Skip to main content
NCBI home page
Howard Hughes Medical Institute Author Manuscripts logoLink to Howard Hughes Medical Institute Author Manuscripts
. Author manuscript; available in PMC: 2020 Aug 17.
Published in final edited form as: ACS Chem Biol. 2019 Aug 2;14(8):1802–1810. doi: 10.1021/acschembio.9b00384

A Glycoconjugated SIRT2 inhibitor with aqueous solubility allows structure-based design of SIRT2 inhibitors.

Jun Young Hong †,#, Ian Robert Price †,#, Jessica Jingyi Bai †, Hening Lin †,‡,*
PMCID: PMC6942458  NIHMSID: NIHMS1061122  PMID: 31373792

Abstract

Small molecule inhibitors for SIRT2, a member of the sirtuin family of nicotinamide adenine dinucleotide-dependent protein lysine deacylases, have shown promise in treating cancer and neurodegenerative diseases. Developing SIRT2-selective inhibitors with better pharmacological properties is key to further realize the therapeutic potential of targeting SIRT2. One of the best SIRT2-selective inhibitors reported is a thiomyristoyl lysine compound called TM, which showed promising anticancer activity in mouse models without much toxicity to normal cells. The main limitations of TM, however, are the low aqueous solubility and lack of X-ray crystal structures to aid future drug design. Here, we designed and synthesized a glucose-conjugated TM (glucose-TM) analog with superior aqueous solubility. Although glucose-TM is not cell permeable, the excellent aqueous solubility allowed us to obtain a crystal structure of SIRT2 in complex with it. The structure enabled us to design several new TM analogs, one of which, NH4-6, showed superior water solubility and better anticancer activity in cell culture. The results of these studies provided important insights that will further fuel the future development of improved SIRT2 inhibitors as promising therapeutics for treating cancer and neurodegeneration.

Introduction

Sirtuins are class III histone deacetylases that use nicotinamide adenine dinucleotide (NAD+) as a co-substrate. Initially thought to remove only acetyl groups, sirtuins have been shown to have additional enzymatic activities. For instance, SIRT2 removes fatty acyl groups on lysine with catalytic efficiency similar to its deacetylase, and SIRT5 possesses demalonyl, desuccinyl, and deglutaryl activities.1–5 Among all the seven sirtuins, SIRT2 is the only sirtuin that is mainly localized in the cytosol.6 SIRT2 has been reported to deacetylate various substrate proteins, including transcription factors, metabolic enzymes and signaling proteins. For instance, SIRT2 regulates cellular iron levels through de-acetylation of transcription factor, NRF2.7 Through removing acetyl group from K116 of Slug, SIRT2 prevents its degradation and consequently promotes growth of basal-like breast cancer.8 Besides these transcription factors, SIRT2 promotes tumor growth through deacetylating and activating metabolic enzymes, like pyruvate kinase M2 (PKM2) and lactate dehydrogenase A (LDH-A).9,10 Furthermore, SIRT2 inhibitors have been reported to have beneficial effects in neurodegenerative diseases, such as Alzheimer’s and Parkinson’s disease.11,12 Lastly, SIRT2 stabilizes oncoprotein c-Myc and further regulates transformation of tumor cells by removing fatty acyl groups on K-Ras4a.13–15 Because SIRT2 regulates numerous biological functions and promotes tumors, interests in inhibitor development have increased significantly.

Several SIRT2-selective inhibitors have been reported from various groups, including Diketopiperazine-Containing 2-Anilinobenzamides (Compound 53), NCO-90/140, KPM-1/2, AGK2, Tenovin-6 and SirReal2.12,16–21 Previously, we also reported a mechanism-based SIRT2-selective inhibitor, TM.14 The compound contains a thiomyristoyl group, which forms a covalent 1’-S-alkylimidate intermediate to detain the activity of SIRT2 (Figure 1). TM is cytotoxic in various cancer cell lines, including breast, pancreatic, colon, lung and many more, with relatively little toxicity in normal mammary epithelial cells. In addition, the compound inhibited tumor growth in a xenograft model and a genetic model of breast cancers in mice.14

Figure 1.

Figure 1.

Open in a new tab

Glucose-TM is a SIRT2-selective inhibitor in vitro. (A) Schematic showing how SIRT2 mechanism-based inhibitors work. (B) Structures and measured IC50 values (μM) of TM and Glucose-TM.

Although TM shows promising anticancer activity in mouse models, it has poor aqueous solubility due to its long hydrophobic thiomyristoyl group. The poor aqueous solubility is a practical problem and causes inconveniences in many studies. For example, when attempting to crystalize SIRT2 with TM for the purpose of structure-based inhibitor design, TM often precipitate out by itself. In addition, drugs with poor aqueous solubility often cause delivery problems, which can impede cellular and in vivo experiments. Thus, improving TM’s aqueous solubility became of interest to us.

