Based Quantification of Hydrolyzed Urinary Steroids

Published OnlineFirst January 19, 2010; DOI: 10.1158/1055-9965.EPI-09-0581

Cancer Research Article Epidemiology, Biomarkers & Prevention Systematic Error in Gas Chromatography-Mass Spectrometry– Based Quantification of Hydrolyzed Urinary Steroids

Ju-Yeon Moon1,2, Young Wan Ha1, Myeong Hee Moon2, Bong Chul Chung1, and Man Ho Choi1

Abstract Gas chromatography-mass spectrometry–based metabolite profiling can lead to an understanding of various disease mechanisms as well as to identifying new diagnostic biomarkers by comparing the metabolites related in quantification. However, the unexpected transformation of urinary steroids during enzymatic hydrolysis with Helix pomatia could result in an underestimation or overestimation of their concentrations. A comparison of β-glucurondase extracted from Escherichia coli revealed 18 conversions of 84 steroids tested as an unexpected transformation under hydrolysis with β-glucuronidase/arylsulfatase extracted from Helix pomatia. In addition to the conversion of 3β-hydroxy-5-ene steroids into 3-oxo-4-ene steroids, which has been reported, the transformation of 3β-hydroxy-5α–reduced and 3β-hydroxy-5β–reduced steroids to 3-oxo- 5α–reduced and 3-oxo-5β–reduced steroids, respectively, was newly observed. The formation of by-products was in proportion to the concentration of substrates becoming saturated against the enzyme. The substances belonging to these three steroid groups were undetectable at low concentrations, whereas the corresponding by-products were overestimated. These results indicate that the systematic error in the quantification of urinary steroids hydrolyzed with Helix pomatia can lead to a misreading of the clinical implications. All these hydrolysis procedures are suitable for study purposes, and the information can help prevent false evaluations of urinary steroids in clinical studies. Cancer Epidemiol Biomarkers Prev; 19(2); 388–97. ©2010 AACR.

Introduction duce deconjugated steroids because sulfate conjugates are not hydrolyzed by E. coli (12-16). There are many naturally occurring steroids that are However, unexpected transformations of steroids excreted mainly through the urine by their water-soluble during hydrolysis could obstruct the analysis. The con- conjugates formed by the substitution of 3- or 17-hydrox- version of 3β-hydroxy-5-ene steroids leads to both 3- yl groups with either sulfate or β-glucuronide (1), and oxo-4-enes as the major products and 6-oxy metabolites their direct measurements have been introduced to clini- as the minor ones with Helix pomatia (14, 16-21). These cal studies (2-5). Gas chromatography-mass spectrometry two actions suggest that they are caused by the presence – β Δ (GC-MS) based profiling is a proven technique in steroid of 3 -hydroxysteroid dehydrogenase/ 5-4-ene steroid analysis, whereas immunoassays have limited applicabil- isomerase and 6-hydroxylase as additional enzymes (17, ity due to cross-reactions (6, 7). However, GC-MS–based 18, 20) in the Helix pomatia extracts. The generation of techniques mainly require the hydrolysis of steroid con- 3-oxo-4-ene steroids might be converted by cholesterol jugates due to their low volatility before instrumental oxidase because 3β-hydroxysteroid dehydrogenase does analysis (8-11), and the need to treat samples with one not require oxygen (16, 17, 21-24), but this has not been of two enzyme solutions, β-glucuronidase and a mixture proven. The variability of the selectivity and reactivity of β-glucuronidase/arylsulfatase, which are extracted of Helix pomatia is also affected by the reaction tempera- from Escherichia coli and Helix pomatia, respectively. The ture, incubation time pH, and amount of enzyme added enzyme solution of Helix pomatia is used widely to pro- (14, 25-27). Metabolite profiling in biological fluids can help un- derstand the metabolic perturbation of biological systems with comprehensive insight by comparing many metabo- Authors' Affiliations: 1Life/Health Division, Korea Institute of Science lites of individual or populations simultaneously (28-31). and Technology and 2Department of Chemistry, Yonsei University, However, the unwanted transformation of steroids dur- Seoul, Korea ing hydrolysis can result in an underestimation or over- Note: Supplementary data for this article are available at Cancer Epide- estimation of their concentrations. Because most of the miology, Biomarkers & Prevention Online (http://cebp.aacrjournals.org/). cancer progression is correlated with either the inhibition Corresponding Author: Man Ho Choi, Life/Health Division, Korea Institute of Science and Technology, 39-1 Hawolkok-dong, Seoul or promotion of targets in specific molecular pathways, 136-791, Korea. Phone: 82-2-958-5081; Fax: 82-2-958-5059. E-mail: the tests of multiple biomarkers would be focused on [email protected] metabolic pathways and not on a single molecule (32). doi: 10.1158/1055-9965.EPI-09-0581 Although the analytic conditions for this particular study ©2010 American Association for Cancer Research. were optimized, the relative amounts of biomarkers are

388 Cancer Epidemiol Biomarkers Prev; 19(2) February 2010

Systematic Error in Hydrolysis of Urinary Steroids

Table 1. GC-MS information for quantitative analysis of the steroids studied

Compounds Abbreviation Exact Molecular TMS-derivitized Ion Retention (trivial name) mass ion ions* selected† time (min)

