Cardiovascular Research

Cardiovascular Research 2004 62(2):256-267; doi:10.1016/j.cardiores.2003.12.021
© 2004 by European Society of Cardiology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Request Permissions Google Scholar Articles by Berthoud, V. M Articles by Beyer, E. C Search for Related Content PubMed PubMed Citation Articles by Berthoud, V. M Articles by Beyer, E. C Social Bookmarking          What's this?

Copyright © 2004, European Society of Cardiology

Pathways for degradation of connexins and gap junctions

Viviana M Berthoud^*,a, Peter J Minogue^a, James G Laing^b and Eric C Beyer^a

^aDepartment of Pediatrics, Section of Hematology/Oncology, University of Chicago, 5841 S. Maryland Ave., MC 4060, Chicago, IL 60637, USA
^bDepartment of Internal Medicine, Washington University School of Medicine, St. Louis, MO 63110, USA

^* Corresponding author. Tel.: +1-773-834-1498; fax: +1-773-702-9881. Email address: vberthou{at}peds.bsd.uchicago.edu

Received 30 October 2003; revised 12 December 2003; accepted 18 December 2003

	Abstract

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Gap junctional proteins, connexins, and gap junctional plaques are short-lived. Three pathways for their degradation have been proposed: (1) misfolded/abnormally oligomerized connexins are retrogradely translocated and degraded by the proteasome through endoplasmic reticulum-associated degradation; (2) connexins (as monomers or oligomers) may traffic directly from an early secretory compartment to the lysosome for degradation without reaching the plasma membrane; (3) connexins within gap junction plaques are degraded by the lysosome after endocytotic internalization. Degradation of gap junction plaques is proteasome-dependent in some cell types. Degradation may be regulated by ubiquitinylation, phosphorylation, or polypeptide domains that act as sorting signals.

KEYWORDS Proteasome; Lysosome; Half-life; Intercellular communication; Connexin; Gap junction

	1. Introduction

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Gap junction channels allow intercellular propagation of current-carrying ions between excitable cells. They also allow intercellular transfer of small molecules including second messengers and metabolites. Gap junction channels are made of a family of proteins called connexins (Cx). Several stimuli that cause cellular stress or metabolic alterations can affect intercellular communication by altering the abundance of functional gap junction channels. Significant changes in gap junction function may alter the velocity or anisotropy of cardiac conduction or affect vasoregulation. Changes in gap junction abundance may be achieved by altering the number of gap junction channels (or plaques) at the plasma membrane. This, in turn, can be modified through regulation of connexin synthesis and/or degradation that could alter the half-life of connexin/gap junctions.

In the next sections, we will summarize the various experiments that have determined the half-lives of connexins and their regulation. Then, we will consider the data on different pathways of connexin and gap junction degradation; these will be summarized in a model. Finally, we will consider how degradation of connexins and gap junctions is regulated.

	2. Connexin half-life

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
The balance between the rate of synthesis and the rate of degradation of a protein determines its steady-state levels; consequently, the rate of synthesis equals the rate of degradation at steady-state. The time required for degradation (or synthesis) of half of the total amount of a protein at steady-state is termed its half-life. A protein's half-life is commonly determined based on pulse-chase experiments.

Because connexins are integral membrane proteins, their half-lives were expected to be long (>20 h). However, many experiments have demonstrated that connexins have relatively short half-lives. In cultured cells, the half-life determined for most connexins is 1.5–4 h (Table 1). Other experiments have suggested that the turnover of connexins may be similarly rapid in vivo. In pioneering experiments performed on preparations enriched in gap junctions isolated from mouse liver after intraperitoneal injection of Na[¹⁴C]-HCO₃, Fallon and Goodenough [1] determined that the major liver gap junction protein (probably Cx32) has a half-life of 5 h. More recently, Beardslee et al. [10] found that the half-life of Cx43 in perfused rat hearts was 1.3 h.

View this table: [in this window] [in a new window]   Table 1 Connexin half-lives

 
Experiments have also been performed in HeLa cells transfected with tetracysteine-tagged Cx43 using biarsenical derivatives of two different fluorophores applied with different intervals of time ("pulse-chase") [20]. After application of the first fluorophore, cells were allowed to continue growing in the absence of the fluorophore; any tetracysteine-Cx43 synthesized subsequently should react with the second fluorophore. When the time elapsed between application of the first and second fluorophores was 8 h, no labeling with the first fluorophore was detected at gap junctional plaques but only in small intracellular punctae (presumably endocytotic vesicles) [20]. These results imply that Cx43 has a very short half-life (i.e., 8 h must represent several Cx43 half-lives).

Lens gap junctions may represent an exception to the rapid turnover of gap junctions. In rat lens cultures, the half-life of Cx46 could not be determined [21]. In chicken lens cultures, the decay of pulse-labelled Cx56 followed a double exponential suggesting the presence of two pools of Cx56, one with a half-life of a few hours and the other of 2–3 days [13]. In chicken lens organ culture, approximately 70–80% of radioactively labeled Cx45.6 had been degraded after 72 h of chase [22]. However, after retroviral infection of chicken lens cultures with Cx45.6, a lens gap junction protein, the degradation rate of the protein was consistent with a half-life of 1.5 h [23].

	3. Regulation of connexin half-life and degradation rate

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
The short half-life determined for connexins has been used to argue that the degree of intercellular communication between cells may be regulated by changes in synthesis or degradation rates.

Studies performed on naturally occurring mutant connexins or mutants obtained by site-directed mutagenesis have started to elucidate the importance of different amino acid residues or polypeptide domains in regulating connexin degradation. Several non-functional Cx32 mutants associated with X-linked Charcot-Marie Tooth disease (CMTX, a peripheral neuropathy) are retained in intracellular compartments [24]. These mutants show a rate of degradation in transfected cells that is at least as fast as that of wild type Cx32 [16]. In contrast, a mutant Cx50 associated with cataracts (Cx50P88S) accumulates in transfected cells and shows decreased degradation when compared with wild type Cx50 [25]. A decrease in the rate of Cx43 degradation is observed after site-directed mutagenesis of the tyrosine at position 286 (a residue not known to be phosphorylated) to alanine [18].

