Protective role of G-CSF in dextran sulfate sodium-induced acute colitis through generating gut-homing macrophages

Cytokine 78 (2016) 69–78

Contents lists available at ScienceDirect

Cytokine journal homepage: www.journals.elsevier.com/cytokine

Protective role of G-CSF in dextran sulfate sodium-induced acute colitis through generating gut-homing macrophages q Shahab Meshkibaf a,b, Andrew J. Martins a,1, Garth T. Henry a, Sung Ouk Kim a,b,⇑ a Department of Microbiology and Immunology and Infectious Diseases Research Group, Siebens-Drake Research Institute, University of Western Ontario, London, Ontario N6G 2V4, Canada b Center for Human Immunology, University of Western Ontario, London, Ontario N6G 2V4, Canada

a r t i c l e

i n f o

Article history: Received 27 May 2015 Received in revised form 19 October 2015 Accepted 29 November 2015 Available online 10 December 2015 Keywords: G-CSF G-CSF receptor Colitis Macrophage

a b s t r a c t Granulocyte colony-stimulating factor (G-CSF) is a pleiotropic cytokine best known for its role in promoting the generation and function of neutrophils. G-CSF is also found to be involved in macrophage generation and immune regulation; however, its in vivo role in immune homeostasis is largely unknown. Here, we examined the role of G-CSF in dextran sulfate sodium (DSS)-induced acute colitis using G-CSF receptor-deficient (G-CSFR
/
) mice. Mice were administered with 1.5% DSS in drinking water for 5 days, and the severity of colitis was measured for the next 5 days. GCSFR
/
mice were more susceptible to DSS-induced colitis than G-CSFR+/+ or G-CSFR
/+ mice. G-CSFR
/
mice harbored less F4/80+ macrophages, but a similar number of neutrophils, in the intestine. In vitro, bone marrow-derived macrophages prepared in the presence of both G-CSF and macrophage colony-stimulating factor (M-CSF) (G-BMDM) expressed higher levels of regulatory macrophage markers such as programmed death ligand 2 (PDL2), CD71 and CD206, but not in arginase I, transforming growth factor (TGF)-b, Ym1 (chitinase-like 3) and FIZZ1 (found in inflammatory zone 1), and lower levels of inducible nitric oxide synthase (iNOS), CD80 and CD86 than bone marrow-derived macrophages prepared in the presence of M-CSF alone (BMDM), in response to interleukin (IL)-4/IL-13 and lipopolysaccharide (LPS)/interferon (IFN)-c, respectively. Adoptive transfer of G-BMDM, but not BMDM, protected G-CSFR
/
mice from DSS-induced colitis, and suppressed expression of tumor necrosis factor (TNF)-a, IL-1b and iNOS in the intestine. These results suggest that G-CSF plays an important role in preventing colitis, likely through populating immune regulatory macrophages in the intestine. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Abbreviations: CSF, colony-stimulating factor; DCs, dendritic cells; DSS, dextran sulfate sodium; ELISA, enzyme-linked immunosorbent assay; FIZZ1, found in inflammatory zone 1; G-BMDM, granulocyte colony-stimulating factor-treated bone marrow-derived macrophages; G-CSF, granulocyte colony-stimulating factor; G-CSFR, granulocyte colony-stimulating factor receptor; GM-CSF, granulocyte/ macrophage colony-stimulating factor; H&E, hematoxylin and eosin; IFN-c, interferon-c; IL, interleukin; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; M-CSF, macrophage colony-stimulating factor; BMDM, bone marrowderived macrophages; MFI, mean fluorescent intensity; MPO, myeloperoxidase; PDL2, programmed death ligand 2; qRT-PCR, quantitative real-time polymerase chain reaction; TGF-b, transforming growth factor b; TNF-a, tumor necrosis factor-a. q Summary sentence: G-CSF generates gut-homing macrophages and protects mice from colitis. ⇑ Corresponding author at: Department of Microbiology and Immunology, University of Western Ontario, London, Ontario N6G 2V4, Canada. E-mail address: [email protected] (S.O. Kim). 1 Current address: Systems Genomics and Bioinformatics Unit, Laboratory of Systems Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892, USA.

http://dx.doi.org/10.1016/j.cyto.2015.11.025 1043-4666/Ó 2015 Elsevier Ltd. All rights reserved.

G-CSF is primarily known for its role in the generation, mobilization and function of neutrophils [1–3] which are key innate immune cells protecting against invading microbes [4–7]. G-CSF also plays an important role in regulating both innate and adaptive immune responses through modulating activation of macrophages and dendritic cells (DCs) [8–10], and promoting generation of tolerogenic DCs [11] and regulatory T cells [12]. These dual modalities of G-CSF in immune regulation could be important in maintaining intestinal immune homeostasis, where both avid immune surveillance and tolerance against infiltrating commensal microbes are required. Previous studies have demonstrated that administration of recombinant G-CSF ameliorates colitis in acute and chronic experimental colitis [13,14] and Crohn’s disease patients [15–17], suggesting its beneficial role in immune homeostasis of the intestine. However, the role of endogenous G-CSF in intestinal immune homeostasis is largely unknown.

70

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

Intestinal macrophages are highly populated in the intestinal lamina propria, residing underneath the single epithelial layer, and Peyer’s patches [18,19]. These cells form an intestinal immune barrier for invading microbes, and also orchestrate immune and wound repair responses that prevent inadvertent development of colitis [20–22]. However, in pathologic conditions, monocytes/ macrophages rapidly recruited into the inflamed intestines can mediate inflammatory responses through releasing proinflammatory cytokines or promoting inflammatory T cell differentiation [23,24]. Unlike most tissue macrophages derived from local progenitor cells [25,26], intestinal macrophages are mainly replenished by circulating monocytes originating from bone marrow cells [19,27]. Circulating monocytes released from the bone marrow migrate to the intestine where they differentiate into resident macrophages under the influence of local cytokine milieu [24]. Recruited monocytes or resident macrophages can then be polarized to either classically activated (M1) or alternatively activated (M2) macrophages in response to LPS/IFN-c or IL-4/IL-13, respectively [28–31]. M1 macrophages are characterized as inflammatory macrophages due to their high capacity in producing proinflammatory cytokines, such as TNF-a, IL-1b, IL-12, IL-23 and iNOS [29,32]. In contrast, M2 macrophages highly express regulatory cytokines, such as TGF-b and IL-10 [32], and various markers, including arginase I [33], Ym1 [34], FIZZ1 [34,35], CD71 (also known as transferrin receptor) [36–38], CD206 [38] and PDL2 [39–41], which are involved in anti-inflammatory, woundhealing or anti-parasitic responses [28,29,31]. In addition to the generation of neutrophils, G-CSF was also shown to be involved in the generation of macrophages. It was shown to support the survival of macrophage progenitor cells in in vitro bone marrow cell cultures [1,42] and mice deficient in G-CSF harbor 50% reduced circulating monocytes, compared to wild-type mice [43]. Recently, we showed that G-CSF plays an important role in the generation of regulatory Gr-1high/F4/80+ macrophages, which are localized mainly in the intestine after adoptive transfer [44]. Here, we further examined the role of G-CSF in intestinal immune homeostasis using a DSS-induced acute colitis model in G-CSFR
/
mice. We found that G-CSFR
/
mice were more susceptible to DSS-induced colitis than G-CSFR+/
or G-CSFR+/+ mice. Macrophages derived in the presence of G-CSF showed enhanced expression of several M2-like markers, such as PLD2, CD71 and CD206, and suppressed LPS/IFN-c-induced expression of iNOS, CD80 and CD86. Adoptive transfer of these macrophages significantly ameliorated the severity of colitis in G-CSFR
/
mice. These results suggest that G-CSF plays a protective role from developing inadvertent colitis, likely through generating gut-homing macrophages.

