Reverse transcription-quantitative PCR (RT-qPCR) without the need … – Nature.com

Description of the proposed method

Our proposed method, schematically depicted in Fig.1, takes advantage of using a modified primer (Modified Specific Primer, PSM) during the reverse transcription step of the protocol. Such a primer is specific for the RNA molecules to be quantified and its nucleotide sequence is designed to lack a perfect homology to the retro-transcribed template DNA. Generally, it is enough to add few mismatches with respect to the original sequence, preferably located in the close proximity to the 3'-OH terminal region. These modifications make the primer partially complementary to the target sequence but still able to hybridize at the temperatures of 3742C used during the reverse transcription step. Nevertheless, the PSM will dissociate from the partially homologous genomic DNA sequence during the PCR step, once the operating temperature reaches around 60C. The aim of using such conveniently modified specific primer is to achieve amplification specifically from cDNA template while successfully avoiding genomic DNA targets. The correct number of modifications to be applied, their effectiveness and proper discriminating temperatures should be experimentally tested for each and every transcript to be analyzed, by selecting those parameters that show negative and positive amplification tendencies towards DNA and cDNA targets, respectively. This optimization phase represents a preliminary step of our method that enables the setup of negative and positive controls and, advantageously, has to be carried out only once, since it always remains valid for a specific amplicon and can be applied to a varying number of replicates under different experimental conditions. Indeed, in current protocols the negative control (NC: RT, without reverse transcriptase) should ideally be prepared for each new sample to be tested, even though the target is the same, due to the random effectiveness of DNase I treatment. Using a PSM we are able to generate cDNA slightly different from its genomic DNA counterpart, due to the nucleotide mismatches present in the sequence.

Schematic illustration of the new method. (A) Basic model of nucleic acid metabolism from DNA to cDNA. Integration of modified specific primer into cDNA by means of reverse transcription makes it a permanent part of the sequence. (B) Amplification of target sequence by means of polymerase chain reaction. cDNA converted by modified specific primer is properly amplified at certain discriminating temperature, while genomic DNA targets are successfully avoided.

During the phase following reverse transcription (Fig.1B), the amplification of cDNA by PCR takes place using the same modified primer (PSM) from the previous step in addition to the unmodified specific primers (SP) starting from the opposite direction. Consequently, the resulting amplicon is a copy of the cDNA and not the DNA, due to the specifically selective annealing temperatures usually ranging from 55C to 62C. Therefore, with this procedure, there is no need to eliminate the co-purified DNA from the RNA sample since it is no longer a competing target and will not affect the final result of the assay. Indeed, in certain experimental conditions it could be useful and advantageous to have both DNA and RNA present together in the same sample if, for example, the results need to be normalized with respect to the gene copy number variation.

Our proposed new method can be utilized in various experimental investigations and for the purposes of this paper, it has been tested by analyzing three bacterial E. coli genes: ssb, sulA and recA (Figs.2, 3 and 4), and two satellite DNA transcripts: human alpha-satellite (ASAT) (Figs.5 and 6, and Suppl. Fig.2) and TCAST1 satellite from Tribolium castaneum (Suppl. Fig.1).

Transcription of ssb gene in exponentially growing E. coli cells harbouring ssb overexpression plasmid pID2 obtained by dPCR using current (A) and new method (B). Columns represent number of copies/l and the plotted error bar shows whether or not the events differ with 95% Poisson confidence interval.

Transcription of recA gene in exponentially growing E. coli cells obtained by dPCR using current and new method. Columns represent number of copies/l and the plotted error bar shows whether or not the events differ with 95% Poisson confidence interval.

Transcription of sulA gene in exponentially growing E. coli cells obtained by dPCR using current and new method. Columns represent number of copies/l and the plotted error bar shows whether or not the events differ with 95% Poisson confidence interval.

Delta Rn vs Cycle plot of alpha satellite DNA isolated from HeLa cells obtained by qPCR using current method (A) and new method (B).+RT and RT represent positive and negative controls, with and without reverse transcription, respectively.

Transcription level of alpha satellite DNA obtained by qPCR using current method (A) and new method (B). Columns show average of 2 different loaded samples in qPCR experiments performed in triplicate. N0 represents normalized average N0 value for alpha satellite. C represent alpha samples with reverse transcription and NC represents negative controls without reverse transcription and M is 100bp size marker.

