Mitochondrial DNA is not fragmented during apoptosis

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Communication

Vol. 267 No. 16,Issue of June 5 pp. 10939-10941 1992 0 1992 by The American Socieiy for Biochemistry and Molecular Biolod, Inc. Printed in U.S. A.

present study we have induced apoptosis of two cell types, mouse thymocytes and P-815 mastocytoma cells, by three means: 1) culture in the absence of mitogenic factors (4); 2) incubation in the presence of dexamethasone (5); 3) stimula(Received for publication, December 27, 1991) tion with external ATP (6, 7). By using a cDNA probe for Marta MurgiaSB,Paola PizzoS, Dorianna SandonaS, mitochondrial DNA, we show that inthese cells mitochondrial Paola Zanovelloq, Rosario RizzutoS, and DNA does not undergo fragmentation during apoptosis and Francesco Di Virgilioll runs as intact molecule. This evidence support the specific From the $National Research Council for the Study of the nuclear location of DNA fragmentation duringapoptosis. Physiology of Mitochondria and Institute ofGeneral Pathology, via Trieste 75 and llChair of Immunology, via

Mitochondrial DNA Is Not Fragmented during Apoptosis”

Gattamelata 64, University of Padova, Padova, Italy and the 11Institute of General Pathology, University ofFerrara, via Borsari 46, Ferrara, Italy

MATERIALS ANDMETHODS

Thymocytes, obtained from 3-week-old BALB/c mice, were resuspended at a final concentrationof 5 X 106/ml in RPMI 1640 medium containing the indicated FCS’ concentration; 1-ml samples were then We have exposed mouse thymocytes and P-815 mas- incubated as described in the legends to Figs. 1 and 2. P-815 cells tocytoma cells to four different conditions reported to were grown in culture in RPMI medium as previously described (6), cause apoptosis: 1) incubation in the absence of mito- resuspended in 1ml of RPMI containing 10% FCS at a concentration of 2 X lo6 cells/sample, and incubated in the presence or absence of genic factors; 2) incubation in the presence of dexaindicated valinomycin and ATPconcentration for 6 h. Cells were methasone;3) stimulationwith external ATP; 4) treat- the lysed by addition to theculture medium of 5 X lysis buffer (2 M NaC1, mentwithhighconcentrationsofthe K+ ionophore 50 mM Tris/Cl, 10 mM NaEDTA, pH 8, 1%sodium dodecyl sulfate, valinomycin. These treatments caused DNA fragmen- 0.25 mg/ml proteinase K), and DNA was extracted by the salting-out tation to a varying extent in the two cell types. High method. Electrophoresis was carried out on 1 or 0.8% agarose gel stringency hybridization with a cDNA probe specific containing 0.5 pg/ml ethidium bromide. The same amount of DNA, to a mitochondrialDNA sequence revealed that during as assessed by spectrophotometric measurement, was loaded in each apoptosis induced by lack of mitogenic factors, dexa- lane. Isolation of cell nuclei and mitochondria from mouse liver was methasone, or extracellular ATP, mitochondrialDNA performed according to standard procedures (8, 9). Hybridization with the 32P-labeled mitochondrial DNA probe was carried out at was notfragmented. On thecontrary,valinomycin “C. The rat cDNA probe was kindly provided by Dr. Cecilia caused extensive degradation of mitochondrial DNA. 60 Saccone, University of Bari (10).

These results support the notion that DNA fragmentation during apoptosisis a specific nuclearevent and suggest that other agents, such as valinomycin, may act less selectively.

