TMEM25 is a Par3-binding protein that attenuates claudin assembly during tight junction development.

The tight junction (TJ) in epithelial cells is formed by integral membrane proteins and cytoplasmic scaffolding proteins. The former contains the claudin family proteins with four transmembrane segments, while the latter includes Par3, a PDZ domain-containing adaptor that organizes TJ formation. Here we show the single membrane-spanning protein TMEM25 localizes to TJs in epithelial cells and binds to Par3 via a PDZ-mediated interaction with its C-terminal cytoplasmic tail. TJ development during epithelial cell polarization is accelerated by depletion of TMEM25, and delayed by overexpression of TMEM25 but not by that of a C-terminally deleted protein, indicating a regulatory role of TMEM25. TMEM25 associates via its N-terminal extracellular domain with claudin-1 and claudin-2 to suppress their cis- and trans-oligomerizations, both of which participate in TJ strand formation. Furthermore, Par3 attenuates TMEM25-claudin association via binding to TMEM25, implying its ability to affect claudin oligomerization. Thus, the TJ protein TMEM25 appears to negatively regulate claudin assembly in TJ formation, which regulation is modulated by its interaction with Par3.

Hypoxia stabilizes the H2 O2 -producing oxidase Nox4 in cardiomyocytes via suppressing autophagy-related lysosomal degradation.

The hydrogen peroxide (H2 O2 )-producing NADPH oxidase Nox4, forming a heterodimer with p22phox , is expressed in a variety of cells including those in the heart to mediate adaptive responses to cellular stresses such as hypoxia. Since Nox4 is constitutively active, H2 O2 production is controlled by its protein abundance. Hypoxia-induced Nox4 expression is observed in various types of cells and generally thought to be regulated at the transcriptional level. Here we show that hypoxia upregulates the Nox4 protein level and Nox4-catalyzed H2 O2 production without increasing the Nox4 mRNA in rat H9c2 cardiomyocytes. In these cells, the Nox4 protein is stabilized under hypoxic conditions in a manner dependent on the presence of p22phox . Cell treatment with the proteasome inhibitor MG132 results in a marked decrease of the Nox4 protein under both normoxic and hypoxic conditions, indicating that the proteasome pathway does not play a major role in Nox4 degradation. The decrease is partially restored by the autophagy inhibitor 3-methyladenine. Furthermore, the Nox4 protein level is upregulated by the lysosome inhibitors bafilomycin A1 and chloroquine. Thus, in cardiomyocytes, Nox4 appears to be degraded via an autophagy-related pathway, and its suppression by hypoxia likely stabilizes Nox4, leading to upregulation of Nox4-catalyzed H2 O2 production.

GPR125 (ADGRA3) is an autocleavable adhesion GPCR that traffics with Dlg1 to the basolateral membrane and regulates epithelial apicobasal polarity.

The adhesion family of G protein-coupled receptors (GPCRs) is defined by an N-terminal large extracellular region that contains various adhesion-related domains and a highly-conserved GPCR-autoproteolysis-inducing (GAIN) domain, the latter of which is located immediately before a canonical seven-transmembrane domain. These receptors are expressed widely and involved in various functions including development, angiogenesis, synapse formation, and tumorigenesis. GPR125 (ADGRA3), an orphan adhesion GPCR, has been shown to modulate planar cell polarity in gastrulating zebrafish, but its biochemical properties and role in mammalian cells have remained largely unknown. Here, we show that human GPR125 likely undergoes cis-autoproteolysis when expressed in canine kidney epithelial MDCK cells and human embryonic kidney HEK293 cells. The cleavage appears to occur at an atypical GPCR proteolysis site within the GAIN domain during an early stage of receptor biosynthesis. The products, i.e., the N-terminal and C-terminal fragments, seem to remain associated after self-proteolysis, as observed in other adhesion GPCRs. Furthermore, in polarized MDCK cells, GPR125 is exclusively recruited to the basolateral domain of the plasma membrane. The recruitment likely requires the C-terminal PDZ-domain-binding motif of GPR125 and its interaction with the cell polarity protein Dlg1. Knockdown of GPR125 as well as that of Dlg1 results in formation of aberrant cysts with multiple lumens in Matrigel 3D culture of MDCK cells. Consistent with the multilumen phenotype, mitotic spindles are incorrectly oriented during cystogenesis in GPR125-KO MDCK cells. Thus, the basolateral protein GPR125, an autocleavable adhesion GPCR, appears to play a crucial role in apicobasal polarization in epithelial cells.

