Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M407382200 on July 16, 2004

J. Biol. Chem., Vol. 279, Issue 50, 51793-51803, December 10, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/50/51793    most recent
M407382200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meads, M. B.
Right arrow Articles by Medveczky, P. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meads, M. B.
Right arrow Articles by Medveczky, P. G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Kaposi's Sarcoma-associated Herpesvirus-encoded Viral Interleukin-6 Is Secreted and Modified Differently Than Human Interleukin-6

EVIDENCE FOR A UNIQUE AUTOCRINE SIGNALING MECHANISM*

Mark B. Meads{ddagger} and Peter G. Medveczky§

Received for publication, July 1, 2004


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Viral interleukin-6 (vIL-6) is a homolog of cellular IL-6 that is encoded by the Kaposi's sarcoma-associated herpesvirus (KSHV) genome. vIL-6 binds to the IL-6 signal transducer gp130 without the cooperation of the IL-6 high affinity receptor to induce STAT3 DNA binding and cell proliferation. Although vIL-6 is believed to be important in the pathogenesis of KSHV-induced diseases, its secretion and post-translational modifications have not previously been characterized. Pulse-chase analysis revealed that the half-time of vIL-6 secretion is ~8-fold longer than that of human IL-6. Yet, the vIL-6 signal sequence targets human IL-6 secretion to nearly wild-type levels. Surprisingly, vIL-6 was not secreted from a cell line that does not express gp130 but expression of human gp130 in these cells enabled the secretion of vIL-6. Consistent with this observation, complete maturation of gp130 N-glycans is inhibited by vIL-6 coexpression, suggesting that the binding of the receptor to vIL-6 occurs intracellularly in early or pre-Golgi compartments. Furthermore, a vIL-6 mutant containing an endoplasmic reticulum retention signal is not secreted but does still induce receptor activation and signaling. Secreted vIL-6 is completely glycosylated at both possible N-glycosylaton sites and contains a large proportion of immature high-mannose glycans that is not typical of cytokines. These findings suggest that vIL-6 may induce gp130 signaling by an exclusively autocrine mechanism that relies on intracellular binding to its receptor. During KSHV infection, vIL-6 may only induce signaling in KSHV-infected cells to benefit the viral life cycle and promote oncogenic transformation.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Kaposi's sarcoma-associated herpesvirus (KSHV)1 is the causative agent of Kaposi's sarcoma and is associated with two lymphoproliferative disorders: multicentric Castleman's disease (MCD) and AIDS-related primary effusion lymphoma (PEL). DNA sequences and latency-associated nuclear antigen protein from the virus have been found in all of the epidemiological forms of Kaposi's sarcoma (1), in PEL (2), and in a significant percentage of MCD (3, 4).

The virus encodes a homolog of IL-6, viral interleukin-6 (vIL-6). vIL-6 is the only KSHV protein that has been shown to induce signal transducer and activator of transcription-3 (STAT3) phosphorylation and DNA binding. Importantly, STAT3 activation by itself can induce cell transformation and tumorigenesis (5), promote cell survival by inhibiting apoptosis (5–7), and induce mitogenesis by up-regulating c-Myc (8). Recently, Aoki et al. (9) showed that inhibition of STAT3 signaling in PEL cells leads to decreased survivin expression and caspase-dependent apoptosis (9). Furthermore, STAT3 activation by vIL-6 induces vascular endothelial growth factor expression (10, 11), which is important for the progression of MCD (12). vIL-6 expression has been demonstrated in Kaposi's sarcoma, PEL, and MCD (4, 13–15). Therefore, vIL-6 may be a key component leading to KSHV-induced tumorigenesis.

Human IL-6 (hIL-6) is an important autocrine growth factor leading to pathogenesis in several lymphoproliferative disorders (6, 16). Whereas some investigators have found that vIL-6, but not hIL-6, is required for the growth of PEL cells (17–19), others have found that hIL-6 is required (20). In support of this latter finding, vIL-6 has been shown to induce the expression of hIL-6 (21). Recombinant vIL-6 must be used in extracellular concentrations 1000–5000-fold higher than those of the cellular protein to achieve equivalent activity in cell proliferation and STAT3 activation assays (22, 23). Circulating vIL-6 levels in Kaposi's sarcoma, MCD, and PEL were found to be in the picogram/milliliter to very low nanogram/milliliter range, representing concentrations that are thousands of times lower than those required to show activity with purified recombinant protein (24). As a result, the transfection of cells with vIL-6 DNA or concentrated supernatant from transfected cells is often used to demonstrate STAT3 induction and cell proliferation (25, 26). These facts demonstrate that extracellular vIL-6 has very low activity compared with hIL-6 and suggest that extracellular vIL-6 may have comparatively less biological significance than hIL-6.

Several groups have demonstrated that vIL-6 can bind to gp130 alone and that it does not require the high affinity IL-6 receptor (23, 27–29). This is in stark contrast to hIL-6, which must bind to its high affinity receptor before a signaling complex can be formed with its signal transducer, gp130 (30, 31). Using an enzyme-linked immunosorbent assay-based in vitro binding assay, Aoki et al. (27) showed that vIL-6 binds to the extracellular domain of gp130 with a dissociation constant of 2.5 µM, representing a 1000-fold lower affinity for gp130 than that of the hIL-6·IL-6R complex (30). A crystal structure of vIL-6 complexed with the extracellular domain of gp130 was described to be a tetrameric complex containing only vIL-6 and gp130 in a 2:2 stoichiometry (28). Recently, Chatterjee et al. (32) explained that the IL-6R is down-regulated by interferon-{alpha} and that vIL-6 expression protects B-cells from interferon-{alpha}-induced proliferation arrest.

Despite the importance of vIL-6 in the progression of KSHV diseases, secretion efficiency and post-translation modification of the protein have never been studied. Although large quantities of extracellular vIL-6 are needed to elicit cytokine-like activity, no one has demonstrated that such large amounts of the protein are present in tissue culture medium from KSHV cell lines or extracellularly in KSHV-induced tumors. Our work shows that vIL-6 is secreted very slowly from cells compared with hIL-6. Furthermore, wild-type vIL-6 expression inhibits the maturation of gp130 glycans and a mutant vIL-6 that is not secreted can induce the phosphorylation of gp130 and STAT3. Secreted vIL-6 is modified with immature high-mannose asparagine-linked glycans rarely observed in extracellular growth factors. These facts are consistent with a unique autocrine signaling mechanism involving intracellular binding of vIL-6 to its receptor, gp130.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cloning and Chimera Construction—vIL-6 cDNA was amplified by PCR from DNA derived from the PEL cell line BC-1 using primers 5'-GCGTGGATCCCCATGTGCTGGTTCAAGTTGTG-3' and 5'-TCACGTCTCGAGTTACTTATCGTGGACGTC-3'. The cDNA was cloned to vector pGEX5X-3 (Promega, Madison, WI). Chimeras were constructed by overlap-extension PCR. Two fragments were created using outside primers and primers containing an overlapping linker coding for Gly-Gly-Ala-Gly-Gly at the junction point. These fragments were then extended to full-length in a second round of PCR. A stop codon was removed, and a Kozak sequence (33) was added to each chimera and wild-type hIL-6 and vIL-6 by PCR amplification and then subcloned to pCDNA3.1 upstream of a FLAG linker sequence containing a stop codon.

An oligonucleotide-generated linker fragment coding for the COOH-terminal endoplasmic reticulum (ER) retention motif of protein-disulfide isomerase (PDI), ~MEEDDDQKAVKDEL-COOH, was cloned in-frame with the wild-type vIL-6 open reading frame in pCDNA3-vIL-6-FLAG, replacing the FLAG linker to create pCDNA3-vIL-6-KDEL. A vIL-6 construct containing a mutant ER motif was also constructed, pCDNA3-vIL-6-KDQA. The mutant construct codes for point mutations in the last two amino acids of the protein (EL to QA is underlined above). All of the DNA clones were sequenced and contain no mutations.

Cell Lines and Media—A stable cell line that expresses human gp130 and an empty vector control cell line were derived from the BAF-B03 murine IL-3-dependent pro-B-cell line, which does not express gp130 (31). These cells were the generous gift of Dr. Masashi Narasaki (Osaka University Medical School, Osaka, Japan) and were stably transfected with vIL-6 in the context of the expression vector pZEOsv and selected with 400 µg/ml Zeocin (Invitrogen) to generate vIL-6-expressing BAF-B03 cells with and without gp130. BAF-B03 cells were grown in RPMI 1640 medium (Mediatech, Herndon, VA) supplemented with 10% WEHI-3B-conditioned medium as a source of IL-3. Selection for gp130 expression was with 300 µg/ml G418 (Mediatech). BCBL-1 is a B-cell line derived from PEL that was used to study vIL-6 in the context of disease. BCBL-1 and human embryonic kidney (HEK) 293 cells were grown in RPMI 1640 medium and Dulbecco's modified Eagle's medium (Mediatech), respectively. All of the medium was supplemented with 10% fetal calf serum and 500 µg/ml gentamycin (Mediatech).

Materials and Generation of vIL-6 Antisera—Recombinant vIL-6-glutathione S-transferase fusion protein was produced in Escherichia coli and purified using glutathione S-transferase-agarose beads (Sigma). New Zealand White rabbits were immunized with 500 µg of protein in complete Freund's adjuvant and boosted six times with 200 µg of protein in incomplete Freund's adjuvant at one-month intervals. The serum was harvested 10 days after each boost. Antiserum did not cross-react with recombinant hIL-6. {alpha}-FLAG (M2) was purchased from Sigma, {alpha}-gp130 (AM64) was purchased from BD Biosciences, and {alpha}-phospho-tyrosine (PY20) was purchased from BD Transduction Laboratories (San Jose, CA). {alpha}-Actin (C-2), {alpha}-pSTAT3 (B-7), and {alpha}-STAT3 (F-2) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). In vitro transcription/translation kit was purchased from Promega, and protein G-agarose beads were purchased from Amersham Biosciences.

