Identification of Juvenile Hormone-Induced Posttranslational Modifications of Methoprene tolerant and Krüppel homolog 1 in the Yellow Fever Mosquito, Aedes aegypti

Abstract

Recent studies reported that JH regulated phosphorylation status of the JH-receptor complex contributes to its transcription activity in Aedes aegypti. However, phosphorylation sites of these proteins have not yet been identified. In this study, we found that the fusion of an EGFP tag to Ae. aegypti Kr-h1 (AaKr-h1) and Met (AaMet) improved their stability in mosquito Aag-2 cells, which allowed their purification. The liquid chromatography and tandem mass spectrometry analysis of the purified AaKr-h1 showed that the phosphoserine residue at position 694, located in the evolutionarily conserved SVIQ motif, is dephosphorylated when the cells are exposed to JH. The AaKr-h1 dephosphorylation mutant (S694V) showed significantly higher activity in inducing the luciferase gene regulated by JH response elements. The phosphorylation profile of Met also changed after exposing Aag-2 cells to JH III. The Ser-77 and Ser-710 residues of Met were phosphorylated after JH III treatment. In contrast, the two phosphoserine residues at positions 73 and 747 were dephosphorylated after JH III treatment. JH exposure also induced transient and reversible phosphorylation of Thr-664 and Ser-723 residues. Overall, these data show that JH induces changes in post-translational modifications of AaMet and AaKr-h1.

Graphical Abstract

Introduction

Ligand-activated transcription factors (TFs) regulate critical biological processes, including homeostasis, development, metabolism, and reproduction, by modulating gene expression via physical interaction with specific DNA sequences in the upstream regions of their target genes [1]. Their activation often requires heterodimerization with cofactors [2–4]. Many investigations have demonstrated that the activity of TFs is significantly influenced by various post-translational modifications (PTMs), including phosphorylation, acetylation, and ubiquitination [5–7]. However, identifying the PTMs of a TF has not been an easy process, mainly due to a lack of methods to prepare them in amounts and purities fully compatible with normally used protein PTM identification techniques such as Liquid Chromatography with tandem mass spectrometry (LC-MS/MS). Considering that sequence-specific DNA-binding factors, including TFs are estimated to represent only 0.01–0.001% of total cell protein in general [8], the isolation of a specific endogenous TF through conventional chromatographic methods has been very challenging. It often requires preparation of nuclear extract from an extremely high number of cultured cells (several thousand liters of conditioned media) to purify substantial quantities of the TF sufficient for analytical characterization or functional analysis unless an efficient method to promote the enrichment of the specific target protein (e.g., antibody-mediated immunoprecipitation) is used [8–10]. Also, the expression of TFs in heterologous expression systems is usually not easy because of the requirement of proper PTMs for their activity, nuclear translocation, or/and stability. However, the required PTM process could be absent or significantly different, depending on the host expression system used [11–18].

Methoprene-tolerant protein (Met) is regarded as a receptor for juvenile hormone (JH) [19]. Met binds to natural JHs with high affinity [20–22] and forms a heterodimer with a steroid receptor co-activator (SRC), which induces transcription of the genes that mediate anti-metamorphic action of JH [23, 24]. Krüppel-homolog 1 (Kr-h1), a transcription factor with a DNA-binding domain consisting of eight zinc fingers, is an early JH-response gene downstream of Met [25]. The JH-receptor complex (Met/SRC) activates Kr-h1 by directly binding to the JH response elements (JHREs) present in its promoter [22, 26–31]. The Kr-h1 acts as a vital transcription repressor of two key metamorphic genes, the pupal specifier Broad-Complex (BR-C) [32–37] and the adult specifier ecdysone-induced protein 93F (E93) [38–40].

A growing body of evidence suggests that the JH modulates the JH-receptor complex’s transcription activity by altering its PTM status. Liu et al. have reported that JH activation caused the phosphorylation of Met and SRC in Aedes aegypti via activation of the phospholipase C (PLC) pathway, which significantly improved the DNA binding affinity of the JH-receptor complex to JHRE and increased its transcriptional activity [41]. Ojani et al. further demonstrated that protein kinase C (PKC) is a key intermediate of the JHPLC pathway. This pathway is conserved in both Ae. aegypti and Aag-2 cells, which are derived from the embryos of Ae. aegypti [41, 42]. However, the JH-induced phosphorylation sites of Met remain unidentified. Moreover, it has been reported that the JH also induces phosphorylation of a BR-C Z7 isoform (BrZ7) in Helicoverpa armigera, which represses the 20E-dependent gene activation [43]. Overall, these findings suggest JH-induced changes in post-translational modifications of Met and other downstream TFs may play an important role in JH action. We optimized methods to purify the amounts of Ae. aegypti Kr-h1 (AaKr-h1) and Met (AaMet) proteins sufficient for LC-MS/MS analysis. LC-MS/MS analysis of purified AaKr-h1 and AaMet identified phosphorylation sites and revealed JH-induced changes in these two proteins’ phosphorylation levels.

Materials and Methods

Plasmid construction

The full-length coding sequence of AaKr-h1a (XM_021845557.1) was amplified by PCR and cloned into pIEx4 vector at Kpn I and Not I restriction sites. The full-length coding sequence of Met (XM_021840598.1), encoding amino acid residues 1–977, was cloned into pIEx4 vector at Nco I and Hind III restriction sites. For the generation of AaMet and AaKr-h1 fused with EGFP at its C-terminus, the EGFP sequence with no stop codon was cloned into pIEx4/Met at the Hind III site or pIEx4/AaKr-h1 at the Not I site. For preparing a construct with N-terminal EGFP fusion to Met, the Nco I site was used. The primers used in gene cloning and mutagenesis are shown in Supplementary Table 1.

Cell culture

Sf9 cells were maintained in Sf-900 III SFM medium (Thermo Fisher Scientific, Waltham, MA). Aag-2 cells were maintained at 28 °C in Schneider’s insect medium (MilliporeSigma, Burlington, MA) supplemented with 8% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA).

