Nanotech fraud theres nothing small about it

A while ago, I posted about the existence of new blog based in India, focused on exposing fraud in the nanoparticle drug delivery field. In particular, the work of SK Sahoo was put under scrutiny.

Shortly afterward, the blog went silent for undisclosed reasons, but much of the evidence of image manipulation has come into my possession. After carefully pulling the original articles and validating each case, I present to you here an absolute shit-load of fraud (and yes, just in case you need to ask, wherever an image is reproduced it is used to represent two totally undrelated experimental conditions in each case)

PLoS One. 2011;6;e26803; PMID22110595

ACS Appl Mater Interfaces 2011;3;842-56; PMID21370886

Biomaterials. 2012;33;2936-51; PMID22264522

Acta Biomater. 2012 Jul;8(7):2670-87. PMID22484149

Acta Biomater. 2012 Feb;8(2):704-19. PMID22051236

Biomaterials. 2010 May;31(13):3694-706. PMID20144478

Biomaterials. 2011 Aug;32(24):5643-62; PMID21600647

Acta Biomater. 2011 Jan;7(1):355-69. PMID20727991

Biomaterials. 2009 Oct;30(29):5737-50. PMID19631377

Mol Pharm. 2011 Jun 6;8(3):852-66. PMID21480667

Mol Vis. 2011;17:2724-37 PMID22065926

Nanomedicine. 2011 Dec;7(6):859-70. PMID21550422

Eur J Pharmacol. 2011 Nov 30;670(2-3):372-83. PMID21951969

Eur J Pharm Sci. 2010 Jan 31;39(1-3):152-63. PMID19961929

Biomaterials. 2010 Sep;31(25):6597-611. PMID20553984

If you werent keeping count, thats 15 papers with image manipulation. There are also other accusations beyond my technical understanding (e.g. the accuser claims problems with gating voltages used for flow cytometry). Sahoo appears to have about 60 papers, so has certainly crossed the red flag 25% threshold, entitling anyone to ask serious questions about all his work.

This case is also interesting from the gender perspective Take a look at Sahoos lab website, specifically the members of his research group.  All 12 of them are women. Quite the harem!  As a female scientist myself, I dont see anything wrong with that, but I do think its a little creepy for a male senior scientist to have a lab entirely staffed by females. In a male-dominated society like India (which routinely kills female babies), does anyone think these subordinates originated the fraud, or did the male P.I. tell them to?


An isolated case of image re-use?

Dragana Filipovic of the Vinca Institute of Nuclear Sciences in Belgrade, Serbia, studies the effects of isolation stress on rats. It seems the author himself may have been a little stressed during the preparation of these figures

J Neurosci Res 2011; 89;1461-1470; PMID21656845

Neuroscience 2012;223;238;245; PMID22885231

The second example is a little benign, but does raise an important point when blotting for 2 proteins of almost identical molecular weight, its not always good practice to blot/strip/re-blot, in-case some of the signal from the first protein bleeds through into the second. Sometimes its unavoidable (e.g. blotting for the phosphorylated version of a protein, then re-blotting for the total protein to normalize the phosphorylation level), but best practices really call for blotting 2 separate membranes when the proteins are unrelated and of similar mass.

In the methods section of this manuscript, the authors descibe blotting for one protein OR the other, implying different membranes were used. Furthermore, the bands for the 2 proteins of interest exhibit different shape/spacing/slope/features, suggesting they did originate from different blots. However, the loading controls are the same for both proteins (Bax on the left, Bcl-2 on the right), which implies only membrane was used.  Very frustrating.


Aggarwal fails miserably at correction

Bharat Aggarwal is on the defensive. Hes already withdrawn a couple of papers that were found to contain fraudulent images before they even hit the press.  His latest effort is a correction to an article in Antioxidants & Redox Signaling, and its both tragic and f***ing hilarious at the same time

First, lets take a look at the original problem, as reported by blogger Juichii Jigen

Antioxid Redox Signal 2012;16;413-27; PMID22004570

This is Figure 7 of the paper, and its pretty obvious the problem is re-use of a B-actin loading control western blot in panels 7B and 7C for different experimental conditions.  Remember those panels, 7B and 7C.


Now heres the correction published last week in the journal

Antioxid Redox Signal. 2012 Sep 21. [Epub ahead of print] AUTHOR REPORTED CORRECTION OF WESTERN BLOT DATA: Kim JH, Park B, Gupta SC, Kannappan R, Sung B, Aggarwal BB. Antioxid Redox Signal. 2012 Mar 1;16(5):413-27. PMID22998640

This is a non peer-reviewed author-reported erratum addressing: Zyflamend sensitizes tumor cells to TRAIL-induced apoptosis through up-regulation of death receptors and down-regulation of survival proteins: role of ROS-dependent CCAAT/enhancer-binding protein-homologous protein pathway. Kim JH, Park B, Gupta SC, Kannappan R, Sung B, Aggarwal BB. Antioxid Redox Signal. 2012 Mar 1;16(5):413-27.

