TREM-1 regulates macrophage polarization in ureteral obstruction

basic research

http://www.kidney-international.org & 2014 International Society of Nephrology

TREM-1 regulates macrophage polarization in ureteral obstruction Tzu-Han Lo1,12, Kai-Yu Tseng2,12, Wen-Shan Tsao2, Chih-Ya Yang2,3, Shie-Liang Hsieh2,3,4,5,6,7, Allen Wen-Hsiang Chiu8, Toshiyuki Takai9, Tak W. Mak10, Der-Cherng Tarng1,4,5,6,11 and Nien-Jung Chen2,5 1

Institute of Physiology, School of Medicine, National Yang-Ming University, Taipei, Taiwan (ROC); 2Institute of Microbiology and Immunology, School of Life Sciences, National Yang-Ming University, Taipei, Taiwan (ROC); 3Genomic Research Center, Academia Sinica, Taipei, Taiwan (ROC); 4Institute of Clinical Medicine, School of Medicine, National Yang-Ming University, Taipei, Taiwan (ROC); 5 Inflammation and Immunity Research Center, National Yang-Ming University, Taipei, Taiwan (ROC); 6Immunology Center, Taipei Veterans General Hospital, Taipei, Taiwan (ROC); 7Institute for Cancer Biology and Drug Discovery, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan; 8Department of Urology, School of Medicine, National Yang-Ming University, Taipei, Taiwan (ROC); 9Department of Experimental Immunology, Institute of Development, Aging and Cancer, Tohoku University, Sendai, Japan; 10The Campbell Family Institute for Breast Cancer Research, Ontario Cancer Institute, University Health Network and Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada and 11Division of Nephrology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan (ROC)

Chronic kidney disease (CKD) is an emerging worldwide public health problem. Inflammatory cell infiltration and activation during the early stages in injured kidneys is a common pathologic feature of CKD. Here, we determined whether an important inflammatory regulator, triggering receptor expressed on myeloid cells (TREM)-1, is upregulated in renal tissues collected from mouse ureteral obstruction–induced nephritis. TREM-1 is crucial for modulating macrophage polarization, and has a pivotal role in mediating tubular injury and interstitial collagen deposition in obstructive nephritis. Lysates from nephritic kidneys triggered a TREM-1-dependent M1 polarization ex vivo, consistent with the observation that granulocytemacrophage colony-stimulating factor (GM-CSF)-derived M1 macrophages express higher levels of TREM-1 in comparison with M-CSF-derived cells. Moreover, agonistic TREM-1 crosslink significantly strengthens the inductions of iNOS and GMCSF in M1 cells. These observations are validated by a strong clinical correlation between infiltrating TREM-1-expressing/ iNOS-positive macrophages and renal injury in human obstructive nephropathy. Thus, TREM-1 may be a potential diagnostic and therapeutic target in human kidney disease. Correspondence: Tak W. Mak, The Campbell Family Institute for Breast Cancer Research, Ontario Cancer Institute, University Health Network and Department of Medical Biophysics, University of Toronto, University Avenue, Room 7-706, Toronto, Ontario, Canada M5G 2C1. E-mail: [email protected] or Der-Cherng Tarng, Department and Institute of Physiology, School of Medicine, National Yang-Ming University, No.155, Sec.2, Linong Street, Taipei 112, Taiwan (ROC). E-mail: [email protected] or Nien-Jung Chen, Institute of Microbiology and Immunology, School of Life Sciences, National Yang-Ming University, No.155, Sec.2, Linong Street, Taipei 112, Taiwan (ROC). E-mail: [email protected] 12

These authors contributed equally to this work.

Received 22 April 2013; revised 21 April 2014; accepted 24 April 2014; published online 11 June 2014 1174

Kidney International (2014) 86, 1174–1186; doi:10.1038/ki.2014.205; published online 11 June 2014 KEYWORDS: chronic kidney disease; fibrosis; immunology and pathology; macrophages; obstructive nephropathy

Chronic kidney disease (CKD) is an emerging health problem that poses a growing socioeconomic burden for societies in North America and the Asia-Pacific regions.1–5 A common pathologic feature of CKD is inflammatory cell infiltration occurring at early stages in the injured kidneys, followed by tubulointerstitial fibrosis at later stages of disease progression.2–4 Among CKD, upper urinary tract obstruction resulting in renal dysfunction is a significant clinical problem in both adult and pediatric populations. Pathologic manifestations of renal obstructive injury include tubular dilation, tubular cell apoptosis,5 as well as progressive interstitial fibrosis and massive macrophage infiltration in the obstructed kidney.2 Classically activated macrophages (M1), characterized by high major histocompatibility complex class II expression with strong interleukin-12 (IL-12) and IL-23 production, produce nitric oxide, reactive oxygen species, and proinflammatory cytokines, such as IL-1, IL-6, and tumor-necrosis factor (TNF),2,6,7 leading to tubular cell apoptosis and renal tissue injury. In contrast, alternatively activated macrophages (M2) typically have immune-suppressive activity and express arginase (Arg)-1, promoting cell proliferation and collagen production. Besides, YM-1,8,9 Mannose receptor,10 galectin-3,11 and transforming growth factor-b12 also have been reported to be preferentially expressed on M2 cells. Microenvironmental cytokines, glucocorticoid hormones, and surrounding pathogens and apoptotic cells regulate M1/M2 differentiation.10,13,14 Kidney International (2014) 86, 1174–1186

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

Triggering receptor expressed on myeloid cells (TREM) is an immunoglobulin-like family whose members have a critical role in modulating infection-induced inflammation.15,16 A common feature of TREM downstream signaling is the link with adaptor DNAX activation protein (DAP)-12.16,17 Activation of TREM leads to the production of M1 proinflammatory cytokines.16,18 TREM-1 is the best-characterized member in the TREM family. Treatment with soluble TREM-1 decreases inflammation and mortality in high-dose lipopolysaccharide-mediated septic shock model.15 Clinically, the levels of soluble TREM-1 in serum and bronchioalveolar lavage fluid have been suggested as a sensitive and specific predictor of bacterial pneumonia in humans.19 Previously, we also reported that TREM-1-mediated bacterial clearance in the small intestine is an important immune response against K. pneumoniae.20 A putative pathogenic role of TREM-1 was reported in rheumatoid arthritis21 and inflammatory bowel disease.22 However, the role of TREM-1 in the nephropathy of CKD has not yet been characterized. A mouse model of experimental unilateral ureteral obstruction (UUO) is characterized by infiltrating macrophages and the different stages of obstructed nephropathy, which accelerates with disease progression.23 In this study, we examine TREM-1 expression and function in UUO kidneys by using Trem-1 knockout (KO) mice. Our results demonstrate a novel role of TREM-1 in regulating granulocytemacrophage colony-stimulating factor (GM-CSF)/iNOS and M1 polarization. In addition, correlations of TREM-1 expression, macrophage profiling, and severity of renal injury in kidney specimens from patients of obstructive nephropathy further support that TREM-1 is a critical factor in obstructive nephritis. RESULTS Depleting TREM-1 ameliorates the UUO renal pathology

