Carbon Monoxide (CO) Alleviate Diabetic Nephropathy (DN) Effectively
Mar 16, 2022
for more information:ali.ma@wecistanche.com
Part Ⅰ： Carbon monoxide alleviates senescence in diabetic nephropathy by improving autophagy
Guibin Meil; Chunjie Jiang; Xueer Cheng; & at al.V
INTRODUCTION
Diabetic nephropathy (DN) is one of the most common and serious diabetic microvascular complications. Statistically,20%-40% of patients with DN (Diabetic nephropathy) will develop into end-stage renal disease (ESRD) at a rate of 14 times faster than other renal diseases. Additionally, the increased cardiovascular risk of DN (Diabetic nephropathy) patients greatly contributes to high mortality.2The complex pathogenesis of DN (Diabetic nephropathy) is still unclear with the limited treatment approach available. Hence, it is urgent to identify the pathogenesis and therapeutic agents to prevent the progressive loss of renal function of DN (Diabetic nephropathy).
Senescent cells, characterized by permanent cell cycle arrest, have attracted extensive attention as 'zombie cells. Due to resistance to apoptosis and the continued production of senescence-associated secretory phenotype (SASP), senescent cells are increasingly recognized as the crucial cause of age-related diseases. Senescence-linked renal dysfunction has been commonly observed in DN (Diabetic nephropathy) patients and experimental models, such as STZ induced (type 1 diabetes (T1D))and db/db(type 2 diabetes(T2D))mice. Importantly, the clearance of p16-positive cells of aged mice improved glomerulosclerosis. Furthermore, knocking out p21 or p27 in T1D mice relieved proteinuria and glomerular dilation. These results suggested that senescence plays an important role in the pathogenesis of DN (Diabetic nephropathy). However, the regulation of senescence in DN (Diabetic nephropathy) remains vague.
Emerging evidence showed that autophagy, a mechanism of the degradation of intracellular contents, is a valid senescence regulator." Although it has been reported that autophagy promotes the acquisition of senescent phenotype, the anti-senescence effect of autophagy is more widely accepted. Upregulation of autophagy was reported to extend the lifespan of aged mice and elder flies.10 Further studies showed that knocking out Atg7 aggravated stem cell senescence.11 More importantly, the deletion of Atg5 aggravated proteinuria and fibrosis the main performance of kidney aging.12.13 Thus, drugs targeting autophagy to alleviate senescence may be one of the most promising strategies for DN (Diabetic nephropathy) prevention and treatment.
Carbon monoxide (CO) generated via the catabolism of haem by haem oxygenase enzymes is an endogenously gaseous molecule. A number of publications revealed the anti-inflammatory,anti-apoptotic and other protective properties of CO (Carbon monoxide) when applied at low doses.14 Although CO (Carbon monoxide) has been well-studied to confer renoprotection to ischemia-reperfusion or kidney transplantation mice,15 the effect and mechanism of CO (Carbon monoxide) on DN (Diabetic nephropathy) are unclear. Recently, the literature showed that CO (Carbon monoxide) imparted cytoprotective roles as an autophagy activator in islets challenged by hypoxia,16 sepsis mice7and aged rats with cardiac arrest.18 Moreover, the administration of CO (Carbon monoxide) alleviated the senescence of endothelial cells caused by drug toxicity.1These findings apparently advance the potential mediation of autophagy targeted by CO (Carbon monoxide) to ameliorate senescence.
Hence, we hypothesized that the accumulation of senescent cells in the kidney of DN (Diabetic nephropathy) mice could be reversed by CO (Carbon monoxide) via autophagy activation, subsequently improving renal dysfunction. To test our hypothesis and explore the underlying mechanisms, we applied the treatment of carbon monoxide releasing molecule-2 (CORM-2) in vivo(the experimental DN (Diabetic nephropathy) mice) and in vitro (rat mesangial cells, human tubular epithelial cells, and human podocyte).
