novel aspects of understanding molecular working ... · the salivary system is exocrine glands that...

Author's personal copy

Novel aspects of understanding molecular workingmechanisms of salivary glands of worker honeybees(Apis mellifera) investigated by proteomicsand phosphoproteomics☆

Mao Feng, Yu Fang, Bin Han, Lan Zhang, Xiaoshan Lu, 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 A B S T R A C T

Article history:Received 25 March 2013Accepted 13 May 2013Available online 27 May 2013

Honeybee salivary glands (SGs) are important exocrine glands. However, the molecular basisof how SGs fulfill their biological duty is still elusive. Proteomics and phosphoproteomics ofcephalic SG (HSG) and thoracic SG (TSG) were compared between normal and single-cohorthoneybee colonies. Of 113 and 64 differentially regulated proteins and phosphoproteins, 86and 33 were identified, respectively. The SGs require a wide spectrum of proteins to supporttheir multifaceted functions and ensure normal socialmanagement of the colony. Changes ofprotein expression and phosphoproteins are key role players. The HSG triggers labor transitionfrom in-hive work to foraging activities via the regulation of juvenile hormone and ethyl oleatelevels. The stronger expression of proteins involved in carbohydrate and energy metabolism,protein folding, proteinmetabolism, cellular homeostasis and cytoskeleton in TSG, supports thegland to efficiently enhance honey processing by synthesis and secretion of saliva into nectar.The age structure of the colony is vital for increased production efficiency. This data reveals themolecular underpinning of SGs to accomplish their biological missions and provides newknowledge for the beekeeping industry for enhancing the management and productionefficiency of the colony and honey quality through manipulation of potential target proteins.

Biological significanceThis study comprehensively analyses the characteristic of the proteomeandphosphoproteomeof honeybee salivary glands (SGs) between normal and single-cohort honeybee colonies. TheSGs need a wide spectrum of proteins to support their multifaceted functions and ensurenormal socialmanagement of the colony. The cephalic SG triggers labor transition from in-hivework to foraging behavior via the regulation of juvenile hormone and ethyl oleate titer. Thethoracic SG stronger expressed of proteins related to carbohydrate and energy metabolism,protein folding, proteinmetabolism, cellular homeostasis and cytoskeleton to support the glandto efficiently enhance honey processing by synthesis and secretion of saliva into nectar. Thisdata reveals the molecular underpinning of SGs to accomplish their biological missions and

Keywords:HoneybeeSalivary gland2-DEProteomePhosphoproteomeSingle-cohort colony

J O U R N A L O F P R O T E O M I C S 8 7 ( 2 0 1 3 ) 1 – 1 5

☆ This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-No DerivativeWorksLicense, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source arecredited.⁎ Corresponding author at: Institute of Apicultural Research/Key Laboratory of Pollinating Insect Biology, Ministry of Agriculture, Chinese

Academy of Agricultural Science, Beijing 100093, China. Tel./fax: +86 10 6259 1449.E-mail address: [email protected] (J. Li).

1874-3919/$ – see front matter © 2013 The Authors. Published by Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.jprot.2013.05.021

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


provides new knowledge for the beekeeping industry for enhancing the maintenance andproduction efficiency of the colony and honey quality throughmanipulation of potential targetproteins.

© 2013 The Authors. Published by Elsevier B.V. All rights reserved.

1. Introduction

Age-dependent division of labor is a striking pattern ofbehavioral plasticity in honeybee colonies [1]. Honeybeeworkers begin their work within the hive by spending thefirst few weeks in activities including queen tending, broodrearing, and hive maintaining, and then later foraging forpollen, nectar, and propolis outside the hive [2]. The transitionfrom in-hive work to foraging activities is a major change inhoneybee lifestyle, which is influenced and modulated byseveral factors such as changes in endocrine and exocrinegland secretions [3], brain structure [4] and gene expression [5].The salivary system is exocrine glands that show a functionalflexibility with age of honeybee workers [6].

The honeybee salivary glands (SGs), also known as labialglands, are composed of two parts, the cephalic SG (HSG) in thehead and the thoracic SG (TSG) in the thorax, both of whichconnect through a common duct to themouth [7]. The HSG andTSG begin to develop during the pupal period and progressivelyincrease from the newly emerged workers to foragers, reachingtheir maximal size when the workers begin to forage [7].

The larval SGs are the reservoir for the fatty acid esters thatconstitute brood pheromones, which modulate the behavioraldevelopment of young bees and stimulate the workers forpollen foraging [8]. The secretions of the adult females arefunctionally different from the larvae. The adult HSG producesan oily substance involved in wax manipulation, mouthpartlubrication [9] and releases of scent trails to mark the foodsource location [10]. The TSG produces an aqueous secretionmainly containing digestive enzymes that are involved inhoney and sugar digestion as well as pollen and wax moisten-ing [9]. However, the volatile secretions produced by workerhoneybees are hydrocarbons which are important chemicalsignals for the worker bees' nestmate discrimination, colonyidentification, regulation of the age when honeybee workersbegin to forage [7] and for adult behavior maturation [11].

Single-cohort colonies are artificially established coloniesthat are composed entirely of the same aged young bees,without appropriate aged nurses and foragers. It is an idealsystem to investigate the functional flexibility of SGs of theworker honeybees by comparison with the same agedworkersof their parent colonies (normal ones). The secretion amountof the young nurse bees in single-cohort colonies does notdiffer from the old nurse bees in normal colonies [7]. Thistask-related rather than age-dependent mechanism is owedto the younger precocious forgers in single-cohort coloniesstriving to work earlier in order to overcome the imbalancedage distribution [7]. In the normal colony, the HSGs of nursebees secrete an array of enzymes involved in carbohydratemetabolism that mainly function as a carbohydrate preserva-tive [6]. In addition, the salivary system can synthesize growthfactors such as imaginal disk growth factor 4 (Idgf4), which isthen secreted into royal jelly and honey to further influencethe growth and physiology of the other colony members [6].

To date, proteomics has become an efficient interdisciplin-ary platform in the studies of honeybee biology, such as droneembryogenesis [12], larval and hypopharyngeal gland develop-ment [13,14], and so on. However, only a few limited studieshave so far been undertaken to compare the differencesbetween theHSG andTSGofworker bees in the normal colonies[6] including targeted non-proteomic analysis of single or fewproteins [15]. Given the importance of the salivary system of theworker bees in the colony organization, comparative analysis ofthe HSG and TSG between normal and single-cohort coloniesusing proteomic and phosphoproteomic approaches will gainin-depth understanding of the molecular basis and how theglands achieve their functionality. This may potentially beuseful for the improvement of the glandular activities of theworker bees at molecular level, thus promoting honeybeecolony maintenance, production efficiency and honey qualityfor the beekeeping industry.

2. Materials and methods

2.1. Biological sample

Honeybees (Apismellifera L.)were kept in the experimental apiaryof Institute of Apicultural Research, Chinese Academy ofAgricultural Sciences in Beijing. Five colonieswithmatedqueensand sufficient broods were selected as the normal (control)colonies. Five other single-cohort colonies were establishedartificially with ~2000 one-day-old honeybee workers eachfrom the same normal parent colonies, one mated sister queenof the normal colony, one comb ~one-quarter filled with nectarand one-eighth filled with pollen and one empty frame for thequeen to lay eggs [16]. One-day-old beeswere obtained by takingframes containing old pupae and placing them into an incubator(34 °C and 80% relative humidity) for their eclosion.

