© 1995 American Heart Association, Inc.
Articles |
Effects of Hypoxanthine–Xanthine Oxidase on Ca2+ Stores and Protein Synthesis in Human Endothelial Cells
From the Respiratory Division, Hôpital Cantonal Universitaire de Genève (Switzerland).
Correspondence to D. Dreher, Hôpital Cantonal Universitaire, Division de Pneumologie, 24, rue Micheli-du-Crest, 1211 Geneva 14, Switzerland.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abstract We have investigated the effects of reactive O2 metabolites generated by the hypoxanthine–xanthine oxidase (HX-XO) system on intracellular Ca2+ and its relation with protein synthesis in human umbilical vein endothelial cells (HUVECs). Spectrofluorometry with fura 2 showed that the oxidative stress induced a rapid transient rise in cytosolic [Ca2+], followed by a sustained elevation above the baseline value. In the presence of La3+, which blocks Ca2+ influx from the extracellular medium, a transient [Ca2+] increase was still observed, but the sustained rise was suppressed. The HX-XO–related [Ca2+] changes were completely prevented by pretreatment with thapsigargin, which depletes intracellular Ca2+ stores. Hence, the effects of HX-XO on Ca2+ homeostasis were due to mobilization of Ca2+ from the intracellular stores with subsequent influx of extracellular Ca2+. HX-XO mobilized more of sequestered Ca2+ than did thrombin, a receptor agonist that depletes only a part of the intracellular Ca2+ stores (the hormone-sensitive stores). To determine the relevance of the HX-XO–related depletion of Ca2+ stores for cell function, we investigated the role of Ca2+ mobilization in the regulation of protein synthesis. Overall protein synthesis in HUVECs was markedly reduced by thapsigargin, which depletes both hormone-sensitive and -insensitive stores, but was not substantially affected by thrombin. Manipulation of the refilling of the Ca2+ stores via the availability of extracellular Ca2+ significantly influenced the thapsigargin-related and the HX-XO–related inhibition of overall protein synthesis. A corresponding effect of extracellular [Ca2+] was seen in polyribosome distribution profiles, which reflected an inhibition of translation initiation in both treatments. Thus, depletion of Ca2+ stores appeared to be involved in the inhibition of protein synthesis at the initiation level by both thapsigargin and HX-XO. These results indicate that (1) the cytosolic [Ca2+] changes induced by HX-XO result from mobilization of Ca2+ from intracellular stores and subsequent influx of extracellular Ca2+, (2) the HX-XO–related mobilization of sequestered Ca2+ includes hormone-insensitive pools, and (3) the depletion of hormone-insensitive Ca2+ stores appears to be in part responsible for the inhibition of protein synthesis by HX-XO.
Key Words: human umbilical vein endothelial cells • intracellular Ca2+ • reactive O2 species • thapsigargin • thrombin
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reactive O2 species (ROSs) cause cell damage in inflammation, ischemia/reperfusion, and atherosclerosis. In these conditions, ROSs are generated by sequential monovalent reductions of molecular oxygen, yielding the superoxide radical (O2-), hydrogen peroxide (H2O2), and the hydroxyl radical (OH · ). The vascular endothelium is often the prime target for oxidative stress.1 2 To investigate the molecular pathways of endothelial cell response or injury, the effects of oxidants have been tested in primary endothelial cell culture. Although these systems have provided ample evidence that oxidative stress alters Ca2+ homeostasis in endothelial cells, the effect on intracellular Ca2+ is complex and depends on the ROSs that are involved.3 4 5 6 7 The hypoxanthine–xanthine oxidase (HX-XO) system, which generates O2- and H2O2, was shown to induce a rapid rise in cytosolic free Ca2+ ([Ca2+]i) in pig aortic,3 bovine aortic,4 and human umbilical vein endothelial cells (HUVECs).5 7 The [Ca2+]i increase was markedly reduced in the absence of extracellular Ca2+ and was therefore thought to result from changes in membrane permeability.4 5 Nonetheless, the mechanism of the HX-XO–induced [Ca2+]i changes remains unclear.
Experiments in intact smooth muscle,8 renal epithelial,9 and, most recently, venous endothelial cells10 indicate that ROSs can mobilize Ca2+ from intracellular stores. HX-XO–derived O2 intermediates were shown to inhibit the Ca2+-ATPase in isolated endoplasmic reticulum (ER) membranes.11 12 Inhibition of the ER Ca2+- ATPase in the endothelial cell would lead to the release of Ca2+ from the intracellular stores in the cytosol and cause a marked [Ca2+]i increase.13 Both cytosolic Ca2+ and Ca2+ sequestered in intracellular pools have emerged as key regulators of cell function.14 15 Thus, changes in Ca2+ homeostasis might explain much of the endothelial cell dysfunction observed in oxidative stress.16 17 Ca2+ sequestered in intracellular stores plays an important role in the regulation of protein synthesis, and depletion of Ca2+ stores results in a marked fall in overall protein synthesis in different cell types.15 This relation makes protein synthesis a sensitive parameter to assess the consequences of a disruption of Ca2+ homeostasis.18 However, the role of intracellular Ca2+ in the regulation of protein synthesis has yet not been defined in endothelial cells.