Most connexins are phosphoproteins, and specific phosphorylation events may regulate the rate of connexin degradation. Several bands of Cx43 can be resolved in SDS-containing polyacrylamide gels. These bands collapse to the one with the fastest electrophoretic mobility after treatment with phosphatase; consequently, this band has been designated as the non-phosphorylated form of Cx43. The bands with slower electrophoretic mobilities correspond to phosphorylated forms (Cx43-P₁ and Cx43-P₂) [4]. Cx43-P₂ has been associated with the formation of gap junctional plaques [26]. The tumor promoter phorbol ester, 12-O-tetradecanoylphorbol-13-acetate (TPA) is a direct activator of protein kinase C (PKC). The effects of TPA on several Cx43-expressing cell lines have been extensively studied. In most cases, TPA induces a decrease in gap junctional intercellular communication within minutes after its application associated with an increase in the phosphorylated Cx43 forms; a decrease in immunoreactive gap junctional plaques has been observed after TPA treatment for longer periods of time. In some cases, the effects have been associated with changes in the degradation rate of the connexin. Treatment of Novikoff hepatoma cells with TPA during chase periods of 1–4 h increased the degradation rate of Cx43 suggesting involvement of a Cx43 phosphorylation event after activation of a PKC-dependent pathway [6]. In lens cultures infected with a Cx45.6 retrovirus, the rate of degradation of Cx45.6 is decreased when the serine residue at position 363 (a phosphorylation site for caseine kinase II) is mutated to alanine, suggesting that phosphorylation of Ser363 regulates Cx45.6 degradation [23]. On the other hand, mutation of several serine residues in the carboxyl terminus of Cx45 (mainly, Ser381 and Ser382 or Ser384 and Ser385) to other amino acids decreases the half-life of Cx45 suggesting that phosphorylation of these serines stabilizes the protein in transfected cells [11]. The sequence of events leading from connexin phosphorylation to changes in degradation is unknown.

	4. Degradation of connexins and gap junctions

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Similar to other plasma membrane proteins, gap junction proteins are delivered to the cell surface through the secretory pathway. Connexins are translated in the rough endoplasmic reticulum (ER) and are co-translationally inserted into ER membranes. Most of them subsequently traffic through the Golgi compartment. On their transit from the ER through the trans-Golgi network, connexins oligomerize into hexamers (or connexons). The intracellular compartment in which oligomerization occurs is connexin-specific (e.g., Cx32 oligomerizes in the ER and Cx43 oligomerizes in a post-ER/Golgi compartment) [27,28]. Connexon-containing vesicles traffic from the trans-Golgi network to the plasma membrane. Once they reach and fuse with the plasma membrane, connexons move towards appositional membranes, dock with connexons in the adjacent cells and become part of gap junctional plaques (Fig. 1).

View larger version (44K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 1 Model for the degradation pathways of connexins. Connexins are synthesized in the endoplasmic reticulum (ER) and then traffic through the Golgi towards the plasma membrane (P.M.). Connexins are oligomerized into connexons in the ER (e.g., Cx32) or in a post-ER/Golgi (e.g., Cx43) compartment. Misfolded (or inappropriately oligomerized) connexins can be retrogradely transported (from post-ER/Golgi compartments; curved arrows between Golgi and ER compartments) to the ER and dislocated to be degraded by the proteasome; these proteins may be polyubiquitinylated ((Ub)n). Properly oligomerized connexins are transported in vesicles and then inserted into the plasma membrane. Some connexins/connexons (possibly ones that are inappropriately folded/oligomerized) traffic from an early secretory compartment directly to lysosomes [52] (arrows between ER/Golgi intermediate or Golgi compartments and the lysosome). Connexons at the plasma membrane move towards appositional membranes to dock with connexons in the plasma membrane from the adjacent cell to form gap junction channels. Then, gap junction channels cluster to form gap junctional plaques. Connexins at gap junctional plaques may contain tyrosine-based sorting signals (Y286) [18] or other sorting signals or may be modified by ubiquitin (Ub) or by phosphate (P). Then, these modifications/signals must be recognized and induce internalization (endocytosis) of a portion of the gap junction plaque (a process that likely depends on a functional proteasomal pathway). Internalized gap junctions fuse with lysosomes, and degradation of connexins takes place. The exact role of the proteasomal pathway in degradation of gap junction plaques is not yet clear. It has been postulated that: (a) the proteasome pathway can act together with the lysosome in a sequential or parallel manner [51], (b) there is an unknown short-lived protein that targets the conformationally mature protein for proteasomal degradation [53], or (c) that the proteasome is involved in the degradation of a protein that participates in internalization of gap junctions [52].

 
The fate of gap junction proteins has been followed in live cells transfected with green fluorescent protein (GFP)-connexin chimeras [17,29–32]. Time-lapse microscopy has shown that GFP-connexin chimeras are delivered to the non-junctional plasma membrane from the Golgi compartment in vesicles traveling along microtubules [32]. Then, GFP-labeled connexons move laterally until they reach the edges of gap junctional plaques [32]. The plasma membrane in these cells shows diffuse fluorescence, except at regions of membrane apposition where fluorescence is brighter [30]. Some gap junctional plaques are relatively immobile [29] while others are very mobile [17], fusing with each other or segregating from one another. Sometimes junctional fragments are taken into the cytoplasm [17,31].

The endoplasmic reticulum contains a number of chaperones involved in the proper folding and oligomerization of proteins. If newly synthesized proteins are not correctly folded, they are expelled from the ER where they are degraded in a proteasome-dependent manner through a process called ER-associated degradation. In some cases, ER-associated degradation is regulated by ubiquitinylation; in some cases, it is not. Proper folding of connexins (which likely requires the participation of several chaperones) may be inefficient. Proper oligomerization of connexins is also required for formation of hemichannels and their transport to the plasma membrane. Folding and oligomerization of connexins appear to be important steps for quality control systems; connexins that are not successfully folded or oligomerized may be targeted for degradation.

Two major pathways for the degradation of cellular proteins have been described, through the lysosome and the proteasome. The lysosomal pathway degrades endocytosed material including membrane proteins, and the proteasomal pathway is the route of degradation for cytosolic and nuclear proteins. The ubiquitin-activating system and/or the proteasome pathway are also involved in endocytosis of some membrane proteins, including the epithelial Na⁺ channel (ENaC) and growth hormone receptors (reviewed in Refs. [33,34]).

