Biomedical Properties of a Series of Ruthenium-N-Heterocyclic Carbene Complexes Based on Oxidant Activity In Vitro and Assessment In Vivo of Biosafety in Zebrafish Embryos

Abstract

N-Heterocyclic carbene (NHC) ligands have attracted great interest over the last decade for their use in the design of homogenous catalysts. NHC-based metal complexes have interesting potential biomedical applications, such as in antimicrobial and cancer therapy, which are beginning to be explored more fully. We have studied here the oxidant activities of a series of Ru(II) complexes in vitro and zebrafish (Danio rerio) have been used as a model in vivo to investigate and characterize the toxicity of some of these compounds. Dual behavior was observed for the NHC-based complexes as they behaved as antioxidants at low concentrations but showed pro-oxidant capacity at higher concentrations. Zebrafish embryos were exposed to Ru(II) complexes under several different conditions (0 or 24 h postfertilization, with or without the chorion) and various parameters, such as viability, edema, heart rate, blood coagulation, pigmentation, scoliosis, malformation, and hatching, were tested. In general, zebrafish embryos were not harmed by exposure to Ru(II) complexes whatever the experimental conditions. Several toxicity profiles were observed depending upon the chemical structure of the compound in question. Their characteristics as pro-oxidant and/or antioxidant agents together with their biosafety may point to their having biomedical applications as antitumoral or neuroprotective drugs.

Introduction

Since Rosenberg's findings concerning the antineoplastic properties of cisplatin,^1,2 many efforts have been made to develop clinically improved analogs. Apart from carboplatin and oxaliplatin, now used worldwide for solid tumor therapy, very few platinum-based drugs have been investigated to date in clinical trials. In an effort to develop new efficient molecules with biomedical activity, novel complexes with other metals such as silver, gold, ruthenium, gallium, and titanium are attracting increased attention.^3,4

N-Heterocyclic carbene (NHC)-based metal complexes are being proposed as potent antimicrobial [Ag(I)-NHC],^5–7 radiotherapic [¹⁰⁵Rh(III)-NHC],^8,9 and antitumoral [Au(I)-NHC] agents.^10–13 Within the same context, Ray and coworkers have described the antimicrobial activity of Au(I) and Ag(I)-NHC complexes (Fig. 1A), and the potential antitumoral activity of two trans-dichloro Pd(II)-NHC complexes (Fig. 1B, C).¹⁴ The cytotoxicity¹⁵ and antimicrobial activity¹⁶ of other Au(I)-NHC compounds have been studied by Horvath et al. and Ozdemir et al., respectively.

FIG. 1.

Structures of antimicrobial (A) and anticancer (B, C) complexes described by Ray and coworkers.¹⁴

Although the literature shows an increasing biomedical interest in NHC-metal complexes, only a few reports regarding their toxicity are currently available and these limit themselves to analyzing their potential to kill tumoral cells or pathogens^6,14,17; according to our knowledge, no animal models have been used for this purpose. As far as ruthenium compounds are concerned, it has been shown¹⁸ that exposure to a chiral organoruthenium half-sandwich compound gives rise to differential zebrafish (Danio rerio) embryo phenotypes according to which enantiomer is used. The effects ranged from no symptoms to pathological effects (decrease in head size, heads without eyes, stunted and crooked tail, or enlarged, misshapen yolk). The popularity of zebrafish as an animal model in toxicology as well as in developmental biology is due to the fact that they are a vertebrate species, their genome is sequenced, and their embryos are mainly fully transparent, so that any change can be observed directly through an optical microscope.¹⁹ Indeed, Eimon and Rubinstein²⁰ have recently reviewed several assays conducted with zebrafish, including cardiotoxicity, ototoxicity, seizure liability, developmental toxicity, and gastrointestinal motility, the results of which suggest that such assays are acceptably predictive as far as human drugs are concerned, ranging from “sufficient” (65–75% predictability) to “good” (75–85% predictability).

In this work we have studied in vitro the antioxidant/pro-oxidant activity of a series of ruthenium(II) arene complexes with different NHC ligands (NHC = imidazolin-2-ylidene, imidazolin-4-ylidene, and pyrazolin-3-ylidene). For comparative purposes we also studied the oxidant activity of two ruthenium precursors, [RuCl₂(p-cymene)]₂ and RuCl₂(dmso)₄. We assessed the biosafety of three complexes selected from the series in zebrafish embryos, an established animal model for studying toxicity.

Materials and Methods

Preparation of the ruthenium compounds

[RuCl₂(p-cymene)]₂,²¹ RuCl₂(dmso)₄,²² and complex 1²³ were prepared according to the methods described in the literature (Fig. 2). NHC-based complexes 2a, 2b, 3, and 4 were prepared as previously reported by us.²⁴

FIG. 2.

Structures of the studied Ru(II)-N-heterocyclic carbene compounds 1–6.

