Why LOF was lost?

Why LOF was lost?

When a protein aggregates, two things happen, one is certain and the other is uncertain.

- The certain consequence is that the protein loses its function, because any protein needs its native conformation and solubility to function → loss-of-function (LOF)

- The uncertain consequence is that the resulting aggregates become particularly more toxic → gain-of-function (GOF)

While the two mirror-image possibilities are scientifically valid, the majority of the field studying amyloids, especially within neurodegenerative diseases, chose to focus on the uncertain consequence (GOF) and almost completely ignore the certain one (LOF). A friend of mine asked me why? Why something as obvious as LOF became almost unthinkable and is only mentioned shyly as a heretic belief on the outskirts of the literature?

Here, I will try to discuss the historical factors that I think contributed to this huge asymmetry in amyloid science, and how this unilateral focus is probably the main reason behind the continuous failure to find therapeutics for these diseases for so many years.

GOF Roots

Not long after the discovery that the familial forms of Alzheimer’s disease (AD) are linked to mutations in the APP gene; the amyloid cascade hypothesis (ACH) was born (Hardy, 2017). The APP gene encodes the Aβ peptide, which forms fibrillar amyloid aggregates in AD patients. Until this day, the ACH is the most accepted framework for understanding amyloid pathogenesis. It goes something like this:

- Amyloids of Aβ form first →This causes other amyloids to form (tau fibrils) → Both amyloids are cytotoxic and lethal → Neurodegeneration

A straight-forward GOF hypothesis that formed the basis of understanding AD pathogenesis for decades, and was extrapolated to  nearly all other neurodegenerative diseases such as Parkinson’s disease, ALS, Huntington’s disease, and many others.

Structural Contradiction

The ACH is supposed to be about the toxicity of amyloids, which are a well-defined structural species of proteins characterized by a cross-β conformation that gives it its characteristic fibrillar morphology. Also, amyloids constitute the plaques that are the pathological hallmarks of the disease. Thus, one would expect a GOF amyloid theory such as the ACH to link toxicity to the structure in its title; amyloid. However, the benign nontoxic nature of the amyloid fibers is almost as accepted as the ACH itself (Haass and Selkoe, 2007; Ke et al., 2020; Prusiner, 2013; Selkoe and Hardy, 2016). It is even widely accepted that the amyloid fibrils might be protective, a defense mechanism to sequester the real toxic species, the so-called oligomers (see below). 

Let’s just pause here for a moment, a theory that is based on the toxicity of a particular structure that is the hallmark of the disease is simultaneously OK with the fact that this particular structure is nontoxic! How can a theory survive such a blatant contradiction?

Oligomers to the Rescue

Multiple lines of evidence have clearly demonstrated the non-toxic nature of amyloid fibrils. These include the presence of plaques in the brains of people who died without having any symptoms of disease, the well-known poor correlation between the amyloid burden and disease onset or severity and therapeutic interventions that successfully cleared plaques in humans did not result in improvement in clinical symptoms (Ke et al., 2020). Based on these well-accepted facts, one would expect that the attitude towards GOF theories would change favoring the more likely LOF mechanisms in this case. However, this never happened, GOF remains the dominant framework for understanding amyloid diseases and devising therapeutics. The reason it is still very much alive is the magical word: oligomers.

Unlike amyloid, which denotes a defined structure, the word oligomer literally just means a group of molecules, with no particular structure and no defined mechanism of toxicity so far. Additionally, they are extremely transient in nature (Dear et al., 2020). In that sense, oligomers are as biochemically vague as anything can get. However, despite initial acknowledgment that not finding a well-defined toxic species is a major weakness for the ACH, more than 25 years on, referring to the enigmatic oligomers continues to save it from completely collapsing. 

In figure 1, I tracked the description of the most urgent challenge in the field of amyloid pathologies over a period of 18 years by the same influential authors in very influential papers. In a 2002 paper that is cited more than 4500 times, the authors admit that the ACH is controversial “in part because a specific neurotoxic species of Aβ and the nature of its effects on synaptic function have not been defined in vivo (Walsh et al., 2002)” (Fig 1A).  However, in 2016, 14 years after the influential 2002 paper and 25 years after the introduction of the ACH, the authors highlight these questions: “What are the toxic species of Aβ and tau? What is the connection between Aβ and tangle pathology? (Selkoe and Hardy, 2016)” among the major pending issues for the ACH (Fig. 1B). In 2020, after the continued and consistent failure of the therapeutics that are based on ACH and its presumptive GOF implications, the authors continue to defend the ACH while emphasizing that “the biggest challenge is to better define and understand the toxic forms of Aβ and tau which drive cellular dysfunction in the disease (Walsh and Selkoe, 2020)” (Fig. 1C).

