New data on an assortment of therapeutic possibilities appear this week, featuring the clearance of vascular amyloid by passive immunization, a calpain protease inhibitor that seems to block the synaptic effects of soluble Aβ, and a new approach to inhibiting tau phosphorylation by stimulating its glycosylation. Here follows a roundup of the studies, ranked from most clinically advanced candidate to the newest entrant.

The strategies of active and passive vaccination against Aβ have shown a lot of promise, but also present enormous complexities. Human clinical trials continue, as do persistent questions about how to design vaccines to have the best chance of enhancing cognitive function with the fewest side effects. One problem, observed in several mouse models of AD, is that mobilization of parenchymal plaques by Aβ antibodies results in an increase in vascular Aβ deposition and the appearance of microhemorrhages (see ARF related news story; Wilcock et al., 2004; Racke et al., 2005), although how these events are related remains unclear.

To try to sort out the relationship between parenchymal plaque clearance, vascular amyloid, and the risk of hemorrhages, Dora Games and colleagues from Elan Pharmaceuticals, South San Francisco, California, looked at the effect of Aβ antibodies on vascular amyloid and microhemorrhages in PDAPP mice. The work, published in the July 2 issue of the Journal of Neuroscience, shows that an antibody targeted to the N-terminal region of the peptide can clear vascular Aβ and does not cause microhemorrhage when used at moderate doses. The results raise the possibility that current active and passive immunization protocols, which involve N-terminal reactive antibodies, could also result in clearance of vascular Aβ, and that Aβ can be removed from the vasculature without triggering vessel leakage.

The research team, headed up by first author Sally Schroeter, compared the effects of chronic six-month treatment of year-old mice with two different Aβ antibodies. 3D6, which recognizes the amino-terminal 5 residues of Aβ, binds to plaques and clears parenchymal Aβ deposits, while 266, which recognizes central region residues 16-23, binds to soluble AAβ but does not clear plaques. The investigators found that 3D6 cleared vascular Aβ in a dose-dependent manner, while 266 did not. At the highest 3D6 dose (3 mg/kg), vascular Aβ was nearly completely prevented or cleared in the mice. At lower doses, the antibody caused a partial removal of vascular amyloid. Only the high dose of 3D6 was associated with a significantly elevated incidence of microhemorrhage as indicated by hemosiderin staining. The authors conclude that 3D6 can clear vascular amyloid, and that lowering antibody exposure lowers the risk of hemorrhage, presumably by slowing the clearance process.

One caveat to the study is that PDAPP mice deposit only small amounts of vascular amyloid compared to some other mouse models and to humans with cerebral amyloid angiopathy, which is associated with spontaneous hemorrhage. The study leaves open the question of what happens to vessels more heavily laden with amyloid, a situation that occurs in the majority of AD cases.

Taking Aim at Synaptic Changes
A second paper approaches treatment from the synaptic angle, providing evidence that inhibition of the calcium-activated protease calpain might be a way to prevent the impairment in synaptic function that occurs in AD. In addition to demonstrating the promising effects of two calpain inhibitors in vitro, the researchers, from the lab of Ottavio Arancio at Columbia University in New York, show that calpain inhibition improves cognitive performance on two different memory tests in the APP/PS1 mouse model of AD. The results appeared July 1 in the Journal of Clinical Investigation.

The calpain family of proteases has been implicated in AD in multiple ways. Calpain 1 is located at synapses, and its activity is increased in AD brain. Substrates for calpains include a host of proteins that play a role in APP production and tau phosphorylation, and calpain cleaves the Cdk5 regulator p35 to the constitutively active p25 form, which is also elevated in AD brain and appears to be involved in neurodegeneration.

In the new study, first author Fabrizio Trinchese and colleagues found that two different calpain inhibitors were able to restore normal synaptic function to cultured hippocampal neurons from APP/PS1 mice. The neurons show elevated spontaneous neurotransmitter release and fail to respond to glutamate. Treatment of the cells with E64, a general inhibitor of cysteine proteases, or the calpain-specific, orally administrable inhibitor BDA-410 (Li et al., 2007), returned synaptic activity to normal. The inhibitors had no effect on normal cells.