In the present study, we investigated the disruption of Ca2+ homeostasis by HX-XO and its consequences for protein synthesis in human endothelial cells. The effects of HX-XO were compared with those of two Ca2+ agonists, thapsigargin and thrombin, which mobilize intracellular Ca2+ stores in HUVECs through inhibition of the ER Ca2+-ATPase13 and through a receptor-mediated mechanism,19 respectively. It is shown that the HX-XO–related [Ca2+]i rise in endothelial cells is not simply a result of cell membrane damage but is due to the mobilization of Ca2+ from intracellular pools. We differentiated the Ca2+ stores affected by the oxidative stress and tested the physiological relevance of the different Ca2+ pools in HUVECs by investigating their role in the regulation of protein synthesis. Our findings show that the depletion of hormone-insensitive Ca2+ stores by both thapsigargin and the oxidative stress was associated with a marked inhibition of overall protein synthesis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Materials
All chemicals were of analytical grade and came from Sigma Chemical Co or Fluka. Bovine milk xanthine oxidase was obtained from Calbiochem or Sigma. The fura 2 acetoxymethyl ester (fura 2-AM), ionomycin calcium salt, bovine erythrocyte superoxide dismutase, and bovine liver catalase were from Calbiochem. Thapsigargin was obtained from LC Services Co. RPMI 1640 medium was from GIBCO or Biological Industries. Fetal calf serum (FCS) came from Seromed. Endothelial growth factor (EGF) from bovine hypothalamus was from Serva or UBI. L-[2,6-3H]Phenylalanine (3H-Phe, 40 to 60 Ci/mmol) was from Amersham. For liquid and filter scintillation counting, we used Aquassure from NEN and Filter-Count from Packard.
Cell Culture
HUVECs were isolated from human umbilical cords as described
previously.20 Cells were cultured in RPMI 1640 culture
medium (0.4 mmol/L Ca2+) supplemented with 25 mmol/L
HEPES, 50 U/mL penicillin, 50 µg/mL streptomycin, 30 µg/mL EGF, 90
µg/mL heparin, and 10% FCS on gelatin-coated 10-mm glass coverslips
for fluorometry or on gelatin-coated 35-mm plastic Petri dishes for
determination of protein synthesis. In all experiments, cells were used
after the second passage, at least 1 day after confluence.
Measurement of [Ca2+]i
Measurements of [Ca2+]i were
performed with the fluorescent Ca2+-sensitive probe
fura 2 in intact cell monolayers by using the method of Wickham et
al.21 For incubation with fura 2 and fluorometric
readings, a Krebs-Ringer bicarbonate/HEPES buffer (KHB containing
[mmol/L] CaCl2 0.4, NaCl 118, KCl 4.75,
KH2PO4 1.2, MgSO4 1.2,
NaHCO3 25, glucose 5, and HEPES 20; pH 7.4) was used. Cell
monolayers were loaded for 30 minutes at 37°C in KHB containing 4
µmol/L fura 2-AM, followed by incubation with fresh KHB for 20
minutes at 37°C. The loading conditions may lead to sequestration of
fura 2-AM into the ER and mitochondria. However, these organelles have
high intraorganellar [Ca2+] (>1 µmol/L), where
fura 2 would be saturated with Ca2+ and would not
influence the measured [Ca2+]i
changes.22 Possible artifacts by
Ca2+-induced fura 2 release from secretory vesicles
were not observed, because measurements of
[Ca2+]i changes in
Ca2+-free buffer were avoided. Before measurements,
cells were stored up to 90 minutes in KHB at room temperature.
Fluorometric readings were performed with a Perkin-Elmer LS-3
fluorescence spectrometer at 37°C with excitation at 340 nm and
emission at 505 nm. For calibration, maximum
[Ca2+]i levels were obtained by
increasing [Ca2+]o to 2 mmol/L in the
presence of 10 µmol/L ionomycin, and minimum
[Ca2+]i was measured after the
addition of 10 mmol/L EGTA and 60 mmol/L Tris at pH 8.8. Key
experiments were repeated on a Jasco CAF-110 fluorometer, which allowed
calculation of [Ca2+]i changes from
the ratio of emission after excitation of 340 and 380 nm. No
appreciable differences in the results were found between the
calculation of [Ca2+]i from mono- or
double-wavelength excitation, respectively.
45Ca2+ Efflux
Ca2+ efflux across the plasma membrane was
determined in cells preloaded with 5 µCi/mL
45Ca2+ for 24 hours. Monolayers were
washed and exposed to HX-XO in nonradioactive RPMI culture medium. At
different time intervals up to 1 hour, small aliquots of the medium
were withdrawn and counted for 45Ca. Net
45Ca2+ release from the cells was
expressed in percentage of total radioactivity (supernatant+cells).
Extracellular Ca2+
To achieve low [Ca2+]o,
1 mmol/L EGTA was added to the medium, resulting in a calculated free
[Ca2+] of 49 nmol/L, whereas high
[Ca2+]o was obtained by the addition
of CaCl2. For evaluation of sequestered
Ca2+ released by ionomycin, the free
[Ca2+]o was buffered to a
[Ca2+] close to
[Ca2+]i to minimize ionomycin-induced
Ca2+ exchange across the plasma membrane (150 nmol/L
[Ca2+]o, 71 µmol/L
CaCl2, and 100 µmol/L EGTA added to a
Ca2+-free KHB). To block plasma membrane
Ca2+ channels, 0.5 mmol/L LaCl3 was
added to the standard KHB, resulting in a concentration of free
La3+ of 20 µmol/L (18.4 to 24.6 µmol/L) in the
bicarbonate buffer.23 In our cell system, La3+
was as effective as EGTA in blocking Ca2+ entry and
was preferentially used during fluorometry and for polyribosome
distribution profiles (see below), because EGTA promotes cell
detachment in the calibration procedure and may destabilize the
ribosome complex, respectively.
