Adriano Aguzzi is prion researcher, director of the Institute for Neuropathology at the University Hospital (Universitätsspital) Zürich in Switzerland, member of several academies and a self-professed champion of publication ethics. Critics of his papers are described as a lynch mob and “lowlifes”, because apparently all those data irregularities do not matter in the greater scheme of things. Aguzzi’s prion research namely saved many lives. Or so he says.
The image integrity sleuth Elisabeth Bik now followed up on some year-old whistleblower concerns once posted on PubPeer. Back then, Aguzzi and his coauthors reacted with aggressive insults towards his critics. That said, every group leader can unknowingly become victim of a rogue PhD student or postdoc. But evading responsibility when faced with evidence of data manipulation, refusing to correct the scientific record while viciously attacking your critics is not an ethical response. Some define it as research misconduct. It seems, however, that Aguzzi has changed his stance now.
Update 14. and 21.12.2019: Aguzzi now announced to sue me for defamation. His inspiration comes from my sentencing in Berlin court over the deadly trachea transplants performed by Paolo Macchiarini and assisted by Philipp Jungebluth (the plaintiff). Aguzzi wrote this in the comment below:
“I will have no option but a lawsuit. Get prepared for another conviction, and start saving for your legal expenses. You may learn your lesson this time.”
Now Aguzzi made his threat official. He demands an immediate removal of this article and all tweets referencing him, my lifelong commitment never to say anything critical about him, otherwise he will pursue a lawsuit, with damage compensation of €15,000. A translation of this letter is provided at the end.
Readers of my site might recall that Aguzzi as panel member recently awarded the France-based German cancer and ageing researcher Guido Kroemer with the €1mn “Nobel Lombardo” (read here), despite or maybe actually because of Kroemer’s notoriety as peddler of Photoshopped research data. When awarding Kroemer, Aguzzi had no problem joining a quasi-all-male panel with the most toxic peer of all: Carlo Croce, with 9 retractions so far and a lost lawsuit against his critic David Sanders and the New York Times. To be fair, Aguzzi left the panel afterwards, officially to make room for women next year.
Aguzzi’s world is weird. The Zürich professor occasionally claims having both MD and PhD degrees, but his own CV proves that he never did any PhD training, but graduated as MD in 1983, while on leave in USA during his medicine studies in Switzerland and Germany. In fact, Aguzzi explained that he sees his Doctor honoris causa (Dr h.c.) from University of Bologna as a kind of quasi-valid PhD degree. It is not really the same, years of researching in a lab and defending a doctorate thesis versus dressing up in a fancy hat for a party organised by admiring peers.
Update 27.12.2019. Aguzzi insists his Bologna degree is a PhD, even if a honorary one. Actually, what his own diploma says is that he received a “diploma di Laurea”, which is equivalent either to a Bachelor or a Master degree. In Italy, laurea graduates are traditionally addressed as “dottore” (or dottoressa), but a PhD degree is called in Italy “dottere di ricerca”, and is not to be confused with the dottore of the laurea. Hence, according to the diploma he posted, Aguzzi seems to have only a BSc h.c. or a MSc h.c. title from the University of Bologna, and not the honorary PhD at all.
But now, let’s discuss Aguzzi’s scientific record. It seems, his worst enemies are the envious lynch mob “lowlifes” who discuss his papers on PubPeer. Like here:
Johannes Haybaeck , Mathias Heikenwalder , Britta Klevenz , Petra Schwarz , Ilan Margalith , Claire Bridel , Kirsten Mertz , Elizabeta Zirdum , Benjamin Petsch , Thomas J. Fuchs , Lothar Stitz , Adriano Aguzzi Aerosols transmit prions to immunocompetent and immunodeficient mice PLoS Pathogens (2011) doi: 10.1371/journal.ppat.1001257
In July 2013, an anonymous PubPeer critic voiced these concerns:
“Figs 3 and 4 use four differently cropped, exposed and rotated images for the JH-/- (Fig. 3) and c3c4-/- (Fig. 4) mice. Is one of these mice both the same?
Fig. 5E left lane has been spliced onto the gel.
Fig. 5G right panel. Oddly blotchy pale, pixellated background with less detail than for the dark matter in the lanes. Multiple straight line effects at edges of bands. This gel appears to have been composed from multiple individual components inserted into a lower resolution background.
Fig. 6C left panel. Right vertical edge has a jaggie revealing that a shorter lane has been spliced in.
