Quantitative PCR primer design affects quantification of dsRNA‐mediated gene knockdown

Abstract RNA interference (RNAi) is a powerful tool for studying functions of candidate genes in both model and nonmodel organisms and a promising technique for therapeutic applications. Successful application of this technique relies on the accuracy and reliability of methods used to quantify gene knockdown. With the limitation in the availability of antibodies for detecting proteins, quantitative PCR (qPCR) remains the preferred method for quantifying target gene knockdown after dsRNA treatment. We evaluated how qPCR primer binding site and target gene expression levels affect quantification of intact mRNA transcripts following dsRNA‐mediated RNAi. The use of primer pairs targeting the mRNA sequence within the dsRNA target region failed to reveal a significant decrease in target mRNA transcripts for genes with low expression levels, but not for a highly expressed gene. By contrast, significant knockdown was detected in all cases with primer pairs targeting the mRNA sequence extending beyond the dsRNA target region, regardless of the expression levels of the target gene. Our results suggest that at least for genes with low expression levels, quantifying the efficiency of dsRNA‐mediated RNAi with primers amplifying sequences completely contained in the dsRNA target region should be avoided due to the risk of false‐negative results. Instead, primer pairs extending beyond the dsRNA target region of the mRNA transcript sequences should be used for accurate and reliable quantification of silencing efficiency.


Abstract
RNA interference (RNAi) is a powerful tool for studying functions of candidate genes in both model and nonmodel organisms and a promising technique for therapeutic applications. Successful application of this technique relies on the accuracy and reliability of methods used to quantify gene knockdown. With the limitation in the availability of antibodies for detecting proteins, quantitative PCR (qPCR) remains the preferred method for quantifying target gene knockdown after dsRNA treatment.
We evaluated how qPCR primer binding site and target gene expression levels affect quantification of intact mRNA transcripts following dsRNA-mediated RNAi. The use of primer pairs targeting the mRNA sequence within the dsRNA target region failed to reveal a significant decrease in target mRNA transcripts for genes with low expression levels, but not for a highly expressed gene. By contrast, significant knockdown was detected in all cases with primer pairs targeting the mRNA sequence extending beyond the dsRNA target region, regardless of the expression levels of the target gene. Our results suggest that at least for genes with low expression levels, quantifying the efficiency of dsRNA-mediated RNAi with primers amplifying sequences completely contained in the dsRNA target region should be avoided due to the risk of false-negative results. Instead, primer pairs extending beyond the dsRNA target region of the mRNA transcript sequences should be used for accurate and reliable quantification of silencing efficiency.

