Krüppel-homolog 1 exerts anti-metamorphic and vitellogenic functions in insects via phosphorylation-mediated recruitment of specific cofactors

The zinc-finger transcription factor Krüppel-homolog 1 (Kr-h1) exerts a dual regulatory role during insect development by preventing precocious larval/nymphal metamorphosis and in stimulating aspects of adult reproduction such as vitellogenesis. However, how Kr-h1 functions both as a transcriptional repressor in juvenile metamorphosis and an activator in adult reproduction remains elusive. Here, we use the insect Locusta migratoria to dissect the molecular mechanism by which Kr-h1 functions as activator and repressor at these distinct developmental stages. We report that the kinase PKCα triggers Kr-h1 phosphorylation at the amino acid residue Ser154, a step essential for its dual functions. During juvenile stage, phosphorylated Kr-h1 recruits a corepressor, C-terminal binding protein (CtBP). The complex of phosphorylated Kr-h1 and CtBP represses the transcription of Ecdysone induced protein 93F (E93) and consequently prevents the juvenile-to-adult transition. In adult insects, phosphorylated Kr-h1 recruits a coactivator, CREB-binding protein (CBP), and promotes vitellogenesis by inducing the expression of Ribosomal protein L36. Furthermore, Kr-h1 phosphorylation with the concomitant inhibition of E93 transcription is evolutionarily conserved across insect orders. Our results suggest that Kr-h1 phosphorylation is indispensable for the recruitment of transcriptional cofactors, and for its anti-metamorphic and vitellogenic actions in insects. Our data shed new light on the understanding of Kr-h1 regulation and function in JH-regulated insect metamorphosis and reproduction.

In an effort to elucidate how Kr-h1 functions in repressing precocious nymph metamorphosis and stimulating adult reproduction in L. migratoria, we investigated Kr-h1 phosphorylation and its involvement in transcriptional repression and activation. The migratory locust L. migratoria is a destructive insect pest worldwide as well as a representative of evolutionarily basal insects with hemimetabolous development and JHdependent vitellogenesis. We found that PKCα triggers Kr-h1 phosphorylation. Phosphorylated Kr-h1 recruited C-terminal binding protein (CtBP), consequently inhibiting E93 expression and nymphal-adult metamorphosis. Phosphorylated Kr-h1 interacted with CREBbinding protein (CBP), which stimulated the transcription of Ribosomal protein L36 (RL36) and reproduction. We also provide evidence that the essential role of phosphorylated Kr-h1 in recruiting CtBP and repressing E93 expression is evolutionarily conserved in other representative insects including the silkworm Bombyx mori, the beetle T. castaneum and the fruit fly D. melanogaster.

