(Received for publication, May 31, 1996, and in revised form, December 6, 1996)
From the Department of Medicine, Case Western Reserve University School of Medicine and the Ireland Cancer Center, University Hospitals of Cleveland, Cleveland, Ohio 44106
grp78/grp94 induction is critical for maintaining the viability of epithelial cells and fibroblasts following treatment with thapsigargin (TG), an inhibitor of Ca2+ uptake into the endoplasmic reticulum. In contrast to these cell types, WEHI7.2 mouse lymphoma cells undergo apoptosis when treated with TG, prompting us to examine the grp78/grp94 stress response in WEHI7.2 cells. TG treatment failed to induce grp78/grp94 transcription in WEHI7.2 cells, measured by Northern hybridization and nuclear run-on assays, even if the cells were protected from apoptosis by overexpressing bcl-2. However, grp78/grp94 transcription was induced by the glycosylation inhibitor tunicamycin, suggesting that there are at least two grp78/grp94 signaling pathways, one in response to TG-induced endoplasmic reticulum Ca2+ pool depletion, which is inoperable in WEHI7.2 cells, and one in response to glycosylation inhibition, which is operable in WEHI7.2 cells. Studies of additional lymphoid lines, as well as several nonlymphoid lines, suggested a correlation between grp78/grp94 induction and resistance to apoptosis following TG treatment. In conclusion, the vulnerability of TG-treated WEHI7.2 cells to apoptosis may be due to failure to signal a grp78/grp94 stress response.
The endoplasmic reticulum (ER)1 is the major intracellular reservoir of Ca2+ in nonmuscle cells (1). The ER Ca2+ pool is essential for a number of vital cellular functions, which include protein processing within the ER (2, 3), maintaining high translation rates of newly transcribed messages (4), preserving the structural integrity of the ER (5, 6), and regulating cell proliferation and cell cycle progression (7). Under physiological conditions, the ER Ca2+ pool is maintained by an associated Ca2+-ATPase that pumps Ca2+ into the ER lumen from the cytoplasm (8). The ER Ca2+ pool can be depleted by treating cells with the Ca2+ ionophore A23187 or the selective ER Ca2+-ATPase inhibitor thapsigargin (TG) (9).
ER function is mediated, in part, by intraluminal Ca2+-binding proteins, which include the glucose-regulated proteins GRP78 and GRP94 (5, 10, 11). GRP78 and GRP94 are found constitutively within the ER, and transcription of the genes for these proteins is elevated in response to malfolded proteins, inhibition of glycosylation, and ER Ca2+ pool depletion (12-14). GRP78 is a highly conserved 78-kDa protein that shares 60% amino acid homology with the 70-kDa heat shock protein (HSP70). GRP78 (also known as BiP) associates transiently with nascent proteins as they traverse the ER and aids in their folding and transport (15-20). The binding of immature proteins by GRP78 requires ATP, and GRP78 has both ATP binding and ATPase activities (21). GRP94 is a 94-kDa glycoprotein that shares 50% amino acid homology with HSP90 (11, 22). GRP94 acts in concert with GRP78 to fold nascent proteins and also exhibits ATPase activity (22-24).
In epithelial cells and fibroblasts, grp78 and grp94 are coordinately regulated through common Ca2+-responsive promoter elements that respond to ER Ca2+ pool depletion (10, 25). Thus, ER Ca2+ pool depletion, induced by either A23187 or TG, signals an increase in grp78/grp94 transcription, producing a 5-20-fold elevation of grp78/grp94 mRNA levels (25). In these cells, the loss of ER Ca2+ induced by TG or A23187 does not result in a loss of viability, unless the grp78/grp94 stress response is repressed by antisense, promoter competition, or ribozyme techniques (26-28). Moreover, grp78/grp94 induction restores protein synthesis under conditions where intracellular Ca2+ is depleted (29). This indicates that grp78/grp94 gene induction is a protective response mechanism by which cells accommodate to potentially lethal stress caused by the disruption of intracellular Ca2+ homeostasis.
