of April 12, 2019.This information is current as

Immunity, and Phagocytic Uptake of CellsImpacts Chaperone Secretion, Innate Endoplasmic Reticulum Calcium Depletion

Larry Robert Peters and Malini Raghavan

http://www.jimmunol.org/content/187/2/919doi: 10.4049/jimmunol.11006902011;

2011; 187:919-931; Prepublished online 13 JuneJ Immunol 

MaterialSupplementary

0.DC1http://www.jimmunol.org/content/suppl/2011/06/13/jimmunol.110069

Referenceshttp://www.jimmunol.org/content/187/2/919.full#ref-list-1

, 23 of which you can access for free at: cites 60 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2011 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Endoplasmic Reticulum Calcium Depletion ImpactsChaperone Secretion, Innate Immunity, and PhagocyticUptake of Cells

Larry Robert Peters* and Malini Raghavan†

A number of immunological functions are ascribed to cell surface-expressed forms of the endoplasmic reticulum (ER) chaperone

calreticulin (CRT). In this study, we examined the impact of ER stress-inducing drugs upon cell surface CRT induction and the

resulting immunological consequences. We showed that cell surface expression of CRT and secretion of CRT, BiP, gp96, and PDI

were induced by thapsigargin (THP) treatment, which depletes ER calcium, but not by tunicamycin treatment, which inhibits

protein glycosylation. Surface expression of CRT in viable, THP-treated fibroblasts correlated with their enhanced phagocytic

uptake by bone marrow-derived dendritic cells. Incubation of bone marrow-derived dendritic cells with THP-treated fibroblasts

enhanced sterile IL-6 production and LPS-induced generation of IL-1b, IL-12, IL-23, and TNF-a. However, extracellular CRT is

not required for enhanced proinflammatory responses. Furthermore, the pattern of proinflammatory cytokine induction by THP-

treated cells and cell supernatants resembled that induced by THP itself and indicated that other ER chaperones present in

supernatants of THP-treated cells also do not contribute to induction of the innate immune response. Thus, secretion of various

ER chaperones, including CRT, is induced by ER calcium depletion. CRT, previously suggested as an eat-me signal in dead and

dying cellular contexts, can also promote phagocytic uptake of cells subject to ER calcium depletion. Finally, there is a strong

synergy between calcium depletion in the ER and sterile IL-6, as well as LPS-dependent IL-1b, IL-12, IL-23, and TNF-a innate

responses, findings that have implications for understanding inflammatory diseases that originate in the ER. The Journal of

Immunology, 2011, 187: 919–931.

The endoplasmic reticulum (ER) is an important site ofprotein folding, calcium storage, and intracellular sig-naling (1). A number of ER chaperones are important in

the functions of the ER. Calreticulin (CRT) is a chaperone thatmaintains quality control of glycoprotein folding by bindingmonoglucosylated proteins in the ER. CRT also contributes tocalcium storage in the ER (reviewed in Ref. 2). Several recentstudies showed that the cell surface expression of CRT can beinduced in different cell types by different cell treatments(reviewed in Refs. 3, 4). Cell death stimuli suggested to inducecell surface CRT on dying cells include UV light, gamma irra-diation, anthracyclin chemotherapeutics, such as mitoxantrone

(MTX), and platinum-based chemotherapeutics, such as oxali-platin (5–8). The presence of CRT on the surface of dead anddying tumor cells was suggested to stimulate therapeutic and pro-tective antitumor immune responses in mice (6–8). CRT on thesurface of apoptotic cells and dying tumor cells is also suggestedto function as an eat-me signal in the phagocytosis of these cells(5, 7, 9); by this mechanism, CRT could promote the presentationof Ags derived from dying cells to T cells to induce antitumorimmunity. Other mechanisms could also account for the immu-nostimulatory effects of cell surface CRT. Purified ER chaper-ones, such as heat shock protein (HSP) 90, CRT, (10) and gp96(HSPC4), have been implicated in induction of costimulatorymolecule expression and cytokine production by dendritic cells(DCs) (reviewed in Ref. 11). Although it was suggested that TLRligand contamination could account for the reported immunoge-nicity of these and other soluble HSPs, studies demonstrating theimmunogenicity of cell-associated HSPs, including HSP90 andgp96, suggested that HSPs can be immunomodulatory indepen-dently of microbial contaminants (12–14).A number of studies also suggested links between ER calcium

depletion and detection of ER-resident proteins, including CRT,in the extracellular space. Treatment of NIH3T3 cells with thecalcium ionophore A23167, which depletes intracellular calciumstores (15), resulted in secretion of gp96, an ER-resident chaper-one, and reduced intracellular levels of CRT (referred to in thatarticle as CRP55) (16). Furthermore, treatment of NIH3T3 cellswith the sarco-ER Ca2+-ATPase inhibitor thapsigargin (THP) (17)resulted in a decrease in CRT ER staining intensity and an in-crease in the colocalization of CRT with wheat germ agglutininin a non-ER cellular compartment suggested to be the Golgi (16).Another study showed that THP increased levels of surface CRTin the SH-SY5Y neuroblastoma cell line (18). In HeLa cells, adirect correlation between an agent’s ability to deplete ER

*Graduate Program in Immunology, University of Michigan Medical School, AnnArbor, MI 48109; and †Department of Microbiology and Immunology, University ofMichigan Medical School, Ann Arbor, MI 48109

Received for publication March 9, 2011. Accepted for publication May 13, 2011.

This work was supported by NIH Grant AI066131 (to M.R.), University of MichiganRheumatic Diseases Core Center Grant NIH P30 AR048310, and Immunology andCellular Biotechnology training program fellowships (to L.R.P.). This work utilizedthe Hybridoma Core(s) of the Michigan Diabetes Research and Training Centerfunded by Grant NIH P60 DK20572 from the National Institute of Diabetes andDigestive and Kidney Diseases.

Address correspondence and reprint requests to Dr. Malini Raghavan, Depart-ment of Microbiology and Immunology, 5641 Medical Science Building II,University of Michigan Medical School, Ann Arbor, MI 48109-5620. E-mailaddress: malinir@umich.edu

The online version of this article contains supplemental material.

Abbreviations used in this article: 7AAD, 7-aminoactinomycin D; AnnV, Annexin V;BFA, brefeldin A; BMDC, bone marrow-derived dendritic cell; CM, conditionedmedia; CMFDA, 5-chloromethylfluorescein diacetate; CRT, calreticulin; DC, den-dritic cell; ER, endoplasmic reticulum; HSP, heat shock protein; MEF, mouse em-bryonic fibroblast; MTX, mitoxantrone; THP, thapsigargin; TUN, tunicamycin; UPR,unfolded protein response; WT, wild-type.

Copyright� 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1100690

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

calcium and its ability to induce surface CRT expression was alsoshown (18). Furthermore, two other groups showed that ER cal-cium depletion by THP resulted in secretion and surface expres-sion of BiP (19, 20). Finally, we recently showed that amino acidresidues that contribute to the polypeptide-specific chaperoneactivity of CRT also influence its surface expression in THP-treated mouse embryonic fibroblasts (MEFs) (21).Calcium depletion in the ER impairs protein folding (1, 22) as

a result of its impact on the functional activities of calcium-dependent chaperones (23). In calcium-depleted cells, the un-folded protein response (UPR) pathway, a cellular stress responsepathway activated by the presence of misfolded proteins in theER, is induced (24). Different arms of the UPR pathway functionto stop protein translation, increase syntheses of molecular chap-erones, and induce other cellular changes toward the restorationof the folding capacity of the ER (reviewed in Ref. 24). THP isa commonly used drug to induce the UPR pathway and, as notedabove, THP induces calcium depletion (17). Another drug thatis commonly used to induce the UPR pathway is tunicamycin(TUN), which inhibits protein glycosylation, thereby causingprotein misfolding in the ER (25). In the present study, we com-pared the impact of these two ER stress-inducing drugs, as wellas the chemotherapeutic MTX, on cell surface expression of CRTin different cell types. We showed that cell surface expression ofCRT and secretion of other ER proteins, including CRT, are spe-cific to the ER calcium depletion-induced form of ER stress. Cellsurface expression of CRT in THP-treated cells involves a failurein effective ER retention mechanisms for a number of ER-residentproteins. This mode of CRT release from the ER is distinct fromthat described for anthracyclin-induced surface CRT expression,for which cotranslocation with ERp57 is suggested (26, 27).Nonetheless, surface CRT induced by THP treatment does en-hance the phagocytic uptake of cells. The presence of variousextracellular chaperones, including CRT, allowed us to examinefunctions of the extracellular forms of CRT and other ER chap-erones in inducing an innate immune response. We found thatincubation of bone marrow-derived DCs (BMDCs) with THP-treated fibroblasts or conditioned media (CM) from the treatedcells enhanced production of IL-6 under sterile conditions but notthe production of IL-1b, IL-12, or TNF-a, cytokines previouslyshown to be induced by macrophages following exposure to ex-tracellular gp96 (14) or a CRT fragment (10). Furthermore, directTHP treatment of BMDCs enhanced IL-6 production under sterileconditions, as previously shown in macrophages (28), and en-hanced various proinflammatory responses to LPS more broadlyand significantly than did BMDC treatment with TUN. Together,the findings pointed to strong synergies between ER calcium de-pletion and innate immune responses and suggested that the com-bined presence of several ER chaperones in the extracellular en-vironment per se is not strongly proinflammatory.

Materials and MethodsAnimals

Mice were maintained in a specific pathogen-free facility and cared foraccording to the University Committee on Use and Care of Animals-approved protocols. Mice were euthanized by CO2 asphyxiation.

Drugs and Abs

TUN and MTX were purchased from Calbiochem or Sigma and used at10 mg/ml or 1 mM, respectively, unless indicated otherwise. THP was pur-chased from both Sigma and Enzo Life Sciences and used at 5 mM, unlessindicated otherwise. MTX, THP, and TUN were dissolved in DMSO andstored in single-use aliquots at 220˚C at stock concentrations of 1 mM, 5mM, and 10 mg/ml, respectively. Frozen aliquots and freshly dissolved drugsdid not show differences in measured activities. Chicken anti-CRT and rabbit

anti-CRT were purchased from Affinity Bioreagents, now part of ThermoScientific (PA1-902A), and Abcam (ab2907), respectively, and were used forflow cytometry at a concentration of 2.5 mg/ml and a dilution of 1:200, re-spectively. Annexin V (AnnV)-FITC was purchased from Biovision and BDBiosciences. 7-Aminoactinomycin D (7AAD) was purchased from Sigma.All other fluorescent secondary Abs were purchased from Jackson Immu-noresearch Laboratories. For immunoblots, goat anti-CRT (sc-7431; SantaCruz) was used at a dilution of 1:2500, and secondary Abs conjugated to HRPwere purchased from Jackson Immunoresearch Laboratories and used ata dilution of 1:30,000. Mouse anti-BiP (610978; BD Biosciences) and rabbitanti-CRT (ab2907 Abcam) were used in immunoprecipitations at a concen-tration of 1:1000 or 1:1666, respectively. Rabbit anti-PDI (1:2500; sc-20132;Santa Cruz), mouse anti-BiP (610978; BD Biosciences), rabbit anti-HMGB1(ab18256; Abcam), and rabbit anti-GP96 (1:1000; 2104S; Cell SignalingTechnologies) were used in immunoblots.

