Posttranslational modifications of epigenetic modifiers provide a flexible and timely mechanism for rapid adaptations to the dynamic environment of cancer cells. SIRT1 is an NAD+-dependent epigenetic modifier whose activity is classically associated with healthy aging and longevity, but its function in cancer is not well understood. Here, we reveal that 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3, calcitriol), the active metabolite of vitamin D (VD), promotes SIRT1 activation through auto-deacetylation in human colon carcinoma cells, and identify lysine 610 as an essential driver of SIRT1 activity. Remarkably, our data show that the post-translational control of SIRT1 activity mediates the antiproliferative action of 1,25(OH)2D3. This effect is reproduced by the SIRT1 activator SRT1720, suggesting that SIRT1 activators may offer new therapeutic possibilities for colon cancer patients who are VD deficient or unresponsive. Moreover, this might be extrapolated to inflammation and other VD deficiency-associated and highly prevalent diseases in which SIRT1 plays a prominent role.
This is an important study demonstrating that vitamin D-bound VDR both increases SIRT1 expression and directly interacts with SIRT1 to cause auto-deacetylation on Lys610, thereby regulating activation of SIRT1 catalytic activity. The data presented to support the presented conclusions are convincing, and the findings are particularly relevant to the actions of VDR on colorectal cancer.
Over the last decades the study of genetic lesions in tumour cells provided key insights into the deregulated signalling pathways that promote cell proliferation and invasiveness and suppress senescence. In addition, increasing evidence from epidemiological studies suggests a significant impact of non-genetic modifiable factors on the oncogenic process by affecting cell epigenetics and thus, gene expression and phenotype. The key metabolic sensor SIRT1 acts an NAD+-dependent deacetylase and a critical post-translational modifier and epigenetic regulator. As such, SIRT1 promotes efficient energy utilisation and cellular defences in response to environmental challenges, whereas its dysregulation accelerates age-related diseases such as diabetes and cancer (Bonkowski and Sinclair, 2016).
Colorectal cancer (CRC) is the second most diagnosed malignancy in women, the third in men, and a major leading cause of cancer-related deaths worldwide, with an incidence estimated to increase by 60% by 2030 (Bray et al., 2018Cancer Genome Atlas Network, 2012). Many observational/epidemiological studies suggest that vitamin D (VD) deficiency is a risk factor for developing and dying of cancer, particularly for CRC (Feldman et al., 2014Grant et al., 2022Kim et al., 20222021), albeit data from supplementation studies in the human population are controversial. Confirmation of a clinically relevant anti-CRC effect of VD in well-designed prospective randomized studies is still pending (Chandler et al., 2020Henn et al., 2022Manson et al., 2020Muñoz and Grant, 2022Song et al., 2021)
VD in humans has two origins: synthesis in the skin by the action of solar ultraviolet radiation on 7-dehydrocholesterol and the diet. The active VD metabolite is 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3, calcitriol), which results from two consecutive hydroxylations of VD, the first in the liver and the second in the kidney or in many epithelial and immune cell types (Bikle and Christakos, 2020Feldman et al., 2014). VD actions are mediated by 1,25(OH)2D3 binding to a member of the nuclear receptor superfamily of transcription factors, the VD receptor (VDR). Upon ligand binding, VDR regulates the expression of hundreds of target genes, many of which are involved in calcium and phosphate homeostasis, bone biology, immune response, metabolism, detoxification, and cell survival, proliferation and differentiation (Bikle and Christakos, 2020Carlberg and Muñoz, 2022Carlberg and Velleuer, 2022Feldman et al., 2014Fernández-Barral et al., 2020).
Consistent with a protective role of VD against CRC, the colonic epithelium expresses high levels of VDR. Many studies in CRC cells and associated fibroblasts and in experimental animals have shown a large series of VDR-mediated VD anti-CRC effects (Carlberg and Muñoz, 2022Carlberg and Velleuer, 2022Feldman et al., 2014Fernández-Barral et al., 2020Ferrer-Mayorga et al., 2019) In support, germline deletion of Vdr in mice with constitutively active Wnt/β-catenin signalling, as a main driver of CRC, results in increased intestinal tumour load (Larriba et al., 2011Zheng et al., 2012).
Notably, certain evidence links SIRT1 and VD. First, decreased SIRT1 activity has been linked to the pathogenesis of CRC (Ren et al., 2017Strycharz et al., 2018). However, using SIRT1 level as a marker of malignancy is controversial since both increased or decreased SIRT1 tumour expression have been described in different studies (Carafa et al., 2019Chen et al., 2014Firestein et al., 2008), which points to a discrepancy between SIRT1 level and activity. Second, SIRT1 deacetylates the VDR to enhance its activity in kidney and bone cells (Sabir et al., 2017) and, consequently, decreased SIRT1 activity may drive VD insensitivity. Third, promoter-bound VDR increases SIRT1 gene expression in kidney and liver cells (Yuan et al., 2022) and thus, VD deficiency may decrease SIRT1 levels leading to reduced activity. However, whether VD alters SIRT1 activity per se and/or SIRT1 protein expression in CRC remains unclear.
Here we reveal that 1,25(OH)2D3 induces specific SIRT1 activity at multiple levels in human CRC cells. Notably, liganded VDR post-translationally increases SIRT1 activity through auto-deacetylation. Mechanistically, SIRT1 K610 is identified as a critical player for 1,25(OH)2D3-enhanced VDR-SIRT1 interaction, SIRT1 auto-deacetylation and activation. This occurs independently of increased SIRT1 RNA and protein expression and elevated cofactor (NAD+) levels. Discrepant SIRT1 level and activity may explain the controversial role of SIRT1 as tumor suppressor or tumor promoter in CRC. Our results provide a new mechanistic insight into the association of VD deficiency with CRC and suggest potential therapeutic benefit for SIRT1 activators in SIRT1-positive CRC.
