Transcriptome Analysis Reveals the Potential Antioxidant Defense Mechanisms of Myzus persicae in Response to UV-B Stress

Background: As an environmental stress factor, ultraviolet-B (UV-B) radiation directly affects the growth and development of Myzus persicae. Excessive UV-B stress leads to DNA, membrane lipid, and protein damage by the production of reactive oxygen species. However, M. persicae can adaptively respond to such environmental stress by activating the relevant mechanisms in the body. How M. persicae responds to UV-B stress and the molecular mechanisms underlying this adaptation remain unknown. Results: Here, we compared and analyzed transcriptome data for M. persicae following exposure to a light-emitting diode fluorescent lamp and UV-B radiation for 30 min. We identified 758 significant differentially expressed genes (DEGs) following exposure to UV-B stress, including 423 upregulated and 335 downregulated genes. In addition, enrichment analysis using the Gene Ontology and Kyoto Encyclopedia of Genes and Genomes databases illustrated that these DEGs are associated with antioxidation and detoxification, metabolic and protein turnover, immune response, and stress signal transduction. Simultaneously, these DEGs are closely related to the adaptability to UV-B stress. Conclusions: Our results suggest that UV-B stress is associated with a wide range of physiological effects in M. persicae. Our research can raise awareness of the mechanisms of insect responses to UV-B stress.


Background
In recent years, the destruction of the stratospheric ozone layer in the Earth's atmosphere has sharply increased interest in the effects of solar ultraviolet 3 radiation on the Earth's surface, especially ultraviolet-B (UV-B) radiation with a wavelength of 280-320 nm [1]. UV-B is considered a widespread environmental stress factor that induces oxidative stress in organisms through the production of reactive oxygen species, causing DNA, membrane lipid, and protein damage [2][3][4].
Numerous studies illustrated that UV-B radiation has a wide range of effects on the growth, physiology, biochemistry, and population structure of organisms, and organisms can adopt both protection and repair strategies to adapt to UV-B stress [5,6]. As an environmental stress factor, UV-B can cause oxidative stress and genetic mutations, leading to death of insects [7][8][9][10]. However, the molecular mechanisms by which insects adapt to UV-B stress are unclear.
The green peach aphid Myzus persicae (Sulzer) (Homoptera: Aphididae) is a worldwide pest that seriously harms > 400 plants such as tobacco, cruciferous vegetables, peppers, potatoes, eggplants, and melons. This pest can cause the leaves of plants to curl, wither, and even die, thereby reducing crop yields [11,12].
It can also spread > 100 plant viruses as a viral vector and cause sooty blotch and mold parasitic infection through the secretion of honeydew, causing great losses in the production of cash crops [11]. M. persicae lives year-round under direct sunlight, but it is unable to escape from UV-B as an environmental stress factor.

mRNA sequencing, sequence assembly, and functional annotation
In total, 177,070,180 (26.56 Gb) raw reads were generated from the two libraries (control and UV-B radiation groups) using Illumina HiSeq™ 4000 sequencing technology, and 80,281,002 and 90,660,496 clean reads were obtained from the two groups, respectively, after quality control decontamination ( Table 1). The data quality of the clean reads for the control and UV-B radiation groups were separately evaluated. The results illustrated that the Q30 quality score exceeded 92% for both groups. The GC contents of the two groups were 41.6% and 41.5%, respectively, and >93% of the clean reads were independently aligned on the M. persicae reference genome, and <4% of the clean reads of both groups had multiple alignment positions on the reference genome. In addition, the intron region, 5′ UTR, 3′ UTR, and CDS comprised 1.62%, 6.10%, 4.7%, and 83.72% of the clean reads in the control group, respectively, versus 2.01%, 7.03%, 4.4%, and 83.48%, respectively, in the UV-B radiation group. These clean reads were assembled to obtain 40,699 unigenes, and the length distribution of all unigenes is shown in Fig. 1. These results demonstrated that the sequencing quality was relatively high, indicating that the unigenes were suitable for subsequent annotation analysis. We then annotated our unigenes using six functional databases and found that 37

