DU Hui,SUN Li-li,LIU Peng,CAO Chuan-wang

Key Laboratory of Sustainable Forest Ecosystem Management of Ministry of Education,School of Forestry,Northeast Forestry University,Harbin 150040,P.R.China

Abstract The G protein-coupled receptor (GPCR) regulates downstream genes by binding to a heterotrimeric G protein.However,the function of sex peptide receptor (SPR) in lepidopteran species is mostly unknown.Understanding the physiological functions of SPR in insects is essential for exploring new insecticidal targets.In the present study,the functions of an SPR in Lymantria dispar (Asian gypsy moth;LdSPR) were investigated.The expression of LdSPR was the highest in the 6th instar larval stage,and there was a large difference in expression between male and female adults.After LdSPR gene silencing,L.dispar larvae showed increased sensitivity to high temperature,starvation,and oxidative stress,indicating that LdSPR enhances stress resistance.These results enrich our knowledge of the function of the insect SPRs,which will lead to a better understanding of other insect GPCR family members and the identification of new targets for the development of environmentally friendly pesticides.

Keywords:Lymantria dispar,sex peptide receptor,RNAi,physiological function,stress resistance

1.Introduction

Across the evolutionary spectrum,from mating yeast to humans,various physiological functions are controlled by G protein-coupled receptors (GPCRs) (Herskowitz 1995;Morris and Malbon 1999).Signals from extracellular ligands are passed to intracellular signal transduction proteinsviaGPCRs (Straderet al.1994).InDrosophila,mutations in the GPCR methuselah gene extend the adult lifespan,revealing the role of the G-protein signaling system in aging (Linet al.1998).The sex peptide receptor (SPR) is a member of GPCR family A and can activate trimeric G proteins with either Gαi or Gαo subunits when heterologously expressed in cell culture (Poelset al.2010).SPR interacts with two evolutionarily unrelated ligands,the sex peptide (SP)and the myoinhibiting peptide (MIP).SP and MIP control reproductive behaviors and sleep behavior,respectively(Kimet al.2010;Poelset al.2010).The induction of female post-mating behavioral responses (PMR),e.g.,initiation of egg laying and suppression of mating receptivity,occursviamultiple SP pathways that involve SP and its receptor SPR(Haussmannet al.2013).

InDrosophila,the 36 amino acids of SP is the primary regulator of the post-mating response in females,and is transmitted by the maleDrosophilaalong with sperm and accessory gland fluid during copulation (Chenet al.1988;Liu and Kubli 2003).Increased egg production and oviposition,and rejection of courting males,are then induced by SP (Solleret al.1999);however,it can also lead to other behavioral and physiological changes,such as increased feeding,immune stimulation,and altered food choices and sleep patterns (Penget al.2005;Carvalhoet al.2006;Domanitskayaet al.2007;Isaacet al.2010;Ribeiro and Dickson 2010).SP provides significant benefits to males in terms of sperm competition by reducing the receptivity of mating females and increasing egg production rates,which can promote male reproductive success and increase paternity (Chapmanet al.2003;Frickeet al.2009;Tsuda and Aigaki 2016).The high-affinity receptor for SP,SPR,has been identified and its presence is necessary for late switching,resulting in reduced receptivity (easy to mate) and increased oviposition (Yapiciet al.2008).AmongDrosophilaspecies,it appears that the SP/SPR system is highly variable.Genome sequencing revealed that among 12Drosophilaspecies,the gene encoding SP is missing inD.mojavensisandD.grimshawi,suggesting either that the SP gene has been lost during evolution or that the SP-dependent system is not present in these two species (Tsudaet al.2015).Additionally,the presence of additional copies of SP homologs suggests an enhanced SP/SPR-mediated system in some species (Kimet al.2010).The evolutionarily conserved MIP-like peptide family is an ancestral ligand for SPR (Kimet al.2010).The evolutionarily conserved MIPs activate SPRs.In MIPs and SPs,the structural determinants that affect this dual receptor activation have been characterized.InDrosophila,SPR is expressed in embryonic and larval stages,and in the nervous system of the adult male;however,SP is only expressed in the male reproductive system.SPR has an important function in female reproduction and in sleep stabilization in both males and females (Ohet al.2014).

