Accepted Manuscript Title: Quantitative determination of 26 steroids in eggs from various species using liquid chromatography–triple quadrupole-mass spectrometry Author: Xiaoxia Mi Sicong Li Yanhua Li Kaiqiang Wang Dan Zhu Gang Chen PII: DOI: Reference:

S0021-9673(14)00886-3 http://dx.doi.org/doi:10.1016/j.chroma.2014.05.084 CHROMA 355481

To appear in:

Journal of Chromatography A

Received date: Revised date: Accepted date:

11-2-2014 13-5-2014 31-5-2014

Please cite this article as: X. Mi, S. Li, Y. Li, K. Wang, D. Zhu, G. Chen, Quantitative determination of 26 steroids in eggs from various species using liquid chromatographyndashtriple quadrupole-mass spectrometry, Journal of Chromatography A (2014), http://dx.doi.org/10.1016/j.chroma.2014.05.084 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 

Quantitative determination of 26 steroids in eggs from various species using liquid

2 

chromatography–triple quadrupole-mass spectrometry Xiaoxia Mi, Sicong Li, Yanhua Li, Kaiqiang Wang, Dan Zhu, Gang Chen*

4  5 

Key Laboratory of Agro-product Quality and Safety, Institute of Quality Standards & Testing Technology for Agro-products, Chinese Academy of Agricultural Sciences, Beijing, 100081, China

6  7  8 

* Corresponding author. Tel.: +86-10-82106559 Fax: +86-10-82106560 E-mail address: [email protected] (Gang Chen)

9 

Abstract

us

cr

ip t

3 

A method for analyzing 26 types of steroids in egg matrix was developed. The method used liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) in electrospray ionization mode (ESI). The procedure involved extraction with acetonitrile and removal of phospholipids by zinc chloride (ZnCl2) followed by SPE cleanup with a Plexa cartridge. The effect of ZnCl2 on phospholipid removal was directly observed using the post column infusion procedure. The SPE washing and elution conditions were optimized using a shallow gradient procedure. The free and conjugated steroids forms were determined using enzyme hydrolysis. The developed method resulted in satisfactory precision (RSD ≤ 15%), and the limits of quantification were between 0.05 and 25 ng/g depending on the steroid types. The recoveries ranged from 63.2% to 121.5%. Finally, the developed method was successfully applied to compare the steroids in eggs from different species (i.e., hen, duck, quail and pigeon eggs) or different raising system (i.e., normal vs. organic eggs). The steroids can be clearly clustered according to species and raising system. The hierarchical clustering analysis indicated similarity of the steroids among the species. The developed method is sensitive and useful for detection and quantification of steroids in eggs and can be used for residue control programs. In addition, the observed steroid content will provide a fundamental reference for food risk assessment analysis.

29 

1. Introduction

30  31  32  33  34  35  36  37  38  39 

Steroidal hormones control the physiological functions of mammals. These hormones act as body chemical messengers that are transported in the blood and activate signal transduction via binding to androgen, estrogen, or corticosteroid receptors. In addition, these hormones are vital to normal development, maturation and cell senescence of the body. However, the toxic effect of steroidal hormones has been well documented for human exposure to certain circumstances or excessive amounts. Animal products are a major source of human exposure to steroids. Because steroidal hormones can increase the performance of food-producing animals [1, 2], many steroidal hormones as well as synthesized steroids have been illegally used in the animal husbandry industry. The occurrence of steroidal hormone residue in animal products negatively influences human health. Therefore, monitoring of the illicit use of hormones has been performed by many countries’ national

Ac ce p

te

d

M

an

10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28 

Keywords: Steroids; Eggs; Liquid chromatography-tandem mass spectrometry; Principal component analysis

1   

Page 1 of 23

surveillance programs [3].

us

cr

ip t

Steroids contain a large homologous and isomeric group, and most of the steroids exist at extremely low levels (μg kg-1 to ng kg-1) or in a conjugated form. To efficiently monitor the residual levels of steroids in animal products, a robust analytical method for multiple detections and with high sensitivity is required. An immunoassay based method has been applied for medical diagnostic screening analysis, but the low specificity and instability of the method limit its further application with food matrices. Mass spectrometry with chromatographic separation has been used to analyze steroid residues in animal products [4-7]. GC-MS has been a valuable tool for steroid detection. However, LC-MS has prevailed for analytical method development, especially after significant improvement in the instrument sensitivity in recent years [8]. In comparison to GC-MS, LC-MS does not require a tedious derivatization procedure and can provide more information about the isoforms of steroids. In addition, LC-MS can directly analyze the conjugated form of the steroids, and the sample extraction procedure is easier [9-11].

te

d

M

an

Eggs are important animal derived food for humans. The steroid exposure via egg products has been a concerned to susceptible risk groups (e.g., infants, children, pregnant women and patients with estrogen related breast cancer) [12]. From an analytical perspective, eggs have a complicated matrix. USDA reported that eggs contain 12.6% of protein, 10.6% of fat, and the phospholipids amount to 10% of the yolk weight [13]. The analysis of the steroid profile in the egg matrix has been a challenging task. Liquid–liquid extraction (LLE) [14-16] and solid-phase extraction (SPE) [17-20] has been typically used for biological sample preparation. LC-MS with an ESI source requires a particularly clean sample cleanup procedure due to the matrix interference problem. Although the application of a stable isotope labeled internal standard (IS) can “mask” the matrix effects, the co-eluting matrix still needs to be eliminated to avoid the loss of sensitivity [21, 22]. In addition, all of the ISs are not commercially available, and the use of one IS to calibrate other analytes may provide the wrong result.

Ac ce p

40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60  61  62  63  64  65  66  67  68  69  70  71  72  73  74  75  76  77  78  79  80 

To the best of our knowledge, three publications on steroids analysis in egg matrix have been reported. Hartmann et al. (1998) analyzed steroids in hen eggs by using a GC-MS method [23]. Other studies used LC-MS/MS methods and found certain amount of estrogens [24], and progesterone in egg matrix [11]. Compared with previous studies, our study is a comprehensive method for analyzing a wide range of steroids was developed. The steroids include androgens, estrogens, progestogens, corticosteroids, and steroids-like chemicals. Liquid chromatography coupled with tandem mass spectrometry and an ESI source (LC-MS/MS-ESI) was used. The sample extraction procedures were comprehensively validated to achieve cleaner sample cleanup, lower organic solvent consumption, and efficiency. Eggs from different species were analyzed to validate the practical use of the method. The available data from the literature were compared to support the determined values. We believe the established method can facilitate the analysis of steroid residues in egg products. The obtained steroid data in eggs from different species can provide a fundamental reference to nutritionists for risk assessment.

