Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chen, D. Articles by Nicholas, J. Search for Related Content PubMed PubMed Citation Articles by Chen, D. Articles by Nicholas, J.

 Previous Article  |  Next Article 

Journal of Virology, January 2009, p. 722-733, Vol. 83, No. 2
0022-538X/09/$08.00+0     doi:10.1128/JVI.01517-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Intracellular Signaling Mechanisms and Activities of Human Herpesvirus 8 Interleukin-6

Daming Chen, Gordon Sandford, and John Nicholas^*

Sidney Kimmel Comprehensive Cancer Center, Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287

Received 18 July 2008/ Accepted 27 October 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Human herpesvirus 8 (HHV-8)-encoded viral interleukin-6 (vIL-6) has been implicated as a key factor in virus-associated neoplasia because of its proproliferative and survival effects and also in view of its angiogenic properties. A major difference between vIL-6 and human IL-6 (hIL-6) is that vIL-6, uniquely, is largely retained and can signal intracellularly. While vIL-6 is generally considered to be a lytic gene, several reports have noted its low-level expression in latently infected primary effusion lymphoma (PEL) cultures, in the absence of other lytic gene expression. Thus, intracellular autocrine signal transduction by the viral cytokine may be of particular relevance to the growth and survival of latently infected cells and to pathogenesis. Here we report that most intracellular vIL-6 is located in the endoplasmic reticulum (ER), signals via the gp130 signal transducer in this compartment, and does so independently of the gp80 -subunit of the IL-6 receptor, required for hIL-6 signal transduction. Signaling and biological assays incorporating ER-retained vIL-6 and hIL-6 confirmed vIL-6 activity, specifically, in this compartment. Knockdown of vIL-6 expression in PEL cells led to markedly reduced cell growth in normal culture, independently of extracellular cytokines. This could be reversed by reintroduction via virus vector of exclusively ER-retained vIL-6. These data indicate that in virus biology vIL-6 may act to support the growth and survival of cells latently infected with HHV-8 in an autocrine manner via intracrine signaling and that these activities may contribute to the maintenance of latently infected cells and to virus-induced neoplasia.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The cancers with which human herpesvirus 8 (HHV-8) is associated, Kaposi's sarcoma (KS), multicentric Castleman's disease, and primary effusion lymphoma (PEL), are all diseases in which interleukin-6 (IL-6) signaling is believed to play an important role, either as a driver of cell growth and survival or as an inducer of angiogenesis (1, 11, 15, 20, 29). The potential of viral IL-6 (vIL-6), in particular, to contribute to HHV-8 malignant pathogenesis is highlighted by reports of the expression of vIL-6 in KS, PEL, and multicentric Castleman's disease tissues, coupled with demonstrations of vIL-6 support of growth of PEL and other cell types in culture (3, 5, 15, 24, 26). Furthermore, vIL-6 can induce the expression of human IL-6 (hIL-6) in some cells, thereby potentially amplifying pathogenically relevant autocrine and paracrine signaling at sites of HHV-8 infection (22). The viral cytokine, unlike other proteins expressed as part of the productive, lytic cycle of virus replication, may also be expressed at low levels during latency, providing a means by which it could potentially influence latently infected cells in a strictly autocrine fashion via intracellular signaling (3, 9, 23). That such "intracrine" signal transduction can occur is apparent from the demonstration by Meads and Medveckzy (19) that endoplasmic reticulum (ER)-retained, KDEL-tagged vIL-6 is able to transduce signal.

The mechanism by which vIL-6 mediates signal transduction is distinct from that used by hIL-6 and all other cellular IL-6 proteins in that the viral cytokine requires only the gp130 signal-transducing subunit of the IL-6 receptor for stable and functional complex formation; the nonsignaling gp80 -subunit, to which cellular IL-6 proteins first bind, is not needed in the case of vIL-6 (21). The molecular basis of gp80 independence has not been established, but it appears to be conferred by the particular three-dimensional conformation adopted by vIL-6 rather than the specific amino acid composition of the receptor-binding interfaces (8, 10). However, vIL-6 is able to incorporate gp80 into signaling complexes, and indeed, experimental evidence suggests that gp80 stabilizes vIL-6-mediated dimerization of gp130 and enhances signaling (2, 17, 18, 28). Furthermore, STAT activation and biological activities mediated by vIL-6-induced hexameric complexes (gp130₂:gp80₂:vIL-6₂) can be distinguished from effects induced via tetrameric complexes lacking gp80 (14). Thus, STAT signaling amplitude and duration are greatly increased and prolonged in the presence of gp80, and support of BAF-130 cell growth enhanced.

This report is focused on the operation and biological relevance of intracellular signaling by vIL-6. We show that most intracellular vIL-6 localizes naturally to the ER and is functional in this compartment, signaling exclusively via tetrameric (gp80 deficient) signaling complexes, and that this intracrine signaling allows autocrine promotion of PEL cell proliferation and survival. The data presented suggest an important role of vIL-6 during viral latency, implicate autocrine vIL-6 signaling as a contributor to viral pathogenesis, and provide further insights into the subtleties of vIL-6 signal transduction.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell culture, transfections, and lentiviral transduction. BCBL-1 and JSC-1 PEL cells were maintained in RPMI 1640 medium supplemented with 10% fetal calf serum. HEK293T cells were grown in Dulbecco's modified Eagle's medium with 10% fetal calf serum. For transfection of HEK293T cells, cultures were passaged 12 to 24 h prior to transfection to produce 60%-confluent monolayers, to which were added calcium phosphate-DNA coprecipitates that were generated by mixing DNA, CaCl₂, and HEPES-buffered saline (using standard procedures). For infection of PEL cells with lentiviral vectors (see below), 1 x 10⁶ cells were mixed with the concentrated lentivirus in the presence of 5 µg/ml polybrene and incubated at 37°C. After incubation for 5 h, the medium containing virus and polybrene was replaced with fresh medium. The cells were incubated for 2 days before being used in experiments to assess growth rates and apoptosis. For these experiments 1 x 10⁵ cells in 1 ml medium (>90% or approximately 50% green fluorescent protein [GFP]-positive, virus vector-transduced cells) were seeded in each well of a 12-well plate. For growth curves, viable cells were identified by trypan blue exclusion and cell densities determined by using a hemocytometer. GFP-positive cells were identified by using an inverted UV microscope, with total cell numbers being determined under white light; data were collected from several random fields per sample to determine the percent GFP positivity. Cell cycle analysis of cultures was undertaken by flow cytometry following Hoechst 33258 staining of DNA, using appropriate gating based on DNA content to determine the proportions of cells in G₀, S, and G₂/M. For measurements of apoptosis, cells were washed once with phosphate-buffered saline (PBS) containing 0.1% sucrose and resuspended in 1 ml of binding buffer containing 10 mM HEPES (pH 7.4), 140 mM NaCl, and 2.5 mM CaCl₂. Cells were incubated at room temperature for 5 min in the dark after the addition of 1 µl of annexin V-Cy3 stock solution (catalog no. 1002-200; BioVision, Mountain View, CA). After being washed with binding buffer, cells were placed on a glass slide and visualized by UV microscopy. For experiments involving dual infection with vIL-6 and short hairpin RNA (shRNA) lentiviral vectors, cells were first infected at high efficiency (>90% GFP fluorescence) with vIL-6 vectors or empty lentivirus control and subsequently infected, after 2 days, with shRNA-containing lentiviruses. Efficiencies of secondary infection (>90%) were established by infecting parallel naïve PEL cultures and determining the proportion of GFP-positive (GFP⁺) cells within the population.

