Yves here. A striking feature of KLG’s discussion of some of the problem with Alzheimer’s research is anchoring: that once a plausible-seeming theory is adequately substantiated, it can soon determine how scientists and clinicians view a phenomenon. That in turn makes them prey to confirmation bias and blinkering. Yet another manifestation of the Max Planck observation that science advances one funeral at a time.
By KLG, who has held research and academic positions in three US medical schools since 1995 and is currently Professor of Biochemistry and Associate Dean. He has performed and directed research on protein structure, function, and evolution; cell adhesion and motility; the mechanism of viral fusion proteins; and assembly of the vertebrate heart. He has served on national review panels of both public and private funding agencies, and his research and that of his students has been funded by the American Heart Association, American Cancer Society, and National Institutes of Health
As we age, as individuals and as a society, is there anything more frightening than Alzheimer’s Disease (AD)?
The answer was given more than 115 years ago by Auguste D, who was the first AD patient described by the German psychiatrist and pathologist Dr. Alois Alzheimer, when during an examination of Auguste D he asked her to write her name. Trying but failing she replied, “Ich habe mich verloren – I’ve lost myself.” What can be worse? Both for the patient, and his or her family. Frau Auguste D died at the relatively young age of 51, perhaps because she had familial AD, after an extended stay in a mental institution in Frankfurt, years that were financially ruinous to her husband Carl August Wilhelm Deter.
The dementia that became known as Alzheimer’s Disease was delineated in detail by Dr. Alzheimer in a lecture (recalled here) in Tübingen in1906, in which he described his patient and the neuropathology associated with her disease – including the amyloid plaques (Ab) and neurofibrillary tangles (tau) observed by Alzheimer and the Italian physician Gaetano Perusini (pdf) that since then have been diagnostic of AD.
The literature on Alzheimer’s disease is large (>200,000 articles in PubMed, 1913-August 2022), so in the interest of accessibility and concision, the description given here will follow that in Robbins & Cotran Pathologic Basis of Disease, 10thEdition, 2020, which is a standard comprehensive pathology textbook used in medical schools in the US. Remarkably, the current description in Robbins is largely that of Alzheimer and Perusini, albeit with a few additional details that could not have been known in 1906.
The known biochemistry of AD is straightforward, even simple. Amyloid precursor protein (APP) is a protein of unknown function found on the surface of cells in many tissues, including the central nervous system (CNS). Cleavage of APP by specific proteolytic enzymes (a-, b-, and g-secretases) can produce a non-amyloidogenic soluble fragment of APP or the amyloidogenicfragment that first forms soluble oligomers of a relative few Abpeptides (i.e., smaller fragments of APP). These oligomers then aggregate into the insoluble Abplaques in AD brain tissue that were first observed over 100 years ago. It is thought that the oligomers lead to synaptic dysfunction and cell death and activate enzymes (kinases) that add phosphoryl groups to tau, which then leads to the formation of neurofibrillary tangles that are left behind after the death of affected brain cells. Familial (inherited, early-onset) forms of AD support the hypothesis that Abproduction is critical for the initiation of AD. Mutations in two genesPSEN1and PSEN2encode the proteins presenilin-1 and presenilin-2, which lead to a gain-of-function in the g-secretase complex that produces increased amounts of Ab, particularly the Ab42 thought to be a major source of Aboligomers and plaques.
In contrast, familial forms of amyotrophic lateral sclerosis (ALS) account for a small percentage of cases of ALS, the vast majority of which are sporadic and of no currently known origin. Tau, which is the primary component of the neurofibrillary tangles in AD, is a microtubule-binding protein required for normal function of neurons of the CNS. Microtubules can be thought of as railroad tracks that traffic cellular cargo long distances, relative to average cell sizes, from the neuronal cell body to a distant perimeter defined by axons and dendrites and from the axons and dendrites back to the cell body, where the nucleus generally remains (unparalleled illustrations of neurons in situ by Santiago Ramón y Cajal can be found here). Without this structural and transport mechanism neurons cannot function properly.
So, to summarize from Robbins & Cotran, p. 1275:
The fundamental abnormality in AD is the accumulation of two proteins (Aband tau) in specific brain regions, likely due to excessive production and defective removal…The two pathologic hallmarks of AD, particularly evident in the end stages of illness are amyloid plaques and neurofibrillary tangles. Plaques are aggregates of Abpeptides in the neuropil (gray matter), while tangles are aggregates of the microtubule binding protein tau that persist after neuronal death. Both plaques and tanglesappear to contribute to the neural dysfunction, and the interplay between the processes that lead to the accumulation of these two types of abnormal protein aggregates is a critically important aspect of AD pathogenesis that has yet to be fully unraveled (emphasis added, to which we will return).
