Mercuric mercaptide of penicillenic acid, a novel hapten for relevant immunoassay, synthesized from penicillin

Journal of Immunological Methods 353 (2010) 1–7

Contents lists available at ScienceDirect

Journal of Immunological Methods j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m

Research paper

Mercuric mercaptide of penicillenic acid, a novel hapten for relevant immunoassay, synthesized from penicillin Peng Xie a, Xi Tao b, Wu Xu a, Liu-Yin Fan a, Wei Zhang a, Yue-E Zhi b, Pei Zhou b,⁎, Cheng-Xi Cao a,⁎ a Laboratory of Analytical Biochemistry & Bio-separation, Key Laboratory of Microbiology of Educational Ministry, School of Life Science and Biotechnology, Shanghai Jiao Tong University, 800 Dongchuan Rd., Shanghai 200240, China b School of Agriculture and Biology, Shanghai Jiao Tong University, Shanghai 200240, China

a r t i c l e

i n f o

Article history: Received 1 May 2009 Received in revised form 28 October 2009 Accepted 9 December 2009 Available online 24 December 2009 Keywords: Hapten Immunoassay Mercuric ion Penicillin Environment ELISA

a b s t r a c t The synthesis of mercuric mercaptide of penicillenic acid (MMPA) has been the basis for detection of penicillin for nearly 40 years (J. Pharm. Pharmacol., 1972, 24, 790; Chinese Pharmacopoeia Ed. II, 1995). Herein, experiments were performed on: (1) synthesis of MMPA used as a novel mercuric hapten, (2) preparation of mercuric antigen of MMPA-BSA, (3) production of antibodies by rabbits immunized with the antigen, and (4) properties of the antibodies studied by ELISA. The results show that: (1) the antigen is safe for immunized animals; (2) high titer antibodies against MMPA are obtained implying good immunogenicity of the antigen; (3) antisera show slightly higher affinity to OVA-GHS-HgCl than OVA-GSH, indicating weak specific affinity of antisera against mercuric ion. Even the weak specific affinity, the hapten and its antigen have potential uses in immunoassays of mercuric ion in environment and food samples, because of easy chemical selective conversion from mercuric ion to MMPA and complete decomposition of un-reacted penicillin in acidic solution. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Immunoassay of heavy metal ions has become an interesting field for the analyses of environment and food samples (Lerner et al., 1991; Blake et al., 1996a,b; Chakrabarti et al., 1994; Blake et al., 1996a,b; Khosraviani et al., 1998; Blake et al., 1998; Blake et al., 2001; Blake et al., 2003; Khosraviani et al., 2000; Delehanty et al., 2003; Love et al., 1993; Wylie et al., 1992; Wylie et al., 1991; Prudent and Hausman, 2003). In 1984–89, Meares et al. (Meares et al., 1984; Reardan et al., 1985; Moi et al., 1985; Meares, 1986) and Mukkala et al. (Mukkala et al., 1989) firstly developed the method of preparing complexes of metal–chelate–antibodies for clinical medical radio-imaging analyses. In 1994-03, Blake et al. (Blake et al., 1996a,b; Chakrabarti et al., 1994; Blake

⁎ Corresponding author. Cao is to be contacted at Tel.: +86 21 3420 5682; fax: +86 21 3420 5820. E-mail addresses: [email protected] (P. Zhou), [email protected] (C.-X. Cao). 0022-1759/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jim.2009.12.005

et al., 1996a,b; Khosraviani et al., 1998; Blake et al., 1998; Blake et al., 2001; Blake et al., 2003; Khosraviani et al., 2000; Delehanty et al., 2003) performed systemic studies on preparation of monoclonal antibodies (mAbs) by immunizing BALB/c mice with the relevant antigens, antibody-based sensors for numerous metal ions, and relevant immunoassay. Furthermore, the interaction between mAb and its relevant hapten has been studied. Love et al. (1993) illustrated the crystal structure of antigen-binding fragment of antibodies and explained the reasons for the binding properties between mAbs and haptens of metal–chelates. Blake et al. (2003) observed the extraordinary binding property of mAb against hapten of Pb(II) complex. However, seldom mAbs were generated for the relevant Hg(II)–chelates probably due to a low lethal dose of Hg(II) on immunized animal. In vivo, Hg(II) attached to metal–chelate– protein complex may be deprived by numerous functional enzymes and proteins. The accumulation of Hg(II) in vivo after several boosting immunizations usually leads to a toxic response or even death of animals. In addition, the mAbs induced by metal–chelate–protein complex preferably bind

