JOURNAL OF CLINICAL MICROBIOLOGY, May 1989, p. 1002-10070095-1137/89/051002-06$02.00/0Copyright © 1989, American Society for Microbiology

Comparison of Colorimetric, Fluorescent, and EnzymaticAmplification Substrate Systems in an Enzyme Immunoassay for

Detection of DNA-RNA HybridsFRANCOIS COUTLEE, RAPHAEL P. VISCIDI, AND ROBERT H. YOLKEN*

Department of Pediatrics, Eudowvood Division of Infechtios Diseases, The Johns Hopkins University School of Medicine,Baltimore, Maryland 21205

Received 4 November 1988/Accepted 23 January 1989

The monoclonal antibody solution hybridization assay is a novel enzyme immunoassay for detection of RNAwith a biotinylated DNA probe. To increase the sensitivity of this test, a fluorescent substrate and an enzymaticamplification cycling system were compared with a conventional colorigenic substrate for alkaline phosphatase.The fluorescent, cycling, and colorigenic substrates detected, respectively, 10, 10, and 100 amol of unboundalkaline phosphatase in 2 h. With a prolonged incubation period of 16.6 h, the conventional substrate measured10 amol of the enzyme. In the immunoassay for RNA detection, the fluorescence and cycling assays were fasterthan that using the colorigenic substrate and reached an endpoint sensitivity of 3.2 pg/ml (0.16 pg per assay)of cRNA. However, longer incubation periods (16.6 h) for optimal generation of the colorigenic product led toa comparable level of sensitivity for the conventional substrate.

Solid-phase enzyme immunoassays (EIA) have beenwidely applied for rapid and sensitive detection of microbialantigens in body fluids (30, 31). We have described ahomogeneous hybridization assay which uses the EIA for-mat for detection of RNA viruses (27; F. Coutlee, R.Yolken, and R. Viscidi, submitted for publication; C. New-man, J. F. Modlin, R. H. Yolken, M. Bowman, and R.Viscidi, Program Abstr. 27th Intersci. Conf. Antimicrob.Agents Chemother., abstr. no. 490, 1987; R. Viscidi andR. H. Yolken, Abstr. VII Int. Congr. Virol., abstr. no.R.30.5, 1987). As for all EIA, the overall sensitivity of thetest is determined by the kinetics of the antigen-antibodyinteraction, by the detectability of bound labeled immuno-complexes, and by the level of nonspecific reactivity of theantibody (background noise) (1, 21, 32).

In conventional EIA, the presence of a bound immunore-agent is magnified by conversion by a single enzyme mole-cule (here, alkaline phosphatase) of a large number ofsubstrate molecules, leading to a detectable colored com-pound. The lowest detectable concentration of' the endproduct determines the detection limit of an EIA (1, 36).Efficient enzymatic labels generating products, such as flu-orescent (3, 4, 7-10, 12, 14, 25, 35), chemiluminescent (15),or radioactive (11) compounds, that are measurable at lowerconcentrations than visibly colored products could providean alternative to increase the sensitivity of EIA. Enzymati-cally amplified cycling reaction represents another methodof magnifying the signal produced by enzymatic reactions(16, 17, 19, 28). In the most extensively studied cycling assay(5, 6, 13, 22, 24, 26), each bound enzyme does not directlydegrade a substrate to a colored compound but rather to a

coenzyme which amplifies the signal by activation of cou-

pled enzymatic reactions involving NAD-NADH recycling.This substrate system has been reported to be more sensitiveand faster than conventional assays (26).The purpose of this study was to compare the sensitivities

of colorimetric, fluorescence, and cycling systems in themonoclonal antibody solution phase hybridization assay for

* Corresponding author.

detection of RNA. The most sensitive fluorescent substrate,methylumbelliferone (9), and a cycling assay based on NAD-NADH redox cycles were used here.

