The topoisomerase II inhibitor teniposide (VM-26) induces apoptosis in unstimulated mature murine lymphocytes

EXPERIMENTAL

CELL RESEARCH

200,416-424

(1992)

The Topoisomerase II Inhibitor Teniposide (VM-26) Induces Apoptosis in Unstimulated Mature Murine Lymphocytes CHITRA ROY,’ DAVID L. BROWN,JUDY E. LITTLE, BEATRICE K. VALENTINE, P. ROY WALKER,* MARIANNA SIKORSKA,* JULIE LEBLANC,* AND NATHALIE CHALYt Department of Biology, University of Ottawa, Ottawa, Canada, KIN 6N5; *Molecular Cell Biology Group, Institute for Biological Sciences, National Research Council of Canada, Ottawa, Canada, KlA OR6; and tDepartment of Biology, Carleton University, Ottawa, Canada KlS 5B6

ulates DNA topology by creating an enzyme-linked double strand break in DNA, thus enabling the passing of another double strand through the break, and then resealing it ([l, 21 for reviews). In addition there is indirect evidence that topo II plays a structural role in the organization of chromatin loops in the interphase nucleus. Topo II has been shown to be associated with newly replicated DNA [3] and to be a component of the nuclear matrix [4-61. The regions of DNA believed to represent the bases of chromatin loops, termed matrix-associated regions [ 71 or scaffoldassociated regions [8], have been shown to be rich in topo II consensus cleavage sequences. Taken together this evidence has suggested that topo II is at least one factor mediating the attachment of DNA to the nuclear matrix (for recent reviews see [5, 9, lo]). Topo II, which is present at high levels in proliferating cells, has been a major target for the development of cancer chemotherapeutic drugs such as teniposide (VM-26), etoposide (VP-16), and amascrine (mAMSA). These drugs inhibit topo II by forming drugstabilized enzyme-DNA complexes and thereby interfere with the breakage-resealing activity of the enzyme. By an as yet unestablished mechanism this DNA damage is lethal to proliferating cells ([ll, 121 for reviews). In addition to their value as chemotherapeutic agents, these inhibitors have been used also to study the role of topo II in processes such as chromosome condensation [13, 141. We have been using these inhibitors to assess the role of topo II in the restructuring of chromatin that accompanies lymphocyte stimulation [ 151. Although topo II levels increase greatly during lymphocyte stimulation, the amount and activity are undetectable or barely detectable in unstimulated lymphocytes from several species [16-201. We were surprised, therefore, to find in our recent studies that brief (2 h) treatments of unstimulated splenic lymphocytes with VM-26 resulted in cell death by a process similar to that observed earlier in topo II-rich rat thymocytes and determined to be apoptosis [21]. Apoptosis is characterized by morphological changes, such as chromatin condensation and nuclear fragmentation, and by internucleosomal cleavage of DNA ([22,23] for reviews). These characteristics are

This study shows that not only concanavalin A-stimulated proliferating lymphocytes but also unstimulated mouse splenic lymphocytes are sensitive to the topoisomerase II (topo II) inhibitor teniposide (VM-26). When unstimulated lymphocytes are pretreated with VM-26 for a 2-h period and are then incubated in drugfree medium, cell viability, as determined by trypan blue exclusion, decreases to 40% of the control by 6 h. The drug-treated cultures show two to three times the level of detergent soluble DNA than the control cultures and agarose gel electrophoresis of the soluble DNA shows the presence of oligonucleosomal-sized fragments, a feature considered to be a hallmark of apoptosis. Phase contrast microscopy, Hoechst staining for DNA, and immunofluorescence microscopy of various nuclear and cytoplasmic antigens (nucleolar fibrillarin, snRNP, ubiquitin, vimentin, tubulin) in the VM-26treated cells characterize the morphological changes during apoptosis of these cells. The role of topo II as the mediator of the VM-26 effects is supported by pulsed field gel electrophoresis, which shows the typical topo II-induced cleavage of supercoiled DNA into loop-sized 300- and SO-kbp fragments. We conclude that the cancer chemotherapeutic agent VM-26 interacts with topo II and induces apoptosis in unstimulated lymphocytes. 0 1992 Academic Press, Inc.

INTRODUCTION Topoisomerase II (topo II)2 is known primarily for its enzymatic role in the regulation of the topology of supercoiled DNA. Topological interconversions are required for several functions of DNA, such as replication, transcription, recombination, and repair, as well as for chromosome condensation and disjunction. Topo II reg’ To whom correspondence and reprint requests should be addressed. Fax: (613) 564-5608. 2 Abbreviations used: Con A, concanavalin A, FITC, fluoroscein isothiocyanate; topo II, topoisomerase II; TRITC, tetramethyl rhodamine isothiocyanate; VM-26, teniposide; snRNP, small nuclear ribonuclear protein. 0014-4827J92

$5.00

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

416

VM-26-INDUCED

APOPTOSIS

IN UNSTIMULATED

documented in this study and lead us to conclude that VM-26 induces apoptosis in unstimulated splenic lymphocytes. Moreover, using pulsed-field electrophoresis to reveal the high molecular weight DNA damage induced by VM-26, we show that unstimulated splenic lymphocytes contain the enzyme topoisomerase II. MATERIALS

