A gene can look perfectly stable until the experiment changes. This study shows that the safest RT-qPCR reference gene for a parasitoid wasp depends on whether researchers are comparing tissues, development, sex, or diapause. RT-qPCR remains a vital approach for quantifying gene expression in physiological or pathological settings, but appropriate reference genes are essential for meaningful results.

Anastatus japonicus is an important egg parasitoid wasp and natural enemy of fruit bugs and forest caterpillars, yet no validated reference genes had been established for it. The study screened eight candidate reference genes under developmental-stage, tissue, sex, and diapause comparisons, using RefFinder to integrate four algorithms.

The resulting stable genes differed by condition: RPS6 and RPL13 for development, ACTIN and EF1α for sex, RPL13 and EF1α for tissues, and TATA and ACTIN for diapause. RT-qPCR can measure gene expression rapidly, sensitively, efficiently, accurately, and reproducibly under different analytical conditions.

But reliability depends on template quality and yield, primer amplification efficiency, and the underlying biological signal. Internal reference genes are indispensable for normalization and correction among different samples. Common housekeeping genes include eighteen S ribosomal RNA, tubulin, ACTIN, and GAPDH, but these genes may not remain stable across all experimental conditions.

The practical implication is that reference genes must be selected systematically for the particular RT-qPCR application. The study evaluated eight commonly used reference genes for normalizing RT-qPCR data from Anastatus japonicus specimens. The candidates were RPS3a, GAPDH, ACTIN, EF1α, TATA, RPS6, TUB, and RPL13.

The selection focused on genes transcribed by RNA polymerase II whose transcripts can be isolated using oligo dT, including GAPDH, ACTIN, and ribosomal proteins. RefFinder integrated BestKeeper, NormFinder, geNorm, and the delta Ct method, comparing stability across sex, tissue type, developmental stage, and diapause state.

The developmental comparison included mature larvae, pupae, and male and female adults sampled on day one of the corresponding stage. The sample counts were eight mature larvae, ten pupae, twenty male adults, and ten female adults. The tissue comparison used head, chest, and abdomen samples from about forty adult females.

To induce diapause, Antheraea pernyi eggs parasitized by Anastatus japonicus were kept for two days under standard conditions, then transferred to seventeen degrees Celsius with a ten-hours-light, fourteen-hours-dark photoperiod. The diapause-inducing treatment lasted twenty-five, thirty-five, forty-five, or fifty-five days.

Diapause was terminated after fifty-five days by returning the individuals to normal developmental conditions for twelve days. RT-qPCR examined the expression of eight commonly used reference gene candidates, with the candidates listed in Table 1. Primers were designed with the PrimerQuest Tool from sequences derived from the researchers’ transcriptomic analysis of Anastatus japonicus.

Table one lists the forward and reverse primer sequences for eight reference-gene candidates in A. japonicus, along with GenBank identifiers, amplicon lengths, amplification efficiencies, R-squared values, and linear-regression equations.

Amplicon lengths range from one hundred eight to one hundred forty base pairs, while the reported efficiencies and R-squared values document the performance of each assay. These specifications matter because the authors used these primers in RT-qPCR to evaluate reference genes across developmental stages and adult tissues.

The RT-qPCR reactions used TB Green Premix Ex Taq and a CFX Connect Real-Time Instrument in a total volume of fifty microliters. The amplification program began at ninety-five degrees Celsius for thirty seconds, followed by thirty-nine cycles of ninety-five degrees for five seconds and fifty-five degrees for thirty seconds.

Melting curves were measured from fifty-five to ninety degrees Celsius in sequential half-degree steps, and standard curves used a two-fold cDNA dilution series. All eight candidate genes were detected in the Anastatus japonicus samples, and each transcript produced a single amplicon of the expected size on one-percent agarose gels.

Melt curves confirmed amplification specificity through a single peak without evidence of primer-dimer peaks. Amplification efficiencies ranged from ninety-nine percent to one hundred seven point sixty-seven percent, with R squared values greater than zero point nine nine zero nine.

Figure one shows melt curves for eight candidate reference genes in A. japonicus: ACTIN, EF1α, GAPDH, RPL13, RPS3a, RPS6, TATA, and TBP. Each panel displays fluorescence derivative against temperature, with the amplification products producing a prominent melt peak; the authors used these curves to confirm amplification specificity by checking for a single peak without evidence of additional products.

This quality check supports the subsequent reference-gene stability analysis. For the developmental-stage comparison, RPS6 and RPL13 were identified as the most stable reference genes by the delta Ct and related analytical approaches.

BestKeeper instead identified EF1α and RPS3a as the most stable, while ACTIN and GAPDH were the least stable in that analysis. RefFinder’s overall stability order began with RPL13 and RPS6, followed by RPS3a, EF1α, TATA, TUB, ACTIN, and GAPDH. The geNorm pairwise V two over three value was below the established zero point one five cutoff, and RefFinder therefore suggested RPL13 and RPS6 for developmental-stage analysis.

