Laboratory Exercises

Introducing Undergraduate Students to Real-Time PCR

Received for publication, December 10, 2009, and in revised form, March 24, 2010; accepted 2 April 2010

Dale Hancock,‡ Alister Funnell, Briony Jack, and Jill Johnston

From the From the School of Molecular and Microbial Biosciences, University of Sydney, New South Wales, Aus-tralia

An experiment is conducted, which in four 3 h laboratory sessions, introduces third year undergraduateBiochemistry students to the technique of real-time PCR in a biological context. The model used is amurine erythroleukemia cell line (MEL cells). These continuously cycling, immature red blood cells,arrested at an early stage in erythropoiesis, can be induced to progress further through the process by72 h exposure to 1.8% DMSO. This gives a control cell sample and a DMSO-treated preparation. Stu-dents isolate RNA from both cell cultures, check its purity, yield and integrity by UV spectrophotometryand denaturing gel electrophoresis, then synthesized cDNA. The relative levels of three sequences: b glo-bin, amino levulinate synthase, and carbonic anhydrase-1 are estimated by real-time PCR, using 18SrRNA as the reference sequence. The changes in gene expression are robust and reproducible, enablingstudents to experience a ‘‘cutting edge’’ research technique in an undergraduate lab setting. While theundergraduate student experience in practical classes with such sensitive techniques is often mixed, thechanges in gene expression in this model are sufficiently great that students can gain the satisfaction ofconsistent results. In addition they gain experience at setting up checks and controls at stages through-out a multistep process and an appreciation of the difference between a reaction which has gone tocompletion with one that is measured as a rate. This experiment would also complement cell biologyprojects involving red cell development. It could also be extended to more thoroughly investigate thetechnique of real-time PCR.

Keywords: gene expression, real-time PCR, QPCR, undergraduate education.

INTRODUCING UNDERGRADUATE STUDENTS TO REAL TIME PCR

As research in molecular biology increasingly focuseson the regulation of gene expression in eukaryotic cells,techniques such as real time PCR, which enable sensi-tive measurement of the cellular steady state level ofspecific mRNA sequences, are becoming common placeanalyses [1, 2]. Courses in molecular biology need toinclude this technology both at a theoretical and practicallevel if they are to reflect current directions in the field [3,4]. To this end we present an experiment which, overfour 3 h sessions, measures changes in the steady statemRNA levels during differentiation of four sequences in acultured erythroid cell line.

The RNA source used for the experiment is the murineerythroleukemia cell or MEL cell [5, 6], a model previ-ously described for use in teaching microarray analysis[7]. This continuously cycling immature erythrocyte hasbeen arrested early in the differentiation program but canbe induced to progress further through the process by anumber of chemicals such as DMSO [8]. Culturing in

1.8% DMSO for 72 h produces cells which display ery-throid characteristics, that is, they now produce signifi-cant quantities of haemoglobin although they still retaintheir nucleus. They are particularly well suited to teachingas they are not nutritionally fastidious, they are nonadher-ent and can be grown to high concentrations without anyill-effects on the cells. The induction with chemicals suchas DMSO or sodium butyrate provides a clear differentia-tion model with large, well-characterized changes ingene expression. The up-regulation of haemoglobin canalso be monitored visually as the cells appear red aftertreatment.

Unlike microarray analysis, where changes in geneexpression are identified by examining the behaviour ofthousands of genes, real time PCR takes a candidategene approach. Certain target sequences, often identi-fied by microarray analysis, are selected and the relativeabundance of these sequences is monitored with treat-ment. In many cases the level of a particular sequence isexpressed relative to a reference gene. 18S rRNA waschosen in this experiment as the reference gene as itdoes not change with DMSO treatment, but commonly-used reference genes are glyceraldehyde 3-phosphatedehydrogenase (GAPDH) and b-actin [9]. In this model,both b-actin and GAPDH gene expression change withDMSO treatment, according to array results. Three

‡To whom correspondence should be addressed. Bldg G08,Biochemistry Building, Darlington, Sydney, New South Wales,Australia 2006. Tel.: 61-2-9351-4024; Fax: 61-2-9351-4726E-mail: d.hancock@usyd.edu.au.