The strategy of glycoconjugation has been widely used to improve the cancer cells selectivity (cancer cells often heavily rely on glycolysis, a phenomenon termed the Warburg effect) and aqueous solubility of parental drugs.22,23 For instance, glycoconjugates of docetaxel, which has poor aqueous solubility, were synthesized to improve its poor water solubility and to target cancer cells selectively. As a result, the conjugated compounds exerted 3- to 18- fold improvements in inhibitory activity compared to the aglycones.24 Therefore, we decided to synthesize a glycoconjugated SIRT2-selective inhibitor named Glucose-TM to target cancer cells selectively and improve solubility. Glucose-TM maintains the SIRT2 inhibition potency and selectivity while at the same time has dramatically improved aqueous solubility. Contrary to our expectation, Glucose-TM has decreased cell permeability compared to TM, which leads to decreased cellular activity. However, because of the significantly improved aqueous solubility, we were able to co-crystallize Glucose-TM with SIRT2. The structure shows that the N- and C-terminal ends of the thiomyristoyl lysine structure do not significantly contribute to binding of Glucose-TM. This structure insight allowed us to design various modifications on the N- and C-terminals of thiomyristoyl lysine, leading to compounds that maintained SIRT2-selective inhibition but with improved solubility.

Results and Discussion

Glucose-TM Design and Synthesis.

TM contains a thiomyristoyl lysine moiety with a carboxybenzyl protecting group (Cbz) on the N-terminal, and a phenylamide on the C-terminal.14 TM analogues with shorter fatty acids lost SIRT2 selectivity and inhibited SIRT1 and SIRT3. Thus, the thiomyristoyl group of TM is important for selective SIRT2 inhibition. Adding a hydroxyl group on the para-position of C-terminal aniline also led to loss of SIRT2 selectivity.25,26 Because of these observations, we decided to modify the N-terminal of TM to increase its water solubility.

We decided to conjugate glucose to TM using an azido-PEG linker, which is known to increase solubility (Figure 1).27 Click chemistry was utilized to link the azido-PEGylated TM and alkyne tagged glucose. The 1β position of glucose was chosen to attach the alkyne functional group (Figure 1). Using DataWarrior Software, the simulated cLogP value of Glucose-TM was approximately 4.05, while that of TM was approximately 8.81. This simulation suggests that glucose-TM should have significantly improved aqueous solubility.

Glucose-TM demonstrated increased solubility in PBS.

We tested whether there is an improvement in aqueous solubility by adding TM or Glucose-TM to Phosphate Buffered Saline (PBS). The compound was first dissolved at 50 mM in DMSO. Then, the DMSO stock was further diluted to 5 mg/ml (5.5 mM for Glucose-TM and 8.6 mM for TM) or 10 mg/ml (11 mM for Glucose-TM) in PBS. As expected, when TM was diluted to 5 mg/ml in PBS, white precipitation formed instantly. In comparison, Glucose-TM at 5 mg/ml and 10 mg/ml in PBS did not form any precipitation – instead, the mixtures remained transparent (Figure 2). Furthermore, both TM and Glucose-TM were added to Dulbecco’s Modified Eagle Media (DMEM) at 5 mg/mL and 10 mg/mL. Like the previous results, TM precipitated out, while Glucose-TM remained dissolved in DMEM (Supplementary Figure 1). These experiments proved that the glycoconjugation strategy for TM did enhance the aqueous solubility significantly.

Figure 2.

Figure 2.

Open in a new tab

Glucose-TM showed increased solubility in PBS but decreased cancer cell cytotoxicity. (A) Solubility tests of Glucose-TM and TM at indicated concentrations in PBS (B) Cell viability of MCF7 and MDA-MB-468 cells after treatment of TM and Glucose-TM for 72 hours.

Glucose-TM maintained SIRT2-selective inhibition in vitro but has poor cellular activity.

After confirming the solubility improvement, Glucose-TM was assayed against the deacetylase activities of SIRT1, 2 and 3 using previously established methods.14,28 Both Glucose-TM and TM impeded SIRT2 deacetylase activity, with IC50 values of 0.019 and 0.093 μM, respectively. In addition, both compounds did not inhibit any SIRT3 activity at 83 μM. Glucose-TM exerted mild inhibition on SIRT1 with IC50 of 7.5 μM, while TM did not inhibit SIRT1 at 83 μM. Nevertheless, the fold of SIRT2 selectivity over SIRT1 was about 400 for Glucose-TM, suggesting that Glucose-TM still inhibits SIRT2 selectively over other sirtuins.

With its SIRT2-selective inhibition in vitro, we tested Glucose-TM in cellular proliferation assays, hoping to see an improvement in cytotoxicity. However, Glucose-TM did not exert any impediment in cellular proliferation, much worse than TM (Figure 2B). Using DataWarrior Software, the simulated topological polar surface area of Glucose-TM was about 260. Compounds with topological polar surface area higher than 140 often portray poor cellular permeability.29,30 Also, glucose transporters might not recognize the glycoconjugated compound as a substrate31, rendering the active transport process inefficient. These factors may lead to very poor cellular permeability/availability of Glucose-TM.

