a rapid two step chromatographic method for the quantitative determination of vitellogenin in fish plasma

Abstract By combining anion-exchange membrane purifi-cation with high-performance size-exclusion chromatogra-phy HPSEC analysis, a two-step chromatographic method was developed for the d

Abstract By combining anion-exchange membrane

purifi-cation with high-performance size-exclusion

chromatogra-phy (HPSEC) analysis, a two-step chromatographic method

was developed for the determination of vitellogenin (Vtg)

in fish plasma Most plasma protein interferences can be

removed during anion-exchange membrane purification

process Vtg is eluted from the size-exclusion

chromatog-raphy column with a retention time of about 9 min and is

characterized based on the native molecular weight, with a

limit of quantification of 20µg Vtg mL–1 plasma The spiked

recovery and interassay variability were better than 80%

and 4.8% This method was successfully applied to

ana-lyze the plasma Vtg levels of loach (Misgurnus

angailli-caudatus) and sea catfish (Enchelyopus elongatus) In

ad-dition to all the female fish, Vtg is detected in 75% of

male loaches and 100% of male sea catfish The result

in-dicates that some chemicals or unknown factors with

es-trogenic activity have induced male fish to produce Vtg

Keywords Vtg · HPSEC · Membrane purification ·

Loach · Sea catfish

Introduction

The presence of endocrine disrupting chemicals (EDCs)

in the environment has recently become a major issue of

concern from the perspectives of both human and

ecosys-tem integrity [1, 2, 3] Most of those EDCs have the

abil-ity to induce responses similar to those caused by estrogens

Vitellogenin (Vtg), a sexual protein inducted by estrogens,

has become a popular biomarker for estrogenic effects of

exposure to estrogen or estrogen mimics [4, 5, 6] There

are a number of reports on the appearance of Vtg in male

fish plasma [7, 8, 9, 10] In order to assess the potential estrogenicity of chemical substances and their effects on fish, it is important to accurately measure the plasma Vtg levels

A number of methods have been developed for mea-suring Vtg levels in fish plasma [11, 12, 13, 14, 15, 16, 17, 18] Enzyme-linked immunosorbent assay (ELISA) and radioimmunoassay (RIA) are the most commonly used ap-proaches among them [12, 13, 14] These antibody-based methods have considerable merits of sensitivity and selec-tivity, but they suffer from problems related to antibody specificity while facing multiple fish species, because an antibody made against Vtg from one species is limited in its application as a probe for another [5, 19] Although cross-reaction was also found in some species, it is not possible to make an accurate quantification of low levels

of Vtg from other species [20] because of the affinity dif-ference Moreover, Vtg is very susceptible to proteolytic degradation [5], so it is important to measure Vtg as soon

as possible However, as the immunoassay method nor-mally need several hours or days, the degradation of Vtg

is inevitable during this period, even with the present of antiproteolytic agents The degradation of Vtg may occur not only in the period of determination, but also during sample collection or storage As the sample degrades, some epitopes recognized by the antibody are destroyed and others normally buried inside the Vtg molecule are ex-posed, thus changing the overall reactivity of the antibody towards Vtg [5] For these reasons, it is desirable to de-velop a rapid and accurate method for Vtg determination High-performance liquid chromatography (HPLC) has been widely applied in the determination of proteins or other biological molecules [21, 22, 23] Recently, a rapid method for the detection fish plasma Vtg using high-performance anion-exchange chromatography column (POROS-HQ) was reported [18] The lowest detectable amount of Vtg was 2µg per assay Although Vtg was eluted as a single peak from the column, it should be further examined by molecular weight High-performance size-exclusion chro-matography (HPSEC) is a simple and widely used tech-nique for protein assay [24, 25, 26] Because the

separa-Jing Shao · Guoqing Shi · separa-Jingfu Liu · Guibin Jiang

A rapid two-step chromatographic method

for the quantitative determination of vitellogenin in fish plasma

Anal Bioanal Chem (2004) 378 : 615–620

DOI 10.1007/s00216-003-2235-0

Received: 2 June 2003 / Revised: 8 August 2003 / Accepted: 18 August 2003 / Published online: 16 October 2003

O R I G I N A L PA P E R

J Shao · G Shi · J Liu · G Jiang (✉)