In contrast to epithelial cells and fibroblasts, we have found that WEHI7.2 mouse lymphoma cells undergo apoptosis in response to TG-induced ER Ca2+ loss, unless protected by overexpression of the anti-apoptotic oncogene bcl-2 (30). Given this observation, we chose to examine the grp78/grp94 stress response in WEHI7.2 mouse lymphoma cells. We report for the first time that TG-induced Ca2+ loss from the ER of WEHI7.2 cells does not induce grp78/grp94 transcription, even if cells are protected from undergoing apoptosis by bcl-2. Interestingly, treatment with tunicamycin (TN), an inhibitor of N-linked glycosylation, does induce grp78/grp94 transcription, suggesting that ER Ca2+ pool depletion and accumulation of underglycosylated proteins signal an increase in grp78/grp94 transcription through independent pathways, the former pathway being inoperative in WEHI7.2 cells. Moreover, in three breast cancer cell lines and two additional lymphoma lines, the induction of grp78 correlated with resistance to TG-induced apoptosis. These findings suggest that inherent differences in the susceptibility of cells to apoptosis induction by TG can be determined, at least in part, by the cell's capacity to mount a grp78/grp94 stress response.
TG was purchased from LC Laboratories, serum from Hyclone Laboratories, and TN from Calbiochem. L-Glutamine, antibiotics, and nonessential amino acids were from Life Technologies, Inc. All other chemicals, unless noted otherwise, were obtained from Sigma.
Cell Culture and Treatment ConditionsThe WEHI7.2 mouse lymphoma cell line, which does not express detectable levels of Bcl-2, was stably transfected with a cDNA encoding full-length human bcl-2, yielding the W.Hb12, W.Hb13, and W.Hb15 clones employed in the present study (30). Two additional Bcl-2-negative mouse lymphoma cell lines, S49.1 and W7.MG1, have been described previously (31). Lymphoma lines were maintained in Dulbecco's modified Eagle's medium (BioWhittaker, Inc.) supplemented with 10% (v/v) heat-inactivated horse serum, 2 mM glutamine, 50 units/ml penicillin, 50 µg/ml streptomycin, and 0.4 mM nonessential amino acids at 37 °C in a 7% CO2 atmosphere. Cells were maintained at a density of 0.2-1.5 × 106/ml by dilution into fresh culture medium three times weekly. Mm5MT mouse mammary cells (obtained from American Type Culture Collection) were maintained in the same culture medium as lymphoma lines, supplemented with 10 µg/ml bovine insulin. MDA-MB-468 human breast cancer cells (from M. Lippman, Georgetown University) were cultured in improved minimal essential medium (Biofluids, Inc.) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 2 mM glutamine, 50 units/ml penicillin, 50 µg/ml streptomycin, and 0.4 mM nonessential amino acids at 37 °C in a 7% CO2 atmosphere. MCF-7 human breast cancer cells (from S. Gerson, Case Western Reserve University) were cultured in RPMI 1640 medium (Cancer Center Tissue Culture Core Facility) supplemented with 10% (v/v) heat-inactivated fetal bovine serum, 25 µg/ml bovine insulin, 2 mM glutamine, 50 units/ml penicillin, 50 µg/ml streptomycin, and 0.4 mM nonessential amino acids at 37 °C in a 7% CO2 atmosphere.
A 1 mg/ml stock of TG was made in dimethyl sulfoxide and stored in
aliquots at 20 °C. A working stock was prepared by diluting TG in
fresh culture medium to a final concentration of 1 µM. TG was then added to the cell cultures to achieve the desired final concentrations as noted below. Untreated cultures received the same
volumes of dimethyl sulfoxide without TG. A 5 mg/ml stock of TN was
prepared in dimethyl sulfoxide and stored at room temperature. TN was
added to the cell cultures to a final concentration of 0.75 µM. Untreated cultures received the same volumes of
dimethyl sulfoxide without TN. Cell viability was assessed by counting cells on a hemocytometer after suspension in trypan blue dye.
