Environmentally sensitive fluorescent probes with improved properties for detecting and imaging PDEδ in live cells and tumor slices

Keling Li,Shncho Wu,Gopn Dong,Yu Li,Wei Wng,Guoqing Dong,Zhnying Hong,?,Minyong Li,Chunqun Sheng,?

a School of Pharmacy,Second Military Medical University,Shanghai 200433,China

b Department of Medicinal Chemistry,Key Laboratory of Chemical Biology (MOE),School of Pharmaceutical Sciences,Cheeloo College of Medicine,Shandong University,Ji’nan 250012,China

Keywords:Antitumor PDEδ Fluorescent probes Environmentally sensitive Binding affinity

ABSTRACT Kirsten rat sarcoma viral oncogene homolog (KRAS)–phosphodiesterase-delta (PDEδ) is a promising target for antitumor drug discovery.Herein,highly efficient and environmentally sensitive fluorescent probes of PDEδ (DS-Probes) were rationally designed.As compared with the reported PDEδ probes,DS-Probes showed higher binding affinity and selectivity,which were able to conveniently and efficiently label PDEδ in live cells as well as tumor tissues.Therefore,these fluorescent probes are expected to facilitate PDEδbased mechanism elucidation,drug discovery and pathologic diagnosis.

Pancreatic cancer is associated with high mortality with fiveyear survival less than 10% [1–3].Kirsten rat sarcoma viral oncogene homolog (KRAS) has the highest mutation rate (90%) in pancreatic cancer,which is the main factor responsible for the occurrence and development of pancreatic cancer [3–5].Targeting KRAS signaling has become an important field in antitumor drug discovery and achieved great success [6–10].In 2021,the first KRAS inhibitor sotorasib was approved for the treatment of non-small-cell lung cancers with KRAS G12C mutations [11].However,monotherapy of sotorasib is limited due to the low proportion of G12C mutations in all KRAS mutations [12].Therefore,the development of pan-KRAS inhibitors targeting multiple KRAS mutants is becoming a promising strategy [13].

Phosphodiesterase-delta (PDEδ) plays an important role in regulating the functions of KRAS [8],which assists the transport of KRAS to cell membrane by binding the farnesyl group of KRAS,thereby promoting the activation of downstream signaling pathways [14–16].Disruption of KRAS–PDEδprotein–protein interaction is a new strategy for drug development targeting mutant KRAS[16,17].Nevertheless,it is disappointing that the existing PDEδinhibitors were generally limited by low anti-tumor efficacy and poor selectivity [17,18].Thus,new chemical tools targeting PDEδare urgently needed to understand the biological functions and druggability of PDEδ.

In recent years,the visualization of target protein functions by fluorescent probes has facilitated the elucidation of biological mechanisms and drug discovery [20,21].Fluorescent imaging has the advantages of low invasiveness,low radiation,low toxicity,and fast spatiotemporal localization [22–26].Small molecule fluorescent probes have become indispensable tools in the fields of molecular biology and medicine [27,28].Particularly,environmentsensitive fluorescent probes have enhanced fluorescence in hydrophobic environment,showing a significant improvement in the signal-to-noise ratio after target binding and consequently label the target proteins more clearly,providing effective visualization tools for protein function research [29–31].

Currently,three types of chemical fluorescent probes targeting PDEδhave been reported [16,19,32].Based on PDEδbinder atorvastatin,Waldmann’s group designed a fluorescein-labeled atorvastatin probe (AT-Probe,Fig.1A) to detect fluorescence properties (FP) and develop assays [16].Additionally,they developed benzenedisulfonamide-based probe (BZ-Probe,Fig.1A) with higher selectivity and better binding ability to PDEδ,whereas it failed to possess environmental sensitivity [19].

Fig.1.Design rationale of environment-sensitive fluorescent probes of PDEδ.(A) Chemical structures of the reported PDEδ fluorescent probes.(B) The design strategy of DS-Probes.(C) The binding mode of DS-P1 (green) with PDEδ.(D) The superimposed conformation of the ligand (orange) with DS-P1.

In our previous studies,the first class of environment-sensitive fluorescent probes (QZ-Probes,Fig.1A) were designed to image PDEδin Capan-1 and MIA PaCa-2 pancreatic cancer cells and tissues [32].Nevertheless,QZ-Probeshad weak binding activity to PDEδ(KD=440–682 nmol/L).Only when the concentration of PDEδwas higher than 0.5 μmol/L,the fluorescence signal could be significantly enhanced.To overcome this limitation,herein,more effective PDEδenvironment-sensitive fluorescent probes were designed and synthesized.

