TheWood Lab

  Research:  Osmosensing, Osmoregulation and Osmotic Adaptation by Escherichia coli

David Goodsell's representation of the cell wall and the crowded interior of E. coli (D.S. Goodsell (1997) The Machinery of Life)... plus ProP!  

 Cells survive and grow in diverse and changing environments.  For Escherichia coli, these environments include food and water as well as human and animal tissues.  For plant cells they include soils with high or fluctuating salinities.  For kidney cells they include dilute or concentrated urine!  We would like to understand how cells sense and respond to changes in the osmotic pressure or osmolality of their environment.  Osmoregulatory mechanisms ensure that cells remain appropriately hydrated - and hydration is critical for life (see Wood 1999; Wood et al. 2001, Poolman et al. 2004, Wood 2006, Wood 2007, Altendorf et al. 2008, Romantsov & Wood 2008)!

We study osmoregulation by E. coli because advanced research tools drawn from microbiology, genetics, molecular biology, biochemistry and biophysics can readily be applied to E. coli.  In addition, the capacity to osmoregulate may assist E. coli to cause infections in humans and animals (Culham et al. 2001).   We continue to identify genes, proteins and osmolytes that confer osmotic, urea, thermal and oxidative stress tolerance on E. coli and other organisms (Ly et al. 2006).  But we would also like to know how cells sense and respond to osmotic stress.

We were the first to demonstrate that a particular protein, ProP, can act as an osmosensor (Racher et al. 1999).  ProP is integral to the cytoplasmic membrane of E. coli.  It senses changes in osmotic pressure and responds by changing cytoplasmic composition, restoring hydration. Osmotic stress causes many changes in living cells.  We aim to understand which change(s) are sensed by ProP, and how ProP responds to those changes.  We would also like to know how cytoplasmic RNA-binding protein ProQ controls the level of transporter ProP in cells.
 

ProP is H+-osmoprotectant symporter and a member of the major facilitator superfamily (MFS).  The first crystal structures for members of the MFS were published in 2003, enabling Bob Keates (University of Guelph) to build our structural model for ProP (Wood et al. 2005).  

This model represents a snapshot of ProP as it releases its organic substrate (e.g. proline or glycine betaine) and a proton to the cytoplasm.  The two six-helix bundles of ProP would then re-orient around an axis in the membrane plane (perpendicular to your screen), preparing to bind a substrate and a proton from the periplasm.  Transport would occur as this "alternating access" process was repeated. 

We are testing this structural model in collaboration with the lab of Dr. Joan Boggs (Hospital for Sick Children, Toronto).   To do this we explore the chemical reactivities of cysteine (Cys) residues introduced to a fully functional, Cys-less ProP variant, the cross-linking of those residues in ProP dimers, and the impact of their chemical modification on ProP function. Such studies have validated our structural model (Wood 2005, Liu 2007), shown that ProP is a dimer in vivo (Hillar 2005, Hillar 2003) and revealed that the conformation of Loop P1, connecting transmembrane helices I and II, is osmolality-sensitive (Culham et al., In Press).

A structural model for the cytplasm-facing conformation of ProP.  Its central, cytoplasmic loop and extended C-terminal domain are not shown.
 

ProP activity increases with osmolality in cells and in proteoliposomes reconstituted with pure ProP-His6.  Black:  ProP, Red:  ProP*

ProP activity increases with osmolality while most other membrane-based processes (e.g. respiration, other transport) are inhibited (Culham et al. 2008).

But how can transport be controlled by osmotic pressure?  Osmotic pressure changes affect many properties of living cells (Wood 1999).  We think ProP is a direct osmosensor that responds to changes in water activity and/or the crowding of macromolecules in the cytoplasm (Culham et al. 2003).  In proteoliposomes, ProP activity is osmolality-sensitive only if the membrane potential (DY) exceeds about 100 mV (negative inside). Above that voltage threshold, water acts as a ProP inhibitor!

In collaboration with Dr. Boggs' lab, we're now using a fully functional, His-tagged, Cys-less ProP variant, ProP*, as a platform to explore the structural changes that accompany osmosensing by ProP (e.g. Wood et al. 2005, Culham et al. In Press). 

But there's more.  We've found two groups of ProP orthologues (Poolman et al. 2004).  Members of the group including E. coli ProP have an extended, cytoplasmic carboxyl terminus that can form a homodimeric, antiparallel, a-helical coiled-coil (Culham et al. 2004, Hillar et al. 2005, Tsatskis et al. 2008).  Our collaborator Dr. Robert Hodges (University of Colorado Health Sciences Center) studies this structure as a prototypical, free-standing antiparallel coiled-coil.  The ProP coiled-coil is stabilized by interactions of R488 on one strand with D475 and D478 on the other (Zoetewey et al. 2004, Tsatskis et al. 2008). This structure helps tune ProP so that the range of its osmolality response coincides with that of the bacterial growth medium (Culham et al. 2004, Tsatskis et al. 2008).

NMR structure of the antiparallel coiled-coil formed by a peptide replica of ProP residues 468-497 (Zoetewey et al. 2003). 

 

Cardiolipin (also known as diphosphatidyl glycerol) is concentrated at the poles and near the septa of E. coli cells.   This can be seen by microscopy of bacteria stained with NAO, a fluorescent acridine orange derivative.  We studied the subcellular localization of ProP by staining cells with the biarsenical fluorescein derivative FlAsH.  The C-terminal domain plays a role in concentrating ProP molecules with cardiolipin at the poles and near the septa of E. coli cells (Romantsov et al. 2007, Romantsov et al. 2008). This cardiolipin-rich environment raises the osmolality at which ProP activates. 

 

We're now examining the relationships among ProP localization, ProP dimerization and coiled-coil interactions.

ProP colocalizes with cardiolipin at the poles and near the septa of E. coli cells (Romantsov et al. 2007, Romantsov et al. 2008).
 

Structural models for the N- and C-terminal domains of ProP amplifier protein, ProQ (Smith et al. 2004, Smith 2007)

 ..and still more.  ProP can sense and respond to osmolality on its own, in proteoliposomes (Racher et al. 1999), but mutations at proQ attenuate ProP activity in cells (Kunte et al. 1999). Mutations in proQ alter ProP levels under particular conditions.  ProQ is a small, cytoplasmic protein.  Its N- and C-terminal domains can be expressed separately, and the properties of the resulting proteins reflect structures modeled on the crystal structures of RNA binding proteins (Smith 2004, Smith 2007).  We once thought that ProQ would interact directly with ProP to amplify its activity…now we think that ProQ may be an RNA-binding translational regulator!  Stay tuned…
 

...for more information contact Janet Wood.

  

Updated November 2008