Interleukin 10 suppresses lysosome-mediated killing of Brucella abortus in cultured macrophages

Brucella abortus is a Gram-negative zoonotic pathogen for which there is no 100% effective vaccine. Phagosomes in B. abortus–infected cells fail to mature, allowing the pathogen to survive and proliferate. Interleukin 10 (IL10) promotes B. abortus persistence in macrophages by mechanisms that are not fully understood. In this study, we investigated the regulatory role of IL10 in the immune response to B. abortus infection. B. abortus–infected macrophages were treated with either IL10 siRNA or recombinant IL10 (rIL10), and the expression of phagolysosome- or inflammation-related genes was evaluated by qRT-PCR and Western blotting. Phagolysosome fusion was monitored by fluorescence microscopy. We found that the synthesis of several membrane-trafficking regulators and lysosomal enzymes was suppressed by IL10 during infection, resulting in a significant increase in the recruitment of hydrolytic enzymes by Brucella-containing phagosomes (BCPs) when IL10 signaling was blocked. Moreover, blocking IL10 signaling also enhanced proinflammatory cytokine production. Finally, concomitant treatment with STAT3 siRNA significantly reduced the suppression of proinflammatory brucellacidal activity but not phagolysosome fusion by rIL10. Thus, our data provide the first evidence that clearly indicates the suppressive role of IL10 on phagolysosome fusion and inflammation in response to B. abortus infection through two distinct mechanisms, STAT3-independent and -dependent pathways, respectively, in murine macrophages.

Brucella spp. are facultative intracellular Gram-negative bacteria that cause brucellosis in a variety of mammalian hosts; particularly, they cause more than 500,000 new human cases annually (1). They can prevent phagosome maturation by not fully understood mechanisms, leading to successful survival within professional and non-professional phagocytes (2,3). Vaccination seems to be a predominant manner for the control of infectious diseases; however, there is no 100% efficacious vaccine for brucellosis so far. Thus, identification of host defense mechanisms is essential to design rational approaches to eliminate brucellosis (4).
Interleukin 10 (IL10) is a pleiotropic immunomodulatory cytokine that is mainly produced by activated Th2 cells, monocytes, macrophages, and B cells. It was reported earlier to be a key inhibitor of inflammation and Th1-dependent cell-mediated immunity, especially production of proinflammatory cytokines such as interleukin 1␤ (IL1␤), tumor necrosis factor (TNF), interleukin 6 (IL6), granulocyte-macrophage colonystimulating factor (GM-CSF), chemokines such as macrophage inflammatory peptide 1␣ (MIP-1␣), generation of nitric oxide (NO), and up-regulation of surface antigen expression (MHC class II, CD80, and CD86) in LPS-activated macrophages (5,6). Additionally, various studies have also shown that IL10 plays a suppressive role in host response to Brucella abortus when the survival of intracellular B. abortus was markedly decreased in the absence of IL10 (7,8). A recent study indicated that IL10 is beneficial for avoiding the phagolysosome fusion of intracellular Brucella, resulting in prolonged persistence of Brucella in macrophages (9). However, the underlying mechanisms of how IL10 could suppress the lysosome-mediated killing of Brucella in host cells still need to be clarified.
Phagolysosome fusion with possibly hundreds of proteins is the most crucial effector in host responses against bacterial infection. To date, lysosomal membrane glycoproteins 1 and 2 (LAMP1 and LAMP2) 2 and members of the RAS oncogene family (RABs) were the only known regulators to be important in controlling this process (10 -12). Thus, our aim in the present study was to identify which regulators are controlled by IL10 during infection. Furthermore, expression of different hydrolytic enzymes was also evaluated because sufficient recruitment of lysosomal enzymes is essential to restrict bacteria within phagosomes. Our findings revealed that IL10 inhibits lysosome-mediated killing through governing a variety of important proteins such as the RAB family, LAMP1, LAMP2, and the cathepsin (CTS) family through a STAT3-independent pathway in B. abortus-infected macrophages.