Plasmids. pSG5-based eukaryotic expression vectors for vIL-6-chitin-binding domain (CBD) fusion proteins have been described previously (8). Expression vectors for KDEL motif-tagged vIL-6-CBD fusion proteins were generated by cloning PCR-amplified vIL-6-CBD coding sequences into a cloning site-modified pSG5 vector (pSG5.S) (8) into which a double-stranded oligonucleotide (ds-oligonucleotide) encoding the KDEL motif, along with additional sequences to maximize ER retention (MEEDDDQKAVKDEL) (25), had been cloned. BamHI and BglII sites were used for ds-oligonucleotide cloning into pSG5.S, and SmaI and BamHI sites for the subsequent introduction of vIL-6-CBD sequences, in-frame with the KDEL-containing sequence. The viral vectors utilized in this study comprised modified versions of pTYB6 (6), which contains the H1 promoter for shRNA expression (pYNC352), or the simian virus 40 promoter for protein expression (pYNC352-SV); both vectors contain cytomegalovirus-driven GFP coding sequences for identification of transduced cells. The complementary oligonucleotides for vIL-6 knockdown corresponded to GGCTTCAGGGACTTAAGTA; annealed products were cloned between the BamHI and MluI sites of pYNC352. Nonspecific, nonsilencing (NS) ds-oligonucleotide (GCGCGCTTTGTAGGATTCG) was cloned into the lentiviral vector to provide a negative control. The coding sequences for vIL-6-CBD and vIL-6-CBD+KDEL fusions were cloned between the BamHI and MluI sites of pYN352-SV. The shRNA target sequences in these vIL-6 open reading frames were mutated to GATTGCAAGGATTAAAATAT, to generate shRNA-resistant mRNAs without amino acid sequence changes.

Lentivirus production. Lentiviral vectors pYNC352-NS and pYNC352-477 (8 µg) were cotransfected with gag-pol and vesicular stomatitis virus G protein plasmids (32 µg and 8 µg, respectively) into 60%-confluent monolayers of HEK293T cells in 75-cm² culture flasks. Two days posttransfection, the viruses were collected from the medium by centrifugation in an SW28 rotor at 24,000 rpm for 1.5 h at 4°C. Virus pellets were resuspended in 1 ml RPMI 1640 supplemented with 10% fetal bovine serum and 5 µg/ml polybrene. The infectious titers were established by using GFP-based assays in infected PEL cultures prior to the use of lentiviral vectors in experiments employing PEL cells.

Subcellular fractionation. PBS-washed cells were suspended in ice-cold homogenization buffer (0.25 M sucrose, 1 mM EDTA, 10 mM HEPES, pH 7.4) and disrupted by repeated passage through a 27-gauge syringe needle. Homogenates were centrifuged at 1,000 x g for 10 min at 4°C to remove nuclei, and supernatants subjected to density gradient ultracentrifugation for membrane fractionation. Discontinuous iodixanol gradients were prepared by using OptiPrep (Accurate Chemical and Scientific; Westbury, NY) layered successively at 30, 25, 20, 17.5, 15, 12.5, 10, 7.5, 5, and 2.5%. Postnuclear supernatants were applied, and centrifugation carried out at 25,000 rpm in an SW41 rotor for 3 h at 4°C. The gradient was removed in 10 to 12 equal fractions from the top of the gradient. For the isolation of rough ER membranes, PBS-washed cell pellets were resuspended in 3 packed cell volumes of ice-cold hypotonic buffer (10 mM HEPES, pH 7.8, 25 mM KCl, 1 mM EGTA) and cells incubated at 4°C for 20 min. Cell pellets, obtained by low-speed centrifugation, were resuspended in 2 packed cell volumes cold isotonic buffer, and cells lysed by Dounce homogenization, followed by 12 to 15 passages through a 27-gauge needle. Homogenates were centrifuged (at 4°C) at 1,000 x g for 10 min, and supernatants subjected to 12,000 x g centrifugation for 15 min to obtain postmitochondrial fractions. To these were added, dropwise, 7.5 volumes of 8 mM CaCl₂, to a final concentration of 7 mM. Mixtures were centrifuged at 8,000 x g for 10 min to obtain ER-enriched microsomal fractions (pellets) which were resuspended in lysis buffer (20 mM Tris-HCl, pH 8.0, 5 mM EDTA, 50 mM NaCl, 0.2% NP-40, and proteinase inhibitors) for immunoprecipitation, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and Western blot analysis.

Western blotting and coprecipitations. Immunoblotting was carried out to detect organelle membrane markers (p58K [Golgi compartment], Na⁺/K⁺-ATPase [plasma membrane], calnexin [ER], EEA1 [early endosome]), gp130-Fc-coprecipitated CBD-tagged ligands (vIL-6 and hIL-6), IL-6 receptor subunits (gp80 and gp130), and phosphotyrosine residues present on protein A-agarose-precipitated gp130-Fc. The antiserum to vIL-6, generated in this laboratory, has been described previously (28). Commercially obtained antibodies used for immunoblotting were as follows: hIL-6, R&D Systems (Minneapolis, MN), catalog no. AB-206-NA; CBD, New England Biolabs (Beverly, MA), catalog no. E8034S; gp80, Santa Cruz Biotechnology (Santa Cruz, CA), catalog no. sc-661; gp130, BD Biosciences (Rockville, MD), catalog no. 555755; phosphotyrosine, Cell Signaling Technology (Beverly, MA), catalog no. 9411; p58K, Sigma (St. Louis, MO), catalog no. G2404; Na⁺/K⁺-ATPase, Abcam (Cambridge, MA), catalog no. ab7671-50; and EEA1, BD Biosciences (Rockville, MD), catalog no. 610457. Tricine-SDS-PAGE and electrophoretic transfer to nitrocellulose membranes was carried out by using standard procedures. Membranes were blocked in PBS-T buffer (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na₂HPO₄, 1.4 mM KH₂PO₄, 0.1% Tween 20) containing 5% nonfat milk prior to the addition of primary antibody (0.1 to 1.0 µg/ml) and incubation at 4°C overnight. PBS-T-washed membranes were incubated with horseradish peroxidase-conjugated anti-rabbit (vIL-6), anti-goat (gp80), or anti-mouse secondary antibodies in PBS-T-5% nonfat milk and washed with PBS-T, and membrane-bound antibody visualized by using chemiluminescence assay. For coprecipitation-based vIL-6-receptor complexing assays, gp130-Fc was precipitated with protein A-agarose after overnight incubation with cell lysates or ER membrane fractions, sedimented material washed with PBS, and pellets heated in SDS-PAGE load buffer prior to gel loading.

Fluorescence microscopy. For studies of the subcellular localization of vIL-6, 293T cells were seeded onto poly-L-lysine-coated chamber slides and transfected with expression plasmids for vIL-6-red fluorescent protein (RFP) and/or ER-directed KDEL motif-tagged GFP. At 24 h after transfection, cells were fixed with 4% paraformaldehyde in PBS and permeabilized with 0.25% Triton X-100 for immunofluorescence assays. Cells were blocked in PBS containing 1% bovine serum albumin and subsequently incubated sequentially at room temperature with primary and secondary antibodies (1 h each) and were rinsed in between incubations and finally with PBS. Mouse monoclonal antibodies to Golgi compartment 58K protein and early endosome EEA1 were used as primary antibodies to identify the respective compartments and their staining patterns relative to vIL-6-RFP localization. Fluorescein isothiocyanate-conjugated anti-mouse immunoglobulin G (1:100) (catalog no. 816511; Zymed laboratories, San Francisco, CA) was used as the secondary antibody. Cells were counterstained with 4',6'-diamidino-2-phenylindole (DAPI) and viewed by confocal microscopy. For assessment of vIL-6 expression in PEL cells, JSC-1 and BCBL-1 suspension cells or Akata cells (negative control) were centrifuged onto microscope slides and fixed and permeabilized in chilled methanol prior to antibody probing. Rabbit antiserum specific for vIL-6 (28) and/or rat monoclonal antibody to HHV-8 latency-associated nuclear antigen (LANA) (catalog no. 13-210-100; Advanced Biotechnologies, Inc., MD) were incubated with slides before rinsing and addition of Cy3-conjuated goat anti-rabbit and/or fluorescein isothiocyanate-conjugated goat anti-rat secondary antibodies (catalog nos. 81-6115 and 81-9511; Invitrogen, Carlsbad, CA) for detection of the respective antigens.

