| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Department of Haematology, University of Liverpool, Liverpool L69 3GA, United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
It is generally accepted that the killing of resting lymphocytes by purine analogues requires nucleoside phosphorylation (2) , involves the accumulation of nucleoside-induced DNA breaks (2) , and occurs by apoptosis (3) . However, it remains unclear exactly how cell death is triggered.
In some (4 , 5) but not other (6) studies of nucleoside cytotoxicity in CLL, activation of the tumor suppressor protein p53 has been implicated as a link between nucleoside-induced DNA breaks and subsequent killing. Recent work from this laboratory has reconciled these apparently contradictory findings by demonstrating that the nucleoside-induced killing of mouse splenocytes (7) and CLL cells (8) can occur by both p53-dependent and -independent mechanisms.
Regarding possible p53-independent mechanisms of cell killing, activation of PARP has previously been implicated in nucleoside cytotoxicity (9) . Before the recent demonstration that cells from PARP-knockout mice can synthesize poly(ADP-ribose) (10) , it was assumed that PARP was a single gene product. However, several structurally unrelated PARP enzymes have now been identified (11) . Among these, the original PARP (now known as PARP-1) is by far the most abundant and well characterized (11) . This nuclear enzyme, in addition to being a target for proteolytic cleavage during apoptosis (12) , is thought to play a role in DNA repair by signaling genomic damage (13) . Thus, PARP-1 binds to DNA breaks, becomes activated, and, using NAD+ as a substrate, adds ADP-ribose polymers to a range of nuclear proteins including itself, histones, and DNA repair enzymes (11 , 14) .
The idea that poly(ADP-ribosyl)ation might be involved in the cytotoxic action of purine analogues is based on work showing that the nucleoside-induced killing of normal resting lymphocytes (detected as loss of vital dye exclusion) was preceded by NAD+/ATP depletion and that these events were prevented by the PARP inhibitor 3AB (9) . These and other findings have led to a proposed model of nucleoside action involving the following sequence of events: (a) inhibition of ongoing DNA repair; (b) accumulation of DNA breaks; (c) PARP activation; (d) NAD+ consumption; and (e) ATP depletion (15) .
However, there are several problems with this explanation of nucleoside cytotoxicity. For example, ATP depletion is associated with cellular necrosis (16 , 17) , whereas purine analogues induce apoptosis (3) . In addition, PARP-mediated NAD+ depletion is known to occur as a consequence of apoptosis (12) . Given that the cell membrane requires ATP for many of its functions, it seems likely that PARP-mediated NAD+/ATP depletion might contribute to the cell membrane disruption that occurs during the late stages of apoptosis (18) . If so, the previously demonstrated inhibitory effect of 3AB on nucleoside-induced cell membrane disruption (9) might be confined to cells that have already undergone apoptosis. To complicate matters further, in these early experiments, 3AB was used at a high concentration (5 mM) now known to inhibit mono(ADP-ribosyl)ation (19) .
Whereas the role of poly(ADP-ribosyl)ation in the nucleoside-induced killing of normal resting lymphocytes is unclear, its role in the nucleoside-induced killing of CLL cells is completely unknown. It was therefore the aim of the present study to address this important question.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
Cell Culture
Mononuclear cells were prepared from whole blood by
centrifugation over Lymphoprep (Life Technologies, Inc., Paisley,
United Kingdom) and cultured at 37°C in RPMI + 1% BSA in
the presence of 5% CO2. Culture vessels were
precoated with poly(2-hydroxyethyl methacrylate) (Sigma, Poole, Dorset,
United Kingdom) to prevent cell adhesion (20)
.
Pharmacological Agents
The purine analogues CdA and
9-ß-D-arabinosyl-2-fluoroadenine (fludarabine)
monophosphate were kind gifts from Janssen-Cilag (High Wycombe,
Buckinghamshire, United Kingdom) and Schering (Burgess Hill, West
Sussex, United Kingdom), respectively. CdA and fludarabine were used at
the pharmacologically relevant concentrations of 0.2 and 2.0
µM, respectively (21
, 22)
. These low
nucleoside concentrations are not acutely toxic to CLL cells but
instead produce a delayed killing response (8
, 23) .
