Renoprotective effect of berberine via regulating the PGE 2‐EP1‐Gαq‐Ca2+ signalling pathway in glomerular mesangial cells of diabetic rats

Abstract G‐protein coupled receptor‐mediated pathogenesis is of great importance in the development of diabetic complications, but the detailed mechanisms have not yet been clarified. Therefore, we aimed to explore the roles of the prostaglandin E2 receptor 1 (EP1)‐mediated signalling pathway and develop a corresponding treatment for diabetic nephropathy (DN). To create the DN model, rats fed a high‐fat and high‐glucose diet were injected with a single dose of streptozotocin (35 mg/kg, i.p.). Then, rats were either treated or not with berberine (100 mg/kg per day, i.g., 8 weeks). Cells were isolated from the renal cortex and cultured in high‐sugar medium with 20% foetal bovine serum. Prostaglandin E2 (PGE 2) levels were determined by ELISA, and cells were identified by fluorescence immunoassay. We measured the biochemical characteristics and observed morphological changes by periodic‐acid‐Schiff staining. The expression of the EP1 receptor and the roles of GRK2 and β‐arrestin2 were identified using western blotting and flow cytometry. Downstream proteins were detected by western blot, while molecular changes were assessed by ELISA and laser confocal scanning microscopy. Berberine not only improved the majority of biochemical and renal functional parameters but also improved the histopathological alterations. A significant increase in PGE 2 level, EP1 membrane expression and Gαq expression, and concentration of Ca2+ were observed, accompanied by increased GRK2 and β‐arrestin2 levels soon afterwards. Berberine decreased the abnormal concentration of Ca2+, the increased levels of PGE 2, the high expression of EP1 and Gαq and suppressed the proliferation of mesangial cells. The EP1 receptor, a critical therapeutic target of the signalling pathway, contributed to mesangial cell abnormalities, which are linked to renal injury in DN. The observed renoprotective effects of berberine via regulating the PGE 2‐EP1‐Gαq‐Ca2+ signalling pathway indicating that berberine could be a promising anti‐DN medicine in the future.


Introduction
Diabetic nephropathy (DN) is believed to be the major cause of morbidity and mortality in patients with diabetes and is highly prevalent in end-stage renal disease [1]. It is characterized by complete or partial microalbuminuria, glomerular mesangial cell (GMC) proliferation with deposition of extracellular matrix (ECM) at the glomerular level, global glomerular sclerosis and atrophy, which ultimately result in progressive renal failure [2,3]. Hyperglycaemia is a necessary precondition in the process of DN, whereas dyslipidaemia is an equally important aggravating factor [4]. Many factors, including systemic hypertension with hyperfiltration [5], increased advanced glycation end products (AGEs), protein kinase C (PKC) pathway activation and cytokines, have been identified as critical deteriorating factors that accelerate DN based on genetic susceptibility [6]. However, the precise mechanisms are not yet fully elucidated. Prostaglandins are derived from the 20-carbon chain fatty acid, arachidonic acid (AA) stored in the plasma membrane of cells [7]. Prostanoids, including prostaglandins (PGs) PGE2, PGF2a, PGI2, PGD2 and thromboxane A2 (TXA2), are the oldest members of the eicosanoid family. Prostanoids can be biosynthesized from three related fatty acid precursors, 8,11,14-eicosatrienoic acid (homo-linolenic acid), 5,8,11,14-eicosatetraenoic acid (EAA) and 5,8,11,14,17-eicosapentaenoic acid (EPA or timodonic acid) [8]. Once synthesized, prostanoids are transported into the extracellular microenvironment by specific multidrug resistance-associated proteins (MRPs). After being exported to the microenvironment, prostanoids bind to G-protein-coupled receptors that contain seven transmembrane-spanning domains [9]. Among the numerous prostanoids, PGE 2 has been shown to play significant roles in regulating renal physiology, including glomerular filtration, renin release, tubular salt and water metabolism [10]. Multiple actions of PGE 2 were found to be mediated by specific G-protein coupled receptors (GPCRs) [11], and local production of PGE 2 has been shown to be associated with glomerular inflammation and renal injury [12]. In glomerular microcirculation, E prostanoid receptor (EP) is one of the most abundant GPCRs, when coupled to the corresponding G protein, regulates renal functional alterations [13]. One study showed that EP1, an important EP receptor subtype, plays important roles in diabetic progression, together with PGE 2, as suggested by a reduced glomerular size in EP1-deficient mice [14]. Meanwhile, pathophysiological disruptions by the PGE 2 /EP1 axis suggest that selective blockade of EP1 may be a useful therapeutic option in DN [15]. Therefore, research on the PGE 2 -EP1-related signalling pathway in diabetic rats is urgently needed. Berberine (BBR; [C 20 H 18 NO 4 ] + ), a type of isoquinoline alkaloid, is the major constituent of Rhizoma coptidis [16]. Research demonstrated that BBR exhibited various pharmacological activities, including antioxidation, anti-inflammatory, antilipemic and hypoglycaemic activities, which indicated potential clinical and research value as a future therapeutic anti-DN drug [17]. Most recently, research showed that BBR could effectively reduce the PGE 2 levels in various diseases, which suggested that there might be a relationship between BBR and the PGE 2 /EP axis in the process of DN [18]. However, to date, the precise mechanisms are still being investigated. This study aimed to explore the renoprotective effects of BBR and the detailed mechanism via the PGE 2 -EP1-Gaq-Ca 2+ signalling pathway in GMCs of diabetic rats.