Levels of Bcl-2 protein were measured by
Western blotting as described previously (30), using a human monoclonal
anti-Bcl-2 antibody (Pharmingen). GRP78 and GRP94 levels were measured
by Western blotting as described previously (31), using a monoclonal antibody to GRP78 (provided by D. Bole, University of Michigan) (15) or
a rabbit polyclonal antibody to GRP94 (-HS3) that immunocross-reacts with GRP78 (provided by M. Green, St. Louis University) (32).
Total RNA was isolated from cells
using Trizol (Life Technologies, Inc.). 10 or 20 µg of RNA were
separated according to size on a 1.2% agarose gel with 2.2 M formaldehyde (final concentration). After separation, the
RNA was transferred to a Zeta-Probe membrane (Bio-Rad) for 3 h
(Schleicher & Schuell Turboblotter). Plasmids encoding cDNA for
grp78 (p3C5 from A. Lee, University of Southern California),
grp94 (pcDER99-2 from M. Green), and CHO-B (from M. Wilson,
Scripps Clinic and Research Foundation) were used as the templates for
polymerase chain reaction amplification of the specific cDNA
inserts of interest. Probes were prepared by labeling with
[-32P]dCTP via random priming of the polymerase chain
reaction-amplified insert (Stratagene Prime-it kit). Use of polymerase
chain reaction-amplified fragments as the template for random-primed
probes eliminated a cross-reacting second band generated by the plasmid
vector. Zeta-Probe membranes were prehybridized for 30 min at 65 °C
in a buffer containing 1 mM EDTA, 0.5 M
NaH2PO4, pH 7.2, and 7% SDS. The membranes
were incubated overnight at 65 °C in fresh hybridization buffer
containing one of the labeled probes. After hybridization, the
membranes were washed twice at 65 °C for 15 min each in a buffer
containing 1 mM EDTA, 40 mM
Na2HPO4, pH 7.2, and 5% SDS. After the first
two washes, the membrane was monitored for background. If needed, a
final wash using 1 mM EDTA, 40 mM
Na2HPO4, pH 7.2, and 1% SDS was done at 65 °C for 15 min. The membranes were exposed to Kodak XAR-5 x-ray film at
80 °C. Blots were probed with radiolabeled DNA complementary to
grp78 or grp94 mRNA, exposed to film, and then subsequently probed with radiolabeled DNA complementary to CHO-B
mRNA. Blots were not stripped between hybridizations and were
stored moist at 4 °C. Pre- and post-treatment levels of the mRNA
for grp78 and grp94 were quantitated by
densitometry using a SciScan 5000 (U. S. Biochemical Corp.) with
Oberlin Scientific Bioanalysis software and were normalized according
to the constitutive level of CHO-B mRNA. Statistical differences
were derived using a paired t test of the mean values.
