Biochemical Origins ofAlzheimer’s Disease

With Treatment Techniques

Nikolai Wajda

Senior Comprehensive Paper

The CatholicUniversity of America

Spring 2007

Abstract:

Alzheimer’s disease (AD) is a neurodegenerative disease caused by irregular protein formations in the brainleading to neuronal loss and ultimately affecting thepatient’s cognitive ability and memory. ADaffects nearly 4.5 million Americans, and this number is expected to continue to rise1. The pathological manifestations of AD occur in the neurons and are two-fold;the primary cause is the accumulationamyloid (Aprotein depositions, which aggregate into pathogenic plaques. The second is the accumulation of paired helical filaments that form into neurofibrillary tangles (NFTs). A plaques result from the sequential cleavage of the amyloid precursor protein (APP) by -secretase and secretase. NFTs result from the hyperphosphorylation of tau, a stabilizing component of microtubules. Based on current understanding of the Apathway, two major strategies will be discussed that aim at decreasing the deposition of Aplaques in the brain. In the first approach, non-streroidal anti-inflammatory drugs alter the APP cleavage site by -secretase to produce less amyoidogenic plaques. A second method aims at inhibiting-secretase activity on APP through allosteric inhibition of ATP binding.

Introduction:

The first diagnosis of the disease that would eventually carry his name was done by Alois Alzheimer while treating a woman whoshowed early signs of dementia, which included progressive memory loss, delusions, and hallucinations2. Alzheimer, a German psychiatrist working at a mental institution in Frankfurt, Germany, made the critical breakthrough leading to the discovery of the pathological causes of this disease when he applied Max Bielschowsky’s silver staining technique tobrain slides of the woman after her death. In the slides, Alzheimeridentified two abnormal protein products accumulating in the neurons. While the composition was not known at that time, the components would later be identified as -amyloid plaques and neurofibrillary tangles (NFTs), the hallmark causes of AD. amyloid is an irregular protein product that accumulates into large plaques in extra-cellular spaces and inhibits intercellular communication, ultimately leading to neuronal death. NFTs result from the hyperphosphorylation of tau, which leads to the destabilization of microtubules in neurons. This allows the hyperphosphorylatedtau to accumulate into tangled protein masses that blocks both intracellular communication and transportation of nutrients in the cell.

Drugs are being developed that target thesepathological causes. Many biochemical techniques have been utilized for the development of therapeutics that will inhibit the production of these pathological proteins. In this paper, nucleotide binding site inhibition and cleavage site alteration by nonsteroidal anti-inflammatory drugs (NSAIDs) will be discussed.

Production of Amyloid Plaques by Protease Activity

Amyloid-plaques result from the cleavage of the amyloid precursor protein (APP) by a family of proteinases, termed secretases. The gene encoding APP is located on chromosome 21, which is consistent with the predominance of plaques developing in patients with Down’s Syndrome3. The normal function of APP is not well understood, but it is believed that APP could act as a cell-surface receptor, could participate in cell-cell adhesions, or could promote neuritic growth. APP is a transmembrane protein consisting of 770 Amino acids;the protein has both intra- and extra- cellular segments. Three sites for secretase cleavage occur on APP, secretase cleaves APP at AA 687 in the extracellular region, secretase cleavesAPP at AA 671 in the extracellular region, and -secretasecleavage occurs at various locations in the transmembrane region. cleavage precludes production of the amyloidogenic A proteins, however, sequential cleavage of APP by and secretases produces the amyloidogenic A proteins4.

Amyloid-is a normal protein product, with an unknown function, that is formed by the cleavage of APP by secretases. Two routes exist for the sequential cleavage of APP, the first pathway produces the normally secreted product, p3, while the second results in the pathological amyloid plaques6. The transmembrane region of APPis between amino acids 700-723; the N-terminus, amino acids 1-699, is extracellular while the COOH terminus, amino acids 724-770, is in the cytoplasm of the cell. When APP is cleaved by -secretase afterLys687, the result is the release of a large soluble ectodomain fragment(-APPs) leaving the 83 residue COOH-terminal fragment (APP-CTFor C83)with a transmembrane, extracellular, and an intracellular region. APP-CTF issubsequently cleaved by -secretase at either Val711 or Ala713, which results in p3, a non-pathogenic peptide fragment consisting of either 26 or 24 amino acids; a second protein fragment, AICD, remains embedded in the membrane (Figure 1).

