| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Oak Ridge Conference |
1 Abbott Laboratories, Department 9FQ, Abbott Park IL.
aAddress correspondence to this author at: Department 9FQ, AP20, 100 Abbott Park Rd., Abbott Park IL 60064-6015. Fax 847-938-3271; e-mail john.c.russell{at}abbott.com.
Abstract
Background: Conventional methods for conjugation of macromolecules, such as antibodies and reporter groups, typically yield a mixture ranging from unconjugated starting materials to large aggregates. We explored the use of a solid-phase process to allow improved control in conjugation of macromolecules for use in immunodiagnostic reagents.
Methods: Activated components were sequentially delivered to an immobilized core protein, linking in concentric layers. For immunodiagnostic reagents, proteins with the desired signaling properties were added as interior layers and binding proteins were placed in the final surface layer. After assembly, the conjugates were released into solution by cleaving the linker holding the core protein to the support. Conjugates were prepared with use of three different reporter agents: R-phycoerythrin for microsphere fluorescence flow immunoassay, alkaline phosphatase for enzyme immunoassay, and acridinium for magnetic chemiluminescence immunoassay. For each reporter, six conjugates were prepared with various concentrations of both the reporter and an antibody directed against the 
-subunit of thyroid-stimulating hormone (TSH), and the complexes were tested in appropriate assay formats for measurement of TSH.
Results: Products ranged in mass from 
1 to 20 MDa. HPLC analysis of the conjugates on a gel-permeation column showed sizes and chromophore contents highly consistent with the intended structures. In appropriate assay formats, the signal generated by a conjugate increased with incubation time, then plateaued at an intensity approximately proportional to the reporter content but relatively independent of the antibody con-tent of the conjugate. The time required to reach this maximum decreased with increasing antibody content.
Conclusion: The high degree of structural control available with solid-phase assembly and the close correlation of structure with desired function of the resulting conjugates make this an attractive method for preparation of an important class of in vitro diagnostic reagents.
The field of in vitro diagnostics makes extensive use of reagents consisting of binding agents chemically linked to substances that provide signals detectable by instruments. Many methods have been devised to perform this linkage without damaging the desired properties of the linked materials (1)(2). In cases in which both binding and signaling components are proteins, such as linkage of an antibody to an enzyme, conventional coupling methods usually yield a mixture ranging from unconjugated starting materials to large aggregates. Even after extensive purification, the product usually contains species of several different sizes. These mixtures often perform quite well in the intended assay systems, but standardization of conditions to give a product with consistent performance can be difficult.
Our recently described method of protein conjugation on solid phase provides homogeneous preparations of well-defined assemblies in which the desired protein constituents are located at the chosen positions (3). This ability to control the conjugate structure allows detailed study of the effects of structural features on the properties of the assembly and makes possible the rational optimization of these properties to meet specified goals. A common goal might be to prepare a high-performance reagent, i.e., one with rapid, high-affinity, specific binding to its target; minimal nonspecific binding to other materials expected to be present; and a strong signal for each binding event. Another goal might be to minimize the consumption of an expensive binding or signaling protein while maintaining the performance goals of the product.
The solid-phase conjugation process is diagramed in Fig. 1
. In the first step a suitable protein, treated with reagents that provide reactive functional groups, is immobilized on a solid support via a cleavable linker. This immobilized protein becomes the core, or reaction center for attachment of other proteins. Proteins with the desired signaling activity are activated with complementary functional groups and then added to the support, where they link to the core protein. Several layers of appropriately activated proteins can be added sequentially to increase the signaling capacity. The layer proteins are added in excess to assure that each reaction center absorbs all that will fit on the surface generated by the previous step. After each step, the unbound excess protein is washed from the support and measured to determine the quantity bound. The increasing size of the construct allows each subsequent layer to contain more protein. The final layer, at the surface of the assembly, contains the desired binding agent. Finally, remaining active groups on the proteins are quenched, and the bond holding the core protein to the support is cleaved, releasing the conjugate into solution.
|
In this work we use the solid-phase method to prepare reagents for three different optical detection systems: absorbance, fluorescence, and chemiluminescence. For each of these we prepared a series of six conjugates, varying the quantities of signaling agent and binding agent in each molecular unit. Characterization of these conjugates by gel-permeation HPLC gave size and spectral results consistent with those calculated from their construction. Tests of the conjugates in appropriate assay formats demonstrated activities consistent with their structures: signal intensity increased with content of reporter, and the incubation time required to bind to the assay target decreased with increasing antibody content.