TREM-1 mRNA is not detectable in sham control kidneys but is significantly upregulated in the obstructed kidneys harvested from wild-type (WT) C57BL/6 mice at day 7 and day 14 after unilateral ureteral obstruction (UUO; Figure 1a), and TREM-1 protein is mainly detected in a subpopulation of F4/80 þ cells within the renal tubulointerstitium after UUO (Figure 1c and d). To assess the in vivo role of TREM-1 in UUO, we compared the histological renal changes between WT and Trem-1 KO mice (Supplementary Figure S1 online). TREM-1 deletion in KO kidneys is confirmed by quantitative reverse transcriptase-PCR (Figure 1a) and immunohistochemical analysis (Figure 1b). Histologically, UUO induces marked renal injury in WT mice (characterized by cortical tubular dilation with tubular epithelial cell necrosis, brush border loss, intratubular cast formation, and interstitial fibrosis), whereas Trem-1 KO mice show significantly less renal damage after UUO (Supplementary Figure S2 online). Analyses by PAS and Masson’s trichrome staining also show drastic tubular injuries at day 7 (54±9%) and day 14 (81±6%) in WT kidneys after UUO in comparison with the sham group Kidney International (2014) 86, 1174–1186

basic research

(Figure 2a and b), accompanied by increased interstitial collagen deposition after UUO at day 7 (9.1±2.6) and day 14 (34.9±5.7) (Figure 2c and d). In contrast, TREM-1 deficiency results in significantly less tubular injuries (30±7%; 50±13%) at day 7 and day 14 after UUO, and lower interstitial fibrosis (12.8±5.4) at day 14 after UUO (Figure 2a–d). Tubular cell injury was further examined by using terminal deoxynucleotidyl transferase–mediated dUTP nickend labeling and Ki67 staining methods. In Trem-1 KO kidneys, the numbers of apoptotic and proliferating cells are significantly lower than those in WT ones (Supplementary Figure S3a and b online). In addition, massive a-SMA þ myofibroblasts accumulate in WT UUO kidneys but are relatively less in Trem-1 KO kidneys isolated from day-14UUO mice (Figure 2e and f). TREM-1 is critical for regulating macrophage polarization but not recruitment

Renal inflammatory cell infiltration is a prominent feature associated with the pathogenesis of UUO. Renal macrophage recruitment was accessed by determining the distribution of F4/80 þ macrophages. Interestingly, similar levels of F4/80 þ cells are found in WT and Trem-1 KO-obstructed kidneys (Supplementary Figure S4a–c online). A recent report suggests that TREM-1 has an important role in regulating neutrophil trafficking to amplify inflammation against bacterial infection,24 leading us to examine the neutrophil infiltration after UUO. However, only very few Ly6G þ neutrophils are observed in the obstructed kidneys, and no significant difference can be found in the numbers of infiltrating neutrophils between WT and Trem-1 KO groups (Supplementary Figure S4d and e online). Thus, TREM-1 is dispensable for macrophage and neutrophil renal recruitment upon UUO. M1 polarization has been suggested as a key step in UUOmediated renal damage.2 Whether TREM-1 regulates the balance of ‘classical’ and ‘alternative’ activation of macrophages in UUO remains to be investigated. M1 cells were evaluated by iNOS expression, which is presented in F4/80 þ infiltrating cells and tubular epithelial cells on WT renal sections, particularly at day 14 after UUO. By contrast, it is markedly attenuated in the cortex and medulla of UUO Trem-1 KO kidneys (Figure 3a and b). Accordingly, the mRNAs of iNOS, TNF, IL-1b, and IL-6 (M1 markers) are highly expressed in WT UUO kidneys, but the expression is significantly lower in Trem-1 KO UUO kidneys (Figure 3c). Furthermore, the expression of iNOS, TNF, and IL-1b mRNA is significantly lower in Trem-1 KO renal macrophages freshly isolated from UUO kidneys compared with WT cells (Supplementary Figure S5a online). In parallel, M2 differentiation was evaluated by Arg-1 staining on renal sections. Unlike in WT samples, which only show a mild staining, Arg-1 is highly induced and significantly elevated in renal F4/80 þ infiltrating cells in Trem-1 KO kidneys at day 7 and day 14 after UUO (Figure 3d and e). The mRNA expression levels of Arg-1, YM-1, IL-10, and Mannose receptor (M2 markers) are significantly higher in Trem-1 KO UUO kidneys in 1175

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

a

b TREM-1

Sham

WT KO P<0.001 mRNA (relative fold)

UUO Day 7

Day 14

WT 200 100 0

P<0.001

KO Sham 7 14 UUO (days)

c

F4/80 IHC

TREM-1 IHC

UUO kidney from WT mouse at day 14. (×1000)

d

F4/80

TREM-1

DAPI

Merge

Figure 1 | TREM-1 is upregulated and colocalized with macrophages in UUO kidneys. Unilateral ureteral obstruction (UUO) surgery was performed on wild-type (WT) and Trem-1 knockout (KO) mice and they were killed at day 7 and day 14. (a) Quantitative reverse transcriptase-PCR analysis and (b) immunohistochemical (IHC) staining of TREM-1 expression in the obstructed kidney after UUO, compared with sham control. (c) F4/80 and TREM-1 IHC analyses on renal sections harvested from WT mice at 14 days after UUO (1000). (d) Renal tissue was obtained at 14 days after UUO and stained for 4’6-diamidino-2-phenylindole (DAPI; blue), F4/80 (red), and TREM-1 (green). Representative merge images are shown at an original magnification of 400. (a) Representative results were shown as mean±s.d. (b and d) Scale bars ¼ 50 mm. (WT, n ¼ 9; Trem-1 KO, n ¼ 6 in each group.)

comparison with WT UUO ones (Figure 3f), whereas the expressions of IL-4 and transforming growth factor-b are comparable between the two groups (data not shown). However, the increase of M2 marker mRNA expression is not observed in isolated Trem-1 KO UUO renal macrophages (Supplementary Figure S5b online), suggesting that a complete renal microenvironment is crucial to maintain the M2-related marker gene expression. The renal protein levels of M1 and M2 markers were also determined by western blotting. Trem1 KO kidneys contain lower iNOS and TNF but higher Arg-1 and IL-10 proteins in comparison with WT samples at day 7 and day 14 after UUO (Supplementary Figure S6 online). UUO-injured kidney lysates trigger TREM-1-dependent M1 differentiation

