The increase in cytosolic Ca2+ can be a consequence of external Ca2+ influx via the Cch1p⁄ Mid1p Ca2+ channel on the plasma membrane [1,3,6] or release of vacuolar Ca2+ into the cytosol t
Trang 1yeast Saccharomyces cerevisiae
Claudia-Valentina Popa, Ioana Dumitru, Lavinia L Ruta, Andrei F Danet and Ileana C Farcasanu Faculty of Chemistry, University of Bucharest, Romania
Introduction
Eukaryotic cells, from yeast to mammals, respond and
adapt to environmental stress by evolutionarily
con-served multicomponent endogenous systems that utilize
a network of signal transduction pathways to regulate
the adaptive and protective phenotype Changes in the
chemical or physical conditions of the cell that impose
a negative effect on growth demand rapid cellular
responses, which are essential for survival Molecular
mechanisms induced upon exposure of cells to such
adverse conditions are commonly designated as stress
responses
Ca2+-mediated signaling of stress conditions is used
by virtually every eukaryotic cell to regulate a wide
variety of cellular processes through transient increases
in cytosolic Ca2+ Budding yeast (Saccharomyces
cere-visiae) cells use Ca2+as a second messenger when they
are exposed to various environmental stress conditions, such as hypotonic and cold stress [1,2], hyperosmotic and salt stress [3], b-phenylethylamine-induced intracel-lular H2O2generation [4], or high pH [5] The increase
in cytosolic Ca2+ can be a consequence of external
Ca2+ influx via the Cch1p⁄ Mid1p Ca2+ channel on the plasma membrane [1,3,6] or release of vacuolar
Ca2+ into the cytosol through the vacuole-located
Ca2+ channel Yvc1p [7,8] After acting as a second messenger, cytosolic Ca2+ is restored to the normal very low levels through the action of Ca2+pumps and exchangers Thus, the Ca2+-ATPase Pmc1p [9,10] and the vacuolar Ca2+⁄ H+exchanger Vcx1p [11,12] inde-pendently transport cytosolic Ca2+ into the vacuole, whereas Pmr1p, the secretory Ca2+-ATPase, pumps cytosolic Ca2+ into the endoplasmic reticulum and
Keywords
aequorin; cytosolic calcium; oxidative stress;
Saccharomyces cerevisiae; yeast
Correspondence
I C Farcasanu, University of Bucharest,
Faculty of Chemistry, Sos Panduri 90-92,
Bucharest, Romania
Fax: +40 021 410 01 40
Tel: +40 721067169
E-mail: farcasanu.ileana@unibuc.ro
(Received 16 April 2010, revised 2 July
2010, accepted 28 July 2010)
doi:10.1111/j.1742-4658.2010.07794.x
The Ca2+-dependent response to oxidative stress caused by H2O2 or tert-butylhydroperoxide (tBOOH) was investigated in Saccharomyces cerevisiae cells expressing transgenic cytosolic aequorin, a Ca2+-dependent photopro-tein Both H2O2 and tBOOH induced an immediate and short-duration cytosolic Ca2+ increase that depended on the concentration of the stres-sors Sublethal doses of H2O2 induced Ca2+ entry into the cytosol from both extracellular and vacuolar sources, whereas lethal H2O2 shock mobi-lized predominantly the vacuolar Ca2+ Sublethal and lethal tBOOH shocks induced mainly the influx of external Ca2+, accompanied by a more modest vacuolar contribution Ca2+transport across the plasma membrane did not necessarily involve the activity of the Cch1p⁄ Mid1p channel, whereas the release of vacuolar Ca2+into the cytosol required the vacuolar channel Yvc1p In mutants lacking the Ca2+transporters, H2O2or tBOOH sensitivity correlated with cytosolic Ca2+ overload Thus, it appears that under H2O2-induced or tBOOH-induced oxidative stress, Ca2+ mediates the cytotoxic effect of the stressors and not the adaptation process
Abbreviations
BAPTA, 1,2-bis(2-aminophenoxy)ethane-N,N,N ¢,N ¢-tetraacetic acid; tBOOH, tert-butylhydroperoxide; WT, wild-type.
