Choline kinase: An unexpected journey for a precision medicine strategy in human diseases
Abstract
© 2021 by the authors.
Full text
pharmaceutics Review Choline Kinase: An Unexpected Journey for a Precision Medicine Strategy in Human Diseases Juan Carlos Lacal 1,2,* , Tahl Zimmerman 3and Joaquín M. Campos 4,5,* Citation: Lacal, J.C.; Zimmerman, T.; Campos, J.M. Choline Kinase: An Unexpected Journey for a Precision Medicine Strategy in Human Diseases. Pharmaceutics 2021,13, 788. https:// 10.3390/pharmaceutics13060788 Academic Editor: Luisa Carlota Lopez-Cara Received: 27 April 2021 Accepted: 19 May 2021 Published: 25 May 2021 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). 1Instituto de Investigaciones Biomédicas, CSIC, 28029 Madrid, Spain 2Instituto de Investigación Sanitaria Hospital La Paz, IDIPAZ, 28046 Madrid, Spain 3Food Microbiology and Biotechnology Laboratory, Department of Family and Consumer Sciences, College of Agriculture and Environmental Sciences, North Carolina University, 1601 East Market Street, Greensboro, NC 27411, USA; [email protected] 4Departamento de Química Farmacéutica y Orgánica, Facultad de Farmacia, c/Campus de Cartuja, s/n, Universidad de Granada, 18071 Granada, Spain 5Instituto Biosanitario de Granada (ibs. GRANADA), SAS-Universidad de Granada, 18071 Granada, Spain *Correspondence: [email protected] (J.C.L.); [email protected] (J.M.C.); Tel.: +34-914975438 (J.C.L.); +34-958243850 (J.M.C.) Abstract: Choline kinase (ChoK) is a cytosolic enzyme that catalyzes the phosphorylation of choline to form phosphorylcholine (PCho) in the presence of ATP and magnesium. ChoK is required for the synthesis of key membrane phospholipids and is involved in malignant transformation in a large variety of human tumours. Active compounds against ChoK have been identified and proposed as antitumor agents. The ChoK inhibitory and antiproliferative activities of symmetrical bispyridinium and bisquinolinium compounds have been defined using quantitative structure–activity relationships (QSARs) and structural parameters. The design strategy followed in the development of the most active molecules is presented. The selective anticancer activity of these structures is also described. One promising anticancer compound has even entered clinical trials. Recently, ChoK α inhibitors have also been proposed as a novel therapeutic approach against parasites, rheumatoid arthritis, inflammatory processes, and pathogenic bacteria. The evidence for ChoK α as a novel drug target for approaches in precision medicine is discussed. Keywords: phospholipids metabolism; choline kinase; bispyridinium compounds; bisquinolinium compounds; QSAR; anticancer drugs; rheumatoid arthritis; parasites; pathogenic bacteria; inflammatory disease 1. Introduction Alteration of cell metabolism is a frequent event in human diseases [ 1 ]. In some instances are a necessary requirement, such as in the case of cancer onset and progression. A typical feature of cancer cells is an increased metabolic rate which supports unregulated growth and higher duplication rates. In keeping with this requirement, aerobic glycolysis, glutamine catabolism and lipid metabolism are frequently up-regulated in tumours [ 1 ]. As a key event, the pathways involved in the generation of the major phospholipids, such as cytidine diphosphate-choline (CDP-choline) and cytidine diphosphate-ethanolamine (CDP-ethanolamine), are frequently altered in human tumours. One of the enzymes that regulate these pathways is Choline kinase α (ChoK α ). Evidence that ChoK α plays a critical role in many human diseases is increasingly being accumulated, and ChoK α has become the focus of a targeted therapeutic strategy [2]. As a consequence, small molecule inhibitors and small interfering RNA (siRNA) that interfere with the function of ChoK α have been designed. These reagents have proven to be effective and selective anticancer drugs [ 2 ]. We describe here how the most active antiproliferative and antitumoral agents against ChoK were designed by relying on Pharmaceutics 2021,13, 788. https://doi.org/10.3390/pharmaceutics13060788 https://www.mdpi.com/journal/pharmaceutics
Pharmaceutics 2021,13, 788 2 of 28 quantitative structure–activity relationships (QSARs) and structural parameters. The initial framework of these inhibitors was based on symmetrical bispyridinium and bisquinolinium structures. ChoK α inhibitors have also been proposed as a novel therapeutic approach against malaria, rheumatoid arthritis, pathogenic bacteria and inflammatory processes [ 3 ]. The aim of this review is to recount the history of the development of ChoK α inhibition therapies against human diseases and to summarize the most striking findings in this field. It is a fascinating journey from bench to the bedside. In addition, there are many paths yet to be fully explored in areas of illnesses other than cancer. 2. The ChoK Family in Humans In humans, two genes code for the enzymatic activity responsible for the generation of phosphorylcholine (PCho) designated as CHKA and CHKB. Located in chromosomes 11q13.1 (CHKA) and 22q13.33 (CHKB), they code for proteins of ca. 50 kDa (ChoK α ) and 45 kDa (ChoK β , 395 amino acids) [ 4 – 6 ]. The CHKA gene originates by differential splicing ChoKα1 (52 kDa, 457 amino acids) and ChoKα2 (50 kDa, 439 amino acids) (Figure 1). Pharmaceutics2021,13,x 2of29 activeantiproliferativeandantitumoralagentsagainstChoKweredesignedbyrelying onquantitativestructure–activityrelationships(QSARs)andstructuralparameters.The initialframeworkoftheseinhibitorswasbasedonsymmetricalbispyridiniumand bisquinoliniumstructures. ChoKαinhibitorshavealsobeenproposedasanoveltherapeuticapproachagainst malaria,rheumatoidarthritis,pathogenicbacteriaandinflammatoryprocesses[3].The aimofthisreviewistorecountthehistoryofthedevelopmentofChoKα inhibition therapiesagainsthumandiseasesandtosummarizethemoststrikingfindingsinthis field.Itisafascinatingjourneyfrombenchtothebedside.Inaddition,therearemany pathsyettobefullyexploredinareasofillnessesotherthancancer. 2.TheChoKFamilyinHumans Inhumans,twogenescodefortheenzymaticactivityresponsibleforthegeneration ofphosphorylcholine(PCho)designatedasCHKAandCHKB.Locatedinchromosomes 11q13.1(CHKA)and22q13.33(CHKB),theycodeforproteinsofca.50kDa(ChoKα)and 45kDa(ChoKβ,395aminoacids)[4–6].TheCHKAgeneoriginatesbydifferentialsplic‐ ingChoKα1(52kDa,457aminoacids)andChoKα2(50kDa,439aminoacids)(Figure1). Figure1.ChoKcharacteristics.ChoKcatalyzesthephosphorylationofcholine,rendering phosphorylcholine.ThisreactionrequiresMg 2+ andATP.Inhumans,twodistinctgenes, CHKAandCHKB,codefortheChoKenzymes.Thelocusofthesegenesisindicated. CHKAgeneratesChoKα1andChoKα2bydifferentialsplicing.Theirmolecularweight andsizeareincludedaswellastheirintracellularlocation. Thereisanoverallsequenceidentityof56%betweenChoKαandChoKβproteins. Nevertheless,theyseemtoservedifferentmetabolicandbiologicalfunctions.ChoKαis essentialformouse[7]andplant[8]embryodevelopment.AChoKαknockoutcannotbe rescuedbyChoKβ[7].Meanwhile,ChoKβisdispensableinmiceandaChoKβKnockout (ChoKβKO)leadstoanon‐lethal,rostrocaudalmusculardystrophyandforelimbde‐ formityduetodecreasedlevelsofphosphatidylcholine[9]thatcanberescuedbyChoKα [10].Morerecently,theroleofChoKβinbonehomeostasisinmicehasbeendemon‐ strated[11].Thisresultisconsistentwiththefactthatpharmacologicalinhibitionof ChoKactivityinhumanosteoblastsresultsindefectivemineralizationinvitro[12].In humans,mutationsintheCHKBgeneleadtosimilarpathology[13].Finally,inAra‐ bidopsis,therearefourCholine‐EthanolamineKinase(CEK)genesrelatedtocholine (Cho)andethanolamine(Etn)phosphorylation,withdifferentrolesinmetabolismand development[14]. AdditionalenzymaticanalysesrevealeddifferentialactivitypatternsforChoKαand ChoKβ, φυρτηερ συππορτινγ differentphysiologicalroles. AlthoughbothhaveChokinaseandEtnkinaseactivitiesunderinvitroconditions,in wholecellsassaystheydivert.ChoKαstillhasbothactivitiesbutChoKβhasonlythe ethanolaminekinaseactivity[15].Furthermore,ChoKβ doesnotcompensateforde‐ Figure 1. ChoK characteristics. ChoK catalyzes the phosphorylation of choline, rendering phosphorylcholine. This reaction requires Mg 2+ and ATP. In humans, two distinct genes, CHKA and CHKB, code for the ChoK enzymes. The locus of these genes is indicated. CHKA generates ChoK α 1 and ChoK α 2 by differential splicing. Their molecular weight and size are included as well as their intracellular location. There is an overall sequence identity of 56% between ChoK α and ChoK β proteins. Nevertheless, they seem to serve different metabolic and biological functions. ChoK α is essential for mouse [ 7 ] and plant [ 8 ] embryo development. A ChoK α knockout cannot be rescued by ChoK β [ 7 ]. Meanwhile, ChoK β is dispensable in mice and a ChoK β Knockout (ChoK β KO) leads to a non-lethal, rostrocaudal muscular dystrophy and forelimb deformity due to decreased levels of phosphatidylcholine [ 9 ] that can be rescued by ChoK α [ 10 ]. More recently, the role of ChoK β in bone homeostasis in mice has been demonstrated [ 11 ]. This result is consistent with the fact that pharmacological inhibition of ChoK activity in human osteoblasts results in defective mineralization in vitro [ 12 ]. In humans, mutations in the CHKB gene lead to similar pathology [ 13 ]. Finally, in Arabidopsis, there are four Choline-Ethanolamine Kinase (CEK) genes related to choline (Cho) and ethanolamine (Etn) phosphorylation, with different roles in metabolism and development [14]. Additional enzymatic analyses revealed differential activity patterns for ChoK α and ChoK β , φυρτηερσυππ o ρτινγ different physiological roles. Although both have Cho kinase and Etn kinase activities under in vitro conditions, in whole cells assays they divert. ChoK α still has both activities but ChoK β has only the ethanolamine kinase activity [ 15 ]. Furthermore, ChoK β does not compensate for decreased phosphatidylcholine (PC) biosynthesis in ChoK α+/− heterozygous mice. ChoK α production is sufficient to maintain normal PC levels in most tissues, but not in limb muscles of ChoK β KO mice. Furthermore, phosphatidylethanolamine (PE) levels are unaffected in ChoK β KO mice, a
Pharmaceutics 2021,13, 788 3 of 28 result which suggests that PE homeostasis is fully maintained when the ChoK α protein is intact [ 9 ]. Again, in Arabidopsis, substrate specificity for Cho and Etn are distinct for the four CEK enzymes [14] and resembles the situation described in mammalian cells [15]. The crystal structure of human ChoK demonstrates that this enzyme forms a dimer (doi:10.2210/pdb2CKQ/pdb; doi:10.2210/pdb5EQY/pdb) [ 16 , 17 ]. This structural feature, confirmed in enzymatic analysis may have important biochemical and biological consequences since different levels of enzymatic activity for homoand heterodimers have been reported, with the ChoK α homodimer as the most active and the ChoK β homodimer as the least active [ 18 , 19 ]. Furthermore, a balance of expression of the two isoforms may have a differential effect on the regulation of the cell cycle [ 18 , 19 ]. Dimerization ratios may therefore serve as a complex regulatory mechanism for enzyme function, though further work has to be done to clarify this system. 3. ChoKαin Cancer: A Promissing Therapeutic Target ChoK α is overexpressed in a large diversity of human tumours including breast [ 20 ], lung [ 21 – 23 ], colorectal [ 21 , 24 ], bladder [ 25 ], prostate [ 21 , 26 – 28 ], ovary [ 29 ], endometrial [ 30 ], pancreas [ 31 ], liver [ 32 – 35 ], esophagus [ 36 ], and T-cell lymphoma [ 37 ]. ChoK α overexpression has also been reported in osteosarcoma tumour-derived cancer cells [ 12 ], Hepatitis B Virus (HBV)-induced hepatocarcinomas [ 38 , 39 ], breast, colonand liver-derived cancer cells [ 40 ], glioblastoma- [ 41 ] and glioma-derived cell lines [ 42 ], pancreatic tumourderived cell lines [ 31 , 43 ] and T-cell acute lymphoblastic leukemia (T-ALL) primary cells and commercial cell lines [ 44 ]. In addition, ChoK α expression levels appears to be a marker of cancer prognosis, aggressiveness or metastasis, in lung [ 22 , 23 ], breast [ 45 , 46 ], bladder [ 25 ], prostate [ 26 , 27 ], ovary [ 29 ], colorectal [ 24 ], esophageal squamous cell carcinomas [ 36 ], liver [ 34 , 35 ] and B-Cell lymphomas [ 47 , 48 ]. In breast cancer, it has been associated with ER−status, increased invasiveness, and drug resistance [20,45,46,49,50]. Since ChoK α has an established role in the onset and progression of human cancers, this enzyme was proposed as a novel therapeutic target for cancer [ 2 ]. Therefore, a program for the design and synthesis of specific inhibitors (see Section 4below) was established. siRNA attenuation of ChoK production was also used to support the proof of concept [51–53] . Thus, small molecules and specific siRNAs with potent antitumor activities both in vitro and in experimental animal models have been developed. To further support the importance of ChoK α as a target for cancer therapy, ChoK α inhibition has been used in combinatorial regimes and showed a potent antitumor effect that synergizes with 5-FU in breast and colon cancer cells [ 54 – 56 ] and reverses resistance to TRAIL in colorectal [ 57 ] and ovarian cancer cells [ 58 ]. Similar synergic effects have been obtained with pancreatic ductal adenocarcinoma-derived cancer cells with gemcitabine, 5-FU or oxaliplatin [ 31 ], and for lung cancer cells with acid ceramidase inhibitors [ 59 ] or cisplatin (Lacal, unpublished). PC synthesis is an absolute requirement of cancer cells in order to proliferate. Thus, inhibition of ChoK α has been shown to have therapeutic effects in a variety of tumorderived cancer cell lines and tumor xenografts [ 2 , 3 ]. Blocking this metabolic pathway has been shown to be non-toxic to normal, primary cells as well as immortalized nontumorigenic cells, implying that different responses are activated in tumorigenic and non-tumorigenic cells in different cell systems [60–63]. A large number of studies use magnetic resonance spectroscopy (MRS), high-resolution magic angle spinning MRS (HR-MAS MRS) or positron emission tomography (PET) to determine ChoK levels or their products as biomarkers in cancer diagnosis and prognosis have been reviewed recently. Therefore, although this is a very important field in development, these techniques will not be the subject of this review. 4. Design of ChoKαInhibitors as a Precision Medicine Strategy The cost of drug discovery is far too high to be led solely by trial and error. In computer-aided drug design, quantitative structure–activity relationship (QSAR) studies