Experimental protocol
Male Sprague-Dawley rats (180 AE 20 g. Grade SPF, No. 34000200000613) were obtained from the Animal Department of Anhui Medical University (Hefei, China). Rats were acclimatized and then randomly assigned to four groups: Normal control (NC) group, DN control (DNC) group, DNC plus BBR (100 mg/kg) treatment (DNC+BBR) group and DNC plus enalapril (1 mg/kg) treatment (DNC+Enalapril) group. Fifteen rats were assigned to the NC group, while the other three groups, respectively, to ensure that at least 15 rats survived the entire experiment. After an overnight fast, rats were intraperitoneally administered a single injection of STZ (35 mg/kg) dissolved in a 0.1 mM chilled citrate-phosphate buffer (pH 4.5). NC rats were injected with citrate-phosphate buffer. The development of hyperglycaemia (diabetes) was confirmed by fasting blood glucose (FBG) estimation (≥11.1 mmol/l) 72 hrs after injection, and only uniformly diabetic rats were included in the DNC, DNC+BBR and DNC+Enalapril groups. NC and DNC rats were given equal volumes of sodium carboxymethyl cellulose (CMC-Na), while the rest were administered BBR (100 mg/kg/day) and enalapril (1 mg/kg/day) solution intragastrically, for 8 weeks. All experiments were approved by the Ethics Review Committee for Animal Experimentation of the Anhui Medical University.

Biochemical characteristics analysis
After the experiment, the rats were weighed and placed in metabolic cages for 24-hr urine collection. Blood samples were collected, and the supernatants were used for the measurement of blood urea nitrogen (BUN), serum creatinine (SCr) and the ratio of urine total protein to urine creatinine (UTP/C) using an automated biochemistry analyzer (Model 7600 Series Automatic Analyzer; Hitachi Corporation, Tokyo, Japan).

Morphological analysis
Kidney samples were processed using the standard protocol for paraffin embedding. The specimens were sectioned at 4 lm-thick and stained with periodic acid-Schiff stain (PAS) prior to examination.

ELISA assay
Kidney tissue homogenates and cultured media supernatants were collected for PGE 2 measurement using an ELISA kit following the standard protocol. PGE 2 levels were normalized to the amount of cellular protein to which the conditioned media were exposed. The cross-reactivity between PGE 2 and PGF 2a was <0.01%.

Western blot assay
Kidney tissues and cultured cells were washed with precooling phosphate buffer, and then lysed with lyolysis. Proteins were extracted and subjected to SDS-PAGE and subsequently transferred to a polyvinylidenfluoride membrane. The membrane was probed with rabbit monoclonal anti-EP1 antibody (tissue-derived samples, 1:500; cell-derived samples, 1:800), anti-GRK2 antibody (1:500), mouse polyclonal anti-b-arrestin2 and anti-Gaq antibody (1:500), and b-actin was used as an internal control. After incubation with anti-rabbit or mouse IgG (1:30,000), immune complexes were detected using the Super Signal Chemiluminescence Kit (Thermo Scientific, Rockford, IL, USA).