The amyloid hypothesis of AD was first put forward in the 1980s when APP fragments were identified as the primary constituent of amyloid plaques. The hypothesis has dominated a large fraction of AD research ever since, especially after the use of transgenic mouse models of AD, in which human APP/Abforms plaques in the brains of the experimental mice and the toxic oligomers described in Robbins & Cotran may lead to brain dysfunction.
In 2006 a seminal paper was published in Natureon a somewhat larger, apparently toxic Abfragment showing that:
(M)emory deficits in middle-aged Tg2576 mice are caused by the extracellular accumulation of a 56-kDa soluble amyloid-β assembly, which we term Aβ*56(Aβ star 56). Aβ*56 purified from the brains of impaired Tg2576 mice disrupts memory when administered to young rats. We propose that Aβ*56 impairs memory independently of plaques or neuronal loss, and may contribute to cognitive deficits associated with Alzheimer’s disease.
A subsequent paper from the same laboratory was published in Brain in 2013 and concluded that
(I)n cognitively normal (human) adults Aβ*56 increased ahead of amyloid-β dimers or amyloid-β trimers, and pathological tau proteins and postsynaptic proteins correlated with Aβ*56, but not amyloid-β dimers or amyloid-β trimers.We propose that Aβ*56 may play a pathogenic role very early in the pathogenesis of Alzheimer’s disease.
The first authorof each of these apparently very strong papers published in very good journalswas Dr. Sylvain Lesné, who was a postdoctoral research fellow in the laboratory of Dr. Karen Ashe, the corresponding author and principal investigator of the laboratory.
In the large scientific literature on AD, these two papers had a major impact on AD research. The first paper from Nature has been cited in the scientific literature more than 2,300 times,and the paper in Brain has been cited more than 200 times. This would be expected of seminal contributions that strengthened the foundation of the amyloid hypothesis of AD.
If Abpeptides are the “cause” of brain damage and consequent dementia in AD, then preventing their production should prevent AD or slow disease progression.
Alas, all is no longer well in this fundamental area of research on Alzheimer’s Disease. As described in a recent extensive news article by Charles Piller in Science, “Blots on a Field,” much of the data supporting the involvement of Ab*56 in the genesis of AD may have been fabricated.
The following notice has been added to the Naturearticle online: 14 July 2022 Editor’s Note: The editors of Nature have been alerted to concerns regarding some of the figures in this paper. Nature is investigating these concerns, and a further editorial response will follow as soon as possible. In the meantime, readers are advised to use caution when using results reported therein. Similarly, the Brain paper has been corrected extensively, nine years after publication.
“Blots on a Field” (BOAF) covers this apparent mess in depth, so I will only summarize the facts as they are currently understood. “Fabricated” is a strong accusation.
For the most part, Dr. Lesné has been accused of “manipulating” images. Research on Abusing cells and tissues as the source materials requires that the amount and molecular size of Abpeptides be monitored carefully throughout the project. These measurements are accomplished using a technique called immunoblotting (Western blot) in which carefully prepared samples of proteins from brain tissue in this case are separated by size and transferred to a permanent membrane where both the amount and size of the target peptide/protein can be determined using antibodies specific for Ab.
The remarkable utility of Western blotting is that it identifies one protein/peptide (with proper controls under defined conditions) in a mixture of thousands of distinct proteins that are present in varying quantities in the sample.
Western blotting is straightforward and simple when a qualitative assay is required (e.g., present or not present). With minor technical differences, immunoblotting is the basis of the rapid antigen tests for COVID-19 that are widely used. The free test kits sent to me by the US government state “The iHealth COVID-19 Antigen Rapid Test is intended for the SARS-CoV-2 nucleocapsid protein antigen in anterior nasal (nares) swab samples.” A single band result (C for control) indicates the test is working properly. A double band result (C plus T for test) tells you the test is working and that you have SARS-CoV-2 nucleocapsid protein, and therefore the virus, in your nose, despite having been vaccinated and boosted and boosted. But I digress.
The COVID test kit provides a robust yes/no answer, but Western blots used to study Ab(or any other peptide/protein) must provide information on peptide size and quantity during the initiation and progression of AD. This is determined using any number of quantitative image analysis techniques. Because of this, digital image files from Western blots are the data and they are sacrosanct.
