A PROPOSAL FOR NAMING HOST CELL DERIVED INSERTS
IN RETROVIRUS GENOMES

1 2

John M. Coffin” and Harold E. Varmus

lDepartment of Molecular Biology and Microbiology
Tufts University School of Medicine

Boston, Massachusetts 02111

“Department of Microbiology
University of California

San Francisco, California 94143
ABSTRACT

We propose a system for naming inserted sequences in transforming
retreviruses (i.e. onc genes), based on using trivial names

derived from a prototype strain of virus.
A number of retroviruses have been isolated from naturally occurring
or laboratory-induced tumors. Some of these are able to induce rapid
disease in laboratory animals and to induce transformation of morphological
and/or growth properties of appropriate tissue culture cells (for
review see ........). All such viruses whose genomes have been closely
examined have been found to share a common feature: the presence
of a nucleotide sequence which encodes a protein unnecessary for
viral replication but required for the induction of the transformed
phenotype (...c.eeeeeseeeceeee}s Such sequences have been generally
referred to as onc genes (...............). AS shown in Table 1,
there are at least twelve distinct onc genes which have been identified
in at least twenty isolates of transforming retroviruses. Where
tested, all such genes have been found to be closely related to a
sequence present in the uninfected host cell, yet distinct from any
endogenous viruses which might be present. It has been proposed
that the transforming viruses have arisen by a mechanism involving
recombination between virus and cellular information, with the consequence
that an apparently normal cellular gene has come under the replicative
and expression controls provided by the viral genome (........-e cee eee eee eeads
and by virtue of modification in structure and/or mode of expression

has acquired the ability to cause cell transformation.

While there is general agreement among workers in the field
concerning the nature of onc genes and their relationship to the ~
host cell, there is substantial confusion surrounding the names of

these sequences and their cellular relatives. For example, the name
-3-

src, originally used to designate the onc gene of Rous sarcoma virus
(cece eccccecsenccaes eve ae esses }, has recently been applied generally
to sequences which are completely unrelated in sequence, in nature
of the gene product, and in location in the genome. The use of identical
names for genes of unrelated sequence and function can lead to serious
problems in communication. An additional problem has arisen in the
description of the endogenous sequence related to an onc gene. The
sequences related to the various sre genes, for example, have been
often called "sarc", with the result that the virus and cellular
sequences have identical pronounciation. More Eoifbersome designation
for such sequences have been proposed, but not widely accepted.
Retrovirus genes encoding replicative function’ (i.e. gag, pol,
and env) been accorded three letter names derived from their function
of some other feature (.........0.- peeseeveeees ). We propose, for
simplicity and readability, that this system be extended to include
the non replicative inserts found in many strains of retrovirus.
According to this proposed system, such inserts (or onc genes) will
be given trivial three letter designations. These names are not
meant to imply specific diseases, target cells, or functions, rather
they are to be simply names of sequences which are not derived from
viral replicative information, and which encode a protein (or a portion
of a polyprotein) likely to be involved in transformation of the
infected cell. We also propose a system for distinguishing the viral
from the related cellular sequence and, where necessary, the sequences

in related viral strains from one another.
-4-

The names for these sequences are to be assigned according to

the following guidelines:

l.
2.

The names should be 3 letters, lower case italics.

The names should be trivial; that is no target cell specificity

or functional significance is implied, and they are to

be considered as names of coding sequences only.

They are to be derived in some mellifluous, yet mnemonic

way from the name of the prototype virus or viruses or

some other memorable feature of them.

Related sequences in different viruses from the same species

are to be called by the same name, in a way Y net oe should gZ
wea eH

(when completely resolved) point to the sane cell sequence

and the same or a closely related protein product, although

it should not be necessary to have identified all of these

to assign a name.

When necessary for clarity, the differences between inserts

in related viruses can be indicated by prefixing the name

with the abbreviation or name for the virus or virus strain.

The related sequence found in the cell of origin will be

designaed with a lower case c- preceding the sequence name,

e.g. c-src. The animal species of the cellular homologue

should be indicated in paenthesis following the name of

the sequence (e.g. c-src (chicken)). The unadorned name

will always indicate the viral sequence only.

Protein products will be designated according to previous

convention except that no superscripts will be used; thus,

pp6O0src, Pl50c-ab1, PllO0gag-abl stand for the product of
10.

11.

12.

~5-

src, the product of the endogenous cell sequences related

to abl, and the polyprotein containing both gag and abl
specific information, respectively.

Should the same virus be found to have two independently
expressed inserts (i.e. coding for different proteins through
distinct mRNAs), then they can be distinguished by affixing
~A, -B, etc. to the name.

Such names should be reserved for nonviral related sequences
only. Such situations as spleen focus~forming virus, which
seems to have only variants of viral replicative genes

) and the 30S region of Ha and

Ki MSV which is apparently derived from an endogenous virus —
Tike element (...... ccc ce cee cece eee ) should not be so
named. In this way, it can be assured that the names are
unique.

Names along the same lines can also be given to nontransforming
inserts if found in retroviruses or deliberately put there,
but should be limited to genetically significant regions,
i.e. those with protein (or functional RNA) product.

An exception to rule 4 can be made (although it need not)

in the case where somewhat different yet related inserts

are found in viruses of different species.

