Sepsis Mortality in the Innate Immune Response to OX40 Ligand

Published September 15, 2010, doi:10.4049/jimmunol.1000404 The Journal of Immunology

OX40 Ligand Regulates Inﬂammation and Mortality in the Innate Immune Response to Sepsis

Matthew Karulf,* Ann Kelly,* Andrew D. Weinberg,† and Jeffrey A. Gold*

The initial phase of sepsis is characterized by massive inflammatory cytokine production that contributes to multisystem organ failure and death. Costimulatory molecules are a class of receptors capable of regulating cytokine production in adaptive immunity. Recent studies described their presence on neutrophils and monocytes, suggesting a potential role in the regulation of cytokine production in innate immunity. The purpose of this study was to determine the role for OX40–OX40 ligand (OX40L) interaction in the innate immune response to polymicrobial sepsis. Humans with sepsis demonstrated upregulation of OX40L on monocytes and neutrophils, with mortality and intensive care unit stay correlating with expression levels. In an animal model of polymicrobial sepsis, a direct role for OX40L in regulating inflammation was indicated by improved survival, decreased cytokine production, and a decrease in remote organ damage in OX40L2/2 mice. The finding of similar results with an OX40L Ab suggests a potential therapeutic role for OX40L blockade in sepsis. The inability of anti-OX40L to provide significant protection in macrophage-depleted mice establishes macrophages as an indispensable cell type within the OX40/OX40L axis that helps to mediate the clinical signs of disease in sepsis. Conversely, the protective effect of anti-OX40L Ab in RAG12/2 mice further confirms a T cell-independent role for OX40L stimulation in sepsis. In conclusion, our data provide an in vivo role for the OX40/OX40L system in the innate immune response during polymicrobial sepsis and suggests a potential beneficial role for therapeutic blockade of OX40L in this devastating disorder. The Journal of Immunology, 2010, 185: 000–000.

epsis, defined as the systemic inflammatory response to Recent studies attribute a potentially important role to costimu- infection, is a devastating condition with high morbidity and latory molecules in regulating cytokine production and mortality in S mortality. It remains a significant burden on the health care animal models of polymicrobial sepsis (7, 8). The OX40 (CD134)/ system, affecting .700,000 people/y in the United States, resulting OX40 ligand (OX40L; CD252) system has been well described in in 250,000 deaths (1, 2). This results in a net cost of $17 billion/y. the regulation of T cell proliferation, effector cell function, and Over the past decade, mortality from sepsis has remained .25%, survival within the adaptive immune response (9). Growing evi- despite adequate antimicrobial therapy (3). This highlights the need dence suggests that OX40L is present on activated macrophages and for newer adjuvant therapies. mononuclear cells, and OX40L is capable of directly activating The early phases of sepsis are characterized by activation of the intracellular signaling cascades and cytokine production (10, 11). innate immune response, with production of multiple pro- and anti- Furthermore, OX40 has been described on activated neutrophils inflammatory cytokines. However, numerous attempts have been (polymorphonuclear leukocytes [PMNs]), as well as existing as made to inhibit individual cytokines in sepsis, including TNF-a a shed soluble form; the latter is also capable of binding to, and and IL-1b. These strategies, although effective in limiting LPS- activating, OX40L, with subsequent upregulation of IL-6 on air- mediated inflammation, failed to improve mortality in phase III ways smooth muscle and c-fos and c-jun in transfected HUVECs clinical trials or in a mouse model of polymicrobial sepsis, such as (12, 13). Together, these data suggest that OX40/OX40L signaling cecal ligation and puncture (CLP) (4–6). One explanation for their within innate immune cells could play a role in the innate immune failure is the redundant and overlapping actions of the individual response to sepsis. In this report, we describe a pivotal role for cytokines. Consequently, investigators have begun to focus on cell- OX40/OX40L in the regulation of mortality and inflammation in the surface receptors capable of simultaneously regulating multiple innate immune response to mouse polymicrobial sepsis. inflammatory cascades. Materials and Methods Mice