This paper is available on line at http://www.bambed.org DOI 10.1002/bmb.20414309

Q 2010 by The International Union of Biochemistry and Molecular Biology BIOCHEMISTRY AND MOLECULAR BIOLOGY EDUCATION

Vol. 38, No. 5, pp. 309–316, 2010

sequences were selected to compare between the con-trol and treated cells. Two were significantly up-regulatedwith DMSO treatment: b-globin (bHb) and aminolevuli-nate synthase (ALAS), while one was down-regulated:Carbonic Anhydrase 1 (CA-1).

These sequences were chosen because they showedlarge reproducible changes (>50-fold) in gene expressionand could be linked to the process of erythroid differen-tiation and function. They were originally selected fromthe results of a set of spotted arrays performed on sam-ples from similarly-treated cells [7] in senior Biochemistrylaboratory classes. b-globin is one of the most abundantstructural proteins in mature erythocytes, comprising twoof the subunits of tetrameric adult haemoglobin. ALAS isthe rate limiting enzyme in heme synthesis, a pathwaywhich is significantly up-regulated in erythroid differentia-tion. Carbonic anhydrases catalyse the hydration of CO2,(the by-product of cellular fuel oxidation) to bicarbonate,making the CO2 more soluble in the blood and contribut-ing to the control of blood pH. There are a number ofcarbonic anhydrases, two of which are expressed in MELcells, CA-1 and CA-2. While CA-2 expression does notchange with DMSO exposure, CA-1 is significantlydown-regulated [10].

REAL TIME PCR OR Q PCR (QUANTITATIVE PCR)

Real time PCR in its various forms is a techniquedesigned to measure the steady state level of a specificmRNA sequence in an RNA sample [9, 11]. It is essen-tially a PCR, where the amount of product synthesizedafter each amplification cycle is estimated in real time.Quantifying the amount of product can be done byexploiting the binding properties a DNA binding dye,SYBR Green I. This intercalator and minor groove binderonly associates with double stranded DNA, where it fluo-resces quite strongly. Estimating the fluorescent outputafter each cycle reflects the amount of double strandedDNA present in the PCR at the time. While the detectionis specific for double stranded DNA (single strandedDNA shows very little fluorescence with SYBR green I),the use of specific forward and reverse primers to initiateeach round of PCR provides the specificity to the partic-ular gene.

Other more sophisticated probe-based detectionmethodologies are available, such as Taqman probes,which rely for their specificity on primers and a probecomplementary to a region in the middle of the ampliconof interest. This probe contains a fluorescent reporter atthe 50 end and a quencher at the 30 end. As the Taq poly-merase progressively synthesises the copy from theprimer it will encounter the 50 end of the probe and, dueto its 50 exonuclease activity, release the reporter. Onceaway from the quencher this reporter fluoresces; this sig-nal increasing with each cycle. The advantage of thesedetection systems is their specificity. The probe, onlycomplementary to a portion of the sequence of interest,ensures that the fluorescence observed truly reflects onlythe level of the amplified product. SYBR Green I,because it simply detects double stranded DNA, relieson the choice of primer sequence for specificity. Other

double stranded DNA artefacts such as primer dimerscan contribute to the overall fluorescence and confoundthe results. SYBR green I is however, cheaper and easierto design primers for and so is the method of choice forstudent laboratory classes. It also affords melting curveanalysis, which enables students to further investigatethe properties of the amplified product and also the pri-mers.

This experiment gives students experience at a‘‘cutting edge’’ technology in the field of molecular biol-ogy. But it does so much more. It provides an experi-mental environment to develop generic spreadsheetskills, strategies for managing data exported from propri-etary software and the design of controls and progres-sive checkpoints at critical points in a multistep experi-ment.

TIMETABLE

The experiment requires three 3–4 hour ‘‘wet’’ labora-tory sessions with a 4th session for results analysis.

Session 1

RNA isolation from prepared MEL cell cultures andassessment of yield, purity and integrity (UV spectropho-tometry and formaldehyde agarose gel electrophoresis).

Session 2

Complementary DNA synthesis using random hexam-ers as primers.

Session 3

Set-up of real time PCR assays.

Session 4

Results analysis and gel electrophoresis check onamplified products.

