exoskeleton formation in apis mellifera: cuticular hydrocarbons profiles and expression of...

Q3

lable at ScienceDirect

Insect Biochemistry and Molecular Biology xxx (2014) 1e14

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354

55

IB2576_proof ■ 10 May 2014 ■ 1/14

Contents lists avai

Insect Biochemistry and Molecular Biology

journal homepage: www.elsevier .com/locate/ ibmb

565758596061626364656667686970717273747576

Exoskeleton formation in Apis mellifera: Cuticular hydrocarbonsprofiles and expression of desaturase and elongase genes during pupaland adult development

Tiago Falcón a, Maria Juliana Ferreira-Caliman b, Francis Morais Franco Nunes c,Érica Donato Tanaka a, Fábio Santos do Nascimento b, Márcia Maria Gentile Bitondi b,*a Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes 3900, 14049-900 Ribeirão Preto, SP, Brazilb Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes 3900, 14040-901 Ribeirão Preto, SP, BrazilcCentro de Ciências Biológicas e da Saúde, Universidade Federal de São Carlos, Rod. Washington Luís, km 235, 13565-905 São Carlos, SP, Brazil

7778
79808182838485868788899091
a r t i c l e i n f o

Article history:Received 11 March 2014Received in revised form23 April 2014Accepted 25 April 2014

Keywords:Cuticular hydrocarbonsApis melliferaDesaturaseElongaseCuticular envelopeInsect exoskeleton

* Corresponding author. Tel.: þ55 1636023805.E-mail address: [email protected] (M.M. Gentile Bit

http://dx.doi.org/10.1016/j.ibmb.2014.04.0060965-1748/� 2014 Published by Elsevier Ltd.

9293949596979899

100

Please cite this article in press as: Falcón, T.desaturase and elongase genes during pup10.1016/j.ibmb.2014.04.006

a b s t r a c t

Cuticular hydrocarbons (CHCs) are abundant in the superficial cuticular layer (envelope) of insects wherethey play roles as structural, anti-desiccation and semiochemical compounds. Many studies haveinvestigated the CHC composition in the adult insects. However, studies on the profiles of these com-pounds during cuticle formation and differentiation are scarce and restrict to specific stages of a fewinsect species. We characterized the CHCs developmental profiles in the honeybee workers during anentire molting cycle (from pupal-to-adult ecdyses) and in mature adults (forager bees). Gas chroma-tography/mass spectrometry (GC/MS) analysis revealed remarkable differences in the relative quantitiesof CHCs, thus discriminating pupae, developing and newly-ecdysed adults, and foragers from each other.In parallel, the honeybee genome database was searched for predicted gene models using known aminoacid sequences of insect enzymes catalyzing lipid desaturation (desaturases) or elongation (elongases) asqueries in BLASTP analysis. The expression levels of six desaturase genes and ten elongase genespotentially involved in CHC biosynthesis were determined by reverse transcription and real time poly-merase chain reaction (RT-qPCR) in the developing integument (cuticle and subjacent epidermis).Aiming to predict roles for these genes in CHC biosynthesis, the developmental profiles of CHCs anddesaturase/elongase transcript levels were evaluated using Spearman correlation coefficient. This anal-ysis pointed to differential roles for these gene products in the biosynthesis of certain CHC classes. Basedon the assumption that homologous proteins may share a similar function, phylogenetic trees werereconstructed as an additional strategy to predict functions and evolutionary relationships of the hon-eybee desaturases and elongases. Together, these approaches highlighted the molecular complexityunderlying the formation of the lesser known layer of the cuticular exoskeleton, the envelope.

� 2014 Published by Elsevier Ltd.

101102
103 104105106107108109110111112
1. Introduction

The cuticular exoskeleton of insects ismainly formed by proteins,the polysaccharide chitin, and lipids arranged as a complex multi-layered structure: the inner procuticle comprising the endocuticleand exocuticle, the epicuticle and an outermost envelope. Thesefunctional layers are sequentially produced and are secreted by theepidermis at each molt cycle, and differ from each other inbiochemical composition andphysiological properties. The envelope

ondi).

113114115116

, et al., Exoskeleton formational and adult development, In

mainly consists of lipids that form a barrier against water loss andinvading pathogens, and also serve as important cues for chemicalcommunication, besides acting as sex pheromones (Wigglesworth,1970; Blomquist and Dillwith, 1985; Gibbs, 2002; Châline et al.,2005). This lipid layer is largely enriched with hydrocarbons(Hepburn, 1985), which are synthesized in specialized cells calledoenocytes (Piek,1964; Diehl,1973; Schal et al., 1998; Fan et al., 2003;Billeter et al., 2009). In honeybeeworkers the oenocytes are localizedin close associationwith the epidermis and the parietal fat body thatinternally coat the exoskeleton, andaremore frequently found closerto the sternites than the tergites (Ruvolo and Landim, 1993).

Intermediates and end-products of metabolic pathways, such asfatty acids, in addition to specific enzyme classes, are involved in

117118119

in Apis mellifera: Cuticular hydrocarbons profiles and expression ofsect Biochemistry and Molecular Biology (2014), http://dx.doi.org/

mailto:[email protected]

www.sciencedirect.com/science/journal/09651748

http://www.elsevier.com/locate/ibmb

http://dx.doi.org/10.1016/j.ibmb.2014.04.006



T. Falcón et al. / Insect Biochemistry and Molecular Biology xxx (2014) 1e142

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 2/14

CHC biosynthesis. Key enzymes in the CHCs biosynthetic pathwaysare the fatty acid synthases, elongases, desaturases, reductases anda P450 decarbonylase. Two forms of fatty acid synthases are likelyinvolved in the synthesis of unbranched andmethyl-branched fattyacids. Elongases catalyze the chain elongation of saturated andunsaturated fatty acids, which are converted to aldehydes byreductase enzymes. Aldehydes are substrates for the last step ofhydrocarbons biosynthetic pathway, i.e., the oxidative decarbon-ylation catalyzed by a P450 enzyme. Desaturation, i.e., the insertionof carbonecarbon double bonds into the saturated fatty acid chainand consequent conversion to an unsaturated fatty acid is catalyzedby desaturases (Kolattukudy, 1965, 1968; Howard and Blomquist,2005; Blomquist et al., 2012; Qiu et al., 2012). After being synthe-sized and released from the oenocytes, the CHCs are transportedthrough the hemolymph by lipophorins (Chino and Kitazawa,1981;Chino et al., 1981). CHCs reach the insect surface via the pore canalsof the cuticular exoskeleton (Blomquist and Dillwith, 1985), wherethey integrate the envelope layer.

The multicomponent CHC blend shows a great variation amonginsect species. CHCs also display both quantitative and qualitativeintraspecific variation depending on the developmental stage,environment changes and food availability. The majority of theinformation on CHC composition comes from studies on adult so-cial insects and is focused on their roles in chemical communicationfor sex-, kin- and caste-recognition (Dallerac et al., 2000; Roelofset al., 2003; Liu et al., 2004; Châline et al., 2005; Nunes et al.,2008; Ferreira-Caliman et al., 2010).

We have been studying the expression of genes, proteins andenzymes in the honeybee integument as part of a major projectaiming to characterize the molecular elements involved in theexoskeleton formation and the regulation of this process (Bitondiet al., 1998; Santos et al., 2001; Zufelato et al., 2004; Soares et al.,2007, 2011, 2013; Elias-Neto et al., 2010). Such information gener-ated using polyacrylamide gel electrophoresis (SDS-PAGE), westernblot, RT-qPCR, gene sequencing, cDNA microarrays and fluores-cence in situ hybridization (FISH), mainly highlighted the differ-ential gene expression dynamics needed for the adult exoskeletonconstruction. These approaches, however, have neglected theoutmost functional exoskeletal layer, i.e., the envelope, its forma-tion and molecular composition. Trying to fill this gap in part wehere used GC/MS to explore the composition as well as develop-mental profiles of CHCs during exoskeleton formation and matu-ration. Concomitantly, genes potentially encoding desaturases andelongases were searched in the honeybee genome and theirexpression patterns were characterized using RT-qPCR. To get aninsight on the roles of the honeybee desaturases and elongasesgenes on cuticular envelope formation, the strength of the corre-lation between developmental profiles of transcripts and CHCswere estimated. In addition, we built molecular relationship treesfor gene function prediction.

2. Material and methods

2.1. Sample collection

Africanized Apis mellifera workers were obtained in the experi-mental apiary of the Faculty of Medicine, University of São Paulo inRibeirão Preto, SP, Brazil. The samples were collected at successivedevelopmental stages, from pupal ecdysis to a late adult stage, andincluded newly-ecdysed pupae (Pw phase: white eyes and unpig-mented cuticle); pupae-in-apolysis (Pp phase: pink eyes andunpigmented cuticle); early, intermediate and late brown-eyedpharate adults showing unpigmented cuticle (Pb phase), partiallypigmented cuticle (Pbm phase) and intensely pigmented cuticle(Pbd phase); newly-ecdysed adults (Ne) and foragers (Fg). These

Please cite this article in press as: Falcón, T., et al., Exoskeleton formationdesaturase and elongase genes during pupal and adult development, I10.1016/j.ibmb.2014.04.006

developmental time points include the sequential events of thepupal-to-adult molt and adult cuticle synthesis, deposition anddifferentiation/maturation. The pre-ecdysial phases were identifiedaccording Michelette and Soares (1993) and all these develop-mental phases were used to analyze CHCs profiles and theexpression of genes encoding enzymes potentially involved in theirbiosynthetic pathways.

