Subscriber access provided by UNIV OF LOUISIANA
Article
Rational design of a long-wavelength fluorescent probe for highly selective sensing carboxylesterase 1 in living systems Zhenhao Tian, Lele Ding, Kun Li, Yunqing Song, Tongyi Dou, Jie Hou, Xiangge Tian, Lei Feng, Guangbo Ge, and Jing-Nan Cui Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b05417 • Publication Date (Web): 10 Apr 2019 Downloaded from http://pubs.acs.org on April 10, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Rational design of a long-wavelength fluorescent probe for highly selective sensing carboxylesterase 1 in living systems Zhenhao Tian,†⊥Lele Ding,†⊥Kun Li,＃Yunqing Song,‡Tongyi Dou,＃Jie Hou,§Xiangge Tian,§Lei Feng,§Guangbo Ge,*‡Jingnan Cui*† †
State Key Laboratory of Fine Chemicals, Dalian University of Technology, Dalian, 116024, China. E-mail:
[email protected] ‡Institute
of Interdisciplinary Integrative Medicine Research, Shanghai University of Traditional Chinese Medicine, Shanghai, 201203, China. E-mail:
[email protected] ＃
School of Life Science and Medicine, Dalian University of Technology, Panjin, 124221, China.
§
Dalian Medical University, Dalian 116044, China.
ABSTRACT: Rational design of practical probes with excellent specificity and improved optical properties for a particular enzyme is always a big challenge. Herein, a practical and highly specific fluorescent probe for carboxylesterase 1 (CES1) was rationally designed using meso-carboxyl-BODIPY as the basic fluorophore based on the substrate preference and catalytic properties of CES1. Following molecular docking-based virtual screening combined with reaction phenotyping-based experimental screening, we found that MMB (probe 7) exhibited the optimal combination of sensitivity and specificity towards human CES1 in contrast to other ester derivatives. Under physiological conditions, MMB could be readily hydrolyzed by CES1 and released MCB, such biotransformation brought great changes in electronic properties at meso-position of the fluorophore and triggered a dramatic increase in fluorescence emission around 595 nm. Moreover, MMB was cell membrane permeable and was successfully applied to monitor the real activities of CES1 in various biological samples including living cells, tissue slices, organs and zebrafish. In summary, this study showed a good example for constructing specific fluorescent probe(s) for a target enzyme, and also provided a practical and sensitive tool for real-time sensing of CES1 activities in complicated biological samples. All these findings would strongly facilitate high-throughput screening of CES1 modulators and the studies on CES1-associated physiological and pathological processes.
INTRODUCTION Optical probes are essential tools in life science for realtime sensing of target biological molecule(s), which strongly facilitate drug discovery and fundamental researches in biological and biomedical related fields.1,2 Especially, deciphering the biological functions of a particular enzyme and its relevance to human diseases requires reliable tools for highly selective and sensitive sensing the real activities of the target enzyme in complex biological systems. Although significant breakthroughs have been made in the development of specific optical probe substrates for target enzyme(s), rational design of practical fluorescent probes with excellent specificity and improved optical properties for a particular enzyme is still a big challenge. Over the past decade, some strategies and approaches (such as computer aided design and virtual screening technique) have been used to rational design optical probe substrates with excellent specificity towards target enzyme(s), and part of them have been successfully used for sensing the target enzymes in complicated biological systems.3 Such achievements strongly encourage the biochemists to develop more practical and specific optical probe substrates for target enzyme(s) and use them to replace
the physiologically relevant substrates for both drug discovery and basic researches. Serine hydrolases (SHs), one of the largest families of known enzymes in mammalian cells, comprising approximately ~240 enzymes in the human proteome.4 SHs can efficiently catalyse the cleavage of a large number of endogenous and xenobiotic substrates with carboxylic acid ester, thioester, phosphate ester, and amide bonds, thus play key roles in a wide range of patho- and physiological processes, such as lipid homeostasis and drug metabolism.5,6 As one of the most abundant SHs distributed in the human liver, CES1 plays crucial roles in the hydrolytic metabolism of many endogenous esters and a vast number of ester drugs, as well as other biological processes (such as trafficking and retention of proteins in the endoplasmic reticulum).7-12 Recent researches have indicated that the abnormal expression or dysfunction of CES1 is closely associated with several metabolic diseases, such as atherosclerosis, hyperlipidemia and obesity.13,14 Therefore, developing practical optical tools for highly sensitive and selective detecting the real activities of CES1 in complex biological systems is necessary, which will be great helpful to decipher the physiological functions of this key enzyme and its relevance to human diseases, as well as useful for high-throughput screening and characterization of
ACS Paragon Plus Environment
Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
CES1 modulators to regulate both endogenous metabolism and the pharmacokinetic behaviors of ester drugs, such as clopidogrel. Scheme 1. The proposed sensing mechanism of MMB for CES1.
