Nicotinamide reduces renal interstitial fibrosis by suppressing tubular injury and inflammation

Abstract Renal interstitial fibrosis is a common pathological feature in progressive kidney diseases currently lacking effective treatment. Nicotinamide (NAM), a member of water‐soluble vitamin B family, was recently suggested to have a therapeutic potential for acute kidney injury (AKI) in mice and humans. The effect of NAM on chronic kidney pathologies, including renal fibrosis, is unknown. Here we have tested the effects of NAM on renal interstitial fibrosis using in vivo and in vitro models. In vivo, unilateral urethral obstruction (UUO) induced renal interstitial fibrosis as indicated Masson trichrome staining and expression of pro‐fibrotic proteins, which was inhibited by NAM. In UUO, NAM suppressed tubular atrophy and apoptosis. In addition, NAM suppressed UUO‐associated T cell and macrophage infiltration and induction of pro‐inflammatory cytokines, such as TNF‐α and IL‐1β. In cultured mouse proximal tubule cells, NAM blocked TGF–β‐induced expression of fibrotic proteins, while it marginally suppressed the morphological changes induced by TGF‐β. NAM also suppressed the expression of pro‐inflammatory cytokines (eg MCP‐1 and IL‐1β) during TGF‐β treatment of these cells. Collectively, the results demonstrate an anti‐fibrotic effect of NAM in kidneys, which may involve the suppression of tubular injury and inflammation.

been used as a dietary supplement. 9 In addition, therapeutic effects of NAM have been reported in a range of diseases from pellagra, sepsis, to type I diabetes and fatty liver. 10,11 Mechanistically, NAM may modulate fatty acid metabolism, inflammation, oxidative stress, cell proliferation and apoptosis. In kidneys, Tran et al showed that supplementation of NAM may reverse established ischaemic acute kidney injury (AKI). 13 Remarkably, Poyan et al further demonstrated the therapeutic potential in human AKI patients. 14 However, the effect of NAM in chronic kidney disease and related renal interstitial fibrosis, has not been reported. In this study, we conducted in vivo and in vitro experiments to examine the effects of NAM on the development of chronic renal pathologies, including renal interstitial fibrosis.

| Antibodies and reagents
Following primary antibodies were used in this study: anti-α-SMA

| Mouse model of unilateral urethral obstruction
Male C57BL/6 mice (8 weeks to 10 weeks) were purchased from Hunan Slack King Experimental Animal Company (Changsha, China). Unilateral ureteral obstruction surgery was performed by the procedure described before. 15,16 Briefly, mice were anaesthetized with pentobarbital (60 mg/kg) and mouse body temperature was maintained at approximately 36.5°C by a rectal probe of the Homeothermic Blanket Control Unit (507220F, Harvard Apparatus).
A small incision was made on the left side of the mouse to expose and separate the ureter. In the middle of left ureter, 2 points were sutured with 4-0 silk thread. The control mice only exposed the left ureter to separate without ligation. For NAM treatment, mice were intraperitoneally injected with different doses of NAM (200 mg/kg, 400 mg/kg, 800 mg/kg) one hour before unilateral urethral obstruction (UUO) surgery, and then injected daily at the same point in time until the day before being sacrificed. The mice were sacrificed at 14 days after UUO surgery and their left kidneys were harvested for biochemical and morphological examinations.

| Cells and TGF-β treatment
The Boston University mouse proximal tubular cell line (BUMPT) was originally provided by Dr Wilfred Lieberthal (Boston University School of Medicine). The cells were cultured in DMEM medium containing 10% foetal bovine serum. Briefly, BUMPT cells were uniformly plated in a density of 0.3 × 10 6 cells/dish in a 35mm dish and starved overnight in serum-free DMEM media. The cells were then incubated with 5 ng/mL TGF-β for 48 hours in serum-free DMEM medium, while control cells were maintained in serum-free medium without TGF-β. To test the effect of NAM, 25 µmol/L of NAM was added. After 48 hours treatment, cell morphology was monitored and cells were harvested for biochemical analysis.

| Apoptosis detection in BUMPT cells and kidney tissues
Apoptosis in BUMPT cells and kidney tissues were determined by two methods. First, apoptosis of cells and kidney tissues were determined by Western blot analysis of cleaved-caspase3 as recently described. 17,18 Second, morphological analysis was used to analyse apoptosis in cultured cells. The cells were stained with Hoechst 33342 (Molecular Probes, H1399) and the cells which showed typical apoptotic morphological features, including nuclear condensation and chromatin fragmentation, were counted to determine the percentage of apoptosis. TUNEL staining was performed in kidney tissues according to the protocol of the Cell Death Detection kit (Roche Applied Science, 12156792910) as described in our previous work. 17,18 For quantification, 10 to 20 fields were selected randomly from each tissue section to count TUNEL-positive cells/mm 2 .

