| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Anesthesiology and the
Institute for Surgical Research, Klinikum Grosshadern, Ludwig-Maximilians-Universität, Munich, Germany and
Sidney Kimmel Cancer Center, San Diego, California
Correspondence: Manfred Thiel, M.D., Department of Anesthesiology, Klinikum Grosshadern, Ludwig-Maximilians-University, 81377, Munich, Germany. E-mail: manfred.thiel{at}ana.med.uni-muenchen.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: sodium chloride • osmolarity • neutrophils • CD18 • CD62L
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A more recent advance in the therapy of hemorrhagic and traumatic shock is represented by the concept of small-volume resuscitation. Intravenous application of hypertonic saline was shown to rapidly restore macrohemodynamic functions and tissue perfusion. Although there is accumulating evidence for the effectiveness of small-volume resuscitation in preventing organ dysfunction after posttraumatic and hemorrhagic shock under clinical conditions [6 ], the mode of action of hypertonic saline at the cellular level is still not completely understood. Besides mobilizing water from swollen endothelial cells, thereby restoring capillary diameters and flow, intravital microscopy has demonstrated that hypertonic saline can also inhibit the adhesion of PMNLs particularly to the endothelia of postcapillary venules [7 ]. Because adherence of PMNLs to postcapillary venules was shown to correlate with increases of flow resistance and vascular leakage [7 ], prevention of PMNL adhesion by hypertonic saline might significantly ameliorate tissue injury after ischemia reperfusion. Moreover, small-volume resuscitation in hemorrhagic shock was shown to prevent neutrophil margination [8 ] and intra-alveolar sequestration [9 ].
Although modulation of neutrophil adhesion molecules by hypertonic saline has already been addressed in two other studies, we set out to study the effects of hypertonic saline on the expression of ß2-integrins and L-selectin on human PMNLs in more detail. In particular, we questioned whether possible effects on adhesion molecule expression are related specifically to changes in the sodium cation itself or nonspecifically to the mere increase in the extracellular osmolarity. In this study, we present data suggesting that an increase in the concentration of the sodium cation rather than hypertonicity itself is responsible for the N-formyl-methionyl-leucyl-phenylalanine (fMLP)-stimulated inhibition of the numerical and functional up-regulation of ß2-integrins.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
80
mosmol/L. Specific stimuli, chemotactic tripeptide fMLP,
phorbol myristate acetate (PMA), and ionophore A12387 were from Sigma
Chemicals (Deisenhofen, Germany). Lipopolysaccharide (LPS)
(Escherichia coli O55:B) was from Gibco BRL Life
Technologies (Cambridge, UK). All basic chemicals and materials needed
for sodium dodecyl sulfate (SDS)-gel electrophoresis were from Bio-Rad
Laboratories (Munich, Germany). Cell lysis buffer, polyclonal rabbit
anti-phospho-p38 mitogen-activated protein (MAP) kinase, anti-p38 MAP
kinase (MAPK), anti-phospho-p44/42, anti-p44/42, and a luminol-enhanced
chemiluminescence-based Phototope-HRP Western blot detection system
were purchased from New England Biolabs (Frankfurt, Germany). Secondary
polyclonal horseradish peroxidase-labeled sheep anti-rabbit antibodies
were obtained from Dianova (Hamburg, Germany).
Determination of adhesion molecules
Heparinized venous blood (final activity, 10 U/mL) (Vetren Na,
Promonta GmbH, Hamburg, Germany) was drawn from healthy volunteers.
Anticoagulated blood samples were washed two times with HBSS of desired
composition at room temperature to remove plasma proteins. Washed blood
cells were reconstituted to the initial hematocrit with the same HBSS
used for removal of plasma constituents. Aliquots of washed blood cells
were incubated in a water bath at 37°C for 30 min. Thereafter, PMNLs
were stimulated with different concentrations of fMLP for 15 min,
inducing either half-maximal (10-8 M) or maximal
(10-7 M) changes in the expression of adhesion molecules
[10
]. In addition, experiments were carried out in which
receptor-independent activation of cells was induced by addition of PMA
and ionophore A23187 to achieve protein kinase C (PKC)- and
calcium-dependent stimulation of PMNLs, respectively.