	5. Evidence for the involvement of the lysosomal pathway in connexin/gap junction degradation

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Participation of endocytosis and lysosomes in the degradation of gap junctional plaques was proposed more than 20 years ago after the observation of intracellular vesicle-like double-membrane structures resembling gap junctions in transmission electron micrographs. These structures were termed annular gap junctions. The first images of these structures were published by Bjorkman [35] from granulosa cells of the ovarian follicle. Annular gap junctions have been observed in several different cell types [36–42]. They contain the spacing between particles observed in gap junctions [36,38], and they can be decorated with anti-connexin antibodies [31,42–44]. Various approaches were used to verify that some of the annular gap junction profiles corresponded to internalized gap junctions instead of cross-sections of interdigitations or invaginations of gap junctional plaques including freeze-fracture replicas, serial sectioning or lack of labeling of the intercellular gap by lanthanum [36,38,40,41]. These studies led to the proposal that gap junctional plaques (or fractions of them) were internalized by endocytosis. Moreover, during differentiation of the chick otocyst sensory epithelium, annular gap junctions can be found in phagolysosomes [45]. Furthermore, bristle coats (resembling clathrin coats) were found in invaginations of the junctional membrane in granulosa cells [37] and Cx43 in gap junctions was identified in close proximity to clathrin-coated pits in Cx43-transfected C6 glioma cells [42]. Co-localization of Cx43 and clathrin has been observed in adult guinea pig cardiomyocytes [46]. Once the gap junctional plaque vesicle is in the cytoplasm, it appears to be surrounded by a halo of 4–7-nm filaments believed to be actin [37]. Double immunofluorescence suggested a close association between intracellular Cx43 in vesicle-like structures and actin or myosin II filaments in SW-13 human adrenocortical tumor cells [47]. Moreover, the abundance of these structures was decreased after treatment with cytochalasin B, a drug that disrupts microfilaments, consistent with a role for actin microfilaments in their formation [47]. However, coating on annular gap junctions has not been observed in all systems, suggesting that there may be variations in the signals/mechanisms leading to internalization and degradation.

More recently, endocytosis and internalization of gap junctions have been observed in cells expressing GFP-connexin chimeras [17,31]. Formation of annular gap junctions has also been shown in HeLa cells transfected with tetracysteine-tagged Cx43. Using this construct, it has been observed that internalization of gap junctional plaques occurs in the center of the plaque while newly formed channels are incorporated at the periphery of gap junctional plaques [20]. A similar conclusion can be drawn from data obtained by studying fluorescence recovery after photobleaching using a Cx43-GFP chimera [32].

After being internalized, gap junctions may fuse with lysosomes [39,41,42] and be degraded by lysosomal enzymes. Partially lysed annular and fragmented junctional profiles have been observed in several cell types [39,41,42,48]. Acid phosphatase has been localized in gap junction-like structures in the cytoplasm and in compound structures that may represent an intermediate step in their degradation [36,42,49].

A pharmacological approach using inhibitors of the lysosomal pathway has further supported the participation of this proteolytic pathway in degradation of gap junctional plaques and connexins. These studies have looked at the levels, distribution, immunoblot pattern, and/or the rate of degradation of connexins.

In several cell types, treatment with lysosomal inhibitors induces an increase in the levels of connexins [12,18,50–52] and a decrease in their rates of degradation [18,53]. Treatment of several Cx43-expressing cell lines with lysosomal inhibitors also increased immunoreactive intracellular staining and decreased immunostaining of gap junctional plaques at the plasma membrane [18,51–53]. A similar redistribution was observed in HeLa cells transfected with wild type human Cx50 after treatment with a lysosomal inhibitor (Fig. 2B). These results are consistent with the expectations for a treatment that blocks lysosomal function, but not gap junction internalization or fusion with lysosomes.

View larger version (41K): [in this window] [in a new window] [Download PowerPoint slide]   Fig. 2 Effects of protease inhibitors on the distribution of wild type human Cx50. HeLa cells that had been stably transfected with human wild type Cx50 were left untreated (A), or they were treated with 100 µM chloroquine for 48 h (B), or 2 µM epoxomicin (a proteasomal inhibitor) for 24 h (C). Untreated cells showed immunoreactive Cx50 at appositional membranes (arrows). In cells treated with chloroquine, most of the immunoreactive Cx50 was found in vesicle-like compartments within the cell (arrowheads) while the immunoreactivity at appositional membranes virtually disappeared. After treatment with epoxomicin, Cx50 immunoreactivity collapsed into a single large area in the cytoplasm (arrowheads); some staining at appositional membranes could still be observed (arrows). Bar represents 14.3 µm.

 
However, not all cell types behave similarly. In WB-F344 rat liver epithelial cells, levels of Cx43 and immunoreactivity at appositional membranes were unaffected after treatment with chloroquine [54]. Similarly, in cultured neonatal rat cardiac myocytes and in perfused adult rat hearts treated with lysosomal inhibitors for 3–4 h, Cx43 immunoreactivity remained at intercellular appositions [10,12]. The number of gap junctions per cell area and the length of the gap junction profile were both significantly increased in cardiac myocytes treated with lysosomal inhibitors [12]. Moreover, in adult rat hearts perfused with lysosomal inhibitors, the increase in Cx43 levels was predominantly in the phosphorylated forms of Cx43 [10]. Treatment of BWEM cells (a rat heart-derived cell line) with Brefeldin A (BFA), a drug that disrupts the Golgi and protein trafficking to the plasma membrane, led to a decrease in Cx43 levels [51]. Treatment of these cells with lysosomal inhibitors prevented the BFA-induced decrease in Cx43 levels, and the Cx43 immunoreactivity was found in an intracellular vesicular compartment [51]. Cx43-containing gap junction plaques were sparse in BCIR-M1R_k cells (a rat mammary tumor cell line that assembles endogenous Cx43 into functional gap junctional plaques) treated with NH₄Cl (a lysosomal inhibitor) and BFA for 6 h [52]. These treated cells showed a decrease in the relative abundance of phosphorylated Cx43 forms that was comparable to that of cells treated with BFA alone, but significantly lower than that observed in cells treated with NH₄Cl alone [52]. Treatment of Cx43-retrovirus infected MDA-MB-231 cells (a human breast tumor cell line, poorly coupled even after infection) with lysosomal inhibitors induced an increase in total Cx43 levels without changing the ratio of phosphorylated/non-phosphorylated Cx43 forms [52]. In these cells, Cx43 localized to lysosomes; treatment with BFA did not change the cellular distribution of Cx43. Yet, in SKHep1 cells transfected with wild type Cx43, treatment with lysosomal inhibitors induced an increase in the non-phosphorylated form of Cx43 and the protein showed an intracellular localization [18]. Transfection of these cells with a mutant Cx43 in which the tyrosine residue at position 286 (Tyr286) was mutated to alanine rendered the Cx43-gap junctional plaques insensitive to treatment with lysosomal inhibitors or BFA. Because Tyr286 in Cx43 is part of a putative tyrosine-based sorting signal (YxxN, where N corresponds to a hydrophobic residue), it has been suggested that Tyr286 is involved in targeting Cx43 for degradation by the lysosomal pathway [18].

