Tài liệu Báo cáo khoa học: Osmosensing and signaling in the regulation of mammalian cell function docx

MINIREVIEW
Osmosensing and signaling in the regulation
of mammalian cell function
Freimut Schliess, Roland Reinehr and Dieter Ha
¨
ussinger
Clinic for Gastroenterology, Hepatology and Infectiology, Heinrich-Heine-University, Du
¨
sseldorf, Germany
Introduction
Sudden exposure of cells to hypo- or hyperosmotic
solutions induces a rapid osmotic swelling or shrink-
age, respectively. Extensive swelling or shrinkage is
counteracted by induction of a regulatory volume
decrease (RVD) or increase, respectively [1–3]. Most
hypoosmotically swollen cells perform RVD by a
release of inorganic ions, including K
+
,Na
+
,Cl
–
, and
HCO
3
–
and organic osmolytes (e.g. taurine, betaine).
Hyperosmotic regulatory volume increase (RVI) at a
short-term time scale is performed by activation of
electrolyte uptake (e.g. via Na
+

⁄ K
+
⁄ 2Cl
–
cotransport
and Na
+
⁄ H
+
exchange). Long-term adaption to
hyperosmolarity includes an isoosmotic exchange of
inorganic ions against compatible organic osmolytes,
which preserve protein function even at high concen-
trations [4].
Transport systems involved in RVD or RVI can be
activated also by hormones, substrates, second messen-
gers and oxidative stress under isoosmotic conditions.
In these cases, moderate and well-tolerated cell volume
changes are created. For example, insulin produces
a phosphoinositide 3-kinase (PI 3-kinase)-dependent
hepatocyte swelling by inducing a net accumulation of
ions inside the cell, which results from a concerted
activation of Na
+
⁄ H
+
exchange, Na
+
⁄ K
+

⁄ 2Cl
–
sym-
port and the Na
+
⁄ K
+
-ATPase [5].
In the early 1990s, it was recognized, that cell vol-
ume changes trigger signals involved in the regulation
of metabolism, gene expression and the susceptibility
to different kinds of stress [6]. For example, the inhibi-
tion of autophagic proteolysis by insulin, glutamine
and ethanol in the perfused liver critically depends on
the degree of hepatocyte swelling induced by these
stimuli and can be mimicked by hypoosmotic swelling
Keywords
apoptosis; bile acids; CD95; cell volume;
epidermal growth factor; insulin; integrins;
osmolytes; oxidative stress; proliferation
Correspondence
F. Schliess, Heinrich-Heine-Universita
¨
t,
Universita
¨
tsklinikum, Klinik fu
¨
r
Gastroenterologie und Infektiologie,

Moorenstrasse 5, D-40225 Du
¨
sseldorf,
Germany
Fax: +49 211 81 17517
Tel: +49 211 81 18941
E-mail:
(Received 2 July 2007, accepted 29 August
2007)
doi:10.1111/j.1742-4658.2007.06100.x
Volume changes of mammalian cells as induced by either anisoosmolarity
or under isoosmotic conditions by hormones, substrates and oxidative
stress critically contribute to the regulation of metabolism, gene expression
and the susceptibility to stress. Osmosensing (i.e. the registration of cell
volume) triggers signal transduction pathways towards effector sites (osmo-
signaling), which link alterations of cell volume to a functional outcome.
This minireview summarizes recent progress in the understanding of how
osmosensing and osmosignaling integrate into the overall context of growth
factor signaling and the execution of apoptotic programs.
Abbreviations
EGF, epidermal growth factor; MAPK, mitogen-activated protein kinase; PI 3-kinase, phosphoinositide 3-kinase; RGD, arginine-glycine-
aspartic acid; ROS, reactive oxygen species; RVD, regulatory volume decrease; RVI, regulatory volume increase.
FEBS Journal 274 (2007) 5799–5803 ª 2007 The Authors Journal compilation ª 2007 FEBS 5799
[7]. On the other hand, hyperosmotic shrinkage pre-
vents insulin-induced hepatocyte swelling and proteo-
lysis inhibition, indicating that swelling-dependent
signaling essentially contributes to the entire response
to insulin [5]. In general, cell swelling stimulates ana-
bolic metabolism and proliferation and provides cyto-
protection, whereas cellular shrinkage leads to