Fig. 6D right panel. strong splice between lanes 6/7″
The commenter also wondered if it made sense to declare that every single author has contributed everything to the paper.
Aguzzi immediately announced to investigate, and promised to share the results:
“My policy is that all raw data be kept securely for at least 10 years, hence I foresee no problems with data retrieval. I will post the results of the audit.“
More than two years, passed, and in November 2015 Aguzzi replied again:
“I desisted from my original intent to publish further materials on this forum once it became clear that some of the self-appointed “peers” are more interested in mud-slinging than in post-publication review and quality control of science. I see no need to defend my work against a group of anonymous detractors who negate a priori my integrity, yet are not even prepared to identify themselves. Suffice it to say that Mathias Heikenwälder and I have gone carefully through all original data and we confirm that none of the alleged “manipulations” were performed. We stand fully by the findings which we have reported. Both Mathias and I will be prepared to make additional materials available to whoever will contact us directly and identify himself/herself, be it the Editor of PLoS Pathogens or anybody else. What we are NOT going to do is to engage in shouting match with anonymous individuals who make libelous and unfounded accusations. Please note that I will NOT make any further comments or communications on this forum.“
The anonymous commenters pleaded with Aguzzi to reply again, and eventually the paper’s first author Johannes Haybäck (now director of Institute of Pathology at the University of Magdeburg in Germany, with a second affiliation at University of Graz in Austria), did:
“I apologize for not having reacted earlier, but I had not understood the precise nature of the issue with the contested histology figures. I agree with that a mistake may have occurred while preparing the histology panel for publication. These are BSL4 experiments, which are carefully documented for reasons of both biosafety and animal welfare. I have planned to travel to Zurich on Dec. 7th in order to re-inspect all histology together with a coworker from the Aguzzi lab. The paraffin blocks from which the histology was prepared are certainly still existent, and I am confident that the issue can be clarified. If indeed there are errors, a correction will be published in the PLoS Pathogens paper. “
In February 2015, PLOS Pathogens issued a Correction, for figures 3, 5 and 6. On PubPeer, Aguzzi then posted a lengthy comment, where he shared his views on anonymous post-publication peer review. He regretted that his “former student Johannes Haybäck committed an error while mounting Fig. 3A” , but saw little responsibility on his own part. Instead, this:
“Now, I have been active in research since 1982. Over the past 34 years, I have published >450 paper. Notwithstanding the error in Fig. 3A, the Haybäck paper is one of those that I am proudest of. Our discovery of aerosols as prion vectors has led to a significant reassessment of the safety procedures for lab work with prions. By drastically improving the safety of lab workers, this work will likely save human lives. […]
In summary, if my scientific legacy will turn out to be a better protection from occupational hazards for students, postdocs, and lab technicians, well, that is a legacy that was worth working for, and one that I am proud of.”
Well, it is not like lab personnel used to get sick and die from prion exposure in epidemic proportions until Aguzzi’s life-saving paper came along. Come to think of it, there never were any cases of a dangerous prion infection in the lab. Aguzzi was exaggerating his scientific legacy, again.
That same whistleblower flagged in July 2013 other Aguzzi papers on PubPeer, but not even corrections have happened. Like here:
Brady Michel , Adam Ferguson , Theodore Johnson , Heather Bender , Crystal Meyerett-Reid , Bruce Pulford , Adriana Von Teichman , Davis Seelig , John H. Weis , Glenn C. Telling , Adriano Aguzzi, Mark D. Zabel Genetic depletion of complement receptors CD21/35 prevents terminal prion disease in a mouse model of chronic wasting disease Journal of immunology (2012) doi: 10.4049/jimmunol.1201579
Bik provided good visualisation of the problem, namely some very insolent reuse of western blot bands in Figure 4:
“We have commissioned an independent forensic analyses of the figures in this manuscript and will update this post with a report when it is complete. For now, we are sharing all of the raw data images used for these experiments and to create the figure here https://www.sugarsync.com/pf/D6450680_0347642_9170813
We are committed to the highest scientific rigor and integrity in our research. We stand by the data and the conclusions drawn from them in the paper, and will submit an errata to this paper to update the figure if these duplications are confirmed.”
Another PubPeer critic however retorted:
“the problem already exists in the raw data since the two first Ig H blots (page 7 of above pdf) are actually manipulated versions of the same blot.”