Metabolic Labeling and Immunoprecipitation—HEK 293 cells were transiently transfected using Geneporter transfection reagent (Gene Therapy Systems, San Diego, CA) according to the manufacturer's instructions. Transiently transfected HEK 293 or the PEL cell line BCBL-1 were rinsed twice and then incubated in methionine/cysteine-free Dulbecco's modified Eagle's medium containing 10% dialyzed fetal calf serum for 10 min. HEK 293 cells were used 24 h post-transfection. BCBL-1 cells were used 24 h after 12-O-tetradecanoylphorbol-13-acetate induction (100 ng/ml). Cells were pulsed with [35S]methionine/cysteine (PerkinElmer Life Sciences) at 50 µCi/ml for 20 min in the same medium and chased in serum-free AIM V (Invitrogen) for the indicated times. The medium was centrifuged to remove cells. 1 mM Phenylmethylsulfonyl fluoride and 15 mM phenanthroline were added to inhibit proteases (Sigma). Cells were rinsed twice with Tris-buffered saline containing sodium azide (TSA) and lysed in TSA containing 1% Nonidet P-40 and Complete protease inhibitors (Roche Applied Science). Cellular debris and nuclei were removed by centrifugation.

Media and lysates were incubated overnight at 4 °C in a rotator with 1 µg/ml {alpha}-FLAG (M2) monoclonal antibody, 1 µg/ml {alpha}-gp130 (AM64) monoclonal antibody, or 2 µl/ml {alpha}-vIL-6 antiserum. This was followed by incubation for 2 h with 5 µl of protein G-agarose. Beads were rinsed twice in TSA containing 1% Nonidet P-40 and 0.1% bovine hemoglobin and three times in TSA and then incubated at 100 °C for 5 min in 20 µl of 2x SDS-PAGE sample buffer. Visualization of the labeled protein was done with 15% SDS-PAGE and fluorography (Dupont, Wilmington, DE). For membrane preparations, cells were disrupted by hypotonic lysis using a Dounce homogenizer and centrifuged at 500 x g for 5 min to remove nuclei. Supernatant was then centrifuged at 100,000 x g for 30 min to pellet plasma membrane and membranous organelles. The membrane pellet was dissolved in TSA containing 1% Nonidet P-40, and protein was immunoprecipitated as described above.

Western Blotting—Cell lysates or immunoprecipitated protein was incubated at 100 °C for 5 min in SDS-PAGE sample buffer. Samples were applied to 15% polyacrylamide gels for vIL-6, 8% for STAT3, and 5% for gp130 and electrophoresed at 100 V. Gels, nitrocellulose membranes (Amersham Biosciences), and filter paper were soaked in transfer buffer (20% methanol, 20 mM glycine, and 25 mM Tris, pH 8.5) for 5–10 min, and then gel and membrane were sandwiched between filter paper and placed in a transfer tank containing pre-chilled transfer buffer. Complete transfer was achieved at 20 V overnight at 4 °C. The nitrocellulose membrane was blocked by incubation for 1 h at room temperature in blocking buffer (Tris-buffered saline containing 5% non fat dry milk and 0.1% Tween 20). Membranes were probed for 20 min at room temperature with antibody in blocking solution and rinsed four times in Tris-buffered saline + 0.1% Tween 20 between primary and secondary antibody and after secondary antibody. Visualization was done by enhanced chemiluminescence (Amersham Biosciences) according to the manufacturer's instructions.

Endoglycosidase and Inhibitor Studies—Cells were treated with brefeldin A (BFA) (34) or tunicamycin (Sigma) (35) for 1 h prior to metabolic labeling and throughout labeling and chase. BFA was used at a concentration of 1 or 10 µg/ml, and tunicamycin was used at 5 µg/ml unless otherwise stated. Endo-{beta}-N-acetylglucosaminidase H (Endo H) and peptide N-glycosidase F (PNGase F) were purchased from New England Biolabs (Beverly, MA) and used to treat immunoprecipitated proteins according to the manufacturer's instructions.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
N-Linked and O-Linked Glycosylation—vIL-6 contains two possible asparagine-linked (N-)-glycosylation motifs (Asn-Xaa-Ser/Thr). In these two consensus sequences, Xaa is either aspartate or glutamate followed by threonine. Although the predicted molecular mass of the protein is 23.4 kDa, its apparent molecular mass by SDS-PAGE is ~28–30 kDa. This difference could be due to the presence of two N-linked oligosaccharides. To address this possibility, HEK 293 cells transiently expressing COOH-terminally FLAG-tagged vIL-6 were treated with tunicamycin, a specific inhibitor of N-linked glycosylation (35), concentrations ranging from 0.2 to 10 µg/ml, and metabolically labeled with [35S]methionine/cysteine. vIL-6 was immunopre-cipitated from lysates with {alpha}-FLAG and visualized by fluorography. Greater than 95% recombinant (Fig. 1a) and virally produced (Fig. 1b) vIL-6 from untreated cells migrates as a single band with an apparent molecular mass of 28–30 kDa in untreated cells. vIL-6 from these untreated cells also migrates as two minor bands at 24 and 26–27 kDa, representing unglycosylated and monoglycosylated forms, respectively. In contrast, treatment with 2.0–10.0 µg/ml tunicamycin caused a shift in band mobility to 24 kDa and an intermediate band was present between 24 and 29 kDa at low tunicamycin concentrations (0.2–1.0 µg/ml), indicating utilization of both glycosylation sites (Fig. 1a). These shifts corresponded to the mobility of the two minor bands observed in untreated cells. These data indicated that N-glycosylation of vIL-6 at both possible sites is complete.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 1.
vIL-6 is modified with two asparagine-linked oligosaccharides. a, vIL-6 was transiently expressed in [35S]Met/Cys-labeled HEK 293 cells treated with increasing amounts of tunicamycin, a specific inhibitor of N-linked glycosylation (35). Tunicamycin was used in concentrations ranging from 0.2 to 10 µg/ml 1 h prior to labeling, during labeling, and throughout chase period. vIL-6 was immunoprecipitated from lysates with {alpha}-FLAG (M2). Empty vector-transfected and untreated cells were used as controls. b, metabolically labeled vIL-6 was immunoprecipitated from BCBL-1 lysates and media with vIL-6 antiserum after a 4-h chase period and exhaustively digested with Endo H. Digested protein (+) was compared with untreated protein (–). An intermediate band (mixed) reflects the presence of one Endo H-sensitive (high-mannose) and one insensitive (complex) N-glycan. Immunoprecipitation (IP) with preimmune serum served as a control. Fluorogram was overexposed to show light bands representing Endo H-digested vIL-6 from medium. Due to overexposure, a nonspecific band is present in the control lane containing protein immunoprecipitated from medium with preimmune serum. c, the relative percentages of Endo H-sensitive (high-mannose) or Endo H-insensitive (complex) N-glycans were determined by densitometric analysis of band intensities. Results were identical in two independent experiments. L, lysate; M, medium.

 
N-Linked glycans are added to nascent polypeptide chains en bloc in the endoplasmic reticulum. At this point, all N-glycans are identical in composition and contain mostly mannose residues. The structure of the common core high-mannose oligosaccharide is GlcNAc2Man9Glc3, where GlcNAc is N-acetylglucosamine, Man is mannose, and Glc is glucose. These glycans are modified in the Golgi apparatus, where mannose residues are replaced with other sugar units, creating more complex structures. Endo H only cleaves immature high-mannose N-glycans (36, 37) and was used to assess the extent of vIL-6 N-glycan maturation. Because vIL-6 contains only two possible N-glycans, its glycan profile can be deduced from SDS-PAGE mobility shifts produced by Endo H treatment by measuring band intensities. Figs. 1b and 2a show that, in lysates of HEK 293 cells transfected with a vIL-6 expression vector and BCBL-1 cells, both N-glycans can be completely removed by the treatment of immunoprecipitated vIL-6 with Endo H, indicating that they are modified by immature high-mannose N-linked oligosaccharides (36, 37). In HEK 293 cells, all of the secreted vIL-6 contains at least one high-mannose, Endo H-sensitive glycan, and nearly 50% contains two immature glycans (Fig. 2a).



View larger version (46K):
[in this window]
[in a new window]
  		FIG. 2.
The vIL-6 amino-terminal hydrophobic sequence targets hIL-6 for secretion but is not cleaved. a, wild-type vIL-6, transiently expressed in HEK 293 cells, was treated with indicated endoglycosidases and compared with unglycosylated in vitro translated protein. For each lane, 3–4 x 106 cells were labeled with [35S]Met/Cys for 30 min and harvested after a 5-h chase period. Protein was immunoprecipitated with vIL-6 antiserum. Empty vector-transfected cells, untreated cells, and immunoprecipitation (IP) with preimmune serum were used as controls. Visualization was done by fluorography. b, in vitro (iv) translated hIL-6 and vhIL-6 were compared with protein transiently expressed in HEK 293 cells. For each lane, 3–4 x 106 cells were labeled with [35S]Met/Cys for 30 min and harvested after a 1-h chase period. Cells were treated with 5 µg/ml tunicamycin for 1 h prior to and during pulse and chase. Protein was immunoprecipitated from normalized lysates (L) and media (M) with {alpha}-FLAG (M2) and visualized by fluorography.

