Fluorimetric and CD Recognition Between Various Ds-DNA/RNA Depends on a Cyanine Connectivity in Cyanine-Guanidiniocarbonyl-Pyrrole Conjugate

molecules

Article Fluorimetric and CD Recognition between Various ds-DNA/RNA Depends on a Cyanine Connectivity in Cyanine-guanidiniocarbonyl-pyrrole Conjugate

1, 1 2 2 Tamara Šmidlehner †, Marta Košćak , Ksenija Božinović , Dragomira Majhen , 3, 1, Carsten Schmuck ‡ and Ivo Piantanida * 1 Division of Organic Chemistry and Biochemistry, Ruder¯ BoškovićInstitute, BijeniˇckaCesta 54, 10000 Zagreb, Croatia; [email protected] (T.Š.); [email protected] (M.K.) 2 Division of Molecular Biology, Ruder¯ BoškovićInstitute, Bijeniˇckacesta 54, 10000 Zagreb, Croatia; [email protected] (K.B.); [email protected] (D.M.) 3 Institute of Organic Chemistry, University of Duisburg-Essen, 45141 Essen, Germany; [email protected] * Correspondence: [email protected]; Tel.: +385-1-4571-326 Present address: National Institute of Chemistry, Hajdrihova 19, P.O. Box 660, SI-1001 Ljubljana, Slovenia. † This work is dedicated to the late Prof. Dr. Carsten Schmuck, in memory of his invaluable contribution to ‡ our research. Academic Editor: Ugo Caruso Received: 7 September 2020; Accepted: 27 September 2020; Published: 29 September 2020

Abstract: Two novel isosteric conjugates of guanidiniocarbonyl-pyrrole and 6-bromo-TO (thiazole orange) were prepared, differing only in linker connectivity to cyanine (benzothiazole nitrogen vs. quinoline nitrogen). The quinoline analog was significantly more susceptible to aggregation in an aqueous medium, which resulted in induced circular dichroism (ICD; λ = 450–550 nm) recognition between A-T(U) and G-C basepair containing polynucleotides. The benzothiazole-isostere showed pronounced (four-fold) fluorimetric selectivity toward ds-RNA in comparison to any ds-DNA, at variance to its quinoline-analogue fluorescence being weakly selective to GC-DNA. Preliminary screening on human tumor and normal lung cell lines showed that both dyes very efficiently enter living cells and accumulate in mitochondria, causing moderate cytotoxic effects, and thus could be considered as lead compounds toward novel theragnostic mitochondrial dyes.

Keywords: cyanine dyes; guanidiniocarbonyl-pyrrole; ds-DNA/RNA sensing; ﬂuorescence; circular dichroism; mitochondria

1. Introduction Versatile applications of small molecule fluorescent probes targeting DNA/RNA in biochemical and biomedicinal applications have attracted enormous interest and, due to the versatility of applications, have become unavoidable tools for monitoring biological processes [1]. The research on DNA/RNA dyes has been mostly focused on the impact on the DNA/RNA function or selective/specific DNA/RNA labeling [2–4]. However, the huge complexity of DNA-coded processes, which do not depend only on coding DNA basepair sequences, but also include epigenetics, has only recently attracted attention [5,6]. Cyanine analogs are likely the most extensively used family of fluorogenic dyes, their particular advantage being non-emissive in free solution but strong emission fluorescence upon binding to the target [7–9]. On the other hand, our systematic work on aryl-guanidiniocarbonyl-pyrrole (GCP) derivatives characterized GCP moiety as a very useful building block in the design of new small molecules targeting DNA/RNA, particularly when conjugated with large aromatic fluorophores [10–12]. For instance, pyrene-GCP analogs have been shown to be very efficient single-molecule- multipurpose

Molecules 2020, 25, 4470; doi:10.3390/molecules25194470 www.mdpi.com/journal/molecules Molecules 2020, 25, x FOR PEER REVIEW 2 of 15 molecules targeting DNA/RNA, particularly when conjugated with large aromatic fluorophores [10– 12]. For instance, pyrene-GCP analogs have been shown to be very efficient single-molecule- multipurpose probes, being able to recognize between various ds-DNA and ds-RNA by simultaneous specific responses in fluorescence and circular dichroism [13,14]. Moreover, recently reported pyrene- GCP analog fluorimetric recognition of DNA/RNA could be reversibly controlled by pH [15]. We recently showed that isosteric cyanine conjugates with pyrene show some different properties in interaction with DNA/RNA and also proteins (BSA) as a result of different connectivity of the linker to the cyanine chromophore [16]. Very recently, we also prepared the first cyanine-GCP conjugate, which revealed promising selective fluorimetric and circular dichroism sensing between various ds-DNA or ds-RNA polynucleotides, which was additionally controllable by pH due to GCP Molecules 2020, 25, 4470 2 of 14 [17]. Led by the aforementioned promising results, we designed and prepared novel analogs of cyanine-GCP conjugates, conjugates 1 and 2, characterization of which we presented in this work. probes,In the being design able of to new recognize conjugates between (Scheme various 1), we ds-DNA introduced and an ds-RNA elongated by simultaneouslinker with one specific more responsesCH2− group in, fluorescencewith respect and to the circular afore- dichroismstudied analog [13,14 [1].7 Moreover,] to increase recently flexibility reported, and pyrene-GCPwe changed analogcyanine fluorimetric substituent recognition“Cl” with “Br” of DNA. A larger/RNA volume could beof reversibly“Br” is likely controlled to block byintercalation pH [15]. of cyanine moietyWe between recently DNA/RNA showed that base isosteric pairs, cyanine thus directing conjugates small with molecule pyrene show to the some DNA di orfferent RNA properties grooves, inwhich interaction could result with DNAin the/RNA high andsensitivity also proteins to binding (BSA) site as aproperties result of di (namely,fferent connectivity grooves of ofvarious the linker ds- toDNA the or cyanine ds-RNA chromophore differ considerably [16]. Very — recently, see Table we S1, also Supplementary prepared the firstMaterials cyanine-GCP). Conjugate conjugate, 1 was whichpreviously revealed tested promising for inhibitor selective properties fluorimetric of dipeptidyl and circularpeptidase dichroism III (DPP III) sensing [18]. between various ds-DNAThe orconjugates ds-RNA 1 polynucleotides, and 2 are actually which isosteres, was additionally differing in controllable a reversed byconnectivity pH due to to GCP cyanine [17]. Ledchromophore by the aforementioned (benzothiazole promising vs. quinoline), results, wewhich designed resulted and in prepared more (1) novel or less analogs (2) shielded of cyanine-GCP position conjugates,of permanent conjugates positive 1chargeand 2 ,in characterization the molecule (Scheme of which 1, we bottom). presented Moreover, in this work. both molecules also have Inprotonation the design-site of new (GCP) conjugates, which tends (Scheme to 1insert), we into introduced DNA/RNA an elongated grooves only linker at with weakly one moreacidic CHconditions,2− group, thus with allowing respect further to the afore-studied control of DNA/RNA analog [17 interaction] to increase by flexibility, outer stimuli and we (pH changed change cyaninebetween substituent5 and 7). Since “Cl” DNA/RNA with “Br”. are Alarger polyanions volume, and of “Br”electrostatic is likely interactions to block intercalation with small of molecule cyanine moietycations between usually contribute DNA/RNA significantly base pairs, thus to complex directing formation, small molecule it is to to be the expected DNA or RNAthat although grooves, whichisosteres, could 1 and result 2 inwould the high show sensitivity significantly to binding different site propertiespatterns in (namely, DNA/RNA grooves recognition, of various allowing ds-DNA orexploration ds-RNA di offfer the considerably—see importance of Table steric, S1, electrostatic Supplementary and Materials). H-bonding Conjugate impacts1 onwas DNA/RNA previously testedrecognition. for inhibitor properties of dipeptidyl peptidase III (DPP III) [18].