Androgens 5β-Androstan-3α,17α-diol βαα-diol 292.24 436.32 256, 241, 346, 331, 436, 421 256 11.68 5β-Androstan-3β,17α-diol ββα-diol 292.24 436.32 256, 241, 346, 331, 436, 421 256 12.41 5α-Androstan-3α,17α-diol ααα-diol 292.24 436.32 241, 256, 331, 346, 436, 421 241 12.54 5β-Dihydrotestosterone 5β-DHT 290.22 434.30 434, 405, 419 434 12.98 Androsterone An 290.22 434.30 419, 434, 329 434 14.78 Etiocholanolone Etio 290.22 434.30 419, 434, 329 434 14.96 5β-Androstan-3β,17β-diol βββ-diol 292.24 436.32 256, 241, 346, 331, 421, 436 256 15.15 5α-Androstan-3α,17β-diol ααβ-diol 292.24 436.32 241, 256, 331, 346, 436, 421 241 15.52 5β-Androstan-3α,17β-diol βαβ-diol 292.24 436.32 256, 241, 346, 421, 331, 436 256 15.61 5α-Androstan-3β,17α-diol αβα-diol 292.24 436.32 421, 241, 346, 256, 331, 436 241 16.52 Epidihydrotestosterone Epi-DHT 290.22 434.30 434, 405, 419 434 16.95 11-Keto-androsterone 11-keto-An 304.20 520.32 415, 520, 505 520 17.05 11-Keto-etiocholanolone 11-keto-Etio 304.20 520.32 415, 505, 520 520 17.15 Dehydroepiandrosterone DHEA 288.21 432.29 432, 417, 327 432 17.34 Epiandrosterone Epi-An 290.22 434.30 419, 434, 329 419 17.59 Androstenediol A-diol 290.22 434.30 239, 344, 254, 329, 434, 419 434 18.08 Androstanedione 5α-dione 288.21 432.29 275, 432, 417, 290 432 18.10 Epitestosterone Epi-T 288.21 432.29 432, 417, 327 432 18.27 5α-Androstan-3β,17β-diol αββ-diol 292.24 436.32 241, 421, 346, 256, 331, 436 241 18.35 5αDihydrotestosterone 5αDHT 290.22 434.30 434, 405, 419 434 18.83 Androstenedione A-dione 286.19 430.27 430, 415, 325 430 19.28 Testosterone T 288.21 432.29 432, 417, 301 432 20.02 11β-Hydroxyandrosterone 11β-OH-An 306.22 522.34 522, 327, 507, 417 522 20.23 11β-Hydroxyetiocholanolone 11β-OH-Etio 306.22 522.34 522, 417, 507, 327 522 20.55 16α-Hydroxy-DHEA 16α-OH-DHEA 304.20 520.32 505, 520, 415 505 28.05 16α-Hydroxy- 16α-OH-A- 302.19 518.32 503, 518, 430 503 30.43 androstenedione‡ dione Estrogens 17α-Estradiol 17α-E2 272.18 416.26 416, 285, 401 416 18.12 Estrone E1 270.16 414.24 414, 399, 309 414 18.63 17β-Estradiol 17β-E2 272.18 416.26 416, 285, 401 416 19.48 4-Methoxyestrone 4-MeO-E1 300.17 444.25 444, 429, 414 444 22.24 4-Methoxy-17β-estradiol 4-MeO-E2 302.19 446.27 446, 315, 325, 416 446 23.22 2-Methoxyestrone 2-MeO-E1 300.17 444.25 444, 429, 414 444 24.06 2-Methoxy-17β-estradiol 2-MeO-E2 302.19 446.27 446, 315, 416, 431 446 25.05 2-Hydroxyestrone 2-OH-E1 286.16 502.28 502, 487, 397 502 25.42 2-Hydroxy-17β-estradiol 2-OH-E2 288.17 504.29 504, 489, 373 504 26.26 4-Hydroxyestrone 4-OH-E1 286.16 502.28 502, 487, 397 502 26.87 4-Hydroxy-17β-estradiol 4-OH-E2 288.17 504.29 504, 373, 489 504 27.97 17-Epiestriol 17-epi-E3 288.17 504.29 504, 345, 311, 386, 297, 489 504 28.72 Estriol E3 288.17 504.29 504, 345, 311, 386, 297, 489 504 29.40 16-Keto-17β-estradiol 16-keto-E2 286.16 502.28 487, 502, 399 487 29.70 16α-Hydroxyestrone 16α-OH-E1 286.16 502.28 487, 502, 399 487 29.70 16-Epiestriol 16-epi-E3 288.17 504.29 504, 345, 311, 386, 297, 489 504 30.76 2-Hydroxyestriol 2-OH-E3 304.17 592.33 592, 433, 385 592 36.97 Progestins 5β-Dihydroprogesterone 5β-DHP 316.20 388.24 445, 460, 355 445 19.49 Epipregnanolone Epi-P-one 318.26 462.33 447, 462, 357 447 22.81 Pregnanolone P-one 318.26 462.33 447, 462, 357 447 23.12

(Continued on the following page)

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 389

Moon et al.

Table 1. GC-MS information for quantitative analysis of the steroids studied (Cont'd)

Compounds Abbreviation Exact Molecular TMS-derivitized Ion Retention (trivial name) mass ion ions* selected† time (min)

