CHAPTER 13 – STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

263

13
CHAPTER

STRUCTURE AND FUNCTION OF
OSMOREGULATED ABC
TRANSPORTERS
BERT POOLMAN AND
TIEMEN VAN DER HEIDE

INTRODUCTION
GENERAL
In their natural habitats, microorganisms are
often exposed to osmolality changes in the
environment. For instance, soil bacteria such as
Bacillus subtilis are alternately exposed to periods of drought and rain, to which they have to
adapt. Since the cytoplasmic membrane of bacteria is highly permeable to water but forms an
effective barrier for most solutes present in the
medium and metabolites present in the cytoplasm, water will flow out of the cell when the
outside osmolality increases (‘osmotic upshift’).
As a consequence of an osmotic upshift, the
turgor pressure will decrease and ultimately
the cell may plasmolyze. Upon osmotic downshift, water will flow into the cell and thereby
increase the turgor pressure. Maintenance of
cell turgor is a prerequisite for almost any form
of life, as it provides a mechanical force for the
expansion of the cell envelope and regulates cell
growth. Generally, (micro)organisms respond
to an osmotic upshift by rapidly accumulating
compatible solutes to prevent the loss of water

and loss of turgor pressure. Upon osmotic
downshift, the cells need to rapidly export the
solutes to prevent the turgor pressure becoming too high, which, ultimately, may lead to
breakage of the cells. Since changes in protein
expression (biosynthesis) are relatively slow, it
ABC Proteins: From Bacteria to Man
ISBN 0-12-352551-9

is evident that the primary response to osmotic
stress needs to be one in which transport systems or channel proteins already present in the
cytoplasmic membrane are activated in order to
accumulate or release solutes (Glaasker et al.,
1996a, 1996b). This implies that osmotic stress
must be sensed by these systems and converted
into a change in the appropriate activity such
that the osmotic imbalance is rapidly restored.
Most osmotically controlled uptake systems
are regulated at both the genetic (induction of
gene expression) and the enzymatic level
(direct ‘activation’ of the transport protein).
The degree of induction can vary considerably
(from 2- up to 500-fold), whereas in vivo activation of the transport protein by an osmotic
upshift is usually in the range of 5- to 35-fold.
The transport systems known to be activated
by osmotic upshift are either ATP-binding cassette (ABC) transporters or so-called secondary
transporters, that is, transport proteins driven
by an electrochemical ion gradient (Wood,
1999). The majority of both types of these
energy coupling mechanisms have the quaternary ammonium compound glycine betaine as
the preferred (high affinity) substrate. Wellcharacterized osmotic downshift activated

systems are the mechanosensitive channels,
most notably the MscL protein, which open
and thereby release osmolytes when the turgor
pressure becomes too high. This chapter only
summarizes our knowledge of the ABC-type
osmoregulated transporters.
Copyright 2003 Elsevier Science Ltd
All rights of reproduction in any form reserved


264

ABC PROTEINS: FROM BACTERIA TO MAN

BOX 13.1. DEFINITIONS
The term hyperosmotic stress is often used to indicate
an increased osmolality of the external medium. This
is somewhat confusing as the cytoplasm of growing
cells in complete osmotic equilibrium is generally
hyperosmotic relative to the outside, even after an
osmotic upshift. Therefore, we prefer to use the terms
osmotic upshift and downshift for conditions that are
often referred to as hyper- and hypo-osmotic stress.
Turgor pressure (⌬P) is the hydrostatic pressure
difference that balances the difference in internal and
external osmolyte concentration.
⌬P ϭ (RT/Vw) ln(ao/ai) Ϸ RT(ci Ϫ co)
in which Vw is the partial molal volume of water, a is
the water activity, c is the total osmolyte concentration,
and the subscripts i and o refer inside and outside,

respectively. A cell plasmolyzes when ⌬P becomes
negative.
Osmolality is the osmotic pressure of a solution at a
particular temperature, expressed as moles of solute per
kilogram of solvent (osmol kgϪ1 or osmolal). The often
used term osmolarity is an approximation for osmolality
and is expressed as osmol lϪ1 or osmolar (for a more
extensive overview of definitions, see Wood, 1999).

BOX 13.2. COMPATIBLE SOLUTES
Compatible solutes are molecules that can be
accumulated to high cytoplasmic (often molar)
concentrations without affecting vital cellular
processes. In many organisms, the compatible solutes
glycine betaine and carnitine (and other quaternary
ammonium compounds) offer the highest protection
against osmotic stress.

Glycine betaine
CH3

ϩ

N

CH3 CH3

Carnitine
Ϫ

COO

CH3

ϩ

N

under high osmolality conditions, but it is
difficult to derive a quantitative relationship
between the internal concentrations of compatible solutes and the external osmolality as
multiple (macro)molecules are involved. The
compatible solutes to be accumulated to high
intracellular levels are restricted to a few categories (for reviews see Csonka, 1989; Csonka
and Hanson, 1991; Galinski and Trüper, 1994),
and they include (i) potassium ions, (ii) amino
acids (glutamate, alanine, proline), (iii) amino
acid analogues (taurine, N-acetylglutaminylglutamine amide), (iv) methyl-amines and
related compounds (glycine betaine, carnitine,
ectoines), and (v) polyols and sugars (glycerol,
sucrose, lactose). In general, compatible solutes
should not interact specifically with the (mostly
negatively charged) cellular macromolecules,
nor should they perturb cytoplasmic solutes
via (de)hydration, precipitation, or any other
(charge) interaction. Therefore, under steady
state conditions most compatible solutes that
are present in large amounts in the cytoplasm
have no net charge. The accumulation of potassium ions in response to an osmotic upshift
in enteric bacteria is usually only transient.

Following the accumulation of potassium ions,
other compatible solutes are synthesized (e.g.
trehalose), and the uptake systems for glycine
betaine and proline are induced. The accumulation of these solutes then eventually replaces
potassium (see also Poolman and Glaasker,
1998). An exception to the rule of preferring
neutral compatible solutes is found in thermophilic organisms (Bacteria and Archaea),
where negatively charged compounds seem to
be accumulated in response to osmotic upshifts.
These compounds include mannosylglycerate, glutamate, cyclic-2,3-bisphosphoglycerate,
1,3,4,6-tetracarboxyhexane, and myo-inositolphosphate derivates (Martins and Santos, 1995;
Martins et al., 1996).

