ORI: Self-Plagiarism is NOT Misconduct

Yes, you read that correctly. This revelation (to me at least) comes from a recent interaction with the Federal Office of Research Integrity (ORI), namely this reporting of several image duplications across papers from the same lab. Importantly, in most cases*** the images were re-used within a similar context and to describe the same experiments.  Here is the response I received from ORI…

Dear Dr. Brookes:
Thank you for your thorough email regarding the re-use of images and data across nine (9) different publications, over the span of fourteen (14) years. The 2012-2018 publications and after, are within the six year period of limitations, thus are under ORI’s jurisdiction.  In addition, the Current Drug Targets 2008 and Biocatal Biotransformation 2010 publications would also be under ORI’s definition, per 42 C.F.R. Part 93.105 (b)(1) Subsequent use exception.

However, re-use of images and data does not meet the definition of research misconduct.  Per 42 C.F.R. Part 93.103: Research misconduct means fabrication, falsification, or plagiarism in proposing, performing, or reviewing research, or in reporting research results. Falsification is manipulating research materials, equipment or processes such that the research is not accurately represented in the research record.

In each instance of re-use, the research is accurately represented in the research record; thus there is no falsification.  The re-use is consistent with self-plagiarism, which also does not meet ORI’s definition of plagiarism.  Thus ORI does not have jurisdiction.  This also may be consistent with a copyright violation, depending on the specific journal’s policies.

As ORI does not have jurisdiction, ORI considers this a closed matter.

Thank you,
XXXXXXXXXX, Scientist Investigator,
Division of Investigative Oversight,
Office of Research Integrity

This is news to me! It’s the polar opposite of what we teach in our mandatory research ethics course (i.e., self-plagiarism is bad). Even if the experiments are the same, simply republishing the same data and images more than once raises two important ethical issues:

(1) Double-Dipping. When you publish the same data twice you get to <i>game</i>  the metrics system by getting more publications for the same or less work (vs. others who do new experiments each time). Most would agree this is “not fair”, as important events in academia such as promotions and tenure depend on such metrics.

(2) Copyright. When you publish a figure in one journal, typically that journal owns the copyright to the image. Even if you’re a CC-BY hawk and do everything open access, publishing the same figure again elsewhere without acknowledgement is a breach of someone’s copyright.

This issue clearly raises a dilemma from the journal editorial perspective…. Say for example something is published first in Journal A and then Journal B.  If A sues B for copyright infringement and it results in retraction of the paper in B, all good.  But, if B acts alone and retracts the paper, they have to be VERY careful. If they implies any kind of misconduct occurred, and the authors are savvy about the above-mentioned ORI policy, this could open the door for a defamation lawsuit from disgruntled authors.  Caveat editor!


***In some cases the images were not reporting the same experimental conditions. In other papers there were clear examples of image manipulaton such as splicing together unrelated western blots. But, all those papers were >6yrs old and so fell outside the ORI statute of limitations. Thus, overall there was nothing that both met the definition of misconduct AND was within the S.O.L.  Oh well.

Ugh! TFW you just can’t escape image manipulation…

Last week I had the great pleasure of visiting Sabzali Javadov at the University of Puerto Rico, to participate in the PhD thesis defense of Rebecca Parodi-Rullan (Rebecca had spent some time in my lab a few years ago on an SfRBM mini-felowship).  She passed (yay!) and is off to start a post-doc at NYU soon.  Here’s a picture of Dr. Javadov’s awesome lab group, who I greatly enjoyed meeting with and discussing science:

But, as often happens during such visits, there was some down-time, so I found myself waiting in a corridor outside someone’s lab.  What else is there to do but read a poster?

Well, let’s just say the poster had some “imaginative approaches to the re-use of loading controls on western blots”.  I snapped a cellphone pic of the author banner and decided to pull some papers when I got home.

Here is what I found. Ugh!  Details are posted on PubPeer here, here, here and here, for those of you who don’t want to download a PDF with all the details.  Essentially it’s mutiple examples of images being re-used across 9 different papers spanning 2004-2018, all from a single PI’s lab – Dipak K. Banerjee in the Department of Biochemistry at UPR (Dr. Banerjee is the only author common to all 9 papers). He was and is funded by NIH, and some of the papers fall within the ORI 6 year statute of limitations. Thus, ORI has been informed, as have all the journal editors, and COPE (since all the journals are COPE members).