Pro-oxidant assay

The radical cation of 2 [2′-azinobis (3-ethylbenzothiazoline 6-sulfonate)] (ABTS; Sigma, Madrid, Spain) is formed by the interaction of ABTS with the ferrylmyoglobin radical species, generated by the activation of metmyoglobin with a pro-oxidant compound. We conducted the pro-oxidant assay in 96-well microtiter plates, in the presence and absence of the ruthenium compounds 1–6, at 1, 10, and 100 μg/mL. We assayed 3.5 μg/mL of myoglobin (Sigma), used as the source of metmyoglobin, which was oxidized by H₂O₂ (2 mM) in the presence of ABTS (1.8 mM) and dissolved in 0.05 M phosphate-citrate buffer solution. ABTS was then oxidized by the ferrylmyoglobin radical generated, thus producing the ABTS radical cation. After 3 min incubation at room temperature the end point of absorbance at 405 nm was measured using a Benchmark Plus spectrophotometer (Biorad, Hercules, CA). The pro-oxidant activity of the compound in question was calculated as its capacity to form the ABTS radical cation, the results being compared with those obtained when using 2 mM of H₂O₂.

Trolox equivalent antioxidant capacity assay

In the Trolox equivalent antioxidant capacity (TEAC) assay, first reported by Miller and Rice-Evans²⁵ and later modified by Re et al.,²⁶ Trolox (6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid) serves as a standard or control for the antioxidant capacity. The TEAC assay measures the ability of a compound to reduce the ABTS radical cation, giving useful information regarding its antioxidant activity. As described earlier, the ABTS radical cation is formed by interaction of ABTS with the ferrylmyoglobin radical species, generated by the activation of metmyoglobin with H₂O₂. It has a characteristic long-wavelength absorption spectrum maximum at 405 nm. A standard curve for Trolox was constructed by plotting the absorbance of different concentrations (in μM) of Trolox at 405 nm. Antioxidants suppress the production of the ABTS radical cation in a concentration-dependent way and the color of the solution decreases proportionally. The antioxidant capacity of the compound in question is obtained by comparison with the concentration of Trolox, using the standard curve, in μM Trolox equivalent (TE).

We performed the TEAC assay in 96-well microtiter plates with Trolox standard concentrations or ruthenium compounds at 1, 10, and 100 ng/mL. We assayed 3.5 μg/mL of myoglobin (Sigma), used as the source of metmyoglobin, which was oxidized by H₂O₂ (2 mM) in the presence of ABTS (1.8 mM) dissolved in 0.05 M phosphate-citrate buffer solution. ABTS was then oxidized by the ferrylmyoglobin radical thus generated, producing the radical cation of ABTS. After 3 min incubation at room temperature, the end point of absorbance at 405 nm was measured using a Benchmark Plus spectrophotometer (Biorad).

Zebrafish egg production and treatments

Fertilized eggs were obtained by natural mating of adult zebrafish (AB line) in our laboratory, carried out according to the Zebrafish book.²⁷ A total of 4–5 pairs were set up for each mating and, on average, 200–250 embryos per pair were generated. Eggs were collected within 1 h of spawning, rinsed, and placed into a clean Petri dish, where they were kept at 28.5°C. Twenty-four hours after spawning they were rinsed in dilution water (CaCl₂ · 2H₂O, 0.294 g/L; MgSO₄ · 7H₂O, 0.12325 g/L; NaHCO₃, 0.06475 g/L; and KCl, 0.00575 g/L) at pH 7.8 ± 0.2 according to Organization for Economic and Co-operation Development (OECD) TG 212, placed into a clean Petri dish, and kept at 28.5°C. Twenty-four hours after spawning, only the fertilized eggs (a minimum of 10 embryos) were exposed to 24-well plates containing either the compound vehicle (dimethyl sulfoxide [DMSO], 0.1%) or different concentrations of the compounds being assayed (ruthenium compounds or t-butyl hydroperoxide [TBH], ranging from 0.001 to 1.000 μg/mL). Apart from this, to complete the toxicology study we used two reference toxic compounds: sodium dodecyl sulfate (SDS) and copper (II) sulfate (CuSO₄). The assays were performed using the same procedure as described earlier (with DMSO, 0.1%) and using a range of concentrations between 0.01 and 100 μg/mL. Strict precautions were taken to prevent contamination during manipulation. Each experiment was performed in triplicate.

Early-stage testing to assess teratological effects of the compounds

To observe any possible teratological effect, the fish-embryo test was applied and endpoints for the assessment of acute chemical toxicity such as coagulation of embryos, alteration of heart rate, edema, blood coagulation, weak pigmentation, scoliosis, malformations, and hatching success were recorded. For this assay, 30 min postfertilization embryos were incubated for 48 h with various compounds (ruthenium compounds, TBH, and reference products) or solvent (control), at which time the endpoints were observed. Trolox was used as a positive antioxidant control.

