Exploratory study on sequestration of some essential metals by indigo carmine food dye

Indigo carmine forms a stable complex with different ions, and the stability constant of the complexes were evaluated as log K equal to 5.75; 5.00; 4.89 and 3.89 for complexes with Cu(II), Ni(II), Co(II) and Zn(II) ions, respectively, in 0.1 mol L-1 carbonate buffer solution at pH 10. The interaction between Cu(II) ions and indigo carmine (IC) in alkaline medium resulted in the formation of the Cu2(IC) complex, measured by the spectrophotometric method, with a stoichiometric ratio between indigo carmine and metal ions of 2:1 (metal-ligand). The reported method has also been successfully tested for determination of copper in pharmaceutical compounds based on copper-gluconate without pre-treatment.


INTRODUCTION
Copper, zinc, cobalt, iron and nickel play a vital role in human metabolism, since they are involved in almost all known important reactions and metabolic routes (Devlin, 2002).Therefore, the knowledge of their interactions with a food dye can be very important since they can appropriate important metallic compounds to the human organism.As one of the most important essential transition metals, copper is involved in a variety of biological processes such as embryo development, connective tissue formation, temperature control and nerve cell function (Wang, Guo, 2006).Copper is transported from dietary intake through the serum and into the organism via a variety of transporters and patients with deficiency of this metal need to be treated with copper supplements.Menkes disease is characterized by a systemic copper deficiency resulting from a recessive mutation in copper transporter.However, it is known that copper can cause toxic effects at elevated concentrations (White, Rainbow, 1985), whose toxicity depends on environmental factors such as pH, salinity, biological media and organic matter (Campbell, 1995;Allen, Hansen, 1996).Taking into consideration that treatment of both diseases involves copper chelation therapy, the study of new complexing agents of copper with low toxicity seems relevant.
It is well known that most types of dyes form complexes with metal ions in aqueous media (Zollinger, 1991).Interactions between food dyes and metal ions may change their stability, toxicity and other physico-chemical properties of both metal and the dye.The toxicological evidence for synthetic colorants has led to reviewed safety of several compounds in most countries (Marmion, 1991).Accordingly, it is important for human health to know these complexes.
The present paper investigated the feasible complexation between indigo carmine (Figure 1) and some essential metals including Cu(II), Ni(II), Zn(II), Co(II), Fe(II), Fe(III) and K(I) by the spectrophotometric method.The possibility of accomplishing the determination of copper (II) via a complexation reaction with indigo carmine, and its application in pharmaceutical formulations of copper complex containing Cu(II)-gluconate, is also discussed.

Apparatus and procedures
Analytical grade reagents supplied by Merck and using high purity water from a Milli-Q system (Millipore Inc., USA) were used in the preparation of all solutions.Spectrophotometric analysis employed 0.1 mol L -1 carbonate buffer solution at pH 10, 0.1 mol L -1 phosphate buffer solution at pH 6.0 and 8.0, 0.1 mol L -1 sodium hydroxide solution and 0.1 mol L -1 acetate buffer solution.Indigo carmine (Aldrich Company Inc.) and copper stock so-lutions (1x10 -2 mol L -1 ) were prepared from the dry pure substances in water.
Measurements of pH were made using a Micronal pH meter B222 model with a Micronal combined pH reference electrode.Except when stated otherwise, all the experiments were done at room temperature (25 ºC).Experiments involving controlled temperature were carried out in an ultra-thermostatic bath (Nova Técnica, Brazil).
Spectrophotometric measurements were performed on a Hewlett Packard 8453 spectrophotometer operating from 190 -1000 nm using a quartz cell of 1 mL with an optical path of 1 cm.Different volumes of indigo carmine stock solution were transferred to a volumetric flask of 10 mL containing an appropriate buffer solution to which copper or cooper-gluconate solution was added, and aliquot of the final solution was placed in the spectrophotometric cell.

