Figure 1.
Schematic representation of the mitogen-activated protein kinase (MAPK) cascade.
Activation through phosphorylation of the substrate kinase at each step is mediated by the upstream kinase, the single and doubly phosphorylated states of the substrate being denoted by the superscripts “*” and “**”, respectively. Activated kinases are eventually dephosphorylated into their inactive forms by the corresponding phosphatases (indicated by the suffix “-P’ase”). Activity in the three-component pathway is initiated by a signal S regulating the activation of the MAPK kinase kinase (MAP3K). Activated MAP3K controls the activation of MAPK kinase (MAP2K), which in its turn regulates the activation of MAPK. Note that unlike the single phosphorylation of MAP3K, both MAP2K and MAPK require double phosphorylation in order to become active, i.e., capable of acting as the enzyme for the corresponding downstream substrate protein. The eventual response of the cascade is quantified by the concentration of activated MAPK, which can be used in initiating transcription, activating other protein molecules, etc.
Figure 2.
The branched MAPK cascade network motif.
(A) A schematic diagram representing a simple branched cascade with two parallel pathways that is seen in many experimental systems, e.g., the T-cell receptor network [8]. The initial signal S activates a common MAP3K that can phosphorylate two different types of MAP2K molecules, viz., MAP2KA and MAP2KB. Each MAP2K type activates a particular type of MAP kinase, MAPKA and MAPKB, respectively. Specific examples of branched MAPK cascade motifs obtained from the experimental literature are shown in (B–D). They correspond to systems with the specific MAP3K (the branching point of the motif) being (B) MEKK1 that activates both MKK1/2 [15] and MKK4/7 [16], [17], (C) MEKK2 that activates both MKK4/7 [18], [19] and MKK6 [19], and (D) MEKK3/4 that activates both MKK4/7 [18]–[22] and MKK3/6 [19]–[21].
Figure 3.
Amplification of response in unperturbed branch through retrograde propagation of information in a branched motif.
(A) Relative change in the response of different molecular species in branches A and B as a function of signal strength, when phosphorylation of terminal kinase MAPKA in branch A is inhibited. Inset shows a schematic diagram of the branches, with the “stop” sign indicating blocking of activation of MAPK in branch A. (B–F) The perturbation results in amplification (relative to the unperturbed condition) of the concentrations of (B) MAP3K* (D) MAP2K and (F) MAPK
, shown as a function of signal strength and concentration of the phosphatase MAP3K-Pase. Relative increase in concentrations of (C) MAP2K
and (E) MAPK
as a result of the perturbation are also shown as a function of signal strength and the total concentration of MAP3K.
Figure 4.
The branched MAPK cascade responds differently to to distinct perturbations which inhibit the activation of a terminal kinase in one branch.
The normalized response, i.e., the ratio of activated to total kinase concentration, of different molecular species in the reaction cascade when (A) the total concentration of MAPKA is varied and (B) when the product formation rate () of MAPKA is varied shows opposing behavior in the activity of molecules in the unperturbed B branch, although both types of perturbations have the same functional goal of decreasing the activation of MAPKA. As [MAPKA]
is decreased, both MAPKB and MAP2KB decrease in activity. However, when the product formation rate of MAPKA is decreased, the activity of both MAPKB and MAP2KB are increased. The activity of the kinase MAP3K which forms the branch-point also shows different response to the two perturbations unlike the result for single-branch cascades: in (A), the activity is unchanged, whereas in (B), the activity of MAP3K increases, on decreasing the activation of MAPKA. Note that the curves corresponding to MAPKA and MAPKB intersect in (A) when [MAPKA]total = [MAPKB]total = 1.2
M and they intersect in (B) when the product formation rates for MAPKA and MAPKB, i.e.,
and
respectively, are both 0.5 s−1. Except for [MAPKA]total and
, all other total molecular concentrations and reaction rates are kept fixed at the values given in Table S1. The broken lines indicate the physiologically plausible range of values for [MAPKA]total in (A) and
in (B), as used in the Huang-Ferrell model [45].
Figure 5.
Experimental validation of the amplification of activity in one branch on inhibiting the activity of the terminal kinase in the other branch.
Western blots of JNK and p38MAPK phosphorylation in primary macrophages when subjected to CD40 stimuli of strength 3 g/ml are shown under normal (JNK: A, p38MAPK: B) and perturbed (JNK: C, p38MAPK: D) conditions. Perturbation is applied by inhibiting the phosphorylation of either p38MAPK (C) or JNK (D) by using pharmaceutical agents. In each figure, the upper and lower panels show the phosphorylated and total concentrations of the different molecular species (represented by “p JNK” and “JNK”, and “p- p38” and “p38” for JNK and p38MAPK respectively). The control condition shows an unstimulated system, while the times 3, 10, 15 and 30 minutes refer to observations after the system has been exposed to stimuli of the corresponding duration. All experiments have been carried out in triplicate and a representative set of blots are shown. The densitometric analysis of the blots is indicated in Table 1.
Table 1.
Densitometric analysis of western blots for JNK and p38MAPK under normal and perturbed conditions at different time points.
Figure 6.
Relative increase in response as a function of signal concentration on blocking MAPKA activation when the binding reaction rates for (A) branch A (,
and
) or (B) branch B (
,
and
) are 10 times higher than those in the other branch, and, when the total concentrations of MAPK and MAP2K in (C) branch A or (D) branch B are 10 times higher than the corresponding values for the other branch ( = mean value in the Huang-Ferrell range). Note that in (A) the MAPKB activation can increase by more than 1000 times for a particular range of signal strength. In contrast, there is relatively little change in the activity of the two branches when the activation of the terminal kinase for the branch having lower values of reaction rates is blocked.
Figure 7.
Role of competitive inhibition.
Relative increase with stimulus strength of the steady-state response of different molecular species in the unperturbed branch B (MAP2KB and MAPKB) shown as a function of signal concentration, when phosphorylation of MAPKA is prevented. In (A) both MAP2K and MAPK are singly phosphorylated, while in (B) MAP2K are singly phosphorylated but MAPK are doubly phosphorylated. Note that there is a small increase in the steady-state response of MAP2KB and MAPKB in (B) compared to (A). (C) When MAP2K are doubly phosphorylated whereas MAPK are singly phosphorylated, the relative increase in steady-state activity of MAP2KB and MAPKB is more prominent.
Figure 8.
(A–B) Relative increase in steady-state response in a four-branch cascade which allows only single phosphorylation of MAP2K while MAPK is doubly phosphorylated as a function of stimulus strength for the cases (A) when MAPKA phosphorylation is prevented and (B) when phosphorylation of in two branches, i.e., of both MAPKA and MAPKA, are blocked. The effect of having four branches but with singly phosphorylated MAP2K is similar to having two branches with double phosphorylated MAP2K but with a lower relative change in the response. This suggests that competitive inhibition is playing a role but is not solely responsible for the retrograde propagation in the branched cascade model. (C–D) Relative increase in steady-state response of doubly phosphorylated MAP2K and MAPK in the unperturbed branches as a function of signal strength when phosphorylation of MAPKA is prevented in a (C) a four-branch and (D) a three-branch cascade. As the number of branches are increased, the relative change in MAP2K activity on perturbation decreases faster than that of MAPK.