Allopregnanolone Allo-P-one 318.26 462.33 447, 462, 357 447 23.46 Pregnanediol P-diol 320.27 464.35 117, 269, 284, 347, 449 269 24.52 Pregnanetriol P-triol 336.27 552.39 255, 435, 345, 552 435 25.85 Pregnenolone Preg 316.20 460.32 445, 460, 355 445 26.89 Isopregnanolone Iso-P-one 318.26 462.33 447, 462, 357 447 27.25 5α-Dihydroprogesterone 5α-DHP 316.20 460.32 445, 460, 355 445 28.13 Progesterone Prog 314.22 458.30 458, 443, 370, 353 458 29.46 20α-Hydroprogesterone 20α-DHP 316.20 388.24 460, 445, 370 445 29.80 17α-Hydroxypregnenolone 17α-OH-Preg 332.24 548.35 548, 443, 458 548 32.22 17α-Hydroxyprogesterone 17α-OH-Prog 330.22 546.34 546, 316, 441 546 35.37 11β-Hydroxyprogesterone 11β-OH-Prog 330.22 546.34 546, 531, 458 531 41.08 21-Hydroxypregnenolone‡ 21-OH-Preg 332.24 548.35 548, 458, 533 548 40.46 Corticoids Tetrahydrodeoxycortisol THS 350.25 638.40 548, 281, 458, 355 548 34.61 Tetrahydrodeoxycorticosterone THDOC 334.25 550.37 550, 535, 460 550 35.91 β-Cortolone 366.24 726.43 205, 341, 431, 521, 610 341 37.62 Tetrahydrocortisone THE 364.23 724.42 634, 619, 529 634 38.35 β-Cortol 368.26 728.45 253, 207, 343, 523, 445, 343 39.30 433, 355 α-Cortolone 366.24 726.43 205, 341, 431, 521, 610 341 39.34 Tetra-11-dehydrocorticosterone THA 364.23 724.42 636, 621, 531, 451 636 39.44 Tetrahydrocortisol THF 366.24 726.43 636, 546, 621 636 41.13 Tetrahydrocorticosterone THB 350.25 638.40 638, 623, 548 638 41.42 α-Cortol 368.26 728.45 253, 207, 343, 523, 445, 343 41.60 433, 355 5βDihydrodeoxycorticosterone 5βDHDOC 332.24 548.35 548, 533, 460 548 41.88 Allotetrahydrocortisol Allo-THF 366.24 726.43 636, 546, 621, 531 636 42.22 21-Deoxycortisol 21-deoxyF 346.21 634.37 634, 404, 544 634 42.35 11-Deoxycortisol 11-deoxyF 346.21 634.37 544, 529, 456 544 42.60 11-Deoxycorticosterone 11-deoxyB 330.22 546.34 546, 531, 301 546 43.32 Cortisone E 360.19 720.39 615, 634, 527 615 45.94 11-Dehydrocorticosterone 11-dehydroB 344.20 632.36 617, 632, 401 617 46.75 Allodihydrocorticosterone Allo-DHB 348.23 636.39 636, 621, 531, 546 636 46.85 Allodihydrocortisol Allo-DHF 364.23 724.42 634, 619, 529, 544 634 46.98 20α-Dihydrocortisone 20α-DHE 362.21 722.41 439, 617, 517, 527 617 47.46 Corticosterone B 346.21 634.37 634, 619, 544, 529 634 47.80 Cortisol F 362.21 722.41 632, 617, 542, 527 632 47.92 20α-Dihydrocortisol 20α-DHF 364.22 724.42 531, 519, 429, 339 531 48.45 Sterols Cholesterol Chol 386.35 458.39 329, 368, 353, 458, 443 458 40.55 Desmolesterol 384.33 456.38 343, 327, 366, 441, 456 343 42.30 Lanosterol 428.40 498.43 393, 498, 483 393 47.67 20α-hydroxycholesterol 20α-OH-Chol 402.35 546.43 201, 461, 281 461 48.06 24S-Hydroxycholesterol 24S-OH-Chol 402.35 546.43 413, 503, 456, 546 413 49.05 Cholestenone‡ 384.33 456.38 456, 441 456 43.54

NOTE: Principal ions are given as within 30% of the base peak. *All steroids were derivitized with the trimethylsilylation agents, N-methyl-N-trifluorotrimethylsilyl acetamide/ammonium iodide/ dithioerythritol (500:4:2, v/w/w) for both the hydroxyl and keto groups of the steroids. †Quantitative ions as the TMS derivatives of steroids. ‡Additional steroids were analyzed to confirm the transformation or identify the by-product derived from Helix pomatia.

390 Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 Cancer Epidemiology, Biomarkers & Prevention

Systematic Error in Hydrolysis of Urinary Steroids

affected by the presence of unrelated and noncancer cell HLB SPE cartridges placed in a device fitted with a type from the samples. In turn, this would ultimately small peristaltic pump and operated at a low flow rate provide a means to establish a threshold concentration (<1 mL/min). After loading the sample onto the car- in absolute terms, beyond which a test sample would tridge, it was washed with 2 mL water and eluted twice be deemed to contain biomarker levels indicative of the with 2 mL of methanol. The combined eluate was evap- disease state. orated under a nitrogen stream for the next two differ- Understanding the actions of steroid hormones in ent enzymatic hydrolysis steps. (a) For the hydrolysis of mammary carcinogenesis is critical for developing meth- both glucuronide and sulfate conjugates, the dried elu- ods of diagnosing, preventing, and treating breast, thy- ate was added to 1 mL of 0.2 mol/L acetate buffer (pH roid, and prostate cancers (33-39). In addition, the exact 5.2), 100 μL of 0.2% ascorbic acid, and 50 μLofβ-glu- quantification is at the center of clinical applications; curonidase/arylsulfatase solution. The resulting mixture without reliable methods to accurately quantify differen- was then incubated at 55°C for 3 h. (b)Tohydrolyze tially expressed biomolecules, it would not be possible to glucuronide conjugates only, the dried eluate was added identify disease biomarkers. Here, we describe the unex- to 1 mL of 0.2 mol/L acetate buffer (pH 7.2), 100 μLof pected transformation phenomena of urinary steroids, in- 0.2% ascorbic acid, and 50 μLofβ-glucuronidase and cluding androgens, estrogens, corticoids, progestins, and then incubated at 55°C for 1 h. After enzymatic hydro- sterol, in the presence of two difference enzyme systems lysis, the solution was extracted twice with 2.5 mL of to establish experimental protocols in GC-MS–based ethyl acetate/n-hexane (2:3, v/v). The organic solvent quantitative steroid profiling. was evaporated in an N2 evaporator at 40°C and further

