Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press Technical

The quantitation of mRNA expression dress these shortcomings. Syvanen et Comparison has the potential to be of critical value in al. (7) hybridized biotinylated PCR prod- of the a clinical setting but, to date, has been ucts to an probe and im- limited by current quantitative PCR mobilized the complex to avidin-coated EnZ nle-Linked techniques. Many techniques evaluate beads. Saiki et al. (1~ showed that a PCR the relative differences in target product could be identified specifically Olig ucleotide amounts rather than determine the copy by hybridization to an oligonucleotide Sorbent Assay number, and there are safety and time probe immobilized on a nylon mem- concerns associated with 32p labeling of brane. The same principle has been ap- to the 32 PCR products. The recently developed plied to microtiter plate technology us- fluorescein-antifluorescein-based en- ing ELOSA by Inouye and Hondo (a~ and labeled PCR/ zyme-linked oligonucleotide sorbent as- Keller et al. (9'1~ Recently it was demon- Southern Blotting say (FAF-ELOSA) technique overcomes strated that a double-stranded PCR prod- many of these difficulties. We describe a uct obtained with a biotinylated primer Technique in comparison of results obtained by the and a fluorescein-labeled primer could 32p-labeled PCR/Southern blotting tech- be detected upon immobilization in an Quantitative nique to the FAF-ELOSA. Key features of avidin-coated plate and conjugation Analysis of the FAF-ELOSA technique include quan- with a FAF-ELOSA. (~ This technique titation by nonradioactive detection has been successfully used to measure Human and Rat based on relative amounts of PCR prod- drug resistance of human immunodefi- mRNA uct and a synthesized standard target. It ciency virus type 1 (HIV-1) isolates by is reproducible from a single cDNA prep- determining relative HIV-1 copy number aration or from samples derived from in the supernatant of cultured peripheral separate preparations. The technique blood monocytes. (12~ can easily be applied to clinical samples To our knowledge, none of these Joseph A. Carcillo, ~'2 Robert for the rapid determination of many dif- studies have used the FAF-ELOSA proto- A. Parise, 2 and Marjorie ferent specific mRNA expression studies. col to quantitate mRNA, nor have they Romkes-Sparks 2,3 We are currently applying this technique compared this quantitation with stan- for the detection of CYP450 mRNA ex- dard 3zp methodology. There have been 1Department of Anesthesiology and pression in a variety of human tissues reports that 32p incorporation into the Critical Care Medicine, 2Center for from control and cancer patient popula- PCR product shows a good correlation Clinical Pharmacology, and 3Center for tions. with slot (3~ and (4~ for Environmental and Occupational Since the PCR method was first intro- the quantitation of the multidrug resis- Health and Toxicology, University of duced in 1985, (1~ the combination of re- tance gene mdr-1 expression and with in Pittsburgh, Pittsburgh, Pennsylvania verse transcription (RT) with PCR (RT- situ hybridization for progesterone re- 15261 PCR) has increasingly been utilized to ceptor mRNA. ~ In the present work we study gene expression. (2-s~ However, the compare the relative sensitivity of the use of mRNA quantitation as a clinical FAF-ELOSA with the sensitivity of the tool has been hampered by several limi- 32p-labeled PCR/Southern blotting tech- tations of currently utilized techniques, nique in quantitation of PCR products including the 32P-labeled PCR/Southern used to measure rat and human actin blotting technique. First, the 32P-labeled mRNA. Cytochrome P450 CYP2D6 and technique is considered semi-quantita- CYP2E1 mRNA expression in human he- tive rather than quantitative because of patic tissue are evaluated as an example the narrow linear range of film exposure of applying this technique to the analy- and scanning densitometry. Also, the sis of specific mRNAs of interest in clin- use of 32p precludes automated mass ical samples. analysis of clinical samples. Third, the quantitation of the target mRNA is de- pendent on the accuracy of the initial MATERIALS AND METHODS RNA quantitation and the relative effi- Primer and Standard DNA ciency of the RT. Finally, the blotting Sequences technique requires time periods that are not compatible with same day, diagno- The MacVector 4.1 (Kodak International sis-directed therapeutic interventions. Biotechnologies, Inc.) software package In addition to varying cycle number, was used to select oligonucleotide and using amplification of a control gene in standard sequences. The nucleotide se- parallel, or using a competitive template quences of the primers selected for am- as an internal control, (6) several investi- plification of [3-actin in rat (Table 1) and gators have developed biotin, fluores- [3-actin, CYP2D6, and CYP2E1 in human cein, and streptavidin techniques to ad- samples (Tables 1 and 2) are presented.

