Molecular Genetics of Development in Drosophila 
encoded proteins, can specify many unique body 
segments. One answer lies in the fact that these 
genes are expressed in different but overlapping 
sets of segments, leading to different combina- 
tions of homeotic proteins in the various seg- 
ments. For example, of the three homeotic pro- 
teins from the BX-C, only Ubx proiein is detected 
in the second and third thoracic segments; both 
Ubx and abd-A proteins are present in the sec- 
ond-through-fourth abdominal segments; and all 
three BX-C proteins are present in the fifth- 
through-eighth abdominal segments. These three 
combinations would of course define only three 
segmental identities if all cells within segments 
were expressing the same combination of the ho- 
meotic genes. 
In fact, cells within a segment do not express 
the same combination. We find, for example, that 
some cells within the fifth abdominal segment 
express abd-A protein exclusively or predomi- 
nantly; other cells express Ubx protein; and yet 
other cells, Abd-B protein. In the sixth abdomi- 
nal segment, Ubx and abd-A expression appears 
similar to that of the fifth abdominal segment, but 
Abd-B expression is different from that of the ad- 
jacent segment. This kind of "mosaic" expres- 
sion can, in theory, specify an unlimited number 
of segmental identities. We have therefore asked 
the question. Is the mosaic expression necessary 
for correct specification of segment identities? 
To answer this, we constructed genetically al- 
tered flies in which we could express Abd-B pro- 
tein at will. When Abd-B protein is expressed in 
all cells, abolishing the normal mosaic pattern of 
Abd-B expression, thoracic and abdominal seg- 
ments exhibit characteristics of the eighth abdom- 
inal segment. Surprisingly, we do not observe 
completely nonsensical identities, as might have 
been expected from this abnormal situation. One 
reason for this result could be that the eighth seg- 
ment identity requires only Abd-B protein (al- 
though Ubx and abd-A proteins are present at 
low levels) and that the high level of Abd-B pro- 
tein in other parts of the body is sufficient to sup- 
press the effects of Ubx and abd-A, leading to a 
repetition of the eighth abdominal segment iden- 
tities. This suppression of Ubx and abd-A activi- 
ties could be due to repression of Ubx and abd-A 
transcription by Abd-B, or alternatively, the Abd- 
B protein could simply override the effect of Ubx 
or abd-A protein in the same cell. 
To distinguish these possibilities, we asked 
what segmental identity would result if we engi- 
neered the fly so that both Ubx and Abd-B pro- 
teins are expressed in all cells. Our preliminary 
results indicate that Ubx protein exerts its effect 
over Abd-B protein in anterior body segments, 
while in posterior segments the reverse applies: 
the Abd-B protein predominates. Since the level 
of Ubx or Abd-B proteins should be the same in 
all cells, these results suggest that the response 
depends on where the cell is located. This in turn 
suggests that there may be other factors in addi- 
tion to the homeotic genes that contribute to the 
correct identities of body segments. 
380 
