Research Article

Simultaneous determination of threenaturally occurring estrogens inenvironmental waters by high-performanceliquid chromatography

A simple, sensitive and accurate reversed-phase high-performance liquid chromato-

graphic (HPLC) method for simultaneous determination of three naturally occurring

estrogenic steroids including estrone (E1), 17b-estradiol (E2) and estriol (E3) in environ-

mental water samples was developed. Analytes were extracted with ethyl acetate solvents

and preconcentrated prior to HPLC analysis. Separations were accomplished in o20 min

using a reversed-phase C18 column (4.6� 250 mm id, 5 mm) with a gradient elution of

mobile phase containing 3.0 mM ammonium acetate/acetonitrile mixtures (flow rate,

1.0 mL/min). UV light absorption responses at 205 nm were linear over a wide concen-

tration range from 100 000 mg/L to the detection limits of 0.96 mg/L E1, 0.64 mg/L E2 and

0.78 mg/L E3. Quantitation was carried out by the peak area method. The relative standard

deviation for the analysis of three estrogens was o3.0%. This method was applied for the

simultaneous determination of estrogens in environmental water samples collected in

Zhejiang, China. The higher concentrations of both E2 and E3 were found in Tang River

and West Lake waters, and E1 was detected in lake water only. All three estrogens were

below the detection limits in rain waters.

Keywords: 17b-Estradiol / Estriol / Estrone / Environmental waters / HPLCDOI 10.1002/jssc.201100445

1 Introduction

Estrogens, as an important group of endocrine-disrupting

compounds (EDCs), have attracted a great deal of scientific

and public attention during the past decades due to their

occurrence in environmental waters and their potential

adverse effects on the development and reproduction of

aquatic wildlife [1–3]. Estrogenic steroids are not only

excreted by human and animals naturally, but also produced

due to their usage in pharmaceutical and livestock-farming

fields [4]. Owing to incomplete removal in wastewater

treatment plants, estrogenic steroids may enter into the

environment and pose a threat to ecosystem and human

health [5, 6]. Many sewage treatment plant effluents in Asia,

Europe, and North America have been reported to contain

potentially estrogenic components [7–10]. Environmental

concentrations of estrogenic steroids have usually been

detected down to the low ng/L levels. The analytical

difficulties associated with the determination of such low

estrogen concentrations in complex aqueous matrices have

limited extensive surveys on the occurrence and abundance

of estrogens in the environment. However, even at such low

concentration levels, estrogenic steroids, which have extre-

mely high physiological activity, still can affect the growth

and reproductive success of fish [11–13]. Various analytical

separation techniques such as gas chromatography (GC)

and liquid chromatography (LC) techniques have been

reported for the separation and quantification of estrogenic

steroid hormones in aquatic environments [14–22]. These

techniques, particularly, GC coupled with flame ionization

(FID) and mass spectrometric (MS) detection have provided

a main horsepower in the analysis of volatile organic

components in aqueous samples because of their high

selectivity and sensitivity. However, GC method is limited

by the volatility or thermal stability of the analytes. There is

usually the need to perform the derivatization of analytes

from sample matrix. Normally, the derivatization process is

a laborious and time-consuming step in the sample

preparation, and it could increase the possibility of

contamination as a consequence of undesirable derivative

reactions [23]. In contrast to the GC method, HPLC

chromatographic approaches enable direct separation and

quantification of many compounds in aqueous samples

without sample pretreatments, and has gained in popularity

Chengjun Wang1

Chunmei Xu1

Fan Chen1

Xuejiao Tang2

1College of Chemistry andMaterials Engineering WenzhouUniversity, Wenzhou, P. R. China

2College of EnvironmentalScience and EngineeringNankai University, Tianjin,P. R. China

Received May 19, 2011Revised June 7, 2011Accepted June 7, 2011

Abbreviations: E2, 17b-estradiol; E3, estriol; E1, estrone

Correspondence: Dr. Chengjun Wang, College of Chemistry andMaterials Engineering Wenzhou University, Wenzhou 325035,P. R. ChinaE-mail: wang.chengjun@yahoo.comFax: 186-577-86689300

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

J. Sep. Sci. 2011, 34, 2371–2375 2371

as an alternative to the GC method. The techniques of

HPLC coupled with different detectors, such as LC-UV, LC-

MS, and LC-MS/MS methods have been widely used for

analyzing the estrogens in environmental samples [24–26].

However, most of the reported HPLC analytical methods for

estrogens focus on the synthetic estrogens such as 17a-

ethynylestradiol (EE2) [27, 28]. This could be due to the

recently increasing contribution of anthropogenic endo-

crine-disrupting chemicals to environmental pollution.

