- Email: [email protected]
Roles of calcium in the regulation of apoptosis in HL-60 promyelocytic leukemia cells
Life Sciences, Vol. 57, No. 23, pp. 2091-2099, 1995 Copyright Q 1995 Elxvier Science Inc. Printed in the USA. All rights reserved 0024-3205/95 $9.50 + .Xl
Pergamon 0024-3205(95)02202-3
ROLES OF CALCIUM IN THE REGULATION OF APOPTOSIS PROMYELOCYTIC LEUKEMIA CELLS
IN HE-60
Wen-Hui Zhu Br Tatt-Tuck Lob’ Department
of Physiology,
Faculty of Medicine, The University of Hong Kong, Hong Kong (Received
in linal form September
20, 1995)
Summarv Increase in intracellular calcium concentrations ([Ca2+]i) is critical for the initiation of apoptosis in cells such as thymocytes and in other cells, calcium chelators may promote apoptosis. However, calcium modulators, such as calcium ionophore 4bromo-calcium ionophore (Br-A23187) and thapsigargin (TG), induce apoptosis in different cells, including HL-60 cells in which the induction of apoptosis seems a calcium-independent process. These observations imply that the disturbance of calcium homeostasis is probably the most important factor in the regulation of apoptosis. In this article, reagents with different potencies of modulating calcium homesstasis were used to study the possible role of [Ca2+]i and the status of intracellular calcium stores in the causation of HE-60 cell apoptosis. We found that. an increase in [Caz+]i alone did not result in apoptosis, while the depletion of TGsensitive calcium stores in the endoplasmic reticulum was closely related with the induction of apoptosis. In HL-60 cells, extracellular and intracellular calcium chelators promoted apoptosis. Calmodulin antagonist did not attenuate apoptosis induced by other reagents. Our results suggest that the BepPetion of Ca2+ stores is an important mean to modulate calcium homeostasis and that the mobilization of calcium (Ca2+) from intracellular stores, rather than an increase in [Caa+]i, provides the signal for the induction of apoptosis in HL-60 cells. Key Words:
apoptosis,
leukemia,
calcium,
thapsigargin,
HL-60
cell line
Apoptosis is an active cell death process occurring in both physiologic and pathologic conditions, ranging from embryological development and tissue size regulation to anti-cancer therapies (1). It is characterized by certain morphologic changes such as cell shrinkage, chromatin condensation, membrane blebs as well as intemucleosomal cleavage of DNA. The latter is the result of the action of constitutive endonuclease, producing the specific apoptotic DNA “ladder” on agarose gel (2). Ca2+ is an important second signal which participates in the regulation of many cellular functions. The involvement of Caa+ in apoptosis has been widely documented. In thymocytes, an increase in [Caz+]i is associated with the induction of apoptosis which can be abolished by intracellular/extracellular Ca2+ chelators or calmodulin antagonists, suggesting that the work of Caa+-dependent endonuclease is responsible for intemucleosomal cleavage of DNA (3-6). An increase in [Caa+]i has also been documented in other cells (7-9) responsible for the initiation of apoptosis. However, contradictory findings that an increase in [Caz+]i was not required for the induction of apoptosis have been reported (10,ll) and extracellular or intracellular CaZ+ chelators could promote apoptosis in different cells (10,12,13). These findings imply that Ca2+ may not be ubiquitously required for apoptosis. In HL-60 cells, induction of apoptosis is also a Caz+independent process (14). Regulation of Ca2+ homeostasis is a complex process and the change in *Correspondent
author.
2092
Ca”’ and HL-60 Cell Apoptosis
Vol. 51, No. 23, 1995
[Caz+]i is only one aspect of this complex process. Depletion of ER Ca2* pool by thapsigargin (TG) has been reported to be responsible for the inhibition of growth in smooth muscle cells (15). Disturbance of Caz+ homeostasis has also been proposed to be responsible for the initiation of apoptosis in cells where increase in [C$+]i is not required (10,13). In this study, with the use of agents with different potencies of influencing intracellular Ca2+ homeostasis, we further investigated the possible roles of [Ca*+]i and the status of intracellular Ca*+ stores in the causation of HL-60 cell apoptosis. We demonstrated that the solitary increase in [Caz+]i did not result in apoptosis, while the depletion of TG-sensitive Caz+ pool in the endoplasmic reticulum was closely related with the induction of apoptosis in ML-60 cells. Cal+ chelators EGTA and 1,2-bis(o-aminophenoxy)ethane tetraacetic acid (BAPTA) promoted apoptosis and calmodulin antagonist did not attenuate apoptosis in ML-60 cells These results suggest that the mobilization of Ca2+ from intracellular pools, rather than an increase in [Ca*+]i, provide signal for the initiation of apoptosis in HL-60 cells. Materials and methods Materials: Br-A23 187, TG, ATP, histamine, EGTA. BAPTA, N-(4-aminobutyl)-5-chlorolnaphthalenesulfonamide (W-13), Fura-2/AM, and RPMI I640 medium powder were purchased from Sigma, USA. Fetal calf setum (FCS) was from Gibco, USA Cell culture: HL-60 cell line was obtained from American Type Culture Collection. Cells were cultured in RPMI 1640 medium (containing 100 U/ml penicillin and 100 pg/mI streptomycin) supplemented with 10% FCS at 37°C in humidified air containing 5% C02. They were subcultured twice a week and only those in an exponential growth period were used in these experiments. Evaluation of HE-60 cell apoptosis: Apoptosis was confirmed by checking the morphology of treated cells (the formation of apoptotic body in apoptotic cells) and by analyzing the specific DNA fragmentation. Cell viability was also examined after the scheduled treatments to verify the absence of necrosis. All treatments in this study did not significantly alter cell viability. Qualitative and quantitative analysis of internucleosomal fragmented DNA: The internucleosomal DNA fragmentation was assayed by a modified method of Bhalla et al (16). ML-60 cells were treated with the test agents for 4 hours. Then the cells were washed with isotonic phosphatebuffered saline (PBS, pH 7.4) and disrupted by a lysis buffer (5 mM/L Tris-HCl, 0.5% (v/v) Triton X-100 and 20 mM/L EDTA). The cellular lysates were centrifuged at 13,000 g for 20 minutes to separate the low molecular weight DNA from the intact chromatin. Fragmented DNA in the supematant was extracted with phenol, and chloroform/isoamyl alcohol (25:24:1). The total purified DNA from each sample of 5 x 106 cells was dissolved in I5 pl of Tris/EDTA loading buffer (pH 8.0) and 3 ~1 of tracking dye (50% glycerol, 1% xylene cyanol) and electrophoresed through 1% agarose gel. DNA was visualized by UV illumination For quantitative DNA analysis, lo7 cells were treated with test reagents for 4 hours. At the end of incubation, cells were disrupted and the supernatant was collected as described above. The diphenylarnine method (16) was used to measure the DNA content in the supernatant. The amount of spontaneously occurred fragmented DNA in either serum-free medium or medium with 10% FCS was approximately 2 pg/107 cells. Measurement of [Caz+]i: [Caz+]i was measured by a method as described by Lambert & Nahorski (17). Briefly, HL-60 cells at approximately 80% confluence were collected, washed and resuspended to a cell density of 2 x 107/ml with RPM1 medium. Finally they were loaded with 5 l.tM Fura- for 45 min at 37°C. At the end of loading, the cells were washed three times with Ca*+-free (with IO0 I.~M EGTA) or Ca2+-containing Hepes-buffered Hanks’ balanced salts (HBSS, pH 7.4) as defined in the text and were re-suspended to 106/ml with the same buffer. Two ml suspension was added to the quartz cuvette and the fluorescence was measured at 37°C using a Hitachi F4000 fluorescence spectrometer with a thermally controlled cuvette holder and a magnetic stirrer. The excitation and emission wavelengths used were 340 and 500 nm, respectively. Reagents were
