Scientists clash in battle over honesty in research


They started as partners in a quest to solve the riddle of deadly diseases, and they became enemies in a contest over wiggly lines. In the pitch of battle, there were papers pulled at the last minute from scientific journals, anonymous tips, allegations of stolen data in the halls of the School of Pharmacy and ugly confrontations at annual scientific meetings.

It might be absurd, except that now the reputations of two strong-willed scientists hang in the balance. Pitted against each other are a young researcher, Carmen M. Arroyo, and the head of her former lab at the University of Maryland at Baltimore, Gerald M. Rosen, in a case that goes to the heart of science: honesty in published research.

Dr. Rosen was cleared of misconduct by a university committee in April, and Dr. Arroyo is seeking a second opinion from the National Institutes of Health. Nine days ago, their clash boiled over in a lawsuit by Dr. Rosen that also names this reporter and The Baltimore Sun as defendants.

She is a 33-year-old chemist in the prime of her career who earned her stripes in a published work involving the heart that fellow scientists say still stands as among the earliest and best of its kind. He is a 48-year-old chemist and lawyer, chairman of pharmacology and toxicology, whose contributions include the discovery of an enzyme and its effect in cells.

They labor in an obscure field of science whose practitioners, were an international conference called today, would total no more than 130 people. Their passion is finding and trapping free radicals in living things.

Free radicals, formed from oxygen and other things, such as sunlight on the skin, toxic chemicals, and drugs, are thought to be important in the human disease process, perhaps even precursors to cancer. Ultimately, scientists hope to find how these cause disease and introduce chemical compounds to stop them.

Free radicals have a fleeting life span, but for the past 15 years, scientists have been able to detect their presence in animals and test tubes with an electron spin resonance machine. This machine produces a written record of the structure of the trapped free radical in the form of a chart of peaks and valleys.

Between the peaks and valleys on this chart, or spectra, are some wiggly lines. Noise is all they represent, random static during the experiment.

Because the noise in each experiment is different, these wiggly lines are a kind of fingerprint, most scientists say. Thus no two experiments, even when they result in the same peaks and valleys, have the same wiggly lines.

Last spring, something prompted Dr. Arroyo to inspect the wiggly lines on 13 papers published by Dr. Rosen while he was at Duke University. She grouped them two or three to a set and compared the wiggly lines. The captions under the spectra said they came from different experiments, but to her the wiggly lines looked similar.

So it was that she alleged misconduct by Dr. Rosen to the university and the federal government.

From the beginning, it was a contest of wills.

According to the lawsuit filed by Dr. Rosen last week, this case amounts to false charges stemming from a personal vendetta. In February 1990, he withdrew a paper Dr. Arroyo had prepared for FEBS Letters, a journal, because of problems he found in her research and after she had refused to show him reviewers' comments. Angered by this, the lawsuit says, Dr. Arroyo initiated a campaign to discredit him. (The university, consulting an outside expert, later ruled that Dr. Rosen had appropriately withdrawn the paper. It has since been changed and published by Dr. Arroyo in a British journal.)

Dr. Arroyo maintains her charges were motivated by the need tpreserve the integrity of science. She says it was Dr. Rosen who first became angry. This was over her challenge to his research results three months earlier, shortly after Dr. Rosen recruited her from George Washington University. It began, she says, when Dr. Rosen asked her to write up an experiment from a photocopy of the data.

Dr. Arroyo, in an interview with The Sun, said she asked him for the original data, and when this was not forthcoming, tried unsuccessfully to reproduce the experiment to get the data. This was not unusual, since scientists can't always reproduce their -- own work.

So she asked Dr. Rosen to show her what she was doing wrong. "I thought I would learn something from him," she said. Instead, Dr. Arroyo said, "he got very mad and asked for my resignation."

Later, when her paper was withdrawn by Dr. Rosen, Dr. Arroyo said, she did indeed get upset -- enough to yell and scream in the dean's office -- in part because she believed Dr. Rosen had falsely claimed that the paper had been accepted for publication in his application for a National Science Foundation grant. After further confrontations between the pair in May 1990, the dean changed the locks on some of the doors in Pharmacy Hall, limited Dr. Arroyo's access to the building's science labs and found her an office in the medical school.