Kidney function: how Carbon monoxide (CO) affects Diabetic nephropathy (DN)
Click to Cistanches and Cistanche products for kidney disease
MATERIALS AND METHODS
2.1 | Animal experimental design
Eight-week-old male C57BL/6J mice(Charles River)were kept in a standard light/dark cycle (12:12 hours) with a normal diet(ND) or high-fat diet (HFD)(60% energy from fat) for 16 weeks.20 At week 17 of the study, mice fed with HFD received intraperitoneal injections of STZ (50 mg/kg) for 7 days, while mice with ND were injected with vehicle (citrate buffer, pH=4.5). One week after the injections, blood glucose levels of a tail prick were measured twice at a 24 hours interval. Next, the DN (Diabetic nephropathy) mice with glucose levels of 16.9 mmol/L or greater were divided into three groups randomly, with 15 animals in each group (DN (Diabetic nephropathy), DN (Diabetic nephropathy)+CO, and DN (Diabetic nephropathy)+iCO): a group termed as DN control and other groups treated with intraperitoneal injections(twice a week for 16 weeks) of CORM-2 (3 mg/kg)or in-valid CORM-2 which was produced by releasing CO (Carbon monoxide) from CORM-2 at room temperature. At the end of the experimental period, mice were anesthetized, blood samples were collected by enucleation of eyeballs, and kidneys were harvested for analysis. All animals were treated in accordance with the Guiding Principles in the Care and Use of Laboratory Animals published by the US National Institutes of Health, and all animal procedures were approved by the Tongji Medical College Council on Animals Care Committee.
2.2 | Plasma biochemical and histological analysis
The collected whole blood and serum were stored at-80℃ for analysis. Blood urea nitrogen (BUN)was measured with commercially available assay kits(Jiancheng Bioengineering Institute, China). After kidneys were removed and the surface was washed with saline. The kidney tissue was stained with hematoxylin-eosin and Masson's tri-chrome stain and finally observed with a microscope.
2.3 | Electron microscopy
Small pieces of the renal cortex were fixed in glutaraldehyde (2.5%) and embedded in Araldite. The tissue was polymerized cut into ultrathin sections (80-100 nm) using an ultramicrotome (Leica EM UC7). The thin slices on the copper mesh grid were stained and observed under a transmission electron microscope (Tecnai G220 TWIN).
2.4 | Shear wave elastography
At the end of the experimental period, the mice were anesthetized and sent to the ultrasonic laboratory(Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China) for shear wave elastography using a soon scope(Supersonic Imagine). The mean Young's modulus (Eman) was used to measure renal parenchymal stiffness.
2.5 | SA-β-gal staining
Frozen kidney sections and cells seeded in a 6-well plate were used for detection of SA-β-gal activity by a commercial kit (Beyotime Biotechnology)according to the manufacturer's instructions. The blue stain was considered as the accumulation area of senescent cells.
2.6 | Real-time quantitative PCR for mRNA expression
According to the instructions (TaKaRa BIO INC), renal tissue RNA was extracted using the TRlzol reagent. The expression of mRNA was quantified with TB green-based gRT-PCR kit and specific primers (Table 1). Each gene expression was assessed with its own standard curve, and the mRNA level of Gapdh was quantified as an endogenous control.
2.7 | Western blotting and immunoprecipitation
Renal tissue or cell was homogenized and lysed, then quantified by BCA protein assay kit (Beyotime Biotechnology). The protein was separated and subsequently transferred onto the PVDF membranes (Millipore). After being blocked, the membranes were incubated overnight with primary antibodies(Table 2). After washing, the membranes were incubated with a corresponding secondary antibody. The density of each target band was quantified by Image Pro-Plus 6.0 software and normalized to GAPDH as optical density. All sample sizes of animals or cells were greater than or equal to 3.
Lysate of kidney tissue was centrifuged (14 000 g at 4℃C for 30 minutes）， and the supernatant was pre-cleared with 20 μL protein A/G agarose beads for 2 hours and incubated overnight with Bcl-2 or mouse lgG antibody. Immunoprecipitates were washed, re-suspended, boiled, and analyzed by Western blot using anti-Beclin-1 (Cell Signaling Technology) and anti-Bcl-2 (Proteintech) as described earlier.21
TABLE 1 Real-time quantitative PCR primer sequences
2.8 | Immunofluorescence
The renal tissue embedded in OCT media was cut into6-8 μm thick frozen sections. After blocking with 10% normal goat serum, the sections were incubated overnight with primary antibodies. The following day, the tissue was labeled with secondary antibodies. After being rinsed with PBS, the slides were incubated with DAPI and subsequently photographed under a fluorescent microscope (Olympus).