To validate our experimental setup and methods, we tested the effects
of hypertonic sodium chloride-modified HBSS at 500 mosmol/L on the
LPS-stimulated numerical up-regulation of ß2-integrins and shedding
of L-selectin. To this end, washed blood cells were
incubated with HBSS of desired osmolarities containing 10%
heat-inactivated fetal calf serum. After incubation at 37°C for 30
min, LPS was added at a final concentration of 100 ng/mL. Activation of
cells was stopped after 20 min by putting cells on ice. Expression of
adhesion molecules of PMNLs was determined prior to and at the end of
the incubation period by flow cytometry. For this purpose, blood cells
were incubated for 20 min on ice with fluorescein isothiocyanate
(FITC)-labeled monoclonal antibody IB4 [immunoglobulin (Ig) G2a] or
Dreg 200 (IgG1), which specifically bind either to the ß2
chain of human integrins (CD18) or L-selectin (CD62L). The
monoclonal antibody CBRM1/5 was used to detect an extracellularly
located neoepitope on the human 
2 chain CD11b, which is
expressed only when activation-dependent conformational changes of this
ß2-integrin occur [11
]. Consequently, the epitope
CBRM1/5 can be considered to represent an activation reporter epitope.
The monoclonal antibody CBRM1/5 was a kind gift from T. A.
Springer (Harvard Medical School, Boston, MA). Because the antibody was
not labeled by a fluorochrome, PMNL-bound CBRM1/5 was detected by a
secondary polyclonal anti-mouse FITC-labeled antibody (La Roche,
Mannheim, Germany). Leukocyte populations and specificity of binding of
antibodies were assessed as previously described [10
].
CBRM1/5 was used at twice the saturating concentration, and specificity
of binding was demonstrated by an isotype-matched control mouse IgG1
antibody (DAKO, Glostrup, Denmark). There was no binding of the
secondary anti-mouse FITC-labeled antibody to PMNLs either in the
absence or in the presence of the control mouse IgG1 antibody.
Additional control experiments revealed that binding of antibodies was
not affected by the sodium- or choline-dependent osmolarity.
Fluorescence was analyzed using a Becton Dickinson (Heidelberg,
Germany) FACScan equipped with an argon laser emitting 15 mW of light
at 488 nm. Linear list mode data were collected for 5,000 events from
each sample. The linear fluorescence amplifier gain at constant
photomultiplier voltage was adjusted as necessary to position the
signal on scale for each sample. With FACScan software, mean
fluorescence channel numbers were calculated and normalized to a gain
of 1.0. The corrected mean channel number (CMCN) was computed according
to the following formula: CMCN = observed mean channel
number/amplifier gain.
Estimation of cell volume
The volume of cells was estimated by forward light scatter
values (FSCs) [12
] calculated as mean channel numbers at
constant photomultiplier voltage normalized to a gain of 1.0. To
validate estimations of cell volume by FSCs under our experimental
conditions, cell volumes of isolated neutrophils were manipulated by
incubation in hypertonic or hypotonic sodium choline-modified HBSS, and
cell volume and FSCs were determined simultaneously by use of a
Multisizer II Coulter Counter (Coulter Electronics, Krefeld, Germany)
and by flow cytometer measurements. The correlation between the
absolute cell volume and the FSC was as follows: cell volume
(fl) = 222.2 + 0.246 x FSC (CMCN); N = 15; r = 0.955; and P = 0.000
(correlation coefficient according to the method of Pearson).