Studies confirmed the restoration of normal synaptic activity and long-term potentiation (LTP) in hippocampal slices from seven-month-old APP/PS1 mice that had been treated for the previous five months with either inhibitor, suggesting the strategy works in vivo. The effect of the calpain inhibitors depended on the production of Aβ, not PS1 overexpression, since the same effect on LTP was seen in APP mice, which do not overexpress PS1. There was no effect of inhibitors on LTP in normal mice. In addition, BDA-410 was able to block the inhibition of LTP seen when normal hippocampal slices were perfused with soluble Aβ oligomers.

The synaptic normalization observed by electrophysiological means was also reflected in improved behavioral measures. The animals’ performance in a radial arm water maze or in an associative learning test was both normalized by treatment with E64 or BDA-410. There was no effect on normal mice.

These results suggest that calpain activation plays a role in the synaptic toxicity of Aβ oligomers. The inhibitors appear to work downstream of Aβ production, since treatment did not change the levels of Aβ or plaque load in the mice. Treatment was associated with normalization of markers of synaptic remodeling, including phosphorylation of the transcription factor and calpain target CREB, and redistribution of the synaptic protein synapsin I. “Collectively, these data strongly support the possibility that calpain inhibitors act by reestablishing the increase in pCREB [phosphorylated CREB], thus rescuing the impairment of synaptic plasticity caused by overexpression of the APP and PS1 transgenes,” the authors write.

The data support the further development of calpain inhibitors to treat AD, Arancio told ARF. He sees the calpain approach as complementary to therapies aimed at reducing Aβ levels. “We feel it’s possible to improve disease by acting downstream of Aβ, in addition to decreasing Aβ. If you improve memory, that’s what counts,” he said. In March, Arancio received an NIH grant to work with chemist Gregory Thatcher of the University of Illinois at Chicago to synthesize and test new calpain inhibitors, and he said they are trying the first compounds now. Their efforts appear especially timely in light of recent developments highlighting the potential role of abnormally high calcium levels in brain to AD (see ARF related news story).

Hitting the Sweet Spot: Tau Glycosylation
Inhibiting the pathological phosphorylation of the microtubule-associated protein tau could provide a way to detoxify AD tangles and treat other tauopathies. In this regard, kinase inhibitors have gotten a lot of attention (see ARF related news story). In this week’s online edition of Nature Chemical Biology, David Vocadlo and colleagues at Simon Fraser University, Burnaby, British Columbia, Canada, offer up an alternative way to block tau phosphorylation. Taking advantage of the reciprocal relationship between tau phosphorylation and O-linked glycosylation (the two modifications both occur on serine/threonine residues and thus are mutually exclusive; see Liu et al., 2004), Vocadlo’s group shows that boosting glycosylation with an inhibitor of the sugar-removing enzyme O-GlcNAcase inhibits tau phosphorylation.

In the new work, first author Scott Yuzwa and colleagues describe the structure-based design of a specific O-GlcNAcase inhibitor and its effects on tau in vivo. The compound, thiamet-G, caused large increases in O-GlcNAc-modified proteins in PC12 cells with no sign of toxicity. Increased glycosylation was associated with a two- to threefold reduction in tau phosphorylation at the pathological residues Ser396 and Thr231. Thiamet-G was orally available and crossed the blood-brain barrier in rats, where it resulted in increased total brain O-GlcNAc-modified proteins and decreased tau phosphorylation at Ser396, Thr231, and Ser422. Immunohistochemistry showed that tau phosphorylation was decreased in the hippocampus, alongside a general increase in O-GlcNAc-specific staining.

Still to be determined is whether the changes in tau phosphorylation observed in healthy rats can be achieved in animals with established tau pathology. The authors write that these experiments are now underway. Another concern is the large number of substrates for the target enzyme, a problem that “needs to be addressed by further study,” Vocadlo told ARF by e-mail. “A number of proteins are modified with O-GlcNAc, making on-target toxicity a potential complication. Some work suggests that O-GlcNAc is involved in regulating glucohomeostasis and modulating insulin resistance (see, e.g., Yang et al., 2008),” he wrote. In the meantime, the specificity of thiamet-G will aid chemical biology efforts to understand the function of O-GlcNAc in brain.—Pat McCaffrey

Comments

Make a Comment

To make a comment you must login or register.