*Division of Pulmonary and Critical Care Medicine, Oregon Health and Science 2/2 University; and †Robert W. Franz Cancer Research Center, Providence Portland Wild-type (WT) C57BL/6, MaFIA, and RAG1 mice were obtained from 2/2 Medical Center, Portland, OR 97239 The Jackson Laboratory (Bar Harbor, ME). OX40L mice on a C57BL/6 background were graciously provided by Dr. Naoto Ishii (Tohoku Uni- Received for publication February 5, 2010. Accepted for publication August 9, 2010. versity, Sendai, Japan) and bred in the Oregon Health and Science Uni- Address correspondence and reprint requests to Dr. Jeffrey Gold, Division of versity animal facility. All mice were sex- (female) and age-matched (6–8 Pulmonary and Critical Care, Oregon Health and Science University, 3181 SW wk) and allowed to acclimatize for 1 wk prior to use. All studies were Sam Jackson Park Road, Mail Code UHN67, Portland, OR 97239. E-mail address: performed in accordance with the Oregon Health and Science University [email protected] Institutional Animal Care and Use Committee. The online version of this article contains supplemental material. Abbreviations used in this paper: APACHE, Acute Physiology and Chronic Health Cecal ligation and puncture Evaluation; BAL, bronchoalveolar lavage; CLP, cecal ligation and puncture; FSC, CLP was performed as previously described (8, 14). Briefly, mice were forward light scatter; HC, healthy controls; IC, intensive care unit controls; ICU, intensive care unit; MFI, mean fluorescence intensity; MPO, myeloperoxidase; anesthetized with 2.5% isoflurane and underwent CLP with a 19-gauge OX40L, OX40 ligand; PL, peritoneal lavage; PMN, polymorphonuclear leukocytes; needle. Mice received 1 ml 0.9% saline s.c. for resuscitation. At specified sOX40, soluble isoform of OX40; SSC, side scatter; WT, wild-type. times, the mice were used to collect plasma, bronchoalveolar lavage (BAL), and peritoneal lavage (PL; 3 ml), as previously described (8). For Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 survival experiments, mice were monitored for a total of 14 d. For Ab

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1000404 Downloaded by guest on October 1, 2021 2 OX40L IN SEPSIS

Table I. Clinical characteristics of enrolled human subjects

Sepsis (n = 31) ICU Controls (n =8)a Healthy Controls (n =9) Age (y) 58.2 6 14.5 52.2 6 22.8 35.7 6 9.3 Sex (% female) 41.9 37.5 67 APACHE II 17.7 6 6.5 14.5 6 9.3 N/A Mechanical ventilation (%) 54.8 50 N/A Vasopressors (%) 67.7 25 N/A Bacteremia (%) 54.8 N/A N/A 28-d mortality (%) 16.1 12.5 N/A Source of infection (%) N/A N/A Abdomen 9 Blood 9 Lung 45 Genitourinary 21 Brain 9 Skin 6 aDiagnoses of ICU controls included congestive heart failure and cardiogenic shock (n = 3), hypovolumic shock (n =2), pancreatitis (n = 2), and airway obstruction (n = 1).

inhibition, 250 mg anti-OX40L (hybridoma provided by Dr. Naoto Ishii) with FlowJo software (Tree Star, Ashland, OR). All reagents were pur- was injected i.p. 4 h prior to surgery. Rat IgG (BioLegend, San Diego, CA) chased from BD Pharmingen (San Jose, CA). BD compensation beads was used as an Ab control, and PBS was used as a vehicle control. Because were used to calibrate the instrument before each use. PMNs were iden- there was no difference between the two, these groups were combined tified by forward light scatter (FSC)/side scatter (SSC) characteristics and where indicated. For experiments involving MaFIA mice, they were ad- Ly6G+ (mice only), mononuclear cells by FSC/SSC, CD11b+, and F4/80 ministered AP20187 i.p. for 5 d, which results in total macrophage de- (mice) or CD14+ (human). The T cells were identified by FSC/SSC and pletion (15). further subgrouped by CD4+ and CD8+ labeling. Isotype Ab-labeled cells Cytokines and tissue myeloperoxidase (MPO) were determined by were used to control for nonspecific staining. commercially available immunoassays (R&D Systems, Minneapolis, MN; Hycult Biotechnology, Uden, The Netherlands), which were performed Human studies according to the manufacturers’ specifications. NF-kB DNA binding was All studies were approved by the Oregon Health and Science University determined by DNA-binding ELISA (Active Motif, Carlsbad, CA). Pul- Institutional Review Board. All patients meeting Society of Critical Care monary capillary leak was determined by Evans blue dye, as previously Medicine/American College of Chest Physicians criteria for sepsis in the described (8). first 24 h of intensive care unit (ICU) stay were eligible for inclusion (8). Flow cytometry ICU controls were selected at random from patients admitted to the medical ICU, and they did not have a suspected site of infection. Patients Flow cytometry was performed as previously described (8). Splenocytes, were excluded for the following reasons: presence of a do not resuscitate whole blood, or PL was collected, and 1 3 106 cells were incubated with order or decision to institute comfort care measures, hemoglobin , 7g/dl, 100 ml Fc block (mouse cells only) for 15 min and then labeled with the or the presence of active bleeding requiring .2 U packed RBCs. After following Abs: OX40, OX40L, CD4, CD8, LY6g (mouse), CD11b, F4/80 obtaining informed consent, 25 ml blood was collected into glass (serum) (mouse), and CD14 (human) at optimal concentrations for 45 min in the or EDTA-coated tubes (platelet-poor plasma) on days 1 (defined as first 24 dark. RBCs were lysed with RBC lysis buffer, and the cells were fixed with h of admission to ICU), 3–5, 7, and 14, or until death or hospital discharge. 0.1% paraformaldehyde and analyzed on a BD LSRII 8-color analyzer After obtaining preliminary data on soluble mediators and optimization of

