Introduction

Hydronephrosis is a disease of the kidney or urinary tract caused by several mechanisms, including obstruction of urine outflow. This urinary tract obstruction leads to dilated renal pelvis and calyces, and progressive atrophy of the renal parenchyma, with or without impaired renal functions1. Hydronephrosis can occur as a result of congenital anomalies or various pathophysiological conditions, such as kidney stones, enlarged prostate glands, bladder tumors, gynecologic malignancy, and pregnancy2. Generally, two types of hydronephrosis have been described—obstructive and non-obstructive types. The obstructive hydronephrosis refers to structural and functional changes in the kidneys caused by difficulties in urination. If left untreated, the impairment of renal function is often irreversible. Non-obstructive hydronephrosis, on the other hand, generally stems from congenital anomalies of renal pelvic growth that results in dilated renal pelvis1. Non-obstructive hydronephrosis is also found in patients with nephrogenic diabetes insipidus3,4.

The kidneys not only play an important role in urinary excretion, but also contribute significantly to calcium homeostasis and bone turnover through the conversion of 25-hydroxyvitamin D3 [25(OH)D3] to the active 1,25-dihydroxyvitamin D3 [1,25(OH)2D3]5,6. This active hormone is required for transcriptional upregulation of several genes involved in intestinal calcium absorption, e.g., transient receptor potential vanilloid superfamily Ca2+ channel 6 (TRPV6), calbindin-D9k, Na+/Ca2+-exchanger 1 (NCX1), and plasma membrane Ca2+-ATPase (PMCA1b) (for review, please see5,6,7). The renal production of 1,25(OH)2D3 is tightly regulated by parathyroid hormone (PTH), which also enhances renal calcium reabsorption via TRPV5 and phosphate excretion. At the molecular level, PTH stimulates the expression of 1α-hydroxylase (Cyp27b1), the major mitochondrial enzyme that converts 25(OH)D3 to 1,25(OH)2D3, while downregulating sodium-phosphate co-transporters in the proximal tubule6,7,8. Impaired renal function has long been known to be associated with low bone mineral density (BMD) and increased bone loss9,10. Therefore, hydronephrotic patients have a higher risk of developing compromised bone health although the age-dependent changes in the intestinal calcium absorption and bone microstructure in these patients are largely unknown. A recent case report of a 16-year-old hydronephrotic patient presented abnormal mineral metabolism including hypercalcemia, hypophosphatemia, elevated levels of PTH and hypercalciuria, as well as prolonged healing of a traumatic fracture11. However, a small Japanese cohort study of congenital hydronephrosis revealed that patients with normal renal function exhibited intact BMD12. Thus, the relationship between hydronephrosis and dysregulated calcium and bone metabolism is still elusive.

An animal model of hydronephrosis is essential for understanding of pathology and future development of effective treatments for calcium dysregulation and osteopathy. Fujita and co-workers reported a genetically modified claudin-4 knockout mouse model with hydronephrotic condition13. These mice manifested diffuse hyperdysplasia with thickening pelvic structure and urothelium indicating of progressive hydronephrosis. Furthermore, deficiency in the miR-143/145 cluster, which was highly expressed in smooth muscle cells of the renal vasculature and also present in the pelvicalyceal system and ureter, led to abnormal ureteral peristalsis and increased rate of ureter contractions, thereby resulting in hydronephrosis14. Moreover, deletion of the Shh receptor Patched 1 (Ptch1) led to intrinsic ureteropelvic junction obstruction, a common cause of congenital hydronephrosis in mice15.

ICR/Mlac-hydro mice used in the present study were established by selective inbreeding of ICR mice that carried spontaneous hydronephrosis phenotype for many generations16. These mice developed severe bilateral non-obstructive hydronephrosis without evidence of interstitial fibrosis or glomerulosclerosis17,18. Our previous pilot study of 8-week-old male ICR/Mlac-hydro mice revealed deterioration of tibial bone microstructure as well as abnormally low bone turnover19suggesting an association between pathological damage of the kidney and bone loss in these young mice. Nevertheless, whether ICR/Mlac-hydro exhibited endocrinopathy or defective intestinal calcium absorption has not been elucidated yet.

Although the disease was apparently severe and had consequences on bone health, little is known regarding progressive changes in the profile of calcium-regulating hormones, intestinal calcium transport or three-dimensional bone microstructure. Without this knowledge, it would not be possible to make use of this animal model to study and develop appropriate treatments or preventive approaches for hydronephrotic patients. Therefore, the present study aimed to demonstrate in ICR/Mlac-hydro mice (i) changes in the intestinal calcium absorption, bone microstructural parameters, bone mechanical properties, and alteration in calcium-regulating hormone levels, and (ii) whether the consequence of these phenotypes worsened with age (1, 2, 3 and 6 months of age). We expected to obtain clear evidence that the ICR/Mlac-hydro mice would exhibit impaired intestinal calcium absorption and deterioration of bone microstructure.