1 2 Scheme 1. TOP: General structure of studied compounds 1 and 2; DOWN: ball-and-stick models with Scheme 1. TOP: General structure of studied compounds 1 and 2; DOWN: ball-and-stick models with superimposed van der Waals (VdW) radius surface denoting atom charge distribution (red-negative; superimposed van der Waals (VdW) radius surface denoting atom charge distribution (red-negative; blue positive), stressing the difference in the exposure of positive charge (red circles). blue positive), stressing the difference in the exposure of positive charge (red circles). The conjugates 1 and 2 are actually isosteres, differing in a reversed connectivity to cyanine chromophoreBiorelevant (benzothiazole applications vs. of quinoline),novel dyes whichare often resulted related in moreto their (1) interactivity or less (2) shielded with living position cells. of permanentThe series of positive analogs charge of parent in the cyanine molecule-amino (Scheme acid 1(Cy, bottom).-aa, Scheme Moreover, 2) for the both here molecules prepared also 1 and have 2 protonation-site (GCP), which tends to insert into DNA/RNA grooves only at weakly acidic conditions, thus allowing further control of DNA/RNA interaction by outer stimuli (pH change between 5 and 7). Since DNA/RNA are polyanions, and electrostatic interactions with small molecule cations usually contribute significantly to complex formation, it is to be expected that although isosteres, 1 and 2 would show significantly different patterns in DNA/RNA recognition, allowing exploration of the importance of steric, electrostatic and H-bonding impacts on DNA/RNA recognition. Biorelevant applications of novel dyes are often related to their interactivity with living cells. The series of analogs of parent cyanine-amino acid (Cy-aa, Scheme2) for the here prepared 1 and 2 already showed very efficient uptake in human cells and selective accumulation in mitochondria, accompanied by moderate cytotoxicity against human tumor cell lines (IC50 10–90 µM) [19]. The here Molecules 2020, 25, x FOR PEER REVIEW 3 of 15

already showed very efficient uptake in human cells and selective accumulation in mitochondria, accompanied by moderate cytotoxicity against human tumor cell lines (IC50 10–90 M) [19]. The here prepared 1 and 2 are actually upgraded by the addition of a GCP unit, which itself is non-toxic and also known for efficiently entering cells, as well as acting as a transfection agent [20,21]. Thus, for novel Cy-amino acid-GCP conjugates 1 and 2, questions remained whether they will retain efficient cellular uptake and, more important, whether mitochondrial selectivity of cyanine-amino acid part will also bring GCP unit to mitochondria. If so, GCP as a very rich H-bonding moiety and being pH- sensitive could influence the bioactivity of new conjugates.

2. Results and Discussion

2.1. Chemistry Molecules 2020, 25, 4470 3 of 14 The 2 was prepared as shown on Scheme 2, starting from the corresponding cyanine-amino acids (Cy-aa) [19] and guanidiniocarbonyl-pyrrole (GCP) [10–12], following a similar procedure used preparedpreviously1 andfor 2theare preparation actually upgraded of its analog by the 1 addition [18]. Namely, of a GCP the unit, reaction which mixture itself is non-toxicof Cy-amino and acid also knownand GCP for effi inciently acetonitrile entering with cells, asO- well(1H- as6-Chlorobenzotriazole acting as a transfection-1-yl) agent-1,1,3,3 [20-,tetramethyluronium21]. Thus, for novel Cy-aminohexafluorophosphate acid-GCP conjugates (HCTU) and1 and Et32,Nquestions was left stirring remained at room whether temperature they will retainunder eargonfficient for cellular three uptakedays. Boc and,-protect moreed important, product was whether deprotected mitochondrial by the selectivityaddition of of TFA, cyanine-amino the solvent acid was part evaporated will also, bringand the GCP crude unit product to mitochondria. was recrystallized If so, GCP from as a the very methanol rich H-bonding-ether mixture moiety and and washed being pH-sensitive with water couldto give influence a pure product the bioactivity in 60% yield of new as conjugates.a red solid.

Scheme 2.2. Synthetic route for the preparation of 2, analogous to previous preparation ofof 11 [[1818].].

2.2.2. Results Characteri andzation Discussion of 1 and 2 in Aqueous Solutions 2.1. ChemistryBoth compounds were well-soluble in water, allowing preparation of stock solutions at mM concentrations, stable over longer periods stored in a refrigerator. The UV/Vis spectra of 1 and 2 were The 2 was prepared as shown on Scheme2, starting from the corresponding cyanine-amino collected at pH 7.0 and pH 5.0 to monitor both the neutral and protonated form of GCP. At both pH acids (Cy-aa)[19] and guanidiniocarbonyl-pyrrole (GCP)[10–12], following a similar procedure (Figure 1) 1 and 2 revealed two pronounced absorption regions, one (λmax  300 nm) originating from used previously for the preparation of its analog 1 [18]. Namely, the reaction mixture of Cy-amino overlapped absorbancies of guanidiniocarbonyl-pyrrole (GCP) and triazole moieties and another acid and GCP in acetonitrile with O-(1H-6-Chlorobenzotriazole-1-yl)-1,1,3,3-tetramethyluronium absorption band in the visible range attributed to cyanine dye (λmax = 480–510 nm). For both hexaﬂuorophosphate (HCTU) and Et3N was left stirring at room temperature under argon for three compounds at micromolar concentrations, absorbancies were proportional to the concentration days. Boc-protected product was deprotected by the addition of TFA, the solvent was evaporated, (Supplementary Materials.), allowing determination of molar extinction coefficients (Supplementary and the crude product was recrystallized from the methanol-ether mixture and washed with water to Materials in Table S2). give a pure product in 60% yield as a red solid.

2.2. Characterization of 1 and 2 in Aqueous Solutions Both compounds were well-soluble in water, allowing preparation of stock solutions at mM concentrations, stable over longer periods stored in a refrigerator. The UV/Vis spectra of 1 and 2 were collected at pH 7.0 and pH 5.0 to monitor both the neutral and protonated form of GCP. At both pH (Figure1) 1 and 2 revealed two pronounced absorption regions, one (λmax 300 nm) ≈ originating from overlapped absorbancies of guanidiniocarbonyl-pyrrole (GCP) and triazole moieties and another absorption band in the visible range attributed to cyanine dye (λmax = 480–510 nm). For both compounds at micromolar concentrations, absorbancies were proportional to the concentration (Supplementary Materials), allowing determination of molar extinction coeﬃcients (Supplementary Materials in Table S2). Closer analysis of UV/Vis data revealed distinct diﬀerences between 1 and 2, which could be correlated to the connectivity of cyanine to the tether. Namely, the benzothiazole-tethered 1 UV/Vis spectrum has a dominant maximum at λ = 505 nm, the constant ratio r (505 nm/485 nm) was retained over the entire span of micromolar concentrations (Supplementary Materials), and also this ratio did not change with pH (Figure1a), nor it changed with the temperature increase (Supplementary Materials in Figure S5, LEFT); all these properties typical for non-aggregated cyanines [9,22]. Contrary to 1, the UV/Vis spectrum of quinoline-tethered 2 changed within the micromolar concentrations (Figure1b), whereby a decrease of the ratio r (505 nm/485 nm) could be attributed to the increasing H-type cyanine aggregation, also supported by equivalent changes upon temperature increase (Supplementary Materials in Figure S5, RIGHT) [9,22]. Molecules 2020 25 Molecules 20, 20,, 447025, x FOR PEER REVIEW 4 of4 14 of 15

30000 0.14 r (505nm / 485 nm) = 0.98 25000 20000 2, pH 7 0.12 -6 2, pH 5 10 M 15000 -6

2 2*10 M 10000 0.10 3*10-6 M cm 5000

-1 -6 0 0.08 4*10 M 30000 5*10-6 M

25000 Abs 0.06 -6

/ mmol  6*10 M 20000 1, pH 7 7*10-6 M 15000 1, pH 5 0.04 10000 r (505nm / 485 nm) = 1.08 5000 0.02 0 450 475 500 525 550 575 600 0.00 450 475 500 525 550 575 600 625 650  / nm  / nm

(a) (b)

FigureFigure 1. ( a1.) The(a) The UV-Vis UV-V spectrais spectra of 1 ofand 1 and2 (c =2 (2c = 102 × 610M)−6 M) at pHat pH 7.0 7.0 and and pH pH 5.0 5.0 (sodium (sodium cacodylate cacodylate × − buffbuffer,er, I = I0.05 = 0.05 M). M). Note Note the the didifferencefference inin ratio rr (505(505 nm/485 nm/485 nm) nm) at atdifferent different pH pH only only for for2 but2 butnot for not1 for; (b1);( Theb) The absorbance absorbance dependence dependence on onc(2);c( 2note); note the the concentration concentration-dependent-dependent change change of ofratio ratio r(505 r(505nm/485 nm/485 nm). nm).