	6. Evidence for the involvement of the proteasomal pathway in connexin/gap junction degradation

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Participation of the proteasomal pathway in the degradation of Cx43 was first demonstrated by Laing and Beyer [50]. In CHO cells treated with the proteasomal inhibitor, N-acetyl-L-leucyl-L-leucinyl-norleucinal (ALLN), for 16 h, levels of Cx43 were six times those of untreated cells [50]. A similar pharmacological approach has been applied to several other cell lines expressing Cx43, including BWEM, NRK (a normal rat kidney cell line), S180L (mouse sarcoma cells stably transfected with the adhesion molecule L-CAM), L929 fibroblasts and BICR-M1R_k cells. An increase in the levels of Cx43 [51,52] and a decrease in the rate of Cx43 degradation are observed after treatment with proteasomal inhibitors [12,50–53]. In CHO cells, treatment with proteasomal inhibitors (for 6–7 h) induces a dramatic increase in the amount of Cx43 immunoreactivity at gap junctional plaques associated with an increase in the phosphorylated forms of Cx43 and an increase in intercellular dye transfer [53]. In other cell types, the effect of proteasomal inhibitors on staining intensity at gap junctional plaques is much less pronounced (e.g., NRK, S180L and BWEM cells, chicken lens epithelial cells, neonatal rat cardiac myocytes) [12,51,53]). In contrast, HeLa cells transfected with Cx32 showed reduced immunoreactivity at gap junctional plaques after treatment with lactacystin, a proteasomal inhibitor, for 16 h [55]. Proteasomal inhibitors were also able to prevent the BFA-induced decrease in immunodetectable Cx43 at appositional membranes in BWEM cells [51]. Moreover, in BICR-M1R_k cells, no difference in the number of immunoreactive plaques was observed after treatment with lactacystin [52]. Perfusion of adult rat hearts with ALLN induced an increase in Cx43 levels and immunoreactivity at appositional membranes [10]. Most of the Cx43 in homogenates from these hearts migrated as the non-phosphorylated form [10]. However, in MDA-MB-231 cells infected with Cx43-carrying retrovirus, treatment with lactacystin induced an increase in the relative amount of phosphorylated forms of Cx43 without changes in the Cx43 levels [52]. In contrast, treatment of BICR-M1R_k cells with lactacystin increased the levels of Cx43, but did not change the ratio between the phosphorylated and non-phosphorylated forms [52]. In addition, lactacystin treatment partially prevented the BFA-induced loss of immunoreactive gap junctional plaques; this was associated with a higher phosphorylated Cx43/non-phosphorylated Cx43 ratio compared to that determined for cells treated with BFA alone [52]. The participation of the proteasomal pathway in Cx32 degradation has also been demonstrated [16,56]. Because the rate of degradation of wild type Cx32 or Cx43 was not affected by a brief treatment with BFA, and connexin degradation could still be blocked by proteasomal inhibitors, it has been concluded that these connexins are subject to proteasome-mediated degradation in the ER [16].

One of the signals for connexin degradation may be protein modification by ubiquitinylation. Proteasomal degradation can be ubiquitin-dependent or -independent. Ubiquitinylation is a three-step process catalyzed by an ubiquitin-activating enzyme (E1, the rate-limiting step), an ubiquitin-conjugating enzyme (E2) and an ubiquitin-protein ligase (E3). Recent work has shown that lysosomal degradation of some membrane proteins is initiated by ubiquitin modification (reviewed in Ref. [33]). Modification of Cx43 by ubiquitin has been demonstrated, and degradation of Cx43 in cells containing a temperature sensitive mutant E1 is impaired at the restrictive temperature [50]. Moreover, Rütz and Hülser [57] have demonstrated by immunogold staining of freeze-fracture replicas that a high proportion of gap junctional plaques binds anti-ubiquitin antibodies. Thus, it is possible that degradation of Cx43-containing gap junctional plaques is at least in part ubiquitin-dependent.

Phosphorylation may be a pre-requisite for ubiquitinylation as shown for ligand-induced uptake of several membrane proteins (e.g., Ste2p, ENaC) (reviewed in Ref. [58]). Phosphorylation of connexins has been implicated in connexin degradation. Treatment of WBF344 cells with TPA for 1–2 h decreased levels of Cx43 and induced the appearance of a slower migrating Cx43 form in alkali-insoluble material [59]. Because this TPA effect occurred more rapidly than the Cx43 decrease induced by blocking protein synthesis with cycloheximide, it has been suggested that the TPA-induced phosphorylation led to degradation of Cx43 [59]. In cultured bovine lens epithelial cells, treated with TPA for 30 min and allowed to recover for 2 h, a marked decrease in levels of Cx43 was observed [60]. Addition of proteasomal inhibitors (but not lysosomal inhibitors) during the recovery phase prevented the TPA-induced decrease in Cx43 levels. In these cells, treatment with MG132, a proteasomal inhibitor, increased the abundance of immunoreactive Cx43 at appositional membranes and partially prevented the TPA-induced decrease in dye coupling [60].

In summary, both proteasomal and lysosomal pathways are involved in degradation of connexins and gap junctions. The predominance of one pathway over the other as well as the changes in connexin distribution and phosphorylation (as detected by immunoblot) induced by treatment with proteasomal or lysosomal inhibitors are cell type (organ)-specific.

	7. Model for connexin/gap junction degradation

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
The results reviewed in the previous sections can be incorporated into a model that describes the degradation of connexins and gap junctions and involves the participation of both the proteasome and the lysosome (Fig. 1). Connexins that are misfolded are dislocated from the ER membrane and degraded by the proteasome by ER-associated degradation (approximately 40% of newly synthesized Cx32 is degraded in this manner) [16]. In some cells, connexins may also traffic from an early secretory compartment directly to lysosomes for degradation without necessarily reaching the plasma membrane [52]. Gap junctional plaques are internalized (in part or in whole) from the plasma membrane. Internalized gap junctions fuse to lysosomes where connexins are degraded. The internalization process may be proteasome-dependent in some cell lines, but the mechanism has not yet been elucidated.