2. Materials and methods 2.1. Mice C57BL/6j wild-type mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). Homozygote G-CSFR
/
mice with C57BL/6j background were obtained originally from Dr. D. C. Link (Washington University Medical School, St. Louis, MO, USA) and maintained in the animal care and veterinary facility at the University of Western Ontario (London, ON, Canada) under conventional conditions. C57BL/6j and G-CSFR
/
were crossed, and G-CSFR+/
offspring were then crossed to generate G-CSFR+/+, +/
and
/
littermates for use in experimental procedures. G-CSFR+/+, +/
, and
/
were housed in conventional conditions, and were intermixed randomly during experiments such that each would be co-housed with other mice of different

genotypes. Aged-matched mice between 6 and 12 weeks of age were used in all experimental procedures. All experimental protocols were approved by the University of Western Ontario Animal Use Subcommittee, which follows the regulations of the Animals for Research Act (Ontario) and the Canadian Council on Animal Care. 2.2. Reagents, antibodies and cytokines ELISA for TNF-a, IL-1b and IL-10 were obtained from eBioscience (San Diego, CA, USA). LPS from Escherichia coli O111:B4 was obtained from List Biological Laboratories (Campbell, CA, USA). FITC-anti-mouse F4/80, PE-Cy7-anti-mouse Gr-1, PE-antimouse CD71, PE-anti-mouse PDL2, APC-anti-mouse CD206, APC-anti-mouse CD80, and PE-anti-mouse CD86 antibodies were purchased from eBioscience (San Diego, CA, USA). IL-4, IL-13 and IFN-c cytokines were obtained from PeproTech (Rocky Hill, NJ, USA). The murine recombinant M-CSF and G-CSF were obtained from eBioscience (San Diego, CA, USA). 2.3. Dextran sulfate sodium colitis mouse model 1.5% weight/volume DSS (molecular weight: 40,000–50,000 Da, USB Corporation, Cleveland, Ohio) was dissolved in autoclaved drinking water and administered ad libitum to the mice for 5 days. Control mice were given normal drinking water. On day 5, the DSS solution was replaced with water without DSS for an additional 4–5 days. Mice were monitored daily for weight loss, diarrhea and rectal bleeding. Disease activity scoring was performed as follows: weight loss (percentage of starting weight): <10% = 1, <15% = 2, <20% = 3; diarrhea: slightly loose feces and/or slight rectal prolapse = 1, loose feces and/or obvious prolapse = 2, watery diarrhea and/or severe prolapse = 3; bleeding: spotty blood in feces or around anus = 1, slightly bloody feces or moderate blood around anus = 2, bloody stool or severe bleeding around anus = 3; activity: still moving spontaneously but reduced = 1, little spontaneous movement = 2, moves only with gentle touch = 3. The sum of the all values constitutes the colitis score. Mice were euthanized if the total score reached higher than 9 or if more than 20% body weight loss was observed. 2.4. Histology and immunohistochemistry After the mice were sacrificed, the caecum was removed and the remaining colon tissue was cut into 0.5 cm pieces. Tissues were then fixed in 10% neutral buffered formalin, paraffin-embedded, sectioned (5 lm), and stained with hematoxylin and eosin (H&E) for visualization of intestinal tissue damage, or used for immunohistochemical staining for myeloperoxidase (MPO). H&E stained sections were evaluated by a pathologist in a blinded fashion. Histological scoring was performed as follows (modified from (18)): 0 = normal; 1 = lesions that are not transmucosal; 2 = 0–25% of mucosa affected; 3 = 25–50% of mucosa affected; 4 = >75% of mucosa affected. Immunohistochemical staining was performed after heat-induced epitope retrieval by heating slides to 95 °C in 10 mM Tris buffer (pH 10) containing 0.05% Tween20. Staining was performed using the Santa Cruz ABC staining kit (Santa Cruz Biotechnology, Santa Cruz, California) according to the manufacturer’s instructions. Briefly, slides were incubated in 1% hydrogen peroxide in PBS for 5 min to quench endogenous peroxidase activity. Slides were then incubated in 1.5% blocking serum in PBS for 1 h, followed by overnight incubation at 4 °C with 10 lg/ mL anti-MPO antibody (R&D Systems, Minneapolis, Minnesota). Staining was developed the next day and slides were counterstained with hematoxylin. For counting of the number of MPO positive cells per field of view, a single tissue section containing the highest degree of staining on the slide was chosen, and densely

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

stained cells in each field of view at 100 magnification were counted. 2.5. Generation of BMDM and G-BMDM BMDM and G-BMDM were generated as previously described [44]. Briefly, bone marrow cells were harvested from femurs and tibia of G-CSFR+/+ and/or G-CSFR
/
mice using a 25.5-gauge needle and PBS. Red blood cells were lysed using ammonium chloride and EDTA. Bone marrow cells were then cultured in RPMI 1640 medium (Sigma–Aldrich), supplemented with 10% heat-inactivated fetal bovine serum (Sigma–Aldrich), 100 U ml
1 penicillin, 100 lg ml
1 streptomycin (Sigma–Aldrich), 5 mM sodium pyruvate (Sigma–Aldrich), 5 mM MEM non-essential amino acids (Sigma–Aldrich), and M-CSF (20 ng/ml) or M-CSF and G-CSF (10 ng/ml). Cells were then maintained at 37 °C in a humidified atmosphere with 5% CO2 for 7 days. The culture media were replaced with fresh media every 2 days. 2.6. Flow cytometry BMDM and G-BMDM were harvested, washed twice, and labeled with fluorescently conjugated antibodies against PDL2, CD206, CD71, CD80 and CD86. Labeled cells were washed and analyzed with a FACSCalibur flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) using CellQuest software (Becton Dickinson, Franklin Lakes, NJ, USA). Flow cytometry analysis was done using FlowJo software (version 7.6.5; TreeStar, Ashland, OR, USA). 2.7. Adoptive transfer of BMDM/G-BMDM and estimation of the number of cells localized in the intestine Five  106 BMDM or G-BMDM were injected intraperitoneally into G-CSFR
/
recipient mice 1 day before and 2 days after DSS treatment. Blood samples were obtained immediately after sacrifice from each mouse in a 1.5 ml tube via cardiac puncture and allowed to clot at room temperature for 30 min. The blood samples were then centrifuged (2000 g, 4 °C, 10 min) and the serum was collected and stored at
80 °C until analyzed for serum cytokine levels. In addition, colon tissue was excised (feces was gently forced out), washed extensively with PBS, and measured for weight and length. The entire colon tissues were then opened longitudinally, cut into 0.5 cm pieces, and then 150 mg of the tissues were weighed and kept in TRIzol reagent (Life Technologies) for further quantitative real-time PCR (qRT-PCR) screening. The number of BMDM and G-BMDM were calculated by extrapolating a standard curve generated by series of dilutions of G-CSFR transcripts obtained from 1  107 BMDM or G-BMDM injected, using qRT-PCR. Estimation of % of recovered BMDM and G-BMDM in intestinal tissues were then calculated after calculating numbers of cells presented in random intestinal tissue samples (150 mg), using the same qRT-PCR procedures. 2.8. Quantitative real-time PCR Total cellular RNA was isolated from the colon tissues using TRIzol according to manufacturer’s instructions. One lg of total RNA was then reverse-transcribed with M-MuLV reverse transcriptase (New England BioLabs) according to the manufacturer’s instruction in the presence of oligo (dT) primer. cDNA was then used for qRTPCR analysis using the Brilliant SYBR Green PCR Master Mix (Applied Biosystems) on the Rotor-Gene RG3000 quantitative multiplex PCR instrument. Mouse oligonucleotide primers used in qRT-PCR analysis are: GAPDH: forward, 50 -GCATTGTG GAAGGGCTCATG-30 & reverse, 50 -TTGCTGTTGAAGTCGCAGGAG-30 ; TNF-a: forward, 50 -CTTGGAAATAGCTCCCAGAA-30 & reverse, 50 -CA