Pulse-chase. 293T cells were seeded in six-cm culture dishes and transfected with pSG5-vIL-6-CBD, pSG5-hIL-6-CBD, or pSG5. After 24 h, the cells were rinsed twice with PBS and then incubated in methionine-free/cysteine-free medium for 30 min at 37°C and pulse labeled with 200 µCi/ml [³⁵S]methionine/cysteine (catalog no. SJG0079; GE Healthcare, Piscataway, NJ) for 20 min. The cells were rinsed twice with warm PBS and chased with 5 ml warm culture medium containing 2 mM methionine and 2 mM cysteine. Medium and cells were collected at various times during the chase. vIL-6-CBD and hIL-6-CBD in medium or cells were captured by chitin bead precipitation prior to PAGE and autoradiography of dried gels.

Endo H digestion. Recombinant endo-β-N-acetylglucosaminidase H (endo H-maltose binding protein fusion protein [endo H_f]) was obtained from New England Biolabs (catalog no. P0703S; Ipswich, MA). For endo H analysis of high-mannose glycosylation of vIL-6, vIL-6-CBD was precipitated from transfected cell lysates with chitin beads, resuspended in 1x glycoprotein denaturation buffer (0.5% SDS, 40 mM dithiothreitol), and incubated at 100°C for 10 min prior to the addition of 0.1 volume of 10x reaction buffer (50 mM sodium citrate, pH 5.5) and 1,000 units of endo H_f and incubation at 37°C for 1 h. Pre- and posttreatment products were compared by SDS-PAGE/Western blot analysis.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Secretion of vIL-6 versus hIL-6. It has previously been reported that vIL-6 is secreted much less efficiently than hIL-6 in transfected HEK293 cells and BCBL-1 PEL cells latently infected with HHV-8 and that gp130 coexpression is required for vIL-6 secretion in transfected Ba/F3 cells (19). To confirm and compare the relative secretion versus retention of vIL-6 and hIL-6 in HEK293T cells, used in the present study, and to verify comparable properties of CBD-tagged versions of the cytokines, also utilized experimentally, we undertook Western blot analyses of lysates and growth media of transfected cultures expressing the respective native and CBD-fused cytokines. The results (Fig. 1A) showed that while hIL-6 could be detected readily in the medium of hIL-6- and hIL-6-CBD-expressing cells, hIL-6 levels in the lysates were far lower (indeed, difficult to detect); the converse was true for native and CBD-tagged vIL-6 proteins.

View larger version (31K): [in this window] [in a new window]

FIG. 1. Secretion of hIL-6 and vIL-6 in transfected HEK293T cells. (A) Comparisons of hIL-6 and vIL-6 secretion in cells transfected with expression vectors for the native and CBD-fused versions of the human and viral cytokines. Total cell lysates (L) and media (M) were analyzed by Western blotting using antibodies to vIL-6 and hIL-6 (left panel) or to CBD (right panel). CBD antibody detection of vIL-6 and hIL-6 fusion proteins allowed direct comparison of their relative levels; the CBD fusion had no apparent effect on secretion of the two cytokines. CBD-tagged IL-6 proteins were precipitated from culture media by using chitin beads prior to SDS-PAGE and Western blotting; for untagged vIL-6 and hIL-6, 40 µl of medium (4% of total) was analyzed relative to 5% of the corresponding total cell lysate. , anti; long ex., long exposure. (B) Pulse-chase analysis of CBD-tagged vIL-6 and hIL-6 secretion in HEK293T cells. Proteins were pulse-labeled for 20 min with [35S]methionine/cysteine before being rinsed and chased with fresh medium and monitored for intracellular and secreted IL-6 over an 8-h period by CBD immunoblotting of cell lysates and chitin bead-precipitated secreted cytokines from growth media. +, present; –, absent.

The secretion kinetics of vIL-6 and hIL-6 (CBD-tagged) in HEK293T cells were compared using pulse-chase analysis. Proteins were labeled for 20 min prior to washing and chase, and cell lysates and culture media analyzed for the presence of vIL-6 using Western blotting. Cell lysates were either treated or not treated with endo H to allow the identification of immature high-mannose-linked (endo H sensitive and likely ER associated) cytokines. Notably different secretion rates of hIL-6 (approximately 30-min intracellular half-life) and vIL-6 (>8-h half-life of retention) were detected (Fig. 1B), in general agreement with data reported previously for secretion rates of vIL-6 from PEL (BCBL-1) and HEK293 cells (19). Of note is that all of the vIL-6 protein detected in the cell lysates was endo H sensitive, indicative of ER association.

vIL-6 localizes to the ER. Independent investigations of the subcellular distribution of vIL-6 were undertaken by Western blot analysis of membrane fractions and by confocal microscopy. For the first approach, membranes were fractionated on iodixanol gradients and the respective endosomal, Golgi membrane, and ER components identified by immunoblotting for detection of the specific markers EEA1, p58K, and calnexin (see Materials and Methods). vIL-6 (CBD tagged) was found to cofractionate with calnexin, suggesting an ER association of vIL-6 (Fig. 2A). Confocal microscopy was used to provide additional evidence of the ER localization of vIL-6 and to exclude the possibility of significant vIL-6 localization to other compartments. A vIL-6-RFP fusion protein was coexpressed with KDEL-tagged GFP in transfected cells. The results (Fig. 2B) showed that there was very high coincidence of vIL-6 and GFP localization; in contrast, there was a general lack of colocalization of vIL-6 with either Golgi compartment or early endosomal markers (p58K and EEA1, respectively), used as controls. The data demonstrate a predominant ER localization of vIL-6 within the cell.

View larger version (65K): [in this window] [in a new window]

FIG. 2. Colocalization of vIL-6 with ER membranes and protein marker. (A) Iodixanol gradient-based membrane fractionation experiment for analysis of vIL-6 subcellular distribution. vIL-6-transfected HEK293T cell extract was loaded onto the gradient and subjected to ultracentrifugation before SDS-PAGE and immunoblot analysis of gradient-derived fractions (left to right, top to bottom of gradient) for detection of vIL-6 and membrane markers (see Materials and Methods). vIL-6 cofractionated with ER membrane marker calnexin and had a distribution pattern discordant with the distribution patterns of fractions containing Golgi compartment and early endosome markers p58K and EEA1. fractn, fraction. (B) Confocal fluorescence and immunofluorescence microscopy analysis of vIL-6 intracellular localization. To assess ER localization of vIL-6, expression vectors for vIL-6-RFP and ER-targeted, KDEL-motif-tagged GFP (GFP-KDEL) were cotransfected into HEK293T cells; a strong correspondence of RFP and GFP signals was detected (top panels). Cells transfected with vIL-6-RFP were immunostained for detection of p58K (Golgi compartment) and EEA1 (early endosome) markers to provide controls (middle and bottom panels). DAPI was used for fluorescence visualization of nuclei.

vIL-6 signals in ER. While previously published data had shown signaling by ER-targeted, KDEL-tagged vIL-6 (19), we wished to demonstrate directly that signal transduction was initiated by untargeted vIL-6 within ER membranes. Rough ER fractions, free of other membranous components, e.g., plasma membrane and endosomes, were obtained by differential centrifugation (see Materials and Methods) (Fig. 3A). These fractions were derived from HEK293T cells transfected with pSG5 (empty vector), vIL-6, or hIL-6 expression plasmids together with an expression vector for immunoglobulin G-Fc-tagged gp130 signal transducer (gp130-Fc), with or without a gp80 -subunit expression plasmid. gp80 is required for hIL-6 signaling but not for vIL-6 activity, although vIL-6 can form signaling complexes that include gp80 (17, 18). gp130-Fc was precipitated with protein A-agarose from either total cell lysates or ER fractions, and Western blotting undertaken for detection of phosphotyrosine residues, indicative of gp130 activation. The results of these experiments (Fig. 3B) demonstrated that vIL-6, uniquely, was able to induce gp130 phosphorylation, the initial event in signal transduction, within the ER in both gp130-Fc- gp130-Fc-plus-gp80-overexpressing cells, whereas hIL-6 was unable to support detectable ER-initiated signaling (in the presence of overexpressed gp80). Interestingly, analysis of vIL-6-induced ER signaling complexes revealed that gp80 was not coprecipitated with gp130-Fc, as it was from total cell lysate in both vIL-6- and hIL-6-expressing cells (Fig. 3B). Taken together, these data demonstrate that vIL-6 induces signal transduction in the ER and that this is mediated exclusively via tetrameric signaling complexes (vIL-6₂:gp130₂).