Cell culture and DC preparations

All cell lines and primary thymocytes and splenocytes were maintainedin RPMI 1640 media (Life Technologies) supplemented with 100 mg/mlstreptomycin, 100 U/ml penicillin (some media additionally contained0.25 mg/ml the antimycotic amphotericin B), and 10% FBS (v/v) (Invi-trogen). One exception was that during experiments in which immuno-blotting analyses of ER proteins in the supernatants or CM of drug-treatedcells were undertaken, the samples were generated in 1% serum instead of10% serum to minimize background bands from serum proteins. Onlycells testing negative for mycoplasma using a PCR-based detection assay(VenorGem; Sigma) were used in experiments measuring cytokine pro-duction. Mycoplasma2 lines were maintained in media containing a pro-phylactic concentration of the antimycoplasma antibiotic, plasmocin(5 mg/ml; InvivoGen). CRT-deficient (CRT2/2 K42) and wild-type (WT)(K41) MEFs were a kind gift from Dr. Marek Michalak (University ofAlberta, Edmonton, AB, Canada). The CT26 mouse colon cancer cell linewas purchased from American Tissue Culture Collection. Primary BMDCswere derived by culturing mouse femur and tibia bone marrow in GM-CSF–containing media for 5–7 d in 24-well plates after treating with redcell lysis buffer for 2 min on ice (Sigma). BMDC media were made up ofRPMI 1640 supplemented with 100 mg/ml streptomycin, 100 U/ml ofpenicillin, 10% FBS (v/v), 1 mM HEPES, 1 mM sodium pyruvate, 0.1 mMMEM Non-Essential Amino Acids (Life Technologies), and 50 mM 2-ME.BMDC media were changed every other day, and nonadherent and looselyadherent cells were harvested by pipetting. CRT-deficient MEFs expressingWT or mutant CRT or an expression vector with no inserted gene wasgenerated as described (29). Human multiple myeloma lines NCI H929(H929) and RPMI8226 were the kind gift of Drs. Malathi Kandarpa andAndrzej Jakubowiak (University of Michigan, Ann Arbor, MI). Murinethymi from C57BL/6 or BALB/c mice were homogenized with a syringeplunger and cell strainer. Murine spleens were prepared as the thymi, withan additional 2 min of treatment with red cell lysis buffer and subsequentwash before being plated. All cells were maintained at 37˚C and 5% CO2.

Flow cytometry

ForCRTsurfaceanalysisbyflowcytometry,cell linesweredetachedwithPBS+ 8mMEDTA, washed in PBS, and resuspended in PBS containing 2% FBS(flow cytometry buffer). A total of 1.5–23 105 cells was stained in 70ml flowcytometry buffer containing the respective Ab. For CRT + block stains, thesame dilution of chicken anti-CRTwas preincubated for 15–30 min at roomtemperature with a saturating concentration of peptide (KEQFLDG-DAWTNRWVESKHK) corresponding to the sequence in CRT that the Abwas generated against and affinity purifiedwith.Cellswerewashed two timeswith flow cytometry buffer, resuspended in 70 ml buffer containing donkeyanti-chicken Ig conjugated to PE (1:250), and incubated for 30–60 min withrocking at 4˚C. Cells were then washed two times. Finally, cells wereresuspended in a 13 dilution of AnnV-FITC and 7AAD in AnnV bindingbuffer (10mMHEPES [pH7.4], 140mMNaCl, and 2.5mMCaCl2), and datafor each sample were collected on a FACSCanto flow cytometer (BD Bio-sciences). Analysis was performed using FlowJo 8.8.6 (TreeStar).

Immunoprecipitations

The day before drug treatment, 1–2 3 106 cells were seeded in a 10-cmdish and allowed to grow overnight. The next day, the media were re-moved, and cells were treated with different drugs for 5 h, as describedabove. Supernatants from each treatment were centrifuged to remove celldebris, transferred to fresh tubes, and incubated overnight with the in-dicated Abs at 4˚C, with gentle agitation. The following day, samples werecentrifuged to remove precipitated proteins, transferred to new tubes, andincubated for $5 h with Protein G beads (GE Healthcare). The beads were

920 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

washed three times with 1 ml 1% Igepal CA-630 in PBS and then boiled inthe presence of SDS and DTT. The samples were analyzed by SDS-PAGEand immunoblotting with the indicated Abs. In parallel, cells from eachplate were harvested and lysed in Triton X-100 lysis buffer (10 mMNa2HPO4, 10 mM Tris, 130 mM NaCl, 1% Triton X-100, complete EDTA-free protease inhibitors [pH 7.5]) on ice for $1 h, followed by a 30-mincentrifugation at 4˚C to remove cell debris. Protein concentrations ofcleared lysates were determined with Pierce bicinchoninic acid assay, andthese lysates were included in the immunoblotting analysis of the immu-noprecipitated proteins.

Cytokine profiling of BMDCs coincubated with drug,drug-treated MEFs, or MEF-derived CM

For data in Fig. 5, MEFs were drug treated as indicated for 5–6 h; sub-sequently, cells were harvested, washed two or three times (first in 25 mlmedia, then either twice in 10 ml or once in 25–40 ml media), and in-cubated alone or with BMDCs, as described below. For data shown in Fig.6, MEFs were treated with indicated drugs for 5–6 h, supernatants werecollected (CM1), cells were washed three times in 7 ml PBS, and 4 mlBMDC media was added to the cells. After a 5–6-h collection period, CM2were harvested. CM1 and CM2 were centrifuged to clear any floating cellsor cellular debris. For Fig. 6A, a portion of CM2 was added to BMDCs(MEF CM2 + DC), and another portion was directly tested for the presenceof specific cytokines (MEF CM2). For Fig. 6B, CM1 or CM2 were addeddirectly to BMDCs, as in Fig. 6A. For Supplemental Fig. 3, CM1 and CM2were dialyzed using 3- or 10-kDa Centricon filters, as indicated (using atleast three successive concentrations and dilutions), before adding retentateor flow-through to BMDCs. Separate wells of DCs were directly treatedwith the indicated concentrations of the described drugs (direct treatment)or left untreated. In parallel analyses, a second set of MEF plates wastreated as described above for generation of CM2. MEFs were then har-vested and stained with AnnV and 7AAD (to measure cell viability at theend of the MEF CM2 collection). Protocols for BMDC incubation withtreated cells, CM, or drug were adapted from Torchinsky et al. (30). Day-5or -6 BMDCs (5 3 105) were mixed with 1 3 106 washed target cells,cosedimented, and incubated for 19–23.5 h (Fig. 5) or were treated sep-arately with drugs for 21–23 h (Fig. 6B) or with MEF CM for 15–23 h (Fig.6) in 24-well plates in the presence or absence of described amounts ofLPS before centrifuging the plates for 5 min at 1200 rpm and harvestingthe resulting supernatants. Supernatants were submitted to the Universityof Michigan Cancer Center Immunology Core to detect the indicatedcytokines by ELISA. Duoset ELISA development systems were purchasedfrom R&D Systems (Minneapolis, MN).

Phagocytosis assays

Target cells were labeled with CellTracker Green 5-chloromethyl-fluorescein diacetate (CMFDA) (Invitrogen) in 10-cm plates that wereseeded with 2–2.5 3 106 target MEFs/plate 12–20 h before labeling. Eachplate was labeled as directed with 0.1–0.35 mM CMFDA diluted in Opti-MEM serum-free media for 20–40 min at 37˚C. Media were removed,replaced with previously described tissue culture media (RPMI 1640 with10% serum), and incubated at 37˚C for 30–60 min. After this, the mediawere removed and replaced with fresh media. Next, the plates were exposedto 3 min of light on a UV light box or left untreated, and all plates wereincubated overnight in a tissue culture incubator. The following morning,the plates were treated with 5 mM THP or 10 mg/ml TUN for 5–6 h or leftuntreated as indicated. Cells were finally harvested (UV) or washed twicein 10–12 ml PBS (untreated and drug treated), harvested with 8 mM EDTAin PBS, and diluted in RPMI 1640 for counting (a third wash) beforepelleting and resuspending at the appropriate concentration. A total of 1 3105 day-5 or -6 BMDCs were mixed with 4 3 105 targets in 96-well plates,and the plates were incubated at 37 or 4˚C for 60 min. Cells were thenfixed for 10 min with formalin (fisher PROTOCOL), diluted 1:10 in flowcytometry buffer (PBS + 2% FBS), washed twice, and stained with anti–CD11c-allophycocyanin (1:220; BD Biosciences) for 20–40 min at 4˚C.Cells were washed twice, and data for each sample were collected on theFACSCanto flow cytometer (BD Biosciences). Analysis was performedusing FlowJo 8.8.6 (TreeStar). CMFDA was read on the FITC channel.

