Anal. Chem. 2005, 77, 7724-7734
Species-Specific Isotope Dilution Analysis and Isotope Pattern Deconvolution for Butyltin Compounds Metabolism Investigations Pablo Rodrı´guez-Gonza´lez, Andre´s Rodrı´guez-Cea, J. Ignacio Garcı´a Alonso,* and Alfredo Sanz-Medel
Department of Physical and Analytical Chemistry, Faculty of Chemistry, University of Oviedo. Julia´ n Claverı´a 8, 33006 Oviedo, Spain
A methodology for the study of the absorption and metabolism of butyltin compounds in laboratory animals using isotopically enriched species was developed. The method is based on the oral administration of 119Snlabeled monobutyltin (MBT), 118Sn-labeled dibutyltin (DBT), and 117Sn-labeled tributyltin (TBT) to the animals and the measurement of both the concentration and isotopic composition of these compounds in the different tissues by GC-ICPMS. The degradation of butyltin compounds during their metabolism was computed using least-squares isotope pattern deconvolution, and their concentration was measured by reverse isotope dilution analysis using natural-abundance MBT, DBT, and TBT standards. Male Wistar rats were used as models to evaluate the proposed methodology. Preliminary toxicological results obtained with one rat indicate that TBT is highly absorbed (64.4%), and it is found in all organs with relatively high levels in stomach and intestines. The apparent absorption of DBT was 27.3% and was mainly found in liver, kidney, and intestines. However, a large proportion of the found DBT is formed from the degradation of TBT (∼40% of the found DBT in liver is degraded TBT). The apparent absorption of MBT was found to be 12.5%, and the originally administered MBT was mainly recovered in the feces. However, MBT was clearly detected in liver, kidney, stomach, intestines, and urine as degradation products of DBT and TBT. Although a significant variability from rat to rat is expected to be obtained, the analytical variability provided by this methodology is small enough to yield meaningful biological results. The results obtained demonstrate that the developed methodology is able to follow qualitatively, quantitatively, and simultaneously the specific metabolic pathways of different species of a given element.
The use of isotope tracers in nutritional, toxicological, and clinical sciences has permitted researchers to explore the metabolic pathways of a wide variety of elements and organic molecules. Early work has been traditionally carried out using * To whom correspondence should be addressed. E-mail:
[email protected]. Fax: +34-985-103125.
7724 Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
radiotracers with which it is possible to unambiguously determine the amount of tracer absorbed and remaining in the body by whole body counting while using minimal doses of radioisotopes.1 However, the instrumental advances in mass spectrometry, as well as the concern about exposure to ionizing radiation,2 led to a general shift away from radioisotopes and an increase in the use of stable isotopes for metabolic studies. Additionally, due to their safety, stable isotopes allow the use of multiple tracers, which increases the information content obtained in each study.3 Research in this field has been traditionally divided4 between authors employing light isotopes (such as 2H, 18O, 15N, and 13C) to study the metabolism of organic molecules such as proteins, lipids, or carbohydrates5 and those employing heavier isotopes, which is mainly focused on the metabolism of nutrient minerals6 and toxic metals. It is well known that the chemical form of a given element determines its actual toxicity to a living organism. In other words, the absorption, metabolism, accumulation, and toxicity of this element will depend on the particular chemical species ingested.7 Innovative chemical speciation strategies and methodologies are more and more demanded in the field of life sciences, since total metal analysis is not sufficient any longer to provide such information.8,9 However, despite the evidence, there is still a lack of developed methodologies able to provide specific information of the absorption, bioaccumulation, and fate of the different chemical forms of a given element throughout the metabolic system in living organisms.10 Only for selenium have metabolic studies with stable isotopes been carried out so far using different chemical species such as selenite,11-13 selenate,14 or seleno(1) Iyengar, V. Food Nutr. Bull. 2002, 23, 3-10. (2) Koletzko, B.; Demmelmair, H.; Hartl, W.; Kindermann, A.; Koletzko, S.; Sauerwald, T.; Szitanyi, P. Early Hum. Dev. 1998, 53, S77-S97. (3) Bier, D. M. Emerging Technologies for Nutrition Research; National Academy Press: Washington DC, 1997; pp 203-213. (4) Walczyk, T.; Coward, A.; Schoeller D. A.; Preston T.; Dainty J.; Turnlund J. R.; Inyegar V. Food Nutr. Bull. 2002, 23, 69-75. (5) McCabe, B. J.; Previs, S. F. Metab. Eng. 2004, 6, 25-35. (6) Patterson, K. Y.; Veillon, C. Exp. Biol. Med. 2001, 226, 271-282. (7) Fairweather-Tait, Susan J. Fresenius J. Anal. Chem. 1999, 363, 536-540. (8) Windisch, W. Anal. Bioanal. Chem. 2002, 372, 421-425. (9) Sanz-Medel, A. Spectrochim. Acta B 1998, 53, 197-211 (10) Templeton, D. M. Fresenius J. Anal. Chem. 1999, 363, 505-511. (11) Janghorbani, M.; Martin, R. F.; Kasper L. J.; Sun X. F.; Young V. R. Am. J. Clin. Nutr. 1990, 51, 670-677. (12) Suzuki, K. T.; Itoh, M. J. Chromatogr., B 1997, 692, 15-22. (13) Shiobaran, Y.; Ogra, Y.; Suzuki, K. T. Life Sci. 2000, 67, 3041-3049. 10.1021/ac051091r CCC: $30.25
© 2005 American Chemical Society Published on Web 10/25/2005