contaning biological systems, MMB could be readily hydrolyzed to MCB, such biotransformation triggered a dramatic increase in the fluorescence emission around 595 nm (Scheme 1). Moreover, MMB exhibited ultrahigh sensitivity and excellent specificity towards CES1 over various hydrolases and biomolecules found in the human body. These findings encouraged us to further investigate the performance and applicability of this fluorescence probe for highly specific and sensitive monitoring CES1 in various biological samples. EXPERIMENTAL SECTION
Up to now, several methods including qRT-PCR, proteomic techniques and western blotting have been developed for CES1 detection and quantification. Unfortunately, these methods are time-consuming, fairly complex, and limited by some strict requirements including skilled operators and expensive equipment. Moreover, these routinely used techniques can only assess the mRNA or protein levels of target enzyme(s) rather than its real function in biological systems.15-17 Compared with these routinely used techniques, fluorescence-based biochemical assays have drawn increasing attention for their many inherent advantages, such as ultrahigh sensitive, nondestructive and good applicability to highthroughput screening, as well as being compatible for various biological samples.1,18-32 Unfortunately, constructing practical fluorescent probes capable of sensitive and specific detecting CES1 activities in complex biological samples is still a big challenge, due to most of fluorophores are polycyclic phenol compounds, whose corresponding ester derivatives prefer to be metabolized by carboxylesterase 2 (CES2) or by multiple members of the serine hydrolase superfamily.33-36 Although an isoform-specific fluorescence probe (termed BMBT) for CES1 has been developed by us, the emission wavelength was too short to use it for detecting CES1 activities in complex biological samples.37 Therefore, it is necessary to construct more practical fluorescent probes for monitoring mammalian CES1, which may strongly facilitate real-time monitoring CES1 activities and high-throughput screening of CES1 modulators in living systems. This study aimed to design and develop a novel longwavelength fluorescence probe for highly specific and sensitive monitoring CES1 in complex biological samples. To this end, meso-carboxyl-BODIPY was used as a basic fluorophore on the basis of its desirable spectroscopic properties and the catalytic features of CES1. From the view of the chemical structure, the ester derivatives of mesocarboxyl-BODIPY contained a large carboxyl group and a small alcohol group, which was highly consistent with the substrate preference of CES1. Furthermore, the fluorescence emission and fluorescence quantum yields of BODIPY could be greatly affected by the electronic properties of mesosubstituents,38,39 thus the hydrolysis of more electronwithdrawing esters to the less electron-withdrawing carboxylate anions would bring a remarkable change in fluorescence emission under physiological conditions (pH 7.4). To validate this assumption, a panel of ester derivatives were rationally designed and synthesized by introduction of methyl or ethyl ester into the fluorophore skeleton. Following molecular docking-based virtual screening combined with reaction phenotyping-based experimental screening, we found that MMB (probe 7) exhibited the optimal combination of sensitivity, specificity and fast response towards CES1 in contrast to other derivatives. Upon addition of CES1 or CES1-