On June 16, in what she said was an effort to clear her sullied name, Dr. Arroyo took the unorthodox step of writing to 52 scientists in her field to tell them she had lodged misconduct complaints against Dr. Rosen with the Office of Scientific Integrity at the National Institutes of Health, alleging plagiarism and misrepresentations in federal grant applications, poor management, carelessness and improper distribution of credit.

Shortly thereafter, she said, she received an anonymous letter, postmarked Boston, accompanied by two of Dr. Rosen's papers that the writer said contained identical wiggly lines. One of the papers already had been challenged by two sets of scientists in published journal articles.

The writer suggested that she compare the "noise" -- the wiggly lines -- in other Rosen-authored papers, figure by figure, to see if there were more examples. "I often thought of doing this," the writer added, "but had neither the time nor the motivation."

In the summer of 1990, Dr. Arroyo added a collection of Rosen papers to her complaint at the federal government's OSI office in Rockville. When university officials discovered this, they asked her to file a formal charge with them.

On Oct. 9, after consulting at least one outsider, three University of Maryland professors concluded in a preliminary investigation that the wiggly lines in four of the six sets of papers given them by Dr. Arroyo, papers written between 1980 and 1984 while Dr. Rosen was associated with Duke University Medical Center, were identical. They threw out the charges of plagiarism but called for a full investigation of the papers.

This spring, the second University of Maryland panel ruled that the charges were "unsubstantiated and without foundation." This time, four professors, including one from another university, took testimony from six people, including two outside experts. They also reviewed "substantial scientific documentation," according to a statement by the university. The university will not make public the report.

By last November, Dr. Arroyo, no longer a university employee, had been suspended from Pharmacy Hall for "continuing verbal abuse and foul language," according to a memo by David A. Knapp, acting dean of the School of Pharmacy.

Until May, when she collapsed while doing the laundry and spent several days in treatment for depression at Howard County General Hospital, she continued to be paid from a Rosen grant from the Veterans Administration under a federal law protecting whistle-blowers.

Last week, three months after first being asked by The Sun for his side of the story and on the day he filed his lawsuit, Dr. Rosen discussed his work in a two-hour interview at his lawyer's office. However, he declined to comment on any specific allegations of Dr. Arroyo or facts involving the lawsuit.

In the interview, he called his work both creative science and good science, and pointed out that only one of his findings has been criticized by other scholars in published papers, the normal vehicle for scientific debate. He said the question of how to present spectra has been debated in his field almost since it began 15 years ago.

As for the wiggly lines in his papers, he said most of them are "probably not" the same, meaning that except for some review articles where Dr. Rosen said their reuse was appropriate, they all came from different experiments. He pointed to a space where a pen appears to have skipped on the spectra in one paper to show that it must be different from its alleged match.

He also said all the questioned experiments had been conducted, and in the manner described. The findings, he said, are not in error.

But Dr. Rosen said neither he nor anyone else can prove whether the wiggly lines are the same or different, because he doesn't have the originals. (In his lawsuit, he claims that Dr. Arroyo stole the backup data, a charge she denied in a letter to the university last August.)

He conceded, however, that the captions under his figures are "incomplete." Had he added the word "representative" to describe the spectra, he said, "we would not be having this conversation today."

Is this an instance of sloppy science? he was asked.

"No," Dr. Rosen replied. "And I'll tell you why. Science is actually doing the experiments. I think my science is exceptionally creative." In his view, he said, spectra are only illustrations intended to draw the reader to the text where the real data are presented.

Yesterday, Dr. Rosen contacted The Sun to say that he now has definitive proof that the spectra are different. He said the measurements of "the ratio to peak heights" in the spectra have shown each one to be unique.

In the interview last week, he likened the spectra to apples from a Macintosh tree. No matter which branch the apple comes from, it has the same genetic code. So if the answer to an experiment is a Macintosh apple, it doesn't matter which apple you use to illustrate it.

"I think the issue in these papers . . . is how one represents spectra. Does one go ahead and use the same spectra 10 million times in experiments on this apple tree because the Macintosh is always the same, or do you always go ahead and do it, and I don't know what the answer is."

