2055RESEARCH ARTICLE

INTRODUCTIONRegulation of the body shape has been a central issue indevelopmental and evolutionary biology. The skeleton provides theframework of animal body shape. In vertebrates, bones act as theinner core of the body, although there are some skeletal elementsthat appear and function on the exterior, such as teeth, beaks, scalesand antlers. Skeletons of invertebrates are predominantly external:cuticles on the body surface. Intensive studies have revealed thatgenetically programmed body plans ultimately dictate skeletalpatterns. However, cellular mechanisms that directly controlskeletal morphology downstream of those genetic programs remainunknown.

The skeleton is composed of extracellular matrix (ECM),consisting of molecules secreted from cells. Bones comprisecollagen and proteins secreted by osteoblasts (Hall, 2005), whereasinvertebrate cuticles consist of proteins and chitin, a long-chainsugar polymer, released by the underlying epidermal cells(Chapman, 1998). The final morphology of skeletal ECM shouldreflect not only the terminal arrangement of ECM-secreting cellsbut also the history of how their number, position, shape andactivity have changed during the entire ECM formation period. Tounderstand the cellular mechanisms for skeletal morphogenesis, wetherefore need to look at the dynamic behaviors of those cells andto determine how they relate to ECM morphology.

Stringent regulation of ECM morphology is particularlyimportant and prominent in the joints, which are the flexible partsof the skeleton that permit movement. The human skeleton

possesses joints of various configurations, such as ball-and-socket,hinge, saddle and pivot (Standring et al., 2005). In eachconfiguration, opposing sides of the ECM must be shaped in aprecisely reciprocal manner to form a tightly interlocking structurefor efficient and controlled motion. Although the vertebratesynovial joints have long received attention for their medicalimportance, the mechanisms by which they actually form duringembryogenesis have been poorly understood (Khan et al., 2007;Pacifici et al., 2005).

The exoskeletal system offers some technical advantages toaddress this issue. The epidermis consists of a single, continuoussheet of epithelial cells that are polarized along the apicobasal axis:the apical surfaces of the cells face the body surface and secretecuticle. The epidermis maintains its integrity and apicobasalpolarity and continues to adhere to the cuticle without mixing withit (Chapman, 1998). Within the epidermal sheet, cell identities arespecified by two-dimensional positional cues (Cohen, 1993; Payre,2004). This specification process is separated in time from the laterdifferentiation, when cells actually make specific parts of thecuticle (Fristrom and Fristrom, 1993). Moreover, morphogenesis ofthe adult cuticle appears to be completed before the muscles startto contract just prior to eclosion. All of these features make theexoskeleton a promising system for studying the dynamicbehaviors of cells during ECM formation and how they are linkedto the final ECM morphology.

We focus on the ball-and-socket joints in the tarsus of theDrosophila leg. How their positions are determined in the legprimordium before pupariation has been studied extensively(Kojima, 2004), and the central role of the Notch signalingpathway therein has been demonstrated (de Celis et al., 1998;Rauskolb and Irvine, 1999). The subsequent folding (Manjon et al.,2007), unfolding and re-constriction (Fristrom and Fristrom, 1993;Mirth and Akam, 2002) of the epidermis in the joint regions havealso been described. By contrast, the way in which the joint cuticlesdevelop afterwards and how the cells behave during that period arelargely unknown. As it takes more than three days from

Development 137, 2055-2063 (2010) doi:10.1242/dev.047175© 2010. Published by The Company of Biologists Ltd

1Laboratory for Morphogenetic Signaling, and 2Electron Microscope Laboratory,RIKEN Center for Developmental Biology, Kobe 650-0047, Japan. 3Department ofBiology, Kobe University Graduate School of Science, Kobe, Hyogo 657-8501, Japan.

*Author for correspondence (shayashi@cdb.riken.jp)

Accepted 21 April 2010

SUMMARYAnimal body shape is framed by the skeleton, which is composed of extracellular matrix (ECM). Although how the body planmanifests in skeletal morphology has been studied intensively, cellular mechanisms that directly control skeletal ECM morphologyremain elusive. In particular, how dynamic behaviors of ECM-secreting cells, such as shape changes and movements, contribute toECM morphogenesis is unclear. Strict control of ECM morphology is crucial in the joints, where opposing sides of the skeletonmust have precisely reciprocal shapes to fit each other. Here we found that, in the development of ball-and-socket joints in theDrosophila leg, the two sides of ECM form sequentially. We show that distinct cell populations produce the ‘ball’ and the ‘socket’,and that these cells undergo extensive shape changes while depositing ECM. We propose that shape changes of ECM-producingcells enable the sequential ECM formation to allow the morphological coupling of adjacent components. Our results highlightthe importance of dynamic cell behaviors in precise shaping of skeletal ECM architecture.

KEY WORDS: Joint, Ball-and-socket, Cuticle, Chitin, Epidermis, Cell shape change, Leg, Drosophila

Dynamic shape changes of ECM-producing cells drivemorphogenesis of ball-and-socket joints in the fly legReiko Tajiri1, Kazuyo Misaki2, Shigenobu Yonemura2 and Shigeo Hayashi1,3,*

DEVELO

PMENT

2056

specification of the joint regions to the completion of the jointcuticle architecture, the essence of ECM morphogenesis must liein how cells act during this period.

In this study, we describe the development of the ball and thesocket cuticles and the concomitant cellular morphogenesis, whichoccur long after cell specification. Through chronologicalobservation of cellular and cuticular architectures, we reveal thatthe ball and the socket cuticles develop in a sequential manner, andthat the two parts are formed by distinct cell populations thatundergo extensive shape changes while secreting cuticle. Wepropose that the shape changes of ECM-producing cells result inthe sequential formation of the ball and the socket, permitting theirmorphological coupling in a ‘mold-and-cast’ manner.

MATERIALS AND METHODSFly strains and confocal microscopyFly strains used were: sqh-GFP-Moesin (Kiehart et al., 2000), MyosinRegulatory Light Chain (MRLC)-GFP (Royou et al., 2004), UAS-Serp(CBD)GFP (Luschnig et al., 2006), UAS-GFP-TTras (Kato et al.,2004), UAS-krotzkopf verkehrt (kkv) RNAi (Dietzl et al., 2007), Distal-lessmd23 (Dll-GAL4) (Calleja et al., 1996), big brainNP5149 (bib-GAL4) andfringeNP6583 (fng-GAL4) (Hayashi et al., 2002), neuralizedP72 (neur-GAL4)(Bellaiche et al., 2001), UAS-KO, ubi-KOnls (Kaido et al., 2009), ubi-GFP.S65Tnls, UAS-GFP.S65T and dauterless (da)-GAL4 (BloomingtonStock Center). sqh-GFP-Moesin or MRLC-GFP was used as wild typeunless otherwise noted.

