Site-Specifically Labeled Photoprotein-Thyroxine Conjugates Using

Site-Specifically Labeled Photoprotein-Thyroxine Conjugates Using Aequorin Mutants Containing Unique Cysteine Residues: Applications for Binding Assay...
0 downloads 0 Views 119KB Size
Download PDF
140

Bioconjugate Chem. 2000, 11, 140−145

Site-Specifically Labeled Photoprotein-Thyroxine Conjugates Using Aequorin Mutants Containing Unique Cysteine Residues: Applications for Binding Assays (Part II) J. C. Lewis, L. C. Cullen, and S. Daunert* Department of Chemistry, University of Kentucky, Lexington, Kentucky 40506-0055. Received June 21, 1999; Revised Manuscript Received November 29, 1999

The jellyfish Aequorea victoria produces a protein, aequorin, which belongs to the class of Ca2+dependent photoproteins known for their ability to emit visible light. This property of aequorin has allowed for its as a bioluminescent label in binding assays for a variety of analytes. Due to the excellent detection limits we demonstrated in assays for small peptides using a fusion protein between the peptide of interest and the photoprotein, our next goal was to expand the range of possible analytes for producing homogeneous populations of conjugates with the aequorin label to those that were nonpeptidic in nature. Recently, we prepared and characterized four aequorin mutants containing unique cysteine residues at various positions in the polypeptide chain. In the work reported here, the four aequorin mutants were each conjugated with a maleimide-activated methyl ester derivative of thyroxine, a hormone frequently determined to evaluate thyroid function. The thyroxine-aequorin mutant conjugates were characterized in terms of the bioluminescence activities and binding properties with an anti-thyroxine monoclonal antibody for possible future employment in either heterogeneous or homogeneous binding assays for thyroxine and/or other desired analytes.

INTRODUCTION

The photoprotein aequorin was originally isolated from the jellyfish Aequorea victoria located in Friday Harbor, Washington, by Shimomura and co-workers (1). The photoprotein was shown to emit blue light with the requirement of a single cofactor, calcium. Due to its calcium-dependent, nontoxic nature, aequorin has found widespread application as an intracellular calcium indicator (2, 3). Aequorin is a photoprotein complex consisting of an apoprotein, a luciferin (coelenterazine), and molecular oxygen. The binding of calcium ions triggers the protein to undergo a conformational change, which leads to the oxidation of the noncovalently bound coelenterazine. The products of the reaction are the blue fluorescent protein, CO2, and light (λmax = 469 nm) (4). The bioluminescent signal produced by the photoprotein has virtually no background signal and a quantum yield of 15%, allowing the photoprotein to be detected down to attomol levels (5). The lack of need to irradiate the sample eliminates problems associated with light scattering, unselective excitation, and source stability (6). Bioluminescence can have a linear dynamic range up to 6 orders of magnitude and can be measured readily using inexpensive luminometers. Additionally, employing luminometry offers the advantage that while many biological samples contain a variety of fluorescent molecules, bioluminescence occurs relatively rarely in the biological kingdom, thus, furthering decreasing possible background problems in the sample (7). Because of all these advantages of bioluminescence detection, aequorin has not only been used as a calcium indicator but also as a highly sensitive bioluminescent label in binding assays as well (8-10). * To whom correspondence should be addressed. Phone: (606) 257-7060. Fax: (606) 323-1069. E-mail: [email protected].

The photoprotein has been employed indirectly as a label through biotinylation of aequorin, and subsequent use in conjunction with avidin or streptavidin. In one such assay to detect Salmonella antigen, the bioluminescent immunoassay demonstrated 50 times better sensitivity in comparison to the corresponding enzyme assay (11). The photoprotein has also been used directly through chemical conjugation to either the antigen (12) or antibody (13). A particular advantage of using a relatively small protein as a quantitative label is that the gene encoding for the protein can be fused to another desired protein or peptide. Thus, one-to-one homogeneous populations of conjugates can be produced employing genetic engineering techniques (14). Such conjugates yield better assay performance, by overcoming many problems associated with conventional chemical conjugation methods. Our group and others have successfully fused peptides (15) and proteins (16) to the N-terminus of aequorin to yield homogeneous populations of aequorin-labeled conjugates for assay development. Recently, we reported the preparation and characterization of four aequorin mutants containing unique cysteine residues for site-specific conjugation and immobilization [part I (17)]. The four aequorin mutants obtained were based on a previously described cysteinefree mutant that demonstrated increased bioluminescence activity in comparison to the native photoprotein (18). The four mutants contained single cysteine residues at positions 5, 53, 71, and 84 in the polypeptide chain. To produce homogeneous populations of aequorin conjugates for assay development using nonpeptidic analytes, the photoprotein mutants were conjugated to a maleimide-activated methyl ester derivative of thyroxine synthesized in our laboratory. Thyroxine was chosen as a model analyte because it is one of two hormones assayed routinely in clinical laboratories along with triiodothyronine. These two hormones are typically determined

