Deficient assembly and function of gap junctions in Trf1, a trafficking mutant of the human liver-derived cell line HuH-7.

The Trf1 cell line, selected from the human hepatoma cell line HuH-7, manifests altered trafficking of various plasma membrane proteins. In particular, there is a striking loss of State 2 asialoglycoprotein receptors. This cell line is shown here to also manifest defects in function and assembly of gap junctions comprising connexin43 (Cx43). No alteration of Cx43 expression or phosphorylation was apparent. Nevertheless, immunostaining of Cx43 revealed that fewer and smaller gap junctions were present at appositional membrane areas in Trf1 cells as compared with parental HuH-7. This correlated with a significant attenuation in gap junction-mediated communication between Trf1 cells as demonstrated by markedly decreased dye transfer and their reduced ability to propagate mechanically evoked Ca(2+) waves. Isoelectric focusing (IEF) of Cx43 in HuH-7 cells indicated that the pIs of this protein were significantly lower than that predicted from its amino acid sequence; no differences in pI were evident in Cx43 from Trf1 cells and the HuH-7 cell line. The effects of the Trf1 mutation on assembly and function of gap junctions indicate that this mutation influences trafficking of Cx43. Connexins differ in several respects from other membrane proteins thus far analyzed in Trf1 mutants: gap junctions localize exclusively to the lateral cell surface; they are not glycoproteins; and they do not play a role in endocytic pathways. The disruption of trafficking of Cx43 by this mutation suggests that the Trf1 phenotype is a defect at a common point along the trafficking pathway of cell-surface proteins, irrespective of their ultimate destination on the cell surface or their glycosylation profile.

Localization of procollagen I in the lysosome/endosome system of human fibroblasts.

A significant amount of newly synthesized collagen is degraded intracellularly rather than secreted, but there is controversy about whether this process occurs in the lysosomes. We addressed this problem using confocal microscopy and immunofluorescence imaging to study the distribution of procollagen I in the Golgi and the lysosome/endosome system of cultured human fibroblasts. Cells were incubated under basal conditions and then permeabilized and exposed to fluorescently tagged probes for procollagen, Golgi markers (Helix pomatia binding protein or beta-coatamer protein), and lysosome/endosome markers (cathepsin B or LAMP-2). Strong signals for procollagen codistributed with the Golgi and lysosome/endosome markers. Of note, many structures were positive for procollagen and lysosome/endosome markers but not for Golgi markers. When cells were incubated with the proline analog cis-hydroxyproline, which inhibits correct triple helix formation and increases intracellular degradation, the amount of procollagen codistributing with the lysosome/endosome markers increased greatly. Similar results were obtained in I-cells, which do not have functioning lysosomal hydrolases. These findings strongly indicate that the lysosome/endosome system participates in the intracellular degradation of newly synthesized procollagen and that trafficking of procollagen to the lysosome/endosome system does not depend on the cells having active lysosomal hydrolases. We present a model that integrates our findings with other work and resolves inconsistencies in the literature. This model postulates the existence of three separate degradation paths for newly synthesized procollagen. In addition to the endosome/lysosome system, degradation also takes place in the proximal region of the secretory pathway such as the endoplasmic reticulum, cis-Golgi network, or cis-Golgi and in a distal region of the secretory pathway such as the trans-Golgi or trans-Golgi network.

Brefeldin A inhibits degradation as well as production and secretion of collagen in human lung fibroblasts.

Brefeldin A (BFA) inhibits protein secretion, collapses the Golgi complex into the endoplasmic reticulum (ER), causes redistribution of processing enzymes normally resident in the Golgi to the ER, and uncouples the proximal and distal regions of the secretory pathway. We used BFA to determine where intracellular degradation of newly synthesized collagen degradation occurs. In normal human fetal lung fibroblasts, BFA (50 ng/ml) completely blocked collagen secretion and reduced collagen production by two-thirds. In cells synthesizing collagen under normal conditions, intracellular degradation was about 16%; BFA (50 ng/ml) reduced degradation to less than 5%. In cells induced to synthesize structurally abnormal collagen (by incubation with the proline analog cis-hydroxyproline), degradation was approximately 33%; BFA reduced this level to less than 10%. When the y axes were scaled appropriately, the dose-response curves for collagen degradation +/- cis-hydroxyproline versus BFA concentration coincided. A pulse-chase experiment demonstrated that BFA did not inhibit hydroxylation of prolyl residues, a major posttranslational modification of collagen that occurs in the ER, and that inhibition of degradation was independent of inhibition of collagen synthesis. Immunofluorescence examination revealed that BFA redistributed Golgi glycoproteins to the ER. At the ultrastructural level, Golgi complex could not be found in fibroblasts exposed to BFA for 1 h; however, clusters of small vesicles were observed. A different structure, comprising one or two lamellae and resembling a partial Golgi complex, was observed in cells incubated with BFA for 6 h. This structure was adjacent to ER but far from the nucleus. In addition, the ER was devoid of ribosomes. The inhibition of intracellular collagen degradation by BFA indicates that collagen degradation does not occur in the ER. Rather, it suggests that collagen degradation occurs beyond the BFA block, perhaps in the trans-Golgi network.

Collagen degradation in I-cells is normal.

There is evidence that lysosomal proteases mediate the intracellular degradation of structurally abnormal collagen. I-Cell disease (Mucolipidosis II) is characterized by marked deficiency of many lysosomal hydrolases, including the collagenolytic enzyme cathepsin B. The experiments reported here tested the hypothesis that degradation of abnormal collagen would be severely impaired in I-cells. Skin fibroblasts from 3 patients with I-cell disease were incubated with and without cis-hydroxyproline, a proline analog that causes structural abnormalities in collagen, and [14C]proline. The amount of [14C]hydroxyproline in a low molecular weight fraction relative to total [14C]hydroxyproline was used as a measure of intracellular collagen degradation. Levels of degradation were significantly higher in I-cells exposed to cis-hydroxyproline than in cells incubated without the analog. Similar data were obtained for normal human fetal lung fibroblasts incubated under the same conditions. Degradation of [125I]-epidermal growth factor was used to assess the functionality of the lysosomal pathway for protein degradation, and it was much lower in I-cells than in normal cells. It can be concluded that a completely functional complement of lysosomal enzymes is not necessary for structurally abnormal collagen to be degraded intracellularly; the data suggest that a nonlysosomal pathway exists.