Oxidative Stress
Extracellular O2- was generated by the
HX-XO system: in the presence of 2 mmol/L hypoxanthine, the
concentration of xanthine oxidase was adjusted to an initial production
of 10 nmol O2-/mL per minute, by
following the reduction rate of cytochrome C with a double-beam
spectrophotometer at 550 nm. Exposure to HX-XO was done in KHB, except
for experiments with incorporation of 3H-Phe, where phenol
red–free RPMI (GIBCO) without EGF, heparin, or FCS was used. In RPMI,
the concentration of xanthine oxidase required to achieve the same
O2- production rate was about two times higher
than that in KHB. However, xanthine oxidase activity was not affected
by the presence or absence of Ca2+.5 If
not otherwise stated, 2 mmol/L hypoxanthine was included in the medium
in all parallel experiments during the time of the exposure to xanthine
oxidase. For fluorometric readings, hypoxanthine was added to the
medium after loading with fura 2.
Incorporation of 3H-Phe
Overall protein synthesis was measured from the incorporation of
3H-Phe into total proteins in cells incubated for 15
minutes in RPMI without EGF, heparin, or FCS, supplemented with 2
µCi/mL 3H-Phe. Cell monolayers were washed and scraped in
phosphate buffer, pH 7.4, and sonicated for 10 seconds. In one aliquot
of the sonicate, proteins were precipitated with 10% trichloroacetic
acid, collected on Whatman GF/C filters, and counted for
3H. Another aliquot was solubilized in 0.8N NaOH for
determination of total proteins according to Lowry et
al.24 All 3H-Phe counts were related to total
protein content.
Polyribosome Size Distribution
Polyribosome distribution profiles were obtained by
centrifugation of cell extracts through linear sucrose density
gradients (15% to 55% [wt/vol]) and determination of absorbance
at 260 nm in equal fractions.25 Monosome-to-polysome
ratios were determined after integration of gaussian distribution
functions fitted to the 80S monomer peak and the polyribosome peak,
respectively.
Statistical Analysis
In the experimental design, each series was performed on HUVECs
from a different donor. To achieve nonparametric comparisons reflecting
the paired/repeated measures experimental design, ANOVA was performed
on ranks within series (blocks).26 Statistical
significance was set at P<.05. If not otherwise stated,
data are shown as mean±SEM.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Effect of HX-XO on [Ca2+]i
[Ca2+]i was recorded by spectrofluorometry in cell monolayers loaded with fura 2. Characteristic on-line readings are shown in Fig 1
,
while the statistics from repeated experiments are given in the
Table. After stabilization of
[Ca2+]i for several minutes, a median
baseline value of 147 nmol/L was measured in the presence of 0.4 mmol/L
[Ca2+]o (n=44). The addition of
xanthine oxidase (10 nmol O2-/mL per
minute) induced a transient increase in
[Ca2+]i, which reached a
maximum within the first minute (mean
[Ca2+]i increase at the peak, 699
nmol/L), followed by a slow decrease to a new level above the baseline
(Fig 1A, Table). The HX-XO–stimulated
[Ca2+]i increase was dose dependent,
with saturation of the response at doses above 10 nmol
O2-/mL per minute (35±3%, 87±12%,
and 100% of the maximum [Ca2+]i
response with 5, 10, and 20 nmol O2-/mL
per minute, respectively). To test for the specificity of the
HX-XO–stimulated [Ca2+]i increase,
the effect of xanthine oxidase was measured in the absence of
hypoxanthine or after preincubation of the enzyme in 3 mmol/L of the
specific inhibitor allopurinol for 30 minutes at 37°C (final
concentration of allopurinol, 100 µmol/L). In these conditions, the
[Ca2+]i response was <10% of the
increase induced by HX-XO in control conditions (data not shown).
Superoxide dismutase (200 U/mL) plus catalase (200 U/mL) or the free
radical scavenger 5,5-dimethyl-1-pyrroline-N-oxide (5
mmol/L) reduced the HX-XO–related
[Ca2+]i increase by 84±12% and
75±8%, respectively (n=4).
|
|
To determine the role of intracellular Ca2+
mobilization in the HX-XO–related
[Ca2+]i changes, the
Ca2+ influx from the extracellular space was
specifically blocked with 20 µmol/L La3+. In this
condition, the HX-XO–induced peak was reduced to 181 nmol/L, and no
consistent sustained change from the baseline value was found (Fig 1B
,
Table). Correspondingly, the addition of La3+ during the
sustained plateau phase after HX-XO returned
[Ca2+]i toward baseline values (not
shown). Depletion of intracellular stores with 1 µmol/L thapsigargin
caused a mean [Ca2+]i rise of 693
nmol/L. The pretreatment with thapsigargin completely abolished the
[Ca2+]i change by HX-XO (Fig 1C,
Table). Inversely, pretreatment with HX-XO reduced the average
thapsigargin-induced rise to 133 nmol/L (Fig 1A, Table). To control for
inactivation of thapsigargin by the oxidant, the drug was preincubated
for 30 minutes at 37°C with HX-XO (initial
O2-, 200 nmol/mL per minute), followed
by scavenging of the accumulated H2O2 with
catalase (20 U/µL). The oxidant-treated thapsigargin retained the
full [Ca2+]i activity compared with
the drug incubated in HX without XO. Next, we compared the action of
HX-XO on intracellular stores with that of the receptor agonist
thrombin. The maximal effective dose of thrombin (0.5 U/mL) stimulated
a mean [Ca2+]i increase of 433 nmol/L
(Fig 1, Table). In the continuous presence of the drug, no further
response to a second dose of thrombin or to another receptor agonist
like bradykinin or histamine was elicited (Table). In contrast, HX-XO,
given 5 to 10 minutes after thrombin, stimulated a mean
[Ca2+]i rise of 56 nmol/L (Fig 1D,
Table). Because the thrombin receptor rapidly desensitizes after
activation, this effect might have been due to refilling of the
Ca2+ stores. However, a consistent
[Ca2+]i increase from HX-XO after
activation was still observed in the presence of La3+ (mean
rise, 33 nmol/L, Table), which suppresses Ca2+
influx and thus prevents refilling of the Ca2+
stores (see below). Hence, the oxidative stress addressed intracellular
stores not depleted by thrombin or another receptor agonist
(hormone-insensitive stores).
