Moretti AIS et al. (2017), Conserved Gene Microsynteny Unveils Functional ...

ECB-ART-49980

Sci Rep 2017 Dec 08;71:17262. doi: 10.1038/s41598-017-16947-5.

Show Gene links Show Anatomy links

Conserved Gene Microsynteny Unveils Functional Interaction Between Protein Disulfide Isomerase and Rho Guanine-Dissociation Inhibitor Families.

Moretti AIS , Pavanelli JC , Nolasco P , Leisegang MS , Tanaka LY , Fernandes CG , Wosniak J , Kajihara D , Dias MH , Fernandes DC , Jo H , Tran NV , Ebersberger I , Brandes RP , Bonatto D , Laurindo FRM .

Abstract
Protein disulfide isomerases (PDIs) support endoplasmic reticulum redox protein folding and cell-surface thiol-redox control of thrombosis and vascular remodeling. The family prototype PDIA1 regulates NADPH oxidase signaling and cytoskeleton organization, however the related underlying mechanisms are unclear. Here we show that genes encoding human PDIA1 and its two paralogs PDIA8 and PDIA2 are each flanked by genes encoding Rho guanine-dissociation inhibitors (GDI), known regulators of RhoGTPases/cytoskeleton. Evolutionary histories of these three microsyntenic regions reveal their emergence by two successive duplication events of a primordial gene pair in the last common vertebrate ancestor. The arrangement, however, is substantially older, detectable in echinoderms, nematodes, and cnidarians. Thus, PDI/RhoGDI pairing in the same transcription orientation emerged early in animal evolution and has been largely maintained. PDI/RhoGDI pairs are embedded into conserved genomic regions displaying common cis-regulatory elements. Analysis of gene expression datasets supports evidence for PDI/RhoGDI coexpression in developmental/inflammatory contexts. PDIA1/RhoGDIα were co-induced in endothelial cells upon CRISP-R-promoted transcription activation of each pair component, and also in mouse arterial intima during flow-induced remodeling. We provide evidence for physical interaction between both proteins. These data support strong functional links between PDI and RhoGDI families, which likely maintained PDI/RhoGDI microsynteny along > 800-million years of evolution.

PubMed ID: 29222525
PMC ID: PMC5722932
Article link: Sci Rep
Grant support: [+]