MATERIALS AND METHODSMaterials. Polyclonal goat anti-biotin antibody and p-

nitrophenylphosphate (p-NPP) were purchased from SigmaChemical Co., St. Louis, Mo. The fluorogenic substrate4-methylumbelliferyl phosphate was purchased from Re-search Organics Inc., Cleveland, Ohio. The enzyme-linkedimmunosorbent enzyme amplification assay (catalog no.9589SA) was kindly donated by Bethesda Research Labora-tories, Inc., Gaithersburg, Md. The nick translation kit andbio-11-dUTP were purchased from Bethesda Research Lab-oratories. The F(ab') fragment of a mouse monoclonalantibody to DNA-RNA hybrids and labeled with alkalinephosphatase was kindly provided by Robert J. Carrico,Ames division, Miles Laboratories, Inc., Elkhart, Ind. Plas-mid pSP65 and unconjugated alkaline phosphatase wereobtained from Boehringer Mannheim Biochemicals, India-napolis, Ind. The Riboprobe system for RNA transcription,RNasin RNase inhibitor, SP6 RNA polymerase, and DNaseRQ came from Promega. 5-Diethylpyrocarbonate-treatedwater was produced by autoclaving deionized water with0.1% 5-diethylpyrocarbonate.

Production of biotinylated probes by a nick translationreaction. Nick-translated probes were prepared with bio-11-dUTP by a standard protocol (23) supplied by the manu-facturer. The reaction was performed at 15°C for 90 min.Unincorporated nucleotides were separated from the bioti-nylated probe by sodium acetate (final concentration, 0.3M)-95% ethanol precipitation at -70°C for 2 h (19). Aftercentrifugation at 10,000 x g for 15 min at 40C, the pellet waswashed with 70% ethanol. The sediment was air dried andsuspended in 100 pd of Tris hydrochloride (50 mM; pH7.2)-EDTA (2 mM).

Transcription reaction for production of target RNA. Sin-gle-stranded RNA targets were produced from plasmidpSP65 DNA. One microgram of pSP65 template was tran-scribed with SP6 polymerase in a 100-pul reaction volume by

1002

Vol. 27, No. 5

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

EIA FOR DETECTION OF DNA-RNA HYBRIDS 1003

a standard protocol (20). RNasin was added at 1 U/,ul toprotect single-stranded RNA. After 1 h of transcription, theDNA template was removed with addition of 1 U of RQDNase for 15 min at 37°C. The RNA produced was extractedwith an equal volume of phenol, followed by extraction inchloroform-isoamyl alcohol (24:1) (18). It was recovered byprecipitation in sodium acetate-cold ethanol at -70°C asdescribed above. The RNA pellet was suspended in 5-diethylpryocarbonate-treated water with 0.5%s sodium dode-cyl sulfate as an RNase inhibitor. The amount of RNA wasmeasured in a spectrophotometer, and its purity was evalu-ated by measuring the A6(o/A_8() ratio. A ratio of >1.95 wasaccepted as an RNA preparation of high purity.

