Le Comit~ de R~daction a regu des organisateurs du Colloque ,,Avances en Biologie Clinique,, les r~sum~s des communications qui seront pr6sent~es ~ I'occasion
du Salon du Laboratoire du 4 au 8 d~cembre 1989 ~ Paris Nord-Villepinte; il est heureux de publier ces textes qui refl~tent les progr~s technologiques
en mati6re de dosages utilisant une phase solide.
Advanced concepts in biology
Paris, 5-8 December 1989
Topic : solid phase technology in biospecific assays
The next Salon du Laboratoire will be held from December 4-8, 1989 in Paris-Nord (Villepinte). This International Scientific Conference will include participants from industry, universities and the clinical field. This will give an excellent opportunity to exchange ideas and review the most recent advances in the research and development of reagents and instruments for clinical biology.
A high level of scientific knowledge will be ensured by the participation of well known international specialists in research and development in clinical biology and by the support of the French Society of Clinical Biology.
Held under the auspices of the IFCC program established in cooperation with the AACC.
SCIENTIFIC AND ORGANIZING COMMITTEES
Mr. C. AUBRY
Mr. d. BIENVENU Pr. P.A. BONINI Dr. C. BURTIS Mr. J.F. DELAGNEAU Mr. Y. GOURMELIN
Mr. D.S. KABAKOFF Pr. R. MASSEYEFF
Mr. C. POLICARD Mr. P. SCHNIPELSKY
Dr. E. SOlNI Dr. A. TRUCHAUD
Pr. J.P. YVERT
Research and Development, Biom#rieux, France Biologist, International Relations, Soci6t6 Frangaise de Biologie Clinique, France Clinical Chemist, Institute San Raffaele, IFCC Congress Committee, Italy Oak Ridge Laboratory, President of the AACC, USA Research and Development, Pasteur Diagnostics, France Biologist, SFBC Instrumentation Commission, France Research and Development, Immunology, Hybritech, USA Director, Labs, SFBC Immunoanalysis Commission, France Direction G6n#rale, Pasteur Diagnostics, France Research and Development, Kodak, USA Research and Development, Wallac Oy, Finland Biologist Chairman of the IFCC Committee on Analytical Systems, France Biologist, SFBC Instrumentation Commission, SFBC France
68 Derni~re minute
I Tuesdayaftemoon I
Interface between solid phase and immunoreagents
1. Criteria for the selection of a solid phase to be used in immunoassays. R. Masseyeff (Labs, France) & J.F. Delagneau (R.&D., Pasteur Diagnostics, France)
2. Engineering human antibodies for use as im- munologic reagents. R. Hamilton (J. Hopkins University, MD, USA)
3. Orientation of antibodies; binding to solid phase. M. Delaage (Immunotech, France)
4. Interface between solid phase and immunoreagents. S.E. Rasmussen (R.&D., Nunc, Denmark)
5. Multi-microspot, multianalyte, immunoassay. R. Ekins (Middlesex Hospital Medical School, UK)
Conclusion J.E Delagneau & R. Masseyeff (France)
Criteria for the selection of a solid phase to be used in immunoassays. J.F. Delagneau 1 & R. Masseyeff 2 (1Diagnostics Pasteur, Marnes la Coquette, France; 2Labs, Nice, France)
Immunoassays have made a unique contribution to the development of biological analyses during the past decade. Hormone testing, therapeutic drug monitoring, rapid screening of infectious diseases are nowadays performed with high reliability owing to the selection and use of high specificity and affinity antibodies, non-isotopic labels, new solid supports or liquid phase assay formats.
Current development efforts aim at simplifying procedures, eliminating tedious steps, automating assay procedures as well as increasing antibody qualification and quality control.
Very simple assay configurations have been proposed. They aim at adding liquid im- munological reagents to a tube, followed by one mixing, and at providing final signal recording (homogeneous assays such as CEDIA, EMIT, fluorescence polarization-based assays or ag- glutination tests, conventional microparticle support based nephelemetric or particle-counting assays). These separation-free technologies generally depend on modulation of the signal which, under standard conditions, is not negli- gible. The measurement is based upon a signal variation, which considerably limits assay sen- sitivity.
On the other hand, undesired serum inter- ferences, mainly at the signal level, necessitate a significant sample predilution. Of course, no one- step or homogeneous techniques meet the requirements of ultra-sensitive assays. However, they are widely used for therapeutic drug monitoring and readily automated.
By contrast, heterogeneous phase immuno- assay techniques (RIA, ELISA, FEIA) are only limited - in terms of sensitivity - by inevitable non-specific binding of conjugates. They require separation of free from bound fractions and concomitant use of a solid phase. They are more difficult to automate. The improvement of the solid phase is crucial and still remains a problem.
Diagnostic companies have responded by a great variety of technologies aiming at different objectives (alternate site unit testing, high throughput testing, automated testing, etc.). Schematically, solid phases can be categorized into non-dispersable (such as cuvette, tube wall, surface area of any device, bead, disc, porous membrane) and dispersable (acrylamide, poly- styrene, gelatin microbeads, paramagnetic microbeads, etc.). The development of new solid phases involves a great deal of know-how and capabilities with a view to improving test performance characteristics. Amoung the tasks to be implemented are : optimization of the spatial orientation of the final coupled reagents, selection of surface-neutral hydrophilic supports, ac- celeration of reactions increasing the reactive surface area of the supports, efficiency and simplicity of procedural steps. These various aspects will be fully described and emphasized by the lecturers at the first and second session of this congress.
