H. Robert Guy

The mission of Amyloid Research Consultants is to facilitate research that targets root causes of diseases such as Alzheimer’s disease, Parkinson’s disease, and Type 2 diabetes. The ideal treatment would detect these diseases at the early stages before irreversible damage has occurred; e.g., for Alzheimer’s at or before the mild cognitive impairment stage. Relatively small amyloid assemblies called oligomers develop first and are more toxic than much larger fibril structures associated with plaques, tau tangles, and Lewy bodies. The most toxic of these oligomers interact with membranes to form transmembrane channels. Evidence is rapidly increasing that permeation and/or disruption of synaptic and mitochondrial membranes by amyloids are the root causes of these diseases. Thus, these smaller assemblies should be prime targets for developing and improving treatments.

Nonetheless most treatment development efforts have focused either on fibrils or on suppressing formation of amyloid proteins and peptides. Almost all have failed. The first approach may have failed because it targets the wrong assemblies when it is too late to prevent irreversible damage. The second approach has problems because some forms of these proteins are functionally important and eliminating them may cause serious side effects.

Why have oligomers and channels received so little attention from the biomedical community? Two only partially true dogmas have impeded this research. The first is that fibrils are the major culprits. The second is that smaller assemblies are intrinsically disordered and thus are not good targets for drugs. This is true for some oligomers in some environments, but there are exceptions; e.g., the most toxic form of amyloid beta (Aβ) forms channels with highly-ordered β-barrel structures. The other major impediments to targeting oligomers and channels are our ignorance about their three-dimensional structures and polymorphism; experimental studies indicate that numerous assemblies form.  Structure-based drug design is most likely to succeed when: one knows which toxic structures to target and which functionally important structures to avoid; one knows their high resolution three-dimensional structures, and one knows when to target them.

We strive to fill some of this void. Our approach is to develop three-dimensional models of amyloid oligomers in the aqueous phase, when they form protofibrils, when they interact with lipids, and when they form transmembrane channels. We attempt to model not only structures but the processes by which the structures develop beginning with oligomers in the aqueous phase and ending with transmembrane channels. While at the NIH, our group began collaborating with Harvey Pollard and Nelson Arispe in the early 90’s when they were the first to discover that Aβ forms membrane channels. We published concentric β barrel models of Aβ oligomers, annular protofibrils, and channels in 2010. Recently we extended this concept to α-Syn and amylin channels, are revising and updating our Aβ models, and are developing models of Aβ interactions with α-synuclein, a peptide from the PRP prion, amylin, humanin, and cystatin C. Some of these interactions are protective and others are toxic. Understanding molecular mechanisms of both protection and toxicity could be valuable for drug development.

We are the most experienced group modeling membrane channel structures. We began modeling three-dimensional structures of membrane channels in the mid 70’s beginning with β barrel models of bacterial porin, then moved on to transmitter receptors, voltage-gated K+, Na+, and Ca2+ and inward rectifying channels, K+ transporters, VDAC, mechanosensitive channels (MscL and Trek), antimicrobial peptides (magainin and cecropins), toxic peptides (melittin), and even shark repellent (pardaxin). Many of our predictions were subsequently verified experimentally. Our papers have been cited over 6500 times.

Our models are consistent with numerous experimental findings and constrained by well-established modelling criteria and theories. Nonetheless, they are hypothetical because current data are insufficiently precise to solve the structures directly. Hypothesis-based research is often effective. Our models are designed to be tested rather than believed.They can be tested in a multitude of ways and suggest what experimental approaches could be used to solve these structures. Although we are a theoretical group, we still prefer experimentally determined structures. The major impediment to our work is the quality and quantity of structural data. The experimental work on oligomer and channel structures should be funded much better.

My mother had Alzheimer’s, so this is personal. I will consult or collaborate with any legitimate research group or company and provide coordinates of our atomic scale models free of charge if inordinate expense, time, and legal commitments are not incurred. Otherwise, fees can be negotiated.