19 September 2022

Deception decades long

Systematic falsifications have been found in articles on the mechanism of development of Alzheimer's disease

Georgy Kurakin, "Elements"

A decades—long scientific deception has been revealed - a neuroscientist systematically falsified data on the role of beta-amyloid in the development of Alzheimer's disease. Two independent image analyses at once showed that the author was simply "sketching" the beta-amyloid strips he needed on the results of protein electrophoresis. Perhaps this will force us to reconsider the leading theory of the occurrence of this disease.

Lumps and tangles

"I get lost in the yard of my own house. I do not know where I was going, and I forget how to get home. <...> While I managed to get back. <...> It is not ignorance that scares me, but the question of when I will reach the line, having crossed it, I will not be able to return." This is how one of the heroes of the documentary series "The Alzheimer's Project: Recordings of Lost Memory" (The Alzheimer's Project) describes his symptoms. Alzheimer's disease is the most common cause of dementia in old age. It accounts for 60 to 80% of cases of senile dementia, and according to US statistics, more than 10% of people over the age of 65 suffer from this disease.

The above quote well describes how terrible Alzheimer's disease is — its steady progression inevitably leads a person to a complete loss of consciousness, personality, ability to communicate and serve himself. And to death, if he does not die from other causes during this time. Such a clinical picture resembles another group of diseases — prion diseases, which develop due to improper stacking of proteins in the neurons of the brain. The only difference is that Alzheimer's disease develops much more slowly, and it is impossible to get infected with it.

The similarity is not accidental. It has long been known that Alzheimer's disease is also associated with an incorrect structure of proteins, and first of all, the APP protein located on the membranes of neurons. This protein has many functions — for example, it is involved in the formation of synapses and protection from infections. It is cleaved by the enzymes α-, β- and γ-secretases, which leads to the formation of peptides of different lengths depending on the cleavage point. Some of these peptides are able to stick together into dense lumps (it is more correct to call them plaques), which, when stained with histological dyes, resemble starch grains (Fig. 1). Therefore, the substance of these plaques is called amyloid (from the Greek word ἄμυλον — starch).


Fig. 1. Amyloid plaques in the brains of people who died from Alzheimer's disease are clearly visible even with routine histological staining and really look like a starch-like dense substance, which is why they got their name. Image from the website commons.wikimedia.org .

Amyloid is deposited outside of neurons, in the intercellular space. When this protein "garbage" accumulates too much, nerve cells begin to die.

Since there are many amyloids (and they can be formed not only from APP and not only in the brain), the amyloid formed from APP is called beta-amyloid, or Aß. Actually, the name of the APP protein itself stands for Amyloid-beta precursor protein — the precursor protein of beta-amyloid.

The role of this protein in the development of the disease has been studied the most — firstly, beta-amyloid (as already mentioned) is clearly visible under the microscope during pathoanatomic examination. Secondly, all familial forms of early onset Alzheimer's disease are caused by mutations in the APP gene or in the genes of its metabolism (encoding proteins presenilin-1 and presenilin-2).

But the fatal sequence of events in the brain does not end with beta-amyloid (Fig. 2). Inside the neurons themselves, another protein begins to stack incorrectly. We are talking about tau protein, which normally stabilizes microtubules. They, like rails, pass inside the long processes of neurons (axons), and various substances are transported along them along the axons. In Alzheimer's disease, the tau protein begins to tangle into tangles, and the functioning of the brain becomes similar to the operation of a railway with construction debris on the rails every kilometer. The neurotransmitter glutamate also enters the process — it is released uncontrollably and over-activates its receptors, which leads to cell death.


2. Alzheimer's disease is based on a violation of the structure of brain proteins — APP protein forms amyloid plaques, and tau protein tangles into tangles. This leads to the death of neurons and a steady loss of cognitive functions. Drawing from the website biorender.com , with changes.

To date, the most reasonable point of view is that beta-amyloid is the root cause of all disorders, and it is the accumulation of beta-amyloid plaques outside neurons that triggers a cascade of pathological changes inside them, including the tangling of tau protein into tangles, and the leakage of glutamate... This theory is called the amyloid cascade theory.