Vol. 51, No. 23, 1995
Ca*’ and HL-60 Cell Apoptosis
m3
added after a stable fluorescence base was obtained. F,, was obtained by adding 0.1% Triton X100 and Fmi,, by 10 mM EGTA (after the addition of 1.25 mM of Ca*+ in the case of Caz+-free medium). [Ca2+]i was calibrated with the following formula: [Ca*+]i (nM) = (F - Fen)/(Fmax - F) x 224 (F = recorded fluorescence value). Statistical analysis: Student’s t-test was used to analyze the data and the values were expressed as Mean * SD.. Results and apoptosis in l-IL-60 cells
Effects of different Ca2+ modulators on Ca2+ h-is
TG, Br-A23187, ATP and histamine mobilize Ca*+ from intracellular Ca2* stores with a subsequent increase in [Ca2+]i. However, these agents differ in the ability to mobilize intracellular Ca2+. [CsZ+]i (nM)
2min
A
[CaZ+]l (aM)
[Ca2+]i (mM)
@
3w-
90 -
k.I_
_
Fig. I Effects of different Ca2+ modulators on the regulation of intracellular CaP+ homeostasis: HL-60 cells were loaded with Fura-2/AM and then t-e-suspended in Ca2+free HBSS buffer. A. Different abilities of ATP, TG, and Br-A23187 to mobilize of Ca*+ from intracellular stores: Optimal concentrations of ATP (100 PM), TG (6 nM) and BP-A23 187 (BrA, 1pM) were sequentially added to cell suspension and changes in fluorescence were recorded. B. The different effects of ATP and Br-A23187 on TGsensitive Ca2+ pool To check the status of TG-sensitive Ca2+ pool, 3Q nM TG was administrated after the first stimulation with optimal concentrations of either ATP (100 pM) or Br-A23 187 (1 PM). Increase in [Ca*+]i induced by TG was compared with that obtained in intact cells. C. Effects of ATP, TG and Br-A23187 on Caz+ influx. After stimulation with optimal concentrations of ATP (100 FM), TG (6 nM) or Br-A23187 (YpM), 1.25 mM Ca*+ (Ca) was added into the cell suspension. Increase in fluorescence was used as an index to represent Ca2+ influx from external medium. in .As shown . . Fig. lA, after stimulation with maximal concentration~_ (100 gmM) of ATP (Or hutamme, data not shown), Tti still induced a further increase in [CaL+]i. Iu a similar manner, after stimulation with maximal concentration (6 r&I) of TG. Br-A23187 induced a further increase in [Ca*+]i. On the other hand, if the sequence was reversed, there was no further increase in [Ca2+]i. TG is an inhibitor of the Ca*+-ATPase on the endoplasmic reticulum membrane which blocks replacement of Ca*+ in the endoplasm reticulum and consequently depletes its Ca2+ pool. The effects of Br-A23187 and ATP on TG-sensitive Ca2+ poop are presented in Fig. 1B. Maximal concentration (100 pM) of ATP depleted only a small part of the TG-sensitive Ca*+ no01
Vol. 57, No. 2,
Ca2+ and HL-60 Cell Apoptosis
2094
1995
(approximately lo%), while 1 FM Br-A23187 almost completely depleted it. These different effects of ATE, TG and Br-A23 187 on intracellular Ca2+ pool were accompanied by a diversity in plasma membrane permeability to external Ca 29. As shown in Fig. I@, with TG and Br-A23187 there was a marked enhancement of Ca2+ influx, while with ATP the increase was less significant. TG- or Br-A23187-induced increase in Ca2+ influx was the result of an extensive depletion of intracellular Ca2+ pools by opening the specific membrane channel for Ca2+. The increase in Ca2+ influx contributed largely to the sustained increase in [Caz+]i observed in Ca2+-containing medium. Fig. 2 Induction of DNA fragmentation in HL60 cells by different agents: 5 x 106 HL60 cells were exposed to different agents for 4 hours. Fragmented DNA was extracted, purified and run on agarose gel with the method described in Materials and Methods. Lane 1, @XI74 Hae III DNA marker; Lane 2, control cells; Lane 3, 2.5 nM TG; Lane 4, 250 nM Br823187; Lane 5, 100 @I ATP and Lane 6, 100 l.tM histamine.
[Ca2+]i 260-
2 min
(nM) k
40&
_..__..--.._-..--..
.._....
.,._-.._
a
v
0
Concentrations
of TG (nM)
Fig. 3 Effects of TG on TG-sensitive Cal+ pool and the induction of DNA fragmentation: HL-60 cells were treated with TG for 4 h in a serum-free medium. The cells were then loaded with Fura-2/AM and the status of TG-sensitive Ca2+ pool was checked by the addition of 30 nM TG. Alternatively, the amount of fragmented DNA in treated ceBls was measured to evaluate the extent of apoptosis. A. TypicaI recordings of Ca2+ measurement (TG pretreatment: a. Control; b. 0.5 nM; c. 1 nM; d. 2.5 nM and e. 6 nM). B. Relationship between TG-sensitive Ca2+ pool depletion and the induction of DNA fragmentation. The values are Mean f SD., n = 4. * P < 0.01 compared with control. Morphologically, apoptotic HL-60 cells were characterized by extensive membrane bleb formation 4 hour’s treatment of the cells with different apoptotic agents did not significantly alter cell viability (data not shown). Upon this observation, apoptosis was evaluated qualitatively and
Ca*+ and HL-60 Cell Apoptosis
Vol. 57, No. 23, 1995
2CM
quantitatively by analysis of specific DNA fragments. Even under normal conditions, some degree of spontaneous apoptosis (approximately 2 @IO7 cells) occurs in both a serum-free medium and a medium with 10% FCS. The effects of Br-A23187, TG, ATP and histamine on HL60 cell apoptosis are seen in Fig. 2. Where cells were treated in a serum-free medium with either Br-A23187 or TG there was a pronounced apoptosis after 4 hours. There was no evidence of apoptosis with ATP and histamine treatment. These results suggest that apoptosis will not occur when there is only an increase in [Caz+]i but no significant depletion in the TG-sensitive Ca*+ pool. To further understand the effect of TG-induced intraceifular Cal<- pool depletion on the induction of apoptosis, the relationship between TG-induced Ca*+ pool depletion and the induction of apoptosis was examined (Fig. 3). TG induced considerable DNA fragmentation at 4 h. A typical result demonstrating the condition of the TG-sensitive Ca*+ pool after 4 h’s treatment with different concentrations of TG, is seen in Fig. 38. 6 r&l TG depleted the TG-sensitive Ca*+ pool completeely. Fig. 318 shows the relationship between the amount of TG-sensitive Ca*+ pool depletion and DNA fragmentation after 4 h’s treatment with different concentrations of TG. The depletion of the TG-sensitive Ca*+ pool was dose-dependent and was related with induction of DNA fragmentation. An approximately 50% depletion of TG-sensitive Ca** pool was required to promote DNA fragmentation.