Materials and methods

Animals

Male and female wild-type and ICR/Mlac-hydro mice [Imprinting Control Region (ICR) mouse strain] were provided by the National Laboratory Animal Center (NLAC), Mahidol University, Nakhon Pathom, Thailand. Their biological data, including phenotype, blood profile and renal pathology, have been reported previously16,17. Mice were housed in shoebox with bedding of autoclaved corn cob, and fed pasteurized standard chow containing 1.0% wt/wt calcium, 0.9% wt/wt phosphorus, and 4,000 IU/kg vitamin D (CP, Bangkok, Thailand) and reversed osmosis water ad libitum under 12 h/12 h light/dark cycle. Room temperature was controlled at 22 ± 2 °C, and relative humidity was 50–60%. Their phenotype was verified before use, as described previously16,17. Since the reproductive system maturation could affect bone turnover, we used animals of similar age in each group for our experiments and inspected the appearance of their reproductive organs prior to the experiments. The experimental protocols have been approved by the Institutional Animal Care and Use Committee (IACUC), NLAC, Mahidol University and Faculty of Science, Mahidol University. All studies related to animals were performed in accordance with relevant guidelines and regulations, including the ARRIVE guideline (https://arriveguidelines.org).

Experimental design

To investigate whether the aberrant calcium and bone metabolism in hydronephrotic mice was progressive with age, wild-type and ICR/Mlac-hydro male mice aged 1, 2, 3 and 6 months were used in the present study. Female 3-month-old mice (wild-type and ICR/Mlac-hydro) were used to investigate whether dysregulation of calcium and bone metabolism was different between genders. Mice were euthanized by intraperitoneal injection of 150 mg/kg sodium pentobarbital (Abbott Laboratories, North, Chicago, IL, USA) followed by cardiac removal. After being euthanized, blood and bones were collected and frozen at − 80 °C for serum samples and − 20 °C for bone specimens until analyses. Bone length was measured with a vernier caliper. To keep structure of swollen kidney intact, ureter was tied with silk and cut below the knot before kidney collection. Thereafter, swollen kidneys were fixed at 4 °C in 4% paraformaldehyde for 12 h and later embedded in paraffine for hematoxylin & eosin staining. The histopathology of renal sections was independently examined by two experts (PS and KW). All ICR/Mlac-hydro mice were found to exhibit overt hydronephrosis (+), which was not observed in wild-type mice (–). Femora were used for analyses of mechanical property and 3-dimensional microstructure by 3-point bending apparatus and micro-computed tomography (µCT), respectively. In certain experiments, nano-computed tomography (nanoCT) was used to examine the sites of trabecular separation. Tibiae were analyzed for static and dynamic bone parameters by bone histomorphometry. As for the dynamic bone histomorphometric study, mice were administered subcutaneously with fluorescent dye (10 mg/kg body weight calcein) in 2 doses, i.e., on day 6 and day 1 prior to euthanasia. In another experiment, different set of wild-type and ICR/Mlac-hydro male mice aged 3 and 6 months were used for studying intestinal calcium transport and epithelial electrical parameters. For this experiment, after median laparotomy was performed in anesthetized animal, duodenal segment was removed for a study of calcium transport by ex vivo Ussing chamber technique.

Analyses of blood chemistry, calcium-regulating hormones and bone turnover markers

The concentrations of blood urea nitrogen (BUN) and blood creatinine were determined by enzymatic colorimetric/fluorometric method, while serum albumin was determined by bromocresol purple assay kit. Total calcium and inorganic phosphate were analyzed by o-cresolphthalein complexone and phosphomolybdate-based kit using a Dimension RxL analyzer (Dade Behring, Marburg, Germany), respectively. Serum intact PTH (iPTH), intact fibroblast growth factor-23 (iFGF-23), c-terminal FGF-23, 1,25(OH)2D3, N-terminal propeptide of type 1 procollagen (P1NP) and C-terminal telopeptides of type 1 collagen (CTX) were analyzed by commercial enzyme-linked immunosorbent assay (ELISA) or enzyme immunoassay (EIA) kits. Catalog numbers of iPTH, intact FGF-23, c-terminal FGF-23 kits were 60–2305, 60–6800 and 60–6300, respectively (Immutopics, San Clemente, CA, USA). The catalog number of 1,25(OH)2D3 kit was MBS731103 (MyBioSource, San Diego, CA), and those of P1NP and CTX kits were AC-33F1 and AC-06F1, respectively (Immunodiagnostic Systems, Boldon, UK).

Measurement of intestinal calcium by 45Ca radioactive tracer

Ussing chamber technique was employed to determine the transepithelial calcium flux ex vivo, as described previously20. In brief, the duodenum was first mounted and equilibrated for 10 min between apical and basolateral hemichambers, which were filled with an isotonic bathing solution composed of (in mM) 118 NaCl, 4.7 KCl, 1.1 MgSO4, 1.25 CaCl2, 23 NaHCO3, 12 D-glucose, 2.5 L-glutamine and 2 mannitol (all purchased from Sigma). The solution was continuously gassed with humidified 5% CO2 in 95% O2 and maintained at 37 °C with a pH of 7.4. The osmolality was measured by a freezing point-based osmometer (model 3320; Advanced Instruments, Norwood, MA, USA) and found to be 290–293 mmol/kg water. Thereafter, the bathing solution in the apical hemichamber (mucosal compartment) was replaced with fresh bathing solution containing 45Ca, with initial amount of 0.451 Ci/mL and final specific activity of 90 mCi/mol (catalog no. NEZ013; PerkinElmer, Boston, MA, USA). The basolateral side was replenished with fresh normal bathing solution. The 45Ca radioactivity, expressed in counts per minute, was measured by a liquid scintillation spectrophotometer (model Tri-Carb 3100; Packard, Meriden, CT, USA). Radiotracer samples were collected from Ussing chamber, and the unidirectional calcium flux in the apical-to-basolateral direction was calculated as previously described21.