It has been hypothesized that cells differ in their intrinsic ability to cluster (or assemble) gap junction channels into functional plaques based on the differential response of Cx43-expressing cells to treatment with lysosomal or proteasomal inhibitors [53]. Based on this hypothesis, cells have been classified as "assembly-incompetent", "assembly-inefficient" or "assembly-efficient". Connexins in assembly-inefficient cells are highly sensitive to treatment with proteasomal inhibitors while connexins in assembly-efficient cells are sensitive to both lysosomal and proteasomal inhibitors. For example, in CHO cells (classified as assembly-inefficient), proteasomal inhibitors induced an elevation of Cx43 levels [50], an increase in immunoreactive Cx43 gap junctional plaques and dye coupling [53]; lysosomal inhibitors affected neither clustering nor function [53]. One possible interpretation of these data is that Cx43-containing gap junctional plaques in CHO cells are degraded mainly by the proteasomal pathway [50]. Because inhibitors of protein synthesis mimicked the proteasomal inhibitor-induced increase in immunoreactive Cx43 gap junctional plaques and gap junctional intercellular communication, it has been suggested that a short-lived protein is involved in targeting conformationally mature connexins for proteasomal degradation [53].

The direct participation of the proteasome in degradation of Cx43-containing gap junctional plaques is controversial. It is not entirely clear how a multi-protein complex like the proteasome could participate in the degradation of a complex of integral membrane proteins such as a gap junctional plaque. Internalization/endocytosis of other membrane proteins (e.g., ENaC, growth hormone receptor) requires an active ubiquitin-conjugating system, even though their degradation is thought to be mediated by lysosomes [33,58]. In the case of the growth hormone receptor, proteasomal inhibitors prevent both internalization and endosome-to-lysosome transport, but the receptor itself does not need to be ubiquitinylated [61]. In the case of the ENaC, ubiquitinylation of mature channels at the cell surface regulates its stability [62]. This could also be the case for connexins. The data showing the presence of anti-ubiquitin immunoreactivity at gap junctional plaques [57] and the effects of proteasomal inhibitors on immunoreactive connexins at gap junctional plaques suggest a proteasome-dependent step in degradation of gap junctional plaques [51,53]. The possibility that both the lysosome and the proteasome participate in gap junction plaque degradation has been proposed [51]. Participation of the proteasome could involve partial degradation of free cytoplasmic domains of connexins before degradation of the membrane-embedded domains by the lysosome. However, the immunoblot pattern of Cx43 from BICR-M1R_k cells treated with proteasomal or lysosomal inhibitors was nearly identical when using antibodies raised against the amino or the carboxyl terminus of Cx43, suggesting that neither terminus of Cx43 was selectively cleaved by the proteasome [52]. Based on these results (and the association of a phosphorylated form of Cx43 with gap junctional plaques), an alternative proposal has been made that the proteasome regulates the stability of phosphorylated Cx43 and internalization of Cx43 from the cell surface, without Cx43 being a direct substrate of the proteasome [52]. The tyrosine at position 286 as part of a tyrosine-based sorting signal has been implicated in internalization of gap junctions [18]. This tyrosine may also be part of a PY motif (xPPxY). However, based on mutation analysis, Thomas et al. [18] concluded that the PY motif plays at most a limited role in Cx43 turnover. Interestingly, a similar PY motif regulates endocytosis of the ENaC, a short-lived protein that is also subject to proteasomal and lysosomal degradation [33]. In ENaC, this PY motif is part of the sequence pPPxYxxL. It has been hypothesized that Nedd4, an ubiquitin-protein ligase, binds ENaC first and ubiquitinylates it; and, after its release, the PY motif is free to interact with adaptor proteins allowing endocytosis [33]. Although this hypothesis has not been proven, the striking similarity of the tyrosine-containing sequence of ENaC and the Tyr286-bearing sequence (SPPGYKLV) of Cx43 raises the possibility that a similar mechanism may be involved in endocytosis of Cx43-containing gap junctions.

However, it is not clear what effect(s) the proteasome would have on gap junction plaques other than degradation. It is possible that the sensitivity of Cx43-containing gap junctional plaques to proteasomal inhibitors is due to proteasomal degradation of other undefined molecules that regulate Cx43 turnover [52]. These "undefined molecules" (likely proteins) resemble the unidentified short-lived protein involved in targeting conformationally mature Cx43 for proteasomal degradation proposed by Musil et al. [53]. The main differences between these hypotheses are the role of the unidentified protein(s) and the pathway by which Cx43 is degraded.

Qin et al. [52] proposed a "by-pass pathway" for connexin degradation to explain the results obtained in Cx43-retrovirus infected MDA-MB-231 cells including the lysosomal localization of Cx43, the lack of Cx43 redistribution or changes in levels of Cx43 after treatment with BFA, the insensitivity of Cx43 levels to treatment with lactacystin, and their increase after treatment with lysosomal inhibitors, associated with a paucity of gap junction plaques and poor intercellular communication. In this pathway, newly translated connexins may be routed from an early secretory compartment to lysosomes without the protein ever reaching the plasma membrane. Interestingly, the availability of some membrane proteins at the cell surface can be regulated at the trans-Golgi network from where proteins can be diverted to lysosomes instead of being transported to the cell surface [34]. A polyubiquitin signal may specify this intracellular targeting [34].

	8. Regulation of connexin/gap junction degradation under physiological and pathological conditions

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
Several physiological processes are associated with assembly and disassembly (and/or degradation) of gap junctions.

During mitosis, the levels of intercellular communication decrease significantly [63]. A reduction in Cx32 has been observed in proliferating hepatocytes after partial hepatectomy and in cultured hepatocytes induced to proliferate [64–66]. Proliferation of hepatocytes is also associated with changes in gap junctional intercellular communication [66]. Studies of several Cx43-expressing cell lines indicate a redistribution of Cx43 immunoreactivity from appositional membranes to an intracellular location in mitotic cells (obtained by treatment with nocodazole, a drug that disrupts the integrity of microtubules) [67]. This redistribution was associated with the presence of a specific phosphoform of Cx43 that was dependent on an active p34cdc2/cyclin B kinase-dependent pathway. Upon release from mitotic block, this phosphoform gradually disappeared, a process that could be partially prevented by treatment with proteasomal and lysosomal inhibitors [67]. These results suggest that the decrease in intercellular communication observed during mitosis correlates with internalization and degradation of gap junctions.

Degradation of gap junctions appears to be involved in differentiation of the chick otocyst sensory epithelium, because morphologically detectable gap junctions are lost and the number of annular gap junctions increases just prior to and during the first appearance of hair cells [45]. Increased formation of annular gap junctions has also been observed after dissociation of adult rat myocardium [38] and after anoxia and ischemia in liver [68]. Barker et al. [69] have demonstrated that endocytosis of Cx43 due to disaggregation of ventricular muscle is associated with increased interactions of Cx43 with ZO-1.