2

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

to cagelike metal–chelates structure, rather than certain heavy metal ion (Khosraviani et al., 2000). The following two principles ought to be considered for design on mercuric antigen (Wylie et al., 1992): (1) the covalent binding of Hg(II) to ‘spacer arm’ (linkage between metal ion and carrier protein), and (2) enough exposure of Hg(II) to animal immune system. Accordingly Wylie et al. (1992) modified keyhole limpet hemocyanin (KLH) with reduced L-glutathione via amide bond, then combined HgCl2 to the group of –SH in glutathione covalently. The results showed that antibodies induced by the antigen of HgCl– glutathione–KLH could specially react with mercuric ion regardless of the presence of other metal ions. At the same time, Wylie et al. (1991) further conducted the ELISA analysis of Hg(II) in water. Recently Prudent and Hausman (2003) synthesized a novel organomercury conjugate which had potential use in the heavy metal ion immunoassay. Herein, a series of experiments was conducted on: (1) synthesis on a novel hapten of MMPA; (2) preparation of antigen MMPA-BSA; (3) immunization of rabbits with the antigen; and (4) ELISA analyses of antibodies in animal serum. The antigen has some advantages, e.g., safety to animals, good immunogenicity and simple syntheses. Below are the relevant results. 2. Materials and methods 2.1. Chemicals and reagents L-Glutathione (GSH) reduced, imidazole, HgCl2, ascorbic acid and 30% hydrogen peroxide were purchased from Shanghai Chemical Reagents Co. (Shanghai, China). Bovine serum albumin (BSA) was obtained from Lizhu Dongfeng Biotech Co. (Shanghai, China). 1-Ethyl-3-(3-di-methyl-aminopropyl) carbodiimide hydrochloride (EDC·HCl) was purchased from BBI Company. Penicillin G (1650 units/mg), ovalbumin egg (OVA), complete and incomplete Freund's adjuvants were purchased from Sigma-Aldrich Chemical Co. Tween 20, TMB, goat anti-rabbit IgG conjugated with HRP were purchased from Shanghai ShenHang Chemical Reagents Co. (Shanghai, China). New Zealand Rabbits were purchased from Shanghai SLK experimental animal Co. (Shanghai, China). All reagents available were at their highest purity grade.

2.2. Synthesis of hapten MMPA and antigens (Scheme 1) MMPA was synthesized accordingly (Bundgaard and Ilver, 1972; Committee of of Chinese pharmacopoeia, 1995). Briefly, 50 mg penicillin G and 1 g imidazole were dissolved in 10 mL water in a beaker. The pH value of solution was adjusted with 1 M HCl to pH 6.8. Then, 7 mL 8.0 mg/mL HgCl2 was added dropwise. The mixture was incubated in a water bath for 2 h at 60 °C. After that, the solution was acidified with 1.0 M HCl forming precipitate. The precipitate was washed three times by 1 M HCl. UV–vis and IR spectra analyses were conducted to identify MMPA structure and purity. The synthesized MMPA was coupled to BSA or OVA via the catalyst of EDC·HCl (Harlow and Lane, 1988). In order to achieve ideal coupling ratio of MMPA to BSA, some factors (e.g., reaction time and reaction molar ratio of MMPA and BSA) were investigated. Briefly, 40 mg hapten MMPA was dissolved in 20 mL 4% NaHCO3 in a beaker. After that, 200 mg

EDC∙HCl was added and the mixture was stirred for 10 min. Then the mixture was added with 50 mg BSA and stirred gently for 2, or 7, or 12 h. After that, the mixture was filtered two times to remove the precipitate, the filtrate containing the synthesized antigen was dialyzed exhaustively against 10 mM pH 7.4 PBS. UV–vis spectra could be used to monitor whether the coupling of MMPA and protein occurs or not. ICP-AES was used to determine the coupling ratio of MMPA and BSA. The synthesis and characterization of coating antigen MMPA-OVA were similar to these of MMPA-BSA. 2.3. Synthesis of coating antigen OVA-GSH-HgCl and OVA-GSH The syntheses of OVA-GSH-HgCl and OVA-GSH were described in Wylie et al. (1992). Here is a succinct description. 50 mg OVA was dissolved in 20 mL 4% NaHCO3 solution. Then 100 mg vitamin C (used to prevent oxidation of reduced GSH), 100 mg reduced L-GSH and 200 mg EDC∙HCl were added to OVA solution in order. The mixture was slowly stirred at room temperature for 5 h for complete conjugation and then dialyzed exhaustively against 10 mM pH 7.4 PBS. The purified antigen solution was divided into 2 equal portions, one portion was diluted to 50 mL which was coating antigen OVA-GSH; the other portion was added dropwise into 20 mL 0.3 mg/mL HgCl2 solution while stirring. After dialyzing against 10 mM pH 7.4 PBS, the pure coating antigen of OVA-GSH-HgCl was synthesized. ICP-AES was used to identify coupling ratio of OVA-GSH-HgCl (Wylie et al., 1992). 2.4. Rabbit and mouse immunization Four strong male New Zealand rabbits (∼3 kg) were used for yielding the relevant antibodies (the first two rabbits were injected with the immunizing antigen MMPA-BSA with 25/1 coupling ratio of hapten to carrier protein, the second two for the MMPA-BSA with 10/1 coupling ratio). 500 µg immunizing antigen (1 mg/mL) dissolved in 10 mM pH 7.4 PBS was emulsified with complete Freund's adjuvant (1/1, v/v). Then each rabbit was injected with the 500 µg immunizing antigen at hypodermic multi-sites as the basic immunization. After two weeks, each rabbit was boosted with an additional 300 µg antigen emulsified with incomplete Freund's adjuvant. The 2nd, 3rd and 4th boosting was performed every 14 days and rabbits were bled 7 days after each boosting. BALB/c mouse was used for the immunization with the antigen MMPA-BSA used above. The immunization of BALBb/c mouse was performed in accordance with the procedure described by Wylie et al. (1992). The immunizing antigen (1 mg/mL) in 10 mM pH 7.4 PBS was emulsified with complete Freund's adjuvant (1/1, v/v). After that the mice were injected with the 50 µg immunizing antigen as the basic immunization. After fourteen days of the first injection, the mice were boosted with an additional 50 µg antigen emulsified with incomplete Freund's adjuvant. The 2nd and 3rd boosting immunizations were given every two weeks. After seven days of the third boosting, the mice were bled via their trail veins. 2.5. Titration test 96-well microtiter plates were coated with 100 µL/well 4 µg/mL coating antigen of MMPA-OVA in 50 mM pH 9.6