Kr-h1 expression and phosphorylation are in response to JH
To explore the dynamics of p-Kr-h1 before the onset of locust metamorphosis, we conducted western blot using proteins extracted from the penultimate 4th and final 5th instar nymphs. As shown in Fig. 2A, p-Kr-h1 levels were high in mid and late 4th instar nymphs but markedly declined in 5th instar nymphs. The decreased levels of p-Kr-h1 at final nymphal instar appeared to correlate with the decline of JH titer in this phase [50], suggesting a possible effect of JH on Kr-h1 phosphorylation. It . m/z indicates the mass to charge ratio. E Upper panel: Pro-Q Diamond Phosphoprotein Gel Stain of purified bacterially-expressed GST-tagged peptides of Kr-h1(aa1-290), Kr-h1 S154A (aa1-290), Kr-h1(aa89-312), Kr-h1 S154A (aa89-312), and Kr-h1(aa291-591) preincubated with or without PKCα. Lower panel: Coomassie brilliant blue staining was used as the loading controls should be noted that Kr-h1 is expressed in response to JH [21,38]. The abundance of total Kr-h1 also decreased in 5th instar nymphs ( Fig. 2A). To evaluate the responsiveness of Kr-h1 phosphorylation to JH in juvenile stage, western blot was performed using protein extracts from mid-5th instar nymphs as well as those further treated with methoprene for 5-60 min. Application of methoprene caused increase of both Kr-h1 and p-Kr-h1 levels, and longer exposure to methoprene tended to have a relatively more pronounced effect on Kr-h1 expression and phosphorylation (Fig. 2B). Notably, p-Kr-h1 levels increased more rapidly than total Kr-h1 after 15-min exposure to methoprene ( Fig. 2B and Additional file 1: Fig. S3), implying a role of JH in stimulating Kr-h1 phosphorylation. Dose-response experiments demonstrated that higher doses of methoprene induced higher levels of Kr-h1 and p-Kr-h1 (Fig. 2C). The data suggest that JH promotes Kr-h1 expression and phosphorylation in nymphs, and the high levels of Kr-h1 phosphorylation are generally observed with more abundant Kr-h1 proteins.
We next studied the temporal abundance of p-Kr-h1 after adult ecdysis using protein extracts from the fat body of adult females at 0-6 days post adult emergence (PAE). Compared to that on the day of adult emergence, p-Kr-h1 levels increased at 1-4 days PAE and remained high on days 5-6, resembling that of total Kr-h1 (Fig.   2D). As JH is undetectable in the hemolymph at adult emergence but sharply increases thereafter [51], the enhanced levels of Kr-h1 and p-Kr-h1 appeared to positively correlate with elevated hemolymph JH titer. To elucidate the responsiveness of Kr-h1 phosphorylation to JH in adult locusts, western blot analysis was carried out using protein extracts isolated from the fat body of newly emerged adult females and those further treated with methoprene. As observed in nymphs, methopreneinduced Kr-h1 expression and phosphorylation were also seen in adults (Fig. 2E, F). Likewise, p-Kr-h1 abundance increased more rapidly than total Kr-h1 in the fat body of adult females treated with methoprene for 15 min ( Fig. 2E and Additional file 1: S3). Taken together, our data suggest that JH-induced Kr-h1 expression is accompanied by increased levels of Kr-h1 phosphorylation in both nymphal and adult locusts.
Kr-h1 phosphorylation is required for its role in stimulating reproduction Kr-h1 has a dual role in preventing precocious nymphal/ larval metamorphosis and in promoting adult reproduction. In L. migratoria, Ribosomal protein L36 (RL36) (GenBank: MT081313) was previously found to express in response to the JH-Met-Kr-h1 pathway [52]. RL36 is a component of the 60S subunit of ribosomes involved in ribosome biogenesis and protein translation as well as extra-ribosomal functions in various cellular processes [53]. Knocking down RL36 resulted in blocked ovarian growth and arrested oocyte maturation (Additional file 1: Fig. S6). As shown in Fig. 4A, Kr-h1 knockdown caused 54% reduction of RL36 mRNA levels. Similarly, NPC15437 treatment and PKCα knockdown resulted in 41% and 58% decrease of RL36 transcripts, respectively (Fig. 4A), suggesting a possible role of p-Kr-h1 in RL36 expression. For luciferase reporter assay, RL36 promoter region (nt -1647 to -1632) comprising a KBS motif (Additional file 1: Fig. S5A) was cloned into pGL4.10 vector. Co-transfection of pAc5.1/Flag-Kr-h1 and pGL4.10-4×RL36 -1647 to -1632 in S2 cells treated with methoprene brought about 2-fold induction of RL36 reporter activity compared to the empty vector control (Fig. 4B). When pAc5.1/Flag-Kr-h1 S154A was cotransfected with pGL4.10-4×RL36 -1647 to -1632 , no significant induction of RL36 reporter activity was observed (Fig. 4B). However, the induction of RL36 reporter activity was restored by overexpression of Flag-Kr-h1 S154D (Fig. 4B). The data indicate an essential role of Kr-h1 phosphorylation in RL36 transcription. We next performed ChIP assays to quantify in vivo binding of p-Kr-h1 to KBS-containing promoter region of RL36 in the fat body of adult females. Compared to the day of adult emergence, p-Kr-h1 was more enriched with the KBScontaining promoter sequence of RL36 on day 3, and even more on day 6 (Fig. 4C). However, NPC15437 treatment and PKCα knockdown in 6-day-old adult females resulted in significant reduction of p-Kr-h1 enrichment with RL36 promoter (Fig. 4D). Furthermore, application of methoprene to newly emerged adult females led to significantly enhanced precipitation of p-Kr-h1 in RL36 promoter region (Fig. 4E). These results together indicate a pivotal role of Kr-h1 phosphorylation in induction of RL36 transcription during female reproduction.
The Kr-h1 sequence contains eight C 2 H 2 zinc-finger domains. In addition to potentially recognizing a variety of DNA sequences, the zinc-fingers act as a hub for protein-protein interaction [61,62]. The Ser 154 residue is localized at the 3rd zinc-finger domain of Kr-h1. Phosphorylation modification is likely to induce a conformational change that is optimal for Kr-h1 to recruit cofactors. In the present study, CtBP and CBP were found to bind with phosphorylated Kr-h1 in repressing E93 transcription and activating RL36 transcription, respectively. Nevertheless, phosphorylated Kr-h1 could also interact with other cofactors in transcriptional activation or repression of target genes. In Ae. aegypti, Kr-h1 acts synergistically with Hairy, thereby mediating the action of Met in gene repression during previtellogenic development of adult females [27,63]. A study in N. lugens has demonstrated that Hairy directly interacts with the N-terminus zinc-finger domains of Kr-h1 in modulating gene transcription [64]. Dual functions of transcriptional activation and repression are widely observed with transcription factors [65][66][67][68][69][70]. In mammals, Krüppel-like factor 4 promotes the transcription of cyclin B1 via interacting with CBP, but downregulates cyclin B1 transcription by recruiting HDAC3 [66].
We have additionally shown Kr-h1 phosphorylation in other insects belonging to divergent orders, including the lepidopteran B. mori, the coleopteran T. castaneum, and the dipteran D. melanogaster. The requirement of phosphorylation for Kr-h1 action on suppressing E93 transcription was found to be also conserved. The findings provide a clear indication that Kr-h1 phosphorylation and its indispensable role in regulating target gene expression are evolutionarily conserved across distant insect orders. These observations further highlight the significance of Kr-h1 phosphorylation in eliciting transcriptional activity. Previously, JH-dependent Ae. aegypti Kr-h1 dephosphorylation at Ser 694 has been demonstrated to enhance the transcriptional activity [60]. The phosphoserine residue Ser 694 is conserved in some holometabolous insects but not in L. migratoria. The Ser 154 of L. migratoria Kr-h1 is homologous to Ser 206 of Ae. aegypti Kr-h1. Thus, Kr-h1 orthologues likely bear multiple phosphorylation sites with differential responses to JH. While evolutionarily conserved Kr-h1 phosphorylation sites occur in divergent insect species, the lineage-and species-specific Kr-h1 phosphorylation residues may exist in some insects. It is of interest to address these questions in future research.