Methuselah (MTH) is a GPCR associated with longevity inDrosophila,and MTH downregulation can extend theDrosophilalife-span,e.g.,using MTH signal transduction inhibitors (Jaet al.2007).Mthhypomorphic mutants showed a 35% increase in their average life-span and improved resistance to starvation,high temperature,and oxidative stress inDrosophila(Linet al.1998).MTH peptide agonists includeStunted(Sun),andDrosophilaSP and the peptide serendipitous peptide activator of MTH also activates MTH signaling (Tuet al.2000;Jaet al.2009).InLymantria dispar,our previous study showed thatLdmthl1plays a complex and critical role in longevity regulation (Caoet al.2019).SP is an SPR ligand;therefore,we speculated that SPR and MTH have similar functions.Further study of SPR will deepen our understanding of the functional interactions among agonists and GPCRs.

The Asian gypsy moth,Lymantriadispar(Lepidoptera:Erebidae),damages more than 500 plants and is widely distributed worldwide.The Asian gypsy moth seriously affects forest plants,especially birch,poplar,and oak (Royet al.1995;Lazarevicet al.1998).This moth has been responsible for defoliating approximately two million acres of forest annually during the past 20 years in northern and eastern China (Sunet al.2014).A number of strategies have been employed to manage gypsy moths,such as chemical control,biological control (e.g.,usingBacillus thuringiensis),and transgenic engineering of host plants(Houet al.2009;Caoet al.2010).However,environmentally friendly insecticides increasingly require avoidance of the“resistance,resurgence,residue”(3Rs) of chemical insecticides (Audsley and Down 2015).In the present study,we investigated the physiological functions of a GPCR family member,SPR,and provided a molecular target for the development of novel insecticides using the“reverse pharmacology”method.

2.Materials and methods

2.1.Insects

TheL.dispareggs and artificial diet were obtained from the Research Institute of Forest Ecology,Environment and Protection,Chinese Academy of Forestry (Beijing,China).In this research,L.disparwas maintained in the laboratory for more than five years without exposure to pesticides.Lymantria disparlarvae were reared in an insect incubator with a relative humidity of 75% and a photoperiod of 16 h:8 h(light:dark) at (25±1)°C.The healthy 3rd instar larvae ofL.disparwere used for the dsRNA microinjection and bioassays experiments.Eggs,larvae (1st-6th instars),pupae and adults were sampled on the first day of those developmental stages,and then snap-frozen using liquid nitrogen,and stored at -80°C until RNA extraction.

2.2.Identification of LdSPR and bioinformatics analysis

Total RNA was isolated using a RNeasy Mini Kit (Qiagen,Valencia,CA,USA) following the manufacturer’s guidelines,and then treated with RNase free DNase I (Ambion,Austin,TX,USA).TheL.dispartranscriptome was profiled using Illumina Solexa sequencing by the GAII platform at the Beijing Genomics Institute (BGI,Shenzhen,China) (Caoet al.2015).TheLdSPRgene was identified according to its functional annotation and was further confirmed using reverse transcription PCR (RT-PCR) and sequencing.The open reading frame (ORF) ofLdSPRwas identified by ORF finder program from the NCBI (https://www.ncbi.nlm.nih.gov/orffinder/).The ProtParam Software (http://au.expasy.org/tools/protparam.html) was used to predict the isoelectric point and molecular weight of LdSPR.The conserved domain program from the NCBI was used to predict the conserved regions of LdSPR and examine its domains and functional motifs.The ProtScale program was used to predict protein hydrophobicity and the TMHMM2.0 Server program was used to predict transmembrane regions.The BLAST Program from the NCBI was used to perform sequence similarity searches,then the amino acid sequences of GPCR proteins with similar sequences among different insects were selected,and the program Clustalx(1.83) was used to create multiple sequence alignments.The MEGA 7.0.21 Software was used to generate a phylogenetic tree by the neighbor-joining algorithm.