 

Figure 1.

In spite of the time that passed, the evidence that accumulated, and the consistent therapeutic failure, the mysterious nature of oligomers keeps saving the ACH and its GOF assumptions from dying. The real problem, however, is that the mystery about oligomers is engrained in the oligomer concept itself. Its vague and slippery nature can escape any verification or falsification and accommodate any explanation. There can always be a particular set of oligomers that we have not yet discovered that are the real culprit. Stuff that are difficult to prove they exist, are also difficult to prove they don’t. We invented a concept that is almost impossible to define or track and we can only blame ourselves for that, and not biology.

Genetics ≠ GOF

But the question still begs the answer, why? Why a concept so vague and slippery such oligomers needed to be invented to explain away the lack of amyloid toxicity but maintain GOF? 

Mainly, two reasons.

First and most important is the false equivalence between genetic evidence and GOF. The discovery that the Aβ peptide is encoded by the APP gene, which is located on chromosome 21 that is triplicated in Down syndrome, is often cited as the ultimate ACH proof (Selkoe and Hardy, 2016; Walsh and Selkoe, 2020). This is because this gene triplication leads to amyloid plaque accumulation in Down syndrome sufferers and increases the risk of dementia. However, while indisputable genetic evidence ties the disease to a particular protein, it does not indicate how this protein is causing the disease. It demonstrates the cause, not the mechanism. For the mechanism, more than genetics is needed, especially when dealing with a problem that is mainly biophysical in nature. Assuming that overexpression has to always lead to GOF is the real reason why the ACH is so immortal, and that is why when GOF couldn’t be found in the characteristic fibrillar plaques, it had to be found somewhere else; the oligomers.

However, since amyloid aggregation is a biophysical phenomenon that is governed by the laws of thermodynamics and not only genetics, overexpression leading to GOF is not always the case. As a nucleation-dependent phenomenon, overexpression lowers the nucleation barrier, kick-starting a phase transformation event that results in the formation of very stable (not very toxic) solid fibers, while depleting all the soluble, natively-folded protein subunits in the process (more details in our recent review (Malmberg et al., 2020). That is why, even in diseases which involve duplication of the genes encoding amyloidogenic proteins, the soluble protein fraction is decreased and not increased. Lower levels of soluble Aβ are present in patients with APP gene triplication in Down syndrome (Portelius et al., 2014; Tapiola et al., 2001; Zammit et al., 2020). Thus, while overexpression is a genetic GOF in terms of producing more protein, it leads to biophysical LOF as the soluble protein is consumed in the fibrillar form due to uncontrolled nucleation and phase transformation. And since the fibrils are biophysically very stable; and thus, not very reactive (toxic), the pathophysiology is more likely to be due to LOF than GOF.  This explanation fits with genetics, biophysics, and clinical data without the need for the oligomer concept.

Moreover, the focus on genetics without considering the biophysical implications limits the explanatory power of the ACH to familial forms of the disease, leaving the more prevalent sporadic disease forms virtually unexplained. However, the biophysical mechanistic framework holds much more explanatory power, for example by highlighting surface-assisted heterogeneous nucleation as a fundamental force of inducing uncontrolled nucleation in the absence of genetic mutations (Malmberg et al., 2020). This opens the door for many risk factors (such as infections for example) to be included mechanistically in the pathophysiology, after being fiercely dismissed by the ACH as irrelevant.

The second GOF defense is that initial data from knock-out animal models did not show clear disease phenotypes. This was sometimes the case due to compensatory mechanisms that often obscured the outcome. However, careful study of the animal models in recent studies together with the development of the RNAi technology, which enabled adult protein knock-down, demonstrated LOF very clearly. We cite at least 30 publications showing disease phenotypes in knock-out/down animal models of different amyloid pathologies in our review (Malmberg et al., 2020). Moreover, another recent development was finding that the amyloid phenomenon affects as many as 40 proteins of very-well known functions, such as insulin, amylin, and P53. Importantly amongst them, P53 amyloid formation leads to cancer (Navalkar et al., 2019), clearly demonstrating an amyloid leading to enhanced cell proliferation due to LOF rather than cell-death due to GOF.  Oligomer toxicity won’t be able to explain that.

