NOBELIST'S DATA DRAW MORE FIRE
Author: By Judy Foreman, Globe Staff
WASHINGTON -- Dr. James B. Wyngaarden, director of the National Institutes
of Health, is expected to tell a congressional subcommittee today that the
handling of data and a controversial research paper by a team of Boston
scientists was "unorthodox."
After an initial investigation that cleared the scientists of allegations of
fraud, the NIH last Friday decided to reopen its investigation because of
information uncovered by Rep. John Dingell, a Michigan Democrat who heads a
House subcommittee on oversight and investigations.
In a statement, a copy of which was provided last night to the Globe,
Wyngaarden characterized the new information as "extremely disturbing."
The research team includes Nobel laureate David Baltimore, director of the
Whitehead Institute for Biomedical Research in Cambridge; Thereza Imanishi-
Kari, a pathologist at Tufts University School of Medicine; Moema Reis, now of
the Instituto Biologico in Sao Paulo; David Weaver of Boston's Dana Farber
Cancer Institute; and two others.
Dr. Margot O'Toole, at the time a postdoctoral researcher in Imanishi-
Kari's lab, was the first scientist to raise serious questions about the
paper, which was published three years ago in the journal Cell.
Wyngaarden is expected to testify that O'Toole, during NIH's investigation
late last year, "raised the possibility that apparently original laboratory
data had been fabricated at a later time." Initially, NIH considered this
A key component in Baltimore's defense is expected to be the distinction
between honest error and fraud.
But Wyngaarden is expected to testify that "allegations of error may lead
us to findings of falsification or misrepresentation."
According to Wyngaarden's statement, the inspector general of the Department
of Health and Human Services has been asked to help analyze forensic evidence
in this case.
Wyngaarden is expected to voice sympathy for what he calls "the fate of the
"At present, we have only limited authority to protect those who, in good
faith, raise allegations of scientific error or misconduct. . . . Clearly Dr.
O'Toole has taken considerable risk in . . . insisting that the truth be
learned. We are concerned that Dr. O'Toole's scientific career has been
damaged simply because she has pursued her conviction."
Wyngaarden says he has established two special offices to review questions
of scientific misconduct.
The scientific questions in the case revolve around genetic control of the
Whenever a person is infected by a foreign substance, the body responds by
producing proteins called antibodies that attack the foreign invader. The body
also produces immune cells that aid in its defense.
One of the most intriguing mysteries of biology is how the body produces
such a vast array of antibodies, each one different, so that it can mount a
response against all sorts of invaders, even those that have never been
The secret to this ability lies in the genes in B cells, the immune cells
that make antibodies.
Beginning in the fetus and continuing throughout life, certain antibody
genes inside B cells "reshuffle." Each time they line up differently, the
protein or antibody the genes control changes. The result is a huge diversity
of genetic rearrangements and a huge diversity of antibodies.
Antibodies are shaped like an upright letter Y. Each side of the cup of the
Y consists of two chains, a light chain and a longer, heavier chain. One gene
commands the production of the short chain, another the heavy chain.
The researchers were most interested in the gene for the heavy chain. They
took a heavy chain gene from one kind of mouse called a Balb-C and put it into
fertilized egg cells of another kind of mouse called B-6. They did not remove
any genes from the B-6 mice, which meant that the mice that eventually grew
from the fertilized eggs ended up with their own heavy chain genes plus the
inserted gene from the Balb-C mouse.
The question was, would both genes make antibodies? Would the two genes
interact? Would the new gene somehow regulate the host gene? The researchers
hoped to tell by looking at the resulting antibodies.
They used a chemical reagant to test the antibodies from the now-grown
mice. If the test worked as researchers hoped, it should have been able to
tell the difference between antibodies made by the B-6 gene and those inserted
into the B-6 mice.
They believed the reagant would pick out and bind only to antibodies from
the inserted gene.
But O'Toole ran many of those tests and kept finding that the reagant bound
to antibodies from both the inserted gene and the host gene in such a way that
one could not tell which one was which.
This was not what researchers hoped to find.
Imanishi-Kari, however, stuck to the idea that the inserted gene
influenced the host gene to make antibodies that "looked like" the inserted
genetic material. The team of authors, which did not include O'Toole, printed
their finding that the inserted gene "can activate powerful cellular
O'Toole took her concerns to MIT higher-ups. Backed by Charles Maplethorpe,
then a graduate student, she questioned how Imanishi-Kari could draw the
conclusions she did. At a congressional hearing last year, Maplethorpe
testified he believed Imanishi-Kari committed fraud.
After the NIH investigated urged the team to correct itself did the group
acknowledged in a letter to Cell in November 1988 that the test reagent was
To critics, this changes the conclusion of the paper. The team, however,
insists to this day it does not.