Moreover,Closer analysis the for 2, ofthe UV/ pH-dependentVis data revealed difference distinct in differences ratio r (505 nmbetween/485 nm) 1 and (Figure 2, which1a) clearly could be suggestedcorrelated more to etheffi cientconnectivity aggregation of cyanine at neutral to the conditions tether. Namely, in comparison the benzothiazole to pH 5, likely-tethered because 1 UV/Vis the repulsivespectrum interactions has a dominant between maximum the protonated at λ = guanidiniocarbonyl-pyrrole505 nm, the constant ratio r (505 (GCP) nm/485 moieties nm) hamper was retained the effiovercient the aromatic entire stackingspan of micromolar (necessary forconcentrations aggregation). (Supplementary Materials), and also this ratio did notPreviously change with studied pHCy-GCP (Figure revealed 1a), nor aggregation it changed properties with the temperaturein UV/Vis spectrum increase [17 (]Supplementary comparable to itsMaterials close analog in Figure2, pointing S5, LEFT); out that all quinoline-tethered these properties cyanines typical for conjugated non-aggregated with GCP cyanines moiety are[9,22]. moreContrary prone to to aggregation 1, the UV/Vis in comparison spectrum to of benzothiazole-analogs quinoline-tethered 2 ( 1).changed within the micromolar concentrationsBoth 1 and 2 have (Figure negligible 1b), whereby fluorescence a decrease emission of the due ratio to well-knownr (505 nm/485 non-emissive nm) could be relaxation attributed of to cyaninesthe increasing in a free stateH-type [9] cyanine and, although aggregation, both compounds also supported are chiral, by equivalent their CD changes spectra are upon very temperature weak at studiedincrease conditions (Supplementary (data not Materials shown) due in toFigure the distance S5, RIGHT) between [9,22] chiral. center and chromophore. Moreover, the for 2, the pH-dependent difference in ratio r (505 nm/485 nm) (Figure 1a) clearly 2.3. Interactions with ds-DNA, ds-RNA, and ss-RNA suggested more efficient aggregation at neutral conditions in comparison to pH 5, likely because the repulsiveThe choice interactions of ds-DNA between and ds-RNA the protonated in this study guanidiniocarbonyl was driven by-pyrrole significant (GCP) diff erencesmoieties in hamper the secondarythe efficient structure aromatic of particular stacking (necessary homogeneous for agg polynucleotideregation). sequences (for detailed structural features,Previously see Supplementary studied MaterialsCy-GCP in revealed Table S1): aggregation poly dAdT-poly properties dAdT are in characterized UV/Vis spectrum by typical [17] B-helicalcomparable structure to its and close minor analog groove 2, pointing ideal for out small that molecule quinoline binding;-tethered poly cyanines dGdC-poly conjugated dGdC with minor GCP groovemoiety is sterically are more blocked prone to by aggregation guanine amino in comparison groups significantly to benzothi hamperingazole-analogs small molecule(1). insertion; poly A—polyBoth 1 U and presenting 2 have negligible typical RNA fluorescence A-helical emission structure due with to a well major-known groove non as- aemissive potential relaxation target forof small cyanines molecules. in a free Also, state for comparison[9] and, although reasons, both we compounds used naturally are isolatedchiral, their DNA CD from spectracalf thymus are very (ct-DNA),weak at characterized studied conditions by a typical (data B-helical not shown) structure due and to an equalthe distance amount betweenof AT- and chiral GC-base center pairs. and Allchromophore chosen polynucleotides. are 100–200 base pairs long, ensuring a large quantity of identical, mutually independent binding sites for a small molecule and thus making the contribution of molecule binding at polynucleotide2.3. Interactions terminiwith ds- negligible.DNA, ds-RNA, and ss-RNA

2.3.1. ThermalThe choice Denaturation of ds-DNA Experiments and ds-RNA in this study was driven by significant differences in the secondary structure of particular homogeneous polynucleotide sequences (for detailed structural featuresThe thermal, see Supplementary denaturation experiments Materials in provide Table informationS1): poly dAdT about-poly the dAdT ds- polynucleotide are characterized helix by thermaltypical stability B-helical as astructure function and of interaction minor groove with ideal added for smallsmall moleculesmolecule binding; [23]. The poly diﬀ erencedGdC- betweenpoly dGdC theminorTm value groove of i frees sterically ds-polynucleotide blocked by guanine and a complex amino groups with asignificantly small molecule hampering (∆Tm value)small molecule is an importantinsertion; factor poly in A the — characterization poly U presenting of small typical molecule RNA/ ds-polynucleotide A-helical structure interactions. with a major For groove instance, as a moderatepotential to target strong for stabilization small molecules. (∆Tm >Also,5 ◦C) for supports comparison pronounced reasons, intercalativewe used naturally or minor isolated groove DNA bindingfrom interactioncalf thymus (ct [24-],DNA), whereas characterized weak or negligible by a typical stabilization B-helical (structure∆Tm = 0–5 and◦C) an suggests equal amount a binding of AT- processand GC driven-base mostly pairs. byAll hydrophobicchosen polynucleotides eﬀect accompanied are 100–200 with base weak pairs H-bonding long, ensuring and/or a electrostatic large quantity interactions—usuallyof identical, mutually excluding independe classicalnt intercalation binding sites as for a binding a small mode. molecule and thus making the contribution of molecule binding at polynucleotide termini negligible.

Molecules 2020, 25, 4470 5 of 14

The impact of 1 and 2 conjugates on stabilization of ds-polynucleotides at neutral conditions (pH 7) was rather weak and similar for all studied DNA/RNA (Table1), excluding classical intercalation as a dominant binding mode. However, for both compounds the stabilization eﬀect increased considerably at pH 5, which can be attributed to the protonation of GCP and its additional electrostatic interactions with the negatively charged DNA/RNA backbone. Intriguingly, acidic pH had a less pronounced eﬀect on AT-DNA stabilization, likely due to the deep insertion of GCP into a convenient minor groove [25,26].

a b Table 1. ∆Tm-Values (◦C) for diﬀerent ratios r of 1 and 2 added to polynucleotide.

b r ct-DNA p(dAdT)2 pApU 0.1 2.0 0.6 1.0 1, pH 7.0 0.2 2.0 0.3 1.0 0.1 2.0 1.3 c 7.0; 2.0 1, pH 5.0 0.2 2.0 2.6 c 6.0; 1.0 0.1 1.1 1.0 2.0 2, pH 7.0 0.2 1.7 1.1 3.0 0.1 2.0 1.0 c 5.0; 2.0 2, pH 5.0 0.2 6.0 1.1 c 8.0; 1.0 a b c ∆Tm-Value error 0.5 C; r = (compound)/(polynucleotide); Biphasic transitions: the ﬁrst transition at Tm ± ◦ 28.5 ◦C is attributed to denaturation of polyA–polyU and the second transition at Tm 80.1 ◦C is attributed to denaturation of polyAH+–polyAH+ since poly A at pH 5 is mostly protonated and forms ds-polynucleotide [25–27].

2.3.2. Fluorimetric Titrations The tendency of one studied dye to aggregate at c > 2 µM hampered accurate titrations in UV/Vis medium. However, the well-known property of cyanine dyes to switch-on strong fluorescence upon DNA/RNA binding allowed fluorimetric titration at non-aggregative conditions. Indeed, buffered solutions of 1 and 2, intrinsically non-emissive, exhibited strong fluorescence upon the addition of any ds-DNA or ds-RNA. However, the intensity of emission was strongly dependent on a compound structure as well as on the secondary structure of particular ds-polynucleotide (Figure2). Conjugate 1 showed strong, four-fold emission preference toward ds-RNA, at variance to its isostere 2 being generally less selective, but emitting the most brightly for GC-DNA. Both dyes reveal the weakest emission for AT-DNA. Such different responses suggest that both dyes bind in a similar manner to AT-DNA, whereas their binding to GC-DNA or AU-RNA significantly differs as a consequence of a fine interplay between dye structure and peculiar properties of DNA vs. RNA binding sites. Namely, only AT-DNA has a minor groove convenient for small molecule binding, in which both 1 and 2 bind similarly, giving a similar response. The AU-RNA major groove (common binding site for small molecules [4]) is of similar width as-AT-DNA, but much deeper (Table S1)—obviously allowing the difference in binding of cyanine-chromophore of 1 in respect to 2; while GC-DNA minor groove is sterically hindered by protruding amino groups of guanine which significantly limit the accommodation of small molecules, again being sensitive in structure difference between 1 and 2. We performed fluorimetric titrations also at pH 5, at which the GCP unit is protonated and, with an additional positive charge, could interact with negatively charged DNA/RNA backbone, increasing overall affinity (Supplementary Materials in Figure S24). Intriguingly, the fluorimetric response of 1 was identical to experiments done at pH 7, while 2 showed a slightly different fluorescence recognition pattern at pH 5, showing a small fluorimetric preference for mixed ct-DNA, followed by GC-DNA and AU-RNA. The reason why both new dyes (1, 2) did not show a pronounced impact of pH on their fluorimetric response, as previously noted for e.g., series of pyrene—GCP analogs [13–15], is probably because cyanine fluorophore is cationic itself, and thus not so dependent in binding to DNA/RNA on the additional positive charge of GCP present only at pH 5. MoleculesMolecules2020 2020, ,25 25,, 4470 x FOR PEER REVIEW 66 of of 14 15

ctDNA,  = 532 nm em ctDNA,  = 528 nm em p(dAdT) ,  = 532 nm 450 2 em p(dAdT) ,  = 526 nm 2 em p(dGdC) ,  = 530 nm 450 2 em p(dGdC) ,  = 528 nm 400 2 em pApU,  = 534 nm em pApU,  = 533 nm 400 em 350 350 300 300 250 250 200 200