The factor(s) in UUO kidney that can trigger TREM-1 thereby further mediating the downstream M1 polarization remains unknown. Homogenates prepared from WT UUO kidneys 1176

were used for ex vivo stimulation of bone marrow–derived macrophages (BMMs) from WT or Trem-1 KO mice (Figure 4a). Treating WT BMMs with day-14-UUO kidney lysate (KL) significantly augments higher TREM-1 expression than sham lysate does (Figure 4b). We next evaluate the role of GM-CSF, a well-known M1-like macrophage differentiation factor,25–28 in TREM-1-mediated cell polarization in response to UUO KL treatment. WT but not Trem-1 KO BMMs show a higher level of GM-CSF induction in response to day-7- and day-14-UUO KLs (Figure 4c). Accordingly, M1-associated iNOS is highly induced in day-14-UUO KL–treated WT cells but not in the sham lysate–treated ones. In contrast, day-14-UUO KL–treated Trem-1 KO BMMs only express a lower level of iNOS (Figure 4d) accompanied by a significant upregulation of Arg-1 while stimulate cells with day-7-UUO KL (Figure 4e), supporting that UUO-lysates contain stimuli that trigger TREM-1-mediated regulation on M1/M2 differentiation. Kidney International (2014) 86, 1174–1186

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

a

UUO

b Sham

Day 7

Day 14

Tubular injury (%)

WT KO P<0.001

100

WT

P<0.001

80 60 40 20 0

KO 7 14 Sham UUO (days) UUO

d

c

WT KO

Interstitial collagen deposition (%)

P<0.001

50

Sham

Day 7

Day 14

WT

40 30 20 10

KO

0 Sham 7 14 UUO (days)

e

(×400)

f

α-SMA positive area (%)

WT KO

UUO Sham

Day 7

Day 14

P <0.001

30 25 20 15 10 5 0

WT

KO Sham 7 14 UUO (days) (×400)

Figure 2 | TREM-1 has an important role in the progression of renal pathogenesis after UUO. Unilateral ureteral obstruction (UUO) surgery was performed on wild-type (WT) and Trem-1 knockout (KO) mice and they were killed at day 7 and day 14. Representative kidney tissue sections stained with (a and b) periodic acid–Schiff, (c and d) Masson’s trichrome, and (e and f) immunohistochemical staining of the a-smooth muscle actin located in the tubulointerstitium at day 7 and day 14 after UUO, compared with sham control. Quantitative positive area of (a) tubule injury, (c) interstitial collagen deposition, and (e) interstitial fibrosis (a-SMA positive) were shown as mean±s.d. (WT, n ¼ 9; Trem-1 KO, n ¼ 6 in each group). (b, d, and f) Scale bars ¼ 50 mm. (b) Black arrowhead, brush border; white arrowhead, loss of brush border.

GM-BMMs preferentially express TREM-1, which mediates iNOS and proinflammatory cytokine induction

M1-like macrophages can be in vitro differentiated by treating bone marrow cells with GM-CSF (GM-BMMs).25,27,28 Unexpectedly, we observed that GM-BMMs preferentially express higher TREM-1 than M-CSF-derived macrophages (M-BMMs) (Figure 5a and b). To evaluate the contribution of TREM-1 in the plasticity of macrophages, differentiated M-BMMs were treated with GM-CSF to convert them into the inflammatory stage, and the inflammatory GM-BMMs Kidney International (2014) 86, 1174–1186

were then reversed back to the resting stage by replacing them in M-CSF culture condition. Consistently, GM-CSF treatment induces higher levels of TREM-1 in M-BMMs (Figure 5c), whereas replacing GM-BMMs in M-CSF culture attenuates their TREM-1 expression (Figure 5d). The effects of TREM-1 triggering on the induction of inflammation cytokines and M1/M2 markers were investigated by cross-linking TREM-1 with plate-bond-specific agonistic antibody. Cross-linked TREM-1 alone triggers TNF secretion from WT but not from KO GM-BMMs 1177

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

a

b

iNOS IHC

KO

WT

KO

Day 14

F4/80 iNOS DAPI

UUO

iNOS

Day 7

F4/80

Sham

WT

c

iNOS mRNA (relative fold)

TNF P<0.01

40

d

WT KO

Sham 7 14 UUO (days)

P<0.01

P< 0.05

200

40 100

20 0

Sham 7 14 UUO (days)

e

KO

Sham 7 14 UUO (days)

Arg-1 IHC

0

Sham 7 14 UUO (days)

WT

KO

Day 14

F4/80 Arg-1 DAPI

UUO

Arg-1

Day 7

F4/80

Sham

WT

0

IL-6

IL-1β P<0.001

60

20 0

120

f mRNA (relative fold)

Arg-1 400

WT KO

P<0.01

30

YM-1

IL-10 150 100

15

200

0

MR 8

Sham 7 14 UUO (days)

0

P<0.05

P<0.05

4

50

P<0.05

0 Sham 7 14 UUO (days)

P<0.01

P<0.01

Sham 7 14 UUO (days)

0

Sham 7 14 UUO (days)

Figure 3 | TREM-1 is crucial in modulating macrophage polarization upon UUO. (a and d) Renal sections were obtained at day 14 after unilateral ureteral obstruction (UUO) and stained for 4’6-diamidino-2-phenylindole (blue), F4/80 (red), and (a) iNOS (green) or (d) Arg-1 (green). Representative merge images are shown for both genotypes at an original magnification of 1000. Scale bars ¼ 10 mm. n ¼ 4 in each group. (b and e) Representative renal sections were obtained at day 7 and day 14 after UUO and immunohistochemically analyzed for (b) iNOS and (e) Arg-1 expression (400). Scale bars ¼ 50 mm. WT, n ¼ 9; Trem-1 KO, n ¼ 6 in each group. (c and f) Expressions of (c) M1 markers (iNOS, TNF, IL-1b, and IL-6) and (d) M2 markers (Arg-1, YM-1, IL-10, and Mannose receptor) in UUO kidneys at day 7 and day 14 were determined by quantitative reverse transcriptase-PCR (WT, n ¼ 9; Trem-1 KO, n ¼ 6 in each group). IL, interleukin; KO, knockout; TNF, tumor-necrosis factor.