If images have been manipulated, the data are no longer valid, and any conclusions based on these data are suspect. Image manipulation is not the equivalent of a photographic “touch up” to make the data “prettier.” Much as scientists want their data to look good, Western blotting can be messy and convincing at the same time. Depending on the difficulty of the experiment, slightly ragged data are often more impressive than a perfect picture. And research on Abis very difficult and demanding, from the biochemistry and biophysics of Abin solution to connecting brain pathology to measurable defects in memory and cognition.
If the underlying data have been manipulated, how could this have happened? The failures are both institutional and personal. Despite 2300 citations of the Naturepaper and 200+ citations of the Brain paper in the Alzheimer’s Disease literature, why has this taken 16 years to come to light? The first reason is undoubtedly that AD has been a stunningly difficult problem for more than 100 years, and perhaps more so in the past 40 years as molecular concomitants of the disease have been identified. From BOAF:
The experience (of Matthew Schrag, the primary investigator of Dr. Lesné’s publications) introduced him to a disquieting element of Alzheimer’s research. With this enigmatic, complex disease, even careful experiments done in good faith can fail to replicate, leading to dead ends and unexpected setbacks.
There is also research momentum to consider. For nearly 40 years the amyloid hypothesis, despite its provisional nature, has largely dominated Alzheimer’s research.
As noted above, a standard pathology textbook now in its 10thedition and used by thousands of medical students and physicians, hedges on Alzheimer’s Disease. The fundamental abnormality in AD is the accumulation of two proteins (Aband tau) in specific brain regions…The two pathologic hallmarks of AD…..are amyloid plaques and neurofibrillary tangles…which appear to contribute to AD but remain a critically important aspect of AD pathogenesis that has yet to be fully unraveled.
I will simply note that a “fundamental abnormality” associated with a disease is not necessarily the cause of a disease. Nor are “pathologic hallmarks” which “appear to contribute” to pathogenesis but “have yet to be fully unraveled.” We should also note that the early toxic oligomers of Abare considered by much of the Alzheimer’s research community to be the toxic forms of Ab. If so, perhaps the focus on amyloid plaques and neurofibrillary tangles, which appear after oligomers have begun to do their damage and may be an irrelevant endpoint, is misdirected?
Nevertheless, although amyloid-directed therapies have yielded few tangible results. From BOAF:
NIH spent about $1.6 billion on projects that mention amyloids in this fiscal year, about half of its overall Alzheimer’s funding. Scientists who advance other potential Alzheimer’s causes, such as immune dysfunction or inflammation or an environmental cause, complain they have been sidelined…
The amyloid hypothesis has led directly to several monoclonal antibody-based therapeutics such as aducanumab (Aduhelm, Biogen), which was approved by the FDA in 2021. The literature on Aduhelm is large, and the drug has been justifiably controversial.
A recent summary from the American Medical Association states:
Aducanumab does not cure or reverse Alzheimer disease. In 2 clinical trials, after 18 months it reduced amyloid plaque levels, but that did not translate to any clinical effect in 1 trial or a noticeable effect in the other. Potentially serious harms are common. The FDA has required that another trial be completed by 2030 to decide whether aducanumab has a meaningful patient benefit.
Aduhelm may be weak for now, but the amyloid hypothesis remains strong. $1.6 billion contributes to a substantial research momentum, and grantsmanshipis essential if researchers are to be successful in gaining access to that rather large pot of money.
A cardinal rule of grantsmanship is to go with the flow, while presenting conventional wisdom as new, improved, promising, and innovative. Ab*56 was new in 2006, and it has remained promising in 2022. Dr. Karen Ashe, who was the senior/corresponding author and the only other author in common on the Nature and Brain papers, called the investigation of Dr. Lesné “sobering.” But she also replied that she still has “faith in Ab*56” and
….we have promising initial results, I remain excited about this work, and believe it has the potential to explain why Abtherapies may yet work despite recent failures targeting amyloid plaques.
This is grantsmanship at its finest. The first practical advice every novice biomedical faculty member is to “generate preliminary data and get promising results” and use them to write a grant proposal. How an Abpeptide identified when George W. Bush was President has provided “promising initial results” in 2022 is something of a puzzle. But that is the way grants are procured.
At the personal level, it must be emphasized that the results published by Dr. Sylvain Lesné are currently under investigation, but no final conclusions have been reached and no other scientist or coauthor has been implicated in the apparent data manipulation.
The available information outlined in Blots on a Field (BOAF) doesn’t look good for Dr. Lesné, however. Several “personal” anecdotes included in the article are especially troubling for the practice of science.
It appears that Dr. Lesné has had a reputation of sorts since he was a graduate student. One senior co-author who wrote five papers with Dr. Lesné early in his career as a graduate student identified “dubious” results provided by Lesné, that could not be replicated by other students. The senior scientist at Caen subsequently “withdrew the paper before publication ‘to preserve (his) scientific integrity’ and broke off contact with Lesné.”