ResultsDifferential impact of ER stress-inducing drugs on cell surfaceCRT

THP inhibits sarco-endoplasmic reticulum Ca2+-ATPase pumpsand causes a reduction in levels of ER calcium (17). We recentlyshowed that THP treatment of MEFs for short times (5–6 h) in-

duced cell surface expression of CRT (21), as previously shown inneuroblastoma cells (18), and that THP treatment of MEFs alsoinduced CRT secretion (21). In the present study, we askedwhether treatment of cells with another drug commonly used toinduce ER stress, TUN, would also result in cell surface and se-creted CRT. Additionally, the effect of treating cells with theanthracyclin chemotherapeutic MTX was tested, because previousstudies suggested that MTX is a strong inducer of cell surfaceCRT in preapoptotic and apoptotic cells (7, 8, 26).We first examined cells subjected to short drug exposures, which

left the large majority of cells intact. AnnV was used to distinguishthe apoptotic population that had exposed phosphatidylserine onthe outer leaflet of the plasma membrane (31). Furthermore, 7AADwas used, which is impermeable to live and early apoptotic cellsbut stains late apoptotic/secondary necrotic cells that have lostmembrane integrity. The analyses were initially conducted withWT MEFs or CRT2/2 MEFs to control for the specificity of Abbinding. Untreated cells or cells exposed to TUN, THP, or MTXfor 5–7 h were triple stained with AnnV-FITC, chicken anti-CRTfollowed by anti–chicken-PE, and 7AAD and analyzed by flowcytometry. As previously shown (21), in the AnnV2, 7AAD2

gate, which makes up 80–90% of cells in all treatment groups(Fig. 1A), THP treatment significantly enhanced surface CRTrelative to untreated cells (Fig. 1B). A THP-induced increase inCRT surface staining of treated cells relative to untreated cells wasobserved in WT MEFs but not in CRT-deficient MEFs, demon-strating specific staining for CRT (Fig. 1B). The THP-inducedincrease in CRT staining was also inhibited by pretreatment ofthe anti-CRT Ab with the CRT-derived peptide epitope (peptideblock) that had been used as an immunogen for generating theanti-CRT Ab (anti-CRT + block), demonstrating that the stainingwas specific to the Fab region of the Ab. Consistent with a studyby Obeid et al. (7), TUN did not induce surface CRT expressionin viable cells. However, in contrast with the same study, we sawlittle preapoptotic induction of surface CRT in response to shortMTX treatments relative to the CRT upregulation induced by shortTHP treatments of cells (Fig. 1B).Because of the contrast of our data with the findings presented by

Obeid et al. (7), we measured changes in surface CRT levels in theCT26 mouse colon cancer cell line used in that study. The stronginduction of cell surface CRT seen in THP-treated WT MEFs wasalso seen in CT26 cells. In contrast, MTX treatment did not ex-pose surface CRT in CT26 cells, similar to the results obtainedwith WT MEFs (Fig. 1B). It is possible that expression of cellsurface CRT in MTX-treated cells occurs with different kineticsthan in THP-treated cells or in cells that are at a different stage ofcell death. To examine the impact of longer-duration drug treat-ment on surface CRT induction, we next examined cells treatedwith each drug for 17 h, which induced ∼2–5 times more apo-ptotic cells (Fig. 1C). This allowed us to compare surface CRTlevels on AnnV+ apoptotic cells and AnnV2 preapoptotic cellswith those on viable, AnnV2 untreated cells (Fig. 1D). None ofthe long drug treatments induced a significant increase in CRTstaining in apoptotic or preapoptotic MEFs, although the MTX-treated, apoptotic cells had a trend suggesting a slight upregula-tion of surface CRT relative to viable, untreated cells (Fig. 1D).However, interpretation of some of the data presented in Fig. 1D

may be complicated by the fact that apoptotic cells are smallerthan viable cells (Fig. 2A). To further examine this point, wecompared the level of surface CRT staining in the presence orabsence of the anti-CRT peptide block within a given viable orapoptotic cell population (Fig. 2B, 2C). Fig. 2B and 2C includesome of the same data shown in Fig. 1B and 1D, but analyzeddifferently as described, and they include analyses of additional

The Journal of Immunology 921

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

cell types. These analyses showed similar trends of induction inAnnV2 cells as measured in Fig. 1B. Interestingly, MTX-treatedAnnV2 cells had slightly higher levels of CRT than did untreatedAnnV2 WT MEFs. This difference was statistically significant, aswell as specific to the WT MEFs (Fig. 2B). These data analysesalso confirmed that apoptotic cells resulting from long THPtreatments do not have elevated levels of cell surface CRT, sug-gesting that the association of CRT with the cell surface is tran-sient or unstable (Fig. 2C). Consistent with the data analyses fromFig. 1D, the analyses of Fig. 2C suggested that apoptotic cells

resulting from longer MTX treatments have elevated levels ofCRT, although the differences did not achieve statistical signifi-cance. Overall, it seems that MTX-induced surface CRT is lowerin AnnV2 cells at early time points relative to that induced byTHP and is quite variable in AnnV+ cells at later time points.Additionally, although THP-induced CRT surface expression wasobservable in WT MEFs and CT26 cells following short drugtreatments, similar treatments of other cell types, including pri-mary murine splenocytes, thymocytes, and other cancer cell lines,did not strongly induce cell surface CRT (Fig. 2B, 2C). It ispossible that the cancer cell lines display cell surface CRT in theabsence of drug treatments, as recently suggested (32), whichaccounts for the lower levels of THP-induced surface CRT (Fig.2B). Further investigations will be needed to understand the celltype-specific effects of THP treatments.

Several ER-resident proteins are released in THP- but notTUN-treated cells, and CRT surface expression is independentof ERp57 cotranslocation

We examined secretion of ER-resident proteins in response to 5-htreatments of WT MEFs with different drugs. Supernatants fromthe different drug-treated cells were separated by SDS-PAGE,and immunoblotting analyses were performed with Abs directedagainst gp96 and PDI. In addition, because neither CRT nor BiPwas detectable by direct immunoblotting analyses, proteins presentin supernatants were immunoprecipitated with Abs against CRTand BiP, followed by immunoblotting analyses to detect thoseproteins. THP, but not TUN or MTX, induced secretion of at leastfour proteins that are normally retained within the ER: CRT, gp96,PDI, and BiP (Fig. 3A). Cells treated with 200 nM or 5 mM THPsecreted similar levels of gp96 and PDI (Fig. 3B) and inducedsimilar elevations in surface CRT expression (data not shown).Although depletion of ER calcium is expected to occur almostimmediately following THP treatment (17), THP-induced secre-tion of ER chaperones was not detectable at 1.25 or 2.5 h fol-lowing THP treatment (Fig. 3C). These findings suggested that ERcalcium depletion alone is insufficient to induce a detectable lossof ER retention of chaperones. Previous findings from other celltypes indicated that BiP (33, 34) and CRT (35, 36) mRNA levelswere upregulated following 2–5 h of TUN and THP treatment andthat BiP protein levels were increased within 5 h of TUN treat-ment (34). Consistent with these kinetics of UPR induction, anexamination of changes in BiP and CRT protein levels revealedsmall increases in CRT and BiP expression at 3 h and a strongerinduction 5 h post-TUN treatment (Fig. 3D). Over the same timeframe, THP treatment decreased the levels of cellular CRT andBiP, with a greater reduction seen at 5 h compared with 3 h (Fig.3D), consistent with the observation of increases in ER chaperonesecretion over this time frame (Fig. 3C). Taken together, thesefindings suggested that both depletion of ER calcium and in-creases in ER chaperone expression are required for chaperones toescape ER retention in THP-treated cells.The data in Fig. 3A indicate that THP treatment does not result

in a specific release of CRT but rather in a generalized loss ofER retention of various ER proteins. This mechanism of CRT re-lease from the ER is distinct from that suggested for anthracyclin-induced CRT surface expression, for which the specific cotrans-location of CRT–ERp57 complexes was indicated (26, 27). In theER, CRT and ERp57 interact via the P-domain of CRT, and aminoacid W244 within the P-domain of CRT is important for binding toERp57 (29, 37, 38). To test the requirement for CRT–ERp57 in-teraction for CRT surface expression in THP-treated cells, WTCRT or CRT mutants unable to bind to ERp57 (29) (a W244Apoint mutant and a CRT truncation mutant lacking the P domain

FIGURE 1. A specific form of ER stress induces cell surface CRT. Cell

death profiles (A, C) and surface CRT-expression profiles (B, D) of dif-

ferent treatment conditions were analyzed by flow cytometry. WT or

CRT2/2 MEFs were incubated in 5 mM THP, 1 mMMTX, 10 mg/ml TUN,

or media alone (UNT) for 5–7 h (A, B) or 17 h (C, D). Cells were then

stained with anti-CRT, anti-CRT preincubated with the CRT-derived pep-

tide used as the immunogen and used to affinity purify the Ab (anti-CRT

+block), or secondary Ab alone (control). All cells were also stained with

AnnV and 7AAD and analyzed by flow cytometry. The median fluores-

cence values of anti-CRT staining of treated cells were measured for the

AnnV27AAD2 (B, D, as indicated) or the AnnV+7AAD2 populations (D,

as indicated) and plotted as a ratio relative to the median fluorescence of

AnnV27AAD2, untreated cells stained under the same condition (B, D).

Inset in B shows representative overlays of untreated (UNT) or THP-

treated (dashed) CT26 cells stained with anti-CRT Ab. THP-treated cells

were stained in the presence (block [line]) or absence (THP [dashed]) of

the peptide block. A and C, Percentage of cells in the AnnV2/7AAD2

(viable/preapoptotic), AnnV+/7AAD2 (early apoptotic), and AnnV+/7AAD+

(late apoptotic/secondary necrotic) populations are shown. Graphs show

averages of 5–10 (A), 5–7 (B), 3–5 (C), or 2–3 (D) experiments. The p

values from two-tailed, paired t tests are indicated.

922 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

[CRT(DP)]) were expressed in CRT-deficient MEFs using retro-viral infections. Both mutant CRT proteins were induced on thecell surface in response to THP treatment at levels that correlatedwith their total expression in the lysates rather than with theirabilities to interact with ERp57 (Fig. 3E). Thus, CRT does notrequire association with ERp57 to be expressed on the cell surfacein response to ER calcium depletion. It is unclear whether ERp57is present in supernatants of THP-treated cells, because it was notdetectable in cell supernatants by immunoblotting analyses (datanot shown); however ERp57 migrates in close proximity tobackground bands present in the cell supernatants, and low levelsof secreted ERp57 may be masked by the background signals.

In THP-treated cells, ER chaperones follow the secretory routeto gain access to the extracellular compartment

As previously described (39), analysis of cell lysate-derived CRTby immunoblotting revealed that THP treatment of cells inducedCRT glycosylation (Fig. 3A, CRT blot, lane 4) in a subset of CRTmolecules present in the cell. The glycosylation site of CRT seemsto be buried under normal conditions, but it becomes partly ex-posed under calcium-depleting conditions (21). It was of note thatthe glycosylated CRT species was absent from cell supernatants,suggesting that it was either ER retained or preferentially retainedon the cell surface (Fig. 3A) (21).Because CRT glycosylation is partial and induced by specific

cellular conditions, we assessed the glycosylation status of gp96,a constitutively glycosylated protein, to further understand thecellular exit route for ER chaperones in THP-treated cells. Acytosolic route was recently described for CRT release from ap-optotic cells (40). Because cytosolic localization of proteins in-duces their deglycosylation, we assessed the glycosylation statusof extracellular gp96 to further understand whether a secretory orcytosolic route was used in the trafficking of gp96 (and by in-ference, other ER proteins) to the extracellular space in THP-treated cells. Previous findings indicated that gp96 secreted inresponse to cellular calcium perturbation exits the cell throughthe secretory pathway and becomes differentially glycosylated asit travels through the Golgi (16). Consistent with these findings,gp96 from cell supernatants migrated more slowly than did gp96in cell lysates (Fig. 4A). When supernatants from THP-treatedcells were digested with peptide:N-glycosidase F, which cleavescarbohydrate residues from glycoproteins, and analyzed by im-munoblotting with an anti-gp96 Ab, a decrease in m.w. was ob-served, indicating that secreted gp96 is indeed glycosylated (Fig.4B, gp96 blot, lanes 8, 9). Finally, the induction of cell surfaceCRT in response to THP was reduced when the cells were pre-treated with brefeldin A (BFA), which blocks ER–Golgi traffick-ing (Fig. 4C). Together, these findings indicated that THP treat-ment of cells perturbs normal ER retention mechanisms and thatER proteins exit the cell via a secretory route in response to THPtreatment.

