Environment-sensitive fluorescent inhibitors of histone deacetylase
Keywords: Histone deacetylase HDACs inhibitor Environmental sensitivity Fluorescence inhibitor Cell imaging
Abstract
Histone deacetylases (HDACs) are proteases that can catalyze the deacetylation of histones to inhibit gene transcription. Since mutations and/or aberrant expression of various HDACs are frequently associated with human diseases, particularly cancers, HDACs are important therapeutic targets for many human tumors. However, there are still relatively few studies on HDAC small molecule fluorescent probes. Herein, we designed and synthesized a class of environment-sensitive fluorescent inhibitors with a switch mechanism to study HDAC activity. In vitro, the enzyme inhibition activity of compound 6b was comparable to the positive control drug SAHA, and it presented suitable imaging in living cells and tumor-tissue slices. This environment-sensitive fluorescent inhibitor provides a new idea for the diagnosis and treatment of HDACs-related diseases.
The acetylation of histones is essential for epigenetic regulation, and post-translational modifications of histones include acetylation, me- thylation, phosphorylation and ubiquitination [1,2]. Allefrey et al. have demonstrated the relationship between histone acetylation levels and cellular transcriptional activity in 1964 [3]. Histone acetyltransferases (HATs) catalyze the acetylation of the ε-NH2 group of histone lysine residues [4]. In contrast, histone deacetylases (HDACs) catalyze the deacetylation of histones [5]. The acetylation of histones loosens the chromatin structure, leading to transcriptional activation, and the deacetylation of histones concentrates the chromatin structure, re- sulting in inhibition of gene expression [6]. This is a dynamic equili- brium process in normal cells; however, in many cancer cells, this balance is broken, and the activity of histone deacetylases is sig- nificantly increased, resulting in decreased activity of tumor suppressor genes and uncontrolled cell proliferation. Mutations or aberrant ex- pression of HDACs is often observed in tumor cells. Studies have shown that lysine in cancer cells is subjected to histone H4 deacetylation, suggesting that HDACs can become an important therapeutic target for many malignancies [7]. Many studies have shown that HDACs mediate transcriptional inhibition by overexpression [8,9] or abnormal inter- action of transcriptional regulators [10–12]. Moreover, different HDACs have been shown to have cell-environment-specific effects and to play different roles in pro- or anti-cancer activities [13].
At present, probes for HDACs are mainly divided into radio-labeled probes and fluorescent probes, in which radio-labeled probes are mainly used for detecting the biodistribution and Positron Emission Tomography (PET) imaging of HDAC inhibitors in vivo. [18F]-FAHA can
be used to detect the activity of class II HDACs in animals [14,15], and can also be used for PET imaging in rat breast cancer, human glio- blastoma multiforme and cerebral palsy. [11C]-Martinostat is an ima- ging probe that is selective for class I HDACs, which firstly showed gene changes in brain activity in living human brains [16]. Fluorescent probes are mainly used to detect the activity and cell imaging of HDACs. The existing HDAC fluorescent probes are mainly based on the mechanism of enzymatic reaction. The enzymatic reaction deacetylates the probes, leading to release of the fluorophores [17–19], and ag- gregation-induced emission [20], increasing the ability to bind to DNA [21], or stimulating intramolecular self-crosslinking reactions [22,23], in order to achieve the aim of fluorescent switching. In recent years, various researches on tumor diagnosis and treatment have emerged increasingly. However, the probes of HDACs are mainly used for the enzyme activity detection and cell imaging. HDACs are the important target for malignant tumors, but, due to lack of researches on the in- tegration of HDAC diagnosis and treatment of tumors, researches on HDACs should be carried out as soon as possible.
Most HDAC inhibitors consist of three components: surface re- cognition region (Cap structure), zinc ion chelation group (ZBG struc- ture) and Linker [24]. We designed a series of environment-sensitive fluorescent inhibitors with a switch mechanism (Fig. 1. B), and selected SAHA – the most commonly used hydroXamic acid inhibitor of HDAC – as the recognition motif. Considering that fluorophores in small volume can minimally affect the binding affinity of the parent ligand, a rela- tively small environment-sensitive fluorophore SBD [25] is designed to be incorporated into the structure of SAHA (Fig. 1. A). In classic design strategies, the fluorophore is not involved in the binding of the target protein, so that the binding activity of the parent ligand is somewhat reduced. While in our design strategy, SAHA was used as lead comophores in the structure of such fluorescent inhibitors are among the pharmacophores, thereby avoiding the effect of fluorophores on the binding of probes to target proteins in conventional design strategies.
