The proline/arginine-rich domain is a major determinant of dynamin self-activation

Abstract

Dynamins induce membrane vesiculation during endocytosis and Golgi budding in a process that requires assembly-dependent GTPase activation. Brain-specific dynamin 1 has a weaker propensity to self-assemble and self-activate than ubiquitously expressed dynamin 2. Here we show that dynamin 3, which has important functions in neuronal synapses, shares the self-assembly and GTPase activation characteristics of dynamin 2. Analysis of dynamin hybrids and of dynamin 1-dynamin 2 and dynamin 1-dynamin 3 heteropolymers reveals that concentration-dependent GTPase activation is suppressed by the C-terminal proline/arginine-rich domain of dynamin 1. Dynamin proline/arginine-rich domains also mediate interactions with SH3 domain-containing proteins and thus regulate both self-association and heteroassociation of dynamins.

Figure 1

Biochemical analysis of Dyn3. A. GTPase activities of Dyn3, Dyn2, and Dyn 1. B. Concentration-dependent GTPase activity of Dyn3. GTPase activity was measured at 37°C in a solution containing 45 mM NaCl, 20 mM Hepes, pH 7.5, 2 mM MgCl₂ and 1 mM GTP. Data points represent the averages of triplicate measurements of at least two preparations and the error bars show the standard error of the mean. C. Self-assembly of Dyn3 as measured by turbidity of 1 μM dynamin at 350 nm, 45 mM NaCl and at 37°C.

Figure 2

GTPase activities of dynamin constructs assayed as a function of dynamin concentration. A. Schematic presentation of domain structure of dynamins indicating the residues of interchanged domains (see Supplementary Materials and Methods). B. Activity profiles of wild-type Dyn1, Dyn1 with GED of Dyn2 (Dyn1(GED2)), and Dyn1 with PRD of Dyn2 (Dyn1(PRD2)). C. Activity profile of wild-type Dyn2, Dyn2 with GED of Dyn1 (Dyn2(GED2)), Dyn2 with PRD of Dyn1 (Dyn2(PRD1)), and Dyn3 with PRD of Dyn1 (Dyn3(PRD1)). D. Effect of Dyn1 on the GTPase activities of Dyn2 and Dyn3. Concentrations of Dyn2 and Dyn3 were 0.5 μM. GTPase activity was measured at 37°C in a solution containing 45 mM NaCl, 20 mM Hepes, pH 7.5, 2 mM MgCl₂ and 1 mM GTP. Data points represent the averages of triplicate measurements of at least two preparations and the error bars show the standard error of the mean.

Figure 3

Polymerization of dynamin constructs. A. Absorbance at 350nm of solutions containing wild-type Dyn2, Dyn2 with PRD of Dyn1 (Dyn2(PRD1)), and Dyn3 with PRD of Dyn1 (Dyn3(PRD1)). B. Absorbance at 350nm of solutions containing wild-type Dyn1 and Dyn1 with PRD of Dyn2 (Dyn1(PRD2)). Turbidity was measured at 1μM dynamin concentration at 37°C upon reduction of NaCl concentration from 300 mM to 45 mM. After reaching plateau, 50 mM GTP and 100 mM MgCl₂ were added (arrows) to obtain final concentrations of 1 and 2 mM GTP and MgCl₂, respectively.