Materials and Methods
materials
R-Phycoerythrin (RPE)1
was obtained from Prozyme. Calf intestinal alkaline phosphatase (AP) was obtained from Boehringer Mannheim. The monoclonal antibody directed against the subunit of human thyroid-stimulating hormone (TSH; MIT 0414) was obtained from Genzyme. Beaded agarose (Biogel A-50m, medium) was obtained from Bio-Rad. Bovine serum albumin (BSA); sodium periodate; N-ethylmaleimide; 5,5'-dithio-bis(2-nitrobenzoic acid); tris(hydroxymethyl)aminomethane; CHAPS; triethanolamine hydrochloride; 1-ethyl-3,3-(dimethylaminopropyl)carbodiimide; 2-mercaptoethanesulfonic acid, sodium salt; and 
-maleimidobutyric acid N-hydroxysuccinimide ester were obtained from Sigma. N-Succinimidyl S-acetylthioacetate and 
-[N-maleimidocaproic acid]hydrazide were obtained from Pierce. Hydroxylamine (50% solution) was obtained from Aldrich. 3-[9-({{4-[(2,5-dioxo-1-pyrrolidinyl)oxy]-4-oxobutyl}[(4-methylphenyl)-sulfonyl]amino}-carbonyl)-10-acridiniumyl]-1-propanesulfonate; the monoclonal antibody against the ß subunit of TSH, calibration solutions with known concentrations of TSH; and Architect Pretrigger (cat. no. 6E25-65) and Trigger (cat. no. 6C55-60) solutions were obtained from Abbott Laboratories. Paramagnetic microspheres (4.8-µm diameter, containing 10.6% iron) were obtained from Polymer Laboratories. Carboxylate-modified latex microspheres (5.6-µm diameter) were obtained from Bangs Laboratories.
BSA conjugated to acridinium (BAc) was prepared by incubating 11 mg of BSA, 9.5 mg of CHAPS, and 2.3 mg of 3-[9-({{4-[(2,5-dioxo-1-pyrrolidinyl)oxy]-4-oxobutyl}[(4-methylphenyl)-sulfonyl]amino}-carbonyl)-10-acridiniumyl]-1-propanesulfonate (4) for 90 min in 180 µL of 100 mmol/L sodium phosphate at pH 7.8, followed by desalting into phosphate-buffered saline (PBS) adjusted to pH 6.0. Spectrophotometric analysis indicated an incorporation ratio of 13.2 acridinium chromophores per BSA molecule.
Modification of the proteins to give activated forms containing sulfhydryl and maleimide groups was performed as described previously (3). Maleimide groups were added to BSA, BAc, and AP by use of -maleimidobutyric acid N-hydroxysuccinimide ester at concentrations ranging from 0.5 to 20 mmol/L in triethanolamine or phosphate buffer at pH 8.0, incubating at room temperature for 1 h, and then desalting into PBS containing 5 mmol/L EDTA. Sulfhydryl groups were added by incubating the protein with N-succinimidyl S-acetylthioacetate, at 20 mmol/L for RPE and at 0.2 mmol/L for anti-TSH- monoclonal antibody, in triethanolamine buffer at pH 8.0 for 1 h and then adding hydroxylamine to a concentration of 10 g/L and additionally for the antibody, dithiothreitol to 5 mmol/L. After incubation for 15 min at room temperature, the mixture was passed through a desalting column equilibrated with PBS containing 5 mmol/L EDTA.
