Isolation of a human repetitive sequence and its application to ...

Proc. NatL Acad. Sci. USA Vol. 79, pp. 7390-7394, December 1982 Genetics

Isolation of a human repetitive sequence and its application to regional chromosome mapping (recombinant DNA/human chromosome 12/single human chromosome cell hybrid)

MARTHA LIAO LAW*, JEFFREY N. DAVIDSON*t, AND FA-TEN KAO*t *Eleanor Roosevelt Institute for Cancer Research and Departments of tMedicine and tBiochemistry, Biophysics and Genetics, University of Colorado Health Sciences Center, Denver, Colorado 80262

Communicated by Theodore T. Puck, August 4, 1982

ABSTRACT Recombinant A phage Charon 4A with repetitive human DNA inserts have been constructed by using cellular DNA from a human-Chinese hamster ovary cell hybrid retaining the complete hamster genome and a single human chromosome 12. One recombinant phage, 12-11, contains several repetitive sequences, each with a different repetition pattern in the human genome. A 2.2-kilobase (kb) EcoRI fragment ofthis phage was subcloned in pBR325. This sequence has fewer than 5,000 copies in the human genome and does not cross-hybridize with Chinese hamster DNA. When the labeled 2.2-kb probe was hybridized to human chromosome 12 DNA digested with EcoRI, there was an intense band at the 2.2-kb position and a series of other discrete bands. The band pattern at positions other than 2.2 kb appears to be distinct for each human chromosome. The 2.2-kb fragment is composed of at least three subregions. The ends of the fragment are repeated more frequently in the genome than is the middle portion. Hybridization of chromosome 12 DNA with probes made to these subregions yielded simpler band patterns. By using a series of cell hybrids containing various deletions of human chromosome 12, five sequences related to the 2.2-kb fragment have been assigned regionally to a specific portion of the short arm of chromosome 12. These results demonstrate that certain repetitive sequences in the human genome can be used as genetic markers and may permit detailed regional mapping ofhuman chromosomes.

be useful as a marker for a number of chromosomal locations. This approach has not been extensively exploited because most of the better characterized repetitive sequences have copy numbers of 104-105 or more in the human genome (10-13)-far too many to be useful as specific genetic markers. A less repetitive sequence with 103-104 copies might permit the creation of detailed regional maps for all human chromosomes if these chromosomes can by suitably isolated, for example by techniques of cell fusion. In addition, a sequence with such a moderate level of repetition might serve important functions in the human genome. In the present paper, we describe the isolation of a 2.2-kilobase (kb) fragment of human DNA from a partial A phage library constructed from a human-Chinese hamster cell hybrid containing a single human chromosome 12. This 2.2-kb fragment is repeated several thousand times in the human genome. We have further demonstrated that it can be used in conjunction with somatic cell hybrids containing different human chromosomes to generate unique DNA hybridization maps for several chromosomes and with hybrids carrying various deletions of human chromosome 12 to map five related sequences to specific regions of chromosome 12.

With the availability of techniques for cloning genes and DNA fragments into various vectors like bacteriophage or plasmids (1), it has become feasible to undertake structural analysis of eukaryotic genes and the molecular organization of mammalian genomes including that of man (2). In particular, the numerous DNA fragments derived from the human genome (3) can be used as valuable genetic markers for linkage analysis and for the eventual construction of a complete human gene map (4). DNA fragments can be obtained either directly from a specific human chromosome (5-8) or from the total human genomic library, followed by assignment to a specific human chromosome by using somatic cell hybrids and synteny analysis (9). Furthermore, regional assignment of these DNA fragments within a chromosome also can be achieved by using cell hybrids or other suitable cells containing various terminal deletions of the particular chromosome concerned. For example, five DNA fragments derived from human chromosome 11 have been assigned to three specific regions of the chromosome by using this approach (5). So far, the majority of the human chromosome mapping studies with cloned DNA fragments have concentrated on unique sequences (5, 6, 8, 9). Each unique sequence can only mark one chromosomal location. On the other hand, a cloned sequence that is related to many similar sequences in the genome might