Hybridization solution assay. The enzyme immunoassayfor detection of RNA with biotinylated DNA probes (mono-clonal antibody solution hybridization assay) was performedas described previously (27). Briefly, 100 ptI of half-logdilutions of pSP65 RNA transcripts in 5-diethylpyrocarbon-ate-treated HO with 0.5% sodium dodecyl sulfate weremixed with 100 jil of a biotinylated pSP65 DNA probe at aconcentration of 0.4 p.g/ml in a hybridization buffer contain-ing 4x SSC (lx SSC is 150 mM NaCI plus 15 mM sodiumcitrate at pH 7.0), 40 mM HEPES (N-2-hydroxyethylpiper-azine-N'-2-ethanesulfonic acid) buffer (pH 7.4), and 4 mMEDTA. The nucleic acid mixtures were boiled for 3 min andhybridized at 75°C for 16.6 h. Ten microliters of 10%, TritonX-100 was then added. After the samples were vortexed, 50pI ofthe hybridized nucleic acids per well was distributed ona black Microfluor B 96-well U-microplate (no. 011-010-7201) for the fluorescent substrate and on clear polyvinylchloride microtiter plates (no. 001-010-2401) for colorimetricsystems. Both plate types came from Dynatech Laborato-ries, Inc., Chantilly, Va. Black polystyrene plates were usedto minimize nonspecific fluorescence. Black and clear plateshad been coated overnight at 4°C with 50 ,uI of polyclonalgoat anti-biotin antibody at a concentration of 1 p.g/ml in 0.06M carbonate buffer (pH 9.6). Before the plates were used,they were washed six times with a solution of phosphate-buffered saline-0.05% Tween 20 (PBST). Following additionof the samples to triplicate wells (50 p-I per well) for eachsubstrate system, the plates were left at 37°C for 2 h. Afteranother wash with PBST, 50 pI of an alkaline phosphatase-labeled F(ab') fragment of a monoclonal antibody to DNA-RNA hybrids was added to each well. The antibody wasdiluted to a concentration of 0.05 p-g/ml in a solution ofPBST-0.5% gelatin containing 0.5%î, normal mouse serum.After 2 h of incubation at 37°C, the plates were treated asdetailed below. For the noncycyling substrates, the plateswere washed six times with PBST to separate bound fromunbound conjugate, and 50 ,ul of the enzyme substrate wasadded to each well. For the fluorogenic substrate, 50 ,ul of0.1 mM 4-methylumbelliferyl phosphate per well in 50 mMdiethanolamine buffer (pH 9.6)-10 mM MgCl, was added andincubated at room temperature. The amount of fluorescentmethylumbelliferone generated by enzymatic hydrolysis ofthe substrate was measured periodically for up to 16.6 h in aDynatech Microfluor microtiter plate fluorometer (detectionwavelength, 365 nm; emission wavelength, 450 nm). For thecolorimetric substrate, 50 p-l of 1 mM p-NPP (prepared onthe day of use) per well in 50 mM diethanolamine buffer (pH9.6)-10 mM MgCl, was added and left at room temperature.The amount of nitrophenol generated by the enzymaticreaction was monitored periodically for up to 16.6 h on aspectrophotometer (Titertek Multiscan MCC/340 MK II) at awavelength of 405 nm.For the samples studied with the enzymatic amplification

system, the final wash after the conjugate incubation wasdone six times with 50 mM Tris-hydrochloride (pH 7.4)-0.15M NaCI-0.1% sodium azide. This washing buffer was pre-pared daily to minimize the risk of phosphatase contamina-tion due to microbial growth. The standard washing solution(PBST) was not used, since the presence of phosphates caninhibit the enzymatic amplification reaction.

Enzyme-linked immunosorbent enzyme cofactor amplifica-tion system. All reagents were kept at 4°C and brought toroom temperature before use. The substrate was reconsti-tuted by addition of 12 mi of substrate diluent directly into avial containing a fixed amount of lyophilized NADP. Thissolution was gently mixed until complete dissolution of thesubstrate. Fifty microliters of the reconstituted substratewas then added to each well and incubated for 20 min. Theamplifier solution was reconstituted by dissolving the lyoph-ilized enzymes (alcohol dehydrogenase and diaphorase) in 12ml of amplifier diluent, and 50 pul was added to the subsCatesolution in each well without removing the substrate. Assuggested in the instructions provided by the compdh'y,incubation times for substrate and amplifier steps were keptequal and the A495 was measured in a spectrophotometer. Allplates were blanked automatically by the reader against anantibody-coated well with no sample.

For all assays, a result was considered positive if the meanfluorescence or mean absorbance value exceeded the meanactivity of the blank (DNA probe reactivity without RNA)plus 3 standard deviations. This threshold for positivity isillustrated on the graphs as a dark inverted triangle (detec-tion cutoff). The values of fluorescence or absorbance plot-ted on the graphs are expressed as averages of four valuesfrom two independent experiments for each sample.

