Anal. Chem. 1991. 63,890-893
890
resonators coated with the photoresist polymer as much as 10-15% of the integral sensitivity stems from the region beyond the electrode boundary. This behavior is attributed to the dielectric contribution and mass loading of the liquid medium and polymer film, as well as the mass loading of the electrode tabs in this region. Our results clearly indicate that calibration of the QCM based on geometric measurements alone can lead to erroneous results. It is therefore important to calibrate the QCM, preferably by electroplating and stripping, to guarantee accurate measurements of mass with the QCM in liquids. In addition, the observation of appreciable mass sensitivity at r > Ir,l for plano-plano crystals is especially significant for calibration of QCM sensors, which commonly employ designs in which active films are present, and therefore mass changes occur, in nonelectroded regions of the resonator. Our results suggest that plano-convex resonators, which exhibit negligible sensitivity at r > IrJ, may be more suitable for mass-sensing applications. ACKNOWLEDGMENT We gratefully acknowledge the assistance of J. Howe (Du Pont). Registry NO.CU, 7440-50-8;Ag, 7440-57-5; quartz, 14808-60-7. LITERATURE CITED (1) (a) Bruckensteln, S.; Shay, M. Electrochim. Acta 1985, 30, 1295. (b) Buttry. D. A. I n Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1990; Vol. 17, p 1. (c) Ward, M. D.: Buttry, D. A. Science 1990, 249, 1000. (2) Lu, C.; Lewis, 0. J . Appl. fhys. 1972, 43, 4385. (3) Sauerbrey, G. 2.Phys. 1959, 155, 206. (4) (a) Gullbault. G. 0. Ion-Sel. Electrode Rev. 1880, 2 , 3. (b) Guilbault, G. G.; Jordan, J. M. CRC Crn. Rev. Anal. Chem. 1988, 79, 1 and references therein.
(5) (a) Ngeh-Ngwainbi, J.; Foley, P. H.; Kuan, S. S.; Gullbault, 0. 0. J . Am. Chem. Soc. 1988, 108, 5444. (b) Muramatsu. H.; Dicks, J. M.; Tamiya, E.; Karube, 1. Anal. Chem. 1987, 59. 2760. (c) Shons, A.; Dorman, F.; Najarian, J. J . Biomed. Mater. Res. 1972, 6 , 565. (d) U. S. Patent 4,236,893. (e) Roederer, J. E.; Bastiaans. 0. J. Anal. Chem. 1983, 55, 2333. (f) U. S. Patent 4,242,096. (9) Ebersole, R. C.; Ward, M. D. J . Am. Chem. SOC. 1988, 710, 8623. (h) Ebersole, R. C . ; Miller, J. A.; Moran. J. R.; Ward, M. D. J. Am. Chem. Soc. 1990, 772, 3239. (I) Lasky, S. J.; Buttry, D. A. ACS Symp. Ser. 1989, 403, 237. (6) (a) Melroy, 0.: Kanazawa, K.; Gordon, J. G., 11.; Buttry, D. Langmulr 1988, 2, 697. (b) Deakln. M. R.; Melroy, 0. J . Electroanal. Chem. Interfacial Electrochem. 1988. 239, 321. (7) (a) Kaufman, J. H.; Kanazawa, K. K.; Street, G. B. fhys. Rev. Lett. 1984, 53, 2461. (b) Varineau. P. T.; Buttry, D. A. J . Phys. Chem. 1987, 9 7 , 1292. (c) Ward, M. D. J. Electrochem. SOC. 1988, 735, 2747. (d) Orata. D.;Buttry, D. A. J . Am. Chem. Soc. 1987, 709, 3574. (8) Masuda, H.; Baba, N. Chem. Lett. 1987, 1877. (9) Ward, M. D. J . fhys. Chem. 1988, 9 2 , 2049. (10) (a) Schumacher, R.; Mueller, A.; Stoeckel. W. J . Electroanal. Chem. InterfacialElectrochem. 1987, 279, 311. (b) Schumacher, R.; Gordon, J. G.; Melroy. 0. J . Electrmnal. Chem. Interfacial Electrochem. 1987, 276, 127. (c) Schumacher, R.; Borges, G.; Kanazawa, K. K. Surf. Sci. 1985, 163, L261. (11) Baker, C. K.; Reynolds, J. R. J . Electroanal. Chem. InterfaclalElectrochem. 1988, 257, 307. (12) Ullevig, D. M.; Evans, J. F.; Albrecht, M. G. Anal. Chem. 1982, 5 4 , 2341. (13) (a) Koga, I.; Fukuyo. H. J . Instrum. Electr. Commun. Eng. Jpn. 1953, 36. 59. (b) Fukuyo, H.; Yokoyama, A.; Ooura, N.; Nonaka, S. Bull. Tokyo Inst. Tschnol. 1965, 72, 1. (c) Koga, I; Tsuzuki, Y.; Wln, S. N., Jr.; Bennett, A. L. Roc. Annu. Freq. Control Symp, 1960, 74, 53. (14) (a) van Dyke, K. S. Roc. Annu. Freq. Control Symp. 1957, 7 7, 000. (b) van Dyke, K. S. Roc. Annu. Freq. Control Symp. 1958, 10, 1. (15) Martin, B. A.; Hager, H. E. J . Appl. Fhys. 1989, 65, 2630. (16) Kanazawa, K. K.: Gordon, J. G., 11. Anal. Chem. 1985, 5 7 , 1770.
