Knockdown of a nutrient amino acid transporter gene LdNAT1 reduces free neutral amino acid contents and impairs Leptinotarsa decemlineata pupation

A Leptinotarsa decemlineata SLC6 NAT gene (LdNAT1) was cloned. LdNAT1 was highly expressed in the larval alimentary canal especially midgut. LdNAT1 mRNA levels were high right after the molt and low just before the molt. JH and a JH analog pyriproxyfen activated LdNAT1 expression. RNAi of an allatostatin gene LdAS-C increased JH and upregulated LdNAT1 transcription. Conversely, silencing of a JH biosynthesis gene LdJHAMT decreased JH and reduced LdNAT1 expression. Moreover, 20E and an ecdysteroid agonist halofenozide repressed LdNAT1 expression, whereas a decrease in 20E by RNAi of an ecdysteroidogenesis gene LdSHD and disruption of 20E signaling by knockdown of LdE75 and LdFTZ-F1 activated LdNAT1 expression. Thus, LdNAT1 responded to both 20E and JH. Moreover, knockdown of LdNAT1 reduced the contents of cysteine, histidine, isoleucine, leucine, methionine, phenylalanine and serine in the larval bodies and increased the contents of these amino acids in the larval feces. Furthermore, RNAi of LdNAT1 inhibited insulin/target of rapamycin pathway, lowered 20E and JH titers, reduced 20E and JH signaling, retarded larval growth and impaired pupation. These data showed that LdNAT1 was involved in the absorption of several neutral amino acids critical for larval growth and metamorphosis.


Methods and Materials
Insects rearing and chemicals. L. decemlineata larvae and adults were routinely reared in an insectary according to a previously described method 28 , and were supplied with potato foliage at vegetative growth or young tuber stages in order to assure sufficient nutrition. At this feeding protocol, L. decemlineata larvae progressed through four distinct instars, with the approximate periods of the first-, second-, third-and fourth-instar stages being 2, 2, 2 and 4 days, respectively. Upon reaching full size, the fourth larval instars spent an additional 4-7 days as non-feeding prepupae. The prepupae then dropped to the soil and burrowed to a depth of 3-5 cm to pupate.
The representative NAT sequences were retrieved from NCBI, and were aligned with the predicted LdNAT1 using ClustalX (2.1) 32 . The neighbor-joining (NJ) tree was constructed using MEGA6 33 under the Poisson correction method. The reliability of NJ tree topology was evaluated by bootstrapping a sample of 1000 replicates.
Preparation of dsRNA. The same method as previously described 28 was used to express dsSHD, dsE75, dsFTZ-F1, dsJHAMT, dsAS-C, dsNAT1-1, dsNAT1-2 and dsegfp derived from a 141 bp fragment of LdSHD, a 361 bp common fragment in the three LdE75 isoforms, a 319 bp common fragment in both LdFTZ-F1-1 and LdFTZ-F1-2 cDNAs, a 261 bp fragment of LdJHAMT, a 206 bp fragment of LdAS-C, a 665 bp and a 357 bp fragments of LdNAT1, and a 414 bp fragment of enhanced green fluorescent protein gene. These dsRNAs were individually expressed with specific primers in Table S1, using Escherichia coli HT115 (DE3) competent cells lacking RNase III. Individual colonies were inoculated, and induced to express dsRNA by addition of 0.1 mM isopropyl β -D-1-thiogalactopyranoside. The expressed dsRNA was extracted and confirmed by electrophoresis on 1% agarose gel (data not shown), and quantified using a spectrophotometer (NanoDrop Technologies, Wilmington, DE). Bacteria cells were centrifuged at 5000 × g for 10 min, and resuspended in 0.05 M phosphate buffered saline (PBS, pH 7.4) at the ratio of 1:1. The bacterial suspensions (at dsRNA concentration of about 0.5 μ g/ml) were used for bioassay.