Nuclear run-off assays were performed using
modifications of previously published techniques (33-35). After
culture with the appropriate agent, WEHI7.2 and W.Hb12 cells (5 × 107 cells/reaction) were centrifuged at 500 × g for 5 min at 4 °C and washed twice in ice-cold
phosphate-buffered saline. Following the second wash, the cells were
resuspended in cell lysis buffer (10 mM Tris-Cl, pH 7.4, 3 mM CaCl2, and 2 mM
MgCl2) in a total volume of 40 ml. The cells were
centrifuged at 1000 × g for 5 min, resuspended in 1 ml
of cell lysis buffer, and added to 1 ml of Nonidet P-40 lysis buffer
(10 mM Tris-Cl, pH 7.4, 10 mM NaCl, 3 mM MgCl2, and 0.5% Nonidet P-40). After
resuspension, the cells were homogenized using a Dounce homogenizer
fitted with a B pestle. Phase-contrast microscopic examination of the
cell suspension was used to determine that the cells were free of
membrane components. Nuclei were collected by centrifugation of the
homogenized cells at 500 × g for 5 min. The nuclear
pellet was resuspended in 100 µl of glycerol storage buffer (50 mM Tris-Cl, pH 8.3, 40% (v/v) glycerol, 5 mM
MgCl2, and 0.1 mM EDTA), frozen in liquid nitrogen, and stored at 80 °C. Run-off transcription was initiated by resuspending the frozen nuclei in 100 µl of a reaction mixture containing 10 mM Tris-Cl, pH 8.0, 5 mM
MgCl2, 0.3 M KCl, 1 mM dithiothreitol, 40 units/ml RNasin, 1 mM ATP, 1 mM GTP, 1 mM CTP, and 25 µl of
[
-32P]UTP (3000 Ci/mmol) at 30 °C for 30 min. The
DNA was then digested by adding 1 µl of 20,000 units/ml RNase-free
DNase. Yeast tRNA (5 µl of 10 mg/ml) was added after the DNA
digestion. Newly transcribed RNA was purified by adding 500 µl of a
guanidinium thiocyanate solution (4 M guanidinium
thiocyanate, 25 mM sodium citrate, pH 7.0, 0.5%
N-lauroylsarcosine, and 0.1 M
-mercaptoethanol). The 32P-labeled RNA was isolated by
phenol/chloroform extraction followed by precipitation with sodium
acetate (70 µl of 2.0 M) and 1 volume of isopropyl
alcohol at
20 °C for 1 h in a microcentrifuge tube. The RNA
was pelleted by microcentrifugation at 12,000 rpm for 30 min at
4 °C. The pellet was washed once with 70% ethanol and repelleted at
12,000 rpm for 5 min at 4 °C. The 32P-labeled RNA was
denatured by heating at 60 °C in 6 × SSPE, 1 × Denhardt's solution, 0.5% (w/v) SDS, and 50% (v/v) formamide. Total
cpm of 32P were equilibrated and hybridized to slot-blotted
cDNAs for 60 h at 42 °C in 6 × SSPE and 1 × Denhardt's solution (0.5% SDS, 50 µg/ml denatured salmon sperm DNA,
and 50% formamide). Following hybridization, the filters were washed
in 2 × SSPE and 0.1% SDS at 42 °C for 30 min and then
rewashed once for 20 min in 0.2 × SSPE and 0.1% SDS at 56 °C.
After washing, the filters were exposed to Kodak XAR-5 film.
65 µg of plasmid DNA (for grp78, grp94, pBR322, and pCHO-B) were linearized with the appropriate restriction enzyme in a final volume of 300 µl. The DNA was denatured by addition of 33 µl of 1 N NaOH and boiled for 10 min. The linearized DNA was neutralized by addition of 6 × SSC to a final volume of 1.5 ml and placed on ice for 5 min. 125 µl of DNA were added per slot (final concentration of 5 µg/slot) and drawn onto the filter by vacuum. After adding the plasmid DNA, the filter was washed once with 6 × SSC and once with 2 × SSC and baked for 30 min at 80 °C.
The susceptibility to cell death following TG treatment was
investigated in WEHI7.2 cells, which do not express Bcl-2, and in
stable transfectants that express either a low level of Bcl-2 (W.Hb13)
or a high level of Bcl-2 (W.Hb12 and W.Hb15) (Fig.
1A). Consistent with earlier findings (30),
WEHI7.2 cells rapidly lost viability following treatment with 100 nM TG, whereas a derivative expressing a low level of Bcl-2
(W.Hb13) was killed more slowly, and derivatives expressing a high
level of Bcl-2 (W.Hb12 and W.Hb15) were resistant to TG-induced cell
death (Fig. 1B).
Steady-state levels of grp78 mRNA following TG treatment
were assessed by Northern blot analysis (Fig. 2). The
constitutively expressed marker CHO-B was used to control for minor
loading differences as described under "Experimental Procedures."
