Years of research have fingered presenilin as the γ-secretase component that slices amyloid precursor protein (APP) to release amyloid-β. But the evidence left nagging doubts as to whether presenilin is truly the perpetrator. Now, scientists have reconstituted γ-secretase activity in a test tube using only presenilin made in bacteria and liposomes to mimic the cell membrane. The secretase works independently of the other γ-secretase complex constituents. “I think we gave final proof that PS1 is self-sufficient for catalysis,” said lead investigator Yueming Li, Memorial Sloan-Kettering Cancer Center, New York, in an ARF interview. The findings were published November 29 in the Proceedings of the National Academy of Sciences online. Beyond settling a number of pressing questions about γ-secretase activity, the reconstitution system offers an unprecedented opportunity for future research on this baffling membrane complex.

Yeast reconstitution assays showed that γ-secretase activity can be recapitulated by four proteins—presenilin (PS), nicastrin, Aph-1, and PEN-2 (Edbauer et al., 2003), and work by Li and others pointed heavily to presenilin as the business end of the APP-cutting complex (see De Strooper et al., 1998; Wolfe et al., 1999; Li et al., 2000). To be fully convinced, though, scientists needed to show that presenilin could indeed cut APP all by itself, without help from its associated cofactors. In the present study, first author Kwangwook Ahn and colleagues accomplish this longstanding goal using a bacterial overexpression approach.

Doing so required ingenuity and a bit of luck, because presenilin springs into action only once cleaved internally at a specific residue. To deal with this quirk, Li’s team chose to overexpress a familial AD-linked PS1 mutant (PS1ΔE9) that is constitutively active because of an exon 9 deletion spanning the region that codes for the endoproteolytic site. The deletion also results in a serine-to-cysteine substitution (S290C) that drives up production of Aβ42, relative to its less pathogenic cousin Aβ40. Reversing this mutation restores wild-type γ-secretase activity to PS1Δ (Steiner et al., 1999), creating, in essence, a constitutively active PS1 with wild-type APP processing. Ahn and colleagues expressed this PS1 variant (PS1ΔE9-C290S) as a fusion with maltose-binding protein (MBP) to allow purification on amylose affinity columns. The purified proteins, sans MBP tag, were introduced into liposomes and tested for γ-secretase activity by measuring the amount of Aβ produced when PS1-liposomes were given biotinylated APP. As a negative control, Ahn and colleagues used a PS1-ΔE9 with alanine displacing aspartate 385, which is critical for γ-secretase activity (i.e., PS1ΔE9-D385A).

Other labs have expressed presenilin in bacteria but, until now, failed to detect γ-secretase activity in the purified material. The present study embedded the recombinant presenilins into lipid vesicles, simulating more closely the protein’s natural environment in cells. “I think that was the trick,” said Michael Wolfe of Brigham and Women’s Hospital in Boston.

As judged by Aβ40 and Aβ42 production, PS1ΔE9-C290S liposomes were only about 11 percent as active as the entire tetrameric complex purified from HEK293 cells. “On the one hand, you can say ‘only,’ but I’m impressed it's 11 percent,” Wolfe said. “Nobody has been able to show activity with bacterially expressed PS.”

Furthermore, Li and colleagues “bent over backwards” to show that the reconstituted PS1 liposomes don’t have “some funky activity but really behave like PS1 when it’s in the entire γ-secretase complex,” Wolfe said. For example, the researchers generated PS1ΔE9 mutants with either or both of the FAD mutations L166P and G384A, put the proteins into liposomes, and checked them for γ-secretase activity. In these experiments, the double mutant boosted Aβ42:Aβ40 ratios to a greater extent than did either of the single mutants, consistent with studies done in less artificial models (Citron et al., 1998; Levitan et al., 2001). In addition, the researchers tested several FAD-associated APP mutants (I45F or V46F) in their in-vitro system and saw elevated Aβ42:Aβ40 ratios relative to PS1ΔE9-C290S, in line with studies of these APP substrates in transfected cells (Lichtenthaler et al., 1999; Steiner et al., 1999).

Li’s study also demonstrated formally that activation of wild-type PS1 requires PEN-2. In the in-vitro assays, wild-type PS1 alone showed no intrinsic γ-secretase activity, but adding recombinant PEN-2 led to proteolytic cleavage and activation of PS1. All told, the system “faithfully recapitulates the essential biochemical and molecular features of γ-secretase activation that has [sic] heretofore only been observed in living cells,” the authors write. “Moreover, our studies reveal that PEN-2 plays an obligatory role in the activation of PS, and does so in the absence of other subunits of the complex.”

Bart De Strooper of KU Leuven, Belgium, called the study “a breakthrough in γ-secretase research” (see full comment below). In his view, the in-vitro reconstitution assay “allows us to investigate the function of individual subunits to an extent that is not possible in any other type of assay.”

Moreover, the recent findings suggest it is feasible to make boatloads of biologically relevant material for future studies. “It's nice that you can show the proteins you're expressing in bacteria are proteolytically active,” Wolfe said. “You can now go ahead and do structural and functional studies feeling comfortable that the protein is behaving as it does when PS is in the whole [γ-secretase] complex.”