292 PCR Methods and Applications 3:292-2979 by Cold Spring Harbor Laboratory Press ISSN 1054-9803/94$5.00 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press 11|1111 Technical Tips

TABLE 1 Oligonucleotide Primers and Probes for Human and Rat J3-actin

Primer Sequence Rat capture (ract.774Fb) GAGGCTCTCTTCCAGCCTT Rat reporter (ract.827Rf) TCATGGATGCCACAGGATTCC Rat 5' primer (ract.612F) TGAGAGGGAAATCGTGCGT Rat synthetic target ATCGGCAATGAGCGGTTCCGATGCCCCGAGGCTCTCTTCCAGCCTTCC TTCCTGGGTATGGAATCCTGTGGCATCCATGA Human capture (hact.intFb) GACAGGATGCAGAAGGAGAT Human reporter (hact.1013Rf) ATGATCTTGATCTTCATTGT Human 5' primer (hact.874F) AAACTACCTTCAACTCCATC Human synthetic target CCATGTACCCTGGCATTGCCGACAGGATGCAGAAGGAGATCACTGCC CTGGCACCCAGCACAATGAAGATCAAGGATCAT

The capture probes for the PCR product tion method using an RNA isolation kit 32p End-labeling of the S' are listed, whereas the capture probes for (Stratagene, La Jolla, CA). The purity and oligonucleotide the standards are the reverse comple- quantity of the total RNA was deter- The 5' oligonucleotide was end-labeled ment. The 3' primer for human actin is a mined spectrophotometrically (Shi- with [32p]ATP with T4 polynucleotide ki- splice-junction-specific primer designed madzu UV-2101 PC UV-Vis scanning nase (U.S. Biochemical) by standard to avoid the amplification of any con- spectrophotometer). The RNA prepara- techniques. (14) taminating genomic DNA. The standard tion was treated with RNase-free DNase I for both human and rat (Stratagene Cloning Systems) to digest [3-actin were synthesized to span the last any contaminating genomic DNA. PCR Amplification of cDNA 80 bp of the expected target PCR product sequence, including the 3' primer. Bi- To 1 I~1 of the cDNA product, a final con- RT of Total RNA to cDNA otin and fluorescein 5'-end labels were centration of I x PCR buffer [50 mM KC1, incorporated during oligonucleotide Total RNA was reverse transcribed and 10 mM Tris-HCl, (pH 8.3)]; 1 mM MgC12; synthesis. The biotin-labeled primers the cDNA product amplified. Briefly, 1 0.2 mM each of dATP, dCTP, dGTP, and were HPLC purified and the fluorescein- ~g of total RNA was mixed with 0.5 p.1 of dTTP; fluorescein-labeled 3' primer (0.25 labeled, and standard oligonucleotides RNasin RNase inhibitor (40 U/p.l) I~M); unlabeled 5' primer (0.25 R,M); and were purified on NAP-5 columns (Phar- (Promega) and random hexamers (2.5 2.5 units of AmpliTaq DNA polymerase macia) (refer to Fig. 1). p.M) in a total volume of 9.5 p.1. The mix- (Perkin Elmer Cetus) was added. The fi- ture was heated to 94~ for 2 min and nal reaction volume was 100 ixl (refer to cooled on ice for 5 min. To this mixture, Fig. 1). For the 3ZP-labeled PCR/Southern Tissue Collection and Total RNA a final concentration of RT buffer (lX), blotting technique, 106 cpm of 3zp end- Isolation dithiothreitol (0.01 M), and dNTP mix (1 labeled 5' primer was also added. The so- For inactivation of RNases, all solutions mM) was added to a total volume of 20 lution was overlaid with mineral oil and were treated for at least 12 hr with O. 1% ILl. The mixture was incubated at 41~ heated at 94~ for 5 min before the ad- (by volume) aqueous diethyl pyrocar- for 15 min, and 2 ~1 of Superscript RNase dition of AmpliTaq. The rat actin cDNA bonate solution. A human liver sample H- RT enzyme (200 U/pd); (GIBCO-BRL) was amplified under the following con- and a rat heart ventricle sample were fro- was added and incubated for an addi- ditions: 94~ for I min, 57~ for I min, zen at -80~ until total RNA was iso- tional 60 min at 41~ The mixture was and 72~ for 3 min, for up to 40 cycles. lated by the guanidinium thiocyanate heated to 99~ for 5 min to inactivate The human actin cDNA was amplified phenol-chloroform single-step extrac- the RT and cooled to 4~ under similar conditions, except the an-

