CHROMATOGRAPHIC SEPARATION OF ADENOSINE DEAMINASE PROTEINS 
monkey endogenous ADA by fast protein 
liquid-ion-exchange chromatography. It 
allows the detection of human ADA activity 
at levels lower than endogenous ADA levels 
for several reasons. The ADA proteins from 
different species are first separated by FPLC 
under nondenaturing conditions, and the 
partially purified protein fractions can then 
be concentrated and assayed individually 
using a radioactively labeled substrate which 
has a sensitivity greater than that of starch gel 
electrophoresis and in situ staining of the 
gels. As a measure of sensitivity, this method 
has been used successfully in detection of low 
levels of human ADA equivalent to 2 units 
of enzyme activity (1 unit - 1 nm inosine 
produced/min) from peripheral blood 
mononuclear cells of cynomolgus macaque 
monkeys infected with the SAX vector (33). 
F)G. 3. Slurb gd electrophoresis of ADA activity in 
mouse 3T3 cells following the introduction of ■ human 
ADA cDNA by infection with the SAX retroviral vector. 
A whole-cell lysate from the vector-treated cells was 
fractionated on an FPLC Mono Q column to separate 
human and mouse ADA isozymes. Unfractionated 
human CEM cells and uninfected 3T3 cells were used as 
cootrois. Expression of human ADA was detected in 
combioed fractions 3 and 4 and 3 and 6 from the FPLC 
column, having been separated from the endogenous 
mouse XDA isozymes appearing in combined fractions 
17 and IS. Lanes designated Blank did not have any 
protein loaded onto the starch geL 
FlG. 4. Monkey CV-I ceUs infected with the SAX 
vector. A whole-cell lysate was fractionated oo an FPLC 
Mono Q column and the presence of human and endog- 
enous monkey ADA was analyzed by starch gel electro- 
phoresis. Unfractionated human CEM cells and unin- 
fected CV-I cells in lanes I and 2 were used as controls. 
In the CV-I cells infected with SAX, human ADA iso- 
zymes arc present primarily in fractions 3 and 4 and 3 
and 6 with the endogenous monkey ADA activity in 
fractions 17 and IS and 19 and 20. The lane d es ignated 
Blank did not have any protein loaded onto the starch 
gd and the Lane designated CV-I ♦ SAX contained a 
portion of the unfractionated cell lysate. 
In future gene-transfer studies this method 
should prove useful for the separation of 
human ADA from that of other animal spe- 
cies. 
ACKNOWLEDGMENTS 
We thank Dr. Brian Safer and Wayne Kemper for 
their valuable assistance with the FPLC in the initial 
stages of this project. Dr. Rochelle Hinchhorn, Dr. Ste- 
phen O'Brien, and Janice Manenson for technical ad- 
vice about starch gel electrophoresis, and Dr. Phil Kan- 
toff and Dr. Martin Eglins for their support and helpful 
discussions throughout 
REFERENCES 
1. Conway. E. J„ and Cooke, R_( 1939) Biochem. / 3J, 
479-492. 
2. Thompson. I- F., and Seegmiller. J. E. (1980) in 
Recombinant DNA Research, Volume 1 2 
(277] 