The alternate route for proteolytic cleavage of APP begins with -secretase cleavage after Met671. This cleavage resultsin the secretion of a large soluble ectodomain fragment (-APPs) and the retention of a99-residue COOH-terminal fragment (APP-CTFor C99). APP-CTF is then cleaved by-secretase at either Val711 or Ala713. Thiscleavage is important because if C99 is cut at position 711 it will result in A40 while if APP is cut position 713 it will result in A424(Figure 1). Cleavage also results in AICD.

Figure 1: The cleavage of amyloid precursor protien by  secretases.4

The amyloidogenic nature of A fragments is dependent upon the -secretase cleavage site. In studies of the brains of Down’s syndrome patients, who often develop the trademark plaques young in age, it was found that A42 aggregates much more rapidly than A40. In patients fifty years or younger it was found that only 6.3% of senile plaques were A40, while in older brains, fifty or older, the amount of A40 increased to 42%5. A42has two additional hydrophobic amino acids than A40; the presence of isoleucine and alanine increase the amyloidogenic property of the fragments leading to increased aggregation4.

Tau forms Neurofibrillary Tangles Following Hyperphosphorylation:

Tau proteins are a normal cellular product existingin six different isoforms. In healthy neurons, tau binds to and stabilizes microtubules (MTs) as well as promotes tubulin assembly and polymerization. Tau is most prominent in neurons because MTs form theaxons that allow for intracellular transportation of neurotransmitters and nutrients. The disruption of microtubules leads to the inability to transport nutrients and other vital materials in the cell,and through the accumulation of neurofibrillary tau the ultimate result will be neuronal death.

Tau is phosphorylated in areas thatsurround MT binding repeats; thus increasedtau phosphorylationnegatively affectstau’s ability to bind and stabilize MTs7. Hyperphosphorylation leads toboth gain and loss of tau activity that results in neurotoxicity. Hyperphosphorylated tau destabilizes MT binding and increases tau-tau interactions. Destabilized MT does not transport the necessary nutrients to other parts of the neuron. An example of this is that destabilized MTs cannot transport neurotransmitters from the cell body to the synapse where they can be released. Cutting off a neuron’s ability to communicate with other cells effectively isolates it, causing it to die. The increased tau-tauinteractions of hyperphosphorylated tau causes the formation of paired helical filaments that constitute neurofibrillary tangles (NFT’s). Paired helical filaments are composed of two strands of filament twisted around one another with a periodicity of 80 nm and a width varying from 8 to 20 nm7.

Cyclin dependent kinase 5 (Cdk5) is one of manykinasesactive in the hyperphosphorylation of Tau. A cyclin-like membrane protein, p35,regulates Cdk5’s phosphorylation activity by anchoring it to the plasma membrane. Calpain, a protease, cleaves p35 into two fragments, p10 and p25; P10 anchors the active p25-Cdk5 complex to the plasma membrane. Upon cleavage, the active p25-Cdk5 complex is unregulated and free to phosphorylate throughout the cytoplasm (Figure 2). The p25-Cdk5 complex increases phosphorylation and decreases the binding ability of tau on microtubules8,9. Tau that is phosphorylated byP25-Cdk5has reduced binding to microtubules and causes the collapseof microtubule and formation of NFTs10.

Figure 2: The Cdk5, a kinase responsible for phosphorylation of Tau, is regulated by cofactor p35.8

Risks Factors

The largest risk factor for the development of AD is increasing age. After 65 years of age, the incidence and prevalence of Alzheimer’s disease doubles every 5 years. Genetic studies have been able to isolate inheritable factors contributing to AD, as well as genetic mutations, which increase the risk for developing AD. Other than the fact that autosomally dominant variety of AD is early in onset, the phenotypic expression of familial versus sporadic AD is difficult to distinguish. There are four known genetic factors which are related to the amyloid- pathway, the inheritance of the E4 allele of apolipoprotein, or mutations in either Presenilin-1 (PS1), Presenilin-2(PS2), or APP11,12,13.

Mutations in certain genes influence the development of A plaques

Missense mutations in the APP gene, in the PS1gene, or in the PS2 gene increase the likelihood for developing Alzheimer’s disease. The gene encoding the amyloid precursor protein is located on chromosome 21; missense mutations arising in this gene that have been shown to confer an increased risk to AD. Missense mutations in the APP gene increase the overall amount of Athrough the alteration of cleavage sites for the processing of proteases (Figure 3). Missense mutations in APP, however, are considered very rare and are found in only about two dozen families. It is believed that these mutations effect proteolytic activity by the secretases because the mutations are all located either directly before or directly after the secretase cleavage sites4.