oxidation of agarose support
Beaded agarose (200 mL) solids were washed with water and resuspended in water to a total volume of 300 mL and cooled in an ice bath. With the mixture stirring, 6.0 mL of 10 mmol/L sodium periodate was added over 20 min. The mixture was left at 4 °C for 12 h, after which 500 µL of glycerol was added to consume any remaining sodium periodate. After 3 h at room temperature, the solids were washed with water and then stored at 4 °C.
activation of oxidized agarose
To a 20-mL column, we added 10 mL of oxidized agarose. This was washed with 20 mL of PBS containing 2 g/L CHAPS and 5 mmol/L EDTA. The washed support was vortex-mixed with 3 mL of the same buffer and 150 µL of 100 mmol/L -[N-maleimidocaproic acid]hydrazide in dimethylformamide. After 18 h at 4 °C, the column was drained, washed with 15 mL of cold Tris-CHAPS-EDTA (TCE; 10 mmol/L Tris, 2 g/L CHAPS, 5 mmol/L EDTA), pH 7.0, and placed in ice.
immobilization of rpe-sh
To each of six 2-mL columns (Table 1
), we added 600 µL of the -[N-maleimidocaproic acid]hydrazide-treated support, and the mixtures were washed with 2 mL of TCE, pH 7.0. The columns were cooled in ice, the indicated quantity of sulfhydryl-RPE (RPE-SH) was added to each, and the solids were thoroughly dispersed by vortex-mixing. After the columns had incubated for 20 min in ice with occasional vortex-mixing, we added 10 µL of 100 mmol/L 2-mercaptoethanesulfonic acid, sodium salt, to each to deactivate the remaining maleimide groups bound to the support. After incubation for 5 min in ice, the columns were drained and washed with 4.0 mL of cold TCE, pH 7.5. The absorbance of the first 2.0 mL of the effluent was measured at 565 nm to determine the quantity of RPE-SH in the effluent. This was subtracted from that originally added to determine the quantity bound to the support (Table 1
).
|
addition of activated proteins to immobilized rpe-sh
Except for the time spent draining and washing, the columns were kept in an ice bath throughout the conjugation procedure. The quantity of activated protein indicated in Table 1
was added to each, along with 100 µL of 1 mol/L magnesium chloride and sufficient TCE (pH 7.5) for complete dispersal of the solids on vortex-mixing. The columns were incubated in ice for 30 min with occasional vortex-mixing and then drained and washed with TCE (pH 7.5); the next activated protein was then added. The quantity of activated protein remaining in the effluent was determined by measuring the absorbance at 565 nm for RPE [molar absorptivity at 565 nm (565 nm) = 1 960 000 (mol/L)–1cm–1], at 370 nm for BAc [370 nm = 184 000 (mol/L)–1cm–1, assuming 13.2 acridinium chromophores per BSA molecule], and at 280 nm for other proteins. For BSA, the 280 nm was 43 600 (mol/L)–1cm–1, but the addition of 24 maleimide (Mal) groups increased this, giving, for BSA-Mal, an 280 nm of 54 600 (mol/L)–1cm–1. For AP-Mal, the 280 nm was 140 000 (mol/L)–1cm–1, and for the antibody conjugated with maleimide (Ab-Mal) and the sulfhydryl group (Ab-SH), an 280 nm of 210 000 (mol/L)–1cm–1 was used.
release of conjugates from support
After washing after the final protein addition, the solids were treated with 5 µL of 100 mmol/L 2-mercaptoethanesulfonic acid, sodium salt, to deactivate the residual maleimide groups. After 5 min, the columns were washed with TCE (pH 7.5), 10 µL of 100 mmol/L N-ethylmaleimide was added to deactivate the residual sulfhydryl groups, and the mixtures were left on ice until the protein additions were complete for all reactions. The columns were washed with TCE (pH 7.0), the solids were then dispersed by vortex-mixing with an additional 200 µL of TCE (pH 7.0) and 5 µL of 500 g/L hydroxylamine, and the mixtures were left at 4 °C for 14 h. The columns were then drained directly into desalting columns equilibrated with PBS, and the products were washed through both columns with TCE (pH 8.0) and PBS; 2.0 mL of eluate containing product was collected from each column.