The nursing and foraging bees were collected from thenormal and single-cohort colonies according to standardmethodsof identification:workerswithheads in cells containingsmall larvae were identified as nurses, and bees flying into thehive with pollen loads or abdomens distended with nectar wererecognized as foragers [17]. More than 50 honeybee nurses andforagers were captured from each control colony, then the TSGsof nurse bees (NNT) and foragers (NFT) and HSGs of nurse bees(NNH) and foragers (NFH) were dissected and each kind ofsample was pooled together, respectively. The TSGs of nursebees (SNT) and foragers (SFT) and HSGs of nurse bees (SNH) andforagers (SFH) from the single-cohort colonies were sampled onthe same way. Three independent batches were collected andstored at −80 °C until further analysis.

2.2. Protein preparation and 2-DE

Protein extraction was carried out according to a previouslydescribed method [18]. A 400 μg protein sample was dissolved

2 J O U R N A L O F P R O T E O M I C S 8 7 ( 2 0 1 3 ) 1 – 1 5


in 90 μL lysis buffer (8 M urea, 2 M thiourea, 4% CHAPS, 20 mMTris-base, 30 mM DTT, and 2% Biolyte pH 3–10) and mixed with360 μL of rehydration buffer (8 M urea, 2% CHAPS, 0.001%bromophenol blue, 45 mM DTT, 0.2% Bio-lyte pH 3–10). Themixture was loaded onto a 17 cm IPG strips (pH 3–10, linear,Bio-Rad). IEF (Protean IEF Cell, Bio-Rad) and the second dimen-sion electrophoresis, SDS-PAGE (Protean II Xi Cell, Bio-Rad), wereperformed according to documented method [18].

2.3. Fluorescent and CBB gel staining and image analysis

After 2-DE, the gels were fixed, followed by staining withPro-Q Diamond (Invitrogen, Eugene, OR) to detect phospho-proteins. The total proteins were visualized by CBB G-250 [18].

The best three gels of each sample were selected forsubsequent gel analysis using Progenesis SameSpots (v 4.1,Nonlinear Dynamic, UK). The gels matching, alignment, spotsaveraging and normalization as well as the protein expressionlevel determination of a given spot and multi-test were doneaccording to Han et al. [19]. The expression of protein spotswas considered to be statistically significant when they hadp < 0.05 and at least 1.5-fold changes.

2.4. Tryptic digestion and protein identification by MS

The differentially expressed protein spots were manuallyexcised from the CBB-G 250 stained gels and protein digestion

and peptides extraction were done according to our previous-ly established protocol [18].

The digested peptides were then analyzed by high perfor-mance LC-Chip/ESI-QTOF-MS (Q-TOF 6520, Agilent Technolo-gies). The LC-Chip (Agilent Technologies) consisted of a Zorbax300SB-C18 enrichment column (40 nL, 5 μm) and a Zorbax300SB-C18analytical column (75 μm × 43 mm, 5 μm).An8 μL ofdigested peptides was injected into MS. The following chroma-tography and MS analysis were carried out as in our alreadyestablished protocol [20].

Tandem mass spectra were retrieved using the MassHuntersoftware (version B. 04, Agilent Technologies). Before MS/MSdata searching, a peak-list was generated by Mascot Distiller(version 2.3.2.0, Matrix Science, UK). The data were storedin a combined mgf file and searched against a sequencedatabase totaling to 72,672 entries and generated fromprotein sequences of Apis (downloaded May, 2011) augment-ed with sequences from Drosophila melanogaster (downloadedMay, 2011), Sacharomyces cerevisiae (downloaded May, 2011) andthe common repository of adventitious proteins (cRAP, fromThe Global Proteome Machine Organization, downloadedMay, 2011) using in-house Mascot (version 2.3.2, MatrixScience, UK.). The searching parameters were as follows:fixedmodification: carbamidomethyl (C); variablemodifications:oxidation (M); taxonomy: all entries; enzyme: trypsin; missedcleavages: 1; peptide tolerance: ±50 ppm, MS/MS tolerance: ±0.05 Da.

Fig. 1 – 2-DE images of honeybee salivary glands. a is 2-DE gels stained with the CBB G-250 and phosphoprotein-specificfluorescent dye (Pro-Q Diamond) of cephalic salivary glands of nurse bees from normal colonies (NNH) and single-cohortcolonies (SNH), foragers from normal colonies (NFH) and single-cohort colonies (SFH), respectively. b is 2-DE gels stained withCBB and Pro-Q Diamond of thoracic salivary glands of nurse bees from normal colonies (NNT) and single-cohort colonies (SNT),foragers from normal colonies (NFT) and single-cohort colonies (SFT), respectively. Proteins are separated on 17 cm IPG gelstrips (pI 3–10, Linear) with 400 μg of sample loading, followed by 12.5% SDS-PAGE. Differentially expressed protein spots ofidentified proteins are labeled with color codes, where black, red, blue and green indicate up-regulated in each gland.

3J O U R N A L O F P R O T E O M I C S 8 7 ( 2 0 1 3 ) 1 – 1 5


2.5. Quantitative and bioinformatics analysis

The expression profiles of all the identified proteins of theHSG and TSG from the normal and single-cohort colony ofnursing and foraging workers were analyzed by using clusteranalysis software (version 3.0).

The identified proteins and phosphoproteins were anno-tated by searching against the Uniprot database (http://www.uniprot.org/) and grouped on the basis of their biologicalprocess of Gene Ontology (GO) terms. The biological interac-tion network (BIN) of the identified proteins was predictedusing Pathway Studio (Ariadne Genomics, version 8.0) soft-ware as in a previous report [21].

To enrich the identified proteins to specific GO terms, theprotein list was analyzed by CluoGo software applied to theDrosophila database downloaded from the GO database(release date, April 19, 2012). Ontology was selected as thebiological process. Enrichment analysis was done by right-side hypergeometric statistical testing and the probabilityvalue was corrected by Bonferroni method.

2.6. Quantitative real-time PCR (qRT-PCR)

To test the protein expression at gene level, total RNA wasextracted from the dissected samples (around 20 mg for eachsample) using TRIzol regent (Takara Bio, Kyoto, Japan). Eachsample was analyzed individually and processed in triplicate.Eight differentially expressed proteins from six major func-tional groups (carbohydrate and energy metabolism, protein

folding, development, protein metabolism, cytoskeleton andcellular homeostasis) were selected to detect the correspond-ing mRNA expression levels on the basis of their connectivityin the network. Reverse transcription was performed using anRNA PCR Kit (Takara Bio, Kyoto, Japan), according to themanufacturer's instructions. Beta-tubulin was used as thereference gene. The real-time PCR experiment and followingdata analysis were carried out as in our previously reportedprotocol [22].