Pharmaceutics 2021,13, 788 4 of 28 are an effective and useful approach. A QSAR study attempts to understand differences in the biological activities of a group of congeners in terms of molecular changes induced by substituent changes. The most potent ChoK inhibitor known in the early 1990s was hemicholinium-3 ( 1 , HC-3) with an IC 50 for ChoK inhibition ex vivo of 500 µ M using recombinant ChoK, and an IC 50 against the human HT-29 cell line of 2.5 mM [ 64 ]. A high affinity paralyzing respiratory effect is induced by HC-3 [ 65 ], which is a competitive inhibitor of Cho transport. As a result, 1 cannot be used as an antiproliferative agent in vivo . Improving the potency and diminishing the toxicity of 1 , by changing its specificity and antiproliferative activity, was the first step in developing potential anticancer drugs that target ChoK. The ChoK inhibitory and antiproliferative activities of symmetrical bispyridinium and bisquinolinium compounds have been defined using quantitative structure–activity relationships (QSARs) and structural parameters. We will review this strategy in the following sections. 4.1. QSAR Studies between Ex Vivo ChoKαInhibition and The Electronic Effects of Substituents at Position 4 of The Pyridinium Moieties Two choline-like chains comprising quaternary groups are bound to the central biphenyl system in the HC-3 structure, forming oxazonium rings in solution [ 65 ]. The 1,4-oxazonium moieties of 1 were replaced with pyridinium rings with various substituents at position 4 in order to investigate their electronic effects. The substitution could be carried out at position 2 of the pyridinium ring, in which the possible electronic effects (inductive and mesomer) could also be operational; however, in the latter position, the sterical effect might have masked or altered the electronic effect. Therefore, in our earlier sudies we eliminated the possibility of substitution at the position adjacent to the N + . We have used the 1,2-ethylene(bisbenzyl) moiety as a linker to maintain the number of atoms between the two positively charged nitrogen atoms of 1(Figure 2). Pharmaceutics2021,13,x 4of29 agnosisandprognosishavebeenreviewedrecently.Therefore,althoughthisisavery importantfieldindevelopment,thesetechniqueswillnotbethesubjectofthisreview. 4.DesignofChoKαInhibitorsasaPrecisionMedicineStrategy Thecostofdrugdiscoveryisfartoohightobeledsolelybytrialanderror.In computer‐aideddrugdesign,quantitativestructure–activityrelationship(QSAR)studies areaneffectiveandusefulapproach.AQSARstudyattemptstounderstanddifferences inthebiologicalactivitiesofagroupofcongenersintermsofmolecularchangesinduced bysubstituentchanges. ThemostpotentChoKinhibitorknownintheearly1990swashemicholinium‐3(1, HC‐3)withanIC50forChoKinhibitionexvivoof500μMusingrecombinantChoK,and anIC50againstthehumanHT‐29celllineof2.5mM[64].Ahighaffinityparalyzingres‐ piratoryeffectisinducedbyHC‐3[65],whichisacompetitiveinhibitorofChotransport. Asaresult,1cannotbeusedasanantiproliferativeagentinvivo.Improvingthepotency anddiminishingthetoxicityof1,bychangingitsspecificityandantiproliferativeactivity, wasthefirststepindevelopingpotentialanticancerdrugsthattargetChoK.TheChoK inhibitoryandantiproliferativeactivitiesofsymmetricalbispyridiniumandbisquino‐ liniumcompoundshavebeendefinedusingquantitativestructure–activityrelationships (QSARs)andstructuralparameters.Wewillreviewthisstrategyinthefollowingsec‐ tions. 4.1.QSARStudiesbetweenExVivoChoKαInhibitionandTheElectronicEffectsofSubstituents atPosition4ofThePyridiniumMoieties Twocholine‐likechainscomprisingquaternarygroupsareboundtothecentralbi‐ phenylsystemintheHC‐3structure,formingoxazoniumringsinsolution[65].The 1,4‐oxazoniummoietiesof1werereplacedwithpyridiniumringswithvarioussubstit‐ uentsatposition4inordertoinvestigatetheirelectroniceffects.Thesubstitutioncould becarriedoutatposition2ofthepyridiniumring,inwhichthepossibleelectroniceffects (inductiveandmesomer)couldalsobeoperational;however,inthelatterposition,the stericaleffectmighthavemaskedoralteredtheelectroniceffect.Therefore,inourearlier sudiesweeliminatedthepossibilityofsubstitutionatthepositionadjacenttotheN+.We haveusedthe1,2‐ethylene(bisbenzyl)moietyasalinkertomaintainthenumberofatoms betweenthetwopositivelychargednitrogenatomsof1(Figure2). Figure2.Molecularvariationscarriedoutfromhemicholinium‐3(HC‐3,1). Thecompoundswerescreenedinanexvivosystemusingthehumanrecombinant ChoKasatarget.WedeterminedtheeffectofourcompoundsontheChoKenzymatic Figure 2. Molecular variations carried out from hemicholinium-3 (HC-3, 1). The compounds were screened in an ex vivo system using the human recombinant ChoK as a target. We determined the effect of our compounds on the ChoK enzymatic activity regardless of other confounding factors such as permeability into cells, special cellular conditions, intracellular modifications of compounds or compartmentalization. ChoK inhibitors were also tested on the HT-29 cell line to see how they affected cell proliferation in tumor cells [ 66 ]. This cell line was developed from a colon adenocarcinoma, one of the most common solid human cancers that are resistant to chemotherapy, making these cells suitable for the production of new antitumor drugs. The Ghose–Crippen modified atomic contribution method (ATOMIC5 option) [ 67 ] of the PALLAS 2.0 software [ 68 ] was used to measure the clog P(calculated log P) values of the bis-salts. πspacer is the linker sub-
Pharmaceutics 2021,13, 788 5 of 28 stituent constant and πcat head is the –CH 2 –C 6 H 4 –(CH 2 ) n –C 6 H 4 –CH 2 – group substituent constant, determined with the PALLAS 2.0 program [ 67 ] Ghose–Crippen modified atomic contribution method (ATOMIC5 option) [67]. The activity against the HT-29 cell line is, on average, higher than the activity against ChoK. Thus, the symmetrical bisquaternized salts may be affecting a different part of the pathway in this tumor cell line (see Tables 1–3). Table 1. Compounds 2–12: structures, biological effects and parameter values. Pharmaceutics2021,13,x 5of29 activityregardlessofotherconfoundingfactorssuchaspermeabilityintocells,special cellularconditions,intracellularmodificationsofcompoundsorcompartmentalization. ChoKinhibitorswerealsotestedontheHT‐29celllinetoseehowtheyaffectedcellpro‐ liferationintumorcells[66].Thiscelllinewasdevelopedfromacolonadenocarcinoma, oneofthemostcommonsolidhumancancersthatareresistanttochemotherapy,making thesecellssuitablefortheproductionofnewantitumordrugs.TheGhose–Crippen modifiedatomiccontributionmethod(ATOMIC5option)[67]ofthePALLAS2.0soft‐ ware[68]wasusedtomeasuretheclogP(calculatedlogP)valuesofthebis‐salts.πspaceris thelinkersubstituentconstantandπcatheadisthe–CH2–C6H4–(CH2)n–C6H4–CH2–group substituentconstant,determinedwiththePALLAS2.0program[67]Ghose–Crippen modifiedatomiccontributionmethod(ATOMIC5option)[67]. TheactivityagainsttheHT‐29celllineis,onaverage,higherthantheactivityagainst ChoK.Thus,thesymmetricalbisquaternizedsaltsmaybeaffectingadifferentpartofthe pathwayinthistumorcellline(seeTables1–3). Table1.Compounds2–12:structures,biologicaleffectsandparametervalues. Comp.R4(IC50)exvivo (μM)a (IC50)HT‐29 (μM)aσRbclogPcπR4d 2,MN58B–NMe217.02.00−0.88−2.830.18e 3–NH223.04.00−0.80−4.67−1.23e 4–CH2OH100>100−0.07−4.25−1.03e 5–CH310020.0−0.16−1.820.56e 6–COOH136.7>10000.11−3.50−0.32e 7–C≡N>10002000.08−3.15−0.57e 8–N(Allyl)217.00.55−0.80f−0.441.34e 9‐pyrrolidino20.01.00−0.85f−1.940.59e 10‐piperidino9.600.40−0.89f−0.930.85e 11‐perhydroaz epino15.00.40−0.86f0.091.60e 12–NMePh6.40.34−0.78f0.371.67e aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombi‐ nanthumanChoKwasusedasatarget.cinvitroassaycarriedoutontheHT‐29cellline.bσR: Electronicparameterforresonanceeffects(ref.[69]).cPredictedbyusingtheGhose‐Crippenmod‐ ifiedatomiccontributionsystem(ATOMIC5option,ref.[67])ofthePALLAS2.0program[68].dπR4 =clogPR4−clogPH;clogPvalueshavebeencalculatedusingtheCDRoptionofPALLAS2.0pro‐ gram.eSeeref.[70].fThesevalueswereestimatedby13CNMRspectroscopy(seeref.[71]). AlloftheR4substituentsareeitherelectron‐releasing,neutral,orelec‐ tron‐withdrawing(–NMe2,–NH2,–CH2OH,–Me,–COOH,–C≡N).Theelectronicpa‐ rametersforresonanceeffect(σR)werepublishedbyCharton[69],andthefollowing correlationEquation(1)emergesfromdataofTable1[64,72]: p(IC50)exvivo=3.92(±0.00)−0.92(±0.06)σR n=5,r=0.994,s=0.052,F1,3=250.13(significanceatα<0.001)(1) where,pIC50=−logIC50,bearinginmindthatthehigherthevalueofpIC50themorepotent isthecompound,nisthenumberofcompounds,risthecorrelationcoefficient,sisthe standarddeviation,FistheFratiobetweenthevariancesofobservedandcalculatedac‐ tivities,anddatawithinparenthesesarestandarderrorsofestimate. Comp. R4(IC50)ex vivo (µM) a (IC50)HT-29 (µM) aσRbclog P cπR4 d 2, MN58B –NMe217.0 2.00 −0.88 −2.83 0.18 e 3–NH223.0 4.00 −0.80 −4.67 −1.23 e 4–CH2OH 100 >100 −0.07 −4.25 −1.03 e 5–CH3100 20.0 −0.16 −1.82 0.56 e 6–COOH 136.7 >1000 0.11 −3.50 −0.32 e 7–C≡N >1000 200 0.08 −3.15 −0.57 e 8–N(Allyl)217.0 0.55 −0.80 f−0.44 1.34 e 9-pyrrolidino 20.0 1.00 −0.85 f−1.94 0.59 e 10 -piperidino 9.60 0.40 −0.89 f−0.93 0.85 e 11 -perhydroazepino 15.0 0.40 −0.86 f0.09 1.60 e 12 –NMePh 6.4 0.34 −0.78 f0.37 1.67 e a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK was used as a target, σR : Electronic parameter for resonance effects (ref. [ 69 ]). cin vitro assay carried out on the HT-29 cell line, Predicted by using the Ghose-Crippen modified atomic contribution system (ATOMIC5 option, ref. [ 67 ]) of the PALLAS 2.0 program [ 68 ]. dπR4 = clog P R4 − clog P H ; clog Pvalues have been calculated using the CDR option of PALLAS 2.0 program. e See ref. [70]. f These values were estimated by 13 C NMR spectroscopy (see ref. [71]). Table 2. Structures, activity data and parameter used for the generation of QSAR Equation (3). Pharmaceutics2021,13,x 6of29 Theinhibitorypotencyofthecompoundsisunrelatedtotheirinductiveeffect[64], butiswellcorrelatedwiththeirresonanceeffect.Thisisinlinewithtraditionalchemical principles.ThepositivelychargedringnitrogenisindirectconjugationwithR4.The highertheresonanceeffectofR4,thebetterthedelocalizationofthepositivecharge. Followingthesefindings,researcherslookedforasubstituentwithahigherelec‐ tron‐releasingeffectthanthe–NH2or–NMe2groups.Asaresult,endocyclicamino groupslikepyrrolidino,piperidino,andperhydroazepinomayhaveastrongerelec‐ tron‐releasingeffect.TheN‐methylanilinoandtheN‐diallylgroupswerealsotested. Furthermore,whencomparedtothe–NH2and–NMe2groups,theirhigherlipophilicities canaidintheantiproliferativeactivitiesofevenmoreactivecompounds[70]. 13CNMRspectroscopywasusedtoestimatetheunknownσRdescriptorsforthedi‐ allylamino,pyrrolidino,piperidino,perhydroazepino,andN‐methylanilinomoieties [71].ThetotalityofthetencompoundssynthesizeduntilthengiverisetoEquation(2): p(IC50)exvivo=3.92(±0.11)−1.09(±0.15)σR n=10,r=0.928,s=0.181,F1,8=49.78(significanceatα<0.001)(2) MN58bisamemberofthefirstgenerationofHC‐3derivativesthatwereusedasa prototypeforstudyingChoKαeffectsinbothnormalandtumorcells[2]. 4.2.ExVivoChoKInhibitionandClogP:QSARStudies Wechosestructureswiththefollowingcharacteristicstoinvestigatethepotential effectoflipophilicityonChoKinhibitioninexvivoconditions[72]: (a) Cationicheadswitha“zeroelectroniceffect”,i.e.,hydrogenatposition4,thatallow thepositivechargetobedispersedtoalargeextent.Weusedunsubstitutedquino‐ liniumandisoquinoliniumringstoachievethis. (b) Aralkylspacerswithdifferentnumberofmethylenegroups. ChoKinhibitionactivityisfoundtocorrelatewithlipophilicityforthesecom‐ pounds,asshowninEquation(3): p(IC50)exvivo=4.83(±0.09)+0.81(±0.15)clogP n=7,r=0.922,s=0.151,F1,5=28.57(significanceatα<0.005)(3) OnthebasisofEquation(3),weproposethathydrophobicinteractionsbetweenthe bisalts13–20andChoKcanoccur.TheQSARgreatestcontributionisthatithasprovided acomprehensiveandreasonablycompletequantitativeunderstandingoftheroleofhy‐ drophobicityindrugaction[73].Hydrophobicityisrelatednotonlytoabsorptionand distributionbutalsototheinteractionswithChoKactivesite.Apartfromelectrostatic interactions,itappearsthatthecationicheadsandspacersbetweenthetwopositiveni‐ trogenatoms,hydrophobilicallybindtotheenzymemorestrongly.Non‐covalentinter‐ actionsinaqueoussolutionaredominatedbyhydrophobicinteractions.Athydrophobic surfaces,looselyconnectedwatermoleculeshaveadegreeorderandarethereforeinan undesirableentropicdegree.Theinteractionofadrughydrophobicareaswithitsbind‐ ingsitereleasestheorganizedwatermolecules,resultinginanincreaseinentropy.The planarquinoliniumandisoquinoliniumcationshavebeenshowntobindtoartificialre‐ ceptors,witharomaticrings,morestronglythanthealkylammoniumcations[74]. Table2.Structures,activitydataandparameterusedforthegenerationofQSAREquation(3). Comp.XYn(IC50)exvivo(IC50)HT‐29ClogPb Comp. X Y n(IC50)ex vivo (µM) a (IC50)HT-29 (µM) aClog Pb 13 =N+– =CH– 0 50 10 − 14 =N+– =CH– 1 100 6.02 −0.86 15 =N+– =CH– 2 34 4 −0.43 16 =N+– =CH– 3 9 2.5 0.08 17 =CH– =N+– 0 >100 ND c−0.85 18 =CH– =N+– 1 60 20 −1.00 19 =CH– =N+– 2 60 20 −0.57 20 =CH– =N+– 3 20 2 −0.06 a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK was used as a target, Predicted by using the Ghose-Crippen modified atomic contribution system (ATOMIC5 option, ref. [ 67 ]) of the PALLAS 2.0 program [ 68 ]. cin vitro assay carried out on the HT-29 cell line, ND: Not determined.