Organization

Students are divided into two groups to isolate RNAfrom either control or DMSO-treated MEL cell cultureswhich have been prepared for them before the class.Following integrity checks, isolated RNA is pooled toproduce two RNA samples: control and treated. Comple-mentary DNA is produced by all pairs of students fromone of the two pooled RNA samples. The cDNA copiedfrom each RNA sample is then pooled to produce twocDNA samples: control and treated. Each pair of stu-dents is then assigned one cDNA sample and onesequence of interest, hence set of primers, from which toset up their real time PCR assays. Duplicate assays oftwo separate dilutions of the cDNA are carried out.

RNA ISOLATION AND ASSESSMENT OF INTEGRITY (SESSION 1)

RNA is isolated by Trizol, using a variation of themethod of Chomczynski and Sacchi [12]. Essentially a

310 BAMBED, Vol. 38, No. 5, pp. 309–316, 2010

MEL cell pellet is suspended and lysed in Trizol (a solutioncontaining guanidinium isothiocyanate and phenol), centri-fuged to remove cell debris then extracted with chloro-form. The aqueous top phase of the resultant mixture, con-taining the RNA, is carefully removed and the RNA precipi-tated with isopropanol. The pellet is resuspended in waterafter washing with 75% ethanol and the yield estimated byUV spectrophotometry. Good quality RNA should have anA260: A280 of >1.9 .The integrity of the RNA is furtherassessed by formaldehyde agarose gel electrophoresis[13] followed by ethidium bromide staining and visualisa-tion using a UV transilluminator. Good quality RNApresents with two large bands at 28S (4712 nt) and 18S(1869 nt) and a further small band made up of 5.8S and 5SrRNA and tRNA (75–120 nt). The fluorescence of the larger28S band should be approximately twice the intensity ofthe smaller 18S band. There should be little smearing oth-erwise. RNA can also be checked with an Agilent Bioana-lyzer, if one is available. Figure 1 shows typical studentresults.

cDNA SYNTHESIS (SESSION 2)

Complementary DNA is then synthesized with ReverseTranscriptase (Invitrogen SuperScript III2) essentiallyaccording to the manufacturer’s protocol, using the controland DMSO-treated RNA preparations (5 lg per 40 lL reac-tion) as template and random hexamers as primers. Ran-dom hexamers were chosen over oligo (dT) as 18S rRNAwas used as the reference gene for all subsequent relativemeasurements and this RNA is not polyadenylated. All thecontrol cDNA preparations are pooled together, as are allthe DMSO-treated cDNA preparations, providing two cDNAsamples for analysis. If time and resources are available stu-dents could analyse their own cDNA preparation without

pooling. To do this each pair must, at the very least, measurethe level of the reference sequence 18S and one other geneof interest in their cDNA sample. This doubles the number ofassays which adds considerably to the reagent costs andtime on the real time instrument.

REAL TIME PCR (SESSION 3)

Each pair of students is assigned one of the four sequen-ces (18S rRNA, bHb, ALAS or CA-1) and either control ortreated cDNA. Four reactions are set up; duplicates with nodilution of the cDNA and duplicates with a 1 in 8 dilution ofthe cDNA preparation. Each reaction is set up according tothe manufacturer’s instructions, using a 20 lL total volume,300 nM for each primer and 1 lL of the cDNA (both undilutedand diluted). The amplification in our classes is performed inan ABI StepOne plus instrument, using ABI Fast SYBR1

Master Mix and a two-step thermocycler program (5 min at508C preincubation to remove any contaminating amplifiedproduct from previous PCR assays (see controls) followedby 40 cycles of 3 s at 958C then 30 s at 608C). A melt curveis performed subsequently. The data presented in the follow-ing figures (Fig. 2–7) are typical of pooled student resultsobtained in our classes. The sequences of the primers usedand the size of the amplicons produced are presented inTable I.

ANALYSIS OF DATA (SESSION 4)

The Ct (threshold cycle or cycle needed to achieve apreset threshold amount of amplified product) value isestimated for each reaction by the ABI proprietary soft-ware and then exported to Excel for student analysis.The melting temperature (Tm) is calculated for eachamplicon and both the Tm and the image of the meltcurve are exported.

