The DLG5 gene encodes a member of the membrane-associated guanylate kinase (MAGUK) family and participates in the regulation of several signaling pathways, including Hippo (see STK3, 605030; SHH, 600725; and TGFB, 190180). DLG5 also maintains cell polarity through interactions with beta-catenin (116806) and the vinexin (610795)-vinculin (193065) complex and plays an important role in cilia function (summary by Marquez et al., 2021).

Vertebrate homologs of the Drosophila discs large (dlg) gene are members of the MAGUK (membrane-associated guanylate kinase) family. See 602887. MAGUK proteins contain PDZ motifs, an SH3 domain, and a guanylate kinase (GUK)-homologous region. Both the PDZ and GUK domains are thought to contribute to protein-protein interactions. By searching an EST database for sequences related to Drosophila dlg, Nakamura et al. (1998) identified cDNAs encoding a novel human homolog. Northern blot analysis revealed that the 9.4-kb transcript was highly expressed in placenta and prostate, as well as in several other tissues, leading the authors to designate the gene PDLG (placenta and prostate DLG). An additional 8.8-kb PDLG mRNA was detected in thyroid. The predicted 859-amino acid PDLG protein contains 3 PDZ domains, an SH3 domain, and a GUK region. PDLG is 45% and 40% identical to DLG1 (601014) and Drosophila dlg, respectively. Western blot analysis of extracts of human prostate tissue and various cell lines showed that PDLG had an apparent molecular mass of 105 kD. Immunofluorescence experiments localized PDLG at the plasma membrane and cytoplasm, and it was expressed in the gland epithelial cells of normal prostate tissue but not in prostate cell lines.

Independently, Nagase et al. (1998) cloned DLG5, which they called KIAA0583. RT-PCR analysis showed high expression in placenta, moderate expression in kidney, prostate, testis, and ovary, and low expression in heart, brain, lung, liver, skeletal muscle, and small intestine. Little to no expression was detected in pancreas, spleen, and thymus.

Using mouse vinexin-beta (610795) as bait in a yeast 2-hybrid screen of a human placenta cDNA library, followed by 5-prime RACE, Wakabayashi et al. (2003) cloned DLG5, which they called LPDLG (long type of PDLG). The deduced 1,764-amino acid protein contains an N-terminal coiled-coil region, 4 PDZ domains, an SH3 domain, and a C-terminal GUK domain. Northern blot analysis detected a major 8.5-kb transcript that was highly expressed in placenta and more weakly expressed in brain, heart, skeletal muscle, spleen, kidney, liver, small intestine, and lung. Little to no expression was detected in colon, thymus, and peripheral blood leukocytes. Western blot analysis revealed high expression of LPDLG proteins with apparent molecular masses of 250 and 200 kD in porcine kidney cells, with lower expression in other cell lines, including HeLa cells.

By genomic sequence analysis, RT-PCR, and Northern blot analysis, Nechiporuk et al. (2007) identified an additional upstream coding exon in mouse and human DLG5. The full-length mouse and human DLG5 proteins contain 1,921 and 1,920 amino acids, respectively, and both have an N-terminal CARD domain, followed by a Duff domain, a coiled-coil domain, 4 PDZ domains, an SH3 domain, and a C-terminal GUK domain. DLG5 is evolutionarily conserved, with orthologs in Drosophila, chicken, and zebrafish.

Using a yeast 2-hybrid screen, Nakamura et al. (1998) determined that PDLG interacted with the GUK domain of p55 (MPP1; 305360), a palmitoylated erythrocyte membrane MAGUK protein. They suggested that PDLG and p55 form a heteromeric MAGUK complex at the plasma membrane and cluster various intracellular molecules to play roles in maintaining the structure of epithelial cells and transmitting extracellular signals to the membrane and cytoskeleton.

By mutation analysis, Wakabayashi et al. (2003) showed that the third SH3 domain of mouse vinexin-beta bound to a proline-rich sequence between the second and third PDZ domains of human LPDLG. Vinexin-beta and Lpdlg colocalized at sites of cell-cell contact in porcine kidney cells, and Lpdlg also colocalized with beta-catenin (CTNNB1; 116806), a major adherens junction protein. Coimmunoprecipitation experiments revealed that endogenous Lpdlg associated with beta-catenin in porcine kidney cells. In transfected COS-7 cells, a ternary complex was formed between LPDLG, beta-catenin, and vinexin-beta.