2   

Page 2 of 23

81 

2. Experimental

82 

2.1. Chemicals and reagents Diethylstilbestrol, hydrocortisone, prednisolone, progesterone, prednisone, dehydroepiandrosterone, dienestrol, hexestrol and stanozolol were purchased for use as standards from Dr. Ehrenstorfer GmbH (Augsburg, Germany). β-Estradiol, estrone, estriol, androstanolone, testosterone, dexamethasone, pregnenolone, 17a-hydroxyprogesteron, corticosterone, 21-hydroxyprogesterone, 17β-estradiol-17β-D-glucuronide sodium salt, estrone-3-sulfate sodium salt, β-glucuronidase from bovine liver and sulfatase from abalone entrails were purchased from Sigma-Aldrich (St. Louis, MO, USA). Epi-androsteron, 4-androstenedione, androsterone, androstenediol, 5β-androstanedione, methyltestosterone, cortisone and β-glucuronidase/aryl sulfatase (from helix pomatia) were obtained from Merck (Darmstadt, Germany). The purity of the steroids reference standards was more than 99%. Methanol and acetonitrile, which HPLC grade, were purchased from Fisher Scientific (Pittsburgh, PA, USA). Ultrapure water was obtained from a Millipore Milli-Q water purification system (Millipore, Bedford, MA, USA). Formic acid (96%) and ammonia water (25%) were purchased from the Beijing Chemical Reagent Company (Beijing, China). Zinc chloride (ZnCl2), sodium phosphate monobasic dehydrate (NaH2PO4·2H2O) and sodium phosphate dibasic dodecahydrate (Na2HPO4·12H2O) were purchased from Sinopharm Chemical Reagent Co, Ltd (Beijing, China). SPE cartridges of Bond Elut Plexa (60 mg, 3 cc), Bond Elut C18 (200 mg, 3 cc), Bond Elut C8 (200 mg, 3 cc), and Bond Elut PLRP-S on-line SPE cartridge (2.1 x 12.5 mm) was from Agilent (MA, USA). Oasis® HLB cartridge was from Waters (60 mg, 3 cc, MA, USA). Hybrid SPE®-Phospholipid Ultra Cartridge was from Sigma-Aldrich (500 mg, 6 cc, MO, USA). Steroid stock solutions were prepared in methanol at a concentration of 1 mg/mL. The working solutions were prepared in water/methanol (1:1, v/v) at concentrations of 10 μg/mL and 1 μg/mL. 

104 

2.2. Sample extraction and cleanup

105  106  107  108  109  110  111  112  113  114  115  116  117  118  119  120  121 

2 g of a homogenized fresh egg sample was weighed into a polypropylene centrifuge tube (50 mL) and extracted with 6 mL of acetonitrile. After vortexing for 20 s, the sample was further extracted ultrasonically at a power 550 W for 10 mins in an ice bath. After centrifuging at 12,000 rpm and 4°C for 10 mins, the supernatant was transferred, and the extraction procedure was repeated one more time. The supernatant was combined, and 1 g of ZnCl2 was added. The tube was tightly sealed, shaken vigorously and centrifuged for 5 mins at 12,000 rpm and 4°C. The supernatant was transferred and evaporated to approximately 2 mL under a nitrogen stream at 30°C. 1 mL of methanol was added to the samples, which were further diluted to approximately 5 mL with pure water. An additional centrifugation step at 12,000 rpm and 4°C for 10 mins was required to remove the precipitates, and the supernatant was used for cleanup.

Ac ce p

te

d

M

an

us

cr

ip t

83  84  85  86  87  88  89  90  91  92  93  94  95  96  97  98  99  100  101  102  103 

For sample cleanup, a Bond Elut Plexa SPE cartridge was employed. The determination of the optimal cartridge washing and elution condition used an online SPE cartridge with a similar sorbent. The procedure is described below. The online SPE cartridge was assembled standalone on the LC system, and a shallow gradient (from 5% to 95% of organic solvent during 30 min) at a flow rate of 300 μL/min was employed. 10 μL of the steroid neat standards was injected. The optimal condition for washing was the mobile fraction prior to elution of the steroids from the SPE cartridge, and the 3   

Page 3 of 23

elution condition was the mobile fraction at which the steroids can be totally eluted. The cleanup procedure for the Bond Elut Plexa SPE cartridge was performed as follows: the cartridge was pre-conditioned with 3 mL of methanol and 3 mL of ultrapure water. After sample loading, the Plexa cartridge was washed sequentially with 3 mL of water and 3 mL of methanol/water (20:80, v/v) and vacuum-dried for 2 mins. The elution of the analytes was performed with 6 mL of acetonitrile. The eluate was evaporated to dryness and reconstituted in 1 mL of methanol/water (50:50, v/v).

128 

2.3. Enzyme hydrolysis

129  130  131  132  133  134  135 

For the determination of the steroid conjugates, three types of hydrolase for β-glucuronide and aryl sulfate conjugate hydrolyzation were used. The hydrolases included a β-glucuronidase/aryl sulfatase mixture from helix pomatia, β-glucuronidase from bovine liver and sulfatase from abalone entrails. The procedures were performed as follows: 2 g of the homogenized egg sample was added to 4 mL of acetate buffer (0.2 M, pH 5.0) and 100 μL of enzyme. The mixed samples were further incubated for 16 h at 37°C. The samples were extracted by 12 mL and 6 mL of acetonitrile. The detailed extraction and cleanup procedures were the same as those described above.