Trang 2Golgi, and is responsible for Ca2+ extrusion from the
cell [13,14] These responses are mediated by the
uni-versal Ca2+sensor protein calmodulin, which can bind
and activate calcineurin; this inhibits, at the
post-tran-scriptional level, the function of Vcx1p [11,15,16] and
induces the expression of PMC1 and PMR1 genes via
activation of the Crz1 transcription factor [15,16] The
release of Ca2+ from intracellular stores stimulates
extracellular Ca2+influx, a process known as
capacita-tive calcium entry [17] Conversely, the release of
vacu-olar Ca2+ via Yvc1p can be further stimulated by
Ca2+ from outside the cell as well as that released
from the vacuole by Yvc1p itself in a positive feedback
process called Ca2+-induced Ca2+release [8,18–20]
S cerevisiaeis a very useful model and an attractive
alternative to mammalian cell lines for studying the
effect of oxidative stress on the eukaryotic cells and
also an interesting model for studying antioxidants
in vivo [21–28] Living in an oxidant-rich medium
under natural conditions, it is expected that yeast cells
will respond promptly to variations in the oxidative
state of the environment How fast the cells respond to
oxidative stress and whether this type of response
is Ca2+-mediated are issues that are not clearly
understood
In this study, evidence is presented that the cytotoxic
effect of the exogenous oxidative stress induced by
H2O2or by the less hydrophilic oxidant
tert-butylhydr-operoxide (tBOOH) is triggered by transient elevations
in cytosolic Ca2+ as a result of both rapid influx from
outside the cell and of release from vacuolar stores
Results
H2O2induces a transient increase in cytoplasmic
Ca2+
As a free radical generator, H2O2 is known to have deleterious effects on cell growth, and exposure to high concentrations requires an immediate response for cell survival To determine whether the cell response to high concentrations of H2O2 is mediated by Ca2+ in
S cerevisiae, the wild-type (WT) strain BY4741 was transformed with a plasmid expressing the luminescent
Ca2+ reporter apoaequorin from a constitutive pro-moter After reconstitution of functional aequorin by addition of its cofactor coelenterazine, cells were exposed to H2O2 shock directly in the luminometer tube The cells responded promptly, as H2O2 initiated immediate and transient luminescence peaks, indicating sudden elevations in the cytosolic Ca2+ Ca2+ pulses could be recorded for H2O2 concentrations as low as 0.5 mm, and a sharp rise followed by a rapid fall in the cytosolic Ca2+-caused luminescence was noted for concentrations of 2 mm and over (Fig 1A) The pulse amplitude increased with H2O2 concentration (Fig 1B)
H2O2 treatment of the cells transformed with the empty vector under the same conditions did not elicit any luminescence response (data not shown) The luminescence spectra were reproducible for the range
of H2O2 concentrations used (0.5–10 mm), which covered toxic but nonlethal concentrations (0.5–2 mm) and lethal concentrations (3–10 mm)
Fig 1 Changes in cytosolic Ca 2+ upon exposure to H2O2 (A) Effect of H2O2on the intensity of the Ca 2+ -dependent cell luminescence WT BY4147 cells were transformed with the plasmid pYX212–cytAEQ The cells expressing coelenterazine-reconstituted aequorin were exposed
to various H2O2concentrations directly in the luminometer tube, as described in Experimental procedures The arrow indicates the addition
of H2O2 (B) Maximal response of H2O2-treated cells The WT cells transformed with pYX212–cytAEQ were subjected to H2O2shocks as in (A) (closed circles) The data are presented as the ratio of the maximum relative light units (RLU) to the average RLU recorded before H 2 O 2
shock Closed diamonds correspond to the response of control cultures not expressing aequorin (transformed with the empty vector pYX212) Each determination was repeated at least three times on different days, with no significant variations being seen (P < 0.05).
Trang 3Under H2O2stress, Ca2+is mobilized from both
external and internal sources
To further characterize the Ca2+ response to
exoge-nous H2O2, we examined whether the Ca2+flux comes
from an external or an internal source In the WT
cells, H2O2exposure induced sharp and transient
lumi-nescence peaks caused by the increase in cytosolic
Ca2+ (Fig 2A) The peak intensity was attenuated by
the presence of the membrane-impermeable Ca2+
che-lator
1,2-bis(2-aminophenoxy)ethane-N,N,N¢,N¢-tetra-acetic acid (BAPTA) (Fig 2D), suggesting that, at
least in part, the Ca2+ flux comes from outside the
cells Following the addition of BAPTA, the pattern of
the luminescence spectrum recorded for the WT strain
changed from a short-duration Ca2+ pulse (Fig 2A)
to a prolonged elevation of cytosolic Ca2+, which
gradually diminished within 2–3 min (Fig 2D)
Never-theless, addition of BAPTA did not completely
sup-press the cytosolic Ca2+ elevation, suggesting that, on
exposure to H2O2, internal stores are also mobilized
To test this possibility, the cytosolic Ca2+ burst was
also monitored in cells lacking the genes encoding the components of the main plasma membrane Ca2+ channel, Cch1p⁄ Mid1p It was found that, upon expo-sure to H2O2, the cch1D and mid1D null mutants exhibited robust, albeit different, responses Thus, whereas the luminescence peak caused by the Ca2+
flux decreased to approximately half in the cch1D cells (Fig 2B), in the mid1D mutant it was (surprisingly) slightly higher than in the WT cells (Fig 2C) This observation suggested that, if involved, Cch1p alone, independently of Mid1p, may be partially responsible for the Ca2+ influx under H2O2 stress As the differ-ence between the WT and mid1D cells was small, the tests described above were repeated on seven different days, with no obvious variations being seen from day
to day Statistical analysis showed that mid1D cells exhibited luminescence maximum peaks that were con-stantly higher than that recorded for the WT cells, with an average that was 10 ± 1.6% higher than the normal luminescence maximum intensity (P < 0.05)
In the parallel study, cch1D cells exhibited lumines-cence maximum peaks that were constantly lower than
F
50 s
Fig 2 Effect of mutations affecting Ca 2+ fluxes in response to H2O2 Isogenic strains expressing coelenterazine-reconstituted cytAEQ were exposed to H 2 O 2 (2 m M ; a toxic but sublethal concentration) directly in the luminometer tube When used, the membrane-impermeant Ca 2+
chelator BAPTA was added (5 m M final concentration) 1 min before the H2O2shock The arrows indicate the addition of H2O2 Each determi-nation was repeated at least three times on different days, with no significant variations being seen (P < 0.05) One typical luminescence spectrum is presented for each strain (A) WT strain (B) Null mutant cch1D strain (C) Null mutant mid1D strain (D) WT strain pretreated with BAPTA (E) Null mutant cch1D strain pretreated with BAPTA (F) Null mutant mid1D strain pretreated with BAPTA (G) Null mutant yvc1D strain (H) Null mutant yvc1D strain pretreated with BAPTA.