Hepatocyte polarity establishment and apical lumen formation are organized by Par3, Cdc42, and aPKC in conjunction with Lgl.

Hepatocytes differ from columnar epithelial cells by their multipolar organization, which follows the initial formation of central lumen-sharing clusters of polarized cells as observed during liver development and regeneration. The molecular mechanism for hepatocyte polarity establishment, however, has been comparatively less studied than those for other epithelial cell types. Here, we show that the tight junction protein Par3 organizes hepatocyte polarization via cooperating with the small GTPase Cdc42 to target atypical protein kinase C (aPKC) to a cortical site near the center of cell-cell contacts. In 3D Matrigel culture of human hepatocytic HepG2 cells, which mimics a process of liver development and regeneration, depletion of Par3, Cdc42, or aPKC results in an impaired establishment of apicobasolateral polarity and a loss of subsequent apical lumen formation. The aPKC activity is also required for bile canalicular (apical) elongation in mouse primary hepatocytes. The lateral membrane-associated proteins Lgl1 and Lgl2, major substrates of aPKC, seem to be dispensable for hepatocyte polarity establishment because Lgl-depleted HepG2 cells are able to form a single apical lumen in 3D culture. On the other hand, Lgl depletion leads to lateral invasion of aPKC, and overexpression of Lgl1 or Lgl2 prevents apical lumen formation, indicating that they maintain proper lateral integrity. Thus, hepatocyte polarity establishment and apical lumen formation are organized by Par3, Cdc42, and aPKC; Par3 cooperates with Cdc42 to recruit aPKC, which plays a crucial role in apical membrane development and regulation of the lateral maintainer Lgl.

Intramolecular interaction in LGN, an adaptor protein that regulates mitotic spindle orientation.

Proper mitotic spindle orientation requires that astral microtubules are connected to the cell cortex by the microtubule-binding protein NuMA, which is recruited from the cytoplasm. Cortical recruitment of NuMA is at least partially mediated via direct binding to the adaptor protein LGN. LGN normally adopts a closed conformation via an intramolecular interaction between its N-terminal NuMA-binding domain and its C-terminal region that contains four GoLoco (GL) motifs, each capable of binding to the membrane-anchored Gαi subunit of heterotrimeric G protein. Here we show that the intramolecular association with the N-terminal domain in LGN involves GL3, GL4, and a region between GL2 and GL3, whereas GL1 and GL2 do not play a major role. This conformation renders GL1 but not the other GL motifs in a state easily accessible to Gαi To interact with full-length LGN in a closed state, NuMA requires the presence of Gαi; both NuMA and Gαi are essential for cortical recruitment of LGN in mitotic cells. In contrast, mInsc, a protein that competes with NuMA for binding to LGN and regulates mitotic spindle orientation in asymmetric cell division, efficiently binds to full-length LGN without Gαi and induces its conformational change, enhancing its association with Gαi In nonpolarized symmetrically dividing HeLa cells, disruption of the LGN-NuMA interaction by ectopic expression of mInsc results in a loss of cortical localization of NuMA during metaphase and anaphase and promotes mitotic spindle misorientation and a delayed anaphase progression. These findings highlight a specific role for LGN-mediated cell cortex recruitment of NuMA.

Differential cell surface recruitment of the superoxide-producing NADPH oxidases Nox1, Nox2 and Nox5: The role of the small GTPase Sar1.