10.1021/bc990081s CCC: $19.00 © 2000 American Chemical Society Published on Web 02/04/2000

Site-Specifically Labeled Aequorin-Ligand Conjugates

Bioconjugate Chem., Vol. 11, No. 2, 2000 141

using radioimmunoassays (19). The four thyroxineaequorin conjugates were evaluated in terms of their bioluminescence activities and binding properties to a anti-thyroxine monoclonal antibody. The ability to sitespecifically label aequorin to desired nonpeptidic analytes has possible applications in the development of highly sensitive heterogeneous or homogeneous binding assays for important small biomolecules such as drugs, vitamins, or hormones. To the best of our knowledge, this is the first time that aequorin has been genetically modified and then site-specifically conjugated to a desired analyte for the development of an immunoassay. EXPERIMENTAL PROCEDURES

Reagents. All restriction endonucleases, T4 DNA ligase, Luria Bertani (LB) broth, LB agar, DNA mass ladder, ampicillin, and kanamycin were purchased from Gibco-BRL (Gaithersburg, MD). Tris(hydroxymethyl) amino methane (Tris), ethylenediaminetetraacetic acid (EDTA) disodium salt, sodium chloride, dithiothreitol (DTT), thyroxine, and all other reagents were obtained from Sigma (St. Louis, MO). HQH, quaternized polyethyleneimine anion exchanger, was purchased from Perseptive Biosystems (Cambridge, MA). Coelenterazine was obtained from Biosynth International (Naperville, IL). All oligonucleotides (primers) employed for PCR were provided by Operon Technologies (Alameda, CA). The polymerase selected for PCR was Pfu polymerase from Stratagene (La Jolla, CA). The bicinchoninic acid (BCA) protein assay kit, Sulfo-GMBS, and Ellman’s Reagent were purchased from Pierce (Rockford, IL). The antihuman thyroxine monoclonal antibody was obtained from Cortex Biochem (San Leandro, CA). The goat anti-mouse IgG (Fc) particles were from Bangs Laboratories (Fishers, IN). All solutions were prepared using deionized (Milli-Q Water Purification system, Millipore, Bedford, MA) distilled water. All chemicals were of reagent grade or better and were used as received. Bacterial Strains and Plasmids. The expression vector pSbt and bacterial strain Bacillus subtilis (trpC2, MetB10, npr, apr::cat) was provided by Dr. Philip N. Bryan, University of Maryland. The mammalian expression vector, pMtAEQ, containing the apoaequorin gene was purchased from Molecular Probes (Eugene, OR). The Escherichia coli strain, JM109, employed for initial cloning reactions was from Gibco-BRL (Gaithersburg, MD). Apparatus. Bioluminescence measurements were made on an Optocomp I luminometer from GEM Biomedical (Carrboro, NC) using a 100 µL fixed-volume injector. Absorbance measurements were conducted on a HP Diode-Array Spectrophotometer from Hewlett-Packard (Palo Alto, CA). All luminescence measurements are the average of a minimum of three replicates and have been corrected for the blank. All PCR reactions were performed using a Perkin-Elmer GeneAmp PCR System 2400 (Norwalk, CT). PCR-Based Site-Specific Mutagenesis. The five mutant aequorins including one mutant in which the cysteine residues at positions 145, 152, and 180 were replaced with serine (mutant S), and four additional mutants containing unique cysteine residues at positions 5, 53, 71, and 84 in the polypeptide chain were prepared as previously described [part I (17)]. The site-specific mutations were confirmed through DNA sequencing performed at the Macromolecular Center (University of Kentucky). The final DNA products containing the apoaequorin mutants were cloned into the pSbt expres-

Figure 1. Schematic representation of the pSD110-114 expression vectors using PCR-based site-directed mutagenesis to obtain the desired mutations.

