6564 Genetics: Kantoff et al. 
Proc. Null. Acad. Sci. USA 83 11986) 
Repository (Camden. NJ). Control ADA-positive B-cell line 
JM was established from cells of a normal donor by 
Epstein-Barr virus infection. CEM cells, also used as an 
ADA-positive control, are a well-characterized lymphoblas- 
toid T-cell line. 
SAX Vector Transduction of ADA-Deficient Cells. Conflu- 
ent monolayers of PA-12 cells producing SAX-containing 
viral particles were changed from Dulbecco's modified Ea- 
gle's medium with 10% FBS to RPMI 1640 with 10% FBS 24 
hr before transduction. The SAX-producing cells were irra- 
diated (1500 rads; 1 rad = 0.01 Gy) just prior to transduction, 
in order to eliminate fibroblast contamination of T and B cells 
after cocultivation. ADA-deficient Tor B cells (lO'-lO 7 ) were 
cocultured with the monolayers (containing 5 x 10* SAX- 
producing PA-12 cells) for 24 hr in the presence of Polybrenc 
(2 /xg/ml). After transduction, the lymphocytes were re- 
moved from the monolayer, grown in culture for 24 hr to 
allow adherence of any residual fibroblasts, pelleted, and 
resuspended in fresh medium. These cells were then either 
grown in mass culture or. in some cases, grown in medium 
containing G418 (1.0 mg/ml) for 3 weeks, to enrich for 
SAX-containing cells, prior to analysis. 
Southern Blotting. DNA was prepared from cell lines by the 
method of Gross-Bellard et al. (22). Thirty micrograms of 
genomic DNA was digested with Sac I or EroRI restriction 
endonuclease under the conditions specified by the supplier 
(S«r I. Boehringer Mannheim: EcwRI, New England Biolabs) 
and then subjected to Southern blot analysis (23) as previ- 
ously described (11). The probe was a nick-translated 1.6- 
kilobase (kb) ////idlll-SrwiHI fragment containing the neo R 
gene isolated from pNeo (P-L Biochemicals). The copy 
number of approximately one was estimated based on band 
intensity compared with the intensity of other single-copy 
genes run in comparison. 
Assays for ADA Activity. The [ M C]adenosine assay for 
ADA activity was performed essentially as described by Van 
der Weyden and Bailey (24). The results were calculated as 
nanomoles of inosine produced per min per 10" cells. The 
starch gel electrophoretic analysis for ADA isozymes was 
performed by the method of Spencer el al. (25). 
Assay of 2'-Deoxyadenosine Resistance. As a functional 
assay for restoration of ADA activity, the inhibition of cell 
proliferation by 2'-deoxyadenosine was compared between 
ADA-deficient T (or B) cells before and after ADA gene 
transduction. Exponentially growing cells were plated in 
96-well flat-bottomed microwell plates (Costar, Cambridge. 
M A) at 50,000 cells per well in 200 /xl of RPMI 1640 with 10% 
heat-inactivated horse serum, 10% (vol/vol) interleukin 2, 
and various concentrations of 2'-deoxyadenosine (Sigma) in 
triplicate. Cells were then incubated for 24 hr at 37°C in 5% 
CO;, after which [ 5 H)thymidine incorporation over 4 hr was 
measured. Results are expressed as percent inhibition of 
[ J H]thymidine incorporation by cells in the presence of 
deoxyadenosine compared to untreated cells. The concen- 
tration of deoxyadenosine that led to 50% inhibition (IC ?0 ) 
was calculated from the dose-response curves. 
RESULTS 
Construction of the SAX Vector. The ADA gene-containing 
vector SAX was cloned in a multistep fashion. The parental 
plasmid vector N2 is a Moloney murine leukemia virus-based 
vector with the region coding for viral structural genes 
deleted and the bacterial neomycin-resistance gene (neo R ) 
inserted as a dominant selectable marker (11). A fusion gene 
was created between the simian virus 40(SV40) promoter and 
the ADA structural gene as described in the legend to Fig. 1. 
This SV40 promoter-ADA gene fusion product was inserted 
into the single Xlio I site of the N2 vector with the same 
polarity as the parental vector, so that the initiation of 
s 
...!. > 
LTR V 
P P 
| »• P P 
l» I i s 
NEO" SV40 hADA 
LTR 
1 
0 •» 
] 
1 0 
i *» ?i) ? *i 3 n :i •» 4 ii 
kb 
4 «» Ml ••!» t. • 
Fig. 1. 