Pharmaceutics 2021,13, 788 6 of 28 Table 3. Compounds 21 and 22: structures, biological effects and parameter values. Pharmaceutics2021,13,x 7of29 (μM)a(μM)a 13=N+–=CH–05010− 14=N+–=CH–11006.02−0.86 15=N+–=CH–2344−0.43 16=N+–=CH–392.50.08 17=CH–=N+–0>100NDc−0.85 18=CH–=N+–16020−1.00 19=CH–=N+–26020−0.57 20=CH–=N+–3202−0.06 aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombinanthumanChoKwas usedasatarget.cinvitroassaycarriedoutontheHT‐29cellline.bPredictedbyusingtheGhose‐Crippenmodifiedatomic contributionsystem(ATOMIC5option,ref.[67])ofthePALLAS2.0program[68].cND:Notdetermined. 4.3.CombiningTheElectronicandTheLipophilicEffectsinTheSameMolecules Theelectron‐releasingpotentialofthesubstituentatposition4oftheheteroaromatic cationichead(Table1)andthelipophilicityofthebissalts(Table2)hadasignificantef‐ fectinChoKinhibition.Therefore,inordertoimprovetheinhibitoryefficacy,wecom‐ binedthesetwopropertiesinthesamemoleculeandpreparedcompounds21and22 (Table3).The‐NH2groupisanelectron‐releasinggroupthatshouldbestrong,whilethe pivaloylaminomoietyshouldbeweak.Weusedthecorrespondingvalueofthevery similaracetamidogroupinstead(σR=−0.35)[69]becausethelatterσRvaluewasuna‐ vailable.Theresultsmetourstandards,sowetriedtolinktheChoKinhibitorypotency withthetwodescriptors,resultingEquation(4),inwhichallthecompoundsofTables1– 3wereincluded: p(IC50)exvivo=4.44−0.73(±0.14)σR+0.12(±0.04)clogP n=19,r=0.836,s=0.241,F2,16=18.61(significanceatα<0.001)(4) OurQSARstudiesrevealedthattheefficacyofinhibitorsisdeterminedbyπata particularlocationonthemoleculesratherthantheoveralllogP.Equation(5)suitsvery wellwiththeuseofthesesite‐specificπparameters: p(IC50)exvivo=0.55−1.04(±0.13)σR+0.63(±0.15)πspacer+0.30(±0.08)πcathead n=19,r=0.917,s=0.181,F3,15=26.46(significanceatα<0.001)(5) Table3.Compounds21and22:structures,biologicaleffectsandparametervalues. Comp.R4(IC50)exvivo (μM)a (IC50)HT‐29 (μM)aσRbclogPcπR4d 21–NH210.02.00−0.80−2.13−1.23 22–NHCOBut10.54.74−0.35e1.402.18 aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombi‐ nanthumanChoKfromyeastwasusedasatarget.cinvitroassaycarriedoutontheHT‐29cellline. bσR:Electronicparameterforresonanceeffects(ref.[69]).cPredictedbyusingtheGhose‐Crippen modifiedatomiccontributionsystem(ATOMIC5option,ref.[67])ofthePALLAS2.0program[68]. dπR4=clogPR4−clogPH;clogPvalueshavebeencalculatedusingtheCDRoptionofthePALLAS 2.0program.eWehaveusedtheacetamidovalueinsteadofthepivaloylaminogroupσRvaluebe‐ causethelattervaluewasunavailable(ref.[69]). InordertoobtainEquation(5),wehadtoacceptthatforcompounds2–12,theπcat headwasequaltozero(π2HduetotheH‐2andH‐3protonsofthepyridiniummoiety), Comp. R4(IC50)ex vivo (µM) a (IC50)HT-29 (µM) aσRbclog PcπR4 d 21 –NH210.0 2.00 −0.80 −2.13 −1.23 22 – NHCOBut10.5 4.74 −0.35 e1.40 2.18 a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK from yeast was used as a target, σR : Electronic parameter for resonance effects (ref. [ 69 ]). cin vitro assay carried out on the HT-29 cell line, Predicted by using the Ghose-Crippen modified atomic contribution system (ATOMIC5 option, ref. [ 67 ]) of the PALLAS 2.0 program [ 68 ]. dπR4 = clog P R4 − clog P H ; clog Pvalues have been calculated using the CDR option of the PALLAS 2.0 program. e We have used the acetamido value instead of the pivaloylamino group σRvalue because the latter value was unavailable (ref. [69]). All of the R 4 substituents are either electron-releasing, neutral, or electron-withdrawing (–NMe 2 , –NH 2 , –CH 2 OH, –Me, –COOH, –C ≡ N). The electronic parameters for resonance effect ( σR ) were published by Charton [ 69 ], and the following correlation Equation (1) emerges from data of Table 1[64,72]: p(IC50)ex vivo = 3.92 (±0.00) −0.92 (±0.06) σR n= 5, r= 0.994, s= 0.052, F1,3 = 250.13 (significance at α< 0.001) (1) where, pIC 50 = − log IC 50 , bearing in mind that the higher the value of pIC 50 the more potent is the compound, nis the number of compounds, ris the correlation coefficient, sis the standard deviation, Fis the Fratio between the variances of observed and calculated activities, and data within parentheses are standard errors of estimate. The inhibitory potency of the compounds is unrelated to their inductive effect [ 64 ], but is well correlated with their resonance effect. This is in line with traditional chemical principles. The positively charged ring nitrogen is in direct conjugation with R 4 . The higher the resonance effect of R 4 , the better the delocalization of the positive charge. Following these findings, researchers looked for a substituent with a higher electron-releasing effect than the –NH 2 or –NMe 2 groups. As a result, endocyclic amino groups like pyrrolidino, piperidino, and perhydroazepino may have a stronger electron-releasing effect. The Nmethylanilino and the N-diallyl groups were also tested. Furthermore, when compared to the –NH 2 and –NMe 2 groups, their higher lipophilicities can aid in the antiproliferative activities of even more active compounds [70]. 13 C NMR spectroscopy was used to estimate the unknown σR descriptors for the diallylamino, pyrrolidino, piperidino, perhydroazepino, and N-methylanilino moieties [ 71 ]. The totality of the ten compounds synthesized until then give rise to Equation (2): p(IC50)ex vivo = 3.92 (±0.11) −1.09 (±0.15) σR n= 10, r= 0.928, s= 0.181, F1,8 = 49.78 (significance at α< 0.001) (2) MN58b is a member of the first generation of HC-3 derivatives that were used as a prototype for studying ChoKαeffects in both normal and tumor cells [2]. 4.2. Ex Vivo ChoK Inhibition and Clog P: QSAR Studies We chose structures with the following characteristics to investigate the potential effect of lipophilicity on ChoK inhibition in ex vivo conditions [72]: (a) Cationic heads with a “zero electronic effect”, i.e., hydrogen at position 4, that allow the positive charge to be dispersed to a large extent. We used unsubstituted quinolinium and isoquinolinium rings to achieve this.
Pharmaceutics 2021,13, 788 7 of 28 (b) Aralkyl spacers with different number of methylene groups. ChoK inhibition activity is found to correlate with lipophilicity for these compounds, as shown in Equation (3): p(IC50)ex vivo = 4.83 (±0.09) + 0.81 (±0.15) clog P n= 7, r= 0.922, s= 0.151, F1,5 = 28.57 (significance at α< 0.005) (3) On the basis of Equation (3), we propose that hydrophobic interactions between the bisalts 13 – 20 and ChoK can occur. The QSAR greatest contribution is that it has provided a comprehensive and reasonably complete quantitative understanding of the role of hydrophobicity in drug action [ 73 ]. Hydrophobicity is related not only to absorption and distribution but also to the interactions with ChoK active site. Apart from electrostatic interactions, it appears that the cationic heads and spacers between the two positive nitrogen atoms, hydrophobilically bind to the enzyme more strongly. Non-covalent interactions in aqueous solution are dominated by hydrophobic interactions. At hydrophobic surfaces, loosely connected water molecules have a degree order and are therefore in an undesirable entropic degree. The interaction of a drug hydrophobic areas with its binding site releases the organized water molecules, resulting in an increase in entropy. The planar quinolinium and isoquinolinium cations have been shown to bind to artificial receptors, with aromatic rings, more strongly than the alkylammonium cations [74]. 4.3. Combining The Electronic and The Lipophilic Effects in The Same Molecules The electron-releasing potential of the substituent at position 4 of the heteroaromatic cationic head (Table 1) and the lipophilicity of the bissalts (Table 2) had a significant effect in ChoK inhibition. Therefore, in order to improve the inhibitory efficacy, we combined these two properties in the same molecule and prepared compounds 21 and 22 (Table 3). The -NH 2 group is an electron-releasing group that should be strong, while the pivaloylamino moiety should be weak. We used the corresponding value of the very similar acetamido group instead ( σR = − 0.35) [ 69 ] because the latter σR value was unavailable. The results met our standards, so we tried to link the ChoK inhibitory potency with the two descriptors, resulting Equation (4), in which all the compounds of Tables 1–3were included: p(IC50)ex vivo = 4.44 −0.73 (±0.14) σR+ 0.12 (±0.04) clog P n= 19, r= 0.836, s= 0.241, F2,16 = 18.61 (significance at α< 0.001) (4) Our QSAR studies revealed that the efficacy of inhibitors is determined by π at a particular location on the molecules rather than the overall log P. Equation (5) suits very well with the use of these site-specific πparameters: p(IC50)ex vivo = 0.55 −1.04 (±0.13) σR+ 0.63 (±0.15) πspacer + 0.30 (±0.08) πcat head n= 19, r= 0.917, s= 0.181, F3,15 = 26.46 (significance at α< 0.001) (5) In order to obtain Equation (5), we had to accept that for compounds 2 – 12 , the πcat head was equal to zero ( π2H due to the H-2 and H-3 protons of the pyridinium moiety), while for bisquinolinium and bisisoquinolinium structures 13 – 22 , πcat head is 1.27. πspacer is the –CH 2 –C 6 H 4 –(CH 2 ) n –C 6 H 4 –CH 2 – group substituent constant, which ranges between 5.14 (n = 0 of the spacer) and 6.28 (n= 3 of the spacer). The fact that πspacer has a higher coefficient and values than πcat head leads to the inference that the former is much more significant than the latter in terms of hydrophobicity. Finally, the positive π -term coefficients indicate that hydrophobic moieties and electron-donating groups support ChoK inhibitory action, at least within the context of spacers and heteroaromatic rings used. 4.4. Influence on The Antiproliferative Activity Against The HT-29 Cancerous Cell Line The final research focused on two aspects: (a) the impact that a variation of the linker that links the quinolinium cations with electron-releasing groups at position 4 with other
Pharmaceutics 2021,13, 788 8 of 28 different groups at positions 3, 7, and 8 of the heterocycle (compounds 23 – 63 ) would have on the ex vivo human inhibitory activity; and (b) the impact of the factors that regulate the antiproliferative properties of such compounds [66]. Tables 4–6show the biological results of compounds 23 – 63 , which correspond to series A,B, and C, as a function of the group at position 7 of the quinolinium ring. Table 4. IC50 ChoK and HT-29 values for the bisquinolinium derivatives belonging to series A. Pharmaceutics2021,13,x 8of29 whileforbisquinoliniumandbisisoquinoliniumstructures13–22,πcatheadis1.27.πspaceris the–CH2–C6H4–(CH2)n–C6H4–CH2–groupsubstituentconstant,whichrangesbetween 5.14(n=0ofthespacer)and6.28(n=3ofthespacer).Thefactthatπspacerhasahigher coefficientandvaluesthanπcatheadleadstotheinferencethattheformerismuchmore significantthanthelatterintermsofhydrophobicity.Finally,thepositiveπ‐termcoeffi‐ cientsindicatethathydrophobicmoietiesandelectron‐donatinggroupssupportChoK inhibitoryaction,atleastwithinthecontextofspacersandheteroaromaticringsused. 4.4.InfluenceonTheAntiproliferativeActivityAgainstTheHT‐29CancerousCellLine Thefinalresearchfocusedontwoaspects:(a)theimpactthatavariationofthelinker thatlinksthequinoliniumcationswithelectron‐releasinggroupsatposition4withother differentgroupsatpositions3,7,and8oftheheterocycle(compounds23–63)would haveontheexvivohumaninhibitoryactivity;and(b)theimpactofthefactorsthatreg‐ ulatetheantiproliferativepropertiesofsuchcompounds[66]. Tables4–6showthebiologicalresultsofcompounds23–63,whichcorrespondto seriesA,B,andC,asafunctionofthegroupatposition7ofthequinoliniumring. Table4.IC50ChoKandHT‐29valuesforthebisquinoliniumderivativesbelongingtoseriesA. Comp.LinkerR3R4(IC50)exvivo (μM)a,b (IC50)HT‐29 (μM)a,c 23 Biphenyl‐3,3′‐diyl HAmino1.201.90 24MeAmino11.94.40 25HDimethylamino4.401.60 26HPerhydroazepino0.500.50 27HAnilino1.301.60 28HN‐Methylanilino0.400.80 29H4‐Chloro‐N‐methylanilino2.101.50 30 Biphenyl‐4,4′‐diyl HAmino81.12.20 31MeAmino>2003.30 32HDimethylamino39.71.70 33HPerhydroazepino2.200.50 34HAnilino17.80.70 35HN‐Methylanilino3.000.60 36H4‐Chloro‐N‐methylanilino2.001.20 37 Bibenzyl‐4,4′‐diyl HAmino80.02.00 38HDimethylamino10.20.50 39HPerhydroazepino0.600.30 40HAnilino2.300.30 41HN‐Methylanilino1.400.70 42H4‐Chloro‐N‐methylanilino4.800.70 aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombi‐ nanthumanChoKwasusedasatarget.cinvitroassaycarriedoutontheHT‐29cellline. Comp. Linker R3R4(IC50)ex vivo (µM) a,b (IC50)HT-29 (µM) a,c 23 Biphenyl-3,30-diyl H Amino 1.20 1.90 24 Me Amino 11.9 4.40 25 H Dimethylamino 4.40 1.60 26 H Perhydroazepino 0.50 0.50 27 H Anilino 1.30 1.60 28 HN-Methylanilino 0.40 0.80 29 H 4-Chloro-N-methylanilino 2.10 1.50 30 Biphenyl-4,40-diyl H Amino 81.1 2.20 31 Me Amino >200 3.30 32 H Dimethylamino 39.7 1.70 33 H Perhydroazepino 2.20 0.50 34 H Anilino 17.8 0.70 35 HN-Methylanilino 3.00 0.60 36 H 4-Chloro-N-methylanilino 2.00 1.20 37 Bibenzyl-4,40-diyl H Amino 80.0 2.00 38 H Dimethylamino 10.2 0.50 39 H Perhydroazepino 0.60 0.30 40 H Anilino 2.30 0.30 41 HN-Methylanilino 1.40 0.70 42 H 4-Chloro-N-methylanilino 4.80 0.70 a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK was used as a target. cin vitro assay carried out on the HT-29 cell line. Lipophilicity, molar volume, and steric bulk are all factors that influence molar refractivity (MR) [ 75 ]. To get rational values for the regression coefficients of the resulting QSAR equations, the MR values are usually scaled by a factor of 0.1. With the support of 3D structures, the importance of MR in QSAR equations of certain ligand–enzyme interactions were interpreted. These studies revealed that π -modeled substituents bind in a hydrophobic space. Interactions between these substituents with a polar surface will give rise to a positive sign of MR in a QSAR equation [ 76 , 77 ], while a negative sign or a nonlinear relationship indicates a limited area or steric hindrance at this binding site [75].