Bioassay. Our preliminary results revealed that feeding the second-instar larvae with 0.1 μ g/mL of 20E-, Hal-, JH-or Pyr-immersed foliage did not affect larvae growth, pupation and adult emergence. In this survey, nine independent bioassays were carried out as previously described 25,29 , using newly-ecdysed second-and third-instar larvae. The first bioassay was planned to test the effects of 20E and Hal on LdNAT1 expression. For 1 day, ten third-instar larvae were fed leaves which have been immersed in: (1) water (control), (2) 0.1 μ g/mL 20E, or (3) 0.1 μ g/mL Hal. Each treatment was replicated three times, and was collected to extract total RNA. The second to fourth bioassays were to knock down LdSHD, LdE75 and LdFTZ-F1 and each bioassay had three treatments: (1) PBS-, (2) dsegfp-, (3) dsSHD-, dsE75-or dsFTZ-F1-immerged leaves. Each treatment (ten larvae) was replicated six times and was fed for 3 days. Three replicates were collected to extract total RNA and the other three replicates were used to extract 20E. The fifth bioassay was intended to determine the effects of JH and Pyr on LdNAT1 expression by confining ten third-instar larvae for 1 day in petri dishes with: (1) water (control)-, (2) 0.1 μ g/mL JH-, or (3) 0.1 μ g/mL Pyr-immersed leaves. Three replicates in each treatment were collected to extract total RNA. The sixth and seventh bioassays were intended to silence LdAS-C and LdJHAMT and had three treatments: (1) PBS-, (2) dsegfp-, (3) dsAS-C-or dsJHAMT-immerged leaves. Six replicates in each treatment (ten larvae) were fed for 3 days to extract total RNA and JH respectively. The eighth and ninth bioassays were planned to knock down LdNAT1 by allowing ten second-and third-instar larvae to ingest: (1) PBS-, (2) dsegfp-, (3) dsNAT1-1-, (4) dsNAT1-2-dipped leaves. Each treatment was repeated 15 times. For extraction of total RNA, free amino acid, 20E and JH, twelve replicates were respectively collected after continuously fed for 3 days. The remaining 3 replicates were used to measure the larval weight, and to observe the larval developing period and the pupation using the methods as described previously 28 . The surviving larvae were individually weighed 3, 4, 5 and 6 days after treatment. Their development was observed at 4-hr intervals. Larval instars were identified by head capsule width, the appearance of exuviae, the black color of the pronotum, and the anterior beige and posterior black stripe visible on the pronotum of the 3 rd and 4 th instars respectively. Prepupae were distinctive from larvae by their disappeared black pigmentation, their relative inactivity and their curved body shape. The initiation of pupation was indicated by the soil-digging behavior. The pupation and the adult emergence were recorded during a 4-week trial period. For each bioassay, three biological replicates were carried out.

Real-time quantitative PCR (qPCR).
For tissue expression analysis, RNA samples were extracted from epidermis, foregut, midgut, hindgut, Malpighian tubules, fat body, hemocytes and brain-corpora cardiaca-corpora allata complex of the day 1 fourth-instar larvae. For temporal expression analysis, RNA templates were derived from the first, second, third larval instars at the interval of one day, and from fourth larval instars at the interval of eight hours. For analysis the effects of bioassay, total RNA was extracted from treated larvae. Each sample contained 5-30 individuals and repeated 3 times. The RNA was extracted using SV Total RNA Isolation System Kit (Promega). Purified RNA was subjected to DNase I to remove any residual DNA according to the manufacturer's instructions. Quantitative mRNA measurements were performed by qRT-PCR, using internal control genes (LdRP4, LdRP18, LdARF1 and LdARF4, the primers listed in Table S1) according to our published results 31 . An RT negative control (without reverse transcriptase) and a non-template negative control were included for each primer set to confirm the absence of genomic DNA and to check for primer-dimer or contamination in the reactions, respectively. Each sample was repeated three times. Data were analyzed by the 2 −ΔΔCT method, using the geometric mean of internal control genes for normalization. All methods and data were confirmed to follow the MIQE (Minimum Information for publication of Quantitative real time PCR Experiments) guidelines 34 . Free amino acid analysis. The larvae were collected from 3 replicates, whereas the feces were collected from all 15 replicates after continuously ingested dsNAT1-1 and dsNAT1-2 for 3 days. The foliage samples repeated three times. The collected samples were immediately ground in liquid nitrogen and stored at − 80 °C until assayed. Hydrolytic amino acids in the foliage were extracted by acid hydrolysis method. Free amino acids in the bodies and feces were extracted with 80% (v/v) ethanol at 25 °C for 10 min. Samples were centrifuged for 15 min at 10000 × g and 4 °C. Free amino acid contents in the bodies, feces and foliage were analyzed with a Beckman 6300 Amino Acid Analyzer (Beckman Instruments Inc., Fullerton, CA, USA). The amino acid contents were given as μ m per gram.