RESULTS AND DISCUSSION

The enzymatic cycling amplification assay (Fig. 1) is basedon the conversion by bound alkaline phosphatase of thephosphorylated form of NADP to free NAD. The NAD thenserves as an enzyme cofactor for two enzymes whichcatalyze oxidation-reduction reactions. Alcohol dehydroge-nase concomitantly transforms the NAD to NADH and alsooxidizes ethanol into acetaldehyde. The cycle is completedwith the oxidation of NADH back to NAD by diaphorase,which simultaneously reduces iodonitrotetrazolium violet topurple formazan. This dye is quantitated by measurement ofits A495. The selection of a redox cycle strictly specific forNAD-NADH in the presence of high concentrations ofNADP is necessary for initial effective detection of alkalinephosphatase without subsequent removal of NADP. Theenzymatic cycle can be completed at least 25,000 times per h(16).The colorigenic, fluorogenic, and enzymatically amplified

substrate systems were compared for the ability to detecthalf-log dilutions of unconjugated alkaline phosphatase. Forevaluation of the enzyme concentration, a molecular weightof 140,000 for alkaline phosphatase was assumed. Twenty-five-microliter volumes of serial dilutions of alkaline phos-phatase were added (in quadriplicate for each substrate), towells of uncoated black and clear microtiter plates. Twenty-fivç microliters of each of the following substrates was thenadded per well: 1 mM p-NPP prepared in diethanolaminebuffer; 0.1 mM 4-methylumbelliferyl phosphate in diethanol-amine buffer; amplifier substrate (NADP) for 60 min, fol-lowed by the amplifier reagent (alcohol dehydrogenase anddiaphorase) for an equal duration of time. With the fluores-cent and cycling substrates, 10-18 mol of the enzyme was

VOL. 27, 1989

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

1004 COUTLEE ET AL.

TETRAZOLIUM SALT

DNA/RNA -15.0 -15.5 -16.0 -16.5 -17.0 -17.5 -18.0 -18.5 blank

moles of AP, loglObiotin biotUn biotin

FIG. 1. Principles of enzymatic amplification based on cycling ofNAD-NADH. Biotinylated DNA-RNA hybrids are bound to amicrotiter plate by an antibody to biotin. DNA-RNA hybrids aredetected by an antibody to DNA-RNA labeled with alkaline phos-phatase (alk. phos.). Alkaline phosphatase activity is measured witha cycling assay involving initial dephosphorylation of NADP toNAD and initiation of redox cycles with alcohol dehydrogenase anddiaphorase. A colored formazan dye measured at 492 nm is gener-ated by reduction of iodonitrotetrazolium.

detected after 120 min (Fig. 2). In agreement with previouspublications (12, 13, 26, 30), the p-NPP substrate was lesssensitive, detecting only 3.2 x 10-17 mol of alkaline phos-phatase in 120 min. However, extension of the incubationtime up to 16.6 h (Fig. 3) allowed the colorigenic substrate toreach a detection level similar to that of the other substratesystems (3.2 x 10-18 mol of the enzyme). The fluorogenicsubstrate incubated overnight provided the most sensitivesubstrate, detecting 3.2 x 10-'9 mol of alkaline phosphatase.The monoclonal antibody solution hybridization assay,

described in more detail elsewhere (27; Coutlee et al.,submitted; Newman et al., 27th ICAAC; Viscidi and Yolken,Abstr. VII Int. Congr. Virol.), combines the sensitivity andspecificity of nucleic acid probes with the convenience of anEIA. In this assay, a biotinylated DNA probe is hybridizedin solution with complementary RNA sequences. Biotin-labeled DNA-RNA hybrids are then treated as antigens andare first captured on a solid phase coated with an antibody tobiotin. Following removal of unbound nucleic acids, analkaline phosphatase-labeled monoclonal antibody to DNA-RNA (2, 29) is added to react with bound DNA-RNAhybrids. After a washing step, the substrate is added and themeasurement of the product generated by enzymatic degra-dation of the substrate allows quantitation of biotin-labeledhybrids immobilized on the solid phase. In this study, weused a model system in which biotinylated plasmid pSP65DNA was hybridized to complementary single-strandedpSP65 RNA transcripts. The influences of different substratesystems on the ability of the assay to detect biotinylatedDNA-RNA hybrids as antigens were investigated.The influence of incubation time on substrate and amplifier

reactions was determined first for the enzymatic amplifica-tion procedure. Incubation times of 20, 40, and 120 min for

