All animal experiments for this study followed the Korean Food and Drug Administration (KFDA) animal care and use guidelines. Animal experiments were performed according to procedures approved by the Institutional Animal Care and Use Committee of Chungnam National University, Daejeon, South Korea (permit number: CNU-01043).

C57BL/6 (H-2Kb and I-Ab), C57BL/6J TLR2 knockout (TLR2^-/-; B6.129-Tlr2tm1Kir/J), C57BL/10 TLR4 knockout (TLR4^-/-; C57BL/10ScNJ), and BALB/c mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). All mice were maintained under specific pathogen-free conditions in the Medical Research Center of Chungnam National University.

Bone marrow-derived macrophages (BMDMs) were cultured in Dulbecco’s modified Eagle’s medium (Welgene Co., Daegu, Korea) supplemented with 10% fetal bovine serum (FBS) (Welgene), 100 unit/ml penicillin/100 μg/ml streptomycin (Welgene) and 50 ng/ml mouse macrophage colony stimulating factor (M-CSF) (R&D Systems, Minneapolis, MN, USA) at 37°C and 5% CO₂ for 6 days.

Transformation with recombinant proteins and recombinant Rv2145c-expressing M. smegmatis strains was performed as described previously (24). Fragments of Rv2145c, Ag85B, Rv2145c D1 and Rv2145c D2 were amplified from Mtb H37Rv ATCC 27294 genomic DNA by PCR using the appropriate primers (Table S1). The recombinant plasmids containing Rv2145c, Ag85B, Rv2145c D1 and Rv2145c D2 were transformed into Escherichia coli (E. coli) BL21 cells by heat shock for 1 min at 42°C.

M. smegmatis strain mc²155, grown in 7H9 medium supplemented with 10% oleic acid albumin dextrose catalase (OADC; BD Biosciences, San Jose, CA, USA) was harvested, washed with cold 10% glycerol and resuspended in the same buffer. The plasmids (pVV16 and pVV16-Rv2145c) were electroporated into the bacteria according to the standard procedure of the Gene Pulser apparatus (Bio-Rad, Hercules, CA, USA). The recombinant M. smegmatis strains were selected on 7H10 agar plates containing 50 μg/ml kanamycin (Sigma-Aldrich, St. Louis, MO, USA, 60615). Plates were incubated at 37°C for 3–4 days to obtain the recombinant strains.

Mtb H37Rv (ATCC 27294, Mtb) and M. smegmatis strain mc²155, obtained from American Type Culture Collection (ATCC, Manassas, VA), were grown in 7H9 medium supplemented with 0.5% glycerol and 10% OADC at 37°C. Kanamycin (50 μg/ml) was added to culture the recombinant M. smegmatis strains. Aliquoted bacteria stored at -80°C were thawed and plated onto 7H10 agar plates with or without 50 μg/ml kanamycin to enumerate the colony-forming units (CFUs).

For in vitro infections with Mtb, BMDMs were infected with Mtb (multiplicity of infection, MOI = 1) for 4 h were treated with gentamicin (200 μg/ml; Sigma) for 2 h kill noningested Mtb, washed with medium, and then incubated with antibiotic-free complete medium containing Rv2145c for the indicated additional time periods. To measure the number of bacteria, the cell lysates were plate onto 7H10 agar plates. For M. smegmatis, BMDMs were infected at an MOI of 10, treated with gentamicin (200 μg/ml) for 2 h to remove any remaining extracellular bacteria and then incubated in medium with 10% FBS containing gentamicin (20 μg/ml) for the indicated additional time period. The bacterial counts were performed using 7H10 agar plates containing kanamycin (50 μg/ml).

For in vivo infection with M. smegmatis, C57BL/6 mice were infected via the tail vein with the bacteria (using 1 × 10⁶ CFU/mouse) for 24, 72 or 168 h. To measure the bacterial burden in mice, the lung and spleen were homogenized in 1 ml PBS, and serial dilutions of the homogenates were plated on 7H10 agar plates containing kanamycin (50 μg/ml). CFUs were determined after incubation at 37°C.

