Volume 23, Issue 23 e202200536
Research Article
Open Access

Synthesis, Photophysical Properties and Biological Evaluation of New Conjugates BODIPY: Dinuclear Trithiolato-Bridged Ruthenium(II)-Arene Complexes

Dr. Oksana Desiatkina

Dr. Oksana Desiatkina

Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Freiestrasse 3, 3012 Bern, Switzerland

Search for more papers by this author
Dr. Ghalia Boubaker

Dr. Ghalia Boubaker

Institute of Parasitology, Vetsuisse Faculty, University of Bern, Länggassstrasse 122, 3012 Bern, Switzerland

Search for more papers by this author
Dr. Nicoleta Anghel

Dr. Nicoleta Anghel

Institute of Parasitology, Vetsuisse Faculty, University of Bern, Länggassstrasse 122, 3012 Bern, Switzerland

Search for more papers by this author
Dr. Yosra Amdouni

Dr. Yosra Amdouni

Institute of Parasitology, Vetsuisse Faculty, University of Bern, Länggassstrasse 122, 3012 Bern, Switzerland

Laboratoire de Parasitologie, Université de la Manouba, Institution de la Recherche et de l'Enseignement Supérieur Agricoles, École Nationale de Médecine Vétérinaire de Sidi Thabet, 2020 Sidi Thabet, Tunisia

Search for more papers by this author
Prof. Dr. Andrew Hemphill

Corresponding Author

Prof. Dr. Andrew Hemphill

Institute of Parasitology, Vetsuisse Faculty, University of Bern, Länggassstrasse 122, 3012 Bern, Switzerland

Search for more papers by this author
Prof. Dr. Julien Furrer

Corresponding Author

Prof. Dr. Julien Furrer

Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Freiestrasse 3, 3012 Bern, Switzerland

Search for more papers by this author
Dr. Emilia Păunescu

Corresponding Author

Dr. Emilia Păunescu

Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, Freiestrasse 3, 3012 Bern, Switzerland

Search for more papers by this author
First published: 11 October 2022
Citations: 3

Graphical Abstract

Fifteen new dinuclear arene-ruthenium (II)-BODIPY conjugates were synthesized. Despite a fluorescence quenching effect, these conjugates can be used as fluorescent tracers. Their antiparasitic activity but also their toxicity against healthy human foreskin fibroblasts (HFFs) was reduced compared to the parent ruthenium compounds. TEM images revealed profound alterations of the tachyzoite mitochondrion.

Abstract

The synthesis, photophysical properties and antiparasitic efficacy against Toxoplasma gondii β-gal (RH strain tachyzoites expressing β-galactosidase) grown in human foreskin fibroblast monolayers (HFF) of a series of 15 new conjugates BODIPY-trithiolato-bridged dinuclear ruthenium(II)-arene complexes are reported (BODIPY=4,4-difluoro-4-bora-3a,4a-diaza-s-indacene, derivatives used as fluorescent markers). The influence of the bond type (amide vs. ester), as well as that of the length and nature (alkyl vs. aryl) of the spacer between the dye and the diruthenium(II) complex moiety, on fluorescence and biological activity were evaluated. The assessed photophysical properties revealed that despite an important fluorescence quenching effect observed after conjugating the BODIPY to the diruthenium unit, the hybrids could nevertheless be used as fluorescent tracers. Although the antiparasitic activity of this series of conjugates appears limited, the compounds demonstrate potential as fluorescent probes for investigating the intracellular trafficking of trithiolato-bridged dinuclear Ru(II)-arene complexes in vitro.

1 Introduction

The study of ruthenium complexes for therapeutical purposes is an active area of research for more than two decades.1 Developed initially as a potential alternative to platinum based anticancer drugs,2 ruthenium(II)-arene complexes were also considered for other pharmacological properties, notably as antiparasitic,3 and antibacterial compounds.4

Cationic trithiolato-bridged dinuclear ruthenium(II)-arene complexes (general formula for symmetric [(η6-arene)2Ru2(μ2-SR)3]+, and mixed [(η6-arene)2Ru2(μ2-SR1)2(μ2-SR2)]+ complexes) are highly cytotoxic against human cancer cells (low micromolar range IC50 values (half maximal inhibitory concentration))5 and present interesting antiparasitic efficacy against Toxoplasma gondii,6 Neospora caninum7 and Trypanosoma brucei.8 Yet little is known about the traffic and fate of these complexes in cells and how it relates to their anticancer or antiparasitic effect. Transmission Electron Microscopy (TEM) of T. gondii treated with various trithiolato-bridged dinuclear ruthenium(II)-arene compounds had identified the parasites mitochondrion as a potential target.6 Nonetheless, a better understanding of their mechanism of action and the identification of specific biological targets would allow to design and develop more efficient compounds.

ICP-MS9 (inductively coupled plasma mass spectrometry) or fluorescence microscopy10 of complexes that are fluorescent per se or that are tagged with fluorescent dyes can be used as support for the identification of the cellular localization of metal-based bioactive compounds. The development of traceable therapeutic agents as fluorophore-labelled conjugates of metal-based drugs is a promising approach.11 For example, in the quest of compounds presenting both therapeutic and imaging properties, the Ru(II)-arene moiety was coupled to numerous organic fluorophores including anthracene,12 pyrene,13 naphthalimide,14 coumarins,15 rhodamine,16 BODIPYs,11c or porphyrins.17 However, some of these conjugates were not suitable for bioimaging purposes, mostly due to emission quenching after coupling the metal unit to the fluorophore (e. g., through a photoinduced electron transfer (PET) or by de-excitation of the triplet excited state15b) was too important.

From the large library of fluorescent dyes, BODIPYs (boron dipyrromethene, 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene) remain among the most attractive. They are chemically inert (stable in physiological pH-range, only decomposing in strong acidic and basic conditions18), and well soluble in common organic solvents with photophysical properties mostly independent of solvent polarity. BODIPYs display high photostability, small Stokes shifts, high fluorescence quantum yields, neutral charge, and sharp absorption and emission bands,19 while their properties can be tuned via chemical modifications.20 Since BODIPY derivatives are also non-toxic,21 they are especially useful for cell imaging studies and as biological probes.19a, 22 Although BODIPY derivatives exhibit reduced activity in cellular experiments, anchoring this fluorophore to an organometallic moiety could not only significantly modify the physicochemical properties of the metal complex and change their intracellular localization, but also alter the mode of action of the conjugates. For instance, platinum-fluorophore complexes were not localized in the cellular nucleus,23 and some platinum complexes bearing BODIPYs exhibited preferential mitochondrial distribution.24

Numerous examples of biologically active metal-based compounds tethered with BODIPY dyes have been reported.10, 25 Various platinum,11a, 24b ruthenium,11c osmium,11c iridium,26 gold,27 titanium,28 and copper,29 complexes with BODIPY appendices at the level of the ligands were identified as potential bioactive traceable fluorescent probes, some examples being summarized in Figure 1.