methionine.15-17 However, these studies were based on singleisotope and single-species experiments except for one case where selenite and selenate were simultaneously administered.14 Concerning other elemental species of toxicological interest, only experiments using radiotracers have been published18 including studies on methyl mercury19,20 or tributyltin (TBT),21 but those studies do not provide information on the chemical form in which the metal is found in the different organs. In any case, the simultaneous absorption, distribution, and degradation of several chemical species of a given toxic element have not been studied so far by using either stable or radioactive isotopes. The presence of butyltin compounds in dietary foods, particularly in seafood, has been widely reported throughout the years.22 However, the detection of these compounds in some plastic products and their subsequent transfer to foodstuffs23 and drinking water24 indicate that the exposure of humans and wildlife is derived not only from their use as marine antifouling agents but also from other uses. A recent study25 showed detectable levels of organotin compounds in house dust collected in houses of several European countries, indicating that these compounds can be also present in household products. Additionally, their presence has been reported in human liver26 and blood,27 but some of these results were questioned by other authors.28 On the other hand, Morabito et al.29 reported on the degradation of butyltin compounds during cooking. They found that TBT present in mussels was hardly affected by either boiling or frying, indicating that these compounds are persistent in foodstuffs. Similarly, a recent study from our laboratory using simulated gastrointestinal digestion30 showed that TBT present in mussels was recovered almost quantitatively after in vitro gastrointestinal digestion (∼90% recovery) indicating a low degradation of TBT. Additionally, most of the TBT present in the original solid sample was found in the supernatant after digestion (∼60%), indicating a high potential for absorption. For these reasons, we have considered that the metabolism of these toxic tin species deserves a careful investigation to clarify their toxic action in mammals. (14) Kobayashi, Y.; Ogra, Y.; Suzuki K. T. J. Chromatogr., B 2001, 760, 73-81. (15) Veillon, C.; Patterson, K. Y.; Button, L. N.; Sytkowski, A. J. Am. J. Clin. Nutr. 1990, 52, 155-158. (16) Mangels, A. R.; Moser-Veillon, P. B.; Patterson, K. Y.; Veillon, C. Am. J. Clin. Nutr. 1990, 52, 621-627. (17) Swanson, C. A.; Patterson, B. H.; Levander, O. A.; Veillon, C.; Taylor, P. R.; Helzlsouer, K.; McAdam, P. A.; Zech, L. A. Am. J. Clin. Nutr. 1991, 54, 917-926. (18) Cornelis, R. Analyst 1992, 117, 583-588. (19) Nielsen, J. B.; Andersen, O. Pharmacol. Toxicol. 1996, 79, 60-64. (20) Smith, J. C.; Allen, P. V.; Turner, M. D.; Most, B.; Fisher, H. L.; Hall, L. L. Toxicol. Appl. Pharmacol. 1994, 128, 251-256. (21) Rouleau, C.; Xiong, Z. H.; Pacepavicius, G.; Huang, G. L. Environ. Sci. Technol. 2003, 37, 3298-3302. (22) Hoch, M. Appl. Geochem. 2001, 16, 719-743. (23) Takahashi, S.; Mukai, H.; Tanabe, S.; Sakayama, K.; Miyazaki, T.; Masuno, H. Environ. Pollut. 1999, 106, 213-218. (24) Forsyth, D. S.; Jay, B. Appl. Organomet. Chem. 1997, 11, 551-558. (25) Greenpeace Research Laboratories Technical Note 01/2003 (GRL-TN-012003). (26) Nielsen, J. B.; Strand, J. Environ. Res. 2002, 88, 129-133. (27) Kannan, K.; Senthilkumar, K.; Giesy, J. P. Environ. Sci. Technol. 1999, 33, 1776-1779. (28) Robinson, S.; Kluck, M. Environ. Sci. Technol. 2000, 34, 1877-1879. (29) Massanisso, P.; Di Rosa, F.; Willemsen, F.; Morabito, R. Umpublished work presented at VIth International Conference on Environment and Biological Aspects of Main-Group Organometals (ICEBAMO), Pau, France, 2003. (30) Rodrı´guez-Gonza´lez, P.; Ruiz Encinar, J.; Garcı´a Alonso, J. I.; Sanz-Medel, A. Anal. Bioanal. Chem. 2005, 381, 380-387.
Metabolic studies using enriched isotopes require adequate mathematical approaches to complete and support the information provided by the experimental work. The use of mathematical modeling in nutrition has been concerned mainly with compartmental modeling, which is especially suited to mineral metabolism.31 Similarly, strategies based on the mathematics of combinatorial probabilities have been employed, including mass isotopomer distribution analysis32,33 for the quantification of precursor-product events and its application to study in vivo metabolic fluxes.34 Isotope pattern deconvolution is a mathematical technique that has been applied for the characterization of complex mass spectra containing molecular fragments with polyisotopic elements,35-38 for the elucidation of fragmentation pathways for the accurate measurement of isotope ratios from molecular clusters,39,40 or for the metabolic studies of the serineto-glycine interconversion in humans.41 In this work, we apply isotope pattern deconvolution in combination with isotope dilution analysis to follow simultaneously the degradation, tissue distribution and apparent absorption of three butyltin compounds (monobutyltin, MBT, dibutyltin, DBT, and TBT) throughout their metabolic pathways in a living organism. To the best of our knowledge, this is the first study providing specific and simultaneous metabolic information of several elemental species of a given element by administering adequate isotopically enriched species labeled with different isotopes. The advantages provided by the technique species-specific isotope dilution analysis GCICPMS30 have allowed the administration of very low doses of butyltin compounds without perturbing the system being investigated and without losing the capabilities of obtaining highly accurate and precise determinations. Although the methodology is exemplified here with the study of the metabolism of butyltin compounds in laboratory rats, it could also be extended to other compounds of polyisotopic elements of toxicological interest (such as mercury or lead).
EXPERIMENTAL SECTION Animals. Four-week-old male Wistar rats, weighing 80-100 g, were used throughout the present study. The animals were housed under 12 h dark/12 h light cycles, kept in individual metabolic cages and allowed free access to food (standard diet from PanLab, Barcelona, Spain) and tap water during the experiments. Ambient temperature during the study was maintained at 20 °C. Treatment and Sample Collection. Butyltin species were dissolved in 2% acetic acid in tap water and administered using oral gavage to the animals. In all cases, the butyltin species were administered in a single dose in the range of 2-20 µg of Sn. The (31) Dainty, J. R. Nutr. Res. Rev. 2001, 14, 295-315. (32) Hellerstein, M. K.; Neese, R. A. Am. J. Physiol. 1992, 263, E998-E1001. (33) Kelleher, J. K.; Masterson, T. M. Am. J. Physiol. 1992, 262, E118-E125. (34) Hellerstein, M. K. Metab. Eng. 2004, 6, 85-100. (35) Brauman, J. I. Anal. Chem. 1966, 38, 607-610. (36) Hilmer, R. M.; Taylor J. W. Anal. Chem. 1974, 46, 1038-1044. (37) Roussis, S. G.; Proulx, R. Anal. Chem. 2003, 75, 1470-1482. (38) Meija, J.; Caruso J. A. J. Am. Soc. Mass Spectrom. 2004, 15, 654-658. (39) Hachey, D. L.; Blais, J. C.; Klein, P. D. Anal. Chem. 1980, 52, 1131-1135. (40) Meija, J.; Centineo, G.; Garcı´a Alonso, J. I.; Sanz-Medel, A.; Caruso, J. A. J. Mass Spectrom. 2005, 40, 807-814. (41) Culea, M.; Hachey D. L. Rapid Commun. Mass Spectrom. 1995, 9, 655659.
