Apis mellifera ligusticaapisomics.com/files/201712/f9a06c4f-da34-438e-88e9... · the honeybee life cycle. However, the molecular mechanism of this life transition is still unknown.

J O U R N A L O F P R O T E O M I C S 7 8 ( 2 0 1 3 ) 4 2 8 – 4 4 6

Ava i l ab l e on l i ne a t www.sc i enced i r ec t . com

www.e l sev i e r . com/ loca te / j p ro t

Changes of proteome and phosphoproteome triggerembryo–larva transition of honeybee worker(Apis mellifera ligustica)

Alemayehu Gala, Yu Fang, Dereje Woltedji, Lan Zhang, Bin Han, Mao Feng, Jianke Li⁎

Institute of Apicultural Research/Key Laboratory of Pollinating Insect Biology, Ministry of Agriculture, Chinese Academy of Agricultural Science,Beijing, China

A R T I C L E I N F O

⁎ Corresponding author. Tel./fax: +86 10 6259E-mail address: [email protected] (J. Li).

1874-3919/$ – see front matter © 2012 Elseviehttp://dx.doi.org/10.1016/j.jprot.2012.10.012

A B S T R A C T

Article history:Received 1 August 2012Accepted 12 October 2012Available online 23 October 2012

The development of the last day embryo to the first instar larva is an essential process inthe honeybee life cycle. However, the molecular mechanism of this life transition is stillunknown. The proteome and phosphoproteome of last day embryos (72 h) and first instarlarvae (24 h, post hatching) were analyzed using 2-DE, multiplex fluorescent staining, massspectrometry, bioinformatics, and qRT-PCR. Sixty-five proteins and 34 phosphoproteinschanged their expression across the shift of embryos to larvae. The embryo strongerexpression of proteins related to energy metabolism, development and amino acidmetabolism suggests its high metabolic energy demand during active embryogenesis.While, the newly hatched larvae escalated the expression of proteins related tocytoskeleton, biosynthesis, protein folding, fatty acid and oxidative metabolism, particu-larly the higher phosphorylation of cytoskeleton and biosynthesis indicates their roles toensure the fast growing larvae. These differences in protein expression level illustrate thatspecific protein functions are restricted to particular developmental stage. Our data suggestthe essential changes of proteome and phosphoproteome to trigger the transition ofembryo to larvae. This unravels the molecular event behind the first life cycle transition ofhoneybees and is potentially helpful for future reverse genetic studies in this model insect.

© 2012 Elsevier B.V. All rights reserved.

Keywords:HoneybeeEmbryoLarvaProteomePhosphoproteomeTransition

1. Introduction

The honeybees (Apis mellifera L.) exhibit four developmentalstages in their life cycle, i.e. egg, larva, pupa and adult, which gothrough three weeks before reaching adulthood. This includesthree days as an egg, a period of active embryogenesis [1]; aboutone week as a feeding larva; and the remaining time as a pupaduring which metamorphosis occurs within a sealed cell [2,3].The development of the embryo involves 10 synchronouscleavage mitosis where the resulting nuclei migrate into theperipheral egg layer [1,4]. As a result, conspicuous and relativelyabrupt physical transformation occurs in the embryo structurethrough cell growth and differentiation [5]. Thus, from the

1449.

r B.V. All rights reserved

initial laid egg to the time of hatching, the size,morphology andincubation period of the embryo significantly vary on the basisof both genetic and environmental components [3,6–8]. How-ever, 72 h is generally considered the average incubation timefor the honeybee embryo development into a hatching larva.During this eclosion period before hatching, the eggs graduallysag until it finally rests on the cell floor. The hatching of the egginto the first larval stage is almost indiscernible, and the larvaslowly becomes exposed as the embryo moves and the eggmembrane (chorion) dissolves [3,9]. After successful embryo-genesis, consecutive larval development will continue to en-hance the natural life transition mediated by the stimulation oftwo hormones, juvenile hormone and 20-hydroxyecdysone

.

http://dx.doi.org/10.1016/j.jprot.2012.10.012

mailto:[email protected]


429J O U R N A L O F P R O T E O M I C S 7 8 ( 2 0 1 3 ) 4 2 8 – 4 4 6

(ecdysterone) that control the level of proteins and biologicalactivities associated with its maturation [10,11]. Regardless ofwhether the larva is male or female, it molts five times duringeach consecutive instar and develops into a pupa [12,13].

In general, the developmental changes at early stages are inprincipal arising in response to shifts in the life history of theorganism [14] and are crucial for basic and applied scientificperspective. To this fact, the honeybee embryo has been widelystudied in various scientific approaches such as in vitro rearing[15–17], in-situ hybridization [18,19], cryopreservation [7,20,21],gene based segmental patterning [22,23] and genetic control ofthe early development [24,25]. On the other hand, the honeybeelarvae have been commonly used as a system to investigate theimmune response of certain pathogenic diseases [26–28],genetic manipulation [29] and for queen and worker castdifferentiation in various scientific techniques [30–33]. Further-more, the proteome analysis of the honeybee embryogenesis[34–37], larval development [38,39] and larva cast differentia-tions [40–43] has already been well documented.

To date, proteomics has become amore effective platform tostudy post translational modifications (PTMs). Reversible phos-phorylation is one of the major and widely studied PTMs as it isvital to regulate cellular processes such as signal transduction,and cell differentiation and development [44–46]. Recently, thecombination of proteome and phosphoproteome studies usingfluorescence-based detection technology has become a noveltechnique to identify proteins that change their expression andphosphorylation level in several polychaete species [47–50]. Also,the changes of proteome and phosphoproteome during thelarva–pupa metamorphosis of the cotton bollworm (Helicoverpaarmigera) [51] and the salivary glands of Drosophila melanogasterhave reported [52]. However, despite rapid development inproteomics technologies and their application in honeybee andother insect development, no such study has been conducted onprotein expression and phosphorylation changes associatedwith the first life cycle transition of honeybees. Therefore, thepurpose of this study was to examine the proteome andphosphoproteome changes and deciphers the molecular mech-anism that modulates the transition of the honeybee embryo tolarva. This will help us to gain an in-depth understanding of thehoneybee developmental biology and will be potentially helpfulfor genetic manipulation of the honeybee at the early stagedevelopment.

2. Materials and methods

2.1. Chemical reagents

All the chemicals used for two-dimensional gel electrophoresis(2-DE)were purchased fromSigma (St. Louis,MO,USA) except forBiolyte and immobilized pH gradient (IPG) strips from Bio-Rad(Hercules, CA, USA). Pro-Q Diamond and Sypro Ruby were fromInvitrogen (Eugene, OR, USA). Modified sequencing grade trypsinwas fromRoche (Roche,Mannheim,Germany). All the chemicalsused for RNA isolation and quantitative real-time PCR (qRT-PCR)were from Bio-Rad (Hercules, CA, USA). Chemicals used but notspecified here are noted with their sources in the text. Allreagents used were analytically grade or better.

2.2. Biological sample

Specific aged eggs at 72 h and larvae at 24 h (post-hatching)of the worker honeybees (Apis mellifera L.) were collectedfrom the experimental apiary of Institute of ApiculturalResearch, Chinese Academy of Agricultural Sciences inBeijing. To ensure the exact aged of embryo and larvae to besampled, the queens of five bee colonies were confined to asingle wax comb frame containing worker cells for 5 h with acage made of a queen excluder, through which workers butnot the queen could pass. Subsequently, the queen wasremoved and the eggs contained in the frame weremaintained in the honeybee colony for further development.The developing embryos were collected randomly from thecells of the above five colonies. A total of 3000 eggs and 1000larvae were sampled at each time point. Notably, the larvaewere washed with phosphate buffered saline (PBS) to removethe royal jelly on the surface. All the collected samples werestored at −80 °C until used.

2.3. Protein sample preparation

Protein extraction was carried out according to our previousmethod [38] withminormodifications. Briefly, the fresh sample(1 mg of embryos or larvae/10 μl of buffer) was mixed with alysis buffer (LB, 8 M urea, 2 M thiourea, 4% CHAPS, 20 mMtris-base, 30 mM DTT, and 1.25% Biolyte, pH 3–10). The mixturewas homogenized for 5 min on ice, sonicated for 2 min andcentrifuged twice at 15,000 g at 4 °C for 10 min. The debris wasdiscarded and the supernatant was removed and placed into anew tube. Acetone was added to the collected supernatant for afinal concentration of 80% (V/V) and themixturewas kept on icefor 30 min for protein precipitation. Subsequently, the mixturewas centrifuged twice at 15,000 g at 4 °C for 10 min. Thesupernatant was discarded, and the pellets were used for thenext analysis. Protein concentration was determined accordingto the Bradford method [53] with a DU800 spectrophotometer(Backman Coulter, Los Angeles, CA).

2.4. 2-DE

A 420 μg of protein sample was suspended in 90 μL of LB andmixed with 360 μL of rehydration buffer (8 M urea, 2%CHAPS, 0.001% bromophenol blue, 45 mM DTT, 0.2% Bio-lyte pH 3–10). The mixture was loaded onto a 17 cm IPG strip(immobilized pH gradient, pH 3–10, linear, Bio-Rad). Isoelec-tric focusing (IEF) was performed (Protean IEF Cell, Bio-Rad)at 18 °C according to the following program: 14 h at 50 V,250 V for 30 min (four times), 1000 V for 60 min, 9000 V for5 h, and 9000 V for 60,000 Vh. Prior to SDS-PAGE (sodiumdodecyl sulfate polyacrylamide gel electrophoresis), the IPGstrips were first equilibrated for 15 min in equilibrationbuffer 1 (6 M urea, 0.375 M Tris–HCl pH 8.8, 20% glycerol, 2%SDS, 2% DTT) and later in equilibration buffer 2 (6 M urea,0.375 M Tris–HCl pH 8.8, 20% glycerol, 2% SDS, 2.5%iodoacetoamide) for 15 min. After equilibration, the stripswere transferred to SDS polyacrylamide gels, 12% T sepa-rating gel (1.00 mm). The second dimension electrophoresis,SDS-PAGE, was performed in a Protean II Xi Cell (Bio-Rad) at25 mA/gel for about 5.5 h.