AND

METHODS

CeU culture. Splenic lymphocytes of Balb/c mice (Charles River, Montreal, Quebec) were pooled and cultured at 2.5 X 10’ cells/ml in RPM1 1640 (Flow Laboratories, Mississauga, Ontario) containing 6% fetal calf serum (Oxoid Canada Inc., Nepean, Ontario), 50 rglml gentamicin sulfate (Sigma, St. Louis, MO), and 50 pM 2-mercaptoethanol, buffered to pH 7.4 with 20 m&4 Hepes. The cells were incubated at 37°C in a humidified incubator containing 5% CO,. Cells were stimulated by the addition to the culture of 4 pg/ml concanavalin A (Con A, Calbiochem-Behring Corp., La Jolla, CA). Drug treatment. A 15 m&f VM-26 (Bristol Laboratories, Candiac, Quebec) stock solution was prepared in ethanol and stored at -20°C. For treatment of S-phase cells, Con A-stimulated splenocytes were incubated with various concentrations of VM-26 for 2 h starting 46 h after stimulation. The unstimulated lymphocytes were treated with VM-26 for 2 h, washed, and then recultured in drug-free medium with or without Con A for various lengths of time. Complement-mediated cell lysis. To isolate T cells, splenocytes were incubated with magnetic beads coated with goat anti-mouse IgG (H and L) at 4°C for 30 min in a petri dish, which was then placed on a magnet (BioMag Separator, Advanced Magnetics Inc., Cambridge, MA) for about 5 min. The cell suspension containing the T cells was then collected and the T cells were pelleted and incubated at 4’C for 30 min with the 30H12.4 rat anti-Thy 1.2 monoclonal antibody [24]. Following this the cells were washed, pelleted, treated with Low-ToxM rabbit complement (Cedarlane Ltd., Hornby, Ontario) for 1 h at 37”C, and then washed with medium three times and used for experiments. Cell viability. Using a hemocytometer, cells excluding trypan blue were counted as viable. DNA synthesis. Incorporation of [3H]thymidine (NEN, New England Nuclear, Boston, MA; sp act 84.2 Ci/mmol) was measured between 46 and 48 h of stimulation. For S-phase cells, VM-26 and the radiolabel (4 &i/ml) were added simultaneously, and the cells were harvested after a 2-h incubation at 37’C. Unstimulated cells were pretreated with VM-26 for 2 h, washed, and recultured in drug-free medium containing Con A. Incorporation of [3H]thymidine was then measured between 46 and 48 h of stimulation as above. Following either treatment, the cells were harvested onto glass-fiber filters using a manifold and the radioactivity on the filters was determined by liquid scintillation counting in a Packard liquid scintillation counter, Microscopy. Effects of VM-26 treatments on cell morphology were determined by immunofluorescence staining with the following primary and secondary antibodies: (1) Y12 antibody for the Sm antigen of small nuclear ribonucleoproteins (snRNP) [25] and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG (H and L); (2) serum Di (Scl C) for the fibrillar component of the nucleolus [26] and tetramethylrhodamine isothiocyanate (TRITC)-conjugated rabbit anti-human Ig; (3) anti-ubiquitin (Sigma) and FITC-conjugated goat anti-rabbit Ig; (4) anti-vimentin [27] and FITC-conjugated goat anti-rabbit Ig; (5) 5A6 anti-tubulin [28] and FITC-conjugated goat anti-mouse Ig. All secondary antibodies were purchased from Cappel Laboratories, Inc. (Malvern, PA). For immunostaining cells were layered on poly-L-lysine-coated coverslips and processed as described by Chaly et al. [29] for the detection of nuclear antigens and vimentin and by Roy et al. [ 301 for microtubule detection. Chromatin was stained with the DNA-specific fluorochrome Hoechst 33258 (American Hoechst Corp., San Diego, CA),