The PARP inhibitors 3AB, 4AN, and 2NP were obtained from Sigma. 3AB was used at a concentration (200 µM) known to inhibit poly(ADP-ribosyl)ation but not mono(ADP-ribosyl)ation in intact cells (19) , whereas 4AN and 2NP were used at approximately 1, 10, and 100 times their respective IC50 values for PARP (24) .
Measurement of Cell Killing
Loss of Membrane Integrity.
Cultured cells were gently resuspended and added to an equal volume of
PBS containing 10 µg/ml PI (Sigma). After incubating the cells on ice
for 10 min, cells were analyzed by flow cytometry. Cells with an intact
plasma membrane do not stain with PI (a red DNA-binding fluorochrome).
In contrast, cells that have lost their membrane integrity take up the
fluorochrome and therefore fluoresce bright red (25)
.
Cell Shrinkage.
Cell death can be detected flow cytometrically as a change in FSC and
SSC. Thus, dead cells have a lower FSC and higher SSC than do live
cells (25)
.
Mitochondrial Depolarization.
Cultured cells (50 µl) were gently resuspended and added to 150 µl
of PBS containing 40 nM DiOC6
(Sigma). After a 15-min incubation at 37°C, the cell suspension was
added to an equal volume of PBS containing 10 µg/ml PI. After an
additional 30 min of incubation on ice, cells were analyzed by flow
cytometry. DiOC6 is a cell-permeable green
fluorochrome that is selectively taken up by charged mitochondria and
therefore stains live cells but not apoptotic cells (26
, 27)
. Dual staining with PI enables the identification of early
apoptotic cells that have undergone mitochondrial depolarization but
have not yet lost their membrane integrity (25)
.
Exposure of PS.
Cultured cells were gently washed in PBS and resuspended in annexin
V-FITC (PharMingen, Cowley, Oxford, United Kingdom) diluted 1:20 in a
buffer consisting of 10 mM HEPES/sodium hydroxide (pH 7.4),
140 mM sodium chloride, and 2.5 mM calcium
chloride. After a 15-min incubation at room temperature, the
cell suspension was diluted in 300 µl of the same buffer containing
10 µg/ml PI. After an additional 30 min of incubation on ice, cells
were analyzed by flow cytometry. Annexin V binds specifically to PS, a
phospholipid that becomes exposed on the surface of cells undergoing
apoptosis (28)
. Apoptotic cells therefore bind
FITC-labeled annexin V and, as a result, fluoresce bright green. Dual
staining with PI enables the identification of early apoptotic cells
that have not yet lost their membrane integrity (25)
.
DNA Fragmentation.
This was determined by the method of Nicoletti et al.
(29)
. Briefly, 50 µl of CLL cells were gently
centrifuged and resuspended in 400 µl of a solution containing 0.1%
Triton X-100, 0.1% citrate, and 10 µg/ml PI. Cells were incubated on
ice for at least 60 min before being analyzed by flow cytometry. During
apoptosis, endonuclease activity produces low molecular weight DNA
(30)
that is lost from the cell after permeabilization.
Permeabilized apoptotic cells therefore have a lower residual DNA
content and stain less intensely with PI than do live or necrotic
cells.
Morphological Analysis.
Cytocentrifuge preparations of cultured CLL cells were stained with
May-Grunwald-Giemsa and examined microscopically.
PARP-1 Cleavage.
Cultured cells (2.5 x 106)
were gently washed in PBS and lysed in 100 µl of buffer containing
62.5 mM Tris-HCl (pH 6.8), 2% SDS, 6 M urea,
5% ß-mercaptoethanol, 10% glycerol, and 0.00125% bromphenol blue.
Lysates were sonicated for 15 s and heated at 65°C for 15 min
before being subjected to SDS-PAGE and Western blotting. Membranes were
sequentially reacted with an anti-PARP-1 mouse monoclonal antibody
(clone A6.4.12; Insight Biotechnology Ltd., Wembley, Middlesex, United
Kingdom) and a peroxidase-conjugated antimouse second layer antibody
(Transduction Laboratories, Lexington, KY). The reactive protein bands
were visualized using the enhanced chemiluminescence system (Amersham,
Buckinghamshire, United Kingdom). During apoptosis, intact
Mr 113,000 PARP-1 is cleaved by
caspases, producing a characteristic
Mr 89,000 COOH-terminal fragment
(11
, 12)
.
|
Determination of p53 Status |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
). As expected, CdA and fludarabine induced extensive killing, as
detected by both parameters (Tables 1
2)
. 3AB delayed nucleoside-induced membrane disruption in all six cases
(Table 1)
but inhibited cell shrinkage only in case 5 (Table 2)
.
