Transcriptome-Based Selection and Validation of Reference Genes for Gene Expression in Goji Fruit Fly (Neoceratitis asiatica Becker) under Developmental Stages and Five Abiotic Stresses

Abstract

Goji fruit fly, Neoceratitis asiatica, is a major pest on the well-known medicinal plant Lycium barbarum. Dissecting molecular mechanisms of infestation and host selection of N. asiatica will contribute to the determination of best management practices for pest fly control. Gene expression normalization by Real-time quantitative PCR (qPCR) requires the selection and validation of appropriate reference genes (RGs). Hence, 15 candidate RGs were selected from transcriptome data of N. asiatica. Their expression stability was evaluated with five algorithms (∆Ct, Normfinder, GeNorm, BestKeeper, and RefFinder) for sample types differing in the developmental stage, sex, tissue type, and in response to five different abiotic stresses. Our results indicated that the RGs β-Actin + GST for sex, RPL32 + EF1α for tissue type, RPS13+ EF1α for developmental stages along with odor stimulation, color induction, and starvation-refeeding stresses, EF1α + GAPDH under insecticide stress, RPS13 + RPS18 under temperature stress, respectively, were selected as the most suitable RGs for qPCR normalization. Overall, RPS18 and EF1α were the two most stable RGs in all conditions, while RPS15 and EF1β were the least stable RGs. The corresponding suitable RGs and one unstable RG were used to normalize a target odorant-binding protein OBP56a gene in male and female antennae, different tissues, and under odor stimulation. The results of OBP56a expression were consistent with transcriptome data. Our study is the first research on the most stable RGs selection in N. asiatica, which will facilitate further studies on the mechanisms of host selection and insecticide resistance in N. asiatica.

Keywords: Neoceratitis asiatica; expression stability; qPCR normalization; reference genes; target gene expression.

Figure 1

Lycium barbarum L. and the symptoms of its fruits harmed by Neoceratitis asiatica. (A) Mature fruits of L. barbarum. (B) Dried ripe fruits of L. barbarum L. (C) Harmful symptoms of N. asiatica female adult laying eggs on young fruit (5–7 days after falling flower) of L. barbarum L. (D) Harmful symptoms of N. asiatica larva on mature fruits (28–31 days after falling flower) of L. barbarum L.

Figure 2

Specificity of primer pairs for qPCR amplification in N. asiatica. The melt peaks of primers for qPCR amplification of 15 candidate RGs, including α-TUB, β-Actin, EF1α, EF1β, GAPDH, G6PDH, UBC, AK, GST, SDHA, TBP, RPL32, RPS13, RPS15 and RPS18.

Figure 3

Candidate reference genes expression profiles in N. asiatica. The expression data are presented as mean Ct values of candidate reference genes across all samples under eight different experimental conditions. Whiskers represent the maximum and minimum values. The lower and upper borders of the boxes represent the 25th and 75th percentiles, respectively. The line across the box indicates the median Ct value.

Figure 4

GeNorm analysis of paired variation (V) values of 15 candidate reference genes. Vn/Vn + 1 values are used to determine the optimal number of reference genes. The cut-off value to determine the optimal number of reference genes for qPCR normalization is 0.15.

Figure 5

Stability of 15 candidate reference genes in N. asiatica under eight experimental conditions by RefFinder. In a RefFinder analysis, increasing Geomean values correspond to decreasing gene expression stability. The Geomean values for the following N. asiatica samples are presented: (A) Odor stimulation: samples treated with different odor compounds; (B) Color induction: samples treated with different colors; (C) insecticide treatment: samples treated with imidacloprid insecticide; (D) Starvation-refeeding: samples treated with chlorpyrifos; (E) Temperature: samples treated with different temperature; (F) Developmental stages: samples for all developmental stages; (G) Sex: samples for male adults and female adults; (H) Tissues: samples for different tissues; (I) All samples: all samples for all treatments. The details of 15 candidate RGs are listed in Table 1. The most stable genes are listed on the left, while the least stable genes are listed on the right.

Figure 6

The expression profiles of OBP56a in both sexes and in different tissues of N. asiatica adults. (A) The expression levels of OBP56a in the antennae of male and female flies presented by the FPKM value of transcriptome data. FPKM: fragments per kilobase of exon model per million mapped reads. (B) The expression levels of OBP56a in male and female antennae by qPCR. The expression levels of OBP56a in male antennae were used for normalization. Namely, the relative expression levels were presented as fold changes relative to the transcript levels of OBP56a in the male antennae. β-Actin, GST, and β-Actin + GST were used as the one or two most stable reference genes. TBP was used as the least stable reference gene. (C) The expression levels of OBP56a in different tissues (female antennae, female legs, and female ovipositor) presented by the FPKM value of transcriptome data. (D) The expression levels of OBP56a in female antennae, female legs, and female ovipositor by qPCR (most stable RGs: RPL32 and EF1α, least stable RG: EF1β). The OBP56a expression in female ovipositor was used for normalization. Data are represented as the mean ± SE. Bars labeled with different letters are significantly different (p < 0.05, ANOVA followed by Tukey’s HSD multiple comparison test, n = 3).

Figure 7

The expression analysis of OBP56a in three olfactory organs of female N. asiatica adults after odor stimulation. (A) The expression levels of OBP56a in female antennae, female legs, and female ovipositor before/after odor-stimulation presented by FPKM value of transcriptome data. Control showed the female adults without odor stimulation. Odor sources were collected from the volatiles of wolfberry fruits (green fruits, 5–7 days after falling flower). (B–F) The expression levels of OBP56a in three organs of female adults normalized to the top two stable genes (RPS13 and EF1α) and an unstable gene (RPS15) by qPCR. The expression levels of OBP56a in female ovipositor were used for normalization. Asterisk (*) indicated a significant difference between control vs. odor stimulation (mean ± SE, n = 3, Student’s t-test, * p < 0.05; ** p < 0.01). Bars labeled with different letters are significantly different (mean ± SE, n = 3, Tukey’s HSD, p < 0.05).