examine the significance of epinephrine as an adre- 
nal hormone and the function of adrenergic neurons 
in the central nervous system. 
The DBH gene promoter has been tested with 
genes encoding a variety of interesting molecules to 
ascertain whether and how the promoter-gene com- 
bination might influence development of the sympa- 
thetic nervous system. Experiments are under way 
to test the expression of growth factors (e.g., nerve 
growth factor, NGF; fibroblast growth factor, FGF; 
and stem cell factor, SCF), oncogenes (e.g., myc, 
ras, and/o5), and proteases (e.g., plasminogen acti- 
vator and stromelysin) . 
For example, the promoter and NGF resulted in 
sympathetic ganglia that were much larger than nor- 
mal and had more neurons than controls. The sympa- 
thetic nerve fibers were also larger than normal. 
They generally appeared to take normal routes to 
peripheral target tissues, although some nerves 
grew inappropriately into the adrenal gland, vagus 
nerve, and dorsal spinal roots. Despite the increased 
number of neurons reaching peripheral tissues, in- 
nervation within the tissue was decreased, as judged 
by catecholamine histofluorescence, neurofilament 
immunoreactivity, or catecholamine uptake. These 
observations suggest that NGF gradients are not re- 
quired to guide sympathetic neurons to targets but 
are required for establishment of the normal density 
of innervation within the target tissue. 
Another experimental goal is to change geneti- 
cally the neurotransmitters that sympathetic neu- 
rons make. In the first experiment of this type, the 
DBH promoter was fused to the structural gene for 
PNMT with the aim of converting noradrenergic neu- 
rons to adrenergic neurons. Experimentally the strat- 
egy worked well. Sympathetic neurons had more 
PNMT activity than is normally present in the adre- 
nal gland, and epinephrine levels were high in sym- 
pathetic ganglia and their peripheral targets. How- 
ever, the elevated levels of epinephrine were not 
accompanied by a significant decrease in norepi- 
nephrine. Thus this experiment resulted in norad- 
renergic neurons that also make epinephrine, rather 
than a complete conversion. Considering that the 
neuronal PNMT activity was very high, the results 
suggest that this cytoplasmic enzyme may not have 
the opportunity to metabolize the norepinephrine 
that is synthesized by DBH within secretory 
granules. 
Sympathetic neurons are known to change the 
neurotransmitters they make in response to certain 
environmental cues. For example, the neurons that 
innervate the sweat glands in the toe pads of rodents 
initially use catecholamines as their neurotransmit- 
ter, but shortly after their nerve terminals reach the 
sweat glands, they switch and begin to synthesize 
acetylcholine instead. This type of switching can be 
mimicked in culture by a variety of molecules, in- 
cluding cholinergic differentiation factor (CDF, 
also known as leukocyte inhibitory factor) and cili- 
ary neurotrophic factor. Thus, in one experiment, 
the DBH promoter was used to direct CDF expres- 
sion to all sympathetic neurons. However, all the 
transgenic mice died at birth, perhaps indicating 
that massive neuronal switching is not compatible 
with maintenance of vital functions. 
In the next study, using the insulin gene pro- 
moter, CDF expression was directed to pancreatic 
islets, a normal target of sympathetic neurons, to 
ascertain whether these neurons would switch trans- 
mitters in vivo. In most experiments the mice were 
crossed with others expressing NGF from the insulin 
promoter. NGF expression results in massive sympa- 
thetic innervation of the pancreas, thereby facilitat- 
ing the analysis. 
Measurement of tyrosine hydroxylase, the rate- 
limiting enzyme in catecholamine biosynthesis, and 
of catecholamine levels in pancreas indicated that 
the neuronal content of both was greatly decreased. 
Furthermore, choline acetyltransferase activity, the 
rate-limiting enzyme in acetylcholine biosynthesis, 
was greatly increased, indicating that switching had 
occurred. These experiments are being pursued to 
determine when switching occurs and whether 
other molecules will function similarly. 
Metallothionein Gene Regulation 
The most important observation in the area of me- 
tallothionein (MT) gene regulation is the discovery 
of a new member of the MT gene family. A Japanese 
group recently discovered a small protein that they 
called growth inhibitory factor (GIF), which is defi- 
cient in people with Alzheimer's disease. They 
showed that brain extracts from these people will 
support neuronal survival in culture better than 
control extracts and that GIF inhibits this activity. 
They purified the protein and published its amino 
acid sequence. 
GIF is remarkably similar to MT, except that it is 
seven amino acids larger, having two insertions. The 
human and mouse genes encoding this protein were 
cloned and shown to resemble the other known 
mammalian MT genes in intron/exon structure. Fur- 
thermore, these genes, now called MT-III, are 
closely linked to the other MT genes on human 
chromosome 16 and mouse chromosome 8. The 
genes appear to be expressed exclusively in glial 
cells (astrocytes) in the brain. Transgenic mouse ex- 
GENETICS 249 
