vectors is achieved by Ml 3-mediated coinfection of 
the phasmid-carrying strain, first to isolate the 
pMAD/genomic-ColEl /genie plasmid cointegrate 
and then, by galactocide selection against supF, 
to delete the 5Mp/^-containing pMAD/genomic 
sequence and obtain the desired ColEl /genie se- 
quence. The existence of homology between a geno- 
mic and a genie sequence demonstrates that the 
genomic sequence encodes a transcribed region. 
Dr. Kurnit and his colleagues illustrated this tech- 
nology for the fragile X sequence, demonstrating 
that the fragile X sequence is transcribed in adult 
jejunum, in an 11 -week fetus, and in a variety of 
20-week human fetal tissues, including brain, spinal 
cord, eye, liver, kidney, and skeletal muscle. 
Isolation of few- copy sequences. Copy number is 
determined by the frequency with which a given 
insert in a supF plasmid retrieves sequences from a 
genomic library constructed in phage X. A higher 
degree of recombination is associated with a higher 
degree of repetition. This rapid methodology en- 
ables separation of few-copy, moderately repetitive, 
and highly repetitive sequences. Since hybridiza- 
tion with radiolabeled human DNA does not distin- 
guish few-copy from moderately repetitive se- 
quences, the technique is essential for rapid analysis 
of copy number in this range. The strategy has been 
used to construct a multiplex nonrepeated probe for 
the distal region of chromosome 2 1 that should be 
useful for prenatal diagnosis. 
Determination of the transcriptional activity of 
few- copy fragments in different tissues at different 
times, coupled with concomitant isolation of the 
gene. A bank of human fetal (brain, spinal cord, eye, 
kidney, and voluntary muscle) and HeLa cDNA li- 
braries was screened via recombination with few- 
copy genomic probes (see above) subcloned from 
several YACs localized to the distal region of 2 lq22, 
which is responsible for many of the phenotypic de- 
fects seen in Down syndrome. Whether transcrip- 
tion occurred in particular tissues at particular 
times of development was thereby determined. Sev- 
enty transcripts (some of which must overlap) have 
been rescued. Many of the clones were represented 
in multiple cDNA libraries. 
The high degree of transcription observed via re- 
combination is in accord with previous hybridiza- 
tion results in which 12 out of 20 few-copy se- 
quences were represented in a complex library of 
fetal brain cDNA that Dr. Kurnit's laboratory con- 
structed. These results indicate that a significant 
proportion of single-copy sequences are tran- 
scribed. The ease, generality, and rapidity of appli- 
cation combine to make the recombination-based 
assay a method of choice for coupling a genie initia- 
tive to the genomic initiative. This permits efficient 
screening for the time and tissue of transcription as 
well as isolation of the transcribed sequences. 
The laboratory's finding that a significant plural- 
ity of a large number of few-copy sequences are 
transcribed, which was not feasible by other tech- 
nologies, has several corollaries: 
1. Since a considerable proportion of the few- 
copy segment of the genome is transcribed, the 
search for genes must be efficient. This point re- 
inforces the need for a rapid assay to search for tran- 
scription in multiple gene libraries. 
2. As a necessary corollary to the complexity of 
transcription in point 1, the transcription that oc- 
curs is seldom abundant. If many different se- 
quences are transcribed, then the frequency with 
which each is present in a gene library must be low. 
3. It is incumbent to decipher the timing and tis- 
sue of transcription of a given sequence. The pres- 
ence of clones in some but not all cDNA libraries 
underscores the necessity of searching for transcrip- 
tion in different libraries. 
Dr. Kurnit is also Professor of Pediatrics and 
Human Genetics at the University of Michigan 
Medical School. 
Articles 
Kurnit, D.M. 1992. Identifying transcribed se- 
quences: the state of the art. Biotechnology 
10:36-39. 
Sherman, S.L., Takaesu, N., Freeman, S.B., Grant- 
ham, M., Phillips, C, Blackston, R.D., Jacobs, 
P.A., Cockwell, A.E., Freeman, V., Uchida, I., Mik- 
kelsen, M., Kurnit, D.M., Buraczynska, M., Keats, 
B.J.B. , and Hassold, T.J. 1 99 1 . Trisomy 2 1 : associa- 
tion between reduced recombination and nondis- 
junction. Am J Hum Genet 49:608-620. 
Stewart, G.D., Hauser, M.A., Kang, H., McCann, 
D P., Osemlak, M.M., Kurnit, D.M., and Hanz- 
lik, A.J. 1991 . Plasmids for recombination-based 
screening. Gene 106:97-101. 
GENETICS 213 