Comments on News and Primary Papers

  1. Prevention Rather Than Cure of CAA May Be the Best Way Forward
    Increased severity of cerebral amyloid angiopathy (CAA) has been highlighted as a complication of immunotherapy for Alzheimer disease in both human subjects (1) and in transgenic mouse models (2). In their paper in The Journal of Neuroscience, Sally Schroeter et al. showed that passive immunization of 12-month-old PDAPP transgenic mice with the 3D6 antibody directed against the N-terminal of the Aβ molecule prevented deposition or cleared Aβ from artery walls in a dose-dependent manner. At the moment, however, it is not clear whether Aβ was eliminated from artery walls by macrophages, analogous to the removal of Aβ plaques from brain parenchyma by microglia (1), or by some other mechanism. Perivascular microhemorrhages were increased in animals treated with the higher dose of the antibody. Fewer hemorrhages were detected at lower doses of the 3D6 antibody, although clearance of Aβ was not as complete.

    By treating the PDAPP transgenic mice with passive immunization at the relatively early age of 12 months, Schroeter and her colleagues have emphasized the importance of preventing the deposition of Aβ in vessel walls rather than removing it. CAA appears to have two major complications. One is the replacement of smooth muscle cells by Aβ that may result in intracerebral hemorrhage or may interfere with autoregulation of cerebral blood flow. The other complication is the blockage of pathways by which interstitial fluid and solutes drain from the brain. Basement membranes between vascular smooth muscle cells are the perivascular route by which interstitial fluid and solutes, such as Aβ, drain out of the brain (3). In the early stages of CAA, fibrils of insoluble Aβ are deposited within vascular basement membranes and disrupt the structure of the basement membranes involved (4,5). The effects of basement membrane disruption in CAA on the drainage of fluid and solutes from the brain has not been quantified, but it does seem to be associated with deposition of Aβ plaques in gray matter and with increased fluid retention in cerebral white matter (6). It is probable that the elimination of other brain metabolites, in addition to Aβ, is blocked in CAA, leading to a loss of homeostasis of the neuronal environment and possible neuronal malfunction. Thus, it could be vital for normal functioning of the brain to preserve the structure of vascular basement membranes by preventing the deposition of Aβ. In this way, the integrity of the drainage pathways for solutes and fluid from the brain would be maintained. Preserving vascular basement membranes is one reason why the approach of Schroeter et al. in preventing CAA could be so valuable in the management of Alzheimer disease.

    References:

    . Neuropathology of human Alzheimer disease after immunization with amyloid-beta peptide: a case report. Nat Med. 2003 Apr;9(4):448-52. PubMed.

    . Passive immunotherapy against Abeta in aged APP-transgenic mice reverses cognitive deficits and depletes parenchymal amyloid deposits in spite of increased vascular amyloid and microhemorrhage. J Neuroinflammation. 2004 Dec 8;1(1):24. PubMed.

    . Solutes, but not cells, drain from the brain parenchyma along basement membranes of capillaries and arteries: significance for cerebral amyloid angiopathy and neuroimmunology. Neuropathol Appl Neurobiol. 2008 Apr;34(2):131-44. Epub 2008 Jan 16 PubMed.

    . Capillary and arterial cerebral amyloid angiopathy in Alzheimer's disease: defining the perivascular route for the elimination of amyloid beta from the human brain. Neuropathol Appl Neurobiol. 2003 Apr;29(2):106-17. PubMed.

    . Perivascular drainage of amyloid-beta peptides from the brain and its failure in cerebral amyloid angiopathy and Alzheimer's disease. Brain Pathol. 2008 Apr;18(2):253-66. PubMed.

    . Cortical and leptomeningeal cerebrovascular amyloid and white matter pathology in Alzheimer's disease. Mol Med. 2003 Mar-Apr;9(3-4):112-22. PubMed.

    View all comments by Roy Weller
  2. I recommend this paper as a persuasive examination of the normalizing effects of calpain inhibitors on synaptic and learning deficits in PS1/APP mice—but some clarification or further electrophysiological analysis is first needed to characterize the signaling deficits. In particular, I'm referring to increased frequency of mEPSCs (it would also be useful to see the traces) in the PS/APP cultured neurons (Fig 1D), and the normalized mEPSCs measured in glutamate in 1E. Since the baselines are vastly different between the wild-type and APP/PS1, it actually appears that the frequency of mEPSCs maxes out to a similar level across all conditions. There appears to be no increase in the APP/PS1 cells just because they are already releasing transmitter at a high level. Paired pulse facilitation experiments would have been helpful here to further evaluate presynaptic effects; one might expect a reduction in paired pulse facilitation (PPF) in the APP/PS1 cultures under these conditions, which is consistent with increased synaptic strength—which is not observed here. The input/output curves suggest reduced synaptic strength.