Confocal microscopy and leg stainingOur method is based on that described by Mirth and Akam (Mirth andAkam, 2002). Briefly, pupal thoraces were dissected in PBS and transferredto 4% paraformaldehyde in PBS. After removal of pupal cuticles, the legswere fixed for 20-30 minutes at room temperature and washed with PBST.For GFP and KO, the legs were directly stored overnight in Vectashield withor without DAPI (Vectashield) and then mounted on glass slides. Forquantification of GFP (see Fig. S3D in the supplementary material), confocalimages of the legs were captured [Olympus FV-1000, �60 water immersionlens, NA1.2; LD473 nm laser (15 mW) at 0.1% intensity, PMT 600 V,aperture 120 m] and fluorescence intensities were measured by manuallyoutlining cells of interest using ImageJ (National Institutes of Health, USA).

For tissue staining, fixed legs were blocked in PBST containing 3%normal goat serum for 30 minutes, incubated overnight at 4°C withantibodies or fluorescent probes and washed with PBST before mounting.The following probes and dilutions were used: mouse anti--galactosidase

antibody (Promega, 1:100), Cy3-conjugated anti-mouse IgG (MolecularProbes, 1:400), rhodamine-conjugated chitin-binding probe (New EnglandBiolabs, 1:100), and Fluostain (Calcofluor White Stain, Fluka, 2 g/ml).Antibody staining was effective for stages before cuticle deposition [~48hours after puparium formation (APF)]. Chitin binding probe labels the balland the socket cuticles. The staining within the ball was non-uniform,probably reflecting incomplete penetration of the probe due to its size (48kDa; see Figs 1 and 3). Fluostain gave more uniform and reproduciblepatterns in the ball (Moussian et al., 2005).

Verification of kkv RNAiIn order to verify the effect of RNAi against kkv, progeny from the crossingof males with two copies of UAS-kkv RNAi and da-GAL4 females werestained with Fluostain (Moussian et al., 2005) or processed for cuticlepreparation with or without the vitelline membrane in Hoyer’s medium(Sullivan et al., 2000).

The effect of kkv RNAi on chitin accumulation in the ball cuticle wasquantified as follows (see Fig. S3A-C in the supplementary material). Legsfrom RNAi (bib-GAL4/UAS-kkv RNAi; UAS-GFP-TTras/+) and siblingcontrol (CyO/UAS-kkv RNAi; UAS-GFP-TTras/+) pupae (at least four foreach genotype) were pooled and divided into two batches: one forFluostain staining and the other for negative control for backgroundmeasurement. Confocal images were taken [Olympus FV-1000, �60 waterimmersion lens, NA1.2; LD405 nm laser (50 mW) at 0.1% intensity,PMT 550 V, aperture 120 m] and fluorescence intensity was measuredusing ImageJ (see Fig. S3A� in the supplementary material).

Time-lapse imagingPupae were removed from pupal cases and placed on glass-bottomeddishes. Time-lapse images were taken with an Olympus FV1000 confocalscanning microscope, and processed using iSEMS (Kato and Hayashi,2008).

Electron microscopyThoraces were dissected from pupae in PBS, and legs were immediatelytransferred to fixation buffer (2.5% glutaraldehyde, 2% formaldehyde,0.1 M cacodylate buffer). Pupal cuticles were removed in the buffer, andlegs were incubated for >2 hours at room temperature. Fixed legs wereembedded in 1% agarose and subjected to post-fixation with 1% OsO4 in0.1 M cacodylate buffer for 8 hours on ice, and then to en bloc staining in0.5% uranyl acetate aqueous solution overnight at room temperature.Samples were subsequently dehydrated in ethanol and embedded inPoly/Bed812 (Epon 812, Polysciences). Sections stained with uranylacetate and lead citrate were viewed in a JEOL JEM-1010 transmissionelectron microscope at 100 kV.

RESEARCH ARTICLE Development 137 (12)

Fig. 1. Mature joint morphology. (A)Cuticle preparation of the adult leg (tarsus). ta1-ta5, tarsomeres 1-5. Arrows indicate the joints focused onin this study; arrowhead indicates the joint between ta4 and 5. (B-D)Mature structure of the three proximal inter-tarsomere joints (arrows in A).(B-B�) Electron micrograph of the mature joint. Boxed areas in B are enlarged in B� and B�. (C,C�) Confocal microscope image of the mature joint.(C)Differential interference contrast (DIC) image. (C�)The ball and socket cuticles are labeled by a chitin-binding probe (magenta). GFP-Moesinlabels actin enrichment at the cell cortices (green). (D)Serp(CBD)GFP (green) is overexpressed using Dll-GAL4 in the mature joint. Cuticles arelabeled by a chitin-binding probe (magenta). (E,E�) The joint between ta4 and ta5, labeled as in C and C�. Dorsal is up and distal is to the right inthis and all subsequent figures unless otherwise noted. so, socket cuticle; lu, lubricant matrix. Scale bars: 2m in B; 1m in B�, B�; 5m in C-E�. D

EVELO

PMENT

Nikkomycin injectionPupae at 40-45 hours APF (a few hours prior to the onset of cuticleformation) were sterilized by soaking in 70% ethanol and placed on plasticdishes with adhesive tape. Nikkomycin Z (Sigma-Aldrich) was dissolvedin water (2.5 mg/ml) and 0.5~1.0 l of the solution was injected into thethoracic region of pupae by piercing a hole in the pupal case with a glassneedle. Water injection was used as a control. Approximately 24 hours afterinjection at 25°C, the legs were prepared for viewing under a confocalmicroscope. Cuticle preparations in Gary’s Magic Mountant (Roberts,1986) were made from pupae incubated for more than 48 hours afterinjection.

