| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
-Inducible Lysosomal Thiol Reductase Expression by Inflammatory Cytokines1
* Department of Cell Biology and
Section of Immunobiology, Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, CT 06520
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-inducible lysosomal thiol reductase (GILT), which plays a role in MHC class II-restricted processing and presentation of Ags containing disulfide bonds, can be induced in various cell types by the cytokine IFN-. APCs, including circulating macrophages, constitutively express high levels of GILT, although the pathways regulating its expression in these cells have not been characterized. In this study, we used the promonocytic cell line THP-1, an established model for monocyte to macrophage differentiation, to investigate the induction of GILT upon exposure to bacteria. We show that contact with LPS or intact Escherichia coli causes THP-1 cells to undergo programmed differentiation, characterized by adhesion, cytokine secretion, and up-regulation of Ag processing and presentation components, including GILT. Unlike GILT induction in response to IFN- treatment, induction by bacteria is dependent on new protein synthesis, NF-
B signaling, and secretion of the inflammatory cytokines TNF and IL-1
. Furthermore, we show that both cytokines are sufficient for GILT induction in the absence of a microbial stimulus. The majority of GILT synthesized by differentiated THP-1 cells is secreted as the precursor form rather than being transported to, and maturing in, lysosomes, suggesting a novel role for GILT in cells of the macrophage lineage. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-inducible lysosomal thiol reductase (GILT)3 is a soluble protein localized in late endosomes and lysosomes of APCs, where it catalyzes the reduction of disulfide bonds (1). This activity, maximal at acidic pH, is required for the processing and presentation of specific epitopes of Ags containing disulfide bonds by MHC class II molecules (2, 3). GILT is synthesized as a proenzyme, with N- and C-terminal propeptides which are cleaved to yield the mature protein as GILT is transported through the endosomal system (1, 4). We have shown previously that 
20% of newly synthesized GILT is secreted as a disulfide-linked dimer of the precursor form by tissue culture cells (4, 5), and circulating GILT has been detected in mouse serum (our unpublished data). Although precursor GILT has been shown to possess thiol reductase activity, the functional consequences of its secretion are unknown (6).
GILT expression is induced by IFN- through a CIITA-independent, Stat-1-dependent pathway (7). IFN- induces several components of the Ag processing and presentation machinery, and leads to the activation of APC function in many cell types, a step necessary for the identification and clearance of invading pathogens from peripheral tissues (8). Although GILT is expressed constitutively in professional APCs, little is known about the factors that induce its expression in these cells during their development.
Monocytes are bone marrow-derived precursor cells that differentiate into macrophages upon entry into peripheral tissue and serve as professional APCs that play key roles in both innate and adaptive immune responses. They are highly phagocytic and express diverse pattern recognition receptors (PRRs), including TLRs, which allow the identification and phagocytosis of foreign pathogens (9). Recognition by TLRs initiates signaling events leading to the degradation of IB, which binds and sequesters the NF-B transcription factor in the cytosol of unstimulated cells, allowing the translocation of NF-B to the nucleus where it stimulates gene expression. Bacterial LPS stimulation is TLR4 dependent, and leads to NF-B activation through two signaling pathways (10). One is dependent on the adaptor protein MyD88, and activates proinflammatory cytokine and chemokine transcription, and the second is MyD88 independent and leads to the induction of IFN-
, and subsequently type-1 IFN-inducible genes, as well as inflammatory cytokines and chemokines (11). Additionally, it has recently been shown that TLR4 engagement rapidly leads to the activation of the transcription factor IFN regulatory factor 3, which stimulates TNF expression required for MyD88-independent activation of gene transcription (12, 13). Overall, TLR4 engagement leads to the recruitment of inflammatory cells to the site of infection, activation of macrophage killing activity, and the induction of Ag presentation function in macrophages.