Nevertheless, Sylvain Lesné apparently completed his PhD in good standing and proceeded to make a career for himself in Alzheimer’s research. A junior scientist who worked for Professor Lesné at the University of Minnesota “calls him passionate, hardworking, and charismatic. She and others in the lab often ran experiments and produced Western blots…but in their papers together, Lesné prepared all the images for publication.” This behavior is clearly reminiscent of another case that was covered well in Retraction Watch. That established scientist and professor would take data out of the hands of techniciansand use it to prepare figures for his papers and grant applications. The fraud was discovered only after the principal investigator hired a senior scientist with a PhD, who had a continuing career interest in the research and necessarily became a whistleblower.
Dr. Lesné received from the National Institute on Aging (NIH) with a start date of June 2022. NIH has many funding mechanisms, with the R01 grant being the primary route to a successful and sustained career as an academic biomedical scientist. In my experience as a grant reviewer, the rule for evaluating the work of a former collaborator is that the relationship must have ended 3-5 years before. Dr. Yang was not a reviewer of the proposal by Dr. Lesné, but the current situation must be uncomfortable for him. This total mess should also be uncomfortable for scientists who now state that the amyloid hypothesis was always a stretch. If so, as reviewers and members of the research community, they were complicit in the expenditure of hundreds of millions on what may well be a chimera.
Although biological scientists have sometimes mistaken the agent of a disease for the cause of the disease (here and here, for example), there is still reason to believe that Abis fundamental to the etiology, development, and progression of AD, but perhaps with an environmental antecedent?
As pointed out to me in an email conversation not long ago, the pathology of Alzheimer’s seems to be something new and the result of an environmental insult, perhaps an infection. If toxic Aboligomers are the culprit, prevention of their formation should prevent Alzheimer’s Disease. However, if “current phase 3 clinical trials for three drugs targeting amyloid oligomers fail…the Abhypothesis is very much under duress” (BOAF).
Nevertheless, several things are certain. Institutional inertia, however plausible, should never be allowed to dominate any one field of biomedical research, particularly one as large and as important as Alzheimer’s Disease. Another is that a biological hypothesis that eludes capture, seemingly indefinitely, may not be correct. Plus, scientists who prove early they should be watched carefully cannot be allowed to escape notice. Ever. Otherwise, they “might further undercut public trust in science during a time of increasing skepticism and attacks” by being responsible for years of wasted money and effort by hundreds of scientists.” Dennis Selkoe is correct in this statement.
But public trust in scientists is indeed a trust. Once broken, it will be exceedingly difficult to regain, as we are reminded once more during a strange time.
 Amyloidogenic proteins are those that lead to the formation of insoluble protein aggregates in tissues. The amyloidosis caused by these proteins leads to distinct diseases (yes, Wikipedia, but this seemed to be the most useful non-paywalled link, especially the table included).
 By convention, the first author of a paper in the biomedical sciences made the largest contribution to the research and is generally viewed as the scientist “who did the work.” Papers often have several “first authors” who made equal contributions to the research. For example, the recent paperabout “super-vicious hamsters” (sic), naturally in Daily Mail, has three “first authors.” The last author, or corresponding author, is usually the principal investigator with overarching responsibility for the research project. Both the first and last author(s) receive equivalent credit for the work.
 Both papers appear upon first reading to be solid, in that they are well written and the data support their very significant conclusions. They were also published in excellent scientific journals. Any paper published in a legacy journal (something to which I plan to return) is cause for celebration. Brain: A Journal of Neurology was established in 1878 and is published by Oxford University Press. A biomedical paper published in Nature (1869) or Science(1880) or Cell (1974) is cause for champagne and dinner for the laboratory on the principal investigator’s credit card.
 To put this in context, most scientific papers are rarely cited. A paper with 100 citations may be considered a “citation classic,” and this does indicate the paper had a significant impact on related research.
 Grantsmanship (1961): The art of obtaining grants. As I mentioned in my previous poston why basic research is essential, “grantsmanship” has truly become a very important thing. There is even a Grantsmanship Training Program, ready and willing to help one become proficient in grant procurement; this is probably one of many. While I have had success in “procuring” grants, I confess to remaining clueless about what really matters the most in any given situation.
 Technicians and similar personnel are essential lab members but often have no continuing career interest in the research and are therefore unlikely to pay close attention to publications from the laboratory. In many laboratories, technicians are not expected to be coauthors, although this is changing. As a former research technician, I approve of this practice!