Trang 4that recorded for the WT cells, with an average that
was 33.5 ± 2.2% smaller than the normal
lumines-cence maximum intensity (P < 0.05)
Interestingly, BAPTA treatment of both cch1D
(Fig 2E) and mid1D (Fig 2F) cells attenuated the
Ca2+ influx to levels comparable to those detected for
the BAPTA-treated WT cells (Fig 2D), indicating
that, under H2O2 stress, external Ca2+ may enter the
cytosol via a transporter other than the Cch1p⁄ Mid1p
channel Similar results were obtained when BAPTA
was replaced with another Ca2+chelator, EGTA (data
not shown)
The addition of BAPTA or other Ca2+ chelators to
the medium equalized the H2O2-induced cytosolic
Ca2+peak in WT, cch1D and mid1D cells (Fig 2D–F),
but it still allowed a robust Ca2+ burst, suggesting
that, on exposure to the oxidant, internal stores are
also mobilized In yeast, the vacuole is the main
stor-age location for intracellular Ca2+, so the H2O2
-induced Ca2+ flux was monitored in cells lacking
Yvc1p, the vacuolar membrane channel responsible for
release of Ca2+ from the vacuole to the cytosol The
YVC1 gene deletion drastically reduced the amplitude
of the Ca2+ burst, indicating that most of the Ca2+
released into the cytosol after H2O2 exposure comes
from the vacuole via the Yvc1p channel
(approxi-mately 93%) However, a luminescence peak was still
noticeable in the yvc1D cells, reaching approximately
1⁄ 15 of the normal value (Fig 2G) This residual peak
was practically abolished by the addition of BAPTA
(Fig 2H), suggesting that the small Ca2+burst seen in
yvc1D cells was the result of Ca2+ influx from outside
the cell As BAPTA addition removes 50% of the
signal in the case of WT cells (Fig 2A,D), it seems
probable that, in the presence of H2O2, the vacuolar
Yvc1p channel is activated by cytoplasmic Ca2+ that
may come from outside the cell or be released from
the vacuole by Yvc1p itself in a positive feedback
pro-cess, as previously reported [8,18–20] Thus, by
dele-tion of YVC1, both the H2O2-induced component and
the Ca2+-induced component are removed If this is
so, the signal that disappears in BAPTA-treated WT
cells (Fig 2D) may be interpreted as the signal
through Yvc1p secondarily induced by external Ca2+
Adaptation to H2O2-induced oxidative stress is
favored under low-Ca2+conditions
As the elevation of cytosolic Ca2+ in response to
oxidative stress was different in the mutant cells
used, growth in media supplemented with H2O2 was
investigated It was found that the cch1D and yvc1D
cells were more tolerant to H2O2 than the WT cells,
suggesting that the lower level of cytosolic Ca2+ burst achieved during oxidative shock may result in a pro-tective effect on cell growth In this regard, mid1D cells, which exhibited the highest Ca2+ burst when exposed to H2O2shock, were slightly more sensitive to
H2O2 than WT cells and considerably more sensitive than cch1D and yvc1D cells (Fig 3A, upper right) At the same time, supplementation of the medium with the Ca2+ chelator EGTA augmented cell tolerance to
H2O2 (Fig 3A, middle), whereas supplementary Ca2+ augmented sensitivity to H2O2(Fig 3A, bottom right) Paradoxically, yvc1D cells, which exhibited the lowest cytosolic Ca2+ burst, were the most tolerant to exoge-nous H2O2 (Fig 3B), suggesting that lower cytosolic
Ca2+ bursts during oxidative shocks may actually increase the chances of cell survival To check this pos-sibility, transgenic cells expressing the YVC1 gene from
a galactose-inducible promoter were monitored in terms of sensitivity to oxidative stress WT cells over-expressing YVC1 are known to release more Ca2+into the cytosol [7] and, indeed, such cells were more sensi-tive to H2O2 than cells expressing the control vector (Fig 3C) Moreover, sensitivity increased with induc-tion time, probably because of increasing gene expres-sion induced by activation of the GAL1 promoter by galactose In contrast, cells overexpressing the PMC1 gene from a galactose-inducible promoter became more tolerant to H2O2than cells expressing the control vector (Fig 3C) Pmc1p transports Ca2+ to the vacu-ole, and its overproduction probably results in faster restoration of cytosolic Ca2+
Other mutations that alter cytosolic Ca2+were also investigated For example, whereas pmc1D and vcx1D null mutants were both more tolerant than the wild type, the double mutant pmc1D vcx1D was found to be hyper-sensitive to H2O2 (Fig 3B) This double mutant lacks