We next attempted to test whether the poor cellular cytotoxicity of Glucose-TM arises from its poor cellular permeability/availability. Traditional Caco-2 permeability assay or PAMPA transwell assay do not work for such compounds because the hydrophobicity of the thiomyristoyl groups make these compounds readily bind to the plastic surfaces, membranes of the transwell plates, and cellullar membranes. Thus, to better reflect the overall cellular permeability/availability of such compounds, we treated MCF7 cells with the compounds for 6 hours, and then collected the cells, extracted the compound, detected and quantified using LC-MS. Only approximately 0.085 μg/million cells of Glucose-TM was detected in the treated cells. Meanwhile, approximately 0.30 μg/million cells of TM was detected, which was about 3.5 fold greater than that of Glucose-TM. The decreased cellular permeability/availability plus the decreased SIRT2 inhibition potency of Glucose-TM could explain why the compound did not exert much cytotoxicity.

Crystal structure of SIRT2(56-356) with Glucose-TM and NAD+.

To guide further development of TM-based inhibitors, we decided to obtain the crystal structure of SIRT2 in complex with Glucose-TM. Our previous attempts to crystalize SIRT2 in complex with TM all failed due to the poor aqueous solubility of TM. Co-crystals of SIRT2(56-356), NAD+, and glucose-TM were obtained, and the structure was solved and refined to 2.45-Å resolution. The overall structure is similar to previously-determined structures of SIRT2.33 In the structure, the NAD+ and Glucose-TM have reacted to form a covalent intermediate at the reaction site cleft (Figure 3A). The electron density at the active site indicates a 3-way junction between the lysine, the myristoyl, and the N-ribose of the ADP-ribose, supporting the mechanism-based nature of this class of thiomyristoyl SIRT2 inhibitors (Figure 3B and Supplemental Figure 2).

Figure 3.

Figure 3.

Open in a new tab

Crystal structure of SIRT2 in complex with Glucose-TM (PDB 6NR0). (A) Overall structure. Two copies of SIRT2(56-356) are found in the asymmetric unit. The Glucose-TM binds at the active site cleft between the Rossmann and Zn2+-binding domains in each SIRT2 molecule. (B) Glucose-TM at the active site. The 2FO-FC map at 1.0 σ shows continuous electron density for the head group, lysine, myristoyl, and 1’-SH-ADP-ribose. (C) The modeled head group of the Glucose-TM hydrogen bonds at the peptide binding site. 2FO-FC map is shown at 1.0 σ. (D) Close-up of the captured Intermediate III with 3’-O-myristoyl lysine linkage. The 2FO-FC map is shown at 1.5 σ. The N-ribose stacks below Phe96 and the 2’-OH hydrogen bonds with His187.

The thiomyristoyl lysine moiety of Glucose-TM binds at a cleft between residues 235-237 and 267-268 (Figure 3C). There is strong electron density from the C1 of the phenyl ring to the carbonyl adjacent to the PEG linker. The direction of the peptide backbone was determined by the rounded density for the phenyl ring, the longer continuous density toward the PEG linker, the shape of the backbone, and by comparison to the backbone in previously-reported peptide-bound SIRT2 structures.33 Directly analogous to the binding of peptide at this site, Glucose-TM forms hydrogen bonds to the main chain of SIRT2, with the lysine carbonyl binding to the backbone nitrogen of Gly236 (3.3 Å O-N distance), the lysine nitrogen binding to the carbonyl of Glu237 (3.0 Å), and the phenyl amide nitrogen binding to the backbone carbonyl of Gln267 (3.3 Å).

Beyond the hydrogen bonding, we also observed that the phenyl ring of Glucose-TM is within ~4 Å of the phenyl ring of Phe235 (6 Å center-to-center distance). This is near the limit for potential “pi-pi” or hydrophobic interactions. Though some interaction may be possible, the electron density for the aniline ring does not suggest a strong stable conformation of the ring, at least at this resolution. Additionally, there is little electron density beyond the beginning of the PEG linker. Hence, the rest of the PEG linker and glucose were not modeled, and they likely do not contribute to the binding affinity.

This information suggests that it is really the thiomyristoyl lysine moiety of the TM-based inhibitors that makes the binding contributions to SIRT2, leaving the periphery of the inhibitor structure open for modifications to improve other pharmacological properties of the inhibitors, such as solubility.

Interestingly, the 2Fo-Fc electron density map around the N-ribose (Figure 3D) suggests formation of a 2’-O myristoyl inhibitor linkage (Intermediate III, see Figure 1A), rather than the 1’-SH-thioalkylimidate linkage (Intermediate I) as originally predicted for this mechanism-based inhibitor.14 This observation of a 2’-O linkage was also made for a peptide-based thiomyristoyl SIRT2 inhibitor.33 Though it is possible that a small portion of asymmetric units contain a 1’-S linkage, several features of the structure support the 2’-O linkage as the dominant species in the crystal. First, the concave shape of the electron density around the ribose can be seen at this resolution, with the 1’, 2’, and 3’ groups pointing up and left in Figure 3D. Second, the 3’-OH is documented in several other sirtuin structures to form a hydrogen bond with His187. If we model Intermediate I in the structure, the electron density in this structure indicates that the ribose would have to be flipped, due to the clear position of the myristoyl linkage.34 In this case, the hydrogen bond with His187 would be lost and the concave shape of the electron density would not fit the ribose as well. Indeed, in the SIRT3 structure with intermediate I, the ribose is flipped34. After modeling and refining the structure both ways, intermediate III better fits our data and the recent structure with BHJH-TM1.33