To understand the factors leading to GILT expression in macrophages, we have studied GILT induction in the promonocytic leukemia cell line THP-1, a well-characterized model for the differentiation of immature monocytes to macrophages (14). Exposure to diverse stimuli, including PMA and LPS, causes the cells to become adherent, alter their gene expression, and secrete proinflammatory cytokines and chemokines (14). In this study, we observed differentiation-dependent induction of GILT in THP-1 cells incubated with LPS or live E. coli bacteria. GILT induced in this manner differs in its maturation and distribution compared with THP-1 cells stimulated with IFN-. We show that induction by bacteria requires de novo protein synthesis, the activation of NF-B-mediated transcription, and depends at least partially on the inflammatory cytokines TNF and IL-1. Furthermore, we show that both TNF and IL-1 are sufficient for modest differentiation and GILT induction in the absence of microbial stimuli.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
THP-1 cells were purchased from American Type Culture Collection (ATCC) and grown in RPMI 1640 medium (Invitrogen Life Technologies) with 5–7% FCS (HyClone). For stimulation experiments, cells were seeded at 0.5 x 106 cells/ml in growth medium containing 1000 U/ml penicillin/streptomycin (Invitrogen Life Technologies) and 0.04 mg/ml gentamicin sulfate (Gemini Bio-Products), and were stimulated with LPS at 1 µg/ml (Escherichia coli type 0111:B4, Sigma-Aldrich), subcloning efficiency DH5 bacteria (Invitrogen Life Technologies; Ref. 15) at a multiplicity of infection of 2 and/or IFN- at 200 U/ml (R&D Systems). HEK-293T cells were purchased from ATCC and grown in IMDM (Invitrogen Life Technologies) with 10% bovine calf serum (HyClone). Abs used in these studies were rabbit anti-human GILT antiserum (R.GILT) (1), mouse anti-human GILT mAb (MAP.IP30) (5), SPA-850 mAb (anti-GRP94; Stressgen), TRK5G4-6C5 mAb (anti-GAPDH; RDI), and rabbit anti-cathepsin D serum (DakoCytomation).
Quantitative Western blots
Cells were stimulated as described above, harvested by pipetting and scraping (as necessary) at the indicated times, washed once in PBS, and solubilized in 1% Triton X-100 (American BioAnalytical) in 0.15 M NaCl, 0.01 M Tris-Cl (pH 7.4) (TBS) at 8 x 106 cells/ml for 30 min on ice. Nonreducing SDS-PAGE sample buffer (10x) was added to 12 µl of the postnuclear supernatant, the samples were heated for 5 min at 95°C, and resolved by SDS-PAGE on a 12% gel. Separated proteins were transferred to polyvinylidene difluoride membranes (Millipore) and blocked with Blotto (5% milk (Carnation), 5% bovine calf serum in PBS with 0.2% Tween 20 (Sigma-Aldrich)) for 20 min at room temperature. Membranes were cut and probed individually for 1 h at room temperature with R.GILT serum (1/20,000) and SPA-850 (0.05 µg/ml) in Blotto. After washing, membranes were incubated with species-specific alkaline-phosphatase-coupled secondary Abs (1/5000) (Jackson ImmunoResearch Laboratories) for 30 min at room temperature, washed and imaged using a Fluorimager with ECF substrate (Amersham Biosciences). Analysis was performed with ImageQuant software version 5.2 (Molecular Dynamics). Under these conditions, the measured signal was linear with respect to GILT and GRP94 concentrations.
Metabolic radiolabeling and immunoprecipitations
THP-1 cells were stimulated with IFN- or bacteria as described for 48 h, washed once in PBS, and incubated in medium without cysteine and methionine for 1 h at 37°C. Cells were labeled for 1 h with 1 mCi [35S]methionine/cysteine labeling mix (PerkinElmer), and chased in the presence of 15-fold excess methionine/cysteine. Aliquots were collected at the indicated time points, cells were harvested by centrifugation, and supernatants were saved. Cell pellets were washed with cold PBS and solubilized in 1 ml of 1% Triton X-100 in TBS for 30 min on ice. The postnuclear supernatant and saved cell supernatant were precleared with preimmune mouse serum (Sigma-Aldrich) and protein G-Sepharose (GE Healthcare). Immunoprecipitations were conducted with MAP.IP30, TRK5G4-6C5, rabbit anti-cathepsin D, or preimmune serum and protein G-Sepharose. After washing with TBS containing 0.1% Triton X-100, the beads were eluted in reducing SDS-PAGE sample buffer, and resolved on 12% gels followed by autoradiography. Analysis was performed using ImageQuant software as above.