In a CRT-independent manner, THP-treated cells, cellsupernatants, and THP itself induce and enhance innateimmune responses more broadly and significantly thancorresponding TUN treatments

Using THP-treated WT or CRT-deficient MEFs, we investigatedwhether cell surface and/or extracellular CRT was linked to en-hanced proinflammatory cytokine production. We first examined

FIGURE 2. Impact of cell size, cell type, and drug treatment on surface

CRT expression and detection. A, Representative forward scatter (FSC-A)

and side-scatter (SSC-A) density plot of MTX-treated apoptotic WT MEFs

(left panel) showing the gating of larger, higher forward scatter cells and

smaller, lower forward scatter cells that are AnnV low (larger cells rep-

resented by the gray filled peak) and AnnV high (smaller cells represented

by the black line), respectively (right panel). Bar graphs show cell surface

CRT expression of indicated cell types subjected to the indicated treat-

ments for short (B) or long (C) times, unless described otherwise. Levels of

cell surface CRT are represented by the median fluorescence of the anti-

CRT stain relative to that of the anti-CRT + block stain. WT MEF, CRT2/2

MEF, and CT26 bars show data from Fig. 1 (with the exception of CT26

long treatment time point/AnnV+ data not shown in Fig. 1D, which is

shown here) reanalyzed as the CRT/CRT + block ratio for a given cell

population. Drug treatment times for the data not shown in Fig. 1 were

THP (H929 and RPMI8226), 4.5 or 6 h (B, C) and TUN (CT26) 17 h, THP

(CT26) 17 or 21 h, MTX (CT26) 14–21 h, and MTX (splenocytes and

thymocytes), 7–15 h (C). The average cell death profiles (AnnVand 7AAD

double negative/AnnV single positive) and n values for cell types not

described in Fig. 1 were H929 (n = 2) UNT-82/3.6, THP-56/25; RPMI8226

(n = 2) UNT-63/4.3, THP-51/10 (B, C); splenocytes UNT-58/16.5 (n = 5);

thymocytes UNT-72/17.9 (n = 11) (B); CT26, UNT 94/0.5 (n = 4), TUN

93/0.3 (n = 1), MTX 68/2.9 (n = 3), THP 76/4.3 (n = 2); splenocytes MTX-

14.4/17.8 (n = 4); thymocytes MTX-7.3/62 (n = 7) (C). Unless indicated

otherwise, long drug treatments analyzed the combination of nonadherent

and adherent cells, and short drug treatments analyzed adherent cells only.

RPMI8226 and H929 are naturally nonadherent cells. C, In one experi-

ment, only adherent CT26 cells were analyzed following a 14-h MTX

treatment. The p values from two-tailed, paired t tests are indicated.

The Journal of Immunology 923

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

production of IL-6, a cytokine previously shown to be induced byTHP (28) and TUN (41) treatments of murine macrophages. WTor CRT-deficient MEFs were treated with TUN, MTX, and THPfor short (5–6.5 h) or long (20.5 h) times, as described in Fig. 1,harvested, washed extensively with PBS, and coincubated withBMDCs or alone, and cytokines in the supernatants of thesecultures were quantified by ELISA. BMDC coincubations withMEFs treated with THP resulted in higher levels of IL-6 comparedwith BMDC coincubations with other target cells (Fig. 5A);however, these effects were independent of CRT expression. ATHP-induced increase in IL-6 production was also seen withMEFs subjected to longer drug treatments (Supplemental Fig. 1B),although the levels of IL-6 measured were lower (SupplementalFig. 1B compared with Fig. 5A). These findings are consistentwith higher levels of MEF apoptosis following long drug treat-ments (Supplemental Fig. 1A compared with Fig. 1A) and theexpected immunosuppressive effects of apoptotic cells (42). It isnoteworthy that MTX-treated, preapoptotic (Fig. 5) or apoptotic(Supplemental Fig. 1) cells did not induce significant proin-flammatory cytokine production by BMDCs, despite being re-ported to induce strong antitumor immune responses in mice (7).We also examined production of IL-12, IL-1b, and TNF-a,

cytokines that were shown to be induced by a membrane-linkedextracellular form of gp96 (14) or a bacterially expressed CRTfragment (10). Additionally, we compared induction of IL-23,

because both THP and TUN were recently reported to enhanceexpression of this cytokine in the presence of TLR agonists (43).Coincubating THP-treated MEFs with BMDCs strongly enhancedTLR-dependent (0.5–2 ng/ml LPS) production of IL-1b, IL-12p70, and IL-23 compared with coincubating other drug-treated or untreated MEFs with BMDCs (Fig. 5B–D). CRT wasnot required for enhanced cytokine production in response toTHP-treated cells; in fact, CRT deficiency in the THP-treatedMEFs seemed to correlate with higher levels of cytokine pro-duction. However, observed differences between WT and CRT-deficient cells were not statistically significant (Fig. 5B–D). Com-pared with other drug treatments, THP-treated cells subjected tolong (20.5 h) drug treatments also stimulated IL-1b and IL-23(Supplemental Fig. 1C, 1D), although at reduced levels comparedwith coincubations with MEFs subjected to short drug treatments(Fig. 5). Immunostimulatory effects of TUN-treated apoptoticcells upon IL-23 production were also measurable followinglonger drug treatments but at reduced levels compared withTHP-treated cells (Supplemental Fig. 1C).To better understand the molecular basis for the immunosti-

mulatory effects of THP-treated cells (Fig. 5), we next examinedthe impact of CM from various drug-treated cells, as well as directdrug treatments of BMDCs, on induction or enhancement ofproinflammatory cytokine production (Fig. 6). We were particu-larly interested in determining whether secreted ER chaperones

FIGURE 3. Various ER chaperones are released from THP-treated MEFs, and THP-induced surface CRT expression is independent of CRT–ERp57

binding. A, Immunoblotting of cell supernatants and lysates for the presence of various ER chaperones. Supernatants (CM1) from cells treated with TUN,

MTX, or THP or untreated cells were harvested in parallel with the corresponding cell lysates. Direct immunoblotting analyses of each fraction were

performed for PDI and gp96. For CRT and BiP, proteins in cell supernatants were immunoprecipitated (IP) with anti-CRT or anti-BiP prior to immuno-

blotting analyses. Approximately 8500 cells/lane contributed to the CM1 used to blot for PDI and gp96. The blots of corresponding lysates show ∼30,000cells’ worth of lysate. B, Comparison of PDI and gp96 release in response to low (200 nM) or high (5 mM) THP dose. C, Kinetics of PDI and gp96

secretion. Analyses were performed as in A, but immunoblotting analyses were undertaken at the indicated time points. D, Kinetics and magnitude of BiP

and CRT induction in response to TUN and THP. Indicated microgram amounts of total cell lysates were analyzed as in A at the indicated timevpoints

postdrug treatment. E, Flow cytometric assessments of ERp57 binding dependence of CRT surface expression. Cell surface CRT, measured as in Fig. 1,

before and after THP (5–7 h of 5 mM THP) treatment of CRT2/2 MEFs infected with retroviruses encoding WT CRT (WT), a CRT construct lacking the P-

domain (DP), a CRT point mutant that is unable to interact with ERp57 (W244A), or a virus lacking CRT (vec). Representative blots of three or four

experiments are shown in A. Data in B and C are based on one experiment Data in D are from one experiment, representative of two total experiments. Data

in E are based on three to five experiments. A–D, Vertical black lines in blots indicate places where lanes of the blot were rearranged to match the labeling

scheme used in adjacent immunoblots or to digitally remove lanes containing size standard (relevant bands from size standard are shown digitally for each

blot). The p values from two-tailed, paired t tests are indicated.

924 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

present in supernatants of THP-treated cells or residual THP itselfmight be responsible for the observed proinflammatory effectsof THP-treated cells. For these analyses, following each drugtreatment of 5–6.5 h, media containing drugs (CM1) were re-moved, and cells were washed three times with $7 ml PBS toremove soluble drug present in the media. Following these washsteps, fresh media were collected from the washed cells over a 5–6-h period to generate different MEF CM (CM2). We noted thatsecretion of PDI and gp96 was detectable for several hours fol-lowing THP removal from target cells, although at significantlyreduced levels compared with those detected during drug treat-ment (Supplemental Fig. 2A, CM1 compared with CM2). None-theless, THP CM2 was able to induce proinflammatory cytokineproduction by BMDCs, with patterns resembling those induced byTHP-treated cells (Figs. 5, 6A). Only trace amounts of cytokines,if any, were directly detectable in CM2 from various drug-treated,washed MEFs (Fig. 6A). BMDCs mixed with THP CM2, but notBMDCs coincubated with other CM2, produced IL-6 under sterileconditions (Fig. 6A). Additionally, LPS-dependent production ofIL-12p70, IL-23, and IL-1b strongly synergized with stimulationby THP CM2 to yield more IL-1b, IL-12p70, and IL-23 than otherBMDC treatments (Fig. 6A). However, BMDC cytokine pro-

duction in response to THP CM2 was clearly independent of CRTexpression in MEFs (Fig. 6A). The different drug-treated, washedcells used to generate CM2 were also harvested and stained withAnnV and 7AAD and analyzed by flow cytometry. These analysesindicated that, compared with other treatments, the THP CM2generation protocol did not enhance levels of apoptotic/secondarynecrotic cells (Supplemental Fig. 2B). Therefore, necrosis-relatedfactors were not a likely explanation for the immunostimulatoryproperties of THP-treated cell supernatants. Additionally, therewas no correlation between the level of secondary necrotic cells ina given treatment group and its ability to stimulate the BMDCs(Fig. 6A, Supplemental Fig. 2B).It is noteworthy that IL-12p70, IL-23, and IL-1b were not

significantly induced by THP CM2 or THP-treated cells understerile conditions (Figs. 5B–D, 6A, THP2 LPS bars). In contrast,IL-12p40 and IL-1b were both previously reported to be inducedunder sterile conditions by a membrane-linked form of gp96 (14).Furthermore, a recombinant, N-terminally truncated CRT fragmentpurified from Escherichia coli induced TNF-a production bymouse peritoneal macrophages and human PBMCs in the absenceof a TLR agonist (10). However, TNF-a was also not inducedby THP CM2 or THP-treated cells (WT or CRT-deficient) understerile conditions (Supplemental Fig. 2C, 2D). THP CM2 en-hanced TNF-a production mediated by LPS, but it was in-dependent of CRT expression by the target cells (SupplementalFig. 2C, 2D). Thus, despite containing various extracellular ERchaperones, THP-treated cell supernatants and cells induced onlythe production of IL-6 under sterile conditions and, therefore,were not as broadly immunostimulatory as predicted by the pres-ence of extracellular gp96 or CRT (10, 14).Previous studies showed that treatment of cells with TUN or