1,25(OH)2D3 increases SIRT1 level via the VDR in CRC cells
The question of whether VD governs SIRT1 activity in CRC was first approached in two well characterized CRC cell lines, HCT 116 and HT-29. Immunofluorescence analyses (Fig. 1A) revealed that SIRT1 protein level doubled in response to 1,25(OH)2D3, a result confirmed by Western blotting of nuclear extracts (Fig. 1B). The increased expression of SIRT1 protein by 1,25(OH)2D3 could be explained by a 2-fold upregulation of SIRT1 RNA revealed by RT-qPCR (Fig.1C). The induction of CYP24A1 (Fig. 1D), a major VDR target gene in many cell types served as control for 1,25(OH)2D3 activity.
VDR expression is the main determinant of cell responsiveness to 1,25(OH)2D3 and is downregulated in a proportion of patients with advanced CRC, implying that these patients would probably not benefit from the anticancer effects of 1,25(OH)2D3 (Ferrer-Mayorga et al., 2017Larriba and Muñoz, 2005Pálmer et al., 2004Peña et al., 2005). To model CRCs patients with low VDR tumour level or with VD deficiency, we used an HCT 116-derived cell line stably depleted of VDR using shRNA (hereafter named ShVDR) (Larriba et al., 2011). Induction of SIRT1 gene expression by 1,25(OH)2D3 was confirmed by RT-qPCR analysis in ShControl but not in ShVDR cells, although basal SIRT1 levels were similar (Fig. 1E). At the protein level, confocal imaging showed reduced SIRT1 protein in ShVDR as compared to ShControl cells (Fig. 1F), suggesting that VDR contributes to maintaining basal SIRT1 protein expression. Western blot analyses on nuclear extracts confirmed that ShControl cells responded to 1,25(OH)2D3 with increased SIRT1 protein levels in contrast to the unresponsiveness of ShVDR cells (Fig. 1G). Moreover, the upregulating effect of 1,25(OH)2D3 was specific for SIRT1, since the expression of the other nuclear sirtuin, SIRT7, was unaffected (Fig. 1H). Together, these data indicated that 1,25(OH)2D3 specifically increased SIRT1 mRNA and protein levels in CRC cell lines via VDR. This is important because SIRT1 depletion and/or sirtuin activity decay might represent an important neglected mediator of VD deficiency and a target for intervention in CRC. Whether VD controls sirtuin activity independently of sirtuin level in CRC is a critical issue that deserved further exploration.
1,25(OH)2D3 induces SIRT1 deacetylase activity in CRC cells
Evaluation of global sirtuin activity in the nuclei of CRC cells revealed that treatment with 1,25(OH)2D3 increased NAD+-dependent deacetylase activity 1.5-fold as compared to control untreated cells (Fig. 2A). NAD+ is a cofactor required for sirtuin activity. Metabolic control of the level of NAD+ represents an important node for sirtuin regulation and could explain the induction of general sirtuin activity by 1,25(OH)2D3. Indeed, nuclear NAD+ level doubled in response to 1,25(OH)2D3 in human CRC cells (Fig. 2B). However, whether 1,25(OH)2D3 could specifically increase SIRT1 activity remained unknown.
Liganded VDR interacts with SIRT1 in kidney and bone cells (Sabir et al., 2017) suggesting that 1,25(OH)2D3 might alter SIRT1 expression or activity via post-transcriptional mechanisms. Since protein-protein interactions regulate stability and activity of proteins, the ability of 1,25(OH)2D3 to promote VDR-SIRT1 interaction in CRC cells was evaluated by co-immunoprecipitation assays. The results revealed that 1,25(OH)2D3 potently (4-fold) enhanced VDR/SIRT1 complex formation (Fig. 2C). To test if this interaction alters SIRT1 activity, the acetylation status of two specific SIRT1 substrates (FOXO3a and H3K9) was evaluated as a read out of SIRT1 deacetylase activity in response to 1,25(OH)2D3. Acetyl-lysine immunoprecipitation followed by Western blotting demonstrated efficient deacetylation of FOXO3a in response to 1,25(OH)2D3 (Fig. 2D). Likewise, the level of acetylated histone H3K9 (Ace H3K9) dramatically decreased in response to 1,25(OH)2D3 (Fig. 2E), indicating specific activation of SIRT1 by 1,25(OH)2D3. The effect of 1,25(OH)2D3 was mediated by VDR since ShVDR cells did not show Ace H3K9 deacetylation in response to 1,25(OH)2D3, whereas ShControl cells maintained the response (Fig. 2F). Importantly, the specific SIRT1 activator SRT1720 reduced Ace H3K9 levels in both ShControl and ShVDR cells, thus mimicking 1,25(OH)2D3 action independently of VDR (Fig. 2F).
The ability of 1,25(OH)2D3 to increase protein deacetylation mediated by SIRT1 could be driven either by elevated SIRT1 level or by increased SIRT1 activity. To distinguish between these two possibilities, we performed an in vitro deacetylase assay using equivalent amounts of immunoprecipitated SIRT1 from cells treated or not with 1,25(OH)2D3. SIRT1-specific deacetylase activity was enhanced 3-fold in cells treated with 1,25(OH)2D3 despite similar levels of immunoprecipitated SIRT1 (Fig. 2G).
Together, these results indicated that 1,25(OH)2D3 induces nuclear SIRT1 activity. This implies that VD deficiency and/or loss of tumour VDR expression (both observed in a proportion of CRC patients) (Evans et al., 1998Ferrer-Mayorga et al., 2019Giardina et al., 2015Larriba et al., 20132009Pálmer et al., 2004Peña et al., 2005) may negatively impact SIRT1 deacetylase activity and prompted us to examine samples of CRC patients.
CRC biopsies exhibit discrepancies between SIRT1 protein levels and deacetylase activity
Changes in SIRT1 RNA and protein expression and enzymatic activity were analysed in samples of CRC patients. First, we compared available data (TCGA-COAD database) from 160 pairs of human colon adenocarcinoma samples matched with their normal adjacent tissue. TNM plot analysis revealed that both VDR and SIRT1 gene expression decreased from normal to tumour colonic tissue in a highly significant fashion (P<0.001) (Fig. 3A-B).