 
The glycosylation profile of secreted vIL-6 was also analyzed in the PEL cell line BCBL-1 to assess the maturity of naturally produced vIL-6 glycans. Protein was immunoprecipitated with vIL-6 antiserum and exhaustively digested overnight with an excess of Endo H, and then fresh enzyme was added in the morning and digestion was continued for 3 h more. In these cells, secreted vIL-6 was partially sensitive to Endo H (Fig. 1b). Endo H digested vIL-6 from lysate served as a control for complete digestion. Overexposure of lysate lanes was necessary to show low levels of secreted protein in the medium. It was determined that 28% vIL-6 secreted from BCBL-1 contains two Endo H-sensitive glycans and that 17% contains two glycans that are not sensitive to the enzyme by densitometric analysis (Fig. 1c). The remaining 55%, represented by an intermediate band, contains one sensitive and one insensitive glycan. 12-O-Tetradecanoylphorbol-13-acetate treatment did not alter the secreted glycan profile of vIL-6 but did enhance vIL-6 expression (data not shown). Therefore, in HEK 293 cells, all of the secreted recombinant vIL-6 contains at least one high-mannose glycan compared with 83% secreted naturally produced vIL-6 in BCBL-1 cells. Furthermore, ~75% N-glycans from secreted vIL-6 produced in HEK 293 cells are high-mannose and ~56% from vIL-6 produced in KSHV-infected BCBL-1 cells are high-mannose.

Treatment with BFA, a well characterized inhibitor of the classical secretory pathway, inhibits vesicle trafficking that generates the Gogi apparatus, causing the Golgi apparatus to be disassembled and absorbed by the endoplasmic reticulum (34, 38). This treatment caused a tunicamycin-sensitive (Fig. 3c) SDS-PAGE band representing singly N-glycosylated hIL-6 to shift to a higher mobility (Fig. 3b). This result was expected, because the vast majority of N-glycosylated hIL-6 species contains one complex N-glycan, and maturation of N-glycans to the complex form is known to occur in the Golgi apparatus (39, 40). Furthermore, BFA has been shown by others to inhibit the terminal glycosylation of oligosaccharide chains (34). BFA treatment did not cause a similar shift in vIL-6 band mobility (Fig. 3a). Because vIL-6 contains two N-glycans compared with one for hIL-6 and the two proteins are nearly the same size (hIL-6 is slightly larger by 8 amino acids), extensive modification of vIL-6 oligosaccharides in the Golgi would be expected to yield a comparable or larger bandshift on BFA treatment than that observed with hIL-6. Bands representing N-glycosylated hIL-6 were broad due to the heterogeneous nature of its N-glycans (Fig. 3b) (39, 41). vIL-6 bands were very compact by comparison (Fig. 3a), suggesting that its N-glycans are homogeneous. These results are consistent with Endo H experiments, which showed that secreted vIL-6 contains mostly immature high-mannose glycans (Figs. 1b and 2a), and suggest that Endo H-insensitive forms of vIL-6 are not extensively modified in the Golgi apparatus and may contain hybrid rather than highly complex structures.



View larger version (41K):
[in this window]
[in a new window]
  		FIG. 3.
Secretion of recombinant hIL-6 and vIL-6 in HEK 293 cells. vIL-6 (a) and hIL-6 (b) are not present in medium from transiently transfected HEK 293 cells treated with BFA, an inhibitor of the classical secretion pathway (34). Cells were labeled with [35S]Met/Cys for 20 min 1 h after drug treatment and then chased in "cold" medium for 2 h Each lane represents protein immunoprecipitated from 3–4 x 106 cells with vIL-6 antiserum (vIL-6) or {alpha}-FLAG (M2) (hIL-6). Visualization was done by fluorography. Lysates (L) and media (M) were normalized so that band intensities can be compared with the determined percentage of secreted protein. Vector-transfected cells (c) and immunoprecipitation (IP) with preimmune sera (Pre) were used as controls. c, the glycosylation species of hIL-6 represented by three major SDS-PAGE bands have previously been identified by Gross et al. (39) and are indicated. An increase in the mobility of these bands on BFA treatment is marked with an asterisk. c, cells expressing hIL-6 were treated with 5 µg/ml tunicamycin (T) or not (U) to specifically inhibit N-linked glycosylation but not O-linked glycosylation (35).

 
vIL-6 transiently expressed in HEK 293 cells was immunoprecipitated with vIL-6 antiserum and treated with PNGase F, which removes all N-linked glycans, regardless of their modification (36). The apparent molecular mass of PNGase F-treated vIL-6 was comparable with that of in vitro translated protein (Fig. 2a). This finding suggests that vIL-6 does not contain O-linked glycans or other large post-translational modifications and verifies the presence of N-linked glycans. O-Glycosylation of hIL-6 at an unknown but small number of sites (39, 41) leads to a shift in band mobility of mature protein on SDS-PAGE (Fig. 3b). This shift did not take place in cells treated with BFA (Fig. 3b), which disrupts the Golgi apparatus where O-glycosylation of secretory proteins takes place (34, 40). A lower concentration of BFA caused incomplete inhibition of O-glycan addition observed as a doublet, representing O-glycosylated and unglycosylated hIL-6 (Fig. 3b). This treatment did not cause a change in the band mobility of vIL-6, indicating that vIL-6 does not likely contain O-linked glycans to an extent comparable with hIL-6. Therefore, results from experiments with an inhibitor of O-glycan addition and endoglycosidase-treated protein are in agreement. These results taken together with the observation that both secreted and intracellular vIL-6s are observed as one sharp homogeneous band on SDS-PAGE, unlike hIL-6 (Fig. 3), offer evidence that vIL-6 is not extensively modified by O-linked glycans.

Signal Sequence Cleavage—Secreted proteins have an NH2-terminal signal sequence that is removed co-translationally in the endoplasmic reticulum (42). The removal of this signal causes secretory proteins to be released into the lumen of the endoplasmic reticulum. Inefficiently cleaved or uncleaved signal sequences are usually signal anchors that tether protein to membrane. A comparison of endoglycosidase-treated vIL-6 with in vitro translated vIL-6 suggested the possibility that the signal sequence of secreted vIL-6 might not be cleaved (Fig. 2a). In that experiment, immunoprecipitated wild-type vIL-6 from HEK 293 cells and medium was treated with endoglycosidases Endo H and PNGase F and its SDS-PAGE mobility was compared with in vitro translated vIL-6. Endoglycosidase-treated vIL-6 from cell culture medium had an apparent molecular mass identical to in vitro translated protein (Fig. 2a).

Attempts were made to obtain an NH2-terminal amino acid sequence of secreted vIL-6 purified from PEL cell culture supernatants to verify cleavage of the signal sequence of the protein, but vIL-6 was not present in extracellular concentrations high enough to make this possible. It was also not possible to purify enough secreted vIL-6 for matrix-assisted laser desorption ionization time-of-flight mass spectroscopy analysis.

Therefore, to determine whether or not the signal sequence of vIL-6 is cleaved, the apparent molecular mass of transiently expressed vhIL-6 and hIL-6 immunoprecipitated from lysates and medium of tunicamycin-treated HEK 293 cells was compared with that of in vitro translated protein (Fig. 2b). Tunicamycin specifically inhibits the addition of N-glycans to nascent polypeptides; thus, protein mobility can be compared with unglycosylated in vitro translated protein by SDS-PAGE. O-Glycosylation causes only minor shifts in band mobility; however, N-glycosylation causes very large shifts compared with those produced by signal sequence cleavage. In vitro translated protein is not modified by glycosylation or signal sequence cleavage in the absence of microsomes. Because hIL-6 has a signal sequence that is cleaved (43–45), it was used as a positive control. hIL-6 migrated faster than vhIL-6 and in vitro translated hIL-6 (Fig. 2b). Conversely, the mobility of vhIL-6 secreted from HEK 293 cells was identical to in vitro translated vhIL-6. The 2.5–3.0-kDa shift in the apparent molecular mass of recombinant hIL-6 compared with in vitro translated protein corresponded to the predicted molecular mass of its cleaved signal sequence (~2.8 kDa). The putative signal sequence of vIL-6 is approximately the same length as the cleaved signal peptide of hIL-6 (22 amino acids compared with 27 amino acids, respectively) (Fig. 5a) and has a predicted molecular mass of 2.5 kDa. The two proteins have identical amino acid sequences downstream of their small NH2-terminal signal sequences and should have identical post-translational modifications. This is consistent with the observation that the SDS-PAGE banding patterns of vhIL-6 and hIL-6 were identical (Fig. 2b). Therefore, these data suggest that the vIL-6 signal sequence is not cleaved.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 5.
Generation of chimeric proteins. a, sequence alignment of vIL-6 and hIL-6 signal sequences. Identical (|) and similar (:) residues as well as positively charged (+), negatively charged (–), and proline residues (^) are indicated. The hydrophobic (H) regions of the signals are underlined. The NH2-terminal (N) and COOH-terminal (C) regions of the signal sequences are shown. The hydrophobic region of the vIL-6 signal is much shorter than that of hIL-6 and is flanked by positively charged amino acids. The proposed cleavage site of the vIL-6 signal according to the program SignalP, version 1.1 (56) and the known scissile bond of the hIL-6 (41, 43) signal are indicated (asterisk). The junction points of the chimeric proteins vhIL-6 and hvIL-6 are marked (~). b, cartoon depicting wild-type and chimeric proteins. The first 40 amino acids of each protein were swapped and attached by a five amino acid linker.

