Biological evaluation of amidine derivatives: In vitro cytotoxicity and cellular antioxidant capacity

TL;DR

This paper evaluates amidine derivatives for cytotoxicity and antioxidant effects, finding them non-toxic and potentially useful in drug development.

Contribution

The study introduces amidine and formamidine urea derivatives as non-toxic and functionally relevant for further drug development.

Findings

Amidine derivatives showed LC50 values above 0.3 mM, indicating low cytotoxicity.

Most compounds suppressed antioxidant capacity or increased ROS generation.

Cell morphology changes aligned with viability measurements.

Abstract

Amidines and related compounds are well known intermediates and protecting groups in organic synthesis. New methodological approaches and obvious structural and functional relevance to guanidines and imidazoles have also prompted interest in the biological activity of these compounds. Here we report a preliminary cytotoxicty evaluation of a set a formamidines and formamidine ureas obtained by convenient and modular synthetic routes. Standard epithelial (Vero, MDCK-SIAT) and fibroblast cell lines (COS-1, COS-7) were employed. All compounds were found to be relatively non-toxic, with LC50 values all in excess of 0.3 mM, but found to vary over the range of compound structures. Cell morphological changes were in good agreement with cell viability. Most of the compounds either suppressed the cellular antioxidant capacity or promoted reactive oxygen species (ROS) generation. The nontoxic nature of these molecules at low to moderate concentrations suggests that the amidine and formamidine urea functional groups are suitable for continued investigation in drug development.

Figures

Compounds explored in this study.(A) Formamidine and (B) formamidine urea compounds with the code number for the most cytotoxic in red and the least cytotoxic in blue.

Optical microscopy images of Vero (left column) and COS-7 (right column) cells exposed to 24 h-extracts.(A) Control cells, (B) compound 2 (1.25 mg/mL, 3.7 mM), (C) compound 5 (1.25 mg/mL, 2.9 mM), and (D) compound 10 (4.25 mg/mL, 13.3 mM), each for 24 h. Scale bar = 200 µm.

SEM images of Vero cells exposed to 24 h-extracts.(A) Untreated (control) cells, (B) compound 5 at 1.25 mg/mL and (C) compound 10 at 4.25 mg/mL for 24 h. In (A) the arrows show healthy cell-substrate and cell-cell interactions; in contrast, in (B) and (C), * indicates cell detachment, while the loss of the interactions between the cell and other cells or the substrate.

SEM images of COS-7 cells exposed to 24 h-extracts.(A) Untreated (control) cells, (B) compound 5 at 1.25 mg/mL and (C) compound 10 at 4.25 mg/mL for 24 h. In (A) the arrows show healthy cell-substrate and cell-cell interactions; in contrast, in (B) and (C), * indicates cell detachment/rounding, while the loss of the interactions between the cell and other cells or the substrate.

ROS levels (%).Values measured for the 24 h DMEM extract (A) and the 7-day DMEM extracts (B) in contact with four different cell lines for 24 h.

Equations

Full text

Introduction

Formamidines are the class of amidines bearing a hydrogen atom at the central carbon (R^1^N=CH-NR^2^R^3^). Since the publication of the first conversion of formamides into formamidines by Mandel and Hill in 1954 [1], the synthesis of substituted formamidine derivatives has been the subject of many reports [2,3]. Especially significant have been the contributions of Raczyńska and Oszczapowicz, Cotton and Junk, and Meyers, each of whom have developed original synthetic methodologies for acyclic N,N’-substituted-formamidines. Key difficulties sidestepped by these advances include long preparation times with different intermediate steps, and limited stability of formamidines toward standard methods of purification.

The applications of the formamidine nucleus in organic chemistry are based on its role as protective group for primary amines [4], as auxiliary group in asymmetric synthesis [5,6], as ligand in metal-mediated catalysis in heterocycle or polymer synthesis [7,8], and as building block in functional materials due to hydrogen-bonding interactions [9].