RESULTSANDDISCUSSION

We have induced apoptosis of mouse thymocytes by three procedures: ( a ) short-term in uitro culture in the absence of mitogens; ( b ) incubation in thepresence of 1 p~ dexamethasone; ( c ) incubation inthe presence of high valinomycin There is increasing interest in the mechanism of apoptosis, concentrations. Both dexamethasone and incubation in the the active process whereby cell death occurs during embryo- absence of mitogens (“spontaneous fragmentation”)triggered genesis, hormone and growth factor deprivation, thymic se- extensive DNA degradation whichwas easily detected by lection, cell-mediated cytotoxicity, and glucocorticoid treat- agarose gel electrophoresis of cellular DNA (Fig. 1).Spontament (1). Apoptosis is understood to require gene expression neous DNA fragmentation occurred after a 4-6-h incubation and activation of nuclear endonucleases, these latter respon- and was prevented by increasing the serum concentration of sible for the pattern of DNA degradation into nucleosome- the incubation medium (Fig. 1,lane 6).Valinomycin, a widely size fragments considered to be the hallmark of apoptosis (1, used and powerful DNA-fragmenting agent ( l l ) , was a rather 2). DNA-degrading events are believed to occur early after weak stimulus for DNA degradation in mouse thymocytes the delivery of the lethal hitand before intracellular organelles under our experimental conditions (not shown). and the plasma membrane are damaged (3). Thus, according DNA wastransferred to nitrocellulose filters and hybridized t o this hypothesis, DNA degradation should be a specific a t high stringency with a rat cDNA probe complementary to nuclear event. However, whether DNA contained within other a mitochondrial DNA sequence of 3000 base pairs (10). This subcellular organelles, e.g. the mitochondria, also undergoes cDNA hybridized with purified mouse mitochondrial but not fragmentation or rather remains intact, is unknown. In the nuclear DNA, giving a major band of 21,000 base pairs and a minor band of apparent lower M,, probably corresponding to * This work was supported by grants from the National Research Council (Target Project Biotechnology and Bioinstrumentation; Spe- supercoiled DNA (Fig. 2, panel A ) . However, as shown in Fig. cial Project Biology and Pathology of Calcium), Ministry of Scientific 2, panels B and C, the cDNA probe did not hybridize with low Research (MURST), and Associazione Italiana per la Ricerca sul M , DNA fragments isolated from apoptotic thymocytes (lanes Cancro (AIRC). The costs of publication of this article were defrayed 1 and 3 in panel B; lanes 1 and 4 in panel C), suggesting that in part by the payment of page charges. This article must therefore the ladder of DNA fragments contained no mitochondrial be hereby marked “advertisement” in accordance with 18 U.S.C. DNA. To prove the ability of the probe to specifically recogSection 1734 solely to indicate this fact. The nucleotide sequencefs)reported in thispaper has been submitted nize the mitochondrial DNA in the totalcellular DNA extract, to the GenBankTM/EMBL Data Bank with accession numberfs) 14848. we digested cellular DNA with three different restriction enzymes, PuuII, PstI, andHindIII. After digestion of cellular § To whom correspondence should be addressed Institute of General Pathology, University of Padova, via Trieste 75,I-35121 Padova, Italy. Tel.: 49-828-6573;Fax: 49-828-6576.

10939

’ The abbreviation used is: FCS, fetal calf serum.

Apoptosis Mitochondrial and DNA

10940 1 2 3

4 5 6

1

2

3

1

2

3

1 2 3 4

5

1 Q-

FIG. 1. A g a r o s e gel electrophorcsis of total cellular DNA f r o m m u r i n e t h y m o c y t e s s h o w i n g t h e p a t t e r n o fDNA b a n d s t y p i c a l of apoptosis. Cells were incutmted at : G "C in RI'MI medium containing either IO"0 (lanes 1 - 3 ) or 20"; ( l a n ~ 4-6) s FCS. I A I ~ P S2 and 5 . DNA extracted from control thymocytes immediately after the isolation from thethymus; lanes 1 and 4 , DNA extracted from t hymocytes incuhated for 6 h in the presenceof 1 p~ dexamethasone; lanes 3 and 6. DNA extracted from thymocytes after 6 h of incuhation in the ahsenceof dexamethasone (spontaneous fragmentation).

