Background
Brain-derived neurotrophic factor (BDNF) belongs to the neurotrophin superfamily which promotes the proliferation, development, survival, and differentiation of neurons in the peripheral and central nervous systems [
1]. BDNF is synthesized as pro-BDNF and is secreted as a mixture of pro-BDNF and mature BDNF. Mature BDNF binds to the tropomyosin-related kinase receptor B (tyrosine kinase B, TrkB) and thereby initiates phosphorylation through the mitogen-activated protein kinases, phosphatidylinositol 3-kinase, and phospholipase C-gamma signaling pathways, which ultimately accelerate protein synthesis, axonal growth, dendritic cell maturation, synaptic plasticity, and neuroprotective effect. By contrast, pro-BDNF binds preferentially to the p75 neurotrophin receptor which mediates apoptosis [
2]. Low BDNF level or impairment of its signaling pathway is thought to be implicated in a variety of neuropsychiatric and neurological disorders. Although BDNF is strongly expressed in the neocortex, hippocampus, amygdale, and cerebellum of the central nervous system, it is also expressed in nonneuronal tissues including heart, retina, urethral sphincter, liver, airway smooth muscle, ovary, and fetal kidney [
3‐
8]. Nevertheless, the role and expression of BDNF in the kidney remains underexplored.
Chronic cyclosporine A (CsA) nephropathy is a common complication observed during treatment for solid-organ transplantation and autoimmune diseases. This complication is manifested by afferent arteriolopathy, striped interstitial fibrosis, tubular atrophy, and progressive renal insufficiency [
9]. CsA treatment also induces renal tubular injury manifested by polyuria, magnesium wasting, renal distal tubular acidosis, and hyperkalemia. Of these, impaired urinary-concentrating ability is a predominant feature of chronic CsA nephropathy in clinical practice [
10]. Several factors, such as the renin-angiotensin system, nitric oxide, apoptosis, inflammatory mediators, innate immunity, transforming growth factor-β1 (TGF-β1), aquaporins, and urea transporters have been suggested as possible mediators of the evolution of chronic CsA nephropathy [
9,
11]. However, the precise mechanism remains unclear.
Recent publications have delineated that calcineurin inhibitor (CsA or FK506) regulates the expression of the BDNF/TrkB pathway in the hippocampus and midbrain of astrocytes and that differential expression of BDNF/TrkB triggers different roles in the central nervous system, such as neurotoxicity or neuroprotection [
12‐
14]. Considering the above mentioned background, we hypothesized that BDNF would be expressed in the kidney and may play a function because the renal tubules are vulnerable to CsA-induced insult. To test this hypothesis, this study was designed to assess the expression and localization of BDNF and Trk receptors in a well-established animal model of chronic CsA nephropathy.
Methods
Animals and drugs
Male Sprague-Dawley rats (Charles River, Wilmington, MA) weighing 200 to 220 g were placed in cages (Nalge Co., Rochester, NY) at 37 °C with a fixed light-dark cycle and were permitted food and water ad libitum. Starting 7 days before and continuing throughout the experimental period, rats were provided a low-salt diet (0.05% sodium, Teklad Premier, Madison, WI).
CsA (Novartis Pharma Ltd., Basle, Switzerland) was dissolved in olive oil to 15 mg/mL concentration. Exogenous 1-desamino-8-D-arginine vasopressin (DDAVP; Sigma, dissolved in 0.9% NaCl solution at 10 mM) was given via an implantable osmotic minipump.
Study design
This protocol was approved by the Animal Care Committee of the Catholic University of Korea. Two animal studies were carried out.
Protocol1
The first experiment was conducted to examine the influence of CsA treatment on the expression of BDNF and Trk receptors. After acclimatization and a low-salt diet for 1 week, weight-matched rats were randomly divided into two groups and treated daily for 4 weeks as follows: 1) vehicle group (VH): rats were injected subcutaneously with olive oil (1 mL/kg); 2) CsA group: rats were injected subcutaneously with CsA (15 mg/kg). Each group contained 8 rats.
Protocol2
The second experiment was performed to examine the influence of exogenous vasopressin treatment on the expression of BDNF and AQP-2, and the urinary concentration in CsA-induced nephrotoxicity. The weight-matched rats were randomly divided into four groups and treated daily for 4 weeks as follows: 1) VH (
n = 8); 2) VH + DDAVP (
n = 8); 3) CsA (
n = 8); 4) CsA + DDAVP (
n = 8). The treatment protocol of olive oil and CsA was similar to that of protocol1. Exogenous DDAVP was administered for the last 7 study days as previously described [
10,
15].
Under ketamine anesthesia, animals were euthanized at the end of study, and kidney specimens were withdrawn for further evaluations.