DNA transfection and affinity chromatography

The day before transfection, 1.8 × 10⁶ Aag-2 or Sf9 cells were seeded in each T-25 flask. After reaching 80–90% confluency, the cells were transfected with expression constructs using the TransIT-PRO Transfection Kit (Mirus Bio LLC, Madison, WI). The transfected cells were incubated with JH III at a final concentration of 10 μM for 1 or 24 h before harvesting at 2 days post-transfection. The His and S-tagged recombinant proteins were purified using subcellular fractionation followed by metal affinity chromatography under denaturing conditions [44, 45]. Briefly, the nuclear protein fraction was isolated from the harvested cells using the NE-PER kit (Thermo Fisher Scientific, Waltham, MA) following the manufacturer’s instructions. A Ni-charged MagBeads from GenScript (Piscataway, NJ) were then employed to isolate recombinant proteins. The nuclear extract was suspended in an extraction buffer (50 mM Tris, pH 7.4, 150 mM NaCl, and 6 M guanidine HCl) at 2 mg/ml concentration followed by incubation for 10 min with shaking at 4^oC. The suspension was then centrifuged at 100,000 × g for 30 min. The resultant supernatant was loaded onto Ni-charged MagBeads (GenScript), which have been pre-equilibrated with a washing buffer (50 mM Tris, pH 7.4, 150 mM NaCl, 6 M guanidine HCl, and 10 mM imidazole). After extensive washing with the washing buffer, bound His-tagged proteins were eluted by stepwise gradient elution with imidazole in the presence of 6 M guanidine HCl and 150 mM NaCl. For removing salt in the eluents, methanol/chloroform precipitation was carried out, and SDS-PAGE separated the purified proteins. After Coomassie Brilliant Blue staining, the protein bands were cut from the gels and subjected to LC-MS/MS analysis.

Phosphorylation Analysis by LC-MS/MS

The LC-MS/MS analyses were carried out at Taplin Mass Spectrometry Facility (Harvard Medical School). Briefly, the protein bands were retrieved and digested with trypsin as described previously [46]. The tryptic peptides were reconstituted in 5–10 μl of HPLC solvent A (2.5% acetonitrile, 0.1% formic acid) and loaded onto a nano-scale C18 reverse-phase HPLC capillary column, which is connected to a UHPLC [47]. A gradient was formed, and peptides were eluted with increasing concentrations of solvent B (97.5% acetonitrile, 0.1% formic acid). All the fractions collected were then subjected to electrospray ionization and analyzed by LTQ Orbitrap Velos Pro ion-trap mass spectrometer (Thermo Fisher Scientific, San Jose, CA). Eluting peptides were detected, isolated, and fragmented to produce a tandem mass spectrum of specific fragment ions for each peptide. Peptide sequences were determined by matching protein or translated nucleotide databases with the acquired fragmentation pattern by the software program, Sequest (ThermoFinnigan, San Jose, CA) [48]. The modification of 79.9663 mass units to Ser, Thr, and Tyr was included in the database searches to identify phosphopeptides. Phosphorylation assignments were determined by the Ascore algorithm [49]. All databases include a reversed version of all the sequences and the data was filtered to between a one and two percent peptide false discovery rate.

Reverse transcriptase quantitative PCR (RT-qPCR)

The total RNA from Ae. aegypti tissues or Aag-2 cells was extracted with TRIzol reagent (Life Technologies) following the manufacturer’s instructions. Two μg of total RNA was used to synthesize cDNAs using SuperScript II reverse transcriptase (Life Technologies). RT-qPCR reactions were run in quadruplicate on the iCycler iQ system using iQ SYBR Green Supermix (Bio-Rad). The relative mRNA levels were normalized to the expression of the reference gene, RPS2 and analyzed by the ‘Student’s t-test for significance. Primers used for RT-qPCR are shown in Supplementary Table 1.

Western blot assay

The purified proteins’ identities were confirmed by western blot analysis using rabbit anti-S-tag IgG obtained from Novus Biologicals (Centennial, CO) (1:2000 dilution as described in the manufacturer’s instructions). Rabbit anti-β-actin IgG (Cell Signaling Technology, Danvers, MA) was utilized (1:3000 dilution as described in the manufacturer’s instructions). HRP-conjugated goat anti-rabbit IgG (Cell Signaling Technology) was used at 1:2000 as a secondary antibody. The proteins were detected by chemiluminescence using the SuperSignal West Dura Extended Duration Substrate from Pierce Biotechnology (Waltham, MA).

Luciferase assay

To evaluate recombinant Met proteins’ transcriptional activity, we used pGL3/6XJHRE construct as described previously [31]. This pGL3 basic vector contains the luciferase gene under the control of a minimal promoter and six repeats of the juvenile hormone response elements (6XJHRE) identified in AaKr-h1 promoter. Briefly, Aag-2 cells were transfected with pGL3/6xJHRE vector along with AaMet expression construct. The luciferase assay was performed at 48 h post-transfection, as described previously [31].

To identify Ae. aegypti KBS sequences, the upstream regions of Ae. aegypti BR-C and E93 were scanned using the two Bombyx mori KBS sequences [35, 39] and their reverse complementary sequences using Pairwise Sequence Alignment tool at EBI (www.ebi.ac.uk/Tools/psa). The predicted putative KBS sequences are shown in Table 2. Briefly, the three promoter regions of BR-C (BP1, nucleotides −2660 to −2701; BP2, −3728 to −3769; BP3, −8249 to −8289) and the two promoter regions of E93 (EP1, nucleotides −2377 to −2418; EP2, −2692 to −2733; EP3, −3084 to −3125) were predicted as Ae. aegypti KBS candidates and are shown in Table 2. Each KBS site was cloned into the modified GL3 basic vector which has the luciferase gene under the control of a minimal promoter. Aag-2 cells were then co-transfected with reporter vector constructs and the gene expression construct encoding VP16 activation domain-fused to AaKr-h1 protein. The luciferase activity assays were carried out at 48 h post-transfection. To compare the transcription activity of the AaKr-h1 S694 dephosphorylation mutant with that of the wild-type, the gene expression constructs were co-transfected into Aag-2 cells along with either pGL3/BP1 or pGL3/EP2. The cells were then incubated at 28 °C for 48 h and subjected to the luciferase assays. All samples were prepared in triplicate.