FIGURE 1. Upregulation of OX40 and OX40L in human sepsis. A, Representative scattergram of whole blood taken from a septic patient on day 1 gated on PMNs by FSC/SSC (left panel), as well as graphs of PMN OX40L (CD252; middle panel) and CD80 expression (right panel). B, Representative dot plot of CD14+ monocytes’ (identified by FSC/SSC) expression of OX40L in a septic subject. Gates determined by isotype-stained controls. C, Quantification of PMN OX40L expression (mean fluorescence intensity [MFI]) from healthy controls (HC; n = 8), ICU controls (IC; n = 8), and septic patients at day 1 (n = 32). D, Quantification of OX40L MFI on circulating CD14+ monocytes in the same groups as in C. E, Monocyte OX40L expression (MFI) on admission to the ICU in survivors (n = 26) and nonsurvivors (n = 6). F, Correlation between day-1 PMN OX40L expression (MFI) and ICU-free days among septic patients. Downloaded by guest on October 1, 2021 The Journal of Immunology 3

FIGURE 2. sOX40 is expressed in human sepsis and activates OX40L. A, Expression of sOX40 in healthy controls, ICU controls, and septic patients within 24 h of admission to the ICU. B, THP-1 monocytes were treated with LPS (100 ng/ml) or PBS for 24 h, and cells were lysed for immunoblot (left panel) or stained for OX40L via ﬂow cytometry (right panel). Lanes nor- malized for protein (75 mg/lane). Experiment was performed three times. C, THP-1 macrophages were primed with LPS for 24 h, after which supernatant was removed, and cells were treated with vehicle (PBS), OX40:Ig (10 mg/ml), or control Ig for 24 h. Super- natants were collected and assayed for IL-6 via ELISA. The experiment was performed three times.

labeling protocols, all subsequent subjects had additional blood collected dendritic cells and macrophages and increased expression of for flow cytometry. Clinical data, including Acute Physiology and Chronic OX40L on circulating monocytes from patients with acute coronary Health Evaluation (APACHE) II scores, were recorded at the time of each syndromes, patients with sepsis exhibited upregulation of OX40L on blood draw. Patients were followed for 28 d or until death or hospital discharge. ICU and ventilator-free days were defined as the number of days circulating monocytes compared with healthy controls (Fig. 1B,1C) alive outside the ICU or off the ventilator, respectively, during the 28- (16). Of even greater interest was the expression of OX40L on d observation period. circulating PMNs (Fig. 1D). The presence of OX40L on PMNs Cell culture was specific, as evidenced by lack of staining for other costimulatory molecules, CD40 and CD80 (Fig. 1A). Overall, expression THP-1 cells were differentiated with PMA for 24 h. Cells were subsequently of monocyte and PMN OX40L was greater in septic subjects stimulated with LPS (100 ng/ml) for 24 h and harvested for immunoblot, as previously described (7). For stimulation studies, cells were blocked with compared with healthy and ICU controls on day 1 in the ICU (Fig. CD16/32 for 1 h and then OX40:Ig (10 mg/ml), Fc control, or PBS was 1C,1D); it decreased over time with resolution of symptoms added for 24 h. Supernatants were collected for ELISA. (Supplemental Fig. 1). Finally, among septic subjects there was no Statistics correlation between monocyte or PMN OX40L expression on day 1 and level of IL-6, IL-10, or IL-12. However, for the entire cohort Survival was analyzed by Kaplan–Meier analysis. All comparisons between groups were performed using the Mann–Whitney or Kruskal–Wallis (septic patients and controls), there was a significant correlation (ANOVA) test, depending on the number of groups analyzed. For multi- between day 1 OX40L expression and IL-6 (r = +0.35; p = 0.009) group analysis, intergroup comparisons were performed with the Dunn and IL-10 (r = +0.35; p = 0.02). This is consistent with the known test. Correlations were made with the Spearman test. All statistics were ability of OX40L to regulate IL-6 in other disease systems (12). done with GraphPad Prism 5.0 software (GraphPad, La Jolla, CA). We next sought to determine whether day-1 expression of OX40L could potentially serve as a biomarker for outcome among Results septic patients. Levels of monocyte OX40L were greater in non- Expression of OX40/OX40L in human sepsis survivors compared with survivors (Fig. 1E), and expression levels We first examined OX40 and OX40L expression in human sepsis. loosely correlated with severity of illness, as determined by We obtained blood from subjects admitted to the ICU with sepsis, APACHE II score (r = +0.35; p = 0.04). Further, PMN OX40L subjects admitted to the ICU for nonseptic conditions (ICU con- negatively correlated with ICU-free days (Fig. 1F). However, trols), and healthy controls. Overall, healthy controls were more there was no association between monocyte or PMN OX40L ex- likely to be younger than septic patients or ICU controls (Table I). pression levels on day 1 and the presence of bacteremia, renal In concert with prior studies documenting OX40L expression on failure, circulatory failure, or respiratory failure (data not shown).