Ca2+in the vacuole, and therefore cannot release Ca2+ via Yvc1p Nevertheless, this strain is deficient in restor-ing cytosolic Ca2+levels after a stress burst [12], and is consequently less tolerant to exogenous oxidants Cytosolic Ca2+was reported to upregulate the calci-neurin-dependent degradation of Yap1p [29], the main transcription factor inducing gene expression in response to oxidative stress [19,20,30] If the high cyto-solic Ca2+achieved by cells when they are exposed to
H2O2 can result in calcineurin-dependent Yap1p degradation and hence sensitivity to oxidants, cells lacking functional calcineurin should gain a certain tolerance to H2O2 To check this possibility, the growth of cnb1D cells (which lack the gene encoding the regulatory subunit of calcineurin) under oxidative stress was tested It was noted that CNB1 deletion resulted in an increase of tolerance to H2O2 (Fig 3B),
Trang 5suggesting that the Ca2+-mediated toxicity of H2O2
may be also the result of Ca2+-mediated degradation
of Yap1p
Exposure to lethal H2O2concentrations results in
Ca2+mobilization predominantly from the
vacuole
In the experiments described above, Ca2+ appearance
in the cytosol was monitored in cells exposed to toxic,
but nonlethal, concentrations (as seen from the cell
viability following exposure; data not shown) of H2O2
To determine whether cells behaved differently when
subjected to stronger oxidative insults, WT cells were
exposed to a lethal concentration of H2O2(10 mm) In
this case, it was found that BAPTA pretreatment of
cells did not attenuate the Ca2+ burst (Fig 4A,B), indicating that, under lethal oxidative conditions, cells rely mainly on internal Ca2+ stores Under the same stress conditions, yvc1D cells exhibited only faint
Ca2+-mediated luminescence (Fig 4A,B), suggesting that when cells are confronted with higher, lethal con-centrations of H2O2, they utilize the vacuole as the main source for Ca2+to be directed to the cytosol
Alkylhydroperoxide also induces a transient increase in cytosolic Ca2+
To determine whether the Ca2+-mediated response to exogenous H2O2 is part of a more general oxidative stress response mechanism, a less hydrophilic oxidant, tBOOH, was considered This substance is largely used
A
B
C
Fig 3 Effect of H 2 O 2 -induced oxidative
stress on the cell growth of Ca 2+ channel
mutants (A) Growth properties of Ca 2+
channel mutants Isogenic strains
(WT, cch1D, mid1 and yvc1D) were spotted
(approximately 4 lL) in 10-fold serial
dilutions (from 10 7 cellsÆmL)1, left, to
103cellsÆmL)1, right) onto YPD plates
supplemented with the indicated chemicals
(4 m M H2O2, 20 m M EGTA, 10 m M CaCl2).
Early exponential growth phase cultures
were used to prepare the diluted cultures,
which were spotted on plates by means of
a replicator Cells were photographed after
3 days of incubation at 28 C The
concen-trations used were 4 m M H2O2, 20 m M
EGTA and 10 m M CaCl 2 (B) Effect of YVC1
and PMC1 overexpression on sensitivity to
oxidative stress WT cells transformed with
plasmids pGREG506DSalI (empty vector,
with fragment SalI–SalI excised),
pGAL11–YVC1 and pGAL11–PMC1 were
shifted from a fresh preculture to SG-Ura for
galactose induction of the genes, 6 or 16 h
before being stamped [approximately 4 lL,
in 10-fold serial dilutions from 10 7 cellsÆmL)1
(left) to 103cellsÆmL)1(right)] onto SG-Ura
agar plates containing H2O2 Time variation
was not observed in the case of the PMC1
galactose-induced phenotype Similar results
were obtained for tBOOH (1.5 and 2 m M ,
data not shown) (C) H2O2effect on growth
of mutants with altered Ca 2+ -related
phenotype The strains were spotted
(approximately 4 lL, 10 6 cellsÆmL)1) onto
YPD plates containing the indicated
concentrations of H 2 O 2
Trang 6to mimic the oxidative stress that causes formation of
alkyl hydroperoxides (e.g oxidation of unsaturated
lipid chains) As the minimal inhibitory concentration
for the WT strain is approximately 1 mm, Ca2+pulses
were recorded in aequorin-expressing WT cells exposed
to 0.25–1.5 mm tBOOH (Fig 5A) The pulse
ampli-tude increased with tBOOH concentration (Fig 5B)
When exposed to toxic, but nonlethal,
concentra-tions of tBOOH, WT cells promptly responded with a
transient elevation in cytosolic Ca2+, as seen by the