136 

2.4. Instrumentation

137  138  139  140  141  142  143  144  145  146  147  148  149  150  151  152  153  154  155  156  157  158  159  160  161  162 

LC-MS/MS analysis was carried out on a HPLC system (1200SL Series, Agilent, Palo Alto, CA, USA) coupled with tandem mass spectrometry with an ESI ion source (API 5000, Applied Biosystems, Foster City, CA, USA). The instrument is controlled by the Analyst software (version 1.4.2), and chromatographic separation was achieved with an Extend-C18 column for positive mode and an EclipsePlus-C18 column (3.5 μm, 100 x 2.1 mm, Agilent, USA) for negative mode. Both of the chromatographic separations were performed using a linear gradient, and the injection volume was 10 μL. The optimized LC conditions are shown in Table 1.

te

d

M

an

us

cr

ip t

122  123  124  125  126  127 

Ac ce p

The optimized parameters for the mass spectrometry in ESI+ were as follows: curtain gas, 20 psi; ion spray voltage, 5500 V; collision gas, 5 psi; ion source gas 1 and gas 2 were 70 and 50 psi, respectively; and source temperature, 500°C. The steroids analyzed in ESI+ mode included hydrocortisone, cortisone, corticosterone, prednisolone, prednisone, dexamethasone, stanozolol, progesterone, pregnenolone, 17a-hydroxyprogesteron, 21-hydroxyprogesterone, epi-androsteron, 4-androstenedione, androstenediol, 5β-androstanedione, androsterone, androstanolone, testosterone, methyltestosterone, and dehydroepiandrosterone. The operating MS parameters in ESI- were as follows: curtain gas, 25 psi; voltage, -4500 V; collision gas, 5 psi; ion source gas 1 and gas 2 were both 50 psi; and source temperature, 450°C. The steroids analyzed in ESI- mode included estrone, 17β-estradiol, estriol, diethylstilbestrol, dienestrol and hexestrol. Data acquisition for identification was performed by working in the multiple reaction monitoring (MRM) mode. The product ions were identified in a collision cell (Q2) using nitrogen as the collision gas. Representative product ions were selected to set up a MRM transition. The parameters of the precursor ion, product ions, declustering potential (DP), collision energy (CE), and collision cell exit potential (CXP) of each compound are listed in Table 2. 4   

Page 4 of 23

2.5. Matrix interference

164  165  166  167  168  169  170  171  172 

Electrospray ionization is susceptible to matrix interference. In order to evaluate the effect of sample extraction on removal of matrix interference, a post column infusion system was used as described by Bonfiglio et al. (1999) [25]. Two types of sample extraction were prepared, i.e. samples extracted with or without ZnCl2. The mixed reference standards solution (1 μg/mL) was infused via a Harvard syringe pump (Holliston, MA) at a flow rate of 5 μL/min, the infusion tube was connected to electrospray interface via a Tee connector. The extracted egg samples (10 μL) were injected into the HPLC column. The chromatographic separation follows the same as for real sample analysis. Effluent from the HPLC column was combined with infused analytes and entered the electrospray interface.

173 

2.6. Method validation

174  175  176  177  178  179  180  181  182  183  184  185  186  187 

Calibration curves were prepared with six concentration points for each analyte within the range of 0.1 to 500 ng/mL. Recoveries were assessed by analyzing the spiked egg samples at three concentrations (i.e., 10, 20 and 50 ng/g) and with six replicates at each concentration. The equivalence between the concentrations at which the eggs were spiked and the concentrations in the final extract is 2 (Conc. extract = 2 x Conc. sample). The pregnenolone was spiked at 50, 100 and 250 ng/g due to its low sensitivity in mass ionization. The steroid content in the blank egg matrix was subtracted for recovery calculation. The precision of the method was evaluated using the relative standard deviation (RSD). The limit of detection (LOD) and limit of quantification (LOQ) were defined as the concentration at 3 and 10 times the signal intensity of noise in the spiked egg samples. However, for the steroids with a high endogenous content in the blank egg samples, the neat standard solution was used for determining the LOD and LOQ values. The selectivity and repeatability of the method were assessed by analyzing the egg samples from different species (i.e., hen eggs, quail eggs, duck eggs, pigeon eggs) or from different raising systems (i.e., organic eggs and normal eggs). Five eggs were taken for each type..

188 

3. Results and Discussion

189 

3.1. Sample preparation and cleanup

190  191  192  193  194  195  196  197  198  199  200  201  202 

The eggs contain high amounts of protein and lipids, which interfere with steroid analysis. For the selection of extraction solvents, we compared the efficiency of acetonitrile and methanol. Both solvents resulted in a similar recovery for most of the steroids. However, acetonitrile resulted in cleaner extracts and less background noise, which could be observed because acetonitrile precipitates protein better than does methanol. The precipitation of protein with 3% trichloroacetic acid was also tested, but this method did not improve the cleanup performance. Liquid-liquid extraction (LLE) followed by SPE cartridges was employed for sample cleanup, and the Bond Elut Plexa SPE cartridge was used. Other SPE cartridges (i.e., HLB, C18, C8 and Hybrid) were also tested, but the analytical performance was not significantly improved (Fig. 1). The optimizations of the SPE washing and elution steps are time-consuming and require many cartridges [26]. Here, we used an online SPE cartridge to optimize the washing and elution conditions. The online SPE cartridge is a fairly newly developed product, and is designed for fast sample preparation. The online cartridge can be used either standalone or in tandem with an analytical column, a choice not available with offline

Ac ce p

te

d

M

an

us

cr

ip t

163 

5   

Page 5 of 23

SPE (Phenomenex, Torrance, USA). We assembled the online SPE cartridge standalone on LC system, and run with a shallow gradient to optimize the wash and elution conditions, which was convenient and faster than the conventional method. A significant matrix ion suppression effect was observed in our study, which caused low recovery for the analysis of the spiked samples. The matrix interference was significantly reduced after using LLE and SPE for sample cleanup. However, the high content of phospholipids in the eggs cannot be removed by the above procedures due to their biphasal character, and these phospholipids interfere with steroid analysis. It has been reported that divalent metal ions, such as Zn2+, Pb2+ and Cd2+, can coordinate and precipitate phospholipids from eggs [27], and this method can be used for egg sample cleanup [28]. Another study reported that ZnCl2 can precipitate lipids in eggs and improve the recovery of steroids [24]. To observe the direct effect of ZnCl2 on matrix cleanup, we used the post column infusion system. We observed a significant improvement in the chromatogram (Fig. S1A & Fig. S1B). The ion suppression was significantly reduced in the chromatographic region corresponding to 7-11 min in the ESI+ mode (Fig. S1A) and 5-12 min in ESI- mode (Fig. S1B). This result indicated that the co-eluting matrix components (both polar and nonpolar components) decreased after addition of ZnCl2. The reduction of the ion suppression region corresponds with the significant improvement in the recoveries of several steroids.