Spectrophotometric behavior of Cu(II)/indigo carmine
The interaction between indigo carmine and Cu(II) ions was followed in the pH range 2-12, by monitoring changes in the absorption spectrum over the range from 800 to 190 nm.The optimal conditions of the formation of the complex ions were also evaluated.
Figure 2 shows the absorption spectra (in Absorbance, A) for 1x10 -5 mol L -1 of indigo carmine (A) in aqueous solution, for 1x10 -3 mol L -1 Cu (II) in aqueous solution and also for mixture of both in aqueous solutions (C).The absorption band of indigo carmine (Figure 2A) shows a maximum absorption at 615, 320 and 270 nm, attributed to the indigoide group present as chromophore center (Gordon et al., 1983).Selecting 615 nm as the measuring wavelength a linear relationship was obtained from 1x10 -5 to 1x10 -4 mol L -1 , using the equation: A = 0.00901 + 0.202x10 5 C (C = mol L -1 ), r = 0.9997.Copper (II) exhibits two characteristic bands at 300 nm and 807 nm only for concentrations increasing from 1x10 -3 to 7.5x10 -2 mol L -1 (Figure 3B).The band at 300 nm increased linearly with concentration of Cu(II) in agreement with the equation: A = -0.00207+ 15.52 C (C = mol L -1 ), r = 0.9923.In the UV-Vis spectrum (Figure 3C) obtained for the product of the reaction between indigo carmine and Cu(II) ion, both in aqueous solutions, shows a broad absorption band with maximum absorbance at 715 nm that appears concomitantly with significant decreases in the absorption band of the indigo carmine (615 nm).As Cu(II) or indigo carmine does not absorb in this range of wavelengths, this behavior is indicative that there is formation of a complex between indigo carmine and Cu(II).

Optimum conditions for complex formation
The influence of pH on the formation of yielding complex was followed by recording the UV-Vis spectrum of the reaction mixture between 1x10 -5 mol L -1 IC and 1x10 -4 mol L -1 Cu(II) solutions at 0.1 mol L -1 acetate buffer solution, pH 2.0 and 4.0; 0.1 mol L -1 phosphate buffer solution, pH 6.0 and 8.0; 0.1 mol L -1 carbonate buffer solution pH 10 and 0.1 mol L -1 sodium hydroxide solution at pH 12.The increased intensity of the absorption band at 715 nm (IC-copper II) and decrease at 615 nm (IC) at different pH values was followed by measuring of the absorbance, which was shown in Figure 3.   Indigo carmine showed no change in UV spectrum during 180 min of analysis at pH 2-12.But in the presence of Cu(II) and pH ≥ 8 a marked decrease in the absorbance at 615 nm was seen, concomitant to the increase of absorption band at 715 nm.Under this condition, Cu(OH) 2 will start to precipitate in aqueous solution under alkaline conditions, but this did not occur in the presence of indigo carmine.This fact confirms the existence of soluble complex ions between Cu(II) and indigo carmine under alkaline conditions.
The stability of the resulting complex formed under alkaline conditions was investigated by testing a mixture solution of indigo carmine and Cu(II) in equimolar concentrations of 1x10 -4 mol L -1 at 0.1 mol L -1 carbonate buffer solution at pH 10, whose recording spectra were recorded every 10 min.The results are shown in Figure 4.The complex reached maximum absorption after 5 min of preparing the reaction solution and remained stable for 3 hours at 25 o C. The absorbance of indigo carmine at 615 nm also decreased in function of time and reached maximum decrease after 20 min.
The influence of temperature on the formed complex was investigated by keeping the mixture of 1x10 -4 mol L -1 IC and Cu(II) solutions for a controlled period of 10 min under controlled temperatures of 25, 35, 45 and 55 o C. The solution was then cooled and the absorbance measured at 715 nm.The results indicated that the complex did not exhibit significant changes up to 35 o C, but absorbance is 58% lower after 10 min under 55 o C, indicating that there is a great influence by temperature, probably through shifting the equilibrium toward formation of the hydroxy complex.
The metal-to-ligand ratio in the complex was determined by spectrophotometric measurements of the solutions having 1x10 -4 mol L -1 indigo carmine and different initial metal concentrations from 1x10 -5 to 1x10 -4 mol L -1 Cu(II) solutions, with the respective plot depicted in Figure 5.The results indicated the formation of a complex with a 2:1 metal-to-ligand ratio, which suggests the formation of complex species (Cu) 2 IC.Using the stoichiometric ratio, the stability constant was determined by the molar ratios method (Bulatov, 1986) at pH 10.The values of log K= 5.75 indicated that a stable complex was formed between Cu(II) and IC.
The significant influence of the sequence of the added reagents on complex formation was observed during spectrophotometric measurements.A qualitative difference in complexation was evident when the sample containing ligand concentration in buffer solution at pH 10 received the copper solution, indicating that hydroxy complex are avoided and that the complex metal-indigo carmine is preponderant.