IL10 represses brucellacidal activity through promoting bacterial ability to avoid late endosomes in macrophages
To monitor the impact of B. abortus infection on IL10 expression, we first used RT-PCR to evaluate the transcriptional profile of Il10 mRNA in macrophages at different time points of infection. Interestingly, as shown in Fig. 1A, B. abortus infection markedly induced the transcription of Il10 ϳ2.2-, 1.8-, and 4.5-fold at 4, 24, and 48 h postinfection, respectively, as compared with uninfected cells. Additionally, the production of IL10 protein measured by indirect ELISA was also consistently obtained with a marked increase of intracellular and secreted IL10 after infection (Fig. 1B).
To investigate the roles of IL10 in the brucellacidal activity, RAW 264.7 cells were pretreated with IL10 siRNA prior to infection with B. abortus. As shown in Fig. 1C, inhibition of IL10 signaling caused a significant decrease of intracellular Brucella survival that paralleled the observation of colocalization of LAMP1 and Brucella-containing phagosomes (BCPs). In the IL10-blocked cells, colocalization was ϳ1.4 times higher as compared with control cells (Fig. 1, D and E). Taken together, these data clearly indicated that IL10 prevents the recruitment of lysosomes by BCPs, leading to the persistence of bacteria within macrophages.

Interference of IL10 signaling alters normal acquisition of membrane-trafficking regulators by B. abortus phagosomes
Fusion of phagosomes with late endocytic organelles (lysosomes) is a central effector of antimicrobial immunity that requires hundreds of proteins, and this process is suppressed by IL10; thus, identification of regulators mediated by IL10 could provide insights into its underlying mechanism during Brucella infection. Based on previous reports on phagolysosome regulation (12,13) and our previous study on microarray analysis of gene expression profiling of B. abortus-infected macrophages, 3 30 trafficking regulators of interest were selected. These transcripts were then assessed by RT-PCR in cells with or without treatment with IL10 siRNA at different phases of infection. As shown in Fig. 2A, inhibition of IL10 signaling caused the induction of Lamp1, Lamp2, and Rab34 mRNA at the early phase; however, no influence of IL10 on these genes was obtained at 24 h postinfection (Fig. 2B). Interestingly, transcripts of Lamp1, Lamp2, Rab5a, Rab7, Rab20, Rab22a, Rab34, and Stx11 were found to be remarkably increased in IL10-deficient cells in comparison with the controls at late infection (Fig. 2C).
To complement these data, we checked the expression of selected proteins by Western blotting at 4 and 48 h postinfection. As expected, the expression of LAMP2, RAB34, and

Functional characterization of IL10 in B. abortus infection
RAB22A proteins was shown to be consistent with RT-PCR results when blocking IL10 signaling significantly increased these proteins compared with the control (Fig. 2D). Furthermore, evaluation of acquisition of these trafficking regulators by B. abortus phagosomes at 48 h postinfection revealed that inhibition of IL10 in macrophages significantly induced the fraction of Brucella phagosomes labeled for LAMP2 (Fig. 3, A and B) and RAB22A (Fig. 3, C and D) at late infection. Altogether, our findings suggest that the effect of suppression of IL10 on phagolysosome fusion might be through inhibition of LAMP1, LAMP2, RAB5A, RAB7, RAB20, RAB22A, RAB34, and STX11 during B. abortus infection.

IL10 mediates the expression and recruitment of hydrolytic enzymes during B. abortus infection in macrophages
The acquisition of acidic lysosomal enzymes by pathogencontaining phagosomes and their activation result in efficient killing of intracellular pathogens (12,14). Thus, we hypothesized that IL10 also manipulates the expression of these hydrolytic enzymes. To address this hypothesis, transcriptional profiling of 30 lysosomal enzymes was initially assessed by RT-PCR at different times. Intriguingly, the expression of all enzymes was independent of the deficiency of IL10 at early and middle phases of infection (Fig. 4, A and B); however, a number of genes, including Hexb, Gla, Ctsa, Ctsd, Ctsl, Man1a, and Man2a1, were uncovered to be negatively controlled by IL10 at the late stage (Fig. 4C). To validate these data, we next evaluated the expression of proteins encoded by these genes by Western blotting at 4 and 48 h postinfection. Consistent with the observation from RT-PCR, the marked induction of HEXB and CTSD but not CTSZ proteins in IL10-deficient cells was only observed at 48 h postinfection, whereas no difference was observed during earlier infection as compared with the control (Fig. 4D).
The observation that inhibition of IL10 induced the phagolysosome fusion suggested that not only the expression but the delivery of lysosomal enzymes to Brucella phagosomes might also be increased in IL10-lacking cells. Therefore, we monitored the fraction of B. abortus phagosomes that could be labeled for CTSA and CTSD markers. At 48 h after infection, the colocalization of B. abortus phagosomes with both CTSA (Fig. 5, A and B) and CTSD (Fig. 5, C and D) was notably elevated in IL10-deficient cells relative to controls, suggesting that IL10 controls the recruitment of CTSA and CTSD by BCPs. However, these results also raised the question whether this effect is general to all lysosomal enzymes, including those that were not altered by IL10. To answer this question, colocalization of CTSZ was observed by microscopy. However, the percentage of colocalization for these proteins was not influenced by IL10 (Fig. 5