    The role of calcium in the presynaptic transmitter release dynamics is fundamental, and may affect calpain activation. As an aside, but perhaps noteworthy, is the inclusion of 60 micromolar calcium in the patch pipette used to record postsynaptic events. This concentration is 100 times the normal cytosolic levels in neurons, and I am curious how calpain activity may differ in the APP/PS1 neurons if more physiological levels were used.

    Obviously, these are complicated dynamics, but a few extra experiments to characterize the synaptic dynamics would have been very enlightening. Since the role of calpain in the neuropathology is evident, more detailed information is certainly welcomed.

    View all comments by Grace Stutzmann
  3. Inhibition of Alzheimer Neurofibrillary Degeneration by Inhibition of O-GlcNAcase: A Sweet Approach With Some Bitter Hurdles Ahead
    The development of a potent O-GlcNAcase inhibitor and its ability to inhibit abnormal hyperphosphorylation of tau by Yuzwa et al. (2008), while very promising, might at the same time produce contraindicated effects by inhibiting phosphorylation of PI-3 kinase cascade enzymes upstream of glycogen synthase kinase-3 (GSK3).

    Tau is a major microtubule-associated protein in the neuron. It is abnormally hyperphosphorylated and aggregated into neurofibrillary tangles in AD brains (Grundke-Iqbal et al., 1986a; Grundke-Iqbal et al., 1986b). Unlike normal tau, which promotes assembly of tubulin into microtubules and stabilizes them, the AD abnormally hyperphosphorylated tau sequesters normal microtubule associated proteins and inhibits microtubule assembly as well as self-assembling into paired helical filaments (Alonso et al., 1994; 2001a). Many studies have demonstrated that abnormal hyperphosphorylation of tau is crucial to neurodegeneration in AD and other tauopathies (Iqbal et al., 2005). Thus, inhibiting and/or reversing tau hyperphosphorylation has been one of the major objectives of research in the AD field.

    The exact causes leading to abnormal hyperphosphorylation of tau in AD are poorly understood. We and others have found that protein phosphatase 2A, the major tau phosphatase in the brain, is downregulated in AD brain (Gong et al., 1993; Gong et al., 1995; Vogelsberg-Ragaglia et al., 2001; Sontag et al., 2004; Liu, F et al., 2005; Liu, R et al., 2008; Zhou et al., 2008), suggesting that this downregulation may be a possible cause of tau hyperphosphorylation. In addition to being regulated by tau kinases and phosphatases, tau phosphorylation is also regulated by O-GlcNAcylation. In 2004, we found that human brain tau is modified by O-GlcNAc in addition to phosphates (Liu, F et al., 2004). O-GlcNAcylation regulates phosphorylation of tau inversely both in cultured cells and in metabolically active rat brain slices. More importantly, O-GlcNAcylation is decreased in AD brain. In a mouse model of decreased brain glucose metabolism induced by fasting, we observed decreased O-GlcNAcylation and concurrently increased tau phosphorylation at multiple phosphorylation sites (Liu, F et al., 2004; Li et al., 2006). On the basis of these observations, we proposed a mechanism by which impaired brain glucose metabolism leads to Alzheimer neurofibrillary degeneration via decrease in O-GlcNAcylation and consequently hyperphosphorylation of tau (Gong et al., 2006).

    According to the above proposed mechanism, upregulation of tau O-GlcNAcylation could be a novel approach to inhibit and reverse hyperphosphorylation of tau and thus to treat AD and other tauopathies. The new O-GlcNAcase inhibitor, thiamet-G, developed in Dr. Vocadlo’s laboratory (Yuzwa et al., 2008) appears to be an excellent agent for upregulation of protein O-GlcNAcylation. Compared to previously developed O-GlcNAcase inhibitors, thiamet-G is very potent (Ki = 21 nM) and highly specific to human O-GlcNAcase. It exhibits 37,000-fold selectivity for human O-GlcNAcase over human lysosomal β-hexosaminidase and does not inhibit other glycoside hydrolases at as high as 500 μM concentration. Furthermore, its competitive inhibition to O-GlcNAcase, extreme stability, and ability to cross the blood-brain barrier makes thiamet-G very attractive for drug development for treating AD. Furthermore, treatment of PC12 cells with thiamet-G induced increased protein O-GlcNAcylation and decreased tau phosphorylation at Thr231, Ser396, and Ser422. Similar results were observed in vivo when rats are treated with thiamet-G either intravenously or orally. This in vivo study with oral administration is especially attractive for drug development.