Degradation of connexins is not necessarily an all-or-none phenomenon. Some connexins undergo partial proteolysis generating a truncated form of the connexin that may lack some regulatory sites. Cx32 is cleaved at the carboxyl terminus by calpain, and this is prevented by phosphorylation with protein kinase C [70]. The carboxyl terminus of lens Cx50 is cleaved by calpain in the lens nucleus [71]; this form can make functional channels that are significantly less sensitive to cytoplasmic acidification [72,73]. Cleavage of the carboxyl terminus of Cx45.6 by a caspase-3-like protease is developmentally regulated and is inhibited by phosphorylation of Cx45.6 at Ser363 [74].

Pesticides (e.g., DDT and lindane) induce decreased Cx43 levels, Cx43-P₂ and dye coupling in WB-F344 cells [54]. The decrease in Cx43 levels and alterations in the immunoblot pattern of Cx43 can be prevented by treatment with chloroquine; however, the lysosomal inhibitor did not prevent the effect on dye coupling [54]. Because colchicine, a drug that disrupts microtubules, prevented the lindane- and DDT-induced effects on levels of Cx43 and immunoreactive spots in the cytoplasm, it has been suggested that microtubules are involved in degradation of Cx43, probably by their involvement in trafficking of vesicles [54]. These results suggest that the DDT- and lindane-induced effects are due to internalization of gap junctions and lysosomal degradation of the connexin [54].

During wound healing and in lesioned vessels and cardiac myocardium, increased levels of the transforming growth factor, TGF-β1, are observed. Prolonged treatment of bovine aortic endothelial cells with TGF-β1 causes increased Cx43 expression, synthesis and content per cell [9], and accumulation of the protein in a lysosomal compartment [75]. This may result from TGF-β1 altering lysosomal functioning; however, no internalized gap junctions were observed within the cytoplasm or within lysosomes [75]. Thus, it is possible that under these conditions, Cx43 was routed directly to lysosomes before reaching the plasma membrane through the by-pass pathway proposed by Qin et al. [52].

Several pathological conditions have been associated with cardiac gap junction remodeling. The number of gap junctions per intercalated disk length and the number of cells connected by intercalated disks to a single myocyte are decreased in infarct border zones [76]. The gap junction surface area in hearts subjected to chronic ischemia and hypertrophy is significantly decreased [77]. Moreover, the estimated gap junction content per cell is decreased in ischemic ventricle [77]. Cardiac myocytes subjected to heat-shock show a decrease in Cx43 levels and immunoreactivity at gap junctional plaques [12]. This decrease is prevented by proteasomal and lysosomal inhibitors [12]. Cytosolic stress can decrease the degradation of connexins by ER-associated degradation and enhance formation of functional gap junctions [56]. Thus, it is very likely that remodeling of cardiac gap junctions under pathological conditions (or heart injury) involves degradation of pre-existing gap junctions and their re-formation at the same or different sites of membrane apposition depending on the type of insult. This type of remodeling may contribute to increased anisotropy (with different alterations of conduction velocities in longitudinal versus transverse directions) predisposing to re-entrant arrhythmias.

These results imply that metabolic alterations or disruption of cell contacts lead to gap junctional plaque internalization and eventually degradation. However, because rapid reformation of gap junctions has been observed in several systems (in some cases even in the presence of protein synthesis inhibitors) [78–81], it is still unknown whether internalized gap junctions can be reutilized. The signals and mechanisms that target formation and/or degradation of internalized gap junctions are largely unknown.

	9. Fate of disease-associated connexins

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
A substantial number of diseases are associated with mutant proteins that do not traffic properly. Some of these proteins (including the cystic fibrosis transmembrane conductance regulator) accumulate in the cytoplasm in detergent-insoluble structures called aggresomes [82–85]. Aggresomes may form when the proteasome becomes overloaded with misfolded proteins that it normally degrades [86,87]; small protein aggregates that are initially dispersed throughout the cell travel along microtubules towards the microtubule organizing center where they coalesce to form aggresomes. This process can be mimicked by treatment of cells with proteasomal inhibitors [82,83,85]. In most cases, aggresomes composed of connexins have not been observed, but proteasomal inhibitor treatment of cells transfected with wild type human Cx50 induces the formation of aggresomes (Fig. 2C). In contrast, a cataract-associated Cx50 mutant accumulates in intracellular structures in transfected cells that do not correspond to aggresomes [25].

Several CMTX-associated Cx32 mutants do not traffic properly and localize to the ER or Golgi compartments [24]. Some of these mutants are retrogradely transported to and dislocated from the ER to be degraded by the proteasome [16]. Proteasome substrates may or may not be ubiquitinylated. Because many misfolded proteins are targeted for proteasomal degradation by poly-ubiquitinylation, it is assumed that these Cx32 mutants are modified by ubiquitin; however, that remains to be demonstrated. Chloroquine treatment of HeLa cells transfected with CMTX-associated Cx32 mutants did not alter the distribution of ER-retained mutants, but caused intracellular accumulation of Cx32-positive granules for all Golgi-retained mutants. Some of these granules co-localized with a lysosomal-associated membrane protein (LAMP-2) [55]. These results suggest that Golgi-retained Cx32 mutants may degrade in the lysosome after following the by-pass pathway described by Qin et al. [52].

	10. Pitfalls and future directions

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 
The short half-life determined for connexins is still puzzling. Could the short half-life result from the two-dimensional distribution of the cells in culture as compared to the three-dimensional structure of the tissue? The few experiments performed in whole tissue argue against that. However, the question still remains, is there a very stable pool of connexin that does not label in the pulse-chase experiments? Because some experiments suggest the presence of a connexin pool readily available for reformation of gap junction plaques, does this pool incorporate label in pulse-chase experiments? Given that cells seem to have different abilities to form gap junction plaques, might the dynamics of gap junction plaques observed in the imaging studies be different in other cell types?

The current model cannot account for several results and needs to be refined. According to the model, treatment with lysosomal inhibitors induces accumulation of connexin in intracellular compartments. However, in cultured cardiac myocytes or perfused hearts treated with lysosomal inhibitors, Cx43 immunoreactivity is increased at intercellular appositions. Treatment of BWEM cells with BFA and proteasomal inhibitors shows an increase in staining at appositional membranes.

The current model for connexin/gap junction degradation implies that connexins that can be dislocated from the membrane are targets for proteasomal degradation while membrane-embedded connexins are targets for lysosomal degradation. The model also implies that more than one step in the gap junction biosynthetic/degradative pathway is subject to quality control, including connexin folding, oligomerization, and targeting to the plasma membrane as well as gap junction function, internalization and degradation. The signals, mechanisms and effectors (proteins/cofactors) that regulate these processes (targeting connexins or gap junctions for degradation) are largely unknown. The availability of membrane proteins at the cell surface may be regulated (in part) by diverting proteins directly from the trans Golgi network to the lysosomes [34]. This pathway is similar to the one proposed by Qin et al. [52], except that Cx43 may be diverted from an early secretory compartment. What controls whether connexins are directed to the plasma membrane or to lysosomes is not yet known, but it is possible that the proportion of the cell surface involved in appositional membranes may play a role.