Details are in the caption following the image

Structure and physicochemical properties of reported metal-complex – BODIPY conjugates.

For the BODIPY-labelled platinum compound A30 (Figure 1), highly antiproliferative against cancer cells, the complexation to the metal led to fluorescence quantum yield decrease, which was ascribed to the oxidative photoinduced electron transfer (PET) from the excited core of BODIPY to the pyridyl group.31 Using confocal laser scanning microscopy, A showed a distinct mitochondrial distribution in cancer cells and its fluorescence intensity was higher than that of the ligand, paralleling the enhanced cellular uptake of the complex vs. the free ligand.

Derivatives like B11a, 24a and C24a exhibited cytotoxic and DNA-damaging properties, along with robust in vivo fluorescence. For B microscopy revealed a predominately cytosolic/perinuclear localization, with nuclear distribution at higher concentrations,11a while C localized in the cytoplasm near the nucleus.24a

The nature of the ligand and that of the coordinated metal center can strongly influence the photophysical properties of the conjugates. Half-sandwich Ru(II) complex D32 (Figure 1) was highly fluorescent, photostable while exhibiting negligible cytotoxicity at concentrations used for imaging purposes. Live cell imaging investigations revealed that D localized specifically in the mitochondria, and coordination to the metal center led to fluorescence quenching.

Other properties of the BODIPY fluorophores, as the generation of singlet oxygen on light activation for photodynamic therapy/photocytotoxicity, were exploited.11b, 24b, 24c, 33 For example, BODIPY-functionalized Ru(II)-arene complex E11d (Figure 1) behaved as a potential theranostic agent, accumulating in the cancer cells lysosome, exhibiting high photo-cytotoxicity under visible light on cancer cells, while being less toxic in the dark. Similarly, dyad F (Figure 1) was shown to effectively enable cancer cells photo-inactivation,25a, 34 the coordination of the Ru(II)-arene unit being followed by an important fluorescence quenching.25a A partial quenching was noticed for F upon the complexation of the organometallic entity, an effect explained by intramolecular PET.25a, 34

For the half-sandwich iridium(III) complex G35 (Figure 1), the introduction of the pyridyl-BODIPY ligand increased the lipophilicity and the cytotoxicity. Living cell fluorescence imaging indicated both G and its free ligand were membrane permeant and accumulated in cells,35 being detected indistinctly as large diffuse zones and small bright spots in the cytoplasm, the amount of the dyad being higher compared to that of the ligand.

Complex H36 exhibited medium cytotoxicity towards cancer cells, with a strong influence of the BODIPY on the cellular uptake and preferred cell membranes accumulation without reaching the nuclei.

The BODIPY–phosphane ligand in I11c (Figure 1) led to theranostics featuring Ru(II), Os(II) and Au(I). In I the fluorescence was moderately quenched, attributed to the fact that the compound is prone to PET,37 and can promote BODIPY phosphorescence.38 The in vitro imaging showed that I and the BODIPY-phosphane ligand rapidly bind to the biological membranes, with no clear specificity, the uptake and distribution properties of I being determined by the BODIPY moiety.

Previous studies have shown that trithiolato-bridged diruthenium complexes are highly stable constituting good substrates for derivatization using the ′chemistry on the complex′ strategy. Conjugates can be obtained by anchoring molecules of interest on the bridge thiols, and hybrids with peptides,39 the anticancer drug chlorambucil,40 coumarin fluorophores,41 antimicrobial drugs42 and nucleobases43 were recently synthesized and assessed for their anticancer or antiparasitic properties. Interestingly, trithiolato-bridged diruthenium conjugates tagged with coumarin fluorophores showed promising anti-Toxoplasma properties, but also completely quenched fluorescence.41 Alternatively, BODIPY analogues have been used for the study of lipid metabolism in T. gondii.44

The current study was focused on the obtainment of new conjugates BODIPY-trithiolato-bridged diruthenium(II)-arene units as potential antiparasitic and intracellular fluorescent tracking agents. To have a better input on the parameters that can influence the photophysical properties and/or the cytotoxicity/antiparasitic activity of the dyads, different structural elements were varied as: (i) the type of the bond connecting the two entities (ester vs. amide), (ii) the length of the spacer, and iii) the nature of the BODIPY dye (with an meso-aliphatic or an aromatic substituent).

The photophysical properties of the new conjugates as well as those of the corresponding free dyes were studied. The antiparasitic activity of the compounds was assessed against the apicomplexan parasite T. gondii β-gal and the cytotoxicity of the compounds was determined on HFF. Representative derivatives were also submitted to TEM (Transmission Electron Microscopy) and fluorescence microscopy studies to gain more insight on the compounds′ potential targets and intracellular localization.

2 Results and Discussion

2.1 Chemistry

Three mix trithiolato diruthenium(II)-arene derivatives bearing hydroxy 2, amino 3 and carboxy 4 groups on one of the bridging thiols were synthesized following a two-step reported procedure41, 45 (Scheme 1), using the ruthenium dimer [Ru(η6-p-MeC6H4Pri)Cl]2Cl246 and 4-tert-butylbenzenemethanethiol for the first step, and the resulted dithiolato intermediate 1 and 4-mercaptophenol, 4-aminobenzenethiol and 2-(4-mercaptophenyl)acetic acid for the second step.

Details are in the caption following the image

Synthesis of the dinuclear dithiolato 1 and OH, NH2, CH2CO2H functionalized trithiolato ruthenium(II)-arene complexes 2, 3 and 4.

The synthetic approaches to obtain the borondipyrromethene core are based on the porphyrin chemistry research.47 A first library of carboxy-functionalized meso-BODIPY compounds 912 with aliphatic chains of various length between the fluorophore and the CO2H group was synthesized by adapting reported protocols.48 The BODIPYs were obtained by the condensation of 3-ethyl-2,4-dimethylpyrrole with four commercially available acid chlorides as acylium equivalent: methyl 4-chloro-4-oxobutyrate, methyl 6-chloro-6-oxohexanoate, methyl 8-chloro-8-oxooctanoate and methyl 10-chloro-10-oxodecanoate (Scheme 2).48a The acylpyrrole intermediates (not isolated) were reacted DIPEA (N,N-diisopropylethylamine) and BF3⋅OEt2 (boron trifluoride etherate) to afford the methyl protected carboxy BODIPY dyes 58 (yields 32–52 %). In a second step, the esters 58 were hydrolyzed in basic conditions (KOH)48b-48d leading to the isolation of the BODIPY carboxy analogues 912 in medium yields (45 %-quantitative, Scheme 2).

Details are in the caption following the image

Synthesis of the carboxy-BODIPY dyes 912.