FIGURE 3. Expression and function of OX40L in mouse sepsis. A, WT mice underwent no surgery (n =3)orCLP(n = 3) and harvested at 18 h. PL analyzed by ﬂow cytometry for OX40L expression on macrophages (F4/80+; left panel)or PMNs (Ly6G+; middle panel). Repre- sentative dot plot of PMN expression (right panel). B, Immunoblot for OX40L from TG PMNs. C, Expression of OX40 on peritoneal PMNs (left panel) and representative dot plot (right panel). Downloaded by guest on October 1, 2021 4 OX40L IN SEPSIS

FIGURE 4. OX40L contributes to lethality of polymicrobial sepsis. A,WT(n = 13) and OX40L2/2 mice (n = 9) were monitored for survival for 14 d. No additional deaths occurred beyond day 6. B, WT and OX40L2/2 mice underwent CLP and plasma harvest at 18 h for cytokine analysis (n = 4–6/group).

In contrast with our findings with OX40L, the typical reservoir of attenuation in cytokine production was associated with a decrease in OX40, CD4+ T cells, expressed only low levels of OX40 (,10% remote organ injury, as evidenced by pulmonary capillary leak (Fig. of cells stained positive), with no difference between septic sub- 5A), hepatic NF-kB induction (Fig. 5B), and hepatic PMN accu- jects and either control group at any time point (data not shown). mulation assessed by MPO 18 h after CLP (Fig. 5C). However, levels of the soluble isoform of OX40 (sOX40) were in- We next wished to establish whether physiologic blockade of creased in septic patients compared with controls on day 1 (Fig. 2A). OX40L would have a similar effect, to confirm the genetic model, We further demonstrated a proinflammatory role for sOX40 on as well as to provide important preclinical data. Administration of macrophages. We first established that OX40L could be upregu- an anti-OX40L mAb significantly improved survival compared lated on human macrophages in vitro with stimuli consistent with with isotype-treated controls (Fig. 6A). Similar to results with that observed with human sepsis. LPS stimulation of THP-1 cells OX40L2/2 mice, mice treated with anti-OX40L showed a re- significantly upregulated OX40L, by immunoblot and flow cytom- duction in plasma IL-6, IL-10, and IL-12 18 h after CLP compared etry, compared with vehicle controls (Fig. 2B). Further stimulation with controls (Fig. 6B). Finally, use of anti-OX40L as a rescue of LPS-primed cells with an OX40:Ig fusion protein significantly therapy, administered 4 h after the onset of sepsis, resulted in upregulated IL-6 compared with vehicle- or Ig-treated controls, a significant improvement in survival (Fig. 6C). confirming that OX40L is capable of reverse signaling in macrophages via stimulation with soluble ligands (Fig. 2C). OX40L activation is macrophage dependent and T cell independent in vivo OX40L regulates lethality of polymicrobial sepsis The upregulation of OX40L on PMNs and mononuclear cells To better understand the biologic role of the OX40/OX40L system suggests a pivotal role for myeloid cells in mediating OX40L in- in sepsis, we used a mouse model of polymicrobial sepsis: CLP. We flammation in vivo; however, numerous other reservoirs for OX40L first wished to confirm expression of OX40 and OX40L on mouse have been described, including vascular endothelium and smooth innate immune effector cells in polymicrobial sepsis, thereby muscle (17, 18). To confirm the requirement of macrophages in confirming our human observations and further validating the our system, we used MaFIA mice, which contain a FasR (CD95) mouse model as a useful preclinical model for OX40L-targeted expressing transgene driven by the CSF1 promoter (19). These mice, therapeutic interventions. Eighteen hours after CLP, OX40L was when treated with the dimerizing compound AP20187, get targeted upregulated on peritoneal macrophages and PMNs compared with FasR dimerization and subsequent systemic macrophage apoptosis controls (Fig. 3A). We confirmed PMN expression of OX40L via and total animal macrophage depletion by day 5 (15, 19). Unlike immunoblot (Fig. 3B). PMNs were also the predominant source of our observations with WT mice or macrophage-intact MaFIA mice OX40 in CLP, with peritoneal PMNs expressing high levels of (data not shown), anti-OX40L failed to provide any meaningful OX40 (Fig. 3C). This is similar to what was observed by other protection in macrophage-depleted MaFIA mice (Fig. 7A), suggest- investigators in human PMNs (13). Finally, similar to our human ing macrophages as a central effector cell type of OX40/OX40L- observations, circulating and splenic CD4+ T cells expressed low mediated inflammation in sepsis. levels of OX40 (,10% positive), with no difference between sep- We next sought to establish that OX40L stimulation was indeed tic and control mice (data not shown). T cell independent (either through PMN-expressed OX40 or gen- To test the in vivo significance of OX40L upregulation in sepsis, eration of sOX40), further establishing its role in innate immunity. To 2 2 we used OX40L / mice, which exhibited a dramatic improvement establish a T cell-independent role for OX40L activation in CLP, we in survival compared with WT mice after CLP (Fig. 4A). This was repeated the experiments in RAG12/2 mice. Surprisingly, similar to associated with decreased levels of IL-6, IL-10, and IL-12 in plasma our observations in WT mice, anti-OX40L significantly improved (Fig. 4B), BAL fluid, and PL fluid (IL-6 only) compared with sepsis-specific survival in RAG12/2 mice compared with controls WT mice 6 and 18 h after CLP (Supplemental Figs. 2, 3). The (Fig. 7B). Improvement in survival was associated with attenuation