sion vector as EcoRI-SalI fragments to yield the pSD110114 vectors shown in Figure 1. All molecular biology procedures were conducted using standard protocols (20). Expression and Purification of Mutant Apoaequorins. The five apoaequorin mutants were expressed and purified as described [part I (17)]. The fractions containing the mutant protein were combined and concentrated using a Centriplus 3000 device from Millipore (Bedford, MA). The desired buffer exchange was achieved employing a polyacrylamide 6000 desalting column from Pierce (Rockford, IL). The purity of the mutant proteins was verified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) using an electrophoresis PhastSystem from Pharmacia Biotech (Uppsala, Sweden). Polyacrylamide (12.5%) gels were developed by the method of silver staining. The protein concentrations were estimated by using the BCA protein assay, with BSA as the standard. Conversion of Apoaequorin to Aequorin. The mutant aequorins were generated from the apoproteins by mixing a set amount of the purified proteins in a glass test tube with a molar excess of coelenterazine in 10 mM Tris-HCl, 0.15 M NaCl, 2.0 mM EDTA, 5 mM DTT, pH 7.4, (TBS/EDTA/DTT) buffer. The mixture was briefly vortexed, and then placed on ice for a minimum of 4 h. Calibration Plots of Aequorin Mutants. To compare the bioluminescence activities of the mutant aequorins containing unique cysteine residues with mutant S, calibration plots were constructed by serially diluting stock solutions of the aequorin mutants in TBS/EDTA buffer containing 1% BSA. Luminescence was triggered by injecting 100 µL of a 100 mM CaCl2, 100 mM TrisHCl, pH 7.5, buffer (activation buffer) to a volume of 100 µL of the different concentrations of the aequorin mutants. The bioluminescence signal was integrated over a 3-s time period. Synthesis of Methyl L-Thyroxyl-(γ-maleimido)butanamide (maleimide-activated thyroxine). The free acid of thyroxine was methylated by placing 1 g of the acid in the bottom of a test tube followed by the immediate addition of 10 mL of methanol. HCl gas was then bubbled through the suspension producing a yellow solid that was cooled with an ice bath. After allowing the mixture to reach RT, the above step was repeated. Since treatment with HCl/methanol produces the salt of the amino acid, the mixture was reacted with base (KOH) to yield the free amino acid. The solid was isolated by filtration, and washed with cold absolute ethanol and cold

142 Bioconjugate Chem., Vol. 11, No. 2, 2000

ether. The product was subjected to high vacuum overnight in a P2O5 desiccator to dry the solid. To synthesize the maleimide-activated thyroxine suitable for conjugation to the photoproteins, 75 mg of the methylated L-thyroxine was placed in a round-bottom flask with 5 mL of dry dimethyl sulfoxide (DMSO). The mixture was stirred for 10 min before 36.25 mg of N-[γ-maleimidobutyryloxy]sulfosuccinimide ester (Sulfo-GMBS) was added, and the mixture was allowed to further stir overnight at RT under argon. The solvent was removed from the product in a rotavaporator under high vacuum. 1H NMR was performed to verify the desired structure in (DMSOd6): δ 1.27 (triplet, J ) 5.66 Hz, CH2), 1.64 (multiplet, J ) 5.66 Hz, 7.51 Hz, CH2), 2.06 (triplet, J ) 7.51 Hz, CH2), 2.90 (quartet, J ) 4.36 Hz, CH2), 3.60 (singlet, OCH3), 4.45 (J ) 4.36 Hz, CH), 6.98 (singlet, CH olefinic), 7.03 (singlet, CH aromatic), 7.78 (singlet, CH aromatic), 8.33 (doublet, J ) 8.06 Hz, NH), 9.20 (broad OH). Site-Specific Conjugation of Aequorin Mutants with Maleimide-Activated Methyl Ester Derivative of Thyroxine. A volume of 2 mL of each of the purified apoaequorin mutants was incubated with fresh DTT at a final concentration of 5 mM for 1 h at RT. The mixtures were concentrated to 0.5 mL using a Centriplus 3000 device and then desalted with a polyacrylamide 6000 desalting column into a phosphate-buffered saline solution (PBS) with a pH of 7.0. A 20-fold molar excess of the maleimide-activated methyl ester derivative of thyroxine in dimethyl formamide (DMF) was added to the protein solution, and the mixture was allowed to stir overnight at 4 °C in the dark. The excess thyroxine was removed through desalting as described above. The thyroxine:aequorin molar ratio was calculated before addition of the chromophore (coelenterazine) using an extinction coefficient for thyroxine of E1%,1cm ) 73.7 (21). A 5-fold molar excess of coelenterazine was then added to the mixture, which was left at 4 °C in the dark for an additional 24 h to form the photoprotein. The excess coelenterazine was removed through dialysis for 12 h at 4 °C against TBS/EDTA buffer. Care was taken at each step not to expose the labile thyroxine to light. Calibration Plots of Thyroxine-Aequorin Mutant Conjugates. Stock solutions of the thyroxine-aequorin mutant conjugates were serially diluted in TBS/EDTA buffer containing 1% gelatin (assay buffer). Luminescence was triggered by injecting 100 µL of a 100 mM CaCl2, 100 mM Tris-HCl, pH 7.5, buffer (activation buffer) to a volume of 100 µL of the different concentrations of the thyroxine-aequorin mutant conjugates. The bioluminescence signal was integrated over a 3 s time period. Homogeneous Association Study. The anti-thyroxine antibody was diluted from a stock solution of 500 µg/ mL in assay buffer to various concentrations ranging from 5 µg/mL to 150 µg/mL. Next, a fixed concentration of the thyroxine-aequorin mutant in a volume of 100 µL was added to the same volume of the various antibody concentrations followed by an incubation step of 2 h at 4 °C with shaking (150 rpm). To determine whether an increase or decrease in the bioluminescence signal occurred upon binding of the conjugate to the antibody, a blank containing 100 µL of the assay buffer in place of the antibody plus the thyroxine-aequorin mutant was used. The luminescence signal was measured as described above. Heterogeneous Association Study. The anti-thyroxine antibody was diluted from a stock solution of 500 ng/mL in assay buffer to various concentrations ranging from 10 to 200 ng/mL. Goat anti-mouse IgG (Fc specific) magnetic particles (10 mg/mL) were diluted with assay