To assess the influence of the functional group present on the fluorophore on the photophysical properties of the BODIPY-trithiolato-diruthenium conjugates, one hydroxy (14) and two amino-functionalized (17 and 18) meso-substituted BODIPY compounds with short alkyl spacers were synthesized following the reaction sequence presented in Scheme 3. 14 was obtained in two steps using reported protocols48a (Scheme 3 (top)). First, the acetyl protected intermediate 13 was synthesized by the reaction of 3-ethyl-2,4-dimethylpyrrole with 2-chloro-2-oxoethyl acetate (commercially available) as acylium equivalent. Addition of DIPEA to the intermediate acylpyrrole formed in situ, followed by application of BF3⋅OEt2 afforded the BODIPY dye 13 (35 %). Ester hydrolysis in basic conditions (LiOH) released the hydroxy BODIPY 14, isolated in 43 % yield. 17 and 18 were obtained following the reaction pathway presented in Scheme 3 (bottom) using previously described procedures.49 First, 3-ethyl-2,4-dimethylpyrrole was reacted with phthalimide protected derivatives 1,3-dioxo-2-isoindolineacetyl chloride and 1,3-dioxo-2-isoindolinebutanoyl chloride as acylium equivalents, affording intermediates 15 and 16 isolated with 44 and 26 % yield, respectively. In a second step, compounds 15 and 16 were deprotected using hydrazine in refluxing EtOH,49a and the BODIPY amino alkyl derivatives 17 and 18 were isolated in 25 and 18 % yield.

Details are in the caption following the image

Synthesis of the BODIPY dyes containing a hydroxy group 14 (top) and an amino group 17 and 18 (bottom).

The carboxy BODIPYs 912 were reacted with the diruthenium hydroxy and amine derivatives 2 and 3 affording the ester 1922, and respectively, amide dyads 2326 (Scheme 4).

Details are in the caption following the image

Synthesis of the ester 1922 (left) and amide 2326 (right) conjugates BODIPY-dinuclear trithiolato ruthenium(II)-arene complexes containing alkyl spacers of various lengths.

For the esters conjugates, EDCI (N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride) was used as coupling agent and DMAP (N,N-dimethylpyridin-4-amine) as base50 and 1922 were isolated in 12–93 % yield. For the amide analogues, the reactions were realized in the presence of EDCI and HOBt (1-hydroxybenzotriazole) as coupling agents and of DIPEA as basic catalyst,51 conjugates 2326 being isolated in 22–74 % yield.

The BODIPY hydroxy 14 and the amino 17 and 18 derivatives were reacted with carboxy diruthenium complex 4, using the reaction conditions presented in Scheme 5.

Details are in the caption following the image

Synthesis of the ester (left) and amide (right) conjugates BODIPY-dinuclear trithiolato ruthenium(II)-arene complexes 27, 28 and 29 containing short alkyl spacers.

The ester conjugate 27 was obtained using EDCI as coupling agent and DMAP as base and was isolated in 35 % yield. Amide dyads 28 and 29 were obtained in the presence of EDCI and HOBt as coupling agents and DIPEA as base and both were isolated in 52 % yield.

To further vary the photophysical properties, the use of structurally different BODIPY dyes was also considered. BODIPY derivatives presenting an aromatic unit in the meso-position were largely studied and represent an interesting option.19c

BODIPY derivative 30 was synthesized by reacting 3-ethyl-2,4-dimethylpyrrole with 4-(chloromethyl)benzoyl chloride as acylium equivalent (Scheme 6). Addition of DIPEA to the intermediate acylpyrrole formed in situ, followed by application of BF3⋅OEt2 afforded the BODIPY dye 30 isolated in 85 % yield.

Details are in the caption following the image

Synthesis of the meso-arene BODIPY dyes 30, 31 and 32 functionalized with chloromethylene, hydroxy and carboxy groups, respectively.

The condensation of pyrrole derivatives with aromatic aldehydes, followed by oxidation and complexation is a largely used method for the obtainment of meso-aryl BODIPY fluorophores,52 which led to abundant use of the meso-aryl group as a synthetic handle for the introduction of various functionalities.53 The TFA (trifluoroacetic acid) catalyzed condensation of 4-hydroxybenzaldehyde and 4-carboxybenzaldehyde with 3-ethyl-2,4-dimethylpyrrole (Scheme 6) afforded the corresponding dipyrromethane intermediates (not isolated) which were further oxidated with DDQ (2,3-dichloro-5,6-dicyano-p-benzoquinone) to yield dipyrrin structures. The dipyrrins were subjected to TEA (triethylamine) and BF3⋅OEt2 to afford the boron difluoride complexes 31 and 32 in low yield of 28 and 25 %, respectively (Scheme 6).

Nucleophilic substitution of the chlorine atom in intermediate 30 with the diruthenium amino derivative 3 in the presence of KI as activator, in basic conditions (DIPEA) (Scheme 7 (top)), allowed the obtainment of the BODIPY amino conjugate 33 isolated in 50 % yield. Hydroxy functionalized BODIPY derivative 31 was reacted with the diruthenium carboxy compound 4 using EDCI as coupling agent and DMAP as basic catalyst, to afford the ester conjugate 34 isolated in 30 % yield (Scheme 7 (bottom)).

Details are in the caption following the image

Synthesis of the amine (top) and ester (bottom) conjugates meso-aryl BODIPY-dinuclear trithiolato ruthenium(II)-arene complexes 33 and 34.

The carboxy meso-aryl BODIPY dye 32 was reacted with the diruthenium hydroxy and amine intermediates 2 and 3 leading to the obtainment of the ester and amide conjugates, 35 and, respectively, 36 (Scheme 8). In the first case, the reaction was run in the presence of EDCI as coupling agent and DMAP as basic catalyst and the ester dyad 35 was isolated in 25 % yield. The synthesis of the amide analogue 36 was realized in the presence of EDCI and HOBt as coupling agents and of DIPEA as basic catalyst, the conjugate being isolated in 56 % yield.

Details are in the caption following the image

Synthesis of the ester (left) and amide (right) dyads meso-aryl BODIPY-dinuclear trithiolato ruthenium(II)-arene complexes 35 and 36.

All compounds were fully characterized by 1H, 13C and, where suitable, 19F and 11B nuclear magnetic resonance (NMR) spectroscopy, high resolution electrospray ionization mass spectrometry (HR ESI-MS) and elemental analysis (see the Experimental Section Chemistry in Supporting Information for full details). Mass spectrometry corroborated the spectroscopic data with the trithiolato diruthenium conjugates 1929 and 3336 exhibiting molecular ion peaks corresponding to [M−Cl]+ ions.

For the assessment of the biological activity, the compounds were prepared as stock solutions in dimethylsulfoxide (DMSO), a solvent in which the derivatives present good solubility. Similarly, to previous reports,41, 45 1H-NMR spectra of conjugates 24, 1929 and 34 solubilized in DMSO-d6, recorded at 25 °C, 5 min and more than 1 month after sample preparation showed almost no changes (see Figures S6 and S8 in the Supporting Information), demonstrating the compounds high stability in this highly complexing solvent.