, E and F).
To clarify whether IL10 could also have an inhibitory effect on the expression of phagolysosome-related genes in normal conditions, we treated RAW 264.7 cells with IL10 siRNA followed by incubation for 2 days. The expression of 14 representative proteins was assessed by qRT-PCR; however, our data showed that no difference between IL10 siRNA-treated cells and control was observed (Fig. S1A), suggesting that the inhibitory effect of IL10 only occurs in the B. abortus-infected condition. Furthermore, the above data suggested that IL10 might play a similar role in primary mouse macrophages. To address this question, we collected and differentiated bone marrowderived macrophages (BMMs) from BALB/c mice and treated them with recombinant IL10 (rIL10) during B. abortus infection. As expected, treatment with rIL10 induced B. abortus persistence within BMM cells at late infection (Fig. S1B). In parallel, down-regulation of trafficking regulators and lysosomal enzymes was also observed in rIL10-treated cells at 4 h postin-

Functional characterization of IL10 in B. abortus infection
fection (pi) compared with control ( Fig. S1, C and D). Moreover, reduced colocalization of LAMP1 with BCPs was observed when IL10 signaling was enhanced (Fig. S1E), suggesting that IL10 also suppresses a phagolysosome fusion event to promote bacterial survival in primary mouse macrophages. However, the target phagolysosomal genes of IL10 signaling were different between BMM and RAW 264.7 cells, which may result from the different regulatory mechanisms activated in the responses to pathogens in these cells (15). In summary, these results are the first evidence that clearly shows the regu-

Functional characterization of IL10 in B. abortus infection
latory role of IL10 in the lysosomal-mediated killing of B. abortus in murine macrophages.

Repressive effect of IL10 on the phagolysosome fusion in Brucella-infected macrophages is through STAT3-independent pathway
Binding of IL10 to the extracellular domain of IL10 receptor activates various subsequent signaling pathways; however, JAK1/STAT3 is the best understood pathway to be mainly responsible for subsequent transduction (16,17). Thus, we hypothesized that the suppressive effect of IL10 on the phagolysosome event is through the JAK1/STAT3 pathway. To test this hypothesis, we first used fluorescence microscopy to monitor the translocation of STAT3 to the nucleus upon treatment of B. abortus-infected RAW 264.7 cells with either rIL10 or IL10 siRNA. Interestingly, the translocation of STAT3 into the nucleus was remarkably enhanced when the infected cells were treated with rIL10, whereas the reverse was observed with IL10 siRNA treatment (Fig. 6, A and B). Likewise, the results of evaluation of the activation of STAT3 by Western blot assay were also consistent with the microscopy observation (Fig. 6C).
To determine the role of the JAK1/STAT3 pathway in IL10 signaling, the infected cells were concomitantly treated with rIL10 and STAT3 siRNA, and the transcriptional levels of IL10regulated trafficking regulators and hydrolytic enzymes were then assessed at 48 h postinfection. Surprisingly, the transcriptional levels of all IL10-regulated trafficking regulators and hydrolytic enzymes were found to be unchanged when the STAT3 pathway was blocked (Fig. 6, D and E). In addition, the acquisition of LAMP2 by BCP was not different between STAT3-blocked and -producing rIL10-treated cells (Fig. 6F), suggesting that the JAK1/STAT3 pathway is not responsible for the suppressive effect of IL10 on phagolysosome fusion in B. abortus-infected macrophages.