The observation of intermediate III raises the question of why this is the structure seen rather than intermediate I. It was previously hypothesized that intermediate I would be the stalled intermediate likely because the S-alkylamidate is more stable than an O-alkylamidate, which would form with regular substrate. However, the structure actually captured an O-alkylamidate intermediate III. Perhaps a sulfhydryl at the 1’ position interferes sterically with nucleophilic attack of water to form intermediate IV. Or, perhaps a hydroxyl at position 1’ is important in abstracting a proton from the attacking water to form intermediate IV, while a sulfydryl is not as efficient. It could also be that the crystallization condition required for co-crystallization favors the later intermediate III rather than Intermediate I.

Development of new SIRT2 inhibitors based on the structure of SIRT2 in complex with Glucose-TM.

As seen from the crystal structure of Glucose-TM in complex with SIRT2, neither end of the thiomyristoyl lysine structure contributes much to the binding to SIRT2. Thus, we decided to modify the N-terminal and C-terminal groups of TM to further test this. An additional goal is to obtain SIRT2 inhibitors that are similar to TM in potency, selectivity, and cellular activity, but with improved solubility and perhaps even metabolic stability for future animal and clinical studies. Using TM as a reference structure, we removed the Cbz protecting group (NH3), replaced Cbz with a β-alanine (NH3-6), removed the C-terminal aniline (NH4-3), removed Cbz and replaced the phenyl amide with a methyl ester (NH4-4) introduced a trimethylamine group on the C-terminal to improve aqueous solubility (NH4-6), and lastly, added methyl ester group on the C-terminal (NH4-8) (Figure 4A).

Figure 4.

Figure 4.

Open in a new tab

New TM analogs synthesized based on the structural information and their properties. (A) Synthesized TM analogs and their IC50 (μM) values for SIRT1-3. (B) Cell viability of MCF7 and MDA-MB-231 cells after treatment of TM, NH4-6 and NH4-8 for 72 hours. (C) Calculated concentrations (μg/million cells) of TM and NH4-6 detected from MCF7 cell permeability assay. MCF7 cells were treated with 25 and 50 μM of indicated inhibitors for 6 hours. (D) Soft agar colony formation assays of MCF7 cells treated with inhibitors at indicated concentrations for 10 days.

Interestingly, all six compounds inhibited SIRT2 deacetylase at IC50 values lower than 0.5 μM (Figure 4A). Compounds without the C-terminal phenyl amide, NH4-3, NH4-6 and NH4-8 inhibit SIRT2 at much lower IC50 at 0.012, 0.032 and 0.018 μM, respectively. These values are slightly better than that for TM, suggesting that the C-terminal aniline does not contribute to the inhibition and can be replaced. This is consistent with the structure of SIRT2 in complex with Glucose-TM. In contrast, compared to TM, compounds without the Cbz group at the N-terminal (NH3, NH3-6, NH4-4) all showed slightly worse IC50 values for SIRT2 (~0.4 μM). This suggests that the Cbz group contributes a little to the inhibition efficiency. Because the structure of Glucose-TM did not have the Cbz group, the crystal structure of SIRT2 with Glucose-TM could not reveal the interaction between Cbz and SIRT2.

We next examined the specificity of the new SIRT2 inhibitors. All the new compounds, except NH4-6, did not inhibit SIRT3 even at 83 μM (Figure 4A). For SIRT1, at 83 μM, three compounds did not inhibit SIRT1 activity (Figure 4A). NH3-6, NH4-3 and NH4-6 had IC50 for SIRT1 of 12, 4.4 and 3 μM, respectively. Nevertheless, even these three compounds still showed strong selectivity towards SIRT2 (NH3-6, NH4-3 and NH4-6 showed 30-, 366-, and 94-fold SIRT2 selectivity over SIRT1, respectively). Furthermore, NH3, NH3-6 and NH4-8 did not exert any inhibition against SIRT6 demyristoylase, while NH4-3, NH4-4 and NH4-6 demonstrated very weak inhibitory behaviors against SIRT6 with IC50 of ~50 μM. Considering all the IC50 values, these five TM analogues all displayed SIRT2-selective inhibition, supporting that the thiomyristoyl lysine is essential, but the N- and C-terminal are unimportant or less important.

We have chosen two compounds, NH4-6 and NH4-8, to evaluate their effects on the viability of MCF7 and MBA-MB-231 breast cancer cells. NH4-8 was designed to act as “semi” pro-drug. It can inhibit SIRT2 by itself but can also transform to NH4-3 by non-specific esterases in the cells. After the transformation, NH4-3 with carboxylic acid moiety, which often shows poor cellular permeability, can be trapped inside of the cell. NH4-6 contains charged trimethylamine group, which increases aqueous solubility significantly (simulated cLogP of 3.86). Through improved aqueous solubility, more compounds will remain dissolved at higher concentrations, which could increase cellular cytotoxicity.