2.2 X-ray crystallography

The crystal structures of BODIPY derivatives 16 and 30 were established in the solid state by single-crystal X-ray diffraction substantiating the expected structure (ORTEP representation are shown in Figure 2, see Supporting Information for full experimental details and more related information). Data collection and refinement parameters are given in Table S2, whereas selected structural data are presented in Table 1.

Details are in the caption following the image

ORTEP representation of BODIPY derivatives 16 and 30 (thermals ellipsoids are shown with 50 % probability).

Table 1. Selected structural parameters for the crystal structures of 16 and 30.

Compound

B−N (Å)

B−F (Å)

N−C(CH3) (Å)

N−B−N (°)

F−B−F (°)

16

N1−B1 1.5396(18) N2−B1 1.5399(18)

F1−B1 1.3947(17) F2−B1 1.3918(17)

N1−C1 1.3534(16) N2−C9 1.3456(16)

N1−B1−N2 107.11(11)

F2−B1−F1 108.47(11)

30

N1−B1 1.5472(19) N2−B1 1.543(2)

F1−B1 1.391(2) F2−B1 1.385(2)

N1−C1 1.3471(19) N2−C9 1.3505(18)

N2−B1−N1 106.87(11)

F2−B1−F1 109.64(13)

Slow evaporation of solutions of 16 and 30 in CHCl3 afforded pink-violet single crystals suitable for X-ray diffraction. The crystallographic data revealed that 16 crystallizes in the triclinic system, space group Purn:x-wiley:14394227:media:cbic202200536:cbic202200536-math-0001 , while 30 crystallizes in the monoclinic system, space group P21/n. In both structures, the central six-membered ring of the BODIPY moiety is almost coplanar with the adjacent pyrrole rings, with a π-electron delocalization in the BODIPY core, as often observed for this class of compounds.54 The two B−N distances are similar, while for N1−C1 and N2−C9 the measured bond lengths reveal a pronounced double bond character.

2.3 Photophysical properties

The photophysical properties of the BODIPY derivatives 518 and 3032, and of the conjugates BODIPY-trithiolato ruthenium(II)-arene complexes 1929 and 3336 are summarized in Table 2. The absorption and emission spectra of compounds 7, 11, 21, 25, 32, 35 and 36 (10 μM solution in CHCl3) are comparatively shown in Figures 345.

Table 2. Photophysical properties of compounds 536 (10 μM solution in CHCl3).

Compound

urn:x-wiley:14394227:media:cbic202200536:cbic202200536-math-0002 [nm]

ϵ [M−1 cm−1]

urn:x-wiley:14394227:media:cbic202200536:cbic202200536-math-0003 [nm]

Δλ [nm]

ΦF [%]

Rhodamine 6G[a]

532.5

60832.8

557

24.5

75*

BODIPY derivatives with aliphatic handles

5

527

62440.3

543

16

79

6

524

57593.2

538

14

82

7

523

71699.1

538

15

87

8

522.5

53674.9

537

14.5

84

9

526.5

35884.7

540

13.5

77

10

523.5

24534.4

536

12.5

92

11

523

55178.2

536

13

83

12

522.5

51363.4

537

14.5

86

13

549

36932.2

569

20

72

14

544.5

55663.2

565

20.5

64

15

546, 309.5

8045.19

554

8

14

16

525

46390.0

539

14

64

17

534

56442.9

551

17

78

18

527.5

43566.5

542

14.5

63

Diruthenium-BODIPY conjugates with aliphatic spacers

19

529, 245

56519.5, 63901.6

543

14

11

20

524, 244.5

63619.5, 63006.0

539

15

17

21

523, 244.5

62739.9, 69086.4

538

15

14

22

522.5, 247

68351.7, 70393.6

538

15.5

18

23

524.5, 245.5

59063.5, 69777.7

540

15.5

12

24

521.5, 245

64531.5, 73078.6

536

14.5

12

25

521.5, 245

58855.7, 67572.0

536

14.5

14

26

521.5, 247.5

66142.7, 74163.7

536

14.5

18

27

536, 247.5

1598.3, 4660.3

552

16

10

28

539, 244.5

50505.5, 67274.0

556

17

5

29

521.5, 247.5

64983.9, 68330.6

537

15.5

6

BODIPY compounds with aromatic handles

30

527.5

56484.9

544

16.5

63

31

525.5

59696.8

541

15.5

74

32

528.5

48019.8

547

18.5

40

Diruthenium-BODIPY conjugates with aromatic spacers

33

526.5, 247.5

65652.2, 74004.7

545

18.5

6

34

527.5, 247.5

64733.1, 67122.7

545

17.5

15

35

529.5, 247.5

38450.4, 62860.5

548

18.5

17

36

527, 247.5

60159.8, 71312.0

544

17

19

  • [a] Values taken from ref. [55].
Details are in the caption following the image

UV-Vis absorption (left) and emission spectra (right) of rhodamine 6G, BODIPY compounds 7, 11, and the corresponding ester 21 and amide 25 conjugates BODIPY – trithiolato diruthenium complex, at 10 μM in CHCl3.

Details are in the caption following the image

UV-Vis absorption (left) and emission spectra (right) of rhodamine 6G, BODIPY compound 32 and the corresponding ester 35 and amide 36 conjugates BODIPY – trithiolato diruthenium complex, at 10 μM in CHCl3.

Details are in the caption following the image

UV-Vis absorption (left) and emission spectra (right) of ester 21 conjugate BODIPY – trithiolato diruthenium complex at various concentrations in CHCl3.

The BODIPY intermediates 58, in which the ester group is separated from the dye moiety by aliphatic chains of different lengths, exhibit typical19c absorption and emission bands between 522–527 nm and 537–543 nm respectively, with high quantum yields (ΦF=77–87 %). Slight hypsochromic shifts in both absorption and emission spectra can be observed after deprotection to acids 912, with minor changes in quantum yields. Interestingly, the length of the spacer chain affects both the absorption and emission wavelength, with shorter chains leading to more important bathochromic shifts. For example, compounds 1318 absorb at 525–549 nm, emit at 539–569 nm, and exhibit quantum yields lover (ΦF=63–78 %) compared to derivatives with longer chains. Compound 15 shows a low quantum yield of 14 %, probably due to the presence of the phthalimide group in close vicinity to the BODIPY unit. Meso-aryl compounds 3032 absorb at 525–528 nm and emit at 541–547 nm. As expected,19c changing the nature of the meso substituent from aliphatic to aromatic neither affects the absorption nor the emission of BODIPYs, but the quantum yields are significantly lower (ΦF=40–74 %).