1178

Kidney International (2014) 86, 1174–1186

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

a

Sham KLs 7-day UUO KLs 14-day UUO KLs

WT or TREM-1 KO BM cells

qRT-PCR Day 5

Day 0

Day 6

M-CSF (20 ng/ml)

mRNA (relative fold)

TREM-1 6

WT KO

c mRNA (relative fold)

b

P<0.05

4 2 0

16

P<0.05

8

P<0.01

6 4 2 0 Sham

7 14 UUO (days) Treatment with KL

Sham

7 14 UUO (days) Treatment with KL

e mRNA (relative fold)

mRNA (relative fold)

P<0.01

WT KO

P<0.01

0

7 14 UUO (days) Treatment with KL iNOS

P<0.05 P<0.01

Sham

d

WT KO

GM-CSF

Arg-1 P<0.01

8

WT KO P<0.01

6 4 2

The induction of iNOS under TREM-1 crosslink in M-BMMs with M-CSF, GM-CSF, or day-14 UUO-KL stimulation was next examined. As predicted, TREM-1 triggers more iNOS in M-BMMs cultured with GM-CSF than M-CSF (Figure 5i), and it also further augments iNOS induction in UUO KL–treated M-BMMs (Figure 5j). HMGB1 was recently reported to be a putative ligand for TREM-1.29 We investigated whether treating GM-BMMs with recombinant HMGB1 is sufficient to induce iNOS through TREM-1. However, neither group of cells showed significant iNOS induction after HMGB1 treatment (data not shown), suggesting that recombinant HMGB1 alone is not sufficient to trigger iNOS via TREM-1. More investigations are needed to identify the TREM-1 ligand contributing to M1 cell polarization in UUO. The role of DAP12 in TREM-1-mediated M1 polarization was also determined by comparing the iNOS and Arg-1 mRNA expression in WT, Dap12 KO,30 and Trem-1 KO GM-BMMs with TREM-1 crosslink. Although TREM-1 expression is comparable in WT and Dap12 KO GM-BMMs (Supplementary Figure S8a and b online), the inductions of iNOS and Arg-1 mediated by TREM-1 are completely abolished in Dap12 KO GM-BMMs (Supplementary Figure S8c and d online), supporting that DAP12 is crucial for mediating TREM-1-triggered cell activation.

0 Sham 7 14 UUO (days) Treatment with KL

Figure 4 | UUO-injured WT kidneys contain an unidentified TREM-1 ligand that triggers M1 macrophage differentiation. (a) Experimental design of ex vivo kidney lysate–induced macrophage polarization. Homogenates of WT UUO kidney (KL) in phosphatebuffered saline were isolated and used as stimuli for the subsequent ex vivo BMM stimulation. M-BMMs derived from WT and Trem-1 KO mice were incubated with KLs from sham, day-7, and day-14 UUO–injured kidneys for 24 h. Expression levels of (b) TREM-1, (c) GM-CSF, (d) iNOS, and (e) Arg-1 mRNA in stimulated M-BMMs were determined by quantitative reverse transcriptase-PCR analysis. Representative results of three independent experiments are shown as mean±s.d. BMM, bone marrow–derived macrophage; KL, kidney lysate; KO, knockout; UUO, unilateral ureteral obstruction; WT, wild-type.

(Supplementary Figure S7a online), and it synergistically boosts the lipopolysaccharide-mediated secretion of TNF and IL-6 in WT GM-BMMs, but not in Trem-1 KO ones (Supplementary Figure S7a and b online). Similar results are also obtained from mRNA analyses of TNF and iNOS expressions (Figure 5e and f). By contrast, Arg-1 is slightly induced in WT GM-BMMs under TREM-1 stimulation (Figure 5g) in a much milder way. Protein levels of iNOS and Arg-1 were further checked by immunoblotting. Accordingly, TREM-1 crosslink only induced iNOS induction in WT but not in Trem-1 KO GM-BMMs. Arg-1 induction is not observed in both groups of GM-BMMs, and neither iNOS nor Arg-1 protein are detected in WT and Trem-1 KO M-BMMs (Figure 5h). Kidney International (2014) 86, 1174–1186

basic research

GM-CSF is a crucial factor downstream of TREM-1 for M1 induction

M-BMMs could be polarized into M1 under UUO-KL treatment through a TREM-1-dependent process (Figure 4d). Accordingly, crosslink of TREM-1 triggers the induction of GM-CSF (Figure 6a). To determine the role of GM-CSF in UUO KL–mediated M1 polarization, antagonist antibody against GM-CSF was added in the culture. Depletion of GMCSF significantly reduces UUO KL–triggered M1 differentiation (Figure 6b). Neutralization of GM-CSF also suppresses UUO KL–induced TREM-1 and GM-CSF expression in WT BMMs (Figure 6b). Thus, modulation of GM-CSF activity in UUO-KL stimulation affects TREM-1-mediated GM-CSF induction and blocks the positive feedback of TREM-1 enhancement and M1 differentiation. Adoptive transfer of WT GM-BMMs into Trem-1 KO mice deteriorates UUO pathogenesis

To verify the importance of TREM-1 expressed on M1-like macrophages in mediating the UUO pathogenesis in vivo, we adaptively transferred carboxyfluorescein succinimidyl ester-labeled WT or Trem-1 KO GM-BMMs into UUOtreated Trem-1 KO mice (Figure 7a).31,32 Reconstitution of carboxyfluorescein succinimidyl ester-labeled cells, which are TREM-1 þ (Figure 7b) and F4/80 þ (data not shown), can be observed in the kidneys of Trem-1 KO mice that received the WT GM-BMMs. In agreement with our previous findings, transferring WT GM-BMMs, but not Trem-1 KO ones, into Trem-1 KO recipient mice can intensify the tubular injury, 1179

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

b

TREM-1

GM-BMM

P <0.001

8 6

M-BMM

Counts

mRNA (relative fold)

a

4 2

0 100

0 M-BMM

GM-BMM

0 101

103 104

102

100

101

102

103 104

TREM-1

d

TREM-1 M-CSF GM-CSF

P <0.01

2

0

TREM-1 4

2

0

WT KO M-BMM TNF

f

P <0.01

2

1

0

g P <0.01

80

40

0

Ctrl anti-TREM-1 Crosslink

h

KO WT GM-BMM

iNOS mRNA (relative fold)

mRNA (relative fold)

e

GM-CSF M-CSF

P <0.01

IgG

40

0 Ctrl anti-TREM-1 Crosslink

GM-BMM

–

M-BMM TR1 KO

TR1

WT KO

80

Ctrl anti-TREM-1 Crosslink

WT Ab

Arg-1 mRNA (relative fold)

4

mRNA (relative fold)

mRNA (relative fold)

c

–

IgG

WT TR1

–

TR1 KO

TR1

–

TR1 iNOS Arg-1 α-tubulin

i

j

iNOS P <0.05

WT KO

Fold (crosslink / KLS alone)

mRNA (relative fold)