Primary culture of GMCs
Kidneys were aseptically harvested and then stored in precooled Hanks' buffered salt solution (HBSS; 137 mM NaCl, 5.4 mM KCl, 1.2 mM CaCl 2 , 0.8 mM MgSO 4 , 5.6 mM D-glucose, 4.1 mM NaHCO 3 , Na 2 HPO 4 Á12H 2 O and 15 mM Hepes, pH 7.4). Glomeruli were isolated using the sieving method (100 mesh sieve above and 200 mesh sieve below). The fragments were collected and centrifuged until the purity of the glomeruli exceeded 98%; then, the supernatant was discarded. After digestion, the process was terminated using Dulbecco's modified Eagle's Medium containing 10% foetal bovine serum. The cells were uniformly scattered into culture flasks with antibiotics under appropriate conditions. Cultured cells were identified by hallmarks using a fluorescence immunoassay.

MTT assay and trypan blue staining
The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay was used to measure cell proliferation. Briefly, cells were seeded at 10 4 cells/well in 96-well plates. The medium was replaced with serum-free medium for a 24-hr incubation, and the cells were incubated in the presence or absence of different concentrations of BBR (5, 10, 30, 60, 90, 120 and 240 lM) for 20 and 44 hrs. MTT was added and incubated at 37°C for additional 4 hrs. The medium was carefully removed so as not to disturb the formazan crystals formed. Dimethylsulfoxide was added to each well, and the absorbance of solubilized blue formazan was read at a wavelength of 490 nm using a microplate reader. The optical densities of six wells in each group were measured and normalized to those of the normal group. Meanwhile, cell viability was determined by trypan blue staining to choose the appropriate concentration of BBR for such experiments. The same experiment was used to determine the proper concentration of PGE 2 .

Desensitization of EP1 receptor
Prostaglandin E 2 was used to stimulate the desensitization of the EP1 receptor in cultured cells. Briefly, the medium was replaced with serumfree medium for 24 hrs and then incubated with PGE 2 at different intervals. Flow cytometry and western blot assay were used to detect the EP1 expression on the cell membrane at different time points (0, 5, 10, 15, 20, 30, 60 and 120 min.). Fluorescent intensities were compared.

Laser confocal scanning microscopy analysis
Briefly, after treatment with or without BBR, cells were washed with phosphate-buffered saline, suspended in HBSS and then loaded for 40 min. with 4 lM fluo-3/AM mixed with Pluronic F127 in the dark. Then, the cells were washed, resuspended and assayed by laser confocal scanning microscopy (LCSM) (Leica TCS-SP2, Wetzlar, Germany). The fluo-3 fluorescence intensity was monitored with 488 nm excitation and 526 nm emission wavelengths in cell suspension in the dark.

Statistical analysis
Data were assessed using SPSS 16.0 (IBM Corporation, Armonk, NY, USA). Significant differences were evaluated using one-way ANOVA using a post hoc Bonferroni correction (GraphPad Prism 5.0; GraphPad Software, La Jolla, CA, USA). A two-sided P < 0.05 was considered significant. The data are presented as the mean AE S.D. (n ≥ 3).

Results
Effect of BBR on metabolic parameters A significant increase in FBG was detected after the induction of diabetes. Administration of BBR had an undistinguishable hypoglycaemic effect in short-term treatment (0-4 week), while the longterm treatment (6-8 week) was shown to be satisfactory compared with the DNC group. Enalapril had little impact on reducing the FBG level (Fig. 1A). Diabetic rats had lower body weights, while BBR treatment had no effect on regaining the abnormal weight (Fig. 1B). The ratio of kidney weight to body weight in diabetic rats was significantly increased compared with the NC group (P < 0.01), while BBR effectively decreased the ratio (Fig. 1C). The results of enalapril treatment were similar to the results of BBR treatment for both parameters. The levels of UTP/C, BUN and SCr in the DNC group were significantly higher than those of the NC rats. Berberine and enalapril exhibited a remarkable decrease in these indexes ( Fig. 1D-F).