2.7. Western blot analysis

To verify the results of differentially expressed proteins atthe protein level, vha55, Hsp60 and Hsp83 were selected forwestern blot analysis. The primary rabbit polyclonal antibod-ies were anti-vha55, Hsp60, Hsp83 (Abcam, Cambridge, MA,USA) at dilutions of 1:2000, 1:3000, 1:3000, respectively. Thesecondary antibody was horseradish peroxidase-conjugatedgoat anti-rabbit at a dilution of 1:10000. Each 14 μg samplewas separated by stacking (4%) and separating (12%) theSDS-PAGE gels. Beta-tubulin was used as the reference andthe western blot and data analysis were performed accordingto our previous method [22].

3. Results

In order to gain better understanding of how the honeybeeSGs carry out their biological missions during the life

Fig. 1 (continued).



transition from nursing to foraging, the proteome andphosphoproteome of honeybee worker SGs from normal andsingle-cohort colonies were analyzed. The number of totalproteins varied from 242 to 317 of the SGs of nurse bees andforagers from the normal and single-cohort colonies, whilethe phosphoprotein spots visualized by Pro-Q Diamond variedfrom 83 to 101 (Fig. 1), indicating that ~32% of the totalproteins were modified by phosphorylation. In total, 113 totalprotein spots and 64 phosphoprotein spots were altered intheir expression (>1.5-fold and p < 0.05). Of which 86 totalprotein spots and 33 phosphoprotein spots were successfullyidentified (Tables S1 and S2). The remaining proteins andphosphoproteins were unidentified due to their low abun-dance which resulted either in inability to produce enoughMS/MS spectra or, in search scores too low to yield unambig-uous results. Moreover, most of the highly sensitive fluores-cent stained proteins were present in low abundance andunder the detection limit of CBB stain during spot excision.

3.1. Qualitative comparisons of differentially expressedproteins

The 86 identified proteins (including 36 phosphoproteins), wereclassified into nine functional categories based on theirbiological activities (Fig. 2). The most highly represented groupwas that of proteins related to carbohydrate and energymetabolism (37.2%, 23 proteins/32 spots), followed by proteinsassociated with cytoskeleton (16.3%, seven proteins/14 spots),protein folding (15.1%, 11 proteins with 13 spots), proteinmetabolism (8.1%, six proteins/seven spots), development(8.1%, seven proteins), lipid metabolism (5.8%, four proteins/five spots), antioxidant activities (3.5%, three proteins), cellularhomeostasis (3.5%, three proteins), and molecular transporter(2.4%, two proteins) (Fig. 2, Table S1). The different species ofsome proteins, such asMlc2, and their different distributions ingelsmight be caused by some post-translational modifications,e.g. phosphorylation, thus resulting in a shift of Mr and pI ofthese proteins [23].

Although the number of up-regulated proteins was variedwith the task dependent manner of the honeybee workers,the overall tendency was that the majority (48 out of 76 ofHSG, Fig. 3a and 41 out of 69 in TSG, Fig. 3b) of thosedifferentially expressed proteins showed stronger expressionin the head and thoracic glands of the workers in the normalcolonies.

Fig. 3 – Comparisons of the up-regulated protein number inthe salivary gland. a represents the number of up-regulatedprotein spots in the head salivary gland of nurse bees fromnormal colonies (NNH) and single-cohort colonies (SNH),foragers from normal colonies (NFH) and single-cohortcolonies (SFH), respectively. b represents the number ofup-regulated protein spots thoracic salivary gland of nursebees from normal colonies (NNT) and single-cohort colonies(SNT), foragers from normal colonies (NFT) and single-cohortcolonies (SFT), respectively.

Fig. 2 – Function category of all the differently expressed total proteins of head and thoracic salivary glands from normal andsingle-cohort colonies, respectively. Color codes represent different protein functional groups.



As for the 33 phosphoproteins, cytoskeleton was the mostrepresented form (10 proteins), followed by seven with foldingfunctions. The others were involved in carbohydrate and

energy metabolism (6), protein metabolism (4), development(3), molecular transporters (2), and cellular homeostasis (1)(Table S2).



Fig. 4 – Unsupervised hierarchical biclustering analysis of the differentially expressed proteins in the salivary glands of honeybeeworkers. a is the results of the cephalic salivary gland of nurse bees fromnormal colonies (NNH) and single-cohort colonies (SNH),foragers from normal colonies (NFH) and single-cohort colonies (SFH), respectively. b is the results of thoracic salivary gland ofnurse bees from normal colonies (NNT) and single-cohort colonies (SNT), foragers from normal colonies (NFT) and single-cohortcolonies (SFT), respectively. The columns represent the different samples, and the rows represent the individual protein. Theup- or down-regulated proteins are indicated in red and green, respectively. The intensity of the color increases with increasingexpression differences as noted on the key bar on the top left side. Protein name is indicated on the right.



3.2. Quantitative analysis of proteomeand phosphoproteome

Interestingly, the NFH and SFH, NNH and SNH (Fig. 4a), NFTand SFT, NNT and SNT (Fig. 4b) were each clustered together,respectively, based on the unsupervised biclustering [24].

For the total proteome, 76 identified protein spots signif-icantly changed their expression in the HSG of honeybeeworkers in the normal and single-cohort colonies (Fig. 4a). Ofthe 36 proteins up-regulated in the NFH, metabolism ofcarbohydrate and energy, and protein metabolism were themajor protein families, accounting for 58.3% and 11.1%,respectively (Fig. S1, Table S3). In the SFH, of the 18 proteinswhich elevated their expression, proteins involved in carbo-hydrate and energy metabolism (38.8%), protein folding(22.2%) and cytoskeletal proteins (16.7%) were the majorforms (Fig. S1, Table S3). In addition, twelve proteins inseven functional groups and ten proteins in six functionalgroups were up-regulated in the NNH (Fig. S1, Table S3) andSNH (Fig. S1, Table S3), respectively.

In the TSG, 69 proteins were differentially expressed in thenormal and single-cohort colonies (Fig. 4b). Of those 24up-regulated in the NFT, proteins related to carbohydrate andenergy metabolism (37.5%), protein folding (20.8%) and cyto-skeletal proteins (20.8%) were the majority (Fig. S1, Table S3).Among the 15 proteins with escalated levels of expression inthe SFT, the most highly represented group was involved incarbohydrate and energy metabolism (46.6%) (Fig. S1, Table S3).In regards to the NNT, 17 proteins implicated in carbohydrateand energy metabolism (35.3%), protein folding (17.6%) anddevelopment regulation (17.6%)were themain functional group(Fig. S1, Table S3). Thirteen proteins were up-regulated in theSNT, the main functional categories were carbohydrate andenergy metabolism (23.1%), cytoskeleton (23.1%), protein fold-ing (15.4%) and cellular homeostasis (15.4%) (Fig. S1, Table S3).

In respect to the 33 differentially expressed phosphopro-teins in the honeybee SGs (Table S2), 27 proteins weredifferentially expressed in the HSG. Of which 13 proteinswere strongly expressed in the NNH, and were associatedwith four cytoskeletal proteins, three carbohydrate and

Fig. 5 – Predicted biological interaction network of the differentially expressed proteins and phosphoproteins in honeybeecephalic and thoracic salivary glands of normal and single-cohort colonies using Pathway Studio software. Those highlightedin green represent the key node proteins which altered their expression level. Protein entities which belong to distinctfunctional groups were represented in different shapes according to the default settings of the software as described in thelegend.



energy metabolisms, three developments, two protein metab-olisms and one with folding functions. In the NFH, six proteinshad higher levels of expression, including four cytoskeletalproteins, one molecular transporter and one with foldingfunctions (Table S4). In the SFH, seven differential proteinswere enhanced in their expression and they were associatedwith one cytoskeletal protein, two protein metabolisms, onemolecular transporter, two proteins with folding functions andone carbohydrate metabolism. In contrast, only one cytoskele-tal protein was up-regulated in the SNH (Table S4).

