2-Azahetaryl-2-(oxoindolin-2-ylidene)acetonitriles as Colorimetric Probes for Zn: Synthesis and Optical Properties

A new one-pot approach for the synthesis of the Zn2+-sensitive probes 2-azahetaryl-2-(oxoindolin-2-ylidene)acetonitriles 3a–c and 4 is described. The method includes the in situ formation of imidoylchloride and its further condensation with azahetarylacetonitrile 1. The structure of the obtained compounds is studied using 1H nuclear magnetic resonance (NMR), 13C NMR, infrared (IR), high-resolution mass spectrometry (HRMS), and UV–Vis spectroscopy techniques. Two model ligands both exhibiting the highest extinction coefficient and the best solubility in a Tris buffer pH 7.2/dimethyl sulfoxide (DMSO) solution, namely 5-methyl-benzothiazole derivative 3b and benzoxazole derivative 4, are thoroughly studied as colorimetric probes for Zn2+. The probe 3b has the highest sensitivity to Zn2+, showing a limit of ion detection (LOD) calculated by the 3S criterion of 0.43 μM and selectivity upon masking Cu2+ ions with Na2S2O3. The composition of the complexes in the solution was determined by the limited logarithm method. The stability constant (lg K) values of 3b-Zn of 10.27 ± 0.02 and 4-Zn of 12.5 ± 0.2 indicate the formation of complexes of average stability.


INTRODUCTION
Zinc is widely recognized as a "gatekeeper of the immune system" 1 that protects the human body against diseases and possesses antioxidant, anti-inflammatory, and antiviral activities, 2 including against Sars-Cov2. 3−7 To assess the individual or population zinc status, various biomarkers are recommended, in particular, concentration of zinc in plasma or serum. 8−10 From a research perspective, the recommended methods for the determination of Zn 2+ in biofluids are atomic absorption spectroscopy and inductively coupled plasma mass spectrometry. 11,12 However, these methods are too expensive to be involved in routine clinical analysis. Fluorescence analysis is also a widely used technique; 13 however, UV−VIS spectrophotometry (SP) remains the state-of-the-art technique for biochemical and clinical research due to its accuracy, sensitivity, and simplicity. Despite the large number of chromophore probes for zinc, 14 a limited number of them are used for serum and plasma SP analysis. Well-known ones in the range are PAR 15 and Zincon, 16 but they either do not have contrasting color transitions or sufficient selectivity or require toxic masking agents such as cyanide ions, for example, in serum analysis with nitro-PAPS 17 (Figure 1).
It is known that porphyrinoids because of their large conjugated system have a strong absorption peak in the visible region. 18 This feature correlates with their wide use as a sensitive chromogenic reagent for the detection of metal traces. 18−22 However, their coordination to metals often requires elevated temperature, auxiliaries, or pH far from biologically relevant conditions, 18,19 probably because of the rigidity of the cavity and the required deprotonations ( Figure  2a). Moreover, fast functionalization of porphyrins suffers from significant synthetic limitations; thus, switching their coordination and optical properties is troublesome. Multistep synthesis of the probe is indeed usually required. 23 Here, we report simplified nonmacrocyclic analogues of porphyrins, 2-azahetaryl-2-(oxoindolin-2-ylidene)acetonitriles 3a−c and 4 that are easily available by a one-pot threecomponent reaction. Featuring the dipyrrin-like skeleton, they exhibit the potential for metal chelation. The representative probe 3b can chelate Zn 2+ at room temperature (hereafter rt) in Tris buffer pH 7.2/DMSO (1/5, v/v) with a LOD equal to 0.43 μM and selectivity upon masking Cu 2+ ions (Figure 2b).