The authors claim that Figure 7 reporting Western blot data was erroneous. Specifically, the beta-actin panel of Fig. 7B was found to be switched with that of Fig. 7D. The corrected version is reported here. The authors claim that this correction does not influence the conclusion of the study.

Accompanying the correction was this figure

Aggarwal claims that the problem was the blots in panels 7B and 7D were switched, and in doing so COMPLETELY MISSES THE POINT that the blots in 7B and 7C were duplicated. FAIL!

What makes this whole thing even more crazy is we now have a corrected figure in which the blot in panel 7C is still duplicated, only this time its with the one in panel 7D instead of 7B.  Heres what it looks like now

This brings up serious questions about Aggarwals control over his own data sets.  If you cant even publish a simple correction without botching it up, what does that say about the integrity of the rest of your work? Well, we have these 65 papers plus 20 more called into question, which give us a pretty good idea about data integrity in the Aggarwal lab.

As an aside, I would imagine there will be a collective sigh of relief at the journal, because it was specifically stated this correction was not peer-reviewed. Nice bullet-dodging there ARS!


Yet more Prion fraud

Its the gift that keeps on giving!

Prions just cant catch a break lately. First we had Claudio Hetz, then his mentor Claudio Soto, then Stanley Prusiner, then Bradley Hyman, amounting to nearly 20 papers in the field being called into question.  Well now it seems we can add two more labs to the growing list of those producing questionable data

The first is David Westaway, of the Prion center at the University of Alberta.  These are interesting examples, because they all required messing with levels in the images to reveal where they were joined together.  Clearly, after splicing, someone took care to make sure it wasnt visible to the naked eye.

Plos Pathogens 2011;7;e1002391; PMID22114562

J Neurochem 2010;113;92-104; PMID20067571


The second lab is Maria Gasset, of CSIC in Madrid Spain, and part of the NeuroPrion consortium. The force is strong with this one, amounting to blatant re-use of multiple bands within panels of a single figure.  Did they honestly think nobody would notice?

J Biol Chem 2012;in-press; PMID22955286 (doi: 10.1074/jbc.M112.398776)

Im seriously considering whether to rename this blog  Id register the domain name, but theres probably enough money been wasted on this joke of a scientific field already!


Does collaboration foster misconduct?

Today we showcase a paper from Molecular Cell, which includes authors from no less than 8 separate departments in a university.  While normally such collaborative efforts are praised, its increasingly apparent that with authors spread over multiple places (and often in different cities), keeping track of all the data can be difficult. This can provide fertile ground for the kind of naughtiness you see below.

Molecular Cell 2006;23;641-650; PMID16949361

The first example is common-or-garden western blot re-use, with a bit of re-sizing and alteration of levels to make the blots look sufficiently different, but the dots above and below the bands give the game away.

The second example (bonus 2 kinds of fraud in 1 paper!) is from electrophysiology (patch clamping) work, where traces have been re-used between different conditions.  Notably, the traces have been re-sized ever so slightly, making an innnocent copy/paste error seem unlikely.

So, does this type of collaborative effort involving multiple labs make it easier for fraudsters to peddle their wares?  What can a lead-author do to ensure the integrity of data in a paper coming from multiple groups?  Do you trust the people you collaborate with?


Pathological mis-use of western blots

Todays case concerns two papers in the American Journal of Pathology, highlighting shockingly poor standards of peer review. Both papers come from the lab of Sheeran Ezzat at the University of Toronto, and concern epigenetic regulation of fibrobloast groth factor receptors (FGFRs).

Am J Pathol 2010;176;2333–2343; PMID20348248

Am J Pathol 2010;177;2860-2869; PMID21037081


Aside from both papers being in the same issue of the journal, with the same authors (that should always be a red flag for editors), the bigger issue here is that these image manipulations are pretty obvious to the naked eye, even without the highlighting tricks used here. Heck, some of them are even within the same panel of a single figure!  Its going to be very hard for the journal to make the case that their peer review process is up to scratch, when such obvious fraud gets through.

Its also worth nothing that if Dr. Ezzats lab website is any indication, this is not someone who cares alot about image. It hasnt been updated since 2002!



Breaking news Dimitrios Iliopoulos (of miniscule error bars fame) has figured out a way to get cancer stem cells to come back to life!

Anonymous tipster TC reports on two papers, in which Iliopoulos was able to culture cancer stem cells from frozen human breast cancer specimens. Thats nothing special, until you realize that the specimens were purchased from commercial suppliers who snap freeze their samples in liquid nitrogen without any cryopreservative. Here are the papers, and their relevant methods sections

Proc Natl Acad Sci 2011;108;1397-402; PMID21220315

Five human invasive ductal carcinoma tissues (stage III) were purchased from AMS Biotechnology and Biochain Inc.

Cancer Res 2011;71;3196-201; PMID21415163

Three snap-frozen ductal carcinoma tissues were purchased from AMS Biotechnology and Biochain Inc. The purification process of CD44high and CD24highderived from human breast tumors [...] is detailed in the Supplementary Methods.