Engineering human antibodies for use as immunologic reagents. R. G. Hamilton (Clinical Immunology, Dermatology, Allergy and Clinical Immunology Reference Laboratory, Johns Hopkins University School of Medicine, Asthma and Allergy Center, 301 Bayview Blvd. Baltimore, MD 21224, USA)
Variable and constant region domains of antibodies are encoded by distinct genes. Molecular biology methods allow the recom- bination of these genes and their transfection into lymphocytes to engineer antibodies for use as drugs and as diagnostic assay reagents. While non-human antibodies (e.g., mouse) can be made with these recombinant DNA techniques, they are more readily prepared by hybridoma cell
Derni6re minute 69
fusion. Gene transfection techniques have had their greatest impact on the production of human antibodies with defined specificity and effector functions. Chimeric antibodies, for example, have been prepared with DNA from mouse variable regions from murine hybridomas that have been linked to DNA coding for each of the human isotypic constant regions (IgG, IgA, IgM, IgD and IgE). Alternatively, human antibodies have been constructed by inserting complementarity- determining region DNA into DNA segments that code for the human antibody V and C region framework (e.g., CDR-grafted antibody).
Engineered human antibodies have become useful : 1), as characterized human antibody standards for constructing reference curves that are used in solid phase antibody immunoassays; 2), as primary standards to define antibody concentrations in daily working "secondary" antibody pools; and 3), as immunochemical reagents to study immunoglobulin biological effector functions. This presentation will overview basic terminology and the methods employed in the engineering of chimeric and CDR-grafted "human" antibodies from cells transfected with reconstructed DNA. Second, an application of human IgE and IgG chimeric antibodies as antibody standards will be presented for the construction of dose-response curves. Finally, heterologous interpolation from these dose- response curves will be presented as an alternative standardization approach to existing solid phase elution and depletion methods for the standardization of secondary human reference sera in weight per volume units of specific immunoreactive antibody.
Orientation of antibodies : binding to solid phase. M. Delaage & P. Prince (Immunotech, Case 915, 13288 Marseille Cedex, France)
In modern immunoassays, the antibodies are most often fixed on a solid support. This technology is easy to handle; however, it introduces in the test an additional variability due to variations in : 1), the amount of immobilized antibody; 2), the accessibility of the antigen binding site. As a matter of fact, the environment of the binding site is modified as compared to the situation in solution, especially if the solid phase is dried. The immobilized antibody appears to be placed in an evolutive situation which is affected by • 1), intrinsic factors (physicochemical pro- perties of the antibody, type of linkage); and 2),
external factors (presence of additives, capillary forces during draining, kinetics of the drying process).
We have succeeded in controlling the antibody amount by using an avidin-biotin system presenting a capacity largely in excess of the needs. The problem of controlling and stabilizing the accessibility of binding sites is more difficult to resolve, and has to be investigated case by case. Some general rules will be discussed.
Interface between solid phase and immuno* reagents. S. E. Rasmussen (A/S NUNC, Kamstrupvej 90, Kamstrup, DK-4000 Roskilde, Denmark)
The performance of solid phase immuno-assays is dependent on several parameters, one of these being the immobilization of antibodies or antigens on the solid phase, either by hydrophobic interaction or bycovalent binding.
The binding strength in hydrophobic interaction is dependent on the type of solid phase, the nature of the molecules to be bound, and the binding conditions, Le. buffer type, temperature and time. Although immobilization by hydrophobic interaction has been successful in solid phase immuno-assays, there are limitations illustrated by the fact that some molecules show instability when bound by hydrophobic interactions, either because of denaturation or by desorption. Some molecules will adsorb with an unwanted orien- tation, and finally, some molecules cannot be adsorbed by hydrophobic interaction, e.g. because they are too hydrophilic or because they are too small.
These problems can be overcome by anchoring the biospecific molecules onto the solid phase by covalent binding. Different approaches have been made, in order to modify polymer materials for covalent coupling.
A new product for covalent coupling has recently been introduced to the field of bio- specific assays. This product type has a well- defined linker, covalently attached onto a poly- styrene matrix. The benefits of covalent binding using this product have been demonstrated by immobilization of small haptens which can be adsorbed to a limited degree by hydrophobic interactions.
70 Derni~re minute
Multi-microspot, multi-analyte immunoassay. R. Ekins, F. Chu, E. Biggart (Department of Molecular Endocrinology, University College and Middlesex School of Medicine Mortimer Street, London, UK)
This presentation primarily centres on the development of a multianalyte or "random access" immunoassay methodology capable of measuring tens, hundreds or thousands of substances in a small sample, e.g. a drop of blood. The possibility of this quantum leap in immunoassay methodology stems from a recognition of certain fundamental physico- chemical concepts. The first is that all immuno- assay methods rely on a measurement of the fractional occupancy of antibody binding-sites by antigen. The second is that, provided the antibody concentration used in an immunoassay system does not exceed 0.01/K (where K is the affinity constant of the antibody/antigen reaction), fractional antibody occupancy is essentially independent of : 1), the amount of antibody present; and 2), sample volume. Recognition of these principles is the key to the perception that sensitive immunoassays may be developed using a "vanishingly small" amount of antibody located on a "microspot" exposed to the analyte- containing fluid. Following such exposure, measurement of antibody occupancy by analyte (using, for example, a second antibody) reveals the analyte concentration in the sample. Furthermore, the recognition that "micro-immuno- assays" of high sensitivity are feasible leads to the concept of "multianalyte assays" based on multi-microspot arrays, each individual antibody microspot being directed against a different analyte.
