Molecular Cloning and Characterization of Small Heat Shock Protein Genes in the Invasive Leaf Miner Fly, Liriomyza trifolii

Small heat shock proteins (sHSPs) comprise numerous proteins with diverse structure and function. As molecular chaperones, they play essential roles in various biological processes, especially under thermal stresses. In this study, we identified three sHSP-encoding genes, LtHSP19.5, LtHSP20.8 and LtHSP21.7b from Liriomyza trifolii, an important insect pest of ornamental and vegetable crops worldwide. Putative proteins encoded by these genes all contain a conserved α-crystallin domain that is typical of the sHSP family. Their expression patterns during temperature stresses and at different insect development stages were studied by reverse-transcription quantitative PCR (RT-qPCR). In addition, the expression patterns were compared with those of LtHSP21.3 and LtHSP21.7, two previously published sHSPs. When pupae were exposed to temperatures ranging from −20 to 45 °C for 1 h, all LtsHSPs were strongly induced by either heat or cold stresses, but the magnitude was lower under the low temperature range than high temperatures. Developmentally regulated differential expression was also detected, with pupae and prepupae featuring the highest expression of sHSPs. Results suggest that LtsHSPs play a role in the development of the invasive leaf miner fly and may facilitate insect adaptation to climate change.


Introduction
Liriomyza trifolii (Burgess) (Diptera: Agromyzidae) is an economically important and highly polyphagous pest in fields and greenhouses [1]. Both larvae and adults damage crop plants. The larvae tunnels in the leaves, whereas female adults puncture the leaf tissues for oviposition [2][3][4]. Originating from America, it has rapidly spread throughout the world [5]. In mainland China, L. trifolii was first recorded in Guangdong in 2005 [6], and now it has distributed to more than ten provinces [7]. L. trifolii, L. sativae and L. huidobrensis are the most important vegetable leaf mining pests in China [8][9][10]. L. sativae is the dominant species and has been detected throughout the country, whereas L. huidobrensis only occurs in regions of relatively high altitude [11,12]. L. trifolii mostly occurs in southeast coastal regions, but recently has been identified in certain northern provinces as well [7,13]. Temperature, particularly low temperature, appears to be the most important environmental factor that affects the

RNA Isolation and Cloning Experiments
Total RNA was extracted using the SV Total RNA Isolation System (Promega, Fitchburg, WI, USA) and treated with DNase I to eliminate DNA contamination, following the manufacturer's protocol. Integrity and purity of RNA was determined by agarose gel electrophoresis and spectrophotometry (Eppendorf Bio Photometer plus, Hamburg, Germany). First-strand cDNA was synthesized from 1 µg of total RNA using the First Strand cDNA Synthesis Kit (Fermentas, Ontario, Canada). Three putative sHSP genes, selected based on the analysis of our unpublished transcriptome data, were PCR amplified with gene-specific primers (Table S1). Touchdown PCR conditions were as follows: 94 • C for 3 min, 19 cycles of 94 • C for 30 s, 65-45 • C (annealing temperature decreased by 1 • C /cycle, from 65 • C to a "touchdown" 45 • C) for 30 s, 72 • C for 1 min, and then 25 cycles of 94 • C for 30 s, 45 • C (annealing temperature) for 30 s, and 72 • C for 1 min, followed by extension at 72 • C for 10 min. 5' and 3' rapid amplification of cDNA ends (5' and 3' RACE) were performed to obtain full-length cDNAs using a SMART RACE cDNA Amplification Kit (Clontech, CA, USA) according to the manufacturer's instructions. LA Taq DNA Polymerase (Takara, Japan) was used for the PCR amplification and PCR parameters were as follows: 94 • C for 3 min, 35 cycles of 94 • C for 30 s, 68 • C for 30 s, and 72 • C for 3 min, followed by extension at 72 • C for 10 min. After obtaining the sequence information of the three sHSP genes, specific primers were designed to amplify the full-length of cDNAs. Gene-specific primers are shown in Table S1. The full-length cDNAs were purified using a gel extraction kit (Axygen, New York, NY, USA), cloned into gam-T Easy Vector (Promega, Fitchburg, WI, USA) and subjected to sequencing.

Reverse Transcription Quantitative PCR
Total RNA (0.5 µg) was reverse-transcribed into first-strand cDNA using the Bio-Rad iScript™ cDNA Synthesis Kit (Bio-Rad, CA, USA). The RT-qPCR reactions were performed using a CFX96 Real-Time PCR System (Bio-Rad Laboratories, Berkeley, CA, USA) in 20 µL reaction volume, as previously described [13]. The relative quantifications of LtsHSPs were assessed using the 2 −∆∆Ct method [45] and ACTIN was used as a reference gene, because it is commonly used and the most optimal reference gene in L. trifolii under different experimental conditions [43]. Each sample was assessed in triplicate (technical replicates).