In both MCF7 and MDA-MB-231 cells in 2D cell proliferation assays, treatment of NH4-8 showed a similar trend to that of TM (Figure 4B). NH4-6 interestingly showed a slightly worse effect than that of TM at lower concentrations in both MCF7 and MDA-MB-231 cells. This is most likely because the charged trimethylamine group of NH4-6 increases the overall polar surface area, which impairs the cellular permeability (Figure 4B). However, at higher concentrations, NH4-6 showed stronger cytotoxicity than TM in both MCF7 and MDA-MB-231 cells. This is likely because NH4-6 is more soluble and could reach higher concentrations and thus show stronger cytotoxicity than TM, which has very poor aqueous solubility and will precipitate out at higher concentrations.

To test this hypothesis, we treated MCF7 cells with TM and NH4-6 at 25 and 50 μM, and then detected the amount of inhibitors in cells (Figure 4C). The levels of TM detected in the cells were about 0.29 and 0.30 μg/million cells after 6 hours of 25 and 50 μM treatment, respectively. Interestingly, the concentrations of TM in the cells did not change significantly between the two tested concentrations. This shows why there was no significant difference in cellular proliferation between 25 and 50 μM of TM in both MCF7 and MDA-MB-231 cells. The concentrations of NH4-6 inside the cells were about 0.19 and 0.86 μg/million cells after treatment with 25 and 50 μM, respectively. Thus, at 25 μM, the concentration of NH4-6 in the cells was slightly lower than TM, while at 50 μM treatment, the concentration of NH4-6 in the cells was about 2.8 times higher than that of TM. This explains why NH4-6 had weaker cytotoxicity than TM at 25 μM but stronger cytotoxicity at 50 μM than TM. Simply put, at higher concentrations, more NH4-6 was dissolved in the media and had entered the cells, which led to increased cytotoxicity, while TM was already saturated at 25 μM and thus higher amount of TM did not result in higher cytotoxicity.

Next, we examined the effect of TM and NH4-6 against anchorage-independent growth, a cancer cell-specific phenotype, through soft agar colony formation assay (Figure 4D).35 We specifically chose NH4-6 to test, because it not only showed improved solubility but also improved cytotoxicity in the cellular proliferation assays at higher concentration. MCF7 cells were suspended in soft agar mixture with various concentrations of inhibitors for 10 days to observe the colony formation. At 12 μM treatment, TM-treated cells formed many colonies, while cells treated with NH4-6 did not have any colonies at all. This suggests that NH4-6 has stronger effects than TM against anchorage-independent growth of cancer cells.

To further confirm the cellular permeability/availability trend and on-target effect of TM, NH4-6 and NH4-8, we evaluated acetylation of ɑ-tubulin as SIRT2’s target in cells, confirmed by previous reports.36 We treated MCF7 cells with either DMSO, TM, NH4-6, or NH4-8 at indicated concentrations for 6 hours, and monitored acetylated ɑ-tubulin by immunofluorescence. As expected, acetylation levels of ɑ-tubulin were increased upon treatment of TM at all concentrations. Similar trend was observed for NH4-8. Treatment with 25 μM of NH4-6 did not increase the acetylation levels of α-tubulin, but 50 and 100 μM treatment of NH4-6 increased the levels, comparable to that of TM treated samples. This result is consistent with the cellular permeability/ availability assay results described above.

Next, we examined whether TM, NH4-6 and NH4-8 inhibit SIRT1 in cells by looking at acetylated p53 (lysine 382), a previously reported SIRT1 substrate.37 Consistent with previously reported results, treatment of EX527 significantly increased the acetylation level of p53, compared to that of DMSO control.34,38 At all the tested concentrations, TM and NH4-8 did not increase acetylation levels of p53. This was consistent with in vitro SIRT1 IC50 values of TM (>83 μM) and NH4-8 (~50 μM).

Because the IC50 value of NH4-6 for SIRT1 is 3.0 μM, we expected NH4-6 to increase the acetylation levels of p53. However, 10, 25 and 50 μM of NH4-6 did not increase acetylation levels of p53, but 100 μM of NH4-6 did. Thus, at lower concentrations, NH4-6 could not inhibit SIRT1, but its improved solubility allows more NH4-6 to accumulate in cells to inhibit SIRT1 at 100 μM. It is possible that simultaneous inhibition of SIRT1, 2 and 3 may have contributed to improved cytotoxicity of NH4-6 at higher concentrations.

