Validation of reference genes for quantitative PCR in the forest pest, Ips calligraphus

The six-spined ips, Ips calligraphus, is a North American bark beetle that can exploit most eastern North American Pinus species and can cause mortality. Biotic and abiotic disturbances weaken trees, creating breeding substrate that promotes rapid population growth. Management historically relied on silvicultural practices, but as forests become increasingly stressed, innovative management is needed. Manipulation of the cellular RNA interference (RNAi) pathway to induce gene silencing is an emerging means of insect suppression, and is effective for some bark beetles. Quantitative PCR (qPCR) is a powerful tool for analysis of gene expression, and is essential for examining RNAi. To compare gene expression among individuals, stably expressed reference genes must be validated for qPCR. We evaluated six candidate reference genes (18s, 16s, 28s, ef1a, cad, coi) for stability under biotic (beetle sex, developmental stage, and host plant), and abiotic (temperature, photoperiod, and dsRNA exposure) conditions. We used the comprehensive RefFinder tool to compare stability rankings across four algorithms. These algorithms identified 18s, 16s, and 28s as the most stably expressed. Overall, 16s and 28s were selected as reference genes due to their stability and moderate expression levels, and can be used for I. calligraphus gene expression studies using qPCR, including those evaluating RNAi.


Results
Across all treatments combined (Fig. 2) the most highly expressed gene is 18s ribosomal RNA (18s) with Cq values ranging from 12.7 to 15.6, while Ca2 carbamoyl-phosphate synthetase 2 (cad) has the lowest expression with Cq values ranging from 28.8 to 34.9.
Cytochrome oxidase I (coi), has the greatest range of expression, with Cq values ranging from 23.9 to 35.1. Of the remaining genes evaluated, 16s ribosomal RNA (16s), 28s ribosomal RNA (28s), and elongation factor 1-alpha (ef1a), are moderately expressed, with average Cq values ranging from 19.2 to 24.7.
The average Cq value of each candidate reference gene was determined for each condition, including beetle sex, developmental stage, host plant, temperature, light conditions, and dsRNA exposure (Fig. 3a-f). The relative level of expression for each candidate gene was consistent across all conditions, with 18s having the highest expression, and cad being the least expressed. Across all conditions tested, coi was the most variable, having the largest range of Cq values.
GeNorm 28 uses a gene expression stability value (M) to determine stability; M < 1.5 is required for a gene to be considered stable. All candidate genes tested had an M value below 1.5, with 16s and 18s being ranked equally as most stable (Fig. 4). NormFinder 28 ranks candidate genes based on an overall stability value (SV); SV > 1 indicates  www.nature.com/scientificreports/ lower stability. 18s and 16s had the lowest SV scores, with 28s and ef1a being the only other candidate genes to meet the threshold. BestKeeper 28 uses standard deviation (SD) to determine stability, with a SD < 1 more stable. Using this approach, all candidate genes except coi met the threshold for stable expression, but 18s, 28s, and 16s were ranked as the most stable. The comparative delta-Ct method also uses SD to determine stability, with a lower value indicating a more stable expression. Using the delta-Ct method, 18s and 16s were the most stable. The geomean value was determined using the web tool RefFinder 28 , which gave a comprehensive ranking across the four algorithms, with 18s ranking first, 16s ranking second, and 28s ranking third. The four top-ranking reference gene candidates met the stability parameters of each of the algorithms (Table 1). Of these, 16s and 28s were selected as reference genes, as they have similar levels of stability to 18s across the algorithms used, while having a more moderate Cq values, allowing them to be more applicable to evaluating genes with moderate levels of expression.