A [Ca2+]i (nM)
2 min
; 1
TG 30 nM
b
II e
Fig. 4 Effects of serum on TG-induced Ca*+ pool depletion and apoptosis: HL-60 cells were pretreated with TG under different conditions. Untreated cells, (a); cells treated with 6 nM TG for 4 h in serum-free medium (b) or in medium supplemented with 20% FCS (c); cell treated with 6 nM TG for 10 min, followed by throughly washing and incubated for up to 4 h in serum-free medium (d) or in medium supplemented with 20% FCS (e). Cells were then loaded with Fura-2/AM and the status of TG-sensitive Ca2+ pool was checked by the addition of 30 nM TG (Fig. 4A). The amounts of fragmented DNA in treated cells are shown in Fig. 4B. The amounts of DNA fragments of untreated cells cultures in serum-free medium or medium with 20% FCS were used as controls, respectively. There was no difference in the amount of DNA fragments in untreated cells cultured in serum-free medium or medium with 20% FCS. The bars are Mean f S.D., n = 4. * P < 0.05. Short et al (15) reported that serum promoted the refilling of TG-depleted Caz+ pool when TG was washed away. We observed the effect of serum on TG-induced Ca*+ pool depletion and apoptosis (Fig. 4). In cells treated with 6 r&l TG for 4 h, the TG-sensitive Ca*+ pool was completely depleted and remarkable DNA fragmentation was induced (Fig. 4A-b and 4B-b). The presence of
Ca2’ and HL-60 Cell Apoptosis
2096
Vol. 51, No. 23, 1995
20% of serum did not reverse the depletion of Ca2+ pool by TG if TG was continuously present (Fig. 4A-c); it also did not decrease TG-induced DNA fragmentation (Fig. 4B-c). In cells treated with 6 nM TG for only 10 min followed by complete washing, no pool refilling was detected in 4 h in serum-free medium (Fig. 4A-d); the amount of DNA fragments was only slightly decrease compared with that in cells treated with TG for 4 h (Fig. 4B-d). However, if added after TG was completely washed away, serum promoted the refilling of the depleted Caa+ pool (Fig. 48-e). There was a pool recovery of approximately 40% in 4 hours in cells tmated with 6 nM TG for IO min followed by washing and addition of 20% FCS. The refilling of TG-sensitive Ca2* pool was accompanied by significant inhibition DNA fragmentation (Fig. 4B-e). Effects of Ca2+ chelators and calmodulin
antagonist on anontosis
Br-A23187 and TG not only deplete intracellular Ca2+ pool, but also increase the [Caz+]i. To determine whether the [Ca2+]i increase induced by these agents was required for the induction of apoptosis, the effects of Ca2+ chelators and calmodulin antagonist on apoptosis were examined. In a Ca2+-containing medium, Br-A23187- or TG-induced increase in [Caz+]i was partially due to an increase in the permeability of the membrane to Ca 2+. Therefore, the extracellular Ca2+ chelator EGTA was used to eliminate Ca2+ in the medium. EGTA decreased the magnitude of increase in [Caz+]i and obviously shortened the time of sustained [Ca2+]i increase induced by TG and BrA23187 (Fig. 5A). If an increase in [Ca2+]i were essential for the initiation of apoptosis, EGTA would decrease the apoptotic effect of TG and Br-A23187. As shown in Fig. 5B, however, EGTA itself induced slight DNA fragmentation in HL-60 cells and significantly increased the apoptotic effect of both Br-A23 187 and TG. IQWi
0W
B
A
550 soo-
r
a
2mln
60-
rt BPA 500 nM
TG2.5nM
b
z
L
I
I
Control
TG
I
Br-A
Fig. 5 Effects of extracellular Ca2+ chelator EGTA on [Ca2+]i increase and DNA fragmentation induced by Br-A23 187 and TG: A. HL-60 cells were loaded with Fura2/AM and then suspended in Caa+-containing HBSS. Increase in [Ca2+]i induced by Br-A23187 or TG was recorded in the absence (a) or presence (b) of 2 mM EGTA. B. 1 x 107 cells were suspended in serum-free RPMI 1640 medium (with 0.424 mM Ca2+ and 0.406 mM Mg2+). Cells were treated with 250 nM of Br-A23187 or 2.5 nM of TG for 4 h in the presence (shadow bars) or absence (open bars) of 2 mM of EGTA. The fragmented DNA was then extracted and measured. The bars represent Mean f S.D., n = 6. * P c 0.01 compared with the respective group without EGTA. Calmodulin is the major intracellular receptor for Ca2+ (18). Calmodulin antagonist W-13 which blocks the Caa+/calmodulin signal pathways, inhibited HL-60 cell growth in a dose-dependent
Vol. 57, No. 23, 1995
Ca2+ and HL-60 Cell Apoptosis
20!97
manner with the maximal effect at 40 pM. However, 40 l.tM W-13 did not cause apoptosis. It even slightly enhanced DNA fragmentation induced by Br-A23187 and TG (data not shown). BATPA is also an intracellular Ca2+ chelator. As shown in Fig. 6, instead BAPTA itself produced a dose-dependent DNA fragmentation. BAPTA [Caz+]i induced by TG.~In the presence of both EGTA and BAPTA, the [Caz+]i was abolished (Fig. 7A). However, additive apoptotic effects were EGTA and TG (Fig. 7B).
of inhibiting apoptosis, reduced the increase in TG-induced increase in observed with BAPTA,
TT
r ,Je, 1
Y
Fig. 6 Effect of intracellular Ca2+ chelator BAPTA on HE-60 cell apoptosis: HE60 cells were treated with different concentrations of BAPTA for 4 b in serum-free medium. The amount of fragmented DNA was then measured. The bars are Mean + SD., n = 4. * P e 0.05, ** P < 0.01 compared with control.
*
*
I
6.5
Concentrations
6
5.5
5
(
4.5
of BAPTA (- log M)
B
A [Ca2+]i
270-
(nM)
r---
2 min
TG+E+Bi TG+B-
B-
TG -
Control I
I
I
0
250
500
Fig. 7 Effects of Ca2+ chelators on TG-induced [Ca2+]i increase and apoptosis: A. I-K-60 cells were loaded with Fura-2/AM and suspended in Ca2+-containing HBSS. TGinduced increase in [Caz+]i was recorded in the presence of 10 pM BAPTA and/or 2 mM EGTA. B. 1 x 107 cells were treated for 4 h in serum-free medium with 2.5 nM TG, 2 mM EGTA (E), 10 pM BAPTA (B) or the combination of these agents. The amount of fragmented DNA was then measured. The bars are Mean k S.D., n = 6.
