Fig 1.
IGSF1 domain structure and known mutations in IGSF1.
A) IGSF1 is co-translationally cleaved into the NTD and CTD by signal peptidase and signal peptide peptidase. The NTD is retained in the ER, while the CTD is trafficked to the plasma membrane. The figure was generated in BioRender. B) Known mutations in IGSF1 cluster in the CTD coding region.
Fig 2.
Ig loops in the NTD cluster with specific loops from the CTD.
A) Ig loops with the same colour have the highest sequence identity. B) Sequence alignment of each Ig loop with highly conserved residues highlighted (blue = hydrophobic, red = positive charge, magenta = negative charge, green = polar, pink = cysteine, orange = glycine, yellow = proline, cyan = aromatic, non-conserved = white). Ig loops are ordered based on sequence identity; Ig1 with Ig6, Ig2 with Ig7, and so on. Sequences are from human accession number Q8N6C5. C) A phylogenetic tree of the 12 Ig domains in human IGSF1. The branches are labelled with PhyML aLRT support values.
Table 1.
Identity and similarity between the Ig loops in human IGSF1 NTD and CTD.
Fig 3.
IGSF1 has high sequence similarity in all mammals and only occurs in chordates.
Phylogenetic tree based on a Clustal Omega alignment of IGSF1 and its human paralogs and their sets of orthologs across eutheria made using PhyML default parameters, except that starting tree topology and invariable sites were optimized, and the PhyML option for using the best of the available tree searching operations was applied. Bi-directional best hits (BBH) from a selection of non-eutherian amniotes were included as outgroups. The IGSF1-NTD and -CTD regions were aligned separately. The clades for IGSF1-NTD and -CTD, A1BG, OSCAR, FCAR, TARM1, NCR1, LILRA5, and VSTM1, and ‘Other Amniote’ sequences are labelled on the appropriate branches, and collapsed to wedge shapes, except for the IGSF1 and A1BG clades, which are highlighted in grey. The nodes at which branches end were color-coded by PHyML aLRT branch support value as indicated by the heat map. The tree was drawn using FigTree.
Fig 4.
IGSF1 loop 4 is a highly conserved member of the LRC family.
Phylogenetic tree based on a Clustal Omega alignment of IGSF1 Ig loop 4 and its human paralogs and their sets of orthologs across eutheria. In addition, BBH hits from a selection of non-eutherian amniotes were included as outgroups. The clades for IGSF1-NTD and -CTD regions, A1BG, OSCAR, FCAR, TARM1, NCR1, LILRA5, and VSTM1, and ‘Other Amniote’ sequences are labelled on the appropriate branches. The nodes at which branches end are labelled and colour-coded by PHyML aLRT branch support value as indicated by the heat map.
Fig 5.
IGSF1 Ig loops cluster with specific Ig loops from other LRC members.
A schematic depiction of the Ig-like domains in IGSF1-NTD and -CTD, A1BG, OSCAR, FCAR, TARM1, NCR1, LILRA5, and VSTM1. These domain alignments were taken from the Clustal Omega multiple sequence alignment. The endpoints of the domains from Pfam [or SMART, if there was no Pfam annotation; these are both taken from InterPro [27]] are labelled in italics.
Fig 6.
Full-length and CTD-only Igsf1 isoforms encode the same protein.
Protein lysates from HEK293 cells transfected with pcDNA3.0, murine full-length IGSF1-1 or murine CTD-coding IGSF1-4 expression vectors, or from murine pituitaries were treated with PNGase F (+) or left untreated (-) and immunoblotted using an IGSF1-CTD antibody. Arrows indicate the mature and immature glycoforms, as well as deglycosylated IGSF1.
Fig 7.
The extant IGSF1 gene results from duplication of the ancestral IGSF1 gene.
A schematic depiction of the evolution of the IGSF1 gene. The ancestral gene encoded only the IGSF1-CTD. The region encoding Ig loop 8 in the CTD was lost in the NTD. Additionally, loop 11 in the CTD was duplicated and inserted upstream of itself creating loop 9 (or vice versa). Loop 9 was duplicated and emerged in the NTD as loop 3. Loop 11 was lost in the NTD.