catabolism and insulin resistance and sensitizes cells to
apoptotic stimuli.
The influence of cell volume on cell function requires
structures that register volume changes (‘osmosensing’)
and trigger signaling pathways towards effector sites
(‘osmosignaling’). Anisoosmotically exposed cells and
tissues were frequently used as a model in order to
study osmosensing and osmosignaling. By this
approach, the activation of signaling pathways by cell
volume changes could be linked to specific functional
outcomes [8]. Here, we summarize some recent pro-
gress concerning the understanding of how ‘osmosen-
sing’ and ‘osmosignaling’ integrate into the overall
context of signal transduction, which is activated by
growth factors, substrates or (pro)-apoptotic stimuli in
mammalian cells with some focus on the hepatocyte. A
more general treatize on osmosensing and osmosignal-
ing is provided elsewhere [9].
Osmosensing in mammalian cells
The investigation of osmosensing processes and struc-
tures in mammalian cells considers, among others,
macromolecular crowding, stretch-activated ion chan-
nels, cholesterol-enriched microdomains of the plasma
membrane (caveolae), intracellular organelles, ligand-
independent activation of growth factor- and cytokine
receptors and autocrine stimulation of signal transduc-
tion by release of mediators such as ATP [10,11].
Recent studies have identified the integrin system as
one major sensor of hepatocyte swelling [12–14].
Integrin-inhibitory peptides exhibiting an arginine-gly-

cine-aspartic acid (RGD) motif abolish hypoosmotic
osmosignaling towards Src-type kinases, mitogen-acti-
vated protein kinases (MAPKs) and downstream
metabolic events, including the stimulation of bile for-
mation, proteolysis inhibition and volume-regulatory
K
+
-efflux [12,13]. It should be noted that RGD pep-
tides do not inhibit hypoosmotic hepatocyte swelling
[13], indicating that inhibition of osmosensing at the
integrin level uncouples hepatocyte swelling from
osmosignaling and its functional consequences.
Like hypoosmotic hepatocyte swelling [12,13], insu-
lin-induced hepatocyte swelling is registrated by the
integrin system, leading to a Src-dependent activation
of the p38-type MAPK and thereby inhibition of
autophagic proteolysis [14]. Thus, integrin-dependent
cell volume sensing and signaling integrates into the
overall context of insulin signaling. Similarly, sensing
of glutamine-induced hepatocyte swelling by integrins
feeds into Src-dependent p38 activation, which is criti-
cally required for autophagic proteolysis inhibition by
glutamine [13].
As demonstrated recently [15], hyperosmotic hepato-
cyte shrinkage may be sensed by the endosomal com-
partment. Mild hyperosmolarity (405 mosmolÆL
)1
)in
rat hepatocytes induces a rapid endosomal acidification
by activation of vacuolar-type H

+
-ATPase, probably
driven by an increase in intracellular Cl
–
concentration
due to osmotic water loss and Cl
–
accumulation in the
course of a RVI, respectively [15]. The endosomal
compartment acidified by hyperosmolarity colocalized
with the acidic sphingomyelinase [15]. Hyperosmolarity
in hepatocytes triggers a rapid production of reactive
oxygen species (ROS), which critically depends on ser-
ine phosphorylation of the NADPH oxidase regulatory
subunit p47
phox
, which again depends on a acid sphin-
gomyelinase-catalyzed ceramide production and subse-
quent activation of the PKCf [15]. Bafilomycin A1 (an
inhibitor of vacuolar-type H
+
-ATPases) and the anion
channel blocker 4,4¢-diisothiocyanostilbene-2,2¢-disulf-
onic acid disodium salt not only prevent endosomal
acidification by hyperosmolarity, but also block the
hyperosmotic increase of ceramide, p47
phox
phosphory-
lation and ROS [15]. The findings localize endosomal
acidification most upstream in the signaling cascade

underlying hyperosmotic ROS production.
Osmosignaling in proliferation
and apoptosis
Growth factors stimulate a rapid osmolyte uptake that
is important for mitogenesis [16]. For example, a rapid
and transient Na
+
and amino acid influx is essential
for the mitogenic response of 3T3 fibroblasts to growth
factors [17,18]. Activation of Na
+
⁄ K
+
⁄ 2Cl
–
cotrans-
port via NKCC1 was shown to be essential for cell
cycle progression in 3T3 fibroblasts [17,19]. The impor-
tance of ion uptake for cell cycle progression was
strengthened by the finding that NKCC1 overexpres-
sion in 3T3 fibroblasts induces a transformed pheno-
type [20].
Cell swelling due to isoosmotic osmolyte uptake
could be one mechanism that contributes to cell cycle
progression. It was shown that cell water increases
during the cell cycle of 3T3 fibroblasts resulting from
Na
+
⁄ K
+