Among the 26 proteins that changed their expression inthe TSG, 14 proteins involved in five functional groups hadelevated expression in SFT, namely cytoskeletal (6), proteinmetabolism (3), protein folding (3), development (1), carbohy-drate and energy metabolism (1), and protein metabolism (1).Six proteins were up-regulated in the NFT, such as fivecytoskeletal proteins and one related to carbohydrate andenergy metabolism. Of the five up-regulated proteins in theNNT, four were related to protein folding and one wasinvolved in developmental regulation. Only one protein wasup-regulated in the SNT (Table S4).

3.3. BIN analysis and GO term enrichment

In total, 43 proteins were highly linked to the BIN as key nodeproteins in the SGs (Fig. 5). Proteins associatedwith carbohydrateandenergymetabolismwere themost highly represented (41.8%,18 proteins). They were CG1516, Zw, CG7430, Eno, CG7920, Pgd,Idh, CG12262, Ald, l(1)g0334, Argk, Ter94, ATPsyn-beta, CG5362,Pglym78, CG8036, CG6084, and ATPsyn-d. The second largestgroup (18.6%, eight proteins) was proteinswith folding functions,such as ERp60, Hsp60, Hsc70-4, Hsc70-5, Hsc70-3, Gp93, Hsp83

and PDI. The third largest was four protein metabolism relatedproteins (9.3%), i.e. pros29, prosbeta2, ef1gamma and eIF-5a.The other groups (30.3%) were: development (l(2)37 cc, Idgf4,14-3-3zeta), cellular homeostasis (vha55, vha100-1, vha26), lipidmetabolism (CG10527, MFE-2, Men), cytoskeletal proteins (Tm1and Tm2) and two proteins with antioxidant activities (jafrac1and trx-1).

The identified proteins were significantly enriched in fivefunctional groups, namely protein folding, carbohydrate andenergy metabolism, cellular homeostasis, protein metabolismand antioxidant (Fig. 6). The leading term (a term which isstatistically highly significant or with the lowest p-value ofeach functional group) in protein folding was “protein folding”in which Crc, ERp60, Gp93, Hsc70-3, Hsc70-4, Hsc70-5, Hsp60,Hsp83, PDI and Clc were significantly enriched. With the“cellular homeostasis” as the leading term, CG7430, ERp60,Jafrac1, PDI, Trxr-1, vha100-1, and vha55 were significantlyenriched. As for carbohydrate and energymetabolism, “glucosemetabolic process” was the leading term and Ald, CG1516, Eno,Pgd, Pglym78, transaldolase, Zw, l(1)G0334 and CG7430 weresignificantly enriched. The leading term in protein metabolismwas “translation elongation” and Ef1gamma, RpLP1, RpLP2, Cpaand eIF-5a were significantly enriched. Proteins related toantioxidant activities were significantly enriched for “responseto oxygen level” with Ald and Trxr-1.

3.4. Test of differentially expressed proteins by qRT-PCR

To test the tendency of protein expression at the transcrip-tional level, eight key node proteins were selected in the BINfor qRT-PCR analysis. The mRNA expression levels of thosegenes, Pglym78, Argk, Hsp 60, eIF-5a, RpLp2, 14-3-3zeta, vha55

Fig. 6 – Function enrichment analysis of the differentially expressed proteins. The * or ** indicate significant enriched geneonthology (GO) terms at the p < 0.05 and p < 0.01 statistical levels, respectively. Numbers on the horizontal axis indicatenumber of proteins of each term. The percentages in the bars indicate the ratio of associated proteins to GO process in eachterm.



and Act42a were consistent with their protein expressionaltendency from the 2-DE results (Fig. 7).

3.5. Western blot analysis

The expressional tendency of the three proteins was furtherverified at the protein level by western blot analysis. Hsp60,Hsp83 and vha55 were generally consistent with the results of2-DE analysis (Fig. 8).

4. Discussion

In order to investigate the molecular basis underlying howthe honeybee SGs achieve the multifaceted biologicalmission of efficient colony management, the proteome andphosphoproteome were compared between the SGs of nursebees and foragers from the normal and single-cohort colony.The alterations of protein and phosphoprotein expressionlevels among the HSG and TSG are the major driving forces for

the worker bees to accomplish their variety of duties inaccordance with the different physiological status. The proteinexpression in the SG is task-related rather than age-dependentand HSG is the key role player in the regulation of transitionfrom in-hive to foraging behavior through stronger expressionof proteins related to metabolism of juvenile hormone (JH) andethyl oleate (EO). The up-regulations of several arrays ofproteins in the TSG play supporting roles in helping foragersprocessing honey by synthesis and secretion of saliva intonectar.

Most of the differentially expressed phosphoproteins aremainly implicated in cytoskeletal and proteins with foldingactivities. The phosphorylations of those proteins either playimportant roles in trafficking, or improving the foldingefficiency of the chaperones and folding enzymes [25], so asto satisfy the synthesis, transportation and secretion ofproteins associated with regulation of labor-division in theHSG, or proteins involved in sugar digestion in the TSG tosufficiently convert polysaccharide (sucrose) to monosaccha-ride (glucose and fructose) thus improving the honey quality.

RpLp2 RpLp2 14-3-3zeta 14-3-3zeta

Vha55 vha55 act42a act42a

Pglym78 Pglym78 Argk Argk

Hsp60 Hsp60 eIF-5a eIF-5a

Fig. 7 – Test of the eight differentially expressed proteins at the mRNA level by quantitative real time PCR analysis. The“closed” and “open” circles represent the fold changes of protein and mRNA, respectively. Error bar is standard deviation.



Unsupervised hierarchical clustering is based on thedegree of similarity among different datasets [26], so thesimilar profiles among the datasets indicate intrinsic relationsbetween them. The NFH and SFH, NNH and SNH, NFT and SFT,NNT and SNT were clustered together in pairs (Fig. 4),indicating that they have similar protein expression patternsin each pair. Due to the fact that the young bees, especiallyforagers (~7-day-old), in single-cohort colonies begin theirfield activities much younger than that of normal colonies[27], it is thus believed that the protein expression feature ofthe SG is task-related rather than age-dependent so as toguarantee the gland to perform their normal functionality inthe single-cohort colony [16]. This provides the first evidenceat the protein level in the aid of understanding this importantaspect of honeybee biology.