Synthesis.
To modulate the spectral characteristics of colorimetric probes, we aimed at simplifying the structure of the well-known porphyrins. Thus, we envisaged that 2azahetaryl-2-(oxoindolin-2-ylidene)acetonitriles 3a−c and 4 might be competitive with porphyrins in terms of light absorption, while they could be synthesized from commercially available isatines 2 and N-hetarylacetonitriles 1 easily accessed synthetically from the readily available starting materials. Moreover, kinetics of the coordination of Zn 2+ might be enhanced because of the decreased steric hindrance at the coordination site.
It is known that the reaction of isatines with nucleophiles is directed at position 3 of isatin. To direct the reaction at position 2, a preactivation can be achieved by refluxing isatin and PCl 5 in benzene. The resulting imidoylchloride is so far, requiring isolation and dissolution in more polar solvents to perform the condensation reaction. This two-step approach decreases the yield of the product due to the poor stability of imidoylchloride, promoting the formation of byproducts. 24   To overcome this limitation, we have reported a one-pot method consisting first ofin situ formation of imidoylchloride followed by its condensation with hetarylacetonitrile 1 (Scheme 1). 27 Similar three-component reactions were applied for the synthesis of 2-azahetarylenaminonitriles as reported earlier by our group. 28 The limiting reaction step of the process is the generation of imidoylchloride, yet the reaction is over within an hour. The yield of the products ranges from 70 to 87%, and their purification is achieved by simple filtration through a plug of silica gel using dichloromethane as an eluent. In the 1 H NMR spectra of 3a−c and 4, the characteristic broad peak of the proton of the amino group is significantly deshielded (δ = 11.4−12.0 ppm). Most likely, the reason for this downfield shift is an intramolecular hydrogen bond that forms between the nitrogen of the azaheterocycle and the hydrogen of the indolinone backbone. Single crystals of 3b are grown by a gas diffusion method from CHCl 3 /hexane; XRD allows the full assessment of the molecular structure ( Figure 3).
According to the XRD of 3b, the estimated length of the hydrogen bond N3H3A···N1 is 2.15 Å and the angle N3− H3A···N1 is equal to 123.3°. The dihedral angle between heterocycle planes is equal to 8.8°, which indicates an effective charge delocalization in the molecule, promoting the shift of the NH in 1 H NMR to the deshielded side.
A complete assignment of the peaks at 1 H and 13 C NMR was performed using 2D NMR techniques (COSY, HMBC, and HMQC).
The method showed a wide scope allowing to generate a library of ligands to modulate the donating properties of the R substituent and the nature and electronic properties of the attached heterocycle. Thus, a library of compounds has been accessed based on either benzothiazole (3) or benzoxazole (4) from a reaction with unsubstituted isatin (a), isatin featuring a σ-donor such as a methyl group (b), or a π-donor such as a methoxy group (c).

Photophysical
Properties. 2-Azahetaryl-2-(5-R-3oxoindoline-2-ylidene)acetonitriles are characterized by the rich color ranging from red to purple. The nonsubstituted isatin compounds 3a and 4 in DMSO featured absorption spectra with a long wave band at 425−600 nm and the maximum centered around 520 and 515 nm, respectively ( Figure 4 and Table 1). Methyl-and methoxy-substituted compounds 3b and 3c in DMSO featured absorption spectra with the maximum centered around 530 and 550 nm, respectively, and featuring less intensive peaks at 650−750 nm. The increased bathochromic shift of the substituted dyes compared to nonsubstituted ones may be rationalized by the presence of electron density donors at the oxoindolinone core consequently shrinking of the HOMO−LUMO gap.

Photostability Test.
To assess the stability of dyes in DMSO solutions, UV−Vis absorption spectra were recorded after 24 h, 72 h, 96 h, and 6 months of exposure to indirect light. The absorbance of the solutions remains unchanged, which indicates a high photostability of the dye solutions (see Figure S14).