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

carbonate buffer for 3 h in 37 °C water bath. After three washes with 10 mM pH 7.4 PBS plus 0.05% Tween 20 (PBST), the plates were blocked with 300 µL/well 1% OVA (w/v) in 10 mM pH 7.4 PBS. After 1 h incubation in 37 °C water bath, the plates were washed as described previously. Serial dilutions (1/1000, 1/27,000 and 1/83,700 etc.) of samples were prepared in 10 mM pH 7.4 PBS containing 1% OVA (w/v). 100 µL/well of diluted serum was transferred to the plates. After 1 h incubation in 37 °C water bath, the plates were washed, and 100 µL/well goat anti-rabbit IgG conjugated with HRP (1/1000 dilution) in 10 mM pH 7.4 PBS containing 1% OVA (w/v) was added. The plates were incubated for 1 h in 37 °C water bath. After PBST washing, 100 µL of a substrate solution (10 µL 30% H2O2 and 400 µL 0.6% TMB in dimethyl sulfoxide (DMSO) were added to 10 mL pH 5.6 citrate-acetate buffer) was added to each well. After 15 min, 50 µL 2 M H2SO4 were added and the absorbance value was read at 450 nm using a Shanghai KHB ST-360 Microplate Reader. 2.6. Analyses of antibodies specific affinity to MMPA by indirect ELISA 96-well microtiter plates were coated with 100 µL/well coating antigen MMPA-OVA and blocked by incubation for 1 h in 37 °C water bath with 1% OVA in 10 mM pH 7.4 PBS, the blank wells were coated with 1% OVA in 50 mM pH 9.6 mM carbonate buffer. 100 µL/well 1/1000 dilution antisera containing 1% OVA were added. The remaining steps were similar to those mentioned in titration test. The absorbance value was read at 450 nm. 2.7. Analyses of antibodies specific affinity to mercuric ion by indirect ELISA (Wylie et al., 1992; Harlow and Lane, 1988) OVA-GSH and OVA-GSH-HgCl were used as coating antigen and were both diluted to 20 µg/mL by 50 mM pH 9.6 carbonate buffer. 100 µL/well OVA-GSH or OVA-GSH-HgCl was added. After 1 h incubation in 37 °C water bath, the plates were blocked by 1%OVA. Then 100 µL/well 1/300 dilution antisera containing 1% OVA were added and incubated for 1 h. The remaining steps were similar to those mentioned above. 3. Results 3.1. Characterization of hapten MMPA synthesized The structure of MMPA is different from those of penicillin and its degradation products, because MMPA has a unique oxazolone ring instead of β-lactamic ring in penicillin (see Scheme 1). The structure of oxazolone ring in MMPA results in characteristic ultraviolet absorption at about 325 nm (Smith et al., 1967; de Weck and Eisen, 1960; Longridge and Timms, 1971). Thus, the analysis of UV–vis spectra can be used to monitor the successful synthesis of MMPA. Fig. S1 shows the comparisons of UV–vis spectra between MMPA and penicillin G. The comparisons in Fig. S1 manifest the characteristic 325 nm UV absorption of produced precipitate, which indicates the synthesis of MMPA. The analysis of UV– vis spectra of MMPA conjugated with BSA and OVA further implies the unique oxazolone ring of MMPA (see Section 3.2).