2.3.Quantitative real-time reverse transcription PCR(qRT-PCR)

The RNeasy Mini Kit (Qiagen) was used to extract total RNA from insect samples.DNase I (Ambion Inc.,Austin,TX,USA) was used to remove genomic DNA contamination from the total RNA.cDNA was produced from the total RNA(approximately 0.5 µg)viareverse transcription using 1 µmol L-1of oligo(dT) primers.The synthesized cDNA (10 µL) was diluted 10-fold with sterile water and used as a template for quantitative real-time reverse transcription PCR (qRT-PCR)using an MJ Opticon™2machine (BioRad,Hercules,CA,USA).TheL.disparreference genes used comprisedEF1α(MK926771),TUB(MK926772),andActin(MK926773)(Sunet al.2014).The qRT-PCR reaction mixture (20 µL)comprised forward and reverse primers (0.5 µmol L-1each),SYBR Green Real-time PCR Master Mix (10 µL;Toyobo,Osaka,Japan),and cDNA template (2 µL representing 100 ng of total RNA).The amplification program comprised:94°C for 30 s;followed by 45 cycles of 94°C for 12 s,60°C for 30 s,72°C for 40 s,and 1 s at 82°C for plate reading.The purity of the amplicon was assessed using the melting curve generated at the end of the run.Relative gene expression was calculated using the threshold cycle with the 2-ΔΔCT method (Pfaffl 2001).The qRT-PCR was carried out in triplicate biological repeats to ensure the reproducibility of the results.

2.4.RNA interference (RNAi) in L. dispar larvae

RNAi was used to investigate the function ofLdSPR.A MEGAscript T7 High Yield Transcription Kit (Ambion) was used to synthesize theLdSPRdsRNA (dsLdSPR,559 bp)using the full-lengthLdSPRcDNA as a templateinvitro.A green fluorescent protein (GFP) dsRNA (dsGFP) was generated from plasmid pMW1650 as the template (kindly donated by Dr.Nannan Liu,Auburn University,Alabama,USA).The specific primers were designed to incorporate the T7 promoter (Appendix A).Water-saturated phenol/chloroform was used to purify the dsRNA,which was then precipitated with ethanol and diluted to 1-2 µg µL-1using nuclease-free water.The dsRNA (1 µg) was microinjected into the penultimate section of the abdomen of 3rd instar larvaeviaa sterile 33-gauge needle (MICROLITER™ #65,Hamilton Co.,Reno,NV,USA).dsGFP(1 µg) was injected as a control.Anesthesia using CO2was employed during the microinjection ofL.disparlarvae.Next,qRT-PCR was used to assess the effect of dsRNA microinjection onLdSPRmRNA levels in larvae after 24,48,96,and 120 h.Ten larvae were used for each replicate,and three replicates were performed.

2.5.Stress resistance analysis of L.dispar larvae after RNA silencing

To identify whether theLdSPRgene ofL.disparis related to stress tolerance resistance,3rd instarL.disparlarvae microinjected with dsLdSPRwere subjected to high temperature stress,starvation stress,and oxidative stress.For high temperature stress,theL.disparlarvae microinjected with dsLdSPRor dsGFPwere reared in an insect incubator with a relative humidity of 75% and a photoperiod of 16 h:8 h (light:dark) at 36°C.TheL.disparlarvae were fed on fresh artificial diet,and the cumulative mortality was assessed every 3 h.For starvation stress,theL.disparlarvae microinjected with dsLdSPRor dsGFPwere reared in an incubator with a relative humidity of 75%and a photoperiod of 16 h:8 h (light:dark) at (25±1)°C without feeding any diet.The cumulative mortality was assessed every 3 h.TwentyL.disparlarvae were used for each replicate,and three replicates were performed.For oxidative stress,bioassays were conducted by feedingL.disparlarvae with paraquat-treated artificial diets containing different concentrations of paraquat (30,15,7.5,3.25,1.63 and 0 mg L-1).Paraquat acts as a free radical generator and induces oxidative damage,which is implicated in neuronal cell death (Vettrainoet al.2001).Paraquat is a classic source of exogenous peroxide radicals used to generate oxidative stress inDrosophila(Arkinget al.1991;Linet al.1998).The cumulative mortality of microinjectedL.disparlarvae was determined after 48 h of paraquat treatment.The LC10,LC30and LC50for 3rd instar larvae were calculated using Polo Software (LeOra Software Inc.USA).TenL.disparlarvae were used for each replicate,and three replicates were performed.

2.6.Measurement of individual performance

The wild-type and silencedL.disparlarvae were reared on the artificial diet and their performances were assessed by determining the following traits:cumulative mortality,fresh weight,and developmental durations of the 3rd and 4th instars.During a period of eight days,development and total survival were determined daily.Cumulative mortality was calculated as follows:

whereXiis the number of dead larvae at theith day,andNis the total number of larvae.

For the normal and silenced treatment groups,at least five replicates were conducted,and each replicate comprised 10 larvae.