View larger version (39K): [in this window] [in a new window]

FIG. 3. vIL-6-induced signaling and signaling complexes in the ER. (A) Immunoblot quality control of rough ER-enriched microsomal fractions derived by differential centrifugation of transfected HEK293T cell extracts (see Materials and Methods). The rough ER preparations contained ER-resident calnexin but little or no Golgi compartment, early endosome, and plasma membrane markers (p58K, EEA1, and Na+/K+-ATPase, respectively) detected in complete cell lysates. (B) Coprecipitation-based analysis of IL-6-induced gp130 phosphorylation and gp130:vIL-6 association. Extracts of HEK293T cells transfected with expression vectors for gp130-Fc or gp130-Fc plus gp80 and vIL-6 or hIL-6 (or empty pSG5 vector) were subjected to protein A-agarose precipitation, and bound proteins were fractionated by SDS-PAGE and analyzed by Western blotting for detection of tyrosine-phosphorylated gp130 (p-gp130-Fc) and gp130-bound IL-6 (IL-6-CBD). Phosphorylation (activation) of gp130 and gp130 association of IL-6 in ER fractions was detected only in vIL-6-expressing cells. However, no gp80 association with gp130 was detected in ER fractions derived from cells coexpressing the two transfected receptors. In contrast, total cell lysates from vIL-6- and hIL-6-expressing cells contained phosphorylated gp130 and gp130-associated gp80 (indicative of hexameric signaling complex formation). The data indicate that only tetrameric signaling complexes are induced by vIL-6 in the ER compartment and that only vIL-6 and not hIL-6 can activate gp130 in ER. (C) Testing of KDEL-tagged hIL-6 and vIL-6 (CBD fusions) for intracellular retention in transfected HEK293T cells. Cell lysates and media were analyzed for IL-6 expression by immunoblotting using CBD-specific antibody. (Note: the lysate and media blots are not directly comparable, due to different proportions of the two loaded onto the gel and the different film exposure times used to visualize the bands.) (D) Analysis of IL-6-KDEL-induced gp130 dimerization and gp130:gp80 complexing by coprecipitation assay. Experiments analogous to those outlined in the panel B legend were undertaken but with a substitution of ER-targeted hIL-6-KDEL (and vIL-6-KDEL) and detection of gp130 dimerization, indicative of signaling complex formation; gp130-CBD expression vector was cotransfected with other receptor and ligand plasmids to allow detection of gp130 dimerization. Only vIL-6 could induce gp130 complexing in the ER. RER, rough ER; v, vIL-6; h, hIL-6; –, empty vector; vK, vIL-6-CBD+KDEL; hK, hIL-6-CBD+KDEL; IP, immunoprecipitation; IB, immunoblotting.

To exclude the possibility that the lack of signaling complex formation in the ER by hIL-6 was due merely to limiting amounts of hIL-6 in this compartment, we utilized an ER retention motif (KDEL)-tagged hIL-6-CBD fusion protein in an equivalent experiment. Examination of transfected cell lysates and culture media expressing KDEL-tagged and untagged hIL-6 confirmed increased intracellular accumulation of ER-targeted cytokine (Fig. 3C). Coprecipitation assays were then undertaken to identify IL-6-induced complexes in ER fractions, as before, using KDEL-tagged hIL-6 and vIL-6. In these experiments we used cotransfected gp130-Fc and gp130-CBD expression plasmids in the presence and absence of the gp80 expression vector to allow subsequent detection of gp130 dimerization and gp80:gp130 complexing in response to the cytokines. Coprecipitation of gp130-CBD with gp130-Fc by protein A-agarose from ER preparations derived from cells expressing vIL-6, but not from those transfected with pSG5, confirmed the induction of gp130 dimerization by vIL-6 in the ER compartment (Fig. 3D). As before, gp80, when coexpressed, could not be coprecipitated with gp130-Fc as a function of vIL-6 expression, indicative of vIL-6 induction of tetrameric signaling complexes exclusively. hIL-6-KDEL induced only minimal gp130:gp80 complexing, evidenced by a very faint coprecipitated gp80 band but no detectable gp130 dimerization within the ER. Thus, vIL-6 is uniquely able to induce gp130:gp130 complexing and signal transduction in the ER, presumably in part because it does not require gp80 for gp130 dimerization, in contrast to hIL-6. Because vIL-6-induced tetrameric and hexameric complexes have distinct signaling properties (14), this finding also indicates that there are qualitative differences in vIL-6 signal transduction in the ER and at the cell surface.

ER versus non-ER distribution of vIL-6:gp130 complexes. Having established the intracellular localization of vIL-6 as predominantly within the ER and the ability of vIL-6 to signal via tetrameric complexes in this compartment, we next wanted to determine the proportion of vIL-6-induced signaling complexes located in the ER. To this end we examined the relative levels of endo H-sensitive (ER localized) versus endo H-resistant (exclusively Golgi compartment/post-Golgi compartment localized) vIL-6 coprecipitated with gp130. Experiments were undertaken using cotransfected gp130-Fc to facilitate precipitation of the signal transducer from cell lysates. All detectable gp130-associated vIL-6 was found to be endo H sensitive, in contrast to gp130-associated hIL-6 (in cells cotransfected with the gp80 expression plasmid), which was completely resistant to endo H (Fig. 4A). These data indicated that the vast majority of vIL-6:gp130 complexes were present in the ER. However, the distribution of g130-Fc as determined by density gradient centrifugation was more equally localized between ER and non-ER (likely plasma membrane) compartments in both the absence and presence of vIL-6 (Fig. 4B). Thus, the majority of vIL-6:gp130 interactions occur in the ER, reflecting the distribution of vIL-6 (in contrast to gp130) within the cell.

View larger version (21K): [in this window] [in a new window]

FIG. 4. Analysis of ER localization of gp130-associated vIL-6 by endo H sensitivity. (A) vIL-6 and hIL-6 proteins (CBD tagged) were expressed in HEK293T cells together with gp130-Fc or gp130-Fc plus gp80, respectively. Gp130-Fc was precipitated from cell lysates with protein A-agarose, and coprecipitated material was analyzed for the presence of endo H-resistant (post-ER) and endo H-sensitive (ER associated) vIL-6 and hIL-6 by SDS-PAGE and Western blotting using CBD antibody. All detectable gp130-precipitated vIL-6 was endo H sensitive, indicating predominant ER signaling by the viral cytokine. All hIL-6 precipitated with gp130 was endo H resistant. +, present; –, absent; IB, immunoblotting; IP, immunoprecipitation. (B) Iodixanol gradient cell membrane fractionation and immunoblotting to assess the subcellular distribution of gp130 in the presence and absence of vIL-6. The data reveal distinct distribution profiles of the signal transducer and ligand and the lack of influence of vIL-6 on the gross distribution of gp130.