To improve the binding affinity,highly active PDEδinhibitors were required to be attached with fluorophores.First,the fluorescent groups 7-nitrobenzo-2-oxa-1,3-diazole (NBD) with alkyl chains,which possessed environmental sensitivity,good watersolubility and small size were used [33].Then,among the PDEδinhibitors,benzenedisulfonamides showed the highest binding activity to PDEδ,which antagonized the allosteric effect of PDEδmediated by Arl2 [19].On the basis of excellent binding activity,benzenedisulfonamide PDEδinhibitorDS-7(KD<2 nmol/L,Fig.1B)was selected to design novel fluorescent probes of PDEδ.The binding mode of compoundDS-7with PDEδreveals that its terminal carboxyl group is located in the Tyr149 pocket and forms hydrogen bonding interaction with Met118 (PDB code: 5ML4) [19].Notably,this group also points to the outside of the binding pocket,offering a favorable site for probe design.Thus,the carboxyl group of compoundDS-7was extended by an alkyl side chain and subsequently connected with NBD,affording three new PDEδfluorescent probes (herein namedDS-Probes,Fig.1B).

Next,in order to clarify the binding mode ofDS-Probeswith PDEδ,DS-P1(Fig.1B) was selected for molecular docking and the results indicated that the newly designed probe maintained the binding mode of the ligand and key hydrogen bonds with Cys56,Arg61,Met118 and Tyr149 were retained (Fig.1C).Furthermore,the fluorophores had no interactions with the residues of PDEδ,which had little effect on the binding affinities ofDS-Probes.As shown in the superimposed structures (Fig.1D),the binding mode ofDSP1was highly similar to that of ligandDS-7,which validated the rationality of the probe design.

The synthetic route ofDS-Probes(DS-P1,DS-P2,DS-P3) is shown in Scheme 1.Starting from compound1(NBD-Cl),NBD fluorescent fragments3was prepared by substitution reaction with diamines followed by removing the Boc protection,which can be used directly without further purification.Using cyclopentylamine andp-chlorobenzaldehyde as starting materials,intermediate6was obtained through reductive amination reaction,and then condensed withp-bromobenzenesulfonyl chloride to obtain intermediate7.Compound7was coupled with benzylthiol under the catalysis of metal palladium to afford intermediate8,which was further oxidized by dichlorohydantoin to give key intermediate9.Compound10was substituted by methylamine hydrochloride to obtain intermediate11,which was further converted to intermediate12viareductive amination reaction with 1-Boc-4-(aminomethyl)piperidine.Then,compound12was condensed with key intermediate9to give intermediate13.After hydrolysis of compound13with lithium hydroxide (LiOH),the demethylated compound14was condensed with key intermediate3in the presence ofO-benzotriazol-1-yl-tetramethyluronium hexafluorophosphate (HBTU) and triethylamine (TEA).Finally,the target compoundsDS-Probeswere obtained after removing the protective group of the compound14.

Scheme 1.Synthesis of DS-Probes.Reagent and conditions: a) DIPEA,NMP,microwave,110 °C,1.5 h,70%–87%;b) TFA,DCM,r.t.,0.5 h;c) MgSO4,NaBH4,MeOH,r.t.,14 h,97%;d) TEA,DCM,40 °C,12 h,69%;e) benzyl mercaptan,Pd2(dba)3,X-Phos,DIPEA,1,4-dioxane,110 °C,12 h,75%;f) MeCN-AcOH-H2O,0 °C,2 h,76%;g) K2CO3,MeNH3Cl,NMP,100 °C,24 h,52%;h) NaBH(OAc)3,AcOH,1,2-dichloroethane,r.t.,overnight,60%;i) TEA,DCM,0 °C,1 h,85%;j) LiOH,THF-MeOH-H2O,50 °C,1 h,85%;k) HBTU,TEA,DMF,r.t.,2 h;l) TFA,DCM,r.t.,0.5 h,35%–39%.

Initially,the PDEδ-binding activities ofDS-Probeswere assayed by FP and surface plasmon resonance (SPR) methods using PDEδinhibitorDeltazinone 1and our previously reported probeQZ-P1as positive controls (Table S1 in Supporting information).In the FP assay,all the three probes showed potent activities in binding PDEδ(KDrange: 16.08–34.17 nmol/L),which were significantly more potent than probeQZ-P1(KD=682 nmol/L).The SPR assay further confirmed the PDEδbinding ability of the probes (KDrange: 0.061–0.25 μmol/L).DS-P1was proven to be the most potent probe in both assays (FP,KD=16.08 nmol/L;SPR,KD=0.061 μmol/L).