MATERIALS AND METHODS Source of Cloning Vectors. Escherichia coli DP50supF and bacteriophage A Charon 4A were obtained from Fred Blattner, who also provided the procedures for maintenance and growth of the bacterial strains and for extraction of bacteriophage DNA. E. coli HB101 and plasmid pBR325 were provided by John R. Sadler. Cultures of Mammalian Cells and Cell Hybrids. Human cells, Chinese hamster ovary (CHO) cell line CHO-K1, and human-CHO-KL somatic cell hybrids were grown in standard culture conditions (9). The human fibrosarcoma cell line HT1080, provided by Carlo Croce (14), was used as the source of human DNA unless otherwise specified. The cell hybrids were developed and characterized as described (9). In particular, the hybrid 12A, containing a complete CHO-K1 genome plus a single human chromosome 12 (15), was used for cloning human DNA fragments into A phage Charon 4A. In addition, the following subclones derived from 12A and possessing various deletions of chromosome 12 were used for regional mapping of DNA sequences on the chromosome: 12A-1, previously designated A9 (16, 17), has a deletion in the region pter-p1205 ofthe short arm with the phenotype TPI-GAPD-LDHB-ENO2-SHMT+PEPB+; 12A-2, previously designed MA2 (16, 17), has a deletion in the region ql2-qter of the long arm with the phenotype TPI+GAPD+LDHB+ENO2+SHMT-PEPB-; 12A-3 and 12A4 are segregants from 12A that have no cytogenetically detect-

The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviations: kb, kilobase; CHO, Chinese hamster ovary.

7390

Genetics: Law et aL

able human chromosome 12 and have lost all chromosome 12 markers assayed (TPI-GAPD-LDHB-ENO2-SHMT-PEPB-). Another hybrid, 60A2, used in the study was derived from the fusion between x-irradiated 12A (15) and the glycine-requiring mutant gly-A of CHO-K1 cells (18). This hybrid has apparently lost all human chromosomes except a dot-like chromosome including the centromere region and the serine hydroxymethyltransferase (SHMT) marker of human chromosome 12 (unpublished data). Cloning of Hybrid 12A DNA into Phage Charon 4A. DNA from human-CHO-Ki hybrid 12A cells was extracted as described (9, 19). For cloning the hybrid DNA fragments into A phage Charon 4A, we used procedures similar to those previously described (5, 20-22). Identification of Recombinant Phage Clones Containing Human DNA. Methods for distinguishing clones containing human DNA from those containing hamster DNA were similar to those described by Gusellaet aL (5). Duplicate nitrocellulose filters were prepared from the phage plaques by using the Benton and Davis method (23) and were hybridized to nick-translated 32P-labeled total human HT-1080 DNA or Chinese hamster CHO-K1 DNA. The clones that hybridized to human probe but not to hamster probe were isolated and proved to contain human DNA inserts. Isolation of DNA Restriction Fragments from Recombinant Phage. Restriction fragments from a recombinant phage containing human DNA were isolated either by direct electroelu. tion ofthe fragments from agarose gels (24, 25) or by subcloning of the fragments into plasmids. Bacterial alkaline phosphatase was used to remove the phosphate group from the 5' end ofthe phage DNA prior to digestion, thus rendering the phage arms totally incapable of ligating to the plasmid DNA (26), and the resulting recombinant plasmids contained only human DNA. Restriction Endonuclease Digestion, Agarose Gel Electrophoresis, and Southern Transfer. Restriction endonuclease digestion was carried out by procedures provided by the suppliers (New England BioLabs and Miles). The digested DNA fragments were fractionated by horizontal 0.75% agarose gel (Sigma) electrophoresis and stained with ethidium bromide- (5 ,g/ml). After electrophoresis, the DNA was transferred from gels to nitrocellulose filters by the method of Southern (27), with modifications of Wahl et aL (28). DNA Labeling, Filter Hybridization, and Autoradiography. DNA was labeled with [a-32P]dCTP (New England Nuclear) by nick-translation (29). Specific activity of >108 cpm/,tg of DNA was routinely obtained. Filter hybridization was carried out by using dextran sulfate as described by Wahl et aL (28). After hybridization, the filters were dried and exposed to DuPont Cronex IV x-ray film at -70°C for 16 hr to 3 wk with a DuPont Lightning Plus intensifying screen. Estimation of Copy Number in the Human Genome. A DNA dot blot procedure similar to that of Brandsma and Miller (30) was used to estimate the number of times a particular cloned DNA sequence is represented in the human genome.

