Molecular Analysis of Muscle Contraction 
ily suggests that this factor, or a close relative, is a 
cardiocyte lineage-determining gene. Experi- 
ments are in progress to determine its physiologi- 
cal role, its mode of action, and its developmen- 
tal and physiological regulation. 
Regulation of Protein Diversity by 
Alternative Splicing 
To explore the mechanisms involved in the pro- 
duction of different proteins from a single gene, 
we have continued to focus on the a-tropomyosin 
(TM) gene. This gene generates a minimum of 1 0 
different isoforms that are tissue-specific and de- 
velopmentally regulated. These are produced by 
the alternative use of two different promoters and 
poly (A) addition sites, together with three pairs 
of exons that are used in a mutually exclusive 
fashion. We have focused on one of these exon 
pairs (exons 2 and 3) to elucidate the elements 
involved in this type of regulation. 
Using a combination of in vivo and cell-free 
splicing assays, we have analyzed the regulatory 
features of this system. The mutual exclusive be- 
havior is due to a competition between the two 
exons for the common flanking splicing sites. In- 
clusion of exon 3 is the default pattern and oc- 
curs in all cell types, except smooth muscle cells. 
If exon 3 is deleted, however, inclusion of exon 2 
becomes the default pattern and occurs in all cell 
types. 
The basis for this behavior is the polypyrimi- 
dine tract located at the 3' end of the intron, be- 
tween the branch site and the splice site. The role 
of the tract in splice site selection is determined 
by its ability to bind to a polypyrimidine binding 
protein (PBP) . This factor has proven to be essen- 
tial for splicing. During the past year we have 
biochemically purified and characterized it. 
In addition, we have purified and cloned a ribo- 
nucleoprotein that copurifies with PBP. Experi- 
ments are now in progress to determine the stage 
in spliceosome assembly that requires PBP as 
well as its interaction with other components of 
the splicing complex. 
The role of PBP seems to explain the default 
splicing pattern. However, since PBP is a ubiqui- 
tous splicing factor richly present in smooth mus- 
cle cells, the obvious question of how the regu- 
lated pattern is established remains unanswered. 
In vivo and in vitro experiments have clearly 
demonstrated a negative regulatory mechanism. 
Factor(s) present in smooth muscle cells block 
the default splicing pattern, whereupon exon 2 
becomes the default pathway and is included in 
the mature mRNA. The sequences involved in this 
negative regulation have been mapped to three 
different elements in and around exon 3. Each of 
these elements is required, but neither is suffi- 
cient to produce this form of regulation. Experi- 
ments are in progress to isolate and characterize 
the trans-acting factors that interact with these 
elements, using a combination of biochemical 
and genetic approaches. 
Maintenance of the Terminally 
Differentiated State 
One of the more intriguing characteristics of 
muscle cells is their terminally differentiated phe- 
notype. In the process of differentiation, these 
cells withdraw irreversibly from the cell cycle 
and are therefore unable to regenerate. Moreover, 
expression of muscle-specific genes is dependent 
on this terminally differentiated state. In its ab- 
sence, these genes are only transiently expressed 
and repressed in response to growth factor stimu- 
lation. In an attempt to understand this process, 
we have reversed the terminally differentiated 
state through the expression of several DNA tu- 
mor virus oncogenes. All these oncogenes inter- 
act with the product of the retinoblastoma gene. 
Using antibodies against the transforming pro- 
tein, Rb, p53, and muscle-specific transcription 
factors, we have determined the involvement of 
these several gene products in the production of 
the terminally differentiated state. 
Structure-Function Relationships in 
Potassium Channels 
The contractile cycle of the sarcomere is trig- 
gered by the action potential. Potassium channels 
are fundamental to the repolarization of the cell 
membrane. For this reason, we recently initiated 
an in-depth analysis of the structure of a mamma- 
lian potassium channel cloned in our laboratory. 
This channel belongs to the family of delayed rec- 
tifiers and has very low inactivation kinetics, 
making it a valuable model to study the voltage 
sensor mechanism in this molecule. 
516 