Endotoxin removal resin (END-X B15) and an endotoxin filter (END-X) were purchased from the Associates of Cape Cod (East Falmouth, MA, USA). LPS from Escherichia coli O111:B4 (cat code. tlrl-eblps) and palmitory-3-Cys-Ser-(Lys)4 (Pam3CSK4, cat code. tlrl-pms) were purchased from InvivoGen (San Diego, CA, USA). A fluorescein isothiocyanate (FITC)-Annexin V Apoptosis Detection Kit (cat. 556547) and DAPI (cat. D3571) were purchased from BD Biosciences and Molecular Probes (Eugene, OR, USA), respectively. Specific inhibitors of STAT3 (S31-201, cat. 573102), ERK1/2 (U0126, cat. 662005), p38 (SB203580, cat. 559389), JNK (SP600125, cat. 420119) and NF-κB (BAY 11-7082, cat. 196870) were purchased from Calbiochem. Silencer negative control siRNA (AM4611), silencer predesigned siRNA STAT3 (cat. AM16708) and TurboFect transfection reagent were purchased from Thermo Fisher Scientific. Detailed information on the antibodies is listed in Table S2.

BMDMs treated with Rv2145c or LPS (100 ng/ml) for 24 h were stained with Annexin V and PI according to manufacturer’s instruction and analyzed using a FACSCanto flow cytometer (BD Biosciences). FlowJo data analysis software (BD Biosciences) was used data processing.

The cytokine levels were determined from the culture supernatants of BMDMs stimulated with various stimuli, including Rv2145c, or infected with mycobacteria with a Vmax kinetic microplate reader (Molecular Devices Co., Sunnyvale, CA, USA) according to the manufacturer’s instructions (eBioscience; San Diego, CA, USA).

After the appropriate stimulation or infection, the cells were harvested and preincubated with 0.5% BSA in PBS for 30 min. The cells were stained with fluorescence-conjugated antibodies for 30 min at room temperature. The intensity of expression of surface molecules and cytokine receptors was analyzed by flow cytometry (FACSCanto), and the data were processed using FlowJo data analysis software (BD Biosciences).

For heat-denaturation, LPS or Rv2145c were incubated at 100°C for 1 h. For digestion of proteinase K (PK), LPS or Rv2145c were incubated along with 50 μg/ml soluble PK followed by heating for 20 min at 100°C to deactivate the enzyme, and then prepared stimulant were treated in BMDMs. For pretreatment with polymyxin B (PMB; Sigma), LPS and Rv2145c were incubated in medium containing 50 μg/ml of PMB for 1 h at room temperature. After 24 h, cytokine levels in the supernatant were analyzed by sandwich ELISA.

BALB/c mice were immunized intraperitoneally three times with incomplete Freund’s adjuvant-emulsified Rv2145c (25 μg) at 2-week intervals. Serum collection was performed 1 week after the final immunization.

After the appropriate treatment or infection, the cells fixed on coverslips were permeabilized in 0.1% Triton X-100 for 10 min, incubated with primary antibodies for 24 h at 4°C, and then incubated with secondary antibodies for 1 h at room temperature. DAPI was used to stain nuclei. After mounting, immunofluorescence images were acquired using a confocal laser-scanning microscope (TCS SP8, Leica, Wetzlar, Germany) with consistent excitation, emission, pinhole, and exposure time parameters.

Immunoprecipitation was performed as described previously (24). In brief, the cell lysate and 20 μg His-tagged Rv2145c (20 μg) were mixed and incubated at 4°C overnight, followed by mixing and precipitating with the relevant antibodies and Ni-NTA agarose (Qiagen, Hilden, Germany) beads. The beads were washed and the bound proteins were resolved by SDS-PAGE for immunoblotting with anti-TLR2 and anti-TLR4 antibodies.

After the appropriate treatment or infection, the cell lysates were loaded onto SDS-polyacrylamide gels, and the separated protein were transferred to a nitrocellulose membrane. The membranes were blocked in 5% skim milk for 1 h at room temperature and then incubated with antibodies for 24 h at 4°C and subsequently with HRP-conjugated secondary antibodies for 1 h at room temperature. Target proteins were visualized using the ECL advance kit (GE Healthcare, Little Chalfont, UK).

Mtb was cultured in Sauton’s synthetic medium and harvested at 20 days and 56 days. The 5 μg of the culture filtrate was coated onto plates at 4°C for 24 h and then reacted with an anti-Rv2145c mouse antibody for 1 h, followed by an enzyme-conjugated secondary antibody. TMB substrate reagent (BD) was then added for color development. The plates were read on a Vmax kinetic microplate reader (Molecular Devices Co., Sunnyvale, CA, USA) at 450 nm.