RESULTSMature joint morphologyThe tarsus comprises five segments (tarsomeres), separated byjoints (Fig. 1A). On the dorsal side of the inter-tarsomere joints, thecuticle has the ball-and-socket architecture (Fig. 1). The jointbetween tarsomeres 4 and 5 had a slightly different ball-and-socketmorphology (Fig. 1E), with a smaller ball and a thicker socket thanthe other three (Fig. 1C,D). In this study, we focused on the threeproximal inter-tarsomere joints, which develop simultaneously.Both the ball and the socket were labeled by a chitin-binding probe(Fig. 1C�,D,E�). The ball-and-socket structure was tucked under theridge, on which the socket cuticle spreads (Fig. 1B-C�). On theventral aspect of the ball, a distally oriented process, which wenamed the ‘lip’, grew into the epithelium (Fig. 1B-C�). Electronmicroscopy (EM) revealed a strikingly regular stratification of theball cuticle, consisting of 70-80 layers of approximately 100-150 nm thickness (the variation might depend on the angles of thinsections; Fig. 1B�). It also uncovered the matrix that filled the spacebetween the ball and the socket, sealed by the membranous cuticle(Fig. 1B). A secreted form of GFP [Serp(CBD)GFP (Luschnig etal., 2006)] was enriched in this matrix when it was expressed byunderlying cells (Fig. 1D). This matrix was labeled poorly with thechitin-binding probe and we speculate that it acts as a lubricant inthe moving adult legs.

Sequential development of the jointThe leg primordium is composed of an epithelium with deepsegmental folds during the prepupal period (de Celis et al., 1998)that then elongate to become a straight and smooth cylinder by theprepupa-pupa transition (apolysis) (Fristrom and Fristrom, 1993;Mirth and Akam, 2002). At this stage (~20 hours APF, Fig. 2A),cells in the joint region were aligned along the circumference (Fig.2A�). They underwent apical constriction (data not shown),resulting in the constriction of the leg in the joint regions (Fig.2B,B�). When the entire leg subsequently became thinner (Fig. 2C),the cells in the joint region remained neatly aligned and constricted(Fig. 2C�). Cells on the dorsal side of the joint constricted furtherand infolded proximally (Fig. 2D-D�). By ~45 hours APF, thefurrow grew deeper and resulted in a deep cavity (Fig. 2E-E�).

Cuticle deposition in the joint region began at 45-48 hours APFwhen the chitin-binding probe revealed signals covering both thedorsal and the ventral sides of the cavity (Fig. 3A-A�). The cuticlein the deepest part (bottom) of the cavity then thickened to formthe ball cuticle (Fig. 3B-D�). Electron micrographs indicated thatlayers of consistent thickness were added successively to the ballcuticles (Fig. 4B,C). At 63-66 hours APF, the ball cuticle reachedits full size.

At this same stage, the socket cuticle was formed only on thedorsal side of the cavity (Fig. 3D,D�; Fig. 4C-C�). The socketcuticle then expanded gradually in the dorsal-to-ventral direction

towards the bottom of the cavity (Fig. 3E-E�), until eventuallycovering the entire cavity at ~75 hours APF (Fig. 3F-F�; Fig. 4D).During this time, the ‘lip’ of the ball cuticle also developed (Fig.3E-F�; Fig. 4D).

Thus, the ball and the socket are produced in a sequentialmanner: the socket cuticle develops throughout the cavity onlyafter the ball has reached its full size. We propose that the serialformation of the ball and the socket serves as ‘mold casting’, whichpermits their morphological coupling (see Discussion).

It has been reported that most of the final cell divisions arecomplete by about 30 hours APF (Fristrom and Fristrom, 1993),suggesting that the sequential formation of the ball and the socketcuticles shown in Fig. 3 progresses without cell proliferation. Wenext addressed the cellular mechanism underlying suchsequentiality.

Sites of ball/socket production revealed byelectron microscopyIn order to study the spatio-temporal pattern of cuticle secretion atthe cellular level, we used EM analysis to monitor the appearanceof the histological structure called plasma membrane plaques

2057RESEARCH ARTICLEBall-and-socket joint morphogenesis

Fig. 2. Cellular morphogenesis before cuticle formation. Thetarsus morphology under DIC (A-E), and cell arrangements in the jointregions (A�-E�,D�,E�) at several stages before the onset of cuticleformation. Cells are outlined by MRLC-GFP (A�-D�,D�) or GFP-Moe(E�,E�). Boxes in A-E indicate the joint regions shown in A�-E�,D�,E�.(A,A�) At 18-22 hours APF, cells in the joint region are aligned along theleg circumference (arrows). (B,B�) By 24-26 hours APF, the leg hasshrunk in the joint regions (brackets). (C,C�) Cells in the joint region at32-34 hours APF continue ordered alignment (brackets). (D-D�) Aprojected image showing the leg surface (D�) and a single section at themidline of the leg (D�) at 38-40 hours APF shows that cells on thedorsal side invaginate proximally (arrows). (E-E�) By 45-48 hours APF, adeep cavity has formed on the dorsal side of the joint region (arrows).E� is a projected image as in D�; E� is a single section as D�. Scale bars:in A, 50m for A,B,C,D,E; in A�, 20m for A�-E�,D�,E�.

DEVELO

PMENT

2058

(PMPs). PMPs are electron-dense regions of the plasma membranelocated at the tips of microvilli in contact with newly synthesizedcuticles of various insects (Chapman, 1998; Locke and Huie,1979), and are regarded as the sites of chitin microfibril releaseand/or cuticle-cell adhesion.

At the onset of ball formation, the apical surfaces of the cellssurrounding the ball cuticle contained many PMPs (Fig. 4A,A�).The PMPs were in contact with fibrous materials, presumablychitin microfibrils, arranged parallel to the ball cuticle (Fig. 4A�).Thus, these cells contacted the ball cuticle, and were probablyadding chitin microfibrils to it. The thin and scattered socket cuticlewas evenly associated with PMPs of the cells on the dorsal side(Fig. 4A).