Though not essential, it is advantageous to use double-label techniques in this context, one label attached to the "capture" antibody molecules situated in a microspot, the second to a "recognition" antibody, enabling identification of either occupied or unoccupied "capture-antibody" binding sites. Observation of the ratio of signals emitted by the two labels reveals capture- antibody occupancy, and hence the analyte concentration to which the microspot has been exposed. Fluorescent antibodies are especially useful since simultaneous measurement of fluorescent signals emitted by dual fluorophors constitutes well-established methodology. Current studies in the authors' laboratory rely on the use of a laser-based confocal microscope and conventional fluorescent labels; however, time resolution techniques for differentiation between fluorophors may be adopted in future studies.
With these techniques, immunoassays using microspots <50 sq I~m in area may be performed, permitting, in principle, construction of immuno- arrays capable of measuring some 2 million different substances in a small sample. Though developments of this kind clearly lie in the more distant future, the construction of limited antibody arrays for selected groups of analytes (e.g. all hormones present in blood) is more im- mediately feasible. Such technology - which is closely analogous to that underlying the compact disks used for sound recordings - is likely to completely revolutionize the immunodiagnostics field.
Wednesday afternoon I
Faster solid phase immunoassays
1. Solid phase immunoassays : biophysical approach. P. Montagne (Facult~ de M~decine, Nancy, France)
2. Dispersion of the solid phase : contribution of microparticles in solid phase immunoassays. P. Varcin (Sanofi Recherche, Montpellier, France)
3. Acceleration of solid phase immunoassays by physical processes, especially microwaves. Dr. F. Patat (CHRU Bretonneau, France)
4. Activity of antibodies immobilized on polystyrene copolymer beads. M. Bale Oenick (R.&D., Kodak, USA)
5. Use of chromium dioxide particles in an immunoassay system. D.M. Obzansky (R.&D., Du Pont de Nemours, DE, USA)
6. Rapid formats for heterogenous tests for high molecular weight analytes. E. Krodel (R.&D., Ciba Coming, MA, USA)
Solid phase immunoassays : biophysical and biochemical approach. P. Montagne (Labo- ratoire d'lmmuno/ogie, Facu/t~ de M~decine, BP 184, F 54505 Vandoeuvre-/&s-Nancy Cedex, France)
Antigen-antibody (Ag-Ab) reaction is an essential tool for today's biology but Ag-Ab interactions are often inadequately tackled by many physicists or chemists. The uncommon nature of Ag-Ab bonds, inadequacy of stoichiometry and variable composition of Ag-Ac complexes indeed complicate the Ag-Ab reactions. This complexity is still further increased
Derni~re minute 71
if Ag or Ab are bound to a solid phase, though the most highly sensitive immunoassays are based upon this technical adaptation.
The Ag-Ab bonds (hydrogen, electrostatic, Van der Waals and hydrophobic bonds) are weak. Thermodynamic values (free energy, enthalpy and entropy) associated with the formation of Ag-Ab complexes results in a large number of these involved Ag-Ab bonds. However, the dissociable and reversible nature of non covalent Ag-Ab bonds permits discussion of the Ag-Ab reaction in the usual terms of chemical equilibrium (association and dissociation constants, kinetic parameters, law of mass action, Le Chatelier principle). But the ideal Ag-Ab reaction situation (a monovalent hapten reacting with a monoclonal Ab)and the chemical standard conditions are more often than not absent. Ag valency, Ag/Ab molar ratio, surface tension of the liquid medium, amphoteric nature of proteins, pH, ionic strength, presence of chemical surfactants and temperature must be considered in order to understand Ag-Ab interactions. Some of these parameters are altered by the coating of Ag or Ab on solid phase : modification of the spatial configuration of Ag or Ab, accessibility of Ag determinants or Ab active sites, water exclusion, surface area, charge and ionic environment, dispersion and agitation of the solid phase can at the same time favorize Ag-Ab association or dissociation.
It is of utmost importance to understand the physical and chemical parameters of the Ag-Ab reaction, to take into account the modifications brought about by the solid phase, and their causes. From these theoretical considerations, it is possible to choose the most appropriate solid phase, the Ag or Ab binding conditions and the reaction medium for successful results in a solid phase immunoassay.
Dispersion of the solid phase : contribution of microparticles to immunoassays. P. Varcin (Sanofi Recherche, rue du Professeur Joseph Blayac, F 34082 Montpellier Cedex, France)
Due the low free energy of the antigen-antibody reaction and the minute quantities of reagents, direct detection of the antigen-antibody reaction is not suitable for routine conditions.
In contrast, the selectivity of the immunological processes greatly exceeds that of classical chemical reactions. Immunoassays are possible
when a highly performing detection system is coupled to an immunological system :
In the case of radioimmunoassays, the radioisotopes provide high energy quantums which are easily detectable.