2.9 | Cell culture
HBZY-1 was maintained in DMEM(Gibco) and HK-2 was maintained in DMEM/F12 (Gibco) supplemented with 10% fetal bovine serum,100 U mL-1 penicillin/streptomycin (Gibco) at 37°C in a humidified atmosphere containing 95% air and 5% CO (Carbon monoxide). HPC provided by Professor Chun Zhang from Huazhong University of Science and Technology was maintained in RPMI 1640 (Gibco). HPC was first proliferated at 33℃ and then transferred to 37℃ for differentiation before it can be used in experiments as described earlier.22
2.10 | ELIS Assay
After HPCwas treated differently and cultured for 5 days, half of the supernatant was taken out for detection (released VEGF). The cells were broken using an ultrasonic cell crusher (SONICS), allowing the proteins to be released completely into the remaining supernatant (total VEGF). Then, the VEGF concentrations were measured using a commercial ELISA kit (MEIMIAN). The ratio of released VEGF to total VEGF was the VEGF leakage rate.
2.11 | Gene silencing
Knockdown of Atg7 in HK-2 cells was achieved by using a re-verse siRNA transfection procedure performed in six-well plates. Once grown to 70% confluence, cells were transfected with siRNA or scrambled siRNA(RiboBio)using Lipofectamine RNAiMAX (Invitrogen) according to the manufacturer's protocol. After 48 hours of transfection, the transfection efficiency was verified by Western blot to validate the sequence (ACTCGAGTCTTTCAAGACT).
TABLE 2 All antibody lists
2.12 | Monitor the autophagic flux
The plasmids containing lentiviral vector RFP-GFP-LC3(Genecopoeia)were constructed and then packaged in HEK293T cells according to the manufacturer's instructions (GeneCopoeia). The virus supernatant was collected to infect HK-2 and HBZY-1 cells to monitor autophagy flow. This probe could distinguish autophagosomes (GFP*/RFPt, yellow puncta) and autolysosomes (GFP-/RFPt, red puncta).
2.13 | Data analysis
The data were analyzed using Graph Pad Prism8 and showed as the mean ± SEM. Differences among the groups were determined by a one-way analysis of variance. Significance was set as P<.05.
RESULTS
3.1 | CO (Carbon monoxide) attenuated renal dysfunction of DN (Diabetic nephropathy) mice
To explore the renoprotection of CO (Carbon monoxide), a DN (Diabetic nephropathy) model was induced by HFD andSTZ(Figure 1A). After34 weeks of experimental feeding, blood glucose, weight, the kidney-body ratio, creatinine, and BUN were measured. In the DN (Diabetic nephropathy) group, kidney-body ratio, creatinine, and BUN were increased, whereas the treatment of CO (Carbon monoxide) reversed abnormalities (Table 3). Consistently, the renal morphology of DN (Diabetic nephropathy) mice was extremely disordered with epithelial cells desquamation, vacuolar degeneration, glomerular Bowman's space enlargement, and mesangial expansion (Figure 1B). Furthermore, the ultrastructure of the glomerulus displayed basement membrane thickening and podocyte foot process effacement (Figure 1B). On the contrary, CO (Carbon monoxide) treatment normalized above negative morphological changes (Figure 1B).In addition, fibrillar collagen deposited in the renal cortex(Figure 1C) and renal hardness, measured by shear wave elastography, were both increased in DN (Diabetic nephropathy) mice (Figure 1D). As expected, CO (Carbon monoxide) significantly improved renal dysfunction of DN (Diabetic nephropathy) including fibrosis.
FIGURE 1 CO (Carbon monoxide) attenuated renal dysfunction of DN (Diabetic nephropathy) mice induced by HFD and STZ.
A Schema of experimental design. B, Representative images of kidney tissue stained with H&E (scale bar: 50 μm), PAS (scale bar: 10 μm), and ultrastructural changes in glomerular morphology assessed by transmission electron microscopy (scale bar: 1 μm).
And the statistical data of the ratio of PAS-positive to the glomerular area, GBM thickness, foot process width, and the number of foot processes per mm of GBM (n = 3).
C, Representative images of the renal tissue stained with Masson's trichrome and the statistical data of positive area (n = 3). Scale bar: 25 μm.