Discussion
qPCR is an indispensable tool for evaluating gene expression, an important aspect of many genetic studies, and essential for demonstrating RNAi-induced gene silencing, as its sensitivity permits detection of even minor differences between samples 27,29,30 . However, this sensitivity can also be a limitation, as any differences in baseline transcription levels, variation in sample preparation, or technical inconsistencies become evident. To normalize gene expression between samples for qPCR quantification of mRNA levels, internal controls that are stably expressed in various experimental conditions are needed 26 . The selection and use of stable reference genes allow researchers to normalize variation between samples, thus giving a more accurate representation of gene expression differences and avoiding error. Despite some variation in ranking, the top four reference genes (18s, 16s, 28s, and ef1a) were consistently stable regardless of the algorithm used. In addition, they were consistently stable across the biotic and abiotic conditions we tested, and these conditions may vary across different experimental designs, increasing the versatility of our selected reference genes for use in gene expression studies. The conserved nature of these essential genes should also allow for their use in genetically variable individuals, though further investigation is required to confirm this among different populations or subspecies. Stability in dsRNA-exposed beetles and individuals of different developmental stages could allow for these same reference genes to be used for normalizing gene expression in RNAi experiments using oral delivery of dsRNA in a sucrose solution, with treatment at multiple life stages, an important factor given that I. calligraphus causes damage both through larval feeding, and during dispersal as adults by carrying damaging microorganisms to naïve hosts. Additionally, RNAi manipulations allow for investigations into gene function using reverse genetics, serving as a technique that can illuminate the effects of reduced gene expression at a variety of life stages, including genes for which knockouts prevent complete development.
This study provides an essential foundation for future gene expression work in the experimental conditions analyzed. Additionally, the selection of reference genes that are stable for dsRNA treated beetles will serve as an important guide for their selection in future RNAi studies, a valuable tool of reverse genetics, and importantly a developing management tool, though additional efforts to verify reference gene stability across the specific conditions used in future RNAi studies are required. Due to the requirement of a ≥ 16 base pair match for gene silencing to occur 24 , with careful selection of the target gene sequences dsRNAs can be engineered to be highly specific to the target insect. Thus, utilizing RNAi for pest suppression could offer a means for managing for I. calligraphus while minimizing harmful effects on non-target organisms, one barrier to use of traditional insecticides to prevent and reduce outbreak populations. Further work establishing the practicality of potential   www.nature.com/scientificreports/ methods of delivering dsRNA, such as root drenches, trunk injections, or transgenically expressed dsRNA, will be critical to deployment of this innovative technology, and will enhance our understanding of how it may be implemented in a forest ecosystem. Climate change, with associated shifts in temperature and rainfall, have made forests more susceptible to large outbreaks in which stressed, and even healthy trees are killed. In these conditions, traditional management techniques like silvicultural control measures are not sufficient to promote healthy forests, and begin to lose their efficacy. These concerns, in addition to the threat that the beetle may pose in its expanding introduced ranges, mean that a new technique must be added to our existing integrated pest management strategies, for which RNAi could be invaluable. Moreover, I. calligraphus represents just one species in a genus comprised of multiple devastating forest pests, including the massively destructive Ips typographus. Outbreaks of I. typographus have caused unparalleled damage in European forests, killing millions of trees and even nearly eliminating its primary host, Norway spruce (Picea abies), in affected areas 31 . These outbreaks are devastating ecologically, representing the loss of key carbon sinks and wildlife habitat, as well as a loss of genetic diversity associated with old growth forests that will be vital to adaptation and survival of these forests in the face of rapidly changing environmental conditions 31 . I. typographus has been intercepted repeatedly at North American ports 32 , and when coupled with its potential for expanding suitable ranges with climate change 33,34 , it becomes evident that forests across the globe are at great risk of damage by this beetle. As I. calligraphus outbreak populations respond to increasingly frequent disturbance events and invade naïve ranges, and other Ips species continue to devastate forests globally, the ability to apply modern molecular techniques in innovating new management approaches is critical.

Materials and methods
Gene selection. Candidate reference genes ( Table 2) were selected because they are transcribed in all cells and have essential functions that are ubiquitously expressed across cell types. Sequences were obtained from previous entries to NCBI, and primers were designed using the online tool Primer3Plus 35 . Primer pairs with an amplicon length between 80-120 bp, a GC% of ~ 50%, and a melt temperature of 60 °C, as well as with the lowest self and any scores, were selected. Primers were only selected if the linear regression coefficient (R 2 ) was > 0.99 and efficiency percentage was 90-110%.

Insects. Experimental insects (Ips calligraphus calligraphus)
were obtained by suspending I. calligraphus lures (Synergy Semiochemicals Corporation, Delta, BC) on a plantation located on Florida Forestry Service land in Newnans Lake State Forest in central Florida for a two-week period in January 2020. Infested trees were felled by Florida Forestry Service personnel in compliance with institutional, national, and international guidelines addressing collections of plant material, and in compliance with IUCN provisions. Stems were immediately transported to the University of Kentucky, Lexington, KY, and stored at 4 °C. Stems were sectioned and transferred to rearing bins (55.6 × 62.7 × 81.3 cm) as needed. Beetles used to evaluate stability of gene expression based on sex and developmental stage were processed immediately, otherwise newly emerged beetles were maintained under specified conditions for 72 h to assess the stability of potential reference genes 18,19 . Beetles were sexed using the presence or absence of the pars stridens and the number of protibial spurs 36 . To evaluate developmental stage, adults, pupae, and larvae were collected by debarking an infested log and preserving for RNA extraction immediately. To evaluate reference stability of gene expression based on host plant, beetles were starved for 24 h, then fed white (P. strobi) or loblolly pine bark; beetles for each treatment fed on the same piece of bark, and feeding was confirmed by visual inspection of beetles within galleries and the presence of frass. Beetles were maintained at 20 °C and 25 °C to evaluate stability of gene expression based on temperature, and 16:8 L:D and total darkness to evaluate stability based on photoperiod. dsRNA treated beetles were individually fed 10 μg of green fluorescent protein dsRNA, and then after their four-hour feeding period kept for 72 h before RNA preservation. Excluding the sex and developmental stage treatments, all beetles were maintained in each respective , and PCR products were purified using the QIAquick PCR purification kit (Qiagen, Germantown, MD). This purified PCR product was then used to synthesize dsRNA using the MEGAscript RNAi Kit (Invitrogen, Waltham, MD). The quality and quantity of dsRNA was tested using NanoDrop spectrometry and gel electrophoresis.
Administering dsRNA. To administer the dsRNA treatments, 10 μg of dsRNA was suspended in a 0.5% sucrose solution containing 0.5% food dye in a 0.5 mL microcentrifuge tube, for a total of 4 μL of solution. The anterior end of each individual beetle was immersed in the solution up to the pronotum, so as to not interfere with spiracles and oxygen uptake. A KimWipe (Kimtech, Neenah, WI) was placed into the tube to prevent beetles from backing out until the solution was completely consumed (~ 4 h) 20 . After treatment exposure, beetles were placed in petri dishes (60 × 15 mm) lined with moistened KimWipes, and oriented vertically in a cylindrical humidity chamber (~ 23 × 21 cm) for 72 h at 23℃.  TAA TAC GAC TCA CTA TAG GGC GAT GCC ACC TAC GGC AA   TAA TAC GAC TCA CTA TAG GGT GTC GCC CTC GAA CTT CA