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
7725
animals were sacrificed 24 h after the administration of the butyltin species using CO2 anesthesia and exsanguination via cardiac puncture to allow blood collection. Samples of feces, urine, whole blood, and different organs (brain, thymus, lungs, heart, spleen, testicles, kidneys, liver, stomach, and intestine) were immediately isolated or dissected, lyophilized for 72 h, and stored frozen at -18 °C until analysis. Treatment of the animals in the present study followed the ethical guidelines established by the University of Oviedo, and approval of all animal procedures was obtained by the local ethics committee. Instrumentation. A Hewlett-Packard (Palo Alto, CA) gas chromatograph model 6890, fitted with a split/splitless injector and a HP-5 capillary column (cross linked 5% phenyl methyl siloxane, 30 m × 0.32 mm × 0.25 µm thickness), was used for the separation of the organotin species. The gas chromatograph was coupled to a HP-4500 inductively coupled plasma mass spectrometer (Yokogawa Analytical Systems, Tokyo, Japan) using the laboratory-made transfer line described in detail previously.42 The lyophilization of the samples was carried out in a freeze-drier model Lyolab 3000 (Jouan, France). For the extraction of the organotin compounds from the samples, a digital control immersion thermostat model Digiterm 100 (J.P. Selecta, Barcelona, Spain) was used. For the optimization of the extraction procedure, a microwave oven, model 1200 (Milestone, Socisole, Italy), equipped with middle pressure PTFE vessels was used. The homogenization of the tissues during the optimization of the extraction procedure was carried out by means of an Ultra-Turrax T-25 (Ika Labortechnik, Staufen, Germany). The pH measurements were carried out using a pH meter model Micro pH 2000 (Crison, Barcelona, Spain). Reagents and Materials. Tributyltin chloride (95.9%) was obtained from Aldrich (Steinheim, Germany). Stock solutions were prepared by dissolving the corresponding salt in acetic acidmethanol (3:1) (Merck, Darmstadt, Germany). All organometallic standard solutions were kept in the dark at -18 °C, and diluted working solutions were prepared daily in acetic acid-methanol (3:1) when performing reverse isotope dilution analysis or in 2% acetic acid in tap water for the administration experiments. Solid 119Sn-enriched and 118Sn-enriched tin metal were purchased from Cambridge Isotope Laboratories (Andover, MA), and a solution of 117Sn-enriched tributyltin was supplied by the Laboratory of the Government Chemist (Teddington, U.K.). The single-isotope spike solution consisting of a 119Sn-enriched mixture of MBT, DBT, and TBT was provided by ISC-Science (Gijo´n, Spain). The preparation of the triple-spike (117Sn-TBT,118Sn-DBT, 119Sn-MBT) has been described in a previous publication.43 Acetic acid (Merck) and methanol (Merck) were used for the preparation of a (3:1) mixture, which was employed as extractant for the extraction of the organotin compounds from the lyophilized samples. Ethylation of the butyltin species was performed using sodium tetraethylborate (Galab, Geesthacht, Germany). Ultrapure water was obtained from a Milli-Q 185 system (Millipore, Mosheim, France). Procedures. Extraction of the Butyltin Species from the Samples. The lyophilized samples were spiked with an appropri-
ate amount of spike solution (natural abundance or isotopically enriched), and 5 mL of a mixture of acetic acid and methanol (3:1) (g/g) was added to a weighed amount of sample in 15-mL glass vials with screw caps (Supelco, Bellefonte, PA). The vials were introduced into a thermostatic bath at 37 °C for at least 2 h under mechanical shaking, left to cool to room temperature, and centrifuged at 3000 rpm for 15 min. Finally, the extract was transferred to another glass vial and ethylated as described below. For the calculation of the isotopic composition of the samples and the isotope pattern deconvolution studies, no spike was added to the sample before extraction. Ethylation, Separation, and Determination by GCICPMS. Ethylation of the tin species was carried out in 15-mL glass vials with screw caps (Supelco). The pH was adjusted to 5.4 with an appropriate volume of 1 M acetic acid/sodium acetate buffer, and ethylation was performed using 1.5 mL of 2% w/v sodium tetraethyl borate in 0.1 M NaOH. Then, 1 mL of hexane was added for liquid-liquid extraction, which was performed by manual shaking for 10 min. To break the resulting emulsions, 1 mL of hexane was subsequently added to the vial, and the sample was centrifuged at 3000 rpm for 15 min. Cleanup of the samples was found to be necessary, particularly for the most fatty tissues, and it was carried out by passing the organic layer through an alumina column as described in a previous publication.44 Finally, the organic layer was transferred to a 7-mL chromatographic vial with screw cap (Supelco) and stored at -18 °C until analysis. This final volume was preconcentrated until almost dryness under a gentle stream of nitrogen, just before the GC-ICPMS measurement. Typical operating conditions and analytical features of these determinations by GC-ICPMS have been described previously.43 Measurement of Isotope Ratios Using GC-ICPMS. Integration of the chromatographic peaks was carried out using the commercial chromatographic software supplied with the ICPMS instrument. Isotope ratios were always computed as peak area ratios. The integration time per isotope depended on the number of isotopes monitored while keeping the total integration time at 200 ms to be able to follow accurately the chromatographic peak profile.45 When using the 119Sn-enriched mixed spike solution, the isotopes selected were 118, 119, and 120. For the optimization of the extraction conditions using the triple spike, the isotopes monitored were 117, 118, 119, and 120, and for the isotope pattern deconvolution approach, the isotopes monitored were 116, 117, 118, 119, and 120. In all cases, mass bias was corrected by bracketing of a natural butyltin standard mixture of MBT, DBT, and TBT between each triplicate of samples. Triple-Spike Isotope Dilution for the Optimization of Extraction Conditions. The use of the triple-spike isotope dilution approach for the optimization of extraction conditions has been described in a previous publication.43 Briefly, this spike solution contains each butyltin species enriched with a different tin isotope: MBT is labeled with the isotope 119, DBT with 118, and TBT with 117. When this solution is added to a sample containing natural butyltin species and the mixture is fully equilibrated, it is possible to measure nine independent isotope ratios (120/117, 120/118, and 120/119 for each of the three
(42) Montes-Bayo´n, M.; Gutierrez Camblor, M.; Garcı´a Alonso, J. I.; Sanz-Medel, A. J. Anal. At. Spectrom. 1999, 14, 1317-1322. (43) Rodrı´guez-Gonza´lez, P.; Ruiz Encinar, J.; Garcı´a Alonso, J. I.; Sanz-Medel, A. J. Anal. At. Spectrom. 2004, 19, 685-691.
(44) Ceulemans, M.; Witte, C.; Lobinski, R.; Adams, F. C. Appl. Organomet. Chem. 1994, 8, 451-461. (45) Ruiz Encinar, J.; Garcı´a Alonso, J. I.; Sanz-Medel, A. J. Anal. At. Spectrom. 2000, 15, 1233-1239.