| Histological analysis of kidney tissues
Kidney tissue was fixed in 4% paraformaldehyde solution, embedded in paraffin and sectioned at 4µm. Haemotoxylin and eosin (H&E) staining and Masson trichrome staining were performed according to the standard protocols provided by the manufacturer (Servicebio). The atrophic renal tubules were characterized by thinning of tubular cell body, dilation of the tubule, necrotic debris in the lumen, and significant expansion of the interstitial space. Tubular atrophy score was calculated by using a scoring system based on the percentage of atrophic tubules (0; 1, <25%; 2, 25% to 50%; 3, 50%-75%; 4, >75%). The type I, type III and type IV collagen fibrils were stained aniline blue by Masson staining. To quantify the fibrotic area, 10 to 20 stained areas (magnification 100×) were randomly selected. ImageProPlus software was then used to assess the ratio of positively (blue) stained area to the entire area (excluding glomerular, small vena cava and blood vessels), which was expressed as the percentage of fibrotic area.

| Immunohistochemical analysis of T cells and macrophages
Paraffin-embedded tissue sections were deparaffinized, rehydrated and placed in 0.1 mol/L sodium citrate (pH 6.0). Then the paraffin sections were heated in an oven for antigen retrieval. These sections were incubated in 3% hydrogen peroxide and blocked in 2% normal goat serum. The sections were then exposed to 1:250 anti-CD3 or 1:1000 anti-F4/80 antibodies overnight at 4°C, followed by rewarming at room temperature and exposure to biotinylated

| Western blot analysis
Cells and kidney tissues were lysed in a lysis buffer containing 2% SDS and protease inhibitor cocktail (Sigma-Aldrich, P8340). The lysates were assayed for protein concentration by the BCA Protein Assay Kit (Thermo Scientific, 23225) and then loaded for reduced SDS-gel electrophoresis by standard procedures. The proteins were then transferred onto polyvinylidene difluoride membranes. These membranes The ratio of FN/GAPDH or α-SMA/GAPDH of sham control was arbitrarily set as 1, and the ratios of other groups were normalized with control to determine fold changes. These data were expressed as mean ± SD (n = 3). *P < 0.05, ***P < 0.001, significantly different from the sham group (with 0 NAM), # P < 0.05, ## P < 0.01, significantly different from UUO (with 0 NAM) were incubated with blocking buffer containing 5% bovine serum albumin and then exposed to specific primary antibodies overnight at 4°C. Finally, the membranes were incubated with secondary antibodies to reveal signal with the ECL kit (EMD Millipore, WBKLS0500).

| Real-time RT-PCR
For quantitative analysis of gene expression, total RNA was isolated using the Trizol kit (12183555, Invitrogen,). RNA concentration was measured by NanoDrop2000 and cDNA was synthesized by using the Takara kit (RR047A, Takara). RT-PCR was performed according to the standard procedure of the Takara kit (RR820A, Takara). All data were analysed by the LightCycler ® 96 SW 1.1 software, the threshold period (Ct) was determined and the mRNA expression of the specific gene in the target sample was calculated by Ct values. Primer sequences for the specific genes used in this study are listed in Table 1.

| Statistics
Qualitative data including Western blots and various morphological images were representatives of at least three experiments.
Quantitative data were expressed as means ± SD (standard deviation) and statistical analysis were performed using GraphPad Prism software. The 2-tailed unpaired or paired Student's t tests were used to determine the statistical difference between two groups. ANOVA followed by Tukey's post hoc tests were used to determine statistical differences between multiple groups. The value of P < 0.05 was considered significantly different.