The model has not taken into consideration the proteins that interact with connexins. Several proteins (e.g., ZO-1, and β-catenin, p120-catenin, occludin, claudin-1, caveolin-1) co-localize with connexins at gap junctional plaques, and, in some cases, connexins can be co-immunoprecipitated with antibodies against these proteins (or vice versa) [88–95]. Additional interacting proteins that need to be identified include chaperones that influence folding, trafficking and targeting to the plasma membrane. Other events, besides ubiquitinylation (possibly including phosphorylation) may be involved in targeting connexin/gap junction plaques for degradation. How is the balance between trafficking of connexins/connexons towards the plasma membrane versus targeting for degradation achieved? The intracellular and plasma membrane pools of connexins should be isolated and analyzed for mono- or poly-ubiquitinylation and phosphorylation. Such experiments, if performed under non-dissociating conditions, might also allow identification of proteins involved in targeting connexins for degradation by the proteasome or lysosome pathway.

The possibility that connexins internalize in lipid rafts could be entertained, since several connexins interact with caveolin-1 [93]. Interestingly, the TGF-β receptor is internalized both in lipid rafts and in clathrin-coated pits; while the former pathway is involved in receptor turnover, the latter pathway regulates TGF-β signaling [96]. A similar mechanism might have been adapted to specify the fate of internalized gap junctions (degradation versus re-utilization).

The interpretation of the results obtained with proteasomal and lysosomal inhibitors must consider effects of these drugs on degradation of other cellular proteins in addition to gap junctional proteins. Proteins that could be affected include those involved in trafficking and targeting connexins/connexons to the plasma membrane and those involved in their degradation.

	Notes

 
Time for primary review 23 days

	References

Top Abstract 1. Introduction 2. Connexin half-life 3. Regulation of connexin... 4. Degradation of connexins... 5. Evidence for the... 6. Evidence for the... 7. Model for connexin/gap... 8. Regulation of connexin/gap... 9. Fate of disease-associated... 10. Pitfalls and future... References

 