430 J O U R N A L O F P R O T E O M I C S 7 8 ( 2 0 1 3 ) 4 2 8 – 4 4 6

2.5. Multiplex fluorescent gel staining and image analysis

After electrophoresis, a series of stains including Pro-Q Dia-mond, Sypro Ruby and Comassiee Blue Brilliant (CBB, G-250)were applied to each gel according to the previously describedmethod [49] with minor modifications. In brief, the 2-DE gelswere fixed overnight in 40% (v/v) ethanol and 10% (v/v) aceticacid and washed three times with ultrapure water (15 min perwash). To stain the phosphoproteins, the gelswere incubated inPro-Q Diamond solution in the dark for 3 h followed bydestaining with three successive washes of destaining solution(20% acetonitrile (ACN) in 50 mM of sodium acetate, pH 4.0(30 min per wash)). After destaining, the gels were againwashed two times in de-ionized water for 5 min per wash inorder to reduce the possible corrosion on images due to thedestaining solution. The gels were scanned for phosphoproteinspots using a Pharos FX plus system (Bio-Rad, Hercules, CA) atan excitation of 532 nm with a 610 bandpass (BP) 30 emissionfilter. Following this, the images were acquired and then thegels were incubated overnight in the dark with Sypro Ruby fortotal proteindetection. The gelswere subsequently destained inSypro Ruby destaining solution (10% methanol and 7% aceticacid) three times for 30 min each and then washed twice withdistilled water (5 min per wash). They were again scannedusing the Pharos FX Plus system (Bio-Rad, Hercules, CA) at anexcitation of 582 nmwith a 610 BP 30 bandpass emission filter.To prepare for protein spot excision and subsequent massspectrometry (MS) analysis, the gels were further stained withCBB G-250 and the protein spots were visualized by imageScanner III (GE Healthcare, Piscataway, NJ, USA) at 16 bit and300 dpi resolution. The gels from three independent biologicalreplicates were used for the analysis of phosphoproteins andtotal proteins by PDQuest software (version 8.0, Bio-Rad,Hercules, CA, USA). Automatic spot detection in each gel wasverified manually to correct the mismatched and unmatchedspots. The expression levels were determined by the relativevolume of each spot in the gel and expressed as %Vol=[spotvolume/total volumes of all spots resolved in the gels−the totalspot volumes of all spots detected as major royal jelly proteins(MRJP)]. The spot intensities were normalized to equalize thetotal density of each gel. The means and standard deviationsfrom the triplicate experiments were calculated, and thestatistical significance of the level of expression of the proteinsbetween the embryos and larvae was assessed with one-wayANOVA (SPSS version 16.0, SPSS Inc.), and a Duncan's multiple-range test was used to compare the difference between themeans of the expression level of the two developmental stages.Differentially expressed or phosphorylated protein spots with achange in abundance of at least 1.5-foldwith an error probabilityof p<0.05 were considered to be statistically significant.

2.6. Tryptic digestion for MS

The differentially expressed protein spots were manuallyexcised from CBB stained gels of embryo and larva samplesand destained for 30 min using 100 mL of acetonitrile (50%) and25 mM NH4HCO3 (pH 8, 50%) until the gels were transparent.The gels were dehydrated for 10 min with acetonitrile (100%)and dried for 30 min using a Speed-Vac system. To preparetrypsin solution, 2.5 mL of 25 mMNH4HCO3 was added to 25 μg

of trypsin (final concentration 10 ng/μL). Protein digestion andpeptide extraction were done according to our previouslyestablished protocol [54].

2.7. MS analysis and protein identification

The digested protein spots were then analyzed by liquidchromatography-chip/electrospray ionization-quadrupoletime-of-flight/mass spectrometry (LC-Chip/ESI-QTOF-MS)(QTOF G6520, Agilent Technologies), equipped with a capillarypump G1382A, a nano pump G2225A, an autosampler G1377Dand the Chip Cube G4240A. The LC-Chip used (Agilent Technol-ogies) was constituted of a Zorbax 300SB-C18 enrichmentcolumn (40 nL, 5 μm) and a Zorbax 300SB-C18 analytical column(75 μm×43 mm, 5 μm). Loading flow rate was 4 μL/min andloading mobile phase was water with 0.1% formic acid. Elutionfrom the analytical column was performed by a binary solventmixture composedofwaterwith 0.1% formic acid (solventA) andacetonitrile with 0.1% formic acids (solvent B). The followinggradient programwasused: from3% to 8%B in 1 min, from8% to40% B in 5 min, from 40% to 85% B in 1 min and 85% B for 1 min.Chip flow rate was 300 nL/min. MS conditions were: positive ionmode; Vcap: 1900 V; drying gas flow rate: 5 L/min; drying gastemperature: 350 °C; fragmentor voltage: 175 V; skimmer volt-age: 65 V. The digested sampleswere diluted in 20 μLwaterwith0.1% formic acid; centrifuged for 5 min at 10,000×g and injectedwith 8 μL of upper solution. Spectra were calibrated by massreference standard purine and HP-0921 (121. 050873, 922.009798,Agilent Technologies).Tandem mass spectra were retrievedusing the MassHunter software (Version B. 02. 01, AgilentTechnologies). Before MS/MS data searching, peak-list wasgenerated by Mascot Distiller software (Version 3.2.1. 0, MatrixScience). The data were stored in a combined mgf file andsearched against sequence database generated from proteinsequences of Apis mellifera (downloaded May, 2011) augmentedwith sequences from Drosophila melanogaster (downloaded May,2011), Saccharomyces cerevisiae (downloaded May, 2011) andcommon repository of adventitious proteins (cRAP, from TheGlobal ProteomeMachineOrganization, downloadedMay, 2011)using in-house Mascot (version 2.3, Matrix Science,UK.). Searchparameters: carboxymethyl (C) and oxidation (M) were selectedas variable modifications and no fixed modification wasselected. The other parameters used were: taxonomy: allentries; enzyme: trypsin;missed cleavages: 1; peptide tolerance:±1.2 Da, and MS/MS tolerance: ±0.6 Da. The searches wereagainst 72,672 sequences and 37,194,247 residues in thedatabase.

When the identified peptides matched to multiple mem-bers of a protein family, or a protein appeared under the samename and accession number, the match wasmade in terms ofdifferential patterns of protein spots on 2-DE gels. Proteinidentifications were accepted if they contained at least 2identified peptides having both minimal cutoff Mascot scoreof 78 and probability of 95%.

2.8. Biological interaction network (BIN) and functionalenrichment analysis

The identified proteins and phosphoproteins were annotatedby searching against the Uniprot database (http://www.uniprot.

http://www.uniprot.org/


org/) and Flybase (http://flybase.org/) and grouped on the basisof their biological process of Gene Ontology terms. The BIN ofthe identified proteins was predicted using Pathway Studio(Ariadne Genomics) software (http://www.ariadnegenomics.com). Briefly, the protein list was blasted against the Drosophiladatabase that was implemented with the functional relation-ships of proteinmolecules supported by the scientific literature.The applied filters included “all shortest paths between selectedentities”. The information received was narrowed down to ourprotein list of interest, namely, proteins whose involvement inregulatory functions had been observed. Each link was builtwith evidence from at least three publications. Protein entitieswhich belong to different functional groups were representedby different shapes according to the default settings of thesoftware as shown in the legend.

To enrich the identified proteins to specific functional terms,the protein list was analyzed by CluoGo software [55] applyingto the Drosophila database downloaded from the Gene Ontol-ogy database (release date, January 10, 2012). Ontology wasselected as a biological process. Enrichment analysis was doneby right-side hypergeometric statistical testing and the proba-bility value was corrected by Bonferroni method.

2.9. Test of protein expression by qRT-PCR

Total RNA was extracted from both honeybee embryos (n=100)and larvae (n=100) using TRIzol reagent from the five honeybeecolonies as described in Section 2.2. Reverse transcription wasperformed using an RNA PCR Kit (Takara Bio), according to themanufacturer's instructions. The analysis of qRT-PCR wasconducted on 11 differentially expressed proteins selected fromfive major functional groups (carbohydrate metabolism andenergy production, development, protein biosynthesis, cytoskel-eton, and antioxidation) based on their connectivity in thenetwork. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH)was used as the reference (Table S1 of the supporting informa-tion). Real-time PCR was performed using an iQ5 MulticolorReal-Time PCR Detection System (Bio-Rad) in a 25 μL reactionsystem containing 1 μL of cDNA, 5 pmol of forward and reverseprimers, 12.5 μL of SYBR Green Supermix (Bio-Rad), and water.Fold-change was calculated using 2−ΔΔCt method. Each samplewas analyzed independently and processed in triplicate. Thestatistical analysis of gene expression was performed byone-way ANOVA (SPSS version 16.0, SPSS Inc.) using Duncan'smultiple-range test. An error probability p<0.05 was consideredstatistically significant.

3. Results

3.1. Identification of differentially expressed proteins andphosphoproteins

To analyze the proteome and phosphoproteome during thetransition of embryo–larva of honeybee, the 2-DE gels werevisualized by the multiplexed fluorescent gel staining. Fig. 1is a representative 2-DE gel of sequentially stained phos-phoproteins with Pro-Q Diamond dye and total proteinswith Sypro Ruby dye of the last stage of the embryo (72 h)and the first instar larva (24 h post hatching). About 297 and

280 protein spots and 111 and 102 phosphoprotein spotswere detected in the embryo and newly hatched larvae,respectively (Fig. 2A) indicating that about 37% of the totalproteins were modified by phosphorylation at both devel-opmental stages. Also, the majority of the phosphoproteinswere focused at pI range of 4.0–6.7 except one phosphopro-tein spot (spot 67) with experimental pI of 9.0; while theSypro-Ruby stained total proteins focused evenly at pI rangeof 3–10 (Fig. 1).

About 95 total protein and 54 phosphoprotein spots thataltered their expression (>1.5 fold, p<0.05) at the twodevelopmental stages were selected for further analyses.Among these, 65 total proteins and 34 phosphoproteinswere successfully identified as being of Apis mellifera origin(Tables 1 and 2). The remaining proteins and phosphopro-teins were not identified either because of their lowabundance in producing enough spectra or because thedatabase search scores cannot yield unambiguous results(<95%). Moreover, most of the highly sensitive fluorescentstained proteins were present in low abundance and underthe detection limit of CBB stain during spot excision. In allour further analysis, we excluded the MRJPs detectedexclusively in the larvae to avoid biasness as they areexogenous proteins secreted into larval diet by the nursebees.

3.2. Qualitative comparison of differentially expressed andphosphorylated proteins

A total of 99 differentially regulated andphosphorylatedproteins(65 total proteins and 34 phosphoproteins) were classified intoeight functional categories. Among them, proteins involved incarbohydrate metabolism and energy production were themostabundant group (20), followed by proteins associated withdevelopment (17), protein folding (15), cytoskeleton (15), proteinbiosynthesis (11), antioxidant system (10), amino acid andnucleotide metabolism (6) and fatty acid metabolism (5) (Fig. 3).About 85% of the identified protein species were shared incommon by the two developmental stages and there were noconsiderable differences in protein species. However, there weresignificant variations between the two developmental stages interms of the number of up-regulated proteins and phosphopro-teins across each functional category (Fig. 4). In regard to thisfact, a total of 54 (54.2%) and 45 (45.5%) of differentially expressedand phosphorylated proteins were upregulated by the embryosand the larvae, respectively. Briefly, 37 total proteins and 17phosphoproteins were downregulated and 28 total proteins and17 phosphoproteins were upregulated during embryo–larvaetransition (Fig. 2B).