 
Comparison of vIL-6 and hIL-6 Secretion—The possibility that the signal sequence of vIL-6 is not cleaved suggested that the protein might not be efficiently secreted. Because low levels of vIL-6 found in the medium could have been the result of cell lysis, BFA, a well characterized inhibitor of the classical secretory pathway, was used to determine whether or not the secretory apparatus is required for translocation of vIL-6 to the medium (34, 38). In the medium from untreated cells, the levels of vIL-6 were much lower than the levels of hIL-6 at the same 2-h time point (Fig. 3). vIL-6 was not detected in culture medium of cells treated with BFA, indicating that vIL-6 is secreted from cells, albeit very slowly (Fig. 3a). The secretion of hIL-6, which was used as a control, was also completely inhibited by BFA, even though significantly more hIL-6 than vIL-6 was secreted at the 2-h time point in the absence of drug treatment.

To determine the efficiency of vIL-6 secretion, BCBL-1 B-cells derived from PEL were analyzed by pulse-labeling of proteins with [35S]methionine/cysteine followed by "cold" chase in medium containing a vast excess of unlabeled amino acids. After the chase, aliquots of membrane preparations or medium were immunoprecipitated with vIL-6 antiserum. These membrane preparations also contain membranous organelles such as endoplasmic reticulum and Golgi apparatus. Densitometry of bands from pulse-chase time courses revealed that vIL-6 has a half-life of ~5–7 h in membrane preparations from BCBL-1 cells and was still detectable after 21 h (Fig. 4). vIL-6 was not present in the soluble cytoplasmic cell fraction (Fig. 4) and was not present in nuclear BCBL-1 fractions (data not shown), indicating exclusive targeting of vIL-6 to microsome-containing membrane fractions. Relatively low amounts of vIL-6 were present in the medium.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 4.
Expression of vIL-6 in BCBL-1 cells over time. Cells were labeled with [35S]Met/Cys for 20 min and then chased in "cold" media for the indicated times for pulse-chase analysis. For each time point, protein was immunoprecipitated from a membrane preparation and medium from 107 cells with vIL-6 antiserum. Immunoprecipitation (IP) of membrane preparation with preimmune sera (Pre) and soluble cytoplasmic fraction (S) with vIL-6 antiserum were used as controls. Visualization was done by fluorography.

 
The secretion rate of vIL-6 was more stringently analyzed by comparing the biosynthesis of vIL-6 and hIL-6 transiently expressed in [35S]methionine/cysteine-labeled HEK 293 cells by pulse-chase analysis. COOH-terminally FLAG-tagged vIL-6 and hIL-6 were constructed and used for these experiments so that the two proteins could be evaluated using the same {alpha}-FLAG (M2) antibody (Fig. 5b). Proteins were immunoprecipitated from aliquots of normalized lysates and media taken at several time points. This experiment showed that recombinant vIL-6 is secreted much less efficiently than hIL-6 (Fig. 6). Densitometric analysis showed that the half-time of vIL-6 secretion was at least eight times longer than that of hIL-6. The half-time of hIL-6 secretion was ~0.5 h, whereas vIL-6 had a secretion half-time of ~4 h, consistent with the half-life of vIL-6 in BCBL-1 membrane preparations (Fig. 4).



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 6.
Secretion kinetics of wild-type vIL-6 and hIL-6 compared with chimeras. Pulse-chase time courses of [35S]Met/Cys-labeled HEK 293 cells transiently expressing FLAG-tagged vIL-6, hIL-6, hvIL-6, or vhIL-6 were used to compare intracellular and secreted protein levels by fluorography. For each time point, 3–4 x 106 cells were pulsed with labeled amino acids for 20 min and then chased in medium containing an excess of unlabeled amino acids for the indicated times. Protein was immunoprecipitated (IP) with {alpha}-FLAG (M2) monoclonal antibody or irrelevant isotype control antibody (Control). Empty vector-transfected cells (C) were also used as controls. Lysates (L) and media (M) were normalized so that band intensities can be compared with the determined percentage of secreted protein.

 
Because much lower proportions of recombinant vIL-6, in comparison to hIL-6, were detected in the media from HEK 293 cell cultures compared with amounts found in lysates, the possibility that vIL-6 might be preferentially reabsorbed by these cells after secretion was investigated. HEK 293 cells were labeled for 1 h with [35S]methionine/cysteine and chased in cold medium for 4 h. Conditioned medium from these cells, which contained labeled vIL-6, was applied to unlabeled HEK 293 cells (Fig. 7). Unlabeled cells were then harvested at 1- or 5-h time points, and vIL-6 was immunoprecipitated from lysates and media with vIL-6 antiserum. The levels of vIL-6 present in the lysates and media of unlabeled cells were compared with the levels of the protein in labeled vIL-6-expressing cells. vIL-6 levels did not decrease over time in the media from unlabeled cells, and vIL-6 was not detected in the lysates from these cells. However, the cells absorbed other labeled proteins from the medium by the 5-h time point. Therefore, the prolonged association of vIL-6 with cells is not due to uptake by cells but rather to slow secretion.



View larger version (46K):
[in this window]
[in a new window]
  		FIG. 7.
Secreted vIL-6 is not reabsorbed by cells. HEK 293 cells were transiently transfected with pCDNA3.1-vIL-6 and metabolically labeled 24 h later with [35S]Met/Cys for 1 h and then chased in medium containing an excess of unlabeled amino acids for 4 h (input). 2 ml of conditioned medium from 3–4 x 106 labeled cells was applied to an equal amount of unlabeled cells for 1 or 5 h. Protein was immunoprecipitated (IP) from normalized lysates (L) and media (M) with vIL-6 antiserum or preimmune serum (Pre). Visualization was done by fluorography.

 
The Amino-terminal Sequence of vIL-6 Is Not Responsible for Slow Secretion—To characterize the putative signal sequence of the viral protein, two chimeric genes were constructed. One contains the coding sequence for the first 40 amino acids of vIL-6 and the coding sequence for amino acids 41–209 of hIL-6 (vhIL-6). The other contains the coding sequence for the portion of vIL-6 COOH-terminal to its putative signal sequence for amino acids 41–205 and the coding DNA for the first 40 amino acids of hIL-6 (hvIL-6) (Fig. 5). The exchanged NH2-terminal regions contain the 28 amino acid signal sequence of hIL-6 and the 22 amino acid putative signal sequence of vIL-6. These constructs do not have any deleted coding sequence beyond the first 120 bases that were swapped. The DNAs were constructed using a primer-generated linker sequence coding for Gly-Gly-Ala-Gly-Gly.

The two chimeras were compared with both wild-type proteins to test whether or not the vIL-6 NH2-terminal sequence is responsible for the slow secretion of the viral protein compared with that of hIL-6 (Fig. 6). HEK 293 cells were transfected with the two constructs, and 24 h later, pulse-chase experiments were performed as described earlier. The NH2-terminal sequence of each protein does not seem to determine the efficiency of secretion, as both the vIL-6 and hIL-6 sequences can target hIL-6 for secretion at similar levels. Conversely, vIL-6 is still secreted nearly as slowly as wild type when targeted by the hIL-6 signal (Fig. 6). Secretion of hvIL-6 is slightly improved compared with wild-type vIL-6 but not to levels of wild-type hIL-6. Likewise, the secretion rate of hIL-6 is only slightly reduced by the vIL-6 signal sequence. This may indicate that the signal sequence of vIL-6 is marginally less effective than the hIL-6 signal sequence or that the secretion kinetics of the chimeric proteins are altered moderately due to structural changes caused by the addition of heterologous signals. Therefore, the NH2-terminal sequence of vIL-6 does not contribute significantly to the slow secretion rate of vIL-6. Secretion of chimeras via the classical route was verified by BFA inhibition of vesicle traffic (data not shown), as described previously, to ensure that results were not due to mistargeting of chimeric proteins.

Retention of vIL-6 in the Absence of Receptor—To determine whether the cause of slow secretion of vIL-6 is associated with its receptor, human gp130, secretion of vIL-6 was measured over time in a stably transfected IL-3-dependent murine pro-B-cell line BAF-B03 that does not express gp130 (31). A BAF-B03-derived cell line that was stably transfected with human gp130 and an empty vector control cell line was used to generate vIL-6-expressing cells. Protein was immunoprecipitated with vIL-6 antiserum from the lysates and media of vIL-6-expressing BAF-B03 cells that express gp130 and cells that do not express the receptor by pulse-chase experiments (Fig. 8), as described previously.



View larger version (38K):
[in this window]
[in a new window]
  		FIG. 8.
Secretion kinetics of vIL-6 in the presence and absence of gp130. a, pulse-chase time courses were used to compare vIL-6 expression in normalized lysates (L) and media (M) of stably transfected, [35S]Met/Cys-labeled gp130-negative and gp130-positive BAF-B03 cells. Cells were harvested at indicated time points, and protein was immunoprecipitated with vIL-6 antiserum or preimmune serum (Pre). Empty vector-transfected cells (C) were also used as controls. For gp130-positive BAF-B03 cells, secreted vIL-6 is compared with in vitro translated protein (iv). Fluorograms were exposed to film for 90 days. b, flow cytometry showing gp130 surface expression in BAF-B03 cells. Cells were stained with 3 µg/ml AM64 monoclonal antibody to human gp130. Antibody to irrelevant protein was used as an isotype control for all three cell lines and produced peaks that overlap with gp130-negative BAF-B03 cells stained with {alpha}-gp130 (AM64).