7726
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
butyltin species). These isotope ratios enable the calculation of nine unknown parameters: the original concentrations of the three butyltin species regardless of degradation and six factors corresponding to six interconversion reactions. The interconversion factors that can be calculated with this approach are three debutylation factors: degradation of TBT to DBT (F1), degradation of DBT to MBT (F2), and direct degradation of TBT to MBT (F4) as well as three butylation factors: butylation of DBT to form TBT (F7), butylation of MBT to form DBT (F8), and direct butylation of MBT to form TBT (F10). Studies carried out under different extraction conditions showed that only factors F1 and F2 needed to be considered, while the rest of the factors proved to be negligible under all experimental conditions tested.46 Single-Spike Isotope Dilution for the Studies with NaturalAbundance TBT. Once extraction conditions were optimized and no degradation of the butyltin compounds was detected, all preliminary studies in laboratory rats using natural-abundance TBT were carried out by measuring the amounts of MBT, DBT, and TBT in the different organs by isotope dilution analysis using a mixed spike solution in which all butyltin compounds were enriched in 119Sn. Details of this procedure have been given in previous publications.47 Isotope Pattern Deconvolution for the Study of the Species Degradation Reactions. The triple-spike methodology described above and applied to optimize the extraction procedure can only be employed for rearrangement reactions in closed “static” systems (e.g., a sample vial or a microwave vessel in which a certain energy is applied to carry out the extraction of the species from a solid matrix). In such cases, no external contribution to the initial concentration of the butyltin species existing in the sample occurs. Thus, our methodology is able to calculate the extent of possible butylation and debutylation reactions. However, if transformations of the species occur inside a dynamic system, based on different connected or correlated compartments (e.g., the human body), this isotope ratio methodology is not valid because of the mutual influences among degradation reactions occurring in each of the different interconnected pools. To study the transformation reactions occurring in this latter type of system, a methodology able to calculate the extent to which each initial isotopically enriched species (administered to the living organism) contributes to the measured species in the different compartments under investigation was developed. This methodology makes use of an isotope pattern deconvolution approach similar to that proposed previously by Meija and Caruso.38 for the deconvolution of isotope patterns in GC/MS spectra of butyltin compounds. For this purpose, the isotope abundances of the species determined in the samples are expressed as a linear combination of the isotope abundances of the initially administered isotopically enriched species and the natural abundances of the corresponding element. Although this methodology can be applied to a wide variety of elements, the development of the mathematical approach will be explained for the particular case of the three different species of tin studied (MBT, DBT, TBT), each enriched in a different Sn isotope. (46) Rodrı´guez-Gonza´lez, P.; Garcı´a Alonso, J. I.; Sanz-Medel, A. J. Anal. At. Spectrom. 2004, 19, 767-772. (47) Ruiz Encinar, J.; Monterde Villar, M. I.; Gotor Santamarı´a, V.; Garcia Alonso, J. I.; Sanz-Medel, A. Anal. Chem. 2001, 73, 3174-3180.
In this particular case, the Sn isotope abundances will be calculated by considering only the isotopes 116, 117, 118, 119, and 120, as the abundances for the isotopes 112, 114, 115, 122, and 124 are negligible in the isotopically enriched species employed. Then, three different mass balances for the number of moles of each species in a given sample m can be expressed as follows: TBT TBT TBT NTBT ) NTBT m MBT + NDBT + NDBT + Nnat
(1)
DBT DBT DBT NDBT ) NDBT m MBT + NDBT + NTBT + Nnat
(2)
MBT MBT MBT ) NMBT NMBT m MBT + NDBT + NTBT + Nnat
(3)
In eq 1, NTBT is the number of moles of TBT in the sample m, m NTBT MBT is the amount of TBT in the sample m formed from the initially administered MBT, NTBT DBT the possible amount of TBT formed from the initially administered DBT, NTBT TBT the amount of TBT coming from the initially administered TBT, and NTBT nat the amount of TBT coming from natural-abundance TBT, which could be endogenously present in the samples. Analogous notations are employed in eqs 2 and 3, and similar mass balances can be also obtained for the five isotopes in each species, as illustrated by eq 4 for the isotope 117 of TBT (in sample m): TBT TBT TBT TBT Nm,117 ) NMBT,117 + NTBT DBT,117 + NTBT,117 + Nnat,117
(4)
Therefore, eq 4 can be expressed as a linear combination of the original isotope abundances in the spike or in natural-abundance tin: TBT TBT TBT NTBT m Am,117 ) NMBTAMBT,117 + NDBTADBT,117 + TBT NTBT DBTATBT,117 + Nnat,117Anat,117 (5)
where AMBT,117, ADBT,117, and ATBT,117 are the isotope abundances of the isotope 117 in the isotopically enriched MBT, DBT, and TBT, respectively, and Anat,117 is the natural isotope abundance of 117Sn. Rearranging and applying eq 1. the following expression is obtained: TBT ) RAMBT,117 + βADBT,117 + γATBT,117 + δAnat,117 (6) Am,117
where
R)
β)
γ)
δ)
NTBT MBT TBT TBT TBT NTBT MBT + NDBT + NDBT + Nnat
NTBT DBT TBT TBT TBT NTBT MBT + NDBT + NDBT + Nnat
NTBT DBT TBT TBT TBT NTBT MBT + NDBT + NDBT + Nnat
NTBT nat TBT TBT TBT NTBT MBT + NDBT + NDBT + Nnat
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
(7)
(8)
(9)
(10) 7727
In the expressions 7-10, the variables R, β, γ, and δ indicate the molar fraction of TBT arising from the different compounds (enriched MBT, DBT, and TBT and natural-abundance TBT). If we assume that there are no other contributions to TBT than those derived from the isotopically enriched MBT, DBT, and TBT and that from the natural-abundance TBT, namely R, β, γ, and δ, respectively, we can state that
R+β+γ+δ)1
(11)
It is to be noted that the calculation of δ contribution makes this methodology applicable not only to study xenobiotic species (which are not expected to be found in the organism investigated) but also to study other species that could be already endogenously present. Similar equations can be obtained for the rest of the isotopes and they can be expressed, for the case of TBT, in matrix notation as
[] TBT Am,116
TBT Am,117
TBT Am,118 ) TBT Am,119
[ ][
TBT Am,120
AMBT,116 ADBT,116 ATBT,116 Anat,116
R β γ
1-R-β-γ
AMBT,117 ADBT,117 ATBT,117 Anat,117
‚ AMBT,118 ADBT,118 ATBT,118 Anat,118 AMBT,119 ADBT,119 ATBT,119 Anat,119 AMBT,120 ADBT,120 ATBT,120 Anat,120
]
[][ R
β ‚ γ
7728
TBT nat,i
TBT 2 - AMBT,i )
i
∑(A
∑A ∑A
MDBT sp CDBT ) βCMBT m MMBT m
∑A ∑A
MTBT sp CTBT ) γCMBT m MBT Mm
∑A ∑A
Mnat MBT CMBT nat ) δCm MMBT m
∑A ∑A
TBT nat,i
TBT TBT TBT - AMBT,i ) ‚ (Anat,i - ADBT,i )
∑(A
TBT nat,i
TBT TBT TBT - ADBT,i ) ‚ (Anat,i - AMBT,i )
TBT nat,i
TBT TBT TBT - AMBT,i ) ‚ (Anat,i - ATBT,i )
(14)
MBT sp
MBT m
(15)
DBT sp
MBT m
(16)
TBT sp MBT m
(17)
nat
∑(A
∑(A
TBT nat,i
TBT TBT TBT - ATBT,i ) ‚ (Anat,i - AMBT,i )
i
TBT nat,i
TBT 2 - ADBT,i )
i
∑(A
MBT m
where CMBT, CDBT, CTBT, and CMBT are the resulting corrected nat concentrations, Msp is the atomic weight of the corresponding initially administered isotopically enriched species, Mm is the atomic weight of the corresponding species in the sample m, and ΣAsp and ΣAm are the sum of the isotope abundances of the measured isotopes with respect to the total isotopes of the element of each species in the initial enriched species and in the sample, respectively. It is worth stressing that the sum of the isotope abundances is in all cases close to 1 except for the case of natural
i
i
i
MMBT sp CMBT ) RCMBT m MMBT m
(12)
The unknowns R, β, and γ represent the relative contribution of the initial isotopically enriched MBT, DBT, and TBT to the TBT that is present in the sample m. As we have more parameters than unknowns, least-squares fitting can be applied to calculate the values for these unknowns. It can be demonstrated (see Supporting Information) that the three unknown parameters for each compound can be obtained by resolving the following set of equations after matrix inversion:
∑(A
Similar equations can also be obtained for DBT and MBT, and hence, R, β, γ, and δ can be calculated for each species in each compartment of the dynamic system under study by measuring the isotopic composition of each compound. Determination of the Concentration of the Original Species after Degradation. The calculation of the concentrations of MBT, DBT, and TBT in the samples treated with the enriched species was carried out by reverse isotope dilution analysis using natural-abundance butyltin compounds taking into account the isotopic composition measured previously. However, when calculating the corrected concentrations after degradation we have to take into account the molar fractions (R, β, γ, δ), the atomic weights of the species in the sample, the atomic weights of the initial isotopically enriched species, and the sum of the isotope abundances of the measured isotopes with respect to the total isotopes of the element (both in the samples and in the initial enriched species). This can be expressed for MBT concentration in a given sample (CMBT m ) by the following equations:
∑(A
∑(A
TBT nat,i
TBT nat,i
TBT TBT TBT - ADBT,i ) ‚ (Anat,i - ATBT,i )
i
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
∑(A
[
TBT nat,i
TBT 2 - ATBT,i )
i
∑(A
TBT nat,i
TBT TBT TBT - Am,i ) ‚ (Anat,i - AMBT,i )
i
∑(A
TBT nat,i
TBT TBT TBT - Am,i ) ‚ (Anat,i - ADBT,i )
i
∑(A i
] ]
TBT TBT TBT - ATBT,i ) ‚ (Anat,i - ADBT,i ) )
i
TBT nat,i
TBT TBT TBT - Am,i ) ‚ (Anat,i - ATBT,i )
(13)
Figure 1. General scheme for the optimization of the extraction conditions.