Proteins related to carbohydrate and energy metabolismare necessary to produce energetic fuels for the livingorganism. The up-regulations of 21 proteins in the NFH aremainly involved in gluconeogenesis, tricarboxylic acid cycle,pentose-phosphate shunt, glycolysis, oxidative phosphoryla-tion and ATP generation. They work together to cope with theextremely carbohydrate-rich diet [28] so as to satisfy theenergy demand for normal functionality of HSG of workerbees. The large number of proteins up-regulated in the NFH isprobably because the enhanced carbohydrate metabolism inthe HSG demands high energetic fuels during foragingactivities [29]. This is in line with a 15% increase in themetabolic rate in foragers at the onset of foraging compared tonurse bees [30]. In addition, the generated energy is likelybeing used to synthesize, sort, and transport the substancesproduced by the HSG involved in wax manipulation, mouth-part lubrication [9], marking of a food source [10] and for theprogressive development of the gland itself [31]. In contrast,the small number of proteins up-regulated in the SFH that

coincide with the flight capacity of precocious foragers of thesingle-cohort colony is much lower and requires less energyconsumption than that of normal-aged counterparts [32,33]. Itis well documented that the TSG is a gland where theenzymes involved in honey and sugar digestion are produced[9]. Those enzymes not identified in this study are likely dueto the lower abundance that could not detected by 2-DE basedproteomic, which is one of limitations of this traditionalgel-based proteomic method [34]. However, we hypothesisthat all the enzymes associated with successful conversion ofsugar into honey should up-regulate both in NFT and SFT toachieve these biological roles [9]. This requires furtherresearch, which employs mass spectrometry based proteo-mics that could provide unbiased identification of both lowand high abundant proteins in a biological sample [35]. Theidentification of differentially expressed proteins not relatedto sugar conversion may play a supporting role as an energyresource for the biosynthesis and secretion of those proteins.In the TSG, the strong expression of zwischenferment (Zw,spot 21), enolase (spot 29), aldolase (Ald, spots 40 and 41) andphosphoglyceromutase (Pglym78, spot 68) in the foragingworkers of the normal colonies is used to generate metabolicfuel [29] in order to guarantee the digestive activities, andmoistening of pollen and wax [9], rather than for their owndevelopment since no morphology changes occur among theyoung bees and older foragers [36]. Almost all of the sixphosphoproteins in this family are enzymes. The phosphor-ylation of these enzymes is thought to enhance theirenzymatic activity and to supply sufficient energy, and helpthe glands to better fulfill their biological duties [37,38].

Protein metabolism is a bidirectional process involvingprotein synthesis and protein breakdown. Eukaryotic transla-tion initiation factor 5A (eIF-5a), 60S acid ribosomal protein P1and P2 (RpLP1 and RpLP2), and translation elongation factor

Vha55 Vha55

Hsp60 Hsp60 Hsp83 Hsp83

Fig. 8 – Western blot analysis of heat shock protein 60 (Hsp60), heat shock protein 90 (Hsp 83) and vacuolar H+-ATPase 55 kDa(vha55). Protein samples were subjected to SDS-PAGE followed by western blot analysis. Hsp60, Hsp83 and vha55 weredetected using corresponding antibodies, respectively. Beta-tubulin was used as reference control. The “closed” and “open”circles represent the fold changes of 2-DE and western blot analysis, respectively. Error bar is standard deviation.



1-gamma (Ef1gamma) all participate in the protein synthesisprocess [39,40]. On the other hand, proteasome, such asproteasome 2 subunit CG3329 (prosbeta2) and proteasomesubunit alpha type 4 (pros29), are involved in proteindegradation under the control of proteasome–ubiquitin sys-tem [41]. The up-regulation of eIF-5a (spot 79), RpLP2 (spot 82),pros29 (spot 74) and prosbeta2 (spot 76) in the NFH is to satisfythe intensification of glandular activity in foragers [42]. Thesimilar number of proteins elevated their expression in theNFT, SFT, NNT and SNT and may be attributed to their similarprotein metabolism activities, which is in line with littlevariation in the amount of rough endoplasmic reticulum (RER)in TSG [43]. The up-regulation of phosphorylated eIF-5a (spot79), RpLP1(spot 81) and RpLP2 (spot 82) in SFT, is to improvethe synthesis of proteins related to honey and sugar di-gestions to satisfy the colony needs because the synthesisrate of precocious foragers is lower relative to their normalcounterparts [7]. This in line with the findings that theiractivities are improved after phosphorylation [44].

As for the exocrine glands, the secretary pathway ofhoneybee SGs demands a strict quality control (QC) mecha-nism for protein synthesis. Under the guidance of this QCsystem, the newly synthesized proteins are folded into theirnative conformations before their transportation [45]. Sincethe foragers have higher amounts of secretion than the nursesin the HSG [31], more chaperones and folding enzymes areinvolved in the QC system, such as the heat shock proteinfamily (Hsps), lectins (clathrin, Clc and calreticulin, Crc) andprotein disulfide-isomerase (PDI) are needed in the NFH andSFH than in the NNH and SNH to facilitate the proper foldingof those nascent proteins. Toward this goal, the up-regulationof phosphoproteins in the NFH (Clc, spot 60) and SFH (PDI,spots 14 and 15) is to increase the folding efficiency byenhancing their association to the unfolded and/or misfoldedproteins [25]. The high number of proteins strongly expressedin the NFT is thought necessary to satisfy the folding demandbecause the secretion amount in the TSG reaches the peaklevel in the foraging period [43].

During the honeybee behavioral development, increasingthe JH titer makes bees begin foraging earlier than average[16]. In contrast, EO can delay the onset of foraging byinhibiting the biosynthesis of JH [11]. The protein CG10527plays a key role in the biosynthesis of JH precursor [46,47],while the fatty acyl-CoA reductase (FAR) is related to theproduction of EO [48]. The up-regulation of CG10527 (spot 25)and down-regulation of FAR1 (spot 75) in SFH indicate a higherJH level but lower EO level in the small-cohort colony thanthat of the normal one. As a result, the foraging behavior ofyoung bees is earlier in the small-cohort but the premature isdelayed in the normal colony. This is in well agreement withthe “activator–inhibitor” model in nursing–foraging transition[16] and indicates that the HSG plays important roles in thelabor division of honeybee workers.

Chemosensory proteins (Csps) have multiple functionsranging from embryonic development to chemosensorysignal transduction [49]. The phosphorylated Csp1 (spot 84)in the NNH is supposed to enable nestmates recognition bysolubilizing cuticular hydrocarbons as in ants [50]. While thestrongly expressed Csp3 (spot 85) in the NFH might act as achemical messenger carrier to deliver semiochemicals by the

foragers and guide the nestmate to a food-rich place in thefield more quickly and efficiently than that of SFH [51,52].

Cytoskeletal proteins play important roles in intracellulartransport, cellular division and maintaining the stability ofthe cell shape [53,54]. Actin related protein (Acp) contributesto actin filament assembly and intracellular vesicular trans-port [55]. Myosin II, whose activity is controlled by thephosphorylation of myosin regulatory light chain 2 (Mlc2)[56], is required for formation and release of specific vesiclesubtypes in Golgi and trans-Golgi networks [57]. Tropomyosin1 (Tm1), together with Tm2, can remarkably improve micro-filament organization [58]. The phosphorylation of Tm1 canincrease cell contractility [59]. The up-regulation and phos-phorylation of Acp1 (spot 77) and Mlc2 (spots 53, 54, 57 and 58)in the NNH and Mlc2 (spots 53, 54, 57 and 58) in the NFH arenot only critical for the trafficking of the oily substanceproduced by the HSG [9] because vesicles are the main carrierfor lipid droplet transportation, but it also is involved in theprogressive development of the HSG itself. The similarnumber of proteins up-regulated in the NFT, SFT, NNT andSNT, respectively, is in line with the report that no significantmorphology changes occur from young bees to older foragers[60]. The up-regulations of protein and phosphoproteins,Mlc2, Tm1 and Tm2 in NFT, SFT, NNT and SNT are supposedto facilitate secretion by maintaining the proper shape of thesecretory cells [61–63] or increasing cell contractility duringthe secretory activities [59].

