Elucidating the role of nicotinamide N-methyltransferase-p53 axis in the progression of chronic kidney disease

Animal model

Eight-week old male C57BL6J mice were purchased from Vitong Lihua (Beijing). Mice were raised in specific pathogen-free (SPF) conditions and housed with ad libitum access to food and water. Mice were randomly divided into four groups and subjected to either sham operation or unilateral ureteral obstruction (UUO) operation. To investigate the role of NNMT in renal fibrosis, mice were administered 0.25 mg/g/d NAM (Sigma-Aldrich, St. Louis, MO, USA) via intraperitoneal injection, starting 3 days before surgery and continuing daily for 7 days post-surgery. The renal fibrosis model was established as described previously (Zhen et al., 2021). The data presented in this manuscript are based on a separate mouse study, repeated under the same conditions as used before. For the time course study of NNMT, animals were sacrificed on days 3, 7, 10 and 14 after UUO. Other groups of mice were euthanized on day 7 post-surgery. The number of mice in each group was five to six and fed in animal cages, and a total of about 42 mice were used. Sample sizes is based on our previous experiments. The method of euthanasia was caused by cervical spine dislocation after intraperitoneal injection of an overdose of 1.5% pentobarbital sodium anesthetic (100 g/ml(w/v)). One of the following condition was used as benevolent endpoint and euthanized animals prior to the planned end of the experiment: animals lose 15–20% of their body weight rapidly, persistent hypothermia, obvious near-death manifestations. Mice were excluded from statistical analysis, that were euthanized prematurely or had large individual differences. All animal experiments were approved by the Ethics Committee for Animal Experiments at the Third Affiliated Hospital of Sun Yat-Sen University.

Cell culture and treatment

The rat renal interstitial fibroblasts (NRK-49F) cell line was obtained from ATCC. NRK-49F cells were cultured in DMEM/F12 medium supplemented with 10% fetal bovine serum (Gibco/Life Technologies, New York, NY, USA). To assess the effect of NNMT on the TGF-β1-associated fibrosis, NRK-49F cells were starved for 12 h and then exposed to 1, 2, 5,10 or 20 ng/mL TGF-β1 (R&D Systems, Minneapolis, MN, USA) for 24 h. To transfect NNMT siRNA into NRK-49F cells, cells at 60–70% confluence were transfected with NNMT or scrambled siRNA using Lipofectamine 2000 for 24 h before stimulation with 10 ng/mL TGF-β1 for an additional 24 h. To investigate the role of NNMT in vitro, different concetration of NAM (1, 5, 10, or 20 mM) were applied for 1 h before stimulation with TGF-β1 (10 ng/mL) for 24 h.

RNA extraction and quantitative real-time PCR

Total RNA was isolated from NRK-49F cells using TRIzol reagent and reverse transcribed according to the manufacturer’s instructions (Vazyme, Nanjing, Jiangsu, China). Real-time PCR was performed to detect the expression of NNMT, CTGF, and GAPDH using SYBR Green Mastermix (Vazyme, Nanjing, Jiangsu, China). The primer sequences used in the experiments were as follows: (1) NNMT (rat) forward 5′-GAATCAGGCTTCACCTCCAA-3′; reverse 5′-TCACACCGTCTAGGCAGAAT-3′; (2) GAPDH (rat) forward 5′-ACCATCTTCCAGGAGCGAGA-3′; reverse 5′-CTCGTGGTTCACACCCATCA-3′; (3) CTGF (rat) forward 5′-CTGACCTAGAGGAAAACATT-3′; reverse 5′-AGAAAGCTCAAACTTGACAG-3′. (4) NNMT (mice) forward 5′-TGTGCAGAAAACGAGATCCTC-3′; reverse 5′-AGTTCTCCTTTTACAGCACCCA-3′; (5) GAPDH (mice) forward 5′-ACTCCACTACGGCAAATTC-3′; reverse 5′-TCTCCATGGTGGTGAAGACA-3′.