Because it appeared that 3AB was producing two different effects, case
5 and two of the other cases (cases 2 and 4) were examined in more
detail to further define these two effects. To do this, cell death was
measured using several additional parameters (Figs. 2
3
).
|
|
|
|
|
Effect of 3AB on Other Parameters of Nucleoside-induced
Killing.
In cases 2 and 4 (Fig. 2
), fludarabine-induced cell membrane disruption
(increased PI staining) was accompanied by extensive mitochondrial
depolarization (decreased DiOC6 staining) and
exposure of PS (increased annexin staining). Importantly, the number of
early apoptotic (DiOC6-dim/PI-dim and
annexin-bright/PI-dim) cells as well as late apoptotic/necrotic
(PI-bright) cells was increased. Fludarabine also induced extensive DNA
fragmentation (decreased PI staining of permeabilized cells) and PARP-1
cleavage, confirming that cell death was occurring by apoptosis. The
majority of fludarabine-treated cells had a smeared morphology
reflecting cell membrane disruption.
As expected, 3AB produced a marked decrease in the number of late apoptotic/necrotic (PI-bright) cells and a corresponding increase in the number of early apoptotic (DiOC6-dim/PI-dim and annexin-bright/PI-dim) cells. Importantly, the inhibitor did not increase the proportion of live (DiOC6-bright/annexin-dim) cells. Furthermore, there was no diminution in the extent of DNA fragmentation or PARP-1 cleavage. The morphology, although less smeared, was apoptotic rather than live. Taken together, these findings clearly indicate that 3AB was delaying the membrane disruption of apoptotic cells without affecting the induction of killing.
Case 5 differed from cases 2 and 4 in two major respects. First,
although fludarabine-induced cell membrane disruption was again
accompanied by extensive mitochondrial depolarization, exposure of PS,
and cell smearing, the extent of DNA fragmentation and PARP-1 cleavage
(two hallmarks of apoptosis) did not reflect the extent of killing as
determined by the other parameters measured (Fig. 3
). This
indicates that killing was occurring in part by necrosis. Secondly, 3AB
had a pronounced inhibitory effect on all manifestations of
fludarabine-induced killing (Fig. 3
). Similar results were obtained for
CdA (Fig. 4
).
|
). Importantly, both of these inhibitory effects were demonstrable at
concentrations of 4AN and 2NP close to their IC50
values for PARP and considerably lower than their
IC50 values for MART. These experiments therefore
confirm that the inhibitory effects observed in the present study are
due to inhibition of poly(ADP-ribosyl)ation rather than inhibition of
MART activity.
|
Frequency of PARP-mediated Killing.
To determine the frequency of PARP-mediated killing in CLL, the effect
of 3AB on nucleoside cytotoxicity (as measured by the
PI/DiOC6 method) was examined in 24 additional
cases. As expected, 3AB inhibited nucleoside-induced cell membrane
disruption (Fig. 6A
). However, the inhibitor did not prevent mitochondrial
depolarization in any of these additional cases (Fig. 6B
).
Therefore, poly(ADP-ribosyl)ation made a major contribution to
nucleoside-induced killing in only 1 of the 30 cases of CLL studied.
|
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
Although p53 gene defects in CLL predict for poor survival and failure to respond clinically to purine analogues (31 , 32) , recent work from this laboratory (8) and elsewhere (6) has shown that the in vitro killing of CLL cells by nucleosides can be p53 independent. Because PARP-mediated NAD+/ATP depletion has been implicated in the nucleoside-induced killing of normal resting lymphocytes (9 , 15) , it seemed important to determine whether or not this enzymatic activity contributed to the cytotoxic action of purine analogues in CLL. To do this, we used 3AB at a concentration known to produce selective inhibition of poly(ADP-ribosyl)ation in intact cells (19) and measured cell death by a number of different methods.
In keeping with its previously reported effects on normal lymphocytes (9) , 3AB delayed nucleoside-induced killing (as detected by loss of vital dye exclusion) in the great majority of CLL cases studied, indicating that PARP activity is indeed involved in such killing.