RESULTS Construction and Isolation of Charon 4A Recombinant Phage Containing DNA Fragments from a Somatic Cell Hybrid. EcoRI fragments of 15- to 20-kb length, purified from the total DNA of the cell hybrid 12A containing a single human chromosome 12, were ligated to Charon 4A phage A arms, and the concatemeric molecules were packaged in vitro to produce viable phage. In the first series of experiments, 1.4 x 104 plaque-forming units were obtained, and >95% of the phage were shown to contain recombinant. DNA derived from either hamster or human.

Proc. Nati Acad. Sci. USA 79 (1982)

7391

Identification of Recombinant Phage Containing Only Human DNA. Several hundred recombinant phage were screened for the presence of human DNA inserts by hybridizing to labeled total human DNA. Under the conditions used (5), human middle repetitive sequences do not cross-hybridize with hamster sequences. Fourteen recombinants containing only human DNA were identified. One recombinant phage, designated 1211, was chosen for detailed study. Characterization of Recombinant Phage 12-11 Containing Human DNA. The DNA from phage 12-11 was digested with EcoRI, electrophoresed on a 0. 75% agarose gel, transferred to a nitrocellulose filter, and hybridized to 32P-labeled total human HT-1080 DNA. In Fig. 1, the digested phage DNA fragments stained with ethidium bromide (lane 1) are compared with the fragments that hybridized to the middle repetitive probe (lane 2). The human DNA insert in phage 12-11 consists of five EcoRI fragments 5.6, 4.0, 2.2, 1.0, and 0.65 kb in size. The 5.6-kb fragment hybridized strongly with total human DNA (Fig. 1, lane 2), indicating that it is frequently repeated in the human genome. The 1.0-kb fragment, which showed less hybridization, is likely to be a sequence less repetitive than the 5.6-kb fragment. The 4.0- and 2.2-kb fragments showed even less hybridization and probably represent low copy number sequences. In order to facilitate further analysis, the EcoRI fragments from phage 12-11 were subcloned into pBR325. In particular, the plasmid containing the 2.2-kb fragment was chosen for further study. Hybridization of the 2.2-kb Fragment to Digested DNA from CHO and Human Cells. The 2.2-kb fragment from the plasmid containing the 2.2-kb fragment-was electroeluted, nicktranslated with [32P]dCTP to high specific activity, and hybridized to Southern blots of digested DNA from various cell types (Fig. 2). There was absolutely no hybridization with the CHO, DNA (Fig. 2, lane 1), indicating that the 2.2-kb sequence is human specific. When the probe was hybridized to total human DNA, there was an intense hybridization band at the 2.2-kb position, with a continuum of bands above that position and 30.719.81095.64.02.21.0-

.65-

1 2 FIG. 1. Ethidium bromide-stained gel showing EcoRI-digested phage 12-11 DNA separated by agarose gel electrophoresis (lane 1). The fragments were transferred to nitrocellulose filter and hybridized to labeled total human DNA (autoradiogram, lane 2). In lane 1, the three dark bands on the top are the A phage Charon 4A arms (19.8 and 10.9 kb and thejoined 30.7-kb form). The two weaker bands below are possibly human DNA notcompletely digested from the phage. The five bands of 5.6,4.0,2.2, 1.0, and 0.65 kb in the lower section are from the human insert.

7392

Genetics: Law et aL

2 3 4 5 6 7 8 FIG. 2. Autoradiogramsshowinghybridization patterns of the 2.2kb probe hybridized to DNA from various cellular sources. Lanes: 1, CHO-K1; 2, human HT-1080; 3, 12A; 4, 12A-1; 5, 12A-2; 6, 12A-4; 7, 60A2; 8, 12A-3.

