Male C57BL/6J mice were bred at the Bichat Medical School’s breeding facility. MC-deficient (kit^Wsh/Wsh) in the C57/BL/6J background, named thereafter W^sh/sh, have been obtained from Juan Rivera (NIH) and are bred at Bichat Medical School’s animal facility. Mice deficient in MCPT4 chymase in the C57/BL/6J maintained in Bichat Medical School’s breeding facility have been described previously (21).

Histological kidney specimens of 21 pediatric patients (12 males and 9 females, age 2–168 months) with severe obstructive uropathy of diverse origin who had undergone kidney resection were assayed for the inflammatory infiltrate and fibrosis in retrieved kidney specimens as part of a routine procedure to evaluate kidney damage.

Newborn mice at day 3 of life were placed on a warming pad under a microsurgical microscope. Under anesthesia by isoflurane and oxygen, a right longitudinal flank incision was made to expose the right ureter before embedding it in the psoas muscle by two points, 11-0 nylon sutures, according to the technique described by Ulm and Miller (33). After closure of the incision (9-0 nylon sutures), the mice were allowed to recover on the warming pad with oxygen before returning them to their mother (29). Due to the surgery procedure, there was an important postoperative mortality [wild-type (WT) 82%; W^sh/sh 81%; and Mcpt4^−/− 70%] most of which was due through maternal cannibalism. Postoperative deaths were excluded from the study.

Twenty-four hours before euthanization, imaging sequences were performed with a 7-T MRI, 12 cm diameter, horizontal bore magnet (Pharmascan, Bruker-Biospin SA, Germany). All MRI sequences were respiratory triggered in order to limit artifact induced by respiratory movement using anesthesia by isoflurane and oxygen (1.5% isoflurane in 1 l/min air/O₂) administered through a nose cone. Mice body temperature was maintained at 36–37°C with a warm water circulator. The mouse was in the prone position inside a quadrature bird cage coil (Bruker). Morphological analysis of kidneys was performed using a coronal T2-weighted 2D RARE sequence (anatomic acquisition) with the following parameters: TR/TE: 1,500/11 ms; field of view: 40 mm × 40 mm; acquisition matrix: 128²; 3 coronal slices; slice thickness: 1 mm; no gap between slices, RARE factor: 8, 2 signal averages, fat suppression. The acquisition was synchronized with respiration using balanced acquisitions over several respiratory periods with an effective TR of about 2.5 s. The acquisition time was around 5 min, depending on breath rate for each mouse. This sequence was optimized to compare the operated kidney with the non-operated kidney. On the coronal plan, the maximum length kidney was measured from the superior to the inferior parenchyma pole. Coronal T2-weighted images were made with a slice thickness of 1 mm without gaps. The kidney parenchyma volume was calculated using ImageJ (W. Rasband, NIH, USA) and the additional Magnetic Resonance Urography plugin, by adding every parenchyma area (contours were manually drawn) multiplied by slice thickness. The number of pixels was determined in each slice through the parenchyma excluding cavities (2D surface). The kidney volume (V_t) was calculated as follows: V_t = (A₁ + A₂ + … + A_n) × PS × ST^h where A_n is the number of pixel of the parenchyma in the slice n, PS is the pixel area, and ST^h is the slice thickness. The studies were conducted following the recommendations of the European Convention for the protection of Vertebrates Animals used for Experimentation.

At day 75, before euthanization, we ensured ureteric permeability on anesthetized mice (100 mg/kg ketamine and 20 mg/kg xylazine, i.p.), by injecting Methylene blue as described (30). Mice with a complete ureteral obstruction were excluded from the study. After vascular perfusion by heart puncture (20 ml PBS), kidneys were removed by blunt dissection and weighed without the upper tract using the same laboratory balance.

