Role of the high-affinity leukotriene B4 receptor signaling in fibrosis after unilateral ureteral obstruction in mice

Leukotriene B4 (LTB4) is a lipid mediator that acts as a potent chemoattractant for inflammatory leukocytes. Kidney fibrosis is caused by migrating inflammatory cells and kidney-resident cells. Here, we examined the role of the high-affinity LTB4 receptor BLT1 during development of kidney fibrosis induced by unilateral ureteral obstruction (UUO) in wild-type (WT) mice and BLT1 knockout (BLT1-/-) mice. We found elevated expression of 5-lipoxygenase (5-LOX), which generates LTB4, in the renal tubules of UUO kidneys from WT mice and BLT1-/- mice. Accumulation of immunoreactive type I collagen in WT UUO kidneys increased over time; however, the increase was less prominent in BLT1-/- UUO kidneys. Accumulation of S100A4-positive fibroblasts increased temporally in WT UUO kidneys, but was again less pronounced in-BLT1-/- UUO kidneys. The same was true of mRNA encoding transforming growth factor-β (TGF)-β and fibroblast growth factor (FGF)-2. Finally, accumulation of F4/80-positive macrophages, which secrete TGF-β, increased temporally in WT UUO and BLT1-/- UUO kidneys, but to a lesser extent in the latter. Following LTB4 stimulation in vitro, macrophages showed increased expression of mRNA encoding TGF-β/FGF-2 and Col1a1, whereas L929 fibroblasts showed increased expression of mRNA encoding α smooth muscle actin (SMA). Bone marrow (BM) transplantation studies revealed that the area positive for type I collagen was significantly smaller in BLT1-/—BM→WT than in WT-BM→WT. Thus, LTB4-BLT1 signaling plays a critical role in fibrosis in UUO kidneys by increasing accumulation of macrophages and fibroblasts. Therefore, blocking BLT1 may prevent renal fibrosis.


Introduction
Unilateral ureteral obstruction (UUO) is an experimental animal model of renal fibrosis that mimics the pathogenesis of chronic obstructive nephropathy in humans. The hydrostatic pressure resulting from the obstruction triggers expression of chemokines in tubular epithelial cells [1], followed by increased interstitial capillary permeability [2], infiltration by interstitial inflammatory cells [3], myofibroblast activation, and extracellular matrix deposition [4][5][6]. Progressive fibrosis, loss of renal parenchyma due to capillary rarefaction [4], and tubular cell death via apoptosis and necrosis [7] also occur. Despite these severe changes in the obstructed kidney, the animal remains healthy because the contralateral kidney is fully functional. Indeed, unlike renal ablation models anesthesia of isoflurane mixed with air and its adequacy was monitored from the disappearance of the pedal withdrawal response. A median abdominal incision was made, and the left proximal ureter was ligated at two points using 3-0 silk. The incision was closed with wound clips (AUTOCLIP, 9 mm; ALZET, Cupertino, CA, USA). Sham-operated mice had the ureter exposed but not ligated [35]. All experimental procedures were approved by the Animal Experimentation and Ethics Committee of the Kitasato University School of Medicine (2018-166), and were performed in accordance with the guidelines for animal experiments established by the Kitasato University School of Medicine and conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). The mice were maintained at constant humidity (60 ± 5%) and temperature (22˚C ± 1) on a 12-h light/ dark cycle. All animals were provided with food and water ad libitum. The total number of mice used in this experiment is 166. The number of mice per group is from 4 to 20. At the end point of the experiments, mice were sacrificed under inhalation anesthesia of isoflurane mixed with air. Mice exhibiting symptoms of infection including suppressed appetite, purulent discharge from the wound were removed from the study prior to the study endpoint.

Tissue harvesting
Kidney samples were collected on Days 0, 1, 3, 5, 7, 10, and 14 after UUO. Day 0 kidney samples were collected without the need for surgical procedures. All mice were anesthetized with isoflurane and perfused with PBS via the left ventricle. The left kidney was harvested immediately and cut into transverse sections for RT-PCR, paraffin embedding, freezing, and Sircol collagen assays.