    To date, more than 40 phosphorylation sites have been identified in tau protein isolated from AD brain (Gong et al., 2005; Hanger et al., 2007). It is generally believed that hyperphosphorylation at multiple sites converts the normal tau into the pathological tau, and that different phosphorylation sites of tau play different roles in this conversion (Iqbal et al., 2005; Wang et al., 2007). The current study investigated only seven phosphorylation sites of tau, among which only three sites were found decreased upon treatment with thiamet-G. It will be important to examine the effects of thiamet-G on tau phosphorylation at other sites as well, especially at those sites known to be involved in its pathological activity, i.e., sequestration of normal tau, MAP1, and MAP2, and its self-assembly into paired helical filaments (Iqbal et al., 1986; Alonso et al., 1994; 1997; 2001b; Iqbal et al., 2005).

    Thiamet-G can cause decreased tau phosphorylation at certain phosphorylation sites directly via elevation of tau O-GlcNAcylation. Thiamet-G might also elevate O-GlcNAcylation level of other neuronal proteins in the brain. Whether elevation of O-GlcNAcylation of these proteins has significant diverse effects remains to be elucidated. These affected proteins could include several protein kinases, especially those of PI-3 kinase pathway, that regulate tau phosphorylation. Glycogen synthase kinase-3β (GSK3β) is one of the most important tau kinases in the brain (Takashima, 2006; Avila and Hernandez, 2007). The activity of GSK3β is mainly regulated by its upstream kinase AKT via phosphorylation at Ser9. It has been reported that AKT and several components of the AKT signaling pathway are also modified by O-GlcNAc and that O-GlcNAcylation of AKT inhibits its kinase activity (Vosseller et al., 2002; Luo et al., 2008; Yang et al., 2008). Thus, it is possible that thiamet-G could also elevate AKT O-GlcNAcylation, which, in turn, leads to inhibition of AKT activity and, consequently, results in activation of GSK3β. It will not be surprising if increased phosphorylation at some of tau phosphorylation sites is induced in the brain by thiamet-G treatment. The ultimate changes of tau phosphorylation with thiamet-G treatment will be the combined consequence of direct effect through increase in tau O-GlcNAcylation and indirect effects through modifying activities of various tau kinases and phosphatases.

    In the end, if the net effect results in converting the abnormally hyperphosphorylated tau to a protein with normal-like biological activity or non-inhibitory molecule, which remains to be determined, it will be a major step forward.

    References:

    . Abnormal phosphorylation of tau and the mechanism of Alzheimer neurofibrillary degeneration: sequestration of microtubule-associated proteins 1 and 2 and the disassembly of microtubules by the abnormal tau. Proc Natl Acad Sci U S A. 1997 Jan 7;94(1):298-303. PubMed.

    . Role of abnormally phosphorylated tau in the breakdown of microtubules in Alzheimer disease. Proc Natl Acad Sci U S A. 1994 Jun 7;91(12):5562-6. PubMed.

    . Interaction of tau isoforms with Alzheimer's disease abnormally hyperphosphorylated tau and in vitro phosphorylation into the disease-like protein. J Biol Chem. 2001 Oct 12;276(41):37967-73. PubMed.

    . Hyperphosphorylation induces self-assembly of tau into tangles of paired helical filaments/straight filaments. Proc Natl Acad Sci U S A. 2001 Jun 5;98(12):6923-8. Epub 2001 May 29 PubMed.

    . GSK-3 inhibitors for Alzheimer's disease. Expert Rev Neurother. 2007 Nov;7(11):1527-33. PubMed.

    . Post-translational modifications of tau protein in Alzheimer's disease. J Neural Transm. 2005 Jun;112(6):813-38. PubMed.

    . Impaired brain glucose metabolism leads to Alzheimer neurofibrillary degeneration through a decrease in tau O-GlcNAcylation. J Alzheimers Dis. 2006 Mar;9(1):1-12. PubMed.

    . Phosphatase activity toward abnormally phosphorylated tau: decrease in Alzheimer disease brain. J Neurochem. 1995 Aug;65(2):732-8. PubMed.

    . Phosphoprotein phosphatase activities in Alzheimer disease brain. J Neurochem. 1993 Sep;61(3):921-7. PubMed.

