Cloning and expression of genes encoding heat shock proteins in Liriomyza trifolii and comparison with two congener leafminer species

The polyphagous agromyzid fly, Liriomyza trifolii, is a significant and important insect pest of ornamental and vegetable crops worldwide. The adaptation of insects to different environments is facilitated by heat shock proteins (HSPs), which play an important role in acclimation to thermal stress. In this study, we cloned and characterized five HSP-encoding genes of L. trifolii (Lthsp20, Lthsp40, Lthsp60, Lthsp70, and Lthsp90) and monitored their expression under different thermal stresses using real-time quantitative PCR. Pupae of L. trifolii were exposed to 19 different temperatures ranging from -20 to 45°C. The results revealed that Lthsp20, Lthsp40, Lthsp70 and Lthsp90 were significantly upregulated in response to both heat and cold stress, while Lthsp60 was induced only by heat temperatures. The temperatures of the onset (Ton) and maximal (Tmax) expression of the five Lthsps were also determined and compared with published Ton and Tmax values of homologous genes in L. sativae and L. huidobrensis. Although L. trifolii occurs primarily in southern China, it has cold tolerance comparable with the other two Liriomyza species. Based on the heat shock proteins expression patterns, L. trifolii has the capacity to tolerate extreme temperatures and the potential to disseminate to northern regions of China.


Introduction
Liriomyza trifolii is an economically important invasive insect pest in China [1]. It was initially discovered in Guangdong in 2005 [2] and has since proliferated throughout the southern region of China [3]. Both larvae and adults of L. trifolii can cause damage to crop plants. The larvae mine tunnels in the leaf tissues, and female adults puncture the leaf tissues for oviposition. Both activities can reduce photosynthesis and increase leaf drop, resulting in lower crop a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 quality and yield [4][5]. In recent years, L. trifolii has spread rapidly throughout the country, causing significant damage to various vegetable and horticultural crops [6][7][8].
Insects are poikilothermic organisms, and their physiological activities can be greatly affected by temperatures [9]. The tolerance of insects to temperature stress is a definitive factor in their survival [10][11]. Multiple studies have shown that insect tolerance to thermal stress is multifactorial and has genetic, physiological, and biochemical components [12][13][14][15][16][17]. Insects exposed to temperature stress may exhibit alterations in behavior, such as seeking shelter. Additionally, changes in morphology, life history and physiological characteristics, which include changes in membrane fluidity, the accumulation of carbohydrate alcohols, and the generation of heat shock proteins (HSPs) and antioxidant enzymes are also shown to effect in tolerate extreme temperatures [18][19]. Alternations of hsps expression in insects as affected by temperature stress has been widely studied and is one of the best predictors of insect tolerance to temperature stress [20].
Previous studies have examined the response of hsp90, hsp70, hsp60, hsp40 and hsp20 to temperature stress in L. sativae and L. huidobrensis [27], and hsp90 and hsp70 in L. trifolii [28][29]. However, the expression profiles of hsp60, hsp40 and hsp20 in L. trifolii during temperature stress has not yet been investigated. In this study, we characterized the five hsps in L. trifolii, hsp90, hsp70, hsp60, hsp40 and hsp20 to better understand hsp expression in response to both high and low temperature stress. In addition, we also compared the expression of hsps in L. trifolii with the homologous genes in L. sativae and L. huidobrensis, which provides insights into the competition between Liriomyza spp. and the distribution and dissemination of leaf mining insects in response to temperature.

Study insect
L. trifolii were originally collected on celery in Yangzhou (32.39˚N, 119.42˚E) in 2010 and reared on beans in the laboratory at 26˚C with a 16:8 h (L: D) photoperiod as described by Chen & Kang [30]. Beans (Vigna unguiculata) were seeded at the rate of 5-6 plants per pot (12 cm in diameter) and moved into cages (40×40×65cm) for insect feeding when plants had five to six true leaves. About 150 adults were reared per cage and the larvae inside the tunneling leaves were collected in plastic bags until pupation. The pupae were collected in glass tubes and no field populations were added during experimental period. No specific permissions were required for these activities and the field studies did not involve endangered or protected species.

RNA isolation and cloning experiments
Total RNA was extracted from L. trifolii using the SV Total RNA isolation system (Promega, USA). The integrity and purity of RNA was determined by agarose gel electrophoresis and spectrophotometry (Eppendorf Bio Photometer plus, Germany). Total RNA (1 μg) was transcribed into cDNA using oligo (dT) primers. Degenerate primers (Table 1) were used to amplify partial segments of the five hsps, and then 5 0 and 3 0 RACE were utilized to obtain the full-length cDNAs as recommended by the manufacturer (SMART RACE cDNA Amplification Kit, Clontech, USA).
Quantitative real-time reverse transcriptase PCR (qRT-PCR) RNA (0.5 μg) was reverse-transcribed into first-strand cDNA using the Bio-Rad iScript™ cDNA Synthesis Kit (Bio-Rad, CA, USA). Reactions were conducted in a 20 μl reaction volume consisting of 10 μl Bio-Rad iTaq™ Universal SYBR 1 Green Supermix (2×), 1 μl of each genespecific primer (10 μM) ( Table 1), 2 μl of each cDNA template, and 6 μl of ddH 2 O. Real-time PCR were performed using an Applied Biosystems 7500 real-time PCR system (Thermo Fisher Scientific, USA) under the following conditions: 3 min at 95˚C, 40 cycles of denaturation at 95˚C for 30 s, and annealing at the T m of primer pairs (Table 1) for 30 s. Each treatment contained four replications, and each reaction was run in triplicate. β-actin was cloned from L. trifolii (GenBank accession no: KY231150) and used as a reference gene.