GO and KEGG analyses of DEGs
We focused on 423 upregulated and 335 downregulated genes to further understand the biological mechanism by which M. persicae responds to UV-B stress. In GO analysis, we grouped DEGs into three categories, namely, biological process, cellular component, and molecular function (Fig. 3). In the biochemical process category, "metabolic process," "cell process," and "single biological process" were significantly enriched. In the cell component category, "membrane," "cell part," "membrane part," and "cell" were significantly enriched. In the molecular function category, "catalytic activity" and "binding" were significantly enriched. In addition we also found that 31 DEGs were significantly associated with the "response to stimulus." KEGG is a bioinformatics database for the systematic analysis of gene function [13].
In this study, 576 DEGs were annotated into 194 pathways, which were further divided into six categories as follows: cellular processes, environmental information processing, genetic information processing, metabolism, organismal systems, and human metabolism diseases. The first 62 KEGG pathways are shown in Fig. 4.
Among these pathways, the immune and antioxidant defense, transport and catabolic, and signal transduction pathways were mainly enriched.

Antioxidant and detoxification
In this study, we identified several genes involved in the regulation of antioxidant and detoxification mechanisms (Fig. 5). Seven DEGs in the peroxisome pathway were associated with antioxidant systems [14]. In this pathway, fatty acyl-CoA

Metabolic and protein turnover
In this study, many metabolic reactions were significantly enhanced under UV-B stress in M. persicae, including carbohydrate (12 pathways), lipid metabolism (13 pathways), cofactor and vitamin metabolism (6 pathways), and amino acid metabolism (12 pathways). Among them, genes induced in carbohydrate metabolism included those involved in pyruvate metabolism (three DEGs), tricarboxylic acid cycle (TCA cycle) (three DEGs), glycolysis/gluconeogenesis (two DEGs), ascorbic acid and metabolism (four DEGs), and starch and sucrose metabolism (five DEGs).
Moreover, we found that some genes related to amino acid metabolism were induced, including those involved in alanine, aspartate, and glutamate metabolism (three DEGs); glycine, serine, and threonine metabolism (two DEGs); cysteine and methionine (two DEGs); valine, leucine, and isoleucine degradation (seven DEGs); lysine degradation (three DEGs); and arginine and proline metabolism (one DEG).

Immune response
In our study, many immune-related DEGs were significantly enriched, most of which were upregulated in response to UV-B stress. These genes were mainly involved in five pathways, including antigen processing and presentation (17 DEGs), platelet activation (six DEGs), and chemokine signaling pathways (three DEGs) (Fig. 6).
Nine, five, and two unigenes related to antigen processing and presentation, platelet activation, and chemokine signaling pathways, respectively, were upregulated. Among these, cathepsin B (CTSB) has 14 DEGs (6 unregulated and 8 downregulated) involved in the antigen processing and presentation pathways. In addition, some antiviral genes such as serine proteinase inhibitor 2 (Unigene 7537) and scavenger receptor class B (Unigene 10115) genes were induced. Fc gamma Rmediated phagocytosis (Unigene 7353) gene, which is involved in insect immunity, was upregulated.

Validation of expression profiles via qRT-PCR
To verify the transcriptome data, we randomly selected 15 genes and further tested their relative expression levels via qRT-PCR. According to comparative analysis, the trend of qRT-PCR results was consistent with the results of DEG expression analysis ( Fig. 7), which verified the accuracy and reliability of the sequencing data.

