Zhang BN et al. (2020), rad21 Is Involved in Corneal Stroma Development...

XB-ART-57474

Int J Mol Sci 2020 Oct 21;2120:. doi: 10.3390/ijms21207807.

Show Gene links Show Anatomy links

rad21 Is Involved in Corneal Stroma Development by Regulating Neural Crest Migration.

Zhang BN , Liu Y , Yang Q , Leung PY , Wang C , Wong TCB , Tham CC , Chan SO , Pang CP , Chen LJ , Dekker J , Zhao H , Chu WK .

???displayArticle.abstract???
Previously, we identified RAD21R450C from a peripheral sclerocornea pedigree. Injection of this rad21 variant mRNA into Xenopus laevis embryos disrupted the organization of corneal stroma fibrils. To understand the mechanisms of RAD21-mediated corneal stroma defects, gene expression and chromosome conformation analysis were performed using cells from family members affected by peripheral sclerocornea. Both gene expression and chromosome conformation of cell adhesion genes were affected in cells carrying the heterozygous rad21 variant. Since cell migration is essential in early embryonic development and sclerocornea is a congenital disease, we studied neural crest migration during cornea development in X. laevis embryos. In X. laevis embryos injected with rad21 mutant mRNA, neural crest migration was disrupted, and the number of neural crest-derived periocular mesenchymes decreased significantly in the corneal stroma region. Our data indicate that the RAD21R450C variant contributes to peripheral sclerocornea by modifying chromosome conformation and gene expression, therefore disturbing neural crest cell migration, which suggests RAD21 plays a key role in corneal stroma development.

???displayArticle.pubmedLink??? 33096935
???displayArticle.pmcLink??? PMC7594026
???displayArticle.link??? Int J Mol Sci
???displayArticle.grants??? [+]

Genes referenced: arhgap44 col24a1 ctcf foxg1 myh6 pitx2 rad21 septin10 tfap2a twist1

Phenotypes: rad21 mut RNA injection [+]

???attribute.lit??? ???displayArticles.show???