Measurement of fMLP receptor expression
The number of cell surface fMLP receptors expressed by
peripheral blood neutrophils was estimated by the binding of a
fluorescein derivative of formyl-Nle-Leu-Phe-Nle-Tyr-Lys (fNLPNTL-FITC)
(Molecular Probes Inc., Eugene, OR). Heparinized whole blood was washed
two times with sodium chloride- or choline chloride-modified basic HBSS
of desired osmolarity. After reconstitution to the initial hematocrit,
the cells were incubated for 30 min at 37°C. At 5 min before the end
of the incubation period, LDS 751 (Molecular Probes) was added at a
final concentration of 10-5 M. Each cell suspension was
aliquoted, diluted with ice-cooled modified basic HBSS of appropriate
osmolarity, and incubated for an additional period of 30 min on ice.
Thereafter, fNLPNTL-FITC was added at concentrations ranging from 0 and
10-10 M to 10-7 M in the absence or presence
of unlabeled fMLP in excess (10-4 M) to block nonspecific
binding. After 30 min, neutrophil-bound fNLPNTL-FITC was assessed in
blood cell suspensions by linear fluorescein fluorescence intensities
on FL-1 after gating on neutrophils in FSC/SSC dot blots obtained by
setting the threshold on FL-3 for LDS-stained nuclear cells. Specific
binding of fNLPNTL-FITC was calculated by the difference between FL-1
fluorescence intensities measured in the absence and presence of
blocking concentrations of unlabeled fMLP.
Isolation of PMNLs and determination of intracellular calcium
PMNLs were isolated from heparinized whole blood by continuous
Percoll density gradient centrifugation as previously described
[1
]. Granulocytes were washed two times with isotonic
HBSS, cell numbers were counted, and the concentrations were adjusted
to 106 cells/mL. Because cell separation procedures were
previously shown to activate PMNLs, we determined expression of
ß2-integrins and L-selectin on PMNLs prior to and after
cell isolation. During cell isolation, only a slight increase in the
numerical expression of ß2-integrins and no changes of
L-selectin levels occurred when compared with baseline
values determined before cell purification (ß2-integrins, 136.2 ± 28.4 vs. 105.4 ± 21.7; L-selectin, 54.9 ±
3.95 vs. 56.8 ± 8.8; expression after cell isolation compared
with baseline, n = 6).
To determine relative changes in intracellular concentrations of calcium ions, the fluorescence intensities of the calcium-sensitive dye Fluo-3 (Molecular Probes, Inc., Eugene, OR) and of SNARF-1 (Molecular Probes) obtained at its isoemissive point were measured, and the ratio of the two fluorescence intensities was calculated. This method has been shown to overcome variation in basal fluorescence intensity because of heterogeneity in intracellular dye concentration or compartmentalization [13 ]. Because granulocytes change their cell volume when incubated in media with different osmolarities, any effects of cell volume on Fluo-3 signals were eliminated by calculating the ratio to the fluorescence intensity of the concomitantly loaded fluorescence dye SNARF-1 recorded at its isoemissive point. At that point, the intensity of the fluorescence of SNARF-1 is independent of activation-dependent changes in intracellular pH. For loading cells with Fluo-3 and SNARF-1, acetoxy-methyl ester derivatives of both compounds were added to 106 cells/mL at final concentrations of 5 x 10-6 M and 1 x 10-6 M, respectively. After incubation of cells for 20 min at 37°C, the dye-loaded cells were washed two times with either sodium chloride- or choline chloride-modified HBSS of desired osmolarity. Cells were incubated for an additional 10 min at 37°C and activated by the addition of fMLP at final concentration of 10-7 or 10-8 M. Kinetics of fluorescence signals were recorded intermittently for 5-s periods of measurement separated by 10-s periods of cytometer downtime for Fluo-3 on FL-1 and SNARF-1 on FL-3 using CHRONYS software (Becton Dickinson). Cell suspensions were kept at 37°C during measurements. From linear list mode data collected for 5,000 events from each sample, the CMCN was computed for FL-1 and FL-3 as described above. Although cells were double stained, there was no need to set compensation on the instrument.