K E Y W O R D S
double-stranded RNA, gene knockdown, Primer design, RNAi, RT-qPCR antisense strands can recruit enzymes that methylate the histone proteins leading to the formation of a silenced chromatin, thereby inhibiting transcription (Novina & Sharp, 2004).
Naturally, RNAi regulates development and physiology, suppresses transposon activity, and provides defense against RNA virus infections in many organisms using endogenously expressed microR-NAs or exogenously introduced viral dsRNA (Ambros, 2004;Obbard, Gordon, Buck, & Jiggins, 2009). However, this mechanism can also be exploited artificially to study functions of endogenous eukaryotic genes of interest through the introduction of synthetic dsRNA or siRNA molecules that trigger the host's natural RNAi machinery to silence the respective genes, which allows investigation into their specific functions. The artificial injection of dsRNA for a gene encoding the myofilament protein into the nematode Caenorhabditis elegans led to the discovery of dsRNA-mediated RNAi silencing (Fire et al., 1998). Following its discovery, dsRNA-mediated RNAi became a powerful research tool for understanding gene functions as well as a promising therapeutic candidate for the management of genetic disorders (Agrawal et al., 2003;Seyhan, 2011).
The success and extent of RNAi silencing differs between hosts, life stages, and genes of the same organism. This may be due to variability in the stability of dsRNA molecules in vivo, their uptake by target cells, and in vivo amplification and transmission of the silencing signal between cells to facilitate systemic or transgenerational silencing (Scott et al., 2013;Wang et al., 2016). In addition to these endogenous challenges, successful use or application of RNAi can be influenced by external technical factors, including the methods used for administering the dsRNA and measuring its success (Herbert, Coppieters, Lasham, Cao, & Reid, 2011;Holmes, Williams, Chapman, & Cross, 2010;Scott et al., 2013). It is therefore important that reliable and precise methods are used to evaluate efficacy and specificity of gene silencing following RNAi to avoid false-positive or false-negative results and consequently wrong conclusions (Herbert et al., 2011;Holmes et al., 2010). In addition to phenotypic observations, there are two standard methods for assessing the success of RNAi-mediated gene silencing: real-time quantitative PCR (RT-qPCR) for quantifying depletion of relevant mRNA transcripts relative to controls, and western blotting or immunofluorescence for measuring the reduction in the amount of target proteins (Agrawal et al., 2003;Scott et al., 2013).
While it is known that accurate quantification of target mRNA transcript levels by RT-qPCR after gene knockdown with exogenous siRNAs depends on the selection of primer binding site relative to the siRNA cleavage site (Herbert et al., 2011;Holmes et al., 2010), it is unclear how RT-qPCR primer selection affects quantification of dsRNA-mediated RNAi gene knockdown, especially in relation to the expression levels of target genes. Our first insights into the importance of primer design in the accurate measurement of dsRNA gene silencing came about when we were studying the role of Dysdercus fasciatus' (Hemiptera: Pyrrhocoridae; Figure 1) immune genes in the regulation of its gut bacterial symbionts (Onchuru & Kaltenpoth, 2019). Following dsRNA-mediated RNAi, transcript levels of genes under low expression remained unchanged or were even higher in the treatment groups compared to the controls, while transcript levels of the highly expressed target genes decreased significantly as expected.
To identify if this unexpected result was due to primer design and/or target gene expression levels, we designed two primer pairs for three genes with different levels of expression, respectively; one primer pair targeted the part of the mRNA transcript sequence that was entirely complementary to the dsRNA used for silencing and the other amplified a sequence extending beyond the region complementary to the dsRNA construct within the target mRNA. We report that the use of qPCR primers targeting a sequence that is completely contained within the dsRNA construct can lead to false negatives or an underestimation of gene knockdown in genes with low expression levels. F I G U R E 1 The African cotton stainer (Dysdercus fasciatus) adult insect (left) and its juveniles (right) maintain a symbiotic relationship with gut bacterial mutualists. These symbionts supplement the insect's cotton seed diet with limiting B-vitamins and protect against trypanosomatid infections. Once established, the gut microbiota appears to be resistant to the host's immune effectors, as revealed by RNA interference of immune effector and upstream signaling genes. © Martin Kaltenpoth, Johannes Gutenberg University Mainz