Generation of stable THP-1/GILT cells
Wild-type human GILT was amplified from pcDNA3.1(-)puro (1) using primers GILTFHIND (5'-ATA AAG CTT GCC ACC ATG GAT AGT CGC CAC ACC TTT GCC CCT GCT-3') and GILTRVCLA (5'-ATA ATC GAT CTA CTT GAA GCA AAC ACT CCT GAG GGA GCT GGT TGA-3') and cloned into the HindIII and ClaI sites of the pLHCX retroviral vector (BD Biosciences) to generate pLHCX/GILT. Retroviral supernatant was generated by cotransfection of pLHCX/GILT with pCL-Ampho (Imgenex) into HEK-293T cells using Lipofectamine 2000 (Invitrogen Life Technologies). Polybrene (8 µg/ml; Sigma-Aldrich) was added to supernatant harvested from overnight culture and three rounds of spinfection were conducted on THP-1 cells at 32°C for 90 min at 1260 x g, followed by overnight incubation at 32°C to generate THP-1/GILT cells. Cells were subsequently selected in growth medium containing 0.3 mg/ml hygromycin (Roche), and assayed for GILT expression and secretion. Retroviral infection did not cause THP-1 cell maturation.
Inhibitor/activator studies
THP-1 cells were treated with 10 µg/ml cycloheximide (Sigma-Aldrich) or 10 µM parthenolide (Calbiochem) for up to 2 h before stimulation, and the agents were maintained in the medium for the duration of the experiment. The NF-B activator potassium bisperoxo (1,10-phenanthroline) oxovanadate (V) (bpV(phen); Alexis Biochemicals) was added at 10 µM to THP-1 seeded at 0.5 x 106 cells/ml for 15 h. Recombinant human TNF sRI/TNFRSF1A/Fc chimera (90 ng/ml; R&D Systems) or recombinant human IL-1 sRII (5 µg/ml; R&D Systems) was added to THP-1 cells simultaneously with the bacteria. Recombinant human TNF (20 ng/ml; eBiosciences) and recombinant human IL-1 (80 ng/ml; R&D Systems) were added to THP-1 cells seeded at 0.5 x 106 cells/ml for the indicated times.
Quantitative RT-PCR
Cells were stimulated as described above and harvested at various time points by pipetting and scraping. Total RNA was extracted using the QIAQuick RNA miniprep kit (Qiagen), 200 ng of RNA was used to generate cDNA with the Prostar First Strand DNA Synthesis kit (Stratagene), and 5 µl of this reaction was used for quantitative RT-PCR using the TaqMan Gene Expression system (Applied Biosystems) on the Stratagene MX3000P.