Detection of phospho-p38 and phospho-p44/42 in polymorphonuclear
leukocytes
Isolated PMNLs were incubated with sodium chloride- or choline
chloride-modified basic HBSS of the desired osmolarities and stimulated
with fMLP (10-7 M) as described above for determination of
adhesion molecules. Activation of cells was stopped 2 min after the
addition of the chemotactic tripeptide by adding 10 volumes of ice-cold
incubation medium and then subjecting the cells to rapid
centrifugation. In control experiments with LPS, performed to
demonstrate the reliability of our Western blotting system, isolated
PMNLs were incubated in the same way as the other stimuli to assess the
effects of hypertonic sodium-modifed HBSS on the expression of adhesion
molecules.
For immunoblotting, cell pellets were lysed with lysing buffer and
sonicated in a water bath for 2 min. Lysed samples were kept at
-80°C until use. After thawing, protein concentrations in samples
were determined by photometry (Protein Assay ESL, Boehringer Mannheim,
Mannheim, Germany), and the specimens were boiled in Laemmli sample
buffer for 10 min before analysis by 10% sodium dodecyl
sulfate-polyacrylamide gel electrophoresis. The amounts of protein
loaded per slot were 60 µg for determination of phospho-p44/42 and
200 µg for determination of phospho-p38. Protein loading was kept
constant for lanes of an individual gel. After transfer to
nitrocellulose membranes and blockage by nonfat dry milk at room
temperature for at least 1 h, membranes were incubated with the
appropriate antibodies at room temperature overnight. The dilution was
1:500 for anti-phospho-p38, 1:1,000 for anti-p38, 1:4,000 for
anti-phospho-p44/42 [extracellular regulated kinase (ERK) 1/2], and
anti-p44/42 (ERK1/2). After washing, primary antibodies were detected
by binding of secondary horseradish peroxidase-conjugated antibodies
and luminol-enhanced chemiluminescence illuminating X-ray films. After
development, the films were subjected to densitometry. Optical
densities (ODs) were corrected for background density, and the results
are expressed as OD units (ODus) per square millimeter. In addition to
direct measurements, equal loading of lanes with proteins was also
determined intermittently by stripping membranes and reprobing for
phosphorylation-independent total anti-p38 or anti-p44/42.
Statistics
All data were tested for normal distribution by the one-sample
Kolmogorov-Smirnov test. Effects of incubation conditions (resting vs.
stimulated and changes in osmolarities) were analyzed by
repeated-measures analysis of variance, comparing the mean of each
level to the mean of the level of interest with levels of significance,
corrected according to the method of Bonferroni. For comparison of
phosphorylation of MAPKs, a paired t-test and correction of
levels of significance according to the method of Bonferroni were
performed. Data are presented as means ± SD. All
calculations were done using SPSS 10.0 software (SPSS Inc, Chicago,
IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
When hypertonic choline chloride solution was tested in relation to the shedding of L-selectin stimulated by the lower concentration of fMLP (10-8 M), hypertonic choline chloride, like hypertonic saline, inhibited the loss of this type of adhesion molecule from the surface of PMNLs (Fig. 2C) but again could not inhibit shedding because of the high fMLP concentration (10-7 M).
Similar to the results obtained for choline chloride, increasing osmolarities of the nonionic compound mannitol did not inhibit the fMLP-stimulated numerical up-regulation of ß2-integrins (Fig. 3A ). In contrast to hypertonic sodium and choline chloride solutions, hypertonic mannitol-modified medium did not attenuate the fMLP-induced shedding of L-selectin (Fig. 3B) . Similiar results (Table 1 ) were obtained when osmolarity was increased by stimulation with sucrose instead of mannitol—neither of which enters cells. When we tested urea, to which the cell membrane is permeable, significant hemolysis occurred even when it was added to basic HBSS in amounts to achieve an osmolarity of 290 mosmol/L. For this reason, the effects of urea-dependent osmolarities on expression of adhesion molecules by neutrophils were not assessed.
|
|
Effects of sodium chloride and choline chloride on cell surface
binding of fNLPNTL-FITC
Because the contrasting effects of hypertonic saline and choline
chloride on the up-regulation of ß2-integrins might be explained by
differences in the ability of the solutions to interfere with the
interaction of fMLP and their cell surface receptors, the effects of
sodium chloride-modified and choline chloride-modified media were
compared in relation to the specific binding of fNLPNTL-FITC to PMNLs.