To compare the relative representation of phosphoproteins,we analyzed the percentage proportion of phosphoproteins ineach respective functional category of total proteins. Accord-ingly, the cytoskeleton group involved the most abundantlyphosphorylated protein spots, 53% (8 of 15), followed bymetabolismof amino acids andnucleotides, 50% (3 of 6), proteinbiosynthesis, 45% (5 of 11) and protein folding, 40% (6 of 15) ineach of their respective functional group. Similarly, proteinsrelated to carbohydrate and energy metabolism and develop-ment were shared 30% (6 of 20) and 18% (3 of 17) ofphosphoproteins in their group, respectively (Fig. 3).

http://www.uniprot.org/

http://flybase.org/

http://www.ariadnegenomics.com

http://www.ariadnegenomics.com

Fig. 1 – Representative 2-DE gel analysis of last day embryo (72 h) and first instar larvae (24 h post-hatching) of honeybeeworkers (Apis mellifera L.). Upper panel is 2-DE gels stained with the total protein fluorescent dye (Sypro Ruby). Lower panel is2-DE gels stained with the phosphoprotein-specific fluorescent dye (Pro-Q Diamond). Proteins are separated on 17 cm IPG gelstrips (pI 3–10 Linear) with 420 μg of sample loaded, followed by 12.5% SDS-PAGE on a vertical slab gel. Differentially expressedprotein spots of known identity are labeled with color codes, where red indicates up-regulation and blue indicatesdown-regulation at each developmental stages.


3.3. Quantitative comparisons of differentially expressedproteins and phosphoproteins

To estimate the extent of biological function of each protein, theexpression level of all identified proteinswas performed for both

embryos and larvae using the ANOVA log ratio (|log1.5 Ratio|≥1).The p-values for the differentially expressed and phosphorylat-ed proteinswere calculated as the ratio of the protein abundance(≥1.5 fold changes and p≤0.05) (larvae/embryo) according toTables 1 and 2 (Fig. 5). As a result, all proteins up-regulated in

Fig. 2 – Comparison of proteome and phosphoproteome between the last day embryos (72 h) and first instar larvae (24 hpost-hatching) of honeybee workers (Apis mellifera L.). A is the number of protein and phosphoprotein spots reproduciblydetected in embryos and larvae of honeybee worker. B is the number of differentially expressed total protein spots andphosphoprotein spots (p<0.05, >1.5 fold) during the period of embryo–larva transition.


both developmental stages showed stronger expressions (Fig. 5),which were more than the minimum threshold of 1.5 foldchange. Subsequent analysis of the 54 proteins upregulated bythe embryo showed eight protein spots with >10 fold higher intheir expression: vitellogenin (spot 42), 14-3-3 zeta (spot 47),ubiquitin-protein ligase (Prp19) (Spot 21), TCP-1-alpha (Spot16),yellow-g proteins (Spot 58, 66), RuvB-1 (Spot 30), MACD (Spot 44)and 26S proteasome subunit P45 (Spot 39). On the other hand, ofthe 45 proteins upregulated by the newly hatched larvae, nineproteins; (GH07301p (spot 57), dynamitin 2 (p50) (spot 40), HSP60(Spot 65), β-Tubulin at 56D (spot 48), cathD (spot 51), Trxr-1(spot69) trxr-2 (spot 98), translation elongation factor1(spot 73),ACAD1(Spot 46) and SEC13(spot 71), showed >10 fold expression.Moreover, the divergence in the level of each protein expressionbetween the two developmental stages was more profound ateach biological function.

Accordingly, of the 20 differentially expressed proteins andphosphoproteins related to the carbohydrate metabolism andenergy production, 14 (70%) were upregulated by the embryocompared to the 6 upregulated by the larvae (30%). Likewise,of the 17 proteins associatedwith development, 12 (70.6%) and 5(29.4%) were upregulated by the embryos and the larvae,respectively. Also, of the 6 proteins implicated in amino acidmetabolism, 5 (83.3%) were overregulated in the embryoscompared to the 1 overregulated by the larvae (16.6%). On theother hand, the larvae upregulated more proportions of pro-teins than in the embryos such as protein folding (8 of 15),cytoskeleton (10 of 15), antioxidant system (6 of 10), proteinsynthesis (6 of 11) and fatty acid metabolism (3 of 5) (Fig. 5,Tables 1 and 2).

3.4. Functional enrichment of differentially expressedproteins

Gene Ontology (GO), the de facto standard in gene function-ality description, is used widely in functional annotation andenrichment analysis. To enrich the identified proteins to

specific functional terms, the entire protein spectrumfrombothembryo and larvae was analyzed by CluoGo software. Accord-ingly, the identified proteins were significantly enriched to fourmajor functional groups, i.e. carbohydrate metabolism andenergy production (10, or 46%), folding associated proteins (6, or27%), protein biosynthesis (5, or 23%) and cytoskeleton (1, or4.5%) (Fig. 6).

3.5. Biological network analysis

Since proteins perform their function in networks, we analyzedall the pathways and interactions connected to all the identifiedproteins hoping to find the possible key node proteins during theembryo–larva transition using Pathway Studio software. Hence,from the 99 identified proteins, 56were linked to theBIN throughthe shortest path (Ariadne Genomics) on the basis of diverselinkage relationships such as protein–protein interactions (PPI),modifications, regulation of expression, etc. Consequently, pro-teins involved in carbohydrate metabolism and energy produc-tion were the most abundant in the BIN (15, or 26.7%). Of which,nine proteins, vha55 (spot 38), nurf-38 (spot 59), Aldo/keto (spot53), scs-fp (spot 61), Eno (spot 36), argk (spot 49), gbp(prp19) (spot21), l(1)g0022 (spot 14) and Mdh1(spot 67), were upregulated inthe embryo, and six proteins, Atpsyn-d (spot 88), GH07301p(spot 57), Aldh (Spot 34), CG12140 (ETF-QO) (spot 11), l(1)g0230(spot 92) and cg8036 (spot 7), were upregulated in thelarvae. Developmental proteins were the second most repre-sented group (10, or 17.9%) in theBIN inwhich seven, sn (spot 13),Tcp-1η (spot 19), idgf4 (spot 29), TER94 (spot 1), Ben (spot 96), caf1(spot 26) and phosphoprotein,14-3-3zeta (spot 74, 75), wereupregulated in the embryo, while three, awd (spot 93), l(2)37cc(spot 77) and nacα (spot 60), upregulated in the larvae. On theother hand, cytoskeletonwas networked through seven proteins(12.5%) inwhich five, sec13 (spot 71), tsr (cadf) (spot 91),βTub56D(spot 48), sop2 (spot 99) and cyp1 (spot 94), were upregulated inthe larvae and twoproteins,αTub84B (spot 35) andyellow-g (spot58), were upregulated in the embryo. Likewise, of eight (14.3%)

image of Fig.�2

Table 1 – Identification of differentially expressed proteins during transition of the last day embryo to first instar larvae ofhoneybee worker (Apis mellifera L.).

Spotno.

ExperimentalMr(kDa)/pI

TheoreticalMr(kDa)/pI

Sequencecoverage

(%)

Matched Score Accessionnumber

Protein name Logratio(1.5)(+/−)

p-Value

Development1 83.78/5.12 89.47/5.18 51 60 2555 gi|66534286 Transitional endoplasmic reticulum

(TER94), (CG2331)−3.0 0.002

9 70.26/6.48 36.09/4.60 52 27 832 gi|66530527 Annexin IX (AnnIX) CG5730-PC −1.6 0.00010 68.27/8.36 81.71/9.40 27 36 1642 gi|110761874 Insulin-like growth factor-II

(IGF-II mRNA)(imp) CG1691-PA1.9 0.000

13 63.85/6.38 58.78/6.06 28 15 511 gi|66549818 Singed (Sn) CG32858-PA −2.4 0.00119 61.06/6.34 60.41/6.03 52 33 828 gi|66540596 Chaperone, tailless complex

polypeptide 1(Tcp-1η) [EC.3.6.1.3]CG8351-PA

−4.0 0.015

26 55.22/4.75 48.49/4.72 27 11 210 gi|66534191 Chromatin assembly factor 1subunit (Caf1) , CG4236-PA

−2.2 0.000

29 55.74/7.67 48.99/8.06 31 14 189 gi|66514614 Imaginal disc growth factor 4 (Idgf4)[EC 1.2.1.3] CG1780-PA

−2.0 0.005

42 45.61/5.62 20.21/6.29 3 12 103 gi|58585104 Vitellogenin (Vg), CG11129 −1047 40.8/4.90 28.17/4.79 41 11 554 gi|48097086 Leonardo protein (14-3-3 zeta)

isoform 1, CG17870−10

60 33.38/4.75 22.73/4.70 2 2 93 gi|66500427 NAC-alpha (Nacα) CG8759-PB 1.8 0.01872 33.35/5.83 20.21/6.29 5 9 229 gi|58585104 Vitellogenin (Vg) −2.1 0.00577 28.39/6.75 29.99/6.54 65 33 1207 gi|48097857 lethal (2) 37Cc (l(2) 37Cc) CG10691-PA 2.4 0.00093 16.66/6.69 17.69/6.75 70 14 296 gi|66520497 Nucleoside diphosphate kinase

(awd), [EC.2.7.4.6]2.5 0.003

96 16.39/5.42 17.27/5.71 29 7 222 gi|66564615 Bendless (Ben) [EC=6.3.2.19 ]CG18319-PA

−3.7 0.000

Carbohydrate and energy metabolism2 84.57/7.0 99.32/6.50 26 18 292 gi|66550870 Iron regulatory protein 1B (Irp-1B),

CG6342-PA, [EC=4.2.1.3]−2.8 0.008

7 73.47/7.01 67.77/7.62 33 42 1154 gi|110751363 Transketolase CG8036-PB,[EC=2.2.1.1]

1.6 0.009

11 68.6/7.1 67.62/7.26 30 9 236 gi|66540209 Electron transfer flavoprotein-ubiquinone oxidoreductase (ETF-QO),CG12140, [EC=1.5.5.1]

1.3 0.016

14 64.01/6.5 44.21/6.58 56 35 812 gi|66513205 lethal (1) G0022 CG8231-PA (Tcp-1ζ)[EC:3.6.3.14]

−3.8 0.018

21 59.33/6.73 55.30/6.36 36 21 569 gi|66513511 GTP-binding-protein (Prp19)CG5519-PA, EC=6.3.12.9

−10

34 53.74/6.08 55.94/6.69 40 28 992 gi|66530423 Aldehyde dehydrogenase (Aldh)CG3752-PA, [EC=1.2.1.3]

1.4 0.004

36 51.17/6.6 40.18/5.51 50 31 1571 gi|110761968 Enolase (Eno), [EC=4.2.1.11],CG17654-PA

−4.7 0.000

38 48.08/8 55.39/5.41 39 18 615 gi|66531434 Vacuolar H+-ATPase 55kD B subunit(Vha55), CG17369-PB

−4.9 0.039

49 39.46/5.9 40.33/5.66 56 27 782 gi|58585146 Arginine kinase (argk), CG32031[EC=2.7.3.3]

−1.8 0.001

53 34.71/6.6 36.46/6.26 41 16 488 gi|66525576 Aldehyde reductase (Aldo/keto)[EC=1.1.1.21],CG6084-PA

−4.5 0.014

54 34.67/6.84 36.46/6.26 50 29 897 gi|66525576 Aldehyde reductase (Aldo/keto)[EC=1.1.1.21],CG6084-PA

−3.2 0.009

57 32.99/6.37 35.67/6.09 20 5 182 gi|48099074 GH07301p, CG9119 [EC=3.1.-.–] PA 1059 32.98/5.15 85.57/6.31 20 15 515 gi|66507623 Nucleosome remodeling factor -38kD

(Nurf-38) [EC=3.6.1.1], CG4634-PA−4.7 0.000

61 33.16/4.6 73.27/6.75 30 13 311 gi|66505480 Succinate dehydrogenase (Fp)CG17246, [ EC 1.3.5.1]

−2.8 0.000

Protein folding5 79.41/6.06 83.79/4.98 12 8 230 gi|229892248 heat shock protein 90 ( HSP90),

CG27202.9 0.000

16 63.59/5.75 60.59/5.69 44 38 294 gi|66560172 T-complex protein 1 subunit alpha(TCP-1-alpha), CG5374

−10


Table 1 (continued)

Spotno.



