Location of the Second Gene Required for Expression of the Leukemia-Associated Mouse

Antigen G$_{\text{IX}}$
Author(s): Elisabeth Stockert, Hidetoshi Sato, Katsuaki Itakura, Edward A. Boyse, Lloyd J.
Old and John J. Hutton
Source: Science, New Series, Vol. 178, No. 4063 (Nov. 24, 1972), pp. 862-863
Published by: American Association for the Advancement of Science
Stable URL: http://www.jstor.org/stable/1734916
Accessed: 30-12-2016 06:44 UTC
REFERENCES
Linked references are available on JSTOR for this article:
http://www.jstor.org/stable/1734916?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

American Association for the Advancement of Science is collaborating with JSTOR to digitize,
preserve and extend access to Science

This content downloaded from 202.3.77.206 on Fri, 30 Dec 2016 06:44:34 UTC
All use subject to http://about.jstor.org/terms

thymocytes of other (GIX ) strains

ance of virus ln vivo, and the ability
of the cells Ito be induced by IdU are
functions of the same genetic locus.
Thus, it can be concluded that one
of the virus-inducing loci of AKR mice

is on linkage group 6, 12 nzap units

from Gpi-l, with the gene order cG pi-l -Vl, and that this locus deter-

This appfoach is complicated by the

designation of this locus as Akv-l (AKR

virus-inducLng locus-1).
The finding of relatively close linkage betweeln Akv-l and Gpi-l is important in several respects. First, it
provides direct proof that the virllsinducing factor being studied in these

croisses is a chromosomal locus. Second,

testing for Gpi-l linkage provides a
convenient and rapid means to test
ior allelism (or identity) between Akv-l
and other, still unmapped loci involved
in expression of MuLV virus and anti-

gens (14). Third, since Gpi-l is ex-

deHarven, G. Geering, Virology 34, 617

(1968); J. W. Hartleys W. P Rcywe, W. I.
Capps, R. J. Huebner, l. Virol. 3, 126
(1969); R. J. Huebner, G. J. Todaro, P. S.
Sarma, J. W. Hartley, A. E. Freeman, R.
L. Petersfi C. li. Whitmire, H. Meier, R. Y,
Gilden, in Second Internatzonat Sym po*ium
on Tumor Viruses 1969 (Editions du Centre

somal site containing viral genestic material (V2) 1(9). However, if a similar
linked marker -can be found for this
locu.s, the genetic approach tc) the oncogene hypothesis should be feasible.
WALLACE P. ROWE

mines both the spontaneous and IdU

induction rates. We propose the formal

5. R. C. Nowinski, L. J. Old, E. A. Boyse, E.

existence of at least one other chromo-

3ANET W. HARTLEY
Laboratory oJ Xf iral Diseases,
National Institllte of Allergy and
Infectious Diseases
Bethesdaf Maryland 20014

THEODORE BREMNE:R

National de la Recherche Scientifique, Paris

1970), p. 33.
6. R. J. Huebner and G. J. Todaro, Proc. Nat.
Acad. Sci. U.S.A. 64, 1087 (I969).
7. S. A. Aaronson, G. J. Todaro, E. M. Scolnick, Science 174, 157 (1971).
8. J. W. Hartley and W. P. Rowe, unpubllshed
data.
9e W P Rowe, J. Exp. Med., in press.
10. -- - d J. W. Hartley, ibid., in press.
11. R. J. Delorenzo and E. H. Ruddle, Biochemo
Genet. 3, 151 (1969).
12. J. J. Hutton and T. H. Roderick, #wid. 4

Departnent of Botany, Howard

University, Washington,! D.C. 2100l;)
References and Notes

339 (1970).
13. W. P. Rowe, W. E. Pugh, J. W. HartIet,
Verology 42, 1136 (1970).
4. B. A. Taylor, H. Meier, D. D. Myerss Proc.
Nat. Acad. Sci. U.S.A. 68, 3190 (1971); E
Stockqrt, L. J. Old, E. A. Boyse, J. Exp.
Med. 133, 1334 (1971); J. R. Stephenson and

1. J. Furth, H. R. Seibold, R. R. Rathbone,

Amer. J. Cancer 9, 521 (1933).
2. W. P. Rowe and T. Pmcus, J. Exp. Mede
135, 429 (I972).
3. W. P. Rowe, J. W. Hartley, M. R. lLander,
W. li. Pugh, N. Texch, Virology 4hi, 866
(1971).
4. D. R. Lowy, W. P. Rowe, N. Teichs J. W.
Hartley, 5cience 174, 155 (1971).