In these experiments, which were conducted under the same experimental
conditions used for the study of the effects of osmolarity on the
fMLP-induced changes in adhesion molecule expression, no differences
were observed with respect to the type of major cations used (sodium
vs. choline) or the osmolarity selected (230 vs. 410 mosmol/L). As
expected for the interaction of the fluorescent ligand with a receptor
site, dose-response curves were of sigmoidal shape, reaching
half-maximal and maximal saturating concentrations of binding
at 2–3 x 10-9 M and at 1 x
10-8 M, respectively (Fig. 4
).
|
|
|
|
|
|
The effects of sodium chloride- or choline chloride-dependent osmolarity on the phosphorylation state of MAPKs as assessed by densitometry (Fig. 8 and 9) are also shown by individual blots (Fig. 10 and Fig. 11 ).
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
From these observations, the following conclusions can be drawn: (1) hypertonic saline exerts its inhibitory action on the fMLP-stimulated up-regulation of ß2-integrins by a sodium-specific effect rather than by an effect induced by hyperosmolarity itself; (2) hypertonic saline appears to prevent activation-dependent shedding of L-selectin by an increase in osmolarity rather than by a sodium cation-specific effect; and (3) there might be a specific role of chloride anions in the inhibition of the shedding of L-selectin, given the finding that hypertonic mannitol- and sucrose-modified HBSS did not attenuate the fMLP-stimulated shedding of L-selectin.
Although hypertonic saline has been shown to induce shedding of L-selectin in in vitro [16 ] and in vivo models of hemorrhagic shock [8 , 9 ] and hence has been suggested to account for prevention of injury to microvascular endothelial cells by impediment of selectin-dependent leukocyte adhesion [2 ], we did not further pursue the possible specific role of chloride anions in the regulation of shedding of L-selectin but focused on the difference observed between hypertonic-saline-modified and hypertonic-choline chloride-modified HBSS in inhibiting the expression of ß2 integrins. To this end, we compared the effects of the two hypertonic solutions on signaling pathways known to mediate activation of human polymorphonuclear leukocytes through the chemotactic tripeptide fMLP.
Effects of hypertonic saline and hypertonic choline chloride on
binding of fMLP to its receptor and receptor-dependent classical
signaling
To discern the mechanism by which sodium may specifically
affect the up-regulation of ß2-integrins, we compared the effects of
hypertonic saline-modified medium and hypertonic choline-modified
medium on different steps in the classical fMLP receptor-signaling
cascade. As one might have expected from the observation that
hypertonic saline and choline chloride solutions did not differ with
respect to the inhibition of the fMLP-stimulated shedding of
L-selectin, direct analysis of specifically bound
fNLPNTL-FITC confirmed no differences in binding of the
fluorescein-conjugated fMLP derivative regarding the type of the major
cations used (sodium vs. choline) or the osmolarity selected (230 vs.
410 mosM/L) (Fig. 4)
.
There were also no differences between the effects of hypertonic-saline-modified and choline chloride-modified HBSS on fMLP-induced calcium changes (Fig. 5) and the numerical up-regulation of ß2-integrins stimulated by the calcium ionophore A23187 (Fig. 6) . These findings are in agreement with the suppression of the fMLP-stimulated increase in intracellular calcium caused by an increase in extracellular tonicity irrespective of whether hypertonic N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid buffered salt solution, sodium sulfate, or sucrose was used [17 ].