several discrete bands below it (Fig. 2, lane 2). To obtain an estimate of the number of copies of sequences related to the 2.2-kb fragment in the human genome, a dot blot analysis similar to that of Brandsma and Miller (30) was performed. The results indicated that there are between 2,000 and 5,000 sequences related to the 2.2-kb probe in the human genome. This result may underestimate the actual number of related sequences because less closely related sequences hybridize the probe less efficiently. Hybridization of the 2.2-kb Probe to Digested DNA from a Series of Cell Hybrids Containing the Entire Single Human Chromosome 12 or Deletion Mutants Thereof. When the 2.2kb probe was hybridized to cell hybrid 12A containing a single human chromosome 12, an intense band at the 2.2-kb position was evident. In addition, another dark band at 2.7 kb and a series of discrete bands above and below the 2.2-kb position were also present (Fig. 2, lane 3). The 2.2-kb probe was hybridized also to a series of cell hybrids derived from 12A and carrying various deletions of chromosome 12. The 12A-1 hybrid, which has a terminal deletion in the short arm ofchromosome 12, still retained the dark bands at the 2.2- and 2.7-kb positions and the bands above and below 2.2 kb (Fig. 2, lane 4). The band patterns appeared to be similar in the two hybrids, and no missing bands were readily discernible in 12A-1. For 12A-2, which has a large deletion in the long arm of chromosome 12, the band at the 2.2-kb position was missing, but a series of distinct bands at other molecular sizes were present (Fig. 2, lane 5). In the case of hybrid 12A-4, which has apparently lost human chromosome 12 completely, a series of bands still appeared (Fig. 2, lane 6), indicating that this segregant still retains some human DNA, although cytogenetic analysis failed to detect its presence. In the hybrid 60A2 (Fig. 2, lane 7), which retains a dot-like chromosome and the chromosome 12 marker serine hydroxymethyltransferase, a series of bands showed up, but the dark 2.2-kb band was again missing. Finally, in another segregant from 12A, 12A-3 (Fig. 2, lane 8), which has lost chro-

Proc. Nati. Acad. Sci. USA 79 (1982)

mosome 12, no detectable bands could be seen in the autoradiogram, indicating the probable loss of all human DNA. These findings suggest that the cloned 2.2-kb EcoRI fragment belongs to a family of human repetitive sequences that are oftwo types: type A sequences are those that have the same size (within the limits ofresolution) as that ofthe cloned 2.2-kb fragment when cleaved with EcoRI, and type B sequences are those that possess homologies to the cloned 2.2-kb fragment but yield sizes other than 2.2 kb when cleaved with EcoRI. In the case of chromosome 12, most of the type A sequences are mapped to the region ql2-qter of the long arm, whereas the type B sequences appear to be scattered along the entire chromosome. Hybridization of the 2.2-kb Probe with DNA from a Series of Human-Chinese Hamster Cell Hybrids Containing Other Human Chromosomes. 32P-Labeled 2.2-kb DNA probe was hybridized to a Southern blot of EcoRI-digested DNA from a series of cell hybrids containing various human chromosomes other than chromosome 12 (Fig. 3). The hybridization showed, for all of the hybrids, a strong hybridization band at 2.2 kb (type A sequence) and a series ofdiscrete bands above and below 2.2 kb (type B sequence). The pattern of the latter sequences appeared to be unique for each human chromosome. These results indicate that sequences with homology to the 2.2-kb repetitive fragment are present on other human chromosomes in addition to chromosome 12. Using hybrids with other single human chromosomes, we have shown that the sequences homologous to the 2.2-kb fragment are present on human chromosomes 11, 21 (Fig. 3, lanes 1 and 6), and 9 or 14, or both (Fig. 3, lane 3). Electroelution of Subfragments of the 2.2-kb Fragment. In order to reduce the complexity ofthe bands on chromosome 12 so that regional assignment of certain bands can be made, the

I. .1 .1

_

I

:.I

I

~40 MW

2.2_ --o

1 2

3

4

5

6

FIG. 3. Autoradiogramsshowinghybridizationofthe2.2-kbprobe

to DNA from cell hybrids containing the following human chromosomes. Lanes: 1, chromosome 11; 2, chromosomes 3, 5, 7, 11, 16, 17, 19, 20, and 21; 3, chromosomes 9 and 14; 4, chromosomes 2, 14, 17, and 20;

5, chromosomes 9 and 21; 6, chromosome 21.

Genetics: Law et aL

Proc. Nati. Acad. Sci. USA 79 (1982) 12A

kb 2.2 6.0 5.4 5.0 4.64.2 FIG. 4. Diagram showing the recombinant plasmid p12-2.2 containing the human insert of 2.2-kb DNA, which is cleaved by Pvu into three subfragments, 0.6, 1.2, and 0.4 kb in the order as shown.