Measurement of duodenal electrical parameters

The epithelial electrical parameters, including transepithelial potential difference (PD or voltage), short-circuit current (Isc) and transepithelial resistance (TER), were measured as previously described21. In brief, PD and Isc were recorded using two pairs of electrodes made of Ag/AgCl half cells connecting with Ussing chamber through salt bridges (2 M KCl in 3 g% agar). The PD-sensing electrodes were positioned near the Caco-2 monolayer and connected to a preamplifier (model EVC-4000; World Precision Instruments, Sarasota, FL, USA) and a PowerLab digital recording system (model 4/30; ADInstruments, Colorado Springs, CO, USA). The Isc-passing electrode was placed at the rear end of each hemichamber, connected in series to the EVC-4000 current-generating unit and PowerLab 4/30 operated with Chart version 5.2.2. Fluid resistance was consistently subtracted using the EVC-4000 system. TER was then calculated using Ohm’s equation.

mRNA expression of calcium transport-related genes by quantitative real-time PCR

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) from duodenal mucosal scrapings and kidney. RNA purity was assessed with a NanoDrop 2000c spectrophotometer (Thermo Scientific, Waltham, MA, USA) by measuring the 260/280 nm ratio, which consistently ranged from 1.8 to 2.0. One microgram of total RNA was reverse transcribed into cDNA using an oligo(dT20) primer and an iScript kit (Bio-Rad, Hercules, CA, USA). The primer sequences and full names of the studied transcripts are listed in Table 1. GAPDH, a housekeeping gene, was used as a control to ensure consistent reverse transcription (coefficient of variation of GAPDH mRNA expression < 5%). These primers had been validated for specificity and efficiency using conventional PCR (Bio-Rad). Quantitative real-time PCR (qRT-PCR) and melting curve analysis were conducted using a QuantStudio 3 real-time PCR system (Applied Biosystems, Waltham, MA, USA) with SsoFast EvaGreen Supermix (Bio-Rad). Relative gene expression was calculated based on 2–∆∆CT method.

Table 1 Mouse (Mus musculus) primer sequences.
Full size table

Determination of TRPV6 protein expression by immunohistochemistry

Duodenal segment was cleared of luminal content by flushing with chilled 0.1 M phosphate-buffered saline (PBS) and fixed overnight in PBS containing 4% paraformaldehyde. Tissues were dehydrated in graded alcohol and embedded in paraffin, and then the specimens were cut longitudinally into 4-µm sections, which were subsequently exposed to antigen retrieval solution (0.01 mg/mL proteinase K, 50 mM Tris-HCl pH 8.0 and 5 mM EDTA). Tissues were incubated for 1 h with 3% H2O2 to inhibit endogenous peroxidase activity. Non-specific bindings were blocked for 2 h with 10% normal goat serum and 0.1% Tween-20 in PBS. Thereafter, sections were incubated at 4 °C overnight in moist chamber with anti-TRPV6 antibody (RRID: AB_2039791; Cat#ACC-036; Alomone Labs, Jerusalem, Israel). After washing with 0.1% Tween-20 in PBS, sections were incubated for 1 h at room temperature with anti-rabbit conjugated with biotin (Cat# 31820; Invitrogen), washed, and then incubated for 1 h with streptavidin-conjugated horseradish peroxidase (HRP) solution (Cat# 434323, Invitrogen). The red-brown color was developed with 3,3´-diaminobenzidine (DAB) substrate chromogen system (Cat# K3468, Dako; Glostrup, Denmark).

Bone microstructure assessment by microcomputed tomography (µCT) and NanoCT

Skyscan 1178 µCT (Bruker, Kontich, Belgium) with isotopic voxel size of 85 µm3 enabled vBMD measurement. Calibration was performed with two conventional hydroxyapatite (HA) cylinder phantoms (i.e., 250 and 750 mg HA/cm3) scanned at peak x-ray tube voltage of 50 kV and current of 615 µA. A femoral specimen was wrapped with saline-soaked gauze and fixed with Styrofoam to secure specimens in place during scanning. The regions of interest were distal femoral spongiosa and cortical midshaft, i.e., 0.765–1.7 mm and 5.950–6.885 mm from the distal growth plate, respectively. After scanning, three-dimensional region was reconstructed by NRecon software (version 1.6.4.8) with ring artifact correction of 10 and a beam hardening correction of 30%. The serial 8-bit binary images were analyzed by CTAn software (version 1.14.4). The principal parameters obtained from Skyscan 1178 were trabecular vBMD at distal femoral metaphysis, cortical vBMD, cortical thickness, and femoral midshaft. Thereafter, bone specimens were subjected to ultra-high-resolution µCT scanning by using an E-Class VECTor6CT system (MILabs, Netherlands) at 10-µm3 voxel size, peak voltage of 55 kV and current of 0.17 mA, according to the procedure previously reported by Suntornsaratoon et al.22.

Femora were also examined using ultra-high-resolution nanoCT (Skyscan 2214 nanoCT system; Bruker, Kontich, Belgium), which is capable of operating at a resolution as low as 500 nm. Femora were scanned at 80 kV and 130 µA using a 1-mm aluminum filter, with a resolution of 10 × 10 µm2. Image reconstruction was conducted using CTRecon, and 3D visualization was performed with CTVox (Bruker).

Determination of bone microstructure by bone histomorphometry

Bone specimens were cleaned off adhering muscles and connective tissues. Histological changes were determined in ex vivo tibiae obtained from the wild-type and ICR/Mlac-hydro mice by using computer-assisted bone histomorphometry with Goldner’s trichrome staining. Briefly, tibiae were dehydrated in 70%, 95% and 100% vol/vol ethanol for 3, 3 and 2 days, respectively23. Dehydrated bone specimens were then embedded in methyl methacrylate plastic resin at 42 °C for 48 h. The resin-embedded tibiae were first adjusted to obtain the same orientation, and longitudinal sections of 7-µm thickness were cut with the use of rotary microtome equipped with a tungsten carbide blade (model RM2255; Leica, Nussloch, Germany).