Article Images: [+] show captions

	Figure 1. Evolutionary trajectory of the PDI/RhoGDI gene cluster across the animal phylogeny. The emergence of the gene cluster dates back to the last common ancestor shared between bilateria and the cnidarians (N. vectensis). In earlier branching animal, e.g. sponges (A. queenslandica) or ctenophores (not shown), orthologs to the participating genes are unlinked. In the arthropods, represented by D. melanogaster, the cluster has been disrupted. Two subsequent rounds of cluster duplications in the last common ancestor of the vertebrates gave rise to the three contemporary paralogous gene clusters. The tree topology follows the accepted animal phylogeny, the vertebrate subtrees are arranged according to Supplementary Figure 1 (maximum likelihood PDI family tree). H. sapiens â Sarcopterygii; L. oculatus â Actinopterygii; C. milii â Chondrichthyes; S. purpuratus â Echinodermata; C. elegans â Nematoda; D. melanogaster â Arthropoda; N. vectensis â Cnidaria; A. queenslandica â Porifera. The genomic localizations in the individual species are given next to the cluster representation. In D. melanogaster, the two genes reside on the same chromosome, yet at a distance of >1âMb.
	Figure 2. Evolutionary histories of the PDI and RhoGDI paralogs, respectively, in the three human PDI/RhoGDI gene clusters. The trees are based on the maximum likelihood phylogeny of the PDIs and RhoGDIs, respectively, but were adapted for congruency with the species phylogeny. Topology testing revealed that the adapted topology is not significantly worse compared to the ML topology (SH test: pâ>â0.05). Branch labels denote percent bootstrap support. The Pfam83 domain architectures of the individual PDI and RhoGDI proteins are shown next to the leaf labels: PF02115 â Rho_GDI; PF13848 â Thioredoxin_6; PF00085 â Thioredoxin. The architectures are connected if the genes reside next to each other in the genome of the respective species. The orthologous groups, PDIA1, PDIA2, PDIA8 and correspondingly, RhoGDIÎ±, RhoGDIÎ³ and RhoGDIÎ² are indicated by the same background color. In two cases, an ortholog was not predicted, while gene order indicates its presence (âmissed orthologâ).
	Figure 3. Alignment of the genomic region containing PDIA1/RhoGDIÎ± cluster in Eukaryota. Gene names and positioning are based upon the Genomicus database (see methods). All species are aligned against Homo sapiens. The map is centralized in PDIA1 (P4HB)/RhoGDIÎ± (ARHGDIA) gene pair. Genes are aligned in columms and kept in the order in which they appear in chromosomes (Chr) and scaffolds (Scf), without consideration for distance, while the transcriptional sense is represented by the pentagon tip. All orthologs are drawn with the same color and the lettering on the top refers to the Homo sapiens genes. In addition to the main genes of interest, the neighboring genes are included for reference.
	Figure 4. Alignment of the genomic region containing PDIA2 /RhoGDIÎ³ cluster in vertebrates. Gene names and positioning are based upon the Genomicus database (see methods). All species are aligned against Homo sapiens. The map is centralized in PDIA2/RhoGDIÎ³ (ARHGDIG) gene pair. Genes are aligned in columms and kept in the order in which they appear in chromosomes (Chr) and scaffolds (Scf), without consideration for distance, while the transcriptional sense is represented by the pentagon tip. All orthologs are drawn with the same color and the lettering on the top refers to the Homo sapiens genes. In addition to the main genes of interest, the neighboring genes are included for reference.
	Figure 5. Alignment of the genomic region containing PDIA8/RhoGDIÎ² cluster in vertebrates. Gene names and positioning are based upon the Genomicus database (see methods). All species are aligned against Homo sapiens. The map is centralized in PDIA8 (Erp27)/RhoGDIÎ² (ARHGDIB) gene pair. Genes are aligned in columms and kept in the order in which they appear in chromosomes (Chr) and scaffolds (Scf), without consideration for distance, while the transcriptional sense is represented by the pentagon tip. All orthologs are drawn with the same color and the lettering on the top refers to the Homo sapiens genes. In addition to the main genes of interest, the neighboring genes are included for reference.
	Figure 6. Investigation of PDIA1 and RhoGDIÎ± co-regulation in different systems. (A) RhoGDIÎ± protein levels following PDIA1 overexpression in VSMC using a doxycyclin-inducible lentiviral vector carrying a myc-tag; (B) Effects of PDIA1 overexpression upon RhoGDIÎ± levels in a transgenic mouse model. Tissue from transgenic PDIA1-overexpressing mice: brain, kidney, liver, aortae and heart. Data representative of nââ¥â3, 2 (heart) or 1 (brain). (C) RhoGDIÎ± protein and mRNA levels in VSMC from transgenic PDIA1-overexpressing mice. Representative of 3 independent experiments. Uncropped western blots are shown in Supplementary FigsÂ S8 and S9.
	Figure 7. Investigation of PDIA1 and RhoGDIÎ± co-regulation in different systems. (A) Effects of induced PDIA1 transcriptional activation on RhoGDIÎ± protein levels (see methods); three distinct guide RNAs, depicted on the map above, were used to drive CRISPR dCas9 VP64-mediated PDIA1 transcription; (B) Design similar to (A): effects of induced RhoGDIÎ± transcriptional activation on PDIA1 protein levels; three distinct guideRNAs, depicted on the map above, were used to drive CRISPR dCas9 VP64-mediated RhoGDIÎ± transcription. In both cases, data are representative of three sets of independent experiments. Results are mean and SEM. (C) Results of mRNA expression levels from experiments depicted in (A) and (D). Representative from 3 independent set of experiments; results are mean and SEM. Uncropped western blots are shown in Supplementary Fig.Â S8. (D) Correlation coefficient (r2) between mRNA expressions of PDIA1 vs. RhoGDIÎ± from left carotid arteries (L) 48âh after partial carotid ligation and their contralateral right carotid arteries (R). Left graph depicts intimal samples and right graph depicts left over tissue (mediaâ+âadventitia). Data represent Delta Ct (gene target - 18âS). Each dot represents an individual artery from nâ=â4â5 mice.
	Figure 8. Physical interaction between PDIA1 and RhoGDIÎ± in endothelial cells. (A) and (B) RhoGDIÎ± immunoprecipitation followed by Western blot for PDIA1. RhoGDIÎ± was immunoprecipitated using IgG against N-terminus (A) or C-terminus (B). (C) PDI and RhoGDIÎ± interaction by pulldown assay. Using RhoGDIÎ±-GST which was either reduced (DTT 20âmM) or oxidized (H2O2 20âmM), we pulled down PDI from HUVEC lysates under reduced conditions. Reduced RhoGDIÎ± was able to pull-down PDIA1 as a monomer and as an apparent heterodimer, while oxidized RhoGDIÎ± incubation resulted in the appearance of its dimeric form. Results representative from 3 experiments. (D) Confocal microscopy images for PDIA1 and RhoGDIÎ± showing co-localization. Representative results of three experiments which were replicated using 2 different antibodies to PDIA1 and 2 different antibodies to RhoGDIÎ± (N-terminus and C-terminus). Magnification 40x and zoom 2.5x; Uncropped western blots are shown in Supplementary Fig.Â S10. (E) Representative confocal images of proximity ligation assay (PLA) analysis showing the interaction between PDI and RhoGDIAÎ± in HUVEC. Positive signal of protein interaction is represented as a red dot; nuclei are stained with DAPI (blue). Cells in panels E1-E4 were incubated with a mix of antibodies against PDIA1 and RhoGDIÎ±. Panel B shows negative control in which the primary antibodies were omitted. Magnification: in E, 40x; in F1/F3, 40x and zoom 2x; in F2/F4, 63x and zoom 2x; (F) Representative ortho-images (F1 and F2) and 3D reconstruction (F3 and F4) of confocal z-stacks of PLA showing PDI/RhoGDIÎ± dimer distribution into HUVEC. The center image of each panel (shown by crossing lines) is the X-Y view, cross section at the green line is X-Z view and cross section at red line is Y-Z view.

References [+] :

Abascal, ProtTest: selection of best-fit models of protein evolution. 2005, Pubmed