Zabel then provided his final decision in October 2014 that nothing at all wrong was with the paper. He explained that loading controls were irrelevant anyway, and even accused the whistleblowers of having fabricated evidence in Photoshop:
“An independent analysis of the files and blots, as well as interviews with authors of the original paper, concluded that “duplication or manipulation of raw blots was deemed unlikely, unintentional or inconclusive. Explanations given by the technicians who performed the experiments are plausible and no further action was recommended. In fact, it was demonstrated that unrelated blots could actually be made to look extremely similar by the same manipulations that raised concern about the images published in the manuscript.
In short, it could not be determined definitively whether these images were inadvertently duplicated blots, but that it was unlikely that any intentional manipulation of the blots had occurred. Moreover, the bands in question were mainly controls, which would not have significantly altered data interpretation. We therefore stand by the data and the conclusions drawn from them in the manuscript.“
Of course the paper was not corrected, and rest assured it never will be. Journal of Immunology made it their business model of doing nothing at all about falsified research (read here and here). How else can infectious diseases be cured if not with the modern technology of Photoshop?
Another coauthor of Aguzzi was more constructive, after Bik flagged a badly manipulated figure. Background patches were copy-pasted, presumably to cover up results which stood in the way of the fight against a deadly infectious prion disease.
Ciriaco Ligios, Maria Giovanna Cancedda , Antonello Carta , Cinzia Santucciu , Caterina Maestrale , Francesca Demontis , Mariangela Saba , Cristiana Patta , James C DeMartini , Adriano Aguzzi, Christina J Sigurdson Sheep with scrapie and mastitis transmit infectious prions through the milk Journal of Virology (2011) doi: 10.1128/jvi.02022-10
“We are looking into these blots as a high priority.”
Let’s hope Sigurdson will not pull a Zabel on this one, and that Journal of Virology is more ethical than Journal of Immunology, which is actually not that difficult. In the next case however, it is perfectly clear that the society journal Blood is not interested in a correction: Blood has a policy of ignoring everything older than a year or so, regardless of how bad. Hence, no need for Aguzzi or his Universitätsspital Zürich colleagues to bother about this reused and shifted image:
Stefan Baenziger , Mathias Heikenwalder , Pål Johansen , Erika Schlaepfer , Ursula Hofer , Regina C. Miller , Simone Diemand , Kenya Honda , Thomas M. Kundig , Adriano Aguzzi , Roberto F. Speck Triggering TLR7 in mice induces immune activation and lymphoid system disruption, resembling HIV-mediated pathology Blood (2009) doi: 10.1182/blood-2008-04-151712
Being such an important man, and likely unbothered by the journals, Aguzzi so far saw no reason whatsoever to address the concerns posted in parallel in 2013 regarding this paper:
Megan Larson , Mathew A Sherman , Fatou Amar , Mario Nuvolone , Julie A Schneider , David A Bennett, Adriano Aguzzi, Sylvain E Lesné The complex PrP(c)-Fyn couples human oligomeric Aβ with pathological tau changes in Alzheimer’s disease Journal of Neuroscience (2012) doi: 10.1523/jneurosci.1858-12.2012
Also the last author Sylvain Lesné of University of Minnesota in USA did not reply. Maybe he or Aguzzi will now, after Bik provided some illustrations? Or is this evidence also falsified in Photoshop, like previously uncovered by Zabel? How is this approach to curing Alzheimer’s to explain:
Bik’s concerns about this Nature paper from Aguzzi lab were so far not dignified with a comment:
Michael B. Fischer , Christiane Roeckl , Petra Parizek , Hans Peter Schwarz , Adriano Aguzzi Binding of disease-associated prion protein to plasminogen Nature (2000) doi: 10.1038/35044100
A paper containing such a falsified figure would not be not reliable anymore. Falsifying data in Photoshop, regardless which of the authors did it, does affect the conclusions, and it is the responsibility of the corresponding author to act upon the evidence. It does not matter how many lives Aguzzi thinks he has saved with his prion research.
Update 13.12.2019. Aguzzi has discussed this Nature figure in the comment section below, and now first author Michael Fischer replied by email:
“We had been able to answer questions about the paper to the satisfaction of everyone (Nature and University of Zurich) 4 years ago. I therefore do not want to comment again.“
Aguzzi did reply to this commenter, who raised unspecified concerns about “Figures 1, 3, 4, 5” in March 2017. The first author was meanwhile finally tasked with retrieving raw data.