A difference from the original process is that in this work the quantity of RPE-SH added to each reaction mixture in the first step (immobilization of core protein) was varied depending on the number of cycles of conjugation intended. If the immobilized core groups are located too close together on the support, there is a risk that added activated protein will bridge between them, giving an assembly containing two core groups. With multiple conjugation cycles, as the assembly increases in size this risk increases. By using a lower density of core protein, we made the average distance between immobilized core groups larger, allowing more cycles and, therefore, larger conjugates to be assembled while minimizing the number of neighboring reaction centers that became cross-linked by added protein.
hplc
To 45 mL of each released conjugate we added 5 µL of 100 g/L CHAPS in PBS. A 30-µL sample was passed through a Whatman MacrosphereTM GPC 1000A column [250 x4.6 mm (i.d.)] with 10 g/L CHAPS in PBS as the mobile phase at 0.2 mL/min, with simultaneous detection at 280, 370, and 566 nm using a Beckman Coulter System GoldTM 168 diode array detector.
microsphere coating
Magnetic and latex microspheres were coated with antibody directed against the ß subunit of TSH by a modification of the method described in Bangs Laboratories Technote 205 (5). Microspheres were incubated 1 h with antibody at 10 mg of antibody per 1 g of microsphere in 10 mmol/L sodium phosphate (pH 5.9). The microspheres were then centrifuged or collected on a magnet as appropriate and resuspended in 50 mmol/L MES, pH 6.0. We added 1-ethyl-3,3-(dimethylaminopropyl)carbodiimide at 10 mg per 1 g of microspheres, and the mixture was rotated at least 60 min. The microspheres were washed with a buffer consisting of 10 g/L BSA and 5 g/L Tween 20 in PBS and resuspended in the same buffer. The Luminex100TM microsphere fluorescence analyzer was used to determine the concentrations of the microspheres.
microsphere flow studies
Coated microspheres at 4000/µL in a buffer consisting of 10 g/L BSA and 5 g/L Tween 20 in PBS were incubated at least 30 min with an equal volume of TSH calibrator. To 20 µL of this mixture was added 20 µL of the RPE conjugate in the same buffer to give a final concentration of 1.0 nmol/L. After incubation, 5-µL samples of the mixture were diluted in 150 µL of the same buffer and incubated at room temperature for 90 s; 10 µL of this mixture was then aspirated into a Luminex100 microsphere fluorescence analyzer, and the median fluorescence intensity of 100 microspheres was reported.
enzyme immunoassay studies
Coated microspheres at 10 000/µL in a buffer consisting of 10 g/L BSA and 5 g/L Tween 20 in PBS were incubated at least 30 min with an equal volume of TSH calibrator. To 50 µL of this mixture was added 50 µL of the AP conjugate in a buffer containing 10 mmol/L bis-tris propane, 150 mmol/L NaCl, 1 mmol/L MgCl2, 0.1 mmol/L ZnCl2, 10 g/L BSA, and 0.4 g/L Tween 20 (pH 7.0) to give a final conjugate concentration of 1.0 nmol/L. After incubation for 30 min, the mixture was diluted with 500 µL of the same buffer and centrifuged, and the sedimented microspheres were washed with 500 µL of the same buffer and resuspended in 50 µL. To 40 µL of this mixture was added 100 µL of p-nitrophenylphosphate substrate, and the change in absorbance at 405 nm was monitored on a Molecular Devices plate reader over 20 min.
magnetic chemiluminescence immunoassay studies
A microplate luminometer (EG&G Berthold Microlumat PlusTM Model LB96V) was used to assess the acridinium conjugates. A KingFisherTM magnetic particle processor (Thermo LabSystems) was used to manipulate the coated magnetic microspheres, analyte, and conjugate before triggering the luminescence. For the study, 100 mL of magnetic microspheres at 4000/µL and 50 µL of TSH calibrator were incubated for 20 min, and the microspheres were washed and transferred to 50 µL of the conjugate solution. After incubation for 5 min, the microspheres were washed and transferred to 50 µL of Pretrigger solution. This is an acidic solution that releases the conjugate from the microspheres. After 2 min, the microspheres were removed. The microplate was placed on the microplate luminometer, 100 µL of Trigger (a solution of alkaline hydrogen peroxide) was added, and the luminescence signal was measured for 6 s.