FIGURE 5. OX40L contributes to remote organ injury in polymicrobial sepsis. A, WT and OX40L2/2 mice underwent CLP, and lung leak was evaluated by Evans blue dye (n = 5/group) at 18 h postsurgery. WT and OX40L2/2 mice underwent CLP, and liver was harvested at 18 h for NF-kB(B) and MPO (C)(n = 5/group). Downloaded by guest on October 1, 2021 The Journal of Immunology 5

FIGURE 6. Efﬁcacy of anti-OX40L Ab in mouse sepsis. A, WT mice were administered control IgG or anti-OX40L (250 mg) i.p. 4 h pre-CLP and monitored for survival (n = 10–13/group). Mice were monitored for 14 d, with no additional deaths after day 6. B, WT mice were administered vehicle or anti-OX40L 4 h pre-CLP and harvested at 18 h for plasma cytokine analysis (n = 5/group). C, WT mice were administered vehicle control (n = 7) or anti- OX40L (250 mg) (n = 7) i.p. 4 h post–22-gauge CLP and monitored for survival. Mice were monitored for 14 d, with no additional deaths after day 6.

of IL-6, IL-10, and IL-12 in plasma (Fig. 7C) and BAL (data not of OX40L in the knockout mice. However, it remains unclear shown). To the best of our knowledge, these data strongly suggest, why such a large discrepancy exists between OX40L2/2 and anti- for the first time, a T cell-independent mechanism of OX40L ac- OX40L–treated mice. Whether this represents incomplete blockade tivation in vivo. and/or compensatory dysregulation of other pathways with congenital OX40L deficiency remains to be determined. More impor- Discussion tantly, the ability of an OX40L Ab to provide protection after the Sepsis is a devastating disorder for which there are limited adjuvant onset of disease further established OX40L inhibition as a potential therapies. Numerous attempts at anticytokine therapy have failed in therapeutic intervention in human sepsis. Enthusiasm for the po- animal models and humans. The redundant and overlapping effects tential of OX40L blockade in human sepsis is bolstered by the of many proinflammatory cytokines suggest that strategies tar- finding of increased OX40L expression in our cohort of septic geting transcription factors or upstream receptors that control mul- patients. This seemed to be relatively specific for sepsis in a medical tiple inflammatory pathways may be more effective, such as ob- ICU population, because similar findings were not observed in servations with macrophage migration inhibitory factor and high healthy or ICU controls, and the increase in expression levels mobility group box 1 (20–22). Costimulatory molecules represent returned to that of healthy controls with resolution of illness. The another potential target, based on their presence on macrophages, association between OX40L expression levels and mortality, se- a potent source of cytokine production in sepsis, and their well- verity of illness, and ICU-free days further supports the hypothesis documented ability to regulate numerous transcriptional pathways that OX40L contributes to the lethality of polymicrobial sepsis. in a bidirectional manner (7, 8, 23). Although numerous studies investigated the role of the OX40/ One of the major findings of this study is the improved survival OX40L system in graft rejection, autoimmune disease, and anti- in OX40L2/2 mice subjected to polymicrobial sepsis. The ability tumor immunity, to our knowledge, this is the first description of to recapitulate this with pharmacological inhibition of OX40L im- OX40L blockade in the innate immune response to an acute bac- plies that this is not an epiphenomenon due to congenital absence terial infection (sepsis) (24–26).

FIGURE 7. Cellular source of OX40/OX40L activity in sepsis. A, MaFIA mice were treated with AP20187 daily for 5 d and allowed a 2-d recovery. Mice were administered vehicle control or anti-OX40L (250 mg) i.p. 4 h pre-CLP (18-gauge) and monitored for survival (n = 7–8/ group). B,RAG12/2 mice were administered vehicle control or anti- OX40L (n = 19–20/group) 4 h pre- CLP and monitored for survival for 14 d. No additional deaths occurred after day 6. C,RAG12/2 mice were administered control or anti-OX40L 4 h pre-CLP and harvested 18 h post- CLP for plasma cytokine analysis (n = 4–5/group). Downloaded by guest on October 1, 2021 6 OX40L IN SEPSIS