However, when cell death was examined in more detail, it became clear that, in most cases of CLL, the major effect of 3AB was to delay the loss of membrane integrity of cells that had already undergone apoptosis. This indicates that PARP activity was accelerating the transition of cells from the early stage to the late stage of apoptosis (18) but was not contributing to the induction of this process. This observation illustrates the importance of using the appropriate analytical techniques in experiments involving inhibition of cell killing and calls into question the results of previous studies involving PARP inhibition in which cytotoxicity was detected as loss of vital dye exclusion.
Importantly, we identified one case of CLL in which cell death occurred by a mixture of apoptosis and necrosis and in which all manifestations of nucleoside-induced killing were dramatically inhibited by 3AB. Interestingly, this was the only case among the 30 studied in which a biallelic p53 gene defect was detected. This indicates that poly(ADP-ribosyl)ation can occasionally be central to nucleoside-induced killing and that such PARP-mediated killing is p53 independent. Our demonstration that this type of killing is largely non-apoptotic is consistent with the notion that it is mediated by ATP depletion, an event known to predispose cells to a necrotic mode of cell death (16 , 17) .
It is highly unlikely that the inhibitory effects of 3AB were due to inhibition of MART, given that 3AB inhibits this enzyme with an IC50 that is 20-fold higher than the concentration used in the present study (19) . However, to exclude this possibility, we used two other PARP inhibitors, 4AN and 2NP. Both compounds produced a pattern of inhibition identical to that of 3AB. The inhibitory effects of 4AN and 2NP were not due to inhibition of MART because they were demonstrable at inhibitor concentrations 1000-fold (4AN) or 200-fold (2NP) lower than the IC50 values for this enzyme (24) .
It is unclear which specific enzymes are responsible for mediating the PARP-dependent events demonstrated in the present study. However, owing to its relative abundance and known propensity for activation by DNA breaks (11) , PARP-1 is likely to be the major enzyme involved in PARP-dependent killing and cell membrane disruption. Nevertheless, the possibility that other PARP enzymes might contribute to either or both of these events cannot be excluded.
PARP activation has been implicated in the cytotoxic action of
DNA-damaging agents other than purine analogues. For example, 3AB has
been shown to antagonize the cytotoxic effects of free radicals in
myeloid leukemia cells (33)
, neurones (34)
,
and pancreatic islet cells (35)
. Indeed, PARP-1 knockout
mice are highly resistant to diabetes induced by the ß-cell-specific
genotoxic drug streptozocin (36)
. In contrast, thymocytes
from such animals undergo a normal apoptotic response to
-irradiation (37)
.
Therefore, whether or not DNA damage results in PARP-mediated cytotoxicity clearly depends on the cell type and/or the nature of the genotoxic agent. Because PARP-mediated NAD+/ATP depletion is likely to require sustained PARP activation, this mechanism of killing may be restricted to genotoxic agents that induce persistent rather than transient DNA breaks. Furthermore, PARP-mediated killing is likely to be most prominent in cells in which other cytotoxic DNA damage response pathways (e.g., p53 activation) are blocked.
Taken together with previous work, our findings in CLL cells are consistent with a model of purine analogue cytotoxicity in which unrepaired DNA breaks result in the activation of two potentially cytotoxic pathways: (a) p53-mediated apoptosis; and (b) PARPmediated NAD+/ATP depletion. The fact that p53 dysfunction is associated with delayed nucleoside-induced killing (7 , 8) suggests that p53-mediated apoptosis is the most rapid cytotoxic pathway activated by purine analogues and that PARP-mediated killing is therefore only likely to occur in cells with p53 dysfunction.
However, we have demonstrated in the present study that PARP activity is not always responsible for the nucleoside-induced killing of CLL cells with p53 dysfunction (detected as an impaired p53/p21 response to ionizing radiation). There are, we believe, two possible explanations for this observation. First, nucleosides might constitute a more potent stimulus for p53 activation than radiation. This seems plausible, given that purine analogues induce the progressive accumulation of DNA breaks (9) , whereas radiation-induced DNA breaks are transient (38) . Purine analogues might therefore be capable of triggering p53-mediated apoptosis in cases of CLL that are refractory to radiation-induced p53 activation but contain some wild-type p53 protein. Alternatively, nucleosides might activate a third cytotoxic pathway that kills cell more slowly than p53-mediated apoptosis but more quickly than PARP-mediated NAD+/ATP depletion. PARP-mediated killing would then only occur in cells in which the other two pathways were blocked. Further work is in progress to identify which of these possibilities is correct.