Fig. 1. (A) The design strategy of compounds; (B) The structures of four compounds.
Through various analyses of the crystal structure of HDAC [26–29], it has been found that there is a hydrophobic channel in the active region of HDAC. Based on the environment-sensitive fluorophore, the fluorescent probe can detect the hydrophobic pocket. Therefore, when a compound enters a low-polarity HDAC active site from a highly polar aqueous environment, the fluorescence intensity changes significantly, which constitutes of a fluorescent switch mechanism.
Spectroscopic properties of the Compounds
The results displayed that compounds possessed excellent fluor- escent properties (Figs. S1–S3). The maximum ultraviolet absorption wavelengths of all compounds were about 440 nm, and the maximum excitation wavelengths about 440 nm and the maximum emission wa- velengths about 600 nm. The relative fluorescence quantum yield of the compounds in the Phosphate Buffer Saline (PBS) solution (highly polar) was extremely low at an almost quenching level; but the quantum yield in the acetonitrile solution (low polarity) was increased by more than 100 times (Table 1), which preliminarily proved that four compounds were all environment-sensitive. The fluorescence intensity of com-
fluorescence intensity increased significantly with the decrease in the polarity of the solution and the increase in the viscosity, further con- firming that compounds were environment-sensitive, mainly for the polarity and viscosity of the environment.
HeLa cell extract inhibition of Compounds
Compounds 6a (204 nM) and 6b (108 nM) had the best in vitro inhibitory activities, and compound 6b (108 nM) was slightly better than the positive control drug SAHA (134 nM) (Table 2). The com- parison of compounds 6a and 6b showed that the siX-carbon linker was more conducive to the inhibition of enzyme activity. It also indicated that our design strategy was successful and the introduction of the fluorophore did not reduce the activity of the compounds. The com- parison of compounds 11a and 11b with 6a and 6b indicated that the presence of urea groups reduced the inhibitory activity. The compar- ison of compounds 11a and 11b indicated that the presence of a branched chain at the hydroXamic acid alpha position greatly reduced the activity of the compounds. These were all consistent with the study inhibitors [30–32].
In vitro HDAC Isoform-Selectivity of Compounds
The isoform selectivity of compounds 6a and 6b were essentially identical to the positive control drug SAHA. The result presented that both compounds 6a and 6b were the broad spectrum HDAC inhibitors (Table 3).
In vitro antiproliferative Assay
The experimental results showed that the antiproliferative effect of hydroXamic acid HDAC inhibitors on hematoma was better than on solid tumors. The most sensitive one among the four cell lines is MOLT4 cell (Table 4). The anti-tumor proliferative activities of compounds were inferior to the positive control drug SAHA, but it represented that the compounds were less cytotoXic and more beneficial for cell imaging.
Protein-Binding Assay
We can see (Fig. 2) that the fluorescence intensity increased with the increase in the enzyme concentration until saturation, while the concentration of compound 6b was kept constant. These results in- dicated that compound 6b exhibited its environmental sensitivity at the enzyme level, and had a fluorescent switch function, which might be used for inhibitors screening and determination of binding kinetic parameters.
Flow-Cytometry Analysis
To future explore the antiproliferative mechanism, the MOLT4 cells were treated with compounds 6a, 6b and SAHA, respectively, and then were stained with Annexin V-PE. The results indicated that, as the an- titumor drugs, compounds 6a, 6b and SAHA could induce apoptosis of tumor cells. At the same concentration (5 μM), compounds 6a and 6b induced apoptosis (23.30%, and 20.02%, respectively), and the apop- tosis rate was lower than that induced by SAHA (29.58%) (Fig. 3). The results of apoptosis experiments performed by flow cytometry were consistent with the anti-tumor proliferation activity test in vitro, in- dicating that compounds 6a and 6b were less toXic and more suitable for cell imaging.
Fig. 2. The fluorescence intensity saturation curve of 6b (1 nM) as the enzyme concentration changes.
Hematoxylin-Eosin (HE) and tissue section Staining
To investigate the ability of 6b for imaging in mouse tumor tissue sections, we established a subcutaneous PC-3-Xenograft mouse model where the tumor was paraffin-embedded when reaching about 1 cm in diameter. HematoXylin-Eosin (HE) staining was performed using the sections provided by Qilu Hospital of Shandong University. The results showed that the tumor sections were in significant pathological state compared to normal tissues (Fig. 6).