 
In the absence of gp130, vIL-6 secretion was not detected for up to 4 h but vIL-6 was secreted very efficiently to the medium as early as 0.5 h in cells that express gp130 (Fig. 8a). vIL-6 secretion was more efficient in gp130-expressing BAF-B03 cells than in any other cell type analyzed. These cells expressed much more gp130 than BCBL-1 cells (Fig. 8b) and much less vIL-6 than BCBL-1 or transiently transfected HEK 293 cells (data not shown). Because gp130-expressing BAF-B03 cells express lower levels of vIL-6 than control BAF-B03 cells, the lack of detectable vIL-6 secretion in control cells cannot be attributed to low levels of expression. Even though vIL-6 can induce STAT3 phosphorylation in murine fibroblasts and gp130-expressing BAF-B03 cells (data not shown), it cannot induce STAT3 phosphorylation in control cells that do not express the IL-6 signal transducer, gp130 (data not shown). This finding is consistent with the fact that BAF-B03 cells do not express murine gp130, as described by Hibi et al. (31).

Effect of vIL-6 on gp130 N-Linked Glycan Maturation—The observation that vIL-6 is not secreted in the absence of gp130 suggests that vIL-6 interacts with the receptor intracellularly. To test this hypothesis, the effect of vIL-6 expression on the maturation of gp130 N-glycans was measured to demonstrate a direct interaction between gp130 and vIL-6 in early secretory compartments. gp130 is modified by nine N-linked glycans (46). Maturation of these glycans was analyzed in the presence and absence of vIL-6 expression. Endogenous gp130 immunoprecipitated from HEK 293 cells stably transfected with vIL-6 or vector control plasmid (pCDNA3) was treated or not with Endo H, which only cleaves immature high-mannose glycans (Fig. 9). If gp130 glycan maturation were inhibited by association with vIL-6, an increase in the proportion of immature gp130 glycans, represented by an increase in band mobility of Endo H-treated protein, would be observed. The removal of one unmodified high-mannose N-linked glycan resulted in a 2.5–3.0-kDa increase in band mobility, as observed with Endo H-treated vIL-6 (Fig. 2a).



View larger version (44K):
[in this window]
[in a new window]
  		FIG. 9.
vIL-6 prevents the maturation of gp130 glycans. HEK 293 cells were stably transfected with pCDNA3-vIL-6 or empty pCDNA3 vector. Left panel, protein was immunoprecipitated (IP) from lysates of 107 cells with {alpha}-gp130 (AM64) or isotype control antibody (Iso). Precipitated protein was then digested with Endo H (+) or not (–) and visualized by Western blotting (WB) using {alpha}-gp130 (AM64). gp130 was resolved for 6 h on a large 5% polyacrylamide gel to enable the detection of relatively small molecular mass differences caused by immature N-glycan removal, resulting in band-broadening. Right panel, whole-cell lysates from stably transfected HEK 293 cell lines were analyzed for vIL-6 expression by Western blotting with vIL-6 antiserum.

 
Because gp130 is a large protein, it was resolved for over 6 h on a large gel by SDS-PAGE to observe relatively small molecular mass differences. Endo H treatment of immunoprecipitated gp130 did cause an increase in band mobility in both vIL-6-expressing and control cells, but the increase in band mobility was more pronounced in vIL-6-expressing cells. gp130 bands were also broader in vIL-6-expressing cells, suggesting that the glycosylation state of the receptor may be more heterogeneous in these cells (Fig. 9). This experiment was repeated using another vIL-6-expressing stable HEK 293 cell line and yielded identical results (data not shown). Very high vIL-6 expression was shown in both stable HEK 293 cell lines by Western blotting of whole cell lysates with vIL-6 antiserum (Fig. 9c).

vIL-6 Induces gp130 and STAT3 Signaling in the Absence of Secretion—To determine whether intracellular vIL-6 can induce STAT3 signaling and activation of the hIL-6 signal transducer gp130, the well characterized 14 amino acid ER retention motif of PDI, ~MEEDDDQKAVKDEL-COOH, was appended to the carboxyl terminus of vIL-6 (vIL-6-KDEL). The ER retention motif of PDI contains a COOH-terminal KDEL motif (underlined), but since this motif alone does not completely prevent hIL-6 secretion (44), the larger sequence was used. The 14 amino acid PDI sequence was previously shown to completely prevent secretion of hIL-6 (44).

Secretion of PDI-tagged vIL-6 and a control vIL-6 containing a mutant PDI tag, ~MEEDDDQKAVKDQA-COOH (vIL-6-KDQA), was measured by pulse-chase analysis, as described previously. HEK 293 cells were transfected with vIL-6-KDEL and vIL-6-KDQA in the context of pCDNA3 and pulsed with [35S]methionine/cysteine for 30 min 24 h later. After a 4-h chase period in "cold" medium, the lysates and media were collected and vIL-6 was immunoprecipitated with vIL-6 antiserum. This analysis showed that vIL-6-KDEL was not secreted at detectable levels for up to 4 h but that vIL-6-KDQA was secreted at levels similar to wild type (Fig. 10a).



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 10.
vIL-6 containing an ER retention motif is not secreted but can induce gp130 and STAT3 phosphorylation. a, a 14 amino acid OOOH-terminal ER retention motif (KDEL) or a mutant ER retention motif (KDQA) was appended to the COOH terminus of vIL-6. HEK 293 cells were transiently transfected with the indicated expression vectors and labeled for 30 min with [35S]Met/Cys 24 h. later. vIL-6 was immunoprecipitated (IP) from normalized lysates (L) and media (M) with vIL-6 antiserum or preimmune serum (Pre) after a 4-h chase period and visualized by fluorography. b, HEK 293 cells were transiently transfected with the indicated constructs and harvested 24 h later. Samples were split so that gp130 and STAT3 phosphorylation could be assessed in parallel. The status of gp130 phosphorylation was assessed by immunoprecipitation with {alpha}-gp130 (AM64) followed by Western blotting with {alpha}-phospho-tyrosine (pTyr) (PY20). The status of STAT3 phosphorylation was determined by Western blotting of whole-cell lysates using {alpha}-pSTAT3 (B-7). Membranes were stripped and reprobed with {alpha}-STAT3 (F-2) or {alpha}-gp130 for STAT3 and gp130 controls, respectively. vIL-6 expression was also detected in these whole-cell lysates using vIL-6 antiserum. As a loading control, the vIL-6 blot was also probed with {alpha}-actin (C-20).

 
STAT3 and gp130 phosphorylation was measured in parallel to determine whether vIL-6-KDEL could induce the activation of these molecules. HEK 293 cells were transfected with a human gp130 expression vector, pEF-BOS-gp130, and either wild-type vIL-6, vIL-6-KDEL, vIL-6-KDQA, or empty vector (Fig. 10b). 24 h later, gp130 was immunoprecipitated from lysates with {alpha}-gp130 and gp130 phosphorylation was detected by Western blotting with {alpha}-phospho-tyrosine. This blot was stripped and reprobed with {alpha}-gp130 to control for gp130 expression. Aliquots of these lysates were also analyzed by Western blotting to measure STAT3 phosphorylation at tyrosine 705 and vIL-6 expression using {alpha}-pSTAT3 and vIL-6 antiserum, respectively. For loading controls, the STAT3 blot was stripped and reprobed with {alpha}-STAT3 and the vIL-6 blot was probed with {alpha}-actin.

This experiment showed that vIL-6-KDEL can induce STAT3 and gp130 phosphorylation (Fig. 10b), despite its inability to be secreted. This is in contrast to PDI-tagged hIL-6, which is unable to signal through gp130, despite binding to the IL-6R{alpha} subunit intracellularly (44). vIL-6-KDEL and its secreted control vIL-6 KDQA induced STAT3 and gp130 phosphorylation at identical levels, but wild-type vIL-6 induced slightly higher levels of phosphorylation. This may reflect steric problems or conformational changes caused by the addition of a 14 amino acid PDI tag.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Similar to hIL-6 (45), vIL-6 has two potential N-linked glycosylation sites, but unlike hIL-6, vIL-6 is completely glycosylated at both sites. Only ~46% hIL-6 is N-glycosylated at a single site, but nearly all of it is O-glycosylated at an unknown number of sites (39, 41, 47). A very large proportion of secreted vIL-6 contains immature high-mannose glycans (Figs. 1c and 2a) compared with a very small proportion of secreted hIL-6 (39, 41). The N-glycan profile of hIL-6 secreted from human primary monocytes was described by Parekh et al. (41). 54% were the complex (not sensitive to Endo H), 8% were high-mannose (Endo H-sensitive), and the remainder was a small tetrasaccharide di-mannose core (Endo-H resistant) (41). Because the hIL-6 glycosylation profile observed by us in HEK 293 cells is identical to that described by others in primary human monocytes (39, 41), our observations concerning vIL-6 glycosylation in HEK 293 cells are not likely to be due to an anomaly of this cell line (Fig. 3b). Furthermore, similar results were found in vIL-6 secreted from BCBL-1 cells and HEK 293 cells. ~75% N-glycans from secreted vIL-6 produced in HEK 293 cells are high-mannose and ~56% from vIL-6 produced in KSHV-infected BCBL-1 cells are high-mannose. This high proportion of immature N-glycans is consistent with the lack of SDS-PAGE band heterogeneity observed for secreted vIL-6. Greater than 95% of the protein migrates in one very sharp band at 28–30 kDa. This is in stark contrast to hIL-6, which migrates as a more diffuse band of variably modified complex N-glycans (Fig. 3b) (39).

The presence of high-mannose glycans may indicate that vIL-6 is limited to local extracellular concentrations by circulating and liver mannose receptors (48, 49). Proteins containing high-mannose sugars are removed from circulation by the liver and spleen, whereas proteins with complex N-glycans are not (48–50). For example, liver mannose receptors immaturely clear glycosylated amylase A from mouse serum with a t1/2 of 9 min, whereas unglycosylated amylase is not cleared and uptake is specifically blocked by yeast mannan and mannose-albumin (51).