Zebrafish embryo test: embryotoxicity

The development of embryos from blastula to early larval stages was monitored at 24, 48, 72, 96, and 120 h postfertilization (hpf ) throughout the experiment. For the assays, 24 hpf embryos were incubated with various compounds (ruthenium compounds, TBH, and reference products) and solvent (control) for 48 h. At the time of observation, dead embryos were removed from the exposure chambers to prevent contamination of the surviving embryos. Endpoints used for assessing the effects of the compounds included egg and embryo mortality, alteration of heart rate, edema, blood coagulation, weak pigmentation, scoliosis, malformations, and hatching success. The morphological effects were observed and recorded among the larvae from both control and treated groups using a stereomicroscope (7–115×) connected to a camera.

Toxicity of the ruthenium compounds toward dechorionated embryos

The chorions of 24-h-old embryos were disrupted mechanically using two very fine pairs of forceps and experiments were conducted as described earlier.

Results

Antioxidant/pro-oxidant capacities

The oxidant/pro-oxidant capacities of a series of NHC-based Ru(II)-arene complexes (NHC = imidazolin-2-ylidene, imidazolin-4-ylidene, and pyrazolin-3-ylidene [1–4]) as well as [RuCl₂(p-cymene)]₂ and RuCl₂(dmso)₄ (5 and 6, respectively) (Fig. 2) were assessed using the TEAC assay^25,26 (see Materials and Methods for details). We first studied the pro-oxidant capacity of the ruthenium complexes 1–6 by dissolving them in DMSO at three different concentrations (1, 10, and 100 μg/mL). The NHC-based complexes 1–4 showed a significant pro-oxidant activity at the highest concentration (100 μg/mL), whereas the metal precursors 5 and 6 did not (Fig. 3). The pro-oxidant potential of complexes 2b and 3 was the highest and showed similar values throughout the entire experimental range from 1 to 100 μg/mL, followed by 2a, 4, and 1. These results suggested that complexes 2b and 3 might be the most interesting molecules of the series in terms of cytotoxic activity within the range of 1–100 μg/mL. Pro-oxidant activity was dose-dependent in all cases.

FIG. 3.

Pro-oxidant activity in vitro of the ruthenium compounds in the range 1–100 μg/mL. The results are shown as the oxidation of the ABTS radical cation caused by the compounds compared with oxidation by 2 mM H₂O₂. The data are the mean ± SD of two independent experiments. The statistical significance of the differences was examined with Student's t-test (*p < 0.05 compared with 5; ^#p < 0.05 compared with 6). ABTS, 2′-azinobis (3-ethylbenzothiazoline 6-sulfonate); SD, standard deviation.

With respect to antioxidant activity, complex 6 [RuCl₂(dmso)₄] was found to be the most effective, showing an antioxidant level close to 80 μM TE at the lowest dose assayed (1 ng/mL) (Fig. 4). In contrast, complex 2a had the lowest antioxidant potential, whereas there was little difference in antioxidant potential between the rest. The observation that complex 6 presented the highest antioxidant activity at the lowest dose suggested good possibilities of medical applications for this compound. Antioxidant activity was dose-dependent with all the compounds.

FIG. 4.

Antioxidant activity in vitro of the ruthenium compounds in the range of 1–100 ng/mL. The antioxidant activity of the compounds is expressed as TE antioxidant capacity values (μM TE) measuring the capacity of the compounds to prevent the formation of the ABTS radical cation by 2 mM H₂O₂ and quantified using the Trolox standard curve (see Materials and Methods). The data are the mean ± SD of two independent experiments. The statistical significance of the differences was examined with Student's t-test (*p < 0.05 compared with 5; ^#p < 0.05 compared with 6). TE, Trolox equivalent.

Embryotoxicity of the ruthenium compounds

To demonstrate the utility of these organometallic compounds for experiments in whole organisms we conducted phenotypic experiments in zebrafish embryos following the OECD Test Guideline 212 (with minor changes). Complexes 3, 4, and 6 were chosen for the biosafety tests because of their antioxidant/pro-oxidant capacities as revealed in the in vitro assays (compound 3: the most pro-oxidant; compound 4: pro-oxidant at high concentrations, antioxidant at low doses; and compound 6, the most antioxidant agent). In embryotoxicity studies, the embryos were always exposed to the compounds at 24 hpf, and although the assays usually lasted for 48 h posttreatment (hpt), depending upon the test design, they could be prolonged to 96 hpt to include more sublethal endpoints such as hatching success and larval abnormalities. To study the effects of the different compounds, a preliminary set of experiments were carried out at 24 hpf in zebrafish embryos exposed in 24-well plates containing either vehicle compounds or different concentrations of treatments (ruthenium compounds or TBH, a known pro-oxidant agent) in the range of 0.001–100 μg/mL. At these concentrations we observed neither mortality nor detectable abnormalities in any of the treated embryos. Nevertheless, a delay in egg development was observed in the assays with compounds 3 and 4 at 100 μg/mL, at which concentration the embryos remained within the chorion after 48 hpt (Fig. 5). No other symptoms related to the treatment, such as teratogenic malformations, tail defects, edema, or weak pigmentation, were seen, indicating the high safety of these compounds.

