Reduced endocytosis and altered lysosome function in cisplatin-resistant cell lines.

We isolated human KB adenocarcinoma cisplatin-resistant (CP-r) cell lines with multidrug-resistance phenotypes because of reduced accumulation of cisplatin and other cytotoxic compounds such as methotrexate and heavy metals. The uptake of horseradish peroxidase (HRPO) and Texas Red dextran was decreased several-fold in KB-CP-r cells, indicating a general defect in fluid-phase endocytosis. In contrast, although EGF receptors were decreased in amount, the kinetics of EGF uptake, a marker of receptor-mediated endocytosis, was similar in sensitive and resistant cells. However, 40-60% of the (125)I-EGF released into the medium after uptake into lysosomes of KB-CP-r cells was TCA precipitable as compared to only 10% released by sensitive cells. These results indicate inefficient degradation of internalised (125)I-EGF in the lysosomes of KB-CP-r cells, consistent with slower processing of cathepsin L, a lysosomal cysteine protease. Treatment of KB cells by bafilomycin A(1), a known inhibitor of the vacuolar proton pump, mimicked the phenotype seen in KB-CP-r cells with reduced uptake of HRPO, (125)I-EGF, (14)C-carboplatin, and release of TCA precipitable (125)I-EGF. KB-CP-r cells also had less acidic lysosomes. KB-CP-r cells were crossresistant to Pseudomonas exotoxin, and Pseudomonas exotoxin-resistant KB cells were crossresistant to cisplatin. Since cells with endosomal acidification defects are known to be resistant to Pseudomonas exotoxin and blocking of endosomal acidification mimics the CP-r phenotype, we conclude that defective endosomal acidification may contribute to acquired cisplatin resistance.

The Stat3/5 locus encodes novel endoplasmic reticulum and helicase-like proteins that are preferentially expressed in normal and neoplastic mammary tissue.

The signal transducers and activators of transcription (STAT) 5 and 3 are critical for mammary alveolar development during pregnancy and remodeling during involution. In the mouse, STAT3, STAT5a, and STAT5b are encoded by adjacent genes on chromosome 11 (60.5 cM). To identify additional genes in the Stat3/5 locus that may participate in normal and neoplastic development of the mammary gland, we have cloned and sequenced 500 kb and searched for genes preferentially expressed in mammary tissue. We identified six known genes and cloned two new genes, termed D11Lgp1 and D11Lgp2. Both genes are most highly expressed in normal mammary tissue and mammary tumors from several transgenic mouse models. LGP1 consists of 532 and 530 amino acids in mouse and human, respectively (88% similarity). A region in the carboxy-terminal half of LGP1 has limited homology with Arabidopsis thaliana GH3-like proteins. Immunofluorescence studies demonstrated that LGP1 is located in the nuclear envelope and the endoplasmic reticulum. LGP2 is a cytoplasmic protein of 678 amino acids.

An isoform of branched-chain aminotransferase is a novel co-repressor for thyroid hormone nuclear receptors.

The functions of thyroid hormone receptors (TRs) are regulated by a host of co-regulatory proteins. Tissue-specific expression of these co-regulators leads to distinct expression patterns and regulation of thyroid hormone (T3) target genes in tissues. Previously we have found that human colon carcinoma RKO cells exhibit strong T3-independent transcriptional activity. We therefore searched for co-regulatory proteins in RKO cells using a yeast two-hybrid system with the intact TRbeta1 as bait. One of the three positive clones, designated as P3, was identified to be an isoform of human mitochondria branched-chain aminotransferase (BCATm). P3 was a spliced variant of BCATm with an internal 12-amino acid deletion near the carboxyl-terminal region and was abundantly expressed in RKO cells. The expressed protein localized both to the mitochondria and the nucleus of transfected CV1 cells. P3 physically interacted with TRbeta1 in a T3-independent manner that led to the inhibition in binding of TRbeta1 to thyroid hormone-responsive element. P3 not only enhanced the repressor activity of the unliganded TR but also repressed the ligand-dependent activation of TR. This repression was reversed by treatment of cells with trichostatin A, suggesting that in addition to the inhibition of DNA binding, the repression activity of P3 on TR may also be mediated by histone deacetylase activity. Thus, unlike the currently known co-repressors, P3 is a novel ligand-independent co-repressor for TR.

Glycan-dependent signaling: O-linked N-acetylglucosamine.