Beer's Law and sensitivity
A calibration graph for the determination of copper was prepared under optimum experimental conditions (0.1 mol L -1 carbonate buffer solution, pH 10, indigo carmine concentration of 1.0x10 -4 mol L -1 , reaction time = 20 min and temperature = 25 o C).The respective spectra obtained for Cu(II) concentrations ranging from 1x10 -5 to 1x10 -4 mol L -1 is shown in Figure 6.
The absorbance of indigo carmine decreased in function of Cu(II) addition, concomitant to an increase of an extra peak at 715 nm attributed to the Cu(II) complex, showing an isobestic point at 650 nm.Beer's law is obeyed within a range of 1x10 -5 to 5x10 -5 mol L -1 of copper measuring absorbance at 715 nm, as shown in the graphs of Figure 7.The peak of indigo carmine also decreased linearly up to 5x10 -5 mol L -1 , and reached a plateau at higher values of Cu (II).A wide linear relationship is obtained for lower concentrations of indigo carmine.For an indigo carmine concentration of 1x10 -5 mol L -1 in 0.1 mol L -1 carbonate buffer solution at pH 10 it is possible to produce a calibration graph represented by the following linear regression equation: A= 0.00103 + 1.15 x 10 3 C (C = mol L -1 ), r = 0.9991, n = 9.The molar absorptivity is 1.17x104 mol L -1 cm -1 .The detection limit obtained by measurement at 715 nm (three times the standard deviation of the intercept/slope) is FIGURE 5 -Plots of absorbance of complex versus molar ratio of components: 1x10 -4 mol L -1 indigo carmine and 1x10 -6 to 1x10 -4 mol L -1 Cu(II) solutions.
FIGURE 7 -Effect of Cu(II) on the absorbance of maximum absorbance of (A) 1x10 -4 mol L -1 indigo carmine solution in 0.1 mol L -1 carbonate buffer solution at pH 10 (l = 615 nm) and (B) the complex at 715 nm.
4.28x10 -7 mol L -1 .The precision of the method was checked by testing five replicate measurements using solution containing 1x10 -4 mol L -1 of indigo carmine and 1x10 -5 mol L -1 Cu(II) solution.The relative standard deviation is 1.3%.

