Retinoic acid has been shown to be an anticancer agent, and a growing
literature suggests that it is the nuclear retinoic acid receptor beta
2 (RAR beta 2) that is primarily responsible for mediating this effec
t, at least in some systems, To determine whether partial inactivation
of RAR beta 2 would predispose to lung cancer in mice, we generated t
hree transgenic lines expressing antisense sequences. When killed at 1
3-3/4-18 months of age, 21/36 animals had a total of 43 pulmonary tumo
rs superficially visible upon necropsy, whereas among 23 nontransgenic
mice, only 1 had a single visible lung tumor, A twofold higher incide
nce of lung tumors was seen in homozygous vs, hemizygous antisense mic
e, The endogenous RAR beta 2 message level was reduced in transgenic l
ung tissue and further reduced in the tumors, RAR beta 4, a truncated
isoform derived from the same transcript as RAR beta 2, does not carry
the sequence identified by the antisense construct and its message wa
s not as strongly affected, Immunofluorescence studies showed that RAR
beta was virtually undetectable in the tumors, but present in normal
tissue, We conclude that RAR beta 2, but probably not RAR beta 4, play
s an important role in suppression of murine lung tumorigenesis.