luminescence spectrum of the aequorin-expressing cells
(Fig 6A) This observation suggested that, as in the
response to H2O2, cells utilize Ca2+-mediated signaling
to initiate the defense mechanisms CCH1 deletion lowered the amplitude of the Ca2+ burst to approxi-mately 75% of the normal peak (Fig 6B), but MID1 deletion resulted in a Ca2+ cytosolic burst that was slightly higher than in the WT cells (Fig 6C); this was paralleled by an increased sensitivity of the mid1D mutant to tBOOH (Fig 7A,C) As the difference between WT and mid1D cells was small, the tests described above were repeated on seven different days, with no major variations being seen from day to day Statistical analysis showed that mid1D cells exhibited luminescence maximum peaks that were consistently greater than that recorded for WT cells, with an aver-age that was 12.14 ± 0.8% higher than the normal luminescence maximum intensity (P < 0.05) In the parallel study, cch1D cells exhibited luminescence maxi-mum peaks that were consistently lower than that recorded for WT cells, with an average that was 25.56 ± 1.2% lower than the normal luminescence maximum intensity (P < 0.05)
Surprisingly, BAPTA addition prior to exposure to tBOOH strongly attenuated the Ca2+ burst to less than one-quarter of the normal level, both in WT and mutant cch1D and mid1D cells (Fig 6D–F) This obser-vation suggested that, under tBOOH stress, external
Ca2+ may be predominantly responsible for the cyto-solic bursts, but that the major route of entry into the cell may, again, not be through the Cch1p⁄ Mid1p channel The contribution of vacuolar stores could not, however, be ruled out, because although the
Ca2+-dependent luminescence was drastically reduced
by the presence of BAPTA, it still peaked at approxi-mately one-fifth to one-quarter of the normal
Fig 4 Ca 2+ mobilization under lethal H2O2 conditions Isogenic
strains (WT and yvc1D) expressing coelenterazine-reconstituted
cytAEQ were exposed to 10 m M H2O2(lethal) in the absence (A) or
presence (B) of 5 m M BAPTA directly in the luminometer tube The
arrow indicates the addition of H 2 O 2 Each determination was
repeated at least three times on different days, with no significant
variations (P < 0.05) being seen One typical luminescence
spectrum is presented for each type of experiment RLU, relative
light units.
Fig 5 Changes in cytosolic Ca 2+ concentration upon exposure to tBOOH (A) Effect of tBOOH on the intensity of the Ca 2+ -dependent cell luminescence WT cells expressing coelenterazine-reconstituted cytAEQ were exposed to various tBOOH concentrations directly in the lumi-nometer tube The arrow indicates the addition of tBOOH (B) Maximal response of tBOOH-treated cells The WT cells transformed with pYX212–cytAEQ were subjected to tBOOH shocks as in (A) (closed squares) The data are presented as ratio of the maximum relative light units (RLU) to the average RLU recorded before tBOOH shock Closed diamonds correspond to the response of control cultures not expressing aequorin (transformed with the empty vector pYX212) Each determination was repeated at least three times on different days, with no significant variations being seen (P < 0.05).
Trang 7maximum, and this was followed by a steady state,
which was observed throughout the recording time To
determine whether the Ca2+also comes from the
vacu-ole, the cytosolic Ca2+ release under tBOOH stress
was monitored in yvc1D cells It was seen that, as
com-pared with WT cells, the Ca2+burst was attenuated in
yvc1D cells (towards approximately one-quarter of the
normal peak; Fig 6G), suggesting that the vacuolar
stores may also contribute to the total cytosolic pool
The burst was completely abolished when yvc1D cells
were preincubated with BAPTA (Fig 6H), indicating
that, under tBOOH shock also, the transient elevation
in cytosolic Ca2+ is the result of cooperative influx
from both outside the cell and from the vacuole, even
though, in this case, the vacuolar contribution seems
to be lower In terms of growth, cch1D and yvc1D cells
were more tolerant to tBOOH than WT cells, whereas
mid1D cells were more sensitive to tBOOH than WT
cells (Fig 7A,C) Supplementation of the medium with
the Ca2+chelator EGTA slightly augmented cell
toler-ance to tBOOH (Fig 7A, right), but external Ca2+
