The endoplasmic reticulum: integration of protein folding, quality control, signaling and degradation

The endoplasmic reticulum is the entry point into the secretory pathway. To acquire a correct conformation, secretory proteins encounter the endoplasmic reticulum molecular machines of folding, quality control, signaling and disposal, which function as an integrated mechanism. The creation of such a molecular network, spatially regulated, suggests how the endoplasmic reticulum promotes the release of correctly folded secretory proteins.

Phosphorylation by CK2 and MAPK enhances calnexin association with ribosomes

Calnexin was initially identified as an endoplasmic reticulum (ER) type I integral membrane protein, phosphorylated on its cytosolic domain by ER-associated protein kinases. Although the role of the ER luminal domain of calnexin has been established as a constituent of the molecular chaperone machinery of the ER, less is known about the role of the cytosolic phosphorylation of calnexin. Analysis by two-dimensional phosphopeptide maps revealed that calnexin was in vitro phosphorylated in isolated microsomes by casein kinase 2 (CK2) and extracellular-signal regulated kinase-1 (ERK-1) at sites corresponding to those for in vivo phosphorylation. In canine pancreatic microsomes, synergistic phosphorylation by CK2 and ERK-1 led to increased association of calnexin with membrane-bound ribosomes. In vivo, calnexin-associated ERK-1 activity was identified by co-immunoprecipitation. This activity was abolished in cells expressing a dominant-negative MEK-1. Activation of ERK-1 in cells by addition of serum led to a 4-fold increase in ribosome-associated calnexin over unstimulated cells. Taken together with studies revealing calnexin association with CK2 and ERK-1, a model is proposed whereby phosphorylation of calnexin leads to a potential increase in glycoprotein folding close to the translocon.

Association of insulin-degrading enzyme with a 70 kDa cytosolic protein in hepatoma cells

We have investigated the biosynthesis, subcellular location and expression of insulin-degrading enzyme (IDE). a type-I peroxisomal protease, in semi-permeabilized hepatoma cells using pulse-chase experiments, non-denaturing immunoprecipitation protocols and Northern-blot analyses. In HcpG2 cell lysates prepared from cells radiolabelled with Tran[35S]-label, immunoprecipitated IDE was observed immediately after a 5 min pulse and subsequently declined during chase with t1/2 of approx. 33 h. In addition to the 110 kDa IDE protein, a protein of 70 kDa (p70) was identified in radiolabelled immunoprecipitates when using a monoclonal anti-IDE antibody 9B12 under non-denaturing conditions. This same antibody did not recognize p70 on Western blots of whole-cell lysates nor in sequential immunoprecipitates of immunocomplex-bead eluates from anti-IDE immunoprecipitations. Likewise, cross-linking studies performed on intact HepG2 and H35 hepatoma cells in vivo revealed the existence of a hetero-oligomeric complex of 180 kDa in which IDE and p70 were physically associated. Digitonin-permeabilization studies in normal and 35S-labelled HepG2 cells have defined a predominant association of IDE and its associated protein p70 with cytosol (supernatant); only a minor amount of the protein IDE was detected in peroxisomes (cellular pellet). Immunoprecipitation of IDE from 35S-labelled cell lysates of normal and stably transfected Chinese hamster ovary cells overexpressing IDE failed to detect p70. Treatment of HepG2 cells with clofibrate, a peroxisome proliferator, resulted in a dose-dependent increase of the two human IDE transcripts of 3.6 and 3.2 kb. This effect was not accompanied by a similar change at the protein level, nor by a change in the subcellular location of the proteins IDE and p70. Based on these findings we propose that in hepatoma cells: (1) IDE mainly exists in a stable cytoplasmic pool that is unchanged in cells undergoing peroxisomal proliferation; and (2) p70 binding to IDE may serve to maintain the dual cytosolic and peroxisomal pools of IDE in a stable equilibrium.

Association of folding intermediates of glycoproteins with calnexin during protein maturation

Calnexin, an endoplasmic reticulum transmembrane protein, represents a new type of molecular chaperone that selectively associates in a transient fashion with newly synthesized monomeric glycoproteins in HepG2 cells. Calnexin only recognizes glycoproteins when they are incompletely folded. Dissociation of glycoproteins from calnexin occurs at different rates and is related to the time taken for their folding, which may then initiate their differential transport rates from the endoplasmic reticulum.