Mtb were grown in 7H9 medium supplemented with 0.5% glycerol and 10% OADC to an OD₆₀₀ of 0.8. Cells were harvested by centrifugation, washed with DPBS, and lysed with lysis buffer containing 1 mM phenylmethylsulfonyl (PMSF), 5 mM DTT, 0.1 mg/ml lysozyme and protein cocktail inhibitor by sonication for 20 min. Lysates were centrifuged at 3,000 g for 30 min to remove unbroken cells. Whole-cell lysates were ultracentrifuged at 27,000 g for 30 min to separate and obtain the pellet fraction (cell wall) and supernatant (cytosol). The cell wall was washed with lysis buffer, recentrifuged, and resuspended in lysis buffer. All centrifugation steps were performed at 4°C.

BMDMs were transfected with above siRNA (siCON, 50 nM) and STAT3 siRNA (siSTAT3, 50 nM) using TurboFect according to the manufacturer’s instructions. After transfection, the cells were exposed to Mtb and Rv2145c, and then bacterial growth and cytokine production were analyzed.

The mice were injected intraperitoneally with 1 mg of either an anti-IL-10R antibody or IgG1 isotype control antibody and after 24 h, were infected intravenously with M. smegmatis (using 1 × 10⁶ CFU/mouse). After 24 or 72 h, the mice were sacrificed.

Histopathology was performed as described previously (25). In brief, the prepared tissue sections were stained with hematoxylin and eosin (H&E) and acid-fast bacilli (AFB). All tissue analyses were performed through blind assessment. ImageJ software (National Institutes of Health, Bethesda, MD) was used to analyze histological image to quantify the granuloma area and bacterial count.

All performed experiments were repeated at least three times. Tukey’s multiple comparison test distribution or two-way ANOVA using statistical software (GraphPad Prism Software, version 4.03; GraphPad Software, San Diego, CA) were used to determine statistical significance between samples. Data in the graphs are presented as the mean values ± SD, and *p < 0.05, **p < 0.01 or ***p < 0.001 were considered statistically significant.

A recombinant Rv2145c protein purified from Escherichia coli (E. coli) appeared as a major band at 35 kDa on SDS-PAGE and reacted with an anti-His antibody (Figure 1A). Purified Rv2145c was not cytotoxic in bone marrow-derived macrophages (BMDMs) up to a concentration of 10 μg/ml (Figure 1B). We previously reported that MAP1889c, which is homologous with Rv2145c, stimulates DCs and macrophages to increase surface molecule presentation and secrete cytokines (23). Therefore, we investigated whether Rv2145c could also induce macrophage activation. As shown in Figure 1C, Rv2145c induced significant production of IL-10, IL-6, TNF-α, and IL-12p70 in a dose-dependent manner when compared to that in untreated BMDMs. At a concentration of 10 μg/ml, Rv2145c-stimulated BMDMs produced significantly higher levels of IL-10, IL-6, and TNF-α but not IL-12p70 than LPS (100 ng/ml)-stimulated BMDMs. The expression of co-stimulatory molecules and MHC class molecules in Rv2145c-treated BMDMs was significantly increased in a dose-dependent manner, which was comparable with that after LPS treatment (Figure 1D), when compared to that of the untreated controls. We next confirmed that Rv2145c-induced macrophage activation was not due to LPS contamination using heat denaturation or treatment with PK or PMB. The activity of Rv2145c was significantly decreased by boiling and PK treatment but not pretreatment with PMB (Figure S1). Taken together, these results suggest that Rv2145c induces macrophage activation with elevated IL-10, IL-6, and TNF-α, and moderate IL-12p70 production.

We next determined whether Rv2145c stimulated BMDMs through TLR recognition. Confocal microscopy showed that Rv2145c interacted with the surface of BMDMs from WT and TLR2^-/- BMDMs but did not bind to TLR4^-/- BMDMs (Figure 2A). Immunoprecipitation analysis with an anti-His antibody also showed that Rv2145c directly interacted with TLR4 molecules (Figure 2B). Furthermore, Rv2145c-mediated cytokine production and surface molecule expression were significantly depressed in TLR4^-/- BMDMs compared to those in WT and TLR2^-/- BMDMs, while LPS and Pam3 activities were decreased in TLR4^-/- BMDMs and TLR2^-/- BMDMs, respectively (Figures 2C, D). These results suggest that Rv2145c activated macrophages via TLR4 pathway.

