Genetic Defects in the Metabolic Pathways Interconnecting the Urea and 
Tricarboxylic Acid Cycles 
tion of the urea cycle-related enzymes, we re- 
cently have embarked on a project to clone the 
genes for additional enzymes that are metaboli- 
cally related to OAT. For several of these, purified 
protein or sequence data are not available; there- 
fore we are cloning these genes by functional 
complementation of mutant variants of Saccha- 
romyces cerevisiae. To date we have cloned a 
full-length cDNAfrom humanpyrroline-5-carbox- 
ylate reductase by this method. Preliminary 
studies support the feasibility of using the same 
strategy to clone pyrroline-5-carboxylate dehy- 
drogenase. These two enzymes catalyze reactions 
involving the product of the OAT reaction, and 
the latter is the site of the primary defect in the 
human disorder, type II hyperprolinemia. We an- 
ticipate that the genes for these enzymes will be 
subject to transcriptional regulation that interacts 
with that of OAT. 
We are continuing an extensive molecular anal- 
ysis of the OAT mutations that cause GA in the 
probands of 79 families from around the world. 
The functional consequences of the detected mu- 
tations are tested by expressing the mutant allele 
in a Chinese hamster cell line that lacks endoge- 
nous OAT protein or mRNA. Of the 27 alleles we 
have detected, 1 5 have been confirmed as delete- 
rious, 10 are presumed to be deleterious, 1 is 
neutral, and 1 is synonymous. In addition to pro- 
viding a means for accurate molecular diagnosis 
of GA, we anticipate that our examination of 
these defects, particularly the 18 missense muta- 
tions, will enhance our understanding of OAT 
structure and function. We are exploring strate- 
gies to express OAT in quantities sufficient for 
crystallographic analysis of OAT structure. 
Our studies of the mutations causing GA have 
highlighted the remarkable allelic heterogeneity 
at the OAT locus. With one exception (L402P, 
the most prevalent allele in Finnish GA patients), 
each mutant allele accounts for less than 10 per- 
cent of the total. Survey of the genotype of pro- 
bands from our 79 GA pedigrees indicates that 
the 35 mutant alleles so far discovered by our- 
selves and others account for only 70 percent of 
the total mutant OAT genes in this population. 
A final OAT-related project depends on the uti- 
lization of a cluster of OAT-related sequences (all 
located on the proximal portion of the short arm 
of the X chromosome) to develop a large-scale 
genetic map of this region of the human genome. 
The genes for several diseases involving the retina 
map to this region. At this point we have cloned 
and organized two regions, each about a mega- 
base, in Xpl 1 .2-Xpl 1.3, which are separated by 
a 1-3 megabase interval. We are beginning to 
identify the active genes in these areas and will 
examine their possible role in inherited retinal 
degenerations. 
We have also begun an investigation of inborn 
errors of peroxisome biogenesis and function. 
Zellweger syndrome, a neurodevelopmental dis- 
order fatal in infancy, is the disease paradigm. 
Cells and tissue from these patients exhibit defi- 
ciency of virtually all peroxisomal enzymes. In 
collaboration with Hugo Moser, we are cloning 
the genes for integral membrane proteins unique 
to peroxisomes. The first of these, the 70-kDa 
peroxisome membrane protein (PMP), is a 
member of the multiple-drug resistance (MDR) 
superfamily of proteins, which also includes the 
CFTR protein involved in cystic fibrosis. We have 
cloned and sequenced a full-length cDNA for the 
human PMP and are in the process of determining 
if mutations in this gene are responsible for any of 
the inborn errors of peroxisomal biogenesis. 
458 