Organelle-specific phosphorylation. Identification of unique membrane phosphoproteins of the endoplasmic reticulum and endosomal apparatus

Highly purified endoplasmic reticulum fractions from rat liver and dog pancreas harbor membrane-associated kinases that phosphorylate integral membrane proteins of 90, 56, 35, and 15 kDa with [gamma-32P]GTP and of 90, 56, and 35 kDa with [gamma-32P]ATP. Of these, only the 35-kDa phosphoprotein was N-glycosylated. Screening of Golgi fractions, endosomes, plasma membranes, lysosomes, and mitochondria revealed phosphoproteins unique to each organelle. In particular, endosomes were found to harbor a 48-kDa extrinsic membrane protein and two or more integral membrane phosphoproteins of 30-35 kDa. None of these were N-glycosylated as judged by their insensitivity to digestion by N-glycosidase F and a lack of binding to concanavalin A or wheat germ agglutinin. Since the 30-35 kDa membrane phosphoproteins present in Golgi-free endosomal fractions were not detected in endosome-free, highly purified Golgi fractions and were found exclusively in horseradish peroxidase-containing endosomes as determined by the diaminobenzidine shift protocol, then these membrane phosphoproteins are unique to endosomes. Since membrane phosphoproteins unique to the endoplasmic reticulum have been shown to have important functional significance in calcium binding and as a membrane chaperone(s) (Wada, I., Rindress, D., Cameron, P.H., Ou, W.-J., Doherty, J.-J., II, Louvard, D., Bell, A.W., Dignard, D., Thomas, D.Y., and Bergeron, J.J.M. (1991) J. Biol. Chem. 266, 19599-19610; Ahluwalia, N., Bergeron, J.J.M., Wada, I., Degen, E., and Williams, D.B. (1992) J. Biol. Chem. 267, 10914-10918), then the unique endosomal phosphoproteins may serve equally important functions in addition to serving as novel markers for the organelle.

SSR alpha and associated calnexin are major calcium binding proteins of the endoplasmic reticulum membrane

GTP phosphorylation of rough microsomes in vitro is limited to four integral membrane proteins. Two of these, a phosphoprotein (pp90) and a phosphoglycoprotein (pgp35) were purified as a complex with two nonphosphorylated membrane glycoproteins, gp25H and gp25L. The authenticity of this complex was confirmed using two different purification procedures and by coimmunoprecipitation. By immunofluorescence a reticulated cytoplasmic network was revealed for the proteins which was similar to that for Louvard et al. (Louvard, D., Reggio, H., and Warren, G. (1982) J. Cell Biol. 92, 92-107) marker antisera which also recognized purified pp90 on immunoblots. Amino acid sequencing of peptides derived from pgp35 identified this protein as SSR alpha, an endoplasmic reticulum constituent as identified by cross-linking of translocating nascent chains (Görlich, D, Prehn, S., Hartmann, E., Herz, J., Otto, A., Kraft, R., Wiedmann, M., Knespel, S., Dobberstein, B., and Rapoport, T. A. (1990) J. Cell Biol. 111, 2283-2294). The sequence of gp25H was determined from cDNA clones and was identical with SSR beta identified by Görlich et al. (1990) as being tightly bound to SSR alpha. Sequencing of gp25L revealed no similarity of the deduced sequence with other proteins. However, pp90 revealed a high degree of sequence identity with the Ca(2+)-binding protein, calreticulin. 45Ca2+ overlay studies indicated that pp90 bound Ca2+ and the name calnexin is proposed. Surprisingly, pgp25 (SSR alpha) also bound Ca2+ although gp25H (SSR beta) and gp25L did not. Triton X-114 partitioning of the integral membrane proteins of rough microsomes suggested that pgp35 (SSR alpha) and calnexin were major Ca(2+)-binding proteins of the endoplasmic reticulum membrane. We propose that the function of the complex is to regulate Ca(2+)-dependent retention mechanisms for luminal proteins of the endoplasmic reticulum.