71

TTTGGGAACTTCTCATCC-30 ; IL-10: forward, 50 -TGCTATGCTGCCTG CTCTTA-30 & reverse, 50 -TCATTTCCGATAAGGCTTGG-30 ; arginase I: forward, 50 -CAGAAGAATGGAAGAGTCAG-30 & reverse, 50 -CAGA TATGCACGGAGTCACC-30 ; TGF-b: forward, 50 -GGAGAGCCCTGGA TACCAAC-30 & reverse, 50 -CAGGGTCCCAGACAGAAGTT-30 ; Ym1: forward, 50 -CATGATGCAGAATGCGCTGTGGAA-30 & reverse, 50 -AG GGCCCTTATTGAGAGGAGCTTT-30 ; FIZZ1: forward, 50 -ACTGCCTGT GCTTACTCGTTGACT-30 & reverse, 50 -AAAGCTGGGTTCTC CACCTCTTCA-30 ; IL-1b: forward, 50 - GCTTCAGGCAGGCAGTATCAC-30 & reverse, 50 -CGACAGCACGAGGCTTTTT-30 ; iNOS: forward, 50 -CAG CTGGGCTGTACAAACCTT-30 & reverse, 50 -CATTGGAAGT 0 0 GAAGCGTTTCG-3 ; G-CSFR: forward, 5 -CATCCAACCTGGGGACA GAC-30 & reverse: 50 -CTGGCAGGGGGATAGCCT-30 . 2.9. Statistical analysis Statistical analysis was carried out using Prism 5.0c for Mac OS X (GraphPad Software, La Jolla, CA, USA). Student’s t-tests, one-way ANOVA with Tukey’s multiple comparison post hoc test and two-way ANOVA with Bonferroni post-tests were performed as described in the figure legends. 3. Results 3.1. G-CSFR deficiency exacerbates DSS-induced colitis in mice To examine the role of endogenous G-CSF in colitis, we first examined susceptibility of G-CSFR
/
and G-CSFR+/
mice to DSSinduced acute colitis in comparison to G-CSFR+/+ mice. As shown in Fig. 1A, non-DSS-treated G-CSFR
/
mice (dotted line) showed no difference in body weight relative to G-CSFR+/
mice. Treatments of DSS through drinking water (ad libitum) for 5 days started causing severe body weight loss in Day 6 in all mice; however, GCSFR
/
mice showed more pronounced and rapid body weight loss than G-CSFR+/+ and +/
mice by Day 7, and some G-CSFR
/
mice lost more than 20% of their starting body weight by Day 8 (Fig. 1A, left panel). Consistent with these results, DSS-treated GCSFR
/
mice exhibited higher diarrhea, rectal bleeding and lethargy scores during the course of DSS treatment (Day 7), which then decreased to the levels of G-CSFR+/+ mice on Day 9 (Fig. 1A, middle panel). The reason for the decrease observed on Day 9 was due to early termination of these mice because of high scores on Day 8, rather than recovery from colitis. In agreement with these results, overall colitis scores including body weight, diarrhea, rectal bleeding and lethargy were significantly higher in G-CSFR
/
than G-CSFR+/+ or G-CSFR+/
mice toward to end of DSS-treatment (Fig. 1, right panel). Colons from mice euthanized on Day 9 were stained with H&E for histological evaluation. Non-DSS-treated G-CSFR+/
and
/
mice retained intact colonic structure (Fig. 1B, upper panels), whereas all DSS-treated mice had extensive epithelial cell destruction, crypt loss, and tissue infiltration (Fig. 1B, lower panels). However, colons from DSS-treated G-CSFR
/
mice showed more extensive tissue damage than those from DSS-treated G-CSFR+/+ mice. In agreement with these results, DSS treatment led to a significant increase in histopathological scores (0–4, 4 being the most tissue damage) of G-CSFR
/
colons compared to their G-CSFR+/+ counterparts (Fig. 1C). Indeed, all G-CSFR
/
colons examined scored 2–4, while G-CSFR+/+ tissue scores ranged from 0 to 4. Additionally, 3 out of the 5 G-CSFR
/
mice examined histologically also revealed the formation of a pseudomembranous structure (Fig. 1B, lower right panel), which appeared to be admixed with a large population of bacteria and pockets of bacterial clusters surrounded by infiltrating cells (arrow). These results suggest the presence of a secondary bacterial infection in the DSS-treated G-CSFR
/
mice. In

72

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

B

100 90 80 70

DSS

* *

* 0 1 2 3 4 5 6 7 8 9 Day G-CSFR+/-

4 3 *

2

*

1 0

0 1 2 3 4 5 6 7 8 9 Day

Overall colitis score

110

Diarrhea/rectal bleeding

Body Weight (% of starting body weight)

A 10 8

G-CSFR+/+ n=14 G-CSFR+/- n=11 G-CSFR-/- n=16 G-CSFR+/- No DSS n=4 G-CSFR-/- No DSS n=3

* *

6

*

4

2 0 0 1 2 3 4 5 6 7 8 9 Day

C

G-CSFR-/-

3 2 1 0 G-CSFR-/-

DSS-treated G-CSFR-/-

4

G-CSFR+/+

DSS-treated G-CSFR+/+

Histological Score

* 5

Fig. 1. G-CSFR deficiency exacerbates DSS-induced colitis in mice. (A) G-CSFR+/+, +/
or
/
mice were administered 1.5% dextran sulfate sodium (DSS) in the drinking water for 5 days, and fresh water thereafter. Body weight loss, diarrhea/rectal bleeding/lethargy scores and overall colitis scores (encompassing weight loss, stool consistency, rectal bleeding, and loss of activity) were plotted from Day 0 to 9. Figures show pooled data of more than three independent experiments (n P 10 per group, except Day 9 (n = 5 for G-CSFR+/
and
/
, n = 2 for G-CSFR+/+)). ⁄, p < 0.05 by two-way ANOVA with Bonferroni post-tests. Symbols represent mean ± SEM. (B) Colon sections stained with H&E. Non DSS-treated G-CSFR+/
and
/
colon tissue showed healthy, normal appearance. Nine days after DSS treatment, colon tissue of all mice groups showed intestinal tissue damage; however, G-CSFR+/+ tended to show a lesser degree of tissue damage than G-CSFR
/
. DSS-treated G-CSFR
/
mice showed the presence of a pseudomembranous structure with clusters of bacteria, digitally magnified inset of DSS-treated G-CSFR
/
; bacteria indicated by arrow. Original images were obtained at 400 magnification. (C) Histological damage scores (0–4, 4 being the most tissue damage) from 4 to 6 tissue sections of each group were plotted. All G-CSFR
/
tissues examined scored 2–4, whereas G-CSFR+/+ scores ranged from 0 to 4. ⁄, p < 0.05 by Student’s t-test.

contrast, no pseudomembranous structures or bacterial infections were detected in any of the colon segments from G-CSFR+/+ or +/
mice. 3.2. G-CSFR
/
mice harbor less basal levels of F4/80+ macrophage-like cells without defects in F4/80
/Gr-1+ neutrophil-like cells in basal and DSS-exposed intestines. We then examined whether F4/80+ macrophage-like and F4/80
/Gr-1+ neutrophil-like cells are similarly populated in the intestine between G-CSFR+/+ and G-CSFR
/
mice. As shown in Fig. 2A and B, G-CSFR
/
intestines were similarly populated by neutrophil-like cells (20% of all intestinal lamina propria cells), but significantly lower in macrophage-like cells (10%), when compared to those (20% in both cell types) of G-CSFR+/+ mice. To determine whether neutrophil deficiency participated in the exacerbation of inflammatory conditions in the DSS-treated G-CSFR
/
mice, we evaluated neutrophil infiltration into the intestine of DSS-treated mice using immunohistochemical staining for MPO. Non-DSS treated mice showed on average less than one positive cell per entire tissue section (data not shown). However, the average of MPO+ cells of colon segments from DSS-treated G-CSFR
/
mice (Fig. 2C; right panels shows a digitally magnified inset of left column, arrows indicate example MPO+ cells) showed a trend for slightly higher numbers of MPO+ cells relative to DSS-treated G-CSFR+/+ or +/
mice (Fig. 2D), although this did not reach statistical significance. Taken together, these data imply that macrophage, but

not neutrophil, infiltration may be responsible for the increased DSS-induced colitis susceptibility in G-CSFR
/
mice.

3.3. G-CSF induces enhanced expression of several M2-like cell surface markers in BMDM in response to IL-4/IL-13 To examine the effects G-CSF on macrophage activation, primary macrophages were generated by culturing bone marrow cells with M-CSF (BMDM) or G-CSF and M-CSF (G-BMDM) for 5 days and treated with IL-4 (20 ng/ml) and IL-13 (20 ng/ml) for an additional 2 days. On day 7, expression of M2 cell surface markers including PDL2, CD71 and CD206 (mannose receptor, C type 1) were examined on F4/80+ cells using flow cytometry. As shown in Fig. 3A, IL-4/IL-13 significantly induced expression of PDL2, CD206 and CD71 in both BMDM and G-BMDM. However, expression levels of these markers were significantly higher in GBMDM than BMDM. Similar experiments using G-CSFR
/
mice showed no such G-CSF effects on these markers in G-CSR
/
BMDM, suggesting that G-CSFR was the receptor responsible for exogenous G-CSF responses in BMDM. We further quantified mRNA expression levels of intracellular M2 markers in these cells, including Arginase I, TGF-b, Ym1 and FIZZ1. However, unlike surface markers, expression levels of these intracellular markers were not significantly different between G-BMDM and BMDM (Fig. 3B). These results suggest that G-CSF generated a distinct macrophage population with different capabilities in responding to M2-stimuli, particularly in expression surface M2 markers.