Given their resemblance to biological functional groups in which two basic nitrogen atoms are bridged by a single carbon center, the biological activities of amidines and formamidines have also attracted much attention. In the case of formamidines, such activity is greatly influenced by their rates of hydrolysis [10], which are much greater than for acyclic amidines or cyclic imidazoles. Formamidines have been reported to be antihistamines [11], ligands for adrenergic and neurochemical receptors [12,13], and inhibitors of monoamine oxidase [14], among other applications. More broadly, recent years have seen sustained interest in a wide range of nitrogen-rich small molecules for biomedical applications, including enzyme inhibition and cytotoxicity-related studies. Structurally diverse systems such as sulfadrug-pyrrole conjugates, Mannich bases, pyrazoline sulfonamides, and pyrimidine derivatives have been investigated as carbonic anhydrase or acetylcholinesterase inhibitors, as well as for their cytotoxic or antiproliferati.e properties, often accompanied by in silico analyses to explore structure-activity relationships [15–18], These studies highlight the relevance of evaluating cellular toxicity and biocompatibility as part of the early biological profiling of small-molecule scaffolds. Within this context, amidines and formamidines represent a particularly interesting class of compounds due to their basicity, hydrogen-bonding capability, and structural similarity to biologically relevant motifs such as guanidines and imidazoles, while systematic data on the mammalian cytotoxicity of formamidines and formamidine ureas remain comparatively limited.

Early applications of formamidines were as pesticides displaying acaricidal, insecticidal, and antifeedant responses (e.g., amitraz [19], chlordimeform [20], formetanate [21]). Later, compounds showing potent analgesic [22], peripheral antinociceptive [23], and antiinflammatory activities [24] (e.g., clonidine, arylformamidines, indole derivatives) were developed, as well as a few examples of antiviral [25] and antimalarial drugs [26]. An interesting example was provided by Tarasiuk and co-workers, who described a new family of formamidine sugar-modified daunorubicin derivatives as potential anti-cancer drugs in cases of multidrug resistance [27]. Formamidine ureas, a class of formamidines bearing an aminoacyl group on one nitrogen (R^1^N=CH-NR^2^CONR^3^R^4^), also display a rich chemistry [28]. Their tunable hydrolytic instability makes them potential “soft drug” functionalities that can be broken down under physiological conditions [29].

To our knowledge, few reports on the mammalian cytotoxicity of formamidine compounds have appeared to date. In the context of polynucleotide delivery, amidine groups have been found to provide reduced cytotoxicity and improved transfection efficiency [30], as well as simulated cell penetrating properties that enhance cell uptake and intracellular delivery [31]. Pursuing this theme, Fischer et al. prepared in a one-pot reaction dextran-formamidine esters with different degrees of substitution for potential application to the cellular delivery of plasmid DNA [32]. The influence of the number of formamidine groups on in vitro cyto- and hemotoxicity was evaluated, and in this case higher degrees of formamidine loading was associated with reduced cellular viability. In a different type of application, del Pino and colleagues summarized current understanding of the molecular mechanisms of amitraz mammalian toxicity, pointing out the need for information such as dose−response relationships and toxic effects [33]. The study described below represents a survey of the effects of several formamidines and formamidine ureas on cultured cell lines as a preliminary assessment. We found these compounds in general to be only mildly cytotoxic (with the lowest LC_50_ values in the range of 300 µM and greater than 1 mM), but to exhibit interesting variations in cellular response with structure.

We prepared formamidines 1–5 and formamidine ureas 6–10 (Fig 1) by previously reported procedures [21,34,35]. The formamidines are derived from hydrazones and show particularly good stability towards aqueous hydrolysis. All of these compounds have a high density of hydrogen bond donor and acceptor centers.

Compounds explored in this study.(A) Formamidine and (B) formamidine urea compounds with the code number for the most cytotoxic in red and the least cytotoxic in blue.