6

4 2

0 anti-TREM-1

–

+ M-CSF

–

+ GM-CSF

iNOS induction 2.0

P <0.05

1.5 1.0 0.5 0 KLs

–

Sham

14-day UUO

Figure 5 | TREM-1 is preferentially expressed in GM-BMMs and modulates macrophage polarization toward M1. (a and b) TREM-1 expression in WT M-BMMs and GM-BMMs was determined by (a) quantitative reverse transcriptase-PCR and (b) fluorescence-activated cell sorting analyses (gray filled: nonstained; dashed line: isotype control; black line: TREM-1). (c and d) TREM-1 mRNA expression in WT and Trem-1 KO (c) M-BMMs and (d) GM-BMMs treated with M-CSF or GM-CSF for 24 h. (e) TNF, (f) iNOS, and (g) Arg-1 mRAN expression levels in WT and Trem-1 KO GM-BMMs were left untreated or stimulated with anti-TREM-1 agonistic antibody for 4 h. (h) iNOS and Arg-1 expression levels were determined by immunoblotting analyses on WT and Trem-1 KO GM-BMMs and M-BMMs. a-tubulin served as a loading control. Representative results of three independent experiments are shown. (i) iNOS expression in WT and Trem-1 KO M-BMMs (maintained in M-CSF or treated with GM-CSF) were left untreated or costimulated with anti-TREM-1 agonistic antibody for 24 h. The expression levels of indicative genes were determined by quantitative reverse transcriptase-PCR. (j) iNOS induction in WT M-BMMs (left untreated or treated with KLs from sham and day-14 UUO samples) costimulated with or without anti-TREM-1 agonistic antibody for 24 h. The fold induction of iNOS was calculated by normalizing with results from cells without TREM-1 crosslink. (a, c–g, i, and j) Representative results of three independent experiments are shown as mean±s.d. BMM, bone marrow–derived macrophage; GM-CSF, granulocyte-macrophage colony-stimulating factor; KL, kidney lysate; KO, knockout; UUO, unilateral ureteral obstruction; WT, wild-type.

1180

Kidney International (2014) 86, 1174–1186

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

interstitial fibrosis, and myofibroblast accumulation in kidneys of UUO Trem-1 KO mice (Figure 7c and d). Association of TREM-1 expression, iNOS expression, and renal injury in human obstructive nephropathy

To validate the pathogenic role of TREM-1 in human noninfectious nephritis, nephrectomized specimens of the affected kidneys were obtained from nondiabetic, nonhypertensive patients with unilateral partial (n ¼ 3) and complete (n ¼ 6) ureteral obstruction (Table 1). The causes of obstruction are renal stone and ureteral cancer. The normal tissues of nephrectomized kidneys from renal cell carcinoma patients served as controls (n ¼ 4). There are no differences in age and gender distribution among the three groups. However, baseline glomerular filtration rate (GFR) measured by the technetium-99m-diethylene-pentaacetate method is significantly lower in the affected kidneys than in the contralateral nonaffected kidneys. On the basis of the histopathologic analysis, tubular injury score and collagen deposition in the interstitium are positively correlated with the severity of ureteral obstruction (Supplementary Figure S9 online), and negatively correlated with glomerular filtration rate of the affected kidneys before surgery (Table 1). Consistent with the results from the UUO mouse model, numbers of CD68-positive macrophages and TREM-1positive cells are the highest in the specimens of completely obstructed kidneys, followed by partially obstructed kidneys and then normal renal tissues (Figure 8a and b and Table 1). TREM-1-positive cells in obstructed kidneys are mainly observed in the interstitia, which are also positively stained for the macrophage cell marker CD68 by immunofluorescent staining (Figure 8c). The TREM-1-staining profile is also highly correlated with the profile of iNOS in samples of completely ureteral obstructed kidneys (Figure 8d). By contrast, very few TREM-1 and iNOS-stained cells are observed in normal renal tissues. Double staining demonstrates that the numbers of infiltrating cortical interstitial

8 mRNA (relative fold)

DISCUSSION

TREM-1 is crucial in infection-associated inflammation,20 but its role in sterile inflammation remains to be determined. Deletion of TREM-1 in mice attenuates diethylnitrosamineinduced hepatocellular carcinogenesis. Loss of TREM-1 attenuates Kupffer cell activities including inflammatory cytokine production and signal induction, resulting in diminished liver injury after diethylnitrosamine exposure.29 In the present study, we report an independent Trem-1-deficient mouse model and reveal a novel role of TREM-1 in the pathogenesis of UUO nephritis. Similar to a previous model,24,29 the basic immune parameters, such as immune cell counts, WBC population, and the weight of lymphoid organs, in our Trem-1 KO mice are comparable with WT mice. In addition, we also show that TREM-1 is dispensable for renal macrophage and neutrophil recruitment upon UUO. Our finding regarding the involvement of TREM-1/ DAP12 in UUO nephritis was further supported by a recent report that shows that DAP12 (partly through TREM-1/3) is involved in renal inflammation during UUO.33 However, in contrast to our results that Trem-1 deficiency ameliorates both renal injury/inflammation and fibrosis in mice with UUO, Tammaro A. et al. reported that the renal fibrosis was comparable in WT and Trem-1/3 double KO mice. Of note, there are about 40 DAP12-associated receptors expressed on various types of cells.18 Some of these receptors have opposite roles in modulating inflammation and tissue fibrosis (such as TREM-1 vs. TREM2). Whether TREM-3 has inflammatory

Blocking with isotype Ctrl Blocking with anti-GM-CSF

b GM-CSF

P <0.001

WT KO

6 4 2

6 mRNA (relative fold)

a

macrophages positive for both TREM-1 and iNOS are positively associated with tubular injury and interstitial fibrosis, and negatively with glomerular filtration rate in the ureteral obstructed kidneys (Table 1). Taken together, our results support that TREM-1 has a crucial role in mediating M1-mediated pathogenesis of renal injury in obstructive nephropathy.

iNOS P <0.001

3

Ctrl anti-TREM-1 Crosslink

TREM-1 P <0.01

GM-CSF P <0.01

0

0 14-day UUO KLs

12

6

3

0

0

6

14-day UUO KLs

14-day UUO KLs

Figure 6 | Neutralization of GM-CSF inhibits TREM-1-mediated M1 polarization. GM-CSF mRNA expression in WT and Trem-1 KO (a) GM-BMMs left untreated or stimulated with anti-TREM-1 agonistic antibody for 4 h. (b) iNOS, TREM-1, and GM-CSF expressions in WT M-BMMs stimulated with day-14-UUO KLs in the presence of GM-CSF-neutralizing antibodies (1 mg/ml, R&D systems) or isotype control (1 mg/ml) for 24 h. The expressions of indicated genes were determined by quantitative reverse transcriptase-PCR. Representative results of three independent experiments were shown as mean±s.d. BMM, bone marrow–derived macrophage; GM-CSF, granulocyte-macrophage colony-stimulating factor; KL, kidney lysate; KO, knockout; UUO, unilateral ureteral obstruction; WT, wild-type. Kidney International (2014) 86, 1174–1186

1181

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

or anti-inflammatory role in modulating tissue fibrosis during UUO remains to be investigated. Thus, the controversy observed between Trem-1 KO and Trem-1/3

a

WT/TREM-1KO GM-BMM (CFSE) UUO transfer

Day 0

WT/TREM-1KO GM-BMM (CFSE) transfer

1

Sacrifice

8 Saline ↓ TREM-1–/–

Saline ↓ TREM-1+/+

14 TREM-1+/+ ↓ TREM-1–/–

TREM-1–/– ↓ TREM-1–/–

Merge

TREM-1

CFSE

b

double KO mice may be a result of functional compensation between TREM-1- and TREM-3. We have examined the Trem-3 expression in WT and Trem-1 KO mice by