Other than vIL-6, no other cytokine has been shown to contain such a high proportion of immature glycans. Granulocytecolony stimulating factor (52) and hIL-6 (39, 41) are two of only a handful of cytokines with characterized N-glycosylation, and both contain an extremely high proportion of complex N-glycans. Interferon-{gamma} from primary human T-cell culture supernatants also has a very high proportion of complex N-glycans at two sites. Only a very small percentage of molecules contain Endo H-sensitive hybrid (13%) or high-mannose (10%) structures at a single site (53). Some other types of secreted proteins, such as immunoglobulins IgM and IgA, do contain a very small proportion of high-mannose glycans, but they have been shown to be inaccessible to exoglycosidases that would normally modify them in the Golgi apparatus due to blocking by polymerization or association with J chain that occurs in early secretory compartments, respectively (54, 55). These immunoglobulin high-mannose glycans are also inaccessible to mannose receptors.

Signal sequences are cleaved co-translationally in the endoplasmic reticulum (42). Attempts to identify the signal sequence cleavage site failed, but other experiments revealed that, despite its secretion, the vIL-6 signal peptide does not seem to be cleaved. When the mobility of secreted hL-6 and vhIL-6 was compared with that of in vitro translated protein by SDS-PAGE, secreted hIL-6 was shown to be increased relative to in vitro translated protein, but the mobility of secreted vhIL-6 was not increased (Fig. 2b). Furthermore, when secreted vIL-6 was treated with the endoglycosidases, it migrated at the same rate as in vitro translated vIL-6 (Fig. 2a). The cleaved hIL-6 signal sequence is 27 amino acids in length (41, 43) compared with the predicted vIL-6 signal peptide length of 22 amino acids using the program SignalP, version 1.1 (56). The first 22 amino acids of vIL-6 have a predicted molecular mass of 2.5 kDa or greater than 10% total molecular mass of the protein. A shift of this size would be as easily seen on the gel as the 2.8-kDa shift observed for hIL-6 (Fig. 2a). Other than the small hydrophobic domain of the vIL-6 signal sequence, hIL-6 and vhIL-6 only differ by 13 and 18 amino acids in their C-regions immediately downstream of their known or putative cleavage sites, respectively (Fig. 5a). This region of the vIL-6 sequence does not contain any consensus sites for large post-translational modifications such as glycosylation that could preferentially retard the migration of vhIL-6. Therefore, it is unlikely that such modifications could confound the interpretation of these results.

Fibroblast growth factors-9 and -16 and carp retinal-binding protein are the only other examples of secreted proteins that have uncleaved true amino-terminal signal sequences (57–59). These proteins are all secreted via the classical route, but they have unusual unrecognizable signal sequences that are much less hydrophobic than classical leucine-rich amino-terminal signal sequences (56, 60). Unlike vIL-6, the fibroblast growth factor-16 signal cannot target heterologous protein for secretion (58) and probably contains an additional required internal signal sequence. Therefore, vIL-6 is unique in that it has a classical highly hydrophobic amino-terminal signal sequence that does not appear to be cleaved.

Although very high levels of extracellular recombinant vIL-6 have been shown to promote the growth of B-cells (22, 23, 25, 26), bona fide secretion of the protein has not been previously demonstrated. The data presented here show that vIL-6 is secreted much more slowly than hIL-6. Furthermore, we demonstrated for the first time that the secretion kinetics of hIL-6 is similar to that of interferon-{gamma}, which also has a half-time of secretion of ~20–25 min (53). To our knowledge, the secretion kinetics of other cytokines have not been determined on a small time-scale measured in hours, similar to the one presented here.

Even though little vIL-6 was found in tissue culture medium, secretion does take place. This was verified by using a well characterized inhibitor of secretion, demonstrating the need for an intact Golgi apparatus for vIL-6 translocation to the medium. Furthermore, the protein was found to be N-glycosylated, indicating passage through the endoplasmic reticulum. Secretion of wild-type vIL-6 was much slower than the secretion of hIL-6 (Fig. 6), but because the secretion rate of vIL-6 and hIL-6 was virtually unchanged when their signal sequences were exchanged, the retention of wild-type vIL-6 could not be attributed to an inefficient secretion signal.

The vIL-6 NH2-terminal sequence was able to target hIL-6 for secretion. hIL-6 is well characterized and cannot be secreted without a viable signal sequence (44, 45). Whether or not vIL-6 contains a functional internal secretion signal in addition to the amino-terminal signal is yet to be determined, but its NH2-terminal signal sequence is clearly sufficient for efficient secretion of heterologous protein. Uncleaved signal sequences usually function as signal anchors, tethering protein to membrane (61). The hydrophobic domains of signal anchors are typically between 20 and 30 amino acids in length. The short length of the hydrophobic region of the vIL-6 signal, only 15 amino acids compared with a more typical 20–30 amino acids, may prevent stable membrane insertion (62, 63). In contrast, the hydrophobic region of the hIL-6 signal is 24 amino acids in length (Fig. 5a). This might explain the presence of full-length vIL-6 found extracellularly (Fig. 2b).

Because vIL-6 has been shown to bind to gp130 directly without help from the high affinity IL-6 receptor (23, 27–29), the possibility that binding to gp130 could be responsible for intracellular retention of the protein was investigated in BAF-B03 cells (31). Surprisingly, our data showed that vIL-6 secretion could not be detected in the absence of gp130 but secretion was very efficient in cells that express high levels of gp130. Therefore, in these cells, gp130 protein expression may be required for vIL-6 secretion. This interesting possibility should be investigated further. The slow secretion of vIL-6, coupled with the observation that detectable vIL-6 secretion did not take place in the absence of gp130, suggests that vIL-6 may contain a novel ER retention motif that is masked by binding to its receptor intracellularly. Intracellular binding of vIL-6 to its receptor was also investigated by assessing gp130 N-glycan maturation. Those results suggested that the binding of vIL-6 to gp130 in pre-Golgi compartments prevents the maturation of one or more gp130 N-linked glycans and that vIL-6 binding may increase gp130 glycan heterogeneity by making these glycans inaccessible to Golgi exoglycosidases. It is important to note that changes in band mobility in this experiment were only observed when gp130 was treated with the endoglycosidase, Endo H. Therefore band shifts are only due to changes in gp130 carbohydrate composition and not truncated protein.

Rose-John et al. (44) have demonstrated that complete retention of hIL-6 in the ER could be achieved when the 14 COOH-terminal amino acids of protein-disulfide isomerase that contain the well characterized KDEL ER retention motif were added to its COOH terminus. This mutant IL-6 was not secreted and prevented the translocation of IL-6 receptor to the cell surface but could not induce signaling (44). Those results are consistent with those of Wang et al. (64) who have shown that IL-6 and other IL-6 family members only activate mature gp130 at the cell surface.

The experiments described here show that modification of vIL-6 with the same COOH-terminal ER retention motif used by Rose-John et al. (44) prevents its secretion (Fig. 10a) but that this vIL-6 mutant can induce phosphorylation of gp130 and STAT3 (Fig. 10b). Furthermore, inhibition of gp130 N-glycan maturation by vIL-6 (Fig. 9) strongly suggests that wild-type vIL-6 binds to the receptor in pre-Golgi compartments where N-glycan maturation occurs. Signaling by gp130 from intracellular compartments is theoretically possible because the association of Jak1, which imparts kinase activity to the receptor, to the cytoplasmic domain of gp130 occurs at the endoplasmic reticulum, and this binding is required for gp130 translocation to the cell surface (65).

The lack of detectable vIL-6 secretion in the absence of gp130, inhibition of gp130 glycan maturation by vIL-6, and gp130 activation by a non-secreted vIL-6 mutant are consistent with a model in which gp130 binds to vIL-6 in intracellular compartments and the two are transported together to the cell surface for signaling. vIL-6 may be released from its receptor at the cell surface due to extremely low binding affinity. This is reminiscent of the oncogenic viral platelet-derived growth factor homolog v-Sis encoded by simian sarcoma virus that causes fibrosarcoma in woolly monkeys. v-Sis is also secreted very slowly (66) and does not transform cells when applied to cells extracellularly (67). v-Sis interacts with its receptor intracellularly (68) but must travel to the cell surface for signaling (69). The v-Sis homodimer is modified with two N-linked oligosaccharides, one on each chain, that are both Endo H-sensitive in 70–80% molecules (66). The remainder contains one high-mannose and one complex glycan. Unlike vIL-6, N-glycans on secreted v-Sis molecules are removed by proteolytic processing of the protein in post-Golgi compartments (66). Because maturation of N-glycans occurs in the Golgi, modification of v-Sis N-glycans is also blocked by some mechanism. Therefore, vIL-6 shares several traits with v-Sis that make these proteins unique among secreted growth factors. They are secreted very slowly, decorated with an extremely high proportion of immature high-mannose glycans, and are associated with their receptors intracellularly.

vIL-6 is partially retained intracellularly by an unknown mechanism. Although it is secreted much slower than its cellular counterpart, its signal sequence efficiently targets hIL-6 for secretion at nearly wild-type levels. The protein is decorated with a high proportion of high-mannose N-glycans compared with other cytokines (39, 41, 52, 53) and must be present at extraordinarily high extracellular levels to promote growth of IL-6-dependent myeloma cells (22, 23, 70). The immature glycosylation of vIL-6 could be explained by intracellular interactions with its receptor, gp130, that render the protein inaccessible to Golgi exoglycosidases. This explanation is consistent with our findings that vIL-6 secretion could only be detected in cells that express gp130 and that vIL-6 expression can inhibit the maturation of gp130 asparagine-linked glycans. Finally, our results show that a vIL-6 mutant can activate gp130 and STAT3 despite the fact that it is not secreted. Taken together, these facts suggest a unique mode of action for vIL-6 compared with cellular IL-6 and other cytokines and indicate that the biologically significant binding of vIL-6 to its receptor may only occur intracellularly in vIL-6-expressing cells.