The effects of these compounds on the viability of two cultured epithelial-like cell lines (MDCK-SIAT and Vero) and two fibroblast-like cell lines (COS-1 and COS-7) were examined. Morphological changes were characterized by optical and scanning electron microscopy. In addition, to further explore the suitability of these compounds in a biomedical context, the total cellular antioxidant capacity was determined by a reactive oxygen species (ROS) assay.

Experimental section

Materials

Materials and experimental procedures related to the preparation of formamidines (1−5) and formamidine ureas (6−10) are described in the Supporting Information in S1 File. MDCK-SIAT, Vero, COS-1, and COS-7 cells were obtained from ATCC (USA). Dulbecco’s phosphate buffered saline solution (PBS) without calcium chloride and magnesium chloride, Dulbecco’s modified Eagle’s medium (DMEM, with 4500 mg/L of glucose, 110 mg/L of sodium pyruvate and 2 mM L-glutamine), penicillin–streptomycin, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, 97.5%) and trypsin–EDTA solution (0.05% trypsin, 0.02% EDTA) were all purchased from Sigma-Aldrich (USA). Trypan blue stain (0.4%) was purchased from Gibco, UK. Dimethyl sulfoxide (99.0%) and methanol were purchased from Panreac Quimica S.A.U. (Spain). For the antioxidant capacity assay, the following cell lines were used: COS-1, MCF-7, MDCK-SIAT, and PNT2. All other chemical reagents were used as received without further purification, including ABTS (2,2’-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) from Merck, and hydrogen peroxide from Sigma-Aldrich.

Synthesis and characterization of amidine derivatives

Briefly, formamidines 1–5 were prepared by a two-step exchange process using acetonitrile as solvent and DMAP as catalytic base (Scheme 1A and Scheme S1). Formamidine ureas were obtained from the addition of a substituted urea to a mixture of isocyanide and acid chloride in THF or MeCN, which gives formamidine urea salts in pure form as precipitates (Scheme 1B). The desired products were characterized by standard techniques (Electronic Supplementary Information, ESI in S1 File).

Scheme 1. General synthetic procedures. Synthesis of (A) formamidines from formamide acetals and (B) formamidine ureas. $[eqn]$

Cytotoxicity of amidine compounds

Cell viability assays.

Cell viability was evaluated in the presence of compounds incubated in complete culture medium (DMEM) for periods of 24 h and 7 days, in order to assess the effects of intact and hydrolysed samples, respectively. Thus, each formamidine compound was first sterilized in the solid state by irradiation with UV light (254 nm) for 15 min in a laminar flux cabinet. DMEM was added to achieve concentrations of 2.5 mg/mL (5.7–9.3 mM, depending on the compound) and 8.5 mg/mL (26.6–41.9 mM) for formamidines and formamidine ureas, respectively. After homogenizing by vortex mixing for several minutes, each sample tube was sealed with parafilm to prevent evaporation and placed in a shaking incubator (80 rpm) at 37 °C for 24 h to promote solubilization. Each mixture was then centrifuged at 12000 rpm for 15 minutes, and the supernatant filtered (0.22 μm) to obtain the extract used in subsequent tests. All such extracts were frozen at –20 °C for storage and thawed immediately before use in cell culture assays. The extent of hydrolysis that may have occurred under these conditions was not directly quantified. While electron-rich formamidine derivatives are generally reported to exhibit moderate stability under neutral aqueous conditions, formamidine ureas are expected to be more susceptible to hydrolysis [29]. As such, the biological assays described herein intentionally assess the cellular response to the full composition of the extracts after defined incubation periods, encompassing both intact compounds and any degradation products. Accordingly, the LC50 values reported represent effective cytotoxicity values for the extracts and should be interpreted within this context.

The immortalized MDCK-SIAT, Vero, COS-1, and COS-7 cell lines were chosen for their stability and fast in vitro growth, providing a good initial indication of cytotoxicity (but not predictive of the response of primary cells or cells in vivo). Cells were cultured in DMEM high glucose supplemented with 10% FBS, penicillin (100 units/mL), and streptomycin (100 μg/mL). Further details on cell culture are provided in the Supporting Information in S1 File.