Endogenously expressed vIL-6 in PEL cells localizes to and signals in ER. To confirm the ER localization of endogenously expressed vIL-6 in PEL cells, we undertook density gradient membrane fractionation and Western blot analysis of BCBL-1 cell extracts. As seen with transfected HEK293T cells (Fig. 2A and 4B), vIL-6 cofractionated with ER-resident calnexin, with a distribution pattern distinct from that of the plasma membrane marker Na⁺/K⁺-ATPase present in "lighter" fractions closer to the top of the iodixanol gradient (Fig. 5A). Independent evidence that vIL-6 was present in the ER was provided by Western blot analysis of ER membrane-enriched microsomal preparations derived from BCBL-1 cell lysates, obtained by differential ultracentrifugation (Fig. 5B). Consistent with the results of a similar analysis of such preparations from transfected HEK293T cell extracts (Fig. 3B), vIL-6 was detected in the rough ER fraction at a level equivalent to that detected in whole-cell lysates (note the similar amounts of calnexin, used as a control, in both preparations).

View larger version (46K): [in this window] [in a new window]

FIG. 5. ER localization and activity of endogenously produced vIL-6 in PEL cells. (A) Density gradient (iodixanol) fractionation of membranes derived from BCBL-1 cells, showing cofractionation of vIL-6 with calnexin-containing ER and distinct (but overlapping) distribution relative to plasma membrane marker (Na+/K+-ATPase). (B) Western blot analysis of rough ER preparations derived from BCBL-1 extracts by differential centrifugation (see Materials and Methods). The ER preparations, showing little or no contamination with markers of early endosomes (EEA1), Golgi compartment (p58K), or plasma membranes (Na+/K+-ATPase), contained a high proportion of the total cellular vIL-6 (compare lysate and rough ER samples, loaded proportionately). (C) Western blot and qRT-PCR analyses of the efficacy of shRNA-mediated vIL-6 depletion in lentivirus vector-transduced BCBL-1 cells. Cells were infected at an efficiency of 80% with lentivirus (GFP+), and this led to significant depletion of vIL-6 mRNA and protein at 48 h postinfection relative to the levels in NS shRNA-transduced cells (infected at equivalent efficiency). For qRT-PCR data, error bars represent differences between actin-normalized values obtained from duplicate reactions for each sample. (D) Analysis of vIL-6-dependent tyrosine phosphorylation (activation) of ER-localized gp130 in BCBL-1 cells by phosphotyrosine immunoblot of gp130 immunoprecipitated (IP) from ER preparations (as described for panel B) in the absence (NS shRNA transduced) and presence of shRNA-mediated vIL-6 depletion. The data show an inverse correlation between vIL-6 depletion and gp130 activation in the ER. RER, rough ER; , anti; PY, phosphotyrosine.

For analysis of vIL-6 intracellular signaling and biological activity in PEL cells, we generated lentivirus vectors (GFP⁺) containing vIL-6-directed shRNA or NS shRNA to allow the generation of PEL cultures specifically depleted of vIL-6 or providing an appropriate control. The efficacy of vIL-6 mRNA and protein depletion by the selected vIL-6 shRNA was tested initially in transfected HEK293T cells (data not shown) and subsequently confirmed by quantitative reverse transcription-PCR (qRT-PCR) and Western blot analysis of extracts of lentivirus-transduced BCBL-1 PEL cultures (Fig. 5C). For analysis of ER signaling by vIL-6, BCBL-1 cultures were infected with each virus vector using titers sufficient to achieve >90% transduction efficiency, as determined by visualizing GFP fluorescence 48 h after infection, and ER fractions were derived for analysis of gp130 activation. Western blotting was undertaken to identify tyrosine phosphorylation of immunoprecipitated gp130 in the control and vIL-6-depleted cells. The results of these experiments revealed that active, tyrosine-phosphorylated gp130 was present in BCBL-1 ER membranes and that gp130 activation was reduced in the vIL-6 shRNA-transduced cells (Fig. 5D). These data indicate that endogenously expressed vIL-6 can signal in the ER of PEL cells.

Autocrine vIL-6 signaling supports PEL cell growth. PEL cells have been reported previously to be responsive to vIL-6 with respect to growth in culture, but the viral cytokine was not necessary for cell growth (15). In light of our results indicating predominant intracellular vIL-6 signaling in HEK293T cells, coupled with the known expression of vIL-6 in latently infected PEL cultures (7, 9, 23) and our detection of the viral cytokine and vIL-6 signaling in PEL ER (Fig. 5), we wanted to test the relevance of endogenously produced vIL-6 and its autocrine signaling to the growth of PEL cells. Initially, we monitored the growth of JSC-1 PEL cells (infected with HHV-8 and Epstein-Barr virus [EBV]), transduced at high efficiency (>90%) with lentivirus-delivered shRNA, relative to the growth of cells transduced with the control NS shRNA. The EBV-infected B-cell line Akata was used as a control. We found that vIL-6 shRNA, specifically, induced markedly slowed JSC-1 cell growth relative to the rates of proliferation of NS shRNA-transduced cells (Fig. 6A). Akata cells were unaffected by vIL-6 shRNA transduction; vIL-6- and control (NS) shRNA-containing cells grew with similar kinetics.

View larger version (18K): [in this window] [in a new window]

FIG. 6. Biological activity of vIL-6 in latently infected PEL cells. (A) Effect of vIL-6 depletion on growth of JSC-1 cells. Cells were transduced with vIL-6-directed or NS shRNAs by infection of separate cultures with the respective (GFP+) lentivirus vectors, used at titers sufficient to give >90% transduction (as evidenced by GFP fluorescence). Viable, trypan blue-excluding cells from triplicate cultures were counted daily starting two days after lentiviral infection. The EBV-infected B-cell line Akata was used as a control in the same experiment. vIL-6 depletion in JSC-1 cells led to markedly reduced rates of cell growth. (B) Determination of the role of autocrine signal transduction by vIL-6 on PEL cell growth. BCBL-1, JSC-1, and Akata cells were infected with vIL-6 or NS shRNA lentivirus vectors to achieve a transduction efficiency of 50%, as determined by GFP fluorescence. Cultures were then monitored, after 48 h, for their growth rates (left panels). The proportions of GFP+ (shRNA transduced) cells in the cultures were also determined (right panels). The data show PEL-specific, vIL-6 depletion-dependent decreases in the rates of growth predominantly of the GFP+ (vIL-6 shRNA transduced) populations, demonstrating strictly autocrine effects of vIL-6 on PEL cell growth. For all experiments, data were collected from triplicate cultures and from multiple samplings from each at every time point; error bars represent standard deviations from average sampling values derived from each of the triplicate cultures. sh, shRNA.

We next wanted to distinguish autocrine from paracrine effects of vIL-6. To address this experimentally, we utilized suboptimal infectious doses of our lentivirus (GFP⁺)-shRNA vectors, vIL-6 and NS (control), to achieve around 50% infection efficiency so that the cell growth properties of transduced GFP⁺ cells could be monitored relative to those of cocultured, untransduced PEL cells. BCBL-1 (infected with HHV-8) and JSC-1 (infected with HHV-8 and EBV) cultures were used for these experiments; the cell growth rates and proportions of GFP⁺ cells were compared to those of untransduced cells. vIL-6-depleted cultures had markedly reduced overall growth rates relative to those of NS-shRNA-transduced controls, and the growth rates of vIL-6-depleted (GFP⁺) cell populations were impaired relative to those of untransduced cells in the same cultures (Fig. 6B). In contrast, the proportion of NS-shRNA-transduced (GFP⁺) cells in the cultures was stable over the course of the experiment. Neither NS nor vIL-6 shRNA transduction had any effect on Akata cell growth in analogous assays. These results indicate that autocrine effects independent of the activities of secreted vIL-6 or other cytokines are involved in support of PEL cell growth by vIL-6.