FIG. 5.

Representative example of an early control larva (72 hpf ) (A) and another treated at 24 hpf + 48 hpt with 100 μg/mL of compound 3 (B). Compound 3 did not cause any abnormality or death but did result in the embryos' not hatching. Each compound was tested in at least two independent experiments performed in triplicate. hpf, hours postfertilization; hpt, hours posttreatment.

On the basis of the results of these initial assays, the dependence of embryotoxicity on the concentration of ruthenium compounds 3, 4, and 6 was analyzed in comparison to the three reference compounds SDS, CuSO₄, and TBH, all regularly used for toxicological studies (Fig. 6).^18,28 SDS is an anionic surfactant classified as a membrane-damaging chemical, used as a reference chemical in the ACuteTox project,²⁹ whereas CuSO₄ is a powerful oxidizing agent and irritant to mucous membranes. Their viability was not diminished when embryos were incubated with ruthenium compounds up to the dose of 100 μg/mL, whereas TBH started to become toxic at concentrations above 100 μg/mL. In fact, compounds 3 and 4 presented a LC₅₀ (defined as the concentration of a substance that causes the death of 50% of a group of test animals) at lower concentrations than TBH, showing that these compounds are more toxic than this latter pro-oxidant agent (Table 1). Compound 6, on the other hand, caused no loss of viability at any concentration from 0.1 to 1000 μg/mL, indicating that it is the safest compound of the series. This result accords with the in vitro results for pro-oxidant activity (Fig. 3). To compare these results with those of other toxicological substances, CuSO₄ and SDS were evaluated (Fig. 6 and Table 1). The toxicity range of the compounds evaluated was the following (from higher to lower toxicity): SDS, CuSO₄, compound 3, compound 4, TBH, and compound 6. The toxicological parameters of the two ruthenium compounds showing significant toxicity (3 and 4) were very similar to each other, indicating that structural changes in the NHC complex did not affect embryos viability. Moreover, this agrees with the results obtained from the analysis of their pro-oxidant activity, where the closest concentration to their LC₅₀ (100 μg/mL) presented similar pro-oxidant activities.

FIG. 6.

Percentage of zebrafish-embryo viability after exposure to different concentrations of compounds. The embryos were exposed at 24 hpf + 48 hpt to ruthenium compounds or reference substances. Values are the mean ±SD of two independent experiments performed in triplicate.

Table 1.

Toxicological Parameters of the Evaluated Compounds

	SDS	CuSO₄	TBH	Compound 3	Compound 4	Compound 6
LC₅₀^a (μg/mL)	6.3	28.3	384.6	195.5	217.4	>1000
LC₅^b (μg/mL)	4.2	20.8	141.6	99.2	93.0	>1000
LC₉₅^c (μg/mL)	9.6	39.9	1270.5	345.5	668.1	>1000
LOEC^d (μg/mL)	2.0	10.0	400.0	100.0	100.0	1000.0
NOEC^e (μg/mL)	1.0	1.0	200.0	10.0	10.0	400.0

Dose of a chemical that kills 50% of a sample population.

Dose of a chemical that kills 5% of a sample population.

Dose of a chemical that kills 95% of a sample population.

The lowest tested concentration of a substance with a significant effect compared with the control.

The concentration immediately below LOEC. These data were obtained in embryos exposed to the compounds at 24 hpf and 48 hpt.

SDS, sodium dodecyl sulfate; CuSO₄, copper (II) sulfate; TBH, t-butyl hydroperoxide; hpf, hours postfertilization; hpt, hours posttreatment.

Morphological changes in embryos incubated with ruthenium compounds

Once the viability curves had been obtained and the toxicological parameters defined, we studied the induction of morphological changes in embryos. Embryos were incubated with three concentrations of all six compounds (10, 100, and 1000 μg/mL), corresponding to low, mid, and high toxicity (Fig. 7). At 10 μg/mL, the embryos incubated with ruthenium compounds or with TBH were in no way different from the controls, having emerged from the chorion at the early larva stage (72 hpf = 24 hpf + 48 hpt) and clearly showing a well-developed body. At the same concentration, however, the embryos treated with CuSO₄ remained inside the chorion, whereas the embryos treated with SDS had died and turned opaque because of coagulation of their organs. At a concentration of 100 μg/mL the embryos treated with compound 6 or TBH were no different in appearance from the controls, whereas those incubated with SDS or CuSO₄ were dead. The embryos treated with ruthenium compounds 3 and 4 remained within the chorion, suggesting either a delay in egg development or some sort of defense mechanism against the toxic substance in the medium. When the maximum concentration was used (1000 μg/mL), all of the embryos died, except the controls and those treated with compound 6, which remained within the chorion, a similar behavior to that observed in those incubated with 100 μg/mL of compounds 3 and 4.

FIG. 7.

