NlSPARC identification
Based on N. lugens transcriptome data constructed in our laboratory and genome data from National Center for Biotechnology Information (assembly ASM1435652v1), one specific PCR primer pairs for SPARC were designed and used to clone this gene from N.lugens. The cDNA of N.lugens SPARC (NlSPARC) consists of 1288 nucleotides, including the 5′ and 3′ untranslated regions, and an open reading frame of 891 nucleotides (from 138 bp to1028 bp). It ensodes a protein of 296 amino acids, with a theoretical molecular weight of 34.7 kDa and an isoelectric point of 5.3. The cDNA sequences of NlSPARC were deposited in GenBank under accession number MZ983402. Blast against NCBI showed that NlSPARC shared high identity with the SPARC predicted from Thrips palmi (up to 70.9% identity with an E value 4e-74 and 45% coverage). Blast analysis with N.lugens genome data showed that NlSPARC was located on chromosome 2 and was composed of at least 7 exons.
The sequence of the NlSPARC protein is similar with other insect SPARC orthologs, showing 87.5% and 70.6% identity to the SPARC from Laodelphax striatellus and Blattella germanica, respectively. When compared with SPARC from D. melanogaster, its identity decreases to 50.0%. As in other insect SPARC proteins, NlSPARC is structured in the three characteristic domains, Domain I (the acidic domain in the N-terminal), Domain II (the follistatin-like domain, the conserved putative N-glycosylation site) and Domain III (EF-hand calcium binding domain). NlSPARC sequences begin at the N-terminal methionine of the signal sequences (MERKAYLLFALLACFLLIDVTSS) (Fig. 1). Domain I (residues 24–76) have 12 acidic residues, the overall acidity is reduced by the presence of 9 basic residues (pI 6.3). NlSPARC contains 16 cysteine residues, 11 of which are conserved in domain II (residues 77–191). The C-terminal half in domain III is most highly conserved containing two EF-hand domains in all organisms. EF-hand II domain in NlSPARC is stabilized by a disulfide bridge between cysteine residues 258 and 274 and an additional cysteine residue (286) located near the C-terminus which is found only in invertebrate SPARC proteins (Fig. 1). The name of the species, the corresponding accession numbers for SPARC, length and the pIs of Domain I are listed in Table S1.
To clarify the evolutionary relationship of NlSPARC, we used a neighbor-joining tree construction program Mega 7 based on distances of 18 SPARC sequences from 18 arthropod including Coleoptera, Lepidoptera, Diptera, Hemiptera, Dictyoptera, Hymenoptera and Psocodea. The dendrogram obtained places the NlSPARC with Hemiptera as a distinct cluster (Figure S1).
Developmental and tissue-specific expression of NlSPARC in N. lugens
Egg and nymphal development normally lasts 6 ~ 7 days and 12 ~ 14 days, respectively. Nymphal development is comprised of five nymphal instars, from N1 to N5, spanning about 48 h for each. The RT-qPCR analysis revealed that NlSPARC was expressed at all developmental stages of N. lugens. Expression of NlSPARC was highest on average in 3- and 4-day-old eggs, then declined rapidly in 5- and 6-day-old eggs and reached to the lowest level in newly hatched nymphs (Fig. 2A). The results of the expression analysis of NlSPARC in various tissues showed that the highest expression was in fat body and hemolymph and there was no significant difference between females and males (Fig. 2B).
Knock-down of NlSPARC resulted in nymph mortality
Based on the sequence alignment of NlSPARC, the regions (95–891 bp) of NlSPARC with great divergence from other insects were selected for the synthesis of dsRNA. Preliminary study conducted for dose-determination showed NlSPARC expression was downregulated by more than 70% and the nymph mortality within 24 h was less than 20% after injection of dsNlSPARC in nymphs with a concentration of 700 ng/μL (i.e., 70 ng dsRNA in each insect), which concentration was used for further study. In order to elucidate the function of NlSPARC in nymphal development, 3rd instar and 4th instar nymphs were injected with dsNlSPARC, respectively. There was a significant increase in nymph mortality at day 8 after injection. The mortality was 36.0% and 46.4% at day 9 after injection in 3rd instar and 4th instar nymphs respectively, whereas it was 20.4% and 22.1% in control with dsGFP (P = 0.0104, P = 0.0004, respectively) (Fig. 3A).