This sucks!  What should otherwise have been an enoyable trip now leaves a sour after-taste. Scientific misconduct never sleeps, it just keeps following me into the most unlikely places. When you’ve seen it once, it’s impossible not to see it everywhere you look. Ugh!

AHA comes through, and other news…

PSBLAB post-doc’ Chaitanya Kulkarni (Chaitu) found out this week that his AHA post-doctoral fellowship (the one he wrote just 2 months after joining the lab!) is gonna be funded!  Congratulations Chaitu!

Next month Paul is speaking at Drug Discovery 2018 in San Francisco CA, and in January we’re presenting posters at this Keystone CO meeting on Cardiac and Skeletal Muscle Mitochondria.

Yves Wang’s paper on the mitochondrial unfolded protein response and ATF5 was resubmitted recently (after getting the run around elsewhere. The pre-print on BioRxiv has been updated.

A few years ago, Rebecca Parodi-Rullan, a grad’ student in the lab of Sabzali Javadov at the University of Puerto Rico  was awarded an SFRBM “mini-fellowship” to spend time in our lab learning the mouse Langendorff perfused heart method. In December Rebecca will defend her PhD thesis, and then she’s off to NYU for a post-doc!

Also in December is the beginning of Module 5 – Metabolism – in the new IND431  “Foundations in Modern Biology” course. In addition to making the lecture notes available to URMC students on BlackBoard, I am investigating the possibility to try to host some of the course materials here.

Upcoming Conferences

The lab will be at a few conferences during the fall…

(1) The Society for Heart & Vascular Metabolism (SHVM) conference in Charleston SC (Paul speaking).

(2) The Translational Research in Mitochondria Aging and Disease (TRiMAD) conference at Penn State, State College PA. One of the speakers, Dave Brown from Virginia Tech, is going to stop by Rochester to viit us on his way there.

(3) Drug Discovery 2018 in San Francisco CA (Paul speaking).

We’re also probably going to this Keystone meeting next January, which looks freakin’ awesome!

Unfortunately due to other committments we’re probably going to miss the AHA and SFRBM meetings (both in Chicago).

Our latest paper on NMN (and changing publication norms)

Our latest paper on the acute effects of nicotinamide mononucleotide (NMN) is out in J. Mol. Cell. Cardiol.  For those without access to the journal, a pre-print of the paper was placed in BioRxiv and is still there.

The paper stems from some work we started a few years ago, trying to find out if the lysine deacetylase SIRT1 can regulate intracellular pH.  In a previous paper, we had shown that the SIRT1 inhibitor splitomicin can cause cytosolic alkalinization in cardiomyocytes (Figure 4 of that paper). So, the next logical step was to see if stimulating SIRT1 can do the opposite. In addition, we’d published several papers showing that SIRT1 activation is cardioprotective against ischemia-reperfusion injury, so we had a general interest in seeing if SIRT1 stimulators might be protecting the heart by altering pH.

Behold, NAD+ supplements!

There’s a giant poo-storm in the literature about how best to stimulate SIRT1, with resveratrol being everyone’s least favorite molecule, and other alleged compounds (e.g., the Sirtris molecules) being of questionable efficacy. But, many people have been paying attention to the growing field of NAD+ supplementation. Since the sirtuin enzymes use NAD+ as a substrate, there’s a lot of interest in orally bioavailable precursors of NAD+, including nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR).

Naturally, with any small molecule that has potential ability to prevent or reverse diseases of aging, there are significant commercial interests at play.  Namely, Elysium (Leonard Guarente and David Sinclair) is shilling NR pills under the name “Basis”, at $40 for a month’s supply.  Chromadex is hawking a similar product under the name “Tru-Niagen” at $50. Other companies are selling NMN with all sorts of health claims, and if you’re cheap like me you can even buy NMN on Amazon. Naturally, the mainstream press is beginning to pay attention to all this fountain of youth chatter.

The data…

So, what happened when we put NMN into hearts and heart cells? It acidified the ever-loving crap out of them! Here are the key data, using a fluorescent pH indicator in adult mouse primary cardimyocytes…

So all good right?  Hypothesis proven – stimulating SIRT1 makes cells acid, just as inhibiting it makes them alkaline. Not to just figure out how SIRT1 is regulating pH and we’re all set.  Not so fast!  It turned out that blocking SIRT1 didn’t do anything to NMN’s ability to acidify, so it must have been working via another mechanism.