The addition of O-linked N-acetylglucosamine (O-GlcNAc) to target proteins may serve as a signaling modification analogous to protein phosphorylation. Like phosphorylation, O-GlcNAc is a dynamic modification occurring in the nucleus and cytoplasm. Various analytical methods have been developed to detect O-GlcNAc and distinguish it from glycosylation in the endomembrane system. Many target molecules have been identified; these targets are typically components of supramolecular complexes such as transcription factors, nuclear pore proteins, or cytoskeletal components. The enzymes responsible for O-GlcNAc addition and removal are highly conserved molecules having molecular features consistent with a signaling role. The O-GlcNAc transferase and O-GlcNAcase are likely to act in consort with kinases and phosphatases generating various isoforms of physiological substrates. These isoforms may differ in such properties as protein-protein interactions, protein stability, and enzymatic activity. Since O-GlcNAc plays a critical role in the regulation of signaling pathways of higher plants, the glycan modification is likely to perform similar signaling functions in mammalian cells. Glucose and amino acid metabolism generates hexosamine precursors that may be key regulators of a nutrient sensing pathway involving O-GlcNAc signaling. Altered O-linked GlcNAc metabolism may also occur in human diseases including neurodegenerative disorders, diabetes mellitus and cancer.

Calreticulin Is a receptor for nuclear export.

In previous work, we used a permeabilized cell assay that reconstitutes nuclear export of protein kinase inhibitor (PKI) to show that cytosol contains an export activity that is distinct from Crm1 (Holaska, J.M., and B.M. Paschal. 1995. Proc. Natl. Acad. Sci. USA. 95: 14739-14744). Here, we describe the purification and characterization of the activity as calreticulin (CRT), a protein previously ascribed to functions in the lumen of the ER. We show that cells contain both ER and cytosolic pools of CRT. The mechanism of CRT-dependent export of PKI requires a functional nuclear export signal (NES) in PKI and involves formation of an export complex that contains RanGTP. Previous studies linking CRT to downregulation of steroid hormone receptor function led us to examine its potential role in nuclear export of the glucocorticoid receptor (GR). We found that CRT mediates nuclear export of GR in permeabilized cell, microinjection, and transfection assays. GR export is insensitive to the Crm1 inhibitor leptomycin B in vivo, and it does not rely on a leucine-rich NES. Rather, GR export is facilitated by its DNA-binding domain, which is shown to function as an NES when transplanted to a green fluorescent protein reporter. CRT defines a new export pathway that may regulate the transcriptional activity of steroid hormone receptors.

Hormone binding induces rapid proteasome-mediated degradation of thyroid hormone receptors.

The thyroid hormone 3,3',5-triiodo-l-thyronine (T3) is essential for growth, differentiation, and development. Its biological activities are mediated by T3 nuclear receptors (TRs). At present, how T3 regulates TR proteins and the resulting functional consequences are still unknown. Immunofluorescence analyses of endogenous TR in the growth hormone-producing GC cells showed that the T3-induced rapid degradation of TR was specifically blocked by lactacystin, a selective inhibitor of the ubiquitin-proteasome degradation pathway. Immunoblots demonstrated that the transfected TRbeta1 was ubiquitinated and that the ubiquitination was T3 independent. Studies with a series of truncated TRbeta1 showed that the hormone-binding domain was sufficient for the T3-induced rapid degradation of TRbeta1 by the proteasome degradation pathway. T3 also induced rapid degradation of TRbeta2 and TRalpha1. In contrast, the stability of the non-T3-binding TRalpha2 and naturally occurring TRbeta1 mutants that do not bind T3 was not affected by T3 treatment, indicating that hormone binding to receptor was essential for the degradation of the wild-type receptors. In the presence of proteasome protease inhibitors, the levels of both total and ubiquitinated TRbeta1 protein increased, yet T3-dependent transcriptional activation and the expression of the growth hormone gene were diminished, suggesting that proteasome-mediated degradation played a novel role in modulating transcriptional activation by TR. The present study reveals a role of T3 in modulating the functions of TR by regulating its receptor level via the ubiquitin-proteasome degradation pathway.

The long signal peptide isoform and its alternative processing direct the intracellular trafficking of interleukin-15.