had no obvious effect on cell growth (data not shown)
It was found that the Ca2+-mediated cytosol lumi-nescence increased when the tBOOH concentration was varied from 0.75 mm (toxic, but nonlethal) to 1.5 mm (lethal) (Fig 7B, dotted lines), but did not change when the same conditions were applied to BAPTA-pretreated cells (Fig 7B, solid line) This observation indicated that, unlike the case with H2O2, the extent of vacuolar Ca2+ mobilization may not be dependent on tBOOH concentration
The susceptibility to tBOOH of other transgenic strains with altered Ca2+homeostasis was tested Simi-lar to what was seen with H2O2, YAP1 overexpression reduced, whereas PMC1 overexpression augmented, the tolerance of WT cells to tBOOH (data not shown) Other mutations that alter cytosolic Ca2+ were also investigated, and the response to tBOOH paralleled the response to H2O2 For example, whereas pmc1D and vcx1D null mutants were both more tolerant than the wild type, the double mutant pmc1D vcx1D was found to be hypersensitive to tBOOH (Fig 7C) Also, CNB1 deletion resulted in an increase in tolerance to tBOOH (Fig 7C) Thus, it seems that Ca2+ mediates
Fig 6 Effect of mutations affecting Ca 2+ fluxes in response to tBOOH Isogenic strains expressing coelenterazin-reconstituted cytAEQ were exposed to tBOOH (0.5 m M , a toxic but sublethal concentration) directly in the luminometer tube When used, the membrane-imper-meant Ca 2+ chelator BAPTA was added (5 m M final concentration) 1 min before the tBOOH shock The arrows indicate the addition of tBOOH Each determination was repeated at least three times on different days, with no significant variations being seen (P < 0.05) One typical luminescence spectrum is presented for each strain (A) WT strain (B) Null mutant cch1D strain (C) Null mutant mid1D strain (D)
WT strain pretreated with BAPTA (E) Null mutant cch1D strain pretreated with BAPTA (F) Null mutant mid1D strain pretreated with BAPTA (G) Null mutant yvc1D strain (H) Null mutant yvc1D strain pretreated with BAPTA RLU, relative light units.
Trang 8the cytotoxic effect of exogenous oxidants rather than
the adaptative process
Discussion
Aerobic cells are continuously exposed to oxidative
insults that have to be perceived and scavenged rapidly
to allow normal growth and development Studies
con-cerning the antioxidant arsenal of the cell are
numer-ous, and many defense mechanisms have been
elucidated, but how exactly the cell senses the threat of
oxidative species that are above the non-noxious thresholds is still not clearly understood Ca2+ is an important second messenger in the eukaryotic cell, and there is increasing evidence that cytosolic Ca2+ entry
in yeast is critical for survival under a variety of stress conditions, including hypo-osmotic or hyperosmotic shock [1,3,5,7], protein-unfolding agents [30], or anti-fungal drugs [31,32] The results presented in this study provide evidence that exogenous oxidative stress induces a transient increase in cytosolic Ca2+ that originates from both outside the cell and from the
A
B
C
Fig 7 (A) Effect of tBOOH-induced oxida-tive stress on cell growth Isogenic strains (WT, cch1D, mid1D and yvc1D) were spotted (approximately 4 lL) in 10-fold serial dilutions [from 10 7 cellsÆmL)1(left) to
10 3 cellsÆmL)1(right)] onto YPD plates supplemented with the indicated chemicals (0.75 m M tBOOH, 20 m M EGTA) Early exponential growth phase cultures were used to prepare the diluted cultures, which were spotted on plates by means of a repli-cator Cells were photographed after 3 days
of incubation at 28 C (B) Ca 2+
mobilization under lethal tBOOH concentrations WT cells expressing coelenterazine-reconstituted cytAEQ were exposed to toxic but nonlethal (0.75 m M , left) or lethal (1.5 m M , right) concentrations of tBOOH directly in the luminometer tube The arrow indicates the addition of tBOOH When used, BAPTA was added to a final concentration of 5 m M Each determination was repeated at least three times on different days, with no significant variations being seen (P < 0.05) One typical luminescence spectrum is pre-sented for each type of experiment (C) The effect of tBOOH on growth of mutants with altered Ca 2+ -related phenotype The strains were spotted (approximately 4 lL,
10 6 cellsÆmL)1) onto YPD plates containing the indicated concentrations of tBOOH.