How would scientists be sure he conducted the experiments if they spotted old data?

Dr. Rosen said they should trust him. "That's what science is all about," he said. "Science is trust."

To some scientists, which branch the apple comes from makes a difference. One of them is Garry R. Buettner. He knows neither Carmen Arroyo nor Gerald Rosen but for casual chats at professional meetings, and once, through a mutual colleague, his laboratory and Dr. Rosen's laboratory shared some second-hand equipment. But he knows wiggly lines.

In his job as director of the Electron Spin Resonance Center at the University of Iowa College of Medicine, it is not uncommon for him to review up to 75 spectra and their wiggly lines each day. It was he who devised a data base to catalog spectra from different experiments that has since been computerized and expanded by NIH for scholarly use.

At the request of The Sun, Dr. Buettner reviewed the same four sets of Rosen papers investigated by the University of Maryland, spending 20 hours examining the papers and consulting the literature. In a 10-page paper received by the newspaper last week, he concluded that the spectra in each set are identical.

In two sets, this is not a problem, he said, since the experiments were also the same or because the data was labeled accordingly. In two others, he found the similarity troubling.

For instance, there is the set of three papers including the spectra with the pen skip. Not only are they identical, Dr. Buettner said, but they came from the same scan of the same experiment. In the course of researching this set, Dr. Buettner uncovered a fourth Rosen paper with wiggly lines that matched. In each case, he said, the papers describe a different experiment as the source of the spectra.

In contrast to Dr. Rosen, Dr. Buettner regards spectra as actual data and more important than the commentary accompanying it because, he said, interpretations can be wrong or modified as science moves along and new facts emerge. "Thus the key aspect of a paper is the careful presentation of the data and the methods used to collect that data," Dr. Buettner said.

He said the use of identical data to represent the results of different experiments, even when the same chemistry is observed and the scientist is reporting what happened, "is not the usual practice of scientists."

(Asked whether it is common to publish old results for neexperiments, William A. Catterall, chairman of pharmacology at the University of Washington in Seattle, said no. "That is not a common practice and would be considered unacceptable if detected by the reviewers," said Dr. Catterall, who edited Molecular Pharmacology for five years, though not when the Rosen paper was published.)

But of all the Rosen papers, only one appears to be at odds with scientific literature, Dr. Buettner said. This is an article published 1984 in Molecular Pharmacology in which Dr. Rosen and other scientists proposed a mechanism by which acetaminophen, or Tylenol, can cause damage to the liver and kidneys.

In this case, Dr. Buettner said, the fact that three independent research groups have not been able to verify Dr. Rosen's work -- one group has proposed a different structure entirely -- raises the possibility that an experimental error was made or that the experiment was not done as described. (Dr. Rosen said the experiment was done as described and is not in error. He said a paper soon to be published by Japanese researchers will support his theory.)

On Monday, Dr. Arroyo started a new job at the U.S. Medical Research Institute of Chemical Defense at Aberdeen Proving Ground.

The Rosen papers, prepared with federal grants worth at least several hundred thousand dollars, and the results of the investigation were forwarded to NIH by the University of Maryland for review in April. Since then, Dr. Arroyo has asked NIH to look at more of Dr. Rosen's published work.

In the informal networks of science, the combatants continue to defend their honor.

At a May 14 meeting of the Oxygen Society in Washington, Dr. Arroyo said, she and her husband were confronted by a Johns Hopkins scientist who once worked with Dr. Rosen and who defended his work. "This is not about friendship," she said she and her husband told the scientist; "this is about truth in science."

According to Dr. Rosen's lawsuit, Dr. Arroyo told the Hopkins scientist that Dr. Rosen had committed scientific fraud, even though she knew he had just been cleared by the university. (Dr. Arroyo maintains the university didn't brief her until a week after the gathering.) The next day, the Rosen lawsuit claims, she mailed the scientist a copy of the by then-overruled university findings in the case.

Dr. Arroyo says the truth will come from an NIH review.

Dr. Rosen says the truth will be told in a court of law.

Copyright © 2020, The Baltimore Sun, a Baltimore Sun Media Group publication | Place an Ad