Two isoforms of interleukin (IL)-15 exist: one with a short and another with a long signal peptide (LSP). Experiments using combinations of the LSP and mature proteins IL-2, IL-15, and green fluorescent protein revealed complex pathways of intracellular trafficking. In one pathway, the LSP was unprocessed, and IL-15 was not glycosylated, remained in the cytoplasm, and was degraded. The second trafficking pathway involved endoplasmic reticulum entry, N-linked glycosylation, and alternative partial LSP processing. The third pathway involved endoplasmic reticulum entry, followed by glycosylation, complete processing, and ultimately secretion. The complex intracellular trafficking patterns of LSP-IL-15 with its impediments to secretion as well as impediments to translation may be required due to the potency of IL-15 as an inflammatory cytokine. In terms of a more positive role, we propose that intracellular infection may relieve the burdens on translation and intracellular trafficking to yield effective IL-15 expression.

Calcium, ATP and nuclear pore channel gating.

Nuclear envelope (NE) cisternal Ca2+ and cytosolic ATP are required for nuclear-pore-complex-(NPC-) mediated transport of DNAs, RNAs, transcription factors and other large molecules. Isolated cardiomyocyte nuclei, capable of macromolecular transport (MMT), have intrinsic NPC ion channel behavior. The large ion conductance (gamma) activity of the NPC channel (NPCC) is blocked by the NPC monoclonal antibody mAb414, known to block MMT, and is also silenced during periods of MMT. In cardiomyocytes, neither cytosolic Ca2+ nor ATP alone directly affects NPCC gating. To test the role of Ca2+ and ATP in NPCC activity, we carried out the present patch-clamp study with the pipette attached to the outer NE membrane of nuclei isolated from cultured Dunning G prostate cancer cells. Our investigations demonstrate that in these isolated nuclei neither cytosolic Ca2+ nor ATP alone directly affects NPCC gating. However, when simultaneously applied to the bath and pipette, they transiently silence NPCC activity through stimulation of MMT by raising the Ca2+ concentration in the NE cisterna ([Ca2+]NE). Our fluorescence microscopy observations with nuclear-targeted macromolecular fluorochromes (B-phycoerythrin and plasmid for the enhanced green fluorescence protein EGFP, pEGFP-C1) and with FITC-labeled RNA support the view that channel silence accompanies MMT. Repeated Ca2+ loading of the NE with Ca2+ and ATP, after unloading with 1-5 microM inositol 1,4,5-trisphosphate (IP3), thapsigargin (TSG) or 5 mM BAPTA or EGTA, failed to affect channel gating. This result indicates that other factors are involved in this phenomenon and that they are exhausted during the first cycle of NE Ca2+ loading/unloading--in agreement with current theories of NPC-mediated MMT. The results explain how Ca2+ and IP3 waves may convert the NE into an effective Ca2+ barrier and, consequently, affect the regulation of gene activity and expression through their feedback on MMT and NPCC gating. Thus, [Ca2+]NE regulation by intracellular messengers is an effective mechanism for synchronizing gene activity and expression to the cellular rhythm.

Phosphorylation and glycosylation of nucleoporins.

The nuclear pore complex mediates macromolecular transport between the nucleus and cytoplasm. Many nuclear pore components (nucleoporins) are modified by both phosphate and O-linked N-acetylglucosamine (O-GlcNAc). Among its many functions, protein phosphorylation plays essential roles in cell cycle progression. The role of O-GlcNAc addition is unknown. Here, levels of nucleoporin phosphorylation and glycosylation during cell cycle progression are examined. Whereas nuclear pore glycoproteins are phosphorylated in a cell-cycle-dependent manner, levels of O-GlcNAc remain constant. The major nucleoporin p62 can be phosphorylated in vitro by protein kinase A and glycogen synthase kinase (GSK)-3alpha but not by cyclin B/cdc2 or GSK-3beta. The consensus sites of these kinases resemble sites which can be glycosylated by O-GlcNAc transferase. These data are consistent with a model that O-GlcNAc limits nucleoporin hyperphosphorylation during M-phase and hastens the resumption of regulated nuclear transport at the completion of cell division.

Reconstitution of HIV-1 rev nuclear export: independent requirements for nuclear import and export.

