Dispersed repetitive sequences of the mouse genome are differentially represented in extrachromosomal circular DNAs in vivo

PLASMID

17.257-260

(1987)

SHORT COMMUNICATIONS Dispersed Repetitive Represented

Sequences of the Mouse Genome Are Differentially in Extrachromosomal Circular DNAs in Vivo

SONIA C. FLORES, TERRY

KAY MOORE,

AND JAMES W. GAUBATZ

Department of Biochemistry, University of South Alabama, College of Medicine, Mobile, Alabama 36688 Received November 17, 1986; revised February 17, 1987 Eukaryotic cells contain extrachromosomal circular (ccc) DNAs which are homologous to chromosomal sequences. We have isolated eccDNAs from whole tissues of C57BL/6 mice. Using hybridization techniques, we show that R-, MIF-, Bl-, and B2dispersed repetitive sequences of the mouse genome are differentially represented in heart, brain, and liver tissues. Moreover, we show that the relative abundance of B2 sequences in heart and liver eccDNAs is higher than the relative abundance of the other repetitive sequence families studied. o 1987 Academic

Prcsl Inc.

Most eukaryotic cells which have been studied contain a population of extrachromosomal circular (ccc)’ DNA molecules that can be physically separated and distinguished from chromosomal DNA. The size distribution of these circular molecules can be very heterogeneous, ranging from 150 to over 20,000 base pairs (bp) (I), although distinct size classes have been observed in some cells (2). Most eccDNA sequences are homologous to sequences which are highly repeated in nuclear DNA, for example, Alu sequences, satellite sequences, and long interspersed repetitive (LINE) sequences such as KpnI and R. A smaller fraction of eccDNAs has a significantly higher and undefined sequence complexity, suggesting the presence of unique or single-copy chromosomal sequences (3-5). Some families of dispersed repetitive sequences appear to be transposable elements, and it has been suggested that they are propagated throughout the genome by a retroposition mechanism; thus, they have been termed retroposons (6). Retroposons belonging to the short repetitive DNA (SINE) fami’ Abbreviations used: ccc, extrachromosomal circular bp, base pairs; LINE, long intemper& repetitive; SINE, short interspersed repetitive; kb, kilobases.

lies have many similar structural features. The Alu family of primates and the Bl and B2 families of rodents contain RNA polymerase III promoters, have long polydA tracts at their 3’ ends, are flanked by short direct repeats (7), and are abundantly transcribed into nuclear RNA (8). Bacterial transposable elements appear to move from one area of the genome to the other by a process of recombination and replication between two separate regions of the genomic DNA (9). However, the transposable copia elements of Drosophila (IO), yeast Ty elements (11), and mouse intracisternal A particles (12) appear to transpose via an RNA intermediate. Reverse transcription of this RNA intermediate may generate extrachromosomal circular or linear forms of the element that could subsequently insert at different chromosomal locations. Indeed, copia elements and Ty elements are found extrachromosomally as circular elements (13). Sunnerhagen el al. (14) and others (15) recently have shown the presence of repetitive sequences in eccDNAs of cultured mouse cells. We have looked for the presence of eccDNA molecules in whole tissues of C57BL/6 mice in order to study possible in vivo events of extrachromosomal circular molecules such as transposition or other ge-

257

0147-619X/87

$3.00

Copyrighl8 1987 by Acwkmic Pms. Inc. All rights of mpmduction in any form reserved.

258

SHORT COMMUNICATIONS

netic rearrangements that might reflect genomic instability or remodeling. Using hybridization techniques, we report here the existence of eccDNA sequences homologous to the major dispersed repetitive DNA sequence families of the mouse genome. Heart, liver, and brain tissues were removed from mice and homogenized. After selective alkaline denaturation-renaturation (16) and subsequent centrifugation, the supernatant was collected, and nucleic acids were precipitated and treated with RNase A and proteinase K. The digest was then subjected to equilibrium centrifugation in ethidium bromide/cesium chloride gradients as described (I 7). To determine the location of eccDNAs in the gradient, samples from each fraction were blotted onto nitrocellulose filters and hybridized with nick-translated 32P-labeled chromosomal DNA. Two regions of DNA were detected in the gradient by hybridization. The peak of material hybridizing at the higher density of 1.6 to 1.63 g/ml, as determined by refractometry, was pooled and rebanded in ethidium bromide/cesium chloride equilibrium gradients two additional times. DNAs thus isolated were shown to be resistant to digestion with Sl nuclease, exonuclease III, RNase H, Pronase, and RNase A. However, the preparations were sensitive to the action of DNase I and digestion with restriction enzymes led to a redistribution of DNA in ethidium bromide/cesium chloride gradients that was not observed in untreated controls. The restriction enzyme treatment was consistent with linearizing part of a population of closed circular DNAs (manuscript in preparation). The eccDNA preparations were slot blotted (18) at various dilutions onto quadruplicate nitrocellulose filters. The filters were then hybridized to cloned mouse B 1, B2, R, and MIF repetitive sequences. Figure 1 shows that sequences homologous to the major repeat families of the mouse genome are present in eccDNAs in vivo. Figure 1 also shows that the relative abundance of these repetitive sequences in eccDNAs is different