Map of SAX vector. SAX was 
made by inserting a 
SV40-promoted human ADA cDNA into the previously described 
(II) parental vector N2. A fusion gene was created between the SV40 
promoter and the ADA structural gene by placing the 400-base-pair 
(bp) Kpn I-//mdlll fragment containing the enhancing and promoting 
elements of the SV40 early promoter immediately upstream of a 
1300-bp sequence containing the full-length ADA cDNA |Ec«RI-Acc 
I fragment of clone ADA 211 (26)|. The following regions are 
indicated: 0-1.5 and 4. 7-5. 5 kb. Moloney murine leukemia virus 
sequences; 1.5-2. 8 and 4. 6-4. 7 kb. neomycin-resistance gene (rteo K ) 
from Tn5 transposon (the hatched area is the coding sequence): 
2. 9-3. 3 kb. Kpn 1-W/ndlll fragment of the SV40 early promoter: 
3. 3-4. 6 kb. human ADA cDNA (hADA. black box); LTR. viral long 
terminal repeat; <b. viral packaging signal. Restriction sites: S. Sac I; 
P. Psi 1: E. £V«R1: C. Cla I. 
transcription of the ADA gene would occur in the SV40 
promoter and terminate in the 3' long terminal repeat (LTR) 
of the vector (Fig. 1). The name SAX thus stands for S (SV40 
promoter), A (human ADA gene). X (inserted into the Xlio 1 
site of N2). 
Characterization of the SAX Vector. The SAX vector was 
initially used to infect 3T3 cells. After infection and selection 
in medium containing 0.5 mg of G418 per ml. the cells were 
biochemically analyzed. By blot hybridization analysis of 
electrophoretically fractionated poly(A)* RNA, three vector- 
specific transcripts were shown to be present, two presum- 
ably initiating within the 5' LTR and the other, shorter 
transcript presumably initialing within the SV40 promoter 
(data not shown). 
Gene Transfer into T and B Cells. This SAX retroviral 
vector was used to transduce the human ADA gene into 
ADA-deficient T and B cells. The ADA-deficient T-cell line 
TJF-2 was established from a 2-ycar-old patient (JF) with 
ADA-deficient SCID by infection of peripheral blood mono- 
nuclear cells with human T-lymphotrophic virus type I 
(HTLV-I), using the procedure of Mitsuya et al. (21). Two 
control normal T-cell lines (K7 and HM) were similarly 
established by HTLV-1 infection. Southern blot analysis of 
the transduced T- and B-cell lines showed that the integrated 
proviral vector DNA is intact (Fig. 2). Since Sac I cuts once 
in each LTR (Fig. 1), an intact provirus would releasea4.9-kb 
fragment. This fragment was present in Sac I-digested DNA 
from SAX-transduced TJF-2 (lane 1) and two SAX- 
transduced B-cell lines (lanes 2 and 3) but was absent in the 
nontransduced T-cell line (lane 5). Further analysis with the 
restriction endonuclease EcoRI. which produces a 3.3-kb 
internal fragment (lanes 6-8). confirmed that at this level of 
analysis the vector DNA present in the transduced cells is 
intact. Quantitative analysis of Southern blots from a popu- 
lation of SAX-transduced TJF-2 cells revealed that they 
contain, on the average, one proviral copy per cell. The 
overall efficiency of infection ranged between 23% and 50% 
in different experiments as determined by DNA dot blot 
analysis of individual clones of T cells derived by limiting 
dilution following SAX infection (data not shown). 
Expression of the Transferred Genes in T and B Cells. Starch 
gel electrophoresis for ADA isozymes was performed on cell 
lysates to assess the expression of the introduced gene (Fig. 
3). The control B- (lane 1) and T- (lane 7) cell lines show a 
normal pattern of three human ADA isozymes. ADA-defi- 
cient B (lane 2) and T (lane 4) cells show no ADA activity. 
However, after transduction with the SAX vector, the 
nonselected uncloned population of cells shows ADA activity 
[ 246 ] 
Recombinant DNA Research, Volume 12 