Long story short – the mechanism is simple bulk stimulation of glycolysis. As you will recall from grade school biochemistry, a key step in glycolysis is GAPDH, which uses NAD+.  The reason we have lactate dehydrogenase (LDH) is to convert the NADH generated by GAPDH back into NAD+, so the pathway can continue working.  Ergo, without NAD+ glycolysis cannot work, and so boosting NAD+ would be predicted to make it work faster…

(If you don’t know this stuff by heart, do yourself a favor and grab a copy of the excellent “Metabolism at a Glance” by J.G. Salway. It’s a required desktop accessory in any serious metabolism/mitochondria lab). Anyway, using various methods including 13C-glucose flux tracing, we showed that NMN boosts glycolysis.

We also looked at lysine acetylation, as a surrogate marker for sirtuin activity (since the SIRT enzymes are lysine deacetylases). In a quite surprising result, we also found that NMN caused robust deacetylation inside mitochondria, without any impact in the cytosol.

A probable explanation for this is that SIRT1 (which is mostly cytosolic in myocytes) doesn’t really get a look-in on the bolus of NAD+ made from NMN.  There’s an crap-ton of GAPDH around (after all, what do we all use as a housekeeping marker protein on our western blots?), so it just chews through all the NAD+, depriving SIRT1 of that sweet nucleotide action. We assume that in the mito’, the NAD+ pool is protected, and so deacetylation can still occur via mito’ sirtuins such as SIRT3.  Some excellent recent work by Joe Baur on NAD compartmentalization and synthesis is probably relevant to this point.

We also know from some work done by Jim Downey and others that acidic pH in the heart can be protective in the setting of reperfusion injury (acid pH keeps the mito’ permeability transition pore closed). So, we were able to show that delivering a burst of acidosis by giving NMN at the moment of reperfusion is also cardioprotective. We also showed that blocking glycolysis (by removing glucose from the media or by using galactose to prevent glycolysis from making any net ATP) can block the cardioprotective effects of NMN. Overall, although it had already been shown that NMN is cardioprotective, the mechanism was thought to involve SIRT1, but our work now shows that glycolysis and acid pH can also play a role.

So, what does this all mean for the burgeoning NAD+ supplementation industry?

First off, I should stress that the effects of NMN we saw here were all acute, taking place within 10-30 minutes of addition. Nevertheless, I think there are some implications for the human usage of NMN or NR supplements….

(i) The Warburg effect is a key feature of cancer.  To put it in simple terms, cancer cells are addicted to glycolysis. There’s also some literature suggesting that stem cells (which share some metabolic features with cancer cells) can benefit from NAD+ supplementation. So, what do you think would happen if we acutely boosted glycolysis in cancer cells?  Is it wise to boost NAD+ levels in cancer patients?

(ii) Although the overall tone of this post might seem negative toward the entire field of NAD+ supplementation, it’s hard to argue with the rather impressive effects that have so far been reported in animal models and even in humans. As such, it’s worth asking whether acute, transient metabolic acidosis might actually be part of the beneficial mechanism of these molecules?  On that topic, there’s another common intervention that causes acute metabolic acidosis, and is rather famous for having beneficial effects in  various diseases… exercise!

Evolving open publication strategies

As suggested in the post title, it’s worth highlighting that this is now the 6th paper for which we’ve posted a pre-print on BioRxiv. A couple of years ago when we dipped a toe in the pre-print water for the first time (with this paper), it was with some trepidation! The surprise for me is how this has now become the standard publication workflow for my lab – we just post pre-prints without really thinking about it.  Similarly, almost 3 years ago for the first time we published the complete raw data set for a paper on FigShare. Now it’s just something we do as part of the normal paper preparation process. Sure, it’s a lot of work to curate everything into an accessible format, but it also really makes you look at your data carefully, so is helpful in preventing mistakes. As a bonus – mentioning both of these strategies (pre-prints and open data) in the Rigor & Reproducibility section of the NIH grant proposal forms has so far worked very much in our favor!  It’s a tangible way to show reviewers that you’re walking-the-walk, not just talking-the-talk.