Table 2.

Prediction of Ae. aegypti Kr-h1 interacting DNA motifs

BR-C promoter KBS1 (BP1) (−2660 to −2701)	BmBR-C_KBSa -TGACCTACGCTAACGCTAAATAGAGTTCCGA---------- BP1 CTTGTTCTACTAGTCTCAAACTAGAGAATCATCACTACTCCA  ** ***** *
BR-C promoter KBS2 (BP2) (−3728 to −3769)	BmBR-C_KBSa ----------TGACCTACGCTAACGCTAAATAGAGTTCCGA- BP2 CTGTAGCCTTGAGGTTACGCTTTCGCTTCATAAGCGGAAGGT  ****** **** *** *
BR-C promoter KBS3 (BP3) (−8249 to −8289)	BmE93_KBSb ----------------TGACCTTTCTCAAC----------- BP3 CCATCCATCGGGTTCATGTCCTTTCGCGCTTCTAGTGCAAA  ** ****** *
E93 promoter KBS1 (EP1) (−2377 to −2418)	BmBR-C_KBSa ----TGACCTACGCTAACGCTAAATAGAGTTCCGA------- EP1 TAAATGAATCAAAGTAACGCAAAGCAGAGGCAATTTTGAAGA  *** * ****** ** ****
E93 promoter KBS1 (EP2) (−2692 to −2733)	BmE93_KBSb -----------------TGACCTTTCTCAAC---------- EP2 GCCTGGGATTGAACCCATGACCTTCTGCATATGAAGCAGAAG  ******* **
E93 promoter KBS1 (EP3) (−3084 to −3125)	BmBR-C_KBSa TGACCTACGCTAACGCTAAATAGAGTTCCGA----------- EP3 GTGCCATTGTTGCCTAAAAATAGAAGATCACCAGCACTTATA  ** * * * ******* *

^a[35]

^b[39]

Statistical analysis

Statistical analyses were performed with the GraphPad Prism 5.01 software (San Diego, CA). Differences between groups were examined using Student’s t-test. p < 0.05 was considered statistically significant.

Data availability

The raw mass spectrometry proteomics data have been deposited at Zenodo. (https://zenodo.org/deposit?page=1&size=20) (Username: MK32145 & Password: WJ22C2A.bcytUwp).

Results

Expression and purification of recombinant AaKr-h1 protein

The full-length AaKr-h1α fused to both S and 8xHis tags (denoted as AaKr-h1) was transiently expressed in Spodoptera frugiperda Sf9 cells. The cells were then incubated with JH III for 0, 1 or 24 h before harvesting at 48 h post-transfection. Immobilized metal affinity chromatography (IMAC) was carried out for protein purification under denaturing conditions as described in the Materials and Methods section. The SDS-PAGE analysis showed that most of the recombinant AaKr-h1 protein running at ~110 kDa regardless of JH treatment was eluted by 200 mM imidazole (Supplementary Fig. 1) as detected by western blot using S-tag antibodies (Fig. 1A & B). To produce AaKr-h1 protein in Ae aegypti cells, the recombinant AaKr-h1 was also produced using the Aag-2 cells. Aag-2-expressed AaKr-h1 appeared to be very similar in size to the one produced in Sf9 cells (Fig. 1C). Considering that the molecular weight of the recombinant AaKr-h1 before any modifications is expected to be ~85 kDa, AaKr-h1 seems to be highly modified when expressed in either of the two cell lines. However, despite its decent expression levels in Aag-2 cells, it was difficult to purify the recombinant AaKr-h1 from the cell lysate. The total amount or concentration of the AaKr-h1 protein in the nuclear extract of Aag-2 cells is probably not sufficient for purification using IMAC due to a relatively lower transfection efficiency and smaller nucleus size of Aag-2 cells compared to Sf9 cells.

Fig. 1. SDS-PAGE and Western blot analysis of recombinant AaKr-h1 expression.

AaKr-h1–8xHis-S tag was transiently expressed in Aag-2 and Sf9 cells. (A) SDS-PAGE and (B) Western blot analysis of the recombinant Kr-h1 protein extracted from Sf9 cells. (C) Western blot analysis of the recombinant AaKr-h1 expressed in Aag-2 and Sf9 cells. (D) Western blot analysis of AaKr-h1 and AaKr-h1-EGFP expressed in Aag-2 cells. (E) SDS-PAGE and Western blot analysis of purified AaKr-h1-EGFP protein isolated from Aag-2 cells cultured in T-25 flasks. 1.8 × 10⁶ Aag-2 cells were seeded into each T-25 flask on the day before transfection. 48 h after transient transfection, a nuclear extract containing AaKr-h1-EGFP was fractionated from the cells, followed by protein purification by IMAC under a guanidine denaturing condition. 50 ng of purified Kr-h1 or 20 μg of total cell lysate was utilized for Western blot analysis against S-tag. β-actin was used in Western blotting as a loading control.

To purify the amount of Aag-2-expressed AaKr-h1 protein sufficient for LC-MS/MS analysis, the fusion protein consisting of AaKr-h1 and EGFP attached the C-terminus of AaKr-h1 (denoted as AaKr-h1-EGFP) was produced. The results of western blot analysis using protein isolated from cells at 48 h post-transfection showed that the C-terminal EGFP fusion considerably improved the stability of AaKr-h1 in Aag-2 cells. (Fig. 1D). Importantly, this simple protein engineering enabled us to produce AaKr-h1 protein in Aag-2 cells grown in one T-25 flask and purify the amount sufficient for LC-MS/MS analysis. This recombinant protein was successfully purified from the nuclear extract of Aag-2 cells by a single-step denaturing IMAC (Fig. 1E).

