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

A common sentiment among scientists is that their vocation, which it remains for a steadily shrinking population that will soon be a remnant,[1] is self-correcting.  It is not always obvious what is meant by self-correcting, but any scientist who has spent more than a few years in the lab has corrected his or her own work.  This goes with the territory.

And then there are students and scientists who believe, a verb I use advisedly here, that “if it is in the literature, it is a fact.”  This can apply to their own work and that of others.  In the first instance, every specialized field has its members who are always in a hurry for whatever reason and are thus forced, if they are honest, to correct their work.  This is sometimes accomplished by failing to cite their previous research that was, how shall we put it, “premature.”  And then there are others for whom anything and everything in the literature is fact.  These people, you need to watch out for them.

Several years ago, I was asked to review a paper for a leading cell biology journal.  This manuscript described novel findings built upon previous work that had been done in the laboratory of which I was a member at the time.  I had had nothing to do with the research, however, and after more than 10 years this did not represent a conflict of interest.  I thought the manuscript under review was very good and represented a genuine advance in the field of protein-protein interactions in multicomponent assemblies responsible for cell adhesion and motility, possibly one more piece of the puzzle to explain cancer progression/metastasis.  Thus, I had only a few minor suggestions for the authors and the Senior Editor of the journal, which other than the exceedingly rare “Thank you for your manuscript, which is accepted without revision,” is the review we all hope to receive.

Reviewer #2 had different ideas, so it is not always Reviewer #3 (NSFW; with apologies to the late, great Bruno Ganz).  Since the research under review seemed to contradict what had been published previously, it just could not be correct according to Reviewer #2, and this meant that the results had to be “wrong” and should not be published in such a leading journal.

In the third round of review in a journal that generally allowed only two, I recommended that the authors state, which they did in the final revision (paraphrase), “Because we have used previously unavailable imaging and biophysical techniques, our results extend rather than contradict those of the previous research.”  Eventually Reviewer #2 relented, and the paper was published as an example of “self-correcting science in action” by going beyond what was previously possible.  Of course, the editor could have made that decision on his own after Round 2, but I digress.

This example was a “win” for the authors and for other scientists working on cell adhesion and motility.  But occasionally a situation comes along that is simply unfathomable.  Eleven years ago, bear with me, a paper appeared in the journal Science[2] with the title “A Bacterium That Can Grow by Using Arsenic Instead of Phosphorous.”  The primary and first author was Dr. Felisa Wolfe-Simon, then of the NASA Astrobiology Institute and the US Geological Survey and the paper made a very big impression, as it should have based on the Abstract:

Life is mostly composed of the elements carbon, hydrogen, nitrogen, oxygen, sulfur, and phosphorus. Although these six elements make up nucleic acids, proteins, and lipids and thus the bulk of living matter, it is theoretically possible that some other elements in the periodic table could serve the same functions. Here, we describe a bacterium, strain GFAJ-1 of the Halomonadaceae, isolated from Mono Lake, California, that is able to substitute arsenic for phosphorus to sustain its growth. Our data show evidence for arsenate in macromolecules that normally contain phosphate, most notably nucleic acids and proteins. Exchange of one of the major bio-elements may have profound evolutionary and geochemical importance. (emphasis added)

What are the authors saying here?  That alternative life forms are possible!  Yes, but is it really conceivable that on earth, where it is generally accepted due to overwhelming evidence from the evolutionary biology of molecules and cells that life arose once and that all living organisms are descended from that common ancestor, that arsenate can substitute for phosphate in the sugar-phosphate backbone of DNA (and RNA) in the Watson-Crick model of DNA that all of us learned about in high school biology.

The first reaction to this paper of every biochemist and molecular biologist I asked was “No way!”  If this were to be true, then the enzymes responsible for the synthesis and degradation of RNA and DNA in this bacterium named GFAJ-1, which is closely related to other bacteria that do not live in arsenic-rich Mono Lake in California, would also have co-evolved to be active with a sugar-arsenate backbone instead of a sugar-phosphate backbone.