Cellular homeostasis in the external environment iscritical for facilitating normal cellular function. An appropri-ate pH value is vital for the performance of secretion and theposttranslational modifications and processing of secretedproteins are impaired [64], cargo is misdirected [65] when thepH gradient across Golgi membranes collapsed. Therefore, theup-regulations of V-ATPase, vacuolar H+-ATPase 55 kDa Bsubunit (vha55, spot 23) in the NFH, vacuolar ATP synthasesubunit E (vha26, spot 67) in the SNH and NFT, vha55 (spot 23)and vha100-1 CG1709 (vha100-1, spot 78) in the SNT arethought to be responsible for the acidification of secretoryvesicles in order to guarantee successful protein sorting,trafficking and modification [66]. The strong expression ofphosphoprotein vha100-1 (spot 78) in the SNT is believed torestore the activity of this V-ATPase [67] because its activitycan be impaired by starvation in the case of food shortage inthe single-cohort colony [68].

Growth factors are important for the regulation of theorgan development in honeybees [69]. The identification ofIdgf4 in this study is consistent with the fact that it issynthesized by the honeybee SGs [6]. The strongly expressedIdgf4 (spot 32) in the NNH is secreted into royal jelly [6] tostimulate the proliferation, polarization and motility ofimaginal disk cells of honeybee larvae with the aiding ofinsulin-like peptides [70] during the nursing activities. Thestrong expression of TCTP (spot 63) in SNH and NNT suggestsa role in regulation of the cell cycle since the silencing of TCTPcan lead to growth defects in Drosophila [71]. The expression ofAnn IX (spot 48) in this study is likely to be involved inexocytosis [72] and its strong expression in the NNH and NNTmanifests its role in assistance of secretion cargo viaparticipation of intralumenal vesicle formation [73]. Typically,the secretion activity of honeybee SGs is innervated [36] and



the expression of 14-3-3 proteins may participate in theregulation of secretion signal [74]. The up-regulation of14-3-3epsilon (spot 61) in the NFH may ascribe to the relativehigher secretion activity which needs more regulatory pro-teins to supervise the signal transduction. The phosphoryla-tion of 14-3-3 may make it more sensitive to the secretionsignals to secrete enzymes involved in nectar conversionduring foraging because phosphorylated 14-3-3 zeta makescells more susceptible to apoptotic signals [75]. The identifiedMRJP1 in the NNH and SNH indicates that the HSG has theability to synthesizeMRJPs while TSG does not [6] It may relateto regulating the development of the HSG and the phosphor-ylation could enhance the cell proliferation as it did in thehoneybee brain [76].

Antioxidants are used to protect the tissues and organsfrom the oxidative damage of the ROS [77]. It is known thatthioredoxin reductase 1 (Trxr-1), glutathione S transferase S1(GstS1), and thioredoxin peroxidase 1 (Jafrac1) are all acted asantioxidant proteins [78–80]. The up-regulation of Trxr-1 (spot30) in the NFH and NFT, and GstS1 (spot 72) in the SFH andSFT, indicates roles to cope with the enhanced ROS resultedfrom the higher metabolic rate in foragers than in nurse bees[29]. The up-regulation of Jafrac1 (spot 73) in the NNH andNNT is to intensify the resistance to toxicity of hydrogenperoxide [81] produced during the nursing activities in orderto maintain their normal functions.

As is well known, the majority of proteins in the living cellinteract with others for proper biological activities and thesignificant overrepresented GO terms could determine hy-potheses for the biological events behind the proteome data.Therefore, the proteins networked in the BIN and thesignificant enriched GO terms signify their centrality forfunctionality as key node proteins and their principal rolesat pathway level for the honeybee SGs. For example, proteinsare involve in carbohydrate and energy metabolism, proteinfolding, protein metabolism, cellular homeostasis and anti-oxidant activities. For those key node proteins identified inthe BIN, they have potential to be used as target proteins forfurther functional assays through RNAi, to improve the glandperformance on the basis of the tested results at gene andprotein levels. This provides a sound clue for the reversegenetic manipulation of honeybee SGs to selectively enhancetheir performance via target proteins thus promoting produc-tivity of the honeybee colony.

5. Conclusions

This study of the proteome and phosphoproteome enables usfor the first time to gain a new insight into the molecularunderpinning of how the SGs of the honeybee workersaccomplish their biological missions, through comparison ofnormal and single-cohort honeybee colonies. To fulfill thebiological duty, the SGs require a wide spectrum of proteins tosupport their multifaceted functions and ensure normalsocial management of the colony. Change of protein expres-sion and phosphorylation status aremajor role players relatedto carbohydrate and energy metabolism, protein folding,cellular homeostasis, protein metabolism, cytoskeleton, mo-lecular transporter, lipid metabolism, development, major

royal jelly protein and antioxidant activities. The HSG triggerslabor transition from in-hive work to foraging activities via theregulation of JH and EO levels, while the stronger expressions ofproteins involved in carbohydrate and energy metabolism,protein folding, protein metabolism, cellular homeostasisand cytoskeleton in TSG, support the gland to efficientlyenhance honey processing by synthesis and secretion of salivainto nectar during foraging. To maintain efficient social colonymanagement, appropriate demographic structure with correctage is of paramount importance to increase the colonyproduction efficiency.

Supplementary data to this article can be found online athttp://dx.doi.org/10.1016/j.jprot.2013.05.021.

Acknowledgments

We thank Katrina Klett from University of Minnesota, USA, forher help with the language of the manuscript. This work issupported by the earmarked fund for Modern Agro-industryTechnology Research System (CARS-45), Key Projects of theNational Scientific Supporting Plan of the 12th Five-YearDevelopment (2011-2015) (2011BAD33B04), and Science &Technology Fund from the Chinese Academy of AgriculturalScience (201207).

R E F E R E N C E S

[1] Robinson GE. Regulation of division of labor in insectsocieties. Annu Rev Entomol 1992;37:637–65.

[2] Winston ML. The biology of the honey bee. Cambridge, MA:Harvard University Press; 1987.

[3] Robinson GE, Vargo EL. Juvenile hormone in adult eusocialHymenoptera: gonadotropin and behavioral pacemaker. ArchInsect Biochem Physiol 1997;35:559–83.

[4] Fahrbach SE, Moore D, Capaldi EA, Farris SM, Robinson GE.Experience-expectant plasticity in the mushroom bodies ofthe honeybee. Learn Mem 1998;5:115–23.

[5] Toma DP, Bloch G, Moore D, Robinson GE. Changes in periodmRNA levels in the brain and division of labor in honey beecolonies. Proc Natl Acad Sci U S A 2000;97:6914–9.

