Autores
Santos, V.F. (UNIVERSIDADE FEDERAL DO RIO DE JANEIRO) ; Carneiro, G.R.A. (UNIVERSIDADE FEDERAL DO RIO DE JANEIRO) ; Coelho, M.C.C. (UNIVERSIDADE FEDERAL DO RIO DE JANEIRO) ; Machado, S.P. (UNIVERSIDADE FEDERAL DO RIO DE JANEIRO) ; Pereira, H.M.G. (UNIVERSIDADE FEDERAL DO RIO DE JANEIRO)
Resumo
All World Anti-Doping Agency (WADA) accredited laboratories must follow a List
of Prohibited Substances and Methods, updated every year. Some analytical
strategies are developed to cover most of the list. However, the analysis of
cobalt is yet a challenge because of its incompatibility with the classical
strategies available. Thus, the project aims to develop a cost-effective
approach for the detection of cobalt in athletes’ urine employing liquid
chromatography coupled with mass spectrometry (LC-MS). The adopted strategy for
achieving the goal was the synthesis and characterization of a cobalt complex
with diethyldithiocarbamate and further application of Design of Experiments in
different steps of the analysis. The method developed will be validated
according to the WADA guidelines.
Palavras chaves
Anti-doping; Cobalt; Design of Experiments
Introdução
1. Anti-doping control in sports: The anti-doping control in sports is an
interdisciplinary science that includes different areas of knowledge, such as
Chemistry, Biology, Pharmacy and Toxicology, to monitor a growing number of
substances and methods capable of promoting performance enhancement in athletes
(PEREIRA, 2015). Such substances have different physicochemical characteristics
and pharmacological activities, with the aim of providing the sports community
with a range of options to circumvent the anti-doping system (THEVIS and
SCHÄNZER, 2014). The fight against the use of doping agents in sports has been
constantly improving since anti-doping control began in the 1960s (BADOUD et
al., 2011). In this context, since 2004, WADA publishes annually a List of
Prohibited Substances and Methods, which includes hundreds of substances
prohibited in and out of competition or only in competition. These are
classified into nine classes (S1 to S9), a group of analytes prohibited only in
specific sports (beta-blockers, class P1), and three prohibited methods (i.e.,
manipulation of blood and its components, chemical and physical manipulation,
and genetic doping). The list also includes the S0 class, which englobes drugs
under development or discontinued, design drugs or substances for veterinary
use, many of which do not have well-established analytical targets (WADA, 2022).
Therefore, to cover the diversity of substances in the list, several analytical
strategies are required to detect them in biological matrices. However, the
request for more sensitive and specific methods to detect an increasing number
of substances continues to grow (BADOUD et al., 2011). 2. Use of cobalt in
sports: Cobalt was incorporated into the WADA Prohibited List in 2015 (WADA,
2015), although the use of cyanocobalamin (vitamin B12) is permitted (WADA,
2014). The supplementation with cobalt salts stimulates erythropoiesis through
the stabilization of hypoxia-inducible factor (HIF) (LIPPI et al., 2005; THEVIS
and SCHÄNZER, 2014). An exposure to 120 or 150 mg/day of cobalt chloride, for
example, results in the development of polycythemia, with a substantial increase
in hematocrit and hemoglobin up to 20% above the pretreatment levels (LIPPI et
al., 2006). This allows the muscles to become more resistant, which promotes the
increase of physical performance of athletes (LIPPI et al., 2005). In addition
to its effect on erythropoiesis, cobalt chloride supplementation exhibits
beneficial effects on protein biosynthesis (KRUG et al., 2014), on lipid and
glucose metabolism parameters (SIMONSEN et al., 2011), and prevents oxidative
stress induced by high altitudes (SIMONSEN et al., 2012). In 2015, Thevis et al.