Key Laboratory of Environmental Chemistry and Ecotoxicology,

Research Center for Eco-Environmental Sciences,

Chinese Academy of Sciences,

P.O Box 2871, 100085 Beijing, China

e-mail: gbjiang@mail.rcees.ac.cn

© Springer-Verlag 2003

tion is based on the molecular weight, its potential for the

rapid and non-destructive analysis of Vtg with both

quan-titative and qualitative was of interest Waagboe and

Sandnes successfully used HPSEC for the determination

of Vtg in the plasma of estrogen-treated rainbow trout

The Vtg determined by this method was significantly

cor-related with the alkali-labile protein phosphorus assay in

naturally maturing fish [25] However, owing to the

inter-ference of plasma proteins, HPSEC, for the time being,

cannot separate and detect trace levels Vtg in fish plasma

In our previous work, we found that Vtg could be

rapidly separated from fish plasma by step-gradient

elu-tion with anion-exchange membrane chromatography [27]

The Vtg fraction by this method shows almost a single

peak when analyzed with HPSEC This indicates that the

anion-exchange membrane can be used to extract Vtg from

fish plasma before further analysis In the present study,

we developed a two-step method for detecting Vtg with a

commercially available anion-exchange membrane and

HPSEC column The method was applied to analyze the

plasma Vtg levels of loach (Misgurnus angaillicaud atus)

and sea catfish (Enchelyopus elongatus) collected from a

fish farm or estuary of the Haihe River during February to

April 2003

Experimental

Chemicals and instruments

17 β -Estradiol (E2), thyroglobulin (640 kDa), ferritin (440 kDa),

im-munoglobulin (158 kDa), bovine serum albumin (BSA, 66 kDa), and

heparin were obtained from Sigma (St Louis, MO, USA); aprotinin

was from Boehringer (Mannheim, Germany) All other chemicals

were reagent-grade compounds obtained from commercial sources.

Buffers and sample solutions were filtered through a 0.2- µ m

cellu-lose acetate filter.

The ready-to-use unit Sartobind MA Q15 (Sartorius,

Goettin-gen, Germany) is a strongly basic anion exchanger with quaternary

ammonium groups The membrane material of this unit is

sup-ported cross-linked regenerated cellulose with an effective

adsorp-tion area of 15 cm 2 The HPLC system for protein analysis consists

of Isopump (Agilent, G1310A, Germany), VWD detector (Agilent,

G1314A, Japan), and high-performance size-exclusion

chroma-tography column: TSK G3000SWXLcolumn (30 cm × 7.8-mm ID,

TOSOH Biosep, Japan) with HPLC workstation (Agilent HP1100

system) An Allitech 426 HPLC Pump (Allitech, USA), which can

achieve a higher flow rate, was used for sample cleaning.

Fish

Adult loaches for method developing were cultured in the

labo-ratory with an average weight of 15 g Another batch of loaches

(20 males and 8 females) with body weights of 10–25 g were

caught from a fish farm where the water was polluted by industrial

effluent Sea catfish (5 males and 7 females) with body weights of

100–200 g were captured from the estuary of the Haihe River.

Plasma preparation

The fish was anesthetized with quinaldine sulfate (40 mg L –1 ).

Blood was then collected from the caudal vein with heparinized

syringes, and transferred to 1.5-mL centrifuge tubes containing

1 aprotinin (2.5 TIU) and 6 heparin (30 USP units) Then, the blood

was centrifuged at 3,000 g, 4°C for 20 min, and plasma was col-lected and stored at –80°C until analysis.

Preparation of vitellogenin standard curve Vitellogenin in plasma from E2-treated fish was purified and vali-dated by SDS-PAGE according to the described method [27] The protein concentration was determined by following the method of Bradford at 595 nm [28] The Vtg standard curve was established with concentrations ranging from 0 to 400 µ g mL –1

Sample purification

A commercially available ready-to-use anion-exchange membrane unit, Sartobind MA Q15, was connected to the high-flow-rate HPLC pump and the procedure for sample purification was as fol-lows: 1) initialize the anion-exchange membrane with 5 mL of buffer A (20 mM phosphate buffer, pH 6.5, containing 0.32 mol L –1 NaCl); 2) dilute 100 µ L of fish plasma with 3 mL of buffer A and pass through 0.2- µ m filter, then load the solution on the anion-ex-change membrane; 3) wash the membrane with 10 mL of buffer A; 4) elute Vtg with 3 mL of buffer B (20 mM phosphate buffer, pH 6.5, containing 0.42 mol L –1 NaCl) and immediately analyze the frac-tions by HPSEC; 5) regenerate the membrane with 5 mL of buffer