Then there is the question of the stability of the sugar-arsenate backbone in the bacterium.  The water concentration will be the same inside the cell and DNA or RNA with arsenate instead of phosphate in the backbone will be much more susceptible to hydrolysis, i.e., the cleavage of sugar-arsenate bonds by hydroxyl ions in water.  This does not happen at a measurable rate with sugar-phosphate backbones in the absence of an enzyme (nuclease) to catalyze the reaction.  DNA susceptible to cleavage by water is not likely to be useful as the genetic material.  On Planet Earth, that is.  Recalling our high school chemistry, the hypothesis that arsenic can substitute for phosphorous because it is below phosphorous in Group 5A of the Periodic Table of Elements is as plausible as a silicon-based instead of the carbon-based organic chemistry we have because the former is below carbon in Group 4A.  The latter was a subject of nerd bull sessions back in the day.

As one might imagine, this paper is an example of the adage that “extraordinary claims require extraordinary evidence” and as such it led to something of a “storm.”[3]  It turns out that the techniques used to “prove” that arsenate substituted for phosphate in GFAJ-1 were not well suited to the task, and most of the evidence in support of the conclusions was indirect.  It is not difficult to purify DNA from bacteria, even slow-growing bacteria such as GFAJ-1.  The authors did not do this and show that arsenate was a constituent of DNA from GFAJ-1.  Two papers appeared together in Science a year later.  The first, from the laboratory of Julia A. Vorholt at ETH-Zurich, showed that GFAJ-1 Is an Arsenate-Resistant, Phosphate-Dependent Organism:

The bacterial isolate GFAJ-1 has been proposed to substitute arsenic for phosphorus to sustain growth. We have shown that GFAJ-1 is able to grow at low phosphate concentrations (1.7 μM), even in the presence of high concentrations of arsenate (40 mM), but lacks the ability to grow in phosphorus-depleted (<0.3 μM), arsenate-containing medium. High-resolution mass spectrometry analyses revealed that phosphorylated central metabolites and phosphorylated nucleic acids predominated. A few arsenylated compounds, including C6 sugar arsenates, were detected in extracts of GFAJ-1, when GFAJ-1 was incubated with arsenate, but further experiments showed they formed abiotically. Inductively coupled plasma mass spectrometry confirmed the presence of phosphorus in nucleic acid extracts, while arsenic could not be detected and was below 1 per mil relative to phosphorus. Taken together, we conclude that GFAJ-1 is an arsenate-resistant, but still a phosphate-dependent, bacterium. (emphasis added)

Or to put it in anthropomorphic terms, bacteria are remarkably ingenious and very patient, as any biochemist or microbiologist will know.  They can figure out a way to grow under the most extreme conditions: scarce food, high salt, high arsenic, temperatures approaching the boiling point of water in hot springs and next to geothermal vents in the ocean floor, which is a good thing.  The polymerase chain reaction (PCR) is dependent on enzymes that evolved in the latter (e.g., Thermus aquaticus and Pyrococcus furiosus).  The culture medium used in the original paper from Felisa Wolfe-Simon et al. was phosphate depleted but not phosphate-free.  No, arsenate cannot replace phosphate in bacterial DNA.  And there is a reason the water in arsenic-rich lakes is seemingly pristine and that no macrofauna live in Mono Lake, other than a species of brine shrimp and nematode plus a few other small creatures.  Attempts to introduce fish into Mono Lake were unsurprisingly unsuccessful.

The other paper from the laboratory of Rosemary J. Redfield at the University of British Columbia addressed the key question:

A strain of Halomonas bacteria, GFAJ-1, has been reported to be able to use arsenate as a nutrient when phosphate is limiting, and to specifically incorporate arsenic into its DNA in place of phosphorus. However, we have found that arsenate does not contribute to growth of GFAJ-1 when phosphate is limiting and that DNA purified from cells grown with limiting phosphate and abundant arsenate does not exhibit the spontaneous hydrolysis expected of arsenate ester bonds. Furthermore, mass spectrometry showed that this DNA contains only trace amounts of free arsenate and no detectable covalently bound arsenate. (emphasis added)

