- Email: [email protected]
Alzheimer’s disease: Channeling APP to non-amyloidogenic processing
BBRC Biochemical and Biophysical Research Communications 331 (2005) 375–378 www.elsevier.com/locate/ybbrc
Breakthroughs and Views
AlzheimerÕs disease: Channeling APP to non-amyloidogenic processing Bor Luen Tang * Department of Biochemistry, National University of Singapore, 8 Medical Drive, Singapore 117597, Singapore Received 27 February 2005 Available online 22 March 2005
Abstract A good number of pharmacologic agents have over the years been touted as potentially beneficial in either preventing the onset or delay the progression of AlzheimerÕs disease. These include compounds such as non-steroidal anti-inflammatory drugs (NSAIDs) (HMG-CoA reductase inhibitors (statins)) and flavonoids. The underlying mechanisms for the beneficial effect of these agents are by and large attributed to their ability to reduce b-amyloid (Ab) production and amyloid load in the brain, via inhibition of amyloidogenic c-secretase activity. Recent reports have now provided mechanistic insights as to how non-amyloidogenic processing might also be enhanced by these seemingly unrelated treatments. Intriguingly, this appears to involve the inhibition of the activity of small GTPase Rho and its effector, the Rho-associated kinase, ROCK. Dietary caloric restriction (CR) also enhances non-amyloidogenic processing of APP, and this may be part of a more general anti-aging effect of CR mediated by gene expression changes downstream of the activity of the histone deacetylase SIRT1. Ó 2005 Elsevier Inc. All rights reserved. Keywords: AlzheimerÕs disease; b-Amyloid; Non-steroidal anti-inflammatory drugs; Rho; Rho-associated protein kinase; Statins
Proteolytic processing of the amyloid precursor protein (APP) can proceed via two opposing paths, with vastly different outcomes (Fig. 1, see [1] for a recent review). The amyloidogenic path involves sequential cleavages by b-secretase (BACE) and the c-secretase complex, with the generation of b-amyloid (Ab) fragments—the etiological agents of Alzheimic pathology. The non-amyloidogenic path, however, involves APP cleavage by a-secretases [2] at a site which will preclude BACE cleavage, and which releases a neuroprotective sAPPa fragment [3,4]. Much attention has been paid to BACE and c-secretase in the development of anti-Alzheimic strategies. However, it has become clear in recent reports that the channeling of APP to the non-amyloidogenic pathway underlies at least part of the beneficial action of compounds such as non-steroidal anti-inflammatory drugs (NSAIDs) [5,6], HMG-CoA reductase inhibitors (statins) [7,8], flavonoids [9,10], as well as *
Fax: +65 67791453. E-mail address: [email protected].
0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.03.074
dietary caloric restriction [11]. Some recent reports also provided mechanistic insights as to how non-amyloidogenic processing might be enhanced by these seemingly unrelated treatments. Intriguingly, this appears to involve modulation of the activity of the small GTPase Rho and its effector, the Rho-associated kinase (ROCK).
The effect of NSAIDs and statins on the isoprenoid pathway, Rho/ROCK activity, and APP processing The notion of a potential beneficial effect of NSAIDs on AlzheimerÕs disease has been around for some time [5,6]. While it is clear that NSAIDsÕ ability to lower Ab42 levels could be attributed to its direct modulation of c-secretase activity [12,13], these drugs have also been shown to stimulate the secretion of sAPPa into the conditioned media of cultured cells [14]. Zhou et al. [15] had also recently demonstrated that NSAIDs may function by perturbing the isoprenoid pathway, particularly geranylgeranylation. In the authorsÕ experiments,
376
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
Fig. 1. Schematic diagram depicting the inhibition of Rho-ROCK axis as a possible underlying cause of NSAID and statin effects on the proteolytic fate of APP. The drugs could inhibit the isoprenylation of Rho, therefore its membrane attachment and ability to activate its downstream effector ROCK. Inhibition of the Rho-ROCK axis enhances the generation of sAPPa via indirect, unknown mechanisms (indicated as ?). These drugs may also have a direct inhibitory effect on the amyloidogenic pathway by inhibiting c-secretase. The net effect is a channeling of APP processing towards the non-amyloidogenic pathway and a decrease in b-amyloid (Ab) production. Caloric restriction (CR) enhances the production of sAPPa, probably via an increased expression of a-secretase (see text).
geranylgeranyl pyrophosphate (GGpp) (but not farnesyl pyrophosphate) treatment of SH-SY5Y cells expressing the Swedish APP mutant resulted in an increase in Ab42 production. This increase is blocked by the NSAIDs sulindac sulfide and ibuprofen. The downstream factor targeted by the drugs is delineated and narrowed down to the small GTPase Rho, as only the dominant-negative form of Rho, but not Cdc42 or Rac1, caused a decrease in Ab42. Inactivation of Rho by C3 transferase also reduced Ab42. In addition, treatment of both APP-expressing cells and mice with the ROCK inhibitor Y-27632 also reduced Ab42 both in vitro and in vivo. Furthermore, there is a clear correlation between the ability of individual NSAIDs to negatively regulate Rho activity and their Ab42 lowering capacity. Pedrini et al. [8], on the other hand, demonstrated that the a-secretase-mediated shedding of sAPPa induced by statins in Swedish APP-expressing N2a mouse neuroblastoma cells is modulated by ROCK. A farnesyl transferase inhibitor, FTI-1, promotes sAPPa shedding in a manner that is synergistic with statins. Contrastingly, arachidonic acid, which activates ROCK, reduced sAPPa shedding. A constitutively active ROCK mutant diminished sAPPa shedding from both untreated and statin-treated cells, whereas a kinase-dead ROCK mutant on its own
activated sAPPa shedding. It appears, therefore, that both NSAIDs and statins might enhance sAPPa production through the attenuation of a signaling pathway that is modulated by Rho and ROCK. Since sAPPa production implicates an increase in a-secretase processing, such preclusion of amyloidogenic, b-secretase processing could at least partly explain the ability of NSAIDs and statins to reduce Ab42 levels.