The investigators were blinded to the identity of the treatment group allocation and data analyses.
Basic measurements
Body weight (BW) of pair-fed rats were recorded, and 24-h urine and blood samples for each rat were collected for measurement of urine osmolality (Fiske Associates, Norwood, MA), serum creatinine (Scr), blood urea nitrogen (BUN), and CsA concentration, as we have previously reported [
16].
Systolic blood pressure (SBP) was measured with a tail manometer-tachometer system (BP-2000, Visitech system, Apex, NC). In brief, rats were placed into circular cases, and then SBP was detected using the tail-cuff system after rats were quiescent for 3–5 min. To obtain accurate measurements, three measurements for each rat were averaged.
Histology
Kidney specimens were stained with Masson’s trichrome and hematoxylin. Evaluation of tubulointerstitial fibrosis (TIF) was performed as previously described [
17,
18].
Immunohistochemistry
Sections were incubated for BDNF (ab108383, Abcam’sRabMAb® technology, USA), TrkB (ab33655, Abcam’sRabMAb® technology USA), and TrkC (ab33656, Abcam’sRabMAb® technology, USA).
Double immunofluorescence
Immunostaining of BDNF, TrkB and TrkC was performed using horseradish peroxidase conjugate as secondary antibody and 3,3′-diaminobenzidine as chromogen. Sections primary stained with BDNF, TrkB and TrkC were incubated with Na-K-ATPase-α1 (Abcam’sRabMAb®technology USA) antibody. Na-K-ATPase-α1 labeling was examined with fluorescein isothiocyanateY conjugated donkey anti-rabbit IgG antibody (Jackson ImmunoResearch Laboratories). The procedure of immunofluorescence for AQP1 (ab65837 Abcam’sRabMAb® technology USA), AQP2 (ab15116, Abcam’sRabMAb® technology USA), and Bcl-2 (Bioworld Co.,Minneapolis, MN, USA) was similar to that for Na-K-ATPase-α1. Images of the tissue sections were captured using a Carl Zeiss photomicroscope equipped with differential interference contrast optics and fluorescence imaging capabilities (Axio Imager M2; Carl Zeiss, Jena, Germany).
Immunoblotting
Immunoblotting analyses were performed as previously described [
17].BDNF, TrkB, TrkC, and AQP-2 were detected with specific antibodies. Densitometric analyses were referred to the VH group as 100%.
Apoptotic cell death
In situ TdT-mediated dUTP-biotin nick end-labeling (TUNEL) assay was performed according to Apoptosis Detection Kit instruction manual (Intergen, NY). TUNEL-positive cells were auto-counted using a digital camera-based image analyzer at 20 nonoverlapped fields in each section.
Examination of urinary BDNF and 8-Hydroxy-2′-Deoxyguanosine (8-OHdG)
Urinary total BDNF (R&D Systems, Inc., Minneapolis, MN, USA) level and oxidative stress biomarker 8-OHdG (Institute for the Control of Aging, Shizuoka, Japan) concentrations in serum and urine were examined using competitive enzyme-linked immunosorbent assay (ELISA).
Statistical analysis
The data are presented as mean ± SEM. Comparisons between groups were accomplished using Student’s t-test for protocol1 and one-way ANOVA with post hoc Bonferroni test for protocol 2 (SPSS software version 21.0, Microsoft Corp.). Correlation coefficient analyses were used to examine the relationships among TUNEL-positive cells, BDNF expression, and urinary 8-OHdG. Significance was considered as p < 0.05.
Discussion
The present study clearly demonstrates that BDNF and Trk receptors (TrkB and TrkC) are constitutively expressed in the collecting duct of normal rat kidneys, whereas the expression is significantly lower in the kidneys of CsA-treated rats. Suppression of BDNF expression by CsA was paralleled by an increase in urine volume and a decrease in urinary osmolality. These effects were reversed after administration of DDAVP to the CsA-treated rats. Our findings suggest that the BDNF/Trk receptor system plays a role in the regulation of urine-concentrating capacity in chronic CsA nephropathy.
In addition to expression in the central nervous system, BDNF and Trk receptors are expressed in a wide range of nonneuronal tissues. However, few studies have examined the expression of BDNF and Trk receptors in the kidney. In the fetus, BDNF is expressed in the primitive glomeruli and primitive tubules, and Trk receptors (TrkB and TrkC) are expressed in the proximal and distal tubules and collecting tubule epithelial cells [
8]. As development progresses, their expression becomes more prominent in the collecting duct system except for the glomeruli [
20‐
22]. In the present study, double immunofluorescent staining showed that BDNF and Trk receptors were abundant in the collecting tubules of the cortex and medulla, but were confined to the principal cells and epithelial cells, whereas other structures were negative for BDNF and Trk receptors. Our findings are consistent with those of previous reports that found similar localization of BDNF and Trk receptors in the kidneys of teleosts, frogs, lizards, and humans [
20‐
22], and suggest that the pattern of BDNF and Trk receptor expression changes during kidney development.