Sensitivity and low detection limit of enzyme immunoassays are related to the kinetics of the dye producing reaction.
In both cases, the signal reading is the final step in multi-stage processes; solid phases are used merely to separate the free and bound fractions and are generally not involved in the signal generating system.
When used in nephelemetric detection assays, polymeric colloidal particles amplify the diffused light; moreover, the surface energy of the solid-liquid interface is a part of the energetic balance of the system.
Electric charges, hydrophilic and hydrophobic properties of the particles as well as pH and ionic strength of the liquid phase are the governing parameters.
The performances of nephelemetric assays are greatly improved with a detection limit of about 1 nmol/I and a 15-min reaction time.
Near-equilibrium reacting conditions are limiting factors.
Acceleration of solid-phase immunoassays by physical processes, especially microwaves. F. Patat, Z.Q. Wang, P. Makris, L. Boucher & L. Pourcelot (Laboratoire de Biophysique M6dicale BP 32 23, 37032 Tours Cedex, France)
Ultrasonic waves generate non-linear forces usually referred to as radiation pressure. When applying an acoustic wave of sufficient intensity on small particles such as red blood cells or latex spheres, these are moved to the nodal planes. This effect has been used to accelerate an aggregation or an agglutination process. This idea has then been applied to agglutination immuno-assays in microplates of various shapes. A kinetic model of particle movement is presented and the reduction in the image formation time is shown. The sensitivity of the technic is com- pared by reference to the gravity sedimentation assay.
The disadvantages related to the use of ultrasonic energy are discussed and compared with the benefits and disadvantages of other physical treatments.
72 DerniOre minute
Activity of antibodies immobilized on polystyrene copolymer beads. M.D.B. Oenick, S.J. Danielson, J.L. Daiss & R.C. Sutton (Eastman Kodak Company, Rochester, NY 14650-2113, USA)
Monodisperse polystyrene copolymeric latex particles offer a well-defined, high surface area substrate for immobilization of antibodies for immunological assays. Our work has focused on ways of preserving antibody activity following covalent or noncovalent immobilization. Our methods of analyzing adsorption of the specific antibody following coimmobUization with adjuvant proteins and measuring activity of the immobilized antibody towards small molecules and antigenenzyme conjugates will be discussed. Our results show that small changes in the copolymer composition, such as addition of 1% acrylamide or 5% acrylic acid, significantly influence the extent of adsorption and retention of biological activity. For some copolymers, antibody concentration during immobilization is an impor- tant factor in determining the antibody activity; similarly, coimmobilization with an adjuvant protein improves activity.
Use of chromium dioxide particles in an immunoassay system. D.M. Obzansky (E.L Dupont de Nemours & Company Inc., Wilmington, DE, USA)
Five key properties of MagniSep TM chromium dioxide particles provide the foundation for a highly versatile immunoassay technology. The combination of large surface area (40 m2/g) and high protein uptake capacity (40 mg/g) of chromium dioxide particles produce solid supports with rapid capture kinetics and high binding capacity. Magnetic and physical properties of chromium dioxide particles are optimized to yield rapid separation and complete resuspension, highly desirable characteristics for efficient automation of immunoassays. A unique coating process, that includes covalent at- tachment of haptens and antibodies, yields particles with low non-specific binding and enhanced stability. Chromium dioxide particles are adaptable to a variety of assay formats that include competitive, forward, reverse and simultaneous two-site immunoassays.
The Vista ® immunoassay system has been developed to completely automate highly sensitive and rapid heterogeneous immuno- assays using chromium dioxide particle techno-
Iogy. Selected performance characteristics for representative hormone, tumor marker and infectious disease assays are shown in the table below. No "hook effect" was observed for the HBsAg, HCG and CEA assays at 500,000 l~g/I, 1,000,000 IU/I and 25,000 l~g/I, respectively.
Method Sample Units Reporting Sensitivity W/thin-run Method "size (pl) range reproducibility comparison
CEA 50 I~g/I 0.7-100 0.7 2.9%at5 0.96x+0.36" 2.1% at 75 r = 0.960
HBsAg 175 - 0.2 I~g/I ad - Specificity 99.7 % 0.3 pg/I ay Auszyme ® II
* Comparison vs Hybritech immunoradiometric assays.
Rapid formats for heterogenous tests for high molecular weight analytes. E. Krodel, M. Kwiatkowski, J. Boland & G. Carey (Ciba Coming Diagnostics Corp., 333 Coney Street, E. Walpole, MA 02032, USA)
The development of rapid immunoassays in heterogenous formats places strict requirements on the properties of antibodies, label, and solid phase and also on the selection of engineering devices to allow rapid and precise recovery of the solid phase from the reaction mixture.
We have developed rapid sandwich immuno- assays by employing acridinium ester as the label, since it offers a detection limit of 10 attomol; and signal generation and detection occur within 5 s. A suspension of paramagnetic particles with a size distribution of 0.01-1.5 microns offers surface area of 100 m2/g for the capture of antibody. Post coupling treatments reduce nonspecific binding to very low levels (0.1%). The collection of these particles from the reaction mixture can be accomplished rapidly by employing a reaction vessel where the distance of travel to stationary magnets is minimized. When neodinium-boron magnets are used, the field gradients generated cause rapid clearance of the particles from the liquid.
