Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Feb 5;104(3):613-21.
doi: 10.1016/j.bpj.2012.12.002.

Defining the membrane-associated state of the PTEN tumor suppressor protein

Affiliations

Defining the membrane-associated state of the PTEN tumor suppressor protein

Craig N Lumb et al. Biophys J. .

Abstract

Phosphatase and tensin-homolog deleted on chromosome 10 (PTEN) is a tumor-suppressor protein that regulates phosphatidylinositol 3-kinase (PI3-K) signaling by binding to the plasma membrane and hydrolyzing the 3' phosphate from phosphatidylinositol (3,4,5)-trisphosphate (PI(3,4,5)P3) to form phosphatidylinositol (4,5)-bisphosphate (PI(4,5)P2). Several loss-of-function mutations in PTEN that impair lipid phosphatase activity and membrane binding are oncogenic, leading to the development of a variety of cancers, but information about the membrane-associated state of PTEN remains sparse. We have modeled a membrane-associated state of the truncated PTEN structure bound to PI(3,4,5)P3 via multiscale molecular dynamics simulations. We show that the location of the membrane-binding surface agrees with experimental observations and is robust to changes in lipid composition. The level of membrane interaction is substantially reduced in the phosphatase domain for the triple mutant R161E/K163E/K164E, in line with experimental results. We observe clustering of anionic lipids around the C2 domain in preference to the phosphatase domain, suggesting that the C2 domain is involved in nonspecific interactions with negatively charged lipid headgroups. Finally, our simulations suggest that the oncogenicity of the R335L mutation may be due to a reduction in the interaction of the mutant PTEN with anionic lipids.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Structure of PTEN. (A) Domain structure of PTEN. (B) The crystal structure of PTEN, with the phosphatase domain (PD; residues 14–185) in pink and the C2 domain (residues 186–351) in cyan. The tartrate molecule bound in the phosphatase active site is shown as van der Waals spheres. The loop missing from the crystal structure between P281 and K313 is shown as a dotted line. (C) Coarse-grained (CG) representation of PTEN. The bound tartrate has been replaced with a CG representation of PI(3,4,5)P3, with the PI(3,4,5)P3 molecule positioned as described in the main text.
Figure 2
Figure 2
Self-assembly and simulation of the PTEN-membrane complex, outlining the serial multiscale MD workflow. (A) Initial simulation setup. Lipids are randomly distributed throughout the simulation box. Lipid phosphate headgroups are shown as transparent orange spheres, while lipid tails, water, and ions are omitted for clarity. (B) Simulation snapshot after 0.5 μs CG simulation. Bilayer self-assembly and PTEN-membrane complex formation occurs within the first few tens of nanoseconds of simulation in all cases. (C) The final frame of the CG simulation shown in panel B is converted to an atomistic representation, which is then simulated for a further 50 ns.
Figure 3
Figure 3
Membrane binding of PTEN. Protein-lipid heavy atom contacts within 4 Å from the atomistic simulations for the wild-type PTEN protein and (A) a zwitterionic bilayer, and (B) a bilayer containing 20% anionic lipids. The region containing residues R161, K163, and K164 is expanded in each case. (C and D) Respective contacts mapped onto the PTEN structure, illustrating the location of the interaction surface with the lipid bilayer. The region missing from the crystal structure is shown as a dotted line, and is seen to be distant from the lipid-binding interface. Note that panels A and C are annotated to indicate regions (a–e) of the sequence that interact most strongly with the lipid bilayer. The main peaks occur in approximately the same residues for both the zwitterionic and the 20% anionic bilayer. Thus, the largest peaks are for (a) D22, (b) R41 (R47 for the 20% anionic bilayer), (c) R161, (d) K263, and (e) R335.
Figure 4
Figure 4
Membrane binding of a mutant PTEN. The electrostatic potential around (A) the wild-type protein and (B) the R161E/K163E/K164E mutant protein are shown, calculated using the Adaptive Poisson-Boltzmann Solver (38). The location of the cationic patch formed by R161, K163, and K164 is indicated by the dotted line. (C) Protein-lipid heavy atom contacts within 4 Å from the atomistic simulations for the R161E/K163E/K164E mutant protein and a bilayer containing 20% anionic lipids. The region containing E161, E163, and E164 is highlighted to facilitate comparison with Fig. 3. Electrostatic potential isocontours are shown at contour values of −1 kT/e (red) to +1 kT/e (blue).
Figure 5
Figure 5
Anionic lipid association with membrane bound PTEN. The results of analysis of the frequency of occurrence of anionic (PS) lipid headgroups in the bilayer (xy) plane adjacent to PTEN based on simulations of (A) the wild-type and (B) the R161E/K163E/K164E mutant proteins. The analysis is derived from the CG MD simulations in the presence of 20% PS in the bilayer. The color contours show the surface density of lipid headgroups. For 20% PS in an equilibrated bilayer with even distribution of lipids one would expect a surface density of ∼0.003 Å−2 assuming the area per lipid for POPS is 55 Å−2 (54). This would correspond to yellow/green contours in the density maps. The black trace shows the position of PTEN, used as a reference frame in this analysis, with the PD and C2 domains labeled.
Figure 6
Figure 6
Potential of mean force (i.e., free energy profile) along a reaction coordinate defined by the distance between the PTEN and PI(3,4,5)P3 centers of mass along the bilayer normal (z). This PMF was evaluated using CG MD simulations and the bilayer containing 20% PS. The PMF was calculated over the intervals 400–500 ns for each window on z: this 100-ns interval was divided into ten segments each of 10-ns duration (400–410, 410–420 ns, etc.) and the PMF evaluated for each segment. The resultant ten PMF profiles were used to calculate a mean profile (red line) and standard deviations (gray bars), as shown. The windows at 4 Å, 10 Å, and 18 Å are labeled along the PMF by u, v, and w, respectively, and snapshots from the final frame of each of these windows are shown on the right of the figure.
Figure 7
Figure 7
Locations of membrane-interacting clinically important mutations. Most oncogenic mutants occur at the PD:C2 domain interface or at the phosphatase active site. A selection of these is shown as yellow van der Waals spheres. However several mutants occur at sites spatially separated from these locations. In the PD (pink), N48K is shown in red and labeled. In the C2 domain (cyan), R234Q and R335L are also shown in red and labeled. The lipid bilayer is shown as a translucent white surface.

Similar articles

Cited by

References

    1. Lemmon M.A. Membrane recognition by phospholipid-binding domains. Nat. Rev. Mol. Cell Biol. 2008;9:99–111. - PubMed
    1. McLaughlin S. The electrostatic properties of membranes. Annu. Rev. Biophys. Biophys. Chem. 1989;18:113–136. - PubMed
    1. Lumb C.N., Sansom M.S.P. Finding a needle in a haystack: the role of electrostatics in target lipid recognition by PH domains. PLoS Comp. Biol. 2012;8:e1002617. - PMC - PubMed
    1. McLaughlin S., Wang J., Murray D. PIP2 and proteins: interactions, organization, and information flow. Annu. Rev. Biophys. Biomol. Struct. 2002;31:151–175. - PubMed
    1. McLaughlin S., Murray D. Plasma membrane phosphoinositide organization by protein electrostatics. Nature. 2005;438:605–611. - PubMed

Publication types