Little information is available in the literature about the

simultaneous determination of three naturally occurring

estrogens in environmental water samples. However, the

naturally occurring steroids estrone (E1), 17b-estradiol (E2),

and estriol (E3) play a predominant role in the estrogenic

activity found in domestic sewage effluents and environ-

mental waters [29]. Therefore, development of a sensitive

and reliable analytical technique is still essential for

studying the occurrence, transportation, biological effects,

and fate of these natural estrogenic steroids.

The aims of this work were to investigate the procedures

for extraction and preconcentration of three naturally

occurring estrogens E1, E2, and E3 in environmental water

samples and to develop the suitable HPLC method for the

simultaneous determination of three estrogenic steroids.

2 Materials and methods

2.1 Chemicals and standards

Standards of E1 and 17b-estradiol were obtained from Tokyo

Chemical Industry (Tokyo, Japan) and E3 was purchased

from Sigma-Aldrich (St. Louis, MO, USA). The chemical

structures of three estrogens are shown in Fig. 1.

Ammonium acetate, ethyl acetate, and hydrochloric acid

were obtained from Jiani Chemistry (Wuxi, China).

Acetonitrile and methanol was supplied by Pharmco

Products (Brookfield, CT, USA). The stock standard

solutions of three estrogens were first prepared at a

concentration of 100 mg/mL in methanol. The working

standard solutions containing three standard compounds

were prepared in methanol at concentrations of 0.0, 5.0, 10,

15, 20, and 25 mg/mL by combining and diluting the

individual stock standard solutions. Except where noted,

all reagents were of analytical grade and all solution

preparations were made using double distilled-deionized

water.

2.2 Sample collection and preparation

Rain water samples were collected with a Teflon container

or film on the grass of the Wenzhou University campus.

River and lake water samples were collected from Tang

River in Wenzhou and West Lake in Hangzhou, respec-

tively. All sites of sample collection are located in Zhejiang

province, China. After sampling, all samples were stored at

41C in dark until used. All samples were centrifuged and

filtered through 0.45 mm membrane filters (Fisher Scien-

tific brand) to remove the impurities in samples before

extraction. Twenty milliliters of water samples were

acidified by adding 0.05 M HCl to maintain pH value at 2

and extracted with 5 mL ethyl acetate thrice. The combined

ethyl acetate extracts (15 mL) were centrifuged and then

decanted into a 20-mL glass vial and then completely dried

under a stream of nitrogen gas. The dried residues of ethyl

acetate extracts of all water samples were dissolved in 200 mL

of methanol and transferred into a 2 mL mini-vial for HPLC

analysis. Precautions were always taken to minimize sample

contamination. All sample containers, glassware and filtra-

tion devices were thoroughly cleaned with 0.1 M HCl

solution and then finally rinsed with double distilled-

deionized water.

2.3 HPLC analysis

An Agilent 1200 high-performance liquid chromatograph

(Agilent Technologies, Palo Alto, CA, USA) equipped with a

quaternary pump, a well plate autosampler, a column oven,

a diode array detector, and ChemStation software was used

for all experiments. The analytical column used was an

Agilent SB-C18 reversed-phase column (4.6� 250 mm id,

5 mm particle size) at column temperature 251C, and

guarded by a 10 mm C18 guard column. The estrogens of

interest were separated using a gradient elution program of

mobile phase consisting of 3.0 mM ammonium acetate

solution and acetonitrile organic modifier, as shown in

Table 1. The flow rate was 1.0 mL/min. Twenty microliters

of samples or standard solutions was injected onto the

HO

O

H

H

H

HO

OH

H

H

H

HO

OH

OH

H

H

H

E1 E2 E3

Figure 1. Chemical structures of three naturally occurring estrogens.

J. Sep. Sci. 2011, 34, 2371–23752372 C. Wang et al.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

HPLC system. Detection of estrogenic steroids was carried

out by direct UV absorbance at 205 nm.

2.4 Identification and quantification

The three naturally occurring estrogens in environmental

water samples were identified by matching retention times

against those of standards and standard addition. All

quantification was performed by the external calibration

method based on peak areas. Calibration curves were

constructed by linear regression of the peak area individual

standard versus the concentration. All the calibration

standards and the rain, river, and lake water samples were

run in triplicate.

3 Results and discussion

3.1 Chromatographic separation and identification

of estrogenic steroids in water samples

Prior to employing HPLC for the determination of estrogens

in rain, river, and lake water extracts, the efficacy of the

separation and detection of three estrogens using the HPLC

technique was tested on a standard mixture. After systema-

tic experiments on the several instrumental parameters,

including the gradient elution program, mobile-phase

composition, column temperature, and flow rate, the

separation conditions were optimized and described in

Section 2. Figure 2A shows a representative separation of a

standard mixture of three estrogens. A baseline separation

of the E1, E2, and E3 was achieved within 20 min using the

method described. Three estrogens were eluted generally in

order of decreasing hydrophilic OH functional groups, i.e.