IL10 represses expression of proinflammatory cytokines by up-regulating Sosc3 expression during B. abortus infection
Although our data clearly indicated that the inhibitory role of IL10 on lysosomally mediated killing is independent of the JAK1/STAT3 pathway, we still assessed STAT3 function in the inflammatory response upon infection when the IL10 pathway was inhibited. For this, we evaluated the expression of Socs3 and

Functional characterization of IL10 in B. abortus infection
proinflammatory cytokines (Il6, Tnf, Mcp1, Il1a, and Il1b) by RT-PCR and indirect ELISA in infected cells concomitantly treated with rIL10 and STAT3 siRNA. Interestingly, the addition of rIL10 caused a significant increase of Socs3 that was accompanied by a decrease of Il6, Tnf, and Il1a but not Mcp1 and Il1b mRNA levels at 48 h postinfection. However, this consequence was blocked when the cells were concomitantly treated with STAT3 siRNA (Fig. 7A). In parallel, the presence of secreted cytokines in culture supernatant was also shown to be consistent with the observation by RT-PCR (Fig. 7B). These findings suggest that IL10 mediated the activation of STAT3/ SOCS3 that inhibits the inflammatory response in B. abortusinfected macrophages.
Our findings showed two distinct regulatory mechanisms of IL10 in Brucella-infected macrophages that are through STAT3-dependent and -independent pathways, leading to the question of what actual role STAT3 plays in anti-Brucella suppression by IL10. For this, we concomitantly treated RAW 264.7 cells with rIL10 and STAT3 siRNA and then evaluated bacterial survival. Surprisingly, treatment with STAT3 siRNA markedly reduced the rIL10-promoted bacterial persistence in RAW 264.7 cells (Fig. 7C), suggesting that STAT3 is required for the antimicrobial suppression of IL10 during B. abortus infection in macrophage cells. Taken together, our data clearly indicate that IL10 regulates the proinflammatory anti-Brucella immunity through controlling the STAT3/SOCS3 pathway in RAW 264.7 macrophages.

Discussion
B. abortus, a causative agent of brucellosis, is one of the pathogens that have acquired the ability to survive and replicate within host cells by mechanisms that still needed to be eluci-

Functional characterization of IL10 in B. abortus infection
dated. To date, several studies have shown that the fusion of bacteria-containing phagosomes with late endosomes/lysosomes is the most important innate immune effector against intracellular Brucella; however, how this process is carried out by the host and subverted by bacteria is not fully understood. Thus, identification of potential molecules that may control this process will provide insights into rational therapeutic design for brucellosis elimination.
In agreement with previous observations, we also showed that B. abortus infection markedly induces expression of IL10 (7,9). In this study, we proved that this induction is beneficial for survival of intracellular Brucella because suppression of IL10 signaling by siRNA treatment enhanced killing and restricted bacteria within macrophages. Consistent with a previous study (9), our study also showed that IL10 promotes survival of intracellular Brucella by increasing bacterial ability to prevent the recruitment of lysosomes by BCPs, and this regulatory role of IL10 was found at both early and late stages of infection.
It has been shown that IL10 negatively regulates the expression of three regulators, Lamp1, Lamp2, and Rab34, at the early phase, whereas regulators at the late stage were Lamp1, Lamp2, Rab5a, Rab7, Rab20, Rab22a, Rab34, and Stx11. The observed up-regulation at the mRNA level resulted in a higher content of these proteins, suggesting that IL10 mediates these proteins to control the fusion of phagosomes with late endosomes/lysosomes during B. abortus infection. These data could be rationalized by the findings that eight potent membrane-trafficking molecules are subverted by IL10 during infection.
To date, LAMP1 and RAB7 were the only regulators that have been proven to be crucial in regulating the fusion of BCPs with lysosomes (18), whereas the roles of other regulators are unknown. Our results showed that IL10 negatively regulates RAB5A, RAB20, and RAB22A during infection; however, these proteins have been found to be mainly associated to phagosomes containing intracellular pathogens such as Listeria and Mycobacterium and to stimulate the maturation of these phagosomes at early infection (13), leading to the question of