TABLE 20ligonucleotide Primers and Probes for Human CYP2D6 and CYP2E1

Primer Sequence CYP2D6 capture (2D6.148Fb) TGGGCAACCTGCTGCATGTG CYP2D6 reporter (2D6.170Rf) GGTCGAAGCAGTATGGTGTG CYP2D6 5' primer (2D6.ATG) ATGGGGCTAGAAGCACTGGT CYP2D6 synthetic target CCCACCAGGCCCCCTGCCACTGCCCGGGCTGGGCAACCTGCTGCATG TGGACTTCCAGAACACACCATACTGCTTCGACC CYP2E1 capture (2E1.67Fb) TGGAGGCAGGTGCACAGCAG CYP2E1 reporter (2E2.160Rf) TGGGAAGCGGGAAAGGGCCT CYP2E1 S' primer (2E1.ATG) ATGTCTGCCCTCGGAGTCAC CYP2E1 synthetic target GGCCTTCCTCCTGCTGGTGTCCATGTGGAGGCAGGTGCACAGCAGCT GGAATCTGCCCCCAGGCCCTTTCCCGCTTCCCA Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press Technical TipslllllllEE

5' 3' RNA film (XAR-2). The film was developed af- ter various exposure times and then an- reverse alyzed by scanning densitometry using transcription the Collage Analyst System (Fotodyne). 51 3' cDNA ! ...... 5 I Quantitative Analysis of the PCR amplification Fluorescein-labeled PCR Product 5' 3' with ELOSA 5'1 t"" 3' 3' 5' PCR product A standard curve was determined for 3'[~5 ' both the human and rat actin-synthe- sized 80-mer oligonucleotides with ap- ELOSA DETE(;TION propriate dilutions. Ten-microliter ali- quots of three separate fluorescein- PCR product Standard labeled PCR products from a single 51 3 I cDNA preparation were taken at 20, 22, 3' 5' 3' 5'1 I 24, 30, 35, and 40 cycles of PCR amplifi- 3' 3' 5' cation and diluted to various dilutions with deionized, double-distilled water to a total volume of 20 i~l. The PCR product was heat denatured for 10 min at 100~ LEGEND and then cooled at 4~ Twenty microli- ters of the standard dilutions, or dena- 5' primer for PCR I 1 tured sample, was aliquoted to the streptavidin-coated microtiter plate 3' primer for PCR and reporter for PCR product, also reporter for standard wells (DuPont). One hundred microli- - fluorescein labeled ters of hybridization buffer (10% forma- mide, 6x SSC, 1.2% Triton X-IO0, 0.12% capture for PCR product BSA) containing 10 nm of biotin-labeled - biotin labeled captured probe [and fluorescein-labeled reporter probe (5 nM) for the standard only] was added to each well and incu- capture for standard bated at 37~ for 1 hr. The plate was - biotin labeled washed six times with 1• DuPont plate wash buffer (DuPont). One hundred mi- croliters of antifluorescein horseradish 3 I standard = truncated complement of peroxidase (HRP) conjugate (l:200)(Du- fluorescein labeled PCR product Pont) in antifluorescein diluent buffer 3,~5, 5' (PBS, 2% Tween, 5% FCS, 0.2% casein) was added to the wells and incubated at room temperature for 15 min. The wells FIGURE 1 FAF-ELOSAusing 5'-fluorescein-labeled RT-PCR product and internal biotin capture were washed six more times and 100 i~l probe, and standard DNA with fluorescein reporter probe and biotin capture probe. of trimethylbenzidene blue (TMB) sub- strate (ScyTek Laboratories) was added and incubated for 60 min in the dark at nealing step was performed at 51~ PCR preparation were performed. Ten-micro- room temperature. The reaction was samples underwent agarose gel electro- liter aliquots of the 32p-labeled PCR stopped with TMB stop solution (ScyTek phoresis and sequencing to confirm the product were taken at 20, 22, 24, 30, 35, Laboratories), and substrate develop- amplification of a single-band product. and 40 cycles and applied to a 1% agar- ment was read on a microtiter plate Control samples in which total RNA was ose (ME Agarose, FMC Bioproducts) gel. reader (Bio-Tek automated microplate not reverse transcribed and reaction The DNA products were electrophoresed reader, model EL340) at an absorbance mixture alone were also PCR amplified in 1X TBE buffer at 100 V for 45 min. The of 450--630 nm. (ls,16) A standard curve to verify that the reaction products were gels were washed 2x15 rain in 0.5 M was generated for each PCR reaction by dependent on RT and not contaminated NaOH and 1.5 M NaC1, 2x15 min in 0.5 using a serial dilution of a synthesized with extraneous DNA. MTris, and 1.5 MNaC1 (pH 7.4), I x 5 rain single-stranded DNA oligonucleotide in 2x SSC, and transferred to a Nytran identical in sequence to the target DNA. filter (Micron Separation, Inc.) over- Semiquantitative Analysis of The concentration of the synthetic DNA night. The filters were UV-cross-linked ~zP-labeled PCR Product was determined spectrophotometrically. (UV Stratalinker, Stratagene) and then By comparing the absorbance obtained Three PCR reactions from a single cDNA autoradiographed using Kodak x-omat from a sample containing an unknown