#

B

2 3

4-5

Kb

-

2 3.1 -9.4 -6.5 -4.4

-2.0 -0.56

DNA, Southern hybridization with the mitochondrial DNA probe identified discrete bands corresponding to the known 1 2 3 4 5 6 . 7 1 2 3 4 5 6 7 .I. restriction map of the intact mitochondrialDNA (lunes 2 and KD 1 4 in panel R, and lanes 2,3,5, and 6 in panel C).These results 2 3.1 provide conclusiveevidence thatnofragmentation of the C -9.4 -6.5 mitochondrial DNA occurred in the apoptoticcells. However, -4.4 they do not rule out the possibility that failure of mitochon2.0 drial DNA to undergo fragmentation during apoptosis could be due to an intrinsic resistance of native mitochondrial DNA -a56 t o intracellular nucleases. To exclude this possibility, we investigated the effect of valinomycin in a cell type, the P815 mastocytoma cells, which is known to undergo massive FIG. 2 . S o u t h e r n hlot a n a l y s i s of total c c ~ l l u l a rI)Sjl f r o m DNA fragmentation when stimulated with this ionophore (11,a p o p t o t i c m u r i n e t h y m o c y t e sby h y b r i d i z a t i o n t o a r a t c n N A 12). As shown in Fig. 3, panel A, valinomycin promoted probe specific for mitochondrial D N A . Agnrnse gel elrctrophoextensive fragmentation of cellular DNA (lune I, left); how- resis ( l e f t ) and corresponding autoradiogrnphy (rrcht I are shown in each panel. Panel A: lanes 1, purified mouse liver mitochondrial DSA ever, under these conditions mitochondrial DNA was also specifically recognized hv the prohe; lnncs 2. purifiedmouse liver degraded, as shown by hybridization of the mitochondrial nuclear DNA which did not react with the prohe: l a n c 3 . molecular probe with the DNA (panel A, lune 1, right). It is likely that, weight markers (X-DNA digest of I h K I and HindIlll. Panrl H : total at the high concentration used to induceDNA fragmentation, cellular DNA from thymocytes treated with dexamethasone (Iancs I valinomycin also caused a generalized alteration of intracel- and 2 ) and thymocytes that underwent spontaneous fragmentation lular membranes, thusallowing degradation of mitochondrial (lanes 3 and 4 ) ; DNA in lanes 2 and 4 was also digested with rest rirt ion endonuclease PuulI, while DNA in /ones 1 and .'I was undigested: DNA. On the contrary, external ATP, recently reported be to lanrs 5 , molecular weight markers (X-DNA digest of Hintilll~.I'nncl a DNA-fragmenting agent in P-815 and other cell types (6, C: total cellular DNA from thymocytes treated with dexamethasone 7 ) ,caused no mitochondrial DNA breakdown(Fig. 3, panel R, (lanes 1-3) and thymocytes that underwent spontaneous fragrnentnlanes I ). DNA fragmentation was specifically due to ATP in tion (lanes 4 - 6 ) ; DNA digestedwith restrictionendonuclease /'st1 (lanes 2 and 5 ) ; DNA digested with restriction endonrlclease \findlll its fully dissociated (ATP") formsinceaddition of M$+ ( l a n ~ 9s and 6 ) ;undigested DNA (lnnc.9 1 and 4 ) . Equal amounts of completely prevented the effect (panel R, lunes 2 ) . DNA from the various samples ( 5 and 2.5 p g in panrls R anti (', An increasing number of physiologic and pathologic events respectively) were digested with the indicated restrictionenzyme. Kh. are known to involve cell death by apoptosis (131, but although kilohasds). this mechanism is thought to be active and to require protein synthesis and an intact energy metabolism, how this conprompted speculation that specific endonucleasesareactitrolled suicide is triggered and regulated is still mysterious. A vated in apoptotic cells. On the other hand, it is difficult to handful of surface molecules whose ligation triggers apoptosis rule out that DNA is rather digested as a consequence of an is known, e.g. APO-1 (14), the TNF receptor (IS), and a still early and generalized damage to intracellularmembranes ill defined membrane molecule bound by extracellular ATP which could permit digestionnotonlv of nuclear but also ( 6 , 7 ) . In addition, the recent observation that overexpression of a protein located in the mitochondria, the product of the mitochondrial DNA. Furthermore, it has also been suggested nuclear gene bcl-2, prevents lymphoid cells from undergoing that DNA fragmentation could be directly caused bv free apoptosis also suggested intriguing links between apoptosis radicals which are generated duringseveral processes leading we would to cell death (18). Werethesesuggestionstrue, and the mitochondria (16, 17). One of the most surprising of mitochondrialas well as aspects of apoptosis is the early degradationof cellular DNA expecttofindfragmentation nuclear DNA, since collapse of intracellular membranes or that is considered to be the hallmark of this process. DNA degradation in apoptotic cells is highly reminiscent of DNA generation of highlv reactive oxvgen intermediates would cause unrestricted DNA damage. On the contrary, the three digestion by bacterialrestriction enzymes, andthishas

-

-

Mitochondrial Apoptosis D N A and

A r h n o r l , l r d ~ r n e n t s - ~ wish ~ e thank Prof. to C. Saccone (I’niversity o f Hari) for the kind gift of the mitochondrial cDNA prohe and Prof. T . I’ozzan (LJniversityof[’adova)forhelpfulsuggestions.