3

To further demonstrate the synthesis of MMPA, we performed IR analyses on both MMPA and penicillin. Fig. S2 shows the IR spectra of MMPA and penicillin. Clearly, βlactamic ring feature band 1774 cm− 1 (Fernández-González et al., 2005) disappeared in the spectrum of MMPA, indicating that β-lactamic ring was degraded completely after the synthesis of MMPA. The characteristic bands of 2500 cm− 1 and 1677 cm− 1of MMPA indicated the existences of –SH (or –S –Hg+) and –C C–C O (α, β unsaturated ketone) in MMPA, respectively. Other nonspecific bands both existed in penicillin G sodium and MMPA, such as 750 cm− 1, 700 cm− 1 (monosubstitution on aromatic ring), 2927 cm− 1 (–CH2), 1400, 1368 cm− 1 [–C(CH3)2]. Evidently, Fig. S2 supplied numerous results further implying the existence of oxazolone structure and the disappearance of β-lactamic ring in the hapten of MMPA. Scheme 1 also shows the existence of –S-HgCl in MMPA. This means that the analysis of ICP-AES can be used to directly monitor the synthesis of MMPA and the conjugation of MMPA to a carrier protein of BSA or OVA. This will be shown by the ICP-AES data in Table 1 in Section 3.2. Evidently, the UV–vis spectra in Fig. S1 and IR spectra in Fig. S2, together with the quantitative detection of mercuric element in MMPA-BSA and MMPA-OVA by ICP-AES in Table 1, demonstrate the successful synthesis of MMPA. 3.2. Synthesis and characterization of MMPA-BSA and MMPA-OVA According to the procedure of antigen in Section 2, we synthesized the immunizing antigen of MMPA-BSA and coating antigen of MMPA-OVA. The UV–vis spectra could be used to monitor the conjugation between the hapten of MMPA and carrier protein of BSA or OVA. As clearly shown in Fig. 1, there was a maximum absorption peak at about 325 nm in the UV spectra of BSA-MMPA and OVA-MMPA, indicating the link between MMPA and BSA (OVA). ICP-AES was used to quantitatively investigate the coupling ratio between MMPA and BSA during the preparation of antigen (see Table 1). It was clearly shown in Table 1 that coupling ratios were stable at 6/1 after 2 h reaction if the reaction molar ratio was set at M = 10/1, further increased reaction time had no obvious influence on the ratio. However, if M was set at 100/1, the coupling ratios were changed from 5/1 to 10/1 and to 25/1 if the reaction time was increased from 2 h to 4 h and finally to 12 h, respectively. Under the given conditions for the preparation of BSA-MMPA with the ratio of 25/1, the coupling ratio of MMPA-OVA was 9/1. These results evidently demonstrated that the synthesis conditions (e.g., reaction time and molar ratio between hapten and carrier protein) have strong influence on the coupling ratio. 3.3. Titration of antisera After the final boosting immunization with the antigen of MMPA-BSA (coupling ratio = 25/1), the antibody titration for each rabbit was analyzed by indirect ELISA described in Section 2. Fig. 2 shows the analyses on the titration of antisera. In Fig. 2, the negative control was achieved with normal serum. It was evidently revealed in Fig. 2 that (1) the titers of two rabbits were both at least up to 83,700 after the

4

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

Scheme 1. Synthesis procedure of MMPA and BSA-MMPA from penicillin G.

final boosting immunization, indicating the strong immunogenicity of the synthesized antigen of MMPA-BSA to rabbits; and (2) rabbit 2 presented a slightly higher concentration of antibodies than rabbit 1. However if MMPA-BSA with 10/1 coupling ratio was used for immunization of rabbits, the titers of antisera were not good enough. This implies the poor immunogenicity of MMPA-BSA with 10/1 coupling ratio. Hence, MMPA-BSA with 25/1 coupling ratio was used for the following experiments. 3.4. Affinity of antiserum to MMPA The affinity of rabbit antiserum to MMPA was further investigated by indirect ELISA after the final boosting immunization. Fig. 3 shows the results on the affinity against the hapten of MMPA. The negative control was achieved with normal serum, the optical density (OD) values were 0.27 for the normal serum of rabbit one and 0.21 for that of rabbit two, both were very low. The OD values for antiserum of rabbit one and two were respectively 0.51 and 0.42, when 1% OVA was used as the coating antigen. Greatly, the values for the antiserum of rabbit one and two were respectively increased

Table 1 Influence of synthesis conditions on the coupling ratios between MMPA and BSA. Conjugation reaction time (h)

Coupling ratio (M = 10/1) a

(M = 100/1) a

2 4 12

6/1 6/1 6/1

5/1 10/1 25/1

a M represents reaction molar ratio of hapten and carrier protein used for the preparation of antigen.