One interesting finding in this study is that CsA treatment significantly decreased BDNF expression, and this was accompanied by a decrease in TrkB and TrkC expression. The mechanism by which CsA suppresses BDNF and Trk receptor expression in this model may be direct or indirect. CsA has been shown to directly downregulate BDNF and Trk receptor expression in cultured SH-SY5Y cells in the hippocampus and midbrain, and that this may be related to the depressive symptoms of chronic CsA neurotoxicity [
14]. In addition, CsA may indirectly decrease BDNF-Trk receptor expression by inducing apoptotic cell death. Amore et al. reported that CsA induces apoptotic cell death in cultured renal cells and murine tubular epithelial cells [
23]. CsA is associated with hypoxia injury caused by afferent artery constriction, which ultimately results in oxidative stress [
18]. It is accepted that oxidative stress stimulates apoptotic cell death, and that antioxidant therapy inhibits this program [
24‐
26]. There is overwhelming evidence of a close relationship between oxidative stress and apoptosis in chronic CsA nephropathy [
27,
28]. In the present study, treatment of rats with CsA increased the number of TUNEL-positive cells that are localized mainly to the proximal tubules, thick ascending limb, and collecting duct cells. Double immunofluorescence staining showed that BDNF and its receptors colocalized with Bcl-2 in the collecting duct cells in kidneys of CsA-treated rats. The existence of TUNEL-positive cells in collecting duct cells was ascertained by double immunolabeling for AQP-2 and the TUNEL assay, as previously reported by our laboratory [
10]. Moreover, the number of TUNEL-positive cells correlated negatively with BDNF protein expression (
r = − 0.866,
p < 0.001) and positively with urinary 8-OHdG excretion (
r = 0.884,
p < 0.001). These findings suggest that loss of renal cells by apoptosis may partially account for the decreased BDNF expression in chronic CsA nephropathy.
BDNF functions as a neuroprotective agent against various neural insults through its role in neuronal cell proliferation, differentiation, and survival [
1,
29]. These actions have been confirmed by studies of exogenous BDNF administration or BDNF-deficient animals, which have shown the neuroprotective effects of BDNF in the central nervous system [
30‐
32]. However, the role of BDNF in the kidney is largely unknown. To identify its function in CsA toxicity, we infused exogenous DDAVP into CsA-treated rats. We found that BDNF expression was restored by DDAVP treatment and that this was accompanied by recovery of the urine-concentrating ability, as shown by improved urine osmolality and volume. The alterations in BDNF expression in this study lead us to speculate that the function of BDNF may be related to the regulation of urine-concentrating ability.
One potential limitation when interpreting the results of our study is the role of AQP-2 in urine-concentrating ability. AQP-2 is a member of a membrane protein family that plays a critical role in reabsorption of water in the kidney [
33]. Therefore, changes in urine volume and osmolality during CsA or DDAVP treatment may arise from the actions of AQP-2 rather than BDNF. However, several items of evidence should be considered. First, we previously reported that AQP-2 was expressed in the apical region of collecting duct cells in normal rat kidneys, but that CsA treatment decreased AQP-2 protein expression only in the inner medulla of the collecting duct. However, AQP-2 protein expression within the cortex and outer medulla of the collecting duct did not differ between the VH- and CsA-treated groups [
10,
34]. Second, BDNF expression was decreased in the CsA-treated rat kidneys throughout the collecting duct system in our current study (cortex, inner medulla, and outer medulla, Fig.
1b), and upregulation of BNDF expression by DDAVP was unaccompanied by restoration of AQP-2 expression. Third, BDNF may act as a neuromodulator [
3,
34‐
36]. Using an animal model of bladder pain syndrome/interstitial cystitis, Frias et al. found that intrathecal injection of BDNF improved bladder function and relieved cyclophosphamide-induced cyctitis by modulating detrusor overactivity [
33]. Murray et al. showed that increased BDNF expression in dorsal root ganglion cells of bladder afferents is involved in bladder innervations [
36]. These findings suggest that BDNF plays a neuromodulatory role in urinary tract symptoms including urgency, frequency, and incontinence. Similar neuromodulatory effects of BDNF have been observed in congestive heart failure, insulin secretion after chronic exercise, and carbachol-induced contraction of intestinal longitudinal smooth muscle [
3,
37,
38]. Thus, regulation of urine-concentrating ability, shown in this study, may contribute to the action of BDNF.