IL-27 Negatively Regulates Tip-DC Development during Infection

ABSTRACT Tumor necrosis factor (TNF)/inducible nitric oxide synthase (iNOS)-producing dendritic cells (Tip-DCs) have profound impacts on host immune responses during infections. The mechanisms regulating Tip-DC development remain largely unknown. Here, using a mouse model of infection with African trypanosomes, we show that a deficiency in interleukin-27 receptor (IL-27R) signaling results in escalated intrahepatic accumulation of Ly6C-positive (Ly6C+) monocytes and their differentiation into Tip-DCs. Blocking Tip-DC development significantly ameliorates liver injury and increases the survival of infected IL-27R−/− mice. Mechanistically, Ly6C+ monocyte differentiation into pathogenic Tip-DCs in infected IL-27R−/− mice is driven by a CD4+ T cell-interferon gamma (IFN-γ) axis via cell-intrinsic IFN-γ signaling. In parallel, hyperactive IFN-γ signaling induces cell death of Ly6C-negative (Ly6C−) monocytes in a cell-intrinsic manner, which in turn aggravates the development of pathogenic Tip-DCs due to the loss of the negative regulation of Ly6C− monocytes on Ly6C+ monocyte differentiation into Tip-DCs. Thus, IL-27 inhibits the dual-track exacerbation of Tip-DC development induced by a CD4+ T cell–IFN-γ axis. We conclude that IL-27 negatively regulates Tip-DC development by preventing the cell-intrinsic effects of IFN-γ and that the regulation involves CD4+ T cells and Ly6C− monocytes. Targeting IL-27 signaling may manipulate Tip-DC development for therapeutic intervention.

remove debris in a steady condition (4,5) and exert an anti-inflammatory function in some disease settings (6)(7)(8). Ly6C 1 monocytes are also referred to as inflammatory monocytes. During infection and inflammation, Ly6C 1 monocytes are rapidly released from the bone marrow in a CCR2-dependent manner and recruited to infected tissues (2,9). The recruited Ly6C 1 monocytes can further differentiate into CD11b 1 Ly6C 1 CD11c 1 tumor necrosis factor (TNF)-and/or inducible nitric oxide synthase (iNOS)-producing dendritic cells (Tip-DCs) (2). Tip-DCs play a central role in defense against a broad range of pathogenic microorganisms, including viruses (10), bacteria (11), fungi (12), and protozoal parasites (13), but also contribute to the pathogenesis of infections such as African trypanosomiasis (14) and influenza (15). Thus, the development of Tip-DCs during infection has a profound impact on disease outcome; however, mechanisms involved in Tip-DC development remain incompletely understood.
In the current study, we characterized the interactions among Ly6C 1 monocytes, Ly6C 2 monocytes, and CD4 1 T cells in the context of IL-27 signaling in a mouse model of infection with African trypanosomes. We found that infected IL-27R 2/2 mice exhibited increased Ly6C 1 monocyte recruitment and Tip-DC development in the liver compared to infected wild-type (WT) mice and that Tip-DCs mediated the early mortality of infected IL-27R 2/2 mice. We further demonstrated that Tip-DC development is negatively regulated by IL-27 by preventing the cell-intrinsic effects of interferon gamma (IFN-g), a complex process that involves Ly6C 1 monocytes, Ly6C 2 monocytes, and CD4 1 T cells.