Materials and Methods

Chemicals The 84 steroids examined in this study (Table 1) were obtained from Sigma-Aldrich, Steraloids, and NARL. The d internal standards (IS), 16,16,17- 3-testosterone and d methyltestosterone for 25 androgens, 2,4,16,16- 4-estradiol d for 17 estrogens, 9,11,12,12- 4-cortisol for 23 corticoids, α d 2,2,4,6,6,17 ,21,21,21- 9-progesterone and 2,2,4,6,6,21,21, d α 21- 8-17 -hydroxyprogesterone for 14 progestins, and d 2,2,3,4,4,6- 6-cholesterol for 5 sterols were purchased from NARL and C/D/N isotopes. Sodium phosphate monobasic (reagent grade), sodium phosphate dibasic (reagent grade), sodium acetate (reagent grade, anhydrous), acetic acid (glacial, 99.99+%), and L-ascorbic acid (reagent grade) was obtained from Sigma-Aldrich. A solution of β-glucuronidase/arylsulfatase from Helix pomatia [aqueous solution stabilized with 0.01% thiomerosal: β-glucuronidase (100,000 Fishman U/mL) and sulfatase (800,000 Roy U/mL)] and a 50% glycerol solution of β-glucuronidase extracted from E. coli (140 U/mL) were obtained from Roche Diagnostics GmbH. The trimethylsilylating agents, N-methyl-N- trifluorotrimethylsilyl acetamide (MSTFA), ammonium iodide, and dithioerythritol, were purchased from Sigma. All organic solvents used in analytic and high perfor- mance liquid chromatography grade were purchased from Burdick & Jackson. The deionized water was prepared using a Milli-Q purification system. Figure 1. Representative chemical structures of the steroids showing a Urinary Steroid Profiling relationship between the substrate and by-products with Helix pomatia. There are three different types of structures in the steroids producing The quantitative metabolite profiling of urinary ster- by-products after enzymatic hydrolysis: (A) 3β-hydroxy-5-ene oids was achieved based on previous reports (11, 40). steroid (e.g., DHEA, A-diol, Preg, 17α-OH-Preg, 16α-OH-DHEA, Briefly, the urine samples (2 mL) were added to 20 μL 21-OH-Preg, Chol, 24S-OH-Chol, 20α-OH-Chol, and desmosterol), μ d d (B) 3β-hydroxy-5α–reduced steroid (e.g., αββ-diol, αβα-diol, Epi-An, of the 7 ISs (1 g/mL 3-testosterone and 4-estradiol; 5 μ d d α Iso-P-one, lanosterol), and (C) 3β-hydroxy-5β–reduced steroid (e.g., g/mL 4-cortisol and 8-17 -hydroxyprogesterone; and ββα-diol, βββ-diol, Epi-P-one). See Table 1 for the full names of the steroid μ d d 10 g/mL methyltestosterone, 9-progesterone, and 6- hormones and Supplementary Fig. S1 for mass spectral identification of cholesterol). The samples were extracted with Oasis the by-products generated by Helix pomatia.

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 391

Moon et al.

Figure 2. The effect of incubating β-glucuronidase/arylsulfatase in an acetate buffer with increasing concentrations of substrates. Solid and dotted lines, the substrates (A, DHEA; B, A-diol; C, Preg; D, 17α-OH-Preg; E, αββ-diol; F, αβα-diol.

dried in a vacuum desiccator over P2O5-KOH for at and quality control were prepared in house as steroid- least 30 min. Finally, the dried residue was derivatized free urine (41). with N-methyl-N-trifluorotrimethylsilyl acetamide/ammonium iodide/dithioerythritol (40 μL; 500:4:2, v/w/ Enzyme-Based Transformation of Steroids w) at 60°C for 20 min, and 2 μL of the resulting mixture The pure steroid standards were examined individual- weresubjectedtoGC-MSinselected-ionmonitoring ly to confirm the conversion phenomena of steroids in the (SIM) mode. presence of β-glucuronidase/arylsulfatase and β-glucuronidase solutions only. After evaporating the standard solution added to known amounts, the dried standard Standard Solution and Quality-Control Sample was incubated, extracted, and derivatized using the Each stock solution of the reference standards was pre- methods described above. Acquisition was done in scan pared at a concentration of 1,000 μg/mL in methanol and mode (m/z 100-650) to detect the by-products, and their the working solutions were made up with methanol at peak identification was achieved by comparing the reten- various concentrations ranging from 0.1 to 10 μg/mL. tion times and matching the mass spectra with those of L-ascorbic acid (1 mg/mL) was used to prevent the oxi- the reference standards. dation of catechol estrogens. All standard solutions were Each working solution of steroids was prepared at six stored at − 20°C until needed and they were stable for a different concentrations (1, 5, 20, 50, 100, and 200 ng/mL) minimum of 3 mo. The urine samples for the calibration to examine the calibration linearity of the deconjugated

392 Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 Cancer Epidemiology, Biomarkers & Prevention

Systematic Error in Hydrolysis of Urinary Steroids

Figure 2 Continued. G, Epi-An; H, Iso-P-one; I, βββ-diol; J, Epi-P-one; K, ββα-diol) lost and the corresponding by-products generated with Helix pomatia in each sample, respectively. Semiquantitative results plotted as a ratio of the analyte to the IS show the extent of unwanted transformations in the presence of increasing amounts of substrate.

