combinations of mutations, how these genes inter- 
act to form a developmental network for specifying 
positional information. 
To date, Dr. Capecchi and his colleagues have de- 
scribed the phenotype associated with inactivating 
two closely linked ^o:xr genes, box-1.5and hox-1.6, 
and have initiated analysis of mice with six addi- 
tional ^oxgene mutations. Interestingly, the pheno- 
types of box- 1 . 5' /box- 1 . 5~ and box- 1 . 6~ /box- 1 . 6~ 
mice are completely different and nonoverlapping. 
Whereas box-1.5~ homozygotes exhibit defects in 
muscle, cartilage, and bone in a narrow region of the 
mouse, box-1.6~/box-1.6~ mice exhibit primarily 
neuronal deficiencies in the hindbrain and cranial 
nerves. The box- 1 . 5~ / box- 1 . 5~ mice also exhibit 
defects of the heart, major heart arteries, and cranio- 
facial structures. This complex set of deficiencies is 
remarkably similar to the pathology of humans with 
the congenital disorder DiGeorge's syndrome. Early 
death of box- 1 .6~ I box- 1 .6~ mice may be attribut- 
able to the hindbrain defects, whereas box-l.5~ / 
box-1.5~ mice appear to die from cardiovascular 
failure. 
iwf-related Genes 
In addition to disrupting genes participating in a 
transcriptional developmental program, the labora- 
tory is also focusing on genes that mediate localized 
developmental decisions through cell-cell signal- 
ing. The m?-related genes are excellent candidates 
for this second class. These proto-oncogenes were 
first identified in virally induced mammary carci- 
nomas. Expression studies, however, suggested that 
their normal role is in development. The m?-/ gene, 
now designated wnt-1, is a homologue of the Dro- 
sopbila wingless gene required to establish the pos- 
terior compartment of each parasegment. Clonal 
analysis of wingless mutations indicate that they are 
not cell autonomous, which is consistent with the 
wingless protein being made by one cell type, being 
secreted, and affecting the fate of neighboring cells. 
Previously Dr. Capecchi and his colleagues 
showed that disruption of wnt-1 in the mouse leads 
to a range of phenotypes from death at birth to sur- 
vival with severe ataxia, as a result of a cerebellar 
deficiency. Histological analysis revealed severe ab- 
normalities in the development of the midbrain and 
cerebellum. The range in phenotype was puzzling 
and provided concern as to whether the targeted 
mutation could be "leaky." More recently the labo- 
ratory identified a separate mutant allele of wnt-1, 
swaying, which could be shown by sequence analy- 
sis to be a null mutation. Its range of expressivity 
was the same as that of the original targeted muta- 
tion. Comparative analysis of the two mutants dem- 
onstrated that the primary cause of the wnt- 1 null 
mutation is the lack of formation of the anterior lobe 
of the cerebellum. However, with variable expres- 
sivity, the effects of the mutation can spread anteri- 
orly into the midbrain and posteriorly to encompass 
the posterior lobe of the cerebellum. It is hypothe- 
sized that a ^^;/2^i-related gene can complement, 
with variable expressivity, the function of wnt-1 in 
the midbrain and posterior lobe of the cerebellum. 
Phenotypic analysis of int-2~ /int-Z~ mice has just 
been completed. This proto-oncogene is a member 
of the fibroblast groviT;h factor family of genes. The 
expression pattern of int-2 is complex, but the de- 
fects associated with disruption of this gene are re- 
stricted to formation of the tail and inner ear. The 
defect in tail formation shows 1 00% penetrance, but 
the inner ear defect shows variable penetrance and 
variable expressivity. Both the vestibule and the co- 
chlea are malformed. In some mice, however, one 
inner ear can be completely malformed, whereas 
the other appears normal. Classically, variability in 
expressivity was attributed to either leakiness of the 
mutation or variability in the genetic background 
of the affected individuals. Here the laboratory 
showed that the variability in expressivity in the 
int- 2~ / int- 2" mice cannot be attributed to either of 
the above reasons. Rather, Dr. Capecchi and his col- 
leagues postulate that the developmental program, 
among individuals with identical genetic back- 
grounds, must be interpreted differently from one 
individual to another. 
Dr. Capeccbi is also Professor of Human Genet- 
ics at tbe University of Utab Scbool of Medicine 
and Professor of Biology at tbe University of Utab. 
Articles 
Chisaka, O., Musci, T.S., and Capecchi, M.R. 
1992. Developmental defects of the ear, cranial 
nerves and hindbrain resulting from targeted 
disruption of the mouse homeobox gene Hox- 
1.6. Nature 355:516-520. 
Deng, C, and Capecchi, M.R. 1992. Reexamina- 
tion of gene targeting frequency as a function of 
the extent of homology between the targeting 
vector and the target locus. Mol Cell Biol 
12:3365-3371. 
Thomas, K.R., Deng, C, and Capecchi, M.R. 
1992. High-fidelity gene targeting in embryonic 
stem cells by using sequence replacement vec- 
tors. Mol Cell Biol 12:2919-2923. 
Thomas, K.R., Musci, T.S., Neumann, P.E., and Ca- 
pecchi, M.R. 1991. Swaying is a mutant allele of 
the proto-oncogene Wnt-\. Cell 67:969-976. 
GENETICS 163 