FIG. 2. Measurement of alkaline phosphatase (AP) with colori-genic, fluorogenic, and cycling substrates. Fifty-microliter volumesof half-log dilutions of unconjugated alkaline phosphatase wereadded in quadriplicate to wells of an uncoated microtiter plate. Anequal volume of a substrate was added. Symbols: +, 1 mM p-NPPin diethanolamine buffer (pH 9.8, colorigenic substrate); a, 0.1 mMmethylumbelliferyl phosphate in diethanolamine buffer (pH 9.8;fluorogenic substrate); x, the substrate reagent (NADP) followed bythe amplifier reagents (alcohol dehydrogenase and diaphorase) forthe cycling assay substrate. After incubation for 120 min at 23°C. theamount of the end product was measured (at 405 nm for nitrophenol,at 492 nm for formazan dye, and on a fluorometer for methylumbel-liferone). These quantities are expressed as percent activity of theenzyme detected, 100% representing the fluorescence or absorbancevalue of the first dilution of the enzyme. The data are the means offour values from two experiments. Detection cutoff (V), meanactivity of the blank plus 3 standard deviations. *, Results superim-posed.

the complete cycling reaction (10, 20, and 60 min for eachreaction) were compared for the ability to detect half-logdilutions of RNA from 1,000 to 0.3 pg/ml (50 to 0.015 pg perassay). The results (Fig. 4) demonstrated that prolongedincubation led to higher optical densities for each RNAdilution. The 40- and 120-min incubations reached the sameendpoint sensitivity of 3.2 pg/ml (0.16 pg per well). How-

-15.0 -15.5 -16.0 -16.5 -17.0 -17.5 -18.0 -18.5 blank

moles of AP, loglOFIG. 3. Influence of substrate incubation time on detection of

unconjugated alkaline phosphatase (AP). The procedure was asdetailed in the legend to Fig. 2, except for extension of theincubation time to 16.6 h for p-NPP (+) and methylumbelliferylphosphate (-) and 120 min for the cycling assay (*). Y, Detectioncutoff.

J. CLIN. MICROBIOL.

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

EIA FOR DETECTION OF DNA-RNA HYBRIDS 1005

OD units, 100 logiO2.5r

2.0V 2.0V

1.5

1.0

0.51-

1.5V

1.0

0.5

_._

3 2.5 2.0 1.5 1.0 0.5 0.0 -0.5 probe

RNA concentration, logiO lpg/ml)FIG. 4. Comparison of incubation times with the enzymatic

amplification assay system for detection of RNA in the monoclonalantibody solution hybridization assay. The reactivities of biotiny-lated DNA-RNA hybrids with the monoclonal antibody to DNA-RNA with various amplification incubation times are shown. Aftersolution hybridization, biotinylated DNA-RNA hybrids were cap-

tured on a plate and reacted with anti-DNA-RNA antibody labeledwith alkaline phosphatase. Equal incubation times for the substrateand amplifier steps of 10 (M), 20 (+). and 60 (*) min were used toquantitate the bound label. Absorbance values were measured at 492nm. The data are the means of four values from two experiments.Detection cutoff (V), mean reactivity of DNA probe without RNAplus 3 standard deviations; OD, optical density.

3.0 2.5 2.0 1.5 1.0 0.5 0.0 probe

RNA concentration, loglO ( pg/ml)FIG. 5. Colorimetric substrates for detection of DNA-RNA hy-

brids in the monoclonal antibody solution hybridization assay.

Half-log dilutions of RNA were hybridized in solution with a

biotinylated DNA probe. Hybrids were sampled on a microtiterplate coated with anti-biotin antibody and detected with anti-DNA-RNA antibody. The bound conjugate was quantitated byaddition of various substrates. p-NPP (1 mM) in diethanolaminebuffer was incubated for 2 h (W) or 16 h (+), and the cycling assay

was used with an overall incubation time of 40 min (*). The data are

the means of four values from two experiments. Detection cut-off(Y). mean reactivity of DNA probe plus 3 standard deviations; OD,optical density.

ever, a total reaction time of 20 min was less sensitive, withan endpoint of 10 pg/ml (0.5 pg per assay). The greatestsensitivity was achieved when incubation times for theamplification and substrate steps were identical (data notshown).