The effect of thrombin on intracellular Ca2+ takes
place within seconds,14 whereas Ca2+
mobilization by thapsigargin is a slower process.13
Although the shape of the Ca2+ transient was
comparable to that induced by thapsigargin, the kinetics of the
Ca2+ release by HX-XO are unknown. Fluorometry does
not allow reliable long-term recording of
[Ca2+]i because of loss of fura 2.
Therefore, to compare the mobilization of intracellular
Ca2+ by HX-XO and thapsigargin over longer time
periods, Ca2+ efflux across the plasma membrane was
traced with 45Ca2+. The results are
depicted in Fig 2. The kinetics of the
45Ca2+ release were comparable with that
described in monolayers of bovine pulmonary artery endothelial
cells.27 Net 45Ca2+ efflux
in HX-XO–treated monolayers shows an increase to 61%, 72%, and 77%
of total cell Ca2+ after 10, 20, and 30 minutes
compared with 29%, 34%, and 41% in control cells. The
45Ca2+ efflux in thapsigargin (100
nmol/L)–treated cells (61%, 77%, and 82%) was not significantly
different from that in cells treated with HX-XO. These data were
consistent with mobilization of the same pools of sequestered
Ca2+ by HX-XO and thapsigargin.
|
Effect of HX-XO on Intracellular Ca2+ Stores
To assess the filling state of the Ca2+ stores,
total intracellularly sequestered Ca2+ was
determined from the transient [Ca2+]i
increase induced by ionomycin in the presence of 150 nmol/L
[Ca2+]. In cells exposed to thapsigargin (100
nmol/L), thrombin (0.5 U/mL), or HX-XO, the ionomycin-related
[Ca2+] increase was measured immediately after
exposure in KHB (0.4 mmol/L Ca2+) for 15 or 30
minutes or after 30-minute exposure followed by 30-minute incubation in
KHB without the drug. In Fig 3, the content of the
Ca2+ stores is expressed as percentage of that
observed in the control cells. All three treatments markedly reduced
the ionomycin-releasable Ca2+ pools after exposure
for 15 minutes ([Ca2+]i transient,
340±81, 670±39, and 302±25 nmol/L after thapsigargin, thrombin, or
HX-XO, respectively). Immediately after the 30-minute exposure,
sequestered Ca2+ was unchanged in
thapsigargin-treated and HX-XO–treated cells compared with the
15-minute exposure ([Ca2+]i transient,
307±46 and 316±55 nmol/L, respectively) but had increased in the
continued presence of thrombin (1103±66 nmol/L). The following
30-minute incubation with fresh KHB did not change sequestered
Ca2+ in the thapsigargin-treated cells
([Ca2+]i transient, 297±45 nmol/L),
whereas partial and complete recovery of the stores was observed in
cells that had been exposed to HX-XO (804±160 nmol/L) or thrombin
(1320±312 nmol/L), respectively. The refilling of
Ca2+ stores in thrombin-exposed cells was
suppressed, even in the absence of the drug, when extracellular
Ca2+ was chelated with 1 mmol/L EGTA
([Ca2+]i transient, 481±74, 468±41,
and 367±41 at 15, 30, and 60 minutes, respectively). This is in
accordance with previous findings that repletion of agonist-sensitive
Ca2+ pools does not occur in the absence of
extracellular Ca2+.28 ANOVA at 15 and
30 minutes showed no difference between the filling state of the
stores after exposure to HX-XO or thapsigargin (mean
[Ca2+]i transient, 301±82 and
339±118 nmol/L, respectively). In contrast, HX-XO was significantly
more effective in reducing sequestered Ca2+ than
thrombin (mean [Ca2+]i transient,
886±252 nmol/L). There was still a significant difference between
HX-XO and thrombin when refilling of the stores in thrombin-exposed
cells was inhibited with EGTA (mean
[Ca2+]i transient, 475±112
nmol/L).
|
To determine whether changes in extracellular Ca2+
could influence the depletion of Ca2+ stores by
HX-XO, cells were exposed for 15 minutes to HX-XO at different levels
of [Ca2+]o. After the exposure, the
filling state of the Ca2+ stores was determined at
150 nmol/L [Ca2+]o as described above.
As shown in Fig 4, we found a significant correlation
between [Ca2+]o and the HX-XO–related
depletion of sequestered Ca2+, with smaller
ionomycin-induced [Ca2+]i transients,
indicating a more complete depletion at low
[Ca2+]o. Thus, a
[Ca2+]i transient of 176±52 and
125±24 nmol/L was measured when the Ca2+ influx
from the extracellular space was inhibited by 20 µmol/L
La3+ or 1 mmol/L EGTA, respectively, compared with a
326±62-nmol/L transient in the control condition in 0.4 mmol/L
[Ca2+]o. Inversely, there was higher
Ca2+ content in intracellular stores when exposure
to HX-XO was performed in the presence of 2.5 mmol/L
[Ca2+]o
([Ca2+]i transient, 441±115
nmol/L).