2.2. CHCs extraction, identification and statistical analysis

The samples (36 bees per developmental phase, except for Pwphase that comprised 35 bees) were individually added to 1.5 ml of95% n-hexane (Mallinckrodt Chemicals) and bathed for 1 min toextract the CHCs. The extracts were then dried under N2 flow,resuspended in 160 ml of 95% n-hexane and analyzed in a GasChromatograph/Mass Spectrometer (Shimadzu GCMS modelQP2010), equipped with a 30 m DB-5MS column and helium as thecarrier gas (1 ml/min), using the electronic ionization (EI) method.The injection volume was 1 ml at an initial temperature of 150 �Celevated at a rate of 3 �C/min to 280 �C and keeping this temperaturefor 15 min. Compounds identification was based on their diagnosticions and in a standard solution containing different synthetic hy-drocarbons (Fluka). To analyze the chromatograms we used thesoftware GCMS solutions for Windows version 2.6 (Shimadzu Cor-poration). The positions of unsaturations in alkenes and alkadieneswere identified according to the dimethyl disulfide derivatizationtechnique (Carlson, 1989) and analyzed at the same GCMS systemabove mentioned. The initial temperature was 80 �C for 2 min, thenincreased to 180 �C at a rate of 30 �C/min and then to 300 �C at a rateof 3 �C/min, keeping 300 �C for 80 min. The identification of com-pounds and positions of unsaturations were done in splitless mode,which is recommended for low concentrated samples (Hü;bschmann, 2009), as those composed by a mixture of CHCs. Quan-tification of CHCs was based on their peak area in each chromato-gram (Singer and Espelie, 1992). We adjusted the compoundspercentage to 100%andeachpeak areawas transformed according tothe formula Z¼ ln[Ap/g(Ap)], where Ap is the peak area, g(Ap) is thegeometric mean of the peak for each bee sample and Z is the trans-formed peak area (Aitchison, 1982). To discriminate developmentalstages according to their CHCs profile, we performed a PerMANOVAtest in R software (version 3.0.0; package vegan: version 2.0e7). Thistest allowed us to analyze the dispersion of the samples around eachgroup (developmental phase) centroid giving us back a result basedon permutation tests (here we utilized 10000 permutations). Weplotted the results in a scatterplot with the two first Principal Co-ordinates as axes. The variation in the proportion of each compoundduring development was also verified using the One-Way ANOVAassociated to the Tukey’s Honestly Significance Difference (Tukey’sHSD) post hoc test (R software). We used this same test to comparevariations in the proportions of CHC groups: n-alkanes, methyl-alkanes, dimethyl-alkanes, alkenes, alkadienes and non-identifiedcompounds.

2.3. Identification of desaturases and elongases potentially involvedin hydrocarbons biosynthesis

Known desaturase and elongase amino acid sequences ofDrosophila melanogaster and lepidopterans, available in the Na-tional Center for Biotechnology Information (NCBI) data bank(www.ncbi.nlm.nih.gov/), were used to search for homologous se-quences in the honeybee genome (http://hymenopteragenome.org/beebase/) (version 4.5) using a BLASTP search tool (Altschulet al., 1990). The evolutionarily conserved functional motifs wereconfirmed in all the honeybee desaturase and elongase amino acidsequences, and the structural architecture of the respective

in Apis mellifera: Cuticular hydrocarbons profiles and expression ofnsect Biochemistry and Molecular Biology (2014), http://dx.doi.org/

http://www.ncbi.nlm.nih.gov/

http://hymenopteragenome.org/beebase/

http://hymenopteragenome.org/beebase/

T. Falcón et al. / Insect Biochemistry and Molecular Biology xxx (2014) 1e14 3

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 3/14

nucleotide sequences, including intron/exon boundaries, wascharacterized using Artemis 7.0 software (Rutherford et al., 2000).

2.4. RNA extraction and quantification

Honeybees at the Pw, Pp, Pbm, Ne and Fg developmental phases(3 bees per phase) were dissected in 0.9% NaCl solution, saving onlythe integument lining thorax and abdomen. Total RNA wasextracted from individual integuments using 1 ml TRIzol� (Invi-trogen) according to manufacturers’ instruction. After extraction,the RNA was suspended in autoclaved ultra-pure water treatedwith 0.1% diethylpyrocarbonate v/v (DEPC, Sigma). RNA sampleswere quantified in NanoDrop� ND-1000 (NanoDrop Technologies)and A260nm/A280nm wavelength ratios between 1.9 and 2.0 wereadopted as indicative of RNA quality and purity (Ausubel et al.,1995). The samples were diluted to a concentration of 0.6 mg/ml(considering that 1 OD unit measured at 260 nm corresponds to40 mg RNA per ml; Sambrook et al., 1989). To control for genomicDNA contamination the samples were incubated with DNAse (0.1unit RQ RNAse-free DNAse, Promega) for 30 min at 37 �C, followedby 15 min incubation at 70 �C to inactivate the enzyme.

2.5. First strand cDNA synthesis

First strand cDNA was synthesized from total RNA using theSuperScript� system (Invitrogen�, Life Technologies). The reversetranscription reaction was performed by incubating 0.6 mg of totalRNA with 1 ml Oligo(dT)12-18 (500 mg/ml) and 1 ml dNTP (10 mM) at65 �C for 5 min, and then adding 1 ml SuperScript� II ReverseTranscriptase 200U (Invitrogen) plus water to make 20 ml (finalvolume). This mixture was incubated at 42 �C for 50 min, and thenat 72 �C for 15 min to inactivate the enzyme. As negative controlsfor semi-quantitative RT-PCR and RT-qPCR analyses, the Super-Script� II enzyme was omitted in some reaction mixtures.

2.6. Semi-quantitative RT-PCR analysis

The PCR reactions were performed in a Thermo Cycler Veriti�

(Applied Biosystems). Each cDNA sample was used as template inreaction mixtures containing 10 ml of 2.5x Promega Master Mix[TaqeDNA Polymerase (0.06 unit/ml); 2.5x Taq-reaction buffer(125mMKCl, 75mMTriseHCl pH 8.4, 4mMMg2þ, 0.25%Nonidet�-P40); 500 ml of each dNTP]; 10 pmol of each primer, 1 ml of cDNA andautoclaved ultrapure water to make a final volume of 20 ml.

Specific primers were designed for six desaturase sequences andten elongase sequences in the honeybee genome using the soft-ware Primer3 (version 0.4.0) (http://frodo.wi.mit.edu/primer3/).The respective accession numbers, primer sequences and the ex-pected amplicon sizes are listed in Table S1. A primer pair (forward:50-TGC CAA CAC TGT CCT TTC TG-30 and reverse: 5 -AGA ATT GACCCA CCA ATC CA-3 ) was designed for the gene encoding a ribo-somal protein, AmRP49, currently renamed as ribosomal proteinL32 (RpL32) (GenBank-NCBI accession number NM_001011587.1),which is expressed in similar levels during the honeybee devel-opment, and was validated as being a suitable reference gene forPCR normalization (Lourenço et al., 2008). All primers weredesigned to span an intron, thereby serving as control for genomicDNA contamination. PCR conditions were: 5 min at 94 C, 30 cyclesof 30s at 94 �C, 30 s at 60 �C, 30 s at 72 �C and a final extension for7 min at 72 �C. PCR-amplified cDNA aliquots (20 ml) were mixedwith DNA loading buffer (0.25% bromophenol blue, 0.25% xylenecyanol FF and 30% glycerol) and analyzed by electrophoresis per-formed on GelRed� (Biotium) stained 1% agarose gel (0.1 ml/mlagarose gel) using 1x TBE (0.45 M Tris base, 0.45M boric acid, 0.5 MEDTA, pH 8.0) as running buffer. The electrophoresis was

Please cite this article in press as: Falcón, T., et al., Exoskeleton formationdesaturase and elongase genes during pupal and adult development, In10.1016/j.ibmb.2014.04.006

performed under constant voltage (100V). The gel was analyzed inan image scanner KODAK EDAS 290. A 100 bp lDNA/Hind III(Invitrogen�, Life Technologies) was used as molecular weightmarker.

2.7. RT-qPCR analysis

RT-qPCR analysis was performed in a 7500 Real Time PCR Sys-tem (Applied Biosystems) using primers for desaturase and elon-gase genes (Table S1) and for the reference gene. Standard curveswere performed for each primer to verify amplification efficiency(parameters used: �3.6 � Slope � �3.3; R2 � 0.99). Reactionmixtures were prepared with 10 ml of 2x Sybr Green PCR MasterMix (Applied Biosystems), 0.8 ml of each specific primer of a pair,and 2 ml of cDNA in a final volume of 20 ml completed with DEPCtreated-autoclaved milli-Q water. The amplification program was:2 min at 52 �C and 10 min at 95 �C; 40 cycles of 1 min at 94 �C, 30 sat 60 �C and 30 s at 72 �C. Dissociation curve for each primer pairallowed verifying the quality of amplification. Triplicates wereprepared for each sample (technical triplicates), and three inde-pendent samples were used for each developmental phase (bio-logical samples). The relative quantification was calculated using2�DDCt (Schmittgen and Livak, 2008). The transcript levels werecompared between phases using One-Way ANOVA and Tukey’sHSD post hoc test (p < 0.05) (R software).

Desaturase and elongase transcripts profiles were correlatedwith CHCs profiles using Spearman test and R software (p < 0.05).

2.8. Amplicons validation

PCR amplicons were sequenced using the primers cited above.After amplification for 5 min at 94

C, 40 cycles of 30 s at 94 �C, 30 sat 60 �C, 30 s at 72 �C and a final extension for 7 min at 72 �C, thecDNA was quantified in NanoDrop and 1 ml was mixed with 1 ml offorward primer (F) or reverse primer (R), 2 ml of Big Dye� Termi-nator solution v3.1 Cycle Sequencing Kit (Perkin Elmer), 2 ml ofbuffer solution (5x Sequencing Buffer) and 4 ml of ultrapure auto-clavedwater, totalizing 10 ml of reaction solution.We used 25 cyclesof 10 s at 96 �C, 5 s at 50 �C and 4 min at 60 �C. The samples wereprecipitated with 40 ml of 75% isopropanol, incubated at roomtemperature for 15 min and centrifuged at 12,100 � g during20 min. The supernatant was discarded, the pellet was dried at55 �C, resuspended in 12 ml of Hi-DiTM formamide and transferredto sequencing tubes to be submitted to a thermal shock (2 min and5 s, at 95 �C, followed by 5 min on ice). Samples were analyzed bythe method of Sanger in an automatic sequencer ABI Prism 310(Applied Biosystems). We also confirmed the sequences through insilico PCR analysis (http://genome.ucsc.edu).