The mechanism by which OX40L regulates lethality lies in its This is thought to be mediated, in part, by T cells; other costimu- ability to regulate multiple inflammatory pathways. The OX40/ latory molecules were also shown to regulate this phase of sepsis OX40L system has been well described to control T cell pro- (2, 31). We cannot exclude a role for the OX40 system in the liferation in the adaptive immune response. However, a growing transition to this phase because of the high early lethality of our literature describes a putative receptor function for OX40L as well. animal model. In addition, our animal model and our human cohort Stimulation of OX40L on dendritic cells and airway smooth had predominantly bacterial sepsis, with multiple bacterial species muscle cells results in JNK activation and subsequent IL-6 pro- represented. It should be acknowledged that the contribution duction (12, 13). We now show that macrophages are capable of of OX40L to the host-inflammatory response may be pathogen similar OX40L-mediated IL-6 production in vitro. This correlates dependent. This is even more of an issue for invasive fungal dis- with the attenuation of IL-6, IL-10, and IL-12 observed in eases, which were not represented in our studies and account for up the OX40L2/2 and anti-OX40L–treated mice after CLP. Our to 5% of isolatable pathogens in human sepsis, and the incidence of in vitro findings and the correlation between OX40L expression fungal sepsis is continuing to increase (3). Finally, we acknowledge levels and circulating IL-6 and IL-10 in our human cohort suggest that the ability of OX40L and sOX40 expression to serve as a bio- that this is valid in humans as well. Clinically, the attenuation of marker for the presence or outcome in sepsis is preliminary and inflammatory cytokines improves survival through a reduction in must be validated in a larger cohort, with a multivariate analysis remote organ injury. In our mice, this was represented by a re- to better control for other confounders. A recent study suggested duction in pulmonary capillary leak and hepatic NF-kB and PMN that monocytes’ OX40L upregulation may be present in patients accumulation in the OX40L2/2 mice compared with controls. with acute coronary syndromes, and other studies described sOX40 Macrophages seem to be critical for mediating the lethal effects of upregulation in patients with autoimmune disease (29, 32). Neither OX40L in sepsis in vivo. The ability of LPS to upregulate OX40L on of these populations was represented in our cohort. macrophages provides a mechanism for the specificity of OX40L In conclusion, our data describe an integral role for the OX40/ upregulation in septic humans. However, it is likely that LPS is not OX40L system in the regulation of inflammation and mortality in the only mediator of OX40L upregulation in vivo. The known ability the early stages of the innate immune response to polymicrobial of CD154 and IL-12, both of which are upregulated in sepsis, to sepsis. The results of this study provide important preclinical in- upregulate OX40L expression on dendritic cells and T cells, re- formation in terms of OX40/OX40L specific biomarker devel- spectively, suggests multiple potential pathways toward the increased opment and therapeutic interventions for the treatment of sepsis expression seen in our patients and mice (8, 10, 27). Finally,the failure and septic shock in the future. of OX40L blockade to rescue macrophage-depleted mice suggests that macrophages are necessary for mediating the toxic effects of OX40L in sepsis in vivo. However, we acknowledge that the in- Acknowledgments duction of systemic macrophage apoptosis as a method of macro- We thank Meghan Lindauer for assistance with experiments using MaFIA mice. phage depletion may result in dysregulation of other pathways that might have been OX40L dependent in our model. Our finding of OX40L on circulating PMNs in septic mice and humans provides Disclosures another potential source of OX40L. The significance of this remains The authors have no financial conflicts of interest. unclear and will be the subject of future investigations. Finally, the description of OX40L expression on endothelial cells provides an- References other mechanism by which OX40L can regulate lethality (17). 