Ligand-mediated autophosphorylation activity of the epidermal growth factor receptor during internalization

The association of EGF with its receptor in endosomes isolated from rat liver homogenates was assessed biochemically by polyethylene glycol precipitation and morphologically by electron microscope radioautography. The proportion of receptor-bound ligand in endosomes at 15 min after the injection of doses of 0.1 and 1 microgram EGF/100 g body weight was 57%. This value increased to 77% for the dose of 10 micrograms EGF injected. Quantitative electron microscope radioautography carried out on endosomes isolated at 15 min after the injection of 10 micrograms 125I-EGF demonstrated that most radiolabel was over the endosomal periphery thereby indicating that ligand-receptor complexes were in the bounding membrane but not in intraluminal vesicles of the content. EGF receptor autophosphorylation activity during internalization was evaluated in plasmalemma and endosome fractions. This activity was markedly but transiently reduced on the cell surface shortly after the administration of saturating doses of EGF. The same activity, however, was augmented and prolonged in endosomes for up to 30 min after EGF injection. The transient desensitization of cell surface activity was not due to prior in vivo phosphorylation since receptor dephosphorylation in vitro failed to restore autophosphorylation activity. Transient desensitization of cell surface autophosphorylation activity coincided with a diminished capacity for endocytosis of 125I-EGF with endocytosis returning to normal after the restoration of cell surface autophosphorylation activity. The inhibition of cell surface autophosphorylation activity and the activation of endosomal autophosphorylation activity coincident with downregulation suggest that EGF receptor traffic is governed by ligand-regulated phosphorylation activity.

Ligand-mediated internalization, recycling, and downregulation of the epidermal growth factor receptor in vivo

EGF receptor internalization, recycling,a nd downregulation were evaluated in liver parenchyma as a function of increasing doses of injected EGF. The effect of ligand occupancy in vivo on the kinetics and extent of internalization was studied with changes in the receptor content of isolated plasmalemma and endosome fractions evaluated by direct binding, Scatchard analysis, and Western blotting. For all doses of injected EGF, receptor was lost from the plasmalemma and accumulated in endosomes in a time- and dose-dependent fashion. However, at doses of injected EGF equivalent to less than or equal to 50% surface receptor occupancy (i.e., less than or equal to 1 microgram/100 g body weight), receptor levels returned by 120 min to initial values. This return was resistant to cycloheximide and therefore did not represent newly synthesized receptor. Neither was the return due to replenishment by an intracellular pool of low-affinity receptors as such a pool could not be detected by Scatchard analysis or Western blotting. Therefore, receptor return was due to the recycling of previously internalized receptor. At doses of injected EGF greater than 50% receptor occupancy, net receptor loss-i.e., downregulation-was observed by evaluating the receptor content of total particulate fractions of liver homogenates. At the higher saturating doses of injected EGF (5 and 10 micrograms/100 g body weight), the majority of surface receptor content was lost by 15 min and remained low for at least an additional 105 min. As the kinetics of ligand clearance from the circulation and liver parenchyma were similar for all doses of EGF injected, then the ligand-mediated regulation of surface receptor content and downregulation were not a result of a prolonged temporal interaction of ligand with receptor. Rather, the phenomena must be a consequence of the absolute concentrations of EGF interacting with receptor at the cell surface and/or in endosomes.

Transplacental and milk transfer of polybrominated biphenyls to perinatal guinea pigs from treated dams

Timed-pregnant albino Hartley strain guinea pigs of approximately 65 days gestation or lactating animals within 6-12 h of parturition received a single oral dose of Firemaster FF-1 (50 mg/kg body wt). The pregnant animals and their fetuses were killed 2 days later at term while the lactating animals and their pups were killed at intervals of 2, 4, 7, 14, 28, 42, and 60 days. Tissues (liver, kidney, lung, perirenal fat) were removed for the analysis of the 2,4,5,2',4',5'-hexabromobiphenyl (HBB) isomer content by gas-liquid chromatography following extraction. Microsomes were prepared from samples of fresh liver for the analysis of hepatic mono-oxygenase (p-nitroanisole O-demethylase, aniline hydroxylase) and UDP-glucuronosyl-transferase activities. Transplacentally-acquired residues of the order of 45 micrograms HBB/g were found in both maternal and fetal adipose tissue and in fetal liver. HBB residues in maternal kidney, lung and liver were of the order of 4-7 micrograms/g while, in the fetuses and pups, levels in the kidney and lung were of the order of 1-2 micrograms/g. Levels of HBB in breast milk 2 days after treatment averaged 22.4 +/- 7.8 micrograms/g (mean +/- S.D.). A marked induction of hepatic microsomal mono-oxygenases was observed in guinea pig pups concomitant with elevated hepatic levels of HBB (9-19 micrograms/g) which decreased with time as the agent was redistributed into adipose tissue. By 7 days of age the pups had been weaned and by 14 days of age the enzymatic activities were comparable to those measured in control pups. HBB levels in the pup kidney, lung and adipose tissue reflected redistribution or sequestration in the body fat. The biological half-life of HBB in tissues of both dams and pups appeared to be approximately 22 days.