Effect of BBR on renal histopathology
Untreated rats exhibited marked ECM accumulation, mesangial expansion, glomerular sclerosis and atrophy. When given BBR, pathological changes, such as GBM thickening and ECM accumulation, were scarcely detected, and mesangial expansion was also extremely reduced. However, glomerular atrophy and sclerosis were observed to some extent. Meanwhile, enalapril exhibited a great outcome of improving histological features (Fig. 2).

Effect of BBR on the level of PGE 2 in kidney tissue
As shown, PGE 2 secretion was markedly elevated in kidneys of the DNC group by more than fourfold, and such an induction was remarkably blunted by BBR treatment compared with the DNC rats (P < 0.05) (Fig. 3B).

Effect of BBR on the expression of EP1 in kidney tissue
Western blotting revealed a significant increase in renal EP1 expression in DNC rats (1.7-fold of NC; P < 0.05), whereas protein expression was attenuated by BBR treatment (0.7-fold of DNC; P < 0.05) (Fig. 4A).

Identification of cultured cells
Fluorescence immunoassay using an inverted microscope showed that the cultured cells had positive staining for of vimentin, desmin, asmooth muscle actin and myosin but negative staining for factor VIII antigen. The results confirmed that the cultured cells are GMCs (Fig. 5).

Effect of BBR on proliferation of GMCs
An MTT assay was used to assess the impacts of BBR and PGE 2 on the proliferation of GMCs. Interestingly, it was noted that administration of BBR for 24 or 48 hrs significantly suppressed the proliferation of GMCs, with a minimum concentration of 30 lM (P < 0.05). Trypan blue staining showed that cell mortality was less than 15% at the peak concentration of 90 lM, while cell mortality was 5% at a concentration of 60 lM (Table 1). Based on Figure 6B, 60 lM was considered a therapeutic concentration, and 10 lM was the optimum concentration of PGE 2 according to Figure 6A.

Desensitization of EP1 receptor in GMCs
After stimulation by PGE 2 , membrane EP1 expression in GMCs was gradually increased for 10 min. and then continuously decreased. Membrane EP1 expression was almost flat from that of the control level at 0 min. (Fig. 7). Western blotting showed that membrane EP1 expression reached a peak at 10 min. in response to PGE 2 stimulation, and then flow cytometry revealed that membrane EP1 expression decreased to near control level at 0 min. (Fig. 8).

Changes of cytoplasm GRK2 and membrane b-arrestin2
The expression level of GRK2 in the cytoplasm was clearly enhanced after 10 min. with PGE 2 stimulation compared with 0 min. (P < 0.01). Interestingly, membrane b-arrestin2 continuously increased after 10 min. under PGE 2 induction, and it remained flat close to that level (Fig. 9).

Effect of BBR on the level of PGE 2 in GMCs
The data indicated that the level of PGE 2 in GMCs of the DNC group was significantly elevated and was higher than that of the NC group. Berberine could effectively inhibit the increased PGE 2 level (P < 0.01) compared with that of the DNC group (Fig. 3C).

Effect of BBR on expression of EP1 and Gaq in GMCs
Compared with the NC group, increased expression of EP1 in GMCs with DN was observed, and increased expression of the Gaq protein was also observed. Berberine could apparently inhibit the expression levels of EP1 (Fig. 4B) and Gaq (Fig. 10A), as observed in the kidneys.

Effect of BBR on the concentration of Ca 2+ in GMCs
The results showed that, in the DNC group, the fluorescence intensity of cytosolic calcium increased and was significantly higher than that of the NC group, which meant that the concentration of free calcium in the cells of the DNC group was markedly higher than that in the NC group. Meanwhile, BBR-treated cells showed markedly decreased fluorescence intensity in GMCs, which indicated that BBR could significantly decrease the level of cytoplasmic calcium (Fig. 10B).