tin (ΣAnat ) 0.8761 as, as said before, the isotopes 112, 114, 115, 116, 122, and 124 are not taken into account). RESULTS AND DICUSSION Preliminary Experiments Administering Natural-Abundance TBT. Preliminary experiments were carried out by administering to the animals a single dose of natural-abundance TBT dissolved in 2% acetic acid in tap water. Using isotope dilution analysis with the spike solution enriched in 119Sn, the stability of these solutions was demonstrated to be at least one week. Nevertheless, the doses were freshly prepared from concentrated standards. The concentration of these single doses was selected trying to keep the amount of TBT as low as possible and within the range of the “no observed adverse effect level” for continuous administration, which has been established as 0.010 µg of Sn g-1 day-1 (1 µg per day of TBT as Sn for a 100-g rat).48 Also, we had to consider the small amount of isotopically enriched species available for the study and the detection limits of the procedure. Finally, two different doses of 2 and 18.6 µg of TBT (as Sn) were administered to check the levels in which the butyltin species occur in the different organs. Optimization of the Extraction Procedure. The first aim of these preliminary experiments was the optimization of the extraction procedure. For this purpose, a comparison of different extraction strategies was carried out by analyzing different aliquots of a rat’s liver exposed to the single dose of 18.6 µg of TBT (as Sn) applying the triple-spike methodology, as employed in previous publications30,43,46 and explained in the procedures section. Liver tissue was selected for this optimization due to the high amount of sample (7 g of wet tissue) and its high fat content, which is claimed to compromise the extraction of the butyltin species from the solid. For the extraction of the butyltin compounds, we compared microwave assisted extraction at 150 W and mechanical shaking at 37 °C, both employing a mixture of acetic acid and methanol as extractant. These two procedures have (48) Penninks, A. H. Food Addit. Contam. 1993, 10, 351-361.
shown to provide little or no degradation of the species and quantitative extraction when analyzing the Certified Reference Material CRM-477 (lyophilized mussel tissue).46 A flowchart summarizing the experiments carried out on this optimization is given in Figure 1. Using both extraction techniques, the influence of the lyophilization, tissue homogenization, and moment of the spike addition was studied. To obtain a more representative sample, the tissue was cut in small pieces and mixed. Then, eight different representative aliquots were taken and treated as indicated in Figure 1. As can be observed, aliquots A1, A2, and A3 followed the same extraction procedures but with different MW assisted extraction times whereas aliquots A4, A5, A6, and A7 were extracted using mechanical shaking at 37 °C with a previous tissue homogenization (aliquots A4 and A5) and without such homogenization (aliquots A6 and A7). On the other hand, when extracting aliquot B, the triple spike was added to the sample before the lyophilization and the tissue homogenization. The final results for all the experiments, in terms of concentration found (ng/g as Sn, dry weight) and degradation factors F1 and F2, are listed in Table 1. Several conclusions can be obtained from these results: (a) No degradation was observed when using mechanical shaking at 37 °C in all the experiments whereas MW assisted extraction at 150 W promoted a significant degradation of the species after 16 min of extraction (aliquot A3), in agreement with previous results.46 (b) The degradation-corrected concentrations obtained for MW assisted extraction (aliquots A1, A2, and A3) are in general agreement with those obtained under mechanical shaking at 37 °C with or without performing the homogenization of the lyophilized tissue with the extractant (aliquots A4, A5, A6, and A7). So we can safely assume that quantitative extraction can be obtained under all such conditions. (c) The addition of the “spike” solution to the tissue before or after homogenization and lyophilization provided similar results, indicating that both steps do not promote any loss or degradation of the butyltin species neither in the sample nor in the spike. Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
7729
Table 1. Results Obtained for the Extraction of MBT, DBT, and TBT from Different Aliquots of a Rat Liver under Different Extraction Conditionsa aliquotb
TBT
DBT
MBT
F1 (%)
F2 (%)
A1 A2 A3 A4 A5 A6 A7 B
18.4 20.5 19.4 20.8 22.2 19.2 21.3 21.8
472 497 504 465 497 529 523 487
144 132 161 125 134 165 147 131
-1.6 2.1 9.3 0.0 0.0 -1.2 -0.7 0.0
0.9 -0.2 8.7 0.9 0.9 0.4 1.3 0.9
a Concentrations are expressed in ng/g of Sn, dry weight. The degradation factor F1 corresponds to the percent degradation of TBT to DBT and F2 to the percent degradation of DBT to MBT. b A1, A2, and A3: microwave-assisted extraction at 150 W for 4, 8, and 16 min, respectively. A4 and A5: mechanical shaking at 37 °C for 2 and 4 h, respectively, with a previous tissue homogenization. A6 and A7: mechanical shaking at 37 °C for 2 and 4 h, respectively, without a previous tissue homogenization. B: mechanical shaking at 37 °C for 2 h adding the triple spike before tissue homogenization and lyophilization.