150 150

100 100

Rel. Fluo. Int. (a.u.) Int. Rel.Fluo. 50

Rel. Fluo. Int. (a.u.) Int. Rel. Fluo. 50

0 0 0,0 1,0x10-5 2,0x10-5 3,0x10-5 4,0x10-5 5,0x10-5 6,0x10-5 0,0 1,0x10-5 2,0x10-5 3,0x10-5 4,0x10-5 5,0x10-5 6,0x10-5 c(polynucleotide) / M c(polynucleotide) / M

(a) (b)

7 FigureFigure 2.2. Changes in fluorescence of: ( (aa)) 11 ((λexcexc = =505505 nm nm,, c =c 5= ×5 10−710 M) orM) (b or) 2 ((b)exc2 =(λ 505exc =nm505, c = nm, 5 × × − c10= −57 M10); upon7 M); addition upon addition of polynucleotides. of polynucleotides. Done Doneat pH at 7.0; pH sodium 7.0; sodium cacodylate cacodylate buffer, buffer, I = 0.05I = 0.05 M. M. × − AllWe titrationperformed data fluorimetric were processed titrations by also non-linear at pH 5, at fitting which to the the GCP Scatchard unit is protonated equation (McGhee, and, with vonan Hippeladditional formalism) positive [28 charge,29],, giving could binding interact constants with negatively (Table2). charged A ffinities DNA/RNA of both dyes backbone, to all ds-polynucleotidesincreasing overall were affinity similar, (Supplementary thus suggesting Materials that the in observedFigure S24 diff).erence Intriguingly, in emission the fluorimetric response is notresponse due to ofthe 1 strength was identical of interactions to experiments but more done likely at the pH result 7, while of fine 2 dishowedfferences a inslightly positioning different of cyaninefluorescence within recognition the binding pattern site and at pH consequence 5, showing freedom a small offluorimetric rotation around preference methine for mixed bond. Further,ct-DNA, somewhatfollowed stronger by GC-DNA binding and at pH AU 5-RNA. in comparison The reason to pH why 7 agrees both nicely new to dyes the aforementioned (1, 2) did not stronger show a stabilizationpronounced e impactffects (Table of pH1) on and their could fluorimetric be attributed response, to the protonationas previously of noted GCP moiety for e.g. at, series weakly of acidicpyrene conditions,— GCP analogs yielding [13 additional–15], is probably binding because interaction. cyanine fluorophore is cationic itself, and thus not so dependent in binding to DNA/RNA on the additional positive charge of GCP present only at pH 5. Table 2. Binding constants and spectroscopic properties of complexes (alog K /bInt) of conjugates 1 and All titration data were processed by non-linear fitting to the Scatchards equation (McGhee, von 2 with ds-polynucleotides calculated by processing fluorimetric titrations (c = 5 10 7 M), at pH = 7.0 Hippel formalism) [28,29], giving binding constants (Table 2). Affinities× of − both dyes to all ds- and pH = 5.0, sodium cacodylate buffer, I = 0.05 M. polynucleotides were similar, thus suggesting that the observed difference in emission response is

not due to the strength of interactionsctDNA but more p(dAdT) likely2 the p(dGdC)result of2 fine differencespApU in positioning of cyanine within the1 binding, pH 7.0 site and6.1/ 120consequence6.2 freedom/40 of6.4 rotation/160 around6.2/450 methine bond. Further, somewhat stronger1, pH binding 5.0 at pH6.2/ 600 5 in comparison6.3/610 to pH6.4 7 /220 agrees nicely6.5/600 to the aforementioned 2, pH 7.0 6.2/300 6.1/270 6.2/480 5.6/500 stronger stabilization effects (Table 1) and could be attributed to the protonation of GCP moiety at 2, pH 5.0 6.5/870 6.5/420 6.5/760 6.0/930 weakly acidic conditions, yielding additional binding interaction. aProcessing of titration data by means of Scatchard equation [28,29] gave values of ratio n[bound dye]/[polynucleotide] = 0.1 and 0.2, for easier comparison all log Ks values were re-calculated for fixed n = 0.1. Thus, the error of binding constantTable 2. value Binding varies withinconstants the halfand order spectroscopic of magnitude, properties and only diofff erencescomplexes of at least(alog one Ks order/bInt) ofof magnitude conjugates can 1 b beand considered 2 with ds as- significant.polynucleotides Correlation calculated coefficients by wereprocessing> 0.99 for fluorimetric all calculated titrations Ks; Int—fluorescence (c = 5 × 10−7 intensity M), at pH of dye/polynucleotide complex calculated by Scatchard equation, taking into account that free dyes were non-emissive. = 7.0 and pH = 5.0, sodium cacodylate buffer, I = 0.05 M.

2.3.3. Circular Dichroism (CD) ExperimentsctDNA p(dAdT)2 p(dGdC)2 pApU 6.1/120 6.2/40 6.4/160 6.2/450 To study in more structural1, pH 7.0 detail the properties of complexes formed, we used circular dichroism (CD) spectrophotometry [30]. In addition, the small molecule could, upon binding to polynucleotides, 6.2/600 6.3/610 6.4/220 6.5/600 acquire an induced (I)CD1, signal,pH 5.0 positioned at the absorption bands of the small molecule, in this case in the range 300–550 nm, not coinciding with the DNA/RNA CD signals. The character and the magnitude of the ICD2, signalpH 7.0 o ﬀer6.2/ valuable300 6.1/ information270 6.2/ for480 the determination5.6/500 of binding modes (intercalation, agglomeration, groove binding, etc.) [31]. Minor groove binding to ds-DNA orientates 2, pH 5.0 6.5/870 6.5/420 6.5/760 6.0/930 the ligand approximately at 45◦ with respect to the DNA chiral axis, giving a strong positive ICD band. Intercalation brings the aromatic moiety of the ligand in a coplanar arrangement with the base pairs, givingaProcessing only a weak of titration ICD band data (in by themeans majority of Scatchard of cases equation with a[2 negative8,29] gave sign values due of toratio parallel n[bound orientation dye]/ of the[polynucleotide] transition vector = 0.1 of and the ligand0.2, for easier and the comparison longer axis all log of theKs values surrounding were re- basecalculated pairs) for [31 fixed,32]. n = 0.1. Thus, the error of binding constant value varies within the half order of magnitude, and only

Molecules 2020, 25, 4470 7 of 14

The studied compounds are chiral, but the intensities of their CD spectra in the 230–500 nm range are negligible with respect to CD spectra of polynucleotides, allowing accurate correction of CD titrations. The addition of studied dyes resulted in a general decrease of DNA/RNA CD bands (240–290 nm, Supplementary Materials in Figures S25–S28), somewhat more pronounced for 2 in comparison with 1. Such a decrease of DNA/RNA chirality is usually associated with the small molecule-induced unwinding of polynucleotide double helix [31,32]. Intriguingly, 1 induced by far the strongest decrease in the AU-RNA CD spectrum (Supplementary Materials in Figure S26d), which could be correlated to the strongest ﬂuorescence emission increase of 1 upon addition of AU-RNA (Figure2). On the other hand, 2 induced similar changes in CD spectra of all DNA/RNA (Figure3) in accordance with similar emissionMolecules 20 increase20, 25, x FOR (Figure PEER2 REVIEW). 8 of 15

6 AU-RNA 3 x 5 0

2 AT-DNA 0 -2 r = 0.3 GC-DNA r = 0.5 2 0 -2

CD (mdeg) -4 ct-DNA 2 0 -2 250 275 300 325 350 375 400 425 450 475 500 525 550

 (nm)