We analyzed whether Rv2145c induced macrophage activation via STAT3, MAPKs and NF-κB. As shown in Figures 3A, B, Rv2145c triggered strong phosphorylation of STAT3, ERK1/2, p38 and JNK and phosphorylation and degradation of IκB-α in BMDMs. STAT3 was activated later than MAPKs. In addition, a confocal assay showed that Rv2145c induced robust nuclear translocation of p65 from the cytosol (Figure 3C). To confirm the functional roles of STAT3, MAPK and NF-κB signaling in Rv2145c-mediated macrophage activation, pharmacological inhibitors were pretreated before Rv2145c stimulation. Predictably, these inhibitors significantly suppressed Rv2145c-mediated IL-10, TNF-α, and IL-12p70 production (Figure 3D) and expression of surface molecules such as CD80 and MHC-I (Figure 3E). Our data indicate that the STAT3, MAPKs and NF-κB signaling pathways are involved in Rv2145c-mediated macrophage activation.

It has been reported that Rv2145c is related to cell wall and cell processes in functional categories and is associated with the cell membrane fraction (26, 27). We also found that Rv2145c was detected in the supernatants of Mycobacterium tuberculosis (Mtb) cultured for 8 weeks but not 3 weeks (Figure 4A). Immunoblot assays showed that Rv2145c was located mostly in the cell wall fraction of Mtb but was also weakly detected in the cytosol fraction (Figure 4B). GroEL was also detected as an indicator of the cytosolic fraction. As described above, Rv2145c induced relatively higher IL-10 production in macrophages, and some mycobacterial proteins involved in cell wall biogenesis are responsible for Mtb virulence and contribute to its ability to grow in macrophages (28, 29). Therefore, we determined whether Rv2145c could affect Mtb growth in macrophages. Mtb-infected BMDMs were stimulated with LPS, Rv2145c, and Ag85B, which were used as unrelated control Mtb antigens, and then intracellular survival was determined. As shown in Figure 4C, intracellular Mtb growth was significantly higher in Rv2145c-stimulated BMDMs than in untreated or LPS- or Ag85B-treated BMDMs. In addition, Rv2145c-mediated enhancement of intracellular Mtb growth was observed in BMDMs from WT and TLR2^-/- mice but not in BMDMs from TLR4^-/- mice (Figure S2). Next, we tested whether Rv2145c basically expressed in the Mtb cell wall could affect bacterial survival by using an anti-Rv2145c antibody. Intracellular Mtb growth was significantly reduced in BMDMs pretreated with the anti-Rv2145c antibody before Mtb infection compared with that in untreated or isotype control-treated cells (Figure S3). To further confirm the role of Rv2145c in bacterial survival, Rv2145c was overexpressed in nonpathogenic Mycobacterium smegmatis (M. smegmatis) by using the pVV16 shuttle vector. We confirmed that the recombinant M. smegmatis strain expressed Rv2145c, and there was no significant difference in vitro growth rates between the Rv2145c-expressing recombinant strain (Ms_Rv2145c) and the vector control strain (Ms_vector) (Figure S4). The vector control M. smegmatis was more rapidly cleared in macrophages than the Rv2145c-expressing strain (Figure 4D). Taken together, these data suggest that Rv2145c may play an important role in Mtb survival in macrophages.

STAT3 pathways are involved in the survival of Mtb in macrophages (14), and Rv2145c induced the phosphorylation of STAT3 from 12 h after stimulation in BMDMs (Figure 3A). Therefore, we investigated the effects of Rv2145c on STAT3 phosphorylation in macrophages during Mtb infection. As shown in Figure 5A, phosphorylation of STAT3 was weakly induced in BMDMs infected with Mtb at 24 and 48 h but was strongly enhanced by Rv2145c treatment. STAT1 is required for intracellular signaling and cellular responses to IFN-γ, and IFN-γ-mediated NO production is an intracellular Mtb-killing mechanism in macrophages (30, 31). Rv2145c did not enhance STAT1 activation and iNOs expression, but suppressed STAT1 activation and iNOs expression mediated by Mtb at 48 h (Figure 5A). M. smegmatis expressing Rv2145c also strongly induced activation of STAT3 but not STAT1 activation and iNOs expression compared to those with the vector control strain (Figure 5B). In addition, confocal microscopy assays showed that nuclear translocation of phosphorylated STAT3 in Mtb-infected BMDMs was increased by Rv2145c treatment (Figure 5C), and this translocation was also higher in cells infected with Rv2145c-expressing M. smegmatis than in cells infected with the vector control strain (Figure 5D).