We have produced 15-min assays for HCG, TSH ferritin, and CKMB by exploiting the advantage of a highly dispersed magnetic solid
Derni~re minute 73
phase and an acridinium ester label together with the implementation of robotics for rapid and precise particle recovery.
2. Target amplification for DNA analysis by the polymerase chain reaction. K. Mullis (Biotechnology Consultant, CA, USA)
3. DNA hybridization : liquide phase versus solid phase. H. Soderlund (Orion Genetic Engineering, Finland)
4. Rapid non isotopic detection of nucleic acids at subattomole levels by Q beta replicase amplification. G. Parsons (Gene-Trak, USA)
5. Transcription-based amplification system and the detection of its RNA products by a bead-based sandwich hybridization system. T. Gingeras (Sibia, CA, USA)
6. Non isotopic probe labeling : advantages and drawbacks for diagnostic specification. P. Lebacq (Bioprobe, France)
7. Use of simplified cold probing for gene detection : amplification in sexually transmitted diseases. M. Herzberg (PBS Orgenics, Israel)
Nucleic acid hybridization techniques have been used for several decades in basic research to isolate genes, to determine their structure or analyse their mechanism. The possibility of using them in clinical biology has been considered for several years. The transfer from basic to applied research can now take place due to the new technology of non-radioactive probes (cold probes).
Hybridization features the use of a probe nucleic acid molecule and of a target nucleic acid molecule. It leads to the formation of a double- stranded molecule, also called duplex, which can be demonstrated with great sensitivity. For this to be done, the probe can be tagged with a variety of labels such as a radio-isotope or a non-isotopic marker which can be detected for its fluorescence, luminescence or enzyme activity or by immuno reaction(s).
Hybridization can be performed either in a liquid solution or on a solid support. In the first case, hybridization is generally followed by a separation step which makes use of a solid phase to isolate the hybrid (duplex) formed. This separation can be achieved by biochemical means (adsorption chromatography, differential precipitation, electro- phoresis, etc.) or by immunological means (affinity chromatography, immuno-capture). In the second case, the support can be used in different ways according to the assay format employed to perform hybridization. The target DNA can be immobilized by contacting the sample to be tested with the support. Simple specimen deposit with a pipette (for low volumes), or by aspiration under vacuum or by (capillary or electrophoretic) transfer from an agarose gel. In the sandwich assay format, probe DNA is immobilized onto the support to specifically extract the target DNA from the medium under assay.
Supports to be employed for hybridization purpose can be selected from the group consisting of : nylon (N), nitrocellulose (NC) or polyvinylidene difluoride (PVDF) membranes; N+NC mixed membranes; cellulose and its derivatives; polyoside supports (Sephacryl, Sepharose, etc.), latex particles, polystyrene microplates. The coupling procedure and capacity vary according to the type of support used.
Target amplification for DNA analysis by the polymerase chain reaction. K.B. Mullis (6767 Neptune Place, ApL 4, La Jolla, CA, USA 92037)
The polymerase chain reaction (PCR) has become a standard laboratory technique. An enzymatic reaction, as simple to perform as it is satisfying to contemplate, the PCR solves two of the more universal problems in the chemistry of natural nucleic acids. It allows for the physical separation of any particular sequence of interest from its context; and then provides for an in vitro amplification of this sequence which is virtually without limit. The surprising robustness of PCR derives from its fortuitous combination of three familiar phenomena, each of which is intrinsically powerful. The first of these is the impressive ability of almost all oligodeoxynucleotides to bind tightly and specifically to their complementary nucleic acid sequences, discriminating easily between hundreds of thousands of sites. The second familiar phenomenon is illustrated by the notion that the probability for the occurrence of a compound action is the product of the individual probabilities for the occurrence of each of its
74 DerniOre minute
components. The third phenomenon embodied in the polymerase chain reaction relates to the branching structure of its propagation and the inherent robustness attached to such a form.
Consideration of the above leads to certain generalities regarding the relative utility of various protocols for carrying out the PCR. Specific conditions of time, temperatures, concentrations, etc. wil l be described, as well as sample preparation and analytical methods.
DNA hybridization : liquid phase v e r s u s solid phase. H. SSderlund (Orion Pharmaceu- tica-Biotechnology, Valimotie 7, SF00380 Hel- sinki, Finland)
In the annealing reaction between nucleic acid strands the two molecules collide with each other initiating the zipper-like base pairing reaction. The reaction kinetic is thus of second order. The concentration of the reactants may be at a femtomolar level leading to slow reactions. In practical hybridization experiments this problem is solved by adding a large excess of the probe, thus increasing the reaction rate by changing the kinetic to pseudo-first order. A problem arising from this approach is the separation of excess probe from hybridized probe. A million-fold excess of free probe should be separated from the hybrid which is a demanding biochemical task. The target nucleic acid is frequently immobilized onto a solid support prior to hybridization to allow efficient fractionation but this introduces non-specific reactions and reduced hybridization rates due to diffusion barriers. Our approach is to combine the fast reaction in solution to the fractionation power of solid phase techniques. In addition to the labeled detector probe a second probe with an affinity tag is introduced into the hybridization reaction to create a sandwich hybrid. With the aid of this capturing probe the hybrids can be conveniently separated from free probe with the required efficiency.