FIG. 1. Quality control checks on isolated RNA. (A) shows the electrophoretic pattern of isolated RNA. RNA samples (~1 lg) areapplied to a 1% agarose gel in 2.2 M formaldehyde, 20 mM MOPS, 5 mM acetate and 1 mM EDTA. Electrophoresis is carried outat 80 V for~1 h. RNA is stained with ethidium bromide. (B) shows the UV spectrum of a typical pooled student RNA sample.

311

All Subsequent Calculations are Based on theFollowing Principle

The higher the concentration of a sequence, the fewerthe number of cycles required to achieve a preset orthreshold level of product (Ct). As each cycle representsa doubling of the amount of amplified product, mathe-matically the Ct has an inverse relationship to the log2[mRNA sequence].

An Example Calculation

Suppose there are 16 times more copies of 18S than of aparticular target sequence, A, in the control MEL cell tran-scriptome. Because the amount of product doubles witheach cycle, it will require four additional cycles (16 ¼ 24) ofthe target A PCR to make the equivalent amount of productas in the 18S reaction. So, if in a particular real time PCR runthe Ct for 18S is, say 8, then the Ct for target A would be 12.

TABLE I

Details of the primers used in this experiment

PrimerPrimer

sequencePrimerTm

a (8C)Ampliconsize (bp)

AmpliconTm

b (8C)

Fwd 18S CACGGCCGGTACAGTGAAAC 63.7 71 76Rev 18S AGAGGAGCGAGCGACCAA 62.3Fwd b-Hb ACCACCTGGGCAAGGATTTC 63.4 64 79Rev b-Hb ACTCCAGCCACCACCTTCTG 62.5Fwd ALAS CATCGTCCTGTGCACTTTGAA 62.2 63 75Rev ALAS TCCCATGTTCCCAAAGTAGGA 61.6Fwd CA1 ACCCAAGCCTGCAGAAAGTACT 61.5 85 74Rev CA1 GATGGGTCAAAATTTGTGAATGG 62.5

a The Tms for the primers are calculated from the primer design program: Primer3 [14].b The amplicon Tms are determined experimentally from the melt curve (see Fig. 4). Theoretical values based on the equation used in

Primer3 yielded similar values.

FIG. 2. Amplification plots for each of the sequences; reference gene 18S (A) and target sequences bHb (B), ALAS (C) andCA-1(D). The plots are presented as the change in reaction relative to the baseline rate, DRn, on a logarithmic scale, (y axis) versuscycle number (x axis). The treated measurements are shown in red, while the control curves are in blue, with undiluted and 1 in 8dilutions of the cDNA preparations. The y-axis values range from 0.01 to 10. The maximum value on the x-axis is 40. For the exactCt values and the difference between them, refer to Figure 3.

312 BAMBED, Vol. 38, No. 5, pp. 309–316, 2010

Student Data Processing

The duplicate Ct values exported from the ABI realtime instrument are compared for closeness (see Fig. 3,column E), then the undiluted duplicate average Ct (Fig.3, Column F) is subtracted from the 1 in 8 diluted aver-age Ct to check for quantitation (Fig. 3, Column G). A 1in 8 dilution of a cDNA preparation should give a Ct 3greater than the undiluted cDNA (8 ¼ 23). (Only thosemeasurements which fitted these criteria are used for fur-ther analysis.) Differences which vary greatly from 3could indicate impurities in the cDNA preparation, ineffi-cient PCR or poor pipetting.

DDCt Analysis

To ensure that equivalent amounts of cDNA are beingcompared between control and treated samples the tar-get gene Ct values are normalized against the referencegene, 18S. As the Ct, in effect, represents a log2 quantity,normalization is carried out as follows: normalized Ct ¼Ct (target sequence) – Ct (18S) (Fig. 3, Column H).

Once this correction has been performed, the normalizedCt values (DCt) for the treated are subtracted from the re-

spective control Ct values. By raising this difference (DDCt)to the power of 2, an estimate of the fold change betweencontrol and treated can be obtained [15]. The fold changesare shown in Figure 3, Columns J (Treated: Control) and K(Control: Treated). The formula is as follows:

Fold change (treated: control), 2DDCt ¼ 2(normalized Ct

control – normalized Ct treated).