⁄ 2Cl
–
cotransport and glutamine uptake [17].
Likewise, hepatocyte swelling due to the activation of
system A-type amino acid transporters was observed
Osmosensing and signaling in mammalian cell function F. Schliess et al.
5800 FEBS Journal 274 (2007) 5799–5803 ª 2007 The Authors Journal compilation ª 2007 FEBS
in vivo following partial hepatectomy, and inhibition of
cell swelling antagonized liver regeneration [21]. Hypo-
osmolarity in many cell types activates the MAPKs
Erk1 ⁄ Erk2 and the PI 3-kinase [8], which play a major
role in mitogenic signaling. Hypoosmotic exposure of
HepG2 cells potentiates proliferation by a PI 3-kinase-
meditated activation of the transcription factor activa-
tor protein AP-1 [22], corroborating a critical role of
cell swelling for cell cycle progression. Consistently,
the cell volume was increased in 3T3 cells expressing
oncogenic Ha-ras [23] and cell hyperhydration has
been linked to tumor growth [24].
A volume decrease resulting from osmolyte release
through specific transport proteins at the beginning of
apoptosis (apoptotic volume decrease) is an early pre-
requisite for the execution of apoptotic programs [25].
Signaling mechanisms upstream of apoptotic volume
decrease depend on the cell type and stimulus under
investigation and have been discussed previously [26–
28]. The contribution of apoptotic volume decrease to
apoptotic signal transduction is currently not well
understood. Using hyperosmotically treated cells as a
model, it was shown that efficient volume regulation

can protect cells from apoptosis [25,29] and it was sug-
gested that the impairment of mechanisms antagoniz-
ing cell shrinkage may be a general feature of
apoptosis. For example, renal tubular epithelial cell
apoptosis was accompanied by a caspase-dependent
cleavage of the Na
+
⁄ H
+
exchanger NHE1 [30]. How-
ever, cell shrinkage is not always sufficient to trigger
apoptosis [25,29]. Mechanisms that could protect cells
from shrinkage-induced apoptosis include activation of
the protein kinase B survival pathway [31], p53 activa-
tion [32], induction of the serum- and glucocorticoid-
inducible kinase Sgk [33], expression of the heat shock
protein Hsp70 [34] and cyclooxygenase-2 [35], and a
high antioxidant capacity [36].
In hepatocytes, a close interrelation between osmotic
shrinkage and ROS production has been established.
On the one hand, hyperosmotic shrinkage induces
ROS production (see above) and, on the other, ROS
mediate cell shrinkage. Thus, a vicious circle results,
which, when not interrupted, results in apoptosis. Such
a vicious circle may be activated by osmotic shrinkage
or pro-apoptotic bile acids [37]. As shown in Fig. 1,
hepatocyte shrinkage and ROS production are con-
nected in an autoamplificatory signaling loop. Mutual
amplification of shrinkage and ROS triggers apoptosis,
which could be prevented by NAPDH oxidase inhibi-

tors and the availability of antioxidants and osmolytes.
In rat hepatocytes, mild hyperosmolarity (405
mosmolÆL
)1
) activates the CD95 system, which local-
izes downstream of the ROS production triggered by
endosomal acidification mentioned above [15,38]. Hyp-
erosmotic CD95 activation includes trafficking of the
CD95 from inside the hepatocyte to the plasma mem-
brane [39], which depends on a ROS-mediated tyrosine
phosphorylation of the epidermal growth factor
(EGF)-receptor, the association of CD95 with the
EGF-receptor, and phosphorylation of CD95 on
Tyr232 and Tyr291 by the EGF-receptor tyrosine
kinase activity [15]. Although the appearance of CD95
at the plasma membrane was associated with death
inducing signaling complex formation and activation
of caspases 3 and 8, mild hyperosmolarity was not suf-
ficient to induce hepatocyte apoptosis [39], suggesting
that apoptotic signals under this condition are counter-
balanced by yet unknown survival signals. However,
more severe hyperosmolarity (‡ 505 mosmolÆL
)1
) shifts
the balance towards hepatocyte apoptosis [15].
Like hyperosmolarity, CD95 ligand in hepatocytes
via generation of ROS induced EGF-receptor tyrosine
phosphorylation, CD95 ⁄ EGF-receptor association,
CD95 tyrosine phosphorylation, trafficking of the
CD95 to the plasma membrane surface, and death