Functional characterization of IL10 in B. abortus infection
whether these functional proteins also play undescribed roles in late Brucella infection. The other particularly interesting result is that LAMP2, RAB34, and STX11 are also highly likely to regulate the fusion of phagosomes and lysosomes in macrophages. LAMP2 was demonstrated to have overlapping functions with LAMP1 in recruitment of RAB7, movement toward the microtubule-organizing center, and subsequent fusion with lysosomes (11). Likewise, RAB34 was also reported to be required for the fusion of phagosomes and lysosomes because a deletion mutant of this gene markedly reduced fusion ability (13,19). STX11 is another membrane-trafficking regulator that is suppressed by IL10; however, the general function of this protein in the phagolysosome event is still unknown. Taken together, these findings and our results showing that IL10 subverts these proteins followed by inhibition of phagolysosome fusion suggest the potential of these trafficking regulators to govern phagolysosome fusion.
Likewise, several hydrolytic enzymes, including HEXB, GLA, CTSA, CTSD, CTSL, MAN1A, and MAN2A1, were clearly shown to increase when IL10 signaling was inhibited at the late stage of infection. Different cathepsins, including CTSB, CTSD, CTSG, CTSL, and CTSS, are known to interact and contribute to killing intracellular Mycobacterium tuberculosis (20,21), Mycobacterium bovis (22), Streptococcus pneumoniae (23), and Listeria monocytogenes (24). Additionally, in macrophages, HEXB was also proven to protect them against Mycobacterium marinum (25). Thus, the observation of an induced fraction of phagosomes that are labeled by HEXB and CTSD in BCPs in parallel with an elevated fraction of phagosomes labeled with LAMP1 suggests that IL10 may mediate these enzymes to inhibit lysosome-mediated killing of Brucella within macrophages, and this also opens up the discussion on the roles of these lysosomal enzymes in killing Brucella.
Conversely, our observation is similar to previous reports on the anti-inflammatory role of IL10 in which blocking of IL10 up-regulated production of inflammatory cytokines during infection. However, to date, only TNF has been recently proven to participate in brucellacidal activity (26). The actual roles of other proinflammatory cytokines in IL10 signaling and brucellacidal immunity have yet to be investigated. Additionally, one of the striking findings in our study is that IL10 inhibits phagolysosome fusion and proinflammatory brucellacidal activity through two distinct signaling mechanisms in which STAT3 importantly uses the proinflammatory brucellacidal suppression effect without phagolysosome interference. Thus, further investigations of the JAK/STAT pathways might reveal the relationship of inflammation and anti-Brucella activity, which could be useful for further understanding host resistance to Brucella infection.
In summary, our findings reveal a possible novel role of IL10 to suppress the synthesis and delivery of molecules involved in phagolysosome fusion, which prevents killing of intracellular B. abortus in macrophages. In addition, further investigation of our identified molecules (eight trafficking regulators and seven lysosomal enzymes regulated by IL10) might provide insights into how this process operates.

Reagents
Mouse IL10 and STAT3 siRNAs and rat anti-LAMP1 and -LAMP2; mouse anti-RAB34, -RAB22A, and -CTSZ; and goat anti-CTSA antibodies were obtained from Santa Cruz Biotechnology. Goat anti-CTSD antibody was obtained from R&D Systems. Texas Red-rabbit anti-goat IgG antibodies and the mouse IL10 ELISA kit were purchased from Abcam. Mouse IL10 recombinant protein and rabbit phospho-STAT3 polyclonal, rabbit anti-HEXB, and FITC-goat anti-mouse IgG antibodies were obtained from Thermo Fisher Scientific. Texas Red-goat anti-rat IgG antibody and Lipofectamine RNAiMAX were purchased from Life Technologies. FITCconjugated goat anti-rabbit IgG antibody was obtained from Sigma-Aldrich.

Bacterial strain and cell culture
B. abortus 544 biovar 1 strain was routinely cultured in Brucella broth (BD Biosciences) at 37°C until stationary phase. The murine macrophage RAW 264.7 cells were grown at 37°C in 5% CO 2 atmosphere in RPMI 1640 medium containing 10% heat-inactivated fetal bovine serum (FBS) with or without 100 units/ml penicillin and 100 g/ml streptomycin.