Sequencecoverage

(%)



p-Value

17 61.05/5.75 59.89/5.70 56 50 1514 gi|66522349 T-complex Chaperonin 5,(Cct5)CG8439-PA

2.9 0.023

18 61.1/6.18 60.41/6.03 54 34 1147 gi|66540596 T-complex protein 1, beta subunit(Tcp-1eta ) [ EC.3.6.1.3], CG8351-PA

−3.4 0.044

20 60.89/6.45 58.32/6.22 59 45 1130 gi|66533395 T-complex protein 1 subunit alpha(TCP-1 alfpha) CG7033-PA

−3.9 0.005

22 59.72/7.52 57.85/7.10 48 27 683 gi|66558942 T-complex protein 1 subunit delta(tcp-1-delta, [EC=.3.6.1.3] CG5525-PA

−6.5 0.001

40 45.9/6.4 46.20/6.93 66 33 162 gi|110755392 Dynamitin 2 (p50), CG8269 1065 32.3/6.3 31.54/6.65 62 87 538 gi|66547450 60 kDa heat shock protein, (HSP60)

CG1210110

83 24.6/4.53 17.22/4.55 61 15 339 gi|66510528 Fkbp13 [EC=5.2.1.8] CG9847-PA 1.8 0.022

Cytoskeleton12 66.64/6.58 67.42/6.32 50 20 522 gi|66538420 Putative actin-interacting protein 1

(AIP1), CG10724−5.0 0.000

48 43.0/5.1 45.59/5.75 13 13 212 gi|48095525 β-Tubulin at 56D (βTub56D)CG9277

10

66 32.29/7.13 41.48/5.38 15 27 187 gi|48137874 yellow-g CG5717-PA −1089 19.51/5.02 14.06/5.14 52 23 106 gi|66510581 tubulin-specific chaperone (tbcE),

CG78614.0 0.007

94 17.59/9.27 23.02/9.27 45 15 373 gi|66534750 Cyclophilin 1(cyp1)[EC=5.2.1.8]CG9916-PA

3.8 0.015

95 16.2/4.53 16.58/4.65 50 10 292 gi|66558818 Myosin-2 essential light chain(MEC-2), CG3201

3.1 0.001

99 14.2/5.59 13.94/5.64 58 9 266 gi|56404802 Suppressor of profilin 2 (Sop2),CG8978

2.3 0.016

Antioxidant system31 54.34/7.4 53.67/6.90 42 19 424 gi|33089108 thioredoxin reductase (trxr-1) CG2151

[ EC=1.8.1.9]−3.7 0.000

32 50.4/7.5 54.69/6.90 48 22 512 gi|295842224 thioredoxin reductase (trxr-1), CG2151[ EC=1.8.1.9]

−3.8 0.000

51 38.1/5.0 40.65/5.6 49 17 262 gi|66560290 CathD CG1548-PA 1064 32.37/5.24 34.96/5.34 45 8 348 gi|48132928 epsilon subunit of coatomer protein

complex (β'Cop) CG66991.6 0.014

69 30.5/4.7 32.03/4.94 37 53 485 gi|66524108 thioredoxin reductase (trxr-1), CG2151[ EC=1.8.1.9]

10

87 23.36/8.34 26.61/6.45 36 11 483 gi|110749015 Short-chain dehydrogenase/reductase(SDR), CG10962-PB

1.7 0.011

98 14.6/5.4 12.18/4.82 38 7 460 gi|48104680 thioredoxin-2 (Trx-2) [EC=1.8.1.8]CG31884-PA

10

Protein biosynthesis30 54.81/6.83 50.16/6.32 24 13 14 gi|48106137 RuvB-like protein 1(RuvB-l1) (EC=3.6.1.-) −1041 47.69/7.38 53.01/8.16 55 34 989 gi|66518848 Elongation factor T u mitochondrial

(eftum) CG6050-PA-2.5 0.005

55 33.23/5.6 34.56/5.75 48 23 870 gi|66559310 60S acidic ribosomal protein P0 (rplp0),CG7490, [EC=4.2.99.18]

−2.5 0.014

70 30.66/7.65 36.25/7.62 52 30 569 gi|48104663 Guanine nucleotide-binding proteinsubunit beta-(Gbeta 13 F)

3.7 0.000

73 30/6.8 49.24/6.02 8 3 78 gi|110761214 translation elongation factor 1-gamma(Ef1γ) isoform 2

10

97 14.46/5.14 14.79/5.16 20 3 77 gi|48104167 Nuclear transport factor-2 (Ntf-2)CG1740-PA

−1.9 0.003

Fatty acid metabolism3 78.99/8.34 376.750/8.34 11 57 2084 gi|110758758 Retinoid and fatty acid binding protein

(Rfabg), CG11064-PA isoform 12.1 0.013

37 50.83/7.05 53.45/7.93 34 21 556 gi|66531425 Acetyl-CoA hydrolase/transferase,CG7920-PA

2.0 0.007

(continued on next page)

Protein folding


Table 1 (continued)

Spotno.



Sequencecoverage

(%)



p-Value

44 43.88/6.59 46.83/8.51 39 13 329 gi|66499429 Probable medium-chain specificacyl-CoA dehydrogenase,mitochondrial (MCAD) [EC:1.3.99.3]

−10 0.000

46 43.9/6.6 46.83/6.51 45 248 499 gi|66499429 Acyl-CoA dehydrogenase CG12262-PA 1063 32.4/5.08 34.52/5.15 33 17 470 gi|48141571 4-nitrophenyl phosphatase (Phos5567)

CG5567-PA (EC:3.1.3.41)−2.3 0.001

Amino acid and nucleotides metabolism58.5/7.46 62.06/8.24 60 27 641 gi|66523390 Glutamate dehydrogenase (GDH)

[EC 1.4.1.3] CG5320-PF−3.2 0.000

39 49.12/5.99 48.83/5.81 32 17 79 gi|66536893 26S proteasome subunit P45 (Rpt1)[EC=3.6.1.3] CG1341-PA

−10

43 84.06/7.4 62.06/8.24 57 26 520 gi|66523390 Glutamate dehydrogenase, (GDH),[ EC 1.4.1.3], CG5320-PF

−1.2 0.033

Fatty acid metabolism


proteins belonging to the biosynthesis, five were upregulated inthe larvae ef1γ (spot, 73), gbeta13f (spot 70) crb (spot 82), eif-4a(spot 28) and eif-5a (spot 90), whereas three protein spots, Ntf-2(spot 97), eftum (spot 41) and rplp0 (spot 55), were upregulated inthe embryo. Similarly, of the six (10.7%) folding proteins in theBIN, four of them, tcp-1alfpha (spot 20), tcp-1-delta (spot 22),hsc70-4 (spot 8) and hip-r (spot 33), were upregulated in theembryo but, two spots, dmn (spot 40) and hsc70-3 (spot 6), wereupregulated in the larvae. Also, of the four (7%) antioxidantproteins in the BIN, equal proportions of proteins (2, 50%) wereupregulated at both developmental stages. Likewise, four (7%)proteins related to the metabolism of amino acids wereupregulated, three (75%) in the embryo and only one (25%) inthe larvae. On the other hand, two (3.5%) proteins associatedwith the metabolism of fatty acids networked in the BIN wereupregulated by the larvae (Fig. 7).

3.6. Test of differentially expressed proteins by qRT-PCR

To test the level of protein expression at the transcript level, 11key node proteins in the BIN from five major functional groups(carbohydratemetabolismandenergyproduction, development,protein biosynthesis, cytoskeleton, and antioxidation) wereselected for qRT-PCR analysis. The trend of mRNA expressionshowed that the expression of eight proteins, transketolase (spot7), idgf4 (spot 29), argk (spot 49), trx-2 (spot 81), vg (spot 72), ntf-2(spot 97), jafrac1 (spot 84) and eIF-5A (spot 90) was consistentwith their encoding genes. However, three genes, βTub56D (spot48), aldehyde reductase (spot 53) and awd (spot 93) showedmRNA–protein expression inconsistency, which may be dueto the lack of a direct relationship, or unsynchronized genetranscription and translation (Fig. 8).

4. Discussions

The last day embryo developing into the first instar larva afterhatching is the first stage of transition in the complete life

cycle of the honeybee. At this critical stage, the embryo fullydevelops into an immature larva and hence, vital proteins andphosphoproteins to be involved in supporting this transitionalprocess. To gain an in-depth understanding of molecularevents behind this process, we compared the global proteomeand phosphoproteome of the honeybee embryos (72 h) andnewly emerged larvae (24 h post hatching). The existences ofsignificant differences in protein expression level and phos-phorylation status between the two developmental stagescontribute to the transitional process from the last day embryoto the first instar larva. In addition, specific expressions ofembryo and larval stage proteins indicate that each of the twodevelopmental stages has their own definite quantitative andqualitative biological requirement to perform specific functions.Accordingly, 99 differentially expressed and phosphorylatedproteins (65 total protein spots and 34 phosphoprotein spots)were detected to be vital for the honeybee in the earliestdevelopmental shift. Of these, proteins related to energy produc-tion and metabolism, development and amino acid metabolismwere strongly expressed in the embryo (Fig. 4), suggesting theembryo demand of high metabolic enhancing proteins duringactive embryogenesis [1], specifically for muscle contraction andabdominal peristalsis upon hatching [8,9]. The higher levels ofprotein expression and phosphorylation were involved in cyto-skeleton, protein biosynthesis, protein folding, fatty acid andoxidative metabolism in the newly hatched larvae (Fig. 4) ingeneral, signify the synthesis and translation of proteins inresponse to high rate of biogenesis and structural constituent offast developing larvae [56]. This is useful for studying earlydevelopmental stages where regulatory mechanisms are trig-gered by protein expression changes and phosphorylation status[57,58]. Moreover, the large number of key node proteins (56)networked in the BIN ismost likely towork together and play thepotential role in the regulatory pathway during the embryotransition to larvae.

The 34 differentially regulated phosphoproteins (Fig. 5B,Table 2) at both developmental stages signify the potentialrole of phosphorylation in the cellular and physiological

Table 2 – Identification of differentially expressed phosphoproteins during transition of the last day embryo to first instarlarvae of honeybee worker (Apis mellifera L.).

Spotno.