Cell viability was evaluated by the colorimetric MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay [36]. Specifically, the culture plates were washed twice using 200 µL of PBS before adding 20 µL of a MTT solution (5 mg/mL in PBS) to each well. After 3 h of incubation, the medium from each well was aspirated, 200 µL of DMSO/methanol/water (70/20/10% v/v) was added, and the plate was gently agitated for 15 minutes. Absorbance at 570 mm was measured using a microplate reader (Biochrom, UK). The viability in each well was calculated and normalized to the control (culture cells without extracts) as relative percentages.

Cell morphology.

Morphological analysis was performed by optical and scanning electron microscopy. Briefly, VERO and COS-7 cells were seeded onto glass cover slips in 24-well culture plates as described above. After cell attachment, the plates were washed with PBS (200 µL, twice), and cells were exposed to formamidine extracts for 24 h under culture conditions. Optical morphologic observations of cells stained with hematoxylin and eosin stain (H&E) were performed with a Zeiss Axioskop 40 microscope, and micrographs were taken with a Zeiss AxiosCam MRC5 digital camera. Identically exposed cells were also observed using a Focused Ion Beam Zeiss Neon40 scanning electron microscope (SEM) operating at 5 kV. These cells were fixed in a 2.5% glutaraldehyde PBS solution (pH = 7.2) overnight at 4 °C, dehydrated by washing in an alcohol battery (30°; 50°; 70°; 90°; 95° and 100°) at 4 °C for 30 min per wash, air-dried, mounted on a double-side adhesive carbon disc, and sputter-coated with a carbon layer of 6–10 nm thickness (K950X Turbo Evaporator) to prevent sample charging problems.

Reactive oxygen species (ROS) assay.

The influence of amidine compounds on the cellular generation and management of reactive oxygen species (ROS) was evaluated using variation of a standard assay with 2,2’-azinobis (3-ethylbenzthiazoline-6-acid) (ABTS) [37]. Aliquots containing 10^4^ cells in 150 µL of fresh medium were maintained under culture conditions for 24 h. The medium in each well was then aspirated to withdraw non-adherent cells, and the wells were washed twice with 200 µL PBS. The cells were then cultured with the appropriate formamidine or formamidine urea extract at a concentration equivalent to the LC_50_ determined by MTT assay. The negative control was DMEM only, and the positive control was 3% hydrogen peroxide to induce oxidative stress in the cells. After incubation for 24 h, the cells were washed twice with PBS and treated with 100 µL DMSO and 100 µL ABTS solution (10 mg ABTS + 2 mg potassium persulfate in 1 mL water and kept in the dark at room temperature for 12–16 h before use, and then diluted with 60 mL methanol). The plates were incubated for 30 min on a shaker at room temperature, and the absorbance was read in a microplate reader at 570 and 405 nm. The ROS level (%) was determined by percentage of inhibition of absorbance (ABTS radical cation decolorization assay), and was calculated as follows:

$$[eqn]$$

where control A_0_ and sample A_1_ are absorbance values of the ABTS with DMEM only and compound extracts, respectively.

Statistical analysis.

Biocompatibility was evaluated using three replicates in independent experiments, and the results were averaged (mean ± SD). Data were plotted as relative viability (%) vs. extract concentration, and the curve obtained was adjusted to fit dose-response using OriginPro v10 software (Origin Microcal, USA). The 50% lethal cytotoxicity (LC_50_) was defined as the concentration of the extract (in mg/mL) at which the relative cell viability equals 50%.

Results and discussion

The synthetic methods applied are notable in their production of pure compounds, isolated as white solids by filtration in moderate yields [34,35]. Table 1 summarizes the properties of the compounds tested in this work.

Table 1: Formamidine and formamidine urea compounds explored in this study.

Biocompatibility evaluation of amidine derivatives

Cell viability assays.