S. A. Aaronson, Proc. Nat. Acad. Sci. U.S.A.,

in press.
15. We ate indebted to J. B. Hunzphrey for
management of the animal breeding. Sub
ported in part by the SpecIaI Virus Cancer
Program of NCI.
l

September

1972

"

pressed in tissue culture cells, iit can be

used as a marker for following the
Akv-l locus in somatic hybridization
studies with tlssue culture cells; questions such a-s whether this type of locus
is the integration site for a superinfect-

Location of the Second Gene Required for Expression

of the LeukensiaAssociated Mouse Antigen GIX

ing MuLV genome may be answerable

by this means. And fourth, Gpi-l test-

AAbstract. Somle mlouse strains express GIX antigen on their thymocytes; others

ing may provide a unique means of

do not. Expression depends on two genes, Gv-1 and Gv-2, in linkage groups IX

examining the most crucialS and the

and 1 respectively. Cells producing leukemia vitus, howeverJ express GIX sntigen

most diflicult to test, portion of the

regardless of their inherited -Gv-l and Gv-2 genotypee

oncogene hypothesis (6), that is, that

subinfectious expression of the inherited
MuLV genome is a major determinant

GIS is a cell surface antigen found

on the thymocytes of some (GIX+ )

of malignancy- --not only of leukemia,

mouse strains and absent from the

but of solid tumors as well. Inbred-

kemia cells, cells of the spleen, and

possibly other cell types of any mouse
strain will express Glx antigen lf they
become productively infected with mu-

mouse strains differ markedly in the in-

(1). Thymocytes are typed for GIX

rine leukemia virus (MuLV) (4), re-

cidence of various spontaneous and car-

antigen by the cytotoxicity test5 with

gardless of whether the cells originated

from a GIX mouse or a GIX mouse.

cinogen-induced tumors; the oncogene

antiserum to GIN: in the same way as

hypothesis would presum-ably predict

II-2, TL, Thy-I ( O ), and Ly alloanti

Moreover, when rats (which do not

that these diSerences are due, in large

gens, which are also found on thymo-

normally possess GIX antigen) are in-

part, to genetic differences between the

cytes (2 ) . Expression of GI2( antigen

oculated with MuLV at birth, their

integrated viral genomes in the various

on the thymocytes of normal mice is

thymocytes and ensuing leukemias become GIX + . Thus MuLV causes GIX

strains. If the integrated defective and

controlled by tz o unlinked Mendeliarl

nondefective viral genomes are at allel-

genes, Gv-1 and Gv-2 (3); at each

antigen to be expressed on cells of

ic sites in different mouse strains, trac-

locus, every mouse carries cither the

genotypes which -normally yield the

ing their transmission in segregating

positive allele for expression of GIN

crosses by means of a closely linked

antigen or ( presumably ) the alternative

GIX- phenotype, that is, on cells of

mice that lack either one or both of

genetic marker provides a way to ex-

allele for nonexpression of GIS anti

the Gv-1 + and Gv-2 + alleles. [There

amine whether inheritance of a particu-

gen. In order for a mouse to have GIN

is a provocative parallel here with the

lar viral genome is correlated with sus-

antigen on its thymocytes, it must in-

anomalous expression of TL thymocyte

ceptibility to a particular type of {umori-

herit the positive allele at both the

antigens on leukemia cells of mice

genesis. Since the expression of the

Gv-l and Gv-2 loci (which we shall

whose thymocytes are normally TL

(5).]
To recapitulate: In the absence of

marker, in this case Gpi-l, is indepen

refer to here simply as the Gv l and

dent of the viral genome, this test could

Gv-2 + alleles, as contrasted with alleles

be done even with mouse strains in

Gv-1 = and Gv-2 ^).

overt MuLV infection, GIX antigen ap-

which the viral genome is so highly

The special importance of GIx anti-

pears as a simple MendeIian character

defective that its expression i5- not de-

gen in relation to leukemia virus and

controlled by two chromosomal genes

leukemegenesis is the followiing: Letl

whereas productive MuLV infection

. tectable by any available technique.