   FOOTNOTES
 
* This work was supported by Grant 1R01CA75894 from the National Institutes of Health (to P. G. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} Supported by Training Grant T32DA07245 from the National Institutes of Health. Back

§ To whom correspondence should be addressed. Tel.: 813-974-2372; Fax: 813-974-4151; E-mail: pmedvecz{at}hsc.usf.edu.

1 The abbreviations used are: KSHV, Kaposi's sarcoma-associated herpesvirus; IL, interleukin; IL-6, interleukin-6; vIL-6, viral interleukin-6; hIL-6, human interleukin-6; PEL, primary effusion lymphoma; MCD, multicentric Castleman's disease; STAT, signal transducer and activator of transcription; Endo H, endo-{beta}-N-acetylglucosaminidase H; HEK, human embryonic kidney; PDI, protein-disulfide isomerase; PNGase F, peptide N-glycosidase F; gp, glycoprotein; BFA, brefeldin A; ER, endoplasmic reticulum; TSA, Tris-buffered saline containing sodium azide. Back


   ACKNOWLEDGMENTS
 
We thank Dr. Masashi Narasaki for the generous gift of stable BAF-B03 pro-B-cell lines and the human gp130 expression plasmid. We thank Dr. G. Ciliberto (Istituto di Ricerche di Biologia Moleculare P. Angeletti, Pomezia (Roma), Italy) for human IL-6 cDNA and Dr. Rodney Arcenas for flow cytometry assistance. DNA sequencing was performed by the Molecular Biology Core Facility at H. Lee Moffitt Cancer Center.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Boshoff, C., and Weiss, R. A. (1998) Adv. Cancer Res. 75, 57–86[Medline] [Order article via Infotrieve]
  2. Dupin, N., Diss, T. L., Kellam, P., Tulliez, M., Du, M. Q., Sicard, D., Weiss, R. A., Isaacson, P. G., and Boshoff, C. (2000) Blood 95, 1406–1412[Abstract/Free Full Text]
  3. Dupin, N., Fisher, C., Kellam, P., Ariad, S., Tulliez, M., Franck, N., van Marck, E., Salmon, D., Gorin, I., Escande, J. P., Weiss, R. A., Alitalo, K., and Boshoff, C. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 4546–4551[Abstract/Free Full Text]
  4. Parravinci, C., Corbellino, M., Paulli, M., Magrini, U., Lazzarino, M., Moore, P. S., and Chang, Y. (1997) Am. J. Pathol. 151, 1517–1522[Abstract]
  5. Bromberg, J. F., Wrzeszczynska, M. H., Devgan, G., Zhao, Y., Pestell, R. G., Albanese, C., and Darnell, J. E., Jr. (1999) Cell 98, 295–303[CrossRef][Medline] [Order article via Infotrieve]
  6. Catlett-Falcone, R., Landowski, T. H., Oshiro, M. M., Turkson, J., Levitzki, A., Savino, R., Ciliberto, G., Moscinski, L., Fernandez-Luna, J. L., Nunez, G., Dalton, W. S., and Jove, R. (1999) Immunity 10, 105–115[CrossRef][Medline] [Order article via Infotrieve]
  7. Grandis, J. R., Drenning, S. D., Zeng, Q., Watkins, S. C., Melhem, M. F., Endo, S., Johnson, D. E., Huang, L., He, Y., and Kim, J. D. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 4227–4232[Abstract/Free Full Text]
  8. Bowman, T., Broome, M. A., Sinibaldi, D., Wharton, W., Pledger, W. J., Sedivy, J. M., Irby, R., Yeatman, T., Courtneidge, S. A., and Jove, R. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 7319–7324[Abstract/Free Full Text]
  9. Aoki, Y., Feldman, G. M., and Tosato, G. (2003) Blood 101, 1535–1542[Abstract/Free Full Text]
  10. Aoki, Y., Jaffe, E. S., Chang, Y., Jones, K., Teruya-Feldstein, J., Moore, P. S., and Tosato, G. (1999) Blood 93, 4034–4043[Abstract/Free Full Text]
  11. Liu, C., Okruzhnov, Y., Li, H., and Nicholas, J. (2001) J. Virol. 75, 10933–10940[Abstract/Free Full Text]
  12. Nishi, J., and Maruyama, I. (2000) Leuk. Lymphoma 38, 387–394[Medline] [Order article via Infotrieve]
  13. Cannon, J. S., Nicholas, J., Orenstein, J. M., Mann, R. B., Murray, P. G., Browning, P. J., DiGiuseppe, J. A., Cesarman, E., Hayward, G. S., and Ambinder, R. F. (1999) J. Infect. Dis. 180, 824–828[CrossRef][Medline] [Order article via Infotrieve]
  14. Parravicini, C., Chandran, B., Corbellino, M., Berti, E., Paulli, M., Moore, P. S., and Chang, Y. (2000) Am. J. Pathol. 156, 743–749[Abstract/Free Full Text]
  15. Staskus, K. A., Sun, R., Miller, G., Racz, P., Jaslowski, A., Metroka, C., Brett-Smith, H., and Haase, A. T. (1999) J. Virol. 73, 4181–4187[Abstract/Free Full Text]
  16. Screpanti, I., Musiani, P., Bellavia, D., Cappelletti, M., Aiello, F. B., Maroder, M., Frati, L., Modesti, A., Gulino, A., and Poli, V. (1996) J. Exp. Med. 184, 1561–1566[Abstract/Free Full Text]
  17. Smart, E. J., Ying, Y. S., Mineo, C., and Anderson, R. G. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 10104–10108[Abstract/Free Full Text]
  18. Jones, K. D., Aoki, Y., Chang, Y., Moore, P. S., Yarchoan, R., and Tosato, G. (1999) Blood 94, 2871–2879[Abstract/Free Full Text]
  19. Drexler, H. G., Meyer, C., Gaidano, G., and Carbone, A. (1999) Leukemia 13, 634–640[CrossRef][Medline] [Order article via Infotrieve]
  20. Foussat, A., Wijdenes, J., Bouchet, L., Gaidano, G., Neipel, F., Balabanian, K., Galanaud, P., Couderc, J., and Emilie, D. (1999) Eur. Cytokine Netw. 10, 501–508[Medline] [Order article via Infotrieve]
  21. Mori, Y., Nishimoto, N., Ohno, M., Inagi, R., Dhepakson, P., Amou, K., Yoshizaki, K., and Yamanishi, K. (2000) J. Med. Virol. 61, 332–335[CrossRef][Medline] [Order article via Infotrieve]
  22. Burger, R., Neipel, F., Fleckenstein, B., Savino, R., Ciliberto, G., Kalden, J. R., and Gramatzki, M. (1998) Blood 91, 1858–1863[Abstract/Free Full Text]
  23. Mullberg, J., Geib, T., Jostock, T., Hoischen, S. H., Vollmer, P., Voltz, N., Heinz, D., Galle, P. R., Klouche, M., and Rose-John, S. (2000) J. Immunol. 164, 4672–4677[Abstract/Free Full Text]
  24. Aoki, Y., Yarchoan, R., Wyvill, K., Okamoto, S., Little, R. F., and Tosato, G. (2001) Blood 97, 2173–2176[Abstract/Free Full Text]
  25. Molden, J., Chang, Y., You, Y., Moore, P. S., and Goldsmith, M. A. (1997) J. Biol. Chem. 272, 19625–19631[Abstract/Free Full Text]
  26. Wan, X., Wang, H., and Nicholas, J. (1999) J. Virol. 73, 8268–8278[Abstract/Free Full Text]
  27. Aoki, Y., Narazaki, M., Kishimoto, T., and Tosato, G. (2001) Blood 98, 3042–3049[Abstract/Free Full Text]
  28. Chow, D., He, X., Snow, A. L., Rose-John, S., and Garcia, K. C. (2001) Science 291, 2150–2155[Abstract/Free Full Text]
  29. Hoischen, S. H., Vollmer, P., Marz, P., Ozbek, S., Gotze, K. S., Peschel, C., Jostock, T., Geib, T., Mullberg, J., Mechtersheimer, S., Fischer, M., Grotzinger, J., Galle, P. R., and Rose-John, S. (2000) Eur. J. Biochem. 267, 3604–3612[Medline] [Order article via Infotrieve]
  30. Boulanger, M. J., Chow, D. C., Brevnova, E. E., and Garcia, K. C. (2003) Science 300, 2101–2104[Abstract/Free Full Text]
  31. Hibi, M., Murakami, M., Saito, M., Hirano, T., Taga, T., and Kishimoto, T. (1990) Cell 63, 1149–1157[CrossRef][Medline] [Order article via Infotrieve]
  32. Chatterjee, M., Osborne, J., Bestetti, G., Chang, Y., and Moore, P. S. (2002) Science 298, 1432–1435[Abstract/Free Full Text]
  33. Kozak, M. (1986) Cell 44, 283–292[CrossRef][Medline] [Order article via Infotrieve]
  34. Misumi, Y., Miki, K., Takatsuki, A., Tamura, G., and Ikehara, Y. (1986) J. Biol. Chem. 261, 11398–11403[Abstract/Free Full Text]
  35. Powell, L. D. (1999) Current Protocols in Molecular Biology (Chanda, V. B., ed) Vol. 3, John Wiley & Sons, Inc., New York
  36. Maley, F., Trimble, R. B., Tarentino, A. L., and Plummer, T. H., Jr. (1989) Anal. Biochem. 180, 195–204[CrossRef][Medline] [Order article via Infotrieve]
  37. Varki, A., and Kornfeld, S. (1983) J. Biol. Chem. 258, 2808–2818[Abstract/Free Full Text]
  38. Chardin, P., and McCormick, F. (1999) Cell 97, 153–155[CrossRef][Medline] [Order article via Infotrieve]
  39. Gross, V., Andus, T., Castell, J., Vom Berg, D., Heinrich, P. C., and Gerok, W. (1989) FEBS Lett. 247, 323–326[CrossRef][Medline] [Order article via Infotrieve]
  40. Spiro, R. G. (2002) Glycobiology 12, 43R-56[Abstract/Free Full Text]
  41. Parekh, R. B., Dwek, R. A., Rademacher, T. W., Opdenakker, G., and Van Damme, J. (1992) Eur. J. Biochem. 203, 135–141[Medline] [Order article via Infotrieve]
  42. Sakaguchi, M. (1997) Curr. Opin. Biotechnol. 8, 595–601[CrossRef][Medline] [Order article via Infotrieve]
  43. May, L. T., Shaw, J. E., Khanna, A. K., Zabriskie, J. B., and Sehgal, P. B. (1991) Cytokine 3, 204–211[CrossRef][Medline] [Order article via Infotrieve]
  44. Rose-John, S., Schooltink, H., Schmitz-Van de Leur, H., Mullberg, J., Heinrich, P. C., and Graeve, L. (1993) J. Biol. Chem. 268, 22084–22091[Abstract/Free Full Text]
  45. Simpson, R. J., Hammacher, A., Smith, D. K., Matthews, J. M., and Ward, L. D. (1997) Protein Sci. 6, 929–955[Medline] [Order article via Infotrieve]
  46. Moritz, R. L., Hall, N. E., Connolly, L. M., and Simpson, R. J. (2001) J. Biol. Chem. 276, 8244–8253[Abstract/Free Full Text]
  47. Hasegawa, M., Orita, T., Kojima, T., Tomonoh, K., Hirata, Y., and Ochi, N. (1992) Eur. J. Biochem. 210, 9–12[Medline] [Order article via Infotrieve]
  48. Steube, K., Gross, V., Haussinger, D., Tran-Thi, T. A., Decker, K., Gerok, W., and Heinrich, P. C. (1986) Biochem. Biophys. Res. Commun. 141, 949–955[CrossRef][Medline] [Order article via Infotrieve]
  49. Gross, V., Heinrich, P. C., vom Berg, D., Steube, K., Andus, T., Tran-Thi, T. A., Decker, K., and Gerok, W. (1988) Eur. J. Biochem. 173, 653–659[Medline] [Order article via Infotrieve]
  50. Lee, S. J., Evers, S., Roeder, D., Parlow, A. F., Risteli, J., Risteli, L., Lee, Y. C., Feizi, T., Langen, H., and Nussenzweig, M. C. (2002) Science 295, 1898–1901[Abstract/Free Full Text]
  51. Niesen, T. E., Alpers, D. H., Stahl, P. D., and Rosenblum, J. L. (1984) J. Leukocyte Biol. 36, 307–320[Abstract]
  52. Oheda, M., Hase, S., Ono, M., and Ikenaka, T. (1988) J. Biochem. (Tokyo) 103, 544–546[Abstract/Free Full Text]
  53. Sareneva, T., Mortz, E., Tolo, H., Roepstorff, P., and Julkunen, I. (1996) Eur. J. Biochem. 242, 191–200[Medline] [Order article via Infotrieve]
  54. Royle, L., Roos, A., Harvey, D. J., Wormald, M. R., van Gijlswijk-Janssen, D., Redwan, El-R. M., Wilson, I. A., Daha, M. R., Dwek, R. A., and Rudd, P. M. (2003) J. Biol. Chem. 278, 20140–20153[Abstract/Free Full Text]
  55. Cals, M. M., Guenzi, S., Carelli, S., Simmen, T., Sparvoli, A., and Sitia, R. (1996) Mol. Immunol. 33, 15–24[CrossRef][Medline] [Order article via Infotrieve]
  56. Nielsen, H., Engelbrecht, J., Brunak, S., and von Heijne, G. (1997) Protein Eng. 10, 1–6[Medline] [Order article via Infotrieve]
  57. Revest, J. M., DeMoerlooze, L., and Dickson, C. (2000) J. Biol. Chem. 275, 8083–8090[Abstract/Free Full Text]
  58. Miyakawa, K., and Imamura, T. (2003) J. Biol. Chem. 278, 35718–35724[Abstract/Free Full Text]
  59. Bellovino, D., Morimoto, T., Mengheri, E., Perozzi, G., Garaguso, I., Nobili, F., and Gaetani, S. (2001) J. Biol. Chem. 276, 13949–13956[Abstract/Free Full Text]
  60. von Heijne, G. (1985) J. Mol. Biol. 184, 99–105[CrossRef][Medline] [Order article via Infotrieve]
  61. Sakaguchi, M., Mihara, K., and Sato, R. (1987) EMBO J. 6, 2425–2431[Medline] [Order article via Infotrieve]
  62. Sakaguchi, M., Tomiyoshi, R., Kuroiwa, T., Mihara, K., and Omura, T. (1992) Proc. Natl. Acad. Sci. U. S. A. 89, 16–19[Abstract/Free Full Text]
  63. Armstrong, R. T., Kushnir, A. S., and White, J. M. (2000) J. Cell Biol. 151, 425–438[Abstract/Free Full Text]
  64. Wang, Y., and Fuller, G. M. (1995) J. Cell. Biochem. 57, 610–618[CrossRef][Medline] [Order article via Infotrieve]
  65. Heinrich, P. C., Behrmann, I., Haan, S., Hermanns, H. M., Muller-Newen, G., and Schaper, F. (2003) Biochem. J. 374, 1–20[CrossRef][Medline] [Order article via Infotrieve]
  66. Lokeshwar, V. B., Huang, S. S., and Huang, J. S. (1990) J. Biol. Chem. 265, 1665–1675[Abstract/Free Full Text]
  67. Bejcek, B. E., Li, D. Y., and Deuel, T. F. (1989) Science 245, 1496–1499[Abstract/Free Full Text]
  68. Keating, M. T., and Williams, L. T. (1988) Science 239, 914–916[Abstract/Free Full Text]
  69. Hannink, M., and Donoghue, D. J. (1988) J. Cell Biol. 107, 287–298[Abstract/Free Full Text]
  70. Dela Cruz, C. S., Lee, Y., Viswanathan, S. R., El-Guindy, A. S., Gerlach, J., Nikiforow, S., Shedd, D., Gradoville, L., and Miller, G. (2004) J. Exp. Med. 199, 503–514[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Virol.Home page
D. Chen, Y. B. Choi, G. Sandford, and J. Nicholas
Determinants of Secretion and Intracellular Localization of Human Herpesvirus 8 Interleukin-6
J. Virol., July 1, 2009; 83(13): 6874 - 6882.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
N. Adam, B. Rabe, J. Suthaus, J. Grotzinger, S. Rose-John, and J. Scheller
Unraveling Viral Interleukin-6 Binding to gp130 and Activation of STAT-Signaling Pathways Independently of the Interleukin-6 Receptor
J. Virol., May 15, 2009; 83(10): 5117 - 5126.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
D. Chen, G. Sandford, and J. Nicholas
Intracellular Signaling Mechanisms and Activities of Human Herpesvirus 8 Interleukin-6
J. Virol., January 15, 2009; 83(2): 722 - 733.
[Abstract] [Full Text] [PDF]