Regarding modulation of PKC-dependent pathways, hypertonic saline-modified and choline chloride-modified media inhibited the numerical up-regulation of ß2-integrins to the same extent (Fig. 7) . Similar findings have been reported on the effects of hypertonic saline on the fMLP- and PMA-stimulated numerical expression of CD11b/CD18 [9 ] and on the production of oxygen radicals elicited by the chemotactic peptide or the phorbol ester [18 , 19 ], but to our knowledge there are no comparable published data relating to the effects of hypertonic choline on adhesion molecule expression and oxygen radical production elicited by these compounds.
Considered together, these findings suggest that hypertonicity, irrespective of the cation (sodium or choline), can inhibit calcium- and PKC-dependent pathways in the fMLP receptor-induced signaling. Because fMLP-stimulated numerical up-regulation of ß2-integrins and expression of the activation reporter epitope CBRM1/5 were not inhibited by hypertonic choline chloride, the attenuation of the classical pathways of fMLP signaling (Ca+ and PKC) appears not to solely account for the down-regulation of ß2-integrin expression by hypertonic saline.
Effects of hypertonic saline and hypertonic choline chloride on the
fMLP-stimulated tyrosine-specific phosphorylation of
MAPKs p38 and p44/42
Recent in vitro studies suggest a strong role for the activation
of p38 not only in the shedding of L-selectin induced by
hypertonicity [16
] but also for the up-regulation of
ß2-integrins after stimulation of PMNLs by endotoxin
[14
] or tumor necrosis factor [20
].
Endotoxin was shown to strongly stimulate phosphorylation of p38 and to
induce up-regulation of ß2-integrins. Blockage of p38 activity by the
specific inhibitor SB203580 abolished the expression of ß2-integrins
stimulated by hypertonicity [16
], endotoxin, or tumor
necrosis factor [20
].
Hypertonic saline alone was shown in neutrophils to slightly stimulate phosphorylation of p38, which was followed by a strong inhibition of the endotoxin- and fMLP-induced phosphorylation-dependent activation of this enzyme [21 ]. It was suggested that hypertonic prestimulation of p38 rendered the MAPK refractory for subsequent activation [21 ]. We were not able to demonstrate an increase in the phosphorylation of p38 by hypertonic saline at 350 or 410 mosmol/L, a finding also reported by others [22 ].
Phosphorylation of p38 on stimulation with fMLP was not inhibited by 350 mosM/L of sodium chloride-modified HBSS but was attenuated at a concentration of 410 mosmol/L (Fig. 8A and 8B) . Because fMLP-stimulated tyrosine phosphorylation of p38 was shown in human neutrophils to partially depend on activation of PKC and cytosolic calcium transients [23 ], our observation of inhibition of the calcium- and PMA-induced up-regulation of ß2-integrins by hypertonic saline (Fig. 6A and 7A) is consistent with the inhibition of the activation of p38 by sodium chloride-modified HBSS at the higher tonicity. However, since hypertonic choline-modified HBSS also diminished the A23187- and PMA-induced up-regulation of ß2-integrins (Fig. 6B and 7B) , one could have expected that p38 phosphorylation would be inhibited by hypertonic choline chloride-modified medium as well. Accordingly, there was no difference between hypertonic sodium chloride- and choline chloride-modified HBSS with respect to their effects on basal and fMLP-stimulated phosphorylation of p38 at osmolarities of 350 and 410 mosmol/L.
Thus, the sodium-specific effect in the hypertonic inhibition of fMLP-stimulated up-regulation of ß2-integrins is unlikely to be explained at the level of activation of p38. To add complexity, hypertonic saline at 350 mosmol/L already significantly decreased fMLP-stimulated ß2-integrin expression but not p38 phosphorylation (Fig. 1A and Fig. 8A ), making p38 alone unlikely to account for fMLP-stimulated expression of ß2-integrins. It is interesting that inhibition of p38 by SB203580 failed to inhibit fMLP-stimulated, integrin-dependent adhesion of neutrophils [20 ].