2.9. Molecular relationship analysis of desaturases and elongases

Amino acids sequences of desaturases and elongases from ar-thropods were searched using the same approach described in thesection “2.3.”. Only complete amino acid sequences were used forthe analysis, except for the desaturase sequences corresponding tothe GB42217 and GB48194 accession numbers. The multiplealignments were performed using the software MAFFT v. 7 (Katohet al., 2002) (http://mafft.cbrc.jp/alignment/server/) with defaultparameters. This software has a better performance than othersoftwares when analyzing a great number of terminals (>50)(Katoh et al., 2005). These alignments were visualized using thesoftware CLustalX 2.1 (Larkin et al., 2007) to identify the enzymesmotifs. The arthropod species used in the analysis as well as thedesaturases and elongases accession numbers at NCBI are listed inTable S2.


http://frodo.wi.mit.edu/primer3/

http://genome.ucsc.edu

http://mafft.cbrc.jp/alignment/server/

Fig. 1. Cuticular hydrocarbons (CHCs) profiles during A. mellifera development. Compounds were extracted with n-hexane from the cuticle of pupae (Pw and Pp phases), developingadults (Pb, Pbm and Pbd phases), newly-ecdysed adults (Ne) and forager (Fg) bees, and identified by GC/MS.


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 4/14

Please cite this article in press as: Falcón, T., et al., Exoskeleton formation in Apis mellifera: Cuticular hydrocarbons profiles and expression ofdesaturase and elongase genes during pupal and adult development, Insect Biochemistry and Molecular Biology (2014), http://dx.doi.org/10.1016/j.ibmb.2014.04.006

Fig. 1. (continued).


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 5/14

We performed a model test for both desaturase and elongasesequencesusing the softwareProtTest v. 3.2 (Darribaet al., 2011)withdefault parameters. The indicated model for both enzymes was LG(Le and Gascuel, 2008), which represents a standard matrix of sub-stitution of amino acids, with an estimative of invariable sites (þI)and a gamma heterogeneity rate between sites (þG) for both groupsof enzymes. The eLnL (log likelihood) values were �39,255 fordesaturases and�41,945 for elongases. For both groups of enzymes,we used the software PhyML, version 3.0 (Guindon et al., 2010).

The molecular phylogeny of desaturases was performed usingmaximum likelihood analysis (Felsenstein, 1981) with 500 boot-strap replications. Desaturase protein sequences of the crustaceanDaphnia pulex and the tick Amblyomma americanum (Liu et al.,1999; Roelofs and Rooney, 2003) were used as external groups.The tick desaturase sequence was used for rooting the tree, onceunlike D. pulex it has only one desaturase sequence. We also usedBLASTP (Altschul et al., 1990) to compare each sequence with thefollowing desaturase sequences whose functions were alreadyvalidated: D. melanogaster desat1 (NP_731710), a D9 desaturasethat uses preferentially palmitic acid (C16) over stearic acid (C18)(16 > 18) as substrate; D. melanogaster desat2 (NP_650201), a D9desaturase that prefers myristic acid (C14) and synthesizes D5 hy-drocarbons (Dallerac et al., 2000); D. melanogaster desatF(NP_651996), a desaturase that uses the u-7 product from desat1


unsaturation as substrate (Chertemps et al., 2006); Ostrinia furna-calis D14 desaturase (AAL35746); Helicoverpa zea D9 desaturase(AAF81790) that acts on stearic acid (18 > 16) (reviewed in Roelofsand Rooney, 2003); Trichoplusia ni D9 desaturase (AAB92583) witha high specificity for stearic acid (18 > 16) (Liu et al., 1999), and D11desaturase (AAD03775) (Knipple et al., 1998); Choristoneura rosa-ceana D9 desaturase (AAN39697) that acts on palmitic acid(16 > 18) (Hao et al., 2002); Danaus plexippus D14 desaturase(EHJ69993); Bombyx mori D11 desaturase (NP_001040141); Lamp-ronia capitella D14-26 desaturase (ABX71629); Choristoneuraparalela D14-26 desaturase (AAQ12887); Helicoverpa assulta D14-26 desaturase (AF482905) (Liénard et al., 2008).

For the molecular phylogeny of elongases we performed acluster analysis using an approximate likelihood ratio test (aLRT)(Anisimova and Gascuel, 2006). The generated trees were visual-ized and edited using the software FigTree v.1.3.1 (http://tree.bio.ed.ac.uk/software/figtree/).

3. Results

3.1. CHC profiles during cuticle formation and maturation

Five CHC classes including n-alkanes, methyl-alkanes, dimethyl-alkanes, alkenes and alkadienes, with chain-length of the carbon


http://tree.bio.ed.ac.uk/software/figtree/

http://tree.bio.ed.ac.uk/software/figtree/


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115

IB2576_proof ■ 10 May 2014 ■ 6/14

backbones ranging from C19 to C35, were identified by GC/MS in thecuticle of newly-ecdysed pupae (Pw phase), pupae-in-apolysis (Ppphase) and in the cuticle of developing adults (Pb, Pbm and Pbdphases), newly-ecdysed adults (Ne) and mature bees (Fg). Indi-vidually, the five peaks that better explain the total variation herefound were Z-?- C35 (3.754%)/Z-7-C23, Z-9-C23 (3.746%)/ZZ-?-C33a(3.551%)/Z-10-C35, Z-12-C35, Z-14-C35, Z-15-C35 (3.223%)/5-MeC25(2.995%). Fig.1 shows the variation in the proportions of the specificcompounds in each CHC class during this developmental period(see mean values and statistics in Table S3). The most representedn-alkane in the developing and adult cuticle is C27. C27 is propor-tionally more abundant in foragers, and C25 is the second moreabundant n-alkane in foragers. C24 and C26, which are less repre-sented in the cuticle, also showed increased percentages in for-agers. The other n-alkanes showed diverse developmental profiles:C19, C21, and C23 proportions increased significantly at the Pbm orPbd phases and then decreased in foragers. C28, C29 and C30 wererelatively more abundant in the cuticle of the immature develop-mental phases and newly-ecdysed adults than in foragers. C31showed similar proportions in the cuticle of some immature phasesand foragers (Fig. 1A).

The most represented methyl-alkanes in the developing cuticlewere 9-;11-;13-MeC27, 9-;11-; 13-;15-MeC29 and 13-;15-MeC31.Except for 7-;9-MeC33, the other sixteen identified methyl-alkanes(94%) were proportionally more concentrated in the immature cu-ticles (often including the cuticle of the newly-ecdysed adults) thanin the cuticle of foragers (Fig. 1B). This developmental pattern isrepeated for the dimethyl-alkanes, all of them (100%) presentingrelatively higher quantities in the immature cuticle (Fig. 1C). There-fore, the great majority of the methyl-alkanes and all identifieddimethyl-alkanes lowered to basal levels in the cuticle of foragers. Incontrast to these branched compounds (and similarly to the C24, C25,C26 and C27 n-alkanes), most of the alkenes (Fig. 1D) and alkadienes(Fig. 1E) were overrepresented in the cuticle of foragers. Althoughtwoof the alkenes (Z-7-C23, Z-9-C23) haspeaked in the Pbdphase andanother peak of alkenes (Z-8-C28, Z-9-C28) has decreased in foragers,and one of the alkadienes (ZZ-?-C27) showed the higher concentra-tion in thePbd phase andnewly-ecdysed adults, our data confidentlyshow that in general, the proportion of both unsaturated CHCs, al-kenes and alkadienes, increases in the mature cuticle of foragers.

Table 1 summarizes the variation in the proportions of n-al-kanes, methyl-alkanes, dimethyl-alkanes, alkenes and alkadienesin the cuticle during development. Foragers clearly differentiatedfrom immature bees and newly-ecdysed adults regarding pro-portions of CHC classes. The overrepresented CHC class in thecuticle of foragers is the class of n-alkanes, which makes up 71.92%of the total CHCs. It is followed by the alkenes, representing 22.28%of the CHCs. The immature cuticles of the developing bees andnewly-ecdysed adults also contain a relatively high proportion of n-alkanes, comprising 44.88e57.43%, but the methyl-alkanes are also

Table 1Proportions of CHC chemical classes during adult cuticle formation and maturation.

Proportions of CHC groups (%)

Phase Saturated CHCs Unsaturated CHCs Non-identified

n-Alkanes Methyl-alkanes

Dimethyl-alkanes

Alkenes Alkadienes

Pw 47.82 44.10 3.87 3.86 0.01 0.34Pp 50.87 43.42 2.66 2.60 0.01 0.44Pb 44.88 47.47 3.72 3.34 0.01 0.57Pbm 45.20 46.57 3.28 4.47 0.01 0.47Pbd 57.44 35.15 1.99 4.97 0.19 0.26Ne 51.53 38.19 2.80 6.39 0.60 0.49Fg 71.93 2.59 0.11 22.28 2.60 0.49

116117118119120121122123124125126127128129130


abundant (35.15e47.47%) in these cuticles. It was evident thatmethyl-alkanes and dimethyl alkanes exist in significant higheramounts in the immature cuticle in comparison to foragers. Incontrast, alkenes and alkadienes are both proportionally moreabundant in the cuticle of foragers than in the cuticle of the earlierdevelopmental phases.

Although qualitative variations in CHC profiles during devel-opment have not been detected, all developmental phases quan-titatively differed from each other (PerMANOVA test, Pseudo-F ¼ 214.64. p < 1e-05) (Fig. 2). This analysis, which included all theidentified and non-identified compounds, revealed significant dif-ferences across all developmental phases and clearly separatedthem in three groups according to the relative quantities of CHCs:one group including pupae (Pw and Pp phases) and the successivePb and Pbm phases, a second group formed by the pre- and post-ecdysial phases (Pbd and Ne), and a third group formed exclu-sively by foragers (Fg). Thus, the obtained data indicate that thepupal and adult (immature and mature) cuticles share a similarCHC blend, but with great differences in the relative quantities ofthe majority of the compounds. All CHC chemical classes wereidentified by GC/MS in the developing and adult cuticles and wererepresented in the principal component analysis (Table S3). Anexample of the chromatograms is shown in Fig. S1.

