March 21, 1971

Dr. Peter Vogt
Department of Microbiology
University of Washington
School of Medicine
Seattle, Washington 98105

Dear Peter,

Enclosed with this letter you will find copies of papers
recently written by us and by Malcolm Martin's group at NIH.
As you will see from these preprints, both mouse and chick
virus polymerases appear to synthesize double-stranded DNA,
which is principally of low complexity - representing less
than 57. (our interpretation based on Fd RF and X standards)
or about 209. (their value based on an SV 40 standard) of the
viral 70s RNA genome. There is also at least one additional
population of double-stranded DNA molecules produced, this
population being of greater complexity (12% of the genome
in our hands, perhaps 100% in Martin's). Measuring the rate
of reassociation of the large population (RR fraction, for
"rapidly reassoclating") in the presence of unlabeled cell
DNA's, we find about 20 copies of the RR sequences in both
chick embryo DNA and in DNA extracted from SR-RSV induced
tumors. Salmon sperm DNA four viscosity control) has no
copies.  As you can see from their paper, Gelb et s.  also
find multiple copiesoof their rapidly reassociating fraction
in normal and transformed (including "non-producing") rat and
mouse cells, and in African green monkey kidney cells. We are
about th look for the sequences in HeLa cells and may look at
other animal species as well. In addition, we are trying to
determine the chromosomal location of the RR sequences by using
chick DNA extracted from chromosome populations fractionated in
sucrose on the basis of size. Because the RR sequences repre-
sent a small fraction of the viral genetic information, we are
also using single-stranded DNA product (prepared from RNA:DNA
hybrids and presumably containing all the viral sequences) as
a probe for viral information in normal and transformed cells.
Preliminary experiments in our lab and Martin's indicate that
the transformed DNA does contain more viral sequences than
normal cells.

Of course, at this point the significance of the RR copies in
normal chick cells is uncertain - we do not yet have any real

Page 2
Vogt/Varmus
March 21, 1971

idea what the gene products of the RR DNA might be.
sure you will agree that it would be of particular interest to know
whether c/o' cells, which appear to lack viral antigens and
cryptic RAV-60, are also lacking the RR eequences,  It is possible,
of course, that we are not yet using the best RR probe, since ours
is synthesized by SR-RSV polymerase; we would be interested in
your opinion about the feasibility of preparing sufficient amounts
of RAV 60 to use for RR synthesis.

But we are

We consider thebbiochemical experiments we are suggesting with
c/o' cells to be principally attempts to extent your biological
findings and Hanafusa's from another potnt of view.
pleased that you are willing to assist us in these experiments
by sending us the cells and hope we will soon be in touch about
questions or results.

We are very

Yours,

Harold E. Varmus, M.D. Warren Levinson, M.D., Ph.D.

J. Michael Bishop, M.D.
Department of Microbiology

HEV : WL : JMB/ j s
encl