Histological examination
Kidney tissues were fixed overnight at 4˚C in 4% paraformaldehyde and embedded in paraffin. Paraffin-embedded tissues were cut into 4 μm sections and stained with H&E and Sirius red. Kidney cortex thickness was measured by investigators blinded using Image J software.

Sircol collagen assay
The total amount of soluble collagen was measured using a Sircol collagen assay kit (Biocolor, Antrim, UK). Kidney samples were frozen immediately with in liquid nitrogen immediately and stored at -80˚C until use. All measurements were performed in duplicate and results were expressed as μg of collagen/mg of kidney cortex.

Real-time RT-PCR
Total RNA was extracted from decapsulated kidney tissues using TRIzol reagent (Gibco-BRL; Life Technologies, Rockville, MD, USA), and single-stranded cDNA was generated from 1 μg of total RNA via reverse transcription using the ReverTra Ace qPCR RT Kit (TOYOBO CO., LTD., Osaka, Japan), according to the manufacturer's instructions. Real-time PCR was performed using SYBR Premix Ex Taq II (Tli RNaseH Plus; Takara Bio, Inc., Shiga, Japan). The gene-specific sequences are described in Table 1. Expression of target genes was normalized to that of GAPDH.

Serum levels of blood urea nitrogen and creatinine
Blood was drawn from the heart and then centrifuged and stored at −20˚C until use. Levels of serum blood urea nitrogen (BUN) and creatinine (Cre) were measured using a Dri-Chem 7000 Chemistry Analyzer System (Fujifilm, Tokyo, Japan).

Collection of peritoneal macrophages
Thioglycolate-induced peritoneal macrophages were collected from 8-12-week-old WT mice. In brief, 2 ml of 4% thioglycolate medium was injected into the peritoneal cavity. After 3 days, the peritoneal cavity was washed three times with 5 ml of PBS. Cells in the lavage fluid were washed and suspended in RPMI 1640 medium containing 10% FCS and then placed in 12-well culture plates (1 × 10 6 cells/well). After incubation at 37˚C in 5% humidified CO 2 for 16 h, the plates were washed with PBS to remove non-adherent cells. Approximately 60% of cells remained adherent and were used for subsequent experiments.

Cell culture and treatments
At 16 h after plating, cells were washed twice with PBS and incubated for 2 h in serum-free medium. Cells were then stimulated for 12 h with LTB 4 (0.1, 1, or 10 nM) or serum-free medium (control). Total mRNA was isolated from cells using TRIzol reagent, and mRNA expression was measured by real-time RT-PCR. Thioglycolate-induced peritoneal macrophages (after removal of non-adherent cells) were stimulated with LTB 4 , and expression of TGF-β, FGF-2, αSMA, and Col1a1 mRNA was measured. Murine fibroblasts (L929) were purchased from the Cell Bank at RIKEN BioResource Center (Ibaraki, Japan). Cells were suspended in DMEM containing 10% FCS, plated in 6-well culture plates (3 × 10 5 cells/well), and stimulated with LTB 4 and TGF-β, and expression of TGF-β, FGF-2, αSMA, and Col1a1 mRNA was measured.

BM transplantation
BM transplantation experiments were carried out as described previously [36]. In brief, donor BM was obtained by flushing the femoral and tibial cavities of WT mice and BLT1 -/mice with PBS. The flushed BM cells were dispersed and resuspended in PBS at a density of 1 × 10 6 cells/100 μl. Both WT and BLT1 -/mice were lethally irradiated with 9.5 Gy X-rays using an MBR-1505 R Xray irradiator (Hitachi Medico, Tokyo, Japan) equipped with a filter (copper, 0.5 mm; aluminum, 2 mm). The cumulative radiation dose was monitored. BM mononuclear cells from WT and BLT1 -/mice (2 × 10 6 cells/200 μl) were transplanted into irradiated WT and BLT1 -/mice via the tail vein.

Statistical analysis
All results are expressed as the mean ± SEM. Comparisons between two groups were performed using Student's t test. Comparisons between multiple groups were performed using one-way ANOVA, followed by Tukey's post-hoc test. P values <0.05 were considered statistically significant.