73

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

B

Intestinal neutrophils (%)

*

25 20 15 10 5

30 20 10

G-CSFR-/-

G-CSFR-/-

C

G-CSFR+/+

0

0 G-CSFR+/+

Intestinal macrophages (%)

A

D

60 40 20

G-CSFR-/-

G-CSFR+/-

0 G-CSFR+/+

DSS-treated G-CSFR-/-

# of MPO+ cells per field of view

DSS-treated

G-CSFR+/+

+ DSS Fig. 2. G-CSFR
/
mice harbor less basal levels of F4/80+ macrophage-like cells without defects in F4/80
/Gr-1+ neutrophil-like cells in basal and DSS-exposed intestines. (A and B) The intestinal lamina propria cells were isolated from G-CSFR+/+ and G-CSFR
/
mice and hematogenous cells were enriched on a Percoll step gradient. Percentage of F4/80+ macrophage-like (A) and F4/80
/Gr-1+ neutrophil-like cells (B) were quantified and plotted. Data shown as mean ± SEM (n P 3; ⁄, p < 0.05 by Student’s t-test). (C and D) G-CSFR+/+ and
/
mice colon tissue sections prepared 9 days after the initiation of 1.5% dextran sulfate sodium (DSS) treatment were stained for MPO. (C) While G-CSFR+/+ tissue sections showed a small number of cells with strong MPO staining (digitally magnified inset of DSS-treated G-CSFR+/+; MPO staining = brown color; arrow indicates example MPO+ cell), G-CSFR
/
showed higher numbers of MPO+ cells but with weak staining (digitally magnified inset of DSS-treated G-CSFR
/
; arrow indicates example MPO+ cell). Original images were obtained at 400 magnification. (D) Quantification of MPO+ cells per field of view from more than 4 tissue sections from each group was plotted. Line represents the mean. No significant changes in between groups were detected. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3.4. G-BMDM are less responsive to M1-stimuli in expressing iNOS, CD80 and CD86 than BMDM To examine the effects of G-CSF on the response of BMDM to M1-stimuli, BMDM were similarly cultured as above, except with IFN-c (20 ng/ml) and LPS (50 ng/ml) for an additional 2 days, and examined for the expression of M1 markers including IL-1b, iNOS, CD80 and CD86. As shown in Fig. 4, IFN-c and LPS increased the expression of all these markers in BMDM. However, in G-BMDM, only the expression of IL-1b, but not iNOS, CD80 and CD86, was similarly induced. These results suggest that G-BMDM are less responsive to IFN-c and LPS in expressing several key proinflammatory markers. 3.5. Adoptive transfer of G-BMDM exerts a protective effect on DSSinduced mouse colitis in G-CSFR
/
mice Next, we examined the role of G-BMDM in murine DSSinduced colitis in mice. To this end, 5  106 G-BMDM or BMDM

were adoptively transferred into G-CSFR
/
mice twice, 1 day before and 2 days after starting DSS treatment. Adoptive transfer of G-BMDM significantly prevented body weight loss, which was comparable to that of G-CSFR+/+ mice, whereas BMDM had no effect on body weight loss (Fig. 5A, left panel). Consistent with these results, adoptive transfer of G-BMDM, but not BMDM, into DSS-treated G-CSFR
/
mice also showed significantly lower diarrhea, rectal bleeding and overall colitis scores, which were similar to those of DSS-treated G-CSFR+/+ mice (Fig. 5A, middle and right panels). In addition, DSS caused more extensive colon shortening, which is an additional disease severity indicator in colitis, in G-CSFR
/
than G-CSFR+/+ mice. However, adoptive transfer of G-BMDM into G-CSFR
/
mice prevented the colon shortening to a similar extent of G-CSFR+/+ mice (Fig. 5B). We estimated that about 28% (2.8  106) of total adoptively transferred-G-BMDM 107 were localized in the intestine of these mice, based on qRT-PCR-based estimation as described in the Material and Methods, and provided cells were not dividing (Fig. 5C).

74

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

150 50 0

b

b c c

b

c c

+ IL-4 + IL-13

+ IL-4 + IL-13

+ IL-4 + IL-13

a

300 200

0

b b

b

100 c c

BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM

100

a 250 200 b b b 150 100 c c 50 c c 0

WT KO

c c

BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM

MFI of CD71

a

MFI of CD206

200

BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM BMDM G-BMDM

MFI of PDL2

A

+ IL-4 + IL-13

+ IL-4 + IL-13

+ IL-4 + IL-13

b

b

90 50 10 3 2 1 0

a a b

BMDM BMDM G-BMDM G-BMDM

b

19000 a 10000 a 1000 b 6 4 2 b 0

None IL-4/IL-13

BMDM BMDM G-BMDM G-BMDM

a

FIZZ1 mRNA fold change

a

Ym1 mRNA fold change

6 5 4 3 2 1 0

BMDM BMDM G-BMDM G-BMDM

TGF-β mRNA fold change

a a 500 400 300 5 b 2.5 b 0 BMDM BMDM G-BMDM G-BMDM

Arginase I mRNA fold change

B

a

600 400

b

200 bc 0

None IFN-γ + LPS

bc

BMDM BMDM G-BMDM G-BMDM

MFI of CD86

500 a 400 ab ab 300 200 b 100 0 BMDM BMDM G-BMDM G-BMDM

MFI of CD80

ab 10.0 7.5 bc 5.0 2.5 c c 0.0

BMDM BMDM G-BMDM G-BMDM

iNOS mRNA fold change

10 a 8 ab 6 bc 4 2 c 0

BMDM BMDM G-BMDM G-BMDM

IL-1β mRNA fold change

Fig. 3. G-CSF induces enhanced expression of several M2-like cell surface markers in BMDM in response to IL-4/IL-13. G-CSFR+/+ (WT) and G-CSFR
/
(KO) bone marrow cells were treated with M-CSF (20 ng/ml) (BMDM) or M-CSF + G-CSF (10 ng/ml) (G-BMDM) for 5 days. Then, cells were treated with IL-4 (20 ng/ml) + IL-13 (20 ng/ml) for an additional 2 days. (A) Mean fluorescence intensity (MFI) of PDL2, CD71 and CD206 cell surface markers of WT and KO M- and G-BMDM were examined. (B) mRNA expression of Arginase I, TGF-b, Ym1 and FIZZ1 was analyzed in IL-4/IL-13-treated WT M- or G-BMDM. Data shown as mean ± SEM [n P 3; ⁄, p < 0.05 by one-way ANOVA with Tukey’s multiple comparison post hoc test; columns accompanied by the same letter (a, b or c) are not significantly different from each other].

Fig. 4. G-BMDM are less responsive in expressing iNOS, CD80 and CD86 than BMDM to M1-stimuli. G-CSFR+/+ bone marrow cells were treated with M-CSF (20 ng/ml) (BMDM) or M-CSF + G-CSF (10 ng/ml) (G-BMDM) for 5 days. Then, cells were treated with IFN-c (20 ng/ml) + lipopolysaccharide (LPS; 50 ng/ml) for an additional 2 days. mRNA expression of IL-1b and iNOS and mean fluorescence intensity (MFI) of CD80 and CD86 cell surface markers were examined. Data shown as mean ± SEM [n P 3; ⁄, p < 0.05 by one-way ANOVA with Tukey’s multiple comparison post hoc test; columns accompanied by the same letter (a, b or c) are not significantly different from each other].

3.6. Adoptive transfer of G-BMDM prevents expression of proinflammatory cytokines and iNOS in DSS-treated G-CSFR
/
mice

is at least in macrophages.