NSAIDs and statins may also affect brain inflammation through the isoprenoid pathway and Rho/ROCK The benefits of NSAIDs and statins in countering the progression of AlzheimerÕs go beyond reducing Ab42 production by neurons. AlzheimerÕs is often viewed as a neuroinflammatory disease and Ab42 is a potent activator of microglia-mediated CNS tissue inflammation [16,17]. The anti-inflammatory action of NSAIDs is of course well known, but statins could also suppress neuronal inflammation. A recent report by Cordal and Landreth [18] showed that statin robustly inhibited Ab-stimulated expression of interleukin-1b and NO production in microglia and monocyte cells. This inhibition is apparently linked to the suppression of the isoprenoid
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
pathway and Rho activation, as mevalonate and GGpp attenuated this inhibition, whereas both a geranylgeranyl transferase inhibitor and C3 transferase block Ab-induced inflammation. The results suggest that, in addition to reducing Ab production by neurons, statins may also counter the disease by attenuating Ab-induced inflammation. There are some caveats to the above finding. It should be pointed out that the clinical indication for use of statins as a treatment option for AlzheimerÕs is still controversial. Recent community-based prospective cohort study did not reveal a clear benefit of statin use in dementia [19]. It should also be noted that statins may not necessarily suppress microglia-mediated inflammation in vivo. Bi et al. [20] have found that statins may actually activate microglia in some manner in culture rat hippocampal slides, inducing changes in morphology and the upregulation of TNF-a. Statins, acting via suppression of the isoprenoid pathway, may also evoke the phosphorylation of tau, which may be undesirable as far as AlzheimerÕs disease progression is concerned [21]. While the results alluded to above implicate the Rho-ROCK pathway in modulating the prevalence of amyloidogenic versus non-amyloidogenic APP cleavage, the details in this connection are still missing (see Fig. 1). It is unclear if the drugs elevate a-secretasesÕ expression or activity, or they simply facilitated their subcellular colocalization with APP (thereby enhancing non-amyloidogenic processing). Neither APP nor a-secretases are known as direct targets of ROCKÕs kinase activity. Investigations on the details of how the Rho-ROCK axis modulates non-amyloidogenic APP cleavage and identification of the relevant intermediates in this pathway should now be given some research priority.
Caloric restriction and APP processing Another noteworthy recent finding is with regard to a possible mechanistic explanation as to how caloric restriction (CR) may be beneficial to AlzheimerÕs disease. Caloric intake [22,23] and metabolic defects in terms of insulin resistance [24,25] have been linked to AlzheimerÕs, but the mechanism underlying these links is not particularly clear. Wang et al. [11] demonstrated that a CR regimen in mice diminished Ab generation and amyloid plaque deposition (compared to control, well-fed mice). Importantly, the authors observed a larger than twofold increase in the concentration of brain sAPPa in CR animals compared to control. This increase can potentially be accounted for by a statistically significant 30% increase in ADAM10 levels in the CR mice over fed mice. There is also a statistically significant increase in the levels of the insulin degrading enzyme (IDE), the loss of function of which may impair amyloid clearance [26].
377
It is unclear at the moment if CR could affect the isoprenoid pathway or Rho/ROCK in a manner that is reminiscent of NSAIDs and statins above. CR, however, has been known to prolong the life span of many model organisms [27,28]. One of the physiological effects of CR that could directly explain its anti-aging effect is the elevation of the nicotinamide adenine dinucleotide (NAD)dependent histone deacetylase Sir2p/SIRT1 [28]. SIRT1 modulates the functions of transcription factors such as p53 (reviewed in [29]), NF-jB [30], and the FOXO family of transcription factors (reviewed in [31]). These are all key regulators of transcriptional control of death/ survival genes. In mammals, SIRT1 activity modulates metabolic processes that contribute to aging-associated defects, such as fat mobilization [32]. Enhanced SIRT1 expression or activity in mammals may also be directly neuroprotective. One recent illustration of this notion is the increase in NAD generating nicotinamide mononucleotide adenylyltransferase 1 (Nmnat1) activity in Wallerian degeneration slow (wlds) mice that is apparently responsible for axonal protection in these animals [33]. In this mouse, neuroprotection is blocked by the Sir2 inhibitor sirtinol and when SIRT1 expression was knocked down by SIRT1 siRNA. It is conceivable that CR elevated SIRT1 could have resulted in changes in transcriptional profile of neurons and glial that is antiamyloidogenic, manifested by increases in a-secretase activities and other gene product that prevents or clears amyloid deposition.