reported, through Inductively Coupled Plasma-Mass Spectrometry (ICP-MS)
analysis, significant amounts of undeclared cobalt and nickel in products sold
as erythropoiesis-stimulating agents (THEVIS et al., 2015). Nickel, like cobalt,
is also known to induce hypoxia, but is not on the WADA Prohibited List (MAXWELL
and SALNIKOW, 2004). 3. Motivation: The detection of cobalt in human urine for
anti-doping control purposes is yet a challenge due to its incompatibility with
the classical analytical strategies available in accredited laboratories around
the world. To achieve this goal through liquid chromatography coupled with high
resolution mass spectrometry (LC-HRMS), an analytical technique widely used in
the area, it is necessary the complexation of cobalt from the matrix with a
ligand. In this way, the electrospray ionization (ESI) of diethyldithiocarbamate
(DTC, (C2H5)2NCSS-) complexes with transition metal ions has been known for over
twenty years (SCHOENER et al., 1999; ROSS et al., 2000) and was the strategy
adopted for method development. Thus, the goals of the project are to (1)
synthesize the cobalt complex with the DTC ligand, (2) characterize it by
spectrophotometry in the Ultraviolet-Visible (UV-Vis) region, Density Functional
Theory (DFT) and MS, and (3) to optimize experimental conditions of sample
preparation and instrumental analysis through Design of Experiments.
Material e métodos
1. Synthesis and characterization of tris(diethyldithiocarbamate)cobalt(III)
[Co(DTC)3]: The synthesis of [Co(DTC)3] complex was based on Eagle et al (EAGLE
et al., 1999). Briefly, a solution of NaDTC.3H2O in ethanol:water (1:1) was
added, slowly and with stirring, to a solution of cobalt(II) acetate
tetrahydrate [Co(CH3COO)2.4H2O], with a ratio of 4 NaDTC (in excess) to 1
Co(II). O2 was added to the green precipitate for 10 minutes to ensure a
complete oxidation process. The [Co(DTC)3] complex was filtered, washed with
ethanol and water, and recrystallized in chloroform. Calculations employing the
Density Functional Theory (DFT) and the Time Dependent Density Functional Theory
(TD-DFT) were performed with Gaussian09 for the prediction of structural and
spectroscopic properties of [Co(DTC)3] complex. The experimental geometry was
obtained from the X-Ray structure available in Cambridge Structural Database
(CSD refcode: ETDCCO02). Different functionals and basis set were employed to
validate the best methodology for both geometry optimization and simulation of
the complex electronic absorption spectra. The computational results were
confirmed experimentally with a spectrophotometer in the UV-Vis region. 2.
Screening and modeling experimental designs - optimization of the experimental
set-up: The previous synthetized and characterized [Co(DTC)3] complex was spiked
in a blank urine at a concentration of 200 ng mL-1 and was utilized for the
optimization of ESI source, MS and liquid-liquid extraction (LLE) conditions.
For each of these steps, a multivariate screening and modeling experimental
designs were performed to determine the most important variables and the robust
experimental region, respectively. After the establishment of the instrumental
and extraction set-up, the same methodology was followed for the complexation in
situ in urine of spiked cobalt at a concentration of 200 ng mL-1. The optimized
conditions were used throughout the work. 3. Systematic evaluation of the LC-
HRMS mobile phase composition: Cobalt was spiked to water (2.00 mL) and urine
(2.00 mL). DTC and citric acid were added, and the mixture was shaken at 300 rpm
for 10 minutes. Liquid-liquid extraction with tert-butyl methyl ether (TBME) was
performed, the mixture was shaken at 300 rpm for 5 minutes and centrifuged at
3000 rpm for 10 minutes. The organic layer was transferred to another tube and
evaporated to dryness in a thermostatic bath under a nitrogen stream at 40°C.
The samples were reconstituted with 30 µL of mobile phase A and 70 µL of mobile
phase B. The experimental design for evaluation of the final mobile phase
composition resulted in eight possible combinations between mobile phases A
(i.e. pH 3, pH 4, pH 5 and pH 6) and B (i.e. methanol and acetonitrile). A
Thermo Dionex Ultimate 3000 UHPLC system coupled to a QExactive hybrid
quadrupole-orbitrap mass spectrometer equipped with an ESI source was used.