3.2. Molecular phylogeny of desaturases and elongases

All seven desaturases included in the phylogenetic analysis aremembers of the subfamily First Desaturase (Hashimoto et al.,2008). The deduced amino acid sequences of all desaturasesherein characterized contain the eight conserved histidine residuesarranged as three box domains, which are typical of integralmembrane desaturases and are essential for the catalytic activityand function (Table S2 and Fig. S2). The desaturases grouped ac-cording to their functions (position where the unsaturation iscreated in the fatty acid) (Fig. 3). Six of them (GB42218, GB51236,GB51238, GB48194, GB42217, GB48195) displayed the highestsimilarity with D9-desaturases (16 > 18) of other organisms, whichcatalyses the insertion of a double bond at the 9th position fromthecarboxyl group of a fatty acid. The desaturase encoded by thegene GB48193 clustered with D4-desaturases creating a carbon/carbon double bond at the 4th position.

The honeybee elongase sequences have the conserved domainformed by three histidine residues and clustered in two groups(Fig. 4, Table S2 and Fig. S2) with different functional motifs. Elevenelongase sequences were included in the phylogenetic analysis.Three of them (GB53873, GB53872 and GB45596) belong to the S/MUFA cluster and should have roles in elongating Saturated/Monounsaturated Fatty Acids. Eight elongases (GB51249, GB51247,GB54401, GB54302, GB54396, GB54399, GB40681 and GB46038)clustered into the PUFA class, which has roles in PolyunsaturatedFatty Acids elongation. Interestingly, except for the PUFA elongasegenes GB54399 and GB40681, all the other PUFA genes withexpression quantified by RT-qPCR showed a positive correlationwith the alkadienes.

3.3. Structure and expression profiles of desaturase and elongasegenes in the integument during cuticle formation and maturation

Six of the seven desaturase sequences and ten of the elevenelongase sequences included in the phylogenetic analyses werecharacterized in terms of gene structure and expression in theintegument. The six desaturase coding sequences (CDSs) have 663to 1371 nucleotides (stop codon included) and encompass 5 to 8exons that potentially encode 220 to 456 amino acid residues. Theten elongase CDSs contain 420 to 996 nucleotides distributed into 2


Fig. 2. Principal coordinate analysis of CHC profiles discriminating developmental phases of A. mellifera workers. Axes are represented by principal coordinates. PerMANOVA test(Pseudo-F ¼ 214.64, p < 1e-05). The centroid is marked in red for each developmental phase. Dashed lines highlight link the more distant samples in each group. Blue lines linksamples to centroid. Developmental phases are represented by symbols: pupae (Pw V); pupae-in-apolysis (Pp e >); developing adult phases: Pb (D), Pbm (x) and Pbd (þ); newly-ecdysed adult: Ne (⌧); forager bee: Fg (B). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 7/14

to 7 exons that encode proteins formed by 139e331 amino acids.Expected canonical splice sites (GT/AG) were found at all exon/intron boundaries. The interprimer regions (amplicons) weresequenced and validated (Fig. S3).

Some of the honeybee desaturase and elongase genes are orga-nized in tandemin the chromosomes. ThedesaturasegenesGB42217and GB42218, both encoding amino acid sequences that groupedtogether with D9-desaturases in the phylogenetic analysis (Fig. 3),are clustered on chromosome 1. Six of the elongase genes are ar-ranged into two clusters on chromosome 16, one cluster includingthe genes GB54401, GB54302, GB54399 and GB54396, and the otherformed by theGB51249 andGB51247, all of themencodingelongasesof the PUFA class as suggested by the phylogenetic analysis (Fig. 4).

Four D9-desaturase genes (GB51238, GB48195, GB42218 andGB42217) and the D4-desaturase gene (GB48193) showed signifi-cant transcript levels increase in the integument of newly-ecdysed(Ne) and/or forager (Fg) bees in comparison to the earlier devel-opmental phases (Pw, Pp and Pbm phases). The desaturase geneidentified as GB51236 (D9 desaturase) was the only gene present-ing similar levels of transcripts during development (Fig. 5).

Ten of the genes potentially encoding elongases showed vari-able expression profiles. Genes encoding elongases of the PUFAclass (GB51249, GB51247 and GB54396) and genes encoding S/MUFAs (GB53872 and GB45596) showed higher expression in thenewly-ecdysed adults and/or forager bees than in the earlierdevelopmental phases. Two other elongase genes in the PUFA classalso showed significantly higher expression in foragers but onlywhen compared with the Pp phase (gene GB54401) or Pbm phase(gene GB46038). The gene GB40681potencially encoding an elon-gase of the PUFA class showed a different expression profile with ahigher level of transcripts in the Pbm phase than in the earlier orlater phases. The expression profiles of the other PUFA elongasegenes, GB54399 and GB53873, did not show significant variation inthe developing integument (Fig. 6).

3.4. Desaturase and elongase transcript levels variation versusCHCs profiles during cuticle formation and maturation

Sperman’s correlation coefficient was used to measure theinterdependence between the changing levels of CHCs and desa-turase or elongase transcripts during cuticle development (Fig. 7;


Table S4). In general, we found significant positive correlations be-tween the increase in desaturase transcript levels during develop-ment and the increase in the proportions of 13 alkenes peaks (Z-7-C23, Z-9-C23/Z-?-C25/Z-6-C27, Z-7-C27, Z-9-C27/Z-?-C27/Z-7-C29, Z-9-C29/Z-?-C29/Z-5-C25, Z-7-C25, Z-9-C25/Z-7-C31, Z-8-C31, Z-9-C31, Z-10-C31, Z-11-C31, Z-15-C31/Z-8-C33, Z-9-C33, Z-10-C33, Z-12-C33/Z-9-C33,Z-10-C33, Z-13-C33/Z-?-C33/Z-10-C35, Z-12-C35, Z-14-C35, Z-15-C35/Z-?-C35) and 6 peaks of alkadienes (ZZ-?-C27/ZZ-?-C31/ZZ-?-C33a/ZZ-?-C33b/ZZ-?-C35a/ZZ-?-C35b), representing respectively 93% and 100%of the total of the identified alkenes and alkadienes. Therefore, theincrease in the products of desaturation reactions is possibly due toincreased transcription of desaturase genes. These correlation dataallow us to infer that the six desaturase genes, or rather, theirrespective protein products, promote an increase in the biosynthesisof alkenes and alkadienes in adult bees by the insertion of doublebonds into the fatty acid (hydrocarbon) backbone chains.

Interestingly, the expression of five of the desaturase genes(GB42218, GB42217, GB48195, GB51238 and GB48193), with a fewexceptions, were positively correlated with the same pool of un-saturated CHCs, alkenes (Z-10-C35, Z-12-C35, Z-14-C35, Z-15-C35/Z-?-C29/Z-?-C35/Z-?-C33/Z-5- C25, Z-7-C25, Z-9-C25/Z-9-C33, Z-10-C33,Z-13-C33/Z-7-C23, Z-9-C23/Z-8-C33, Z-9-C33, Z-10-C33, Z-12-C33/Z-7-C31, Z-8-C31, Z-9-C31, Z-10-C31, Z-11-C31, Z-15-C31/Z-?-C25) andalkadienes (ZZ-?-C33a/ZZ-?-C33b/ZZ-?-C31/ZZ-?-C35a/ZZ-?-C35b) (seethe dark blue area at the left of Fig. 7). The other desaturase gene,GB51236, which in contrast to the five desaturase genes abovementioned did not show a significant increase in expression inforagers, exhibited a more restrictive correlation pattern. Its tran-scriptional modulation over time was positively correlated onlywith some of these alkenes (Z-8-C33, Z-9-C33, Z-10-C33, Z-12-C33/Z-?-C25) and alkadienes (ZZ-?-C31/ZZ-?-C33b/ZZ-?-C35a/ZZ-?C35b) andwith other peaks of alkenes (Z-?-C27/Z-7-C29, Z-9-C29/Z-6-C27, Z-7-C27, Z-9-C27) (Fig. 7). Such differences in correlation patterns mayreflect enzyme specificities in desaturation reactions.

Unexpectedly, significant positive correlation was also found be-tween the levels of four of the desaturase transcripts (GB42217,GB48195, GB51238, GB48193) and some alkanes (C23/C24/C25/C31),and methyl-alkanes (7-MeC33/9-MeC33). Except for the alkane C25,the expression of the desaturase gene GB42218 was also positivelycorrelated with the levels of these same alkanes andmethyl-alkanes,and the expression of the desaturase gene GB51236 was positively


Fig. 3. Evolutionary relationships of insect desaturases. Different clusters are markedin different colors with their putative function. Branches: Hymenoptera (blue); Diptera(red); Lepidoptera (green); Coleoptera (yellow); Crustacea (purple); Chelicerata (or-ange). The A. mellifera enzymes are written in blue. The horizontal bar indicates thescale of amino acids substitutions in branches. The values in the nodes are the branchsupport after 500 bootstrap replications. The first letter in each enzyme name indicatesthe gender and the following three letters indicate the species, followed by the


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125

IB2576_proof ■ 10 May 2014 ■ 8/14


correlatedwithC27andC26 levels (Fig. 7). All thesealkanescorrespondto a fraction (26%) of the total alkanes identified, and the changinglevels of the majority of them showed no correlation, or showednegative correlation, with desaturase transcripts levels modulation.