Quantitative determination of JH and 20E. Hemolymph was collected and JH was extracted following the methods described previously 28 . A liquid chromatography tandem mass spectrometry was used to quantify JH titers (ng per ml hemolymph) 35 .
20E was extracted according to an ultrasonic-assisted extraction method 29 , and its titer (ng per g body weight) was analyzed by a liquid chromatography tandem mass spectrometry-mass spectrometry system using a protocol the same as described 36 . Data analysis. The data were given as means ± SE, and were analyzed by analyses of variance (ANOVAs) followed by the Tukey-Kramer test, using SPSS for Windows (SPSS, Chicago, IL, USA). Pupation rate and amino acid composition were subjected to arc-sine transformation before ANOVAs.

Results
Identification of a putative nutrient amino acid transporter transcript. We sequenced a full-length transcript encoding a putative nutrient amino acid transporter in Leptinotarsa decemlineata, and provisionally designated LdNAT1. The correctness of the cDNA was confirmed by end-to-end amplification and sequencing. It was submitted to GenBank with the accession number of AHH29249.
LdNAT1 had a 1920 bp open reading frame encoding a 640-amino acid protein. The LdNAT1 protein was predicted to have 12 transmembrane domains and intracellular C and N termini using TMHMM 2.0 ( Figure S1A), which was in agreement with the general structure of transporters in the SLC6 family 11 . Based on the sequence alignment with selected characterized NATs and a crystallized bacterial NAT 13,37 , the possible substrate-binding moieties and first and second sodium-binding sites of LdNAT1 were also predicted ( Figure S1A).
The evolutionary relationship of NAT-like representatives derived from 6 insect species was evaluated ( Figure  S1B). As expected, LdNAT1 belongs to the Coleopteran clade. It was first grouped with that from Tribolium castaneum (XP_973741), with 100% of bootstrap value, and then the two and T. castaneum XP_008196787 joined together, supported by 97% of bootstrap value ( Figure S1B).
The expression of LdNAT1. The tissue expression patterns of LdNAT1 were tested by quantitative real-time PCR (qRT-PCR). LdNAT1 was detectable in the epidermis, foregut, midgut, hindgut, Malpighian tubules, fat body, hemocytes and brain-corpora cardiaca-corpora allata complex of the day 1 fourth-instar larvae. LdNAT1 mRNA was high in the midgut and moderate in the foregut and hindgut, whereas it was expressed at low levels in the Malpighian tubules, epidermis, fat body, hemocytes and brain-corpora cardiaca-corpora allata complex (Fig. 1A).
Scientific RepoRts | 5:18124 | DOI: 10.1038/srep18124 The temporal expression profiles of LdNAT1 were also determined in the larvae. LdNAT1 was expressed throughout all larval stages. Within the first, second and third larval instars, the expression levels of LdNAT1 were high right after the molt than just before the molt. In the fourth larval instar, a peak of LdNAT1 occurred 24 hours after ecdysis. Moreover, LdNAT1 showed two troughs 80 and 96 hours after ecdysis (Fig. 1B).
Juvenile hormone activates the expression of LdNAT1. The expression patterns showed that LdNAT1 mRNA levels were correlated with circulating JH. To determine whether JH induces LdNAT1 in vivo, LdNAT1 mRNA level in newly-ecdysed L. decemlineata third-instar larvae were tested after ingestion of water (control)-, JH-and Pyr-immersed foliage for 1 day. Compared with control specimens, LdNAT1 expression levels were significantly increased by 3.4 and 4.0 folds in the larvae that had ingested JH and Pyr ( Fig. 2A).
Moreover, LdNAT1 transcription was significantly enhanced by 3.6 fold in the LdAS-C RNAi hypomorphs, in which JH titer was significantly increased ( Fig. 2B-D).
Furthermore, JHAMT plays a major role in JH biosynthesis 27,28 . In this study, we knocked down LdJHAMT by RNAi (Fig. 2E) to lower JH titer (Fig. 2F), and found that LdNAT1 mRNA level was significantly diminished by 61.1% in the LdJHAMT RNAi larvae (Fig. 2G).