Caihong Zhu, Uli S. Herrmann , Jeppe Falsig , Irina Abakumova , Mario Nuvolone , Petra Schwarz , Katrin Frauenknecht , Elisabeth J. Rushing , Adriano Aguzzi A neuroprotective role for microglia in prion diseases The Journal of Experimental Medicine (2016) doi: 10.1084/jem.20151000
Back in 2017, the PubPeer critic also voiced concerns about shifted baselines, high mortality among the alleged healthy controls mice, and also this bizarre approach to statistics:
“Experiments are largely underpowered: serve as examples N=4 or 3 from figures 3, 4, which are central to the argument of the paper. The dispersion of the data indicates extensive variability, calling for a substantial increase of the N. I am very surprised these datasets comply with the requirements for normality and homocesdasticty needed for the ANOVA“.
The editors and reviewers at the elite journal somehow missed all that, or just trusted Aguzzi to know what he is doing. Aguzzi’s reply arrived years later, in July 2019, typos his:
“The sentence “I wonder if somebody more expert could have a look at westerns at Figures 1, 3, 4, 5? Some of them look odd to me, but I’m not an expert in picking up image manipulation” provides a vivid demonstratin of everything that’s wrong with PubPeer.
The commenter is unhappy with this publication, most likely because he has some personal beef with one of the authors. However, as much as he would wish to prove the paper wrong, he cannot find any flaws in the figures. Therefore, he claims that they “look odd” to him, hoping that the mob will rise and lynch the authors – or at the very least that something disreputable will stick.
Sadly (but unsurprisingly), the world of research is not devoid of lowlifes. The meanness, spite, the posturing and the defamation under the protection of anonymity, you find them on Reddit and you find them on PubPeer as well.“
That professorial reply to the lynch mob “lowlifes” actually provides a vivid demonstration of everything that’s wrong with Aguzzi. The statistics concerns are actually very much valid, and as for western blots: the loading controls do not always look like they show same gel. Despite Aguzzi et al having written in the methods section:
“To avoid variation in loading, the same blots were stripped and incubated with an anti-actin antibody”
That was apparently a fib, or an honest error of oversight, for the noble purpose of saving lives. Like here, in Figure 5, where an grid was placed to visualise that, sadly, and unlike Aguzzi et al explicitly wrote, GFAP signal and Actin loading control did not derive from same gel:
But as we learned before from Aguzzi’s colleague Zabel, loading controls are irrelevant anyway, presumably in prion research at least. Although, Figure 1E does contain undisclosed splicing, in a paper from 2016:
Aguzzi eventually understood that maybe it was he who was wrong. The gels are possible not what they are presented to be. Hence this new reply:
“I agree that this is weird. I’ll try to retrieve whatever original materials are around. My former postdoc Caihong Zhu has recently moved to Shanghai, but his records are meticulous and we are in touch every week. In fact we are still doing a lot of experiments together and he is still maintaining a large mouse colony in my institute. I am confident that he can shed light on this issue.”
With a paper barely 3 years old, the raw data should be available. Different with Aguzzi’s 11 year old paper, flagged by Bik for some very naughty Photoshoppery aimed at extirpating prion disease transmission.
Nicolas Genoud , David Ott , Nathalie Braun , Marco Prinz , Petra Schwarz , Ueli Suter , Didier Trono , Adriano Aguzzi Antiprion prophylaxis by gene transfer of a soluble prion antagonist American Journal Of Pathology (2008) doi: 10.2353/ajpath.2008.070836
Such things as in Figure 1C and 5h do not happen by mistake, those are intentional data manipulations. Regardless of who did it, as as corresponding last author and lab head Aguzzi is very much responsible for what he publishes. And no, having published many papers in Science, Nature and Cell is not an excuse.
Aguzzi warned however:
“Thank you for pointing out these issues. I agree with your concern and I will try to retrieve the original histology slides. I cannot promise that I will find anything though. These experiments were performed >15 years ago. My policy is to retain all original materials for at least 10 years. Older materials have been occasionally retained, but not in a systematic manner.“
Aguzzi also replied in similar vein on another paper, where an image was duplicated without change. That might be one of the rare cases of an honest mistake.
Mathias Heikenwalder , Christian Federau , Lotta Von Boehmer , Petra Schwarz , Mareike Wagner , Nicolas Zeller , Johannes Haybaeck , Marco Prinz , Burkhard Becher , Adriano Aguzzi Germinal center B cells are dispensable in prion transport and neuroinvasion Journal of Neuroimmunology (2007) doi: 10.1016/j.jneuroim.2007.09.022
Aguzzi now announced to recover original data and correct the literature. Great start! He also promised to stop calling himself a PhD (something he used to do even in papers and book chapters as corresponding author).