[6] Fujita T, Kozuka-Hata H, Uno Y, Nishikori K, Morioka M,Oyama M, et al. Functional analysis of the honeybee (Apismellifera L.) salivary system using proteomics. BiochemBiophys Res Commun 2010;397:740–4.

[7] Katzav-Gozansky T, Soroker V, Ionescu A, Robinson GE,Hefetz A. Task-related chemical analysis of labial glandvolatile secretion in worker honeybees (Apis melliferaligustica). J Chem Ecol 2001;27:919–26.

[8] Conte YL, Bécard JM, Costagliola G, de Vaublanc G, MaataouiME, Crauser D, et al. Larval salivary glands are a source ofprimer and releaser pheromone in honey bee (Apis melliferaL.). Naturwissenschaften 2006;93:237–41.

[9] Simpson J. The functions of the salivary glands of Apismellifera. J Insect Physiol 1960;4:107–21.

[10] Jarau S, Hrncir M, Zucchi R, Barth F. A stingless bee uses labialgland secretions for scent trail communication (Trigonarecursa Smith 1863). J Comp Physiol A Neuroethol Sens NeuralBehav Physiol 2004;190:233–9.

[11] Leoncini I, Le Conte Y, Costagliola G, Plettner E, Toth AL,Wang M, et al. Regulation of behavioral maturation by aprimer pheromone produced by adult worker honey bees.Proc Natl Acad Sci U S A 2004;101:17559–64.



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

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

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

[15] Costa RAC, Cruz-Landim C. Enzymes present in the thoracicgland extracts from workers and males of Apis mellifera(Hymenoptera: Apidae). Sociobiology 2001;37:563–70.

[16] Huang ZY, Robinson GE. Honeybee colony integration:worker–worker interactions mediate hormonally regulatedplasticity in division of labor. Proc Natl Acad Sci U S A 1992;89:11726.

[17] Robinson GE. Regulation of honey bee age polyethism byjuvenile hormone. Behav Ecol Sociobiol 1987;20:329–38.

[18] Gala A, Fang Y, Woltedji D, Zhang L, Han B, Feng M, et al.Changes of proteome and phosphoproteome triggerembryo–larva transition of honeybee worker (Apis melliferaligustica). J Proteomics 2012;78:428–46.

[19] Han B, Zhang L, Feng M, Fang Y, Li J. An integrated proteomicsreveals pathological mechanism of honeybee (Apis cerena)Sacbrood disease. J Proteome Res 2013;12:1881–97.

[20] Zhang L, Fang Y, Li R, Feng M, Han B, Zhou T, et al. Towardsposttranslational modification proteome of royal jelly.J Proteomics 2012;75:5327–41.

[21] 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.

[22] Li J, Feng M, Begna D, Yu F, Aijuan Z. Proteome comparison ofhypopharyngeal gland development between Italian androyal jelly producing worker honeybees (Apis mellifera L.).J Proteome Res 2010;9:6578–94.

[23] Qian PY, Wong YH, Zhang Y. Changes in the proteome andphosphoproteome expression in the bryozoan Bugula neritinalarvae in response to the antifouling agent butenolide.Proteomics 2010;10:3435–46.

[24] Caldas J, Kaski S. Hierarchical generative biclustering formicrorna expression analysis. J Comput Biol 2011;18:251–61.

[25] Michalak M, Groenendyk J, Szabo E, Gold LI, Opas M.Calreticulin, a multi-process calcium-buffering chaperone ofthe endoplasmic reticulum. Biochem J 2009;417:651–66.

[26] Kull M, Vilo J. Fast approximate hierarchical clustering usingsimilarity heuristics. BioData Min 2008;1:1–14.

[27] Ben-Shahar Y, Dudek NL, Robinson GE. Phenotypicdeconstruction reveals involvement of manganesetransporter malvolio in honey bee division of labor. J Exp Biol2004;207:3281–8.

[28] Kunieda T, Fujiyuki T, Kucharski R, Foret S, Ament S, Toth A,et al. Carbohydrate metabolism genes and pathways ininsects: insights from the honey bee genome. Insect Mol Biol2006;15:563–76.

[29] Harrison JF, Fewell JH. Environmental and genetic influenceson flight metabolic rate in the honey bee, Apis mellifera. CompBiochem Physiol A Mol Integr Physiol 2002;133:323–33.

[30] Bloch G, Wheeler DE, Robinson GE. Endocrine influences onthe organization of insect societies. Horm Brain Behav 2002;3:195–235.

[31] Poiani SB, Cruz-Landim C. Changes in the size of cephalicsalivary glands of Apis mellifera and Scaptotrigona postica(Hymenoptera: Apidae) queens and workers in different lifephases. Zoologia 2010;27:961–4.

[32] Vance JT, Williams JB, Elekonich MM, Roberts SP. The effectsof age and behavioral development on honey bee (Apismellifera) flight performance. J Exp Biol 2009;212:2604–11.

[33] Page RE, Peng CYS. Aging and development in social insectswith emphasis on the honey bee, Apis mellifera L. ExpGerontol 2001;36:695–711.

[34] Chevalier F. Highlights on the capacities of “Gel-based”proteomics. Proteome Sci 2010;8:23.

[35] Monteoliva L, Albar JP. Differential proteomics: an overviewof gel and non-gel based approaches. Brief Funct GenomicProteomic 2004;3:220–39.

[36] Schnitzer K, Seifert P. Anatomy and ultrastructure of thesalivary gland in the thorax of the honeybee worker, Apismellifera (Insecta, Hymenoptera). Zoomorphology 1990;109:211–22.

[37] Miyazaki T, Neff L, Tanaka S, Horne WC, Baron R. Regulationof cytochrome c oxidase activity by c-Src in osteoclasts. J CellBiol 2003;160:709–18.

[38] Fang JK, Prabu SK, Sepuri NB, Raza H, AnandatheerthavaradaHK, Galati D, et al. Site specific phosphorylation of cytochromec oxidase subunits I, IVi1 and Vb in rabbit hearts subjected toischemia/reperfusion. FEBS Lett 2007;581:1302–10.

[39] Kang HA, Hershey J. Effect of initiation factor eIF-5A depletionon protein synthesis and proliferation of Saccharomycescerevisiae. J Biol Chem 1994;269:3934–40.

[40] Tchórzewski M. The acidic ribosomal P proteins. Int JBiochem Cell Biol 2002;34:911–5.

[41] Trombetta ES, Parodi AJ. Quality control and protein foldingin the secretory pathway. Annu Rev Cell Dev Biol 2003;19:649–76.

[42] Poiani SB, Cruz-Landim CD. Morphological changes in thecephalic salivary glands of females and males of Apis melliferaand Scaptotrigona postica (Hymenoptera, Apidae). J Biosci2010;35:249–55.

[43] Deseyn J, Billen J. Age-dependent morphology andultrastructure of the hypopharyngeal gland of Apis melliferaworkers (Hymenoptera, Apidae). Apidologie 2005;36:49–57.

[44] Rich BE, Steitz JA. Human acidic ribosomal phosphoproteinsP0, P1, and P2: analysis of cDNA clones, in vitro synthesis, andassembly. Mol Cell Biol 1987;7:4065–74.