Results
construction of conjugates
The conjugates in this work were assembled with sulfhydryl-modified RPE as the core protein. These were followed by an alternating sequence of maleimide- and sulfhydryl-modified proteins to provide the appropriate linking groups at each step. Measurement of the quantity of activated protein added and washed off of the support allowed quantitative assessment of the protein uptake in each step. For each reporter type, a series of conjugates was prepared, with the number of reporter groups and antibodies varied to investigate the effect of each on conjugate performance. Table 2
shows the unit construction of each conjugate, along with calculated and measured properties. The unit construction is the molar quantity of protein bound to the support in each step, normalized to that of the core protein. Assuming that the assemblies do not join together or fragment during or after preparation, the unit construction gives a meaningful description of the number and positions of each protein component in the final assembly and provides a basis for calculating such properties as molecular weight and absorptivity. Nomenclature of the conjugates reflects their reporter and antibody content.
|
characterization of conjugates
The molar concentration of each conjugate was determined based on its absorbance at 565 nm, the absorbance maximum for the RPE chromophore. For the series containing RPE as the reporter group, the total calculated RPE was divided by the number of RPE units bound per core to obtain the conjugate concentration. The series containing AP and acridinium as reporter contained only a single core RPE in each conjugate; therefore, the measured RPE concentration was the same as the conjugate concentration.
HPLC traces of the conjugates separated on a size-exclusion column and monitored at 565 nm to detect RPE, 370 nm to detect acridinium, and at 280 nm to detect all proteins are shown in Fig. 2
. Although no protein calibrators were available for the molecular mass ranges of these conjugates, the peak widths and relative retention times are consistent with narrow size distributions at masses corresponding to the calculated unit constructions. For all three of the series, the absorbance at 280 nm reflected both the reporter and the antibody content of the conjugates. For the middle portion of each series, in which the reporter content was held constant and the antibody content increased, the retention time decreased and the absorbance at 280 nm increased as expected for the increasing size and protein content. For RPE conjugates a-f, the absorbance intensities at 565 nm increased with increasing RPE content. Conjugates b-e had the same RPE content but an increasing antibody content, apparent in the increasing absorbance at 280 nm. Both the AP and acridinium conjugate series contained only the single core RPE group, which is reflected in the low absorbances at 565 nm. The absorbance at 280 nm increased with reporter content and with antibody content. In the case of the acridinium conjugates, the absorbance at 370 nm provided an independent measure of the reporter content.
|
performance of conjugates
All conjugates were tested in immunoassay formats in which microspheres coated with antibodies bound TSH through its ß subunit to form a sandwich with the conjugate, which bound the TSH subunit. The analytical instrument then quantitatively detected the portion of conjugate bound to the microsphere. Assuming that a single molecule of microsphere-bound analyte would bind a single unit of conjugate, the signal should be proportional to the number of reporter groups carried by the conjugate. The antibody content of the conjugate was expected to affect the incubation time needed to achieve full binding and, in cases of limited antibody affinity, the position of equilibrium.
In the flow immunoassay system used to compare the RPE conjugates, coated microspheres were incubated with analyte for a time sufficient to give equilibrium binding and then exposed to conjugate. At various times of exposure, the microspheres were sampled by the instrument, which measured the fluorescence intensity of conjugate bound to each of a predetermined number of the microspheres and reported the median of these signals (6).