ELISA
Supernatants collected from stimulated cells were analyzed for TNF production using BD OptEIA Human TNF ELISA Set (BD Biosciences) or IL-1 production using the human IL-1/IL-1F2 DuoSet ELISA Development System (R&D Systems) on Immulon 4HBX 96-well plates (ThermoLabsystems) and developed with Turbo substrate (Pierce) according to the manufacturer’s instructions.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
or bacteria in THP-1 cells
Although circulating human peripheral blood-derived CD14-positive monocytes constitutively express high levels of GILT protein, THP-1 cells represent an early stage of monocyte differentiation (14, 16), are CD14 negative (17), and express low levels of Ag-processing enzymes (18), including GILT. Treatment of the cells with IFN- led to robust GILT induction, in agreement with previous reports (4, 7) (Fig. 1A). Maximal expression was observed at 48-h posttreatment, and mature GILT represented the majority of the steady state cellular GILT pool, consistent with the rapid processing of precursor GILT to its mature form. To determine whether GILT could be induced by microbial stimuli, we stimulated THP-1 cells with E. coli-derived LPS (Fig. 1B). Within the first hour of exposure, the cells became adherent and subsequently GILT protein was induced. Induction was delayed relative to IFN- treatment, and less GILT protein was detected, although the amount continued to increase at 60 h. Surprisingly, the pool of intracellular GILT was composed of almost equal amounts of precursor and mature protein, suggesting altered processing or stability. Because whole bacteria are more potent inducers of TLR-mediated signaling, we next exposed the cells to gentamicin-treated live E. coli (Fig. 1C). Similar to LPS treatment, whole bacteria induced GILT with delayed kinetics relative to IFN-, with a nearly equal ratio of precursor and mature protein. Bacterial treatment led to significantly higher induction of GILT relative to LPS treatment (compare the y-axes of the graphs in Fig. 1, B and C), reaching levels comparable to IFN- treatment by 72 h, and therefore we used bacterial treatment in subsequent experiments. Simultaneous stimulation with both IFN- and bacteria led to an intermediate phenotype, where expression began at approximately the same time as with IFN- treatment alone, yet continued to increase up to 72 h after treatment. Again similar amounts of precursor and mature GILT were present in the cell (Fig. 1D). Of note, higher GILT protein levels were observed upon dual stimulation. Taken together, these results suggest that induction of GILT by IFN- and LPS/bacteria occur through distinct, potentially additive pathways that lead to different cellular phenotypes as well as different pathways of GILT maturation.
|
or bacteria
To further address the kinetics of maturation of precursor GILT, THP-1 cells were stimulated for 48 h with either IFN- or bacteria, washed, starved of cysteine and methionine, metabolically labeled for 1 h with [35S]cysteine/methionine, and chased for various time periods. Precursor and mature GILT were immunoprecipitated from the total cell lysate after detergent extraction or directly from the supernatant and resolved by SDS-PAGE. Precursor GILT from IFN--stimulated cells was processed intracellularly to its mature form or secreted as precursor GILT in roughly equal amounts, with a half-life of 5 h (Fig. 2A). In contrast, in bacterially stimulated cells, the half-life of precursor GILT was < 2 h, a large proportion of newly synthesized GILT was rapidly secreted, and little mature GILT accumulated intracellularly (Fig. 2B). This profile likely accounts for the near-equivalent amounts of precursor and mature GILT in total cell lysates present at steady state, as observed in Fig. 1C. Little of the extracellular GILT was in the mature form.
|
and bacteria also resulted in precursor GILT secretion rather than maturation, with a cellular half-life of roughly 3 h for precursor GILT, intermediate between either treatment alone. In all cases, approximately half of the protein signal was lost during the chase, likely due to degradation of newly synthesized GILT (Fig. 2C). This phenomenon was observed previously in B cells (Ref. 5 and data not shown). Exposure of the cells to LPS for 48 h before metabolic labeling led to a similar secretion phenotype (data not shown). Similar secretion profiles were observed for the proforms of the soluble lysosomal enzymes cathepsin D (Fig. 2D) and cathepsin S (data not shown), suggesting that bacterial stimulation of THP-1 cells might result in a general alteration in trafficking of lysosomal cargo. Importantly, the cytosolic protein GAPDH did not accumulate in the supernatant, demonstrating that the observed secretion was not a result of cell lysis (Fig. 2E).