steroids treated with 50 μL of the enzyme solutions. For gle-quadrupole Agilent 5975 MSD. The electron energy the within-day repeatability, triplicates were analyzed, was 70 eV and the ion source temperature was 230°C. whereas the reproducibility was measured from the sam- Each sample (2 μL) was injected in split mode (10:1) ples run over 4 different days. In addition, the unexpected at an injector temperature of 280°C and was separated transformation of steroids was evaluated using a steroid- through an Ultra-1 capillary column (25 m × 0.2 mm profiling procedure in the same urine samples obtained i.d., 0.33 μm, film thickness; Agilent Technologies). from two healthy male and female volunteers (ages 21 The oven temperature was initially 215°C, which and 20 y, respectively) in triplicate. The differences in was ramped to 260°C at 1°C/min and then finally in- the steroid concentrations derived from the two different creased to 320°C (hold for 1 min) using a 15°C/min enzymes are represented as fold units by dividing the ramping program. Ultrahigh purity helium was used concentrations of β-glucuronidase/arylsulfatase by those as the carrier gas with a column head pressure of of β-glucuronidase. Data processing and illustration were 210.3 kPa (column flow, 1.0 mL/min at an oven tem- carried out using Microsoft Office Excel 2007 (Microsoft perature of 215°C). For quantitative analysis, the char- Corp.) and SigmaPlot (version 10.0, Systat Software, Inc.). acteristic ions of each steroid were determined as their trimethylsilyl (TMS) derivatives. Peak identification Gas Chromatography-Mass Spectrometry was achieved by comparing the retention times and GC-MS analysis was carried out using an Agilent by matching the peak height ratios of the character- 6890 Plus gas chromatograph interfaced with a sin- istic ions (Table 1).

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 393

Moon et al.

Results and Discussion Supplementary Fig. S1). In particular, both desmolsterol (A-10) and lanosterol (B-5), which have a double bond During enzymatic hydrolysis using β-glucuronidase/ at C-24 in contrast to 24S-OH-Chol (A-8) or 20α-OH- arylsulfatase of Helix pomatia, many transformation phe- Chol (A-9), showed a small amount of by-product that nomena were observed and the calibration linearity of might be affected by steric hindrance. Among the five some steroids was in a narrow dynamic range. This is sterols, cholestenone as a by-product of Chol (A-7) was in contrast to β-glucuronidase of E. coli, which did not confirmed using the reference standard and the other lead to by-products, and the signal of the substrate at four corresponding by-products were identified from concentrations <1 ng/mL was also detectable. To con- their mass spectra. In addition, Iso-P-one (B-4), as a firm these unwanted transformation phenomena, the 3β-hydroxy-5α-reduced steroid, transformed into 5α- pure reference standards were examined individually DHP, as a 3-oxo-5α–reduced steroid, resulting in two with β-glucuronidase/arylsulfatase and were analyzed chromatographic peaks (27.25 and 28.72 minutes) be- to identify the major by-products. Because β-glucuroni- cause of the nonselective derivatization. Pregnane deri- dase from E. coli does not lead to by-products (12-14, vatives with a 20-keto-21-methyl side chain without a 16), the individual experiment with this enzyme 17α-hydroxy or 21-hydroxy group never led to single was excluded. Among the 84 steroids monitored, 18 reaction products due to the two most stable side chain compounds were transformed into substrate-derived conformations (42, 43). The retention times (28.13 and by-products. The total ion chromatograms and mass 29.35 minutes, respectively) and mass spectra of the spectra of the TMS derivates of the substrates lost and two peaks were compared with those of the reference by-products generated by Helix pomatia were compared standards. As 3β-hydroxy-5β–reduced steroids were (Supplementary Fig. S1). In addition to the previously also believed to have been converted to 3-oxo-5β–reduced reported conversion of DHEA (A-1) to A-dione, A-diol steroids, EpiP-one (C-2, 19.38 and 20.90 minutes) was (A-2) to T, and Preg (A-3) to Prog (14, 16-20), additional transformedinto5β-DHP (22.71 and 24.16 minutes). transformation was found as follows: the major product Although it was not directly identified with the reference of 17α-OH-Preg (A-4) was clearly consistent with a standards, the product of ββα-diol (C-3) from incubation molecular ion at m/z 546 corresponding to 17α-OH- with Helix pomatia was assumed to be a 3-oxo-5β–reduced Prog. The by-products of 5α-androstane-3β,17β-diol steroid (5β-androstan-17α-ol-3-one) because the signifi- (B-1), 5α-androstane-3β,17α-diol (B-2), and 5β-andros- cant ion at m/z 434 was obtained in the mass spectrum of tane-3β,17β-diol (C-1) were also observed as 5α-DHT, the product. In the case of estrogens, no transformation Epi-DHT, and 5β-DHT, respectively. These three pro- was observed, and they could be evaluated in any enzy- ducts showed a molecular ion and major fragment at matic hydrolysis to quantify either the glucuronide or m/z 434 and m/z 405 and 419. In addition, the major sulfate conjugates, or both. product of Epi-An (B-3) showed a molecular ion of The effects of incubating β-glucuronidase/arylsulfa- m/z 432 and fragment ions of m/z 417 and 327. Accord- tase in an acetate buffer along with increasing substrate ing to the general scheme of the relationship between concentrations were examined by comparing the peak the substrate and by-products with Helix pomatia,the height ratios of the analyte to that of the IS (Fig. 2). By conversion of 3β-hydroxy-5-ene steroids to 3-oxo-4-ene plotting the analyte to IS ratio, the semiquantitative re- steroids was previously reported (16-19). However, sults showed the extent of the unwanted transformations unexpected transformations of 3β-hydroxy-5α–reduced described above in the presence of increasing amounts of and 3β-hydroxy-5β–reduced steroids into 3-oxo-5α– substrate. The 3β-hydroxy-5-ene (Fig. 2A-D), 3β-hy- reduced and 3-oxo-5β–reduced steroids, respectively, droxy-5α–reduced (Fig. 2E-H), and 3β-hydroxy-5β– were newly defined in this study (Fig. 1). However, reduced (Fig. 2I-K) steroids could not be detected at no steroids with 3α-hydroxy-5-ene, 3α-hydroxy-5α concentrations as low as 1 to 20 ng/mL, whereas the or 5β–reduced, and 3-oxo-5α or 5β–reduced structure corresponding by-products were generated. All the by- produced any by-products, which are accordance with products derived from the substrates tended to saturate a previous report (16). These results suggest that Helix in the concentration range of 50 to 100 ng/mL (Fig. 2). pomatia also contains cholesterol oxidase in addition to This suggests that the loss of substrate and the formation 3β-hydroxysteroid oxidoreductase/3-oxosteroid-4,5-ene of by-products from Helix pomatia are dependent on the isomerase and 6-hydroxylase, 6-hydroxysteroid as addi- substrate concentration becoming saturated with 50 μLof tional enzymes (16-18, 20-24). the enzyme in the present conditions. In addition, the As 3β-hydroxy-5-ene steroids, 16α-OH-DHEA (A-5) loss of steroid and the generation of by-products could and 21-OH-Preg (A-6) were converted to 3-oxo-4-ene reach a saturation point, which also decreased in the steroids of 16α-OH-A-dione and 21-OH-Prog, respec- presence of a competing substrate (16). tively. The sterol compounds, such as Chol, 24S-OH- Both enzyme systems with β-glucuronidase and β-glu- Chol, 20α-OH-Chol, desmolsterol, and lanosterol also curonidase/arylsulfatase were applied to real urine sam- produced by-products, but these compounds were as- ples obtained from two healthy male and female sumed to be less affected than the other 3β-hydroxy- volunteers. The resulting concentrations of the 84 urinary 5-ene steroids or 3β-hydroxy-5α–reduced steroids. (see steroids were compared (Supplementary Table S1), and