Western blot analysis

Cells and kidney tissues were homogenized and electrophoresed as previously described in Li et al. (2019). Specifically, samples were electrotransferred to nitrocellulose membranesafter electrophoresis. The membranes were immunoblotted with primary antibodies, followed by horseradish peroxidase-conjugated or fluorescent-labeled secondary antibodies. An enhanced chemiluminescence detection system or Odyssey scanning was used to visualize blots. Primary antibodies such as anti-NNMT, anti-fibronectin, anti-α-SMA, anti-collagen I, anti-tenascin C, anti-p-p53, anti-p53 and anti-GAPDH used in this experiment were consistent with those described in Zhen et al. (2021). Anti-pan-methylation was purchased from Abcam (Cambridge, UK).

Immunoprecipitation

An immunoprecipitation kit was used to extract total protein from kidney tissues following the standard procedure (Shu et al., 2022). Protein complexes were obtained by incubating tissue lysates with either p53 antibodies or normal IgG overnight at 4 °C. Protein A/G agarose beads were added to the protein lysis for 4 h. Immunoprecipitates were washed in wash buffer for three times before dissolved in buffer containing sodium dodecyl sulfate (Sigma, St. Louis, MO, USA).

Histological and immunohistochemical analyses

Masson-trichrome staining was carried out according to the manufacturer’s protocol using 2 μm paraffin-embedded kidney sections. At least 10 random high-power fields (×400) in the cortex were evaluated to quantitatively assess renal fibrosis. The area was measured using Image Pro-Plus Software, and the average staining positive area (%) was calculated for each microscopic field.

NNMT activation measurement

NNMT activity assay was based on previous studies (Rudolphi et al., 2018) which determined that the NNMT product reacts with ketones to form a fluorescent compound under alkaline conditions. The reaction solution was prepared by combining two parts of 1 mol/L nicotinamide and one part of 100 mmol/L SAM (Sigma-Aldrich, St. Louis, MO, USA). The fluorescence solution was prepared by mixing one part of 2 mol potassium hydroxide and 20% acetophenone in ethanol. A total of 97 µL of kidney tissue homogenate was blended with 3 µL substrate solution in a 96-well plate. Blank samples were prepared without substrate solution. The plate was incubated for 60 min at 37 °C. A series of 1-MNA dilutions (starting from 50 µmol/L) were prepared for the standard curve. After 60 min, 30 µL from each reaction well were transferred to a new black plate pre-filled with the 1-MNA standard, 3 µL substrate solution were transferred to each no-reaction wells. A total of 120 µL fluorescence solution were added to each well, and the plate was incubated away from light at room temperature. After 10 min, 150 µL formic acid were added to each well and incubated for an additional 30 min at room temperature. Fluorescence parameter was set with λex 375 nm and λem 430 nm for 0.5 s. NNMT activity (pmol/min/mg) was calculated by determining the absolute amount of 1-MNA generated in samples, then divided by the reaction time and the absolute amount of sample total protein.

Methylation-specific PCR

MSP was used to evaluate the methylation status of p53 promoter region in kidney tissue. The detail methods were based on previous study (Najafipour et al., 2021). A total of 2 µL of converted DNA with methylated and unmethylated primers was amplified in 20 µL reaction mixture. The MSP condition was: 94 °C for 5 min, followed by 35 cycles at 94 °C for 30 s (denaturation), 53 °C for 30 s (annealing), and 72 °C for 30 s (extension). The methylated primers were Forward: 5′-GGGAACGAGTGTTTAAAGTTAAGC-3′ and Reverse: 5′-AAAAAAATACGAAAAACCTATCGAA-3′, the unmethylated primers were Forward: 5′-GGAATGAGTGTTTAAAGTTAAGTGT-3′ and Reverse: 5′-AAAAAAATACAAAAAACCTATCAAA-3′.

Single-cell RNA-sequencing and data processing

Kidneys from sham and UUO mice were minced into 1-mm pieces on ice and incubated in dissociation buffer, as previously described (Si et al., 2019). Specifically, samples were added into HBSS containing 10% fetal bovine serum on ice for 10 min to stop reaction. The resulting suspension was centrifuged at 500 g for 5 min after being passed through 70-um cell strainer (Falcon). Red blood cell lysate was added to the pellet for 3 min, then centrifuged at 500 g for 5 min and resuspended in phosphate-buffered saline. A hemocytometer with trypan blue was used to count the single-cell the suspension that was produced. About 10,000 cells were loaded to each channel and then partitioned into Gel Beads in emulsion in the Chromium instrument using the Single-Cell 3′ Reagent Kit v3 according to the manufacturer’s protocol (10× Genomics). Library preparation including reverse transcription, barcoding, cDNA amplification, and purification was performed according to Chromium 10× v3 protocols. Libraries were sequenced on the Illumina HiSeq4000 platform.