Figure 3: Missense mutations occurring in APP which lead to the production of A4

Double Mutations occurring before the -secretase site (K670N and L671M) induce -secretase cleavage, and allows for increased amounts of the Aprecursor. Five mutations occur on the COOH-terminal side of the -secretase cleavage site(I714T, V715M, I716V, and I,G,F,L717V); these mutations increase the productionof A, which is the Aform that most readily forms into plaques. Three APP mutations occurinside the  and  cleavage sites (A692G, Q693E, and N,K,G694D); these mutations increase the aggregation properties of all Aresidues

Missense mutations in the presenilin genes also increase the risk for developing AD. So far, there are 75 known missense mutations in the PS-1, which is located on chromosome 14,and only three to PS-2, located on chromosome 1, which are known to cause AD. Mutations in the PS-1 gene cause the most common form of dominant early onset familial AD. PS-1 mutations selectively increase precursor C99 cleavage by-secretase to produce moreA42.; A42 is a more amyloidogenic peptide compared with A4014. In its complete and functional form, -secretase consists of PS-1 or PS-2, nicastrin (nct), Aph-1 and PEN-2. PS-1 is found to playa vital role in the enzymatic cleavage of APP by -secretase. PS-1 is made up of 9 transmembrane domains, andis activatedas the result of endoproteolysis. Endoproteolysis occurs in domain 7 of presenilin, and creates the N-terminal fragment (NTF) and the C-terminal fragment (CTF). The full length PS-1 protein is short lived, wheras the processed protein has a much longer half life, suggesting that cleavage is necessary for function15.

Determining the structure of -secretase

The necessity of all four of these proteins for -secretase activity was proven following the expression of the genes encoding for these proteins in Saccharomyces cerevisiaeby Steiner et al16. S. cerevisiae does not contain -secretase, so it is possible to address the questionof which components of the complex are necessary for activity by selectively introducing the four components. An APP based protein that contains a sequence analogous to the-secretase cleavage site was synthesized in order to report -secretase activity. This protein is bound to GAL4, a transcription factor for β-galactosidase (β-gal). Cleavage of C1–55–GAL4 after co-expression of functional γ-secretase is expected to liberate GAL4 from the cytosolic side of the membrane. GAL4 then translocates to the nucleus where it activates transcription of the Escherichia coli LacZ gene, which encodes β-gal14. β-gal breaks down the sugar lactose, producing a blue color that can be measured and quantified. In this experiment, the presence and intensity of the blue coloring is a test for-secretasecleavage of C1–55–GAL4 (Figure: 4).

Figure 4: B-Gal activity to determine components of functional -secretase.14

GAL4: Expression of GAL4 protein as a control shows high B-Gal activity

PS1wt: Expression of the PS-1 without Nct, APH-1, or Pen 2 showed no b-gal activity

PS1wt: Expression of the PS-1 with Nct, APH-1 and Pen 2 showed high b-gal activity

PS1d3854A: Expression of a functionally inactive PS-1 with Nct, APH-1, and Pen 2 showed almost no activity

No NCT: Expression of PS-1, APH-1, and Pen 2 without NCT showed almost no activity

No APH-1: Expression of PS-1, Nct, and Pen 2 without APH-1 showed almost no activity

No PEN-2: Expression of PS-1, Nct, and APH-1 without PEN-2 showed almost no activity

No PS1: Expression of Nct, APH-1, and PEN2 without PS-1 showed no activity.

The only complex that showed significant cleavage of the APP-like substrate leading to -galactoside activity was the complete complex containing PS-1, nicastrin (nct), Aph-1 and PEN-2. When the complex lacked even one of these components, activity was significantly decreased16. In further experiments with mice, the results showed that deleting PS-1 leads to the reduction of -secretase activity16. The vital discovery that PS-1 was required for producing A led researchers to question how PS-1cleavesAPP.