Since adoptive transfer of G-BMDM protected DSS-induced colitis, we further examined the expression of cytokines and iNOS in serum and/or colon tissues from Day 10 samples of DSS and nontreated G-CSFR+/+ and G-CSFR
/
mice. As shown in Fig. 6A, DSS treatments significantly induced TNF-a, IL-1b and iNOS mRNAs in colon tissues, which was more pronounced in G-CSFR
/
than G-CSFR+/+ mice. G-CSFR
/
colon tissues adoptively transferred with G-BMDM showed lower expression of these inflammatory markers, which were comparative to those of G-CSFR+/+ colons. Similarly, production of the pro-inflammatory cytokines TNF-a and IL-1b were enhanced in G-CSFR
/
mice, when compared with those in G-CSFR+/+ mice, and adoptive transfer of G-BMDM to GCSFR
/
mice suppressed expression of these cytokines to the levels of G-CSFR+/+ mice (Fig. 6B). However, no significant changes were detected in the levels of IL-10 production regardless of DSS treatments (Fig. 6B; right panel). Overall, these results indicate that G-CSF plays a protective role in DSS-induced colitis in mice, which

4. Discussion

part

mediated

by

generating

gut-homing

Systemic administration of recombinant G-CSF induces immune suppressive and wound healing responses [45–47], which can have beneficial effects on animal and human colitis [16,48–51]. However, the in vivo role of G-CSF in immune homeostasis of the intestine is not well understood. Here, we show that G-CSFR
/
mice did not develop spontaneous colitis, but these mice developed pseudomembranous structures and more severe colitis than G-CSFR+/+ or G-CSFR+/
mice when 1.5% DSS was administered (Fig. 1), suggesting a protective role of G-CSF in acute colitis. In fact, it is often the case for mice genetically mutated in key innate immune factors that phenotypes only manifest after certain challenges such as DSS or microbial infections [52–54]. Different intestinal microflora and/ or the presence of unique pathogenic microbes could have contributed to the phenotypes of G-CSFR
/
mice [55,56]. However, G-CSFR
/
mice were housed intermixed with G-CSFR+/+ and +/


75

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

100 90

* *

80

DSS

70 0 1 2 3 4 5 6 7 8 9 10 Day

5 4 3 2

*

1

* * *

0

Overall colitis score

110

Diarrhea/rectal bleeding

12 10 8 6 4 2 0

G-CSFR +/+ n=10 *

*

0 1 2 3 4 5 6 7 8 9 10 Day

0 1 2 3 4 5 6 7 8 9 10 Day

G-CSFR -/- n=15 G-CSFR -/- + BMDM n=15 G-CSFR -/- + G-BMDM n=10

C

B a b

4

4

+ DSS

+ DSS

G-CSFR -/+ G-BMDM

G-CSFR -/-

G-CSFR +/+

G-CSFR -/-

G-CSFR +/+

G-CSFR -/+ G-BMDM

G-CSFR -/-

G-CSFR +/+

G-CSFR -/-

G-CSFR +/+

00

40 30 20 10 0 G-CSFR -/+ BMDM G-CSFR -/+ G-BMDM

c

G-CSFR -/+ G-BMDM

b 88

G-CSFR -/+ BMDM

a

G-CSFR -/-

Colon length (cm)

2 12

Recovered adaptively transferred BMDM (%)

Body Weight (% of starting body weight)

A

+ DSS

Fig. 5. Adoptively transferred G-BMDM exerts a protective effect on DSS-induced mouse colitis in G-CSFR
/
mice. G-CSFR+/+, and
/
mice were given 1.5% DSS in the drinking water for 5 days, and fresh water thereafter. G-CSFR
/
mice were injected with 5  106 M-CSF (20 ng/ml)-treated (BMDM) or M-CSF + G-CSF (10 ng/ml)-treated (GBMDM) cells 1 day before and 2 days after DSS treatment. (A) Body weight loss, diarrhea/rectal bleeding and overall colitis scores (encompassing weight loss, stool consistency, rectal bleeding, and loss of activity) were plotted from Day 0 to 10. ⁄, p < 0.05 by one-way ANOVA with Tukey’s multiple comparison post hoc test. Symbols represent mean ± SEM. (B) Colon length of mice after 10 days of DSS treatment was measured and plotted. (C) Gut-homing phenotype of the adoptive transferred-M- and GBMDM in DSS-treated G-CSFR
/
mice was evaluated by investigating the presence of mRNA expression of G-CSF receptor wild-type (G-CSFR exon 4 and 5) in colon tissues. Data shown as mean ± SEM [n P 3; ⁄, p < 0.05 by one-way ANOVA with Tukey’s multiple comparison post hoc test; columns accompanied by the same letter (a, b or c) are not significantly different from each other].

mice during the course of the experiment; therefore, G-CSFR deficiency, rather than differences in microbial exposure, was likely responsible for the development of pseudomembrane structures and severe colitis. G-CSFR
/
mice did not show apparent defects in the numbers of resident (Fig. 2A) and recruited neutrophils (MPO+ cells; Fig. 2B– D). These results are consistent with previous studies shown that, although circulating neutrophil counts are decreased up to 85%, emergency generation of neutrophils and their recruitments to local infection/inflammatory sites during infection or inflammation are not compromised in G-CSF
/
or G-CSFR
/
mice [57– 59]. However, since G-CSF also enhances the bactericidal function of neutrophils, development of severe colitis and the formation of pseudomembranous structures in G-CSFR
/
mice could be due to functional defects of intestinal neutrophils. Indeed, low intensities of MPO staining in G-CSFR
/
mice may indicate a functional defect in G-CSFR
/
neutrophils, as MPO is important for the bactericidal activity of neutrophils [60] and their ability to produce neutrophil extracellular traps [61]. Further studies are required to explore whether and how these defects influence the outcome of severe colitis in G-CSFR
/
mice. Unlike neutrophils, we found lower numbers of intestinal macrophages in G-CSFR
/
mice than wildtype mice (Fig. 2A). This observation is consistent with the facts that G-CSF
/
mice harbor 50% lower circulating monocytes than wild-type mice [43], resident intestinal macrophages are mainly originated from circulating monocytes [19,27], and G-CSF

promotes generation of gut-homing Gr-1high/F4/80+ monocytes [44]. Therefore, we further examined the characteristics of macrophages cultured in the presence of G-CSF and their role in colitis. Of more than 20 soluble hematopoietic factors, M-CSF and granulocyte/macrophage CSF (GM-CSF) play crucial roles in the generation and characteristic differentiation of myeloid cells in the bone marrow [62,63]. In general, M-CSF is involved in the generation and maintenance of ‘‘M2-like” macrophages, which potently produce anti-inflammatory/regulatory cytokines upon activation; whereas, GM-CSF generates ‘‘M1-like” DCs or macrophages skewed toward pro-inflammatory responses [63,64]. Previously, we showed that G-CSF enhances the generation of M2-like macrophages from M-CSF-derived, but not GM-CSF-derived BMDM, which are mainly localized in the colon after the adoptive transfer [44]. In line with these observations, macrophages cultured together with G-CSF expressed higher levels of M2 markers (PDL2, CD71 and CD206; Fig. 3), but lower levels of M1 markers (iNOS and CD80/86; Fig. 4), in response to IL-4/-13 and IFN-c/ LPS, respectively. However, G-CSF had no effects on the capability in expressing several other M1 and M2 markers, such as arginase I, Ym1, FIZZ1, TGF-b and IL-1b, upon activation (Figs. 3B and 4). These results suggest that the effect of G-CSF on macrophages was distinct from a generalized dichotic view of macrophage activation/differentiation. At this moment, it is not clear whether G-CSF directly induced expression or trafficking of these surface markers, and/or contributed to the generation of specific subsets

76

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

A 6

b

4

b

+ DSS

+ DSS

G-CSFR-/G-CSFR-/+G-BMDM

G-CSFR+/+

b G-CSFR-/-

2 b 0

G-CSFR-/G-CSFR-/+ G-BMDM

a

8

G-CSFR+/+

c

iNOS mRNA fold change

b

b

G-CSFR+/+

G-CSFR-/G-CSFR-/+ G-BMDM

G-CSFR+/+

0

a

G-CSFR-/-

c

980 500 20 2 c 1 G-CSFR+/+

b

G-CSFR-/-

G-CSFR+/+

TNF-α mRNA fold change

0

b

IL-1β mRNA fold change

a

120 70 20 2 c 1

+ DSS

0

200 100

+ DSS

G-CSFR-/G-CSFR-/+ G-BMDM

G-CSFR+/+

G-CSFR+/+ G-CSFR-/-

G-CSFR-/G-CSFR-/+ G-BMDM

G-CSFR+/+

G-CSFR-/-

G-CSFR+/+

0

+ DSS

G-CSFR-/+ G-BMDM

200

300

1000 800 600 400 200 0

G-CSFR-/-

400

*

G-CSFR+/+

600

*

400

G-CSFR-/-

*

G-CSFR+/+

*

IL-10 (pg/ml)