The inferior part of kidney (~25%) was cut and used for analysis of αSMA expression. The remainder was cut into two longitudinal parts. One half was fixed in 4% formalin and embedded in paraffin. Tissue sections (4 µm) were stained with hematoxylin-eosin-safran, Masson’s trichrome, and Sirius Red. Images were observed with a Leica microscope (Leica Microsystems, Rueil-Malmaisons, France) coupled to a MD2000 camera (Leica Microsystems) using a 10×, 20×, 40× auxiliary lens and a direct X1C-mount. Objective magnifications are as indicated. Renal fibrosis, interstitial infiltration, and the number of glomeruli in vertical cortex range were scored. All histological quantifications were performed by an experienced pathologist under magnification 10× (fibrosis) or 20× (infiltration) in a blinded fashion according to standard scores: 0 absence of lesions, 1 when <25% of the slide area, 2 when >25 and <50%, and 3 when >50%. Photographs of Sirius Red stained sections to illustrate renal fibrosis and glomerular ranks were made with an image segmentation software after converting the glass slides into digital slides (ImageScope, Aperio, Vista, CA, USA). MC staining with toluidine blue was performed as described (34). Immuno-histochemical staining was performed according to standard procedures using 4 µm formalin fixed sections. T cell infiltrate in mouse kidney sections was evaluated using a rabbit anti-CD3 (Dako). Macrophage infiltrate in mouse kidney sections was checked using rat anti-F4/80 (clone A3-1) and anti-CD11b (clone M1/70) antibody (BD Biosciences). MC and T cell infiltration in samples from UPJ patients was evaluated using mouse anti-chymase, mouse anti-tryptase, and mouse anti-c-kit antibodies (all AbDSerotec) for MC and a mouse anti-CD3 antibody (clone UCHT-1; DAKO A/S) for T cells as well as Fab’2 anti-mouse-HRP secondary antibodies (Jackson Immuno research). For inflammatory infiltration, the (CD3⁺) Banff score was used to identify three groups of severity of kidney disease with low, intermediate, and high inflammatory cell infiltration (Banff scores 1, 2, and 3). The Banff score (35), which classically is used to measure the degree of rejection in kidney allografts by determining among other criteria the mononuclear T cell infiltrate, was adapted here to score the degree of inflammation in UPJ kidneys. A minimum of three high power fields (hpf) were analyzed per patient. Fibrosis scores were determined using Masson’s trichrome stained sections.

Lysates from left and right kidney tissues were prepared in 50 mmol/l Tris–HCl (pH 8.8) containing 1% sodium dodecyl sulfate and 5% glycerol. A total of 20 µg of kidney lysates were migrated on a 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by transfer onto nitrocellulose membrane (Schleicher and Schuell, Dassel, Germany). Membranes were blocked with 4% bovine serum albumin for 1 h followed by incubation for 1 h at room temperature (RT) with the primary antibodies: a purified mouse anti-αSMA-1 (clone 1A4, Thermo Fisher Scientific, France) and anti-αtubulin (Sigma). After several washes, blots were incubated with goat anti-mouse IgG HRP (1/5,000) (Jackson Immunoresearch, Newmarket, UK) for 45 min and were developed by enhanced chemiluminescence; GE, Paris, France. Quantitative analysis of blots was performed by densitometry using NIH ImageJ software. The ratio between αSMA and α tubulin (loading control) was determined and compared to a kidney lysate from a sham-operated WT mouse always run in parallel and arbitrarily set to 1.

CCL2 concentrations in serum collected at the day of euthanization as well as TGFβ and IL6 concentrations in IgE-sensitized MC stimulated with specific antigen (DNP-HSA at 30 ng/ml) were quantified using commercial ELISA according to the manufacturer’s instructions (Duoset cytokine Elisa Kits, R&D System, Lille, France).

Mouse proximal tubule cells were isolated from 3- to 4-week-old C57Bl/6 mice as described (36). Briefly, kidneys were removed aseptically from anesthetized mice. The cortex was separated from the medulla and incubated for 30 min in a 1 mg/ml collagenase solution. Homogeneous populations of nephron segments were separated on a Percoll gradient (Percoll 42%; centrifugation 17,000 rpm for 30 min at 4°C). The F₄ layer, composed almost exclusively of proximal tubules, was seeded on glass slides (76 mm × 26 mm) in culture medium [Ham’s F-12/Dulbecco’s modified Eagle’s medium (1:1, vol/vol), 25 mM HEPES, 21.5 mM HCO₃, 1 mM Na pyruvate, 10 ml/l of a 100 × non-essential amino acid mixture, 4 mM l-glutamine, 50 U/ml penicillin, 50 µg/ml streptomycin, and 50 nM Na selenite] supplemented with 5 µg/ml insulin, 35 µg/ml transferrin, 5 nM triiodothyronine, 0.5 µM retinoic acid, 25 ng/ml prostaglandin E₁, and 50 nM dexamethasone. Fetal calf serum (1%) was present in the medium for the first 48 h. To analyze induction of alpha-SMA, expression cultures were used at day 7. BMMCs were obtained by isolating bone marrow precursors from femurs and tibias of WT mice and grown as described in medium containing recombinant murine IL-3 and stem cell factor at 10 ng/ml (Peprotech, Paris, France) for 5 weeks to obtain differentiated BMMCs (34). To test the capacity of supernatants for induction of αSMA expression in MPTC in co-culture assays, BMMCs (1 × 10⁶) were sensitized with 1 µg/ml IgE anti-DNP (37) for 3 h in MPTC culture medium and were then either left unstimulated or were stimulated with 30 ng DNP-HSA. Supernatants were collected at indicated time points and added to MPTC cultures for 72 h. TGFβ (Peprotech) at 5 ng/ml was added to MPTC cultures as a positive control.

