Alzheimers Disease And Cyclophilin D Treatment Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


The increasing prevalence of neurodegenerative diseases (ND) has attracted more research into their pathophysiology. Alzheimer's disease (AD) affects around 2% of the population and has a predicted incidence increase of three folds in the next 50 years (Mattson, 2004). The understanding of the molecular basis of AD started after the sequencing of Aβ which allowed for localising and cloning the gene encoding APP (Mattson, 2004). Other familial causes of AD were found to be presenilin-1 (PS1) and its homolog (PS2) (Blennow et al, 2006). However, APP, PS1 and PS2 are all linked to the familial form of the disease which constitutes a small prevalence compared to the sporadic form (Blennow et al, 2006). The sporadic form of AD appears to be linked to the differential expression of one of the 3 apolipoprotein E with ε4 being the one linked to increased AD prevalence (Blennow et al, 2006).

Much of the literature relies on animal models of familial AD possibly due to the ability to predict the increased production of Aβ, the hallmark of the disease. The increase of Aβ production results from defective processing of APP as shown in figure 1.

Figure. 1 taken from Mattson (2004) Showing the pathways leading to the production of Aβ40/42. Normally β-secretase and γ-secretase cleave APP producing Aβ40 and sAPPβ (Mattson, 2004). APP mutations favour its cleavage by β and γ-secretases to form Aβ42 (Mattson, 2004). Mutations to PS1 and PS2 alter the activity of γ-secretase leading also to increased production of the pathogenic Aβ42.

Amyloid β

The increase in Aβ40 and Aβ42 as a result of APP and/or PS1&2 mutations coupled with the impaired clearance appears to be at the heart of the problem in AD (Butterfield and Bush, 2004). The difference between Aβ40 and Aβ42 is 2 amino acid residues (IA) yet it makes the longer product more structured due to the lower flexibility of the C-terminus inferred by the hairpin loops it has between residues 31-34 and 38-41 (Sgourakis et al, 2007). Sgourakis et al (2007), after utilising molecular dynamics and replica exchange molecular dynamics, suggested that the improved structure and stability of Aβ42 makes it more prone to aggregating and forming amyloids.

Aβ production and aggregation leads to increased oxidative stress and neurotoxicity (Butterfield, 2002; Mattson, 2004; Blennow et al, 2006; Lin and Beal, 2006) as figure 2 explains.

The oxidative stress and neurotoxicity induced by Aβ42 was reported to be due to the S atom in Met35 residue of Aβ42 as it was found to be critical for protein and lipid oxidation, free radical formation and ultimately neurotoxicity (Butterfield and Bush, 2004). However, the perturbed Ca2+ homeostasis, energy failure and morphological pathologies of mitochondria in AD and other NDs suggested mitochondria as a critical point in NDs.

Figure. 2 taken from Mattson (2004)showing Aβ induced oxidative stress. Aβ oligomers interaction with ions generate H2O2 and (OH*) radicals (Mattson, 2004). Moreover, at the membrane Aβ aggregates leading to the production of 4HNE that covalently modifies proteins (Mattson, 2004). A diverse array of proteins is affected (Mattson, 2004). Aβ acting from within the mitochondria can also increase ROS, superoxides, cytochrome c release and decrease ATP production with dire consequences on respiration efficiency (Petrozzi et al, 2007). Amyloid-β, mitochondria and oxidative stress

Aβ is thought to be imported into the mitochondria through the TOM import machinery (Petersen et al, 2008) and was linked directly to mitochondrial toxicity through interacting with amyloid binding alcohol dehydrogenase (ABAD). This leads to memory and learning impairments, possibly due to the increase in reactive oxygen species (ROS) and cytochrome c release and consequent oxidative stress and DNA fragmentation which may lead to cell death (Lustbader et al, 2004).

