γ-secretase, the multiprotein complex that unleashes amyloid-β1-42 (Aβ1-42) from the amyloidβ precursor protein (AβPP), seems an obvious place to start if you want to develop a drug to prevent or slow Alzheimer disease. There’s only one problem; γ-secretase is also needed to process other proteins, including the cell surface receptors Notch and ErbB-4, so axing γ-secretase may not be the best approach. But don’t fret; all may not be lost. Recent reports have suggested that rather than cripple the protease, some NSAIDs, or nonsteroidal antiinflammatory drugs, jolt it sufficiently so that it cuts AβPP at a different spot, yielding a shorter, less toxic amyloid peptide, Aβ1-38 (see ARF related news story and ARF news story). Now, reporting in this month’s Nature Medicine, researchers from the Alzheimer Research Unit at Massachusetts General Hospital offer a mechanistic explanation for this selective inhibition.

Brad Hyman, Brian Bacskai and colleagues did FRET. They used a nifty technique called fluorescence resonance energy transfer (FRET) to measure the conformation of presenilin 1 (PS1), the core protease of γ-secretase, in the presence or absence of different NSAIDs. FRET is capable of measuring the proximity of one fluorescent molecule to another, but in a variation on this theme, first author Alberto Lleo and colleagues labeled the N- and C-terminal ends of PS1 with fluorophores. The ends of the protease are sufficiently close, it turns out, so that photons emitted from the donor fluorophore at the N-terminus are absorbed by, and excite, the acceptor fluorophore at the C-terminus, allowing FRET to occur intermolecularly.

FRET depends on proximity, but Lleo found that resonance transfer between the N- and C-termini of PS1 was less efficient in the presence of ibuprofen. The fluorescence lifetime of the donor jumped from 1,800 picoseconds to 2,200 picoseconds when the NSAID was added to cells expressing PS1 and AβPP, indicating that the N- and C-terminal ends of presenilin had been pushed apart. Significantly, aspirin and naproxen—NSAIDs that have not been found to alter AβPP cleavage (see ARF related news story on the failure of a naproxen clinical trial)—had no effect on the process.

The authors also found that ibuprofen (both optical isomers) reduced proteolysis at the 42 position by about 50 percent, and that this was compensated by cleavage, yielding peptides shorter than 40 amino acids. They also found that Notch processing was unaffected. To determine if NSAIDs have similar effects in vivo, Lleo and colleagues fed mice an ibuprofen diet for 15 days, fixed the brain tissue, and used fluorescent labeled antibodies to the N- and C-termini of PS1 to measure their proximity. Again, FRET showed that in PS1 exposed to ibuprofen, a conformational change pushed the termini apart.

“Our results suggest that NSAIDs have an allosteric effect on γ-secretase,” write the authors, and they offer a model whereby the distance between the N- and C-termini of PS1 is crucial in determining the proteolytic cleavage site of AβPP. This is welcome news for those who are trying to find drugs that selectively reduce Aβ1-42 but do not affect processing of other γ-secretase targets. It also helps explain why NSAIDs that do not affect AβPP processing have so far proven ineffective in halting or preventing AD (see ARF related news story and ARF news story), while ibuprofen may prove better suited (see ARF related news story).—Tom Fagan

Comments

  1. The findings are significant for three main reasons:

    First: They provide strong evidence that NSAIDs bind to and modulate γ-secretase function in the brain. This extends findings from Dr. Koo's and my laboratories (Weggen et al., Eriksen et al. references in paper), Greg Cole's laboratory (Lim et al.), Dave Morgan's Lab (Jantzen et al.), and Neill Kowall's laboratory (Yan et al.) that show that some NSAIDs 1) reduce Aβ deposition in transgenic mice, 2) acutely lower Aβ42 levels in the brain, and 3) appear to target γ-secretase.

    Second: What this study appears to show is that despite concentrations in the brain that are well below the levels needed to lower Aβ42 in cell culture (Eriksen et al.), certain NSAIDs appear capable of altering PS conformation in the brain. This would support the notion that NSAIDs are concentrated in the compartments in the brain where γ-secretase resides. This provides direct evidence that certain NSIADs could lower Aβ42 in the brain.