Conclusion

In conclusion, we showed that the glycoconjugation strategy on TM lead to a SIRT2 inhibitor Glucose-TM with enhanced aqueous solubility. Although it has poor cell permeability, the high-water solubility allowed us to obtain a co-crystal structure with SIRT2. Crystallography data suggested that the C-terminal phenyl amide and N-terminal Cbz groups are not required for the interaction between the inhibitor and SIRT2. Several TM analogues with various modifications on both N and C-terminals further confirmed and refined this hypothesis. The finding has led to SIRT2 inhibitors with improved solubility and will facilitate the introduction of even further modifications to the thiomyristoyl lysine core to obtain SIRT2-selective inhibitors with improved chemical and biological characteristics. For example, comparing NH4-6 to TM or NH4-8, our results showed that the positively charged quaternary ammonium is responsible for SIRT1 and SIRT3 inhibition. Thus, future work to improve selectivity for SIRT2 would be to replace the quaternary ammonium or change its relative position by incorporating a PEG linker. Improved SIRT2-specific inhibitors will find important applications given the interesting biological functions of SIRT2 and the potential of SIRT2 inhibitors in treating cancer and neurological diseases.

Supplementary Material

Supplemental Info

Figure 5.

Figure 5.

Open in a new tab

Evaluation of TM derivatives on inhibition of SIRT1 and SIRT2 in cancer cells. (A) Immunofluorescence detection of acetyl α-tubulin (K40) in MCF7 cells treated with indicated concentrations of DMSO, TM, NH4-6 and NH4-8. (B). Immunoblots for the acetylation of p53 (K382) in MCF7 cells treated with indicated concentrations of DMSO, TM, NH4-6 and NH4-8.

Acknowledgments

Funding Sources

The Berkeley Center for Structural Biology is supported in part by the Howard Hughes Medical Institute. The Advanced Light Source is a Department of Energy Office of Science User Facility under Contract No. DE-AC02-05CH11231. This work is based upon research conducted at the Northeastern Collaborative Access Team beamlines, which are funded by the National Institute of General Medical Sciences from the National Institutes of Health (P30 GM124165). The Pilatus 6M detector on 24-ID-C beam line is funded by a NIH-ORIP HEI grant (S10 RR029205). This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357. This work had use of the Cornell University NMR facility, which is supported, in part, by the NSF through MRI award CHE-1531632.

Footnotes

ASSOCIATED CONTENT

Supporting Information. Additional Information on methods, NMR spectra of synthesized compounds, Table S1, Supplementary Figure 1–3, and Supplementary Scheme 1–4. The material is available free of charge via the internet at http://pubs.acs.org.