Trang 9vacuole Toxic, but sublethal, concentrations of
exoge-nous H2O2 (1–2 mm) induced very rapid and transient
Ca2+ bursts Cytosolic Ca2+ elevations dropped to
roughly half the normal value when cells were
pretreat-ed with Ca2+ chelators (BAPTA or EGTA) shortly
before the oxidative shock, so part of the Ca2+ must
come from outside the cell However, the external
Ca2+ seems to be taken up through a system that is
different from the Cch1p⁄ Mid1p channel, as Ca2+
bursts were still clearly detectable in cch1D cells and
were higher than normal in mid1D mutants At the
same time, the process seems to be assisted by the
Yvc1 vacuolar channel, as the H2O2-triggered cytosolic
Ca2+ elevation was rather modest in the yvc1D
mutant The residual Ca2+ increase observed in the
yvc1D mutant could be explained by the existence of a
low-affinity Ca2+ system of unknown identity [30,33]
Nevertheless, as the H2O2 concentration increased
from very toxic to lethal doses (3–10 mm), the cells
gradually turned to internal vacuolar stores rather
than external Ca2+
The tolerance to high H2O2correlated inversely with
the amount of cytosolic Ca2+ achieved during
oxida-tive shock Thus, the mid1D mutant, which exhibited
the highest cytosolic Ca2+burst, was the least tolerant,
and the yvc1D mutant, with the lowest Ca2+load, was
the most tolerant Excessive or unregulated levels of
Ca2+ in the cytosol can lead to cell death [34] and,
indeed, there was a good correlation between
hyper-sensitivity to H2O2 or tBOOH and excessive cytosolic
Ca2+ elevation The triad Ca2+⁄ oxidative stress ⁄ cell
death has already been reported for other systems In
mammals, for instance, Ca2+ overload apparently
leads to mitochondrial dysfunction, reactive oxygen
species production and cell death [35] However, Ca2+
has also been reported to upregulate the
calcineurin-dependent degradation of Yap1p [29], the transcription
factor that regulates gene expression in response to
oxidative stress in yeast [24,25,36] In this respect, the
high cytosolic Ca2+ achieved by the cells when
exposed to lethal concentrations may result in Yap1p
degradation, a situation in which the cells become
hypersensitive to oxidative stress
The cell response to high H2O2 was paralleled by
the response to the less hydrophilic oxidant tBOOH
Here also, the cytosolic Ca2+ transients could be
detected upon exposure, with both external and
vacuo-lar contributions Again, the mid1D mutant exhibited
the highest Ca2+ load and the lowest tolerance,
whereas the yvc1D mutant exhibited the lowest Ca2+
burst, correlating with the highest tolerance to
tBOOH The only difference seemed to reside in the
prevalence of the external Ca2+ contribution to the
cytosolic pool, as seen by the low Ca2+burst in BAP-TA-pretreated cells Release of vacuolar Ca2+through Yvc1p requires activation by Ca2+ on the cytosolic side of the vacuolar membrane [8], and the low level of cytosolic Ca2+ achieved in BAPTA-pretreated cells might be accounted for by the inactive Yvc1p One can also speculate that tBOOH, being less hydrophilic than H2O2, diffuses more slowly in the yeast aqueous environment, exerting a milder signaling effect On the other hand, being unable to synthesize polyunsaturated fatty acids, the yeast has not evolved special protective mechanisms against lipid peroxidation, and is less pre-pared to respond to a stressor that mimics lipid hydro-peroxides
Special attention deserves to be paid to the discrep-ancy between the mid1D and cch1D mutants Although Mid1p and Cch1p cooperate to form a high-affinity influx system, differences between mid1D and cch1D mutants have already been reported [4,37] Among other things, Mid1p is known to localize not only in the plasma membrane, but also in the endoplasmic reticulum membrane [38] Recently, MID1 mutations have been reported to cause a defect in the cell wall structure [39] that might render mid1D cells more per-meable to Ca2+ or more susceptible to stressors such
as H2O2or tBOOH
The data above suggest that Ca2+ pulses serve fre-quently, but not invariably, to transduce an oxidative burst signal However, Ca2+ seems unlikely to be involved in the adaptation to oxidative stress Yap1p, the major yeast transcription factor responsible for the transcriptional regulation of many genes involved in the response to oxidative stress, is upregulated directly
by the presence of oxidants [40–43], whereas Ca2+ mediates its calcineurin-dependent degradation [29] Thus, it is probable that, under strong oxidative condi-tions, the cytosolic Ca2+ elevations actually signal the ultimate possible way out, which is cell death This would explain why, under extreme H2O2stress, the cell blocks the import of external Ca2+ and turns in a last effort to internal stores
Experimental procedures
Strains and culture conditions The S cerevisiae strains used in this study were isogenic with the WT parental strain BY4741 (MATa; his3D1; leu2D0; met15D0; ura3D0) [44] The deletion mutant strains used were Y04847 (BY4741, cch1::kanMX4), Y01153 (BY4741, mid1::kanMX4), Y01863 (BY4741, yvc1::kan-MX4), Y04374 (BY4741, pmc1::kanMX4), Y03825 (BY4741, vcx1::kanMX4), Y13825 (MATa; his3D1; leu2D0;
Trang 10lysD0; ura3D0 vcx1::kanMX4) and Y05040 (BY4741,
cnb1::kanMX4) All strains were obtained from