RESULTS
IL-27 prevents intrahepatic accumulation of Ly6C + monocytes and their differentiation to moDCs and Tip-DCs upon infection. Tip-DCs contribute to immunopathology during infection with African trypanosomes (14). We have previously shown that IL-27R 2/2 mice infected with African trypanosomes died significantly earlier than infected WT mice, associated with unaltered parasitemia but enhanced liver immunopathology (27). Thus, we examined Tip-DC development in the liver of IL-27R 2/2 mice during infection with African trypanosomes. As IL-27R 2/2 mice started to die on day 11 after infection with Trypanosoma congolense, we purified liver leukocytes on days 0, 5, 7, and 10 postinfection and examined different subpopulations (see Fig. S1 in the supplemental material). Infected IL-27R 2/2 mice exhibited significantly higher frequencies and absolute numbers of Ly6C 1 monocytes in the liver than infected WT mice ( Fig. 1A and B; Fig. S2A). In addition, the frequency and absolute number of monocyte-derived DCs (moDCs) were also significantly increased in the liver of infected IL-27R 2/2 mice compared to infected WT mice ( Fig. 1C and D; Fig. S2B). Importantly, dramatic increases in the frequency and absolute number of Tip-DCs were observed in the liver of infected IL-27R 2/2 mice compared to infected WT mice ( Fig. 1E and F; Fig. S2C and D). We next infected mice with Trypanosoma brucei and observed enhanced numbers of Ly6C 1 monocytes, moDCs, and Tip-DCs in infected IL-27R 2/2 mice compared to infected WT mice ( Fig. 1G to I). Thus, deficiency of IL-27 signaling promoted intrahepatic accumulation of Ly6C 1 monocytes and their differentiation to moDCs and Tip-DCs during infection with African trypanosomes.
Excessive Tip-DC development leads to shortened survival of infected IL-27R 2/2 mice. As CCR2 signaling is required for Ly6C 1 monocyte emigration from the bone marrow (9), we next compared the levels of production of the CCR2 ligands monocyte chemoattractant protein 1 (MCP-1) and MCP-3 between infected IL-27R 2/2 and WT mice. IL-27R 2/2 mice showed significantly higher plasma levels of these two chemokines than WT mice upon infection with African trypanosomes ( Fig. 2A and B). Previous work has shown that Tip-DCs contributed to immunopathology (14) and that IL-27R 2/2 mice died earlier than infected WT mice during infection with African trypanosomes (27). To determine whether the early mortality of infected IL-27R 2/2 mice was attributed to Tip-DCs, we generated IL-27R and CCR2 double-knockout (CCR2 RFP/RFP IL-27R 2/2 ) mice (Fig. S3). Compared to IL-27R 2/2 mice, CCR2 RFP/RFP IL-27R 2/2 mice survived significantly longer during infections with T. congolense (Fig. 2C) or T. brucei (Fig. 2D), suggesting that disruption of CCR2 signaling prevented the early mortality of infected IL-27R 2/2 mice. The enhanced survival of infected CCR2 RFP/RFP IL-27R 2/2 mice was associated with reduced plasma levels of alanine aminotransferase (ALT), suggestive of reduced liver pathology ( Fig. 2E and F).
Taking advantage of intravital microscopy (IVM), here, we observed that limited red fluorescent protein-positive (RFP 1 ) cells were scattered across the liver in the infected CCR2 RFP/RFP IL-27R 2/2 mice. In contrast, infected CCR2 1/RFP IL-27R 2/2 mice displayed a distinct phenotype with substantial RFP 1 clustering in the liver (Fig. 2G). As expected, disruption of CCR2 signaling in IL-27R 2/2 mice almost abolished the emigration of Ly6C 1 monocytes from the bone marrow to the blood, as demonstrated by the reduced frequency and absolute number of Ly6C 1 monocytes in the blood, spleen, and liver but the enhanced frequency and absolute number of this subset in the bone marrow ( Fig. 2H and I). Accordingly, we detected significantly lower frequencies of moDCs and Tip-DCs in the liver of infected CCR2 RFP/RFP IL-27R 2/2 mice than in infected IL-27R 2/2 mice (Fig. 2I). Interestingly, disrupting CCR2 signaling did not affect the frequency of activated intrahepatic T cells but led to a reduction in the number of activated T cells (Fig. S4). This may reflect the overall attenuated liver inflammation and reduced numbers of total leukocytes recruited to the liver in the absence of CCR2 signaling. Collectively, blocking Tip-DC development by disrupting CCR2 signaling prevented the early mortality of infected IL-27R 2/2 mice.