Transmembrane glycoproteins, synthesized at the endoplasmic reticulum (ER), generally reach the Golgi apparatus in COPII-coated vesicles en route to the cell surface. Here, we show that the bona fide nonglycoprotein Nox5, a transmembrane superoxide-producing NADPH oxidase, is transported to the cell surface in a manner resistant to co-expression of Sar1 (H79G), a GTP-fixed mutant of the small GTPase Sar1, which blocks COPII vesicle fission from the ER. In contrast, Sar1 (H79G) effectively inhibits ER-to-Golgi transport of glycoproteins including the Nox5-related oxidase Nox2. The trafficking of Nox2, but not that of Nox5, is highly sensitive to over-expression of syntaxin 5 (Stx5), a t-SNARE required for COPII ER-to-Golgi transport. Thus, Nox2 and Nox5 mainly traffic via the Sar1/Stx5-dependent and -independent pathways, respectively. Both participate in Nox1 trafficking, as Nox1 advances to the cell surface in two differentially N-glycosylated forms, one complex and one high mannose, in a Sar1/Stx5-dependent and -independent manner, respectively. Nox2 and Nox5 also can use both pathways: a glycosylation-defective mutant Nox2 is weakly recruited to the plasma membrane in a less Sar1-dependent manner; N-glycosylated Nox5 mutants reach the cell surface in part as the complex form Sar1-dependently, albeit mainly as the high-mannose form in a Sar1-independent manner.

Ric-8A, an activator protein of Gαi, controls mammalian epithelial cell polarity for tight junction assembly and cystogenesis.

Correct cyst morphogenesis of epithelial cells requires apical-basal polarization, which is partly regulated by mitotic spindle orientation, a process dependent on the heterotrimeric G protein subunit Gαi and its binding protein LGN. Here, we show that in three-dimensional culture of mammalian epithelial Madin-Darby canine kidney (MDCK) cells, the Gαi-activating protein Ric-8A is crucial for orientation of the mitotic spindle and formation of normal cysts that comprise a single layer of polarized cells with their apical surfaces lining an inner lumen. Consistent with the involvement of LGN, cystogenesis can be well organized by ADP-ribosylated Gαi, retaining the ability to interact with LGN, but not by the interaction-defective mutant protein Gαi2 (N150I). In monolayer culture of MDCK cells, functional tight junction (TJ) assembly, a process associated with epithelial cell polarization, is significantly delayed in Ric-8A-depleted cells as well as in Gαi-depleted cells in a mitosis-independent manner. Ric-8A knockdown results in a delayed cortical delivery of Gαi and the apical membrane protein gp135, and an increased formation of intercellular lumens surrounded by membranes rich in Gαi3 and gp135. TJ development also involves LGN and its related protein AGS3. Thus, Ric-8A regulates mammalian epithelial cell polarity for TJ assembly and cystogenesis probably in concert with Gαi and LGN/AGS3.

Ric-8A-mediated stabilization of the trimeric G protein subunit Gαi is inhibited by pertussis toxin-catalyzed ADP-ribosylation.

The heterotrimeric G protein subunit Gαi can be activated by G protein-coupled receptors and the cytosolic protein Ric-8A, the latter of which is also known to prevent ubiquitin-dependent degradation of Gαi. Here we show that the amounts of the three Gαi-related proteins Gαi1, Gαi2, and Gαi3, but not that of Gαq, are rapidly decreased by cell treatment with pertussis toxin (PTX). The decrease appears to be due to ADP-ribosylation of Gαi, because PTX treatment does not affect the amount of a mutant Gαi2 carrying alanine substitution for Cys352, the residue that is ADP-ribosylated by the toxin. The presence of endogenous and exogenous Ric-8A increases Gαi stability as shown in cells treated with the protein synthesis inhibitor cycloheximide; however, Ric-8A fails to efficiently stabilize ADP-ribosylated Gαi. The failure agrees with the inability of Ric-8A to bind to ADP-ribosylated Gαi both in vitro and in vivo. Thus PTX appears to exert its pathological effects at least in part by converting Gαi to an unstable ADP-ribosylated form, in addition to the well-known inability of ADP-ribosylated Gαi to transduce signals triggered by G protein-coupled receptors.

Structural basis of cofactor-mediated stabilization and substrate recognition of the α-tubulin acetyltransferase αTAT1.