Furthermore, we determined if the C-terminal EGFP fusion inhibits the nuclear translocation of AaKr-h1. Confocal microscopic observations revealed punctate nuclear localization of AaKr-h1-EGFP in both Aag-2 and Sf9 cells (Fig. 2A), which agrees with the findings of Mao et al. [50, 51] showing the very similar nuclear expression of endogenous Kr-h1 in Brown Planthopper. JH treatment also did not significantly affect the subcellular distribution profile of the recombinant AaKr-h1 (Supplementary Fig. 2). These results suggest that the fusion of EGFP to the C-terminus of AaKr-h1 does not significantly inhibit its nuclear transport.

Fig. 2. Alignment of the Kr-h1 protein from Ae. aegypti and its orthologs from other insect species and its subcellular localization.

(A) Multiple alignments of the amino acid sequence C-terminus of Kr-h1 from various insect species showing a conserved SVIQ motif (694–697). The red-colored S indicates the phosphoserine of Ae. aegypti Kr-h1 at position 694, which undergoes dephosphorylation in Aag-2 cells treated with JH. (B) Fluorescence micrographs of Kr-h1-EGFP expressing Aag-2 and Sf9 cells showing the punctate nuclear localization of Kr-h1-EGFP. Cells were stained with RNA-sensitive fluorescent dye to mark the cytosol and counterstained with DAPI to identify the nuclei (Scale bar = 5 μm). The images are representative of multiple fields and experiments.

LC-MS/MS identification of AaKr-h1 Ser-694 phosphorylation site

SDS-PAGE was used to separate the Aag-2-expressed AaKr-h1-EGFP protein isolated by IMAC. The specific protein bands were then cut from the gel and eluted for LC-MS/MS analysis to identify JH-induced post-translational modifications of AaKr-h1. The LC-MS/MS data are shown in Supplementary Data files 1 & 2, and the identified phosphorylation and acetylation sites are shown in Table 1. LC-MS/MS analysis identified a phosphorylated peptide, MSS#VIQYAK. The Ser-694 residue, which is located outside of the multiple zinc finger containing DNA binding domain, was phosphorylated in Aag-2 cells. Interestingly, this specific phospho-serine residue was not detected in the Aag-2 cells exposed to JH III for 1 or 24 h while the other two phosphorylation sites (S561/T613) remained unaffected by JH treatment up to 24 h. These data suggest that activation of the JH signaling pathway is accompanied by removal of the phosphate group from S694 of AaKr-h1. Interestingly, the results of LC-MS/MS of Sf9-expressed AaKr-h1-EGFP revealed extensive modifications. Eight phosphorylation sites (S148/T163/S561/T574/S586/S595/T613/S694) and an acetylated lysine residue at position 571 were detected in the AaKr-h1 protein when it was isolated from the Sf9 cells exposed to JH III for 24 h (Table 1 & Supplementary Fig. 3). These results imply that AaKr-h1 protein may undergo extensive post-translational modifications when expressed heterologously in insect cells. Multiple alignments of the amino acid sequences of AaKr-h1 with its orthologous proteins from other insect species revealed widespread conservation of the RMSSVIQYA motif, where the Ser-694 is located (Fig. 2B). Significantly, the four amino acids SVIQ were perfectly conserved in all the six proteins compared suggesting that the SVIQ motif is crucial for AaKr-h1 function.

Table 1.

Post-translational modifications of Ae. aegypti Kr-h1 identified by tandem mass spectrometry analysis.

ScanF	Host cell	PTM sites		Peptide_A	ModScore for Position_A1	Peptide_B	ModScore for Position_B1
		Phospho-	Ace-
16588	Aag-2	S561		R.FKREES#PVNADPFLYKPLTIM*K.H	110.8	R.FKREES#PVNADPFLYKPLTIM*K.H	118.2
16830	Aag-2	T613		R.QVEAAIAGTENLLT#PPR.S	81.7	R.QVEAAIAGTENLLT#PPR.S	105.5
11649	Aag-2	S694		R.MSS#VIQYAK.A	24.4	R.MSS#VIQYAK.A	30.5
14981	Aag-2 + JH (1 h)	S561		R.FKREES#PVNADPFLYKPLTIM*K.H	16.6	R.FKREES#PVNADPFLYKPLTIM*K.H	23.2
15163	Aag-2 + JH (1 h)	T613		R.QVEAAIAGTENLLT#PPR.S	38.2	R.QVEAAIAGTENLLT#PPR.S	52.7
15120	Aag-2 + JH (24 h)	S561		R.FKREES#PVNADPFLYKPLTIM*K.H	28.7	R.FKREES#PVNADPFLYKPLTIM*K.H	49.4
15260	Aag-2 + JH (24 h)	T613		R.QVEAAIAGTENLLT#PPR.S	51.0	R.QVEAAIAGTENLLT#PPR.S	68.6
9409	Sf9 + JH (24 h)	S148		K.QQQM*VAES#K.G	1000.0	K.QQQM*VAES#K.G	1000.0
16031	Sf9 + JH (24 h)	T163		K.GLLNGPAT#PADPYQCDVCK.K	38.2	K.GLLNGPAT#PADPYQCDVCK.K	52.7
18795	Sf9 + JH (24 h)	S561		R.FKREES#PVNADPFLYKPLTIM*K.H	75.7	R.FKREES#PVNADPFLYKPLTIM*K.H	59.4
22497	Sf9 + JH (24 h)	S561	K571	K.REES#PVNADPFLYK@PLTIM*K.H	23.0	R.FKREES#PVNADPFLYKPLTIM*K.H	31.8
27180	Sf9 + JH (24 h)	T574		R.FKREESPVNADPFLYKPLT#IM*K.H	30.3	R.FKREES#PVNADPFLYKPLTIM*K.H	39.9
16216	Sf9 + JH (24 h)	S586		K.HHGYFAPAS#QVPEFR.S	60.4	K.HHGYFAPAS#QVPEFR.S	72.2
7244	Sf9 + JH (24 h)	S595		R.SSS#DLVR.Q	24.4	R.SSS#DLVR.Q	30.5
18818	Sf9 + JH (24 h)	T613		R.QVEAAIAGTENLLT#PPR.S	51.0	R.QVEAAIAGTENLLT#PPR.S	68.6
11897	Sf9 + JH (24 h)	S694		R.MSS#VIQYAK.A	24.4	R.MSS#VIQYAK.A	30.5

Note:

¹–If the ModScore values for Position_A and Position_B are both above 19, the location is considered confidently assigned. If the value is listed as 1000, there are no other possible sites to assign the phosphate and the assignment is unequivocal.