Liver

Bdl

Heart

FIG. 1. Hybridization of repetitive sequence probes to eccDNAs of heart, brain, and liver tissues of &monthold C57BL/6 mice. Extrachromosomal circular DNAs were purified by alkaline denaturation-renaturation followed by three consecutive rounds of cesium chloride/ ethidium bromide density gradient centrifugations. Form I material was collected, quantitated, and slot blotted at various dilutions. Aliquots of 2, 12,30, and 60 gl of each eccDNA preparation were treated as described (18) and blotted onto quadruplicate nitrocellulose filters. The filters were then hybridized to the indicated radiolabeled probes pMR225 (Bl), pMRl42 (B2), pMR134 (MIF), and pMR290 (R), each having a specific radioactivity of 5 X 10’ cpm/Mg. The plasmid vector sequence of recombinants pMR134 and pMR290 is pBR322. The vector sequences of recombinants pMR225 and pMR142 were from pSP64, a derivative of the pUCl2 plasmid carrying the SP6 polymerase promoter from Sulmone/Iu. The vector pSP64 shares approximately 70% homology with pBR322 (20). Insert sizes in these recombinant plasmids ranged from 100 to 200 bp (I 9). Hybridization and washes were done at high stringency conditions and no hybridization to mitochondrial DNA was detected (data not shown). Hybridization of pBR322 plasmid to eccDNA preparations gave background values which were subtracted from filters hybridized with repetitive sequences. Filter strips representing each slot were cut and counted, counts in all the strips representing a treatment were pooled and plotted. A linear relationship between the amount of DNA applied to each slot and the number of counts in that slot was observed. Radioactivity is expressed as cpm per gram of total tissue DNA. The results shown are from a representative experiment.

for each tissue, with heart having the highest representation of all the sequences studied. In liver and heart tissues, B2 sequence families are found at a much higher level in eccDNAs than are the Bl or the other repetitive sequence families. The percentage of B 1 sequences in genomic DNA is approximately

259

SHORT COMMUNICATIONS

0.9 to 1.3%, while the percentage of B2 sequences is 0.5 to 0.8%. In both cases the consensus size of the repeated sequence is ap proximately 200 bp (19). R and MIF sequences belong to the same family of long interspersed repeats of the mouse. These repetitive sequences have been reclassified as Ll sequences. The size of Ll sequences is highly variable, but can be as large as 7 kilobases (kb). This size variability is due to truncation at apparently random distances from a common 3’ end (ZZ). The percentage of R sequences in genomic DNA is approximately 1.O to 1.5%, while that of MIF varies from 4.0 to 5.0% (19). Although titration studies to determine the percentage of these repetitive sequences in eccDNA were not performed, the fact that the relative abundance of B2 is significantly higher than that of B 1 when compared with their ratios in the genome appears interesting. If, as has been suggested, repetitive sequences move about the genome through an RNA intermediate that is reverse transcribed to yield doublestranded circular or linear DNA molecules, the identification of eccDNA molecules carrying repetitive sequences could be interpreted as evidence of their role as transposition intermediates. In support of this notion, Bl and B2 sequences are transcribed by RNA polymerase III. However, the B2 sequence promoter resembles more closely the established RNA polymerase III consensus promoters associated with tRNA genes (22). It is possible that B2 is therefore transcribed more efficiently to yield higher levels of B2homologous eccDNA molecules. Regardless of the mechanism by which eccDNAs are generated, their presence and possible mobility suggest that they might insert into different regions of the genome and alter the cellular phenotype. Tissue-related polymorphisms in an androgen-regulated gene of the mouse have been ascribed to the insertion of Bl sequences in the 3’-untranslated regions of the gene (23). In addition, a moderately repetitive endogenous transposon-like sequence of the mouse genome (in-

tracistemal A particles) has been reported to activate the c-mos oncogene in a mouse plasmacytoma by virtue of its insertion into the coding region (24). Knowledge about the possible roles and mechanisms of formation of eccDNAs in vivo will require the cloning and sequencing of these circular DNA molecules to determine their relationship to segments of the genome. ACKNOWLEDGMENTS The authors thank Dr. Karen L. Bennett and Dr. Nicholas D. Hastie for kindly providing the cloned repetitive sequences B I, B2, MIF, and R, and Dr. Gary Daniels for critically reviewing this manuscript. This work was supported by grants from the Hearst Foundation, American Federation for Aging Research, and The Council for Tobacco Research (GN 16 I 1). S.C.F. is the recipient of a Procter & Gamble Co. Graduate Fellowship.