In Vitro Inhibition of HDACs Isoforms of Compounds.
Fig. 3. Results of MOLT-4 cells apoptosis experiments of 5 μM 6a, 6b and SAHA by flow cytometry analysis. (A) Control; (B) SAHA; (C) 6a; (D) 6b.
Fig. 4. (A) Fluorescence imaging after in- cubation of 5 μM 6b with PC-3 cells (A1: bright field, A2: green channel, A3: merged image.); (B) Fluorescence imaging after in- cubation with 5 μM 6b and MOLT4 cells (B1: bright field, B2: green channel, B3: merged image.); (C) Fluorescence imaging after incubation of 5 μM 6b with HEK293 cells (C1: bright field, C2: green channel, C3: merged image.); Objective lens: 63 × . Scale bar = 67 μm.
Fig. 5. (A) Co-staining results of 5 μM 6b and 2.5 μM cell membrane dye DID for MOLT-4 cells; (B) Co-staining results of 5 μM 6b and 0.25 μM nuclear dye Hoechst 33,342 for MOLT-4 cells.
Fig. 6. HematoXylin − eosin staining of (A) tumor tissue and (B) normal tissue of male nude mice.
To confirm whether our probes could selectively label HDACs and located tumor sites, these tissue sections were subjected to fluorescence imaging measurements. The experimental results showed that, under the same experimental conditions, the fluorescence intensity of tumor tissue sections was much stronger than that of normal tissue sections, and the fluorescence intensity of tumor tissue sections was significantly reduced when co-incubated with inhibitor SAHA (see Fig. 7), indicating that compound 6b could selectively label HDACs and had potential value in clinical cancer diagnosis.
Fig. 7. (A) Fluorescence imaging of tumor tissues incubated with 10 μM 6b. (B) Fluorescence imaging of tumor tissues incubated with 10 μM 6b and 100 μM SAHA. (C) Fluorescence imaging of normal tissues incubated with 10 μM 6b.
We, for the first time, designed, synthesized and evaluated the en- vironment-sensitive fluorescent inhibitors of HDACs, which exhibited good optical properties and environmental sensitivity. In vitro in- hibitory activity and isoform selectivity of the compounds were similar to those of the positive control drug SAHA. The anti-tumor cell pro- liferation activity in vitro of the compounds was slightly lower than that of SAHA, which made the compounds more suitable for cell imaging. The staining with compound 6b was good for cells with high expression of HDACs (MOLT4 and PC-3), while the cells with low expression of HDAC (HEK293) were almost unstained, indicating that 6b had good selectivity. Furthermore, 6b was successfully applied to the detection of tumor tissue sections. The imaging results of tumor tissue and normal tissue showed significant differences, and the addition of inhibitors significantly reduced the imaging effect of tumor tissue, indicating that 6b could selectively identify HDACs and tumor sites. Therefore, we believe that this compound may be used in the pathology and phy- siology studies of HDACs, and provide a new idea for the integration of clinical diagnosis and treatment of HDAC-related tumors.
Supplementary material
Materials and instruments
All reagents were used in this work without further purification unless otherwise specified. Boc-Lys (acetyl)-AMC and HeLa Cell EXtract were all purchased from Bachem AG, Switzerland. The melting points were determined on an electrothermal melting-point apparatus. 1H NMR and 13C NMR of the compounds were obtained on Bruker 400 or 600 MHz NMR spectrometer with TMS as the internal standard. High- resolution mass spectra were performed by Shandong Analysis and Test Center. HPLC examination was performed on an Agilent 1260 HPLC system. Absorption spectra were performed by a PUXI TU-1901 spectrophotometer, and fluorescence spectra were performed by a PerkinElmer EnSight microplate reader.
Fluorescence-Spectroscopy Test
We determined the ultraviolet absorption, fluorescence quantum yield and fluorescence spectra in 10 μM solutions of PBS Buffer (pH 7.4), and studied the environmental sensitivity of the compounds in solvents of different polarities, viscosities and pH. The fluorescent properties of compounds were tested by a PUXI TU-1901 spectro- photometer and a PerkinElmer EnSight microplate reader. More details can be found in the Supporting Information.