Immunohistochemical analysis of human colon tumor tissue microarrays (TMAs) from the Biobank of Hospital Clínico San Carlos (HCSC, Madrid) revealed a significant (P=0.017) positive association between the levels of VDR and SIRT1 proteins (Fig. 3C). An additional TMA containing healthy colon samples was used to compare tumor and nontumor tissues. The level of VDR protein was usually lower in tumour samples (Fig. 3D), while SIRT1 protein expression showed high variability between tumors and high cellular intratumor heterogeneity (Fig. 3D-E). Globally, TMA data suggested a trend to decreased SIRT1 protein in tumor samples (Fig. 3E), but high variability hampered statistical significance even though for SIRT1 gene expression, differences were statistically significant as shown in Figure 3B.
Global acetylation also showed high cellular intratumor heterogeneity and variability between samples without statistically significant differences between tumor and nontumor samples (data not shown), showing that is not a valid, specific read out of SIRT1 activity. As an alternative approach, the levels of SIRT1 protein and its substrate Ace H3K9 were compared first in intestinal cell lines and then in human tumours. Surprisingly, SIRT1 protein level increased in CRC cells as compared to normal HIEC6 cells (Fig. 3F). However, increased SIRT1 levels coincided with elevated Ace H3K9 (Fig. 3F), indicating SIRT1 inactivation in CRC cells. The results in cell lines encouraged us to evaluate the potential discordance of SIRT1 and Ace H3K9 levels in fresh-frozen samples from paired tumor and nontumor colonic tissue of patients from Hospital Universitario Fundación Alcorcón (HUFA). All tissues were evaluated to have at least 85% of tumor cells by immunohistochemistry (IHC). Western blotting allowed simultaneous detection of SIRT1 and Ace H3K9. Notably, whereas the level of SIRT1 protein variably increased or decreased from nontumor to tumor tissues in different patients, acetylation of its specific substrate H3K9 consistently increased in tumor samples (Fig. 3G), reflecting SIRT1 inactivation as found in carcinoma cell lines (Fig. 3F). In vitro deacetylation assays performed on SIRT1 immunoprecipitated from SIRT1-positive patient samples confirmed a common decrease of SIRT1 activity in colon tumor tissue (Fig. 3H). Western blotting illustrated the variability of SIRT1 level among tumors and its correlation with VDR level (Fig. 3I), while Ace H3K9 levels are generally high, but variable. Importantly, in vitro deacetylation assays on SIRT1 immunoprecipitated from the same samples (Fig. 3J) confirmed Ace H3K9 as a bona fide read out of SIRT1 inactivation. The discrepancy between the levels of SIRT1 protein expression and activity in tumor samples may solve the controversy previously reported regarding the use of SIRT1 as a tumor biomarker and poses the question of how 1,25(OH)2D3 specifically enhances SIRT1 deacetylase activity.
1,25(OH)2D3 induction of SIRT1 activity is mediated through auto deacetylation
Since SIRT1 catalyses its auto deacetylation as a mechanism to increase its deacetylase activity towards other substrates (Fang et al., 2017) and 1,25(OH)2D3 facilitates VDR-SIRT1 interaction, we explored the possibility that liganded VDR increases SIRT1 activity by facilitating SIRT1 auto deacetylation. To evaluate SIRT1 acetylation status upon exposure to 1,25(OH)2D3, nuclear extracts from CRC cells challenged or not with 1,25(OH)2D3 were immunoprecipitated using anti-acetyl-lysine antibodies and subjected to Western blotting. 1,25(OH)2D3 greatly reduced the nuclear pool of acetylated SIRT1 (Fig. 4A). Remarkably, the specific SIRT1 activator SRT1720 also drove SIRT1 deacetylation in HCT 116 CRC cells (Fig. 4B). Consistently, ShVDR cells did not reduce acetylated SIRT1 in response to 1,25(OH)2D3 but retain their response to SRT1720 (Fig 4C). These results indicated that 1,25(OH)2D3 activation of SIRT1 is mediated through SIRT1 deacetylation.
Next, we sought to get further insight into the mechanism of regulation of SIRT1 activity by 1,25(OH)2D3. As the acetyltransferase EP300 acetylates and inactivates SIRT2 (Han et al., 2008) and is critical for CRC cell signalling (Chocarro-Calvo et al., 2013Gutiérrez-Salmerón et al., 2020), we used the ASEB engine (Wang et al., 2012) to search for SIRT1 residues potentially acetylated by EP300. This approach identified K610, a lysine conserved from mouse to human, as a potential EP300 target (Fig. 4D). The role of K610 in the regulation of SIRT1 activity was assessed by generating a non-acetylable K610R mutant, which should be resistant to inactivation through acetylation. We compared the enzymatic activity of K610R SIRT1 with that of wild type (WT) and inactive H363Y SIRT1 (Vaziri et al., 2001). Nuclear extracts from cells transfected with MYC-tagged SIRT1 WT or mutants were immunoprecipitated using anti-acetyl-lysine antibodies. Western blotting using an anti-MYC antibody revealed that exogenous WT SIRT1 was acetylated (Fig. 4E), as it was endogenous SIRT1 (Fig 4A). Of note, catalytically inactive H363Y SIRT1 was highly acetylated, in striking contrast with the lack of acetylation of the K610R SIRT1 mutant (Fig. 4E), This highlights the importance of K610 for the regulation of SIRT1 activity.
Since interaction with VDR was important for SIRT1 deacetylation activity, the ability of each mutant to interact with VDR was compared in immunoprecipitation assays. The results revealed that the K610R SIRT1 mutation facilitated the interaction with VDR, whereas the inactive H363Y SIRT1 showed low interaction with VDR (Fig 4F). This suggested that acetylation of SIRT1 might interfere with or destabilize its interaction with VDR, and that the ability of 1,25(OH)2D3 to favour VDR-SIRT1 interaction may allow SIRT1 auto deacetylation, and thus enhance its activity. The activity of mutant SIRT1 enzymes was first inferred from the level of Ace H3K9, which was lower in cells expressing K610R SIRT1 (Fig 4G). In vitro NAD+-dependent deacetylase assays were next used to compare the activity of WT and mutant SIRT1. Using equivalent amounts of immunoprecipitated MYC-tagged enzymes, we found a 10-fold higher deacetylase activity for the K610R mutant than for WT SIRT1, whereas as expected the H363Y inactive mutant exhibited very low activity as expected (Fig. 4H).