D, Shear wave elastography detected with a sono scope and quantification of Young's modulus as the degree of renal fibrosis (n = 7). *P < .05, **P < .01, ***P < .001
TABLE 3 Metabolic data of mice in all groups
3.2 | CO (Carbon monoxide) alleviated senescence in DN (Diabetic nephropathy) mice and renal cells challenged by high glucose
Senescence is increasingly considered the main cause of renal fibrosis, a hallmark of aging.23 To investigate the anti-senescence effect of CO (Carbon monoxide), we performed SA-β-gal staining on the kidney. The increased SA-β-gal* and p16* cells were accumulated throughout the renal cortex in the DN (Diabetic nephropathy) group, while control and CO (Carbon monoxide)-treatment kidneys showed occasional positivity (Figure 2A, B). From the anatomical position of the kidney, the positive expression of p16 was detected in mesangial cells (yellow arrow), renal tubular epithelial cells (red arrow), and podocytes (black arrow). Senescence that occurred in DN (Diabetic nephropathy) mice was further supported by the evident upregulation of classical senescence-associated proteins, including p53,p21, and p16 (Figure 2C). Conversely, the administration of CO (Carbon monoxide) alleviated the above senescent performance.
senescence-related secretory phenotype, the distinctive secretome of senescent cells, was reported to disrupt the microenvironment and facilitate disease progression.2 Hence, mRNA levels of some representative SASP were measured, including I-1β, Il-6, Tgf-β, Tnf-a, Vegf, Icam-1, and Vcam-1. Results showed that the abnormal in-creases of this SASP were averted by CO (Carbon monoxide) intervention(Figure 2D).
Given the suppressive senescence of CO (Carbon monoxide) in vivo, we further explore its anti-senescence effect in vitro model of 3 types of renal cells, including rat mesangial cells(HBZY-1), human tubular epithelial cells(HK-2), and human podocyte (HPC). After exposure to high glucose (HG,35 mmol/L),3 types of cells displayed the augmentation of SA-β-gal with larger morphology at day 7(Figure 2E), day 5, and day 14(Figure 2l), respectively. Notably, the increase in the LDH leakage rate of HBZY-1 indicated that HG-induced cell senescence and damage (Figure 2F). In order to investigate the appropriate dose of CO (Carbon monoxide), HBZY-1 was treated with different concentrations of CORM-2. By measuring the leakage rate of LDH(Figure 2F), EdU-positive cells (Figure 2G), and SA-β-gal (Figure 2H), we found that the concentration of 1 μmol/L had a better intervention effect. Consistently, CO (Carbon monoxide) (1μmol/L）also decreased the positive expression of SA-β-gal in HK-2 and HPC (Figure 2). Taken together, these findings unraveled that CO (Carbon monoxide) played a remarkable anti-senescence role both in vivo and in vitro.
FIGURE 2 CO (Carbon monoxide) alleviated senescence in DN (Diabetic nephropathy) mice and renal cells challenged by HG.
A, Representative image of kidney tissue stained with SA-β-gal and the statistical data of positive area (n = 3). The glomerulus was indicated by the red dotted circle. Scale bar: 25 μm.
B, The positive expression of p16 in the kidney was shown by immunohistochemistry (n = 3). The yellow arrows indicated mesangial cells, red indicated renal tubular epithelial cells and black arrows indicated podocytes. Scale bar: 25 μm.
C, The protein levels of senescence markers determined by immunoblots and densitometric analysis of p53, p21, and p16 (n = 3).
D, Fold changes in expression levels of SASP mRNA (Il-1β, Il-6, Tgf-β, Tnf-α, Vegf, Icam-1, and Vcam-1) (n ≥ 4).
E, HBZY-1 exposed to NG (5 mmol/L) or HG (35 mmol/L) was stained with SA-β-gal from day 0 to day 7 and statistical data of SA-β-gal-positive cells at 5 and 7 d (n = 3). Scale bar: 50 μm.
F, The LDH release rate of HBZY-1 treated with different concentrations of CORM-2 (0.1, 1, 10 and 100 μmol/L) (n = 6).
G, The proportion of EdU-positive HBZY-1 treated with different concentrations of CORM-2 (n = 3).
H, SA-β-gal staining in HBZY-1 treated with different concentrations of CORM-2 (n = 3). Scale bar: 50 μm.