Strict genetic evidence is not required to assign a name,

but it should be shown A) that the region is non-viral,

and B) that it has either a protein (or functional RNA)

product or a genetically identifiable funtion.
-6-

A list of suggested names is shown in Table 1. We note that
many of the assignments are tentative and that more names will likely
be added in the future. Three of the names on this Vist (src, myb,
erb) are already in use. Erb and myb were originally proposed with
a different rationale; ie. that. they were indicative of. transformed
cell type (...... ccc ec c we ee eee }. We do not consider transformed
cell type to be useful criterion for such assignments, since many
of the viruses cause a variety of diseases, since at least seven
of the onc sequences are in viruses that cause sarcoma as their most
common disease, and since even in these viruses that do cause a relatively
unique definable disease (such as Abelson MuLV), there is 7 general
agreement concerning the nature of the transformed cel]. The three
names mentioned, however, should in this context be considered as
trivial names derived from the name of the prototype virus, and we
suggest they be so used. We do suggest changing the name proposed
for the transforming insert of avian myelocytomatosis virus MC29and
related viruses (MaC3....... ce see cece eeevcees ) to myc to match more
closely the name of the protoype virus.

If the name of an onc "gene" is considered to desribe a name
of inserted sequence, all or at least part of wc enpuces a functional
product, then (at least in principle) it can be precisely defined
as that sequence which is unrelated to the genome of any replication-
competent nontransforming virus (i.e. not belonging to a gag, pol,
or env gene or to some noncoding internal or terminal region of such
a virus). With many of these sequences, it is quite difficult to

obtain a definition by purely genetic techniques, since they are

usually found in replication-defective viruses. In all cases, however,
-7-

it is possible to use physical, biochemical, and recombinant DNA
techniques to define the limits of onc sequences with precision,

for example by comparing nucleotide sequences of a transforming virus,
its nontransforming but replication competent helper, and the related
cellular sequence or sequences with each other and with the amino

acid sequence of the suspected gene product. A region of a genome
defined in this way is not, in the strictest sense, a "gene". However,
to refer to a defined sequence as an onc gene, while imprecise, should
not create serious confusion, so long as it is understood that not

all of the sequence may be directly involved in encoding a product

and that additional viral sequnces may encode part of the final gene

product.

Some of the names proposed may not at first seem as mellifluous
as might be desirable. However, with practice they seem to be fairly
easy to pronounce; for example, abl can be pronounced like "able"
and fps like "fips". We also suggest that mas be pronounced "mass"
to avoid confusion with mos ("mos"). _—_—

The fotlowing investigators have agreed to these guidelines: —
S. Aaronson, P. Balduzzi, J. Ball, D. Baltimore, H. Bauer, J. M.
Bishop, D. Dina, R. Eisenman, R. Friis, D. Fujita, A. Goldberg, H.
Hanafusa, S. Hughes, W. Joklik, G.S. Martin, S. Rasheed, F. Reynolds,

N. Rosenberg, C. Sherr, J. Stephenson, H. Temin, G. Theilen, K. Toyoshima,

G. Vande Woude, I. Verma, P. Vogt, M. Weber, R. Weinberg and M. Yoshida.
Viral Insert

rel

RSV-sre
B77-sre
rASV-sre
PR-RSV-src

AMV-myb
E26-myb

MC29-myc
CMII-myc
MH2-myc

OK10-myc

AEV-erb-A
AEV-erb-B

FSV-fps
PRCIT-fps

Moloney-mos
Gazdar-mos

Rasheed-ras
Kirsten-ras
Harvey-ras
abl

ST-fes
GA-fes

MSs. Cras

WOS
CG

Y73~yés
ESV-yes

TABLE 1. PROPOSED NAMES FOR onc GENES

Virus Strain

avian reticuloendotheliosis virus-T

Rous sarcoma virus

B77 avian sarcoma virus
recovered avian sarcoma virus
Prague strain Rous sarcoma virus

avian myeloblastosis virus strain BAI-1
avian leukemia virus strain E26

avian myelocytoma virus MC29
avian myelocytoma virus CMII
avian myelocytoma and carcinoma virus MH2
avian myelocytoma virusO0K10

avian erythroblastosis virus
avian erythroblastosis virus

Fujinami sarcoma virus
PRCII sarcoma virus

Moloney murine sarcoma virus
Gazdar murine sarcoma virus

Rasheed rat sarcoma virus
Kirsten murine sarcoma virus
Harvey murine sarcoma virus
Abelson murine leukemia virus

Snyder-Theilen feline sarcoma virus
Gardner-~Arnstein feline sarcoma virus

McDonough feline sarcoma virus
Woolly monkey sarcoma virus

¥73 avian sarcoma virus
Esh sarcoma virus

Probable Animal Origin

 

turkey

chicken
chicken
chicken, Japanese quail
chicken

chicken
chicken

chicken
chicken
chicken
chicken

chicken
chicken

chicken
chicken

mouse
mouse

rat
rat
rat
mouse

cat
cat

cat
woolly monkey

chicken
chicken

Protein. Product

?

pp60src.
ppo0src
pp60src
pp60src

?
?
Ly

P1l0gag-mac
P90gag-mgc

PL0Ugagrnéc

P75gag-erb-A
p4sgag-erb-B
P140gag-fps
P105gag-fps

?
?

P29gag-ras
P2lras
P2lras
P120gag-ab1

P85gag-fes
P1l0gag-fes

P170gag-mag JI

?

P90gag- yes
PaOeae SES

ip