Consistent with the notion of exclusively or predominantly autocrine effects of vIL-6 was our finding of detectable, low-level vIL-6 immunofluorescence staining in the majority of JSC-1 and BCBL-1 cells relative to the staining seen in Akata cells (Fig. 7A). A low proportion (1 to 2%) of cells in these cultures stained strongly for vIL-6, indicative of spontaneous lytic reactivation. To distinguish definitively between low-level vIL-6 expression and nonspecific background staining, we mixed Akata cells with JSC-1 cells and then undertook costaining for vIL-6 and HHV-8-encoded LANA, the latter allowing the identification of JSC-1 cells. The results of these experiments confirmed higher vIL-6 immunofluorescence signals in LANA⁺ JSC-1 cells than in Akata cells on the same slides (Fig. 7B). In concurrent assays, JSC-1 cells were found to be universally positive for LANA staining (Fig. 7C). Therefore, our data provide evidence for vIL-6-specific staining and latent expression of vIL-6 protein in PEL cultures.

View larger version (34K): [in this window] [in a new window]

FIG. 7. Detection of vIL-6-specific staining in the majority of PEL cells. (A) vIL-6 antiserum used in immunofluorescence assays was applied to JSC-1, BCBL-1, and Akata (negative control) cultures. Other than strong staining in 1 to 2% of PEL cells, there was evidence of faint cytoplasmic fluorescence signal in the general population relative to the signal in HHV-8-negative Akata cells. (B) Specificity of the vIL-6 fluorescence signal was determined by vIL-6 and HHV-8 LANA costaining in mixed cultures of JSC-1 and Akata cells, allowing low-level vIL-6 staining (in LANA+ cells) to be distinguished from any nonspecific background signal. General staining above background specifically in PEL cells is consistent with functional data indicating widespread latent expression of vIL-6. (C) LANA staining, demonstrating 100% LANA positivity in JSC-1 populations and therefore validating this immunofluorescence approach to positively identify PEL cells in mixed cultures of JSC-1 and Akata cells.

vIL-6 function in PEL cell growth. The basis of the observed effects of vIL-6 depletion on cell growth was investigated by cell cycle and apoptosis analyses of vIL-6 shRNA-transduced and control cultures of JSC-1 cells. For these studies, high titers of vIL-6 and NS (control) shRNA lentivirus vectors were used to ensure efficient (>90%) transduction, as determined by GFP fluorescence. As before, the growth of vIL-6-depleted cells was markedly reduced relative to that of the controls (Fig. 8A). Cell cycle profiling using Hoechst staining and flow cytometry revealed a significant reduction in the proportion of cells in S-phase and G₂/M in the vIL-6 shRNA-transduced cultures relative to the proportion in the NS shRNA-transduced cultures, 32% versus 45% (day 2) and 35% versus 44% (day 3) (Fig. 8B). Annexin V-Cy3 staining to detect apoptosis in day 3 cultures showed a twofold increase in the rate of apoptosis in vIL-6-depleted culture relative to that in the control culture (8% and 4%, respectively) (Fig. 8C). An additional experiment was carried out to investigate apoptosis in BCBL-1 cells, using suboptimally infected cultures to achieve 50% transduction efficiency. Measurements of annexin V-Cy3 and GFP⁺ fluorescence in these NS and vIL-6 shRNA-containing cultures revealed apoptosis predominantly in the GFP⁺ cell population, with more than double the rate of apoptosis in the vIL-6-depleted as in the NS shRNA-transduced cultures (Fig. 8D). Combined, these data indicate that the effects on cell growth of vIL-6 depletion are the result of both diminished rates of cell proliferation and increased cell death and that endogenously produced vIL-6 exerts its influence via autocrine signaling.

View larger version (27K): [in this window] [in a new window]

FIG. 8. Analysis of effects of vIL-6 on PEL cell proliferation and survival. (A) Measurements of JSC-1 cell growth in cultures transduced at high efficiency (>90%) with either vIL-6 or NS shRNA by lentiviral vector infection. (B) Cell cycle analysis by Hoechst 33258 staining and flow cytometry was undertaken for day 2 and day 3 samples of the JSC-1 cultures used for growth analysis (panel A). vIL-6 shRNA-transduced cultures showed a decreased proportion of cells in S and G2/M phases relative to the proportions in NS-transduced cultures. (C) Apoptosis in the day 3 culture was identified and quantified by annexin V-Cy3 staining (example shown), revealing a twofold increase in the percentage of apoptotic cells in the vIL-6-depleted cultures. (D) Apoptosis as a function of vIL-6 depletion was also studied in BCBL-1 cultures, using an shRNA transduction rate of 50% to allow the discernment of autocrine versus paracrine effects. Cells were analyzed 5 days after lentiviral infection. Most (>70%) of the apoptotic (Cy3+) cells in the vIL-6 shRNA-transduced population were GFP+, compared to 40% in the NS shRNA-transduced cultures, and there was a greater-than-twofold increase in the percentage of apoptotic cells in GFP+ vIL-6 shRNA-transduced populations relative to the percentage in NS shRNA-containing controls (17% versus 7.5%). These data indicate predominantly autocrine effects of vIL-6 depletion on PEL cell survival. For growth curves (panel A) and analysis of apoptosis (panels C and D), data were derived from triplicate cultures as outlined in the legend to Fig. 6; standard deviations from mean values are indicated. sh, shRNA.

The biological effects of vIL-6 can be mediated exclusively via intracellular mechanisms. To examine the relevance of intracellular signaling for the support of PEL cell growth, shRNA-resistant vIL-6-KDEL (and vIL-6) sequences were cloned into lentiviral vectors (GFP⁺) for use in complementation experiments. Codon-synonymous mutations were introduced into the shRNA target sequence to allow the expression of virus vector-delivered vIL-6 in endogenous vIL-6-depleted cells. JSC-1 and BCBL-1 cells were infected at high efficiency with empty lentivirus vector (control) or vectors specifying vIL-6 or vIL-6-KDEL and then superinfected with either NS or vIL-6 shRNA-containing vectors (>90% transduction efficiency, as determined by GFP fluorescence in parallel-infected naïve cultures). Measurements of the growth of the various cultures over time determined that lentivirus-transduced vIL-6-KDEL, like vIL-6, could rescue the cells from the growth-inhibitory effects of vIL-6 depletion, seen in the cells infected with virus vector alone (Fig. 9). The ability of ER-targeted vIL-6-KDEL to complement the vIL-6-depletion phenotype provides evidence of the sufficiency of intracellular, strictly autocrine signal transduction by vIL-6 for the growth-promoting function of the viral cytokine in PEL cells.

View larger version (19K): [in this window] [in a new window]

FIG. 9. Biological activity of ER-retained vIL-6. Functional complementation of shRNA (sh)-mediated depletion of endogenous vIL-6 was tested in JSC-1 and BCBL-1 cells expressing lentivirus-transduced shRNA-refractory vIL-6 or ER-targeted vIL-6-KDEL or infected with empty lentivirus vector (control). High transduction rates (>90% GFP+ cells) were achieved for vIL-6-expressing and empty lentivirus infections and subsequent superinfections with either NS (control) or vIL-6 shRNA lentiviral vectors. Viable, trypan blue-excluding cells from triplicate cultures were counted at daily time points to generate growth curves (error bars show standard deviations from mean values). Lentivirus-delivered vIL-6 and vIL-6-KDEL were able to prevent growth inhibition mediated by depletion of endogenous vIL-6.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
While published investigations of vIL-6 function have focused mainly on the influence of soluble, extracellular vIL-6 on cell growth and survival (4, 7, 13, 15, 21, 23), the role of endogenously produced vIL-6 in PEL cells has not been studied extensively, and in particular the activities of intracellular as distinct from extracellular vIL-6 have not been addressed directly. This was the primary goal of the present study. Previously, investigations touching on this issue have demonstrated that vIL-6 is capable of signaling in the ER when directed to this compartment by virtue of "KDEL" tagging (19), that ER-targeted single-chain antibodies specific for vIL-6 are able to inhibit vIL-6 signaling in HepG2 cells (16), and that vIL-6 depletion in PEL cells negatively affects cell growth (30). However, these results do not demonstrate that vIL-6 is naturally located predominantly in the ER, nor do they address the nature of the vIL-6-induced signaling complexes in this compartment or the biological activities specifically associated with ER signaling by vIL-6. The data presented in this report show that vIL-6 localizes normally and almost exclusively to the ER compartment and that here it signals via tetrameric signal-transducing complexes (vIL-6₂:gp130₂) rather than through hexameric complexes (vIL-6₂:gp130₂:gp80₂) that can form at the cell surface (2, 14). Our data provide evidence that intracellular signaling by vIL-6, as distinct from extracellularly mediated signal transduction, contributes significantly to the proliferation and survival of latently infected PEL cells. To our knowledge, this is the first demonstration of the influence of endogenous, latently expressed vIL-6 on PEL cell growth via intracrine signaling. Our findings suggest that vIL-6 may contribute significantly to the maintenance of latently infected cell pools and perhaps also to the development of PEL and other HHV-8-associated malignancies.