Ϫ

OH

COO

CH3 CH3

RATIONALE FOR THE ACCUMULATION OF
COMPATIBLE SOLUTES
After an osmotic upshift, cells benefit from the
accumulation of compatible solutes by balancing the osmotic disturbance via an increase in
cytoplasmic osmolality. High intracellular concentrations of compatible solutes are common

STRUCTURAL ANALYSIS
OF OSMOREGULATED
ABC TRANSPORTERS

Typical ABC-type binding protein-dependent
transporters are composed of five protein(s)
(domains), that is, an extracellular binding protein (receptor), two ATP-binding subunits and
two integral membrane subunits. The integral
membrane subunits provide the translocation


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

Gram-negative bacteria
X
Gram-positive bacteria
C

X

C

Out
W
V

W
V

B

B

A

A

BC
A

A

ATP

ATP

ATP

In
ATP

ATP

ProUEc

ATP

OpuABS

BC

OpuALI

Figure 13.1. Structural organization of osmoregulated ABC transporters. Depicted are the ProU system from

E. coli, OpuABs from B. subtilis and OpuALl from L. lactis. In Gram-negative bacteria, the binding protein is
present in the periplasm, whereas in Gram-postive bacteria, the protein is anchored to the outer surface of
the membrane via a lipid moiety (OpuAC) or fused to the translocator domain (OpuABC). For ProX and
OpuAC, the equilibrium between unliganded and liganded binding protein is depicted.

pathway. Except for the substrate-binding
protein, the other subunits can be present as
distinct polypeptides or fused to one another
but always each entity is present twice. The
osmoregulated ABC transporters belong to
the OTCN family (Dassa and Bouige, 2001), of
which the members transport substrates as
diverse as quaternary ammonium compounds
(glycine betaine, carnitine), proline, alkylsulfonates and -phosphonates, phosphites,
cyanate, and nitrate.
Although biochemical evidence is not available, the osmoregulated ABC transporters most
probably have two identical copies of the ATPbinding subunit. This suggestion is based on
the finding that, in contrast to other families
of ABC transporters, the operons specifying
osmoregulated transporters only have a single
gene for an ATP-binding subunit. The integral
membrane components are either present as
two identical copies or two homologous proteins, but they are not fused to each other or
to the ATP-binding subunit (Figure 13.1). The
ligand-binding subunit is present as a free protein in the periplasm of Gram-negative bacteria,
whereas in Gram-positive bacteria the protein
can be anchored to the membrane through a
1

fatty acid modification of the amino-terminal

cysteine (Kempf et al., 1997; Sankaran and Wu,
1994). Recently, it has been shown that a subset of ABC transporters has the ligand-binding
protein fused to the integral membrane subunit (Figure 13.1). OpuA1 of Lactococcus lactis is
the only well-studied representative of this subset of binding protein-dependent ABC transporters, but database searches indicate that
similar systems are present in Streptomyces coelicolor, Streptococcus pneumoniae, Streptococcus pyogenes, Chlamydia pneumoniae, Helicobacter pylori
and Staphylococcus aureus. By analogy with
other ABC transporters, functional OpuA will
most likely be composed of two integral membrane subunits and two ATP-binding subunits.
Since the translocator subunit is fused to the
substrate-binding protein (OpuABC subunit),
the oligomeric structure implies that two receptor domains are present per functional complex.
This raises questions about the observations
that only a single substrate-binding protein
interacts with the dimeric membrane complex
of an ABC transporter, and that two lobes
of a single substrate-binding protein interact
with different integral membrane protein(s)
(domains) (Ehrmann et al., 1998; Liu et al., 1999).

OpuA of L. lactis is composed of two different types of subunits, that is, the ATPase subunit OpuAA and the chimeric
ligand-binding/translocator protein OpuABC. OpuA of B. subtilis has three different subunits, OpuAA, OpuAB
and OpuAC, and here the translocator (OpuAB) and binding protein (OpuAC) are separate polypeptides (see also
Figure 13.1). To discriminate the L. lactis system from OpuA of B. subtilis, we use the subscripts Ll and Bs, respectively.

265


266

ABC PROTEINS: FROM BACTERIA TO MAN

THE ABC COMPONENT
The ABC or ATPase component of known
osmoregulated ABC transporters has a molecular mass of about 45 kDa. Similar to the ATPase
subunit (MalK) of the maltose transporters
of Escherichia coli and Thermococcus litoralis
(Diederichs et al., 2000), the ABC component
of most osmoregulated systems consists of an
amino-terminal ␣/␤ type ATPase domain
(ϳ27 kDa) and a carboxyl-terminal regulatory
domain (ϳ18 kDa). The carboxyl-terminal
domain in MalK of E. coli provides the system
with a means to control the transport activity as
well as the expression of the mal genes through
interaction of MalK with regulatory proteins
(Chapter 9). Whether the regulatory domain of
the ATPase of the osmoregulated ABC transporters has such a role is not known. Database
searches indicate that some homologues of the
family of osmoregulated ABC transporters have
an ATPase component with a truncated regulatory domain. These systems, e.g. ChoQ of
L. lactis, have a molecular mass of about 35 kDa.

THE INTEGRAL MEMBRANE COMPONENT
Hydropathy profiling of the sequences of the
integral membrane components of the osmoregulated ABC transporters shows that the systems fall in four subfamilies (Figure 13.2). Each
Subfamily I

of the transporters has the conserved EAA motif
in the equivalent cytoplasmic loop (loop V–VI
in subfamilies I and III and loop III–IV in subfamilies II and IV, see Figure 13.2). Members

of subfamily I, represented by ProW of the
E. coli ProU system and OpuAB of B. subtilis
OpuABs (Figure 13.1), are predicted to have
seven transmembrane ␣-helical segments with
the N-terminus facing the external surface
of the membrane and the C-terminus facing
the cytoplasm. This membrane topology is
supported by phoA and lacZ fusions, albeit a
limited set of data (Haardt and Bremer, 1996).
ProW differs from OpuABBs (and other subfamily I homologues) by having an unusually long
amino-terminus of about 100 residues (depicted
as dotted circle in Figure 13.2), which protrudes into the periplasmic space of E. coli. The
members of subfamily II, represented here
by OpuCB and OpuCD of the OpuC system of
B. subtilis, have five transmembrane ␣-helical
segments with the N- and C-termini at the
external and internal side of the membrane,
respectively. Thus, compared to the members of
subfamily I, the first two transmembrane segments are missing.
Subfamily III, exemplified by OpuABC of the
OpuA system of L. lactis, is similar to subfamily
I except that the member proteins have the
ligand-binding domain fused to the integral
membrane component (Figure 13.1). In this case,
Subfamily II

Out

Out

In

In

EAA

I

EAA

II III IV V VI VII

I

Subfamily III

II III IV V VI VII

Subfamily IV

Out

Out

In

In
EAA

EAA

I II III IV V VI VII VIII

I

II III IV V VI VII VIII

Figure 13.2. Topology models of the members of the four subfamilies of osmoregulated ABC transporters.
The additional periplasmic domain in some members of subfamily I is indicated as a dotted circle.
The Pacman-like structure connected to the last transmembrane segment of the members of subfamilies
III and IV represents the ligand-binding domain. The EAA motif in the cytoplasmic loop is also depicted.