Resultado e discussão
1. Characterization of the [Co(DTC)3] - DFT, TD-DFT and UV-Vis
spectrophotometry: Due to the d6 configuration of Co(III), electrons can occupy
orbitals in two ways according to the Pauli Exclusion Principle: (1) by filling
all orbitals before electron pairing, or (2) by pairing all electrons before
filling all the orbitals. Situation 1 results in a high spin complex and
situation 2 in a low spin complex. However, for complexes with d4 to d7
configuration, the phenomenon known as “Spin Crossover” (SCO) may occur. Such
complexes undergo a change in their multiplicity (High Spin to Low Spin or Low
Spin to High Spin) when stimulated by external factors (e.g. variation in
temperature, pressure and light irradiation). This phenomenon was first observed
by Cambi and collaborators in 1931 for the tris(N,N-dialkyldithiocarbamate)
iron-(III) complex (POUGY and MACHADO, 2020), from the same family as the one
proposed for the detection of cobalt in urine. The initial structure for the
[Co(DTC)3] complex was obtained from the CSD database, with the code ETDCCO02
(HEALY et al., 1990). The functionals B3LYP, CAM-B3LYP, WB97XD and PBEPBE and
three basis sets were tested to find the best methodology to describe both
geometry and electronic absorption spectra of the studied complex. From these
results, CAM-B3LYP with 6-31G** proved to be the most accurate methodology and
was used throughout this work. Geometry optimization in chloroform showed little
increase in bond lengths compared to experimental data. On the other hand, to
assess whether the complex was low or high spin, the multiplicity was evaluated:
singlet for low spin or quintet for high spin. The energies and bond lengths for
the complex were compared to the experimental data obtained from the CSD after
geometry optimization and both indicated that the complex is low spin (singlet).
The bond lengths for the singlet complex presented a relative error of up to 3%
in relation to the experimental data and, for the quintet, this error increased
to up to 11%. Also, the quintet presented an energy 16.5 kcal mol-1 higher than
the singlet structure, which indicated that no SCO is observed for this complex
(POUGY and MACHADO, 2020). Comparing the simulated and experimental UV-Vis
spectra, there was a great similarity between them, which validates the
calculation methodology for the complex. 2. Screening and modeling experimental
designs - optimization of the experimental set-up: When it is necessary to
optimize a method that involves several factors, a possible optimization
strategy is the multivariate one, through Design of Experiments (DoE). In this
strategy, as a first step in the selection of the most important factors (i.e.,
those that most affect the response), a screening design is applied. Then, a
modeling design is carried out to elaborate a response surface to verify the
most robust experimental region (i.e., region in which small variations in the
factor can occur without any significant change in the response). In a screening
design, each factor is investigated using fixed levels, which can take on two
values in a two-level design: a high level (1) and a low level (-1). The number
of experiments will be given by 2^k, where k is the number of design factors.
Thus, the number of experiments grows geometrically with the increase in the
number of factors, which guides the type of screening design (e.g. Simple
Factorial, Fractional Factorial, Plackett-Burman). The greater the number of
factors, more approximations must be carried out. From this initial selection,
the most important variables are studied at more levels to develop a response
surface and find the robust experimental region. Therefore, the strategy adopted
in the elaboration of the method for the detection of cobalt in urine by LC-HRMS
was to separate the different stages of preparation and instrumental analysis to
optimize the factors of each stage. After optimization of one stage, the factors
were fixed at the optimum values for the next optimization. 3. Systematic
evaluation of the LC-HRMS mobile phase composition: Due to problems in retention
time reproducibility and to the fact that the [Co(DTC)3] complex was eluting in
the return of the gradient to the initial chromatographic condition (5% methanol
and 95% water pH 3), a systematic study involving the variation of both mobile
phases A and B has been proposed. For this optimization, an isocratic
chromatographic run was performed to neglect a possible effect of the apparent
pH change due to a gradient run. It was observed a variation in the retention
time of the complex as a function of the pH of mobile phase A, which was much
more pronounced when methanol was used as mobile phase B. For this reason, the
final mobile phases were changed for water pH 4 (mobile phase A) and
acetonitrile (mobile phase B). After such change, there was no need for a new
optimization of the instrumental set-up and the data were much more robust with
small changes of mobile phase composition.
Conclusões
As a conclusion, some of the critical steps in a method development for detection
of cobalt in urine by LC-HRMS was overcome by previous characterization of a
synthetized tris(diethyldithiocarbamate)cobalt(III) complex and further
application of Design of Experiments. The geometry and electronic absorption
spectra simulated by DFT and TD-DFT was confirmed experimentally with ultraviolet-
visible spectrophotometry. After characterization, the synthetized material was
used for optimization of the ESI source, the mass spectrometer, and the extraction
procedure. The final steps of optimization englobed the in situ complexation with
spiked cobalt in urine and the systematic study of mobile phase composition. The
method will be validated according to the WADA guidelines for further application
in anti-doping control.
Agradecimentos
The authors thank CNPq, FAPERJ and Autoridade Brasileira de Controle de Dopagem
(ABCD) for financial support.
