A new LKB1 activator, piericidin analogue S14, retards renal fibrosis through promoting autophagy and mitochondrial homeostasis in renal tubular epithelial cells

Background: Liver kinase B1 (LKB1) is the key regulator of energy metabolism and cell homeostasis. LKB1 dysfunction plays a key role in renal fibrosis. However, LKB1 activators are scarce in commercial nowadays. This study aims to discover a new drug molecule, piericidin analogue S14 (PA-S14), preventing renal fibrosis as a novel activator to LKB1. Methods: Our group isolated PA-S14 from the broth culture of a marine-derived Streptomyces strain and identified its binding site. We adopted various CKD models or AKI-CKD model (5/6 nephrectomy, UUO, UIRI and adriamycin nephropathy models). TGF-β-stimulated renal tubular cell culture was also tested. Results: We identified that PA-S14 binds with residue D176 in the kinase domain of LKB1, and then induces the activation of LKB1 through its phosphorylation and complex formation with MO25 and STRAD. As a result, PA-S14 promotes AMPK activation, triggers autophagosome maturation, and increases autophagic flux. PA-S14 inhibited tubular cell senescence and retarded fibrogenesis through activation of LKB1/AMPK signaling. Transcriptomics sequencing and mutation analysis further demonstrated our results. Conclusion: PA-S14 is a novel leading compound of LKB1 activator. PA-S14 is a therapeutic potential to renal fibrosis through LKB1/AMPK-mediated autophagy and mitochondrial homeostasis pathways.


PA and PA-S14
The strain Streptomyces psammoticus SCSIO NS126, isolated from a mangrove sediment sample collected from the Pearl River estuary to South China Sea, was fermented for 100 L, according to a previously described method [1]. The culture broth was extracted with ethyl acetate three times, and then concentrated under vacuum. The extract (75.0 g) was chromatographed on silica gel to give eight fractions (Frs.1~8). Frs.2 was purified by silica gel to obtain a pure compound (355 mg), which was identified as piericidin A (PA) as reported in our previous study [1].
With the guide of HPLC analysis, Frs.7 was purified by silica gel again to give six sub-fractions (Frs.7-1~7-6). Frs.7-4 was purified by semipreparative HPLC to obtain a piericidin analogue compound labeled as S14 (PA-S14, 52 mg). PA-S14 was suggested to be a hydroxylative product of glucopiericidin A because of its molecular weight 593 given by MS analysis ( Figure S2). It was determined to be 13-hydroxyglucopiericidin A by comparison of its 1 H and 13 C NMR date ( Figure S2) with literature data [2]. The obtained compounds, determined to have ≥95% purity by analytical HPLC, were stored at -20°C until use and dissolved in DMSO to a stock concentration of 10 mM.

Animal models
Mouse models of renal fibrosis were constructed, including ADR, UIRI, UUO and 5/6NX, as described previously [3,4]. Male BALB/c or CD-1 mice (weighing 20-25g) were obtained from Southern Medical University Animal Center, Guangzhou, China, and housed in the standard environment with regular light/dark cycles and free access to water and chow diet.
For ADR model, BALB/c mice were administered a single intravenous injection of ADR (doxorubicin hydrochloride; Sigma, St. Louis, MO) at 11 mg/kg body weight.
For UIRI model, BALB/c mice were subjected to unilateral ischemia-reperfusion (IRI) surgery by an established protocol as described previously [3]. Briefly, left renal pedicles were clipped for 35 minutes using microaneurysm clamps (item no.18051-35; Fine Science Tools, Cambridge, UK) for IRI injury. After removing clamps, reperfusion was visually confirmed. During the ischemic period, body temperature was maintained at approximately 37°C~38°C using a temperature-controlled heating system. Mice were randomly assigned to one of three groups (n = 5): 1) sham control, 2) UIRI + vehicle control, and 3) UIRI + PA-S14 (1.0 mg/kg/d). At 4 days after IRI, mice were subjected to daily intraperitoneal injections of PA-S14 for 7 days. After 10 days post-IRI, the intact right kidney was removed via a right flank incision. Mice were sacrificed at 11 days post-IRI, respectively.
For UUO model, BALB/c mice were carried out by double-ligating the left ureter following a midline abdominal incision. Mice were randomly assigned to one of three groups (n = 5): 1) sham control, 2) UUO + vehicle control, and 3) UUO + PA-S14 (1.0 mg/kg body weight). Another set of experiments was performed, in which PA-S14 was given for 3 times (2, 4, 6 days after UUO) via intraperitoneal injections at 1.0 mg/kg body weight. Mice were euthanized at 7 days after UUO.
For the 5/6NX model, male CD-1 mice were randomly divided into three group(n = 5)：1) sham control, 2) 5/6NX + vehicle control, and 3) 5/6NX + PA-S14 (1.0 mg/kg body weight). Briefly, the general operation is as follows. The upper and lower poles of the left kidney, two thirds of the left kidney, were removed. The remnant left kidney was immediately compressed using gelatin sponge to stop bleeding (about 30s). One week later, the right renal vessels were ligated and the right kidney was removed (week 0). At 4 weeks after operation, mice were subjected to daily intraperitoneal injections of PA-S14 for 2 weeks. All mice were euthanized and sacrificed at week 6.
In some experiments, the healthy BALB/c mice were also treated with PA-S14 at the dosage of 1.0 mg/kg via intraperitoneal injection. Mice were randomly assigned to one of two groups (n = 5): 1) vehicle control mice, 2) PA-S14 mice (1mg/kg body weight). Mice were euthanized at 7 days after PA-S14 injection.
All animal studies were performed adhering to the Health Guide for the Care and Use of Laboratory Animals and approved by the Experimental Animal Committee at the Nanfang Hospital, Southern Medical University (NFYY-2017-0425).