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

an eight transmembrane segment allows the
binding domain to face the exterior of the cell. It
is worth noting that the majority of multidomain
membrane transport proteins have the individual domains fused to another via so-called flexible linker regions (Sutrina et al., 1990). In the case
of OpuA of L. lactis, the predicted end of the
eighth transmembrane segment is followed by
a sequence that over its entire length is highly
similar (Ͼ50% identity) to the ligand-binding
protein ProX of the E. coli ProU system. On the
assumption that the amino-terminus of ProX
forms an intrinsic part of the binding protein, a
flexible linker will not be present between the
transmembrane and ligand-binding domains of
the OpuABC polypeptide of the OpuA system.
Finally, members of subfamily IV, exemplified

here by ProWX of H. pylori, and the membrane
component of the ChoQ complex of L. lactis, are
similar to those of subfamily II, except that also
in this case the ligand-binding protein is fused
to the integral membrane component. This
gives rise to a membrane protein with six predicted transmembrane ␣-helical segments and
an amino- and carboxyl-terminus that are facing the outer surface of the membrane. At present, it is unclear whether or not the smaller sizes
of the proteins of subfamilies III and IV when
compared to those of I and II are related to a different functioning of the systems.

THE SUBSTRATE-BINDING PROTEIN
Members of subfamily I and II have a soluble
substrate-binding protein that in Gram-negative bacteria resides in the periplasm. In Grampositive bacteria, the ‘soluble’ binding protein
of subfamily I and II type transporters is
anchored to the outer surface of the cytoplasmic membrane via an amino-terminal lipid
moiety. Recently, ProX of E. coli has been crystallized (Breed et al., 2001) and the structure,
although not published, has been presented at
scientific meetings. As expected, ProX has an
overall tertiary fold typical of ligand-binding
proteins belonging to ABC transporters
(Quiocho and Ledvina, 1996; see Chapter 10),
which is indicative of a Venus fly-trap mechanism for substrate binding. Interestingly, the
substrate, glycine betaine, is bound to Trp
residues via cation-␲ interactions similar to the
observed binding of quaternary ammonium
compounds to acetylcholine esterase of human
brain (Bartolucci et al., 2001; Bremer and Welte,
unpublished). The ␲-electrons interact with

the quaternary ammonium group of glycine

betaine. One of the three Trp residues that interact with the substrate is highly conserved in the
glycine betaine-binding proteins for which the
primary sequences are available to date.

SPECIFICITY OF
OSMOREGULATED ABC
TRANSPORTERS
The substrate specificity of osmoregulated
(ABC) transporters has been investigated
most extensively in B. subtilis (Kempf and
Bremer, 1998). Some systems seem very specific, e.g. OpuB only selects choline, whereas
others such as OpuABs accept a wide range of
substrates. It should be stressed, however, that
in neither case has the specificity been determined directly, that is, as dissociation constants
(Kd values) of ligand binding to the receptor.
The specificity of bacterial osmoregulated
ABC transporters has been determined either
as apparent affinity constants of transport (Km)
or it has been inferred from the ability of a compound to offer protection during growth under
hyperosmotic conditions. Given our experience with this type of analysis for the
oligopeptide-binding protein of the Opp ABC
transporter (Detmers et al., 2000; Lanfermeijer
et al., 1999, 2000), we expect that differences
may be more subtle than suggested by the data
presented in the literature. Nevertheless, in
terms of cell physiology, it is important to
know if a compound does or does not offer protection against osmotic stress when taken up
via a particular system. The differences in
narrow versus broad specificity are not readily
revealed in the primary sequences of the ligandbinding proteins/domains when analyzed by

multiple sequence alignments.
For Listeria monocytogenes and Lactobacillus
plantarum, it has been shown that preaccumulated (trans) substrate inhibits the corresponding osmoregulated transport systems
(Glaasker et al., 1998; Verheul et al., 1997).
Upon raising the medium osmolality, the systems are rapidly activated through a diminished
inhibition by trans substrate. Once turgor pressure has been restored, the cells are in osmotic
equilibrium again and the transporters need to
be inactivated or switched off. The so-called
trans-inhibition may serve as a control mechanism to prevent the accumulation of these

267


268

ABC PROTEINS: FROM BACTERIA TO MAN

compatible solutes to too high levels and thereby
the turgor pressure from becoming too high.
In the case of L. monocytogenes, carnitine is
taken up via an ABC transporter that is specific
for this substrate but is inhibited by high intracellular concentrations of both carnitine and
glycine betaine (and perhaps other compatible
solutes) depending on the osmotic status of the
cells. In kinetic terms, the osmotic activation of
the system results in an increase in apparent
inhibition constant (KI) for glycine betaine and
carnitine at the inner surface of the membrane.
Apparently, as a consequence of the water efflux
following an osmotic upshift, the internal binding site for glycine betaine and carnitine, of a

system specific for carnitine in the uptake reaction, is altered. Binding of compatible solutes to
an internal site thus seems to represent a key
step in the activation–inactivation mechanism.
There is yet no molecular data on the proposed
regulatory binding site in the transporters, and
it is not therefore clear whether this is in the
translocator protein or the ATPase component.

OSMOTIC REGULATION
OF EXPRESSION
OF GENES ENCODING
ABC TRANSPORTERS
KINETICS OF OSMOTIC REGULATION AND
OSMOTIC SIGNALS

The ProU system of E. coli is the best-studied
ABC transporter in terms of osmotic regulation
of gene expression, but despite intensive
efforts, a full understanding of the molecular
mechanism(s) underlying proU expression has
not yet been achieved. In a coupled transcription–translation system, the kinetics of induction of the proU operon involves a lag phase of
15 to 20 min, followed by a rapid increase in
expression, and, subsequently, a slow decay in
the expression rate (Jovanovich et al., 1989).
This genetic regulation is slow in comparison
to the activation of existing ProU (see below),
which occurs within seconds following a
change in osmolality. Therefore, the increased
expression of proU could be mediated by
signaling molecules (second messengers) that

have to build up in the cytoplasm, rather than
by activation directly by signal transduction

pathways. Consistent with a role for specific
second messenger molecules are observations
that potassium glutamate is (at least partially)
responsible for the induction of proU (Booth,
1992; Ramirez et al., 1989), but other studies
have rejected these claims (Csonka et al., 1994;
Jovanovich et al., 1989). It is now thought that
the stimulation of transcription of proU in vitro
by potassium glutamate is a manifestation of the
generally favorable effect of these osmolytes on
macromolecular function (e.g. RNA polymerase–promoter interaction) and is not unique
to osmotic regulation of the proU promoter
(Csonka et al., 1994). Taken together, the data
are consistent with changes in intracellular
osmolality as the signal for proU transcription,
which would lead to maximal levels of ProU at
a time that the turgor has already (largely) been
restored. Such a signaling mechanism is in line
with the observation that E. coli responds to
osmotic upshift by rapidly accumulating potassium glutamate and subsequent replacement of
these ionic osmolytes for neutral ones such as
the substrate of the ProU system.