394 Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 Cancer Epidemiology, Biomarkers & Prevention

Systematic Error in Hydrolysis of Urinary Steroids

Figure 3. The changes in the concentrations of steroids derived from two different enzymes using the same urine samples. The results are represented as fold unit by dividing the concentrations of β-glucuronidase/arylsulfatase by those of β-glucuronidase, which showed a >2-fold increase in the 24 steroids.

the extraction yield of each steroid in Helix pomatia was can affect the accuracy of the assay. Although this ex- generally higher than that of E. coli. Some urinary ster- periment has a limitation on the small number of sam- oids detected in the β-glucuronidase system could not ples, there was a >2-fold difference in the concentrations be detected in the β-glucuronidase/arylsulfatase sys- of the 24 steroids obtained from β-glucuronidase/aryl- tem, which might be decomposed into 3β-hydroxy ster- sulfatase and β-glucuronidase only. In the cases of oids, because ββα-diol, αβα-diol, Iso-P-one, and DHEA, Epi-An, A-diol, αββ-diol, A-dione, 16α-OH- lanosterol were found in the male samples, and 20α- DHEA, Preg, 20α-DHP, and B (corticosterone), a >15- OH-Chol was found in female samples. Therefore, the fold change was obtained (Fig. 3). It should be noted use of β-glucuronidase/arylsulfatase can result in a that these steroids may be more prominent in the sulfate lower yield of 3β-hydroxy steroids, whereas the amount conjugates than free and glucuronic conjugates. of 3-oxo steroids can be overestimated. This indicates In summary, the transformation of 3β-hydroxy-5-ene that the use of a Helix pomatia extract in steroid analysis steroids into 3-oxo-4-ene steroids has been observed

Table 2. Conversion of steroid compounds after enzymatic hydrolysis with Helix pomatia

Substrate By-product

A series 3β-Hydroxy-5-ene steroids 3-Oxo-4-ene steroids Dehydroepiandrosterone Androstenedione Androstenediol Testosterone Pregnenolone Progesterone 17α-Hydroxypregnenolone 17α-Hydroxyprogesterone 16α-Hydroxy DHEA 16α-Hydroxyandrostenedione 21-Hydroxypregnenolone 21-Hydroxyprogesterone Cholesterol Cholest-4-en-3-one (cholestenone) 24S-Hydroxycholesterol Cholest-4-en-24S-ol-3-one 20α-Hydroxycholesterol Cholest-4-en-20α-ol-3-one Desmolsterol Cholest-4,24-diene-3-one B series 3β-Hydroxy-5α-reduced steroids 3-Oxo-5α-reduced steroids 5α-Androstan-3β,17β-diol Dihydrotestosterone 5α-Androstan-3β,17α-diol Epidihydrotestosterone Epiandrosterone 5α-Androstenedione Isopregnanolone 5α-Dihydroprogesterone Lanosterol Cholest-4,4-dimethyl-8,24-diene-3-one C series 3β-Hydroxy-5β-reduced steroids 3-Oxo-5β–reduced steroids 5β-Androstan-3β,17β-diol 5β-Dihydrotestosterone Epipregnanolone 5β-Dihydroprogesterone 5β-androstan-3β,17α-diol 5β-Androstan-17α-ol-3-one

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 395

Moon et al.