20-Hydroxyecdysone inhibits the expression of LdNAT1. The expression patterns suggest that 20E
pulse at the end of each instar inhibits LdNAT1. To verify the suggestion, LdNAT1 mRNA levels in L. decemlineata third-instar larvae were tested after ingestion of water (control)-, Hal-and 20E-immersed foliage for 1 day. Compared with control specimens, LdNAT1 expression levels were significantly decreased by 68.4% and 78.8% in the larvae that had ingested Hal and 20E (Fig. 3A).
Since RNAi of LdSHD reduced 20E titer in L. decemlineata 25,29 , LdNAT1 mRNA level was tested in the third-instar larvae in which LdSHD had been silenced by RNAi (Fig. 3B). As expected, 20E titer had been reduced (Fig. 3C), whereas LdNAT1 mRNA level was significantly increased by 6.1 fold in the LdSHD RNAi hypomorphs, compared with specimens that had ingested dsegfp (Fig. 3D).
Knockdown of LdNAT1 affects the absorption of several neutral amino acids. Hydrolytic amino acids in potato foliage, and free amino acids in bodies and feces of the day 1 third-and fourth-instar larvae were , hemocytes (HE) and brain-corpora cardiaca-corpora allata complex (BR) of the day 1 fourth-instar larvae. For temporal expression analysis, cDNA templates are derived from the first, second, third larval instars at the interval of one day, and from fourth larval instars at the interval of eight hours (D0/H0 indicated newly ecdysed larvae). For each sample, 3 independent pools of 5-30 individuals are measured in technical triplicate using qPCR. The bars represent 2 -ΔΔCt method (± SE) normalized to the geometrical mean of housekeeping gene expression. Different letters indicate significant difference at P value < 0.05.
analyzed (Tables S2-S4). The compositions of nineteen proteinogenic amino acids (tryptophan is undetectable) were listed in Table 1. The compositions of cysteine, histidine, lysine, methionine, serine, tyrosine and valine in the feces of the third-and fourth-instar larvae were significantly lower than those in the body and the potato foliage. Similarly, the compositions of glutamate, isoleucine, leucine and phenylalanine in the feces of the fourth-instar larvae were significantly lower than those in the body and the potato foliage (Table 1). These data suggest that a total of 11 amino acids (cysteine, histidine, lysine, methionine, serine, tyrosine, valine, glutamate, isoleucine, leucine, phenylalanine) may be actively absorbed by the larval gut.
The contents of the nineteen proteinogenic amino acids in the bodies and feces of the LdNAT1 RNAi larvae were measured (Table 2, Tables S2, S3). The contents of cysteine, histidine, isoleucine, leucine, methionine, phenylalanine and serine in the bodies of the LdNAT1 RNAi hypomorphs were significantly lower than those in the bodies of PBS-and dsegfp-fed larvae, whereas the contents of these amino acids in the feces of the LdNAT1 RNAi hypomorphs were significantly higher (Table 2, Tables S2, S3). These data indicate that knockdown of LdNAT1 reduces the absorption of the seven amino acids by the larval gut.
Knockdown of LdNAT1 retards larval growth and impairs pupation. When the larvae were weighed after 3, 4, 5 and 6 days of the initiation of the bioassays, the fresh weights were significantly reduced in the LdNAT1 RNAi larvae (Figs 4C and 5C). Exposure to dsNAT-immersed foliage significantly delayed larval developing stage. The mean periods of those on water-, dsegfp-, dsNAT1-1-, dsNAT1-2-dipped leaves were 14.0, 13.8, 16.0 and 16.3 days in the larvae that had been treated at the second instar stage, and were 12.0, 12.1, 15.0 and 15.0 days in the larvae having been treated at the third instar stage. In addition, the average periods of 2 nd , 3 rd and 4 th instars, and prepupae were further measured. Exposure to dsNAT1-1-, dsNAT1-2-dipped foliage significantly delayed the developing periods of the fourth-instar and prepupae (Table 3).
Moreover, the pupation rates were significantly decreased in the LdNAT1 RNAi larvae, compared with those in the PBS-and dsegfp-fed larvae (Figs 4D and 5D).