RESEARCH ARTICLE Development 137 (12)

Fig. 3. Sequential formation of the ball and the socket cuticles.Sequence of ball-and-socket cuticle formation observed by confocalmicroscopy. (A-F)DIC images. (A�-F�) Cuticles are stained with a chitin-binding probe (magenta), and myosin or actin enrichment at cellcortices is labeled by MRLC-GFP in B�-E�, or by GFP-Moe in F� (green).(A�-F�) Schematic representations. (A-A�) At 45-48 hours APF, cuticlesare detected on the dorsal (arrow) and ventral (asterisk) sides of thecavity. (B-B�) By 52-55 hours APF, the ball cuticle on the ventral side ofthe cavity has thickened (B�, asterisk). An arrow marks the edge of thesocket cuticle. (C-C�) The ball cuticle has further enlarged (asterisk) by55-63 hours APF. The socket cuticle has not yet extended significantly(arrow). (D-D�) At 63-66 hours APF, the ball has reached its full size(asterisk). The socket is detected only on the dorsal side of the cavity(arrow). (E-E�) By 70-72 hours APF, the socket is extending along thecavity into the lateral side of the ball (arrow) and the ‘lip’ is developing(arrowhead). (F-F�) By 72-75 hours APF, the socket has extended to theventral side of the ball (arrow) and the lip has fully formed (arrowhead).Scale bar: 5m.

Fig. 4. Distribution of plasma membrane plaques (PMPs) atdifferent stages of cuticle formation. (A-D)Electron micrographsof joints at different stages of cuticle formation. Yellow arrowsindicate PMPs. (A)At 55 hours APF, the ball cuticle is thickening(asterisk) and the socket cuticle is composed of thin and scatteredpieces (bracket). White arrow indicates the edge of the socket.(A’)Magnification of the boxed area in A. Brackets indicatepresumptive chitin microfibrils. (B)At 58 hours APF, the ball continuesto enlarge (asterisk). The edge of the socket is indicated by whitearrow. (B�,B�) Enlargement of the boxed regions in B. White arrowsindicate pieces of the socket cuticle. (C)By 65 hours APF, the ball hasreached its full size (asterisk). The lip has not yet formed in thissample. (C�)Enlargement of the boxed region in C showing that PMPsare scarce on the lateral side of the ball. (C�)Magnification of theboxed region in C showing that the socket cuticle is associated withPMPs. (D)By 72 hours APF, the socket has formed throughout thecavity, reaching the ventral side of the ball cuticle (edge indicated byarrow). (E)Changes in the distribution of PMPs contacting the ballcuticle and those contacting the socket. Scale bars: 1m in A,B,C,D;500 nm in A�,B�,B�,C�,C�. D

EVELO

PMENT

During its growth, the ball cuticle continued to be surrounded byPMPs (Fig. 4B,B�). However, when it had fully grown, the PMPswere relatively sparse on its lateral side (Fig. 4C,C�). The socketcuticle, which was still thin and scattered at this point, continuedto be associated with PMPs (Fig. 4B,B�,C,C�). When the socketcuticle formed in the bottom of the cavity, it was uniformly linedwith PMPs (Fig. 4D). PMPs were also found enriched along theelongated lip.

These results from electron microscopy experiments aresummarized in Fig. 4E. Notably, the positions of the ball-contactingarea (blue) and the socket-contacting area (red), which presumablycomprise ball-producing activity and socket-producing activity,respectively, shift extensively along the cavity during cuticleformation. If cells remain stationary, the ball-producing and socket-producing activities must sequentially progress across the cells, ina wave-like fashion. If, conversely, the ball-producing cells and thesocket-producing cells retain their respective identities as separatepopulations, they must shift their apical surfaces along the cavityby cell movement or shape changes. With this in mind, we nextexamined the cell dynamics during cuticle formation.

Extensive cell shape changes in the joint regionduring cuticle formationWe sought to label subpopulations of cells within the joint regionusing the GAL4-UAS method (Brand and Perrimon, 1993) and tomonitor their locations and shapes at various stages. Among GAL4drivers inserted near the genes reported to show segmentalexpression in the tarsus, we found three drivers which wereexpressed in different but overlapping populations of cells in thejoint region: big brain (bib)-GAL4, neuralized (neur)-GAL4 andfringe (fng)-GAL4 (de Celis et al., 1998; Mirth and Akam, 2002;Rauskolb and Irvine, 1999; Shirai et al., 2007). bib and neur wereexpressed in the deep region of the cavity, flanked by fngexpression in cells projecting into the ridge (see Fig. 1C�) and cellson the ventral side of the cavity (Fig. 5D,G,J). Simultaneousdetection of bib (bib-GAL4 and UAS-GFP) and neur (neur-lacZnls) revealed that bib was expressed distal to neur, with anoverlap of about two cells (Fig. 5A-A�).

GAL4-expressing cells were outlined by a membrane-boundGFP (Fig. 5B-J). Shortly after the onset of ball formation, the bib-expressing cells were in contact with the ball cuticle (Fig. 5B). The

2059RESEARCH ARTICLEBall-and-socket joint morphogenesis

Fig. 5. Cell shape changes during cuticleformation. (A-A�) Comparison of bib and neurexpression patterns. (A)GFP (green) and anti--galimmunostaining (magenta, shown in grey-scale inA�) in the joint region of a bib-GAL4/+; neur-lacZ/UAS-membrane-targeted-GFP pupa. Thesample was fixed at a stage comparable to thatshown in Fig. 3C,C�. neur expression is hardlydetected in the distal portion of bib-expressingcells (bracket). (B-J)Membrane-targeted GFP isexpressed under the control of bib-GAL4 (B,E,H),neur-GAL4 (C,F,I) or fng-GAL4 (D,G,J). Samplesfixed at different stages of cuticle formation areshown. (B-D)The ball cuticle has started to grow(asterisk). (B)bib-expressing cells contact thedorsal part of the ball cuticle (arrow). (C)neur isexpressed in both the dorsal subset of bib-expressing cells (arrow) and the cells located dorsalto them (arrowhead). (D)fng is expressed in cellsprojecting into the ridge. Arrowhead indicates theventral edge of the fng expression domain.(E-G)The ball has reached its full size (asterisk).The lip and the socket are extending. (E)bib-expressing cells contact the ventral side of the ball(arrow). (F)The ventral subset of neur-expressingcells contact the ventral side of the ball (arrow).The dorsal subset contact the extending socket(arrowhead). (G)fng-expressing cells haveexpanded their apical surfaces and broadly contactthe dorsal part of the socket cuticle (arrowhead).(H-J)The socket cuticle has formed along theentire cavity. Its edge is indicated by the ‘v’.(H)bib-expressing cells extend past the ventral sideof the cavity and contact the lip (arrow). Theirbasal sides remain in the proximal position(asterisk). (I)The ventral neur-expressing cellsextend along the ventral side of the cavity (arrow).The dorsal portion contacts the ventral part of thesocket (arrowhead). (J)fng-expressing cells havefurther expanded their apical surfaces along thecavity (arrowhead). (K)Summary of cell shapechanges, overlaid with the distribution of ball- andsocket-contacting PMPs (see Fig. 4).