Aβ was found to act through many pathways causing mitochondrial dysfunction like activity of SOD and inhibition of α-ketoglutarate dehydrogenase (Lin and Beal, 2006). However, recently the role of the mitochondrial permeability transition pore (mPTP) regulation has been explored.

The mPTP, Cyclophilin-D, ANT and VDAC

The mPTP is multi-component channel that links the mitochondrial matrix to the cytosol. It is a non-specific pore that allows solutes into and out of the mitochondria (Leung and Halestrap, 2008). During periods of increased intramitochondrial calcium overloading and oxidative stress the mPTP opens to efflux Ca2+ (DU and Yan, 2009). However, the mPTP plays a role in apoptosis and necrosis.

Necrosis occurs due to the severe decrease in ATP due to the loss in membrane potential leading to ATPase function reversal hence, no ATP produced and ATP present is broken down leading to activation of phospholipases, proteseases and endonucleases causing necrosis (Leung and Halestrap, 2008). Apoptosis occurs due to the swelling that causes the outer membrane to rupture allowing cytochrome c and pro-apoptotic factors to be released into the cytosol (Leung and Halestrap, 2008).

Figure 3. (A) showing the components of the MPTP. There is a number of transmembrane proteins as well as the pro-apoptotic proteins of the Bcl2 family. The green labels facilitate opening, the red inhibit opening (Zamzami and Kroemer, 2001). (B) shows a schematic diagram of the opening of the pore that suggests a conformational change in the VDAC to open the pore induced by CypD binding and high Ca2+ detected by the ANT (Abo-Sleiman et al, 2006) .Fig. (A) is from Zamzami and Kroemer (2001) and fig (B) from Abo-sleiman et al (2006)Some of the mPTP components play a regulatory role and some are structural as can be seen in fig3 (a and b).

The role of CypD in mPTP opening has been shown in a study that used Ppif (gene encoding CypD) knockout (KO) mice and showed that the lack of CypD served as a protective mechanism against mitochondrial swelling and cell death where as overexpressing CypD caused higher levels of swelling, rupture, oxidative damage and cell death in the hearts, livers and brains of mice tested (Baines et al, 2005). Moreover, inhibiting CypD using Cyclosporin A (CSA) showed similar protective effects (Halestrap et al, 1997. As cited by Du and Yan, 2009). The studies into the other components of the mPTP have shown conflicting results. VDAC's involvement in pore opening was challenged by a study on VDAC KO mice by Krauskopf et al (2006) as it showed no decrease in mPTP formation. Moreover, Baines et al (2007) showed that VDAC is actually dispensable for MPTP formation. A study on ANT gene ablation mice by Kokoszka et al (2004) showed that ANT deficiency has little effect on preventing mPTP opening. Leung and Halestrap (2008) showed that in the inner mitochondrial membrane (IMM), the mitochondrial phosphate carrier (PiC) plays the pivotal role in forming the pore not ANT and they suggest that ANT maybe a regulatory component. Overall, the consensus appears to be that CypD plays an important role in pore opening.

Cyclophilin D structure and function, interactions with Aβ and ABAD

CypD belongs to a ubiquitous family of other cyclophilins which include A,B, C, D and E (Schlatter et al, 2005). Cyclophilins catalyse peptidyl-prolyl cis-trans isomerisation (PPIase) of their ligands and play a role in protein folding (Kajitani et al, 2007) and that can be associated to the molecular role of CypD in changing the confirmation(s) of the mPTP components to open the pore (Leung and Halestrap, 2008).