Development of fibrosis in WT mouse kidneys after UUO
Following induction of UUO, the thickness of the kidney cortex in WT mice decreased gradually ( Fig 1A) and was significantly smaller than that in sham-operated mice at Days 7 and 14 (  Fig 1B). Sirius red staining demonstrated that areas of collagen deposition around dilated renal tubules increased in a temporal manner (Fig 1C), and the Sircol collagen assay showed that collagen levels on Days 7 and 14 were significantly higher than sham-operated mice (Day 7: 8.85±1.20 vs.3.61±0.55μg/mg of kidney weight, P = 0.018, Day 14: 11.62±1.15 vs.3.47 ±0.16μg/mg, P = 0.0007, Fig 1D).

Expression of BLT1 and 5-LOX increases in UUO kidneys
To study the role of LTB 4 -BLT1 signaling in the UUO kidney, we examined expression of 5-LOX (an enzyme upstream of LTB 4 ) and BLT1. Expression of BLT1 mRNA in WT UUO kidneys on Day 1 was markedly higher than that on Day 0 (P<0.0001), while BLT1 mRNA levels in BLT1 -/-UUO kidneys were negligible throughout the experimental period ( Fig 1E). By contrast, expression of 5-LOX increased in both WT and BLT1 -/mice from Day 3, although levels in BLT1 -/mice were significantly lower than those in WT mice at Days 3 and 5 (Day 3: P = 0.0002, Day5: P = 0.03, Fig 1F). Immunostaining of WT UUO kidneys and BLT1 -/-UUO kidneys for 5-LOX at Day 3 revealed that dilated tubule epithelial cells in WT mice were positive, as were some interstitial cells. By contrast, there were fewer 5-LOX-positive cells in dilated tubules and tubulointerstitial areas of BLT1 -/mice ( Fig 1G).

Tubulointerstitial fibrosis in BLT1 -/mice is reduced after UUO
To examine the role of BLT1 signaling in collagen accumulation, we examined kidney fibrosis in WT and BLT1 -/mice after UUO. Immunostaining of type I collagen increased in WT kidneys after induction of UUO (Fig 2A). Quantitative analysis of the immunoreactive renal interstitial area revealed an increase in the percentage positive area after induction of UUO in both WT and BLT -/mice; however, the area of type I collagen was significantly lower in BLT1 -/mice than in WT mice from Day 3 (  Fig 2B). Furthermore, expression of mRNA encoding Col1a1 in WT UUO and BLT -/-UUO kidneys increased in a temporal manner, but was significantly lower in BLT1 -/mice from Day 1 (Day 1: <0.0001, Day 3: P<0.0001, Day 5: P = 0.008, Day 7: P = 0.0002, WT vs. BLT1, respectively; Fig 2C). Moreover, expression of mRNA encoding Col3a1 in WT UUO kidneys significantly increased as compared to BLT1 -/-UUO kidneys (Fig 2D)

Accumulation of macrophages in UUO kidneys is BLT1-dependent
It is well known that macrophages, neutrophils and lymphocytes are related to fibrosis formation. Interstitial macrophages promote fibrosis in the UUO kidney [38]. Therefore, we asked what type of cells is most accumulated in the interstitial spaces within UUO kidneys.
In contrast, immunofluorescence analysis against Gr1 showed that there was no significant difference in the number of accumulated Gr1 positive cells between WT and BLTI -/mice except on Day 3 (Fig 3D and 3E). In addition, there was no significant change in the expression of mRNA encoding CD3 (Fig 3F), a specific marker of T lymphocytes. These results showed that the accumulation of macrophages in interstitial tissues was involved in fibrosis, which was dependent on LTB 4 -BLT1 signaling.