As most of the desaturase genes, the expression of five of theelongase genes (GB53872, GB54396, GB51249, GB45596 andGB51247) also increased significantly in newly-ecdysed adults and/or foragers. Consequently, the correlation pattern of these elongasegenes mimicked the correlation pattern of the desaturase genes.The expression profiles of these elongase genes, and also theexpression profiles of two other elongases genes, GB53873 andGB54401, showed a positive correlation with the same CHCs asspecified above, mainly alkenes and alkadienes, except Z-?-C27 (seethe dark blue area at the left of Fig. 7). Such compounds representrespectively 86% and 100% of the total of alkenes and alkadienesherein identified. This suggests that these seven elongase genes actsynchronically with the five desaturase genes above mentioned inthe biosynthesis of these unsaturated CHCs.

The transcript profiles of the other three elongase genes werepositively correlated with the profiles of other types of hydrocar-bons. Thus, the expression of GB54399 was correlated with themethyl–alkanes 9-;11-;13-;15-MeC29, 5-MeC25 and 7-MeC25, andthe expression of GB40681 was correlated with the dimethyl-alkanes alkanes 3,7-diMeC31, 9,13-diMeC29, 11,15-diMeC27 and thealkenes Z-7-C31, Z-8-C31, Z-9-C31, Z-10-C31, Z-11-C31, Z-15-C31. Theelongase gene GB46038 showed a more extensive correlationpattern including alkanes (C27/C26/C25), alkenes (Z-7-C29, Z-9-C29/Z-6-C27, Z-7-C27, Z-9-C27/Z-8-C33, Z-9-C33, Z-10-C33, Z-12-C33/Z-7-C31,Z-8-C31, Z-9-C31, Z-10-C31, Z-11-C31, Z-15-C31) and alkadienes (ZZ-?-C31/ZZ-?-C33b/ZZ-?-C35a/ZZ-?-C35b). Interestingly, the develop-mental profiles of some compounds were specifically correlatedwith the changing expression of a single elongase gene. Forexample, the methyl-alkanes 9-;11-;13-MeC23, 9-;11-;13-MeC25, 7-MeC25, 5-MeC25 and 9-;11-;13-; 15-MeC29 showed a high positivecorrelation exclusively with the expression of the GB54399 elon-gase gene. Similarly, the dimethyl-alkanes 11,15-diMeC27, 9,13-diMeC29 and 3,7-diMeC31 were positively correlated exclusivelywith the GB40681 elongase gene (see Fig. 7). This may indicate thatthe enzymes encoded by these genes act preferentially on thebiosynthesis of methyl-branched CHCs.

The developmental profiles of a fraction of alkanes (C30/C28/C29),methyl- or dimethyl alkanes (13- MeC31,15- MeC31,17-MeC31/3-MeC31,5-MeC31,7-MeC31,10-MeC31,12-MeC31,13-MeC31,15-MeC31/9-MeC31,12-MeC31,14-MeC31/5-MeC33,11-MeC33,13-MeC33,14-MeC33,15-MeC33,16-MeC33/13-MeC35,15-MeC35,16-MeC35,17-MeC35/10,15-diMeC29/11-MeC26,12-MeC26/7,11diMeC27, 7,15diMeC27/6,10-diMeC27, 11,15-diMeC27/4,8-diMeC27, 13,15-diMeC27/7-MeC27/5-MeC27/9- MeC27,11- MeC27,13MeC27/12- MeC28,13- MeC28,14-MeC28) (see green area in Fig. 7) in general showed negative cor-relation with the expression of all the elongase genes hereincharacterized. This suggests that the products of these genes do nothave roles in the biosynthesis of these specific saturated CHCs.

4. Discussion

4.1. Overrepresented CHC classes in the immature and maturecuticle

We characterized the CHC profiles of honeybee workers duringan entire molting cycle, from the pupal ecdysis to the adult ecdysis,

accession number of their respective amino acid sequences (see Table S2). (For inter-pretation of the references to color in this figure legend, the reader is referred to theweb version of this article.)

126127128129130


Fig. 4. Evolutionary relationships of insect elongases. Enzymes were clustered basedon their functional domains: S/MUFAs (blue area) and PUFAs (green area). Branches:Hymenoptera (blue); Diptera (red); Lepidoptera (green); Coleoptera (yellow);


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122

IB2576_proof ■ 10 May 2014 ■ 9/14


and also in mature adults. The genuine honeybee worker pupa,represented by the Pw phase, exists during a relatively short period,lasting approximately 40 h from the pupal ecdysis. Apolysis, i.e., thedetachment of the pupal cuticle occurs at the Pp phase and signalsthe end of the pupal phase and the onset of adult cuticle depositionthroughout the Pb to Pbd phases. The developing adult cuticlebecomes increasingly thicker, and this is followed by the ecdysisand emergence of the adult bee from the comb cell. Such sequentialevents characterize the molting cycle that transforms the pupa intoan adult bee.

CHC profiles determination through GC/MS allowed us todiscriminate the cuticles of all these molting cycle-phases and alsothe mature cuticle of forager bees. The Principal Coordinate Anal-ysis clustered these developmental time points into three groups:one including the pupal cuticle (Pw and Pp phases) and the cuticleof the subsequent Pb and Pbm phases, one other formed by the pre-and post-ecdysial phases (Pbd phase and the newly-ecdysedadults), and the other exclusively formed by mature foragers. Thismeans that these groups significantly differ from each other in therelative quantity of CHCs. Variations in the relative quantities of anumber of CHCs between the pupal cuticle (Pw and Pp phases) andthe cuticle of the subsequent Pb and Pbm phases may be inter-preted as being due to the incorporation of hydrocarbons into thegrowing (nascent) adult cuticle.

Our analysis showed that the mature cuticle of the forager beesis by far the most dissimilar in CHC proportions. These results goalong with the hypothesis that, in social bees, the cuticle is fullymature only later in the adult life when they start foraging activ-ities. This hypothesis is based on comparative morphological studythat revealed striking differences in the cuticular maturation, andthus in cuticle properties, of workers of social and solitary beespecies. In contrast to the solitary bee species that emerge from thenest with a heavily pigmented/sclerotized cuticle and immediatelystarts foraging, the social species showed the higher degree ofpigmentation/sclerotization only several days after the emergencewhen they leave the intranidal activities for foraging for nectar andpollen (Elias-Neto et al., 2013). In this context, the next step wouldconsist in comparing the CHCs profiles of social and solitary bees atdefinite time points of the adult life, which certainly is an inter-esting way of testing this hypothesis.

Approximately half of the identified n-alkanes, and almost all ofthe identified unsaturated CHCs (alkenes and alkadienes) showedhigher proportions in the mature cuticle of forager bees than in theearlier developmental phases. The opposite was verified for thesaturated, branched CHCs (methyl-alkanes and dimethyl-alkanes),which were less represented in the cuticle of foragers, but over-represented in the immature cuticles of pupae (Pw and Pp phases),developing adults (Pb, Pbm and Pbd phases) and newly-ecdysedadults. Such differences seem due to the intensification of thebiosynthesis and/or deposition of n-alkanes, alkenes and alka-dienes on the cuticle of foragers, since 40% of the total of theidentified compounds (n ¼ 55) were proportionally more concen-trated in foragers than in the earlier developmental phases. Thehigher relative quantities of methyl-alkanes and dimethyl-alkanesin the immature cuticles suggest that branched compounds areimportant for the structure of the envelope at these stages, or forthe communication between brood and worker bees inside the

Crustacea (purple); Chelicerata (orange). A. mellifera elongases are written in blue. Thehorizontal bar indicates the scale of amino acids substitutions in branches. The valuesin nodes are the aLRT support. The first letter in each enzyme name indicates thegender and the following three letters indicate the species, followed by the accessionnumber of their respective amino acid sequences (see Table S2). (For interpretation ofthe references to color in this figure legend, the reader is referred to the web version ofthis article.) Q4

123124125126127128129130


Fig. 5. Expression of desaturase genes in the pupal integument (Pw and Pp phases) and in the integument of developing adults (Pbm phase), newly-ecdysed adult (Ne) and foragerbees (Fg). Transcript levels (columns) were determined by RT-qPCR. Different letters above each error bar indicate significant statistical difference (ANOVA associated to post hocTukey’s HSD test, p < 0.05) between developmental phases. The genes are identified by their accession numbers in the honeybee genome (version 4.5).


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

666768697071727374757677787980818283848586878889909192

IB2576_proof ■ 10 May 2014 ■ 10/14

hive. Interestingly, new types of compounds were not identified asthe cuticle became mature indicating that the blends of CHCs (butnot the proportions of individual CHCs) are similar in the pupal andadult cuticles. Therefore, the CHC classes and types present in thepupal cuticle are also present in the growing and mature adult

Fig. 6. Expression of elongase genes in the pupal integument (Pw and Pp phases) and in the(Fg). Transcript levels (columns) were determined by RT-qPCR. Different letters above each eHSD test, p < 0.05) between developmental phases. The genes are identified by their acces


cuticles, but show clear differences in relative quantities. The re-sults of our CHC analysis in the newly-ecdysed adults and foragersare comparable with the results of Blomquist et al. (1980a). Theseauthors similarly found a higher proportion of saturated than un-saturated compounds and an increase in the proportions of

integument of developing adults (Pbm phase), newly-ecdysed adults (Ne) and foragersrror bar indicate significant statistical difference (ANOVA associated to post hoc Tukey’ssion numbers in the honeybee genome (version 4.5).

93949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130


Fig. 7. Correlation between the developmental profiles of desaturases and elongases transcript levels and the relative proportions of saturated and unsaturated CHCs during adultcuticle formation and maturation (Spearman test, p < 0.05).


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 11/14

pentacosane (C25), heptacosane (C27), pentacosenes (C25:1) andheptacosenes (C27:1) in older bees (26 day-old), possibly foragers, incomparison to the younger bees (7-day old) used in their study. LikeBlomquist et al. (1980a, 1980b) we detected 23 to 35 carbons chain-length alkenes, and 31, 33 and 35 carbons chain-length alkadienes,but another alkadiene (27 carbons chain-length) was also detectedin our analysis of the adult cuticle. Concerning some discrepanciesbetween both data, they could be due to the use of adult bees ofdifferent ages and subspecies: our data were obtained with Afri-canized A. mellifera, the hybrid produced by cross-breeding of theAfrican A. m. scutellata and European honeybees, whereasBlomquist et al. (1980a, b) worked with the Italian race.