previously, whereas the 3α-hydroxy and 3-oxo-steroids Disclosure of Potential Conflicts of Interest did not produce any by-products during enzymatic hydrolysis with Helix Pomatia (Table 2). This study dealt No authors declared any potential conflicts of interest. with 84 urinary steroids, including 3β-hydroxy-5-ene steroids, which can be analyzed by GC-MS combined with hydrolysis procedures. The 3β-hydroxy-5α-re- Grant Support duced and 3β-hydroxy-5β–reduced steroids showed a transformation to 3-oxo-5α–reduced and 3-oxo-5β– Intramural grant from the Korean Institute of Science and Technology, and by grants from the National R&D Program of the Korean Ministry of reduced steroids, respectively. Although the use of anti- Education, Science and Technology and the Korean Science and Engineer- oxidant improves yield in some urinary steroids, it is ing Foundation (KOSEF). The costs of publication of this article were defrayed in part by the not easy to suggest the best condition of enzymatic hy- payment of page charges. This article must therefore be hereby marked drolysis for experimental purposes. However, these re- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate sults could indicate variability in different enzymatic this fact. – hydrolyses combined with GC-MS based steroid profil- Received 6/17/09; revised 11/4/09; accepted 11/30/09; published ing in clinical applications. OnlineFirst 1/19/10.

References 1. You L. Steroid hormone biotransformation and xenobiotic induction conditions using response surface methodology. Analyst 2000;125: of hepatic steroid metabolizing enzymes. Chem Biol Interact 2004; 2255–9. 147:233–46. 15. Choi MH, Hahm JR, Jung BH, Chung BC. Measurement of corticoids 2. Choi MH, Kim KR, Chung BC. Simultaneous determination of urinary in the patients with clinical features indicative of mineralocorticoid androgen glucuronides by high temperature gas chromatography- excess. Clin Chim Acta 2002;320:95–9. mass spectrometry with selected ion monitoring. Steroids 2000;65: 16. Christakoudi S, Cowan DA, Taylor NF. Sodium ascorbate improves 54–9. yield of urinary steroids during hydrolysis with Helix pomatia juice. 3. Choi MH, Kim JN, Chung BC. Rapid HPLC-electrospray tandem Steroids 2008;73:309–19. mass spectrometric assay for urinary testosterone and dihydrotes- 17. Vanlucheneb E, Eechaute W, Vandekerckhove D. Conversion of free tosterone glucuronides from patients with benign prostate hyperpla- 3β-hydroxy-5-ene-steroids by incubation with Helix pomatia. J Ste- sia. Clin Chem 2003;49:322–5. roid Biochem 1982;16:701–3. 4. Raffaelli A, Saba A, Vignali E, Marcocci C, Salvadori P. Direct deter- 18. Messeri G, Cugnetto G, Moneti G, Serio M. Helix pomatia induced mination of the ratio of tetrahydrocortisol+allo-tetrahydrocortisol to conversion of some 3β-hydroxysteroids. J Steroid Biochem 1984; tetrahydrocortisone in urine by LC-MS-MS. J Chromatogr B 2006; 20:793–6. 830:278–85. 19. Schmidt NA, Borburgh HJ, Penders TJ, Weykamp CW. Steroid pro- 5. Vandenput L, Mellström D, Lorentzon M, et al. Androgens and glu- filing-an update. Clin Chem 1985;31:637–9. curonidated androgen metabolites are associated with metabolic risk 20. Houghton E, Grainger L, Dumasia MC, Teale P. Application of gas factors in men. J Clin Endocrinol Metab 2007;92:4130–7. chromatography/mass spectrometry to steroid analysis in equine 6. Wong T, Shackleton CH, Covey TR, Ellis G. Identification of the ster- sports: problems with enzyme hydrolysis. Org Mass Spectrom oids in neonatal plasma that interfere with 17 α-hydroxyprogesterone 1992;27:1061–70. radioimmunoassays. Clin Chem 1992;38:1830–7. 21. Higashi T, Takayama N, Shimada K. Enzymic conversion of 3β-hy- 7. Hsing AW, Stanczyk FZ, Bélanger A, et al. Reproducibility of serum droxy-5-ene-steroids and their sulfates to 3-oxo-4-ene-steroids for sex steroid assays in men by RIA and mass spectrometry. Cancer increasing sensitivity in LC-APCI-MS. J Pharm Biomed Anal 2005; Epidemiol Biomarkers Prev 2007;16:1004–8. 39:718–23. 8. Lee SH, Kim SO, Kwon SW, Chung BC. Androgen imbalance in pre- 22. Smith AG, Brooks CJ. Cholesterol oxidases: Properties and applica- menopausal women with benign breast disease and breast cancer. tions. J Steroid Biochem 1976;7:705–13. Clin Biochem 1999;32:375–80. 23. Labaree D, Hoyte RM, Hochberg RB. A direct stereoselective synthe- 9. Caulfield MP, Lynn T, Gottschalk ME, et al. The diagnosis of congen- sis of 7β-hydroxytestosterone. Steroids 1997;62:482–6. ital adrenal hyperplasia in the newborn by gas chromatography/mass 24. Hauser B, Schulz D, Boesch C, Deschner T. Measuring urinary testos- spectrometry analysis of random urine specimens. J Clin Endocrinol terone levels of the great apes-problems with enzymatic hydrolysis Metab 2002;87:3682–90. using Helix pomatia juice. Gen Comp Endocrinol 2008;158:77–86. 10. Homma K, Hasegawa T, Masumoto M, et al. Reference values for 25. Shackleton CHL. Profiling steroid hormones and urinary steroids. urinary steroids in Japanese newborn infants: gas chromatography/ J Chromatogr 1986;379:91–156. mass spectrometry in selected ion monitoring. Endocr J 2003;50: 26. Shibasaki H, Tanabe C, Furuta T, Kasuya Y. Hydrolysis of conjugated 783–92. steroids by the combined use of β-glucuronidase preparations from 11. Moon JY, Jung HY, Moon MH, Chung BC, Choi MH. Inclusion com- Helix pomatia and ampullaria: determination of urinary cortisol and its plex-based solid-phase extraction of steroidal compounds with en- metabolites. Steroids 2001;66:795–801. trapped β-cyclodextrin polymer. Steroids 2008;73:1090–7. 27. Taylor JI, Grace PB, Bingham SA. Optimization of conditions for the 12. Wakabayashi M, Fishman WH. The comparative ability of β-glucu- enzymatic hydrolysis of phytoestrogen conjugates in urine and plas- ronidase preparations (liver, Escherichia coli, Helix pomatia, and ma. Anal Biochem 2005;341:220–9. Patella vulgata) to hydrolyze certain steroid glucosiduronic acids. 28. Fiehn O. Metabolomics-the link between genotypes and phenotypes. J Biol Chem 1961;236:996–1001. Plant Mol Biol 2002;48:155–71. 13. Graef V, Furuya E, Nishikaze O. Hydrolysis of steroid glucuronides 29. Lee SH, Woo HM, Jung BH, et al. Metabolomic approach to evaluate with β-glucuronidase preparations from bovine liver, Helix pomatia, the toxicological effects of nonylphenol with rat urine. Anal Chem and E. coli. Clin Chem 1977;23:532–5. 2007;79:6102–10. 14. Ferchaud V, Courcoux P, Le Bizec B, Monteau F, André F. Enzymatic 30. Lindon JC, Holmes E, Wilson ID, Nicholson JK. Metabolic phenotyp- hydrolysis of conjugated steroid metabolites: search for optimum ing in health and disease. Cell 2008;134:714–7.