SIRT1 activation mimics the antiproliferative effects of 1,25(OH)2D3 in CRC cells unresponsive to VD
Collectively, our data indicated that the acetylation status of SIRT1 K610 is critically controlled and governs SIRT1 activity in the nuclei of CRC cells. By promoting VDR-SIRT1 interaction, 1,25(OH)2D3 facilitates SIRT1 auto deacetylation and enhances SIRT1 activity. These results suggest that SIRT1 is an effector of 1,25(OH)2D3 and prompted us to ask whether the pharmacological activation of SIRT1 would rescue CRC cells that are unresponsive to 1,25(OH)2D3.
The anticancer effects of 1,25(OH)2D3 are exerted at least partially through the inhibition of tumor cell proliferation (Carlberg and Muñoz, 2022Muñoz and Grant, 2022). 1,25(OH)2D3 strongly reduced the percentage of CRC cells in S-G2M cell-cycle phases (Fig. 5A) and consequently, it lengthened the cell cycle and reduced CRC cell proliferation. We aimed to analyse the effect of the modulation of SIRT1 activity on the proliferation of CRC cells and on the antiproliferative action of 1,25(OH)2D3. Inhibition of general sirtuin activity with Nicotinamide (NAA) abolished the antiproliferative effects of 1,25(OH)2D3 on CRC cells (Fig. 5B). Moreover, activation of SIRT1 by the small molecule SRT1720 mimicked the response to 1,25(OH)2D3 and efficiently reduced the percentage of cells in S-G2M cell cycle phases (Fig. 5C). These data indicate that induction of SIRT1 activity mediates 1,25(OH)2D3-driven cell cycle lengthening in CRC cells.
As expected, in contrast with ShControl cells, ShVDR CRC cells did not extend the cell cycle in response to 1,25(OH)2D3 (Fig. 5D). Accordingly, an early and strong reduction in the proliferation rate in response to 1,25(OH)2D3 was noted in ShControl cells but it was almost absent in ShVDR cells (Fig. 5E). Specific SIRT1 activation with SRT1720 reduced the proliferation of ShControl cells at a comparable magnitude to 1,25(OH)2D3 and, importantly, also reduced proliferation of 1,25(OH)2D3-unresponsive ShVDR cells (Fig.5E).
Collectively, these results show that SIRT1 acts downstream of 1,25(OH)2D3-activated VDR in the inhibition of CRC cell proliferation. They also suggest that activation of SIRT1 in CRC patients with vitamin D deficiency or with VD unresponsive tumors (conditions frequently observed in advanced CRC), could be a therapeutic strategy to effectively antagonize cell proliferation.
Here we reveal that 1,25(OH)2D3 activates the epigenetic regulator SIRT1 by acting at multiple levels. Ligand-activated VDR post-translationally upregulates SIRT1 activity by facilitating its auto deacetylation. Mechanistically, SIRT1 K610 is identified as essential to control VDR-SIRT1 interaction and SIRT1 acetylation status and activity. Notably, both SRT1720 and 1,25(OH)2D3 activate SIRT1 through auto deacetylation. Modulation of SIRT1 activity by 1,25(OH)2D3 through deacetylation occurs in addition to increased mRNA and protein levels as well as to an elevation of cofactor (NAD+) availability. Thus, 1,25(OH)2D3 activates SIRT1 acting at multiple levels, from gene expression to activity, and these findings broaden our understanding of how VD deficiency is associated with CRC.
We also show that SIRT1 deacetylase activity mediates the anti-proliferative action of 1,25(OH)2D3 in CRC, which is in line with the recent proposal that SIRT1 activity opposes several cancer cell hallmarks (Yousafzai et al., 2021). In cases of VD deficiency and/or unresponsiveness such as advanced colorectal tumor lacking VDR expression due to transcriptional repression by SNAIL1/2 (Pálmer et al., 2004Larriba et al., 2009), the activation of SIRT1 by compounds other than 1,25(OH)2D3 may be especially relevant. Clinical trials that explore SIRT1 activators in CRC are rare (https://clinicaltrials.gov). Some trials are designed to study doses and pharmacokinetics of resveratrol, a known sirtuin activator that has many other targets (Patel et al., 2011). Currently there are not known clinical trials using SRT1720 or alternative SIRT1 specific activators.
Consideration of SIRT1 protein level as a prognostic marker for cancer has received much attention, but conflicting results reporting both, tumor SIRT1 increases and decreases raised controversy (Ren et al., 2017Wu et al., 2017). Accordingly, some studies propose SIRT1 either as tumour suppressor or as tumor promoter in CRC (Carafa et al., 2019Ren et al., 2017). Importantly, SIRT1 protein level might not always reflect SIRT1 activity, and some tumors could possibly increase SIRT1 protein expression to counteract inactivation. Our results, which indicate the existence of discrepancies between SIRT1 levels and activity in human colorectal tumors, may help to resolve the controversy. Interestingly, alternative posttranslational SIRT1 modifications may have a similar outcome. For example, proteolytic cleavage of SIRT1 induced by TNFα in osteoarthritis renders accumulation of a catalytically inactive 75 kDa resistant fragment (Oppenheimer et al., 2012) that could be detected by immunohistochemistry. Since inflammation (in general) and especially TNFα is an important mediator of CRC, understanding if this fragment is generated in CRC patients and whether its presence would also lead to decreased SIRT1 activity in samples strongly positive for SIRT1 remains as an interesting unexplored issue.