Wolfe et al. determined that two aspartate residues on PS1 were required for the endoproteolytic and γ-secretase function. Introducing mutations to two intermembrane aspartates in PS-1,D275A and D385Aor D275E, reduced the production of both A40 and A42 while increasing theamounts of Aprecursor substrates C83 and C99. The Asp to Ala double mutations led to substantially less total A (mean 57 +/-3%) and A42(53 +/- 4%) than cell lines expressing wild-type PS-117. D275E mutations was done to see if removing the negative charge of the Asp had a detrimental effect on protein folding that could have resulted in decreased cleavage. However, Glutamate, with its negative charge still resulted in decreased A production 17. An increase in concentrations of both APP-CTFand-APPs shows thatboth the and  secretases are still active, while thedecreased concentrations of A means that-secretase is unable to process these fragments, contributing evidence that the aspartates are required for proteolytic activity. The double mutations of the aspartates also halted endoproteolytic activity of presenilin17. Levels of Aproduced from cells with aspartate double mutations were found to be similar to levels where PS1 was deleted17. Furthermore, the activity of the enzyme requires a slightly acidic pH, which contributesmore evidence to this theory of PS-1 as an aspartyl protease18.

CurrentTreatment:

Current therapeutics are insufficient for the treatment of AD. The most popular drugs used to treat AD are acetylcholinesterase (AChE) inhibitors. AChE inhibitors block the enzyme responsible for the degradation of the neurotransmitter acetycholine in the synapse. As a result of neuronal loss associated with AD, patients suffer from decreased levels of this neurotransmitter, thus inhibiting the activity of AChE, more acetylcholine will be present at the synapse. The second method used for treating AD is memantine, which is an uncompetitive NMDA receptor antagonist. Memantine regulates glutamate, a messenger chemical required for information processing, storage and retrieval. Glutamate binds to NMDA receptors, and allows calcium to enter nerve cells in controlled amounts. Calcium influx is required for information storage, however, too much glutamate overstimulates NMDA receptors and allows too much calcium into cells, disrupting cells and causing death. Memantine protects against excess glutamate activity by partially blocking NMDA receptors. The current treatments for AD only serve to ameliorate the side-effects without curing the disease, therefore researchers seeknovel methods for curing AD1.

-Secretase Inhibitors

Researchtargeting theenzyme-secretaseas a potential target for combating AD has shown some promise as well as some setbacks. Inhibiting -secretase cleavage of APP would prevent the second step in the sequential cleavage of the Aandthus decrease theamount of amyloid plaque developing in the brain. Research has focused on two approaches for altering -secretase activity. The first method is by inhibiting the creation of A of all lengths, while another is to increase the formation of shorter, less pathogenic Afragments. Several inhibitors have been found to be successful in halting A production; developers, however, hit a roadblock when complications arose due to the fact that-secretase is also being responsible for cleaving other membrane proteins vital to cellular survival. LY-450139, the first -secretase inhibitor to go through clinical trials was found to decrease plasma levels of Ahowever; it was unable to lowercerebrospinal fluid levels19. While this drug was found to bepartially effective and well tolerated,LY-450139lackedthe desired specificityfor the cleavage of APP and dosages had to be kept low because of the detrimental side effects that it would have have. -secretase is active in the cleavage of several substrates other than APP, including Notch, a transmembrane cellular protein active in cell signaling. Notch first requires a signal tobind to its extracellular domain that will causecleavage and the release of this domain. Notch then undergoes transmembrane cleavage by -secretase, which releases an intracellular fragment termed, Notch Intracellular Domain, NICD, which enters the nucleus and causes transcriptional changes in the cell20-21. In vivo studies ofa-secretase inhibitor,LY-411,575, by Wong et al., showed decreased production of A, but also showed deleterious effects in mice. Large impairmentsof the spleen, thymus, and intestine were observed. The most pronounced side effects of the -secretase inhibitor were caused by the inhibition of Notch cleavage, ultimately leading togastrointestinal toxicity and interference with the maturation of B- and T-lymphocytes21-22. From these initial studies into the inhibition of secretase, a new direction must be taken for drugs aimed at this enzyme. The new direction must be one that can inhibit the production of amyloid plaques without affecting the cleavage of other -secretase targets. Two approaches will be discussed below, one will be the use of NSAID’s which aim to alter the cleavage site of -secretase and thus produce shorter, less amyoidal plaques while the second will seek to inhibit -secretase activity on APP through allosteric inhibition of ATP binding.

Nonsteroidal Anti-Inflammatory Drugs

The formation of less amyloidogenic plaques is an approach taken by researchers. Instead of inhibiting the secretion of all amyloid plaques, certain drugs will aim at decreasing the percentage of the longer, and more amyloidogenic plaques.