2.2-kb EcoRI fragment was cleaved with restriction endonuclePvu II, which yields three subfragments of sizes 1.2, 0.6, and 0.4 kb. The orientation of these subfragments have been determined as shown in Fig. 4. The subfragments were electroeluted from agarose gels and labeled by nick-translation. Identification and Regional Mapping of DNA Sequences That Are Related to the 2.2-kb Fragment (Type B Sequences). As shown in Fig. 3, the pattern of the type B sequences appeared to be unique for each chromosome. Thus, it might be possible to map each type B sequence to a particular region on a chromosome. Each labeled Pvu II subfragment was hybridized to DNA from 12A (containing an entire chromosome 12) and 12A-1 (containing a partial chromosome 12 including pl205-qter). The results are shown in Fig. 5. By using the subfragments as probes, the band patterns were clearly simplified as compared with that from the use of the entire 2.2-kb fragment as the probe (Fig. 5, lanes A). By using the 1.2. -kb probe, two type B sequences of 5.0 and 4.6 kb were

7393

12A-1 12A-2 6OA2

+

+

+

-

+ + + +

-

-

-

-

-

FIG. 6. Diagram summarizing the hybridization results from Fig. 5. The regional map of human chromosome 12 on the right shows the assignment of the five bands (type B sequences) to the region pter-p1205 of the short arm and the 2.2-kb type A sequence to the region ql2-qter of the long arm.

ase

A

B

12A-1 12A

12A-1 12A

D

C 12A-1

12A

12A-1 12A

60 -5.4=6 50 I l~~~~~~-

g

i~~~~~~~~~~~~~~~~~~~~~~~~~~fr w

4 6

2.2

->t

-- 4|2

*9;

.6

1.2

.4

P

P,_ RF

R

s

;i -4A

$

R

R

p

p

i~

p

p

P

FIG. 5. Diagram showing hybridization of the following labeled probes to DNA from either 12A (right lane in each pair) or 12A-1 (left lane in each pair). Lanes: A, 2.2 kb; B, 1.2 kb; C, 0.6 kb; D, 0.4 kb. R, EcoRI; P, Pvu II.

shown to be present in 12A but not in 12A-1 (Fig. 5, lanes B). Similarly, by using the 0.6- and 0.4-kb probes, another three type B sequences of6.0, 5.4 and 4.2 kb were foundto be present in 12A and absent in 12A-1 (Fig. 5, lanes C and D). From these results, it can be concluded that at least five bands with molecular sizes of 6.0-, 5.4-, 5.0-, 4.6-, and 4.2-kb hybridized to DNA from 12A but not to 12A-1. Therefore, these five bands can be provisionally assigned regionally to the short arm ofchromosome 12 in the region pter-p1205 (Fig. 6). It is also clear from Fig. 5 that the majority ofthe sequences related to the 2.2-kb fragment (type B sequences) hybridized to the ends ofthe 2.2-kb fragment (i.e., the 0.6- and 0.4-kb Pvu II subfragments) and not to the middle portion ofthis fragment (i.e., the 1.2-kb Pvu II subfragment). DISCUSSION Gusella et aL (5) described a method for the construction of cloned DNA fragments from human-Chinese hamster hybrid cells containing a single human chromosome 11 and for distinguishing recombinant phage possessing human DNA inserts from those with hamster inserts. The method was based on the high degree of species-specificity of middle repetitive sequences among different mammalian species (31, 32). Recombinant phage containing human unique sequences have been isolated, and their regional assignments on chromosome 11 have been determined (5). In the present paper, we extended the approach to human chromosome 12 and also demonstrated that human repetitive sequences of relatively low copy number can be identified and isolated for chromosomal mapping analysis of the human genome. In these studies, a 2.2-kb EcoRI fragment was cloned from human chromosome 12. We showed that this 2.2-kb sequence originally derived from human chromosome 12 has homologous copies ofthe same size (type A sequences) and related copies at other molecular sizes (type B sequences) in the human genome. The total number of copies per genome is in the range of 2,000-5,000. We further demonstrated that the type A sequence present on chromosome 12 is also present on other human chromosomes that have been tested, including chromosomes 11 and 21. The type B sequences are abundant on chromosomes 11, 12, and 21 and are likely to be present in multiple copies in most if not all human chromosomes. Although the 2.2-kb probe displayed a continuum or smear when hybridized to total human DNA, it produced discrete