Conclusions
Kr-h1 functions both as a transcriptional repressor in preventing precocious larval/nymphal metamorphosis and a transcriptional activator in stimulating adult reproduction in insects. PKCα phosphorylated Kr-h1 at a serine residue localized in the 3 rd zinc-finger domain. While Kr-h1 phosphorylation levels increased along with JH-induced total Kr-h1 expression, more rapid increase of Kr-h1 phosphorylation than total Kr-h1 was observed in locusts treated with methoprene. JH-induced Kr-h1 phosphorylation was also seen in methoprene-exposed S2 cells. Phosphorylated Kr-h1 recruited CtBP in nymphs, which inhibited E93 expression and metamorphosis. Phosphorylated Kr-h1 recruited CBP in adults, consequently stimulating RL36 transcription and vitellogenesis. Kr-h1 phosphorylation and its essential role in recruiting CtBP and repressing E93 expression are evolutionarily conserved in L. migratoria, B. mori, T. castaneum, and D. melanogaster. Thus, our present study fills a knowledge gap of phosphorylation modification of Kr-h1, an intermediate regulator in the JH/Met-response gene expression hierarchy.

RNA isolation and qRT-PCR
Total RNAs were extracted from insects and S2 cells using TRIzol reagent (Invitrogen), and first-strand cDNAs were reverse transcribed using FastQuant RT kit with gDNase (Tiangen). qRT-PCR was performed using a RealMasterMix SYBR Green kit (Tiangen) with a LightCycler 96 system (Roche), initiated at 95°C for 15 min, and followed by 40 cycles of 95°C for 10 s, 58°C for 20 s, and 72°C for 30 s. Relative expression levels were calculated using 2 −ΔΔCt method, normalized by ribosomal protein 49 (Rp49). Primers for qRT-PCR are listed in Table S3 (Additional file 1).
RNA interference and tissue imaging cDNA templates were amplified by PCR, cloned into pGEM-T vector (Tiangen), and confirmed by sequencing. dsRNAs were synthesized by in vitro transcription with T7 RiboMAX Express RNAi System (Promega). Locusts were intra-abdominally injected with 15 μg dsRNA, and boosted once on day 5. Phenotypes were photographed by Canon EOS550D camera and Leica M205C stereomicroscope. Primers used for RNAi are given in Table S3 (Additional file 1).

Chromatin immunoprecipitation
ChIP assays were performed using an EZ-Magna ChIP A/G Kit (Millipore). Briefly, fat bodies collected from nymph and adult females were fixed with 1% formaldehyde to crosslink chromatin for 10 min at 37°C. After addition of 125 mM glycine, chromatin was sonicated to shear into 200-1000 bp DNA fragments. The complexes were then immunoprecipitated with antibody against Kr-h1, phospho-Kr-h1 (Ser 154 ) or IgG, followed by qPCR. Primers used for ChIP are listed in Table S3 (Additional file 1).

Statistical analysis
Statistical analyses were performed by Student's t-test or one-way analysis of variance (ANOVA) with Tukey's post hoc test using the SPSS22.0 software. Significant difference was considered at P < 0.05. Values were reported as mean ± S.E.
Additional file 1: Figure S1. Identification of Kr-h1 phosphorylation site. Figure S2. Identification of kinase triggering Kr-h1 phosphorylation. Figure S3. Effect of 15-min exposure of methoprene on Kr-h1 phosphorylation. Figure S4. Effect of E93 knockdown on locust metamorphosis. Figure S5. Responsiveness of Kr-h1 phosphorylation to JH. Figure S6. Effect of RL36 knockdown on locust reproduction. Figure S7. Effect of CtBP or CBP knockdown on E93 or RL36 expression. Figure S8. Alignment of the 3 rd zinc-finger domain of Kr-h1 and the partial promoter sequences of E93 with KBS motifs. Table S1. Primers used for cloning and gene expression. Table S2. Primers used for site-directed mutagenesis. Table S3. Primers used for qRT-PCR, RNAi and ChIP.