IL-27 inhibits CD4 + T cell-IFN-c axis-induced Tip-DC development. Having shown the enhanced development of Tip-DCs in infected IL-27R 2/2 mice and their contribution to early mortality, we next examined how IL-27 regulated Tip-DC development. We purified Ly6C 1 monocytes from the liver and bone marrow and cultivated the cells in the absence or presence of IL-27 to examine their differentiation into Tip-DCs in vitro.
IL-27 limits CD4 + T cell-IFN-c axis-induced cell death of Ly6C 2 monocytes. We hypothesized that the reduction of Ly6C 2 monocytes is attributed to cell death induced by the excessive production of IFN-g in infected IL-27R 2/2 mice. To this end, we examined the apoptosis and necrosis of Ly6C 2 monocytes in infected IL-27R 2/2 mice. As expected, deficiency of IL-27R signaling led to a significantly higher percentage of Ly6C 2 monocyte necrosis in the liver during infection (Fig. 5A and B). Compared to infected WT mice, Ly6C 2 monocytes in the liver of infected IL-27R 2/2 mice displayed significantly low levels of expression of CX3CR1 (Fig. 5C), a molecule known for the survival of Ly6C 2 monocytes (31). Interestingly, depletion of CD4 1 T cells or neutralization of IFN-g significantly reduced cell necrosis of intrahepatic Ly6C 2 monocytes in infected IL-27R 2/2 mice, leading to enhanced viable cells (Fig. 5D). In addition, depletion of CD4 1 T cells or neutralization of IFN-g significantly enhanced the expression of CX3CR1 on Ly6C 2 monocytes (Fig. 5E) and dramatically increased the percentage and absolute number of intrahepatic Ly6C 2 monocytes in infected IL-27R 2/2 mice (Fig. 5F). Taken together, these results suggested that IFN-g, likely produced by CD4 1 T cells, drives cell death of Ly6C 2 monocytes in infected IL-27R 2/2 mice.
IFN-c promotes differentiation of Ly6C + monocytes and cell death of Ly6C 2 monocytes in a cell-intrinsic manner. Having demonstrated that IFN-g drives the differentiation of Ly6C 1 monocytes into Tip-DCs and cell death of Ly6C 2 monocytes, we next examined whether these functions are cell intrinsic. Ly6C 1 monocytes were isolated from the bone marrow of naive WT and IFN-g receptor-deficient (IFN-gR 2/2 ) mice and labeled with PKH26 and CellVue, respectively. The two types of cells were mixed (at a 1:1 ratio) and adoptively transferred to infected IL-27R 2/2 mice to determine their differentiation into Tip-DCs (Fig. 6A). Prior to the transfer, almost all monocytes from WT and IFN-gR 2/2 mice expressed Ly6C (Fig. 6B). After transfer, most of the Ly6C 1 monocytes from WT mice, but not IFN-gR 2/2 mice, maintained the Ly6C 1 phenotype (Fig. 6B and C). Importantly, WT Ly6C 1 monocytes exhibited a significantly higher rate of differentiation into MHCII 1 (major histocompatibility complex class II-positive) CD11c 1 moDCs and Tip-DCs than IFN-gR 2/2 Ly6C 1 monocytes (Fig. 6B, D, and E), suggesting that IFN-g signaled directly on Ly6C 1 monocytes to drive their differentiation into moDCs and Tip-DCs.