Sequencecoverage

(%)



p-Value

Cytoskeleton15 63.3/6.6 67.420/6.32 51 19 619 gi|66538420 Putative actin-interacting protein 1

(AIP1), CG10724−4.0 0.007

23 26.7/6.7 50.62/5.01 40 30 1115 gi|66521545 Tubulin alpha-1chain (αTub84B),CG1913

1.6 0.006

27 57.6/5.1 50.6/4.75 73 101 1860 gi|48095525 Tubulin at 56D, isoform B (βTub56D),CG9277-PB

1.6 0.003

35 53.2/4.5 50.6/5.01 28 8 415 gi|66521545 Tubulin alpha-1 chain (αTub84B),CG1913

−2.7 0.001

45 48.4/5.4 42.20/5.3 47 31 782 gi|66509769 Actin-87E isoform 2 (Act87E), CG18290 2.3 0.00258 36.9/4.99 41.48/5.38 13 10 373 gi|48137874 Yellow-g, CG5717-PA −1071 30.0/4.7 34.74/4.99 25 12 392 gi|110756630 SEC13-like 1 isoform b (Sec13),CG6773 1091 18.81/4.86 17.05/6.17 43 7 261 gi|110751158 Actin-depolymerizing factor

homolog (Cadf), CG42543.0 0.001

Carbohydrate and energy metabolism50 41.9/6.3 40.32/5.66 52 50 781 gi|58585146 Arginine kinase (argk), CG32031

[EC 2.7.3.3]−2.3 0.005

52 39.6/5.8 40.33/5.66 30 4 118 gi|58585146 Arginine kinase (argk) CG32031[EC 2.7.3.3]

−2.0 0.007

56 33.35/5.83 36.46/6.26 26 6 261 gi|66525576 Oxidoreductase CG6084 [EC 1.1.1.21] −1.6 0.00567 33.7/9.00 35.95/9.33 47 28 1216 gi|66513092 mitochondrial malate dehydrogenase

precursor (Mdh1) [EC 1.1.1.37]−1.8 0.026

88 22.6/5.15 20.32/5.00 67 29 773 gi|48098315 ATP synthase D chain, mitochondrial(ATPsyn-d) [EC=3.6.3.14], CG6030

1.8 0.002

92 17.1/4.3 17.60/5.09 34 6 139 gi|3399724 Similar to lethal(1) GO230 CG2968-PA 1.5 0.002

Protein folding4 73.9/5.2 83.79/4.98 50 67 2182 gi|229892248 Heat shock protein 90 (HSP90) −1.6 0.0006 70.3/5.4 72.87/5.29 60 96 1649 gi|229892214 Heat shock protein cognate 3

(Hsc70-3), CG41472.7 0.004

8 67.1/5.70 71.38/5.43 46 43 969 gi|229892210 Heat shock protein cognate 4(hsc70-4), CG4264

−4.3 0.002

25 56.8/4.45 47.49/4.45 51 28 688 gi|66545506 Calreticulin (Crc), CG9429-PA 3.4 0.00333 54.4/4.84 42.68/4.84 36 24 572 gi|66511001 Hsc70-interacting protein 2 (hip-r),

CG2947-PA−1.8 0.011

79 28.5/5.32 23.56/5.2 52 16 434 gi|48121613 Immunoglobulin E-set (RhoGDI),CG7823-PA

2.3 0.002

Protein biosynthesis28 55.3/5.6 48.52/5.29 37 37 231 gi|66551115 Eukaryotic initiation factor 4A

(eIF-4A) isoform 2, CG90752.1 0.009

80 26.7/6.73 26.08/6.97 66 26 859 gi|66564402 Calcyclin binding protein (Cacybp) −1.9 0.00682 23.4/4.23 20.22/4.34 22 7 137 gi|66547438 Crumbs, (crb), CG16817-PA 2.0 0.00785 23.8/4.7 19.83/4.57 79 48 1510 gi|66515987 Translationally controlled tumor

protein (Tctp) CG4800-PA isoform 13.7 0.000

90 18.5/5 17.92/5.19 58 6 773 gi|110767655 Eukaryotic translation initiationfactor 5A (eIF-5A) CG3186-PA

2.0 0.004

Development68 31.8/4.81 28.17/4.79 27 43 257 gi|48097086 14-3-3 protein zeta (14-3-3EZ) CG17870 4.4 0.00174 28.0/4.8 28.17/4.79 59 25 683 gi|48097086 14-3-3 protein (14-3-3 zeta) −2.4 0.03275 28.18/4.97 28.17/4.79 41 11 554 gi|48097086 Leonardo protein (14-3-3 zeta)

isoform 1−2.7 0.000

Antioxidant system81 24.7/5.3 25.51/5.14 61 31 980 gi|48103506 Thioredoxin-2 (Trx-2) [EC.1.8.1.8] 1.7 0.00484 24.2/5.7 21.94/5.65 69 23 834 gi|66548188 Thioredoxin peroxidase 1(jafrac1)

[EC.1.11.1.15]−1.9 0.046

86 23.6/5.81 21.944/5.65 69 36 1212 gi|66548188 Thioredoxin peroxidase 1(jafrac1) [EC.1.11.1.15]

−2.0 0.000

(continued on next page)


Table 2 (continued)

Spotno.



Sequencecoverage

(%)



p-Value

Amino acid and nucleotides metabolism62 36.9/4.42 27.52/4.5 26 13 508 gi|66508906 CG10527-PA 2.0 0.04076 28.5/5.7 28.11/5.75 34 22 444 gi|48096769 Proteasome subunit α type 3

(Prosα3T), G1736−1.9 0.038

78 27.0/4.7 26.82/4.83 66 45 441 gi|66541426 Proteasome subunit α type 5(Prosα5) CG10938

−1.8 0.011

Notes: Spot number corresponds to the number of protein spots in Fig. 1. Theoretical molecular weight (Mr) and isoelectric point (pI) of theidentified proteins were retrieved from the protein database generated from protein sequences of Apis mellifera (downloaded May, 2011)augmented with sequences from Drosophila melanogaster (downloaded May, 2011), Sacharomyces cerevisiae (downloaded May, 2011) and commonrepository of adventitious proteins (cRAP, from The Global Proteome Machine Organization, downloaded May, 2011) using in-house Mascot(version 2.3, Matrix Science,UK.). Experimental Mr and pI were calculated with PDQuest software and internal standard molecular massmarkers. Sequence coverage is the ratio of the number of amino acids in every peptide that matches with the mass spectrum divided by thetotal number of amino acids in the protein sequence. Matched peptide is the number of paring an experimental fragmentation spectrum to atheoretical segment of protein. Mascot score is search against the A. mellifer database. Protein name is given when proteins were identified byLC-Chip ESI-QTOF-MS. Accession number is the unique number given tomark the entry of a protein in the NCBInr database. The ratios (R) of theprotein abundance (larva/embryo) were transformed and proteins with |log1.5 R|>1 and p-value Pe0.05 were considered as differentiallyexpressed proteins and phosphoproteins. The log ratios of uniquely expressed proteins at each developmental stage were limited to 10. Positive(+) log ratios indicate proteins upregulated in the larvae and the negative (−) by the embryo.

Fig. 3 – Functional annotations and percentage representation of total proteins and phosphoproteins altered their expressionduring the embryo–larvae transition of the honeybee worker (Apis melifera L.). Red and blue color codes represent for comparedpercentage of proteins and phosphoproteins, respectively as normalized to the total number of differentially expressedproteins in each functional category.


image of Fig.�3

Fig. 4 – Comparison of upregulated protein numbers of lastday embryo and first instar larvae, where the red color codesstand for the total proteins and the black color codes standfor phosphoproteins. And the first bar of similar color codefrom each functional category represents value for the“embryo” and the second bar for the “larva” of honeybeeworker (Apis melifera L.).


regulatory mechanisms of embryo–larva transition as phos-phorylation plays an important role in cell differentiation,development, transcription and metabolism in eukaryoticorganisms [59–61].In this study, cytoskeleton protein thatcomprises the most phosphorylated groups (Fig. 3) suggeststhat they control a multitude of processes, including cell shapeand integrity, division, movement, and intracellular transportduring embryo–larvae transition [62]. Likewise, protein biosyn-thesis is the second most phosphorylated group as involved inthe synthesis and translation of proteins in the newly hatchedlarvae [63]. The spatial focusing of the phosphoproteins on 2-DEat pI of 4.0–6.7 and the Sypro-Ruby stained total proteins at pI of3–10 (Fig. 1) indicates that most phosphoproteins are acidic innature and phosphorylation might increase the Mr of proteinsto more acidic pI value [51].

Interestingly, the downregulation of 54 (54.5%) differentiallyregulated proteins (37 total proteins and 17 phosphoproteins)during embryo–larva transition (Fig. 2B) could be related to therupturing and digestion of the eggshell (chorion) upon hatchingand the degradation of some embryo specific tissues that are nolonger needed by the larvae [7,9,12,13]. This reveals that somestage specific proteins show a shift in the expression level in linewithdevelopmental stages as a signof task accomplishment. Forexample, vitellogenin (vg) is a precursor of the egg-yolk proteinsthat has flexible regulatory functions primarily as the source offood during embryonic development [30]. Nevertheless, thisprotein is downregulated as the larvae entirely change theirsource of nutrition to the royal jelly immediately after hatching[64–66]. Interestingly, the isoforms of yellow-g protein (spot 58and 66) (Tables 1 and 2) that involves in the formation andhardening of the embryo eggshell (chorion) [67] were uniquely

expressed in the embryo (>10 fold change), while exclusivelyabsent in the larvae demonstrate the degradation of the embryoeggshell after hatching.

The embryo overexpression of carbohydrate and energymetabolism proteins (14 of 20) compared with the larvae (6 of20) (Fig. 4) is likely linked to high metabolic rates duringembryogenesis, from oviposition to hatching [1,12]. This alsosuggests that much metabolic energy is required to producethe honeybee embryo since some tissuesmust respond to theinitiation of metamorphosis by increasing their metabolicactivity for additional cell divisions, tissue differentiations andgerm band segmentation [7,9]. These results are in accordancewith observations that the tracheal network becomes visibleabout 2 h before egg hatch, and the upright embryo flexing andabdominal peristalsis occur [7]. To this end, important proteinsimplicated in carbohydrate metabolism and energy productionsuch as prp19 (spot 21), vha55 (spot 38), aldehyde reductase(spot 53, 54), eno (spot 36), nurf-38 (spot 59), argk (spot 49, 50, 52),Mdh1 (spot 67) and succinate dehydrogenase (spot 61), werestrongly expressed in the embryo (Fig. 5). Specifically, theenzymes of GTP-binding-protein (prp19) with the highestexpression (>10 fold change) involves in the catalytic nuclearmRNA splicing, via spliceosome may support tissue differenti-ationby cell divisionduring embryogenesis [68]. The isoformsofaldehyde reductase most likely act as a key metabolic fuel inresponse to high metabolic rates to ensure the muscularcontractions and abdominal peristalsis by the embryo uponhatching [9]. Arginine kinase, catalyzing the reversible phos-phorylation of arginine ATP generation, and the physiologicalimportance of this reaction is in the provision of metaboliccapacitance [69]. This enzyme was phosphorylated in a currentstudy and all of its isoforms were upregulated in the embryoindicating the corresponding upregulation of phosphoarginineand the consequent greater energy release that is moreapparent during active embryonic cell division and organogen-esis [70] as act in tobacco hornworm (Manduca sexta) [71].

