Fig 1.
Schematic representation of the knotted (A) and slipknotted (B) chains.
(A) The knotted chain (left) and protein backbone (PDBID: 4kpp) (right) with a knotted core marked with blue color. This knot is called open trefoil. (B) The slipknotted chain (left) and its representation in the sodium-dependent citrate symporter (KpCitS), a protein with slipknotted backbone identified in this paper (PDBID: 5xar). The knotted core is shown in blue, the slipknot loop in orange and the slipknot tail in green.
Fig 2.
Conserved helical region (core) found in the monovalent cation-proton antiporter superfamily.
Conservation of this region suggests that three different fold types, including one possessing a non-trivial topology (a slipknot), evolved from a common, single-domain ancestor. The putative ancestor is shown in light green box in the middle. Three arrows from the ancestor navigate to three proteins with different folds: 1) left—two-domain slipknotted protein; 2) middle—one-domain unknotted protein; 3) right—two-domain unknotted protein. On the bottom left of the figure is shown a schematic diagram of the entangled region of the slipknotted protein colored from blue (N-terminal) to red (C-terminal). On the bottom right there is a similar schematic diagram that shows the topology of the unknotted protein backbone.
Fig 3.
Structure of the slipknotted transporter KpCitS.
(A) Location of the knotted core (N115-N401) starts at TM4 of domain A, loop-linker and TM8-TM11 and half of TM12 hairpin-like helix of domain B. Slipknot loop (R402-S421) is formed by half of TM12 hairpin-like helix and part of TM13. The rest of TM13 (residues Y422-I446) is slipknot tail. (B) Schematic representation of the slipknot topology. (C, D) Structure of the knotted core and slipknot’s loop and tail based on PDBID 5xar. (E) Knot fingerprint calculated based on the KpCitS structure (PDBID: 5xar).
Table 1.
Known protein families from monovalent cation-proton antiporter superfamily investigated toward identification of possible evolution of the slipkotted topology.
IDs of families and clans are from Pfam database. ND—not determined.
Fig 4.
Identified structural differences between the slipknotted and unknotted proteins.
(A) Schematic representation of the full length two-domain slipknotted protein. Two domains are connected by the long loop. The conserved core region is colored in green. TM helices are enumereted as in the structure (PDBID: 5a1s). (B) Schematic representation of the unknotted protein. Similarly, unknotted protein is composed of two domains. Conserved core is colored in green. TM helices are enumereted as in the structure (PDBID: 4bwz). (C) Structure of a single domain of the slipknotted protein (PDBID: 5a1s). (D) Superposition of the domains A of the slipknotted and unknotted proteins. (E) Structure of a single domain of the unknotted protein (PDBID: 4bwz). (F) Superposition of the fragments of domains A of the slipknotted and unknotted proteins.
Fig 5.
Sequence and phylogenetic analysis of the domains suggests a common evolutionary origin.
Panel A shows clustering of domain profiles. Each family shown in a different color, connections indicate similarity found with e-value below 1e−5. Two domains of the slipknotted protein are shown as orange stars. All families from CL0062 are shown as circles colored according to the family. The family IDs with known unknotted topology are highlighted in blue font. The one-domain proteins (PF03788 and PF04172) are shown as triangles and family ID are colored in magenta font. Two domains of PF05145 (CL0142) are shown as red squares. Panel B shows the phylogenetic tree of the domains. Color-coding of tree branches is the same as for domains clustering on the panel A. Additionally, colored background areas serve to group the tree branches according to some properties: 1) families with known unknotted topology are highlighted with blue; 2) the slipknotted family is highlighted in orange; 3) the families with unknown topology are highlighted in light gray; 4) the families of one-domain proteins are highlighted with magenta.
Fig 6.
Seven evolutionary scenarios found within the transporters.
1) No fusion—diversification of the ancestor gene leading to one-domain protein. 2) Long diversification—gene duplication followed by an extended period of diversification which lead to two domains with low sequence similarity (shown with red and blue). 3) Duplication of already fused protein (PF13593 speciated from PF01758, PF06965 from PF00999). 4) Reverse-order fusion—gene duplication and fusion in a reverse order. 5) Fusion of the domains from different lineages (shown with blue and green). 6) Fast fusion—gene duplication followed by an instant fusion (in families PF05145, PF06826 and PF03956). 7) Fusion of unrelated lineages (PF05982).