We next isolated Ly6C 2 monocytes from the liver of infected WT and IFN-gR 2/2  IL-27 Inhibits Tip-DC Development ® mice and labeled them with PKH26 and CellVue, respectively. The two types of cells were mixed (at a 1:1 ratio) and adoptively transferred to infected IL-27R 2/2 mice to evaluate their viability (Fig. 6F). The results showed that the percentage of viable WT Ly6C 2 monocytes was significantly lower than that of IFN-gR 2/2 Ly6C 2 monocytes ( Fig. 6G and H). Accordingly, the percentage of apoptotic and necrotic WT Ly6C 2 monocytes was significantly higher than that of IFN-gR 2/2 Ly6C 2 monocytes, suggesting that IFN-g signaled directly on Ly6C 2 monocytes to trigger cell death of this subpopulation.
Collectively, these results demonstrated that IFN-g promoted the differentiation of Ly6C 1 monocytes and the cell death of Ly6C 2 monocytes in a cell-intrinsic manner.

DISCUSSION
During infection, Ly6C 1 monocytes emigrate from the bone marrow and then migrate to infected tissues, where they can differentiate into Tip-DCs (11). Tip-DCs make essential contributions to immune defense against a broad range of microbial pathogens, including Listeria monocytogenes (11,32), Brucella melitensis (33), Mycobacterium tuberculosis (34), Toxoplasma gondii (13,35), Plasmodium chabaudi (36), Leishmania major (37), and Cryptococcus neoformans (12,38), but they can also lead to immunopathology during infection with Trypanosoma brucei (14,39) and influenza virus (15). It is known that Ly6C 1 monocytes use CCR2 signaling to emigrate from the bone marrow into the blood (9, 32). However, less is known about how Ly6C 1 monocyte differentiation into Tip-DCs in infected tissues is regulated. We have previously shown that deficiency of IL-27 signaling resulted in early mortality of mice infected with African trypanosomes, associated with excessive secretion of IFN-g and liver pathology (27). In the current study, we demonstrate that mice deficient in IL-27 signaling displayed escalated Tip-DC development and that blocking Tip-DC development by disrupting CCR2 signaling significantly extended the survival of infected IL-27R 2/2 mice. Thus, our studies reveal that Tip-DCs mediate the early death of infected mice and that the development of Tip-DCs is negatively regulated by IL-27 signaling during infection.
Previous work has shown that IL-27 functions on CD4 1 T cells to exert its antiinflammatory property by either directly inhibiting CD4 1 T cell activation (21,22) or promoting CD4 1 T cell secretion of IL-10 (24-26). However, less is known about the impact of IL-27 on myeloid cells, including monocytes. Taking advantage of intravital imaging, we revealed that recruited Ly6C 1 monocytes formed clusters in the liver necrosis sites of infected IL-27R 2/2 mice and that CD4 1 T cells accumulated in the clusters. Depletion of CD4 1 T cells or neutralization of IFN-g diminished Tip-DC development in infected IL-27R 2/2 mice, suggesting that IL-27 negatively regulates Ly6C 1 monocyte differentiation into Tip-DCs by preventing the effects of IFN-g, likely secreted by CD4 1 T cells (40,41). In this context, we have previously shown that neutralization of IFN-g or depletion of CD4 1 T cells prevented the early mortality of infected IL-27R 2/2 mice (27). In line with previous findings, blocking Tip-DC development significantly reduced liver pathology and extended the survival of infected IL-27R 2/2 mice. Obviously, IL-27 inhibits Ly6C 1 monocyte differentiation into Tip-DCs, preventing early mortality during infection with African trypanosomes.