    . Microtubule-associated protein tau. A component of Alzheimer paired helical filaments. J Biol Chem. 1986 May 5;261(13):6084-9. PubMed.

    . Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A. 1986 Jul;83(13):4913-7. PubMed.

    . Novel phosphorylation sites in tau from Alzheimer brain support a role for casein kinase 1 in disease pathogenesis. J Biol Chem. 2007 Aug 10;282(32):23645-54. PubMed.

    . Tau pathology in Alzheimer disease and other tauopathies. Biochim Biophys Acta. 2005 Jan 3;1739(2-3):198-210. PubMed.

    . Defective brain microtubule assembly in Alzheimer's disease. Lancet. 1986 Aug 23;2(8504):421-6. PubMed.

    . Concurrent alterations of O-GlcNAcylation and phosphorylation of tau in mouse brains during fasting. Eur J Neurosci. 2006 Apr;23(8):2078-86. PubMed.

    . Contributions of protein phosphatases PP1, PP2A, PP2B and PP5 to the regulation of tau phosphorylation. Eur J Neurosci. 2005 Oct;22(8):1942-50. PubMed.

    . O-GlcNAcylation regulates phosphorylation of tau: a mechanism involved in Alzheimer's disease. Proc Natl Acad Sci U S A. 2004 Jul 20;101(29):10804-9. PubMed.

    . Phosphorylated PP2A (tyrosine 307) is associated with Alzheimer neurofibrillary pathology. J Cell Mol Med. 2008 Jan-Feb;12(1):241-57. PubMed.

    . Protein modification by O-linked GlcNAc reduces angiogenesis by inhibiting Akt activity in endothelial cells. Arterioscler Thromb Vasc Biol. 2008 Apr;28(4):651-7. PubMed.

    . Altered expression levels of the protein phosphatase 2A ABalphaC enzyme are associated with Alzheimer disease pathology. J Neuropathol Exp Neurol. 2004 Apr;63(4):287-301. PubMed.

    . GSK-3 is essential in the pathogenesis of Alzheimer's disease. J Alzheimers Dis. 2006;9(3 Suppl):309-17. PubMed.

    . PP2A mRNA expression is quantitatively decreased in Alzheimer's disease hippocampus. Exp Neurol. 2001 Apr;168(2):402-12. PubMed.

    . Elevated nucleocytoplasmic glycosylation by O-GlcNAc results in insulin resistance associated with defects in Akt activation in 3T3-L1 adipocytes. Proc Natl Acad Sci U S A. 2002 Apr 16;99(8):5313-8. PubMed.

    . Kinases and phosphatases and tau sites involved in Alzheimer neurofibrillary degeneration. Eur J Neurosci. 2007 Jan;25(1):59-68. PubMed.

    . Phosphoinositide signalling links O-GlcNAc transferase to insulin resistance. Nature. 2008 Feb 21;451(7181):964-9. PubMed.

    . A potent mechanism-inspired O-GlcNAcase inhibitor that blocks phosphorylation of tau in vivo. Nat Chem Biol. 2008 Aug;4(8):483-90. PubMed.

    . Tau hyperphosphorylation correlates with reduced methylation of protein phosphatase 2A. Neurobiol Dis. 2008 Sep;31(3):386-94. PubMed.

    View all comments by Khalid Iqbal
  4. The manuscript by Schroeter et al. demonstrates that even in middle-aged mice, anti-Aβ immunotherapy can cause increased vascular leakage. Importantly, the Schroeter et al. report indicates that high dose antibody treatment can prevent the formation of new vascular deposits and clear the existing deposits when treatment is continued for six months. Critically, they demonstrate that low doses of antibodies do not appear to increase microhemorrhage, although both the highest and the intermediate antibody doses did reduce/prevent the vascular deposits. Unfortunately, it appears from figure 3 and the results mentioned in the text that only the highest antibody dose succeeded in clearing the parenchymal deposits, presumably the target of anti-amyloid therapy. Prior reports showed that older mice harboring significant parenchymal amyloid deposits developed microhemorrhage when treated with one of several different anti-Aβ antibodies (Pfeifer et al., 2002; Wilcock et al., 2004; Racke et al., 2005; Wilcock et al., 2006). In some, but not all instances this modest amount of vascular leakage was associated with increased vascular amyloid deposits.