The functions of microtubules are controlled in part by tubulin post-translational modification including acetylation of Lys4° in α-tubulin. αTAT1 (α-tubulin acetyltransferase 1), an enzyme evolutionarily conserved among eukaryotes, has recently been identified as the major α-tubulin Lys4° acetyltransferase, in which AcCoA (acetyl-CoA) serves as an acetyl group donor. The regulation and substrate recognition of this enzyme, however, have not been fully understood. In the present study, we show that AcCoA and CoA each form a stable complex with human αTAT1 to maintain the protein integrity both in vivo and in vitro. The invariant residues Arg¹³² and Ser¹6° in αTAT1 participate in the stable interaction not only with AcCoA but also with CoA, which is supported by analysis of the present crystal structures of the αTAT1 catalytic domain in complex with CoA. Alanine substitution for Arg¹³² or Ser¹6° leads to a drastic misfolding of the isolated αTAT1 catalytic domain in the absence of CoA and AcCoA but not in the presence of excess amounts of either cofactor. A mutant αTAT1 carrying the R132A or S160A substitution is degraded much faster than the wild-type protein when expressed in mammalian Madin-Darby canine kidney cells. Furthermore, alanine-scanning experiments using Lys4°-containing peptides reveal that α-tubulin Ser³8 is crucial for substrate recognition of αTAT1, whereas Asp³9, Ile4², the glycine stretch (amino acid residues 43-45) and Asp46 are also involved. The requirement for substrate selection is totally different from that in various histone acetyltransferases, which appears to be consistent with the inability of αTAT1 to acetylate histones.

The cell polarity protein mInsc regulates neutrophil chemotaxis via a noncanonical G protein signaling pathway.

Successful chemotaxis requires not only increased motility but also sustained directionality. Here, we show that, during neutrophil chemotaxis via receptors coupled with the Gi family of heterotrimeric G proteins, directional movement is regulated by mInsc, a mammalian protein distantly related to the Drosophila polarity-organizer Inscuteable. The GDP-bound, Gßγ-free Gαi subunit accumulates at the front of chemotaxing neutrophils to recruit mInsc-complexed with the Par3-aPKC evolutionarily conserved polarity complex-via LGN/AGS3 that simultaneously binds to Gαi-GDP and mInsc. Both mInsc-deficient and aPKC-blocked neutrophils exhibit a normal motile activity but migrate in an undirected manner. mInsc deficiency prevents neutrophils from efficiently stabilizing pseudopods at the leading edge; the stability is restored by wild-type mInsc, but not by a mutant protein defective in binding to LGN/AGS3. Thus, mInsc controls directional migration via noncanonical G protein signaling, in which Gßγ-free Gαi-GDP, a product from Gαi-GTP released after receptor activation, plays a central role.

Ubiquitination of the heterotrimeric G protein α subunits Gαi2 and Gαq is prevented by the guanine nucleotide exchange factor Ric-8A.

The cytosolic protein Ric-8A acts as a guanine nucleotide exchange factor for Gα subunits of the Gi, Gq, and G12/13 classes of heterotrimeric G protein in vitro, and is also known to increase the amounts of these Gα proteins in vivo. The mechanism whereby Ric-8 regulates Gα content, however, has not been fully understood. Here we show that Ric-8 Astabilizes Gαi2 and Gαq by preventing their ubiquitination. Ric-8A interacts with and stabilizes Gαi2, Gαq, Gα12, but not Gαs, when expressed in COS-7 cells. The protein levels of Gαi2 and Gαq appear to be controlled via the ubiquitin-proteasome degradation pathway, because these Gα subunits undergo polyubiquitination and are stabilized with the proteasome inhibitor MG132. The ubiquitination of Gαi2 and Gαq is suppressed by expression of Ric-8A. The suppression likely requires Ric-8A interaction with these Gα proteins; the C-terminal truncation of Gαq and Gαi2 completely abrogates their interaction with Ric-8A, their stabilization by Ric-8A, and Ric-8A-mediated inhibition of Gα ubiquitination.

Interaction of NuMA protein with the kinesin Eg5: its possible role in bipolar spindle assembly and chromosome alignment.