REFERENCES SMITH,

C. A., AND VINOGRAD,

J.,

J. Mol. Biol. 69,

163-178 (1972). STANFIELD,

S., AND

HELINSKI,

D. R.,

Cell

9,

333-345 (I 976). KROLEWSKI, J. J., BERTELSEN, M. Z., AND RUSH, M. G.,

399-415 (1982). 4. BERTELSEN, A. H., HUMAYUN, POULOS,

S. G., AND RUSH,

A. H., HUMAYUN,

J. Mol. Biol. 154, M.

Z., KARFO-

M. G., Biochemistry

21.2076-2085 (1982). 5. JONES, R. S., AND POTTER, S. S., Proc. Natl. Acad. Sci. USA 82, 1989-1993 (1985). 6. WEINER, A. M., DEININGER, P. L., AND EFSTRATIADIS, A., Annu. Rev. Biochem. 55, 631-661 ( 1986). 7. JELINEK, W. R., AND SCHMID, C. W., Annu. Rev. B&hem. 51,8 13-844 (1982). 8. HAYNES, S. R., AND JELINEK, W. R., Proc. Natl. Acad. Sci. USA 78,6 130-6 134 ( 198 I). 9. CAU)S, M. P., AND MILLER, J. H., Cell 20,579-595 (1980).

IO. FLAVELL, A. J., AND ISH-HOROWICZ, D., Cell 34, 415-419 (1983). 11. BOEKE, J. D., GARRNKEL, D. J., STYLES, C. A., AND FINK,

G. R., Cell 40,49

l-500

(1985). K., SMITH, SHULMAN,

12. KUFF, E. L., FEENSTRA, A., LUEDERS, L., HAWLEY, R., HOZUMI, N., AND M., Proc. Nat/. Acad. Ski. USA 80, (1983). 13. FLAVELL, A. J., AND ISH-HOROW~, (London) 292.591-595 (1981).

1992-1996 D., Narure

260

SHORT

14. SUNNERHAGEN, A-L., LUNDH,

P., SJOBERG, R-M., L., AND BJURSELL,

Acids Rex 14,7823-7837

COMMUNICATIONS

KARLSON, G., Nucleic

(1986).

15. FLJJIMOTO, S., TSUDA, T., TODA, M., AND YAMAGISHI, H., Proc. Natl. Acad. Sci. USA 85, 2072-2076 (1985). 16. BIRNBOIM, H. C., AND DALY, J., Nucleic Acids Rex 7, 1513-1523 (1979). 17. SMITH, C. A., JORDAN, J. M., AND VINOGRAD, J., J. Mol. Biol. 59,255-272 (197 1). 18. KAFATOS, F. C., WELD~N JONES, C., AND EFSTRATIADIS, A., Nucleic /Icids Res. 7, 1541-I 552 (1979). 19. BENNETT, K. L., HILL, R. E., PIETRAS, D. F., WOODWORTH-GUTAI, M., KANE-HAAS, C.,

HOUSTON, J. M., HEATH, J. I(., AND HASTIF. N. D., Mol. Cell. Biol. 4, 156 I - I57 I ( 1984). 20. VIEIRA, J., AND MESSING, J., Gene 19, 259-268 (1982). 21. FANNING, T. G.. Nucleic Acids Rex I I, 5073-509 1 ( 1983). 22. DANIELS, G. R., AND DEININGER, P. L., Nature

(London) 317.819-822 23. KING,

D., SNIDER,

(1985).

L. D., AND LINGREL.

J. B.,

Mol.

Cell. Biol. 6, 209-2 I7 (I 986). 24. CANAANI, E., DREAZEN, 0.. KLAR, A., RECHAVI, G., RAM, D., COHEN, J. B., ANDGIVOL, D., Pioc.

Natl. Acad Sci. USA 80,7 I 18-7 I22 ( 1983). Communicated by Donald R. Helinski