Table two compares eight candidate reference genes across developmental stage, tissue, sex, diapause, and all conditions. For each setting, geNorm, NormFinder, BestKeeper, and delta Ct report stability values and ranks, while the recommendation column summarizes selected genes, such as RPL13 for developmental stage and RPS3a for all conditions.

This matters because the authors found that different analytical methods can identify different candidates, so reference-gene choice depends on the biological condition and validation approach. Figure three summarizes RefFinder’s stability rankings for eight candidate reference genes across developmental stage, tissue, sex, diapause, and all conditions.

The ordering changes by context: for example, RPL13 and RPS6 appear first for developmental stage, ACTIN leads in tissue, and RPL13 leads for sex, while the combined analysis begins with RPS3a and RPL13. This matters because it shows that reference-gene selection in A. japonicus should be matched to the specific experimental condition.

For tissue comparisons, GAPDH was identified as the least stable reference gene in the geNorm, NormFinder, and BestKeeper analyses. RefFinder’s tissue ranking placed ACTIN first, followed by RPS6, RPL13, EF1α, RPS3a, TUB, TATA, and GAPDH. The pairwise V two over V three value was below zero point one five, although the V seven over V eight value was above that threshold.

Overall, ACTIN and RPS6 were indicated as the most appropriate reference genes for normalizing expression across different tissue types. For the diapause treatments, geNorm identified ACTIN and TATA as the most stable genes and RPS6 as the least stable.

The delta Ct and NormFinder analyses produced similar results, while BestKeeper instead identified RPL13 and EF1α as the most stable. RefFinder ranked TATA first and ACTIN second, followed by RPL13, RPS3a, EF1α, TUB, GAPDH, and RPS6. All pairwise variation values were below zero point one five, and ACTIN and TATA were therefore the most reliable choices for the tested diapause conditions.

Figure four uses geNorm pairwise variation to show how many reference genes are needed for reliable normalization across developmental stage, sex, tissue, and diapause groups. The chart compares sequential values from V two-three through V seven-eight against the dashed zero point one five cutoff.

The authors report that V two-three is below this threshold in some analyses, while all pairwise values are below it when all treatment conditions are considered, illustrating how the recommended number of reference genes depends on the grouping examined. FD, or farnesol dehydrogenase, was used to validate the reference genes across the diapause treatments.

FD expression levels on diapause day twenty-five were fifty-four-fold higher than the non-diapause treatment when normalized to the most stable genes, and eighty-five-fold higher when normalized to the least stable genes. With ACTIN and TATA, significant increases in FD expression appeared in day fifty-five relative to day twenty-five; with unstable genes such as GAPDH and RPS6, that difference was not significant.

Figure five validates the reference-gene normalization by measuring FD, or farnesol dehydrogenase, across non-diapause, diapause, and diapause-termination larvae. The authors show two normalizations: ACTIN with TATA in panel A, and GAPDH with RPS6 in panel B; both report significant differences among conditions, with letters marking statistically distinct groups at p less than zero point zero five.

This matters because FD is linked to diapause biology, so its expression provides a test of whether the selected reference genes support consistent RT-qPCR interpretation. The results revealed marked differences in the optimal internal reference genes for different sample types, supporting the use of more than one reference gene for normalization.

geNorm establishes the optimal number of reference genes through pairwise variation based on successive normalization factors. The optimal number is defined when V sub n over V sub n plus one is below zero point one five. The identified combinations were RPL13 and RPS6 for developmental stages, ACTIN and RPS6 for tissue types, RPL13 and EF1α for sexes, and ACTIN and TATA for diapause treatments.

Table Three summarizes reference-gene stability rankings across Hymenoptera, organizing each species by treatment and listing the most stable, least stable, and all genes tested. The table shows that rankings are context-dependent: for example, Aphidius gifuensis changes from ACTB, RPL13, and PPI across developmental stages to RPL13, RPL27, and ACTB under diverse diets, while Cotesia chilonis has different top genes under low- and high-temperature treatments.

This matters because selecting an appropriate reference gene is essential for reliable gene-expression comparisons. The conclusion is that application-specific reference genes are important for Anastatus japonicus RT-qPCR because no single reference gene is optimal under all conditions.

Across the tested experimental conditions, RPS3a and RPL13 were identified as the most appropriate reference genes on average. The ideal combinations still varied: RPL13 and RPS6 for developmental stages, ACTIN and RPS6 for tissues, ACTIN and TATA for diapause treatments, and RPL13 and EF1α for female and male specimens.

There is no universal reference gene for Anastatus japonicus. Reliable RT-qPCR normalization requires condition-specific combinations, with RPL13 and RPS6, ACTIN and RPS6, ACTIN and TATA, or RPL13 and EF1α depending on the comparison.