The Rev protein of HIV-1 actively shuttles between nucleus and cytoplasm and mediates the export of unspliced retroviral RNAs. The localization of shuttling proteins such as Rev is controlled by the relative rates of nuclear import and export. To study nuclear export in isolation, we generated cell lines expressing a green fluorescent protein-labeled chimeric protein consisting of HIV-1 Rev and a hormone-inducible nuclear localization sequence. Steroid removal switches off import thus allowing direct visualization of the Rev export pathway in living cells. After digitonin permeabilization of these cells, we found that a functional nuclear export sequence (NES), ATP, and fractionated cytosol were sufficient for nuclear export in vitro. Nuclear pore-specific lectins and leptomycin B were potent export inhibitors. Nuclear export was not inhibited by antagonists of calcium metabolism that block nuclear import. These data further suggest that nuclear pores do not functionally close when luminal calcium stores are depleted. The distinct requirements for nuclear import and export argue that these competing processes may be regulated independently. This system should have wide applicability for the analysis of nuclear import and export.

An evaluation of sialation of the nucleoporin p62.

Many nuclear and cytosolic proteins are modified by single residues of O-linked N-acetyl-D-glucosamine. These include many proteins found in nuclear pore complexes required for transport of macromolecules between the nucleus and the cytoplasm. The best characterized pore glycoprotein, p62, mediates its function as one component of a protein complex essential for nuclear transport. Although p62 sugar residues are not essential for nuclear transport, they appear to oppose protein phosphorylation occurring at sites predicted to destabilize protein-protein interactions of the p62 complex. Recently, a p62-like protein isolated from mouse neuroblastoma cells was reported to be modified by both GlcNAc and sialic acid. As there is little precedent for nucleoplasmic sialation, the finding that a characterized nuclear pore protein is sialated is significant because it may regulate pore function. To assess the biological importance of p62 sialation, GlcNAc and sialic acid-specific lectins were used to examine the state of p62 glycosylation in cells commonly used to study nuclear transport: frog eggs and normal rat kidney and HeLa fibroblasts. In addition, four mouse neuroblastoma cell lines derived from the same tumor were examined. The glycosylation of p62 in these cells appears to involve only single O-linked GlcNAc moieties; no significant sialation was detected.

Generation of secretable and nonsecretable interleukin 15 isoforms through alternate usage of signal peptides.

Two isoforms of human interleukin 15 (IL-15) exist. One isoform has a shorter putative signal peptide (21 amino acids) and its transcript shows a tissue distribution pattern that is distinct from that of the alternative IL-15 isoform with a 48-aa signal peptide. The 21-aa signal isoform is preferentially expressed in tissues such as testis and thymus. Experiments using different combinations of signal peptides and mature proteins (IL-2, IL-15, and green fluorescent protein) showed that the short signal peptide regulates the fate of the mature protein by controlling the intracellular trafficking to nonendoplasmic reticulum sites, whereas the long signal peptide both regulates the rate of protein translation and functions as a secretory signal peptide. As a consequence, the IL-15 associated with the short signal peptide is not secreted, but rather is stored intracellularly, appearing in the nucleus and cytoplasmic components. Such production of an intracellular lymphokine is not typical of other soluble interleukin systems, suggesting a biological function for IL-15 as an intracellular molecule.

Calmodulin activates nuclear protein import: a link between signal transduction and nuclear transport.

In addition to the well-characterized GTP-dependent nuclear transport observed in permeabilized cells, we detected a mode of nuclear transport that was GTP-independent at elevated cytoplasmic calcium concentrations. Nuclear transport under these conditions was blocked by calmodulin inhibitors. Recombinant calmodulin restored ATP-dependent nuclear transport in the absence of cytosol. Calmodulin-dependent transport was inhibited by wheat germ agglutinin consistent with transport proceeding through nuclear pores. We propose that release of intracellular calcium stores upon cell activation inhibits GTP-dependent nuclear transport; the elevated cytosolic calcium then acts through calmodulin to stimulate the novel GTP-independent mode of import.

Patch clamp detection of transcription factor translocation along the nuclear pore complex channel.