An Example Calculation

Consider again the target sequence A. As mentionedpreviously, if there are 16 times more copies of 18S thanof target A in the control MEL cell transcriptome, thenthe normalized Ct for target A would be 4. Now target Ais up-regulated with DMSO treatment in MEL cells. TheCt for the treated target A is 11 and the Ct for treated18S is 9, giving a normalized Ct for target A of 2 (11-9).How much up-regulation has occurred with treatment?Intuitively it is a fourfold increase with DMSO treatment.Mathematically it is the difference between the normal-ized Ct for the control target A and the normalized Ct forthe treated target A, raised to the power of 2 or(expressed mathematically) ¼ 2(4–2) ¼ 22 ¼ 4.

FIG. 3. Data from a MEL cell DMSO Induction Experiment. Columns (A)–(E) show the format of the data as exported to Excel.Columns (F)–(K) have been added to display further analysis. These values are calculated in Excel, although the proprietary soft-ware can do these calculations. The control data are shaded gray for each sequence.

313

These data can be expressed simply as fold change(Fig. 4A) or converted to number of copies per 100,00018S (Fig. 4B). To express the normalized Ct as the num-ber of copies per 100,000 18S, consider again the con-trol target sequence A with a normalized Ct of 4. Thismeans that target A mRNA is present at 1/16th the num-ber of copies as 18S or that there are 16 copies of 18Sper copy of target A. Converting this to number of copiesper 100,000 18S, there will be 100,000/16 or 6,250 cop-ies, that is for every 100,000 copies of 18S, there will be6,250 copies of the target sequence. Once up-regulated,there will be 25,000 copies of target A per 100,000copies, of 18S. The general formula is: Number of copiesper 100,000 18S ¼ 100,000/2(normalized Ct).

The products of the amplification reactions can be fur-ther analysed. Melt curves of the amplified products areprovided for the students from the software (Fig. 5). Theproducts are also separated by agarose gel electropho-resis, using a~3% gel (Fig. 6).

COMPARISON OF TRADITIONAL PCR WITH REAL TIME PCR

The concept of the difference between measuring therate of a reaction (real time PCR) and estimating theamount of product after a reaction has gone to comple-tion (traditional or end point PCR) can be well illustratedby comparing the real time PCR data (see Figs. 2 and 3)with the amplified product agarose gel (see Fig. 6), De-spite the very large changes in gene expression for somesequences with DMSO treatment, the amount of ampli-fied product, as judged by the intensity of the product onthe agarose gel, at the completion of the PCR (after 40cycles) is normally very similar. The purely qualitative na-ture of traditional PCR is particularly noteworthy whencomparing lanes 8 and 9 and lanes 5 and 6. Lane 8(treated bHb) has a >50 fold increase in gene expressionover lane 9 (control bHb), according to the real timeamplification plots, yet the intensity of the bands on thegel are similar. Likewise lane 5 (treated ALAS) and lane 6(Control ALAS) show very similar intensities despite largedifferences in the Ct values as judged by real time analy-sis. It is interesting to note that the forerunner of realtime PCR, known as semi-quantitative PCR, relied onquantifying the amount of amplified product after an

appropriate number of cycles, when the amplificationwas still in the linear phase that is before the amplifica-tion curves ‘‘saturated’’ [16].

THE CONTROLS

A ‘‘no template’’ control (NTC) PCR is set up for eachsequence. The reaction mixture in each case containsMaster mix, primers and water, but no template DNA.The NTC is an indicator of contamination and Ct valuesof >30 are considered acceptable (see Figure 7). The18S NTC Ct is typically lower than the other sequencesdue to its abundance, both in the genome and the tran-scriptome, and sequence homology with other species,thus any contamination is detectable. To minimise thecontamination from previous PCR assays, dUTP is sup-plied instead of TTP in real time PCR amplifications.This has the effect of producing PCR products withdUMP incorporated rather than TMP. Template cDNAsynthesized by reverse transcriptase, however, usesTTP in its polymerisation. The real time Master mix con-tains the enzyme uracil DNA glycosylase (UDG orampErase) which destroys residual products from previ-ous real time PCR assays. It requires a short incubationat 508C before the amplification cycles. This enzyme isdestroyed as the temperature is increased to 958C. Theuse of this enzyme reduces false positives and ensuresthat the NTC produces very little product. To check forgenomic DNA contamination, a ‘‘no reverse transcrip-tase’’ control can be included in the reverse transcrip-tase step also.