5 Figure 3. Circular dichroism (CD) titration of ds-DNAs and ds-RNA (—; c = 2 10 −5 M) with 2 at molar Figure 3. Circular dichroism (CD) titration of ds-DNAs and ds-RNA (▬; c = 2×  10− M) with 2 at molar ratios r / = 0.3 (—) and 0.5 (—). Note that marked range 450–550 nm was multiplied by 5 ratios r[2] [2][polynucleotide]/[polynucleotide] = 0.3 (▬) and 0.5 (▬). Note that marked range 450–550 nm was multiplied by 5 forfor betterbetter visibility.visibility. DoneDone atat pHpH 7,7, sodiumsodium cacodylatecacodylate bubuffer,ffer, I I= = 0.050.05 M.M. At variance to all previously studied aryl-GCP conjugates [10–15] and even the Cy-GCP [17] 2.3.4. Biological Evaluation of 1 and 2 on Human Cell Lines (very close analog of 2), no ICD bands of GCP chromophore (at λ = 300 nm) were observed, pointing out thatThe foraim both, of the1 and biological2, GCP experiments moiety was notperformed uniformly was oriented to verify in respectthe actual to thecapability polynucleotide of these chiralDNA/RNA axis [ 32 binders], similar to as penetrate previously into noted cells, for to nucleobase-GCP visualize their intracellular analogs [33 ]. location and subcellular targets,Detailed and evaluate analysis their of λ anti= 450–600-proliferative nm range effect, (Figure to identify3, Supplementary their further Materials,potential applications S27) revealed as veryeither weak lead compounds or negligible toward ICD bandss fluorescent for 1 at anti both-prol pH,iferative at variance drugs, with or as the non much-cytotoxic more dyes pronounced suitable ICDfor intracellular bands for 2 applications(Figure3), the or latterin biochemical correlating studies with on strong DNA ICD and bands RNA [1,2,5 of cyanine,6]. chromophore, as alsoMTT seen assay previously. The studied for its close compounds analog Cy-GCP were screened[17]. Since by all the compounds MTT assay (1 , for2, and antiCy-GCP-proliferative [17]) bindactivity to ds-DNAagainst /humanRNA with lung similar carcinoma affinity (A549) (Table and2), itnormal seems lung that quinoline-tethered(WI38) cell line (Figure cyanines 4). Results ( 2 and Cy-GCPshowed moderate [17]) attain cytotoxic more uniform effects orientation on the A549 in cell respect line to(IC50 the polynucleotideof compounds 1 chiral and 2 axis is approximately (thus yielding well-defined8 and 10 μM, ICDrespectively bands)) [,32 and] in somewhat comparison weaker to poorly activity oriented on the cyanineWI-38 cell of line benzothiazole-tethered (IC50 of 15 μM for 1 analogand 10 μM1. for 2). The series of parent cyanine-amino acids showed similar cytotoxicity [19]; thus, the additionThe of ICD GCP bands did ofno2t havewere a exclusively significant influence positive for on AT-DNAthe anti-proliferative and AU-RNA, activity which of isconjugates. typical for single-molecule minor/major groove binding [31,32], at variance to bisignate ICD bands observed only for GC-containing DNAs, which are characteristic for groove aggregation/dimerization of cyanine dyes [31,32]. These bisignate bands can be correlated to the propensity of 2 for aggregation (see Figure1), and likely to occur within DNA major groove, since GC-DNA minor groove is sterically blocked by alternatively protruding amino groups of guanine [25,26].

Molecules 2020, 25, 4470 8 of 14

2.3.4. Biological Evaluation of 1 and 2 on Human Cell Lines The aim of the biological experiments performed was to verify the actual capability of these DNA/RNA binders to penetrate into cells, to visualize their intracellular location and subcellular targets, and evaluate their anti-proliferative effect, to identify their further potential applications as either lead compounds towards fluorescent anti-proliferative drugs, or as non-cytotoxic dyes suitable for intracellular applications or in biochemical studies on DNA and RNA [1,2,5,6]. MTT assay. The studied compounds were screened by the MTT assay for anti-proliferative activity against human lung carcinoma (A549) and normal lung (WI38) cell line (Figure4). Results showed moderate cytotoxic effects on the A549 cell line (IC50 of compounds 1 and 2 is approximately 8 and 10 µM, respectively), and somewhat weaker activity on the WI-38 cell line (IC50 of 15 µM for 1 and 10 µM for 2). The series of parent cyanine-amino acids showed similar cytotoxicity [19]; thus, theMolecules addition 2020, of25, GCP x FOR did PEER not REVIEW havea significant influence on the anti-proliferative activity of conjugates.9 of 15

Figure 4. Cytotoxic assay of compounds 1 and 2 performed on A549 and WI-38 cell line. Measured absorbanceFigure 4. Cytotoxic (λ = 600 nm) assay directly of compounds correlates with1 and cell 2 performed survival. Cells on A549 with addedand WI DMSO-38 cell in line. corresponding Measured concentrationabsorbance (λ represent = 600 control nm) directly (note toxicity correlates of DMSO with at cell 100 survival.µM but negligible Cells with eﬀects added at lower DMSO conc.). in Datacorresponding are presented concentration as mean SDrepresent madein control three replicates.(note toxicity The of representative DMSO at 100 dataµM but of two negligib independentle effects ± experimentsat lower conc.) that. Data yielded are similarpresented results as mean are shown. ± SD made in three replicates. The representative data of two independent experiments that yielded similar results are shown. Confocal Microscopy. All tested compounds penetrated the cell membrane of the A549 cell line withinConfocal a 90 min Microscopy incubation at. All 37 tested◦C, 5% compounds CO2. Both conjugates,penetrated 1theand cell2, membrane being non-ﬂuorescent of the A549 in cell a freeline state,within revealed a 90 min strong incubation green emissionat 37 °C, upon5% CO accumulating2. Both conjugates in mitochondria,, 1 and 2, being as proven non-fluorescent by co-localization in a free withstate MitoTracker, revealed strong Deep Red green (Figure emission5). upon accumulating in mitochondria, as proven by co- localizationIn order with to quantify MitoTracker the co-localization,Deep Red (Figure we 5) used. Pearson’s correlation coefficient (PCC) [34], correctedIn order for noise to quantify by Replicate-based the co-localization noise correction, we used correlation Pearson’s (RBNCC) correlation [35 ].coefficient The results (PCC) indicated [34], excellentcorrected co-localization for noise by of Replicatethe studied-based dyes with noise the correction commercial correlation mitochondrial (RBNCC) tracker [ (PCC35]. The0.91 results0.02). indicated excellent co-localization of the studied dyes with the commercial mitochondrial ±tracker (PCC 0.91 ± 0.02).

Molecules 2020, 25, 4470 9 of 14 Molecules 2020, 25, x FOR PEER REVIEW 10 of 15

Figure 5. 5. IntracellularIntracellular distribution distribution of of Compounds 1 and 22 comparedcompared to to MitoTracker. MitoTracker. Confocal Confocal microscopy of live A549 cells taken on Leica SP8 X confocal microscope, stained with 10 μMµM of compounds 1 andand 22 ((λexcexc = 505505 nm, λem == 550550 nm); nm); RIGHTRIGHT: :overlay overlay with with the the MitoTracker MitoTracker Deep RedRed ((λexcexc == 644644 nm, nm, λemem = 665= 665 nm). nm).

Obtained results results pointed pointed out out that that1 and 1 and2 specifically 2 specifically accumulate accumulate in mitochondria, in mitochondria, same as same parent as cyanine-aminoparent cyanine- acidsamino [19 acids], thus [1 showing9], thus that showing conjugation that conjugation with GCP did with not GCP influence did not intracellular influence uptakeintracellular and targeting. uptake and It shouldtargeting. be stressedIt should that be stressed the original that fluorophoresthe original fluorophores TO and YO, asTO well and as YO, their as derivativeswell as their (TOTO, derivatives YOYO, (TOTO, and their YOYO analogs),, and their are analogs), typical DNA are typical intercalators, DNA intercalators and if able to, and enter if cells,able theyto enter act cells, as nuclei they fluorescenceact as nuclei stainsfluorescence [7]. Further, stains [ novel7]. Further, conjugates novel1 conjugatesand 2 are slightly1 and 2 are more slightly toxic thanmore parenttoxic than cyanine-amino parent cyanine acids-amino [19] and acids thus [19 possess] and thus somewhat possess increasedsomewhat theragnostic increased theragnostic ability—to fluorescentlyability — to fluorescently mark mitochondria mark mitochondria and impair cellular and impair function. cellular function.