inducing signaling complex formation, leading to the
execution of apoptosis in this case [38]. Although inef-
fective to induce apoptosis by itself, hyperosmolarity
(405 mosmolÆL
)1
) sensitized the hepatocytes towards
CD95 ligand-induced apoptosis [39], indicating a
Fig. 1. Hepatocyte shrinkage and the production of reactive oxygen
species constitute an autoamplificatory signaling loop. Hyperosmot-
ic shrinkage triggers a NADPH oxidase-catalyzed ROS formation.
ROS, again by stimulating K
+
-efflux, antagonize processes under-
lying the RVI, and thereby increase hepatocyte shrinkage by hyper-
osmolarity. Mutual amplification of swelling and oxidative stress
may be limited by the hepatocyte’s antioxidant and volume-regula-
tory capacity. Adapted from [37].
F. Schliess et al. Osmosensing and signaling in mammalian cell function
FEBS Journal 274 (2007) 5799–5803 ª 2007 The Authors Journal compilation ª 2007 FEBS 5801
synergistic interplay between signals triggered by
hyperosmotic shrinkage and CD95 ligand, respectively.
Likewise, hyperosmolarity sensitized H4IIE rat hepa-
toma cells to pro-apoptotic signaling by the protea-
some inhibitor MG-132 [40].
These and other studies on hyperosmotically shrun-
ken cells support the view that the apoptotic volume
decrease, if not even producing de novo death signals,
at least can amplify apoptotic signals released by dif-
ferent apoptotic stimuli. Thus, the apoptotic volume
decrease may further disarrange the balance between

survival and death signals, thereby promoting execu-
tion of the apoptotic program.
Concluding remarks
It is well acknowledged that cell volume fluctuations
release signals of (patho)physiological relevance. The
understanding of how cell swelling integrates into the
cell cycle machinery and how cell shrinkage sensitizes
cells to apoptotic stimuli requires further scientific
effort. Cell hydration may markedly affect the action
of drugs. For example cell hydration changes may
switch the outcome of proteasome inhibitors from a
nontoxic or even protective one to injury and apopto-
sis [40]. Although routine monitoring of cell hydration
in patients would provide valuable information in clin-
ical medicine, this is currently limited by methodologi-
cal difficulties.
Acknowledgements
Our own studies were supported by Deutsche Fors-
chungsgemeinschaft through Sonderforschungsbereich
575 ‘Experimentelle Hepatologie’ (Du
¨
sseldorf).
References
1 Chamberlin ME & Strange K (1989) Anisosmotic cell
volume regulation: a comparative view. Am J Physiol
257, C159–C173.
2 Parker JC (1993) In defense of cell volume? Am J Phys-
iol 265, C1191–C1200.
3 Hoffmann EK & Pedersen SF (2006) Sensors and signal
transduction pathways in vertebrate cell volume regula-

tion. Contrib Nephrol 152, 54–104.
4 Burg MB (1995) Molecular basis of osmotic regulation.
Am J Physiol 37, F983–F996.
5 Schliess F, vom Dahl S & Ha
¨
ussinger D (2001) Insulin
resistance by loop diuretics and hyperosmolarity in
perfused rat liver. Biol Chem 382, 1063–1069.
6Ha
¨
ussinger D & Lang F (1992) Cell volume and hor-
mone action. Trends Pharmacol Sci 13, 371–373.
7 vom Dahl S & Ha
¨
ussinger D (1996) Nutritional state
and the swelling-induced inhibition of proteolysis in
perfused rat liver. J Nutr 126, 395–402.
8Ha
¨
ussinger D & Schliess F (1999) Osmotic induction of
signaling cascades: role in regulation of cell function.
Biochem Biophys Res Comm 255, 551–555.
9Ha
¨
ussinger D & Sies H, eds (2007) Osmosensing and
osmosignaling. Methods Enzymol. Vol 428, Academic
Press, in press.
10 Hamill OP & Martinac B (2001) Molecular basis of
mechanotransduction in living cells. Physiol Rev 81,
685–740.