| NAM reduces UUO-induced renal interstitial fibrosis
We examined the effect of NAM on renal fibrosis during UUO.
Different doses of NAM were given to mice with UUO. After 14 days, the mice were sacrificed and the obstructed kidney was taken for morphological and biochemical analysis. We first used Masson trichrome staining to detect collagen deposition in tissues of the obstructed kidneys. As shown in Figure 1A, sham control kidneys had minimal staining regardless of NAM. UUO induced significant increases in positive Masson staining (blue), which were notably reduced by 200-800 mg/kg NAM. We further quantified the fibrosis staining by morphometric analyses using ImageProPlus. After two weeks of UUO, there was about 15% of renal tissue area stained positive, which was reduced to about 7% by NAM ( Figure 1B).
F I G U R E 2 Nicotinamide (NAM) reduces unilateral urethral obstruction (UUO)-induced tubular atrophy. C57BL/6 mice were subjected to UUO surgery or sham operation. Different doses of NAM or saline were intraperitoneally injected an hour before the surgery and daily thereafter. The mice were sacrificed at 14 days after surgery to collect obstructed kidneys for haematoxylin and eosin histological staining.
The immunoblot results showed that both FN and α-SMA accumulated dramatically in the obstructed kidney tissues, and this accumulation was significantly suppressed by the administration of NAM ( Figure 1C). Semi-quantification by densitometry further verified this conclusion ( Figure 1D). Interestingly, the effect of NAM on renal fibrosis did not show a good dose-dependence, suggesting the possibility of reaching the maximal effect at 200 mg/kg. These results suggest that NAM has inhibitory effects on renal interstitial fibrosis. The administration of NAM partially reduced the amount of atrophic tubules ( Figure 2A). This result was confirmed by statistical analysis of tubular atrophy score ( Figure 2B). Thus, NAM can help reduce UUO-induced tubular atrophy.

| NAM suppresses UUO-induced tubular cell apoptosis
Tubular apoptosis contributes significantly to the progressive cell loss and renal degeneration in renal interstitial fibrosis. It is also closely related to tubular atrophy and particularly its progression to tubular atresia. 2,3,23 To determine the effect of NAM on tubular apoptosis in UUO, we first performed TUNEL (terminal deoxynucleotidyl transferase-mediated dUTP nick end labelling) staining in kidney tissues. As shown in Figure 3A Figure 3D). Together, these results indicate that NAM reduces apoptosis in UUO.

| UUO-induced renal inflammation is alleviated by NAM
Inflammation is another important factor for the initiation and progression of renal interstitial fibrosis diseases. 2,3,5,24 In this aspect, infiltration and phenotypic change of macrophages play a critical role. 25,26 To examine the role of NAM in UUO-associated inflammation, we examined the infiltration of T cells and macrophages in kidney tissues. We further analysed the expression of pro-inflammatory cytokines, including tumour necrosis factor alpha (TNF-α) and interleukin-1β (IL-1β). As shown in Figure

| NAM attenuates TGF-β-induced pro-fibrotic changes in cultured proximal tubular cells
TGF-β is a key factor in renal interstitial fibrosis that may affect multiple cell types including renal proximal tubules. 28

| NAM inhibits TGF-β-induced apoptosis in proximal tubular cells
We showed the inhibitory effect of NAM on renal tubular apoptosis during UUO in mice (Figure 3). In vitro, TGF-β treatment induced a marginal level of apoptosis in BUMPT cells. We recorded cellular and nuclear morphologies following Hoechst staining, which revealed apoptotic cells with cellular and nuclear condensation and fragmentation ( Figure 6A). In counting, there were about 20% apoptotic cells after TGF-β treatment for 48 hours. Interestingly, the addition of NAM significantly decreased apoptosis during TGF-β treatment ( Figure 6A,B). This morphological finding was verified by Western blot analysis of cleaved caspase 3. As shown in Figure 6C,D, TGF-β increased the expression of cleaved caspase3, which was reduced by 25 μmol/L NAM. Thus, both in vivo and in vitro data support the conclusion that NAM reduces renal tubular apoptosis in renal fibrosis.

| Inflammatory response of BUMPT cells induced by TGF-β is inhibited by NAM
The inflammatory response of renal tubular epithelial cells is a key point to the development of renal interstitial fibrosis. 2,3,5,23 To determine the effect of NAM on the inflammatory response of renal tubular cells, we analysed the expression of monocyte chemoattractant protein 1 (MCP-1) and interlukin-1 beta (IL-1β). BUMPT cells were treated with TGF-β with or without 25 μmol/L NAM for 48 hours to extract RNA for real-time PCR analysis. As shown in Figure 7A,B, TGF-β induced significantly increases in MCP-1 and IL-1 mRNA expression, which was partially yet significantly reduced by NAM. The results indicate that NAM may directly suppress the pro-inflammatory response in renal tubular cells during renal interstitial fibrosis.