Lewis et al.

buffer to 1 mg/mL. A volume of 100 µL of the antibody dilutions was added to 100 µL of the diluted anti-mouse magnetic particles followed by an incubation step of 1 h at 4 °C with shaking (150 rpm). Following a wash step to remove unbound antibody, a volume of 100 µL of a fixed concentration of the thyroxine-aequorin mutant was added and incubated with magnetic particles for 2 h at 4 °C with shaking (150 rpm). The particles were washed three times with assay buffer before measuring the bioluminescence signal. RESULTS AND DISCUSSION

The photoprotein aequorin has been successfully employed by us and others as a highly sensitive bioluminescent label in binding assays for important biomolecules. The bioluminescence signal produced requires only the addition of a single cofactor and calcium and occurs as a rapid flash of light lasting less than 10 s. The characteristics of the bioluminescence reaction allow the photoprotein to be detected down to attomol levels. Aequorin has been used to assay a diverse array of analytes such as hormones (22, 23), serum antibodies (7), cytokines (24), proteins and nucleic acids in Western and Southern blots (25), and vitamins (26). In most of these assays aequorin has been covalently attached to either the antigen or antibody being used by employing chemical conjugation methods that produce multi-labeled heterogeneous populations of conjugates. Such conjugation techniques often result in a large decrease in the bioluminescence activity of the photoprotein (22, 27). Also, it has been demonstrated that it is advantageous in the development of binding assays to use one-to-one homogeneous populations because such populations result in more sensitive assays (14). Recently, we have developed both homogeneous and heterogeneous binding assays for small biomolecules using aequorin as a label (15, 26). In the case of the homogeneous assay, we employed a biotinylated-aequorin in a competitive assay format that produced detections limits for biotin of 1 × 10-14 M (4 attomoles). For the heterogeneous assay, we used the technique of gene fusion to produce a one-to-one population of octapeptideaequorin conjugates that retained the full bioluminescence activity of the photoprotein. Even with a relatively low association constant between the octapeptide and the corresponding antibody of =10-7 M-1, detection limits for the octapeptide were on the order of 1 × 10-9 M. To further expand the range of assays to nonpeptidic analytes while still producing homogeneous population of conjugates, we prepared four different aequorin mutants. The mutants produced were based on a previous study in which all three cysteine residues present within the native protein had been replaced with serine producing a mutant aequorin that demonstrated increased bioluminescence activity (16). On the basis of this information, it was foreseable that aequorin mutants could be genetically engineered to contain unique cysteine residues at various positions for site-specific conjugation of a desired biomolecule. In part I (17) of this study, we produced four such mutants with single cysteine residues at positions 5, 53, 71, and 84. In the current study, we have chosen the thyroid hormone, thyroxine, as a model analyte for the site-specific conjugation to the photoprotein mutants in order to demonstrate the feasibility of using these conjugates in binding assays. To produce the aequorin mutants with unique cysteine residues, we chose the technique of PCR-based sitedirected mutagenesis [see part I (17)]. Table 1 lists the