7394 'Genetics: Law et al. 7sProc. Natl. Acad. Sci. USA 79 (1982)

bands when hybridized to DNA from a hybrid containing a single human chromosome (Fig. 2, lane 3; Fig. 3, lanes 1 and 6). Certain other human repetitive sequences, such as the 6.4-kb family found in the region 3' to the f3-globin gene (33, 34) and another family of sequences larger than 2.2kb at the 5' flanking region of the y-globin gene (35), may be present in the human genome in a fashion similar to that of the sequence described here. Because the band patterns of the type B sequences differ in different chromosomes that have been tested, particular bands may serve as specific DNA markers for particular regions of a chromosome. Thus, the potential number of DNA markers for each chromosome can be substantially increased by use of the 2.2-kb sequence described here, which may permit mapping multiple regions of various chromosomes. Availability of such probes could facilitate detection of deletions or other alterations of DNA sequences on specific regions of a chromosome. Such probes may be particularly useful in the analysis ofdiseases such as Wilms tumor and aniridia syndrome, which has been associated with a small deletion of the band p13 in the short arm of chromosome 11 (36); predisposition to retinoblastoma, which has been attributed to the loss of the chromosomal band q14 in the long arm of chromosome 13 (37); or the deletion in the short arm of chromosome 3, region p14-p23, which has been associated with small cell lung carcinoma (38). It also may be helpful in ascertainment of subtle molecular changes in certain cancers involving consistent chromosomal translocations, as in the 9q34/ 22ql1 translocation ofchronic myelocytic leukemia. (39) and the 8q24/14q32 translocation in Burkitt lymphoma (40). With the use of the 2.2-kb sequence and similar DNA sequences, and appropriate somatic cell hybrids containing the single relevant human chromosome with and without the affected alterations, it might be possible to define specific markers for the disease and to elucidate molecular changes in these patients. This approach can be further extended by appropriate restriction enzyme cleavage of the 2.2-kb fragment into subfragments and use of those subfragments that contain less copies in the human genome for mapping purposes. These subfragments not only will yield fewer and more distinct bands when hybridized to single human chromosome DNA but also should form identifiable bands when hybridized to total human DNA. Such subfragment probes could be used to resolve specific bands and may permit them to be mapped to a specific region of a chromosome, as demonstrated in this paper for the five bands mapped on chromosome 12. We thank Dr. S. J. Funderburk for valuable discussion and suggestions and Dr. T. T. Ruck for most helpful advice and criticism. We also thank Karen Kroon and Mary Beth Seeliger for competent technical assistance. This investigation is contribution no. 392 from the Eleanor Roosevelt Institute for Cancer Research and was aided by grants from the American Cancer Society (CD-105), the National Institutes of Health (GM-26631), and the National Foundation-March of Dimes (5276). This paper is no. 36 in the series entitled "Genetics of Somatic Mammalian Cells." The preceding paper is ref. 9. 1. Wu, R., ed. (1979) Methods EnzymoL 68, 1-555. 2. Abelson, J. & Butz, E. (1980) Science 209, 1317-1438. 3. Lawn, R. W., Fritsch, E. F., Parker, R. C., Blake, G. & Maniatis, T. (1978) Cell 15, 1157-1174. 4. Botstein, D., White, R. L., Skolnick, M. & Davis, R. W. (1980) Am. J. Hum. Genet. 32, 314-331. 5. Gusella, J., Keys, C., Varsanyi-Breiner, A., Kao, F. T., Jones, C., Puck, T. T. & Housman, D. (1980) Proc. Nati Acad. Sci. USA 77, 2829-2833.

6. Wolf, S. F., Mareni, C. E. & Migeon, B. R. (1980) Cell 21, 95102. 7. Schmeckpeper, B. J., Smith, K. D., Dorman, B. P., Ruddle, F. H. & Talbot, C. C., Jr. (1979) Proc. Nati Acad. Sci. USA 76, 6525-6528. 8. Davies, K. E., Young, B. D.,'Elles, R. G., Hill, M. E. & Williamson, R. (1981) Nature (London) 293, 374-376. 9. Kao, F. T., Hartz, J. A., Law, M. L. & Davidson, J. N. (1982)

Proc. Nati Acad. Sci. USA 79, 865-869.