    James Nicoll has mentioned that in autopsy cases from the Phase 1 Elan-Wyeth vaccine trial, both increased vascular amyloid and microhemorrhage were found at one to two years after the vaccine treatment. In fact, these observations were used as evidence that active clearance was occurring in those regions (presentation at New York Academy of Sciences, 24 March 2008). However, two patients coming to autopsy four years or more after the treatments had no apparent hemorrhages and both vascular and parenchymal amyloid deposits were cleared.

    A major question regards whether the clearance of pre-existing parenchymal Aβ deposits by immunotherapy will lead to increased vascular deposits and/or microhemorrhage, especially in older individuals where vessels are less compliant. The available data, including those of Shroeter et al., suggest that effective clearance of parenchymal deposits leads initially to localized vascular leakage around vessels, possibly associated with increased vascular amyloid, but ultimately all deposits can be cleared, and the evidence of prior hemorrhage gradually disappears. Shroeter et al. argue that one can titrate the antibody dose to achieve sufficiently slow rates of clearance that the increased vascular leakage does not occur. However, they still do not provide evidence for a dosage that effectively reduces the parenchymal deposits without resulting in microhemorrhage. It would also be of value to have examined shorter survival times to check for an increase in vascular deposits earlier in the therapy.

    Given these data, it is commendable that the Elan-Wyeth bapineuzumab trial is using low doses of antibody with infrequent dosing (every 90 days). While they run the risk of insufficient antibody exposure for a therapeutic effect, they also diminish the likelihood of adverse events associated with development of hemorrhage. I believe everyone hopes immunotherapy can be made safe and successful. The procedures to achieve this goal are already underway.

    References:

    . Cerebral hemorrhage after passive anti-Abeta immunotherapy. Science. 2002 Nov 15;298(5597):1379. PubMed.

    . Exacerbation of cerebral amyloid angiopathy-associated microhemorrhage in amyloid precursor protein transgenic mice by immunotherapy is dependent on antibody recognition of deposited forms of amyloid beta. J Neurosci. 2005 Jan 19;25(3):629-36. PubMed.

    . Passive immunotherapy against Abeta in aged APP-transgenic mice reverses cognitive deficits and depletes parenchymal amyloid deposits in spite of increased vascular amyloid and microhemorrhage. J Neuroinflammation. 2004 Dec 8;1(1):24. PubMed.

    . Deglycosylated anti-amyloid-beta antibodies eliminate cognitive deficits and reduce parenchymal amyloid with minimal vascular consequences in aged amyloid precursor protein transgenic mice. J Neurosci. 2006 May 17;26(20):5340-6. PubMed.

    View all comments by Dave Morgan
  5. I have carefully read this very good paper from David Vocadlo’s group. The data presented are clear and very interesting since the study opens a new strategy for a therapy against Alzheimer disease. To my knowledge it is one of the first times, together with the work of John Chatham’s group in the field of cardioprotection, that some scientists propose to target the O-GlcNAc dynamism with the intention to treat a disease.

    Contrary to complex glycosylations, O-GlcNAc is confined within the cytosol and the nucleus of eukaryotes. It is highly dynamic and it can counteract the effect of phosphorylation by modifying the same sites on the peptide backbone. Two enzymes are responsible for the versatility of O-GlcNAc: the O-GlcNAc transferase, simply named OGT, and the O-GlcNAcase, named OGA. To be aware of the impact of the work presented in the paper by Yuzwa and collaborators, one should know that O-GlcNAc level is tightly dependent upon glucose metabolism since UDP-GlcNAc, the OGT substrate that gives the GlcNAc moiety, comes from the extracellular glucose.