On the other hand, the newly emerged larvae experiencerelatively slow metabolic rate and less gene expressioninstantly after transition from the egg [72]. It also furthercoincides with the down-regulation of the gene encodingenergy metabolism enzymes during larval ecdysone pulse inDrosophila [73]. Nevertheless, the upregulation of some larvalspecific proteins involved in carbohydrate and energy metab-olism is supposed to ensure the energy demand of newlyhatched larvae to facilitate proper cell divisions toward propermorphological and physiological development. For instance,ATPsyn-d, a mitochondrial proton transport for ATP synthesisand lethal (1) G0230 that produces ATP through an oxidativephosphorylation process [74,75], plays major roles in themetabolic activity of growing larvae by increasing theirexpression level.

For most organisms, early stage development is vital as itis the initial stage for life transition and complex cell signalingpathways [76]. In holometabolous insects including honey-bees, the imaginal disks containing the precursor cells thatwill give rise to the appendages of the adult, i. e. the eyes,head appendages, legs, genitalia and wings, are known to beformed during embryonic development [77]. Therefore, theupregulation of large number of developmental proteins bythe embryo (12 of 17) suggests that they are vital for the late

image of Fig.�4

Fig. 5 – Quantitative comparisons of differentially expressed proteins and phosphoproteins during the embryo–larva transitionof honeybee worker (Apis melifera L.). The ratios of the protein abundance (larvae to embryo) are transformed, and the proteinspots with |log1.5 ratio|≥1 (p≤0.05) are selected as the differentially expressed proteins. A and B show the level of differentiallyexpressed proteins and phosphoproteins, respectively. Protein names (in abbreviations) and protein numbers (in theparenthesis) are listed as in Tables 1 and 2. Positive value indicates higher expression in larvae and negative values denotehigher expression proteins in the embryo. The |log1.5 ratio| of the uniquely expressed proteins is limited to 10.


developing embryo to enhance cellular activities of theorganogenesis, tissue elongation and body segmentation[36,54,78]. This is in agreement with the digestive system,the head and body segments are specifically seen within thelate developing embryo of honeybees [3,12]. Hence, develop-mental proteins such as vitellogenin (vg) (spot 42, 72), 14-3-3zeta (spot 47, 68, 74, 75), Tcp-1η (spot 19), singed (spot 13),TER94 (spot 1), idgf4 (spot 29) and bendless (ben) (spot 96) weremore strongly upregulated in the embryos as compared to thelarvae. For instance, Idgf4 is a cofactor of insulin or insulin-

like peptides that stimulate the proliferation, polarization andmotility of imaginal disk cells [79]. The stronger expression ofidgf4 in the late embryo suggests that it is highly involved inembryonic organogenesis as in the embryonic yolk cells andin the fat body of the embryo and larva of Drosophila [79]. Theisoforms of 14-3-3 zeta, encoding ubiquitous family of highlyconserved eukaryotic proteins that bind to phosphoserine/phosphothreoninemotifs, are involved in signal transduction,cell cycle regulation and apoptosis [80–82]. Interestingly, threeisoforms of 14-3-3 zeta family (spot 68, 74, and 75) were

image of Fig.�5

Fig. 6 – Functional enrichment analysis of the differentially expressed proteins during embryo–larva development usingClueGO software. * and ** indicate significant enrichment at the p<0.05 and p<0.01 statistical levels, respectively.

Fig. 7 – Predicted biological interaction network of the differentially expressed proteins and phosphoproteins in the last day(72 h) embryo and first instar larvae (24 h post hatching) of the honeybee worker (Apis melifera L.) using Pathway Studiosoftware. Those highlighted in blue and purple represent the key node proteins upregulated in the embryo and the larvae,respectively. Protein entities which belong to distinct functional groups were represented in different shapes according to thedefault settings of the software as described in the legend.


image of Fig.�6

image of Fig.�7

Fig. 8 – Analysis of the differentially expressed proteins and phosphoproteins of the embryo (72 h before hatching) and thelarvae (24 h post hatching), respectively with qRT-PCR. The mRNA expression is normalized with reference gene (GAPDH). Theblack and the gray bars represent the embryo and the larvae, respectively, where “a” is significantly higher than “b”. Error baris standard deviation. Abbreviated protein names indicate different proteins as in Tables 1 and 2.


phosphorylated and upregulated at both developmentalstages (Fig. 5B), suggesting they are also key regulators of celldivision, signaling and apoptosis during early stage develop-ment of honeybee embryos and larvae. The upregulation ofl(2)137cc in the larvae is most likely involve in the develop-ment of larval hypopharyngeal gland [65,83].

The functions of proteins that act as molecular chaper-ones, mainly assist in the folding of proteins to accomplishtheir functional shape or conformation through polypeptidefolding [84]. Hsps and the subcomponents of chaperonescontaining t-complex protein (TCP-1) are ubiquitous moleculesamong eukaryotes [85]. Hsps function primarily as molecularchaperones, facilitating the protein folding, preventing proteinaggregation, or targeting improperly folded proteins in specificdegradative pathways [86]. In the honeybees, Hsps have widelybeen identified as molecular chaperone in the embryos ofhoneybee workers and drones [35,36], worker and queen larvae

[43], hemolymph [40], and venom gland [87]. Because ourexperiment was also carried out under normal physiologicalcircumstances, their expression is assumed to serve the embryosand the larvae asmolecular chaperones aiding organogenesis byfolding newly synthesized proteins [88,89]. TCP-1 is known todirectly fold cytoskeletal proteins, such asα,β, and γ-tubulin andactin [90]. Its upregulation in the embryo might be to sustainuninterrupted cellular divisions during the late phase of em-bryogenesis. Dynactin subunit 2 (p50), a multi protein complexassociated with dynein, has a function in pro-metaphasechromosome alignment and spindle organization during mito-sis and plays a role in synapse formation during braindevelopment [91]. The exclusive expression of p50 in the larvais believed to support brain development to equip futurememory and its learning ability [92].

Cytoskeletal proteins aremainly involved inmaintaining thestability of the cell's shape and structure and play important

image of Fig.�8


roles in intracellular transport and cellular division [62]. Theyperform imperative nuclear functions in the honeybee queenintended developing larvae during cast differentiation [93].Hence, the higher number of these proteins phosphorylatedand escalated their expression in the newly hatched larvae66.7% (10 of 15) than in the embryo supports the action that thelarva maintains its normal C-shape after hatching, motility,intracellular transportation and cellular division [12,94]. Forinstance, the higher expression of α-tubulin subunit phosphor-ylation suggests a role for phosphorylation in remodeling of thecytoskeletal proteins [57,95] as the remodeling of newlysynthesized tissues and organs have been triggered by proteinphosphorylation [46]. The exclusive expression of β-Tubulin at56D (spot 48) and sec 13 (spot 71) in the larvae is likely attributedto the structural constituent of cytoskeletons involved in thebiological processes of the developing larvae.

Antioxidant systems are to prevent cellular componentsfrom oxidative damages by removing free radicals subsequentlyinhibiting other oxidative reactions [96–98]. The expression ofantioxidant proteins is generally available to protect cellularcomponents fromoxidative damages during the development ofhoneybee embryo [36,99], hypopharyngeal gland [100], larvalcast determination [43], and sperm storage facilitation [101].Likewise, the high demand for oxygen during embryonic andearly larval development leads to increased reactive oxygenspecies (ROS) production through increasing oxidative damage[102]. Consequently, both the embryo and the larvae upregulatedspecific antioxidant proteins are supposed to be in response tothese ROS oxidative damages. Notably, the upregulation ofcathD (spot 51), trxr-1(spot 69) and trxr-2 (spot 98) by the larvaesignifies their role in the protection of ROS-mediated organdamage to ensure their fast development. On the other hand, theembryoupregulation of thioredoxin peroxidase (jafrac1) (spot 84,86) and isoform of trxr-1(spot 31, 32) is suggested to rescue thecellular components from oxidative damage during extrememetabolic rate of embryogenesis that calls for high oxygendemand.

Protein biosynthesis involves a complex translational pro-cess carried out in ribosomes, where activated amino acids areadded to the nascent polypeptide chain [63]. In this study, of the11 protein biosynthesis spots expressed at both developmentalstages, five were phosphorylated in line with synthesis andtranslation of proteins that are largely mediated by reversiblephosphorylation [59]. Specifically, eif-5a (spot 90), eif-4a (spot28), cacybp (spot 80), crb (spot 82), and Tctp (spot 85) werephosphorylated and all except cacybp were upregulated in thelarvae. The phosphorylation of eif-4a in particular suggests itsvital roles in the 20E signal transduction pathway, preparing forthe early larval metamorphosis as it is induced by 20E andinhibited by the protein kinase C specific inhibitor inDrosophilasalivary glands [52]. Similarly, eif-5a is most likely involved inthe synthesis and translation of larval specific proteins as it isthe first step of peptide bond formation in translation andessential for cell proliferation and cell-cycle regulation [103]. Ingeneral, the over-expression of biosynthesis proteins duringembryo–larva transition is closely linked to the high rate ofbiogenesis to cope with programmed cell deaths and the highstructural constituent of fast developing larvae [56].

Fatty acid transport proteins are key regulators in theuptake and enzymatic activation of fatty acids to generate and

transport energy [69]. The expression of these proteins at bothdevelopmental stages implies the critical necessity of energytransporting proteins for effective development of new organsand tissues to take place during the early stage of develop-ment. The upregulation of acyl-CoA dehydrogenase (spot 46),a mitochondrial fatty acid oxidation protein [104], in thelarvae and MCAD, a mitochondria cells fatty acid β-oxidationcatalyzing protein [105], in the embryo (>10 fold change)suggests the oxidation of fatty acids as in the adult honeybeemandibular gland [105], queen intended developing larvae [43]and drone embryos at 72 h of age [54].

Proteasome subunit alpha type3 (prosα3T, spot 76) and alphatype5 (prosα5, spot 78) and rpt1 are members of the 20Sproteasome, which are related to the metabolism of aminoacids and nucleotides. The 20S proteasome, ATP-independentribosomal nucleotide, has several distinct catalytic activitiesinvolved in the ubiquitin proteasome system [106,107].Predominantly, the import of proteasome subunits into thenucleus is controlled by phosphorylation [108]. The upregulationof phosphorylated prosα3T and prosα5 in the embryo suggeststhe participation of proteasomes in tissue remodeling [106]during active embryonic development. This is in line withphosphorylation of proteasomes during the larval molting ofcotton bollworm (Helicoverpa armigera) [109].

Most proteins exist as integral parts of protein complexes towork together rather than a single entity in a living cell. Thepredicted possible key node proteins in the BIN indicate theircentrality in the regulatory pathway of embryo–larva transition.Of these 56 key node proteins and induced phosphoproteinsnetworked in the BIN (Fig. 7), proteins related to carbohydratemetabolism and energy production, development, proteinbiosynthesis, cytoskeleton and protein folding are the centralrole players, which were further confirmed by functionalenrichment analysis (Fig. 6). The results provide a systematicperspective on the protein and the biochemical pathwaysinvolved in the honeybee embryo–larvae transition.

