New insights into the phylogenetic distribution and evolutionary origins of the septins

Abstract

Until recently, it had appeared that the septin family of proteins was restricted to the opisthokont eukaryotes (the fungi and animals and their close relatives the microsporidia and choanoflagellates). It has now become apparent that septins are also present in several other widely divergent eukaryotic lineages (chlorophyte algae, brown algae, and ciliates). This distribution and the details of the non-opisthokont septin sequences appear to require major revisions to hypotheses about the origins and early evolution of the septins.

Figure 1. Distribution of septins in eukaryotes

A simplified version of the eukaryotic tree of life proposed by (Baldauf, 2003) is shown. Magenta, taxonomic groups that are known to include organisms with septin genes; green, groups for which no septin genes are found in the annotated sequence databases although some members have had their whole genomes sequenced; black, groups for which the presence or absence of septin genes cannot be determined because of a lack of whole-genome sequences.

Figure 2. Alignment of the amino-acid sequences of three non-opisthokont septins with that of a canonical animal septin

The sequence of D. melanogaster (Dme) Pnut was aligned with those of representative septins from the ciliate P. tetraurelia (Pte), the brown alga E. siliculosus (Esi), and a chlorophyte alga (C. reinhardtii: Cre) as described in Materials and Methods. Four manual gap adjustments were made near the C-terminal end of the septin-unique element to improve the alignment in that region. Numbers indicate amino acid positions. Amino acids that are identical or similar (L,I,V; S,T; D,E; K,R; N,Q) between Pnut and one of the other proteins are shaded in black and gray, respectively (with black overriding gray in the Pnut sequence). The dashed box indicates the approximate position of the polybasic region that is found in many septins. The G1, G3, and G4 motifs of the predicted guanine-nucleotide-binding sites are indicated by black arrows. White boxes and letters therein indicate motifs and single residues that are highly conserved in opisthokont septins (Pan et al., 2007; Momany et al., 2008), including the "septin-unique element" (Versele et al., 2004). The C-terminal six amino acids (EQARPT) of the Esi sequence are omitted (#) to save space.

Figure 3. Absence of C-terminal coiled-coil domains in most non-opisthokont septins

Coiled-coil probabilities were calculated as described in Materials and Methods and are shown for D. melanogaster Pnut, S. cerevisiae Cdc3; the septins from C. reinhardtii (Cre), E. siliculosus (Esi), P. tetraurelia (Pte), and T. thermophila (Tth); and one of the septins from C. variabilis (Cva). The other C. variabilis septin and those from V. carteri (Vca) and N. bacillaris (Nba) contained no predicted coiled coils (data not shown).

Figure 4. Hydropathy plots of septin proteins

Hydropathy values were calculated as described in Materials and Methods using a window size of 19 with the exception of Tth_EAS01103, for which a window of 9 was used to make the C-terminal hydrophobic patch easier to see. Species names are abbreviated as in Figure 3. Black bars represent septin core domains (G1, G3, and G4 motifs plus the septin-unique element). Hydrophobic patches just C-terminal to the core domains are shaded black. A second hydrophobic patch in Esi_CBJ31641 is shaded gray.

Figure 5. Phylogeny of septin proteins

A majority consensus phylogeny for 65 septins was inferred by Bayesian analysis as described in Materials and Methods. The septins analyzed included 19 non-opisthokont proteins (species names are abbreviated as in Figure 3) plus 46 animal and fungal septins (see list in Materials and Methods). The 37 of these animal and fungal septins that were previously assigned to Groups 1A, 1B, 2A, 2B, 3, and 4 by Momany and co-workers (Pan et al., 2007; Momany et al., 2008) also formed a single clade (with multiple branches) in our analysis and are represented here by the line with black triangle. Posterior probabilities are indicated on selected branches. The S. cerevisiae (Sce) Myo2 motor domain served as outgroup. The scale bar represents evolutionary distance in substitutions per residue.