UUO (14 days) Saline ↓ TREM-1+/+

TREM-1+/+ ↓ TREM-1–/–

TREM-1–/– ↓ TREM-1–/–

α-SMA

Masson trichrome

PAS

c

Saline ↓ TREM-1–/–

UUO (14 days)

P<0.05

TREM-1 S S +/+ –/– ↓ ↓ ↓ ↓ TREM-1 +/+ –/– –/– –/–

1182

50 40 30 20 10 0

f P<0.001

TREM-1 S S +/+ –/– ↓ ↓ ↓ ↓ TREM-1 +/+ –/– –/– –/–

Positive α-SMA stain area (%)

e

Positive masson’s trichrome stain area (%)

100 90 80 70 60 50 40

Tubular injury (%)

d

40 30

P<0.01

20 10 0 TREM-1 S S +/+ –/– ↓ ↓ ↓ ↓ TREM-1 +/+ –/– –/– –/–

Kidney International (2014) 86, 1174–1186

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

quantitative reverse transcriptase-PCR, and found that the expression of Trem-3 is comparable in WT and Trem-1 KO GM-BMMs and UUO kidney–isolated macrophages (data not shown), which further supports the plausibility of this hypothesis. Infiltrated M1 and M2 macrophages can mediate renal injury and tissue repair, respectively.2,3 Initially, infiltrating macrophages are predominantly in M1 type to secrete proinflammatory cytokines and induce the death of tubular

Table 1 | Baseline clinical characteristics, pathology index, and cell population analyses among patients without or with obstructive nephropathy due to partial and complete ureteral obstruction

Clinical characteristics Age (years) Gender (M/F) Causes Stone, n (%) Ureteral cancer, n (%)

No ureteral obstruction (n ¼ 4)

Partial ureteral obstruction (n ¼ 3)

59±25 2/2

58±29 2/1

— —

2 (66.7) 1 (33.3)

4 (66.7) 2 (33.3)

31.2±18.7a 52.2±12.7

9.4±3.4a,b 56.2±24.5

4.5±3.6 0.1±0.0

25.6±2.6c 1.4±0.6c

84.3±12.0b,d 3.2±0.4b,d

0.0±0.0 0.0±0.0 0.0±0.0 0.0±0.0

2.1±0.3c 1.6±0.1c 0.8±0.8 1.6±0.2c

18.9±4.8b,d 13.5±3.4b,d 10.7±2.3b,d 13.3±3.2b,d

0.0±0.0

0.2±0.1

9.8±1.7b,d

GFR by Tc-99m-DTPA (ml/min per 1.73 m2) Affected kidney — Nonaffected kidney 53.5±4.5 Renal pathology index Tubular injury (%) Interstitial fibrosis score Cell population CD68 þ (cells/HPF) TREM-1 þ (cells/HPF) iNOS þ (cells/HPF) CD68 þ , TREM-1 þ (cells/HPF) TREM-1 þ , iNOS þ (cells/HPF)

Complete ureteral obstruction (n ¼ 6) 60±23 4/2

Abbreviations: F, female; GFR; glomerular filtration rate; HPF, high-powered field; M, male; Tc-99m-DTPA, technetium-99m-diethylene-pentaacetate; TREM, triggering receptor expressed on myeloid cells. Values are means ± s.d. of positive cells/400 high-powered field. The statistical analysis was performed by ANOVA with post hoc pairwise comparison. a Po0.05, versus nonaffected kidney in partial and complete ureteral obstruction, respectively. b Po0.001, versus partial ureteral obstruction. c Po0.05, versus no ureteral obstruction. d Po0.001, versus no ureteral obstruction.

basic research

cells, leading to tissue injury and ultimately to the development of scarring and fibrosis. Intriguingly, the absence of TREM-1 ameliorates M1-mediated tissue injury but induces a switch in infiltrated macrophages from M1 to M2. TREM-1-deficient macrophages in obstructive kidneys express Arg-1 and lead to tissue remodeling and repair (Supplementary Figure S10 online). Our data demonstrate that TREM-1 critically modulates M1 polarization both in vivo and ex vivo. Accordingly, TREM-1 is preferentially expressed by M1-like GM-BMMs, and crosslink of TREM-1 significantly strengthens the expression of iNOS, TNF, and GM-CSF. Clinically, TREM-1 has been identified as a novel biomarker in the assessment of pediatric malaria34 and pneumonia disease severity.19 In the present study, the histological analyses of kidney specimens from patients with obstructive nephropathy prove that the frequency of TREM-1 expression in infiltrating macrophages is significantly correlated with the disease progression from partial to complete obstruction, demonstrating that TREM-1 may serve as a new local diagnostic surrogate of human obstructive nephropathy. The identity of the natural TREM-1 ligand(s) remains a mystery.35 The genes encoding the putative ligands expressed on platelets and macrophages remain uncharacterized.36,37 HMGB1 was recently found as a TREM-1-interacting protein.29 However, treating cells with recombinant HMGB1 alone is not sufficient to mediate TREM-1regulated M1 polarization. It cannot be ruled out that the activity of recombinant HMGB1 is different from native HMGB1 (with various posttranslation modulation), and treating cells with isolated HMGB1 from inflamed tissues may be required for inducing TREM-1-mediated M1 polarization. Further fractionation and characterization on UUO KLs will enable us to identify the novel TREM-1triggering danger-associated molecular pattern(s). In summary, a novel positive feedback cycle mediated by TREM-1 and GM-CSF was revealed to enhance M1 polarization under UUO (Supplementary Figure S10 online). Upon UUO, injured tissue-derived GM-CSF initially induces the expression of TREM-1, followed by TREM-1-mediated augmentations of GM-CSF and other M1-promoting factors, and finally amplifies the polarization of M1 responses leading to kidney damages. This amplification cycle mediated by TREM-1/GM-CSF may be a putative target for CKD therapeutic drug design.