    Third: They utilize the FLIM technique to show that the drugs bind and induce a conformational change in γ-secretase. This is a really cool demonstration of how imaging techniques can be used to view molecular alterations.

    The notion that this technique can be used for screening is not unreasonable. However, one would want to show that it actually works with a much larger number of compounds. In this case, we can directly look for compounds that lower Aβ42 so in this instance there are more direct alternatives. However, there may be cases where it would be preferable to analyze structural changes by this technique (for example when one is trying to inhibit a protein-protein interaction) …

    References:

    . A subset of NSAIDs lower amyloidogenic Abeta42 independently of cyclooxygenase activity. Nature. 2001 Nov 8;414(6860):212-6. PubMed.

    . Ibuprofen suppresses plaque pathology and inflammation in a mouse model for Alzheimer's disease. J Neurosci. 2000 Aug 1;20(15):5709-14. PubMed.

    . Microglial activation and beta -amyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal anti-inflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice. J Neurosci. 2002 Mar 15;22(6):2246-54. PubMed.

    . Anti-inflammatory drug therapy alters beta-amyloid processing and deposition in an animal model of Alzheimer's disease. J Neurosci. 2003 Aug 20;23(20):7504-9. PubMed.

    . NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower Abeta 42 in vivo. J Clin Invest. 2003 Aug;112(3):440-9. PubMed.

    . Evidence that nonsteroidal anti-inflammatory drugs decrease amyloid beta 42 production by direct modulation of gamma-secretase activity. J Biol Chem. 2003 Aug 22;278(34):31831-7. PubMed.

  2. Allosteric modulation of γ-secretase activity offers a new avenue for specific inhibition of Aβ42 production
    Using fluorescence lifetime imaging (FLIM), the authors demonstrate that certain NSAIDs known to specifically reduce Aβ42 levels without altering total Aβ induce a structural rearrangement in the catalytic center of γ-secretase (presenilin, PS). Binding of these compounds also alters the interaction between PS1 and APP, thus suggesting that the allosteric rearrangement of PS allows APP to be cleaved at an alternate site. These findings are supported by recent enzyme kinetic studies that demonstrate that sulindac sulfide and R-flurbiprofen act as allosteric modulators of γ-secretase (Beher et al., 2004).

    Identification of the allosteric site at which NSAIDs bind should facilitate the generation of highly specific inhibitors of Aβ42 generation and thus minimize unwanted inhibition of the γ-secretase processing of other important substrates.

    References:

    . Selected non-steroidal anti-inflammatory drugs and their derivatives target gamma-secretase at a novel site. Evidence for an allosteric mechanism. J Biol Chem. 2004 Oct 15;279(42):43419-26. Epub 2004 Aug 10 PubMed.

  3. I agree that this is a very impressive attempt to use the power of energy transfer techniques to study the effects of NSAIDs on the organization of the presenilin/γ-secretase complex. I am concerned that the fluorophores used to measure lifetimes could conceivably be two antibody lengths away from the target antigens of the presenilin molecules, since each fluorophore was attached to secondary antibodies that were then bound to primary anti-PS1 antibodies. Since there were differences in response as a result of exposure to specific NSAIDs, something interesting is going on, but the changes in PSI conformation proposed by the authors are only one of many possible explanations.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Anti-inflammatory Drugs Side-Step COX Cascade to Target Aβ
  2. New Orleans: Out Go Classic γ-Secretase Inhibitors, In Come More Dexterous NSAIDs?
  3. Stockholm: Bad News Official: The Rofecoxib and Naproxen Clinical Trial Has Failed
  4. Stockholm: Not All NSAIDs Are Equally Good When it Comes to Alzheimer’s
  5. Ibuprofen Reduces Plaques and Inflammation in "Hsiao" Mice

Further Reading

No Available Further Reading

Primary Papers

  1. . Nonsteroidal anti-inflammatory drugs lower Abeta42 and change presenilin 1 conformation. Nat Med. 2004 Oct;10(10):1065-6. PubMed.