The ability of 1,25(OH)2D3 to activate SIRT1 may have further major implications for physiological and pathological processes other than CRC since SIRT1 target proteins control several processes such as mitochondria physiology, apoptosis, and inflammation (Strycharz et al., 2018). Thus, 1,25(OH)2D3 may not only regulate gene expression via direct transcriptional regulation of VDR-bound targets, but it can secondarily impact a wide range of events and processes via the control of SIRT1 activity. This may explain how 1,25(OH)2D3 regulates the activity of FOXO3a, an important modulator of cell proliferation and the immune response that is targeted by SIRT1 (An et al., 2010). SIRT1 also modulates the activity of the tumour suppressor p53, whose mutation is a relatively early event in a high proportion of CRCs (Polidoro et al., 2013Strycharz et al., 2018). In this way, SIRT1 activation greatly broaden the known vitamin D regulatory action, albeit whether 1,25(OH)2D3 directs SIRT1 activity towards specific substrates remains to be elucidated. The accumulating evidence for SIRT1 involvement in chronic inflammation (Mendes et al., 2017) suggests a potential benefit of SIRT1 activation by 1,25(OH)2D3 in chronic inflammation-linked CRC and other diseases. Moreover, the mechanism revealed here for SIRT1 activation by 1,25(OH)2D3 may be relevant for diseases associated with VD deficiency beyond CRC, including autoimmune disorders and diabetes.
In conclusion, the multilevel activation of SIRT1 deacetylase by 1,25(OH)2D3 broadens the range of known 1,25(OH)2D3 targets and positions SIRT1 as an important mediator of the protective action of VD against CRC and potentially other neoplasias and non-tumoral diseases. These data support SIRT1 activators as useful agents in situations of dysfunction of the VD system.
Materials and methods
All materials and resources can be found in Table 1.
Experimental model and subject details
Patient samples were derived from surgical removal at Fundación Jimenez Diaz University Hospital, General and Digestive Tract Surgery Department: 95 patient samples diagnosed with stage II CRC. The Institutional Review Board (IRB) of the Fundación Jimenez Diaz Hospital, reviewed and approved the study, granting approval on December 9, 2014, act number 17/14.
Samples from seven healthy individuals were obtained from Hospital Clínico San Carlos (HCSC). IRB-HCSC act number 21/498-E granting approval on June 25th, 2021. Paired fresh-frozen samples from tumor and nontumor adjacent tissue were collected from 18 patients at Hospital Universitario Fundación Alcorcón (HUFA). IRB-HUFA, act number 17/68 granting approval on June 8th, 2017. Colon cancer tissues were collected using protocols approved by the corresponding Ethics Committees and following the legislation. All patients gave written informed consent for the use of their biological samples for research purposes. Fundamental ethical principles and rights promoted by Spain (LOPD 15/1999) and the European Union EU (2000/C364/01) were followed. All patients’ data were processed according to the Declaration of Helsinki (last revision 2013) and Spanish National Biomedical Research Law (14/2007, July 3).
Colorectal cell panels
Human colorectal adenocarcinoma HT-29 and HCT 116 and HCT 116 derived ShControl and ShVDR cells were cultured under standard conditions, in 5%CO2 at 37°C with DMEM containing 25 mM glucose (unless specifically indicated) and 40 mM LiCl to mimic Wnt signalling and supplemented with 10% foetal bovine serum (FBS) and 1% penicillin-streptomycin. Cells were treated as indicated for 24h. HCT 116 ShControl and ShVDR cell lines were derived previously (Larriba et al., 2011)
cells were seeded in plates at 50% confluence for plasmid transfection, using JetPei PolyPlus reagent (Genycell Biotech), following the manufacturer’s instructions. After 24 h cells were treated as indicated for 24 h.
For Sirt1 siRNA, cells plated in six well plates at 50% confluence were transfected with JetPRIME PolyPlus reagent (Genycell Biotech) following the manufacturer’s instructions. After 2 days cells were treated 24 h with LiCl and then another 24 h with 1,25(OH)2D3 before harvesting to analyse by western blot.
Site directed mutagenesis
Directed mutagenesis was performed using the QuickChange Site-Directed Mutagenesis Kit (Stratagene). Oligonucleotides containing the mutation (Table 1) were designed following the manufacturer’s recommendations. 2.5 U of PfuTurbo DNA polymerase was added to each sample and they were amplified by performing the following cycles: 1 denaturation cycle at 95 ° C for 1 minute, 18 amplification cycles of 30 seconds at 95 ° C, 50 seconds at 60 ° C, 1 minute / kilobase of plasmid at 68 °C and the last extension cycle 7 minutes at 68aC. Once amplified, 10 U of Dnp I enzyme were added and incubated at 37 ° C for one hour to digest the supercoiled double-stranded DNA and eliminate the parental molecules that are not mutated.
The correct introduction of the mutation was evaluated by plasmid sequencing using the BigDye Cycle Sequencing Kit (Applied Biosystems).
Preparation of cell extracts
Whole cell extracts
Cells were washed with iced PBS before extract preparation and scraped in RIPA buffer (10 mM Tris HCl [pH 7.4], 5 mM EDTA, 5 mM EGTA, 1% Tryton X100, 10 mM Na4P2O7 [pH 7.4], 10 mM NaF, 130 mM NaCl, 0.1% SDS, 0,5% Na-deoxycholate). After 5 min on ice, cells were pelleted (12000 rpm for 5 min, 4°C) and the supernatant was directly used as whole cell extract or frozen at −80 °C.
Fractionated cell extracts
After washing as before, cells were scraped in hypotonic buffer (20 mM Hepes, [pH 8.0], 10 mM KCl, 0,15 mM EDTA, 0,15 mM EGTA, 0,05% NP40 and protease inhibitors) and incubated on ice for 10 min before adding 1:2 vol of sucrose buffer (50 mM Hepes [pH=8.0], 0.25 mM EDTA, 10 mM KCl, 70% sucrose). Lysates were fractionated (5000 rpm for 5 min at 4°C) to obtain the cytoplasmic fraction in the supernatant. Nuclear pellets were washed twice with washing buffer (20 mM Hepes [pH 8.0], 50 mM NaCl, MgCl2 1.5 mM, 0,25 mM EDTA, 0,15 mM EGTA, 25% glycerol and protease inhibitors), pelleted at 5000 rpm, 5 min at 4°C and resuspended in nuclear extraction buffer (20 mM Hepes[pH 8.0], 450 mM NaCl, MgCl2 1.5 mM, 0,25 mM EDTA, 0,15 mM EGTA, 0,05% NP40, 25% glycerol and protease inhibitors) before centrifugation at 12,000 rpm for 5 min at 4°C to pellet and discard cell debris. The supernatants were used as nuclear fractions.
