approximately twofold higher expression in the forebrain than the endogenous BACE1 gene. The mice retain their own BACE genes.  Due to the subtle expression of the human BACE1 gene and lack of APP overexpression in the PLB4 mice, the researchers performed an extensive battery of cognitive and biochemical experiments designed to tease out the slightest of phenotypes.

To look for differences in non-associative learning, the researchers placed the mice in new cages and monitored their activity. PLB4 mice moved around their new environs for longer periods of time than wild-type mice did, indicating that they failed to habituate to their new surroundings—a sign of poor learning. To test the animals’ spatial memory, the researchers subjected them to an open-field water maze test. The PLB4 mice took slightly longer paths than normal mice to reach a hidden platform, even after several days of training. This deficit was only apparent after six months of age. When the researchers tracked the trajectories taken to reach the platform, they found that PLB4 mice were more likely to search by circling the entire perimeter of the pool or scanning randomly, rather than swimming directly toward safety. This strategic difference indicates that the PLB4 mice may have resorted to non-hippocampal approaches to reach their target, Platt said. “The path length differences were not great, but the strategy was,” she said.

PLB4 mice had deficits in working and semantic memory, based on other cognitive tests. They were less anxious and explored open environments more than normal mice. They had no overt motor deficits. In all cases, behavioral phenotypes were most obvious at six months, a factor that Platt attributes to worsening pathology with age. By 12 months, the PLB4 mice still underperformed in cognitive tests, but wild-type mice were also starting to slip at that point so the differences between the strains were less pronounced, Platt said.

The researchers next parsed the biochemical changes that could cause the cognitive deficits. Using an antibody specific for mouse APP, Plucinska identified elevated levels of several APP protein fragments via western blots of whole brain extracts. These included bands corresponding to hexameric Aβ, the reportedly toxic 56kDa, “Aβ*56” oligomer (see Mar 2006 news story), as well as CTFα and CTFβ—intracellular products of α-secretase and BACE1 cleavage, respectively. The researchers also detected putative  oligomers of Aβ when they probed western blots with the 6E10 antibody, which cross-reacts with human and mouse APP. Compared to brain extracts from mice overexpressing human APP and PS1 (see APPSwe/PSEN1(A246E)), which contained only fibrillary and monomeric Aβ, the PLB4 mice expressed several intermediate Aβ species but not high molecular weight species or the monomer. When the researchers treated their extracts with hexafluoroisopropanol, a solvent known to disaggregate Aβ oligomers, they detected monomers and a reduction in putative oligomeric species including Aβ*56, suggesting that the bands represented bona fide Aβ oligomers. An antibody specific for oligomeric Aβ also detected several Aβ species, including those corresponding to Aβ*56.

“We think in our model the aggregation tendency of Aβ is so strong that there is only a small pool of monomer, and it forms oligomeric species very fast,” Platt said. She speculated that mouse Aβ was not prone to form fibrillar amyloid species, and that the oligomeric forms most likely caused the cognitive deficits observed in PLB4 mice.

The researchers stained hippocampal and cortical tissue slices with the 6E10 antibody—which reacts with both Aβ and full-length APP—and detected an accumulation of the protein both in- and outside of cells in PLB4 mice. Plaques were present, but were small and sparsely distributed. Degenerated neurons were common in the PLB4 mice, but rare in the wild-type.

“The PLB4 mouse model is a great basis for further studies,” commented Stefan Lichtenthaler of the German Center for Neurodegenerative Disease in Munich. “Given that BACE1 cleaves many different neuronal substrates, the authors should next cross their mouse with an APP knockout mouse in order to figure out whether the behavioral/memory changes in their mouse are indeed only due to the increased amyloidogenic processing of APP,” he wrote. Other researchers agreed with Lichtenthaler, and further noted that the PLB4 mice still express endogenous BACE1 gene. “It could be possible that these mice produce more Aβ because there is more BACE1 activity, and the results of this study might not be specific to human BACE1,” commented Frank LaFerla of the University of California, Irvine. “A good strategy to resolve this issue is to cross hBACE1 mice with BACE1 knockout mice.”

Platt said her lab is already doing this. She also plans to obtain BACE1 inhibitors from pharmaceutical companies, and measure their effects on the production of various oligomeric Aβ species in the mice. BACE1 inhibitors are being tested in clinical trials, but given the plethora of other BACE1 substrates, research is still needed to confirm their safety (see Dec 2013 conference coverage series). Testing out BACE1 inhibitors in these mice could help researchers understand potential side effects more clearly, Platt suggested. She also hopes the mice will serve as a useful model to measure the effects of lifestyle factors, such as diet and environmental enrichment, on the progression of disease.—Jessica Shugart

Comments

1. • Frank LaFerla
     University of California, Irvine
   • Posted: 21 Aug 2014
   • Paper: Knock-in of human BACE1 cleaves murine APP and reiterates Alzheimer-like phenotypes.

   This study, executed by Dr. Bettina Platt's laboratory, is an important step toward understanding the role of human BACE1 in Alzheimer’s disease. It is an elegant study that provides a better understanding of the physiological role of human BACE1. In fact, our laboratory demonstrated that BACE reduction resulted in a significant decrease of soluble Aβ-oligomers, preventing the development of wild-type human tau pathology and restoring cognition (Chabrier et al., 2012). Therefore, it would be of high interest to consider BACE1 as a potential target for therapeutic intervention.

   However, we need to interpret carefully the results obtained from this study.  PLB4 mice present both endogenous and human BACE1.  Therefore, it is possible that these mice produce more Aβ because they have more BACE1 activity; then the results of this study would not be specific to human BACE1.  A good strategy to resolve this issue is to cross hBACE1 mice with BACE knockout mice.