^#–Phosphorylation

^@–Acetylation

^*–Oxidation of methionine

Effect of AaKr-h1 dephosphorylation at Ser-694 on its transcriptional activity

Changes in the post-translational modification status of a transcription factor often significantly influence its transactivation activity and nuclear transport [52–55]. For instance, 20E, another key insect hormone, induces the phosphorylation of Broad-Complex (BR-C) in the silkworm moth Bombyx mori, which substantially reduces the affinity of BR-C to its DNA targets and its transcription activity [56]. These observations led us to determine if the JH-dependent dephosphorylation at the Ser-694 of AaKr-h1 significantly alters its transcription activity. Kr-h1 is identified as a transcription factor responsible for repressing the 20E-induced expression of BR-C and E93 in multiple insect species [35, 39, 57]. 5′ regulatory region of BR-C and E93 gene in Bombyx mori was reported to contain Kr-h1 binding sites (KBS) 5′-GACCTACGCTAACGCTAAATAGAGTTCCGA-3′ [35] and 5’-TGACCTTTCTCAAC-3’ [39], respectively. However, the KBS in Ae. aegypti genes have not yet been identified despite previous attempts to identify them by sequence search [58]. JH III or methoprene, a JH analog, treatment repressed 20E-induced activation of aegypti BR-C and E93 genes in Aag-2 cells (Supplementary Fig. 4), suggesting that Kr-h1 also acts as a strong repressor of BR-C and E93 genes in Ae. aegypti. These findings encouraged us to identify Ae. aegypti KBS in the upstream regions of the mosquito BR-C and E93. The identified KBS sequences can be used in the experiments to evaluate the effects of the JH-induced dephosphorylation of Kr-h1 on its transactivation activity.

The 5-kb 5’upstream regions of Ae. aegypti BR-C and E93 genes were searched with the two Bombyx mori KBSs and their reverse complementary sequences. The predicted putative Ae. aegypti KBS sequences are shown in Table 2. The putative KBS sequences were cloned into the modified pGL3 basic vector, which contains the luciferase gene under a minimal promoter’s control. The constructs were co-transfected into Aag-2 cells with gene expression constructs encoding the activation domain of VP16 (denoted as VP16) or the VP16 fused with Kr-h1 (denoted as Kr-h1-VP16). The cells were then subjected to the luciferase assay at 48 h post-transfection. The results revealed that the upstream regions of Ae. aegypti BR-C and E93 genes contain the cis-regulatory elements to which AaKr-h1-VP16 could bind and transactivate the reporter gene (Fig. 3A & B).

Fig. 3. Impacts of S694C mutation on the activity of AaKr-h1.

Identification of the cis-regulatory elements used by AaKr-h1 in the upstream region of Ae. aegypti BR-C (A) and E93 (B). 5-kb 5” upstream regions of Ae. aegypti BR-C and E93 were first scanned for the presence of two Bombyx mori KBS elements and their reverse complementary sequences using the Pairwise Sequence Alignment tool at EBI. Each predicted KBS site was cloned into the modified pGL3 basic vector and then co-transfected into Aag-2 cells along with AaKr-h1-VP16 expression construct. (C) The luciferase reporter assays measured activity of the Kr-h1 wild-type and S694V dephosphorylation mutant with the reporter driven by the Kr-h1-response elements of BR-C and E93 (pGL3/BP2 and pGL3/EP2, respectively). The luciferase activity was normalized to total protein concentration. Each lane refers to an average of triplicates or quadruplicates, and its values were expressed as the mean ± standard deviation. Asterisks, significant differences in the mean values between two groups. p values were determined with Student’s test: *, p < 0.05. (D) 20 μg of total cell lysate was used for western blot analysis using S-tag antibodies.

We further examined the effect of AaKr-h1 dephosphorylation on its transactivation ability. Constructs containing AaKr-h1 and AaKr-h1 mutants (with no VP16 fusion) along with KBS reporter constructs were transfected into Aag-2. The reporter assay results showed that AaKr-h1 wild-type or mutant overexpression significantly increased the luciferase expression regulated by BP2, EP1, EP2, and EP3 KBS elements without the activation domain VP16 (Supplementary Fig. 5), suggesting that all the Kr-h1 wild-type and mutants tested have a transcriptional activator activity. We also observed that AaKr-h1 dephosphorylation mutants (S694V and S694C) showed significantly lower expression levels than the wild-type and phosphoserine mimic mutant (S694E) when the same amount of gene expression plasmids were transfected (Supplementary Fig. 5). However, despite their substantially lower expression levels, the overexpression of AaKr-h1 S694V and S694C induced comparable or somewhat higher luciferase activity for these reporter plasmids when compared to the wild-type and phosphoserine mimic mutant (Supplementary Fig. 5). When the measured luciferase activities were normalized to Kr-h1 expression, both AaKr-h1 dephosphorylation mutants (S694V and S694C) clearly revealed their more vigorous transcriptional activity than the wild-type and S694E mutant (Supplementary Fig. 5). Consistent with these findings, S694C dephosphorylation mutant induced more substantial luciferase activity than the wild type when produced in a similar quantity (Fig. 3C & D). In contrast, no significant difference was observed between the wild-type and phosphoserine mimic mutant after normalization (Supplementary Fig. 5).

