A Chinese hamster ovary cell mutant with a heat-sensitive, conditional-lethal defect in vacuolar function.

We describe a mutant derived from Chinese hamster ovary cells that is offt-sensitive for viability and for resistance to certain protein toxins. This mutant, termed G.7.1, grows normally at 34 degrees C but does not grow in Dulbecco's modified Eagle's medium at 39.5 degrees C. However, when this medium is supplemented with FeSO4, the mutant cells will grow at the elevated temperature. At 39.5 degrees C, G.7.1 cells acquire resistance to diphtheria toxin, modeccin, and Pseudomonas aeruginosa exotoxin A, all of which are protein toxins that require endocytosis and exposure to a low pH within vesicles before they can invade the cytosol and kill cells. The properties of mutant G.7.1 could result from a heat-sensitive lesion that impairs vacuolar acidification. We assayed the ATP-stimulated generation of pH gradients across the membrane of vesicles in cell-free preparations from mutant and parental cells by the partitioning of acridine orange into acidic compartments and found that the acidification response of the mutant cells was heat-labile. Altogether the evidence suggests that G.7.1 cells contain a heat-sensitive lesion that impairs vacuolar acidification and that they fail to grow in normal medium at 39.5 degrees C because they cannot extract Fe+3 from transferrin, a process that normally requires exposing transferrin to a low pH within endosomal vesicles.

Three new complementation groups of temperature-sensitive Chinese hamster ovary cell mutants defective in the endocytic pathway.

We describe here the results of complementation studies with six mutant Chinese hamster ovary cells expressing temperature-sensitive lesions affecting the endocytic pathway. The mutants were crossed with representatives of the End1 and End2 complementation groups identified previously by Robbins et al. (J. Cell Biol. 99:1296-1308, 1984). Two mutants, G.8.1 and 31.1, were members of the End1 complementation group. One mutant, 25.2, was a member of the End2 complementation group. The other three mutants each defined new complementation groups, which we have designated End3 (mutant G.7.1), End4 (mutant V.24.1), and End5 (mutant 42.2). Previous work on mutants of the End1, End2, and End3 classes had shown that these mutants were defective in endosomal acidification. We prepared postnuclear supernatants from mutants harvested at the nonpermissive temperature and compared their acidification activities, assessed by ATP-stimulated quenching of acridine orange. Members of the End1, End2, and End2 groups had reduced acidification activity, correlating with the acidification defects known to be expressed by these mutants. Strain V.24.1 (End4) also expressed a 40% reduction in acidification activity, while strain 42.2 (End5) had no reduction of acidification activity.

The cytoplasmic domain of the polymeric immunoglobulin receptor contains two internalization signals that are distinct from its basolateral sorting signal.

The C-terminal cytoplasmic domain of the polymeric immunoglobulin receptor (pIgR) contains two tyrosine residues, Tyr668 and Tyr734. Previous work identifying Tyr734 as a critical residue in the endocytosis of the pIgR in Madin-Darby canine kidney (MDCK) cells also suggested that a second functional internalization signal was present (Breitfeld, P. P., Casanova, J. E., McKinnon, W. C., and Mostov, K. E. (1990) J. Biol. Chem. 265, 13750-13757). To test this hypothesis, Tyr668 and Tyr734 were mutated singly or together by oligonucleotide-directed mutagenesis of pIgR cDNA, and the mutants were expressed in MDCK cells. The amount of ligand internalized within 5 min from the basolateral membrane by the pIgR in which cytoplasmic tyrosines were mutated separately to Cys668 or Ser734 or together to Cys668, Ser734 was 58, 39, and 20%, respectively, of the internalized by the wild-type pIgR. The cytoplasmic and transmembrane domains of the pIgR, when joined to the external domain of the influenza virus hemagglutinin, retained the capacity to mediate rapid internalization. As with the full-length pIgR, mutation of either tyrosine in the chimera resulted in impairment of endocytosis, with mutation of Tyr734 having a significantly greater effect than mutation on Tyr668 on the initial rate of endocytosis (3 and 44% of control values, respectively). However, unlike the full-length pIgR, mutation of both tyrosines together in the chimera did not reduce internalization further. The two tyrosines in the cytoplasmic sequence of the pIgR, although widely separated in the linear amino acid sequence, both contribute to internalization of the protein, suggesting that both can function as internalization signals. In addition, the correlation between endocytosis and basolateral targeting of the pIgR in MDCK cells was investigated. Neither tyrosine of the cytoplasmic domain was necessary for basolateral targeting of the pIgR.

Evidence for penetration of diphtheria toxin to the cytosol through a prelysosomal membrane.

To kill mammalian cells, diphtheria toxin must be endocytosed and encounter a low pH within intracellular vesicles. The low pH initiates penetration of the catalytically active A fragment of the toxin through a membrane and into the cytosol where the A fragment arrests protein synthesis. To investigate whether penetration occurred through a prelysosomal or a lysosomal membrane, we studied the effect of low temperature on the entry of the toxin into the cytosolic and lysosomal compartments. The toxin arrested protein synthesis at 15 degrees C, indicating entry into the cytosol; however, access to lysosomes was apparently blocked at 15 degrees C, suggesting that the toxin had encountered a low pH before reaching lysosomes and had penetrated a prelysosomal membrane. To further investigate the possibility of prelysosomal acidification, we measured the time required for the toxin to encounter a low pH after endocytosis. Acidification occurred within 3 to 4 min after the toxin was internalized into vesicles. This interval is consistent with prelysosomal acidification since the entry of endocytosed ligands into secondary lysosomes usually takes more than 3 to 4 min.

The effects of foreign transmembrane domains on the biosynthesis of the influenza virus hemagglutinin.

Eleven chimeric proteins were created in which the transmembrane, the cytoplasmic, or both topological domains of the influenza virus hemagglutinin (HA) were replaced with those from five other glycoproteins. All of the chimeric HAs reached the cell surface but appeared to differ in the degree to which they were stably folded. Comparisons of the rates of folding, passage into the Golgi, and arrival at the plasma membrane of wild-type HA and the chimeric proteins suggest that formation of a stable HA trimer is not an absolute requirement for export from the endoplasmic reticulum. In addition, there appear to be at least two steps at which the rate of transport can be altered during exocytosis, one occurring before and the other after the trimming of oligosaccharides by Golgi mannosidases. Certain of the chimeras differed from HA in their ability to pass through each of these steps. Replacement of the HA transmembrane domain with the analogous sequences from other proteins affected folding and transport of the chimeric HAs in ways that suggest that the HA transmembrane sequences form a specific structure in the membrane that differs from that formed by analogous sequences from the other proteins.