As shown in Figure 3, Figure 4 and Figure 5 and in accordance with previous reports (see selected photophysical parameters for various compounds in Figure 1), upon excitation at 450 nm, the conjugates 1929 and 3336 emitted at 500–570 nm (green area) due to the presence of the BODIPY chromophores. Absorption spectra of all dyads (1929 and 3336) are similar (Figures 3 and 5) and correspond well to an overlapping of the absorption spectra of the two units (diruthenium intermediates and BODIPY dyes), with strong peaks at 200–300 nm and 450–570 nm regions. With the exception of compounds 11, 12 and 22, the BODIPY derivatives and conjugates present rather comparable Stokes’ shifts values between 13–16 nm (Table 2) similar to reported compounds.53

The introduction of the diruthenium moiety does not influence the absorption of the dye moiety, but however it induces an important fluorescence quenching irrespective of the nature of the polar group anchored on the BODIPY (e. g., CO2H in 912, OH in 14, or NH2 in 1718) or of the type of bond connecting the dyads’ units (ester vs. amide, e. g., 1922, 27 vs. 2326, 28, 29). A less important quenching effect was observed in the case of conjugates 3436 with more rigid meso-aryl BODIPY dyes. No direct correlation between the quenching intensity and length of the spacer was observed, even if shorter chains favor this outcome (e. g., 19, 2324, 2729). A similar quenching effect was observed for formerly reported BODIPY-labelled metal complexes.11c, 24b, 25a, 32, 35 This effect was explained by the fact that this type of compounds are prone to PET,37 and that they can promote phosphorescence of the BODIPY unit.38 In contrast, the fluorescence of a previously reported series of conjugates coumarin – trithiolato-bridged diruthenium complexes was almost completely quenched and the compounds could not be exploited for intracellular visualization.41

Despite a general substantial fluorescence quenching, a sufficient intense emission was measured for the dyads in solution, which enabled the study of compound 20 as potential tracking agents using fluorescence microscopy.

2.4 In vitro activity against T. gondii tachyzoites

The antiparasitic activity of all compounds regarding their ability to inhibit the in vitro proliferation of tachyzoites was evaluated on T. gondii β-gal (a transgenic strain constitutively expressing β-galactosidase) grown in HFF (human foreskin fibroblast) host cell monolayers. The cellular toxicity of these compounds was separately assessed on non-infected HFFs6 and quantified by alamarBlue assay. The compounds (conjugates, BODIPY dyes and respective intermediates) were subjected to a sequential biological screening strategy already applied in previous studies.45a For the primary screening, cultures were exposed to 0.1 and 1 μM of each compound for 72 h. As controls (100 % viability and 100 % proliferation), non-infected HFF and T. gondii infected HFF cultures were treated with 0.1 % DMSO. In the second screening, the selected compounds (molecules that when applied at 1 μM inhibited T. gondii β-gal proliferation by at least 90 % and did not impair HFF viability by more than 50 %) were submitted to dose response studies to determine the IC50 values, while cytotoxicity in HFF was assessed at 2.5 μM.

The non-modified trithiolato-bridged dinuclear ruthenium(II)-arene complexes 24 were formerly evaluated against T. gondii β-gal under identical conditions,41, 45 and the corresponding values were provided for comparison in Table 3 and Figure 6. The library of tested compounds included the BODIPY dyes and the corresponding intermediates 518 and 3032 (results summarized in Table S1 and Figure S1 in Supporting Information) and the diruthenium-BODIPY conjugates with aliphatic and aromatic spacers 1929 and, respectively, 3336 (Table 3 and Figure 6).

Table 3. Primary efficacy/cytotoxicity screening of the diruthenium compounds in non-infected HFF cultures and T. gondii β-gal tachyzoites cultured in HFF monolayers. The compounds selected for the determination of IC50 values against T. gondii β-gal are tagged with an asterisk.

Compound

HFF viability (%)

T. gondii β-gal growth (%)

0.1 μM

1 μM

0.1 μM

1 μM

Diruthenium intermediates

2[a]*

76±6

46±6

66±14

2±0

3[a]*

74±2

48±1

57±1

2±0

4[a]*

91±4

73±1

114±2

110±2

Diruthenium-BODIPY conjugates with aliphatic spacers

19*

99 ±5

87±7

13±2

0±0

20*

102±2

83±4

74±4

0±0

21*

102±2

75±11

22±0

0±0

22*

97±11

94±7

40±5

0±1

23

106±0

96±1

33±4

11±2

24

106±3

99±1

125±9

70±6

25

98±16

100±1

122±9

101±8

26

84±1

71±4

94±5

83±7

27*

107±5

97±3

126±16

1±1

28*

105±1

102±0

113±7

1±1

29

107±3

101±1

85±9

76±13

Diruthenium-BODIPY conjugates with aromatic spacers

33*

100±3

93±1

123±7

0±0

34

126±4

129±3

140±1

62±1

35

110±4

108±2

85±4

28±2

36

102±1

93±2

105±2

76±3

  • [a] Data for compounds 24 were previously reported.45a
Details are in the caption following the image

Clustered column chart showing the in vitro activities at 0.1 (A) and 1 (B) μM of the trithiolato diruthenium compounds on HFF viability and T. gondii β-gal proliferation. As controls (100 % viability and 100 % proliferation), non-infected HFF and T. gondii infected HFF cultures were treated with 0.1 % DMSO. For each assay, standard deviations calculated from triplicates are displayed on the graph. Data for compounds 24 were previously reported.45a

The biological activity of the hydroxy, amino and carboxy diruthenium compounds 24 was investigated in a former study.45a While 4 had very limited effect on HFF and parasites at both concentrations, 2 and 3 at 1 μM almost completely abolished parasite proliferation but also exhibited considerable cytotoxicity.

Both esters 1922 and amide analogues 2326 exerted limited cytotoxicity when applied at 1 μM. However, ester derivatives 1922 were significantly more efficient in inhibiting T. gondii β-gal proliferation compared to the respective amides 2326 especially when applied at 1 μM. Conjugates 1922 also appear to be more active and less toxic compared to corresponding hydroxy intermediate 2. Only amide 23 exhibited some activity against T. gondii, while derivatives 2426 with longer spacers between the BODIPY and the diruthenium unit did not affect the parasite proliferation. Amide 23 presented an improved activity/cytotoxicity profile as well compared to the respective diruthenium amino intermediate 3.

For conjugates 27, 28 and 29, both the nature of the bond between the two units and the length of the spacer strongly influenced the biological activity. Neither ester 27 nor its structurally related amide 28 affected the host cell viability even when applied at 1 μM, while both compounds completely inhibited parasite proliferation at the same concentration. Like the corresponding diruthenium carboxy intermediate 4, amide 29 did not exhibit cytotoxicity or antiparasitic effects at both tested concentrations.

Conjugates 3336, in which the fluorophore presents a meso-aryl handle, did not affect HFF viability at the tested concentrations and, except for amino-linked compound 33, they demonstrated only poor anti-Toxoplasma efficacy.