Representative images of zebrafish embryos after various treatments. The embryos were exposed at 24 hpf + 48 hpt to different concentrations of compounds and any morphological change was duly recorded.

To discover whether the embryos remaining within the chorion was the result of a delay in egg development or a stage previous to the death, we undertook additional experiments. Embryos incubated with 100 μg/mL of TBH or ruthenium compounds 3, 4, and 6 were studied at four different times: 24 hpf + 24 hpt, 24 hpf + 48 hpt (just as the previous results shown in Fig. 5), 24 hpf + 72 hpt, and 24 hpf + 96 hpt (Fig. 8). Embryos treated with TBH or compound 6 did not die and had emerged from the chorion by 48 hpt, in a similar way to the controls, indicating normal egg development. When compounds 3 and 4 were used, both produced a similar effect: the embryos remained inside the chorion until the end of the experiment, but then started to die at 72 hpt. Compound 3 showed minimally higher toxicity than 4, coinciding with the toxicological parameters set out in Table 1 and with the pro-oxidant activities in Figure 3.

FIG. 8.

Percentage of viability and cumulative hatching of zebrafish embryos after exposure to different compounds (at 100 μg/mL) and recorded at different times: 24 hpf +24 hpt, 24 hpf + 48 hpt, 24 hpf + 72 hpt, and 24 hpf + 96 hpt.

Cardiotoxicity of ruthenium compounds

The influence of ruthenium compounds upon the heart rate of zebrafish embryos was compared with the reference compounds (Fig. 9). The results indicate that the most toxic compound was SDS, followed by CuSO₄, thus confirming the results of the viability assays (Fig. 6 and Table 1). The safest substance assayed was once again compound 6, which caused no alteration except at the highest concentration (1000 μg/mL), where a minor decrease in heart rate was detected. A comparison of compounds 3 and 4 revealed that the former was more cardiotoxic: 3 caused a significant decrease in heart rate from 100 μg/mL, whereas more than 200 μg/mL of 4 was required to produce a similar effect. These results agree with the pro-oxidant activity shown in Figure 3, with the toxicological data in Figure 6 and Table 1, and with the cumulative-hatching results set out in Figure 8.

FIG. 9.

Total heart rate in zebrafish embryos treated with increasing doses of ruthenium and reference compounds. The heart rate of the embryos was recorded at 24 hpf + 48 hpt after exposure to the compounds and the average was plotted. The black horizontal line represents the average value of the heart rate of the controls. The asterisks represent the significant differences between the treated groups compared with the controls (*p < 0.01, Student's t-test). Values are the mean ± SD for two independent experiments performed in triplicate.

Toxicity of ruthenium compounds in dechorionated embryos

To discover whether the effects of the compounds were due to direct toxicity or a differential permeability through the chorion, 24 hpf dechorionated embryos were exposed to a single concentration (100 μg/mL) of the compounds and several endpoints were recorded (Table 2: 24 hpf + 48 hpt − chorion), which were compared with previous results (summarized in Table 2: 24 hpf + 48 hpt + chorion). Once again compound 3 was the most toxic of them all, inducing mortality in 50% of the embryos and provoking sublethal effects such as edema (7%), reduced heart rate (67%), blood coagulation (20%), weak pigmentation (7%), malformations (67%), and scoliosis (embryo's spine is curved) in all of them. The toxicity of compound 4 was similar to that of compound 3, although the survival index was higher (93%) and no malformations were observed. Compound 6 was again the safest, causing neither generalized mortality nor sublethal or teratological effects. Its toxicity profile was also similar in the presence or absence of the chorion. In summary, compounds 3 and 4 proved more toxic to dechorionated embryos, although the toxicity profiles themselves were the same: compound 3 was more toxic than 4 and 6, the latter being the least toxic of the three. The toxicity of compounds was in general quite low (LC₅₀ < 100 μg/mL), showing that even without the barrier of the chorion all three were safe.

Table 2.

Types and Percentages of Abnormalities Observed Among Zebrafish Embryos Exposed to Compounds Under Different Experimental Conditions

Conditions	Parameters studied	Control	Compound 3	Compound 4	Compound 6	TBH	Trolox
24 hpf + 48 hpt + chorion	n	26	18	18	18	27	—
	Viability (%)	100	100	100	100	100	nd
	Sublethal developmental and teratogenicity endpoints (%)
	Edema	0	0	0	0	0	nd
	Reduced heart rate	0	67	6	0	22	nd
	Blood coagulation (clots)	0	0	0	0	0	nd
	Weak pigmentation	0	0	0	0	0	nd
	Scoliosis	0	0	0	0	0	nd
	Malformation	0	0	0	0	0	nd
	Hatching (72 hpf ) (%)	100	0	0	100	100	nd
24 hpf + 48 hpt − chorion	n	9	15	28	29	—	—
	Viability (%)	100	50	93	97	nd	nd
	Sublethal developmental and teratogenicity endpoints (%)
	Edema	0	7	0	7	nd	nd
	Reduced heart rate	0	67	100	3	nd	nd
	Blood coagulation (clots)	0	20	0	0	nd	nd
	Weak pigmentation	0	7	25	3	nd	nd
	Scoliosis	0	100	100	0	nd	nd
	Malformation	0	67	0	3	nd	nd
	Hatching (72 hpf ) (%)	—	—	—	—	—	—
0 hpf + 48 hpt + chorion	n	46	36	57	60	58	29
	Viability (%)	100	60	95	100	97	97
	Sublethal developmental and teratogenicity endpoints (%)
	Edema	0	8	2	0	0	0
	Reduced heart rate	0	100	100	0	0	0
	Blood coagulation (clots)	0	0	0	0	0	0
	Weak pigmentation	0	8	0	8	2	17
	Scoliosis	0	0	0	0	0	0
	Malformation	0	25	4	0	0	0
	Hatching (72 hpf )^a (%)	88	0	0	67	60	nd