The method of choice depends on a number of factors such as purpose of the analysis, analyte concentration, cost-effectiveness, time limitations and label used. The possibility of automation is important when designing routine diagnostic tests.
Parsons (Gene-Trak Systems, 31 New York Avenue, Framingham, MA 01701, USA)
We have developed a sensitive, general assay for specific nucleic acids in biological samples. These tests use reversible target capture, to reduce assay background, in combination with replicatable RNA probes, to amplify the assay signal. This system is based on the RNA replicase and the small RNA molecule, MDV-1, which are produced by infection of E. coil with Q beta phage. The enzyme copies MDV-1 plus RNA to make a complementary minus strand, which the replicase also uses as a template. The doubling time in vitro at 37°C is 20 sec, so 109 copies of the RNA can be made in 10 minutes. Recombinant MDV-1 RNAs bearing a probe sequence for a particular target nucleic acid are generated by transcription of cloned cDNA templates. These MDV-1 probes are copied efficiently by Q beta replicase. To perform the assay, the recombinant probes are hybridized in chaotropic salts to RNA or DNA targets. These hybrid complexes are repeatedly captured on and released from fresh oligo dT-magnetic particles ("target cycling") via hybridization with poly dA- tailed oligonucleotides complementary to target sequences flanking the MDV probe. After the final capture, the hybrid complexes are eluted and the MDV probe is replicated with Q beta replicase. The yield of replicated MDV probe varies directly with input target level, and can easily be quantitated by addition of fluorescent intercalating dye. Known levels of the recombinant RNA probe used in each assay can be run as standards during the amplification reaction to facilitate quantitation. Our assay for HIV-1 RNA employing this approach detects <103 pol region transcripts, with a total assay time of 2.5 h.
Transcription-based amplification system and the detection of its RNA products by a bead- based sandwich hybridization system. (T. Gingeras 1, G. Davis 1, N. Riggs 1, H. Chappelle 1, K. Whitfield 1, L. DiMichele 1, D. Kwoh 1, D. Richman 2 & J. Guatelli 2 (1 Salk Institute Biotechnology/Industria/ Associates Inc., 505 Coast Blvd. S., La Jolla, CA 92037, USA; 2 Dept. of Pathology & Medicine, UCSD School of Medicine, San Diego, CA 92161, USA)
Rapid, nonisotopic detection of nucleic acids at subattomole levels by Q beta replicase amplification. C. Pritchard, J. Stefano & G.
The in vitro amplification of specific target nucleic acids (RNA or DNA) can proceed by means of an RNA transcription-based amplification system (TAS). In comparison to the PCR method of
Derni~re minute 75
amplification, which amplifies by a DNA replication mechanism, TAS achieves similar levels of amplification, primarily by an RNA transcription mechanism. Increases in the copy number of 105- to 10S-fold are routinely observed in 3 to 6 cycles of TAS. The specificity of the TAS amplification can be enhanced by the use of bead-based sandwich hybridization system (BBSHS). The single-stranded nature of the RNA product fits well with this detection formation and permits quantitative detection of the amplified product. The efficiency, reproducibility, and shortcomings of both the TAS and BBSHS technologies will be discussed.
Both the TAS and BBSHS technologies have been used to study HIV-1 in peripheral blood lymphocyte samples obtained from AIDS patients. These samples have been characterized by co-culture, p24 ELISA, PCR- and TAS- mediated hybridization assays. These data have permitted the following comparisons to be made : a) compare the results obtained by each of these assay systems; b) compare the results obtained by TAS- and PCR-mediated amplification methods as measured by several different hybridization assays; and c) using amplification- mediated hybridizations, co-culture, and p24 ELISA assays, to follow the course of HIV-1 infection over several month in patients receiving AZT drug treatment.
Non-isotopic probe labelling : advantages and drawbacks for diagnostic specifications. P. Lebacq (Bioprobe Systems, 26 bis rue Kl~ber, 93100 Montreuil-sous-Bois, France)
Nucleic acid probes represent an acute and elegant means of investigation in molecular biology. Until recently, use of these remarkable tools was limited by the need to use radioactivity for labelling nucleic acids and by the long and fastidious procedure to extract and analyse nucleic acids. Interesting procedures have been developed over the past in order to obtain non- isotopic labelling of nucleic acid probes. In addition, new techniques now allow crude extracts of target DNA to be obtained starting from clinical samples without heavy and time- consuming procedures available for further analysis with "cold" probes. Different strategies have been developed to obtain labelling of nucleic acids : direct or indirect labelling, use of antibodies to recognize specific epitopes on the probe or use of protein and ligand as
avidin-biotin or streptavidin-biotin systems for the same purpose. It is now possible to achieve correct and specific detection of minute quantities of target nucleic acids with approximately the same results as those obtained with radiolabelled probes. Applications in the microbiological as well as in the cancer or genetic disease fields are currently under assay by using non-isotopic probes. A wide description of non-isotopic labelling techniques and their advantages and drawbacks for the diagnostic specifications are described in this paper.