Bacterial genes are a good experimental model to test our method because they do not contain introns in their coding region, removing the possibility of discriminating between transcripts and the DNA according to their different sizes. Hence, the technique could be applied to test the expression of all genes organized with a short or null intron (e.g. viral genes).

The bacterial strain used in this test was transformed with multicopy plasmid carrying a cloned ssb gene9, which could compete for amplification with ssb-cDNA during the transcripts quantification by PCR, unless additional DNase I treatments were implemented. The results indicated in Fig.2 show a large difference (more than 40-fold) in ssb transcription levels measured by our method, as compared to the currently used method. This really high level of amplified ssb sequence in the latter approach, when reverse transcription was not carried out, and the DNA was eliminated in both RNA isolation and RT steps (Fig.2A), is likely due to low efficiency of elimination of covalently closed circular plasmid DNA, meaning that it is false (i.e. it does not accurately represent the process of transcription) and is actually caused by DNA contamination.

This is likely a reason for all the observed cases of high levels of ssb sequence amplification using classical primers (Fig.2A). In contrast, ssb sequence was amplified by our new method only in those cases when reverse transcription was performed, i.e. when cDNA was created (Fig.2B). The level of ssb sequence amplification did not depend on DNA elimination (Fig.2B), thus confirming insensibility of our method to the presence of genomic (and plasmid) DNA. Next, we quantified expression of recA and sulA genes, which are present as single copies in the E. coli genome. In accord with the previous assay, no recA sequence amplification was observed using our method unless cDNA was created by reverse transcription (Fig.3). The level of recA sequence amplification was, again, independent from genomic DNA elimination from the sample (Fig.3). Conversely, the current method, which uses standard primers, showed a false positive signal even when reverse transcription step was skipped and the genomic DNA was (obviously incompletely) eliminated by DNase I treatment (Fig.3).

Finally, analysis of sulA gene expression using a modified primer was in accord with the previous assays since amplification of sulA sequence occurred only after reverse transcription, i.e. it was specific for cDNA (Fig.4). Accordingly, no effect was observed after genomic DNA elimination (Fig.4). In contrast, amplification of sulA sequence using standard primers was very different, and was not abolished even in situations where genomic DNA was eliminated and reverse transcription was not performed (Fig.4); theoretically, the RT/+DNase I sample should not contain any cDNA or genomic DNA.

The presented results clearly demonstrate that our method of using a modified primer during cDNA synthesis produces a cDNA-specific PCR signal, which is independent of genomic DNA, and therefore much more accurately quantifies gene expression when compared to the standard, commonly used method, which, unfortunately, does not produce real negative control since there is always possibility to have contaminating DNA in the sample.

Satellite DNA represents one of the best target candidates for demonstrating the effectiveness of our methodology since it is a highly repetitive non-coding genomic DNA, ever-present in large quantities in the sample and therefore difficult, if not impossible, to remove during RNA purification.

Alpha satellite DNA is the most abundant human satellite DNA of 171bp long, comprising up to 10% of the genome14. Figure5A, shows qPCR results obtained by following the current standard protocol (old method) which implies the elimination of DNA both during the RNA purification and reverse transcription phase. In spite of that, alpha satellite DNA continues to persist in the negative control samples ( RT). Furthermore, since it is not organized into exons and introns, satellite DNA cannot be discriminated from satellite cDNA based on its length; therefore, even a slightest trace of DNA contamination often produces false-positive results. The new method, however, successfully demonstrated the disappearance of the alpha satellite DNA contamination from the qPCR amplification results (Fig.5B, RT), as it can be clearly seen also by loading the amplicons on agarose gel (Suppl. Fig.2): ASAT amplicon of 126bp long is present only in+RT samples (C: controls) respect to RT samples (NC: negative control). The same results could be represented as in Fig.6A (current method) and Fig.6B (new method), where N0 value is the starting concentration of amplicon in the sample and columns show average of 2 different loaded samples in qPCR experiments performed in triplicate (see Materials & method section).

The highly abundant satellite DNA TCAST1 has previously been characterized as the major satellite that makes up to 30% of the beetle Tribolium castaneum genome, organizing the centromeric as well as pericentromeric regions of all 20 chromosomes10,13. Again, using the new method only cDNA was amplified (+RT samples) and almost nothing of genomic DNA contamination was detected in RT samples (Suppl. Fig.1). The results clearly show they are exactly the same as those obtained for human alpha satellite DNA.

Here is the original post:
Reverse transcription-quantitative PCR (RT-qPCR) without the need ... - Nature.com