I, Representative images of SA-β-gal of HPC (day 5) and HK-2 (day 14) with the administration of CORM-2 (1μmol/L) and statistical data (n = 3). Scale bar: 50 μm. *P < .05, **P < .01, ***P < .001
3.3 | CO (Carbon monoxide) activated autophagy and improved autophagy flow in vivo and in vitro
A growing body of research revealed the negative regulation of autophagy to senescence, and CO (Carbon monoxide) has been considered as an autophagy activator. To explore the effects of COon autophagy in DN (Diabetic nephropathy), a series of autophagy-related proteins were measured. Compared with diabetic mice, CO (Carbon monoxide) treatment significantly increased the expression of autophagy initiation protein Beclin-1(Figure 3A) and decreased the autophagy substrate p62(Figure 3B), but reduced the ratio of LC3Il toLC3l, the marker of autophagosome(Figure3C). Previous evidence suggested that this contradiction may be due to blocked autophagy flow caused by the dysfunction of the lysosome.25Results showed that lysosome-related proteins Cathepsin B and LAMP2 were increased in the DN (Diabetic nephropathy)+COgroup compared to the DN (Diabetic nephropathy) group (Figure 3D). Moreover, the co-localization analysis of LC3 puncta and LAMP2 supported that CO (Carbon monoxide) markedly reduced the accumulation of autophagosomes and increased the number of autolysosomes(Figure 3E).
Consistently, the dysfunction of autophagy and lysosome was found in vitro, as shown by protein levels and lysosomal probes (Figure 3F, G). Further, HK-2 and HBZY-1 were transfected with lentivirus RFP-GFP-LC3 to monitor autophagy flow. Under the condition of HG, puncta of autophagosomes were increased and autolysosomes were decreased in HBZY-1 and HK-2, which was not further changed with chloroquine (CQ)treatment. CO (Carbon monoxide) increased the number of autolysosomes and decreased autophagosomes, which was inhibited by the addition of CQ(Figure 3H, I).In summary, these results demonstrated that CO (Carbon monoxide) activated autophagy and improved autophagy flow in vivo and in vitro.
3.4 | CO (Carbon monoxide) alleviated senescence through improving autophagy in vitro
To further explore the role of autophagy in the anti-senescence mechanism of CO (Carbon monoxide), a combination of CORM-2 and autophagy inhibitors was used in renal cells exposed to HG. Two types of inhibitors, wortmannin(WORT) to block autophagy initiation and CQ to re-strain autophagy flow, effectively reversed the reduction of SA-β-gal by CO (Carbon monoxide) in 3 types of renal cells(Figure 4A). Consistently, CO (Carbon monoxide) significantly increased the proportion of EdU cells, which was inhibited by two inhibitors in HBZY-1(Figure 4B)and HK-2(Figure 4C).In HBZY-1, the senescence-alleviated effect of CO (Carbon monoxide), shown by the reduced expression of p53,p21, and p16, was absent after the addition of autophagy inhibitors(Figure 4D). Similarly, immunofluorescence analysis showed that autophagy inhibitors blocked the decreases of p53- and p16-positive cells by CO (Carbon monoxide) in HK-2 （Figure 4印） and HPC (Figure 4F). Further, the silence of Atg7 (Figure 4G)blocked the protective effects of CO (Carbon monoxide), shown by increased SA-β-galt cells and de-creased EdU*cells in HK-2(Figure 4H). The aforementioned results suggested that CO (Carbon monoxide) suppressed senescence by improving autophagy.
3.5 | Activated autophagy may degrade SASP in DN (Diabetic nephropathy) mice
We further explored the specific relationship between autophagy and senescence. Research showed that autophagy alleviated senescence by selectively degrading GATA4, which positively regulated NF-KB to release SASP.26In line with this, a large amount of SASP was developed in the DN (Diabetic nephropathy) group (Figure 2D), and the expression and nuclear translocation of p65 were increased (Figure 5A). However, the protein level of GATA4 stayed unchanged, suggesting GATA4 may not mediate the generation of SASP in a high glucose state (Figure 5B). Additionally, studies revealed that TOR-autophagy spatial coupling compartment (TASCC) formed by the combination of mTOR, late autophagosome, and lysosome accelerated senescence by increasing the secretion of SASP.27However, immunofluorescence results showed that there was no triple colocalization of mTOR, LC3, and LAMP2, indicating the failed formation of TASCC in the kidneys (Figure 5C).