Urinary albumin, serum creatinine and BUN assay
Urinary albumin was measured using a mouse albumin enzyme-linked immunosorbent assay quantitation kit according to the manufacturer's protocol (Bethyl Laboratories, Montgomery, TX). Serum creatinine and BUN levels were determined by an automatic chemistry analyzer (AU480 Chemistry Analyzer, Beckman Coulter, Atlanta, Georgia). Urinary albumin was standardized to urine creatinine and expressed as mg/mg urine creatinine. The levels of serum creatinine and BUN were expressed as mg/dl.

Cell culture and treatment
Human proximal tubular epithelial cells (HKC-8) were provided by Dr. Lorraine C.

mRFP-GFP-LC3B lentivirus infection
Autophagic flux was monitored using the mRFP-GFP-LC3B lentivirus (Hanbio Biotechnology, HB-LP210) based on different pH stabilities of RFP and GFP. HKC-8 cells were infected with lentiviruses expressing a tandem RFP-GFP-LC3B fusion protein and were then treated with TGF-β1 or were cotreated with PA-S14. After different treatments, the cells were observed under the confocal microscope.

Western blot analysis
Protein samples were prepared and loaded onto SDS-PAGE gel to running. When the running finished, protein samples were electrotransfered onto PVDF membrane.
PVDF membrane was then incubated in 5% of milk for one hour at room temperature, and incubated with different primary antibodies overnight at 4℃. The next day, PVDF membrane was incubated with a responding secondary antibody for one hour at room temperature. Chemiluminescent detection was performed. The whole process was performed as described previously [6]. The primary antibodies used were as follows:

Histology and immunohistochemical staining
Paraffin-embedded kidney sections were performed with periodic acid-Schiff (PAS) and Sirius red staining to identify injured tubules and collagen deposition. The lesions were quantified by a computer-aided technique [9]. Immunohistochemical staining was performed as described previously [6]. Antibodies used were as follows: p-LKB1 h. After different treatments, cells were then collected and subjected to analyses.

Transmission Electron Microscopy
To assess mitochondrial morphology and autophagic vacuoles, kidney cortex and HKC-8 cells were collected and fixed in 1.25% glutaraldehyde (0.1 mol/L) in phosphate buffer. Ultrathin sections (60 nm) were prepared by a routine procedure [10] and were examined under an electron microscope (JEOL JEM-1010, Tokyo, Japan).

ATP assay
ATP concentrations were tested using enhanced ATP assay kit (Beyotime Biotechnology, S0027) according to the manufacturer's protocol.

Complex I Activity Assay
Complex I activity was determined using a commercial kit (Abcam, ab109721) according to the procedures specified by the manufacturer.

JC-1 Analysis
The mitochondrial membrane potential was determined by JC-1 staining and operated according to the previous literature [11].

RNA-seq analysis
RNA-seq was performed to characterize the transcriptome of groups of mice treated with UIRI or UIRI/PA-S14. Trizol reagent (Invitrogen, Carlsbad, CA) was used to isolate the total RNA of each sample. RNA integrity was assessed using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2100 system (Agilent Technologies, CA, USA). Then, a total of six cDNA libraries were constructed. PCR products were purified (AMPure XP system) and library quality was assessed on the Agilent immobilized on a bare gold SPRi chip (3D Dextran chip). Then, the chip was incubated in 4 °C overnight with more than 60% humidity. The SPRi chip was washed three times with PBS and blocked by using 1 M ethanolamine aqueous solution (pH=8.5). Samples of PA-S14 (purity >95%) as the analyte were diluted with PBS to concentrations of 0.5, 5, 25, and 50μM. Data were fitted to a 1:1 binding model to obtain KD. All the samples were tested by a PlexArray HT A100 (plexera) system, and data were analyzed using BIAevaluation Software.

Statistical analyses
All data examined were expressed as mean ± SEM. Statistical analysis of the data was carried out using SPSS 20.0 (SPSS Inc.). Comparison between groups was made using one-way ANOVA followed by Student-Newman-Kuels test or Dunnett's T3 procedure. P < 0.05 was considered to represent a significant difference.