Discussion
Diabetic nephropathy is characterized by progressive renal failure, which is mainly caused by GMC proliferation and augmented deposition of ECM proteins at the glomerular level, thus inducing mesangial expansion, glomerulosclerosis and atrophy [19]. During the process, abnormal changes of renal function were mainly manifested as significant increases in kidney indexes, such as SCr, BUN and UTP/C, from a macroscopic viewpoint [20]. Current interventions for DN, namely, control of blood glucose, hypertension and blockade of the renin-angiotensin system, can hinder but not prevent the progressive loss of renal function that frequently accompanies DN [21]. Accordingly, the identification of new anti-DN drugs continues to be pursued. In this study, we explored the renoprotective roles of BBR and found that ratios of KW/BW and UTP/C as well as levels of SCr and BUN were significantly increased compared with the NC group, which indicated that the model was successfully established and glomerular injury and renal dysfunction emerged. Compared with the DNC group, BBR significantly ameliorated abnormal renal function beginning at the sixth week, which suggested that BBR exerted its renoprotective effect by improving functional parameters, but prolonged treatment is needed. In addition, BBR improved pathological alterations such as ECM accumulation, mesangial expansion, glomerulosclerosis and glomerular atrophy based on histological changes in the kidneys of DN rats. These findings supported the view that BBR can attenuate the progression of DN. Mesangial hypercellularity or cell proliferation preceded an increase in ECM proteins, subsequent ECM accumulation, and importantly, mesangial hypertrophy, which were associated with eventual glomerulosclerosis and atrophy [22]. Of course, there is an increasing awareness and consensus that the abnormal growth and activity of GMCs plays an important role in the pathophysiology of the early stages of DN. In addition, GMCs are considered specialized smooth muscle cells, which play pivotal roles in the regulation of glomerular haemodynamics. In glomeruli, GMCs formed a tree-like network, branching from the hilar site to the glomerular capillary loops and connecting with each other [23]. According to one report, BBR has multiple pharmacological activities, which indicate its wide clinical and research value as a therapeutic anti-DN drug [24]. According to the pathological analysis, BBR reduced the proliferation of GMCs in the DN state, thus reducing ECM accumulation and ameliorating glomerular sclerosis and atrophy. What is the mechanism of BBR when targeting GMCs in DN rats? In cellular activities, signalling transduction is a critical element and the foundation of the pathophysiological process, which is quite complex and has not been completely interpreted at present. Research demonstrated that PGE 2 is an inflammatory mediator that has pleiotropic effects on signalling cascades via specific GPCRs, the seven transmembrane proteins that form the largest single family of integral membrane receptors. In particular, excessive PGE 2 has been shown to be associated with an increased glomerular filtration rate in the early stage of DN and with subsequent glomerular hypertrophy, proteinuria and renal injury [25]. Further analysis using an ELISA assay showed that the production of PGE 2 in both kidney tissue and the supernatant of cultured GMCs were significantly elevated. On the basis of molecular and pharmacological research, the corresponding GPCRs, namely, the 'EP receptors', are now divided into four subtypes, EP1, EP2, EP3 and EP4, which are similar in structure, but differ in terms of tissue distribution. E-prostanoid receptors transduce information provided by extracellular stimuli into intracellular second messengers via coupling to heterotrimeric G proteins and the subsequent regulation of a diverse variety of effector systems after binding to ligand [26]. In this study, the expression of the EP1 receptor increased significantly both in glomeruli and cultured GMCs, which indicated that the PGE 2 /EP1 signalling pathway was activated and may transmit signals that regulate the activities of GMCs in the DN state. In the transduction process, the reversible phosphorylation of G protein kinase (GRK) played critical roles in signal identification and transfer in cells. There are seven known GRKs, and GRK2 is the most widely expressed. GRKs, how-ever, preferentially phosphorylate receptors that are in the agonistoccupied conformation. Under such circumstance, homologous desensitization or internalization of GPCR was mediated by phosphorylation of the receptor and subsequent binding of b-arrestin, which were concentrated inside of the cell membrane along with GRKs [27]. In addition, the recently emerging and constantly expanding field of heptahelical receptor GRK-b-arrestin-dependent signalling also offered several exciting new opportunities for therapeutic intervention in many diseases. Despite still being in its infancy, as far as identification of specific physiological and