Pharmaceutics 2021,13, 788 9 of 28 Table 5. IC50 ChoK and HT-29 values for the bisquinolinium derivatives belonging to series B. Pharmaceutics2021,13,x 9of29 Table5.IC50ChoKandHT‐29valuesforthebisquinoliniumderivativesbelongingtoseriesB. Comp.LinkerR4(IC50)exvivo (μM)a,b (IC50)HT‐29 (μM)a,c 43 Biphenyl‐3,3′‐diyl Amino20.61.90 44Dimethylamino9.600.70 45Pyrrolidino1.200.40 46N‐Methylanilino3.101.00 474‐Chloro‐N‐methylanilino5.701.90 48 Biphenyl‐4,4′‐diyl Amino63.33.20 49Dimethylamino20.60.80 50Pyrrolidino19.82.40 51N‐Methylanilino11.40.50 52,RSM‐932A4‐Chloro‐N‐methylanilino11.41.20 53 Bibenzyl‐4,4′‐diyl Amino80.02.00 54Dimethylamino9.000.27 55Pyrrolidino1.000.20 56N‐Methylanilino3.500.50 574‐Chloro‐N‐methylanilino5.700.80 aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombi‐ nanthumanChoKwasusedasatarget.cinvitroassaycarriedoutontheHT‐29cellline. Lipophilicity,molarvolume,andstericbulkareallfactorsthatinfluencemolarre‐ fractivity(MR)[75].Togetrationalvaluesfortheregressioncoefficientsoftheresulting QSARequations,theMRvaluesareusuallyscaledbyafactorof0.1.Withthesupportof 3Dstructures,theimportanceofMRinQSARequationsofcertainligand–enzymeinter‐ actionswereinterpreted.Thesestudiesrevealedthatπ‐modeledsubstituentsbindina hydrophobicspace.Interactionsbetweenthesesubstituentswithapolarsurfacewillgive risetoapositivesignofMRinaQSARequation[76,77],whileanegativesignoranon‐ linearrelationshipindicatesalimitedareaorsterichindranceatthisbindingsite[75]. Table6.IC50ChoKandHT‐29valuesforthebisquinoliniumderivativesbelongingtoseriesC. Comp.LinkerR4(IC50)exvivo (μM)a,b (IC50)HT‐29 (μM)a,c 58Biphenyl‐3,3′‐diylN‐Methylanilino56.83.80 594‐Chloro‐N‐methylanilino14733.0 60Biphenyl‐4,4′‐diylN‐Methylanilino96.125.2 614‐Chloro‐N‐methylanilino46.119.4 Comp. Linker R4(IC50)ex vivo (µM) a,b (IC50)HT-29 (µM) a,c 43 Biphenyl-3,30-diyl Amino 20.6 1.90 44 Dimethylamino 9.60 0.70 45 Pyrrolidino 1.20 0.40 46 N-Methylanilino 3.10 1.00 47 4-Chloro-N-methylanilino 5.70 1.90 48 Biphenyl-4,40-diyl Amino 63.3 3.20 49 Dimethylamino 20.6 0.80 50 Pyrrolidino 19.8 2.40 51 N-Methylanilino 11.4 0.50 52, RSM-932A 4-Chloro-N-methylanilino 11.4 1.20 53 Bibenzyl-4,40-diyl Amino 80.0 2.00 54 Dimethylamino 9.00 0.27 55 Pyrrolidino 1.00 0.20 56 N-Methylanilino 3.50 0.50 57 4-Chloro-N-methylanilino 5.70 0.80 a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK was used as a target. cin vitro assay carried out on the HT-29 cell line. Table 6. IC50 ChoK and HT-29 values for the bisquinolinium derivatives belonging to series C. Pharmaceutics2021,13,x 9of29 Table5.IC50ChoKandHT‐29valuesforthebisquinoliniumderivativesbelongingtoseriesB. Comp.LinkerR4(IC50)exvivo (μM)a,b (IC50)HT‐29 (μM)a,c 43 Biphenyl‐3,3′‐diyl Amino20.61.90 44Dimethylamino9.600.70 45Pyrrolidino1.200.40 46N‐Methylanilino3.101.00 474‐Chloro‐N‐methylanilino5.701.90 48 Biphenyl‐4,4′‐diyl Amino63.33.20 49Dimethylamino20.60.80 50Pyrrolidino19.82.40 51N‐Methylanilino11.40.50 52,RSM‐932A4‐Chloro‐N‐methylanilino11.41.20 53 Bibenzyl‐4,4′‐diyl Amino80.02.00 54Dimethylamino9.000.27 55Pyrrolidino1.000.20 56N‐Methylanilino3.500.50 574‐Chloro‐N‐methylanilino5.700.80 aAllvaluesarethemeanoftwoindependentdeterminationsperformedinduplicate.bRecombi‐ nanthumanChoKwasusedasatarget.cinvitroassaycarriedoutontheHT‐29cellline. Lipophilicity,molarvolume,andstericbulkareallfactorsthatinfluencemolarre‐ fractivity(MR)[75].Togetrationalvaluesfortheregressioncoefficientsoftheresulting QSARequations,theMRvaluesareusuallyscaledbyafactorof0.1.Withthesupportof 3Dstructures,theimportanceofMRinQSARequationsofcertainligand–enzymeinter‐ actionswereinterpreted.Thesestudiesrevealedthatπ‐modeledsubstituentsbindina hydrophobicspace.Interactionsbetweenthesesubstituentswithapolarsurfacewillgive risetoapositivesignofMRinaQSARequation[76,77],whileanegativesignoranon‐ linearrelationshipindicatesalimitedareaorsterichindranceatthisbindingsite[75]. Table6.IC50ChoKandHT‐29valuesforthebisquinoliniumderivativesbelongingtoseriesC. Comp.LinkerR4(IC50)exvivo (μM)a,b (IC50)HT‐29 (μM)a,c 58Biphenyl‐3,3′‐diylN‐Methylanilino56.83.80 594‐Chloro‐N‐methylanilino14733.0 60Biphenyl‐4,4′‐diylN‐Methylanilino96.125.2 614‐Chloro‐N‐methylanilino46.119.4 Comp. Linker R4(IC50)ex vivo (µM) a,b (IC50)HT-29 (µM) a,c 58 Biphenyl-3,30-diyl N-Methylanilino 56.8 3.80 59 4-Chloro-N-methylanilino 147 33.0 60 Biphenyl-4,40-diyl N-Methylanilino 96.1 25.2 61 4-Chloro-N-methylanilino 46.1 19.4 62 Bibenzyl-4,40-diyl N-Methylanilino 133 7.00 63 4-Chloro-N-methylanilino 57.5 19.5 a All values are the mean of two independent determinations performed in duplicate. b Recombinant human ChoK was used as a target. cin vitro assay carried out on the HT-29 cell line. Equation (6) was obtained after taking into account the volume effects (MR 8 ) (the subscript refers to the location of the substituent), the measured global lipophilicity (clog P), the linker substituent constant, and the electronic parameters ( σR ) of the R 4 substituent for antiproliferative activity: p(IC50)HT-29 =−2.66 −0.03 (±0.00) MR82+ 0.10 (±0.02) clog P+ 1.05 (±0.31)πlinker −3.73 (±0.71) σR n= 40 (series A,Band C), r= 0.920, s= 0.223, F4,35 = 47.856, α< 0.001 (6)
Pharmaceutics 2021,13, 788 16 of 28 Pharmaceutics2021,13,x 17of29 circle.TheATPbindingsiteresiduesaremarkedwitharedcircle.Whilethereispooroverallalignmentofsequences, thereisstrongconservationintheareasoftheATPandcholinebindingsite,ascanbeseenbythenumberofresiduesthat areeitherconservedorsemiconserved. Figure5.AlignmentofthecrystalstructuresofhChok(blue,RCSBaccession#2CKO,dimer); pfChok(cyanRCSBaccession#6YXS[140],monomer);andsChoK(purple.RCSBaccession4R77, monomer).ThebasicN‐terminalandC‐terminaldomainsareshowntobegenerallyconserved. AlignmentandfiguregenerationcarriedoutwiththePyMolPackage. 8.3.TrypanosomabruceiandCholineKinase TrypanosomabruceiisthecausativeagentofHumanAfricanTrypanosomiasisalso knownasAfricnsleepingsickness.ThisparasiteistransmittedbytheTsetsefly.Several hundredthousandsub‐SaharanAfricansareinfectedyearlywiththisparasite,leadingto 10,000deaths.Thisparasiteproducesasinglecholinekinaseisoform(tbChoK)[144].Con‐ ditionalknockoutstbChoKhasdemonstratedthatthisenzymeiscriticalforcellgrowth [145].Theonlyavailabletreatmentsareoftentoxicanddrugresistantstrainshaveemerged [146].Todate,wehaveidentifiedasingledrugdiscoverystudywhichutilizedfragment basedscreeningagainsttbChoKtoidentifyleadsthatwereeffectiveagainstT.bruceiatIC 50 s inthelowμM[145].TheseresultssuggestthattbChoKisapromisingdrugtarget. 9.BacterialPathogensandTheirChoKsandChoKIs Theemergenceofbacterialpathogensthatareresistanttocurrentantibioticthera‐ piesunderlinestheneedforcontinuousresearchintoalternativetherapiesandthedis‐ coveryofnoveldrugtargets.Apromisingstrategyistoexploitinformationregarding drugtargetsineukaryoticsystemsandexploretheiranalogsinbacterialsystems.Thisis anideal,cost‐effective,andefficientstrategy,becausethesametherapiesthatwerede‐ velopedatconsiderableexpenseineukaryoticsystemscanberepurposedasantibiotics andantimicrobials.Promisingexamplesofthisstrategyarebacterialpathwaysofwhich ChoKisapart.ThesepathwaysarebestcharacterizedintheGram‐positiveStreptococcus pneumoniaeandGram‐negativeHaemophiliusinfluenzae. Figure 5. Alignment of the crystal structures of hChok (blue, RCSB accession #2CKO, dimer); pfChok (cyan RCSB accession #6YXS [ 140 ], monomer); and sChoK (purple. RCSB accession 4R77, monomer). The basic N-terminal and C-terminal domains are shown to be generally conserved. Alignment and figure generation carried out with the PyMol Package. Repurposed hChoK inhibitors which selectively inhibited the ethanolamine kinase function of pf-ChoK were also discovered. These blocked the growth of parasites at IC 50 s in the low nanomolar range. These inhibitors also caused a concomitant drop in cellular PE, leading to cell death [139]. Bisquinolinium bromide salt derivatives were discovered via high-throughput compound screening. These inhibitors were found to have affect P. falciparum growth at IC50s<1µM and had IC50s against pf-ChoK in the low µM [141]. 8.2. Leishmania infantum and Choline Kinase Leishmania infantum is the causative agent of Leishmaniasis, a disease found in the tropics and subtropics and transmitted by phlebotomine sand fly. Leishmaniasis affects nearly 1 million people a year [ 142 ]. There has been an initial study whether or not ChoK inhibition is an effective strategy for eliminating L, infantum parasites. A number of quaternary ammonium salts were screened, however the most effective inhibitor, (Niodomethyl-N,N-dimethyl-N-(6,6-diphenylhex-5-en-1-yl) ammonium iodide, only inhibited liChoK in the mM range. Therefore, more work needs to be carried out to discover more powerful inhibitors against liChoK and to assess their ability to kill the parasite [143]. 8.3. Trypanosoma brucei and Choline Kinase Trypanosoma brucei is the causative agent of Human African Trypanosomiasis also known as Africn sleeping sickness. This parasite is transmitted by the Tsetse fly. Several hundred thousand sub-Saharan Africans are infected yearly with this parasite, leading to 10,000 deaths. This parasite produces a single choline kinase isoform (tbChoK) [ 144 ]. Conditional knockouts tbChoK has demonstrated that this enzyme is critical for cell growth [ 145 ]. The only available treatments are often toxic and drug resistant strains have emerged [ 146 ]. To date, we have identified a single drug discovery study which utilized fragment based screening against tbChoK to identify leads that were effective against T. brucei at IC 50 s in the low µ M [ 145 ]. These results suggest that tbChoK is a promising drug target.