Expression of CXCL12 in UUO kidneys is BLT1-dependent
Macrophages and fibroblasts are recruited to the kidneys following UUO [39]. Therefore, we examined chemokine levels in UUO kidneys from WT and BLT1 -/mice. Immunofluorescence staining revealed that CXCL12 was expressed primarily in the interstitial spaces of WT UUO kidneys on Day 3 (Fig 4A). Expression of CXCL12 mRNA in WT mice fell transiently on Day 1, before increasing again on Day 3; however, this increase was significantly lower in BLT1 -/mice (Day 3,5,7: P<0.001, WT vs. BLT1 -/-, respectively; Fig 4B). Moreover, expression of CXCR4, a specific ligand for CXCL12, in BLT1 -/-UUO kidneys was significantly lower than that in WT UUO kidneys on Day 7 (P = 0.004, WT vs. BLT1 -/-, Fig 4C). These results suggest that BLT1-induced macrophage infiltration into the UUO kidneys is dependent on the CXCL12/CXCR4 axis. There was no significant change between serum levels of BUN (Day 0: P = 0.811, Day 7: P = 0.241 Fig 4D) and Cre (Day 0: P = 0.285, Day 7: P = 0.929 Fig 4E) in WT and BLT1 -/mice.

LTB 4 increases expression of TGF-β, FGF-2, and collagen mRNA by macrophages in vitro
To evaluate whether expression of pro-fibrotic cytokines and collagen by macrophages is dependent on LTB 4 -BLT1 signaling, we isolated macrophages of WT mice from the peritoneal cavity and incubated them with LTB 4 . Expression of mRNA encoding TGF-β and FGF-2 increased 12 h after LTB 4 treatment (Fig 5A and 5B). Increased expression of collagen gene Col1a1 was also detected 12 h after addition of LTB 4 (Fig 5C), although expression of αSMA did not increase after LTB 4 treatment (Fig 5D). These results suggest that LTB 4 induces expression of collagen, TGF-β, and FGF-2 by macrophages.

LTB 4 increases expression of mRNA encoding αSMA in connective tissuederived mouse L929 fibroblast-like cells in vitro
When L929 murine fibroblasts were incubated with LTB 4 , we observed no increase in expression of mRNA encoding TGF-β or FGF-2 (Fig 6A and 6B). Unlike in macrophages, LTB 4 did not induce increased expression of mRNA encoding collagen 1a1, although αSMA mRNA levels increased slightly at 12 h after addition of 0.1 nM LTB 4 (Fig 6C and 6D). These results suggest that LTB 4 does not trigger secretion of collagen 1, TGF-β, or FGF-2 by L929 cells. When L929 cells were stimulated with TGF-β, we saw no change in expression of mRNA encoding Col1a1 and αSMA (Fig 6E and 6F).