Knockdown of LdNAT1 disrupts insulin/TOR, JH and 20E signaling pathways. InR, 4EBP, FOXO and TOR are core components in insulin/TOR signaling pathway 38 . We sequenced cDNAs encoding InR, 4EBP, FOXO and TOR from L. decemlineata. After the initiation of the bioassay for 3 days, the expression levels of the four genes in the fourth-instar alive larvae were tested. LdInR, LdFOXO and Ld4EBP mRNA levels in dsNAT1-1-and     dsNAT1-2-fed larvae were significantly upregulated. In contrast, LdTOR mRNA level was significantly downregulated in the LdNAT1 RNAi larvae (Fig. 6A-D).

Discussion
In this survey, a putative LdNAT1 were cloned from L. decemlineata. LdNAT1 had a high amino acid similarity to homologs from other insect species. The phylogenetic result indicated that LdNAT1 was distantly related to other insect NAT1-like proteins. Moreover, LdNAT1 was predicted to have 12 transmembrane domains, which was in agreement with the general structure of transporters in the SLC6 family 11 . It appears that LdNAT1 may be among insect NAT-SLC6 members, and may play a principal role in active absorption and distribution of essential amino acids in L. decemlineata.
Tissue expression of LdNAT1. In insects, amino acids taken up from midgut need to traverse at least three membranes to reach their intracellular site of use: (1) uptake into an epithelial gut cell, (2) basolateral efflux from the gut cell into the hemolymph, and (3) uptake into somatic cell. In general, SLC6 transporters potentially mediate the uptake steps (1) and (3), whereas other SLC members are involved in the efflux step (2) 1, 39 .
In this survey, tissue expression profiles revealed that LdNAT1 mRNA levels were high or moderate in the larval gut, and lower in other surveyed tissues. Similarly, NAT genes in mammalians, other insect species and nematodes are highly expressed in the apical membranes of the alimentary canal, as well as in other organs with elevated requirements for essential amino acids 1,[11][12][13]37,39 . For example, MsKAAT1 in M. sexta 7 , AeAAT1 11 and AeNAT5 14 in A. aegypti, AgNAT6 37 and AgNAT8 12 in A. gambiae were highly expressed in the larval guts. Moreover, insect NATs showed unique expression patterns in neurons of the central ganglia and sensory system, which suggest their role as substrate providers for the synthesis of monoamine neurotransmitters 12,37 .
Therefore, the tissue expression patterns of LdNAT1 are compatible with the common idea that NATs function in: (1) active epithelial uptake of amino acids from the lumen of the gut to support systemic amino acid requirements; (2) active uptake of amino acids into specific cells to support their specialized metabolism or growth; and (3) control of the extracellular concentration of neurotransmitter amino acids in the context of synaptic transmission 39 . JH triggers whereas 20E represses the expression of LdNAT1. Essential amino acids are absorbed by midgut in insects [11][12][13] . In L. decemlineata midgut, several cysteine proteases such as intestain A through E have been reported to digest foliage protein 40,41 . Vacuolar H + -ATPases energize Na + and/or K + /H + antiport 42 to actively   Table 2. The contents of seven amino acids in the bodies and feces of the Leptinotarsa decemlineata larvae fed on PBS, dsegfp, dsNAT1-1 and dsNAT1-2-immersed foliage for 3 days. Free amino acid contents in the bodies and feces are analyzed with a Beckman 6300 Amino Acid Analyzer. Difference of amino acid content within each line is analyzed by ANOVA followed by the Tukey-Kramer test. Data that do not share the same letters are significantly different at P values of 0.05.

Larval instar 2 nd 3 rd 4 th Prepupae
Initiation of the bioassay at the early second instar stage  Table 3. The developing stage of L. decemlineata surviving larvae subjected to dietary dsRNA introduction.