11 Ha
¨
ussinger D, Reinehr RM & Schliess F (2006) The
hepatocyte integrin system and cell volume sensing.
Acta Physiol 187, 249–255.
12 Ha
¨
ussinger D, Kurz AK, Wettstein M, Graf D, vom
Dahl S & Schliess F (2003) Involvement of integrins
and Src in tauroursodeoxycholate-induced and
swelling-induced choleresis. Gastroenterology 124,
1476–1487.
13 vom Dahl S, Schliess F, Reissmann R, Go
¨
rg B,
Weiergra
¨
ber O, Kocalkova M, Dombrowski F &
Ha
¨
ussinger D (2003) Involvement of integrins in osmo-
sensing and signaling toward autophagic proteolysis in
rat liver. J Biol Chem 278, 27088–27095.
14 Schliess F, Reissmann R, Reinehr R, vom Dahl S &
Ha
¨
ussinger D (2004) Involvement of integrins and Src
in insulin signaling towards autophagic proteolysis in
rat liver. J Biol Chem 279, 21294–21301.
15 Reinehr R, Becker S, Braun J, Eberle A, Grether-Beck

S&Ha
¨
ussinger D (2006) Endosomal acidification and
activation of NADPH oxidase isoforms are upstream
events in hyperosmolarity-induced hepatocyte apoptosis.
J Biol Chem 281, 23150–23166.
16 Rozengurt E (1986) Early signals in the mitogenic
response. Science 234, 161–166.
17 Bussolati O, Uggeri J, Belletti S, Dallasta V & Gazzola
GC (1996) The stimulation of Na ⁄ K ⁄ 2Cl cotransport
and of system AS for neutral amino acid transport is a
mechanism for cell volume increase during the cell cycle.
FASEB J 10, 920–926.
18 Berman E, Sharon I & Atlan H (1995) An early tran-
sient increase of intracellular Na
+
may be one of the
first components of the mitogenic signal. Direct detec-
tion by 23Na-NMR spectroscopy in quiescent 3T3
mouse fibroblasts stimulated by growth factors. Biochim
Biophys Acta 1239, 177–185.
19 Koch KS & Leffert HL (1979) Increased sodium influx
is necessary to initiate rat hepatocyte proliferation. Cell
18, 153–163.
20 Panet R & Atlan H (2000) Overexpression of the
Na
+
⁄ K
+
⁄ 2Cl

–
cotransporter gene induces cell prolifera-
tion and phenotypic transformation in mouse fibro-
blasts. J Cell Physiol 182, 109–118.
Osmosensing and signaling in mammalian cell function F. Schliess et al.
5802 FEBS Journal 274 (2007) 5799–5803 ª 2007 The Authors Journal compilation ª 2007 FEBS
21 Freeman TL, Ngo HQ & Mailliard ME (1999) Inhibi-
tion of system A amino acid transport and hepatocyte
proliferation following partial hepatectomy in the rat.
Hepatology 30, 437–444.
22 Kim RD, Roth TP, Darling CE, Ricciardi R, Schaffer
BK & Chari RS (2001) Hypoosmotic stress stimulates
growth in HepG2 cells via protein kinase B-dependent
activation of activator protein-1. J Gastrointest Surg 5,
546–555.
23 Fu
¨
rst J, Haller T, Chwatal S, Wo
¨
ll E, Dartsch PC,
Gschwentner M, Dienstl A, Zwierzina H, Lang F, Paul-
michl M et al. (2002) Simvastatin inhibits malignant
transformation following expression of the Ha-ras onco-
gene in NIH 3T3 fibroblasts. Cell Physiol Biochem 12,
19–30.
24 McIntyre GI (2006) Cell hydration as the primary factor
in carcinogenesis: a unifying concept. Med Hypotheses
66, 518–526.
25 Maeno E, IshizakiY, Kanaseki T, Hazama A &
OkadaY (2000) Normotonic cell shrinkage because of