RECEIVED for review November 26, 1990. Accepted January 29,1991. This paper is contribution no. 5622 from the Central Research and Development Department (Du Pont).
Determination of Boron in Tissues and Cells Using Direct-Current Plasma Atomic Emission Spectroscopy Rolf F. Barth,* Dianne M. Adams, Albert H. Soloway, Eugene B. Mechetner,' Fazlul Alam? and Abul K. M. Anisuzzaman
Department of Pathology and College of Pharmacy, The Ohio State University, Columbus, Ohio 43210 We have developed a safe, slmpie, and efflclent method for boron determlnatlon by means of dlreckurrent plasma atomic emlsslon spectroscopy. Tissues were soiubllized by using concentrated sulfuric acld and 70 % hydrogen peroxlde to dlgest the samples wlthout the need of hlgh temperatures and pressures. Boron duster compounds coukl be measured wlth sensitlvlty, preclslon, and accuracy slmllar to those of boric acld standards. Results obtalned with [(C,H5)3NH]2B12H12, Cs2Bl,H11SH*H,0, and ClSH,,Bl,O, show that thls analytical method ls applicable to a variety of compounds with different chemlcal structures. A sensltlvlty of 0.1 ppm has been obtained with known standards alone and In a varlety of tissue matrices includlng tumor, blood, liver, skin, and cell suspensions. The measurement of total boron by dlrect-current plasma atomlc emlssion spectroscopy (DCP-AES) has been achleved wlth as little as 50 mg of tissue or as few as 5 X lo7 cells. The procedure ls applkabk to the analysis of boron In the ppm range with a hlgh degree of precision and accuracy. Present address: Department of Genetics, University of Illinois, College of Medicine, 808 S. Wood St., Chicago, IL 60612. Present address: US. Borax, 412 Crescent Way, Anaheim, CA
92801.
INTRODUCTION The accurate measurement of total boron content in biological samples with a sensitivity in the ppm range is essential for evaluating the potential usefulness of various tumorlocalizing boron-containing compounds for boron neutron capture theory (BNCT) ( I ) . Among the procedures that have been used is spectrophotometric analysis involving various complexing agents such as 1,l'-dianthrimide (2-4), methylene blue (5), and curcumin (6, 7). These methods are time consuming and require that the boron compounds be oxidized to boric acid. Relatively low sensitivity and interference from various contaminants are further limitations in the use of colorimetric assays. A second group of analytical procedures for boron are those involving nuclear methods. These include the detection of a particles resulting from the ' O B ( ~ , C Y , ~reaction ) ~ L ~ by means of a track autoradiography (8-11) and the measurement of y photons, by means of prompt y analysis (12,13). A major advantage in the use of nuclear methods is the fact that sample decomposition is not required. However, sample geometry is important and the a particles and y photons resulting from the 'OB(n,cu,~)~Li reaction are produced only from boron-10, which comprises 19.8% of natural elemental boron. The remaining 8O.2%, consisting of boron-11, is not detected by
0003-2700/91/0363-0890$02.50/00 1991 American Chemical Society
ANALYTICAL CHEMISTRY, VOL. 63, NO. 9, MAY 1, 1991
these methods, thereby reducing their overall sensitivity. Furthermore, a thermal neutron source such as a nuclear reactor or a neutron-emitting radionuclide such as californium-252 also is required. Analysis of boron by means of atomic absorption spectroscopy has been reported. However, this technique is subject to interference and lacks sufficient sensitivity to be analytically useful for the assay of boron in biological samples. Recently, the use of inductively coupled plasma atomic emission spectroscopy (ICP-AES) has been shown to be sensitive enough for the detection of microgram quantities of boron in biological samples (14-17). The sensitivity is adversely affected by high concentrations of inorganic salts in the samples (17). An advantage of this method, however, is that both boron-10 and boron-11 can be detected, regardless of the isotopic form. Since alkaline fusion is not suitable for use with ICP (18), samples were digested by exposure to either perchloric acid (14, 15) or nitric acid (16). With the former, there may be a danger of explosion and with the latter, it may be necessary to decompose the tissues in Teflon-lined digestion bombs. Both procedures, therefore, have their limitations. The objectives of the present study were to determine whether DCP-AES could be used t o quantify boron in a variety of chemical compounds, including polyhedral boranes and carboranes, to improve the procedure for the digestion of tissue samples, to determine if this method could be used to quantify cellular uptake of boron, and finally to define the limits of boron detection in biologic samples by means of this method. EXPERIMENTAL SECTION Instrumentation. A Spectraspan VB direct-current plasma atomic emission spectrometer (Applied Research Laboratories, Brea, CA) was used for boron determinations. This instrument combines a high-resolution spectrometer with a high-resolution Echelle grating and prism. The plasma source is argon gas heated to a temperature of 6000-7000 K. The instrument settings used for analysis were wavelength 249.773 nm, argon flow 7 L/min, sleeve 50 psi, ceramic nebulizer 