In general, the fibroblast-like COS-1 and COS-7 cell lines were more susceptible than the epithelial VERO or MDCK-SIAT cells to the compounds tested. Among the amidine compounds, DMEM extracts from 4, 5, and 10 showed the greatest dose-dependent cytotoxicity, with LC_50_ values (the concentration of extract that inhibits relative cell viability by 50%) below 1 mg/mL for all four cell lines (Table 2, and Supporting Information for further details in S1 File). Most other compounds were less toxic by approximately 5–10-fold (Table 2). Compounds 1 and 2 were exceptions, showing very little toxicity toward all four cell lines at the maximum concentrations tested.

Table 2: LC50 values for the amidine derivatives extracts.

In most cases, incubation for 24 h and 7 days gave very similar cytotoxic effects. Notable exceptions were compound 5 (and to a lesser extent, 4), which were several-fold less harmful to the cells over longer incubation time. Constant or diminishing apparent toxicity over longer exposure times are both unusual observations for standard cytotoxic agents: longer exposure usually leads to greater cell death. We propose that the hydrolytic instability of these molecules, either in the aqueous medium or after cellular uptake, could be responsible, if the hydrolysis products (amines, formamides, and formaldehyde) are less harmful than the assembled amidines or formamidines.

The most benign structures (1 and 2) share a common feature of a benzylic or pyridinylmethyl amine component instead of an amide – for example, compounds 2–3 differ by only one oxygen atom, but have notably different effects on two cell lines. Also striking is the difference between 4, which exhibits the same level of mild toxicity towards all cell lines tested, and 3, is significantly less harmful to two cell lines: these structures are identical except for the pyridyl nitrogen in the latter molecule. Taken together, these comparisons suggest that relatively subtle structural variations can have a measurable impact on cellular response. In particular, the presence of benzylic or pyridinylmethyl amine motifs appears to be associated with lower cytotoxicity compared to closely related amide-containing analogues. Similarly, small heteroatom substitutions, such as the introduction of a pyridyl nitrogen, can result in noticeable differences in cytotoxic behavior across specific cell lines. Although these trends are based on a limited set of compounds and cell models, they provide a qualitative framework for understanding structure-dependent effects within this compound family.

Cell morphology.

Cell morphologies in the presence of amidine and formamidine extracts, examined by optical microscopy, were consistent with the MTT assays of cytotoxicity, as shown in the representative example of Fig 2. Here, Vero and COS-7 cells exposed to 24 h-extracts of four compounds showed cell numbers and morphologies to be largely unchanged for compound 2, but not for 5 and 10. In this experiment, most of the cells exposed to 10 died after 24 h of culture regardless the cell type, in addition to exhibiting abnormal morphology.

Optical microscopy images of Vero (left column) and COS-7 (right column) cells exposed to 24 h-extracts.(A) Control cells, (B) compound 2 (1.25 mg/mL, 3.7 mM), (C) compound 5 (1.25 mg/mL, 2.9 mM), and (D) compound 10 (4.25 mg/mL, 13.3 mM), each for 24 h. Scale bar = 200 µm.

Inspection of cell morphology by scanning electron microscopy (SEM) allowed for the observation of details of cellular shape at high magnification. Figs 3 and 4 present SEM images of Vero and COS-7 cells, respectively, exposed to the more cytotoxic compounds, 5 and 10. In comparison to untreated cells (Figs 3A and 4A), which are well spread onto the surface establishing cell-to-cell interconnections, the amidine derivatives induced cell detachment from the surface and negative effects on the cytoplasm. The long and thin actin filaments (filopodia) oriented towards the surface and responsible for cell adhesion are largely absent in the treated samples. Similar morphologies were observed for cells exposed to 7d-extracts (images not shown). All these distinctive morphological features between control and treated cells have been indicated in the SEM images.