Taken together, these results suggest that treatment of THP-1 cells with IFN- or bacteria induce distinct states of cellular activation. Stimulation by IFN- leads to GILT maturation seen typically in cells that express the protein constitutively, while stimulation with bacteria leads to a distinct maturation and trafficking profile, likely due to differentiation-induced reprogramming of the secretory pathway. In further support of this hypothesis, GILT expressed ectopically in retrovirally transduced unstimulated THP-1 cells matured normally, but treatment of those cells with bacteria resulted in predominant secretion of precursor GILT as early as 1 h after stimulation (Fig. 3).
|
Fig. 1, A and B, suggests that GILT protein first appears in IFN--treated cells earlier than in cells stimulated with bacteria. To verify that a temporal difference exists at the level of transcription, quantitative RT-PCR analysis was performed at various times poststimulation (Fig. 4A). GILT mRNA could be detected within 2 h of treatment with IFN-, and reached a maximum by 10 h. In contrast, in cells stimulated with bacteria, the transcript did not accumulate appreciably within the first 6 h. This suggests that IFN- directly induces transcription, while induction by bacteria may require new protein synthesis. To test this, THP-1 cells were stimulated with bacteria for 15 h in the presence of cycloheximide, an inhibitor of protein translation. Although addition of the inhibitor itself induced a low level of GILT mRNA in unstimulated cells, consistent with an earlier report (4), cycloheximide completely abrogated GILT induction by bacteria (Fig. 4B). To determine whether novel protein synthesis is required for GILT induction by IFN-, cells were incubated for 6 h (Fig. 4C) or 3 h (data not shown) with the cytokine in the presence of cycloheximide. Although cycloheximide did cause a slight inhibition of GILT induction at both time points, significant transcript was generated in the presence of the inhibitor, implying that novel protein synthesis is not absolutely required for IFN--mediated induction, consistent with previous results (7).
|
B activity
Because GILT induction in THP-1 cells is a secondary effect of bacterial stimulation, we next sought to determine the signal transduction pathway involved. LPS recognition leads to TLR4 signaling, which activates the NF-B pathway. Accordingly, THP-1 cells were preincubated with parthenolide, which inhibits IB degradation and prevents nuclear translocation of NF-B and subsequent expression of NF-B-responsive genes. Parthenolide-treated cells were exposed to bacteria for 12 h, followed by quantitative RT-PCR analysis of GILT induction. As shown in Fig. 5A, parthenolide had no effect on basal GILT transcription in unstimulated cells, but strongly inhibited GILT induction in cells stimulated with bacteria. Similar results were obtained with the peptide inhibitor SN50, which masks the NF-B nuclear localization signal (data not shown). To determine whether NF-B activity is sufficient for GILT induction, THP-1 cells were incubated with the NF-B activator bpV(phen). As shown in Fig. 5B, during a 15-h incubation GILT was induced by bpV(phen) to levels comparable to bacterial stimulation, implying a direct role for NF-B activity in GILT induction. Interestingly, the relative adherence of THP-1 cells treated with inhibitors or activators correlated with GILT induction (data not shown). As a further measure of NF-B signaling, we examined the effects of the above treatments on the secretion of TNF, an inflammatory cytokine stimulated by TLR4-mediated NF-B signaling (Fig. 5C). Untreated cells did not secrete measurable TNF, while cells stimulated with bacteria secreted robust amounts. TNF secretion correlated strongly with NF-B activity and GILT induction. Similar correlation was observed for the secretion of IL-1 (data not shown).