Materials and instruments. All chemicals were commercial products of analytical grade and used without further purification. NMR (1H and 13C) spectra were recorded on Bruker spectrometer (400 MHz and 500 MHz). Highresolution mass data were obtained with LTQ Orbitrap XL and GC–HRTOF MS spectroscopy. All fluorescence tests were carried out on Synergy H1 Hybrid Multi-Mode Microplate Reader (BioTek). MMB and its hydrolytic product MCB were determined by a HPLC-UV system (Waters e2695 equipped with 2998 PDA Detector). Pooled human liver microsomes (HLMs) and a panel of HLMs from twelve individuals were purchased from RILD (Shanghai, China). Paraoxonases including PON-1 and PON-2 were purchased from Bioworld (MN, USA), human serum albumin (HSA), trypsin, acetylcholinesterase (AChE), butyryl-cholinesterase (BChE), carbonic anhydrase (CA), lipase, α-chymotrypsin (α-CT), bovine serum albumin (BSA), Proteinase K and dipeptidylpepeidase4 (DPP-4) were obtained from Sigma, lysozyme was obtained from Solarbio, recombinant human CEs isoforms including human CES1 and human CES2 were obtained from BD company (MA, USA). Loperamide (LPA) and bis-pnitrophenyl phosphate (BNPP) were obtained from TCI. Galanthamine (GA) and ethylene diaminetetraacetic acid (EDTA) were purchased from J&K Chemical Ltd. (China), 3O-(β-Carboxypropionyl)-ursolic acid (UKA) was synthesized by the authorand described previously.40 Stock solution of probe MMB (10 mM) was prepared in dimethyl sulfoxide (DMSO), storing at -80 ˚C for future use. General procedure for monitoring CES1 activity. All the measurements of CEs or other hydrolytic enzymes activity were carried out in following incubation system. In brief, the incubation mixture (a total volume of 200 μL) was consisted of PBS (100 mM, pH 7.4), and above mentioned enzyme or human liver microsomes which were mixed gently. The reactions were initiated by adding of probe substrate MMB (dissolved in DMSO previously), following by incubated in a shaking water bath at 37 ˚C. The final concentration of DMSO did not exceed 1% (v/v) which had no influence on the enzymatic activity of CES1 (Figure S13). The reactions were terminated by addition of equal volume of ice-cold acetonitrile, then the mixtures were centrifuged at 20,000 × g for 10 min. Aliquots of supernatant were then taken for fluorescence analysis (Gain 100). Control assays (without enzyme sources) were performed to avoid false positive results. Chemical inhibition assays. To ensure that the hydrolysis of MMB was CES1 dependent, chemical inhibition assays were carried out by incubating MMB (final concentration 10 μM) with human CES1 and HLMs in the presence of several selective esterase inhibitors, including LPA (CES2 selective
ACS Paragon Plus Environment
Page 2 of 9
Page 3 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry inhibitor),41 BNPP (a known inactivator of CEs), UKA (CES1 selective inhibitor), GA (AChE selective inhibitor) and EDTA (a selective inhibitor of PONs). The chemical inhibitors were firstly pre-incubated with human CES1 or HLMs at (0.2 μg/mL) at 37 ˚C for 10 min, then, MMB was added to start the reaction, following by incubated in a shaking water bath at 37 ˚C. Termination step and sample preparation were carried out followed procedures described previously. Cell culture and cellular imaging. Caco-2 cells at logarithmic phase were seeded into glass polylysine-coated confocal cell culture dishes (Φ 20 mm) and incubated with MEM/EBSS culture medium containing 10% fetal bovine serum (FBS) in a 5% CO2 humidified incubator at 37 ˚C overnight. After wash twice with culture medium, the adherent cells were incubated with/without 100 µM BNPP (prepared in FBS-free medium) in 5% CO2 incubator for 30 min at 37 ˚C. The probe MMB stock solution (10 mM) was diluted to a final concentration of 10 μM with the FBS-free cell culture media, and then the adherent cells were incubated with MMBcontaining FBS-free cell culture media for another 30 min at 37 ˚C. After washed with PBS for three times, fluorescence imaging was performed on confocal microscope (Leica SP8, Germany) with excitation at 515 nm and the collected windows were set as 575-615 nm. Fluorescence imaging in zebrafish. Zebrafishs were maintained in holt buffer (6 mM NaCl, 0.07 mM KCl, 0.03 mM NaHCO3, and 0.09 mM CaCl2) , when grown for 5 days, they were pretreated with/without 100 uM BNPP for 30 min. After that, the zebrafishs were incubated with MMB (10 µM) at 28 ˚C for another 30 min. After washing with PBS (pH 7.4) to clear the remaining MMB, the zebrafishs were imaged on confocal microscope (Olympus FV1000) for excitation at 515 nm. The fluorescence was collected in the ranges of 550−650 nm. Scheme 2. Chemical structures of the designed probes.