SIRT1 Activity Assay
NAD+-dependent deacetylase activity was measured on nuclear extracts of HT-29 cells treated as indicated, using SIRT-Glo Assay kit (Promega), following the manufacturer’s instructions. For each reaction, 1 μg of protein was incubated with SIRT-Glo Reagent Mix at room temperature for 3 min, and the luminescence was measured using a GloMax Microplate Reader (Promega). Relative luciferase units (RLU) were calculated as fold induction relative to the corresponding control.
Nuclear NAD+ abundance
The NAD/NADH-Glo Assay Kit (Promega) was used to measure NAD+ abundance in HCT 116 nuclear extracts. 20 μg of protein were diluted in 50 μl of lysis buffer (described for cell extracts) for each reaction and 25 μl of 0.4 N HCl were added before incubation at 60 °C for 15 minutes. This exploits the high stability of NAD+ under acidic conditions, while NADH breaks down. Samples were cooled down to room temperature for 10 min and neutralized with 25 μL 0.5 M Trizma®. NAD+ levels were then measured using a GloMax microplate reader (Promega) and following the manufacturer’s instructions.
For immunoprecipitation from fractionated extracts, the hypotonic buffer was modified by adding 100mM NaCl and 0.1% NP40. For immunocomplex formation, protein A/G-coated magnetic beads (Invitrogen) were washed 3 times with the extraction buffer before coating with the primary antibody for 2 hr at 4 °C in a rotating wheel, followed by 2 washes with the same buffer to eliminates unbound antibody and then extracts were added O/N at 4°C in the rotating wheel. Immunocomplexes were washed twice and used for western blotting.
Proteins from lysed cells or immunoprecipitates were denatured and loaded on sodium dodecyl sulfate polyacrylamide gels and then transferred to polyvinylidene difluoride membranes (Bio-Rad). After blocking with 5% (w/v) BSA or milk the membrane was incubated with the corresponding primary and secondary antibodies (Bio-Rad). The specific bands were analyzed using Thyphoon or ChemiDoc Imaging Systems (Bio-Rad).
Cells in cover slips were washed three times and fixed with 4% paraformaldehyde in PBS [pH 7.4] for 10min, washed again, permeabilized (PBS [pH7.4], 0.5% Triton X-100, 0.2% BSA) for 5min, blocked (PBS [pH7,4, 0,05% Triton X-100, 5% BSA) for 1h at room temperature and incubated with primary antibody over night at 4°C. After three washes (5 min) cells were incubated with secondary antibody for 1h at room temperature. Slides were mounted and images were acquired using a SP5 confocal microscope (Leica) with a 63x objective. Fluorescence intensity was quantified using Image J software. For each experiment, 3 different fields were evaluated per slide.
Cell-growth curves and cell cycle
Cell growth was determined by colorimetry. For that, cells were seeded at a density of 20,000 cells per well in a Corning 12-well plate and treated as indicated for 1 to 6 days. After that cells were treated with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Sigma-.Aldrich) in a 1:10 ratio in a culture medium and incubated at 37 °C for 3 h. In living cells, MTT is reduced to formazan, which has a purple color. The medium was removed, and the formazan was resuspended in dimethylsulfoxide (DMSO), transferred to p96 plates and analyzed by a Spectra FLUOR (Tecan) at 542 nm. Viability was measured in four independent experiments and then duplicated.
For cell cycle analysis, cells were harvested by trypsinization, washed with PBS, fixed in 70% ethanol, stained with 7-AAD (Santa Cruz Biotechnology) for 10 min at 37°C, and analyzed by flow cytometry (FACSCalibur, Becton-Dickinson). The percentages of cells at each cell cycle phase were analyzed using CXP software (Becton-Dickinson).
Total RNA was extracted from 3 replicates of colorectal cells using TRIzol reagent (Invitrogen). Reverse transcription of 1μg of RNA was performed according to the manufacturer’s instructions Reagents and detection systems were from Applied biosystems. 18S ribosomal RNA primers served as a nonregulated control. Relative expression was calculated using the Ct method, expressed as 2−ΔΔCt (Livak and Schmittgen, 2001). The PCR efficiency was approximately 100%.
TMA, immunohistochemistry, and quantification
Tissue microarrays (TMA) containing 92 cores of stage 2 human CRC samples, were constructed using the MTA-1 tissue arrayer (Beecher Instruments, Sun Prairie) for immunohistochemistry analysis. Each core (diameter 0.6 mm) was punched from pre-selected tumour regions in paraffin-embedded tissues. We chose central areas from the tumour, avoiding foci of necrosis. Staining was conducted in 2-μm sections. Slides were deparaffinized by incubation at 60°C for 10 min and incubated with PT-Link (Dako, Agilent) for 20 min at 95°C in low pH to detect VDR, SIRT1 and acetyl lysines. To block endogenous peroxidase, holders were incubated with peroxidase blocking reagent (Dako, Agilent) and then with (1:50) dilutions of antibodies: anti-VDR, anti-SIRT1 and (1:300) anti-acetyl lysine, overnight at 4°C. All previously described antibodies presented high specificity. After that, slides were incubated for 20 min with the appropriate anti-Ig horseradish peroxidase-conjugated polymer (EnVision, Dako, Agilent). Sections were then visualized with 3,3’-diaminobenzidine (Dako, Agilent) as a chromogen for 5 min and counterstained with Harrys’ Haematoxylin (Sigma Aldrich, Merck). Photographs were taken with a stereo microscope (Leica DMi1). According to the human protein atlas (available at http://www.proteinatlas.org), a human colon tissue was used as a positive control for immunohistochemical staining to determine anti-VDR concentration and a human testis tissue for anti-SIRT1. Immunoreactivity was quantified blind with a Histoscore (H score) that considers both the intensity and percentage of cells stained for each intensity (low, medium, or high) following this algorithm (range 0–300): H score = (low%) × 1 + (medium%) × 2 + (high %) × 3. Quantification for each patient biopsy was calculated blindly by 2 investigators (MJFA and JMU). VDR and SIRT1 showed nuclear staining, whereas acetyl lysines were nuclear and extranuclear. Clinicopathological characteristics of patients are summarized in Table 2.