Transcription factors (TFs) are cytoplasmic proteins that play an essential role in gene expression. These proteins form multimers and this phenomenon is thought to be one of the mechanisms that regulate transcription. TF molecules reach their DNA binding sites through the large central channel of the nuclear pore complex (NPC). However, the NPC channel is known to restrict the translocation of molecules > or = 20-70 kD. Therefore, during their translocation, TF molecules and/or their multimers may plug the NPC channel and thus, interrupt ion flow through the channel, with a concomitant reduction in the ion conductance of the channel (gamma). Here we show with patch clamp that gamma is reduced during translocation of three major TFs: c-Jun (40 kD), NF-kappa B (approximately equal to 50 kD), and SP1 (approximately equal to 100 kD). Within a minute, femtomolar concentrations of these proteins reduced gamma suggesting a purely mechanical interaction between single TF molecules and the inner wall of the NPC channel. NPCs remained plugged for 0.5-3 hr in the absence of ATP but when ATP was added, channel plugging was shortened to < 5 min. After unplugging, channel closures were rarely observed and the number of functional channels increased. The transcription factors also stabilized the NPCs as shown by the extended duration of the preparations which allowed recordings for up to 72 hr. These observations are the first direct demonstration of the important role of NPCs in mediating nuclear translocation of TFs and, therefore, in forming part of the mechanisms regulating gene expression.(ABSTRACT TRUNCATED AT 250 WORDS)

Nuclear pore complex ion channels (review).

It is currently thought that nuclear pore complexes (NPCs) primarily govern nucleocytoplasmic interactions via selective recognition and active transport of macromolecules. However, in various nuclear preparations, patch-clamp and fluorescence, luminiscence and ion microscopy support classical microelectrode measurements indicating that monoatomic ion flow across the nuclear envelope (NE) is strictly regulated. Gating of large conductance nuclear envelope ion channels (NICs) somewhat resembles that of gap junctional channels. In other respects, NICs are distinct in that they require cytosolic factors, are blocked by wheat germ agglutinin and are blocked and/or modified by antibodies to epitopes of NPC glycoproteins. Therefore, NIC activity, recorded as electrical current/conductance is likely to be intrinsic to NPCs. This observation suggests a potential use for the patch-clamp technique in establishing the mechanisms underlying nuclear pore gating in response to cytosolic and nucleosolic factors such as transcription and growth factors, oncogene and proto-oncogene products and receptors for retinoids, steroids and thyroid hormone. NIC activity may also be useful in evaluating the mechanisms of nuclear import of foreign nucleic acid material such as that contained in virons and viroids. Finally, in consideration to the electrophysiological data accumulated so far, the study of nuclear pore ion channel activity may help our understanding of other important issues such as cell suicide, programmed cell death or apoptosis.

Glucocorticoid receptor binding to rat liver nuclei occurs without nuclear transport.

The binding of cell-free activated glucocorticoid receptor-steroid complexes from HTC cells to various preparations of HTC and rat liver nuclei has been examined under conditions that did or did not support the nuclear translocation of macromolecules via nuclear pores. To the best of our knowledge, this is the first such study with functionally active isolated nuclei. Conventionally prepared HTC nuclei were found to be porous, as determined from their inability to exclude the fluorescent macromolecule phycoerythrin (PE) at 4 degrees C. Thus the nuclear binding of activated complexes to these nuclei can not involve nuclear translocation. Further studies, using established conditions with sealed nuclei prepared from rat liver, revealed that nuclear translocation of PE containing a covalently linked, authentic nuclear translocation sequence could be obtained at 22 degrees C, but not at 4 degrees C. However, under the same conditions, activated glucocorticoid complexes displayed equal levels of nuclear binding at both temperatures. We therefore conclude that the current translocation conditions with intact rat liver nuclei are not sufficient to reproduce the nuclear transport of glucocorticoid complexes observed in intact cells. The nuclear binding that was seen with intact rat liver nuclei was not affected by aurintricarboxylic acid, which selectively inhibits protein-nucleic acid interactions. The antibody AP-64, shown to be specific for amino acids 506-514 of the nuclear translocation sequence of the rat glucocorticoid receptor, inhibited the nuclear binding of activated complexes, apparently by blocking receptor access to the nuclear membrane. Collectively, these data argue that activated complex binding to nuclei capable of nuclear translocation involves only an association with nuclear membrane components such as nuclear pores. Thus this system, and these reagents, may be useful in future studies of activated complex binding to nuclear pores.

Antibodies against the SV40 large T antigen nuclear localization sequence.