LYMPHOCYTES

used at 1 rig/ml for 1 min. The controls included untreated and VM26-treated cells processed for immunofluorescence detection of nuclear antigens, vimentin, or tubulin but incubated with the secondary antibodies only. The cells were examined using a Zeiss universal microscope equipped with epifluorescence optics and photographed using Ilford XPl-400 film. DNA fragmentation. DNA fragmentation in control and VM-26treated lymphocytes was quantified by a modification of Burton’s method [31], as described by Newell et al. [32]. In brief, cell pellets (3 x 10’ cells) were suspended in 0.5 ml of hypotonic lysing buffer (10 mM Tris buffer, 1 n&f EDTA, 0.2% Triton X-100, pH 7.5) and immediately centrifuged (13,OOOg, 10 min). The pellet (high molecular weight DNA) and the supernatant (low molecular weight DNA) were incubated overnight at 4’C in 0.5 ml of lysis buffer and 0.125 ml of 50% trichloroacetic acid and centrifuged at 13,000g for 4 min. The supernatants from both fractions were discarded and the DNA in the precipitates was hydrolyzed by heating to 90°C in 0.24 ml of 5% trichloroacetic acid. After one more 13,000g centrifugation, the supernatants were assayed calorimetrically. Following an overnight incubation with 0.48 ml of diphenylamine reagent, the absorbance was measured at 570 nm. The detergent-soluble DNA (DNA in the supernatant) was calculated as a percentage of the total DNA content (DNA in the supernatant and in the pellet). To assess the pattern of DNA cleavage, pulsed field and conventional agarose gel electrophoresis were used for the large and small fragments of DNA, respectively [21]. To prepare the DNA fractions pellets of 8 X lo6 cells were embedded in 1.5% low melting point agarose plugs, lysed, and digested for 18 h at 37°C in a solution containing 10 m&f NaCl, 10 mA4 Tris-HCl (pH 9.5), 25 m&f EDTA, 1% N-lauroylsarcosine, and 0.1 mg/ml proteinase K (Sigma). During this digestion fragments of DNA smaller than 20,000 bp diffuse out of the agarose plug [2I]. The supernatant, containing this low molecular weight DNA, was collected for the assay of detergent-soluble DNA as described below. The agarose plugs were washed for 4-6 h in TE buffer (10 n&f Tris-HCl [pH 81, 2 mM EDTA) at 4°C and finally suspended in 50 mM EDTA, pH 8.0. The plugs could be stored this way for several weeks. One 3-mm-long piece was then cut from the agarose block and subjected to pulsed field gel electrophoresis, as described in detail earlier [21]. Two sets of standards with overlapping molecular sizes were used: yeast chromosomes (250-1100 kbp from Bio-Rad Laboratories, Richmond, CA), and polymerized X phage DNA (50-700 kbp from Clontech Labs, Inc., Palo Alto, CA). Following electrophoresis the gels were stained with ethidium bromide and photographed in uv light with a Polaroid camera using Polaroid positive/negative film No. 55. To analyze the low molecular weight DNA contained in the detergent soluble fraction, the supernatant collected after incubation of the agarose plugs at 37°C described above was used. DNA precipitated from the supernatant with 2.5 M sodium acetate (pH 5.2) and absolute ethanol was resuspended in 100 pl TC buffer (10 mM Tris, pH 8, 1 ti CDTA [trans-1,2-diaminocyclohexane-N,N,N’,N’-tetraacetic acid]) and subjected to conventional gel electrophoresis, using 0.8% agarose gels and Tris-acetate buffer (0.04 M Tris acetate, pH 8.0, 0.002 M EDTA). The molecular weight standards were a Hind111 digest of X DNA and a 1-kb ladder (GIBCO-BRL). The results described in this paper were reproduced in at least three experiments, except for the DNA quantification experiments, which were performed only twice.

RESULTS

Effects of WI-26 treatment on stimulated lymphocytes. Figure 1 is a dose-response curve of the effects of VM26 on DNA synthesis and cell viability of Con-A-stimulated splenic lymphocytes; the results are expressed as percentage of control. Treating cells with doses of the

ROY ET AL.

418

I 90

n

75

45 30

j

60

15

5

0t

0

I I

I 2

1 3

CONCENTRATION OF VM-26

4

10 5

Treatment with VM-26 of Con A-stimulated S-phase lymphocytes. Incorporation of [‘H]thymidine and cell viability, as percentage of control, measured at 48 h post-Con A in cells treated for 2 h with various doses of VM-26 between 46 and 48 h of stimulation. Each point represents the mean of triplicates which did not differ by more than 17% for [3H]thymidine incorporation and by 18% for cell viability.

drug as high as 5 PM between 46 and 48 h of stimulation had no effect on lymphocyte viability. In contrast, these treatments strongly affected DNA synthesis. At 0.05 PM VM-26 inhibited DNA synthesis by 35% and concentrations of 1 PM or more inhibited synthesis by over 80%. These results are consistent with several earlier studies showing that DNA replication is particularly sensitive to topo II inhibitors [ll, 121. Effects of VM-26 pretreatment on unstimulated lymphocytes. In the course of our studies of VM-26 effects on Con A-stimulated lymphocytes we observed that VM-26 also had effects on the unstimulated cells. When unstimulated lymphocytes (G,,) were treated with VM-26 for 2 h, there was no immediate effect of the drug on cell viability. When such pretreated cells were then incubated with Con A in drug-free medium for 48 h, however, there was a severe loss of cell viability, and DNA synthesis in the surviving cells was also inhibited (Fig. 2). These results were surprising since the target of the inhibitor, topo II, has been reported to be absent or at barely detectable levels in unstimulated lymphocytes [ 16-201. This apparent contradiction prompted us to examine the effects of VM-26 on unstimulated lymphocytes in more detail. Time course of cell death in VM-26-treated lymphocytes. To characterize the mode of death induced by VM-26, we first followed a time course of cell viability in the drug-treated unstimulated lymphocytes (Fig. 3). As reported above, no effect on cell viability was seen at the end of the 2-h drug treatment. Following the drug treatment, however, the number of cells excluding trypan blue decreased to about 80% of the control after

e

0 I

0

(PM)

FIG. 1.

’ z i iii “s

2

3

CONCENTRATION

4

OF VM-26

5

(PM)

FIG. 2. Treatment with VM-26 of unstimulated lymphocytes. Unstimulated cells were pretreated with various doses of VM-26 and then Con A was added. Incorporation of [3H]thymidine andcell viability, as percentage of control, measured at 48 h. Each point represents the mean of triplicates which did not differ by more than 9% for [3H]thymidine incorporation and by 12% for cell viability.