Gene expression (RNA-seq) data from the TCGA colon adenocarcinoma cohort (COAD) was downloaded from the cBioportal for Cancer Genomics (http://www.cbioportal.org) using CGDS-R (Cerami et al., 2012Gao et al., 2013) following TCGA guidelines (http://cancergenome.nih.gov/publications/publicationguidelines). SIRT1 and VDR gene expression from samples of colorectal cancer and healthy tissue adjacent to the tumour were analysed with TNMplot.com (https://tnmplot.com/analysis/) tool (Bartha and Győrffy, 2021). An asymptotic Spearman correlation test using original log2 expression values was used to determine the significance of the Spearman rank correlation.
Results are presented as fold induction, mean ± SEM from three biological replicates. To determine whether calculated Hscores for protein levels were well-modelled by a normal distribution, Kolmogorov-Smirnov test was used. Association between SIRT1 and VDR protein levels was performed with Chi-square test. Comparisons between two independent groups (healthy versus tumors) were performed with Mann-Whitney U-test. For statistical association, Hscore of antigens were categorized as high- or low-expression levels using the median as a cut-off point. Tests for significance between two sample groups were performed with Student’s t test and for multiple comparisons, ANOVA with Bonferroni’s post-test.
Funding for this work was provided by the Agencia Estatal de Investigación (PID2019-104867RB-I00 / AEI/10.13039/501100011033, RTI2018-099343-B-100 and PID2021-127645OA-I00); Instituto de Salud Carlos III (CIBERONC, CB16/12/00273 and CB16/12/00326); Comunidad de Madrid (Ayudas Atracción de Talento 2017-T1/BMD-5334 and 2021-5A/BMD-20951; A385-DROPLET Young Reserchers R&D Project 2019 CAM-URJC; PRECICOLON-CM, P2022/BMD7212); Universidad Rey Juan Carlos (ADIPOMELM, Proyecto Puente de Investigación 2020).
We thank Lucille Banham for her valuable assistance in the preparation of the English manuscript and María Gutiérrez Salmerón for help making figures.
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication
CGJ, JMGM, AM, and MJL conceived the experiments, JMGM, ACC, AB, JMU, MJFA, MCF, JCR performed the experiments. JMGM and ADLV performed the Figs. All authors participated in the discussion. CGJ, AM and MJL wrote the manuscript and provided supervision.
The authors declare that they have no conflict of interest. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) as specified under the paragraph Human samples in the Materials and Methods Section and with the Helsinki Declaration of 1975, as revised in 2000. lnformed consent was obtained from all patients for being included in the study.
- Stimulation of Sirt1-Regulated FoxO Protein Function by the Ligand-Bound Vitamin D ReceptorMol Cell Biol 30:4890–4900https://doi.org/10.1128/MCB.00180-10
- Tnmplot.Com: A web tool for the comparison of gene expression in normal, tumor and metastatic tissuesInt J Mol Sci 22:1–12https://doi.org/10.3390/ijms22052622
- New aspects of vitamin D metabolism and action — addressing the skin as source and targetNat Rev Endocrinol 16:234–252https://doi.org/10.1038/s41574-019-0312-5
- Slowing ageing by design: The rise of NAD+ and sirtuin-activating compoundsNat Rev Mol Cell Biol 17:679–690https://doi.org/10.1038/NRM.2016.93
- Comprehensive molecular characterization of human colon and rectal cancerNature 487:330–337https://doi.org/10.1038/nature11252
- Dual Tumor Suppressor and Tumor Promoter Action of Sirtuins in Determining Malignant PhenotypeFront Pharmacol 10:38https://doi.org/10.3389/fphar.2019.00038
- An update on vitamin D signaling and cancerSemin Cancer Biol 79:217–230https://doi.org/10.1016/j.semcancer.2020.05.018
- Vitamin D and the risk for cancer: A molecular analysisBiochem Pharmacol 196:https://doi.org/10.1016/J.BCP.2021.114735
- The cBio Cancer Genomics Portal: An open platform for exploring multidimensional cancer genomics dataCancer Discov 2:401–404https://doi.org/10.1158/2159-8290.CD-12-0095
- Effect of Vitamin D 3 Supplements on Development of Advanced CancerJAMA Netw Open 3:https://doi.org/10.1001/jamanetworkopen.2020.25850
- High levels of SIRT1 expression enhance tumorigenesis and associate with a poor prognosis of colorectal carcinoma patientsSci Rep 4:https://doi.org/10.1038/SREP07481
- Glucose-Induced β-Catenin Acetylation Enhances Wnt Signaling in CancerMol Cell 49:474–486https://doi.org/10.1016/j.molcel.2012.11.022
- Vitamin D receptor expression as a predictive marker of biological behavior in human colorectal cancerClinical Cancer Research 4:1591–1595
- Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countriesCA Cancer J Clin 68:394–424https://doi.org/10.3322/caac.21492
- Sirt7 promotes adipogenesis in the mouse by inhibiting autocatalytic activation of Sirt1Proc Natl Acad Sci U S A 114:https://doi.org/10.1073/pnas.1706945114
- The role of vitamin D in reducing cancer risk and progressionNat Rev Cancer https://doi.org/10.1038/nrc3691
- Vitamin D Effects on Cell Differentiation and Stemness in CancerCancers (Basel) 12:2413–2432https://doi.org/10.3390/cancers12092413