    Alzheimer disease is characterized by two hallmarks—the formation of senile plaques and the aggregation of hyperphosphorylated tau into paired helical filaments. It has been recently proposed by Suzanne de la Monte that Alzheimer disease represents a brain-specific form of diabetes, i.e., type 3 diabetes. Starting from this hypothesis, a link between the nucleocytoplasmic-specific glycosylation from O-GlcNAc and the impairment in glucose metabolism found in Alzheimer patients could be drawn. Tau proteins are extensively modified with O-GlcNAc residues (Arnold et al., 1996) and our group has observed a reciprocal relationship between this glycosylation and phosphorylation for this neuronal protein (Lefebvre et al., 2003). So, it is strongly suspected that the hyperphosphorylation of tau is a consequence of its own hypoglycosylation, which itself ensues from glucose impairment. The idea of Yuzwa’s paper is based on the specific inhibition of OGA, the enzyme that hydrolyses the O-GlcNAc moiety from the target proteins. Numerous OGA inhibitors have been described in the literature, PUGNAc being the most widely used. The major problem of PUGNAc is that it is not sufficiently selective to inhibit the cytosolic and nuclear hexosaminidase OGA (indeed numerous other glycosidases are found in other organelles such as the lysosome, for example) and moreover it cannot reach the blood-brain barrier. But the authors have designed and synthesized a new molecule they named thiamet-G. Thiamet-G is highly stable under aqueous conditions and is selective for nucleocytoplasmic OGA with a very interesting Ki of 21 nM. Experiments were conducted either on PC-12 cells, in which thiamet-G rapidly increases the O-GlcNAc levels, and in rats. The administration of thiamet-G in vivo clearly reduces tau phosphorylation in the CA-1 region of the hippocampus and in the cortex of these animals. Accordingly, thiamet-G reduces the phosphorylation at Thr231, Ser396, and Ser422, two of these sites being key priming sites that control the pathological hyperphosphorylation of tau in Alzheimer disease.

    In conclusion, the paper creates an opening in the treatment of Alzheimer disease by targeting one of the enzymes of the cycling O-GlcNAc. Nevertheless, and as the authors clearly recognize, the experiments conducted with thiamet-G were realized only on healthy, non-pathological animals, and at the moment it is not known if the drug could reverse or counteract the pathological phosphorylation process acting on tau. It would be also interesting to test if the opposite strategy—inhibition of OGT with specific inhibitors—would lead to an increase in tau phosphorylation in vivo. What we also do not know is what effect thiamet-G has on the O-GlcNAc levels in tissues other than brain, since O-GlcNAc is widely expressed in the entire organism and controls many crucial cellular processes such as transcription, trafficking, and protein stability.

    References:

    . The microtubule-associated protein tau is extensively modified with O-linked N-acetylglucosamine. J Biol Chem. 1996 Nov 15;271(46):28741-4. PubMed.

    . Evidence of a balance between phosphorylation and O-GlcNAc glycosylation of Tau proteins--a role in nuclear localization. Biochim Biophys Acta. 2003 Jan 20;1619(2):167-76. PubMed.

    View all comments by Tony Lefebvre

References

News Citations

  1. Mini-strokes from Passive Immunization?
  2. Channel Surfing—Two Studies Strengthen Calcium-AD Connection
  3. Aβ Busters and Other Ploys Show Promise for Treating Neurodegeneration

Paper Citations

  1. . Passive immunotherapy against Abeta in aged APP-transgenic mice reverses cognitive deficits and depletes parenchymal amyloid deposits in spite of increased vascular amyloid and microhemorrhage. J Neuroinflammation. 2004 Dec 8;1(1):24. PubMed.
  2. . Exacerbation of cerebral amyloid angiopathy-associated microhemorrhage in amyloid precursor protein transgenic mice by immunotherapy is dependent on antibody recognition of deposited forms of amyloid beta. J Neurosci. 2005 Jan 19;25(3):629-36. PubMed.
  3. . BDA-410: a novel synthetic calpain inhibitor active against blood stage malaria. Mol Biochem Parasitol. 2007 Sep;155(1):26-32. PubMed.
  4. . O-GlcNAcylation regulates phosphorylation of tau: a mechanism involved in Alzheimer's disease. Proc Natl Acad Sci U S A. 2004 Jul 20;101(29):10804-9. PubMed.
  5. . Phosphoinositide signalling links O-GlcNAc transferase to insulin resistance. Nature. 2008 Feb 21;451(7181):964-9. PubMed.

Further Reading

Papers

  1. . Untangling tau hyperphosphorylation in drug design for neurodegenerative diseases. Nat Rev Drug Discov. 2007 Jun;6(6):464-79. PubMed.

Primary Papers

  1. . A potent mechanism-inspired O-GlcNAcase inhibitor that blocks phosphorylation of tau in vivo. Nat Chem Biol. 2008 Aug;4(8):483-90. PubMed.
  2. . Immunotherapy reduces vascular amyloid-beta in PDAPP mice. J Neurosci. 2008 Jul 2;28(27):6787-93. PubMed.
  3. . Inhibition of calpains improves memory and synaptic transmission in a mouse model of Alzheimer disease. J Clin Invest. 2008 Aug;118(8):2796-807. PubMed.