The larval growth is checked at 4-hr intervals. See text for detail explanation for the identification of instars and prepupae. The data are given as means ± SE, and are subjected one-way ANOVA and followed by the Tukey-Kramer test. Means on the same column followed by the same letters are not significantly different at P < 0.05. transport Na + /K + from the hemolymph into the midgut lumen. Na + /K + ions are expected to subsequently trigger Na + /K + -dependent NATs to absorb amino acids in L. decemlineata, similar to the NAT-SLC6 members reported in other insect species 43 . In some insect species, genes encoding proteases, ATPase subunits and some NAT members are regulated by JH/20E. In Spodoptera litura larvae, for example, a putative serine protease gene Slctlp2 was induced by JH III but not 20E 44 . In Helicoverpa armigera, the expression of a trypsin-like serine protease gene HaTLP was upregulated by a JH analog methoprene and downregulated by 20E in vivo 45 . In A. aegypti, the transcription of a serine-type protease gene JHA15 is activated by JH in the newly emerged female adults 46 . Moreover, the expression of V-ATPase subunit genes were repressed by 20E in M. sexta 47 . Furthermore, two SLC7 genes, JhI-21 and minidiscs (mnd), were JH-inducible in D. melanogaster 48 . However, whether NAT-SLC6 genes in insects respond to JH/20E remains unproven. In several insect species, JH peaks and troughs are observed at each ecdysis. In a termite Cryptotermes secundus, for example, within each instar the JH titer rose shortly before or right after the molt, and then dropped sharply 49 . Similar phenomenon was observed in L. decemlineata 27 and M. sexta 47 . In this study, temporal expression patterns indicate that LdNAT1 transcript level appears to be positively correlated with the titer of JH. Thus, we determined whether the correlation had any biological significance. As expected, we discovered that either JH or Pyr induced the expression of LdNAT1 in an in vivo bioassay. Moreover, we found that knockdown of LdAS-C to increase JH titer 27 activated LdNAT1 expression. Conversely, silencing LdJHAMT to inhibit JH biosynthesis reduced LdNAT1 transcript in L. decemlineata final instar larvae.
Temporal expression patterns also imply that 20E inhibits LdNAT1 transcription. As expected, we found that LdNAT1 expression was dramatically decreased in L. decemlineata specimens having ingested 20E or an ecdysteroid agonist Hal. Conversely, a decrease in 20E in the LdSHD RNAi hypomorphs activated the expression of LdNAT1. Thus, 20E represses the transcription of LdNAT1 in L. decemlineata. In holometabolous insects such as D. melanogaster, 20E signal directly induces transcription of early 20E-response genes such as DmBR-C, DmE74A and DmE75A, and upregulates an early-late gene DmHR3 during larval-pupal transition 50 . DmHR3 then induces DmβFTZ-F1 expression in mid-prepupae 51 . In this survey, knockdown of either LdE75 or LdFTZ-F1 reduced 20E titer, as our previously reported results 29 , the expression levels of LdNAT1 in both LdE75 and LdFTZ-F1 RNAi larvae were significantly increased. Therefore, our results suggested that LdNAT1 repression required complete 20E signaling pathway.
In response to JH and 20E, LdNAT1 transcription is activated right after the molt, and is repressed just before the molt. Thus, its protein may function in absorption of amino acids at the mid instar stage when L. decemlineata larvae are actively feeding. Involvement of LdNAT1 in uptake of neutral amino acids. Xenopus oocyte-expressed insect transporters, such as M. sexta MsKAAT1 5 and MsCATCH1 10 , A. aegypti AaNAT1 11 and D. melanogaster DmNAT1 13 , shared relatively broad substrate spectra. MsKAAT1 absorbed phenylalanine, tryptophan, isoleucine, leucine, valine, methionine and alanine 5,7-9 , whereas MsCAATCH1 preferred threonine in the presence of K + but preferred proline in the presence of Na + 10 . AeAAT1 from A. aegypti had notable apparent affinities and transport velocities for phenylalanine, cysteine, histidine, alanine, serine and methionine 11 . DmNAT1 in D. melanogaster transported threonine, isoleucine, leucine, valine, histidine, phenylalanine, tyrosine, tryptophan, methionine, cysteine, alanine, proline, serine, asparagine and glycine, with virtually equal apparent affinities and transport velocities 13 . In contrast, several Dipteran NAT-SLC6 members showed narrow specialization for absorption of essential amino acids 12,14,37 . However, their orphan orthologs are absent in D. melanogaster or outside Diptera species 14 .