| D ISCUSS I ON
Renal interstitial fibrosis is a pathological hallmark of progressive kidney diseases in almost all etiologies. At present, there is an urgent How does NAM inhibit renal interstitial fibrosis? With this question, we examined the effects of NAM on tubular injury and inflammation. During the development of renal interstitial fibrosis, tubular injury is characterized by tubular degeneration, including tubular cell death as well as tubular atrophy and atresia. 2,3,23,33 In our experiments, UUO induced atrophic tubules in mouse kidneys that was apparently reduced by NAM ( Figure 2). In addition, NAM significantly reduced renal apoptosis during UUO, as shown by TUNEL assay and immunoblot analysis of active/cleaved caspase 3 ( Figure 3).
Moreover, TGF-β-induced apoptosis in cultured proximal tubular cells was also decreased in NAM ( Figure 6). These observations suggest that NAM may antagonize renal fibrosis by reducing tubular injury and/degeneration, including tubular atrophy and apoptosis.
Chronic or persistent inflammation is another important factor driving the progression of renal fibrosis. 2,3,5,24,25 In our study, NAM showed significant anti-inflammation effects. We specifically examined macrophage and T cell infiltration in renal tissues during UUO, which was significantly suppressed by NAM. In addition, NAM reduced the expression of pro-inflammatory cytokines, such as TNF-α and IL-1β ( Figure 3). Interestingly, cultured tubular cells produced pro-inflammatory cytokines in response to TGF-β1 treatment, and this production was also suppressed by NAM (Figure 7). These results, together, suggest that NAM may antagonize renal interstitial fibrosis by attenuating inflammation.
While our study demonstrates the effects of NAM on both tubular injury and inflammation, we postulate its primary or proximal action on renal tubular cells. In UUO, urine flow is blocked, resulting in the backup of urine in the obstructed kidney, which leads to tubular injury and cell death, inflammation and disruption of renal haemodynamics for fibrogenesis. 34 Under this and other related conditions, tubular changes are considered the driving force for the development of renal interstitial fibrosis. It is known that injured or regenerating tubular cells may change to a secretory phenotype for the production and release of pro-fibrotic factors that not only stimulate the activation and expansion of interstitial resident fibroblasts but also trigger inflammation. As such, amelioration of tubular injury or stress may have enormous effects on the occurrence and progression of renal interstitial fibrosis. The recent studies by Parikh and colleagues have demonstrated a critical role of NAD biosynthesis in maintaining renal tubular cell viability and function in AKI in both mice and humans. 13,14 In cells, NAM is a key precursor for NAD biosynthesis. Our current study indicates that supplementation of NAM can reduce UUO-associated tubular atrophy and apoptosis, further supporting the role of NAM-mediated NAD biosynthesis in cellular homeostasis and viability of renal tubules. By reducing tubular injury and stress, NAM may suppress inflammation and interstitial fibroblast activation.
It remains unclear how NAM protects renal tubular cells. As mentioned above, NAM is a key precursor for the biosynthesis of NAD+, a critical coenzyme and electron acceptor in glycolysis and the Krebs cycle. It is therefore possible that NAM may help the cells maintain their bioenergetics status for survival during injury and stress. In addition, NAD + is a substrate for several non-redox enzymes including poly ADP-ribose polymerase (PARP) and sirtuins. Interestingly, these enzymes have been implicated in kidney injury. For example, PARP-1 has been reported to inhibit glycolysis during ischaemic kidney injury by poly(ADP-ribosyl)ation of the key glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH) 35 and inhibition of PARP-1 can protect against ischaemic kidney injury by preserving tubular cell ATP. 36 NAM has been reported to have protective effects in a rat model of Alzheimer's disease by inhibiting PARP-1. 37 Sirtuins are a family of NAD-dependent deacetylases that play regulatory functions in a variety of tissues and organs, including kidneys. 38 Of much relevance to our current study, NAM was recently shown to reduce ageing-associated susceptibility to AKI in a Sirt1-dependent manner. 39 Future studies should investigate the effects of NAM on cellular bioenergentics, PARP-1 and sirtuins in experimental models of renal fibrosis to gain better understanding of its anti-fibrotic effect.

ACK N OWLED G EM ENTS
The work was supported partly by the grants from National Key R&D Program of China (2018YFC1312700) and National Natural Science Foundation of China (81720108008).

CO N FLI C T O F I NTE R E S T
The authors declare no competing financial interests in relation to the work described in this study.