   Furthermore, BACE1 has many putative functions and substrates involved in cognitive function. Therefore, this model would be useful to determine the role of BACE1 in cognitive impairments.  Obviously, Aβ is a key molecular factor that might mediate BACE1 cognitive deficits, however, BACE1 also regulates cAMP/PKA signaling and controls voltage-gated sodium channels and serotonergic transmission.

   New findings reveal that BACE1 activity increases early in AD pathogenesis and it is elevated in patients with mild cognitive impairment. BACE1 appears to be an important factor for MCI patients to convert to AD (Cheng et al., 2014; Jiang et al., 2011; Zetterberg et al., 2008).  Therefore, this novel model will provide a great opportunity to better understand this phenomenon.

   References:

   Chabrier MA, Blurton-Jones M, Agazaryan AA, Nerhus JL, Martinez-Coria H, Laferla FM. Soluble aβ promotes wild-type tau pathology in vivo. J Neurosci. 2012 Nov 28;32(48):17345-50. PubMed.

   Cheng X, He P, Lee T, Yao H, Li R, Shen Y. High activities of BACE1 in brains with mild cognitive impairment. Am J Pathol. 2014 Jan;184(1):141-7. PubMed.

   Jiang H, Hampel H, Prvulovic D, Wallin A, Blennow K, Li R, Shen Y. Elevated CSF levels of TACE activity and soluble TNF receptors in subjects with mild cognitive impairment and patients with Alzheimer's disease. Mol Neurodegener. 2011;6:69. PubMed.

   Zetterberg H, Andreasson U, Hansson O, Wu G, Sankaranarayanan S, Andersson ME, Buchhave P, Londos E, Umek RM, Minthon L, Simon AJ, Blennow K. Elevated cerebrospinal fluid BACE1 activity in incipient Alzheimer disease. Arch Neurol. 2008 Aug;65(8):1102-7. PubMed.

   View all comments by Frank LaFerla
2. • Stefan Lichtenthaler
     German Center for Neurodegenerative Diseases (DZNE)
   • Posted: 21 Aug 2014
   • Paper: Knock-in of human BACE1 cleaves murine APP and reiterates Alzheimer-like phenotypes.

   I really like this paper. It shows that a mild overexpression of human BACE1 in mice can induce pathological changes similar to those seen in a human AD brain. It is amazing that the changes happen so early (behavioral and memory changes after a few months and pathology after about one year), although only endogenous APP is used. The paper is a great step toward the use of AD mouse models not overexpressing APP, and complements the recent Nat Neuroscience paper from Takaomi Saido’s lab (Saito et al., 2014). 

   The mouse model is a great basis for further studies. Given that BACE1 cleaves many different neuronal substrates, the authors should next cross their mouse with an APP knock-out mouse in order to figure out whether the behavioral/memory changes in their mouse are indeed only due to the increased amyloidogenic processing of APP or also due to increased cleavage of additional BACE1 substrates with brain functions. Additionally, it would be helpful if the mouse could be crossed with BACE1-deficient mice, such that only the overexpressed human BACE1 remains. Another point to study is the unexpected increase of C83. Is there really an increase in this a-secretase fragment or is this rather C89, i.e. the β-prime cleavage fragment, which results from the BACE1 cleavage site?

   As an additional point: the title "Knock-in of human BACE1…." is a bit oversimplified because it suggests that human BACE1 is knocked into the mouse BACE1 locus. In fact, their mouse mildlyover expresses human BACE1 that was knocked into a specific genetic locus, but not that of murine BACE1. Thus, the mouse has both the endogenous murine BACE1 plus the transgenically expressed human BACE1.  

   References:

   Saito T, Matsuba Y, Mihira N, Takano J, Nilsson P, Itohara S, Iwata N, Saido TC. Single App knock-in mouse models of Alzheimer's disease. Nat Neurosci. 2014 May;17(5):661-3. Epub 2014 Apr 13 PubMed.

   View all comments by Stefan Lichtenthaler

Make a Comment

To make a comment you must login or register.

References

Research Models Citations

1. PLB4 (hBACE1)
2. hBACE54
3. hBACE
4. APPSwe/PSEN1(A246E)

Alzpedia Citations

1. BACE1

News Citations

1. Aβ Star is Born? Memory Loss in APP Mice Blamed on Oligomer 17 Mar 2006
2. Cloistered Retreat Takes the Pulse of BACE Research 7 Dec 2013

Antibody Citations

1. APP or β-Amyloid (6E10)

Paper Citations

1. Bodendorf U, Danner S, Fischer F, Stefani M, Sturchler-Pierrat C, Wiederhold KH, Staufenbiel M, Paganetti P. Expression of human beta-secretase in the mouse brain increases the steady-state level of beta-amyloid. J Neurochem. 2002 Mar;80(5):799-806. PubMed.
2. Lee EB, Zhang B, Liu K, Greenbaum EA, Doms RW, Trojanowski JQ, Lee VM. BACE overexpression alters the subcellular processing of APP and inhibits Abeta deposition in vivo. J Cell Biol. 2005 Jan 17;168(2):291-302. PubMed.

Further Reading

Papers

1. Lesné SE, Sherman MA, Grant M, Kuskowski M, Schneider JA, Bennett DA, Ashe KH. Brain amyloid-β oligomers in ageing and Alzheimer's disease. Brain. 2013 May;136(Pt 5):1383-98. PubMed. Correction.