2 h and to 40% of the control after 6 h of incubation in drug-free medium. This effect on viability was also seen when VM-26 was present continuously in the culture (data not shown) and was independent of the presence of Con A in the culture (Fig. 3). In contrast with this extended time course of VM-26-induced death, complement treatment, which induces necrosis [33], resulted in over 90% cell death within 1 h (data not shown). This

3.0 -

T

2.6 7 2 b

2.2 -

”

1.8-

Y g

1.4-

Y $

1.0 -

5 0.6 0.2 -

0

2

TIME AFTER V‘M-26 FIG. 3.

4

TREATMENT

6

(h)

Time course of cell death in VM-26-treated unstimulated lymphocytes. Viable cell counts determined at 0,2, 4, and 6 h following a 2-h treatment of unstimulated lymphocytes with 5 @ VM-26 are shown. Counts represent mean and standard deviation of triplicates. (13) -VM-26 - Con A, (B) -VM-26 + Con A, (8) +VM-26 Con A, (B) +VM-26 + Con A.

VM-26-INDUCED

APOPTOSIS

IN

28 r

W

$

20

3 8

16

2

TIME

AFTER

VM-26

4

TREATMENT

6

(h)

of fragmented DNA. The percentage of FIG. 4. Quantification DNA cleaved into detergent soluble fragments was determined at 0,2, 4, and 6 h after a 2-h treatment of unstimulated lymphocytes with 5 j&f VM-26. The drug-pretreated cells were incubated with or without Con A. (Cl) -VM-26 - Con A, @) +VM-26 + Con A, ( Con A, (m) +VM-26 + Con A.

suggested that the VM-26-induced death of unstimulated splenic lymphocytes, like that of rat thymocytes [21], might be occurring by apoptosis rather than by necrosis. DNA fragmentation. It is thought that the final common pathway of apoptosis involves the activation of an endogenous endonuclease that breaks DNA down to oligonucleosomal fragments [34, 351. To determine whether VM-26 induces apoptosis in unstimulated lymphocytes, DNA of the drug-treated cells was analyzed at 2-h intervals, following a 2-h treatment with the drug. Figure 4 shows that the amount of soluble DNA following VM-26 treatment was consistently greater than that in untreated cells. Between 4 and 6 h after VM-26 treatment, the drug-treated samples showed two to three times the control levels of detergent-soluble DNA whether or not Con A had been added to the cultures. The timing of this increase in fragmented DNA correlates with the decrease in cell viability determined by trypan blue exclusion (Fig. 3). Gel electrophoresis showed that the DNA was cleaved to oligonucleosome-sized fragments (Fig. 5). The cleavage was first detected in samples harvested 4 h after the drug treatment and was substantially increased in the 6-h samples. Addition of Con A to the cultures had no effect on the extent or pattern of DNA cleavage. Morphology of VM-26-treated cells. In addition to internucleosomal DNA cleavage, apoptosis is characterized by a sequence of morphological alterations of the nucleus and cytoplasm. In particular, in the early stages of apoptosis the nucleus shrinks, chromatin becomes condensed and marginated, and later the nucleus is

UNSTIMULATED

419

LYMPHOCYTES

fragmented ([22, 231 for reviews). As a further test of whether VM-26 induces apoptosis in unstimulated splenic lymphocytes, and to extend the morphological characterization of the process of apoptosis, we have examined the behavior of DNA and of certain nuclear and cytoplasmic components by immunofluorescence microscopy. All immunofluorescence samples were counterstained with the DNA-specific fluorochrome Hoechst 33258 and were also examined by phase contrast microscopy to assess changes in overall chromatin organization during apoptosis. Figures 6 to 10 show cells that were treated with VM-26 for 2 h and then incubated in drug-free medium for 4 h. Cells showing the apoptotic features of nuclear shrinkage and chromatin condensation (micrographs b’ and b” in Figs. 6, 7, 9), chromatin margination (Figs. 6b’, lob’), and nuclear fragmentation (b’ and b” in Figs. 8,lO) were evident in the Hoechst and phase contrast images. The distributions of certain nuclear and cytoplasmic components in these apoptotic cells were then examined. Nucleolar morphological changes have repeatedly been recognized as a characteristic of apoptosis ([34] for review). Breakdown and dispersal of nucleolar components have also been detected by electron microscopy of apoptotic rat thymocytes [22]. Figure 6 shows an example of immunofluorescence staining with the autoimmune serum Di (SC1 1) for the nucleolar antigen fibrillarin. This antiserum detects the M, 34,000 protein of the U3 RNP particle, which has been localized to the fibrillar region of the nucleolus [26]. In control unstimulated lymphocytes this antigen appears distributed as a spot or ring within each nucleolus (Fig. 6a). In apoptotic nuclei with marginated chromatin the distribution of this antigen did not appear to be altered; however, in

-‘con r-02

p,1

Ac”,lc6+lrSTD1

MW

--Con r-02

Al ~22

~42

rw6T

KBP 23.1 9.4 6.5 4.3

FIG. 5. Internucleosomal cleavage of DNA. Standard agarosegel electrophoresis of DNA isolated from control cells (-) and from cells treated with 5 pM VM-26 for 2 h (+) and then cultured for 0,2,4, and 6 h, in the presence or absence of Con A. Molecular weight standards (MW-STD) were: 1 kb standard and Hind111 digest of X DNA.