3. Conclusions 3. Conclusions The here presented study showed that the linker tethering position on cyanine moiety has a strong The here presented study showed that the linker tethering position on cyanine moiety has a impact on the intrinsic properties of cyanine-amino acid-guanidiniocarbonyl-pyrrole conjugates in strong impact on the intrinsic properties of cyanine-amino acid-guanidiniocarbonyl-pyrrole water. Namely, a derivative with a positive charge more exposed to the solvent (Scheme1, 2) showed conjugates in water. Namely, a derivative with a positive charge more exposed to the solvent (Scheme significantly stronger aggregation propensity with respect to the analog with a more centered positive 1, 2) showed significantly stronger aggregation propensity with respect to the analog with a more charge (Scheme1, 1). Also, 1 and 2 showed significantly different fluorimetric and circular dichroism centered positive charge (Scheme 1, 1). Also, 1 and 2 showed significantly different fluorimetric and (induced CD) sensing of various ds-DNA and ds-RNA. The analog with a more centered positive circular dichroism (induced CD) sensing of various ds-DNA and ds-RNA. The analog with a more charge (Scheme1, 1) showed pronounced fluorimetric and CD selectivity toward AU-RNA, while centered positive charge (Scheme 1, 1) showed pronounced fluorimetric and CD selectivity toward the positive charge of 2 was more exposed to the solvent almost completely abolished fluorimetric AU-RNA, while the positive charge of 2 was more exposed to the solvent almost completely selectivity. However, as a result of more pronounced aggregation ability, only 2 showed measurable abolished fluorimetric selectivity. However, as a result of more pronounced aggregation ability, only induced (I)CD bands, yielding different ICD band patterns for AT(U)-polynucleotides with respect 2 showed measurable induced (I)CD bands, yielding different ICD band patterns for AT(U)- to GC-containing-polynucleotides. polynucleotides with respect to GC-containing-polynucleotides. Interactions of both studied compounds with DNA/RNA were, to some extent, controllable by Interactions of both studied compounds with DNA/RNA were, to some extent, controllable by pH due to GCP protonation, yielding somewhat stronger affinity and thermal stabilization effects at pH due to GCP protonation, yielding somewhat stronger affinity and thermal stabilization effects at pH 5. However, pH control did not result in pronounced differences like noticed for pyrene analogs pH 5. However, pH control did not result in pronounced differences like noticed for pyrene analogs PE1 (at pH 5, an opposite emission change of GC- or AT-DNA, respectively) [15], as well as for other PE1 (at pH 5, an opposite emission change of GC- or AT-DNA, respectively) [15], as well as for other pyrene-GCP analogs [13,14]. The reason is cyanine fluorophore being cationic itself, and thus not so pyrene-GCP analogs [13,14]. The reason is cyanine fluorophore being cationic itself, and thus not so dependent in binding to DNA/RNA on the additional positive charge of GCP present only at pH 5. dependent in binding to DNA/RNA on the additional positive charge of GCP present only at pH 5. Preliminary screening on human tumor and normal lung cell lines showed that both dyes very efficiently enter living cells and specifically accumulate in mitochondria. Although spectrophotometric results showed that studied compounds bind equally efficiently to all ds-DNA

Molecules 2020, 25, 4470 10 of 14

Preliminary screening on human tumor and normal lung cell lines showed that both dyes very efficiently enter living cells and specifically accumulate in mitochondria. Although spectrophotometric results showed that studied compounds bind equally efficiently to all ds-DNA or ds-RNA, intracellular localization is quite specific for mitochondria. Since cyanine dye used in the construction of compounds (TO) usually localizes in the cell nucleus, it seems that amino acid residue with GCP unit changes the target, likely because of the fine equilibrium of steric, hydrophobic, and H-bonding interactions. To determine the exact set of interactions, which take place in cellulo, further experiments are needed, which would in detail characterize cellular uptake mechanism (likely related to some of the membrane protein transporters, binding studied molecules and transporting them into mitochondria), and then in mitochondria determining whether ds-DNA or protein is the main target. The introduction of GCP did not have a pronouncedly different biological effect in comparison to parent cyanine-amino acid; nevertheless, the novelty is that for the first time, the GCP unit was introduced into mitochondria, thus proving that cyanine-amino acids can be used as mitochondria- targeting transporters even for rather large molecular moieties. Thus, compounds 1 and 2 are behaving as intriguing novel mitochondria-specific theragnostic dyes. Namely, similar molecules or supramolecular constructs characterized by simultaneous highly selective intracellular fluorescence localization and biological effects (many of them cyanine analogs), are increasingly studied for so-called theragnostic (therapy and diagnostic) applications [36–38]. Further development of here presented lead compounds (1, 2) would explore the applicability of pH (based on GCP moiety) or thermochemical control of mitochondrial function as a tool of controlling cell life cycles.

4. Materials and Methods

4.1. General Information Solvents were distilled from appropriate drying agents shortly before use. TLC was carried out on TLC Silica gel 60 F254 Plastic sheets, and preparative thin layer (2 mm) chromatography was done on Merck 60 F254 plates (Merck KGaA, Darmstadt, Germany). NMR spectra were recorded on AV600 and AV300 MHz spectrometers (Bruker BioSpin GmbH, Rheinstetten, Germany), operated at 600.13 or 1 13 300.13 MHz for H nuclei and at 150.92 MHz or 75.46 MHz for C, using DMSO-d6 (δH: 2.50 ppm, δC: 39.52 ppm) or CDCl3 (δH: 7.26 ppm, δC: 77.16 ppm) as the internal standard. Mass spectrometry was performed on an Agilent 6410 Triple Quad mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). High-resolution mass spectra (HRMS) were obtained using a Q-TOF2 hybrid quadrupole time-of-ﬂight mass spectrometer (Micromass, Cary, NC, USA).

4.2. Synthesis Compound 1 (E)-5-bromo-3-(4-(4-(2-(5-(carbamimidoylcarbamoyl)-1H-pyrrole-2-carboxamido)-3-methoxy-3-oxopropyl)- 1H-1,2,3-triazol-1-yl)butyl)-2-((1-methylquinolin-4(1H)-ylidene)methyl)benzo[d]thiazol-3-ium) was prepared previously.13 Compound 2 (E)-5-bromo-2-((1-(4-(4-(2-(5-(carbamimidoylcarbamoyl)-1H-pyrrole-2-carboxamido)-3-methoxy-3- oxopropyl)-1H-1,2,3-triazol-1-yl)butyl)quinolin-4(1H)-ylidene)methyl)-3-methylbenzo[d]thiazol-3-ium)

The reaction mixture of Cy-amino acid and GCP in acetonitrile with HCTU and Et3N was left stirring at room temperature under argon for 3 days. Boc-protected product was deprotected by the addition of TFA, the solvent was evaporated, and the crude product was recrystallized from the methanol-ether mixture and washed with water to give a pure product in 60% yield as a red solid. Molecules 2020, 25, 4470 11 of 14

1H NMR (600 MHz, DMSO) δ 12.69 (s, 1H, NH), 12.46 (s, 1H, NH), 10.95 (s, 1H, NH), 8.83 (dd, J = 24.5, 13.1 Hz, 1H, Ar), 8.65 (d, J = 38.3 Hz, 1H, Ar), 8.16 (d, J = 12.2 Hz, 2H, Ar), 8.06–8.00 (m, 3H, Ar + NH), 7.98–7.91 (m, 2H, Ar), 7.76 (d, J = 7.3 Hz, 2H, Ar), 7.67 (d, J = 8.2 Hz, 1H, Ar), 7.56 (d, J = 5.0 Hz, 1H, Ar), 7.38 (d, J = 15.1 Hz, 1H, Ar), 7.03 (t, J = 25.5 Hz, 1H, Ar), 6.93–6.82 (m, 2H, Ar), 6.70 (s, 1H, CH), 4.63 (dd, J = 57.4, 14.8 Hz, 2H, CH2), 4.43–4.30 (m, 1H, CH), 4.00 (s, 3H, CH3), 3.84 (d, J = 22.3 Hz, 2H, CH2), 3.60 (s, 3H, CH3), 2.97–2.86 (m, 2H, CH2), 1.97–1.69 (m, 4H, 2XCH2) ppm.13C NMR (151 MHz, DMSO) δ 162.81, 160.09, 154.96, 149.00, 144.84, 142.34, 142.07, 137.15, 133.61, 127.54, 126.84, 125.85, 125.00, 124.30, 123.21, 121.52, 121.06, 116.70, 115.57, 108.36, 88.38, 83.51, 54.22, 52.43, 34.30, 33.95, 26.11 ppm. HRMS (MALDI-TOF/TOF): m/z calcd for C H BrN O S2+ ([M H]+) 35 37 10 4 − 773,1980, found 773,1982.