Oops.  GFAJ-1 is not a bacterium that can substitute arsenic for phosphorous, but it is able to survive in an environment rich in arsenic and low in phosphorus, provided there is just enough phosphorous available.  Despite that the results in “A Bacterium That Can Grow by Using Arsenic Instead of Phosphorous” are wrong in virtually every respect, the paper remains online, no less with the tagline “Evidence is offered for arsenate replacing phosphate as a molecular building block in a Mono Lake, California, bacterium.”  Yes, insufficient and mistaken evidence.  Is this a problem with science or scientific publishing?  Perhaps both, but in this particular case the consequences are, well, inconsequential.

What of literature connected to human health?  From a post by Patricia Murray, who is Professor of Stem Cell Biology and Regenerative Medicine at the University of Liverpool, and Peter Wilmshurst, a cardiologist who has been a scientific gadfly for years, respected by some as a founding member of COPE, the Committee of Publication Ethics, and reviled by others:

The editor of a medical journal that charges readers for access to articles whilst knowingly keeping fraudulent articles on its website is as guilty of financial fraud as an art dealer who knowingly sells forged artworks, but there is no moral equivalence. The complicity in fraud by the editor of the medical journal may also cause death and harm to patients.

That is a bold statement.  The primary article in question was published in the Lancet in 2008: Clinical transplantation of a tissue-engineered airway (paywall, sorry).  The primary authors of this paper were the surgeon Paolo Macchiarini, then in Barcelona and later at the Karolinska Institutet in Stockholm, and Martin Birchall, currently an otolaryngologist and head and neck surgeon in London.  This revolutionary surgical intervention involved the preparation of a decellularized trachea from a cadaver that was seeded with the recipient’s stem cells and thus regenerated into a functional implant that was then transplanted into the patient whose left main bronchus (the rigid airway that branches from the trachea into the left lung) had been severely narrowed by tuberculosis.

So basically, the stiff structural matrix of the cadaveric trachea was denuded of cells from the donor, which would likely have caused rejection of the transplant by the recipient, and then transformed into a rigid airway[4] made functional by the recipient’s cells.  In theory, this autologous transplant would not lead to rejection of the implant.  Unfortunately, despite a statement in a follow-up paper published in 2014 (another paywall preventing access to an old article for some reason), “These clinical results provide evidence that a tissue-engineering strategy including decellularisation of a human trachea, autologous epithelial and stem-cell culture and differentiation, and cell-scaffold seeding with a bioreactor is safe and promising (emphasis added),” the graft was unsuccessful.  

It is simply not true, as stated in the original 2008 paper, that “The graft immediately provided the recipient with a functional airway, improved her quality of life, and had a normal appearance and mechanical properties at 4 months.  The patient had no anti-donor antibodies and was not on immunosuppressive drugs.”  Instead, the graft collapsed and a stent was inserted three weeks after the transplant surgery.  Despite a perfectly good rationale for the preparation of the engineered implant, it did not workNor was it safe and promising.

Moreover, subsequent attempts at repeating the operation were unsuccessful and transplant recipients died after surgery.  Later, while he was at Karolinska, Dr. Macchiarini switched from decellularized tracheas seeded with cells from the transplant recipient onto plastic tubes coated with a suspension of the same cells.  One does not have to be a biologist to respond to that with disbelief.  The rationale for the original approach was that autologous epithelial and stem cells purified from the transplant recipient would respond to cues of their environment and form the functional cellular matrix of the rigid airway.  No plastic matrix is likely to elicit this biological response, certainly not in 2008, which was still the early days of functional stem cell biology.  And while great progress has been made in tuning stem cell environments to produce the required biological/clinical differentiated cells for a given intervention, this continues to be a problem in regenerative medicine.[5]

Moreover, the images in the 2014 follow-up paper in the Lancet have been manipulated.  This kind of apparent misconduct has been linked to the papers on the amyloid hypothesis of Alzheimer’s disease, which was addressed here in August.  But those manipulations, if they are such, came to light only after fairly sophisticated image analysis.  The image manipulation here is as obvious as it is clumsy, and to be fair to the Lancet one of the papers with Dr. Macchiarini as an author has been retracted.