Site-Specifically Labeled Aequorin-Ligand Conjugates

Bioconjugate Chem., Vol. 11, No. 2, 2000 143

Table 1. Characterization of Aequorin Mutants Containing Unique Cysteine Residues aequorin mutants

amino acid substitution

% relative bioluminescencea

no. of free sulfhydryl groupsb

5 53 71 84

Ser5 f Cys Glu53 f Cys Met71 f Cys Glu84 f Cys

102 98 13 14

1.3 0.98 0.90 0.93

a Relative bioluminescence activities are reported as the percentage of the maximum signal obtained for mutant S using 1 µg of each purified mutant aequorin in 350 µL of TBS/EDTA/DTT, pH 7.6, buffer containing 0.2 µg of coelenterazine. All measurements were performed in triplicate with 10 µL aliquots of the mutant aequorin mixtures. b The number of free sulfhydryl groups was determined using Ellman’s reagent according to the manufacturer’s instructions.

apoaequorin mutants prepared and the amino acid substitutions that were made by introducing the single cysteine residues into the apoprotein. Overlap extension PCR was used in the final step of the genetic manipulation of the apoaequorin gene to fuse the “pre”-signal peptide of the protein subtilisin to the N-terminus of the photoprotein. The peptide facilitates the secretion of a desired protein outside the cell into the culture medium, thus, simplifying the purifcation procedure. The final PCR product containing the DNA sequence encoding for the “pre”-signal peptide and the mutant apoaequorin was cloned into the pSbt expression vector as an EcoRI-SalI fragment (see Figure 1). The newly constructed expression vectors, pSD110-114, were used to express and purify the mutant apoaequorin proteins as previously described in part I (17). Typical yields for the purification procedure ranged from 22 to 43 mg of protein/L of culture. The protein concentrations were estimated using standard BCA assays. The purified apoaequorin mutants were converted to their corresponding photoproteins by mixing the apoproteins with a molar excess of coelenterazine. All five mutant proteins exhibited flash-type emission characteristics with the resulting luminescence signal rising sharply after injection with the activation buffer followed by an exponential decay of the light intensity [data shown in part I (17)]. The bioluminescence activities of the mutants were characterized by generating calibration curves for the mutants including mutant S, which contained no cysteine residues (Figure 2). The bioluminescence signal emitted by all five mutant aequorins extended over 5-6 orders of magnitude. Mutants S, 5, and 53 yielded detection limits on the order of 1 × 10-14 M, corresponding to 1 × 10-18 mol of photoprotein. The detection limits for mutants 71 and 84 were 1 order of magnitude higher at 1 × 10-13 M (1 × 10-17 mol of photoprotein). Further comparison can be made by determining the percent relative bioluminescence activity in of the mutants as shown in Table 1. Mutants 71 and 84 showed a substantial decrease in luminescence activity; however, both of these mutants were still detected down to very low levels. On the other hand, mutants 5 and 53 retained the full bioluminescence activity of the unmodified mutant S. Before conjugating the activated-thyroxine analogue with the photoprotein, it was verified that a single sulfhydryl group had been introduced into each mutant protein. This was accomplished by first reducing each mutant protein with DTT and then employing Ellman’s reagent [5,5′-dithio-bis-(2-nitrobenzoic acid (DTNB)], which reacts with free thiol groups to release TNB (412 ) 14 150 M-1 cm-1 at 412 nm). Reaction of each of the mutant apoaequorins with Ellman’s reagent indicated the pres-

Figure 2. Calibration plots of aequorin mutants including mutant S (+) with no cysteine residues, and mutants with unique cysteine residues at positions 5 ([), 53 (9), 71 (2), and 84 (b) in the polypeptide chain. Light intensity is measured in counts and integrated over a 3 s time period.