The tested eight proteins exhibited a similar expressionpattern at the transcript level (Fig. 8), indicating a prospectiveopportunity for reverse genetic research through gene manip-ulation at early developmental stages of honeybee embryo orlarva. Therefore, understanding the relationship between pro-teins and their encoding genes will potentially facilitate theability to generate desirable phenotypes and genetically pro-ductive honeybees.

5. Conclusions

In conclusion, proteomic analysis of the last day embryos andthe first day larvae has yielded significant insight into themolecular mechanisms behind the embryo–larva transition.Ninety-nine differentially expressed and phosphorylated pro-teinswere identified during the life change of last day embryo tonewly hatched larvae. Proteins and phosphoproteins related toenergymetabolism, development, protein folding, cytoskeletonand protein biosynthesis are major modulators involved in theembryo–larva transitional process and triggered by changes ofprotein expression levels and phosphorylation status. Thetested results of qRT-PCR between expression of proteins andtheir encoding genes from some key node proteins in the BIN


provide us important target proteins or genes to be selected forfurther functional analysis. Thus future studies will be requiredto investigate the possible role of these proteins in response tothe honeybee early development through RNA interference orby inhibiting the proteins' functions by pharmacologicalapproaches (e.g. Hsp, proteasome and cytoskeletal inhibitors).This first proteomic study deciphers themolecular mechanismduring the fascinating transition of embryo to larva and hassignificantly gained new insight into the developmental biologyof honeybees.

Acknowledgment

We thank Ms Katrina Klett from the University of Minnesota,USA, for her helpwith the language of themanuscript. Thisworkis supported by the earmarked fund for Modern Agro-industryTechnology Research System (CARS-45), the National NaturalScience Foundation of China (No. 30972148) and key projects ofthe national scientific supporting plan of the 12th Five-YearDevelopment (2011–2015) (2011BAD33B04).

Appendix A. Supplementary data

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.jprot.2012.10.012.

R E F E R E N C E S

[1] Fleig R, Sander K. Embryogenesis of the honeybee Apismellifera L. (Hymenoptera: Apidae): an SEM study. Int J InsectMorphol Embryol 1986;15:449–62.

[2] Pipan N, Rakovec V. Cell death in the midgut epithelium ofthe worker honey bee (Apis mellifem carnica) duringmetamorphosis. Zoomorphology 1980;94:217–24.

[3] DuPraw E. The honeybee embryo. Methods Dev Biol 1967:183–217.

[4] Yu R, Hagen A, Omholt SW. Biopsied preblastodermhoneybee embryos develop into normal honeybee queens.Apidologie 1998;29:547–54.

[5] Yu R, Omholt SW. Early developmental processes in thefertilised honeybee (Apis mellifera) oocyte. J Insect Physiol1999;45:763–7.

[6] Woyke J. Size change of Apis mellifera eggs during theincubation period. J Apic Res 1998;37.

[7] Collins AM. Variation in time of egg hatch by the honey bee,Apis mellifera (Hymenoptera: Apidae). Ann Entomol Soc Am2004;97:140–6.

[8] Evans JD, Boncristiani H, Chen Y. Scientific note on masscollection and hatching of honey bee embryos. Apidologie2010;41:654–6.

[9] DuPraw E. A unique hatching process in the honeybee.Trans Am Microsc Soc 1961;80:185–91.

[10] Li Y, Zhang Z, Robinson GE, Palli SR. Identification andcharacterization of a juvenile hormone responseelement and its binding proteins. J Biol Chem 2007;282:37605–17.

[11] Rachinsky A, Strambi C, Strambi A, Hartfelder K. Caste andmetamorphosis: hemolymph titers of juvenile hormone andecdysteroids in last instar honeybee larvae. Gen CompEndocrinol 1990;79:31–8.

[12] Winston ML. The biology of the honey bee. Harvard Univ Pr;1991.

[13] Jean-Prost P, Médori P. Apiculture: know the bee, managethe apiary. Intercept Limited; 1994.

[14] Wray GA. Punctuated evolution of embryos. Science1995;267:1115–6.

[15] Bergem M, Norberg K, Aamodt RM. Long-term maintenanceof in vitro cultured honeybee (Apis mellifera) embryonic cells.BMC Dev Biol 2006;6:17.

[16] Peng YSC, Mussen E, Fong A, Montague MA, Tyler T. Effectsof chlortetracycline of honey bee worker larvae reared invitro. J Invertebr Pathol 1992;60:127–33.

[17] Czoppelt C, Rembold H. Effect of parathion on honey beelarvae rearedin vitro. Anz Schädlingskd 1988;61:95–100.

[18] Kaneko K, Hori S, Morimoto MM, Nakaoka T, Paul RK,Fujiyuki T, et al. In situ hybridization analysis of theexpression of futsch, tau, and MESK2 homologues in thebrain of the European honeybee (Apis mellifera L.). PLoS One2010;5:e9213.

[19] Osborne P, Dearden PK. Non-radioactive in-situhybridisation to honeybee embryos and ovaries. Apidologie2005;36:113–8.

[20] Collins AM. Collection of honey bee eggs for cryopreservation.J Apic Res 2002;41:89–95.

[21] Ronglin Y, Hagen A, Omholt SW. Cryopreservation oftotipotent nuclei from honeybee (Apis mellifera) embryos byrapid freezing. Cryobiology 1997;35:41–5.

[22] WilsonMJ, Havler M, Dearden PK. Giant, Kruppel, and caudalact as gap genes with extensive roles in patterning thehoneybee embryo. Dev Biol 2010;339:200–11.

[23] Dearden PK. Germ cell development in the honeybee(Apis mellifera); vasa and nanos expression. BMC Dev Biol2006;6:6.

[24] ANITA MC. Sources of variation in the viability of honey bee,Apis mellifera L., semen collected for artificial insemination.Invertebr Reprod Dev 2004;45:231–7.

[25] Dearden PK, Wilson MJ, Sablan L, Osborne PW, Havler M,McNaughton E, et al. Patterns of conservation and change inhoney bee developmental genes. Genome Res 2006;16:1376–84.

[26] Chan QW,Melathopoulos AP, Pernal SF, Foster LJ. The innateimmune and systemic response in honey bees to a bacterialpathogen, Paenibacillus larvae. BMC Genomics 2009;10:387.

[27] Evans JD. Transcriptional immune responses by honey beelarvae during invasion by the bacterial pathogen,Paenibacillus larvae. J Invertebr Pathol 2004;85:105–11.

[28] Genersch E. American foulbrood in honeybees and itscausative agent, Paenibacillus larvae. J Invertebr Pathol2010;103:S10–9.

[29] Wegener J, Al-Kahtani S, Bienefeld K. Collection of viablehoney bee (Apis mellifera) larvae after hatching in vitro. J ApicRes 2009;48:115–20.

[30] Kamakura M. Royalactin induces queen differentiation inhoneybees. Nature 2011;473:478–83.

[31] Evans JD, Wheeler DE. Expression profiles during honeybeecaste determination. Genome Biol 2000;2research0001.1–research0001.6.

[32] Wheeler DE. Developmental and physiological determinantsof caste in social Hymenoptera: evolutionary implications.Am Nat 1986;128:13–34.

[33] Wirtz P, Beetsma J. Induction of caste differentiation in thehoneybee (Apis mellifera) by juvenile hormone. Entomol ExpAppl 1972;15:517–20.

[34] Li J, Fang Y, Zhang L, Begna D. Honeybee (Apis melliferaligustica) drone embryo proteomes. J Insect Physiol 2011;57:372–84.

[35] Fang Y, Li J. Analysis of developmental proteome at eggstage of drone honeybees (A. m. ligustica). Sci Agric Sin2008;41:3793–800.



[36] Li J, Zhang L, Feng M, Zhang Z, Pan Y. Identification of theproteome composition occurring during the course ofembryonic development of bees (Apis mellifera). Insect MolBiol 2009;18:1–9.

[37] Zhang L, Li J, Wu L. Profile analysis of the proteome of theeggs of the higher royal jelly producing bees (Apis melliferaL.). Sci Agric Sin 2007;40:1276–87.

[38] Li J, Wu J, Rundassa DB, Song F, Zheng A, Fang Y. Differentialprotein expression in honeybee (Apis mellifera L.) larvae:underlying caste differentiation. PLoS One 2010;5:e13455.

[39] Chan QW, Foster LJ. Changes in protein expression duringhoney bee larval development. Genome Biol 2008;9:R156.

[40] Chan QW, Howes CG, Foster LJ. Quantitative comparison ofcaste differences in honeybee hemolymph.Mol Cell Proteomics2006;5:2252–62.

[41] Barchuk AR, Cristino AS, Kucharski R, Costa LF, Simoes ZL,Maleszka R. Molecular determinants of caste differentiationin the highly eusocial honeybee, Apis mellifera. BMC Dev Biol2007;7:70.

[42] Begna D, Han B, Feng M, Fang Y, Li J. Differential expressionsof nuclear proteomes between honeybee (Apis mellifera L.)queen and worker larvae: a deep insight into caste pathwaydecisions. J Proteome Res 2012;11(2):1317–29.

[43] Begna D, Fang Y, Feng M, Li J. Mitochondrial proteinsdifferential expression during honeybee (Apis mellifera L.)queen and worker larvae caste determination. J ProteomeRes 2011;10:4263–80.

[44] Delom F, Chevet E. Phosphoprotein analysis: from proteinsto proteomes. Proteome Sci 2006;4:15.

[45] Mann M, Ong SE, Grønborg M, Steen H, Jensen ON, Pandey A.Analysis of protein phosphorylation usingmass spectrometry:deciphering the phosphoproteome. Trends Biotechnol 2002;20:261–8.

[46] Rewitz KF, Larsen MR, Lobner-Olesen A, Rybczynski R,O'Connor MB, Gilbert LI. A phosphoproteomics approach toelucidate neuropeptide signal transduction controllinginsect metamorphosis. Insect Biochem Mol Biol 2009;39:475–83.

[47] Mok FSY, Thiyagarajan V, Qian PY. Proteomic analysisduring larval development and metamorphosis of thespionid polychaete Pseudopolydora vexillosa. Proteome Sci2009;7:44.

[48] Chandramouli KH, Mok FSY, Wang H, Qian PY.Phosphoproteome analysis during larval development andmetamorphosis in the spionid polychaete Pseudopolydoravexillosa. BMC Dev Biol 2011;11:31.

[49] Zhang Y, Sun J, Xiao K, Arellano SM, Thiyagarajan V, QianPY. 2D gel-based multiplexed proteomic analysis duringlarval development and metamorphosis of the biofoulingpolychaete tubeworm Hydroides elegans. J Proteome Res2012;9(9):4851–60.

[50] Qian PY, Wong YH, Zhang Y. Changes in the proteome andphosphoproteome expression in the bryozoan Bugulaneritina larvae in response to the antifouling agentbutenolide. Proteomics 2010;10:3435–46.