2.7.Nutrition utilization

The dsRNA ofGFPandLdSPR(1 µg) were microinjected into 3rd instarL.disparlarvae (n=40 in each group).After feeding for eight days,the 4th instarL.disparlarvae were starved for 12 h and divided into six groups for nutrition determination.The initial dry weight of each larva was estimated by setting aside aliquots at the beginning of the experiment and then calculating a fresh/dry conversion ratio.The dry masses of diet,larva,and larval feces were determined by drying at 50°C for 24 h and then to a constant weight at 120°C.The calculation of nutritional indicators was as follows,according to the method of Waldbauer with modifications (Waldbauer 1964).

Cumulative body weight growth rate (%)=[(Weight after injection-Body weight before injection)]×100/Body weight before injection

Relative efficiency of the consumption rate (RCR;%)=(A-B)/[(C+D)/2]×100

Relative growth rate (RGR;%)=(D-C)/[(C+D)/2]×100

Efficiency of digested food conversion (ECD;%)=(D-C)/(A-B-E)×100

Efficiency of ingested food conversion (ECI;%)=(D-C)/(A-B)×100

Approximate digestibility (AD;%)=(A-B-E)/(A-B)×100

In these formulae,A and B are the dry masses of the diet before and after treatment,respectively;C and D are the dry masses of the larvae before and after treatment,respectively;and E is the dry mass of larval feces.Each replicate comprised 20L.disparlarvae and each treatment was repeated three times.

2.8.Statistical analysis

In the figures,data are represented as the mean±SD.Student’st-test (two-tailed unpairedt-test) was used to compare the differences between the two treatments.Oneway analysis of variance (ANOVA) followed by Student Newman-Keuls multiple comparisons test were used to calculate the differences among multiple samples.The statistical analyses were performed using GraphPad InStat version 3.05 (GraphPad Software,Inc.La Jolla,CA,USA).P＜0.05 indicated statistical significance.

3.Results

3.1.Characterization of LdSPR in L.dispar

The ORF ofLdSPRcomprised 1 263 bp,encoding a putative protein of 420 amino acids,with a predicted molecular mass 48.79 kDa and an isoelectric point (pI) of 8.79.The LdSPR protein is a basic protein comprising 29 negatively charged residues (e.g.,Asp and Glu) and 37 positively charged residues (e.g.,Arg and Lys).LdSPR is an unstable protein with a fat solubility index of 95.79.The relative amount of valine (Val,41) in LdSPR was the highest,followed by leucine (Leu,38),which accounted for 9.8 and 9.0% of the total amino acids,respectively.LdSPR has a basic seven-transmembrane receptor structure of GPCR family members,and belongs to GPCR family A.Thirty-one SPR proteins from other insects were selected for multiple sequence alignment with LdSPR (Appendix B).The LdSPR amino acid sequence has homology ranging from 68 to 95%with the other SPRs.A phylogenetic tree was generated(Fig.1) using 32 insect SPR sequences,which indicated that the following five species were clustered into a group withL.dispar:Trichoplusiani,Helicoverpaarmigera,Spodoptera lituraandH.assulta.

Fig.1 Phylogenetic analysis of the sex peptide receptor proteins from 32 insects.

3.2.Expression of LdSPR during developmental stages

Analysis of the transcript levels ofLdSPRduring different developmental stages (egg,1st-6th instar larvae,pupae,and male and female adults) showed that its expression was the lowest in 5th instar larvae and the highest in 6th instar larvae (by 42.96-fold compared with that in 5th instar larvae).LdSPR expression was 5.41-fold higher in female adults than in male adults (Fig.2).

Fig.2 Developmental expression levels of the LdSPR gene in Lymantria dispar.The relative mRNA expression levels were normalized by the levels of EF1α,Actin,and TUB.1L-6L indicate 1st-,2nd-,3rd-,4th-,5th-and 6th-instar larvae,respectively.The lowest expression level of LdSPR was detected in the 5th instar of the larval stage,so the value of the relative expression level of the 5th instar larvae is 1.The data shown are the mean±SD (n=3).Different letters indicate significant differences among developmental stages according to ANOVA followed by Student-Newman-Keuls multiple comparisons test (P＜0.05).