SCIENCEt V01.. 178

862

This content downloaded from 202.3.77.206 on Fri, 30 Dec 2016 06:44:34 UTC
All use subject to http://about.jstor.org/terms

r _ * fi . sites of.int'e1gr.a-dJ.T.ExH.p AfRoede.,ri&nkp,reisbi;da.n4dJJ.3W.9

Table 1. Segregation of GV-I, GV-2, Hbb, and Gpi-l in the backcross (C57BL/6 X 129)F1 X
C57BL/6, showing linkage of Hbb with Gv-2. The results of Gpi-l typing--are not included
because there was no linkage with Gv-2, showing that the order must be Gpi-l Hbb Gv-2.
Because no strain of mouse known so far has the genotype G^-l+G-2-5 it was not possible
to avoid the use of a cross in which both genes would be segregating and therefore yielding
only 25 percent GIS+ segregants. The total number of mice typed for Glx in this series was
345, yielding 87 &IS+ 2S8 GIS- segregants (expected 86.2S: 258.75). In thc early part of
the study} only GIX segregants were typed for Gpi-1 and Hbb, hence the preponderance
phenotypes among the 158 mice in this table (85 G}x+ and 73 Glx were typed for Gpi-1 and
Hbb). For these two reasons the calculations of significance and map distance are lengthy and
complicated (copies can be had from the authors if required). The estimated distance of
Gv-2 from Hbh is 33.6 + 5 units. The- references for typing methods are as follows: for GIY

ognized in connection with loci like

GY-1 and GV-2 which may represent
viral genes: The establishment of linkage for such a locus in a particular
strain of mice carries no assurance that
the
same gene or its alleIes will occupy
of GIX+
the same site in the genomes of other
mice Tn fact the linkage of GV-1 with
H-2 (linkage group IX) established in

(13, Gpi-l (9), and Hbb (10) (LG, linkage group).

129 mice ( 1 ) and in A mice (un-

Parental types

published) has not so far been demonstrable in AKR mice; one explanation

C57BL/6 GY-1- Gv-2- HbbS Gpi-lt

129. GY-1+ GY-2+ Hbbd Gpi-la
Phenotype

-Hbb48

GIS+

among otherss is that GY-1 occupies a

Genotypes

LG IX
GV-l+/Gv ]--

G-l+J

6V_2t H^bSx

Gw-l -/

G^-2+ Hbb4/

- HbbdS

Gv-l*/

G v-2- Hbbd /

G^-l-/

G v-?- Hbbd /

Gw-1-/

Gv 2t HbbSX

- _Hbb9S

HIDETOSHI SATO

(No. )

GY-2S Hbbd/Gr-2- Hbbg

Hbb88

GIX

- ELISABETH STOCKERT

served

LG I
E

different site in AKR mice.

Ob-

>

G-l+/

Gv-2 HbbS/

G-l-J

Gv-2- Hbb/

KATSUAKI TTAKURA
3Q

55

EDWARD A. BOYSE
LLOYD J. OLD

28 Division of Immunologys
Sloan-Kettering Institute for

Cancer Research New York 10021

45 JOHN J. HUTTON

T tal 158* Division of Hematology

University of Kentacky Medical Center5

* Comprising 15 prd geny of hybrid males md 143 of hybrid females. Approximatel y equal numbersOof Lexington 405106
males and females. Segregation ratios: For Gpi-l, 75 (bb), 83 (ab) For Hbb, 74
data indi&ate 31.6+3.7 units between Gpi-l and Hbb [compare 32+5 reported in (lI>].

RefBrences and Nefes

l. E. Sto&kert, L. J. Old, E. A. Boyse, J Expr

Med. 133, 1334 (1971).

[which occurs spontaneously in asiso-

inheritance of leukemi la VlrUS (5 7) 2. E A. Boyse and L. J. Old, Annu. Rev Genet.

ciation with aging, and with leukemia

and serves other valuab xle purposes out- 3. The lo&us notation Gv (Gross virus) was

and other forms of malignancy; or is

lined by Rowe ( 8 ) .

produced experimentally by inoculation

adopted at the suggestiorl of Dr. Margaret

The Gv-l gene ha vs alreadzr been Green, Jackson Laboratory, Bar Harbor,

of MuLV-for review see (6)] causes

located in linkage gr oup IX (chro- 4. The designation MuLV is used in this report

GI)E antigen to appear reglardless of the

inherited Gv-l and Gv-2 genotypes of

mosome 17)5 36 units

this is why the antiger [} is called GIX Rauscher, and other subtypes which are

the virus-producing cells.