- Vitamin D receptor expression and associated gene signature in tumour stromal fibroblasts predict clinical outcome in colorectal cancerGut 66:1449–1462https://doi.org/10.1136/gutjnl-2015-310977
- Mechanisms of action of vitamin D in colon cancerJ Steroid Biochem Mol Biol 185:1–6https://doi.org/10.1016/j.jsbmb.2018.07.002
- The SIRT1 Deacetylase Suppresses Intestinal Tumorigenesis and Colon Cancer GrowthPLoS One 3:https://doi.org/10.1371/journal.pone.0002020
- Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortalSci Signal 6:1–34https://doi.org/10.1126/scisignal.2004088
- Regulation of VDR expression in Apc-mutant mice, human colon cancers and adenomasCancer Prevention Research 8:387–399https://doi.org/10.1158/1940-6207.CAPR-14-0371
- Comparing the Evidence from Observational Studies and Randomized Controlled Trials for Nonskeletal Health Effects of Vitamin DNutrients 14:1–31
- Paradoxical activation of AMPK by glucose drives selective EP300 activity in colorectal cancerPLoS Biol 18:https://doi.org/10.1371/journal.pbio.3000732
- Acetylation of Sirt2 by p300 attenuates its deacetylase activityBiochem Biophys Res Commun 375:576–580https://doi.org/10.1016/j.bbrc.2008.08.042
- Vitamin D in Cancer Prevention: Gaps in Current Knowledge and Room for HopeNutrients 14:1–32
- Total Vitamin D Intake and Risks of Early-Onset Colorectal Cancer and PrecursorsGastroenterology 161:1208–1217https://doi.org/10.1053/j.gastro.2021.07.002
- Prediagnostic Vitamin D Status and Colorectal Cancer Survival by Vitamin D Binding Protein Isoforms in US CohortsJ Clin Endocrinol Metab https://doi.org/10.1210/CLINEM/DGAC742
- Vitamin D is a multilevel repressor of Wnt/β-catenin signaling in cancer cellsCancers (Basel) https://doi.org/10.3390/cancers5041242
- Snail2 cooperates with Snail1 in the repression of vitamin D receptor in colon cancerCarcinogenesis 30:https://doi.org/10.1093/carcin/bgp140
- SNAIL vs vitamin D receptor expression in colon cancer: therapeutics implicationsBr J Cancer 92:985–989https://doi.org/10.1038/sj.bjc.6602484
- Vitamin D receptor deficiency enhances Wnt/β-catenin signaling and tumor burden in colon cancerPLoS One 6:https://doi.org/10.1371/journal.pone.0023524
- Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) MethodMethods 25:402–408https://doi.org/10.1006/meth.2001.1262
- Principal results of the VITamin D and OmegA-3 TriaL (VITAL) and updated meta-analyses of relevant vitamin D trialsJ Steroid Biochem Mol Biol 198:105522https://doi.org/10.1016/j.jsbmb.2019.105522
- Nuclear sirtuins and inflammatory signaling pathwaysCytokine Growth Factor Rev 38:98–105https://doi.org/10.1016/j.cytogfr.2017.11.001
- Vitamin D and Cancer: An Historical Overview of the Epidemiology and MechanismsOverview of the Epidemiology and Mechanisms Nutrients 14:1448https://doi.org/10.3390/nu14071448
- 75-kd sirtuin 1 blocks tumor necrosis factor α-mediated apoptosis in human osteoarthritic chondrocytesArthritis Rheum 64:718–728https://doi.org/10.1002/ART.33407
- The transcription factor SNAIL represses vitamin D receptor expression and responsiveness in human colon cancerNat Med 10:917–919https://doi.org/10.1038/nm1095
- Clinical trials of resveratrolAnn N Y Acad Sci 1215:161–169https://doi.org/10.1111/j.1749-6632.2010.05853.x
- E-cadherin and vitamin D receptor regulation by SNAIL and ZEB1 in colon cancer: Clinicopathological correlationsHum Mol Genet 14:3361–3370https://doi.org/10.1093/hmg/ddi366
- Vitamin D protects human endothelial cells from H2O2 oxidant injury through the Mek/Erk-sirt1 axis activationJ Cardiovasc Transl Res 6:221–231https://doi.org/10.1007/s12265-012-9436-x
- Haploinsufficiency of SIRT1 Enhances Glutamine Metabolism and Promotes Cancer DevelopmentCurrent Biology 27:483–494https://doi.org/10.1016/j.cub.2016.12.047
- SIRT1 enzymatically potentiates 1,25-dihydroxyvitamin D3 signaling via vitamin D receptor deacetylationJ Steroid Biochem Mol Biol 172:117–129https://doi.org/10.1016/J.JSBMB.2017.06.010
- No Association Between Vitamin D Supplementation and Risk of Colorectal Adenomas or Serrated Polyps in a Randomized TrialClinical Gastroenterology and Hepatology 19:128–135https://doi.org/10.1016/j.cgh.2020.02.013
- SIRT1 as a Therapeutic Target in Diabetic ComplicationsCurr Med Chem 25:1002–1035https://doi.org/10.2174/0929867324666171107103114
- hSIR2SIRT1 functions as an NAD-dependent p53 deacetylaseCell 107:149–159https://doi.org/10.1016/S0092-8674(01)00527-X
- ASEB: a web server for KAT-specific acetylation site predictionNucleic Acids Res 40:https://doi.org/10.1093/nar/gks437
- Meta-analysis of SIRT1 expression as a prognostic marker for overall survival in gastrointestinal cancerOncotarget 8:62589–62599https://doi.org/10.18632/oncotarget.19880
- Recent advances of SIRT1 and implications in chemotherapeutics resistance in cancerAm J Cancer Res 5233–5248
- Sirt1 Mediates Vitamin D Deficiency-Driven Gluconeogenesis in the Liver via mTorc2/Akt Signalinghttps://doi.org/10.1155/2022/1755563
- Inactivation of the vitamin D receptor in APC(min/+) mice reveals a critical role for the vitamin D receptor in intestinal tumor growthInt J Cancer 130:10–19https://doi.org/10.1002/IJC.25992