The similarity of beta-amyloid and tau protein tangles with prions lies not only in their shape and clinical picture of the disease, but also in the kinetics of the process (J. M. Nussbaum et al., 2012. Alzheimer disease). Apparently, beta-amyloid also behaves like a prion at the molecular level — if small aggregates of beta-amyloid are formed in the brain, they further serve as "condensation nuclei", inducing the transition of other APP-peptides into amyloid form and their further "sticking" to the nascent aggregate. The same thing happens with tau protein. The only difference is that it is apparently impossible to get infected with Alzheimer's disease by eating the brains of those who died from this disease (like Kuru and Creutzfeldt—Jakob disease).

Nevertheless, scientists still sometimes have some doubts — not least related to the discovery of unexpected links and pathogenesis factors that are not yet integrated into the theory of the amyloid cascade, like an extra piece in a puzzle. For example, certain alleles of the apolipoprotein E gene increase the risk of Alzheimer's disease. The surprise is that the main function of this protein is to participate in the transport of cholesterol. Thus, it turns out that cholesterol is also somehow "involved" in the development of Alzheimer's disease. But how is still unclear. In general, doubts concern whether the accumulation of beta-amyloid is the leading link in the pathogenesis of the disease, or whether it is triggered by some other mechanisms, such as, for example, a violation of cholesterol metabolism or even infections and inflammation!

The latter assumption has received a second life in the light of one recent work — on brain organoids, scientists were able to show the possible role of the most banal viruses in triggering Alzheimer's disease (D. M. Cairns et al., 2022. Potential Involvement of Varicella Zoster Virus in Alzheimer’s Disease via Reactivation of Quiescent Herpes Simplex Virus Type 1). We are talking about the herpes simplex virus type 1 (HSV-1, the one that causes a "cold on the lips") and the chickenpox virus (causing "chickenpox", most often in children). About these viruses "like" to infect nerve cells. Individually, they do not cause changes in them characteristic of Alzheimer's disease, but if the chickenpox virus infects neurons in which HSV-1 has already "settled" (which is called superinfection), this leads to rapid deposition of beta-amyloid and the formation of tangles of tau protein. Thus, Alzheimer's disease may also have an infectious origin.

The question of the pathogenesis of the disease is very important, because without knowing it in detail, we are practically deprived of the chance to develop at least some effective medicine for it. Currently, drugs that affect the action of neurotransmitters (for example, rivastigmine and memantine) are used for its treatment. They are able to slow down the deterioration of memory and thinking a little, but not to prevent the death of neurons, so over time they lose effectiveness. Scientists do not give up trying to create a drug aimed at the mechanism of development of Alzheimer's disease itself — and basically "aim" at beta-amyloid.


In 2006, a young neuroscientist Sylvain Lesné and his co-authors published an article in the journal Nature, A specific amyloid-β protein assembly in the brain impairs memory, which at that time seemed revolutionary to the scientific community. Even then it was clear that at the tissue level, Alzheimer's disease is associated with the accumulation of amyloid plaques, but what exactly the role of beta-amyloid was not completely clear.


Lesne and colleagues claimed to have discovered beta–amyloid oligomers with particles weighing 56 kDa, which they briefly called Aß*56. In the first article, it was postulated that the introduction of such an amyloid to experimental rodents contributed to the development of changes in them that characterized Alzheimer's disease in humans. Well, exactly like a prion! In subsequent articles, the topic Aß*56 developed, but one message remained unchanged: these 56-kilodaltone oligomers are the root cause of all the molecular ills befalling the neurons of the diseased.

The main method of detecting protein aggregates of a given molecular weight in Lesne's work was western blotting - electrophoresis of proteins with antibody detection. Simply put, proteins of different molecular weights were "accelerated" in an electric field, forming bands in a polyacrylamide gel. Of course, images of such electrophoregrams were included in every article. Lesna's articles were quoted thousands of times, and no one suspected anything was wrong until everything was revealed randomly.

The story of the exposure began with the fact that one lawyer needed the help of an expert in a delicate process against a pharmaceutical company that launched the drug simufilam. Clients of the lawyer — neuroscientists — tried to challenge the results of studies of the drug and claimed that they were falsified. The lawyer called Matthew Schrag — a neuroscientist and psychiatrist from Vanderbilt University — and brought him in as an expert. Shreg managed to establish that there were indeed falsifications. But along the way, he found something else interesting.