HeLa cell extract inhibition of the Compounds
The experimental detail was conducted with reference [33]. The Hela cell extract was used as the enzyme source and the fluorescence analysis method with small error and high sensitivity was adopted. Firstly, the HDACs enzymes, compounds and the fluorescent substrate Boc-Lys (acetyl)-AMC were incubated together, so that the histone deacetylase catalyzed the substrate to deacetylate and form Boc-Lys- AMC; and then Boc-Lys-AMC was treated with trypsin to form the fluorescent group AMC (4-amino-7-methylcoumarin), and the fluores- cence intensity of the substance was measured at 390/460 nm (ex- citation wavelength/emission wavelength) with a microplate reader. Finally, the IC50 value was calculated by GraphPad Prism 5 software.
In vitro HDAC Isoform-Selectivity of the Compounds
We selected HDAC1, HDAC2, HDAC3, HDAC6, HDAC7 and HDAC8 for in vitro inhibition of enzyme activity to further confirm the HDAC isoform selectivity of 6a and 6b. More experimental details were pro- vided in the Supporting Information.
Protein-Binding Assay
In this experiment, we incubated 6b with different concentrations of Hela cell extract and measured the changes in fluorescence intensity at 450/590 nm with a microplate reader (POLARstar Omega), to verify whether 6b could reflect its environment-sensitivity at the enzyme level. More experimental details were provided in the Supporting Information.
In vitro antiproliferative Assay
The cytotoXicity of 6a and 6b was assayed on solid tumors (PC-3 and A549) and hematological tumor (K562 and MOLT4) cell lines by the Cell Counting Kit-8 (CCK-8) method [34]. In brief, solid tumor cell lines (5 × 103 / well) and hematological tumor cell lines (1 × 104 / well) were cultured for 12 h, and then different concentrations of compounds and SAHA were added, respectively. Then, they were in- cubated for another 24 h. Finally, the absorbance was measured at 450 nm two hours after CCK8 miXture was added and IC50 value was calculated by GraphPad Prism 5.
Flow-Cytometry Analysis
The apoptosis rate of MOLT4 cell line induced by the compounds was detected by Annexin V-PE apoptosis assay kits. 2 × 105 cells/well were added to the 6-well plate and incubated for 12 h. Compounds or SAHA (final concentration 5 μM) was then added to the wells and in- cubated for additional 24 h. Then the cells were collected and washed with 1 × PBS for three times and gently re-suspended and counted. 1 × 105 re-suspended cells were centrifugated (6 min, 1200 rmp) and the supernatant was removed. Then 195 μL Annexin V-PE binding so- lution and 5 μL Annexin V-PE were added sequentially and miXed gently. After incubated for about 20 min in dark, the tubes were ana- lyzed by a Beckman flow cytometer.
Cell staining and fluorescence Imaging
Compound 6b with low cytotoXicity was selected for the cell fluorescence imaging. And we selected PC-3 and MOLT4 cells with high expression of HDACs and HEK293 cells with normal expression of HDACs in the experiments, with reference to https://www.proteinatlas. org/search/HDAC. PC-3 and HEK293 cells (1 × 104/mL, 1 mL, DMEM medium + 10% FBS), as well as MOLT4 cells (2 × 104/mL, 1 mL, RPMI-1640 medium + 10% FBS) were inoculated into confocal culture dishes and incubated for 12 h, respectively. Then the medium was as- pirated, and a solution of 6b (5 μM) diluted with serum-free medium was added, followed by incubation for 30 min. Subsequently, these cells were imaged by Zeiss AXio Observer A1 fluorescence microscope (ob- jective lens: 63 × ). Using the same procedure, 6b (5 μM) was co- stained with the commercial dyes DID (2.5 μM) and Hoechst 33,342 (0.25 μM). The fluorescence imaging was captured by a Zeiss LSM780 confocal fluorescence microscope.
Fluorescence staining of mouse tumor and normal tissue Sections PC-3 (2 × 107) cells were seeded under the armpit of the male nude mice. About one month later, the tumor tissues and normal tissues (skin and thigh muscle tissues) of the nude mice were made into paraffin sec- tions for immunostaining experiment. After baking, dewaxing, hydration, and antigen retrieval, the sections were incubated with 6b (10 μM) or 6b (10 μM) and 100 μM SAHA in Krebs solution overnight at 4 °C. On the next day, it was washed once with Krebs solution, and a drop of anti-fluorescence quencher was added. Subsequently,CWI1-2 the imaging was performed with the Olympus VS120 fluorescence upright microscope (40×).