220 

3.2. Enzymatic hydrolysis

221  222  223 

231  232  233  234  235  236  237  238  239  240  241 

Many endogenous steroids exist in the conjugated form in the tissue, such as blood and liver. The hydroxylation of egg samples by glucuronidase/aryl sulfatase for steroid analysis has been previously performed [23, 24]. In our study, we evaluated the effect of enzyme hydrolysis using β-glucuronidase/aryl sulfatase from helix pomatia. Most of the steroids did not demonstrate a difference with or without enzyme hydrolysis, except for one steroid (i.e., progesterone). The content of progesterone substantially increased after enzyme hydrolysis (Fig. 2A). Progesterone does not form conjugates due to the lack of hydroxyl group in its chemical structure. Moreover, the content of pregnenolone was significantly reduced after enzyme hydrolysis (Fig. 2B), which was not expected. To investigate this result, we used separate enzymes of β-glucuronidase and aryl sulfatase to hydrolyze the egg samples, and we did not observe the change in progesterone and pregnenolone that occurred with the helix pomatia enzyme (Fig. 2). The enzymatic activities of β-glucuronidase and aryl sulfatase have been confirmed by hydroxylation of authentic reference standards (i.e., 17β-estradiol-17β-D-glucuronide sodium salt and estrone-3-sulfate sodium salt. The result demonstrated that both progesterone and pregnenolone did not exist the conjugated form in eggs. To the best of our knowledge, no scientific reports on the conjugated progesterone or pregnenolone in eggs have been published. The variation after enzyme hydroxylation was due to the transformation of pregnenolone to progesterone. This transformation demonstrated that the enzyme from helix pomatia may have steroid reductase activity. It should be notified that the transformation of pregnenolone to progesterone due to the enzyme hydrolysis has been reported previously [29, 30], and our result further confirmed this finding. Therefore, if steroid conjugation is studied, the types of hydrolysis enzymes need to be carefully considered.

242 

3.3 Method validation

243  244 

The linearity of the 26 steroid standard curves are summarized in Table 3. The determination coefficients (r2) for all of the standard curves were more than 0.99 within the linear range. The

229  230 

cr

us

an

M

d

te

Ac ce p

224  225  226  227  228 

ip t

203  204  205  206  207  208  209  210  211  212  213  214  215  216  217  218  219 

6   

Page 6 of 23

selectivity was studied by analyzing eggs from different individuals and different species. No interfering peaks were observed under the monitored mass transition. Table 4 shows the recoveries and relative standard deviation (RSD) of the steroids spiked at three different concentrations. The recoveries were 63.20%-95.64% at low concentrations (RSD, 0.23%-20.57%), 65.99% -115.30% at medium concentrations (0.25%-13.45%), and 66.14%-121.47% at high concentrations (0.31%-12.62%), respectively. The LODs and LOQs ranged from 0.015 to 7.5 ng/g and 0.05 to 25 ng/g, respectively (Table 3). The performance of the validated method was compared with previous work (Table 6). In general, the methods were comparable. The method described by Wang et al. (2010) has recoveries higher than 100% [24]. The LOQs of steroids analyzed in negative ESI ionization mode were lower [24]. The present method had the lowest LOQs of the steroids analyzed in positive ESI ionization mode compared with other methods (Table 6).

256 

3.4. Analysis of real-samples

257  258  259  260  261  262  263  264  265  266  267  268  269  270  271  272  273  274  275  276  277  278  279 

This method has been successfully applied to detect steroids in eggs bought from local supermarkets in Beijing, China. Ten individual eggs were analyzed for each type of sample. Only natural steroids were found in all of the egg samples. Table 5 shows the determined steroids values. Progesterone, 4-androstenedione, testosterone, pregnenolone, 5β-androstanedione and estrone were detected in all of the eggs even though their amount varies based on the sample types. All of the egg samples exhibit a rather large amount of progesterone (23.69-89.80 ng/g) and pregnenolone (18.74-116.03 ng/g), which are precursors of active steroid hormones that can perform essential biological functions during embryo development. The eggs also contain fairly high amounts of 4-androstenedione (ranged from 3.98 to 8.15 ng/g), which is an androgenic steroid that can be converted to testosterone. All of the egg samples contain trace amounts of biologically active steroids i.e., testosterone (0.18-1.42 ng/g) and estrone (0.12-0.84 ng/g). A small amount of the hydroxyl form of progesterone and pregnenolone as well as dehydroepiandrosterone (DHEA) was found in pigeon eggs but not in other species. Moreover, a very high concentration of 5β-androstanedione (ranged from 60.37 to 299.8 ng/g) was also detected in the eggs, which has not been previously reported. In humans, the transformation of 5β-androstanedione to dihydrotestosterone is a key factor in advanced prostate cancer development [31]. The above results implicate that eggs are a considerable source of steroids intake in human food. This could be expected, since eggs are produced in ovaries, one of the main steroidogenic organs [23]. Also, the steroids are required to maintain the normal physiological development of eggs during hatching. A summary of steroid concentrations in normal hen eggs found in the present and previous study is given in Table 6. In general, the steroids concentrations obtained in the present study were consistent with previous reports. Estrone level was lower than that being reported, this might be due to the variation of individual eggs. Hartmann et al. (1998) reported that only one of the five individuals was found to have high estrone level (0.89 ng/g), and the rest was below 0.4 ng/g [23]. 17β-Estradiol was not detected in our study. To our knowledge, the comparison of steroids concentrations in eggs from different species has not been reported. Our results showed that the steroids levels varied among species, possibly due to the genetic background and dietary components. We also found the ratio of yolk to egg white varies between species, this might affect steroids content when analyzed with whole egg, since more steroids were accumulated in yolk due to their lipophilic character. Raising system can influence the steroids levels, and higher amount of progesterone and 5β-androstanedione were found in organic hen eggs than in normal hen eggs. The Chinese organic hens were commonly outdoor raised. We speculated that external factors such as the raising

280  281  282  283  284  285  286  287 

Ac ce p

te

d

M

an

us

cr

ip t

245  246  247  248  249  250  251  252  253  254  255 

7   

Page 7 of 23

288  289  290  291  292  293  294  295  296  297  298  299  300  301  302  303  304  305  306  307 

environment and diet may affect the maturation of steroidogenic organs, and affect steroids levels in eggs. Our results can provide fundamental data for nutritionists consideration.

