The "Protein-Only" Hypothesis: The Emperor Without Clothes

The protein-only hypothesis has dominated the field of amyloid pathologies for more than two decades now, particularly after the Nobel Prize that was awarded to Dr. Stanley Prusiner in 1997, which acknowledged prions as "a new biological principle of infection". The history of the theory of “protein-only conformational replication” goes something like this: 

Brain cells are dying, we went looking for a virus, we couldn't find a virus, instead, we found a protein, only! But, we still assumed it behaves like a virus in two main aspects:

1. It kills the cells → origin of the gain-of-toxic-function hypothesis (GOF).

2. It self-reproduces without nucleic acid → origin of the prion/amyloid strain hypothesis.

Despite being around for about 40 years (since initial publication in 1982), the theory of protein-only conformational replication still lacks a mechanistic description of its two main viral-like assumptions:

1. Amyloid fibrils are not very toxic, and because of that, the field has settled on ascribing toxicity to the not-well-defined, non-amyloid species of oligomers.

2. As late as 2019, the mechanism of how a protein alters the conformation of another protein is still unknown, let alone how this conformational information is preserved and transmitted with fidelity across different prion strains (Fig.1).

How can a field survive on the viral analogy for so long without any defined mechanisms that support its fundamental claims?

Compare this to the Watson and Crick paper on preservation and replication of biological information via DNA. In the last paragraph, they famously highlight the importance of their structural discovery by the fact that it establishes a well-defined mechanism for the copying of genetic material based on a four-letter code and hydrogen bond pairings (Fig. 2). Later on, a plethora of elaborate molecular machinery was discovered that mediate the replication of biological information stored in the DNA and ensure its fidelity in what became known as molecular biology. For the protein-only replication paradigm, however, not a mechanism is known nor any machinery we know exists that mediates anything similar to what happens when biological information is actually preserved and transferred via DNA.  

How can we expect to able to tackle amyloid pathologies without having clarity regarding the pathological mechanisms? On what basis can we design our therapeutics? 

But, are there really no mechanisms that accurately describe the amyloid phenomenon? There are. However, they are based on physical chemistry and the thermodynamics of nucleation and phase transformation, which has nothing to do with how viruses operate. Amyloids are a generic solid-state of proteins, a state that literally any protein can reach if pushed hard enough. That is why it is so prevalent in human diseases with more than 35 proteins known to form amyloids in different pathologies.

The transformation of a protein that is usually soluble into the solid-state of amyloid is dependent on the protein concentration, temperature, pH and the presence of catalyzing surfaces such as lipid membranes, nanoparticles or viruses. No biological information is preserved or transferred during this process, as much as no biological information is preserved or transferred during the process of crystallization. The different shapes, or polymorphs, of amyloids are dependent on the factors mentioned above, such as concentration, pH, and catalysts, all of which are factors that are exogenous to the amyloid structure and not controlled or encoded by it. Thus, preservation or faithful "strain-like" transmission of conformational information that is somehow encoded within the amyloid structure is theoretically impossible. And in my opinion, this is why a mechanism of conformational replication has not been found so far, because frankly, it cannot exist.

That is why I think the viral analogy has not served the field well, and it’s maybe time to move on to the mechanisms that we can define and quantify accurately, no matter how complicated they sound. Biology has to be based on physics and chemistry, there are no shortcuts. Amyloids/prions are not viruses, they are also not proteins that behave like viruses, they are proteins that behave like proteins. They can polymerize, solidify, precipitate and lose their function and there are well-defined physicochemical mechanisms describing these processes.   

Within this framework that we describe in our review here, amyloids are treated as what they are, a physical state of proteins, and the factors that can lead to this physical transformation are assessed in relation to different etiologies. Importantly, by highlighting the physical nature of amyloids as a protein phase, it becomes clear that most proteins cannot perform their functions in this state because all the active domains are sequestered within the solid amyloid structure. This loss-of-function (LOF) should be considered a critical toxicity mechanism, especially due to lack of amyloid fibril direct toxicity and in light of data from disease models that demonstrate pathological phenotypes in knockout and knockdown animals. From this perspective, amyloid pathologies can be considered LOF diseases, opening the way for testing replacement therapies instead of the anti-amyloid therapies that have so far failed in the clinic.

The protein-only conformational replication has been naked of any mechanism for too long, and it’s about time to highlight that this cannot continue to be acceptable at the expense of other physical mechanisms that are far more accurate, useful and relevant for both mechanistic understanding and therapeutic development.

Excuse me, but the emperor has no clothes!