We found that increased Tip-DC development was associated with decreased Ly6C 2 monocytes in the liver of infected IL-27R 2/2 mice. Interestingly, adoptive transfer of Ly6C 2 monocytes diminished Tip-DC development in infected IL-27R 2/2 mice, demonstrating that, in contrast to IFN-g that drives Tip-DC development, Ly6C 2 monocytes inhibit Tip-DC development in our experimental setting. In contrast to Ly6C 1 monocytes, the function of Ly6C 2 monocytes is still a mystery. Ly6C 2 monocytes have been shown to patrol on the vessel wall and act as "blood macrophages" to remove debris on the vessel wall under resting conditions (4,5). It is also proposed that Ly6C 2 monocytes have both anti-inflammatory (6)(7)(8) and proinflammatory (42)(43)(44) properties. The finding that Ly6C 2 monocytes inhibited Tip-DC development is of significance because it highlights a previously undescribed function of Ly6C 2 monocytes and suggests that targeting Ly6C 2 monocytes may manipulate Tip-DC development for therapeutic intervention. However, the mechanism involved in inhibiting Tip-DC development by Ly6C 2 monocytes is unknown and deserves further investigation. It has been reported that Ly6C 1 monocytes can also differentiate into alternative macrophages with the help of CD4 1 T cells (45) and that Ly6C 2 monocytes can facilitate this transition (46). It is likely that Ly6C 2 monocytes promote the differentiation of Ly6C 1 monocytes into macrophages, consequently reducing the transition of Ly6C 1 monocytes into Tip-DCs. Alternatively, Ly6C 2 monocytes may directly inhibit the differentiation of Ly6C 1 monocytes into Tip-DCs through cell-cell contact or secreting antiinflammatory mediators such as IL-10.
We further demonstrated that neutralization of IFN-g or depletion of CD4 1 T cells increased the number of Ly6C 2 monocytes in infected IL-27R 2/2 mice. These results suggest that IFN-g, likely produced by CD4 1 T cells, induced excessive cell death of Ly6C 2 monocytes in infected IL-27R 2/2 mice. Thus, IFN-g not only promoted Ly6C 1 monocyte differentiation into Tip-DCs but also drove cell death of Ly6C 2 monocytes in infected IL-27R 2/2 mice. It is important to note that the functions of IFN-g on both Ly6C 1 monocytes and Ly6C 2 monocytes were cell intrinsic, i.e., directly signaling with IFN-g receptor expressed on the two subsets of monocytes, as demonstrated by adoptive-transfer experiments. These observations are important because they demonstrate that IFN-g is involved in Tip-DC development in the context of IL-27 signaling by multiple mechanisms; i.e., IFN-g promotes Tip-DC development directly through signaling on Ly6C 1 monocytes and indirectly through signaling on Ly6C 2 monocytes. It is expected that IFN-g signals on Ly6C 1 monocytes promote their activation and differentiation into Tip-DCs. However, it is surprising that IFN-g also drives Tip-DC development indirectly by inducing cell death of Ly6C 2 monocytes, highlighting a previously undescribed mechanism and reflecting the complexity of interactions between Ly6C 1 monocytes and Ly6C 2 monocytes.
In summary, IL-27R 2/2 mice displayed increased differentiation of Ly6C 1 monocytes into Tip-DCs in the liver during infection. Blocking Tip-DC development extended the survival of infected IL-27R 2/2 mice, suggesting that Tip-DCs mediates the early mortality of infected mice. We further demonstrated that the differentiation of Ly6C 1 monocytes into Tip-DCs was driven by IFN-g and CD4 1 T cells but inhibited by Ly6C 2 monocytes in infected IL-27R 2/2 mice. IFN-g also induced cell death of Ly6C 2 monocytes in infected IL-27R 2/2 mice. Finally, we revealed that IFN-g promotes Ly6C 1 monocyte differentiation into Tip-DCs and induces Ly6C 2 monocyte death in a cell-intrinsic manner. Thus, IFN-g drives Tip-DC development directly through signaling on Ly6C 1 monocytes and indirectly through signaling on Ly6C 2 monocytes. We conclude that IL-27 negatively regulates Tip-DC development by preventing the cell-intrinsic effects of IFN-g. IL-27 signaling shows potential as a target for manipulating Tip-DC development and therapeutic intervention.

MATERIALS AND METHODS
Ethics statement. All experiments were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (47). The protocols of animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Maryland, College Park.