Table 2. Distribution of MBT, DBT, and TBT in the Different Organs and Samples of Two Rats Treated with (a) 18563 and (b) 2029 ng of Natural-Abundance TBT (as Sn) (a) 18563 ng of TBT (as Sn) sample
TBT
DBT
MBT
TBT
DBT
MBT
blood brain thymus lung heart spleen testicles kidney liver stomach small intestine large intestine feces urine
3.5 21.4 13.2 10.6 10.9 3.5 4.9 23.8 15.5 110.7 546.3 404.2 3389.3 0.1
2.7 2.8 6.9 6.9 1.9 3.2 1.8 80.6 428.8 23.8 143.0 372.0 2014.5 32.3
0.8 0.4 1.3 0.8 0.2 0.3 0.4 27.3 133.6 5.7 30.9 110.6 445.1 45.5
0.5 2.0 0.2 1.1 0.9 0.3 0.5 2.4 2.5 15.4 58.6 43.8 604.9 0.0
0.1 0.2 0.0 0.5 0.1 0.2 0.1 7.3 84.8 3.3 16.0 22.7 156.8 1.0
0.1 0.1 0.1 0.2 0.1 0.1 0.2 3.4 22.4 1.4 5.0 6.4 43.5 1.5
total
4558
3121
803
732.9
293.0
84.5
36.1
14.4
4.2
recovery, %
In light of above results, the extraction procedure employed for aliquot A6 was selected for subsequent experiments as it provided similar concentration values as the degradation-corrected concentrations obtained for aliquot A3 (the small differences in concentration values obtained under the different extraction conditions tested can be attributed to lack of homogeneity of the sample). Results Obtained for Rats Exposed to Natural-Abundance TBT. Three animals (n ) 1 animal per group) were exposed to 0 (control), 2, and 18.6 µg of natural-abundance TBT (as Sn). After 24 h, the animals were sacrificed and the different organs indicated in the procedures plus urine and feces were analyzed for their butyltin content using the 119Sn mixed spike and the extraction conditions selected previously. In brief, all samples were lyophilized, spiked with 119Sn-labeled MBT, DBT, and TBT, and extracted with 5 mL of the mixture of acetic acid and methanol (3:1) for 2 h at 37 °C. The concentration of butyltin compounds in all organs from the control animal was below 0.5 ng g-1 (as Sn, dry weight). Therefore, no correction for endogenous butyltin compounds had to be performed. Table 2 reflects all the quantitative results obtained for animals exposed to either 2 and 18.6 µg of natural-abundance TBT. The results given in Table 2, indicated as total amount of Sn (in ng) found in each organ or sample for mass balance purposes, show a significant correlation for both samples: TBT levels in the rat exposed to 18.6 µg were ∼10 times higher than those obtained for the rat exposed to 2 µg of TBT. Moreover, it is worth noting that significant DBT and MBT levels were observed in most samples (due to the degradation of TBT). Especially in liver and kidney, the found levels of DBT and MBT were higher than those of TBT. This observation is in agreement with other studies in rats and humans26,23 where the main metabolite in liver was found to be DBT. Concerning the recovery values (obtained from the mass balances), it is interesting to note that in both experiments ∼30% of the Sn administered as TBT was recovered as TBT, 15% as DBT, and 4% as MBT. The total recovery for TBT and its metabolites approached 50% on average with an apparent absorp7730 Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
(b) 2029 ng of TBT (as Sn)
total recovery, % apparent absorption, %
24.6
16.8 45.7 68.5
4.3
54.7 60.4
tion (calculated from the amount of the three species in the feces) of 68% for the dose of 18.6 µg and 60% for the 2-µg dose animals. In all cases, most of the butyltin compounds were found in the feces while no TBT was detected in urine (only its metabolites). In contrast to previous single-administration studies49,50 where different metabolites were detected, the presence of metabolites other than MBT or DBT was not observed here except for inorganic tin in urine. Only in the case of liver samples, unknown peaks, close in retention time to TBT, were obtained at much lower levels than the butyltin species. These differences could be attributed to the 100-fold lower amount of TBT administered to the animals in this study, as compared to the other two. Administration of Isotopically Enriched Butyltin Species. The preliminary results showed in Table 2 indicated that even a single dose of 2 µg allowed the quantification of MBT, DBT, and TBT in the different organs investigated. However, the levels found in some organs were too close to the limit of quantification of the method. Therefore, and due the very limited amount of isotopically enriched species available, a single 1-mL dose of a solution containing 3.9, 4.8, and 7.0 µg (as Sn) of MBT, DBT, and TBT, respectively, was administered to one single animal. For this purpose, the isotopically enriched species, which were dissolved in acetic acid and methanol to ensure their long-term stability, were taken to dryness in a vacuum distillator and redissolved in water containing 2% acetic acid. Finally, this solution was characterized, before administration, in terms of isotope composition and concentration by reverse isotope dilution analysis. A GC-ICPMS chromatogram of the resulting solution is given in Figure 2. As can be observed, MBT is enriched in 119Sn, DBT enriched in 118Sn, and TBT enriched in 117Sn. Additionally, the spike contained no detectable levels of inorganic tin. (49) Matsuda, R.; Suzuki, T.; Saito, Y. J. Agric. Food Chem. 1993, 41, 489-495. (50) Ueno, S.; Kashimoto, T.; Susa, N.; Ishii, M.; Toshikazu, C.; Mutoh, K.; Hoshi, F.; Suzuki, T.; Sugiyama, M. Arch. Toxicol. 2003, 77, 173-181.
Figure 2. GC-ICPMS chromatogram of the solution containing the isotopically enriched species administered to the animal. The plots corresponding to the different masses have been shifted for clarity.
For the purpose of this study, after lyophilization all samples were split into two representative aliquots: one for the determination of the isotope abundances of each butyltin compound in the sample and the other, spiked with an appropriate amount of natural-abundance butyltin species, to determine the amount of each butyltin species in all samples (by reverse isotope dilution analysis). Additionally, when the amount of sample allowed it (feces, liver, and small and large intestine), three independent determinations of both the isotope abundance and concentration were carried out to evaluate the repeatability of the applied analytical methodology. Isotope Pattern Deconvolution. The mathematical approach described in the procedures was applied to calculate the degradation of the studied butyltin species in the different organs and samples taken. For this purpose, the isotopic composition measured in each sample was deconvoluted, using a spreadsheet software, where the isotopic composition of the administered spike and the isotopic composition of natural tin were taken into account. As an illustrative example, the original data obtained for DBT in lung and the input quantities for the isotopic composition of the different Sn isotopically enriched species is given in eq 18:
116 117 118 119 120
()() () () () 4.48 51.20 43.53 0.34 0.45
0.65 6.68 ) R. 34.66 55.75 2.17 m
2.49 31.53 + β. 65.07 0.41 0.44 MBT 6.54 92.99 γ. 0.25 0.04 0.15
+
DBT
16.58 8.77 + δ. 27.66 9.80 37.19 TBT
(18)
nat
As can be observed, the isotopic composition of natural tin had to be normalized to 100% for the purpose of these calculations.