After co-culture with supernatants from BMMC, MPTC cells (about 1 × 10⁵) were added to 8-well Lab-Tek chamber slides and allowed to adhere before fixing them for 20 min on ice in 10 mM PIPES pH6.8, 150 mMNaCl, 5 mM EGTA, 5 mM MgCl₂, 5 mM glucose containing 4% paraformaldehyde (IF buffer). After permeabilization in IF buffer containing 0.025% saponin for 20 min at RT, followed by blocking in IF buffer containing 0.012% saponin and 7% horse serum (Invitrogen) for 30 min at RT. Staining with primary mouse anti-αSMA antibody was performed in IF buffer containing 0.012% saponin and 5% horse serum for 2 h at RT followed by incubation with secondary goat anti-mouse FITC Ab for 60 min at RT. Cells were mounted in Prolong-Gold anti-fading reagent (Molecular Probes) and were analyzed under an immunofluorescence microscope. Quantitative analysis of fluorescence was performed by densitometry using NIH ImageJ software.

Statistical analysis was performed using GraphPad Prism software (release 5.00 windows, GraphPad Software, San Diego, CA, USA). Statistical significance between the three mice strains (WT, W^sh/sh, Mcpt4^−/−) was examined by Kruskal–Wallis test. To compare two delta-values of the three strains mice (WT, W^sh/sh, Mcpt4^−/−), a Mann–Whitney test was used. For each mouse strain, comparison between RT and RL was examined by Wilcoxon test. To compare two RT or two RL coming from two different mice strains, a Mann–Whitney test was used. The Spearman correlation test was used to evaluate correlation between volume and weight kidneys for each mouse strain. p-Values of 0.05 were considered significant.

A large fraction of obstructive nephropathies in children, with 50% of them being diagnosed before birth, represent pyelo-ureteral junction pathology. Kidney sections from surgical specimen from pediatric patients having undergone obstructive nephropathy surgery were analyzed for the inflammatory response by evaluating CD3⁺ T cell infiltration (Banff score) and fibrosis as well as MC infiltration using tryptase and chymase and c-kit as a marker. Three representative patients with different degrees of the inflammatory T cell infiltrate are presented in Figure 1A. A quantitative analysis of scores including 21 patients is shown in Figure 1B. The data indicate that patients, showing, respectively, a low, intermediate, and high inflammatory cell infiltrate (Banff scores 1, 2, and 3) are positively correlated with MC numbers. We also evaluated fibrosis scores using Masson’s Trichrome-stained sections. Data show no statistically significant fibrosis between the different Banff scores. Altogether, our data indicate that MCs infiltrate the kidney in UPJ patients and that their presence is correlated with the degree of inflammation, but not with the degree of fibrosis.

To analyze the role of MC in UPJ pathology, we used the previously developed pUUO model in neonatal mice (29, 30, 33). Following surgery at day 3 of life in WT, MC-, and MCPT4-deficient mice, a total of 21 WT, 15 W^sh/sh, and 22 Mcpt4^−/− mice survived. Out of these 17 WT, 12 W^sh/sh and 20 Mcpt4^−/− mice presented pUUO as judged by ureter permeability test (see Materials and Methods) at the day of euthanization. To judge UPJ pathology progression, MRI sequences taken at day 75 after surgery the day before euthanization from WT, MC-, and MCPT4-deficient mice were analyzed. Representative morphologic T2-weighted images of left and right operated kidneys of the various mouse strains compared to sham-operated mice (WT) are presented in Figure 2A. In agreement with the partial obstruction UPJ pathology, WT mice developed marked differences in size between operated right kidneys (RKs) and non-operated left kidneys (LKs). Volume and length measurements of RK and LK in WT animals (Figures 2B,C) revealed a pronounced hypotrophy in operated RK as compared to kidneys from sham-operated mice. At the same time, a compensatory upregulation of length and volume occurred in the non-operated LK. These alterations were not seen in sham-operated animals as described before (30) (for quantitative data, see weight analysis below). Strikingly, our measurements showed that kidney hypotrophy and the compensatory hypertrophy were less pronounced in MC- and MCPT4-deficient strains. This became particularly evident when differences (Δ) between the LK and RK were plotted revealing significant differences between WT and MC-deficient or MCPT4-deficient strains.