Our results on the adult stage also match those of Kather et al.(2011) in demonstrating that the pool of CHCs in forager honey-bees is mainly composed of n-alkanes and that these compoundsand alkenes have higher proportions in foragers than in newly-ecdysed adults. In addition to their structural and physiological


functions in the cuticular envelope, the alkenes may be importantfor nestmate recognition (Châline et al., 2005; Dani et al., 2005;Kather et al., 2011) whereas n-alkanes are less informative in thisaspect of insect sociality (Krasnec and Breed, 2013).

To our knowledge only two previous studies have investigatedCHCs in immature stages of the honeybee, and even so, usingpunctual stages. Nation et al. (1992) and Martin et al. (2001) usedGC/MS to investigate the CHC mimicry of honeybee host by theectoparasitic mite Varroa jacobsoni. At a stage identified as “purple-eye pupa”, apparently equivalent to our pupa-in-apolysis (Ppphase), Nation et al. (1992) found relatively low proportions of al-kanes in comparisonwith the adults, these having significantly lessbranched alkanes than the pupa. This was confirmed by our anal-ysis. Alkenes were found in proportionally higher quantities in theadult cuticle. This was also evident in our analysis, which addi-tionally showed a significant increase in the proportions of alka-dienes on the adult cuticle. Martin et al. (2001) used newly-ecdysed



1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 12/14

adult bees, five-day-old larvae and a stage identified as “eight-day-old pupae”, which seems equivalent to our Pbm phase, in acomparative analysis with CHCs from the mite. Like our findings,these analyses did not reveal qualitative differences, but quantita-tive differences in the CHC pools of developing honeybees.

CHC developmental profiles were also characterized in otherinsect species. Similarly to our findings, the n-alkanes detected inpupae of T. ni were in general also detected in the 3-day-old adultfemales, and the proportions of these compounds were higher inthe adult moths. However, differently from the honeybees, thebranched CHCs of T. ni showed qualitative variation during devel-opment (de Renobales and Blomquist, 1983). There was also noqualitative variation in the CHC profiles in larvae, pupae and adultsof Sarcophaga bullata, but quantitative variations (Armold andReignier, 1975). Taken together, these few studies on characteriza-tion of CHC profiles revealed remarkable differences in the relativequantities of these compounds during development.

There is evidence in T. ni that biosynthesis and transport ofhydrocarbons to the cuticular surface are highest during thefeeding periods of the successive larval instars. As the larva pre-pares for each molt and stops feeding, hydrocarbon biosynthesisand transport decrease at different rates, the decrease in transportbeing even greater, such as the internal hydrocarbons accumulateand are then used as a source of CHC for the next larval instar(Dwyer et al., 1986). During the pupal-to-adult molt of T. ni, hy-drocarbons biosynthesis was limited to two specific time periods:soon after the pupal molt and preceding the adult ecdysis. Theproduced hydrocarbons remained stored internally up to the adultecdysis and thenwere deposited on the adult cuticle (de Renobaleset al., 1988). Similar results were obtained in another lepidopteran,Spodoptera eridania (Guo and Blomquist, 1991) and in the dipteranS. bullata (Armold and Regnier, 1975). Even in hemimetaboloussuch as Blattella germanica, the patterns of hydrocarbons biosyn-thesis were correlated to molting and food intake during nymphalstages (Young and Schal, 1997). It follows that the deposition ofCHCs does not reflect the rate of biosynthesis, sincemost of the CHCpool in a given stage were produced in the previous stage. Thesedata on the dynamics of biosynthesis and deposition of CHCs allowus to infer that the CHCs integrating the pupal cuticle (Pw and Ppphases) and the cuticle of the subsequent developmental phases(Pb, Pbl and Pbm phases) of the honeybee are remnant of larvalsynthesis. The CHCs detected at the time of adult ecdysis, i.e., at thePbd and Ne phases, may result from post-pupal molt biosynthesis.The increase in the proportions of several CHCs in forager bees isthen possibly due to feeding and de novo biosynthesis at the ex-penses of precursors derived from the metabolism of dietarycompounds. This idea receives support from the expression profilesof several desaturase and elongase genes, which showed increasedtranscript levels in forager bees in comparison to the immaturephases.

4.2. Evolution of desaturases and elongases in Hymenoptera

Genes encoding proteins/enzymes using fatty acid as substratesgenerally show duplications and have accumulated mutations, andthis may imply in new functions (Karanth et al., 2009). The desat1and desat2 desaturase genes of D. melanogaster are tandemlyarrayed in a region of the chromosome 3 and could be evolved froma single gene through an initial duplication (Wicker-Thomas et al.,1997). The in tandem chromosomal organization of some of thehoneybee desaturase and elongase genes supports the occurrenceof duplication events during evolution.

Four genes putatively encoding desaturases belonging to theFirst Desaturase subfamily, with higher affinity for stearic acid(C18:0) to form oleic acid (C18:1), were previously characterized in


A. mellifera (Hashimoto et al., 2008). In the current version of thehoneybee genome (version 4.5) we could detect at least sevenmembers of this subfamily. Computational and phylogenetic dataincluding desaturase sequences displaying known functions werethen performed to predict the functions of the honeybee desa-turases. The premise was that, in a phylogenetic analysis of a familyof enzymes encoded by multiple related genes from closely relatedspecies, the sequences will be clustered according to the function ofenzyme/protein, or gene, or by the duplication history, but not byspecies (Roelofs and Rooney, 2003). In addition, the use of boot-strap confidence levels for assessing the accuracy of the phyloge-netic tree allows inferring that members of a cluster of desaturaseswith a bootstrap support �90% would probably display the samefunction (Liu et al., 2004). Such analysis allowed us to classify thehoneybee enzymes as being D9 (16 > 18) or D4 desaturases.Evidently, this needs experimental confirmation, especiallybecause subtle differences between the sequences may implydifferent functions, as detected for the group of the D10,11 desa-turases of lepidopterans (Liu et al., 2004).

Our phylogenetic analysis of desaturases highlighted thefollowing points: (1) the desaturases of some hymenopterans(Nasonia vitripennis and Harpegnathos saltator), a coleopteran (Tri-bolium castaneum) and a lepidopteran (D. plexippus) clusteredtogether as D14 desaturases (desaturation at the position D14) (seeFig. 3, blue clade); (2) the desaturase encoded by the GB48193 geneis possibly a D4 desaturase since it groups with the D4 desaturasesof Bombus terrestris (XP_003395145) and B. impatiens(XP_003492440), with a branch support of 0.963. The B. terrestrisdesaturase is an important enzyme for pheromones production(Bu�cek et al., 2013); (3) one desaturase encoding gene, GB48195,shares an apparent orthology relationship with the D. melanogasterDesatF (see NP_651966 in Fig. 3), which creates a second unsatu-ration in the incipient fatty acid that will originate polyunsaturatedCHCs. DesatF is expressed only in females and is involved in thesynthesis of alkadienes, which are used as pheromones by theseflies (Dallerac et al., 2000). Interestingly, the GB48195 gene ishighly expressed in forager bees and showed a high positive cor-relation with the unsaturated alkenes and alkadienes. Furtherstudies may inform whether this honeybee desaturase is female-specific and involved in the biosynthesis of pheromones.

Two elongase genes of the S/MUFA subfamily and four per-taining to the PUFA subfamily were previously identified inA. mellifera (Hashimoto et al., 2008). In the A. mellifera genomeversion 4.5 we found three S/MUFA members and eight PUFAmembers (see Fig. 4). Seven elongases that clustered into the S/MUFA subfamily (GB53872, GB53873, GB45596) and into the PUFAsubfamily (GB54396, GB51249, GB51247, GB54401), were mainlycorrelated to the production of alkenes and alkadienes, thus rein-forcing the suggestion that they have roles in the elongation ofmonounsaturated or polyunsaturated CHCs. This premise wasstrengthened by the results of RT-qPCR transcript profiles thatrevealed that except for the genes GB53873 and GB54401, theseelongase genes were highly expressed in the adult integumentwhere theymay contribute to the biosynthesis of unsaturated CHCsfor the forager cuticle. In contrast, the elongases encoded by thegenes GB54399 and GB40681 were mainly correlated to the pro-duction of methyl- and dimethyl-alkanes, but were clustered intothe PUFA class. This apparently discrepant result deserves furtherinvestigation.

4.3. Expression of desaturase and elongase genes in the context ofcuticle formation and maturation

In the current work, our attention focused on genes putativelyencoding two enzyme classes in the hydrocarbons biosynthetic


Q1


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465

66676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130

IB2576_proof ■ 10 May 2014 ■ 13/14

pathways, desaturases and elongases, both determining fatty acidstructure and function. The increased expression of five of the sixdesaturase genes and five of the ten elongase genes in newly-ecdysed adults and/or forager bees coincided with the period ofenlargement of the smooth endoplasmic reticulum in the oeno-cytes described by Hepburn et al. (1991). This structural change inthe CHC-synthesizing cells suggests increasing synthesis of thesecompounds and strengthens our premise that the genes hereanalyzed have a role in CHCs biosynthesis.

The increase in desaturase transcript levels in the integument ofadult honeybees correlated positively with the increase in theproportion of most of the unsaturated compounds (alkenes andalkadienes). Although such correlation requires further experi-mental validation, it is important because it has the potential toinform us whether, and how strongly, these variables (levels ofCHCs and transcripts) are interconnected. Together, the phyloge-netic and the correlation analysis were relevant as the first stepstoward the elucidation of the function of desaturase and elongasegenes in the biosynthesis of CHCs of interest. A suitable way toadvance in this direction consists in using RNA interference forsilencing these desaturase or elongase genes, followed by CHCanalysis in the silenced bees.