1. Angus, D. C., W. T. Linde-Zwirble, J. Lidicker, G. Clermont, J. Carcillo, and M. R. Pinsky. 2001. Epidemiology of severe sepsis in the United States: analysis of The source of OX40 in our model remains less clear. The ability of incidence, outcome, and associated costs of care. Crit. Care Med. 29: 1303–1310. 2/2 anti-OX40L to rescue RAG1 mice suggests that there is a T cell- 2. Hotchkiss, R. S., and I. E. Karl. 2003. The pathophysiology and treatment of independent source of OX40 in sepsis and firmly places OX40 sepsis. N. Engl. J. Med. 348: 138–150. 3. Martin, G. S., D. M. Mannino, S. Eaton, and M. Moss. 2003. The epidemiology in the innate-immune response to polymicrobial sepsis. One po- of sepsis in the United States from 1979 through 2000. N. Engl. J. Med. 348: tential source is cell-bound OX40 on circulating PMNs. Baumann 1546–1554. et al. (13) described OX40 expression on human PMNs, and our 4. Remick, D. G., D. E. Newcomb, G. L. Bolgos, and D. R. Call. 2000. Comparison of the mortality and inflammatory response of two models of sepsis: lipopoly- finding of OX40 expression on mouse PMNs is consistent with this saccharide vs. cecal ligation and puncture. Shock 13: 110–116. result. It is unclear why we did not observe high levels of OX40 5. Opal, S. M., C. J. Fisher, Jr., J. F. Dhainaut, J. L. Vincent, R. Brase, S. F. Lowry, J. C. Sadoff, G. J. Slotman, H. Levy, R. A. Balk, et al. 1997. Confirmatory expression on human PMNs. Soluble OX40 represents another po- interleukin-1 receptor antagonist trial in severe sepsis: a phase III, randomized, tential means of OX40L activation in sepsis. Recently, studies double-blind, placebo-controlled, multicenter trial. The Interleukin-1 Receptor documented that OX40 can also be shed as a soluble isoform Antagonist Sepsis Investigator Group. Crit. Care Med. 25: 1115–1124. 6. Cohen, J., J. Carlet; International Sepsis Trial Study Group. 1996. INTERSEPT: (sOX40) that, similar to its cell-bound form, is able to activate and an international, multicenter, placebo-controlled trial of monoclonal antibody to induce cytokine production from OX40L-expressing airway smooth human tumor necrosis factor-alpha in patients with sepsis. Crit. Care Med. 24: muscle and, as described in this paper, macrophages as well (18). It 1431–1440. 7. Gold, J. A., M. Parsey, Y. Hoshino, S. Hoshino, A. Nolan, H. Yee, D. B. Tse, and is believed this is generated through alternative splicing, but the M. D. Weiden. 2003. CD40 contributes to lethality in acute sepsis: in vivo role source(s) of it remain less clear (28). The presence of sOX40 also for CD40 in innate immunity. Infect. Immun. 71: 3521–3528. 8. Nolan, A., M. Weiden, A. Kelly, Y. Hoshino, S. Hoshino, N. Mehta, and J. A. Gold. establishes its potential as a circulating biomarker of the presence 2008. CD40 and CD80/86 act synergistically to regulate inflammation and mortality and/or outcome in human sepsis, similar to its ability to discrimi- in polymicrobial sepsis. Am. J. Respir. Crit. Care Med. 177: 301–308. nate between patients with systemic sclerosis and systemic lupus 9. Weinberg, A. D., K. W. Wegmann, C. Funatake, and R. H. Whitham. 1999. Blocking OX-40/OX-40 ligand interaction in vitro and in vivo leads to decreased erythematosus (29). However, greater numbers of subjects will be T cell function and amelioration of experimental allergic encephalomyelitis. J. required to validate this. Immunol. 162: 1818–1826. However, it is important to point out some limitations to our study. 10. Ohshima, Y., Y. Tanaka, H. Tozawa, Y. Takahashi, C. Maliszewski, and G. Delespesse. 1997. Expression and function of OX40 ligand on human First, and most importantly, we examined the initial phases of sepsis dendritic cells. J. Immunol. 159: 3838–3848. and septic shock. Although these early phases are characterized by 11. Barrios, C. S., B. D. Johnson, J. D Henderson, Jr., J. N. Fink, K. J. Kelly, and V. P. Kurup. 2005. The costimulatory molecules CD80, CD86 and OX40L are a robust proinflammatory response, recovery is associated with up-regulated in Aspergillus fumigatus sensitized mice. Clin. Exp. Immunol. 142: a compensatory immunoparalysis and lymphocyte apoptosis (2, 30). 242–250. Downloaded by guest on October 1, 2021 The Journal of Immunology 7