Supplementary materials Five human invasive ductal carcinoma tissues (stage III) were purchased from AMS Biotechnology and Biochain Inc. The magnetic bead separation is less stressful to the tumor cells in comparison to multiple rounds of cell sorting, thereby allowing the use of these cells in culture experiments.

Results We examined whether metformin selectively inhibits the growth of CSCs from human breast tumors. CSCs (CD44high/CD24low) and NSCCs (CD44low/CD24high) were purified from human mammary adenocarcinomas.

So far so good, until a visit to AMS website brings up the following

Frozen Tissue Blocks and Snap-Frozen Tissue: All fresh frozen tissue samples are collected by certified medical pathologists and are snap-frozen in liquid nitrogen within 30 minutes of surgery excision.

Frozen Tissue Sections: Tissues available include human adult and fetal normal tissues, human diseased and tumor tissues, as well as from mouse, rat, and primates. The tissues are excised, immediately frozen by liquid nitrogen and then stored in -80ºC.

And heres Biochains methodology

The tissues are immediately put in liquid nitrogen after being excised, and then identified by a board-licensed pathologist.

Note that neither supplier mentions anything about cryo-preservatives, and its generally agreed in the field that you need such things (e.g. DMSO) to stand any chance of getting viable cells back from the dead. Further confusing the matter, the Iliopoulos papers reference  two other papers, in which cells were cultured from biopsy samples, but in both those papers the samples were freshly collected, not frozen.

With Halloween just around the corner, maybe now is a good time for Iliopoulos to share his secrets on how to get zombie cells out of liquid N2 frozen tissue samples.

[Update] In between preparing this post, and publishing it,TC informs that he/she messaged the NIH and the journals concerned. Then, PNAS actually asked for an explanation from the authors, and based on this the journal responded that they consider the matter closed. Apparently the authors claimed DMSO was used as a cryoprotectant.  Also theres the problem that DMSO is useless if the freezing process is too fast.  All the samples were snap frozen in liquid N2, so DMSO wont do any good. The rest of the authors explanation makes wild claims about stem cells having special properties that enable them to survive freezing differently to normal cells (how exactly do stem cells resist the physical shearing forces of crystallizing water?).  Theres also the inevitable high profile publications from the top experts in the cancer stem cell field have validated ALL our findings claim.

Of course, I emailed the suppliers about whether they used cryo-protectants, and they had this to say

No cryo-preservative agents are added to our snap frozen tissue samples. We just dissect connective tissue, fat and other unwanted material in ice-cold PBS from freshly-resected tissue of interest. Then we snap frozen them in liquid nitrogen.

Any chance Iliopoulos will come forward with his magic pixie-dust zombie stem cell formula? Did the journal really fall for this BS explanation, which is opposite to what the supplier says?


I hate beta actin!

If I had $1 for every faked beta actin loading control Ive seen, I could buy this bike. It seems some people just cant be bothered to do loading controls the proper way (i.e., stripping and re-probing the blot), so they just cobble together a control from somewhere else and hope it gets past the reviewers. The VERY first thing I look at in a paper is the beta actin loading controls. If theyre dodgy in any way, the whole paper is called into question. Heres another one that recently found its way onto my Scott pile

Proc Natl Acad Sci 2011;108;4129-34; PMID21325052

As the caption in red says, there are several duplicated panels between the 2 figures in this paper.  This alone might be OK, because they represent the same proteins (Tsc2, pS6, and S6) in the same samples (WT and Tsc2 knockout mice, with or without rapamycin).  That being said, WHY ARE THE ACTIN CONTROLS DIFFERENT?

In one of these figures the beta actin control that is presented, cannot possibly belong to the Tsc2, pS6 and S6 blots if one is correct then the other one is wrong.  One figure being valid (i.e. the control is actually from the same gel), by default means the other one is not valid.

I long for the day when we can wistfully look back on the beta actin era and write it all off as a horrible mistake.

EGCG does it

A couple of weeks ago I reported on the case of Mark Stearns, the retired cancer researcher who set up shop as an Herbal Supplements seller.  In addition to the 1 retraction and 3 other papers flagged in the previous post, two more are now flagged as suspect

Mol Cancer Res 2004;2;403-16; PMID15280448

In the Fig. 3 example, the caption really says it all.  Back in the early days of western blotting, maybe 20 years ago, you could get away with spliced up images, but by 2004 this was definitely off limits. The inclusion of this image brings up some serious questions about the rigor of peer review at the journal.

Transl Oncol 2011;4;147-156; PMID21633670

The second paper, on epi-gallo-catechin-gallate (EGCG), a green tea polyphenol, includes the usual blot splicing shenanigans some blots spliced but the loading controls not.  Ergo, the loading controls did not originate from the same gels.

As noted previously, the institutions of both Stearns and his co-author Wang are being tight lipped on the circumstances of their departure earlier this year. Sadly nothing new to report in the intervening period.

In addition, as noted in the comments earlier, someone tried to leave an objective comment on the Yelp review page of Stearns new shop, and it got removed. This suggests someone (Stearns?) is monitoring that page, and so must be aware of this site.