396 Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 Cancer Epidemiology, Biomarkers & Prevention

Systematic Error in Hydrolysis of Urinary Steroids

31. Kaddurah-Daouk R, Kristal BS, Weinshilboum RM. Metabolomics: a steroid hormones in breast adipocytes: methods and implications. global biochemical approach to drug response and disease. Annu Cancer Epidemiol Biomarkers Prev 2008;17:1891–5. Rev Pharmacol Toxicol 2008;48:653–83. 39. Eliassen AH, Missmer SA, Tworoger SS, Hankinson SE. Circulating 32. Sawyers CL. The cancer biomarker problem. Nature 2008;452: 2-hydroxy- and 16α-hydroxy estrone levels and risk of breast cancer 548–52. among postmenopausal women. Cancer Epidemiol Biomarkers Prev 33. Zhao XY, Malloy PJ, Krishnan AV, et al. Glucocorticoids can promote 2008;17:2029–35. androgen-independent growth of prostate cancer cells through a 40. Moon JY, Jung HY, Moon MH, Chung BC, Choi MH. Heat-Map vi- mutated androgen receptor. Nature Med 2000;6:703–6. sualization of gas chromatography-mass spectrometry based quan- 34. Bosland MC. The role of steroid hormones in prostate carcinogene- titative signatures on steroid metabolism. J Am Soc Mass Spectrom sis. J Natl Cancer Inst Monogr 2000;27:39–66. 2009;20:1626–37. 35. Coffey DS. Similarities of prostate and breast cancer: evolution, diet, 41. Cho YD, Choi MH. Alternative sample preparation techniques in gas and estrogens. Urology 2001;57:31–8. chromatographic-mass spectrometric analysis of urinary androgenic 36. Chan EK, Sepkovic DW, Bowne HJY, Yu GP, Schantz SP. A hormon- steroids. Bull Kor Chem Soc 2006;27:1315–22. al association between estrogen metabolism and proliferative thyroid 42. Chambaz EM, Defaye G, Madani C. Trimethylsilyl ether-enol-tri- disease. Otolaryngol Head Neck Surg 2006;134:893–900. methylsilyl ether. New type of derivative for the gas phase study of 37. Micheli A, Meneghini E, Secreto G, et al. Plasma testosterone and hormonal steroids. Anal Chem 1973;45:1090–8. proghosis of postmenopausal breast cancer patients. J Clin Oncol 43. Duax WL, Griffin JF, Rohrer DC. Conformation of progesterone side 2007;25:2685–90. chain: conflict between x-ray data and force-field calculations. J Am 38. Falk RT, Gentzschein E, Stanczyk FZ, et al. Measurement of sex Chem Soc 1981;103:6705–12.

www.aacrjournals.org Cancer Epidemiol Biomarkers Prev; 19(2) February 2010 397

Systematic Error in Gas Chromatography-Mass Spectrometry− Based Quantification of Hydrolyzed Urinary Steroids

Ju-Yeon Moon, Young Wan Ha, Myeong Hee Moon, et al.

Cancer Epidemiol Biomarkers Prev 2010;19:388-397. Published OnlineFirst January 19, 2010.

Updated version Access the most recent version of this article at: doi:10.1158/1055-9965.EPI-09-0581

Supplementary Access the most recent supplemental material at: Material http://cebp.aacrjournals.org/content/suppl/2010/01/18/1055-9965.EPI-09-0581.DC1

Cited articles This article cites 43 articles, 9 of which you can access for free at: http://cebp.aacrjournals.org/content/19/2/388.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cebp.aacrjournals.org/content/19/2/388.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cebp.aacrjournals.org/content/19/2/388. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.