3.3.LdSPR silencing is associated with L.dispar development and stress resistance

TheLdSPRexpression level in 3rd instar larvae is lower than in other larval stages.For the functional investigation ofLdSPR,RNAi was used to knockdownLdSPRexpression in 3rd instarL.disparlarvae.At 48,72,96,and 120 h after the injection of dsLdSPR,theLdSPRmRNA levels had decreased significantly compared with those in the dsGFP-injected larvae,thus confirming successful RNAi ofLdSPR.The decrease ofLdSPRgene expression began at 48 h,and the expression was relatively stable at 24 h.The downregulation ofLdSPRgene expression was most significant at 72 h,at which point its expression was 14.13% of that in the control treatment (Fig.3).The survival rate ofL.disparlarvae injected with dsLdSPRwas 87.5%,which was 10% lower than that of the control group (Fig.4;Appendix C).The cumulative growth rate of the dsGFP-injectedL.disparlarvae was higher compared with that in the dsLdSPR-injected larvae (Table 1).LdSPRgene silencing affected the nutrition utilization of the 4th instarL.disparlarvae.The RCR and AD ofLdSPR-silencedL.disparlarvae were higher compared with those of the control dsGFP-injected larvae (Table 2).By contrast,the RGR,ECD,and ECI values of theLdSPR-silencedL.disparlarva were lower than those of the control larvae.After silencing theLdSPRgene,the stress resistance of 3rd instar larvae was analyzed,including starvation stress,high temperature stress,and paraquat oxidative stress.Under high temperature stress,the survival rate of theLdSPRsilenced 3rd instar larvae was lower than that of the control dsGFP-injected larvae at each time point.The survival rate changed rapidly in the early stage,while the change in the later stage was slow and finally stabilized.At 156 h,the survival rate of the control group was 3.67-fold higher than that of theLdSPR-silenced group,which had a survival rate of only 7.5%.The average death time of theLdSPRsilenced group was 104.80 h,which was 80.62% of that of the control group.We speculated thatLdSPRgene silencing decreased the resistance of theL.disparlarvae to the high temperature stress (Fig.5-A).Under starvation stress,the survival rate of theLdSPR-silenced 3rd instar larvae was lower than that of the dsGFP-injected larvae at each time point,and the survival rate changed rapidly in the early stage.At 108 h,the dsGFPcontrol group showed a 1.35-fold higher survival rate than that of theLdSPR-silenced group.The average death day of theLdSPR-silenced group was 93.11% of that of the control group.Thus,we speculated that theLdSPRgene has a certain regulatory effect on the starvation resistance ofL.dispar(Fig.5-B).

Fig.3 LdSPR gene expression in the 3rd instar Lymantria dispar larvae with LdSPR gene silencing.The gene expression is shown as the mean±SD from three independent assays.＊ indicates a significant difference between the dsLdSPR and dsGFP treatments according to Student’s t-test (P＜0.05).

Fig.4 The survival rate of 3rd instar Lymantria dispar larvae with LdSPR gene silencing.The larvae treated with dsGFP were regarded as the control.The bar indicates SD (n=3).＊indicates a significant difference between the dsLdSPR and dsGFP treatments according to Student’s t-test (P＜0.05).

After treatment of the 3rd instarL.disparlarvae withparaquat for 48 h,the LC10,LC30,and LC50of theLdSPR-silenced group were lower than those of the control group,and the LC10of the control group was 2.14-fold higher than that of the treatment group.The LC50values of theLdSPR-silenced group and the control group at 48 h were 6.75 and 10.54 mg L-1,respectively.The results showed that the antioxidant activity of 3rd instar larvae afterLdSPRgene silencing was weakened (Table 3).

The 3rd instarL.disparlarvae were subjected to the oxidative stress of a sublethal paraquat treatment (48-h LC30=3.02 mg L-1).The survival rate of the dsGFP-injected group was significantly higher than that of theLdSPR-silenced group.Thus,we speculated thatLdSPRgene silencing reduced the oxidative stress resistance ofL.dispar(Fig.5-C).

Fig.5 Effects of high temperature,starvation,and paraquat oxidative stress (48-h LC30=3.02 mg L-1) on the survival rate of the 3rd instar Lymantria dispar larvae with LdSPR gene silencing.A,effects of high temperature on the survival rate of the 3rd instar L.dispar larvae with LdSPR gene silencing.B,effects of starvation on the survival rate of the 3rd instar L.dispar larvae with LdSPR gene silencing.C,effects of paraquat oxidative stress (48-h LC30=3.02 mg L-1) on the survival rate of the 3rd instar L.dispar larvae with LdSPR gene silencing.The larvae treated with dsGFP were regarded as the control.The bar indicates SD (n=3).＊ indicates a significant difference between the dsLdSPR and dsGFP treatments according to Student’s t-test (P＜0.05).