STUDENT ASSESSMENT

Students write a brief report presenting all gel scans,example amplification plots and melt curves, withlegends, and a summary of class results. They must alsocomplete a set of questions and comment on the role ofvarious steps and components in the technique. A copyof the raw data and analysis is included in student labo-ratory notebooks. A fuller description of the in-semesterreport task is described in the on-line resources. Sum-

mative assessment on this topic is carried out in the

end-of-semester examination. Raw data are provided as

Ct values for an experiment similar to the one performed

FIG. 4. Alternative graphical presentations of the results. (A) shows the results expressed as fold change (treated: control) foreach target sequence while (B) presents the number of copies of each target sequence per 100,000 copies of rRNA. For bHb andALAS the control levels are too low to show up on the graph as is the number of copies of CA-1 for the treated sample.

314 BAMBED, Vol. 38, No. 5, pp. 309–316, 2010

in class, that is a cultured cell line treated with a com-

pound which alters the gene expression of the cell. Stu-

dents must identify the up- and down-regulated sequen-

ces, quantify the magnitude of these changes, drawsome conclusions and comment on the purpose of someof the steps in the methodology. A sample of thisassessment is provided with the on-line resources.

FURTHER EXTENSIONS

There are a number of additional aspects that could beincluded in this experiment. Depending on the emphasisof the course, students could investigate the real timePCR technique more thoroughly. Using the model andthe sequences described here comparisons could bemade of the various detection methods, such as Taqman

probes, Amplifluor probes, LUX Fluorogenic probe setsand SYBR green I. The construction and use of a stand-ard curve (dilutions of a known amount of template)could also be included, thereby giving students experi-ence of absolute quantitation. To provide students with athorough grounding in the design and planning aspectsof the technique a ‘‘dry’’ primer design exercise, usingweb-based software, could be added. There are evenmathematical simulations of real time PCR [16] whichcould be introduced for advanced courses.

Alternatively this experiment could form part of aninvestigation into methodologies for measuring geneexpression regulation. In this context other approachesto monitoring gene expression could also be explored,such as microarrays and proteomic analysis. Other tech-niques employed to measure the expression of specific

FIG. 5. Analysis of the amplified products. Melt curves for each of the four sequences: (A) 18S, (B) bHb, (C) ALAS, (D) CA-1.The melt curve of amplified products derived from control cDNA are shown in blue while the melt curve for the amplified productsfrom DMSO treated cDNA are in red.

315

sequences, such as Northern blot analysis or RNase Pro-tection Assays could also be compared.

For courses primarily focusing on cell biology, theinclusion of this experiment would provide a moleculartechnique to accompany microscopic and observationalinvestigations. It would be well suited to studies onerythropoiesis or more general developmental cellulargene expression and differentiation.

CONCLUSIONS

The experiment presented here gives senior under-graduate students experience of a ‘‘cutting edge’’ tech-nique which is extensively used in molecular biologyresearch laboratories. By the time these students havereached the third year of their science degree, they have

covered the required background biochemistry and mo-lecular and cellular biology, and are well placed to appre-ciate the context of the model, as well as the chemicalprinciples behind the techniques. The model used is asimple one involving the familiar process of differentiationof an erythroid cell line. The design of the experiment issuch that students carry out the real time PCR in its en-tirety, giving them experience in and understanding of acomplex technique containing many steps. The use ofcontrols and checks on steps of the procedure is dem-onstrated throughout the experiment.

REFERENCES

[1] S.A. Eaton, A.P. Funnell, N. Sue, H. Nicholas, R.C. Pearson, M.Crossley (2008) A network of Kruppel-like Factors (Klfs). Klf8 isrepressed by Klf3 and activated by Klf1 in vivo, J. Biol. Chem. 283,26937–26947.

[2] A.P. Funnell, C.A. Maloney, L.J. Thompson, J. Keys, M. Tallack,A.C. Perkins, M. Crossley (2007) Erythroid Kruppel-like factordirectly activates the basic Kruppel-like factor gene in erythroidcells, Mol. Cell Biol. 27, 2777–2790.