Raw sequencing data were processed using Cell Ranger Single Cell Software Suite (v3.0.2) (10× Genomics). The Mus musculus genome reference (mm10) was used for reads alignment in accordance with 10× Genomics recommendations and reads were mapped to the reference genome using STAR (Spliced Transcripts Alignment to a Reference) with default setting. The mapped reads with valid barcodes and unique molecular identifiers (UMIs) were used to generate the gene-cell matrix.

The cells were reduced in dimension by principal component analysis (PCA) and then clustered using Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) tool. Cells were classified by using markers as previously described. Downstream analysis was performed in the R statistical software 3.6.3 (R Core Team, 2020; Feng et al., 2022). Quality control (QC) filters were performed using the parameters which has been reported (Park et al., 2018): (1) cells with <200 genes were excluded; and (2) cells with >30% mitochondrial RNA reads were excluded. After QC filters, a total of 12,193 cells from two independent experiments were remained (5,093 cells from healthy and 7,100 cells from UUO mice).

Statistical test

All data were expressed as means ± SD. Two groups’ comparisons were analyzed using unpaired Student’s t-test. For comparisons multiple groups were made using one-way ANOVA. The non-parametric Kruskal–Wallis test was used when data did not meet a normal distribution (SPSS software, version 19.0; SPSS, Inc., Chicago, IL, USA). 95% confidence interval, p < 0.05 was considered as statistically significant.

Increased NNMT expression in the fibrotic kidneys from mice with UUO

We initially investigated renal fibrosis and NNMT expression in the kidneys of mice subjected to UUO for 3, 7, 10, and 14 days. Western blot and real-time PCR results revealed that basal NNMT protein expression was low in the kidney cortices of mice. However, NNMT expression was significantly increased in the kidneys of UUO mice (Figs. 1A–1C), with its expression and renal fibrosis positively correlating with the duration of obstruction (Figs. 1D–1F).

NNMT is primarily expressed in fibroblasts and positively correlates with fibroblast marker gene levels in the kidneys of UUO mice

To further investigate the role of NNMT in renal fibrosis, we conducted single-cell RNA sequencing (scRNA-seq) on UUO mouse kidneys. Figure 2A shows a uniform manifold approximation and projection (UMAP) of unsupervised clusters, and Fig. 2B presents markers for cell classification. An increased infiltration of fibroblasts and immune cells, consistent with UUO characteristics, was observed (Fig. 2C). Notably, scRNA-seq analysis of UUO-7d mouse kidneys revealed a marked increase and specific enrichment of NNMT expression in fibroblasts (Fig. 2D). The scRNA-seq data further supported the Western blot results, indicating a positive correlation between NNMT and fibroblast markers, as well as genes including α-SMA, fibronectin, and collagen I (Figs. 2E–2G). These findings suggest that the increased NNMT expression in the kidney after UUO injury may play a role in the development of renal fibrosis.

NNMT mediates TGF- $\mathbf{\beta }$1-induced renal fibroblast activation in NRK-49F cells

We further investigated the potential role of NNMT in mediating fibroblast activation in vitro. TGF-β1, as a master fibrogenic factor, was used to stimulate injury conditions in NRK-49F cells. Consistent with in vivo experiments, western blot and real time-PCR analysis indicated that the expression of NNMT was dependent on TGF-β1 concentration (Figs. 3A–3C). Western blot analysis of α-SMA, fibronectin (FN), and collagen I demonstrated that matrix deposition was dependent on TGF-β1 concentration (Figs. 3D and 3E). To verify the direct role of NNMT in the activation of renal interstitial fibroblasts and matrix proteins production, NNMT siRNA was transfected either with or without TGF-β1 in NRK-49F cells. Figure 3F showed the transfection efficiency of NNMT in NRK-49F cells. The NNMT siRNA treatment notably attenuated TGF-β1-induced expression of α-SMA and Tenascin C (Fig. 3G), and confirmed by gray-level analysis (Fig. 3H). Additionally, TGF-β1-induced CTGF level were reduced by NNMT interference (Fig. 3I). These data suggest that NNMT is an important mediator of renal fibroblast activation.