TRANSCRIPTION FACTORS AND
OSMOREGULATED PROMOTERS

Contrary to what one would expect for a system that is tightly regulated by the osmolality

of the medium (Ͼ500-fold induction), specific
transcription factors do not seem to be involved
or at least they have not been discovered.
Genetic searches for such proteins have led to
the isolation of mutants with defects in general
DNA-binding proteins such as TopA, GyrAB,
IHF, HU and H-NS (Kempf and Bremer, 1998).
Mutations in these proteins have pleitropic
effects on gene expression, for instance through
alterations in DNA supercoiling, and it is
unlikely that the tight osmotic control of the
proU operon is solely mediated by these proteins. On the basis of transcriptional analysis
of the P1 promoter of the proP gene, which
encodes an osmoregulated secondary proline
transporter, the suggestion has been made that
the cAMP receptor protein (CRP) could function as a general osmoregulator of transcription in E. coli (Landis et al., 1999). Binding of
CRP to a site within the proP P1 and some
other promoters is destabilized after an
osmotic upshift. These studies imply that CRP
could have a general osmoregulatory role in
addition to its function in catabolite control.
Transcription of proU is effected via the
promoters P1 (sigma factor ␴S) and P2 (sigma


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

OSMOTIC REGULATION
OF ACTIVITY OF ABC
TRANSPORTERS

KINETICS OF OSMOTIC (IN)ACTIVATION
In order to cope effectively with osmotic stress,
cells need osmotically controllable systems in
the membrane at all times as de novo synthesis
takes too long and is not adequate for a quick
response. In terms of an osmosensing/regulation mechanism, the only well-characterized
ABC transport protein is OpuA of L. lactis (van
der Heide and Poolman, 2000b; van der Heide
et al., 2001). The complete protein complex
(OpuAA and OpuABC) has been purified and
studied in proteoliposomes. By including ATP
plus an ATP-regenerating system inside the
vesicle lumen (Figure 13.3), uptake of glycine
betaine has been followed in response to
osmotic shifts, as a function of membrane lipid

AA AA
ADP ATP ATP ADP

AA

ADP
ATP

AA

factor ␴70). During exponential growth, transcription from P2 contributes most to the
expression by employing ␴70. Sigma factor ␴S
generally contributes to the expression of genes
in the stationary phase of growth, but the transcription of proU is not significantly increased

under these conditions. The presence of potassium glutamate enhances transcription via ␴S
and ␴70, and increases the selectivity of ␴S for P1
in vitro (Rajkumari et al., 1996).
In organisms other than E. coli, e.g. B. subtilis
and L. lactis, the expression of the genes specifying osmoregulated ABC transporters is also
under osmotic control (Kempf and Bremer, 1998;
van der Heide and Poolman, 2000a), but little
is known about the signals and transcription
factors that regulate expression. In B. subtilis, a
general stress regulon is present, whose expression depends on the alternative sigma factor
SigB (␴B). In addition to salt, heat, oxidative and
pH stresses also affect the expression of the ␴B
regulon (Wood et al., 2001). The induction of the
␴B regulon by osmotic upshift is only transient
and ␴B-controlled proteins cannot adequately
protect cells against prolonged high osmolality
stress. The structural genes for the glycine
betaine (OpuD) and proline (OpuE) secondary
transport proteins are members of the ␴B regulon, but there is no evidence that the osmoregulated ABC transporters (OpuABs, OpuB and
OpuC) of B. subtilis are under the control of ␴B.
The choline-specific transporter OpuB is under
the control of the GbsR repressor, but this transcription factor is a choline sensor rather than an
osmosensor/regulator (Bremer, 2002). Finally, it
should be stressed that maximal rates of uptake
via OpuABs and OpuC of B. subtilis increase only
1.5- to 3-fold when 0.4 M NaCl is added to the
growth medium (Kappes et al., 1996), indicating
that the corresponding genes are not under tight
osmotic control, as for example in the case of
proU of E. coli. In L. lactis IL1403, the glycine

betaine uptake capacity increases more than 10fold when the cells are grown in the presence of
0.5 M KCl (van der Heide and Poolman, 2000a).
The induction by osmotic stress, however, is
only observed in chemically defined media and
not in complex broth, suggesting that factors
other than osmolality of the medium tune the
expression levels to the needs of the cell. A preliminary report on the regulation of expression
of the ABC transporter OpuALl suggests that a
transcriptional regulator of the GnrR family acts
as a repressor of the opuA operon (Obis et al.,
2000), but the signal sensed by the protein is not
known.

ATP
Creatine phosphate
ϩADP

ADP

Creatine kinase
Creatine ϩ ATP

Figure 13.3. Proteoliposomal system to measure
the activity of OpuA from L. lactis. The purified
protein complex was inserted into preformed
liposomes, after which excess detergent was
removed by adsorption onto polystyrene beads. The
ATP-regenerating system (ATP, creatine kinase plus
creatine phosphate) was included in the vesicle
lumen by freeze-thawing, followed by sizing of the

proteoliposomes by extrusion through
polycarbonate filters with 200 nm pores (for details
see van der Heide and Poolman, 2000b; van der
Heide et al., 2001).