|
Role of Ca2+ Stores in HX-XO–Related Inhibition
of Protein Synthesis
To investigate the role of Ca2+ stores in
protein synthesis in endothelial cells, the effects of HX-XO,
thapsigargin, and thrombin on the incorporation of phenylalanine were
compared. Cells were exposed for 30 minutes to HX-XO, 100 nmol/L
thapsigargin, 0.5 U/mL thrombin, or HX-XO plus 100 nmol/L thapsigargin,
and 3H-Phe was added during the last 15 minutes of this
exposure period. As shown in Fig 5, exposure to HX-XO in
0.4 mmol/L [Ca2+]o resulted in a
59±8% decrease in 3H-Phe incorporation. The HX-XO effect
was not different from that of 100 nmol/L thapsigargin (reduction of
3H-Phe incorporation, 74±11%), whereas thrombin had no
significant effect on overall protein synthesis. When thapsigargin was
applied together with HX-XO, the inhibition of protein synthesis
amounted to 89±3%. As shown above, the depletion of
Ca2+ stores can be influenced by the availability of
extracellular Ca2+. Therefore, we used different
levels of [Ca2+]o to investigate the
role of Ca2+ stores in the inhibition of protein
synthesis by HX-XO and thapsigargin. Suppression of
Ca2+ influx by 20 µmol/L La3+ or 1
mmol/L EGTA during the 30-minute period enhanced the inhibition of
protein synthesis by HX-XO to 71±7% and 73±6%, respectively, while
increasing [Ca2+]o to 2.5 mmol/L
diminished the HX-XO–related inhibition to 54±8% compared with the
control condition in the respective medium. A similar dependence of the
inhibition of protein synthesis on
[Ca2+]o was found in cells exposed to
thapsigargin, with 74±11% and 77±9% inhibition in La3+
and EGTA compared with a 47±53% reduction in 2.5 mmol/L
[Ca2+]o. In contrast, the inhibition
of Ca2+ influx (and refilling of the stores) did not
lead to significant inhibition of protein synthesis in thrombin-exposed
cells. Therefore, rapid recovery of Ca2+ stores was
not an explanation for the lack of a thrombin effect on protein
synthesis. 3H-Phe incorporation was not significantly
different between the control cells in La3+, EGTA, and 0.4
mmol/L and 2.5 mmol/L [Ca2+]o.
|
In previous work, we have found that HX-XO inhibits protein
synthesis in endothelial cells at the translational
initiation.29 To examine the role of sequestered
Ca2+ at the initiation step, cells were exposed for
60 minutes to HX-XO or thapsigargin at different levels of
[Ca2+]o or with La3+.
Initiation efficiency was assessed indirectly from the ribosome size
distribution on sucrose gradients. Fig 6 shows an
optical density profile representative for three independent
experiments. HX-XO and thapsigargin markedly increased the
monosome-to-polysome ratio in 0.4 mmol/L
[Ca2+]o (mean value, 0.80 and 1.31,
respectively, versus 0.49 in the control condition), reflecting the
block at the initiation level. The effect of HX-XO and thapsigargin on
the monoribosome-to-polyribosome ratio was enhanced when the influx of
extracellular Ca2+ was blocked by La3+
(mean value, 1.26 and 2.77, respectively, versus 0.38 in the control
condition), whereas increasing [Ca2+]o
to 2.5 mmol/L reduced the accumulation of monoribosomes in both
treatments (mean value, 0.71 and 1.11, respectively, versus 0.51 in the
control condition).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the present study, we have compared the effects of ROSs generated by the HX-XO system with those of thapsigargin and thrombin on Ca2+ homeostasis and protein synthesis in HUVECs. To our knowledge, on-line recordings of the effect of HX-XO on intracellular Ca2+ in human endothelial cells have not been shown before. High doses of xanthine–xanthine oxidase (with the generation of 45 nmol O2-/mL per minute) were used by Geeraerts et al5 to induce oxidative stress in HUVECs. However, measurements in these experiments started 5 to 20 minutes after administration of the oxidative stress, so that an initial [Ca2+]i peak would have been missed. Maximum [Ca2+]i occurred after 48 minutes when most of the O2- production had ceased. That late progressive [Ca2+]i increase, which leads to cell death, presumably results from the accumulation of H2O27 and is clearly different from the early [Ca2+]i rise shown in the experiments presented here. However, the time course and extent of the [Ca2+]i changes observed in the present study were comparable to that found by Franceschi et al,4 who recorded the [Ca2+]i response to HX-XO (4 nmol O2-/mL per minute) in monolayers of bovine aortic endothelial cells.