As shown by the Northern blot in Fig. 2A, the
grp78 mRNA level did not appear to increase following
treatment of WEHI7.2 cells with 100 nM TG. In multiple
experiments, the ratio of post-treatment to pretreatment levels was
quantitated at each time point by densitometry with normalization to
the CHO-B standard. The maximum ratio was 1.7 ± 0.2, which did
not represent a reproducible elevation above pretreatment levels
(p 0.05). The failure of TG treatment to induce an
elevation of the grp78 mRNA level was confirmed at
several other concentrations of TG (10, 50, and 300 nM)
(data not shown).
To determine whether or not the failure of TG treatment to increase
grp78 mRNA levels in WEHI7.2 cells was secondary to
early changes accompanying cell death, we examined the grp78
stress response in W.Hb12 cells, which are protected from apoptosis by bcl-2. As shown by the Northern blot in Fig. 2B,
the grp78 mRNA level did not appear to increase
following treatment of W.Hb12 cells with 100 nM TG. In
multiple experiments, the maximum post-treatment to pretreatment
grp78 mRNA ratio was 2.1 ± 0.4, which did not represent a significant elevation above pretreatment levels
(p 0.05). Northern blot analysis of two other
Bcl-2-expressing clones, W.Hb13 and W.Hb15, confirmed that
grp78 mRNA levels did not increase following treatment
with 100 nM TG (Fig. 3, A and B). Note that in Fig. 3, grp78 mRNA levels
actually decreased relative to CHO-B levels at 16 and 24 h after
TG addition. This observation was variable among experiments, including
those with WEHI7.2 and W.Hb12 cells. Note that we have previously
shown, in WEHI7.2 cells and derivatives expressing Bcl-2, that TG
treatment inhibits the ER Ca2+-ATPase, producing cytosolic
Ca2+ elevation and ER Ca2+ pool depletion (30,
36). Hence, the failure to significantly elevate
grp78/grp94 transcription following TG treatment
is not due to a failure of TG to disrupt Ca2+
homeostasis.
Levels of GRP78 and GRP94 proteins, assessed by Western blotting, were
the same in untreated WEHI7.2 and W.Hb12 cells, indicating that
bcl-2 does not affect basal levels of GRP78/GRP94 expression at the protein level (Fig. 4, A and
B). Furthermore, levels of GRP78 protein did not increase
following TG treatment in either WEHI7.2 or W.Hb12 cells (Fig.
4C).
Both WEHI7.2 and W.Hb12 cells up-regulated grp78 mRNA
levels by 6-7-fold when treated with 0.75 µM TN (Fig.
5). Thus, although ER Ca2+ pool depletion
failed to induce an up-regulation of grp78 mRNA, accumulation of unglycosylated proteins in the ER induced a strong up-regulation of grp78 mRNA levels. These findings
suggest that there is more than one signal transduction pathway for
grp78 induction (see "Discussion").
To assess if grp94 is regulated in the same manner as
grp78 in WEHI7.2 and W.Hb12 cells, we examined the
steady-state level of grp94 mRNA after treatment with
100 nM TG (Fig. 6). A modest elevation of
grp94 mRNA levels appeared to occur at 5 h after TG
addition in both WEHI7.2 and W.Hb12 cells. In multiple experiments, however, the maximum ratio of post-treatment to pretreatment
grp94 mRNA levels in WEHI7.2 cells was only 2.0 ± 0.5, which did not represent a reproducible elevation above base-line
levels (p 0.05). In W.Hb12 cells, the maximum ratio
was only 1.5 ± 0.2, which also did not represent a significant
elevation above base-line levels (p
0.05).
The preceding findings suggest that TG treatment does not signal an
increase in grp78/grp94 transcription in the
WEHI7.2 lymphoma cell line or its derivatives that express Bcl-2. To
confirm that this is the case, we measured the effect of TG treatment
on the transcription rate of grp78 and grp94
genes by nuclear run-off assays using isolated nuclei from WEHI7.2 and
W.Hb12 cells. An increase in newly expressed
grp78/grp94 message after TG treatment was not
detected in WEHI7.2 cells (Fig. 7A) or W.Hb12
cells (Fig. 7B). TN treatment, however, did induce a
significant increase in grp78 and grp94
transcription, which was detected by 5 and 7 h, respectively. This
indicates that grp78/grp94 transcription is not
induced by TG in WEHI7.2 cells or derivatives that express Bcl-2, but
is induced by TN.