Ca2+ and HL-60 Cell Apoptosis
Vol. 57, No. 23, 1995
DiSCUSSiOn
Thapsigargin (TG), a specific inhibitor of the Ca 2+-ATPase of the endoplasmic reticulum (ERJ membrane, blocks the refilling of Ca2+ and irreversibly depletes the Ca*+ stores in ER (19). Ca2* ionophores induce a rapid release of Ca*+ from various intracellular stores, including ER and mitochondria (20). Also ATP and histamine release Ca2+ from intracellular non-mitochondrial Ca2+ stores by generating inositol l&%trisphosphate (InsP3) (21,22). However, ATP and histamine deplete only a small fraction of the TG-sensitive pool in HL-60 cells (23). We domonstrated that the different effects of these Ca2+ modulators on intracellular Ca*+ stores were also accompanied by different changes in Ca2+ influx. Br-A23187 and TG significantly increased Ca*+ influx, while ATP and histamine did not. Also Br-A23187, TG, ATP and histamine had different ability to initiate apoptosis Although all of them increased [Ca2+Ji, only TG and BrA23187, which emptied the ER Caz+ pool, were capable of inducing apoptosis, while ATP and histamine did not. Approximately 50% depletion of the TG-sensitive ER Ca*+ pool was required for the initiation of apoptosis. The role of ER Ca*+ pool depletion is further strengthened by our findings that serum promoted the recovery of the depleted ER Ca2+ pool and inhibited apoptosis. These results suggest that an increase in [Ca*+]i alone does not induced apoptosis, while the depletion of TG-sensitive ER Ca*+ pool is closely related to the induction of apoptosis. In spite of the fact that Br-A23187 or TG increases [Ca2*]i, several lines of evidence here suggest that an increase in [Caz+]i induced by these agents is not associated with apoptosis. First, the calmodulin antagonist W-13 did not attenuate Br-A23187- or TG-induced apoptosis. Second, the apoptotic effects of Br-A23187 and TG were not attenuated in the presence of EGTA, a chelator of extracellular Ca*+ which prevented a sustained increase in [Caz+]i. Third, intracellular Ca2+ chelator BAPTA also induced apoptosis in HL-60 cells and additive apoptotic effects were observed when TG, BAPTA and EGTA were simultaneously administrated. These results differ from those observed in thymocyte where Ca2+ played a crucial role in apoptosis (46,24). The existence of endonucleases with different requirement of Ca2+ for their activation may be responsible for the heterogeneous action of Ca2+ in the induction of apoptosis in various cell. types, as it has been reported that the endonuclease of thymocyte is Ca2*/Mg2+-dependent and can be directly activated by calcium and magnesium (25); while the endonuclease separated from myelogenous cell lines (HL-60, U-937, KG-I) is Ca2+-independent (26). The increase in [Ca*+]i is a result of either mobilization of Ca 2+ from intracellular stores or an increase in the influx of extracellular Ca 2+. Moreover, an increase in Ca*+ influx induced by Ca*+ modulators is generally due to the depletion of intracellular Ca *+ stores resulting in the opening of the second messenger-operated channels (SMOCs) to Ca*+ (27). Therefore, mobilization of intracellular Ca*+ is the original and probably the most important step in the modulation of Ca2+ homeostasis. Lam et al also reported that the apoptosis-inhibiting BCL-2 protein, localized on the ER, reduced Ca*+ efflux from ER and suppressed TG-induced apoptosis in mouse lymphoma cells (28). These results indicate that TG-induced apoptosis was related to Ca*+ release from ER. In conclusion, our data support the interpretation that in some cell types mobilization of intracellular Ca*+ stores, rather than an increase in [Ca*+]ig provides the signal(s) for the initiation of apoptosis, Acknowledpmenrs We a= indebted to Dr. Russel A. Huggins for his review on our manuscript and Mr. W. B. Wong for technical assistance. This work was supported by Hong Kong Research Grant Council No. HKU 372/94M and the University of Hong Kong Research Grants No. 337-034-0032.
1. 2.
J.F.R. KERR, and B.V. HARMON, Anootosis: The molecular basis of cell death, L.D. TOMEI, and F.O. COPE (eds), 5-25, Cold Spring Harbor Laboratory Press, New York (1991). A. H. WYLLIE, Nature 284 555-556 (1990).
Vol. 57, No. 23, 199.5
Cazt
and HL-60
Cell Apoptosis
2O!B
3.
S. DURANT,
4.
g;8%‘CONKEY P HARTZELL, S.K. DUDDY, H. HAKANSSON and S. CRENIUS, S&&e 242 256-259 (1988). M.D. STORY, L.C. STEPHENS, S.P. TOMASOVIC and R.E. MEYN, Int. J. Radiol. Biol. fj.I 234-25 1 (1992). S. JIANG, S.C. CHOW, P. NICOTERA and S. ORRENIUS, Exp. Cell Res. X2 84-92 (1994). M. PEROTTI, F. TODDEI, F. MIRABELLI, M. VAIRETTI, G. BELLOMO, D.J. MCCONKEY and S. ORRENIUS, FEBS L&t. 259 331-334 (1990). 6. BELLOMO, M. PEROTTI, F. TADDEI, F. MIRABELLI, G. FINARDI, P. NICOTERA and S. ORRENIUS, Cancer Res. 2 1342-1346 (1992). Y. FURUYA, P. LUNDMO, A.D. SHORT, D.L. GILL and J.T. ISSACS, Cancer Res. &I 6167-6175 (1994). G. BAFFY, T. MIYASHITA, J.R. WILLIAMSON and J.C. REED, J. Biol. Chem. 248 6511-6519 (1993). I.I. BRUMAN, A.S. GUKOVSKAYA, V.V. PETRUNYAKA, I.P. BELTSKY and ES. TREPAKOVA, J. Cell Physiol. B 112-l 17 (1992). N.J. BANSAL, A.G. HOULE and G. J. MELNYKOVYCH, Cell Physiol. X$105-lQ9 (1990). R.M. KLUCK, C.A. MCDOUGALL, B.V. HARMON and I.W. HALLIDAY, Biocktim. Biophys. Acta. 1223 247-254 (1994). S.V. LENNON, S.A. KILFEATHER, M.B. HALLETT, A.K. CAMPBELL and T.G. COTTER, Clin. Exp. Immunol. fl465-47 1 (1992). A.D. SHORT, J. BIAN, T.K. GHOSH, R.T. WALDRON, S.L. RYBAK and D.L. GILL, Proc. Natl. Acad. Sci. U.S.A. $jQ4986-4990 (1993). K. BHALLA, C. TANG, A.M. BRADO, S. GRANT, E. TOURKINA, C. HOLLADAY, M. HUGHES, M.E. MAHONEY and Y. HUANG, Blood 80 2883-2890 (1992). D.G. LAMBERT and S.R. NAHORSKI, Biochem. J. 265 555-562 (1980). A.R. MEANS and J.R. DEDMAN, Nature m 73-77 (1980). 4. BIAN. T.K. GHOSH, J.C. WANG and D.L. GILL, J. Biol. Chem. 2M 8801-8806 (1991). P. R. ALBERT and JR. A.H. TASHJIAN, Am. J. Physiol. 25L C887-892 (I986). G.R. DUBYAK. D.S. COWEN and L.M. MEULLER. 9. Biol. Cllem. m 18108-18117 (1988) R. SEIFERT, A. HOER, I. SCHWANER and A. BUSCHAUER, Mol. Pharmacol. 42 235241 (1992). Y.M. LEUNG, C.Y. KWAN and TUT. LOH, Biochem. 9. j@ 87-94 (1994). D.R. DOWD, P.N. MACDONALD, B.S. KOMM, MR. HAUSSLER and R. 4. MIESFELD, I. Biol. @hem. 2M 18423-18426 (1991). J.J. COHEN and R.C. DUKE, J. Immunol. m 38-42 (1984). K. MATSUBARA, M. KUBOTA, S. ADACHI, K. KUWAKADO, H. HIRGTA, Y. WAKAZONO, Y. AKIYAMA and H. MIKAWA, Exp. Cell Res. 214 19-25 (1994). T. POZZAN, R. RIZZUTO, P. VOLPE and J. MELDOLESI, Physiol. Rev. B 595-636 (1994). M. LAM, G. DUBYAK, L. CHEN, G. NUNEZ, R.L. MIESFELD and C.W. DISTELMORST, Proc. Natl. Acad. Sci. l_J.S.A. ,%I 6569-6573 (1994).