REFERENCES

  • (1).Feldman JL; Dittenhafer-Reed KE; Kudo N; Thelen JN; Ito A; Yoshida M; Denu JM Kinetic and Structural Basis for Acyl-Group Selectivity and NAD+ Dependence in Sirtuin-Catalyzed Deacylation. Biochemistry 2015, 54, 3037–3050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (2).Teng Y-B; Jing H; Aramsangtienchai P; He B; Khan S; Hu J; Lin H; Hao Q Efficient Demyristoylase Activity of SIRT2 Revealed by Kinetic and Structural Studies. Sci. Rep 2015, 5, 8529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (3).Cui Y; Li X; Lin J; Hao Q; Li XD Histone Ketoamide Adduction by 4-Oxo-2-nonenal Is a Reversible Posttranslational Modification Regulated by Sirt2. ACS Chem. Biol 2017, 12, 47–51. [DOI] [PubMed] [Google Scholar]
  • (4).Du J; Zhou Y; Su X; Yu JJ; Khan S; Jiang H; Kim J; Woo J; Kim JH; Choi BH; He B; Chen W; Zhang S; Cerione RA; Auwerx J; Hao Q; Lin H Sirt5 Is a NAD-Dependent Protein Lysine Demalonylase and Desuccinylase. Science 2011, 334, 806–809. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (5).Pannek M; Simic Z; Fuszard M; Meleshin M; Rotili D; Mai A; Schutkowski M; Steegborn C Crystal structures of the mitochondrial deacylase Sirtuin 4 reveal isoform-specific acyl recognition and regulation features. Nat. Commun 2017, 8, 1513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (6).North BJ; Verdin E Interphase nucleo-cytoplasmic shuttling and localization of SIRT2 during mitosis. PLoS One 2007, 2, e784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (7).Yang X; Park SH; Chang HC; Shapiro JS; Vassilopoulos A; Sawicki KT; Chen C; Shang M; Burridge PW; Epting CL; Wilsbacher LD; Jenkitasemwong S; Knuston M; Gius D; Ardehail H Sirtuin 2 regulates cellular iron homeostasis via deacetylation of transcription factor NRF2. J. Clin. Invest 2017, 127, 1505–1516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (8).Zhou W; Ni TK; Wronski A; Glass B; Skibinski A; Beck A; Kuperwasser C The SIRT2 Deacetylase Stabilizes Slug to Control Malignancy of Basal-like Breast Cancer. Cell Rep. 2016, 17, 1302–1317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (9).Park SH; Ozden O; Liu G; Song HY; Zhu Y; Yan Y; Zou X; Kang HJ; Jiang H; Principe DR; Cha YI; Roh M; Vassilopoulos A; Gius D SIRT2-Mediated Deacetylation and Tetramerization of Pyruvate Kinase Directs Glycolysis and Tumor Growth. Cancer Res. 2016, 76, 3802–3812. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (10).Zhao D; Zou SW; Liu Y; Zhou X; Mo Y; Wang P; Xu YH; Dong B; Xiong Y; Lei QY; Guan KL Lysine-5 acetylation negatively regulates lactate dehydrogenase A and is decreased in pancreatic cancer. Cancer Cell 2013, 23, 464–476. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (11).Herskovits AZ; Guarente L Sirtuin deacetylases in neurodegenerative diseases of aging. Cell Res. 2013, 23, 746–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (12).Outeiro TF; Kontopoulos E; Altmann SM; Kufareva I; Strathearn KE; Amore AM; Volk CB; Maxwell MM; Rochet J-C; McLean PJ; Young AB; Abagyan R; Feany MB; Hyman BT; Kazantsev AG Sirtuin 2 inhibitors rescue alpha-synuclein-mediated toxicity in models of Parkinson’s disease. Science 2007, 317, 516–519. [DOI] [PubMed] [Google Scholar]
  • (13).Liu PY; Xu N; Malyukova A; Scarlett CJ; Sun YT; Zhang XD; Ling D; Su SP; Nelson C; Chang DK; Koach J; Tee AE; Haber M; Norris MD; Toon C; Rooman I; Xue C; Cheung BB; Kumar S; Marshall GM; Biankin AV; Liu T. The histone deacetylase SIRT2 stabilizes Myc oncoproteins. Cell Death Differ. 2013, 20, 503–513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (14).Jing H; Hu J; He B; Negrón Abril YL; Stupinski J; Weiser K; Carbonaro M; Chiang Y-L; Southard T; Giannakakou P; Weiss RS; Lin H A SIRT2-Selective Inhibitor Promotes c-Myc Oncoprotein Degradation and Exhibits Broad Anticancer Activity. Cancer Cell 2016, 29, 297–310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (15).Jing H; Zhang X; Wisner SA; Chen X; Spiegelman NA; Linder ME; Lin H SIRT2 and lysine fatty acylation regulate the transforming activity of K-Ras4a. eLife 2017, 6 e32436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (16).Kozako T; Mellini P; Ohsugi T; Aikawa A; Uchida YI; Honda SI; Suzuki T Novel small molecule SIRT2 inhibitors induce cell death in leukemic cell lines. BMC Cancer 2018, 18, 791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (17).Suzuki T; Khan MN; Sawada H; Imai E; Itoh Y; Yamatsuta K; Tokuda N; Takeuchi J; Seko T; Nakagawa H; Miyata N Design, synthesis, and biological activity of a novel series of human sirtuin-2-selective inhibitors. J. Med. Chem 2012, 55, 5760–5773. [DOI] [PubMed] [Google Scholar]