10. Schmid, C. W. & Deininger, P. L. (1975) Cell 6, 345-358. 11. Deininger, P. L. & Schmid, C. W. (1976) J. Mol Biol. 106, 773-790. 12. Jelinek, W. R., Toomey, T. P., Leinwand, L., Duncan, D. H., Biro, P. A., Choudary, P. V., Weissman, S. M., Rubin, C. M., Houck, C. M., Deininger, P. L. & Schmid, C. W. (1980) Proc. Nati Acad. Sci. USA 77, 1398-1402. '13. Schmid, C. W. & Jelinek, W. R. (1982) Science 216, 1065-1070. 14. Croce, C. M. (1976) Proc. Natl Acad. Sci. USA 73, 3248-3252. 15. Law, M. L. & Kao, F. T. (1978) Somatic Cell Genet. 4, 465-476. 16. Law, M. L.. & Kao, F. T. (1979) Cytogenet. Cell Genet. 24, 102-114. 17. Law, M. L. & Kao, F. T. (1982)J. Cell Sci. 53, 245-254. 18. Kao, F. T., Chasin, L. & Puck, T. T. (1969) Proc. Natl. Acad. Sci. USA 64, 1284-1291. 19. Gusella, J., Varsanyi-Breiner, A., Kao, F. T., Jones, C., Puck, T. T., Keys, C., Orkin, S. & Housman, D. (1979) Proc. Natl Acad. Sci. USA 76, 5239-5243. 20. Blattner, F. R., Williams, B. G., Belchl, A. E., DennistonThomas, K., Faber, H. E., Furlong, L. A., Grunwald, D. J., Kiefer, D. O., Moore, D. D., Schumm, J. W., Sheldon, E. L. & Smithies, 0. (1977) Science 196, 161-169. 21. Blattner, F. R., Blechl, A. E., Denniston-Thompson, K., Faber, H. E., Richards, J. E., Slighton, J. L., Tucker, P. W. & Smithies, 0. (1978) Science 202, 1279-1289. 22. Maniatis, T., Hardison, R. C., Lacy, E., Lauer, J., O'Connell, C., Quon, D., Sim, G. K. & Efstratiadis, A. (1978) Cell 15, 687-701. 23. Benton, W. & Davis, R. (1977) Science 196, 180-182. 24. Dretzen, G., Bellard, M., Sassone-Corsi, P. & Chambon, P. (1981) Anal. Biochem. 112, 295-298. 25. Girvitz, S. C., Bacchetti, S., Rainbow, A. J. & Graham, F. L. (1980) Anal Biochem. 106, 492-496. 26. Ullrich, A., Shine, J., Chirgwin, J., Pictet, R., Tischer, E., Rutter, W. J. & Goodman, H. M. (1977) Science 196, 1313-1319. 27. Southern, E. (1975) J. Mol Biol. 98, 503-517. 28. Wahl, G. M., Stern, M. & Stark, G. R. (1979) Proc. Natl. Acad. Sci. USA 76, 3683-3687. 29. Rigby, P., Dieckmann, M., Rhodes, C. & Berg, P. (1977)J. Mol Biol 1'13, 237-251. 30. Brandsma, J. & Miller, G. (1980) Proc. Natl Acad. Sci. USA 77, 6851-6855. 31. McConaughy, B. L. & McCarthy, B. J. (1970) Biochem. Genet. 4, 425-446. 32. Soeiro, R. '& Darnell, J. E. (1969) J. Mol Biol 44, 551-562. 33. Kaufman, R. E., Kretschmer, P. J., Adams, J. W., Coon, H. C., Anderson, W. F. & Nienhuis, A. W. (1980) Proc. Natl Acad. Sci. USA 77, 4229-4233. 34. Adams, J. W., Kaufman, R. E., Kretschmer, P. J., Harrison, M. & Nienhuis, A. W. (1980) Nucleic Acids Res. 8, 6113-6128. 35. Duncan, C., Biro, P. A., Choudary, P. V., Elder, J. T., Wang, R. R. C., Forget, B. G., De Riel, J. K. & Weissman, S. M. (1979) Proc. Nat. Acad. Sci. USA 76, 5095-5099. 36. Riccardi, V. M., Sujansky, E., Smith, A. C. & Francke, U. (1978) Pediatrics 61, 604-610. 37. Wilson, M. G., Ebbin, A. J., Towner, J. W. & Spencer, W. H. (1977) Clin. Genet. 12, 1-8. 38. Whang-Peng, J., Kao-Shan, C. S., Lee, E. C., Bunn, P. A., Carney, D. N., Gazdar, A. F. & Minna, J. D. (1982) Science 215, 181-182. 39. Rowley, J. (1980) Annu. Rev. Genet. 14, 17-39. 40. Klein, G. (1981) Nature (London) 294, 313-318.