Pharmaceutics 2021,13, 788 17 of 28 9. Bacterial Pathogens and Their ChoKs and ChoKIs The emergence of bacterial pathogens that are resistant to current antibiotic therapies underlines the need for continuous research into alternative therapies and the discovery of novel drug targets. A promising strategy is to exploit information regarding drug targets in eukaryotic systems and explore their analogs in bacterial systems. This is an ideal, cost-effective, and efficient strategy, because the same therapies that were developed at considerable expense in eukaryotic systems can be repurposed as antibiotics and antimicrobials. Promising examples of this strategy are bacterial pathways of which ChoK is a part. These pathways are best characterized in the Gram-positive Streptococcus pneumoniae and Gram-negative Haemophilius influenzae. 9.1. Streptococcus pneumoniae and Choline Kinase Choline is an essential nutrient for S. pneumoniae [ 147 ] and the choline kinase of S. pneumoniae (sChoK) is an essential enzyme [ 148 ]. This pathogen meets its needs for this metabolite in part by scavenging choline molecules from host cells using the Pce phosphodiesterase [ 149 ]. As in eukaryotes, the ChoK of S. pneumoniae (sChoK) phosphorylates choline (Cho) into phosphorylcholine (PCho). However, the subsequent metabolic pathways diverge from the Kennedy pathway: PCho is utilized in the pathways responsible for the synthesis of bacterial teichoic acids (Figure 6). Pharmaceutics2021,13,x 18of29 9.1.StreptococcuspneumoniaeandCholineKinase CholineisanessentialnutrientforS.pneumoniae[147]andthecholinekinaseofS. pneumoniae(sChoK)isanessentialenzyme[148].Thispathogenmeetsitsneedsforthis metaboliteinpartbyscavengingcholinemoleculesfromhostcellsusingthePcephos‐ phodiesterase[149].Asineukaryotes,theChoKofS.pneumoniae(sChoK)phosphorylates choline(Cho)intophosphorylcholine(PCho).However,thesubsequentmetabolic pathwaysdivergefromtheKennedypathway:PChoisutilizedinthepathwaysrespon‐ sibleforthesynthesisofbacterialteichoicacids(Figure6). Figure6.ThesChokpathwayandhowitfitsintotheoverallprocessofteichoicacidassembly. PChomolecules(purple)aretransferredfromCDP‐ChototheGalNacresidues(green)ofglycan precursors(RXN3).Theglycanprecursorsthenassembletoformpre‐teichoicacids(RXN4),which becomeeitherLTAorCTAmoleculesdependingontheirfinaldestination.LTAandCTAareem‐ beddedinthemembraneorcellwall,respectively. Thesearelipoteichoicacid(LTA)andcellwallteichoicacid(CTA)[150,151].The LicCcytidylyltransferasecatalyzestheproductionofCDP‐cholinefromPCho(Figure6, RXN2).TheLicD1andLicD2PChotransferasestransferthePChotothe N‐acetylgalactosamine(GalNac)residuesfoundonpre‐teichoicacidglycanprecursors (Figure6,RXN3)[152]. Theprecursersarethenassembledbyanuncharacterizedenzymeintovariably sizedpre‐teichoicacidmolecules,(Figure6,RXN4)[152].Pre‐teichoicacidtransport acrossthecellmembranethenmediatedbytheteichoicacidflippaseTacFtobecomeei‐ therLTAorCTA.LTAandCTAarechemicallysimilar,withthedifferencethatLTAis embeddedinthecellmembraneviaaglycolipidlipidanchorthroughtheactionofthe TacLligase[153],andCTAiscovalentlyattachedtothecellwallviapeptidoglycan moleculesthroughtheactionofLCPphosphotransferases[145].MeanwhileLTAisa knownvirulencefactorandtheLTAsynthesispathwayhasdemonstratedtobeasource ofdrugtargets[154]. Significantly,LidD2knockoutsofS.pneumoniaediminishvirulenceduetolimitations inthenumberofPChocontainingteichoicacidsfoundonthecellsurface[155].PChodec‐ orationofthecellsurfaceisrequiredfornormalcell‐functioningbecausetheChomoiety anchorscholinebindingproteins(CBPs)suchasmureinhydrolasesLytAandLytaAB, whicharecriticalfactorsforcelldivision[156].Inaddition,itanchorsthecholinebinding proteinA(CBPA)whichisadeterminantofvirulence[157].ManyotherCBPsareinvolved incolonizationandevensepsis[158].Therefore,thecholinemetabolicpathways,ofwhich sChoKisanintegralpart,playanimportantroleinS.pneumoniaegrowthandinvasion. DisruptingtheteichoicacidproductionpathwaycanalsodisruptS.pneumonaiecell growth.Therefore,itfollowsthatdisruptionofanyelementofthispathway,suchas sChoK,isapromisingtherapeuticavenuetofollow.Inaddition,thesametoolsusedthe toinhibiteukaryoticChoKsarelikelytobeeffectiveagainstsChoK,particularlythose knowtointeractwiththesubstratebindingsites.ThisisbecausethecholineandATP Figure 6. The sChok pathway and how it fits into the overall process of teichoic acid assembly. PCho molecules (purple) are transferred from CDP-Cho to the GalNac residues (green) of glycan precursors (RXN 3). The glycan precursors then assemble to form pre-teichoic acids (RXN 4), which become either LTA or CTA molecules depending on their final destination. LTA and CTA are embedded in the membrane or cell wall, respectively. These are lipoteichoic acid (LTA) and cell wall teichoic acid (CTA) [ 150 , 151 ]. The LicC cytidylyl transferase catalyzes the production of CDP-choline from PCho (Figure 6, RXN 2). The LicD1 and LicD2 PCho transferases transfer the PCho to the N-acetylgalactosamine (GalNac) residues found on pre-teichoic acid glycan precursors (Figure 6, RXN 3) [152]. The precursers are then assembled by an uncharacterized enzyme into variably sized pre-teichoic acid molecules, (Figure 6, RXN 4) [152]. Pre-teichoic acid transport across the cell membrane then mediated by the teichoic acid flippase TacF to become either LTA or CTA. LTA and CTA are chemically similar, with the difference that LTA is embedded in the cell membrane via a glycolipid lipid anchor through the action of the TacL ligase [ 153 ], and CTA is covalently attached to the cell wall via peptidoglycan molecules through the action of LCP phosphotransferases [ 145 ]. Meanwhile LTA is a known virulence factor and the LTA synthesis pathway has demonstrated to be a source of drug targets [154]. Significantly, LidD2 knockouts of S. pneumoniae diminish virulence due to limitations in the number of PCho containing teichoic acids found on the cell surface [ 155 ]. PCho decoration of the cell surface is required for normal cell-functioning because the Cho moiety anchors choline binding proteins (CBPs) such as murein hydrolases LytA and LytaAB, which are critical factors for cell division [ 156 ]. In addition, it anchors the choline binding
Pharmaceutics 2021,13, 788 18 of 28 protein A (CBPA) which is a determinant of virulence [ 157 ]. Many other CBPs are involved in colonization and even sepsis [ 158 ]. Therefore, the choline metabolic pathways, of which sChoK is an integral part, play an important role in S. pneumoniae growth and invasion. Disrupting the teichoic acid production pathway can also disrupt S. pneumonaie cell growth. Therefore, it follows that disruption of any element of this pathway, such as sChoK, is a promising therapeutic avenue to follow. In addition, the same tools used the to inhibit eukaryotic ChoKs are likely to be effective against sChoK, particularly those know to interact with the substrate binding sites. This is because the choline and ATP sites of sChoks are highly conserved with respect to eukaryotic isoforms (Figure 4) as are the tertiary structures. With this rationale in mind, the well-known choline analog and competitive hChoKI was studied and found to inhibit sChoK and cell growth with a high IC 50 (in the mM range). This was also found to disrupt lipoteichoic acid production, which was consistent with the cell wall deformations that were observed by scanning electron microscopy. This established sChoK as a putative drug target [ 159 – 161 ]. Later, the hChoK inhibitors MN58b and RSM-932A were assayed and found to inhibit both sChoK activity and S. pneumoniae cell growth at moderate and low µ M concentrations, to modulate lipoteichoic acid production and assembly, and to affect the cell wall. Interestingly, in the case of in vitro studies with purified sChoK, MN58 functions as a competitive inhibitor against both choline and ATP, while RSM932A functions as a competitive and non-competitive inhibitor against choline and ATP, respectively. This result against RSM932A is in sharp contrast to the observed uncompetitive behavior of this drug in the case of pf-ChoK. This indicates that there are likely enough differences in the largely conserved active sites of the different ChoKs to modulate ChoKI behavior. This promising outcome suggests that it will likely be possible to rationally design ChoKI that are effective but selective between the different isoforms. 9.2. Haemophilius influenzae ChoK The choline kinase of Haemophilius influenzae (hiChoK) is another promising target to explore for the design of inhibitors against this pathogen. Cho is not a nutritional requirement in the case of the Gram-negative H. influenzae, as it is in S. pneumoniae. Nevertheless, this pathogen does uptake this metabolite from its surroundings [ 162 ], including from host cells [ 163 ]H. influenzae produces PCho and uses it to decorate its lipopolysaccharides (LPS). PCho containing LPS molecules mediate the interactions between the pathogen and host and help the pathogen avoid host immune responses [ 149 ]. By mimicking the phosphatidylcholine of eukaryotic host cells, these LPS molecules shield the pathogen from attack from host cell produced anti-microbial peptides [ 164 ] and antibodies [ 165 ]. It is important to note that the LicA gene that codes for hiChoK is upregulated during the colonization steps of H. influenzae [ 166 ]. PCho is also important for pathogenesis of H. influenzae, because this molecule mediates cell adhesion to the host PAF receptor [ 162 ] in a step which leads to invasion the respiratory tract [ 149 ]. In addition, PCho decoration of lipoteichoic acid has been shown to reduce the binding of a bacteriocidal antibodies [ 164 ]. 10. Rheumatoid Arthritis and Inflammation In animal models of rheumatoid arthritis (RA), ChoKI α s are also very successful, suggesting that these drugs have the potential to be used as potent therapeutics in inflammatory diseases [ 167 ]. Cell migration and resistance to apoptosis of cultured fibroblast-like synoviocytes (FLS), which are involved in cartilage destruction in RA, were suppressed by MN58b , a powerful specific ChoK α I. ChoK α inhibition with MN58b significantly decreased FLS migration and proliferation, and abrogated joint inflammation and damage in either pretreatment or proven disease protocols in the K/BxN arthritis mouse model [ 167 ]. These findings are in keeping with studies that postulate that synovial inflammation, hyperplasia, and joint destruction are all hallmarks of RA [ 168 ]. Phosphoinositide 3-kinase (PI3K)/Akt and Mitogen-activated protein kinase (MAPK) are involved in the control of
