Molecular Analysis of Down Syndrome 
nomic DNA fragment, then recombination me- 
diated by that homology will ensue. 
Following recombination between the plasmid 
and the bacteriophage, selection for bacterio- 
phages carrying a given plasmid with swpF will 
result in selection for bacteriophages carrying a 
cDNA that is homologous to a genomic DNA 
cloned in the plasmid. In other words, the system 
is designed to select for genomic sequences that 
are transcribed. The system is also designed to 
stand alone or to interdigitate with the genomic 
initiative as it proceeds. In the latter case, as se- 
quencing detects open reading frames, the recom- 
bination-based assay is designed to delineate the 
tissue and timing of transcription quickly and ac- 
curately and isolate the transcribed sequence. 
This methodology has worked in a model ex- 
periment and resulted in the isolation of at least 
one gene on chromosome 2 1 . However, in per- 
forming these studies, two arcane problems were 
noted. 
1. Once recombination indicating transcrip- 
tion has occurred, further recombination be- 
tween the X phage of the cDNA library and the X 
lysogen used originally to insert a single copy of 
the PI ban gene results in frequent scrambling of 
the X phage, carrying the cDNA into an unrecog- 
nizable state. This can be circumvented by clon- 
ing a copy of the V\ ban gene in another way than 
as a X lysogen. The final successful strategy 
(which we accomplished recently) was to con- 
struct a bacterial strain (DK43) carrying a low- 
copy PI lysogen with a functioning ban gene and 
the PI restriction system inactivated insertion- 
ally. This strain can now be used to isolate genes 
on chromosome 21. 
2. The bulk of cDNA (gene) libraries, made in 
other laboratories for other purposes, are contam- 
inated with small amounts of ubiquitous plasmid 
pBR322 DNA sequences. These sequences pre- 
vent the screening of such libraries with genomic 
clones inserted in ColEl (pBR322)-based plas- 
mids by recombination. To circumvent this prob- 
lem, we are cloning the sup¥ gene into an R6K- 
derived plasmid that , lacks homology with 
pBR322. Since the R6K replicon is used rarely 
and need not be present in the strains used for 
propagation of Xgtl 1 libraries, contamination of 
cDNA libraries with this plasmid will not occur. 
This will allow the screening of a wide variety of 
extant cDNA libraries for transcription by chro- 
mosome-specific elements. 
256 