We next investigated whether Rv2145c-mediated STAT3 activation could affect Mtb growth in macrophages. As expected, a STAT3 inhibitor (S31-201) completely suppressed Rv2145c-mediated STAT3 phosphorylation and enhanced the phosphorylation of STAT1 and iNOs expression in Mtb-infected macrophages (Figure 6A). Similarly, S31-201 inhibited STAT3 activation and enhanced STAT1 activation and iNOs expression in macrophages infected with Rv2145c-expressing M. smegmatis, and the same reactive pattern was observed in macrophages infected with the vector control strain (Figure 6B). Pretreatment with S31-201 significantly suppressed intracellular Mtb growth compared to that with the DMSO control (SC; solvent control), abrogated Mtb growth enhancement induced by Rv2145c (Figure 6C), and more prominently inhibited the growth of Rv2145c-expressing M. smegmatis than the vector control strain in macrophages (Figure 6D). Rv2145c induced significant IL-10, IL-6, and TNF-α production in Mtb-infected macrophages when compared to that in Mtb only-infected cells (Figure 6E). Interestingly, pretreatment S31-201 significantly inhibited Rv2145c-mediated IL-10 production but enhanced Rv2145c-mediated IL-6 and TNF-α production compared to those in untreated cells (Figure 6E). BMDMs infected with Rv2145c-expressing M. smegmatis produced significantly higher IL-10 and lower IL-6 and TNF-α than vector control strain-infected cells, and pretreatment with S31-201 significantly inhibited IL-10 production induced by Rv2145c-expressing M. smegmatis, but enhanced IL-6 and TNF-α production when compared to that of untreated BMDMs (Figure 6F). There was no difference in IL-10 production induced by the vector control between S31-201-treated and untreated cells. To further confirm the role of STAT3, BMDMs transfected with siSTAT3 were infected with Mtb and then stimulated with Rv2145c. We first confirmed that Mtb- or Rv2145c-induced STAT3 activation in Mtb-infected macrophages was abrogated by transfection with siRNA specific to STAT3 (siSTAT3) (Figure S5). Similar to STAT3 inhibition, STAT3 gene silencing led to a significant reduction in Mtb growth compared to that in siCON-transfected cells, abrogation of Rv2145c-mediated enhancement of Mtb growth (Figure 6G), and more rapid clearance of Rv2145c-expressing M. smegmatis compared to that in siCON-transfected cells (Figure 6H). In addition, transfection of siSTAT3 resulted in a significant reduction in Rv2145c-mediated IL-10 production and enhancement of Rv2145c-mediated IL-6 and TNF-α production in Mtb-infected BMDMs (Figure 6I). The same patterns in cytokine production induced by Rv2145c-expressing M. smegmatis were observed by transfection with siSTAT3 (Figure 6J). Taken together, these results suggest that STAT3 activation is required for Rv2145c-mediated Mtb growth enhancement.