The time course of signal development on addition of the conjugates to microspheres containing bound TSH is shown in Fig. 3A
. For each conjugate, the measured intensity increased rapidly with the first few minutes of incubation and then plateaued at a maximum value dependent on the RPE content of the conjugate. When the RPE content of the conjugate was held constant, increasing the antibody content had relatively little effect on the maximum signal, but did increase the rate at which this maximum was approached. The increase in signal with increasing analyte concentration for the different conjugates with incubation time held constant at 20 min is shown in Fig. 3B
. The dose response was nearly linear, with the signal per bound analyte strongly dependent on RPE content of the conjugate. The signals for conjugates b, c, d, and e were very similar, even with a 13-fold difference in antibody content, a finding that underlines the potential usefulness of the defined conjugates to make optimum use of a binding agent with limited availability.
|
The AP conjugates were compared in an enzyme immunoassay format using the same microspheres that were in the microsphere flow immunoassay. In this format, after the microspheres bound the analyte, the mixture was exposed to the conjugate, which bound in proportion to the quantity of bound analyte. The unbound conjugate was then washed off, phosphatase substrate was added, and the rate at which product appeared was measured by absorbance spectroscopy. The effect of analyte concentration on signal for each of AP conjugates g–l is shown in Fig. 4A
. As with the RPE conjugates in the flow immunoassay system, the AP conjugates showed dose responses that depended mainly on the reporter content, with little effect of antibody content.
|
The acridinium conjugates were compared in magnetic chemiluminescence immunoassay format using magnetic microspheres coated with the same antibody used on the latex microspheres. The microsphere processing was similar to that of the enzyme immunoassay format, except that the magnetic properties of the microspheres were used in the transfer between incubation and wash solutions. After the unbound conjugate was washed off, the conjugate was released from the microspheres by exposure to the acidic Pretrigger solution, and the exhausted microspheres were removed. Chemiluminescence was induced by adding Trigger solution to the released acridinium in the Pretrigger solution and was measured on a luminometer. The effect of analyte concentration on chemiluminescence signal for each of acridinium conjugates m–r is shown in Fig. 4B
. Additionally, results are shown for a conventional acridinium conjugate composed of a single antibody (from the same clone used with the assembled conjugates) containing six bound acridinium groups. As with the RPE and AP conjugates, the dose responses depended on reporter content, with little influence of antibody content at the incubation time used. As expected from its lower acridinium content, the conventional conjugate showed a much lower response than the larger assembled conjugates. Table 3
shows the signaling characteristics for each of the acridinium conjugates for samples containing 0 and 0.5 mIU/L TSH. The assembled conjugates showed higher background signals than the single-antibody conjugate, presumably attributable in part to the higher reporter content and in part to nonspecific binding of the conjugate to the microsphere. This is reflected in the sensitivities calculated from the results. Although the dose response of the most sensitive conjugate was higher by a factor of 25, the sensitivity was improved by a factor of only 5. Additionally, the expected response in counts per acridinium group decreased by a factor of 2 from the conventional conjugate to the largest assembled conjugate. This may be attributable to damage to the acridinium group in the assembly process or to nonideal behavior of the large layered assembly in the generation of signal under assay conditions. For all acridinium conjugates, the measured sensitivities were well within the range of third-generation TSH assays; however, a meaningful assessment for comparison with commercially available assays would require testing beyond the scope of the present work.
|
Discussion
The solid-phase conjugation method provides a high degree of control in the assembly of nanometer-scale structures using proteins as building blocks. Specified proteins are linked in specified numbers in the structure under mild conditions that retain the needed properties of the proteins. This makes it possible to incorporate multiple activities in a single conjugate and to adjust the amounts of any of these activities independently of the others. The stepwise procedure also allows the proteins to be placed in the structure at the location most suitable for the intended activity. In particular, components intended to interact directly with the surroundings, such as binding proteins, can be placed at the exterior surface of the assembly, whereas those that may have undesired interactions can be placed in the interior.