E3, E2, and E1, on the reverse-phase C18 column. Figure 2B

and C illustrate typical chromatograms of water sample

extracts without and with the addition of the standard

mixtures. These chromatograms revealed that no peak in

the sample extracts overlapped the analyte peaks. Three

estrogen components were identified by (i) matching

retention times against those of standards, (ii) spiking the

sample extracts with authentic standards of each analyte.

The mean values of retention times for three estrogens

determined in standard mixtures were: E1, 15.2170.16 min;

E2, 13.2370.13 min; E3, 4.8570.13 min, respectively. The

relative standard deviation (RSD) values of the retention

times and peak areas were generally o0.5%, indicating that

the separation method developed was very stable and had

high reproducibility.

3.2 Quantitative analysis

Standards mixtures of the three estrogen compounds in the

concentration range of 0.00�25.0 mg/mL were prepared for

calibration curves. Calibration curves were y 5 122x�15.0

for E1, y 5 146x�29.0 for E2, and y 5 153x�38.1 for E3 (y is

the ratio of peak area of standards; x is the concentration of

standards). All of them were linear over the concentration

ranges tested with correlation coefficients 40.999. The

detection limits measured as thrice the background noise

were 0.96 mg/L E1, 0.64 mg/L E2, and 0.78 mg/L E3.

Increasing the injection volume of sample can further

lower the detection limits.

Table 1. Gradient elution program for HPLC analysis of estro-

gens

Time (min) 3 mM ammonium acetatea) (%) Acetonitrile (%)

0 60 40

5 60 40

15 25 75

17 25 75

20 60 40

a) Prepared in double distilled-deionized water.

E3

E3

E3

E2

E2

E2

E1

E1

E1

-402080

140200260320380440 A

B

C

Retention time (min)

mA

U

-25

15

55

95

135

175

Retention time (min)

mA

U

-200

20406080

100120140160180

0 2 4 6 8 10 12 14 16 18

0 2 4 6 8 10 12 14 16 18

0 2 4 6 8 10 12 14 16 18Retention time (min)

mA

U

Figure 2. HPLC chromatograms of (A) E1, E2, and E3 calibrationstandards (25 mg/mL); (B) West lake water extracts; (C) West lakewater extracts spiked with 15 mg/mL standard mixtures. Injectionvolume: 20 mL.

J. Sep. Sci. 2011, 34, 2371–2375 Liquid Chromatography 2373

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

3.3 Determination of E1, E2, and E3 estrogenic

steroids in environmental waters

The described method was tested in several environmental

water matrices with known amounts of three estrogens

added, and the average percentage recovery was found to be

82–95% for all analytes. Figure 2B and C illustrates the

typical elution profile of lake water samples and the spiked

sample with 15 mg/mL standard mixtures. In the chromato-

gram of water sample, besides estrogens, there are some

other peaks that might be due to low-molecular-weight

organic acids but have not been identified. This developed

HPLC method combined with ethyl acetate liquid–liquid

extraction and preconcentration has been successfully

applied for the simultaneous determination of E1, E2, and

E3 in environmental water samples, which were collected in

Zhejiang, China. The concentrations of E1, E2, and E3

estrogens in rain, river, and lake water samples are given in

Table 2. Among all the samples analyzed, the higher

concentrations of both E2 and E3 were found in Tang River

and West Lake waters, and E1 was detected in lake water

only. All three estrogens were below the detection limits in

rain waters. The RSDs for the analysis of three estrogens in

samples were o3% in the general concentration ranges of

about 1–3 mg/L that were found in the environmental water

samples studied.

4 Concluding remarks

The described ethyl acetate liquid–liquid extraction and

preconcentration combined with HPLC method has been

proved a simple, sensitive, and accurate technique for the

simultaneous separation and determination of three natu-

rally occurring estrogens E1, E2, and E3 in environmental

water samples. It can also be applied for the determination

of estrogens in human urines for the medical health studies.

The project is jointly sponsored by the Initial Research Fundof Wenzhou University and Scientific Research Foundation for

the Returned Overseas Chinese Scholars, State EducationMinistry.

The authors have declared no conflict of interest.

5 References

[1] Letcher, R. J., Bustnes, J. O., Dietz, R., Jenssen, B. M.,Jørgensen, E. H., Sonne, C., Verreault, J., Vijayan, M.M., Gabrielsen, G. W., Sci. Total Environ. 2010, 408,2995–3043.