Bipolar spindle assembly in mitotic cells is a prerequisite to ensure correct alignment of chromosomes for their segregation to each daughter cell; spindle microtubules are tethered at plus ends to chromosomes and focused at minus ends to either of the two spindle poles. NuMA (nuclear mitotic apparatus protein) is present solely in the nucleus in interphase cells, but relocalizes during mitosis to the spindle poles to play a crucial role in spindle assembly via focusing spindle microtubules to each pole. In the present study we show that the kinesin-5 family motor Eg5 is a protein that directly interacts with NuMA, using a proteomics approach and various binding assays both in vivo and in vitro. During mitosis Eg5 appears to interact with NuMA in the vicinity of the spindle poles, whereas the interaction does not occur in interphase cells, where Eg5 is distributed throughout the cytoplasm but NuMA exclusively localizes to the nucleus. Slight, but significant, depletion of Eg5 in HeLa cells by RNA interference results in formation of less-focused spindle poles with misaligned chromosomes in metaphase; these phenotypes are similar to those induced by depletion of NuMA. Since NuMA is less accumulated at the spindle poles in Eg5-depleted cells, Eg5 probably contributes to spindle assembly via regulating NuMA localization. Furthermore, depletion of cytoplasmic dynein induces mislocalization of NuMA and phenotypes similar to those observed in NuMA-depleted cells, without affecting Eg5 localization to the spindles. Thus dynein appears to control NuMA function in conjunction with Eg5.

The WD40 protein Morg1 facilitates Par6-aPKC binding to Crb3 for apical identity in epithelial cells.

Formation of apico-basal polarity in epithelial cells is crucial for both morphogenesis (e.g., cyst formation) and function (e.g., tight junction development). Atypical protein kinase C (aPKC), complexed with Par6, is considered to translocate to the apical membrane and function in epithelial cell polarization. However, the mechanism for translocation of the Par6-aPKC complex has remained largely unknown. Here, we show that the WD40 protein Morg1 (mitogen-activated protein kinase organizer 1) directly binds to Par6 and thus facilitates apical targeting of Par6-aPKC in Madin-Darby canine kidney epithelial cells. Morg1 also interacts with the apical transmembrane protein Crumbs3 to promote Par6-aPKC binding to Crumbs3, which is reinforced with the apically localized small GTPase Cdc42. Depletion of Morg1 disrupted both tight junction development in monolayer culture and cyst formation in three-dimensional culture; apico-basal polarity was notably restored by forced targeting of aPKC to the apical surface. Thus, Par6-aPKC recruitment to the premature apical membrane appears to be required for definition of apical identity of epithelial cells.

Structural basis for interaction between the conserved cell polarity proteins Inscuteable and Leu-Gly-Asn repeat-enriched protein (LGN).

Interaction between the mammalian cell polarity proteins mInsc (mammalian homologue of Inscuteable) and Leu-Gly-Asn repeat-enriched protein (LGN), as well as that between their respective Drosophila homologues Inscuteable and Partner of Inscuteable (Pins), plays crucial roles in mitotic spindle orientation, a process contributing to asymmetric cell division. Here, we report a crystal structure of the LGN-binding domain (LBD) of human mInsc complexed with the N-terminal tetratricopeptide repeat (TPR) motifs of human LGN at 2.6-Å resolution. In the complex, mInsc-LBD adopts an elongated structure with three binding modules--an α-helix, an extended region, and a ß-sheet connected with a loop--that runs antiparallel to LGN along the concave surface of the superhelix formed by the TPRs. Structural analysis and structure-based mutagenesis define residues that are critical for mInsc-LGN association, and reveal that the activator of G-protein signaling 3 (AGS3)-binding protein Frmpd1 [4.1/ezrin/radixin/moesin (FERM) and PSD-95/Dlg/ZO-1 (PDZ) domain-containing protein 1] and its relative Frmpd4 interact with LGN via a region homologous to a part of mInsc-LBD, whereas nuclear mitotic apparatus protein (NuMA) and the C terminus of LGN recognize the TPR domain in a manner different from that by mInsc. mInsc binds to LGN with the highest affinity (K(D) ≈ 2.4 nM) and effectively replaces the Frmpd proteins, NuMA, and the LGN C terminus, suggesting the priority of mInsc in binding to LGN. We also demonstrate, using mutant proteins, that mInsc-LGN interaction is vital for stabilization of LGN and for intracellular localization of mInsc.