Epilogue That multiple potentially beneficial drugs and physiologic adjustments all involve an enhancement of the non-amyloidogenic pathway of APP processing is clearly indicative of the therapeutic potential of targeting a-secretase in prevention or delaying disease progression for AlzheimerÕs. Recent findings that overexpression of ADAM10 in transgenic mice reducing amyloid plaque formation and alleviating spatial learning deficits [34] is supportive of this therapeutic strategy of actively channeling APP processing towards a-secretase cleavage. Finding out the details of how the RhoROCK axis mediates non-amyloidogenic APP cleavage could potentially reveal a further spectrum of novel molecular targets for therapeutic intervention. In this regard, it would be of obvious clinical interest to check if this apparent channeling of APP towards the non-amyloidogenic pathway by compounds other than NSAIDs and statins, as well as caloric restriction, also involves Rho-ROCK signaling. It would of course be prudent to initiate studies to look for compounds or dietary regimes that would alter gene expression profiles and metabolic activities that enhance the non-amyloidogenic or suppress the amyloidogenic paths of APP processing.
378
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
References [1] V. Wilquet, B. De Strooper, Amyloid-beta precursor protein processing in neurodegeneration, Curr. Opin. Neurobiol. 14 (2004) 582–588. [2] T.M. Allinson, E.T. Parkin, A.J. Turner, N.M. Hooper, ADAMs family members as amyloid precursor protein a-secretases, J. Neurosci. Res. 74 (2003) 342–352. [3] M.P. Mattson, Secreted forms of beta-amyloid precursor protein modulate dendritic outgrowth and calcium responses to glutamate in cultured embryonic hippocampal neurons, J. Neurobiol. 25 (1994) 439–450. [4] H. Meziane, J.C. Dodart, C. Mathis, S. Little, J. Clemens, S.M. Paul, A. Ungerer, Memory-enhancing effects of secreted forms of the beta-amyloid precursor protein in normal and amnestic mice, Proc. Natl. Acad. Sci. USA 95 (1998) 12683–12688. [5] J.J. Hoozemans, R. Veerhuis, A.J. Rozemuller, P. Eikelenboom, Non-steroidal anti-inflammatory drugs and cyclooxygenase in AlzheimerÕs disease, Curr. Drug Targets 4 (2003) 461–468. [6] C.A. Szekely, J.E. Thorne, P.P. Zandi, M. Ek, E. Messias, J.C. Breitner, S.N. Goodman, Nonsteroidal anti-inflammatory drugs for the prevention of AlzheimerÕs disease: a systematic review, Neuroepidemiology 23 (2004) 159–169. [7] S. Parvathy, M. Ehrlich, S. Pedrini, N. Diaz, L. Refolo, J.D. Buxbaum, A. Bogush, S. Petanceska, S. Gandy, Atorvastatininduced activation of AlzheimerÕs alpha secretase is resistant to standard inhibitors of protein phosphorylation-regulated ectodomain shedding, J. Neurochem. 90 (2004) 1005–1010. [8] S. Pedrini, T.L. Carter, G. Prendergast, S. Petanceska, M.E. Ehrlich, S. Gandy, Modulation of statin-activated shedding of Alzheimer APP ectodomain by ROCK, PLoS Med. 2 (2005) e18. [9] Y. Levites, T. Amit, S. Mandel, M.B. Youdim, Neuroprotection and neurorescue against Abeta toxicity and PKC-dependent release of nonamyloidogenic soluble precursor protein by green tea polyphenol ( )-epigallocatechin-3-gallate, FASEB J. 17 (2003) 952–954. [10] F. Colciaghi, B. Borroni, M. Zimmermann, C. Bellone, A. Longhi, A. Padovani, F. Cattabeni, Y. Christen, M. Di Luca, Amyloid precursor protein metabolism is regulated toward alphasecretase pathway by Ginkgo biloba extracts, Neurobiol. Dis. 16 (2004) 454–460. [11] J. Wang, L. Ho, W. Qin, R.B. Rocher, I. Seror, N. Humala, K. Maniar, G. Dolios, R. Wang, P.R. Hof, G.M. Pasinetti, Caloric restriction attenuates beta-amyloid neuropathology in a mouse model of AlzheimerÕs disease. FASEB J. (in press) (online 13 Jan 2005; 10.1096/fj.04-3182fje). [12] J.L. Eriksen, S.A. Sagi, T.E. Smith, S. Weggen, P. Das, D.C. McLendon, V.V. Ozols, K.W. Jessing, K.H. Zavitz, E.H. Koo, T.E. Golde, NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower Abeta42 in vivo, J. Clin. Invest. 112 (2003) 440–449. [13] S. Weggen, J.L. Eriksen, S.A. Sagi, C.U. Pietrzik, V. Ozols, A. Fauq, T.E. Golde, E.H. Koo, Evidence that nonsteroidal antiinflammatory drugs decrease amyloid b42 production by direct modulation of c-secretase activity, J. Biol. Chem. 278 (2003) 31831–31837. [14] Y. Avramovich, T. Amit, M.B. Youdim, Non-steroidal antiinflammatory drugs stimulate secretion of non-amyloidogenic precursor protein, J. Biol. Chem. 277 (2002) 31466–31473. [15] Y. Zhou, Y. Su, B. Li, F. Liu, J.W. Ryder, X. Wu, P.A. Gonzalez-DeWhitt, V. Gelfanova, J.E. Hale, P.C. May, S.M. Paul, B. Ni, Nonsteroidal anti-inflammatory drugs can lower amyloidogenic Abeta42 by inhibiting Rho, Science 302 (2003) 1215–1217. [16] W.J. Streit, Microglia and AlzheimerÕs disease pathogenesis, J. Neurosci. Res. 77 (2004) 1–8.