294 PCR Methods and Applications Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press Technical Tips

amount of DNA with that obtained from the standard, the concentration of the target DNA can be determined. 0.8 .,,-,,,. o replicate 1 / [] 20 cycles /~ O 0 RESULTS -~lid 0.6 O 0 [3-Actin mRNA was detected in human Ln ,~- 2 liver and rat ventricle using the 3Zp-la- v beled PCR/Southern blot technique or the FAF-ELOSA technique. Figure 2 O 9 0.4 O O shows a representative PCR amplifica- C m t~ tion of human liver [3-actin cDNA. With both techniques, the amplification 0 W o 0.2 curves show logarithmic increases in .,Ot~ product until a plateau phase is reached /e R^2--0"996 at -24 cycles (Fig. 2A, B). The signal with i~ y=-0.05374+.98234x ELOSA could still be detected with a 1:64 ' I ' I ' 0.0 9 | ' l ' l ' dilution of the PCR product detected 0 1 2 3 2 3 4 5 with 32p (Fig. 2B). fmol target DNA ul PCR product Relative quantitation of the 32p end- FIGURE 3 Quantitation of human actin PCR product. (A) Standard curve for human actin. A labeled PCR product is possible, i.e., rel- sensitivity of 0.05 fmole was achieved. A linear curve was obtained with a correlation coefficient ative difference in the amount of target of 0.996. Two replicates are shown, indicating the small deviation among experiments. (B) DNA is quantitated rather than a deter- Following the observation that 20 and 22 cycles were in the exponential range of amplification, mination of femtomole concentration. serial dilutions of the PCR products were made and FAF-ELOSA performed. The linearity of the In contrast, direct quantitation of the quantitation is demonstrated. The number of femtomoles can then be calculated as indicated in fluoresceinated PCR product was accom- Table 3. plished by (1) identifying the cycle num- ber at which logarithmic increases were occurring (Fig. 2B), (2) performing a I : 2 dilution series with the PCR product at absorbance readings with a standard as a mean and standard deviation of the this cycle number, and analyzing absor- curve with a known femtomole amount three dilutions (Table 3). There was a bance readings at three points of signal of 80-mer actin oligonucleotide standard <15% standard deviation among these linearity (Fig. 3A), (3) comparing these (Fig. 3B), and (4) calculating the quantity three dilution measures. We attribute this variation to unavoidable pipetting errors and to potential variations in the streptavidin coating of the microtiter A B plates. The FAF-ELOSA technique allows us to detect <0.05 fmole of actin PCR 10000 10 product. 1:16 dilution In Table 4, three rat actin PCR reac- 132 dilution tions from the same cDNA source were 1:64 dilution compared for coefficient of variation to determine any differences in variance when using the FAF-ELOSA or the 32p_ labeled PCR/Southern blot technique. At W j o "" 1000 20 and 22 cycles, the variance among the O E t,'- three PCR reactions was similar with ~ .I both the two assay techniques. This sug- gests that the FAF-ELOSA adds no greater variability to the differences ob- served among these three separate PCR reactions than was apparent with the

100 9 I ' I ' I .01 9 I 9 I ' s2p/Southern blot technique. 10 20 30 40 50 20 30 40 The quantitation of PCR product by cycle no. cycle no. FAF-ELOSA from known amounts of cDNA is demonstrated in Table 5. With FIGURE 2 Quantitation of human actin PCR product. (A) Typical results obtained for the 3zp_ 400 and 40 ng of human liver cDNA, we labeled PCR/Southern blotting quantitation technique. The intensity value for each band as calculated as area under the curve was plotted versus cycle number. (B) Results obtained for three demonstrated that the absorbance read- separate dilutions of one PCR product by the FAF-ELOSAtechnique. Femtomoles of PCR-product ings of the PCR products at 22 and 24 as calculated from the standard curve are plotted versus cycle number. As shown, the plateau in cycles show the same 10:1 ratio as pre- PCR amplification occurred at a comparable point, as detected by both techniques. dicted from the ratio of starting cDNA

PCR Methods and Applications 295 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press Technical