to 1.37 and 1.42, if the coating antigen was MMPA-OVA following by blocking with 1% OVA. No antiserum showed significant cross-reactivity against OVA alone. The affinity of mouse antiserum to MMPA was also studied by ELISA determination of antibody titer after the third boosting immunization. Table 2 exhibits the results of ELISA analyses on the affinity of antiserum. It was evident in Table 2 that (1) all of the five mice yielded much low affinity antibody towards the OVA, (2) the sera obtained from mouse 3 and 4 did not contain specific antibody having obvious affinity towards the MMPA, and (3) however the sera from mouse 1– 2 and 5 produced the special antibody with high affinity towards the MMPA. Both the experiments with rabbits and these with mice demonstrated the antigen of MMPA-BSA could elicit the tested animals to produce the relevant antibody having high affinity towards the hapten of MMPA. 3.5. Weak specific affinity of serum antibody to Hg(II) We further investigated the specific affinity of antisera against Hg(II). To detect the specific affinity, we synthesized two coating antigens of OVA-GSH-HgCl and OVA-GSH accordingly (Wylie et al., 1992; Harlow and Lane, 1988). The ICP-AES analysis revealed that the coupling ratio was 4/1. With the two coating antigens, we analyzed the specific affinity of antisera to Hg(II) (Wylie et al., 1992). Table 3 displays the specific affinity of antisera against Hg(II). In Table 3, the OD value (=0.445) of OVA-GSH-HgCl for the normal serum of the first rabbit is less than that of OVA-GSH (=0.461), the relative difference is −3.4%. Whereas, the OD value (= 1.07) of OVA-GSH-HgCl for the antiserum of rabbit one is clearly higher than that of OVA-GSH (=0.924), the relative difference of OD values between the OVA-GSH-HgCl

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

5

Fig. 3. Affinity of antisera to the hapten of MMPA. N: normal serum for negative control; O: OVA coating; O-h: OVA-MMPA coating. Before the indirect ELISA analyses, sera were diluted 1000 folds. The blank wells (O) were coated with OVA, the wells (O-h) used for determining specific affinity to MMPA were coated with 4 µg/mL coating antigen OVA-MMPA which were subsequently blocked by 1%OVA.

Fig. 1. UV–vis spectra of MMPA-BSA (the upper) and MMPA-OVA (the bottom). 0.10 mg/mL MMPA-BSA or MMPA-OVA solution, or BSA or OVA solution was prepared, then scanned under 190–450 nm with an instrument of Thermo Electron Evolution 300 UV–vis spectrophotometer.

and OVA-GSH is 15.8%. The final relative difference of specific affinity of the first rabbit's serum against Hg(II) is up to 19.2% (=15.8% + 3.4%) higher than that of normal serum. However, the antiserum of rabbit two shows little affinity difference

between OVA-GSH-HgCl and OVA-GSH. The final difference of the second rabbit's antiserum against Hg(II) is slightly (5.4% = 1.5% + 3.9%) higher than that of normal serum. Considering the standard deviation of OD value, there was no obvious significant difference between OD values of OVA-GSH and OVA-GSH-HgCl for rabbit 2, and the OD value of OVA-GSH for rabbit 1 was slightly higher than that of OVA-GSH-HgCl. All of the results indicate the weak specific affinity of antiserum antibody towards the metal ion of Hg(II). The weak affinity was further observed during the immunization of mouse with the sample antigen of MMPA-BSA. Table 4 shows the ELISA determination of antibody titer in the antiserum of mouse immunized with MMPA-BSA. The antiserum was neutralized with BSA at first. Then, the antibody titers towards the MMPA-BSA, BSA, Hg-GSH-OVA and GSHOVA were detected. It was clearly shown in Table 4 that the five immunized mice (except for the fourth mouse) all produced high titer affinity antibody against the hapten of MMPA. The

Table 2 ELISA determination of antibody titer in antiserum from mouse after the third boosting immunization with MMPA-BSA. Mouse

No. 1 No. 2 No. 3 No. 4 No. 5 Normal

Fig. 2. Titration test of two rabbits' antisera after the final boosting immunization with MMPA-BSA (coupling ratio = 25/1). Indirect ELISA was performed for titration test.

Neutralized with a

Antibody titer b MMPA-OVA c

OVA c

None Penicillin None Penicillin None Penicillin None Penicillin None Penicillin None Penicillin

128,000 64,000 128,000 64,000 64,000 64,000 32,000 32,000 128,000 64,000 <2000 <2000

<2000 <2000 <2000 <2000 <2000 <2000

a The samples were diluted with PBS contained 5% penicillin G sodium salt for neutralization. b Antibody titer was showed as a reciprocal of antiserum dilution. The first dilution of serum sample with PBS buffer was 1/2000. c The antigen (10 μg/mL) was coasted on ELISA plate.