Theinvitroantitumor activity of the probes was further assayed against KRAS-dependent pancreatic cancer cell lines MIA PaCa-2 and Capan-1 using the cell counting kit-8 (CCK-8) method.As shown in Table S1,DS-Probesexhibited moderate antiproliferation activity against MIA PaCa-2 cell line (IC50range: 15.8–18.3 μmol/L) and Capan-1 cell line (IC50range: 17.6–31.3 μmol/L),which were suitable for cellular imaging.

Then,the spectra ofDS-Probeswere measured (Fig.S2 in Supporting information).The maximum absorption wavelength (λmax),maximum excitation wavelength (λex) and maximum emission wavelength (λem) are shown in Table 1.The difference betweenλexandλemindicated that the probes had favorable properties for fluorescence detection.

Table 1Spectral properties of DS-Probes.

Furthermore,the fluorescence quantum yields (Φ) of the probes were measured (Table 1).In the phosphate buffered solution (PBS,pH 7.4),the fluorescence quantum yields ofDS-Probeswere less than 0.1%,while the fluorescence quantum yields of the probes were significantly increased in dimethyl sulfoxide (DMSO,7.27%–23.08%).These results suggested thatDS-Probespossessed the environmentally sensitive turn-on mechanism.Interestingly,PDEδwas tested as non-fluorescent in our previous works [32],the addition of a little amount of PDEδ(0.125 μmol/L),also led to significant enhancement of the fluorescence quantum yields.With the increase of protein concentration,the fluorescence quantum yields were further increased in a concentration-depended manner(Table 1).Similarly,the fluorescence intensity of the probes also depended on the concentration of PDEδ(Fig.2).In contrast,the fluorescence intensity ofQZ-P1was not obviously increased in the presence of the low concentration of PDEδ(0.125 μmol/L).These results indicated thatDS-Probeshad good environmental sensitivity to PDEδ,which were significantly more effective than probeQZ-P1.

Fig.2.Fluorescence emission spectra of DS-Probes and QZ-P1 before and after the addition of PDEδ.(A) DS-P1;(B) DS-P2;(C) DS-P3;(D) QZ-P1.

To verify that theDS-Probesact on PDEδin cells,the effects of different probes on the thermal stability of PDEδin MIA PaCa-2 cells were investigated by cellular thermal shift assays.After the treatment of probes (20 μmol/L) for 2 h,the cell samples were collected and divided into 10 groups.Then,the changes of PDEδin each group were detected by the Western blot at different temperatures.The results revealed that the thermal stability of PDEδwas significantly improved after the treatment ofDS-Probeswhen compared with the blank control (Fig.S3A in Supporting information).When the temperature was above 55 °C,the PDEδbands of theDS-Probeswere still clearly visible.TheDS-P1group showed the highest stability,and there was still a little amount of PDEδat 70 °C.In contrast,the bands in theQZ-P1group and the blank control group (1% DMSO) gradually disappeared above 52 °C.The results indicated thatDS-Probeswere able to bind PDEδin cells,which were significantly more effective than probeQZ-P1.

In order to clarify the effects ofDS-Probeson the KRAS signaling pathway in cells,the changes of the phosphorylation levels of protein kinase B (Akt) and extracellular signal-related kinase(Erk) in the downstream signaling pathways in MIA PaCa-2 cell line were evaluated by Western blot analysis.After the treatment ofDS-Probesfor 6 h,the phosphorylation levels of Akt and Erk were separately down-regulated byDS-ProbesandDS-P3.Furthermore,DS-P1andDS-P2affected Akt andDS-P3affected Erk in a concentration-dependent manner (Fig.S3B in Supporting information).At the same concentration,the down-regulation effects ofDS-Probeson Akt were more obvious than those observed inDeltazinone 1andQZ-P1groups.These results suggested thatDS-Probesacted on PDEδin MIA PaCa-2 cells,interfered with KRAS–PDEδinteraction,and affected the KRAS signaling pathway.

At present,specific fluorescent probes are valuable tools for visualizing the expression and localization of target proteins in cells [34,35].Based on the excellent fluorescence properties,DS-Probeswere used to detect and image PDEδin KRASdependent MIA PaCa-2 cells.The results showed thatDS-Probes(5 μmol/L) rapidly bound to PDEδ,enabling PDEδimaging in live cells,among which the fluorescence ofDS-P1was relatively stronger (Fig.3A).For the fluorescence distribution,green fluorescence was mainly distributed in the cytoplasm and plasma membrane,where PDEδwas mainly located (Fig.3A).Furthermore,MIA PaCa-2 cell line was co-incubated with 5 μmol/LDS-Probesand 100 μmol/LDS-7.The results showed that the fluorescence intensity of the probes was decreased by the competition of PDEδinhibitorDS-7,indicating thatDS-Probescan reversibly bind to PDEδ.WhenDS-Probes(5 μmol/L) were evaluated in the normal HEK-293T cell line,the results showed that probes failed to stain cells,indicating thatDS-Probescan selectively bind to PDEδ.These results demonstrated thatDS-Probescan be used as effective tools for real-time detection and rapid visualization of KRAS–PDEδprotein–protein interaction in live cells.The imaging results of Capan-1 cells were also consistent with those observed in MIA PaCa-2 cell images (Figs.S4–S6 in Supporting information).