Thereafter, each histological section was mounted on a standard glass slide, deplastinated, dehydrated and processed for Goldner’s trichrome staining24. Image captures and analyses were performed under a light microscope (model BX51TRF; Olympus, Tokyo, Japan) using an objective lens 20× and the computer-assisted Osteomeasure system operated with the software version 4.1 (Osteometric Inc., Atlanta, GA, USA). The region of interest (ROI) covered the whole trabecular region of the proximal tibial metaphysis at 0.5–2 mm distal to the growth plate (i.e., secondary spongiosa). The histomorphometric parameters obtained from the Goldner’s trichrome-stained sections consisted of trabecular bone volume normalized by tissue volume (also known as bone volume fraction; BV/TV, %), trabecular thickness (Tb.Th, µm), trabecular separation (Tb.Sp, µm), trabecular number (Tb.N, mm– 1), osteoblast surface normalized by bone surface (Ob.S/BS, %), osteoclast surface (Oc.S/BS, %), active erosion surface (aES/BS, %). Dynamic histomorphometric parameters obtained from calcein-staining sections were double-labeled surface (dLS/BS, %), mineralizing surface (MS/BS, %), mineral apposition rate (MAR, µm/day), and bone formation rate (BFR/BS, µm3/µm2/day).

Determination of bone mechanical properties by 3-point bending test

Three-point bending was used to measure the flexional stiffness and strength of long bone structure, all of which determined an ability of bone to resist fracture. After being collected and cleaned off adhering tissue, femora were wrapped in saline-soaked gauze and frozen at − 20 °C. Two hours before mechanical testing, femora were thawed at room temperature. Three-point bending equipment consisted of three rollers. Each bone was secured on the two lower supports of the anvil (Instron, Canton, MA, USA). Bone was clamped with one of the two rollers to prevent rotation of the bone along its axis. The upper roller contacted at the mid-diaphysis with the load direction perpendicular to the medio-lateral diameter. Load-displacement curves were constructed by Instron 5900 software in order to obtain bone mechanical properties, i.e., stiffness (rigidity of bone; N/mm), maximal load (maximum force that bone withstand before fracture; N), and yield load (a force at the elastic limit or at the beginning of plastic deformation; N).

Statistical analysis

Results were expressed as means ± SE. Unless otherwise specified, the data were analyzed by unpaired Student’s t-test using GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA). Comparisons between male and female mice were performed by two-way analysis of variance (ANOVA). The statistical significance was considered when P-values were less than 0.05.

Results

As shown in Fig. 1A, a representative histological section of the kidney obtained from an ICR/Mlac-hydro mouse exhibited an enlarged pelvicalyceal space and proximal part of the ureter—i.e., hydroureteronephrosis—with severe shrinkage of renal parenchyma. Hydronephrotic phenomenon was observed in all three age groups, i.e., 2-, 3- and 6-month-old male mice, as well as 3-month-old female mice. Denser arrangement of parenchymal cells (such as tubular cells) and mild obliteration of Bowman’s capsule space were also observed (Fig. 1A). Although blood urea nitrogen (BUN)/creatinine ratio of ICR/Mlac-hydro mice was not significantly different from that of the wild-type mice at 2 months of age, an increase in BUN became apparent in 3- and 6-month-old male ICR/Mlac-hydro mice (P = 0.0210, Fig. 1B). Serum and urinary creatinine levels were statistically lower in 3- and 6-month-old ICR/Mlac-hydro mice, respectively (Supplementary Fig. S1). Moreover, serum albumin levels were increased in both male and female 3-month-old ICR/Mlac-hydro mice, as well as in 6-month-old ICR/Mlac-hydro male mice (Fig. 1C). Interestingly, serum total calcium levels were unaltered (Fig. 1D), whereas serum phosphate levels were elevated in 2- and 3-month-old, but not 6-month-old ICR/Mlac-hydro male mice (Fig. 1E).

Fig. 1
figure 1

(A) Representative photomicrographs of cross-sectional renal sections stained with hematoxylin and eosin. Kidneys are collected from 3-month-old male Mlac-hydro mice and wild-type mice (WT). Glomerulus (g) and renal tubules (arrow heads) are identified. Scale bars of low magnification and high magnification images are 1 mm and 30 μm, respectively. Serum levels of (B) blood urea nitrogen (BUN)/creatinine, (C) albumin, (D) total calcium, (E) inorganic phosphate of three age groups, i.e., 2-, 3- and 6-month-old male ICR/Mlac-hydro mice (Mlac-hydro) and wild-type mice (WT). Serum albumin obtained from 3-month-old female mice are shown in panel C. n = 4–7; *P < 0.05, **P < 0.01 vs. age-matched WT control group.