[17] W.J. Streit, R.E. Mrak, W.S. Griffin, Microglia and neuroinflammation: a pathological perspective, J. Neuroinflamm. 1 (2004) 14. [18] A. Cordal, G. Landreth, 3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors attenuate beta-amyloid-induced microglial inflammatory responses, J. Neurosci. 25 (2005) 299–307. [19] G. Li, R. Higdon, W.A. Kukull, E. Peskind, K. Van Valen Moore, D. Tsuang, G. Van Belle, W. McCormick, J.D. Bowen, L. Teri, G.D. Schellenberg, E.B. Larson, Statin therapy and risk of dementia in the elderly: a community-based prospective cohort study, Neurology 63 (2004) 1624–1628. [20] X. Bi, M. Baudry, J. Liu, Y. Yao, L. Fu, F. Brucher, G. Lynch, Inhibition of geranylgeranylation mediates the effects of 3hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors on microglia, J. Biol. Chem. 279 (2004) 48238–48245. [21] V. Meske, F. Albert, D. Richter, J. Schwarze, T.G. Ohm, Blockade of HMG-CoA reductase activity causes changes in microtubule-stabilizing protein tau via suppression of geranylgeranylpyrophosphate formation: implications for AlzheimerÕs disease, Eur. J. Neurosci. 17 (2003) 93–102. [22] J.A. Luchsinger, M.X. Tang, S. Shea, R. Mayeux, Caloric intake and the risk of AlzheimerÕs disease, Arch. Neurol. 59 (2002) 1258– 1263. [23] M.P. Mattson, S.L. Chan, W. Duan, Modification of brain aging and neurodegenerative disorders by genes, diet and behavior, Physiol. Rev. 82 (2002) 637–672. [24] L. Ho, W. Qin, P.N. Pompl, Z. Xiang, J. Wang, Z. Zhao, Y. Peng, G. Cambareri, A. Rocher, C.V. Mobbs, P.R. Hof, G.M. Pasinetti, Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of AlzheimerÕs disease, FASEB J. 18 (2004) 902–904. [25] N. Rasgon, L. Jarvik, Insulin resistance, affective disorders, and AlzheimerÕs disease: review and hypothesis, J. Gerontol. A: Biol. Sci. Med. Sci. 59 (2004) 178–183. [26] W. Farris, S. Mansourian, M.A. Leissring, E.A. Eckman, L. Bertram, C.B. Eckman, R.E. Tanzi, D.J. Selkoe, Partial loss-offunction mutations in insulin-degrading enzyme that induce diabetes also impair degradation of amyloid beta-protein, Am. J. Pathol. 164 (2004) 1425–1434. [27] S.D. Hursting, J.A. Lavigne, D. Berrigan, S.N. Perkins, J.C. Barrett, Calorie restriction, aging, and cancer prevention: mechanisms of action and applicability to humans, Annu. Rev. Med. 54 (2003) 131–152. [28] J. Koubova, L. Guarente, How does calorie restriction work?, Genes Dev. 17 (2003) 313–321. [29] J. Smith, Human Sir2 and the ÔsilencingÕ of p53 activity, Trends. Cell Biol. 12 (2002) 404–406. [30] F. Yeung, J.E. Hoberg, C.S. Ramsey, M.D. Keller, D.R. Jones, R.A. Frye, M.W. Mayo, Modulation of NF-kappaB-dependent transcription and cell survival by the SIRT1 deacetylase, EMBO J. 23 (2004) 2369–2380. [31] A. Antebi, Tipping the balance toward longevity, Dev. Cell 6 (2004) 315–316. [32] F. Picard, M. Kurtev, N. Chung, A. Topark-Ngarm, T. Senawong, R. Machado De Oliveira, M. Leid, M.W. McBurney, L. Guarente, Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-gamma, Nature 429 (2004) 771–776. [33] T. Araki, Y. Sasaki, J. Milbrandt, Increase nuclear NAD biosynthesis and SIRT1 activation prevent axonal degeneration, Science 305 (2004) 1010–1013. [34] R. Postina, A. Schroeder, I. Dewachter, J. Bohl, U. Schmitt, E. Kojro, C. Prinzen, K. Endres, C. Hiemke, M. Blessing, P. Flamez, A. Dequenne, E. Godaux, F. van Leuven, F. Fahrenholz, A disintegrin-metalloproteinase prevents amyloid plaque formation and hippocampal defects in an Alzheimer disease mouse model, J. Clin. Invest. 113 (2004) 1456–1464.