The results of RT-qPCR analysis for BR-C mRNA levels in Aag-2 cells also showed that the S694V dephosphorylation mutant repressed the 20E-dependent activation of BR-C isoforms Z1 and Z4 as efficiently as the wild-type, although its protein levels measured were lower than the wild-type (Fig. 4). These results suggest that Kr-h1 has a transcriptional activator activity and may require another partner like Hairy to act as a transcriptional repressor [58]. Overall, these findings support the hypothesis that the JH-induced dephosphorylation of AaKr-h1 enhances its transcription factor activity.

Fig. 4. Effects of overexpression of AaKr-h1 S694V dephosphorylation mutant on 20E-dependent activation of the Ae. aegypti Broad-Complex gene.

qRT-PCR analysis of BR-C isoform mRNA (Z1, Z2, Z3, and Z4) levels in Aag-2 cells was performed as described in Materials and Methods. Aag-2 cells were transfected with pIEx4, pIEx4/AaKr-h1-EGFP, or pIEx4/AaKr-h1-EGFP S694V. The cells were then incubated with 20E at a final concentration of 1 μg/ml for 6 h. Then the cells were harvested at 48 h post-transfection for qRT-PCR analysis. Error bars indicate the SD of triplicate samples (*, P < 0.05 [Student t test]).

Expression and purification of recombinant AaMet

The expression of full-length AaMet fused with 8xHis and S tags (denoted as AaMet) was analyzed by immunoblotting using S tag antibodies. However, the recombinant AaMet was expressed below the assay’s detection limit in both Aag-2 and Sf9 cells, probably because this protein is unstable. To improve the stability of AaMet, we generated the fusion protein consisting of AaMet and EGPF fused at N- or C-terminus of AaMet (denoted as EGFP-AaMet and AaMet-EGFP, respectively). We found that the fusion of EGFP at the N-terminus of AaMet considerably enhanced the stability of AaMet, whereas the C-terminal EGFP fusion showed little improvement (Fig. 5A). The AaMet purified from proteins isolated from Aag-2 cells transfected with AaMet-EGFP construct and exposed to JH III for 1 or 24 h showed expected size bands in SDS-PAGE gels stained with Coomassie blue (Fig. 5B). Then, we determined if EGFP fusion affected the activity of AaMet. The AaMet-EGFP constructs were transfected into Aag-2 cells along with the pGL3 reporter vector containing the luciferase gene under the control of a minimal promoter and six repeats of the juvenile hormone response elements (6XJHRE), identified in the AaKr-h1 promoter. The transfected cells were incubated for 48 h and the luciferase reporter assays were conducted. Overexpression of AaMet or EGFP-AaMet significantly increased the luciferase activity (Fig. 5C), suggesting that AaMet retained significant transcription activity after EGFP fusion. Confocal microscopic observations also revealed that the majority of EGFP-AaMet protein was preferentially distributed to the nuclei of Aag-2 and Sf9 cells, even in the absence of JH (Fig. 5D), which is consistent with the findings of Greb-Markiewicz et al.[59] that showed the JH-independent nuclear accumulation of Drosophila melanogaster Met in immortalized cell lines. However, JH stimulation did not make any conspicuous changes to the recombinant AaMet in its subcellular distribution in both cell lines (Supplementary Fig. 6).

Fig. 5. Effects of the EGFP fusion to Ae. aegypti Met and SRC on protein stability and nuclear localization.

(A) Western blot analysis of AaMet-8xHis-S, EGFP-AaMet-8xHis-S, and AaMet-EGFP-8xHis-S expressed in Aag-2 cells. 50 and 5 μg of total cell lysate were utilized for Western blot analysis against S-tag and β-actin, respectively. (B) SDS-PAGE analysis of the purified recombinant EGFP-Met extracted from Aag-2 cells. (C) The luciferase reporter assays measured activities of the recombinant Met proteins with the reporter driven by the JH response element (JHRE) of Ae. aegypti Kr-h1 (AKA, pGL3/6xJHRE). The luciferase activity was normalized to protein concentration. Each lane refers to the average of triplicates or quadruplicates, and its values were expressed as the mean ± standard deviation. Asterisks, significant differences of the mean values between two groups. p values were determined with Student’s test: *, p < 0.05. (D) Fluorescence images of EGFP-Met expressing Aag-2 and Sf9 cells an evenly distributed nuclear localization of EGFP-Met. 12 μg of desired gene expression plasmid was transfected when cells reached 80–90% confluence in a T-25 flask. The fluorescent signal was observed under a microscope at 48 h post-transfection. Each cell population was co-stained with RNA-sensitive fluorescent dye to mainly mark the cytosol and counterstained with DAPI to identify the nucleus (Scale bar = 5 μm). All images were representative of multiple fields and experiments.

JH-induced post-translational modifications of AaMet

Recombinant AaMet was purified from the Aag-2 cells harvested from a T-25 flask using denaturing IMAC protocol (Fig. 5B). The LC-MS/MS data obtained from this protein are shown in Supplementary Data 1 & 2, and the identified phosphorylation sites are shown in Table 3. Results obtained in the LC-MS/MS analysis revealed not only the specific residues of AaMet whose phosphorylation status is influenced by the JH signaling pathway but also how JH alters the phosphorylation profiles of these residues overtime. Phosphorylation of Ser-77 and Ser-710 in AaMet was detected after the cells were exposed to JH III while the other two phosphoserine residues at positions 73 and 747 underwent dephosphorylation in response to JH. Interestingly, these modifications were not reconstituted within 24 h after continuous exposure to JH III. Whereas the JH-induced phosphorylation changes at the Thr-664 and Ser-723 residues of AaMet were restored to the original state. In contrast, no significant change in phosphorylation status of Ser-651 was observed at 1 h or 24 h after exposure to JH III, suggesting that a phosphoserine residue at position 651 probably remained unaffected by up to 24 h of JH treatment. Together, these data clearly show that activation of JH signaling not only induces phosphorylation at the specific sites in AaMet but is accompanied by dephosphorylation of other phosphoserine residues. Considering that most JH receptor complexes’ activity is reduced after continuous exposure to JH for 10 h or longer (Supplementary Fig. 7), it seems that some JH-induced post-translational modifications of AaMet are not quickly restored to the original state even after JH signaling is terminated. A multiple protein sequence alignment was carried out using the online ClustalW program to identify the location of phosphorylation residues. The results revealed that AaMet amino acids whose phosphorylation profiles were influenced by JH were neither positioned in the conserved domains nor present in its orthologous proteins (Fig. 6). These results suggest that the identified PTMs of AaMet are species-specific.