NlSPARC mRNA levels were measured in 3rd instar nymphs after injection at day 4 and 9 and in tissues of adults developed from the injected 4th instar nymphs. Compared with dsGFP injection, in the 3rd instar nymphs, expression of NlSPARC significantly decreased by 18.4-fold at day 4 (P = 0.0003) and 4.4-fold at day 9 (P = 0.0035) after injection of dsNlSPARC (Fig. 3B). In the 4th instar nymphs, expression of NlSPARC significantly decreased by 24.2-, 20.6- and 7.3-fold in fat body, female and male internal reproductive organs respectively (P = 0.0034, P = 0.0027 and P = 0.0088) (Fig. 3B). After insect anatomy evaluation, we found that the fat body in adults injected with dsNlSPARC was smaller and less abundant than those from adults injected with dsGFP (Fig. 3C).
Effects of NlSPARC knock-down on reproduction and progeny
In order to elucidate the function of NlSPARC in N.lugens reproduction, the 5th instar nymph, newly emerged adults and gravid females were injected with dsNlSPARC, respectively. As no difference between injection of dsGFP and no injection control in 5th instar nymphs was observed, the no injection control was omitted in other stages. The injection of dsNlSPARC did not significantly affect adult longevity and ovariole maturation. The number of offspring after injection of dsNlSPARC in the 5th instar nymph, newly emerged adults and gravid females was significantly reduced by 51.3%, 59.1% and 52.5%, respectively (P = 0.0001, P = 0.0174 and P = 0.0300). The number of eggs after injection of dsNlSPARC in the 5th instar nymph, newly emerged adults and gravid females was significantly reduced by 49.3%, 59.2% and 46.0%, respectively (P = 0.0001, P = 0.0252 and P = 0.0467), compared to dsGFP injection (Fig. 4A). No difference in hatch rate between the treatment and control was found (data not shown).
There was no significant difference in 10-day survival rate of F1 progeny between injection of dsNlSPARC and dsGFP in 5th nymphs. The 10-day survival rate of F1 progeny of newly emerged adults injected with dsNlSPARC was 58.3%, and that of F1 progeny of gravid females injected with dsNlSPARC was 42.6%. Compared with dsGFP control, it was significantly reduced by 25.5% and 46.1%, respectively (P = 0.1956 and P = 0.0003) (Fig. 4B). Most nymphs die at the 1-2nd instar stage. No obvious morphological abnormalities were observed in injected individuals and their progeny.
NlSPARC mRNA levels were measured in F1 progeny with RNA extracted from 30 newly hatched nymphs and 20 mixed insects of 10 females and 10 males. Compared to dsGFP-injected control, the expression of NlSPARC decreased significantly in newly hatched nymphs by 2.4, 9.0 and 6.4 times after injection of the 5th instar nymphs, newly emerged adults and gravid females, respectively (P < 0.0001),whereas 10 days later no significant difference between treatment and control was found (Fig. 4C).
mRNA levels of NlSPARC, NlVg (Vitellogenin), NlVgR (Vitellogenin receptors) and NlFOXO were also measured in newly emerged females developed from the injected 5th instar nymphs. Compared to dsGFP-injected control, the expression of NlSPARC decreased by 7.0 times and the expression of NlVg, NlVgR and NlFoxO showed no significant difference (Figure S2).
As the number of offspring decreased after injection of dsNlSPARC, we then dissected the adults and found that the ovarian tubules were easily dispersed and the amount and volume of fat body reduced compared to the control (dsGFP) (Fig. 5). There was no obvious difference in development of internal reproductive organ, ovarian tubules and egg granules between dsNlSPARC and dsGFP treatment (Fig. 5).