SEM images of Vero cells exposed to 24 h-extracts.(A) Untreated (control) cells, (B) compound 5 at 1.25 mg/mL and (C) compound 10 at 4.25 mg/mL for 24 h. In (A) the arrows show healthy cell-substrate and cell-cell interactions; in contrast, in (B) and (C), * indicates cell detachment, while ** the loss of the interactions between the cell and other cells or the substrate.

SEM images of COS-7 cells exposed to 24 h-extracts.(A) Untreated (control) cells, (B) compound 5 at 1.25 mg/mL and (C) compound 10 at 4.25 mg/mL for 24 h. In (A) the arrows show healthy cell-substrate and cell-cell interactions; in contrast, in (B) and (C), * indicates cell detachment/rounding, while ** the loss of the interactions between the cell and other cells or the substrate.

Reactive oxygen species (ROS) assay.

To probe the potential effects of amidine compounds on cellular oxidative metabolism, [38] we measured the total antioxidant capacity of several different cell lines (COS-1, MCF-7, MDCK-SIAT, and PNT2) after exposure to the amidine derivatives at their previously-determined LC_50_ concentrations (Fig 5). 24 h DMEM extracts of cpds 3, 4, and 5 induced little to no effect on the cell management of ROS levels regardless of the cell line (Fig 5A). In contrast, for the remaining compounds, ROS levels similar to or higher than those exhibited by H_2_O_2_-treated cells were observed. The most striking effect was observed for DMEM extracts of compound 10, which greatly diminished the ability of all but the COS-1 cell line to manage oxidative stress. Similarly, the 7-day DMEM extract of compound 1 displayed a similar property (for two cell lines).

ROS levels (%).Values measured for the 24 h DMEM extract (A) and the 7-day DMEM extracts (B) in contact with four different cell lines for 24 h.

The measured ROS levels did not correlate with observed cytotoxicity values for the amidine and formamidine urea compounds, as would be expected if ROS-induced apoptosis were an important mechanism [32,39–41]. Thus, the least cytotoxic molecules (1 and 2) all induced significant apparent levels of ROS, but other compounds, which showed greater toxicity, gave very similar ROS values. The most cytotoxic agents (4, 5, and 10) were widely divergent in ROS generation. It must be emphasized again that these experiments were conducted at high compound concentrations (at or above 1 mM) relative to the active concentrations of most drug molecules.

Overall, our findings indicate that the cytotoxic effects induced by the amidine compounds occur independently of ROS, suggesting that reactive oxygen species are not the primary drivers of the observed outcomes and that ROS-dependent pathways are unlikely to be involved. Instead, our observations are consistent with loss of cell adhesion and cytoskeletal stability, including pronounced disruption of actin microfilaments and the microtubule network, along with a reduction in focal adhesion size. These alterations led to cell rounding, detachment, reduced metabolic activity, and ultimately cell death—potentially via anoikis or apoptosis—through a ROS-independent mechanism. A detailed elucidation of the precise mechanism underlying cell death lies beyond the scope of the present study, but our results collectively support a non-oxidative, adhesion- and cytoskeleton-related mode of cytotoxicity for these amidine compounds.

Conclusion

Extensive work has appeared in the literature on the cellular toxicities of formamidine-based pesticides [19,42–47]. These insect-focused studies, which were concerned on such mechanisms as the blocking of neuromuscular transmission and interfering with amine regulation, identified molecular features such as small size, specific aryl ring substitution, and lipophilicity as significant enhancers of insectical effects. The compounds explored here are different and likely to engage cellular mechanisms in different ways. Indeed, these molecules were found to be quite benign in their effects on four cultured cell lines, at least in terms of cell death and changes in observable morphology. Overall, these results provide some confidence that formamidine and formamidine urea derivatives – which have advantageous properties of synthetic accessibility, basicity, and H-bonding ability – can and should be used in the design of molecules that engage biological systems in useful ways.

Supporting information

S1 FileAdditional experimental data regarding the biocompatibility evaluation of amidine derivatives, as well as 1H NMR, 13C NMR, and representative mass spectra (TOF MS-ES^+^) of the compounds.(DOCX)