	Figure 1 Clusters of expression profiles of differentially expressed genes (DEGs) and expression levels. (a) Six different expression clusters were obtained from downregulated genes. (b) One expression cluster was identified from the upregulated genes. (c) Expression level changes were detected in three functional groups in affected lymphoblastoid cell lines (LCLs). The expression levels of three PCDHG gene clusters (PCDHGA, PCDHGB, and PCDHGC) were upregulated in LCLs from affected members. Eight mitochondria-related genes and five MHC class II-related genes showed reduced expression levels in affected-members LCLs. The functional enrichment scores were obtained by using DAVID bioinformatics. Three significant functional groups with false discovery rate (FDR) < 0.05 were obtained, and the expression levels of genes within these groups are shown.
	Figure 2 Local chromosome organization changes were detected in LCLs. (a) The coding genes overlapping with insulation score change (ISC) bins were chosen for functional enrichment analysis. Significant functional groups (FDR < 0.05) identified were cell adhesion, membrane components, glucuronidation, and N,N-dimethylaniline monooxygenase activity. (b) Local chromosome structural changes and gene expression changes in LCLs. Four differentially expressed genes/gene clusters showed changes in isolation scores, indicating local chromosome structural alterations may lead to expression changes in LCLs. (c) RAD21R450C caused insulation score changes in PCDH gene clusters. Insulation score profile across the PCDH gene clusters. Both of these loci showed CTCF and RAD21 binding signals (indicated by red arrowheads) in GM12878 cells.
	Figure 3 rad21mut disrupts the migration of neural crest cells (NCCs) in stage-25 Xenopus laevis. (a) The in situ hybridization pattern of twist1 was altered in the rad21mut-injected side. Panel i and ii show twist1 expression on both sides of the same non-injected embryo. The patterns around the optic cup (oc) and the neural crest migration streams into branchial arches (ba) are intact (indicated by a dash circle and dark solid lines, respectively). Panel iii and iv show the non-injected side and the injected side of the same rad21mut-injected embryo. The optic cup and branchial arch in panel iii are similar to those in panel i and ii, whereas these patterns are disrupted in panel iv. A total of 20 rad21mut-injected embryos were used, and 18 of them showed disrupted patterns in the rad21mut-injected side (18/20). (b) The in situ hybridization pattern of ap2a is altered in rad21mut-injected X. laevis embryos. Panel i shows ap2a is highly expressed in the head region and the spinal cord (sc) in the non-injected side. Panel ii shows the ap2a expression pattern is disrupted in the rad21mut-injected side. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). Panel iii and iv show higher magnifications of the non-injected side and rad21mut-injected side of another embryo. Similar to panel ii, disrupted neural crest migration streams are shown in panel iv. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). (c) Quantification of the mandibular areas showing positive staining of ap2a and twist1. Means and standard deviations for three injected sides were analyzed in each group. Statistics was done by using paired t test; * indicates p < 0.05.
	Figure 3 Cont. rad21mut disrupts the migration of neural crest cells (NCCs) in stage-25 Xenopus laevis. (a) The in situ hybridization pattern of twist1 was altered in the rad21mut-injected side. Panel i and ii show twist1 expression on both sides of the same non-injected embryo. The patterns around the optic cup (oc) and the neural crest migration streams into branchial arches (ba) are intact (indicated by a dash circle and dark solid lines, respectively). Panel iii and iv show the non-injected side and the injected side of the same rad21mut-injected embryo. The optic cup and branchial arch in panel iii are similar to those in panel i and ii, whereas these patterns are disrupted in panel iv. A total of 20 rad21mut-injected embryos were used, and 18 of them showed disrupted patterns in the rad21mut-injected side (18/20). (b) The in situ hybridization pattern of ap2a is altered in rad21mut-injected X. laevis embryos. Panel i shows ap2a is highly expressed in the head region and the spinal cord (sc) in the non-injected side. Panel ii shows the ap2a expression pattern is disrupted in the rad21mut-injected side. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). Panel iii and iv show higher magnifications of the non-injected side and rad21mut-injected side of another embryo. Similar to panel ii, disrupted neural crest migration streams are shown in panel iv. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). (c) Quantification of the mandibular areas showing positive staining of ap2a and twist1. Means and standard deviations for three injected sides were analyzed in each group. Statistics was done by using paired t test; * indicates p < 0.05.
	Figure 4 The expression of pitx2 was decreased by injecting rad21mut into X. laevis embryos. The transcription factor pitx2 is highly abundant in periocular mesenchymes (POM) and was used as a marker of POM. The lens is indicated by a dash circle. (a) Panel i and iii show the in situ hybridization expression pattern of pitx2 (blue) in the non-injected side of two X. laevis embryos at stage 41. The other sides of all five embryos (5/5) injected with rad21mut mRNAs showed a lack of pitx2 staining in the cornea (C) area (panel ii and iv). Normal structures of lens and retina layers could be observed in both the non-injected and the rad21mut-injected sides. (b) The quantification of pitx2-positive areas in the non-injected control eyes and the rad21mut-injected eyes was done using ImageJ. *** A significant decrease of staining corresponding to 57.7% was detected (p = 0.003).
	Figure S1: The top six up- and top four down-regulated genes from the RNA sequencing results were selected for qPCR validation.
	Figure S2: Global chromosome organization of LCLs.
	Figure S3: Local chromosome structural changes in two differentially expressed genes.
	Figure S4: Rescue of disrupted twist1 and apa2 expression patterns.
	Table S6: Primer sequences used in this study.
	Figure 1. Clusters of expression profiles of differentially expressed genes (DEGs) and expression levels. (a) Six different expression clusters were obtained from downregulated genes. (b) One expression cluster was identified from the upregulated genes. (c) Expression level changes were detected in three functional groups in affected lymphoblastoid cell lines (LCLs). The expression levels of three PCDHG gene clusters (PCDHGA, PCDHGB, and PCDHGC) were upregulated in LCLs from affected members. Eight mitochondria-related genes and five MHC class II-related genes showed reduced expression levels in affected-members LCLs. The functional enrichment scores were obtained by using DAVID bioinformatics. Three significant functional groups with false discovery rate (FDR) < 0.05 were obtained, and the expression levels of genes within these groups are shown.
	Figure 2. Local chromosome organization changes were detected in LCLs. (a) The coding genes overlapping with insulation score change (ISC) bins were chosen for functional enrichment analysis. Significant functional groups (FDR < 0.05) identified were cell adhesion, membrane components, glucuronidation, and N,N-dimethylaniline monooxygenase activity. (b) Local chromosome structural changes and gene expression changes in LCLs. Four differentially expressed genes/gene clusters showed changes in isolation scores, indicating local chromosome structural alterations may lead to expression changes in LCLs. (c) RAD21R450C caused insulation score changes in PCDH gene clusters. Insulation score profile across the PCDH gene clusters. Both of these loci showed CTCF and RAD21 binding signals (indicated by red arrowheads) in GM12878 cells.
	Figure 3. rad21mut disrupts the migration of neural crest cells (NCCs) in stage-25 Xenopus laevis. (a) The in situ hybridization pattern of twist1 was altered in the rad21mut-injected side. Panel i and ii show twist1 expression on both sides of the same non-injected embryo. The patterns around the optic cup (oc) and the neural crest migration streams into branchial arches (ba) are intact (indicated by a dash circle and dark solid lines, respectively). Panel iii and iv show the non-injected side and the injected side of the same rad21mut-injected embryo. The optic cup and branchial arch in panel iii are similar to those in panel i and ii, whereas these patterns are disrupted in panel iv. A total of 20 rad21mut-injected embryos were used, and 18 of them showed disrupted patterns in the rad21mut-injected side (18/20). (b) The in situ hybridization pattern of ap2a is altered in rad21mut-injected X. laevis embryos. Panel i shows ap2a is highly expressed in the head region and the spinal cord (sc) in the non-injected side. Panel ii shows the ap2a expression pattern is disrupted in the rad21mut-injected side. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). Panel iii and iv show higher magnifications of the non-injected side and rad21mut-injected side of another embryo. Similar to panel ii, disrupted neural crest migration streams are shown in panel iv. A total of 11 rad21mut-injected embryos were used, and 9 showed disrupted staining patterns in the rad21mut-injected side (9/11, arrowhead). (c) Quantification of the mandibular areas showing positive staining of ap2a and twist1. Means and standard deviations for three injected sides were analyzed in each group. Statistics was done by using paired t test; * indicates p < 0.05.
	Figure 4. The expression of pitx2 was decreased by injecting rad21mut into X. laevis embryos. The transcription factor pitx2 is highly abundant in periocular mesenchymes (POM) and was used as a marker of POM. The lens is indicated by a dash circle. (a) Panel i and iii show the in situ hybridization expression pattern of pitx2 (blue) in the non-injected side of two X. laevis embryos at stage 41. The other sides of all five embryos (5/5) injected with rad21mut mRNAs showed a lack of pitx2 staining in the cornea (C) area (panel ii and iv). Normal structures of lens and retina layers could be observed in both the non-injected and the rad21mut-injected sides. (b) The quantification of pitx2-positive areas in the non-injected control eyes and the rad21mut-injected eyes was done using ImageJ. *** A significant decrease of staining corresponding to 57.7% was detected (p = 0.003).

References [+] :

Akaneya, Ephrin-A5 and EphA5 interaction induces synaptogenesis during early hippocampal development. 2010, Pubmed