Breakthroughs and Views
AlzheimerÕs disease: Channeling APP to non-amyloidogenic processing Bor Luen Tang * Department of Biochemistry, National University of Singapore, 8 Medical Drive, Singapore 117597, Singapore Received 27 February 2005 Available online 22 March 2005
Abstract A good number of pharmacologic agents have over the years been touted as potentially beneficial in either preventing the onset or delay the progression of AlzheimerÕs disease. These include compounds such as non-steroidal anti-inflammatory drugs (NSAIDs) (HMG-CoA reductase inhibitors (statins)) and flavonoids. The underlying mechanisms for the beneficial effect of these agents are by and large attributed to their ability to reduce b-amyloid (Ab) production and amyloid load in the brain, via inhibition of amyloidogenic c-secretase activity. Recent reports have now provided mechanistic insights as to how non-amyloidogenic processing might also be enhanced by these seemingly unrelated treatments. Intriguingly, this appears to involve the inhibition of the activity of small GTPase Rho and its effector, the Rho-associated kinase, ROCK. Dietary caloric restriction (CR) also enhances non-amyloidogenic processing of APP, and this may be part of a more general anti-aging effect of CR mediated by gene expression changes downstream of the activity of the histone deacetylase SIRT1. Ó 2005 Elsevier Inc. All rights reserved. Keywords: AlzheimerÕs disease; b-Amyloid; Non-steroidal anti-inflammatory drugs; Rho; Rho-associated protein kinase; Statins
Proteolytic processing of the amyloid precursor protein (APP) can proceed via two opposing paths, with vastly different outcomes (Fig. 1, see [1] for a recent review). The amyloidogenic path involves sequential cleavages by b-secretase (BACE) and the c-secretase complex, with the generation of b-amyloid (Ab) fragments—the etiological agents of Alzheimic pathology. The non-amyloidogenic path, however, involves APP cleavage by a-secretases [2] at a site which will preclude BACE cleavage, and which releases a neuroprotective sAPPa fragment [3,4]. Much attention has been paid to BACE and c-secretase in the development of anti-Alzheimic strategies. However, it has become clear in recent reports that the channeling of APP to the non-amyloidogenic pathway underlies at least part of the beneficial action of compounds such as non-steroidal anti-inflammatory drugs (NSAIDs) [5,6], HMG-CoA reductase inhibitors (statins) [7,8], flavonoids [9,10], as well as *
Fax: +65 67791453. E-mail address: [email protected].
0006-291X/$ - see front matter Ó 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2005.03.074
dietary caloric restriction [11]. Some recent reports also provided mechanistic insights as to how non-amyloidogenic processing might be enhanced by these seemingly unrelated treatments. Intriguingly, this appears to involve modulation of the activity of the small GTPase Rho and its effector, the Rho-associated kinase (ROCK).
The effect of NSAIDs and statins on the isoprenoid pathway, Rho/ROCK activity, and APP processing The notion of a potential beneficial effect of NSAIDs on AlzheimerÕs disease has been around for some time [5,6]. While it is clear that NSAIDsÕ ability to lower Ab42 levels could be attributed to its direct modulation of c-secretase activity [12,13], these drugs have also been shown to stimulate the secretion of sAPPa into the conditioned media of cultured cells [14]. Zhou et al. [15] had also recently demonstrated that NSAIDs may function by perturbing the isoprenoid pathway, particularly geranylgeranylation. In the authorsÕ experiments,
376
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
Fig. 1. Schematic diagram depicting the inhibition of Rho-ROCK axis as a possible underlying cause of NSAID and statin effects on the proteolytic fate of APP. The drugs could inhibit the isoprenylation of Rho, therefore its membrane attachment and ability to activate its downstream effector ROCK. Inhibition of the Rho-ROCK axis enhances the generation of sAPPa via indirect, unknown mechanisms (indicated as ?). These drugs may also have a direct inhibitory effect on the amyloidogenic pathway by inhibiting c-secretase. The net effect is a channeling of APP processing towards the non-amyloidogenic pathway and a decrease in b-amyloid (Ab) production. Caloric restriction (CR) enhances the production of sAPPa, probably via an increased expression of a-secretase (see text).
geranylgeranyl pyrophosphate (GGpp) (but not farnesyl pyrophosphate) treatment of SH-SY5Y cells expressing the Swedish APP mutant resulted in an increase in Ab42 production. This increase is blocked by the NSAIDs sulindac sulfide and ibuprofen. The downstream factor targeted by the drugs is delineated and narrowed down to the small GTPase Rho, as only the dominant-negative form of Rho, but not Cdc42 or Rac1, caused a decrease in Ab42. Inactivation of Rho by C3 transferase also reduced Ab42. In addition, treatment of both APP-expressing cells and mice with the ROCK inhibitor Y-27632 also reduced Ab42 both in vitro and in vivo. Furthermore, there is a clear correlation between the ability of individual NSAIDs to negatively regulate Rho activity and their Ab42 lowering capacity. Pedrini et al. [8], on the other hand, demonstrated that the a-secretase-mediated shedding of sAPPa induced by statins in Swedish APP-expressing N2a mouse neuroblastoma cells is modulated by ROCK. A farnesyl transferase inhibitor, FTI-1, promotes sAPPa shedding in a manner that is synergistic with statins. Contrastingly, arachidonic acid, which activates ROCK, reduced sAPPa shedding. A constitutively active ROCK mutant diminished sAPPa shedding from both untreated and statin-treated cells, whereas a kinase-dead ROCK mutant on its own
activated sAPPa shedding. It appears, therefore, that both NSAIDs and statins might enhance sAPPa production through the attenuation of a signaling pathway that is modulated by Rho and ROCK. Since sAPPa production implicates an increase in a-secretase processing, such preclusion of amyloidogenic, b-secretase processing could at least partly explain the ability of NSAIDs and statins to reduce Ab42 levels.