420

ROY

ET

AL.

Ubiquitin was chosen as an example of an antigen that might show increased expression during lymphocyte apoptosis. Ubiquitin participates in targeting proteins for removal from cells by proteolysis [36], and activation of the polyubiquitin gene has been observed during developmentally programmed cell death [37]. In the control lymphocytes ubiquitin was detected by immunofluorescence primarily in the nucleus with an interchromatinic distribution very similar to that for snRNPs detected by the Y12 antibody (Figs. 8a-8a”). In contrast, in the apoptotic lymphocytes there was bright staining of the cytoplasm and little or no ubiquitin was detected in regions of condensed chromatin. In addition to the extensive changes in nuclear morphology, reorganization of cytoskeletal elements has also been reported to occur in apoptotic cells [34]. Figures 9 and 10 show, respectively, the response of the vimentin intermediate filament and microtubule systems in apoptotic lymphocytes. In the unstimulated

FIG. 6. Untreated unstimulated lymphocytes (a-a”) and unstimulated lymphocytes pretreated with 5 pM VM-26 for 2 h (b-b”) were cultured for 4 h, immunolabeled with anti-nucleolar fibrillarin (a, b), counterstained with Hoechst (a’, b’), and examined by phase contrast microscopy (a”, b”). Apoptotic cells are indicated by arrows and the nuclei with marginated chromatin by arrowheads. Bar, 5 pm.

cells with shrunken nuclei the antigen appeared uniformly distributed throughout the nucleus (Fig. 6b). As a marker for changes in the organization of the interchromatinic regions of apoptotic nuclei we have used the monoclonal antibody Y12, which recognizes antigens common to the Ul, U2, U4, U5, and U6 snRNPs [25]. In unstimulated control lymphocytes this antibody stains a network of patches or granules that are excluded from the nucleoli and from regions of aggregated chromatin (Figs. 7a-7a” and Ref. [29]). In the apoptotic nuclei of the VM-26-treated lymphocytes it appeared that first the network became disorganized and diffuse and at a later stage the antigen was excluded from the nucleus and was detected in the cytoplasm (Fig. 7b). Later still, the cytoplasmic staining was much less intense, reflecting perhaps antigen degradation in the dying cells.

FIG. 7. Immunofluorescence staining of snRNP (a, b), Hoechst staining (a’, b’), and phase contrast (a”, b”) of cells treated as in Fig. 6. Apoptotic cells are indicated by arrows and the nuclei with marginated chromatin by arrowheads. Bar, 5 pm.

VM-26INDUCED

APOPTOSIS

IN

UNSTIMULATED

LYMPHOCYTES

421

treatment with 5 PLM VM-26, large fragments of DNA, with a peak at about 300 kbp, had accumulated. The band appearing at 800 kbp results from the comigration of larger DNA fragments that are not resolved by these electrophoresis conditions. The 300- and >860-kbp bands are not seen after the drug is removed since in the absence of the drug the topo II molecules reseal the DNA. However, some DNA damage persists and by 4-6 h after the VM-26 treatment, an accumulation of low molecular weight DNA fragments (Fig. 5), which is typical of cells undergoing apoptosis, as well as fragments of 50 kbp was detected (Fig. 11). DISCUSSION

The two major apoptosis, differ in ated, the former by DNA fragmentation

FIG. Hoechst Apoptotic

8. Immunofluorescence (a’, b’), and phase contrast cell indicated by arrows.

modes of cell death, necrosis and the way the two processes are initimembrane damage and the latter by ([23,34] for reviews). Necrosis, in-

staining of ubiquitin (a, b), (a”, b”) of cells treated as in Fig. 6. Bar, 5 grn.

control lymphocytes vimentin appeared as a filamentous network that was partially coincident with the radial pattern of microtubules (Figs. 9a, 10a and Ref. [38]). In the apoptotic cells the vimentin no longer appeared filamentous and a diffuse staining was seen throughout the cytoplasm (Figs. 9b-9b”). The microtubule system also was sensitive in the apoptotic cells. Some microtubules were present in cells with marginated chromatin, but they were completely disassembled when the nuclei appeared fragmented (Figs. lob-lob”). Mechanism of VM-26-induced apoptosis. Pulsed field gel electrophoresis has been used to show that the topo II-specific drug VM-26 induces the cleavage of DNA into large fragments believed to represent chromatin loop domains [ 14,21,39]. To determine if VM-26 induced similar cleavage of DNA in the present study, pulsed field electrophoresis of the DNA of drug-treated unstimulated lymphocytes was also carried out. Figure 11 shows that in samples taken immediately after a 2 h

FI(Yg. staining Apoptotic

Immunofl,-~urescence --------- -A-:-:-YC~IIIIIIL: UI vimenzin (a, D), Hoechst (a’, b’), and ph---RQP ___.-_ contrast (a”, b”) of cells treated as in Fig. 6. cells indicated by arr ows. Bar, 5 pm.

422

ROY

ET

AL.