308 

4. Conclusions

309  310  311  312  313 

A robust LC-MS/MS method for analyzing 26 types of steroids in eggs was developed. The steroids cover a wide range including androgens, estrogens, corticosteroids, progestogens, and illegal steroids. Most of the steroid hormones can be detected at μg kg-1 levels. The method was applied to analyze steroids in real egg samples, and six types of steroids were found. The steroid content varies among the various egg species.

314 

5. Acknowledgments

315  316  317  318  319  320  321  322  323  324  325  326  327  328 

This work was supported by the National High Technology and Science Development Plan of China (2011AA100302), the Natural Science Foundation of China (31171691, 31371779), the International Science & Technology Cooperation Program of China (2012DFA31140), and the Special Fund for Agro-scientific Research in the Public Interest (201203046).

us

cr

ip t

Principal component analysis (PCA) was used to differentiate eggs from different species. The eigen values of the first three components accounted for 39.35%, 28.34%, and 20.18%, respectively. The first two components contributed to 67.7% of the total variance. PCA plot of component 1 vs. component 2 was given in Fig. 3A. It can be seen that the eggs from different species can be clearly separated based on steroid levels (Fig. 3A). The species of eggs can be divided into four groups by four quadrants based on the positions in the PCA scatter plot. The differences between hen eggs and others eggs were captured largely by the first PCA axis. The normal hen eggs were separated from organic hen eggs primarily along PCA axis 2, and duck eggs were clearly separated from quail eggs and pigeon eggs in the same direction along axis 2. The pigeon eggs were scattered in the plot due to the large variation in the steroid content.

Ac ce p

te

d

M

an

The Euclidean distance measurement and single linkage clustering were performed for hierarchical clustering analysis. The dendrogram is shown in Fig. 3B. The results clearly indicated that the eggs from the five species can be categorized into two major groups consisting of normal hen eggs and organic hen eggs in one group and duck and pigeon eggs in the other group. Normal and organic hen eggs have the shortest distance and can be clearly distinguished from all of the remaining species. Quail eggs did not fall into either of the two groups.

Reference [1] R. Ahmadkhaniha, A. Shafiee, N. Rastkari, F. Kobarfard, Accurate quantification of endogenous androgenic steroids in cattle’s meat by gas chromatography mass spectrometry using a surrogate analyte approach, Anal. Chim. Acta 631 (2009) 80-86. [2] L. Rambaud, F. Monteau, Y. Deceuninck, E. Bichon, F. Andre, B.L. Bizec,  Development and validation of a multi-residue method for the detection of a wide range of hormonal anabolic compounds in hair using gas chromatography–tandem mass spectrometry, Anal. Chim. Acta 586 (2007) 93-104. [3] J. Aufartova, C. Mahugo-Santana, Z. Sosa-Ferrera, J.J. Santana-Rodriguez, L. Novakova, P. Solich, Determination of steroid hormones in biological and environmental samples using green 8   

Page 8 of 23

te

d

M

an

us

cr

ip t

microextraction techniques: An overview, Anal. Chim. Acta 704 (2011) 33-46. [4] Y. Yang, B. Shao, J. Zhang, Y. Wu, H. Duan, Determination of the residues of 50 anabolic hormones in muscle, milk and liver by very-high-pressure liquid chromatography–electrospray ionization tandem mass spectrometry, J. Chromatogr. B 877 (2009) 489-496. [5] B. Shao, R. Zhao, J. Meng, Y. Xue, G. Wu, J. Hu, X. Tu, Simultaneous determination of residual hormonal chemicals in meat, kidney, liver tissues and milk by liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 548 (2005) 41-50. [6] F. Bianchi, M. Mattarozzi, M. Careri, A. Mangia, M. Musci, F. Grasselli, S. Bussolati, G. Basini, An SPME–GC–MS method using an octadecyl silica fibre for the determination of the potential angiogenesis modulators 17β-estradiol and 2-methoxyestradiol in culture media, Anal. Bioanal. Chem. 396 (2010) 2639-2645. [7] M.H. Devier, P. Labadie, A. Togola, H. Budzinski, Simple methodology coupling microwave-assisted extraction to SPE/GC/MS for the analysis of natural steroids in biological tissues: Application to the monitoring of endogenous steroids in marine mussels Mytilus sp., Anal. Chim. Acta 657 (2010) 28-35. [8] H.F. De Brabander, B. Le Bizec, G. Pinel, J.-P. Antignac, K. Verheyden, V. Mortier, D. Courtheyn, H. Noppe, Past, present and future of mass spectrometry in the analysis of residues of banned substances in meat-producing animals, J. Mass Spectrom. 42 (2007) 983-998. [9] K. Verheyden, H. Noppe, J. Vanden Bussche, K. Wille, K. Bekaert, L. De Boever, J. Van Acker, C.R. Janssen, H.F. De Brabander, L. Vanhaecke, Characterisation of steroids in wooden crates of veal calves by accelerated solvent extraction (ASE®) and ultra-high performance liquid chromatography coupled to triple quadrupole mass spectrometry (U-HPLC-QqQ-MS-MS), Anal. Bioanal. Chem. 397 (2010) 345-355. [10] H.J. Cho, J.D. Kim, W.Y. Lee, B.C. Chung, M.H. Choi, Quantitative metabolic profiling of 21 endogenous corticosteroids in urine by liquid chromatography–triple quadrupole-mass spectrometry, Anal. Chim. Acta 632 (2009) 101-108. [11] Y. Yang, B. Shao, J. Zhang, Y. Wu, J. Ying, Analysis of eight free progestogens in eggs by matrix solid-phase dispersion extraction and very high pressure liquid chromatography with tandem mass spectrometry, J. Chromatogr. B 870 (2008) 241-246. [12] A. Passantino, Steroid Hormones in Food Producing Animals: Regulatory Situation in Europe, in: Dr. Carlos C. Perez-Marin (Ed.), A Bird's-Eye View of Veterinary Medicine, In Tech, Rijeka, 2012, pp. 33-50. [13] USDA report National Nutrient Database for Standard Reference Release 26. Available from http://ndb.nal.usda.gov/ndb/foods/show/112?fg=&man=&lfacet=&format=&count=&max=25&o ffset=&sort=&qlookup=egg. [14] E.M. Malone, C.T. Elliott, D.G. Kennedy, L. Regan, Development of a rapid method for the analysis of synthetic growth promoters in bovinemuscle using liquid chromatography tandem mass spectrometry, Anal. Chim. Acta 637 (2009) 112-120. [15] C. Van Poucke, C. Detavernier, R. Van Cauwenberghe, C. Van Peteghem, Determination of anabolic steroids in dietary supplements by liquid chromatography–tandem mass spectrometry, Anal. Chim. Acta 586 (2007) 35-42. [16] M.E. Touber, M.C. van Engelen, C. Georgakopoulus, J.A. van Rhijn, M.W. Nielen, Multi-detection of corticosteroids in sports doping and veterinary control using high-resolution liquid chromatography/time-of-flight mass spectrometry, Anal. Chim. Acta 586 (2007) 137-146.