Referências
BADOUD, F.; GUILLARME, D.; BOCCARD, J.; GRATA, E.; SAUGY, M.; RUDAZ, S.; VEUTHEY, J. C. Analytical aspects in doping control: Challenges and perspectives. Forensic Science International, no 213, 49–61, 2011. EAGLE, C. T.; FARRAR, D. G.; HOLDER, G. N.; GOODEN, D. M.; GOODMAN, A. B.; WYATT, S. W. Synthesis and Characterization of Tris (diethyldithiocarbamato) cobalt (III) as an Undergraduate Inorganic Laboratory. The Chemical Educator, no 4, 105-107, 1999. KRUG, O.; KUTSCHER, D.; PIPER, T.; GEYER, H.; SCHÄNZER, W.; THEVIS, M. Quantifying cobalt in doping control urine samples–a pilot study. Drug Test. Analysis, no 6, 1186-1190, 2014. LIPPI, G.; FRANCHINI, M.; GUIDI, G. C. Cobalt chloride administration in athletes: a new perspective in blood doping? British journal of sports medicine, no 39, 872-873, 2005. LIPPI, G.; FRANCHINI, M.; GUIDI, G. C. Blood doping by cobalt. Should we measure cobalt in athletes? Journal of Occupational Medicine and Toxicology, no 1, 18, 2006. HEALY, P. C.; CONNOR, J. W.; SKELTON, B. W.; WHITE, A. H. Alkyl substituent effects in diamagnetic dithiocarbamate cobalt (III) and nickel (II) complexes. Australian Journal of Chemistry, no 43, 1083-1095, 1990. MAXWELL P.; SALNIKOW K. HIF-1 – an oxygen and metal responsive transcription factor. Cancer Biol Ther, no 3, 29–35, 2004. PEREIRA, H. Dopagem no Desporto. In: DINIS-OLIVEIRA, R.J.; CARVALHO, F.D.; BASTOS, M.L. Toxicologia Forense, 375 – 393, 2015. POUGY, K.C.; MACHADO, S.P. Uso da Teoria do Funcional de Densidade na análise de Spin Crossover em aulas de química inorgânica. Química Nova, no 43, 127-130, 2020. ROSS, A.R.; IKONOMOU, M.G.; ORIANS, K.J. Electrospray ionization of alkali and alkaline earth metal species. Electrochemical oxidation and pH effects. J Mass Spectrom., no 35, 981-989, 2000. SCHOENER, D.F.; OLSEN, M.A.; CUMMINGS, P.G.; BASIC, C. Electrospray ionization of neutral metal dithiocarbamate complexes using in‐source oxidation. J Mass Spectrom., no 34, 1069-1078, 1999. SIMONSEN, L. O.; BROWN, A. M.; HARBAK, H.; KRISTENSEN, B. I.; BENNEKOU, P. Cobalt uptake and binding in human red blood cells. Blood Cells, Molecules, and Diseases, no 46, 266-276, 2011. SIMONSEN, L. O.; HARBAK, H.; BENNEKOU, P. Cobalt metabolism and toxicology—a brief update. Science of the Total Environment, no 432, 210-215, 2012. THEVIS, M.; KRUG, O.; PIPER, T.; GEYER, H.; SCHÄNZER, W. Solutions advertised as erythropoiesis-stimulating products were found to contain undeclared cobalt and nickel species. Int J Sports Med., no 36, 82-84, 2015. THEVIS, M.; SCHÄNZER, W. Analytical approaches for the detection of emerging therapeutics and non-approved drugs in human doping controls. Journal of Pharmaceutical and Biomedical Analysis, no 101, 66-83, 2014. WORLD ANTI-DOPING AGENCY (WADA). 2015 Prohibited list - Summary of Major Modifications and Explanatory Notes. Access in: June 20th, 2022. Available at https://www.wada-ama.org/sites/default/files/resources/files/wada-2015-prohibited-list-summary-of-modifications-en.pdf. 2014. WORLD ANTI-DOPING AGENCY (WADA). 2015 Prohibited list. Access in: June 20th, 2022. Available at https://www.wada-ama.org/sites/default/files/resources/files/wada-2015-prohibited-list-en.pdf. 2015. WORLD ANTI-DOPING AGENCY (WADA). Prohibited list. Access in: July 27th, 2022. Available at https://www.wada-ama.org/sites/default/files/2022-01/2022list_final_en_0.pdf. 2022.