For the purpose of examination, he visited the PubPeer post-publication review service, where scientists can leave comments on the works of their colleagues. He was looking for articles on Alzheimer's disease to which critical comments would be written. But it wasn't just the articles on simufilam that attracted his attention. It turned out that Lesne's works were also "popular" — colleagues left numerous doubts on the service about the reliability of the images of Western blotting proteins published by him. Shreg became interested and decided to conduct a detailed analysis of these images to identify signs of "painting".

He subjected the archive of Lesne's publications for many years to computer analysis, combining parts of the images one with another. As a result, numerous exact coincidences of stripes in different places were found in the articles of the neuroscientist, which could not be explained by coincidence. The fact is that in Western blotting, protein strips form "blots" with uneven edges, the shape of which is largely random. Exact coincidences could only point to one thing — Sylvain Lesne "painted" part of the strips by copy-pasting on a computer. The "painting" was localized precisely in the area assigned by the researcher to Aß*56. Figure 3 clearly explains the technique of detecting copied fragments in images, which was used by Shreg.


Fig. 3. How to determine that the data on the electrophoregram is falsified by copying: step-by-step instructions on the example of a model image from the GIMP editor. a) I drew a picture similar to an electrophoregram, applying strips "by hand". Approximately the same appearance will have spots on real electrophoregrams. b) To save effort, some of the strips are simply copied from row to row. Will you be able to determine that I have deceived you? c) Having found some suspicious coincidences in two rows, you highlight them, make a digital negative out of the image and color the rows red and green. Then combine the rows with each other in such a way that the colors are combined when overlaying. d) The copied stripes will match completely and will be monotonously yellow due to the overlap of red and green. Stripes obtained in good faith will have different edges — therefore, a complete match will not work. A red-yellow-green "blob" will be observed. Such actions are repeated many times with different images. A large number of such coincidences most likely indicates a large-scale falsification and casts doubt on the published results. Drawing by Grigory Kurakin.

Who should I trust now?

A description of Schrag's findings was published in a recent issue of the journal Science and for some time became one of the main news of the scientific community. Before publishing the results of this investigation, Science engaged two more image analysis experts who rechecked Lesna's electrophoregrams and generally confirmed Schrag's arguments. There really was a large-scale falsification lasting many years.

This is an unprecedented amount of analytical work to identify falsifications in scientific research that has ever been published in Science. Andrey Zayakin, a Russian physicist and a member of the Dissernet community, which is engaged in identifying scientific falsifications in Russia, commented on it this way: "The amount of work of Shreg is amazing — it can be called a "One—man Dissernet," hundreds of strips have passed through his hands."

Following Schrag's investigation, other inconvenient details about amyloid Aß*56 began to emerge. It turned out that no one could single him out at all! But only one group of researchers has published such negative results (G. M. Shankar et al., 2008. Amyloid β-Protein Dimers Isolated Directly from Alzheimer Brains Impair Synaptic Plasticity and Memory). Simply because negative results are published much less often than positive ones. At the time of publication, no one noticed that Aß*56 was not found for some reason.

So what now — the hypothesis of the amyloid cascade turned out to be wrong? It is premature to draw such conclusions yet. The very idea of beta-amyloid as the most important trigger factor of Alzheimer's disease was "held" not only by the works of Lesna and his mythical Aß*56: there is too much evidence in favor of the guilt of beta-amyloid, from genetics to microscopy. We now only know that specifically Aß*56 is invented and drawn — but the possible role of other forms of beta-amyloid is still written off. Even other oligomers can play the role that Lesne attributed to the fruit of his imagination. Schrag himself says that "the broader story of amyloid oligomers may survive this problem, but one should think about the origins of this situation."

What happened really makes us ask a more important question: how much can we believe science and how can science protect itself from such stories in the future? A thorough check of the images and data of the article before publication could save from such manipulations, but it is not yet known whether the editors of scientific journals will be able to develop some common standard and follow it (although calls for this are not the first time).

The moral of this whole story can be served by the phrase of Matthew Schrag from an interview in the journal Science: "You can cheat to publish an article. You can cheat to get a degree. You can cheat to win a grant. But you can't cheat to cure the disease. Biology doesn't care."

Source: C. Piller. Blots on a field? // Science. 2022

Portal "Eternal youth" http://vechnayamolodost.ru

Found a typo? Select it and press ctrl + enter Print version