Full size image

Serum intact PTH (iPTH) levels of ICR/Mlac-hydro mice were significantly reduced in all age groups, as compared to wild-type mice. In male mice, this change was detectable at the age of 2 months [152.1 ± 11.92 (n = 7) vs. 104.5 ± 19.57 (n = 7); P = 0.030)], and the iPTH levels decreased to approximately one half of the wild-type mice at the age of 3 and 6 months (Fig. 2A). Similar finding was observed in 3-month-old female mice but with less reduction (two-way ANOVA). Significant reduction in 1,25(OH)2D3 levels was seen only in 3-month-old male ICR/Mlac-hydro mice (P = 0.0234), while age-matched female mice exhibited strong tendency to decrease in the 1,25(OH)2D3 level (P = 0.0510; Fig. 2B). Intact and C-terminal FGF23, a bone-derived hypophosphatemic hormone, was examined in 3-month-old female mice. Results showed ICR/Mlac-hydro mice manifesting ~ 50% reduction in both intact and C-terminal FGF23 levels, as compared to wild-type mice (Fig. 2C).

Fig. 2
figure 2

Serum levels of (A) intact PTH (iPTH), (B) 1,25(OH)2D3 of three age groups, i.e., 2-, 3- and 6-month-old male ICR/Mlac-hydro mice (Mlac-hydro) and wild-type mice (WT). Serum levels of iPTH and 1,25(OH)2D3 of 3-month-old female mice are also shown in the right panel (n = 7). (C) Serum levels of intact FGF23 (iFGF23) and C-terminal FGF23 (C-ter FGF23) of 3-month-old female Mlac-hydro mice and WT mice (n = 4). *P < 0.05, **P < 0.01, ***P < 0.001 vs. age-matched WT control group.

Full size image

qRT-PCR analysis in 3-month-old male ICR/Mlac-hydro and wild-type mice showed a downregulation of key calcium transporter genes in the duodenum and kidney, i.e., TRPV6 and TRPV5, respectively. However, the renal expression levels of PMCA1b and NCX1 remained unaltered (Figs. 3A, C). Immunohistochemical analysis of duodenal sections demonstrated positive red-brown signals of TRPV6 proteins predominantly localized at the apical membrane, with some cytoplasmic staining in villous absorptive epithelial cells of wild-type mice, whereas the signals in ICR/Mlac-hydro mice were visibly diminished (Fig. 3B).

Fig. 3
figure 3

(A) mRNA expression of TRPV6 in duodenal tissue from 3-month-old male ICR/Mlac-hydro mice (Hydro) and wild-type (WT) mice (n = 5). (B) Representative immunohistochemistry photomicrographs of TRPV6 protein in WT and Mlac-hydro mice. Positive red-brown signals of TRPV6 protein are predominantly localized at the apical membrane and within the cytoplasm of villous absorptive epithelial cells. Scale bars, 20 μm. (C) mRNA expression of TRPV5, PMCA1b, and NCX1 in kidney tissue from 3-month-old male Hydro and WT mice (n = 5). *P < 0.05, **P < 0.01 vs. age-matched WT control group. TRPV5, transient receptor potential cation channel subfamily V member 5; TRPV6, transient receptor potential cation channel subfamily V member 6; PMCA1b, plasma membrane Ca2+-ATPase 1b; NCX1, Na+/Ca2+ exchanger 1.

Full size image

Furthermore, the duodenal calcium transport was determined in 3- and 6-month-old male ICR/Mlac-hydro mice by using 45Ca radioactive tracer. As shown in Fig. 4A, the transepithelial calcium flux was significantly lower in ICR/Mlac-hydro mice compared to that of wild-type mice. The epithelial electrical parameters (i.e., PD, Isc, and TER) of 3-month-old ICR/Mlac-hydro mice and age-matched wild-type mice were comparable (Fig. 4B–D). However, at 6 months of age, PD and Isc were significantly lower, while TER was higher in ICR/Mlac-hydro mice when compared with wild-type mice (Fig. 4B–D), suggesting that electrogenic ion transport and epithelial integrity were altered in older ICR/Mlac-hydro mice.

Fig. 4
figure 4

Transepithelial calcium flux (A) and epithelial electrical parameters [potential difference (PD; B), short-circuit current (Isc; C), and transepithelial electrical resistance (TER; D)] in the duodenum obtained from male ICR/Mlac-hydro mice (Mlac-hydro) and wild-type mice (WT). n = 5; *P < 0.05, ***P < 0.001 vs. WT control group.

Full size image

Bone specimens were further examined at 1, 2, 3, and 6 months of age. Femoral length and trabecular vBMD of male ICR/Mlac-hydro mice were significantly lower than wild-type mice from 2 to 6 months of age (Fig. 5A–B). While trabecular vBMD of wild-type mice showed clear and continuous increase, that of ICR/Mlac-hydro mice was unchanged from 1 to 2 months of age. After which, it increased at 3 and 6 months old but remained significantly lower than those of age-matched wild-type mice. Cortical vBMD of ICR/Mlac-hydro mice were lower than the wild-type mice at 2, 3 and 6 months of age (Fig. 5C), but no significant differences in Ct.Th was observed between the ICR/Mlac-hydro and wild-type mice in any age group (Fig. 6). Ultra-high resolution µCT analysis of femora obtained from 3-month-old male and female mice revealed that both male and female ICR/Mlac-hydro mice had significant lower BV/TV, Tb.Th, and Conn.D than in wild-type mice. However, an increase in Tb.Sp was observed only in female ICR/Mlac-hydro mice (Fig. 6), and the sites of trabecular separation were observed in nanoCT images, with the separation in females being visibly more pronounced (Fig. 6B). Additionally, Fig. 6G showed no difference in the degree of anisotropy between ICR/Mlac-hydro mice and wild-type mice. Maximum load, yield load and stiffness of male ICR/Mlac-hydro mice were significantly lower than those of wild-type mice at the age of 3 and 6 months (Fig. 7). The same results were observed in 3-month-old female mice.