B

A

.12Kb -23.1 I

.

-6.5

10941

06*5

REFERENCES 1. Willve, A. H.. Kerr. J. F.

R. Xr Currie, A. R. (1980) Int. RPP.Cytol.

68,2 5 1 4 0 5

-2.0

2. Duvalle, E.Rr \#’illye, A. H. (1986) Immunol. Today 7. 115-119 3 . Husch, LV., Kleine, I,. R: Tenniswood. M. (1990, Riochrm. C‘rll Riol. 68,1071-1074 00.56 4. Collins, It. .I., Verschuer, I,. A., Harmon, R. V., Prentice, R. I,.. -0.56 Pope, J. H. Rr Kerr, .J. F. (1989) Hr. J . Hnrrnotol. 71, :l43-3.50 5. Ucker, D. S. (19x7) Y n t u r r 327, 62-64 6. Zanovella, P., Hronte, V., Rosato, A,, Pizzo, P. Rr Di Virgilio, F. (1990)J . Irnrnunol. 145, 1.545-15.50 7 . Zheng, I,. M., Zychlinskv, A,. Liu, C-C.. Ojcius, D. M.X. Young. .J. I>-E. (1991) J . C ’ d l H i d . 112. 279-288 FIG. 3. Effect of valinomycin and external ATP on mito8. Palva, T. K. X; Palva, E. 7’. (198.5) FI.’H.S h t t . 192, 267-270 chondrial DNA extracted fromP-8 15 mastocytoma cells. Aga9. Rusch, H. (1967) Mrthods I;nzyrnol. 12. 321-4.’14 rose gel electrophoresis ( / r / t ) and autoradiography ( r i g h f )are shown 10. Gadaleta, G.. Pepe, G., De Candia, G., Quagliarello. C.. Shisa, E. in each pond. I’onrl A: lonrs I . DNA extracted from cells treated Rr Saccone, C. (1989) J . Mol. F;td. 28. 4Y7-516 with valinomycin (15 pM): lnnrs 2, control cells. I’nnrl H : 1onr.s I , DNA extracted from cells treated with extracellular ATP ( 5 mM): 11. Martz, E.Xr Howell, I). M.(1989) ImmunoI. Today 10, 79-86 12. Allnitton, N. I,.. L‘erret, C. It., Wolley. R. C. X; Risen.H. N. Innc~s2, control cells treated with5 mM A T P in the presence of 5 mM (1988) J . Exp. Mrd. 167, q514-527 MpCI,. Kb. kilohase(s). 19. Williams, G. 7’. (1991) C’rll 65,1097-1098 14. Trauth. R. C., Klas. C., Peters, A. M.. Matzku. S., Moller. P., Falk, Mr., Dehatin,K-M. Rr Krammer, 1’. H. (1980) Srirnce 245, 301-30~5 apoptotic stimuli used in the present experiments left mito15. Laster, S . M., LVood, J. G. X; Gooding, I,. R. (19RR)J . Immunol. chondrial DNA remarkably intact, evenin the presence of 14 1, “29-2634 16. Henderson, S.. Rowe,M., Gregory, C., Croom-Carter. D., Wang, extensive nuclear DNA degradation. This observation lends F., Kieff, E.R: Rickinson, A. (1991) C‘cll65. 1107-131.5 support to the hypothesis that DNA cleavageduring apoptosis 17. Hockenhery, D., Nunez. G . , Milliman. C., Schreiher. H. D. R: is a specific nuclear event likely dependent on enzyme pathKorsmever. S. J. (1990) Noturr 348. R R 4 - R : K 18. Halwell, H. R: Aruoma. 0. 1. (1991) F E R S b f t . 2 8 1 . 9 - 1 9 ways which are exclusively located in the nucleus.