To utilise CypD as a drug target, detailed appreciation of structure-function relationship is needed to potentially engineer inhibitors that stop the uncontrollable pore opening that leads to loss of matrix volume (or increase) (Schlatter et al, 2005). Schlatter et al (2005) carried out a mutagenesis study followed by crystallisation. CypD was mutated at K133 to I133 (Schlatter et al, 2005) as it was demonstrated that changing charged residues can improve crystallisation. Only the truncated tCypD/K133I was crystallised and its structure determined at 1.7 A. The structure of CypD was found to compare to that of CypA crystallised before (Schlatter et al, 2005). The crystal structure showed that the crystal contact point could only be achieved by K133I mutation and cannot be achieved in wild-type protein. Fig 4 shows the crystal structure achieved

Fig 4. Taken from Schlatter et al (2005) showing the crystal structure of tCypD/K133I. The image also shows a putative active site and all the mutations that were performed during the study.

Another study by Kajitani et al (2007) attempted crystallising tCypD/K133I bound to CSA, (Schlatter et al,2005; Kajitani et al, 2007). This study refined the structure at 0.96 A allowing for a better understanding of CSA binding. The overall structure of the complex can be seen in fig 5 and shows that CypD has 8 β-sheets and 2 α-helices and one 310 helix (Kajitani et al, 2007) CSA can be seen to bind with half of the molecule facing the outside and the other half buried in CypD

Fig 5 taken from Kajitani et al (2007)

(A) showing the overall structure of the complex in a ribbon and stick model. (B) showing CSA binding to CypD. (C) showing a diagram of the structure of CSA and where it binds on CypD and (D) showing the distribution of conserved residues in all cyclophilins. Residues conserved in 3 of 4 are orange in 2 of 4 yellow, in 1 of 4 green and unconserved blue.

Du et al studied CypD in AD in a series of studies. CypD, normally in the mitochondrial matrix, associates with the IMM to induce the opening of the MPTP (Du et al, 2008).

Figure 6. Taken from Du et al (2009)

Shows that the increase in CypD in the temporal lope and the hippocampus of AD brains is significantly higher than that of non AD brains.

Shows that the increase in CypD in TgmAPP mice is significantly higher than that of nonTg mice for mice older than 3 months. The increase in CypD appears to be also a factor of age. The expression of CypD in the Aβ rich cortical areas was found to be elevated in AD brains and mAPP transgenic mice (TgmAPP) as seen in figure 6 and appears to increase with age of Tg mice (Du et al, 2009). Their research took the findings further and inspected the possible interaction between Aβ and CypD and confirmed by Surface Plasmon Resonance that oligomeric Aβ40 and Aβ42 bind to CypD in a specific manner as the reversed Aβ peptide showed no binding to CypD. The complex was studied further using immunoprecipitation and immunobloting studies and showed the presence of the complex in AD and TgmAPP mice brains but not in non-AD nonTg mice. The colocalisation of the two proteins was confirmed by confocal and immunogold electron microscopy (Du et al, 2008). This array of in vitro and in vivo studies confirm the interaction between Aβ42 and CypD to form a complex. Du et al (2008) successfully showed that CypD deficiency protects against mitochondrial stress and the results are summarised in figure 7. The results show increased Ca2+ buffering capacity and higher resistance to high Ca2+ concentrations. These results suggest that inhibiting or removing CypD from mitochondria will protect it against stress. The results of MitoSox Red staining, indicating superoxide production in mitochondria, suggested a decreased generation of ROS and hence, greater protection against oxidative stress in the absence of or by blocking CypD. The effect of CypD-Aβ complex on mitochondrial function was also investigated by Du et al (2008) and the results show a significant reduction in respiratory control rate in mAPP animals at 12 months of age suggesting mitochondrial dysfunction. This reduction was absent in the absence of CypD. Moreover, COX IV activity was reduced in the mAPP condition but much of the activity was restored in the absence of CypD. Similar results were found for ATP production in 12 months old animals.

Figure 7 Taken from Du et al (2008)

(A) and (B) show Ca2+ buffering capacity change. a shows a reduced Ca2+ capacity as a function of time in both conditions, however, the rate of reduction is significantly faster in mAPP animals. (B) and (C) show that mAPP animals are significantly less efficient at buffering Ca2+, however, in the absence of CypD (Ppif-/-) and in the case of CypD inhibition by CSA, the Ca2+ buffering capacity is restored and exceeds that of wild type animals.