|
In an effort to identify the protein(s) induced by NF-B-mediated transcription that might mediate GILT induction, we performed a preliminary analysis of cytokines and chemokines detectable in the supernatant of bacterially stimulated cells relative to untreated cells. Similar to activated macrophages (19), THP-1 cells stimulated with bacteria secreted the inflammatory cytokines TNF, IL-1, and IL-6 and the chemokines IL-8, MIP-1, MIP-1, and MCP-1 (data not shown). IFN- was not detected in either untreated or bacterially stimulated cells. We next sought to determine whether neutralizing any of these factors would inhibit GILT induction. Although neutralizing Abs to IL-6, IL-8, and MCP-1 had no effect on GILT induction (data not shown), addition of either a soluble recombinant TNFR chimera (Fig. 6A) or a soluble recombinant IL-1R (Fig. 6B) inhibited GILT induction. Similarly to the studies depicted in Fig. 5, the degree of inhibition of GILT induction correlated with the relative adherence of the cells, further suggesting a link between cytokine secretion, differentiation, and GILT induction. The kinetics of TNF and IL-1 secretion were determined by ELISA analysis of the supernatants of cells treated with bacteria or IFN- (Fig. 6C). Strikingly, bacterially stimulated cells secreted both cytokines but IFN--stimulated cells did not. Both TNF and IL-1 secretion were detectable by 4 h poststimulation, just before the appearance of the GILT message at 6 h (Fig. 4A). The TNF signal peaked after 6 h and then declined slowly; in additional experiments, a second wave of secretion was observed beginning at 36 h, consistent with previous reports (12, 13) (data not shown). The IL-1 signal also peaked at 6 h, but then continued to increase with time. To determine whether TNF or IL-1 were sufficient to induce GILT expression, excess amounts of recombinant cytokines were added separately or in combination to unstimulated THP-1 cells, and GILT induction was monitored by quantitative RT-PCR (Fig. 6D). Each cytokine alone resulted in slight cellular adherence and modest, equivalent GILT induction compared with untreated samples. The simultaneous addition of both cytokines resulted in higher relative levels of adherence and GILT induction. These data clearly indicate that TNF and IL-1 are involved in stimulating GILT expression in response to microbial stimuli, and are sufficient to achieve moderate induction. The level of induction was 5- to 10-fold below that observed with bacterial stimulation, suggesting that additional factors are likely to contribute to the extent of differentiation and GILT induction in THP-1 cells.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
, typically secreted by activated T cells, induces expression of the elements of the MHC class II Ag processing and presentation machinery, including GILT, cathepsin S, MHC class II, and HLA-DM, and facilitates the clearance of infection via the adaptive immune response. Unsurprisingly, therefore, treatment of the promonocytic cell line THP-1 with IFN- led to rapid up-regulation GILT mRNA and protein (Fig. 1A). mRNA induction was independent of new protein synthesis (Fig. 4, A and C), and is consistent with the activation of GILT transcription through the JAK/Stat pathway, which leads to the phosphorylation, dimerization and binding of the Stat-1 transcription factor to the GILT promoter, as reported by O’Donnell et al. (7) for melanoma cells. The time course of induction, and the kinetics of processing and maturation of GILT are in agreement with previous reports for other cell types (1, 4, 7). GILT induction by IFN- has been shown to be necessary for the presentation of a cysteinylated Ig chain-derived peptide by an HLA-DR-positive derivative of THP-1 to CD4-positive T cells (20). Thus, the data are consistent with a model in which treatment of THP-1 cells with IFN- activates them to become competent for Ag processing and presentation, where expression of GILT is required for efficient processing of Ags containing disulfide bonds.