Use of simplified cold probing for gene detection : application to sexually transmitted diseases. I. Nur, R. Elkaim & M. Herzberg (Orgenics Ltd, Israel & PBS-Orgenics, France)
It is possible to label DNA non-radioactively either by incorporation of precursors such as biotin, or direct binding or enzymes such as alkaline phosphatase or by chemically modifying one of the bases such as by sufonation. The chemiprobe system uses such a sulfonation which is then recognized by a monoclonal antibody detecting only the modified basis in the DNA polymer.
Most of the systems of labelling, however, encounter two main problems : the sensitivitY is between 0.1 to 1 pg of target DNA; the probing format is not suitable for clinical work.
At the sensitivity level, various amplification methods such as the polymerase chain reaction (PCR), QB replicase, or "Christmas tree" procedures relieve the problem while putting a burden on specificity of the signal.
At the format level, the problems can be listed as follows : need to work with partially purified DNA (clinical samples); cost of the assay for clinical applications; absence of highly sophisti- cated trained personnel in the regular clinical environment.
It is to answer to the above problems that we developed the "reverse probing assays" in which a specific probe is attached to a solid phase while labelling is done by sulfonation on partially purified sample DNA. We adapted this technique to a simple format allowing for molecular hybridization in any environment and applied it to the detection of sexually transmitted diseases.
76 Derni~re minute
Friday morning
Improvements in detection of biospecific assays
1. Amplify or improve detection. J.P. Yvert (Ch. B~gin, France) & B. Mandrand (Biom~rieux, France)
2. Time resolved fluorescence in biospecific assays. Dr. E. Soini (Wallac Oy, Finland)
3. Bio- and chemiluminescent quantitation of enzyme label in immunoassays. G. Thorpe (Wolfson Labs, UK)
4. A bioluminescent solid phase for immunoassays and DNA probe detection. Pr. Nicolas (INSERM, Montpellier, France)
5. Electrochemical detection of immunoassays. D. Bates (Novo Biolabs, UK)
6. Detection of specific nucleic acid sequences by surface plasmon resonance. D. Pollard-Knight (R.&D., Amersham, UK)
7. Biosensors for directly measuring cell-affecting agents. W. Parce (R.&D., Molecular Devices Corp., USA)
Accurate quantification at the molecular level, by simple means, is a challenge for the new generations of biospecific assays.
In order to attain this, it is necessary to reduce the non specific background of the current assays and to further amplify the detection signal.
In immunoassays, interference reduction may take in account the use of Fab fragments from high affinity antibodies, the development of neutral and dispersed solid phase and im- provements of incubation media.
As far as nucleic acid probes are concerned, target sequence amplification widely increases the specific material keeping the other com- ponents stable.
Concerning the detection side, time resolved fluorescence reduces interference sources while measuring fluorescence decay.
Chemi- and bioluminescence which are practically insensitive to turbidity allow even homogeneous assays.
Electrochemical detection is a definite means of avoiding optical interference.
To date, whatever the detection system used, kinetic measurements are likely to be more reproducible than end point titration.
For the future, a better understanding of stereochemical interactions would probably be a key factor in improving biospecific assays.
Time-resolved fluorescence for immuno- assays and DNA probes. E. Soini (Wallace Oy, PO Box 10, SF-20101 Turku, Finland)
Fluorescent lanthanide chelates and their detection using time-resolved fluorometry for in vitro as well as for microscopic bio-specific assays are reviewed, with reference to other fluorescent dyes and labelling materials. The minimum detectable dose of labelled components and the theoretical ultimate sensitivities of immunoassays and DNA-assays are discussed and compared with experimental results. It is concluded that the time-resolved fluorometry provides the highest potential sensitivity in immuno-assays, and a sensitivity comparable to phosphorus-32 in DNA hybridiza~on assays, it is also concluded that lanthanide chelates are a potential alternative for organic fluorescent dyes in microscopy and that they are most useful in multiparameter assays. The combination of organic fluorescent dyes and lanthanide chelates provides the highest separation efficiency between two assay parameters.
Bio- and chemiluminescent quantitation of enzyme labels in immuno-assays. G. Thorpe (Wolfson Research Laboratories, Queen Eliza- beth Medical Centre, Edgbaston, Birmingham UK)
Bio- or chemiluminescent reactions can be used to sensitively quantitate all of the enzyme labels commonly used in immunoassays. Such reactions produce prolonged light emission and can be measured with a range of detectors. Horseradish peroxidase labels can be assayed using luminol and oxidant in the presence of an enhancer such as para-iodophenol (enhanced chemiluminescence), and luminol can also be employed to monitor xanthine oxidase labels. D- Luciferin-o-phosphate and dioxetane phosphate substrates provide bio- or chemiluminescent end-
Derni6re minute 77
points for alkaline phosphatase labels, and appropriate substrates for other enzymes can be synthesised. Beta-galactosidase can be assayed in this manner or in a coupled enzyme assay involving formation of NADH which is detected with the bioluminescent bacterial luciferase- oxidoreductase reaction. Glucose-6-phosphate dehydrogenase labels can also be quantitated using the latter reaction. The diverse applications of such reactions in immunoassays will be reviewed.