When applied at 1 μM, all BODIPY intermediates and dyes exhibited negligible effect on T. gondii β-gal proliferation and, apart from derivative 6, they did not affect the host cell proliferation by more than 50 % at the same concentration (Table S1 and Figure S1).

Seven new diruthenium-BODIPY conjugates fulfilled the selection criteria (compounds that when applied at 1 μM inhibited T. gondii β-gal proliferation by at least 90 % and did not impair HFF viability by more than 50 %) and were submitted to a second screening IC50 determination on T. gondii and toxicity assessment to host cells at 2.5 μM (Table 4).

Table 4. T. gondii β-gal half-maximal inhibitory concentration (IC50) values (μM) and viability of HFF cultures exposed to 2.5 μM for seven selected diruthenium-BODIPY conjugates, standard drug pyrimethamine and diruthenium intermediates 2 and 3.

Compound

IC50 [μM]

[LS; LI][b]

SE[c]

HFF viability at 2.5 μM (%)[d]

SD[e]

Pyrimethamine[a]

0.326

[0.396; 0.288]

0.052

99

6

Diruthenium intermediates

2[a]

0.117

[0.139; 0.098]

0.051

56

6

3[a]

0.153

[0.185; 0.127]

0.049

51

5

Diruthenium-BODIPY conjugates with aliphatic spacers

19

0.442

[0.524; 0.373]

0.169

79

0

20

0.483

[0.613; 0.381]

0.238

91

0

21

0.255

[0.272; 0.240]

0.060

93

0

22

0.419

[0.518; 0.339]

0.211

95

0

27

0.524

[0.555; 0.495]

0.057

91

0

28

0.385

[0.576; 0.257]

0.156

86

0

Diruthenium-BODIPY conjugates with aromatic spacers

33

0.512

[0.678; 0.387]

0.110

96

0

  • [a] Data for pyrimethamine, 2 and 3 were previously reported;45a 2 and 3 do not correspond to the first screening selection criteria, but the IC50 values and viability of HFF were determined for comparison purpose. [b] Values at 95 % confidence interval (CI); LS is the upper limit of CI and LI is the lower limit of CI. [c] The standard error of the regression (SE), represents the average distance that the observed values fall from the regression line. [d] Control HFF cells treated only with 0.25 % DMSO exhibited 100 % viability. [e] Standard deviation of the mean (six replicate experiments).

The selected conjugates exhibit IC50 values against T. gondii β-gal ranging from 0.255–0.524 μM, which are lower compared to formerly reported diruthenium complexes6, 45a or hybrid molecules in which the this organometallic moiety was functionalized with coumarin fluorophores41 or with antimicrobial drugs.42 This substantiates the critical role of the organic fragment appended to the diruthenium scaffold for the overall biological activity of the conjugates.

Dyads 19, 21, 22 and 28 had a similar impact on in vitro proliferation of T. gondii to that exerted by pyrimethamine (IC50=0.326 μM), a standard treatment for toxoplasmosis. For 19, 21, 22 and 28 the cytotoxicity against HFF was in a close range (79–95 %) and comparable to that of pyrimethamine (99 %). Conjugation of BODIPY dyes to the diruthenium moiety resulted in a significantly reduced cellular toxicity compared to corresponding hydroxy and amino functionalized diruthenium intermediates 2 and 3.

Overall, the screening results indicate that the potential of the BODIPY-diruthenium hybrids as antiparasitic therapeutic agents is moderate. Among the newly reported compounds, conjugate 21 exhibits the lowest IC50 value (0.255 μM). Yet, 21 is less active than the hydroxy diruthenium intermediate 2 but exhibits also lower toxicity to host cells.

Nonetheless, this type of dyads could be of potential interest as fluorescent probes for tracking their uptake and localization in cells and parasites.

2.5 Transmission electron microscopy

The ultrastructural changes induced by BODIPY ester conjugates 21 and 27 (presenting the lowest and the higher IC50 values against T. gondii β-gal from the current compound library) were studied by TEM. HFF monolayers were infected with T. gondii β-gal tachyzoites, and compounds were added after the infection was established. The cultures were either maintained as controls without addition of compounds (Figure 7), or at 24 h post-infection, treatments with 21 or 27 (administrated at 0.5 μM) were initiated (Figure 8). In all cases, cultures were fixed and processed for TEM at 48 h post-infection. In control cultures (Figure 7), T. gondii tachyzoites were found to be located within the cytoplasm of their host cells, within a parasitophorous vacuole that was delineated by a parasitophorous vacuole membrane. At 24 h post-infection, vacuoles contained mostly 2 tachyzoites (Figure 7A), while at 48 h, proliferation had resulted in larger vacuoles with an increased number of tachyzoites (Figure 7B). Proliferation of T. gondii takes place by endodyogeny, thus two daughter zoites emerge from one tachyzoite, as indicated in Figure 7C.

Details are in the caption following the image

TEM of T. gondii β-gal tachyzoites fixed at 24 h post-infection (A, C) and at 48 h post-infection (B). Panel A shows a low magnification view of several parasitophorous vacuoles within HFF host cells, a higher magnification view of the boxed area is shown in C. Tachyzoites proliferate by endodyogeny, and an emerging daughter zoites is indicated with the arrow. ap=apical part, mic=micronemes, dg=dense granules, mito=tachyzoite mitochondrion, hmito=host cell mitochondrion. Note the higher number of parasites within the parasitophorous vacuole fixed at 48 h. The yellow frames indicate the position of the visible parts of the mitochondrion.

Details are in the caption following the image

TEM of T. gondii β-gal tachyzoites treated with 0.5 μM of 21 (A–D) and 0.5 μM of 27 (E–H) fixed and processed 24 h after treatment initiation. A and C show lower magnification views of cultures treated with 21, the boxed area in A is enlarged in B, and a parasite exhibiting profound changes in the mitochondrion (mito) is enlarged in D. Note that matrix remnants are more clearly detectable in B than in D. Larger parasitophorous vacuoles are found in cultures treated with 27 (E, F), and only few changes in the mitochondrion are evident (E, H). Arrows point towards parasites undergoing endodyogeny, ap=apical part, rop=rhoptries. Red frames indicate damaged parts of the mitochondrion, yellow frame shows a part of the mitochondrion with an intact matrix.

In non-treated parasites, the nucleus (nuc), and secretory organelles such as micronemes (mic) and dense granules (dg) are clearly visible. Host cell mitochondria (hmito) are often found in the close vicinity of the parasitophorous vacuole membrane. Tachyzoites contain one mitochondrion (mito) that forms an intracytoplasmic tubular network, of which are visible only the parts exposed on the surface of the section plane (Figure 7C). The matrix of the mitochondrion is electron-dense and contains numerous cristae.