Table 3.

Post-translational modifications of Ae. aegypti Met identified by tandem mass spectrometry analysis.

ScanF	Host cell	Phosphorylation site	Peptide_A	ModScore for Position_A1	Peptide_B	ModScore for Position_B1
7767	Aag-2	S73	R.S#GPSSVATTTATEPGEK.L	19.2	R.S#GPSSVATTTATEPGEK.L	22.9
5712	Aag-2	S651	K.AHS#PALESR.V	65.4	K.AHS#PALESR.V	86.2
12157	Aag-2	S747	R.THSNS#CLDSSDGSSIGHTADFVVQK.R	19.8	R.THSNS#CLDSSDGSSIGHTADFVVQK.R	19.4
5566	Aag-2 + JH (1 h)	S651	K.AHS#PALESR.V	65.4	K.AHS#PALESR.V	86.2
16363	Aag-2 + JH (1 h)	T664	R.VPPLHGT#DHDFIQPFSPAGSTSSSSSSSGSVFSPASHQR.N	23.4	R.VPPLHGT#DHDFIQPFSPAGSTSSSSSSSGSVFSPASHQR.N	39.0
20259	Aag-2 + JH (1 h)	S723	R.SPFGGSIPPTSPPAIAATPSTSVS#R.I	20.1	R.SPFGGSIPPTSPPAIAATPSTSVS#R.I	23.7
7436	Aag-2 + JH (24 h)	S77	R.SGPSS#VATTTATEPGEK.L	21.2	R.SGPSS#VATTTATEPGEK.L	19.0
5246	Aag-2 + JH (24 h)	S651	K.AHS#PALESR.V	65.4	K.AHS#PALESR.V	86.2
15866	Aag-2 + JH (24 h)	S710	R.SPFGGSIPPTS#PPAIAATPSTSVSR.I	18.1	R.SPFGGSIPPTS#PPAIAATPSTSVSR.I	22.9

Note:

¹–If the ModScore values for Position A and Position B are both above 19, the location is considered confidently assigned. If the value is listed as 1000, there are no other possible sites to assign the phosphate and the assignment is unequivocal.

^#–Phosphorylation

^*–Oxidation of methionine

Fig. 6. Alignment of the Met protein from Ae. aegypti and its orthologues from other insect species.

The specific amino acid residues of Ae. aegypti Met whose phosphorylation profiles are influenced by JH treatment in Aag-2 cells (S73/S77/T664/S710/S723/S747) highlighted in yellow color. However, the phosphoserine residue at position 651 seems to remain unaffected up to 24 h exposure to JH III.

Discussion

This study reports a fast and convenient protocol to prepare transcription factors in the amount and quality sufficient for analyzing their post-translational modifications using tandem mass spectrometry. This protocol involves protein engineering for the high-yield and IMAC enrichment process. These methods were used to express, purify and analyze two essential transcription factors in JH action. JH is known for its anti-metamorphic action to maintain the juvenile status of holometabolous insects [60]. The JH signaling pathway is also involved in previtellogenic development, oogenesis, and vitellogenesis in many insect species, which are essential for female insect reproduction. The molecular action of JH for its physiological functions is largely dependent on the JH-receptor complex (Met/SRC) [61, 62] and Kr-h1 [63]. Recent studies have shown that JH stimulation not only promotes the formation of the JH-receptor complex but also directly improves its transcription activity. In the present study, we monitored how JH changed the phosphorylation profiles of AaMet and AaKr-h1 in mosquito cells. Interestingly, our three-time point analysis of AaMet (0, 1, and 24 h after JH treatment at a final concentration of 10 μM) revealed that JH induced alterations in phosphorylation status at specific amino acid residues of AaMet, rather than affecting the total phosphorylation levels (Table 3). Within 1 h after treatment, JH promoted both phosphorylation at The-664 and Ser-723 and dephosphorylation at Ser-73 and Ser-747, which implies that changing phosphorylation status of specific amino acids is probably important in modulating the activity of AaMet during the early stage of the activation by JH. Intriguingly, the two dephosphorylated serine residues were not restored to the original state up to 24 h after JH treatment, while the JH-dependent phosphorylation at The-664 and Ser-723 were reconstituted after 24 h exposure to JH. Also, although Ser-77 and Ser-710 were not significantly phosphorylated within 1 h by the JH treatment, they were phosphorylated at 24 h time point. In other words, the Ser-77 and Ser-710 of AaMet probably undergo slower phosphorylation by JH activation than The-664 and Ser-723, implying the sequential phosphorylation at the different sites of AaMet by the JH signaling pathway. Together, these data provide the first evidence showing that the phosphorylation profiles of the multiple residues of Met are dynamically altered over time by the JH signaling pathway. Simultaneously, present results also implicate the possibility that the total number of phosphorylation sites on AaMet can be measured differently by immunoblotting, depending on the time of exposure and/or JH dose used. Moreover, considering the observed dynamics of the JH-induced PTMs of AaMet, we cannot completely rule out the possibility that other transient modifications occurring between 1 h and 24 h post-treatment with JH are missed in this study. Therefore, it is necessary to check JH-dependent modifications of AaMet at additional time points in future studies.