In conclusion, the present study has identified two roles for PARP in the nucleoside-induced killing of CLL cells: (a) as a p53-independent mechanism of cell killing responsible for cytotoxicity in occasional cases; and (b) as a mediator of cell membrane disruption during the late stages of apoptosis.
|
FOOTNOTES |
|---|
1 Supported by the Leukaemia Research Fund (United
Kingdom) and the North West Cancer Research Fund. 
2 To whom requests for reprints should be
addressed, at Department of Haematology, University of Liverpool,
Liverpool L69 3GA, United Kingdom.
3 The abbreviations used are: CLL, chronic
lymphocytic leukemia; PARP, poly(ADP-ribose) polymerase; MART,
mono(ADP-ribosyl) transferase; 3AB, 3-aminobenzamide; 4AN,
4-amino-1,8-naphthalimide; 2NP,
2-nitro-6(5H)-phenanthridinone; CdA,
2-chloro-2'-deoxyadenosine; PI, propidium iodide; FSC, forward-angle
light scatter; SSC, side-angle light scatter; DiOC6,
3,3'-dihexolyloxacarbocyanine iodide; PS, phosphatidyl serine.
4 A. R. Pettitt, T. Stankovic, P. D.
Sherrington, G. S. Stewart, J. C. Cawley, and A. M. R. Taylor. p53 dysfunction in B-cell CLL: inactivation of ATM as an
alternative to p53 mutation, submitted for
publication.
Received 12/ 6/99. Accepted 6/ 1/00.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSDetermination of p53 StatusRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. R. Mackey, C. M. Galmarini, K. A. Graham, A. A. Joy, A. Delmer, L. Dabbagh, D. Glubrecht, L. D. Jewell, R. Lai, T. Lang, et al. Quantitative analysis of nucleoside transporter and metabolism gene expression in chronic lymphocytic leukemia (CLL): identification of fludarabine-sensitive and -insensitive populations Blood, January 15, 2005; 105(2): 767 - 774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rosenwald, E. Y. Chuang, R. E. Davis, A. Wiestner, A. A. Alizadeh, D. C. Arthur, J. B. Mitchell, G. E. Marti, D. H. Fowler, W. H. Wilson, et al. Fludarabine treatment of patients with chronic lymphocytic leukemia induces a p53-dependent gene expression response Blood, September 1, 2004; 104(5): 1428 - 1434. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hasmann and I. Schemainda FK866, a Highly Specific Noncompetitive Inhibitor of Nicotinamide Phosphoribosyltransferase, Represents a Novel Mechanism for Induction of Tumor Cell Apoptosis Cancer Res., November 1, 2003; 63(21): 7436 - 7442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Lin, P. D. Sherrington, M. Dennis, Z. Matrai, J. C. Cawley, and A. R. Pettitt Relationship between p53 dysfunction, CD38 expression, and IgVH mutation in chronic lymphocytic leukemia Blood, July 30, 2002; 100(4): 1404 - 1409. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. K. Baker, A. R. Pettitt, J. R. Slupsky, H. J. Chen, M. A. Glenn, M. Zuzel, and J. C. Cawley Response of hairy cells to IFN-alpha involves induction of apoptosis through autocrine TNF-alpha and protection by adhesion Blood, June 28, 2002; 100(2): 647 - 653. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Wosikowski, K. Mattern, I. Schemainda, M. Hasmann, B. Rattel, and R. Loser WK175, a Novel Antitumor Agent, Decreases the Intracellular Nicotinamide Adenine Dinucleotide Concentration and Induces the Apoptotic Cascade in Human Leukemia Cells Cancer Res., February 1, 2002; 62(4): 1057 - 1062. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) represents early
apoptotic (DiOC6-dim/PI-dim) cells, whereas the open
portion (
) represents late apoptotic/necrotic
(DiOC6-dim/PI-bright) cells. Very similar results were
obtained for CdA (data not shown).