| MATERIAL S AND ME THODS
The synthesized dsRNA was used for in vivo RNAi gene knockdown that was performed by feeding 15 replicate individuals of second instar D. fasciatus nymphs that had been deprived of water for 24 hr with dsRNA for defensins (i.e., defensin 1 and defensin 2 in combination) or c-type lysozyme, or nonsense dsRNA for GFP for the control group, respectively. In pilot experiments, we found that there was no difference in the degree of silencing when the two defensin genes are silenced independently or together. One week after dsRNA feeding, total RNA was extracted from one nymph per replicate treatment and used for cDNA synthesis as described above to measure gene knockdown success. Two qPCR primer sets (Table 1) were designed for each gene with primer BLAST using their respective sequences obtained from the D. fasciatus transcriptome (Bauer et al., 2014). One primer set amplified the section of the target mRNA sequence that was entirely complementary to the dsRNA sequence used for silencing, while the second primer set amplified the mRNA sequence extending beyond the region that was complementary to the dsRNA construct ( Figure 2). Specificity of the primers was determined by blasting their sequences and those of their respective PCR products against a local BLAST database that was created using the D. fasciatus transcriptome. Blasting of the PCR product sequences also allowed us to exclude the possibility of off-target gene silencing.
Quantitative PCR was set-up using these two primer sets to quantify gene knockdown success. The 10 µl qPCR reaction mixture contained 0.5 µl of each primer (10 µM), 5 µl SYBR-mix, 3 µl of qPCR H 2 O, and 1 µl of either the cDNA template or a standard or a negative control (H 2 O). The reaction mixture was run on the Rotor-Gene Q cycler (Qiagen, Hilden, Germany) with the following cycling conditions: 95°C initial denaturation for 5 min followed by 95°C denaturation for 10 s, touchdown annealing for 15 s at 68-60°C for the first eight cycles, then 60°C annealing for the remaining 37 cycles for 15 s, extension of 72°C for 10 s, and a final melting curve analysis from 65°C to 99°C with a temperature raise of 1°C for each step. Quantification of each target gene was performed with the Rotor-Gene Q software as described by (Onchuru, Martinez, & Kaltenpoth, 2018) using external standard curves amplified with similar conditions from serial dilutions (10 10 -10 2 copies/µl) of purified PCR product of the respective gene. The transcript copy numbers of each gene were normalized with the number of host 18S rRNA transcripts that were quantified by qPCR using previously described primers (Table 1; Onchuru et al., 2018) and conditions described above. Differences in the normalized transcript copy numbers between treatments and controls were evaluated using Wilcoxon-signed rank tests and plotted using boxplots as implemented in SPSS Version 23 (IBM). Part of the data analyzed here, that is, gene expression using primers targeting mRNA sequence extending beyond the dsRNA target site, is published in (Onchuru & Kaltenpoth, 2019).
TA B L E 1 qPCR primers used in measuring dsRNA-mediated gene knockdown   Efficacy and duration of RNAi-mediated gene silencing varies not only between organisms but also between genes of the same organism (Scott et al., 2013;Wang et al., 2016). This disparity may be explained by variation in the RNAi machinery or extracellular enzymatic capacity of different organisms or tissues to degrade exogenous dsRNA molecules, which affects their stability in vivo and consequently silencing efficiency (Scott et al., 2013;Spit et al., 2017;Wang et al., 2016). Additionally, in some organisms, the silencing signal can be amplified by different mechanisms, for example, the RNA-dependent RNA polymerase (RdRp), which uses siRNAs generated from the diced primary dsRNA molecule as primers to copy the target mRNA resulting in the formation of secondary dsRNA (Sijen et al., 2001). Our findings suggest that variation of  (Herbert et al., 2011;Holmes et al., 2010). For example, after siRNA-mediated cleavage, degradation of the generated 3' mRNA fragment may be blocked; hence the use of primers targeting this fragment leads to an underestimation of the degree of RNAimediated gene silencing (Holmes et al., 2010;Mainland, Lyons, Ruth, & Kramer, 2017). On the other hand, using primers too close to the siRNA cleavage site results in false-positive results when contaminating siRNA molecules extracted with total RNA inhibit RT-qPCR (Herbert et al., 2011).

| RE SULTS
RNAi is a useful tool for research and therapeutics (Agrawal et al., 2003;Seyhan, 2011 (Spit et al., 2017), accurate quantification of successful RNAi gene silencing is key to avoiding incorrect conclusions. With RT-qPCR being the most common and widely used method for the quantification of intact target mRNA transcripts after silencing, its reliability is essential regardless of the gene or organism being studied. Based on our findings, we recommend the use of primers amplifying the mRNA transcript sequence extending beyond the dsRNA target region of the mRNA transcript to ensure accurate quantification of RNAi gene knockdown, especially in genes with low expression patterns.

ACK N OWLED G M ENTS
We thank Dagmar Klebsch for the excellent technical assistance. This work was supported by the Jena School of Microbial Communication (JSMC fellowship awarded to TOO), the Max Planck Society (to TOO and MK), and Johannes Gutenberg University-Mainz (to MK). The funders had no role in study design, data collection, and interpretation, or the decision to submit the work for publication.

CO N FLI C T O F I NTE R E S T
None declared.

DATA ACCE SS I B I LIT Y
The gene expression raw data measured by qPCR are available from the Dryad repository (https ://doi.org/10.5061/dryad.28n8d6t).