Figure 3. GC-ICPMS chromatogram of the urine of the animal treated with isotopically enriched butyltin species
The values of R, β, γ, and δ can be calculated then by least squares using the equations given in the procedures. For DBT in lung, the values found were R ) 0.000, β ) 0.664, γ ) 0.325, and δ ) 0.011. That means that 66.4% of DBT in lung (as molar fraction) comes from the originally spiked DBT while 32.5% was formed from the originally spiked TBT. Finally, 1.1% of the isotopic composition in the sample is explained by natural tin abundances. The results obtained in this way for all samples and all compounds under scrutiny are shown in Table 3 for MBT, DBT, and TBT, respectively. (a) MBT Sources in the Different Tissues and Samples. As can be observed in Table 3, the isotopic signature of MBT in the feces is very close to that of MBT in the spike solution (Figure 2). Thus, it is to be expected that 83.9% of MBT in feces comes from the original MBT in the spike with minor contributions from DBT and TBT. Also a large proportion of the original MBT can be found in stomach (56.7%) and in the small and large intestines (22.8 and 29.7% respectively). However, for the urine sample in Figure 3 the situation is quite the opposite: MBT shows an isotopic signature between that of DBT and TBT with very little contribution from the original MBT in the spike. Calculations show that 59.3% of MBT in urine is derived from metabolized TBT and 28.6% from metabolized DBT. Please note also in Figure 3 that almost no TBT was detected in urine while inorganic tin had a signature similar to that of TBT. For the liver and kidney samples, MBT comes mainly from both DBT and TBT with variable contributions. Samples of blood (Figure 4), brain, thymus, lung, heart, spleen, pancreas, and testicles contained very little MBT, the corresponding isotopic signature being between that of TBT and natural tin. Please note also the isotopic composition of inorganic tin detected in blood (Figure 4), which shows almost pure Sn natural abundances. (b) DBT Sources in the Different Tissues and Samples. Table 3 shows the results obtained for the isotopic contribution of the isotopically enriched species to the DBT mass spectra. As can be observed, the contribution from MBT (R values) is close to 0 for all samples, ruling out the existence of butylation reactions. Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
7731
Table 3. Isotope Pattern Deconvolution for MBT, DBT, and TBT in the Different Samples and Organs of the Animal Treated with Isotopically Enriched butyltin Speciesa MBTb
blood brain thymus lung heart spleen pancreas testicles kidney liver (n ) 3) stomach small intestine (n ) 3) large intestine (n ) 3) feces (n ) 3) urine
DBTb
TBTb
R
β
γ
δ
R
β
γ
δ
R
β
γ
δ(%)
3.2 0.1 2.5 7.1 5.8 2.1 6.3 1.0 5.5 1.2 ( 0.1 56.7 22.8 ( 1.3
10.3 -5.6 6.8 14.8 10.3 17.2 12.7 13.8 42.4 23.1 ( 0.3 11.8 15.9 (1.1
38.4 33.3 25.9 57.9 30.5 37.9 45.6 21.1 50.9 73.9 ( 0.4 29.3 55.5 ( 1.5
48.0 72.3 64.7 20.3 53.5 42.8 35.4 64.1 1.2 1.7 ( 0.1 2.2 5.9 ( 1.3
0.0 -0.7 0.1 0.0 0.1 0.0 0.0 0.1 0.0 0.0 ( 0.0 0.0 0.0 ( 0.0
66.8 47.8 68.7 66.4 65.9 68.7 72.0 58.5 74.8 59.8 ( 0.2 74.0 62.5 ( 0.2
30.9 42.8 29.0 32.5 32.8 30.0 26.7 32.4 25.2 40.4 ( 0.2 26.0 36.6 ( 0.9
2.2 10.2 2.1 1.1 1.2 1.3 1.3 9.0 0.0 -0.2 ( 0.1 0.0 0.8 ( 0.8
0.0 -0.2 0.1 0.0 -0.1 -0.6 -0.1 -0.5 -0.1 0.0 ( 0.1 0.0 0.0 ( 0.0
0.3 -0.3 -0.3 -0.2 -0.4 0.1 -0.3 0.8 -0.2 -0.2 ( 0.4 0.0 0.1 ( 0.3
93.7 94.8 87.6 94.6 96.6 93.1 97.1 80.7 98.7 97.4 ( 1.1 100.1 98.6 ( 2.0
6.7 5.7 12.7 5.6 3.8 7.5 3.3 18.9 1.6 2.8 ( 1.2 -0.1 1.4 ( 1.8
29.7 ( 4.6 16.1 ( 0.4 50.6 ( 2.4 3.6 ( 1.9 1.0 ( 1.7 59.7 ( 1.0 39.0 ( 0.8 0.3 ( 0.1
0.1 ( 0.1 -0.1 ( 0.2 99.6 ( 0.6 0.5 ( 0.7
83.9 ( 0.2 11.0 ( 0.1 4.5 ( 0.1 11.2 28.6 59.3
0.0 ( 0.0 -0.2 ( 0.1 99.2 ( 0.9 1.0 ( 1.0
0.6 ( 0.1 0.0 ( 0.0 82.7 ( 0.6 17.1 ( 0.5 0.2 ( 0.2 1.0 0.0 43.2 56.1 0.7
a The uncertainty in the values for the liver, small intestine, large intestine, and feces corresponds to the standard deviation of three independent replicates (n ) 3). b R, molar fraction (%) arising from initially administered MBT; β, molar fraction (%) arising from initially administered DBT; γ, molar fraction (%) arising from initially administered TBT; δ, molar fraction (%) arising from natural-abundance tin.