6

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

Table 3 Comparisons of OD values between the OVA-GSH and OVA-GSH-HgCl coating. Rabbit

OVA-GSH

OVA-GSH-HgCl

Relative difference a

1 Normal serum 1 2 Normal serum 2

0.924 ± 0.016 0.461 ± 0.032 0.976 ± 0.023 0.383 ± 0.036

1.07 ± 0.035 0.445 ± 0.045 0.991 ± 0.045 0.368 ± 0.033

+ 15.8 − 3.4 +1.5 − 3.9

a The relative difference is calculated by the formula: [(OD450 of BSA-GSHHgCl − OD450 of BSA-GSH) / (OD450 of BSA-GSH)] × 100%. Each number represents the average of triplicate determinations.

titer of mouse serum (No. 3) against Hg-GSH-OVA was 128,000 being slightly higher than that against GSH-OVA (=64,000). However, there was no clear difference between the antibody titers towards the two antigens of Hg-GSH-OVA and GSH-OVA for the mouse of No. 1, 4 and 5. The brief experiments in Table 4 further revealed that the antibody produced by immunized mice had weak affinity to the metal ion of Hg(II). 3.6. Possible mechanism of weak affinity of serum antibody to Hg(II) As compared with the much high relative differences of specific affinity of antiserum against Hg(II) (44.8%, 24.3%, 66.9% and 52.6% for mouse 4, 5, 6 and 7, respectively) in the previous work (Wylie et al., 1992), the values of relative difference in Table 3 were only 15.2%, indicating weak affinity of antibody to Hg(II). The weak affinity of antibody to Hg(II) herein might be caused by the following three reasons. At first, MMPA has three epitopes, viz., the phenyl group, the oxazolone ring and the group of –S-HgCl. In contrast to the strong immunogenicity of phenyl group and oxazolone ring, the group of –S-HgCl has weak heterologous character to animal immunization system. However, the hapten of GSHHgCl (see Fig. 1 in Wylie et al., 1992) contains the three epitopes, viz., the –NH2, the –COOH and –S-HgCl. As compared with the weak immunogenicity of –NH2 and –COOH, the epitope of –S-HgCl in GSH-HgCl has stronger immunogenicity. Secondly, the phenyl group and oxazolone ring have large sizes and give strong space shielding efficiency to the epitope of –S-HgCl. The shielding efficiency further isolates the epitope of –S-HgCl from the recognition by animal immunization system. Whereas, the groups of –NH2 and –COOH in GSH-HgCl (Wylie et al., 1992) have small sizes and are of weak shielding efficiency Table 4 ELISA determination on antibody titer in mouse antiserum after 3 boosting immunizations with MMPA-BSA. Mouse

No. 1 No. 2 No. 3 No. 4 No. 5 Normal

Neutralized with

BSA BSA BSA BSA BSA BSA

Antibody titer a MMPABSA b

BSA b

Hg-GSHOVA b

GSHOVA b

128,000 128,000 128,000 32,000 128,000 < 1000

<1000 <1000 <1000 <1000 <1000 <1000

32,000 128,000 – 4000 64,000 <1000

32,000 64,000 – 4000 64,000 <1000

a The antibody titer was showed as a reciprocal of antiserum dilution. The first dilution of serum sample with PBS buffer was 1/1000. b The antigen (10 μg/mL) was coated on ELISA plate.