12. Matsumura, Y., T. Hori, S. Kawamata, A. Imura, and T. Uchiyama. 1999. In- HMGB1 release and improve survival in experimental sepsis. Nat. Med. 10: tracellular signaling of gp34, the OX40 ligand: induction of c-jun and c-fos 1216–1221. mRNA expression through gp34 upon binding of its receptor, OX40. J. Immu- 23. Nolan, A., H. Kobayashi, B. Naveed, A. Kelly, Y. Hoshino, S. Hoshino, nol. 163: 3007–3011. M. R. Karulf, W. N. Rom, M. D. Weiden, and J. A. Gold. 2009. Differential role 13. Baumann, R., S. Yousefi, D. Simon, S. Russmann, C. Mueller, and H. U. Simon. for CD80 and CD86 in the regulation of the innate immune response in murine 2004. Functional expression of CD134 by neutrophils. Eur. J. Immunol. 34: polymicrobial sepsis. PLoS ONE 4: e6600. 2268–2275. 24. Blazar, B. R., A. H. Sharpe, A. I. Chen, A. Panoskaltsis-Mortari, C. Lees, 14. Yousefi, S., J. A. Gold, N. Andina, J. J. Lee, A. M. Kelly, E. Kozlowski, H. Akiba, H. Yagita, N. Killeen, and P. A. Taylor. 2003. Ligation of OX40 I. Schmid, A. Straumann, J. Reichenbach, G. J. Gleich, and H. U. Simon. 2008. (CD134) regulates graft-versus-host disease (GVHD) and graft rejection in al- Catapult-like release of mitochondrial DNA by eosinophils contributes to anti- logeneic bone marrow transplant recipients. Blood 101: 3741–3748. bacterial defense. Nat. Med. 14: 949–953. 25. Chen, M., X. Xiao, G. Demirci, and X. C. Li. 2008. OX40 controls islet allograft 15. Lindauer, M. L., J. Wong, Y. Iwakura, and B. E. Magun. 2009. Pulmonary in- tolerance in CD154 deficient mice by regulating FOXP3+ Tregs. Transplantation flammation triggered by ricin toxin requires macrophages and IL-1 signaling. J. 85: 1659–1662. Immunol. 183: 1419–1426. 26. Nohara, C., H. Akiba, A. Nakajima, A. Inoue, C. S. Koh, H. Ohshima, H. Yagita, 16. Liu, D. M., J. C. Yan, C. P. Wang, G. H. Chen, S. Ding, P. J. Liu, and R. Z. Du. Y. Mizuno, and K. Okumura. 2001. Amelioration of experimental autoimmune 2008. The clinical implications of increased OX40 ligand expression in patients encephalomyelitis with anti-OX40 ligand monoclonal antibody: a critical role with acute coronary syndrome. Clin. Chim. Acta. 397: 22–26. for OX40 ligand in migration, but not development, of pathogenic T cells. J. 17. Imura, A., T. Hori, K. Imada, T. Ishikawa, Y. Tanaka, M. Maeda, S. Imamura, and Immunol. 166: 2108–2115. T. Uchiyama. 1996. The human OX40/gp34 system directly mediates adhesion of 27. Mendel, I., and E. M. Shevach. 2006. Activated T cells express the OX40 ligand: activated T cells to vascular endothelial cells. J. Exp. Med. 183: 2185–2195. requirements for induction and costimulatory function. Immunology 117: 196–204. 18. Burgess, J. K., S. Carlin, R. A. Pack, G. M. Arndt, W. W. Au, P. R. Johnson, 28. Taylor, L., and H. Schwarz. 2001. Identification of a soluble OX40 isoform: J. L. Black, and N. H. Hunt. 2004. Detection and characterization of OX40 li- development of a specific and quantitative immunoassay. J. Immunol. Methods gand expression in human airway smooth muscle cells: a possible role in 255: 67–72. asthma? J. Allergy Clin. Immunol. 113: 683–689. 29. Komura, K., A. Yoshizaki, M. Kodera, Y. Iwata, F. Ogawa, K. Shimizu, 19. Burnett, S. H., E. J. Kershen, J. Zhang, L. Zeng, S. C. Straley, A. M. Kaplan, and T. Wayaku, T. Yukami, M. Murata, M. Hasegawa, et al. 2008. Increased serum D. A. Cohen. 2004. Conditional macrophage ablation in transgenic mice soluble OX40 in patients with systemic sclerosis. J. Rheumatol. 35: 2359–2362. expressing a Fas-based suicide gene. J. Leukoc. Biol. 75: 612–623. 30. Hotchkiss, R. S., C. M. Coopersmith, and I. E. Karl. 2005. Prevention 20. Calandra, T., B. Echtenacher, D. L. Roy, J. Pugin, C. N. Metz, L. Hu¨ltner, of lymphocyte apoptosis—a potential treatment of sepsis? Clin. Infect. Dis. 41 D. Heumann, D. Ma¨nnel, R. Bucala, and M. P. Glauser. 2000. Protection from (Suppl. 7): S465–S469. septic shock by neutralization of macrophage migration inhibitory factor. Nat. 31. Schwulst, S. J., M. H. Grayson, P. J. DiPasco, C. G. Davis, T. S. Brahmbhatt, Med. 6: 164–170. T. A. Ferguson, and R. S. Hotchkiss. 2006. Agonistic monoclonal antibody 21. Qin, S., H. Wang, R. Yuan, H. Li, M. Ochani, K. Ochani, M. Rosas-Ballina, against CD40 receptor decreases lymphocyte apoptosis and improves survival in C. J. Czura, J. M. Huston, E. Miller, et al. 2006. Role of HMGB1 in apoptosis- sepsis. J. Immunol. 177: 557–565. mediated sepsis lethality. J. Exp. Med. 203: 1637–1642. 32. Liu, D. M., J. C. Yan, C. P. Wang, G. H. Chen, S. Ding, P. J. Liu, and R. Z. Du. 22.Wang,H.,H.Liao,M.Ochani,M.Justiniani,X.Lin,L.Yang,Y.Al-Abed, 2008. The clinical implications of increased OX40 ligand expression in patients H. Wang, C. Metz, E. J. Miller, et al. 2004. Cholinergic agonists inhibit with acute coronary syndrome. Clin. Chim. Acta 397: 22–26. Downloaded by guest on October 1, 2021