UV−Vis Spectra in H 2 O/DMSO Solutions.
Water solubility of 2-azahetaryl-2-(5-R-3-oxoindolin-2-ylidene)acetonitriles affects seriously their spectroscopic and chemical properties. The influence of water on the absorption spectra of the dyes 3a, 3b, and 4 in DMSO and the values of their molar absorption coefficient (ε) were studied (see Supporting Information Section 3). The molar absorption coefficient change was calculated as Δε max , % = (ε max DMSO −   , v/v) mixtures. The decrease of the molar absorption coefficient was lower than 10% (Δε max ≤ 10%), while a decrease of the absorbance value up to 50% was observed for dye 3a (see Table S2). Based on this observation, dyes 3b and 4 were chosen for further investigations. The deprotonation of molecules 3b and 4 at the NH site upon titration with alkaline solution is demonstrated in Figures  5 and 6. The isosbestic points located at 584 nm for 3b and 562 nm for 4 indicate the coexistence of two forms (neutral and anionic) of dyes in alkaline solutions. The bathochromic shifts (Δλ) of the absorption bands of neutral forms (λ max = 530 nm (3b) and λ max = 515 nm (4)) to anionic forms are equal to 150 nm for 3b and 130 nm for 4, which correspond to the drastic color change from purple to turquoise visible by the naked eye. These transitions were shown to be reversible. The bathochromic shift occurring under basic conditions may be rationalized by the formation of an anion, characterized by the facilitated oscillation of the π-electrons compared to the neutral form. Therefore, less energy is required to excite the electrons, which causes a long-wavelength π → π* electronic transition (see Scheme S1).
The apparent ionization constant (pK a ) values of molecules 3b and 4 in H 2 O/DMSO mixtures have been obtained with spectrophotometric titrations according to eq 1 (Section 4.3) (Tables 2, S4, and S5).

Interaction with Zn 2+ .
It was envisioned that dyes 3b and 4 can chelate Zn 2+ due to the presence of the dipyrrin-like skeleton. 29 Indeed, upon addition of Zn 2+ to the solution of 3b and 4, the hyperchromic effect occurs at 600 and 565 nm, respectively. Meanwhile, the color changes from purple (3b) and red (4) to blue, which corresponds to the bathochromic shift of λ max in the UV−vis spectra (Figures 7 and 8). Such changes may probably indicate the formation of complexes. The most convenient environment for the interaction of the ligand and Zn 2+ was chosen based on two factors: the largest achievable light absorption of the complex at the appropriate    wavelength (to ensure the highest detection sensitivity) and the lowest possible DMSO content. Keeping this in mind and considering the data presented in Figures S17b and S18 and Table S3, the ratios of Tris buffer (pH 7.2) and DMSO of 1/5 (3b) and 1/3 (4) were the best. The classic Bent and French method 30 was applied for spectral data processing. Considering the shape of the saturation curves in Figures 9 and 10, it can be assumed that a medium stability complex was formed in the solutions. The molar composition of complexes (Zn m R n ) was established as described in Section 4.4. The calculations gave the values of n/ m equal to 1 for both complexes of probes 3b and 4 with Zn 2+ , which confirmed the formation of ZnR complexes in solutions.
Under these conditions, the neutral forms of probes appeared to be predominant in the solutions (see Figure S19).
The stability constants of the complexes 3b-Zn and 4-Zn have been calculated by processing the data from the saturation curves (Figures 9 and 10) using eq 2 (see Supporting Information Section 5). The values of the affinity constants were found to be equal lg K f = 10.27 ± 0.02 and lg K f = 12.5 ± 0.2 for probes 3b and 4, respectively.
2.6. Spectrophotometric Determination of Zn 2+ . Hence, the absorbance of probes 3b and 4 at the maximum of the complexes after treatment with the Zn 2+ solutions was also found to be dependent on their concentration. The dependence of the absorption maxima and absorbance at the wavelength ranges with a minimum spectrum overlap is consistent with the Beer−Lambert law in the wide Zn 2+ concentration ranges (Table 3).
The dye 3b exhibits better characteristics due to the wider linearity of the Zn 2+ concentration range and higher sensitivity. The selectivity of dyes 3b toward Zn 2+ was appraised by titration of the probe with various metal ions under the optimal conditions of zinc complex formation (see Supporting Information Section 6). It was shown that heavy metal ions such as Fe 3+ , Fe 2+ , Ni 2+ , and Cd 2+ did not affect the absorption of dye 3b (see Figure S22). As expected, only Co 2+ and Cu 2+ caused significant change in the colors of the solutions. However, the change of the color in the presence of Co 2+ is a very slow process and the absorption spectrum change could be observed after 30 min of solution treatment with Co 2+ (see Figure S23). Due to this fact, the presence of an equal amount of Co 2+ does not interfere with the Zn 2+ determination during 20 min after the probe treatment. It was found that Cu 2+ ions could be masked using 1.0 mmol·L −1 Na 2 S 2 O 3 solution (see Figure S24).
To prove the reliability of the technique, the probe 3b was applied for the detection of Zn 2+ in the serum model solution (stripped of protein) (see Supporting Information Section 7). The model solution of two times diluted pretreated blood serum was prepared as recommended in ref 31 using standard solutions of Zn 2+ , Cu 2+ , Fe 2+ , Na + , K + , Ca 2+ , Mg 2+ , and Tris buffer (pH = 7.2). The resulting samples were analyzed spectrophotometrically. For this purpose, 2.5 mL of 40 μmol· L −1 of probe 3b solution in DMSO was mixed with 0.5 mL of   model solution in Tris buffer (pH = 7.2) in the presence of Na 2 S 2 O 3 , and the absorbance of the solutions was measured. The recovery value does not exceed 105% for 5.0 and 10.0 μmol·L −1 spiked Zn 2+ in the sample. The data obtained suggest that the proposed technique could be used for Zn 2+ determination in real samples. Full method validation using real samples will be the aim of our future work.
A comparison of probe 3b with other reported colorimetric probes for Zn 2+ in complex matrices has shown that the closest analogue of probe 3b in terms of sensitivity, metal ion selectivity, and contrasting color change accompanying the complexation is Zincon (see Table S9). Moreover, probe 3b demonstrates much better stability in solutions than Zincon. 16 Finally, a high DMSO content in the sample solution does not interfere with the determination of the Zn total amount in the pretreated biosamples using probe 3b.