TABLE 3 Human [3-actin Amplification fers another level of specificity, as any nonspecific PCR product that might Absorbance at 450/620 Absorbance at 450/260 have homology with two 20 bp, that are (calculated fmole) (calculated fmole) -200 bp apart is less likely to have ho- Dilution 20 cycles 22 cycles mology with a third 20 bp that is only 20 4.5 ~l (1:2 dilution of PCR product) 0.208 (4.58) 0.620 (13.40) bp from the 3' fluorescein-labeled 2.25 Vd (1:4 dilution of PCR product) 0.099 (4.28) 0.280 (12.10) primer. 1.13 ~l (1:8 dilution of PCR product) 0.065 (5.62) 0.183 (15.81) Despite the advantages of the ELOSA fmole (mean --- S.D.) 4.83 + 0.70 13.77 + 1.88 assay over the 32p technique, the need The data shown are the actual absorbance readings for each cycle, and in parentheses, the for standardization remains paramount. calculated number of total femtomoles per 100-~1 reaction for both 20 and 22 cycles. The average The valid concerns about variations in deviation is <15%, mainly attributable to inaccuracy in pipetting microliter volumes. the RT step and PCR reaction variability from tube to tube requires that internal standards be run. Furthermore, just as 32p labeling standardization is necessary material. Because the absorbance read- signing primers and probes for ELOSA. across experiments, so too does ELOSA ings are on the linear range of the stan- As a matter of clarity we use the term require standardization across microtiter dard curve, the technique described in capture probe to identify biotin-labeled plates. It is important to determine the Figures 2 and 3 allows quantitation as a oligonucleotides designed to hybridize linear range of amplification for each femtomole amount of PCR product. This to the PCR product and attach to the target mRNA of interest. Differences in accurate quantitation process required streptavidin-coated well. The term re- denaturation and reannealing, hybrid- <4 hr from the moment of PCR ampli- porter probe is used to identify fluores- ization, antibody binding, and color re- fication to the point of quantitation. The cein-labeled oligonucleotides designed action must all be considered if quanti- azp technique generally requires 2 days. to be added to the hybridization mixture tation from plate to plate or day to day is In Table 6, the quantitation of to hybridize to the PCR product and at- to be compared. The simultaneous run- CYP2D6 and CYP2E1 mRNAs in a second tach to FAF antibody, and the term ning of a known cDNA sample across all human liver sample is used as an exam- primer is used to identify any oligonu- plates over time is recommended to al- ple of applying the FAF-ELOSA tech- cleotide (unlabeled or labeled with bi- low for correction of any plate-to-plate nique to specific targets of interest. The otin or fluorescein) designed to be added variation that might occur. The use of a exponential range of amplification and to the PCR mixture to amplify and pro- synthesized oligonucleotide standard for linear range of absorbance for each P450 duce the PCR product (Fig. 1). Landgraf each PCR product should also be consid- were determined by the same approach