Discussion
Most insects grow and multiply under direct sunlight and thus cannot escape the UV-B effect of direct sunlight. However, many insects also develop a variety of mechanisms, including morphological and physiological adaptations, in response to UV-B stress. This may be due to the expression of certain unique genes in the body of insects. In this study, we performed a comparative transcriptional analysis of M.
persicae to identify genes associated with the UV-B adaptability of this species. We identified 758 DEGs under UV-B stress (423 upregulated and 335 downregulated) and analyzed numerous biomarkers for antioxidants and detoxification, metabolic and protein turnover, immune response, and stress signal transduction. In GO analysis, 31 DEGs were significantly associated with the "response to stimulus," which is related to the response of M. persicae to UV-B stress. Similar results have been reported for response of Glyphodes pyloalis to heat stress and the response of Antheraea pernyi to zinc stress [15,16]. In addition, 92 DEGs were associated with "membrane," suggesting that most cells of M. persicae need to be repaired under UV-B stress [17].
UV can result in ROS accumulation in insect cells, and the imbalance between ROS 9 production and antioxidation can directly lead to a variety of toxic effects, including non-specific DNA, protein, and lipid damage [3,4,18,19]. Several antioxidantrelated genes, fatty acyl-CoA reductase (Unigene 15357 and Unigene 11089), hydroxymethylglutaryl-CoA lyase (Unigene 5738), mpv17-like protein 2 (Unigene 7104), and long-chain-fatty-acid-CoA ligase were significantly upregulated after UV-B treatment (Fig. 5). These upregulated genes function as antioxidants to remove ROS in M. persicae. In addition, UV-B stress promotes the accumulation of toxic substances in M. persicae. Regarding detoxification and antioxidative mechanisms, glutathione S-transferase can catalyze the binding of the electrophilic group of endogenous harmful substances produced by UV-B stress to the thiol group of reduced glutathione, thereby forming a more soluble, non-toxic derivative that is easily excreted or decomposed by metabolic enzymes [20]. The activity of glutathione S-transferase in Helicoverpa armigera adults was also significantly increased under UV stress [18]. As an important serine hydrolase in insects, carboxylesterase can effectively catalyze the hydrolysis of various endogenous and exogenous compounds containing carboxyl ester bonds, amide bonds, and thioester bonds [21]. In our study, multiple carboxylesterase genes were significantly induced. The P450 enzyme system can metabolize various harmful endogenous and exogenous substances to protect living cells [22]. However, under UV-B stress in M.
persicae, seven unique CYP sequences were induced, and these sequences were grouped into the CYP6 (four sequences), CYP4 (two sequences), and CYP18 clades (one sequence). CYP6BQ4 and CYP6BQ8 mRNA levels were also significantly increased under UV-A stress in Tribolium castaneum [ 23]. Therefore, the antioxidant and detoxification processes were speculated to be closely related to the response of M. persicae to UV-B stress. Because insects lack adaptive immunity, they can only rely on innate immune reactions for defense; however, these reactions also cause immunological changes during the stress response [28,29]. CTSB is an important proteolytic enzyme in insect lysosomes, which plays an important role in growth and metabolism by degrading protein activity and maintaining normal programmed cell death. As an important digestive protease present in various oviparous animals, CTSB provides nutrients for embryonic development by breaking down vitellin containing abundant amino acid components in the egg [30][31][32][33][34]. Simultaneously, antiviral and FcγRmediated phagocytosis of immune-related genes was induced under UV-B stress.
These upregulated genes associated with immune responses indicate that many immune responses in insects can be activated under UV-B stress, illuminating the strong immune adaptive functions of insects [29,35].
The sensing and transduction of intracellular stress signals are critical for the adaptation and survival of insects following exposure to UV-B radiation. Studies indicated that the PI3K-Akt pathway activates the nuclear factor erythroid 2 related factor 2 pathway, which is a key factor that protects cells against damage induced by UV-B by inhibiting oxidative stress [36]. Therefore, activation of the PI3K-Akt pathway may be critical in the insect response to UV-B stress. As an important pressure-sensing and energy-regulating factor, AMPK is essential for the survival of organisms in harsh environments. Under UV-B stress, the AMPK signaling pathway in M. persicae was activated, and this activation is related to its molecular adaptation mechanism under UV-B stress [37,38]. Activation of the cGMP-PKG signaling pathway mobilizes intracellular Zn 2+ to prevent mitochondrial oxidative damage in cardiomyocytes [39]. However, UV-B stress can cause significant ROS accumulation and damage to organisms. Activation of the cGMP-PKG signaling pathway can protect mitochondria in insect cells. These findings indicated that signal transduction plays an important role in the response of M. persicae to UV-B stress.

Conclusions
In summary, in this study, we used RNA-seq for the first time to narrate the genes associated with the adaptation of M. persicae to UV-B stress. The results illustrated that the adaptive mechanism of M. persicae to UV-B stress is complex, mainly involving genes involved in antioxidation, detoxification, metabolic and protein turnover, immune response, and stress signal transduction. Our results clarify the basis of the adaptive mechanism of insects in response to UV-B stress.

Insects rearing
M. persicae was raised by the Institute of Entomology, Guizhou University in chambers at a temperature of 25℃ ± 1℃ and relative humidity of 70%-80% under a 14-h/10-h photoperiod.