Time for a LOFolution!

LOF has always been in the literature, but never mainstream. The immense power of the ACH and its GOF assumptions meant that LOF researchers are taking a risk by going against the mainstream. That is why it is interesting to see that those researchers who dared to suggest LOF often used question marks in their titles (Fig. 2), proactively admitting that their approach is controversial. This justifiable cautiousness in my opinion is what prevented LOF researchers from taking the next logical step; advocate for replacement therapy. However, this should not be the case anymore, especially at a time when enormous GOF investments have not yielded any benefit for the patients.

 

Figure. 2

 

LOF is a valid scientific theory supported by genetics, biophysics, animal and clinical data. It holds much more explanatory power than GOF, and requires no ad-hoc entities such as the oligomers. LOF is a certain consequence of the amyloid phase transformation, GOF is not. Most importantly, LOF is treatable, GOF is not.

In patients with diabetes, we give them what they lost; insulin and amylin (both can form amyloids, by the way, that’s why amylin is given as an analog, pramlintide). It’s not cured, but it’s no longer a death sentence. The patients with other amyloid pathologies deserve the same chance.

We don’t need more of the same, we need a LOFolution!

References

Dear, A. J., Michaels, T. C. T., Meisl, G., Klenerman, D., Wu, S., Perrett, S., et al. (2020). Kinetic diversity of amyloid oligomers. Proc. Natl. Acad. Sci., 201922267. 

Haass, C., and Selkoe, D. J. (2007). Soluble protein oligomers in neurodegeneration: lessons from the Alzheimer’s amyloid beta-peptide. Nat. Rev. Mol. Cell Biol. 8, 101–12. 

Hardy, J. (2017). The discovery of Alzheimer-causing mutations in the APP gene and the formulation of the “ amyloid cascade hypothesis .” 284, 1040–1044. 

Ke, P. C., Zhou, R., Serpell, L. C., Riek, R., Knowles, T. P. J., Lashuel, H. A., et al. (2020). Half a century of amyloids: past, present and future. Chem Soc Rev. 

Malmberg, M., Malm, T., Gustafsson, O., Wright, A., Andaloussi, S. El, and Ezzat, K. (2020). Disentangling the Amyloid Pathways : A Mechanistic Approach to Etiology. Front. Neurosci. 14, 1–11. 

Navalkar, A., Ghosh, S., Pandey, S., Paul, A., Datta, D., and Maji, S. K. (2019). Prion-like p53 Amyloids in Cancer. 

Portelius, E., Hölttä, M., Soininen, H., Bjerke, M., Zetterberg, H., Westerlund, A., et al. (2014). Altered cerebrospinal fluid levels of amyloid β and amyloid precursor-like protein 1 peptides in Down’s syndrome. NeuroMolecular Med. 16, 510–516. 

Prusiner, S. B. (2013). Biology and genetics of prions causing neurodegeneration. Annu. Rev. Genet. 47, 601–23. 

Selkoe, D. J., and Hardy, J. (2016). The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol. Med. 8, 595–608. 

Tapiola, T., Soininen, H., and Pirttilä, T. (2001). CSF tau and Aβ42 levels in patients with Down’s syndrome. Neurology 56, 979. 

Walsh, D. M., Klyubin, I., Fadeeva, J. V., Cullen, W. K., Anwyl, R., Wolfe, M. S., et al. (2002). Naturally secreted oligomers of amyloid β protein potently inhibit hippocampal long-term potentiation in vivo. Nature 416, 535–539. 

Walsh, D. M., and Selkoe, D. J. (2020). Amyloid b-protein and beyond: the path forward in Alzheimer’s disease. Curr. Opin. Neurobiol. 61, 116–124. 

Zammit, M. D., Laymon, C. M., Betthauser, T. J., Cody, K. A., Tudorascu, D. L., Minhas, D. S., et al. (2020). Amyloid accumulation in Down syndrome measured with amyloid load. Alzheimer’s Dement. Diagnosis, Assess. Dis. Monit. 12, 1–9.