4.Discussion

Herein,we cloned the full-length cDNA of the Asian gypsy moth geneLdSPR,and performed functional and stress resistance analysis of LdSPR using RNAi technology.LdSPR is a GPCR containing a seventransmembrane structure,belonging to GPCR family A (Yapiciet al.2008).When a ligand activates a GPCR,extracellular signals are transformed into intracellular physiological responses (Lappano and Maggiolini 2011).InDrosophila,SP in the seminal fluid has a central function in various aspects of post-mating responses (Chenet al.1988;Liu and Kubli 2003).In addition toDrosophila,SPR has been reported in many insects,such asH.armigera,S.litura,H.assulta,andBombyx mori(Naokiet al.2010;Heet al.2015).Based on multiple sequence alignments,the phylogenetic tree showed that LdSPR could be clustered with the SPRs fromH.assulta,T.ni,S.litura,andH.armigera.SPR is highly expressed in the reproductive tract of femaleDrosophila.SP/SPR-mediated PMR might have evolved in theD.melanogasterspecies ancestor because it increased SPR expression in the female reproductive tract (Tsuda and Aigaki 2016).LdSPRexpression inL.disparvaried at different developmental stages,being higher in the egg stage than in most other developmental stages (except the 4th and 6th instar larvae,pupae,and female adults),in which theLdSPRexpression changed according to larval growth.LdSPRexpression in 6th instar larvae was the highest,and it was higher in female adults than in male adults.Based on this special expression pattern,we speculatedLdSPRmight be involved in the regulation ofL.disparage changes,pupation,and female oviposition.

In 3rd instarL.disparlarvae,RNAi silencing ofLdSPRwas achieved using microinjection.The results showed that the silencing effect was the highest at 72 h,with an 85.87% silencing efficiency.In this study,we studied the effects of silencing theLdSPRgene on the development,growth,and nutrient utilization ofL.dispar.The results showed thatLdSPRsilencing significantly affected the normal development and growth ofL.dispar,and the cumulative growth in fresh weight was lower at each time point compared with that in the control group.The RGR,ECI and ECD values ofL.disparlarvae afterLdSPRsilencing were lower than those of the control group,while the RCR and AD values of the treatment group were higher than those in the control group.We speculated that the reason for this phenomenon is that interference withLdSPRexpression inhibits metabolism,thus retardingL.disparlarval growth and development.LdSPR plays a certain regulatory role in the growth,development,and metabolism ofL.disparlarvae.In our previous study,a similar regulatory function of Methuselah-like (Mthl) was found forL.disparlarval growth and development after silencingMthl(Caoet al.2019).Compared with the larvae injected with dsGFP,the survival rate of theL.disparlarvae injected with dsLdSPRat each time point was lower under high temperature,starvation,and oxidative stress.RNAi silencedL.disparlarvae showed significantly increased sensitivity to high temperature,starvation,and oxidative stress.Thus,LdSPR might be involved in regulatingL.disparstress resistance.However,previous research has addressed the effects of SPR on reproduction and sleep(Kimet al.2010).The functions of LdSPR associated with reproduction and sleep will be investigated in a future study.

5.Conclusion

The SPR inL.disparwas first cloned and might participate in insect growth and development regulation,metabolism,life-span,and stress resistance.TheL.disparlarvae withLdSPRgene silencing showed increased sensitivity to high temperature,starvation,and oxidative stress.Our results further enrich our knowledge of the functions of insect GPCR SPRs,and provide a novel molecular target for developing environmentally friendly insecticides to control pests.

Acknowledgements

This work was supported by grants from the National Key R&D Program of China (2018YFC1200400),the National Natural Science Foundation of China (31570642),the Fundamental Research Funds for the Central Universities,China (2572019CG04) and the Heilongjiang Touyan Innovation Team Program,China (Tree Genetics and Breeding Innovation Team).

Declaration of competing interest

The authors declare that they have no conflict of interest.

Appendicesassociated with this paper are available on http://www.ChinaAgriSci.com/V2/En/appendix.htm