Gd-DTPA

Determination of the Protonation Constant (beta - 111) of the Magnetic Resonance Imaging (MRI) contrast reagent Gadolinium (III) diethyltriaminepentaacetic acid (DTPA) by Inductively Coupled Plasma atomic emission spectroscopy (ICP-AES).

Thomas Manning*, Warren Fiskus, Michael Mitchell, and Leigh Ann Dees
Department of Chemistry
Valdosta State University
Valdosta, Georgia 31698

Submitted to Applied Spectroscopy

ABSTRACT

The first and second protonation constants (log 111, log 121) of the metal-ligand complex Gadolinium (III) - diethyltriaminepentaacetic acid (DTPA) have been determined to be 2.27 and 1.52 at 25oC and I = 0.1. A technique in which the three complexes (H2Gd-DTPA, HGd-DTPA-1, Gd-DTPA-2) are separated according to their relative polarity is utilized. An inductively coupled plasma-atomic emission spectrometer (ICP-AES) is used to determine the relative concentrations of each species by monitoring the gadolinium concentration.

INTRODUCTION

The thermodynamics ( G, H, S) of the lanthanides with aminocarboxylates such as NTA, EDTA, HEDTA, and DTPA have been studied for several decades1,2,3. The lanthanide-aminocarboxylate complex [Gd(DPTA)-2] has drawn attention from the medical community because of its role as a contrast reagent in magnetic resonance imaging (MRI)4,5. The stability constant (log[ML]/[M][L]) of the complex [Gd(DPTA)(H2O)-2] is measured in aqueous solutions to be 22.46,7. The free DTPA complex has five carboxylates and three amine groups capable of being protonated. When complexed to Gd+3, protonation of the complex is possible and will cause the complex to become unstable in very acidic mediums. Research has shown that there exists a strong correlation between the number of amines incorporated in the amino carboxylate and the enthalpy of the metal ligand complexation8. Other studies have shown that the carboxylate arms on molecules such as EDTA and DTPA exchange rapidly between free and bound states when complexed to cations9. These data would indicate that the carboxylates are being protonated when the metal-ligand complex complexes [H+]. The first protonation (log 111) of the [Gd(DPTA)-2] has been measured to be 2.356 at 37oC and I=0.15. In this work, we utilized a solvent extraction approach that differentiated between the relatively nonpolar H2GdDTPA, and the progressively more polar HGdDTPA- and GdDTPA-2. We used atomic emission spectroscopy to measure the concentration of gadolinium. We were able to measure both log 010 and log 020 for this important complex.

EXPERIMENTAL

A 0.01 M solution of Gadolinium (III) DTPA was made from the salt (Aldrich, Diethylenetriamine-pentaacetic acid, gadolinium (III) dihydrogen salt hydrate, Lot # 10114DF, TG) and had its pH adjusted with 0.1 M HCl (Aldrich, 0.0997N, Lot# 04029EZ). An Accumet pH meter was used to measure the [H+] and a 0.1 M NaOH solution was used in the titration. After each addition of NaOH, a 1 mL aliquot was removed and mixed with 10 mLs of 1-butanol (Fisher Chemical, Lot no. 952047). The solutions were shaken an allowed to sit for 5 days. There were 25 aliquots extracted over the pH range 1.4 to 3.3. A Perkin Elmer Optima 3000 DV ICP-AES was used to measure the relative gadolinium concentrations. The gadolinium wavelengths monitored were 342.247 and 335.047 nm.

RESULTS and DISCUSSIONS

Measuring the first two protonation constants of aqueous phase GdDTPA-2 involves three species: GdDTPA-2, HGdDTPA-1, H2GdDTPA. The first species, with a -2 charge, is relatively polar. The second species, with a -1 charge, is less polar than the first species but more polar than the last species, which is neutral.

In this experiment, we start by acidifing the solution, which predominantly gives H2GdDTPA. Sodium hydroxide is added in small increments, making the solution less acidic and transforming the complex from a charged to an uncharged species. At each titration interval, a 1 milliliter aliquot of the solution is drawn off and mixed with 10 milliliters of n-butanol resulting in two layers. Calculations of the [GdDTPA] in solution are adjusted accordingly. n-butanol has the alcohol functional group which provides enough polarity to attract any of the three forms of the water soluble GdDTPA complex. The alcohol also has the nonpolar butyl group which offers some discrimination between the metal-ligand complexes of different polarities. We postulated that the less polar H2GdDTPA should extract at a higher rate from the aqueous solution to the butanol while the charged GdDTPA-2 will extract less efficiently. Titration intervals are small enough to allow a meaningful curve consisting of the relative concentration of Gd verses pH plot to be constructed. We should note that the cation (Na+) is extracted into the alcohol at inverse level to the Gd extraction. We assume that the sodium ion forms a weak, outer sphere electrostatic attraction with the charged GdDTPA complexes (i.e. Na2.Gd.DTPA, Na.H.GD.DTPA). When the titration was rerun with K+ the Gd levels shifted slightly but the trends remained the same. In the butanol we measure the relative concentration of gadolinium by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). The changes in relative concentration of gadolinium in the butanol can be correlated to the protonation of the GdDTPA complex. In figures 1a and 1b show the relative gadolinium concentrations measured by ICP-AES at two wavelengths as the pH increases. The y-axis is in relative log(Gd emission intensity) units. The two pKa's measured are indicated by arrows. Recall that the pKa = pH when the ratio of the protonated to deprotonated species is 1.0. A first derivative operation was performed and the pKa's were determined to be 2.27 (+0.1) and 1.52 (+0.1), respectively in an ionic strength of 0.1 (NaCl) and at a temperature of 25oC.

CONCLUSION

The protonation constants of an biomedically important metal-ligand complex were measured coupling a solvent extraction procedure and atomic emission technique. Using strictly potentiometric data to determine protonation constants of multiprotic molecules and complexes often results in the need for detailed statistical analysis. The results of statistical analysis which involve various approaches to best fits can vary from computer program to computer program and from lab to lab.

REFERENCES

1. R.C. Vickery, J. Chem. Soc. 1895 (1952)
2. E.J. Wheelwright et al., J. Amer. Chem. Soc. 75, 4196 (1953)
3. R. Harder, S. Chaberek, J. Inorg. Nucl. Chem., 11, 197 (1959)
4. R. Lauffer, Chem. Rev. 87, 901 (1987)
5. F. Wehrli, Physics Today, 34, (1992)
6. A. Martell, R. Smith, Critical Stability Constants; Plenum: New York, 1974
7. Micskei, K.; Helm, L.; Brucher, E. ; Merbach, A. E. Inorganic Chemistry., 3844 (1993)
8. G. Choppin, G., Goedken, M., Gritmon, T., J. Inorg. Nucl. Chem., 39, 2025 (1977)
9. R. Day, C. Reilley, Analytical Chemistry, 36, 1073 (1964)

FIGURES

Figure 1a,b: The relative emission intensity (log(I)) of gadolinium extracted into the n-butanol as a function of different aqueuous phase pH's. As the metal-ligand complex protonates and becomes less polar, more Gd-DTPA extracts into the n-butanol increasing the Gd concnetration.

ACKNOWLEDGEMENTS

Valdosta State University and the state of Georgia Technology iniative are gratefully acknowledged for the use of equipment and the purchase of the ICP-AES unit.