Sequence alignment and data analysis
Full-length cDNAs sequences of the five Lthsps were used as queries to search for other insect hsps using the BLAST programs available at the NCBI website (http://www.ncbi.nlm.gov/ BLAST/). Sequence alignments were conducted using Clustal X software [31]. Open reading frames (ORFs) were identified using ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/). Sequence analysis tools of the ExPASy Molecular Biology Server (Swiss Institute of Bioinformatics, Switzerland) were used to analyze the deduced hsps sequences. The 2 −ΔΔ Ct method was used to evaluate fold changes in mRNA expression levels [32]. Geometric means of the reference genes were utilized to normalize expression under different experimental conditions. One-way ANOVA was used to detect significant differences in mRNA levels among treatments, followed by Tukey's multiple comparison (P<0.05) in SPSS v. 16.0 (SPSS, Chicago, IL, USA). For ANOVA tests, original data were log-transformed for homogeneity of variances. The temperature where expression was significantly higher than that at 25˚C was designated as the onset temperature (T on ), whereas the temperature where expression was significantly higher than that of other temperatures was denoted as T max .
The alignment of LtHSP20 with sHSPs from L. sativae and L. huidobrensis revealed a conserved region in the middle, which constitutes an α-crystalline domain (Fig 1A). The N-and , which constitute the most conserved region of HSP40, comprise the DnaJ domain ( Fig 1B). LtHSP60 showed a high degree of identity to related proteins in L. sativae and L. huidobrensis (95.63 and 95.98% identity, respectively). LtHSP60 contained a conserved GGM motif at the C-terminal end (Fig 1C). Multiple ATP/Mg 2+ binding sites were distributed throughout the predicted protein product in L. trifolii, which were consistent with the structure of HSP60s in the two other Liriomyza spp. Amino acid alignments revealed that LtHSP70 was closely related to analogous proteins in L. sativae and L. huidobrensis, which showed 99.06 and 95.96% amino acid identity, respectively. Similarly, HSP90 in L. trifolii showed a high degree of identity relative to that in L. huidobrensis and L. sativae (97.34 and 99.30%, respectively). Conserved EEVD motifs were identified in the C-terminal ends of LtHSP90 and LtHSP70 (Fig 1D and 1E).
The relative mRNA levels of Lthsps were compared and the onset (T on ) and maximal (T max ) temperature values were identified. Under cold temperature stress, the T on values of these five Lthsps were all in -7.5˚C and the T max values were -17.5˚C, except for Lthsp90, which peaked at -15˚C. In response to heat stress, the T on values were 32.5˚C for Lthsp20, 40˚C for Lthsp40 and Lthsp90, and 37.5˚C for Lthsp70. The T max was 40˚C for Lthsp60 and Lthsp20 and 42.5˚C for the other three Lthsps (Fig 2).

Interspecific differences in hsps
A total of ten TATA-box-like regulatory elements were identified in the 5' untranslated regions (5'UTRs) of the five Lthsps. In comparison, five and eleven TATA-box-like elements were identified in the 5'UTRs of hsps in L. sativae and L. huidobrensis, respectively (Fig 3). Liriomyza huidobrensis contained a single TATA-box-like element in the 5'UTR of hsp20, which was not present in the other two Liriomyza spp. (Fig 3A). The 5'UTR of hsp40 contained three, four, and one TATA-box in L. trifolii, L. huidobrensis, and L. sativae, respectively (Fig 3B). In hsp60, L. trifolii and L. sativae possessed a single TATA-box-like element but L. huidobrensis had four (Fig 3C). Five TATA-box-like elements were found in Lthsp70, whereas L. huidobrensis and L. sativae contained one and two, respectively (Fig 3D). All three leafminers contained a single TATA-box-like element in the 5' UTR of hsp90 (Fig 3E). The first temperature where expression was significantly higher than that the control (25˚C) was described as the onset temperature (T on ) or the hsp, and the temperature at which the expression level was significantly higher than expression at other temperatures was denoted as T max . T on and T max are marked by arrows (!), and the notable temperature shifts of T on and T max are indicated on the curves. The relative level of hsp expression represented the fold increase as compared with the expression in controls. The data were denoted as mean ± SE. Under cold stress, the T on and T max values of the five hsps in L. trifolii and L. huidobrensis were 2.5 to 7.5˚C lower than that in L. sativae ( Table 3). The T on and T max values were generally similar among the three leafminer species when exposed to heat stress, with the exception of T on for hsp90, which varied from 30 to 40˚C in the three leafminer species (Table 3).