[42], and tumor necrosis factor [43]. Recently we reported the induction of apoptosis by the topo II inhibitor VM-26 in topo II-rich thymocytes [21]. Now we present data to show that VM-26 also induces apoptosis in unstimulated murine splenic lymphocytes. The time course of cell death after VM-26 treatment of unstimulated lymphocytes indicated that apoptosis, rather than necrosis, was induced. There was little loss of cell membrane integrity (80% of the cells excluded trypan blue) even 2 h after the removal of the VM-26 from the cell culture (Fig. 3). In contrast, in necrotic cells produced by Thy 1 and complement treatment of T lymphocytes [33], uptake of trypan blue occurred immediately after the treatment (data not shown), in keeping with the well-established criterion that membrane damage initiates the process of necrosis [34,42]. The most widely adopted characteristic defining apoptosis is the cleavage of DNA into oligonucleosomesized fragments [34, 35, 421. Topo II inhibitors have been reported to induce such endonucleolytic cleavage of DNA in a number of proliferating cell types [21, 44, 451, including Con A-stimulated murine lymphocytes [19]. Our data show that VM-26 induces the same pattern of DNA fragmentation in unstimulated lymphoq&es (Fig. 5). The shrinkage of the nucleus and the nargination and condensation of chromatin that accom-

TIME(h

t VM-2E kb

FIG. 10. Immunofluorescence staining of tubulin (a, b), Hoecbst (a’, b’), and phase contrast (a”, b”). Apoptotic cells are indicated by arrows and the nuclei with marginated chromatin by arrowheads. Bar, 5 Wm.

700-

duced by complement injury, ischemia, or toxic injury, leads to increased permeability of the plasma membrane, osmotic swelling, and finally to total dissolution of the organized structure of the cell, accompanied by a release of intracellular enzymes into the circulation. Apoptosis follows an energy-dependent programmed pattern of condensation and fragmentation of the nucleus and repackaging of cytoplasm and organelles within a sealed plasma membrane, so that there is no leakage of cellular components into the cellular microenvironment. In the final stage, the apoptotic cell is fragmented into apoptotic bodies which are engulfed by the neighboring cells. These features are common to cells undergoing apoptosis irrespective of how this death process is induced. Apoptosis can occur under normal physiological conditions, especially during morphogenesis and atrophy ([23,34] for reviews), or it may be induced by various external agents, such as glucocorticoids [34, 351, heat [40], radiation [41], lymphotoxins

290-

104.458-

FIG. 11. Cleavage of supercoiled DNA. Pulsed field gel electrophoresis of DNA isolated from control cells (-) and from cells treated with 5 PM VM-26 for 2 h (+) and then cultured in drug-free medium for 0, 2, 4, and 6 h. Molecular weight standards (MW-STD) were yeast chromosomal DNA and polymerized X DNA.

VM-26-INDUCED

APOPTOSIS

IN

pany the double-stranded DNA cleavage into oligonucleosomal fragments ([22, 23, 341 for reviews) was also observed in VM-26-treated unstimulated lymphocytes (Figs. 6-10). Compared to chromatin, the nucleolar protein fibrillarin [26] appeared to be relatively resistant to apoptotic changes. It appeared unaltered when chromatin was condensed and marginated (Figs. 6b-6b”), but at a late stage of apoptosis fibrillarin was dispersed as had been observed by Arends et al. [22] in apoptotic rat thymocytes. The changes in chromatin architecture were accompanied by changes in the distribution of interchromatinic proteins. The snRNP antigens recognized by the antibody Y12 [25] were reorganized early in apoptosis and then appeared to be dispersed to the cytoplasm (Figs. 7b-7b”). Ubiquitin, which in control cells appeared as a nuclear antigen with the same pattern of distribution as the Y12 antigen, also appeared to be dispersed to the cytoplasm when the nucleus was condensed and fragmented (Fig. 8). This loss of nuclear proteins may be a consequence of proteolysis occurring during apoptosis. Kaufman [45] has described the degradation of several nuclear proteins (poly(ADP-ribose) polymerase, lamin B, topoisomerases I and II, and histone Hl) that occurred concomitant with DNA fragmentation in HL-60 cells treated with a variety of cytotoxic drugs including topo II inhibitors. In addition to the redistribution of the nuclear ubiquitin, we also observed a large increase in the staining intensity of cytoplasmic ubiquitin in apoptotic lymphocytes. Overexpression of polyubiquitin genes has been reported to occur during apoptosis [37], when it may participate in removal of degraded proteins from dying cells. Changes in the distribution of cytoskeletal components, in particular of intermediate filament proteins, have also been described in apoptosis ([23, 341 for reviews). Similarly, in the apoptotic lymphocytes the distribution of the vimentin intermediate filament system was extensively altered (Fig. 9). In addition, microtubules were disassembled in the apoptotic cells (Fig. 10). The microtubules did not appear to be affected as early in apoptosis as the nuclear organization. Microtubules were only partially disassembled in cells in which the chromatin was marginated and the snRNP staining pattern was obviously disorganized. All of our observations lead us to conclude that a 2-h treatment with 5 pM VM-26 induces apoptosis in unstimulated mouse splenic lymphocytes and inhibits the response to the mitogen Con A (Fig. 2). We should note, however, that it has been reported that a 2-h treatment of human peripheral blood lymphocytes with 5 to 20 pM VM-26 did not affect the ability of the cells to respond to the mitogen PHA and to incorporate [3H]thymidine 1201. To explain this discrepancy in results we can only suggest that there may be a species difference in sensitivity to VM-26 or in the ability to recover from VM-26 damage.