[51] Fu Q, Liu PC, Wang JX, Song QS, Zhao XF. Proteomicidentification of differentially expressed and phosphorylatedproteins in epidermis involved in larval–pupalmetamorphosisof Helicoverpa armigera. BMC Genomics 2009;10:600.

[52] Sun X, Song Q. PKC-mediated USP phosphorylation isrequired for 20E-induced gene expression in the salivaryglands of Drosophila melanogaster. Arch Insect BiochemPhysiol 2006;62:116–27.

[53] Bradford MM. A rapid and sensitive method for thequantitation of microgram quantities of protein utilizing theprinciple of protein-dye binding. Anal Biochem 1976;72:248–54.

[54] Fang Y, Song F, Zhang L, Aleku DW, Han B, Feng M, et al.Differential antennal proteome comparison of adult

honeybee drone, worker and queen (Apis mellifera L.).J Proteomics 2012;75:756–73.

[55] Bindea G, Mlecnik B, Hackl H, Charoentong P, Tosolini M,Kirilovsky A, et al. ClueGO: a Cytoscape plug-in to decipherfunctionally grouped gene ontology and pathway annotationnetworks. Bioinformatics 2009;25:1091–3.

[56] Gregorc A, Bowen ID. Histochemical characterization of celldeath in honeybee larvae midgut after treatment withPaenibacillus larvae, amitraz and oxytetracycline. Cell Biol Int2000;24:319–24.

[57] Roux MM, Radeke MJ, Goel M, Mushegian A, Foltz KR. 2DEidentification of proteins exhibiting turnover andphosphorylation dynamics during sea urchin egg activation.Dev Biol 2008;313:630–47.

[58] Yurttas P, Morency E, Coonrod SA. Use of proteomics toidentify highly abundant maternal factors that drive theegg-to-embryo transition. Reproduction 2010;139:809–23.

[59] Hunter T. Protein kinases and phosphatases: the yin and yangof protein phosphorylation and signaling. Cell 1995;80:225–36.

[60] Pan C. Phosphoproteomics and proteomic phenotyping toassess signal transduction in cancer cells; 2008.

[61] Raggiaschi R, Gotta S, Terstappen GC. Phosphoproteomeanalysis. Biosci Rep 2005;25:33–44.

[62] Wulfkuhle JD, Petersen NS, Otto JJ. Changes in the F-actincytoskeleton during neurosensory bristle development inDrosophila: the role of singed and forked proteins. Cell MotilCytoskeleton 1998;40:119–32.

[63] Kim HD, Lee JY, Kim J. Erk phosphorylates threonine 42residue of ribosomal protein S3. Biochem Biophys ResCommun 2005;333:110–5.

[64] Haydak MH. Honey bee nutrition. Annu Rev Entomol1970;15:143–56.

[65] Feng M, Fang Y, Li J. Proteomic analysis of honeybee worker(Apis mellifera) hypopharyngeal gland development. BMCGenomics 2009;10:645.

[66] Schonleben S, Sickmann A, Mueller MJ, Reinders J. Proteomeanalysis of Apis mellifera royal jelly. Anal Bioanal Chem2007;389:1087–93.

[67] Li J, Christensen B. Biological function of insect yellow genefamily. In: Liu T, Kang L, editors. Recent advances inentomological research. Springer Berlin Heidelberg; 2011.p. 121–31.

[68] Herold N, Will CL, Wolf E, Kastner B, Urlaub H, Lührmann R.Conservation of the protein composition and electronmicroscopy structure of Drosophila melanogaster and humanspliceosomal complexes. Mol Cell Biol 2009;29:281–301.

[69] Kucharski R, Maleszka R. Arginine kinase is highlyexpressed in the compound eye of the honey-bee, (Apismellifera). Gene 1998;211:343–9.

[70] Yu-mei EW, Esbensen P, Bentley D. Arginine kinaseexpression and localization in growth cone migration.J Neurosci 1998;18:987–98.

[71] Chamberlin M. Mitochondrial arginine kinase in the midgutof the tobacco hornworm (Manduca sexta). J Exp Biol1997;200:2789–96.

[72] Chen J, Li J. Comparative analysis of proteome complementbetween worker bee larvae of high royal jelly producing bees(A. m. lingustica) and Carniolians (A. m. carnica) bees. Sci AgricSin 2008;41:3292–9.

[73] White KP, Rifkin SA, Hurban P, Hogness DS. Microarrayanalysis of Drosophila development during metamorphosis.Science 1999;286:2179–84.

[74] Sardiello M, Licciulli F, Catalano D, Attimonelli M, Caggese C.MitoDrome: a database of Drosophila melanogaster nucleargenes encoding proteins targeted to the mitochondrion.Nucleic Acids Res 2003;31:322–4.

[75] Galloni M, Edgar BA. Cell-autonomous andnon-autonomous growth-defective mutants of Drosophilamelanogaster. Development 1999;126:2365–75.


[76] Heyland A, Moroz LL. Signaling mechanisms underlyingmetamorphic transitions in animals. Integr Comp Biol2006;46:743–59.

[77] Curtiss J, Heilig JS. Establishment of Drosophila imaginalprecursor cells is controlled by the arrowhead gene.Development 1995;121:3819–28.

[78] Keller R. Mechanisms of elongation in embryogenesis.Development 2006;133:2291–302.

[79] Kawamura K, Shibata T, Saget O, Peel D, Bryant PJ. A newfamily of growth factors produced by the fat body and activeon Drosophila imaginal disc cells. Development 1999;126:211–9.

[80] Darling DL, Yingling J, Wynshaw-Boris A. Role of 14-3-3proteins in eukaryotic signaling and development. Curr TopDev Biol 2005;68:281–315.

[81] Muslin AJ, Tanner JW, Allen PM, Shaw AS. Interaction of14-3-3 with signaling proteins is mediated by therecognition of phosphoserine. Cell 1996;84:889–97.

[82] Aitken A, Collinge D, Heusden BPH, Isobe T, Roseboom P,Rosenfeld G, et al. 14-3-3 proteins. a highly conserved,widespread family of eukaryotic proteins. Trends BiochemSci 1992;17:498–501.

[83] Li J, Li H, Zhang Z, Pan Y. Identification of the proteomecomplement of high royal jelly producing bees(Apis mellifera) during worker larval development. Apidologie2007;38:545–57.

[84] Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P.The shape and structure of proteins. Mol Biol Cell 2002;4.

[85] Nomura M, Nakajima A, Inaba K. Proteomic profiles ofembryonic development in the ascidian Ciona intestinalis.Dev Biol 2009;325:468–81.

[86] Ma H, Cherng S, Yang Y. Expression of heat shock proteins25, 60, 70 and 90 in rat myocardium followingtransmyocardial laser revascularization. Researcher 2009;1:86–9.

[87] Peiren N, de Graaf DC, Vanrobaeys F, Danneels EL, DevreeseB, Van Beeumen J, et al. Proteomic analysis of the honey beeworker venom gland focusing on the mechanisms ofprotection against tissue damage. Toxicon 2008;52:72–83.

[88] Borges JC, Ramos CHI. Protein folding assisted bychaperones. Protein Pept Lett 2005;12:257–61.

[89] Wittung-Stafshede P. Protein folding inside the cell. BiophysJ 2011;101:265.

[90] Liang P, MacRae T. Molecular chaperones and thecytoskeleton. J Cell Sci 1997;110:1431–40.

[91] Goshima G, Wollman R, Goodwin SS, Zhang N, Scholey JM,Vale RD, et al. Genes required for mitotic spindle assemblyin Drosophila S2 cells. Science 2007;316:417–21.

[92] Garcia L, Saraiva Garcia CH, Calábria LK, Costa Nunes daCruz G, Sánchez Puentes A, Báo SN, et al. Proteomic analysisof honey bee brain upon ontogenetic and behavioraldevelopment. J Proteome Res 2009;8:1464–73.

[93] Begna D, Han B, Feng M, Fang Y, Li J. Differential expressionsof nuclear proteomes between honeybee (Apis mellifera L.)queen and worker larvae: a deep insight into caste pathwaydecisions. J Proteome Res 2012;11:1317–29.

[94] Waller GD. Honey bee life history. Beekeeping in the UnitedStates, 335; 1980. p. 24.

[95] Luduena RF. Multiple forms of tubulin: different geneproducts and covalent modifications. Int Rev Cytol 1998;178:207–75.

[96] Temneanu O, Zamfir C, Eloaie Zugun F, Cojocaru E, Tocan L.Oxidants and antioxidants relevance in rats' pulmonaryinduced oxidative stress. J Med Life 2011;4:244–9.

[97] Sies H. Oxidative stress: oxidants and antioxidants. ExpPhysiol 1997;82:291–5.

[98] Vertuani S, Angusti A, Manfredini S. The antioxidants andpro-antioxidants network: an overview. Curr Pharm Des2004;10:1677–94.

[99] Zheng A, Li J, Begna D, Fang Y, Feng M, Song F. Proteomicanalysis of honeybee (Apis mellifera L.) pupae headdevelopment. PLoS One 2011;6:e20428.

[100] Yu F, Mao F, Jianke L. Royal jelly proteome comparisonbetween A. mellifera ligustica and A. cerana cerana. J ProteomeRes 2010;9:2207–15.

[101] Corona M, Robinson GE. Genes of the antioxidant system ofthe honey bee: annotation and phylogeny. Insect Mol Biol2006;15:687–701.

[102] Seehuus SC, Norberg K, Gimsa U, Krekling T, Amdam GV.Reproductive protein protects functionally sterile honey beeworkers from oxidative stress. Proc Natl Acad Sci U S A2006;103:962–7.

[103] Yao M, Ohsawa A, Kikukawa S, Tanaka I, Kimura M. Crystalstructure of hyperthermophilic archaeal initiation factor 5A:a homologue of eukaryotic initiation factor 5A (eIF-5A). JBiochem 2003;133:75–81.

[104] Henze K, Martin W. Evolutionary biology: essence ofmitochondria. Nature 2003;426:127–8.

[105] Hasegawa M, Asanuma S, Fujiyuki T, Kiya T, Sasaki T, EndoD, et al. Differential gene expression in the mandibularglands of queen and worker honeybees, Apis mellifera L.:implications for caste-selective aldehyde and fatty acidmetabolism. Insect Biochem Mol Biol 2009;39:661–7.

[106] Mykles DL. Structure and functions of arthropodproteasomes. Mol Biol Rep 1999;26:103–11.

[107] Ciechanover A. The ubiquitin-proteasome pathway: onprotein death and cell life. EMBO J 1998;17:7151–60.

[108] Mykles DL. Intracellular proteinases of invertebrates:calcium-dependent and proteasome/ubiquitin-dependentsystems. Int Rev Cytol 1998;184:157–289.

[109] Dong DJ, He HJ, Chai LQ, Jiang XJ, Wang JX, Zhao XF.Identification of genes differentially expressed during larvalmolting and metamorphosis of Helicoverpa armigera. BMCDev Biol 2007;7:73.

Apis mellifera ligusticaapisomics.com/files/201712/f9a06c4f-da34-438e-88e9... · the honeybee life cycle. However, the molecular mechanism of this life transition is still unknown.

Documents