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
::: 19. 20. 21, 22. 23. 24.
;:: 27. 28.
F. HOMO and D. DUVAL, Biochem. Biophys. Res. Commun. %! 385391
Pergamon 0024-3205(95)02202-3
ROLES OF CALCIUM IN THE REGULATION OF APOPTOSIS PROMYELOCYTIC LEUKEMIA CELLS
IN HE-60
Wen-Hui Zhu Br Tatt-Tuck Lob’ Department
of Physiology,
Faculty of Medicine, The University of Hong Kong, Hong Kong (Received
in linal form September
20, 1995)
Summarv Increase in intracellular calcium concentrations ([Ca2+]i) is critical for the initiation of apoptosis in cells such as thymocytes and in other cells, calcium chelators may promote apoptosis. However, calcium modulators, such as calcium ionophore 4bromo-calcium ionophore (Br-A23187) and thapsigargin (TG), induce apoptosis in different cells, including HL-60 cells in which the induction of apoptosis seems a calcium-independent process. These observations imply that the disturbance of calcium homeostasis is probably the most important factor in the regulation of apoptosis. In this article, reagents with different potencies of modulating calcium homesstasis were used to study the possible role of [Ca2+]i and the status of intracellular calcium stores in the causation of HE-60 cell apoptosis. We found that. an increase in [Caz+]i alone did not result in apoptosis, while the depletion of TGsensitive calcium stores in the endoplasmic reticulum was closely related with the induction of apoptosis. In HL-60 cells, extracellular and intracellular calcium chelators promoted apoptosis. Calmodulin antagonist did not attenuate apoptosis induced by other reagents. Our results suggest that the BepPetion of Ca2+ stores is an important mean to modulate calcium homeostasis and that the mobilization of calcium (Ca2+) from intracellular stores, rather than an increase in [Caa+]i, provides the signal for the induction of apoptosis in HL-60 cells. Key Words:
apoptosis,
leukemia,
calcium,
thapsigargin,
HL-60
cell line
Apoptosis is an active cell death process occurring in both physiologic and pathologic conditions, ranging from embryological development and tissue size regulation to anti-cancer therapies (1). It is characterized by certain morphologic changes such as cell shrinkage, chromatin condensation, membrane blebs as well as intemucleosomal cleavage of DNA. The latter is the result of the action of constitutive endonuclease, producing the specific apoptotic DNA “ladder” on agarose gel (2). Ca2+ is an important second signal which participates in the regulation of many cellular functions. The involvement of Caa+ in apoptosis has been widely documented. In thymocytes, an increase in [Caz+]i is associated with the induction of apoptosis which can be abolished by intracellular/extracellular Ca2+ chelators or calmodulin antagonists, suggesting that the work of Caa+-dependent endonuclease is responsible for intemucleosomal cleavage of DNA (3-6). An increase in [Caa+]i has also been documented in other cells (7-9) responsible for the initiation of apoptosis. However, contradictory findings that an increase in [Caz+]i was not required for the induction of apoptosis have been reported (10,ll) and extracellular or intracellular CaZ+ chelators could promote apoptosis in different cells (10,12,13). These findings imply that Ca2+ may not be ubiquitously required for apoptosis. In HL-60 cells, induction of apoptosis is also a Caz+independent process (14). Regulation of Ca2+ homeostasis is a complex process and the change in *Correspondent
author.
2092
Ca”’ and HL-60 Cell Apoptosis
Vol. 51, No. 23, 1995
[Caz+]i is only one aspect of this complex process. Depletion of ER Ca2* pool by thapsigargin (TG) has been reported to be responsible for the inhibition of growth in smooth muscle cells (15). Disturbance of Caz+ homeostasis has also been proposed to be responsible for the initiation of apoptosis in cells where increase in [C$+]i is not required (10,13). In this study, with the use of agents with different potencies of influencing intracellular Ca2+ homeostasis, we further investigated the possible roles of [Ca*+]i and the status of intracellular Ca*+ stores in the causation of HL-60 cell apoptosis. We demonstrated that the solitary increase in [Caz+]i did not result in apoptosis, while the depletion of TG-sensitive Caz+ pool in the endoplasmic reticulum was closely related with the induction of apoptosis in ML-60 cells. Cal+ chelators EGTA and 1,2-bis(o-aminophenoxy)ethane tetraacetic acid (BAPTA) promoted apoptosis and calmodulin antagonist did not attenuate apoptosis in ML-60 cells These results suggest that the mobilization of Ca2+ from intracellular pools, rather than an increase in [Ca*+]i, provide signal for the initiation of apoptosis in HL-60 cells. Materials and methods Materials: Br-A23 187, TG, ATP, histamine, EGTA. BAPTA, N-(4-aminobutyl)-5-chlorolnaphthalenesulfonamide (W-13), Fura-2/AM, and RPMI I640 medium powder were purchased from Sigma, USA. Fetal calf setum (FCS) was from Gibco, USA Cell culture: HL-60 cell line was obtained from American Type Culture Collection. Cells were cultured in RPMI 1640 medium (containing 100 U/ml penicillin and 100 pg/mI streptomycin) supplemented with 10% FCS at 37°C in humidified air containing 5% C02. They were subcultured twice a week and only those in an exponential growth period were used in these experiments. Evaluation of HE-60 cell apoptosis: Apoptosis was confirmed by checking the morphology of treated cells (the formation of apoptotic body in apoptotic cells) and by analyzing the specific DNA fragmentation. Cell viability was also examined after the scheduled treatments to verify the absence of necrosis. All treatments in this study did not significantly alter cell viability. Qualitative and quantitative analysis of internucleosomal fragmented DNA: The internucleosomal DNA fragmentation was assayed by a modified method of Bhalla et al (16). ML-60 cells were treated with the test agents for 4 hours. Then the cells were washed with isotonic phosphatebuffered saline (PBS, pH 7.4) and disrupted by a lysis buffer (5 mM/L Tris-HCl, 0.5% (v/v) Triton X-100 and 20 mM/L EDTA). The cellular lysates were centrifuged at 13,000 g for 20 minutes to separate the low molecular weight DNA from the intact chromatin. Fragmented DNA in the supematant was extracted with phenol, and chloroform/isoamyl alcohol (25:24:1). The total purified DNA from each sample of 5 x 106 cells was dissolved in I5 pl of Tris/EDTA loading buffer (pH 8.0) and 3 ~1 of tracking dye (50% glycerol, 1% xylene cyanol) and electrophoresed through 1% agarose gel. DNA was visualized by UV illumination For quantitative DNA analysis, lo7 cells were treated with test reagents for 4 hours. At the end of incubation, cells were disrupted and the supernatant was collected as described above. The diphenylarnine method (16) was used to measure the DNA content in the supernatant. The amount of spontaneously occurred fragmented DNA in either serum-free medium or medium with 10% FCS was approximately 2 pg/107 cells. Measurement of [Caz+]i: [Caz+]i was measured by a method as described by Lambert & Nahorski (17). Briefly, HL-60 cells at approximately 80% confluence were collected, washed and resuspended to a cell density of 2 x 107/ml with RPM1 medium. Finally they were loaded with 5 l.tM Fura- for 45 min at 37°C. At the end of loading, the cells were washed three times with Ca*+-free (with IO0 I.~M EGTA) or Ca2+-containing Hepes-buffered Hanks’ balanced salts (HBSS, pH 7.4) as defined in the text and were re-suspended to 106/ml with the same buffer. Two ml suspension was added to the quartz cuvette and the fluorescence was measured at 37°C using a Hitachi F4000 fluorescence spectrometer with a thermally controlled cuvette holder and a magnetic stirrer. The excitation and emission wavelengths used were 340 and 500 nm, respectively. Reagents were