Dilutions ofRNA were tested in parallel in the monoclonalantibody solution hybridization assay using the differentsubstrate systems. The titration curves and sensitivity end-points for RNA detection with the colorigenic substrate andenzymatic amplification system are shown in Fig. 5, and theresults with the fluorogenic substrate are presented in Fig. 6.With the conventional colorigenic substrate, 32 pg of single-stranded RNA per ml was detected after a substrate incuba-tion time of 120 min. The sensitivity of the colorimetricsubstrate improved with overnight incubation (16.6 h),reaching the optimal level of detection of 3.2 pg/ml. How-ever, after incubation for only 20 min, the fluorogenicsubstrate reached the optimum detection limit of 3.2 pg/ml ofRNA. The sensitivity of the fluorescence assay was notimproved by prolonged incubation times of up to 16.6 h (datanot shown). Thus, with optimal incubation times both thecolorigenic and fluorescence assays attained identical end-points. Under optimal conditions, the enzymatic amplifica-tion assay reached the same RNA detection endpoint of 3.2pg/ml in 40 min of substrate incubation.

Publications that advocate the use of more powerfulsubstrate systems to improve EIA sensitivity report experi-ments in which hydrolysis of the colorigenic substrate was

performed for short periods (5, 12, 13, 24, 26, 35). Ourexperiments support this conclusion when an incubationtime of 120 min is used. The colorigenic substrate was then10-fold less sensitive than the fluorogenic substrate. Underoptimal conditions, colorigenic substrates reached sensitiv-ity endpoints comparable to those of fluorogenic substrates(33, 34). As demonstrated in our study, the only advantage offluorogenic substrates resides in the rapidity with which the

assay can be performed. Cycling assays can increase by 250times the absorbance values of enzyme-substrate systems(13). Practical applications of this technique resulted in a 20-to 70-fold increase in sensitivity over assays with colorimet-ric substrates (5, 13, 26). However, not unlike the experi-ments comparing fluorogenic and colorigenic substrates, thetime allowed for hydrolysis of the colorigenic product was

always limited to 30 to 60 min in these reports. In ourexperiments, allowing degradation of the substrate to pro-ceed for longer periods increased the sensitivity of theconventional assay to levels achieved with the enzymaticamplification substrate. In contrast to previous publications,an enzymatic amplification system was not more sensitivethan a conventional assay but only provided faster results.

Fluorescence units, log O

2

3.0 2.5 2.0 1.5 I.0 0.5 0.0 probe

RNA concentration, loglO ( pg/ml)FIG. 6. Fluorogenic substrate for the monoclonal antibody solu-

tion hybridization assay. The experiment was as described in thelegend to Fig. 5, except for the substrate used: 0.1 mM methylum-belliferyl phosphate in diethanolamine buffer incubated for 20 min.

2.5QD units, 10 loglO

VOL. 27, 1989

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

1006 COUTLEE ET AL.

Other factors besides the detectability of the end productaffect the sensitivity of the monoclonal antibody solutionhybridization assay. An RNA concentration of 3.2 pg/mlprobably represents the minimal concentration of nucleicacid which can bind specifically to the microtiter plate. Forexample, amplification of the signal from nonspecific reac-tions can raise the background noise without increasing theendpoint sensitivity. In our experiments, nonspecific reac-tivity was reduced to a minimum by use of the F(ab')fragment of the antibody, which eliminated nonspecificinteractions with the Fc portion of the immunoglobulin.The enzymatic amplification assay studied here was re-

vealed to be sensitive, simple, reliable, and rapid. Theprincipal advantage of fluorescence and cycling assays is toreduce the time required to perform the assay to reachmaximal sensitivity. The availability and widespread use of amicrotiter plate colorimeter for measurement of colored endproducts of cycling assays is an advantage over fluorescentsubstrates. The cycling systems products can be measuredon a standard spectrophotometer, while fluorescent sub-strates require the use of a fluorometer not readily availableto all laboratories. The monoclonal antibody solution hybrid-ization assay completed by cycling reactions represents asimple and adaptable technique for nonisotopic detection ofRNA.

