MOLECULAR MECHANISMS UNDERLYING CHEMICAL COMMUNICATION 
AND TISSUE-SPECIFIC snRNPS 
Michael R. Lerner, M.D., Ph.D., Assistant Investigator 
Sex pheromones trigger mating behavior in many 
members of the animal kingdom. For this activity to 
occur, three criteria must be satisfied. Pheromone 
molecules must get to receptors, they must interact 
v^ith receptors, and they must be inactivated. The 
sensitive and accurate sensory system specializing 
in this process is being studied in the moth 
Manduca sexta. 
Each antenna from a male moth has —45,000 
sensilla, which are lacking in females. Wax-filled tu- 
bules that originate at pores on the chitin-covered 
surface of the sensilla serve as conduits to the inte- 
rior. Inside, the receptor lymph fluid bathes and 
protects the cilia of pheromone-sensitive receptor 
neurons. However, while the lymph effectively 
shields the delicate cilia from the air and desicca- 
tion, it simultaneously poses an aqueous barrier to 
the extremely hydrophobic pheromone molecules. 
The backbone of a typical pheromone is an un- 
branched chain of —15-16 carbon atoms, and 
those for M. sexta are no exceptions. The task of 
solubilization apparently falls on the small phero- 
mone-binding protein (PBP), which, at a concentra- 
tion of > 10 mM, is by far the most abundant pro- 
tein in the receptor lymph. The onset of synthesis 
of PBP, the final fall in pupal ecdysone levels, and 
the ability of antennae to respond electrically to 
pheromone occur at the same time. In vitro, the ex- 
pression of PBP can be blocked by exposing an an- 
tenna to ecdysone. 
Pheromone receptors on dendritic cilia serve not 
only to detect minuscule amounts of particular 
molecules but also to discriminate between closely 
related ones. After PBP has transported pheromone 
across the aqueous lymph, these specific receptors 
are activated. A 68 kDa membrane protein that may 
be a pheromone receptor has been identified in the 
moth Antheraea polyphemus. Dr. Lerner and his 
colleagues are working to clone cDNA encoding 
this protein. 
Once pheromone molecules solubilized in the 
receptor lymph have interacted with receptors, 
they must be prevented from contributing back- 
ground noise by repeatedly stimulating the recep- 
tors. All of the pheromones for M. sexta are alde- 
hydes, and an aldehyde oxidase (AOX) has been 
found in the receptor lymph. The time course of its 
expression is the same as that for the PBP As a re- 
sult of the AOX, the half-life of pheromone in sen- 
silla is estimated at 0.6 ms. 
One way to control gene expression, particularly 
in the central nervous system (CNS), is by alterna- 
tive pre-mRNA processing. The components of 
the small nuclear ribonucleoproteins (snRNPs) that 
are involved in splicing are essentially constant 
from one tissue to the next. Each snRNP particle 
is composed of a small nuclear RNA (snRNA), or in 
one case two snRNAs, complexed with at least 
six common polypeptides: B, D, D', E, F, and G. 
There are also several proteins that are specific to 
certain snRNPs but not to specific cell types. Al- 
though the mechanisms responsible for alternative 
pre-mRNA processing are not well understood, one 
scenario invokes the existence of snRNPs that could 
direct tissue-specific RNA-splicing choices. A snRNP 
fitting this description might have one or more 
components that are expressed in a tissue-specific 
manner. 
A tissue-specific snRNP polypeptide called N has 
been identified and cloned. Although expressed 
primarily in brain, its tissue and cell-type distribu- 
tion closely correlates with the ability of tissues and 
cells to make mRNA for CGRP (calcitonin gene- 
related peptide) from the calcitonin/CGRP gene. N 
is closely related to B. Conceivably snRNPs contain- 
ing N in place of B recognize distinct nucleotide se- 
quences in pre-mRNAs and aid in the selection of 
alternative splice sites. 
Another interesting property of N is that, like B, 
it bears at least one Sm epitope. Besides being the 
highly conserved epitope common to all U snRNPs, 
the presence of circulating antibody to Sm is pa- 
thognomonic for the autoimmune disease systemic 
lupus erythematosus (SLE). A specific peptide 
within the N protein has been found to contain an 
Sm epitope, as it is recognized by anti-Sm sera from 
patients with SLE. 
Dr. Lerner is also Assistant Professor of Molecular 
Neurobiology at Yale University School of Medicine. 
Continued 
517 