NSAIDs and statins may also affect brain inflammation through the isoprenoid pathway and Rho/ROCK The benefits of NSAIDs and statins in countering the progression of AlzheimerÕs go beyond reducing Ab42 production by neurons. AlzheimerÕs is often viewed as a neuroinflammatory disease and Ab42 is a potent activator of microglia-mediated CNS tissue inflammation [16,17]. The anti-inflammatory action of NSAIDs is of course well known, but statins could also suppress neuronal inflammation. A recent report by Cordal and Landreth [18] showed that statin robustly inhibited Ab-stimulated expression of interleukin-1b and NO production in microglia and monocyte cells. This inhibition is apparently linked to the suppression of the isoprenoid
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
pathway and Rho activation, as mevalonate and GGpp attenuated this inhibition, whereas both a geranylgeranyl transferase inhibitor and C3 transferase block Ab-induced inflammation. The results suggest that, in addition to reducing Ab production by neurons, statins may also counter the disease by attenuating Ab-induced inflammation. There are some caveats to the above finding. It should be pointed out that the clinical indication for use of statins as a treatment option for AlzheimerÕs is still controversial. Recent community-based prospective cohort study did not reveal a clear benefit of statin use in dementia [19]. It should also be noted that statins may not necessarily suppress microglia-mediated inflammation in vivo. Bi et al. [20] have found that statins may actually activate microglia in some manner in culture rat hippocampal slides, inducing changes in morphology and the upregulation of TNF-a. Statins, acting via suppression of the isoprenoid pathway, may also evoke the phosphorylation of tau, which may be undesirable as far as AlzheimerÕs disease progression is concerned [21]. While the results alluded to above implicate the Rho-ROCK pathway in modulating the prevalence of amyloidogenic versus non-amyloidogenic APP cleavage, the details in this connection are still missing (see Fig. 1). It is unclear if the drugs elevate a-secretasesÕ expression or activity, or they simply facilitated their subcellular colocalization with APP (thereby enhancing non-amyloidogenic processing). Neither APP nor a-secretases are known as direct targets of ROCKÕs kinase activity. Investigations on the details of how the Rho-ROCK axis modulates non-amyloidogenic APP cleavage and identification of the relevant intermediates in this pathway should now be given some research priority.
Caloric restriction and APP processing Another noteworthy recent finding is with regard to a possible mechanistic explanation as to how caloric restriction (CR) may be beneficial to AlzheimerÕs disease. Caloric intake [22,23] and metabolic defects in terms of insulin resistance [24,25] have been linked to AlzheimerÕs, but the mechanism underlying these links is not particularly clear. Wang et al. [11] demonstrated that a CR regimen in mice diminished Ab generation and amyloid plaque deposition (compared to control, well-fed mice). Importantly, the authors observed a larger than twofold increase in the concentration of brain sAPPa in CR animals compared to control. This increase can potentially be accounted for by a statistically significant 30% increase in ADAM10 levels in the CR mice over fed mice. There is also a statistically significant increase in the levels of the insulin degrading enzyme (IDE), the loss of function of which may impair amyloid clearance [26].
377
It is unclear at the moment if CR could affect the isoprenoid pathway or Rho/ROCK in a manner that is reminiscent of NSAIDs and statins above. CR, however, has been known to prolong the life span of many model organisms [27,28]. One of the physiological effects of CR that could directly explain its anti-aging effect is the elevation of the nicotinamide adenine dinucleotide (NAD)dependent histone deacetylase Sir2p/SIRT1 [28]. SIRT1 modulates the functions of transcription factors such as p53 (reviewed in [29]), NF-jB [30], and the FOXO family of transcription factors (reviewed in [31]). These are all key regulators of transcriptional control of death/ survival genes. In mammals, SIRT1 activity modulates metabolic processes that contribute to aging-associated defects, such as fat mobilization [32]. Enhanced SIRT1 expression or activity in mammals may also be directly neuroprotective. One recent illustration of this notion is the increase in NAD generating nicotinamide mononucleotide adenylyltransferase 1 (Nmnat1) activity in Wallerian degeneration slow (wlds) mice that is apparently responsible for axonal protection in these animals [33]. In this mouse, neuroprotection is blocked by the Sir2 inhibitor sirtinol and when SIRT1 expression was knocked down by SIRT1 siRNA. It is conceivable that CR elevated SIRT1 could have resulted in changes in transcriptional profile of neurons and glial that is antiamyloidogenic, manifested by increases in a-secretase activities and other gene product that prevents or clears amyloid deposition.