Fig. 5
figure 5

(A) Femoral length, (B) trabecular and (C) cortical volumetric bone mineral density (vBMD) in 1-, 2-, 3- and 6-month-old male ICR/Mlac-hydro mice (Mlac-hydro) and wild-type mice (WT) analyzed by ex vivo µCT. Femoral length, trabecular and cortical vBMD of 3-month-old female and male mice are shown on the right panels. n = 7; *P < 0.05, **P < 0.01, ***P < 0.001 vs. age-matched WT control group.

Full size image
Fig. 6
figure 6

(A) Representative µCT images (scale bars, 1 mm) and (B) ultra-high-resolution nanoCT images of the femoral metaphyses. The sites of trabecular separation are indicated by arrows. Microstructural analysis of distal femoral metaphysis in 3-month-old male and female ICR/Mlac-hydro and wild-type mice (WT) as determined by µCT, i.e., (C) trabecular bone volume normalized by tissue volume (BV/TV), (D) trabecular thickness (Tb.Th), (E) trabecular separation (Tb.Sp), (F) connectivity density (Conn.D), (G) degree of anisotropy. (H) Cortical thickness (Ct.Th) of femoral midshaft in 3-month-old male and female Mlac-hydro and WT determined by µCT. n = 7; **P < 0.01, ***P < 0.001 vs. age-matched WT control group.

Full size image
Fig. 7
figure 7

Femoral mechanical properties, i.e., (A) maximum load, (B) yield load, (C) stiffness in 1-, 2-, 3- and 6-month-old male ICR/Mlac-hydro mice (Mlac-hydro) and wild-type mice (WT) analyzed by 3-point bending test. Maximum load, yield load, and stiffness of 3-month-old female and male mice are shown on the right panels. n = 7; *P < 0.05, **P < 0.01, ***P < 0.001 vs. age-matched WT control group.

Full size image

Bone histomorphometric analyses of specimens obtained from 6-month-old mice showed consistent results with a lower trabecular vBMD in ICR/Mlac-hydro mice. Specifically, as compared to wild-type mice, ICR/Mlac-hydro mice showed lower BV/TV, lower Tb.N and higher Tb.Sp, with dramatic decreases in Ob, S/BS, MS/BS, dLS/BS, MAR and BFR (Fig. 8). In addition, a bone resorption parameter (i.e., aES/BS) of ICR/Mlac-hydro mice was much lower than those of wild-type mice (Fig. 8). Furthermore, the circulating levels of serum P1NP and CTX-1 in ICR/Mlac-hydro mice were also significantly decreased (Fig. 9). These profiles of serum bone markers were consistent with the reduced activities of both osteoblasts and osteoclasts shown in bone histomorphometry. Collectively, the aforementioned findings clearly demonstrated bone microstructural defects and low bone turnover in ICR/Mlac-hydro mice, strongly consistent with adynamic bone disease.

Fig. 8
figure 8

(A) Representative photomicrographs of proximal tibial metaphysis of 6-month-old male ICR/Mlac-hydro mice (Hydro) and wild-type mice (WT) stained with Goldner’s trichrome. Scale bars, 1 mm. Epiphyseal plate (Ep), bone trabeculae (arrows), and bone marrow (Ma) are identified. Mineralized bone matrix, erythrocytes, and cytoplasm are stained green, orange, and red, respectively. (B–H) Microstructural analysis of proximal tibial metaphysis in 6-month-old male Hydro and WT mice determined by static bone histomorphometry. (B) trabecular bone volume normalized by tissue volume (BV/TV), (C) trabecular thickness (Tb.Th), (D) trabecular separation (Tb.Sp), (E) trabecular number (Tb.N), (F) osteoblast surface (Ob.S) normalized by bone surface (BS), (G) osteoclast surface (Oc.S) normalized by BS, (H) active erosion surface (aES) normalized by BS. (I–L) Bone microstructural analysis by dynamic bone histomorphometry. (I) double-labeled surface (dLS) normalized by BS, (J) mineralizing surface (MS) normalized by BS, (K) mineral apposition rate (MAR), (L) bone formation rate (BFR) normalized by BS. n = 7; *P < 0.05, **P < 0.01 vs. WT control group.

Full size image
Fig. 9
figure 9

Serum levels of (A) procollagen type I N-terminal propeptide (PINP)—biomarker for bone formation, and (B) C-terminal telopeptide-1 (CTX-1)—biomarker for bone resorption, from 6-month-old male ICR/Mlac-hydro mice (Hydro) and wild-type mice (WT). n = 7; *P < 0.05, **P < 0.01, vs. WT control group.

Full size image

Discussion

Since the kidneys are indispensable for phosphate excretion and calcium homeostasis under the regulation of 1,25(OH)2D3, kidney disease like hydronephrosis was expected to profoundly disturb calcium and bone metabolism. Besides being a kidney disease model, hydronephrosis could also be a useful model for studying endocrine regulation of bone turnover. Previously, we reported that young 8-week-old ICR/Mlac-hydro mice exhibited massive bone loss in both cortical and trabecular compartments with inappropriately low osteoclast-mediated bone resorption19. Herein, we postulated that a disruption of the PTH-vitamin D axis and dysregulation of their target organs (e.g., impaired intestinal calcium absorption) would be associated with hydronephrotic osteopathy. Early-onset hypoparathyroidism observed in both male and female ICR/Mlac-hydro mice, therefore, explained most characteristic manifestations, e.g., decreased serum levels of 1,25(OH)2D3 and FGF23, an impaired intestinal calcium transport and bone microstructural defect.