[45] Nishikawa S, Brodsky JL, Nakatsukasa K. Roles of molecularchaperones in endoplasmic reticulum (ER) quality controland ER-associated degradation (ERAD). J Biochem 2005;137:551–5.

[46] Burtenshaw S, Su P, Zhang J, Tobe S, Dayton L, Bendena W. Aputative farnesoic acid O-methyltransferase (FAMeT)orthologue in Drosophila melanogaster (CG10527): relationshipto juvenile hormone biosynthesis? Peptides 2008;29:242–51.

[47] Zhang H, Tian L, Tobe S, Xiong Y, Wang S, Lin X, et al.Drosophila CG10527 mutants are resistant to juvenilehormone and its analog methoprene. Biochem Biophys ResCommun 2010;401:182–7.

[48] Teerawanichpan P, Robertson AJ, Qiu X. A fatty acyl-CoAreductase highly expressed in the head of honey bee (Apismellifera) involves biosynthesis of a wide range of aliphaticfatty alcohols. Insect Biochem Mol Biol 2010;40:641–9.

[49] Foret S, Wanner KW, Maleszka R. Chemosensory proteinsin the honey bee: insights from the annotated genome,comparative analyses and expressional profiling. InsectBiochem Mol Biol 2007;37:19–28.

[50] Ozaki M, Wada-Katsumata A, Fujikawa K, Iwasaki M,Yokohari F, Satoji Y, et al. Ant nestmate and non-nestmatediscrimination by a chemosensory sensillum. Science2005;309:311.

[51] Jacquin-Joly E, Vogt RG, François MC, Nagnan-Le Meillour P.Functional and expression pattern analysis of chemosensoryproteins expressed in antennae and pheromonal gland ofMamestra brassicae. Chem Senses 2001;26:833–44.

[52] Pelosi P, Calvello M, Ban L. Diversity of odorant-bindingproteins and chemosensory proteins in insects. Chem Senses2005;30:i291–2.

[53] Tuszynski JA, Carpenter EJ, Huzil JT, Malinski W, Luchko T,Luduena RF. The evolution of the structure of tubulin and itspotential consequences for the role and functionofmicrotubulesin cells and embryos. Int J Dev Biol 2006;50:341–58.



[54] Kirschner M, Schulze E. Morphogenesis and the control ofmicrotubule dynamics in cells. J Cell Sci Suppl 1986;5:293–310.

[55] EckleyDM, Schroer TA. Interactions between the evolutionarilyconserved, actin-related protein, Arp11, actin, and Arp1. MolBiol Cell 2003;14:2645–54.

[56] Somlyo AP, Somlyo AV. Ca2+ sensitivity of smooth muscleand nonmuscle myosin II: modulated by G proteins, kinases,and myosin phosphatase. Physiol Rev 2003;83:1325–58.

[57] Heimann K, Percival JM, Weinberger R, Gunning P, Stow JL.Specific isoforms of actin-binding proteins on distinctpopulations of Golgi-derived vesicles. J Biol Chem 1999;274:10743–50.

[58] Shah V, Bharadwaj S, Kaibuchi K, Prasad G. Cytoskeletalorganization in tropomyosin-mediated reversion ofras-transformation: evidence for Rho kinase pathway.Oncogene 2001;20:2112.

[59] Houle F, Rousseau S, Morrice N, Luc M, Mongrain S, TurnerCE, et al. Extracellular signal-regulated kinase mediatesphosphorylation of tropomyosin-1 to promote cytoskeletonremodeling in response to oxidative stress: impact onmembrane blebbing. Mol Biol Cell 2003;14:1418–32.

[60] Deseyn J, Billen J. Age-dependent morphology andultrastructure of the thoracic labial gland of Apis melliferacarnica Pollm. workers (Hymenoptera, Apidae). Belg J Zool2004;134:55–9.

[61] Sellers JR. Myosins: a diverse superfamily. Biochim BiophysActa 2000;1496:3–22.

[62] Sokac AM, Bement WM. Regulation and expression ofmetazoan unconventional myosins. Int Rev Cytol 2000;200:197–304.

[63] Wu X, Jung G, Hammer JA. Functions of unconventionalmyosins. Curr Opin Cell Biol 2000;12:42–51.

[64] Carnell L, Moore HP. Transport via the regulated secretorypathway in semi-intact PC12 cells: role of intra-cisternalcalcium and pH in the transport and sorting of secretograninII. J Cell Biol 1994;127:693–705.

[65] Chanat E, Huttner WB. Milieu-induced, selective aggregationof regulated secretory proteins in the trans-Golgi network.J Cell Biol 1991;115:1505–19.

[66] Yan Y, Denef N, Schüpbach T. The vacuolar proton pump,V-ATPase, is required for notch signaling and endosomaltrafficking in Drosophila. Dev Cell 2009;17:387–402.

[67] Voss M, Vitavska O, Walz B, Wieczorek H, Baumann O.Stimulus-induced phosphorylation of vacuolar H+-ATPase byprotein kinase A. J Biol Chem 2007;282:33735–42.

[68] Kane PM. Regulation of V-ATPases by reversible disassembly.FEBS Lett 2000;469:137–41.

[69] Deuel TF. Polypeptide growth factors: roles in normal andabnormal cell growth. Annu Rev Cell Biol 1987;3:443–92.

[70] 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.

[71] Hsu Y-C, Chern JJ, Cai Y, Liu M, Choi K-W. Drosophila TCTP isessential for growth and proliferation through regulation ofdRheb GTPase. Nature 2007;445:785–8.

[72] Johnstone SA, Hubaishy I, Waisman D. Phosphorylation ofannexin II tetramer by protein kinase C inhibits aggregationof lipid vesicles by the protein. J Biol Chem 1992;267:25976–81.

[73] Futter CE, White IJ. Annexins and endocytosis. Traffic 2007;8:951–8.

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

[75] Sunayama J, Tsuruta F,MasuyamaN, GotohY. JNK antagonizesAkt-mediated survival signals by phosphorylating 14-3-3. J CellBiol 2005;170:295–304.

[76] Garcia L, Saraiva Garcia CH, Calábria LK, Costa Nunes da CruzG, Sánchez Puentes A, Báo SN, et al. Proteomic analysis ofhoney bee brain upon ontogenetic and behavioraldevelopment. J Proteome Res 2009;8:1464–73.

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

[78] Li J, Fang Y, Zhang L, Begna D. Honeybee (Apis melliferaligustica) drone embryo proteomes. J Insect Physiol 2010;57:372–84.

[79] Zheng A, Li J, Begna D, Fang Y, Feng M, Song F. Proteomicanalysis of honeybee (Apismellifera L.) pupae head development.PLoS One 2011;6:e20428.

[80] Woltedji D, Song F, Zhang L, Gala A, Han B, Feng M, et al.Western honeybee drones and workers (Apis melliferaligustica) have different olfactory mechanisms than Easternhoneybees (Apis cerana cerana). J Proteome Res 2012;11:4526–40.

[81] Radyuk SN, Sohal RS, Orr WC. Thioredoxin peroxidases canfoster cytoprotection or cell death in response to differentstressors: over-and under-expression of thioredoxin peroxidasein Drosophila cells. Biochem J 2003;371:743–52.


novel aspects of understanding molecular working ... · the salivary system is exocrine glands that...

Documents