A bioluminescent solid phase for immunoassays and DNA probe detection. J.C. Nicolas, B. T~rouanne, P. Balaguer, A.M. Boussioux & A. Crastes de Paulet (INSERM Unit& 58, 60, rue de Navacelles, Montpellier 34090, France)
A luminescent adsorbent composed of bacterial luciferase, FMN oxidoreductase and a protein such as an antibody or streptavidin coimmobilized on Sepharose has been used to detect a label enzyme (glucose 6 phosphate dehydrogenase). The label enzyme, bound to the solid phase, produces NADH and starts a chain of enzymatic reactions leading to light emission. The dehydrogenase not bound to the solid phase produces NADH in solution which is rapidly oxidized by a scavenger system (lactate dehydrogenase plus pyruvate) and thus does not participate in light emission.
Using this solid phase, binding assays do not required separation of the excess of label, and the assay protocol is limited to the addition of sample, followed by an incubation step and after addition of reagents, light emission is measured.
The immobilization of the different enzymes participating in the bioluminescent reaction increases their stability. The confinement of intermediary reaction products in a small volume of gel results in a high efficiency of the luminescent reaction. Moreover, the effect of interfering compounds is limited by using a small volume of gel, and high concentrations of bioluminescent enzymes.
We have used this solid phase for rapid immunoassays of haptens and proteins but also for the rapid quantitation of amplification reaction catalyzed by DNA polymerase.
Most non-isotopic immunoassays require fairly sophisticated instrumentation to achieve precise quantitation and this limits their use to specialist laboratories or to simple formats which can only give semi-quantitative results. Many of these limitations can be circumvented by using and electrochemical detection system instead of the more usual colorimetric or fluorimetric methods. Electrochemical detection has many additional advantages mostly connected with the relative simplicity of amperometric measurements, the wide dynamic range and the potential for cheap instrumentation.
Enzyme labels may be detected electro- chemically through an appropriate electrogenic substrate. One of the first to be described was p- phenyl-phosphate which is hydrolyzed by the enzyme alkaline phosphatase producing the electroactive species phenol. An alternative approach is to couple the enzyme label (such as glucose oxidase) directly to an electrode via a ferrocene mediator.
More recently a two-stage enzyme-amplified electrochemical detection system has been described which permits even greater sensitivity. The enzyme label alkaline phosphatase de o phosphorylates NADP + to NAD + which then participates catalytically in an enzyme-mediated redox cycle. Alcohol dehydrogenase oxidizes ethanol and the NADH produced is reoxidized to NAD + by the enzyme diaphorase using the ferricyanide ion [Fe(CN)6]3- as the electron acceptor. In a second step, the [Fe(CN)6]4- produced is reoxidized at a carbon electrode and the integrated current is directly proportional to the amount of NAD + produced by the label and hence ultimately to the amount of analyte. Using this method, highly sensitive immunoassays for prostatic acid phosphatase and thyrotrophin have been developed which perform as well as any enzyme immunoassays using conventional colorimetric detection.
Electrochemical detection has many advan- tages over other methodologies. Most notably, perhaps, it can function in an optically opaque environment leading to the possibility of precise and sensitive quantitative measurements being carried out in simple formats such as membrane based devices.
78 Derni~re minute
Detection of specific nucleic acid sequences by surface plasmon resonance. D. Pollard- Knight, M. Downes, D. Yeung, A. McDougall, P. Heaney & P.B. Garland (Pollards Wood Laboratories, Amersham International plc, Nightingales Lane, Chalfont St-Giles, Bucks, UK)
Surface plasmon resonance (SPR) has been used to detect changes of refractive index that occur when an antigen diffuses from solution to bind to an antibody immobilized at a silver surface 1. In theory any molecular binding pair can be substituted for the antigen/antibody pair : for example, enzyme/substrate, hormone receptor/ hormone, DNA/DNA and DNA/DNA binding protein. The possible applications of SPR to molecular biology techniques are therefore being evaluated. In particular the detection of specific nucleic acid sequences by hybridisation or displacement events occurring at the silver surface. Data will be presented demonstrating : the stability of the silver surface with immobilised DNA probe in hybridisation conditions; acces- sibility of the immobilised DNA to its comple- mentary sequence for a hybridisation or displacement assays; sensitivity of SPR for nucleic acid detection.
The advantages of this unenhanced SPR technique are absence of labels, rapid assay times and convenience. The sensitivity may be increased by the use of refractive index probes.
Biosensors for directly measuring cell- affecting agents. J.W. Parce (Molecular Devices Corp., 4700 Bohannon Drive, Menlo Park, CA 94025, USA)
Cellular perfusion chambers have been constructed from the light addressable poten- tiometric sensor (LAPS) previously described (Science (1988) 240, 1182). We have used these chambers to measure the effects of a variety of agents on the metabolic rates of cells. The chambers are used in a stopped flow mode. When flow is on, samples may be introduced into the chamber. When flow is stopped, acidification of the very small volume of medium in the chamber is used to determine the metabolic rate of the cells. Using a variety of types of mammalian cells we have demonstrated the following. The triggering of cellular receptors can be determined in minutes.
Metabolic inhibition of normal human cells can be correlated with the ocular irritancy level of these compounds as determined by the Draize test in rabbits. The efficacy of chemotherapeutic agents on tumor cells exhibiting multidrug resistance can be determined in a few hours. And finally, cytopathic effects of viral infections can be detected as the infection proceeds.
1 Leidberg B., Nylander C. & Lundstrom I. (1983) Sensors and Actuators 4, 299.