269


270

ABC PROTEINS: FROM BACTERIA TO MAN

composition and various water stress related
parameters. It has been shown that OpuALl is
activated instantaneously upon raising the
osmolality of the medium, that is, when the
external medium is made hyperosmotic relative
to the inside. Activation has been measured as
an increase in either the rate of ATP-driven
substrate uptake or the rate of substratedependent ATP hydrolysis; for the latter measurements, ATP rather than the ATP-regenerating
system was included in the proteoliposomes.
Activation is elicited by ionic and non-ionic
osmolytes, provided the molecules do not
equilibrate across the membrane on the time
scale of the transport measurements (van der
Heide and Poolman, 2000b). Since (proteo)liposomes act as osmometers, that is, water diffuses
across the membrane in response to the
osmotic difference between the inner compartment and the outside medium, proteoliposomes are expected to decrease their volume to
surface ratio when the outside osmolality is
increased. The changes in membrane structure

and lumen contents (osmolyte concentration)
in osmotically stressed (proteo)liposomes can
be compared with those in cells that are in a
state of plasmolysis. Proteoliposomes with an
average diameter of 200 nm change their shape
from spherical to sickle-shaped as shown by
cryo-electron microscopy (cryo-EM). These
morphological changes occurred within milliseconds, i.e. on a time scale much shorter than
the interval over which transport was measured. Upon lowering the outside osmolality to
the initial value, yielding iso-osmotic conditions again, the vesicles regained their spherical shape and the transporter was deactivated
(van der Heide et al., 2001). Thus, osmotic activation and inactivation of OpuALl is entirely
reversible, occurs on a time scale of seconds or
less, and follows the shape and volume changes
of the liposomes (see below, for proposed
mechanism of osmosensing).
In contrast to the OpuA system from L. lactis,
the hyperosmotic activation of ProU from
E. coli takes several minutes (Faatz et al., 1988),
which makes it unlikely that decreased external water activity or reduced turgor pressure
triggers the activation. A closer look at the data
suggests that activation of ProU is also dependent on the presence of the transport substrate
(glycine betaine), as the transport activity
increases up to 3 min after an osmotic upshift in
the absence but not in the presence of glycine
betaine (Faatz et al., 1988). Overall, the regulation of ProU transport activity seems different

from that of OpuA of L. lactis and resembles
more that of the transcription of proU, i.e. both
are delayed upon an upshift. It should be
stressed here that OpuALl has been studied in a

well-defined proteoliposomal system whereas
ProU has been analyzed in intact cells, which
complicates a direct comparison of the data.

MECHANISM(S) OF OSMOSENSING
Osmotic activation of membrane transporters
may be triggered through a change in the
hydration state of the transport protein complex, resulting from the altered water activity
(aw), or the signal may be transmitted to the
protein via a specific signaling molecule or a
change in the physicochemical properties of the
surrounding membrane. Since OpuA of L. lactis
can be activated not only by osmotic upshift but
also by the insertion of cationic amphipaths
in the membrane, it seems plausible that the
membrane transduces the activation (van der
Heide and Poolman, 2000b). The observation
that an osmotic upshift leads to sickle-shaped
vesicle structures already indicates that at the
macroscopic level major changes take place at
the membrane. These morphological changes,
however, do not occur upon insertion of
amphipaths into the membrane, indicating that
the macroscopic alterations in membrane folding are not intrinsic to the osmosensing/regulation mechanism of OpuALl. At the molecular
level, physical properties such as membrane
fluidity, bilayer thickness, hydration state of
lipid headgroups, interfacial polarity and charge,
and/or lateral pressure may vary with changes
in the osmolality of the medium. To define the
osmotic signaling process in more detail,

OpuALl has been incorporated into liposomes
of different lipid composition (van der Heide
et al., 2001). It was found that the fraction of
anionic (charged) lipids is of major importance
for the osmosensing mechanism, whereas variations in acyl chain length, degree of fatty acid
saturation, position of the cis/trans double bond,
and the fraction of non-bilayer lipids have
relatively minor effects. By varying the fraction
of anionic lipids (phosphatidylglycerol (PG) or
phosphatidylserine (PS)) from 6 to 13 to 25%,
OpuALl is converted from an inactive or tense (T)
to an intermediate (I) to an osmotically controllable (R) state (Figure 13.4). Moreover, at a
given mol% of PG, OpuALl can be converted
from R to I by adding cationic amphipaths and
from I to R with anionic amphipaths. This suggests that the overall charge of the headgroup


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

R
T

Inactive

“Low”
Cytoplasmic
Ionic Strength

I
R

Active

“High”

[Anionic Lipids]
[Cationic amphiphiles]

[Anionic amphiphiles]

Figure 13.4. Schematic presentation of the
factors affecting the osmotic activation of the
membrane-embedded OpuA complex. A low
fraction of anionic lipids converts OpuA from an
inactive (T) to an intermediate state (C). A further
increase of the anionic lipid fraction leads to an
osmotically controlled (R) state. At the level of the
membrane, the effect of anionic lipids can be
mimicked by the addition of anionic or cationic
membrane-active compounds. The effects of
cytoplasmic ionic osmolytes, shifting the system
from an inactive to an active state, is also depicted.

region of the membrane lipids determines the
activity (kinetic state) of the transporter (van
der Heide et al., 2001).
The change in volume/surface ratio of the
proteoliposomes upon osmotic upshift leads to
an increase in the concentration of intravesicular
osmolytes. By varying the intravesicular composition, it was observed that ionic osmolytes

rather than internal osmolality switch the system from an inactive to an active state. In other
words, the system can be activated at isoosmotic conditions simply by an increase in the
internal ionic strength. Because the effects of the
ions vary with the fraction of anionic lipids in
the membrane, and the threshold for osmotic
activation is lowered by cationic and raised by
anionic amphipaths, these experiments support
the notion that osmotic signaling most probably
occurs via the bilayer in which the protein complex is embedded (van der Heide et al., 2001).

NATURE OF THE ACTIVATING SIGNAL
Why would the cell use ionic strength rather
than intracellular osmolality (affecting protein
hydration) or a specific signaling molecule
(allosteric regulatory site on the protein)? When
the osmolality of the medium is raised, the initial
change in cytoplasmic water activity depends
on the elasticity of the cell wall. Contrary to

what is often thought, the cell wall of bacteria
is not rigid but actually quite elastic (Csonka
and Hanson, 1991; Doyle and Marquis, 1994).
Consequently, even at turgor pressures above
zero, the cytoplasmic volume decreases with
increasing external osmolality, and the ion
(osmolyte) concentrations increase accordingly.
The increase in ionic strength accompanying
the volume decrease is undesirable as too
high concentrations of electrolytes interfere with
macromolecular functioning in eubacteria as