ence of one available thiol group for conjugation (see Table 1). The purified apoproteins were incubated for 1 h with fresh DTT to prevent any disulfide bond formation before buffer exchange into a PBS buffer at pH 7.0 was performed using polyacrylamide desalting columns. A 20fold molar excess of the maleimide-activated methyl ester derivative of thryoxine was used for the conjugation with protein concentrations ranging between 1.0 × 10-6 and 1.5 × 10-6 M. The reaction was carried out overnight at 4 °C in the dark to protect the thyroxine from exposure to light. After removing the excess conjugation reagent, the degree of conjugation of thyroxine to the apoaequorin mutants was determined using the extinction coefficient for thyroxine at 326 nm (21). The thyroxine:aequorin molar ratio for each of the mutants was determined to be approximately 1. Because the chromophoric unit for the photoprotein, coelenterazine, absorbs at 326 nm, conversion of the thyroxine-apoaequorin conjugates to photoproteins was not performed until after the excess thyroxine had been removed. At this time, a 5-fold molar excess of coelenterazine in methanol was added followed by an incubation step of 24 h at 4 °C. To study the effect of conjugation on the bioluminescence activity of the individual aequorin mutants, as well as to select appropriate conjugate concentrations for the remainder of the experiments, calibration plots were constructed for the each of the thyroxine-aequorin conjugates (Figure 3). By comparing the calibration curves generated before and after conjugation, it is evident that mutants 5 and 53 demonstrated little, if any, loss in their bioluminescence activity. However, the calibration curves generated for the thyroxine-aequorin conjugates using mutants 71 and 84 produced detection limits on the order of 1 × 10-11 M for the conjugates, which is 2 orders of magnitude higher than the above detection limits for the unconjugated mutant aequorins. The calibration curves were then used to select the appropriate concentration of the thyroxine-aequorin conjugates for association studies with the anti-thyroxine monoclonal antibody. Homogeneous binding curves were generated using fixed concentrations of the thyroxineaequorin conjugates with various dilutions of the antibody followed by an incubation step. To determine if a change in signal had occurred, blanks were used in which

144 Bioconjugate Chem., Vol. 11, No. 2, 2000

Figure 3. Calibration plots of thyroxine-aequorin mutant conjugates including maleimide-activated thyroxine conjugated to mutants 5 (9), 53 (2), 71 (b), and 84 ([).

Figure 4. Homogeneous association curve employing thyroxineaequorin mutant 71. The curve was obtained by incubating various concentrations of the anti-thyroxine monoclonal antibody with a set amount of the conjugate. Data are the average of ( one standard deviation (n ) 3).

an equal volume of buffer was substituted for the antibody. Of the four thyroxine-aequorin conjugates studied, one conjugate (mutant 71) produced an increase in signal upon binding with the anti-thyroxine antibody (Figure 4). Since this particular conjugate exhibited a loss in bioluminescence activity upon conjugation with thyroxine, it may be postulated that upon binding with the antibody the thyroxine moiety is somehow displaced from a position at which it interacts with certain amino acid residues within the protein structure causing an interference in the production of the luminescence signal by the photoprotein. This position may be in the hydrophobic region of the photoprotein that interacts with the chromophoric unit coelenterazine. Because of the lack of a three-dimensional structure of the protein, the explanation for of this observation cannot be verified. A maximum signal increase of 45% was observed at antibody concentrations of 50 µg/mL and greater using a final concentration of the conjugate of 5 × 10-9 M. Heterogeneous association curves were also generated employing the thyroxine-aequorin conjugates that dem-

Lewis et al.

Figure 5. Heterogeneous association curve employing thyroxine-aequorin mutants 5 (9) and 53 (b). A secondary antibody approach was used to obtain the curve employing magnetic particles coated with goat anti-mouse IgG (Fc specific) antibodies, and then varying the amount of anti-thyroxine antibody with a set amount of the conjugates.

onstrated no loss in bioluminescence activity, namely mutants 5 and 53, and therefore would be more suitable for development of heterogeneous assays for thyroxine. The curves were produced employing a secondary antibody approach using magnetic particles coated with Goat anti-mouse IgG (Fc specific) antibody. This method allows for less consumption of the primary antibody, as well as specific orientation of the antibody on the solid phase. Figure 5 shows both curves generated for the two thyroxine-aequorin conjugates (mutants 5 and 53). It is apparent from the figure that both conjugates did successfully bind to the anti-thyroxine monoclonal antibody. Also, using a final concentration of 8.5 × 10-12 M for the conjugates, the luminescence signal was observed to remain constant at antibody concentrations greater than 125 ng/mL, thus, indicating the presence of an excess amount of antibody. As described in part I (17), we determined that positions 5, 53, 71, and 84 in apoaequorin were viable for introduction of a unique cysteine residue. Here, we have demonstrated that site-specific conjugation to these unique cysteine residues is possible for all four mutants produced and that the four thyroxine-aequorin conjugates exhibit varying bioluminescence and binding properties. Of the four aequorin mutants conjugated to thyroxine, two of the mutants, 71 and 84, that previously showed a dramatic change in their bioluminescence activities, also produced interesting binding properties. The conjugate prepared with mutant 84 did not appear to bind with the anti-thyroxine monoclonal antibody to any degree (data not shown). However, the thyroxineaequorin mutant 71 conjugate produced a homogeneous change (the bioluminescence signal increased following binding to the antibody) in the bioluminescence signal upon binding with the anti-thyroxine antibody (the bioluminescence signal was observed to increase). This mutant has the potential to be employed in the development of homogeneous assays for small biomolecules; these assays require fewer steps to complete and present a number of advantages, for example, when automation of the assay is desired. We also demonstrated that generation of one-to-one homogeneous populations of thyroxine-aequorin conjugates in a site-specific manner enables the production of conjugates that have not been