Certain genetic diseases in human manifest both hydronephrosis and hypoparathyroidism. For example, an autosomal dominant Barakat syndrome with mutations of GATA3 interferes with development of both kidneys and parathyroid gland25,26. ICR/Mlac-hydro mice has been shown to have a polymorphism in Gpi1 locus16, which possibly affected the function of glucose phosphate isomerase 1. Gpi1 normally functions in the cytoplasm as an enzyme for interconversion of glucose-6-phosphate and fructose-6-phosphate, while acting extracellularly as a neurotrophic factor for survival of motor and sensory neurons16. Future investigation is required to elaborate whether Gpi1 is essential for parathyroid gland function and whether GATA3 transcription factor is also impaired in ICR/Mlac-hydro mice.

Pathological examination of all kidney specimens obtained from ICR/Mlac-hydro mice revealed bilateral hydronephrosis (Fig. 1A), consistent with the previous reports that showed bilateral enlargement of renal pelvis and surrounding calyceal system16,17. Hence, degeneration of the renal parenchyma with conspicuous fluid collection in the pelvicalyceal lumen and proximal part of ureter could explain why hydronephrotic mice exhibited an increase in the BUN/creatinine ratio—a sign of declining renal function, which gradually worsened in older mice. It is noteworthy that ICR/Mlac-hydro mice had slightly lower serum creatinine levels (Supplementary Fig. S1), but these remained within the normal range, suggesting that their renal function was still partially retained. The slight decrease in serum creatinine might be due to sarcopenia, which is often observed in osteopenic mice. Renal damage had likely already occurred at a very young age (< 2 months old), but compensatory mechanisms or renal functional reserve helped maintain renal function until the mice were 3 months old or older27,28. In some other animal models with renal impairment, e.g., nephrectomized rats, an increase in serum BUN levels and/or hypertension was not present until renal function was lost by about 80%29,30. Other possible causes of an elevated BUN/creatinine ratio in advanced age might include acute kidney injury, prerenal impairment, or salt-losing nephropathies, but future experiments are required to rule out these possibilities.

Furthermore, we also determined serum inorganic phosphate and total calcium levels as well as the intestinal calcium transport in the ICR/Mlac-hydro mice. Since serum iPTH levels were markedly decreased in hydronephrotic mice, it was not surprising to observe decreases in serum 1,25(OH)2D3, iFGF23 and cFGF23 levels (Fig. 2), as well as the renal expression of TRPV5 (Fig. 3), the latter of which is known to be dependent on PTH. Under normal conditions, PTH positively regulates the conversion of 25(OH)D3 into 1,25(OH)2D3 using 1α-hydroxylase (encoded by Cyp27b1 gene) in the renal proximal tubular cells, thereby enhancing the 1,25(OH)2D3-dependent intestinal calcium absorption (for review, please see7). Meanwhile, 1,25(OH)2D3 is able to stimulate FGF23 production and secretion from osteoblasts and osteocytes31,32. Therefore, a reduction in 1,25(OH)2D3 level could lead to a decrease in transepithelial calcium flux across the duodenum of ICR/Mlac-hydro mice (Fig. 4), consistent with the downregulation of TRPV6 mRNA and protein expression (Fig. 3), which is crucial for calcium uptake into enterocytes. Both PD and Isc of the duodenal epithelium were also decreased in the 6-month-old ICR/Mlac-hydro mice, which might have resulted from the reduction in electrogenic transport of major ions, particularly the transcellular Na+ transport by Na+/K+-ATPase33,34. Indeed, a reduction in the intestinal Na+ absorption, if present, probably helped decrease severity of volume overload in mice with renal impairment35. Nevertheless, serum total calcium levels in ICR/Mlac-hydro were not changed, suggesting that there was a compensatory mechanism(s) coming into play to maintain circulating calcium concentration, which is critical for normal function of excitable cells such as cardiomyocytes and neurons. The elevated serum albumin (Fig. 1C), which could trap calcium in blood vessel36,37, probably one of the compensatory mechanisms to help normalize serum total calcium levels, as shown in Fig. 1D.

Since PTH and FGF23 were both responsible for the enhancement of renal phosphate excretion38, 2- and 3-month-old ICR/Mlac-hydro mice showed higher serum phosphate levels than in wild-type mice. This could be explained by the declining PTH action on the kidney, which led to an upregulation of the activity of Na+-dependent phosphate cotransporters (e.g., NaPi2a and NaPi2c) for phosphate reabsorption and reduction in phosphate excretion in the proximal tubule, thus resulting in hyperphosphatemia39. Under normal conditions, excess phosphate in the circulation was primarily excreted via the kidney40; therefore, a damage of renal parenchyma in hydronephrotic mice inevitably compromised phosphate excretion, thereby aggravating the already occurring hyperphosphatemia. The present results (Fig. 1E) suggested that the degree of phosphate dysregulation in hydronephrotic mice might be age-dependent—i.e., becoming less severe with an increase in age; however, the underlying mechanism was not known. It was tempting to speculate that sub-normal 1,25(OH)2D3 levels also contributed to the maintenance of normophosphatemia by reducing the intestinal phosphate absorption in hydronephrotic mice, and as previously suggested, kidney maturation helped increase the fractional excretion of phosphate per functioning nephron in advanced age41. This hypothesis warrants further validation by the expression profile of Na+-dependent phosphate cotransporters in the intestine and kidney, which is tightly regulated by 1,25(OH)2D3 and FGF23, respectively42,43.