800

IL-1β (pg/ml)

TNF-α (pg/ml)

B

+ DSS

Fig. 6. Adoptive transfer of G-BMDM prevents expression of pro-inflammatory cytokines and iNOS in DSS-treated G-CSFR
/
mice. G-CSFR+/+, and
/
mice were given 1.5% DSS in the drinking water for 5 days, and fresh water thereafter. G-CSFR
/
mice were injected with 5  106 M-CSF (20 ng/ml) + G-CSF (10 ng/ml)-treated (G-BMDM) cells 1 day before and 2 days after DSS treatment. (A) mRNA expression of TNF-a, IL-1b and iNOS was measured using qRT-PCR from colon tissues of mice with or without adoptive transfer of G-BMDM. Data shown as mean ± SEM [n P 3; by one-way ANOVA with Tukey’s multiple comparison post hoc test; columns accompanied by the same letter (a, b, or c) are not significantly different from each other]. (B) Cytokine levels of TNF-a, IL-1b and IL-10 were measured from serum of mice using ELISA. ⁄, p < 0.05 by Student’s t-test. Each point represents an individual mouse, line shows the mean. Pooled data of two independent experiments.

of macrophages. Since, G-CSF, when treated on fully differentiated macrophages, had no effects on the expression of these markers (data not shown), we suspect that G-CSF preferentially promoted generation of macrophages with unique characteristics. It should also be noted that G-CSFR activates STAT3 and STAT 5 signaling cascades which are common to those by IL-10 and IFN-c receptors, respectively [8,65–67]. Therefore, G-CSF pretreatment could provide an early IL-10 or IFN-c-like signal, resulting in a distinct M0 intermediate state. Further experiments using cell proliferation inhibitors or mutant G-CSFR specifically defective in the cell proliferation or differentiation signaling cascade [68,69] will likely provide information on these questions. Regardless of the mechanism of G-CSF in inducing M2-like responses, we examined the role of macrophages in colitis and found that G-BMDM, but not BMDM, protected G-CSFR
/
mice from DSS-induced acute colitis (Fig. 5). DSS treatments induced higher levels of TNF-a, IL-1b and iNOS in the colon, and TNF-a and IL-1b in the serum in G-CSFR
/
mice than wild-type mice, which was prevented by adoptively transferring G-BMDM (Fig. 6). Since these pro-inflammatory cytokines are main mediators of colitis in DSS-induced colitis in animals and inflammatory bowel disease patients [70–73], we speculate that G-CSF at least in part positively contribute to immune homeostasis of the intestine through enhancing intestinal localization of immune regulatory macrophages and suppressing production of pro-inflammatory cytokines. However, it is still unknown how G-BMDM rendered the colitis ameliorating effects. We speculate

that G-BMDM adoptively transferred likely conferred beneficial effects during DSS-induced colitis through enhancing immune barrier function (eg. phagocytosis and killing microbes) and/or wound healing processes without inducing overt inflammation, all of which are functions of intestinal resident macrophages [74–77]. Consistent with the beneficial effects of G-CSF, we previously showed that G-BMDM comprise mainly with Gr-1+/CD11b+ cells [44], which are known as myeloid-derived suppressor cells [78,79] and adoptive transfer of these cells protects allergeninduced airway inflammation [80] and colitis [81]. In summary, this study demonstrates that G-CSFR
/
mice were more susceptible to DSS-induced acute colitis than wild-type mice and adoptive transfer of G-BMDM protected G-CSFR
/
mice from the colitis. These results suggest that G-CSF plays an important role in maintaining intestinal immune homeostasis and preventing colitis, likely through populating immune regulatory macrophages in the intestine. Conflict of interest None. Acknowledgement We thank Ms. Chantelle Reid for editing the manuscript. The study was supported by the Natural Sciences & Engineering Research of Canada (No. RGPIN 312482-2013) to SOK.

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

References [1] D. Metcalf, N.A. Nicola, Proliferative effects of purified granulocyte colonystimulating factor (G-CSF) on normal mouse hemopoietic cells, J. Cell. Physiol. 116 (1983) 198–206. [2] N.A. Nicola, D. Metcalf, M. Matsumoto, G.R. Johnson, Purification of a factor inducing differentiation in murine myelomonocytic leukemia cells. Identification as granulocyte colony-stimulating factor, J. Biol. Chem. 258 (1983) 9017–9023. [3] G. Carulli, Effects of recombinant human granulocyte colony-stimulating factor administration on neutrophil phenotype and functions, Haematologica 82 (1997) 606–616. [4] E. Cario, H. Goebell, A.U. Dignass, Factor XIII modulates intestinal epithelial wound healing in vitro, Scand. J. Gastroenterol. 34 (1999) 485–490. [5] J.C. Hebert, M. O’Reilly, R.L. Gamelli, Protective effect of recombinant human granulocyte colony-stimulating factor against pneumococcal infections in splenectomized mice, Arch. Surg. 125 (1990) 1075–1078. [6] H. Wakiyama, S. Tsuru, N. Hata, M. Shinomiya, N. Shinomiya, M. Noritake, et al., Therapeutic effect of granulocyte colony-stimulating factor and cephem antibiotics against experimental infections in neutropenic mice induced by cyclophosphamide, Clin. Exp. Immunol. 92 (1993) 218–224. [7] H. Yasuda, Y. Ajiki, T. Shimozato, M. Kasahara, H. Kawada, M. Iwata, et al., Therapeutic efficacy of granulocyte colony-stimulating factor alone and in combination with antibiotics against Pseudomonas aeruginosa infections in mice, Infect. Immun. 58 (1990) 2502–2509. [8] S.O. Kim, H.I. Sheikh, S.D. Ha, A. Martins, G. Reid, G-CSF-mediated inhibition of JNK is a key mechanism for Lactobacillus rhamnosus-induced suppression of TNF production in macrophages, Cell. Microbiol. 8 (2006) 1958–1971. [9] A.J. Martins, P. Colquhoun, G. Reid, S.O. Kim, Reduced expression of basal and probiotic-inducible G-CSF in intestinal mononuclear cells is associated with inflammatory bowel disease, Inflamm. Bowel Dis. 15 (2009) 515–525. [10] A.J. Martins, S. Spanton, H.I. Sheikh, S.O. Kim, The anti-inflammatory role of granulocyte colony-stimulating factor in macrophage-dendritic cell crosstalk after Lactobacillus rhamnosus GR-1 exposure, J. Leukoc. Biol. (2011). [11] S. Rutella, G. Bonanno, L. Pierelli, A. Mariotti, E. Capoluongo, A.M. Contemi, et al., Granulocyte colony-stimulating factor promotes the generation of regulatory DC through induction of IL-10 and IFN-alpha, Eur. J. Immunol. 34 (2004) 1291–1302. [12] S. Rutella, L. Pierelli, G. Bonanno, S. Sica, F. Ameglio, E. Capoluongo, et al., Role for granulocyte colony-stimulating factor in the generation of human T regulatory type 1 cells, Blood 100 (2002) 2562–2571. [13] H. Egi, K. Hayamizu, M. Yoshimitsu, F. Shimamoto, K. Oishi, I. Ohmori, et al., Regulation of T helper type-1 immunity in hapten-induced colitis by host pretreatment with granulocyte colony-stimulating factor, Cytokine 23 (2003) 23–30. [14] T. Kudo, T. Matsumoto, I. Nakamichi, S. Yada, M. Esaki, Y. Jo, et al., Recombinant human granulocyte colony-stimulating factor reduces colonic epithelial cell apoptosis and ameliorates murine dextran sulfate sodium-induced colitis, Scand. J. Gastroenterol. 43 (2008) 689–697. [15] J.R. Korzenik, B.K. Dieckgraefe, J.F. Valentine, D.F. Hausman, M.J. Gilbert, Sargramostim for active Crohn’s disease, N. Engl. J. Med. 352 (2005) 2193–2201. [16] C. Dejaco, C. Lichtenberger, W. Miehsler, G. Oberhuber, F. Herbst, H. Vogelsang, et al., An open-label pilot study of granulocyte colony-stimulating factor for the treatment of severe endoscopic postoperative recurrence in Crohn’s disease, Digestion 68 (2003) 63–70. [17] P.J. Mannon, F. Leon, I.J. Fuss, B.A. Walter, M. Begnami, M. Quezado, et al., Successful granulocyte-colony stimulating factor treatment of Crohn’s disease is associated with the appearance of circulating interleukin-10-producing T cells and increased lamina propria plasmacytoid dendritic cells, Clin. Exp. Immunol. 155 (2009) 447–456. [18] P.D. Smith, C. Ochsenbauer-Jambor, L.E. Smythies, Intestinal macrophages: unique effector cells of the innate immune system, Immunol. Rev. 206 (2005) 149–159. [19] C.C. Bain, A.M. Mowat, The monocyte-macrophage axis in the intestine, Cell. Immunol. (2014). [20] A.M. Mowat, C.C. Bain, Mucosal macrophages in intestinal homeostasis and inflammation, J. Innate Immun. 3 (2011) 550–564. [21] J.E. Qualls, A.M. Kaplan, N. van Rooijen, D.A. Cohen, Suppression of experimental colitis by intestinal mononuclear phagocytes, J. Leukoc. Biol. 80 (2006) 802–815. [22] A. Rizzo, I. Monteleone, D. Fina, C. Stolfi, R. Caruso, M.C. Fantini, et al., Inhibition of colitis by IL-25 associates with induction of alternatively activated macrophages, Inflamm. Bowel Dis. 18 (2012) 449–459. [23] C.C. Bain, A.M. Mowat, Macrophages in intestinal homeostasis and inflammation, Immunol. Rev. 260 (2014) 102–117. [24] Y.R. Mahida, The key role of macrophages in the immunopathogenesis of inflammatory bowel disease, Inflamm. Bowel Dis. 6 (2000) 21–33. [25] S. Epelman, K.J. Lavine, G.J. Randolph, Origin and functions of tissue macrophages, Immunity 41 (2014) 21–35. [26] D. Hashimoto, A. Chow, C. Noizat, P. Teo, M.B. Beasley, M. Leboeuf, et al., Tissueresident macrophages self-maintain locally throughout adult life with minimal contribution from circulating monocytes, Immunity 38 (2014) 792–804. [27] S. Yona, K.W. Kim, Y. Wolf, A. Mildner, D. Varol, M. Breker, et al., Fate mapping reveals origins and dynamics of monocytes and tissue macrophages under homeostasis, Immunity 38 (2013) 79–91.