After exposure of T. gondii-infected HFF to 0.5 μM of 21 for 24 h (Figure 8A–D), profound changes occurred most notably in the matrix of the tachyzoite mitochondrion, as indicated by the red frames. In many instances, the mitochondrial matrix was dissolved, with only remnants of cristae found within otherwise empty vacuoles. However, these changes were not evident in all domains of the mitochondrion, as can be seen in Figure 8B (framed in yellow). Mitochondrial alterations were less pronounced compared to similar modifications induced by coumarin-trithiolato diruthenium conjugates41 or by other complexes with no organic molecule appended on one of the bridging thiols.6 In addition, mitochondrial changes were not as readily identified upon treatment with compound 27, for which the impact on the mitochondrial matrix was evident only in about 20 % of tachyzoites (Figure 8E, H). Overall, the shape and size of the parasites was not extensively altered by any of the two compounds. Additionally, the fact that parasites undergoing endodyogeny were readily identified, especially in compound 27 treated cultures (Figure 8E, G), and larger vacuoles were seen in cultures treated with both compounds (Figure 8A, E, F), show that these BODIPY-diruthenium conjugates did not have a notable impact on intracellular proliferation when applied after the infection has already been established.

2.6 Fluorescence microscopy

To assess the potential use of the new conjugates BODIPY-diruthenium unit as cellular traceable probes and to identify possible key targets, fluorescence microscopy was performed on HFF treated with 20 μM of either carboxy BODIPY dye 10 or the corresponding BODIPY-diruthenium conjugate 20. The microtubules of HFF were labelled with mouse monoclonal anti-alpha-tubulin antibody (Red) and DAPI (Blue) was used for staining the nuclei.

Interestingly, the fluorescence intensity of conjugate 20 inside the cells was greater compared to that of the corresponding BODIPY dye 10 (Figure 9), although the latter exhibits a significantly higher fluorescence quantum yield (Table 2). These results suggest that the BODIPY dye 10 was better taken up by HFF when it was conjugated to the diruthenium complex, which might indicate a higher internalization and/or a retention of the dyad 20. A similar effect was reported for other hybrids metal complex-BODIPY dyes.30, 35 As demonstrated in the merged image and in accordance with previous studies,25a within HFF 20 showed a pattern of cytoplasmic, but not nuclear, localization. This is despite the cationic nature of the diruthenium moiety which can interact with anionic molecules like DNA. The images suggest a localization of conjugate 20 on the HFF microtubules. Additional experiments are needed to corroborate these results but are beyond the scope of the current study.

Details are in the caption following the image

Fluorescence microscopy (green) of HFF treated with 20 μM of the BODIPY dye 10, or the corresponding diruthenium conjugate 20 for 1 h at 37 °C. Cells were stained with monoclonal anti-alpha-tubulin antibody (red) and DAPI (blue).

3 Conclusions

A series of 15 new conjugates BODIPY-trithiolato-bridged dinuclear ruthenium(II)-arene complex were synthesized and fully characterized. The influence of the connecting bond (amide vs. ester), as well as that of the type and length of the spacer between the dyads’ units upon the photophysical and biological properties were evaluated.

The considered structural variations, as the connecting bond, the aliphatic spacer‘s length, the presence of an meso-aromatic ring in the BODIPY, had significant impact on both the conjugates’ fluorescence and measured antiparasitic properties, while they influenced in a lower extent the cytotoxicity (most dyads had negligible effect on HFF viability even at 1 μM). For example, when applied at 1 μM the ester conjugates with aliphatic spacers 1922 exhibited noteworthy higher T. gondii antiproliferative activity compared to the corresponding amide analogues 2326.

Despite an important fluorescence quenching due to the conjugation of the BODIPY to the diruthenium unit, the spectral properties of the dyad 20 (relative ΦF=17 %) allowed fluorescence microscopy visualization in HFF cells. Noteworthy, the intracellular fluorescence intensity was higher for conjugate 20 compared to that of the free BODIPY dye 10, suggesting a better internalization and/or retention of the diruthenium conjugate.

The antiparasitic activity of the conjugates (in terms of IC50 values against T. gondii) remained in the same range as that of the standard drug pyrimethamine. Thus, while these compounds could be valuable fluorescent probes for tracking their cellular targets, their potential as antiparasitic therapeutic agents is moderate. However, the present biological evaluation results must be considered with caution, as it cannot be excluded that the cellular uptake, accumulation, and localization of the BODIPY dyads could diverge from those of non-modified diruthenium complexes or of other reported conjugates. Incidentally, TEM also points out the parasite mitochondrion as a potential target, but the observed effects were less pronounced compared to those of previously described trithiolato-bridged dinuclear ruthenium(II)-arene compounds and diruthenium conjugates. Additional experiments on cancer cells and bacteria using these BODIPY-diruthenium dyads as intracellular tracking agents are considered.

Experimental Section

Chemistry

The chemistry experimental part, with full description of synthetic procedures and characterization data for all compounds, as well as the data corresponding to the crystal structures determination are presented in the Supporting Information.

Photophysical measurements

Instruments and methods: UV-Visible spectra were recorded on a Thermo Scientific Evolution 201 UV-Vis spectrophotometer, and the fluorescence emission spectra were recorded on an Agilent Cary Eclipse fluorescence spectrophotometer. The UV-Vis absorption spectra of compounds 536 were recorded in the 200–1100 nm range at r. t. using 1, 2, 5 and 10 μM solutions in CHCl3. Emission spectra were recorded in the 450–650 nm range after excitation at 450 nm at r. t. using the same solutions (1, 2, 5 and 10 μM solutions in CHCl3).

Determination of quantum yields: Relative quantum yields in CHCl3 at r. t. were calculated using equation (1) and rhodamine 6G (suitable for fluorescence, BioReagent, Sigma-Aldrich, Germany) (ΦF=0.75 in CHCl3) as standard.55
urn:x-wiley:14394227:media:cbic202200536:cbic202200536-math-0004(1)

where A is the absorbance at the excitation wavelength, F is the area under the emission curve, n is the refractive index of the solvent (at 20 °C) used in measurements (n=1.446 for CHCl3), and the subscripts s and x represent standard and unknown, respectively.

Stokes shifts were calculated using equation (2) as the difference between the values of maxima of the intense bands in the fluorescence and absorption spectra:
urn:x-wiley:14394227:media:cbic202200536:cbic202200536-math-0005(2)

Biological activity evaluation

In vitro activity assessment against T. gondii tachyzoites and HFF: During the in vitro experimental testing, compounds were protected from the light. All tissue culture media were purchased from Gibco-BRL, and biochemical agents from Sigma-Aldrich. Human foreskin fibroblasts (HFF) were purchased from ATCC, maintained in DMEM (Dulbecco's Modified Eagle's Medium) supplemented with 10 % fetal calf serum (FCS, Gibco-BRL, Waltham, MA, USA) and antibiotics as previously described.3e Transgenic T. gondii β-gal samples (expressing the β-galactosidase gene from Escherichia coli) were kindly provided by Prof. David Sibley (Washington University, St. Louis, MO, USA) and were in vitro maintained by serial passage in HFF as previously described.3e, 58

All the compounds were prepared as 1 mM stock solutions from powder in dimethyl sulfoxide (DMSO, Sigma, St. Louis, MO, USA). For in vitro activity and cytotoxicity assays, HFF were seeded in 96 well plates at 5×103/well and incubated at 37 °C with 5 % CO2 until reaching confluency.