In a minor fraction of our correlation data, we could detect apositive relationship between levels of desaturase transcripts andlevels of saturated CHCs (alkanes and methyl-alkanes). At firstglance, the positive correlation between these variables seemedmerely casual, i.e., not representing causation effects, since theproducts of desaturase reactions are unsaturated, and not saturatedcompounds. However, saturated compounds are substrates fordesaturases, and cell regulatory mechanisms may ensure substrateavailability when the reaction product is needed for the organism.There are examples in biochemistry demonstrating that substratesmay induce the transcription of enzymes-encoding genes, thuscontrolling the production of metabolites through biosyntheticpathways and the rate at which both substrates and enzymes aresynthesized. This may tentatively explain why the expression ofdesaturase genes was positively correlated with the increase in theproportion of certain saturated compounds.

Elongases catalyze the elongation of saturated and unsaturatedfatty acids, however, a fraction of the CHC compounds, all of thembeing alkanes, methyl- or dimethyl-alkanes, showed a negativecorrelation with the expression of all the elongase genes hereincharacterized. Such genes may not have roles in the biosynthesis ofthese saturated CHCs. Interestingly, the elongase gene GB40681,which is significantly more expressed in the developing adultcuticle of the Pbm phase, was exclusively correlated with dimethyl-alkanes, coincidently present in higher proportions in the imma-ture than in the mature cuticle. Similarly, the expression of theelongase gene GB54399 was exclusively correlated with methyl-alkanes, suggesting functional specificity. Thus, our correlativedata points to strong causeeeffect relationships between theexpression of desaturase/elongase genes and levels of CHCs thussuggesting that these genes have roles in the biosynthesis of thesecompounds.

In conclusion, our data identified the hydrocarbon compositionof the developing envelope layer, the less studied structuralcomponent of the exoskeleton, thus increasing our understandingof its formation and maturation. To our knowledge, this is the firstapproach on the changes in CHCs profiles in the immature cuticleduring an entire molting cycle in comparison to the mature cuticle.Such analysis clearly demonstrated that different CHC classes pre-dominate in immature and mature cuticles. In addition, correlationanalysis allowed us to evaluate the strength of interconnectionbetween the changing developmental patterns of CHCs and thefluctuations in the expression of genes potentially involved in CHCs


biosynthesis. Additional insights into specific functions of thesegenes were provided by molecular phylogenetic analyses that alsohighlighted the evolutionary relationships between the desa-turases and elongases of the honeybee and other insects. All thisinformation is important for further studies on the molecularmechanisms leading to the diversity of CHC profiles and theirdevelopmental regulation.

Acknowledgments

This research was supported by FAPESP (São Paulo ResearchFoundation, Processes 2010/16380-9 and 2011/03171-5). Fellow-ships from CAPES and FAPESP (São Paulo Research Foundation,Process 2012/24284-4) were provided to T. Falcon. We are verygrateful to Dr. D. G. Pinheiro for bioinformatics support and to Dr. N.Châline for helpful comments on themanuscript.We also thank L.R.Aguiar for his valuable technical assistance in the apiary.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ibmb.2014.04.006.

References

Aitchison, J., 1982. The statistical analysis of compositional data. J. R. Stat. Soc. B 44,139e177.

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic localalignment search tool. J. Mol. Biol. 215, 403e410.

Le, S.Q., Gascuel, O., 2008. An improved general amino acid replacement matrix.Mol. Biol. Evol. 25, 1307e1320.

Anisimova, M., Gascuel, O., 2006. Approximate likelihood-ratio test for branches: afast, accurate, and powerful alternative. Syst. Biol. 55, 539e552.

Armold, M.T., Regnier, F.E., 1975. Stimulation of hydrocarbon biosynthesis brecdysterone in the flesh fly Sarcophaga bullata. J. Insect Physiol. 21, 1827e1833.

Ausubel, F., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K.,1995. Short Protocols in Molecular Biology. John Wiley & Sons, New York.

Billeter, J.C., Atallah, J., Krupp, J.J., Millar, J.G., Levine, J.D., 2009. Specialized cells tagsexual and species identity in Drosophila melanogaster. Nature 461, 987e991.

Bitondi, M.M., Mora, I.M., Simões, Z.L.P., Figueiredo, V.L., 1998. The Apis melliferapupal melanization program is affected by treatment with a juvenile hormoneanalogue. J. Insect Physiol. 44, 499e507.

Blomquist, G.J., Dillwith, J.W., 1985. Cuticular lipids. In: Kerkut, G.A., Gilbert, L.I.(Eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology,Integument, Respiration and Circulation, vol. 3. Pergamon, Oxford, pp. 117e154.

Blomquist, G.J., Chu, A.J., Remaley, S., 1980a. Biosynthesis of wax in the honeybee,Apis mellofera L. Insect Biochem 10, 313e321.

Blomquist, G.J., Howard, R.W., McDaniel, C.A., Remaley, S., Dwyer, L.A., Nelson, D.R.,1980b. Application of methoxymercuration-demercuration followed by massspectrometry as a convenient microanalytical technique for double-bondlocation in insect-derived alkenes. J. Chem. Ecol. 6, 257e269.

Blomquist, G.J., Jurenka, R., Schal, C., Tittiger, C., 2012. Pheromone Production:biochemistry and molecular biology. In: Gilbert, L.I. (Ed.), Insect Endocrinology.Elsevier, New York, pp. 523e567.

Bu�cek, A., Vogel, H., Matousková, P., Prchalová, D., Zácek, P., Vrkoslav, V., Sebesta, P.,Svatos, A., Jahn, U., Valterová, I., Pichová, I., 2013. The role of desaturases in thebiosynthesis of marking pheromones in bumblebee males. Insect Biochem. Mol.Biol. 43, 724e731.

Carlson, D.A., 1989. Dimethyl disulfide derivatives of long chain alkenes, alkadienes,and alkatrienes for gas chromatography/mass spectrometry. Anal. Chem. 61,1564e1571.

Châline, N., Sandoz, J.C., Martin, S.J., Ratnieks, F.L.W., Jones, G.R., 2005. Learning anddiscriminating of individual cuticular hydrocarbons by honeybees (Apis melli-fera). Chem. Senses 30, 327e335.

Chertemps, T., Duportets, L., Labeur, C., Ueyama, M., Wicker-Thomas, C., 2006.A female-specific desaturase gene responsible for diene hydrocarbon biosyn-thesis and courtship behaviour in Drosophila melanogaster. Insect Mol. Biol. 15,465e473.

Chino, H., Kitazawa, K., 1981. Diacylglycerol-carrying lipoprotein of hemolymph ofthe locustand some insects. J. Lipid Res. 22, 1042e1052.

Chino, H., Katase, H., Downer, R.G., Takahashi, K., 1981. Diacylglycerol-carrying li-poprotein of hemolymph of the American cockroach: purification, character-ization, and function. J. Lipid. Res. 22, 7e15.

Dallerac, R., Labeur, C., Jallon, J.M., Knipple, D.C., Roelofs, W.L., Wicker-Thomas, C.,2000. A delta 9 desaturase gene with a different substrate specificity isresponsible for the cuticular diene hydrocarbon polymorphism in Drosophilamelanogaster. Proc. Natl. Acad. Sci. U. S. A 97, 9449e9454.




Q2


1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162

63646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120

IB2576_proof ■ 10 May 2014 ■ 14/14

Dani, F.R., Jones, G.R., Corsi, S., Beard, R., Pradella, D., 2005. Nestmate recognitioncues in the honey bee: differential importance of cuticular alkanes and alkenes.Chem. Senses 30, 477e489.

Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2011. ProtTest 3: fast selection ofbest-fit models of protein evolution. Bioinformatics 27, 1164e1165.

Diehl, P.A., 1973. Paraffin synthesis in the oenocytes. Nature 243, 468e470.Dwyer, L.A., Zamboni, A.C., Blomquist, G.J., 1986. Hydrocarbon accumulation and

lipid biosynthesis during larval development in the cabbage looper, Trichoplusiani. Insect Biochem 16, 463e469.

Elias-Neto, M., Soares, M.P.M., Simões, Z.L.P., Bitondi, M.M.G., 2010. Developmentalcharacterization, function and regulation of a Laccase2 encoding gene in thehoney bee, Apis mellifera (Hymenoptera, Apinae). Insect Biochem. Mol. Biol. 40,241e251.

Elias-Neto, M., Nascimento, A.L.O., Bonetti, A.M., Nascimento, F.S., Mateus, S.,Garófalo, C.A., Bitondi, M.M.G., 2013. Heterochrony of cuticular differentiationin eusocial corbiculate bees. Apidologie. http://dx.doi.org/10.1007/s13592-013-0254-1.

Fan, Y., Zurek, L., Dykstra, M.J., Schal, C., 2003. Hydrocarbon synthesis by enzy-matically dissociated oenocytes of the abdominal integument of the GermanCockroach, Blattella germanica. Naturwissenschaften 90, 121e126.

Felsenstein, J., 1981. Evolutionary trees from DNA sequences: a maximum likelihoodapproach. J. Mol. Evol. 17, 368e376.

Ferreira-Caliman, M.J., Nascimento, F.S., Turatti, I.C., Mateus, S., Lopes, N.P.,Zucchi, R., 2010. The cuticular hydrocarbons profiles in the stingless bee Meli-pona marginata reflect task-related differences. J. Insect Physiol. 56, 800e804.

Gibbs, A.G., 2002. Lipid melting and cuticular permeability: new insights into an oldproblem. J. Insect Physiol. 48, 391e400.

Guindon, S., Dufayard, J.F., Lefort, V., Anisimova, M., Hordijk, W., Gascuel, O., 2010.New algorithms and methods to estimate maximum-likelihood phylogenies:assessing the performance of PhyML 3.0. Syst. Biol. 59, 307e321.

Guo, L., Blomquist, G.J., 1991. Identification, accumulation, and biosynthesis of thecuticular hydrocarbons of the southern armyworm, Spodoptera eridania(Cramer) (Lepidoptera: Noctuidae). Arch. Insect Biochem. Physiol. 16, 19e30.