et al. incorporated a biotin primer and a ered to further verify that observed as described above for actin. Absorbance fluorescein primer into their PCR mix- changes in experimental samples are readings, calculated concentrations, and ture and amplified a PCR product that in valid. Additional controls for fluctua- the ratio of P450 mRNA to actin mRNA at and of itself attached to the streptavidin tions in the efficiency of the RT step 22 cycles are given. Again, a 10-fold dif- plate and attached to FAF.(11) The advan- should be performed, for example, the ference in starting cDNA template was tage of this technique is the low cost of coamplification of a relatively invariant reflected as -10-fold difference in con- synthesizing multiple probes. The disad- mRNA or amplification of a known centration of PCR product in the expo- vantage of this approach is that excess amount of an unrelated template RNA nential range of amplification. biotin primer in the PCR mixture must with the target RNA. A synthetic internal be removed from the PCR product be- standard that uses the same target fore the sample is added to the well as it primer sequences but yields a product of DISCUSSION will otherwise compete for avidin-bind- different size has also been utilized as a The ELOSA is a sensitive assay for quan- ing sites with the biotin incorporated titation of mRNA through PCR amplifi- into the PCR product. Eron et al. (lz) used cation of a known mRNA sequence. a sandwich assay, designing a capture TABLE 4 Rat {]-actin Amplification Specificity is achieved by primer selec- probe and a reporter probe in addition to tion and an internal capture probe. Mul- the two primers used for the PCR reac- 32p-labeled tiple controls, such as the use of a stan- tion. However, this approach has a dis- FAF--ELOSA PCR/Southern dard sample, and amplification of either advantage of higher costs of synthesiz- Cycle coefficient of blot coefficient an internal or external control should be ing probes and primers. In this paper we number variation variation included to verify the reproducibility incorporated a fluorescein primer into 20 24.1 16.4 and accuracy of the assay. The FAF- our PCR product and designed an inter- 22 18.1 20.7 ELOSA procedure can be performed nal capture probe separate from the 5'- within a relatively short period of time unlabeled primer sequence. This capture From the same cDNA preparation, three sep- arate PCR reactions for each quantitative anal- probe was designed within 20 bp of the from the preparation of the PCR product ysis technique were performed with aliquots with the use of the enzyme-linked im- 3' fluorescein primer incorporated into taken during the exponential phase of ampli- munosorbent assay (ELISA) format, a the PCR product. This has two theoreti- fication at 20 and 22 cycles. The coefficient of commonly used clinical laboratory tech- cal advantages. First, we minimize costs variations for the FAF-ELOSA assay and the nique. and time without requiring a primer sep- 3ZP-labeled PCR/Southern blot results were Several strategies may be used in de- aration step. Second, this strategy con- similar.