To measure proliferation of T. gondii β-gal tachyzoites in presence or absence of compounds, the β-galactosidase colorimetric assay was performed as previously described.3e, 41, 42, 45a Briefly, experiments were conducted in 96 well plates, compounds were evaluated at 0.1 and 1 μM and added to HFFs prior to the infection with freshly isolated tachyzoites (1×103/well). Control untreated parasites received only media containing 0.01 or 0.1 % DMSO. All cultures were incubated for 3 days at 37 °C with 5 % CO2.45b At the end of the treatment, medium was removed, and cells were lysed with 0.05 % Triton X-100 in phosphate-buffered saline (PBS) and 5 mM chlorophenol red-β-D-galactopyranoside (CPRG; Roche Diagnostics, Rotkreuz, Switzerland), the substrate for β-galactosidase was added. Absorbance of released chlorophenol red was measured at 570 nm using an EnSpire® multimode plate reader (PerkinElmer, Inc., Waltham, MA, USA). The 100 % of proliferation was settled to the control untreated T. gondii β-gal-infected HFF.

IC50 determination on T. gondii β-gal was performed as previously described41, 42, 45a based on the β-galactosidase colorimetric assay and by applying a range of concentrations from 0.007 to 1 μM. IC50 values were calculated after the logit-log-transformation of relative growth and subsequent regression analysis.

The cytotoxicity of the compounds on non-infected HFF was assessed by the Resazurin reduction assay (alamarBlue) as previously reported.41, 42, 45a, 59 Briefly, confluent HFF monolayer cultures in 96 well plates were exposed to 0.1, 1 and 2.5 μM of each compound. Non-treated HFF as well as DMSO controls (0.01 %, 0.1 % and 0.25 %) were included. After 72 h of incubation at 37 °C/5 % CO2, the medium was removed, and plates were washed three times with PBS. Then, 200 μL of resazurin (0.01 g/L) were added to each well. Fluorescent resorufin were measured at excitation wavelength 530 nm and emission wavelength 590 nM on an EnSpire® multimode plate reader (PerkinElmer, Inc.). Fluorescence was measured at two time points: 0 min (T0) and after 5 hours (T5 h). After subtraction of T0 values from T5 h, mean fluorescence and the standard deviation are calculated from three (primary screening) or six (secondary screening) replicates. The 100 % viability was settled to control untreated HFF incubated in DMEM complemented medium containing 0.01 %, 0.1 % or 0.25 % DMSO.

Transmission electron microscopy (TEM): Transmission electron microscopy (TEM) was performed as previously described.57 In brief, confluent HFFs grown in T25 flasks were infected with 106 T. gondii ME49 tachyzoites (genotype II, kindly provided from Dr. Furio Spano, Istituto Superiore di Sanità, Roma, Italia) and 0.5 μM solutions of 21 or 27 were added at 24 h post-infection. After 24 and 48 h, medium from the flasks was discarded, cells were washed with a 0.1 M sodium cacodylate buffer (pH 7.3) and then fixed in 2 % glutaraldehyde in cacodylate buffer for 10 min at room temperature r. t. Fixed cultures were then gently scraped, transferred into Eppendorf tubes and fixed for another 2 h at r. t. Specimens were then washed in cacodylate buffer, post-fixed in 2 % osmium tetroxide in cacodylate buffer for 2 h at r. t. Samples were then washed with water, pre-stained in saturated uranyl acetate solution and stepwise dehydration in ethanol. They were then embedded in Epon 812-resin and processed for TEM as described.6 Specimens were viewed on a CM12 transmission electron microscope operating at 80 kV.

Fluorescence microscopy: Glass cover slips of 12 mm in diameter were placed in 24 well culture plate and sterilized in UV Stratalinker 1800 (Stratagene) for 40 min. Cover slips were then coated with HFFs (2×104 cells/mL) and incubated for 3 days at 37 °C with 5 % CO2. Afterwards, the culture medium was removed and replaced with fresh medium (1 mL/well) containing either (i) 0.5 % DMSO, or (ii) 20 μM of BODIPY dye 10, or (iii) 20 μM of ester conjugate 20, the treatment duration being 1 h at 37 °C/5 % CO2. After treatment, wells were washed (3 washes with sterile PBS), fixed with 2 % paraformaldehyde (PFA) in PBS for 20 min and permeabilized with 0.2 % Triton X-100 in PBS applied for 5 min. Unspecific binding sites were blocked using 3 % BSA in PBS for 2 h at r. t.. Glass coverslips were then incubated for 30 min with mouse monoclonal anti-alpha tubulin antibody (Sigma) diluted 1 : 200 in PBS containing 0.3 % BSA, then with a goat anti-mouse antibody conjugated to TRITC (tetramethylrhodamine isothiocyanate) (Sigma, 1 : 200), and after three washes in PBS, coverslips were mounted in Vectashield plus DAPI and viewed on a Nikon Eclipse E800 digital confocal fluorescence microscope. Images were acquired and processed with Openlab 5.5.2 software.

Supporting information

Primary cytotoxicity/efficacy screening of BODIPY compounds 518 and 3032 in non-infected HFF cultures and T. gondii β-gal tachyzoites cultured in HFF, crystal data and structure refinement for 16 and 30, chemistry experimental part with the full description of experimental procedures, stability of compounds 24 and 1521 in DMSO-d6 can be found in the Supporting Information. Deposition Numbers2118529 (for 16) and 2118530 (for 30) contain the supplementary crystallographic data for this paper. These data are provided free of charge by the joint Cambridge Crystallographic Data Centre and Fachinformationszentrum Karlsruhe Access Structures service www.ccdc.cam.ac.uk/structures.

Acknowledgements

This work was financially supported by the Swiss Science National Foundation (SNF, Sinergia project CRSII5-173718 and project 310030_184662) (E. P., O. D., J. F., A. H., G. B., N. A.). The Synergy diffractometer was partially funded by the SNF within the R'Equip program (project 206021_177033). Dr. Jürg Hauser and the X-ray crystal structure determination service unit, Department of Chemistry, Biochemistry and Pharmaceutical Sciences, University of Bern, are acknowledged for measuring, solving, refining, and summarizing the structures of BODIPY compounds 16 and 30.

    Conflict of interest

    The authors declare no conflict of interest.

    Data Availability Statement

    The data that support the findings of this study are available in the supplementary material of this article.