[3] Real Time PCR. Available athttp://pathmicro.med.sc.edu/pcr/real-time-home.htm (revised 2009).

[4] J. Logan, K. Edwards, N. Saunders, Eds. (2009) Real-time PCR:Current Technology and Applications, Caister Academic, Norfolk,UK.

[5] P.A. Marks, Z.X. Chen, J. Banks, R.A. Rifkind (1983) Erythroleuke-mia-cells—Variants inducible for hemoglobin-synthesis withoutcommitment to terminal cell-division, Proc. Natl. Acad. Sci. USA80, 2281–2284.

[6] P.A. Marks, M. Sheffery, R. Ramsay, K. Ikeda, R.A. Rifkind (1987)Induction of transformed-cells to terminal differentiation, Ann. N YAcad. Sci. 511, 246–255.

[7] D. Hancock, L.L. Nguyen, G.S. Denyer, J.M. Johnston (2006) Micro-arrays for undergraduate classes, Biochem. Mol. Biol. Educ. 34,432–437.

[8] R.A. Rifkind, M. Sheffery, H.R. Profousjuchelka, R.C. Reuben, P.A.Marks(1983) Induction of globin gene-expression during erythroidcell-differentiation, Ann. N Y Acad. Sci. 411, 141–149.

[9] S.A. Bustin (2000) Absolute quantification of mRNA using real-timereverse transcription polymerase chain reaction assays, J. Mol.Endocrinol. 25, 169–193.

[10] J.L. Villeval, U. Testa, G. Vinci, H. Tonthat, A. Bettaieb, M. Titeux, P.Cramer, L. Edelman, H. Rochant, J. Breton-Gorius, W. Vainchenker.(1985) Carbonic anhydrase I is an early specific marker of normalhuman erythroid differentiation, Blood 66, 1162–1170.

[11] S.A. Bustin (2002) Quantification of mRNA using real-time reversetranscription PCR (RT-PCR): Trends and problems, J. Mol. Endocri-nol. 29, 23–39.

[12] P. Chomczynski, N. Sacchi(1987) Single-step method of RNA isola-tion by acid guanidinium thiocyanate phenol chloroform extraction,Anal. Biochem. 162, 156–159.

[13] H. Lehrach, D. Diamond, J.M. Wozney, H. Boedtker (1977) RNAmolecular-weight determinations by gel-electrophoresis underdenaturing conditions, a critical re-examination, Biochemistry 16,4743–4751.

[14] S. Rozen, H. J. Skaletsky, in S. Krawetz, S. Misener Eds.(2000) Bio-informatics Methods and Protocols: Methods in Molecular Biology,Humana Press, Totowa, NJ, pp. 365–386.

[15] D. A. Favy, S. Lafarge, P. Rio, C. Vissac, Y. J. Bignon, D. Bernard-Gallon, (2000) Real-time PCR quantification of full-length and exon11 spliced BRCA1 transcripts in human breast cancer cell lines,Biochem. Biophys. Res. Commun. 274, 73–78.

[16] W. Liu, D.A. Saint (2002) Validation of a quantitative method for realtime PCR kinetics, Biochem. Biophys. Res. Commun. 294, 347–353.

FIG. 7. Typical NTC results. Equal volumes of the reactionmixture in wells E1, E3, E5, and E7 have been separated byagarose gel electrophoresis (see Fig. 6).

FIG. 6. Amplification products separated by agarose gelelectrophoresis. Electrophoresis using a 3% agarose gel wascarried out in TAE buffer at 80 V for 30 min. Products werevisualised with a transilluminator, using the intrinsic fluorescenceof the bound SYBR green I. All products loaded on this gel arederived from undiluted cDNA template, both Control andTreated or no template control (NTC see later). Equal volumesof each PCR assay were loaded. Lane 1, NTC CA-1; lane2, Treated CA-1; lane 3, Control CA-1, lane 4, NTC ALAS; lane5, Treated ALAS; lane 6, Control ALAS, lane 7, NTC bHb;lane 8, Treated bHb; lane 9, Control bHb; lane 10, NTC 18S;lane 11, Treated 18S and lane 12, Control 18S.

316 BAMBED, Vol. 38, No. 5, pp. 309–316, 2010