4.3. Study of DNA/RNA Interactions All measurements were performed in aqueous buffer solution (pH = 7.0 or pH 5.0, I = 0.05 M, sodium cacodylate/HCl buffer). The UV-Vis spectra were recorded on a Varian Cary 100 Bio spectrometer, fluorescence spectra were recorded on a Varian Cary Eclipse fluorimeter, and CD spectra were recorded on JASCO J815 spectropolarimeter at 25.0 ◦C using appropriate quartz cuvettes (path length: 1 cm). Polynucleotides were purchased as noted: poly dAdT–poly dAdT, poly A–poly U, poly A, poly G, poly C, poly U (Sigma), calf thymus (ct)-DNA (Aldrich) and dissolved in sodium cacodylate buffer, I = 0.05 M, pH = 7.0. The ct-DNA was additionally sonicated and filtered through a 0.45 mm filter to obtain mostly short (ca. 100 base pairs) rod-like B-helical DNA fragments [39]. Polynucleotide concentration was determined spectroscopically [40] as the concentration of phosphates (corresponds to c(nucleobase)). Circular dichroism (CD) spectra were recorded on JASCO J-815 spectropolarimeter at room temperature using 1 cm path quartz cuvettes with a scanning speed of 200 nm/min (an average of three accumulations). A buffer background was subtracted from each spectrum. CD experiments were performed by adding portions of the compound stock solution into the solution of the polynucleotide (c = 2 10–5 M). × Thermal melting curves for ds-DNA, ds-RNA, and their complexes with studied compounds were determined as previously described [41] by following the absorption change at 260 nm as a function of temperature. The absorbance of the ligands was subtracted from every curve, and the absorbance scale was normalized. Tm values are the midpoints of the transition curves determined from the maximum of the first derivative and checked graphically by the tangent method [15]. The ∆Tm values were calculated subtracting Tm of the free nucleic acid from Tm of the complex. Every ∆Tm value here reported was the average of at least two measurements. The error in ∆Tm is 0.5 C. ± ◦ 4.4. Biological Evaluation of 1 and 2 on Human Cell Lines Study model. Experiments have been performed using two human cell lines, epithelial human lung adenocarcinoma A549 (ATCC®CCL-185™) and human normal lung fibroblast WI-38 (ATCC®CCL-75™). Both cell lines adhere to plastic and glass surface and are maintained in the culture under same conditions. Cells were grown in Dulbecco Modified Eagle0s Medium (DMEM, Sigma Aldrich, St Louis, MO, USA) supplemented with 10% of fetal bovine serum (FBS, Sigma Aldrich, St Louis, MO, USA) at 37 ◦C and 5% CO2 in a humified atmosphere. Cells were passaged twice per week in order to retain maximum confluency of 70–80%. Cells exhibiting normal morphology without any contamination signs, were kept in culture and used in all further experiments. Two biological replicas have been performed for all experiments. Cytotoxicity assay—MTT. Studied compounds 1 and 2 were dissolved in an appropriate volume of dimethyl sulfoxide solution (DMSO, Gram-Mol, Zagreb, Croatia) under sterile conditions, in order to obtain the stock of 10 mM solution. Solutions were kept in the dark and stored at 20 C in order to − ◦ prevent degradation. Prior to each assay, a new fresh working solution has been prepared from the stock solution. For testing the cytotoxic effects of each chemical compound on A549 and WI-38 cells, Molecules 2020, 25, 4470 12 of 14 stock solutions have been diluted with DMEM (10% FBS) to get a range of concentrations (100, 10, 5, 1, and 0.1 µM) which have been tested using a MTT test. Cells were seeded on a 96 well plate at a concentration of 7 103 cells/well in 100 µL of DMEM (10% FBS) and left in the incubator overnight × (37 ◦C, 5% CO2). The next day, 100 µL of the working solution was added to the wells in that manner that the final concentration of tested compounds was obtained in the total volume of 200 µL/well. All conditions were tested in triplicates. Cells treated with the same dilutions of DMSO represented control, while cells treated only with DMEM (10% FBS) represented negative control. The plate was then incubated for the next 72 h (37 ◦C, 5% CO2). After the incubation, the medium was removed, and the 40 µL of MTT solution was added to each well (treated wells, control wells containing DMSO, cells grown only in medium, and the empty well, served as blank). The plate was incubated in the cell incubator for 3 h, allowing the formazan crystals to form. After 3 h, 170 µL of DMSO was added in each well and put on a shaker for 20 minutes, allowing crystals to dissolve. The absorbance of the MTT-formazan product was measured with a microplate reader at 600 nm. Absorbance value directly correlates with a cell survival. Co-localization Assay. Live imaging of the cells treated with compounds was performed on the A549 cell line. Cells were seeded in Ibidi imaging cell chambers (Ibidi®, Germany) in 500 µL of medium, with the concentration of 5 104 cells/well, and left in the cell incubator for 48 h (37 C, 5% CO ). After × ◦ 2 two days, cells were treated with a 10 µM solution of each compound and left in the cell incubator for 90 min to allow the compound to enter the cells. After the incubation, the medium was changed, and 500 µL of 100 nM MitoTracker Deep Red solution (Invitrogen, Molecular Probes) was added to the chambers. Cells were incubated for 20 min (37 ◦C, 5% CO2), allowing MitoTracker to enter the cells. After incubation, the medium was replaced with 500 µL of fresh medium. Co-localization of compounds (λexc = 505 nm, λem = 550 nm) and mitochondria (MitoTracker λexc = 644 nm, λem = 665 nm) was then visualized and confirmed using Leica SP8 X confocal microscope (Leica Microsystems, Germany).

Supplementary Materials: Supplementary Materials are available online, structural properties of studied DNA and RNA; physico-chemical properties of studied compounds aqueous solutions, additional experimental data (fluorimetric and CD titrations, thermal denaturation experiments) of interactions with double-stranded DNA/RNA. Author Contributions: T.Š. and M.K. share the first authorship for equal contributions; T.Š. synthesized compounds and contributed to titration experiments, M.K. performed the titration experiments, analyzed results, prepared samples for biological experiments, and assisted in writing paper. K.B. and D.M. performed and analyzed biological experiments and contributed to the writing of the paper. I.P. and C.S. designed the research concept, and I.P. wrote the paper. All authors have read and agreed to the published version of the manuscript. Funding: The financial support of the Croatian Science Foundation projects IP-2013-11-1477 and IP-2018-01-5475 is gratefully acknowledged. Acknowledgments: Authors would like to dedicate this work to the late Carsten Schmuck for his essential contribution to this work and many previously published. Authors are grateful to Ivo Crnolatac for help with quantification of the co-localization by Pearson’s correlation coefficient and by Replicate-based noise correction correlation (RBNCC). Conflicts of Interest: The authors declare no conflict of interest.

References

1. Drummen, G.P.C. Fluorescent Probes and Fluorescence (Microscopy) Techniques—Illuminating Biological and Biomedical Research. Molecules 2012, 17, 14067–14090. [CrossRef][PubMed] 2. Trinquet, E.; Mathis, G. Fluorescence technologies for the investigation of chemical libraries. Mol. Biosyst. 2006, 2, 381. [CrossRef][PubMed] 3. Bhaduri, S.; Ranjan, N.; Arya, D.P. An overview of recent advances in duplex DNA recognition by small molecules. Beilstein J. Org. Chem. 2018, 14, 1051–1086. [CrossRef][PubMed] 4. Demeunynck, M.; Bailly, C.; Wilson, W.D. (Eds.) DNA and RNA Binders; Wiley-VCH: Weinheim, Germany, 2002. 5. Baylin, S.B.; Schuebel, K.E. The epigenomic era opens. Nature 2007, 448, 548. [CrossRef][PubMed] 6. Li, K.K.; Luo, C.; Wang, D.X.; Jiang, H.L.; Zheng, Y.G. Chemical and biochemical approaches in the study of histone methylation and demethylation. Med. Res. Rev. 2012, 32, 815. [CrossRef] Molecules 2020, 25, 4470 13 of 14