Pharmaceutics 2021,13, 788 19 of 28 FLS activity in RA, including matrix metalloproteinases (MMP) expression, synoviocyte growth and survival, and are the subject of therapeutic intervention in RA [ 169 ]. As previously mentioned, MN58b selectively inhibits ChoK α , which inhibits MAPK and PI3K/Akt signalling [103–105], while PI3K inhibition affects ChoK function [170]. Finally, disruption of ChoK α activity affects phosphorylation of Rb, downmodulates cyclin D1, and interferes with PDGF signaling in FLS, as previously demonstrated in other cell systems [62]. Using the C. Elegans as a model for Parkinson’s disease (PD), it has been suggested that ChoK may be an important player in PD since inhibition of its expression by siRNA significantly reduces the nuclear localization of Daf-16. ChoK silenced worms showed decreased lipid content, a parameter of immense importance in PD-associated endpoints [113]. Furthermore, ChoK α inhibitors have been shown to be very active in the treatment of several animal models for human diseases related to dysregulation of the inflammasome [ 171 ]. Thus, in the LPS-induced septic shock model, ChoK α inhibition dramatically reduced LPS-induced death, an effect associated with reduced levels of circulating IL1 β . Additionally, ChoK α inhibition has therapeutic effects in three models of chronic syndromes associated with cryopyrin (CAPS): Muckle–Wells syndrome (MWS), familial cold autoinflammatory syndrome (FCAS) and neonatal-onset multisystem inflammatory disease (NOMID). These three syndromes are a consequence of mutations in the NLRP3 gene that cause chronic activation of the inflammasome [ 171 ]. Therefore, ChoK α inhibition may play an important role in the treatment of a diversity of human diseases related to an altered inflammatory response. 11. Future Perspectives: ChoKIs Development for the Treatment of Cancer, Arthritis, Inflammation, Infections and Beyond ChoK α is overexpressed in a large number of human tumors, and blocking its function causes cancer cells to die while non-tumorigenic cells undergo a reversible cell cycle arrest. Based on a novel mechanism of action, it makes possible combination of ChoK inhibitors with many current therapeutic approaches. Thus, specific targeting of this enzyme is a novel strategy for the treatment of many cancer types. Other diseases for which ChoK α inhibitors have been shown to be effective include malaria, rheumatoid arthritis, inflammation, parasites, and pathogenic bacteria. One of our ChoK α inhibitors, RSM932A/TCD717 , has reached the first in human Phase I clinical trial [ 79 ]. This first step towards the use of this novel family of drugs in the regular clinical practice has been a milestone that has paved the way for future development of successful drugs. Although further work is needed to finally reach the clinic as a standard of care in any of the several pathological conditions where ChoKIs have been proposed as a therapeutic strategy, the most important issues as the preclinical and the toxicology studies have been already addressed. There is limited research into ChoKIs of parasites and therefore a limited number of promising ChoK drug leads. Likewise, much of this research has not yet reached the pre-clinical stage, nor therefore Phase I trials. Given this dearth in research, repurposing hChoK inhibitors may be good strategy for discovering novel treatments for parasites. As hChoKIs become cleared for use in humans to treat pathologies such as cancer, rheumatoid arthritis and inflammation, the immense amount of research required to reach that point can be exploited to find new treatments for illnesses such as malaria. Despite the importance of hiChoK in pathogenesis of H. influenzae, the idea that hiChoK could function as a therapeutic target, remains untested. While, hiChoK has not yet been identified as an enzyme critical for cell growth, it is likely that inhibitors for this enzyme could be developed and employed as therapeutic adjuncts designed to assist the immune system in defending against infection. Meanwhile, the improvements in sChoK inhibition observed with MN58b and RSM932A fully validated the idea that inhibitors of eukaryotic ChoKs could be repurposed for use against prokaryotes [ 172 , 173 ]. However, sChoK still needs to be fully validated as the
Pharmaceutics 2021,13, 788 20 of 28 drug target responsible for these physiological effects and further metabalonomic and/or mutational studies need to be carried to fully establish this system. In addition, more refined research needs to be carried out to fully distinguish the effects that these drugs might have on other choline binding proteins such as the autolysin LytA and many others [ 156 ]. The autolysin LytA mediates cell autolysis in S. pneumoniae, therefore, as an initial step, any putative sChoK inhibitor should also be assayed for its ability to modulate the autolytic process [ 171 , 172 ]. For example, HC-3 is known to modulate autolysis [ 174 ], but MN58b and RSM932A do not. Precision about the actual drug targets is necessary. However, more than one target for putative sChoK inhibitor would actually be highly beneficial. Multiple disruptions in more than one part of the choline metabolism or choline binding pathway may help prevent the development of new resistance to this novel family of inhibitors. ChoK and other CBPs may be produced by other many bacterial species. Therefore, ChoKIs, like many other drugs [ 175 ] may affect species found in the human microbiome. Future studies will need to take these possible effects into account in determine the effects of ChoKIs on human health. The final goal should be to develop a ChoKI therapeutic that is specific to microbial choline kinases, and preferably targeted against specific pathogenic species. Preventing unintended consequences will require a more complete and rationalized understanding of the structural differences between the different ChoKs and CPBs of the various species. Recently, a functional relationship between the immune checkpoint programmed cell death protein-1 (PD-1) and ChoK α has been established [ 176 ]. PD-L1 regulates metabolism through ChoK α , COX-2, and TGFβ , a new twist in the application of combinatorial therapies targeting both the immune checkpoints and modulation of critical enzymes involved in cancer metabolism. Thus, a new door has been opened to further investigate a potential relationship of ChoKαand the immune response. Finally, since the mechanism of action of ChoKαinhibitors is based on the inhibition of the production of the ILs responsible for these inflammatory processes [ 166 , 170 ] there is another potential application. ChoK α Is reduce macrophage activation and NLRP3 inflammasome attenuating the inflammatory response by modulating the production of IL-6, IL-1 β and IL-18. All these processes are carried out by inhibiting ChoK α , the therapeutic target of our inhibitors. COVID-19 patients present in its most acute phases the Severe Acute Respiratory Syndrome (SARS), caused by strong bilateral lung inflammation, responsible for their death. This inflammatory process is associated with a strong increase in cytokine levels mediated by the inflammasome [ 177 ] process that coincides with studies carried out with SARS-Cov in which a relationship with the activation of the NLRP3 inflammasome was observed [ 178 ]. Based on these observations, it is reasonable to assume that inhibition of the phosphatidylcholine synthesis pathway through inhibition of the ChoK α enzyme could be an effective treatment in COVID-19 patients in the most advanced stages of the disease. This also deserves further investigation. Author Contributions: Conceptualization, J.C.L., T.Z. and J.M.C.; writing—original draft preparation, J.C.L., T.Z. and J.M.C.; writing—review and editing, J.C.L., T.Z. and J.M.C. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by CSIC, grant number PIE202020E041; and in part by the NIFA through the Agricultural Research Program at North Carolina Agricultural and Technical State University (Evans-Allen Program, project number NC.X-291-5-15-170-1). Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: No new data were created or analyzed in this study. Data sharing is not applicable to this article. Conflicts of Interest: The authors declare no conflict of interest.
Pharmaceutics 2021,13, 788 21 of 28 Abbreviations ChoKα, ChoKβCholine Kinase αor β PC phosphatidylcholine PCho phosphorylcholine ChoKI ChoK inhibitor ChoKαI ChoKα-specific inhibitor siRNA small interfering RNA 5-FU 5-Fluorouracil PI3K phosphatidylinositol 3 kinase MAPK mitogen activated protein kinase EGFR epidermal growth factor receptor NMR nuclear magnetic resonance MRS magnetic resonance spectroscopy PET positron emission tomography HR-MAS MRS high-resolution magic angle spinning magnetic resonance spectroscopy FLS fibroblast-like synoviocytes RA rheumatoid arthritis References 1. DeBerardinis, R.J.; Thompson, C.B. Cellular metabolism and disease: What do metabolic outliers teach us? Cell 2012 ,148, 1132–1144. [CrossRef] [PubMed] 2. Lacal, J.C. Choline kinase: A novel target to search for antitumoral drugs. IDrugs 2001,4, 419–426. 3. Lacal, J.C. Choline kinase as a precision medicine target for therapy in cancer, autoimmune diseases and malaria. Precis. Med. 2015,1. [CrossRef] 4. Aoyama, C.; Liao, H.; Ishidate, K. Structure and function of choline kinase isoforms in mammalian cells. Prog. Lipid Res. 2004 ,43, 266–281. [CrossRef] [PubMed] 5. Fagone, P.; Jackowski, S. Phosphatidylcholine and the CDP–choline cycle. Biochim. Biophys. Acta 2013 ,1831, 523–532. [CrossRef] [PubMed] 6. Chen, X.; Qiu, H.; Wang, C.; Yuan, Y.; Tickner, J.; Xu, J.; Zou, J. Molecular structure and differential function of choline kinases CHKalpha and CHKbeta in musculoskeletal system and cancer. Cytokine Growth Factor Rev. 2017 ,33, 65–72. [CrossRef] [PubMed] 7. Wu, G.; Aoyama, C.; Young, S.G.; Vance, D.E. Early Embryonic Lethality Caused by Disruption of the Gene for Choline Kinase α , the First Enzyme in Phosphatidylcholine Biosynthesis. J. Biol. Chem. 2008,283, 1456–1462. [CrossRef] [PubMed] 8. Lin, Y.-C.; Liu, Y.-C.; Nakamura, Y. The Choline/Ethanolamine Kinase Family in Arabidopsis: Essential Role of CEK4 in Phospholipid Biosynthesis and Embryo Development. Plant Cell 2015,27, 1497–1511. [CrossRef] [PubMed] 9. Sher, R.B.; Aoyama, C.; Huebsch, K.A.; Ji, S.; Kerner, J.; Yang, Y.; Frankel, W.N.; Hoppel, C.L.; Wood, P.A.; Vance, D.E.; et al. A Rostrocaudal Muscular Dystrophy Caused by a Defect in Choline Kinase Beta, the First Enzyme in Phosphatidylcholine Biosynthesis. J. Biol. Chem. 2006,281, 4938–4948. [CrossRef] 10. Sayed-Zahid, A.A.; Sher, R.B.; Sukoff, R.; Anderson, L.C.; Patenaude, K.E.; Cox, G.A.; Sayed, A.A. Functional rescue in a mouse model of congenital muscular dystrophy with megaconial myopathy. Hum. Mol. Genet. 2019,28, 2635–2647. [CrossRef] 11. Kular, J.; Tickner, J.C.; Pavlos, N.J.; Viola, H.M.; Abel, T.; Lim, B.S.; Yang, X.; Chen, H.; Cook, R.; Hool, L.C.; et al. Choline Kinase Mutant Mice Exhibit Reduced Phosphocholine, Elevated Osteoclast Activity, and Low Bone Mass. J. Biol. Chem. 2015 ,290, 1729–1742. [CrossRef] 12. Li, Z.; Wu, G.; Van Der Veen, J.N.; Hermansson, M.; Vance, D.E. Phosphatidylcholine metabolism and choline kinase in human osteoblasts. Biochim. Biophys. Acta 2014,1841, 859–867. [CrossRef] [PubMed] 13. Mitsuhashi, S.; Nishino, I. Megaconial congenital muscular dystrophy due to loss-of-function mutations in choline kinase β .Curr. Opin. Neurol. 2013,26, 536–543. [CrossRef] 14. Lin, Y.-C.; Araguirang, G.E.; Ngo, A.H.; Lin, K.-T.; Angkawijaya, A.E.; Nakamura, Y. The Four Arabidopsis Choline/Ethanolamine Kinase Isozymes Play Distinct Roles in Metabolism and Development. Plant Physiol. 2020,183, 152–166. [CrossRef] [PubMed] 15. Gallego-Ortega, D.; De Molina, A.R.; Ramos, M.; Valdes-Mora, F.; Sarmentero-Estrada, J.; Lacal, J.C. Differential role of choline kinase alpha and beta isoforms in human carcinogenesis. Eur. J. Cancer Suppl. 2008,6, 26. [CrossRef] 16. Malito, E.; Sekulic, N.; Too, W.C.S.; Konrad, M.; Lavie, A. Elucidation of Human Choline Kinase Crystal Structures in Complex with the Products ADP or Phosphocholine. J. Mol. Biol. 2006,364, 136–151. [CrossRef] [PubMed] 17. Hong, B.S.; Allali-Hassani, A.; Tempel, W.; MacKenzie, F.; Dimov, S.; Vedadi, M.; Park, H.W. Crystal Structures of Human Choline Kinase Isoforms in Complex with Hemicholinium-3: Single amino acid near the active site influences inhibitor sensitivity. J. Biol. Chem. 2010,285, 16330–16340. [CrossRef] [PubMed] 18. Aoyama, C.; Ohtani, A.; Ishidate, K. Expression and characterization of the active molecular forms of choline/ethanolamine kinase-alpha and -beta in mouse tissues, including carbon tetrachloride-induced liver. Biochem. J. 2002 ,363, 777–784. [CrossRef]