In contrast, endotoxin-induced phosphorylation of p38 was inhibited by
hypertonic saline by 60 % (Fig. 12)
. In parallel, hypertonic saline
decreased the up-regulation of ß2-integrins and the shedding of
L-selectin in endotoxin-stimulated PMNLs. These
observations not only confirm results reported previously by others
[14
] but also demonstrate the sufficient sensitivity of
our methods to assess hypertonic saline-mediated changes in
stimulus-induced p38 phosphorylation.
During studies to determine a possible role of p44/42 in the sodium-specific effect of hypertonic inhibition of ß2-integrin up-regulation, results were obtained that were similar to those for p38 phosphorylation. Hypertonic sodium chloride and choline chloride did not visibly differ in their effects on basal and fMLP-stimulated p44/42 phosphorylation either at 350 mosmol/L or at 410 mosmol/L, although there was a statistically significantly stronger inhibition of p44/42 phosphorylation by hypertonic saline at the higher osmolarity (Fig. 9A and 9B) . There are several findings that make this slightly stronger inhibition by hypertonic saline on p44/42 phosphorylation an unlikely explanation for the sodium specific effect. First, in the case of tumor necrosis factor-induced up-regulation of ß2-integrins, selective inhibition of p44/42 MAPKs ERK1/2 by PD 98059 was without any effect [24 ]. Second, as with p38, hypertonic saline at 350 osmol/L decreased significantly the fMLP-stimulated up-regulation of ß2-integrins but not the phosphorylation of p44/42 (Fig. 1A and 9A) . Third, endotoxin clearly increased expression of ß2-integrins and caused shedding of L-selectin without any phosphorylation of p44/42 (data not shown). Concerning the role of MAPKs, the results of the present study suggest that phosphorylation of neither p38 nor p44/42 can actually account for the sodium-specific inhibitory effect on the fMLP-stimulated up-regulation of ß2-integrins.
Effects of hypertonicity-associated changes in cell volume on
fMLP-stimulated expression of ß2-integrins
Because there is evidence that a cell can detect changes in its
own volume and generate signals for regulation of mechanisms to keep
its volume constant, we also analyzed the effects of hypertonic saline
and choline chloride on FSCs of neutrophils. As shown in Figure 13
,
PMNLs incubated in hypertonic saline for 30 min at 37°C prior to
stimulation with fMLP had significantly higher FSCs than cells exposed
to hypertonic choline of the same osmolarity. This observation
indicates that hypertonic-saline-exposed PMNLs had increased their
cell volumes, a process that was inhibited in hypertonic-choline-bathed
cells because of the inability of the Na+/H+
exchangers to transport choline cations into the cells
[15
]. After stimulation of PMNLs with fMLP, again only
hypertonic-saline-bathed cells further increased their volumes.
Therefore, mechanisms associated with the regulation of cell volume
might generate signals that inhibit the fMLP-stimulated up-regulation
of ß2-integrins. However, when PMNLs were incubated with amiloride at
a concentration shown by others to abolish the gains in cell volumes
induced by fMLP [15
, 25
], the inhibition of
the chemotactically stimulated numerical up-regulation of
ß2-integrins and the expression of the activation-reporter epitope
CBRM1/5 were further enhanced (Fig. 14)
. In addition, prevention of the
increase in cell volumes of osmotically shrunken neutrophils (e.g., by
incubation of cells in hypertonic choline chloride, which is not
transported by the antiport) did not cause inhibition of the
up-regulation of ß2-integrins (Fig. 2)
. Therefore, one can conclude
that neither the signals induced by a decrease in cell volume nor the
transport of sodium into the cell through the antiporter is responsible
for inhibition of the up-regulation of ß2-integrins by hypertonic
saline.