Exposure of THP-1 cells to E. coli LPS or whole bacteria also resulted in the induction of GILT mRNA and protein, but several lines of evidence suggest that this induction is through a unique mechanism resulting from the maturation of THP-1 cells along the macrophage lineage. Treatment with bacteria leads to cellular adherence within the first hour of exposure, while IFN- treatment leaves the cells in suspension. Induction of GILT mRNA is slower than in IFN--treated cells, and requires new protein synthesis. In addition, treatment with bacteria results in the up-regulation of surface CD14, the LPS receptor and a classic monocyte/macrophage marker, while CD14 is not detectable at the surface of IFN--treated cells (Ref. 17 and data not shown). THP-1 cells express TLR4 at the cell surface (Ref. 21 and data not shown), which together with surface expression of CD14 and MD-2 proteins, renders the cells capable of recognizing, binding, and internalizing Gram-negative bacteria. Recognition of LPS via TLR4 results in a signaling cascade through the adaptor protein MyD88, leading to the destruction of the IB protein and the translocation of NF-B to the nucleus where it activates the transcription of genes encoding inflammatory cytokines and chemokines. A second pathway downstream of TLR4 and independent of MyD88 but also dependent on NF-B activity leads to the transcription of type-1 IFNs, and a third pathway leads to the induction of IFN regulatory factor 3 independent of NF-B signaling. Consistent with differentiation of cells, bacterial exposure caused THP-1 cells to secrete inflammatory cytokines, such as TNF and IL-1, and chemokines (Fig. 6 and data not shown), while IFN- treatment did not lead to detectable TNF and IL-1 secretion. GILT induction by bacteria was found to be mediated by NF-B (Fig. 5), and, given that GILT is poorly induced by type 1 IFNs (4), it seemed likely that GILT was induced through inflammatory cytokine activity. Indeed, we were able to partially inhibit GILT induction using either a soluble TNFR chimera or a soluble IL-1 receptor, which respectively sequester TNF or IL-1 secreted into the medium (Fig. 6). Moreover, GILT induction was found to correlate with cell adherence and secretion of TNF and IL-1 in studies using inhibitors and activators. This finding for TNF is consistent with previous reports showing that many genes inducible by IFN-, such as MHC class I and class II, are also inducible by TNF, and in most cases the induction is synergistic (8). The fact that simultaneous treatment with IFN- and bacteria led to increased GILT levels relative to either treatment alone (Fig. 1) suggests a similar relationship. Moreover, IL-1 itself is induced by TNF in mononuclear cells (22), which may explain its increased secretion with time, as well the increasing GILT induction even as TNF levels begin to fall 8 h after bacterial stimulation (Fig. 6C). Treatment of unstimulated THP-1 with excess of either recombinant TNF or IL-1 led to moderate GILT induction that increased with time. The fact that much greater induction was achieved with whole bacteria argues that other factors may contribute to THP-1 cell differentiation. This is supported by the observation that enhanced induction was observed when the cells were stimulated with both TNF and IL-1 (Fig. 6D), and by the incomplete blockade of GILT induction by the soluble receptors (Fig. 6, A and B).
Given that TNF and IL-1 induction and secretion occur before GILT expression, the results suggest a model in which these cytokines regulate the differentiation process, and that GILT expression is acquired once the cells commit to the differentiation program. This is consistent with the identification of TNF as a "differentiation inducing factor" purified from stimulated human PBMC, shown to be capable of causing the differentiation of monocytic cell lines, including THP-1 cells (23), and with the observation that IL-1 stimulates bone marrow stem cell differentiation along the myeloid lineage (24). Interestingly, we have found that purified primary CD14-positive human peripheral blood monocytes constitutively secrete TNF and IL-1 and express high levels of GILT (our unpublished results).
Induction of GILT in THP-1 cells by exposure to bacteria could point to a role for GILT in the innate immune response, to its established role in Ag processing, or to a novel function for the enzyme. The fact that GILT levels rise slowly after exposure to bacteria implies that it is not involved in the initial stages of pathogen neutralization. However, the fact that most newly synthesized precursor GILT is secreted from differentiated THP-1 cells suggests that a major role in Ag processing is unlikely, at least for conventionally internalized exogenous Ags. Moreover, although GILT was shown to be necessary for proper presentation of the cysteinylated -chain peptide, this was observed in IFN--stimulated THP-1 cells, which are more characteristic of APCs than differentiated THP-1. Of note, bacterial exposure induces other proteins with no reported role in Ag processing and presentation, such as guanylate binding protein-1 (our unpublished results), suggesting that the differentiation of THP-1 cells is not solely for enhanced APC function.