to the epitope of –S-HgCl. Thirdly, the chain connecting the mercuric atom and BSA in the antigen of MMPA-BSA is –S–C–C– CO–, and the chain bridging the mercuric atom and KLH in the antigen of KLH-GSH-HgCl is –S–CH2–CH–NH–CH2–CO– (Wylie et al., 1992). Clearly, the former chain is shorter than the latter one. The short chain may enhance the space shielding efficiency to the epitope of –S-HgCl. 3.7. Potential use of MMPA in immunoassay of Hg(II) The weak specific affinity of serum antibody against Hg(II) observed in Section 3.5 indicated the great difficulty to achieve an antibody recognizing Hg(II), and to develop the relevant immunoassay for the direct specific detection of Hg (II). However, the difficulty could be overcome by simple chemical pretreatment of sample. In our lab, Hg(II) and Cu(II) could be equivalently synthesized as MMPA and copper mercaptide of penicillenic acid (CMPA) under the given experimental conditions. The syntheses under the given conditions were very easy for the two metal ions of Hg and Cu, but not for the other metal ions, e.g., Pb(II), Pb(III), Cd(II), Cr(II), Cr(III), Co(II) and Ni(II) etc. The un-reacted penicillin after the syntheses could be completely degraded under addition of acid into the reaction mixture. The ion of Cu(II) in a sample before the syntheses could be almost completely removed by using a specific copper reagent. Thus, with a simple chemical treatment of sample, viz., the removal of Cu (II) with the copper reagent and the degradation of unreacted penicillin, one could equivalently and exclusively synthesize MMPA with Hg(II) and penicillin. The MMPA could be further detected via an ELISA with antibody recognizing MMPA molecule, rather than Hg combined with MMPA. Thus, with proper pretreatment of sample mentioned above, one could use an antibody recognizing MMPA rather than Hg to detect content of Hg(II) in sample matrix. This might be a new strategy for immunoassay of heavy metal ions. Such work of using an antibody recognizing MMPA rather than Hg for an immunoassay of metal ion is being ploughed in our lab. This may lead to the development of novel immunoassay of Hg(II), which do not directly detect the quantity of Hg(II), but do indirectly monitor the concentration of Hg(II) via the direct analysis of MMPA. 4. Discussion A satisfying antigen for mercuric ion should fit the following criteria, viz., no obvious toxicity to immunized animals, high immunogenicity, and enough exposure of heavy metal ion to the immune system of the animal. The mercuric antigen synthesized by using the novel hapten of MMPA meets the former two criteria as will be shown below. Firstly, MMPA is of good chemical stability both in vitro and in vivo, no obvious toxic response was observed in our immunized animals (including rabbits and mice). This is clearly different from the antigen of heavy metal ion synthesized via the bifunction reagents, such as EDTA. Since, our unpublished experiments revealed that the animals immunized with Hg–EDTA–protein died easily. Secondly, the antigen of MMPA-BSA holds strong immunogenicity to animal immune system. We also observed that coupling ratio between MMPA and BSA has great influence on

P. Xie et al. / Journal of Immunological Methods 353 (2010) 1–7

the immune response of animals. If the coupling ratio of MMPA-BSA was less than 10/1, the immune response of animals was much weak. When MMPA-BSA with the coupling ratio of 25/1 was used, the immunized animals could yield high titer antibodies (Tables 2 and 4, and Fig. 3). This observation was in agreement with the finding: the higher coupling ratio usually increases the strength and specificity of the immune response (Singh et al., 2004; Marco et al., 1995). Thirdly, as compared with the antigen of KLH-GSH-HgCl (Wylie et al., 1992), the synthesized MMPA-BSA supplied a less exposure of mercuric ions to the immune system of animal, as analyzed in Section 3.6. The less exposure might induce the weak specific affinity of serum antibody to Hg(II), as observed in Tables 3 and 4. Fourthly, even MMPA resulted in the weak specific affinity of serum antibody to Hg(II) as shown in Section 3.5, it had potential application in the development on immunoassay of mercuric ion (see Section 3.7), with the present chemical techniques, viz., the selective synthesis of metal mercaptide of penicillenic acid, the chemical separation of Hg(II) and Cu (II), and the decomposition of excessive penicillin. Acknowledgements The authors are grateful for the funding of the National ‘863’ Scientific-Technological Key Program of China (No. 2007AA10Z401), the National Key Basic Research Program (973 Program) of China (No. 2009CB118906), the NSFC (No. 20675051, 20821005 and 20805031), and the Instrumental Analytic Centre of SJTU. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jim.2009.12.005. References Blake, D.A., Blake II, R.C., Mehraban, K., Andrey, R.P., 1998. Immunoassays for metal ions. Anal. Chim. Acta 376, 13. Blake, D.A., Chakrabarti, P., Khosraviani, M., Hatcher, F.M., Westhoff, C.M., Goebel, P., Wiley, D.E., Blake II, R.C., 1996a. Metal binding properties of a monoclonal antibody directed toward metal–chelate complexes. J. Biol. Chem. 271, 27677. Blake, D.A., Dawson, G.N., Chakrabarti, P., Hatcher, F.M., Van Emon, J.M., Gerlach, C.L., Johnson, J.C. (Eds.), 1996b. Environmental Immunochemical Methods: Perspectives and Applications. American Chemical Society, Washington, DC. p. 10. Blake, D.A., Jones, R.M., Blake II, R.C., Pavlov, A.R., Darwish, I.A., Yu, H.N., 2001. Antibody-based sensors for heavy metal ions. Biosens & Bioelectron. 16, 799. Blake II, R.C., Delehanty, J.B., Khosraviani, M., Yu, H.N., Jones, R.M., Blake, D.A., 2003. Allosteric binding properties of a monoclonal antibody and its Fab fragment. Biochemistry 42, 497. Bundgaard, H., Ilver, K., 1972. A new spectrophotometric method for the determination of penicillins. J. Pharm. Pharmacol. 24, 790.