296 PCR Methods and Applications Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press 1111111 Technical Tips

TABLE 5 Quantitative Estimate of Actin mRNA Concentration by FAF-ELOSA and M.M. Manak. 1990. Detection of hep- atitis B virus DNA in serum by polymerase Cycle number chain reaction amplification and microti- cDNA sample (ng) 20 22 24 30 ter sandwich hybridization. J. Clin. Micro- biol. 28: 1411-1416. 40 0.003 0.012 0.033 0.67 10. Keller, G.H., D.-P. Huang, and M.M. 400 0.025 0.132 0.339 0.897 Manak. 1991. Detection of human immu- Ratio 8.3 11.0 10.3 1.4 nodeficiency virus type 1 DNA by poly- merase chain reaction amplification and cDNA (40 and 400 ng) from a second human liver sample was PCR amplified and analyzed by the capture hybridization in microtiter wells. FAF-ELOSA technique. The exponential range of amplification is similar for both starting J. Clin. Microbiol. 29: 638--641. amounts. At the logarithmic phase of amplification, there was an -10:1 ratio of absorbance 11. Landgraf, A., B. Reckmann, and A. Pin- readings. Using a standard curve, the calculated femtornole concentration from 400 ng starting goud. 1991. Direct analysis of polymerase template was 1.43 fmole/total PCR product at 22 cycles and 3.62 fmoles/total PCR product at 24 chain reaction products using enzyme- cycles. linked imrnunosorbent assay techniques. Anal. Biochem. 198: 86-91. 12. Eron, J.J., P. Gorczyca, J.C. Kaplan, and RT control. (2) There are, however, sev- clinical tests. Future studies will deter- R.T. D'Aquila. 1992. Susceptibility testing eral limitations to these approaches as mine its accuracy, reproducibility, and by polymerase chain reaction DNA quan- well, including the requirement that sensitivity in this setting. titation: A method to measure drug resis- tance of human immunodeficiency virus both products are amplified with the type 1 isolates. Proc. Natl. Acad. Sci. same efficiency and kinetics and poten- 89: 3241-3245. REFERENCES tial differences in RNA purity and degra- 13. Park, O.-K. and K.E. Mayo. 1991. Tran- dation rates are accounted for/w) 1. Saiki, R.K., S. Scharf, F. Faloona, K.B. Mul- sient expression of progesterone receptor In conclusion, ELOSA is a sensitive as- lis, G.T. Horn, H.A. Erlich, and N. Arn- messenger RNA in ovarian granulosa cells say for the quantitation of mRNA and helm. 1985. Enzymatic amplification of after the preovulatory luteinizing hor- has no disadvantages when compared beta-globin genomic sequences and re- mone surge. Mol. Endocrinol. 5: 967-978. with the 3ZP-labeled PCR/Southern blot- striction site analysis for diagnosis of 14. Ausubel, F.M., R. Brent, R.E. Kingston, ting technique. The clear advantages of sickle cell anemia. Science 230: 1350- D.D. Moore, J.G. Seidman, J.A. Smith, and the ELOSA technique are that (1) it can 1354. K. Struhl, eds. Current protocols in molecu- 2. Wang, A.M., M.V. Doyle, and D.F. Mark. lar biology. 