Because earlier studies of grp78 regulation have emphasized
epithelial cells and fibroblasts (see the Introduction), as a positive
control, we examined the effect of TG treatment on grp78 mRNA levels in three epithelial breast cancer lines, Mm5MT,
MDA-MB-468, and MCF-7. Treatment of Mm5MT cells with 100 nM
TG did not induce cell death (Fig. 8D), but
did induce a 5-fold elevation of grp78 mRNA levels
detectable within 7 h of adding TG (Fig. 8B).
MDA-MB-468 and MCF-7 cells were also much less sensitive than WEHI7.2
cells to TG-induced cell death (Fig. 8D) and displayed
marked induction of grp78 mRNA levels in response to TG
treatment (Fig. 8, B and C).
To determine if the defect in TG-mediated grp78 signaling is
observed in other lymphoid cells, we measured the effect of TG treatment on grp78 mRNA levels in two additional
Bcl-2-negative mouse lymphoma lines, W7.MG1 and S49.1. grp78
transcription is induced by TN treatment in both of these lines (31).
The level of grp78 mRNA failed to increase following TG
treatment in W7.MG1 cells, which rapidly lost viability following TG
treatment, whereas the level of grp78 mRNA did increase
3-4-fold following TG treatment in S49.1 cells, which were relatively
resistant to TG-induced cell death (Fig. 9). These data
are consistent with the concept that a deficiency of grp78
induction increases susceptibility to TG-induced cell death.
We have discovered that the transcription of grp78 and grp94 is not significantly increased in WEHI7.2 cells in response to treatment with the ER Ca2+-ATPase inhibitor TG, even when apoptosis is inhibited by overexpressing grp78. Examination of two additional lymphoma lines revealed an absence of grp78 induction in W7.MG1 cells and 3-4-fold induction of grp78 in S49.1 cells following TG treatment. By comparison, TG treatment induced a marked elevation of grp78 mRNA levels in all three nonlymphoid lines tested (Mm5MT, MDA-MB-468, and MCF-7), consistent with studies indicating that TG treatment substantially induces grp78/grp94 transcription in epithelial cells and fibroblasts (13).
We have previously shown, in WEHI7.2 cells and derivatives expressing Bcl-2, that TG treatment inhibits the ER Ca2+-ATPase, producing cytosolic Ca2+ elevation and ER Ca2+ pool depletion (30, 36). Hence, the failure to significantly elevate grp78/grp94 transcription following TG treatment is not due to a failure of TG to disrupt Ca2+ homeostasis. Moreover, in the present study, we show that TN treatment induces a substantial grp78/grp94 transcriptional response. This observation is important for two reasons. First, it provides evidence that the grp78/grp94 stress response is not already maximally induced in WEHI7.2 cells. Second, it suggests that the grp78/grp94 stress response induced by Ca2+ mobilization may be regulated differently than that induced by TN. Ca2+ mobilization and inhibition of glycosylation have been shown to induce grp78/grp94 transcription through common promoter elements (12). Therefore, the deficiency in the TG-induced grp78/grp94 transcriptional response observed in WEHI7.2 cells is unlikely to reside at the promoter level. One possible explanation for our findings is that two independent ER-to-nucleus grp78/grp94 signaling pathways may exist: one Ca2+-mediated and the other mediated by glycosylation inhibition. Both pathways are operative in fibroblasts and epithelial cells, which induce grp78/grp94 in response to both TG and TN, but only the glycosylation inhibition signaling pathway appears to be operative in WEHI7.2 cells.