Nicotinamide supplementation inhibits NNMT activity and attenuates renal fibrosis in UUO mice

Previous studies have suggested that nicotinamide analogs can inhibit NNMT (Ruf et al., 2018), and our previous study demonstrated that nicotinamide supplementation can decrease NNMT expression (Zhen et al., 2021). To investigate the potential of nicotinamide in slowing CKD progression, mice were treated with daily intraperitoneal injections of nicotinamide for 3 days prior to UUO surgery. Masson and Sirius Red staining revealed that interstitial collagen fiber deposition was reduced following nicotinamide administration (Figs. 4A–4C). As NNMT’s effectiveness relies on its catalytic activity, we also assessed the levels of 1-MNA in UUO-7d mouse kidneys to verify the effect of nicotinamide on NNMT activity. Our results showed that nicotinamide supplementation weakened NNMT activity (Fig. 4D) in the UUO mice model. Our previous study (Zhen et al., 2021) also demonstrated that mRNA expression of NNMT was lessened by nicotinamide supplementation (Fig. 4E). These findings suggest that nicotinamide replenishment can exert anti-fibrotic and anti-NNMT effects in UUO mice.

Impairing NNMT using nicotinamide inhibits activation of renal interstitial fibroblasts in vivo and in vitro

Because the activation of renal interstitial fibroblasts is a key factor in the progression of renal fibrosis, we further investigated the effect of nicotinamide on interstitial fibroblast activation in the UUO model and NRK-49F cells, with a focus on NNMT. Immunoblot analysis revealed that the administration of nicotinamide significantly reduced NNMT expression and also decreased the levels of fibrotic markers, including α-SMA, fibronectin, and collagen I, in obstructed kidneys (Figs. 5A and 5C). To further confirm that NNMT regulates the activation of renal interstitial fibroblasts, we examined the impact of nicotinamide on renal fibroblast activation in vitro. We found that nicotinamide treatment inhibited the TGF-β1-induced increase in NNMT expression and activation of NRK-49F cells (Figs. 5B and 5D). These results suggest that NNMT activation is necessary for promoting renal fibroblast activation, which can be suppressed by nicotinamide treatment.

Nicotinamide inhibits p53 signaling activation in the kidney following UUO injury

To further elucidate the mechanism by which NNMT contributes to pro-fibrotic responses, we utilized the STRING database to identify potential proteins that interact with NNMT. The results indicated a protein-protein interaction between NNMT and p53 (as shown in Fig. 6A). Previous research has demonstrated that nicotinamide supplementation can reduce renal fibrosis by activating the TGF-β/Smad3 signaling pathway (Zhen et al., 2021). Given that p53 can be activated by TGF-β1 and is also thought to be a co-factor in TGF-β1-mediated pro-fibrotic gene transcription, it is possible that NNMT may mediate p53 activation. To test this hypothesis, we assessed the effect of nicotinamide on p53 phosphorylation in the kidney after UUO injury. We found that UUO injury led to increased p53 phosphorylation, which was suppressed by nicotinamide treatment (as demonstrated in Figs. 6B and 6C). Previous studies have shown that NNMT catalyzes N-methylation of pyridine-containing compounds using SAM as a methyl donor, thereby regulating cellular methylation potential to impact various epigenetic processes (van Haren et al., 2016; Babault et al., 2018). Based on this, the researchers investigated whether NNMT could decrease p53 methylation to regulate renal fibrosis. Both immunoprecipitation (as shown in Figs. 6D and 6E) and methylation-specific PCR (as demonstrated in Fig. 6F) indicated that both the p53 protein’s pan-methylation and DNA methylation levels were decreased following UUO injury, but this phenomenon was reversed by NNMT inhibition triggered by nicotinamide treatment. Collectively, these findings suggest that NNMT contributes to renal fibrosis by weakening methylation and increasing p53 activation.