The chemotaxis network of E. coli

Transcription

1 The chemotaxis network of E. coli Ned Wingreen Boulder Summer School 2007 Thanks to: Robert Endres, Clint Hansen, Juan Keymer, Yigal Meir, Mica Skoge, and Victor Sourjik Support from HFSP

2 Adaptati Adaptati uses methylati to adjust Δf total 0, and thereby enhances sensitivity. Tar Δf > 0 Δf > 0 Tsr Δf > 0 Δf Off total 0

3 Scaling of wild-type adapted respse Sourjik and Berg: Δ[MeAsp] ΔFRET{Tar(QEQE)} Free energy scaling: Δ[MeAsp] Δ(F F ) Sourjik and Berg (2002) ΔFRET for Tar(QEQE) strain Doesn t collapse zero-ambient data. Includes zero-ambient data! And yields K D s: K D = 25 µm, K D 0.5 mm

4 Motor output also yields K D s + + Δ D D K C K C f t / 1 / 1 ~ log ~ Berg and Tedesco (1975) K D = 27 µm, K D 0.9 mm

5 2-state receptor model CheB Originally proposed by Asakura and Hda (1984). Modified by Barkai and Leibler (1998) to explain precise and robust adaptati: Receptor complex has 2 states:, i.e. inactive as kinase, and, i.e. active as kinase. Demethylati ly occurs in state, d Methylati = a[cher]-b[cheb] P dt Therefore, at steady state, P = a [CheR] / b[cheb] Which implies precise and robust adaptati of each receptor complex to a fixed activity.

6 Failure of precise adaptati? Barkai-Leibler single-receptor adaptati model: R CheB B d Methylati = dt Steady state a[cher]-b[cheb] P Activity P = a [CheR] / b[cheb] Yields imprecise adaptati of receptor clusters:

7 Help from assistance neighborhoods Antommattei et al. (2004) Li and Hazelbauer (2005) Tethered CheR/CheB act neighborhood of 5-7 receptors.

8 Precise adaptati saved! Assistanceneighborhood model Barkai-Leibler + assistance neighborhoods = precise adaptati:

9 Precisi of adaptati with assistance neighborhoods Assistance neighborhood of ~ 6 receptors sufficient for precise adaptati: Adaptati error: A([ L]) A(0) A(0)

10 Initial respse and sensitivity of adapted receptors Experiment Sourjik and Berg (2002) Simulati (with assistance neighborhoods)

11 Two peaks of sensitivity ΔA / A [ ΔL]/[ L] d(log A) d(log[ L]) FRET Simulati Tar Tsr K a K a K s K s Analytic result for single cluster da d(log[ L]) = da df df d(log[ L])

12 Predicti: Two limits of adaptati Ser Berg and Brown (1972) Tsr:Tar=2:1 fully methylated Asp Serine Aspartate Full methylati before saturati adaptati stops Free Energy On Off Off On Saturati by ligand no respse for [L] > K D Serine [L] Aspartate [L]

13 Open questis What determines cluster size and what is the mechanism of receptor-receptor coupling? Two limits of adaptati? What is being optimized? Stock (2000)

14 Cclusis E. coli chemotaxis network remarkable for: precise and robust adaptati signal integrati sensitivity FRET studies reveal two regimes of receptor activity Model of mixed clusters of 2-state receptors accounts for network properties and for two regimes Precise adaptati of clusters requires assistance neighborhoods Predicti: two possible limits of adaptati

15 Outline Introducti to chemotaxis in E. coli The chemotaxis network Two regimes of activity Receptors functi collectively Modeling Mixed clusters of receptors Precise adaptati through assistance neighborhoods

16 E. coli chemotaxis: runs and tumbles (Thanks to Howard Berg.)

17 The chemotaxis network labs/bacteria/projects_fret.html

18 Chemoreceptor clustering Receptors are clustered globally into a large array, and locally into trimers of dimers. Gestwicki et al. (2000) Kim et al. (1999); Studdert and Parkins (2004)

19 Chemoreceptors Tar - aspartate, glutamate (~900 copies) Tsr - serine (~1600) Sensor Trg - ribose, galactose (~150) Tap - dipeptides (~150) (Aer - oxygen via FAD (150?)) Linker regi Homodimer Transmembrane helices Attractant binding inhibits phosphorylati of CheA Adaptati: More attractant increased methylati by CheR faster phosphorylati of CheA 380 A Cytoplasmic domain Methyl binding sites CheB, CheR Less attractant increased demethylati by CheB slower phosphorylati of CheA Stock (2000) CheA / CheW binding regi

20 In vivo FRET studies of receptor activity Sourjik and Berg (2002) Real-time measurement of rate of phosphorylati of CheY. (FRET also allows subcellular imaging, Vaknin and Berg (2004).)

21 FRET data: two regimes of activity Sourjik and Berg (2002) Regime I: Activity moderate to low at zero ambient MeAsp (0.06,1) K i small and almost cstant Regime I Regime II Regime II: Activity high (saturated) at zero ambient MeAsp ( ) K i1 large and increasing with methylati Plateau in activity K i2 approximately cstant Two regimes of receptor activity csistent with 2-state receptor model.

22 Two regimes of a 2-state receptor On Off But first, a 1-state receptor: Off On Off Regime I Off Ligand Free Energy On Off Regime II Off On K D Ligand Ligand ε e However, Activity = Psingle = receptor does not ε ε C e + e + e K Regime I: Activity low to very low at zero ligand ccentrati K i = K D account for low apparent K i in Regime I. Regime II: Activity high (saturated) 1 at zero Pno ligand bound ccentrati = C K i increasing as 1+ ε D ε K D

23 Receptor-receptor coupling Duke and Bray (1999) Duke and Bray (1999) proposed that receptorreceptor coupling could enhance sensitivity to ligands. MWC model: if N receptors are all or all together, Activity = P =, Δε = ε ε Regime I (Δε > 0): 1+ e NΔε ( 1 1+ Receptor-receptor coupling gives enhanced sensitivity (low K i ) in Regime I, and enhanced cooperativity (high Hill coefficient ) in Regime II. K C D ) N Low activity ~ e - NΔε at zero ligand ccentrati K i = K D / N Hill coefficient = 1 Regime II (Δε < 0): K i = K D e - Δε Hill coefficient = N

24 Mixed cluster MWC model Mello and Tu (2005) Keymer et al. (2006) Tar Tsr Mixed clusters of size Each cluster is an independent 2-state system. Regime I: K i = K D / N. Regime II: Plateaus: some clusters, some. Hill coefficient 1.

25 Receptor homogeneity and cooperativity More Tars Receptors are in Regime II: Hill coefficient increases with Tar homogeneity because more receptors bind ligand at transiti. K i (or K i1 ) decreases with Tar homogeneity because fewer Tsrs need to be switched.