THP stimulated activation of NF-kB in HeLa cells (44) and pro-duction of reactive oxygen species in MEFs (45), as well as sig-nificantly enhanced production of type I IFN, TNF-a (46), ISG15,and IL-6 (41) by LPS-treated macrophages. Furthermore, THPand TUN enhanced IL-23, but not IL-12, production by humanmonocyte-derived DCs in response to a mixture of LPS and a TLR-8 agonist (CL-097) (43). We asked whether direct addition of TUNor THP to murine BMDCs could stimulate production of any of thecytokines measured in Figs. 5 and 6A. The levels of cytokinesresulting from drug treatment of BMDCs and CM treatment ofBMDCs were compared within the same experiment (Fig. 6B).THP and TUN treatment of BMDCs both stimulated production ofIL-6 in the absence of LPS, although THP stimulated, on average,∼9.8-fold more IL-6 than did TUN at the drug concentrations andtime points used in the analyses (Fig. 6B, 9.8-fold was the averageenhancement from five experiments, of which four are shown).THP and TUN treatment of BMDCs stimulated production ofIL-1b and IL-23 (Fig. 6B) in response to 1–10 ng/ml LPS. How-ever, under the conditions tested, THP was a stronger inducer of thecytokines, particularly IL-1b. As noted previously with humanmonocyte-derived DCs (43), lower concentrations of THP resultedin higher levels of IL-23 compared with higher concentrations ofTHP. In contrast to the previous studies with human myeloid DCs,THP (but not TUN) treatment also enhanced IL-12p70 productionin response to LPS (Fig. 6B). Importantly, the patterns of cytokineinduction by THP-treated cell supernatants were strikingly similarto those induced by direct THP treatment of BMDCs (Fig. 6B). Asnoted above, in generating THP CM2 for Fig. 6A, THP was re-moved from the MEFs, and MEFs were washed sufficiently to re-duce the extracellular, soluble drug concentration to ,15 pM, priorto collection of CM2. However, it is possible that THP was se-questered within cell membranes or intracellularly and subsequentlypartitioned into the media. In support of this possibility, we found

FIGURE 4. Release of ER chaperones from THP-treated cells involves

the secretory pathway. A and B, Immunoblotting analyses with anti-gp96

of supernatants (CM1) and lysates from THP-treated or untreated cells to

analyze the presence of glycosylated forms of gp96 in cell supernatants

following THP treatment. A, SDS-PAGE–based protein separation was

undertaken on an 8% polyacrylamide gel, and separation was optimized

compared with that shown in Fig. 3A, to allow for resolution of different

forms of gp96. The supernatant sample (CM1) was also run adjacent to the

lysate sample to better resolve alterations in electrophoretic mobility

resulting from posttranslational modifications. The vertical black line

denotes lanes that were cut and pasted from the same blot. B, Indicated cell

supernatant samples were digested with peptide:N-glycosidase F (+) or left

undigested (2) prior to the immunoblotting analyses. The blot is repre-

sentative of three experiments. C, BFA blockade of the secretory pathway

inhibits THP-induced surface CRT expression. Cells were pretreated with

BFA for 45 min, followed by the addition of THP for 6–7 h. Cells were

harvested, and surface CRT expression was measured in the AnnV2 and

7AAD2 population by flow cytometry, as described. Bar graph shows ratio

of surface CRT median fluorescence of BFA+THP-treated/untreated cells.

The data are the average of two experiments.

The Journal of Immunology 925

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

that when CM1 (cell supernatants from drug-treated cells) and CM2(cell supernatants from drug-treated, washed cells) were extensivelydialyzed in Centricons (3 or 10 KDa) to remove free soluble drugthat was present, the majority of cytokine-inducing activity waspresent in the retentates rather than in the flow through, even thoughTHP has a molecular mass of 650 Da (Supplemental Fig. 3). Basedon these findings, it is likely that the activity of THP-treated cellsupernatants (CM1 and CM2) (Fig. 6, Supplemental Fig. 3) and theresemblance of the stimulatory capacities of the supernatants todirect THP treatment of BMDCs (Fig. 6B) result from sequestra-tion of THP within cell membranes or intracellularly during drugtreatment of MEFs, as well as subsequent partitioning of active drugfrom MEFs into the CM within higher m.w. membrane vesicles orprotein complexes. This would allow THP to directly stimulateBMDCs. THP is hydrophobic and requires organic solvents (such asDMSO) for its solubilization; thus, its partitioning into membranesis perhaps not unexpected.If residual drug in THP-treated cells was responsible for the

BMDC-stimulating activity, we expected that reducing the con-centration of THP used to treat MEFs would reduce the immu-nostimulatory capacity of THP-treated cells. Thus, we comparedabilities of 5 mM or 200 nM THP CM1 and CM2 to induce pro-duction of various proinflammatory cytokines. Although treatingBMDCs with 5 mM THP CM1 or CM2 induced IL-6 production,treatment of BMDCs with 200 nM THP CM1 or CM2 did notinduce sterile IL-6 production (Fig. 6B). Furthermore, reducedlevels of LPS-induced IL-1b and IL-23 (Fig. 6B) were observedwhen 200 nM THP CM1 or CM2 was compared with 5 mM THPCM1 or CM2. Notably, levels of chaperone secretion were verysimilar in 200 nM and 5 mM THP CM1 (Fig. 3B).Together, these findings indicated that production of proinflam-

matory cytokines induced by THP-treated cells or CM was notenhanced by CRT expression by target cells (Figs. 5, 6, Supple-mental Fig. 2C, 2D). Notably, the profiles of proinflammatorycytokine induction by THP-treated cells and dialyzed CM closelyresembled those induced by THP itself, suggesting that residualTHP within protein/lipid complexes was responsible for enhancedcytokine production (Fig. 6B). Significantly, the effects of directBMDC treatment with THP and TUN were different, with directTHP treatment being more strongly proinflammatory.

CRT expression enhances phagocytic clearance of THP-treatedcells but not TUN- or UV-treated cells

Because cell surface CRT is an eat-me signal in the context ofapoptotic cells (5), preapoptotic tumor cells (7), and live tumor cellssubject to CD47 blockade (32), we asked whether THP-treated cellswould also be phagocytosed in a CRT-dependent manner. A pre-viously published flow cytometry-based assay (7) was used to mea-sure phagocytosis. WT or CRT-deficient MEFs were labeled witha fluorophore, treated with drugs or UV light, washed, and in-cubated with BMDCs. The cocultures were then fixed and stainedwith an Ab specific for the DC-specific marker CD-11c. The per-centage of CD-11c+ events that were also positive for the target cellmarker was recorded as a measure of target cell/DC association.The ratio of DC/target cell association in cocultures incubated at37˚C relative to that of cocultures incubated at 4˚C is reported asthe phagocytic index (the 37˚C readings quantify adhesion andphagocytosis, whereas the 4˚C readings quantify adhesion alone).

FIGURE 5. Coculture of THP-treated target cells with BMDCs induces

and enhances proinflammatory cytokine production in a CRT-independent

manner. Cytokine production by DCs, MEFs, and DC + MEF cocultures.

WTor CRT2/2 MEF target cells were treated with TUN, MTX, or THP for

5.5 or 6.5 h or left untreated. Adherent target cells were harvested, washed,

and incubated in media alone (MEF alone) or coincubated with BMDCs

(MEF + DC) in the absence (A) or presence (B–D) of 0.5–2 ng/ml LPS for

18.5–23.5 h. The concentrations of the indicated cytokine were measured

in duplicate by ELISA. Wells of DCs incubated in the presence (B–D) or

absence (A) of LPS were included in each experiment as an additional

control (DC alone). The THP(2)LPS bars indicate conditions in which

THP-treated WT MEFs or CRT2/2 MEFs were incubated with DCs, but in

the absence of LPS, to illustrate the LPS dependence of enhanced pro-

duction of indicated cytokines. The data from coincubations of DCs and

MEFs show the mean and SE of three (WT) or five (CRT2/2) experiments.

The data from MEFs alone or DCs alone show the mean and SE of one to

three experiments in the presence of LPS or two to four experiments in the

absence of LPS. The p values from two-tailed, paired t tests are indicated.