As we know, autophagy functions in the degradation of cytosolic constituents, which is mainly performed by autolysosome formed by the fusion of autophagosome and lysosome. Autophagy has been reported to reduce the secretion of IL-1β by degrading pro-IL-1β in LPS-stimulated macrophages.28 Interestingly, we found that IL-1β(green) was co-located with LC3 puncta (red) and LAMP2 (pink) in the CON and DN (Diabetic nephropathy)+ CO (Carbon monoxide) groups (Figure 5D, the red arrow), indicating the intersection between SASP and autophagy degradation. Similarly, in comparison with the DN (Diabetic nephropathy) group, other cytokines of SASP including IL-6, TGF-β and VEGF also had modest but evident co-localization with LC3 puncta and LAMP2 in the CON and DN+ CO (Carbon monoxide) groups(Figure 5E-G, the red arrow), among which VEGF co-localized with autolysosomes most obviously(Figure 5G). However, the overlapping yellow staining in the DN (Diabetic nephropathy) group showed SASP was only co-located with LC3 puncta(Figure 5D-G, the white arrow). Furthermore, the protein level of VEGF was measured in HPC after adding autophagy agonists (ABT737 and rapamycin) or CORM-2. Although there was no significant difference in the total protein level of VEGF (Figure 5H), the presence of both autophagy agonists and CO (Carbon monoxide) markedly reduced the release of VEGF in HPC induced by HG (Figure 5I, J). Taken together, these results indicated that some SASP is degraded via autophagy.
FIGURE 4 CO (Carbon monoxide) alleviated senescence through improving autophagy in vitro. HBZY-1, HPC, and HK-2 were treated with HG, CORM-2, and autophagy inhibitors (WORT: 200nM and CQ: 5 μmol/L).
A, The staining and positive cells of SA-β-gal in 3 kinds of renal cells (n = 3). Scale bar: 25 μm.
B, C, Cell proliferation assessed by EdU staining in HBZY-1 and HK-2 and quantification of the proportion of positive cells (n = 3). Scale bar: 50 μm.
D, Expression of senescence-related protein (p53, p16 and p21) in HBZY-1 (n = 5).
E, F, Representative immunofluorescence images of p53 and p16 and quantification of the proportion of positive cells in HK-2 and HPC (n = 3). Scale bar: 50 μm.
G, The protein expression levels of Atg7 in HK-2 silenced with siRNA (three sequences).
H, Representative images of SA-β-gal (Scale bar: 50 μm) and EdU staining (Scale bar: 25 μm), and quantification of the proportion of positive cells in HK-2 (n = 3). *P < .05, **P < .01, ***P < .001
3.6 | CO (Carbon monoxide) activated autophagy partly by dissociating Beclin-1-Bcl-2 complex
The dissociation of the Beclin-1-Bcl-2 complex has been reported to improve autophagy as well as prevent premature aging including age-related renal changes.21 Additionally, CO (Carbon monoxide) activated autophagy through Beclin-1 in the sepsis mice.1/Hence, we explored whether dissociating the complex-mediated the activation of autophagy by CO (Carbon monoxide). Co-immunoprecipitation results showed that CO (Carbon monoxide) reduced Beclin-1-Bcl-2 binding in DN (Diabetic nephropathy) mice (Figure6A). Furthermore, compared with the treatment of HG and CO (Carbon monoxide), the combined use of HG, CO (Carbon monoxide), and the Beclin-1-Bcl-2 complex dissociation agent ABT737 further activated autophagy of HBZY-1, as evidenced by unchanged LAMP2, decreased p62, and increased LC3ll/LC3l ratio (Figure 6B). Moreover, the results showed that the combined use of CO (Carbon monoxide) and ABT737 further down-regulated the expression of p53,p21, and p16 in HBZY-1 (Figure 6C)and decreased the SA-β-galt cell in HBZY-1, HK-2, and HPC (Figure 6D), indicating that ABT737 enhanced the anti-senescence effect of CO (Carbon monoxide). These observations suggested that CO (Carbon monoxide) could activate autophagy by dissociating Beclin-1-Bcl-2 complex to alleviate senescence.