EURO-SCARF (European S cerevisiae Archive for Functional
Analysis, Institute of Molecular Biosciences Johann
Wolf-gang Goethe-University Frankfurt, Germany) Strain
VP012 (BY4741, pmc1::kanMX4 vcx1::kanMX4) was
obtained by crossing of strains Y04374 and Y13825
fol-lowed by diploid sporulation and random spore analysis
The absence of PMC1 and VCX1 from VP102 cells was
checked by colony PCR with primers specific for PMC1
(forward primer, 5¢-ATGTCTAGACAAGACGAAAAT-3¢;
reverse primer, 5¢-GAAATGACATCACCGACTAA-3¢)
and for VCX1 (forward primer,
5¢-ATGGATGCAACTA-CCCCACT-3¢; reverse primer,
5¢-GGATAACTCCAATAT-TTTTC-3¢) Internal control primers for ACT1 (forward
primer, 5¢-GAGGTTGCTGCTTTGGTTAT-3¢; reverse
primer, 5¢-GCGGTTTGCATTTCTTGT-3¢) were included
in all PCR reactions Cell growth and manipulation were
performed as previously described [45] Strains were grown
in standard YPD, SD or SG supplemented with the
neces-sary amino acids For solid media, 2% agar was used
Plasmid construction
The expression vectors harboring PMC1 or YVC1 were
obtained as follows The corresponding ORFs were
ampli-fied from the yeast genome by PCR, and were subsequently
cloned into vector pGREG506 [46], which allows for
galac-tose-inducible expression via the GAL1 promoter Plasmid
pGREG506 was purchased from EUROSCARF The
fol-lowing primers were used: for PMC1, 5¢-CATTGGAT
CCATGTCTAGACAAGACGAAAAT-3¢ (forward primer,
introducing the BamHI site, underlined) and 5¢-TACTGTC
GACTTAATAAAAGGCGGTGGACT-3¢ (reverse primer,
introducing the SalI site, underlined); and for YVC1, 5¢-TA
CTGTCGACATGGTATCAGCCAACGGCGA-3¢ (forward
primer, introducing the SalI site, underlined) and 5¢-GAG
ACTCGAGTTACTCTTTCTTATCCTTTA-3¢ (reverse
pri-mer, introducing the XhoI site, underlined) The PMC1 and
YVC1ORFs obtained by PCR were purified with a Wizard
SV gel and PCR Clean-up System (Promega), cut with
the appropriate restriction enzymes, and cloned into the
BamHI–SalI and SalI–XhoI sites of pGREG506,
respec-tively, to generate pGAL1–PMC1 and pGAL1–YVC1 As
control vector, pGREG506 with the SalI–SalI fragment
deleted was constructed (pGREG506DSalI) This deletion
removes the HIS3 gene used as a counterselective marker
of pGREG vectors for the in vivo plasmid recombination
[46]
Growth assessment
Overnight precultures were inoculated in fresh media at a
density of 105cellsÆmL)1, and cells were then incubated
with shaking (200 r.p.m.) at 28C for 2 h before H2O2,
tBOOH, BAPTA or EGTA was added from sterile stocks prepared in 0.1 m Mes⁄ Tris (pH 6.5) The influence of stressors on cell growth was monitored at time intervals
by determining the attenuance (D) of a cell suspension at
600 nm (Shimadzu UV–visible spectrophotometer, UV mini 1240) For spot assays, fresh cell cultures of
D600 nm 1 were diluted 10-fold, 100-fold, 1000-fold and 10 000-fold, and stamped on agar plates with a replicator
In vivo monitoring of oxidative stress-induced
Ca2+pulse Monitoring of cytosolic Ca2+was performed with an apo-aequorin cDNA expression system [47] Yeast strains were transformed with the plasmid pYX212–cytAEQ, containing the apoaequorin gene under the control of the TPI pro-moter, generously provided by E Martegani (Department
of Biotechnology and Biosciences, University of Milano-Bicocca, Milan, Italy) [48] Yeast transformation was per-formed with a modified lithium acetate method [49], and Ura+transformants were selected for growth on SD med-ium lacking uracil For luminescence assays, overnight pre-cultures of cells expressing the apoaequorin gene were diluted in fresh SD-Ura medium to a density of
5· 106 cellsÆmL)1 The cells were incubated with shaking for another 2 h at 28C, before being harvested by centri-fugation The cells were washed three times (by centrifuga-tion at 9167 g, 1 min) and resuspended at a density of approximately 109cellsÆmL)1 in 0.1 m Mes⁄ Tris buffer (pH 6.5) To reconstitute functional aequorin, 50 lm native coelenterazine (Sigma; stock solution of 1 lgÆlL)1in meth-anol) was added to the cell suspension, and the cells were incubated for 2 h at 28C in the dark Aliquots containing approximately 107cells were harvested, and excess coelen-terazine was removed by centrifuging three times (9167 g,
30 seconds) The cells were resuspended in 0.1 m Mes⁄ Tris buffer (pH 6.5) and transferred to the luminometer tube A cellular luminescence baseline was determined by 1 min of recording at 1 s intervals The exogenous oxidants H2O2 and tBOOH were added from sterile stocks to give the final concentrations indicated, and light emission was monitored with a single-tube luminometer (Turner Biosystems,
20n⁄ 20) The light emission was monitored for at least
5 min after the stimulus at 1 s intervals, and reported as relative luminescence units⁄ s When used, the calcium che-lator BAPTA was added to the cell suspension to the final concentration of 5 mm 1 min before application of the stress To ensure that total reconstituted aequorin was not limiting in our assay, at the end of each experiment aequorin expression and activity were checked by lysing cells with 10% Triton X-100, and only the cells with con-siderable residual luminescence were considered Multiple tests, on different days (a minimum of three for large differ-ences between transgenic lines; a maximum of seven for