Bone marrow-derived macrophage preparation
BMMs from female BALB/c mice were prepared as described previously (27). Briefly, bone marrow cells were collected and cultured in BMM medium containing L-cell-conditioned medium for 5 days at 37°C in 5% CO 2 . An equal volume of fresh BMM medium without antibiotics was added, and the cells were incubated further for 5 days. After 10 days of incubation, BMMs were washed three times with PBS and incubated with fresh RPMI 1640 medium containing 10% (v/v) heat-inactivated FBS for further experiments.

Bacterial infection and intracellular replication assay
This assay was performed as described previously (3). Briefly, macrophages (10 6 cells) were seeded in a 96-well plate and incubated for 24 h at 37°C in 5% CO 2 . The cells were then infected with 10 7 cfu of the virulent B. abortus for 1 h at 37°C in 5% CO 2 . RPMI 1640 medium containing 10% (v/v) FBS and gentamicin (30 g/ml) were subsequently added to kill extracellular bacteria. At 2, 24, and 48 h postinfection, cells were lysed and plated on Brucella agar plates for cfu determination. Additionally, the culture supernatant and total proteins or RNA from macrophages were also obtained at different time points.

RNA interference
RAW 264.7 cells were grown to 50% confluence on 6-, 12-, or 96-well plates and transfected with siRNA directed against IL10 using Lipofectamine RNAiMAX. The cells were incubated for 24 h at 37°C and 5% CO 2 prior to performing the intracellular growth assay or protein or RNA isolation. The same concentrations of negative control siRNAs were used throughout as controls. Knockdown efficiency was quantified using RT-PCR.

RNA extraction
Total RNA was isolated from RAW 264.7 cells (uninfected or infected with B. abortus) at different time points using a Qiagen RNeasy kit. DNA was removed before final elution of the RNA sample using the Qiagen on-column DNase digestion protocol.

Functional characterization of IL10 in B. abortus infection
sion profiles were normalized with respect to ␤-actin. -Fold increase of each gene was calculated using the 2 Ϫ⌬⌬CT method.

Western blot assays
The lysates of cells were identified by Western blot assay as described previously (4,12). Briefly, the proteins were boiled for 5 min at 100°C in 2ϫ SDS buffer and subjected to 10% SDS-PAGE. Separated proteins were then transferred onto Immobilon-P membranes (Millipore) using a semidry electroblot assembly (Bio-Rad). Membranes were blocked with 5% skim milk (Difco) and subsequently incubated with primary antibodies (1:5,000 -1:1,000 dilution) in blocking buffer. After washing with PBS with 0.05% Tween 20, membranes were incubated with HRP-conjugated goat anti-mouse IgG antibody (1:10,000 dilution; Sigma) in blocking buffer. The proteins were detected with ECL solution (Thermo Fisher Scientific).

LAMP1, LAMP2, RAB22A, CTSA, CTSD, and CTSZ staining
Colocalization of BCPs with LAMP1, LAMP2, RAB22A, CTSA, CTSD, and CTSZ was performed as reported previously (26,28). Briefly, RAW 264.7 cells were treated with IL10 siRNA prior to infection. The infected cells were incubated for 2 (LAMP1) or 48 h (LAMP2, RAB22A, CTSA, CTSD, and CTSZ), fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and blocked with blocking buffer (2% goat serum in PBS). The samples were stained with primary antibodies that were diluted 1:100 in blocking buffer followed by secondary incubation with Texas Red-goat anti-rat IgG or Texas Red-rabbit anti-goat IgG (1:1,000) in blocking buffer. The samples were stained with anti-B. abortus rabbit serum and FITC-conjugated anti-rabbit IgG to identify the bacteria and placed in mounting medium. Fluorescence images were captured using a laserscanning confocal microscope (Olympus FV1000, Japan) and processed using FV10-ASW Viewer 3.1 software. 100 cells were randomly selected, and the percentage of colocalization of these proteins with the BCPs was determined.

ELISA
The levels of IL10, TNF, IL6, IL1␤, IL1␣, and MCP1 in culture supernatants were determined by sandwich ELISA performed in accordance with the manufacturer's instructions (Thermo Fisher Scientific).

Statistical analysis
The data are expressed as the mean Ϯ (S.D. Student's t test was used to statistically compare the groups. Results with p Ͻ 0.05 were considered significantly different.