Transport of large proteins into the nucleus requires both a nuclear localization signal (NLS) and exposure of that signal to components of the transport machinery. In this report, polyclonal and monoclonal antibodies were generated against the NLS of SV40 large T antigen. Several of these antibodies immunoprecipitated large T antigen produced by in vitro transcription-translation and recognized T antigen expressed in cultured cells. Binding of the antibodies to T antigen was quantified using an indirect radioimmunoassay and found to be specifically inhibited by peptides corresponding to the T antigen NLS. The ability of NLS-specific antibodies to recognize large T antigen suggests that the NLS is exposed on the surface of T antigen. When one of the NLS-specific monoclonal antibodies was introduced into the cytoplasm of cells expressing T antigen, the antibody remained cytoplasmic. These results suggested either that cytoplasmic components compete for binding to the NLS or that the antibody dissociates from T antigen during transport into the nucleus. When an antibody directed against an epitope distinct from the NLS was microinjected into the cytoplasm of cells expressing large T antigen, both the antibody and antigen were transported into the nucleus. The observed stability of the antigen-antibody complex strongly suggest protein unfolding is not required for nuclear protein transport.

Primary sequence and heterologous expression of nuclear pore glycoprotein p62.

The major nuclear pore protein p62 is modified by O-linked N-acetylglucosamine and functions in nuclear transport. We have cloned, sequenced, and expressed the full-length rat p62 cDNA. The rat p62 mRNA is 2,941 nucleotides long and encodes a protein of 525 amino acids containing 30% serine and threonine residues. The amino acid sequence near the amino-terminus contains unique tetrapeptide repeats while the carboxy-terminus consists of a series of predicted alpha-helical regions with hydrophobic heptad repeats. Heterologous expression of rat p62 in African Green Monkey Kidney COS-1 cells and CV-1 cells was detected using a species-specific antipeptide serum. When transiently expressed in COS-1 cells, rat p62 binds wheat germ agglutinin and concentrates at the spindle poles during mitosis. In CV-1 cells cotransfected with rat p62 cDNA and SV40 viral DNA, rat p62 associates with the nuclear membrane without interfering with the nuclear transport of SV40 large T antigen. The ability to express p62 in tissue culture cells will facilitate analysis of the role of this pore protein in nuclear transport.

Glycosylation of nuclear pore protein p62. Reticulocyte lysate catalyzes O-linked N-acetylglucosamine addition in vitro.

The addition of O-linked N-acetylglucosamine (GlcNAc) to the major nuclear pore complex glycoprotein p62 was examined. Expression of the rat p62 cDNA in transfected monkey cells was detected using a rat p62-specific antipeptide antiserum and two previously described nuclear pore-specific monoclonal antibodies which require O-linked GlcNAc for binding. Although the p62 cDNA was predicted to encode a 54-kDa polypeptide, the product expressed in monkey cells migrated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis as two species of 62 and 59-kDa. Cell-free translation of the p62 in vitro transcript yielded a 59-kDa polypeptide using wheat germ extract and a 62-kDa product using a commercially available rabbit reticulocyte lysate. Several lines of evidence indicated that the 62-kDa rabbit reticulocyte lysate translation product was modified by O-linked N-acetylglucosamine; the protein bound specifically to a wheat germ agglutinin affinity column and was converted to 59 kDa when treated with jack bean beta-acetylglucosaminidase. The 59-kDa unglycosylated wheat germ translation product was converted to the 62-kDa glycosylated form upon incubation with reticulocyte lysate demonstrating that O-linked GlcNAc can be added to p62 post-translationally.

Overexpression and biosynthesis of CD4 in Chinese hamster ovary cells: coamplification using the multiple drug resistance gene.

CD4 is a T-cell surface glycoprotein and serves as the receptor for the human immunodeficiency virus. Glycosylation of CD4 has been shown to be necessary for proper surface expression. To study the biosynthesis and assembly of CD4, wild-type and glycosylation-deficient mutant Chinese hamster ovary (CHO) cells were cotransfected with a cDNA encoding CD4 and a cDNA for the human multiple drug resistance gene, which allowed the amplification of the transfected CD4 cDNA sequences. Clones were isolated that exhibited high-level expression of CD4 resulting from the integration of several copies of CD4 cDNA. CD4 synthesized by these cells acquired resistance to endoglycosidase H after 20-30 min of chase, suggesting a rapid translocation of the glycoprotein from the rough endoplasmic reticulum to the medial Golgi apparatus. The sensitivity of CD4 to glycosidases suggested the presence of biantennary unsialylated complex-type oligosaccharides. Consistent with this, CD4 synthesized by the Lec2 mutant, which does not add sialic acid to oligosaccharides, was identical to the glycoprotein produced by wild-type CHO cells. The amplification strategy used to express CD4 at high levels in wild-type and mutant CHO cells will have general utility.