The mechanisms of endogenous vIL-6-induced proliferation and survival are at present unresolved and fall outside the scope of this study. However, it is important to note that such functions cannot be replaced by hIL-6, which we show here, as others have reported (25), cannot mediate signal transduction in the ER compartment. Our data further indicate that tetrameric signaling, with its distinct characteristics relative to hexameric signal transduction (14), is sufficient for these biological activities. Whether the precise differences in hexameric and tetrameric signal transduction at the cell surface from exogenously added vIL-6 reflect differences induced by endogenously produced vIL-6 in the ER and plasma membranes remains to be determined. The main differences reported previously are the longer durations and greater amplitudes of STAT1 and STAT3 activation, coupled with greater STAT1-to-STAT3 activation for vIL-6-induced hexameric complexes relative to that for tetramers (14). It is possible, however, that in the context of ER, signal transduction is altered, conceivably through distinct availabilities of or ability to recruit gp130- and STAT-activating Jaks to ER-localized, vIL-6-induced gp130 dimers or distinct accessibility of gp130 to SHP2 and SOCS, negative regulators of gp130/STAT signaling (12, 27). Regardless, it is significant that in the context of PEL cells, we were able to detect active (tyrosine phosphorylated) gp130 in the ER compartment, consistent with the location and intracrine activities of vIL-6.

The major implications of the work presented here are first, that intracellular sequestration of vIL-6 allows biological activity of vIL-6 even when expressed at very low levels, during latency, relative to those induced during lytic replication, and second, that in latently infected PEL cells, vIL-6 is expressed in most or all cells in functionally significant amounts. The latter can be concluded from our finding of clear effects on cell growth of vIL-6 depletion specifically in shRNA-transduced PEL cells when cocultured with untransduced cells, demonstrating that intracellular rather than secreted vIL-6 is biologically active. Thus, the depletion over time specifically of vIL-6 shRNA-transduced cells provides clear evidence of the operation of intracellular autocrine rather than paracrine effects of vIL-6, which may be insignificant in the context of viral latency. Further, we identified higher rates of apoptosis in vIL-6-depleted BCBL-1 cells than in their cocultured counterparts lacking transduced vIL-6-specific shRNA. The biological relevance of intracrine signaling is reinforced by the finding that ER-targeted, exclusively intracellular vIL-6 is able to "rescue" vIL-6 shRNA-transduced cells equivalently to native vIL-6.

We hypothesize that vIL-6 can function both during latency, to promote the growth and survival of infected cells for the maintenance of latent virus within the host, and in a distinct manner during lytic replication where, due to high expression levels, it can act in a paracrine manner also, potentially altering the microenvironment to enhance the productive replication and spread of the virus. As a by-product, lytically expressed vIL-6 may also influence viral pathogenesis, perhaps particularly in KS, where the angiogenic and proinflammatory properties of vIL-6 indicate potential involvement in disease development (11). Thus, we propose the possibility of two modes of vIL-6 action: during latency as an autocrine growth factor via ER and tetrameric-complex signal transduction and during lytic replication as a paracrine factor acting via both tetrameric and hexameric complexes to provide conditions conducive to virus production and propagation. Further investigation is warranted to evaluate the role of vIL-6 in productive replication and the signaling complexes and pathways involved in mediating its effects.

 
We thank Y. Choi for his advice and critical reading of the manuscript.

This work was supported by grant no. RO1-CA76445 from the National Institutes of Health.

 
* Corresponding author. Mailing address: Sidney Kimmel Comprehensive Cancer Center, Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21287. Phone: (410) 502-6801. Fax: (410) 502-6802. E-mail: nichojo{at}jhmi.edu