Embryos were exposed for 48 h and then the compounds were removed.

n, number of embryos used; nd, not determined.

Early-stage testing to assess the teratological effects of the compounds

Essential information in toxicology studies with new chemical entities is provided by teratogenic tests, in which the drug is applied at a fairly early stage, such as around the time of gastrulation. Thus we replicated fish-embryo test assays in which embryos were treated at 30 min postfertilization (Table 2: 0 hpf + 48 hpt + chorion) to observe any differences from treatments administered at 24 hpf (Table 2: 24 hpf + 48 hpt + chorion). All the treatments were administered at 100 μg/mL. In general, the results were very similar to those found in the previous embryotoxicity studies, although compound 3 resulted in embryo mortality increasing to 40% in this teratogenic study. Further, with compound 3 we detected malformations in 25% of the embryos and edema and weak pigmentation in 8%. Compound 4 had a lower teratogenic effect than 3, giving rise only to a reduced heart rate, thus showing its higher safety profile. The teratogenic assays with 6 produced very similar results to the embryotoxicity tests: we could detect only a weakening of pigmentation in 8% of the embryos in this latter assay. As an internal control we used a pro-oxidant and an antioxidant agent. TBH (pro-oxidant agent) did not induce generalized sublethal or teratogenic phenomena, indicating low toxicity: only 2% of the embryos showed weak pigmentation and 60% had emerged from the chorion by 72 hpf (0 hpf + 48 hpt + 24 h without treatment). Trolox (water-soluble derivative of vitamin E) proved not to be toxic for the embryo either: 97% of the treated embryos reached viability, although 17% presented weak pigmentation. In summary, the toxicity profiles of the compounds were kept the same as in the embryotoxicity tests. Compound 3 was more toxic than 4 and 6, the latter proving to be the least toxic of the three. The toxicity profiles of the compounds were in general fairly low (LC₅₀ < 100 μg/mL).

Discussion

We have assessed the biomedical properties and biosafety of a series of ruthenium-NHC complexes: (i) antioxidant/pro-oxidant capacity (in vitro); (ii) embryotoxicity; (iii) chorion function as a barrier for the uptake of the compound in embryotoxicity; and (iv) teratological effects (early-stage test).

Our in vitro results showed that the ruthenium complexes assayed act as either antioxidants or pro-oxidants at certain doses, suggesting their possible application as antitumoral or neuroprotective agents, for instance. In harmony with our observations, it has recently been demonstrated that a series of ruthenium compounds present a similar dual behavior showing genotoxic (pro-oxidant) effects at higher doses (200 μM), and antioxidant potential at lower doses (0.1–10 μM).³⁰ According to these results, the choice of a specific concentration must determine the balance between the two faces of their oxidative status. Although high doses may be useful to sensitize or kill tumoral cells, lower doses, with antioxidant power, may eventually be used to treat or prevent neuropathological conditions associated with aging.

In summary, our in vitro studies suggest that the most active complexes as far as cytotoxicity is concerned are 2b and 3, which are the most potent candidates of the series for antitumoral agents, and complex 6 displayed significant antioxidant potential, which might be used for neuroprotective strategies. Although the low pro-oxidant capacity of these complexes at low doses points to their low toxicity, and hence low side effects, it was essential to carry out preclinical and clinical studies to double-check their harmlessness. Therefore, we chose to study the embryotoxicity of these complexes using zebrafish embryo models. For further studies, compounds 3 (the most pro-oxidant), 4 (intermediately oxidative), and 6 (the most antioxidant) were assessed and compared with reference toxics (SDS and CuSO₄), a commercial pro-oxidant (TBH), and a known antioxidant (Trolox). Viability studies showed that compound 6 was the safest of the series, followed by TBH and then compounds 4 and 3. SDS and CuSO₄ were the most highly toxic. The toxicological parameters set out in Table 1 corroborate these results and agree with the studies into their oxidative status.