20 psi, and entrance slits 50-300 pm, as predetermined by the manufacturer. The viewing height was 1 mm between the arms of the "V" of the inverted "Y" configuration created by the three electrodes. The liquid uptake rate was approximately 2 mL/min. The operating power, once the plasma was established, was -40 V in the jet power supply with a 7-A constant-current output. With the operating parameters used no background correction was needed. Integration times and gain had been optimized by the manufacturer and were present in the computer software. These particular parameters are not user alterable. Variables such as travel, delay, and rinse times, were not factors, since samples were run manually without the use of an autosampler. Reagents and Test Compounds. All reagents were of analytical grade. Concentrated sulfuric acid and 70% hydrogen peroxide (E. I. du Pont de Nemours & Co., Wilmington, DE) were used for tissue digestion. A stock solution containing 1000 ppm of boron was prepared by dissolving weighed quantities of 95%-enriched loB boric acid (Eagle Picher, Quapaw, OK) in a known volume of deionized, distilled water. This solution has been stored in a plastic container and is stable at ambient temperature for up to 1 year. Working standards were made by diluting the stock solution with distilled water. The polyhedral boranes Na2Bl2Hl1SHand Cs2Bl2HllSH*H20were generously provided by Callery Chemical Co., Callery, PA. The polyhedral and the carborane C1&&$006(Figure borane [ (C2H5)3NH]3B12H12 1) were synthesized in our laboratory. Animal Tumors and Tumor Cells. Male BALB/c mice, weighing -20 gm, were purchased from Charles River Laboratories, Inc. (Wilmington, MA). The animals were injected intramuscularly in the right flank with 1.5 X lo6 Harding-Passey melanoma cells suspended in 0.2 mL of Hanks' balanced salt solution. Ten to 12 days later, when tumors had attained a size of -1 cm3, animals were bled via retrorbital venous plexus and then killed. Tumor, liver, skin, and muscle samples were removed and weighed. Known quantities of boric acid were added, and
.
891
.
BlO H1o
Figure 1.
the samples were processed for boron determination, as described below. For in vitro studies on the cellular uptake of boron-containing compounds, mouse Rauscher leukemia virus-transformed K-2 and human K562 erythroleukemic cell lines were used. Since it was observed in preliminary experiments that an unknown constituent of tissue culture medium RPMI 1640 interfered with the boron signal, cell lines were adapted for growth in Dulbecco's minimum essential medium (DMEM) supplemented with 10% fetal bovine serum. In order to determine the sensitivity of the DCP-AES instrument for measuring boron concentrations in tissue culture medium, varying concentrations of the polyhedral borane Na2Bl2Hl1SH (Callery Chemical Co., Callery, PA), ranging from 0.8 to 1000 pg of boron, were added per milliliter of DMEM. Determinations were performed on duplicate samples. Cellular uptake studies were carried out by separating viable from nonviable cells by means of density gradient centrifugation using Histopaque (Sigma Chemical Co., St. Louis, MO). Cells at a concentration of 106/mL were incubated at 37 "C with Na2B12HllSH for varying times ranging from 17 to 96 h. Sample Preparation for Analysis. Concentrated sulfuric acid (1-2 mL), which was adequate for up to 1 g of tissue, was added to 150 x 16 mm Pyrex culture tubes, and these were placed in a mineral oil bath, heated to 100 "C in an exhaust hood, and stirred intermittently for 1h. Since no interference with the boron signal was found with the sulfuric acid-hydrogen peroxide cocktail, the amount of sulfuric acid used in digestion was determined by the ease of digestion for a particular sample, rather than by any specific ratio of sample weight to acid volume. The amount and type of tissue were critical factors for successful digestion. If the amount of tissue approached 1 g or more, there was more chance for overheating and loss of sample by bubbling. If sample size was small (lo00 mg and containing >5.0 pg of boron/mL of tissue digest, there was a reduction in the amount of boron detected compared with the value predicted, based on the amount of boron that was used. Skin showed a greater suppression of the boron signal a t values of 10 pg and greater. Since the majority of normal tissue samples to be analyzed would be expected to contain 6 0 pg of boron/mL of tissue digest, a slight loss in sensitivity for those with >50 pg of boron/mL would not be of great significance. Furthermore, those samples containing >50 pg of boron/mL can be diluted with water in order to obtain more accurate analytical results. Since many compounds that will be screened as potential agents for BNCT contain either polyhedral borane anions or carboranes, it was important to determine whether these boron clusters could be measured with sensitivity, precision, and accuracy similar to those of boric acid standards. For this purpose, we have compared [ (C2H5)3NH]2B12H12, Cs2B12HI,SH-H20, and C15H32B1006 with H3B03 standards. The results are tabulated in Table I1 and show that this analytical method is applicable to a variety of compounds with different chemical structures, including boron-containing clusters with a precision of