UNSTIMULATED

LYMPHOCYTES

423

We further conclude that VM-264nduced apoptosis in unstimulated lymphocytes is mediated by topo II, although an additional intracellular target for this drug cannot be definitely ruled out. Studies using VM-26-resistant mutant cells have established that topo II is the specific target of the drug [46, 471. In addition, pulsed field gel electrophoresis of DNA from unstimulated lymphocytes treated with VM-26 shows a pattern of DNA cleavage that is characteristic of an interaction between topo II and VM-26 (Fig. ll), namely, the generation of DNA fragments representing 300-kb rosettes and 50-kb chromatin loops [14, 21, 391. The apparent discrepancy between our present results and those of previous studies showing that topo II was present at low to barely detectable levels in unstimulated lymphocytes [ 16-201 may be due to the presence of a second form of topo II. Chung et al. [48] have recently identified two isozymes of topo II in Raji cells, a 170-kDa form and a 180-kDa form, that are coded by separate genes. In NIH-3T3 cells synchronized by serum starvation and then restimulated to enter the cell cycle by addition of fresh growth medium, the amount of the 170-kDa form was undetectable until the cells reached late S phase, peaked in G,-M phase cells, and decreased as the cells completed mitosis [49]. The 180kDa form, on the other hand, was detected at all phases of the cell cycle, including the G, phase. Woessner et al. [49] have also shown that a higher salt concentration is required for the extraction of 180-kDa topo II than for that of the 170-kDa form of the enzyme. They have suggested that this may indicate that the enzyme is located at relatively inaccessible sites, such as the bases of the chromatin loops which are attached to the nuclear matrix. It is possible that the 180-kDa form of topo II is present in unstimulated lymphocytes and that it was not detected in the earlier studies [ 16-201. This suggestion is supported by our observation of an approximately threefold greater sensitivity to VM-26 of stimulated (Fig. 1) versus unstimulated (Fig. 2) lymphocytes. Drake et al. [50] have shown that the 180-kDa form of topo II is three times less sensitive to VM-26 than the 170-kDa form. In summary, our results show that VM-26 cleaves supercoiled DNA and induces apoptosis in unstimulated murine lymphocytes. We speculate that the drug interacts with the 180-kDa form of topo II in the unstimulated lymphocyte and may inhibit both forms of topo II in stimulated lymphocytes. We are currently analyzing the expression and functions of the two forms of topo II in the different stages of lymphocyte activation. We thank Dr. France Guay, Bristol-Myers Squibb Pharmaceutical Group, St. Laurent, P.Q., for providing VM-26. We are grateful to Drs. V. I. Kalnins (University of Toronto), T. Owens (McGill University), and G. Beimer (Bahnhofstr) for their gifts of the antibodies to vimentin, Thy 1.2 (30H12.4), and fibrillarin (serum Di), respectively. We thank also Dr. J. Steitz (Yale University) for generously supply-

ROY ing the hybridoma Y12 which produces the antibody recognizing snRNPs. This project was funded by a grant from the Medical Research Council of Canada to D. L. Brown and N. Chaly.

ET

AL. 24.

Ledbetter, 47,63-90.

25.

Lerner, (1981)

26.

Reimer, G., Pollard, K. M., Penning, C. A., Ochs, R. L., Lischwe, M. A., Busch, H., and Tan, E. M. (1987) Arth. Rheum. 30,793800. Federoff, S., White, R., Neal, J., Subrahmanyan, L., and Kalnins, V. I. (1983) Dev. Brain Res. 7,303-315. Aitchison, W. A., and Brown, D. L. (1986) Cell Motil. Cytoskel. 6, 122-127.

REFERENCES 1.

Vosberg, 102.

2.

Wang,

3.

Nelson, 187-189.

4.

Berrios, M., Osheroff, N., and Fisher, Acad. Sci. USA 82,4142-4146.

P. A. (1985)

Pienta,

D. S. (1989)

5.

H-P.

(1985)

J. C. (1987)

Curr.

Biochim.

Top. Microbial. Biophys.

W. G., Liu, L. F., and Coffey,

K., Partin,

A. W., and Coffey,

114, 19-

Immunol.

Acta 909,

28.

l-9.

D. S. (1986)

Nature

322,

Proc.

Natl.

Cancer

Res.

Cell Res.

192,

19,2525-2532. 6. 7.

Kaufman, 511-523. Cockerill,

8.

Gasser, 22.

9.

Gasser, J. F-X. Gairard,

10.

S. H., and Shaper, P. N., and Garrard, S. M.,

and Laemmli,

J. H. (1991)

Ezp.

W. T. (1986)

Cell 44,

U. K. (1987)

Trends

273-282.

Genet.

3,16-

S. M., Amati, B. B., Cardenas, M. E., and Hoffman, (1989) Znt. Rev Cytol. 119, 57-96. W. T. (1990) Nucleic Acids Mol. Biol. 4, 163-175.

11.

D’Arpa, 163-177.

12.

Liu,

13. 14.

Newport, J., and Spann, T. (1987) Cell 48,205-217. Roberge, M., Th’ng, J., Hamaguchi, J., and Bradbury, (1990) J. Cell Biol. 111, 1753-1762.

15.