There is an important caveat, though. While the present study demonstrates that presenilin has intrinsic γ-secretase activity in a cell-free system, “overexpression of PS alone, wild-type or mutant, does not lead to a rise in γ-secretase activity in any eukaryotic cell culture system or animal model investigated,” noted Harald Steiner of Ludwig-Maximilians-Universitat, Munich, Germany, in an e-mail to ARF. “Thus, in cells, PS is not active unless its complex partners are coexpressed.”

Wolfe thinks the other cofactors may be required to put presenilin into a proteolytically active conformation. “But if you put presenilin into lipid vesicles and control their environment, you may be able to induce a conformation that’s catalytically active without the other γ-secretase components,” he said.

Whether each of the three remaining components is absolutely necessary remains in question, as scientists recently reported that a nicastrin-deficient γ-secretase complex was able to cleave APP and other substrates (Zhao et al., 2010 and ARF related news story). Other studies reveal further nuances of γ-secretase function and modulation. De Strooper and colleagues provided evidence suggesting that Aph-1B-containing γ-secretase contributes highly to brain amyloid burden, whereas complexes containing the Aph-1A isoform seem more geared toward Notch processing (Serneels et al., 2009 and ARF related news story). Furthermore, a recent paper describes the discovery of a γ-secretase activating protein (GSAP) that specifically promotes the interaction of γ-secretase with APP, but not with Notch (He et al., 2010 and ARF related news story).

Ultimately, the present study “is not the end, but rather the end of the beginning, in the long journey to a comprehensive understanding of γ-secretase activity and the spectrum of physiological and pathophysiological roles it serves,” wrote Edward Koo of the University of California, San Diego, in a separate PNAS commentary on Li’s paper.—Esther Landhuis


  1. This is a breakthrough in γ-secretase research. The in vitro reconstitution assay allows us to investigate the function of individual subunits to an extent that is not possible in any other type of assay. The authors show elegantly the proteolytic activity of single presenilin and the activation of presenilin proteolytic activity by PEN-2. The purification method for presenilin is also impressive: It suggests that presenilin can be generated in a preparative scale. Finally, the data suggest that presenilin on its own can be expressed in an active form. It will be much easier to crystallize presenilin alone than in the tetrameric complex.

    View all comments by Bart De Strooper

Make a Comment

To make a comment you must login or register.


News Citations

  1. Think You Know γ-secretase and p25? Think Again
  2. Double Paper Alert—Keystone Presentations Now in Press
  3. There’s a GSAP for That: Novel APP Partner a New Therapeutic Target?

Paper Citations

  1. . Reconstitution of gamma-secretase activity. Nat Cell Biol. 2003 May;5(5):486-8. PubMed.
  2. . Deficiency of presenilin-1 inhibits the normal cleavage of amyloid precursor protein. Nature. 1998 Jan 22;391(6665):387-90. PubMed.
  3. . Two transmembrane aspartates in presenilin-1 required for presenilin endoproteolysis and gamma-secretase activity. Nature. 1999 Apr 8;398(6727):513-7. PubMed.
  4. . Photoactivated gamma-secretase inhibitors directed to the active site covalently label presenilin 1. Nature. 2000 Jun 8;405(6787):689-94. PubMed.
  5. . The biological and pathological function of the presenilin-1 Deltaexon 9 mutation is independent of its defect to undergo proteolytic processing. J Biol Chem. 1999 Mar 19;274(12):7615-8. PubMed.
  6. . Additive effects of PS1 and APP mutations on secretion of the 42-residue amyloid beta-protein. Neurobiol Dis. 1998 Aug;5(2):107-16. PubMed.
  7. . PS1 N- and C-terminal fragments form a complex that functions in APP processing and Notch signaling. Proc Natl Acad Sci U S A. 2001 Oct 9;98(21):12186-90. PubMed.
  8. . Mechanism of the cleavage specificity of Alzheimer's disease gamma-secretase identified by phenylalanine-scanning mutagenesis of the transmembrane domain of the amyloid precursor protein. Proc Natl Acad Sci U S A. 1999 Mar 16;96(6):3053-8. PubMed.
  9. . Amyloidogenic function of the Alzheimer's disease-associated presenilin 1 in the absence of endoproteolysis. Biochemistry. 1999 Nov 2;38(44):14600-5. PubMed.
  10. . Gamma-secretase composed of PS1/Pen2/Aph1a can cleave notch and amyloid precursor protein in the absence of nicastrin. J Neurosci. 2010 Feb 3;30(5):1648-56. PubMed.
  11. . gamma-Secretase heterogeneity in the Aph1 subunit: relevance for Alzheimer's disease. Science. 2009 May 1;324(5927):639-42. Epub 2009 Mar 19 PubMed.
  12. . Gamma-secretase activating protein is a therapeutic target for Alzheimer's disease. Nature. 2010 Sep 2;467(7311):95-8. PubMed.

Other Citations

  1. PS1ΔE9

Further Reading

Primary Papers

  1. . And four equals one: presenilin takes the gamma-secretase role by itself. Proc Natl Acad Sci U S A. 2010 Dec 14;107(50):21236-7. PubMed.
  2. . Activation and intrinsic gamma-secretase activity of presenilin 1. Proc Natl Acad Sci U S A. 2010 Dec 14;107(50):21435-40. PubMed.