By radiation hybrid analysis, Nagase et al. (1998) mapped the DLG5 gene to chromosome 10. Using the same technique, Nakamura et al. (1998) refined the localization of the DLG5 gene to 10q23.

Yuksel-Vogel-Bauer Syndrome

In a 2-month-old girl, born of consanguineous parents, with Yuksel-Vogel-Bauer syndrome (YUVOB; 620703), Yuksel et al. (2019) identified a homozygous frameshift mutation in the DLG5 gene (604090.0001). The mutation, which was found by whole-exome sequencing and confirmed by Sanger sequencing, was present in heterozygous state in each unaffected parent. Functional studies of patient cells were not performed.

In a 7-month-old boy, born of unrelated parents (family IV), with YUVOB, Marquez et al. (2021) identified a homozygous nonsense mutation in the DLG5 gene (R821X; 604090.0002). The mutation, which was found by whole-exome sequencing and confirmed by Sanger sequencing, was present in heterozygous state in each unaffected parent. Expression of the R821X variant in dlg5-null Xenopus embryos failed to rescue abnormal pronephros development and did not significantly improve the ventriculomegaly phenotype, suggesting that it is a loss-of-function allele. Studies of patient cells were not performed. Marquez et al. (2021) identified 3 additional families with some overlapping clinical features associated with other variants in the DLG5 gene. The patients included a 21-week-old fetus (family I) with multiple severe congenital anomalies who carried a de novo heterozygous R249W variant that exerted a dominant-negative effect in Xenopus studies; a father and son (family II) with a mild renal phenotype who carried a heterozygous Q1509L variant, which was able to rescue the abnormal renal phenotype in dlg5-null Xenopus embryos, suggesting it was a functional allele; and a 7-year-old girl (family III) with isolated renal disease who carried compound heterozygous missense variants in the DLG5 gene (R166H and R1072C), that showed a possible loss-of-function effect in Xenopus complementation studies. Each of her unaffected parents was heterozygous for one of the mutations, both of which were present in the gnomAD database. Hamosh (2024) noted that the frequencies of the variants in these 3 additional families were too high in the gnomAD database (v4.0; 2/5/2024) for them to be considered pathogenic.

Associations Pending Confirmation

For discussion of an association between variation in the DLG5 gene and inflammatory bowel disease, see IBD20 (612288).

In Xenopus tropicalis, Marquez et al. (2021) found expression of the dlg5 gene in the brain, pronephros, heart, and neural tube. Morpholino- or CRISPR/Cas9-based knockdown of the dlg5 gene in Xenopus embryos resulted in defects in kidney and brain development, including anasarca, cystic kidneys, enlarged kidney tubules with fewer cilia, and communicating hydrocephalus with loss of midline structures. There was fluid flow within the cerebral ventricles, suggesting some residual ciliary function. Mutant animals showed loss of cilia from the surface of multiciliated epidermal cells due to failure of apical migration, and there was evidence of impaired Shh signaling compared to controls. These findings suggested disrupted cell polarity and impaired downstream signaling.

Nechiporuk et al. (2007) found that Dlg5 -/- mice had growth retardation fully penetrant hydrocephalus due to the closure of the aqueduct of Sylvius. Hydrocephalus was accompanied by loss of ependymal cells and disorganization of the subventricular stem-cell niche. Dlg5 -/- kidney exhibited disruption of epithelial cell polarity, loss of cilia, and cysts. Similarly, Dlg5 -/- neural progenitors exhibited loss of cell polarity and disorganization of apical junctional adhesion complexes. Dlg5 -/- progenitor cells showed impaired cadherin-catenin complex formation and decreased cell-surface levels of N-cadherin (CDH2; 114020). Likewise, primary mouse embryonic fibroblasts (MEFs) from Dlg5 -/- mice showed impaired cell-surface delivery and stabilization of the N-cadherin-catenin complex. Dlg5 localized to adherens junctions (AJs) and N-cadherin-containing vesicles and facilitated delivery of N-cadherin to AJs in wildtype MEFs. Further analysis indicated that Dlg5 bound syntaxin-4 (STX4; 186591) and appeared to facilitate targeted cadherin delivery by linking cadherin-catenin-carrying transport vesicles with the t-SNARE complex (see 600322) at the plasma membrane.