Gene-expression studies are vital to investigations into cellular function and mechanisms of disease development and progression. Reverse transcription quantitative PCR (qPCR) is a common approach to measure mRNA and offers an accurate and sensitive method for gene-expression analysis, accelerating our understanding of changes in gene expression across different tissues, blood types, and individual cells.
Even though qPCR is fast and easy, careful attention to experimental design is essential as numerous critical quality issues throughout the workflow can influence the validity of results. Moreover, considerable variation exists across qPCR protocols that can affect the reproducibility of an experiment from lab to lab. Thus, quality assurance and quality control, from sample preparation and RNA extraction to enzymatic steps and data analysis, must be integral to any qPCR experiment.
Impact of design
A well-designed qPCR assay for gene-expression studies must be quantitative, accurate, and reproducible, requiring meticulous assay design, setup, and optimization. Suboptimal reaction conditions, enzymatic inhibitors or underlying single-nucleotide polymorphisms (SNPs) can cause inefficient or incorrect amplification. “Even attempts to repeat published assays might not work if experimental design is not adjusted for each lab’s specific reaction conditions, wasting time and materials,” says Aurita Menezes, qPCR product manager at Integrated DNA Technologies. “Resources like the MIQE guidelines (Minimum Information for Publication of Quantitative Real-Time PCR Experiments) outline a minimum set of requirements for qPCR assays to enable replication of experiments and ensure quality reproducibility. These guidelines help solve technical deficiencies in experiments to ensure consistent assay performance.”
Considering each step of the workflow can help determine areas of variability and where controls are most important. Key steps include sample acquisition and handling, RNA extraction, RNA concentration, and quality assessment, cDNA synthesis by reverse transcription, assay optimization, qPCR to measure target gene-expression level, and data analysis with normalization. David Bishop, professor at Victoria University in Melbourne, noted in a recent publication that although qPCR is effective in quantifying gene expression and capable of detecting low levels of mRNA, quality of results can be severely impacted by variability in the workflow from poor method design and inadequate quality control.
Challenges in sample handling and preparation, RNA quality, choice of reference genes, and data normalization affect expression analysis accuracy, a key aspect of setting protocols and choosing materials used throughout the process. Frank Bizouarn, market development team manager for the Digital Biology Group at Bio-Rad Laboratories, explains that mRNA as starting material inherently brings a level of unpredictability to an experiment. RNA degrades easily, so experimental design must include careful sample-processing procedures, emphasizing reverse transcription strategy as any reverse transcriptase enzyme biases or RNA template differences can impact results.
“A second consideration,” offers Bizouarn, “is biological variability. Analyzing one sample once can indicate a certain process is occurring but doesn’t show trends or validate that process for that sample type. Running sufficient replicates to get statistically correct information verifies an observed change in expression levels.” Technical replicates confirm that the experiment is accurate but say nothing about reproducibility from a biological standpoint. Biological replicates help determine changes in expression levels and signify cell functions in different cell types, from healthy, diseased, damaged, or treated cells.
Best practices for assay robustness
“Experimental design involves knowing your gene, managing SNP positioning, and ensuring specificity,” explains Menezes. Databases and programs such as Ensembl or Clustal provide transcript variant and exon organization data on a gene of interest so an assay can be designed to best fit that gene. Evaluating SNP positioning avoids issues with previously unidentified SNPs that can affect primer and probe annealing, polymerase extension efficiency, and possibly target specificity. Lastly, ensuring that primer and probe sequences are specific to the area of interest and not complimentary to other sequences helps generate correct amplicons. Using BLAST or similar tools can confirm sequence specificity and identify areas of cross-reaction. Predesigned qPCR assays like PrimeTime Predesigned qPCR Assays from IDT incorporate these general design parameters, integrating BLAST analysis and SNP information.
Best design practices vary depending on the type of experiment and what information is needed. For instance, bulk analysis is relatively straightforward when analyzing hundreds of cells to determine average expression, where individual cell analysis is more challenging but provides clearer insights. Still, warns Bizouarn, it is important to include both positive and negative controls and proper reference genes along with a robust standard curve for qPCR. Controls help determine relative baseline levels and whether samples have contaminants or non-specific PCR amplification products that could skew results. One such contaminant is RNA itself. When Bishop and colleagues tested common mistakes in performing qPCR, they discovered that complete removal of RNA from cDNA samples is essential for obtaining accurate cDNA content used for data normalization.
The Bishop lab also found that use of unstable reference genes led to substantial differences in results. While glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and actin beta (ACTB) have historically been used as reference genes, thought to be stably present in all cells, later research revealed that their expression varied considerably under different experimental settings in a range of tissues. More recent approaches acknowledge that the stability of each reference gene must be evaluated for every assay with more than one reference gene required, and no one-size-fits-all gene can be used in all studies and with all protocols.
Robust assay design for low-abundance levels of targets is especially important simply because the more cycles you amplify the more variability comes into play. “Examining small differences in expression levels compounded with a standard curve and its standard error, normalized to reference genes that have their own variability and standard error among replicates, can make analysis very difficult at the low end of quantitation with qPCR. Digital PCR is an absolute counting assay and forgoes comparison to standard curves increasing accuracy,” says Bizouarn. “Thus, all of these parameters must be carefully optimized, whether using qPCR or higher resolution assays like dPCR.”
Best practices for assay efficiency
Of course, simpler chemistry with pre-optimized reagents and kits makes for a better assay, offering the ability to design and perform gene-expression analysis with the fewest steps possible. Yet supplementing master mixes and calibrated instruments with assay design tricks that focus on amplicon size, primer accessibility and concentration, and annealing temperature can also help increase assay efficiency. For example, amplifying 100 bases is more efficient and precise than trying to amplify 1000 bases. Or simplifying primer design by avoiding areas that create a secondary structure from DNA self-hybridization after denaturing improves primer annealing and thus polymerase extension.
“When working with limited samples or low abundance targets, using two-step RT-qPCR protocols and preamplification of RNA or first-strand cDNA can increase the amount of detectable target,” suggests Menezes. These steps can be useful when performing single-cell analysis or working with clinical samples, fine needle biopsies, microdissection samples, or FACS-generated cells.
The more we understand the biology and behavior of cells and chemistry of reactions, the more we can design experiments to generate better results and gain more significant insights into function and disease. New innovations are more recently enabling faster and easier studies of more complex samples such as solid tumors, tissues, liquid biopsy, or whole blood, examining the average abundance and cross distribution of a target. “For example, introduction of whole transcriptome analysis for isolated individual cells within these sample types generates even more information, opening the field up to greater precision and understanding what is going on at a cellular level as opposed to a tissue level. And while these experiments are more elaborate, they can still be done in a day or two because workflows are becoming more streamlined, migrating to easier processes,” says Bizouarn.