  • (18).Mellini P; Itoh Y; Tsumoto H; Li Y; Suzuki M; Tokuda N; Kakizawa T; Miura Y; Takeuchi J; Lahtela-Kakkonen M; Suzuki T Potent mechanism-based sirtuin-2-selective inhibition by an in situ-generated occupant of the substrate-binding site, “selectivity pocket” and NAD(+)-binding site. Chem. Sci 2017, 8, 6400–6408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (19).Rumpf T; Schiedel M; Karaman B; Roessler C; North BJ; Lehotzky A; Olah J; Ladwein KI; Schmidtkunz K; Gajer M;l Pannek M; Steegborn C; Sinclair DA; Gerhardt S; Ovadi J; Schutkowski M; Sipple W; Einsle O; Jung M Selective Sirt2 inhibition by ligand-induced rearrangement of the active site. Nat. Commun 2015, 6, 6263. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (20).Lain S; Hollick JJ; Campbell J; Staples OD; Higgins M; Aoubala M; McCarthy A; Appleyard V; Murray KE; Baker L; Thompson A; Mathers J; Holland SJ; Stark MJ; Pass G; Woods J; Lane DP; Westwood NJ Discovery, in vivo activity, and mechanism of action of a small-molecule p53 activator. Cancer Cell 2008, 13, 454–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (21).Mellini P; Itoh Y; Elboray EE; Tsumoto H; Li Y; Suzuki M; Takahashi Y; Tojo T; Kurohara T; Miyake Y; Miura Y; Kitao Y; Kotoku M; Iida T; Suzuki T Identification of Diketopiperazine-Containing 2-Anilinobenzamides as Potent Sirtuin 2 (SIRT2)-Selective Inhibitors Targeting the “Selectivity Pocket”, Substrate-Binding Site, and NAD(+)-Binding Site. J. Med. Chem 2019, 62, 5844–5862. [DOI] [PubMed] [Google Scholar]
  • (22).Pohl J; Bertram B; Hilgard P; Nowrousian MR; Stüben J; Wießler M D-19575—a sugar-linked isophosphoramide mustard derivative exploiting transmembrane glucose transport. Cancer Chemother. Pharmacol 1995, 35, 364–370. [DOI] [PubMed] [Google Scholar]
  • (23).Stüben J; Port R; Bertram B; Bollow U; Hull WE; Schaper M; Pohl J; Wiessler M Pharmacokinetics and whole-body distribution of the new chemotherapeutic agent β- D -glucosylisophosphoramide mustard and its effects on the incorporation of [methyl- 3 H]-thymidine in various tissues of the rat. Cancer Chemother. Pharmacol 1996, 38, 355–365. [DOI] [PubMed] [Google Scholar]
  • (24).Mikuni K; Nakanishi K; Hara K; Hara K; Iwatani W; Amano T; Nakamura K; Tsuchiya Y; Okumoto H; Mandai T In vivo antitumor activity of novel water-soluble taxoids. Biol. Pharm. Bull 2008, 31, 1155–1158. [DOI] [PubMed] [Google Scholar]
  • (25).Spiegelman NA; Hong JY; Hu J; Jing H; Wang M; Price IR; Cao J; Yang M; Zhang X; Lin H A Small-Molecule SIRT2 Inhibitor That Promotes K-Ras4a Lysine Fatty-Acylation. ChemMedChem 2019, 14, 744–748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (26).Farooqi AS; Hong JY; Cao J; Lu X; Price IR; Zhao Q; Kosciuk T; Yang M; Bai JJ; Lin H Novel Lysine-Based Thioureas as Mechanism-Based Inhibitors of Sirtuin 2 (SIRT2) with Anticancer Activity in a Colorectal Cancer Murine Model. J. Med. Chem 2019, 62, 4131–4141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (27).Milla P; Dosio F; Cattel L PEGylation of Proteins and Liposomes: a Powerful and Flexible Strategy to Improve the Drug Delivery. Curr. Drug Met 2012, 13, 105–119. [DOI] [PubMed] [Google Scholar]
  • (28).Spiegelman NA; Price IR; Jing H; Wang M; Yang M; Cao J; Hong JY; Zhang X; Aramsangtienchai P; Sadhukhan S; Lin H Direct Comparison of SIRT2 Inhibitors: Potency, Specificity, Activity-Dependent Inhibition, and On-Target Anticancer Activities. ChemMedChem 2018, 13, 1890–1894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (29).Shultz MD Two Decades under the Influence of the Rule of Five and the Changing Properties of Approved Oral Drugs. J. Med. Chem 2019, 62, 1701–1714. [DOI] [PubMed] [Google Scholar]
  • (30).Pajouhesh H; Lenz GR Medicinal chemical properties of successful central nervous system drugs. NeuroRx 2005, 2, 541–553. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (31).Patra M; Awuah SG; Lippard SJ Chemical Approach to Positional Isomers of Glucose–Platinum Conjugates Reveals Specific Cancer Targeting through Glucose-Transporter-Mediated Uptake in Vitro and in Vivo. J. Am. Chem. Soc 2016, 138, 12541–12551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (32).Chong S; Dando SA; Soucek KM; Morrison RA In vitro permeability through caco-2 cells is not quantitatively predictive of in vivo absorption for peptide-like drugs absorbed via the dipeptide transporter system. Pharm. Res 1996, 13, 120–123. [DOI] [PubMed] [Google Scholar]
  • (33).Wang Y; Fung YME; Zhang W; He B; Chung MWH; Jin J; Hu J; Lin H; Hao Q Deacylation Mechanism by SIRT2 Revealed in the 1’-SH-2’-O-Myristoyl Intermediate Structure. Cell Chem. Biol 2017, 24, 339–345. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (34).Gertz M; Fischer F; Nguyen GT; Lakshminarasimhan M; Schutkowski M; Weyand M; Steegborn C Ex-527 inhibits Sirtuins by exploiting their unique NAD+-dependent deacetylation mechanism. Proc. Natl. Acad. Sci. U. S. A 2013, 110, E2772–2781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (35).Mori S; Chang JT; Andrechek ER; Matsumura N; Baba T; Yao G; Kim JW; Gatza M; Murphy S; Nevins JR Anchorage-independent cell growth signature identifies tumors with metastatic potential. Oncogene 2009, 28, 2796–2805. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (36).North BJ; Marshall BL; Borra MT; Denu JM; Verdin E The human Sir2 ortholog, SIRT2, is an NAD+-dependent tubulin deacetylase. Mol. Cell 2003, 11, 437–444. [DOI] [PubMed] [Google Scholar]
  • (37).Vaziri H; Dessain SK; Ng Eaton E; Imai SI; Frye RA; Pandita TK; Guarente L; Weinberg RA hSIR2(SIRT1) functions as an NAD-dependent p53 deacetylase. Cell 2001, 107, 149–149. [DOI] [PubMed] [Google Scholar]
  • (38).Solomon JM; Pasupuleti R; Xu L; McDonagh T; Curtis R; DiStefano PS; Huber LJ Inhibition of SIRT1 catalytic activity increases p53 acetylation but does not alter cell survival following DNA damage. Mol. Cell Biol 2006, 26, 28–38. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Info
NIHMS1061122-supplement-Supplemental_Info.pdf (2.5MB, pdf)

RESOURCES