Fig.3.Fluorescence labeling of DS-Probes in tumor cells and tissues.(A) Fluorescence images of MIA PaCa-2 and HEK-293T cell lines incubated with DS-Probes.Scale bar=67 μm.(B) Flow cytometry results of DS-Probes in MIA PaCa-2 cells.B1: control (1% DMSO);B2: 100 μmol/L DS-7;B3: 5 μmol/L DS-P1;B4: 5 μmol/L DS-P1+100 μmol/L DS-7;B5: 5 μmol/L DS-P2;B6: 5 μmol/L DS-P2+100 μmol/L DS-7;B7: 5 μmol/L DS-P3;B8: 5 μmol/L DS-P3+100 μmol/L DS-7 (Meaning of the figures in B1–B8: a ratio of the area under the horizontal line at a fixed position to the total area under the curve in percent).(C) Fluorescence staining of Capan-1 cell xenograft sections by DS-Probes.The slices were captured by the OLYMPUS VS120 virtual slide microscope (objective lens: 40×).

The staining effect ofDS-Probesto MIA PaCa-2 cell line was further quantitatively analyzed by flow cytometry (FCM).As shown in Fig.3B,the fluorescence intensity of MIA PaCa-2 incubated with 5 μmol/L probes was significantly stronger than that of MIA PaCa-2 incubated with 100 μmol/LDS-7and that of the control group.When the probes were competed with 100 μmol/LDS-7,the fluorescence intensity of cells was significantly reduced.The results further demonstrated the staining effect and selectivity ofDS-Probes.The FCM results of Capan-1 cells were also consistent with those observed in MIA PaCa-2 cells (Fig.S7 in Supporting information).

Additionally,the effects ofDS-Probeson PDEδin tumor tissues were further detected by fluorescence staining of tissue sections.Tumor tissue sections of the Capan-1 cell line were stained with probes at a concentration of 5 μmol/L,andDS-7(100 μmol/L)was used as a competitive ligand to investigate the reversibility and specificity of probes in imaging PDEδ.As shown in Fig.3C,the tumor slices treated with the probes alone had a good response of green fluorescence,among which the fluorescence ofDSP1andDS-P3was relatively stronger.In contrast,the fluorescence intensity of normal mice skin tissue slices treated withDS-Probes(5 μmol/L) was relatively weaker than that in Capan-1 tumor slices.In the competitive binding experiment,the fluorescence intensity was significantly reduced when high concentration (100 μmol/L) ofDS-7was added,which further proved that the probes reversibly and selectively bound to PDEδand labeled tumor cells in tissues.The experimental procedures and the animal use and care protocols were approved by the Committee on Ethics of Biomedicine,Second Military Medical University.

In summary,a new series of environmentally sensitive fluorescent probes for PDEδwas rationally designed.DS-Probeshad stronger affinity,better selectivity and higher sensitivity to PDEδ,leading to better imaging capabilities.Mechanism studies revealed thatDS-Probescould selectively bind to PDEδand down-regulate the phosphorylation levels of Erk and Akt in the KRAS signaling pathway.Furthermore,DS-Probesquickly,efficiently,selectively and reversibly labeled PDEδin living cells and tumor tissues.Thus,DS-Probesare expected to serve as valuable tools for the detection and visualization of KRAS–PDEδinteraction with potential to be applicated in better understanding the biological functions of PDEδand developing assays for drug screening.Admittedly,fluorescence imaging of PDEδwas limited to tumor cells and tissues due to the spectral properties ofDS-Probes,the further optimization of PDEδ-labeling probes is currently underway in our groups.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Key Research and Development Program of China (No.2020YFA0509200 to C.Sheng),National Natural Science Foundation of China (Nos.81903436 to Y.Li,82204211 to W.Wang and 22077138 to S.Wu) and Shanghai Rising-Star Program (No.22QA1411300 to S.Wu).

Supplementary materials

Supplementary material associated with this article can be found,in the online version,at doi:10.1016/j.cclet.2023.108231.