Pharmaceutics 2021,13, 788 22 of 28 19. Gruber, J.; Too, W.C.S.; Wong, M.T.; Lavie, A.; McSorley, T.; Konrad, M. Balance of human choline kinase isoforms is critical for cell cycle regulation. FEBS J. 2012,279, 1915–1928. [CrossRef] 20. Ramírez de Molina, A.; Gutiérrez, R.; Ramos, M.A.; Silva, J.M.; Silva, J.; Bonilla, F.; Sánchez, J.J.; Lacal, J.C. Increased choline kinase activity in human breast carcinomas: Clinical evidence for a potential novel antitumoral strategy. Oncogene 2002 ,21, 4317–4322. [CrossRef] 21. Ramírez de Molina, A.; Rodríguez-González, A.; Gutiérrez, R.; Martínez-Piñeiro, L.; Sánchez, J.J.; Bonilla, F.; Rosell, R.; Lacal, J.C. Overexpression of choline kinase is a frequent feature in human tumor-derived cell lines and in lung, prostate and colorectal human cancers. Biochem. Biophys. Res. Commun. 2002,296, 580–583. [CrossRef] 22. Ramírez de Molina, A.; Sarmentero-Estrada, J.; Belda-Iniesta, C.; Taron, M.; Ramírez de Molina, V.; Cejas, P.; Skrzypski, M.; Gallego-Ortega, D.; de Castro, J.; Casado, E.; et al. Expression of choline kinase alfa to predict outcome in patients with early-stage non-small-cell lung cancer: A retrospective study. Lancet Oncol. 2007,8, 889–897. [CrossRef] 23. Zhang, L.; Chen, P.S.; Li, G.; Bao, W. CHKA mediates the poor prognosis of lung adenocarcinoma and acts as a prognostic indicator. Oncol Lett. 2016,12, 1849–1853. [CrossRef] 24. Hu, L.; Wang, R.-Y.; Cai, J.; Feng, D.; Yang, G.-Z.; Xu, Q.-G.; Zhai, Y.-X.; Zhang, Y.; Zhou, W.-P.; Cai, Q.-P. Overexpression of CHKA contributes to tumor progression and metastasis and predicts poor prognosis in colorectal carcinoma. Oncotarget 2016 ,7, 66660–66678. [CrossRef] [PubMed] 25. Hernando, E.; Sarmentero-Estrada, J.; Koppie, T.; Iniesta, C.B.; De Molina, A.R.; Cejas, P.; Ozu, C.; Le, C.; Sánchez, J.J.; GonzálezBarón, M.; et al. A critical role for choline kinaseα in the aggressiveness of bladder carcinomas. Oncogene 2009 ,28, 2425–2435. [CrossRef] [PubMed] 26. Challapalli, A.; Trousil, S.; Hazell, S.; Kozlowski, K.; Gudi, M.; O Aboagye, E.; Mangar, S. Exploiting altered patterns of choline kinase-alpha expression on human prostate tissue to prognosticate prostate cancer. J. Clin. Pathol. 2015 ,68, 703–709. [CrossRef] [PubMed] 27. Burch, T.C.; Isaac, G.; Booher, C.L.; Rhim, J.S.; Rainville, P.; Langridge, J.; Baker, A.; Nyalwidhe, J.O. Comparative Metabolomic and Lipidomic Analysis of Phenotype Stratified Prostate Cells. PLoS ONE 2015,10, e0134206. [CrossRef] [PubMed] 28. Nanni, S.; Aiello, A.; Salis, C.; Re, A.; Cencioni, C.; Bacci, L.; Pierconti, F.; Pinto, F.; Ripoli, C.; Ostano, P.; et al. Metabolic Reprogramming by Malat1 Depletion in Prostate Cancer (Stage 2). Cancers 2020,13, 15. [CrossRef] 29. Iorio, E.; Mezzanzanica, D.; Alberti, P.; Spadaro, F.; Ramoni, C.; D’Ascenzo, S.; Millimaggi, D.; Pavan, A.; Dolo, V.; Canevari, S.; et al. Alterations of Choline Phospholipid Metabolism in Ovarian Tumor Progression. Cancer Res. 2005 ,65, 9369–9376. [CrossRef] 30. Trousil, S.; Lee, P.; Pinato, D.J.; Ellis, J.K.; Dina, R.; Aboagye, E.O.; Keun, H.C.; Sharma, R. Alterations of Choline Phospholipid Metabolism in Endometrial Cancer Are Caused by Choline Kinase Alpha Overexpression and a Hyperactivated Deacylation Pathway. Cancer Res. 2014,74, 6867–6877. [CrossRef] 31. Mazarico, J.M.; Sánchez-Arévalo Lobo, V.J.; Favicchio, R.; Greenhalf, W.; Costello, E.; Carrillo-de Santa Pau, E.; Marqués, M.; Lacal, J.C.; Aboagye, E.; Real, F.X. Choline Kinase Alpha (CHK α ) as a Therapeutic Target in Pancreatic Ductal Adenocarcinoma: Expression, Predictive Value, and Sensitivity to Inhibitors. Mol. Cancer Ther. 2016,15, 323–333. [CrossRef] 32. Mok, S.R.S.; Mohan, S.; Grewal, N.; Elfant, A.B.; Judge, T.A. A genetic database can be utilized to identify potential biomarkers for biphenotypic hepatocellular carcinoma-cholangiocarcinoma. J. Gastrointest. Oncol. 2016,7, 570–579. [CrossRef] [PubMed] 33. Carrillo-Reixach, J.; Torrens, L.; Simon-Coma, M.; Royo, L.; Domingo-Sàbat, M.; Abril-Fornaguera, J.; Akers, N.; Sala, M.; Ragull, S.; Arnal, M.; et al. Epigenetic footprint enables molecular risk stratification of hepatoblastoma with clinical implications. J. Hepatol. 2020,73, 328–341. [CrossRef] [PubMed] 34. Kwee, S.A.; Hernandez, B.; Chan, O.; Wong, L. Choline Kinase Alpha and Hexokinase-2 Protein Expression in Hepatocellular Carcinoma: Association with Survival. PLoS ONE 2012,7, e46591. [CrossRef] 35. Lin, X.-M.; Hu, L.; Gu, J.; Wang, R.-Y.; Li, L.; Tang, J.; Zhang, B.-H.; Yan, X.-Z.; Zhu, Y.-J.; Hu, C.-L.; et al. Choline Kinase α Mediates Interactions Between the Epidermal Growth Factor Receptor and Mechanistic Target of Rapamycin Complex 2 in Hepatocellular Carcinoma Cells to Promote Drug Resistance and Xenograft Tumor Progression. Gastroenterology 2017 ,152, 1187–1202. [CrossRef] [PubMed] 36. Ma, W.; Wang, S.; Zhang, T.; Zhang, E.Y.; Zhou, L.; Hu, C.; Yu, J.J.; Xu, G. Activation of choline kinase drives aberrant choline metabolism in esophageal squamous cell carcinomas. J. Pharm. Biomed. Anal. 2018,155, 148–156. [CrossRef] 37. Xiong, J.; Bian, J.; Wang, L.; Zhou, J.-Y.; Wang, Y.; Zhao, Y.; Wu, L.-L.; Hu, J.-J.; Li, B.; Chen, S.-J.; et al. Dysregulated choline metabolism in T-cell lymphoma: Role of choline kinase-αand therapeutic targeting. Blood Cancer J. 2015,5, e287. [CrossRef] 38. Gobeil Odai, K.; O’Dwyer, C.; Steenbergen, R.; Shaw, T.A.; Renner, T.M.; Ghorbani, P.; Rezaaifar, M.; Han, S.; Langlois, M.A.; Crawley, A.M.; et al. In Vitro Hepatitis C Virus Infection and Hepatic Choline Metabolism. Viruses. 2020,12, 108. [CrossRef] 39. Li, H.; Zhu, W.; Zhang, L.; Lei, H.; Wu, X.; Guo, L.; Chen, X.; Wang, Y.; Tang, H. The metabolic responses to hepatitis B virus infection shed new light on pathogenesis and targets for treatment. Sci. Rep. 2015,5, 8421. [CrossRef] 40. Ling, C.S.; Yin, K.B.; Cun, S.T.W.; Ling, F.L. Expression profiling of choline and ethanolamine kinases in MCF7, HCT116 and HepG2 cells, and the transcriptional regulation by epigenetic modification. Mol. Med. Rep. 2014,11, 611–618. [CrossRef] 41. Kumar, M.; Arlauckas, S.P.; Saksena, S.; Verma, G.; Ittyerah, R.; Pickup, S.; Popov, A.V.; Delikatny, E.J.; Poptani, H. Magnetic Resonance Spectroscopy for Detection of Choline Kinase Inhibition in the Treatment of Brain Tumors. Mol. Cancer Ther. 2015 ,14, 899–908. [CrossRef] [PubMed]
Pharmaceutics 2021,13, 788 23 of 28 42. Yue, F.; Zou, Y.; Sun, S.; Wang, Z.; Huang, L.; Ma, H. Knockdown of choline kinase α (CHKA) inhibits the proliferation, invasion and migration of human U87MG glioma cells. Chin. J. Cell. Mol. Immunol. 2020,36, 724–728. 43. Penet, M.-F.; Shah, T.; Bharti, S.; Krishnamachary, B.; Artemov, D.; Mironchik, Y.; Wildes, F.; Maitra, A.; Bhujwalla, Z.M. Metabolic Imaging of Pancreatic Ductal Adenocarcinoma Detects Altered Choline Metabolism. Clin. Cancer Res. 2015 ,21, 386–395. [CrossRef] 44. Mariotto, E.; Bortolozzi, R.; Volpin, I.; Carta, D.; Serafin, V.; Accordi, B.; Basso, G.; Navarro, P.L.; López-Cara, L.C.; Viola, G. EB-3D a novel choline kinase inhibitor induces deregulation of the AMPK-mTOR pathway and apoptosis in leukemia T-cells. Biochem. Pharmacol. 2018,155, 213–223. [CrossRef] 45. Ramirez de Molina, A.; Báñez-Coronel, M.; Gutiérrez, R.; Rodríguez-González, A.; Olmeda, D.; Megías, D. Choline kinase activation is a critical requirement for the proliferation of primary human mammary epithelial cells and breast tumor progression. Cancer Res. 2004,64, 6732–6739. [CrossRef] [PubMed] 46. Shah, T.; Wildes, F.; Penet, M.-F.; Jr, P.T.W.; Glunde, K.; Artemov, D.; Ackerstaff, E.; Gimi, B.; Kakkad, S.; Raman, V.; et al. Choline kinase overexpression increases invasiveness and drug resistance of human breast cancer cells. NMR Biomed. 2010 ,23, 633–642. [CrossRef] [PubMed] 47. Gokhale, S.; Lu, W.; Zhu, S.; Liu, Y.; Hart, R.P.; Rabinowitz, J.D.; Xie, P. Elevated Choline Kinase α –Mediated Choline Metabolism Supports the Prolonged Survival of TRAF3-Deficient B Lymphocytes. J. Immunol. 2020,204, 459–471. [CrossRef] [PubMed] 48. Pera, B.; Krumsiek, J.; Assouline, S.E.; Marullo, R.; Patel, J.; Phillip, J.M.; Roman, L.; Mann, K.K.; Cerchietti, L. Metabolomic Profiling Reveals Cellular Reprogramming of B-Cell Lymphoma by a Lysine Deacetylase Inhibitor through the Choline Pathway. EBioMedicine 2018,28, 80–89. [CrossRef] [PubMed] 49. Gadiya, M.; Mori, N.; Cao, M.D.; Mironchik, Y.; Kakkad, S.; Gribbestad, I.S.; Glunde, K.; Krishnamachary, B.; Bhujwalla, Z.M. Phospholipase D1 and choline kinase-αare interactive targets in breast cancer. Cancer Biol. Ther. 2014,15, 593–601. [CrossRef] 50. Kim, H.S.; Tian, L.; Jung, M.; Choi, S.K.; Sun, Y.; Kim, H.; Moon, W.K. Downregulation of Choline Kinase-Alpha Enhances Autophagy in Tamoxifen-Resistant Breast Cancer Cells. PLoS ONE 2015,10, e0141110. [CrossRef] 51. Glunde, K.; Raman, V.; Mori, N.; Bhujwalla, Z.M. RNA Interference-Mediated Choline Kinase Suppression in Breast Cancer Cells Induces Differentiation and Reduces Proliferation. Cancer Res. 2005,65, 11034–11043. [CrossRef] [PubMed] 52. Bañez-Coronel, M.; De Molina, A.R.; Rodríguez-González, A.; Sarmentero, J.; Ramos, M.A.; García-Cabezas, M.A.; García-Oroz, L.; Lacal, J.C. Choline kinase alpha depletion selectively kills tumoral cells. Curr. Cancer Drug Targets 2008 ,8, 709–719. [CrossRef] 53. Krishnamachary, B.; Glunde, K.; Wildes, F.; Mori, N.; Takagi, T.; Raman, V.; Bhujwalla, Z.M. Noninvasive detection of lentiviralmediated choline kinase targeting in a human breast cancer xenograft. Cancer Res. 2009,69, 3464–3471. [CrossRef] [PubMed] 54. de la Cueva, A.; Ramírez de Molina, A.; Álvarez-Ayerza, N.; Ramos, M.A.; Cebrián, A.; Del Pulgar, T.G.; Lacal, J.C. Combined 5-FU and ChoK α inhibitors as a new alternative therapy of colorectal cancer: Evidence in human tumor-derived cell lines and mouse xenographs. PLoS ONE 2013,8, e64961. [CrossRef] [PubMed] 55. Mori, N.; Glunde, K.; Takagi, T.; Raman, V.; Bhujwalla, Z.M. Choline Kinase Down-regulation Increases the Effect of 5-Fluorouracil in Breast Cancer Cells. Cancer Res. 2007,67, 11284–11290. [CrossRef] [PubMed] 56. Mariotto, E.; Viola, G.; Ronca, R.; Persano, L.; Aveic, S.; Bhujwalla, Z.M.; Mori, N.; Accordi, B.; Serafin, V.; López-Cara, L.C.; et al. Choline Kinase Alpha Inhibition by EB-3D Triggers Cellular Senescence, Reduces Tumor Growth and Metastatic Dissemination in Breast Cancer. Cancers 2018,10, 391. [CrossRef] 57. Lacal, J.C.; Andera, L. Choline kinase inhibitors synergize with TRAIL in the treatment of colorectal tumors and overcomes TRAIL resistance. Cancer Transl. Med. 2016,2, 163. [CrossRef] 58. Rizzo, A.; Satta, A.; Garrone, G.; Cavalleri, A.; Napoli, A.; Raspagliesi, F.; Figini, M.; De Cecco, L.; Iorio, E.; Tomassetti, A.; et al. Choline kinase alpha impairment overcomes TRAIL resistance in ovarian cancer cells. J. Exp. Clin. Cancer Res. 2021 ,40, 1–13. [CrossRef] 59. Ramírez de Molina, A.; de la Cueva, A.; Machado-Pinilla, R.; Rodríguez-Fanjul, V.; Gómez del Pulgar, T.; Cebrián, A.; Perona, R.; Lacal, J.C. Acid ceramidase as a chemotherapeutic target to overcome resistance to the antitumoral effect of choline kinase α inhibition. Curr. Cancer Drug Targets 2012,12, 617–624. [CrossRef] 60. Rodríguez-González, A.; Ramírez de Molina, A.; Báñez-Coronel, M.; Megías, D.; Núñez, M.C.; Lacal, J.C. Inhibition of choline kinase renders a highly selective cytotoxic effect in tumor cells through a mitochondrial independent mechanism. Int. J. Oncol. 2005,26, 999–1008. 61. Rodriguez-Gonzalez, A.; De Molina, A.R.; Fernández, F.; Lacal, J.C. Choline kinase inhibition induces the increase in ceramides resulting in a highly specific and selective cytotoxic antitumoral strategy as a potential mechanism of action. Oncogene 2004 ,23, 8247–8259. [CrossRef] [PubMed] 62. Sánchez-López, E.; Menchén, L.E.; Seco, E.; Gómez del Pulgar, T.; Moyer, M.P.; Lacal, J.C.; Cebrian, A. Endoplasmic reticulum stress participates in the cytotoxic effect of choline kinase αinhibitors in tumor cells. Cell Death Dis. 2013,4, e933. 63. Arlauckas, S.P.; Kumar, M.; Popov, A.V.; Poptani, H.; Delikatny, E.J. Near infrared fluorescent imaging of choline kinase alpha expression and inhibition in breast tumors. Oncotarget 2017,8, 16518–16530. [CrossRef] [PubMed] 64. Campos, J.; Núñez, M.C.; Rodríguez, V.; Gallo, M.A.; Espinosa, A. QSAR of 1,1 0 -(1,2-ethylenebisbenzyl)bis(4-substitutedpyridinium) dibromides as choline kinase inhibitors: A different approach for antiproliferative drug design. Bioorg. Med. Chem. Lett. 2000 ,10, 767–770. [CrossRef]
