Unformatted text preview:

Systems Biology 7.81/8.591/9.531 Problem Set 2 Assigned: 10.06.04 Due in class Due: 10.19.04 1. Biochemistry of the chemotaxis network. LT Lk+ + = L L kK fL T2P T3PT2 T3LT2P LT3PLT2 LT3T2PT3P T2T3ka k -a kck -ck -d kb k -b1-fLfL The E. coli chemotaxis network is represented here in simplified form. T represents the Tarreceptor, and L the ligand or attractant. The receptor can by modified by phosphorylation(subscript P) or methylation (subscripts 2 or 3).The Tar receptor binds an extracellular ligand L according to , with . Calculate the fraction of bound receptor. tot tottot totkd Effective model Full model (10) a. L + T kL − kL − (10) b. We now assume that the ligand binding reaction is in rapid equilibrium, and only consider total amounts of each modified form of receptor. For example, T3Ptot = T3P + LT3P , etc. How would you calculate the effective rate constants ka,...,kd and k-a,...,k-d between these total concentration pools in terms of the rate constants of the original methylation /demethylation and phosphorylation/dephosphorylation reactions? 17Systems Biology 7.81/8.591/9.531 (10) c. The assumed rates in the models of Spiro et al. and Barkai et al. are shown in the table below. Write down explicitly the effective rate constants for each model, in terms of the symbols listed in the table. Spiro model Barkai model L-unbound L-bound L-unbound L-bound ka k8 0 ka 0 0 k-a ky ky k-a k0 k0 kb 3 k8 1.1 k8 kb kp1 kp2 k-b ky ky k-b k-p1 k-p2 kc k1 k3 kc k k k-c k-1 k-1 k-c 0 0 kd k1 k3 kd 0 0 k-d k-1 k-1 k-d km km Note that some effective rate constants are now functions of L. This is appropriate, since we know for example that the receptors should become less phosphorylated as L increases. i) In the Spiro model, do ka/k-a and kb/k-b increase or decrease with L? ii) In the Barkai model, we would like kb to decrease and k-b to increase with L. What does this imply about kp1, kp2, k-p1 and k-p2? (10) d. Spiro model. In steady state, after the slow methylation reactions have had time to equilibrate, let α(L) represent the fraction of receptors that are methylated. Consider now the total concentration of phosphorylated and unphosphorylated receptors. Write out explicitly, in terms of α(L), the effective rates of phosphorylation (kp) and dephosphorylation (k-p) using kp = 1( −α)) ( ka +α) ( L −a k L −bL Lkb , k− p = 1( −α)) ( k +α) ( For perfect adaptation to be achieved, the phosphorylated fraction of receptor must be independent of L in steady state. You should have found above that k-p = ky; it is *therefore sufficient for perfect adaptation that kp = kp is a constant. *Set k8 = 15 s-1; KL = 1x106 M-1; and kp = 15 s-1. Plot ka and kb for L = 0, ..., 2 KL. On * *the same graph, draw a horizontal line showing the desired kp . Finally, set kp = kp in the equation above, solve for α(L), and plot this function. This is the magical form of α(L) required for perfect adaptation. The model of Spiro et al. is carefully “tuned” in order to achieve this result. We can contrast this situation with the Barkai model in part f, which is perfectly adapting but requires no fine tuning. 2Systems Biology 7.81/8.591/9.531 (10) e. Barkai model. Biochemical evidence suggests that the methylation reaction (whose rate constant was written as k in part c) operates at saturation with rate v. Show that under this assumption, the entire model reduces to the following reaction scheme: T k0 2 tot T2 topt v km’ km L) ( kb T3 tot totT3 pL) ( k−b Note that v is a constant rate (measured in M s-1) while km is a rate constant (measured in s-1). What is the value of km’ ? ( tot Write down the equation for T d 3 tot +T3 p ) and solve for T3ptot in steady state. Show dt that this value is independent of L if and only if km’ = 0. This is the essence of the Barkai model: perfect adaptation is easy to achieve, as long as only the phosphorylated receptors are demethylated by the CheB protein. 3Systems Biology 7.81/8.591/9.531 2. Adaptation and frequency response of the chemotaxis network. With a slight change of notation, the Barkai model of the chemotaxis network (see Problem 1f) can be represented as v k+ ) ( L C *C ) ( Lk− km’ km *Here, v represents the rate of creation of C, the unphosphorylated receptor; C is the phosphorylated or active form of the receptor, the actual signal which induces bacterial tumbling; km and km’ are the rate constants of demethylation reactions; and finally, k+ and k-are rate constants that represent the effect of ligand binding on the phosphorylation state of the receptor. Set α+ =∂k+< 0 , and α− =∂k−> 0 . This ensures that a sudden increase of ligand ∂L ∂L concentration causes a drop in the phosphorylated fraction of the receptor. *(10) a. Write down the equations for dC/dt and dC/dt. Solve for the steady state * *concentrations Css and Css. Under what conditions will Css be independent of L? *(10) b. Set δC = C − C , δC * = C − C *ss . Derive the linearized equations representing ss fluctuations from steady state, driven by fluctuations δL(t) of the ligand concentration. You should obtain d δC  − (k++ km ) ' δC   1  dt δC * = + k− )* + (α C * −α+ C )δL .− ss ss   + k+− (k−+ km δC  − 1 (10) c. Assume for now that δL = 0, km’ = 0, and km = 0. Calculate the eigenvectors and eigenvalues of the above matrix. You will find that one of the eigenvalues is zero. Recalculate this eigenvalue to first order in km. On a graph of δC vs. δC*, plot the eigenvectors and note the slow and fast eigenvalues. Sketch a few typical timecourses for various initial values of {δC,δC*}. This initial perturbation might arise if the system had first reached steady state for one value of L, but that value was abruptly changed. Sketch out such an event, showing a step *increase in L at time t = 0, and the subsequent evolution of C, C, and CT as functions of time. (10) d. Now assume that δL, δC, δC* ~ eiωt. This corresponds to Fourier transforming the equation above. 4Systems Biology 7.81/8.591/9.531 *δC (ω)Calculate the transfer function T (ω) = . L(ω δ ) Claiming that perfect adaptation holds corresponds to claiming that T(ω) has no dc component (T(ω = 0) = 0). Show that this is true only if km’ = 0. Assume from now on that km’ = 0. (i) What is the behavior of


View Full Document

MIT 8 591J - Problem Set 2

Download Problem Set 2
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Problem Set 2 and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Problem Set 2 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?