Parasites. T. congolense, Trans Mara strain, variant antigenic type (VAT) TC13, was obtained from Jude Uzonna (University of Manitoba, Canada). T. brucei AnTat1.1E was provided by Stefan Magez (Vrije Universiteit Brussel, Belgium). For passages of the parasites, CD1 mice were immunosuppressed with IL-27 Inhibits Tip-DC Development ® cyclophosphamide and intraperitoneally (i.p.) infected with the parasites. Blood was taken for purification of parasites by DEAE-cellulose chromatography as described previously (52).
Purification of intrahepatic leukocytes, spleen cells, and bone marrow cells. Liver leukocytes were purified as described previously (27). In brief, the liver was perfused with phosphate-buffered saline (PBS) until it became pale. The liver was collected, and the gallbladder was removed from the liver. The liver was cut into small pieces with surgical scissors and then mechanically dispersed by using a sterile syringe plunger to press through a 70-mm cell strainer in 50 ml RPMI 1640 medium containing 5% fetal calf serum (FCS). The cell suspension was centrifuged at 30 Â g with the off-brake setting for 10 min at 4°C. The supernatant was collected and centrifuged at 300 Â g for 10 min at 4°C. The pellet was resuspended in 10 ml 37.5% Percoll in Hanks' balanced salt solution (HBSS), followed by centrifugation at 850 Â g with the off-brake setting for 30 min at 23°C. Erythrocytes in the cell pellets were lysed with 0.5 to 1 ml ACK (Ammonium-Chloride-Potassium) buffer at room temperature for 5 min, followed by the addition of 14 ml RPMI 1640 medium containing 5% FCS. After centrifugation at 300 Â g for 10 min at 4°C , cells were resuspended in cold RPMI 1640 medium containing 5% FCS. For splenocyte purification, spleens were removed. The tissues were minced and forced through a 70-mm cell strainer in 50 ml RPMI 1640 medium containing 5% FCS. Erythrocytes were lysed as described above.
To collect bone marrow cells, the femur and the tibia from both hind legs were removed. The extreme distal tip of each extremity was cut off. PBS was forced through the bone with a 23-gauge needle and filtered through 40-mm nylon mesh. Erythrocytes were lysed as described above.
Flow cytometry. Isolated intrahepatic, splenic, and bone marrow leukocytes were suspended in flow cytometry staining buffer. Fc receptors were blocked by incubation of the cells with purified antimouse CD16/CD32 for 15 min on ice. For surface staining, cells were then washed with staining buffer, followed by staining with mAbs for various cell markers. After two washes, cells were fixed and resuspended in staining buffer. For intracellular staining of TNF-a and iNOS, freshly isolated cells were cultivated for 6 h in the presence of a protein transport inhibitor (brefeldin A). After surface staining, the cells were then fixed and washed twice using the intracellular fixation and permeabilization buffer set (Thermo Fisher Scientific), followed by staining of intracellular TNF-a and iNOS for 30 min at room temperature. A final wash using the permeabilization buffer was performed before data acquisition. For cell viability staining, after surface staining with no fixation, cells were incubated with 7-AAD and annexin V in annexin V binding buffer (BioLegend) for 15 min at room temperature, followed by the addition of an extra 400 ml annexin V binding buffer. Data were collected using the FACSCanto II or FACS Celesta system (BD Biosciences) and analyzed with FlowJo (BD Biosciences).
Determination of chemokines and alanine transaminase. The plasma levels of MCP-1 and MCP-3 were determined with enzyme-linked immunosorbent assay (ELISA) kits from Thermo Fisher Scientific. Liver alanine transaminase (ALT) activities were examined using the EnzyChrom alanine transaminase assay kit (BioAssay Systems).