Bone defects in hydronephrotic mice with low iPTH could probably be explained in a straightforward manner. An elevated phosphate level as well as lower calcium supply due to an impaired intestinal calcium absorption hindered bone calcium accretion and matrix mineralization. These findings are consistent with lower trabecular bone volume fraction, double-labeled surface, mineralizing surface and mineral apposition rate as determined by µCT and dynamic bone histomorphometry (Figs. 6 and 8). Meanwhile, inadequate calcium supply and aberrant calcium/phosphate ratio could also impair longitudinal bone growth44, as suggested by shorter femoral length in ICR/Mlac-hydro mice (Fig. 5). Under normal conditions, both PTH and 1,25(OH)2D3 are essential for maintaining normal activities of osteoblasts and osteoclasts. Strikingly, hydronephrotic mice were found to have lower bone formation rate, osteoblast surface and active erosion surface, indicating that both osteoblast-mediated bone formation and osteoclast-mediated bone resorption were suppressed, presumably by sub-normal levels of PTH and 1,25(OH)2D3.

Bone histomorphometry and biomarker data unanimously supported low bone turnover in hydronephrotic mice as shown by reduction in both bone formation and bone resorption parameters. Therefore, ICR/Mlac-hydro mice probably encountered a detrimental phenomenon known as adynamic bone, whereby old bone with micro-cracks or microdamage could not be effectively replaced and healed by bone remodeling process. Both trabecular and cortical vBMD were significantly reduced in hydronephrotic mice as a result of lower bone mineral deposition (Fig. 5). Moreover, deterioration of mechanical properties, i.e., decreased maximal load, yield load, and stiffness (Fig. 7), was consistent with the decrease in cortical vBMD, indicating that the diaphysis of ICR/Mlac-hydro mouse had reduced strength to resist deformation and fractures. In other words, hydronephrotic mice were susceptible to bone fracture after the third month of age.

In murine bone, sexual dimorphism can be observed at both the organ and cellular levels. Specifically, Glatt et al.45 reported that femoral trabecular separation in female mice was more pronounced than in male mice; therefore, significant trabecular changes are expected to occur in females rather than males. In an in vitro osteoclastogenesis study46, the number of osteoclasts differentiated from female bone marrow cells was greater compared to those derived from males. The exact reason why female rodents are often more susceptible to osteoclast-mediated bone resorption is not fully understood, but it may be related to circulating levels of estrogens, androgens, and/or inflammatory mediators (for review, please see47). Nevertheless, it is noteworthy that exogenous estrogen treatment can induce osteoclast apoptosis and inhibit bone resorption47.

Limitations of the present study included the absence of data showing the underlying cellular and molecular mechanism(s) of calciotropic regulation in bone cells and enterocytes. In addition, because the low availability of mutant animals particularly female mutants posted a problem, certain parameters (e.g., serum phosphate and calcium levels) were absent. Since both male and female mice have been reported to develop the same severity of hydronephrosis16, they should exhibit similar endocrine dysregulation and bone defects. Indeed, sex-dependent discrepancy did exist for some parameters. For instance, a greater trabecular separation was observed only in 3-month-old female, but not male ICR/Mlac-hydro mice as compared to their age-matched wild-type control groups (Fig. 6). However, serum estrogen levels and vaginal cytology were not assessed in the present study. Additional experiments, such as exogenous calciotropic hormone administration and rescue experiments, are required to demonstrate how aberration of PTH and 1,25(OH)2D3 activities contributes to calcium dysregulation in ICR/Mlac-hydro mice. Direct measurement of osteoblast and osteoclast activities can further reveal the underlying cellular mechanisms behind adynamic bone phenotype. In general, adynamic bone can be observed in several conditions, such as end-stage renal disease with hypoparathyroidism, chronic use of calcimimetics or active vitamin D, aging, and pediatric patients with extremely low PTH levels. Although it is too early to conclude that ICR/Mlac-hydro mice can represent hydronephrotic pediatric or chronic kidney disease patients, they may serve as an alternative preclinical model for testing vitamin D analogs or PTH treatment.

In conclusion, we showed that the hydronephrotic ICR/Mlac-hydro mice manifested sub-normal iPTH, 1,25(OH)2D3 and FGF23 levels, which probably led to an impaired intestinal calcium absorption and hyperphosphatemia. Defective bone elongation, low vBMD and deteriorated bone microstructure probably resulted from a reduction in mineral apposition and adynamic bone disease, thereby leading to a compromised bone strength (Fig. 10). This mouse model should be potentially useful for future investigations of therapeutic intervention for calcium and phosphate dysregulation and metabolic bone derangement in hydronephrosis with hypoparathyroidism. In terms of knowledge translation, we propose that patients with hydronephrosis—particularly female patients—should be screened for bone turnover markers and BMD to assess the presence of adynamic bone disease.

Fig. 10
figure 10

Graphical illustration demonstrated effects of hydronephrosis on bone and calcium metabolism in ICR/Mlac-hydro mice. Degeneration of renal parenchyma in hydronephrotic mice could impair renal function as manifested by increase BUN and serum phosphate. The unaltered total calcium level in ICR/Mlac-hydro mice might be resulted from elevated serum albumin with unknown mechanism (dash arrow). Hydronephrotic mice manifests sub-normal iPTH, 1,25(OH)2D3 and FGF23 levels, that may lead to impairment of intestinal calcium absorption. Impaired intestinal calcium absorption prevents bone calcium accretion and mineral apposition, resulting in low bone formation and vBMD, deteriorated bone microstructure, defective bone elongation, eventually leading to compromised bone strength.

Full size image