77

[28] S. Giorgio, Macrophages: plastic solutions to environmental heterogeneity, Inflamm. Res. 62 (2013) 835–843. [29] D.M. Mosser, J.P. Edwards, Exploring the full spectrum of macrophage activation, Nat. Rev. Immunol. 8 (2008) 958–969. [30] M.L. Novak, T.J. Koh, Macrophage phenotypes during tissue repair, J. Leukoc. Biol. 93 (2013) 875–881. [31] F.O. Martinez, L. Helming, S. Gordon, Alternative activation of macrophages: an immunologic functional perspective, Annu. Rev. Immunol. 27 (2009) 451–483. [32] C.D. Mills, K. Kincaid, J.M. Alt, M.J. Heilman, A.M. Hill, M-1/M-2 macrophages and the Th1/Th2 paradigm, J. Immunol. 164 (2000) 6166–6173. [33] S.B. Weisser, L.K. Kozicky, H.K. Brugger, E.N. Ngoh, B. Cheung, R. Jen, et al., Arginase activity in alternatively activated macrophages protects PI3Kp110delta deficient mice from dextran sodium sulfate induced intestinal inflammation, Eur. J. Immunol. 44 (2014) 3353–3367. [34] G. Raes, P. De Baetselier, W. Noel, A. Beschin, F. Brombacher, Gh.G. Hassanzadeh, Differential expression of FIZZ1 and Ym1 in alternatively versus classically activated macrophages, J. Leukoc. Biol. 71 (2002) 597–602. [35] D. Ruffell, F. Mourkioti, A. Gambardella, P. Kirstetter, R.G. Lopez, N. Rosenthal, et al., A CREB-C/EBPbeta cascade induces M2 macrophage-specific gene expression and promotes muscle injury repair, Proc. Natl. Acad. Sci. USA 106 (2009) 17475–17480. [36] G. Corna, L. Campana, E. Pignatti, A. Castiglioni, E. Tagliafico, L. Bosurgi, et al., Polarization dictates iron handling by inflammatory and alternatively activated macrophages, Haematologica 95 (2010) 1814–1822. [37] C. Gaetano, L. Massimo, M. Alberto, Control of iron homeostasis as a key component of macrophage polarization, Haematologica 95 (2010) 1801–1803. [38] F. Rey-Giraud, M. Hafner, C.H. Ries, In vitro generation of monocyte-derived macrophages under serum-free conditions improves their tumor promoting functions, PLoS ONE 7 (2012) e42656. [39] S. Huber, R. Hoffmann, F. Muskens, D. Voehringer, Alternatively activated macrophages inhibit T-cell proliferation by Stat6-dependent expression of PDL2, Blood 116 (2010) 3311–3320. [40] P. Loke, J.P. Allison, PD-L1 and PD-L2 are differentially regulated by Th1 and Th2 cells, Proc. Natl. Acad. Sci. USA 100 (2003) 5336–5341. [41] R. Parsa, P. Andresen, A. Gillett, S. Mia, X.M. Zhang, S. Mayans, et al., Adoptive transfer of immunomodulatory M2 macrophages prevents type 1 diabetes in NOD mice, Diabetes 61 (2012) 2881–2892. [42] F. Liu, H.Y. Wu, R. Wesselschmidt, T. Kornaga, D.C. Link, Impaired production and increased apoptosis of neutrophils in granulocyte colony-stimulating factor receptor-deficient mice, Immunity 5 (1996) 491–501. [43] G.J. Lieschke, D. Grail, G. Hodgson, D. Metcalf, E. Stanley, C. Cheers, et al., Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization, Blood 84 (1994) 1737–1746. [44] S. Meshkibaf, M. William Gower, G.A. Dekaban, S. Ouk Kim, G-CSF preferentially supports the generation of gut-homing Gr-1high macrophages in M-CSF-treated bone marrow cells, J. Leukoc. Biol. 96 (2014) 549–561. [45] C.L. Semerad, F. Liu, A.D. Gregory, K. Stumpf, D.C. Link, G-CSF is an essential regulator of neutrophil trafficking from the bone marrow to the blood, Immunity 17 (2002) 413–423. [46] M. Matsumoto, S. Matsubara, T. Matsuno, M. Tamura, K. Hattori, H. Nomura, et al., Protective effect of human granulocyte colony-stimulating factor on microbial infection in neutropenic mice, Infect. Immun. 55 (1987) 2715–2720. [47] N. Nishio, Y. Okawa, H. Sakurai, K. Isobe, Neutrophil depletion delays wound repair in aged mice, Age (Dordr) 30 (2008) 11–19. [48] J.R. Korzenik, B.K. Dieckgraefe, An open-labelled study of granulocyte colonystimulating factor in the treatment of active Crohn’s disease, Aliment. Pharmacol. Ther. 21 (2005) 391–400. [49] D.W. Hommes, J. Meenan, S. Dijkhuizen, F.J. Ten Kate, G.N. Tytgat, S.J. Van Deventer, Efficacy of recombinant granulocyte colony-stimulating factor (rhGCSF) in experimental colitis, Clin. Exp. Immunol. 106 (1996) 529–533. [50] M. Yoshimitsu, K. Hayamizu, H. Egi, J. Okiyama, M. Okajima, T. Itamoto, et al., The neutrophil/Th1 lymphocyte balance and the therapeutic effect of granulocyte colony-stimulating factor in TNBS-induced colitis of rat strains, J. Interferon Cytokine Res. 26 (2006) 291–300. [51] S.K. Sainathan, E.M. Hanna, Q. Gong, K.S. Bishnupuri, Q. Luo, M. Colonna, et al., Granulocyte macrophage colony-stimulating factor ameliorates DSS-induced experimental colitis, Inflamm. Bowel Dis. 14 (2008) 88–99. [52] A. Araki, T. Kanai, T. Ishikura, S. Makita, K. Uraushihara, R. Iiyama, et al., MyD88-deficient mice develop severe intestinal inflammation in dextran sodium sulfate colitis, J. Gastroenterol. 40 (2005) 16–23. [53] T. Shea-Donohue, K. Thomas, M.J. Cody, Z. Aiping, L.J. Detolla, K.M. Kopydlowski, et al., Mice deficient in the CXCR2 ligand, CXCL1 (KC/GROalpha), exhibit increased susceptibility to dextran sodium sulfate (DSS)induced colitis, Innate Immun. 14 (2008) 117–124. [54] T.O. Khor, M.T. Huang, K.H. Kwon, J.Y. Chan, B.S. Reddy, A.N. Kong, Nrf2deficient mice have an increased susceptibility to dextran sulfate sodiuminduced colitis, Cancer Res. 66 (2006) 11580–11584. [55] M.M. Heimesaat, A. Fischer, B. Siegmund, A. Kupz, J. Niebergall, D. Fuchs, et al., Shift towards pro-inflammatory intestinal bacteria aggravates acute murine colitis via Toll-like receptors 2 and 4, PLoS ONE 2 (2007) e662. [56] M. Boirivant, A. Amendola, A. Butera, Intestinal microflora and immunoregulation, Mucosal Immunol. 1 (Suppl 1) (2008) S47–S49. [57] S. Basu, G. Hodgson, H.H. Zhang, M. Katz, C. Quilici, A.R. Dunn, ‘‘Emergency” granulopoiesis in G-CSF-deficient mice in response to Candida albicans infection, Blood 95 (2000) 3725–3733.