C (20 mM phosphate buffer, pH 6.5, containing 1 mol L –1 NaCl) The flow rate of the mobile phase was 10 mL min –1

High-performance size-exclusion chromatography analysis The mobile phase for high-performance size-exclusion chromatog-raphy analysis is 50 mM sodium phosphate buffer, pH 6.5, contain-ing 300 mM NaCl After injectcontain-ing 20 µ L of sample, the separation was carried out at a flow rate of 0.7 mL min –1 at room temperature Absorbance at 210 nm was monitored and the concentration of Vtg was calculated based on the standard curve.

Results and discussion

Selection of wavelength

In order to find the optimum wavelength for Vtg, we mea-sured the serum protein at three wavelengths: 210 nm,

254 nm, and 280 nm The results show that the absorbance

of Vtg at 210 nm is much higher than that at 254 nm and

280 nm (Fig 1) It is well known that the absorption of proteins at 254 nm and 280 nm is mainly due to phenyl chromophores, while absorption at 210 nm is mainly due

to carboxyl groups It can be expected that the Vtg contains more carboxyl groups than phenyl groups Experiments demonstrated that the detection limit for Vtg at 210 nm was much lower than that at 280 nm Therefore, in this study

we choose 210 nm as the detective wavelength

Analysis of Vtg HPSEC has the advantage of accomplishing quantitative and qualitative analysis simultaneously However, Vtg at low concentration levels in fish plasma cannot be detected

by HPSEC because of the interference from non-Vtg pro-teins This drawback has been partly overcome by adding

an anion-exchange membrane purification procedure be-fore HPSEC analysis Figure 2 compares chromatograms 616

obtained before and after membrane purification for the plasma from a female loach Figure 2a illustrates the HPSEC analysis of plasma without being treated by anion-ex-change membrane Although a small Vtg peak could be observed, it cannot be quantification because the peak is severely overlapped with other plasma proteins After mem-brane purification, most of the plasma proteins are elimi-nated (Fig 2b), and a major Vtg peak with a retention time of 8.9 min appeared as shown in Fig 2c The relative molecular mass of Vtg was calculated to be 419 kDa with thyroglobulin, ferritin, IgG, and BSA as molecular mass standard The result is in accordance with those obtained

by size-exclusion chromatography from other fish

spe-cies, for example, 440 kDa for Salmo gairdneri [25], 442,

435, and 424 kDa for Oncorhynchus mykiss, Gobio gobio, and Leuciscus cephalus, respectively [29].

The anion-exchange membrane used in this study is a commercially available cartridge with three pieces of mem-brane inside; the total absorption area is 15 cm2 and the absorption capacity for Vtg is 12 mg The principle for pro-tein separation on the anion-exchange membrane is similar

to that on a conventional anion-exchange column How-ever, because of the more efficient mass transfer charac-teristics of membrane adsorbers (MAs) relative to particle adsorbers that are used in conventional LC column, the MAs can be operated at higher flow rates without imped-ing the bindimped-ing capacity and separation efficiency [30] In this study, Vtg can be separated from most other plasma proteins within 3 min at a flow rate of 10 mL min–1

617

Fig 1a–c Chromatograms of serum protein detected at different

wavelengths: a 210 nm, b 254 nm, c 280 nm Arrows indicate the

Vtg position

Fig 2a–d Comparison of chromatograms of female loach plasma

obtained before and after anion-exchange membrane purification Mobile phase: 50 mM PBS, pH 6.5, containing 300 mM NaCl; 20- µL injection volume a serum diluted with buffer A before membrane cleaning; b serum diluted with buffer A and passed through the membrane; c the Vtg fraction eluted with buffer B;

d the fraction eluted with buffer C

According to the manufacturer’s information, the highest

flow rate of mobile phase that the Sartobind MA Q15 can

tolerate is 50 mL min–1 By connecting a SartobindMA Q15

unit to a syringe one can also carry out the purification

pro-cedure In our experiment, we found that the recovery of

Vtg was still higher than 60% when we push the

corre-sponding sample and buffer through the membrane gently

(the flow rate was about 30 mL min–1)