Other means have been used to increase the reporter content of detection reagents. In the ABC (avidin-biotin complex) method and related procedures, the detection reagent is assembled during the course of the assay by sequential addition of biotin-labeled antibody, avidin, and biotin-labeled enzyme (7)(8). Multiple layers of enzyme can also be built up by stepwise addition of enzyme conjugated to antibody directed against the host species of the antibody in the previous layer (9). These methods can give highly sensitive measurements, but at the cost of increasing the number of steps in the assay. Large aggregates of phycoerythrins can be prepared by covalent linkage (10) or by chemically stabilizing naturally occurring phycobilisomes, a naturally aggregated form of phycoerythrin isolated from algae, and the aggregates conjugated to binding proteins to produce reagents with very high fluorescence response (11)(12). Detection reagents can also be prepared by linking antibodies and reporter proteins directly to the same reactive polymer or polystyrene microspheres (13)(14)(15)(16). Signaling capacity for small-molecule reporters has been increased by use of similarly labeled secondary antibodies that bind to labels on the first antibody (17), and by linkage of dendrimers to the binding protein to increase labeling capacity (18).
In this work we exploited the high degree of structural control of the solid-phase process to assemble signaling agents for three different detection technologies with structural variations expected to affect their performance. For each we prepared a series of conjugates, independently varying the number of signaling proteins in, and binding proteins at the surface of each structural unit. The use of RPE as the core protein provided a distinctive optical absorbance that facilitated both the HPLC characterization and the measurement of molar concentrations of the final released conjugates.
Each of the reporter types had characteristics requiring attention in the conjugation process. We found in earlier work (J. Russell, unpublished studies) that directly linking RPE to RPE decreases its fluorescence yield, whereas use of an intervening protein layer between RPE layers gives conjugates retaining more of the fluorescence. We therefore used the BSA-Mal layers between RPE-SH layers. Additionally, we found that conjugates with a large surface of exposed RPE groups show nonspecific binding to the antibody-coated microspheres but that this effect is decreased by covering the conjugate surface with IgG or BSA. Because some of the intended RPE conjugates contain insufficient antibody to provide this protective function, we included a surface layer of BSA before the antibody layer.
The chemiluminescent conjugates used the small molecule acridinium to generate signal. Acridinium is usually linked directly to exposed lysine groups of antibodies to prepare chemiluminescent reagents. In this work, acridinium was linked to BSA to prepare an intermediate component, which was then treated to add maleimide groups and incorporated into the conjugates. This allowed many more acridinium groups to be placed in a molecular unit of conjugate than the direct linkage permits. It is worth noting that the 13 acridinium molecules per BSA accounted for only 10% of the mass of this material. It is likely that carriers can be developed to increase the loading capacity even higher, to prepare conjugates with yet higher signals. The signaling capacities of both AP and acridinium were found to be damaged by addition of sulfhydryl groups; therefore, conjugates using these materials were prepared with maleimide-modified reporter protein alternating with sulfhydryl-modified BSA.
Results of HPLC characterization of the conjugates were highly consistent with the structures expected from the synthetic steps in their preparation, with elution times and optical absorbances appropriate for the calculated molecular weights and compositions. Functional characterization, using appropriate immunoassay formats, also showed activities consistent with structure. For each reporter type, the response of signal intensity to analyte concentration was approximately proportional to the number of reporter groups per conjugate unit. The antibody content affected the rate at which conjugate bound to the analyte-coated microspheres, but prolonged incubation time gave nearly complete binding even at the lowest antibody content tested.
In summary, the solid-phase process gives structural control that can be exploited to prepare protein assemblies with quantitatively specified properties. Detection reagents can be rationally optimized to have low or high signal intensities appropriate to the analyte and assay format. Use of expensive binding agents can be limited without compromising performance. In this work, we examined the method in an immunoassay format for only a few detection technologies. It is likely that it can be applied beneficially in many other systems as well.
Footnotes
1 Nonstandard abbreviations: RPE, R-phycoerythrin; AP, alkaline phosphatase; TSH, thyroid-stimulating hormone; BSA, bovine serum albumin; BAc, bovine serum albumin linked to acridinium; PBS, phosphate-buffered saline (150 mmol/L sodium chloride, 20 mmol/L sodium phosphate, pH 7.2); TCE, Tris–CHAPS–EDTA (100 mmol/L Tris, 2 g/L CHAPS, 5 mmol/L EDTA); and Mal, maleimide. 
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