introduction of halogen atoms into 2, 6-position of the fluorophore may cause a significant red shift in fluorescence spectra, as reported previously.45 In these cases, mesocarboxyl-BODIPY was selected as a basic scaffold and a series of ester derivatives (probe 1-8) were rationally designed and synthesized as probe candidates for CES1 by introducing methyl or ethyl ester into the meso-position of this fluorophore (Scheme 2). The details of the synthetic procedure for MMB and other probes were described in the supplementary information, and the chemical structures of these probes are fully characterized by HRMS, 1H NMR and 13C NMR, respectively (Supplementary Figure S14-S54). Firstly, molecular docking simulations were performed to evaluate the potentials of these designed probes as CES1 substrates, which might provide key information on rational design and structural modification of specific ligands for the target enzyme.47,48 As shown in Table S1, most of these designed probes could be well-docked into the active cavity of human CES1 with the LibDockScores larger than 80, while the distances between the carboxyl group of these probes and Ser221 (the key serine residual in the active cavity of human CES1) or the oxyanion hole (Gly142 and Gly143) were not exceed 5 Å. Conversely, these probes could also be docked into the active cavity of human CES2, but the distances between the ester bond of these probes and the oxyanion hole (Gly148 and Gly149) of human CES2 were larger than 5.5 Å (Table S2), such distances would hinder the process of CES2-mediated catalysis to a great extent. These results suggested that most of these probes could be potential substrates for CES1 but poor substrates for CES2. Notably, the distances between probe 7 (MMB) and Ser221 or the oxyanion hole of human CES1 were all within 3 Å (Figure S1), which was much shorter than that of other probes (Table S1). Furthermore, as shown in Figure S2, this probe could also create strong interactions with the residuals in the catalytic cavity of human CES1. All these results indicated that MMB might be a good substrate for CES1. After that, the participation of CES1 and CES2 in hydrolysis of these esters was assayed using recombinant enzymes. As shown in Figure S3, all synthesized substrates could be hydrolyzed by CES1. Among all tested substrates, probe 1-4 could be hydrolyzed by both CES1 and CES2, while probe 5-8 showed relatively high specificity towards CES1 over CES2. As expected, the probe 7 (MMB) displayed much faster response than that of probe 5, 6 and 8 under physiological conditions, which was highly consistent with the results from molecular docking simulations.
RESULTS AND DISCUSSIONS Design, synthesis and sensing mechanism of MMB. As previously reported, the substrate preferences of CES1 and CES2 are highly different, while CES1 preferentially metabolizes the substrates with a large acyl group and a small alcohol group.9,42-44 Recent studies have indicated that the fluorescence properties of BODIPYs are strongly dependent on the electron-withdrawing capability of the meso-substituent of the fluorophore.38,39,45 It is expected that the ester bonds of these fluorescence substrates at the meso-position may be hydrolyzed to their corresponding carboxylate anions by CES1 under physiological conditions. Considering that the Hammett constant for these esters (σp=0.45) are much larger than that of carboxylate anions (σp=0.00),46 the changes in the electronic property at meso-position will bring a dramatic change in fluorescence emission. It also should be noted that the
Figure 1. (a) Fluorescence spectra of MMB (5 μM) incubated with increasing concentrations of CES1 (0-8 μg/mL) in PBS (containing less than 1% DMSO) at 37 ˚C for 30 min. (b) The linear relationship between fluorescence intensity and CES1 concentrations (0-8 μg/mL) in PBS (containing less than 1% DMSO). The experiments were conducted in triplicate and the data were obtained as mean (± SD). λex = 530 nm.
The sensing ability of MMB in real samples was then investigated. As shown in Figure S4, MMB was very stable
ACS Paragon Plus Environment
Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
under physiological conditions without any detectable metabolites after incubated for 30 min. By contrast, MMB could be rapidly metabolized and formed a single stable metabolite after incubated with CES1, and such biotransformation triggered a dramatic increase in the fluorescence emission at 595 nm (Figure 1a). The metabolite was fully identified as MCB by comparison of the LC retention time and UV spectra with the standard substance (Figure S4). The proposed sensing mechanism of MMB for CES1 was shown in Scheme 1. The formation of MCB was time- and CES1- dependent. As shown in Figure 1b, the fluorescence intensity at 595 nm exhibited an excellent linearity (R2 = 0.99, p< 0.0001) with the CES1 concentrations within the range of 0–8 µg/mL. The time-dependent fluorescence spectra changes of MMB (5 µM) in CES1 were depicted in Figure S5, upon addition of CES1, the fluorescence responses were linearly related to the reaction times of up to 40 min. In addition, the effects of pH on the fluorescence detection of MMB and its hydrolytic metabolite MCB were also performed. As shown in Figure S6, the fluorescence intensity of both MMB and MCB were stable within the pH range of 4.0–8.0. These results demonstrated that MMB could serve as a practical long-wavelength “turnon” fluorescent probe for sensing the enzymatic activity of CES1 under physiological conditions.
Figure 2. (a) Fluorescence responses of MMB (5 µM) towards various species of hydrolases in PBS (containing less than 1% DMSO) for 30 min at 37 ˚C; (b) Inhibitory effects of esterase inhibitors toward MMB in recombinant CES1 and HLMs. The experiments were conducted in triplicate and the data were obtained as mean (± SD).