Vol. 51, No. 23, 1995
Ca*’ and HL-60 Cell Apoptosis
m3
added after a stable fluorescence base was obtained. F,, was obtained by adding 0.1% Triton X100 and Fmi,, by 10 mM EGTA (after the addition of 1.25 mM of Ca*+ in the case of Caz+-free medium). [Ca2+]i was calibrated with the following formula: [Ca*+]i (nM) = (F - Fen)/(Fmax - F) x 224 (F = recorded fluorescence value). Statistical analysis: Student’s t-test was used to analyze the data and the values were expressed as Mean * SD.. Results and apoptosis in l-IL-60 cells
Effects of different Ca2+ modulators on Ca2+ h-is
TG, Br-A23187, ATP and histamine mobilize Ca*+ from intracellular Ca2* stores with a subsequent increase in [Ca2+]i. However, these agents differ in the ability to mobilize intracellular Ca2+. [CsZ+]i (nM)
2min
A
[CaZ+]l (aM)
[Ca2+]i (mM)
@
3w-
90 -
k.I_
_
Fig. I Effects of different Ca2+ modulators on the regulation of intracellular CaP+ homeostasis: HL-60 cells were loaded with Fura-2/AM and then t-e-suspended in Ca2+free HBSS buffer. A. Different abilities of ATP, TG, and Br-A23187 to mobilize of Ca*+ from intracellular stores: Optimal concentrations of ATP (100 PM), TG (6 nM) and BP-A23 187 (BrA, 1pM) were sequentially added to cell suspension and changes in fluorescence were recorded. B. The different effects of ATP and Br-A23187 on TGsensitive Ca2+ pool To check the status of TG-sensitive Ca2+ pool, 3Q nM TG was administrated after the first stimulation with optimal concentrations of either ATP (100 pM) or Br-A23 187 (1 PM). Increase in [Ca*+]i induced by TG was compared with that obtained in intact cells. C. Effects of ATP, TG and Br-A23187 on Caz+ influx. After stimulation with optimal concentrations of ATP (100 FM), TG (6 nM) or Br-A23187 (YpM), 1.25 mM Ca*+ (Ca) was added into the cell suspension. Increase in fluorescence was used as an index to represent Ca2+ influx from external medium. in .As shown . . Fig. lA, after stimulation with maximal concentration~_ (100 gmM) of ATP (Or hutamme, data not shown), Tti still induced a further increase in [CaL+]i. Iu a similar manner, after stimulation with maximal concentration (6 r&I) of TG. Br-A23187 induced a further increase in [Ca*+]i. On the other hand, if the sequence was reversed, there was no further increase in [Ca2+]i. TG is an inhibitor of the Ca*+-ATPase on the endoplasmic reticulum membrane which blocks replacement of Ca*+ in the endoplasm reticulum and consequently depletes its Ca2+ pool. The effects of Br-A23187 and ATP on TG-sensitive Ca2+ poop are presented in Fig. 1B. Maximal concentration (100 pM) of ATP depleted only a small part of the TG-sensitive Ca*+ no01
Vol. 57, No. 2,
Ca2+ and HL-60 Cell Apoptosis
2094
1995
(approximately lo%), while 1 FM Br-A23187 almost completely depleted it. These different effects of ATE, TG and Br-A23 187 on intracellular Ca2+ pool were accompanied by a diversity in plasma membrane permeability to external Ca 29. As shown in Fig. I@, with TG and Br-A23187 there was a marked enhancement of Ca2+ influx, while with ATP the increase was less significant. TG- or Br-A23187-induced increase in Ca2+ influx was the result of an extensive depletion of intracellular Ca2+ pools by opening the specific membrane channel for Ca2+. The increase in Ca2+ influx contributed largely to the sustained increase in [Caz+]i observed in Ca2+-containing medium. Fig. 2 Induction of DNA fragmentation in HL60 cells by different agents: 5 x 106 HL60 cells were exposed to different agents for 4 hours. Fragmented DNA was extracted, purified and run on agarose gel with the method described in Materials and Methods. Lane 1, @XI74 Hae III DNA marker; Lane 2, control cells; Lane 3, 2.5 nM TG; Lane 4, 250 nM Br823187; Lane 5, 100 @I ATP and Lane 6, 100 l.tM histamine.
[Ca2+]i 260-
2 min
(nM) k
40&
_..__..--.._-..--..
.._....
.,._-.._
a
v
0
Concentrations
of TG (nM)
Fig. 3 Effects of TG on TG-sensitive Cal+ pool and the induction of DNA fragmentation: HL-60 cells were treated with TG for 4 h in a serum-free medium. The cells were then loaded with Fura-2/AM and the status of TG-sensitive Ca2+ pool was checked by the addition of 30 nM TG. Alternatively, the amount of fragmented DNA in treated ceBls was measured to evaluate the extent of apoptosis. A. TypicaI recordings of Ca2+ measurement (TG pretreatment: a. Control; b. 0.5 nM; c. 1 nM; d. 2.5 nM and e. 6 nM). B. Relationship between TG-sensitive Ca2+ pool depletion and the induction of DNA fragmentation. The values are Mean f SD., n = 4. * P < 0.01 compared with control. Morphologically, apoptotic HL-60 cells were characterized by extensive membrane bleb formation 4 hour’s treatment of the cells with different apoptotic agents did not significantly alter cell viability (data not shown). Upon this observation, apoptosis was evaluated qualitatively and
Ca*+ and HL-60 Cell Apoptosis
Vol. 57, No. 23, 1995
2CM
quantitatively by analysis of specific DNA fragments. Even under normal conditions, some degree of spontaneous apoptosis (approximately 2 @IO7 cells) occurs in both a serum-free medium and a medium with 10% FCS. The effects of Br-A23187, TG, ATP and histamine on HL60 cell apoptosis are seen in Fig. 2. Where cells were treated in a serum-free medium with either Br-A23187 or TG there was a pronounced apoptosis after 4 hours. There was no evidence of apoptosis with ATP and histamine treatment. These results suggest that apoptosis will not occur when there is only an increase in [Caz+]i but no significant depletion in the TG-sensitive Ca*+ pool. To further understand the effect of TG-induced intraceifular Cal<- pool depletion on the induction of apoptosis, the relationship between TG-induced Ca*+ pool depletion and the induction of apoptosis was examined (Fig. 3). TG induced considerable DNA fragmentation at 4 h. A typical result demonstrating the condition of the TG-sensitive Ca*+ pool after 4 h’s treatment with different concentrations of TG, is seen in Fig. 38. 6 r&l TG depleted the TG-sensitive Ca*+ pool completeely. Fig. 318 shows the relationship between the amount of TG-sensitive Ca*+ pool depletion and DNA fragmentation after 4 h’s treatment with different concentrations of TG. The depletion of the TG-sensitive Ca*+ pool was dose-dependent and was related with induction of DNA fragmentation. An approximately 50% depletion of TG-sensitive Ca** pool was required to promote DNA fragmentation.