7

Chakrabarti, P., Hatcher, F.M., Blake II, R.C., Ladd, P.A., Blake, D.A., 1994. Enzyme immunoassay to determine heavy metals using antibodies to specific metal–EDTA complexes: optimization and validation of an immunoassay for soluble indium. Anal. Biochem. 217, 70. Committee of Chinese pharmacopoeia, 1995. Chinese Pharmacopoeia Ed. II. Delehanty, J.B., Jones, R.M., Bishop, T.C., Blake, D.A., 2003. Identification of important residues in metal–chelate recognition by monoclonal antibodies. Biochemistry 42, 14173. de Weck, A.L., Eisen, H.N., 1960. Some immunochemical properties of penicillenic acid. J. Exp. Med. 112, 1227. Fernández-González, A., Badía, R., Díaz-García, M.E., 2005. Insights into the reaction of β-lactam antibiotics with copper(II) ions in aqueous and micellar media: kinetic and spectrometric studies. Anal. Biochem. 341, 113. Harlow, E., Lane, D., 1988. Antibodies: A Laboratory Manual. Cold Spring Harbor Lab, Cold Spring Harbor, NY. p. 84. Khosraviani, M., Blake II, R.C., Pavlov, A.R., Lorbach, S.C., Yu, H., Delehanty, J.B., Brechbiel, M.W., Blake, D.A., 2000. Binding properties of a monoclonal antibody directed toward lead–chelate complexes. Bioconjugate. Chem. 11, 267. Khosraviani, M., Pavlov, A.R., Flowers, G.C., Blake, D.A., 1998. Detection of heavy metals by immunoassay: optimization and validation of a rapid, portable assay for ionic cadmium. Environ. Sci. Technol. 32, 137. Lerner, R.A., Benkovic, S.J., Schultz, P.G., 1991. At the crossroads of chemistry and immunology catalytic antibodies. Science 252, 659. Longridge, J.L., Timms, D., 1971. Penicillenic acid, the mechanism of the acid and base catalysed hydrolysis reactions. J. Chem. Soc. B. 852. Love, R.A., Villafranca, J.E., Aust, R.M., Nakamura, K.K., Jue, R.A., Major Jr., J.G., Radhakrishnan, R., Butler, W.F., 1993. How the anti-(metal chelate) antibody CHA255 is specific for the metal ion of its antigen: X-ray structures for two Fab′/hapten complexes with different metals in the chelate. Biochemistry 32, 10950. Marco, M.P., Gee, S., Hammock, B.D., 1995. Immunochemical techniques for environmental analysis II. antibody production and immunoassay development. Trend. Anal. Chem. 14, 415. Meares, C.F., 1986. Chelating agents for the binding of metal ions to antibodies. Int. J. Rad. Appl. Instru. B. 13, 311. Meares, C.F., McCall, M.J., Reardan, D.T., Goodwin, D.A., Diamanti, C.I., McTigue, M., 1984. Conjugation of antibodies with bifunctional chelating agents: isothiocyanate and bromoacetamide reagents, methods of analysis, and subsequent addition of metal ions. Anal. Biochem. 142, 68. Moi, M.K., Meares, C.F., McCall, M.J., Cole, W.C., DeNardo, S.J., 1985. Copper chelates as probes of biological systems: stable copper complexes with a macrocyclic bifunctional chelating agent. Anal. Biochem. 148, 249. Mukkala, V.M., Mikola, H., Hemmilä, I., 1989. The synthesis and use of activated N-benzyl derivatives of diethylenetriaminetetraacetic acids: alternative reagents for labeling of antibodies with metal ions. Anal. Biochem. 176, 319. Prudent, C.K., Hausman, M.C., 2003. Organomercury bioconjugate synthesis and characterization by matrix-assisted laser desorption ionization mass spectrometry. Bioconjugat. Chem. 14, 1270. Reardan, D.T., Meares, C.F., Goodwin, D.A., Mctgue, M., David, G.S., Stonge, M.R., Leung, J.L.P., Bartholomew, R.M., Frincke, J.M., 1985. Antibodies against metal chelates. Nature 316, 265. Singh, K.V., Kaur, J., Varshney, G.C., Raje, M., Suri, C.R., 2004. Synthesis and characterization of papten–protein conjugates for antibody production against small molecules. Bioconjugate. Chem. 15, 168. Smith, J.W.G., de Grey, G.E., Patel, V.J., 1967. The spectrophotometric determination of ampicillin. The Analyst. 92, 247. Wylie, D.E., Carson, L.D., Carlson, R., Wagner, F.W., Schuster, S.M., 1991. Detection of mercuric ions in water by ELISA with a mercury-specific antibody. Anal. Biochem. 194, 381. Wylie, D.E., Lu, D., Carlson, L.D., Carlson, R., Babacan, K.F., Schuster, S.M., Wagner, F.W., 1992. Immunology, monoclonal antibodies specific for mercuric ions. Proc. Natl. Acad. Sci. USA 89, 4104.