ERK2 is a kinase protein that belongs to a Ras/Raf/MEK/ERK signaling pathway, which is activated in response to a range of extracellular signals. Malfunctioning of this cascade leads to a variety of serious diseases, including cancers. This is often caused by mutations in proteins belonging to the cascade, frequently leading to abnormally high activity of the cascade even in the absence of an external signal. One such “gain-of-function” mutation in the ERK2 protein, called a “sevenmaker” mutation (D319N), was discovered in 1994 in Drosophila. The mutation leads to disruption of interactions of other proteins with the D-site of ERK2 and results, contrary to expectations, in an increase of its activity in vivo. However, no molecular mechanism to explain this effect has been presented so far. The difficulty is that this mutation should equally negatively affect interactions of ERK2 with all substrates, activators, and deactivators. In this paper, we present a semiquantitative kinetic network model that gives a possible explanation of the increased activity of mutant ERK2 species. A simplified biochemical network for ERK2, viewed as a system of coupled Michaelis-Menten processes, is presented. Its dynamic properties are calculated explicitly using the method of first-passage processes. The effect of mutation is associated with changes in the strength of interaction energy between the enzyme and the substrates. It is found that the dependence of kinetic properties of the protein on the interaction energy is nonmonotonic, suggesting that some mutations might lead to more efficient catalytic properties, despite weakening intermolecular interactions. Our theoretical predictions agree with experimental observations for the sevenmaker mutation in ERK2. It is also argued that the effect of mutations might depend on the concentrations of substrates.