Site-Specifically Labeled Aequorin-Ligand Conjugates

compromised in terms of the bioluminescence activity of the photoprotein label, as in the case of mutants 5 and 53. This may be especially critical for developing assays for analytes such as thyroxine, which contains four iodine atoms that may quench the luminescence signal (28). It is also essential to retain the full bioluminescence activity of the photoprotein following conjugation in order to achieve current and future requirements for the low detection limits needed for commonly assayed drugs, vitamins, hormones, etc. Finally, this is the first time that a bioluminescent photoprotein label has been genetically manipulated to produce homogeneous conjugates with nonpeptidic analytes for the purpose of developing immunoassays with superior performance. ACKNOWLEDGMENT

We would like to thank the Department of Energy (DEFG05-95ER62010), the National Institutes of Health (GM47915) for support of this work. J. C. Lewis acknowledges the National Science Foundation for an IGERT Predoctoral Fellowship. S. Daunert is a Cottrell Scholar and a Lilly Faculty Awardee. LITERATURE CITED (1) Shimomura, O., Johnson, F. H., and Saiga, Y. (1962) Extraction, purification and properties of aequorin, a bioluminescent protein from luminous hydromedusan, Aequorea. J. Cell. Comput. Physiol. 59, 223. (2) Rizzuto, R., Simpson, A. W. M., Brini, M., and Pozzan, T. (1992) Rapid changes of mitochondrial Ca2+ revealed by specifically targeted recombinant aequorin. Nature 358, 325. (3) Brini, M., Marsault, R., Bastianutto, C., Alvarez, J., Pozzan, T., and Rizzuto, R. (1995) Transfected aequorin in the measurement of cytosolic Ca2+ concentration J. Biol. Chem. 17, 9896-9903. (4) Ohmiya, Y., and Hirano, T. (1998) Shining the light: mechanism of the bioluminescence reaction of calciumbinding photoproteins. Chem. Biol. 3, 337-347. (5) Campbell, A. K., and Sala-Newby, G. (1993) In Fluorescent and Luminescent Probes for Biological Activity (W. T. Mason, Ed.) pp 58-82, Academic Press, New York. (6) Rongen, H. A. H., Hoetelmans, R. M. W., Bult, A., and Van Bennekom, W. P. (1994) Chemiluminescence and immunoassays. J. Pharm. Biomed. Anal. 12, 433. (7) Jackson, R. J., Fujihashi, K., Kiyono, H., and McGhee, J. R. (1996) Luminometry: a novel bioluminescent immunoassay enhances the quantitation of mucosal and systemic antibody responses. J. Immunol. Methods 190, 189. (8) Actor, J. K., Kuffner, T., Dezzutti, C. S., Hunter, R. L., and McNichol, J. M. (1998) A flash-type bioluminescent immunoassay that is more sensitive than radioimaging: quantitative detection of cytokine cDNA in activated and resting human cells. J. Immunol. Methods 211, 65. (9) Rigl, C. T., Rivera, H. N., Patel, M. T., Ball, R. T., Stults, N. L., and Smith, D. F. (1995) Bioluminescence immunoassays for human endocrine hormones based on aqualite, a calcium-activated photoprotein. Clin. Chem. 41, 1363. (10) Kricka, L. (1993) Ultrasensitive Immunoassay Techniques. Clin. Biochem. 26, 325. (11) Stults, N. L., Stocks, N. A., Cummings, R. D., Cormier, M. J., and Smith, D. F. (1991) Applications of recombinant bioluminescent proteins as probes for proteins and nucleic acids. In Bioluminescence and chemiluminescence: Current Status (P. E. Stanley, and L. J. Kricka, Eds.) pp 529-32, Wiley, Chichester. (12) Galvan, B., and Christopoulos, T. K. (1996) Bioluminescence hybridization assays using recombinant aequorin.