A [Ca2+]i (nM)
2 min
; 1
TG 30 nM
b
II e
Fig. 4 Effects of serum on TG-induced Ca*+ pool depletion and apoptosis: HL-60 cells were pretreated with TG under different conditions. Untreated cells, (a); cells treated with 6 nM TG for 4 h in serum-free medium (b) or in medium supplemented with 20% FCS (c); cell treated with 6 nM TG for 10 min, followed by throughly washing and incubated for up to 4 h in serum-free medium (d) or in medium supplemented with 20% FCS (e). Cells were then loaded with Fura-2/AM and the status of TG-sensitive Ca2+ pool was checked by the addition of 30 nM TG (Fig. 4A). The amounts of fragmented DNA in treated cells are shown in Fig. 4B. The amounts of DNA fragments of untreated cells cultures in serum-free medium or medium with 20% FCS were used as controls, respectively. There was no difference in the amount of DNA fragments in untreated cells cultured in serum-free medium or medium with 20% FCS. The bars are Mean f S.D., n = 4. * P < 0.05. Short et al (15) reported that serum promoted the refilling of TG-depleted Caz+ pool when TG was washed away. We observed the effect of serum on TG-induced Ca*+ pool depletion and apoptosis (Fig. 4). In cells treated with 6 r&l TG for 4 h, the TG-sensitive Ca*+ pool was completely depleted and remarkable DNA fragmentation was induced (Fig. 4A-b and 4B-b). The presence of
Ca2’ and HL-60 Cell Apoptosis
2096
Vol. 51, No. 23, 1995
20% of serum did not reverse the depletion of Ca2+ pool by TG if TG was continuously present (Fig. 4A-c); it also did not decrease TG-induced DNA fragmentation (Fig. 4B-c). In cells treated with 6 nM TG for only 10 min followed by complete washing, no pool refilling was detected in 4 h in serum-free medium (Fig. 4A-d); the amount of DNA fragments was only slightly decrease compared with that in cells treated with TG for 4 h (Fig. 4B-d). However, if added after TG was completely washed away, serum promoted the refilling of the depleted Caa+ pool (Fig. 48-e). There was a pool recovery of approximately 40% in 4 hours in cells tmated with 6 nM TG for IO min followed by washing and addition of 20% FCS. The refilling of TG-sensitive Ca2* pool was accompanied by significant inhibition DNA fragmentation (Fig. 4B-e). Effects of Ca2+ chelators and calmodulin
antagonist on anontosis
Br-A23187 and TG not only deplete intracellular Ca2+ pool, but also increase the [Caz+]i. To determine whether the [Ca2+]i increase induced by these agents was required for the induction of apoptosis, the effects of Ca2+ chelators and calmodulin antagonist on apoptosis were examined. In a Ca2+-containing medium, Br-A23187- or TG-induced increase in [Caz+]i was partially due to an increase in the permeability of the membrane to Ca 2+. Therefore, the extracellular Ca2+ chelator EGTA was used to eliminate Ca2+ in the medium. EGTA decreased the magnitude of increase in [Caz+]i and obviously shortened the time of sustained [Ca2+]i increase induced by TG and BrA23187 (Fig. 5A). If an increase in [Ca2+]i were essential for the initiation of apoptosis, EGTA would decrease the apoptotic effect of TG and Br-A23187. As shown in Fig. 5B, however, EGTA itself induced slight DNA fragmentation in HL-60 cells and significantly increased the apoptotic effect of both Br-A23 187 and TG. IQWi
0W
B
A
550 soo-
r
a
2mln
60-
rt BPA 500 nM
TG2.5nM
b
z
L
I
I
Control
TG
I
Br-A
Fig. 5 Effects of extracellular Ca2+ chelator EGTA on [Ca2+]i increase and DNA fragmentation induced by Br-A23 187 and TG: A. HL-60 cells were loaded with Fura2/AM and then suspended in Caa+-containing HBSS. Increase in [Ca2+]i induced by Br-A23187 or TG was recorded in the absence (a) or presence (b) of 2 mM EGTA. B. 1 x 107 cells were suspended in serum-free RPMI 1640 medium (with 0.424 mM Ca2+ and 0.406 mM Mg2+). Cells were treated with 250 nM of Br-A23187 or 2.5 nM of TG for 4 h in the presence (shadow bars) or absence (open bars) of 2 mM of EGTA. The fragmented DNA was then extracted and measured. The bars represent Mean f S.D., n = 6. * P c 0.01 compared with the respective group without EGTA. Calmodulin is the major intracellular receptor for Ca2+ (18). Calmodulin antagonist W-13 which blocks the Caa+/calmodulin signal pathways, inhibited HL-60 cell growth in a dose-dependent
Vol. 57, No. 23, 1995
Ca2+ and HL-60 Cell Apoptosis
20!97
manner with the maximal effect at 40 pM. However, 40 l.tM W-13 did not cause apoptosis. It even slightly enhanced DNA fragmentation induced by Br-A23187 and TG (data not shown). BATPA is also an intracellular Ca2+ chelator. As shown in Fig. 6, instead BAPTA itself produced a dose-dependent DNA fragmentation. BAPTA [Caz+]i induced by TG.~In the presence of both EGTA and BAPTA, the [Caz+]i was abolished (Fig. 7A). However, additive apoptotic effects were EGTA and TG (Fig. 7B).
of inhibiting apoptosis, reduced the increase in TG-induced increase in observed with BAPTA,
TT
r ,Je, 1
Y
Fig. 6 Effect of intracellular Ca2+ chelator BAPTA on HE-60 cell apoptosis: HE60 cells were treated with different concentrations of BAPTA for 4 b in serum-free medium. The amount of fragmented DNA was then measured. The bars are Mean + SD., n = 4. * P e 0.05, ** P < 0.01 compared with control.
*
*
I
6.5
Concentrations
6
5.5
5
(
4.5
of BAPTA (- log M)
B
A [Ca2+]i
270-
(nM)
r---
2 min
TG+E+Bi TG+B-
B-
TG -
Control I
I
I
0
250
500
Fig. 7 Effects of Ca2+ chelators on TG-induced [Ca2+]i increase and apoptosis: A. I-K-60 cells were loaded with Fura-2/AM and suspended in Ca2+-containing HBSS. TGinduced increase in [Caz+]i was recorded in the presence of 10 pM BAPTA and/or 2 mM EGTA. B. 1 x 107 cells were treated for 4 h in serum-free medium with 2.5 nM TG, 2 mM EGTA (E), 10 pM BAPTA (B) or the combination of these agents. The amount of fragmented DNA was then measured. The bars are Mean k S.D., n = 6.