Spectrophotometric behavior of other essential metals and indigo carmine
Taking into consideration that indigo carmine can complex Cu(II) and form a stable product in alkaline medium, the proposed methodology was also tested in the sequestering action of indigo carmine in relation to essential metals including: Zn(II), Ca(II), K(I), Co(II), Ni(II), Fe(III), Fe(II), Pb(II), Cd(II).All these cations are highly relevant to metabolic equilibrium in human beings as essential micronutrients or due to toxicity effect.
To study the effect of various cations on the reaction with indigo carmine, a mixture of 1x10 -4 mol L -1 indigo carmine with 0.1 mol L -1 carbonate buffer solution at pH 10 and variable concentration of metals from 1x10 -5 to 1x10 -3 mol L -1 were examined.The respective spectra obtained for Ni(II) is shown in Figure 8.The spectrophotometric behavior was very similar to that obtained for Cu(II), except for a maximum absorbance of the complex which occurs at 760 nm.The absorbance at 760 nm increases with Ni(II), but a linear relationship is observed only from 1.0 x 10 -5 to 4.0 x 10 -5 mol L -1 .The linear relationship is represented by the equation: A = 0.112 + 1022 C (C= mol L −1 ), r = 0.9994, n = 7.
At higher concentrations of Ni(II) a deviation from linearity is observed indicating saturation of the complexing reaction.UV-Vis spectra obtained for Co(II) in the presence of indigo carmine in 0.1 mol L -1 carbonate buffer at pH 10 also exhibits the same morphology, marked by the rising absorbance peak at 730 nm.This peak has a lower intensity compared to Cu(II) or Ni(II), but also increases linearly with Co(II) concentration from 5x10 -5 mol L -1 up to 6x10 -4 mol L -1 , following the equation: A = 0.0409 + 1064.8C (C = mol l -1 ), r = 0.9777, n = 7.
The effect of the Zn(II) ions on the complex formation is less accentuated than for all metals described previously, but it is also possible to see the extra band at 713 nm at concentration of Zn(II) higher than 7.0x10 -5 mol L -1 using an indigo carmine concentration of 1x10 -4 mol L -1 .The absorbance increased linearly from 7x10 -5 to 7x10 -4 mol L -1 following the equation: A = 0.0589 + 112.8 C (C = mol L -1 ), r = 0.998, n = 5.
Nevertheless, addition of Pb(II), K(I), Cd(II), Fe(III), Fe(II) and Ca(II) ions do not promote any alte-ration in the original spectra recorded for 1x10 -4 mol L -1 of indigo carmine, suggesting that there is no interaction between the food dye and these metallic compounds.
The metal-to-ligand ratio in the complexation of IC with Ni(II), Co(II) and Zn (II) ions determined by spectrophotometric measurements of the solutions containing 1x10 -4 mol L -1 indigo carmine and different initial metal concentrations, presented the typical formation of a complex with a 2:1 metal-to-ligand ratio as verified for Cu(II).In addition, the stability constant determined by the molar ratio method, presented values of log K ranging from 5.00, 4.89 and 3.89 for Ni(II), Co(II) and Zn(II), respectively.These results indicated that stable complexes are formed with food dye.

Copper determination in pharmaceutical formulations
In order to investigate the interference of indigo carmine on copper source administered as a pharmaceutical formulation, its influence was tested in the available complex sample commercialized as cooper-gluconate.The solution is administered to treat symptoms of copper deficiency such as: hypercolesterolemy, anemia, leucopeny, bones demineralization, blood vessel fragility and other nutrition problems.
In order to test the analytical application of this method, the matrix effect was evaluated by additionrecovery experiments carried out using the commercial pharmaceutical formulation.The results obtained by the standard addition method indicated a mean of recoveries ranging from 94.8 to 102.3% of copper using the proposed procedure for testing a triplicate sample.
The proposed procedure was confirmed by spectrophotometric measurement of Oligosol which presented maximum absorbance at 283 nm.The copper content was determined by the standard addition method and compared with the proposed method using indigo carmine.The content of copper calculated by both methods is in agreement, as well as the label values described by the industrial laboratory.The results obtained were also compared by applying the F -test and t -test at 95% confidence level (Miller et al., 1993).In both cases the calculated F or t values did not exceed the theoretical values (F 3.3 = 9.28; t 6 = 2.45), confirming that there are no significant differences between the results obtained by the two procedures.

CONCLUSIONS
The interaction between Cu(II) ions and indigo carmine in an alkaline medium resulted in the formation of Cu 2 (IC) complex.Spectrophotometric methods were used to reveal that the stoichiometric ratio between indigo carmine and copper is 2:1.The molar absorptivity was 1.17x10 4 mol L -1 cm -1 at 715 nm and the stability constant of the complex log K = 5.75, calculated at pH 10, were obtained by spectrophotometric data.Complex of indigo carmine with Ni(II), Co(II) and Zn(II) ions in 0.1 mol L -1 carbonate buffer solution at pH 10 are also formed, which is potential interference reaction in the method.The reported method has been successfully tested for determination of copper in pharmaceutical compounds based on copper-gluconate without expected interferences caused by excipients.

FIGURE 2 -
FIGURE 2 -UV-Vis spectra recorded for 1x10 -5 mol L -1 of indigo carmine in aqueous solution (A), for 1x10 -3 mol L -1 aqueous solution of Cu(II) (B) and after 10 min of mixture of both in aqueous medium (C).