Alterations of [Ca2+]i by oxidative stress may result from Ca2+ mobilization from both extracellular and intracellular sources. Therefore, we followed HX-XO–induced [Ca2+]i changes in the presence of the Ca2+ entry blocker La3+. These experiments showed that HX-XO still induces a transient [Ca2+]i increase when the Ca2+ influx from the extracellular medium is blocked but that the sustained elevation above baseline values is suppressed. The results are in accordance with the previous findings in bovine aortic endothelial cells, where HX-XO still induced a transient [Ca2+]i rise in the absence of extracellular Ca2+.4 These observations suggest that the oxidative stress can mobilize Ca2+ from intracellular sources. However, in the previous study, the effect of HX-XO on the Ca2+ stores was not investigated. In contrast, the present study demonstrates that HX-XO mobilizes Ca2+ from intracellular stores. Thus, HX-XO substantially reduced the total ionomycin-releasable Ca2+ pool, and both thapsigargin- and thrombin-related [Ca2+]i changes were largely precluded in cells pretreated with HX-XO. The latter effect was not simply due to an inhibition of Ca2+ influx by the oxidative stress,6 because the effect was also observed in the presence of La3+. These experiments show that the [Ca2+]i rise induced by HX-XO in endothelial cells was in part directly caused through depletion of the Ca2+ stores. However, comparison of the [Ca2+]i tracings with and without La3+ indicates that an influx of extracellular Ca2+ contributed to the [Ca2+]i changes. This HX-XO–related Ca2+ influx could be a secondary phenomenon related to the mobilization of sequestered Ca2+, according to the capacitative Ca2+ model described by Putney.30 This notion was supported by the fact that a depletion of the stores with thapsigargin (1 µmol/L) completely prevented the effect of a subsequent treatment with HX-XO in the presence of extracellular Ca2+. The model of capacitative Ca2+ entry following oxidative exposure was recently confirmed in canine venous endothelial cells.10
Next, we compared the action of HX-XO on intracellular Ca2+ stores with that of a receptor-mediated Ca2+ mobilization. Thrombin was chosen as agonist, because in the intact endothelium, it stimulates Ca2+ transients that exceed those induced by bradykinin and histamine.21 This is in contrast to most findings in endothelial cell suspensions, where bradykinin and histamine appear as the most effective Ca2+ agonists.6 13 Despite rapid desensitization of the thrombin receptor after activation, the Ca2+ effect of thrombin in endothelial cells is sustained over a prolonged period,19 and the subsequent refilling of Ca2+ stores may be further delayed through a thrombin-specific inhibition of Ca2+ influx.31 However, the recovery of the stores was completely suppressed over a 60-minute period when Ca2+ entry from the extracellular medium was blocked. After pretreatment of the cell monolayers with thrombin, a reduced [Ca2+]i rise through HX-XO was still observed, even when Ca2+ influx was blocked. Also, Ca2+ content of intracellular stores was significantly smaller in cells treated with HX-XO than after treatment with thrombin. These findings indicate that like thapsigargin, HX-XO affected both the hormone-sensitive stores and hormone-insensitive stores. From the experiments presented here, the nature of the hormone-insensitive store that is depleted by both thapsigargin and HX-XO is not determined. Receptor agonists like thrombin mobilize stored Ca2+ via inositol trisphosphate (IP3)–sensitive channels. The other major Ca2+ channel responsible for the release of sequestered Ca2+ is the ryanodine receptor.14 The presence of a Ca2+ pool sensitive to ryanodine (5 µmol/L) has been recently demonstrated in HUVECs. However, this pool appears to overlap with the IP3-sensitive stores.32 Neither ryanodine (5 µmol/L) nor 8-bromo-cGMP (1 mmol/L), which stimulates Ca2+ release through the ryanodine receptor via the generation of cyclic ADP-ribose,33 inhibited protein synthesis in our cell system (data not shown). Therefore, the IP3- or ryanodine-sensitive Ca2+ pool(s) appear(s) not to be involved in the regulation of protein synthesis in HUVECs. There are several possible explanations for this observation based on the hypothetical models by Rossier and Putney34 : (1) The IP3- and ryanodine-sensitive stores are organelles separated from the ER (which regulates protein synthesis). (2) IP3- and ryanodine-sensitive Ca2+ channels are localized on highly specialized regions of the ER. (3) The receptor agonists act, although less effectively, on the same Ca2+ pool as thapsigargin and HX-XO.
HX-XO stimulated no more Ca2+ mobilization after pretreatment with 1 µmol/L thapsigargin, suggesting that the hormone-insensitive store affected by HX-XO could be depleted by the inhibition of the ER Ca2+-ATPase. In experiments with 45Ca2+, HX-XO was as effective as thapsigargin in mobilizing intracellular Ca2+. Experiments with isolated microsomes have shown in other cell types that HX-XO can empty Ca2+ stores by direct inhibition of the Ca2+-ATPase.11 12 Thus, one may speculate that the effect of HX-XO on Ca2+ homeostasis in HUVECs results from inactivation of the ER Ca2+-ATPase, probably via oxidation of sulfhydryl (SH) groups critical to transport function. Indeed, the SH-oxidizing agent thimerosal increases [Ca2+]i in HUVECs by mobilization of intracellular Ca2+ pools and subsequent Ca2+ influx from the extracellular medium.31 Alternatively, ATP depletion through HX-XO 29 or free radical–mediated nonspecific damage may be responsible for reduced Ca2+-ATPase activity. The other principle mechanism that could explain the mobilization of sequestered Ca2+ by the oxidative stress is an increase in Ca2+ permeability of the ER membrane by opening of ion channels or by disruption of the ER membrane.
To assess the role of intracellular Ca2+ in the inhibition of protein synthesis by HX-XO, we examined both overall protein synthesis and the translational initiation. Overall protein synthesis, measured by the amino acid incorporation rate, was markedly reduced by HX-XO or thapsigargin but was not substantially affected by thrombin. In the presence of extracellular Ca2+, depletion of intracellular stores leads to Ca2+ influx through the plasma membrane, thus increasing [Ca2+]i and facilitating refilling of the Ca2+ stores. Therefore, blocking the Ca2+ influx with EGTA or La3+ not only prevents a sustained [Ca2+]i increase but also impairs the reconstitution of sequestered Ca2+. Our data show a significant correlation between the availability of extracellular Ca2+ and the inhibition of protein synthesis by thapsigargin, confirming previous results with thapsigargin and EGTA in HeLa cells.35 As with thapsigargin, the HX-XO–related inhibition of protein synthesis was significantly influenced by [Ca2+]o. The oxidative stress caused more inhibition when the normal refilling of Ca2+ stores was prevented through EGTA or La3+, whereas higher [Ca2+]o correlated with more efficient protein synthesis. These results corresponded to the effect of [Ca2+]o on the HX-XO–related depletion of sequestered Ca2+, where enhanced depletion of Ca2+ stores was documented for lower [Ca2+]o during the exposure. To understand the effect of [Ca2+]o, as well as the additive action of thapsigargin plus HX-XO, it is important to note that neither HX-XO nor thapsigargin was alone sufficient to completely deplete intracellular Ca2+ stores in these experiments. The low concentration of thapsigargin (100 nmol/L instead of the more effective dose, 1 µmol/L) was chosen here to achieve an inhibition of protein synthesis comparable to that of HX-XO. The additive effect of HX-XO plus thapsigargin was largely reduced in experiments where treatments over 1 hour (instead of 30 minutes) led to more complete inhibition of the Ca2+-ATPase by thapsigargin (data not shown). Our findings suggest that depletion of sequestered Ca2+, but not the [Ca2+]i increase (which is inversely influenced by [Ca2+]o), plays a significant role in the inhibition of protein synthesis by thapsigargin and HX-XO.
