Response to: Coskuner and Murray (2014) J Alzheimers Dis 41, 561-574

20 July 2014

I am writing with reference to the recent paper by Coskuner and Murray [1]. I am sorry but I find it very hard to believe that these authors were not aware of our previous research in the field of ATP and amyloid-β (Aβ) [2-4]. In this research, we show unequivocally that ATP, ATP + Mg, and ATP +Al (III) (as well as equivalent preparations using ADP and AMP) influence significantly the propensity for Aβ25-35, Aβ1-40, Aβ1-42, and amylin (IAPP) to form β sheets of amyloid under near physiological conditions. May I suggest that the authors’ claimed novelty of their findings is somewhat diminished in the light of our previous (and novel) research in this field.

Christopher Exley, PhD
Professor in Bioinorganic Chemistry
Honorary Professor, University of the Highlands and Islands

[1] Coskuner O, Murray IV (2014) Adenosine triphosphate (ATP) reduces amyloid-β protein misfolding in vitro. J Alzheimers Dis 41, 561-574.
[2] Exley C, Birchall JD (1996) Biological availability of aluminium in commercial ATP. J Inorg Biochem 63, 241-252.
[3] Exley C (1999) A molecular mechanism of aluminium-induced Alzheimer's disease? J Inorg Biochem 76, 133-140.
[4] Exley C, Korchazhkina O (2001) Promotion of formation of amyloid fibrils by aluminium adenosine triphosphate (AlATP). J Inorg Biochem 84, 215-224.


Letter to the Editor Response

Dear Dr. Exley,

We appreciate your letter. Our data is novel for several reasons:

1. We demonstrated that ATP, especially with the Mg2+ cation, prevents Aβ42 misfolding a pH 7.4. This has not been previously demonstrated.

2. We used a computational modeling combination of docking and energy minimization to identify the ATP binding sites on Aβ – Serine and Tyrosine, and experimentally validated the requirement of Tyr 10 residue. This again has not previously been demonstrated. (We note that you inferred the Ser binding site).

3. To my knowledge, we are the first to report ATP levels in human cerebrospinal fluid, notably in Alzheimer’s disease (AD) and non-AD cases. We were aware of your 1997 publication [1] of ATP interaction with Aβ22–35, and indeed cited it as reference #24 in our publication. We thank you making us aware of your subsequent publications, albeit which have different results than our publication.

There are two important issues to note upon comparing our publications. First, your findings demonstrate that the purity of ATP is crucial. This is especially important for experimental design of future studies of ATP and Aβ. You clearly demonstrate that aluminum contamination of commercial ATP preparations augment Aβ misfolding. Of note: we used ATP of greater than 95% pure (ATP Mg ATP Na, 95 and 99% pure, both from sigma: cat #’s A9137 and A1852 respectively).

Second, differences likely stem from preparation of Aβ and have novel implications. With monomeric Aβ42 as starting product, ATP Mg prevented formation of HMW Aβ species and likely stabilized oligomers. The Aβ42 starting product used in your publication [2] was likely protofibrillar, and ATP Al augmented misfolding. You inferred that this seeding of protofibrils resulted in ThioT detectable mature fibrils (we presume likely via lateral association of protofibrils).

Thus ATP may both prevent and promote Aβ misfolding in vivo. The implication is prevention of initial formation of toxic protofibrils in the first place, and sequestering any protofibrils that do form into more inert plaque fibrils. We thank you for alerting us to this!

Dr. Ian V.J. Murray
Associate Professor
Department of Physiology and Neuroscience
St George’s, Grenada, West Indies

[1] Exely C (1997) ATP-promoted amyloidosis of an amyloid beta peptide. Neuroreport 8, 3411-3414.
[2] Exley C, Korchazhkina O (2001) Promotion of formation of amyloid fibrils by aluminium adenosine triphosphate (AlATP). J Inorg Biochem 84, 215-224.