DEVELO

PMENT

2060

expression of neur appeared to overlap largely with bib-expressingcells and to extend dorsally to them (Fig. 5C; consistent with Fig.5A-A�). bib and neur expression domains were four to six cellswide along the anteroposterior axis (data not shown).

When the ball had fully grown, the apical surfaces of bib-expressing cells came into contact with the ventral side of the ball(Fig. 5E), and then the apical surfaces further shifted distallyeventually contacting the newly formed lip (Fig. 5H).Consistently, the apices of the ventral portion of neur-expressingcells (co-expressing bib) relocated to the ventral side of the ball(Fig. 5F,I), and the apical surfaces of their dorsal subsets movedwith the edge of the extending socket, until covering the ventralpart of the socket (Fig. 5F,I). Meanwhile, the dorsal fng-expressing cells expanded their apical surfaces (Fig. 5G,J), andwhen the socket cuticle formed at the bottom of the cavity, theycovered a large dorsal portion of the socket cuticle (Fig. 5J). Asimilar observation was made on legs in which all of the cells inthe joint region were labeled by Dll-GAL4 (data not shown).Furthermore, tracing the cell shapes in electron micrographsconfirmed extensive changes in the morphology of the cellcontacting the ball and socket cuticles (see Fig. S1 in thesupplementary material).

We also took a live-imaging approach in order to directlymonitor the dynamic behaviors of bib- and neur-expressing cells.Pupal legs were doubly labeled with fluorescent proteins expressedby bib-GAL4 or neur-GAL4, and ubiquitously expressed nuclearlocalized markers (see Movies 1-3 in the supplementary material)and imaged over 5 hours. The movies allowed us to continuouslytrace the bib-GAL4- and neur-GAL4-expressing cells, whichmoved their apical surfaces extensively along the cavity.

Fig. 5K shows a summary of the cell shape changes, overlaid onthe distribution of PMPs at respective stages. Notably, themovement of the apical surfaces of the cells continuouslyparalleled the distribution changes of PMPs. This parallelism isconsistent with the model that the ball-producing cells and thesocket-producing cells retain their respective identities as separatepopulations (see Discussion).

Identification of ball-producing and socket-producing cell populationsTo identify which cutcicular elements are produced by each of thebib-, neur- and fng-expressing cells, we attempted to interfere withcuticle production in a cell-specific manner using transgenic RNAiconstructs of krotzkopf verkehrt (kkv), encoding Chitin Synthase 1.The UAS-kkv RNAi strains used here caused defects phenocopyingthose of kkv mutants: attenuation of chitin accumulation in thetracheal lumen and its failure to expand uniformly (Devine et al.,2005; Tonning et al., 2005), deformation of the head skeleton, andreduction of cuticle strength (Moussian et al., 2005; Ostrowski etal., 2002) (see Fig. S2 in the supplementary material).Measurement of the chitin-staining dye Fluostain demonstrated thatkkv RNAi robustly reduced the amount of chitin in the ball cuticle(see Fig. S3A-C in the supplementary material).

As a positive control for chitin synthesis attenuation, weinjected Nikkomycin Z, a GlcNAc analog that competitivelyinhibits chitin synthase activity (Tellam et al., 2000), into the pupalthoraces before the onset of cuticle formation. Fig. 6B shows thetypical morphology of the joint cuticle that developed in wild-typepupae injected with Nikkomycin. The cuticle consisted of multipleamorphous bulges: one relatively large bulge on the ventral sideof the cavity that presumably derived from the ball, and a fewsmaller ones on the lateral and dorsal sides that probably

correspond to the socket. Thus, the morphology of the entire ball-and-socket cuticle was severely disrupted by the global attenuationof chitin synthesis.

We next used bib- and neur-GAL4 drivers for knock-downexperiments. Activities of those drivers were approximately thesame, as assessed by the measurement of GFP signals expressed bythose drivers (see Fig. S3D in the supplementary material). Whenkkv RNAi was induced by bib-GAL4, the morphology of the ballcuticle was disrupted as severely as seen in Nikkomycin-injectedsamples (Fig. 6C; see also Fig. S3E in the supplementary material).The socket formation was also inhibited, although we could oftendiscern the dorsal portion of the socket. Conversely, when neur-GAL4 was used, the ball cuticle was largely intact, whereas theventral part of the socket was missing (Fig. 6D; see also Fig. S3Ein supplementary material). Thus, neur>kkv RNAi preferentiallyaffected the socket rather than the ball, whereas bib>kkvRNAistrongly affected both (Fig. 6C,D; see also Fig. S3E insupplementary material). The overlap in their phenotypes in thesocket cuticle is likely to reflect the overlap of bib and neurexpression (Fig. 5A-A�; see Discussion). Use of fng-GAL4 causedyet another phenotype: the ball morphology was largely normal,whereas the socket cuticle had a bloated appearance and a fewbulges (Fig. 6E).

Altogether, attenuation of chitin synthesis in specific cellpopulations differentially affected both the ball and the socket, andthe ball-producing activity and the socket-producing activity weremapped to different cell populations along the proximodistal axis,

RESEARCH ARTICLE Development 137 (12)

Fig. 6. Attenuation of chitin synthesis in different cellpopulations exerts differential defects on the ball and thesocket. (A-E)Joint cuticle morphology in situations where chitinsynthase is inhibited. (A)Control (wild-type, injected with water)showing normal mature cuticle morphology. (B)Nikkomycin Z-injectedwild type. The large bulb (asterisk) and the smaller bulbs (arrows)presumably represent the ball and the socket, respectively. (C-E)RNAiagainst kkv is induced by bib-GAL4 (C), neur-GAL4 (D) or fng-GAL4 (E).(C)The presumptive ball cuticle (asterisk) is severely misshaped, similarto in B. The socket cuticle is partly formed (arrow). (D,E)The ball cuticlehas normal morphology. (D)The ventral portion of the socket cuticle ismissing (arrow). (E)The socket cuticle has some bulbs (arrows).(F)Correlation between spatial relationship of bib-, neur- and fng-expressing cells and the types of cuticle that they produce.

DEVELO

PMENT

as shown in Fig. 6F. Importantly, this map is consistent with thepart of the cuticle that each cell population continues to associatewith during cuticle formation (Fig. 5K; see Discussion).

We also attempted to locate the lubricant-producing activity. Asshown in Fig. 1D, Serp(CBD)GFP was enriched in the lubricantwhen it was expressed by Dll-GAL4. When Dll-GAL4 wasreplaced with either bib-, neur- or fng-GAL4, Serp(CBD)GFP stilllocalized in the lubricant (see Fig. S4 in the supplementarymaterial), suggesting that the lubricant-secreting activity isdistributed rather broadly in the joint region. Alternatively, it couldbe that proteins secreted by cells outside of this region aretransported into the lubricant.

Shape change of cuticle-secreting cells proceedseven when cuticle morphology is severelydisruptedAs shown above, the cells progressively relocated their apicalsurfaces in the proximal-to-distal direction (Fig. 5; see also Movies1-3 in the supplementary material), and this was accompanied bypolarized growth of the cuticle in the same direction: stratificationof the ball progressed from dorsal to ventral; the socket and the lipboth elongated from proximal to distal. Interestingly, the apicalsurfaces of the ball-producing cells moved together with the lip,whereas those of the socket-producing cells followed the edge ofthe socket cuticle (Fig. 5). These observations prompted us to testwhether the cuticle acts as a structural guidance cue informing thecells of the direction or the destination of movement.

We used Nikkomycin treatment, which causes severe disruptionof the morphology and continuity of the entire joint cuticles (Fig.6B), to determine whether the cuticle has a guidance role.Nikkomycin was injected at the onset of cuticle formation intopupae in which bib- or fng-expressing cells were labeled bymembrane-bound GFP, and the cell and ECM morphologies wereexamined after 24 hours, when cell shape changes would have beencompleted under normal conditions (cf. Fig. 5G-I). Typically, wesaw that the ball developed as an amorphous bulb without the ‘lip’,and the socket disintegrated into multiple small bulges (Fig. 6B;Fig. 7A,B). In severe cases, the ball even appeared to befragmented into small pieces (Fig. 7C,D).

To our surprise, the overall shapes of bib- and fng-expressingcells in Nikkomycin-injected pupae were comparable to those incontrols. The apical domains of bib-expressing cells had relocatedto the ventral side of the cavity, where they would have normallycontacted the lip (Fig. 7A,C). Thus, the movement of the apices ofball-producing cells appears to be capable of orienting correctly inthe absence of the normal cuticular morphology, including the lip.Likewise, the apical surfaces of the dorsal fng-expressing cellsextended ventrally out of the ridge (Fig. 7B,D), similarly to howthey move in standard conditions. The shape changes of neur-expressing cells in Nikkomycin-injected pupae were alsocomparable to those in the normal situation (data not shown). Thefact that all of the cells relocated their apical surfaces in the correctdirection and to the correct destination, even when the cuticlemorphology was severely disrupted, suggests that the normal shapeand continuity of the cuticle is dispensable for guiding themigration of the apical surface of cuticle-secreting cells.

We also noted that the apical surfaces of socket-producing cellsexpanded so as to wrap around the misshapen or fragmented ballcuticle (Fig. 7B,D), indicating that the final shapes of socket-producing cells respected the morphology of the remaining ballcuticle. This local adjustment of cell shape might reflect the effectof remaining cuticle components. Because the Nikkomycin

treatment did not completely eliminate the cuticle, it is possible thatthe cuticle might play a permissive role in cell movement; forexample, by acting as a substratum for migration.

DISCUSSIONIn this study, we revealed that the ball and the socket cuticlesdevelop sequentially. We have shown that the ball-producingactivity and the socket-producing activity are allocated to distinctcell populations, and have found that shape changes of these cellsthat occur simultaneously with their cuticle-secreting activitiesresult in the interlocking ball-and-socket structure. As the ballcuticle builds up, concurrent cell shape changes drive the apicaldomains of ball-producing cells out of the cavity and bring in theapical domains of the socket-producing cells, resulting in closeenwrapment of the ball by the latter cell population. Accordingly,the shape of the resulting socket cuticle conforms to that of the ball.Synchronization between ECM formation and dynamic relocationof the cell surfaces that mediate it thus underlies the building of thecomplex ECM structure.

Mapping of ball-producing cells and socket-producing cellsThe map of ball-producing and socket-producing cells shown inFig. 6F best summarizes the results of kkv RNAi, and is consistentwith the result indicating their continuous association withrespective parts of the cuticle during their formation (Fig. 5K). Theball morphology was severely disrupted by bib>kkv RNAi but notby neur>kkv RNAi, indicating that the ball-producing activity isrestricted to the distal subset of bib-expressing cells that do notsignificantly express neur. Consistently, these cells are in constantcontact with the ball cuticle throughout its formation. The cuticlephenotype of neur>kkv RNAi shows that neur-expressing cells are

2061RESEARCH ARTICLEBall-and-socket joint morphogenesis

Fig. 7. Cell shape changes occur independently of normal cuticleformation. The final shapes of bib-expressing cells (A,C) or fng-expressing cells (B,D) in Nikkomycin Z-injected animals, visualized bymembrane-targeted GFP. (A,B)Typical cuticular phenotype. The ball andthe socket appear as the large, amorphous bulb on the ventral side(asterisk) and the smaller fragments on the dorsal side (arrowheads),respectively. No obvious lip development. (A)bib-expressing cells areextended along the ventral side of the cavity. Their apical surfaces arelocated distal to the cavity (arrow), where the lip would have formed.(B)The apical surfaces of fng-expressing cells are expanded ventrallyalong the cavity (arrow). (C,D)Severe defects in cuticle morphologyobserved in some Nikkomycin Z-injected animals. The ball isfragmented into a small bulb on the distal side (asterisks) and multiplesmaller bulges (dots). bib-expressing cells are extended distally (C,arrow), and fng-expressing cells expand their apices (D, arrow).

DEVELO

PMENT

2062

responsible for forming the ventral part of the socket cuticle, withwhich they continue to associate. Likewise, fng-expressing cellscontribute to the formation of the remainder of the socket. Partialdisruption of the socket by bib>kkv RNAi (Fig. 6C; see Fig. S3Ein the supplementary material) should be, to some extent, due todirect blocking of socket production in cells co-expressing bib andneur. Additionally, the impairment of ball formation mightsomehow interfere with socket formation. Occasional deformationof the ball by neur>kkv RNAi (see Fig. S3E in the supplementarymaterial) might be caused by marginal expression of neur in thepresumptive ball-producing cells, which is barely detectable in Fig.5A-A�.

Role of dynamic cell behaviors in regulating ECMmorphologyPatterns of ECM-producing tissues do play a major role in theregulation of ECM morphology. Previous studies have unraveledhow global positional information affects skeletal patterns throughthe regulation of specification, differentiation and proliferation ofECM-producing cells (Hall, 2005; Payre, 2004). There, themorphology of ECM was assumed to be synonymous with that ofthe cells or tissues that secrete it. Our present study illustrates thatthe skeletal morphology reflects not only the pattern of those cellsat one point in time, but also the history of their dynamic behaviorsduring ECM formation. Secreted apically by the epidermis, thecuticle is monolayered in most parts. In the joints, however,relocation of the secretory surfaces enables formation of a cuticlebeneath a pre-formed layer. Cell motility thus allows a tissue ofsimple configuration to build a complex and essential three-dimensional ECM structure. We envision that movements of ECM-secreting cells probably play important roles in ECMmorphogenesis in other systems, especially in formation oradjustment of intricate skeletal structures.

Mechanism for cell shape changesThe morphology of the cuticle, as well as how it develops,correlated well with cell shape changes (Fig. 5K). This suggestedeither that the cell-shape changes govern the morphology of thecuticle, and/or vice versa. We found that the movement of theapical surfaces of the cells was correctly oriented even when theshape of the cuticle was disrupted (Fig. 7), indicating that themorphogenesis of the ball-and-socket cuticle is primarily controlledby the way the cells change their shapes as they deposit the cuticle.How do the cells know which way to move? In other words, whatis the molecular mechanism that mediates global proximodistalpolarity of the leg to direct cell movement? In mutants of well-known planar cell polarity genes, such as frizzled, dishevelled andprickled, extra joints of reverse proximodistal polarity are formed(Held et al., 1986). Nonetheless, the ball-and-socket structure ofindividual joints remains largely intact (data not shown), indicatingthat cell shape changes are correctly guided by a mechanism otherthan this pathway. Analysis and local disruption of cytoskeletalarchitecture in the joint region could help us to answer thisquestion.

As mentioned earlier, our results do not rule out the possibilitythat the cuticle plays a permissive role in cell movements. TheECM generally affects cell shape and motility (Alberts et al., 2002),and chitin-based ECM has been shown to regulate epithelialmorphogenesis in some Drosophila tissues (Devine et al., 2005;Moussian et al., 2005; Tonning et al., 2005). Whether the cuticleprovides a permissive environment for cell shape changes in thejoint is an important issue to address in future work.

‘Mold casting’ – a general mechanism for jointdevelopment?The formation of reciprocally shaped interfaces is vital for the sakeof joint function. The serial progression of ball-and-socketmorphogenesis shown here can be compared to ‘mold casting’: (1)the ball enlarges rapidly through stratification, and the cavityexpands to accommodate it; and (2) the enlarged cavity then servesas the ‘mold’ along which the socket cuticle is formed. Hence, theshape of the ball is transmitted to the socket (the ‘cast’). Whetheror not this model also applies to vertebrate synovial joints is anintriguing question. Pacifici et al. speculated that, in the chick digitjoints, chondrogenic cell differentiation on the distal side mightpromote its expansion to become convex; at the same time,proliferation of peripheral cells on the proximal side might permitthem to grow and wrap themselves around the distal side, therebybecoming concave (Pacifici et al., 2005). If this were the case, thatmodel can be regarded as a modified version of our ball-and-socketmorphogenesis, one side fitting to the other through cellproliferation instead of cell shape changes. It will then becomeimportant to study how cells and ECM collectively undergomorphogenesis in other types of joints and in other species.Unraveling similarities and differences in the modes of jointdevelopment would be crucial in a medical sense as well, forunderstanding various joint pathologies and designing therapies totreat them.

AcknowledgementsWe thank Drs S. Luschnig, D. Kiehart and F. Matsuzaki, the Bloomington StockCenter, the Kyoto Stock Center and the Vienna Drosophila RNAi Center for flystrains. We are grateful to Drs S. Kuratani, G. Sheng, N. Niwa, T. Noguchi andT. Kondo for helpful comments on the manuscript, and to all members ofHayashi lab for valuable discussions. This work was supported in part by aResearch Grant for Genome Research from MEXT Japan to S.H. R.T. is a RIKENSpecial Postdoctoral Fellow.

Competing interests statementThe authors declare no competing financial interests.

Supplementary materialSupplementary material for this article is available athttp://dev.biologists.org/lookup/suppl/doi:10.1242/dev.047175/-/DC1

ReferencesAlberts, B., Johnson, A., Lewis, J., Raff, M., Roberts, K. and Walter, P. (2002).

Molecular Biology of the Cell, 4th Edn. New York: Garland Sciences.Bellaiche, Y., Gho, M., Kaltschmidt, J. A., Brand, A. H. and Schweisguth, F.

(2001). Frizzled regulates localization of cell-fate determinants and mitoticspindle rotation during asymmetric cell division. Nat. Cell. Biol. 3, 50-57.

Brand, A. H. and Perrimon, N. (1993). Targeted gene expression as a means ofaltering cell fates and generating dominant phenotypes. Development 118, 401-415.

Calleja, M., Moreno, E., Pelaz, S. and Morata, G. (1996). Visualization of geneexpression in living adult Drosophila. Science 274, 252-255.

Chapman, R. F. (1998). The Insects: Structure and Function. Cambridge:Cambridge University Press.

Cohen, S. M. (1993). Imaginal disc development. In The Development ofDrosophila melanogaster, vol. 2 (ed. M. Bate and A. Martinez-Arias), pp. 747-842. New York: Cold Spring Harbor Laboratory Press.

de Celis, J. F., Tyler, D. M., de Celis, J. and Bray, S. J. (1998). Notchsignalling mediates segmentation of the Drosophila leg. Development 125,4617-4626.