We also found that although the concentration of Vtg

may vary greatly, the elution volume of Vtg fraction

basi-cally did not change Thus, under the protein capacity of

the SartobindMA Q15 unit, it is expectable that one may

load much more fish serum onto the membrane, which is

helpful in increasing the Vtg concentration in the eluted

fraction Because the maximum serum we can get from a

loach is about 200–400µL, we take 100µL fish serum per

assay for Vtg analysis

The limit of quantification (LOQ) of Vtg

Figure 3a is the HPSEC chromatogram of blank plasma

after membrane purification The result indicates that there

are still little interferential proteins that cannot be

elimi-nated In order to examine the limit of quantification (LOQ)

for real samples, various amounts of Vtg were added to

the same elution fraction; the lowest amount of Vtg that can

give a quantitative peak in the vicinity of 9 min is

empiri-cally identified as the LOQ for this method (see Fig 3b)

For loach, the LOQ is 20µg Vtg mL–1of plasma

Reproducibility and recovery

Pooled plasma from female loach was analyzed five times

to test the reproducibility of this method When 100µL plasma was analyzed each time according to the above-described procedure, the mean retention time of Vtg was

8.88 min (RSD=0.6%, n=5) and the mean Vtg peak height was 14.9 (RSD=4.8%, n=5) at a concentration level of

1,532.4µg mL–1 vitellogenin in the plasma

In order to evaluate the assay recovery, different amounts

of purified Vtg (10µg and 30µg) were added to 100µL pooled female serum, and the assay was then performed The recovery of the added Vtg was found at 84.6% and 90%, respectively

Measurement of plasma Vtg levels

in loach and sea catfish

A total of 28 loaches (20 males and 8 females) from a fish farm where the water was polluted by industrial effluent, and 12 sea catfishes (5 males and 7 females) captured from the estuary of the Tianjin Haihe River were analyzed us-ing this method The results are shown in Fig 4 For loach, Vtg was detected in the plasma of 15 male loaches with concentrations of 73–243µg mL–1; 170–3,829µg Vtg mL–1 was also detectable in all of eight female loaches The high Vtg level and high detection ratio (75%, 15 out of 20) in male loach indicates noteworthy EDC pollution in this fish farm The loach usually live at the bottom of wa-ter and sometimes in the sediment, so the route of expo-sure is also of interest

618

Fig 3a,b Typical chromatograms of Vtg obtained by the proposed

method: a blank plasma; b blank plasma spiked with 4µ g of Vtg

Fig 4a,b The Vtg levels of loach (a) and sea catfish (b), the

num-ber on the graph indicates the detection ratio of Vtg

For sea catfish, the Vtg was first identified based on a

retention time of 8.7 min, which corresponds to the

mo-lecular weight of 474 kDa The chromatograms for a female

and a male sea catfish are shown in Fig 5 Plasma Vtg

levels for the sea catfish were in the range 20–320µg mL–1

(166±128µg mL–1, mean±SD) in male and 1,370–5,570µg

mL–1(1,377±1,218µg mL–1, mean±SD) in female catfish

All of those fish have detectable levels Because sea

cat-fish are seldom move too far away from where they usually

live [31], the high Vtg level in male sea catfish plasma

may be induced by exposure to the water from the Tianjin

Haihe River, which is known to be seriously contaminated

However, further investigations are required to identify

which chemicals or factors contribute more to the

estro-genic effect for these fish

Conclusion

A two-step method based on the combination of anion-ex-change membrane purification and high-performance size-exclusion chromatography analysis was developed to de-termine Vtg in fish plasma This method can be used to determine Vtg levels greater than 20µg Vtg mL–1plasma, which is adequate to detect moderate to strong estrogenic effects The identification of Vtg was based on the reten-tion time in the range 8.36–9.36 min on a size-exclusion chromatography column; this retention time corresponds

to 300–600 kDa in native molecular weight, which is gen-erally agreed to be a characteristic of most teleost Vtg [5, 26] As HPLC is a widely used instrument, this study pro-vides an alternative or reference method that is techni-cally easy to carry out for the determination of Vtg in fish plasma

Acknowledgements This work was jointly supported by the

Na-tional Natural Science Foundation of China (20137010), the State High Tech Development Plan (2001AA640610), and the Chinese Academy of Sciences (KZCX2–414).