Role of extracellular and intracellular sodium in the inhibition of
up-regulation of ß2-integrins
A recently published study on osmotic inhibition of exocytosis in
neutrophils suggested that hypertonic saline inhibits exocytosis of all
types of granules by enhanced submembranous F-actin ring formation and
abolishment of depolymerization, irrespective of the stimulus used
(fMLP, PMA, A23187, or LPS) [26
]. Because prevention and
induction of F-actin polymerization decreased and the inhibitory action
of hypertonic saline increased, respectively, the authors concluded
that cytoskeletal remodeling appears to be a key component in the
neutrophil-suppressive effects of hypertonic saline. It is noteworthy
that blockage of tyrosine phosphorylation by specific kinase inhibitors
(e.g., for p38 by SB203580) had no effect on the osmotically induced
actin polymerization. When neutrophils were permeabilized for
monovalent cations by nystatin, addition of NaCl (+100 mM) increased
polymerization of actin in neutrophils kept at constant cell volumes,
whereas addition of sucrose (+ 200 mM) only decreased cell volumes
without induction of F-actin. Based on these findings, the authors
concluded that elevation of intracellular ionic strength might be an
important contributor but that reduced cell volume itself did not seem
to be sufficient for actin modeling. The latter conclusion fits well
with our observation that the expression of ß2-integrins was
inhibited by the addition of NaCl but not by choline chloride,
mannitol, or sucrose, which all decrease cell volume but cannot enter
cells. However, one may argue that in intact cells, shrinking results
in a concomitant increase in intracellular cation concentrations
because of the ensuing water efflux. This should be the same with NaCl
and choline chloride, since the latter induced more shrinkage, as shown
in Figure 13
. Thus, experiments suggest the involvement of the
extracellular rather than (or in addition to) the intracellular sodium
cation concentrations in the inhibition of the fMLP-stimulated
up-regulation of ß2–integrins. Why this requirement for increases in
both extracellular and intracellular sodium (or metal) cation
concentrations is restricted to the inhibition of the fMLP-induced
response remains to be clarified by further experiments.
|
ACKNOWLEDGEMENTS |
|---|
The authors gratefully acknowledge the skillful technical assistance of Sabine Fischer, Gaby Kröger, Marion Hörl, and Ursula Göttler from the Clinic of Anesthesiology. Some of this research was conducted by Kay Eberhardt and Florian Setzer in partial fulfillment of their requirements for doctorial theses from the medical faculty of the Ludwig-Maximilians-University of Munich, Munich, Germany.
Received November 2, 1999; revised April 12, 2001; accepted April 16, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Chen, N. Hashiguchi, L. Yip, and W. G. Junger Hypertonic saline enhances neutrophil elastase release through activation of P2 and A3 receptors Am J Physiol Cell Physiol, April 1, 2006; 290(4): C1051 - C1059. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kaakinen, H. Alaoja, J. Heikkinen, S. Dahlbacka, P. Laurila, K. Kiviluoma, T. Salomaki, H. Tuominen, P. Ohtonen, F. Biancari, et al. Hypertonic Saline Dextran Improves Outcome After Hypothermic Circulatory Arrest: A Study in a Surviving Porcine Model Ann. Thorac. Surg., January 1, 2006; 81(1): 183 - 190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Larkin, D. Murphy, D. F. Reilly, M. Cahill, E. Sattler, P. Harriott, D. J. Cahill, and N. Moran ICln, a Novel Integrin {alpha}IIb{beta}3-Associated Protein, Functionally Regulates Platelet Activation J. Biol. Chem., June 25, 2004; 279(26): 27286 - 27293. [Abstract] [Full Text] [PDF]
|
|
|
|
, 230; 
, 290; 
, 350; and 
, 410; symbols shown represent
mean values of five experiments with neutrophils from different
volunteers per curve.
, P < 0.05 vs. choline chloride modified HBSS and PMLP
10-7m; paired t-test with adjustment for
multiple comparisons according to the method of Bonferroni.