926 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

Uptake of UV-treated MEFs was significantly greater than untreatedMEFs, but it was independent of CRT expression by the MEFs(Fig. 7A). The latter finding was surprising in light of a previouslysuggested role for CRT in the phagocytic uptake of apoptotic UV-treated fibroblasts using fibroblasts or a macrophage cell line as thephagocyte (5). Therefore, surface CRT expression was tested in theUV-treated fibroblasts, which were triple stained for CRT, as well asmarkers of apoptosis and plasma membrane integrity, as definedin the analyses of Fig. 1. Our analyses indicated that CRT is notpresent or significantly induced on the surface of AnnV+PI2 (Fig.7B) or AnnV2PI2 (data not shown) MEFs in response to UV

treatment, consistent with the absence of significant, CRT-depen-dent uptake of UV-treated cells. Several differences in protocolmay explain these discrepancies. These include the use of a differ-ent phagocyte [BMDCs (Fig. 7) rather than macrophages or fibro-blast (5)], a different phagocytosis assay [flow cytometry (Fig. 7)rather than microcopy (5)], and the use of both adherent and non-adherent UV-treated target cells (Fig. 7), rather than nonadherentUV-treated target cells (5).In contrast to UV-treated MEFs, THP-treated WT MEFs were

phagocytosed with significantly higher efficiency than were THP-treated CRT-deficient MEFs and untreated MEFs (Fig. 7A). The

FIGURE 6. Differential induction of proinflammatory cytokines by drugs and by CM from drug-treated MEFs. BMDCs were treated with MEF CM or

drugs either under sterile conditions (top panels for IL-6 measurements) or in the presence of LPS [lower panels for IL-1b, IL-23 and IL-12p70 meas-

urements; 1 ng/ml LPS (A) or 1–10 ng/ml LPS (B) were used] for 21–23 (A) or 15–22 (B) hours. Indicated cytokine measurements followed CM + DC, drug +

DC, or DC-media (DC alone) coincubations in the presence or absence of LPS. A, Concentrations of the indicated cytokines in MEF CM2 + DC coincu-

bations or DC alone were measured in duplicate by ELISA. The THP(–)LPS bars indicate conditions in which CM2 from THP-treated WTor CRT2/2 MEFs

were incubated with DCs. CRT2/2 MEF CM2 and WT MEF CM2 bars indicate direct cytokine measurements of the indicated CM2 samples without further

treatment with LPS. The results with WT MEFs show the average of three independent experiments, and the results with CRT2/2 MEFs are from a single

experiment. B, THP 200 nM CM represents CM prepared as for other CM, but by incubating MEFs with 200 nM THP instead of 5 mM THP. Direct DC

treatment represents cytokine production by BMDCs that were directly treated with THP (black bars) or TUN (vertically striped bars) at their specified

concentrations or by BMDCs that were left untreated (horizontally striped bars). The (2)LPS bars represent measurements from DCs treated with THP CM1

or 5 mM THP, as indicated, but in the absence of LPS. Data are averages of four experiments (for direct treatment of BMDCs with drugs) or eight experiments

(for CM2). Data for CMI are from a single experiment.

The Journal of Immunology 927

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

levels of target cell death were similar between WT and CRT-deficient MEF target cells within a treatment group (Fig. 7C).Additionally, the level of dead target cells in the untreated groupwas least as high as that in the THP-treated group (Fig. 7C), aresult that indicated that cell death in the THP-treated group wasinsufficient to account for the level of phagocytic uptake observed.Finally, although TUN treatment enhanced phagocytic uptakeof target cells relative to untreated cells, there was no CRT de-pendence to this effect, consistent with the inability to observe cellsurface CRT in TUN-treated cells (Figs. 1B, 2B). The fluorescenceintensities of the various labeled target cells were similar andcould not explain the observed differences in the phagocytic in-dices (Fig. 7D). Taken together, these results suggested that THP-treated cells with measurable cell surface CRT are phagocytosedin a manner that is dependent, in part, on CRT.

DiscussionCell surface CRT has been characterized as an important factorin inducing antitumor immune responses (7, 8, 26, 47). Anticancerregimens that induce an antitumor immune response are likely tobe more effective than are nonimmunogenic regimens with similarcancer cell cytotoxicity (4). Identifying conditions that induce cellsurface CRT, as well as understanding mechanisms and functionalconsequences of cell surface CRT expression, could help in thedesign of more effective antitumor therapies. Our findings in-

dicated that THP induces transient, preapoptotic, cell surface CRTexposure in some cell types (Figs. 1, 2); that THP treatment oftarget cells increases proinflammatory cytokine production byBMDCs in a manner independent of target cell-derived CRT (Fig.5); that extracellular CRT, either in a cell surface-bound (Fig. 5,Supplemental Fig. 2D) or soluble form (Fig. 6, Supplemental Fig.2C), is unable to stimulate production of IL-6, IL-1b, TNF-a, IL-23, or IL-12p70; and that THP treatment of cells impacts theirphagocytic uptake in a CRT-dependent manner (Fig. 7).Twomechanisms were previously described for cell surface CRT

exposure (8, 40). The first pathway is dependent on the pancreaticER kinase arm of the UPR and results in the specific cotranslo-cation of CRT–ERp57 complexes to the cell surface (8). Thesecond suggested pathway, relevant to apoptotic cells, involves thebinding of cytosolic CRT to phosphatidylserine on the inner leafletof the plasma membrane prior to apoptotic exposure of phos-phatidylserine–CRT complexes on the cell surface (40). Neither ofthese pathways seems to be related to the mechanisms of surfaceexposure of CRT in viable cells subjected to ER calcium de-pletion, which display a general loss of ER retention of various ERchaperones. The THP-induced mode of CRT surface expressionis independent of CRT–ERp57 binding (Fig. 3E). Rather, we re-cently showed that induction of a generic polypeptide binding siteon CRT facilitates its cell surface expression under calcium-depleting conditions (21).

FIGURE 7. CRT enhances phagocytic uptake of THP-treated target cells but not of UV- or TUN-treated target cells. Cell Tracker Green (CMFDA)

labeling of MEFs, cell death profiles, phagocytic uptake, and surface CRT expression following different treatments. WT or CRT2/2 MEFs were labeled

with Cell Tracker Green, exposed to UV light for 3 min, and then cultured for an additional 16–24 h, exposed to THP or TUN for 5–6 h, or left untreated. A,

Phagocytic uptake: labeled and treated cells were washed three times, mixed with BMDCs at an E:T ratio of 1:5, and incubated for 1 h at 37 or 4˚C.

Cocultures were fixed, stained with anti-CD11c, and analyzed by flow cytometry. For each cell type and treatment condition, the percentages of CMFDA+

events in the CD11c+ gate were measured for the 4 and 37˚C plates; the phagocytic index is plotted as the 37˚C/4˚C ratio of those percentages. The

phagocytic index data are averages of eight experiments for all conditions, with the exception of TUN, for which the average of four experiments is shown.

On average, 3–6% of the total added target cells were phagocytosed. B, Surface CRT expression was measured in UV-treated WT or CRT2/2 MEFs, as

described in Fig. 1; data are representative of three experiments. C, Cell death profiles: CMFDA-labeled MEFs subjected to the indicated treatments were

stained with AnnV-PE and 7AAD to quantify the percentages of MEFs in the indicated populations. Data represent the averages of six experiments for all

conditions, with the exception of TUN treatments, for which the average of three experiments is shown. AnnV/7AAD data were not collected in two of

eight experiments. D, CMFDA fluorescence of various treated MEFs were measured on the FITC channel. Averages of eight total experiments are shown

for all conditions, with the exception of TUN treatment, for which the average of four experiments is shown. The p values from two-tailed, paired t tests are

indicated.

928 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

Our data indicated a secretory route for ER proteins (Fig. 4), asalso suggested for the CRT–ERp57 cotranslocation mechanism(8). It is noteworthy that several of the ER-resident proteinsdetectable in supernatants of THP-treated cells (Fig. 3) are knowncalcium-binding proteins of the ER (48). As is the case with CRT(49), ER retention of BiP, PDI, and gp96 may, in part, involvecalcium-dependent processes. However, calcium-depletion aloneseems to be insufficient to induce secretion of the ER chaperones(Fig. 3C, 3D). Rather, the combined effects of KDEL receptorsaturation and calcium-depletion seem to be the mechanism thatdrives secretion of the ER chaperones in THP-treated cells.Similar mechanisms may be relevant to the extracellular locali-zation of ER chaperones under physiological protein-misfoldingconditions known to induce UPR and alter ER calcium homeo-stasis. To investigate this possibility, we compared chaperonesecretion in WT and cog/cog mice thyroids, which expressa misfolded, mutant thyroglobulin that accumulates in the ER(50). In preliminary findings, we observed enhanced secretion ofCRT, PDI, and gp96 into the apical follicular lumen of cog/cogmice thyroids compared with their WT counterparts, correlatingwith significant enhancement of total chaperone levels in thecog/cog thyroid lysates (E. Jeffery, A.P. Kellogg, P. Arvan, andM. Raghavan, unpublished observations).Previous studies in HeLa cells showed that NF-kB activation in

response to agents that induce ER stress involves the release ofcalcium from the ER and the subsequent generation of reactiveoxygen species (reviewed in Ref. 51). This mechanism likelyaccounts for sterile IL-6 production in response to direct THPtreatment of BMDCs (Fig. 6B). THP is a stronger activator of thesterile IL-6 response compared with TUN (Fig. 6B), which likelyresults from the rapid effect of THP on cytosolic calcium eleva-tion, whereas elevation of cytosolic calcium in response to TUNtreatment may be more modest, as well as kinetically delayed.Many recent studies showed synergies between TLR-derivedsignals and transcription factors induced by UPR, includingXBP-1 and C/EBP homologous protein. Of the cytokines mea-sured in Figs. 5 and 6, XBP-1 was shown to synergize with TLRsignals in enhancing IL-6 production by murine macrophages(41). In human monocyte-derived DCs, C/EBP homologous pro-tein binding to the IL-23p19 promoter enhanced IL-23 productionin response to TLR ligation (43). Similar mechanisms could ac-count for the synergy of ER stress signals and LPS in IL-23production by BMDCs (Fig. 6B). We found that THP alsostrongly synergized with LPS in the generation of IL-1b andIL-12p70. Recent findings showed that high-avidity ligation ofITAM-containing receptors, such as DAP12, can synergize withTLR signals to activate proinflammatory gene expression (reviewedin Refs. 52, 53). High-avidity ligation of DAP12 triggers anacute, transient calcium increase (reviewed in Ref. 52). In thisarticle, we showed that, compared with TUN, THP treatmentstrongly enhanced various proinflammatory responses to LPS(Fig. 6B), consistent with similar findings from previous studies(45, 46). These results point toward direct synergies betweencytosolic calcium and TLR responses, as well as broader synergybetween intracellular calcium and TLR signaling compared withthose between the UPR and TLR signaling. Interestingly, it wassuggested that ER calcium depletion can occur under certainconditions of ER overload and protein polymerization, even in theabsence of classical UPR induction (54). Thus, the findings de-scribed in this article may be relevant to a better understanding ofinflammatory protein-folding disorders that do not induce a clas-sical UPR pathway.A truncated bacterially expressed CRT construct was shown to

induce TNF-a production by murine macrophages and human

PBMCs (10). N-terminal truncation of CRT induced self-associa-tion, likely via the exposure of hydrophobic surfaces (10, 29), andit is possible that protein truncation induces enhanced binding tobacterial pathogen-associated molecular patterns. We showed inthis study that full-length extracellular CRT derived from an en-dogenous source does not enhance the production of various proin-flammatory cytokines in a THP-treated BMDC context (Figs. 5, 6,Supplemental Fig. 2C, 2D). These findings are consistent with aprevious report that a secreted form of calreticulin does not induceproinflammatory cytokine production by BMDC (55). CRT isknown to be present in the extracellular environment under a numberof physiological conditions, and a number of extracellular functionsare described for CRT (reviewed in Ref. 3). Given these extracellularroles of the protein, the lack of innate immune induction by en-dogenously derived, extracellular CRT is perhaps not surprising.CRT-high, THP-treated, target cells were phagocytosed more