References

1 Colborn T, Saal FSV, Soto AM (1993) Environ Health Per-spect 101:378–384

2 Colborn T, Dumanoski D, Myers JP (1996) Our stolen future Dutton, New York

3 Sohoni P, Tyler CR, Hurd K, Caunter J, Hetheridge M, Williams

T, Woods C, Evans M, Toy R, Gargas M, Sumpter JP (2001) Environ Sci Technol 35:2917–2925

4 Heppel SA, Denslow ND, Folmar LC, Sullivan CV (1995) En-viron Health Perspect 103:9–15

5 Denslow ND, Chow MC, Kroll KJ, Green L (1999) Ecotoxi-cology 8:385–398

6 Folmar LC, Denslow ND, Rao V, Chow MC, Grain DA, En-blom J, Marcino J, Guillette LJ Jr (1996) Environ Health Per-spect 104:1096–1101

7 Folmar LC, Denslow ND, Kroll K, Orlando EF, Enblom J, Marcino J, Metcalfe C, Guillette LJ Jr (2001) Arch Environ Contam Toxicol 40:392–398

8 Sheahan DA, Bucke D, Matthiessen P, Sumpter JP, Kirby MF, Neall P, Waldock M (1994) In: Muller R, Lloyd R (eds) Fish-ing news books: sublethal and chronic effects of pollutants on freshwater fish London, pp 99–112

9 Environment Agency for England and Wales (1998) Consulta-tive report environmental issues series

10 Allen Y, Scott AP, Matthiessen P, Haworth S, Thain JE, Feist

S (1999) Environ Sci Technol 18:1791–1800

11 Kramer VJ, Miles-Richardson S, Pierens SL, Giesy JP (1998) Aquat Toxicol 40:335–360

12 Edmunds JS, Mccarthy RA, Ramsdell JS (2000) Environ Health Perspect 108:219–224

13 Diamond SA, Oris J (1995) Environ Toxicol Chem 14:1387– 1388

14 Solé M, Porte C, Barceló D (2001) Trends Anal Chem 20: 518–525

15 Soh N, Nishiyama H, Mishima K, Imato T, Masadome T, Asano Y, Kurokawa Y, Tabei H, Okutani S (2002) Talanta 58:1123–1130

16 Johnsen HK, Tveiten H, Willassen NP, Arnesen AM (1999) Comp Biochem Physiol B 124:355–362

17 Fukada H, Haga Ayumu, Fujita T, Hiramatsu N, Sullivan CV, Hara Akihiko (2001) Comp Biochem Physiol A 130:163–170

619

Fig 5a,b Chromatograms of sea catfish (a) female and (b) male

18 Yamanaka S, Arizono K, Matsuda Y, Soyano K, Urushitani H,

Iguchi T, Sakakibara R (1998) Biosci Biotechnol Biochem

62:1196–1200

19 Jones PD, De Coen WM, Tremblay L, Giesy JP (2000) Wat Sci

Technol 42:1–14

20 Tyler CR, Van der Eerden B, Jobling S, Panter G, Sumpter JP

(1996) J Comp Physiol B 166:418–426

21 Leonil J, Gagnaire V, Molle D, Pezennec S, Bouhallab S

(2000) J Chromatogr A 881:1–21

22 Iwata S (2001) Clin Chim Acta 303:95–104

23 Chollet DF, Gounaz L, Juliano C, Anderegg G (2000) J

Chro-matogr B 746:311–314

24 Hayakawa K, Masuko M, Mineta M, Yoshikawa K, Yamauchi

K, Hirano M, Katsumata N, Tanaka T (1997) J Chromatogr B

696:19–23

25 Waagboe R, Sandnes K (1988) J Chromatogr 427:138–143

26 Liang H, Scott MK, Murry DJ, Sowinski KM (2001) J Chro-matogr 754:141–151

27 Shi GQ, Shao J, Jiang GB, Wang QX, Lu YQ, Liu JF, Liu JM (2003) J Chromatogr B 785:361–368

28 Bradford MM (1976) J Anal Biochem 72:248–254

29 Brion F, Rogerieux F, Noury P, Migeon B, Flammarion P, Thybaud E, Porcher JM (2000) J Chromatogr B 737:3

30 Reif OW, Freitag R (1993) J Chromatogr A 654:29

31 Wang SA, Wang ZM, Li GL, Cao YP et al (eds) The fauna of Hebei, China, Pisces (2001) Hebei Science and Technology Publishing House (in Chinese)

620