Little is known about the ER-to-nucleus signaling pathway that activates grp78/grp94 transcription. ER-to-nucleus signaling may be Ca2+/calmodulin-regulated (37) or may be mediated through tyrosine kinases and/or serine/threonine kinases (38, 39). Recently, it has been shown that IRE1p (Ern1), a yeast transmembrane serine/threonine kinase required for the induction of KAR2, the yeast homologue of grp78, may play a role in the ER-to-nucleus signaling pathway mediating KAR2/grp78 up-regulation in response to malfolded proteins (40, 41). Overexpression of IRE1p in fibroblasts produced a modest increase in the ability of transfectants to up-regulate grp78 in response to TG treatment (39). The WEHI7.2 cell line described in this report may be a useful model for the delineation of ER-to-nucleus signaling pathways. For example, it will be interesting to determine whether or not expression of IRE1p/Ern1 restores Ca2+-mediated grp78/grp94 transcriptional induction in these cells, thus further elucidating the role of IRE1p/Ern1 proteins in the pathway of grp78/grp94 induction.
Understanding ER-to-nucleus signaling pathways should provide insight into mechanisms that regulate apoptosis induction during ER Ca2+ pool depletion. Indeed, our findings suggest that cells deficient in grp78 stress response signaling are more susceptible to TG-induced apoptosis than cells that mount a grp78 stress response. These findings are consistent with those of earlier work by Lee and co-workers (26-28) in fibroblasts and epithelial cells, indicating that up-regulation of grp78 and coordinately regulated grp94, in response to ER Ca2+ pool depletion, prevents cell death. Hence, when the grp78/grp94 response was inhibited, fibroblasts died in response to treatment with agents that mobilize Ca2+ from the ER, including TG and the Ca2+ ionophore A23187. Using a grp78 antisense plasmid, they demonstrated that the inability to up-regulate grp78 resulted in increased cell death following A23187 treatment (26). Similarly, when grp78 induction was inhibited by amplification of the grp78 core promoter region, an increased sensitivity to A23187 was observed (27). Furthermore, when induction of grp78/grp94 was inhibited by ribozyme cleavage of newly transcribed grp94 mRNA, increased sensitivity to A23187 and TG was observed (28). Interestingly, abrogation of the grp78/grp94 stress response did not enhance the cytotoxicity of TN, suggesting that the increase in grp78/grp94 transcription provides specific protection against ER Ca2+ pool depletion (28). In agreement with their findings, we have found that WEHI7.2 and W.Hb12 cells were killed by TN (data not shown), even though TN increased grp78/grp94 transcription.
Understanding the role of grp78/grp94 in regulating cell death may have important implications for cancer therapy. In this regard, there is evidence that prior induction of grp78 can make cells less susceptible to death following treatment with photodynamic therapy (42), the superoxide-generating anti-cancer agent doxorubicin (43), and the topoisomerase inhibitor etoposide (44).
In summary, the findings reported here have three important implications. First, the grp78/grp94 stress response may be differentially regulated among different types of cells, with a much greater response observed in nonlymphoid cells than in lymphoid cells. Second, there may be at least two signal transduction pathways that mediate the grp78/grp94 stress response, one in response to ER Ca2+ mobilization and the other in response to protein glycosylation inhibition. Third, regulation of the grp78/grp94 stress response may be a major factor in deciding whether a cell lives or dies in response to disruption of intracellular Ca2+ homeostasis. Indeed, the absence of a Ca2+-mediated grp78/grp94 stress response may be the basis for the marked susceptibility of WEHI7.2 cells to TG-induced apoptosis.
We thank Roger Miesfeld, Keith Yamamoto, Marc Lippman, and Stanton Gerson for cell lines; Amy Lee for the GRP78 cDNA; Michael Green for the GRP94 cDNA and the GRP78/GRP94 antibody; Michael Wilson for the CHO-B cDNA; and David Bole for the GRP78 antibody. We also thank Mark Distelhorst for preparing figures. We are grateful to Satu Chaterjee, Nancy Oleinick, Dennis Templeton, and Hsing-Jien Kung for critically reviewing the manuscript.