The sad but infuriating case of Dr. Macchiarini has been recounted in several places, here, here, and here.  For his trouble, Peter Wilmshurst was subsequently dismissed from the Committee on Publication Ethics, ostensibly because he was not a journal editor, which he had never been even though he was present at the creation of COPE.  Is this because, in the title of his post he asks whether “COPE membership (has) become a way for unprincipled journals to buy a fake badge of integrity?”  That is borderline inflammatory, but he arguably makes his case.

For me the major similarity between the one publication describing GFAJ-1 and the other describing the apparent fiasco that has been these particular tissue-reengineered airway implants, is that in both situations there was too much enthusiasm for the novelty of the research, and this led to less attention to the data than was warranted.  Where have we seen this play out recently?  Extraordinary claims still require extraordinary evidence, whether the subject is a completely new form of life on earth or a revolutionary use of stem cells in approaches to human disease.  The GFAJ-1 paper is simply wrong based on irrefutable subsequent research.  Thus, science has been self-correcting regarding the bacterium that can use arsenic instead of phosphorous as a constituent of its DNA.  Although in the “discussion” following publication of this paper by Dr. Felisa Wolfe-Simon several of her critics recommended retraction, the paper is still online.  Still, no scientist is likely to “go there” unless compelling evidence emerges that the unlikely hypothesis is plausible.  Science in this case has been self-correcting.

Likewise, practitioners of regenerative medicine and the use of stem cells are exceedingly unlikely to follow Dr. Macchiarini’s particular path.  So, in the case of Dr. Macchiarini, who denies wrongdoing, science has also been self-correcting even if the corrections have come primarily from outside of the scientific establishment.  I would also caution that while “engineering” and “systems” are terms often used to describe work with stem cells in regenerative medicine, this borders on wishful thinking.  There remain too many unknowns for this to be called “engineering,” even when successful, with too many systems of what are essentially nonlinear differential equations involved in the differentiation of stem cells into mature tissues.  On the other hand, the Romans were engineers, and their bridges and buildings are still standing because they knew what they were doing, even if we remain astonished by their ingenuity.

Finally, I can see no reason why, on the evening of 21 November 2022, I clicked on their PubMed links and both the 2008 and 2014 papers on transplantation of a tissue-engineered airway are still there, without a notice of retraction.  The GFAJ-1 paper was simply mistaken in its conclusions.  This particular work on the engineered airway was very likely more than just wrong, however fruitful and pioneering the rationale may be.

I have recommended both Science and the Lancet in this series as two scientific journals at the very top of the heap, although a very disordered heap it is these days.  We all must remember that scientific authority is a fleeting and sometimes intangible thing.  As we have been reminded repeatedly over the past three years, extraordinary claims still require extraordinary evidence.  If the evidence is not there, science suffers, often with untoward consequences for all of us.


[1] Which some of us hope will rise as a phoenix.  But for many others in the Age of Bayh-Dole, a life in scientific research, especially biomedical research, is the route to “cash out quickly with a unicorn” (Comment at 9:53 by PlutoniumKun), the advancement of our understanding of the natural world be damned.

[2] As I have mentioned before, Science, which is published by the American Association for the Advancement of Science, is the leading scientific journal in the United States.  Any number of sources will provide the so-called “Impact Factor” of Science, but it is not clear that Eugene Garfield’s invention has been an unalloyed good for the practice of science.  Even so, any scientist who publishes just one paper in Science (or Nature) has accomplished something special.  Cell is the equivalent journal for molecular and cell biologists.

[3] Comments published in Science can be found at this link for those interested.

[4] A rigid bronchial (and tracheal) airway is essential.  Otherwise, the airway will collapse due to pressure differentials experienced during breathing.

[5] Other than in bone marrow transplants in the treatment of leukemias and other disorders of hematopoietic cells, in which the normal extracellular environment is present in the bone marrow of the transplant recipient.