Sequence Alignment of sHSPs and Data Analysis
Full-length cDNA sequences of the three LtsHSPs were used as queries to search for other homologous sHSPs using the BLAST programs (http://www.ncbi.nlm.gov/BLAST/). Sequence alignments were conducted using Clustal X software [46], and the open reading frames (ORFs) were identified with ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/). The deduced amino acid sequences of sHSPs were analyzed by ExPASy (Swiss Institute of Bioinformatics, Switzerland). The phylogenetic relationship of sHSPs was generated by MEGA 6.0 [47], using a neighbor joining (NJ) method based on the Poisson correction model with a bootstrap value of 1000. The protein structure of sHSP genes was predicted by the SWISS-MODEL (https://www.swissmodel.expasy.org/).
Data were analyzed with one-way ANOVA, followed by Tukey's multiple comparison and analysis with SPSS v. 16.0 (SPSS, Chicago, IL, USA). For ANOVA, data were transformed for homogeneity of variance tests. Differences were considered statistically significant when P < 0.05.
To examine the phylogenetic relationships between various sHSPs, the phylogenetic tree was generated using ten full-length sHSP family members, selected from three congener Liriomyza species using the neighbor-joining method. LtHSP20.8 was grouped with LsHSP20.8 in a separate branch, and LtHSP19.5 and LtHSP21.7b were clustered into a larger group that contained several other orthologs genes, as seen in Figure 3.

Expression of Three LtsHSPs in Response to Temperature Treatments
The relative mRNA levels of the three sHSPs were observed at different temperature stresses. In cold stress, compared with the control group at 25 • C, the expression level of three LtsHSPs were significantly increased after low temperature treatment (LtHSP19.5: F 10, 33 = 23.514, P < 0.001; LtHSP20.8: F 10, 33 = 29.116, P < 0.001; and LtHSP21.7b: F 10, 33 = 4.245, P < 0.05). Gene expression peaked at -17.5 • C, which were 11.12-, 26.27-and 12.16-fold increase relative to the control, as seen in Figure 4. . Relative expression levels of LtHSP19.5, LtHSP20.8, and LtHSP21.7b under low temperature treatments. The relative level of HSP expression represented the fold increase as compared with the expression in controls. The data were denoted as mean ± SE. One-way analysis of variance (ANOVA) was used to analyze the relative expression levels of three sHSPs under low temperature treatments. For the ANOVA, data were tested for homogeneity of variances and normality. Different lowercase letters indicate significant differences among different temperature treatments. Tukey's multiple range test was used for pairwise comparison for mean separation (P < 0.05).

Expression of Three LtsHSPs at the Developmental Stages
We investigated the mRNA level of three sHSPs through the developmental stage of L. trifolii, including 3rd instar larvae, prepupae, 2-day-and 10-day-old pupae, and male and female adults. All three LtsHSPs showed expression variations throughout the developmental stages. The expression of LtHSP19.5 and LtHSP21.7b peaked at prepupae, which were significantly up-regulated by 98.10and 42.27-fold relative to the control-male adults (LtHSP19.5: F 5, 12 = 338.747, P < 0.001; LtHSP21.7b: F 5,12 = 54.757, P < 0.001). Expression of LtHSP20.8 peaked at 2-day-old pupae (up-regulated by 6.83-fold; F 5, 12 = 6.640, P < 0.05) as seen in Figure 6. Relative expression levels of LtHSP19.5, LtHSP20.8, and LtHSP21.7b under high temperature treatments. The relative level of HSP expression represented the fold increase as compared with the expression in controls. The data were denoted as mean ± SE. One-way analysis of variance (ANOVA) was used to analyze the relative expression levels of three sHSPs under high temperature treatments. For the ANOVA, data were tested for homogeneity of variances and normality. Different lowercase letters indicate significant differences among different temperature treatments. Tukey's multiple range test was used for pairwise comparison for mean separation (P < 0.05). . The data were denoted as mean ± SE. One-way analysis of variance (ANOVA) was used to analyze the relative expression levels of three sHSPs in different developmental stages. For the ANOVA, data were tested for homogeneity of variances and normality. Different lowercase letters indicate significant differences among different developmental stages. Tukey's multiple range test was used for pairwise comparison for mean separation (P < 0.05). Abbreviations: FM = females adult; M = males adult; L = third instar larvae; PP = prepupae; P = two-day-old pupae; and OP = ten-day-old pupae.

Comparative Characteristics and Expression Pattern of Five LtsHSPs
We have previously characterized the expression pattern of LtHSP21.3 under temperature stresses [13]. In this study, for comparison, the expression pattern of LtHSP21.3 was also investigated at different developmental stages. As seen in Figure S1, the highest expression stage of LtHSP21.3 occurred at 2-day-old pupal stages (F 5, 12 = 13.935, P < 0.001), similar to that of LtHSP20.8, but different from other sHSPs in L. trifolii.