efficiently than were untreated cells, and enhanced phagocytosiswas dependent, in part, on CRT expression by the target cells (Fig.7). Correlating with undetectable levels of surface CRT in UV-treated cells, we did not observe an effect of CRT on the phago-cytosis of UV-treated cells (Fig. 7). In the context of UV-treatedapoptotic cells, other changes are expected that promote phago-cytic uptake (reviewed in Ref. 56), including the exposure ofphosphatidylserine, which likely explain why significant uptakewas observed in the absence of any CRT expression by the targetcells (Fig. 7). Correspondingly, UV-treated target cells had ahigher level of total association (adhesion and phagocytosis, in-dicated by the 37˚C measurements) with BMDCs than did anyother target cell group (Supplemental Fig. 4). This is likely theresult of molecules that are upregulated on the surface of apoptoticcells and responsible for bridging apoptotic phagocytic cargo topotential phagocytes (reviewed in Ref. 56). It is important toconsider that increases in surface calreticulin by a magnitudemeasurable by flow cytometry were, in general, difficult to detectin apoptotic cells (Figs. 1, 2). Stronger nonspecific protein bindingto apoptotic cells could render a specific signal more difficultto detect by flow cytometry. Microscopic observations could allowfor calreticulin redistribution or upregulation to be more readilyvisualized on apoptotic cells, as described in other studies (5, 9,57). It is also important to note that cell surface CRT functions inthe context of several other key eat-me and don’t-eat-me signalsin viable and apoptotic cells (5, 56, 57). The data in Fig. 7 showthat TUN-treated cells displayed enhanced phagocytic uptakecompared with untreated cells, although this uptake was not CRTdependent. This observation correlates with the absence of de-tectable surface CRT in TUN-treated cells (Fig. 1B). The samelevel of uptake was seen with THP-treated, CRT-deficient cells(Fig. 7A). It is possible that interference with proper proteinfolding resulting from TUN and THP treatments also caused de-creased surface expression of don’t-eat-me signals, such as CD47(5, 32) and plasminogen activator inhibitor-1 (57). This potentialdecreased surface expression of don’t-eat-me signals may explainthe increased level of uptake of TUN-treated and THP-treated,CRT-deficient cells by BMDCs.In contrast to some previous findings (7), we showed in this

article that THP induces cell surface CRT and that THP promotesphagocytic uptake of cells via mechanisms that are, in part, CRTdependent. Thus, the inability of THP-treated cells to induce an-titumor immunity with sufficient potency (7, 47) is likely not re-lated to an absence of surface CRT per se. Other immunogenicsignals absent in THP-treated cells (Supplemental Fig. 2E, 2F),including HMGB1 (58) and elevated levels of ATP (59), may berequired to confer stronger immunogenicity to tumor cells. Con-sequently, although THP has some potentially desirable attributes

The Journal of Immunology 929

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

for a chemotherapeutic, including the abilities to induce strongerphagocytic uptake of treated cells by DCs (Fig. 7) and elicitstronger cytokine production by DCs (Figs. 5, 6), these seem to beinsufficient signals to confer antitumor immune protection. Theseobservations are highly relevant to the ongoing clinical trialsevaluating THP prodrugs for the treatment of advanced solidtumors [ClinicalTrials.gov Identifier: NCT01056029 (60)]. Be-cause phagocytic uptake and cytokine levels can influence T cellpriming steps, our results highlight the importance of further re-search investigating the impact of different forms of ER stress onthe priming of CD8 and CD4 T cell responses.

AcknowledgmentsWe thank the transgenic mouse core for providing mouse bone marrow for

DC preparations; Joanne Sonstein, Natasha Del Cid, and Dr. Jeff Curtis for

assistance with the development of phagocytosis assays; Dr. Yasmina

Laouar for advice on protocols for BMDC preparations and flow cytometry;

Joel Whitfield for performing the ELISA assays; Elise Jeffery for assistance

with the project; and Drs. Malathi Kandarpa and Andrzej Jakubowiak for

the indicated cell lines.

DisclosuresThe authors have no financial conflicts of interest.

References1. Gorlach, A., P. Klappa, and T. Kietzmann. 2006. The endoplasmic reticulum:

folding, calcium homeostasis, signaling, and redox control. Antioxid. RedoxSignal. 8: 1391–1418.

2. Michalak, M., J. Groenendyk, E. Szabo, L. I. Gold, and M. Opas. 2009. Cal-reticulin, a multi-process calcium-buffering chaperone of the endoplasmic re-ticulum. Biochem. J. 417: 651–666.

3. Gold, L. I., P. Eggleton, M. T. Sweetwyne, L. B. Van Duyn, M. R. Greives,S. M. Naylor, M. Michalak, and J. E. Murphy-Ullrich. 2010. Calreticulin: non-endoplasmic reticulum functions in physiology and disease. FASEB J. 24: 665–683.

4. Green, D. R., T. Ferguson, L. Zitvogel, and G. Kroemer. 2009. Immunogenic andtolerogenic cell death. Nat. Rev. Immunol. 9: 353–363.

5. Gardai, S. J., K. A. McPhillips, S. C. Frasch, W. J. Janssen, A. Starefeldt,J. E. Murphy-Ullrich, D. L. Bratton, P. A. Oldenborg, M. Michalak, andP. M. Henson. 2005. Cell-surface calreticulin initiates clearance of viable orapoptotic cells through trans-activation of LRP on the phagocyte. Cell 123: 321–334.

6. Obeid, M., T. Panaretakis, N. Joza, R. Tufi, A. Tesniere, P. van Endert,L. Zitvogel, and G. Kroemer. 2007. Calreticulin exposure is required for theimmunogenicity of gamma-irradiation and UVC light-induced apoptosis. CellDeath Differ. 14: 1848–1850.

7. Obeid, M., A. Tesniere, F. Ghiringhelli, G. M. Fimia, L. Apetoh, J. L. Perfettini,M. Castedo, G. Mignot, T. Panaretakis, N. Casares, et al. 2007. Calreticulinexposure dictates the immunogenicity of cancer cell death. Nat. Med. 13: 54–61.

8. Panaretakis, T., O. Kepp, U. Brockmeier, A. Tesniere, A. C. Bjorklund,D. C. Chapman, M. Durchschlag, N. Joza, G. Pierron, P. van Endert, et al. 2009.Mechanisms of pre-apoptotic calreticulin exposure in immunogenic cell death.EMBO J. 28: 578–590.

9. Kuraishi, T., J. Manaka, M. Kono, H. Ishii, N. Yamamoto, K. Koizumi,A. Shiratsuchi, B. L. Lee, H. Higashida, and Y. Nakanishi. 2007. Identification ofcalreticulin as a marker for phagocytosis of apoptotic cells in Drosophila. Exp.Cell Res. 313: 500–510.

10. Hong, C., X. Qiu, Y. Li, Q. Huang, Z. Zhong, Y. Zhang, X. Liu, L. Sun, P. Lv,and X. M. Gao. 2010. Functional analysis of recombinant calreticulin fragment39-272: implications for immunobiological activities of calreticulin in health anddisease. J. Immunol. 185: 4561–4569.

11. Henderson, B., S. K. Calderwood, A. R. Coates, I. Cohen, W. van Eden,T. Lehner, and A. G. Pockley. 2010. Caught with their PAMPs down? The ex-tracellular signalling actions of molecular chaperones are not due to microbialcontaminants. Cell Stress Chaperones 15: 123–141.

12. Spisek, R., A. Charalambous, A. Mazumder, D. H. Vesole, S. Jagannath, andM. V. Dhodapkar. 2007. Bortezomib enhances dendritic cell (DC)-mediatedinduction of immunity to human myeloma via exposure of cell surface heatshock protein 90 on dying tumor cells: therapeutic implications. Blood 109:4839–4845.

13. Tsan, M. F., and B. Gao. 2009. Heat shock proteins and immune system. J.Leukoc. Biol. 85: 905–910.

14. Liu, B., J. Dai, H. Zheng, D. Stoilova, S. Sun, and Z. Li. 2003. Cell surfaceexpression of an endoplasmic reticulum resident heat shock protein gp96 triggersMyD88-dependent systemic autoimmune diseases. Proc. Natl. Acad. Sci. USA100: 15824–15829.

15. Drummond, I. A., A. S. Lee, E. Resendez, Jr., and R. A. Steinhardt. 1987. De-pletion of intracellular calcium stores by calcium ionophore A23187 induces thegenes for glucose-regulated proteins in hamster fibroblasts. J. Biol. Chem. 262:12801–12805.

16. Booth, C., and G. L. Koch. 1989. Perturbation of cellular calcium induces se-cretion of luminal ER proteins. Cell 59: 729–737.

17. Thastrup, O., P. J. Cullen, B. K. Drøbak, M. R. Hanley, and A. P. Dawson. 1990.Thapsigargin, a tumor promoter, discharges intracellular Ca2+ stores by specificinhibition of the endoplasmic reticulum Ca2(+)-ATPase. Proc. Natl. Acad. Sci.USA 87: 2466–2470.

18. Tufi, R., T. Panaretakis, K. Bianchi, A. Criollo, B. Fazi, F. Di Sano, A. Tesniere,O. Kepp, P. Paterlini-Brechot, L. Zitvogel, et al. 2008. Reduction of endoplasmicreticulum Ca2+ levels favors plasma membrane surface exposure of calreticulin.Cell Death Differ. 15: 274–282.

19. Delpino, A., and M. Castelli. 2002. The 78 kDa glucose-regulated protein(GRP78/BIP) is expressed on the cell membrane, is released into cell culturemedium and is also present in human peripheral circulation. Biosci. Rep. 22:407–420.

20. Zhang, Y., R. Liu, M. Ni, P. Gill, and A. S. Lee. 2010. Cell surface relocalizationof the endoplasmic reticulum chaperone and unfolded protein response regulatorGRP78/BiP. J. Biol. Chem. 285: 15065–15075.

21. Jeffery, E., L. R. Peters, and M. Raghavan. 2011. The polypeptide bindingconformation of calreticulin facilitates its cell-surface expression under con-ditions of endoplasmic reticulum stress. J. Biol. Chem. 286: 2402–2415.