78

S. Meshkibaf et al. / Cytokine 78 (2016) 69–78

[58] Y. Zhan, G.J. Lieschke, D. Grail, A.R. Dunn, C. Cheers, Essential roles for granulocyte-macrophage colony-stimulating factor (GM-CSF) and G-CSF in the sustained hematopoietic response of Listeria monocytogenes-infected mice, Blood 91 (1998) 863–869. [59] M.L. Hibbs, C. Quilici, N. Kountouri, J.F. Seymour, J.E. Armes, A.W. Burgess, et al., Mice lacking three myeloid colony-stimulating factors (G-CSF, GM-CSF, and M-CSF) still produce macrophages and granulocytes and mount an inflammatory response in a sterile model of peritonitis, J. Immunol. 178 (2007) 6435–6443. [60] S. von Vietinghoff, K. Ley, Homeostatic regulation of blood neutrophil counts, J. Immunol. 181 (2008) 5183–5188. [61] K.D. Metzler, T.A. Fuchs, W.M. Nauseef, D. Reumaux, J. Roesler, I. Schulze, et al., Myeloperoxidase is required for neutrophil extracellular trap formation: implications for innate immunity, Blood 117 (2010) 953–959. [62] V. Chitu, E.R. Stanley, Colony-stimulating factor-1 in immunity and inflammation, Curr. Opin. Immunol. 18 (2006) 39–48. [63] A.J. Fleetwood, T. Lawrence, J.A. Hamilton, A.D. Cook, Granulocyte-macrophage colony-stimulating factor (CSF) and macrophage CSF-dependent macrophage phenotypes display differences in cytokine profiles and transcription factor activities: implications for CSF blockade in inflammation, J. Immunol. 178 (2007) 5245–5252. [64] J.A. Hamilton, A. Achuthan, Colony stimulating factors and myeloid cell biology in health and disease, Trends Immunol. 34 (2013) 81–89. [65] C. Berlato, M.A. Cassatella, I. Kinjyo, L. Gatto, A. Yoshimura, F. Bazzoni, Involvement of suppressor of cytokine signaling-3 as a mediator of the inhibitory effects of IL-10 on lipopolysaccharide-induced macrophage activation, J. Immunol. 168 (2002) 6404–6411. [66] A. Chakraborty, K.F. Dyer, D.J. Tweardy, Delineation and mapping of Stat5 isoforms activated by granulocyte colony-stimulating factor in myeloid cells, Blood Cells Mol. Dis. 26 (2000) 320–330. [67] A. Meinke, F. Barahmand-Pour, S. Wohrl, D. Stoiber, T. Decker, Activation of different Stat5 isoforms contributes to cell-type-restricted signaling in response to interferons, Mol. Cell. Biol. 16 (1996) 6937–6944. [68] Q.S. Zhu, L.J. Robinson, V. Roginskaya, S.J. Corey, G-CSF-induced tyrosine phosphorylation of Gab2 is Lyn kinase dependent and associated with enhanced Akt and differentiative, not proliferative, responses, Blood 103 (2004) 3305–3312. [69] M. Sampson, Q.S. Zhu, S.J. Corey, Src kinases in G-CSF receptor signaling, Front. Biosci. 12 (2007) 1463–1474. [70] H.C. Reinecker, M. Steffen, T. Witthoeft, I. Pflueger, S. Schreiber, R.P. MacDermott, et al., Enhanced secretion of tumour necrosis factor-alpha, IL-6,

[71]

[72]

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80]

[81]

and IL-1 beta by isolated lamina propria mononuclear cells from patients with ulcerative colitis and Crohn’s disease, Clin. Exp. Immunol. 94 (1993) 174–181. K.J. Myers, S. Murthy, A. Flanigan, D.R. Witchell, M. Butler, S. Murray, et al., Antisense oligonucleotide blockade of tumor necrosis factor-alpha in two murine models of colitis, J. Pharmacol. Exp. Ther. 304 (2003) 411–424. M.F. Neurath, I. Fuss, M. Pasparakis, L. Alexopoulou, S. Haralambous, K.H. Meyer zum Buschenfelde, W. Strober, G. Kollias, Predominant pathogenic role of tumor necrosis factor in experimental colitis in mice, Eur. J. Immunol. 27 (1997) 1743–1750. Y. Yan, V. Kolachala, G. Dalmasso, H. Nguyen, H. Laroui, S.V. Sitaraman, et al., Temporal and spatial analysis of clinical and molecular parameters in dextran sodium sulfate induced colitis, PLoS ONE 4 (2009) e6073. L.E. Smythies, M. Sellers, R.H. Clements, M. Mosteller-Barnum, G. Meng, W.H. Benjamin, et al., Human intestinal macrophages display profound inflammatory anergy despite avid phagocytic and bacteriocidal activity, J. Clin. Invest. 115 (2005) 66–75. C.C. Bain, C.L. Scott, H. Uronen-Hansson, S. Gudjonsson, O. Jansson, O. Grip, et al., Resident and pro-inflammatory macrophages in the colon represent alternative context-dependent fates of the same Ly6Chi monocyte precursors, Mucosal Immunol. 6 (2013) 498–510. U. Hadis, B. Wahl, O. Schulz, M. Hardtke-Wolenski, A. Schippers, N. Wagner, et al., Intestinal tolerance requires gut homing and expansion of FoxP3+ regulatory T cells in the lamina propria, Immunity 34 (2011) 237–246. Z. Enderlin Vaz da Silva, H.A. Lehr, D. Velin, In vitro and in vivo repair activities of undifferentiated and classically and alternatively activated macrophages, Pathobiology 81 (2014) 86–93. V. Greifenberg, E. Ribechini, S. Rossner, M.B. Lutz, Myeloid-derived suppressor cell activation by combined LPS and IFN-gamma treatment impairs DC development, Eur. J. Immunol. 39 (2009) 2865–2876. E. Ribechini, V. Greifenberg, S. Sandwick, M.B. Lutz, Subsets, expansion and activation of myeloid-derived suppressor cells, Med. Microbiol. Immunol. 199 (2010) 273–281. M. Arora, S.L. Poe, A. Ray, P. Ray, LPS-induced CD11b+Gr1(int)F4/80+ regulatory myeloid cells suppress allergen-induced airway inflammation, Int. Immunopharmacol. 11 (2011) 827–832. L.A. Haile, R. von Wasielewski, J. Gamrekelashvili, C. Kruger, O. Bachmann, A.M. Westendorf, J. Buer, R. Liblau, M.P. Manns, F. Korangy, T.F. Greten, Myeloidderived suppressor cells in inflammatory bowel disease: a new immunoregulatory pathway, Gastroenterology 135 (2008) 871–881. 881 e1–5.