Figure 4. GC-ICPMS chromatogram of the blood of the animal treated with isotopically enriched butyltin species
Also, the contribution from natural-abundance DBT is negligible except for brain and testicles. In general, a combined contribution from original DBT and metabolized TBT was evident in all samples. The largest contribution from original DBT was found in feces (82.7%) while the largest contribution from the original TBT was in urine (56.1%, Figure 3). Similar contributions in stomach, kidney, and pancreas (∼74% of the original DBT) were observed while ∼60% of the original DBT was observed in liver and intestines. Intermediate contributions were obtained in the rest of the samples (except for brain and testicles, where the natural contribution was lager). It is worth stressing the small uncertainties obtained for the isotopic contributions in those samples where three independent measurements were performed (liver, intestines, feces). (c) TBT Sources in the Different Tissues and Samples. Table 3 collects all the results obtained for the isotopic contribu7732
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
tion of the administered enriched species to the TBT mass spectra. As can be observed, both the R and the β contributions are negligible in all cases, indicating the expected lack of butylation reactions. In all samples and tissues, the main contribution was from the original 117Sn-enriched TBT. Only in thymus and testicles was the relative contribution from natural-abundance TBT larger than 10%. Such contributions can be attributed to either blank values, originating from contamination of the reagents, or from endogenous TBT originally present in the sample. Reverse Isotope Dilution Analysis Using Natural-Abundance MBT, DBT, and TBT. Table 4 shows the results obtained for MBT, DBT, and TBT concentrations, respectively, in the different organs and samples analyzed after reverse isotope dilution analysis (as ng of Sn). The first column in each table shows the total amount found for each compound regardless of its source, while in the second column the total amount found referred only to the original compounds administered (after correction for the degradation factors shown in Table 3) is given. That is, when DBT and MBT were increased due to the degradation of TBT, this is taken into account in the calculation of the degradation-corrected amounts. Data for MBT shown in Table 4 indicate that most of MBT found in the different organs comes from the degradation of DBT and TBT. As can be observed, the levels of MBT found in all samples, except for the feces, decrease to almost zero after degradation is taken into account. If we perform a mass balance from the original MBT amount given, we obtain an apparent absorption of only 12.5% with a total recovery of 87.8% in the samples analyzed. The total recovery was higher than 100% if the degradation of TBT and DBT was not taken into account with negative values for the apparent absorption. For DBT (Table 4), the degradation-corrected data are always lower than the corresponding uncorrected data. It is clear also that DBT is formed faster from the original TBT than it degrades to produce MBT. The high levels of DBT found in liver may indicate a preferential accumulation in this tissue. Nevertheless,
Table 4. Distribution of MBT, DBT, and TBT in the Different Organs and Samples of the Animal Treated with Isotopically Enriched Butyltin Species before and after Correction for the Degradationa MBT
DBT
TBT
uncorrected (ng of Sn)
degradationcorrected (ng of Sn)
uncorrected (ng of Sn)
degradationcorrected (ng of Sn)
uncorrected (ng of Sn)
degradationcorrected (ng of Sn)
blood brain thymus lung heart spleen pancreas testicles kidney liver (n ) 3) stomach small intestine (n ) 3) large intestine (n ) 3) feces (n ) 3) urine
0. 4 0.3 0.0 0.5 0.0 0.2 0.1 0.5 13.1 74.1 ( 2.7 4.3 12.2 ( 1.3 13.7 ( 0.4 4088.4 ( 174.9 13.2
0.01 -0.04 0.00 0.04 0.00 -0.01 0.01 -0.01 0.73 0.95 ( 0.15 2.43 2.72 ( 0.33 3.96 ( 0.90 3433 ( 143 1.48
2. 4 2.4 1.4 6.8 1.7 4.3 4.5 4.4 114.0 701.7 ( 28.1 22.9 132.7 ( 2.2 78.8 ( 1.2 3654.3 ( 71.2 13.6
1.7 1.1 1.0 4.6 1.1 3.0 3.3 2.7 90.9 437.3 ( 15.8 17.5 85.2 ( 1.5 49.5 ( 0.4 3469.6 ( 66.9 9.7
1.2 9.3 1.2 8.7 3.6 1.9 3.2 4.9 8.7 9.1 ( 0.6 62.0 184.6 ( 0.8 82.8 ( 0.7 1715.9 ( 16.4 n.d.
2.0 10.0 1.5 10.7 4.0 3.1 4.3 5.5 43.9 346.1 ( 10.7 69.3 237.0 ( 5.3 119.6 ( 0.5 2503.8 ( 26.0 15.4
total ng of Sn administered recovery aparent absorption
4221.0 3922.2 107.6% -4.2%
3445.5 3922.2 87.8% 12.5%
4746.1 4770.1 99.5% 23.4%
4178.1 4770.1 87.6% 27.3%
2097.0 7031.1 29.8% 75.6%
3376.1 7031.1 48.0% 64.4%
a The uncertainty in the values for the liver, small intestine, large intestine, and feces corresponds to the standard deviation of three independent replicates (n ) 3).
most of the original DBT was found in the feces with very little contribution from degraded TBT. This resulted in an apparent absorption of only 27.3% when the degradation of TBT to form DBT and the degradation of DBT to form MBT was taken into account. The total degradation-corrected recovery for DBT was 87.6%, a value very similar to that found for MBT. Data for TBT are also summarized in Table 4. As can be observed, in all cases, here the degradation-corrected values were higher than the uncorrected ones with some striking results. For example, very little TBT was found in kidney and liver; however, if we take into account the contribution of TBT to the isotopic signatures of DBT and MBT in these two tissues (degradation corrected values), the amount found in kidney increases 5-fold while the amount in liver increases more than 30 times. In other words, it seems that these two organs are responsible for most of the TBT degradation. Similar results were found in urine where TBT was not even detected in the uncorrected data. However, when the contributions of MBT and DBT were taken into account, more than 15 ng of the original TBT was shown to be excreted in 24 h. Comparing the total amount found in the feces with the total TBT amount administered, an apparent absorption of 64.4% was found. This figure is in agreement with the values given in Table 2 where natural-abundance TBT was administered. Also, this figure turned out to be close to a “solubilization factor” of 60% we found previously using simulated gastrointestinal digestion.30 Total recovery of TBT was only of 48.0%, in agreement also with data shown in Table 2 for experiments with naturalabundance TBT. This means that more than 50% of TBT was unaccounted for; that is, we did not measure all possible tissue samples containing Sn in the animal (TBT may be also accumulated in other tissues, e.g., muscles, skin, and the fat tissue under the skin). It is also possible than some original TBT was degraded to inorganic tin. However, the isotopic signature of
labeled TBT was only found in urine associated with inorganic tin. In the rest of the analyzed samples, inorganic tin was not present or its isotopic signature was that of natural-abundance tin. The relative standard deviations obtained in the concentration values were in the range of 0.5-7% depending on the concentration level. Similar uncertainties were obtained in most cases in the measurement of the isotopic contributions (except when the isotopic contributions were found to be negligible). Thus, as reflected in the standard deviations obtained both in the amount of species and in the contribution factors, the reproducibility of the analytical methodology can be considered satisfactory for this type of metabolism study. However, it should be stressed that an important biological variability for these parameters is to be expected when more than one animal is included in the study. CONCLUSIONS An analytical methodology has been developed in this work to study organotin metabolism. Using such method, we have been able to demonstrate that the butyltin species more efficiently absorbed is TBT. This conclusion is first derived from the apparent absorption values obtained for the three species and also from the fact that the metabolite products of 117Sn-enriched TBT were found to be the predominant species detected, both in liver and in urine. This fact could be qualitatively demonstrated, with the corresponding GC-ICPMS chromatograms and, also quantitatively, from the significantly similar γ contributions obtained in both types of samples for DBT and MBT. The results shown here also demonstrate that the developed methodology is able to follow qualitatively, quantitatively, and simultaneously the specific metabolic pathways of different species of a given element. It is worth stressing that if the isotope abundances of all the administered species differ from the Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
7733
endogenous abundances of the element under scrutiny, this methodology is not limited to the study of the metabolism of xenobiotic species as endogenous ones can also be simultaneously followed. However, one must ensure that the isotopically enriched species are administered in such a way that the isotope abundances of the endogenous species are significantly altered. Finally, this work clearly demonstrates the present need of synthesizing isotopically enriched elemental species. Their subsequent availability in the future will pave the way to carry out toxicological studies of individual chemical species with realistic statistical populations. These studies may be extremely useful to unravel the metabolic pathways of the different chemical forms of a wide number of elements in the frame of interdisciplinary collaborations between analytical chemists, toxicologists and nutritionists.
7734
Analytical Chemistry, Vol. 77, No. 23, December 1, 2005
ACKNOWLEDGMENT The authors would like to thank Dr. Javier Garcı´a Ferna´ndez from the Department of Organic Chemistry of the University of Oviedo for his help in the distillation of the isotopically enriched species. Funding of this work through the Spanish Ministry of Science and Technology project BQU2000-0221 and BQU200303438 is also acknowledged. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review June 20, 2005. Accepted September 22, 2005. AC051091R