Ac ce p

329  330  331  332  333  334  335  336  337  338  339  340  341  342  343  344  345  346  347  348  349  350  351  352  353  354  355  356  357  358  359  360  361  362  363  364  365  366  367  368  369  370  371  372 

9   

Page 9 of 23

te

d

M

an

us

cr

ip t

[17] N. Gineys, B. Giroud, E. Vulliet, Analytical method for the determination of trace levels of steroid hormones and corticosteroids in soil, based on PLE/SPE/LC-MS/MS, Anal. Bioanal. Chem. 397 (2010) 2295-2302. [18] G. Dusi, M. Gasparini, M. Curatolo, W. Assini, E. Bozzoni, N. Tognoli, E. Ferretti, Development and validation of a liquid chromatography–tandem mass spectrometry method for the simultaneous determination of nine corticosteroid residues in bovine liver samples, Anal. Chim. Acta 700 (2011) 49-57. [19] K.S. Schmidt, R. Hackenberg, C.S. Stachel, P. Gowik, Endogenous and synthetic steroids in bovine urine – Preparation of in-house reference material, stability studies and results of a proficiency test, Anal. Chim. Acta 717 (2012) 85-91. [20] Y. Deceuninck, E. Bichon, F. Monteau, J.P. Antignac, B.L. Bizec, Determination of MRL regulated corticosteroids in liver from various species using ultra high performance liquid chromatography–tandem mass spectrometry (UHPLC), Anal. Chim. Acta 700 (2011) 137-143. [21] A.V. Eeckhaut, K. Lanckmans, S. Sarre, I. Smolders, Y. Michotte, Validation of bioanalytical LC–MS/MS assays: Evaluation of matrix effects, J. Chromatogr. B 877 (2009) 2198-2207. [22] C. Muller, P. Schafer, M. Stortzel, S. Vogt, W. Weinmann, Ion suppression effects in liquid chromatography–electrospray-ionisation transport-region collision induced dissociation mass spectrometry with different serum extraction methods for systematic toxicological analysis with mass spectra libraries, J. Chromatogr. B 773 (2002) 47-52. [23] S. Hartmann, M. Lacorn, H. Steinhart, Natural occurrence of steroid hormones in food, Food Chem. 62 (1998) 7-20. [24] Q.L. Wang, A.Z. Zhang, X. Pan, L.R. Chen, Simultaneous determination of sex hormones in egg products by ZnCl2 depositing lipid, solid-phase extraction and ultra performance liquid chromatography/electrospray ionization tandem mass spectrometry, Anal. Chim. Acta 678 (2010) 108-116. [25] R. Bonfiglio, R.C. King, T.V. Olah, K. Merkle, The effects of sample preparation methods on the variability of the electrospray ionization response for model drug compounds, Rapid Commun. Mass SP. 13 (1999) 1175-1185. [26] J. O'Mahony, L. Clarke, M. Whelan, R. O'Kennedy, S.J. Lehotay, M. Danaher, The use of ultra-high pressure liquid chromatography with tandem mass spectrometric detection in the analysis of agrochemical residues and mycotoxins in food – Challenges and applications, J. Chromatogr. A 1292 (2013) 83-95. [27] K.B. Yu, Z.L. Yang, L. Zhang, F. Wang, S.F. Weng, J.G. Wu, Influence of divalent metal ions on the aggregation properties of EYPC, Acta Phys.-Chim. Sin. 19 (2003) 747-750. [28] A.Z. Zhang, Q.L. Wang, J. Shen, S.F. Zhang, L.R. Chen, Simultaneous determination of seven sex hormones in fish products using ultra performance Hquid chromatography—tandem mass spectrometry, Chin. J. Chromatogr. 28 (2010) 190-196. [29] H. Noppe, B. Le Bizec, K. Verheyden, H.F. De Brabander, Novel analytical methods for the determination of steroid hormones in edible matrices, Anal. Chim. Acta 611 (2008) 1-16. [30] S. Hartmann, H. Steinhart, Simultaneous determination of anabolic and catabolic steroid hormones in meat by gas chromatography mass spectrometry, Arch. Lebensm. hyg. 48 (1997) 105-117. [31] K. Chang, R. Li, M. Papari-Zareei, L. Watumull, Y.D. Zhao, R.J. Auchus, N. Sharifi, Dihydrotestosterone synthesis bypasses testosterone to drive castration-resistant prostate cancer ,

Ac ce p

373  374  375  376  377  378  379  380  381  382  383  384  385  386  387  388  389  390  391  392  393  394  395  396  397  398  399  400  401  402  403  404  405  406  407  408  409  410  411  412  413  414  415  416 

10   

Page 10 of 23

Proc. Natl. Acad. Sci. USA 108 (2011) 13728-13733.