Figure 7 | Histopathologic analyses of kidney tissues in Trem-1 KO UUO mice receiving carboxyfluorescein succinimidyl ester (CFSE)labeled WT or Trem-1 KO GM-BMMs. (a) Experimental design for adoptive transfer of CFSE-labeled GM-BMMs to UUO Trem-1 KO mice. After UUO surgery, cultured WT or Trem-1 KO GM-BMMs labeled with CFSE were intravenously injected (3106 GM-BMMs/mouse) into Trem-1 KO mice at day 1 and day 8 (the same amount of saline was injected in the first two groups of mice as controls). Mice were killed at day 14, and renal tissues were harvested for analyses. (b) Renal sections were stained for TREM-1 (red) and 4’6-diamidino-2-phenylindole (blue). Representative merge images are shown at an original magnification of 400. (c) Representative results of renal sections stained with PAS, Masson’s trichrome, and a-smooth muscle actin. (400) (d–f) Quantitative positive area of (d) tubule injury, (e) interstitial collagen deposition, and (f) interstitial fibrosis. Results were shown as mean±s.d. Scale bars ¼ 50 mm (n ¼ 4 in each group). BMM, bone marrow–derived macrophage; GM-CSF, granulocyte-macrophage colony-stimulating factor; KO, knockout; UUO, unilateral ureteral obstruction; WT, wild-type. Kidney International (2014) 86, 1174–1186

1183

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

a

b

TREM-1 IHC

DAPI

CD68

TREM-1

DAPI CD68 TREM-1

Isotype ctrl

DAPI

iNOS

TREM-1

DAPI iNOS TREM-1

Isotype ctrl

Complete obstruction

Partial obstruction

Normal

c

CD68 IHC

Complete obstruction

Partial obstruction

Normal

d

(×400)

Figure 8 | Immunohistochemical and immunofluorescene staining of kidney specimens in patients with or without obstructive nephropathy. TREM-1 expression was characterized by renal interstitial macrophages in the nephrectomized kidney tissues from patients with partial and complete ureteral obstruction. (a) Accumulation of infiltrated CD68 þ macrophages, as well as (b) the expression of TREM-1, in the renal interstitium of kidney tissues was determined by immunohistochemistry. Double immunofluorescence staining revealed the (c) colocalization of CD68 þ macrophages (green) and TREM-1-expressing cells (red), and (d) colocalization of iNOS-expressing macrophages (green) and TREM-1 þ cells (red) in the tubulointerstitium of patients with partial and complete ureteral obstruction. (c and d) Nuclei were stained with the DNA-intercalating dye 4’6-diamidino-2-phenylindole (blue); isotype Ctrl means the control staining using control antibodies with the same isotype. Scale bars ¼ 50 mm.

MATERIALS AND METHODS Animals Age-matched (8- to 10-week-old) male WT C57BL/6J, Trem-1 KO (B6.129P2-Trem-1tm1Mak), and Dap12 KO (generated in the 129/SvJ and C57BL/6 hybrid background as described29 and backcrossed in C57BL/6J background) were maintained in the animal center of NYMU in accordance with the IACUC guidelines. Experimental UUO animal model UUO surgery was performed by making a mid-abdominal incision under intraperitoneal anesthesia with pentobarbital (50 mg/kg). The 1184

left ureter was dissected free and ligated with 5.0 silk. Age-matched WT and Trem-1 KO mice, incision and suture of the mid-abdominal wall without kidney ligation were used as sham control. Pathological examinations of the kidneys The kidney was cut apart. Half of the tissue was frozen in liquid nitrogen for protein and RNA analyses. The other half was fixed in 4% paraformaldehyde for 24 h and embedded in 4-mm-thick paraffin sections. After deparaffinization, the sections were stained with hematoxylin and eosin, periodic acid–Schiff (PAS) reagent (Sigma-Aldrich, St Louis, MO), and Masson’s Trichrome (Sigma-Aldrich) stains. Kidney International (2014) 86, 1174–1186

basic research

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

Immunohistochemical staining Sections were treated with 0.3% hydrogen peroxide to quench the endogenous peroxidase activity. Titrated primary antibodies against the following antigens were used: for mouse, F4/80, Ly6G (eBioscience, San Diego, CA), TREM-1 (R&D Systems, Minneapolis, MN), Ki67 (BioLegend, San Diego, CA), iNOS, Arg-1, and a-SMA (Santa Cruz Biotechnology, Santa Cruz, CA); for humans, TREM-1 (R&D Systems), CD68 (KP1, DAKO-CD68, Denmark), and iNOS (Santa Cruz Biotechnology). Immunoreactivity was detected with the Envision avidin-biotin-free HRP-labeled polymer (Dako Cytomation) and visualized using the diaminobenzidine kit (Lab Vision, Fremont, CA). Statistical analysis Descriptive statistics included mean±s.d. The comparisons between UUO and sham groups (or WT and KO groups) were performed by independent Student’s t-test or an indicative method for a difference using the Statistical Package of Social Science (SPSS 12.0, 2003; SPSS, Chicago, IL) software.

REFERENCES 1.

2. 3.

4.

5. 6. 7. 8.

9.

10.

DISCLOSURE

All the authors declared no competing interests.

11. 12.

ACKNOWLEDGMENTS

This work was supported by grants to NJC from the National Science Council (NSC97-2320-B010-032-MY2; NSC99-2320-B10-003-MY3; NSC102-2320-B -010-015)), Taipei Veterans General Hospital (V97S5001; V98S5-008; V99S5-002; V100E4-003), Yen Tjing Ling Medical Foundation (CI-100-10), and a grant (98A-C-D117) from the Ministry of Education, Aim for the Top University Plan in Taiwan. DCT is supported by grants from the National Science Council (NSC 99-2314B-010-004-MY3; NSC 102-2314-B-010-004-MY3), Taipei Veterans General Hospital (V101E4-001; V102E4-001), and Bureau of Health Promotion, Department of Health (DOH98-HP-1110). TWM is supported by grants from the Canadian Institute of Health Research, the Leukemia Lymphoma Society, the Canadian Institute of Health Research, and the Terry Fox Cancer Research Foundation.

13.

14. 15.

16. 17.

18. 19.

SUPPLEMENTARY MATERIAL Table S1. Primer sequences for real-time polymerase chain reaction. Figure S1. Generation of Trem-1 KO mice. Figure S2. Depletion of TREM-1 in mice ameliorates the progression of renal pathogenesis after UUO. Figure S3. Deletion of TREM-1 protects mice from UUO-mediated renal tubular cell apoptosis and proliferation. Figure S4. TREM-1 is dispensable in macrophage and neutrophil recruitment upon UUO. Figure S5. M1 and M2 marker mRNA expression levels in WT or Trem-1 KO macrophages isolated from 7-day or 14-day-UUO kidneys. Figure S6. Renal protein levels of M1/M2 markers at day 7 and day 14 after UUO. Figure S7. TREM-1 contributes to M1 proinflammatory cytokine production in GM-BMMs. Figure S8. DAP12 has essential roles in TREM-1-mediated iNOS induction in GM-BMMs. Figure S9. Histological analysis on specimens obtained from the nephrectomized kidneys of patients with unilateral partial and complete ureteral obstruction and the surrounded normal renal tissues of patients with renal cell carcinoma. Figure S10. TREM-1 has a critical role in the regulation macrophage polarization in UUO kidney. Supplementary material is linked to the online version of the paper at http://www.nature.com/ki Kidney International (2014) 86, 1174–1186

20.

21.

22.