Home page
 J. Leukoc. Biol.Home page
P. Gasperini, S. Sakakibara, and G. Tosato
Contribution of viral and cellular cytokines to Kaposi's sarcoma-associated herpesvirus pathogenesis
J. Leukoc. Biol., October 1, 2008; 84(4): 994 - 1000.
[Abstract] [Full Text] [PDF]

Home page
 J. Gen. Virol.Home page
M. de Turenne-Tessier and T. Ooka
Post-translational modifications of Epstein Barr virus BARF1 oncogene-encoded polypeptide
J. Gen. Virol., October 1, 2007; 88(10): 2656 - 2661.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
F.-C. Ye, D. J. Blackbourn, M. Mengel, J.-P. Xie, L.-W. Qian, W. Greene, I-T. Yeh, D. Graham, and S.-J. Gao
Kaposi's Sarcoma-Associated Herpesvirus Promotes Angiogenesis by Inducing Angiopoietin-2 Expression via AP-1 and Ets1
J. Virol., April 15, 2007; 81(8): 3980 - 3991.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
D. Chen and J. Nicholas
Structural Requirements for gp80 Independence of Human Herpesvirus 8 Interleukin-6 (vIL-6) and Evidence for gp80 Stabilization of gp130 Signaling Complexes Induced by vIL-6
J. Virol., October 1, 2006; 80(19): 9811 - 9821.
[Abstract] [Full Text] [PDF]

Home page
 J. Virol.Home page
M. Kovaleva, I. Bussmeyer, B. Rabe, J. Grotzinger, E. Sudarman, J. Eichler, U. Conrad, S. Rose-John, and J. Scheller
Abrogation of Viral Interleukin-6 (vIL-6)-Induced Signaling by Intracellular Retention and Neutralization of vIL-6 with an Anti-vIL-6 Single-Chain Antibody Selected by Phage Display.
J. Virol., September 1, 2006; 80(17): 8510 - 8520.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/50/51793    most recent
M407382200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Meads, M. B.
Right arrow Articles by Medveczky, P. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Meads, M. B.
Right arrow Articles by Medveczky, P. G.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement