GENETICA MOLECOLARE DELLO SVILUPPO DEGLI INSETTI XVII.pdf · DELLO SVILUPPO DEGLI INSETTI Tavole...

ACCADEMIA NAZIONALE

ITALIANA DI ENTOMOLOGIAFirenze 2008

GENETICA MOLECOLAREDELLO SVILUPPO DEGLI INSETTI

Tavole Rotonde sui maggiori problemiriguardanti l’Entomologia Agraria in Italia

Sotto gli auspici del MIPAAF

XVII.

Estratto da:ATTI DELLA

ACCADEMIA NAZIONALEITALIANA DI ENTOMOLOGIA

Rendiconti Anno LVI - 2008

GENETICA MOLECOLAREDELLO SVILUPPO DEGLI INSETTI

Tavole Rotonde sui maggiori problemiriguardanti l’Entomologia Agraria in Italia

Sotto gli auspici del MIPAAF

XVII.

© 2008 Accademia Nazionale Italiana di Entomologia50125 Firenze - Via Lanciola 12/a

ISBN 978-88-96493-01-4

«Dobbiam fare le maraviglie come i Naturalistitentato non abbiano intorno quello soggetto(lo sviluppo) alcune sperienze nella famigliadegli insetti; giacché abbiam luogo di presu-mere che non sarebbono senza esito.»

C. Bonnet (1773)«Contemplazione della Natura»

Nel corso delle riunioni del 2008 questa Ac ca -demia ha discusso alcuni aspetti della genetica mole-colare nello sviluppo degli insetti. In questo campodi studio, l’applicazione delle tecniche di manipola-zione genetica, sviluppate in un insetto modello, ilmoscerino della frutta Drosophila melanogaster,associate alla grande mole di conoscenze disponibilisulla embriologia comparata e sulla filogenesi degliinsetti, stanno permettendo di fare rapidi progressiin campi fino ad ora inesplorati ad un livello moleco-lare (Carla Malva). Un numero relativamente pic-colo di queste vie di segnalazione controlla unastraordinaria diversità di risposte e decisioni chiaveche, nelle cascate di eventi che regolano lo sviluppodi un organismo, portano alla acquisizione di destinicellulari diversi. I componenti principali di questecascate di trasduzione di segnali, sono evolutiva-mente conservati dagli insetti all’uomo.

Come primo esempio Carla Malva ha usato il suostesso lavoro sperimentale. Nel corso di questiultimi studi, soprattutto in quelli relativi a Toxo -neuron nigriceps e al polydnavirus ad esso asso-ciato, svolti in collaborazione con i gruppi diretti daE. Tremblay e F. Pennacchio, è stato scoperto che ilparassitoide, per alterare molti processi biologicivitali del suo ospite, quali ad esempio lo sviluppo e larisposta immunitaria, usa molecole molto simili aquelle descritte in Drosophila ed anche nei mammi-feri. Fra esse, le proteine cactus-IκB, inibitori dei fat-tori di trascrizione della famiglia NFκB, sono similia quelle che in Drosophila controllano una serie diprocessi biologici fondamentali. Esse competonocon le molecole endogene dell’insetto ospite, inquanto simili, ma, non avendo più la loro attivitàbiologica, alterano vie di trasduzione del segnaleimportanti per la sopravvivenza dell’ospite. Daquesti studi emerge che non soltanto sono conser-vate le vie di segnalazione che regolano le rispostecellulari mediate dai fattori di trascrizione, dagliinsetti ai mammiferi, ma anche le strategie utilizzatedai patogeni e parassiti per interferire con l’attiva-zione di questi fattori nei loro ospiti. Infatti proteinesimili agli inibitori cactus-IκB del TnBV sono statetrovate nel genoma di altri polydnavirus, ma anche

in altri virus dei vertebrati che mostrano efficientistrategie di soppressione della risposta immunitaria.

Un altro interessantissimo esempio di conserva-zione evolutiva di molecole implicate nei processi disviluppo è stato presentato e discusso da MiodragGrbic del Department of Biology, University ofWestern Ontario, Canada. La maggior parte deglistudi sono stati compiuti sull’insetto parassitoidepoliembrionico Copidosoma floridanum che paras-sitizza il nottuide Trichoplusia ni, e forma, da unsingolo uovo, fino a 2000 embrioni che daranno ori-gine ad individui differenziati in caste.

Lo sviluppo embrionale del C. floridanum diffe-risce da quello degli altri insetti. Il Co pidosomadepone uova piccole e senza vitello, circondate da unsottile corion. Il gene vasa è stato isolato inCopidosoma e si è osservato che è trascritto nellecellule nutrici dell’ovario. È quindi veramente stu-pefacente che la presenza o assenza di una proteinapossa decidere il destino riproduttivo e, come conse-guenza, un diverso destino di casta. La formazionedel guscio dell’uovo richiede la sintesi coordinata didifferenti componenti proteiche nelle cellule follico-lari della camera ovarica.

Giuliano Callaini, del Dipartimento di BiologiaEvolutiva dell’Università di Siena, si interessa datempo dei processi di duplicazione dei centrioli e deicentrosomi nella divisione meiotica durante la sper-matogenesi. La funzione delle differenti componentidei centrosomi viene studiata in condizioni in cui siproducono fusi con centrosomi multipli o fusimonopolari. In conclusione, lo studio della gameto-genesi in Drosophila permette di dissezionare pro-cessi biologici fondamentali quali la regolazionedella espressione genica tessuto-specifica e la regola-zione del ciclo cellulare, come è stato illustrato daGiuseppe Gargiulo del Dipartimento di BiologiaEvoluzionistica Sperimentale dell’Università diBologna.

Per finire, riallacciandosi alla frase del Bonnetdel 1773 riportata all’inizio, appare evidente che sestudiando la D. melanogaster si sono scoperti mec-canismi di regolazione dello sviluppo così efficientie precisi, immaginiamo quali meraviglie potrannoscaturire dallo studio, ad un livello molecolare e conle sofisticate metodologie oggi disponibili. L’oggettodelle discussioni avvenute nel 2008 in Accademiaha fornito un basilare contributo in questo ambitodi conoscenze.

BACCIO BACCETTI

Presidente dell’Accademia Italiana di Entomologia

PRESENTAZIONE

INDICE

GENETICA MOLECOLARE DELLO SVILUPPO DEGLI INSETTI

CARLA MALVA – Genetica molecolare dello sviluppo degli insetti - Presentazione . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

CARLA MALVA – From D. melanogaster to insect parasitoids: different processes, conserved molecularmechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

MIODRAG GRBIC’ – Evolution of polyembryonic development in parasitic wasps . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

GIUSEPPE GARGIULO – Control of vitelline membrane gene expression during D. melanogaster ooge-nesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

GIULIANO CALLAINI – Drosophila spermatogenesis: a system model to cell cycle analysis . . . . . . . . . . . . . . . . . .

Pag. 33

» 37» 43

» 53» 59

Atti Accademia NazionaleItaliana di EntomologiaAnno LVI, 2008: 33-35

(*) Istituto di Genetica e Biofisica, via Pietro Castellino 111, Napoli; e-mail: [email protected] oggetto delle Sedute pubbliche dell’Accademia tenute nell’Anno accademico 2008.

GENETICA MOLECOLARE DELLO SVILUPPO DEGLI INSETTI

CARLA MALVA (*)

«Dobbiam fare le maraviglie come i Naturalistitentato non abbiano intorno quello soggetto(lo sviluppo) alcune sperienze nella famigliadegli insetti; giacché abbiam luogo di presu-mere che non sarebbono senza esito.C. Bonnet «Contemplazione della Natura»

Nel corso delle riunioni del 2008 abbiamo dis-cusso alcuni aspetti della genetica molecolaredello sviluppo degli insetti. In questo campo distudio, l’applicazione delle tecniche di manipola-zione genetica, sviluppate in un insetto modello, ilmoscerino della frutta Drosophila melanogaster,associate alla grande mole di conoscenze disponi-bili sulla embriologia comparata e sulla filogenesidegli insetti, stanno permettendo di fare rapidiprogressi in campi fino ad ora inesplorati ad unlivello molecolare. Le presentazioni sono stateimperniate sull’osservazione che i meccanismi dirisposta delle cellule a stimoli endogeni o esogenisono conservati nel corso dell’evoluzione e sonousati reiteratamente negli organismi di una stessaspecie in processi diversi. Un numero relativamen-te piccolo di queste vie di segnalazione controllauna straordinaria diversità di risposte e decisionichiave che, nelle delicate cascate di eventi cheregolano lo sviluppo di un organismo, portano allaacquisizione di destini cellulari diversi. I compo-nenti principali di queste cascate di trasduzione disegnali: recettori, molecole adattatrici, attivatoritrascrizionali, complessi enzimatici attivatori-ini-bitori, sono evolutivamente e funzionalmente con-servati dagli insetti all’uomo.

Come primo esempio ho usato il mio stessolavoro sperimentale in cui, dallo studio dello svi-luppo e dell’oogenesi della D. melanogaster, sonoapprodata alla analisi delle basi molecolari delleinterazioni fisiologiche fra alcuni braconidi paras-sitoidi ed i loro ospiti. Nel corso di questi ultimistudi, soprattutto in quelli relativi a Toxoneuronnigriceps e al polydnavirus ad esso associato(TnBV), svolti in collaborazione con i gruppi

diretti da E. Tremblay e F. Pennacchio, abbiamoscoperto che il parassitoide, per alterare moltiprocessi biologici vitali del suo ospite, quali adesempio lo sviluppo e la risposta immunitaria, usamolecole molto simili a quelle descritte in Droso-phila ed anche nei mammiferi. Fra esse, le protei-ne cactus-IκB, inibitori dei fattori di trascrizionedella famiglia NFκB, sono simili a quelle che inDrosophila controllano una serie di processi bio-logici fondamentali, tra cui lo stabilirsi dell’assedorso-ventrale dell’embrione e la risposta antimi-crobica, ma sono troncate e mancano di alcuni sitiattivi nella regolazione della risposta da essi con-trollata. Esse competono con le molecole endoge-ne dell’insetto ospite, in quanto simili, ma, nonavendo più la loro attività biologica, alterano viedi trasduzione del segnale importanti per lasopravvivenza dell’ospite. Da questi studi emergeche non soltanto sono conservate le vie di segnala-zione che regolano le risposte cellulari mediate daifattori di trascrizione NFκB, dagli insetti ai mam-miferi, ma anche le strategie utilizzate dai patogenie parassiti per interferire con l’attivazione di que-sti fattori nei loro ospiti. Infatti proteine simili agliinibitori cactus-IκB del TnBV sono state trovatenel genoma di altri polydnavirus, ma anche in altrivirus dei vertebrati che mostrano efficienti strate-gie di soppressione della risposta immunitaria.

Un altro interessantissimo esempio di conserva-zione evolutiva di molecole implicate nei processidi sviluppo è stato presentato e discusso da MiodragGrbic del Department of Biology, University ofWestern Ontario, Canada. Grbic studia le strategiee modalità di embriogenesi in alcuni parassitoidinei quali si osserva una delle più interessanti e pecu-liari soluzioni adattative sviluppate da parassiti ani-mali: la poliembrionia. La maggior parte degli studisono stati compiuti sull’insetto parassitoide poliem-brionico Copidosoma floridanum che parassitizza ilnottuide Trichoplusia ni, e forma, da un singolouovo, fino a 2000 embrioni che daranno origine adindividui differenziati in caste.

Lo sviluppo embrionale del C. floridanum diffe-risce drammaticamente da quello degli altri inset-ti. Il Copidosoma depone uova piccole e senzavitello, circondate da un sottile corion. La primadivisione dell’uovo è totale e dà luogo alla forma-zione di due blastomeri, la seconda crea un blasto-mero piccolo e altri tre uguali tra di loro e di gran-dezza maggiore. La cellula piccola è diversa dallealtre cellule e non condivide con esse lo stessodestino cellulare. Questo si è potuto dimostrarestudiando l’ espressione del gene vasa identificatoprima in D. melanogaster e poi in altri insetti. Laproteina codificata da questo gene è una RNA eli-casi e in Drosophila è implicata nella specificazio-ne della linea germinale. Il gene vasa è stato isolatoin Copidosoma e si è osservato che è trascritto nellecellule nutrici dell’ovario. La proteina maternaVasa in Copidosoma si localizza in modo asimme-trico nell’uovo e durante l’embriogenesi risultainvariabilmente localizzata solo nel blastomeropiccolo e successivamente nel gruppo delle cellulefiglie che da esso derivano, suggerendo che questecellule rappresentano una specifica linea cellulare,la linea germinale, da cui si formeranno le gonadi.

È quindi veramente stupefacente che la presen-za o assenza di una proteina possa decidere ildestino riproduttivo e, come conseguenza, undiverso destino di casta.

È molto importante che nel corso dello sviluppocontrolli stringenti modulino trascrizione, tradu-zione e processi post-traduzionali, localizzazioneasimmetrica e segregazione delle molecole, pergarantire la presenza di specifici messaggeri e pro-teine, sia con funzione di regolazione che struttu-rale, nel posto giusto ed al momento giusto. Unesempio di regolazione spaziale e temporale moltopreciso è rappresentato dal controllo della espres-sione dei geni che codificano per le proteine dellaMembrana Vitellina (VM), che, insieme al corion,costituisce il guscio dell’uovo di Drosophila mela-nogaster. Questo tema è stato presentato e discus-so da G. Gargiulo, del Dipartimento di BiologiaEvoluzionistica Sperimentale, Università di Bolo-gna.

La formazione del guscio dell’uovo richiede la sin-tesi coordinata di differenti componenti proteichenelle cellule follicolari della camera ovarica. Le pro-teine sono successivamente secrete nello spazio extra-cellulare situato tra le cellule follicolari e la superficiedell’uovo, dove possono andare incontro a tagli pro-teolitici e a trasporto selettivo. Infine esse si aggre-gano in complessi sopramolecolari insolubili attraversola formazione di legami covalenti che coinvolgonospecifici domini proteici. I geni che codificano perle varie componenti della membrana vitellina sonoespressi coordinatamente verso la fine dell’oogenesi

tra lo stadio 8 e lo stadio 10, con una specifica sequenzatemporale correlata alle funzioni svolte dalle singoleproteine.

Mentre quasi tutti i geni VM sono espressi in tuttol’epitelio follicolare, il gene chiamato VM32E, cheè l’ultimo in ordine temporale ad essere trascritto,ha una peculiare localizzazione del messaggero chesi trova solo nelle cellule follicolari colonnari late-rali e mai in quelle anteriori e posteriori.

Solo dopo che il messaggero è stato tradotto e laproteina secreta, questa si muove ai poli dell’uovoed assume una distribuzione uniforme. Quindi,per motivi ancora sconosciuti, è necessario che laproteina VM32E sia presente ai due poli solo allafine del processo di sintesi delle altre proteine VMe appena prima dell’assemblaggio irreversibile delguscio. Questo peculiare profilo di espressionespaziale e temporale si ottiene attraverso l’attivitàdi varie proteine di controllo della trascrizione,che agiscono su specifiche sequenze bersaglio,presenti nel promotore del gene VM32E.

Quindi la proteina VM32E sembra svolgere unruolo molto peculiare non solo nella formazione ecostruzione del guscio dell’uovo, ma anche neiprocessi di localizzazione ed intrappolamento disegnali materni che, secreti durante l’oogenesi,saranno attivati con la fertilizzazione e sarannodeterminanti per la formazione degli assi di sim-metria dell’embrione ed il suo corretto sviluppo.

In Drosophila non solo l’oogenesi ma anche laspermatogenesi rappresenta un ottimo modellosperimentale che permette di dissezionare proces-si biologici fondamentali, primo tra tutti la regola-zione del ciclo cellulare. Giuliano Callaini, delDipartimento di Biologia Evolutiva dell’Universi-tà di Siena, si interessa da tempo dei processi diduplicazione dei centrioli e dei centrosomi nelladivisione meiotica durante la spermatogenesi, par-ticolarmente adatta allo studio dei processi di for-mazione del fuso e della citodieresi. La meiosi neimaschi di D. melanogaster manca, infatti, di«checkpoints». Questo consente alle cellule diproseguire nel ciclo cellulare anche se compaionodifetti di organizzazione del fuso, dei centrosomi odell’allineamento dei cromosomi. Quindi si puòstudiare il ruolo di specifici geni regolatori attra-verso l’effetto che specifiche mutazioni a loro cari-co producono sulla duplicazione e segregazionedei centrosomi. La funzione delle differenti com-ponenti dei centrosomi si studia in condizioni incui si producono fusi con centrosomi multipli ofusi monopolari. Con esperimenti di immunoloca-lizzazione si possono inoltre localizzare questi dif-ferenti componenti nei vari distretti del fuso enelle varie fasi delle divisioni cellulari, avendo cosìinformazioni utili per la loro funzione.

– 34 –

In conclusione, lo studio della gametogenesi inDrosophila permette di dissezionare processi bio-logici fondamentali quali la regolazione dellaespressione genica tessuto-specifica e la regolazio-ne del ciclo cellulare, qui brevemente riassunti eapprofonditamente discussi nelle riunioni accade-miche 2008, ma anche il mantenimento e rinnovodelle cellule staminali, la comunicazione intercel-lulare tra linea germinale e linea somatica, l’acqui-sizione di destini cellulari diversi, la generazioneed il mantenimento della polarità negli epiteli, imeccanismi di migrazione cellulare e di conversio-ne di cellule epiteliali in cellule migratorie invasi-ve, l’apoptosi, il traffico nucleo-citoplasmatico, iltrasporto e la localizzazione asimmetrica di RNA eproteine.

Per finire, riallacciandomi alla frase premonitri-ce del Bonnet del 1773 riportata all’inizio, se stu-diando la D. melanogaster si sono scoperti mecca-nismi di regolazione dello sviluppo così efficienti eprecisi, immaginiamo quali meraviglie potrannoscaturire dallo studio, ad un livello molecolare econ le sofisticate metodologie oggi disponibili,della incredibile diversità di stili di vita, modalitàriproduttive, processi di sviluppo e forme degliinsetti. Questo futuro è già cominciato e ogni gior-no si chiariscono nuovi meccanismi e vengonoprodotti tanti nuovi dati: quanto discusso nel 2008in Accademia fornisce un interessante contributoin questo ambito di conoscenza.

– 35 –


(*) Istituto di Genetica e Biofisica, via Pietro Castellino 111, Napoli; [email protected] tenuta nella Seduta pubblica dell’Accademia - Firenze, 23 febbraio 2008.

From D. melanogaster to insect parasitoids: different processes, conserved molecular mechanismsSeveral parasitic wasps, when ovipositing in host insects, inject factors that disrupt their physiology, development and immune

reaction. Among these factors, the polydnaviruses (PDVs), obligate symbionts of the parasitoids, play a pivotal role in successfulparasitism. We have shown that the genome of the polydnavirus associated with Toxoneuron nigriceps (TnBV) codes fortruncated forms of Cactus/IκB-like proteins. In Drosophila, Cactus regulates the nuclear import of various NF-κB/Rel proteins,which are transcription factors controlling embryonic dorso-ventral patterning and antimicrobial response. Insect immuneresponse shares many components with the mammalian innate immunity, including the intracellular signaling pathways thatactivate NF-κB. The viral cactus-like proteins may act as dominant inhibitors leading to the cytoplasmic retention of NF-κB/Relproteins, thus providing a molecular mechanism accounting for the impairment of insect immune response induced byparasitoids.

We discuss the role of Cactus/IκB molecules as an example of the reiterated use of intracellular signaling pathways conservedduring evolution.

KEY WORDS: Drosophila melanogaster, developmental genetics, insect parasitoids, polydnaviruses, immune response.

FROM D. MELANOGASTER TO INSECT PARASITOIDS:DIFFERENT PROCESSES, CONSERVED MOLECULAR MECHANISMS

CARLA MALVA (*)

THE CACTUS GENE IN DROSOPHILA MELANOGASTER

a) establishment of Dorsal-Ventral axisAntero-Posterior (A/P) and Dorsal-Ventral

(D/V) axis formation of the Drosophila embryo isestablished during oogenesis and requiresreciprocal cell signaling between germline (oocyteand nurse cells) and somatic follicle cells in the eggchamber (SPRADLING, 1993; FULLER and SPRA -DLING, 2007). The first visible asymmetry duringthe egg chamber development is the positioning ofthe oocyte posteriorly in the germline cyst (Fig. 1)and this early event plays a key role in the esta -blishment of both major body axes. The posteriorlocalization of the oocyte relative to the nurse cellsin the egg chamber allows a reciprocal cross-talkbetween the oocyte and the few follicle cells thatcontact it. The key signaling pathway is composedof the ligand Gurken (Grk), the transforminggrowth factor α (TGFα) homologue, present inthe oocyte, and its receptor, the epidermal growthfactor receptor homologue (Egfr-top), expressedby all somatic follicle cells (CLIFFORD andSCHUPBACH, 1992; ROTH et al., 1995; GONZALEZ-REYES et al., 1995). A large number of factorsregulate or amplify the signal generated by thisligand/receptor interaction, leading the folliclecells adjacent to the oocyte to adopt a posterior

fate. Signaling from the follicle cells back to theoocyte induces the reorganization of the corticalcytoskeleton and the polarization of the micro -tubule network whose plus ends become dire ctedtoward the posterior pole. As a conse quence, grkmRNA, associated with the oocyte nucleus, movestowards the future dorsal-anterior corner of theoocyte, where the newly translated Grk proteinacts as dorsalizing signal (Fig. 2a). At the sametime, the microtubule network is also involved inthe intracellular transport of a large number ofRNA and protein molecules establishing the A-Paxis (Fig. 2b, c).

The induction of D-V polarity in the follicularepithelium triggers differential gene activity in thedorsal and ventral sides which, in turns, shapes theegg shell, chorion and vitelline membrane,constructed by the follicle cells around the oocyte(see GARGIULO in this volume). This polarity isvisibly marked by the chorion appendages, locatedin a dorsolateral position of the egg.

The differential expression of genes in theventral part of the follicular epithelium alsoproduces extraembryonic signals secreted only inthe ventral regions of the perivitelline space. Thesesignals, once activated, promote the inductiveprocess that establishes the D-V polarity of theembryo (MOUSSIAN and ROTH, 2005).

The first step of this process consists in theventral activation of a protease cascade culmi -nating in the production of an active ligand for amembrane receptor called Toll. Ventral Tollactivation is transduced into the embryo byadaptor proteins and kinases leading tophosphorylation and degradation of the proteinCactus (Cact), which is the inhibitor of the NF-κBtranscription factor called Dorsal (Dl). (Scheme I).Cactus belongs to the family of IκB proteins,characterized by the presence of multiple ankyrindomains able to bind one or more NF-κB/relfamily members (ROTH et al., 1991, MOUSSIAN andROTH, 2005). Upon Cactus phosphorylation anddegradation, Dorsal is released, becaming able toenter the nucleus and activate zygotic genesrequired for ventral cell fate specification.

b) Immune responseThe NFκB signaling system has been conserved

to operate on divergent genes in many differentspecies (GHOSH et al., 1998; SILVERMAN andMANIATIS, 2001). In Drosophila, beside embryonicD-V patterning, the IκB protein Cactus, intera -cting with a set of NF-κB related transcriptionfactors belonging to the Rel family, regulates

– 38 –

Figure 1Wild type D. mela -nogaster ovariole stainedwith DAPI to marknuclei. Egg chambersfrom wild type femaleswere stained with DAPIand analyzed by conven -tional epifluorescence.Arrow indicates theposterior position of theoocyte in the eggchamber.

Figure 2In situ hybridization experiments on wild type D. melanogasteregg chambers showing the mRNA localization of grk (a), linkedto the oocyte nucleus at the dorsal anterior corner, bicoid (b),localized anteriorly, and oskar (c) in the posterior part of thegrowing oocyte.

multiple cellular responses, including the anti -microbial defences operated by the innate immunesystem (HOFFMANN and REICHHART, 2002;). TheDrosophila genome encodes three Rel proteins,sharing a Rel-homology domain, which consists ofa conserved region of 300-amino-acid (aa) that isresponsible for dimerization and DNA binding.

These proteins are Dorsal, described above inD-V patterning of the embryo, Dorsal-related

Immunity Factor (DIF) and Relish (WU andANDERSON, 1998). Of these three Rel proteins,DIF is the main transactivator of a large numberof genes encoding Drosomycin and otherantifungal and anti-bacterial peptides directedagainst Gram-positive pathogens. The Relishprotein is upregulated upon bacterial infectionand is mainly involved in the defensive reactionsagainst Gram-negative bacteria leading to thetranscription of genes coding for the antibacterialpeptides Diptericin, Cecropin, Drosocin andAttacin. Relish is not inhibited by Cactus, butcarries by itself ankyrin repeat domains located inthe C terminal region (HOFFMANN, 2003).

This short summary obviously represents a hugeoversimplification of the complex process ofimmune response in insects, which involves crosstalks between the different pathways, concomitantactivation of multiple pathways and multipleresponses (FERRANDON et al., 2007).

THE CACTUS-LIKE ANK GENES OF PARASITOIDS

In parallel with our study of Drosophila deve -lopment, several years ago we begun a con structivecollaboration with the groups directed by E. Trem -blay and F. Pennacchio on the insect para sitoidToxoneuron nigriceps (Viereck) (Hymeno ptera,Braconidae) and its associated polydna virus (TnBV).

Several parasitic wasps, when ovipositing in hostinsects, inject factors that disrupt their physiology,

– 39 –

Scheme 1A simplified scheme of the NFκB pathways regulating the Dorso-Ventral embryonic patterning and the adult antimicrobialresponse.

development and immune reaction. Among thesefactors, the polydnaviruses (PDVs), obligatesymbionts of the parasitoids, play a pivotal role insuccessful parasitism. Polydnaviruses are inte -grated as proviruses in the genome of the para -sitoids and vertically transmitted through thegermline. The viral particles contain circular DNAsegments of different sizes, which excise andreplicate in the epithelium of the ovarian calyxand are injected in the host’s body at oviposition.The viruses enter various target tissues, where theyexpress a set of genes redirecting host physiology,thus allowing parasitoid development (for a reviewsee PENNACCHIO and STRAND, 2006).

To contribute to the study of the molecularbases of host-parasitoid interactions in insects, thestrategy we pursued was based on the sequencingof the TnBV genome and on the characterisationof the viral genes expressed in the host. We foundthat, among other genes (FALABELLA et al., 2003,MALVA et al., 2004, PROVOST et al., 2004), theTnBV genome contained three genes, located intwo DNA circles of 10.5 and 4.7 kb (Fig. 3),coding for proteins showing an average 30%

Figure 3Mapping of ank genes on the TnBV genome. Southern blotexperiments on undigested TnBV DNA using as probesTnBVank3 (a) and TnBVank1 (c) cDNAs. In (b) the ethidiumbromide stained gel of undigested TnBV DNA is shown toindicate the correspondence between the hybridization signalsand the DNA circles of the TnBV genome. Below the gels are indicated the linear maps of two circles of theTnBV genome, circle 10.3 of 10,292 bp and circle 4.7 of 4,734bp, where, by sequence analysis, the TnBVank genes have beenidentified. Circle 10.3 contained both TnBVank2 and TnBVank3genes while circle 4.7 contained TnBVank2. The Southern blotexperiments identified a band of 10.3 kb with the TnBVank3probe (a) and of 4.7 kb with the TnBVank1 cDNA, whichexperimentally confirmed the sequence data.

identity with Cactus and other IκB proteins fromseveral species. We called these genes TnBVank1,2 and 3 (FALABELLA et al, 2007). As describedabove, in Drosophila, Cactus regulates multiplecellular responses activated by the nuclear importof various NF-κB/Rel proteins, which controlembryonic D-V patterning and antimicrobialresponse. This central role of Cactus/IκB proteinsin the regulation of key signaling pathwaysprompted us to characterise the structure of theviral genes and their expression pattern inparasitized hosts. In Fig. 4 the structure of thepredicted TnBVank proteins is schematicallyreported and compared to that of DrosophilaCactus and mammalian IκBα and IκBβ. All viralproteins are small, ranging between 146 and 194aa, and consist of 3 ankyrin repeats. Notably,unlike other IκB homologue proteins, the viralproteins do not contain the conserved N-terminalmotif and the relative serine residues recognizedby the IκB kinase (IKK), which triggersCactus/IκB phosphorylation and degradation.Furthermore, none of the viral proteins containsthe C-terminal PEST domain, present in the otherCactus/IκB proteins, correlated with rapid proteinturnover.

The transcription profile of the three TnBVgenes showed that two genes, TnBVank1 andTnBVank3, are expressed very early afterparasitization in the haemocytes of H. virescenslarvae and, at much lower level, in the thorax,which contains abundant fat body, known to betargeted by TnBV. Thus, the transcription profilesare consistent with a possible role of TnBVankgenes in the suppression of the host immuneresponse shortly after parasitization.

– 40 –

Figure 4Structure of the putative TnBV ank proteins. Schematic repre -sentation of the proteins encoded by TnBVank genes, comparedwith Drosophila Cactus, human IκBa and mouse IκBb. Thesmall boxes correspond to the ankyrin repeats and the shortamino acid sequences reported are the IKK target region.Numbers on the right indicate the length of each protein.

The predicted sequence of the viral truncatedCactus-like proteins suggests that they might bindto NF-κB/Rel transactivators, competing withendogenous host IκB proteins. Since they cannotbe phosphorylated, they would behave asdominant inhibitors, retaining the transactivatorsin the cytoplasm and disrupting a number ofcellular pathways in parasitized hosts.

We tested this model with different approachesand we observed that NF-κB/Rel cellular pathwaysare disrupted in parasitized H. virescens larvae, thusproviding a molecular mechanism accounting forthe parasitoid-induced impairment of host immunedefences (FALABELLA et al., 2007).

Given the conservation of the Cactus/IκBstructure and function, we decided to test if theviral cactus-like genes act out of the parasitized hostcontext. We used the powerful Drosophilatransformation system that allows conditional andtissue specific expression of any gene of interest(BRAND and PERRIMON, 1993). We producedDrosophila transgenic lines carrying the TnBVank1gene and intend to investigate if the expression ofthis viral gene in different tissues may affectDrosophila development and-or immune response.

In conclusion we described a mechanism for theimpairment of NF-κB/Rel functions, mediated bythe TnBV-encoded IκB-like proteins, which playsa central role in the cyto-pathological effectsinduced in the host during parasitization. Thisconclusion is strongly supported by the fact thatsimilar genes have been described in otherbracoviruses and also in ichnoviruses (KROEMER

and WEBB, 2005). In addition, some viruses ofvertebrates (REVILLA et al., 1998) appear to usesimilar strategies to manipulate the NF-κBpathways and disrupt the immune response ofinfected organisms.

ACKNOWLEDGEMENTS

I would like to thank all the components of mygroup who have collaborated to this work in thecourse of the years, especially Silvia Gigliotti, whois continuing the projects on insect parasitoidsafter my retirement. Special thanks to E. Tremblay,who introduced me to the world of parasitoidswith an immense entomological culture, passion,open mind and humanity. He strongly supportedthe collaboration with F. Pennacchio, bringingtogether two separated worlds, and allowed us,with the exchange of different expertises, toachieve significant progress in our common effortsto understand the molecular mechanisms ofparasitization in insects.

RIASSUNTO

DALLA DROSOPHILA MELANOGASTERAGLI INSETTI PARASSITOIDI: PROCESSI DIVERSI,

MECCANISMI MOLECOLARI CONSERVATI.

Molti insetti parassitoidi, all’atto di depositare l’uovo nelloro insetto ospite, iniettano in esso fattori che ne alteranola fisiologia, lo sviluppo e la risposta immunitaria. Traquesti fattori i polydnavirus (PDV) sono simbiontiobbligati dei parassitoidi e sono indispensabili per ilsuccesso della parassitizzazione. Noi abbiamo dimostratoche il genoma del polydnavirus associato al Toxoneuronnigriceps (TnBV) codifica per forme troncate di proteineappartenenti alla famiglia Cactus/IκB. In Drosophila,Cactus regola l’importo nucleare di varie proteineappartenenti alla famiglia NFκB/Rel, che sono fattori ditrascrizione che controllano una serie di processi biologicifondamentali, tra cui lo stabilirsi dell’asse dorso-ventraledell’embrione e la risposta antimicrobica. La rispostaimmunitaria degli insetti ha molte tappe in comune con laimmunità innata dei mammiferi, compresa la viaintracellulare di trasduzione del segnale che attiva NFκB.Le proteine virali ank, simili a proteine cactus-IκBtroncate, potrebbero funzionare come inibitori dominantiche, mantenendo intrappolate nel citoplasma proteinedella famiglia NF-κB/Rel dell’ospite, ne alterano larisposta immunitaria. Il ruolo delle proteine Cactus/IκBsarà discusso come un esempio dell’uso reiterato di vie ditrasduzione di segnali molto efficienti e conservate nelcorso dell’evoluzione.

REFERENCES

BRAND A., PERRIMON N., 1993 – Targeted gene expressionas a means of altering cell fates and generating dominantphenotypes. - Development, 118: 401-415 .

CLIFFORD R., SCHUPBACH T., 1992 – The torpedo (DER)receptor tyrosine kinase is required at multiple timesduring Drosophila embryogenesis. Development, 115:853-872

FALABELLA P., VARRICCHIO P., GIGLIOTTI S., TRANFAGLIAA., PENNACCHIO F., MALVA C., 2003 – Toxoneuronnigriceps polydnavirus encodes a putative aspartylprotease highly expressed in parasitized host larvae. -Insect Molecular Biology, 12: 9-17

FALABELLA P., VARRICCHIO P., PROVOST B., ESPAGNE E.,FERRARESE R., GRIMALDI A., DE EGUILEOR M., FIMIANIG., URSINI M.V., MALVA C., DREZEN J.M., PENNACCHIOF., 2007 – Characterization of the IkappaB-like gene familyin polydnaviruses associated with wasps belonging todifferent Braconid subfamilies. - J. Gen. Virol., 88: 92-104.

FERRANDON D., IMLER J., HETRU C., HOFFMANN J., 2007 –The Drosophila systemic immune response: sensing andsignaling during bacterial and fungal infections. - Nature,7: 862-874.

FULLER M.T., SPRADLING A.C., 2007 – Male and femaleDrosophila germline stem cells: two versions ofimmortality. Science 316 (5823).

GONZALES-REYES A., ELLIOT H., ST JOHNSTON D., 1995 –Polarization of both major body axes in Drosophila bygurken-torpedo signalling. - Nature, 375: 654-658.

GHOSH S., MAY M. J., KOPP E. B., 1998 – NF-κB and Relproteins: evolutionarily conserved mediators of immuneresponses. - Annu. Rev. Immunol., 16: 225-260.

HOFFMANN J.A., REICHHART J.M., 2002 – Drosophilainnate immunity: an evolutionary perspective. - Nat.Immunol., 3: 121-126.

HOFFMANN J.A., 2003 – The immune response ofDrosophila. - Nature, 42: 33-38.

KROEMER J.A., WEBB B.A., 2005 – IκB-related vankyringenes in the Campoletis sonorensis ichnovirus: temporaland tissue-specific patterns of expression in parasitizedHeliothis virescens lepidopteran hosts. - J. Virol., 79:7617-7628.

MALVA C., VARRICCHIO P., FALABELLA P., LA SCALEIA R.,GRAZIANI F., PENNACCHIO F., 2004 – Physiological andmolecular interaction in the host-parasitoid systemHeliothis virescens-Toxoneuron nigriceps: current statusand future perspectives. - Insect Bioch. Mol. Biol., 34:177-183.

MOUSSIAN B., ROTH S., 2005 – Dorsoventral AxisFormation in the Drosophila Embryo: Shaping andTransducing a Morphogen Gradient. - Current Biology,15: 887-899.

PENNACCHIO F., STRAND M. R., 2006 – Evolution ofdevelopmental strategies in parasitic hymenoptera. -Annu. Rev Entomol., 51: 233-258.

PROVOST B., VARRICCHIO P., ARANA E., ESPAGNE E.,FALABELLA P., HUGUET E., LA SCALEIA R., CATTOLLICOL., POIRIÉ M., MALVA C., OLSZEWSKI J.A., PENNACCHIOF., DREZEN J.M., 2004 – Bracoviruses contain a largemultigenic family coding for Protein TyrosinePhosphatases. - J. of Virology, 78: 13090-13103.

REVILLA, Y., CALLEJO, M., RODRIGUEZ, J. M., CULEBRAS,E., NOGAL, M. L., SALAS, M. L., VINUELA, E. FRESNO,M., 1998 – Inhibition of Nuclear factor κB activation bya Virus-encoded IκB-like protein. -The Journal ofBiological Chemistry, 273: 5405-5411.

ROTH S., HIROMI Y., GODT D., NUSSLEIN-VOLHARD C.,1991 – Cactus a maternal gene required for properformation of the dorsoventral morphogen gradient inDrosophila embryos. - Development, 112: 371-388

SILVERMAN N., MANIATIS T., 2001 – NF-κB signalingpathways in mammalian and insect innate immunity. -Genes Dev., 15: 2321-2342.

SPRADLING A.C., 1993 – Developmental genetics ofoogenesis. In: Drosophila Development. - (Eds. M. Bateand A. Martinez-Arias). Cold Spring Harbor Press,CSH, N.Y., pp. 1-70.

WU L.P., ANDERSON K.V., 1998 – Regulated nuclear importof Rel proteins in the Drosophila immune response. -Nature, 392: 93-97.

– 41 –


(*) Department of Biology, University of Western Ontario, London N6A 5B7, CanadaLettura tenuta nella Seduta pubblica dell’Accademia - Firenze, 6 giugno 2008.

Evolution of polyembryonic development in parasitic waspsMajor developmental innovations have been associated with adaptive radiations that have allowed particular groups of

organisms to occupy empty ecospace. However, an understanding of the evolutionary forces and molecular mechanisms behinddevelopmental novelties still remains tenuous. A little studied adaptive radiation in insects from the developmental perspective isthe evolution of parasitism. The parasitic lifestyle has allowed parasitic insects to occupy a novel ecological niche where they haveevolved a plethora of life history strategies and modes of embryogenesis, developing on or within the body of the host. One ofthe most striking adaptations to development within the body of the host includes polyembryonic development, where certainwasps form clonally up to 2000 embryos from a single egg. Taking advantage of well-established insect phylogeny, techniquesdeveloped in a model insect, the fruit fly, and a wealth of knowledge in comparative insect embryology, we are starting to teaseapart the evolutionary events that have led to this novel mode of development in insects.

KEY WORDS: embryogenesis, evolution of parasitism, molecular mechanisms.

EVOLUTION OF POLYEMBRYONIC DEVELOPMENT IN PARASITIC WASPS

MIODRAG GRBIC’ (*)

INTRODUCTION

Polyembryonic development represents the for-mation of multiple embryos from a single zygote.The accidental form of polyembryonic develop-ment, where an individual egg occasionally formsmultiple embryos, has been described in almost allanimal groups studied to date (OLSEN 1962; STANS-FIE, 1968; KAUFMAN, 1982; LAALE, 1984; ASHWORT

et al., 1998). This accidental form of polyembryonysuggests that eggs of otherwise monoembryonicspecies have the regulative capacity to generatemultiple embryos. On the other hand, obligatorypolyembryonic development, where a single zygoteof certain species invariably produces multipleembryos, is a relatively rare event in metazoans, butquite frequent in plants (SHAANKER and GANESHA-IAH, 1996; CARMAN, 1997). In metazoans, obligato-ry forms of polyembryonic development are pres-ent in both vertebrates and invertebrates. Speciesexhibiting polyembryonic development are scat-tered in multiple phyla including Cnidaria, Platy-helmintes, Arthropoda, Bryozoa, Echino dermataand Chor data (reviewed in CRAIG et al., 1997). Itshould be noted that in certain groups, the sourceof clones is not the embryo but the larva, as in alldescribed cases of polyembryony in the philaCnidaria and Echinodermata, and in Cestodea andTrematoda (Platyhelmintes) and Crustacea(Arthro poda) (NOBLE et al., 1989; SHOSTAK, 1993;GLENNER and HOEG, 1995; JAECKLE, 1994).

The focus of this review is obligatory polyem-bryony in insects that arises by embryonic cloning.The term polyembryony denotes both the develop-mental process, and the form of reproduction.Developmental processes include complex cellu-lar and molecular events whereby multipleembryos form clonally from a single zygote. Inaddition, polyembryony refers to a unique form ofreproduction in which a single egg results in mul-tiple progeny, maximizing the reproductive capac-ity of the species and increasing its fitness. Alongwith its ecological and reproductive ramifications,study of the phenomenon of polyembryony ininsects has the potential for addressing one of cru-cial questions in the evolution of development:How do developmental novelties arise? Polyem-bryony in insects represents a developmental nov-elty whereby both precursor structure and evolu-tionary processes are basically unknown (type Anovelty sensu WILKINS, 2001). In general, truedevelopmental novelties are rare and often theirevolution is not easily tractable. However, thecombination of a relatively well-established insectphylogeny, embryological studies of insect poly-embryony that span more than a century (MAR-CHAL, 1898), and techniques and concepts estab-lished in a closely-related model Arthropod,Drosophila melanogaster, demonstrate a promis-ing system that could provide clues as to howcomplex developmental novelties are formed.

MULTIPLE EVENTS OF INDEPENDENT EVOLUTION

OF POLYEMBRYONIC DEVELOPMENT IN WASPS

Hymenoptera (wasps) represents a holometa -bolous insect order that consists of two suborders.Suborder Symphita includes basal plant-eatinggroups, and Apocrita, an advanced group of para-sitic species (Figure 1). Hymenoptera posses poly-trophic meroistic ovaries (BUNING, 1994) and basalgroups produce yolky eggs which undergo longgermband embryogenesis (SPEICHER, 1936; FLEIG

and SANDER, 1986, reviewed in SANDER 1976).Apocrita (parasitic wasps plus ants and bees) repre-sents a monophyletic assemblage which includesectoparasitic species (laying the egg on the surfaceof the host), endoparasitic species (ovipositingwithin the body of the host), and free-living pollina-tors including eusocial species (WHITFIELD, 1998).Basal species in all parasitic groups whose life histo-ries are known appear to be ectoparasitic. They laylarge yolky eggs, and undergoing long germbanddevelopment, such as described in the honeybee(FLEIG, 1990; BINNER and SANDER, 1997) and theendoparasitic basal braconid Bracon hebetor(GRBIC and STRAND, 1998). This suggests that thebasal state of embryonic development in parasiticwasps includes canonical long germband develop-ment associated with meroistic polytrophic oogen-

esis, where critical determinants are transcribed innurse cells and transported to the oocyte in a man-ner described in Drosophila. However, many para-sitic lineages contain parasitic species that haveevolved a derived form of development within thebody of the host (endoparisites). This switch in lifehistory strategy subjects them to a different selec-tion regime compared to other terrestrial insects.The evolution of endoparasitism appears to be cru-cial for further evolutionary innovations, such aspolyembryony. Polyembryony evolved independ-ently four times in wasps: in Braconidae, Encyr-tidae, Dryinidae and Platygasteridae (IVANOVA-KAZAS, 1972). The association of endoparasitclifestyle with evolution of polyembryony isstrengthened by the fact that the only other case ofpolyembryony in insects is displayed by endopara-sitic Strepsiptera (NOSKIEWICZ and POLU SZYNSKI,1935).

POLYEMBRYONIC EMBRYOGENESIS:EMBRYOLOGICAL INNOVATIONS

Independent evolution of polyembryony evokesseveral important questions. First, what is qualita-tively novel in polyembryonic development relativeto canonical insect embryogenesis? Second, which

– 44 –

Figure 1Phylogeny of Hymenoptera (modified from Whitfield 1998). Families that display polyembryonic development are high-lighted by gray shading. Drawing of the egg illustrates long germband development of more primitive basal Hymenoptera.

elements of the regulatory mechanisms were modi-fied to result in a novel, obligatory form of embryocloning? Finally, understanding such independent-ly evolved, but similar novelties could inform usabout evolutionary constraints and plasticity. Forexample, are there multiple pathways in the evolu-tion of certain features, or are similar evolutionaryinnovations based on a common program?

Thus far, our model insect for polyembryonicdevelopment has been the polyembryonic encyrtidCopidosoma floridanum (SILVESTRI, 1906; GRBIC etal., 1996; GRBIC et al., 1998, ZHUROV et al., 2004).This wasp parasitizes noctuid moths and producesup to 2000 embryos from a single egg. However, apoor understanding of encyrtid phylogeny and alack of knowledge of closest monoembryonicancestors led us to initiate studies on another inde-pendently-evolved polyembryonic wasp, the bra-conid Macrocentrus grandii. A better understand-ing of the phylogeny of braconids could help us todetermine the closest monoembryonic relatives,and to generate a hypothesis about transitory formsthat may have led to polyembryonic development.In addition, studies of multiple forms of polyem-bryony could uncover common features and possi-ble variations in polyembryonic development.

POLYEMBRYONY IN COPIDOSOMA: A CHALLENGE

FOR THE DROSOPHILA PARADIGM OF DEVELOPMENT

Copidosoma floridanum is a parasitic wasp thatparasitizes the eggs of the host, the moth Trichoplu-sia ni (Fig. 2). After parasitization, the host emergesand undergoes five host instars. During the processof host development Copidosoma undergoesembryonic development within the host body sur-rounded by the nutritive insect blood (heamo -lymph). As a result of embryonic proliferation, upto two thousand larvae are formed synchronouslyduring the fifth host instar. These larvae pupateand emerge as adult wasps.

The embryonic development of this endopara-sitic insect differs dramatically from the develop-ment of other insects. First, Copidosoma ovipositstiny yolkless eggs (50mm in size, a size similar tomouse eggs) that are surrounded by a thin chori-on. The first cleavage of the egg is total, and leadsto the formation of two posterior blastomeres(which will give rise to the embryo proper) and ananteriorly localized polyploid cell (Fig. 2 gray)that results from the fusion of polar nuclei (GRBIC

et al., 1998). This cell will form the polyploid syn-cytial extraembryonic membrane (gray). The sec-ond embryo cleavage creates one small blas-tomere and three equal-size blastomeres. The

– 45 –

Figure 2Embryonic development of Copidosoma floridanum and its hostTrichoplusia ni.. A) Embryonic development of Copidosoma. Polarbody and polar body-derived extraembryonic membrane – grey;germ line (PGC cells) – red; proliferating morulae without PGC(progenitors of the precocious embryos) blue; L1-L5 larval instarsof T. ni.

small cell (Fig 2 red) is different from the othercells as it retains an injected fluorescent tracer,and is thus dye-uncoupled from other cells (GRBIC

et al., 1996). Embryonic blastomeres then under-go cleavages and become enveloped by the syncy-tial extraembryonic membrane, and embryosemerge from the chorion into the hosthaemolymph and form primary morula. The pri-mary morula implants in the host tissue and initi-ates the proliferative phase of development thatincreases cellular mass many fold (GRBIC et al.,1998). In monoembryonic animals, developmen-tal progression from that point in embryogenesiswould include a transition from morula-stage

embryo to gastrulation and segmentation leadingto a completely segmented animal. In contrast,«insertion» of the proliferative phase in thecanonical monoembryonic developmental pro-gram represents the developmental noveltyresponsible for clonal production of thousandembryos in Copidosoma. The proliferative phase isinitiated by the split of the primary morula and cre-ation of the polymorula, which consists of manyproliferative morulae. Each proliferative morula atthis stage consists of hundreds of round, apparent-ly non-differentiated cells (Fig. 2), surrounded bythe extraembryonic membrane (GRBIC et al.,1998). These packages of cells become subdividedby the ingressing extraembryonic membrane intoprogressively smaller clusters of cells. When thenumber of cells per cluster reaches about 20-30 atthe fourth host instar larva, these cells undergo achange in cell shape from round to fibroblastic.The establishment of cell contacts results in cellcompaction and simultaneous de novo formationof 2000 embryonic primordia (GRBIC et al., 1998).Follo wing compaction, each embryo undergoesgastrulation and segmentation to form larva. Thus,polyembryonic embryogenesis in Copidosomashows similarities to mammalian embryonic devel-opment, including early separation of embryonicand extraembryonic lineages, morula morphology,implantation, compaction and most importantlythe net increase of the embryonic mass that isunique to mammals (DAVIDSON, 1990; GURDON,1992). These evolutionary changes apparently rep-resent convergent evolution driven by the similardevelopmental environment: placental develop-ment in mammals and nutritive host environmentin Copidosoma. Obligatory polyembryony evolvedin mammals (armadillo) (FERNANDEZ, 1909) andinsects (parasitic wasps), but while in mammalspolyembryony is conceptually compatible with theregulative development of the mouse embryo,polyembryony in insects is in sharp contrast withmaternal pre-patterning of the Drosophila embryo.

MATERNAL PRE-PATTERNING IN COPIDOSOMA:SPECIFICATION OF THE GERM-LINE

The germ line is one of first developmental fatesspecified in many organisms (SAFFMAN and LASKO,1999). The RNA helicase vasa is the ubiquitousgerm line marker in metazoans involved in thespecification of the primordial germ cell (PGC) lin-eage. PGCs represent the first cells that give riseexclusively to germ cells by clonal mitotic divisions(NIEUWKOOP and SUTASURYA, 1979). PGCs areprogenitors of germ line stem cells (GSCs) that

undergo self-renewal, differentiate into gametes,and ultimately produce all of the cell types in futureoffspring.

Isolation of the Copidosoma vasa mRNA (Cfvas)homologue and examination of its pattern ofexpression showed that Cfvas is transcribed innurse cells in Copidosoma ovaries (ZHUROV et al.,2004). Vasa protein localizes in the structure calledoosome (ZHUROV et al., 2004), which was proposedto be homologous to the Drosophila germ line poleplasma (nouage). Thus, in Copidosoma at least oneasymmetrically localized maternal determinant isdeposited in the forming egg. After ovipositionVasa protein is invariably localized into the dye-uncoupled small cell, showing that this early cellu-lar asymmetry is also paralleled by a molecularasymmetry. In the primary morula this asymmetry isperpetuated in several cells that express Vasa pro-tein (Fig 2, red). Following the initiation of the pro-liferative phase Vasa-positive cells are scattered inindividual proliferating morulae (Figure 2). Duringthe process of division, the daughter cells all inheritVasa protein, suggesting that they represent celllineage. Following the entrance into the morpho-genetic phase each reproductive embryonic pri-mordium receives two Vasa-positive cells. Thesecells remain localized at the posterior and give riseto the embryonic gonads (Fig. 2). Thus, maternalcellular asymmetry marked by the expression ofVasa protein perpetuates throughout the prolifera-tive phase and becomes continuous with the germline, suggesting that Copidosoma specifies the PGCmaternally.

DEVELOPMENT OF THE PRECOCIOUS EMBRYOS:DIFFERENTIAL DISTRIBUTION OF PGC SPECIFIES

THE REPRODUCTIVE POTENTIAL

While it was known that the reproductive larvaegive rise to adults and have a reproductive function,the reproductive potential of the precocious larvaein Copidosoma was uncertain. They do not moltand become consumed by their reproductive sib-ling. This poses the question of whether they have areproductive potential that is simply not realizeddue to their premature death or they entirely lackpotential for the reproduction. In socialhymenoptera, workers are sterile in contrast to fer-tile queens. However, this sterility is often condi-tional. Both queens and workers have germ lineprogenitor cells, but in queens the reproductiveapparatus become hypertrophic while in workersovarioles degenerate (SCHMIDT CAPELLA andHARTFELDER, 2002). However, in some cases work-ers can restore their reproductive potential and

– 46 –

become reproductives (NIJHOUT, 1999). During theproliferative phase in Copidosoma it was noticedthat some proliferative morulae do not containVasa-positive cells (Fig. 2 blue). These morulaeundergo differentiation and give rise into the pre-cocious embryos that do not inherit PGCs. Thus,the mechanism based on segregation of PGC line-age in reproductives, and the failure of the preco-cious embryos to inherit PGCs represents novelcell-sorting mechanism that specifies the caste fate.This mechanism specifies in all-or-none fashion adifferent reproductive capacity in genetically iden-tical embryos.

FUNCTION OF PGCS IN COPIDOSOMA:CASTE FATE AND PROLIFERATION

In contrast to Drosophila, where PGCs undergomigration through the embryo to reach their posi-tion in future gonads (UNDERWOOD et al., 1980),Copidosoam PGCs undergo a complex journeywhich includes cell parceling during the prolifera-tive stage, differential segregation to two castes andfinal localization at the embryonic gonads (ZHUROV

et al., 2004). Clearly PGCs must be involved in theformation of the germ-line, but their complexontogeny poses the question of whether they haveother functions in polyembryonic development?Besides Vasa, these cells likely contain many otherdeterminants that may have a role in embryo germcell specification, proliferation or caste fate. Onepossibility is that these cells have a cell-autonomousfunction in specifying the germ-line as in Drosophi-la. This scenario predicts that the removal of PGCprogenitor cell will result in formation of the repro-ductive larvae without gonads. Alternatively, thiscell could have non-cell autonomous function(s) sothat the germ line specification is coupled withother developmental processes. Laser ablation ofVasa-positive cell at the four cell stage (red Fig 1)has revealed that it has multiple functions (ZHUROV

et al., 2004, DONNELL et al., 2004) As a conse-quence of ablation Copidosoma reproductiveembryos did not proliferate as detected by lack offormation of the reproductive embryos (ZHUROV etal., 2004). However, the precocious larvae develop-ment was not affected, resulting in normal num-bers. Laser ablation of the Vasa-positive cellreduced 95% of polyembryonic proliferation. Incontrast, ablation of any of the large blastomeres atthe same stage (Fig. 2 white cells) restores thedevelopment and proliferation of reproductiveembryos (ZHUROV et al., 2004). This suggests thatthe PGC progenitor has a dual function: it regu-lates proliferation and the reproductive caste fate.

TRANSITORY STEPS PRECEDING

POLYEMBRYONIC DEVELOPMENT

In order to address the question of evolution ofthe polyembryonic development it is necessary toturn to the system that preceded the evolution ofpolyembryonic development and to look at thedevelopment in the closest monoembryonic ances-tor. The putative ancestor has to be an endopara-sitic wasp. Second, it should undergo total eggcleavage. Finally, it should emerge from the chorioninto the host hemocoel, and should utilize the polarbody-derived cell to form the extraembryonicmembrane surrounding the embryo.

The braconid endoparasite Aphidius erviexhibits the predicted features of the hypothesizedancestor of polyembryonic wasps. This wasp laystiny transparent eggs that undergoe total cleavage .Its embryo emerges from the egg shell into the hosthemocoel and remains enveloped by the polarbody-derived extraembryonic membrane. Follow-ing the emergence from the chorion, morphogene-sis is initiated by the formation of an embryonicprimordium that consists of a solid ball of cells,similar to the Copidosoma embryonic primordium.The embryo of Aphidius initially forms just theanterior structures of the embryo. The rest of thetrunk is formed by sequential proliferation ,exhibiting characteristic short germband develop-ment. Since basal braconids display longgermband development, Aphidius developmentrepresents secondarily derived short germbandembryogenesis.

The embryogenesis of polyembryonic parasiticbraconid Macrocentrus grandii wasps could help usto understand how polyembryony evolved in adefined phylogenetic context of braconids. Eggs ofthis species is also transparent, surrounded by tinychorion and in contrast to those of their basal,ectoparasitic relatives, contain almost no yolk. Ini-tial cleavage events in these tiny eggs differ from thecanonical type of insect syncytial cleavage. Bothwasps undergo total (holoblastic) cleavage in whichnuclear division is immediately followed by cyto-plasmic division, forming individual cells (blas-tomeres). This novel type of early cleavage appearsto be common also in polyembryonic platygasterids(IVANOVA-KAZAS, 1972), and its general presence inall polyembryonic species suggests that it repre-sents a prerequisite for the evolution of polyembry-onic development.

Following early cleavages, Macrocentrus embryosemerge from the tiny chorion into the host haemo-coel and enter the proliferative phase of develop-ment. In this phase, the number of cells increasesand cells become subdivided by the extraembryon-

– 47 –

ic membrane into several independent spatialdomains. The proliferative stage in Macrocentrusresults in a smaller rate of proliferation than inCopidosoma to form ultimately up to 20 embryos.Finally, initiation of the morphogenetic phaseresults in the formation of the embryonic primordi-um. In both species embryonic primordia areformed from the very beginning as cellularizedstructures. However, in Copidosoma the embryonicprimordium is solid, without the blastocoel (GRBIC

et al., 1996), while the Macrocentrus primordiumconsists of single layer of cells that surrounds thehollow space of the blastocoel. These species alsodiffer in the type of germband. Copidosomaembryogenesis was hard to classify. It more resem-bled long germband development by its propor-tional growth and expression of molecular markers(GRBIC et al., 1996). On the other hand, Macrocen-trus embryogenesis is clearly of a short germbandtype. The initial primordium consists of anteriorstructures and the remaining trunk is generated byposterior growth.

The comparison of development in two inde-pendently evolved polyembryonic species andtheir putative monoembryonic ancestor suggeststhat evolution of polyembryony is compatiblewith meroistic ovarial apparatus present in basalmonoembryonic wasps. On the other hand, inno-vations that are conserved in both polyembryonicspecies include a novel type of cleavage, and theproliferative phase responsible for creation ofmultiple embryos. It appears that in both polyem-bryonic wasps the proliferative phase has beensimply «inserted» into the monoembryonic devel-opmental program without any consequences forthe later phases of development. Even though theproliferative phase seems to be similar in specificembryological events but different in the amountof proliferation, the late morphogenetic phase dis-plays two completely different trajectories. InCopidosoma three-dimensional tissue specificationproceeds from the morphogenesis of a solid ballof cells, resembling the long germband type ofembryogenesis (GRBIC et al., 1996). In contrast,the Macrocentrus primordium forms a single celllayer, and extension of the embryo trunk repre-sents a form of short germband development, asdescribed in primitive insects. Even though shortgermband development is considered to be aprimitive remnant of insect ancestors, its secon-darly-derived development in Macrocentrus indi-cates that the evolutionary trajectory can beinverted: short germband development can evolvefrom a long germband ancestor.

Collectively, descriptions of embryogenesis inthese wasps illustrate the surprising level of plastici-

ty and modularity of developmental programs.First, meroistic polytrophic ovaries that synthesizedeterminants for syncytial cleavage and longgermband development in Drosophila are compati-ble with specification of determinants for polyem-bryonic development. Second, innovations in thecleavage type and proliferative phase which shouldtheoretically scramble Drosophila localized mater-nal determinants and diffusion-based action of thetranscription factors are perfectly compatible withde novo formation of thousands of embryonic axesmany days after oviposition. On the other hand,these multiple independent evolutionary events ofpolyembryony suggest that evolution of such acomplex developmental program could have a rela-tively simple genetic basis that includes changes invery few genes.

SCENARIOS FOR EVOLUTION OF POLYEMBRYONY

An analysis of multiple independent events ofpolyembryony in wasps within the phylogeneticframework suggests that it consists of a complexand stepwise processes. The ancestral type of devel-opment in all polyembryonic lineages included anectoparasitic life history strategy and a large yolkyegg, exhibiting long germband embryogenesis.With the evolution of endoparasitism, waspembryos gained the advantage of exploiting thenutritive environment of the host not only for larvalfeeding, but also for embryo development. Thisshift resulted in several changes in egg architecture.First, the chorion which consists of elaborate struc-tures in ectoparasites and other terrestrial insectsprotecting them from dessication, decreased in itscomplexity once the embryo evolved emergencefrom the chorion into the host nutritivehaemolymph. In addition, because host nutrientswere utilized for embryo development it was notnecessary to stockpile a large amount of yolk in theeggs. Consequently, endoparasitic egg sizedecreased. In smaller eggs evolution favoured anew type of cleavage: total cleavage, immediatelyforming individual cells.

It is unique that in many endoparasitic waspspolar nuclei do not degenerate as in other terrestri-al insects (TREMBLAY and CALVERT, 1972). Instead,they participate in the formation of extraembryon-ic membranes that completely surround theembryo. It appears that this structure evolvedmany new functions in contrast to the extraembry-onic membranes in terrestrial insects. In manyendoparasitic wasps, at the completion of morpho-genesis the extraembryonic membrane fragmentsinto individual polyploid cells called teratocytes. In

– 48 –

some endoparasitic wasps teratocytes circulate inthe host hemolymph and synthesize proteinswhich are secreted, altering host physiology in sup-port of endoparasitic development (RANA et al.,2002). However, in the polyembryonic embryoge-nesis of Copidosoma, the extraembryonic mem-brane is involved in the proliferative phase ofdevelopment, separating proliferative cells intospatial domains. It never fragments to form the ter-atocytes and continues to surround both embryosand larvae. Even though endoparasitic embryoscan take advantage of the host nutritive environ-ment, they must first evolve a defense against thehost immune system. Findings by CORLEY andSTRAND (2003) that the extraembryonic membranein Copidosoma protects the larvae from the hostimmune system may provide a clue as to the pri-mary reason for the evolution of this structure. Inaddition, it has been proposed that the polar cell-derived extraembryonic membrane plays a role inthe uptake of nutrients from the host haemolymph(KOSCIELSKI and KOSCIELSKA, 1985). Analyzingthe expression pattern of genes in the proliferativephase of development, it was determined that allcells of the extraembryonic membrane in Copido-soma express alkaline phosphatase mRNA (TERZIN

and GRBIC, unpublished). This enzyme is involvedin nutrient absorption and transport mechanismsin insects and vertebrates (EGUCHI, 1995), suggest-ing that the extraembryonic membrane activelyabsorbs nutrients from the host heamolymph.Thus, the primary role of the extraembryonicmembrane initially was probably to protect theemerged embryo of monoembryonic endopara-sites against the host immune system, and toabsorb nutrients. Later on, the existing structurewas likely co-opted to the proliferative phase ofembryogenesis in polyembryonic insects to organ-ize proliferative growth.

Evolution of small egg size, total cleavage, andnovel, multifunctional extraembryonic mem-branes were the prerequisites for the evolution ofthe novel proliferative stage. This stage representsthe true developmental innovation (Type A)because it was derived from novel structures (theextraembryonic membrane) and a cleavage typethat does not have a known precursor in ancestral,ectoparasitic, insects. It is hard to conceptualizethe evolution of a novel stage that disrupts one ofthe crucial paradigms of Drosophila development,maternal specification of the embryonic axis,while at the same time creating de novo 2000 inde-pendent embryonic axes! If the syncytial environ-ment of the Drosophila pre-blastoderm embryohas created complications in understanding howpattern formation proceeds in the cellular milleu

of short and intermediate germband insects(WILKINS, 2001), then polyembryonic develop-ment represents a real challenge for the Drosophilaparadigm. One of first prerequisites for such anevent appears to be the uncoupling of posteriorpatterning and germ cell specification. The secondstep should include the initiation of the prolifera-tion mechanisms to generate at least 40,000 cellsnecessary for initiation of 2000 embryonic primor-dia (GRBIC at al., 1998). There are several relativelysimple possible means how to initiate prolifera-tion. In the monoembryonic ancestor cleavagesmust generate enough cells for the formation ofthe single embryonic primordium. At this pointproliferation has to stop and become coupled withaxial patterning. Thus, a simple change in the reg-ulatory region of the mitogenic signal couldextend the period of proliferation necessary forpolyembryonic development. Another avenuegenerating the same effect would be to produce amutation in the putative suppressor of prolifera-tion that terminates early proliferation and regu-lates entry into the blastoderm stage of themonoembryonic ancestor. Both of these changesare relatively simple and could involve existinggenes without requiring new gene recruitment(WILKINS, 2001). In a likewise manner, removal ofthe mitogenic signal by a similar mechanism at thecompletion of proliferation could regulate the exitfrom the proliferative stage.

It is hard to conceptualize how is the proliferativestage integrated with de novo establishment ofembryonic axes. All 2000 embryo axes appear toform independently with random axial orientationrelative to each other (GRBIC et al., 1996). Thisfavours an independent specification of the axialpolarity within each embryo rather than a globalmechanism specifying simultaneous polarity in2000 embryos.

CONCLUDING REMARKS

Evolution of developmental novelties is a com-plex phenomenon that requires understanding ofboth the ecological processes and developmentalmechanisms responsible for its creation. Analysis ofthe evolution of polyembryonic development with-in the phylogenetic context, and studies of multipleindependent events of polyembryony have beenimportant stepping stones toward beginning tounderstand the processes and mechanisms shapingthe evolution of this novel form of development. Asstated by WILKINS (2001), there is no general ana-lytical method that can be applied to all develop-mental novelties. However, clues derived from a

– 49 –

broader phylogenetic context suggesting the polari-ty of change and an examination of possible ances-tral states are essential in constructing testablehypotheses.

RIASSUNTO

EVOLUZIONE DELLO SVILUPPOPOLIEMBRIONICO DELLE VESPE PARASSITE

La comparsa di nuovi ed innovativi meccanismi di svilup-po embrionale è stata associata a radiazioni adattative chehanno permesso a gruppi particolari di organismi di occu-pare ecosistemi vuoti. Tuttavia, le forze evolutive ed i mec-canismi molecolari responsabili dell’origine di queste nuovemodalità di sviluppo sono poco studiati o ancora del tuttosconosciuti. L’evoluzione del parassitismo negli insetti, adesempio, è un meccanismo adattativo molto interessantema poco approfondito nel campo della embriogenesi. Gliinsetti parassitoidi, sviluppandosi su o dentro il corpo del-l’ospite, hanno evoluto una pletora di strategie di vita emodalità di sviluppo embrionale. Uno dei più affascinantiadattamenti allo sviluppo che avviene all’interno del corpodell’ospite è rappresentato dalla poliembrionia presente inalcune vespe in cui, da un singolo uovo, si formano clonal-mente fino a 2000 embrioni. Traendo vantaggio dalle tec-niche sviluppate in un insetto modello, la Drosophilamelanogaster, e da una grande quantità di conoscenze sullaembriologia comparata degli insetti, stiamo cercando diaffrontare gli eventi evolutivi che hanno portato a questanuova modalità di sviluppo negli insetti.

LITERATURE

ASHOWORT C.J., ROSS A.W., BARRETT P., 1998 – The use ofDNA fingerprinting to assess monozygotic twinning inMeishan and Landrace x Large White pigs. Reprod. Fertil.Dev., 10 (6): 487-490.

BINNER P., SANDER K., 1997 – Pair-rule patterning in thehoneybee Apis mellifera: expression of even-skipped com-bines traits known from beetles and fruit fly. Dev. Genes.Evol., 206: 447-454.

BUNING J., 1994 – The Insect Ovary. Kluwer AcademicPublishers. 416pp.

CARMAN J.G., 1997 – Asynchronous expression of duplicategenes in angiosperms may cause apomixis, bispory,tetraspory, and polyembryony. Biol. J. Linn. Soc. Lond.,61 (1): 51-94.

CORLEY L.S., STRAND M.R., 2003 – Evasion of encapsulationby the polyembryonic parasitoid Copidosoma floridanumis mediated by a polar body-derived extraembryonic mem-brane. J. Invertebr. Pathol., 83 (1): 86-89.

CRAIG S.F., SLOBODKIN L.B., WRAY G.A., BIERMANN C.H.,1997 – The ‘paradox’ of polyembryony: A review of thecases and a hypothesis for its evolution. Evol. Ecology, 11(2): 127-143.

DAVIDSON E.H., 1990 – How embryos work: a comparativeview of diverse mode of cell fate specification. Develop-ment, 108: 365-389.

DONNELL D.M., CORLEY L.S., CHEN G., STRAND M.R.,2004 – Caste determination in a polyembryonic waspinvolves inheritance of germ cells. Proc. Natl. Acad. Sci.USA., 101: 10095-10100.

EGUCHI M., 1995 – Alkaline-phosphatase isozymes in insects

and comparison with mammalian enzyme. Comp.Biochem. Physiol., 111 (2): 151-162.

FERNANDEZ M., 1909 – Beitrage zur Embryologie derGurteltiere. I. Zur Kaimblatter-inversion und spezifischenPolyembryonie der Mulita (Tatusia hybrida Desm.).Morph. Jahrb., 39: 302-333.

FLEIG R., 1990 – Engrailed expression and body segmenta-tion in the honeybee Apis mellifera. Roux’s Arch. Dev.Biol., 198: 467-473.

FLEIG G R., SANDER K., 1986 – Embryogenesis of the honey-bee Apis mellifera L. (Hymenoptera : Apidae): an SEMstudy. Int. J. Ins. Morph. Embryol., 15: 449-462.

GLENNER H., HOEG J.T., 1995 – A new motile, multicellularstage involved in host invasion by parasitic barnacles (Rhi-zocephala). Nature, 377 (6545): 147-150.

GRBIC’, M., NAGY L., STRAND M., 1998 – Development ofpolyembryonic insects: a major departure from typicalinsect embryogenesis. Dev. Genes Evol., 208: 69-81.

GRBIC’ M., STRAND M.R., 1998 – Shifts in the life history ofparasitic wasps correlate with pronounced alterations inearly development. Proc. Natl. Acad. Sci. USA, 95: 1097-1101.

GRBIC’ M., NAGY L.M., CARROLL S.B., STRAND M.R., 1996– Polyembryonic development: insect pattern formation ina cellularized environment. Development,12: 795-804.

GURDON J.B., 1992 – The generation of diversity in patternin animal development. Cell, 68: 185-199.

IVANOVA-KAZAS O.M,. 1972 – Polyembryony in insects. In:Counce SJ, Waddington CH. Ed; Developmental Sys-tems: Insects. Academic Press. p 243-271.

JAEKLE W.B., 1994 – Multiple-modes of asexual reproductionby tropical and subtropical sea star larvae - an unusualadaptation for genet dispersal and survival. - Biol. Bul., 186(1): 62-71.

KAUFMAN M.H., 1982 – 2 examples of monoamnioticmonozygotic twinning in diploid parthenogenetic mouseembryos. J. Exp. Zool., 224 (2): 277-282.

KOSCIELSKA M.K., KOSCIELSKI B., 1985 – Ultra structuralStudies on the polyembryony in Ageniaspis fuscicollis(Chalcidoidea, Hyme noptera). Zool. Poloniae, 32: 203-215.

LAALE H.W., 1984 – Polyembryony in teleostean fishes - dou-ble monstrosities and triplets. J. Fish Bio., 24 (6): 711-719.

MARCHAL P., 1898 – Dissociation de l’oeuf en un cycle evolu-tif chez l’Encyrtus fuscicollis (Hymenoptera). Compt.Rend. Acad. Sci. Paris; 126: 662-664

Nieuwkoop P.D., Suturya L.A., 1979 – Primo rdial germcells in the chordates. Cambridge Univ. Press.

NIJHOUT H.F., 1999 – Control mechanisms of polyphenicdevelopment in insects. Bioscience, 49: 181-192.

NOBLE E.R., NOBLE G.A., SCHAD G.A., MACIN NES A.J.,1989 – Parasitology. The biology of animal parasites. 6thedition 574 pp. Lea&Febiger, Philadelphia, London.

NOSKIEWICZ J., POLUSZYNSKI G., 1935 – Embryo logischeUntersuchungen and Strepsipteren II Teil; Polyembryonie.Zool. Poloniae, 1: 53-94.

OLSEN M.W., 1962 – Polyembryony in unfertilized turkeyeggs. J. Heredity, 53 (3): 125-126.

RANA R.L., DAHLMAN D.L., Webb B.A., 2002 – Expressionand characterization of a novel teratocyte protein of thebraconid, Microplitis croceipes (Cresson). Ins. Biochem.Mol. Biol., 32 (11): 1507-1516.

SAFFMAN E.E., LASKO P., 1999 – Germline development invertebrates and invertebrates. Cell. Mol. Life Sci., 55 (8-9):1141-1163.

SANDER K., 1976 – Specification of the basic body pattern ininsect embryogenesis. Adv. Ins. Phys., 12: 125-238.

– 50 –

SCHMIDT CAPELLA I.C., HARTFELDER K., 2002 – Juvenile-hormone-dependent interaction of actin and spectrin is cru-cial for polymorphic differentiation of the larval honey beeovary. Cell Tissue Res., 307: 265-272.

SHAANKER R.U., GANESHAIAH K.N., 1996 – Polyem bryonyin plants: A weapon in the war over offspring numbers?Trends Ecol. Evol., 11 (1): 26-27.

SHOSTAK S., 1993 – Cnidaria . In: Reproductive Biology ofInvertebrates (Adiyodi, K.G. and Adiyodi R.G. eds).John Wiley and Sons Chichester.

SILVESTRI F., 1906 – Contribuzioni alla conoscenza biologicadegli imenotteri parasiti. Biologia del Litomastix trun-catellus (Dalm.) (2 nota preliminare). Ann. Regia. Sc.Agric. Portici; 6: 3-59.

SPEICHER B.R., 1936 – Oogenesis, fertilization and earlycleavages in Habrobracon. J. Morph.,59: 401-421.

STANSFIE.W.D., 1968 – A serological estimate of monozygot-ic twinning in sheep. J. Heredity, 59 (3): 211-212.

TREMBLAY E., CALVERT D., 1972 – New cases of polar nucleiutilization in insects. Ann. Soc. Ent. Fr., 8 (2): 495-498.

UNDERWOOD E.M., CAULTON J.H., ALLIS C.D., MAHOVALD

A.P., 1980 – Developmental fate of pole cells in Drosophi-la melanogaster. Dev. Biol., 77: 303-314.

WHITFILED J.B., 1998 – Phylogeny and evolution of host-par-asitoid interactions in hymenoptera. Ann. Rev. Entomol.,43: 129-151.

WHITFIELD J.B., 2002 – Estimating the age of the polyd-navirus/braconid wasp symbiosis. Proc. Natl. Acad. Sci.USA, 99 (11): 7508-7513.

WILKINS A., 2001 – The evolution of developmental path-ways. Sinauer Associates Inc. Sunderland, Massachu-setts, pp. 603.

ZHUROV V., TERZIN T., GRBIC’ M., 2004 – Early blastomeredetermines embryo proliferation and caste fate in a polyem-bryonic wasp. Nature, 432: 746-769.

– 51 –


(*) Dipartimento Biologia Evoluzionistica Sperimentale, Università di Bologna, Via Selmi, 3 Bologna. Email: [email protected] tenuta durante la Tavola rotonda «Genetica molecolare dello sviluppo degli insetti».Seduta pubblica dell’Accademia - Firenze, 28 novembre 2008.

Control of vitelline membrane gene expression during D. melanogaster oogenesisThe formation of extracellular structures is a complex process that requires time-coordinate synthesis, cleavage and trans-

port of various proteins, and finally, cross-linking mediated by particular functional domains. Exactly how the precise featuresof such biological structures are constructed remains a fascinating problem. We approach this question by studying the eggshellassembly in Drosophila melanogaster.

Although much is known about the induction and refinement of the signaling pathways involved in the formation of the ante-rior eggshell structures, little is known about the regulation and the function of the genes encoding eggshell structural proteins.

This review will summarize the knowledge on vitelline membrane gene expression, focusing on our results on the expressionpattern and the regulatory elements controlling transcription of a member of this gene family, the VM32E gene. Compared withthe other vitelline membrane genes, this gene shows a peculiar expression pattern that suggests interesting perspectives ofVM32E protein function in eggshell assembly.

KEY WORDS: Drosophila melanogaster, eggshell, vitelline membrane genes, gene expression.

CONTROL OF VITELLINE MEMBRANE GENE EXPRESSIONDURING D. MELANOGASTER OOGENESIS

GIUSEPPE GARGIULO (*)

OVERVIEW OF DROSOPHILA EGGSHELL

The Drosophila eggshell is an extracellular struc-ture functional to the different roles it absolves,from egg fertilization to the hatching of the larva atthe end of embryogenesis. This multilayered extra-cellular matrix, that forms between the oocyte andthe overlying follicle cells, is made up of the fol-lowing layers (Fig.1A): a vitelline membrane,which is the innermost layer, the wax layer, thecrystalline innermost chorionic layer, the tripartiteendochorion (inner part, pillars, outer part), andthe exochorion which is the outer layer of theeggshell (MARGARITIS et al., 1980; reviewed byCAVALIERE et al., 2008). The formation of theseextracellular structures is a complex process thatrequires time-coordinated synthesis, cleavage andtransport of various proteins and, finally, cross-linking at specific functional domains (ANDRE -NACCI et al., 2001; MANOGARAN & WARING, 2004).The eggshell proteins are synthesized and secretedby the follicle cells surrounding the oocyte duringstages 8-14 of egg chamber development(reviewed by CAVALIERE et al., 2008). From stages8 to 10, the synthesis of vitelline membrane com-ponents is predominant. Subsequently, duringstages 11-14, the chorion proteins are synthesizedand secreted. The chorion becomes insoluble dur-

ing stage 14 as a result of a peroxidase-typeenzyme activity that cross-links two or three tyro-sine residues of the chorion proteins (PETRI et al.,1976). The vitelline membrane proteins remainsoluble till stage 14 and become insoluble only inlaid eggs (reviewed by WARING, 2000). Like mostinsect eggs, several regional specializations areapparent on the surface of Drosophila eggshell. Itsanterior end is characterized by specializedeggshell structures such as the micropyle thatallows sperm entry, the operculum that forms theexit hatch for the developed embryo, and two dor-sal respiratory appendages (Fig. 1B) (reviewed byCAVALIERE et al., 2008). This eggshell specializa-tion depends on sequential and in some cases com-bined activities of the known major signaling path-ways during the middle and late stages of oogene-sis. The epidermal growth factor receptor (Egfr)signaling induces dorsal follicle cells fates(reviewed by RAY & SCHÜPBACH, 1996; PERRIMON

& PERKINS, 1997; reviewed by DOBENS &RAFTERY, 2000), leading to the definition of twoseparate populations of dorsal follicle cells(WASSERMAN & FREEMAN, 1998; PERI et al., 1999)that will guide the production of the two dorsalappendages. The patterning of the eggshell alongthe AP axis also requires transforming growth fac-tor-beta (Tgf-β) family member, Decapentaplegic

(Dpp) (PADGETT et al., 1987; TWOMBLY et al., 1996;DENG & BOWNES, 1997; PERI & ROTH, 2000). Thesize and placement of the operculum and dorsalappendages are quite sensitive to altered levels ofDpp signal (TWOMBLY et al., 1996).

The Ecdysone signaling pathway also appearsinvolved on eggshell production by controllingchorion gene transcription and amplification(CHERBAS et al., 2003; ORO et al., 1992; HACKNEY

et al., 2007). Ecdysone activity is mediated by aheterodimer of two nuclear receptors, Ecdysonereceptor (EcR) and Ultraspiracle (Usp) (YAO et al.,1992; THOMAS et al., 1993; YAO et al., 1993). EcRencodes three protein isoforms: EcR-A, EcR-B1and EcR-B2, which have common DNA- and lig-and-binding domains but differ in their N-termi-nal region. The EcR/Usp heterodimer controlsgene transcription by binding to specificsequences named Ecdysone Response Elements(EcREs) (RIDDIHOUGH & PELHAM, 1987; CHERBAS

et al., 1990).

EXPRESSION OF VITELLINE MEMBRANE GENES

Four Drosophila vitelline membrane proteingenes have been characterized so far: VM26A.1,

VM34C, VM26A.2 and VM32E (HIGGINS et al.,1984; MINDRINOS et al., 1985; BURKE et al., 1987;POPODI et al., 1988; GIGLIOTTI et al., 1989).

Although the vitelline membrane protein genesVM26A.1, VM34C, VM26A.2 are expressed in allfollicle cells surrounding the oocyte, from stage 8to stage 10 of oogenesis, the VM32E gene is activeonly at stage 10 of egg chamber development(BURKE et al., 1987; POPODI et al., 1988; GIGLIOTTI

et al., 1989), and it is differently expressed in thedistinct domains of the follicular epithelium(GARGIULO et al., 1991). Within the main body fol-licle cells, the expression of VM32E is temporallyregulated. The expression starts at stage 10A andit appears to spread from the ventral to the dorsalfollicle cell domain (Fig. 2A). By stage 10B thegene is transcribed in all main-body follicle cells,but is absent in the anterior follicle cells (cen-tripetal cells) and posterior follicle cells (Fig. 2B,arrow and arrowhead respectively). Differently, atstage 10B of egg chamber development, the othervitelline membrane genes are expressed in all folli-cle cells surrounding the oocyte. In Fig. 2C isshown the expression pattern of the VM26A.2gene, which encodes one of the most abundantvitelline membrane structural proteins (POPODI etal., 1988; PASCUCCI et al., 1996).

– 54 –

Fig. 1Eggshell morphology. (A) Stage 14 egg chamber showing the building up of the different eggshell layers. The inner layers include thevitelline membrane (VM), the wax layer (WL), and the inner chorion layer (ICL). The outer layers include the endochorion (En), wherea thin fenestrated floor (f) is separated from an outer roof (r) by vertical pillar (p) creating cavities as indicated, and the exochorion (Ex).(B) Dorsal view of Drosophila melanogaster egg illustrating the specialized regions of the anterior shell. The prominent structures indi-cated are the micropyle (mi), the operculum area (op), and the dorsal appendages (da).

Fig. 2 Expression pattern of VM32E and VM26A.2 genes. (A,B) Wholemount in situ hybridization showing the spatial distribution of theVM32E transcript. (A) At stage 10A the expression starts in agroup of ventral columnar follicle cells and gradually spreads fromthe ventral to the dorsal follicle cell domain. (B) At stage 10B allmain body follicle cells express the VM32E gene but the cen-tripetal cells (arrows) and a group of posterior follicle cells aresilent (arrowhead). (C) Whole-mount in situ hybridization show-ing the spatial distribution of VM26A.2 showing that this gene isexpressed in all follicle cells. In all the panels, the egg chambers areoriented with the anterior region toward the left, and with the dor-sal side at the top.

A B

C

– 55 –

VM32E PROTEIN SYNTHESIS AND DISTRIBUTION

In order to analyze the in situ distribution of theVM32E protein during egg chamber developmentwe have raised a polyclonal antibody against theC-terminal VM32E peptide EELRGLGQGIQGQQY (anti-CVM32E), specific to this protein(ANDRENACCI et al., 2000). Our analysis of thelocalization and assembly of the VM32E proteindemonstrated several unique features of this pro-tein that add more detail on the complex processof eggshell morphogenesis. The VM32E proteinsynthesis reflects the VM32E gene transcriptionpattern (Fig. 3A). At stage 10B this protein is syn-thesized in the columnar follicle cells, except themost anterior and posterior ones (Fig. 3C arrowsand arrowhead respectively), and secreted in theextracellular space between the follicle cells andthe oocyte surface. Furthermore we have foundthat once secreted this protein moves to the polesand by stage 11 it appears uniformly distributed inthe vitelline membrane. The VM32E protein dis-tribution was also studied at the ultrastructurallevel by immunogold electron microscopy on eggchamber thin sections (ANDRENACCI et al., 2001).At the late stage 10B the immunogold particlesstrongly labeled the vitelline membrane. At stage12 (Fig. 3B) immunogold particles were detectedin the vitelline membrane and in the formingendochorion pillars. Finally, at stage 14 gold parti-cles labeled both the vitelline membrane and the

endochorion. The VM32E protein is therefore anintegral component of both the vitelline and endo-chorion layers.

SPATIAL AND TEMPORAL REGULATION OF VM32EGENE EXPRESSION

The complex temporal and spatial expressionpattern of VM32E is under the control of the Dppand Egfr signaling pathways. The VM32E gene isnot expressed in the centripetal migrating folliclecells, where the Decapentaplegic (Dpp) pathway isactive in patterning the anterior eggshell structures.By analyzing the expression of VM32E gene ingenetic backgrounds altering the Dpp pathway wehave demostrated that in the centripetal cells theVM32E gene is negatively regulated by the Dppsignaling (BERNARDI et al., 2006). The temporaland spatial VM32E gene expression pattern, inwhich the gene is initially active only in the ventralfollicle cells, is under the control of the Egfr signal-ing which display a gradient of its activity along thedorsoventral axis. We have shown that in folliclecell clones expressing a constitutively active formof the Egfr the VM32E gene is down-regulated,while the loss of the Egfr activity up-regulatesVM32E expression (BERNARDI et al., 2007). Morerecently we have demonstrated that ecdysone sig-naling plays a crucial role in activating the expres-sion of the VM32E gene in the main body follicle

cells. This control is mediated by Ecr-B1 isoformand Usp receptors (BERNARDI et al., 2009).

Interestingly the Dpp, Egfr and Ecdysone sig-naling pathways do not appear to play a generalrole in controlling the expression of the othermembers of the vitelline membrane gene family.These findings highlight the distinguishing charac-teristics of VM32E gene among the other vitellinemembrane genes.

In an effort to elucidate the cis-acting regulatorysignals that dictate the developmental expressionof the VM32E gene, we have performed an exten-sive analysis of its promoter. Results from germ linetransformation experiments using different frag-ments of the VM32E promoter fused to the lacZreporter gene have shown that the spatial and tem-poral expression pattern of VM32E is achieved bythe combined activity of positive and negative con-trolling elements (Fig. 4A). Within the minimalVM32E promoter (-348/-39) we defined the cis-acting regulatory regions dictating VM32E geneexpression in the different follicle domains(CAVALIERE et al., 1997; ANDRENACCI et al., 2000).The 3’ most segment of the minimal promoter(proximal element; -112/-39) interacts with twoother cis-regulatory regions (-253/-113 and -348/-254) and is absolutely required for their output. In

transgenic flies, the -112/-39 region by itself drivesthe expression of the lacZ reporter gene only in thecentripetal cells that normally do not transcribe theVM32E gene (Fig. 4B). When this upstream ele-ment is joined with the -253/-113 region, the lacZreporter gene is expressed in the ventral folliclecells (Fig. 4B). The -253/-113 fragment besidescontaining the positive element(s) that specifyVM32E gene expression in the ventral domain con-tains the cis-acting element(s) repressing the geneactivity in the centripetal follicle cells through theDpp signaling pathway (BERNARDI et al., 2006). Inaddition, the -253/-39 region contains the responseelement(s) to the Egfr pathway that modulates thelevel of VM32E gene expression (BERNARDI et al.,2007). Within the -348/-39 minimal promoter, the-348/-254 region appears essential to drive thedorsal expression of the gene and to enhance theventral expression. When the -348/-254 region isjoined with the -112/-39 segment, the expression ishigher than the one shown by the -348/-39 pro-moter and involves also centripetal follicle cells,due by the absence of the -253/-113 region(ANDRENACCI et al., 2000). When analyzed byitself the -348/-254 region drives expression onlyin the border cells, suggesting that, in these cells,native VM32E gene is switched off by negative

– 56 –

Fig. 3 VM32E distribution in egg chambers and in the forming eggshell layers. (A) Whole-mount egg chambers stained with anti-CVM32Eantibody showing that the VM32E protein is not synthesized in the anterior and posterior follicle cells. (C) Confocal image showing thecross section view of a stage 10B egg chamber expressing the fused His2avDGFP protein (Clarkson and Saint, 1999) that marks all thenuclei (green). VM32E protein (red) is not detected in the centripetal follicle cells (arrows) and in a small group of posterior follicle cells(arrowhead). In the (A) and (C) panels, the egg chambers are oriented with the anterior region toward the left, and with the dorsal sideat the top. (B) Immunoelectron microscopy detection of VM32E protein using anti-CVM32E antibody showing the multilayered dis-tribution of VM32E protein. As shown by the localization of immunoglod particles at stage 12 the protein is localized in the vitellinemembrane (VM) and in the pillars (arrow) of the forming endochorion layer (En).

control element(s) which could be located withinthe -112/-39. Within the -348/-254 region wefound a putative EcRE sequence to which theEcr/Usp transcription complex could bind andactivate gene expression (BERNARDI et al., 2009).

The precise temporal and spatial control of theVM32E eggshell structural gene, mediated by Dpp,Egfr and Ecdysone signalings, suggests a crucialrole of VM32E protein in the complex process ofeggshell assembly. Many aspects of eggshell bio-genesis are still to be elucidated. Besides its struc-tural contribution to eggshell formation, it hasbeen suggested that the vitelline membrane mayperform an important function for the localizationof maternal signals involved in embryonic pattern-ing (reviewed by CAVALIERE et al., 2008). Thevitelline membrane proteins appear not functional-ly redundant suggesting that each structural com-ponent may have a specific role in the eggshellassembly (ANDRENACCI et al., 2001). The peculiartemporal and spatial regulation of VM32E geneactivity may suggest that the VM32E protein couldbe involved in the process that determines theproper localization of maternal cues within thevitelline membrane.

RIASSUNTO

ESPRESSIONE GENICA DI CONTROLLODELLA MEMBRANA VITELLINA DURANTE

L’OOGENESI DI D. MELANOGASTER

La formazione di strutture extracellulari è un processocomplesso e in parte ancora ignoto che ha inizio all’inter-no della cellula con la sintesi coordinata delle sue differen-ti componenti proteiche. Le proteine sono successiva-mente secrete nello spazio extracellulare dove possonoandare incontro a tagli proteolitici e a trasporto selettivo.Infine esse sono assemblate in complessi sopramolecolaristabili attraverso la formazione di legami covalenti checoinvolgono specifici domini proteici. I meccanismi chestanno alla base della formazione di tali strutture rappre-sentano un quesito biologico affascinante e il guscio del-l’uovo di Drosophila melanogaster rappresenta un ottimosistema per indagare questi fenomeni. Numerosi studihanno permesso di identificare il ruolo di varie vie di seg-nalazione nella formazione delle strutture anteriori delguscio. Meno noti sono invece i meccanismi alla base delcontrollo trascrizionale dei geni che codificano per le com-ponenti della membrana vitellina, così come le funzionisvolte dalle singole proteine nell’assemblaggio del guscio.

Questa review riassume le conoscenze attuali riguardan-ti l’espressione dei principali geni della membrana vitellinae in particolare sono descritti i risultati da noi ottenuti suimeccanismi che regolano l’espressione del gene VM32E,uno dei membri di questa famiglia genica.

– 57 –

Figure 4Schematic representation of the interactions among the cis-regulatory elements (patterned boxes) of the VM32E promoter. The -348/-254 and the -253/-113 elements interact with the -112/-39 fragment to control the VM32E expression (continuous arrows) in the dif-ferent follicle cell populations. The -348/-254 region contains an EcRE site (-295/-273) marked by a box. cis-acting regions responsiveto Dpp and Egfr signaling are delimited by brackets. (B) Stage 10B egg chamber showing the VM32E protein expression pattern of the native gene (red) and the ventral lacZ expression pat-tern (green) driven by the -253/-39 VM32E promoter region. (C) Stage 10B egg chamber showing the lacZ expression (green) in thecentripetal follicle cells (indicated by arrows) driven by the –112/-39 VM32E promoter region. In (B) and (C) panels, the egg chambersare oriented with the anterior region toward the left. In the panel (B) the dorsal side is at the top.

L’analisi di tale gene è interessante in quanto esso pre-senta un profilo di espressione peculiare che suggerisceinteressanti prospettive svolte dalla proteina nella for-mazione del guscio dell’uovo.

REFERENCES

ANDRENACCI D., CAVALIERE V., CERNILOGAR F.M.,GARGIULO G., 2000 – Spatial activation and repression ofthe Drosophila vitelline membrane gene VM32E areswitched by a complex cis-regulatory system. Dev. Dyn.,218: 499-506.

ANDRENACCI D., CERNILOGAR F.M., TADDEL C., ROTOLID., CAVALIERE V., GRAZIANI F., GARGIULO G., 2001 –Specific domains drive VM32E protein distribution andintegration in Drosophila eggshell layers. J. Cell Sci., 114:2819-2829.

BERNARDI F., CAVALIERE V., ANDRENACCI D., GARGIULOG., 2006 – Dpp signaling down-regulates the expressionof VM32E eggshell gene during Drosophila oogenesis.Dev. Dyn., 235: 768-775.

BERNARDI F., DUCHI S., CAVALIERE V., DONATI A.,ANDRENACCI D., GARGIULO G., 2007 – Egfr signalingmodulates VM32E gene expression during Drosophilaoogenesis. Dev. Genes Evol., 217: 529-540.

BERNARDI F., ROMANI P., TZERTZINIS G., GARGIULO,G.,CAVALIERE V., 2009 – EcR-B1 and Usp nuclear hormonereceptors regulate expression of the VM32E eggshell geneduring Drosophila oogenesis. doi:10.1016/j.ydbio.2009.01.013.

BURKE T., WARING G.L., POPODI E., MINOO P., 1987 –Characterization and sequence of follicle cell genes selec-tively expressed during vitelline membrane formation inDrosophila. Dev. Biol., 124: 441-450.

CAVALIERE V., SPANO S., ANDRENACCI D., CORTESI L.,GARGIULO G., 1997 – Regulatory elements in the pro-moter of the vitelline membrane gene VM32E ofDrosophila melanogaster direct gene expression in dis-tinct domains of the follicular epithelium. Mol. Gen.Genet., 254: 231-237.

CAVALIERE V., BERNARDI F., ROMANI P., DUCHI S.,GARGIULO G., 2008 – Building up the Drosophilaeggshell: first of all the eggshell genes must be transcribed.Dev. Dyn., 273: 2061-2072.

CHERBAS L., HU X., ZHIMULEV I., BELYAEVA E., CHERBASP., 2003 – EcR isoforms in Drosophila: testing tissue-spe-cific requirements by targeted blockade and rescue.Development, 130: 271-284.

CHERBAS P., CHERBAS L., LEE K., ANDRES A., 1990 –Ecdysone response elements of a Drosophila gene. Prog.Clin. Biol. Res., 342: 112-115.

DENG W.M., BOWNES M., 1997 – Two signalling pathwaysspecify localised expression of the Broad-Complex inDrosophila eggshell patterning and morphogenesis.Development, 124: 4639-4647.

DOBENS L.L., RAFTERY L.A., 2000 – Integration of epithe-lial patterning and morphogenesis in Drosophila ovarianfollicle cells. Dev. Dyn., 218: 80-93.

GARGIULO G., GIGLIOTTI S., MALVA C., GRAZIANI F., 1991–. Cellular specificity of expression and regulation ofDrosophila vitelline membrane protein 32E gene in thefollicular epithelium: identification of cis-acting elements.Mech. Dev., 35: 193-203.

GIGLIOTTI S., GRAZIANI F., DE PONTI L., RAFTI F., MANZIA., LAVORGNA G., GARGIULO G., MALVA, C., 1989 – Sex,tissue-, and stage-specific expression of a vitelline mem-brane protein gene from region 32 of the second chromo-some of Drosophila melanogaster. Dev. Genet., 10: 33-41.

HACKNEY J.F., PUCCI C., NAES E., DOBENS L., 2007 – Ras

signaling modulates activity of the ecdysone receptor EcRduring cell migration in the Drosophila ovary. Dev. Dyn.,236: 1213-1226.

HIGGINS M.J., WALKER V.K., HOLDEN J.A., WHITE B.N.,1984 – Isolation of two Drosophila melanogaster genesabundantly expressed in the ovary during vitelline mem-brane synthesis. Dev. Biol., 105: 155-165.

MARGARITIS L.H., KAFATOS F.C., PETRI W.H., 1980 – Theeggshell of Drosophila melanogaster. J. Cell. Sci., 43: 1-35.

MINDRINOS M.N., SCHERER L.J., GARCINI F.J., KWAN H.,JACOBS K.A., PETRI W.H., 1985 – Isolation and chromo-somal location of putative vitelline membrane genes inDrosophila melanogaster. EMBO J., 4: 147-153.

MANOGARAN A., WARING G.L., 2004 – The N-terminalprodomain of sV23 is essential for the assembly of a func-tional vitelline membrane network in Drosophila. Dev.Biol., 270(1): 261-271.

ORO A.E., MCKEOWN M., EVANS R.M., 1992 – TheDrosophila retinoid X receptor homolog ultraspiraclefunctions in both female reproduction and eye morpho-genesis. Development, 115: 449-462.

PASCUCCI T., PERRINO J., MAHOWALD A.P., WARING G.L.,1996 – Eggshell assembly in Drosophila: Processing andlocalization of vitelline membrane and chorion proteins.Dev. Biol., 177: 590-598.

PERI F., BOKEL C., ROTH S., 1999 – Local gurken signalingand dynamic MAPK activation during Drosophila oogen-esis. Mech. Dev., 81: 75-88.

PERI F., ROTH S., 2000 – Combined activities of gurken anddecapentaplegic specify dorsal chorion structures of theDrosophila egg. Development, 127: 841-850.

PETRI W.H., WYMAN A.R., KAFATOS F.C., 1976 –. Specificprotein synthesis in cellular differentiation. III. Theeggshell proteins of Drosophila melanogaster and theirprogram of synthesis. Dev. Biol., 49: 185-199.

PERRIMON N., PERKINS L.A., 1997 – There must be 50 waysto rule the signal: the case of the Drosophila EGF recep-tor. Cell, 89: 13-16.

POPODI E., MINOO P., BURKE T., WARING G.L., 1988 –Organization and expression of a second chromosome fol-licle cell gene cluster in Drosophila. Dev. Biol., 127: 248-256.

RAY R.P, SCHÜPBACH T., 1996 – Intercellular signaling andthe polarization of body axes during Drosophila oogene-sis. Genes. Dev., 10: 1711-1723.

RIDDIHOUGH G., PELHAM H.R., 1987 – An ecdysoneresponse element in the Drosophila hsp27 promoter.EMBO J., 6: 3729-3734.

THOMAS H.E., STUNNENBERG H.G., STEWART A.F., 1993 –Heterodimerization of the Drosophila ecdysone receptorwith retinoid X receptor and ultraspiracle. Nature, 362:471-475.

TWOMBLY V., BLACKMAN R.K, JIN H., GRAFF J.M,PADGETT R.W, GELBART W.M., 1996 – The TGF-betasignaling pathway is essential for Drosophila oogenesis.Development, 122: 1555-1565.

WARING G.L., 2000 – Morphogenesis of the eggshell inDrosophila. Int. Rev. Cytol., 198: 67-108

WASSERMAN J.D., FREEMAN M., 1998 – An autoregulatorycascade of EGF receptor signaling patterns theDrosophila egg. Cell, 95: 355-364.

YAO T.P., SEGRAVES W.A., ORO A.E., MCKEOWN M.,EVANS R.M., 1992 – Drosophila ultraspiracle modulatesecdysone receptor function via heterodimer formation.Cell., 71: 63-72.

YAO T.P., FORMAN B.M., JIANG Z., CHERBAS L., CHEN J.D.,MCKEOWN M., CHERBAS P., EVANS R.M., 1993 –Functional ecdysone receptor is the product of EcR andUltraspiracle genes. Nature, 366: 476-479.

– 58 –


(*) Department of Evolutionary Biology, University of Siena, via A.Moro 2, 53100 Siena, ItalyLettura tenuta durante la Tavola rotonda «Genetica molecolare dello sviluppo degli insetti».Seduta pubblica dell’Accademia - Firenze, 28 novembre 2008.

Drosophila spermatogenesis: a system model to cell cycle analysisDrosophila melanogaster represents a powerful tool to examine cell division mechanisms and to decipher cen -

triole/centrosome duplication processes. Male meiosis is particularly suitable to study the regulation of cell cycle progressionand the process of spindle assembly and cytokinesis. Male meiosis lacks, indeed, working cell-cycle checkpoints. Thus, germcells can progress through the cell cycle despite the occurrence of abnormalities in spindle assembly, chromosome alignment,centrosome structure. Overriding canonical blocks in the presence of spindle-DNA damages leads to unusual conditions inwhich eventual problems are enlarged, becoming more evident. Drosophila genetics also offers the opportunity to study the roleof specific regulators in centriole duplication/assembly, such as SAK/PLK4 and SAS-6. Making transgenic flies for these genesgives useful information to understand how centriole assembly and duplication are controlled during the cell cycle. Furtherstudies in Drosophila may establish common developmental and cellular pathways underlying the phenotype of several humandiseases.

KEY WORD: Drosophila, spermatogenesis, cell cycle, centrosomes, centrolies.

DROSOPHILA SPERMATOGENESIS:A SYSTEM MODEL TO CELL CYCLE ANALYSIS

GIULIANO CALLAINI (*)

ORGANIZATION OF THE DROSOPHILA TESTIS

Drosophila testis consists of an elongated coiledtube filled of germ cells at different developmentalstages (Fig. 1), that have their origin in a smalldistinct apical region where somatic and stem cellsform a distinct niche. In the male gonad ofDrosophila two stem cell lines, germline (GSCs)and somatic cyst progenitor (CPCs) cells, lie at theapical tip of the testis, anchored by thincytoplasmic extensions to a cluster of non-dividingcells, the hub cells (HARDY et al., 1979). EachGSCs has been proposed to be flanked in theniche by a pair of SSCs, to form a complex ofthree cells that are thought to divide in synchrony.The three daughter cells remaining in contact withthe hub cells retain stem cell fate, while the otherthree cells associate into a cyst. Within these cysts,the cytoplasmic extensions of two squamous cystcells form a thin, microtubule-rich shell thatsurrounds the gonioblast, that undergoes fourrounds of mitotic divisions with incompletecytokinesis.

An interesting feature of the male meiosis inDrosophila is the reduction of centriole numberduring transition from meiosis I to meiosis II.Spermatocytes I have two centrosomes containinga centriole pair each. These centrioles behave

throughout each centrosome during the meioticcell cycle as somatic cell centrioles, and eachdaughter spermatocyte II inherits two centro -somes containing only one centriole. However,unlikely somatic cells, the centrioles of sperma -tocytes II do not reproduce. So, the sper ma tocytesII progress trough meiosis with centro somescontaining only one centriole. This repre sents aninteresting exception to the general rule: thesecells are able to undergo meiosis and cytokinesiswith non-canonical centrosomes.

Sperm formation in Drosophila occurs by twodistinct steps: spermatogenesis and spermio -genesis. During spermatogenesis a germ line stemcell divides asymmetrically to produce a primaryspermatogonium, that after four incompletemitoses gives rise to a progeny of 16 primaryspermatocytes. Spermatocytes are connected bycytoplasmic bridges and reside in a common cystdelimitated by two large and flat somatic cyst cells.After a growth period in which the spermatocytesincrease several times their volume, they undergotwo rounds of meiotic divisions to form cysts of 64interconnected spermatids (FULLER 1993). Spermiogenesis is a complex process that involvesa series of elaborate transformations of thespermatid components. These dramatic transfo -rmations from an immotile cell to a motile one

include chromatin condensation, cell and axo -neme elongation, mitochondrial rearrange ment,and sperm individualization (FULLER, 1993). Afterelongation the syncytial spermatids must beindividualized to become mature sperm.Cytoplasmic bridges are broken and the excesscytoplasmic material within the flagella issequestered, whereas mitochondrial derivatives areremodeled. This process is mediated by 64 distinctconical complexes, or individualization complexes(ICs) (TOKUYASU et al., 1972), mainly composed ofactin (FABRIZIO et al., 1998), that translocate alongthe length of the spermatid bundles from thenuclei to the tail ends.

DROSOPHILA MALE MEIOSIS

One approach to obtain insight into the mecha -nisms of spindle assembly during cell division isthe isolation and characterization of genesinvolved in specific controls of this process.Meiosis in Drosophila males offers the opportunityfor a detailed analysis of the process of spindleassembly and cell division through the combineduse of genetics and cell biology. The phenotype ofsome Drosophila mutants that uncouplecentrosome organization from spindle bipolarityshould therefore be helpful in unraveling thecontributions of centrosomes to spindle orga -

nization (GLOVER et al., 1995; GONZALEZ et al.,1998; ROTHWELL and SULLIVAN, 2000). Due to therelatively large size of primary and secondaryspermatocytes, meiotic spindles and centrioles arevery prominent and can be readily detected bystaining with several antibodies (Fig. 2). Moreover,several studies demonstrated that meioticmechanisms have been broadly conserved and thatsubstantial similarities between meiotic regulationin flies and vertebrates exist (MAINES andWASSERMAN, 1998). Products of Drosophila genessuch as twine (ALPHEY et al., 1992), boule(EBERHART et al., 1996) and polo (LLAMAZARES etal., 1991) have counterparts in the vertebrate germline. Moreover, the cytology of male meiosis hasbeen thoroughly described in Drosophila (CENCI etal., 1994) and many mutants that disrupt specificaspects of this process have been isolated (seeCASTRILLON et al., 1993). Mutations in the genespolo (SUNKEL and GLOVER, 1988), aurora (GLOVER

et al., 1995) and slimb (WOJCIK et al., 2000) impaircentrosome segregation or duplication leading tothe formation of spindles with multiplecentrosomes. Monopolar spindles organized byeither unreplicated centrosomes or multiplecentrosomes which have failed to segregate arecommon to many mitotic mutants in Drosophila.Mutational analysis also suggested that acomponent of the Drosophila mitotic centrosome,centrosomin (cnn), is required for spindle

– 60 –

Fig. 1Preparation of a pupal testis for immunofluorescence microscopy localization of microtubules. spI, primary spermatocytes; spIIsecondary spermatocytes during prophase (arrows) or telophase (arrowheads); spermatids, el, elongationg spermatids, sp, sperm.

organization during male meiosis. Mutations in theabnormal spindle locus (asp) result in abnormalspindle morphology leading to a metaphase arrestin third-instar larval neuroblasts. Asp localizes tothe spindle poles from prophase to anaphase ofthe syncytial divisions and asp mitotic spindles arebipolar but have disorganized broad poles atwhich γ-tubulin has a highly abnormal clump-likedistribution (AVIDES and GLOVER, 1999).However, despite these studies have providedinsight into the function of some centrosomalcomponents, the role of the centrosome in spindleorganization and dynamics still deserves particularattention.

These observations suggest that asp may regulatethe stability or other dynamic properties of thespindle microtubules at the centrosome(RIPARBELLI et al., 2002). In late anaphase-telophase of male meiosis Asp protein localises tothe spindle mid-zone, but unlike othercentrosomal antigens such as Pava rotti-KLP andPolo kinase that become associated with the verycentral region of the spindle midzone, the majorityof Asp protein decorates the very terminal regionsof mid-zone microtubules, which at this stage arepositioned between the telophase nuclei and thecentre of the spindle. These appear to be theminus ends of microtubules that have beenreleased from the centrosomes which at this stagenucleate independent asters of microtubules. This

association of Asp with the spindle mid zoneappears to be required for the correct structuresof the late central spindle and in turn forcytokinesis. Cytokinesis defects are not seen inother tissues of asp mutants which generally causecell division to arrest at metaphase. This couldreflects either the apparent lack of a requirementfor asp function to organize the poles of themeiotic spindle, or perhaps more likely that thespindle integrity checkpoint delays M-phaseprogression less effectively in male meiosis that itdoes in mitosis. The central spindle plays anessential role during cytokinesis and there is acooperative interaction between this structure andthe acto-myosin contractile ring: whenever one ofthe structures is disrupted the other fails toassemble and function. In keeping with this wehave found that many cells within asp mutantcysts have abnormal central spindles lacking thecharacteristic interdigitating microtubules. More -over, molecules that participate in forming partsof the contractile ring, Pavarotti-KLP, the septinPeanut, Polo kinase and Actin, do not localiseproperly in asp mutant spermatocytes.Interestingly, human hortologues of Asp andcentrosomin (cnn) are also mutated in primarymicrocephaly, suggesting that further studies inDrosophila may establish common developmentaland cellular pathways underlying the humanprimary microcephaly phenotype.

– 61 –

Fig. 2Detail of a cyst of primary spermatocytes stained for tubulin (green) and γ-tubulin (red). Note that each germ cells contains two pairs oforthogonal rod-like centrioles (arrows).

CENTROSOME AND SPINDLE ASSEMBLY

An essential feature of cell division is the fidelitywith which the chromosomes are segregated,because errors in chromosome inheritance canlead to severe congenital malformations orcontribute to tumour progression. At the onset ofeach M phase, the assembly and appropriateorganization of the microtubule cytoskeleton in abipolar array is necessary to distribute to thedaughter cells replicated chromosomes (duringmitosis or meiosis II) or homologous chromo -somes (during meiosis I). The role of thecentrosome, in spindle assembly and function wassuggested as far back as the late 1800s, whencentrosomes were first identified.

The centrosome is the primary microtubule-organizing centre (MTOC) of eukaryotic cell and assuch plays a fundamental role in the orga nization ofthe cytoplasm. During interphase, the centrosomeorganizes a microtubule network that is involved infundamental cellular functions such as intracellulartrafficking, cell motility, cell adhesion, and cellpolarity. During cell division, two centrosomes areneeded to correct organize the bipolar spindlerequired to accurately segregate chromosomes andplay main roles in its orientation with respect tocortical cues, and the late events of cytokinesis.Since each daughter cell receives only onecentrosome upon cytokinesis, the centrosome, likeDNA, must duplicate once prior to next mitosis. Inproliferating cells, the centrosome duplicates justbefore, or at, the onset of S phase. Thus, at anygiven time in the cell cycle, cells have either oneunduplicated or two duplicated centrosomes. Inaddition to its function in microtubule organi -zation, the centrosome serves as a scaffold foranchoring several regulatory proteins. Among theseare cell-cycle regulators whose association with thecentrosome is required for several cell-cycletransitions, including G1 to S-phase, G2 to mitosisand metaphase to anaphase.

The centrosomes exert a strong influence on thespindle architecture and dictate the number of thepoles formed (SLUDER and NORDBERG,. 2004;NIGG, 2002). Any deviation from normalcentrosome numbers can result in the formation ofmonopolar or multipolar spindles, with directconsequences on the proper chromosomesegregation, condition often associated withaneuploidy, a hallmark of cancer cells (NIGG,2006). Accordingly, abnormalities in the spindlepole organization function occur in many cancersand can be associated with genomic instabilitybecause extra and irregular centrosomes can giverise to aberrant cell division (NIGG, 2006).

THE CENTRIOLE: THE MASTER

OF CENTROSOME ORGANIZATION

The centrosome typically consists of two compo -nents: a core structure consisting of a pair ofcentrioles, which serve as centrosomal organizer;and a surrounding electron-dense matrix calledpericentriolar material (PCM). PCM serves as aframework to anchor microtubules and containsgamma-tubulin ring complexes, which nucleatemicrotubules, although other proteins also appearto be involved in this process (GERGELY andBASTO, 2008).

The structure of the centrioles is conserved fromancient eukaryotes to mammals. They are barrel-shaped organelles, 100-250 nm in diameter and100-500 nm in length, with ninefold symmetricarray of single, doublet or triplet microtubules.Centriolar characteristics determine mostproperties, such as stability, dynamicity and pola -rity of the centrosome (ACILAN and SAUNSERS,2008). Studies using antibodies to poly gluta -mylated tubulin to ablate centrioles confirmed thatcentrosome size and organization of PCM dependon centriolar integrity (BORNENS, 2002). It has,also, been demonstrated how the acquisition of thepericentriolar material is mechanistically linked tocentriole duplication, thus determining finalcentrosome size (BOBINNEC et al., 1998). Thus,single centrioles, and even small centrioles, canrecruit the PCM, but centriole loss leads in generalto PCM dispersal. The ability of centrioles toduplicate is therefore central to the reproductivecapacity and function of the centrosome.

In addition, the centrioles have another distinctfunction as basal bodies to seed the growth offlagella and to template the assembly either of aprimary cilium or of a beating cilia duringciliogenesis in specialized cells. Recent studieshave demonstrated that cilia have crucial roles inan increasing number of cellular and deve -lopmental processes, establishing a link be tweendysfunctional cilia and several genetic diseases (forreview see DELATTRE et al., 2006; SLUDER et al.,1989; DAVIS et al., 2006). For example, inability ofcentrioles to properly migrate prior to ciliaryassembly has recently been linked to Meckel-Gruber syndrome (DAWE et al., 2007a).Additionally, proper orientation of cilia viacentriole positioning towards the posterior ofembryonic node cells is critical for establishingleft-right asymmetry during mammaliandevelopment (ELEY et al., 2005). However, howcentrioles and basal bodies are assembled and howtheir numbers are controlled within cellsconstitute long-standing unresolved questions.

– 62 –

At the onset of the centrosome duplicationcycle, the two tightly apposed centrioles, motherand daughter, split slightly from one another.Mother centrioles are so-called because they wereassembled in a previous cell cycle to the daughtercentriole. At the G1/S transition new centriolesgrowth orthogonally from each of the two pre-existing centrioles. The new centrioles continue toelongate during late G2 and remain strictlyorthogonal (engaged) until early M phase (DAWE

et al., 2007b). Upon entry into mitosis the twocentrosomes separate from each and participate inmitotic spindle organization. Centriole pairsbecome disengaged at the transition frommetaphase to anaphase, losing their orthogonalconfiguration. According to a recently proposedmodel (NONAKA et al., 2005), the engagement ofnew duplicated centrioles blocks their furtherduplication, whereas their disengagement at theend of mitosis represents a licensing mechanismfor ensuring that centrioles duplicate only once inevery cell cycle. Separase, a protease involved insevering sister chromatid cohesion during mitosisand meiosis, has been recently implicated in thisprocess (KURIYAMA and BORISY 1981). However, itremains to be seen whether the overexpression ofseparase induces ectopic centriole duplication.Interestingly, the overexpression of SAK/PLK-4and SAS-6 in human tissue cultured cells andDrosophila embryos leads to centriole ampli -fication (TSOU and STEARNS, 2006), also in theabsence of pre-existing centrioles (HABEDANCK etal., 2005; BETTENCOURT-DIAS et al., 2005; PEEL etal, 2007), suggesting that these proteins could be apotential players in the licensing mechanism forthe regulation of centriole duplication. HsSAS-6has been, indeed, shown to be a positive regulatorof procentriole formation during the centrosomeduplication cycle of HeLa cells (RODRIGUES-MARTINS et al., 2007a).

CENTRIOLE ASSEMBLY

The general mechanisms that orchestratecentriole assembly remain poorly understood,although substantial progress has recently beenmade in invertebrates. Studies in Caenorhabditiselegans have identified five proteins as beingessential for centriole formation (BETTENCOURT-DIAS and GLOVER, 2007, for a review). Inparticular, the SPD-2 protein acts upstream incentriole assembly pathway by recruiting theZYG-1 kinase to the site of daughter centrioleformation. This process then leads to therecruitment of SAS-5 and SAS-6, two coiled-coil

proteins needed for central tube assembly, astructure onto which singlet microtubules aresubsequently assembled in a SAS-4-dependentmanner (RODRIGUES-MARTINS et al., 2007b).Homologs of nematode SAS-4 and SAS-6 werealso required for centriole biogenesis in Drosophila(HABEDANCK et al., 2005; PEEL et al., 2007). It hasalso been proposed that the Polo kinase familymember SAK/PLK-4, that is essential for centrioleduplication in both human cells and Drosophila(TSOU and STEARNS, 2006) could be related toZYG-1. Downregulation and overexpression ofSAK is associated with cancer in humans, mice andflies. The regulation of SAK is thus crucial in thecontrol of centriole number and tumo rigenesis.

The characteristic orthogonal disposition of thecentrioles within the centrosome has led to thesuggestion that centriole duplication inproliferating cells requires the presence of a pre-existing centriole acting as “template” for theformation of a daughter. However, the findings insome biological systems, that centrioles can beformed de novo, without pre-existing template,have blurred this dogma. Studies of multi-ciliatedepithelial cells (VLADAR and STEARNS, 2007), earlymouse embryos (CALARCO-GILLAM et al., 1983),Chlamydomonas (MARSHALL et al., 2001) andparthenogenetic insects (RIPARBELLI and CALLAINI,2003; RIPARBELLI et al., 2005) clearly show thatcentrioles can be assembled de novo. Strikingly, denovo assembly can also occur in vertebratesomatic cells when the centrosomes are destroyedor removed (LA TERRA et al., 2005). Moreover, thepremise that each maternal centriole serves as atemplate for the formation of one and only onedaughter centriole during each cell division cycledoes not readily explain the rapid induction ofaberrant centriole numbers detected under certainexperimental conditions.

SAK/PLK4, an upstream regulator of canonicalcentriole biogenesis, can promote centrosomeamplification in Drosophila embryos andunfertilised eggs. In embryos this is initiated onthe gonomeric spindle in a process that appears tobe templated from the sperm basal body. Becausecentrioles are eliminated from the oocyte duringoogenesis, their generation in unfertilised eggsexpressing SAK/PLK4 must represent their denovo formation. Both SAK/PLK4 promotedtemplated and de novo centrosome amplificationrequire the activity of DSAS-6 and DSAS-4, twoconserved molecules required for centrioleduplication. This suggests the existence of auniversal self-organizing process of centrioleassembly.

Recent studies in C. elegans have highlighted a

– 63 –

group of molecules necessary for centrioleassembly (BETTENCOURT-DIAS and GLOVER,2007). ZYG-1 kinase recruits a complex of twocoiled-coil proteins, SAS-6 and SAS-5, which arenecessary to form the C. elegans centriolar tube, ascaffold in centriole formation (LEIDEL et al.,2005; DELATTRE et al., 2006; PELLETIER et al.,2006). This complex also recruits SAS-4 which isimportant for the assembly of the centriolarmicrotubules that decorate that tube. DrosophilaSAS-6 is necessary for centriole assembly andcohesion. Overexpression of DSAS-6 in syncitialembryos led to the de novo formation of multiplemicrotubule organizing centers (MTOCs).Strikingly, the center of these MTOCs did notcontain centrioles, as described previously forSAK/PLK4 overexpression (RODRIGUES-MARTINS

et al., 2007b). Instead, tube-like structures werepresent, supporting the idea that centriolarassembly starts with the formation of a tube-likescaffold, dependent on DSAS-6. In DSAS-6 loss offunction mutants, centrioles failed to close and toelongate the structure along all axes of the 9-foldsymmetry suggesting modularity in centrioleassembly.

Overexpression of SAK/PLK4 is sufficient toinduce amplification of centrioles of normalultrastructure in Drosophila embryos and their denovo formation in unfertilised eggs. Both theseprocesses require the activity of DSAS-6 andDSAS-4, two regulators of canonical duplication.Centriole biogenesis is a template-free self-assembly process that is locally triggered andregulated by molecules that ordinarily associatewith the existing centriole. This suggests that therole of the mother centriole is not one of a bona-fide «template» but a platform for a set ofregulatory molecules, hence catalyzing andregulating daughter centriole assembly.

Centrosome amplification is normally inhibitedby degradation of SAK and SAK is targeted fordegradation by the SCF/Slimb ubiquitin ligase.This complex physically interacts with SAK, andin its absence, SAK levels accumulate, leading tothe striking formation of multiple daughtercentrioles surrounding each mother. Thisinteraction is mediated via a conserved Slimbbinding motif in SAK. Mutations in this degronimpair the interaction with Slimb, stabilise SAKlevels and lead to centrosome amplification(CUNHA-FERREIRA et al., 2009). This regulation islikely to be conserved, because knockout ordepletion of the orthologue of Slimb, ß-Trcp1 inmice, also shows centrosome amplification(GUARDAVACCARO et al., 2003). Since the SCF/ß-Trcp complex plays an important role in cell cycle

progression, these results lead to newunderstanding on the control of centrosomenumber and how it may go awry in human disease.Drosophila eggs and embryos should, therefore,provide an ideal experimental system for thefurther analyses of these phenomena and how itmay go awry in cancer. The fact that bothcanonical and de novo centriole formation dependon SAK/PLK4, DSAS-6 and DSAS-4 suggeststhese as good candidates to target to controlaberrant centriole formation in human disease.

RIASSUNTO

LA SPERMATOGENESI DI DROSOPHILA:UN SISTEMA MODELLO PER LO STUDIO

DEL CICLO CELLULARE

Drosophila melanogaster rappresenta un ottimo sistemaper esaminare in dettaglio i meccanismi della divisione cellu-lare e per capire il processo di duplicazione dei centrioli o deicentrosomi. La meiosi maschile è partico larmente adatta allostudio della regolazione del ciclo cellulare e del processo diformazione del fuso e citodieresi. La meiosi maschile manca,infatti, di checkpoints. Questo consente alle cellule di pros-eguire nel ciclo cellulare nonostante possano compariredifetti di organizzazione del fuso, dei centrosomi o allinea-mento incompleto dei cromosomi. Il superamento dei con-sueti blocchi in presenza di difetti nell’assemblaggio del fusoo di danni al DNA, porta ad una particolare condizione in cuii problemi sono ingigantiti, diventando così evidenti e piùfacilmente studiabili. La genetica di Drosophila offre anchel’opportunità di studiare il ruolo di specifici regolatori coin-volti nei processi di assemblaggio e duplicazione dei cen-trioli, come SAK/PLK4 e SAS-6. La produzione di moschetransgeniche per questi geni fornisce utili informazioni percapire come l’assemblaggio e la duplicazione dei centriolisono regolati durante il ciclo cellulare.

REFERENCES

ACILAN C., SAUNSERS W.S., 2008 – A tale of too manycentrosomes. Cell, 134: 572.

ALPHEY L., JIMENEZ J., WHITE-COOPER H., DAWSON I.,NURSE P., GLOVER D., 1992 – twine, a cdc25 homologthat functions in the male and female germline ofDrosophila. Cell, 69: 977.

AVIDES M.C., GLOVER D.M., 1999 – Abnormal spindleprotein, Asp, and the integrity of mitotic centrosomalmicrotubule organizing center. Science, 283: 1733.

BETTENCOURT-DIAS M., GLOVER D.M., 2007 – Centrosomebiogenesis and function: centrosomics brings newunderstanding. Nat. Rev. Mol. Cell Biol. 8: 451.

BETTENCOURT-DIAS M., RODRIGUES-MARTINS A.,CARPENTER L., RIPARBELLI M., LEHMANN L., GATT

M.K., CARMO N., BALLOUX F., CALLAINI G., GLOVER

D.M., 2005 – SAK/PLK4 is required for centrioleduplication and flagella development. Curr. Biol. 15:2199

BOBINNEC Y., KHODJAKOV A., MIR L.M., RIEDER C.L.,EDDÉ B., BORNENS M., 1998 – Centriole disassembly invivo and its effect on centrosome structure and function invertebrate cells. J. Cell Biol. 143: 1575.

BORNENS M., 2002 – Centrosome composition and

– 64 –

microtubule anchoring mechanisms. Curr. Opin. CellBiol. 14: 25.

CALABARCO-GILLAM P.D, SIEBERT M.C., HUBBLE R.,MITCHISON T., KIRSCHNER M., 1983 – Centrosomedevelopment in early mouse embryos as defined by anautoantibody against pericentriolar material. Cell, 35: 621.

CASTRILLON, D.H., GONCZY P., ALEXANDER S., RAWSONR., EBERHART C.G., VISWANATHAN S., DINARDO S.,WASSERMAN S.A., 1993 – Toward a molecular geneticanalysis of spermatogenesis in Drosophila melanogaster:characterization of male-sterile mutants generated bysingle P element mutagenesis. Genetics 135, 489.

CENCI G., BONACCORSI S., PISANO C., VERNI F., GATTI M.,1994 – Chromatin and microtubule organization duringpremeiotic, meiotic and early postmeiotic stages ofDrosophila melanogaster spermatogenesis. J. Cell Sci.107: 3521.

DAVIS E.E, BRUECKNER M, KATSANIS N. 2006 – Theemerging complexity of the vertebrate cilium: newfunctional roles for an ancient organelle Dev Cell. 11: 9

DAWE H.R., FARR H., GULL K.. 2007a – Centriole/basalbody morphogenesis and migration during ciliogenesis inanimal cells. J. Cell Sci. 120: 7.

DAWE H.R., SMITH U.M., CULLINANE A.R., GERRELLI D.,COX P., BADANO J.L., BLAIR-REID S., SRIRAM N.,KATSANIS N., ATTIE-BITACH T., AFFORD S.C., COPP A.J.,KELLY D.A., GULL K., 2007b – The Meckel-GruberSyndrome proteins MKS1 and meckelin interact and arerequired for primary cilium formation. Hum. Mol. Genet.16: 173.

DELATTRE M., CANARD C., GONCZY P., 2006 – Sequentialprotein recruitment in C. elegans centriole formation.Curr. Biol. 16: 1844.

EBERHART C.G., MAINS J.Z., WASSERMAN S.A., 1996 –Meiotic cell cycle requirement for a fly homologue ofhuman Deleted in Azoospermia. Nature 381: 783.

ELEY L, YATES L.M., GOODSHIP J.A., 2005 – Cilia anddisease. Curr. Opin. Genet. Dev. 15: 308.

FABRIZIO J.J., HIME G., LEMMON S.K., BAZINET C., 1998 –Genetic dissection of sperm individualization inDrosophila melanogaster. Development 125: 1843.

FULLER M.T., 1993 – Spermatogenesis. In: Bates, M.,Arias, A.M. (Eds.), The Development of Drosophilamelanogaster, vol. I, Cold Spring Harbor LaboratoryPress, Cold Spring Harbor, New York, 1993, pp.71-147.

GERGELY F., BASTO R., 2008 – Multiple centrosomes:together they stand, divided they fall. Genes Dev. 22:2291.

GLOVER D., LEIBOWITZ M.H., MCLEAN S.A., PARRY H.,1995 – Mutations in aurora prevent centrosomeseparation leading to the formation of monopolarspindles. Cell 81: 95.

GONZALEZ C., TAVOSANIS G., MOLLINARI C., 1998 –Centrosomes and microtubule organisation duringDrosophila development. J. Cell Sci. 111: 2697.

GUARDAVACCARO D., KUDO Y., BOULAIRE J., BARCHI M.,BUSINO L., DONZELLI M., MARGOTTIN-GOGUET F.,JACKSON P.K., YAMASAKI L., PAGANO M., 2003 – Controlof meiotic and mitotic progression by the F box proteinbeta-Trcp1 in vivo. Dev. Cell 4: 799.

HABEDANCK R., STIERHOF Y.D., WILKINSON C.J., NIGGE.A., 2005 – The Polo kinase Plk4 functions in centrioleduplication. Nat. Cell Biol. 7: 1140.

HARDY R.W., TOKUYASU K.T., LINDSLEY D.L., GARAVITOM., 1979 – The germinal proliferation center in the testisof Drosophila melanogaster. J. Ultrastruct. Res. 69: 180.

CUNHA-FERREIRA I., RODRIGUES-MARTINS A., BENTO I.,

RIPARBELLI M., ZHANG W., LAUE E., CALLAINI G.,GLOVER D.M., BETTENCOURT-DIAS M., 2009 – TheSCF/Slimb ubiquitin ligase limits centrosomeamplification through degradation of SAK/PLK4. Curr.Biol. 19: 1.

KURIYAMA R, BORISY G.G., 1981 – Centriole cycle inChinese hamster ovary cells as determined by whole-mount electron microscopy. J. Cell Biol. 91: 814.

LA TERRA S., ENGLISH C.N., HERGERT P., MCEWEN B.F.,SLUDER G., KHODJAKOV A., 2005 – The de novocentriole assembly pathway in HeLa cells: cell cycleprogression and centriole assembly/maturation. J. CellBiol. 168: 713.

LEIDEL S., DELATTRE M., CERUTTI L., BAUMER K.,GÖNCZY P., 2005 – SAS-6 defines a protein familyrequired for centrosome duplication in C. elegans and inhuman cells. Nat. Cell Biol. 7: 115.

LLAMAZARES S., MOREIRA A., TAVARES A., GIRDHAM C.,SPRUCE B.A., GONZALEZ C., KARESS R.E., GLOVER D.,SUNKEL C.E., 1991 – Polo encodes a protein kinasehomolog required for mitosis in Drosophila. Genes Dev.5: 2153.

MAINES J., WASSERMAN S., 1998 – Regulation andexecution of meiosis in Drosophila males. Curr. Top.Dev. Biol. 37: 301.

MARSHALL W.F., VUCICA Y., ROSENBAUM J.L., 2001 –Kinetics and regulation of de novo centriole assembly.Implications for the mechanism of centriole duplication.Curr. Biol. 11: 308.

NIGG E.A., 2002 – Centrosome aberrations: cause orconsequence of cancer progression? Nat. Rev. Cancer. 2:815.

NIGG E.A., 2006 – Origins and consequences of centrosomeaberrations in human cancers. Int. J. Cancer. 119: 2717

NONAKA S., YOSHIBA S., WATANABE D., IKEUCHI S., GOTOT., MARSHALL W.F., HAMADA H.. 2005 – De novoformation of left-right asymmetry by posterior tilt of nodalcilia. PLoS Biol. 3: 268.

PEEL N., STEVENS N.R., BASTO R., RAFF J.W., 2007 –Overexpressing centriole-replication proteins in vivoinduces centriole overduplication and de novo formation.Curr. Biol. 17: 834.

PELLETIER L., O’TOOLE E., SCHWAGER A., HYMAN A.A.,MULLER-REICHERT T.. 2006 – Centriole assembly inCaenorhabditis elegans Nature 444: 619.

RIPARBELLI M.G., CALLAINI G., GLOVER D.M., DO CARMOAVIDES M., 2002 – A requirement for the AbnormalSpindle protein to organise microtubules of the centralspindle for cytokinesis in Drosophila. J. Cell Sci. 115, 913

RIPARBELLI M.G., CALLAINI G., 2003 – Drosophila parthe -no genesis: a model for de novo centrosome assembly. Dev.Biol. 260: 298.

RIPARBELLI M.G., TAGU D., BONHOMME G., CALLAINI G.,2005 – Aster self-organization at meiosis: a conservedmechanism in insect parthenogenesis? Dev. Biol. 278, 220

RODRIGUES-MARTINS A., BETTENCOURT-DIAS M.,RIPARBELLI M., FERREIRA C., FERREIRA I., CALLAINI G.,GLOVER D.M., 2007a. – DSAS-6 organizes a tube-likecentriole precursor, and its absence suggests modularity incentriole assembly. Curr. Biol. 17: 1465.

RODRIGUES-MARTINS A., RIPARBELLI M., CALLAINI G.,GLOVER D.M., BETTENCOURT-DIAS M., 2007b –Revisiting the role of the mother centriole in centriolebiogenesis. Science 316: 1046.

ROTHWELL W.F., SULLIVAN W., 2000 – The centrosome inearly Drosophila embryogenesis. Curr. Top. Dev. Biol.49: 409.

– 65 –

SLUDER G., MILLER F.J., RIEDER C.L., 1989 – Reproductivecapacity of sea urchin centrosomes without centrioles. CellMotil. Cytoskeleton 13: 264.

SLUDER G., NORDBERG J.J., 2004 – The good, the bad andthe ugly: the practical consequences of centrosomeamplification. Curr. Opin. Cell Biol. 16: 49.

SUNKEL C.E., GLOVER D.M., 1988 – Polo, a mitotic mutantof Drosophila displaying abnormal spindle poles. J. CellSci. 89: 25.

TOKUYASU K.T., PEACOCK W.J., HARDY R.W., 1972 –Dynamics of spermiogenesis in Drosophila melanogaster.

I. Individualization process. Z. Zellforsch. Mikrosk. Anat.124: 479.

TSOU M.F., STEARNS T. 2006 – Mechanism limitingcentrosome duplication to once per cell cycle. Nature 442:947.

VLADAR E.K., STEARNS T, 2007 – Molecularcharacterization of centriole assembly in ciliated epithelialcells. J. Cell Biol. 178: 31.

WOJCIK E.J., GLOVER D., HAYS T.S., 2000 – The SCFubiquitin ligase protein slimb regulates centrosomeduplication in Drosophila. Curr. Biol. 10: 1131.

– 66 –

GENETICA MOLECOLARE DELLO SVILUPPO DEGLI INSETTI XVII.pdf · DELLO SVILUPPO DEGLI INSETTI Tavole...

Documents

Transcript of GENETICA MOLECOLARE DELLO SVILUPPO DEGLI INSETTI XVII.pdf · DELLO SVILUPPO DEGLI INSETTI Tavole...