23.

24.

25. 26.

27.

Lopez-Novoa JM, Martinez-Salgado C, Rodriguez-Pena AB et al. Common pathophysiological mechanisms of chronic kidney disease: therapeutic perspectives. Pharmacol Ther 2010; 128: 61–81. Ricardo SD, van Goor H, Eddy AA. Macrophage diversity in renal injury and repair. J Clin Invest 2008; 118: 3522–3530. Schiffer L, Bethunaickan R, Ramanujam M et al. Activated renal macrophages are markers of disease onset and disease remission in lupus nephritis. J Immunol 2008; 180: 1938–1947. Vielhauer V, Kulkarni O, Reichel CA et al. Targeting the recruitment of monocytes and macrophages in renal disease. Semin Nephrol 2010; 30: 318–333. Klahr S, Morrissey J. Obstructive nephropathy and renal fibrosis. Am J Physiol Renal Physiol 2002; 283: F861–F875. Benoit M, Desnues B, Mege JL. Macrophage polarization in bacterial infections. J Immunol 2008; 181: 3733–3739. MacMicking J, Xie QW, Nathan C. Nitric oxide and macrophage function. Annu Rev Immunol 1997; 15: 323–350. Chang NC, Hung SI, Hwa KY et al. A macrophage protein, Ym1, transiently expressed during inflammation is a novel mammalian lectin. J Biol Chem 2001; 276: 17497–17506. Raes G, Noel W, Beschin A et al. FIZZ1 and Ym as tools to discriminate between differentially activated macrophages. Dev Immunol 2002; 9: 151–159. Stein M, Keshav S, Harris N et al. Interleukin 4 potently enhances murine macrophage mannose receptor activity: a marker of alternative immunologic macrophage activation. J Exp Med 1992; 176: 287–292. Dumont P, Berton A, Nagy N et al. Expression of galectin-3 in the tumor immune response in colon cancer. Lab Invest 2008; 88: 896–906. Tsunawaki S, Sporn M, Ding A et al. Deactivation of macrophages by transforming growth factor-beta. Nature 1988; 334: 260–262. Savill J, Dransfield I, Gregory C et al. A blast from the past: clearance of apoptotic cells regulates immune responses. Nat Rev Immunol 2002; 2: 965–975. Gordon S. Alternative activation of macrophages. Nat Rev Immunol 2003; 3: 23–35. Bouchon A, Facchetti F, Weigand MA et al. TREM-1 amplifies inflammation and is a crucial mediator of septic shock. Nature 2001; 410: 1103–1107. Klesney-Tait J, Turnbull IR, Colonna M. The TREM receptor family and signal integration. Nat Immunol 2006; 7: 1266–1273. McVicar DW, Taylor LS, Gosselin P et al. DAP12-mediated signal transduction in natural killer cells. A dominant role for the Syk proteintyrosine kinase. J Biol Chem 1998; 273: 32934–32942. Lanier LL. DAP10- and DAP12-associated receptors in innate immunity. Immunol Rev 2009; 227: 150–160. Gibot S, Cravoisy A, Levy B et al. Soluble triggering receptor expressed on myeloid cells and the diagnosis of pneumonia. N Engl J Med 2004; 350: 451–458. Lin YT, Tseng KY, Yeh YC et al. TREM-1 promotes survival during Klebsiella pneumoniae liver abscess in mice. Infect immun 2014; 82: 1335–1342. Kuai J, Gregory B, Hill A et al. TREM-1 expression is increased in the synovium of rheumatoid arthritis patients and induces the expression of pro-inflammatory cytokines. Rheumatology (Oxford) 2009; 48: 1352–1358. Schenk M, Bouchon A, Seibold F et al. TREM-1–expressing intestinal macrophages crucially amplify chronic inflammation in experimental colitis and inflammatory bowel diseases. J Clin Invest 2007; 117: 3097–3106. Chevalier RL, Forbes MS, Thornhill BA. Ureteral obstruction as a model of renal interstitial fibrosis and obstructive nephropathy. Kidney Int 2009; 75: 1145–1152. Klesney-Tait J, Keck K, Li X et al. Transepithelial migration of neutrophils into the lung requires TREM-1. J Clin Invest 2013; 123: 138–149. Hamilton JA. Colony-stimulating factors in inflammation and autoimmunity. Nat Rev Immunol 2008; 8: 533–544. Verreck FA, de Boer T, Langenberg DM et al. Human IL-23-producing type 1 macrophages promote but IL-10-producing type 2 macrophages subvert immunity to (myco)bacteria. Proc Natl Acad Sci USA 2004; 101: 4560–4565. Fleetwood AJ, Lawrence T, Hamilton JA et al. 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 2007; 178: 5245–5252.

1185

basic research

28.

29.

30.

31.

32.

Wu MF, Chen ST, Yang AH et al. CLEC5A is critical for dengue virusinduced inflammasome activation in human macrophages. Blood 2013; 121: 95–106. Wu J, Li J, Salcedo R et al. The proinflammatory myeloid cell receptor TREM-1 controls Kupffer cell activation and development of hepatocellular carcinoma. Cancer Res 2012; 72: 3977–3986. Kaifu T, Nakahara J, Inui M et al. Osteopetrosis and thalamic hypomyelinosis with synaptic degeneration in DAP12-deficient mice. J Clin Invest 2003; 111: 323–332. Wang Y, Cao Q, Zheng G et al. By homing to the kidney, activated macrophages potently exacerbate renal injury. Am J Pathol 2008; 172: 1491–1499. Cao Q, Wang Y, Zheng D et al. Failed renoprotection by alternatively activated bone marrow macrophages is due to a proliferation-dependent phenotype switch in vivo. Kidney Int 2014; 85: 794–806.

1186

T-H Lo et al.: TREM-1 modulates M1 polarization in UUO nephritis

33. 34.

35.

36.

37.

Tammaro A, Stroo I, Rampanelli E et al. Role of TREM1-DAP12 in renal inflammation during obstructive nephropathy. PLoS One 2013; 8: e82498. te Witt R, van Wolfswinkel ME, Petit PL et al. Neopterin and procalcitonin are suitable biomarkers for exclusion of severe Plasmodium falciparum disease at the initial clinical assessment of travellers with imported malaria. Malar J 2010; 9: 255. Sharif O, Knapp S. From expression to signaling: roles of TREM-1 and TREM-2 in innate immunity and bacterial infection. Immunobiology 2008; 213: 701–713. Gibot S, Buonsanti C, Massin F et al. Modulation of the triggering receptor expressed on the myeloid cell type 1 pathway in murine septic shock. Infect Immun 2006; 74: 2823–2830. Haselmayer P, Grosse-Hovest L, von Landenberg P et al. TREM-1 ligand expression on platelets enhances neutrophil activation. Blood 2007; 110: 1029–1035.

Kidney International (2014) 86, 1174–1186