1. Fallon R.F, Goodenough D.A. Five-hour half-life of mouse liver gap-junction protein. J. Cell Biol. (1981) 90:521–526.[Abstract/Free Full Text]
2. Traub O, Look J, Paul D, Willecke K. Cyclic adenosine monophosphate stimulates biosynthesis and phosphorylation of the 26 kDa gap junction protein in cultured mouse hepatocytes. Eur. J. Cell Biol. (1987) 43:48–54.[ISI][Medline]
3. Traub O, Look J, Dermietzel R, Brümmer F, Hülser D, Willecke K. Comparative characterization of the 21-kD and 26-kD gap junction proteins in murine liver and cultured hepatocytes. J. Cell Biol. (1989) 108:1039–1051.[Abstract/Free Full Text]
4. Musil L.S, Cunningham B.A, Edelman G.M, Goodenough D.A. Differential phosphorylation of the gap junction protein connexin43 in junctional communication-competent and -deficient cell lines. J. Cell Biol. (1990) 111:2077–2088.[Abstract/Free Full Text]
5. Laird D.W, Puranam K.L, Revel J.P. Turnover and phosphorylation dynamics of connexin43 gap junction protein in cultured cardiac myocytes. Biochem. J. (1991) 273:67–72.[ISI][Medline]
6. Lampe P.D. Analyzing phorbol ester effects on gap junctional communication: a dramatic inhibition of assembly. J. Cell Biol. (1994) 127:1895–1905.[Abstract/Free Full Text]
7. Darrow B.J, Laing J.G, Lampe P.D, Saffitz J.E, Beyer E.C. Expression of multiple connexins in cultured neonatal rat ventricular myocytes. Circ. Res. (1995) 76:381–387.[Abstract/Free Full Text]
8. Laird D.W, Castillo M, Kasprzak L. Gap junction turnover, intracellular trafficking, and phosphorylation of connexin43 in Brefeldin A-treated rat mammary tumor cells. J. Cell Biol. (1995) 131:1193–1203.[Abstract/Free Full Text]
9. Larson D.M, Wrobleski M.J, Sagar G.D.V, Westphale E.M, Beyer E.C. Differential regulation of connexin43 and connexin37 in endothelial cells by cell density, growth, and TGF-β1. Am. J. Physiol. (1997) 272:C405–C415.[ISI][Medline]
10. Beardslee M.A, Laing J.G, Beyer E.C, Saffitz J.E. Rapid turnover of connexin43 in the adult rat heart. Circ. Res. (1998) 83:629–635.[Abstract/Free Full Text]
11. Hertlein B, Butterweck A, Haubrich S, Willecke K, Traub O. Phosphorylated carboxy terminal serine residues stabilize the mouse gap junction protein connexin45 against degradation. J. Membr. Biol. (1998) 162:247–257.[CrossRef][ISI][Medline]
12. Laing J.G, Tadros P.N, Green K, Saffitz J.E, Beyer E.C. Proteolysis of connexin43-containing gap junctions in normal and heat-stressed cardiac myocytes. Cardiovasc. Res. (1998) 38:711–718.[Abstract/Free Full Text]
13. Berthoud V.M, Bassnett S, Beyer E.C. Cultured chicken embryo lens cells resemble differentiating fiber cells in vivo and contain two kinetic pools of connexin56. Exp. Eye Res. (1999) 68:475–484.[CrossRef][ISI][Medline]
14. Hertzberg E.L, Sáez J.C, Corpina R.A, Roy C, Kessler J.A. Use of antibodies in the analysis of connexin 43 turnover and phosphorylation. Methods (2000) 20:129–139.[CrossRef][ISI][Medline]
15. Larson D.M, Seul K.H, Berthoud V.M, Lau A.F, Sagar G.D.V, Beyer E.C. Functional expression and biochemical characterization of an epitope-tagged connexin37. Mol. Cell Biol. Res. Commun. (2000) 3:115–121.[CrossRef][Medline]
16. VanSlyke J.K, Deschenes S.M, Musil L.S. Intracellular transport, assembly, and degradation of wild-type and disease-linked mutant gap junction proteins. Mol. Biol. Cell (2000) 11:1933–1946.[Abstract/Free Full Text]
17. Windoffer R, Beile B, Leibold A, Thomas S, Wilhelm U, Leube R.E. Visualization of gap junction mobility in living cells. Cell Tissue Res. (2000) 299:347–362.[CrossRef][ISI][Medline]
18. Thomas M.A, Zosso N, Scerri I, Demaurex N, Chanson M, Staub O. A tyrosine-based sorting signal is involved in connexin43 stability and gap junction turnover. J. Cell Sci. (2003) 116:2213–2222.[Abstract/Free Full Text]
19. Yamaguchi D.T, Ma D. Mechanism of pH regulation of connexin 43 expression in MC3T3-E1 cells. Biochem. Biophys. Res. Commun. (2003) 304:736–739.[CrossRef][ISI][Medline]
20. Gaietta G, Deerinck T.J, Adams S.R, et al. Multicolor and electron microscopic imaging of connexin trafficking. Science (2002) 296:503–507.[Abstract/Free Full Text]
21. Jiang J.X, Paul D.L, Goodenough D.A. Posttranslational phosphorylation of lens fiber connexin46: a slow occurrence. Invest. Ophthalmol. Vis. Sci. (1993) 34:3558–3565.[Abstract/Free Full Text]
22. Jiang J.X, Goodenough D.A. Phosphorylation of lens-fiber connexins in lens organ cultures. Eur. J. Biochem. (1998) 255:37–44.[ISI][Medline]
23. Yin X, Jedrzejewski P.T, Jiang J.X. Casein kinase II phosphorylates lens connexin 45.6 and is involved in its degradation. J. Biol. Chem. (2000) 275:6850–6856.[Abstract/Free Full Text]
24. Deschenes S.M, Walcott J.L, Wexler T.L, Scherer S.S, Fischbeck K.H. Altered trafficking of mutant connexin32. J. Neurosci. (1997) 17:9077–9084.[Abstract/Free Full Text]
25. Berthoud V.M, Minogue P.J, Guo J, et al. Loss of function and impaired degradation of a cataract-associated mutant connexin50. Eur. J. Cell Biol. (2003) 82:209–221.[CrossRef][ISI][Medline]
26. Musil L.S, Goodenough D.A. Biochemical analysis of connexin43 intracellular transport, phosphorylation, and assembly into gap junctional plaques. J. Cell Biol. (1991) 115:1357–1374.[Abstract/Free Full Text]
27. Musil L.S, Goodenough D.A. Multisubunit assembly of an integral plasma membrane channel protein, gap junction connexin43, occurs after exit from the ER. Cell (1993) 74:1065–1077.[CrossRef][ISI][Medline]
28. DasSarma J, Wang F.S, Koval M. Targeted gap junction protein constructs reveal connexin-specific differences in oligomerization. J. Biol. Chem. (2002) 277:20911–20918.[Abstract/Free Full Text]
29. Jordan K, Solan J.L, Dominguez M, et al. Trafficking, assembly, and function of a connexin43-green fluorescent protein chimera in live mammalian cells. Mol. Biol. Cell (1999) 10:2033–2050.[Abstract/Free Full Text]
30. Bukauskas F.F, Jordan K, Bukauskiene A, et al. Clustering of connexin 43-enhanced green fluorescent protein gap junction channels and functional coupling in living cells. Proc. Natl. Acad. Sci. U. S. A. (2000) 97:2556–2561.[Abstract/Free Full Text]
31. Jordan K, Chodock R, Hand A.R, Laird D.W. The origin of annular junctions: a mechanism of gap junction internalization. J. Cell Sci. (2001) 114:763–773.[Abstract]
32. Lauf U, Giepmans B.N.G, Lopez P, Braconnot S, Chen S.C, Falk M.M. Dynamic trafficking and delivery of connexons to the plasma membrane and accretion to gap junctions in living cells. Proc. Natl. Acad. Sci. U. S. A. (2002) 99:10446–10451.[Abstract/Free Full Text]
33. Rotin D, Kanelis V, Schild L. Trafficking and cell surface stability of ENaC. Am. J. Physiol. (2001) 281:F391–F399.[ISI]
34. Strous G.J, van Kerkhof P. The ubiquitin-proteasome pathway and the regulation of growth hormone receptor availability. Mol. Cell Endocrinol. (2002) 197:143–151.[CrossRef][ISI][Medline]
35. Bjorkman N.A. Study of the ultrastructure of the granulosa cells of the rat ovary. Acta Anat. (Basel) (1962) 51:125–147.[ISI][Medline]
36. Larsen W.J, Tung H.N. Origin and fate of cytoplasmic gap junctions in rabbit granulosa cells. Tissue Cell (1978) 10:585–598.[ISI][Medline]
37. Larsen W.J, Tung H.N, Murray S.A, Swenson C.A. Evidence for the participation of actin microfilaments and bristle coats in the internalization of gap junction membrane. J. Cell Biol. (1979) 83:576–587.[Abstract/Free Full Text]
38. Mazet F, Wittenberg B.A, Spray D.C. Fate of intercellular junctions in isolated adult rat cardiac cells. Circ. Res. (1985) 56:195–204.[Abstract/Free Full Text]
39. Gregory W.A, Bennett M.V.L. Gap junctions in goldfish preoptic ependyma: regional variation in cellular differentiation. Brain Res. (1988) 470:205–216.[Medline]
40. Severs N.J, Shovel K.S, Slade A.M, Powell T, Twist V.W, Green C.R. Fate of gap junctions in isolated adult mammalian cardiomyocytes. Circ. Res. (1989) 65:22–42.[Abstract/Free Full Text]
41. Vaughan D.K, Lasater E.M. Renewal of electronic synapses in teleost retinal horizontal cells. J. Comp. Neurol. (1990) 299:364–374.[CrossRef][ISI][Medline]
42. Naus C.C, Hearn S, Zhu D, Nicholson B.J, Shivers R.R. Ultrastructural analysis of gap junctions in C6 glioma cells transfected with connexin43 cDNA. Exp. Cell Res. (1993) 206:72–84.[CrossRef][ISI][Medline]
43. Spray D.C, Moreno A.P, Kessler J.A, Dermietzel R. Characterization of gap junctions between cultured leptomeningeal cells. Brain Res. (1991) 568:1–14.[CrossRef][ISI][Medline]
44. Dermietzel R, Hertberg E.L, Kessler J.A, Spray D.C. Gap junctions between cultured astrocytes: immunocytochemical, molecular, and electrophysiological analysis. J. Neurosci. (1991) 11:1421–1432.[Abstract]
45. Ginzberg R.D, Gilula N.B. Modulation of cell junctions during differentiation of the chicken otocyst sensory epithelium. D