Mice lacking both cathepsin B and L, two lysosomal proteases that have been implicated in the development of Alzheimer's disease, suffer a profound loss of neurons soon after birth, Ute Felbor and colleagues at Harvard Medical School and elsewhere report in the June 11 PNAS, now available online. Earlier knockout mouse strains missing only one of these two cysteine proteases showed few symptoms, suggesting that cathepsin B and L compensate for each other. Taking out both, however, caused unprecedented brain atrophy, the authors write. "The phenotype of these mice was a surprise to us and very dramatic," senior author Bjorn Olsen said. The study ties into ongoing research on whether age-related changes in cathepsin activity play a role in AD, or in neurodegeneration generally.

Felbor et al. crossed mice lacking cathepsin B with another strain lacking cathepsin L. Among the offspring, only those missing all four cathepsin alleles showed the described phenotype. They stopped growing soon after birth and died emaciated after about two weeks. Those reared without competition from healthy littermates survived to around day 50, allowing for more extensive analysis of postnatal development. Curiously, even though cathepsin B and L are expressed in most organs, only the brain was severely affected. The cerebral cortex and cerebellum showed pronounced atrophy and massive apoptosis. Astrogliosis accompanied this neurodegeneration, as seen in humans. Degenerating neurons contained membranous inclusions around their nuclei and in their axons that looked like remnants of lysosomes in the electron microscope.

The relevance of this paper for AD research is unclear. Previous work has implicated cathepsin proteases in AD or, more broadly, in age-related neurodegeneration. Yet clearly delineating the in-vivo function for these enzymes is difficult because their substrate specificities overlap. For example, cathepsins D and E have been implicated in AβPP processing (Gruninger-Leitch et al, 2000), and cathepsin B released from activated microglia is reported to induce apoptosis in neurons (Kingham & Pocock, 2001). Cathepsin B and L inhibitors infused into brain ventricles or applied to cultured brain slices cause age-related changes (Yong et al, 1999), and when applied to hippocampal slices from ApoE-knockout mice they cause AD-like neurofibrillary tangles, suggesting lysosomal dysfunction plays a role in the development of AD (Bi et al., 2001). On the genetic front, earlier studies reporting links between various cathepsin polymorphisms and increased AD risk have not been widely replicated (Bertram et al., 2000; Bhojak et al., 2001).—Gabrielle Strobel


  1. While there may be some evidence that cathepsins might share some BACE1-like properties, our BACE1 KO neurons/mice clearly demonstrated that it is unlikely there exist other BACE1 homologues as far as processing AbPP is concerned. Moreover, there is no evidence that cathepsin B/L are involved in AbPP processing. So I don't think that the information in this paper has any relevance for AD.

  2. I have to take issue on two accounts with Phil Wong's assessment of the Felbor et al. study. First, relevance to Alzheimer’s disease cannot be judged simply in terms of a protease's effect on AβPP processing. Neurodegeneration, the likely basis for memory loss in AD, almost certainly involves proteases that are not AβPP secretases and may be regulated by disease-related factors that are independent of AβPP and its metabolites. Growing evidence implicates the lysosomal system, including aging-related cysteine protease dysfunction in the pathogenesis of Alzheimer's disease (reviewed in J. Alzheimers Dis., 2001) and aging (J Neurochem., 2002).

    Second, even in evaluating relevance to AβPP processing in AD, a direct action of a protease on AβPP should be regarded as only one of several ways in which proteases may be influential. Other ways include modifying AβPP trafficking and modulating the activity of AβPP secretases directly or indirectly. For example, calpains are not considered AβPP secretases, but their activity regulates AβPP delivery to the cell surface and strongly modulates Aβ generation (J Biol Chem., 2002). Overexpression of the cation-dependent mannose-6-phosphate receptor, as occurs in AD brain, increases delivery of a limited set of mannosylated cathepsins to early endosomes, promoting Aβ production in the endocytic pathway (J Biol Chem., 2002;277:5299-307).

    Unless or until the exclusive role of Aβ in the death of neurons in AD is conclusively established and the mediating proteases are defined, any model of neurodegeneration involving proteases known to be altered in aging and Alzheimer's disease has the potential to inform us about disease pathogenesis.


    . The neuronal endosomal-lysosomal system in Alzheimer's disease. J Alzheimers Dis. 2001 Feb;3(1):97-107. PubMed.

    . The neuropathogenic contributions of lysosomal dysfunction. J Neurochem. 2002 Nov;83(3):481-9. PubMed.

    . Calpain activity regulates the cell surface distribution of amyloid precursor protein. Inhibition of calpains enhances endosomal generation of beta-cleaved C-terminal APP fragments. J Biol Chem. 2002 Sep 27;277(39):36415-24. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Rapid induction of intraneuronal neurofibrillary tangles in apolipoprotein E-deficient mice. Proc Natl Acad Sci U S A. 2001 Jul 17;98(15):8832-7. PubMed.

External Citations

  1. Gruninger-Leitch et al, 2000
  2. Kingham & Pocock, 2001
  3. Yong et al, 1999
  4. Bertram et al., 2000
  5. Bhojak et al., 2001

Further Reading

No Available Further Reading

Primary Papers

  1. . Neuronal loss and brain atrophy in mice lacking cathepsins B and L. Proc Natl Acad Sci U S A. 2002 Jun 11;99(12):7883-8. PubMed.