CONCLUSIONS
In summary, we have developed a one-pot approach for the synthesis of 2-azahetaryl-2-(oxoindolin-2-ylidene)acetonitriles via in situ formation of imidoylchloride, and we have found that these dyes are sensitive probes for Zn 2+ ions when Cu 2+ ions are masked with Na 2 S 2 O 3. The best analytical response is achieved with benzothiazole derivative 3b with a LOD of 0.43 μM in a Tris buffer pH 7.2/DMSO solution. The contrasting color change accompanied with the complex formation and the stability of the dye and its complex with zinc facilitate the highly sensitive and selective determination of Zn 2+ ions using UV−Vis spectrophotometry. Therefore, probe 3b is promising for detecting and monitoring Zn 2+ ions in biological fluids, such as serum.
Preparative chromatography was performed manually with silica gel (63−200 μm). 1 H NMR and 13 C{ 1 H} NMR spectra were recorded on a Bruker Avance 300 (300 and 75 MHz, respectively) or Bruker Avance III 400 (400 and 101 MHz, respectively) spectrometer and were referenced to the residual proton resonances of the DMSO-d 6 : 2.50 for 1 H and 40.0 ppm for 13 C. The degree of carbon atom substitution was determined by NMR spectra acquired according to the DEPT-135 or JMOD methods. Two-dimensional (2D) (COSY, HMQC, HMBC) spectra were recorded on a Bruker Avance III 400 spectrometer and Bruker Avance 500. All spectra were recorded at ambient temperature (298 K). Coupling constants (J) were reported in Hz, and chemical shift (δ) were reported in ppm. The multiplicity of signals is indicated using the following abbreviations: s = singlet, bs = broad singlet, d = doublet, t = triplet, q = quartet, quint = quintet, dd = doublet of doublets, ddd = doublet of doublets of doublets, dt = doublet of triplets, m = multiplet, and br = broad.
IR spectra were recorded on a FT-IR spectrometer Perkin Elmer in KBr plates.
High-resolution mass spectra (HRMS) were recorded on a GCT Premier spectrometer upon electron spray ionization (ESI) or chemical ionization (DCI, CH 4 ).
Melting points were determined on a Boetius micro-hot stage apparatus with a VEB Analytik 1399RNMK 05 observation lens.
X-ray crystallography was performed on a single-crystal diffractometer Agilent Gemini. A gas diffusion method was implemented for growing single crystals.
Absorption spectra were recorded on a SHIMADZU UV-2401 PC spectrophotometer. pH measurements were performed with a STARTER 3100 bench pH meter (OHAUS).