well as in higher organisms (Yancey et al., 1982).
As best documented for E. coli (Higgins et al.,
1987; Record et al., 1998), eubacteria expel ionic
compounds if the electrolyte concentration
becomes too high and replace these molecules
with neutral osmolytes such as glycine betaine
to balance the cellular osmolality. The increase
in electrolyte concentration (or ionic strength)
upon a modest decrease in turgor pressure
would thus represent an excellent trigger
(‘osmotic signal’) for the activation of any
osmoregulated transporter for neutral compatible solutes such as OpuALl. Actually, it would
prevent the osmotic stress from turning into
‘electrolyte stress’. Why is the increase in intracellular osmolality less suitable as osmotic signal? In order to maintain a relatively constant
turgor pressure at different external osmolalities, the cell will have to switch on OpuALl and
take up glycine betaine with maximal activity at
different internal osmolalities. In other words,
the ability of (the majority of) microorganisms
to grow at maximal rate over a wide range of
osmolalities of the medium implies that cellular
processes function optimally over wide range of
intracellular osmolalities. Finally, the cell could
use the osmotic upshift-dependent change in
concentration of a specific molecule as signal,
but ionic strength seems a more general signal
for osmoresponsive systems, including signal
transduction pathways to control the expression
of osmoregulated genes.
Although further work is needed to elucidate
all the intricacies of the osmosensing mechanism of the ABC transporter for glycine betaine

in L. lactis, the in vitro studies with the purified
protein complex indicate that all the regulatory
properties observed in vivo are present in the
two polypeptides comprising OpuALl. The fact
that osmotic activation of OpuALl in proteoliposomes mimics the regulation in vivo may seem
surprising as the vesicles do not withstand
turgor, whereas the turgor pressure in cells is
several atmospheres. However, there is mounting evidence that in bacteria turgor pressure

271


272

ABC PROTEINS: FROM BACTERIA TO MAN

exists only across the cell wall (and outer membrane in the case of Gram-negative bacteria) and
not across the cytoplasmic membrane (Cayley
et al., 2000). This implies that osmosensing
devices present in the cytoplasmic membrane
cannot repond to changes in turgor pressure;
rather they must sense the consequences of the
water influx or efflux, e.g. changes in ionic
strength. The consequences of osmotic shifts,
experienced by proteins in membrane model
systems such as proteoliposomses, may thus be
similar to those in the in vivo situation. Osmotic
regulation of OpuALl requires the protein complex to be embedded in a membrane bilayer
with the appropriate lipid composition (ϳ25%
anionic and some non-bilayer lipids) and the

appropriate concentration of small inorganic
electrolytes in the vesicle lumen. We cannot
entirely rule out the possibility that additional
proteinaceous factors influence the osmotic activation of OpuALl in vivo, as has been suggested
for the proton motive force-driven proline transporter ProP (Kunte et al., 1999), but we do not
have any indications or need for such a factor to
relate the in vitro and in vivo data.

CHALLENGES AND
PERSPECTIVES
From the studies presented in the previous section, it is evident that osmosensing and regulation, be it a transporter, channel or signaling
pathway, can only be fully understood if the
system can be analyzed not only in the intact
cell but also in artificial membrane systems
with defined components. The ABC transporter
for glycine betaine has some unique properties
that make the protein complex ideally suitable
for in vitro analyses. Firstly, the system has the
ligand-binding domain covalently linked to the
translocator moiety, which simplifies purification of the system and results in a high efficiency of the transport reaction without having
to use a large excess of (soluble) ligand-binding
protein. Secondly, the subunits of the OpuA
complex (OpuAA and OpuABC) remain associated in the detergent-solubilized state, which
greatly facilitates reconstitution in proteoliposmes. Thirdly, the system has a high catalytic
efficiency, which enables us to measure accurately activities even below 5% of Vmax.
The research on OpuA from L. lactis work
indicates that a change in intracellular ionic
strength serves as primary signal of osmotic

stress for this ABC transporter. We propose that

this signal is not sensed by the protein directly,
i.e. via changes in surface hydration of the protein or direct effects of the ions on the protein
(allosteric site); rather the membrane in which
the protein is embedded serves as a mediator.
Changes in cellular ionic strength are likely to
alter specific interactions between (ionic) lipids
and the protein, thereby affecting the transport
activity. The ATPase and translocation activity
of the OpuA transporter are strictly coupled,
and, at present, it is unclear whether the ATPase
(OpuAA) or the membrane-embedded part
(translocator) of OpuABC is the actual sensor.
A challenge will now be to assign specific
regions or residues in the OpuALl protein complex as sites that actually sense the changes in
membrane structure (e.g. electrostatic interactions between protein and phospholipid molecules), and to translate the in vitro observations
into proposals that can be experimentally tested
in vivo. Moreover, one may wish to study
osmotic signaling pathways that affect the transcription of genes in a manner analogous to the
studies described heretofore for the ABC transporter OpuALl. Past research has been restricted
to mutant selection and isolation, and construction of allelic strains defective in one or more
transcription factors. The proU regulatory circuit is clearly too complicated for a complete
understanding of its osmoregulatory mechanism at this moment, but it is possible to devise
protocols for the in vitro analysis of simpler
pathways, e.g. two-component regulatory systems. One would also like to know how the cell
responds to osmotic stress not only at the level
of activation of transcription and transporter
activity but also at the level of lipid synthesis.
The membrane bilayer composition is intrinsic
to the osmosensing mechanism of OpuALl
and most likely other transporters as well

(Rübenhagen et al., 2000; Wood et al., 2001),
and changes therein will on the longer time
scales be important for volume control of the
cell. Preliminary studies have been reported on
the changes in fatty acid composition of the
membrane of L. lactis in relation to osmotic
stress (Guillot et al., 2000). Unfortunately, no
information is available on variations in headgroup composition, e.g. fractions of anionic
lipids, as these factors seem far more important
for the osmoregulation of OpuALl than acyl
chain length, degree of saturation, etc. Finally,
the relationship between osmotic and other
stresses needs to be evaluated more thoroughly
at the level of transporter activity (in vitro and


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

in vivo) and cellular glycine betaine accumulation levels.
Note added in proof: Recent analysis of
genomic databases has indicated that some
homologues of OpuA of L. lactis have two
substrate-binding domains fused in tandem to
the translocator moiety of the ABC transporter.
These systems thus have four extracellular
substrate-binding sites per functional complex.
For a full description of the newly discovered
chimeric substrate-binding/translocator proteins, one is referred to van der Heide and
Poolman (2002) EMBO reports, in press.

ACKNOWLEDGMENTS
This work was supported by grants from the
Netherlands Organization for Scientific Research
(NWO) under auspices of the Netherlands
Foundation of Life Sciences. We thank Erhard
Bremer for sharing information prior to
publication.