To further investigate the regulation of protein synthesis by thapsigargin and HX-XO, we assessed the effect on translational initiation by examining the polyribosome distribution on sucrose gradients. This method has successfully been used to detect the inhibition of initiation by an ionophore18 and by oxidative stress.29 After exposure to thapsigargin or HX-XO, the accumulation of monoribosomes indicated a block at the initiation step. Corresponding to the results from overall protein synthesis, in both treatments the initiation block was enhanced in conditions that favored the emptying of intracellular Ca2+ stores. Therefore, in both treatments, the depletion of intracellular Ca2+ stores affects translational initiation and may thus lead to the inhibition of overall protein synthesis. These findings are in accordance with results in GH3 pituitary cells, where depletion of Ca2+ stores blocks the translational process at the initiation level.18 The extent of mobilization of intracellularly sequestered Ca2+ by HX-XO was almost identical to that of thapsigargin (100 nmol/L). Thus, the depletion of intracellular stores induced by HX-XO would alone be expected to cause inhibition of protein synthesis. However, it should be pointed out that other (Ca2+-independent) mechanisms may contribute to the effect of the oxidative stress on protein synthesis. In contrast to HX-XO and thapsigargin, the mobilization of sequestered Ca2+ by thrombin did not substantially affect protein synthesis, in accordance with previous findings that the hormone-sensitive stores, ie, the fraction of Ca2+ stores that is mobilized by receptor agonists, are less critical to protein synthesis.35 Therefore, depletion of the hormone-insensitive rather than the hormone-sensitive Ca2+ stores appears to be involved in the regulation of protein synthesis in HUVECs by HX-XO.
|
Acknowledgments |
|---|
This study was supported by a grant (No. 32.27601.89) from the Swiss National Research Foundation, by the Carlos and Elsie De Reuter Foundation for Medical Research, and by the Lancardis Foundation. We thank Dr K.H. Krause and Dr W. Schlegel for discussions and reading the manuscript.
Received March 24, 1994; accepted November 16, 1994.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Gerritsen ME, Bloor CM. Endothelial cell gene expression in response to injury. FASEB J. 1993;7:523-532. [Abstract]
2. Ward PA. Mechanisms of endothelial cell injury. J Lab Clin Med. 1991;118:421-426. [Medline] [Order article via Infotrieve]
3. Hirosumi J, Ouchi Y, Watanabe M, Kusunoki J, Nakamura T, Orimo H. Effect of superoxide and lipid peroxide on cytosolic free calcium concentration in cultured pig aortic endothelial cells. Biochem Biophys Res Commun. 1988;152:301-307. [Medline] [Order article via Infotrieve]
4. Franceschi D, Graham D, Sarasua M, Zollinger RM. Mechanisms of oxygen free radical-induced calcium overload in endothelial cells. Surgery. 1990;108:292-297. [Medline] [Order article via Infotrieve]
5.
Geeraerts MD, Ronveaux-Dupal MF, Lemasters JJ, Herman B.
Cytosolic free Ca2+ and proteolysis in lethal
oxidative injury in endothelial cells. Am J Physiol. 1991;261:C889-C896.
6. Elliot SJ, Doan TN. Oxidant stress inhibits the store-dependent Ca2+-influx pathway in vascular endothelial cells. Biochem J. 1992;292:385-393.
7. Dreher D, Junod AF. Differential effects of superoxide, hydrogen peroxide, and hydroxyl radical on intracellular calcium in human endothelial cells. J Cell Physiol. 1995;162:142-153.
8. Roveri A, Coassin M, Maiorino M, Zamburlini A, van Amsterdam FTM, Ratti E, Ursini F. Effect of hydrogen peroxide on calcium homeostasis in smooth muscle cells. Arch Biochem Biophys. 1992;297:265-270. [Medline] [Order article via Infotrieve]
9. Ueda N, Shah SV. Role of intracellular calcium in hydrogen peroxide-induced renal tubular cell injury. Am J Physiol. 1992; 263:F214-F221.
10. Doan TN, Gentry DL, Taylor AA, Elliott SJ. Hydrogen peroxide activates agonist-sensitive Ca2+-flux pathways in canine venous endothelial cells. Biochem J. 1994;297:209-215.
11. Grover AK, Samson SE. Protection of Ca pump of coronary artery against inactivation by superoxide radical. Am J Physiol. 1989; 256:C666-C673.
12.
Suzuki YJ, Ford GD. Superoxide stimulates
IP3-induced Ca2+ release from vascular
smooth muscle sarcoplasmic reticulum. Am J Physiol. 1991;261:H568-H574.