General Procedure for the Synthesis of 2-Hetaryl-2-(5-R-3-oxoindolin-2-ylidene)acetonitriles 3a−c and 4.
To a strongly stirred suspension of 5-R-isatine (1 mmol) in hot benzene (10−15 mL), phosphorus pentachloride (1.3 mmol) was added. The resulting mixture was refluxed for 1 h. Then, a solution of hetarylacetonitrile (1 mmol) in benzene was added dropwise and a characteristic color change occurred. The mixture was allowed to cool to room temperature upon stirring. The precipitate was then filtered and washed with minimum amounts of ethanol and water and again with ethanol. The dry precipitate was purified by filtration through a plug of silica gel (eluent: CH 2 Cl 2 ).   The calculation of apparent ionization constants was carried out using the absorption data according to the Henderson− Hasselbalch equation

(Z)-2-(Benzo[d]thiazol-2-yl)-2-(5-methyl-3-oxoindo-
where A HR is the absorbance of the solution at the absorption maximum of the molecular form of the dye, A R − is the absorbance of the solution at the absorption maximum of the ionic form of the dye, and A i is the absorbance of the solution at the absorption maximum of the mixture of ionic and neutral forms, corresponding to a certain pH value, which was measured using a pH meter for each solution. The absorption maxima of the neutral and anionic forms of dyes are 530 and 680 nm (3b) and 515 and 645 nm (4), respectively.

Study of Complexation.
The complexation study was realized as recommended in ref 36 using spectrophotometric titration according to the limited logarithm Bent and French method 30 and the Molland method. 37 For this purpose, two series of solutions were prepared. The first series of solutions contained equal amounts of Zn 2+ (10 μmol·L −1 for 3b and 5 μmol·L −1 for 4), whereas the concentration of probes 3b and 4 varied (10−90 μmol·L −1 and 0.5−81 μmol·L −1 , respectively) upon the addition of the small portion of the corresponding probe. The second series of solutions contained equal amounts of probes 3b (38 μmol·L −1 ) or 4 (9.9 μmol· L −1 ), whereas the concentration of Zn 2+ varied (2−90 μmol· L −1 and 1−112 μmol·L −1 , respectively). The absorption spectra of these solutions were recorded, and saturation curves ΔA as a function of C, μmol·L −1 , were obtained, where ΔA = A R − A ZnR at a certain wavelength.
The molar composition of complexes (Zn m R n ) was established using the formula n/m = (ΔA max Zn × C R )/ (ΔA max R × C Zn ), where m and n are the number of Zn 2+ and the ligand attached to a metal ion consequently, respectively; ΔA max Zn and ΔA max R are absorption, corresponding to the saturation area; and C R and C Zn are the initial concentrations of the ligand and Zn 2+ in solutions of each series, respectively.
The stability constants of the complexes were calculated using eq 2 ( ) where ΔA i is the value of absorbance in the initial linear section of the saturation curves; A max Zn is the value of absorbance in the saturation region; C R is the concentration of the probe, which corresponds to ΔA i ; C Zn is the concentration of Zn 2+ ; n is the number of the coordinated ligands; x is the number of protons capable of dissociation of dyes 3b or 4 (x = 1); y is the number of protons detached from the dye (y = 1); and K a is the ionization constant of dyes 3b or 4.