The function of secreted precursor GILT is unknown, although we have detected precursor GILT in mouse serum, suggesting a physiological correlate. In addition, precursor GILT has been shown to exhibit thiol reductase activity, and though optimal at low pH, measurable activity was observed in vitro at neutral pH (6). The fact that several lysosomal proenzymes appear to be strongly secreted from differentiated THP-1 cells suggests a maturation-induced reprogramming of cellular transport pathways. Primary macrophages have been shown to exhibit this secretory phenotype (25, 26, 27), suggesting an extracellular function for secreted lysosomal enzymes. Interestingly, osteoclasts, bone-digesting cells similarly derived from the monocyte lineage, secrete soluble lysosomal enzymes into a specialized bone-resorbing pocket which is sealed off from the extracellular space, maintains an acidic pH, and functions as a localized site of bone degradation (28). Although GILT displays no detectable proteolytic activity, it may play a similar role extracellularly as it is thought to play in Ag processing, i.e., assisting in proteolysis by helping to unfold substrate proteins by reduction of disulfide bonds and thereby exposing protease sites. This could in turn lead to neutralization of extracellular pathogen and/or clearance of cell debris resulting from infection. Indeed, a recent proteomic study of proteins secreted upon the differentiation of the mouse macrophage cells line RAW.247 into osteoclasts identified precursor GILT as an abundantly secreted protein (29). Further studies are currently ongoing to establish the mechanism and functionality of precursor GILT secretion in cells of the macrophage lineage.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by National Institutes of Health Grant R37-AI23081 and the Howard Hughes Medical Institute. R.L.L. was supported by grants from the Natural Sciences and Engineering Research Council of Canada and Le Fonds Québécois de Recherche sur la Nature et les Technologies. 
2 Address correspondence and reprint requests to Dr. Peter Cresswell, Section of Immunobiology, Yale University School of Medicine, P.O. Box 208011, New Haven, CT 06520-8011. E-mail address: peter.cresswell{at}yale.edu
3 Abbreviation used in this paper: GILT, IFN--inducible lysosomal thiol reductase.
Received for publication March 13, 2006. Accepted for publication July 11, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-interferon-inducible lysosomal thiol reductase (GILT). Proc. Natl. Acad. Sci. USA 97: 745-750. -interferon-inducible protein. J. Biol. Chem. 263: 12036-12043. -Interferon-inducible lysosomal thiol reductase (GILT): maturation, activity, and mechanism of action. J. Biol. Chem. 275: 25907-25914. -inducible lysosomal thiol reductase in melanoma cells is STAT1-dependent but CIITA-independent. J. Immunol. 173: 731-735. . Annu. Rev. Immunol. 15: 749-795. [Medline]B activation. Science 309: 1854-1857. in inducing interleukin-8 production in human monocytic THP-1 cells is accompanied by up-regulation of CD14, Toll-like receptor 4, MD-2 and MyD88 expression. J. Endotoxin Res. 9: 145-153. -interferon-inducible lysosomal thiol reductase in melanomas disrupts T cell recognition of select immunodominant epitopes. J. Exp. Med. 195: 1267-1277. B through interleukin-1 signaling mediators in cultured human dermal endothelial cells and mononuclear phagocytes. J. Biol. Chem. 274: 7611-7614. This article has been cited by other articles:
|
|
 |
|
  D. R. Peaper and P. Cresswell The redox activity of ERp57 is not essential for its functions in MHC class I peptide loading PNAS, July 29, 2008; 105(30): 10477 - 10482. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Su, G. Carey, M. Maric, and D. W. Scott B Cells Induce Tolerance by Presenting Endogenous Peptide-IgG on MHC Class II Molecules via an IFN-{gamma}-Inducible Lysosomal Thiol Reductase-Dependent Pathway J. Immunol., July 15, 2008; 181(2): 1153 - 1160. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Demirel, Z. Waibler, U. Kalinke, F. Grunebach, S. Appel, P. Brossart, A. Hasilik, R. Tampe, and R. Abele Identification of a Lysosomal Peptide Transport System Induced during Dendritic Cell Development J. Biol. Chem., December 28, 2007; 282(52): 37836 - 37843. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), and mature GILT (
) are quantified in the right panels. Gels are representative of three experiments. 
), mature GILT (•), secreted GILT (
)). D and E, THP-1 cells were stimulated with E. coli as above, and cathepsin D (D) or GAPDH (E) was immunoprecipitated from the total cell lysate or supernatant. Lanes labeled C denote control immunoprecipitation with preimmune serum. Gels are representative of at least three independent experiments.
), or both (•), total RNA was isolated at the indicated time points, and GILT induction was analyzed by quantitative RT-PCR as above. Values are normalized to GAPDH expression, and calibrated to the basal level of GILT expression in cells grown in the absence of cytokines for 12 h.