Discussion
Here, we demonstrated that BLT1 signaling plays a major role in development of fibrosis in BLT1 -/-UUO model mice. Accumulation of collagen type I in UUO kidneys was significantly lower in BLT1 -/mice than in WT mice. We also observed BLT1-dependent recruitment of macrophages and fibroblasts (Fig 2 and Fig 3); however, BLT1 signaling induced expression of mRNA encoding pro-fibrotic factors and collagen in a cell type-specific manner (Fig 2). Surprisingly, we found that LTB 4 acted on macrophages directly to upregulate expression of Col1a1. BLT1 signaling also induced expression of mRNA encoding TGF-β and FGF-2 in vivo, both of which are pro-fibrotic factors; however, we did not demonstrate production of TGF-β and FGF-2 by LTB 4 -stimulated fibroblasts in vitro (Fig 5 and Fig 6). BM transplantation experiments revealed that the area of type I collagen deposition in UUO kidneys from BLT1 -/-BM!WT was lower than that from WT-BM!WT (Fig 7). Thus, BM-derived cells induced by LTB 4 -BLT1 signals exacerbate renal fibrosis in the UUO kidney. Together, these results demonstrated a finely tuned mechanism underlying BLT1-dependent fibrosis in this model (Fig 8), and suggested that blocking of BLT1 signaling may prevent fibrosis. The UUO model is good for studying tubulointerstitial fibrosis accompanied by cellular infiltration. LTB 4 is a chemoattractant for leukocytes, particularly neutrophils and macrophages [21][22][23][24]. We found that, after UUO treatment, dilated tubular cells expressed 5-LOX (Fig 1G), which may be important for initiating BLT1-dependent fibrosis. Expression of 5-LOX also induces LTB 4 , a chemoattractant for macrophages. These molecules elicit local extravasation of fibroblasts and macrophages, which infiltrate the tubulointerstitial space of the kidney in a BLT1-dependent manner. PGE 2 and PGI 2 , the most abundant eicosanoids, suppress fibrosis in vivo [18,40]; however, few studies have examined involvement of LTs and other metabolites in fibrosis after UUO. Models of pulmonary fibrosis exhibit a synthetic imbalance favoring pro-fibrotic LTs over anti-fibrotic PGE 2, suggesting a role for eicosanoids in fibrotic lung disease [41]. PGE 2 levels in BAL fluid from patients with idiopathic pulmonary fibrosis (IPF) are lower than those from healthy control subjects [42]. In addition, fibroblasts grown from lung tissue isolated from patients with IPF synthesize less PGE 2 than cells from healthy control subjects due to reduced COX-2 expression [43]. This has important pathologic consequences because decreased levels of COX-2 and PGE 2 in these cells contribute to increased collagen synthesis and cell proliferation in response to TGF-β [43]. By contrast, BAL fluid from patients with IPF contains more LTB 4 than that from control subjects [44]. LTB 4 levels in lung tissue homogenates from patients with IPF are 15-fold higher, and those of LTC 4 are 5-fold higher, than those in control subjects, reflecting constitutive activation of 5-LOX in alveolar macrophages [45]. Increased lung LT levels have been observed in mice after intratracheal administration of bleomycin; this is a commonly used animal model of pulmonary fibrosis [46]. In this model, fibrosis was blunted markedly following disruption of cPLA 2 [47] and 5-LOX [46], suggesting that endogenous LTs play a major role in facilitating fibrosis. Several features of bleomycin-induced injury, including pulmonary recruitment of macrophages and neutrophils, alveolar septal thickening, fibroblast accumulation, and collagen deposition, are significantly less severe in LT receptor-deficient mice than in their WT littermates [48]. These results suggest that LTs are pro-fibrotic in these pathological settings. In contrast to lung fibrosis, there is no definitive evidence that LT is involved in UUO-induced fibrosis in the kidneys.
As mentioned above, we showed that 5-LOX expression increased after induction of UUO; we also observed 5-LOX-expressing cells in dilated renal tubules in UUO kidneys (Fig 1G). A previous study in a mouse UUO model shows that COX-2 is upregulated in renal tubule epithelial cells after ligation of the ureter [18].
We observed BLT1-dependent accumulation of type I collagen, S100A4-positive fibroblasts, and αSMA-positive myofibroblasts in UUO kidneys (Fig 2). Accumulation of S100A4-positive fibroblasts in the interstitial spaces within kidney tissues was evident from the early stages of UUO onset, although it was less marked in BLT1 -/mice. The αSMA mRNA levels ( Fig 2H) suggest that αSMA-positive myofibroblasts accumulate in UUO kidneys; again, this was less marked in BLT1 -/mice from Day 3. These results suggest that LTB 4 /BLT1 signaling is important for induction of renal fibrosis in UUO kidneys.