Published ahead of print on 5 November 2008.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Aoki, Y., E. S. Jaffe, Y. Chang, K. Jones, J. Teruya-Feldstein, P. S. Moore, and G. Tosato. 1999. Angiogenesis and hematopoiesis induced by Kaposi's sarcoma-associated herpesvirus-encoded interleukin-6. Blood 93:4034-4043.[Abstract/Free Full Text]
2
2. Boulanger, M. J., D. C. Chow, E. Brevnova, M. Martick, G. Sandford, J. Nicholas, and K. C. Garcia. 2004. Molecular mechanisms for viral mimicry of a human cytokine: activation of gp130 by HHV-8 interleukin-6. J. Mol. Biol. 335:641-654.[CrossRef][Medline]
3
3. Brousset, P., E. Cesarman, F. Meggetto, L. Lamant, and G. Delsol. 2001. Colocalization of the viral interleukin-6 with latent nuclear antigen-1 of human herpesvirus-8 in endothelial spindle cells of Kaposi's sarcoma and lymphoid cells of multicentric Castleman's disease. Hum. Pathol. 32:95-100.[CrossRef][Medline]
4
4. Burger, R., F. Neipel, B. Fleckenstein, R. Savino, G. Ciliberto, J. R. Kalden, and M. Gramatzki. 1998. Human herpesvirus type 8 interleukin-6 homologue is functionally active on human myeloma cells. Blood 91:1858-1863.[Abstract/Free Full Text]
5
5. Cannon, J. S., J. Nicholas, J. M. Orenstein, R. B. Mann, P. G. Murray, P. J. Browning, J. A. DiGiuseppe, E. Cesarman, G. S. Hayward, and R. F. Ambinder. 1999. Heterogeneity of viral IL-6 expression in HHV-8-associated diseases. J. Infect. Dis. 180:824-828.[CrossRef][Medline]
6
6. Chang, L. J., and A. K. Zaiss. 2003. Self-inactivating lentiviral vectors and a sensitive Cre-loxP reporter system. Methods Mol. Med. 76:367-382.[Medline]
7
7. Chatterjee, M., J. Osborne, G. Bestetti, Y. Chang, and P. S. Moore. 2002. Viral IL-6-induced cell proliferation and immune evasion of interferon activity. Science 298:1432-1435.[Abstract/Free Full Text]
8
8. Chen, D., and J. Nicholas. 2006. 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. 80:9811-9821.[Abstract/Free Full Text]
9
9. Chiou, C. J., L. J. Poole, P. S. Kim, D. M. Ciufo, J. S. Cannon, C. M. ap Rhys, D. J. Alcendor, J. C. Zong, R. F. Ambinder, and G. S. Hayward. 2002. Patterns of gene expression and a transactivation function exhibited by the vGCR (ORF74) chemokine receptor protein of Kaposi's sarcoma-associated herpesvirus. J. Virol. 76:3421-3439.[Abstract/Free Full Text]
10
10. Chow, D., X. He, A. L. Snow, S. Rose-John, and K. C. Garcia. 2001. Structure of an extracellular gp130 cytokine receptor signaling complex. Science 291:2150-2155.[Abstract/Free Full Text]
11
11. Ensoli, B., and M. Sturzl. 1998. Kaposi's sarcoma: a result of the interplay among inflammatory cytokines, angiogenic factors and viral agents. Cytokine Growth Factor Rev. 9:63-83.[CrossRef][Medline]
12
12. Fischer, P., U. Lehmann, R. M. Sobota, J. Schmitz, C. Niemand, S. Linnemann, S. Haan, I. Behrmann, A. Yoshimura, J. A. Johnston, G. Muller-Newen, P. C. Heinrich, and F. Schaper. 2004. The role of the inhibitors of interleukin-6 signal transduction SHP2 and SOCS3 for desensitization of interleukin-6 signalling. Biochem. J. 378:449-460.[CrossRef][Medline]
13
13. Hoischen, S. H., P. Vollmer, P. Marz, S. Ozbek, K. S. Gotze, C. Peschel, T. Jostock, T. Geib, J. Mullberg, S. Mechtersheimer, M. Fischer, J. Grotzinger, P. R. Galle, and S. Rose-John. 2000. Human herpes virus 8 interleukin-6 homologue triggers gp130 on neuronal and hematopoietic cells. Eur. J. Biochem. 267:3604-3612.[Medline]
14
14. Hu, F., and J. Nicholas. 2006. Signal transduction by human herpesvirus 8 viral interleukin-6 (vIL-6) is modulated by the nonsignaling gp80 subunit of the IL-6 receptor complex and is distinct from signaling induced by human IL-6. J. Virol. 80:10874-10878.[Abstract/Free Full Text]
15
15. Jones, K. D., Y. Aoki, Y. Chang, P. S. Moore, R. Yarchoan, and G. Tosato. 1999. Involvement of interleukin-10 (IL-10) and viral IL-6 in the spontaneous growth of Kaposi's sarcoma herpesvirus-associated infected primary effusion lymphoma cells. Blood 94:2871-2879.[Abstract/Free Full Text]
16
16. Kovaleva, M., I. Bussmeyer, B. Rabe, J. Grotzinger, E. Sudarman, J. Eichler, U. Conrad, S. Rose-John, and J. Scheller. 2006. 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. 80:8510-8520.[Abstract/Free Full Text]
17
17. Li, H., and J. Nicholas. 2002. Identification of amino acid residues of gp130 signal transducer and gp80 alpha receptor subunit that are involved in ligand binding and signaling by human herpesvirus 8-encoded interleukin-6. J. Virol. 76:5627-5636.[Abstract/Free Full Text]
18
18. Li, H., H. Wang, and J. Nicholas. 2001. Detection of direct binding of human herpesvirus 8-encoded interleukin-6 (vIL-6) to both gp130 and IL-6 receptor (IL-6R) and identification of amino acid residues of vIL-6 important for IL-6R-dependent and -independent signaling. J. Virol. 75:3325-3334.[Abstract/Free Full Text]
19
19. Meads, M. B., and P. G. Medveczky. 2004. 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. J. Biol. Chem. 279:51793-51803.[Abstract/Free Full Text]
20
20. Miles, S. A., A. R. Rezai, J. F. Salazar-Gonzalez, M. Vander Meyden, R. H. Stevens, D. M. Logan, R. T. Mitsuyasu, T. Taga, T. Hirano, T. Kishimoto, et al. 1990. AIDS Kaposi sarcoma-derived cells produce and respond to interleukin 6. Proc. Natl. Acad. Sci. USA 87:4068-4072.[Abstract/Free Full Text]
21
21. Molden, J., Y. Chang, Y. You, P. S. Moore, and M. A. Goldsmith. 1997. A Kaposi's sarcoma-associated herpesvirus-encoded cytokine homolog (vIL-6) activates signaling through the shared gp130 receptor subunit. J. Biol. Chem. 272:19625-19631.[Abstract/Free Full Text]
22
22. Mori, Y., N. Nishimoto, M. Ohno, R. Inagi, P. Dhepakson, K. Amou, K. Yoshizaki, and K. Yamanishi. 2000. Human herpesvirus 8-encoded interleukin-6 homologue (viral IL-6) induces endogenous human IL-6 secretion. J. Med. Virol. 61:332-335.[CrossRef][Medline]
23
23. Nicholas, J., V. R. Ruvolo, W. H. Burns, G. Sandford, X. Wan, D. Ciufo, S. B. Hendrickson, H. G. Guo, G. S. Hayward, and M. S. Reitz. 1997. Kaposi's sarcoma-associated human herpesvirus-8 encodes homologues of macrophage inflammatory protein-1 and interleukin-6. Nat. Med. 3:287-292.[CrossRef][Medline]
24
24. Parravicini, C., B. Chandran, M. Corbellino, E. Berti, M. Paulli, P. S. Moore, and Y. Chang. 2000. Differential viral protein expression in Kaposi's sarcoma-associated herpesvirus-infected diseases: Kaposi's sarcoma, primary effusion lymphoma, and multicentric Castleman's disease. Am. J. Pathol. 156:743-749.[Abstract/Free Full Text]
25
25. Rose-John, S., H. Schooltink, H. Schmitz-Van de Leur, J. Mullberg, P. C. Heinrich, and L. Graeve. 1993. Intracellular retention of interleukin-6 abrogates signaling. J. Biol. Chem. 268:22084-22091.[Abstract/Free Full Text]
26
26. Staskus, K. A., R. Sun, G. Miller, P. Racz, A. Jaslowski, C. Metroka, H. Brett-Smith, and A. T. Haase. 1999. Cellular tropism and viral interleukin-6 expression distinguish human herpesvirus 8 involvement in Kaposi's sarcoma, primary effusion lymphoma, and multicentric Castleman's disease. J. Virol. 73:4181-4187.[Abstract/Free Full Text]
27
27. Terstegen, L., P. Gatsios, J. G. Bode, F. Schaper, P. C. Heinrich, and L. Graeve. 2000. The inhibition of interleukin-6-dependent STAT activation by mitogen-activated protein kinases depends on tyrosine 759 in the cytoplasmic tail of glycoprotein 130. J. Biol. Chem. 275:18810-18817.[Abstract/Free Full Text]
28
28. Wan, X., H. Wang, and J. Nicholas. 1999. Human herpesvirus 8 interleukin-6 (vIL-6) signals through gp130 but has structural and receptor-binding properties distinct from those of human IL-6. J. Virol. 73:8268-8278.[Abstract/Free Full Text]
29
29. Yoshizaki, K., T. Matsuda, N. Nishimoto, T. Kuritani, L. Taeho, K. Aozasa, T. Nakahata, H. Kawai, H. Tagoh, T. Komori, et al. 1989. Pathogenic significance of interleukin-6 (IL-6/BSF-2) in Castleman's disease. Blood 74:1360-1367.[Abstract/Free Full Text]
30
30. Zhang, Y. J., R. S. Bonaparte, D. Patel, D. A. Stein, and P. L. Iversen. 2008. Blockade of viral interleukin-6 expression of Kaposi's sarcoma-associated herpesvirus. Mol. Cancer Ther. 7:712-720.[Abstract/Free Full Text]

---

Journal of Virology, January 2009, p. 722-733, Vol. 83, No. 2
0022-538X/09/$08.00+0     doi:10.1128/JVI.01517-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Dela Cruz, C. S., Viswanathan, S. R., El-Guindy, A. S., Shedd, D., Miller, G. (2009). Complex N-Linked Glycans on Asn-89 of Kaposi Sarcoma Herpes Virus-encoded Interleukin-6 Mediate Optimal Function by Affecting Cytokine Protein Conformation. J. Biol. Chem. 284: 29269-29282 [Abstract] [Full Text]  
• Chen, D., Choi, Y. B., Sandford, G., Nicholas, J. (2009). Determinants of Secretion and Intracellular Localization of Human Herpesvirus 8 Interleukin-6. J. Virol. 83: 6874-6882 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chen, D. Articles by Nicholas, J. Search for Related Content PubMed PubMed Citation Articles by Chen, D. Articles by Nicholas, J.