Specificity of MMB towards CES1. The specificity of MMB was carefully investigated using a panel of enzymes in the human body with esterase or pseudo-esterase activity. As shown in Figure 2a, only CES1 could trigger the increase in fluorescent emission at 595 nm, while other enzymes including CES2, CA, a-CT, trypsin, PON1, PON2, HSA, BChE, AChE, lipase, lysozyme, DPP-4, BSA and Proteinase K could not cause any changes in fluorescence intensity around 590 nm (for MMB) under same conditions. To further confirm whether MMB hydrolysis was specifically mediated by CES1, the effects of diverse selective esterase inhibitors on this hydrolysis reaction were evaluated in both human CES1 and HLMs, the hydrolysis reaction could be strongly inhibited by BNPP (a carboxylesterase inhibitor) and UKA (a specific CES1 inhibitor), while LPA (a specific CES2 inhibitor), EDTA (PONs selective inhibitor) and GA (AChE selective inhibitor) did not inhibit this reaction (Figure 2b). The fluorescence response of MMB toward various biologically relevant species was also carried out to investigate its selectivity in biological systems. As shown in Figure S7, common amino acids and metal ions in human tissue fluids had no effect on the fluorescence spectrum of MMB. These findings demonstrated that MMB showed excellent specificity for CES1 over other biologically relevant species and it could
serve as an efficient tool for highly specific detecting CES1 in complicated biological samples. Enzymatic kinetic of MMB hydrolysis. Considering that enzymatic kinetic behavior was very important for the quantification of the enzyme activatable probes.3,49 The enzymatic kinetic behaviors of MMB were characterized in different enzyme sources including human CES1 and HLMs. As shown in Figure 3a and 3b, MMB hydrolysis in both enzyme sources exhibited typical Michaelis-Menten kinetics, as evidenced by the corresponding Eadie-Hofstee plots. Moreover, the hydrolysis of MMB in both human CES1 and HLMs showed high affinity (km< 5 μM), as well as good reactivity (kcat/km>5 mL·min-1·mg-1 of protein). Meanwhile, the km values of the MMB hydrolysis in human CES1 and HLMs were very closed (Table S3), indicating that CES1 served as the predominant enzyme contributed to the metabolism of MMB in human liver preparations. These findings indicated that MMB hydrolysis exhibited ideal kinetic behaviors and could serve as a practical tool for quantitative detecting CES1 in human biological samples.
Figure 3. Michaelis-Menten kinetic plots of MMB hydrolysis in CES1 (a) and HLMs (b). Dose-inhibition curves of BNPP (c) and UKA (d) on MMB (5 µM) hydrolysis in CES1 and HLMs, respectively. The experiments were conducted in triplicate and the data were obtained as mean (± SD).
We subsequently evaluated the applicability of MMB for rapid screening CES1 inhibitors using both human CES1 and HLMs as the enzyme sources. As shown in Figure 3c and 3d, the two known CEs inhibitors (including BNPP and UKA) displayed strong inhibition toward CES1-mediated MMB hydrolysis via a dose-dependent manner in both CES1 and HLMs. The IC50 values of BNPP and UKA in human CES1 was also similar to that in HLMs. These results suggested that MMB could be applied to high-throughput screening (HTS) of CES1 modulators by using HLMs to replace high-cost recombinant human CES1, which would effectively facilitate CES1-related drug discovery.
ACS Paragon Plus Environment
Page 4 of 9
Page 5 of 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry Figure 4. (a) The hydrolytic activities of MMB in 12 individual HLMs samples. (b) Correlation analysis between the hydrolytic rates of MMB and clopidogrel in 12 individual HLMs samples. The experiments were conducted in triplicate and the data were obtained as mean (± SD).
Quantification of CES1 in HLMs and correlation studies. To further explore the practicability of MMB for highly selective sensing CES1 in human biological samples, we evaluated the metabolic rates of MMB in a panel of 12 individual HLMs samples. As shown in Figure 4a, the catalytic activities of CES1 in HLMs from different individuals were assayed using MMB as a probe substrate. About 4.1-fold individual differences in the catalytic activity was observed, in accordance with the previous literature regarding the variability in CES1 levels among individual HLMs.50,51 Furthermore, the hydrolytic rates of MMB and clopidogrel (a known specific substrate drug of CES1) exhibited a strong correlation with a high correlation coefficient (R2 = 0.93, p