We also observed reduced accumulation of macrophages in BLT1 -/mice during the early stage of UUO (Fig 3). In contrast, there was no difference in the number of accumulated Gr1-positive cells between WT mice and BLT1 -/mice except on Day 3. Although neutrophils are the first responder cell type to be recruited to the inflammatory sites contributing to tissue injury in a BLT1 dependent manner [49], our data demonstrated that infiltration by macrophages in UUO kidneys appeared to be extensive as compared with neutrophils during the development of fibrosis formation. These results suggested that the accumulated macrophages mainly are involved in fibrosis formation. These accumulated macrophages supply TGF-β to sites of fibrosis. CXCL12 is a chemokine that plays a role in migration of BM-derived stem cells to the peripheral blood and from there to sites of tissue injury. CXCL12 is a potent chemoattractant for fibroblasts/myofibroblasts. CXCL12 binds to a specific receptor, CXCR4. Philips et al. report that the CXCL12/CXCR4 axis induces recruitment of BM-derived stem cells to injured lung tissue to induce pulmonary fibrosis [50]. Also, CXCR4 antagonists ameliorate renal fibrosis in the UUO kidney [51]. It would be worth investigating whether LTB 4 -BLT1 signaling interacts with the CXCL12/CXCR4 axis in the UUO kidney. Here, we found that CXCL12-positive cells localized primarily to the interstitial spaces within UUO kidneys (Fig 4A). Furthermore, expression of CXCL12 and CXCR4 increased in WT kidneys more than in BLT1 -/kidneys. These results suggested that the CXCL12/CXCR4 axis contributes to renal fibrosis in the UUO kidney in a LTB 4 -BLT1 signaling-dependent manner. The results also suggested that accumulation of collagen-producing cells is regulated by BLT1 signaling. Lack of BLT1 signaling may explain, at least in part, the reduced fibrosis observed in UUO kidneys. BM-mobilized macrophages and fibrotic cells express CXCR4 and infiltrated CXCL12-enriched tissue; however, we did not identify the type of cell that contributes to renal fibrosis in UUO kidney. Further experiments are needed to answer this question.
TGF-β has the potential to increase collagen biosynthesis by fibroblasts and myofibroblasts [52]. We confirmed reduced expression of TGF-β in UUO kidneys of BLT1 -/mice, along with increased expression of TGF-β by macrophages stimulated with LTB 4 in vitro (Fig 5A). Interestingly, LTB 4 -stimulated L929 cells did not show increased expression of TGF-β and FGF-2 ( Fig 6A and 6B). Moreover, LTB 4 -stimulated macrophages upregulated expression of collagen 1a mRNA, but not that of αSMA mRNA (Fig 5C and 5D). By contrast, LTB 4 -stimulated L929 cells upregulated expression of αSMA mRNA, but not collagen 1a mRNA (Fig 6C and 6D). These results suggest that LTB 4 -BLT1 signaling induces fibrosis via accumulation of macrophages and fibroblasts. A previous report suggested that LTB 4 did not alter expression of collagen-encoding genes in primary mouse lung fibroblasts. Furthermore, bleomycin-treated macrophages produced LTB 4 and showed increased production of TGF-β in a BLT1-dependent manner [53]. Taken together, these results suggest that LTB 4 -BLT1 signaling might promote TGF-β production by macrophages recruited via BLT1 signaling.
The results of the BM transplant experiments suggest that BM-derived cells induced by LTB 4 -BLT1 signaling exacerbate renal fibrosis in the UUO kidney (Fig 7). Immunohistochemical study showed that accumulation of F4/80-positive macrophages in UUO kidneys in BLT1 -/-BM!WT at Day 7 was significantly lower than that in WT-BM!WT (Fig 7E and 7F). Accumulation of BM-derived macrophages expressing BLT1 was associated with collagen deposition in the UUO kidney. These results indicated that BLT1 signaling in BM-derived macrophages is involved in renal fibrosis. In addition, the current study showed that there was no change in the number of Gr1 positive neutrophils in the UUO kidneys between BLT1 -/-BM!WT and WT-BM!WT on Day 7 (Fig 7G and 7H). In this experiment, we did not estimate accumulation of neutrophils in the early phase. Further study is needed to answer whether Gr1 positive neutrophils related to fibrosis formation in the UUO kidney. Taken together, the results from the present experiments suggest that complex synergistic loops may be active during LTB 4 -induced fibrosis in this UUO model.
In conclusion, we showed here that BLT1 signaling plays a role in development of fibrosis in UUO models. Accumulation of collagen type I in UUO kidneys of BLT1 -/mice was significantly lower than that in UUO kidneys of WT mice. BLT1 signaling induced accumulation of macrophages and fibroblasts, and induced collagen biosynthesis directly via induction of TGF-β. Thus, BLT1-dependent fibrosis in this model is finely regulated, suggesting that BLT1 signaling is a good therapeutic target for preventing fibrosis.