Ca2+ and HL-60 Cell Apoptosis
Vol. 57, No. 23, 1995
DiSCUSSiOn
Thapsigargin (TG), a specific inhibitor of the Ca 2+-ATPase of the endoplasmic reticulum (ERJ membrane, blocks the refilling of Ca2+ and irreversibly depletes the Ca*+ stores in ER (19). Ca2* ionophores induce a rapid release of Ca*+ from various intracellular stores, including ER and mitochondria (20). Also ATP and histamine release Ca2+ from intracellular non-mitochondrial Ca2+ stores by generating inositol l&%trisphosphate (InsP3) (21,22). However, ATP and histamine deplete only a small fraction of the TG-sensitive pool in HL-60 cells (23). We domonstrated that the different effects of these Ca2+ modulators on intracellular Ca*+ stores were also accompanied by different changes in Ca2+ influx. Br-A23187 and TG significantly increased Ca*+ influx, while ATP and histamine did not. Also Br-A23187, TG, ATP and histamine had different ability to initiate apoptosis Although all of them increased [Ca2+Ji, only TG and BrA23187, which emptied the ER Caz+ pool, were capable of inducing apoptosis, while ATP and histamine did not. Approximately 50% depletion of the TG-sensitive ER Ca*+ pool was required for the initiation of apoptosis. The role of ER Ca*+ pool depletion is further strengthened by our findings that serum promoted the recovery of the depleted ER Ca2+ pool and inhibited apoptosis. These results suggest that an increase in [Ca*+]i alone does not induced apoptosis, while the depletion of TG-sensitive ER Ca*+ pool is closely related to the induction of apoptosis. In spite of the fact that Br-A23187 or TG increases [Ca2*]i, several lines of evidence here suggest that an increase in [Caz+]i induced by these agents is not associated with apoptosis. First, the calmodulin antagonist W-13 did not attenuate Br-A23187- or TG-induced apoptosis. Second, the apoptotic effects of Br-A23187 and TG were not attenuated in the presence of EGTA, a chelator of extracellular Ca*+ which prevented a sustained increase in [Caz+]i. Third, intracellular Ca2+ chelator BAPTA also induced apoptosis in HL-60 cells and additive apoptotic effects were observed when TG, BAPTA and EGTA were simultaneously administrated. These results differ from those observed in thymocyte where Ca2+ played a crucial role in apoptosis (46,24). The existence of endonucleases with different requirement of Ca2+ for their activation may be responsible for the heterogeneous action of Ca2+ in the induction of apoptosis in various cell. types, as it has been reported that the endonuclease of thymocyte is Ca2*/Mg2+-dependent and can be directly activated by calcium and magnesium (25); while the endonuclease separated from myelogenous cell lines (HL-60, U-937, KG-I) is Ca2+-independent (26). The increase in [Ca*+]i is a result of either mobilization of Ca 2+ from intracellular stores or an increase in the influx of extracellular Ca 2+. Moreover, an increase in Ca*+ influx induced by Ca*+ modulators is generally due to the depletion of intracellular Ca *+ stores resulting in the opening of the second messenger-operated channels (SMOCs) to Ca*+ (27). Therefore, mobilization of intracellular Ca*+ is the original and probably the most important step in the modulation of Ca2+ homeostasis. Lam et al also reported that the apoptosis-inhibiting BCL-2 protein, localized on the ER, reduced Ca*+ efflux from ER and suppressed TG-induced apoptosis in mouse lymphoma cells (28). These results indicate that TG-induced apoptosis was related to Ca*+ release from ER. In conclusion, our data support the interpretation that in some cell types mobilization of intracellular Ca*+ stores, rather than an increase in [Ca*+]ig provides the signal(s) for the initiation of apoptosis, Acknowledpmenrs We a= indebted to Dr. Russel A. Huggins for his review on our manuscript and Mr. W. B. Wong for technical assistance. This work was supported by Hong Kong Research Grant Council No. HKU 372/94M and the University of Hong Kong Research Grants No. 337-034-0032.
1. 2.
J.F.R. KERR, and B.V. HARMON, Anootosis: The molecular basis of cell death, L.D. TOMEI, and F.O. COPE (eds), 5-25, Cold Spring Harbor Laboratory Press, New York (1991). A. H. WYLLIE, Nature 284 555-556 (1990).
Vol. 57, No. 23, 199.5
Cazt
and HL-60
Cell Apoptosis
2O!B
3.
S. DURANT,
4.
g;8%‘CONKEY P HARTZELL, S.K. DUDDY, H. HAKANSSON and S. CRENIUS, S&&e 242 256-259 (1988). M.D. STORY, L.C. STEPHENS, S.P. TOMASOVIC and R.E. MEYN, Int. J. Radiol. Biol. fj.I 234-25 1 (1992). S. JIANG, S.C. CHOW, P. NICOTERA and S. ORRENIUS, Exp. Cell Res. X2 84-92 (1994). M. PEROTTI, F. TODDEI, F. MIRABELLI, M. VAIRETTI, G. BELLOMO, D.J. MCCONKEY and S. ORRENIUS, FEBS L&t. 259 331-334 (1990). 6. BELLOMO, M. PEROTTI, F. TADDEI, F. MIRABELLI, G. FINARDI, P. NICOTERA and S. ORRENIUS, Cancer Res. 2 1342-1346 (1992). Y. FURUYA, P. LUNDMO, A.D. SHORT, D.L. GILL and J.T. ISSACS, Cancer Res. &I 6167-6175 (1994). G. BAFFY, T. MIYASHITA, J.R. WILLIAMSON and J.C. REED, J. Biol. Chem. 248 6511-6519 (1993). I.I. BRUMAN, A.S. GUKOVSKAYA, V.V. PETRUNYAKA, I.P. BELTSKY and ES. TREPAKOVA, J. Cell Physiol. B 112-l 17 (1992). N.J. BANSAL, A.G. HOULE and G. J. MELNYKOVYCH, Cell Physiol. X$105-lQ9 (1990). R.M. KLUCK, C.A. MCDOUGALL, B.V. HARMON and I.W. HALLIDAY, Biocktim. Biophys. Acta. 1223 247-254 (1994). S.V. LENNON, S.A. KILFEATHER, M.B. HALLETT, A.K. CAMPBELL and T.G. COTTER, Clin. Exp. Immunol. fl465-47 1 (1992). A.D. SHORT, J. BIAN, T.K. GHOSH, R.T. WALDRON, S.L. RYBAK and D.L. GILL, Proc. Natl. Acad. Sci. U.S.A. $jQ4986-4990 (1993). K. BHALLA, C. TANG, A.M. BRADO, S. GRANT, E. TOURKINA, C. HOLLADAY, M. HUGHES, M.E. MAHONEY and Y. HUANG, Blood 80 2883-2890 (1992). D.G. LAMBERT and S.R. NAHORSKI, Biochem. J. 265 555-562 (1980). A.R. MEANS and J.R. DEDMAN, Nature m 73-77 (1980). 4. BIAN. T.K. GHOSH, J.C. WANG and D.L. GILL, J. Biol. Chem. 2M 8801-8806 (1991). P. R. ALBERT and JR. A.H. TASHJIAN, Am. J. Physiol. 25L C887-892 (I986). G.R. DUBYAK. D.S. COWEN and L.M. MEULLER. 9. Biol. Cllem. m 18108-18117 (1988) R. SEIFERT, A. HOER, I. SCHWANER and A. BUSCHAUER, Mol. Pharmacol. 42 235241 (1992). Y.M. LEUNG, C.Y. KWAN and TUT. LOH, Biochem. 9. j@ 87-94 (1994). D.R. DOWD, P.N. MACDONALD, B.S. KOMM, MR. HAUSSLER and R. 4. MIESFELD, I. Biol. @hem. 2M 18423-18426 (1991). J.J. COHEN and R.C. DUKE, J. Immunol. m 38-42 (1984). K. MATSUBARA, M. KUBOTA, S. ADACHI, K. KUWAKADO, H. HIRGTA, Y. WAKAZONO, Y. AKIYAMA and H. MIKAWA, Exp. Cell Res. 214 19-25 (1994). T. POZZAN, R. RIZZUTO, P. VOLPE and J. MELDOLESI, Physiol. Rev. B 595-636 (1994). M. LAM, G. DUBYAK, L. CHEN, G. NUNEZ, R.L. MIESFELD and C.W. DISTELMORST, Proc. Natl. Acad. Sci. l_J.S.A. ,%I 6569-6573 (1994).
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
::: 19. 20. 21, 22. 23. 24.
;:: 27. 28.
F. HOMO and D. DUVAL, Biochem. Biophys. Res. Commun. %! 385391