In morphological studies, the embryos treated with the ruthenium compounds did not present side effects, although at high doses those treated with 3 and 4 remained inside the chorion, suggesting either a delay in egg development or a defense mechanism against the toxic substance in the medium. To try to resolve this question the behavior of the embryos was monitored until 96 hpt (120 hpf ). Those treated with TBH or compound 6 did not die and had emerged from the chorion by 48 hpt but those treated with compounds 3 and 4 remained within the chorion until the end of the experiment and started showing signs of morbidity by 72 hpt.

One of the most important parameters to explore in determining the toxicity of a compound is its capacity to induce deleterious effects to the heart, and because of its morphological characteristics the zebrafish embryo is an optimum model for this, allowing us not only to measure heart rate but also to study alterations to heart rhythm, such as tachycardia and bradycardia, or anomalies in its development, such as pericarditis, atrophy, and hypertrophy. All three compounds produced good results in the heart-rate study, showing toxicity with doses over 100 μg/mL: compound 6 being the least toxic, followed by 4 and 3.

In an attempt to clarify the role of the chorion in the embryotoxicity of ruthenium compounds (as a possible barrier for the compounds), 24 hpf dechorionated embryos were exposed to a single concentration (100 μg/mL) and several endpoints were recorded. The toxicity profile of the compounds was the same as in the previous embryotoxicity tests, compound 3 being more toxic than 4 and 6, the latter being the least toxic. The toxicity of the compounds increased, above all that of 3, the LC₅₀ of which fell from ∼200 μg/mL with the chorion in place to 100 μg/mL in its absence. Further, sublethal and teratogenic endpoints increased; phenomena such as edema, blood coagulation, weak pigmentation, scoliosis, and malformation were detected when the chorion was removed. Finally, teratological studies were carried out by administrating the compound 30 min postfertilization and the results showed once again that the toxicological profiles of the compounds remained the same as those of the embryotoxicity test.

In summary, the results of this study clearly reveal that ruthenium complexes are safe, showing LC₅₀ ≥ 100 μg/mL in every experimental parameter studied (according to OECD TG 212, substances responsible for LC₅₀ > 100 μg/mL are safe and can be considered nontoxic). Moreover, the three ruthenium complexes, with different chemical structures, present different toxicity profiles. In conclusion, the oxidant activity of the ruthenium complexes studied in vitro correlates well with the profiles of the toxicological parameters studied in zebrafish embryos, which may be related to their structures.

Footnotes

Acknowledgments

This work was funded by Neuron BPh and supported by funds from the Torres Quevedo Program (J.M.A., M.C.R., and J.S.B). The authors thank Professor Fernando Valdivieso for his continuous encouragement and help. The authors from the Universitat Jaume I gratefully acknowledge financial support from the MEC of Spain (CTQ2008-04460) and Bancaixa (P1.1A2008-02), and the Juan de la Cierva program (to M.P.).

Disclosure Statement

The authors from Neuron BPh have served as paid researchers for the company.

References

Rosenberg

. Platinum complexes for treatment of cancer. Interdiscip Sci Rev, 1978; 3:134–147.

Rosenberg

, Vancamp

, Trosko

, Mansour

. Platinum compounds—a new class of potent antitumour agents. Nature, 1969; 222:385–386.

Jakupec

, Galanski

, Arion

, Hartinger

, Keppler

. Antitumour metal compounds: more than theme and variations. Dalton Trans, 2008; 2:183–194.

Clarke

, Zhu

, Frasca

. Non-platinum chemotherapeutic metallopharmaceuticals. Chem Rev, 1999; 99:2511–2533.

Kascatan-Nebioglu

, Melaiye

, Hindi

, Durmus

, Panzner

, Hogue

et al. Synthesis from caffeine of a mixed N-heterocyclic carbene-silver acetate complex active against resistant respiratory pathogens. J Med Chem, 2006; 49:6811–6818.

Kascatan-Nebioglu

, Panzner

, Tessier

, Cannon

, Youngs

. N-Heterocyclic carbene-silver complexes: a new class of antibiotics. Coord Chem Rev, 2007; 251:884–895.

Melaiye

, Sun

, Hindi

, Milsted

, Ely

, Reneker

et al. Silver(I)-imidazole cyclophane gem-diol complexes encapsulated by electrospun tecophilic nanofibers: formation of nanosilver particles and antimicrobial activity. J Am Chem Soc, 2005; 127:2285–2291.

Quezada

, Garrison

, Panzner

, Tessier

, Youngs

. The potential use of rhodium N-heterocyclic carbene complexes as radiopharmaceuticals: the transfer of a carbene from Ag(I) to RhCl₃ · 3H₂O. Organometallics, 2004; 23:4846–4848.

Garrison

, Tessier

, Youngs

. Synthesis and crystallographic characterization of multi-donor N-heterocyclic carbene chelating ligands and their silver complexes: potential use in pharmaceuticals. J Organomet Chem, 2005; 690:6008–6020.

10.

Barnard

, Baker

, Berners-Price

, Day

. Mitochondrial permeability transition induced by dinuclear gold(I)-carbene complexes: potential new antimitochondrial antitumour agents. J Inorg Biochem, 2004; 98:1642–1647.