REFERENCES
Bartolucci, C., Perola, E., Pilger, C., Fels, G.
and Lamba, D. (2001) Three-dimensional
structure of a complex of galanthamine with
acetylcholinesterase from Torpedo californica:
implications for the design of new antiAlzheimer drugs. Proteins 42, 182–191.
Booth, I.R. (1992) Regulation of gene expression
during osmoregulation: the role of potassium
glutamate as a secondary signal of osmotic
stress. In: Alkali Cation Transport Systems in
Prokaryotes (ed. E.P. Bakker), pp. 205–224,
Boca Raton, FL: CRC Press.
Bourot, S., Sire, O., Trautwetter, A., Touze, T.,
Wu, L.F., Blanco, C. and Bernard, T. (2000)
Glycine betaine-assisted protein folding in a
lysA mutant of Escherichia coli. J. Biol. Chem.
275, 1050–1056.
Breed, J., Kneip, S., Gade, J., Welte, W. and
Bremer, E. (2001) Purification, crystallization
and preliminary crystallographic analysis of
the periplasmic binding protein ProX from
Escherichia coli. Acta Cryst. D57, 448–450.

Bremer, E. (2002) Adaptation to changing osmolarity. In: Bacillus subtilis and its Closest
Relatives: From Genes to Cells (ed. A.L.
Sonenshein et al.), Washington, DC: ASM
Press.

Cayley, D.S., Guttman, H.J. and Record, M.T. Jr.
(2000) Biophysical characterization of changes
in amounts and activity of Escherichia coli cell
and compartment water and turgor pressure
in response to osmotic stress. Biophys. J. 78,
1748–1764.
Csonka, L.N. (1989) Physiological and genetic
responses of bacteria to osmotic stress.
Microbiol. Rev. 53, 121–147.
Csonka, L.N. and Hanson, A.D. (1991) Prokaryotic osmoregulation: genetics and physiology. Annu. Rev. Microbiol. 45, 569–606.
Csonka, L.N., Ikeda, T.P., Fletcher, S.A. and
Kustu, S. (1994) The accumulation of glutamate is necessary for optimal growth of
Salmonella typhimurium in media of high
osmolarity but not induction of the proU
operon. J. Bacteriol. 176, 6324–6333.
Dassa, E. and Bouige, P. (2001) The ABC of
ABCs: a phylogenetic and functional classification of ABC systems in living organisms.
Res. Microbiol. 152, 211–229.
Detmers, F.J.M., Lanfermeijer, F.C., Abele, R.,
Tampé, R., Konings, W.N. and Poolman, B.
(2000) Combinatorial peptide libraries reveal
the ligand binding properties of the oligopeptide receptor OppA of Lactococcus lactis. Proc.
Natl Acad. Sci. USA, 97, 12487–12492.
Diederichs, K., Diez, J., Greller, G., Müller, C.,
Breed, J., Schnell, C., Vonrhein, C., Boos,

W. and Welte, W. (2000) Crystal structure of
MalK, the ATPase subunit of the trehalose/
maltose ABC transporter of the archaeon
Thermococcus litoralis. EMBO J. 19,
5951–5961.
Doyle, R.J. and Marquis, R.E. (1994) Elastic,
flexible peptidoglycan and bacterial cell
wall properties. Trends Microbiol. 2, 57–60.
Ehrmann, M., Ehrle, R., Hofmann, E., Boos,
W. and Schlösser, A. (1998) The ABC maltose transporter. Mol. Microbiol. 29, 685–694.
Faatz, E., Middendorf, A. and Bremer, E. (1988)
Cloned structural genes for the osmotically
regulated binding-protein-dependent glycine
betaine transport system (ProU) of Escherichia
coli K-12. Mol. Microbiol. 2, 265–279.
Galinski, E.A. and Trüper, H.G. (1994)
Microbial behaviour in salt-stressed ecosystems. FEMS Microbiol. Rev. 15, 95–108.
Glaasker, E., Konings, W.N. and Poolman, B.
(1996a) Osmotic regulation of intracellular
solute pools in Lactobacillus plantarum. J.
Bacteriol. 178, 575–582.
Glaasker, E., Konings, W.N. and Poolman, B.
(1996b) Glycine-betaine fluxes in Lactobacillus plantarum during osmostasis and

273


274

ABC PROTEINS: FROM BACTERIA TO MAN

hyper- and hypoosmotic shock. J. Biol. Chem.
271, 10060–10065.
Glaasker, E., Heuberger, E.H.M.L., Konings,
W.N. and Poolman, B. (1998) Mechanism
of osmotic activation of the quaternary
ammonium compound transporter (QacT)
of Lactobacillus plantarum. J. Bacteriol. 180,
5540–5546.
Guillot, A., Obis, D. and Mistou, M.-Y. (2000)
Fatty acid membrane composition and
activation of glycine betaine transport in
Lactococcus lactis subjected to osmotic stress.
Int. J. Food Microbiol. 55, 47–51.
Haardt, M. and Bremer, E. (1996) Use of phoA
and lacZ fusions to study the membrane
topology of ProW, a component of the
osmoregulated ProU transport system of
Escherichia coli. J. Bacteriol. 178, 5370–5381.
Higgins, C.F., Cairney, J., Stirling, D.A.,
Sutherland, L. and Booth, I.R. (1987)
Osmotic regulation of gene expression: ionic
strength as an intracellular signal? Trends
Biochem. Sci. 12, 339–344.
Jovanovich, S.B., Record, M.T. and Burgess,
R.R. (1989) In an Escherichia coli coupled
transcription-translation system, expression
of the osmoregulated gene proU is stimulated at elevated potassium concentrations
and by an extract from cells grown at high
osmolality. J. Biol. Chem. 264, 7821–7825.

Kappes, R.M., Kempf, B. and Bremer, E. (1996)
Three transport systems for the osmoprotectant glycine betaine operate in Bacillus subtilis:
characterization of OpuD. J. Bacteriol. 178,
5071–5079.
Kempf, B. and Bremer, E. (1995) OpuA, an
osmotically regulated binding proteindependent transport system for the osmoprotectant glycine betaine in Bacillus subtilis.
J. Biol. Chem. 270, 16701–16713.
Kempf, B. and Bremer, E. (1998) Uptake and
synthesis of compatible solutes as microbial
stress responses to high-osmolarity environments. Arch. Microbiol. 170, 319–330.
Kempf, B., Gade, J. and Bremer, E. (1997)
Lipoprotein from the osmoregulated ABC
transport system OpuA of Bacillus subtilis:
purification of the glycine betaine binding
protein and characterization of a functional
lipidless mutant. J. Bacteriol. 179, 6213–6220.
Kunte, H.J., Crane, R.A., Culham, D.E.,
Richmond, D. and Wood, J.M. (1999)
Protein ProQ influences osmotic activation
of compatible solute transporter ProP in
Escherichia coli K-12. J. Bacteriol. 181,
1537–1543.