Discussion
In this study, three new sHSP-encoding genes were cloned from L. trifolii. Sequence analysis shows that all three predicted protein sequences contain the characteristic α-crystalline domain. In addition, LtHSP20.8 and LtHSP21.7b also contained V/IXI/V motif. The propensity of the IPI/V motif to form multiple inter-subunit interactions may contribute to the diversity in structure and function seen in the α-crystallin [27]. Moreover, the 3'UTRs of LtsHSPs contain several other typical motifs, such as the poly adenylation signal (AATAAA or ATTAAA) [48] and the AT-rich element (ATTTA), which have been shown to afford greater mRNA stability and to contribute to the maintenance and re-establishment of basal levels of gene expression [42,[49][50][51]. The number of TA-rich regions in 5'UTR of those three LtsHSPs were different; only two TA-rich regions were found in the 5'UTR of LtHSP19.5, but it was lacking in LtHSP20.8 and LtHSP21.7b. The difference in the number of these elements have also been found in HSPs of other insect species [41], and the number of these elements may be related to the expression pattern of HSPs [52][53][54]. The phylogenetic analysis showed that LtHSP19.5 and LtHSP20.8 clustered with representatives of their orthologs, but LtHSP21.7b was clustered with HSP21.3. This clustering pattern was also recorded in previous studies [23,55], which suggested that the evolution of sHSP may be complex. However, the available sHSP data are limited, and the systematic illustration of the evolution of sHSPs is expected to be achieved through the genome-wide analysis in future.
In this study, three new LtsHSPs and two previously identified LtsHSPs are all significantly up-regulated by low and high temperature treatments. The same expression patterns have also been observed in most previously recorded responses of sHSPs [30,[56][57][58]. The sensitivity to temperature stresses of five LtsHSPs was different but the temperatures of the maximal (Tmax) expression were comparable. The same highest expression temperature range suggests similar functionality under temperature stresses, except for LtHSP21.7 at high temperatures. In this study, the response level of three sHSPs to different temperature stresses was higher than that of two published sHSPs (LtHSP21.3 and LtHSP21.7), which was reflected by a higher gene expression fold. The five LtsHSPs varied in terms of temperature sensitivity and suggest a synergistic effect of different sHSP family members with respect to thermal tolerance, which is consistent with recent research of LtHSP70s [59]. Several sHSPs play a role in temperature stress together, and similar expression patterns were also found in Chilo suppressalis and Bemisia tabaci [31,60]. At the same time, it is worth noting that, like in cases of other HSPs, expression level of LtHSPs induced by high temperatures is higher than that induced by low temperatures. In addition, other mechanisms for confronting with the stress may play a role in resisting extreme low temperature stress, such as antioxidation and supercooling phenomenon.
Attention has been paid to the role of HSPs in the regulation of insect development [35]. It seemed that all three new LtsHSPs and two previous LtsHSPs could be expressed in all developmental stages in L. trifolii. However, all of the five sHSPs' expression levels were significantly different in the developmental stages, and the expression fold of LtHSP19.5, LtHSP21.3, and LtHSP21.7 [43] in prepupae was significantly different from that in control, while LtHSP20.8 and LtHSP21.7b reached a peak in the pupal stage. For prepupae, the mature larvae just leave the leaves for pupation, so it may lead to more sensitivity to the environment temperature. Meanwhile, the expression level increased significantly during the transition from larvae to pupae, which is the same as the one observed in LtHSP70s [59]. This suggests that metamorphosis itself can serve as a factor to induce the expression of HSPs. The metamorphosis has large physical changes and HSPs, as important molecular chaperones, are involved in processes such as assembling, removing, folding, and refolding of different kinds of proteins [19]. Thus, the change of protein structure may lead to a relatively high expression level of HSPs [42].

Conclusions
In summary, genes encoding sHSPs of L. trifolii contained several typical conserved domain and motifs. Those LtsHSPs could be significantly induced by temperature stresses and expressed during different stages of insect development. This study provides further insights into physiological responses of L. trifolii under climate change, and the mechanisms of distribution of leaf miner flies in response to temperature. The data were denoted as mean ± SE. One-way analysis of variance (ANOVA) was used to analyze the relative expression levels of three sHSPs in different developmental stages and under temperature treatments. For the ANOVA, data were tested for homogeneity of variances and normality. Different lowercase letters indicate significant differences among different temperature treatments. Tukey's multiple range test was used for pairwise comparison for mean separation (P < 0.05). Abbreviations: FM = females adult; M: males adult; L: third instar larvae; PP: prepupae; P: two-day-old pupae; OP: ten-day-old pupae, Table S1: Primers used in the cDNA cloning and real-time quantitative PCR.