7. Tatikolov, A. Polymethine dyes as spectral-fluorescent probes for biomacromolecules. J. Photochem. Photobiol. C Photochem. Rev. 2012, 13, 55–90. [CrossRef] 8. Shindy, H.A. Basics, Mechanisms and Properties in the Chemistry of Cyanine Dyes: A Review Paper. Mini-Rev. Org. Chem. 2012, 9, 352–360. [CrossRef] 9. Armitage, B. Cyanine Dye-DNA Interactions: Intercalation, Groove Binding and Aggregation. In DNA Binders and Related Subjects; Springer: Berlin/Heidelberg, Germany, 2005; pp. 55–76. 10. Matić,J.; Šupljika, F.; Tandarić,T.; Dukši, M.; Piotrowski, P.; Vianello, R.; Brozović,A.; Piantanida, I.; Schmuck, C.; RadićStojković,M. DNA/RNA recognition controlled by the glycine linker and the guanidine moiety of phenanthridine peptides. Int. J. Biol. Macromol. 2019, 134, 422–434. [CrossRef] 11. Maity, D.; Matković,M.; Li, S.; Ehlers, M.; Wu, J.C.; Piantanida, I.; Schmuck, C. Peptide-Based Probes with an Artificial Anion-Binding Motif for Direct Fluorescence "Switch-On" Detection of Nucleic Acid in Cells. Chem. Eur. J. 2017, 23, 17356–17362. [CrossRef] 12. Schmuck, C. A Journey through 12 Years of Interacting Molecules: From Artificial Amino Acid Receptors to the Recognition of Biomolecules and Switchable Nanomaterials. SYNLETT 2011, 2011, 1798–1815. [CrossRef] 13. Hernandez-Folgado, L.; Baretić,D.; Piantanida, I.; Marjanović,M.; Kralj, M.; Rehm, T.; Schmuck, C. Guanidiniocarbonyl-pyrrole-aryl derivatives: Structure tuning for spectrophotometric recognition of specific DNA and RNA sequences and antiproliferative activity. Chem. A Eur. J. 2010, 16, 3036–3056. [CrossRef] [PubMed] 14. Hernandez-Folgado, L.; Schmuck, C.; Tomić,S.; Piantanida, I. A novel pyrene-guanidiniocarbonyl-pyrrole cation efficiently differentiates between ds-DNA and ds-RNA by two independent, sensitive spectroscopic methods. Bioorg. Med. Chem. Lett. 2008, 18, 2977–2981. [CrossRef][PubMed] 15. Radić-Stojković,M.; Piotrowski, P.; Schmuck, C.; Piantanida, I. Short, rigid linker between pyrene and guanidiniocarbonyl-pyrrole induced new set of spectroscopic responses to ds-DNA secondary structure. Org. Biomol. Chem. 2015, 13, 1629–1633. [CrossRef][PubMed] 16. Smidlehner, T.; Badovinac, M.; Piantanida, I. Pyrene-cyanine conjugates as multipurpose fluorescent probes for non-covalent recognition of ds-DNA, RNA and proteins. N. J. Chem. 2018, 42, 6655–6663. [CrossRef] 17. Šmidlehner, T.; Karaˇcić,Z.; Tomić,S.; Schmuck, C.; Piantanida, I. Fluorescent cyanine-guanidiniocarbonyl- pyrrole conjugate with pH-dependent DNA/RNA recognition and DPP III fluorescent labelling and inhibition properties. Monatsh. Chem. 2018, 149, 1307–1313. [CrossRef] 18. Cehić,M.;´ SućSajko, J.; Karaˇcić,Z.; Piotrowski, P.; Šmidlehner, T.; Jerić,I.; Schmuck, C.; Piantanida, I.; Tomić,S. The guanidiniocarbonylpyrrole—Fluorophore conjugates as theragnostic tools for DPP III monitoring and inhibition. J. Biomol. Struct. Dyn. 2020, 38, 3790–3800. [CrossRef] 19. Šmidlehner, T.; Kurutos, A.; Slade, J.; Belužić,R.; Ang, D.L.; Rodger, A.; Piantanida, I. Versatile Click Cyanine Amino Acid Conjugates Showing One-Atom-Influenced Recognition of DNA/RNA Secondary Structure and Mitochondrial Localisation in Living Cells. Eur. J. Org. Chem. 2018, 2018, 1682–1692. [CrossRef] 20. Li, M.; Ehlers, M.; Schlesiger, S.; Zellermann, E.; Knauer, S.; Schmuck, C. Incorporation of a Non-Natural Arginine Analogue into a Cyclic Peptide Leads to Formation of Positively Charged Nanofibers Capable of Gene Transfection. Angew. Chem. Int. Ed. 2016, 55, 598–601. [CrossRef] 21. Kuchelmeister, H.Y.; Karczewski, S.; Gutschmidt, A.; Knauer, S.; Schmuck, C. Utilizing Combinatorial Chemistry and Rational Design: Peptidic Tweezers with Nanomolar Affinity to DNA can be transformed into Efficient Vectors for Gene Delivery by Addition of a Lipophilic Tail. Angew. Chem. Int. Ed. Engl. 2013, 52, 14016–14020. [CrossRef] 22. Šmidlehner, T.; Piantanida, I. Novel DNA/RNA-targeting amino acid beacon for the versatile incorporation at any position within the peptide backbone. Amino Acids 2017, 49, 1381–1388. [CrossRef] 23. Mergny, J.-L.; Lacroix, L. Analysis of Thermal Melting Curves. Oligonucleotides 2003, 13, 515–537. [CrossRef] [PubMed] 24. Wilson, W.D.; Ratmeyer, L.; Zhao, M.; Strekowski, L.; Boykin, D. The search for structure-specific nucleic acid-interactive drugs: Effects of compound structure on RNA versus DNA interaction strength. Biochemistry 1993, 32, 4098–4104. [CrossRef][PubMed] 25. Cantor, C.R.; Schimmel, P.R. Biophysical Chemistry; WH Freeman and Company: San Francisco, CA, USA, 1980; Volume 3. 26. Egli, M.; Saenger, W. Principles of Nucleic Acid Structure; Springer: Berlin/Heidelberg, Germany, 1983. Molecules 2020, 25, 4470 14 of 14

27. Tumir, L.M.; Piantanida, I.; Cindrić, I.J.; Hrenar, T.; Meić, Z.; Žinić, M. New permanently charged phenanthridinium-nucleobase conjugates. Interactions with nucleotides and polynucleotides and recognition of ds-polyAH+. J. Phys. Org. Chem. 2003, 16, 891–899. [CrossRef] 28. Scatchard, G. The attractions of proteins for small molecules and ions. Ann. N.Y. Acad. Sci. 1949, 51, 660–672. [CrossRef] 29. McGhee, J.D.; Hippel, P.H.V.Theoretical aspects of DNA-protein interactions: Co-operative and non-co-operative binding of large ligands to a one-dimensional homogeneous lattice. J. Mol. Biol. 1974, 86, 469–489. [CrossRef] 30. Rodger, A.; Norden, B. Circular Dichroism and Linear Dichroism; Oxford University Press: New York, NY, USA, 1997; Chapter 2. 31. Eriksson, M.; Nordén, B. Methods in Enzymology; Chaires, J.B., Waring, M.J., Eds.; Academic Press: Cambridge, MA, USA, 2001; Volume 340, pp. 68–98. 32. Šmidlehner, T.; Piantanida, I.; Pescitelli, G. Polarization spectroscopy methods in the determination of interactions of small molecules with nucleic acids–tutorial. Beil. J. Org. Chem. 2018, 14, 84–105. [CrossRef] 33. Ban, Ž.; Žinić,B.; Vianello, R.; Schmuck, C.; Piantanida, I. Nucleobase–Guanidiniocarbonyl-Pyrrole Conjugates as Novel Fluorimetric Sensors for Single Stranded RNA. Molecules 2017, 22, 2213. [CrossRef] 34. Dunn, K.W.; Kamocka, M.M.; McDonald, J.H. A practical guide to evaluating colocalization in biological microscopy. Am. J. Physiol.-Cell Phys. 2011, 300, C723–C742. [CrossRef] 35. Adler, J.; Pagakis, S.N.; Parmryd, I. Replicate-based noise corrected correlation for accurate measurements of colocalization. J. Microsc. 2008, 230, 121–133. [CrossRef] 36. Zhang, J.Y.; Liu, Z.N.; Lian, P.; Qian, J.; Li, X.W.; Wang, L.; Fu, W.; Chen, L.; Wei, X.B.; Li, C. Selective imaging and cancer cell death via pH switchable near-infrared fluorescence and photothermal effects. Chem. Sci. 2016, 7, 5995–6005. [CrossRef] 37. Zhao, X.; Yang, C.X.; Chen, L.G.; Yan, X.P. Dual-stimuli responsive and reversibly activatable theranostic nanoprobe for precision tumor-targeting and fluorescence-guided photothermal therapy. Nat. Commun. 2017, 8, ARTN 14998. [CrossRef] 38. Tan, X.; Luo, S.L.; Wang, D.C.; Su, Y.P.; Cheng, T.M.; Shi, C.M. A NIR heptamethine dye with intrinsic cancer targeting, imaging and photosensitizing properties. Biomaterials 2012, 33, 2230–2239. [CrossRef] 39. Chaires, J.B.; Dattagupta, N.; Crothers, D.M. Studies on interaction of anthracycline antibiotics and deoxyribonucleic acid: Equilibrium binding studies on interaction of daunomycin with deoxyribonucleic. Biochemistry 1982, 21, 3933–3940. [CrossRef][PubMed] 40. Tumir, L.-M.; Piantanida, I.; Novak, P.; Žinić,M. Interactions of novel phenanthridinium-nucleobase conjugates with complementary and non-complementary nucleotides in aqueous media. J. Phys. Org. Chem. 2002, 15, 599–607. [CrossRef] 41. Tumir, L.M.; Piantanida, I.; Juranović,I.; Meić,Z.; Tomić,S.; Žinić,M. Recognition of homo-polynucleotides containing adenine by phenanthridinium bis-uracil conjugate in aqueous media. Chem. Commun. 2005, 2561–2563. [CrossRef][PubMed]

Sample Availability: Samples of the compounds are not available from the authors.

© 2020 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 (http://creativecommons.org/licenses/by/4.0/).