In this survey, by comparing the compositions of nineteen proteinogenic amino acids (tryptophan is undetectable) in potato foliage, and in bodies and feces of the day 1 third-and fourth-instar larvae, we found that a total of 11 amino acids (cysteine, histidine, lysine, methionine, serine, tyrosine, valine, glutamate, isoleucine, leucine, phenylalanine) may be actively absorbed by L. decemlineata larval gut. In the LdNAT1 RNAi larvae, the contents of cysteine, histidine, isoleucine, leucine, methionine, phenylalanine and serine in the bodies were significantly lower, whereas the contents of these amino acids in the feces were significantly higher, compared with those in the bodies and feces of PBS-and dsegfp-fed larvae. Therefore, LdNAT1 in L. decemlineata appears an SLC6-NAT transporter belonging to the B(0) system, like most of its homologs in Dipteran and Lepidopteran insect species 5,10,11,13 .
However, among 11 amino acids being suggested active absorption by the larval gut in L. decemlineata in this survey, lysine, tyrosine, valine and glutamate can not be absorbed by LdNAT1. This indicates that there are other functional NATs in L. decemlineata. Consistent with our results, a total of 9 SLC families are present in insects out of 10 SLC families participating in mammalian amino acid transport 1 . Work is in progress to identify these transporters. Knockdown of LdNAT1 results in the deficiency of seven neutral amino acids, i.e., cysteine, histidine, isoleucine, leucine, methionine, phenylalanine and serine. This delays the onset of critical weight and inhibits insulin/TOR signaling. The onset of critical weight and the inhibited insulin/TOR signaling may change the timing of the 20E pulse and JH level in the final instar larvae. Moreover, the reduced insulin/TOR signaling may also decrease 20E and JH signals. Thus, the growth is retarded and the pupation is impaired in the LdNAT1 RNAi hypomorphs.
Scientific RepoRts | 5:18124 | DOI: 10.1038/srep18124 Knockdown of LdNAT1 impaired larval development. Silencing LdNAT1 caused obvious negative effects: larval growth was retarded; development period was lengthened, and pupation was impaired. Similarly, RNAi of AeNAT5 increased larval mortality during ecdysis and dramatically suppressed adult emergence in A. aegypti 14 . Moreover, SLC6 was prominent in bacteria and archaea, and often served in environmental absorption of tryptophan, phenylalanine, tyrosine and methionine. Depriving the organism of such amino acids, or knocking out bacterial SLC6 transporters, limited the exponential growth of bacterial populations [52][53][54][55][56] . In Caenorhabditis elegans, snf-5 encodes a SLC6 family NAT. A loss-of-function mutation in snf-5 increased dauer formation and reduced dauer maintenance upon starvation 39 .
In insect, nutritional deprivation inhibited insulin/TOR pathway 38 to modulate insect growth and metamorphosis [22][23][24] . Thus, we measured the expression levels of LdInR, Ld4EBP, LdFOXO and LdTOR. As expected, the transcription of LdInR, Ld4EBP and LdFOXO was upregulated in L. decemlineata. In contrast, the mRNA level of LdTOR was reduced.
Our results further revealed that exposure to dsNAT1-1-, dsNAT1-2-dipped foliage significantly delayed the developing periods of the fourth-instar and prepupae, whenever we treated the larvae at the second-instar or at the third-instar stages. Thus, NAT is required for dietary intake of amino acids supporting growth and development to critical weight at the fourth (final) instar stage. Moreover, we further discovered that knockdown of LdNAT1 reduced the mRNA level of an ecdysteroidogenesis gene LdSHD, decreased 20E titer, and lowered the expression of two 20E-response gene LdE75 and LdFTZ-F1 29 . Furthermore, silencing LdNAT1 increased the mRNA level of LdAS-C, reduced the expression of a JH biosynthesis gene LdJHAMT, decreased JH titer, and diminished the transcription of a JH early-inducible gene LdKr-h1 in L. decemlineata larvae.
There are at least two alternative, but not mutually exclusive, explanations for the results. Firstly, delay of the onset of critical weight and reduced insulin/TOR signaling change the timing of 20E pulse and the level of circulating JH in the LdNAT1 RNAi hypomorphs. Alternatively, the reduced insulin/TOR signaling decreases circulating 20E and JH [22][23][24] . Disturbed 20E and JH signals subsequently delay larval development and impaired pupation. Accordingly, a model for the influence of knockdown of LdNAT1 on the larval-pupal metamorphosis is proposed in L. decemlineata (Fig. 7).