To confirm development of kidney hypo- and hypertrophy, we determined kidney weights after euthanization. Surgery in general did not modify the normal growth of mice, as no significant differences in the average weight between each operated and sham-operated strain were found (WT = 24.93 ± 0.76 g versus 23.24 ± 1.43 g, W^sh/sh = 23.44 ± 0.73 g versus 23.08 ± 1.09 g, Mcpt4^−/− = 22.73 ± 0.61 g versus 21.83 ± 0.49 g). By contrast, measurements of kidney weight confirmed the development of weight loss in operated RK and compensatory upregulation of weight in LKs versus those from sham-operated WT animals, while no differences were seen in sham-operated animals of all strains (Figure 3A). The weight loss and compensatory weight increases were diminished in MC- and MCPT4-deficient mice confirming that UPJ pathology was less severe in the absence of MC and MCPT4. For all strains, kidney weight significantly correlated with the kidney volume calculated on MRI volume sequences (Table 1). We also examined whether MC deficiency could influence normal kidney development after pUUO, based on the fact that, at the time of surgery, kidney nephrogenesis is incomplete, by determining the number of glomerular ranks. We observed a slight compression of the cortex in RK with diminution of the number of glomerular ranks, likely due to an arrest in nephrogenesis after pUUO (Figure 3B). This was particularly observed in WT mice, while no significant differences became apparent in MC and MCPT4-deficient strains supporting also a role of MC in nephrogenesis. Together, our data support that MC play a role in the development UPJ pathology including nephrogenesis.

Ureteropelvic junction pathology is characterized by the development of interstitial fibrosis, which is quite variable in patients and may develop over years being hardly to predict. In agreement, our histologic assessment of fibrosis (Figure 4A) after pUUO shows only discrete signs of fibrosis in WT mice appearing rather focal and localized mainly in the medulla in areas close to the pelvis dilations. Interestingly, when examining MC- and MCPT4-deficient strains, very little fibrosis and cell infiltration was apparent when compared to WT mice (Figures 4A,B). These data are in agreement with a fibrosis-promoting role of MC and MCPT4 chymase. Concerning the inflammatory cell infiltrate, MCs were not detectable in kidney parenchyma, but they could be detected in kidney capsules, where they often revealed a degranulated phenotype (Figure 5A). In kidney parenchyma, we evaluated T cell infiltration by staining sections with an anti-CD3 antibody. Although T cell infiltration was detectable after pUUO, their numbers did not differ between the various mouse strains (Figure 5B). No significant macrophage infiltration was found. Next, we evaluated systemic parameters of the associated inflammatory response by measuring serum CCL2 chemokine levels (Figure 5C) known to be increased in patients with UPJ (38, 39). Our results revealed increased levels of serum CCL2 levels in WT mice after pUUO when compared to sham control mice. The levels were also significantly elevated compared to MC- and MCPT4-deficient mice.

As fibrosis development was rather focalized and moderate after pUUO, we concentrated on epithelial–mesenchymal transition (EMT) as an early step that precedes fibrosis development to further examine the MC contribution in pathology. EMT is characterized by the generation of myofibroblasts producing ECM proteins and α-SMA, which is abundantly expressed as shown by western blot analysis (Figure 6). Our quantitative analysis of operated RKs as compared to a sham control kidney loaded each time in parallel shows that consistent with EMT after pUUO WT mice show a marked increase in α-SMA levels. In comparison, MC-deficient mice show significantly less α-SMA, while MCPT4-deficient animals show intermediate levels. These results supported a role of MC in the early stages of myofibroblast generation, which is partially dependent on MCPT4.

To further analyze the possible implication of MC in EMT, we examined the effect of supernatants collected from primary resting or activated MC on the expression of α-SMA by cultured proximal tubular cells. Immunofluorescence analysis (Figure 7A) shows that similar to the incubation with TGFβ, used as a positive control, supernatants from long-term (6 h) cultured MC enhance the expression of α-SMA. Activation does not further increase the expression. Supernatants from short-term activated MC (30 min) do not show this enhancing effect. This supports that MC constitutively secrete a factor promoting EMT. Further analysis (Figure 7B) shows that cultured BMMC upon stimulation for 3 and 6 h via the IgE receptor can indeed produce cytokines such as TGFβ and IL6 known to be implicated in EMT (40, 41).