We report here that Gv -2 has now been and passage A virus and o&cur infrequently

GIX is probably absent from the

virus itself, and it is not known whether

located in linkage grcn zun I (chromo- or not at all as natural infe&tions of mice)

'1

in

reference

to

wild-type

and

passage

As

Gross

rom - \ J, leukemia virus (that is, ex≪ding FriendJ

serologi&ally distinguishable from wild-type

r . 5 E. A. Boyse, L. J. Old, E. Sto&kert, in RNA

some 7 ) . This was a Lscertalned ln a Viruses and Host Genome in Oncogenesis,

GP
i-1g( Holland,
lucose P. Emmelot
and P. Bentvelzens Eds. (NorthAmsterdam, I972), p. 171.

the coding gene belongs to the viral

three-point cross with

genome or to the cellular genome. ThB

phosphate isomerase) a] nd Hbb (/3 chain 6. L. J. Old and E. A. Boyse

the expression of GIX antigen on

of hemoglobin). The

thymocytes of normal mice, which is

marized in Table 1 an d give the order 7. H. Meier, D. V. Myers, Rb J. Huebner, Proc.

independent of MuLV production

Gpi-l Hbb Gv-2, with

could for example, be viewed as partial

33.6 + 5 units between t Hbb and G>-2 Med. 133, 1219 (1971); B. A. Taylor, H. Meier,

expression of a viral genome integrated

(13) (footnote to Tab] le 1). 3190 (1971); J. Hilgers, M. Beya, G. Geering,

data are sum- Lectures (Academic Press,, New York, in

ZYat. Acad. Sci. U.5.A. VJ 759 (1969), T.

a dlstance of Pincus, J. W. Hartley, W. P. Rowe, J. Exp.

D. D. Myers, Proc. Nat. Acad. Sci. U.S.A. 68,

at either the Gv-l locus or the Gv-2

Another locus recen Ltl identified in - E- A Boyse, LF J- Old, in RNA Virus and

locus. Alternatively, Gv-l and Gv-2 mazr

linkage group I is Ak sv-l ( 8 ), which P. Bentvelzen, Eds. (North-Holland, Amster-

be cellular genes with Gv-l and

Gv-2- alleIes that repress GIX antlgen

be one of two unlinked

in GIX mice; and MuLV may alter

Y Host Genome in Oncogenesis, PF Emmelot and

Rowe and his colleag ues consider to dam, 1972),

187, Res.,
F. Lilly
and W.
T. Pincus5
Advan. p.
Cancer
in press;
P. Rowes

this negative control in such a way as

{ 12'
bid.,
in
tion of MuLV in A} K1tmlce
Vl Hartley
JX 8. W. P.
Rowe,
J. press.
W. Hartley, T. Bremner,
Mice which inherit tl he Akv-l allele Science 178, 860 (1972).

to permit expression of GIX antigen.

Such possibillties have already been

carried by AKR exhibi lt hlgh levels of Genetics Ch, 393 (1969).

MuLV. production start ;ing in early life 1o. J. J. Hutton, Bbehem. Genet. 3 507 (1969).

discussed (1 ) .

9. F. H. Ruddle, T. B Shows, T. H. Roderick

and their cells are <

One useful step that can be taken

susceptible to the induct

to approach this difficult problem is to

S-iododeoxyuridine. Hc

CharaCterlStlCally (1970)-Y Aed. 133, 123i (1971).

zwever, the two 13. The authors are much ende

find out where Gv-l and Gv-2 are

lociAkv-l arldGv-2 a

situated in the cellular genome; knowl-

nor even closely associ iated for Akv-l which the estimate of map distance between

edge of their linkage relations enables

us to establish whether they are identi-

has been located on tl

cal with or closely linked to other genes

centromere Akv-l G pi-l b Bbb Gv-2. lowship to H.S.

associated with leukemogenesis or the

One special reservati on must be rec- 29 September 1972

al

Taylor,

Jackson

Laboratory,

for

having

en 1C, independently verified the calculations on

' . Hbb arld ;v-2 is based and to Miss Shelley

he opposlte slde Jambs for technicK assistanc

Or upl-l anc Hob, glV ring the order- part by NCI grant CA 08748 a
AM 160l3-Ql and by a I:>amon Runyon fel

24 NOVEMBER 1972
863

This content downloaded from 202.3.77.206 on Fri, 30 Dec 2016 06:44:34 UTC
All use subject to http://about.jstor.org/terms