Ac ce p

te

d

M

an

us

cr

ip t

417 

11   

Page 11 of 23

Figure Captions: Fig. 1. Comparison of different SPE cartridges on the recoveries of steroids in eggs. Steroids were spiked at a concentration of 20ng/g (pregnenolone at 100 ng/g) in egg matrix. Fig. 2. Ion chromatograms of progesterone (2A) and pregnenolone (2B) in egg samples with different

ip t

enzyme hydrolysis. Progesterone level was significantly increased after hydrolysis by

β-glucuronidase/aryl sulfatase from helix pomatia (blue line in Fig. 2A), whereas pregnenolone was

cr

significantly reduced (blue line in Fig. 2B). Incubation with separate enzymes (β-glucuronidase from bovine liver, and aryl sulfatase from abalone entrails) did not cause the changes.

us

Fig. 3. Principal component analysis (3A) and hierarchical clustering analysis (3B) of steroids in eggs

an

from different species and raising systems. Each group contains ten individual egg samples.

Ac ce p

te

d

M

 

12   

Page 12 of 23

331 

Quantitative determination of 26 steroids in eggs from various species using liquid

332 

chromatography–triple quadrupole-mass spectrometry

333  334 

Highlights: A HPLC-MS/MS method for analysis of 26 steroids in eggs was developed.

336 

A shallow gradient procedure using online SPE cartridge facilitates SPE clean-up development.

337 

Six natural steroids were found in the eggs. 5β-Androstanedione was first time reported.

338 

The steroids content in the eggs was species dependent.

cr

us

te

d

M

an

 

Ac ce p

339 

ip t

335 

19   

Page 13 of 23

Ac

ce

pt

ed

M

an

us

cr

i

Figure 1

Page 14 of 23

Ac

ce

pt

ed

M

an

us

cr

i

Figure 2

Page 15 of 23

Ac

ce

pt

ed

M

an

us

cr

i

Figure 3

Page 16 of 23

Table 1 Chromatographic conditions of LC-MS/MS in positive and negative ionization. Positive

Negative

Time (min)

0 50 50 2 25 75 6 5 95 11 5 95 11.1 50 50 15 50 50 A: 0.1% formic acid in water B: 0.1% formic acid in methanol Flow rate: 300 μL/min Column: Eclipse plus-C18

0 50 2 20 6 5 8 5 8.1 50 12 50 A: 0.1% Ammonia in water B: Methanol Flow rate: 200 μL/min Column: Extand-C18

Ac ce p

te

d

M

an

us

cr

ip t

Mobile Phase A (%)

Mobile Phase B (%) 50 80 95 95 50 50

Time (min) Mobile Phase A (%) Mobile Phase B (%)

Page 17 of 23

Table 2 Optimized MS/MS parameters for the analysis of steroids. Precursor ion (m/z)

Product ions (m/z)

DP (V)

CE (V)

CXP (V)

289.3

109.1*, 97.1

121, 121

33, 45

22, 12

Androstenediol Epi-Androsteron Androsterone Prednisolone

126, 151

63, 29

12, 4

273.2

*

105.1 , 93.0

141, 136

47, 35

22, 18

273.2

*

105.1 , 93.0

141, 136

47, 35

22, 18

273.2

*

141, 136

47, 35

22, 18

96, 86

41, 37

20, 16

35, 33

18, 14

47, 37

20, 12

216, 191

63, 55

16, 12

181, 146

37, 55

12, 22

105.1 , 93.0 *

147.3 , 171.2 *

315.4

Prednisone

97.2 , 109.2 *

147.2 , 171.3

359.3

*

Stanozolol

329.4

81.1 , 95.2

Methyltestosterone

303.4

109.2*, 97.1 *

147.3 , 171.3

393.4

287.4

Dehydroepiandrosterone

271.3

166, 166

39, 35

10, 8

151, 146

33, 41

16, 16

*

186, 181

37, 41

12, 16

121.2 , 147.4

363.3

4-Androstenedione

131, 111

*

163.2 , 121.2

361.4

Hydrocortisone

151, 146

an

Dexamethasone Cortisone

*

97.2 , 109.2

146, 61

35, 33

8, 14

*

146, 171

27, 51

14, 10

*

147.3 , 213.2

136, 136

21, 25

14, 22

197.1 , 105.1

289.4

Corticosterone

347.3

121.0*, 97.0

111, 106

37, 43

12, 14

317.3

*

131, 136

35, 45

20, 16

*

97.1 , 109.1

121, 116

37, 37

16, 12

*

d

5β-Androstanedione

te

Pregnenolone 21α-Hydroxyprogesterone 17α-Hydroxyprogesterone

Ac ce p

Diethylstilbestrol

β-Estradiol

Negative

105.1 , 159.2

361.4

Progesterone

Positive

291.3

ip t

Dihydrotestosterone

*

cr

Testosterone

us

Analytes

M

Ionization

331.5

331.5

267.1 271.2

159.2 , 81.0

97.1 , 109.1

121, 116

37, 37

16, 12

*

-180, -160

-36, -40

-21, -23

*

-195, -195

-54, -50

-15, -13

*

237.2 , 222.0 183.3 , 145.1

Estrone

269.1

145.1 , 183.3

-155, -120

-36, -48

-19, -25

Estriol

287.1

171.3*, 145.1

-155, -135

-56, -68

-13, -17

-135, -195

-22, -36

-11, -5

-190, -190

-40, -40

-17, -17

Dienestrol

Hexestrol

265.1 269.1

*

93.1 , 147.2 *

119.1 , 133.1

* The most abundant product ion used for the quantitative analysis. DP: declustering potential; CE: collision energy; CXP: collision cell exit potential.