UV-B treatment
To exclude the influence of other light sources, M. persicae was fully dark adapted for 2 h, and specimens were then divided into two groups: UV-B radiation and control groups (3 replicates per group, 30 aphids per replicate). In the first group, specimens were irradiated with UV-B (280-320 nm) for 30 min at an intensity of 300 μW/cm 2 . In the second group, specimens were irradiated using light-emitting diode fluorescent lamp bulbs for 30 min at the same intensity. The temperature and humidity during irradiation were consistent with the normal feeding conditions.
Immediately after the end of treatment, insects were quickly frozen in liquid nitrogen and stored at −80℃ until RNA was extracted.

RNA isolation, library construction, and RNA sequencing (RNA-seq)
Total RNA in M. persicae was extracted using TRIzol® according the manufacturer's instructions (Invitrogen), and genomic DNA was removed using DNase I (TaKaRa).
The RNA-seq transcriptome library was constructed using a TruSeq™ RNA sample preparation kit (Illumina, San Diego, CA, USA). Poly-A mRNA was first enriched from 13 5 μg of total RNA using magnetic beads with oligo (dT). Then, fragmentation buffer was added to randomly degrade the mRNA into small fragments of approximately 200 bp. Next, double-stranded cDNA synthesis was performed with mRNA as a template using a SuperScript Double-Stranded cDNA Synthesis Kit (Invitrogen, CA) and a random hexamer primer (Illumina). The double-stranded cDNA structure was blunt-ended by adding End-Repair Mix (Enzymatics, USA), followed by addition of an "A" base at the 3′ end to ligate the Y-shaped link. The specific procedure is described in the specification. After amplifying cDNA via 15 cycles of PCR, a 200-300-bp target band was recovered using 2% agarose gel. After quantification using TBS380 (Picogreen), the library was subjected to high-throughput sequencing using the Illumina HiSeq4000 sequencing platform with a sequencing read length of 2× 150 bp.

Differential expression analysis and functional enrichment
To identify differentially expressed genes (DEGs) between two different samples, fragments per kilobase of transcript per million mapped reads (FPKM) were used to quantify gene expression, and the count of reads was further normalized to the FPKM values. The corresponding significance thresholds for fold change (FC) and Pvalue were estimated using standardized gene expression levels [determined by the 14 control false discovery rate (FDR)]. Based on the expression level, the significance thresholds for DEGs in this study were FDR < 0.05 and FC > 1.5. In addition, the enrichment of DEGs was analyzed using GO and KEGG. GO functional enrichment and KEGG pathway analyses were performed using Goatools (https://github.com/tanghaibao/Goatools) and KOBAS 2.0 (http://kobas.cbi.pku.edu.cn/home.do) [41].

qRT-PCR verification
The 15 annotated unigenes were randomly selected for verification via qRT-PCR.
Total RNA was extracted from specimens from each treatment group using TRIzol.
The primers used for qRT-PCR are shown in Table 3. cDNA was synthesized using a reverse transcription PrimeScript™ RT reagent kit (TaKaRa). qRT-PCR was performed on a C1000 real-time PCR system (Bio-Rad). In total, the 20 μL reaction mixture comprised 1 μL of cDNA (400 ng/μL), 10 μL of LYBR Green Supermix (TaKaRa), 1 μL of each of the primers (10 μmol/L), and 7 μL of ddH 2 O. The 2 − ΔΔCt method was used to analyze the relative differences in transcription levels [42]. Glyceraldehyde-3phosphate dehydrogenase ( GAPDH) and β-actin were used as internal controls, and experiments were performed using three biological replicates. declarations Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
RNA sequencing raw data have been deposited in the NCBI Sequence Read Archive (SRA; accession number PRJNA592018).

Competing interests
The authors declare that they have no competing interests.

Funding
This research was supported by the National Key R&D Program of China (2017YFD0200900) and the National Natural Science Foundation of China (31401754 and 31460483).

Authors' contributions
CLY and CYZ conceived this study. CLY and MSY collected data and data analyses.    Figure 1 Length distribution of the unigenes in the Myzus persicae transcriptome.
25 Figure 1 Length distribution of the unigenes in the Myzus persicae transcriptome.