Brown, D. L., Dudani, R., Little, J. E., Valentine, B. K., Chaly, N., and Sahai, B. M. (1990) in Biochemical and Structural Dynamics of the Cell Nucleus (Wang, E., Wang, J. L., Chien, S., Cheung, W-Y., and Wu, C-W., Eds.), pp. 161-170, Academic Press, San Diego.

16. 17.

P., and Liu,

L. F. (1989)

Annu.

L. F. (1989)

Biochim.

Rev. Biochem.

58,

Biophys.

Acta 989,

19. 20. 21. 22. 23.

30.

Roy, C., Prasad, K. V. S., Reuhl, K. R., Little, B. K., and Brown, D. L. (1991) Exp. Cell Res.

31. 32. 33. 34. 35.

J. G., and Brown,

D. L. (1988)

J. E., Valentine,

195,345-352.

Burton, K. (1956) Biochem. J. 62,315-322. Newell, M. K., Haughn, L. J., Maroun, C. R., and Julius, M. H. (1990) Nature 347, 286-289. Hawkins, H. L., Ericsson, J. L. E., Biberfeld, P., and Benjamin, F. T. (1972) Am. J. Pathol. 68, 255-288. Wyllie, A. H. (1987) Znt. Rev. Cytol. Suppl. 17, 755-785. Cohen, J. J., and Duke, R. C. (1984) J. Immunol. 132, 38-42.

37.

Schwarz, L. M., Myer, A., Kosz, L., Engelstein, M., and Maier, C. (1990) Neuron 5,411-419. Paulin-Levasseur, M., and Brown, D. L. (1987) Cell Motil. Cytoskel. 8, 227-237. Filipski, J., Leblanc, J., Youdale, T., Sikorska, M., and Walker, P. R. (1990) EMBO J. 9, 1319-1327. Harmon, B. V., Corder, A. M., Collins, R. J., Gobe, G. C., Allan, D. J., and Kerr, J. F. R. (1990) Znt. J. Radiat. Biol. 58,845-858. Sellins, K. S., and Cohen, J. J. (1987) J. Zmmunol. 139, 31993206. Schmid, D. S., Tite, J. P., and Ruddle, N. H. (1986) Proc. Natl. Acad. Sci. USA 83, 1881-1885. Laster, S. M., Wood, J. G., and Gooding, L. R. (1988) J. Zmmunol. 141,2629-2634.

39. 40. 41. 42. 43. 44.

Jaxel, C., Taudou, G., Portemer, C., Mirambeau, G., Panijel, J., and Duguet, M. (1988). Biochemistry 27,95-99. Hwang, J., Shyy, S., Chen, A. Y., Juan, C-C., and Whang-Peng, J. (1989) Cancer Res. 49,958-962. Walker, P. R., Smith, C., Youdale, T., Leblanc, J., Whitfield, J. F., and Sikorska, M. (1991) Cancer Res. 51, 1078-1085. Arends, M. J., Morris, R. G., and Wyllie, A. H. (1990) Am. J. Path01 136,593-608. Bursch, W., Kleine, L., and Tenniswood, M. (1990) Biochem. Cell Biol. 68,1071-1074.

47.

6, 1992

Kaplan,

J. A.

Mayer, (1989)

45. 46.

December 3,199l version received February

M.,

Rev.

36.

Kaplan, J. G., Brown, D. L., Chaly, C., Greer, W. L., Prasad, K. V. S., Severini, A., and Sahai, B. M. (1987) J. Mol. Evol. 26, 1-5.

Received Revised

Zmmunol.

E. A., Lerner, M. R., Janeway, C. A., Jr., and Steitz, Proc. Natl. Acad. Sci. USA 78, 2737-2741.

Chaly, N., Cadrin, Biol. Cell 63, 9-17.

E. M.

Taudou, G., Mirambeau, C. L., der Garabedian, A., Vermeersch, J., and Duguet, M. (1990) FEBS L&t. 176,431-435. Heck, M. M. S., and Earnshaw, W. C. (1986) J. Cell Biol. 103,

L. A. (1979)

29.

38.

351-357.

2569-2581. 18.

27.

J. A., and Herzenberg,

48.

49. 50.

A., Siegel, N. R., Schwarz, Science 244, 1480-1483.

A. L., and

Ciechanover,

Villeponreau, B., Pribyl, T. M., Grant, M. H., and H. G. (1986) J. Biol. Chem. 261, 10,359-10,365. Kaufman, S. H. (1989) Cancer Res. 49,5870-5878.

A.

Martinson,

Sullivan, D. M., Latham, M. D., Rowe, T. C., and Ross, W. E. (1989) Biochemistry 28,5608-5687. Fernandes, D. J., Danks, M. K., and Beck, W. T. (1990) Biochemistry 29,4235-4241. Chung, T. D. Y., Drake, F. H., Tan, K. B., Per, S. R., Crooke, S. T., and Mirabelli, C. K. (1989) Proc. Natl. Acad. Sci. USA 86, 9431-9435. Woessner, R. D., Mattern, M. R., Mirabelli, C. K., Johnson, R. K., and Drake, F. H. (1991) Cell Growth Differ. 2,209-214. Drake, F. H., Hofmann, G. A., Bartus, H. F., Mattern, M. R., Crooke, S. T., and Mirabelli, C. K. (1989) Biochemistry 28, 8154-8160.