Studies on post-translational modification analysis of AaKr-h1 revealed that JH activation resulted in the quick removal of a phosphate group at the Ser-694. This dephosphorylation was not reconstituted to the original state up to 24 h JH exposure (Table 1). This may suggest either the activity of the enzyme(s) responsible for the dephosphorylation of AaKr-h1 probably was not restored to the normal state in the presence of JH or JH activation left a long-lasting molecular mark on AaKr-h1 protein. AaKr-h1 transcription activity assays also showed that the AaKr-h1 S694V dephosphorylation has higher transcription activity than the wild-type protein (Figs. 3 & 4). Moreover, the phosphoserine motif (SVIQ) containing Ser-694 is conserved in the orthologous proteins of AaKr-h1 from multiple holometabolous insects (Fig. 2A). Intriguingly, another type of phosphoserine motif (SVIH) was also observed in the Kr-h1 homologs of many hemimetabolous insects (data not shown), which may suggest that JH-depending dephosphorylation in these phosphoserine motifs of Kr-h1 is evolutionally important for Kr-h1 activation. This is a fascinating question to be addressed in future research. Overall, our results show the first evidence that JH activation promotes AaKr-h1 expression in Aag-2 cells and induces the dephosphorylation of AaKr-h1 protein that modulates its activity.

The identified PTMs of AaMet and AaKr-h1 in the present study are not located in their conserved protein domains, such as the basic HLH, PAS-A, and PAS-B of Met or the multiple zinc fingers of AaKr-h1 (Figs. 2 & 6). Currently, we do not know how these modifications can possibly affect their activity. Still, it is not difficult to find examples that show a post-translational modification occurring outside of the conserved domains of a transcription factor regulates its transcription activity [43, 64–67]. These previous reports led us to speculate that the identified modifications on Met and Kr-h1 probably are involved in conformational changes affecting the binding with their partner proteins or target response elements. The concomitant treatment of Aag-2 cells with 20E and JH III significantly repressed the 20E-induced expression of BR-C and E93 (Supplementary Fig. 4). The observed gene repression is probably due to the increased expression of Kr-h1 (Supplementary Fig. 4). In line with these findings, the overexpression of AaKr-h1 or its dephosphorylation mutant (S694V) also repressed BR-C isoform expression (Fig. 4). However, the Kr-h1 wild-type or mutant expression rather increased the luciferase expression regulated by the short KBS elements found in the upstream regions of the mosquito BR-C and E93 (Supplementary Figs. 3&5). These results imply that Kr-h1 has a transcriptional activator activity, but it turns into a transcriptional repressor via interaction with its binding partner on a promoter, such as Hairy. Indeed, both Kr-h1 and Hairy are involved in the repression of many JH responsive genes [58, 63, 68], including adult specifier E93 [69]. EGFP fusion significantly improved the stability of AaMet and AaKr-h1 without conspicuous changes in their nuclear translocation (Fig. 1A, 2B & 5A, D) and mRNA expression levels (Supplementary Fig. 8). The resultant fusion proteins also retained substantial transcription activity (Fig. 3C–D & 5C). These observations might be useful examples showing that simple EGFP fusion allows a high-yield production of an insect transcription factor by enhancing its stability in insect cells.

Conclusions

We studied how the JH alters the post-translational modification status of two important transcription factors in JH action, Met and Kr-h1. For target protein enrichment, we have developed an efficient method to prepare Ae. aegypti Met and Kr-h1 using a mosquito cell expression system. JH-induced changes in their phosphorylation and acetylation sites were mapped and monitored through a liquid chromatography–tandem mass spectrometry (LC–MS/MS)-based approach. The results showed that the JH signaling pathway provoked dynamic alterations in the phosphorylation profiles of the multiple residues of AaMet. Not only phosphorylation and dephosphorylation at the specific residues of AaMet were involved in the alterations, but also their modification status was further changed over time after JH activation. Meanwhile, JH also induced post-translational modification to AaKr-h1, a transcription factor working downstream of Met/SRC in JH signaling pathway. JH induced dephospholrylation of S94 residue which is located in the RMSSVIQYA motif highly conserved in its orthologous proteins from other insect species. In contrast, the JH-induced post-translational modification of AaMet detected are specific to Aedes aegypti. Moreover, no significant levels of acetylation were detected in these transcription factors. Together, we provide an efficient approach to monitor the post-translational modification profiles of insect transcription factors, which are typically difficult to produce and purify in large quantities. These results can help us further understand how the JH signaling pathway modulates its crucial transcription factors and provide a basis for developing a novel insect pest control strategy.

Significance.

Female Aedes aegypti mosquitoes are known to vector many disease agents, including Zika virus, dengue virus chikungunya virus, and Mayaro and yellow fever virus. In the present study, we developed an efficient method to prepare Ae. aegypti Met and Kr-h1, which are typically difficult to produce and purify, using a mosquito cell line expression system. A liquid chromatography–tandem mass spectrometry (LC–MS/MS)-based approaches were utilized to map the phosphorylation profiles of the isolated proteins. We then monitored the changes induced by JH activation in the phosphorylation profiles to check if the JH modulates post-translation modification of its key transcription factors. We found that the JH induced alterations in the phosphorylation profiles of the multiple residues of AaMet. In contrast, activation of the JH signaling pathway was accompanied by dephosphorylation of AaKr-h1 at phosphoserine-694, increasing its transcriptional activity. In addition, S694 of AaKr-h1 was located in the RMSSVIQYA motif highly conserved in orthologous proteins from other insect species. These results can help us further understand how JH modulates its key transcription factors and provide a basis for the development of novel insect control strategies.

Highlights.

• EGFP-fusion highly improves the stability of AaMet and AaKr-h1 proteins in mosquito cells.

• The phosphorylation profiles of the multiple residues of AaMet are dynamically altered over time by the JH signaling pathway.

• Meanwhile, the phosphoserine residue at position 694 of AaKr-h1 is dephosphorylated by the JH signaling pathway, which increases its transcriptional activity.

• The Ser-694 of AaKr-h1 is located in the evolutionarily conserved SVIQ motif.

Abbreviations used:

JH

Juvenile hormone

Kr-h1

Krüppel homolog 1

Met

Methoprene tolerant

SRC

Steroid receptor co-activator

LC-MS/MS

Liquid Chromatography with tandem mass spectrometry

EGFP

enhanced green fluorescent protein

UHPLC

Ultra High-Performance Liquid Chromatography

PAS

Per-Arnt-Sim