Du et al (2008) also showed an increased translocation of CypD to the inner membrane in the presence of Aβ42. Furthermore, Aβ was found in the IMM in the mAPP condition but not in the nonTg animals possibly suggesting the presence of an interaction between Aβ and CypD that facilitates the latter's translocation.

The increased neuronal cell death found by Du et al in 2008 and 2009 was probably due to the interaction between Aβ and CypD. The interaction appears to increase mPTP formation, decrease the ability to cope with stress, Ca2+ buffering capacity and cause increased swelling (Du et al, 2009). The effects were attenuated in the absence of CypD which was seen as a reduced number of TUNEL positive cells (Du et al, 2009).

Putting all the above into context, behavioural and electrophysiological studies were conducted by Du et al in 2008 and 2009 and showed improved, learning, memory, synaptic function and normal LTP production in mAPP-Ppif-/-, CSA treated mAPP and aged AD model mice suggesting that limiting CypD-Aβ complex can attenuate neuronal cell death and improve learning and memory.

Unpublished data collected by Yan and Stern in 2004 (as cited by Ren, 2008) and confirmed by Ren (2008, PhD thesis) showed ABAD and CypD colocalisation in mitochondria and suggested that CypD-ABAD also can form a complex that prevents the translocation of CypD from the matrix into the IMM. However, Ren (2008) suggested that in an Aβ rich environment the CypD-Aβ complex formation will be more prevalent and may displace that of CypD-ABAD and hence cause the ultimate cell death. It could be due to the creation of ABAD-Aβ complexes limiting the amount of ABAD available for CypD-ABAD complex formation.

CypD in other diseases

CypD effects on mPTP formation and the consequent cell death were also found to be implicated in other diseases. In Parkinson's disease (PD), inhibition of proteasome of PC12 cells led to a loss in mitochondrial membrane potential and increased mitochondrial apoptosis, effects that were blocked by CSA suggesting an involvement of CypD-dependent mPTP fromation (Abou-Sleiman et al, 2006). Moreover, CypD deficient mice were found to be more resistant to ischemia and anoxia/reoxygention cells injury suggesting a role for the CypD dependant mPTP cell death in ischemic cells damage in non-neuronal cells as well (Nakagawa et al, 2005).

CypD as a therapeutic target

Given the effects of CypD dependant mPTP opening on cells, it's logical to consider it as a therapeutic target. A useful drug may interact with the CypD-Aβ complex, Aβ or CypD to stop the formation of the MPTP and a known molecule that can stop the formation of mPTP is CSA (Muirhead et al, 2010). However, CSA is an immunosuppressant, large and poorly soluble in water (Muirhead et al, 2010) therefore, using it as a drug maybe inappropriate. With the availability of a crystal structure of CypD it may be possible to engineer drugs that would inhibit its effects and attenuate cell death.


The role played by CypD in mitochondrial induced cell death in NDs and specifically AD, and the increasing understanding of its structure and functions may help to position it as a viable potential drug target. However, and due to the limitations of the known inhibitor (CSA), rational drug design is need and requires a detailed understanding of the structure-function relationship. The crystal structure available is based on a truncated mutant form that gives an idea of the interactions the molecule is capable of but there is no guarantee that the wild-type will behave in exactly the same way. Thus, despite the inability to crystallise the wild-type CypD due to hidden crystal contact points (Schlatter et al, 2005) it is useful to conduct analysis studies on the wild-type as well as the crystallisable protein. The research project undertaken will attempt identifying compounds that bind to and manipulate CypD function in its native and truncated mutant form. Fluorescent thermal shift analysis using a library of compounds and CypD and tCypD/K133I whilst bound to Aβ will be used to generate 'hits'. The compounds producing best 'hits' can then be manipulated to improve their pharmacokinetics.