Adoptive transfer of Ly6C + and Ly6C 2 monocytes. Ly6C 1 monocytes were isolated from the bone marrow of naive wild-type (WT) and IFN-gR 2/2 mice using the bone marrow monocyte isolation kit from Miltenyi Biotec. Ly6C 2 monocytes were isolated from the livers of T. congolense-infected WT and IFN-gR 2/2 mice on days 7 to 10 postinfection using CD11b microbeads (Miltenyi Biotec) and cell sorting (53). For cotransfer experiments, monocytes (Ly6C 1 or Ly6C 2 ) from WT and IFN-gR 2/2 mice were labeled with the lipophilic membrane dyes PKH26 and CellVue, respectively. The labeled monocytes from WT and IFN-gR 2/2 mice were mixed at a 1:1 ratio, and a total of 4 Â 10 6 cells were transferred to IL-27R 2/2 mice via the tail vein on day 7 after infection with T. congolense. Flow cytometry was performed to analyze the donor cells in the liver of the recipient mice 48 h after transfer. In another set of experiments, 2 Â 10 6 WT Ly6C 2 monocytes were transferred to IL-27R 2/2 mice on days 0 and 3 postinfection. Recipient mice were sampled 48 h after transfer.
Intravital microscopy. Intravital microscopy (IVM) was performed as described previously (52). Briefly, mice were anesthetized by i.p. injection of a mixture of ketamine (200 mg/kg of body weight) and xylazine (10 mg/kg). Cannulation of the tail vein was performed for delivery of Abs. A middle laparotomy was performed. A lateral incision was made along the costal margin to the midaxillary line to expose the liver. The mice were placed on a customized acrylic imaging stage, and the liver was covered with a trimmed coverslip. To prevent tissue dehydration, the exposed abdominal tissues were covered with saline-soaked gauze, and the liver was continuously moistened with a saline-soaked Kimwipe. The liver was imaged in real time using the Zeiss Axio Examiner Z1 system. An infrared heating lamp was used to maintain the body temperature of the mouse throughout the imaging.
Generation and genotyping of CXCR6 GFP/GFP IL-27R 2/2 and CCR2 RFP/RFP IL-27R 2/2 mice. For genotyping, we used a PCR-based strategy using a common forward primer with WT-specific and knockout (mutant [MT])-specific reverse primers in one reaction. To generate the double-knockout mice, the reporter mice (CCR2 RFP/RFP or CXCR6 GFP/GFP ) were crossed with IL-27R 2/2 mice. The offspring of the resultant heterozygotes were genotyped, from where the heterozygous or double-knockout littermates were further mated to eliminate the WT gene from the colony. Primers used are as follows: Cell culture. To evaluate the effects of IL-27 on Tip-DC development, Ly6C 1 monocytes were purified from the liver or bone marrow of the WT mice on day 7 postinfection. Cells were cultivated in the presence of 20 ng/ml GM-CSF, 20 ng/ml IFN-g, and various concentrations of IL-27 for 48 h. In another group, cells were pretreated with 100 ng/ml IL-27 for 4 h prior to the introduction of 20 ng/ml GM-CSF, 20 ng/ml IFN-g, and 100 ng/ml IL-27 (designated as 100 1 100 ng/ml). Brefeldin A was added for the last 6 h of cultivation. An ex vivo group was included as a baseline by cultivation for 6 h with brefeldin A without the addition of cytokines.
To evaluate the role of Ly6C 2 monocytes in Tip-DC development, liver Ly6C 1 and Ly6C 2 monocytes were isolated from WT mice on day 7 postinfection. Ly6C 1 monocytes were labeled with PKH26. Ly6C 1 monocytes were cultured with or without Ly6C 2 monocytes for 48 h in the presence of 20 ng/ml GM-CSF and 20 ng/ml IFN-g. Brefeldin A was added to the culture for the last 6 h of cultivation.
Statistical analysis. Data are presented as the means 6 standard errors of the means (SEM). Statistical significance was determined by Student's t test, analysis of variance (ANOVA) plus Dunnett's multiple-comparison test, or a log rank test for curve comparison using GraphPad Prism 8.0 software. A P value of ,0.05 is considered statistically significant.
Data availability. All relevant data are within the manuscript and its supplemental material.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.