Update 12.12.2019. For completeness, I decided to discuss this collaborative paper from Christian Haass lab at LMU Munich in Germany, coauthored by Aguzzi.
Armgard Uelhoff , Jörg Tatzelt , Adriano Aguzzi , Konstanze F. Winklhofer, Christian Haass A pathogenic PrP mutation and doppel interfere with polarized sorting of the prion protein Journal of Biological Chemistry (2005) doi: 10.1074/jbc.c400560200
What we see here, is a western blot experiment where the “no signal” analysis was probed on one gel (Dpl in A and PrP in B), but the “with signal” analysis was made on a different, or maybe even several different gels, spliced together. There is therefore no positive control or evidence that any protein at all was loaded on the “no signal” gel. The authors might want to correct this figure, which is otherwise meaningless.
Before reading Aguzzi’s comments below, you might want to prepare yourself with this eulogy in Tagesanzeiger.
“A postdoc of his held last week a lecture at the Institute of a Nobel laureate in Israel, as the neuropathologist Adriano Aguzzi posted recently on Facebook. After the talk, the Nobel laureate told the postdoc: “Adriano should have received the Nobel Prize together with Stanley Prusiner”. This is not correct, Aguzzi continues in his post, “but my ego grew enormously, today it does not fit through the door anymore. “
The article continues to decry the tragedy that Aguzzi was not given the deserved Nobel Prize.
Update 16.12.2019. Another Aguzzi paper is discussed on PubPeer:
Mario Nuvolone , Nicolas Schmid , Gino Miele , Silvia Sorce , Rita Moos , Christian Schori , Roger R. Beerli , Monika Bauer , Philippe Saudan , Klaus Dietmeier , Ingolf Lachmann , Michael Linnebank , Roland Martin , Ulf Kallweit , Veronika Kana , Elisabeth J. Rushing , Herbert Budka , Adriano Aguzzi Cystatin F is a biomarker of prion pathogenesis in mice PLoS ONE (2017) doi: 10.1371/journal.pone.0171923
Aguzzi’s reply was:
“It is simply one of the many attacks ad personam that Leonid Schneider (who was recently convicted and sentenced to a stiff pecuniary fine for slander by a German court, and was in addition ordered to pay all legal expenses to the person whom he slandered) is launching against me.
I could ask PubPeer to remove the slanderous comments by Schneider and others. But I will not, because I prefer everybody to see and judge for themselves the level of absurdity, incompetence and meanness of his latest smear campaign. Schneider, don’t delude yourself that I will not fight back.”
Aguzzi sees every attempt to critically discuss his papers as a personal attack, a smear campaign and slander, for which he threatens to sue his critics. This is unfortunately a kind of retaliation which every whistleblower has to deal with. Not sure what kind of signal University of Zürich wants to be sending here.
Aguzzi sent me today me this registered letter, the original is shown at the beginning of the article:
Dear Mr. Schneider
You spread untrue claims about me and my employees. I request of you now, informally and extrajudicially, the following acts:
- Please delete your blog entry “https://forbetterscience.com/2019/12/12/aguzzi-and-the-lowlifes”
- Please delete all of your tweets and other online posts which mention my person and my employees
- To refrain lifelong from mentioning my person in the future, from judging, attacking, hinting, speaking of, coercing, and publicising in any form online or otherwise.
This letter is meant as a friendly, unprejudiced request. If you comply with this request I will consider this unfortunate matter as settled and will undertake no further steps.
However, should you do not comply with my request within the next 10 days, you will receive from my legal representative a cease and desist declaration with penalty clause.
As you know from your previous experiences, this will be combined with compensation demands which correspond to the severity of your law violation. We envision a compensation of €15,000, which we intend to donate to a charitable foundation. You probably also know that repeat offenders can expect little understanding from the judiciary.
I sincerely hope that you recognize my benevolent willingness to settle out of court, will comply with my request and save you a lot of trouble.
Christian Schwarzenegger, Univerity of Zürich’s Vice-Rector for faculty affairs and law professor, informed me on 22 December 2019 that the university is not involved with Aguzzi’s announced legal action against me. He however declared:
“the university leadership has initiated an investigation of the accusations you raised. The procedure is as usual based on the presumption of innocence. As soon as we have a report, we will inform you about the next steps.”
If you are interested to support my work, you can leave here a small tip of $5. Or several of small tips, just increase the amount as you like (2x=€10; 5x=€25). Your generous patronage of my journalism, however small it appears to you, will greatly help me with my legal costs.