[2] Jobling, S., Nolan, M., Tyler, C. R., Environ. Sci.Technol. 1998, 32, 2498–2506.

[3] Kavlock, R., Environmental Governance and AnalyticalTechniques: Environmental Issues Related to EDCPollution In East Asia, Symposium at United NationsUniversity Headquarters, Tokyo, 1999.

[4] Markiewicz, L., Gurpide, E., Human Endometrium 1994,734, 285–297.

[5] Williams, R. J., Johnson, A. C., Smith, J. J. L., Environ.Sci. Technol. 2003, 37, 1744–1750.

[6] Ternes, T. A., Kreckel, P., Mueller, J., Sci. Total Environ.1999, 225, 91–99.

[7] Harries, J. E., Sheahan, D. A., Jobling, S., Environ.Toxic. Chem. 1996, 15, 1993–2002.

[8] Komori, K., Tanaka, H., Okayasu, Y., Wat. Sci. Technol.2004, 50, 93–100.

[9] Stuer-Lauridsen, F., Kjølholt, J., Søren, L., Survey ofEstrogenic Activity in the Danish Aquatic Environment,In Danish Ministry of the Environment, EnvironmentalProtection Agency, 2005.

[10] Lei, B., Huang, S., Zhou, Y., Wang, D., Wang, Z.,Chemosphere 2009, 76, 36–42.

[11] Lai, K. M., Scrimshaw, M. D., Lester, J. N., Critic. Rev.Toxic. 2002, 32, 113–132.

[12] Lascombe, I., Beffa, D., Ruegg, U., Environ. HealthPersp. 2000, 108, 621–629.

[13] Li, W., Seifert, M., Xu, Y., Environ. Int. 2004, 30, 329–335.

[14] Basheer, C., Jayaraman, A., Kee, M. K., J. Chromatogr.A 2005, 1100, 37–143.

Table 2. The concentrations of E1, E2, E3 found in environmental water samples (mg/L)

Analyte samples E1 E2 E3

Concentration RSD% Concentration RSD% Concentration RSD%

Tang River I NDa) – 5.32 1.33 2.11 2.31

Tang River II ND – 4.98 0.85 3.24 1.56

Tang River III ND – 6.27 0.96 2.56 1.72

Rain I ND – ND – ND –

Rain II ND – ND – ND –

Rain III ND – ND – ND –

West Lake I 1.82 1.23 1.67 2.14 2.12 1.53

West Lake II 1.98 0.96 2.35 1.65 1.87 1.96

West Lake III 1.22 2.65 3.58 0.56 2.38 1.11

a) ND: not detected

J. Sep. Sci. 2011, 34, 2371–23752374 C. Wang et al.

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com

[15] Carpinteiro, J., Quintana, J. B., Rodriguez, I., J. Chro-matogr. A 2004, 1056, 179–185.

[16] Fine, D. D., Breidenbach, G. P., Price, T. L., J. Chroma-togr. A 2003, 1017, 167–185.

[17] Gibson, R., Tyler, C. R., Hill, E. M., J. Chromatogr. A2005, 1066, 33–40.

[18] Gomez, M. J., Mezcua, M., Martinez, M. J., Int. J.Environ. Anal. Chem. 2006, 86, 3–13.

[19] Jeannot, R., Sabik, H., Sauvard, E., J. Chromatogr. A2002, 974, 143–159.

[20] Isobe, T., Shiraishi, H., Yasuda, M., J. Chromatogr. A2003, 984, 195–202.

[21] Zuo, Y., Zhang, K., Lin, Y., J. Chromatogr. A 2007, 1148,211–218.

[22] Zhang, K., Zuo, Y., Anal. Chim. Acta 2005, 554, 190–196.

[23] Zuo, Y., Lin, Y., Chemosphere 2007, 69, 1175–1176.

[24] Beck, I. C., Bruhn, R., Gandrass, J., J. Chromatogr. A2005, 1090, 98–106.

[25] Gentili, A., Perret, D., Marchese, S., Chromatographia2002, 56, 25–32.

[26] Gomes, R. L., Birkett, J. W., Scrimshaw, M. D., Int. J.Environ. Anal. Chem. 2005, 85, 1–14.

[27] Majima, K., Fukui, T., Yuan, J. L., Anal. Sci. 2002, 18,869–874.

[28] Zuo, Y., Zhang, K., Deng, Y. W., Chemosphere 2006, 63,1583–1590.

[29] Farre, M., Perez, S., Kantiani, L., Barcelo, D., TrendsAnal. Chem. 2008, 27, 991–1007.

J. Sep. Sci. 2011, 34, 2371–2375 Liquid Chromatography 2375

& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.jss-journal.com