11.

Baker

, Barnard

, Berners-Price

, Brayshaw

, Hickey

, Skelton

et al. Cationic, linear Au(I) N-heterocyclic carbene complexes: synthesis, structure and anti-mitochondrial activity. Dalton Trans, 2006; 30:3708–3715.

12.

Hickey

, Ruhayel

, Barnard

, Baker

, Berners-Price

, Filipovska

. Mitochondria-targeted chemotherapeutics: the rational design of gold(I) N-heterocyclic carbene complexes that are selectively toxic to cancer cells and target protein selenols in preference to thiols. J Am Chem Soc, 2008; 130:12570–12571.

13.

Barnard

, Wedlock

, Baker

, Berners-Price

, Joyce

, Skelton

et al. Luminescence studies of the intracellular distribution of a dinuclear gold(I) N-heterocyclic carbene complex. Angew Chem Int Ed, 2006; 45:5966–5970.

14.

Ray

, Mohan

, Singh

, Samantaray

, Shaikh

, Panda

et al. Anticancer and antimicrobial metallopharmaceutical agents based on palladium, gold, and silver N-heterocyclic carbene complexes. J Am Chem Soc, 2007; 129:15042–15053.

15.

Horvath

UEI

, Bentivoglio

, Hummel

, Schottenberger

, Wurst

, Nell

et al. A cytotoxic bis(carbene) gold(I) complex of ferrocenyl complexes: synthesis and structural characterisation. New J Chem, 2008; 32:533–539.

16.

Ozdemir

, Denizci

, Ozturk

, Cetinkaya

. Synthetic and antimicrobial studies on new gold(I) complexes of imidazolidin-2-ylidenes. Appl Organomet Chem, 2004; 18:318–322.

17.

Brabec

, Novakova

. DNA binding mode of ruthenium complexes and relationship to tumor cell toxicity. Drug Resist Updat, 2006; 9:111–122.

18.

Braunbeck

, Bottcher

, Hollert

, Kosmehl

, Lammer

, Leist

et al. Towards an alternative for the acute fish LC50 test in chemical assessment: the fish embryo toxicity test goes multi-species—an update. ALTEX, 2005; 22:87–102.

19.

Alestrom

, Holter

, Nourizadeh-Lillabadi

. Zebrafish in functional genomics and aquatic biomedicine. Trends Biotechnol, 2006; 24:15–21.

20.

Eimon

, Rubinstein

. The use of in vivo zebrafish assays in drug toxicity screening. Expert Opin Drug Metab Toxicol, 2009; 5:393–401.

21.

Bennett

, Smith

. Arene ruthenium(II) complexes formed by dehydrogenation of cyclohexadienes with ruthenium(III) trichloride. J Chem Soc Dalton Trans, 1974; 42:233–241.

22.

Evans

, Spencer

, Wilkinson

. Dichlorotetrakis(dimethylsulphoxide)ruthenium(II) and its use as a source material for some new ruthenium(II) complexes. J Chem Soc Dalton Trans, 1973; 2:204–209.

23.

Herrmann

, Elison

, Fischer

, Kocher

, Artus

GRJ

. N-heterocyclic carbenes: generation under mild conditions and formation of group 8–10 transition metal complexes relevant to catalysis. Chem Eur J, 1996; 2:772–780.

24.

Prades

, Viciano

, Sanau

, Peris

. Preparation of a series of “Ru(p-cymene)” complexes with different N-heterocyclic carbene ligands for the catalytic beta-alkylation of secondary alcohols and dimerization of phenylacetylene. Organometallics, 2008; 27:4254–4259.

25.

Miller

, Rice-Evans

. Factors influencing the antioxidant activity determined by the ABTS radical cation assay. Free Radic Res, 1997; 26:195–199.

26.

, Pellegrini

, Proteggente

, Pannala

, Yang

, Rice-Evans

. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radic Biol Chem, 1999; 26:1231–1237.

27.

Westerfield

. The Zebrafish Book: A Guide for the Laboratory Use of Zebrafish. Eugene, OR: The University of Oregon Press, 1995.

28.

Kishi

, Bayliss

, Uchiyama

, Koshimizu

, Qi

, Nanjappa

et al. The identification of zebrafish mutants showing alterations in senescence-associated biomarkers. PLoS Genet, 2008; 4:e1000152.

29.

Clemedson

, Kolman

, Forsby

. The integrated acute systemic toxicity project (ACuteTox) for the optimisation and validation of alternative in vitro test. Presented at the 2nd Joint Conference of the Estonian Society of Toxicology/Scandinavian Society for Cell Toxicology, Frame, Ida Viru, Estonia, 2005.

30.

Paula

MMD

, Pich

, Petronilho

, Drei

, Rudnicki

, de Oliveira

et al. Antioxidant activity of new ruthenium compounds. Redox Rep, 2005; 10:139–143.