Epilogue That multiple potentially beneficial drugs and physiologic adjustments all involve an enhancement of the non-amyloidogenic pathway of APP processing is clearly indicative of the therapeutic potential of targeting a-secretase in prevention or delaying disease progression for AlzheimerÕs. Recent findings that overexpression of ADAM10 in transgenic mice reducing amyloid plaque formation and alleviating spatial learning deficits [34] is supportive of this therapeutic strategy of actively channeling APP processing towards a-secretase cleavage. Finding out the details of how the RhoROCK axis mediates non-amyloidogenic APP cleavage could potentially reveal a further spectrum of novel molecular targets for therapeutic intervention. In this regard, it would be of obvious clinical interest to check if this apparent channeling of APP towards the non-amyloidogenic pathway by compounds other than NSAIDs and statins, as well as caloric restriction, also involves Rho-ROCK signaling. It would of course be prudent to initiate studies to look for compounds or dietary regimes that would alter gene expression profiles and metabolic activities that enhance the non-amyloidogenic or suppress the amyloidogenic paths of APP processing.
378
B.L. Tang / Biochemical and Biophysical Research Communications 331 (2005) 375–378
References [1] V. Wilquet, B. De Strooper, Amyloid-beta precursor protein processing in neurodegeneration, Curr. Opin. Neurobiol. 14 (2004) 582–588. [2] T.M. Allinson, E.T. Parkin, A.J. Turner, N.M. Hooper, ADAMs family members as amyloid precursor protein a-secretases, J. Neurosci. Res. 74 (2003) 342–352. [3] M.P. Mattson, Secreted forms of beta-amyloid precursor protein modulate dendritic outgrowth and calcium responses to glutamate in cultured embryonic hippocampal neurons, J. Neurobiol. 25 (1994) 439–450. [4] H. Meziane, J.C. Dodart, C. Mathis, S. Little, J. Clemens, S.M. Paul, A. Ungerer, Memory-enhancing effects of secreted forms of the beta-amyloid precursor protein in normal and amnestic mice, Proc. Natl. Acad. Sci. USA 95 (1998) 12683–12688. [5] J.J. Hoozemans, R. Veerhuis, A.J. Rozemuller, P. Eikelenboom, Non-steroidal anti-inflammatory drugs and cyclooxygenase in AlzheimerÕs disease, Curr. Drug Targets 4 (2003) 461–468. [6] C.A. Szekely, J.E. Thorne, P.P. Zandi, M. Ek, E. Messias, J.C. Breitner, S.N. Goodman, Nonsteroidal anti-inflammatory drugs for the prevention of AlzheimerÕs disease: a systematic review, Neuroepidemiology 23 (2004) 159–169. [7] S. Parvathy, M. Ehrlich, S. Pedrini, N. Diaz, L. Refolo, J.D. Buxbaum, A. Bogush, S. Petanceska, S. Gandy, Atorvastatininduced activation of AlzheimerÕs alpha secretase is resistant to standard inhibitors of protein phosphorylation-regulated ectodomain shedding, J. Neurochem. 90 (2004) 1005–1010. [8] S. Pedrini, T.L. Carter, G. Prendergast, S. Petanceska, M.E. Ehrlich, S. Gandy, Modulation of statin-activated shedding of Alzheimer APP ectodomain by ROCK, PLoS Med. 2 (2005) e18. [9] Y. Levites, T. Amit, S. Mandel, M.B. Youdim, Neuroprotection and neurorescue against Abeta toxicity and PKC-dependent release of nonamyloidogenic soluble precursor protein by green tea polyphenol ( )-epigallocatechin-3-gallate, FASEB J. 17 (2003) 952–954. [10] F. Colciaghi, B. Borroni, M. Zimmermann, C. Bellone, A. Longhi, A. Padovani, F. Cattabeni, Y. Christen, M. Di Luca, Amyloid precursor protein metabolism is regulated toward alphasecretase pathway by Ginkgo biloba extracts, Neurobiol. Dis. 16 (2004) 454–460. [11] J. Wang, L. Ho, W. Qin, R.B. Rocher, I. Seror, N. Humala, K. Maniar, G. Dolios, R. Wang, P.R. Hof, G.M. Pasinetti, Caloric restriction attenuates beta-amyloid neuropathology in a mouse model of AlzheimerÕs disease. FASEB J. (in press) (online 13 Jan 2005; 10.1096/fj.04-3182fje). [12] J.L. Eriksen, S.A. Sagi, T.E. Smith, S. Weggen, P. Das, D.C. McLendon, V.V. Ozols, K.W. Jessing, K.H. Zavitz, E.H. Koo, T.E. Golde, NSAIDs and enantiomers of flurbiprofen target gamma-secretase and lower Abeta42 in vivo, J. Clin. Invest. 112 (2003) 440–449. [13] S. Weggen, J.L. Eriksen, S.A. Sagi, C.U. Pietrzik, V. Ozols, A. Fauq, T.E. Golde, E.H. Koo, Evidence that nonsteroidal antiinflammatory drugs decrease amyloid b42 production by direct modulation of c-secretase activity, J. Biol. Chem. 278 (2003) 31831–31837. [14] Y. Avramovich, T. Amit, M.B. Youdim, Non-steroidal antiinflammatory drugs stimulate secretion of non-amyloidogenic precursor protein, J. Biol. Chem. 277 (2002) 31466–31473. [15] Y. Zhou, Y. Su, B. Li, F. Liu, J.W. Ryder, X. Wu, P.A. Gonzalez-DeWhitt, V. Gelfanova, J.E. Hale, P.C. May, S.M. Paul, B. Ni, Nonsteroidal anti-inflammatory drugs can lower amyloidogenic Abeta42 by inhibiting Rho, Science 302 (2003) 1215–1217. [16] W.J. Streit, Microglia and AlzheimerÕs disease pathogenesis, J. Neurosci. Res. 77 (2004) 1–8.