Pharmaceutics 2021,13, 788 24 of 28 65. Cannon, J.G. Structure-activity aspects of hemicholinium-3 (HC-3) and its analogs and congeners. Med. Res. Rev. 1994 ,14, 505–531. [CrossRef] 66. Sánchez-Martín, R.; Campos, J.; Conejo-García, A.; Cruz-López, O.; Báñez-Coronel, M.; Rodríguez-González, A.; Gallo, M.A.; Lacal, J.C.; Espinosa, A. Symmetrical Bis-Quinolinium Compounds: New Human Choline Kinase Inhibitors with Antiproliferative Activity against the HT-29 Cell Line. J. Med. Chem. 2005,48, 3354–3363. [CrossRef] [PubMed] 67. Viswanadhan, V.N.; Ghose, A.K.; Revankar, G.R.; Robins, R.K. Atomic physicochemical parameters for three dimensional structure directed quantitative structure-activity relationships. 4. Additional parameters for hydrophobic and dispersive interactions and their application for an automated superposition of certain naturally occurring nucleoside antibiotics. J. Chem. Inf. Model. 1989,29, 163–172. [CrossRef] 68. Pallas Frame Module, a Prediction Tool of Physicochemical Parameters. Available online: www.compudrug.com (accessed on 15 April 2021). 69. Charton, M. Electrical Effect Substituent Constants for Correlation Analysis. Prog. Phys. Org. Chem. 2007 ,13, 119–251. [CrossRef] 70. Hansch, C.; Leo, A. Substituent Constant for Correlation Analysis in Chemistry and Biology; Wiley: New York, NY, USA, 1979; pp. 51–52. 71. Campos, J.; Núñez, M.C.; Sánchez, R.M.; Gómez-Vidal, J.A.; Rodríguez-González, A.; Báñez, M.; Gallo, M.A.; Lacal, J.C.; Espinosa, A. Quantitative structure–activity relationships for a series of symmetrical bisquaternary anticancer compounds. Bioorg. Med. Chem. 2002,10, 2215–2231. [CrossRef] 72. Campos, J.; Núñez, C.; Díaz, J.J.; Sanchez, R.M.; Gallo, M.A.; Espinosa, A. Anticancer bisquaternary heterocyclic compounds: A ras-ional design. Il Farm. 2003,58, 221–229. [CrossRef] 73. Gupta, S.P. QSAR studies on enzyme inhibitors. Chem. Rev. 1987,87, 1183–1253. [CrossRef] 74. Kearney, P.C.; Mizoue, L.S.; Kumpf, R.A.; Forman, J.; McCurdy, A.; Dougherty, D.A. Molecular recognition in aqueous media. New binding studies provide further insights into the cation-.pi. interaction and related phenomena. J. Am. Chem. Soc. 1993 ,115, 9907–9919. [CrossRef] 75. Kubinyi, H. QSAR: Hansch Analysis and Related Approaches. In Methods and Principles in Medicinal Chemistry; Mannhold, R., Krogsgaard-Larsen, P., Timmerman, H., Eds.; VCH: Weinheim, Germany, 1993; Volume 1, pp. 40–41. 76. Blaney, J.M.; Hansch, C. Quantitative Drug Design. In Comprehensive Medicinal Chemistry. The Rational Design, Mechanistic Study & Therapeutic Application of Chemical Compounds; Ramsden, C.A., Hansch, C., Sammes, P.G., Taylor, J.B., Eds.; Pergamon Press: Oxford, UK, 1990; Volume 4, pp. 459–496. 77. Hansch, C.; Blaney, J.M. Chemische Struktur und Biologische Aktivität von Wirkstoffen. Methoden der Quantitativen Struktur-WirkungAnalyse; Seydel, J.K., Schaper, K.-J., Eds.; Verlag Chemie: Weinheim, Germany, 1979; pp. 185–208. 78. Lacal, J.C.; Campos, J. Preclinical Characterization of RSM-932A, a Novel Anticancer Drug Targeting the Human Choline Kinase Alpha, an Enzyme Involved in Increased Lipid Metabolism of Cancer Cells. Mol. Cancer Ther. 2015,14, 31–39. [CrossRef] 79. Study of Intravenous TCD-717 in Patients with Advanced Solid Tumours. ClinicalTrials.gov Identifier: NCT01215864. Available online: http://clinicaltrials.gov/ct2/show/NCT01215864 (accessed on 15 April 2021). 80. Schiaffino-Ortega, S.; Baglioni, E.; Mariotto, E.; Bortolozzi, R.; Serrán-Aguilera, L.; Ríos-Marco, P.; Carrasco-Jimenez, M.P.; Gallo, M.A.; Hurtado-Guerrero, R.; Marco, C.; et al. Design, synthesis, crystallization and biological evaluation of new symmetrical biscationic compounds as selective inhibitors of human Choline Kinase α 1 (ChoK α 1). Sci. Rep. 2016 ,6, 23793. [CrossRef] [PubMed] 81. Zech, S.G.; Kohlmann, A.; Zhou, T.; Li, F.; Squillace, R.M.; Parillon, L.E.; Greenfield, M.T.; Miller, D.P.; Qi, J.; Thomas, R.M.; et al. Novel Small Molecule Inhibitors of Choline Kinase Identified by Fragment-Based Drug Discovery. J. Med. Chem. 2016 ,59, 671–686. [CrossRef] 82. Rubio-Ruiz, B.; Serrán-Aguilera, L.; Hurtado-Guerrero, R.; Conejo-García, A. Recent advances in the design of choline kinase α inhibitors and the molecular basis of their inhibition. Med. Res. Rev. 2021,41, 902–927. [CrossRef] [PubMed] 83. Rubbini, G.; Buades-Martín, A.B.; Kimatrai-Salvador, M.; Entrena, A.; Gallo-Mezo, M.Á.; Ríos-Marco, P.; Marco, C.; Mattiuzzo, E.; Bortolozzi, R.; Mariotto, E.; et al. Lead optimization-hit expansion of new asymmetrical pyridinium/quinolinium compounds as choline kinase α1 inhibitors. Future Med. Chem. 2018,10, 1769–1786. [CrossRef] 84. Falcon, S.C.; Hudson, C.S.; Huang, Y.; Mortimore, M.; Golec, J.M.; Charlton, P.A.; Weber, P.; Sundaram, H. A non-catalytic role of choline kinase alpha is important in promoting cancer cell survival. Oncogenesis 2013,2, e38. [CrossRef] [PubMed] 85. Mori, N.; Wildes, F.; Kakkad, S.; Jacob, D.; Solaiyappan, M.; Glunde, K.; Bhujwalla, Z.M. Choline kinaseα protein and phosphatidylcholine but not phosphocholine are required for breast cancer cell survival. NMR Biomed. 2015 ,28, 1697–1706. [CrossRef] [PubMed] 86. Asim, M.; Massie, C.E.; Orafidiya, F.; Pértega-Gomes, N.; Warren, A.Y.; Esmaeili, M.; Selth, L.A.; Zecchini, H.I.; Luko, K.; Qureshi, A.; et al. Choline Kinase Alpha as an Androgen Receptor Chaperone and Prostate Cancer Therapeutic Target. J. Natl. Cancer Inst. 2016,108, djv371. [CrossRef] 87. Asim, M.; Massie, C.E.; Neal, D.E. Kinase joins the chaperone club: Androgen-regulated kinome reveals choline kinase alpha as a potential drug target in prostate cancer. Mol. Cell Oncol. 2016,3, e1140262. [CrossRef] 88. Zimmerman, T.; Moneriz, C.; Diez, A.; Bautista, J.M.; Del Pulgar, T.G.; Cebrián, A.; Lacal, J.C. Antiplasmodial Activity and Mechanism of Action of RSM-932A, a Promising Synergistic Inhibitor of Plasmodium falciparum Choline Kinase. Antimicrob. Agents Chemother. 2013,57, 5878–5888. [CrossRef] [PubMed]
Pharmaceutics 2021,13, 788 25 of 28 89. Hudson, C.S.; Knegtel, R.M.; Brown, K.; Charlton, P.A.; Pollard, J.R. Kinetic and mechanistic characterisation of Choline Kinaseα . Biochim. Biophys. Acta Proteins Proteom. 2013,1834, 1107–1116. [CrossRef] [PubMed] 90. Kall, S.L.; Delikatny, E.J.; Lavie, A. Identification of a Unique Inhibitor-Binding Site on Choline Kinase α .Biochemistry 2018 ,57, 1316–1325. [CrossRef] 91. Koch, K.; Hartmann, R.; Schröter, F.; Suwala, A.K.; Maciaczyk, D.; Krüger, A.C.; Willbold, D.; Kahlert, U.D.; Maciaczyk, J. Reciprocal regulation of the cholinic phenotype and epithelial-mesenchymal transition in glioblastoma cells. Oncotarget 2016,7, 73414–73431. [CrossRef] 92. Granata, A.; Nicoletti, R.; Perego, P.; Iorio, E.; Krishnamachary, B.; Benigni, F.; Ricci, A.; Podo, F.; Bhujwalla, Z.M.; Canevari, S.; et al. Global metabolic profile identifies choline kinase alpha as a key regulator of glutathione-dependent antioxidant cell defense in ovarian carcinoma. Oncotarget 2015,6, 11216–11230. [CrossRef] [PubMed] 93. Lin, Y.C.; Kanchara, K.; Nakamura, Y. Arabidopsis CHOLINE/ETHANOLAMINE KINASE 1 (CEK1) is a primary choline ekinase localized at the endoplasmic reticulum (ER) and involved in ER stress tolerance. New Phytol. 2019,223, 1904–1917. [CrossRef] 94. Trousil, S.; Kaliszczak, M.; Schug, Z.; Nguyen, Q.-D.; Tomasi, G.; Favicchio, R.; Brickute, D.; Fortt, R.; Twyman, F.J.; Carroll, L.; et al. The novel choline kinase inhibitor ICL-CCIC-0019 reprograms cellular metabolism and inhibits cancer cell growth. Oncotarget 2016,7, 37103–37120. [CrossRef] [PubMed] 95. Ramírez de Molina, A.; Gallego-Ortega, D.; Sarmentero-Estrada, J.; Lagares, D.; Gómez del Pulgar, T.; Bandrés, E.; García-Foncillas, J.; Lacal, J.C. Choline kinase as a link connecting phospholipid metabolism and cell cycle regulation: Implications in cancer therapy. Int. J. Biochem. Cell Biol. 2008,40, 1753–1763. 96. Moestue, S.A.; Borgan, E.; Huuse, E.M.; Lindholm, E.M.; Sitter, B.; Børresen-Dale, A.L.; Engebraaten, O.; Maelandsmo, G.M.; Gribbestad, I.S. Disctinct choline metabolic profiles are associated with differences in gene expression for basal-like and luminallike breast cancer xenograft models. BMC Cancer 2010,10, 433. [CrossRef] 97. Grinde, M.T.; Skrbo, N.; Moestue, S.A.; A Rødland, E.; Borgan, E.; Kristian, A.; Sitter, B.; Bathen, T.F.; Børresen-Dale, A.-L.; Mælandsmo, G.M.; et al. Interplay of choline metabolites and genes in patient-derived breast cancer xenografts. Breast Cancer Res. 2014,16, R5. [CrossRef] 98. Miyake, T.; Parsons, S.J. Functional interactions between choline kinaseα , epidermal growth factor receptor and c-Src in breast cancer cell proliferation. Oncogene 2012,31, 1431–1441. [CrossRef] 99. Kall, S.L.; Whitlatch, K.; Smithgall, T.E.; Lavie, A. Molecular basis for the interaction between human choline kinase alpha and the SH3 domain of the c-Src tyrosine kinase. Sci. Rep. 2019,9, 1–13. [CrossRef] [PubMed] 100. Liu, P.J.; Chen, Y.H.; Tsai, K.W.; Yeah, H.Y.; Yeh, C.Y.; Tu, Y.T.; Yang, C.Y. Involvement of MicroRNA-1-FAM83A Axis Dysfunction in the Growth and Motility of Lung Cancer Cells. Int. J. Mol. Sci. 2020,21, 8833. [CrossRef] [PubMed] 101. Chang, C.C.; Few, L.L.; Konrad, M.; Too, W.C.S. Phosphorylation of Human Choline Kinase Beta by Protein Kinase A: Its Impact on Activity and Inhibition. PLoS ONE 2016,11, e0154702. [CrossRef] [PubMed] 102. Lai, S.; Safaei, J.; Pelech, S. Evolutionary Ancestry of Eukaryotic Protein Kinases and Choline Kinases. J. Biol. Chem. 2016 ,291, 5199–5205. [CrossRef] [PubMed] 103. Chua, B.T.; Gallego-Ortega, D.; De Molina, A.R.; Ullrich, A.; Lacal, J.C.; Downward, J. Regulation of Akt(ser473) phosphorylation by Choline kinase in breast carcinoma cells. Mol. Cancer 2009,8, 131. [CrossRef] 104. Yalcin, A.; Clem, B.; Makoni, S.; Clem, A.; Nelson, K.; Thornburg, J.; Siow, D.; Lane, A.N.; E Brock, S.; Goswami, U.; et al. Selective inhibition of choline kinase simultaneously attenuates MAPK and PI3K/AKT signaling. Oncogene 2009 ,29, 139–149. [CrossRef] 105. Clem, B.F.; Clem, A.L.; Yalcin, A.; Goswami, U.; Arumugam, S.; Telang, S.; O Trent, J.; Chesney, J. A novel small molecule antagonist of choline kinaseα that simultaneously suppresses MAPK and PI3K/AKT signaling. Oncogene 2011 ,30, 3370–3380. [CrossRef] 106. Ramírez de Molina, A.; Penalva, V.; Lucas, L.; Lacal, J.C. Regulation of Choline kinase activity by Ras proteins involves Ral-GDS and PI3K. Oncogene 2002,21, 937–946. [CrossRef] 107. Glunde, K.; Shah, T.; Winnard, P.T., Jr.; Raman, V.; Takagi, T.; Vesuna, F.; Artemov, D.; Bhujwalla, Z.M. Hypoxia regulates choline kinase expression through hypoxia-inducible factor-1 alpha signaling in a human prostate cancer model. Cancer Res. 2008 ,68, 172–180. 108. Bansal, A.; Harris, R.A.; DeGrado, T.R. Choline phosphorylation and regulation of transcription of choline kinase α in hypoxia. J. Lipid Res. 2012,53, 149–157. [CrossRef] [PubMed] 109. Viswanath, P.; Radoul, M.; Izquierdo-Garcia, J.L.; Luchman, H.A.; Cairncross, J.G.; Pieper, R.O.; Phillips, J.J.; Ronen, S.M. Mutant IDH1 gliomas downregulate phosphocholine and phosphoethanolamine synthesis in a 2-hydroxyglutarate-dependent manner. Cancer Metab. 2018,6, 3. [CrossRef] 110. Marcucci, H.; Paoletti, L.; Jackowski, S.; Banchio, C. Phosphatidylcholine Biosynthesis during Neuronal Differentiation and Its Role in Cell Fate Determination. J. Biol. Chem. 2010,285, 25382–25393. [CrossRef] [PubMed] 111. Domizi, P.; Aoyama, C.; Banchio, C. Choline kinase alpha expression during RA-induced neuronal differentiation: Role of C/EBPβ.Biochim. Biophys. Acta 2014,1841, 544–551. [CrossRef] [PubMed] 112. Domizi, P.; Malizia, F.; Chazarreta-Cifre, L.; Diacovich, L.; Banchio, C. KDM2B regulates choline kinase expression and neuronal differentiation of neuroblastoma cells. PLoS ONE 2019,14, e0210207. [CrossRef] [PubMed]