Hao, G., O’Connor, M., Liu, W., Roelofs, W.L., 2002. Characterization of Z/E11- andZ9-desaturases from the obliquebanded leafroller moth, Choristoneura rosa-ceana. J. Insect Sci. 2, 26.

Hashimoto, K., Yoshizawa, A.C., Okuda, S., Kuma, K., Goto, S., Kanehisa, M., 2008. Therepertoire of desaturases and elongases reveals fatty acid variations in 56eukaryotic genomes. J. Lipid Res. 49, 183e191.

Hepburn, H.R., 1985. Structure of the integument. In: Kerkut, G.A., Gilbert, L.I. (Eds.),Comprehensive Insect Physiology, Biochemistry and Pharmacology, Integu-ment, Respiration and Circulation, vol. 3. Pergamon, Oxford, pp. 1e58.

Hepburn, H.R., Bernard, R.T.F., Davidson, B.C., Muller, W.J., Lloyd, P., Kurstjens, S.P.,Vincent, S.L., 1991. Synthesis and secretion of beeswax in honeybees. Apidologie22, 21e36.

Howard, R.W., Blomquist, G.J., 2005. Ecological, nehavioral, and chemical aspects ofinsect hydrocarbons. Annu. Rev. Entomol. 50, 371e393.

Hübschmann, H.J., 2009. Handbook of GC/MS: Fundamentals and Applications,second ed. WILEY-VCH Verlag GmbH & Co, KGaA, Weinheim.

Karanth, S., Lall, S.P., Denovan-Wright, E.M., Wright, J.M., 2009. Differential tran-scriptional modulation of duplicated fatty acid-binding protein genes by dietaryfatty acids in zebrafish (Danio rerio): evidence for subfunctionalization orneofunctionalization of duplicated genes. BMC Evol. Biol. 9, 219.

Kather, R., Drijfhout, F.P., Martin, S.J., 2011. Task group differences in cuticular lipidsin the honey bee Apis mellifera. J. Chem. Ecol. 37, 205e212.

Katoh, K., Misawa, K., Kuma, K., Miyata, T., 2002. MAFFT: a novel method for rapidmultiple sequence alignment based on fast Fourier transform. Nucleic AcidsRes. 30, 3059e3066.

Katoh, K., Kuma, K., Toh, H., Miyata, T., 2005. MAFFT version 5: improvement inaccuracy of multiple sequence alignment. Nucl. Acids Res. 33, 511e518.

Knipple, D.C., Rosenfield, C.L., Miller, S.J., Liu, W., Tang, J., Ma, P.W., Roelofs, W.L.,1998. Cloning and functional expression of a cDNA encoding a pheromonegland-specific acyl-CoA Delta11-desaturase of the cabbage looper moth, Tri-choplusia ni. Proc. Natl. Acad. Sci. U. S. A 95, 15287e15292.

Kolattukudy, P.E., 1965. Biosynthesis of wax in Brassica oleracea. Biochemistry 4,1844e1855.

Kolattukudy, P.E., 1968. Biosynthesis of surface lipids. Sci. 159, 498e505.Krasnec, M.O., Breed, M.D., 2013. Colony-specific cuticular hydrocarbon profile in

Formica argentea ants. J. Chem. Ecol. 39, 59e66.Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A.,

McWilliam, H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D.,Gibson, T.J., Higgins, D.G., 2007. Clustal W and Clustal X version 2.0. Bioinfor-matics 23, 2947e2948.

Liénard, M.A., Strandh, M., Hedenström, E., Johansson, T., Löfstedt, C., 2008. Keybiosynthetic gene subfamily recruited for pheromone production prior to theextensive radiation of Lepidoptera. BMC Evol. Biol. 8, 270.


Liu, W., Ma, P.W., Marsella-Herrick, P., Rosenfield, C.L., Knipple, D.C., Roelofs, W.,1999. Cloning and functional expression of a cDNA encoding a metabolic acyl-CoA delta 9-desaturase of the cabbage looper moth, Trichoplusia ni. InsectBiochem. Mol. Biol. 29, 435e443.

Liu, W., Rooney, A.P., Xue, B., Roelofs, W.L., 2004. Desaturases from the spottedfireworm moth (Choristoneura parallela) shed light on the evolutionary originsof novel moth sex pheromone desaturases. Gene 342, 303e311.

Lourenço, A.P., Mackert, A., Cristino, A.S., Simões, Z.L.P., 2008. Validation of referencegenes for gene expression studies in the honey bee, Apis mellifera, by quanti-tative real-time RT-PCR. Apidologie 39, 372e385.

Martin, C., Salvy, M., Provost, E., Bagnères, A.G., Roux, M., Crauser, D., Clement, J.L.,Le Conte, Y., 2001. Variations in chemical mimicry by the ectoparasitic miteVarroa jacobsoni according to the developmental stage of the host honey-beeApis mellifera. Insect Biochem. Mol. Biol. 31, 365e379.

Michelette, E.R.F., Soares, A.E.E., 1993. Characterization of preimaginal develop-mental stages in Africanized honey bee workers (Apis mellifera L). Apidologie24, 431e440.

Nation, J.L., Sanford, M.T., Milne, K., 1992. Cuticular hydrocarbons from Varroajacobsoni. Exp. Appl. Acarol. 16, 331e344.

Nunes, T.M., Nascimento, F.S., Turatti, I.C., Lopes, N.P., Zucchi, R., 2008. Nestmaterecognition in a stingless bee: does the similarity of chemical cues determineguard acceptance? Anim. Behav. 75, 1165e1171.

Piek, T., 1964. Synthesis of wax in the honeybee (Apis mellifera L.). J. Insect Physiol.10, 563e572.

Qiu, Y., Tittiger, C., Wicker-Thomas, C., Le Goff, G., Young, S., Wajnberg, E., Fricaux, T.,Taquet, N., Blomquist, G.J., Feyereisen, R., 2012. An insect-specific P450 oxida-tive decarbonylase for cuticular hydrocarbon biosynthesis. Proc. Natl. Acad. Sci.U. S. A 109, 14858e14863.

de Renobales, M., Blomquist, G.J., 1983. A developmental study of the compositionand biosynthesis of the cuticular hydrocarbons of Trichoplusia ni (Lepidoptera:Noctuidae). Insect Biochem 13, 493e502.

de Renobales, M., Nelson, D.R., Mackay, M.E., Zamboni, A.C., Blomquist, G.J., 1988.Dynamics of hydrocarbon biosynthesis and transport to the cuticle duringpupal and early adult development in the cabbage looper Trichoplusia ni(Lepidoptera: Noctuidae). Insect Biochem 18, 607e613.

Roelofs, W.L., Rooney, A.P., 2003. Molecular genetics and evolution of pheromonebiosynthesis in Lepidoptera. Proc. Natl. Acad. Sci. U. S. A 100, 14599.

Rutherford, K., Parkhill, J., Crook, J., Horsnell, T., Rice, P., Rajandream, M.A., Barrell, B.,2000. Artemis, sequence visualization and annotation. Bioinformatics 16, 944e945.

Ruvolo, M., Landim, C., 1993. Morphologic and morphometric aspects of oenocytesof Apis mellifera queens and workers in different phases of life. Mem. Inst.Oswaldo Cruz 88, 387e395.

Santos, A.E., Bitondi, M.M.G., Simões, Z.L.P., 2001. Hormone-dependent proteinpatterns in integument and cuticular pigmentation in Apis mellifera duringpharate adult development. J. Insect Physiol. 47, 1275e1282.

Schal, C., Sevala, V.L., Young, H.P., Bachmann, J.A.S., 1998. Sites of synthesis andtransport pathways of insect hydrocarbons: cuticle and ovary as target tissues.Am. Zool. 38, 382e393.

Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparativeC(T) method. Nat. Protoc. 3, 1101e1108.

Singer, T.L., Espelie, K.E., 1992. Social wasps use nest paper hydrocarbons for nest-mate recognition. Anim. Behav. 44, 63e68.

Soares, M.P.M., Elias-Neto, M., Simões, Z.L.P., Bitondi, M.M.G., 2007. A cuticle proteingene in the honeybee: expression during development and in relation to theecdysteroid titer. Insect Biochem. Mol. Biol. 37, 1272e1282.

Soares, M.P.M., Silva-Torres, F.A., Elias-Neto, M., Nunes, F.M.F., Simões, Z.L.P.,Bitondi, M.M.G., 2011. Ecdysteroid-dependent expression of the tweedle andperoxidase genes during adult cuticle formation in the honey bee, Apis mellifera.PloS one 6, e20513.

Soares, M.P.M., Barchuk, A.R., Simões, A.C.Q., Cristino, A.S., Freitas, F.C.P., Canhos, L.L.,Bitondi, M.M.G., 2013. Genes involved in thoracic exoskeleton formation duringthe pupal-to-adult molt in a social insect model, Apis mellifera. BMC Genomics14, 576.

Wicker-Thomas, C., Henriet, C., Dallerac, R., 1997. Partial characterization of a fattyacid desaturase gene in Drosophila melanogaster. Insect Biochem. Mol. Biol. 27,963e972.

Wigglesworth, V.B., 1970. Structural lipids in the insect cuticle and the function ofthe oenocytes. Tissue Cell. 2, 155e179.

Young, H.P., Schal, C., 1997. Cuticular hydrocarbon synthesis in relation to feedingand developmental stage in nymphs of Blattella germanica (Dictyoptera: Blat-tellidae). Ann. Entomol. Soc. Am. 90, 655e663.

Zufelato, M.S., Lourenço, A.P., Simões, Z.L.P., Jorge, J.A., Bitondi, M.M.G., 2004. Phe-noloxidase activity in Apis mellifera honey bee pupae, and ecdysteroid-dependent expression of the prophenoloxidase mRNA. Insect Biochem. Mol.Biol. 34, 1257e1268.

121122123124


http://dx.doi.org/10.1007/s13592-013-0254-1

http://dx.doi.org/10.1007/s13592-013-0254-1

exoskeleton formation in apis mellifera: cuticular hydrocarbons profiles and expression of...

Documents