The P-loop (see <PDOC00017>) guanosine triphosphatases (GTPases) control a
multitude of biological processes, ranging from cell division, cell cycling,
and signal transduction, to ribosome assembly and protein synthesis. GTPases
exert their control by interchanging between an inactive GDP-bound state and
an active GTP-bound state, thereby acting as molecular switches. The common
denominator of GTPases is the highly conserved guanine nucleotide-binding (G)
domain that is responsible for binding and hydrolysis of guanine nucleotides.
Members of the dynamin GTPase family appear to be ubiquitous. They catalyze
diverse membrane remodelling events in endocytosis, cell division, and plastid
maintenance. Their functional versatility also extends to other core cellular
processes, such as maintenance of cell shape or centrosome cohesion. Members
of the dynamin family are characterized by their common structure and by
conserved sequences in the GTP-binding domain. The minimal distinguishing
architectural features that are common to all dynamins and are distinct from
other GTPases are the structure of the large GTPase domain (~280 amino acids)
and the presence of two additional domains: the middle domain and the GTPase
effector domain (GED) (see <PDOC51388>), which are involved in oligomerization
and regulation of the GTPase activity. In many dynamin family members, the
basic set of domains is supplemented by targeting domains, such as:
pleckstrin-homology (PH) domain (see <PDOC50003>), proline-rich domains
(PRDs), or by sequences that target dynamins to specific organelles, such as
mitochondria and chloroplasts [1,2,3].
The dynamin-type G domain consists of a central eight-stranded β-sheet
surrounded by seven α helices and two one-turn helices (see <PDB:4H1U>).
It contains the five canonical guanine nucleotide binding motifs (G1-5). The
P-loop (G1) motif (GxxxxGKS/T) is also present in ATPases (Walker A motif) and
functions as a coordinator of the phosphate groups of the bound nucleotide. A
conserved threonine in switch-I (G2) and the conserved residues DxxG of
switch-II (G3) are involved in Mg(2+) binding and GTP hydrolysis. The
nucleotide binding affinity of dynamins is typically low, with specificity for
GTP provided by the mostly conserved N/TKxD motif (G4). The G5 or G-cap motif
is involved in binding the ribose moiety [4,5,6].
Some proteins containing a dynamin-type G domain are listed below [2,3]:
- Animal dynamin, the prototype for this family. The role of dynamin in
endocytosis is well established. Additional roles were proposed in vesicle
budding from the trans-Golgi network (TGN) and the budding of caveolae from
the plasma membrane .
- Vetebrate Mx proteins, a group of interferon (IFN)-induced GTPases involved
in the control of intracellular pathogens [6,7].
- Eukaryotic Drp1 (Dnm1 in yeast) mediates mitochondrial and peroxisomal
- Eukaryotic Eps15 homology (EH)-domain-containing proteins (EHDs), ATPases
implicated in clathrin-independent endocytosis and recycling from
endosomes. The dynamin-type G domains of EHDs bind to adenine rather than
to guanine nucleotide [8,9].
- Yeast to human OPA1/Mgm1 proteins. They are found between the inner and
outer mitochondrial membranes and are involved in mitochondrial fusion.
- Yeast to human mitofusin/fuzzy onions 1 (Fzo1) proteins, involved in
mitochondrial dynamics [10,11].
- Yeast vacuolar protein sorting-associated protein 1 (Vps1), involved in
vesicle trafficking from the Golgi.
- Escherichia coli clamp-binding protein CrfC (or Yjda), important for the
colocalization of sister nascent DNA strands after replication fork passage
during DNA replication, and for positioning and subsequent partitioning of
sister chromosomes 
- Nostoc punctiforme bacterial dynamin-like protein (BDLP) [13,14],
The signature pattern that we developed for the dynamin-type G domain is based
on a highly conserved region downstream of the ATP/GTP-binding motif 'A' (P-loop) (see <PDOC00017>). We also developed a profile that covers the entire
PROSITE is copyright. It is produced by the SIB Swiss Institute
Bioinformatics. There are no restrictions on its use by non-profit
institutions as long as its content is in no way modified. Usage by and
for commercial entities requires a license agreement. For information
about the licensing scheme send an email to