Because STAT3 inhibition resulted in a significant reduction in Rv2145c-mediated IL-10 production and an increase in Rv2145c-mediated IL-6 and TNF-α production, we next investigated the role of IL-10 in Rv2145c-induced Mtb growth enhancement. First, we determined the neutralizing effect of IL-10 on Rv2145c activity. Mtb-infected BMDMs were pretreated with anti-IL-6 or anti-IL-10 antibodies prior to adding Rv2145c. IL-10 and IL-6 induced by Rv2145c were undetected after pretreatment with anti-IL-10 and anti-IL-6 antibodies, respectively, and Rv2145c-mediated IL-6 and TNF-α production was significantly increased by pretreatment with the anti-IL-10 antibody in Mtb-infected BMDMs, whereas Rv2145c-mediated IL-10 production was significantly increased by pretreatment with the anti-IL-6 antibody (Figure 7A). However, anti-IL-10 or anti-IL-6 antibodies did not affect intracellular Mtb growth or enhance Rv2145c-mediated Mtb growth in BMDMs (Figure 7B). Next, we hypothesized that Rv2145c might regulate the IL-10 receptor (IL-10R) or IL-6 receptor (IL-6R) signaling pathways. The anti-IL-10R antibody did not affect Rv2145c-mediated IL-10 production in Mtb-infected BMDMs but induced a significant increase in Rv2145c-mediated IL-6 and TNF-α production compared to that in isotype-treated cells (Figure 7C). The anti-IL-6R antibody did not induce any change in Rv2145c-mediated cytokine production. Interestingly, the anti-IL-10R antibody, but not anti-IL-6R antibody, significantly inhibited intracellular Mtb growth and abrogated Rv2145c-mediated Mtb growth enhancement (Figure 7D). These results suggest that the increase in intracellular bacterial survival induced by Rv2145c is closely related to IL-10R signaling in macrophages. Therefore, we next investigated the effect of Rv2145c on the expression of cytokine receptors in Mtb-infected macrophages. Rv2145c induced a significant increase in the expression of IL-10R, which was abolished by pretreatment with the anti-IL-10R antibody (Figure 7E). However, Rv2145c did not affect IL-6R expression in Mtb-infected macrophages. Similarly, the STAT3 inhibitor abolished the enhancement of Rv2145c-mediated IL-10R expression (Figure 7F). There was no difference in the expression of IL-6R regardless of the presence or absence of S31-201. These results indicated that Rv2145c-mediated IL-10R expression via the STAT3 pathway leads to the enhancement of Mtb growth in macrophages.

Rv2145c (Wag31) is a homolog of DivIVA (a cell division and cell shape protein), which regulates cell morphology in gram-positive bacteria (32) and contains two coiled-coil domains on the N-terminal side (32-59 aa) and C-terminal side (166-193 aa), a DivIVA functional site (3-63 aa), and an N-terminal-membrane binding domain. Therefore, we produced two truncated Rv2145c proteins: the N-terminal part (1-90 aa, domain 1; D1) and the C-terminal part (91-260 aa, domain 2; D2) (Figure S6A). Domain 2 of Rv2145c induced significantly higher IL-10, IL-6 and TNF-α production than domain 1 of Rv2145c and produced amounts comparable to those of full-length Rv2145c protein, at a concentration of 10 μg/ml (Figure S6B). We also confirmed that there was no endotoxin contamination in the purified recombinants (Figure S6C). Furthermore, domain 2 but not domain 1 significantly enhanced intracellular Mtb growth in BMDMs compared to that in the medium control (Figure S6D). These results suggest that the DivIVA functional active site that regulates cell morphology is not involved in the modulation of intracellular survival mediated by Rv2145c.

To estimate the role of Rv2145c during bacterial infection, the mice were infected with Rv2145c-expressing M. smegmatis and vector control strains. As shown in Figure 8A, clearance of Rv2145c-expressing M. smegmatis in the lung and spleen was significantly delayed compared with that of the vector control strain. All mice infected with the vector control strain survived, but all mice infected with Rv2145c-expressing M. smegmatis died by 8 days post-infection (Figure 8B). Histopathologic findings showed that pathologic lesions were prominently more severe in the lungs of mice infected with Rv2145c-expressing M. smegmatis than in those of mice infected with the vector control strain (Figure 8C), and AFB counts were higher in tissue sections of lungs infected with Rv2145c-expressing M. smegmatis than in those of the vector control strain-infected mice (Figure 8D).

Finally, we tested whether IL-10R signaling was involved in Rv2145c-mediated bacterial control in the in vivo system. The bacterial loads in the lungs and spleens of the mice infected with Rv2145c-expressing M. smegmatis were significantly reduced by administration of an anti-IL-10R antibody compared with those in the mice injected with an isotype antibody (Figure 8E). The growth of the M. smegmatis vector control in mice injected with the anti-IL-10R antibody was lower than that in mice injected with the isotype antibody, but there was no significant difference between the groups. Overall, these results indicate that overexpression of Rv2145c makes nonpathogenic mycobacteria more virulent in mice and that IL-10R signaling is intimately related.