Page 18 of 23

i cr us

Table 3

Retention time (min) and linearity of steroids. Limits of detection (LODs) and quantification (LOQs) of the method. Experiments were done in triplicates. Intercept (Mean ± SD)

Linear range (ng/mL)

Determination coefficients (r2)

LODs (ng/g)

LOQs (ng/g)

4.98 5.44 5.15 5.85 5.37 6.48 5.76 3.89 4.67 4.60 5.11 5.95 4.17 5.62 4.18 3.75 6.23 6.68 6.02 5.33 6.13 6.01 6.14 6.21 4.70 6.25

2.91x105±7.55x103 7.25x104±1.33x103 1.42x105±1.00x103 5.65x103±1.07x102 1.44x103±3.06x10 9.29x103±6.24x10 5.14x103±5.29x10 6.37x104±3.00x102 4.40x104±3.61x102 2.13x103±2.08x10 2.92x103±3.61x10 8.26x103±1.93x102 2.37x104±2.00x102 1.48x105±1.15x103 2.77x104±2.52x102 1.65x104±2.31x102 2.40x105±3.61x103 2.34x103±6.03x10 1.61x105±1.53x103 1.81x105±2.52x103 1.08x104±1.73x102 1.10x104±1.00x102 2.46x104±1.00x103 1.60x104±1.73x102 6.89x103±9.61x10 3.60x103±7.37x10

1.22x104±5.13x102 5.55x103±5.25x102 9.58x103±2.62x103 4.39x103±1.52x103 6.66x103±2.26x103 5.63x103±2.30x102 -1.33x103±1.73x103 2.60x103±2.33x102 5.16x103±5.39x102 3.43x102±1.26x102 1.55x104±1.86x103 6.70x103±1.64x103 3.09x103±1.30x102 7.03x103±4.97x102 5.96x103±4.65x102 6.36x103±5.86x102 3.07x103±1.72x103 2.95x104±4.63x103 1.24x104±3.79x102 8.79x103±2.88x103 -1.21x102±4.15x102 -3.17x103±1.93x102 -1.11x102±1.49x103 -1.22x102±4.47x102 1.62x103±3.45x103 -3.25x102±1.21x102

0.1-50 0.5-100 0.2-200 5.0-500 20-200 5.0-200 10-200 0.1-100 0.5-100 5.0-200 10-500 5.0-200 0.2-100 0.1-100 0.5-100 1.0-200 0.1-100 50-1000 0.1-100 0.1-100 0.5-200 0.5-500 0.5-200 0.5-200 1.0-200 1.0-200

0.9982 0.9990 0.9988 0.9994 0.9990 0.9994 0.9986 0.9996 0.9992 0.9996 0.9986 0.9990 0.9988 0.9986 0.9990 0.9984 0.9968 0.9994 0.9986 0.9988 0.9996 0.9992 0.9996 0.9998 0.9994 0.9998

0.015 0.075 0.030 0.750 3.000 0.500 1.500 0.015 0.045 0.750 1.500 0.750 0.030 0.015 0.045 0.150 0.015 7.500 0.015 0.015 0.075 0.075 0.045 0.030 0.150 0.150

0.050 0.250 0.100 2.500 10.000 1.500 5.000 0.050 0.150 2.500 5.000 2.500 0.100 0.050 0.150 0.500 0.050 25.000 0.050 0.050 0.250 0.250 0.150 0.100 0.500 0.500

Ac

ed

M

an

Slope (Mean± SD)

ce

4-Androstenedione 17α-Hydroxyprogesterone 21α-Hydroxyprogesterone 5β-Androstanedione Androstenediol Androsterone Epi-androsteron Cortisone Corticosterone Dexamethasone Dehydroepiandrosterone Dihydrotestosterone Hydrocortisone Methyltestosterone Prednisolone Prednisone Progesterone Pregnenolone Stanozolol Testosterone β-Estradiol Diethylstilbestrol Dienestrol Estrone Estriol Hexestrol

Retention time (min)

pt

Analytes

Page 19 of 23

Table 4 Recoveries and relative standard deviations (RSD) of steroids in eggs spiked at three different levels (n=6). Spiked level (10 ng/g)

Spiked level (50 ng/g)

RSD (%)

Recovery (%)

RSD (%)

Recovery (%)

RSD (%)

Testosterone Dihydrotestosterone Epi-Androsteron Androstenediol

87.1 87.7 64.4 83.1

4.8 7.2 14.9 10.7

91.2 93.4 92.3 88.0

3.4 9.8 4.1 8.4

91.1 91.8 69.0 97.6

3.4 4.5 11.2 8.2

Androsterone Prednisolone Progesterone Prednisone Stanozolol

88.3 73.3 80.3 69.7 85.2

11.1 4.3 4.5 9.2 3.5

98.2 87.7 78.9 88.5 94.3

5.9 2.1 4.2 1.9 5.2

93.7 83.8 74.5 85.0 98.8

3.5 3.8 1.9 5.0 3.0

Methyltestosterone Dexamethasone Cortisone Hydrocortisone 4-Androstenedione

88.5 75.1 79.3 63.2 68.2

3.2 6.5 4.7 5.4 3.1

92.6 94.3 89.0 76.2 73.4

4.4 7.2 3.3 6.1 1.8

92.4 90.4 86.0 75.9 74.3

3.3 12.6 3.7 2.7 2.2

5β-Androstanedione Corticosterone Pregnenolonea

95.6 77.6 73.4 87.2

2.2 5.5 3.3 3.4

115.3 86.6 77.5 84.3

4.1 4.2 5.4 8.3

121.5 85.3 90.2 93.0

3.5 3.2 5.9 3.6

83.8 86.7 64.0 87.9 85.1

6.9 6.1 1.4 0.3 0.6

96.2 70.0 66.0 85.6 84.2

2.1 13.5 1.3 0.4 0.8

94.7 92.8 66.1 85.8 81.4

3.9 7.1 6.4 0.5 0.5

94.6 83.6 86.7

0.2 1.6 5.2 100

0.3 1.9 6.1 ng/g,

88.0 0.3 78.8 4.3 83.4 4.0 respectively.

Estriol Dienestrol Hexestrol a Spiked

levels

were

us

M

d

te

Ac ce p

21α-Hydroxyprogesterone 17α-Hydroxyprogesterone Dehydroepiandrosterone Diethylstilbestrol β-Estradiol Estrone

50,

ip t

Recovery (%)

cr

Spiked level (20 ng/g)

an

Analytes

88.7 84.8 85.6 and

250

Page 20 of 23

i cr us an

Table 5

Steroids contents in egg samples (ng/g, n=10). Data were presented as mean ± SD. The values with different superscripts differ at p