[17] W.J. Streit, R.E. Mrak, W.S. Griffin, Microglia and neuroinflammation: a pathological perspective, J. Neuroinflamm. 1 (2004) 14. [18] A. Cordal, G. Landreth, 3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors attenuate beta-amyloid-induced microglial inflammatory responses, J. Neurosci. 25 (2005) 299–307. [19] G. Li, R. Higdon, W.A. Kukull, E. Peskind, K. Van Valen Moore, D. Tsuang, G. Van Belle, W. McCormick, J.D. Bowen, L. Teri, G.D. Schellenberg, E.B. Larson, Statin therapy and risk of dementia in the elderly: a community-based prospective cohort study, Neurology 63 (2004) 1624–1628. [20] X. Bi, M. Baudry, J. Liu, Y. Yao, L. Fu, F. Brucher, G. Lynch, Inhibition of geranylgeranylation mediates the effects of 3hydroxy-3-methylglutaryl (HMG)-CoA reductase inhibitors on microglia, J. Biol. Chem. 279 (2004) 48238–48245. [21] V. Meske, F. Albert, D. Richter, J. Schwarze, T.G. Ohm, Blockade of HMG-CoA reductase activity causes changes in microtubule-stabilizing protein tau via suppression of geranylgeranylpyrophosphate formation: implications for AlzheimerÕs disease, Eur. J. Neurosci. 17 (2003) 93–102. [22] J.A. Luchsinger, M.X. Tang, S. Shea, R. Mayeux, Caloric intake and the risk of AlzheimerÕs disease, Arch. Neurol. 59 (2002) 1258– 1263. [23] M.P. Mattson, S.L. Chan, W. Duan, Modification of brain aging and neurodegenerative disorders by genes, diet and behavior, Physiol. Rev. 82 (2002) 637–672. [24] L. Ho, W. Qin, P.N. Pompl, Z. Xiang, J. Wang, Z. Zhao, Y. Peng, G. Cambareri, A. Rocher, C.V. Mobbs, P.R. Hof, G.M. Pasinetti, Diet-induced insulin resistance promotes amyloidosis in a transgenic mouse model of AlzheimerÕs disease, FASEB J. 18 (2004) 902–904. [25] N. Rasgon, L. Jarvik, Insulin resistance, affective disorders, and AlzheimerÕs disease: review and hypothesis, J. Gerontol. A: Biol. Sci. Med. Sci. 59 (2004) 178–183. [26] W. Farris, S. Mansourian, M.A. Leissring, E.A. Eckman, L. Bertram, C.B. Eckman, R.E. Tanzi, D.J. Selkoe, Partial loss-offunction mutations in insulin-degrading enzyme that induce diabetes also impair degradation of amyloid beta-protein, Am. J. Pathol. 164 (2004) 1425–1434. [27] S.D. Hursting, J.A. Lavigne, D. Berrigan, S.N. Perkins, J.C. Barrett, Calorie restriction, aging, and cancer prevention: mechanisms of action and applicability to humans, Annu. Rev. Med. 54 (2003) 131–152. [28] J. Koubova, L. Guarente, How does calorie restriction work?, Genes Dev. 17 (2003) 313–321. [29] J. Smith, Human Sir2 and the ÔsilencingÕ of p53 activity, Trends. Cell Biol. 12 (2002) 404–406. [30] F. Yeung, J.E. Hoberg, C.S. Ramsey, M.D. Keller, D.R. Jones, R.A. Frye, M.W. Mayo, Modulation of NF-kappaB-dependent transcription and cell survival by the SIRT1 deacetylase, EMBO J. 23 (2004) 2369–2380. [31] A. Antebi, Tipping the balance toward longevity, Dev. Cell 6 (2004) 315–316. [32] F. Picard, M. Kurtev, N. Chung, A. Topark-Ngarm, T. Senawong, R. Machado De Oliveira, M. Leid, M.W. McBurney, L. Guarente, Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-gamma, Nature 429 (2004) 771–776. [33] T. Araki, Y. Sasaki, J. Milbrandt, Increase nuclear NAD biosynthesis and SIRT1 activation prevent axonal degeneration, Science 305 (2004) 1010–1013. [34] R. Postina, A. Schroeder, I. Dewachter, J. Bohl, U. Schmitt, E. Kojro, C. Prinzen, K. Endres, C. Hiemke, M. Blessing, P. Flamez, A. Dequenne, E. Godaux, F. van Leuven, F. Fahrenholz, A disintegrin-metalloproteinase prevents amyloid plaque formation and hippocampal defects in an Alzheimer disease mouse model, J. Clin. Invest. 113 (2004) 1456–1464.