ACKNOWLEDGMENTS

This research was supported by Public Health Service grant AI00625-01 from the National Institute of Allergy and InfectiousDiseases and a Medical Research Council fellowship to F.C.

LITERATURE CITED

1. Belanger, L. 1978. Alternative approaches to enzyme immu-noassays. Scand. J. Immunol. 8(Suppl. 7):33-41.

2. Boguslawski, S. J., D. J. Smith, M. A. Michalak, K. E. Michel-son, C. O. Yehle, W. L. Patterson, and R. J. Carrico. 1986.Characterization of monoclonal antibody to DNA-RNA and itsapplication to immunodetection of hybrids. J. Immunol. Meth-ods 89:123-130.

3. Burd, J. F., R. J. Carrico, M. C. Fetter, R. T. Buckler, R. D.Johnson, R. C. Boguslawski, and J. E. Christner. 1977. Specificprotein-binding reactions monitored by enzymatic hydrolysis ofligand-fluorescent dye conjugates. Anal. Biochem. 77:56-67.

4. Burgett, M. W., S. J. Fairfield, and J. F. Monthony. 1977. Asolid phase fluorescent immunoassay for the quantitation of theC4 component of human complement. J. Immunol. Methods16:211-219.

5. Carr, R. I., M. Mansour, D. Sadi, H. James, and J. V. Jones.1987. A substrate amplification system for enzyme-linked im-munoassays. J. Immunol. Methods 98:201-208.

6. Clayton, A. L., C. Roberts, M. Godley, J. F. Best, and S. M.Chantler. 1986. Herpes simplex virus detection by ELISA:effect of enzyme amplification, nature of lesion sampled andspecimen treatment. J. Med. Virol. 20:89-97.

7. Curry, R. E., H. Heitzman, D. H. Riege, R. V. Sweet, and M. G.Simonsen. 1979. A systems approach to fluorescent immunoas-says: general principles and representative applications. Clin.Chem. 25:1591-1595.

8. Forsman, R. W., R. C. McCarthy, H. Markowitz, and J. F.O'Brien. 1980. A rapid fluorescent enzyme assay of solid-phaseantibody bound PAP. Clin. Chem. 26:1028.

9. Guilbault, G. G. 1968. Use of enzymes in analytical chemistry.Anal. Chem. 40:459-470.

10. Guilbault, G. G., P. J. Brignac, and M. Juneau. 1968. New

substrates for the fluorometric determination of oxidative en-zymes. Anal. Chem. 40:1256-1263.

11. Harris, C. C., R. H. Yolken, H. Krokan, and I. C. Hsu. 1979.Ultrasensitive enzymatic radioimmunoassay: application to de-tection of cholera toxin and rotavirus. Proc. NatI. Acad. Sci.USA 76:5336-5339.

12. Ishikawa, E., and K. Kato. 1978. Ultrasensitive enzyme immu-noassay. Scand. J. Immunol. 8:(Suppl.7):43-55.

13. Johannsson, A., C. J. Stanley, and C. H. Self. 1985. A fast highlysensitive colorimetric enzyme immunoassay system demon-strating benefits of enzyme amplification in clinical chemistry.Clin. Chem. Acta 148:119-124.

14. Koninj, A. M., R. Levy, G. Link, and C. Hershko. 1982. A rapidand sensitive ELISA for serum ferritin employing a fluorogenicsubstrate. J. Immunol. Methods 54:297-307.

15. Konishi, E., S. Iwasa, K. Kondo, and M. Hori. 1980. Chemilu-minescence-linked immunoassay for detection of mumps virusantibodies. J. Clin. Microbiol. 12:140-143.

16. Lowry, O. H. 1980. Amplification by enzymatic cycling. Mpol.Cell. Biochem. 32:135-146.

17. Lowry, O. H., J. V. Passonneau, D. W. Schulz, and M. K. Rock.1961. The measurement of pyridine nucleotide by enzymaticcycling. J. Biol. Chem. 236:2746-2755.

18. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecularcloning: a laboratory manual. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

19. Mansson, M. O., P. O. Larsson, and K. Mosbach. 1979. Recy-cling by a second enzyme of NAD covalently bound to alcoholdehydrogenase. FEBS Lett. 98:309-313.

20. Melton, D. A., P. A. Krieg, M. R. Rebagliati, T. Maniatis, K.Zinn, and M. R. Green. 1984. Efficient in vitro synthesis ofbiologically active RNA and RNA hybridization probes fromplasmid containing a bacteriophage SP6 promoter. NucleicAcids Res. 12:7035-7056.

21. Pesce, A. J., D. J. Ford, and M. A. Gaizutis. 1978. Qualitativeand quantitative aspects of immunoassays. Scand. J. Immunol.8(Suppl. 7):1-6.

22. Rasmussen, H. N., and J. R. Nielsen. 1972. Simple and sensitivephotometric recycling assay for nicotinamide-adenine dinucle-otide. Anal. Biochem. 50:640-647.

23. Rigby, W. J., M. Diechmann, C. Rhodes, and P. Berg. 1977.Labeling deoxyribonucleic acid to high specific activity in vitroby nick translation with DNA polymerase I. J. Mol. Biol.113:237-251.

24. Self, C. H. 1985. Enzyme amplification-a general methodapplied to provide an immunoassisted assay for placental alka-line phosphatase. J. Immunol. Methods 76:389-393.

25. Shalev, A., A. H. Greenberg, and P. J. McAlpine. 1980. Detec-tion of attograms of antigen by high-sensitivity enzyme-linkedimmunoabsorbent assay (HS-Elisa) using a fluorogenic sub-strate. J. Immunol. Methods 38:125-139.

26. Stanley, C. J., A. Johannsson, and C. H. Self. 1985. Enzymeamplification can enhance both the speed and the sensitivity ofimmunoassays. J. Immunol. Methods 83:89-95.

27. Viscidi, R. P., C. O'Meara, H. Farzadegan, and R. Yolken. 1988.Monoclonal antibody solution hybridization assay for detectionof human immunodeficiency virus nucleic acids. J. Clin. Micro-biol. 27:120-125.

28. Woodiey, C. L., and N. K. Gupta. 1971. New enzyme cyclingmethod for determination of oxidized and reduced nicotinamideadenine dinucleotide. Anal. Biochem. 43:341-348.

29. Yehle, C. O., W. L. Patterson, S. J. Boguslawski, J. P. Albarella,K. F. Yip, and R. J. Carrico. 1987. A solution hybridizationassay for ribosomal RNA from bacteria using biotinylated DNAprobes and enzyme-labeled antibody to DNA:RNA. Mol. Cell.Probes 1:177-193.

30. Yolken, R. H. 1980. Enzyme-linked immunosorbent assay(ELISA): a practical tool for rapid diagnosis of viruses and otherinfectious agents. Yale J. Biol. Med. 53:85-92.

31. Yolken, R. H. 1981. Enzymatic analysis for rapid detection ofmicrobial infection in human body fluids: an overview. Clin.Chem. 27:1490-1498.

32. Yolken, R. H. 1982. Enzyme immunoassays for the detection of

J. CLIN. MICROBIOL.

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from

EIA FOR DETECTION OF DNA-RNA HYBRIDS

infectious antigens in body fluids: current limitations and futureprospects. Rev. Infect. Dis. 4:35-61.

33. Yolken, R. H. 1985. Enzyme immunoassays using fluorescentsubstrates, p. 401-407. In K.-O. Habermehl (ed.). Rapid meth-ods and automation in microbiology and immunology. Spring-ler-Verlag, KG, Berlin.

34. Yolken, R. H., and F. J. Leister. 1982. Comparison of fluores-cent and colorimetric substrates for enzyme immunoassays. J.Clin. Microbiol. 15:757-760.

35. Yolken, R. H., and P. J. Stopa. 1979. Enzyme-linked fluores-cence assay: ultransensitive solid phase assay for detection ofhuman rotavirus. J. Clin. Microbiol. 10:317-320.

VoL. 27, 1989 1007

on April 22, 2021 by guest

http://jcm.asm

.org/D

ownloaded from