Lamark, T., Kaasen, I., Eshoo, M.W.,
Falkenberg, P., McDougall, J. and Ström,
A.R. (1991) DNA sequences and analysis of
the bet genes encoding the osmoregulatory
choline-glycine betaine pathway of Escherichia coli. Mol. Microbiol. 5, 1049–1064.
Landis, L., Xu, J. and Johnson, R.C. (1999) The
cAMP receptor protein CRP can function as

an osmoregulator of transcription in Escherichia coli. Genes Dev. 13, 3081–3091.
Lanfermeijer, F., Picon, A., Konings, W.N. and
Poolman, B. (1999) Kinetics and consequences of binding of nona- and dodecapeptides to the oligopeptide binding protein,
OppA, of Lactococcus lactis. Biochemistry 38,
14440–14450.
Lanfermeijer, F.J., Detmers, F.J.M., Konings,
W.N. and Poolman, B. (2000) On the binding mechanism of the peptide receptor of the
oligopeptide transport system of Lactococcus
lactis. EMBO J. 19, 3649–3656.
Liu, C.E., Liu, P.-Q., Wolf, A., Lin, E. and
Ames, G.F.-L. (1999) Both lobes of the soluble receptor of the periplasmic histidine permease, an ABC transporter (traffic ATPase),
interact with the membrane bound complex.
J. Biol. Chem. 274, 739–747.
Lucht, J.M. and Bremer, E. (1994) Adaptation
of Escherichia coli to high osmolarity environments: osmoregulation of the high-affinity
glycine betaine transport system ProU.
FEMS Microbiol. Rev. 14, 3–20.
Martins, L.O. and Santos, H. (1995) Accumulation of mannosylglycerate and di-myoinositol-phosphate by Pyrococcus furiosus in
response to salinity and temperature. Appl.
Environ. Microbiol. 61, 3299–3303.
Martins, L.O., Carreto, L.S., Da Costa, M.S. and
Santos, H. (1996) New compatible solutes
related to di-myo-inositol-phosphate in members of the order Thermotogales. J. Bacteriol.
178, 5644–5651.
McLaggan, D., Naprstek, J., Buurman, E.T. and
Epstein, W. (1994) Interdependence of Kϩ
and glutamate accumulation during osmotic
adaptation of Escherichia coli. J. Biol. Chem.
269, 1911–1917.
Obis, D., Guillot, A. and Mistou, M.-Y. (2000)

Abstract. Comp. Biochem. Physiol. A 126,
S105.
Poolman, B. and Glaasker, E. (1998) Regulation of compatible solute accumulation in
bacteria. Mol. Microbiol. 29, 397–407.
Quiocho, F.A. and Ledvina, P.S. (1996) Atomic
structure and specificity of bacterial periplasmic receptors for active transport and


STRUCTURE AND FUNCTION OF OSMOREGULATED ABC TRANSPORTERS

chemotaxis: variation of common themes.
Mol. Microbiol. 20, 17–25.
Rajkumari, K., Kusano, S., Ishihama, A.,
Mizuno, T. and Gowrishankar, J. (1996)
Effect of H-NS and potassium glutamate on
␴S- and ␴70-directed transcription in vitro
from osmotically regulated P1 and P2 promoters of proU in Escherichia coli. J. Bacteriol.
178, 4176–4181.
Ramirez, R.M., Prince, W.S., Bremer, E. and
Villarjo, M. (1989) In vitro reconstitution of
osmoregulated expression of proU of
Escherichia coli. Proc. Natl Acad. Sci. USA 86,
1153–1157.
Record, M.T. Jr, Courtenay, E.S., Cayley, D.S.
and Guttman, H.J. (1998) Biophysical compensation mechanisms buffering E. coli
protein-nucleic acid interactions against
changing environments. Trends Biochem.
Sci. 23, 190–194.
Rübenhagen, R., Rönsch, H., Jung, H., Krämer,
R. and Morbach, S. (2000) Osmosensor and

osmoregulator properties of the betaine carrier BetP from Corynebacterium glutamicum in
proteoliposomes. J. Biol. Chem. 275, 735–741.
Sankaran, K. and Wu, H.C. (1994) Lipid modification of bacterial prolipoprotein: transfer
of diacylglyceryl moiety from phosphatidylglycerol. J. Biol. Chem. 269, 19701–19706.
Sukharev, S.I., Blount, P., Martinac, B. and
Kung, C. (1997) Mechanosensitive channels
of Escherichia coli: the MscL gene, protein,
and activities. Annu. Rev. Physiol. 59,
633–657.
Sutrina, S.L., Reddy, P., Saier, M.H. and
Reizer, J. (1990) The glucose permease of
Bacillus subtilis is a single polypeptide chain

that functions to energize the sucrose permease. J. Biol. Chem. 265, 18581–18589.
van der Heide, T. and Poolman, B. (2000a)
Glycine betaine transport in Lactococcus
lactis is osmotically regulated at the level of
expression and translocation. J. Bacteriol.
182, 203–206.
van der Heide, T. and Poolman, B. (2000b)
Osmoregulated ABC-transport system of
Lactococcus lactis senses water stress via
changes in the physical state of the membrane. Proc. Natl Acad. Sci. USA 97,
7102–7106.
van der Heide, T., Stuart, M.C.A. and
Poolman, B. (2001) On the osmotic signal
and osmosensing mechanism of an ABC
transport system for glycine betaine. EMBO
J. 20, 7022–7032.
Verheul, A., Glaasker, E., Poolman, B. and

Abee, T. (1997) Betaine and L-carnitine
transport in response to osmotic signals in
Listeria monocytogenes Scott A. J. Bacteriol.
179, 6979–6985.
Wood, J.M. (1999) Osmosensing by bacteria:
signals and membrane-based sensors.
Microbiol. Mol. Biol. Rev. 63, 230–262.
Wood, J.M., Bremer, E., Csonka, L.N., Krämer,
R., Poolman, B., van der Heide, T. and
Smith, L. (2001) Osmosensing and osmoregulatory compatible solute accumulation by
bacteria. Comp. Biochem. Physiol. A. Mol.
Integr. Physiol. 130, 437–460.
Yancey, P.H., Clark, M.E., Hand, S.C., Bowlus,
R.D. and Somero, G.N. (1982) Living with
water stress: evolution of osmolyte systems.
Science 217, 1214–1222.

275