Discussion
Heat shock proteins function in various biological and physiological processes and may be produced in response to temperature, starvation, or disease [33][34][35][36][37]. In this study, we showed that the coding regions of five L. trifolii hsps are highly conserved relative to that in L. huidobrensis and L. sativae. The C-terminal ends of LtHSP90 and LtHSP70 both contain EEVD motifs, which is consistent with their role as molecular chaperones for interaction with other proteins [38]. LtHSP60 also contained C-terminal (GGM) n repeats, which are typical of mitochondrial forms of HSP60 [39]. However, the other Liriomyza HSPs are likely located in the cytosol. LtHSP40 contained a DnaJ domain near the N-terminus, which is consistent with its function in ATPase activity and role as a co-chaperone with HSP70 in multiple processes (e.g. protein folding, trafficking, assembly, and dissociation) [40][41]. The central portion of LtHSP20 contained an α-crystalline domain like other sHSPs, may have essential functions in various processes including diapause and insect immunity [42][43].  Although the coding regions of Liriomyza species hsps are highly conserved, the nucleotide sequences of the 5'UTRs in the Lthsp were different from the other two Liriomyza spp. Regulatory elements in the hsp promoter regions play important roles in hsp expression and may be a contributing factor in establishing specific patterns of hsp expression [44][45]. Ten TATA-boxlike elements were identified in the 5'UTRs of the five Lthsps, and the number of 5'UTRs and their locations varied among the three Liriomyza spp. (Fig 3). Data on different T on and T max values of hsps expression in the three Liriomyza may explain the differences in the ability of three Liriomyza species to tolerate cold and heat stress. Although L. huidobrensis prefers cold climates [46], L. trifolii may be the most cold-tolerant when compared to the other two Liriomyza species based on hsps expression. The differences of heat tolerance among the three species were relatively small. L. trifolii and L. sativae may have comparative thermotolerance than L. huidobrensis according to hsps expression pattern.
Furthermore, the super cooling points (SCPs) of L. trifolii and L. huidobrensis were less than -20˚C, which was much lower than that of L. sativae (-11.7˚C) [30,[47][48]. SCP is an important predictor for cold tolerance [49][50][51] and field populations of Liriomyza appear to enhance their cold tolerance by depressing the SCP of the puparial stage [1,52]. The SCP value for L. trifolii was low enough to enable the leafminer to safely overwinter in most regions of China. However, L. trifolii is primarily distributed in southern China, thus further research is needed to discover underlying reasons for its southern distribution patterns. Other studies have shown that the developmental threshold temperature and effective accumulated temperature of the three leafminers were different [53][54][55]. The developmental threshold temperature of each life stage of L. trifolii was lower than that of L. sativae. Therefore, the first generation of L. trifolii should occur earlier than L. sativae. In addition, the effective accumulated temperature of L. trifolii was higher than that of L. sativae, resulting in the longer dissemination period of L. trifolii relative to L. sativae. In contrast, L. huidobrensis occurs at relatively high latitudes and altitudes. The developmental threshold temperature of L. huidobrensis was the lowest among the three leafminers, and the effective accumulated temperature was in-between L. sativae and L. trifolii. The earlier occurrence of the first generation and longer dissemination in L. trifolii suggest that this pest could probably expose to cold climates in spring/post-winter and autumn/pre-winter, and its relatively lower cold tolerance may play an important role in the survival and development of the population. [1,[56][57].
Although it has the potential to be widely dispersed throughout the country, L. trifolii was currently limited to southern China [3,58]. There may be several reasons for the geographical limitation of L. trifolii. Liriomyza sativae and L. huidobrensis were first identified in China in 1994 [59][60] while L. trifolii was first reported eleven years later in 2005. Liriomyza trifolii may not have had enough time to disperse into other regions of China. In addition, L. sativae has a higher reproductive capacity than L. trifolii [61] and may not be as prolific as L. sativae. Other possible reasons include the occurrence of natural enemies, pesticide resistance, and the availability of host plants [61][62][63][64][65][66][67][68][69].
With the rapid development of facility agriculture in China, L. sativae has been displaced by L. trifolii in the southern region of the country. In Hainan province, L. trifolii has become the dominant species, and it constitutes about 95% of the leafminer population in the city of Sanya [70]. Field investigations of Liriomyza spp. have revealed that the damage caused by L. trifolii has become a more serious pest in the northern part of Jiangsu from 2008 to 2015 [71][72]. In the context of predictions of global warming, and the development of facility agriculture and frequent trade exchange, it is high possible that the range of L. trifolii will expand in China. Therefore, research is critical regarding the factors affecting the distribution of L. trifolii in China.