13. Gericke M, Droogmans G, Nilius B. Thapsigargin discharges intracellular calcium stores and induces transmembrane currents in human endothelial cells. Pflugers Arch. 1993;422:552-557. [Medline] [Order article via Infotrieve]
14. Berridge MJ. Inositol triphosphate and calcium signalling. Nature. 1993;361:315-325. [Medline] [Order article via Infotrieve]
15. Brostrom CO, Brostrom MA. Calcium-dependent regulation of protein synthesis in intact mammalian cells. Annu Rev Physiol. 1990;52:577-590. [Medline] [Order article via Infotrieve]
16. Hoek JB. Intracellular signal transduction and the control of endothelial permeability. Lab Invest. 1992;67:1-4. [Medline] [Order article via Infotrieve]
17. Schilling WP, Elliot SJ. Ca2+ signaling mechanisms of vascular endothelial cells and their role in oxidant-induced endothelial cell dysfunction. Am J Physiol. 1992;262:H1617-H1630. [Abstract]
18.
Chin KV, Cade C, Brostrom CO, Galuska EM, Brostrom MA.
Calcium-dependent regulation of protein synthesis at translational
initiation in eukaryotic cells. J Biol Chem. 1987;262:16509-16514.
19.
Goligorsky MS, Menton DN, Lazlo A, Lum H. Nature of
thrombin-induced sustained increase in cytosolic calcium concentration
in cultured endothelial cells. J Biol Chem. 1989;264:16771-16775.
20. Jornot L, Junod AF. Response of human endothelial cell antioxidant enzymes to hyperoxia. Am J Respir Cell Mol Biol. 1992;6:107-115.
21. Wickham NWR, Vercellotti GM, Moldow CF, Visser MR, Jacob HS. Measurement of intracellular calcium concentration in intact monolayers of human endothelial cells. J Lab Clin Med. 1988; 112:157-167.
22. Kao JPY. Practical aspects of measuring [Ca2+] with fluorescent indicators. Methods Cell Biol. 1994;40:155-181. [Medline] [Order article via Infotrieve]
23. Trombe M, Blaise M, Caro P. Sur la solubilité des carbonates d'yttrium, de scandium et de quelques éléments du groupe des terres rares dans l'eau chargée de gaz carbonique. C R Acad Sci III. 1966;263:521-524.
24. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951; 193:265-275.
25.
Jornot L, Mirault ME, Junod AF. Protein synthesis in hyperoxic
endothelial cells: evidence for translational defect. J Appl
Physiol. 1987;63:457-464.
26. Neter J, Wasserman W. Applied Linear Statistical Models: Regression, Analysis of Variance, and Experimental Designs. Homewood, Ill: Richard D Irwin Inc; 1974.
27. Lewis MS, Whatley RE, Cain P, McIntyre TM, Prescott SM, Zimmerman GA. Hydrogen peroxide stimulates the synthesis of platelet-activating factor by endothelium and induces endothelial cell-dependent neutrophil adhesion. J Clin Invest. 1988;82: 2045-2055.
28. McDonough PM, Eubanks JH, Brown JH. Desensitization and recovery of muscarinic and histaminergic Ca2+ mobilization in 1321N1 astrocytoma cells. Biochem J. 1988;249:135-141. [Medline] [Order article via Infotrieve]
29. Jornot L, Junod AF. Hypoxanthine-xanthine oxidase-related defect in polypeptide chain initiation by endothelium. J Appl Physiol. 1989; 66:450-457.
30. Putney JW. Capacitative calcium entry revisited. Cell Calcium. 1990;11:611-624. [Medline] [Order article via Infotrieve]
31.
Neylon CB, Irvine RF. Thrombin attenuates the stimulatory
effect of histamine on Ca2+ entry in confluent human
umbilical vein endothelial cell cultures. J Biol Chem. 1991;266:4251-4256.
32.
Ziegelstein RC, Spurgeon HA, Pili R, Passaniti A, Cheng L,
Corda S, Lakatta EG, Capogrossi MC. A functional ryanodine-sensitive
intracellular Ca2+ store is present in vascular
endothelial cells. Circ Res. 1994;74:151-156.
33. Galione A, White A, Willmott N, Turner M, Potter BV, Watson SP. cGMP mobilizes intracellular Ca2+ in sea urchin eggs by stimulating cyclic ADP-ribose synthesis. Nature. 1993;365:456-459. [Medline] [Order article via Infotrieve]
34. Rossier M, Putney JW. The identity of the calcium-storing, inositol 1,4,5-triphosphate-sensitive organelle in non-muscle cells: calciosome, endoplasmic reticulum . . . or both? Trends Neurosci. 1991;14:310-314. [Medline] [Order article via Infotrieve]
35. Preston SF, Berlin RD. An intracellular calcium store regulates protein synthesis in HeLa cells, but is not the hormone-sensitive store. Cell Calcium. 1992;13:303-312. [Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  S. Guthikonda, C. Sinkey, T. Barenz, and W. G. Haynes Xanthine Oxidase Inhibition Reverses Endothelial Dysfunction in Heavy Smokers Circulation, January 28, 2003; 107(3): 416 - 421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Diebolt, B. Bucher, and R. Andriantsitohaina Wine Polyphenols Decrease Blood Pressure, Improve NO Vasodilatation, and Induce Gene Expression Hypertension, August 1, 2001; 38(2): 159 - 165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. E. Matheny, T. L. Deem, and J. M. Cook-Mills Lymphocyte Migration Through Monolayers of Endothelial Cell Lines Involves VCAM-1 Signaling Via Endothelial Cell NADPH Oxidase J. Immunol., June 15, 2000; 164(12): 6550 - 6559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Kunsch and R. M. Medford Oxidative Stress as a Regulator of Gene Expression in the Vasculature Circ. Res., October 15, 1999; 85(8): 753 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L Jornot, P Maechler, C. Wollheim, and A. Junod Reactive oxygen metabolites increase mitochondrial calcium in endothelial cells: implication of the Ca2+/Na+ exchanger J. Cell Sci., January 4, 1999; 112(7): 1013 - 1022. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||