1991. Greene Publishing Asso- be performed in a relatively short time, 1989. Quantitation of mRNA by the poly- ciationANiley Interscience, New York. (2) it is nonradiometric, and (3) it can be merase chain reaction. Proc. Natl. Acad. 15. Conway, B., L.J. Bechtel, K.A. Adler, R.T. quantified to a known standard. The Sci. 86: 9717-9721. D'Aquila, J.C. Kaplan, and M.S. Hirsch. ELOSA assay is therefore preferred for 3. Noonan, K.E., C. Beck, T.A. Holzmayer, 1992. Comparison of spot-blot and mi- mRNA quantitation in mass samples for J.E. Chin, J.S. Wunder, I.L. Andrulis, A.F. crotitre plate methods for the detection of Gazdar, C.L. Willman, B. Griffith, D.D. HIV-1 PCR products. Mol. Cell. Probes Von Hoff, and I.B. Roninson. 1990. Quan- 6: 245-249. titative analysis of MDR1 (multidrug re- 16. Adler, K. 1993. Pers. comm. TABLE 6 Human CYP2D6 and sistance) gene expression in human tu- 17. Ferre, F. 1992. Quantitative or semi-quan- CYP2E1 Amplification mors by polymerase chain reaction. Proc. titative PCR: Reality versus myth. PCR Natl. Acad. Sci. 87: 7160-7164. Methods Applic. 2: 1-9. CYP2D6 4. Murphy, L.D., C.E. Herzog, J.B. Rudick, calculated A.T. Fojo, and S.E. Bates. 1990. Use of the cDNA absorbance amole/total polymerase chain reaction in the quanti- Received December 2, 1993; accepted in (ng) reading PCR product tation of mdr-1 gene expression. Biochem- revised form February 1, 1994. /stry 29: 10351-10356. 40 0.065 0.54 5. Innis, M.A., D.H. Gelfand, J.J. Sninsky, 400 0.518 4.56 and T.J. White, eds. PCR protocols: A guide to methods and applications. 1990. Aca- CYP2E1 demic Press, San Diego, CA. calculated 6. Kolls, J., P. Deininger, J.C. Cohen, and J. cDNA absorbance amole/total Larson. 1993. cDNA equalization for re- (ng) reading PCR product verse transcription-polymerase chain re- 40 0.024 0.40 action quantitation. Anal. Biochem. 400 0.283 4.68 208: 264-269. 7. Syvanen, A.C., M. Bengstrom, C.H. Leven- A 10-fold difference in starting cDNA tem- son, and H.A. Erlich. 1989. Quantitation plate was reflected as an -10-fold difference of polymerase chain reaction products by in concentration of PCR product after 22 cy- affinity-based hybrid collection. Proc. cles, which is in the exponential range of am- Natl. Acad. Sci. 86: 6230-6234. plification for CYP2D6, CYP2E1, and actin. 8. Inouye, S. and R. Hondo. 1990. Mi- For 400 ng of starting cDNA template, the cal- croplate hybridization of amplified viral culated ratios for P450 expression in this hu- DNA segment. J. Clin. Microbiol. man liver were 5.7 amoles CYP2D6/fmole ac- 128: 1469-1472. tin and 5.9 amoles CYP2E1/fmole actin. 9. Keller, G.H., D.-P. Huang, J.W.-K. Shih,

PCR Methods and Applications 297 Downloaded from genome.cshlp.org on October 6, 2021 - Published by Cold Spring Harbor Laboratory Press

Comparison of the enzyme-linked oligonucleotide sorbent assay to the 32P-labeled PCR/Southern blotting technique in quantitative analysis of human and rat mRNA.

J A Carcillo, A Parise and M Romkes-Sparks

Genome Res. 1994 3: 292-297

References This article cites 14 articles, 7 of which can be accessed free at: http://genome.cshlp.org/content/3/5/292.full.html#ref-list-1

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the Service top right corner of the article or click here.

To subscribe to Genome Research go to: https://genome.cshlp.org/subscriptions

Copyright © Cold Spring Harbor Laboratory Press