Devine, W. P., Lubarsky, B., Shaw, K., Luschnig, S., Messina, L. and Krasnow,M. A. (2005). Requirement for chitin biosynthesis in epithelial tubemorphogenesis. Proc. Natl. Acad. Sci. USA 102, 17014-17019.

Dietzl, G., Chen, D., Schnorrer, F., Su, K. C., Barinova, Y., Fellner, M., Gasser,B., Kinsey, K., Oppel, S., Scheiblauer, S. et al. (2007). A genome-widetransgenic RNAi library for conditional gene inactivation in Drosophila. Nature448, 151-156.

Fristrom, D. and Fristrom, J. W. (1993). The metamorphic development of theadult epidermis. In The Development of Drosophila melanogaster, vol. 2 (ed. M.Bate and A. Martinez-Arias), pp. 843-898. New York: Cold Spring HarborLaboratory Press.

RESEARCH ARTICLE Development 137 (12)

DEVELO

PMENT

Hall, B. K. (2005). Bones and Cartilage: Developmental and Evolutionary SkeletalBiology: Elsevier Academic Press.

Hayashi, S., Ito, K., Sado, Y., Taniguchi, M., Akimoto, A., Takeuchi, H.,Aigaki, T., Matsuzaki, F., Nakagoshi, H., Tanimura, T. et al. (2002). GETDB, adatabase compiling expression patterns and molecular locations of a collectionof Gal4 enhancer traps. Genesis 34, 58-61.

Held, L. I., Jr, Duarte, C. M. and Derakhshanian, K. (1986). Extra tarsal jointsand abnormal cuticular polaritites in various mutants of Drosophilamelanogaster. Roux’s Arch. Dev. Biol. 195, 145-157.

Kaido, M., Wada, H., Shindo, M. and Hayashi, S. (2009). Essential requirementfor RING finger E3 ubiquitin ligase Hakai in early embryonic development ofDrosophila. Genes Cells 14, 1067-1077.

Kato, K. and Hayashi, S. (2008). Practical guide of live imaging fordevelopmental biologists. Dev Growth Differ 50, 381-390.

Kato, K., Chihara, T. and Hayashi, S. (2004). Hedgehog and Decapentaplegicinstruct polarized growth of cell extensions in the Drosophila trachea.Development 131, 5253-5261.

Khan, I. M., Redman, S. N., Williams, R., Dowthwaite, G. P., Oldfield, S. F.and Archer, C. W. (2007). The development of synovial joints. Curr. Top. Dev.Biol. 79, 1-36.

Kiehart, D. P., Galbraith, C. G., Edwards, K. A., Rickoll, W. L. and Montague,R. A. (2000). Multiple forces contribute to cell sheet morphogenesis for dorsalclosure in Drosophila. J. Cell Biol. 149, 471-490.

Kojima, T. (2004). The mechanism of Drosophila leg development along theproximodistal axis. Dev. Growth Differ. 46, 115-129.

Locke, M. and Huie, P. (1979). Apolysis and the turnover of plasma membraneplaques during cuticle formation in an insect. Tissue Cell 11, 277-291.

Luschnig, S., Batz, T., Armbruster, K. and Krasnow, M. A. (2006). serpentineand vermiform encode matrix proteins with chitin binding and deacetylationdomains that limit tracheal tube length in Drosophila. Curr. Biol. 16, 186-194.

Manjon, C., Sanchez-Herrero, E. and Suzanne, M. (2007). Sharp boundaries ofDpp signalling trigger local cell death required for Drosophila legmorphogenesis. Nat. Cell Biol. 9, 57-63.

Mirth, C. and Akam, M. (2002). Joint development in the Drosophila leg: cellmovements and cell populations. Dev. Biol. 246, 391-406.

Moussian, B., Schwarz, H., Bartoszewski, S. and Nüsslein-Volhard, C. (2005).Involvement of chitin in exoskeleton morphogenesis in Drosophila melanogaster.J. Morphol. 264, 117-130.

Ostrowski, S., Dierick, H. A. and Bejsovec, A. (2002). Genetic control of cuticleformation during embryonic development of Drosophila melanogaster. Genetics161, 171-182.

Pacifici, M., Koyama, E. and Iwamoto, M. (2005). Mechanisms of synovial jointand articular cartilage formation: recent advances, but many lingering mysteries.Birth Defects Res. C Embryo Today 75, 237-248.

Payre, F. (2004). Genetic control of epidermis differentiation in Drosophila. Int. J.Dev. Biol. 48, 207-215.

Rauskolb, C. and Irvine, K. D. (1999). Notch-mediated segmentation and growthcontrol of the Drosophila leg. Dev. Biol. 210, 339-350.

Roberts, D. B. (1986). Drosophila, a Practical Approach. Oxford: IRL Press.Royou, A., Field, C., Sisson, J. C., Sullivan, W. and Karess, R. (2004).

Reassessing the role and dynamics of nonmuscle myosin II during furrowformation in early Drosophila embryos. Mol. Biol. Cell 15, 838-850.

Shirai, T., Yorimitsu, T., Kiritooshi, N., Matsuzaki, F. and Nakagoshi, H.(2007). Notch signaling relieves the joint-suppressive activity of Defectiveproventriculus in the Drosophila leg. Dev. Biol. 312, 147-156.

Standring, S., Wigley, C., Collins, P. and Williams, A. (2005). Gray’s Anatomy,39th edition-The Anatomical Basis of Clinical Practice. Edinburgh: ElsevierChurchill Livingstone.

Sullivan, W., Ashburner, M. and Hawley, R. S. (2000). Drosophila Protocols.New York: Cold Spring Harbor Laboratory Press.

Tellam, R. L., Vuocolo, T., Johnson, S. E., Jarmey, J. and Pearson, R. D. (2000).Insect chitin synthase cDNA sequence, gene organization and expression. Eur. J.Biochem. 267, 6025-6043.

Tonning, A., Hemphala, J., Tang, E., Nannmark, U., Samakovlis, C. and Uv,A. (2005). A transient luminal chitinous matrix is required to model epithelialtube diameter in the Drosophila trachea. Dev. Cell 9, 423-430.

2063RESEARCH ARTICLEBall-and-socket joint morphogenesis

DEVELO

PMENT