disordered volume regulation is an early prerequisite to
apoptosis. Proc Natl Acad Sci USA 97, 9487–9492.
26 Schliess F & Ha
¨
ussinger D (2002) The cellular hydra-
tion state: a critical determinant for cell death and sur-
vival. Biol Chem 383, 577–583.
27 Lang F, Huber SM, Szabo I & Gulbins E (2007) Plasma
membrane ion channels in suicidal cell death. Arch
Biochem Biophys 462, 189–194.
28 Bortner CD & Cidlowski JA (2007) Cell shrinkage and
monovalent cation fluxes: role in apoptosis. Arch Bio-
chem Biophys 462, 176–188.
29 Maeno E, Takahashi N & OkadaY (2006) Dysfunction
of regulatory volume increase is a key component of
apoptosis. FEBS Lett 580, 6513–6517.
30 Wu KL, Khan S, Lakhe-Reddy S, Wang L, Jarad G,
Miller RT, Konieczkowski M, Brown AM, Sedor JR &
Schelling JR (2003) Renal tubular epithelial cell apopto-
sis is associated with caspase cleavage of the NHE1
Na
+
⁄ H
+
exchanger. Am J Physiol 284, F829–F839.
31 Wu KL, Khan S, Lakhe-Reddy S, Jarad G, Mukherjee
A, Obejero-Paz CA, Konieczkowski M, Sedor JR &
Schelling JR (2004) The NHE1 Na
+
⁄ H

+
exchanger
recruits ezrin ⁄ radixin ⁄ moesin proteins to regulate Akt-
dependent cell survival. J Biol Chem 279, 26280–26286.
32 Dmitrieva NI & Burg MB (2005) Hypertonic stress
response. Mutat Res 569, 65–74.
33 Leong ML, Maiyar AC, Kim B, O’Keeffe BA &
Firestone GL (2003) Expression of the serum- and gluco-
corticoid-inducible protein kinase, Sgk, is a cell survival
response to multiple types of environmental stress stimuli
in mammary epithelial cells. J Biol Chem 278, 5871–5882.
34 Shim EH, Kim JI, Bang ES, Heo JS, Lee JS, Kim EY,
Lee JE, Park WY, Kim SH, Kim HS et al. (2002) Tar-
geted disruption of hsp70.1 sensitizes to osmotic stress.
EMBO Rep 3, 857–861.
35 Neuhofer W, Holzapfel K, Fraek ML, Ouyang N, Lutz
J & Beck FX (2004) Chronic COX-2 inhibition reduces
medullary HSP70 expression and induces papillary
apoptosis in dehydrated rats. Kidney Int 65, 431–441.
36 Schliess F & Ha
¨
ussinger D (2005) The cellular hydra-
tion state: role in apoptosis and proliferation. Signal
Transduct 6, 297–302.
37 Becker S, Reinehr R, Graf D, vom Dahl S & Ha
¨
ussinger
D (2007) Hydrophobic bile acids induce hepatocyte
shrinkage via NADPH oxidase activation. Cell Physiol
Biochem 19, 89–98.

38 Reinehr R, Schliess F & Ha
¨
ussinger D (2003) Hyper-
osmolarity and CD95L trigger CD95 ⁄ EGF receptor
association and tyrosine phosphorylation of CD95 as
prerequisites for CD95 membrane trafficking and DISC
formation. FASEB J 17, 731–733.
39 Reinehr RM, Graf D, Fischer R, Schliess F & Ha
¨
us-
singer D (2002) Hyperosmolarity triggres CD95 mem-
brane trafficking and sensitizes rat hepatocytes towards
CD95L-induced apoptosis. Hepatology 36, 602–614.
40 Lornejad-Scha
¨
fer MR, Scha
¨
fer C, Richter L, Grune T,
Ha
¨
ussinger D & Schliess F (2005) Osmotic regulation of
MG-132-induced MAP-kinase phosphatase MKP-1
expression in H4IIE rat hepatoma cells. Cell Physiol
Biochem 16, 193–206.
F. Schliess et al. Osmosensing and signaling in mammalian cell function
FEBS Journal 274 (2007) 5799–5803 ª 2007 The Authors Journal compilation ª 2007 FEBS 5803