22. Stevens, F. J., and Y. Argon. 1999. Protein folding in the ER. Semin. Cell Dev.Biol. 10: 443–454.

23. Sambrook, J. F. 1990. The involvement of calcium in transport of secretoryproteins from the endoplasmic reticulum. Cell 61: 197–199.

24. Zhang, K., and R. J. Kaufman. 2008. From endoplasmic-reticulum stress to theinflammatory response. Nature 454: 455–462.

25. Rutkowski, D. T., and R. J. Kaufman. 2004. A trip to the ER: coping with stress.Trends Cell Biol. 14: 20–28.

26. Panaretakis, T., N. Joza, N. Modjtahedi, A. Tesniere, I. Vitale, M. Durchschlag,G. M. Fimia, O. Kepp, M. Piacentini, K. U. Froehlich, et al. 2008. The co-translocation of ERp57 and calreticulin determines the immunogenicity of celldeath. Cell Death Differ. 15: 1499–1509.

27. Obeid, M. 2008. ERP57 membrane translocation dictates the immunogenicity oftumor cell death by controlling the membrane translocation of calreticulin. J.Immunol. 181: 2533–2543.

28. Bost, K. L., and M. J. Mason. 1995. Thapsigargin and cyclopiazonic acid initiaterapid and dramatic increases of IL-6 mRNA expression and IL-6 secretion inmurine peritoneal macrophages. J. Immunol. 155: 285–296.

29. Del Cid, N., E. Jeffery, S. M. Rizvi, E. Stamper, L. R. Peters, W. C. Brown,C. Provoda, and M. Raghavan. 2010. Modes of calreticulin recruitment to themajor histocompatibility complex class I assembly pathway. J. Biol. Chem. 285:4520–4535.

30. Torchinsky, M. B., J. Garaude, A. P. Martin, and J. M. Blander. 2009. Innateimmune recognition of infected apoptotic cells directs T(H)17 cell differentia-tion. Nature 458: 78–82.

31. Fadok, V. A., D. R. Voelker, P. A. Campbell, J. J. Cohen, D. L. Bratton, andP. M. Henson. 1992. Exposure of phosphatidylserine on the surface ofapoptotic lymphocytes triggers specific recognition and removal by macro-phages. J. Immunol. 148: 2207–2216.

32. Chao, M. P., S. Jaiswal, R. Weissman-Tsukamoto, A. A. Alizadeh, A. J. Gentles,J. Volkmer, K. Weiskopf, S. B. Willingham, T. Raveh, C. Y. Park, R. Majeti, andI. L. Weissman. 2010. Calreticulin is the dominant pro-phagocytic signal onmultiple human cancers and is counterbalanced by CD47. Sci Transl. Med. 2:63ra94.

33. Okada, T., K. Haze, S. Nadanaka, H. Yoshida, N. G. Seidah, Y. Hirano, R. Sato,M. Negishi, and K. Mori. 2003. A serine protease inhibitor prevents endoplasmicreticulum stress-induced cleavage but not transport of the membrane-boundtranscription factor ATF6. J. Biol. Chem. 278: 31024–31032.

34. Winnay, J. N., J. Boucher, M. A. Mori, K. Ueki, and C. R. Kahn. 2010. Aregulatory subunit of phosphoinositide 3-kinase increases the nuclear accumu-lation of X-box-binding protein-1 to modulate the unfolded protein response.Nat. Med. 16: 438–445.

35. Llewellyn, D. H., J. M. Kendall, F. N. Sheikh, and A. K. Campbell. 1996. In-duction of calreticulin expression in HeLa cells by depletion of the endoplasmicreticulum Ca2+ store and inhibition of N-linked glycosylation. Biochem. J. 318:555–560.

36. Waser, M., N. Mesaeli, C. Spencer, and M. Michalak. 1997. Regulation ofcalreticulin gene expression by calcium. J. Cell Biol. 138: 547–557.

37. Frickel, E. M., R. Riek, I. Jelesarov, A. Helenius, K. Wuthrich, and L. Ellgaard.2002. TROSY-NMR reveals interaction between ERp57 and the tip of the cal-reticulin P-domain. Proc. Natl. Acad. Sci. USA 99: 1954–1959.

38. Martin, V., J. Groenendyk, S. S. Steiner, L. Guo, M. Dabrowska, J. M. Parker,W. Muller-Esterl, M. Opas, and M. Michalak. 2006. Identification by mutationalanalysis of amino acid residues essential in the chaperone function of calreti-culin. J. Biol. Chem. 281: 2338–2346.

39. Rizvi, S. M., L. Mancino, V. Thammavongsa, R. L. Cantley, and M. Raghavan.2004. A polypeptide binding conformation of calreticulin is induced by heatshock, calcium depletion, or by deletion of the C-terminal acidic region. Mol.Cell 15: 913–923.

40. Tarr, J. M., P. J. Young, R. Morse, D. J. Shaw, R. Haigh, P. G. Petrov,S. J. Johnson, P. G. Winyard, and P. Eggleton. 2010. A mechanism of release ofcalreticulin from cells during apoptosis. J. Mol. Biol. 401: 799–812.

930 ER CALCIUM IMPACTS CHAPERONE SECRETION AND IMMUNITY

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

41. Martinon, F., X. Chen, A. H. Lee, and L. H. Glimcher. 2010. TLR activation ofthe transcription factor XBP1 regulates innate immune responses in macro-phages. Nat. Immunol. 11: 411–418.

42. Voll, R. E., M. Herrmann, E. A. Roth, C. Stach, J. R. Kalden, and I. Girkontaite.1997. Immunosuppressive effects of apoptotic cells. Nature 390: 350–351.

43. Goodall, J. C., C. Wu, Y. Zhang, L. McNeill, L. Ellis, V. Saudek, andJ. S. Gaston. 2010. Endoplasmic reticulum stress-induced transcription factor,CHOP, is crucial for dendritic cell IL-23 expression. Proc. Natl. Acad. Sci. USA107: 17698–17703.

44. Pahl, H. L., and P. A. Baeuerle. 1995. A novel signal transduction pathway fromthe endoplasmic reticulum to the nucleus is mediated by transcription factor NF-kappa B. EMBO J. 14: 2580–2588.

45. Hsieh, Y. H., I. J. Su, H. Y. Lei, M. D. Lai, W. W. Chang, and W. Huang. 2007.Differential endoplasmic reticulum stress signaling pathways mediated by iNOS.Biochem. Biophys. Res. Commun. 359: 643–648.

46. Smith, J. A., M. J. Turner, M. L. DeLay, E. I. Klenk, D. P. Sowders, andR. A. Colbert. 2008. Endoplasmic reticulum stress and the unfolded proteinresponse are linked to synergistic IFN-beta induction via X-box binding protein1. Eur. J. Immunol. 38: 1194–1203.

47. Martins, I., O. Kepp, F. Schlemmer, S. Adjemian, M. Tailler, S. Shen,M. Michaud, L. Menger, A. Gdoura, N. Tajeddine, et al. 2011. Restoration of theimmunogenicity of cisplatin-induced cancer cell death by endoplasmic reticulumstress. Oncogene 30: 1147–1158.

48. Coe, H., and M. Michalak. 2009. Calcium binding chaperones of the endo-plasmic reticulum. Gen. Physiol. Biophys. 28 Spec No Focus: F96–F103.

49. Sonnichsen, B., J. Fullekrug, P. Nguyen Van, W. Diekmann, D. G. Robinson, andG. Mieskes. 1994. Retention and retrieval: both mechanisms cooperate tomaintain calreticulin in the endoplasmic reticulum. J. Cell Sci. 107: 2705–2717.

50. Kim, P. S., S. A. Hossain, Y. N. Park, I. Lee, S. E. Yoo, and P. Arvan. 1998. Asingle amino acid change in the acetylcholinesterase-like domain of thyro-globulin causes congenital goiter with hypothyroidism in the cog/cog mouse:

a model of human endoplasmic reticulum storage diseases. Proc. Natl. Acad. Sci.USA 95: 9909–9913.

51. Pahl, H. L., and P. A. Baeuerle. 1997. The ER-overload response: activation ofNF-kappa B. Trends Biochem. Sci. 22: 63–67.

52. Ivashkiv, L. B. 2008. A signal-switch hypothesis for cross-regulation of cytokineand TLR signalling pathways. Nat. Rev. Immunol. 8: 816–822.

53. Turnbull, I. R., and M. Colonna. 2007. Activating and inhibitory functions ofDAP12. Nat. Rev. Immunol. 7: 155–161.

54. Davies, M. J., E. Miranda, B. D. Roussel, R. J. Kaufman, S. J. Marciniak, andD. A. Lomas. 2009. Neuroserpin polymers activate NF-kappaB by a calciumsignaling pathway that is independent of the unfolded protein response. J. Biol.Chem. 284: 18202–18209.

55. Bak, S. P., E. Amiel, J. J. Walters, and B. Berwin. 2008. Calreticulin requires anancillary adjuvant for the induction of efficient cytotoxic T cell responses. Mol.Immunol. 45: 1414–1423.

56. Krysko, D. V., and P. Vandenabeele. 2008. From regulation of dying cell en-gulfment to development of anti-cancer therapy. Cell Death Differ. 15: 29–38.

57. Park, Y. J., G. Liu, E. F. Lorne, X. Zhao, J. Wang, Y. Tsuruta, J. Zmijewski, andE. Abraham. 2008. PAI-1 inhibits neutrophil efferocytosis. Proc. Natl. Acad. Sci.USA 105: 11784–11789.

58. Apetoh, L., F. Ghiringhelli, A. Tesniere, M. Obeid, C. Ortiz, A. Criollo,G. Mignot, M. C. Maiuri, E. Ullrich, P. Saulnier, et al. 2007. Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy andradiotherapy. Nat. Med. 13: 1050–1059.

59. Ghiringhelli, F., L. Apetoh, A. Tesniere, L. Aymeric, Y. Ma, C. Ortiz,K. Vermaelen, T. Panaretakis, G. Mignot, E. Ullrich, et al. 2009. Activation ofthe NLRP3 inflammasome in dendritic cells induces IL-1beta-dependent adap-tive immunity against tumors. Nat. Med. 15: 1170–1178.

60. Janssen, S., D. M. Rosen, R. M. Ricklis, C. A. Dionne, H. Lilja,S. B. Christensen, J. T. Isaacs, and S. R. Denmeade. 2006. Pharmacokinetics,biodistribution, and antitumor efficacy of a human glandular kallikrein 2 (hK2)-activated thapsigargin prodrug. Prostate 66: 358–368.

The Journal of Immunology 931

by guest on April 12, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from