Functional characterization of phosphorylation sites in dynamin-related protein 1

Abstract

Dynamin-related protein 1 is member of the dynamin-family of large GTPases. Similar to the endocytosis motor dynamin, Drp1 uses GTP hydrolysis to power constriction of the outer mitochondrial membrane and ultimately mitochondrial division. Dynamin phosphorylation in its unique C-terminal proline-rich domain interferes with binding of accessory proteins that induce membrane curvature and inhibits clathrin-mediated endocytosis. Evidence within the last few years indicates that Drp1 is also regulated by the phosphorylation/dephosphorylation cycle. Drp1 regulation is complex, in that both inhibitory and activating phosphorylations have been described that lead to, respectively, mitochondrial elongation and shortening. In this chapter, we describe methods for the identification and functional characterization of Drp1 phosphorylation sites. Among these methods is replacement of the endogenous protein by phosphorylation-site mutant Drp1 via combined shRNA and RNAi-resistant cDNA expression from the same plasmid. We also discuss primary astrocyte cultures as a model for regulation of cell death and mitochondrial morphology by Drp1 and present ImageJ macro source code for unbiased quantification of mitochondrial shape changes.

Figure 13.1

Drp1 domain diagram with phosphorylation sites. Drp1 consists of an N-terminal GTPase domain, followed by a middle (MID) and variable (VD) domain, as well as a C-terminal GTPase effector domain (GED). Alternative splicing of three exons generates eight splice variants and changes the numbering of the two highly conserved phosphorylation sites that have been characterized thus far. Kinases (↓) and phosphatases (⊥) that target the two phosphorylation sites are discussed in Section1.

Figure 13.2

Stable replacement of endogenous with GFP-tagged, mutant Drp1. PC12 cells were transfected with a vector that expresses a Drp1-targeting shRNA together with the RNAi-resistant, GFP-tagged cDNA. After selection in G418, clonal cell lines were immunoblotted for Drp1, GFP, and the catalytic subunit of PP2A (PP2A/C) as a loading control. Cell lines analyzed in the first two lanes have replaced endogenous with similar levels of GFP-tagged Drp1, whereas the line in lane three failed to silence endogenous Drp1expression.

Figure 13.3

Flowchart of mitochondrial morphometry with ImageJ. The raw image (HeLa cells fluorescently labeled with mitochondrial MTC02 antibody) on the left is first enhanced (deblurred) by 2D deconvolution and then converted to a binary (black and white) image. Individualparticles in the binary image are analyzed for area, perimeter, as well as major and minor axis of a bounding rectangle using the”Analyze Particles” function of ImageJ.These measurements are used to calculate aspect ratio, form factor and area-weighted form factor (product of area and form factor).

Figure 13.4

Comparison of three mitochondrial shape metrics. Rat primary astrocytes were transfected with mitochondrial dsRed, treated for 0^3 h with 1 μM staurosporine, fixed, and analyzed for mitochondrial morphology using the ImageJ macro listed in Section 7. Representative epifluorescence images are shown in (A) (with magnified areas in the insets), while three mitochondrial shape metrics are plotted in (B) (means ± SD of ~130 cells per time point). Staurosporine results in mitochondrial fragmentation that is significant by all metrics (values at the 2 and 3 h time point are not significantly different from one another, while pairwise comparisons between all other time points result in p values <0.01).

Figure 13.5

Effect of Drp1 phosphorylation on mitochondrial morphology is not influenced by alternative splicing. Rat primary astrocytes were transfected with mitochondrial dsRed and plasmids expressing both Drp1-targeting shRNA and GFP-Drp1 cDNAs. Phosphorylation-blocking (S→A) and -mimicking (S→D) substitutions of the PKA/CaMKI phosphorylation site (Ser656 in rat splice variant 111) were compared to wild type (WT) in the context of Drp1 splice variants with (111) and without (100) the two alternative exons in the variable domain (see Fig.13.1). Shown are representative epi-fluorescence images of astrocytes with dsRed-labeled mitochondria (A), as well as mitochondrial shape quantified as form factor (mean ± SD/SEM (lower/upper bars) of ~100 cells/condition). *p <0.005 by Student’s t-test.