Bioconjugate Chem., Vol. 11, No. 2, 2000 145 Application to the detection of prostate-specific antigen mRNA. Anal. Chem. 68, 3545. (13) Rigl, C. T., Rivera, H. N., Patel, R. T., Ball, R. T., Stults, N. L., and Smith, D. F. (1995) Bioluminescence immunoassays for human endocrine hormones based on aqualite, a calcium-activated photoprotein. Clin. Chem. 41, 1363. (14) Witkowski, A., Kindy, M. S., Daunert, S., and Bachas, L. G. (1993) Enzyme-linked immunosorbet assay for an octapeptide based on a genetically engineered fusion protein. Anal. Chem. 65, 1147. (15) Ramanthan, S., Lewis, J. C., Kindy, M. S., and Daunert, S. (1998) Heterogeneous bioluminescence binding assay for an octapeptide using recombinant aequorin. Anal. Chim. Acta 369, 181. (16) Zenno, S., and Inouye, S. (1990) Bioluminescent immunoassay using a fusion protein of protein A and the photoprotein aequorin. Biochem. Biophys. Res. Commun. 171, 169. (17) Lewis, J. C., Lo´pez-Moya, J. J., and Daunert, S. (1999) Bioluminescence and secondary structure properties of aequorin mutants produced for site-specific conjugation and immobilization, Part I. Bioconjate Chem. 11, 65-70. (18) Tsuji, F. I., Inouye, S., Goto, T., and Sakaki, Y. (1986) Sitespecific mutagenesis of the calcium-binding photoprotein aequorin. Proc. Natl. Acad. Sci. U.S.A. 83, 8107. (19) Christofides, N. D., and Sheehan, C. P. (1995) Multicenter evaluation of enhanced chemiluminescence labeled-antibody immunoassay for free thyroxine. Clin. Chem. 41, 24. (20) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed., Cold Spring Harbor Laboratory Press, Plainview, NY. (21) Ferrua, B., Genetet, F., Savaron, M. L., Moulin, C., Salard, J. L., and Masseyeff, R. (1986) A novel enzyme immunoassay for total thyroxine using immobilized antibodies and hydrophobic chromatography purified thyroxine-peroxidase conjugate. J. Immunol. Methods 87, 137. (22) Rigl, C. T., Patel, M. T., Rivera, H. N., Stults, N. L., and Smith, D. F. (1994) A bioluminescent immunoassay for testosterone based on recombinant photoprotein, aequorin. In Biolumin. Chemilumin., Proc. Int. Symp. (A. K. Campbell, L. J. Kricka, and P. E. Stanley, Eds.) pp 345-348, Wiley, Chichester. (23) Rivera, H. N., Patel, M. T., Stults, N. L., Rigl, C. T., and Smith, D. F. (1994) A single-addition bioluminescence immunoassay for HCG that features recombinant aequorin as a dried reagent. Biolumin. Chemilumin., Proc. Int. Symp. (A. K. Campbell, L. J. Kricka, and P. E. Stanley, Eds.) pp 349352, Wiley, Chichester. (24) Actor, J. K., Kuffner, T., Dezzutti, C. S., Hunter, R. L., and McNichol, J. M. (1998) A flash-type bioluminescent immunoassay that is more sensitive than radioimaging: quantitative detection of cytokine cDNA in activated and resting human cells. J. Immunol. Methods 211, 65. (25) Stults, N. L., Stocks, N. F., Rivera, H., Gray, J., McCann, R. O., O’Kane, D., Cummings, R. D., Cormier, M. J., and Smith, D. F. (1992) Use of Recombinant biotinylated aequorin in microtiter and membrane-based assays: purification of recombinant apoaequorin from Escherichia coli. Biochemistry 31, 1433. (26) Witkowski, A., Ramanathan, S., and Daunert, S. (1994) Bioluminescence binding assay for biotin with attomol detection based on recombinant aequorin. Anal. Chem. 66, 1837. (27) Erikaku, E., Zenno, S., and Inouye, S. (1991) Bioluminescent immunoassay using a monomeric Fab-photoprotein aequorin conjugate. Biochem. Biophys. Res. Commun. 174, 1331. (28) Schroeder, H. R., and Yeager, F. M. (1978) Chemiluminescence yields and detection limits of some isoluminol derivatives in various oxidation systems. Anal. Chem. 50, 1114.

BC990081S