pathophysiological effects is concerned, there is a huge potential for exploiting DN for therapeutic purposes. Flow cytometry was used to determine the membrane levels of EP1 expression on GMCs after stimulation by PGE 2 at multiple time points, and the results of western blotting showed that the membrane level of EP1 expression gradually increased up to the peak at 10 min., and then decreased to the control level. At this timepoint, the expression level of GRK2 in the cytoplasm was measured, and the results showed that there was no significant change in GRK2 level in the cytoplasm from 0 to 10 min., and afterward the expression level of GRK2 in the cytoplasm increased significantly. However, the membrane expression of b-arrestin2 continuously increased after 10 min. The fluorescence intensity of GMC stimulated by PGE 2 was enhanced in response to PGE 2 activation. At approximately 10 min., the fluorescence intensity of EP1 on the GMC membrane was markedly increased, and then it gradually decreased, indicating the internalization of EP1. Such expression changes of GRK2 and b-arrestin2 led us to hypothesize that the internalization of EP1 was modulated by GRK2 and b-arrestin2. In the subsequent experiments, BBR was applied to the cultured cells after stimulation with PGE 2 for 15 min. Western bloting was used to detect alterations of the EP1 protein and the corresponding G protein, i.e., Gaq, and the results showed that the expression levels of EP1 and the Gaq protein were markedly increased compared with the NC group, while BBR effectively attenuated the over-expression of EP1 and Gaq. Based on the above findings, we hypothesized that BBR altered the expression of EP1 and then decreased the level of Gaq in the intracellular communication, eventually mediating the PGE 2 -EP1-Gaq signalling pathway by regulating the desensitization of the EP1 receptor. On the basis of the signalling pathway analysis, the concentrations and changes of cytosolic-free calcium ions in the DN condition were examined by LCSM. The results showed that intracellular free calcium fluorescent intensity was markedly elevated due to PGE 2 stimulation in GMCs. By contrast, fluorescent density in the BBR-treated group was markedly decreased compared with the DNC group, which indicated that BBR may effectively regulate the activities of GMCs by reducing the intracellular concentration of free calcium ion. Intracellular Ca 2+ activity, or intracellular Ca 2+ concentration ([Ca 2+ ]i), is an important biological signal in the control of protein secretion, contraction, development and apoptosis in a wide variety of cells. Many cellular functions critically depend on [Ca 2+ ]i and a sustained elevation in [Ca 2+ ]i is capable of activating multiple potentially harmful cellular processes, including phenotypic change, dysregulated cell proliferation, cell injury, apoptosis, and death [28]. Like vascular smooth muscle cells, many different functions of GMCs were controlled by [Ca 2+ ]i. Under such conditions, GMCs changed their phenotype and became proliferative and matrix expanding subsequently, they promoted ECM accumulation, mesangial hypertrophy and even glomerulosclerosis and ultimately accelerated the progression of DN [29]. Meanwhile, one study reported that PGE 2 /EP1 signals mediated cytosolic Ca 2+ channels [30]. Therefore, it  would be reasonable to conclude that BBR may affect the abnormal proliferation, contraction and cytokine and protein biosynthesis of GMCs by regulating PGE 2 -EP1-Gaq-Ca 2+ signalling transduction to ameliorate the above symptoms. Whether targeting the PGE 2 -EP1-Gaq-Ca 2+ signalling pathway represents a new therapeutic option for diabetic patients remains to be established in further studies. To date, the pathogenesis of DN has not been completely clarified, and the exact mechanisms of BBR treatment are not well understood. More in-depth analysis should unravel the profound mystery of and explore the application of BBR. We next plan to test these possibilities by knocking-out or silencing the EP1 gene, and then we can elaborate on the exact renoprotective mechanism of BBR at the gene levels by determining the mRNA expression of EP1 and the corresponding molecules.

Conclusions
In summary, BBR not only attenuates DN by ameliorating changes in renal pathophysiology but also by suppressing the proliferation of GMCs through the PGE 2 -EP1-Gaq-Ca 2+ signalling pathway. The data presented here support the concept that EP1 may be an important therapeutic target for DN treatment and other kidney diseases related to the PGE 2 -EP1 signalling pathway.  The importance of BBR in treating DN leads us to pursue a more detailed analysis, while a better understanding of the mechanisms and physicochemical properties will provide sufficient theoretical support for us to seek the possible therapeutic targets and promote the clinical application of BBR in DN [31]. However, to date, there is no report detailing the effects of BBR on diabetic patients, indicating the enormous potential application in treating diabetic kidney disease in patients.